{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.7048099455852563,
"global_step": 35490,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 9.92063492063492e-07,
"loss": 10.8421,
"theoretical_loss": 20.81281780154715,
"tokens_seen": 65536
},
{
"epoch": 0.0,
"learning_rate": 1.984126984126984e-06,
"loss": 10.794,
"theoretical_loss": 17.566201104328645,
"tokens_seen": 131072
},
{
"epoch": 0.0,
"learning_rate": 2.9761904761904763e-06,
"loss": 10.7002,
"theoretical_loss": 15.939477092836569,
"tokens_seen": 196608
},
{
"epoch": 0.0,
"learning_rate": 3.968253968253968e-06,
"loss": 10.6201,
"theoretical_loss": 14.89231675598857,
"tokens_seen": 262144
},
{
"epoch": 0.0,
"learning_rate": 4.96031746031746e-06,
"loss": 10.1305,
"theoretical_loss": 14.136216937762974,
"tokens_seen": 327680
},
{
"epoch": 0.0,
"learning_rate": 5.9523809523809525e-06,
"loss": 10.084,
"theoretical_loss": 13.552561472550224,
"tokens_seen": 393216
},
{
"epoch": 0.0,
"learning_rate": 6.944444444444444e-06,
"loss": 9.8364,
"theoretical_loss": 13.08180900140119,
"tokens_seen": 458752
},
{
"epoch": 0.0,
"learning_rate": 7.936507936507936e-06,
"loss": 9.897,
"theoretical_loss": 12.690129625483323,
"tokens_seen": 524288
},
{
"epoch": 0.0,
"learning_rate": 8.928571428571428e-06,
"loss": 9.5044,
"theoretical_loss": 12.356592463873625,
"tokens_seen": 589824
},
{
"epoch": 0.0,
"learning_rate": 9.92063492063492e-06,
"loss": 9.6321,
"theoretical_loss": 12.067412607035077,
"tokens_seen": 655360
},
{
"epoch": 0.0,
"learning_rate": 1.0912698412698412e-05,
"loss": 9.4371,
"theoretical_loss": 11.813066231101676,
"tokens_seen": 720896
},
{
"epoch": 0.0,
"learning_rate": 1.1904761904761905e-05,
"loss": 9.6176,
"theoretical_loss": 11.586719208706729,
"tokens_seen": 786432
},
{
"epoch": 0.0,
"learning_rate": 1.2896825396825396e-05,
"loss": 9.199,
"theoretical_loss": 11.383314140186787,
"tokens_seen": 851968
},
{
"epoch": 0.0,
"learning_rate": 1.3888888888888888e-05,
"loss": 8.8935,
"theoretical_loss": 11.199011702111871,
"tokens_seen": 917504
},
{
"epoch": 0.0,
"learning_rate": 1.4880952380952381e-05,
"loss": 9.3252,
"theoretical_loss": 11.030833917977912,
"tokens_seen": 983040
},
{
"epoch": 0.0,
"learning_rate": 1.5873015873015872e-05,
"loss": 8.6232,
"theoretical_loss": 10.87642808645695,
"tokens_seen": 1048576
},
{
"epoch": 0.0,
"learning_rate": 1.6865079365079364e-05,
"loss": 9.0944,
"theoretical_loss": 10.733905740062724,
"tokens_seen": 1114112
},
{
"epoch": 0.0,
"learning_rate": 1.7857142857142855e-05,
"loss": 8.899,
"theoretical_loss": 10.60172987623028,
"tokens_seen": 1179648
},
{
"epoch": 0.0,
"learning_rate": 1.884920634920635e-05,
"loss": 8.9962,
"theoretical_loss": 10.478634172356642,
"tokens_seen": 1245184
},
{
"epoch": 0.0,
"learning_rate": 1.984126984126984e-05,
"loss": 9.3635,
"theoretical_loss": 10.36356394376333,
"tokens_seen": 1310720
},
{
"epoch": 0.0,
"learning_rate": 2.0833333333333333e-05,
"loss": 8.7939,
"theoretical_loss": 10.255632220896747,
"tokens_seen": 1376256
},
{
"epoch": 0.0,
"learning_rate": 2.1825396825396824e-05,
"loss": 8.8135,
"theoretical_loss": 10.15408655327002,
"tokens_seen": 1441792
},
{
"epoch": 0.0,
"learning_rate": 2.2817460317460315e-05,
"loss": 8.5781,
"theoretical_loss": 10.058283561732598,
"tokens_seen": 1507328
},
{
"epoch": 0.0,
"learning_rate": 2.380952380952381e-05,
"loss": 9.316,
"theoretical_loss": 9.967669178840278,
"tokens_seen": 1572864
},
{
"epoch": 0.0,
"objective/train/docs_used": 13112,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 8.743459701538086,
"objective/train/theoretical_loss": 9.881763126393109,
"objective/train/tokens_used": 22098400,
"theoretical_loss": 9.881763126393109,
"tokens_seen": 1638400
},
{
"epoch": 0.0,
"learning_rate": 2.48015873015873e-05,
"loss": 8.8669,
"theoretical_loss": 9.881763126393109,
"tokens_seen": 1638400
},
{
"epoch": 0.0,
"learning_rate": 2.5793650793650793e-05,
"loss": 8.868,
"theoretical_loss": 9.80014659154056,
"tokens_seen": 1703936
},
{
"epoch": 0.0,
"learning_rate": 2.6785714285714284e-05,
"loss": 8.3477,
"theoretical_loss": 9.722452346907446,
"tokens_seen": 1769472
},
{
"epoch": 0.0,
"learning_rate": 2.7777777777777776e-05,
"loss": 8.5573,
"theoretical_loss": 9.648356759081546,
"tokens_seen": 1835008
},
{
"epoch": 0.0,
"learning_rate": 2.876984126984127e-05,
"loss": 8.7737,
"theoretical_loss": 9.577573271145639,
"tokens_seen": 1900544
},
{
"epoch": 0.0,
"learning_rate": 2.9761904761904762e-05,
"loss": 8.9664,
"theoretical_loss": 9.509847046764852,
"tokens_seen": 1966080
},
{
"epoch": 0.0,
"learning_rate": 3.075396825396825e-05,
"loss": 7.8002,
"theoretical_loss": 9.444950537631936,
"tokens_seen": 2031616
},
{
"epoch": 0.0,
"learning_rate": 3.1746031746031745e-05,
"loss": 8.6073,
"theoretical_loss": 9.382679790910457,
"tokens_seen": 2097152
},
{
"epoch": 0.0,
"learning_rate": 3.273809523809524e-05,
"loss": 8.7208,
"theoretical_loss": 9.32285135423398,
"tokens_seen": 2162688
},
{
"epoch": 0.0,
"learning_rate": 3.373015873015873e-05,
"loss": 8.7212,
"theoretical_loss": 9.265299666660276,
"tokens_seen": 2228224
},
{
"epoch": 0.0,
"learning_rate": 3.472222222222222e-05,
"loss": 8.7873,
"theoretical_loss": 9.209874847444755,
"tokens_seen": 2293760
},
{
"epoch": 0.0,
"learning_rate": 3.571428571428571e-05,
"loss": 8.4388,
"theoretical_loss": 9.156440812508292,
"tokens_seen": 2359296
},
{
"epoch": 0.0,
"learning_rate": 3.670634920634921e-05,
"loss": 8.1114,
"theoretical_loss": 9.10487366241335,
"tokens_seen": 2424832
},
{
"epoch": 0.0,
"learning_rate": 3.76984126984127e-05,
"loss": 8.2518,
"theoretical_loss": 9.055060296533734,
"tokens_seen": 2490368
},
{
"epoch": 0.0,
"learning_rate": 3.8690476190476195e-05,
"loss": 8.2561,
"theoretical_loss": 9.006897216643829,
"tokens_seen": 2555904
},
{
"epoch": 0.0,
"learning_rate": 3.968253968253968e-05,
"loss": 8.1331,
"theoretical_loss": 8.960289489909357,
"tokens_seen": 2621440
},
{
"epoch": 0.0,
"learning_rate": 4.067460317460318e-05,
"loss": 8.0787,
"theoretical_loss": 8.915149846640611,
"tokens_seen": 2686976
},
{
"epoch": 0.0,
"learning_rate": 4.1666666666666665e-05,
"loss": 8.3421,
"theoretical_loss": 8.871397892478225,
"tokens_seen": 2752512
},
{
"epoch": 0.0,
"learning_rate": 4.265873015873016e-05,
"loss": 8.0099,
"theoretical_loss": 8.828959418153499,
"tokens_seen": 2818048
},
{
"epoch": 0.0,
"learning_rate": 4.365079365079365e-05,
"loss": 8.2887,
"theoretical_loss": 8.787765792778412,
"tokens_seen": 2883584
},
{
"epoch": 0.0,
"learning_rate": 4.464285714285714e-05,
"loss": 8.5295,
"theoretical_loss": 8.747753428911455,
"tokens_seen": 2949120
},
{
"epoch": 0.0,
"learning_rate": 4.563492063492063e-05,
"loss": 8.2913,
"theoretical_loss": 8.708863309520833,
"tokens_seen": 3014656
},
{
"epoch": 0.0,
"learning_rate": 4.6626984126984126e-05,
"loss": 8.319,
"theoretical_loss": 8.671040568508847,
"tokens_seen": 3080192
},
{
"epoch": 0.0,
"learning_rate": 4.761904761904762e-05,
"loss": 7.8599,
"theoretical_loss": 8.634234117735474,
"tokens_seen": 3145728
},
{
"epoch": 0.0,
"learning_rate": 4.8611111111111115e-05,
"loss": 7.7079,
"theoretical_loss": 8.598396314536323,
"tokens_seen": 3211264
},
{
"epoch": 0.0,
"objective/train/docs_used": 13755,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 7.561188220977783,
"objective/train/theoretical_loss": 8.563482664611069,
"objective/train/tokens_used": 23736800,
"theoretical_loss": 8.563482664611069,
"tokens_seen": 3276800
},
{
"epoch": 0.0,
"learning_rate": 4.96031746031746e-05,
"loss": 7.8004,
"theoretical_loss": 8.563482664611069,
"tokens_seen": 3276800
},
{
"epoch": 0.0,
"learning_rate": 5.05952380952381e-05,
"loss": 8.0798,
"theoretical_loss": 8.529451555895115,
"tokens_seen": 3342336
},
{
"epoch": 0.0,
"learning_rate": 5.1587301587301586e-05,
"loss": 7.9158,
"theoretical_loss": 8.496264019646002,
"tokens_seen": 3407872
},
{
"epoch": 0.0,
"learning_rate": 5.257936507936508e-05,
"loss": 7.5747,
"theoretical_loss": 8.463883515497187,
"tokens_seen": 3473408
},
{
"epoch": 0.0,
"learning_rate": 5.357142857142857e-05,
"loss": 7.8708,
"theoretical_loss": 8.432275737672779,
"tokens_seen": 3538944
},
{
"epoch": 0.0,
"learning_rate": 5.4563492063492063e-05,
"loss": 7.7568,
"theoretical_loss": 8.401408439930716,
"tokens_seen": 3604480
},
{
"epoch": 0.0,
"learning_rate": 5.555555555555555e-05,
"loss": 7.86,
"theoretical_loss": 8.371251277120209,
"tokens_seen": 3670016
},
{
"epoch": 0.0,
"learning_rate": 5.6547619047619046e-05,
"loss": 7.5876,
"theoretical_loss": 8.341775661511075,
"tokens_seen": 3735552
},
{
"epoch": 0.0,
"learning_rate": 5.753968253968254e-05,
"loss": 7.0476,
"theoretical_loss": 8.31295463228533,
"tokens_seen": 3801088
},
{
"epoch": 0.0,
"learning_rate": 5.8531746031746036e-05,
"loss": 7.6287,
"theoretical_loss": 8.284762736781182,
"tokens_seen": 3866624
},
{
"epoch": 0.0,
"learning_rate": 5.9523809523809524e-05,
"loss": 7.8486,
"theoretical_loss": 8.257175922251864,
"tokens_seen": 3932160
},
{
"epoch": 0.0,
"learning_rate": 6.051587301587302e-05,
"loss": 7.7722,
"theoretical_loss": 8.230171437050114,
"tokens_seen": 3997696
},
{
"epoch": 0.0,
"learning_rate": 6.15079365079365e-05,
"loss": 7.7011,
"theoretical_loss": 8.20372774027797,
"tokens_seen": 4063232
},
{
"epoch": 0.0,
"learning_rate": 6.25e-05,
"loss": 7.8601,
"theoretical_loss": 8.177824419053046,
"tokens_seen": 4128768
},
{
"epoch": 0.0,
"learning_rate": 6.349206349206349e-05,
"loss": 7.4565,
"theoretical_loss": 8.152442112639616,
"tokens_seen": 4194304
},
{
"epoch": 0.0,
"learning_rate": 6.448412698412699e-05,
"loss": 7.3253,
"theoretical_loss": 8.1275624427775,
"tokens_seen": 4259840
},
{
"epoch": 0.0,
"learning_rate": 6.547619047619048e-05,
"loss": 7.1366,
"theoretical_loss": 8.10316794961571,
"tokens_seen": 4325376
},
{
"epoch": 0.0,
"learning_rate": 6.646825396825397e-05,
"loss": 7.4872,
"theoretical_loss": 8.07924203272264,
"tokens_seen": 4390912
},
{
"epoch": 0.0,
"learning_rate": 6.746031746031745e-05,
"loss": 7.5183,
"theoretical_loss": 8.055768896701416,
"tokens_seen": 4456448
},
{
"epoch": 0.0,
"learning_rate": 6.845238095238096e-05,
"loss": 7.0588,
"theoretical_loss": 8.032733500989007,
"tokens_seen": 4521984
},
{
"epoch": 0.0,
"learning_rate": 6.944444444444444e-05,
"loss": 7.3674,
"theoretical_loss": 8.010121513461836,
"tokens_seen": 4587520
},
{
"epoch": 0.0,
"learning_rate": 7.043650793650793e-05,
"loss": 7.0203,
"theoretical_loss": 7.987919267509379,
"tokens_seen": 4653056
},
{
"epoch": 0.0,
"learning_rate": 7.142857142857142e-05,
"loss": 7.0155,
"theoretical_loss": 7.966113722271801,
"tokens_seen": 4718592
},
{
"epoch": 0.0,
"learning_rate": 7.242063492063492e-05,
"loss": 7.2468,
"theoretical_loss": 7.944692425767988,
"tokens_seen": 4784128
},
{
"epoch": 0.0,
"learning_rate": 7.341269841269842e-05,
"loss": 7.022,
"theoretical_loss": 7.9236434806675184,
"tokens_seen": 4849664
},
{
"epoch": 0.0,
"objective/train/docs_used": 15099,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 6.9753007888793945,
"objective/train/theoretical_loss": 7.902955512484067,
"objective/train/tokens_used": 25375200,
"theoretical_loss": 7.902955512484067,
"tokens_seen": 4915200
},
{
"epoch": 0.0,
"learning_rate": 7.440476190476191e-05,
"loss": 7.1984,
"theoretical_loss": 7.902955512484067,
"tokens_seen": 4915200
},
{
"epoch": 0.0,
"learning_rate": 7.53968253968254e-05,
"loss": 7.0639,
"theoretical_loss": 7.882617639989203,
"tokens_seen": 4980736
},
{
"epoch": 0.0,
"learning_rate": 7.63888888888889e-05,
"loss": 6.8249,
"theoretical_loss": 7.862619447664628,
"tokens_seen": 5046272
},
{
"epoch": 0.0,
"learning_rate": 7.738095238095239e-05,
"loss": 6.9557,
"theoretical_loss": 7.842950960027937,
"tokens_seen": 5111808
},
{
"epoch": 0.0,
"learning_rate": 7.837301587301588e-05,
"loss": 7.1772,
"theoretical_loss": 7.823602617682313,
"tokens_seen": 5177344
},
{
"epoch": 0.0,
"learning_rate": 7.936507936507937e-05,
"loss": 6.7487,
"theoretical_loss": 7.804565254954165,
"tokens_seen": 5242880
},
{
"epoch": 0.0,
"learning_rate": 8.035714285714287e-05,
"loss": 6.7244,
"theoretical_loss": 7.7858300789950725,
"tokens_seen": 5308416
},
{
"epoch": 0.0,
"learning_rate": 8.134920634920635e-05,
"loss": 7.2012,
"theoretical_loss": 7.767388650235364,
"tokens_seen": 5373952
},
{
"epoch": 0.0,
"learning_rate": 8.234126984126984e-05,
"loss": 6.7945,
"theoretical_loss": 7.749232864086619,
"tokens_seen": 5439488
},
{
"epoch": 0.0,
"learning_rate": 8.333333333333333e-05,
"loss": 6.4889,
"theoretical_loss": 7.731354933799318,
"tokens_seen": 5505024
},
{
"epoch": 0.0,
"learning_rate": 8.432539682539683e-05,
"loss": 7.0238,
"theoretical_loss": 7.71374737438992,
"tokens_seen": 5570560
},
{
"epoch": 0.0,
"learning_rate": 8.531746031746032e-05,
"loss": 6.9503,
"theoretical_loss": 7.696402987558934,
"tokens_seen": 5636096
},
{
"epoch": 0.0,
"learning_rate": 8.630952380952381e-05,
"loss": 6.8864,
"theoretical_loss": 7.679314847528181,
"tokens_seen": 5701632
},
{
"epoch": 0.0,
"learning_rate": 8.73015873015873e-05,
"loss": 6.5425,
"theoretical_loss": 7.662476287731328,
"tokens_seen": 5767168
},
{
"epoch": 0.0,
"learning_rate": 8.82936507936508e-05,
"loss": 6.9681,
"theoretical_loss": 7.645880888297279,
"tokens_seen": 5832704
},
{
"epoch": 0.0,
"learning_rate": 8.928571428571429e-05,
"loss": 6.4495,
"theoretical_loss": 7.629522464270861,
"tokens_seen": 5898240
},
{
"epoch": 0.0,
"learning_rate": 9.027777777777777e-05,
"loss": 6.5397,
"theoretical_loss": 7.613395054519696,
"tokens_seen": 5963776
},
{
"epoch": 0.0,
"learning_rate": 9.126984126984126e-05,
"loss": 6.4287,
"theoretical_loss": 7.59749291128028,
"tokens_seen": 6029312
},
{
"epoch": 0.0,
"learning_rate": 9.226190476190476e-05,
"loss": 6.5709,
"theoretical_loss": 7.581810490299888,
"tokens_seen": 6094848
},
{
"epoch": 0.0,
"learning_rate": 9.325396825396825e-05,
"loss": 6.7154,
"theoretical_loss": 7.5663424415343705,
"tokens_seen": 6160384
},
{
"epoch": 0.0,
"learning_rate": 9.424603174603175e-05,
"loss": 6.4567,
"theoretical_loss": 7.551083600364949,
"tokens_seen": 6225920
},
{
"epoch": 0.0,
"learning_rate": 9.523809523809524e-05,
"loss": 6.351,
"theoretical_loss": 7.536028979299919,
"tokens_seen": 6291456
},
{
"epoch": 0.0,
"learning_rate": 9.623015873015874e-05,
"loss": 6.7204,
"theoretical_loss": 7.521173760129762,
"tokens_seen": 6356992
},
{
"epoch": 0.0,
"learning_rate": 9.722222222222223e-05,
"loss": 6.9272,
"theoretical_loss": 7.506513286506497,
"tokens_seen": 6422528
},
{
"epoch": 0.0,
"learning_rate": 9.821428571428572e-05,
"loss": 6.3143,
"theoretical_loss": 7.492043056920249,
"tokens_seen": 6488064
},
{
"epoch": 0.0,
"objective/train/docs_used": 15777,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 6.954127788543701,
"objective/train/theoretical_loss": 7.4777587180480305,
"objective/train/tokens_used": 27013600,
"theoretical_loss": 7.4777587180480305,
"tokens_seen": 6553600
},
{
"epoch": 0.0,
"learning_rate": 9.92063492063492e-05,
"loss": 6.6163,
"theoretical_loss": 7.4777587180480305,
"tokens_seen": 6553600
},
{
"epoch": 0.0,
"learning_rate": 0.00010019841269841271,
"loss": 6.3178,
"theoretical_loss": 7.463656058451462,
"tokens_seen": 6619136
},
{
"epoch": 0.0,
"learning_rate": 0.0001011904761904762,
"loss": 6.3722,
"theoretical_loss": 7.449731002601916,
"tokens_seen": 6684672
},
{
"epoch": 0.0,
"learning_rate": 0.00010218253968253968,
"loss": 6.2211,
"theoretical_loss": 7.435979605213019,
"tokens_seen": 6750208
},
{
"epoch": 0.0,
"learning_rate": 0.00010317460317460317,
"loss": 6.5174,
"theoretical_loss": 7.422398045861905,
"tokens_seen": 6815744
},
{
"epoch": 0.0,
"learning_rate": 0.00010416666666666667,
"loss": 6.6538,
"theoretical_loss": 7.408982623881875,
"tokens_seen": 6881280
},
{
"epoch": 0.0,
"learning_rate": 0.00010515873015873016,
"loss": 6.1797,
"theoretical_loss": 7.395729753510345,
"tokens_seen": 6946816
},
{
"epoch": 0.0,
"learning_rate": 0.00010615079365079365,
"loss": 6.5865,
"theoretical_loss": 7.3826359592770325,
"tokens_seen": 7012352
},
{
"epoch": 0.0,
"learning_rate": 0.00010714285714285714,
"loss": 6.5482,
"theoretical_loss": 7.369697871618373,
"tokens_seen": 7077888
},
{
"epoch": 0.0,
"learning_rate": 0.00010813492063492064,
"loss": 6.7304,
"theoretical_loss": 7.3569122227050885,
"tokens_seen": 7143424
},
{
"epoch": 0.0,
"learning_rate": 0.00010912698412698413,
"loss": 6.2506,
"theoretical_loss": 7.3442758424706875,
"tokens_seen": 7208960
},
{
"epoch": 0.0,
"learning_rate": 0.00011011904761904761,
"loss": 6.1023,
"theoretical_loss": 7.331785654829519,
"tokens_seen": 7274496
},
{
"epoch": 0.0,
"learning_rate": 0.0001111111111111111,
"loss": 6.6284,
"theoretical_loss": 7.319438674073677,
"tokens_seen": 7340032
},
{
"epoch": 0.0,
"learning_rate": 0.0001121031746031746,
"loss": 6.3081,
"theoretical_loss": 7.307232001438824,
"tokens_seen": 7405568
},
{
"epoch": 0.0,
"learning_rate": 0.00011309523809523809,
"loss": 6.299,
"theoretical_loss": 7.295162821829564,
"tokens_seen": 7471104
},
{
"epoch": 0.0,
"learning_rate": 0.00011408730158730158,
"loss": 6.5472,
"theoretical_loss": 7.283228400695652,
"tokens_seen": 7536640
},
{
"epoch": 0.0,
"learning_rate": 0.00011507936507936508,
"loss": 6.292,
"theoretical_loss": 7.271426081050832,
"tokens_seen": 7602176
},
{
"epoch": 0.0,
"learning_rate": 0.00011607142857142858,
"loss": 6.2414,
"theoretical_loss": 7.259753280626623,
"tokens_seen": 7667712
},
{
"epoch": 0.0,
"learning_rate": 0.00011706349206349207,
"loss": 6.3899,
"theoretical_loss": 7.24820748915387,
"tokens_seen": 7733248
},
{
"epoch": 0.0,
"learning_rate": 0.00011805555555555556,
"loss": 6.3743,
"theoretical_loss": 7.236786265765262,
"tokens_seen": 7798784
},
{
"epoch": 0.0,
"learning_rate": 0.00011904761904761905,
"loss": 5.9537,
"theoretical_loss": 7.225487236512497,
"tokens_seen": 7864320
},
{
"epoch": 0.0,
"learning_rate": 0.00012003968253968255,
"loss": 6.0297,
"theoretical_loss": 7.21430809199212,
"tokens_seen": 7929856
},
{
"epoch": 0.0,
"learning_rate": 0.00012103174603174604,
"loss": 6.5923,
"theoretical_loss": 7.2032465850744005,
"tokens_seen": 7995392
},
{
"epoch": 0.0,
"learning_rate": 0.00012202380952380953,
"loss": 6.116,
"theoretical_loss": 7.192300528730015,
"tokens_seen": 8060928
},
{
"epoch": 0.0,
"learning_rate": 0.000123015873015873,
"loss": 6.4398,
"theoretical_loss": 7.1814677939495155,
"tokens_seen": 8126464
},
{
"epoch": 0.0,
"objective/train/docs_used": 17166,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 6.351791858673096,
"objective/train/theoretical_loss": 7.1707463077509646,
"objective/train/tokens_used": 28652000,
"theoretical_loss": 7.1707463077509646,
"tokens_seen": 8192000
},
{
"epoch": 0.0,
"learning_rate": 0.0001240079365079365,
"loss": 6.4479,
"theoretical_loss": 7.1707463077509646,
"tokens_seen": 8192000
},
{
"epoch": 0.0,
"learning_rate": 0.000125,
"loss": 6.335,
"theoretical_loss": 7.160134051271272,
"tokens_seen": 8257536
},
{
"epoch": 0.0,
"learning_rate": 0.0001259920634920635,
"loss": 6.1976,
"theoretical_loss": 7.149629057937138,
"tokens_seen": 8323072
},
{
"epoch": 0.0,
"learning_rate": 0.00012698412698412698,
"loss": 6.4173,
"theoretical_loss": 7.139229411711638,
"tokens_seen": 8388608
},
{
"epoch": 0.0,
"learning_rate": 0.00012797619047619048,
"loss": 6.0317,
"theoretical_loss": 7.128933245412794,
"tokens_seen": 8454144
},
{
"epoch": 0.0,
"learning_rate": 0.00012896825396825398,
"loss": 6.0573,
"theoretical_loss": 7.118738739100616,
"tokens_seen": 8519680
},
{
"epoch": 0.0,
"learning_rate": 0.00012996031746031748,
"loss": 6.0885,
"theoretical_loss": 7.1086441185293445,
"tokens_seen": 8585216
},
{
"epoch": 0.0,
"learning_rate": 0.00013095238095238096,
"loss": 6.3385,
"theoretical_loss": 7.09864765366177,
"tokens_seen": 8650752
},
{
"epoch": 0.0,
"learning_rate": 0.00013194444444444446,
"loss": 6.1813,
"theoretical_loss": 7.088747657242693,
"tokens_seen": 8716288
},
{
"epoch": 0.0,
"learning_rate": 0.00013293650793650793,
"loss": 6.368,
"theoretical_loss": 7.078942483428749,
"tokens_seen": 8781824
},
{
"epoch": 0.0,
"learning_rate": 0.00013392857142857144,
"loss": 5.9535,
"theoretical_loss": 7.069230526471966,
"tokens_seen": 8847360
},
{
"epoch": 0.0,
"learning_rate": 0.0001349206349206349,
"loss": 6.0473,
"theoretical_loss": 7.059610219454568,
"tokens_seen": 8912896
},
{
"epoch": 0.0,
"learning_rate": 0.0001359126984126984,
"loss": 6.0929,
"theoretical_loss": 7.0500800330726685,
"tokens_seen": 8978432
},
{
"epoch": 0.0,
"learning_rate": 0.0001369047619047619,
"loss": 6.3183,
"theoretical_loss": 7.040638474466625,
"tokens_seen": 9043968
},
{
"epoch": 0.0,
"learning_rate": 0.00013789682539682541,
"loss": 6.2457,
"theoretical_loss": 7.031284086095933,
"tokens_seen": 9109504
},
{
"epoch": 0.0,
"learning_rate": 0.0001388888888888889,
"loss": 6.2984,
"theoretical_loss": 7.022015444656678,
"tokens_seen": 9175040
},
{
"epoch": 0.0,
"learning_rate": 0.0001398809523809524,
"loss": 6.1863,
"theoretical_loss": 7.012831160039609,
"tokens_seen": 9240576
},
{
"epoch": 0.0,
"learning_rate": 0.00014087301587301586,
"loss": 6.3604,
"theoretical_loss": 7.003729874327071,
"tokens_seen": 9306112
},
{
"epoch": 0.0,
"learning_rate": 0.00014186507936507937,
"loss": 6.0677,
"theoretical_loss": 6.994710260827057,
"tokens_seen": 9371648
},
{
"epoch": 0.0,
"learning_rate": 0.00014285714285714284,
"loss": 6.2727,
"theoretical_loss": 6.98577102314278,
"tokens_seen": 9437184
},
{
"epoch": 0.0,
"learning_rate": 0.00014384920634920634,
"loss": 6.2095,
"theoretical_loss": 6.976910894276189,
"tokens_seen": 9502720
},
{
"epoch": 0.0,
"learning_rate": 0.00014484126984126984,
"loss": 6.2549,
"theoretical_loss": 6.968128635764015,
"tokens_seen": 9568256
},
{
"epoch": 0.0,
"learning_rate": 0.00014583333333333335,
"loss": 6.3308,
"theoretical_loss": 6.959423036844894,
"tokens_seen": 9633792
},
{
"epoch": 0.0,
"learning_rate": 0.00014682539682539685,
"loss": 6.364,
"theoretical_loss": 6.950792913656309,
"tokens_seen": 9699328
},
{
"epoch": 0.0,
"learning_rate": 0.00014781746031746032,
"loss": 6.0516,
"theoretical_loss": 6.942237108460029,
"tokens_seen": 9764864
},
{
"epoch": 0.0,
"objective/train/docs_used": 17861,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 6.117774486541748,
"objective/train/theoretical_loss": 6.9337544888949,
"objective/train/tokens_used": 30290400,
"theoretical_loss": 6.9337544888949,
"tokens_seen": 9830400
},
{
"epoch": 0.0,
"learning_rate": 0.00014880952380952382,
"loss": 6.1082,
"theoretical_loss": 6.9337544888949,
"tokens_seen": 9830400
},
{
"epoch": 0.0,
"learning_rate": 0.0001498015873015873,
"loss": 5.8009,
"theoretical_loss": 6.925343947255817,
"tokens_seen": 9895936
},
{
"epoch": 0.0,
"learning_rate": 0.0001507936507936508,
"loss": 5.8948,
"theoretical_loss": 6.917004399797798,
"tokens_seen": 9961472
},
{
"epoch": 0.0,
"learning_rate": 0.00015178571428571427,
"loss": 5.8529,
"theoretical_loss": 6.908734786064147,
"tokens_seen": 10027008
},
{
"epoch": 0.0,
"learning_rate": 0.0001527777777777778,
"loss": 6.0551,
"theoretical_loss": 6.900534068237688,
"tokens_seen": 10092544
},
{
"epoch": 0.0,
"learning_rate": 0.00015376984126984128,
"loss": 6.1428,
"theoretical_loss": 6.89240123051416,
"tokens_seen": 10158080
},
{
"epoch": 0.0,
"learning_rate": 0.00015476190476190478,
"loss": 5.741,
"theoretical_loss": 6.884335278496871,
"tokens_seen": 10223616
},
{
"epoch": 0.0,
"learning_rate": 0.00015575396825396825,
"loss": 5.8161,
"theoretical_loss": 6.87633523861175,
"tokens_seen": 10289152
},
{
"epoch": 0.0,
"learning_rate": 0.00015674603174603175,
"loss": 6.1526,
"theoretical_loss": 6.868400157541997,
"tokens_seen": 10354688
},
{
"epoch": 0.0,
"learning_rate": 0.00015773809523809523,
"loss": 5.9937,
"theoretical_loss": 6.860529101681551,
"tokens_seen": 10420224
},
{
"epoch": 0.0,
"learning_rate": 0.00015873015873015873,
"loss": 6.0325,
"theoretical_loss": 6.85272115660663,
"tokens_seen": 10485760
},
{
"epoch": 0.0,
"learning_rate": 0.0001597222222222222,
"loss": 5.938,
"theoretical_loss": 6.844975426564642,
"tokens_seen": 10551296
},
{
"epoch": 0.0,
"learning_rate": 0.00016071428571428573,
"loss": 6.1597,
"theoretical_loss": 6.8372910339797945,
"tokens_seen": 10616832
},
{
"epoch": 0.0,
"learning_rate": 0.0001617063492063492,
"loss": 5.92,
"theoretical_loss": 6.829667118974749,
"tokens_seen": 10682368
},
{
"epoch": 0.0,
"learning_rate": 0.0001626984126984127,
"loss": 5.4308,
"theoretical_loss": 6.8221028389077185,
"tokens_seen": 10747904
},
{
"epoch": 0.0,
"learning_rate": 0.00016369047619047618,
"loss": 6.2676,
"theoretical_loss": 6.814597367924395,
"tokens_seen": 10813440
},
{
"epoch": 0.0,
"learning_rate": 0.00016468253968253969,
"loss": 6.0017,
"theoretical_loss": 6.807149896524181,
"tokens_seen": 10878976
},
{
"epoch": 0.0,
"learning_rate": 0.00016567460317460316,
"loss": 6.1886,
"theoretical_loss": 6.799759631140145,
"tokens_seen": 10944512
},
{
"epoch": 0.0,
"learning_rate": 0.00016666666666666666,
"loss": 5.7028,
"theoretical_loss": 6.7924257937322245,
"tokens_seen": 11010048
},
{
"epoch": 0.0,
"learning_rate": 0.00016765873015873016,
"loss": 6.1481,
"theoretical_loss": 6.785147621393148,
"tokens_seen": 11075584
},
{
"epoch": 0.0,
"learning_rate": 0.00016865079365079366,
"loss": 5.8603,
"theoretical_loss": 6.777924365966638,
"tokens_seen": 11141120
},
{
"epoch": 0.0,
"learning_rate": 0.00016964285714285717,
"loss": 5.6239,
"theoretical_loss": 6.770755293677423,
"tokens_seen": 11206656
},
{
"epoch": 0.0,
"learning_rate": 0.00017063492063492064,
"loss": 5.6818,
"theoretical_loss": 6.763639684772625,
"tokens_seen": 11272192
},
{
"epoch": 0.0,
"learning_rate": 0.00017162698412698414,
"loss": 5.7517,
"theoretical_loss": 6.756576833174123,
"tokens_seen": 11337728
},
{
"epoch": 0.0,
"learning_rate": 0.00017261904761904762,
"loss": 5.7609,
"theoretical_loss": 6.749566046141486,
"tokens_seen": 11403264
},
{
"epoch": 0.0,
"objective/train/docs_used": 18493,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.560274600982666,
"objective/train/theoretical_loss": 6.7426066439450905,
"objective/train/tokens_used": 31928800,
"theoretical_loss": 6.7426066439450905,
"tokens_seen": 11468800
},
{
"epoch": 0.0,
"learning_rate": 0.00017361111111111112,
"loss": 5.9297,
"theoretical_loss": 6.7426066439450905,
"tokens_seen": 11468800
},
{
"epoch": 0.0,
"learning_rate": 0.0001746031746031746,
"loss": 5.8221,
"theoretical_loss": 6.735697959549075,
"tokens_seen": 11534336
},
{
"epoch": 0.0,
"learning_rate": 0.0001755952380952381,
"loss": 6.0875,
"theoretical_loss": 6.728839338303761,
"tokens_seen": 11599872
},
{
"epoch": 0.0,
"learning_rate": 0.0001765873015873016,
"loss": 5.6213,
"theoretical_loss": 6.722030137647226,
"tokens_seen": 11665408
},
{
"epoch": 0.0,
"learning_rate": 0.0001775793650793651,
"loss": 5.8156,
"theoretical_loss": 6.715269726815689,
"tokens_seen": 11730944
},
{
"epoch": 0.0,
"learning_rate": 0.00017857142857142857,
"loss": 5.6855,
"theoretical_loss": 6.7085574865624125,
"tokens_seen": 11796480
},
{
"epoch": 0.0,
"learning_rate": 0.00017956349206349207,
"loss": 6.0002,
"theoretical_loss": 6.701892808884824,
"tokens_seen": 11862016
},
{
"epoch": 0.0,
"learning_rate": 0.00018055555555555555,
"loss": 5.4675,
"theoretical_loss": 6.695275096759559,
"tokens_seen": 11927552
},
{
"epoch": 0.0,
"learning_rate": 0.00018154761904761905,
"loss": 5.6532,
"theoretical_loss": 6.68870376388518,
"tokens_seen": 11993088
},
{
"epoch": 0.0,
"learning_rate": 0.00018253968253968252,
"loss": 5.6652,
"theoretical_loss": 6.682178234432274,
"tokens_seen": 12058624
},
{
"epoch": 0.0,
"learning_rate": 0.00018353174603174602,
"loss": 5.7053,
"theoretical_loss": 6.675697942800715,
"tokens_seen": 12124160
},
{
"epoch": 0.0,
"learning_rate": 0.00018452380952380953,
"loss": 5.638,
"theoretical_loss": 6.669262333383815,
"tokens_seen": 12189696
},
{
"epoch": 0.0,
"learning_rate": 0.00018551587301587303,
"loss": 5.9672,
"theoretical_loss": 6.662870860339158,
"tokens_seen": 12255232
},
{
"epoch": 0.0,
"learning_rate": 0.0001865079365079365,
"loss": 5.785,
"theoretical_loss": 6.656522987365879,
"tokens_seen": 12320768
},
{
"epoch": 0.0,
"learning_rate": 0.0001875,
"loss": 5.7804,
"theoretical_loss": 6.6502181874881705,
"tokens_seen": 12386304
},
{
"epoch": 0.0,
"learning_rate": 0.0001884920634920635,
"loss": 5.7629,
"theoretical_loss": 6.643955942844831,
"tokens_seen": 12451840
},
{
"epoch": 0.0,
"learning_rate": 0.00018948412698412698,
"loss": 5.7272,
"theoretical_loss": 6.637735744484626,
"tokens_seen": 12517376
},
{
"epoch": 0.0,
"learning_rate": 0.00019047619047619048,
"loss": 5.9396,
"theoretical_loss": 6.631557092167304,
"tokens_seen": 12582912
},
{
"epoch": 0.0,
"learning_rate": 0.00019146825396825398,
"loss": 5.6727,
"theoretical_loss": 6.625419494170049,
"tokens_seen": 12648448
},
{
"epoch": 0.0,
"learning_rate": 0.00019246031746031748,
"loss": 5.9043,
"theoretical_loss": 6.619322467099223,
"tokens_seen": 12713984
},
{
"epoch": 0.0,
"learning_rate": 0.00019345238095238096,
"loss": 5.4898,
"theoretical_loss": 6.613265535707211,
"tokens_seen": 12779520
},
{
"epoch": 0.0,
"learning_rate": 0.00019444444444444446,
"loss": 5.5153,
"theoretical_loss": 6.607248232714213,
"tokens_seen": 12845056
},
{
"epoch": 0.0,
"learning_rate": 0.00019543650793650793,
"loss": 5.534,
"theoretical_loss": 6.60127009863481,
"tokens_seen": 12910592
},
{
"epoch": 0.0,
"learning_rate": 0.00019642857142857144,
"loss": 5.7539,
"theoretical_loss": 6.59533068160918,
"tokens_seen": 12976128
},
{
"epoch": 0.0,
"learning_rate": 0.0001974206349206349,
"loss": 5.633,
"theoretical_loss": 6.589429537238785,
"tokens_seen": 13041664
},
{
"epoch": 0.0,
"objective/train/docs_used": 19731,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.428110122680664,
"objective/train/theoretical_loss": 6.583566228426414,
"objective/train/tokens_used": 33567200,
"theoretical_loss": 6.583566228426414,
"tokens_seen": 13107200
},
{
"epoch": 0.0,
"learning_rate": 0.0001984126984126984,
"loss": 5.3663,
"theoretical_loss": 6.583566228426414,
"tokens_seen": 13107200
},
{
"epoch": 0.0,
"learning_rate": 0.00019940476190476191,
"loss": 5.7325,
"theoretical_loss": 6.5777403252204305,
"tokens_seen": 13172736
},
{
"epoch": 0.0,
"learning_rate": 0.00020039682539682542,
"loss": 5.8639,
"theoretical_loss": 6.571951404663098,
"tokens_seen": 13238272
},
{
"epoch": 0.0,
"learning_rate": 0.0002013888888888889,
"loss": 5.6695,
"theoretical_loss": 6.566199050642863,
"tokens_seen": 13303808
},
{
"epoch": 0.0,
"learning_rate": 0.0002023809523809524,
"loss": 5.5443,
"theoretical_loss": 6.560482853750463,
"tokens_seen": 13369344
},
{
"epoch": 0.0,
"learning_rate": 0.00020337301587301587,
"loss": 5.599,
"theoretical_loss": 6.554802411138745,
"tokens_seen": 13434880
},
{
"epoch": 0.0,
"learning_rate": 0.00020436507936507937,
"loss": 5.3973,
"theoretical_loss": 6.549157326386091,
"tokens_seen": 13500416
},
{
"epoch": 0.0,
"learning_rate": 0.00020535714285714284,
"loss": 5.6837,
"theoretical_loss": 6.54354720936333,
"tokens_seen": 13565952
},
{
"epoch": 0.0,
"learning_rate": 0.00020634920634920634,
"loss": 5.4813,
"theoretical_loss": 6.537971676104026,
"tokens_seen": 13631488
},
{
"epoch": 0.0,
"learning_rate": 0.00020734126984126985,
"loss": 5.6414,
"theoretical_loss": 6.532430348678068,
"tokens_seen": 13697024
},
{
"epoch": 0.0,
"learning_rate": 0.00020833333333333335,
"loss": 5.5939,
"theoretical_loss": 6.5269228550684195,
"tokens_seen": 13762560
},
{
"epoch": 0.0,
"learning_rate": 0.00020932539682539685,
"loss": 5.2664,
"theoretical_loss": 6.521448829050978,
"tokens_seen": 13828096
},
{
"epoch": 0.0,
"learning_rate": 0.00021031746031746032,
"loss": 5.7298,
"theoretical_loss": 6.516007910077416,
"tokens_seen": 13893632
},
{
"epoch": 0.0,
"learning_rate": 0.00021130952380952382,
"loss": 5.7504,
"theoretical_loss": 6.51059974316095,
"tokens_seen": 13959168
},
{
"epoch": 0.0,
"learning_rate": 0.0002123015873015873,
"loss": 5.6754,
"theoretical_loss": 6.50522397876491,
"tokens_seen": 14024704
},
{
"epoch": 0.0,
"learning_rate": 0.0002132936507936508,
"loss": 6.0764,
"theoretical_loss": 6.499880272694068,
"tokens_seen": 14090240
},
{
"epoch": 0.0,
"learning_rate": 0.00021428571428571427,
"loss": 5.7123,
"theoretical_loss": 6.494568285988618,
"tokens_seen": 14155776
},
{
"epoch": 0.0,
"learning_rate": 0.0002152777777777778,
"loss": 5.5138,
"theoretical_loss": 6.489287684820745,
"tokens_seen": 14221312
},
{
"epoch": 0.0,
"learning_rate": 0.00021626984126984128,
"loss": 5.9094,
"theoretical_loss": 6.484038140393699,
"tokens_seen": 14286848
},
{
"epoch": 0.0,
"learning_rate": 0.00021726190476190478,
"loss": 5.6079,
"theoretical_loss": 6.4788193288433105,
"tokens_seen": 14352384
},
{
"epoch": 0.0,
"learning_rate": 0.00021825396825396825,
"loss": 5.5548,
"theoretical_loss": 6.473630931141869,
"tokens_seen": 14417920
},
{
"epoch": 0.0,
"learning_rate": 0.00021924603174603176,
"loss": 5.5485,
"theoretical_loss": 6.468472633004308,
"tokens_seen": 14483456
},
{
"epoch": 0.0,
"learning_rate": 0.00022023809523809523,
"loss": 5.6022,
"theoretical_loss": 6.463344124796616,
"tokens_seen": 14548992
},
{
"epoch": 0.0,
"learning_rate": 0.00022123015873015873,
"loss": 5.715,
"theoretical_loss": 6.45824510144643,
"tokens_seen": 14614528
},
{
"epoch": 0.0,
"learning_rate": 0.0002222222222222222,
"loss": 5.67,
"theoretical_loss": 6.45317526235573,
"tokens_seen": 14680064
},
{
"epoch": 0.0,
"objective/train/docs_used": 20428,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.015859127044678,
"objective/train/theoretical_loss": 6.448134311315593,
"objective/train/tokens_used": 35205600,
"theoretical_loss": 6.448134311315593,
"tokens_seen": 14745600
},
{
"epoch": 0.0,
"learning_rate": 0.00022321428571428573,
"loss": 5.3803,
"theoretical_loss": 6.448134311315593,
"tokens_seen": 14745600
},
{
"epoch": 0.0,
"learning_rate": 0.0002242063492063492,
"loss": 5.4398,
"theoretical_loss": 6.443121956422939,
"tokens_seen": 14811136
},
{
"epoch": 0.0,
"learning_rate": 0.0002251984126984127,
"loss": 5.6742,
"theoretical_loss": 6.438137909999214,
"tokens_seen": 14876672
},
{
"epoch": 0.0,
"learning_rate": 0.00022619047619047618,
"loss": 5.7337,
"theoretical_loss": 6.433181888510964,
"tokens_seen": 14942208
},
{
"epoch": 0.0,
"learning_rate": 0.00022718253968253969,
"loss": 5.5962,
"theoretical_loss": 6.428253612492239,
"tokens_seen": 15007744
},
{
"epoch": 0.0,
"learning_rate": 0.00022817460317460316,
"loss": 5.8581,
"theoretical_loss": 6.4233528064687855,
"tokens_seen": 15073280
},
{
"epoch": 0.0,
"learning_rate": 0.00022916666666666666,
"loss": 5.7247,
"theoretical_loss": 6.418479198883969,
"tokens_seen": 15138816
},
{
"epoch": 0.0,
"learning_rate": 0.00023015873015873016,
"loss": 5.6249,
"theoretical_loss": 6.413632522026391,
"tokens_seen": 15204352
},
{
"epoch": 0.0,
"learning_rate": 0.00023115079365079367,
"loss": 5.6726,
"theoretical_loss": 6.40881251195914,
"tokens_seen": 15269888
},
{
"epoch": 0.0,
"learning_rate": 0.00023214285714285717,
"loss": 5.4304,
"theoretical_loss": 6.404018908450656,
"tokens_seen": 15335424
},
{
"epoch": 0.0,
"learning_rate": 0.00023313492063492064,
"loss": 5.4765,
"theoretical_loss": 6.399251454907132,
"tokens_seen": 15400960
},
{
"epoch": 0.0,
"learning_rate": 0.00023412698412698414,
"loss": 5.865,
"theoretical_loss": 6.394509898306452,
"tokens_seen": 15466496
},
{
"epoch": 0.0,
"learning_rate": 0.00023511904761904762,
"loss": 5.7674,
"theoretical_loss": 6.389793989133574,
"tokens_seen": 15532032
},
{
"epoch": 0.0,
"learning_rate": 0.00023611111111111112,
"loss": 5.2223,
"theoretical_loss": 6.385103481317387,
"tokens_seen": 15597568
},
{
"epoch": 0.0,
"learning_rate": 0.0002371031746031746,
"loss": 5.6511,
"theoretical_loss": 6.380438132168923,
"tokens_seen": 15663104
},
{
"epoch": 0.0,
"learning_rate": 0.0002380952380952381,
"loss": 5.4302,
"theoretical_loss": 6.375797702320966,
"tokens_seen": 15728640
},
{
"epoch": 0.0,
"learning_rate": 0.0002390873015873016,
"loss": 5.4526,
"theoretical_loss": 6.371181955668966,
"tokens_seen": 15794176
},
{
"epoch": 0.0,
"learning_rate": 0.0002400793650793651,
"loss": 5.3307,
"theoretical_loss": 6.366590659313248,
"tokens_seen": 15859712
},
{
"epoch": 0.0,
"learning_rate": 0.00024107142857142857,
"loss": 5.4796,
"theoretical_loss": 6.36202358350248,
"tokens_seen": 15925248
},
{
"epoch": 0.0,
"learning_rate": 0.00024206349206349207,
"loss": 5.4457,
"theoretical_loss": 6.357480501578371,
"tokens_seen": 15990784
},
{
"epoch": 0.0,
"learning_rate": 0.00024305555555555555,
"loss": 5.5368,
"theoretical_loss": 6.352961189921553,
"tokens_seen": 16056320
},
{
"epoch": 0.0,
"learning_rate": 0.00024404761904761905,
"loss": 5.3346,
"theoretical_loss": 6.348465427898629,
"tokens_seen": 16121856
},
{
"epoch": 0.0,
"learning_rate": 0.00024503968253968255,
"loss": 5.2158,
"theoretical_loss": 6.343992997810366,
"tokens_seen": 16187392
},
{
"epoch": 0.0,
"learning_rate": 0.000246031746031746,
"loss": 5.5594,
"theoretical_loss": 6.33954368484097,
"tokens_seen": 16252928
},
{
"epoch": 0.0,
"learning_rate": 0.00024702380952380955,
"loss": 5.5983,
"theoretical_loss": 6.33511727700846,
"tokens_seen": 16318464
},
{
"epoch": 0.0,
"objective/train/docs_used": 21641,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.571507930755615,
"objective/train/theoretical_loss": 6.330713565116083,
"objective/train/tokens_used": 36844000,
"theoretical_loss": 6.330713565116083,
"tokens_seen": 16384000
},
{
"epoch": 0.0,
"learning_rate": 0.000248015873015873,
"loss": 5.6484,
"theoretical_loss": 6.330713565116083,
"tokens_seen": 16384000
},
{
"epoch": 0.0,
"learning_rate": 0.0002490079365079365,
"loss": 5.6139,
"theoretical_loss": 6.326332342704751,
"tokens_seen": 16449536
},
{
"epoch": 0.01,
"learning_rate": 0.00025,
"loss": 5.6404,
"theoretical_loss": 6.32197340600647,
"tokens_seen": 16515072
},
{
"epoch": 0.01,
"learning_rate": 0.0002509920634920635,
"loss": 5.5749,
"theoretical_loss": 6.3176365538987636,
"tokens_seen": 16580608
},
{
"epoch": 0.01,
"learning_rate": 0.000251984126984127,
"loss": 5.176,
"theoretical_loss": 6.313321587860021,
"tokens_seen": 16646144
},
{
"epoch": 0.01,
"learning_rate": 0.00025297619047619046,
"loss": 5.4498,
"theoretical_loss": 6.309028311925785,
"tokens_seen": 16711680
},
{
"epoch": 0.01,
"learning_rate": 0.00025396825396825396,
"loss": 5.059,
"theoretical_loss": 6.304756532645939,
"tokens_seen": 16777216
},
{
"epoch": 0.01,
"learning_rate": 0.00025496031746031746,
"loss": 5.6538,
"theoretical_loss": 6.300506059042775,
"tokens_seen": 16842752
},
{
"epoch": 0.01,
"learning_rate": 0.00025595238095238096,
"loss": 5.4578,
"theoretical_loss": 6.296276702569918,
"tokens_seen": 16908288
},
{
"epoch": 0.01,
"learning_rate": 0.0002569444444444444,
"loss": 5.675,
"theoretical_loss": 6.292068277072099,
"tokens_seen": 16973824
},
{
"epoch": 0.01,
"learning_rate": 0.00025793650793650796,
"loss": 5.6443,
"theoretical_loss": 6.28788059874573,
"tokens_seen": 17039360
},
{
"epoch": 0.01,
"learning_rate": 0.00025892857142857146,
"loss": 4.9578,
"theoretical_loss": 6.283713486100297,
"tokens_seen": 17104896
},
{
"epoch": 0.01,
"learning_rate": 0.00025992063492063497,
"loss": 5.6714,
"theoretical_loss": 6.279566759920507,
"tokens_seen": 17170432
},
{
"epoch": 0.01,
"learning_rate": 0.0002609126984126984,
"loss": 5.4797,
"theoretical_loss": 6.275440243229228,
"tokens_seen": 17235968
},
{
"epoch": 0.01,
"learning_rate": 0.0002619047619047619,
"loss": 5.2515,
"theoretical_loss": 6.271333761251142,
"tokens_seen": 17301504
},
{
"epoch": 0.01,
"learning_rate": 0.0002628968253968254,
"loss": 5.4154,
"theoretical_loss": 6.267247141377137,
"tokens_seen": 17367040
},
{
"epoch": 0.01,
"learning_rate": 0.0002638888888888889,
"loss": 5.5372,
"theoretical_loss": 6.2631802131294085,
"tokens_seen": 17432576
},
{
"epoch": 0.01,
"learning_rate": 0.00026488095238095237,
"loss": 5.4489,
"theoretical_loss": 6.259132808127246,
"tokens_seen": 17498112
},
{
"epoch": 0.01,
"learning_rate": 0.00026587301587301587,
"loss": 5.5914,
"theoretical_loss": 6.255104760053497,
"tokens_seen": 17563648
},
{
"epoch": 0.01,
"learning_rate": 0.00026686507936507937,
"loss": 5.1736,
"theoretical_loss": 6.251095904621689,
"tokens_seen": 17629184
},
{
"epoch": 0.01,
"learning_rate": 0.00026785714285714287,
"loss": 5.6947,
"theoretical_loss": 6.247106079543801,
"tokens_seen": 17694720
},
{
"epoch": 0.01,
"learning_rate": 0.0002688492063492063,
"loss": 5.515,
"theoretical_loss": 6.243135124498652,
"tokens_seen": 17760256
},
{
"epoch": 0.01,
"learning_rate": 0.0002698412698412698,
"loss": 5.531,
"theoretical_loss": 6.239182881100916,
"tokens_seen": 17825792
},
{
"epoch": 0.01,
"learning_rate": 0.0002708333333333333,
"loss": 5.2666,
"theoretical_loss": 6.235249192870732,
"tokens_seen": 17891328
},
{
"epoch": 0.01,
"learning_rate": 0.0002718253968253968,
"loss": 5.5261,
"theoretical_loss": 6.231333905203899,
"tokens_seen": 17956864
},
{
"epoch": 0.01,
"objective/train/docs_used": 22253,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.281318664550781,
"objective/train/theoretical_loss": 6.227436865342643,
"objective/train/tokens_used": 38482400,
"theoretical_loss": 6.227436865342643,
"tokens_seen": 18022400
},
{
"epoch": 0.01,
"learning_rate": 0.0002728174603174603,
"loss": 5.2597,
"theoretical_loss": 6.227436865342643,
"tokens_seen": 18022400
},
{
"epoch": 0.01,
"learning_rate": 0.0002738095238095238,
"loss": 5.6137,
"theoretical_loss": 6.223557922346955,
"tokens_seen": 18087936
},
{
"epoch": 0.01,
"learning_rate": 0.0002748015873015873,
"loss": 5.2549,
"theoretical_loss": 6.219696927066456,
"tokens_seen": 18153472
},
{
"epoch": 0.01,
"learning_rate": 0.00027579365079365083,
"loss": 5.3473,
"theoretical_loss": 6.215853732112821,
"tokens_seen": 18219008
},
{
"epoch": 0.01,
"learning_rate": 0.00027678571428571433,
"loss": 5.2604,
"theoretical_loss": 6.212028191832702,
"tokens_seen": 18284544
},
{
"epoch": 0.01,
"learning_rate": 0.0002777777777777778,
"loss": 5.3027,
"theoretical_loss": 6.208220162281178,
"tokens_seen": 18350080
},
{
"epoch": 0.01,
"learning_rate": 0.0002787698412698413,
"loss": 5.2227,
"theoretical_loss": 6.204429501195701,
"tokens_seen": 18415616
},
{
"epoch": 0.01,
"learning_rate": 0.0002797619047619048,
"loss": 5.4059,
"theoretical_loss": 6.20065606797053,
"tokens_seen": 18481152
},
{
"epoch": 0.01,
"learning_rate": 0.0002807539682539683,
"loss": 5.5029,
"theoretical_loss": 6.19689972363164,
"tokens_seen": 18546688
},
{
"epoch": 0.01,
"learning_rate": 0.00028174603174603173,
"loss": 5.3127,
"theoretical_loss": 6.1931603308120975,
"tokens_seen": 18612224
},
{
"epoch": 0.01,
"learning_rate": 0.00028273809523809523,
"loss": 5.2913,
"theoretical_loss": 6.189437753727901,
"tokens_seen": 18677760
},
{
"epoch": 0.01,
"learning_rate": 0.00028373015873015873,
"loss": 5.3535,
"theoretical_loss": 6.185731858154261,
"tokens_seen": 18743296
},
{
"epoch": 0.01,
"learning_rate": 0.00028472222222222223,
"loss": 5.4967,
"theoretical_loss": 6.182042511402313,
"tokens_seen": 18808832
},
{
"epoch": 0.01,
"learning_rate": 0.0002857142857142857,
"loss": 5.325,
"theoretical_loss": 6.17836958229627,
"tokens_seen": 18874368
},
{
"epoch": 0.01,
"learning_rate": 0.0002867063492063492,
"loss": 5.3921,
"theoretical_loss": 6.1747129411509825,
"tokens_seen": 18939904
},
{
"epoch": 0.01,
"learning_rate": 0.0002876984126984127,
"loss": 4.9391,
"theoretical_loss": 6.171072459749913,
"tokens_seen": 19005440
},
{
"epoch": 0.01,
"learning_rate": 0.0002886904761904762,
"loss": 5.3633,
"theoretical_loss": 6.1674480113235095,
"tokens_seen": 19070976
},
{
"epoch": 0.01,
"learning_rate": 0.0002896825396825397,
"loss": 5.2082,
"theoretical_loss": 6.163839470527964,
"tokens_seen": 19136512
},
{
"epoch": 0.01,
"learning_rate": 0.0002906746031746032,
"loss": 5.3773,
"theoretical_loss": 6.160246713424372,
"tokens_seen": 19202048
},
{
"epoch": 0.01,
"learning_rate": 0.0002916666666666667,
"loss": 5.7411,
"theoretical_loss": 6.156669617458243,
"tokens_seen": 19267584
},
{
"epoch": 0.01,
"learning_rate": 0.0002926587301587302,
"loss": 4.9287,
"theoretical_loss": 6.153108061439397,
"tokens_seen": 19333120
},
{
"epoch": 0.01,
"learning_rate": 0.0002936507936507937,
"loss": 5.493,
"theoretical_loss": 6.149561925522211,
"tokens_seen": 19398656
},
{
"epoch": 0.01,
"learning_rate": 0.00029464285714285714,
"loss": 5.3644,
"theoretical_loss": 6.146031091186222,
"tokens_seen": 19464192
},
{
"epoch": 0.01,
"learning_rate": 0.00029563492063492064,
"loss": 5.3852,
"theoretical_loss": 6.142515441217064,
"tokens_seen": 19529728
},
{
"epoch": 0.01,
"learning_rate": 0.00029662698412698414,
"loss": 5.2977,
"theoretical_loss": 6.1390148596877605,
"tokens_seen": 19595264
},
{
"epoch": 0.01,
"objective/train/docs_used": 23612,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.15675163269043,
"objective/train/theoretical_loss": 6.135529231940326,
"objective/train/tokens_used": 40120800,
"theoretical_loss": 6.135529231940326,
"tokens_seen": 19660800
},
{
"epoch": 0.01,
"learning_rate": 0.00029761904761904765,
"loss": 5.137,
"theoretical_loss": 6.135529231940326,
"tokens_seen": 19660800
},
{
"epoch": 0.01,
"learning_rate": 0.0002986111111111111,
"loss": 5.1525,
"theoretical_loss": 6.132058444567705,
"tokens_seen": 19726336
},
{
"epoch": 0.01,
"learning_rate": 0.0002996031746031746,
"loss": 5.0443,
"theoretical_loss": 6.128602385396022,
"tokens_seen": 19791872
},
{
"epoch": 0.01,
"learning_rate": 0.0003005952380952381,
"loss": 5.4228,
"theoretical_loss": 6.125160943467138,
"tokens_seen": 19857408
},
{
"epoch": 0.01,
"learning_rate": 0.0003015873015873016,
"loss": 5.194,
"theoretical_loss": 6.121734009021521,
"tokens_seen": 19922944
},
{
"epoch": 0.01,
"learning_rate": 0.00030257936507936505,
"loss": 5.2838,
"theoretical_loss": 6.118321473481398,
"tokens_seen": 19988480
},
{
"epoch": 0.01,
"learning_rate": 0.00030357142857142855,
"loss": 5.1757,
"theoretical_loss": 6.114923229434213,
"tokens_seen": 20054016
},
{
"epoch": 0.01,
"learning_rate": 0.00030456349206349205,
"loss": 5.5533,
"theoretical_loss": 6.111539170616359,
"tokens_seen": 20119552
},
{
"epoch": 0.01,
"learning_rate": 0.0003055555555555556,
"loss": 5.0377,
"theoretical_loss": 6.108169191897195,
"tokens_seen": 20185088
},
{
"epoch": 0.01,
"learning_rate": 0.00030654761904761905,
"loss": 5.006,
"theoretical_loss": 6.104813189263336,
"tokens_seen": 20250624
},
{
"epoch": 0.01,
"learning_rate": 0.00030753968253968255,
"loss": 5.2312,
"theoretical_loss": 6.101471059803204,
"tokens_seen": 20316160
},
{
"epoch": 0.01,
"learning_rate": 0.00030853174603174605,
"loss": 5.2477,
"theoretical_loss": 6.098142701691856,
"tokens_seen": 20381696
},
{
"epoch": 0.01,
"learning_rate": 0.00030952380952380956,
"loss": 5.1064,
"theoretical_loss": 6.094828014176053,
"tokens_seen": 20447232
},
{
"epoch": 0.01,
"learning_rate": 0.000310515873015873,
"loss": 5.202,
"theoretical_loss": 6.091526897559593,
"tokens_seen": 20512768
},
{
"epoch": 0.01,
"learning_rate": 0.0003115079365079365,
"loss": 5.0853,
"theoretical_loss": 6.088239253188885,
"tokens_seen": 20578304
},
{
"epoch": 0.01,
"learning_rate": 0.0003125,
"loss": 5.3478,
"theoretical_loss": 6.084964983438763,
"tokens_seen": 20643840
},
{
"epoch": 0.01,
"learning_rate": 0.0003134920634920635,
"loss": 5.2918,
"theoretical_loss": 6.0817039916985465,
"tokens_seen": 20709376
},
{
"epoch": 0.01,
"learning_rate": 0.000314484126984127,
"loss": 5.0812,
"theoretical_loss": 6.078456182358325,
"tokens_seen": 20774912
},
{
"epoch": 0.01,
"learning_rate": 0.00031547619047619046,
"loss": 5.1169,
"theoretical_loss": 6.075221460795472,
"tokens_seen": 20840448
},
{
"epoch": 0.01,
"learning_rate": 0.00031646825396825396,
"loss": 5.279,
"theoretical_loss": 6.071999733361386,
"tokens_seen": 20905984
},
{
"epoch": 0.01,
"learning_rate": 0.00031746031746031746,
"loss": 5.3881,
"theoretical_loss": 6.068790907368448,
"tokens_seen": 20971520
},
{
"epoch": 0.01,
"learning_rate": 0.00031845238095238096,
"loss": 5.26,
"theoretical_loss": 6.0655948910771915,
"tokens_seen": 21037056
},
{
"epoch": 0.01,
"learning_rate": 0.0003194444444444444,
"loss": 5.2783,
"theoretical_loss": 6.062411593683687,
"tokens_seen": 21102592
},
{
"epoch": 0.01,
"learning_rate": 0.00032043650793650796,
"loss": 5.062,
"theoretical_loss": 6.059240925307134,
"tokens_seen": 21168128
},
{
"epoch": 0.01,
"learning_rate": 0.00032142857142857147,
"loss": 5.2426,
"theoretical_loss": 6.056082796977648,
"tokens_seen": 21233664
},
{
"epoch": 0.01,
"objective/train/docs_used": 24330,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.736891746520996,
"objective/train/theoretical_loss": 6.052937120624258,
"objective/train/tokens_used": 41759200,
"theoretical_loss": 6.052937120624258,
"tokens_seen": 21299200
},
{
"epoch": 0.01,
"learning_rate": 0.00032242063492063497,
"loss": 5.1217,
"theoretical_loss": 6.052937120624258,
"tokens_seen": 21299200
},
{
"epoch": 0.01,
"learning_rate": 0.0003234126984126984,
"loss": 5.4246,
"theoretical_loss": 6.049803809063083,
"tokens_seen": 21364736
},
{
"epoch": 0.01,
"learning_rate": 0.0003244047619047619,
"loss": 5.4049,
"theoretical_loss": 6.0466827759857145,
"tokens_seen": 21430272
},
{
"epoch": 0.01,
"learning_rate": 0.0003253968253968254,
"loss": 5.1161,
"theoretical_loss": 6.04357393594778,
"tokens_seen": 21495808
},
{
"epoch": 0.01,
"learning_rate": 0.0003263888888888889,
"loss": 5.2259,
"theoretical_loss": 6.040477204357686,
"tokens_seen": 21561344
},
{
"epoch": 0.01,
"learning_rate": 0.00032738095238095237,
"loss": 5.4008,
"theoretical_loss": 6.037392497465552,
"tokens_seen": 21626880
},
{
"epoch": 0.01,
"learning_rate": 0.00032837301587301587,
"loss": 5.0396,
"theoretical_loss": 6.034319732352309,
"tokens_seen": 21692416
},
{
"epoch": 0.01,
"learning_rate": 0.00032936507936507937,
"loss": 5.1953,
"theoretical_loss": 6.031258826918979,
"tokens_seen": 21757952
},
{
"epoch": 0.01,
"learning_rate": 0.00033035714285714287,
"loss": 5.3939,
"theoretical_loss": 6.0282096998761245,
"tokens_seen": 21823488
},
{
"epoch": 0.01,
"learning_rate": 0.0003313492063492063,
"loss": 5.5128,
"theoretical_loss": 6.025172270733464,
"tokens_seen": 21889024
},
{
"epoch": 0.01,
"learning_rate": 0.0003323412698412698,
"loss": 5.1908,
"theoretical_loss": 6.0221464597896475,
"tokens_seen": 21954560
},
{
"epoch": 0.01,
"learning_rate": 0.0003333333333333333,
"loss": 4.7464,
"theoretical_loss": 6.0191321881221995,
"tokens_seen": 22020096
},
{
"epoch": 0.01,
"learning_rate": 0.0003343253968253968,
"loss": 5.2092,
"theoretical_loss": 6.016129377577614,
"tokens_seen": 22085632
},
{
"epoch": 0.01,
"learning_rate": 0.0003353174603174603,
"loss": 5.3966,
"theoretical_loss": 6.01313795076161,
"tokens_seen": 22151168
},
{
"epoch": 0.01,
"learning_rate": 0.0003363095238095238,
"loss": 4.8061,
"theoretical_loss": 6.010157831029533,
"tokens_seen": 22216704
},
{
"epoch": 0.01,
"learning_rate": 0.00033730158730158733,
"loss": 5.0669,
"theoretical_loss": 6.007188942476907,
"tokens_seen": 22282240
},
{
"epoch": 0.01,
"learning_rate": 0.00033829365079365083,
"loss": 5.2141,
"theoretical_loss": 6.0042312099301425,
"tokens_seen": 22347776
},
{
"epoch": 0.01,
"learning_rate": 0.00033928571428571433,
"loss": 5.1484,
"theoretical_loss": 6.001284558937368,
"tokens_seen": 22413312
},
{
"epoch": 0.01,
"learning_rate": 0.0003402777777777778,
"loss": 5.2602,
"theoretical_loss": 5.998348915759426,
"tokens_seen": 22478848
},
{
"epoch": 0.01,
"learning_rate": 0.0003412698412698413,
"loss": 5.1164,
"theoretical_loss": 5.995424207360987,
"tokens_seen": 22544384
},
{
"epoch": 0.01,
"learning_rate": 0.0003422619047619048,
"loss": 5.0515,
"theoretical_loss": 5.992510361401818,
"tokens_seen": 22609920
},
{
"epoch": 0.01,
"learning_rate": 0.0003432539682539683,
"loss": 5.218,
"theoretical_loss": 5.989607306228168,
"tokens_seen": 22675456
},
{
"epoch": 0.01,
"learning_rate": 0.00034424603174603173,
"loss": 4.9102,
"theoretical_loss": 5.986714970864292,
"tokens_seen": 22740992
},
{
"epoch": 0.01,
"learning_rate": 0.00034523809523809523,
"loss": 5.0805,
"theoretical_loss": 5.983833285004112,
"tokens_seen": 22806528
},
{
"epoch": 0.01,
"learning_rate": 0.00034623015873015873,
"loss": 5.2809,
"theoretical_loss": 5.980962179002983,
"tokens_seen": 22872064
},
{
"epoch": 0.01,
"objective/train/docs_used": 24983,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.134525299072266,
"objective/train/theoretical_loss": 5.978101583869607,
"objective/train/tokens_used": 43397600,
"theoretical_loss": 5.978101583869607,
"tokens_seen": 22937600
},
{
"epoch": 0.01,
"learning_rate": 0.00034722222222222224,
"loss": 5.1396,
"theoretical_loss": 5.978101583869607,
"tokens_seen": 22937600
},
{
"epoch": 0.01,
"learning_rate": 0.0003482142857142857,
"loss": 5.0727,
"theoretical_loss": 5.975251431258057,
"tokens_seen": 23003136
},
{
"epoch": 0.01,
"learning_rate": 0.0003492063492063492,
"loss": 5.1882,
"theoretical_loss": 5.972411653459913,
"tokens_seen": 23068672
},
{
"epoch": 0.01,
"learning_rate": 0.0003501984126984127,
"loss": 4.8682,
"theoretical_loss": 5.9695821833965335,
"tokens_seen": 23134208
},
{
"epoch": 0.01,
"learning_rate": 0.0003511904761904762,
"loss": 5.2343,
"theoretical_loss": 5.966762954611432,
"tokens_seen": 23199744
},
{
"epoch": 0.01,
"learning_rate": 0.0003521825396825397,
"loss": 5.195,
"theoretical_loss": 5.963953901262764,
"tokens_seen": 23265280
},
{
"epoch": 0.01,
"learning_rate": 0.0003531746031746032,
"loss": 5.2282,
"theoretical_loss": 5.961154958115937,
"tokens_seen": 23330816
},
{
"epoch": 0.01,
"learning_rate": 0.0003541666666666667,
"loss": 4.7545,
"theoretical_loss": 5.958366060536315,
"tokens_seen": 23396352
},
{
"epoch": 0.01,
"learning_rate": 0.0003551587301587302,
"loss": 5.1342,
"theoretical_loss": 5.955587144482044,
"tokens_seen": 23461888
},
{
"epoch": 0.01,
"learning_rate": 0.0003561507936507937,
"loss": 5.0144,
"theoretical_loss": 5.952818146496978,
"tokens_seen": 23527424
},
{
"epoch": 0.01,
"learning_rate": 0.00035714285714285714,
"loss": 5.0659,
"theoretical_loss": 5.950059003703704,
"tokens_seen": 23592960
},
{
"epoch": 0.01,
"learning_rate": 0.00035813492063492064,
"loss": 5.1206,
"theoretical_loss": 5.94730965379668,
"tokens_seen": 23658496
},
{
"epoch": 0.01,
"learning_rate": 0.00035912698412698415,
"loss": 4.9633,
"theoretical_loss": 5.944570035035458,
"tokens_seen": 23724032
},
{
"epoch": 0.01,
"learning_rate": 0.00036011904761904765,
"loss": 5.1591,
"theoretical_loss": 5.941840086238027,
"tokens_seen": 23789568
},
{
"epoch": 0.01,
"learning_rate": 0.0003611111111111111,
"loss": 5.3314,
"theoretical_loss": 5.939119746774228,
"tokens_seen": 23855104
},
{
"epoch": 0.01,
"learning_rate": 0.0003621031746031746,
"loss": 5.2119,
"theoretical_loss": 5.936408956559284,
"tokens_seen": 23920640
},
{
"epoch": 0.01,
"learning_rate": 0.0003630952380952381,
"loss": 5.3978,
"theoretical_loss": 5.933707656047414,
"tokens_seen": 23986176
},
{
"epoch": 0.01,
"learning_rate": 0.0003640873015873016,
"loss": 4.9459,
"theoretical_loss": 5.93101578622554,
"tokens_seen": 24051712
},
{
"epoch": 0.01,
"learning_rate": 0.00036507936507936505,
"loss": 5.031,
"theoretical_loss": 5.928333288607086,
"tokens_seen": 24117248
},
{
"epoch": 0.01,
"learning_rate": 0.00036607142857142855,
"loss": 5.1772,
"theoretical_loss": 5.925660105225867,
"tokens_seen": 24182784
},
{
"epoch": 0.01,
"learning_rate": 0.00036706349206349205,
"loss": 4.8824,
"theoretical_loss": 5.92299617863006,
"tokens_seen": 24248320
},
{
"epoch": 0.01,
"learning_rate": 0.0003680555555555556,
"loss": 5.0966,
"theoretical_loss": 5.920341451876267,
"tokens_seen": 24313856
},
{
"epoch": 0.01,
"learning_rate": 0.00036904761904761905,
"loss": 5.1533,
"theoretical_loss": 5.9176958685236585,
"tokens_seen": 24379392
},
{
"epoch": 0.01,
"learning_rate": 0.00037003968253968255,
"loss": 5.2939,
"theoretical_loss": 5.9150593726282015,
"tokens_seen": 24444928
},
{
"epoch": 0.01,
"learning_rate": 0.00037103174603174606,
"loss": 5.0167,
"theoretical_loss": 5.912431908736972,
"tokens_seen": 24510464
},
{
"epoch": 0.01,
"objective/train/docs_used": 26439,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.887441635131836,
"objective/train/theoretical_loss": 5.909813421882534,
"objective/train/tokens_used": 45036000,
"theoretical_loss": 5.909813421882534,
"tokens_seen": 24576000
},
{
"epoch": 0.01,
"learning_rate": 0.00037202380952380956,
"loss": 5.2049,
"theoretical_loss": 5.909813421882534,
"tokens_seen": 24576000
},
{
"epoch": 0.01,
"learning_rate": 0.000373015873015873,
"loss": 5.2346,
"theoretical_loss": 5.907203857577422,
"tokens_seen": 24641536
},
{
"epoch": 0.01,
"learning_rate": 0.0003740079365079365,
"loss": 5.2112,
"theoretical_loss": 5.9046031618086765,
"tokens_seen": 24707072
},
{
"epoch": 0.01,
"learning_rate": 0.000375,
"loss": 5.092,
"theoretical_loss": 5.902011281032472,
"tokens_seen": 24772608
},
{
"epoch": 0.01,
"learning_rate": 0.0003759920634920635,
"loss": 5.2228,
"theoretical_loss": 5.899428162168808,
"tokens_seen": 24838144
},
{
"epoch": 0.01,
"learning_rate": 0.000376984126984127,
"loss": 4.9695,
"theoretical_loss": 5.896853752596286,
"tokens_seen": 24903680
},
{
"epoch": 0.01,
"learning_rate": 0.00037797619047619046,
"loss": 5.1379,
"theoretical_loss": 5.894288000146949,
"tokens_seen": 24969216
},
{
"epoch": 0.01,
"learning_rate": 0.00037896825396825396,
"loss": 5.0639,
"theoretical_loss": 5.891730853101199,
"tokens_seen": 25034752
},
{
"epoch": 0.01,
"learning_rate": 0.00037996031746031746,
"loss": 5.0119,
"theoretical_loss": 5.88918226018278,
"tokens_seen": 25100288
},
{
"epoch": 0.01,
"learning_rate": 0.00038095238095238096,
"loss": 5.0625,
"theoretical_loss": 5.8866421705538325,
"tokens_seen": 25165824
},
{
"epoch": 0.01,
"learning_rate": 0.0003819444444444444,
"loss": 5.2315,
"theoretical_loss": 5.8841105338100155,
"tokens_seen": 25231360
},
{
"epoch": 0.01,
"learning_rate": 0.00038293650793650797,
"loss": 4.8079,
"theoretical_loss": 5.881587299975694,
"tokens_seen": 25296896
},
{
"epoch": 0.01,
"learning_rate": 0.00038392857142857147,
"loss": 5.3085,
"theoretical_loss": 5.8790724194991935,
"tokens_seen": 25362432
},
{
"epoch": 0.01,
"learning_rate": 0.00038492063492063497,
"loss": 5.1826,
"theoretical_loss": 5.876565843248124,
"tokens_seen": 25427968
},
{
"epoch": 0.01,
"learning_rate": 0.0003859126984126984,
"loss": 4.8865,
"theoretical_loss": 5.8740675225047525,
"tokens_seen": 25493504
},
{
"epoch": 0.01,
"learning_rate": 0.0003869047619047619,
"loss": 4.8372,
"theoretical_loss": 5.871577408961457,
"tokens_seen": 25559040
},
{
"epoch": 0.01,
"learning_rate": 0.0003878968253968254,
"loss": 5.1074,
"theoretical_loss": 5.869095454716231,
"tokens_seen": 25624576
},
{
"epoch": 0.01,
"learning_rate": 0.0003888888888888889,
"loss": 5.0627,
"theoretical_loss": 5.866621612268246,
"tokens_seen": 25690112
},
{
"epoch": 0.01,
"learning_rate": 0.00038988095238095237,
"loss": 5.1708,
"theoretical_loss": 5.864155834513486,
"tokens_seen": 25755648
},
{
"epoch": 0.01,
"learning_rate": 0.00039087301587301587,
"loss": 4.9657,
"theoretical_loss": 5.8616980747404295,
"tokens_seen": 25821184
},
{
"epoch": 0.01,
"learning_rate": 0.00039186507936507937,
"loss": 5.1406,
"theoretical_loss": 5.859248286625787,
"tokens_seen": 25886720
},
{
"epoch": 0.01,
"learning_rate": 0.0003928571428571429,
"loss": 5.253,
"theoretical_loss": 5.856806424230314,
"tokens_seen": 25952256
},
{
"epoch": 0.01,
"learning_rate": 0.0003938492063492063,
"loss": 4.9518,
"theoretical_loss": 5.854372441994654,
"tokens_seen": 26017792
},
{
"epoch": 0.01,
"learning_rate": 0.0003948412698412698,
"loss": 4.9569,
"theoretical_loss": 5.851946294735258,
"tokens_seen": 26083328
},
{
"epoch": 0.01,
"learning_rate": 0.0003958333333333333,
"loss": 4.9376,
"theoretical_loss": 5.849527937640345,
"tokens_seen": 26148864
},
{
"epoch": 0.01,
"objective/train/docs_used": 27060,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.355391025543213,
"objective/train/theoretical_loss": 5.8471173262659235,
"objective/train/tokens_used": 46674400,
"theoretical_loss": 5.8471173262659235,
"tokens_seen": 26214400
},
{
"epoch": 0.01,
"learning_rate": 0.0003968253968253968,
"loss": 4.8582,
"theoretical_loss": 5.8471173262659235,
"tokens_seen": 26214400
},
{
"epoch": 0.01,
"learning_rate": 0.0003978174603174603,
"loss": 4.9394,
"theoretical_loss": 5.84471441653186,
"tokens_seen": 26279936
},
{
"epoch": 0.01,
"learning_rate": 0.00039880952380952383,
"loss": 5.2367,
"theoretical_loss": 5.842319164718004,
"tokens_seen": 26345472
},
{
"epoch": 0.01,
"learning_rate": 0.00039980158730158733,
"loss": 5.1756,
"theoretical_loss": 5.83993152746036,
"tokens_seen": 26411008
},
{
"epoch": 0.01,
"learning_rate": 0.00040079365079365083,
"loss": 5.1252,
"theoretical_loss": 5.83755146174731,
"tokens_seen": 26476544
},
{
"epoch": 0.01,
"learning_rate": 0.00040178571428571433,
"loss": 5.0547,
"theoretical_loss": 5.835178924915889,
"tokens_seen": 26542080
},
{
"epoch": 0.01,
"learning_rate": 0.0004027777777777778,
"loss": 5.0688,
"theoretical_loss": 5.832813874648102,
"tokens_seen": 26607616
},
{
"epoch": 0.01,
"learning_rate": 0.0004037698412698413,
"loss": 5.0608,
"theoretical_loss": 5.8304562689673,
"tokens_seen": 26673152
},
{
"epoch": 0.01,
"learning_rate": 0.0004047619047619048,
"loss": 5.0401,
"theoretical_loss": 5.828106066234588,
"tokens_seen": 26738688
},
{
"epoch": 0.01,
"learning_rate": 0.0004057539682539683,
"loss": 5.0167,
"theoretical_loss": 5.825763225145295,
"tokens_seen": 26804224
},
{
"epoch": 0.01,
"learning_rate": 0.00040674603174603173,
"loss": 5.1791,
"theoretical_loss": 5.823427704725473,
"tokens_seen": 26869760
},
{
"epoch": 0.01,
"learning_rate": 0.00040773809523809523,
"loss": 4.8063,
"theoretical_loss": 5.82109946432846,
"tokens_seen": 26935296
},
{
"epoch": 0.01,
"learning_rate": 0.00040873015873015874,
"loss": 4.9278,
"theoretical_loss": 5.818778463631473,
"tokens_seen": 27000832
},
{
"epoch": 0.01,
"learning_rate": 0.00040972222222222224,
"loss": 4.7786,
"theoretical_loss": 5.816464662632243,
"tokens_seen": 27066368
},
{
"epoch": 0.01,
"learning_rate": 0.0004107142857142857,
"loss": 5.1196,
"theoretical_loss": 5.8141580216457065,
"tokens_seen": 27131904
},
{
"epoch": 0.01,
"learning_rate": 0.0004117063492063492,
"loss": 4.6589,
"theoretical_loss": 5.811858501300729,
"tokens_seen": 27197440
},
{
"epoch": 0.01,
"learning_rate": 0.0004126984126984127,
"loss": 4.8227,
"theoretical_loss": 5.809566062536868,
"tokens_seen": 27262976
},
{
"epoch": 0.01,
"learning_rate": 0.0004136904761904762,
"loss": 5.0245,
"theoretical_loss": 5.807280666601191,
"tokens_seen": 27328512
},
{
"epoch": 0.01,
"learning_rate": 0.0004146825396825397,
"loss": 4.9783,
"theoretical_loss": 5.805002275045111,
"tokens_seen": 27394048
},
{
"epoch": 0.01,
"learning_rate": 0.0004156746031746032,
"loss": 5.0457,
"theoretical_loss": 5.8027308497212875,
"tokens_seen": 27459584
},
{
"epoch": 0.01,
"learning_rate": 0.0004166666666666667,
"loss": 5.002,
"theoretical_loss": 5.800466352780546,
"tokens_seen": 27525120
},
{
"epoch": 0.01,
"learning_rate": 0.0004176587301587302,
"loss": 4.7979,
"theoretical_loss": 5.798208746668847,
"tokens_seen": 27590656
},
{
"epoch": 0.01,
"learning_rate": 0.0004186507936507937,
"loss": 5.0751,
"theoretical_loss": 5.795957994124291,
"tokens_seen": 27656192
},
{
"epoch": 0.01,
"learning_rate": 0.00041964285714285714,
"loss": 4.5446,
"theoretical_loss": 5.7937140581741575,
"tokens_seen": 27721728
},
{
"epoch": 0.01,
"learning_rate": 0.00042063492063492065,
"loss": 4.9859,
"theoretical_loss": 5.791476902131985,
"tokens_seen": 27787264
},
{
"epoch": 0.01,
"objective/train/docs_used": 28453,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.114999294281006,
"objective/train/theoretical_loss": 5.789246489594688,
"objective/train/tokens_used": 48312800,
"theoretical_loss": 5.789246489594688,
"tokens_seen": 27852800
},
{
"epoch": 0.01,
"learning_rate": 0.00042162698412698415,
"loss": 4.7322,
"theoretical_loss": 5.789246489594688,
"tokens_seen": 27852800
},
{
"epoch": 0.01,
"learning_rate": 0.00042261904761904765,
"loss": 4.9798,
"theoretical_loss": 5.787022784439701,
"tokens_seen": 27918336
},
{
"epoch": 0.01,
"learning_rate": 0.0004236111111111111,
"loss": 5.1612,
"theoretical_loss": 5.784805750822171,
"tokens_seen": 27983872
},
{
"epoch": 0.01,
"learning_rate": 0.0004246031746031746,
"loss": 5.0818,
"theoretical_loss": 5.782595353172176,
"tokens_seen": 28049408
},
{
"epoch": 0.01,
"learning_rate": 0.0004255952380952381,
"loss": 4.5939,
"theoretical_loss": 5.780391556191977,
"tokens_seen": 28114944
},
{
"epoch": 0.01,
"learning_rate": 0.0004265873015873016,
"loss": 4.9998,
"theoretical_loss": 5.778194324853311,
"tokens_seen": 28180480
},
{
"epoch": 0.01,
"learning_rate": 0.00042757936507936505,
"loss": 5.1566,
"theoretical_loss": 5.776003624394711,
"tokens_seen": 28246016
},
{
"epoch": 0.01,
"learning_rate": 0.00042857142857142855,
"loss": 5.2049,
"theoretical_loss": 5.773819420318858,
"tokens_seen": 28311552
},
{
"epoch": 0.01,
"learning_rate": 0.00042956349206349205,
"loss": 4.7744,
"theoretical_loss": 5.771641678389971,
"tokens_seen": 28377088
},
{
"epoch": 0.01,
"learning_rate": 0.0004305555555555556,
"loss": 4.8862,
"theoretical_loss": 5.769470364631225,
"tokens_seen": 28442624
},
{
"epoch": 0.01,
"learning_rate": 0.00043154761904761905,
"loss": 4.7604,
"theoretical_loss": 5.767305445322201,
"tokens_seen": 28508160
},
{
"epoch": 0.01,
"learning_rate": 0.00043253968253968256,
"loss": 4.9808,
"theoretical_loss": 5.765146886996363,
"tokens_seen": 28573696
},
{
"epoch": 0.01,
"learning_rate": 0.00043353174603174606,
"loss": 4.7763,
"theoretical_loss": 5.762994656438579,
"tokens_seen": 28639232
},
{
"epoch": 0.01,
"learning_rate": 0.00043452380952380956,
"loss": 5.0911,
"theoretical_loss": 5.760848720682651,
"tokens_seen": 28704768
},
{
"epoch": 0.01,
"learning_rate": 0.000435515873015873,
"loss": 5.0289,
"theoretical_loss": 5.758709047008894,
"tokens_seen": 28770304
},
{
"epoch": 0.01,
"learning_rate": 0.0004365079365079365,
"loss": 5.0362,
"theoretical_loss": 5.756575602941732,
"tokens_seen": 28835840
},
{
"epoch": 0.01,
"learning_rate": 0.0004375,
"loss": 5.1589,
"theoretical_loss": 5.75444835624733,
"tokens_seen": 28901376
},
{
"epoch": 0.01,
"learning_rate": 0.0004384920634920635,
"loss": 5.1524,
"theoretical_loss": 5.752327274931249,
"tokens_seen": 28966912
},
{
"epoch": 0.01,
"learning_rate": 0.000439484126984127,
"loss": 4.6286,
"theoretical_loss": 5.750212327236129,
"tokens_seen": 29032448
},
{
"epoch": 0.01,
"learning_rate": 0.00044047619047619046,
"loss": 4.9918,
"theoretical_loss": 5.7481034816394105,
"tokens_seen": 29097984
},
{
"epoch": 0.01,
"learning_rate": 0.00044146825396825396,
"loss": 5.0182,
"theoretical_loss": 5.7460007068510635,
"tokens_seen": 29163520
},
{
"epoch": 0.01,
"learning_rate": 0.00044246031746031746,
"loss": 5.1442,
"theoretical_loss": 5.74390397181136,
"tokens_seen": 29229056
},
{
"epoch": 0.01,
"learning_rate": 0.00044345238095238096,
"loss": 4.9284,
"theoretical_loss": 5.741813245688668,
"tokens_seen": 29294592
},
{
"epoch": 0.01,
"learning_rate": 0.0004444444444444444,
"loss": 4.7274,
"theoretical_loss": 5.739728497877267,
"tokens_seen": 29360128
},
{
"epoch": 0.01,
"learning_rate": 0.00044543650793650797,
"loss": 4.9829,
"theoretical_loss": 5.737649697995197,
"tokens_seen": 29425664
},
{
"epoch": 0.01,
"objective/train/docs_used": 29115,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.875042915344238,
"objective/train/theoretical_loss": 5.7355768158821245,
"objective/train/tokens_used": 49951200,
"theoretical_loss": 5.7355768158821245,
"tokens_seen": 29491200
},
{
"epoch": 0.01,
"learning_rate": 0.00044642857142857147,
"loss": 5.0802,
"theoretical_loss": 5.7355768158821245,
"tokens_seen": 29491200
},
{
"epoch": 0.01,
"learning_rate": 0.00044742063492063497,
"loss": 4.8054,
"theoretical_loss": 5.73350982159724,
"tokens_seen": 29556736
},
{
"epoch": 0.01,
"learning_rate": 0.0004484126984126984,
"loss": 4.9775,
"theoretical_loss": 5.731448685417178,
"tokens_seen": 29622272
},
{
"epoch": 0.01,
"learning_rate": 0.0004494047619047619,
"loss": 4.9123,
"theoretical_loss": 5.729393377833956,
"tokens_seen": 29687808
},
{
"epoch": 0.01,
"learning_rate": 0.0004503968253968254,
"loss": 4.5612,
"theoretical_loss": 5.7273438695529535,
"tokens_seen": 29753344
},
{
"epoch": 0.01,
"learning_rate": 0.0004513888888888889,
"loss": 5.0027,
"theoretical_loss": 5.725300131490888,
"tokens_seen": 29818880
},
{
"epoch": 0.01,
"learning_rate": 0.00045238095238095237,
"loss": 4.9368,
"theoretical_loss": 5.7232621347738455,
"tokens_seen": 29884416
},
{
"epoch": 0.01,
"learning_rate": 0.00045337301587301587,
"loss": 5.0227,
"theoretical_loss": 5.721229850735305,
"tokens_seen": 29949952
},
{
"epoch": 0.01,
"learning_rate": 0.00045436507936507937,
"loss": 5.0473,
"theoretical_loss": 5.719203250914208,
"tokens_seen": 30015488
},
{
"epoch": 0.01,
"learning_rate": 0.0004553571428571429,
"loss": 5.0973,
"theoretical_loss": 5.717182307053037,
"tokens_seen": 30081024
},
{
"epoch": 0.01,
"learning_rate": 0.0004563492063492063,
"loss": 4.757,
"theoretical_loss": 5.715166991095922,
"tokens_seen": 30146560
},
{
"epoch": 0.01,
"learning_rate": 0.0004573412698412698,
"loss": 4.8415,
"theoretical_loss": 5.713157275186761,
"tokens_seen": 30212096
},
{
"epoch": 0.01,
"learning_rate": 0.0004583333333333333,
"loss": 4.9583,
"theoretical_loss": 5.71115313166738,
"tokens_seen": 30277632
},
{
"epoch": 0.01,
"learning_rate": 0.0004593253968253968,
"loss": 5.1126,
"theoretical_loss": 5.709154533075688,
"tokens_seen": 30343168
},
{
"epoch": 0.01,
"learning_rate": 0.00046031746031746033,
"loss": 4.9564,
"theoretical_loss": 5.707161452143879,
"tokens_seen": 30408704
},
{
"epoch": 0.01,
"learning_rate": 0.00046130952380952383,
"loss": 5.0017,
"theoretical_loss": 5.7051738617966326,
"tokens_seen": 30474240
},
{
"epoch": 0.01,
"learning_rate": 0.00046230158730158733,
"loss": 5.1598,
"theoretical_loss": 5.7031917351493515,
"tokens_seen": 30539776
},
{
"epoch": 0.01,
"learning_rate": 0.00046329365079365083,
"loss": 4.7815,
"theoretical_loss": 5.701215045506411,
"tokens_seen": 30605312
},
{
"epoch": 0.01,
"learning_rate": 0.00046428571428571433,
"loss": 5.0404,
"theoretical_loss": 5.699243766359421,
"tokens_seen": 30670848
},
{
"epoch": 0.01,
"learning_rate": 0.0004652777777777778,
"loss": 4.958,
"theoretical_loss": 5.697277871385534,
"tokens_seen": 30736384
},
{
"epoch": 0.01,
"learning_rate": 0.0004662698412698413,
"loss": 4.8348,
"theoretical_loss": 5.695317334445736,
"tokens_seen": 30801920
},
{
"epoch": 0.01,
"learning_rate": 0.0004672619047619048,
"loss": 5.1101,
"theoretical_loss": 5.693362129583184,
"tokens_seen": 30867456
},
{
"epoch": 0.01,
"learning_rate": 0.0004682539682539683,
"loss": 5.1583,
"theoretical_loss": 5.691412231021549,
"tokens_seen": 30932992
},
{
"epoch": 0.01,
"learning_rate": 0.00046924603174603173,
"loss": 5.0497,
"theoretical_loss": 5.689467613163388,
"tokens_seen": 30998528
},
{
"epoch": 0.01,
"learning_rate": 0.00047023809523809523,
"loss": 5.0994,
"theoretical_loss": 5.687528250588518,
"tokens_seen": 31064064
},
{
"epoch": 0.01,
"objective/train/docs_used": 30152,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.741349697113037,
"objective/train/theoretical_loss": 5.6855941180524265,
"objective/train/tokens_used": 51589600,
"theoretical_loss": 5.6855941180524265,
"tokens_seen": 31129600
},
{
"epoch": 0.01,
"learning_rate": 0.00047123015873015874,
"loss": 4.9423,
"theoretical_loss": 5.6855941180524265,
"tokens_seen": 31129600
},
{
"epoch": 0.01,
"learning_rate": 0.00047222222222222224,
"loss": 4.9094,
"theoretical_loss": 5.683665190484683,
"tokens_seen": 31195136
},
{
"epoch": 0.01,
"learning_rate": 0.0004732142857142857,
"loss": 5.0356,
"theoretical_loss": 5.681741442987381,
"tokens_seen": 31260672
},
{
"epoch": 0.01,
"learning_rate": 0.0004742063492063492,
"loss": 4.8249,
"theoretical_loss": 5.679822850833591,
"tokens_seen": 31326208
},
{
"epoch": 0.01,
"learning_rate": 0.0004751984126984127,
"loss": 4.9261,
"theoretical_loss": 5.677909389465831,
"tokens_seen": 31391744
},
{
"epoch": 0.01,
"learning_rate": 0.0004761904761904762,
"loss": 4.5377,
"theoretical_loss": 5.676001034494554,
"tokens_seen": 31457280
},
{
"epoch": 0.01,
"learning_rate": 0.0004771825396825397,
"loss": 4.464,
"theoretical_loss": 5.674097761696653,
"tokens_seen": 31522816
},
{
"epoch": 0.01,
"learning_rate": 0.0004781746031746032,
"loss": 4.6988,
"theoretical_loss": 5.672199547013983,
"tokens_seen": 31588352
},
{
"epoch": 0.01,
"learning_rate": 0.0004791666666666667,
"loss": 4.6733,
"theoretical_loss": 5.670306366551898,
"tokens_seen": 31653888
},
{
"epoch": 0.01,
"learning_rate": 0.0004801587301587302,
"loss": 4.7771,
"theoretical_loss": 5.6684181965778,
"tokens_seen": 31719424
},
{
"epoch": 0.01,
"learning_rate": 0.0004811507936507937,
"loss": 5.0062,
"theoretical_loss": 5.666535013519715,
"tokens_seen": 31784960
},
{
"epoch": 0.01,
"learning_rate": 0.00048214285714285715,
"loss": 4.7684,
"theoretical_loss": 5.6646567939648715,
"tokens_seen": 31850496
},
{
"epoch": 0.01,
"learning_rate": 0.00048313492063492065,
"loss": 4.8464,
"theoretical_loss": 5.6627835146583045,
"tokens_seen": 31916032
},
{
"epoch": 0.01,
"learning_rate": 0.00048412698412698415,
"loss": 4.689,
"theoretical_loss": 5.660915152501465,
"tokens_seen": 31981568
},
{
"epoch": 0.01,
"learning_rate": 0.00048511904761904765,
"loss": 4.5762,
"theoretical_loss": 5.659051684550857,
"tokens_seen": 32047104
},
{
"epoch": 0.01,
"learning_rate": 0.0004861111111111111,
"loss": 4.8833,
"theoretical_loss": 5.657193088016677,
"tokens_seen": 32112640
},
{
"epoch": 0.01,
"learning_rate": 0.0004871031746031746,
"loss": 4.6286,
"theoretical_loss": 5.655339340261474,
"tokens_seen": 32178176
},
{
"epoch": 0.01,
"learning_rate": 0.0004880952380952381,
"loss": 4.791,
"theoretical_loss": 5.653490418798825,
"tokens_seen": 32243712
},
{
"epoch": 0.01,
"learning_rate": 0.0004890873015873016,
"loss": 5.0112,
"theoretical_loss": 5.651646301292022,
"tokens_seen": 32309248
},
{
"epoch": 0.01,
"learning_rate": 0.0004900793650793651,
"loss": 4.7112,
"theoretical_loss": 5.649806965552774,
"tokens_seen": 32374784
},
{
"epoch": 0.01,
"learning_rate": 0.0004910714285714286,
"loss": 4.7717,
"theoretical_loss": 5.6479723895399205,
"tokens_seen": 32440320
},
{
"epoch": 0.01,
"learning_rate": 0.000492063492063492,
"loss": 4.7971,
"theoretical_loss": 5.6461425513581665,
"tokens_seen": 32505856
},
{
"epoch": 0.01,
"learning_rate": 0.0004930555555555556,
"loss": 4.9546,
"theoretical_loss": 5.6443174292568195,
"tokens_seen": 32571392
},
{
"epoch": 0.01,
"learning_rate": 0.0004940476190476191,
"loss": 4.6853,
"theoretical_loss": 5.6424970016285485,
"tokens_seen": 32636928
},
{
"epoch": 0.01,
"learning_rate": 0.0004950396825396826,
"loss": 4.8721,
"theoretical_loss": 5.640681247008156,
"tokens_seen": 32702464
},
{
"epoch": 0.01,
"objective/train/docs_used": 30758,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.8071393966674805,
"objective/train/theoretical_loss": 5.638870144071353,
"objective/train/tokens_used": 53228000,
"theoretical_loss": 5.638870144071353,
"tokens_seen": 32768000
},
{
"epoch": 0.01,
"learning_rate": 0.000496031746031746,
"loss": 4.8628,
"theoretical_loss": 5.638870144071353,
"tokens_seen": 32768000
},
{
"epoch": 0.01,
"learning_rate": 0.0004970238095238095,
"loss": 4.7892,
"theoretical_loss": 5.637063671633564,
"tokens_seen": 32833536
},
{
"epoch": 0.01,
"learning_rate": 0.000498015873015873,
"loss": 4.8607,
"theoretical_loss": 5.635261808648728,
"tokens_seen": 32899072
},
{
"epoch": 0.01,
"learning_rate": 0.0004990079365079365,
"loss": 5.0888,
"theoretical_loss": 5.6334645342081195,
"tokens_seen": 32964608
},
{
"epoch": 0.01,
"learning_rate": 0.0005,
"loss": 4.9737,
"theoretical_loss": 5.631671827539186,
"tokens_seen": 33030144
},
{
"epoch": 0.01,
"learning_rate": 0.0004999899699097292,
"loss": 5.0126,
"theoretical_loss": 5.629883668004389,
"tokens_seen": 33095680
},
{
"epoch": 0.01,
"learning_rate": 0.0004999799398194584,
"loss": 4.9297,
"theoretical_loss": 5.628100035100061,
"tokens_seen": 33161216
},
{
"epoch": 0.01,
"learning_rate": 0.0004999699097291876,
"loss": 5.0938,
"theoretical_loss": 5.626320908455279,
"tokens_seen": 33226752
},
{
"epoch": 0.01,
"learning_rate": 0.0004999598796389167,
"loss": 4.758,
"theoretical_loss": 5.6245462678307385,
"tokens_seen": 33292288
},
{
"epoch": 0.01,
"learning_rate": 0.000499949849548646,
"loss": 4.6107,
"theoretical_loss": 5.622776093117652,
"tokens_seen": 33357824
},
{
"epoch": 0.01,
"learning_rate": 0.0004999398194583751,
"loss": 4.9094,
"theoretical_loss": 5.621010364336651,
"tokens_seen": 33423360
},
{
"epoch": 0.01,
"learning_rate": 0.0004999297893681044,
"loss": 4.7434,
"theoretical_loss": 5.619249061636698,
"tokens_seen": 33488896
},
{
"epoch": 0.01,
"learning_rate": 0.0004999197592778335,
"loss": 4.7376,
"theoretical_loss": 5.61749216529402,
"tokens_seen": 33554432
},
{
"epoch": 0.01,
"learning_rate": 0.0004999097291875627,
"loss": 4.8733,
"theoretical_loss": 5.615739655711037,
"tokens_seen": 33619968
},
{
"epoch": 0.01,
"learning_rate": 0.0004998996990972919,
"loss": 4.8815,
"theoretical_loss": 5.61399151341532,
"tokens_seen": 33685504
},
{
"epoch": 0.01,
"learning_rate": 0.0004998896690070211,
"loss": 4.8804,
"theoretical_loss": 5.6122477190585425,
"tokens_seen": 33751040
},
{
"epoch": 0.01,
"learning_rate": 0.0004998796389167503,
"loss": 4.6509,
"theoretical_loss": 5.610508253415453,
"tokens_seen": 33816576
},
{
"epoch": 0.01,
"learning_rate": 0.0004998696088264795,
"loss": 5.0255,
"theoretical_loss": 5.6087730973828585,
"tokens_seen": 33882112
},
{
"epoch": 0.01,
"learning_rate": 0.0004998595787362087,
"loss": 4.7137,
"theoretical_loss": 5.6070422319786095,
"tokens_seen": 33947648
},
{
"epoch": 0.01,
"learning_rate": 0.0004998495486459378,
"loss": 4.8675,
"theoretical_loss": 5.605315638340606,
"tokens_seen": 34013184
},
{
"epoch": 0.01,
"learning_rate": 0.000499839518555667,
"loss": 4.8488,
"theoretical_loss": 5.603593297725807,
"tokens_seen": 34078720
},
{
"epoch": 0.01,
"learning_rate": 0.0004998294884653962,
"loss": 5.0848,
"theoretical_loss": 5.601875191509249,
"tokens_seen": 34144256
},
{
"epoch": 0.01,
"learning_rate": 0.0004998194583751254,
"loss": 4.4149,
"theoretical_loss": 5.600161301183084,
"tokens_seen": 34209792
},
{
"epoch": 0.01,
"learning_rate": 0.0004998094282848546,
"loss": 5.0088,
"theoretical_loss": 5.598451608355614,
"tokens_seen": 34275328
},
{
"epoch": 0.01,
"learning_rate": 0.0004997993981945837,
"loss": 4.9221,
"theoretical_loss": 5.596746094750342,
"tokens_seen": 34340864
},
{
"epoch": 0.01,
"objective/train/docs_used": 31355,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 5.054786205291748,
"objective/train/theoretical_loss": 5.595044742205037,
"objective/train/tokens_used": 54866400,
"theoretical_loss": 5.595044742205037,
"tokens_seen": 34406400
},
{
"epoch": 0.01,
"learning_rate": 0.0004997893681043129,
"loss": 5.0983,
"theoretical_loss": 5.595044742205037,
"tokens_seen": 34406400
},
{
"epoch": 0.01,
"learning_rate": 0.0004997793380140421,
"loss": 4.5808,
"theoretical_loss": 5.5933475326707995,
"tokens_seen": 34471936
},
{
"epoch": 0.01,
"learning_rate": 0.0004997693079237714,
"loss": 4.5657,
"theoretical_loss": 5.591654448211143,
"tokens_seen": 34537472
},
{
"epoch": 0.01,
"learning_rate": 0.0004997592778335005,
"loss": 4.8711,
"theoretical_loss": 5.589965471001077,
"tokens_seen": 34603008
},
{
"epoch": 0.01,
"learning_rate": 0.0004997492477432298,
"loss": 4.8155,
"theoretical_loss": 5.5882805833262115,
"tokens_seen": 34668544
},
{
"epoch": 0.01,
"learning_rate": 0.0004997392176529588,
"loss": 4.9002,
"theoretical_loss": 5.586599767581859,
"tokens_seen": 34734080
},
{
"epoch": 0.01,
"learning_rate": 0.0004997291875626881,
"loss": 4.9815,
"theoretical_loss": 5.584923006272151,
"tokens_seen": 34799616
},
{
"epoch": 0.01,
"learning_rate": 0.0004997191574724173,
"loss": 4.9042,
"theoretical_loss": 5.583250282009159,
"tokens_seen": 34865152
},
{
"epoch": 0.01,
"learning_rate": 0.0004997091273821465,
"loss": 4.9647,
"theoretical_loss": 5.581581577512031,
"tokens_seen": 34930688
},
{
"epoch": 0.01,
"learning_rate": 0.0004996990972918757,
"loss": 5.0026,
"theoretical_loss": 5.579916875606134,
"tokens_seen": 34996224
},
{
"epoch": 0.01,
"learning_rate": 0.0004996890672016048,
"loss": 4.7549,
"theoretical_loss": 5.578256159222196,
"tokens_seen": 35061760
},
{
"epoch": 0.01,
"learning_rate": 0.000499679037111334,
"loss": 4.6224,
"theoretical_loss": 5.576599411395472,
"tokens_seen": 35127296
},
{
"epoch": 0.01,
"learning_rate": 0.0004996690070210632,
"loss": 4.7725,
"theoretical_loss": 5.574946615264906,
"tokens_seen": 35192832
},
{
"epoch": 0.01,
"learning_rate": 0.0004996589769307924,
"loss": 4.7807,
"theoretical_loss": 5.5732977540723105,
"tokens_seen": 35258368
},
{
"epoch": 0.01,
"learning_rate": 0.0004996489468405216,
"loss": 4.8423,
"theoretical_loss": 5.571652811161542,
"tokens_seen": 35323904
},
{
"epoch": 0.01,
"learning_rate": 0.0004996389167502507,
"loss": 4.9602,
"theoretical_loss": 5.570011769977693,
"tokens_seen": 35389440
},
{
"epoch": 0.01,
"learning_rate": 0.0004996288866599799,
"loss": 4.6312,
"theoretical_loss": 5.568374614066299,
"tokens_seen": 35454976
},
{
"epoch": 0.01,
"learning_rate": 0.0004996188565697091,
"loss": 5.0662,
"theoretical_loss": 5.566741327072535,
"tokens_seen": 35520512
},
{
"epoch": 0.01,
"learning_rate": 0.0004996088264794383,
"loss": 4.7143,
"theoretical_loss": 5.565111892740433,
"tokens_seen": 35586048
},
{
"epoch": 0.01,
"learning_rate": 0.0004995987963891675,
"loss": 4.8642,
"theoretical_loss": 5.563486294912105,
"tokens_seen": 35651584
},
{
"epoch": 0.01,
"learning_rate": 0.0004995887662988968,
"loss": 4.6962,
"theoretical_loss": 5.56186451752697,
"tokens_seen": 35717120
},
{
"epoch": 0.01,
"learning_rate": 0.0004995787362086258,
"loss": 4.7386,
"theoretical_loss": 5.560246544620993,
"tokens_seen": 35782656
},
{
"epoch": 0.01,
"learning_rate": 0.0004995687061183551,
"loss": 4.7368,
"theoretical_loss": 5.558632360325929,
"tokens_seen": 35848192
},
{
"epoch": 0.01,
"learning_rate": 0.0004995586760280842,
"loss": 4.6722,
"theoretical_loss": 5.557021948868571,
"tokens_seen": 35913728
},
{
"epoch": 0.01,
"learning_rate": 0.0004995486459378135,
"loss": 4.8891,
"theoretical_loss": 5.555415294570011,
"tokens_seen": 35979264
},
{
"epoch": 0.01,
"objective/train/docs_used": 32385,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.4478864669799805,
"objective/train/theoretical_loss": 5.553812381844907,
"objective/train/tokens_used": 56504800,
"theoretical_loss": 5.553812381844907,
"tokens_seen": 36044800
},
{
"epoch": 0.01,
"learning_rate": 0.0004995386158475427,
"loss": 4.8836,
"theoretical_loss": 5.553812381844907,
"tokens_seen": 36044800
},
{
"epoch": 0.01,
"learning_rate": 0.0004995285857572718,
"loss": 4.7132,
"theoretical_loss": 5.552213195200755,
"tokens_seen": 36110336
},
{
"epoch": 0.01,
"learning_rate": 0.000499518555667001,
"loss": 4.7041,
"theoretical_loss": 5.550617719237167,
"tokens_seen": 36175872
},
{
"epoch": 0.01,
"learning_rate": 0.0004995085255767302,
"loss": 4.8574,
"theoretical_loss": 5.549025938645155,
"tokens_seen": 36241408
},
{
"epoch": 0.01,
"learning_rate": 0.0004994984954864594,
"loss": 4.743,
"theoretical_loss": 5.547437838206435,
"tokens_seen": 36306944
},
{
"epoch": 0.01,
"learning_rate": 0.0004994884653961886,
"loss": 4.7443,
"theoretical_loss": 5.545853402792717,
"tokens_seen": 36372480
},
{
"epoch": 0.01,
"learning_rate": 0.0004994784353059178,
"loss": 4.8769,
"theoretical_loss": 5.544272617365014,
"tokens_seen": 36438016
},
{
"epoch": 0.01,
"learning_rate": 0.0004994684052156469,
"loss": 4.8343,
"theoretical_loss": 5.542695466972956,
"tokens_seen": 36503552
},
{
"epoch": 0.01,
"learning_rate": 0.0004994583751253761,
"loss": 4.7822,
"theoretical_loss": 5.541121936754111,
"tokens_seen": 36569088
},
{
"epoch": 0.01,
"learning_rate": 0.0004994483450351053,
"loss": 4.7453,
"theoretical_loss": 5.539552011933312,
"tokens_seen": 36634624
},
{
"epoch": 0.01,
"learning_rate": 0.0004994383149448345,
"loss": 4.8192,
"theoretical_loss": 5.537985677821986,
"tokens_seen": 36700160
},
{
"epoch": 0.01,
"learning_rate": 0.0004994282848545637,
"loss": 4.7944,
"theoretical_loss": 5.536422919817495,
"tokens_seen": 36765696
},
{
"epoch": 0.01,
"learning_rate": 0.0004994182547642928,
"loss": 4.8368,
"theoretical_loss": 5.5348637234024824,
"tokens_seen": 36831232
},
{
"epoch": 0.01,
"learning_rate": 0.0004994082246740221,
"loss": 4.8909,
"theoretical_loss": 5.53330807414422,
"tokens_seen": 36896768
},
{
"epoch": 0.01,
"learning_rate": 0.0004993981945837512,
"loss": 4.7606,
"theoretical_loss": 5.5317559576939725,
"tokens_seen": 36962304
},
{
"epoch": 0.01,
"learning_rate": 0.0004993881644934805,
"loss": 4.3933,
"theoretical_loss": 5.530207359786353,
"tokens_seen": 37027840
},
{
"epoch": 0.01,
"learning_rate": 0.0004993781344032096,
"loss": 4.9105,
"theoretical_loss": 5.5286622662386975,
"tokens_seen": 37093376
},
{
"epoch": 0.01,
"learning_rate": 0.0004993681043129389,
"loss": 4.725,
"theoretical_loss": 5.52712066295044,
"tokens_seen": 37158912
},
{
"epoch": 0.01,
"learning_rate": 0.000499358074222668,
"loss": 4.6152,
"theoretical_loss": 5.525582535902489,
"tokens_seen": 37224448
},
{
"epoch": 0.01,
"learning_rate": 0.0004993480441323972,
"loss": 4.8967,
"theoretical_loss": 5.524047871156618,
"tokens_seen": 37289984
},
{
"epoch": 0.01,
"learning_rate": 0.0004993380140421264,
"loss": 4.304,
"theoretical_loss": 5.52251665485486,
"tokens_seen": 37355520
},
{
"epoch": 0.01,
"learning_rate": 0.0004993279839518556,
"loss": 4.8395,
"theoretical_loss": 5.520988873218897,
"tokens_seen": 37421056
},
{
"epoch": 0.01,
"learning_rate": 0.0004993179538615848,
"loss": 4.7291,
"theoretical_loss": 5.519464512549478,
"tokens_seen": 37486592
},
{
"epoch": 0.01,
"learning_rate": 0.0004993079237713139,
"loss": 4.5346,
"theoretical_loss": 5.5179435592258095,
"tokens_seen": 37552128
},
{
"epoch": 0.01,
"learning_rate": 0.0004992978936810431,
"loss": 4.7017,
"theoretical_loss": 5.516425999704987,
"tokens_seen": 37617664
},
{
"epoch": 0.01,
"objective/train/docs_used": 32883,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.716377258300781,
"objective/train/theoretical_loss": 5.514911820521407,
"objective/train/tokens_used": 58143200,
"theoretical_loss": 5.514911820521407,
"tokens_seen": 37683200
},
{
"epoch": 0.01,
"learning_rate": 0.0004992878635907723,
"loss": 4.9009,
"theoretical_loss": 5.514911820521407,
"tokens_seen": 37683200
},
{
"epoch": 0.01,
"learning_rate": 0.0004992778335005015,
"loss": 4.9884,
"theoretical_loss": 5.5134010082861895,
"tokens_seen": 37748736
},
{
"epoch": 0.01,
"learning_rate": 0.0004992678034102307,
"loss": 4.7545,
"theoretical_loss": 5.511893549686616,
"tokens_seen": 37814272
},
{
"epoch": 0.01,
"learning_rate": 0.0004992577733199598,
"loss": 4.7888,
"theoretical_loss": 5.51038943148556,
"tokens_seen": 37879808
},
{
"epoch": 0.01,
"learning_rate": 0.000499247743229689,
"loss": 5.0533,
"theoretical_loss": 5.508888640520928,
"tokens_seen": 37945344
},
{
"epoch": 0.01,
"learning_rate": 0.0004992377131394183,
"loss": 4.8807,
"theoretical_loss": 5.50739116370511,
"tokens_seen": 38010880
},
{
"epoch": 0.01,
"learning_rate": 0.0004992276830491475,
"loss": 4.8592,
"theoretical_loss": 5.505896988024423,
"tokens_seen": 38076416
},
{
"epoch": 0.01,
"learning_rate": 0.0004992176529588767,
"loss": 4.7331,
"theoretical_loss": 5.5044061005385725,
"tokens_seen": 38141952
},
{
"epoch": 0.01,
"learning_rate": 0.0004992076228686059,
"loss": 4.8333,
"theoretical_loss": 5.502918488380116,
"tokens_seen": 38207488
},
{
"epoch": 0.01,
"learning_rate": 0.000499197592778335,
"loss": 4.7589,
"theoretical_loss": 5.501434138753918,
"tokens_seen": 38273024
},
{
"epoch": 0.01,
"learning_rate": 0.0004991875626880642,
"loss": 4.7598,
"theoretical_loss": 5.499953038936635,
"tokens_seen": 38338560
},
{
"epoch": 0.01,
"learning_rate": 0.0004991775325977934,
"loss": 4.7997,
"theoretical_loss": 5.498475176276176,
"tokens_seen": 38404096
},
{
"epoch": 0.01,
"learning_rate": 0.0004991675025075226,
"loss": 4.7163,
"theoretical_loss": 5.497000538191195,
"tokens_seen": 38469632
},
{
"epoch": 0.01,
"learning_rate": 0.0004991574724172518,
"loss": 4.7824,
"theoretical_loss": 5.495529112170568,
"tokens_seen": 38535168
},
{
"epoch": 0.01,
"learning_rate": 0.000499147442326981,
"loss": 4.8341,
"theoretical_loss": 5.494060885772887,
"tokens_seen": 38600704
},
{
"epoch": 0.01,
"learning_rate": 0.0004991374122367101,
"loss": 4.7344,
"theoretical_loss": 5.492595846625951,
"tokens_seen": 38666240
},
{
"epoch": 0.01,
"learning_rate": 0.0004991273821464393,
"loss": 4.7977,
"theoretical_loss": 5.491133982426266,
"tokens_seen": 38731776
},
{
"epoch": 0.01,
"learning_rate": 0.0004991173520561685,
"loss": 4.3782,
"theoretical_loss": 5.489675280938547,
"tokens_seen": 38797312
},
{
"epoch": 0.01,
"learning_rate": 0.0004991073219658977,
"loss": 4.7085,
"theoretical_loss": 5.488219729995227,
"tokens_seen": 38862848
},
{
"epoch": 0.01,
"learning_rate": 0.000499097291875627,
"loss": 4.5818,
"theoretical_loss": 5.486767317495966,
"tokens_seen": 38928384
},
{
"epoch": 0.01,
"learning_rate": 0.000499087261785356,
"loss": 5.0102,
"theoretical_loss": 5.48531803140717,
"tokens_seen": 38993920
},
{
"epoch": 0.01,
"learning_rate": 0.0004990772316950853,
"loss": 4.6955,
"theoretical_loss": 5.483871859761511,
"tokens_seen": 39059456
},
{
"epoch": 0.01,
"learning_rate": 0.0004990672016048144,
"loss": 4.6598,
"theoretical_loss": 5.482428790657449,
"tokens_seen": 39124992
},
{
"epoch": 0.01,
"learning_rate": 0.0004990571715145437,
"loss": 4.822,
"theoretical_loss": 5.480988812258763,
"tokens_seen": 39190528
},
{
"epoch": 0.01,
"learning_rate": 0.0004990471414242729,
"loss": 4.8752,
"theoretical_loss": 5.479551912794086,
"tokens_seen": 39256064
},
{
"epoch": 0.01,
"objective/train/docs_used": 34166,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.898425579071045,
"objective/train/theoretical_loss": 5.478118080556438,
"objective/train/tokens_used": 59781600,
"theoretical_loss": 5.478118080556438,
"tokens_seen": 39321600
},
{
"epoch": 0.01,
"learning_rate": 0.000499037111334002,
"loss": 4.7202,
"theoretical_loss": 5.478118080556438,
"tokens_seen": 39321600
},
{
"epoch": 0.01,
"learning_rate": 0.0004990270812437312,
"loss": 4.6432,
"theoretical_loss": 5.476687303902768,
"tokens_seen": 39387136
},
{
"epoch": 0.01,
"learning_rate": 0.0004990170511534604,
"loss": 4.8362,
"theoretical_loss": 5.475259571253502,
"tokens_seen": 39452672
},
{
"epoch": 0.01,
"learning_rate": 0.0004990070210631896,
"loss": 4.5375,
"theoretical_loss": 5.473834871092089,
"tokens_seen": 39518208
},
{
"epoch": 0.01,
"learning_rate": 0.0004989969909729188,
"loss": 4.6575,
"theoretical_loss": 5.4724131919645576,
"tokens_seen": 39583744
},
{
"epoch": 0.01,
"learning_rate": 0.000498986960882648,
"loss": 4.6854,
"theoretical_loss": 5.470994522479069,
"tokens_seen": 39649280
},
{
"epoch": 0.01,
"learning_rate": 0.0004989769307923771,
"loss": 4.5501,
"theoretical_loss": 5.4695788513054815,
"tokens_seen": 39714816
},
{
"epoch": 0.01,
"learning_rate": 0.0004989669007021063,
"loss": 4.7545,
"theoretical_loss": 5.468166167174912,
"tokens_seen": 39780352
},
{
"epoch": 0.01,
"learning_rate": 0.0004989568706118355,
"loss": 4.714,
"theoretical_loss": 5.466756458879306,
"tokens_seen": 39845888
},
{
"epoch": 0.01,
"learning_rate": 0.0004989468405215647,
"loss": 4.5583,
"theoretical_loss": 5.465349715271013,
"tokens_seen": 39911424
},
{
"epoch": 0.01,
"learning_rate": 0.0004989368104312939,
"loss": 4.7862,
"theoretical_loss": 5.463945925262355,
"tokens_seen": 39976960
},
{
"epoch": 0.01,
"learning_rate": 0.000498926780341023,
"loss": 4.4681,
"theoretical_loss": 5.462545077825214,
"tokens_seen": 40042496
},
{
"epoch": 0.01,
"learning_rate": 0.0004989167502507523,
"loss": 4.9631,
"theoretical_loss": 5.461147161990611,
"tokens_seen": 40108032
},
{
"epoch": 0.01,
"learning_rate": 0.0004989067201604814,
"loss": 4.662,
"theoretical_loss": 5.459752166848292,
"tokens_seen": 40173568
},
{
"epoch": 0.01,
"learning_rate": 0.0004988966900702107,
"loss": 4.7581,
"theoretical_loss": 5.458360081546321,
"tokens_seen": 40239104
},
{
"epoch": 0.01,
"learning_rate": 0.0004988866599799398,
"loss": 4.448,
"theoretical_loss": 5.456970895290674,
"tokens_seen": 40304640
},
{
"epoch": 0.01,
"learning_rate": 0.0004988766298896691,
"loss": 4.548,
"theoretical_loss": 5.455584597344835,
"tokens_seen": 40370176
},
{
"epoch": 0.01,
"learning_rate": 0.0004988665997993982,
"loss": 4.5585,
"theoretical_loss": 5.454201177029395,
"tokens_seen": 40435712
},
{
"epoch": 0.01,
"learning_rate": 0.0004988565697091274,
"loss": 4.6426,
"theoretical_loss": 5.452820623721662,
"tokens_seen": 40501248
},
{
"epoch": 0.01,
"learning_rate": 0.0004988465396188566,
"loss": 4.5398,
"theoretical_loss": 5.45144292685526,
"tokens_seen": 40566784
},
{
"epoch": 0.01,
"learning_rate": 0.0004988365095285858,
"loss": 4.5747,
"theoretical_loss": 5.450068075919752,
"tokens_seen": 40632320
},
{
"epoch": 0.01,
"learning_rate": 0.000498826479438315,
"loss": 4.6306,
"theoretical_loss": 5.44869606046024,
"tokens_seen": 40697856
},
{
"epoch": 0.01,
"learning_rate": 0.0004988164493480441,
"loss": 4.6076,
"theoretical_loss": 5.447326870076996,
"tokens_seen": 40763392
},
{
"epoch": 0.01,
"learning_rate": 0.0004988064192577733,
"loss": 4.6846,
"theoretical_loss": 5.445960494425072,
"tokens_seen": 40828928
},
{
"epoch": 0.01,
"learning_rate": 0.0004987963891675025,
"loss": 4.5425,
"theoretical_loss": 5.444596923213931,
"tokens_seen": 40894464
},
{
"epoch": 0.01,
"objective/train/docs_used": 34866,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.677554607391357,
"objective/train/theoretical_loss": 5.443236146207074,
"objective/train/tokens_used": 61420000,
"theoretical_loss": 5.443236146207074,
"tokens_seen": 40960000
},
{
"epoch": 0.01,
"learning_rate": 0.0004987863590772317,
"loss": 4.517,
"theoretical_loss": 5.443236146207074,
"tokens_seen": 40960000
},
{
"epoch": 0.01,
"learning_rate": 0.0004987763289869609,
"loss": 4.4186,
"theoretical_loss": 5.441878153221662,
"tokens_seen": 41025536
},
{
"epoch": 0.01,
"learning_rate": 0.00049876629889669,
"loss": 4.3927,
"theoretical_loss": 5.440522934128164,
"tokens_seen": 41091072
},
{
"epoch": 0.01,
"learning_rate": 0.0004987562688064192,
"loss": 4.521,
"theoretical_loss": 5.439170478849976,
"tokens_seen": 41156608
},
{
"epoch": 0.01,
"learning_rate": 0.0004987462387161484,
"loss": 4.4519,
"theoretical_loss": 5.437820777363078,
"tokens_seen": 41222144
},
{
"epoch": 0.01,
"learning_rate": 0.0004987362086258777,
"loss": 4.5882,
"theoretical_loss": 5.4364738196956655,
"tokens_seen": 41287680
},
{
"epoch": 0.01,
"learning_rate": 0.0004987261785356068,
"loss": 4.4504,
"theoretical_loss": 5.435129595927794,
"tokens_seen": 41353216
},
{
"epoch": 0.01,
"learning_rate": 0.0004987161484453361,
"loss": 4.1807,
"theoretical_loss": 5.433788096191039,
"tokens_seen": 41418752
},
{
"epoch": 0.01,
"learning_rate": 0.0004987061183550651,
"loss": 4.5904,
"theoretical_loss": 5.432449310668134,
"tokens_seen": 41484288
},
{
"epoch": 0.01,
"learning_rate": 0.0004986960882647944,
"loss": 4.5925,
"theoretical_loss": 5.4311132295926345,
"tokens_seen": 41549824
},
{
"epoch": 0.01,
"learning_rate": 0.0004986860581745236,
"loss": 4.6165,
"theoretical_loss": 5.42977984324857,
"tokens_seen": 41615360
},
{
"epoch": 0.01,
"learning_rate": 0.0004986760280842528,
"loss": 4.49,
"theoretical_loss": 5.428449141970107,
"tokens_seen": 41680896
},
{
"epoch": 0.01,
"learning_rate": 0.000498665997993982,
"loss": 4.5819,
"theoretical_loss": 5.427121116141212,
"tokens_seen": 41746432
},
{
"epoch": 0.01,
"learning_rate": 0.0004986559679037111,
"loss": 4.7141,
"theoretical_loss": 5.42579575619531,
"tokens_seen": 41811968
},
{
"epoch": 0.01,
"learning_rate": 0.0004986459378134403,
"loss": 4.6584,
"theoretical_loss": 5.424473052614967,
"tokens_seen": 41877504
},
{
"epoch": 0.01,
"learning_rate": 0.0004986359077231695,
"loss": 4.7053,
"theoretical_loss": 5.423152995931552,
"tokens_seen": 41943040
},
{
"epoch": 0.01,
"learning_rate": 0.0004986258776328987,
"loss": 4.677,
"theoretical_loss": 5.421835576724906,
"tokens_seen": 42008576
},
{
"epoch": 0.01,
"learning_rate": 0.0004986158475426279,
"loss": 4.5363,
"theoretical_loss": 5.420520785623031,
"tokens_seen": 42074112
},
{
"epoch": 0.01,
"learning_rate": 0.000498605817452357,
"loss": 4.4347,
"theoretical_loss": 5.4192086133017625,
"tokens_seen": 42139648
},
{
"epoch": 0.01,
"learning_rate": 0.0004985957873620862,
"loss": 4.5436,
"theoretical_loss": 5.417899050484451,
"tokens_seen": 42205184
},
{
"epoch": 0.01,
"learning_rate": 0.0004985857572718154,
"loss": 4.5196,
"theoretical_loss": 5.416592087941646,
"tokens_seen": 42270720
},
{
"epoch": 0.01,
"learning_rate": 0.0004985757271815446,
"loss": 4.474,
"theoretical_loss": 5.415287716490787,
"tokens_seen": 42336256
},
{
"epoch": 0.01,
"learning_rate": 0.0004985656970912738,
"loss": 4.9763,
"theoretical_loss": 5.413985926995892,
"tokens_seen": 42401792
},
{
"epoch": 0.01,
"learning_rate": 0.0004985556670010031,
"loss": 4.4879,
"theoretical_loss": 5.412686710367245,
"tokens_seen": 42467328
},
{
"epoch": 0.01,
"learning_rate": 0.0004985456369107321,
"loss": 4.7709,
"theoretical_loss": 5.411390057561097,
"tokens_seen": 42532864
},
{
"epoch": 0.01,
"objective/train/docs_used": 35868,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.189578056335449,
"objective/train/theoretical_loss": 5.410095959579362,
"objective/train/tokens_used": 63058400,
"theoretical_loss": 5.410095959579362,
"tokens_seen": 42598400
},
{
"epoch": 0.01,
"learning_rate": 0.0004985356068204614,
"loss": 4.5885,
"theoretical_loss": 5.410095959579362,
"tokens_seen": 42598400
},
{
"epoch": 0.01,
"learning_rate": 0.0004985255767301905,
"loss": 4.4367,
"theoretical_loss": 5.408804407469308,
"tokens_seen": 42663936
},
{
"epoch": 0.01,
"learning_rate": 0.0004985155466399198,
"loss": 4.697,
"theoretical_loss": 5.407515392323276,
"tokens_seen": 42729472
},
{
"epoch": 0.01,
"learning_rate": 0.000498505516549649,
"loss": 4.6402,
"theoretical_loss": 5.406228905278368,
"tokens_seen": 42795008
},
{
"epoch": 0.01,
"learning_rate": 0.0004984954864593782,
"loss": 4.743,
"theoretical_loss": 5.404944937516161,
"tokens_seen": 42860544
},
{
"epoch": 0.01,
"learning_rate": 0.0004984854563691073,
"loss": 4.4767,
"theoretical_loss": 5.403663480262418,
"tokens_seen": 42926080
},
{
"epoch": 0.01,
"learning_rate": 0.0004984754262788365,
"loss": 4.2795,
"theoretical_loss": 5.402384524786797,
"tokens_seen": 42991616
},
{
"epoch": 0.01,
"learning_rate": 0.0004984653961885657,
"loss": 4.5048,
"theoretical_loss": 5.401108062402562,
"tokens_seen": 43057152
},
{
"epoch": 0.01,
"learning_rate": 0.0004984553660982949,
"loss": 4.285,
"theoretical_loss": 5.399834084466306,
"tokens_seen": 43122688
},
{
"epoch": 0.01,
"learning_rate": 0.0004984453360080241,
"loss": 4.5244,
"theoretical_loss": 5.398562582377666,
"tokens_seen": 43188224
},
{
"epoch": 0.01,
"learning_rate": 0.0004984353059177532,
"loss": 4.4912,
"theoretical_loss": 5.397293547579041,
"tokens_seen": 43253760
},
{
"epoch": 0.01,
"learning_rate": 0.0004984252758274825,
"loss": 4.5286,
"theoretical_loss": 5.396026971555319,
"tokens_seen": 43319296
},
{
"epoch": 0.01,
"learning_rate": 0.0004984152457372116,
"loss": 4.635,
"theoretical_loss": 5.394762845833601,
"tokens_seen": 43384832
},
{
"epoch": 0.01,
"learning_rate": 0.0004984052156469409,
"loss": 4.6563,
"theoretical_loss": 5.393501161982926,
"tokens_seen": 43450368
},
{
"epoch": 0.01,
"learning_rate": 0.00049839518555667,
"loss": 4.5338,
"theoretical_loss": 5.392241911614005,
"tokens_seen": 43515904
},
{
"epoch": 0.01,
"learning_rate": 0.0004983851554663993,
"loss": 4.6497,
"theoretical_loss": 5.390985086378949,
"tokens_seen": 43581440
},
{
"epoch": 0.01,
"learning_rate": 0.0004983751253761284,
"loss": 4.656,
"theoretical_loss": 5.389730677971002,
"tokens_seen": 43646976
},
{
"epoch": 0.01,
"learning_rate": 0.0004983650952858576,
"loss": 4.2885,
"theoretical_loss": 5.388478678124285,
"tokens_seen": 43712512
},
{
"epoch": 0.01,
"learning_rate": 0.0004983550651955868,
"loss": 4.6653,
"theoretical_loss": 5.387229078613521,
"tokens_seen": 43778048
},
{
"epoch": 0.01,
"learning_rate": 0.000498345035105316,
"loss": 4.8537,
"theoretical_loss": 5.385981871253785,
"tokens_seen": 43843584
},
{
"epoch": 0.01,
"learning_rate": 0.0004983350050150452,
"loss": 4.5315,
"theoretical_loss": 5.384737047900243,
"tokens_seen": 43909120
},
{
"epoch": 0.01,
"learning_rate": 0.0004983249749247743,
"loss": 4.4477,
"theoretical_loss": 5.3834946004478965,
"tokens_seen": 43974656
},
{
"epoch": 0.01,
"learning_rate": 0.0004983149448345035,
"loss": 4.2702,
"theoretical_loss": 5.382254520831328,
"tokens_seen": 44040192
},
{
"epoch": 0.01,
"learning_rate": 0.0004983049147442327,
"loss": 4.4734,
"theoretical_loss": 5.381016801024449,
"tokens_seen": 44105728
},
{
"epoch": 0.01,
"learning_rate": 0.0004982948846539619,
"loss": 4.5673,
"theoretical_loss": 5.379781433040252,
"tokens_seen": 44171264
},
{
"epoch": 0.01,
"objective/train/docs_used": 36544,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.2153215408325195,
"objective/train/theoretical_loss": 5.378548408930558,
"objective/train/tokens_used": 64696800,
"theoretical_loss": 5.378548408930558,
"tokens_seen": 44236800
},
{
"epoch": 0.01,
"learning_rate": 0.0004982848545636911,
"loss": 4.3754,
"theoretical_loss": 5.378548408930558,
"tokens_seen": 44236800
},
{
"epoch": 0.01,
"learning_rate": 0.0004982748244734202,
"loss": 4.6866,
"theoretical_loss": 5.377317720785777,
"tokens_seen": 44302336
},
{
"epoch": 0.01,
"learning_rate": 0.0004982647943831494,
"loss": 4.2689,
"theoretical_loss": 5.37608936073466,
"tokens_seen": 44367872
},
{
"epoch": 0.01,
"learning_rate": 0.0004982547642928786,
"loss": 4.558,
"theoretical_loss": 5.374863320944057,
"tokens_seen": 44433408
},
{
"epoch": 0.01,
"learning_rate": 0.0004982447342026079,
"loss": 4.3621,
"theoretical_loss": 5.373639593618675,
"tokens_seen": 44498944
},
{
"epoch": 0.01,
"learning_rate": 0.000498234704112337,
"loss": 4.7059,
"theoretical_loss": 5.372418171000847,
"tokens_seen": 44564480
},
{
"epoch": 0.01,
"learning_rate": 0.0004982246740220663,
"loss": 4.4685,
"theoretical_loss": 5.371199045370283,
"tokens_seen": 44630016
},
{
"epoch": 0.01,
"learning_rate": 0.0004982146439317953,
"loss": 4.502,
"theoretical_loss": 5.369982209043851,
"tokens_seen": 44695552
},
{
"epoch": 0.01,
"learning_rate": 0.0004982046138415246,
"loss": 4.5419,
"theoretical_loss": 5.368767654375327,
"tokens_seen": 44761088
},
{
"epoch": 0.01,
"learning_rate": 0.0004981945837512538,
"loss": 4.6487,
"theoretical_loss": 5.367555373755179,
"tokens_seen": 44826624
},
{
"epoch": 0.01,
"learning_rate": 0.000498184553660983,
"loss": 4.5139,
"theoretical_loss": 5.366345359610327,
"tokens_seen": 44892160
},
{
"epoch": 0.01,
"learning_rate": 0.0004981745235707122,
"loss": 4.6486,
"theoretical_loss": 5.365137604403923,
"tokens_seen": 44957696
},
{
"epoch": 0.01,
"learning_rate": 0.0004981644934804413,
"loss": 4.6411,
"theoretical_loss": 5.363932100635117,
"tokens_seen": 45023232
},
{
"epoch": 0.01,
"learning_rate": 0.0004981544633901705,
"loss": 4.4685,
"theoretical_loss": 5.362728840838843,
"tokens_seen": 45088768
},
{
"epoch": 0.01,
"learning_rate": 0.0004981444332998997,
"loss": 4.3304,
"theoretical_loss": 5.361527817585586,
"tokens_seen": 45154304
},
{
"epoch": 0.01,
"learning_rate": 0.0004981344032096289,
"loss": 4.5782,
"theoretical_loss": 5.360329023481169,
"tokens_seen": 45219840
},
{
"epoch": 0.01,
"learning_rate": 0.0004981243731193581,
"loss": 4.4319,
"theoretical_loss": 5.359132451166534,
"tokens_seen": 45285376
},
{
"epoch": 0.01,
"learning_rate": 0.0004981143430290873,
"loss": 4.6029,
"theoretical_loss": 5.357938093317518,
"tokens_seen": 45350912
},
{
"epoch": 0.01,
"learning_rate": 0.0004981043129388164,
"loss": 4.458,
"theoretical_loss": 5.356745942644645,
"tokens_seen": 45416448
},
{
"epoch": 0.01,
"learning_rate": 0.0004980942828485456,
"loss": 4.3535,
"theoretical_loss": 5.355555991892905,
"tokens_seen": 45481984
},
{
"epoch": 0.01,
"learning_rate": 0.0004980842527582748,
"loss": 4.826,
"theoretical_loss": 5.35436823384155,
"tokens_seen": 45547520
},
{
"epoch": 0.01,
"learning_rate": 0.000498074222668004,
"loss": 4.6619,
"theoretical_loss": 5.353182661303873,
"tokens_seen": 45613056
},
{
"epoch": 0.01,
"learning_rate": 0.0004980641925777333,
"loss": 4.3435,
"theoretical_loss": 5.35199926712701,
"tokens_seen": 45678592
},
{
"epoch": 0.01,
"learning_rate": 0.0004980541624874623,
"loss": 4.7139,
"theoretical_loss": 5.350818044191721,
"tokens_seen": 45744128
},
{
"epoch": 0.01,
"learning_rate": 0.0004980441323971916,
"loss": 4.5414,
"theoretical_loss": 5.349638985412193,
"tokens_seen": 45809664
},
{
"epoch": 0.01,
"objective/train/docs_used": 37915,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.255738258361816,
"objective/train/theoretical_loss": 5.348462083735834,
"objective/train/tokens_used": 66335200,
"theoretical_loss": 5.348462083735834,
"tokens_seen": 45875200
},
{
"epoch": 0.01,
"learning_rate": 0.0004980341023069207,
"loss": 4.2427,
"theoretical_loss": 5.348462083735834,
"tokens_seen": 45875200
},
{
"epoch": 0.01,
"learning_rate": 0.00049802407221665,
"loss": 4.4766,
"theoretical_loss": 5.347287332143064,
"tokens_seen": 45940736
},
{
"epoch": 0.01,
"learning_rate": 0.0004980140421263792,
"loss": 4.6487,
"theoretical_loss": 5.346114723647119,
"tokens_seen": 46006272
},
{
"epoch": 0.01,
"learning_rate": 0.0004980040120361084,
"loss": 4.744,
"theoretical_loss": 5.344944251293852,
"tokens_seen": 46071808
},
{
"epoch": 0.01,
"learning_rate": 0.0004979939819458375,
"loss": 4.4822,
"theoretical_loss": 5.343775908161532,
"tokens_seen": 46137344
},
{
"epoch": 0.01,
"learning_rate": 0.0004979839518555667,
"loss": 4.6656,
"theoretical_loss": 5.342609687360644,
"tokens_seen": 46202880
},
{
"epoch": 0.01,
"learning_rate": 0.0004979739217652959,
"loss": 4.368,
"theoretical_loss": 5.341445582033705,
"tokens_seen": 46268416
},
{
"epoch": 0.01,
"learning_rate": 0.0004979638916750251,
"loss": 4.5362,
"theoretical_loss": 5.3402835853550545,
"tokens_seen": 46333952
},
{
"epoch": 0.01,
"learning_rate": 0.0004979538615847543,
"loss": 4.5253,
"theoretical_loss": 5.339123690530673,
"tokens_seen": 46399488
},
{
"epoch": 0.01,
"learning_rate": 0.0004979438314944834,
"loss": 4.5891,
"theoretical_loss": 5.337965890797989,
"tokens_seen": 46465024
},
{
"epoch": 0.01,
"learning_rate": 0.0004979338014042126,
"loss": 4.3895,
"theoretical_loss": 5.336810179425685,
"tokens_seen": 46530560
},
{
"epoch": 0.01,
"learning_rate": 0.0004979237713139418,
"loss": 4.7045,
"theoretical_loss": 5.335656549713516,
"tokens_seen": 46596096
},
{
"epoch": 0.01,
"learning_rate": 0.000497913741223671,
"loss": 4.3343,
"theoretical_loss": 5.334504994992115,
"tokens_seen": 46661632
},
{
"epoch": 0.01,
"learning_rate": 0.0004979037111334002,
"loss": 4.5711,
"theoretical_loss": 5.333355508622814,
"tokens_seen": 46727168
},
{
"epoch": 0.01,
"learning_rate": 0.0004978936810431293,
"loss": 4.5964,
"theoretical_loss": 5.332208083997459,
"tokens_seen": 46792704
},
{
"epoch": 0.01,
"learning_rate": 0.0004978836509528586,
"loss": 4.3615,
"theoretical_loss": 5.33106271453822,
"tokens_seen": 46858240
},
{
"epoch": 0.01,
"learning_rate": 0.0004978736208625877,
"loss": 4.6588,
"theoretical_loss": 5.329919393697422,
"tokens_seen": 46923776
},
{
"epoch": 0.01,
"learning_rate": 0.000497863590772317,
"loss": 4.8373,
"theoretical_loss": 5.328778114957351,
"tokens_seen": 46989312
},
{
"epoch": 0.01,
"learning_rate": 0.0004978535606820461,
"loss": 4.2761,
"theoretical_loss": 5.327638871830089,
"tokens_seen": 47054848
},
{
"epoch": 0.01,
"learning_rate": 0.0004978435305917754,
"loss": 4.4197,
"theoretical_loss": 5.326501657857326,
"tokens_seen": 47120384
},
{
"epoch": 0.01,
"learning_rate": 0.0004978335005015045,
"loss": 4.435,
"theoretical_loss": 5.32536646661019,
"tokens_seen": 47185920
},
{
"epoch": 0.01,
"learning_rate": 0.0004978234704112337,
"loss": 4.6838,
"theoretical_loss": 5.324233291689069,
"tokens_seen": 47251456
},
{
"epoch": 0.01,
"learning_rate": 0.0004978134403209629,
"loss": 4.3913,
"theoretical_loss": 5.323102126723439,
"tokens_seen": 47316992
},
{
"epoch": 0.01,
"learning_rate": 0.0004978034102306921,
"loss": 4.6,
"theoretical_loss": 5.321972965371691,
"tokens_seen": 47382528
},
{
"epoch": 0.01,
"learning_rate": 0.0004977933801404213,
"loss": 4.1382,
"theoretical_loss": 5.320845801320959,
"tokens_seen": 47448064
},
{
"epoch": 0.01,
"objective/train/docs_used": 38651,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.045684814453125,
"objective/train/theoretical_loss": 5.319720628286955,
"objective/train/tokens_used": 67973600,
"theoretical_loss": 5.319720628286955,
"tokens_seen": 47513600
},
{
"epoch": 0.01,
"learning_rate": 0.0004977833500501504,
"loss": 4.343,
"theoretical_loss": 5.319720628286955,
"tokens_seen": 47513600
},
{
"epoch": 0.01,
"learning_rate": 0.0004977733199598796,
"loss": 4.3051,
"theoretical_loss": 5.318597440013795,
"tokens_seen": 47579136
},
{
"epoch": 0.01,
"learning_rate": 0.0004977632898696088,
"loss": 4.6262,
"theoretical_loss": 5.317476230273831,
"tokens_seen": 47644672
},
{
"epoch": 0.01,
"learning_rate": 0.000497753259779338,
"loss": 4.5661,
"theoretical_loss": 5.316356992867491,
"tokens_seen": 47710208
},
{
"epoch": 0.01,
"learning_rate": 0.0004977432296890672,
"loss": 4.5954,
"theoretical_loss": 5.31523972162311,
"tokens_seen": 47775744
},
{
"epoch": 0.01,
"learning_rate": 0.0004977331995987965,
"loss": 4.4414,
"theoretical_loss": 5.314124410396767,
"tokens_seen": 47841280
},
{
"epoch": 0.01,
"learning_rate": 0.0004977231695085255,
"loss": 4.7758,
"theoretical_loss": 5.31301105307212,
"tokens_seen": 47906816
},
{
"epoch": 0.01,
"learning_rate": 0.0004977131394182548,
"loss": 4.603,
"theoretical_loss": 5.311899643560251,
"tokens_seen": 47972352
},
{
"epoch": 0.01,
"learning_rate": 0.000497703109327984,
"loss": 4.6245,
"theoretical_loss": 5.310790175799497,
"tokens_seen": 48037888
},
{
"epoch": 0.01,
"learning_rate": 0.0004976930792377132,
"loss": 4.5712,
"theoretical_loss": 5.3096826437553,
"tokens_seen": 48103424
},
{
"epoch": 0.01,
"learning_rate": 0.0004976830491474424,
"loss": 4.5459,
"theoretical_loss": 5.308577041420046,
"tokens_seen": 48168960
},
{
"epoch": 0.01,
"learning_rate": 0.0004976730190571715,
"loss": 4.6043,
"theoretical_loss": 5.3074733628129005,
"tokens_seen": 48234496
},
{
"epoch": 0.01,
"learning_rate": 0.0004976629889669007,
"loss": 4.5158,
"theoretical_loss": 5.3063716019796665,
"tokens_seen": 48300032
},
{
"epoch": 0.01,
"learning_rate": 0.0004976529588766299,
"loss": 4.4244,
"theoretical_loss": 5.305271752992619,
"tokens_seen": 48365568
},
{
"epoch": 0.01,
"learning_rate": 0.0004976429287863591,
"loss": 4.5599,
"theoretical_loss": 5.304173809950358,
"tokens_seen": 48431104
},
{
"epoch": 0.01,
"learning_rate": 0.0004976328986960883,
"loss": 4.5297,
"theoretical_loss": 5.303077766977653,
"tokens_seen": 48496640
},
{
"epoch": 0.01,
"learning_rate": 0.0004976228686058175,
"loss": 4.3215,
"theoretical_loss": 5.3019836182252895,
"tokens_seen": 48562176
},
{
"epoch": 0.01,
"learning_rate": 0.0004976128385155466,
"loss": 4.3245,
"theoretical_loss": 5.300891357869929,
"tokens_seen": 48627712
},
{
"epoch": 0.01,
"learning_rate": 0.0004976028084252758,
"loss": 4.2442,
"theoretical_loss": 5.299800980113945,
"tokens_seen": 48693248
},
{
"epoch": 0.01,
"learning_rate": 0.000497592778335005,
"loss": 4.2681,
"theoretical_loss": 5.298712479185288,
"tokens_seen": 48758784
},
{
"epoch": 0.01,
"learning_rate": 0.0004975827482447342,
"loss": 4.5281,
"theoretical_loss": 5.297625849337331,
"tokens_seen": 48824320
},
{
"epoch": 0.01,
"learning_rate": 0.0004975727181544635,
"loss": 4.5414,
"theoretical_loss": 5.296541084848727,
"tokens_seen": 48889856
},
{
"epoch": 0.01,
"learning_rate": 0.0004975626880641925,
"loss": 4.3081,
"theoretical_loss": 5.295458180023262,
"tokens_seen": 48955392
},
{
"epoch": 0.01,
"learning_rate": 0.0004975526579739218,
"loss": 4.1252,
"theoretical_loss": 5.294377129189715,
"tokens_seen": 49020928
},
{
"epoch": 0.01,
"learning_rate": 0.0004975426278836509,
"loss": 4.3003,
"theoretical_loss": 5.293297926701706,
"tokens_seen": 49086464
},
{
"epoch": 0.01,
"objective/train/docs_used": 39958,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.954103946685791,
"objective/train/theoretical_loss": 5.292220566937567,
"objective/train/tokens_used": 69612000,
"theoretical_loss": 5.292220566937567,
"tokens_seen": 49152000
},
{
"epoch": 0.01,
"learning_rate": 0.0004975325977933802,
"loss": 4.5056,
"theoretical_loss": 5.292220566937567,
"tokens_seen": 49152000
},
{
"epoch": 0.01,
"learning_rate": 0.0004975225677031094,
"loss": 4.549,
"theoretical_loss": 5.29114504430019,
"tokens_seen": 49217536
},
{
"epoch": 0.01,
"learning_rate": 0.0004975125376128386,
"loss": 4.2312,
"theoretical_loss": 5.290071353216895,
"tokens_seen": 49283072
},
{
"epoch": 0.01,
"learning_rate": 0.0004975025075225677,
"loss": 4.4122,
"theoretical_loss": 5.288999488139284,
"tokens_seen": 49348608
},
{
"epoch": 0.01,
"learning_rate": 0.0004974924774322969,
"loss": 4.4153,
"theoretical_loss": 5.28792944354311,
"tokens_seen": 49414144
},
{
"epoch": 0.01,
"learning_rate": 0.0004974824473420261,
"loss": 4.5622,
"theoretical_loss": 5.286861213928137,
"tokens_seen": 49479680
},
{
"epoch": 0.02,
"learning_rate": 0.0004974724172517553,
"loss": 4.4577,
"theoretical_loss": 5.285794793817999,
"tokens_seen": 49545216
},
{
"epoch": 0.02,
"learning_rate": 0.0004974623871614845,
"loss": 4.3913,
"theoretical_loss": 5.284730177760077,
"tokens_seen": 49610752
},
{
"epoch": 0.02,
"learning_rate": 0.0004974523570712136,
"loss": 4.0983,
"theoretical_loss": 5.283667360325351,
"tokens_seen": 49676288
},
{
"epoch": 0.02,
"learning_rate": 0.0004974423269809428,
"loss": 4.3874,
"theoretical_loss": 5.2826063361082785,
"tokens_seen": 49741824
},
{
"epoch": 0.02,
"learning_rate": 0.000497432296890672,
"loss": 4.4498,
"theoretical_loss": 5.281547099726654,
"tokens_seen": 49807360
},
{
"epoch": 0.02,
"learning_rate": 0.0004974222668004012,
"loss": 4.2229,
"theoretical_loss": 5.280489645821483,
"tokens_seen": 49872896
},
{
"epoch": 0.02,
"learning_rate": 0.0004974122367101304,
"loss": 4.4956,
"theoretical_loss": 5.279433969056848,
"tokens_seen": 49938432
},
{
"epoch": 0.02,
"learning_rate": 0.0004974022066198595,
"loss": 4.2938,
"theoretical_loss": 5.278380064119782,
"tokens_seen": 50003968
},
{
"epoch": 0.02,
"learning_rate": 0.0004973921765295888,
"loss": 4.2649,
"theoretical_loss": 5.277327925720137,
"tokens_seen": 50069504
},
{
"epoch": 0.02,
"learning_rate": 0.0004973821464393179,
"loss": 4.6124,
"theoretical_loss": 5.276277548590457,
"tokens_seen": 50135040
},
{
"epoch": 0.02,
"learning_rate": 0.0004973721163490472,
"loss": 4.4276,
"theoretical_loss": 5.275228927485855,
"tokens_seen": 50200576
},
{
"epoch": 0.02,
"learning_rate": 0.0004973620862587763,
"loss": 4.4451,
"theoretical_loss": 5.2741820571838804,
"tokens_seen": 50266112
},
{
"epoch": 0.02,
"learning_rate": 0.0004973520561685056,
"loss": 4.4044,
"theoretical_loss": 5.273136932484399,
"tokens_seen": 50331648
},
{
"epoch": 0.02,
"learning_rate": 0.0004973420260782347,
"loss": 4.1472,
"theoretical_loss": 5.272093548209467,
"tokens_seen": 50397184
},
{
"epoch": 0.02,
"learning_rate": 0.0004973319959879639,
"loss": 4.3286,
"theoretical_loss": 5.271051899203207,
"tokens_seen": 50462720
},
{
"epoch": 0.02,
"learning_rate": 0.0004973219658976931,
"loss": 4.4202,
"theoretical_loss": 5.270011980331685,
"tokens_seen": 50528256
},
{
"epoch": 0.02,
"learning_rate": 0.0004973119358074223,
"loss": 4.3812,
"theoretical_loss": 5.268973786482794,
"tokens_seen": 50593792
},
{
"epoch": 0.02,
"learning_rate": 0.0004973019057171515,
"loss": 4.4977,
"theoretical_loss": 5.267937312566123,
"tokens_seen": 50659328
},
{
"epoch": 0.02,
"learning_rate": 0.0004972918756268806,
"loss": 4.4049,
"theoretical_loss": 5.266902553512847,
"tokens_seen": 50724864
},
{
"epoch": 0.02,
"objective/train/docs_used": 40584,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.171596527099609,
"objective/train/theoretical_loss": 5.265869504275602,
"objective/train/tokens_used": 71250400,
"theoretical_loss": 5.265869504275602,
"tokens_seen": 50790400
},
{
"epoch": 0.02,
"learning_rate": 0.0004972818455366098,
"loss": 4.5105,
"theoretical_loss": 5.265869504275602,
"tokens_seen": 50790400
},
{
"epoch": 0.02,
"learning_rate": 0.000497271815446339,
"loss": 4.2898,
"theoretical_loss": 5.264838159828369,
"tokens_seen": 50855936
},
{
"epoch": 0.02,
"learning_rate": 0.0004972617853560682,
"loss": 4.5284,
"theoretical_loss": 5.263808515166355,
"tokens_seen": 50921472
},
{
"epoch": 0.02,
"learning_rate": 0.0004972517552657974,
"loss": 4.1863,
"theoretical_loss": 5.262780565305875,
"tokens_seen": 50987008
},
{
"epoch": 0.02,
"learning_rate": 0.0004972417251755266,
"loss": 4.1704,
"theoretical_loss": 5.261754305284241,
"tokens_seen": 51052544
},
{
"epoch": 0.02,
"learning_rate": 0.0004972316950852557,
"loss": 4.4493,
"theoretical_loss": 5.260729730159641,
"tokens_seen": 51118080
},
{
"epoch": 0.02,
"learning_rate": 0.0004972216649949849,
"loss": 4.1237,
"theoretical_loss": 5.259706835011027,
"tokens_seen": 51183616
},
{
"epoch": 0.02,
"learning_rate": 0.0004972116349047142,
"loss": 4.3399,
"theoretical_loss": 5.2586856149380035,
"tokens_seen": 51249152
},
{
"epoch": 0.02,
"learning_rate": 0.0004972016048144433,
"loss": 4.0849,
"theoretical_loss": 5.257666065060709,
"tokens_seen": 51314688
},
{
"epoch": 0.02,
"learning_rate": 0.0004971915747241726,
"loss": 4.4541,
"theoretical_loss": 5.256648180519708,
"tokens_seen": 51380224
},
{
"epoch": 0.02,
"learning_rate": 0.0004971815446339017,
"loss": 4.353,
"theoretical_loss": 5.255631956475881,
"tokens_seen": 51445760
},
{
"epoch": 0.02,
"learning_rate": 0.0004971715145436309,
"loss": 4.2266,
"theoretical_loss": 5.25461738811031,
"tokens_seen": 51511296
},
{
"epoch": 0.02,
"learning_rate": 0.0004971614844533601,
"loss": 4.4598,
"theoretical_loss": 5.25360447062417,
"tokens_seen": 51576832
},
{
"epoch": 0.02,
"learning_rate": 0.0004971514543630893,
"loss": 4.5701,
"theoretical_loss": 5.252593199238619,
"tokens_seen": 51642368
},
{
"epoch": 0.02,
"learning_rate": 0.0004971414242728185,
"loss": 4.2901,
"theoretical_loss": 5.2515835691946915,
"tokens_seen": 51707904
},
{
"epoch": 0.02,
"learning_rate": 0.0004971313941825477,
"loss": 4.3521,
"theoretical_loss": 5.2505755757531904,
"tokens_seen": 51773440
},
{
"epoch": 0.02,
"learning_rate": 0.0004971213640922768,
"loss": 4.3861,
"theoretical_loss": 5.24956921419458,
"tokens_seen": 51838976
},
{
"epoch": 0.02,
"learning_rate": 0.000497111334002006,
"loss": 4.3723,
"theoretical_loss": 5.248564479818876,
"tokens_seen": 51904512
},
{
"epoch": 0.02,
"learning_rate": 0.0004971013039117352,
"loss": 4.1067,
"theoretical_loss": 5.247561367945544,
"tokens_seen": 51970048
},
{
"epoch": 0.02,
"learning_rate": 0.0004970912738214644,
"loss": 4.3448,
"theoretical_loss": 5.246559873913396,
"tokens_seen": 52035584
},
{
"epoch": 0.02,
"learning_rate": 0.0004970812437311936,
"loss": 4.2188,
"theoretical_loss": 5.245559993080484,
"tokens_seen": 52101120
},
{
"epoch": 0.02,
"learning_rate": 0.0004970712136409227,
"loss": 4.0193,
"theoretical_loss": 5.24456172082399,
"tokens_seen": 52166656
},
{
"epoch": 0.02,
"learning_rate": 0.0004970611835506519,
"loss": 4.537,
"theoretical_loss": 5.243565052540136,
"tokens_seen": 52232192
},
{
"epoch": 0.02,
"learning_rate": 0.0004970511534603811,
"loss": 4.4592,
"theoretical_loss": 5.242569983644074,
"tokens_seen": 52297728
},
{
"epoch": 0.02,
"learning_rate": 0.0004970411233701103,
"loss": 4.2549,
"theoretical_loss": 5.241576509569784,
"tokens_seen": 52363264
},
{
"epoch": 0.02,
"objective/train/docs_used": 41941,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.2740397453308105,
"objective/train/theoretical_loss": 5.240584625769978,
"objective/train/tokens_used": 72888800,
"theoretical_loss": 5.240584625769978,
"tokens_seen": 52428800
},
{
"epoch": 0.02,
"learning_rate": 0.0004970310932798396,
"loss": 4.1928,
"theoretical_loss": 5.240584625769978,
"tokens_seen": 52428800
},
{
"epoch": 0.02,
"learning_rate": 0.0004970210631895686,
"loss": 4.4434,
"theoretical_loss": 5.239594327715992,
"tokens_seen": 52494336
},
{
"epoch": 0.02,
"learning_rate": 0.0004970110330992979,
"loss": 4.5166,
"theoretical_loss": 5.238605610897698,
"tokens_seen": 52559872
},
{
"epoch": 0.02,
"learning_rate": 0.0004970010030090271,
"loss": 3.9728,
"theoretical_loss": 5.237618470823394,
"tokens_seen": 52625408
},
{
"epoch": 0.02,
"learning_rate": 0.0004969909729187563,
"loss": 4.3201,
"theoretical_loss": 5.2366329030197125,
"tokens_seen": 52690944
},
{
"epoch": 0.02,
"learning_rate": 0.0004969809428284855,
"loss": 4.236,
"theoretical_loss": 5.235648903031521,
"tokens_seen": 52756480
},
{
"epoch": 0.02,
"learning_rate": 0.0004969709127382147,
"loss": 4.3479,
"theoretical_loss": 5.2346664664218245,
"tokens_seen": 52822016
},
{
"epoch": 0.02,
"learning_rate": 0.0004969608826479438,
"loss": 4.3036,
"theoretical_loss": 5.233685588771669,
"tokens_seen": 52887552
},
{
"epoch": 0.02,
"learning_rate": 0.000496950852557673,
"loss": 4.3347,
"theoretical_loss": 5.232706265680049,
"tokens_seen": 52953088
},
{
"epoch": 0.02,
"learning_rate": 0.0004969408224674022,
"loss": 4.3565,
"theoretical_loss": 5.231728492763811,
"tokens_seen": 53018624
},
{
"epoch": 0.02,
"learning_rate": 0.0004969307923771314,
"loss": 4.6295,
"theoretical_loss": 5.230752265657554,
"tokens_seen": 53084160
},
{
"epoch": 0.02,
"learning_rate": 0.0004969207622868606,
"loss": 4.3937,
"theoretical_loss": 5.229777580013545,
"tokens_seen": 53149696
},
{
"epoch": 0.02,
"learning_rate": 0.0004969107321965897,
"loss": 4.4336,
"theoretical_loss": 5.228804431501619,
"tokens_seen": 53215232
},
{
"epoch": 0.02,
"learning_rate": 0.000496900702106319,
"loss": 4.2758,
"theoretical_loss": 5.227832815809087,
"tokens_seen": 53280768
},
{
"epoch": 0.02,
"learning_rate": 0.0004968906720160481,
"loss": 4.1819,
"theoretical_loss": 5.226862728640651,
"tokens_seen": 53346304
},
{
"epoch": 0.02,
"learning_rate": 0.0004968806419257774,
"loss": 4.365,
"theoretical_loss": 5.2258941657183,
"tokens_seen": 53411840
},
{
"epoch": 0.02,
"learning_rate": 0.0004968706118355065,
"loss": 4.316,
"theoretical_loss": 5.2249271227812315,
"tokens_seen": 53477376
},
{
"epoch": 0.02,
"learning_rate": 0.0004968605817452358,
"loss": 4.0596,
"theoretical_loss": 5.223961595585755,
"tokens_seen": 53542912
},
{
"epoch": 0.02,
"learning_rate": 0.0004968505516549649,
"loss": 4.3552,
"theoretical_loss": 5.222997579905204,
"tokens_seen": 53608448
},
{
"epoch": 0.02,
"learning_rate": 0.0004968405215646941,
"loss": 4.435,
"theoretical_loss": 5.222035071529845,
"tokens_seen": 53673984
},
{
"epoch": 0.02,
"learning_rate": 0.0004968304914744233,
"loss": 4.4653,
"theoretical_loss": 5.2210740662667945,
"tokens_seen": 53739520
},
{
"epoch": 0.02,
"learning_rate": 0.0004968204613841525,
"loss": 4.4802,
"theoretical_loss": 5.220114559939923,
"tokens_seen": 53805056
},
{
"epoch": 0.02,
"learning_rate": 0.0004968104312938817,
"loss": 4.3459,
"theoretical_loss": 5.219156548389775,
"tokens_seen": 53870592
},
{
"epoch": 0.02,
"learning_rate": 0.0004968004012036108,
"loss": 4.4126,
"theoretical_loss": 5.218200027473481,
"tokens_seen": 53936128
},
{
"epoch": 0.02,
"learning_rate": 0.00049679037111334,
"loss": 3.9646,
"theoretical_loss": 5.217244993064664,
"tokens_seen": 54001664
},
{
"epoch": 0.02,
"objective/train/docs_used": 42708,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.427297115325928,
"objective/train/theoretical_loss": 5.216291441053366,
"objective/train/tokens_used": 74527200,
"theoretical_loss": 5.216291441053366,
"tokens_seen": 54067200
},
{
"epoch": 0.02,
"learning_rate": 0.0004967803410230692,
"loss": 4.2018,
"theoretical_loss": 5.216291441053366,
"tokens_seen": 54067200
},
{
"epoch": 0.02,
"learning_rate": 0.0004967703109327984,
"loss": 4.5827,
"theoretical_loss": 5.215339367345955,
"tokens_seen": 54132736
},
{
"epoch": 0.02,
"learning_rate": 0.0004967602808425276,
"loss": 4.4737,
"theoretical_loss": 5.214388767865036,
"tokens_seen": 54198272
},
{
"epoch": 0.02,
"learning_rate": 0.0004967502507522568,
"loss": 4.2806,
"theoretical_loss": 5.2134396385493815,
"tokens_seen": 54263808
},
{
"epoch": 0.02,
"learning_rate": 0.0004967402206619859,
"loss": 4.5709,
"theoretical_loss": 5.212491975353835,
"tokens_seen": 54329344
},
{
"epoch": 0.02,
"learning_rate": 0.0004967301905717151,
"loss": 4.2733,
"theoretical_loss": 5.211545774249233,
"tokens_seen": 54394880
},
{
"epoch": 0.02,
"learning_rate": 0.0004967201604814444,
"loss": 4.1417,
"theoretical_loss": 5.210601031222324,
"tokens_seen": 54460416
},
{
"epoch": 0.02,
"learning_rate": 0.0004967101303911735,
"loss": 4.3607,
"theoretical_loss": 5.209657742275683,
"tokens_seen": 54525952
},
{
"epoch": 0.02,
"learning_rate": 0.0004967001003009028,
"loss": 4.3571,
"theoretical_loss": 5.208715903427631,
"tokens_seen": 54591488
},
{
"epoch": 0.02,
"learning_rate": 0.000496690070210632,
"loss": 4.1888,
"theoretical_loss": 5.207775510712159,
"tokens_seen": 54657024
},
{
"epoch": 0.02,
"learning_rate": 0.0004966800401203611,
"loss": 4.3009,
"theoretical_loss": 5.2068365601788384,
"tokens_seen": 54722560
},
{
"epoch": 0.02,
"learning_rate": 0.0004966700100300903,
"loss": 4.217,
"theoretical_loss": 5.205899047892753,
"tokens_seen": 54788096
},
{
"epoch": 0.02,
"learning_rate": 0.0004966599799398195,
"loss": 4.2894,
"theoretical_loss": 5.2049629699344075,
"tokens_seen": 54853632
},
{
"epoch": 0.02,
"learning_rate": 0.0004966499498495487,
"loss": 4.1144,
"theoretical_loss": 5.204028322399658,
"tokens_seen": 54919168
},
{
"epoch": 0.02,
"learning_rate": 0.0004966399197592779,
"loss": 4.0488,
"theoretical_loss": 5.203095101399628,
"tokens_seen": 54984704
},
{
"epoch": 0.02,
"learning_rate": 0.000496629889669007,
"loss": 4.272,
"theoretical_loss": 5.202163303060633,
"tokens_seen": 55050240
},
{
"epoch": 0.02,
"learning_rate": 0.0004966198595787362,
"loss": 4.055,
"theoretical_loss": 5.201232923524104,
"tokens_seen": 55115776
},
{
"epoch": 0.02,
"learning_rate": 0.0004966098294884654,
"loss": 4.5155,
"theoretical_loss": 5.20030395894651,
"tokens_seen": 55181312
},
{
"epoch": 0.02,
"learning_rate": 0.0004965997993981946,
"loss": 4.467,
"theoretical_loss": 5.199376405499277,
"tokens_seen": 55246848
},
{
"epoch": 0.02,
"learning_rate": 0.0004965897693079238,
"loss": 4.1735,
"theoretical_loss": 5.198450259368721,
"tokens_seen": 55312384
},
{
"epoch": 0.02,
"learning_rate": 0.0004965797392176529,
"loss": 4.4801,
"theoretical_loss": 5.197525516755965,
"tokens_seen": 55377920
},
{
"epoch": 0.02,
"learning_rate": 0.0004965697091273821,
"loss": 4.0844,
"theoretical_loss": 5.196602173876867,
"tokens_seen": 55443456
},
{
"epoch": 0.02,
"learning_rate": 0.0004965596790371113,
"loss": 4.2085,
"theoretical_loss": 5.195680226961947,
"tokens_seen": 55508992
},
{
"epoch": 0.02,
"learning_rate": 0.0004965496489468405,
"loss": 4.1159,
"theoretical_loss": 5.194759672256309,
"tokens_seen": 55574528
},
{
"epoch": 0.02,
"learning_rate": 0.0004965396188565698,
"loss": 4.256,
"theoretical_loss": 5.19384050601957,
"tokens_seen": 55640064
},
{
"epoch": 0.02,
"objective/train/docs_used": 44116,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.755551815032959,
"objective/train/theoretical_loss": 5.192922724525789,
"objective/train/tokens_used": 76165600,
"theoretical_loss": 5.192922724525789,
"tokens_seen": 55705600
},
{
"epoch": 0.02,
"learning_rate": 0.0004965295887662988,
"loss": 4.1988,
"theoretical_loss": 5.192922724525789,
"tokens_seen": 55705600
},
{
"epoch": 0.02,
"learning_rate": 0.0004965195586760281,
"loss": 4.4217,
"theoretical_loss": 5.19200632406339,
"tokens_seen": 55771136
},
{
"epoch": 0.02,
"learning_rate": 0.0004965095285857573,
"loss": 4.2675,
"theoretical_loss": 5.19109130093509,
"tokens_seen": 55836672
},
{
"epoch": 0.02,
"learning_rate": 0.0004964994984954865,
"loss": 4.1937,
"theoretical_loss": 5.190177651457833,
"tokens_seen": 55902208
},
{
"epoch": 0.02,
"learning_rate": 0.0004964894684052157,
"loss": 4.4199,
"theoretical_loss": 5.189265371962712,
"tokens_seen": 55967744
},
{
"epoch": 0.02,
"learning_rate": 0.0004964794383149449,
"loss": 3.8862,
"theoretical_loss": 5.188354458794902,
"tokens_seen": 56033280
},
{
"epoch": 0.02,
"learning_rate": 0.000496469408224674,
"loss": 4.3709,
"theoretical_loss": 5.187444908313586,
"tokens_seen": 56098816
},
{
"epoch": 0.02,
"learning_rate": 0.0004964593781344032,
"loss": 3.9035,
"theoretical_loss": 5.186536716891892,
"tokens_seen": 56164352
},
{
"epoch": 0.02,
"learning_rate": 0.0004964493480441324,
"loss": 4.1296,
"theoretical_loss": 5.185629880916814,
"tokens_seen": 56229888
},
{
"epoch": 0.02,
"learning_rate": 0.0004964393179538616,
"loss": 4.0278,
"theoretical_loss": 5.18472439678915,
"tokens_seen": 56295424
},
{
"epoch": 0.02,
"learning_rate": 0.0004964292878635908,
"loss": 4.1566,
"theoretical_loss": 5.18382026092343,
"tokens_seen": 56360960
},
{
"epoch": 0.02,
"learning_rate": 0.00049641925777332,
"loss": 4.2582,
"theoretical_loss": 5.182917469747851,
"tokens_seen": 56426496
},
{
"epoch": 0.02,
"learning_rate": 0.0004964092276830491,
"loss": 3.9274,
"theoretical_loss": 5.182016019704204,
"tokens_seen": 56492032
},
{
"epoch": 0.02,
"learning_rate": 0.0004963991975927783,
"loss": 4.1733,
"theoretical_loss": 5.1811159072478095,
"tokens_seen": 56557568
},
{
"epoch": 0.02,
"learning_rate": 0.0004963891675025075,
"loss": 4.1092,
"theoretical_loss": 5.180217128847451,
"tokens_seen": 56623104
},
{
"epoch": 0.02,
"learning_rate": 0.0004963791374122367,
"loss": 3.9516,
"theoretical_loss": 5.17931968098531,
"tokens_seen": 56688640
},
{
"epoch": 0.02,
"learning_rate": 0.0004963691073219659,
"loss": 4.113,
"theoretical_loss": 5.178423560156894,
"tokens_seen": 56754176
},
{
"epoch": 0.02,
"learning_rate": 0.0004963590772316951,
"loss": 4.232,
"theoretical_loss": 5.177528762870973,
"tokens_seen": 56819712
},
{
"epoch": 0.02,
"learning_rate": 0.0004963490471414242,
"loss": 4.2272,
"theoretical_loss": 5.176635285649521,
"tokens_seen": 56885248
},
{
"epoch": 0.02,
"learning_rate": 0.0004963390170511535,
"loss": 4.1245,
"theoretical_loss": 5.175743125027638,
"tokens_seen": 56950784
},
{
"epoch": 0.02,
"learning_rate": 0.0004963289869608827,
"loss": 4.2108,
"theoretical_loss": 5.174852277553498,
"tokens_seen": 57016320
},
{
"epoch": 0.02,
"learning_rate": 0.0004963189568706119,
"loss": 4.4959,
"theoretical_loss": 5.173962739788276,
"tokens_seen": 57081856
},
{
"epoch": 0.02,
"learning_rate": 0.000496308926780341,
"loss": 4.3544,
"theoretical_loss": 5.17307450830609,
"tokens_seen": 57147392
},
{
"epoch": 0.02,
"learning_rate": 0.0004962988966900702,
"loss": 4.1915,
"theoretical_loss": 5.172187579693933,
"tokens_seen": 57212928
},
{
"epoch": 0.02,
"learning_rate": 0.0004962888665997994,
"loss": 4.1823,
"theoretical_loss": 5.1713019505516105,
"tokens_seen": 57278464
},
{
"epoch": 0.02,
"objective/train/docs_used": 44698,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.967449188232422,
"objective/train/theoretical_loss": 5.170417617491682,
"objective/train/tokens_used": 77804000,
"theoretical_loss": 5.170417617491682,
"tokens_seen": 57344000
},
{
"epoch": 0.02,
"learning_rate": 0.0004962788365095286,
"loss": 4.3458,
"theoretical_loss": 5.170417617491682,
"tokens_seen": 57344000
},
{
"epoch": 0.02,
"learning_rate": 0.0004962688064192578,
"loss": 4.5212,
"theoretical_loss": 5.169534577139395,
"tokens_seen": 57409536
},
{
"epoch": 0.02,
"learning_rate": 0.000496258776328987,
"loss": 4.2458,
"theoretical_loss": 5.168652826132623,
"tokens_seen": 57475072
},
{
"epoch": 0.02,
"learning_rate": 0.0004962487462387161,
"loss": 4.1007,
"theoretical_loss": 5.167772361121805,
"tokens_seen": 57540608
},
{
"epoch": 0.02,
"learning_rate": 0.0004962387161484453,
"loss": 4.2059,
"theoretical_loss": 5.166893178769884,
"tokens_seen": 57606144
},
{
"epoch": 0.02,
"learning_rate": 0.0004962286860581746,
"loss": 4.1405,
"theoretical_loss": 5.1660152757522475,
"tokens_seen": 57671680
},
{
"epoch": 0.02,
"learning_rate": 0.0004962186559679037,
"loss": 4.0437,
"theoretical_loss": 5.165138648756665,
"tokens_seen": 57737216
},
{
"epoch": 0.02,
"learning_rate": 0.000496208625877633,
"loss": 3.8546,
"theoretical_loss": 5.164263294483226,
"tokens_seen": 57802752
},
{
"epoch": 0.02,
"learning_rate": 0.0004961985957873621,
"loss": 3.9834,
"theoretical_loss": 5.163389209644287,
"tokens_seen": 57868288
},
{
"epoch": 0.02,
"learning_rate": 0.0004961885656970913,
"loss": 3.9333,
"theoretical_loss": 5.162516390964408,
"tokens_seen": 57933824
},
{
"epoch": 0.02,
"learning_rate": 0.0004961785356068205,
"loss": 4.0711,
"theoretical_loss": 5.1616448351802875,
"tokens_seen": 57999360
},
{
"epoch": 0.02,
"learning_rate": 0.0004961685055165497,
"loss": 4.2007,
"theoretical_loss": 5.160774539040716,
"tokens_seen": 58064896
},
{
"epoch": 0.02,
"learning_rate": 0.0004961584754262789,
"loss": 4.1938,
"theoretical_loss": 5.159905499306511,
"tokens_seen": 58130432
},
{
"epoch": 0.02,
"learning_rate": 0.0004961484453360081,
"loss": 4.0338,
"theoretical_loss": 5.159037712750455,
"tokens_seen": 58195968
},
{
"epoch": 0.02,
"learning_rate": 0.0004961384152457372,
"loss": 4.1705,
"theoretical_loss": 5.158171176157245,
"tokens_seen": 58261504
},
{
"epoch": 0.02,
"learning_rate": 0.0004961283851554664,
"loss": 4.2802,
"theoretical_loss": 5.157305886323435,
"tokens_seen": 58327040
},
{
"epoch": 0.02,
"learning_rate": 0.0004961183550651956,
"loss": 4.2534,
"theoretical_loss": 5.156441840057371,
"tokens_seen": 58392576
},
{
"epoch": 0.02,
"learning_rate": 0.0004961083249749248,
"loss": 4.072,
"theoretical_loss": 5.155579034179144,
"tokens_seen": 58458112
},
{
"epoch": 0.02,
"learning_rate": 0.000496098294884654,
"loss": 4.2144,
"theoretical_loss": 5.15471746552053,
"tokens_seen": 58523648
},
{
"epoch": 0.02,
"learning_rate": 0.0004960882647943831,
"loss": 4.1997,
"theoretical_loss": 5.153857130924929,
"tokens_seen": 58589184
},
{
"epoch": 0.02,
"learning_rate": 0.0004960782347041123,
"loss": 4.2174,
"theoretical_loss": 5.1529980272473175,
"tokens_seen": 58654720
},
{
"epoch": 0.02,
"learning_rate": 0.0004960682046138415,
"loss": 4.3491,
"theoretical_loss": 5.152140151354191,
"tokens_seen": 58720256
},
{
"epoch": 0.02,
"learning_rate": 0.0004960581745235707,
"loss": 4.1085,
"theoretical_loss": 5.151283500123505,
"tokens_seen": 58785792
},
{
"epoch": 0.02,
"learning_rate": 0.0004960481444333,
"loss": 4.1087,
"theoretical_loss": 5.150428070444621,
"tokens_seen": 58851328
},
{
"epoch": 0.02,
"learning_rate": 0.000496038114343029,
"loss": 4.0075,
"theoretical_loss": 5.149573859218261,
"tokens_seen": 58916864
},
{
"epoch": 0.02,
"objective/train/docs_used": 46033,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.9614341259002686,
"objective/train/theoretical_loss": 5.1487208633564405,
"objective/train/tokens_used": 79442400,
"theoretical_loss": 5.1487208633564405,
"tokens_seen": 58982400
},
{
"epoch": 0.02,
"learning_rate": 0.0004960280842527583,
"loss": 4.125,
"theoretical_loss": 5.1487208633564405,
"tokens_seen": 58982400
},
{
"epoch": 0.02,
"learning_rate": 0.0004960180541624875,
"loss": 4.0149,
"theoretical_loss": 5.147869079782423,
"tokens_seen": 59047936
},
{
"epoch": 0.02,
"learning_rate": 0.0004960080240722167,
"loss": 4.2106,
"theoretical_loss": 5.147018505430666,
"tokens_seen": 59113472
},
{
"epoch": 0.02,
"learning_rate": 0.0004959979939819459,
"loss": 4.2445,
"theoretical_loss": 5.146169137246765,
"tokens_seen": 59179008
},
{
"epoch": 0.02,
"learning_rate": 0.0004959879638916751,
"loss": 4.2846,
"theoretical_loss": 5.145320972187402,
"tokens_seen": 59244544
},
{
"epoch": 0.02,
"learning_rate": 0.0004959779338014042,
"loss": 4.0537,
"theoretical_loss": 5.144474007220293,
"tokens_seen": 59310080
},
{
"epoch": 0.02,
"learning_rate": 0.0004959679037111334,
"loss": 4.3207,
"theoretical_loss": 5.143628239324139,
"tokens_seen": 59375616
},
{
"epoch": 0.02,
"learning_rate": 0.0004959578736208626,
"loss": 4.0034,
"theoretical_loss": 5.142783665488567,
"tokens_seen": 59441152
},
{
"epoch": 0.02,
"learning_rate": 0.0004959478435305918,
"loss": 3.8394,
"theoretical_loss": 5.1419402827140885,
"tokens_seen": 59506688
},
{
"epoch": 0.02,
"learning_rate": 0.000495937813440321,
"loss": 4.0548,
"theoretical_loss": 5.141098088012036,
"tokens_seen": 59572224
},
{
"epoch": 0.02,
"learning_rate": 0.0004959277833500501,
"loss": 3.6982,
"theoretical_loss": 5.140257078404524,
"tokens_seen": 59637760
},
{
"epoch": 0.02,
"learning_rate": 0.0004959177532597793,
"loss": 4.1571,
"theoretical_loss": 5.13941725092439,
"tokens_seen": 59703296
},
{
"epoch": 0.02,
"learning_rate": 0.0004959077231695085,
"loss": 4.1842,
"theoretical_loss": 5.138578602615146,
"tokens_seen": 59768832
},
{
"epoch": 0.02,
"learning_rate": 0.0004958976930792377,
"loss": 4.2649,
"theoretical_loss": 5.137741130530934,
"tokens_seen": 59834368
},
{
"epoch": 0.02,
"learning_rate": 0.0004958876629889669,
"loss": 4.1585,
"theoretical_loss": 5.1369048317364685,
"tokens_seen": 59899904
},
{
"epoch": 0.02,
"learning_rate": 0.000495877632898696,
"loss": 4.1038,
"theoretical_loss": 5.13606970330699,
"tokens_seen": 59965440
},
{
"epoch": 0.02,
"learning_rate": 0.0004958676028084253,
"loss": 4.0452,
"theoretical_loss": 5.135235742328217,
"tokens_seen": 60030976
},
{
"epoch": 0.02,
"learning_rate": 0.0004958575727181544,
"loss": 4.2665,
"theoretical_loss": 5.134402945896297,
"tokens_seen": 60096512
},
{
"epoch": 0.02,
"learning_rate": 0.0004958475426278837,
"loss": 4.0482,
"theoretical_loss": 5.133571311117755,
"tokens_seen": 60162048
},
{
"epoch": 0.02,
"learning_rate": 0.0004958375125376129,
"loss": 4.1642,
"theoretical_loss": 5.132740835109448,
"tokens_seen": 60227584
},
{
"epoch": 0.02,
"learning_rate": 0.0004958274824473421,
"loss": 4.2918,
"theoretical_loss": 5.131911514998518,
"tokens_seen": 60293120
},
{
"epoch": 0.02,
"learning_rate": 0.0004958174523570712,
"loss": 4.0217,
"theoretical_loss": 5.131083347922338,
"tokens_seen": 60358656
},
{
"epoch": 0.02,
"learning_rate": 0.0004958074222668004,
"loss": 4.0733,
"theoretical_loss": 5.130256331028474,
"tokens_seen": 60424192
},
{
"epoch": 0.02,
"learning_rate": 0.0004957973921765296,
"loss": 4.5194,
"theoretical_loss": 5.129430461474628,
"tokens_seen": 60489728
},
{
"epoch": 0.02,
"learning_rate": 0.0004957873620862588,
"loss": 4.0056,
"theoretical_loss": 5.128605736428597,
"tokens_seen": 60555264
},
{
"epoch": 0.02,
"objective/train/docs_used": 46630,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.424596786499023,
"objective/train/theoretical_loss": 5.127782153068225,
"objective/train/tokens_used": 81080800,
"theoretical_loss": 5.127782153068225,
"tokens_seen": 60620800
},
{
"epoch": 0.02,
"learning_rate": 0.000495777331995988,
"loss": 4.4411,
"theoretical_loss": 5.127782153068225,
"tokens_seen": 60620800
},
{
"epoch": 0.02,
"learning_rate": 0.0004957673019057172,
"loss": 3.919,
"theoretical_loss": 5.126959708581356,
"tokens_seen": 60686336
},
{
"epoch": 0.02,
"learning_rate": 0.0004957572718154463,
"loss": 4.032,
"theoretical_loss": 5.1261384001657895,
"tokens_seen": 60751872
},
{
"epoch": 0.02,
"learning_rate": 0.0004957472417251755,
"loss": 4.1885,
"theoretical_loss": 5.125318225029231,
"tokens_seen": 60817408
},
{
"epoch": 0.02,
"learning_rate": 0.0004957372116349047,
"loss": 4.0579,
"theoretical_loss": 5.124499180389249,
"tokens_seen": 60882944
},
{
"epoch": 0.02,
"learning_rate": 0.0004957271815446339,
"loss": 3.8165,
"theoretical_loss": 5.12368126347323,
"tokens_seen": 60948480
},
{
"epoch": 0.02,
"learning_rate": 0.0004957171514543631,
"loss": 3.8252,
"theoretical_loss": 5.122864471518334,
"tokens_seen": 61014016
},
{
"epoch": 0.02,
"learning_rate": 0.0004957071213640923,
"loss": 4.0573,
"theoretical_loss": 5.122048801771443,
"tokens_seen": 61079552
},
{
"epoch": 0.02,
"learning_rate": 0.0004956970912738214,
"loss": 4.0186,
"theoretical_loss": 5.121234251489128,
"tokens_seen": 61145088
},
{
"epoch": 0.02,
"learning_rate": 0.0004956870611835507,
"loss": 4.0424,
"theoretical_loss": 5.120420817937591,
"tokens_seen": 61210624
},
{
"epoch": 0.02,
"learning_rate": 0.0004956770310932798,
"loss": 4.2235,
"theoretical_loss": 5.119608498392633,
"tokens_seen": 61276160
},
{
"epoch": 0.02,
"learning_rate": 0.0004956670010030091,
"loss": 3.7588,
"theoretical_loss": 5.118797290139605,
"tokens_seen": 61341696
},
{
"epoch": 0.02,
"learning_rate": 0.0004956569709127383,
"loss": 4.0559,
"theoretical_loss": 5.117987190473361,
"tokens_seen": 61407232
},
{
"epoch": 0.02,
"learning_rate": 0.0004956469408224674,
"loss": 4.3216,
"theoretical_loss": 5.1171781966982195,
"tokens_seen": 61472768
},
{
"epoch": 0.02,
"learning_rate": 0.0004956369107321966,
"loss": 3.99,
"theoretical_loss": 5.116370306127921,
"tokens_seen": 61538304
},
{
"epoch": 0.02,
"learning_rate": 0.0004956268806419258,
"loss": 3.96,
"theoretical_loss": 5.11556351608558,
"tokens_seen": 61603840
},
{
"epoch": 0.02,
"learning_rate": 0.000495616850551655,
"loss": 3.9305,
"theoretical_loss": 5.114757823903647,
"tokens_seen": 61669376
},
{
"epoch": 0.02,
"learning_rate": 0.0004956068204613842,
"loss": 3.885,
"theoretical_loss": 5.113953226923864,
"tokens_seen": 61734912
},
{
"epoch": 0.02,
"learning_rate": 0.0004955967903711133,
"loss": 4.2558,
"theoretical_loss": 5.113149722497221,
"tokens_seen": 61800448
},
{
"epoch": 0.02,
"learning_rate": 0.0004955867602808425,
"loss": 4.0224,
"theoretical_loss": 5.112347307983919,
"tokens_seen": 61865984
},
{
"epoch": 0.02,
"learning_rate": 0.0004955767301905717,
"loss": 4.0078,
"theoretical_loss": 5.111545980753322,
"tokens_seen": 61931520
},
{
"epoch": 0.02,
"learning_rate": 0.0004955667001003009,
"loss": 4.2075,
"theoretical_loss": 5.110745738183919,
"tokens_seen": 61997056
},
{
"epoch": 0.02,
"learning_rate": 0.0004955566700100301,
"loss": 4.2822,
"theoretical_loss": 5.109946577663284,
"tokens_seen": 62062592
},
{
"epoch": 0.02,
"learning_rate": 0.0004955466399197592,
"loss": 3.8552,
"theoretical_loss": 5.109148496588032,
"tokens_seen": 62128128
},
{
"epoch": 0.02,
"learning_rate": 0.0004955366098294884,
"loss": 4.1157,
"theoretical_loss": 5.108351492363779,
"tokens_seen": 62193664
},
{
"epoch": 0.02,
"objective/train/docs_used": 47896,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.253730773925781,
"objective/train/theoretical_loss": 5.107555562405102,
"objective/train/tokens_used": 82719200,
"theoretical_loss": 5.107555562405102,
"tokens_seen": 62259200
},
{
"epoch": 0.02,
"learning_rate": 0.0004955265797392177,
"loss": 3.8586,
"theoretical_loss": 5.107555562405102,
"tokens_seen": 62259200
},
{
"epoch": 0.02,
"learning_rate": 0.0004955165496489468,
"loss": 4.043,
"theoretical_loss": 5.106760704135499,
"tokens_seen": 62324736
},
{
"epoch": 0.02,
"learning_rate": 0.0004955065195586761,
"loss": 3.9028,
"theoretical_loss": 5.105966914987349,
"tokens_seen": 62390272
},
{
"epoch": 0.02,
"learning_rate": 0.0004954964894684052,
"loss": 4.1462,
"theoretical_loss": 5.1051741924018685,
"tokens_seen": 62455808
},
{
"epoch": 0.02,
"learning_rate": 0.0004954864593781344,
"loss": 3.9346,
"theoretical_loss": 5.10438253382908,
"tokens_seen": 62521344
},
{
"epoch": 0.02,
"learning_rate": 0.0004954764292878636,
"loss": 4.1546,
"theoretical_loss": 5.103591936727762,
"tokens_seen": 62586880
},
{
"epoch": 0.02,
"learning_rate": 0.0004954663991975928,
"loss": 4.1114,
"theoretical_loss": 5.102802398565418,
"tokens_seen": 62652416
},
{
"epoch": 0.02,
"learning_rate": 0.000495456369107322,
"loss": 4.2456,
"theoretical_loss": 5.102013916818235,
"tokens_seen": 62717952
},
{
"epoch": 0.02,
"learning_rate": 0.0004954463390170512,
"loss": 4.1117,
"theoretical_loss": 5.101226488971042,
"tokens_seen": 62783488
},
{
"epoch": 0.02,
"learning_rate": 0.0004954363089267803,
"loss": 3.8726,
"theoretical_loss": 5.100440112517276,
"tokens_seen": 62849024
},
{
"epoch": 0.02,
"learning_rate": 0.0004954262788365095,
"loss": 3.9981,
"theoretical_loss": 5.09965478495894,
"tokens_seen": 62914560
},
{
"epoch": 0.02,
"learning_rate": 0.0004954162487462387,
"loss": 4.048,
"theoretical_loss": 5.098870503806567,
"tokens_seen": 62980096
},
{
"epoch": 0.02,
"learning_rate": 0.0004954062186559679,
"loss": 3.8959,
"theoretical_loss": 5.09808726657918,
"tokens_seen": 63045632
},
{
"epoch": 0.02,
"learning_rate": 0.0004953961885656971,
"loss": 3.7875,
"theoretical_loss": 5.097305070804255,
"tokens_seen": 63111168
},
{
"epoch": 0.02,
"learning_rate": 0.0004953861584754263,
"loss": 4.108,
"theoretical_loss": 5.096523914017688,
"tokens_seen": 63176704
},
{
"epoch": 0.02,
"learning_rate": 0.0004953761283851555,
"loss": 4.1691,
"theoretical_loss": 5.095743793763747,
"tokens_seen": 63242240
},
{
"epoch": 0.02,
"learning_rate": 0.0004953660982948846,
"loss": 4.0009,
"theoretical_loss": 5.094964707595047,
"tokens_seen": 63307776
},
{
"epoch": 0.02,
"learning_rate": 0.0004953560682046139,
"loss": 3.7485,
"theoretical_loss": 5.094186653072505,
"tokens_seen": 63373312
},
{
"epoch": 0.02,
"learning_rate": 0.0004953460381143431,
"loss": 4.0331,
"theoretical_loss": 5.093409627765306,
"tokens_seen": 63438848
},
{
"epoch": 0.02,
"learning_rate": 0.0004953360080240723,
"loss": 3.9678,
"theoretical_loss": 5.092633629250866,
"tokens_seen": 63504384
},
{
"epoch": 0.02,
"learning_rate": 0.0004953259779338014,
"loss": 4.031,
"theoretical_loss": 5.091858655114796,
"tokens_seen": 63569920
},
{
"epoch": 0.02,
"learning_rate": 0.0004953159478435306,
"loss": 4.0263,
"theoretical_loss": 5.091084702950868,
"tokens_seen": 63635456
},
{
"epoch": 0.02,
"learning_rate": 0.0004953059177532598,
"loss": 3.9067,
"theoretical_loss": 5.090311770360971,
"tokens_seen": 63700992
},
{
"epoch": 0.02,
"learning_rate": 0.000495295887662989,
"loss": 4.0007,
"theoretical_loss": 5.089539854955088,
"tokens_seen": 63766528
},
{
"epoch": 0.02,
"learning_rate": 0.0004952858575727182,
"loss": 3.7926,
"theoretical_loss": 5.088768954351249,
"tokens_seen": 63832064
},
{
"epoch": 0.02,
"objective/train/docs_used": 48551,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.194922924041748,
"objective/train/theoretical_loss": 5.087999066175502,
"objective/train/tokens_used": 84357600,
"theoretical_loss": 5.087999066175502,
"tokens_seen": 63897600
},
{
"epoch": 0.02,
"learning_rate": 0.0004952758274824474,
"loss": 4.0708,
"theoretical_loss": 5.087999066175502,
"tokens_seen": 63897600
},
{
"epoch": 0.02,
"learning_rate": 0.0004952657973921765,
"loss": 4.289,
"theoretical_loss": 5.0872301880618735,
"tokens_seen": 63963136
},
{
"epoch": 0.02,
"learning_rate": 0.0004952557673019057,
"loss": 4.0031,
"theoretical_loss": 5.086462317652341,
"tokens_seen": 64028672
},
{
"epoch": 0.02,
"learning_rate": 0.0004952457372116349,
"loss": 3.7044,
"theoretical_loss": 5.085695452596788,
"tokens_seen": 64094208
},
{
"epoch": 0.02,
"learning_rate": 0.0004952357071213641,
"loss": 3.8383,
"theoretical_loss": 5.084929590552976,
"tokens_seen": 64159744
},
{
"epoch": 0.02,
"learning_rate": 0.0004952256770310933,
"loss": 3.8943,
"theoretical_loss": 5.0841647291865115,
"tokens_seen": 64225280
},
{
"epoch": 0.02,
"learning_rate": 0.0004952156469408225,
"loss": 4.0131,
"theoretical_loss": 5.083400866170806,
"tokens_seen": 64290816
},
{
"epoch": 0.02,
"learning_rate": 0.0004952056168505516,
"loss": 3.8292,
"theoretical_loss": 5.082637999187046,
"tokens_seen": 64356352
},
{
"epoch": 0.02,
"learning_rate": 0.0004951955867602809,
"loss": 3.8947,
"theoretical_loss": 5.081876125924159,
"tokens_seen": 64421888
},
{
"epoch": 0.02,
"learning_rate": 0.00049518555667001,
"loss": 3.8301,
"theoretical_loss": 5.0811152440787755,
"tokens_seen": 64487424
},
{
"epoch": 0.02,
"learning_rate": 0.0004951755265797393,
"loss": 3.9172,
"theoretical_loss": 5.0803553513552036,
"tokens_seen": 64552960
},
{
"epoch": 0.02,
"learning_rate": 0.0004951654964894685,
"loss": 3.9521,
"theoretical_loss": 5.079596445465386,
"tokens_seen": 64618496
},
{
"epoch": 0.02,
"learning_rate": 0.0004951554663991976,
"loss": 3.874,
"theoretical_loss": 5.078838524128878,
"tokens_seen": 64684032
},
{
"epoch": 0.02,
"learning_rate": 0.0004951454363089268,
"loss": 3.886,
"theoretical_loss": 5.078081585072802,
"tokens_seen": 64749568
},
{
"epoch": 0.02,
"learning_rate": 0.000495135406218656,
"loss": 3.9186,
"theoretical_loss": 5.077325626031826,
"tokens_seen": 64815104
},
{
"epoch": 0.02,
"learning_rate": 0.0004951253761283852,
"loss": 3.8648,
"theoretical_loss": 5.076570644748123,
"tokens_seen": 64880640
},
{
"epoch": 0.02,
"learning_rate": 0.0004951153460381144,
"loss": 4.1708,
"theoretical_loss": 5.075816638971341,
"tokens_seen": 64946176
},
{
"epoch": 0.02,
"learning_rate": 0.0004951053159478435,
"loss": 3.8125,
"theoretical_loss": 5.075063606458576,
"tokens_seen": 65011712
},
{
"epoch": 0.02,
"learning_rate": 0.0004950952858575727,
"loss": 3.7121,
"theoretical_loss": 5.074311544974331,
"tokens_seen": 65077248
},
{
"epoch": 0.02,
"learning_rate": 0.0004950852557673019,
"loss": 3.7576,
"theoretical_loss": 5.07356045229049,
"tokens_seen": 65142784
},
{
"epoch": 0.02,
"learning_rate": 0.0004950752256770311,
"loss": 3.9499,
"theoretical_loss": 5.072810326186285,
"tokens_seen": 65208320
},
{
"epoch": 0.02,
"learning_rate": 0.0004950651955867603,
"loss": 3.7499,
"theoretical_loss": 5.072061164448261,
"tokens_seen": 65273856
},
{
"epoch": 0.02,
"learning_rate": 0.0004950551654964894,
"loss": 3.833,
"theoretical_loss": 5.071312964870252,
"tokens_seen": 65339392
},
{
"epoch": 0.02,
"learning_rate": 0.0004950451354062186,
"loss": 4.1678,
"theoretical_loss": 5.070565725253344,
"tokens_seen": 65404928
},
{
"epoch": 0.02,
"learning_rate": 0.0004950351053159479,
"loss": 3.7484,
"theoretical_loss": 5.069819443405842,
"tokens_seen": 65470464
},
{
"epoch": 0.02,
"objective/train/docs_used": 49128,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.299447536468506,
"objective/train/theoretical_loss": 5.069074117143246,
"objective/train/tokens_used": 85996000,
"theoretical_loss": 5.069074117143246,
"tokens_seen": 65536000
},
{
"epoch": 0.02,
"learning_rate": 0.000495025075225677,
"loss": 4.2265,
"theoretical_loss": 5.069074117143246,
"tokens_seen": 65536000
},
{
"epoch": 0.02,
"learning_rate": 0.0004950150451354063,
"loss": 3.8292,
"theoretical_loss": 5.068329744288216,
"tokens_seen": 65601536
},
{
"epoch": 0.02,
"learning_rate": 0.0004950050150451354,
"loss": 3.9529,
"theoretical_loss": 5.067586322670541,
"tokens_seen": 65667072
},
{
"epoch": 0.02,
"learning_rate": 0.0004949949849548646,
"loss": 3.9573,
"theoretical_loss": 5.0668438501271105,
"tokens_seen": 65732608
},
{
"epoch": 0.02,
"learning_rate": 0.0004949849548645938,
"loss": 3.8014,
"theoretical_loss": 5.066102324501883,
"tokens_seen": 65798144
},
{
"epoch": 0.02,
"learning_rate": 0.000494974924774323,
"loss": 4.0132,
"theoretical_loss": 5.065361743645855,
"tokens_seen": 65863680
},
{
"epoch": 0.02,
"learning_rate": 0.0004949648946840522,
"loss": 3.8756,
"theoretical_loss": 5.064622105417033,
"tokens_seen": 65929216
},
{
"epoch": 0.02,
"learning_rate": 0.0004949548645937814,
"loss": 3.938,
"theoretical_loss": 5.063883407680405,
"tokens_seen": 65994752
},
{
"epoch": 0.02,
"learning_rate": 0.0004949448345035105,
"loss": 3.7738,
"theoretical_loss": 5.063145648307904,
"tokens_seen": 66060288
},
{
"epoch": 0.02,
"learning_rate": 0.0004949348044132397,
"loss": 3.8514,
"theoretical_loss": 5.062408825178388,
"tokens_seen": 66125824
},
{
"epoch": 0.02,
"learning_rate": 0.0004949247743229689,
"loss": 3.8407,
"theoretical_loss": 5.061672936177604,
"tokens_seen": 66191360
},
{
"epoch": 0.02,
"learning_rate": 0.0004949147442326981,
"loss": 3.949,
"theoretical_loss": 5.06093797919816,
"tokens_seen": 66256896
},
{
"epoch": 0.02,
"learning_rate": 0.0004949047141424273,
"loss": 3.9103,
"theoretical_loss": 5.060203952139497,
"tokens_seen": 66322432
},
{
"epoch": 0.02,
"learning_rate": 0.0004948946840521565,
"loss": 3.7828,
"theoretical_loss": 5.059470852907861,
"tokens_seen": 66387968
},
{
"epoch": 0.02,
"learning_rate": 0.0004948846539618856,
"loss": 4.0738,
"theoretical_loss": 5.0587386794162725,
"tokens_seen": 66453504
},
{
"epoch": 0.02,
"learning_rate": 0.0004948746238716148,
"loss": 3.7987,
"theoretical_loss": 5.058007429584498,
"tokens_seen": 66519040
},
{
"epoch": 0.02,
"learning_rate": 0.000494864593781344,
"loss": 4.0708,
"theoretical_loss": 5.057277101339023,
"tokens_seen": 66584576
},
{
"epoch": 0.02,
"learning_rate": 0.0004948545636910733,
"loss": 3.9817,
"theoretical_loss": 5.056547692613021,
"tokens_seen": 66650112
},
{
"epoch": 0.02,
"learning_rate": 0.0004948445336008024,
"loss": 3.9141,
"theoretical_loss": 5.055819201346331,
"tokens_seen": 66715648
},
{
"epoch": 0.02,
"learning_rate": 0.0004948345035105316,
"loss": 3.7643,
"theoretical_loss": 5.055091625485421,
"tokens_seen": 66781184
},
{
"epoch": 0.02,
"learning_rate": 0.0004948244734202607,
"loss": 3.9492,
"theoretical_loss": 5.054364962983367,
"tokens_seen": 66846720
},
{
"epoch": 0.02,
"learning_rate": 0.00049481444332999,
"loss": 3.7609,
"theoretical_loss": 5.053639211799824,
"tokens_seen": 66912256
},
{
"epoch": 0.02,
"learning_rate": 0.0004948044132397192,
"loss": 4.1554,
"theoretical_loss": 5.052914369900997,
"tokens_seen": 66977792
},
{
"epoch": 0.02,
"learning_rate": 0.0004947943831494484,
"loss": 3.6762,
"theoretical_loss": 5.052190435259614,
"tokens_seen": 67043328
},
{
"epoch": 0.02,
"learning_rate": 0.0004947843530591776,
"loss": 3.8481,
"theoretical_loss": 5.051467405854897,
"tokens_seen": 67108864
},
{
"epoch": 0.02,
"objective/train/docs_used": 50658,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4438347816467285,
"objective/train/theoretical_loss": 5.05074527967254,
"objective/train/tokens_used": 87634400,
"theoretical_loss": 5.05074527967254,
"tokens_seen": 67174400
},
{
"epoch": 0.02,
"learning_rate": 0.0004947743229689067,
"loss": 3.7927,
"theoretical_loss": 5.05074527967254,
"tokens_seen": 67174400
},
{
"epoch": 0.02,
"learning_rate": 0.0004947642928786359,
"loss": 3.9986,
"theoretical_loss": 5.050024054704677,
"tokens_seen": 67239936
},
{
"epoch": 0.02,
"learning_rate": 0.0004947542627883651,
"loss": 3.8366,
"theoretical_loss": 5.049303728949859,
"tokens_seen": 67305472
},
{
"epoch": 0.02,
"learning_rate": 0.0004947442326980943,
"loss": 3.6166,
"theoretical_loss": 5.048584300413019,
"tokens_seen": 67371008
},
{
"epoch": 0.02,
"learning_rate": 0.0004947342026078235,
"loss": 3.666,
"theoretical_loss": 5.04786576710546,
"tokens_seen": 67436544
},
{
"epoch": 0.02,
"learning_rate": 0.0004947241725175527,
"loss": 3.6312,
"theoretical_loss": 5.0471481270448155,
"tokens_seen": 67502080
},
{
"epoch": 0.02,
"learning_rate": 0.0004947141424272818,
"loss": 3.6529,
"theoretical_loss": 5.046431378255027,
"tokens_seen": 67567616
},
{
"epoch": 0.02,
"learning_rate": 0.0004947041123370111,
"loss": 4.1165,
"theoretical_loss": 5.045715518766322,
"tokens_seen": 67633152
},
{
"epoch": 0.02,
"learning_rate": 0.0004946940822467402,
"loss": 3.9446,
"theoretical_loss": 5.0450005466151815,
"tokens_seen": 67698688
},
{
"epoch": 0.02,
"learning_rate": 0.0004946840521564695,
"loss": 3.916,
"theoretical_loss": 5.044286459844319,
"tokens_seen": 67764224
},
{
"epoch": 0.02,
"learning_rate": 0.0004946740220661987,
"loss": 4.0487,
"theoretical_loss": 5.043573256502652,
"tokens_seen": 67829760
},
{
"epoch": 0.02,
"learning_rate": 0.0004946639919759278,
"loss": 3.751,
"theoretical_loss": 5.0428609346452795,
"tokens_seen": 67895296
},
{
"epoch": 0.02,
"learning_rate": 0.000494653961885657,
"loss": 4.055,
"theoretical_loss": 5.042149492333452,
"tokens_seen": 67960832
},
{
"epoch": 0.02,
"learning_rate": 0.0004946439317953862,
"loss": 3.8691,
"theoretical_loss": 5.041438927634549,
"tokens_seen": 68026368
},
{
"epoch": 0.02,
"learning_rate": 0.0004946339017051154,
"loss": 3.9087,
"theoretical_loss": 5.040729238622053,
"tokens_seen": 68091904
},
{
"epoch": 0.02,
"learning_rate": 0.0004946238716148446,
"loss": 3.9294,
"theoretical_loss": 5.040020423375525,
"tokens_seen": 68157440
},
{
"epoch": 0.02,
"learning_rate": 0.0004946138415245737,
"loss": 3.8602,
"theoretical_loss": 5.039312479980579,
"tokens_seen": 68222976
},
{
"epoch": 0.02,
"learning_rate": 0.0004946038114343029,
"loss": 3.9095,
"theoretical_loss": 5.038605406528857,
"tokens_seen": 68288512
},
{
"epoch": 0.02,
"learning_rate": 0.0004945937813440321,
"loss": 4.2147,
"theoretical_loss": 5.037899201118005,
"tokens_seen": 68354048
},
{
"epoch": 0.02,
"learning_rate": 0.0004945837512537613,
"loss": 3.6029,
"theoretical_loss": 5.037193861851646,
"tokens_seen": 68419584
},
{
"epoch": 0.02,
"learning_rate": 0.0004945737211634905,
"loss": 3.8454,
"theoretical_loss": 5.03648938683936,
"tokens_seen": 68485120
},
{
"epoch": 0.02,
"learning_rate": 0.0004945636910732196,
"loss": 3.9214,
"theoretical_loss": 5.035785774196654,
"tokens_seen": 68550656
},
{
"epoch": 0.02,
"learning_rate": 0.0004945536609829488,
"loss": 3.6581,
"theoretical_loss": 5.035083022044944,
"tokens_seen": 68616192
},
{
"epoch": 0.02,
"learning_rate": 0.0004945436308926781,
"loss": 3.6618,
"theoretical_loss": 5.034381128511525,
"tokens_seen": 68681728
},
{
"epoch": 0.02,
"learning_rate": 0.0004945336008024072,
"loss": 3.7457,
"theoretical_loss": 5.0336800917295506,
"tokens_seen": 68747264
},
{
"epoch": 0.02,
"objective/train/docs_used": 51428,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.8635945320129395,
"objective/train/theoretical_loss": 5.032979909838007,
"objective/train/tokens_used": 89272800,
"theoretical_loss": 5.032979909838007,
"tokens_seen": 68812800
},
{
"epoch": 0.02,
"learning_rate": 0.0004945235707121365,
"loss": 3.842,
"theoretical_loss": 5.032979909838007,
"tokens_seen": 68812800
},
{
"epoch": 0.02,
"learning_rate": 0.0004945135406218656,
"loss": 4.0085,
"theoretical_loss": 5.032280580981691,
"tokens_seen": 68878336
},
{
"epoch": 0.02,
"learning_rate": 0.0004945035105315948,
"loss": 3.8896,
"theoretical_loss": 5.031582103311187,
"tokens_seen": 68943872
},
{
"epoch": 0.02,
"learning_rate": 0.000494493480441324,
"loss": 3.8326,
"theoretical_loss": 5.030884474982842,
"tokens_seen": 69009408
},
{
"epoch": 0.02,
"learning_rate": 0.0004944834503510532,
"loss": 3.8144,
"theoretical_loss": 5.030187694158739,
"tokens_seen": 69074944
},
{
"epoch": 0.02,
"learning_rate": 0.0004944734202607824,
"loss": 3.797,
"theoretical_loss": 5.02949175900668,
"tokens_seen": 69140480
},
{
"epoch": 0.02,
"learning_rate": 0.0004944633901705116,
"loss": 3.9856,
"theoretical_loss": 5.028796667700159,
"tokens_seen": 69206016
},
{
"epoch": 0.02,
"learning_rate": 0.0004944533600802407,
"loss": 3.8468,
"theoretical_loss": 5.0281024184183405,
"tokens_seen": 69271552
},
{
"epoch": 0.02,
"learning_rate": 0.0004944433299899699,
"loss": 3.706,
"theoretical_loss": 5.0274090093460355,
"tokens_seen": 69337088
},
{
"epoch": 0.02,
"learning_rate": 0.0004944332998996991,
"loss": 3.692,
"theoretical_loss": 5.026716438673677,
"tokens_seen": 69402624
},
{
"epoch": 0.02,
"learning_rate": 0.0004944232698094283,
"loss": 3.9909,
"theoretical_loss": 5.0260247045973045,
"tokens_seen": 69468160
},
{
"epoch": 0.02,
"learning_rate": 0.0004944132397191575,
"loss": 3.9516,
"theoretical_loss": 5.02533380531853,
"tokens_seen": 69533696
},
{
"epoch": 0.02,
"learning_rate": 0.0004944032096288867,
"loss": 4.0669,
"theoretical_loss": 5.024643739044526,
"tokens_seen": 69599232
},
{
"epoch": 0.02,
"learning_rate": 0.0004943931795386158,
"loss": 3.8696,
"theoretical_loss": 5.023954503987998,
"tokens_seen": 69664768
},
{
"epoch": 0.02,
"learning_rate": 0.000494383149448345,
"loss": 3.8331,
"theoretical_loss": 5.023266098367161,
"tokens_seen": 69730304
},
{
"epoch": 0.02,
"learning_rate": 0.0004943731193580742,
"loss": 3.9494,
"theoretical_loss": 5.022578520405721,
"tokens_seen": 69795840
},
{
"epoch": 0.02,
"learning_rate": 0.0004943630892678035,
"loss": 3.4961,
"theoretical_loss": 5.0218917683328534,
"tokens_seen": 69861376
},
{
"epoch": 0.02,
"learning_rate": 0.0004943530591775326,
"loss": 3.8677,
"theoretical_loss": 5.021205840383175,
"tokens_seen": 69926912
},
{
"epoch": 0.02,
"learning_rate": 0.0004943430290872618,
"loss": 3.6711,
"theoretical_loss": 5.020520734796728,
"tokens_seen": 69992448
},
{
"epoch": 0.02,
"learning_rate": 0.0004943329989969909,
"loss": 3.8103,
"theoretical_loss": 5.019836449818957,
"tokens_seen": 70057984
},
{
"epoch": 0.02,
"learning_rate": 0.0004943229689067202,
"loss": 3.9293,
"theoretical_loss": 5.019152983700687,
"tokens_seen": 70123520
},
{
"epoch": 0.02,
"learning_rate": 0.0004943129388164494,
"loss": 3.6354,
"theoretical_loss": 5.018470334698101,
"tokens_seen": 70189056
},
{
"epoch": 0.02,
"learning_rate": 0.0004943029087261786,
"loss": 3.5086,
"theoretical_loss": 5.01778850107272,
"tokens_seen": 70254592
},
{
"epoch": 0.02,
"learning_rate": 0.0004942928786359078,
"loss": 3.5231,
"theoretical_loss": 5.017107481091379,
"tokens_seen": 70320128
},
{
"epoch": 0.02,
"learning_rate": 0.0004942828485456369,
"loss": 3.7571,
"theoretical_loss": 5.016427273026212,
"tokens_seen": 70385664
},
{
"epoch": 0.02,
"objective/train/docs_used": 52769,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.7651798725128174,
"objective/train/theoretical_loss": 5.015747875154622,
"objective/train/tokens_used": 90911200,
"theoretical_loss": 5.015747875154622,
"tokens_seen": 70451200
},
{
"epoch": 0.02,
"learning_rate": 0.0004942728184553661,
"loss": 3.8123,
"theoretical_loss": 5.015747875154622,
"tokens_seen": 70451200
},
{
"epoch": 0.02,
"learning_rate": 0.0004942627883650953,
"loss": 3.593,
"theoretical_loss": 5.015069285759269,
"tokens_seen": 70516736
},
{
"epoch": 0.02,
"learning_rate": 0.0004942527582748245,
"loss": 3.6413,
"theoretical_loss": 5.01439150312804,
"tokens_seen": 70582272
},
{
"epoch": 0.02,
"learning_rate": 0.0004942427281845537,
"loss": 3.6802,
"theoretical_loss": 5.0137145255540405,
"tokens_seen": 70647808
},
{
"epoch": 0.02,
"learning_rate": 0.0004942326980942828,
"loss": 3.6497,
"theoretical_loss": 5.013038351335559,
"tokens_seen": 70713344
},
{
"epoch": 0.02,
"learning_rate": 0.000494222668004012,
"loss": 4.0304,
"theoretical_loss": 5.012362978776057,
"tokens_seen": 70778880
},
{
"epoch": 0.02,
"learning_rate": 0.0004942126379137412,
"loss": 3.6552,
"theoretical_loss": 5.011688406184147,
"tokens_seen": 70844416
},
{
"epoch": 0.02,
"learning_rate": 0.0004942026078234704,
"loss": 3.5983,
"theoretical_loss": 5.011014631873566,
"tokens_seen": 70909952
},
{
"epoch": 0.02,
"learning_rate": 0.0004941925777331996,
"loss": 3.9619,
"theoretical_loss": 5.010341654163167,
"tokens_seen": 70975488
},
{
"epoch": 0.02,
"learning_rate": 0.0004941825476429289,
"loss": 3.6699,
"theoretical_loss": 5.009669471376882,
"tokens_seen": 71041024
},
{
"epoch": 0.02,
"learning_rate": 0.0004941725175526579,
"loss": 3.8407,
"theoretical_loss": 5.008998081843721,
"tokens_seen": 71106560
},
{
"epoch": 0.02,
"learning_rate": 0.0004941624874623872,
"loss": 3.7259,
"theoretical_loss": 5.008327483897736,
"tokens_seen": 71172096
},
{
"epoch": 0.02,
"learning_rate": 0.0004941524573721163,
"loss": 3.5894,
"theoretical_loss": 5.00765767587801,
"tokens_seen": 71237632
},
{
"epoch": 0.02,
"learning_rate": 0.0004941424272818456,
"loss": 3.576,
"theoretical_loss": 5.006988656128635,
"tokens_seen": 71303168
},
{
"epoch": 0.02,
"learning_rate": 0.0004941323971915748,
"loss": 3.9922,
"theoretical_loss": 5.006320422998691,
"tokens_seen": 71368704
},
{
"epoch": 0.02,
"learning_rate": 0.0004941223671013039,
"loss": 3.8547,
"theoretical_loss": 5.00565297484223,
"tokens_seen": 71434240
},
{
"epoch": 0.02,
"learning_rate": 0.0004941123370110331,
"loss": 3.8379,
"theoretical_loss": 5.004986310018252,
"tokens_seen": 71499776
},
{
"epoch": 0.02,
"learning_rate": 0.0004941023069207623,
"loss": 3.7754,
"theoretical_loss": 5.004320426890686,
"tokens_seen": 71565312
},
{
"epoch": 0.02,
"learning_rate": 0.0004940922768304915,
"loss": 3.6593,
"theoretical_loss": 5.003655323828376,
"tokens_seen": 71630848
},
{
"epoch": 0.02,
"learning_rate": 0.0004940822467402207,
"loss": 3.6119,
"theoretical_loss": 5.002990999205057,
"tokens_seen": 71696384
},
{
"epoch": 0.02,
"learning_rate": 0.0004940722166499498,
"loss": 3.9359,
"theoretical_loss": 5.002327451399335,
"tokens_seen": 71761920
},
{
"epoch": 0.02,
"learning_rate": 0.000494062186559679,
"loss": 3.947,
"theoretical_loss": 5.001664678794671,
"tokens_seen": 71827456
},
{
"epoch": 0.02,
"learning_rate": 0.0004940521564694082,
"loss": 4.1257,
"theoretical_loss": 5.001002679779363,
"tokens_seen": 71892992
},
{
"epoch": 0.02,
"learning_rate": 0.0004940421263791374,
"loss": 3.5797,
"theoretical_loss": 5.0003414527465235,
"tokens_seen": 71958528
},
{
"epoch": 0.02,
"learning_rate": 0.0004940320962888666,
"loss": 3.4931,
"theoretical_loss": 4.99968099609406,
"tokens_seen": 72024064
},
{
"epoch": 0.02,
"objective/train/docs_used": 53229,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.443143129348755,
"objective/train/theoretical_loss": 4.999021308224664,
"objective/train/tokens_used": 92549600,
"theoretical_loss": 4.999021308224664,
"tokens_seen": 72089600
},
{
"epoch": 0.02,
"learning_rate": 0.0004940220661985958,
"loss": 3.8379,
"theoretical_loss": 4.999021308224664,
"tokens_seen": 72089600
},
{
"epoch": 0.02,
"learning_rate": 0.0004940120361083249,
"loss": 3.7169,
"theoretical_loss": 4.998362387545782,
"tokens_seen": 72155136
},
{
"epoch": 0.02,
"learning_rate": 0.0004940020060180542,
"loss": 3.7454,
"theoretical_loss": 4.997704232469606,
"tokens_seen": 72220672
},
{
"epoch": 0.02,
"learning_rate": 0.0004939919759277834,
"loss": 3.7546,
"theoretical_loss": 4.997046841413049,
"tokens_seen": 72286208
},
{
"epoch": 0.02,
"learning_rate": 0.0004939819458375126,
"loss": 3.7608,
"theoretical_loss": 4.996390212797728,
"tokens_seen": 72351744
},
{
"epoch": 0.02,
"learning_rate": 0.0004939719157472418,
"loss": 3.9281,
"theoretical_loss": 4.995734345049949,
"tokens_seen": 72417280
},
{
"epoch": 0.02,
"learning_rate": 0.000493961885656971,
"loss": 3.6721,
"theoretical_loss": 4.995079236600686,
"tokens_seen": 72482816
},
{
"epoch": 0.02,
"learning_rate": 0.0004939518555667001,
"loss": 3.5009,
"theoretical_loss": 4.994424885885564,
"tokens_seen": 72548352
},
{
"epoch": 0.02,
"learning_rate": 0.0004939418254764293,
"loss": 4.0443,
"theoretical_loss": 4.993771291344839,
"tokens_seen": 72613888
},
{
"epoch": 0.02,
"learning_rate": 0.0004939317953861585,
"loss": 3.851,
"theoretical_loss": 4.993118451423381,
"tokens_seen": 72679424
},
{
"epoch": 0.02,
"learning_rate": 0.0004939217652958877,
"loss": 3.8891,
"theoretical_loss": 4.992466364570659,
"tokens_seen": 72744960
},
{
"epoch": 0.02,
"learning_rate": 0.0004939117352056169,
"loss": 3.7893,
"theoretical_loss": 4.991815029240721,
"tokens_seen": 72810496
},
{
"epoch": 0.02,
"learning_rate": 0.000493901705115346,
"loss": 3.8364,
"theoretical_loss": 4.991164443892175,
"tokens_seen": 72876032
},
{
"epoch": 0.02,
"learning_rate": 0.0004938916750250752,
"loss": 3.4055,
"theoretical_loss": 4.990514606988173,
"tokens_seen": 72941568
},
{
"epoch": 0.02,
"learning_rate": 0.0004938816449348044,
"loss": 3.4979,
"theoretical_loss": 4.989865516996396,
"tokens_seen": 73007104
},
{
"epoch": 0.02,
"learning_rate": 0.0004938716148445337,
"loss": 3.8367,
"theoretical_loss": 4.98921717238903,
"tokens_seen": 73072640
},
{
"epoch": 0.02,
"learning_rate": 0.0004938615847542628,
"loss": 3.8357,
"theoretical_loss": 4.988569571642756,
"tokens_seen": 73138176
},
{
"epoch": 0.02,
"learning_rate": 0.000493851554663992,
"loss": 3.7075,
"theoretical_loss": 4.98792271323873,
"tokens_seen": 73203712
},
{
"epoch": 0.02,
"learning_rate": 0.0004938415245737211,
"loss": 3.939,
"theoretical_loss": 4.9872765956625615,
"tokens_seen": 73269248
},
{
"epoch": 0.02,
"learning_rate": 0.0004938314944834504,
"loss": 3.8696,
"theoretical_loss": 4.9866312174043035,
"tokens_seen": 73334784
},
{
"epoch": 0.02,
"learning_rate": 0.0004938214643931796,
"loss": 3.8818,
"theoretical_loss": 4.9859865769584335,
"tokens_seen": 73400320
},
{
"epoch": 0.02,
"learning_rate": 0.0004938114343029088,
"loss": 3.8674,
"theoretical_loss": 4.9853426728238315,
"tokens_seen": 73465856
},
{
"epoch": 0.02,
"learning_rate": 0.000493801404212638,
"loss": 3.7658,
"theoretical_loss": 4.984699503503771,
"tokens_seen": 73531392
},
{
"epoch": 0.02,
"learning_rate": 0.0004937913741223671,
"loss": 3.7394,
"theoretical_loss": 4.984057067505898,
"tokens_seen": 73596928
},
{
"epoch": 0.02,
"learning_rate": 0.0004937813440320963,
"loss": 3.8743,
"theoretical_loss": 4.9834153633422105,
"tokens_seen": 73662464
},
{
"epoch": 0.02,
"objective/train/docs_used": 53865,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.384488582611084,
"objective/train/theoretical_loss": 4.982774389529053,
"objective/train/tokens_used": 94188000,
"theoretical_loss": 4.982774389529053,
"tokens_seen": 73728000
},
{
"epoch": 0.02,
"learning_rate": 0.0004937713139418255,
"loss": 4.1487,
"theoretical_loss": 4.982774389529053,
"tokens_seen": 73728000
},
{
"epoch": 0.02,
"learning_rate": 0.0004937612838515547,
"loss": 3.535,
"theoretical_loss": 4.9821341445870875,
"tokens_seen": 73793536
},
{
"epoch": 0.02,
"learning_rate": 0.0004937512537612839,
"loss": 3.769,
"theoretical_loss": 4.981494627041286,
"tokens_seen": 73859072
},
{
"epoch": 0.02,
"learning_rate": 0.000493741223671013,
"loss": 3.6317,
"theoretical_loss": 4.98085583542091,
"tokens_seen": 73924608
},
{
"epoch": 0.02,
"learning_rate": 0.0004937311935807422,
"loss": 3.9226,
"theoretical_loss": 4.980217768259496,
"tokens_seen": 73990144
},
{
"epoch": 0.02,
"learning_rate": 0.0004937211634904714,
"loss": 3.7611,
"theoretical_loss": 4.979580424094836,
"tokens_seen": 74055680
},
{
"epoch": 0.02,
"learning_rate": 0.0004937111334002006,
"loss": 3.765,
"theoretical_loss": 4.978943801468967,
"tokens_seen": 74121216
},
{
"epoch": 0.02,
"learning_rate": 0.0004937011033099298,
"loss": 3.6153,
"theoretical_loss": 4.978307898928149,
"tokens_seen": 74186752
},
{
"epoch": 0.02,
"learning_rate": 0.0004936910732196591,
"loss": 3.7389,
"theoretical_loss": 4.977672715022855,
"tokens_seen": 74252288
},
{
"epoch": 0.02,
"learning_rate": 0.0004936810431293881,
"loss": 3.5885,
"theoretical_loss": 4.97703824830775,
"tokens_seen": 74317824
},
{
"epoch": 0.02,
"learning_rate": 0.0004936710130391174,
"loss": 3.7634,
"theoretical_loss": 4.976404497341676,
"tokens_seen": 74383360
},
{
"epoch": 0.02,
"learning_rate": 0.0004936609829488465,
"loss": 3.62,
"theoretical_loss": 4.975771460687641,
"tokens_seen": 74448896
},
{
"epoch": 0.02,
"learning_rate": 0.0004936509528585758,
"loss": 3.8546,
"theoretical_loss": 4.975139136912794,
"tokens_seen": 74514432
},
{
"epoch": 0.02,
"learning_rate": 0.000493640922768305,
"loss": 3.768,
"theoretical_loss": 4.974507524588424,
"tokens_seen": 74579968
},
{
"epoch": 0.02,
"learning_rate": 0.0004936308926780341,
"loss": 3.9096,
"theoretical_loss": 4.973876622289927,
"tokens_seen": 74645504
},
{
"epoch": 0.02,
"learning_rate": 0.0004936208625877633,
"loss": 3.6973,
"theoretical_loss": 4.973246428596802,
"tokens_seen": 74711040
},
{
"epoch": 0.02,
"learning_rate": 0.0004936108324974925,
"loss": 3.672,
"theoretical_loss": 4.972616942092634,
"tokens_seen": 74776576
},
{
"epoch": 0.02,
"learning_rate": 0.0004936008024072217,
"loss": 3.4177,
"theoretical_loss": 4.971988161365077,
"tokens_seen": 74842112
},
{
"epoch": 0.02,
"learning_rate": 0.0004935907723169509,
"loss": 3.9026,
"theoretical_loss": 4.9713600850058395,
"tokens_seen": 74907648
},
{
"epoch": 0.02,
"learning_rate": 0.00049358074222668,
"loss": 3.5683,
"theoretical_loss": 4.970732711610667,
"tokens_seen": 74973184
},
{
"epoch": 0.02,
"learning_rate": 0.0004935707121364092,
"loss": 3.6196,
"theoretical_loss": 4.97010603977933,
"tokens_seen": 75038720
},
{
"epoch": 0.02,
"learning_rate": 0.0004935606820461384,
"loss": 3.8518,
"theoretical_loss": 4.96948006811561,
"tokens_seen": 75104256
},
{
"epoch": 0.02,
"learning_rate": 0.0004935506519558676,
"loss": 3.8093,
"theoretical_loss": 4.968854795227281,
"tokens_seen": 75169792
},
{
"epoch": 0.02,
"learning_rate": 0.0004935406218655968,
"loss": 3.7438,
"theoretical_loss": 4.968230219726093,
"tokens_seen": 75235328
},
{
"epoch": 0.02,
"learning_rate": 0.000493530591775326,
"loss": 3.6534,
"theoretical_loss": 4.967606340227765,
"tokens_seen": 75300864
},
{
"epoch": 0.02,
"objective/train/docs_used": 54881,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.100285053253174,
"objective/train/theoretical_loss": 4.966983155351962,
"objective/train/tokens_used": 95826400,
"theoretical_loss": 4.966983155351962,
"tokens_seen": 75366400
},
{
"epoch": 0.02,
"learning_rate": 0.0004935205616850551,
"loss": 3.7242,
"theoretical_loss": 4.966983155351962,
"tokens_seen": 75366400
},
{
"epoch": 0.02,
"learning_rate": 0.0004935105315947844,
"loss": 3.7258,
"theoretical_loss": 4.966360663722287,
"tokens_seen": 75431936
},
{
"epoch": 0.02,
"learning_rate": 0.0004935005015045135,
"loss": 3.5934,
"theoretical_loss": 4.96573886396626,
"tokens_seen": 75497472
},
{
"epoch": 0.02,
"learning_rate": 0.0004934904714142428,
"loss": 3.4784,
"theoretical_loss": 4.965117754715307,
"tokens_seen": 75563008
},
{
"epoch": 0.02,
"learning_rate": 0.0004934804413239719,
"loss": 3.4207,
"theoretical_loss": 4.964497334604748,
"tokens_seen": 75628544
},
{
"epoch": 0.02,
"learning_rate": 0.0004934704112337011,
"loss": 3.8166,
"theoretical_loss": 4.963877602273776,
"tokens_seen": 75694080
},
{
"epoch": 0.02,
"learning_rate": 0.0004934603811434303,
"loss": 3.8047,
"theoretical_loss": 4.963258556365449,
"tokens_seen": 75759616
},
{
"epoch": 0.02,
"learning_rate": 0.0004934503510531595,
"loss": 3.7029,
"theoretical_loss": 4.962640195526673,
"tokens_seen": 75825152
},
{
"epoch": 0.02,
"learning_rate": 0.0004934403209628887,
"loss": 3.7634,
"theoretical_loss": 4.962022518408183,
"tokens_seen": 75890688
},
{
"epoch": 0.02,
"learning_rate": 0.0004934302908726179,
"loss": 3.5837,
"theoretical_loss": 4.96140552366454,
"tokens_seen": 75956224
},
{
"epoch": 0.02,
"learning_rate": 0.000493420260782347,
"loss": 3.6947,
"theoretical_loss": 4.9607892099541075,
"tokens_seen": 76021760
},
{
"epoch": 0.02,
"learning_rate": 0.0004934102306920762,
"loss": 3.6329,
"theoretical_loss": 4.9601735759390415,
"tokens_seen": 76087296
},
{
"epoch": 0.02,
"learning_rate": 0.0004934002006018054,
"loss": 3.615,
"theoretical_loss": 4.959558620285274,
"tokens_seen": 76152832
},
{
"epoch": 0.02,
"learning_rate": 0.0004933901705115346,
"loss": 3.7512,
"theoretical_loss": 4.958944341662502,
"tokens_seen": 76218368
},
{
"epoch": 0.02,
"learning_rate": 0.0004933801404212638,
"loss": 3.6382,
"theoretical_loss": 4.958330738744172,
"tokens_seen": 76283904
},
{
"epoch": 0.02,
"learning_rate": 0.000493370110330993,
"loss": 3.7997,
"theoretical_loss": 4.957717810207466,
"tokens_seen": 76349440
},
{
"epoch": 0.02,
"learning_rate": 0.0004933600802407221,
"loss": 3.6714,
"theoretical_loss": 4.957105554733289,
"tokens_seen": 76414976
},
{
"epoch": 0.02,
"learning_rate": 0.0004933500501504513,
"loss": 3.4102,
"theoretical_loss": 4.956493971006253,
"tokens_seen": 76480512
},
{
"epoch": 0.02,
"learning_rate": 0.0004933400200601805,
"loss": 3.6513,
"theoretical_loss": 4.955883057714669,
"tokens_seen": 76546048
},
{
"epoch": 0.02,
"learning_rate": 0.0004933299899699098,
"loss": 3.7063,
"theoretical_loss": 4.955272813550524,
"tokens_seen": 76611584
},
{
"epoch": 0.02,
"learning_rate": 0.0004933199598796389,
"loss": 3.5682,
"theoretical_loss": 4.954663237209477,
"tokens_seen": 76677120
},
{
"epoch": 0.02,
"learning_rate": 0.0004933099297893682,
"loss": 3.859,
"theoretical_loss": 4.954054327390841,
"tokens_seen": 76742656
},
{
"epoch": 0.02,
"learning_rate": 0.0004932998996990972,
"loss": 3.6904,
"theoretical_loss": 4.9534460827975675,
"tokens_seen": 76808192
},
{
"epoch": 0.02,
"learning_rate": 0.0004932898696088265,
"loss": 3.5948,
"theoretical_loss": 4.952838502136241,
"tokens_seen": 76873728
},
{
"epoch": 0.02,
"learning_rate": 0.0004932798395185557,
"loss": 3.6899,
"theoretical_loss": 4.952231584117056,
"tokens_seen": 76939264
},
{
"epoch": 0.02,
"objective/train/docs_used": 55424,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.752882957458496,
"objective/train/theoretical_loss": 4.951625327453812,
"objective/train/tokens_used": 97464800,
"theoretical_loss": 4.951625327453812,
"tokens_seen": 77004800
},
{
"epoch": 0.02,
"learning_rate": 0.0004932698094282849,
"loss": 3.6065,
"theoretical_loss": 4.951625327453812,
"tokens_seen": 77004800
},
{
"epoch": 0.02,
"learning_rate": 0.0004932597793380141,
"loss": 3.69,
"theoretical_loss": 4.951019730863894,
"tokens_seen": 77070336
},
{
"epoch": 0.02,
"learning_rate": 0.0004932497492477432,
"loss": 3.5959,
"theoretical_loss": 4.950414793068266,
"tokens_seen": 77135872
},
{
"epoch": 0.02,
"learning_rate": 0.0004932397191574724,
"loss": 3.7588,
"theoretical_loss": 4.94981051279145,
"tokens_seen": 77201408
},
{
"epoch": 0.02,
"learning_rate": 0.0004932296890672016,
"loss": 3.8397,
"theoretical_loss": 4.94920688876152,
"tokens_seen": 77266944
},
{
"epoch": 0.02,
"learning_rate": 0.0004932196589769308,
"loss": 3.4345,
"theoretical_loss": 4.948603919710088,
"tokens_seen": 77332480
},
{
"epoch": 0.02,
"learning_rate": 0.00049320962888666,
"loss": 3.6935,
"theoretical_loss": 4.948001604372287,
"tokens_seen": 77398016
},
{
"epoch": 0.02,
"learning_rate": 0.0004931995987963893,
"loss": 3.6558,
"theoretical_loss": 4.947399941486762,
"tokens_seen": 77463552
},
{
"epoch": 0.02,
"learning_rate": 0.0004931895687061183,
"loss": 4.0196,
"theoretical_loss": 4.946798929795658,
"tokens_seen": 77529088
},
{
"epoch": 0.02,
"learning_rate": 0.0004931795386158476,
"loss": 3.5714,
"theoretical_loss": 4.946198568044602,
"tokens_seen": 77594624
},
{
"epoch": 0.02,
"learning_rate": 0.0004931695085255767,
"loss": 3.4172,
"theoretical_loss": 4.945598854982698,
"tokens_seen": 77660160
},
{
"epoch": 0.02,
"learning_rate": 0.000493159478435306,
"loss": 3.8399,
"theoretical_loss": 4.944999789362508,
"tokens_seen": 77725696
},
{
"epoch": 0.02,
"learning_rate": 0.0004931494483450352,
"loss": 3.7901,
"theoretical_loss": 4.944401369940043,
"tokens_seen": 77791232
},
{
"epoch": 0.02,
"learning_rate": 0.0004931394182547643,
"loss": 3.5237,
"theoretical_loss": 4.9438035954747495,
"tokens_seen": 77856768
},
{
"epoch": 0.02,
"learning_rate": 0.0004931293881644935,
"loss": 3.4995,
"theoretical_loss": 4.9432064647294975,
"tokens_seen": 77922304
},
{
"epoch": 0.02,
"learning_rate": 0.0004931193580742227,
"loss": 3.8164,
"theoretical_loss": 4.942609976470566,
"tokens_seen": 77987840
},
{
"epoch": 0.02,
"learning_rate": 0.0004931093279839519,
"loss": 3.5615,
"theoretical_loss": 4.942014129467637,
"tokens_seen": 78053376
},
{
"epoch": 0.02,
"learning_rate": 0.0004930992978936811,
"loss": 3.3958,
"theoretical_loss": 4.941418922493774,
"tokens_seen": 78118912
},
{
"epoch": 0.02,
"learning_rate": 0.0004930892678034102,
"loss": 3.5015,
"theoretical_loss": 4.940824354325419,
"tokens_seen": 78184448
},
{
"epoch": 0.02,
"learning_rate": 0.0004930792377131394,
"loss": 3.5466,
"theoretical_loss": 4.940230423742372,
"tokens_seen": 78249984
},
{
"epoch": 0.02,
"learning_rate": 0.0004930692076228686,
"loss": 3.503,
"theoretical_loss": 4.939637129527789,
"tokens_seen": 78315520
},
{
"epoch": 0.02,
"learning_rate": 0.0004930591775325978,
"loss": 3.6357,
"theoretical_loss": 4.939044470468156,
"tokens_seen": 78381056
},
{
"epoch": 0.02,
"learning_rate": 0.000493049147442327,
"loss": 3.7812,
"theoretical_loss": 4.938452445353294,
"tokens_seen": 78446592
},
{
"epoch": 0.02,
"learning_rate": 0.0004930391173520562,
"loss": 3.4835,
"theoretical_loss": 4.937861052976332,
"tokens_seen": 78512128
},
{
"epoch": 0.02,
"learning_rate": 0.0004930290872617853,
"loss": 3.1521,
"theoretical_loss": 4.937270292133704,
"tokens_seen": 78577664
},
{
"epoch": 0.02,
"objective/train/docs_used": 56825,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.739461898803711,
"objective/train/theoretical_loss": 4.9366801616251355,
"objective/train/tokens_used": 99103200,
"theoretical_loss": 4.9366801616251355,
"tokens_seen": 78643200
},
{
"epoch": 0.02,
"learning_rate": 0.0004930190571715146,
"loss": 3.8046,
"theoretical_loss": 4.9366801616251355,
"tokens_seen": 78643200
},
{
"epoch": 0.02,
"learning_rate": 0.0004930090270812437,
"loss": 3.5368,
"theoretical_loss": 4.93609066025363,
"tokens_seen": 78708736
},
{
"epoch": 0.02,
"learning_rate": 0.000492998996990973,
"loss": 3.878,
"theoretical_loss": 4.935501786825457,
"tokens_seen": 78774272
},
{
"epoch": 0.02,
"learning_rate": 0.0004929889669007021,
"loss": 3.6171,
"theoretical_loss": 4.934913540150143,
"tokens_seen": 78839808
},
{
"epoch": 0.02,
"learning_rate": 0.0004929789368104313,
"loss": 3.5845,
"theoretical_loss": 4.934325919040461,
"tokens_seen": 78905344
},
{
"epoch": 0.02,
"learning_rate": 0.0004929689067201605,
"loss": 3.5543,
"theoretical_loss": 4.933738922312413,
"tokens_seen": 78970880
},
{
"epoch": 0.02,
"learning_rate": 0.0004929588766298897,
"loss": 3.3024,
"theoretical_loss": 4.933152548785222,
"tokens_seen": 79036416
},
{
"epoch": 0.02,
"learning_rate": 0.0004929488465396189,
"loss": 3.6027,
"theoretical_loss": 4.932566797281324,
"tokens_seen": 79101952
},
{
"epoch": 0.02,
"learning_rate": 0.0004929388164493481,
"loss": 3.9121,
"theoretical_loss": 4.931981666626351,
"tokens_seen": 79167488
},
{
"epoch": 0.02,
"learning_rate": 0.0004929287863590773,
"loss": 3.5938,
"theoretical_loss": 4.931397155649121,
"tokens_seen": 79233024
},
{
"epoch": 0.02,
"learning_rate": 0.0004929187562688064,
"loss": 3.7354,
"theoretical_loss": 4.930813263181631,
"tokens_seen": 79298560
},
{
"epoch": 0.02,
"learning_rate": 0.0004929087261785356,
"loss": 3.433,
"theoretical_loss": 4.93022998805904,
"tokens_seen": 79364096
},
{
"epoch": 0.02,
"learning_rate": 0.0004928986960882648,
"loss": 3.6705,
"theoretical_loss": 4.929647329119659,
"tokens_seen": 79429632
},
{
"epoch": 0.02,
"learning_rate": 0.000492888665997994,
"loss": 3.3607,
"theoretical_loss": 4.9290652852049455,
"tokens_seen": 79495168
},
{
"epoch": 0.02,
"learning_rate": 0.0004928786359077232,
"loss": 3.5638,
"theoretical_loss": 4.928483855159485,
"tokens_seen": 79560704
},
{
"epoch": 0.02,
"learning_rate": 0.0004928686058174523,
"loss": 3.8926,
"theoretical_loss": 4.927903037830983,
"tokens_seen": 79626240
},
{
"epoch": 0.02,
"learning_rate": 0.0004928585757271815,
"loss": 3.8163,
"theoretical_loss": 4.9273228320702565,
"tokens_seen": 79691776
},
{
"epoch": 0.02,
"learning_rate": 0.0004928485456369107,
"loss": 3.6785,
"theoretical_loss": 4.926743236731218,
"tokens_seen": 79757312
},
{
"epoch": 0.02,
"learning_rate": 0.00049283851554664,
"loss": 3.5302,
"theoretical_loss": 4.926164250670868,
"tokens_seen": 79822848
},
{
"epoch": 0.02,
"learning_rate": 0.0004928284854563691,
"loss": 3.4988,
"theoretical_loss": 4.925585872749284,
"tokens_seen": 79888384
},
{
"epoch": 0.02,
"learning_rate": 0.0004928184553660984,
"loss": 3.6865,
"theoretical_loss": 4.925008101829608,
"tokens_seen": 79953920
},
{
"epoch": 0.02,
"learning_rate": 0.0004928084252758274,
"loss": 3.5066,
"theoretical_loss": 4.9244309367780374,
"tokens_seen": 80019456
},
{
"epoch": 0.02,
"learning_rate": 0.0004927983951855567,
"loss": 3.8955,
"theoretical_loss": 4.923854376463816,
"tokens_seen": 80084992
},
{
"epoch": 0.02,
"learning_rate": 0.0004927883650952859,
"loss": 3.7358,
"theoretical_loss": 4.923278419759217,
"tokens_seen": 80150528
},
{
"epoch": 0.02,
"learning_rate": 0.0004927783350050151,
"loss": 3.7513,
"theoretical_loss": 4.92270306553954,
"tokens_seen": 80216064
},
{
"epoch": 0.02,
"objective/train/docs_used": 57296,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.519014596939087,
"objective/train/theoretical_loss": 4.922128312683096,
"objective/train/tokens_used": 100741600,
"theoretical_loss": 4.922128312683096,
"tokens_seen": 80281600
},
{
"epoch": 0.02,
"learning_rate": 0.0004927683049147443,
"loss": 3.5332,
"theoretical_loss": 4.922128312683096,
"tokens_seen": 80281600
},
{
"epoch": 0.02,
"learning_rate": 0.0004927582748244734,
"loss": 3.5115,
"theoretical_loss": 4.921554160071194,
"tokens_seen": 80347136
},
{
"epoch": 0.02,
"learning_rate": 0.0004927482447342026,
"loss": 3.4584,
"theoretical_loss": 4.920980606588142,
"tokens_seen": 80412672
},
{
"epoch": 0.02,
"learning_rate": 0.0004927382146439318,
"loss": 3.6266,
"theoretical_loss": 4.920407651121222,
"tokens_seen": 80478208
},
{
"epoch": 0.02,
"learning_rate": 0.000492728184553661,
"loss": 3.8078,
"theoretical_loss": 4.919835292560689,
"tokens_seen": 80543744
},
{
"epoch": 0.02,
"learning_rate": 0.0004927181544633902,
"loss": 3.8442,
"theoretical_loss": 4.919263529799759,
"tokens_seen": 80609280
},
{
"epoch": 0.02,
"learning_rate": 0.0004927081243731193,
"loss": 3.8532,
"theoretical_loss": 4.918692361734598,
"tokens_seen": 80674816
},
{
"epoch": 0.02,
"learning_rate": 0.0004926980942828485,
"loss": 3.4396,
"theoretical_loss": 4.91812178726431,
"tokens_seen": 80740352
},
{
"epoch": 0.02,
"learning_rate": 0.0004926880641925777,
"loss": 3.6955,
"theoretical_loss": 4.917551805290929,
"tokens_seen": 80805888
},
{
"epoch": 0.02,
"learning_rate": 0.0004926780341023069,
"loss": 3.5437,
"theoretical_loss": 4.916982414719408,
"tokens_seen": 80871424
},
{
"epoch": 0.02,
"learning_rate": 0.0004926680040120361,
"loss": 3.8737,
"theoretical_loss": 4.9164136144576105,
"tokens_seen": 80936960
},
{
"epoch": 0.02,
"learning_rate": 0.0004926579739217654,
"loss": 3.464,
"theoretical_loss": 4.915845403416299,
"tokens_seen": 81002496
},
{
"epoch": 0.02,
"learning_rate": 0.0004926479438314944,
"loss": 3.6247,
"theoretical_loss": 4.915277780509124,
"tokens_seen": 81068032
},
{
"epoch": 0.02,
"learning_rate": 0.0004926379137412237,
"loss": 3.4444,
"theoretical_loss": 4.914710744652614,
"tokens_seen": 81133568
},
{
"epoch": 0.02,
"learning_rate": 0.0004926278836509528,
"loss": 3.5921,
"theoretical_loss": 4.914144294766169,
"tokens_seen": 81199104
},
{
"epoch": 0.02,
"learning_rate": 0.0004926178535606821,
"loss": 3.5012,
"theoretical_loss": 4.913578429772047,
"tokens_seen": 81264640
},
{
"epoch": 0.02,
"learning_rate": 0.0004926078234704113,
"loss": 3.3114,
"theoretical_loss": 4.913013148595355,
"tokens_seen": 81330176
},
{
"epoch": 0.02,
"learning_rate": 0.0004925977933801404,
"loss": 3.7158,
"theoretical_loss": 4.912448450164041,
"tokens_seen": 81395712
},
{
"epoch": 0.02,
"learning_rate": 0.0004925877632898696,
"loss": 3.7015,
"theoretical_loss": 4.91188433340888,
"tokens_seen": 81461248
},
{
"epoch": 0.02,
"learning_rate": 0.0004925777331995988,
"loss": 3.9955,
"theoretical_loss": 4.911320797263471,
"tokens_seen": 81526784
},
{
"epoch": 0.02,
"learning_rate": 0.000492567703109328,
"loss": 3.8269,
"theoretical_loss": 4.910757840664219,
"tokens_seen": 81592320
},
{
"epoch": 0.02,
"learning_rate": 0.0004925576730190572,
"loss": 3.5523,
"theoretical_loss": 4.910195462550334,
"tokens_seen": 81657856
},
{
"epoch": 0.02,
"learning_rate": 0.0004925476429287864,
"loss": 3.552,
"theoretical_loss": 4.909633661863811,
"tokens_seen": 81723392
},
{
"epoch": 0.02,
"learning_rate": 0.0004925376128385155,
"loss": 3.8094,
"theoretical_loss": 4.909072437549434,
"tokens_seen": 81788928
},
{
"epoch": 0.02,
"learning_rate": 0.0004925275827482447,
"loss": 3.7451,
"theoretical_loss": 4.908511788554753,
"tokens_seen": 81854464
},
{
"epoch": 0.02,
"objective/train/docs_used": 58605,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.980154275894165,
"objective/train/theoretical_loss": 4.907951713830082,
"objective/train/tokens_used": 102380000,
"theoretical_loss": 4.907951713830082,
"tokens_seen": 81920000
},
{
"epoch": 0.02,
"learning_rate": 0.0004925175526579739,
"loss": 3.833,
"theoretical_loss": 4.907951713830082,
"tokens_seen": 81920000
},
{
"epoch": 0.02,
"learning_rate": 0.0004925075225677031,
"loss": 3.5955,
"theoretical_loss": 4.907392212328489,
"tokens_seen": 81985536
},
{
"epoch": 0.02,
"learning_rate": 0.0004924974924774323,
"loss": 3.3892,
"theoretical_loss": 4.906833283005785,
"tokens_seen": 82051072
},
{
"epoch": 0.02,
"learning_rate": 0.0004924874623871615,
"loss": 3.7307,
"theoretical_loss": 4.906274924820515,
"tokens_seen": 82116608
},
{
"epoch": 0.02,
"learning_rate": 0.0004924774322968907,
"loss": 3.7447,
"theoretical_loss": 4.90571713673395,
"tokens_seen": 82182144
},
{
"epoch": 0.02,
"learning_rate": 0.0004924674022066199,
"loss": 3.5136,
"theoretical_loss": 4.905159917710073,
"tokens_seen": 82247680
},
{
"epoch": 0.02,
"learning_rate": 0.0004924573721163491,
"loss": 3.6628,
"theoretical_loss": 4.904603266715578,
"tokens_seen": 82313216
},
{
"epoch": 0.02,
"learning_rate": 0.0004924473420260783,
"loss": 3.7499,
"theoretical_loss": 4.904047182719854,
"tokens_seen": 82378752
},
{
"epoch": 0.02,
"learning_rate": 0.0004924373119358075,
"loss": 3.4528,
"theoretical_loss": 4.903491664694977,
"tokens_seen": 82444288
},
{
"epoch": 0.03,
"learning_rate": 0.0004924272818455366,
"loss": 3.614,
"theoretical_loss": 4.902936711615702,
"tokens_seen": 82509824
},
{
"epoch": 0.03,
"learning_rate": 0.0004924172517552658,
"loss": 3.7331,
"theoretical_loss": 4.902382322459456,
"tokens_seen": 82575360
},
{
"epoch": 0.03,
"learning_rate": 0.000492407221664995,
"loss": 3.5698,
"theoretical_loss": 4.901828496206322,
"tokens_seen": 82640896
},
{
"epoch": 0.03,
"learning_rate": 0.0004923971915747242,
"loss": 3.6838,
"theoretical_loss": 4.90127523183904,
"tokens_seen": 82706432
},
{
"epoch": 0.03,
"learning_rate": 0.0004923871614844534,
"loss": 3.7826,
"theoretical_loss": 4.900722528342988,
"tokens_seen": 82771968
},
{
"epoch": 0.03,
"learning_rate": 0.0004923771313941825,
"loss": 3.6392,
"theoretical_loss": 4.900170384706181,
"tokens_seen": 82837504
},
{
"epoch": 0.03,
"learning_rate": 0.0004923671013039117,
"loss": 3.491,
"theoretical_loss": 4.899618799919256,
"tokens_seen": 82903040
},
{
"epoch": 0.03,
"learning_rate": 0.0004923570712136409,
"loss": 3.4612,
"theoretical_loss": 4.899067772975469,
"tokens_seen": 82968576
},
{
"epoch": 0.03,
"learning_rate": 0.0004923470411233702,
"loss": 3.687,
"theoretical_loss": 4.898517302870679,
"tokens_seen": 83034112
},
{
"epoch": 0.03,
"learning_rate": 0.0004923370110330993,
"loss": 3.7585,
"theoretical_loss": 4.897967388603346,
"tokens_seen": 83099648
},
{
"epoch": 0.03,
"learning_rate": 0.0004923269809428286,
"loss": 3.5586,
"theoretical_loss": 4.897418029174519,
"tokens_seen": 83165184
},
{
"epoch": 0.03,
"learning_rate": 0.0004923169508525576,
"loss": 3.9548,
"theoretical_loss": 4.896869223587828,
"tokens_seen": 83230720
},
{
"epoch": 0.03,
"learning_rate": 0.0004923069207622869,
"loss": 3.8464,
"theoretical_loss": 4.896320970849472,
"tokens_seen": 83296256
},
{
"epoch": 0.03,
"learning_rate": 0.0004922968906720161,
"loss": 3.5297,
"theoretical_loss": 4.895773269968219,
"tokens_seen": 83361792
},
{
"epoch": 0.03,
"learning_rate": 0.0004922868605817453,
"loss": 3.7033,
"theoretical_loss": 4.895226119955386,
"tokens_seen": 83427328
},
{
"epoch": 0.03,
"learning_rate": 0.0004922768304914745,
"loss": 3.9995,
"theoretical_loss": 4.894679519824841,
"tokens_seen": 83492864
},
{
"epoch": 0.03,
"objective/train/docs_used": 59310,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.926790952682495,
"objective/train/theoretical_loss": 4.894133468592984,
"objective/train/tokens_used": 104018400,
"theoretical_loss": 4.894133468592984,
"tokens_seen": 83558400
},
{
"epoch": 0.03,
"learning_rate": 0.0004922668004012036,
"loss": 3.6992,
"theoretical_loss": 4.894133468592984,
"tokens_seen": 83558400
},
{
"epoch": 0.03,
"learning_rate": 0.0004922567703109328,
"loss": 3.6724,
"theoretical_loss": 4.8935879652787495,
"tokens_seen": 83623936
},
{
"epoch": 0.03,
"learning_rate": 0.000492246740220662,
"loss": 3.7634,
"theoretical_loss": 4.893043008903591,
"tokens_seen": 83689472
},
{
"epoch": 0.03,
"learning_rate": 0.0004922367101303912,
"loss": 3.701,
"theoretical_loss": 4.892498598491473,
"tokens_seen": 83755008
},
{
"epoch": 0.03,
"learning_rate": 0.0004922266800401204,
"loss": 3.6549,
"theoretical_loss": 4.891954733068863,
"tokens_seen": 83820544
},
{
"epoch": 0.03,
"learning_rate": 0.0004922166499498495,
"loss": 3.6165,
"theoretical_loss": 4.891411411664727,
"tokens_seen": 83886080
},
{
"epoch": 0.03,
"learning_rate": 0.0004922066198595787,
"loss": 3.682,
"theoretical_loss": 4.890868633310515,
"tokens_seen": 83951616
},
{
"epoch": 0.03,
"learning_rate": 0.0004921965897693079,
"loss": 3.4424,
"theoretical_loss": 4.890326397040158,
"tokens_seen": 84017152
},
{
"epoch": 0.03,
"learning_rate": 0.0004921865596790371,
"loss": 3.6114,
"theoretical_loss": 4.889784701890056,
"tokens_seen": 84082688
},
{
"epoch": 0.03,
"learning_rate": 0.0004921765295887663,
"loss": 3.5193,
"theoretical_loss": 4.8892435468990705,
"tokens_seen": 84148224
},
{
"epoch": 0.03,
"learning_rate": 0.0004921664994984956,
"loss": 3.2842,
"theoretical_loss": 4.88870293110852,
"tokens_seen": 84213760
},
{
"epoch": 0.03,
"learning_rate": 0.0004921564694082246,
"loss": 3.7096,
"theoretical_loss": 4.888162853562166,
"tokens_seen": 84279296
},
{
"epoch": 0.03,
"learning_rate": 0.0004921464393179539,
"loss": 3.5071,
"theoretical_loss": 4.88762331330621,
"tokens_seen": 84344832
},
{
"epoch": 0.03,
"learning_rate": 0.000492136409227683,
"loss": 3.4462,
"theoretical_loss": 4.88708430938928,
"tokens_seen": 84410368
},
{
"epoch": 0.03,
"learning_rate": 0.0004921263791374123,
"loss": 3.7536,
"theoretical_loss": 4.8865458408624285,
"tokens_seen": 84475904
},
{
"epoch": 0.03,
"learning_rate": 0.0004921163490471415,
"loss": 3.493,
"theoretical_loss": 4.8860079067791204,
"tokens_seen": 84541440
},
{
"epoch": 0.03,
"learning_rate": 0.0004921063189568706,
"loss": 3.7587,
"theoretical_loss": 4.885470506195227,
"tokens_seen": 84606976
},
{
"epoch": 0.03,
"learning_rate": 0.0004920962888665998,
"loss": 3.6445,
"theoretical_loss": 4.884933638169014,
"tokens_seen": 84672512
},
{
"epoch": 0.03,
"learning_rate": 0.000492086258776329,
"loss": 3.5871,
"theoretical_loss": 4.88439730176114,
"tokens_seen": 84738048
},
{
"epoch": 0.03,
"learning_rate": 0.0004920762286860582,
"loss": 3.772,
"theoretical_loss": 4.883861496034644,
"tokens_seen": 84803584
},
{
"epoch": 0.03,
"learning_rate": 0.0004920661985957874,
"loss": 3.4383,
"theoretical_loss": 4.88332622005494,
"tokens_seen": 84869120
},
{
"epoch": 0.03,
"learning_rate": 0.0004920561685055166,
"loss": 3.4985,
"theoretical_loss": 4.8827914728898065,
"tokens_seen": 84934656
},
{
"epoch": 0.03,
"learning_rate": 0.0004920461384152457,
"loss": 3.5046,
"theoretical_loss": 4.88225725360938,
"tokens_seen": 85000192
},
{
"epoch": 0.03,
"learning_rate": 0.0004920361083249749,
"loss": 3.682,
"theoretical_loss": 4.881723561286149,
"tokens_seen": 85065728
},
{
"epoch": 0.03,
"learning_rate": 0.0004920260782347041,
"loss": 3.422,
"theoretical_loss": 4.881190394994943,
"tokens_seen": 85131264
},
{
"epoch": 0.03,
"objective/train/docs_used": 60492,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.639910936355591,
"objective/train/theoretical_loss": 4.880657753812926,
"objective/train/tokens_used": 105656800,
"theoretical_loss": 4.880657753812926,
"tokens_seen": 85196800
},
{
"epoch": 0.03,
"learning_rate": 0.0004920160481444333,
"loss": 3.7149,
"theoretical_loss": 4.880657753812926,
"tokens_seen": 85196800
},
{
"epoch": 0.03,
"learning_rate": 0.0004920060180541625,
"loss": 3.5313,
"theoretical_loss": 4.880125636819594,
"tokens_seen": 85262336
},
{
"epoch": 0.03,
"learning_rate": 0.0004919959879638916,
"loss": 3.6137,
"theoretical_loss": 4.879594043096755,
"tokens_seen": 85327872
},
{
"epoch": 0.03,
"learning_rate": 0.0004919859578736209,
"loss": 3.4991,
"theoretical_loss": 4.879062971728534,
"tokens_seen": 85393408
},
{
"epoch": 0.03,
"learning_rate": 0.00049197592778335,
"loss": 3.6767,
"theoretical_loss": 4.87853242180136,
"tokens_seen": 85458944
},
{
"epoch": 0.03,
"learning_rate": 0.0004919658976930793,
"loss": 3.7686,
"theoretical_loss": 4.878002392403959,
"tokens_seen": 85524480
},
{
"epoch": 0.03,
"learning_rate": 0.0004919558676028084,
"loss": 3.5113,
"theoretical_loss": 4.877472882627343,
"tokens_seen": 85590016
},
{
"epoch": 0.03,
"learning_rate": 0.0004919458375125377,
"loss": 3.7634,
"theoretical_loss": 4.8769438915648085,
"tokens_seen": 85655552
},
{
"epoch": 0.03,
"learning_rate": 0.0004919358074222668,
"loss": 3.4673,
"theoretical_loss": 4.876415418311928,
"tokens_seen": 85721088
},
{
"epoch": 0.03,
"learning_rate": 0.000491925777331996,
"loss": 3.562,
"theoretical_loss": 4.875887461966537,
"tokens_seen": 85786624
},
{
"epoch": 0.03,
"learning_rate": 0.0004919157472417252,
"loss": 3.7082,
"theoretical_loss": 4.875360021628733,
"tokens_seen": 85852160
},
{
"epoch": 0.03,
"learning_rate": 0.0004919057171514544,
"loss": 3.558,
"theoretical_loss": 4.874833096400865,
"tokens_seen": 85917696
},
{
"epoch": 0.03,
"learning_rate": 0.0004918956870611836,
"loss": 3.259,
"theoretical_loss": 4.874306685387525,
"tokens_seen": 85983232
},
{
"epoch": 0.03,
"learning_rate": 0.0004918856569709127,
"loss": 3.6862,
"theoretical_loss": 4.873780787695547,
"tokens_seen": 86048768
},
{
"epoch": 0.03,
"learning_rate": 0.0004918756268806419,
"loss": 3.546,
"theoretical_loss": 4.87325540243399,
"tokens_seen": 86114304
},
{
"epoch": 0.03,
"learning_rate": 0.0004918655967903711,
"loss": 3.5487,
"theoretical_loss": 4.872730528714139,
"tokens_seen": 86179840
},
{
"epoch": 0.03,
"learning_rate": 0.0004918555667001003,
"loss": 3.57,
"theoretical_loss": 4.872206165649493,
"tokens_seen": 86245376
},
{
"epoch": 0.03,
"learning_rate": 0.0004918455366098295,
"loss": 3.6859,
"theoretical_loss": 4.871682312355761,
"tokens_seen": 86310912
},
{
"epoch": 0.03,
"learning_rate": 0.0004918355065195586,
"loss": 3.588,
"theoretical_loss": 4.871158967950852,
"tokens_seen": 86376448
},
{
"epoch": 0.03,
"learning_rate": 0.0004918254764292878,
"loss": 3.6822,
"theoretical_loss": 4.870636131554869,
"tokens_seen": 86441984
},
{
"epoch": 0.03,
"learning_rate": 0.000491815446339017,
"loss": 3.4897,
"theoretical_loss": 4.8701138022901045,
"tokens_seen": 86507520
},
{
"epoch": 0.03,
"learning_rate": 0.0004918054162487463,
"loss": 3.5301,
"theoretical_loss": 4.869591979281028,
"tokens_seen": 86573056
},
{
"epoch": 0.03,
"learning_rate": 0.0004917953861584754,
"loss": 3.3777,
"theoretical_loss": 4.8690706616542805,
"tokens_seen": 86638592
},
{
"epoch": 0.03,
"learning_rate": 0.0004917853560682047,
"loss": 3.3358,
"theoretical_loss": 4.868549848538675,
"tokens_seen": 86704128
},
{
"epoch": 0.03,
"learning_rate": 0.0004917753259779337,
"loss": 3.3347,
"theoretical_loss": 4.868029539065176,
"tokens_seen": 86769664
},
{
"epoch": 0.03,
"objective/train/docs_used": 61005,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4489526748657227,
"objective/train/theoretical_loss": 4.867509732366907,
"objective/train/tokens_used": 107295200,
"theoretical_loss": 4.867509732366907,
"tokens_seen": 86835200
},
{
"epoch": 0.03,
"learning_rate": 0.000491765295887663,
"loss": 3.6654,
"theoretical_loss": 4.867509732366907,
"tokens_seen": 86835200
},
{
"epoch": 0.03,
"learning_rate": 0.0004917552657973922,
"loss": 3.8192,
"theoretical_loss": 4.866990427579129,
"tokens_seen": 86900736
},
{
"epoch": 0.03,
"learning_rate": 0.0004917452357071214,
"loss": 3.5657,
"theoretical_loss": 4.866471623839248,
"tokens_seen": 86966272
},
{
"epoch": 0.03,
"learning_rate": 0.0004917352056168506,
"loss": 3.348,
"theoretical_loss": 4.8659533202867955,
"tokens_seen": 87031808
},
{
"epoch": 0.03,
"learning_rate": 0.0004917251755265797,
"loss": 3.4467,
"theoretical_loss": 4.86543551606343,
"tokens_seen": 87097344
},
{
"epoch": 0.03,
"learning_rate": 0.0004917151454363089,
"loss": 3.5397,
"theoretical_loss": 4.864918210312927,
"tokens_seen": 87162880
},
{
"epoch": 0.03,
"learning_rate": 0.0004917051153460381,
"loss": 3.5919,
"theoretical_loss": 4.864401402181173,
"tokens_seen": 87228416
},
{
"epoch": 0.03,
"learning_rate": 0.0004916950852557673,
"loss": 3.5553,
"theoretical_loss": 4.863885090816158,
"tokens_seen": 87293952
},
{
"epoch": 0.03,
"learning_rate": 0.0004916850551654965,
"loss": 3.6125,
"theoretical_loss": 4.863369275367968,
"tokens_seen": 87359488
},
{
"epoch": 0.03,
"learning_rate": 0.0004916750250752258,
"loss": 3.4756,
"theoretical_loss": 4.862853954988781,
"tokens_seen": 87425024
},
{
"epoch": 0.03,
"learning_rate": 0.0004916649949849548,
"loss": 3.5864,
"theoretical_loss": 4.862339128832857,
"tokens_seen": 87490560
},
{
"epoch": 0.03,
"learning_rate": 0.0004916549648946841,
"loss": 3.6655,
"theoretical_loss": 4.861824796056533,
"tokens_seen": 87556096
},
{
"epoch": 0.03,
"learning_rate": 0.0004916449348044132,
"loss": 3.5427,
"theoretical_loss": 4.861310955818219,
"tokens_seen": 87621632
},
{
"epoch": 0.03,
"learning_rate": 0.0004916349047141425,
"loss": 3.5534,
"theoretical_loss": 4.860797607278385,
"tokens_seen": 87687168
},
{
"epoch": 0.03,
"learning_rate": 0.0004916248746238717,
"loss": 3.7963,
"theoretical_loss": 4.86028474959956,
"tokens_seen": 87752704
},
{
"epoch": 0.03,
"learning_rate": 0.0004916148445336008,
"loss": 3.6729,
"theoretical_loss": 4.859772381946323,
"tokens_seen": 87818240
},
{
"epoch": 0.03,
"learning_rate": 0.00049160481444333,
"loss": 3.6349,
"theoretical_loss": 4.859260503485298,
"tokens_seen": 87883776
},
{
"epoch": 0.03,
"learning_rate": 0.0004915947843530592,
"loss": 3.7414,
"theoretical_loss": 4.858749113385144,
"tokens_seen": 87949312
},
{
"epoch": 0.03,
"learning_rate": 0.0004915847542627884,
"loss": 3.2873,
"theoretical_loss": 4.858238210816554,
"tokens_seen": 88014848
},
{
"epoch": 0.03,
"learning_rate": 0.0004915747241725176,
"loss": 3.5819,
"theoretical_loss": 4.8577277949522415,
"tokens_seen": 88080384
},
{
"epoch": 0.03,
"learning_rate": 0.0004915646940822468,
"loss": 3.6767,
"theoretical_loss": 4.857217864966943,
"tokens_seen": 88145920
},
{
"epoch": 0.03,
"learning_rate": 0.0004915546639919759,
"loss": 3.3539,
"theoretical_loss": 4.856708420037402,
"tokens_seen": 88211456
},
{
"epoch": 0.03,
"learning_rate": 0.0004915446339017051,
"loss": 3.8479,
"theoretical_loss": 4.8561994593423705,
"tokens_seen": 88276992
},
{
"epoch": 0.03,
"learning_rate": 0.0004915346038114343,
"loss": 3.5546,
"theoretical_loss": 4.8556909820625975,
"tokens_seen": 88342528
},
{
"epoch": 0.03,
"learning_rate": 0.0004915245737211635,
"loss": 3.8083,
"theoretical_loss": 4.855182987380823,
"tokens_seen": 88408064
},
{
"epoch": 0.03,
"objective/train/docs_used": 61597,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.9370856285095215,
"objective/train/theoretical_loss": 4.854675474481779,
"objective/train/tokens_used": 108933600,
"theoretical_loss": 4.854675474481779,
"tokens_seen": 88473600
},
{
"epoch": 0.03,
"learning_rate": 0.0004915145436308927,
"loss": 3.6345,
"theoretical_loss": 4.854675474481779,
"tokens_seen": 88473600
},
{
"epoch": 0.03,
"learning_rate": 0.0004915045135406218,
"loss": 3.4992,
"theoretical_loss": 4.8541684425521705,
"tokens_seen": 88539136
},
{
"epoch": 0.03,
"learning_rate": 0.0004914944834503511,
"loss": 3.4698,
"theoretical_loss": 4.85366189078068,
"tokens_seen": 88604672
},
{
"epoch": 0.03,
"learning_rate": 0.0004914844533600802,
"loss": 3.5404,
"theoretical_loss": 4.853155818357957,
"tokens_seen": 88670208
},
{
"epoch": 0.03,
"learning_rate": 0.0004914744232698095,
"loss": 3.5519,
"theoretical_loss": 4.852650224476609,
"tokens_seen": 88735744
},
{
"epoch": 0.03,
"learning_rate": 0.0004914643931795386,
"loss": 3.2668,
"theoretical_loss": 4.852145108331205,
"tokens_seen": 88801280
},
{
"epoch": 0.03,
"learning_rate": 0.0004914543630892679,
"loss": 3.7668,
"theoretical_loss": 4.851640469118255,
"tokens_seen": 88866816
},
{
"epoch": 0.03,
"learning_rate": 0.000491444332998997,
"loss": 3.6371,
"theoretical_loss": 4.851136306036219,
"tokens_seen": 88932352
},
{
"epoch": 0.03,
"learning_rate": 0.0004914343029087262,
"loss": 3.5714,
"theoretical_loss": 4.850632618285486,
"tokens_seen": 88997888
},
{
"epoch": 0.03,
"learning_rate": 0.0004914242728184554,
"loss": 3.4973,
"theoretical_loss": 4.850129405068383,
"tokens_seen": 89063424
},
{
"epoch": 0.03,
"learning_rate": 0.0004914142427281846,
"loss": 3.6218,
"theoretical_loss": 4.849626665589156,
"tokens_seen": 89128960
},
{
"epoch": 0.03,
"learning_rate": 0.0004914042126379138,
"loss": 3.3802,
"theoretical_loss": 4.849124399053969,
"tokens_seen": 89194496
},
{
"epoch": 0.03,
"learning_rate": 0.0004913941825476429,
"loss": 3.456,
"theoretical_loss": 4.8486226046709024,
"tokens_seen": 89260032
},
{
"epoch": 0.03,
"learning_rate": 0.0004913841524573721,
"loss": 3.6214,
"theoretical_loss": 4.8481212816499415,
"tokens_seen": 89325568
},
{
"epoch": 0.03,
"learning_rate": 0.0004913741223671013,
"loss": 3.7055,
"theoretical_loss": 4.847620429202967,
"tokens_seen": 89391104
},
{
"epoch": 0.03,
"learning_rate": 0.0004913640922768305,
"loss": 3.6888,
"theoretical_loss": 4.847120046543763,
"tokens_seen": 89456640
},
{
"epoch": 0.03,
"learning_rate": 0.0004913540621865597,
"loss": 3.803,
"theoretical_loss": 4.846620132887992,
"tokens_seen": 89522176
},
{
"epoch": 0.03,
"learning_rate": 0.0004913440320962888,
"loss": 3.392,
"theoretical_loss": 4.8461206874532055,
"tokens_seen": 89587712
},
{
"epoch": 0.03,
"learning_rate": 0.000491334002006018,
"loss": 3.5634,
"theoretical_loss": 4.845621709458831,
"tokens_seen": 89653248
},
{
"epoch": 0.03,
"learning_rate": 0.0004913239719157472,
"loss": 3.6485,
"theoretical_loss": 4.845123198126162,
"tokens_seen": 89718784
},
{
"epoch": 0.03,
"learning_rate": 0.0004913139418254765,
"loss": 3.6994,
"theoretical_loss": 4.844625152678364,
"tokens_seen": 89784320
},
{
"epoch": 0.03,
"learning_rate": 0.0004913039117352056,
"loss": 3.6592,
"theoretical_loss": 4.844127572340455,
"tokens_seen": 89849856
},
{
"epoch": 0.03,
"learning_rate": 0.0004912938816449349,
"loss": 3.6714,
"theoretical_loss": 4.84363045633931,
"tokens_seen": 89915392
},
{
"epoch": 0.03,
"learning_rate": 0.0004912838515546639,
"loss": 3.3055,
"theoretical_loss": 4.843133803903651,
"tokens_seen": 89980928
},
{
"epoch": 0.03,
"learning_rate": 0.0004912738214643932,
"loss": 3.6851,
"theoretical_loss": 4.84263761426404,
"tokens_seen": 90046464
},
{
"epoch": 0.03,
"objective/train/docs_used": 63098,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.415714979171753,
"objective/train/theoretical_loss": 4.842141886652876,
"objective/train/tokens_used": 110572000,
"theoretical_loss": 4.842141886652876,
"tokens_seen": 90112000
},
{
"epoch": 0.03,
"learning_rate": 0.0004912637913741224,
"loss": 3.496,
"theoretical_loss": 4.842141886652876,
"tokens_seen": 90112000
},
{
"epoch": 0.03,
"learning_rate": 0.0004912537612838516,
"loss": 3.7029,
"theoretical_loss": 4.841646620304388,
"tokens_seen": 90177536
},
{
"epoch": 0.03,
"learning_rate": 0.0004912437311935808,
"loss": 3.4109,
"theoretical_loss": 4.841151814454632,
"tokens_seen": 90243072
},
{
"epoch": 0.03,
"learning_rate": 0.00049123370110331,
"loss": 3.5139,
"theoretical_loss": 4.840657468341476,
"tokens_seen": 90308608
},
{
"epoch": 0.03,
"learning_rate": 0.0004912236710130391,
"loss": 3.7375,
"theoretical_loss": 4.84016358120461,
"tokens_seen": 90374144
},
{
"epoch": 0.03,
"learning_rate": 0.0004912136409227683,
"loss": 3.4378,
"theoretical_loss": 4.839670152285526,
"tokens_seen": 90439680
},
{
"epoch": 0.03,
"learning_rate": 0.0004912036108324975,
"loss": 3.4727,
"theoretical_loss": 4.8391771808275195,
"tokens_seen": 90505216
},
{
"epoch": 0.03,
"learning_rate": 0.0004911935807422267,
"loss": 3.4128,
"theoretical_loss": 4.838684666075682,
"tokens_seen": 90570752
},
{
"epoch": 0.03,
"learning_rate": 0.0004911835506519559,
"loss": 3.5232,
"theoretical_loss": 4.838192607276896,
"tokens_seen": 90636288
},
{
"epoch": 0.03,
"learning_rate": 0.000491173520561685,
"loss": 3.5274,
"theoretical_loss": 4.837701003679829,
"tokens_seen": 90701824
},
{
"epoch": 0.03,
"learning_rate": 0.0004911634904714142,
"loss": 3.573,
"theoretical_loss": 4.8372098545349305,
"tokens_seen": 90767360
},
{
"epoch": 0.03,
"learning_rate": 0.0004911534603811434,
"loss": 3.4979,
"theoretical_loss": 4.836719159094422,
"tokens_seen": 90832896
},
{
"epoch": 0.03,
"learning_rate": 0.0004911434302908726,
"loss": 3.7194,
"theoretical_loss": 4.836228916612292,
"tokens_seen": 90898432
},
{
"epoch": 0.03,
"learning_rate": 0.0004911334002006019,
"loss": 3.71,
"theoretical_loss": 4.835739126344298,
"tokens_seen": 90963968
},
{
"epoch": 0.03,
"learning_rate": 0.0004911233701103309,
"loss": 3.4962,
"theoretical_loss": 4.8352497875479505,
"tokens_seen": 91029504
},
{
"epoch": 0.03,
"learning_rate": 0.0004911133400200602,
"loss": 3.4739,
"theoretical_loss": 4.834760899482514,
"tokens_seen": 91095040
},
{
"epoch": 0.03,
"learning_rate": 0.0004911033099297893,
"loss": 3.452,
"theoretical_loss": 4.834272461409001,
"tokens_seen": 91160576
},
{
"epoch": 0.03,
"learning_rate": 0.0004910932798395186,
"loss": 3.8125,
"theoretical_loss": 4.833784472590165,
"tokens_seen": 91226112
},
{
"epoch": 0.03,
"learning_rate": 0.0004910832497492478,
"loss": 3.3993,
"theoretical_loss": 4.833296932290495,
"tokens_seen": 91291648
},
{
"epoch": 0.03,
"learning_rate": 0.000491073219658977,
"loss": 3.5064,
"theoretical_loss": 4.832809839776213,
"tokens_seen": 91357184
},
{
"epoch": 0.03,
"learning_rate": 0.0004910631895687061,
"loss": 3.4702,
"theoretical_loss": 4.832323194315265,
"tokens_seen": 91422720
},
{
"epoch": 0.03,
"learning_rate": 0.0004910531594784353,
"loss": 3.576,
"theoretical_loss": 4.831836995177319,
"tokens_seen": 91488256
},
{
"epoch": 0.03,
"learning_rate": 0.0004910431293881645,
"loss": 3.6835,
"theoretical_loss": 4.831351241633756,
"tokens_seen": 91553792
},
{
"epoch": 0.03,
"learning_rate": 0.0004910330992978937,
"loss": 3.6733,
"theoretical_loss": 4.8308659329576695,
"tokens_seen": 91619328
},
{
"epoch": 0.03,
"learning_rate": 0.0004910230692076229,
"loss": 3.7039,
"theoretical_loss": 4.830381068423856,
"tokens_seen": 91684864
},
{
"epoch": 0.03,
"objective/train/docs_used": 63840,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2570016384124756,
"objective/train/theoretical_loss": 4.8298966473088125,
"objective/train/tokens_used": 112210400,
"theoretical_loss": 4.8298966473088125,
"tokens_seen": 91750400
},
{
"epoch": 0.03,
"learning_rate": 0.000491013039117352,
"loss": 3.5049,
"theoretical_loss": 4.8298966473088125,
"tokens_seen": 91750400
},
{
"epoch": 0.03,
"learning_rate": 0.0004910030090270812,
"loss": 3.7075,
"theoretical_loss": 4.829412668890729,
"tokens_seen": 91815936
},
{
"epoch": 0.03,
"learning_rate": 0.0004909929789368104,
"loss": 3.3632,
"theoretical_loss": 4.8289291324494865,
"tokens_seen": 91881472
},
{
"epoch": 0.03,
"learning_rate": 0.0004909829488465397,
"loss": 3.4331,
"theoretical_loss": 4.828446037266647,
"tokens_seen": 91947008
},
{
"epoch": 0.03,
"learning_rate": 0.0004909729187562688,
"loss": 3.5859,
"theoretical_loss": 4.827963382625454,
"tokens_seen": 92012544
},
{
"epoch": 0.03,
"learning_rate": 0.0004909628886659981,
"loss": 3.4441,
"theoretical_loss": 4.827481167810825,
"tokens_seen": 92078080
},
{
"epoch": 0.03,
"learning_rate": 0.0004909528585757272,
"loss": 3.6517,
"theoretical_loss": 4.826999392109344,
"tokens_seen": 92143616
},
{
"epoch": 0.03,
"learning_rate": 0.0004909428284854564,
"loss": 3.4679,
"theoretical_loss": 4.826518054809259,
"tokens_seen": 92209152
},
{
"epoch": 0.03,
"learning_rate": 0.0004909327983951856,
"loss": 3.6007,
"theoretical_loss": 4.826037155200478,
"tokens_seen": 92274688
},
{
"epoch": 0.03,
"learning_rate": 0.0004909227683049148,
"loss": 3.4858,
"theoretical_loss": 4.825556692574562,
"tokens_seen": 92340224
},
{
"epoch": 0.03,
"learning_rate": 0.000490912738214644,
"loss": 3.5593,
"theoretical_loss": 4.825076666224717,
"tokens_seen": 92405760
},
{
"epoch": 0.03,
"learning_rate": 0.0004909027081243731,
"loss": 3.5697,
"theoretical_loss": 4.824597075445799,
"tokens_seen": 92471296
},
{
"epoch": 0.03,
"learning_rate": 0.0004908926780341023,
"loss": 3.5531,
"theoretical_loss": 4.824117919534297,
"tokens_seen": 92536832
},
{
"epoch": 0.03,
"learning_rate": 0.0004908826479438315,
"loss": 3.5856,
"theoretical_loss": 4.823639197788334,
"tokens_seen": 92602368
},
{
"epoch": 0.03,
"learning_rate": 0.0004908726178535607,
"loss": 3.6203,
"theoretical_loss": 4.823160909507665,
"tokens_seen": 92667904
},
{
"epoch": 0.03,
"learning_rate": 0.0004908625877632899,
"loss": 3.5859,
"theoretical_loss": 4.822683053993664,
"tokens_seen": 92733440
},
{
"epoch": 0.03,
"learning_rate": 0.000490852557673019,
"loss": 3.4464,
"theoretical_loss": 4.822205630549329,
"tokens_seen": 92798976
},
{
"epoch": 0.03,
"learning_rate": 0.0004908425275827482,
"loss": 3.1732,
"theoretical_loss": 4.821728638479267,
"tokens_seen": 92864512
},
{
"epoch": 0.03,
"learning_rate": 0.0004908324974924774,
"loss": 3.6278,
"theoretical_loss": 4.821252077089696,
"tokens_seen": 92930048
},
{
"epoch": 0.03,
"learning_rate": 0.0004908224674022067,
"loss": 3.6351,
"theoretical_loss": 4.820775945688437,
"tokens_seen": 92995584
},
{
"epoch": 0.03,
"learning_rate": 0.0004908124373119358,
"loss": 3.4334,
"theoretical_loss": 4.820300243584913,
"tokens_seen": 93061120
},
{
"epoch": 0.03,
"learning_rate": 0.0004908024072216651,
"loss": 3.4374,
"theoretical_loss": 4.819824970090138,
"tokens_seen": 93126656
},
{
"epoch": 0.03,
"learning_rate": 0.0004907923771313941,
"loss": 3.8253,
"theoretical_loss": 4.819350124516717,
"tokens_seen": 93192192
},
{
"epoch": 0.03,
"learning_rate": 0.0004907823470411234,
"loss": 3.5773,
"theoretical_loss": 4.818875706178841,
"tokens_seen": 93257728
},
{
"epoch": 0.03,
"learning_rate": 0.0004907723169508526,
"loss": 3.7615,
"theoretical_loss": 4.818401714392279,
"tokens_seen": 93323264
},
{
"epoch": 0.03,
"objective/train/docs_used": 65224,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.483311653137207,
"objective/train/theoretical_loss": 4.817928148474378,
"objective/train/tokens_used": 113848800,
"theoretical_loss": 4.817928148474378,
"tokens_seen": 93388800
},
{
"epoch": 0.03,
"learning_rate": 0.0004907622868605818,
"loss": 3.571,
"theoretical_loss": 4.817928148474378,
"tokens_seen": 93388800
},
{
"epoch": 0.03,
"learning_rate": 0.000490752256770311,
"loss": 3.5792,
"theoretical_loss": 4.817455007744052,
"tokens_seen": 93454336
},
{
"epoch": 0.03,
"learning_rate": 0.0004907422266800401,
"loss": 3.4639,
"theoretical_loss": 4.816982291521785,
"tokens_seen": 93519872
},
{
"epoch": 0.03,
"learning_rate": 0.0004907321965897693,
"loss": 3.3508,
"theoretical_loss": 4.816509999129618,
"tokens_seen": 93585408
},
{
"epoch": 0.03,
"learning_rate": 0.0004907221664994985,
"loss": 3.5891,
"theoretical_loss": 4.816038129891151,
"tokens_seen": 93650944
},
{
"epoch": 0.03,
"learning_rate": 0.0004907121364092277,
"loss": 3.4847,
"theoretical_loss": 4.815566683131536,
"tokens_seen": 93716480
},
{
"epoch": 0.03,
"learning_rate": 0.0004907021063189569,
"loss": 3.7072,
"theoretical_loss": 4.815095658177472,
"tokens_seen": 93782016
},
{
"epoch": 0.03,
"learning_rate": 0.000490692076228686,
"loss": 3.6021,
"theoretical_loss": 4.814625054357199,
"tokens_seen": 93847552
},
{
"epoch": 0.03,
"learning_rate": 0.0004906820461384152,
"loss": 3.4324,
"theoretical_loss": 4.814154871000497,
"tokens_seen": 93913088
},
{
"epoch": 0.03,
"learning_rate": 0.0004906720160481444,
"loss": 3.5133,
"theoretical_loss": 4.813685107438679,
"tokens_seen": 93978624
},
{
"epoch": 0.03,
"learning_rate": 0.0004906619859578736,
"loss": 3.5597,
"theoretical_loss": 4.813215763004585,
"tokens_seen": 94044160
},
{
"epoch": 0.03,
"learning_rate": 0.0004906519558676028,
"loss": 3.4239,
"theoretical_loss": 4.812746837032582,
"tokens_seen": 94109696
},
{
"epoch": 0.03,
"learning_rate": 0.0004906419257773321,
"loss": 3.373,
"theoretical_loss": 4.812278328858554,
"tokens_seen": 94175232
},
{
"epoch": 0.03,
"learning_rate": 0.0004906318956870611,
"loss": 3.5417,
"theoretical_loss": 4.811810237819904,
"tokens_seen": 94240768
},
{
"epoch": 0.03,
"learning_rate": 0.0004906218655967904,
"loss": 3.3743,
"theoretical_loss": 4.81134256325554,
"tokens_seen": 94306304
},
{
"epoch": 0.03,
"learning_rate": 0.0004906118355065195,
"loss": 3.5175,
"theoretical_loss": 4.810875304505881,
"tokens_seen": 94371840
},
{
"epoch": 0.03,
"learning_rate": 0.0004906018054162488,
"loss": 3.3378,
"theoretical_loss": 4.810408460912846,
"tokens_seen": 94437376
},
{
"epoch": 0.03,
"learning_rate": 0.000490591775325978,
"loss": 3.8392,
"theoretical_loss": 4.809942031819853,
"tokens_seen": 94502912
},
{
"epoch": 0.03,
"learning_rate": 0.0004905817452357072,
"loss": 3.4628,
"theoretical_loss": 4.809476016571809,
"tokens_seen": 94568448
},
{
"epoch": 0.03,
"learning_rate": 0.0004905717151454363,
"loss": 3.699,
"theoretical_loss": 4.809010414515113,
"tokens_seen": 94633984
},
{
"epoch": 0.03,
"learning_rate": 0.0004905616850551655,
"loss": 3.6345,
"theoretical_loss": 4.808545224997644,
"tokens_seen": 94699520
},
{
"epoch": 0.03,
"learning_rate": 0.0004905516549648947,
"loss": 3.3299,
"theoretical_loss": 4.808080447368766,
"tokens_seen": 94765056
},
{
"epoch": 0.03,
"learning_rate": 0.0004905416248746239,
"loss": 3.5958,
"theoretical_loss": 4.807616080979315,
"tokens_seen": 94830592
},
{
"epoch": 0.03,
"learning_rate": 0.0004905315947843531,
"loss": 3.752,
"theoretical_loss": 4.807152125181597,
"tokens_seen": 94896128
},
{
"epoch": 0.03,
"learning_rate": 0.0004905215646940822,
"loss": 3.7867,
"theoretical_loss": 4.806688579329387,
"tokens_seen": 94961664
},
{
"epoch": 0.03,
"objective/train/docs_used": 65494,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3390982151031494,
"objective/train/theoretical_loss": 4.8062254427779205,
"objective/train/tokens_used": 115487200,
"theoretical_loss": 4.8062254427779205,
"tokens_seen": 95027200
},
{
"epoch": 0.03,
"learning_rate": 0.0004905115346038114,
"loss": 3.5332,
"theoretical_loss": 4.8062254427779205,
"tokens_seen": 95027200
},
{
"epoch": 0.03,
"learning_rate": 0.0004905015045135406,
"loss": 3.6865,
"theoretical_loss": 4.80576271488389,
"tokens_seen": 95092736
},
{
"epoch": 0.03,
"learning_rate": 0.0004904914744232698,
"loss": 3.6984,
"theoretical_loss": 4.805300395005444,
"tokens_seen": 95158272
},
{
"epoch": 0.03,
"learning_rate": 0.000490481444332999,
"loss": 3.7887,
"theoretical_loss": 4.804838482502181,
"tokens_seen": 95223808
},
{
"epoch": 0.03,
"learning_rate": 0.0004904714142427281,
"loss": 3.5508,
"theoretical_loss": 4.8043769767351385,
"tokens_seen": 95289344
},
{
"epoch": 0.03,
"learning_rate": 0.0004904613841524574,
"loss": 3.2765,
"theoretical_loss": 4.8039158770668005,
"tokens_seen": 95354880
},
{
"epoch": 0.03,
"learning_rate": 0.0004904513540621865,
"loss": 3.4666,
"theoretical_loss": 4.803455182861087,
"tokens_seen": 95420416
},
{
"epoch": 0.03,
"learning_rate": 0.0004904413239719158,
"loss": 3.4331,
"theoretical_loss": 4.802994893483348,
"tokens_seen": 95485952
},
{
"epoch": 0.03,
"learning_rate": 0.0004904312938816449,
"loss": 3.4208,
"theoretical_loss": 4.802535008300364,
"tokens_seen": 95551488
},
{
"epoch": 0.03,
"learning_rate": 0.0004904212637913742,
"loss": 3.3395,
"theoretical_loss": 4.802075526680335,
"tokens_seen": 95617024
},
{
"epoch": 0.03,
"learning_rate": 0.0004904112337011033,
"loss": 3.3708,
"theoretical_loss": 4.801616447992888,
"tokens_seen": 95682560
},
{
"epoch": 0.03,
"learning_rate": 0.0004904012036108325,
"loss": 3.6168,
"theoretical_loss": 4.801157771609061,
"tokens_seen": 95748096
},
{
"epoch": 0.03,
"learning_rate": 0.0004903911735205617,
"loss": 3.3959,
"theoretical_loss": 4.8006994969013,
"tokens_seen": 95813632
},
{
"epoch": 0.03,
"learning_rate": 0.0004903811434302909,
"loss": 3.6339,
"theoretical_loss": 4.800241623243467,
"tokens_seen": 95879168
},
{
"epoch": 0.03,
"learning_rate": 0.0004903711133400201,
"loss": 3.3917,
"theoretical_loss": 4.799784150010819,
"tokens_seen": 95944704
},
{
"epoch": 0.03,
"learning_rate": 0.0004903610832497492,
"loss": 3.6765,
"theoretical_loss": 4.799327076580017,
"tokens_seen": 96010240
},
{
"epoch": 0.03,
"learning_rate": 0.0004903510531594784,
"loss": 3.6297,
"theoretical_loss": 4.798870402329115,
"tokens_seen": 96075776
},
{
"epoch": 0.03,
"learning_rate": 0.0004903410230692076,
"loss": 3.4202,
"theoretical_loss": 4.798414126637558,
"tokens_seen": 96141312
},
{
"epoch": 0.03,
"learning_rate": 0.0004903309929789368,
"loss": 3.417,
"theoretical_loss": 4.797958248886179,
"tokens_seen": 96206848
},
{
"epoch": 0.03,
"learning_rate": 0.000490320962888666,
"loss": 3.5827,
"theoretical_loss": 4.797502768457193,
"tokens_seen": 96272384
},
{
"epoch": 0.03,
"learning_rate": 0.0004903109327983952,
"loss": 3.5392,
"theoretical_loss": 4.797047684734192,
"tokens_seen": 96337920
},
{
"epoch": 0.03,
"learning_rate": 0.0004903009027081243,
"loss": 3.744,
"theoretical_loss": 4.796592997102147,
"tokens_seen": 96403456
},
{
"epoch": 0.03,
"learning_rate": 0.0004902908726178535,
"loss": 3.823,
"theoretical_loss": 4.796138704947397,
"tokens_seen": 96468992
},
{
"epoch": 0.03,
"learning_rate": 0.0004902808425275828,
"loss": 3.475,
"theoretical_loss": 4.795684807657649,
"tokens_seen": 96534528
},
{
"epoch": 0.03,
"learning_rate": 0.0004902708124373119,
"loss": 3.6812,
"theoretical_loss": 4.795231304621968,
"tokens_seen": 96600064
},
{
"epoch": 0.03,
"objective/train/docs_used": 66055,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.559401750564575,
"objective/train/theoretical_loss": 4.794778195230787,
"objective/train/tokens_used": 117125600,
"theoretical_loss": 4.794778195230787,
"tokens_seen": 96665600
},
{
"epoch": 0.03,
"learning_rate": 0.0004902607823470412,
"loss": 3.6274,
"theoretical_loss": 4.794778195230787,
"tokens_seen": 96665600
},
{
"epoch": 0.03,
"learning_rate": 0.0004902507522567703,
"loss": 3.5801,
"theoretical_loss": 4.794325478875885,
"tokens_seen": 96731136
},
{
"epoch": 0.03,
"learning_rate": 0.0004902407221664995,
"loss": 3.6061,
"theoretical_loss": 4.793873154950399,
"tokens_seen": 96796672
},
{
"epoch": 0.03,
"learning_rate": 0.0004902306920762287,
"loss": 3.6615,
"theoretical_loss": 4.793421222848808,
"tokens_seen": 96862208
},
{
"epoch": 0.03,
"learning_rate": 0.0004902206619859579,
"loss": 3.3969,
"theoretical_loss": 4.7929696819669365,
"tokens_seen": 96927744
},
{
"epoch": 0.03,
"learning_rate": 0.0004902106318956871,
"loss": 3.8594,
"theoretical_loss": 4.792518531701948,
"tokens_seen": 96993280
},
{
"epoch": 0.03,
"learning_rate": 0.0004902006018054163,
"loss": 3.6387,
"theoretical_loss": 4.792067771452341,
"tokens_seen": 97058816
},
{
"epoch": 0.03,
"learning_rate": 0.0004901905717151454,
"loss": 3.4593,
"theoretical_loss": 4.791617400617948,
"tokens_seen": 97124352
},
{
"epoch": 0.03,
"learning_rate": 0.0004901805416248746,
"loss": 3.5499,
"theoretical_loss": 4.791167418599925,
"tokens_seen": 97189888
},
{
"epoch": 0.03,
"learning_rate": 0.0004901705115346038,
"loss": 3.3165,
"theoretical_loss": 4.790717824800755,
"tokens_seen": 97255424
},
{
"epoch": 0.03,
"learning_rate": 0.000490160481444333,
"loss": 3.5253,
"theoretical_loss": 4.790268618624239,
"tokens_seen": 97320960
},
{
"epoch": 0.03,
"learning_rate": 0.0004901504513540623,
"loss": 3.2939,
"theoretical_loss": 4.789819799475499,
"tokens_seen": 97386496
},
{
"epoch": 0.03,
"learning_rate": 0.0004901404212637913,
"loss": 3.7183,
"theoretical_loss": 4.789371366760961,
"tokens_seen": 97452032
},
{
"epoch": 0.03,
"learning_rate": 0.0004901303911735206,
"loss": 3.3984,
"theoretical_loss": 4.788923319888369,
"tokens_seen": 97517568
},
{
"epoch": 0.03,
"learning_rate": 0.0004901203610832497,
"loss": 3.4614,
"theoretical_loss": 4.788475658266766,
"tokens_seen": 97583104
},
{
"epoch": 0.03,
"learning_rate": 0.000490110330992979,
"loss": 3.2819,
"theoretical_loss": 4.788028381306497,
"tokens_seen": 97648640
},
{
"epoch": 0.03,
"learning_rate": 0.0004901003009027082,
"loss": 3.3862,
"theoretical_loss": 4.787581488419207,
"tokens_seen": 97714176
},
{
"epoch": 0.03,
"learning_rate": 0.0004900902708124374,
"loss": 3.7962,
"theoretical_loss": 4.787134979017832,
"tokens_seen": 97779712
},
{
"epoch": 0.03,
"learning_rate": 0.0004900802407221665,
"loss": 3.3423,
"theoretical_loss": 4.786688852516599,
"tokens_seen": 97845248
},
{
"epoch": 0.03,
"learning_rate": 0.0004900702106318957,
"loss": 3.5008,
"theoretical_loss": 4.786243108331024,
"tokens_seen": 97910784
},
{
"epoch": 0.03,
"learning_rate": 0.0004900601805416249,
"loss": 3.736,
"theoretical_loss": 4.7857977458779,
"tokens_seen": 97976320
},
{
"epoch": 0.03,
"learning_rate": 0.0004900501504513541,
"loss": 3.6437,
"theoretical_loss": 4.785352764575304,
"tokens_seen": 98041856
},
{
"epoch": 0.03,
"learning_rate": 0.0004900401203610833,
"loss": 3.4917,
"theoretical_loss": 4.784908163842585,
"tokens_seen": 98107392
},
{
"epoch": 0.03,
"learning_rate": 0.0004900300902708124,
"loss": 3.6996,
"theoretical_loss": 4.784463943100367,
"tokens_seen": 98172928
},
{
"epoch": 0.03,
"learning_rate": 0.0004900200601805416,
"loss": 3.5085,
"theoretical_loss": 4.7840201017705395,
"tokens_seen": 98238464
},
{
"epoch": 0.03,
"objective/train/docs_used": 66835,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 4.026576995849609,
"objective/train/theoretical_loss": 4.783576639276257,
"objective/train/tokens_used": 118764000,
"theoretical_loss": 4.783576639276257,
"tokens_seen": 98304000
},
{
"epoch": 0.03,
"learning_rate": 0.0004900100300902708,
"loss": 3.6122,
"theoretical_loss": 4.783576639276257,
"tokens_seen": 98304000
},
{
"epoch": 0.03,
"learning_rate": 0.00049,
"loss": 3.6488,
"theoretical_loss": 4.783133555041934,
"tokens_seen": 98369536
},
{
"epoch": 0.03,
"learning_rate": 0.0004899899699097292,
"loss": 3.4521,
"theoretical_loss": 4.782690848493245,
"tokens_seen": 98435072
},
{
"epoch": 0.03,
"learning_rate": 0.0004899799398194583,
"loss": 3.4946,
"theoretical_loss": 4.7822485190571165,
"tokens_seen": 98500608
},
{
"epoch": 0.03,
"learning_rate": 0.0004899699097291876,
"loss": 3.1609,
"theoretical_loss": 4.781806566161723,
"tokens_seen": 98566144
},
{
"epoch": 0.03,
"learning_rate": 0.0004899598796389167,
"loss": 3.2605,
"theoretical_loss": 4.781364989236488,
"tokens_seen": 98631680
},
{
"epoch": 0.03,
"learning_rate": 0.000489949849548646,
"loss": 3.5803,
"theoretical_loss": 4.78092378771208,
"tokens_seen": 98697216
},
{
"epoch": 0.03,
"learning_rate": 0.0004899398194583751,
"loss": 3.6634,
"theoretical_loss": 4.780482961020402,
"tokens_seen": 98762752
},
{
"epoch": 0.03,
"learning_rate": 0.0004899297893681044,
"loss": 3.6734,
"theoretical_loss": 4.780042508594596,
"tokens_seen": 98828288
},
{
"epoch": 0.03,
"learning_rate": 0.0004899197592778335,
"loss": 3.6452,
"theoretical_loss": 4.779602429869035,
"tokens_seen": 98893824
},
{
"epoch": 0.03,
"learning_rate": 0.0004899097291875627,
"loss": 3.4789,
"theoretical_loss": 4.779162724279324,
"tokens_seen": 98959360
},
{
"epoch": 0.03,
"learning_rate": 0.0004898996990972919,
"loss": 3.5272,
"theoretical_loss": 4.7787233912622895,
"tokens_seen": 99024896
},
{
"epoch": 0.03,
"learning_rate": 0.0004898896690070211,
"loss": 3.9395,
"theoretical_loss": 4.778284430255981,
"tokens_seen": 99090432
},
{
"epoch": 0.03,
"learning_rate": 0.0004898796389167503,
"loss": 3.803,
"theoretical_loss": 4.77784584069967,
"tokens_seen": 99155968
},
{
"epoch": 0.03,
"learning_rate": 0.0004898696088264794,
"loss": 3.5448,
"theoretical_loss": 4.777407622033838,
"tokens_seen": 99221504
},
{
"epoch": 0.03,
"learning_rate": 0.0004898595787362086,
"loss": 3.3819,
"theoretical_loss": 4.776969773700181,
"tokens_seen": 99287040
},
{
"epoch": 0.03,
"learning_rate": 0.0004898495486459378,
"loss": 3.5351,
"theoretical_loss": 4.776532295141601,
"tokens_seen": 99352576
},
{
"epoch": 0.03,
"learning_rate": 0.000489839518555667,
"loss": 3.6881,
"theoretical_loss": 4.776095185802211,
"tokens_seen": 99418112
},
{
"epoch": 0.03,
"learning_rate": 0.0004898294884653962,
"loss": 3.8074,
"theoretical_loss": 4.775658445127318,
"tokens_seen": 99483648
},
{
"epoch": 0.03,
"learning_rate": 0.0004898194583751254,
"loss": 3.4411,
"theoretical_loss": 4.775222072563429,
"tokens_seen": 99549184
},
{
"epoch": 0.03,
"learning_rate": 0.0004898094282848545,
"loss": 3.4496,
"theoretical_loss": 4.7747860675582485,
"tokens_seen": 99614720
},
{
"epoch": 0.03,
"learning_rate": 0.0004897993981945837,
"loss": 3.348,
"theoretical_loss": 4.77435042956067,
"tokens_seen": 99680256
},
{
"epoch": 0.03,
"learning_rate": 0.000489789368104313,
"loss": 3.5864,
"theoretical_loss": 4.773915158020776,
"tokens_seen": 99745792
},
{
"epoch": 0.03,
"learning_rate": 0.0004897793380140421,
"loss": 3.4012,
"theoretical_loss": 4.773480252389831,
"tokens_seen": 99811328
},
{
"epoch": 0.03,
"learning_rate": 0.0004897693079237714,
"loss": 3.6203,
"theoretical_loss": 4.773045712120284,
"tokens_seen": 99876864
},
{
"epoch": 0.03,
"objective/train/docs_used": 67525,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.8278653621673584,
"objective/train/theoretical_loss": 4.77261153666576,
"objective/train/tokens_used": 120402400,
"theoretical_loss": 4.77261153666576,
"tokens_seen": 99942400
},
{
"epoch": 0.03,
"learning_rate": 0.0004897592778335005,
"loss": 3.5655,
"theoretical_loss": 4.77261153666576,
"tokens_seen": 99942400
},
{
"epoch": 0.03,
"learning_rate": 0.0004897492477432297,
"loss": 3.6504,
"theoretical_loss": 4.772177725481062,
"tokens_seen": 100007936
},
{
"epoch": 0.03,
"learning_rate": 0.0004897392176529589,
"loss": 3.5848,
"theoretical_loss": 4.77174427802216,
"tokens_seen": 100073472
},
{
"epoch": 0.03,
"learning_rate": 0.0004897291875626881,
"loss": 3.6098,
"theoretical_loss": 4.771311193746191,
"tokens_seen": 100139008
},
{
"epoch": 0.03,
"learning_rate": 0.0004897191574724173,
"loss": 3.2068,
"theoretical_loss": 4.770878472111465,
"tokens_seen": 100204544
},
{
"epoch": 0.03,
"learning_rate": 0.0004897091273821465,
"loss": 3.6295,
"theoretical_loss": 4.770446112577445,
"tokens_seen": 100270080
},
{
"epoch": 0.03,
"learning_rate": 0.0004896990972918756,
"loss": 3.4111,
"theoretical_loss": 4.770014114604756,
"tokens_seen": 100335616
},
{
"epoch": 0.03,
"learning_rate": 0.0004896890672016048,
"loss": 3.5571,
"theoretical_loss": 4.769582477655177,
"tokens_seen": 100401152
},
{
"epoch": 0.03,
"learning_rate": 0.000489679037111334,
"loss": 3.2238,
"theoretical_loss": 4.769151201191641,
"tokens_seen": 100466688
},
{
"epoch": 0.03,
"learning_rate": 0.0004896690070210632,
"loss": 3.501,
"theoretical_loss": 4.768720284678228,
"tokens_seen": 100532224
},
{
"epoch": 0.03,
"learning_rate": 0.0004896589769307924,
"loss": 3.6858,
"theoretical_loss": 4.768289727580161,
"tokens_seen": 100597760
},
{
"epoch": 0.03,
"learning_rate": 0.0004896489468405215,
"loss": 3.6342,
"theoretical_loss": 4.767859529363809,
"tokens_seen": 100663296
},
{
"epoch": 0.03,
"learning_rate": 0.0004896389167502507,
"loss": 3.3918,
"theoretical_loss": 4.767429689496682,
"tokens_seen": 100728832
},
{
"epoch": 0.03,
"learning_rate": 0.0004896288866599799,
"loss": 3.4626,
"theoretical_loss": 4.767000207447417,
"tokens_seen": 100794368
},
{
"epoch": 0.03,
"learning_rate": 0.0004896188565697091,
"loss": 3.4291,
"theoretical_loss": 4.766571082685794,
"tokens_seen": 100859904
},
{
"epoch": 0.03,
"learning_rate": 0.0004896088264794384,
"loss": 3.4262,
"theoretical_loss": 4.766142314682716,
"tokens_seen": 100925440
},
{
"epoch": 0.03,
"learning_rate": 0.0004895987963891674,
"loss": 3.8865,
"theoretical_loss": 4.765713902910214,
"tokens_seen": 100990976
},
{
"epoch": 0.03,
"learning_rate": 0.0004895887662988967,
"loss": 3.4615,
"theoretical_loss": 4.765285846841444,
"tokens_seen": 101056512
},
{
"epoch": 0.03,
"learning_rate": 0.0004895787362086259,
"loss": 3.4272,
"theoretical_loss": 4.76485814595068,
"tokens_seen": 101122048
},
{
"epoch": 0.03,
"learning_rate": 0.0004895687061183551,
"loss": 3.5119,
"theoretical_loss": 4.764430799713314,
"tokens_seen": 101187584
},
{
"epoch": 0.03,
"learning_rate": 0.0004895586760280843,
"loss": 3.6724,
"theoretical_loss": 4.764003807605853,
"tokens_seen": 101253120
},
{
"epoch": 0.03,
"learning_rate": 0.0004895486459378135,
"loss": 3.6045,
"theoretical_loss": 4.763577169105912,
"tokens_seen": 101318656
},
{
"epoch": 0.03,
"learning_rate": 0.0004895386158475426,
"loss": 3.4985,
"theoretical_loss": 4.763150883692218,
"tokens_seen": 101384192
},
{
"epoch": 0.03,
"learning_rate": 0.0004895285857572718,
"loss": 3.5738,
"theoretical_loss": 4.762724950844598,
"tokens_seen": 101449728
},
{
"epoch": 0.03,
"learning_rate": 0.000489518555667001,
"loss": 3.4495,
"theoretical_loss": 4.762299370043984,
"tokens_seen": 101515264
},
{
"epoch": 0.03,
"objective/train/docs_used": 68617,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4468700885772705,
"objective/train/theoretical_loss": 4.761874140772408,
"objective/train/tokens_used": 122040800,
"theoretical_loss": 4.761874140772408,
"tokens_seen": 101580800
},
{
"epoch": 0.03,
"learning_rate": 0.0004895085255767302,
"loss": 3.5015,
"theoretical_loss": 4.761874140772408,
"tokens_seen": 101580800
},
{
"epoch": 0.03,
"learning_rate": 0.0004894984954864594,
"loss": 3.7421,
"theoretical_loss": 4.761449262512993,
"tokens_seen": 101646336
},
{
"epoch": 0.03,
"learning_rate": 0.0004894884653961885,
"loss": 3.8129,
"theoretical_loss": 4.761024734749958,
"tokens_seen": 101711872
},
{
"epoch": 0.03,
"learning_rate": 0.0004894784353059178,
"loss": 3.7874,
"theoretical_loss": 4.76060055696861,
"tokens_seen": 101777408
},
{
"epoch": 0.03,
"learning_rate": 0.0004894684052156469,
"loss": 3.7841,
"theoretical_loss": 4.760176728655345,
"tokens_seen": 101842944
},
{
"epoch": 0.03,
"learning_rate": 0.0004894583751253762,
"loss": 3.4924,
"theoretical_loss": 4.75975324929764,
"tokens_seen": 101908480
},
{
"epoch": 0.03,
"learning_rate": 0.0004894483450351053,
"loss": 3.3683,
"theoretical_loss": 4.759330118384053,
"tokens_seen": 101974016
},
{
"epoch": 0.03,
"learning_rate": 0.0004894383149448346,
"loss": 3.7427,
"theoretical_loss": 4.758907335404221,
"tokens_seen": 102039552
},
{
"epoch": 0.03,
"learning_rate": 0.0004894282848545637,
"loss": 3.6284,
"theoretical_loss": 4.758484899848854,
"tokens_seen": 102105088
},
{
"epoch": 0.03,
"learning_rate": 0.0004894182547642929,
"loss": 3.4244,
"theoretical_loss": 4.7580628112097365,
"tokens_seen": 102170624
},
{
"epoch": 0.03,
"learning_rate": 0.0004894082246740221,
"loss": 3.4994,
"theoretical_loss": 4.7576410689797175,
"tokens_seen": 102236160
},
{
"epoch": 0.03,
"learning_rate": 0.0004893981945837513,
"loss": 3.5959,
"theoretical_loss": 4.757219672652717,
"tokens_seen": 102301696
},
{
"epoch": 0.03,
"learning_rate": 0.0004893881644934805,
"loss": 3.1601,
"theoretical_loss": 4.756798621723712,
"tokens_seen": 102367232
},
{
"epoch": 0.03,
"learning_rate": 0.0004893781344032096,
"loss": 3.6548,
"theoretical_loss": 4.756377915688748,
"tokens_seen": 102432768
},
{
"epoch": 0.03,
"learning_rate": 0.0004893681043129388,
"loss": 3.4761,
"theoretical_loss": 4.755957554044917,
"tokens_seen": 102498304
},
{
"epoch": 0.03,
"learning_rate": 0.000489358074222668,
"loss": 3.5187,
"theoretical_loss": 4.755537536290373,
"tokens_seen": 102563840
},
{
"epoch": 0.03,
"learning_rate": 0.0004893480441323972,
"loss": 3.4874,
"theoretical_loss": 4.755117861924321,
"tokens_seen": 102629376
},
{
"epoch": 0.03,
"learning_rate": 0.0004893380140421264,
"loss": 3.387,
"theoretical_loss": 4.754698530447009,
"tokens_seen": 102694912
},
{
"epoch": 0.03,
"learning_rate": 0.0004893279839518556,
"loss": 3.3774,
"theoretical_loss": 4.754279541359738,
"tokens_seen": 102760448
},
{
"epoch": 0.03,
"learning_rate": 0.0004893179538615847,
"loss": 3.4792,
"theoretical_loss": 4.753860894164845,
"tokens_seen": 102825984
},
{
"epoch": 0.03,
"learning_rate": 0.0004893079237713139,
"loss": 3.5093,
"theoretical_loss": 4.75344258836571,
"tokens_seen": 102891520
},
{
"epoch": 0.03,
"learning_rate": 0.0004892978936810432,
"loss": 3.5069,
"theoretical_loss": 4.753024623466752,
"tokens_seen": 102957056
},
{
"epoch": 0.03,
"learning_rate": 0.0004892878635907723,
"loss": 3.7293,
"theoretical_loss": 4.752606998973421,
"tokens_seen": 103022592
},
{
"epoch": 0.03,
"learning_rate": 0.0004892778335005016,
"loss": 3.5777,
"theoretical_loss": 4.752189714392202,
"tokens_seen": 103088128
},
{
"epoch": 0.03,
"learning_rate": 0.0004892678034102307,
"loss": 3.5236,
"theoretical_loss": 4.7517727692306035,
"tokens_seen": 103153664
},
{
"epoch": 0.03,
"objective/train/docs_used": 69265,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0723447799682617,
"objective/train/theoretical_loss": 4.751356162997164,
"objective/train/tokens_used": 123679200,
"theoretical_loss": 4.751356162997164,
"tokens_seen": 103219200
},
{
"epoch": 0.03,
"learning_rate": 0.0004892577733199599,
"loss": 3.525,
"theoretical_loss": 4.751356162997164,
"tokens_seen": 103219200
},
{
"epoch": 0.03,
"learning_rate": 0.0004892477432296891,
"loss": 3.602,
"theoretical_loss": 4.750939895201443,
"tokens_seen": 103284736
},
{
"epoch": 0.03,
"learning_rate": 0.0004892377131394183,
"loss": 3.4738,
"theoretical_loss": 4.750523965354024,
"tokens_seen": 103350272
},
{
"epoch": 0.03,
"learning_rate": 0.0004892276830491475,
"loss": 3.3909,
"theoretical_loss": 4.750108372966501,
"tokens_seen": 103415808
},
{
"epoch": 0.03,
"learning_rate": 0.0004892176529588767,
"loss": 3.2759,
"theoretical_loss": 4.749693117551491,
"tokens_seen": 103481344
},
{
"epoch": 0.03,
"learning_rate": 0.0004892076228686058,
"loss": 3.7655,
"theoretical_loss": 4.749278198622617,
"tokens_seen": 103546880
},
{
"epoch": 0.03,
"learning_rate": 0.000489197592778335,
"loss": 3.4825,
"theoretical_loss": 4.748863615694514,
"tokens_seen": 103612416
},
{
"epoch": 0.03,
"learning_rate": 0.0004891875626880642,
"loss": 3.7804,
"theoretical_loss": 4.748449368282822,
"tokens_seen": 103677952
},
{
"epoch": 0.03,
"learning_rate": 0.0004891775325977934,
"loss": 3.4161,
"theoretical_loss": 4.748035455904185,
"tokens_seen": 103743488
},
{
"epoch": 0.03,
"learning_rate": 0.0004891675025075226,
"loss": 3.5973,
"theoretical_loss": 4.747621878076252,
"tokens_seen": 103809024
},
{
"epoch": 0.03,
"learning_rate": 0.0004891574724172517,
"loss": 3.5654,
"theoretical_loss": 4.747208634317664,
"tokens_seen": 103874560
},
{
"epoch": 0.03,
"learning_rate": 0.0004891474423269809,
"loss": 3.6798,
"theoretical_loss": 4.746795724148061,
"tokens_seen": 103940096
},
{
"epoch": 0.03,
"learning_rate": 0.0004891374122367101,
"loss": 3.7902,
"theoretical_loss": 4.746383147088078,
"tokens_seen": 104005632
},
{
"epoch": 0.03,
"learning_rate": 0.0004891273821464393,
"loss": 3.2944,
"theoretical_loss": 4.745970902659338,
"tokens_seen": 104071168
},
{
"epoch": 0.03,
"learning_rate": 0.0004891173520561686,
"loss": 3.6076,
"theoretical_loss": 4.745558990384451,
"tokens_seen": 104136704
},
{
"epoch": 0.03,
"learning_rate": 0.0004891073219658976,
"loss": 3.3813,
"theoretical_loss": 4.7451474097870125,
"tokens_seen": 104202240
},
{
"epoch": 0.03,
"learning_rate": 0.0004890972918756269,
"loss": 3.245,
"theoretical_loss": 4.744736160391602,
"tokens_seen": 104267776
},
{
"epoch": 0.03,
"learning_rate": 0.0004890872617853561,
"loss": 3.6455,
"theoretical_loss": 4.744325241723777,
"tokens_seen": 104333312
},
{
"epoch": 0.03,
"learning_rate": 0.0004890772316950853,
"loss": 3.6057,
"theoretical_loss": 4.743914653310073,
"tokens_seen": 104398848
},
{
"epoch": 0.03,
"learning_rate": 0.0004890672016048145,
"loss": 3.2381,
"theoretical_loss": 4.743504394678,
"tokens_seen": 104464384
},
{
"epoch": 0.03,
"learning_rate": 0.0004890571715145437,
"loss": 3.5268,
"theoretical_loss": 4.743094465356039,
"tokens_seen": 104529920
},
{
"epoch": 0.03,
"learning_rate": 0.0004890471414242728,
"loss": 3.4362,
"theoretical_loss": 4.742684864873641,
"tokens_seen": 104595456
},
{
"epoch": 0.03,
"learning_rate": 0.000489037111334002,
"loss": 3.2982,
"theoretical_loss": 4.742275592761223,
"tokens_seen": 104660992
},
{
"epoch": 0.03,
"learning_rate": 0.0004890270812437312,
"loss": 3.5981,
"theoretical_loss": 4.741866648550168,
"tokens_seen": 104726528
},
{
"epoch": 0.03,
"learning_rate": 0.0004890170511534604,
"loss": 3.5663,
"theoretical_loss": 4.741458031772817,
"tokens_seen": 104792064
},
{
"epoch": 0.03,
"objective/train/docs_used": 70556,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0196893215179443,
"objective/train/theoretical_loss": 4.741049741962473,
"objective/train/tokens_used": 125317600,
"theoretical_loss": 4.741049741962473,
"tokens_seen": 104857600
},
{
"epoch": 0.03,
"learning_rate": 0.0004890070210631896,
"loss": 3.3116,
"theoretical_loss": 4.741049741962473,
"tokens_seen": 104857600
},
{
"epoch": 0.03,
"learning_rate": 0.0004889969909729187,
"loss": 3.3268,
"theoretical_loss": 4.740641778653395,
"tokens_seen": 104923136
},
{
"epoch": 0.03,
"learning_rate": 0.0004889869608826479,
"loss": 3.535,
"theoretical_loss": 4.740234141380794,
"tokens_seen": 104988672
},
{
"epoch": 0.03,
"learning_rate": 0.0004889769307923771,
"loss": 3.5609,
"theoretical_loss": 4.739826829680833,
"tokens_seen": 105054208
},
{
"epoch": 0.03,
"learning_rate": 0.0004889669007021063,
"loss": 3.2398,
"theoretical_loss": 4.739419843090626,
"tokens_seen": 105119744
},
{
"epoch": 0.03,
"learning_rate": 0.0004889568706118355,
"loss": 3.6536,
"theoretical_loss": 4.739013181148229,
"tokens_seen": 105185280
},
{
"epoch": 0.03,
"learning_rate": 0.0004889468405215647,
"loss": 3.2403,
"theoretical_loss": 4.738606843392644,
"tokens_seen": 105250816
},
{
"epoch": 0.03,
"learning_rate": 0.0004889368104312939,
"loss": 3.3196,
"theoretical_loss": 4.738200829363815,
"tokens_seen": 105316352
},
{
"epoch": 0.03,
"learning_rate": 0.000488926780341023,
"loss": 3.5581,
"theoretical_loss": 4.737795138602624,
"tokens_seen": 105381888
},
{
"epoch": 0.03,
"learning_rate": 0.0004889167502507523,
"loss": 3.478,
"theoretical_loss": 4.737389770650887,
"tokens_seen": 105447424
},
{
"epoch": 0.03,
"learning_rate": 0.0004889067201604815,
"loss": 3.4703,
"theoretical_loss": 4.736984725051357,
"tokens_seen": 105512960
},
{
"epoch": 0.03,
"learning_rate": 0.0004888966900702107,
"loss": 3.5458,
"theoretical_loss": 4.736580001347717,
"tokens_seen": 105578496
},
{
"epoch": 0.03,
"learning_rate": 0.0004888866599799398,
"loss": 3.5523,
"theoretical_loss": 4.736175599084576,
"tokens_seen": 105644032
},
{
"epoch": 0.03,
"learning_rate": 0.000488876629889669,
"loss": 3.4106,
"theoretical_loss": 4.735771517807473,
"tokens_seen": 105709568
},
{
"epoch": 0.03,
"learning_rate": 0.0004888665997993982,
"loss": 3.6541,
"theoretical_loss": 4.735367757062869,
"tokens_seen": 105775104
},
{
"epoch": 0.03,
"learning_rate": 0.0004888565697091274,
"loss": 3.4943,
"theoretical_loss": 4.734964316398148,
"tokens_seen": 105840640
},
{
"epoch": 0.03,
"learning_rate": 0.0004888465396188566,
"loss": 3.4438,
"theoretical_loss": 4.734561195361609,
"tokens_seen": 105906176
},
{
"epoch": 0.03,
"learning_rate": 0.0004888365095285858,
"loss": 3.5359,
"theoretical_loss": 4.734158393502471,
"tokens_seen": 105971712
},
{
"epoch": 0.03,
"learning_rate": 0.0004888264794383149,
"loss": 3.6042,
"theoretical_loss": 4.733755910370867,
"tokens_seen": 106037248
},
{
"epoch": 0.03,
"learning_rate": 0.0004888164493480441,
"loss": 3.4694,
"theoretical_loss": 4.73335374551784,
"tokens_seen": 106102784
},
{
"epoch": 0.03,
"learning_rate": 0.0004888064192577733,
"loss": 3.3701,
"theoretical_loss": 4.732951898495341,
"tokens_seen": 106168320
},
{
"epoch": 0.03,
"learning_rate": 0.0004887963891675025,
"loss": 3.4071,
"theoretical_loss": 4.7325503688562325,
"tokens_seen": 106233856
},
{
"epoch": 0.03,
"learning_rate": 0.0004887863590772317,
"loss": 3.4026,
"theoretical_loss": 4.732149156154276,
"tokens_seen": 106299392
},
{
"epoch": 0.03,
"learning_rate": 0.000488776328986961,
"loss": 3.5821,
"theoretical_loss": 4.731748259944139,
"tokens_seen": 106364928
},
{
"epoch": 0.03,
"learning_rate": 0.00048876629889669,
"loss": 3.1001,
"theoretical_loss": 4.731347679781386,
"tokens_seen": 106430464
},
{
"epoch": 0.03,
"objective/train/docs_used": 70950,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.365251302719116,
"objective/train/theoretical_loss": 4.730947415222481,
"objective/train/tokens_used": 126956000,
"theoretical_loss": 4.730947415222481,
"tokens_seen": 106496000
},
{
"epoch": 0.03,
"learning_rate": 0.0004887562688064193,
"loss": 3.3721,
"theoretical_loss": 4.730947415222481,
"tokens_seen": 106496000
},
{
"epoch": 0.03,
"learning_rate": 0.0004887462387161484,
"loss": 3.3913,
"theoretical_loss": 4.730547465824781,
"tokens_seen": 106561536
},
{
"epoch": 0.03,
"learning_rate": 0.0004887362086258777,
"loss": 3.477,
"theoretical_loss": 4.730147831146537,
"tokens_seen": 106627072
},
{
"epoch": 0.03,
"learning_rate": 0.0004887261785356069,
"loss": 3.4452,
"theoretical_loss": 4.72974851074689,
"tokens_seen": 106692608
},
{
"epoch": 0.03,
"learning_rate": 0.000488716148445336,
"loss": 3.3808,
"theoretical_loss": 4.729349504185867,
"tokens_seen": 106758144
},
{
"epoch": 0.03,
"learning_rate": 0.0004887061183550652,
"loss": 3.6197,
"theoretical_loss": 4.728950811024383,
"tokens_seen": 106823680
},
{
"epoch": 0.03,
"learning_rate": 0.0004886960882647944,
"loss": 3.2728,
"theoretical_loss": 4.7285524308242355,
"tokens_seen": 106889216
},
{
"epoch": 0.03,
"learning_rate": 0.0004886860581745236,
"loss": 3.4828,
"theoretical_loss": 4.728154363148102,
"tokens_seen": 106954752
},
{
"epoch": 0.03,
"learning_rate": 0.0004886760280842528,
"loss": 3.645,
"theoretical_loss": 4.72775660755954,
"tokens_seen": 107020288
},
{
"epoch": 0.03,
"learning_rate": 0.0004886659979939819,
"loss": 3.5389,
"theoretical_loss": 4.72735916362298,
"tokens_seen": 107085824
},
{
"epoch": 0.03,
"learning_rate": 0.0004886559679037111,
"loss": 3.2972,
"theoretical_loss": 4.7269620309037315,
"tokens_seen": 107151360
},
{
"epoch": 0.03,
"learning_rate": 0.0004886459378134403,
"loss": 3.3874,
"theoretical_loss": 4.726565208967973,
"tokens_seen": 107216896
},
{
"epoch": 0.03,
"learning_rate": 0.0004886359077231695,
"loss": 3.375,
"theoretical_loss": 4.726168697382751,
"tokens_seen": 107282432
},
{
"epoch": 0.03,
"learning_rate": 0.0004886258776328988,
"loss": 3.6601,
"theoretical_loss": 4.725772495715983,
"tokens_seen": 107347968
},
{
"epoch": 0.03,
"learning_rate": 0.0004886158475426278,
"loss": 3.369,
"theoretical_loss": 4.725376603536446,
"tokens_seen": 107413504
},
{
"epoch": 0.03,
"learning_rate": 0.0004886058174523571,
"loss": 3.2596,
"theoretical_loss": 4.724981020413787,
"tokens_seen": 107479040
},
{
"epoch": 0.03,
"learning_rate": 0.0004885957873620863,
"loss": 3.3279,
"theoretical_loss": 4.724585745918505,
"tokens_seen": 107544576
},
{
"epoch": 0.03,
"learning_rate": 0.0004885857572718155,
"loss": 3.4956,
"theoretical_loss": 4.7241907796219635,
"tokens_seen": 107610112
},
{
"epoch": 0.03,
"learning_rate": 0.0004885757271815447,
"loss": 3.4726,
"theoretical_loss": 4.723796121096381,
"tokens_seen": 107675648
},
{
"epoch": 0.03,
"learning_rate": 0.0004885656970912739,
"loss": 3.5414,
"theoretical_loss": 4.723401769914824,
"tokens_seen": 107741184
},
{
"epoch": 0.03,
"learning_rate": 0.000488555667001003,
"loss": 3.2837,
"theoretical_loss": 4.723007725651219,
"tokens_seen": 107806720
},
{
"epoch": 0.03,
"learning_rate": 0.0004885456369107322,
"loss": 3.4393,
"theoretical_loss": 4.722613987880335,
"tokens_seen": 107872256
},
{
"epoch": 0.03,
"learning_rate": 0.0004885356068204614,
"loss": 3.4046,
"theoretical_loss": 4.722220556177792,
"tokens_seen": 107937792
},
{
"epoch": 0.03,
"learning_rate": 0.0004885255767301906,
"loss": 3.3551,
"theoretical_loss": 4.721827430120053,
"tokens_seen": 108003328
},
{
"epoch": 0.03,
"learning_rate": 0.0004885155466399198,
"loss": 3.4638,
"theoretical_loss": 4.721434609284424,
"tokens_seen": 108068864
},
{
"epoch": 0.03,
"objective/train/docs_used": 72030,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.6380910873413086,
"objective/train/theoretical_loss": 4.721042093249051,
"objective/train/tokens_used": 128594400,
"theoretical_loss": 4.721042093249051,
"tokens_seen": 108134400
},
{
"epoch": 0.03,
"learning_rate": 0.000488505516549649,
"loss": 3.2402,
"theoretical_loss": 4.721042093249051,
"tokens_seen": 108134400
},
{
"epoch": 0.03,
"learning_rate": 0.0004884954864593781,
"loss": 3.3689,
"theoretical_loss": 4.720649881592919,
"tokens_seen": 108199936
},
{
"epoch": 0.03,
"learning_rate": 0.0004884854563691073,
"loss": 3.5546,
"theoretical_loss": 4.7202579738958494,
"tokens_seen": 108265472
},
{
"epoch": 0.03,
"learning_rate": 0.0004884754262788365,
"loss": 3.2907,
"theoretical_loss": 4.7198663697384955,
"tokens_seen": 108331008
},
{
"epoch": 0.03,
"learning_rate": 0.0004884653961885657,
"loss": 3.763,
"theoretical_loss": 4.719475068702346,
"tokens_seen": 108396544
},
{
"epoch": 0.03,
"learning_rate": 0.0004884553660982949,
"loss": 3.3741,
"theoretical_loss": 4.719084070369714,
"tokens_seen": 108462080
},
{
"epoch": 0.03,
"learning_rate": 0.0004884453360080241,
"loss": 3.5399,
"theoretical_loss": 4.718693374323747,
"tokens_seen": 108527616
},
{
"epoch": 0.03,
"learning_rate": 0.0004884353059177532,
"loss": 3.5136,
"theoretical_loss": 4.718302980148412,
"tokens_seen": 108593152
},
{
"epoch": 0.03,
"learning_rate": 0.0004884252758274825,
"loss": 3.3874,
"theoretical_loss": 4.717912887428501,
"tokens_seen": 108658688
},
{
"epoch": 0.03,
"learning_rate": 0.0004884152457372117,
"loss": 3.2129,
"theoretical_loss": 4.717523095749626,
"tokens_seen": 108724224
},
{
"epoch": 0.03,
"learning_rate": 0.0004884052156469409,
"loss": 3.1945,
"theoretical_loss": 4.717133604698222,
"tokens_seen": 108789760
},
{
"epoch": 0.03,
"learning_rate": 0.00048839518555667,
"loss": 3.4207,
"theoretical_loss": 4.7167444138615355,
"tokens_seen": 108855296
},
{
"epoch": 0.03,
"learning_rate": 0.0004883851554663992,
"loss": 3.6135,
"theoretical_loss": 4.716355522827633,
"tokens_seen": 108920832
},
{
"epoch": 0.03,
"learning_rate": 0.0004883751253761284,
"loss": 3.7415,
"theoretical_loss": 4.715966931185388,
"tokens_seen": 108986368
},
{
"epoch": 0.03,
"learning_rate": 0.0004883650952858576,
"loss": 3.6946,
"theoretical_loss": 4.715578638524491,
"tokens_seen": 109051904
},
{
"epoch": 0.03,
"learning_rate": 0.0004883550651955868,
"loss": 3.4421,
"theoretical_loss": 4.715190644435435,
"tokens_seen": 109117440
},
{
"epoch": 0.03,
"learning_rate": 0.000488345035105316,
"loss": 3.6584,
"theoretical_loss": 4.714802948509522,
"tokens_seen": 109182976
},
{
"epoch": 0.03,
"learning_rate": 0.0004883350050150451,
"loss": 3.3107,
"theoretical_loss": 4.71441555033886,
"tokens_seen": 109248512
},
{
"epoch": 0.03,
"learning_rate": 0.0004883249749247743,
"loss": 3.5758,
"theoretical_loss": 4.714028449516356,
"tokens_seen": 109314048
},
{
"epoch": 0.03,
"learning_rate": 0.0004883149448345035,
"loss": 3.3739,
"theoretical_loss": 4.713641645635718,
"tokens_seen": 109379584
},
{
"epoch": 0.03,
"learning_rate": 0.0004883049147442327,
"loss": 3.1408,
"theoretical_loss": 4.713255138291454,
"tokens_seen": 109445120
},
{
"epoch": 0.03,
"learning_rate": 0.0004882948846539619,
"loss": 3.6864,
"theoretical_loss": 4.712868927078868,
"tokens_seen": 109510656
},
{
"epoch": 0.03,
"learning_rate": 0.0004882848545636911,
"loss": 3.2615,
"theoretical_loss": 4.712483011594056,
"tokens_seen": 109576192
},
{
"epoch": 0.03,
"learning_rate": 0.0004882748244734203,
"loss": 3.5947,
"theoretical_loss": 4.7120973914339075,
"tokens_seen": 109641728
},
{
"epoch": 0.03,
"learning_rate": 0.00048826479438314946,
"loss": 3.3371,
"theoretical_loss": 4.7117120661961005,
"tokens_seen": 109707264
},
{
"epoch": 0.03,
"objective/train/docs_used": 72698,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.991194009780884,
"objective/train/theoretical_loss": 4.711327035479103,
"objective/train/tokens_used": 130232800,
"theoretical_loss": 4.711327035479103,
"tokens_seen": 109772800
},
{
"epoch": 0.03,
"learning_rate": 0.00048825476429287864,
"loss": 3.2689,
"theoretical_loss": 4.711327035479103,
"tokens_seen": 109772800
},
{
"epoch": 0.03,
"learning_rate": 0.0004882447342026078,
"loss": 3.4525,
"theoretical_loss": 4.710942298882169,
"tokens_seen": 109838336
},
{
"epoch": 0.03,
"learning_rate": 0.000488234704112337,
"loss": 3.3659,
"theoretical_loss": 4.710557856005335,
"tokens_seen": 109903872
},
{
"epoch": 0.03,
"learning_rate": 0.00048822467402206624,
"loss": 3.4411,
"theoretical_loss": 4.710173706449419,
"tokens_seen": 109969408
},
{
"epoch": 0.03,
"learning_rate": 0.00048821464393179536,
"loss": 3.382,
"theoretical_loss": 4.709789849816021,
"tokens_seen": 110034944
},
{
"epoch": 0.03,
"learning_rate": 0.0004882046138415246,
"loss": 3.3798,
"theoretical_loss": 4.7094062857075185,
"tokens_seen": 110100480
},
{
"epoch": 0.03,
"learning_rate": 0.0004881945837512537,
"loss": 3.5293,
"theoretical_loss": 4.709023013727063,
"tokens_seen": 110166016
},
{
"epoch": 0.03,
"learning_rate": 0.00048818455366098296,
"loss": 3.6122,
"theoretical_loss": 4.708640033478584,
"tokens_seen": 110231552
},
{
"epoch": 0.03,
"learning_rate": 0.00048817452357071214,
"loss": 3.3274,
"theoretical_loss": 4.708257344566778,
"tokens_seen": 110297088
},
{
"epoch": 0.03,
"learning_rate": 0.0004881644934804413,
"loss": 3.554,
"theoretical_loss": 4.7078749465971175,
"tokens_seen": 110362624
},
{
"epoch": 0.03,
"learning_rate": 0.0004881544633901705,
"loss": 3.3524,
"theoretical_loss": 4.707492839175837,
"tokens_seen": 110428160
},
{
"epoch": 0.03,
"learning_rate": 0.00048814443329989974,
"loss": 3.5396,
"theoretical_loss": 4.707111021909941,
"tokens_seen": 110493696
},
{
"epoch": 0.03,
"learning_rate": 0.00048813440320962887,
"loss": 3.5515,
"theoretical_loss": 4.706729494407197,
"tokens_seen": 110559232
},
{
"epoch": 0.03,
"learning_rate": 0.0004881243731193581,
"loss": 3.2664,
"theoretical_loss": 4.706348256276138,
"tokens_seen": 110624768
},
{
"epoch": 0.03,
"learning_rate": 0.00048811434302908723,
"loss": 3.4359,
"theoretical_loss": 4.705967307126051,
"tokens_seen": 110690304
},
{
"epoch": 0.03,
"learning_rate": 0.00048810431293881646,
"loss": 3.6256,
"theoretical_loss": 4.705586646566987,
"tokens_seen": 110755840
},
{
"epoch": 0.03,
"learning_rate": 0.00048809428284854564,
"loss": 3.5248,
"theoretical_loss": 4.705206274209751,
"tokens_seen": 110821376
},
{
"epoch": 0.03,
"learning_rate": 0.0004880842527582748,
"loss": 3.5576,
"theoretical_loss": 4.704826189665905,
"tokens_seen": 110886912
},
{
"epoch": 0.03,
"learning_rate": 0.000488074222668004,
"loss": 3.3691,
"theoretical_loss": 4.704446392547759,
"tokens_seen": 110952448
},
{
"epoch": 0.03,
"learning_rate": 0.0004880641925777332,
"loss": 3.6296,
"theoretical_loss": 4.7040668824683785,
"tokens_seen": 111017984
},
{
"epoch": 0.03,
"learning_rate": 0.00048805416248746237,
"loss": 3.5711,
"theoretical_loss": 4.7036876590415755,
"tokens_seen": 111083520
},
{
"epoch": 0.03,
"learning_rate": 0.0004880441323971916,
"loss": 3.1637,
"theoretical_loss": 4.7033087218819105,
"tokens_seen": 111149056
},
{
"epoch": 0.03,
"learning_rate": 0.00048803410230692073,
"loss": 3.4891,
"theoretical_loss": 4.7029300706046895,
"tokens_seen": 111214592
},
{
"epoch": 0.03,
"learning_rate": 0.00048802407221664997,
"loss": 3.4164,
"theoretical_loss": 4.702551704825957,
"tokens_seen": 111280128
},
{
"epoch": 0.03,
"learning_rate": 0.00048801404212637915,
"loss": 3.2079,
"theoretical_loss": 4.702173624162507,
"tokens_seen": 111345664
},
{
"epoch": 0.03,
"objective/train/docs_used": 74080,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.588409900665283,
"objective/train/theoretical_loss": 4.701795828231866,
"objective/train/tokens_used": 131871200,
"theoretical_loss": 4.701795828231866,
"tokens_seen": 111411200
},
{
"epoch": 0.03,
"learning_rate": 0.00048800401203610833,
"loss": 3.4279,
"theoretical_loss": 4.701795828231866,
"tokens_seen": 111411200
},
{
"epoch": 0.03,
"learning_rate": 0.0004879939819458375,
"loss": 3.4107,
"theoretical_loss": 4.701418316652299,
"tokens_seen": 111476736
},
{
"epoch": 0.03,
"learning_rate": 0.0004879839518555667,
"loss": 3.4536,
"theoretical_loss": 4.701041089042813,
"tokens_seen": 111542272
},
{
"epoch": 0.03,
"learning_rate": 0.0004879739217652959,
"loss": 3.7348,
"theoretical_loss": 4.700664145023142,
"tokens_seen": 111607808
},
{
"epoch": 0.03,
"learning_rate": 0.0004879638916750251,
"loss": 3.4685,
"theoretical_loss": 4.700287484213753,
"tokens_seen": 111673344
},
{
"epoch": 0.03,
"learning_rate": 0.0004879538615847543,
"loss": 3.4494,
"theoretical_loss": 4.699911106235849,
"tokens_seen": 111738880
},
{
"epoch": 0.03,
"learning_rate": 0.00048794383149448347,
"loss": 3.5418,
"theoretical_loss": 4.6995350107113545,
"tokens_seen": 111804416
},
{
"epoch": 0.03,
"learning_rate": 0.00048793380140421265,
"loss": 3.5618,
"theoretical_loss": 4.699159197262922,
"tokens_seen": 111869952
},
{
"epoch": 0.03,
"learning_rate": 0.00048792377131394183,
"loss": 3.3049,
"theoretical_loss": 4.698783665513934,
"tokens_seen": 111935488
},
{
"epoch": 0.03,
"learning_rate": 0.00048791374122367107,
"loss": 3.4054,
"theoretical_loss": 4.698408415088491,
"tokens_seen": 112001024
},
{
"epoch": 0.03,
"learning_rate": 0.0004879037111334002,
"loss": 3.5376,
"theoretical_loss": 4.698033445611415,
"tokens_seen": 112066560
},
{
"epoch": 0.03,
"learning_rate": 0.00048789368104312943,
"loss": 3.5404,
"theoretical_loss": 4.6976587567082495,
"tokens_seen": 112132096
},
{
"epoch": 0.03,
"learning_rate": 0.00048788365095285856,
"loss": 3.6666,
"theoretical_loss": 4.697284348005253,
"tokens_seen": 112197632
},
{
"epoch": 0.03,
"learning_rate": 0.0004878736208625878,
"loss": 3.6087,
"theoretical_loss": 4.696910219129402,
"tokens_seen": 112263168
},
{
"epoch": 0.03,
"learning_rate": 0.00048786359077231697,
"loss": 3.5387,
"theoretical_loss": 4.696536369708386,
"tokens_seen": 112328704
},
{
"epoch": 0.03,
"learning_rate": 0.00048785356068204615,
"loss": 3.3868,
"theoretical_loss": 4.696162799370606,
"tokens_seen": 112394240
},
{
"epoch": 0.03,
"learning_rate": 0.00048784353059177533,
"loss": 3.6141,
"theoretical_loss": 4.695789507745176,
"tokens_seen": 112459776
},
{
"epoch": 0.03,
"learning_rate": 0.00048783350050150457,
"loss": 3.5212,
"theoretical_loss": 4.695416494461917,
"tokens_seen": 112525312
},
{
"epoch": 0.03,
"learning_rate": 0.0004878234704112337,
"loss": 3.4791,
"theoretical_loss": 4.695043759151353,
"tokens_seen": 112590848
},
{
"epoch": 0.03,
"learning_rate": 0.00048781344032096293,
"loss": 3.4949,
"theoretical_loss": 4.694671301444722,
"tokens_seen": 112656384
},
{
"epoch": 0.03,
"learning_rate": 0.00048780341023069206,
"loss": 3.6483,
"theoretical_loss": 4.694299120973957,
"tokens_seen": 112721920
},
{
"epoch": 0.03,
"learning_rate": 0.0004877933801404213,
"loss": 3.5765,
"theoretical_loss": 4.693927217371698,
"tokens_seen": 112787456
},
{
"epoch": 0.03,
"learning_rate": 0.0004877833500501505,
"loss": 3.5005,
"theoretical_loss": 4.693555590271282,
"tokens_seen": 112852992
},
{
"epoch": 0.03,
"learning_rate": 0.00048777331995987966,
"loss": 3.3525,
"theoretical_loss": 4.693184239306744,
"tokens_seen": 112918528
},
{
"epoch": 0.03,
"learning_rate": 0.00048776328986960884,
"loss": 3.5055,
"theoretical_loss": 4.692813164112819,
"tokens_seen": 112984064
},
{
"epoch": 0.03,
"objective/train/docs_used": 74817,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.344128131866455,
"objective/train/theoretical_loss": 4.692442364324931,
"objective/train/tokens_used": 133509600,
"theoretical_loss": 4.692442364324931,
"tokens_seen": 113049600
},
{
"epoch": 0.03,
"learning_rate": 0.000487753259779338,
"loss": 3.3455,
"theoretical_loss": 4.692442364324931,
"tokens_seen": 113049600
},
{
"epoch": 0.03,
"learning_rate": 0.0004877432296890672,
"loss": 3.4303,
"theoretical_loss": 4.692071839579201,
"tokens_seen": 113115136
},
{
"epoch": 0.03,
"learning_rate": 0.00048773319959879644,
"loss": 3.4279,
"theoretical_loss": 4.6917015895124425,
"tokens_seen": 113180672
},
{
"epoch": 0.03,
"learning_rate": 0.00048772316950852556,
"loss": 3.7147,
"theoretical_loss": 4.691331613762153,
"tokens_seen": 113246208
},
{
"epoch": 0.03,
"learning_rate": 0.0004877131394182548,
"loss": 3.4796,
"theoretical_loss": 4.690961911966523,
"tokens_seen": 113311744
},
{
"epoch": 0.03,
"learning_rate": 0.0004877031093279839,
"loss": 3.4674,
"theoretical_loss": 4.690592483764427,
"tokens_seen": 113377280
},
{
"epoch": 0.03,
"learning_rate": 0.00048769307923771316,
"loss": 3.4433,
"theoretical_loss": 4.690223328795424,
"tokens_seen": 113442816
},
{
"epoch": 0.03,
"learning_rate": 0.00048768304914744234,
"loss": 3.2091,
"theoretical_loss": 4.689854446699757,
"tokens_seen": 113508352
},
{
"epoch": 0.03,
"learning_rate": 0.0004876730190571715,
"loss": 3.345,
"theoretical_loss": 4.689485837118347,
"tokens_seen": 113573888
},
{
"epoch": 0.03,
"learning_rate": 0.0004876629889669007,
"loss": 3.3691,
"theoretical_loss": 4.689117499692798,
"tokens_seen": 113639424
},
{
"epoch": 0.03,
"learning_rate": 0.00048765295887662994,
"loss": 3.5643,
"theoretical_loss": 4.688749434065389,
"tokens_seen": 113704960
},
{
"epoch": 0.03,
"learning_rate": 0.00048764292878635907,
"loss": 3.241,
"theoretical_loss": 4.688381639879076,
"tokens_seen": 113770496
},
{
"epoch": 0.03,
"learning_rate": 0.0004876328986960883,
"loss": 3.4272,
"theoretical_loss": 4.68801411677749,
"tokens_seen": 113836032
},
{
"epoch": 0.03,
"learning_rate": 0.00048762286860581743,
"loss": 3.3181,
"theoretical_loss": 4.687646864404934,
"tokens_seen": 113901568
},
{
"epoch": 0.03,
"learning_rate": 0.00048761283851554666,
"loss": 3.6051,
"theoretical_loss": 4.687279882406381,
"tokens_seen": 113967104
},
{
"epoch": 0.03,
"learning_rate": 0.00048760280842527584,
"loss": 3.1655,
"theoretical_loss": 4.686913170427477,
"tokens_seen": 114032640
},
{
"epoch": 0.03,
"learning_rate": 0.000487592778335005,
"loss": 3.3195,
"theoretical_loss": 4.68654672811453,
"tokens_seen": 114098176
},
{
"epoch": 0.03,
"learning_rate": 0.0004875827482447342,
"loss": 3.4204,
"theoretical_loss": 4.68618055511452,
"tokens_seen": 114163712
},
{
"epoch": 0.03,
"learning_rate": 0.0004875727181544634,
"loss": 3.3986,
"theoretical_loss": 4.685814651075088,
"tokens_seen": 114229248
},
{
"epoch": 0.03,
"learning_rate": 0.00048756268806419257,
"loss": 3.3853,
"theoretical_loss": 4.685449015644537,
"tokens_seen": 114294784
},
{
"epoch": 0.03,
"learning_rate": 0.0004875526579739218,
"loss": 3.1979,
"theoretical_loss": 4.685083648471835,
"tokens_seen": 114360320
},
{
"epoch": 0.03,
"learning_rate": 0.00048754262788365093,
"loss": 3.3925,
"theoretical_loss": 4.684718549206607,
"tokens_seen": 114425856
},
{
"epoch": 0.03,
"learning_rate": 0.00048753259779338017,
"loss": 3.4082,
"theoretical_loss": 4.6843537174991345,
"tokens_seen": 114491392
},
{
"epoch": 0.03,
"learning_rate": 0.00048752256770310935,
"loss": 3.3881,
"theoretical_loss": 4.6839891530003595,
"tokens_seen": 114556928
},
{
"epoch": 0.03,
"learning_rate": 0.00048751253761283853,
"loss": 3.4909,
"theoretical_loss": 4.683624855361876,
"tokens_seen": 114622464
},
{
"epoch": 0.03,
"objective/train/docs_used": 76169,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.169177770614624,
"objective/train/theoretical_loss": 4.68326082423593,
"objective/train/tokens_used": 135148000,
"theoretical_loss": 4.68326082423593,
"tokens_seen": 114688000
},
{
"epoch": 0.03,
"learning_rate": 0.0004875025075225677,
"loss": 3.1599,
"theoretical_loss": 4.68326082423593,
"tokens_seen": 114688000
},
{
"epoch": 0.03,
"learning_rate": 0.0004874924774322969,
"loss": 3.744,
"theoretical_loss": 4.682897059275422,
"tokens_seen": 114753536
},
{
"epoch": 0.03,
"learning_rate": 0.00048748244734202607,
"loss": 3.5137,
"theoretical_loss": 4.682533560133901,
"tokens_seen": 114819072
},
{
"epoch": 0.03,
"learning_rate": 0.0004874724172517553,
"loss": 3.3442,
"theoretical_loss": 4.682170326465565,
"tokens_seen": 114884608
},
{
"epoch": 0.03,
"learning_rate": 0.00048746238716148443,
"loss": 3.4037,
"theoretical_loss": 4.681807357925257,
"tokens_seen": 114950144
},
{
"epoch": 0.03,
"learning_rate": 0.00048745235707121367,
"loss": 3.6075,
"theoretical_loss": 4.681444654168468,
"tokens_seen": 115015680
},
{
"epoch": 0.03,
"learning_rate": 0.0004874423269809428,
"loss": 3.3067,
"theoretical_loss": 4.68108221485133,
"tokens_seen": 115081216
},
{
"epoch": 0.03,
"learning_rate": 0.00048743229689067203,
"loss": 3.5086,
"theoretical_loss": 4.680720039630617,
"tokens_seen": 115146752
},
{
"epoch": 0.03,
"learning_rate": 0.0004874222668004012,
"loss": 3.1917,
"theoretical_loss": 4.680358128163747,
"tokens_seen": 115212288
},
{
"epoch": 0.03,
"learning_rate": 0.0004874122367101304,
"loss": 3.6048,
"theoretical_loss": 4.679996480108773,
"tokens_seen": 115277824
},
{
"epoch": 0.03,
"learning_rate": 0.0004874022066198596,
"loss": 3.3722,
"theoretical_loss": 4.6796350951243895,
"tokens_seen": 115343360
},
{
"epoch": 0.03,
"learning_rate": 0.00048739217652958876,
"loss": 3.491,
"theoretical_loss": 4.679273972869922,
"tokens_seen": 115408896
},
{
"epoch": 0.03,
"learning_rate": 0.00048738214643931794,
"loss": 3.3156,
"theoretical_loss": 4.678913113005333,
"tokens_seen": 115474432
},
{
"epoch": 0.04,
"learning_rate": 0.00048737211634904717,
"loss": 3.3652,
"theoretical_loss": 4.6785525151912175,
"tokens_seen": 115539968
},
{
"epoch": 0.04,
"learning_rate": 0.0004873620862587763,
"loss": 3.3324,
"theoretical_loss": 4.678192179088802,
"tokens_seen": 115605504
},
{
"epoch": 0.04,
"learning_rate": 0.00048735205616850553,
"loss": 3.6088,
"theoretical_loss": 4.6778321043599425,
"tokens_seen": 115671040
},
{
"epoch": 0.04,
"learning_rate": 0.0004873420260782347,
"loss": 3.3234,
"theoretical_loss": 4.677472290667122,
"tokens_seen": 115736576
},
{
"epoch": 0.04,
"learning_rate": 0.0004873319959879639,
"loss": 3.3235,
"theoretical_loss": 4.677112737673453,
"tokens_seen": 115802112
},
{
"epoch": 0.04,
"learning_rate": 0.0004873219658976931,
"loss": 3.3194,
"theoretical_loss": 4.676753445042669,
"tokens_seen": 115867648
},
{
"epoch": 0.04,
"learning_rate": 0.00048731193580742226,
"loss": 3.0035,
"theoretical_loss": 4.676394412439132,
"tokens_seen": 115933184
},
{
"epoch": 0.04,
"learning_rate": 0.00048730190571715144,
"loss": 3.2693,
"theoretical_loss": 4.6760356395278215,
"tokens_seen": 115998720
},
{
"epoch": 0.04,
"learning_rate": 0.0004872918756268807,
"loss": 3.3124,
"theoretical_loss": 4.675677125974339,
"tokens_seen": 116064256
},
{
"epoch": 0.04,
"learning_rate": 0.0004872818455366098,
"loss": 3.4606,
"theoretical_loss": 4.675318871444908,
"tokens_seen": 116129792
},
{
"epoch": 0.04,
"learning_rate": 0.00048727181544633904,
"loss": 3.3459,
"theoretical_loss": 4.674960875606366,
"tokens_seen": 116195328
},
{
"epoch": 0.04,
"learning_rate": 0.00048726178535606816,
"loss": 3.463,
"theoretical_loss": 4.674603138126168,
"tokens_seen": 116260864
},
{
"epoch": 0.04,
"objective/train/docs_used": 76931,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.525822162628174,
"objective/train/theoretical_loss": 4.674245658672382,
"objective/train/tokens_used": 136786400,
"theoretical_loss": 4.674245658672382,
"tokens_seen": 116326400
},
{
"epoch": 0.04,
"learning_rate": 0.0004872517552657974,
"loss": 3.5205,
"theoretical_loss": 4.674245658672382,
"tokens_seen": 116326400
},
{
"epoch": 0.04,
"learning_rate": 0.0004872417251755266,
"loss": 3.2274,
"theoretical_loss": 4.673888436913694,
"tokens_seen": 116391936
},
{
"epoch": 0.04,
"learning_rate": 0.00048723169508525576,
"loss": 3.524,
"theoretical_loss": 4.673531472519397,
"tokens_seen": 116457472
},
{
"epoch": 0.04,
"learning_rate": 0.000487221664994985,
"loss": 3.5341,
"theoretical_loss": 4.673174765159393,
"tokens_seen": 116523008
},
{
"epoch": 0.04,
"learning_rate": 0.0004872116349047141,
"loss": 3.3575,
"theoretical_loss": 4.672818314504198,
"tokens_seen": 116588544
},
{
"epoch": 0.04,
"learning_rate": 0.00048720160481444336,
"loss": 3.2956,
"theoretical_loss": 4.6724621202249335,
"tokens_seen": 116654080
},
{
"epoch": 0.04,
"learning_rate": 0.00048719157472417254,
"loss": 3.4414,
"theoretical_loss": 4.672106181993324,
"tokens_seen": 116719616
},
{
"epoch": 0.04,
"learning_rate": 0.0004871815446339017,
"loss": 3.4029,
"theoretical_loss": 4.6717504994817,
"tokens_seen": 116785152
},
{
"epoch": 0.04,
"learning_rate": 0.0004871715145436309,
"loss": 3.3539,
"theoretical_loss": 4.671395072362996,
"tokens_seen": 116850688
},
{
"epoch": 0.04,
"learning_rate": 0.00048716148445336014,
"loss": 3.7625,
"theoretical_loss": 4.671039900310747,
"tokens_seen": 116916224
},
{
"epoch": 0.04,
"learning_rate": 0.00048715145436308927,
"loss": 3.2591,
"theoretical_loss": 4.670684982999088,
"tokens_seen": 116981760
},
{
"epoch": 0.04,
"learning_rate": 0.0004871414242728185,
"loss": 3.4151,
"theoretical_loss": 4.670330320102753,
"tokens_seen": 117047296
},
{
"epoch": 0.04,
"learning_rate": 0.00048713139418254763,
"loss": 3.4562,
"theoretical_loss": 4.669975911297072,
"tokens_seen": 117112832
},
{
"epoch": 0.04,
"learning_rate": 0.00048712136409227686,
"loss": 3.4337,
"theoretical_loss": 4.669621756257971,
"tokens_seen": 117178368
},
{
"epoch": 0.04,
"learning_rate": 0.00048711133400200604,
"loss": 3.3579,
"theoretical_loss": 4.669267854661973,
"tokens_seen": 117243904
},
{
"epoch": 0.04,
"learning_rate": 0.0004871013039117352,
"loss": 3.3642,
"theoretical_loss": 4.668914206186189,
"tokens_seen": 117309440
},
{
"epoch": 0.04,
"learning_rate": 0.0004870912738214644,
"loss": 3.4332,
"theoretical_loss": 4.6685608105083265,
"tokens_seen": 117374976
},
{
"epoch": 0.04,
"learning_rate": 0.0004870812437311936,
"loss": 3.5261,
"theoretical_loss": 4.66820766730668,
"tokens_seen": 117440512
},
{
"epoch": 0.04,
"learning_rate": 0.00048707121364092277,
"loss": 3.2805,
"theoretical_loss": 4.667854776260132,
"tokens_seen": 117506048
},
{
"epoch": 0.04,
"learning_rate": 0.000487061183550652,
"loss": 3.5186,
"theoretical_loss": 4.667502137048155,
"tokens_seen": 117571584
},
{
"epoch": 0.04,
"learning_rate": 0.00048705115346038113,
"loss": 3.3778,
"theoretical_loss": 4.667149749350805,
"tokens_seen": 117637120
},
{
"epoch": 0.04,
"learning_rate": 0.00048704112337011037,
"loss": 3.1395,
"theoretical_loss": 4.666797612848723,
"tokens_seen": 117702656
},
{
"epoch": 0.04,
"learning_rate": 0.00048703109327983955,
"loss": 3.3278,
"theoretical_loss": 4.666445727223134,
"tokens_seen": 117768192
},
{
"epoch": 0.04,
"learning_rate": 0.00048702106318956873,
"loss": 3.2112,
"theoretical_loss": 4.666094092155843,
"tokens_seen": 117833728
},
{
"epoch": 0.04,
"learning_rate": 0.0004870110330992979,
"loss": 3.6812,
"theoretical_loss": 4.665742707329238,
"tokens_seen": 117899264
},
{
"epoch": 0.04,
"objective/train/docs_used": 78354,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.753464460372925,
"objective/train/theoretical_loss": 4.665391572426282,
"objective/train/tokens_used": 138424800,
"theoretical_loss": 4.665391572426282,
"tokens_seen": 117964800
},
{
"epoch": 0.04,
"learning_rate": 0.0004870010030090271,
"loss": 3.4575,
"theoretical_loss": 4.665391572426282,
"tokens_seen": 117964800
},
{
"epoch": 0.04,
"learning_rate": 0.00048699097291875627,
"loss": 3.2806,
"theoretical_loss": 4.665040687130518,
"tokens_seen": 118030336
},
{
"epoch": 0.04,
"learning_rate": 0.0004869809428284855,
"loss": 3.2856,
"theoretical_loss": 4.664690051126065,
"tokens_seen": 118095872
},
{
"epoch": 0.04,
"learning_rate": 0.00048697091273821463,
"loss": 3.6047,
"theoretical_loss": 4.664339664097617,
"tokens_seen": 118161408
},
{
"epoch": 0.04,
"learning_rate": 0.00048696088264794387,
"loss": 3.3602,
"theoretical_loss": 4.66398952573044,
"tokens_seen": 118226944
},
{
"epoch": 0.04,
"learning_rate": 0.000486950852557673,
"loss": 3.6397,
"theoretical_loss": 4.663639635710373,
"tokens_seen": 118292480
},
{
"epoch": 0.04,
"learning_rate": 0.00048694082246740223,
"loss": 3.5041,
"theoretical_loss": 4.663289993723826,
"tokens_seen": 118358016
},
{
"epoch": 0.04,
"learning_rate": 0.0004869307923771314,
"loss": 3.6349,
"theoretical_loss": 4.662940599457777,
"tokens_seen": 118423552
},
{
"epoch": 0.04,
"learning_rate": 0.0004869207622868606,
"loss": 3.311,
"theoretical_loss": 4.662591452599774,
"tokens_seen": 118489088
},
{
"epoch": 0.04,
"learning_rate": 0.0004869107321965898,
"loss": 3.3616,
"theoretical_loss": 4.662242552837929,
"tokens_seen": 118554624
},
{
"epoch": 0.04,
"learning_rate": 0.00048690070210631896,
"loss": 3.5774,
"theoretical_loss": 4.661893899860923,
"tokens_seen": 118620160
},
{
"epoch": 0.04,
"learning_rate": 0.00048689067201604814,
"loss": 3.5329,
"theoretical_loss": 4.6615454933579965,
"tokens_seen": 118685696
},
{
"epoch": 0.04,
"learning_rate": 0.00048688064192577737,
"loss": 3.3496,
"theoretical_loss": 4.661197333018957,
"tokens_seen": 118751232
},
{
"epoch": 0.04,
"learning_rate": 0.0004868706118355065,
"loss": 3.1691,
"theoretical_loss": 4.66084941853417,
"tokens_seen": 118816768
},
{
"epoch": 0.04,
"learning_rate": 0.00048686058174523573,
"loss": 3.4726,
"theoretical_loss": 4.6605017495945615,
"tokens_seen": 118882304
},
{
"epoch": 0.04,
"learning_rate": 0.0004868505516549649,
"loss": 3.2301,
"theoretical_loss": 4.660154325891618,
"tokens_seen": 118947840
},
{
"epoch": 0.04,
"learning_rate": 0.0004868405215646941,
"loss": 3.3908,
"theoretical_loss": 4.659807147117382,
"tokens_seen": 119013376
},
{
"epoch": 0.04,
"learning_rate": 0.0004868304914744233,
"loss": 3.5186,
"theoretical_loss": 4.6594602129644525,
"tokens_seen": 119078912
},
{
"epoch": 0.04,
"learning_rate": 0.00048682046138415246,
"loss": 3.5771,
"theoretical_loss": 4.659113523125981,
"tokens_seen": 119144448
},
{
"epoch": 0.04,
"learning_rate": 0.00048681043129388164,
"loss": 3.4647,
"theoretical_loss": 4.6587670772956775,
"tokens_seen": 119209984
},
{
"epoch": 0.04,
"learning_rate": 0.0004868004012036109,
"loss": 3.359,
"theoretical_loss": 4.658420875167799,
"tokens_seen": 119275520
},
{
"epoch": 0.04,
"learning_rate": 0.00048679037111334,
"loss": 3.2057,
"theoretical_loss": 4.658074916437155,
"tokens_seen": 119341056
},
{
"epoch": 0.04,
"learning_rate": 0.00048678034102306924,
"loss": 3.5888,
"theoretical_loss": 4.657729200799105,
"tokens_seen": 119406592
},
{
"epoch": 0.04,
"learning_rate": 0.00048677031093279836,
"loss": 3.3121,
"theoretical_loss": 4.657383727949558,
"tokens_seen": 119472128
},
{
"epoch": 0.04,
"learning_rate": 0.0004867602808425276,
"loss": 3.4127,
"theoretical_loss": 4.657038497584967,
"tokens_seen": 119537664
},
{
"epoch": 0.04,
"objective/train/docs_used": 78905,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3278000354766846,
"objective/train/theoretical_loss": 4.656693509402331,
"objective/train/tokens_used": 140063200,
"theoretical_loss": 4.656693509402331,
"tokens_seen": 119603200
},
{
"epoch": 0.04,
"learning_rate": 0.0004867502507522568,
"loss": 3.5357,
"theoretical_loss": 4.656693509402331,
"tokens_seen": 119603200
},
{
"epoch": 0.04,
"learning_rate": 0.00048674022066198596,
"loss": 3.3471,
"theoretical_loss": 4.6563487630991975,
"tokens_seen": 119668736
},
{
"epoch": 0.04,
"learning_rate": 0.00048673019057171514,
"loss": 3.5287,
"theoretical_loss": 4.656004258373651,
"tokens_seen": 119734272
},
{
"epoch": 0.04,
"learning_rate": 0.0004867201604814443,
"loss": 3.5142,
"theoretical_loss": 4.655659994924323,
"tokens_seen": 119799808
},
{
"epoch": 0.04,
"learning_rate": 0.0004867101303911735,
"loss": 3.2315,
"theoretical_loss": 4.655315972450383,
"tokens_seen": 119865344
},
{
"epoch": 0.04,
"learning_rate": 0.00048670010030090274,
"loss": 3.1649,
"theoretical_loss": 4.65497219065154,
"tokens_seen": 119930880
},
{
"epoch": 0.04,
"learning_rate": 0.00048669007021063187,
"loss": 3.5037,
"theoretical_loss": 4.654628649228041,
"tokens_seen": 119996416
},
{
"epoch": 0.04,
"learning_rate": 0.0004866800401203611,
"loss": 3.2458,
"theoretical_loss": 4.654285347880672,
"tokens_seen": 120061952
},
{
"epoch": 0.04,
"learning_rate": 0.0004866700100300903,
"loss": 3.407,
"theoretical_loss": 4.653942286310749,
"tokens_seen": 120127488
},
{
"epoch": 0.04,
"learning_rate": 0.00048665997993981947,
"loss": 3.4908,
"theoretical_loss": 4.653599464220129,
"tokens_seen": 120193024
},
{
"epoch": 0.04,
"learning_rate": 0.00048664994984954865,
"loss": 3.3596,
"theoretical_loss": 4.653256881311198,
"tokens_seen": 120258560
},
{
"epoch": 0.04,
"learning_rate": 0.00048663991975927783,
"loss": 3.489,
"theoretical_loss": 4.6529145372868745,
"tokens_seen": 120324096
},
{
"epoch": 0.04,
"learning_rate": 0.000486629889669007,
"loss": 3.3819,
"theoretical_loss": 4.652572431850608,
"tokens_seen": 120389632
},
{
"epoch": 0.04,
"learning_rate": 0.00048661985957873624,
"loss": 3.2704,
"theoretical_loss": 4.652230564706377,
"tokens_seen": 120455168
},
{
"epoch": 0.04,
"learning_rate": 0.00048660982948846537,
"loss": 3.2807,
"theoretical_loss": 4.651888935558688,
"tokens_seen": 120520704
},
{
"epoch": 0.04,
"learning_rate": 0.0004865997993981946,
"loss": 3.4143,
"theoretical_loss": 4.651547544112575,
"tokens_seen": 120586240
},
{
"epoch": 0.04,
"learning_rate": 0.00048658976930792373,
"loss": 3.3822,
"theoretical_loss": 4.651206390073597,
"tokens_seen": 120651776
},
{
"epoch": 0.04,
"learning_rate": 0.00048657973921765297,
"loss": 3.0172,
"theoretical_loss": 4.650865473147837,
"tokens_seen": 120717312
},
{
"epoch": 0.04,
"learning_rate": 0.00048656970912738215,
"loss": 3.2458,
"theoretical_loss": 4.650524793041903,
"tokens_seen": 120782848
},
{
"epoch": 0.04,
"learning_rate": 0.00048655967903711133,
"loss": 3.3165,
"theoretical_loss": 4.650184349462922,
"tokens_seen": 120848384
},
{
"epoch": 0.04,
"learning_rate": 0.0004865496489468405,
"loss": 3.5201,
"theoretical_loss": 4.649844142118544,
"tokens_seen": 120913920
},
{
"epoch": 0.04,
"learning_rate": 0.00048653961885656975,
"loss": 3.2466,
"theoretical_loss": 4.6495041707169396,
"tokens_seen": 120979456
},
{
"epoch": 0.04,
"learning_rate": 0.0004865295887662989,
"loss": 3.4625,
"theoretical_loss": 4.649164434966794,
"tokens_seen": 121044992
},
{
"epoch": 0.04,
"learning_rate": 0.0004865195586760281,
"loss": 3.4472,
"theoretical_loss": 4.648824934577313,
"tokens_seen": 121110528
},
{
"epoch": 0.04,
"learning_rate": 0.00048650952858575724,
"loss": 3.454,
"theoretical_loss": 4.648485669258216,
"tokens_seen": 121176064
},
{
"epoch": 0.04,
"objective/train/docs_used": 79692,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.7176690101623535,
"objective/train/theoretical_loss": 4.648146638719739,
"objective/train/tokens_used": 141701600,
"theoretical_loss": 4.648146638719739,
"tokens_seen": 121241600
},
{
"epoch": 0.04,
"learning_rate": 0.00048649949849548647,
"loss": 3.6455,
"theoretical_loss": 4.648146638719739,
"tokens_seen": 121241600
},
{
"epoch": 0.04,
"learning_rate": 0.00048648946840521565,
"loss": 3.2591,
"theoretical_loss": 4.647807842672631,
"tokens_seen": 121307136
},
{
"epoch": 0.04,
"learning_rate": 0.00048647943831494483,
"loss": 3.1074,
"theoretical_loss": 4.647469280828153,
"tokens_seen": 121372672
},
{
"epoch": 0.04,
"learning_rate": 0.00048646940822467407,
"loss": 3.625,
"theoretical_loss": 4.647130952898077,
"tokens_seen": 121438208
},
{
"epoch": 0.04,
"learning_rate": 0.0004864593781344032,
"loss": 3.6184,
"theoretical_loss": 4.646792858594686,
"tokens_seen": 121503744
},
{
"epoch": 0.04,
"learning_rate": 0.00048644934804413243,
"loss": 3.3147,
"theoretical_loss": 4.64645499763077,
"tokens_seen": 121569280
},
{
"epoch": 0.04,
"learning_rate": 0.0004864393179538616,
"loss": 3.3483,
"theoretical_loss": 4.646117369719629,
"tokens_seen": 121634816
},
{
"epoch": 0.04,
"learning_rate": 0.0004864292878635908,
"loss": 3.6571,
"theoretical_loss": 4.645779974575069,
"tokens_seen": 121700352
},
{
"epoch": 0.04,
"learning_rate": 0.00048641925777332,
"loss": 3.392,
"theoretical_loss": 4.6454428119113995,
"tokens_seen": 121765888
},
{
"epoch": 0.04,
"learning_rate": 0.00048640922768304916,
"loss": 3.4246,
"theoretical_loss": 4.6451058814434365,
"tokens_seen": 121831424
},
{
"epoch": 0.04,
"learning_rate": 0.00048639919759277834,
"loss": 3.2547,
"theoretical_loss": 4.644769182886495,
"tokens_seen": 121896960
},
{
"epoch": 0.04,
"learning_rate": 0.00048638916750250757,
"loss": 3.4761,
"theoretical_loss": 4.644432715956399,
"tokens_seen": 121962496
},
{
"epoch": 0.04,
"learning_rate": 0.0004863791374122367,
"loss": 3.3089,
"theoretical_loss": 4.644096480369466,
"tokens_seen": 122028032
},
{
"epoch": 0.04,
"learning_rate": 0.00048636910732196593,
"loss": 3.4689,
"theoretical_loss": 4.643760475842518,
"tokens_seen": 122093568
},
{
"epoch": 0.04,
"learning_rate": 0.0004863590772316951,
"loss": 3.3779,
"theoretical_loss": 4.6434247020928705,
"tokens_seen": 122159104
},
{
"epoch": 0.04,
"learning_rate": 0.0004863490471414243,
"loss": 3.3198,
"theoretical_loss": 4.643089158838341,
"tokens_seen": 122224640
},
{
"epoch": 0.04,
"learning_rate": 0.0004863390170511535,
"loss": 3.4073,
"theoretical_loss": 4.642753845797243,
"tokens_seen": 122290176
},
{
"epoch": 0.04,
"learning_rate": 0.00048632898696088266,
"loss": 3.3571,
"theoretical_loss": 4.642418762688379,
"tokens_seen": 122355712
},
{
"epoch": 0.04,
"learning_rate": 0.00048631895687061184,
"loss": 3.4689,
"theoretical_loss": 4.642083909231053,
"tokens_seen": 122421248
},
{
"epoch": 0.04,
"learning_rate": 0.0004863089267803411,
"loss": 3.1746,
"theoretical_loss": 4.641749285145057,
"tokens_seen": 122486784
},
{
"epoch": 0.04,
"learning_rate": 0.0004862988966900702,
"loss": 3.339,
"theoretical_loss": 4.641414890150675,
"tokens_seen": 122552320
},
{
"epoch": 0.04,
"learning_rate": 0.00048628886659979944,
"loss": 3.2195,
"theoretical_loss": 4.641080723968684,
"tokens_seen": 122617856
},
{
"epoch": 0.04,
"learning_rate": 0.00048627883650952857,
"loss": 3.3498,
"theoretical_loss": 4.6407467863203475,
"tokens_seen": 122683392
},
{
"epoch": 0.04,
"learning_rate": 0.0004862688064192578,
"loss": 3.6232,
"theoretical_loss": 4.640413076927418,
"tokens_seen": 122748928
},
{
"epoch": 0.04,
"learning_rate": 0.000486258776328987,
"loss": 3.3464,
"theoretical_loss": 4.6400795955121374,
"tokens_seen": 122814464
},
{
"epoch": 0.04,
"objective/train/docs_used": 81051,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.561691999435425,
"objective/train/theoretical_loss": 4.639746341797229,
"objective/train/tokens_used": 143340000,
"theoretical_loss": 4.639746341797229,
"tokens_seen": 122880000
},
{
"epoch": 0.04,
"learning_rate": 0.00048624874623871616,
"loss": 3.3018,
"theoretical_loss": 4.639746341797229,
"tokens_seen": 122880000
},
{
"epoch": 0.04,
"learning_rate": 0.00048623871614844534,
"loss": 3.3855,
"theoretical_loss": 4.639413315505905,
"tokens_seen": 122945536
},
{
"epoch": 0.04,
"learning_rate": 0.0004862286860581745,
"loss": 3.3287,
"theoretical_loss": 4.639080516361861,
"tokens_seen": 123011072
},
{
"epoch": 0.04,
"learning_rate": 0.0004862186559679037,
"loss": 3.3515,
"theoretical_loss": 4.638747944089273,
"tokens_seen": 123076608
},
{
"epoch": 0.04,
"learning_rate": 0.00048620862587763294,
"loss": 3.5609,
"theoretical_loss": 4.638415598412799,
"tokens_seen": 123142144
},
{
"epoch": 0.04,
"learning_rate": 0.00048619859578736207,
"loss": 3.1183,
"theoretical_loss": 4.638083479057579,
"tokens_seen": 123207680
},
{
"epoch": 0.04,
"learning_rate": 0.0004861885656970913,
"loss": 3.2753,
"theoretical_loss": 4.637751585749234,
"tokens_seen": 123273216
},
{
"epoch": 0.04,
"learning_rate": 0.0004861785356068205,
"loss": 3.3623,
"theoretical_loss": 4.6374199182138565,
"tokens_seen": 123338752
},
{
"epoch": 0.04,
"learning_rate": 0.00048616850551654967,
"loss": 3.2958,
"theoretical_loss": 4.637088476178025,
"tokens_seen": 123404288
},
{
"epoch": 0.04,
"learning_rate": 0.00048615847542627885,
"loss": 3.3467,
"theoretical_loss": 4.636757259368787,
"tokens_seen": 123469824
},
{
"epoch": 0.04,
"learning_rate": 0.00048614844533600803,
"loss": 3.6196,
"theoretical_loss": 4.636426267513668,
"tokens_seen": 123535360
},
{
"epoch": 0.04,
"learning_rate": 0.0004861384152457372,
"loss": 3.4397,
"theoretical_loss": 4.636095500340669,
"tokens_seen": 123600896
},
{
"epoch": 0.04,
"learning_rate": 0.00048612838515546644,
"loss": 3.3183,
"theoretical_loss": 4.635764957578261,
"tokens_seen": 123666432
},
{
"epoch": 0.04,
"learning_rate": 0.00048611835506519557,
"loss": 3.5385,
"theoretical_loss": 4.635434638955388,
"tokens_seen": 123731968
},
{
"epoch": 0.04,
"learning_rate": 0.0004861083249749248,
"loss": 3.4339,
"theoretical_loss": 4.635104544201465,
"tokens_seen": 123797504
},
{
"epoch": 0.04,
"learning_rate": 0.00048609829488465393,
"loss": 3.4177,
"theoretical_loss": 4.634774673046376,
"tokens_seen": 123863040
},
{
"epoch": 0.04,
"learning_rate": 0.00048608826479438317,
"loss": 3.5387,
"theoretical_loss": 4.634445025220475,
"tokens_seen": 123928576
},
{
"epoch": 0.04,
"learning_rate": 0.00048607823470411235,
"loss": 3.2727,
"theoretical_loss": 4.634115600454582,
"tokens_seen": 123994112
},
{
"epoch": 0.04,
"learning_rate": 0.00048606820461384153,
"loss": 3.5262,
"theoretical_loss": 4.633786398479983,
"tokens_seen": 124059648
},
{
"epoch": 0.04,
"learning_rate": 0.0004860581745235707,
"loss": 3.3889,
"theoretical_loss": 4.6334574190284314,
"tokens_seen": 124125184
},
{
"epoch": 0.04,
"learning_rate": 0.00048604814443329995,
"loss": 3.1077,
"theoretical_loss": 4.633128661832145,
"tokens_seen": 124190720
},
{
"epoch": 0.04,
"learning_rate": 0.0004860381143430291,
"loss": 3.5107,
"theoretical_loss": 4.632800126623803,
"tokens_seen": 124256256
},
{
"epoch": 0.04,
"learning_rate": 0.0004860280842527583,
"loss": 3.4446,
"theoretical_loss": 4.632471813136547,
"tokens_seen": 124321792
},
{
"epoch": 0.04,
"learning_rate": 0.00048601805416248744,
"loss": 3.3011,
"theoretical_loss": 4.632143721103983,
"tokens_seen": 124387328
},
{
"epoch": 0.04,
"learning_rate": 0.00048600802407221667,
"loss": 3.4728,
"theoretical_loss": 4.631815850260173,
"tokens_seen": 124452864
},
{
"epoch": 0.04,
"objective/train/docs_used": 81846,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.6328256130218506,
"objective/train/theoretical_loss": 4.631488200339643,
"objective/train/tokens_used": 144978400,
"theoretical_loss": 4.631488200339643,
"tokens_seen": 124518400
},
{
"epoch": 0.04,
"learning_rate": 0.00048599799398194585,
"loss": 3.3451,
"theoretical_loss": 4.631488200339643,
"tokens_seen": 124518400
},
{
"epoch": 0.04,
"learning_rate": 0.00048598796389167503,
"loss": 3.3085,
"theoretical_loss": 4.63116077107737,
"tokens_seen": 124583936
},
{
"epoch": 0.04,
"learning_rate": 0.0004859779338014042,
"loss": 3.3846,
"theoretical_loss": 4.630833562208797,
"tokens_seen": 124649472
},
{
"epoch": 0.04,
"learning_rate": 0.0004859679037111334,
"loss": 3.4496,
"theoretical_loss": 4.630506573469815,
"tokens_seen": 124715008
},
{
"epoch": 0.04,
"learning_rate": 0.0004859578736208626,
"loss": 3.0359,
"theoretical_loss": 4.630179804596775,
"tokens_seen": 124780544
},
{
"epoch": 0.04,
"learning_rate": 0.0004859478435305918,
"loss": 3.3915,
"theoretical_loss": 4.629853255326481,
"tokens_seen": 124846080
},
{
"epoch": 0.04,
"learning_rate": 0.00048593781344032094,
"loss": 3.1342,
"theoretical_loss": 4.629526925396189,
"tokens_seen": 124911616
},
{
"epoch": 0.04,
"learning_rate": 0.0004859277833500502,
"loss": 3.5865,
"theoretical_loss": 4.6292008145436085,
"tokens_seen": 124977152
},
{
"epoch": 0.04,
"learning_rate": 0.0004859177532597793,
"loss": 3.3983,
"theoretical_loss": 4.628874922506897,
"tokens_seen": 125042688
},
{
"epoch": 0.04,
"learning_rate": 0.00048590772316950854,
"loss": 3.3449,
"theoretical_loss": 4.628549249024666,
"tokens_seen": 125108224
},
{
"epoch": 0.04,
"learning_rate": 0.0004858976930792377,
"loss": 3.3839,
"theoretical_loss": 4.628223793835975,
"tokens_seen": 125173760
},
{
"epoch": 0.04,
"learning_rate": 0.0004858876629889669,
"loss": 3.1372,
"theoretical_loss": 4.627898556680327,
"tokens_seen": 125239296
},
{
"epoch": 0.04,
"learning_rate": 0.0004858776328986961,
"loss": 3.2933,
"theoretical_loss": 4.627573537297678,
"tokens_seen": 125304832
},
{
"epoch": 0.04,
"learning_rate": 0.0004858676028084253,
"loss": 3.506,
"theoretical_loss": 4.627248735428427,
"tokens_seen": 125370368
},
{
"epoch": 0.04,
"learning_rate": 0.00048585757271815444,
"loss": 3.4041,
"theoretical_loss": 4.6269241508134185,
"tokens_seen": 125435904
},
{
"epoch": 0.04,
"learning_rate": 0.0004858475426278837,
"loss": 3.5242,
"theoretical_loss": 4.6265997831939405,
"tokens_seen": 125501440
},
{
"epoch": 0.04,
"learning_rate": 0.0004858375125376128,
"loss": 3.3199,
"theoretical_loss": 4.6262756323117245,
"tokens_seen": 125566976
},
{
"epoch": 0.04,
"learning_rate": 0.00048582748244734204,
"loss": 3.1673,
"theoretical_loss": 4.625951697908944,
"tokens_seen": 125632512
},
{
"epoch": 0.04,
"learning_rate": 0.0004858174523570712,
"loss": 3.5243,
"theoretical_loss": 4.625627979728212,
"tokens_seen": 125698048
},
{
"epoch": 0.04,
"learning_rate": 0.0004858074222668004,
"loss": 3.404,
"theoretical_loss": 4.625304477512584,
"tokens_seen": 125763584
},
{
"epoch": 0.04,
"learning_rate": 0.0004857973921765296,
"loss": 3.2105,
"theoretical_loss": 4.624981191005554,
"tokens_seen": 125829120
},
{
"epoch": 0.04,
"learning_rate": 0.00048578736208625877,
"loss": 3.0283,
"theoretical_loss": 4.624658119951052,
"tokens_seen": 125894656
},
{
"epoch": 0.04,
"learning_rate": 0.00048577733199598795,
"loss": 3.1752,
"theoretical_loss": 4.624335264093447,
"tokens_seen": 125960192
},
{
"epoch": 0.04,
"learning_rate": 0.0004857673019057172,
"loss": 3.5489,
"theoretical_loss": 4.624012623177544,
"tokens_seen": 126025728
},
{
"epoch": 0.04,
"learning_rate": 0.0004857572718154463,
"loss": 3.3505,
"theoretical_loss": 4.623690196948582,
"tokens_seen": 126091264
},
{
"epoch": 0.04,
"objective/train/docs_used": 83055,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.7693979740142822,
"objective/train/theoretical_loss": 4.623367985152234,
"objective/train/tokens_used": 146616800,
"theoretical_loss": 4.623367985152234,
"tokens_seen": 126156800
},
{
"epoch": 0.04,
"learning_rate": 0.00048574724172517554,
"loss": 3.554,
"theoretical_loss": 4.623367985152234,
"tokens_seen": 126156800
},
{
"epoch": 0.04,
"learning_rate": 0.00048573721163490467,
"loss": 3.4829,
"theoretical_loss": 4.623045987534609,
"tokens_seen": 126222336
},
{
"epoch": 0.04,
"learning_rate": 0.0004857271815446339,
"loss": 3.5869,
"theoretical_loss": 4.622724203842246,
"tokens_seen": 126287872
},
{
"epoch": 0.04,
"learning_rate": 0.00048571715145436314,
"loss": 3.3389,
"theoretical_loss": 4.622402633822114,
"tokens_seen": 126353408
},
{
"epoch": 0.04,
"learning_rate": 0.00048570712136409227,
"loss": 3.3537,
"theoretical_loss": 4.622081277221616,
"tokens_seen": 126418944
},
{
"epoch": 0.04,
"learning_rate": 0.0004856970912738215,
"loss": 3.2713,
"theoretical_loss": 4.62176013378858,
"tokens_seen": 126484480
},
{
"epoch": 0.04,
"learning_rate": 0.0004856870611835507,
"loss": 3.4149,
"theoretical_loss": 4.621439203271267,
"tokens_seen": 126550016
},
{
"epoch": 0.04,
"learning_rate": 0.00048567703109327987,
"loss": 3.4411,
"theoretical_loss": 4.621118485418362,
"tokens_seen": 126615552
},
{
"epoch": 0.04,
"learning_rate": 0.00048566700100300905,
"loss": 3.5645,
"theoretical_loss": 4.620797979978978,
"tokens_seen": 126681088
},
{
"epoch": 0.04,
"learning_rate": 0.00048565697091273823,
"loss": 3.4029,
"theoretical_loss": 4.620477686702651,
"tokens_seen": 126746624
},
{
"epoch": 0.04,
"learning_rate": 0.0004856469408224674,
"loss": 3.5134,
"theoretical_loss": 4.620157605339347,
"tokens_seen": 126812160
},
{
"epoch": 0.04,
"learning_rate": 0.00048563691073219664,
"loss": 3.2403,
"theoretical_loss": 4.619837735639452,
"tokens_seen": 126877696
},
{
"epoch": 0.04,
"learning_rate": 0.00048562688064192577,
"loss": 3.2403,
"theoretical_loss": 4.619518077353776,
"tokens_seen": 126943232
},
{
"epoch": 0.04,
"learning_rate": 0.000485616850551655,
"loss": 3.3605,
"theoretical_loss": 4.619198630233547,
"tokens_seen": 127008768
},
{
"epoch": 0.04,
"learning_rate": 0.00048560682046138413,
"loss": 3.6782,
"theoretical_loss": 4.6188793940304205,
"tokens_seen": 127074304
},
{
"epoch": 0.04,
"learning_rate": 0.00048559679037111337,
"loss": 3.291,
"theoretical_loss": 4.618560368496466,
"tokens_seen": 127139840
},
{
"epoch": 0.04,
"learning_rate": 0.00048558676028084255,
"loss": 3.2539,
"theoretical_loss": 4.618241553384175,
"tokens_seen": 127205376
},
{
"epoch": 0.04,
"learning_rate": 0.00048557673019057173,
"loss": 3.3349,
"theoretical_loss": 4.617922948446459,
"tokens_seen": 127270912
},
{
"epoch": 0.04,
"learning_rate": 0.0004855667001003009,
"loss": 3.3863,
"theoretical_loss": 4.617604553436642,
"tokens_seen": 127336448
},
{
"epoch": 0.04,
"learning_rate": 0.00048555667001003015,
"loss": 3.3637,
"theoretical_loss": 4.617286368108466,
"tokens_seen": 127401984
},
{
"epoch": 0.04,
"learning_rate": 0.0004855466399197593,
"loss": 3.2942,
"theoretical_loss": 4.6169683922160925,
"tokens_seen": 127467520
},
{
"epoch": 0.04,
"learning_rate": 0.0004855366098294885,
"loss": 3.4269,
"theoretical_loss": 4.616650625514091,
"tokens_seen": 127533056
},
{
"epoch": 0.04,
"learning_rate": 0.00048552657973921764,
"loss": 3.3937,
"theoretical_loss": 4.616333067757449,
"tokens_seen": 127598592
},
{
"epoch": 0.04,
"learning_rate": 0.00048551654964894687,
"loss": 3.4983,
"theoretical_loss": 4.616015718701563,
"tokens_seen": 127664128
},
{
"epoch": 0.04,
"learning_rate": 0.00048550651955867605,
"loss": 3.3911,
"theoretical_loss": 4.615698578102245,
"tokens_seen": 127729664
},
{
"epoch": 0.04,
"objective/train/docs_used": 83592,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.880964994430542,
"objective/train/theoretical_loss": 4.615381645715717,
"objective/train/tokens_used": 148255200,
"theoretical_loss": 4.615381645715717,
"tokens_seen": 127795200
},
{
"epoch": 0.04,
"learning_rate": 0.00048549648946840523,
"loss": 3.5533,
"theoretical_loss": 4.615381645715717,
"tokens_seen": 127795200
},
{
"epoch": 0.04,
"learning_rate": 0.0004854864593781344,
"loss": 3.4536,
"theoretical_loss": 4.615064921298608,
"tokens_seen": 127860736
},
{
"epoch": 0.04,
"learning_rate": 0.0004854764292878636,
"loss": 3.377,
"theoretical_loss": 4.61474840460796,
"tokens_seen": 127926272
},
{
"epoch": 0.04,
"learning_rate": 0.0004854663991975928,
"loss": 3.5236,
"theoretical_loss": 4.614432095401219,
"tokens_seen": 127991808
},
{
"epoch": 0.04,
"learning_rate": 0.000485456369107322,
"loss": 3.4254,
"theoretical_loss": 4.614115993436242,
"tokens_seen": 128057344
},
{
"epoch": 0.04,
"learning_rate": 0.00048544633901705114,
"loss": 3.4557,
"theoretical_loss": 4.613800098471291,
"tokens_seen": 128122880
},
{
"epoch": 0.04,
"learning_rate": 0.0004854363089267804,
"loss": 3.5225,
"theoretical_loss": 4.613484410265032,
"tokens_seen": 128188416
},
{
"epoch": 0.04,
"learning_rate": 0.0004854262788365095,
"loss": 3.3142,
"theoretical_loss": 4.613168928576538,
"tokens_seen": 128253952
},
{
"epoch": 0.04,
"learning_rate": 0.00048541624874623874,
"loss": 3.3601,
"theoretical_loss": 4.612853653165283,
"tokens_seen": 128319488
},
{
"epoch": 0.04,
"learning_rate": 0.0004854062186559679,
"loss": 3.4458,
"theoretical_loss": 4.612538583791146,
"tokens_seen": 128385024
},
{
"epoch": 0.04,
"learning_rate": 0.0004853961885656971,
"loss": 3.1733,
"theoretical_loss": 4.612223720214407,
"tokens_seen": 128450560
},
{
"epoch": 0.04,
"learning_rate": 0.0004853861584754263,
"loss": 3.2368,
"theoretical_loss": 4.611909062195749,
"tokens_seen": 128516096
},
{
"epoch": 0.04,
"learning_rate": 0.0004853761283851555,
"loss": 3.3807,
"theoretical_loss": 4.61159460949625,
"tokens_seen": 128581632
},
{
"epoch": 0.04,
"learning_rate": 0.00048536609829488464,
"loss": 3.3186,
"theoretical_loss": 4.611280361877393,
"tokens_seen": 128647168
},
{
"epoch": 0.04,
"learning_rate": 0.0004853560682046139,
"loss": 3.4312,
"theoretical_loss": 4.610966319101056,
"tokens_seen": 128712704
},
{
"epoch": 0.04,
"learning_rate": 0.000485346038114343,
"loss": 3.3996,
"theoretical_loss": 4.610652480929515,
"tokens_seen": 128778240
},
{
"epoch": 0.04,
"learning_rate": 0.00048533600802407224,
"loss": 3.3225,
"theoretical_loss": 4.610338847125445,
"tokens_seen": 128843776
},
{
"epoch": 0.04,
"learning_rate": 0.0004853259779338014,
"loss": 3.5473,
"theoretical_loss": 4.610025417451913,
"tokens_seen": 128909312
},
{
"epoch": 0.04,
"learning_rate": 0.0004853159478435306,
"loss": 3.339,
"theoretical_loss": 4.6097121916723856,
"tokens_seen": 128974848
},
{
"epoch": 0.04,
"learning_rate": 0.0004853059177532598,
"loss": 3.5529,
"theoretical_loss": 4.609399169550718,
"tokens_seen": 129040384
},
{
"epoch": 0.04,
"learning_rate": 0.00048529588766298897,
"loss": 3.282,
"theoretical_loss": 4.609086350851165,
"tokens_seen": 129105920
},
{
"epoch": 0.04,
"learning_rate": 0.00048528585757271815,
"loss": 3.3268,
"theoretical_loss": 4.6087737353383655,
"tokens_seen": 129171456
},
{
"epoch": 0.04,
"learning_rate": 0.0004852758274824474,
"loss": 3.2374,
"theoretical_loss": 4.6084613227773605,
"tokens_seen": 129236992
},
{
"epoch": 0.04,
"learning_rate": 0.0004852657973921765,
"loss": 3.2591,
"theoretical_loss": 4.608149112933571,
"tokens_seen": 129302528
},
{
"epoch": 0.04,
"learning_rate": 0.00048525576730190574,
"loss": 3.0899,
"theoretical_loss": 4.607837105572816,
"tokens_seen": 129368064
},
{
"epoch": 0.04,
"objective/train/docs_used": 84814,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1867215633392334,
"objective/train/theoretical_loss": 4.607525300461299,
"objective/train/tokens_used": 149893600,
"theoretical_loss": 4.607525300461299,
"tokens_seen": 129433600
},
{
"epoch": 0.04,
"learning_rate": 0.00048524573721163487,
"loss": 3.1172,
"theoretical_loss": 4.607525300461299,
"tokens_seen": 129433600
},
{
"epoch": 0.04,
"learning_rate": 0.0004852357071213641,
"loss": 3.1282,
"theoretical_loss": 4.607213697365613,
"tokens_seen": 129499136
},
{
"epoch": 0.04,
"learning_rate": 0.0004852256770310933,
"loss": 3.3892,
"theoretical_loss": 4.606902296052739,
"tokens_seen": 129564672
},
{
"epoch": 0.04,
"learning_rate": 0.00048521564694082247,
"loss": 3.2205,
"theoretical_loss": 4.6065910962900425,
"tokens_seen": 129630208
},
{
"epoch": 0.04,
"learning_rate": 0.00048520561685055165,
"loss": 3.4858,
"theoretical_loss": 4.606280097845277,
"tokens_seen": 129695744
},
{
"epoch": 0.04,
"learning_rate": 0.0004851955867602809,
"loss": 3.2573,
"theoretical_loss": 4.60596930048658,
"tokens_seen": 129761280
},
{
"epoch": 0.04,
"learning_rate": 0.00048518555667001,
"loss": 3.3325,
"theoretical_loss": 4.605658703982471,
"tokens_seen": 129826816
},
{
"epoch": 0.04,
"learning_rate": 0.00048517552657973925,
"loss": 3.3328,
"theoretical_loss": 4.6053483081018545,
"tokens_seen": 129892352
},
{
"epoch": 0.04,
"learning_rate": 0.0004851654964894684,
"loss": 3.3861,
"theoretical_loss": 4.605038112614018,
"tokens_seen": 129957888
},
{
"epoch": 0.04,
"learning_rate": 0.0004851554663991976,
"loss": 3.1793,
"theoretical_loss": 4.604728117288631,
"tokens_seen": 130023424
},
{
"epoch": 0.04,
"learning_rate": 0.0004851454363089268,
"loss": 3.275,
"theoretical_loss": 4.604418321895739,
"tokens_seen": 130088960
},
{
"epoch": 0.04,
"learning_rate": 0.00048513540621865597,
"loss": 3.4315,
"theoretical_loss": 4.604108726205774,
"tokens_seen": 130154496
},
{
"epoch": 0.04,
"learning_rate": 0.00048512537612838515,
"loss": 3.3219,
"theoretical_loss": 4.60379932998954,
"tokens_seen": 130220032
},
{
"epoch": 0.04,
"learning_rate": 0.00048511534603811433,
"loss": 3.4426,
"theoretical_loss": 4.6034901330182265,
"tokens_seen": 130285568
},
{
"epoch": 0.04,
"learning_rate": 0.0004851053159478435,
"loss": 3.685,
"theoretical_loss": 4.603181135063394,
"tokens_seen": 130351104
},
{
"epoch": 0.04,
"learning_rate": 0.00048509528585757275,
"loss": 3.2137,
"theoretical_loss": 4.6028723358969845,
"tokens_seen": 130416640
},
{
"epoch": 0.04,
"learning_rate": 0.0004850852557673019,
"loss": 3.3579,
"theoretical_loss": 4.602563735291312,
"tokens_seen": 130482176
},
{
"epoch": 0.04,
"learning_rate": 0.0004850752256770311,
"loss": 3.3985,
"theoretical_loss": 4.602255333019068,
"tokens_seen": 130547712
},
{
"epoch": 0.04,
"learning_rate": 0.00048506519558676024,
"loss": 3.4027,
"theoretical_loss": 4.6019471288533165,
"tokens_seen": 130613248
},
{
"epoch": 0.04,
"learning_rate": 0.0004850551654964895,
"loss": 3.4611,
"theoretical_loss": 4.601639122567497,
"tokens_seen": 130678784
},
{
"epoch": 0.04,
"learning_rate": 0.00048504513540621866,
"loss": 3.2935,
"theoretical_loss": 4.601331313935418,
"tokens_seen": 130744320
},
{
"epoch": 0.04,
"learning_rate": 0.00048503510531594784,
"loss": 3.3577,
"theoretical_loss": 4.601023702731264,
"tokens_seen": 130809856
},
{
"epoch": 0.04,
"learning_rate": 0.000485025075225677,
"loss": 3.3172,
"theoretical_loss": 4.600716288729587,
"tokens_seen": 130875392
},
{
"epoch": 0.04,
"learning_rate": 0.00048501504513540625,
"loss": 3.2354,
"theoretical_loss": 4.600409071705312,
"tokens_seen": 130940928
},
{
"epoch": 0.04,
"learning_rate": 0.0004850050150451354,
"loss": 3.406,
"theoretical_loss": 4.60010205143373,
"tokens_seen": 131006464
},
{
"epoch": 0.04,
"objective/train/docs_used": 85560,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.442735195159912,
"objective/train/theoretical_loss": 4.599795227690505,
"objective/train/tokens_used": 151532000,
"theoretical_loss": 4.599795227690505,
"tokens_seen": 131072000
},
{
"epoch": 0.04,
"learning_rate": 0.0004849949849548646,
"loss": 3.4252,
"theoretical_loss": 4.599795227690505,
"tokens_seen": 131072000
},
{
"epoch": 0.04,
"learning_rate": 0.0004849849548645938,
"loss": 3.2521,
"theoretical_loss": 4.5994886002516635,
"tokens_seen": 131137536
},
{
"epoch": 0.04,
"learning_rate": 0.000484974924774323,
"loss": 3.2135,
"theoretical_loss": 4.599182168893604,
"tokens_seen": 131203072
},
{
"epoch": 0.04,
"learning_rate": 0.0004849648946840522,
"loss": 3.29,
"theoretical_loss": 4.598875933393089,
"tokens_seen": 131268608
},
{
"epoch": 0.04,
"learning_rate": 0.00048495486459378134,
"loss": 3.4736,
"theoretical_loss": 4.5985698935272445,
"tokens_seen": 131334144
},
{
"epoch": 0.04,
"learning_rate": 0.0004849448345035106,
"loss": 3.361,
"theoretical_loss": 4.598264049073565,
"tokens_seen": 131399680
},
{
"epoch": 0.04,
"learning_rate": 0.0004849348044132397,
"loss": 3.1653,
"theoretical_loss": 4.597958399809908,
"tokens_seen": 131465216
},
{
"epoch": 0.04,
"learning_rate": 0.00048492477432296894,
"loss": 3.0195,
"theoretical_loss": 4.59765294551449,
"tokens_seen": 131530752
},
{
"epoch": 0.04,
"learning_rate": 0.0004849147442326981,
"loss": 3.4434,
"theoretical_loss": 4.597347685965897,
"tokens_seen": 131596288
},
{
"epoch": 0.04,
"learning_rate": 0.0004849047141424273,
"loss": 3.579,
"theoretical_loss": 4.597042620943069,
"tokens_seen": 131661824
},
{
"epoch": 0.04,
"learning_rate": 0.0004848946840521565,
"loss": 3.1773,
"theoretical_loss": 4.596737750225311,
"tokens_seen": 131727360
},
{
"epoch": 0.04,
"learning_rate": 0.0004848846539618857,
"loss": 3.3462,
"theoretical_loss": 4.596433073592289,
"tokens_seen": 131792896
},
{
"epoch": 0.04,
"learning_rate": 0.00048487462387161484,
"loss": 3.169,
"theoretical_loss": 4.596128590824026,
"tokens_seen": 131858432
},
{
"epoch": 0.04,
"learning_rate": 0.0004848645937813441,
"loss": 3.3918,
"theoretical_loss": 4.595824301700904,
"tokens_seen": 131923968
},
{
"epoch": 0.04,
"learning_rate": 0.0004848545636910732,
"loss": 3.1064,
"theoretical_loss": 4.595520206003663,
"tokens_seen": 131989504
},
{
"epoch": 0.04,
"learning_rate": 0.00048484453360080244,
"loss": 3.4125,
"theoretical_loss": 4.595216303513399,
"tokens_seen": 132055040
},
{
"epoch": 0.04,
"learning_rate": 0.0004848345035105316,
"loss": 3.5448,
"theoretical_loss": 4.594912594011566,
"tokens_seen": 132120576
},
{
"epoch": 0.04,
"learning_rate": 0.0004848244734202608,
"loss": 3.1756,
"theoretical_loss": 4.594609077279973,
"tokens_seen": 132186112
},
{
"epoch": 0.04,
"learning_rate": 0.00048481444332999,
"loss": 3.3843,
"theoretical_loss": 4.594305753100782,
"tokens_seen": 132251648
},
{
"epoch": 0.04,
"learning_rate": 0.00048480441323971917,
"loss": 3.4335,
"theoretical_loss": 4.594002621256511,
"tokens_seen": 132317184
},
{
"epoch": 0.04,
"learning_rate": 0.00048479438314944835,
"loss": 3.4907,
"theoretical_loss": 4.59369968153003,
"tokens_seen": 132382720
},
{
"epoch": 0.04,
"learning_rate": 0.0004847843530591776,
"loss": 3.3829,
"theoretical_loss": 4.593396933704562,
"tokens_seen": 132448256
},
{
"epoch": 0.04,
"learning_rate": 0.0004847743229689067,
"loss": 3.4217,
"theoretical_loss": 4.593094377563681,
"tokens_seen": 132513792
},
{
"epoch": 0.04,
"learning_rate": 0.00048476429287863594,
"loss": 3.5836,
"theoretical_loss": 4.592792012891314,
"tokens_seen": 132579328
},
{
"epoch": 0.04,
"learning_rate": 0.00048475426278836507,
"loss": 3.1773,
"theoretical_loss": 4.592489839471735,
"tokens_seen": 132644864
},
{
"epoch": 0.04,
"objective/train/docs_used": 86181,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.529838800430298,
"objective/train/theoretical_loss": 4.592187857089571,
"objective/train/tokens_used": 153170400,
"theoretical_loss": 4.592187857089571,
"tokens_seen": 132710400
},
{
"epoch": 0.04,
"learning_rate": 0.0004847442326980943,
"loss": 3.3219,
"theoretical_loss": 4.592187857089571,
"tokens_seen": 132710400
},
{
"epoch": 0.04,
"learning_rate": 0.0004847342026078235,
"loss": 3.0821,
"theoretical_loss": 4.591886065529795,
"tokens_seen": 132775936
},
{
"epoch": 0.04,
"learning_rate": 0.00048472417251755267,
"loss": 3.2018,
"theoretical_loss": 4.591584464577728,
"tokens_seen": 132841472
},
{
"epoch": 0.04,
"learning_rate": 0.00048471414242728185,
"loss": 3.3234,
"theoretical_loss": 4.591283054019041,
"tokens_seen": 132907008
},
{
"epoch": 0.04,
"learning_rate": 0.0004847041123370111,
"loss": 3.3791,
"theoretical_loss": 4.5909818336397485,
"tokens_seen": 132972544
},
{
"epoch": 0.04,
"learning_rate": 0.0004846940822467402,
"loss": 3.2039,
"theoretical_loss": 4.590680803226213,
"tokens_seen": 133038080
},
{
"epoch": 0.04,
"learning_rate": 0.00048468405215646945,
"loss": 3.4776,
"theoretical_loss": 4.590379962565141,
"tokens_seen": 133103616
},
{
"epoch": 0.04,
"learning_rate": 0.0004846740220661986,
"loss": 3.337,
"theoretical_loss": 4.590079311443583,
"tokens_seen": 133169152
},
{
"epoch": 0.04,
"learning_rate": 0.0004846639919759278,
"loss": 3.431,
"theoretical_loss": 4.589778849648934,
"tokens_seen": 133234688
},
{
"epoch": 0.04,
"learning_rate": 0.000484653961885657,
"loss": 3.3345,
"theoretical_loss": 4.589478576968932,
"tokens_seen": 133300224
},
{
"epoch": 0.04,
"learning_rate": 0.00048464393179538617,
"loss": 3.4218,
"theoretical_loss": 4.589178493191655,
"tokens_seen": 133365760
},
{
"epoch": 0.04,
"learning_rate": 0.00048463390170511535,
"loss": 3.4226,
"theoretical_loss": 4.588878598105527,
"tokens_seen": 133431296
},
{
"epoch": 0.04,
"learning_rate": 0.00048462387161484453,
"loss": 3.3897,
"theoretical_loss": 4.588578891499308,
"tokens_seen": 133496832
},
{
"epoch": 0.04,
"learning_rate": 0.0004846138415245737,
"loss": 3.3441,
"theoretical_loss": 4.588279373162101,
"tokens_seen": 133562368
},
{
"epoch": 0.04,
"learning_rate": 0.00048460381143430295,
"loss": 3.1719,
"theoretical_loss": 4.587980042883347,
"tokens_seen": 133627904
},
{
"epoch": 0.04,
"learning_rate": 0.0004845937813440321,
"loss": 3.5295,
"theoretical_loss": 4.587680900452824,
"tokens_seen": 133693440
},
{
"epoch": 0.04,
"learning_rate": 0.0004845837512537613,
"loss": 3.3176,
"theoretical_loss": 4.587381945660653,
"tokens_seen": 133758976
},
{
"epoch": 0.04,
"learning_rate": 0.00048457372116349044,
"loss": 3.1641,
"theoretical_loss": 4.587083178297288,
"tokens_seen": 133824512
},
{
"epoch": 0.04,
"learning_rate": 0.0004845636910732197,
"loss": 3.1573,
"theoretical_loss": 4.5867845981535185,
"tokens_seen": 133890048
},
{
"epoch": 0.04,
"learning_rate": 0.00048455366098294886,
"loss": 3.3356,
"theoretical_loss": 4.586486205020474,
"tokens_seen": 133955584
},
{
"epoch": 0.04,
"learning_rate": 0.00048454363089267804,
"loss": 3.331,
"theoretical_loss": 4.586187998689616,
"tokens_seen": 134021120
},
{
"epoch": 0.04,
"learning_rate": 0.0004845336008024072,
"loss": 3.464,
"theoretical_loss": 4.585889978952741,
"tokens_seen": 134086656
},
{
"epoch": 0.04,
"learning_rate": 0.00048452357071213645,
"loss": 3.1701,
"theoretical_loss": 4.58559214560198,
"tokens_seen": 134152192
},
{
"epoch": 0.04,
"learning_rate": 0.0004845135406218656,
"loss": 3.6266,
"theoretical_loss": 4.585294498429796,
"tokens_seen": 134217728
},
{
"epoch": 0.04,
"learning_rate": 0.0004845035105315948,
"loss": 3.3988,
"theoretical_loss": 4.584997037228986,
"tokens_seen": 134283264
},
{
"epoch": 0.04,
"objective/train/docs_used": 87535,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1884214878082275,
"objective/train/theoretical_loss": 4.584699761792674,
"objective/train/tokens_used": 154808800,
"theoretical_loss": 4.584699761792674,
"tokens_seen": 134348800
},
{
"epoch": 0.04,
"learning_rate": 0.00048449348044132394,
"loss": 3.2752,
"theoretical_loss": 4.584699761792674,
"tokens_seen": 134348800
},
{
"epoch": 0.04,
"learning_rate": 0.0004844834503510532,
"loss": 3.3727,
"theoretical_loss": 4.5844026719143205,
"tokens_seen": 134414336
},
{
"epoch": 0.04,
"learning_rate": 0.00048447342026078236,
"loss": 3.4484,
"theoretical_loss": 4.5841057673877135,
"tokens_seen": 134479872
},
{
"epoch": 0.04,
"learning_rate": 0.00048446339017051154,
"loss": 3.429,
"theoretical_loss": 4.5838090480069695,
"tokens_seen": 134545408
},
{
"epoch": 0.04,
"learning_rate": 0.0004844533600802407,
"loss": 3.1828,
"theoretical_loss": 4.5835125135665375,
"tokens_seen": 134610944
},
{
"epoch": 0.04,
"learning_rate": 0.0004844433299899699,
"loss": 3.2808,
"theoretical_loss": 4.583216163861191,
"tokens_seen": 134676480
},
{
"epoch": 0.04,
"learning_rate": 0.0004844332998996991,
"loss": 3.4666,
"theoretical_loss": 4.58291999868603,
"tokens_seen": 134742016
},
{
"epoch": 0.04,
"learning_rate": 0.0004844232698094283,
"loss": 3.0757,
"theoretical_loss": 4.582624017836489,
"tokens_seen": 134807552
},
{
"epoch": 0.04,
"learning_rate": 0.00048441323971915745,
"loss": 3.5777,
"theoretical_loss": 4.582328221108318,
"tokens_seen": 134873088
},
{
"epoch": 0.04,
"learning_rate": 0.0004844032096288867,
"loss": 3.1727,
"theoretical_loss": 4.5820326082976,
"tokens_seen": 134938624
},
{
"epoch": 0.04,
"learning_rate": 0.00048439317953861586,
"loss": 3.6033,
"theoretical_loss": 4.581737179200739,
"tokens_seen": 135004160
},
{
"epoch": 0.04,
"learning_rate": 0.00048438314944834504,
"loss": 3.3186,
"theoretical_loss": 4.581441933614466,
"tokens_seen": 135069696
},
{
"epoch": 0.04,
"learning_rate": 0.0004843731193580742,
"loss": 3.3087,
"theoretical_loss": 4.581146871335832,
"tokens_seen": 135135232
},
{
"epoch": 0.04,
"learning_rate": 0.0004843630892678034,
"loss": 3.0668,
"theoretical_loss": 4.580851992162214,
"tokens_seen": 135200768
},
{
"epoch": 0.04,
"learning_rate": 0.0004843530591775326,
"loss": 3.2295,
"theoretical_loss": 4.5805572958913086,
"tokens_seen": 135266304
},
{
"epoch": 0.04,
"learning_rate": 0.0004843430290872618,
"loss": 3.5356,
"theoretical_loss": 4.580262782321135,
"tokens_seen": 135331840
},
{
"epoch": 0.04,
"learning_rate": 0.00048433299899699095,
"loss": 3.3037,
"theoretical_loss": 4.579968451250032,
"tokens_seen": 135397376
},
{
"epoch": 0.04,
"learning_rate": 0.0004843229689067202,
"loss": 3.2675,
"theoretical_loss": 4.579674302476661,
"tokens_seen": 135462912
},
{
"epoch": 0.04,
"learning_rate": 0.0004843129388164493,
"loss": 3.2687,
"theoretical_loss": 4.579380335800001,
"tokens_seen": 135528448
},
{
"epoch": 0.04,
"learning_rate": 0.00048430290872617855,
"loss": 3.5504,
"theoretical_loss": 4.579086551019348,
"tokens_seen": 135593984
},
{
"epoch": 0.04,
"learning_rate": 0.00048429287863590773,
"loss": 3.4594,
"theoretical_loss": 4.5787929479343195,
"tokens_seen": 135659520
},
{
"epoch": 0.04,
"learning_rate": 0.0004842828485456369,
"loss": 3.2982,
"theoretical_loss": 4.578499526344848,
"tokens_seen": 135725056
},
{
"epoch": 0.04,
"learning_rate": 0.0004842728184553661,
"loss": 3.1927,
"theoretical_loss": 4.578206286051184,
"tokens_seen": 135790592
},
{
"epoch": 0.04,
"learning_rate": 0.00048426278836509527,
"loss": 3.3044,
"theoretical_loss": 4.5779132268538945,
"tokens_seen": 135856128
},
{
"epoch": 0.04,
"learning_rate": 0.00048425275827482445,
"loss": 3.4211,
"theoretical_loss": 4.577620348553859,
"tokens_seen": 135921664
},
{
"epoch": 0.04,
"objective/train/docs_used": 88189,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8670549392700195,
"objective/train/theoretical_loss": 4.577327650952276,
"objective/train/tokens_used": 156447200,
"theoretical_loss": 4.577327650952276,
"tokens_seen": 135987200
},
{
"epoch": 0.04,
"learning_rate": 0.0004842427281845537,
"loss": 3.3964,
"theoretical_loss": 4.577327650952276,
"tokens_seen": 135987200
},
{
"epoch": 0.04,
"learning_rate": 0.00048423269809428287,
"loss": 3.0862,
"theoretical_loss": 4.5770351338506545,
"tokens_seen": 136052736
},
{
"epoch": 0.04,
"learning_rate": 0.00048422266800401205,
"loss": 3.0829,
"theoretical_loss": 4.57674279705082,
"tokens_seen": 136118272
},
{
"epoch": 0.04,
"learning_rate": 0.0004842126379137413,
"loss": 3.4358,
"theoretical_loss": 4.57645064035491,
"tokens_seen": 136183808
},
{
"epoch": 0.04,
"learning_rate": 0.0004842026078234704,
"loss": 3.4142,
"theoretical_loss": 4.576158663565371,
"tokens_seen": 136249344
},
{
"epoch": 0.04,
"learning_rate": 0.00048419257773319965,
"loss": 3.3851,
"theoretical_loss": 4.575866866484967,
"tokens_seen": 136314880
},
{
"epoch": 0.04,
"learning_rate": 0.0004841825476429288,
"loss": 3.4905,
"theoretical_loss": 4.575575248916767,
"tokens_seen": 136380416
},
{
"epoch": 0.04,
"learning_rate": 0.000484172517552658,
"loss": 3.2962,
"theoretical_loss": 4.575283810664155,
"tokens_seen": 136445952
},
{
"epoch": 0.04,
"learning_rate": 0.0004841624874623872,
"loss": 3.3953,
"theoretical_loss": 4.574992551530822,
"tokens_seen": 136511488
},
{
"epoch": 0.04,
"learning_rate": 0.00048415245737211637,
"loss": 3.4929,
"theoretical_loss": 4.574701471320768,
"tokens_seen": 136577024
},
{
"epoch": 0.04,
"learning_rate": 0.00048414242728184555,
"loss": 3.2127,
"theoretical_loss": 4.574410569838304,
"tokens_seen": 136642560
},
{
"epoch": 0.04,
"learning_rate": 0.00048413239719157473,
"loss": 3.3957,
"theoretical_loss": 4.574119846888045,
"tokens_seen": 136708096
},
{
"epoch": 0.04,
"learning_rate": 0.0004841223671013039,
"loss": 3.4099,
"theoretical_loss": 4.573829302274915,
"tokens_seen": 136773632
},
{
"epoch": 0.04,
"learning_rate": 0.00048411233701103315,
"loss": 3.4686,
"theoretical_loss": 4.573538935804146,
"tokens_seen": 136839168
},
{
"epoch": 0.04,
"learning_rate": 0.0004841023069207623,
"loss": 3.0463,
"theoretical_loss": 4.573248747281273,
"tokens_seen": 136904704
},
{
"epoch": 0.04,
"learning_rate": 0.0004840922768304915,
"loss": 3.7113,
"theoretical_loss": 4.5729587365121365,
"tokens_seen": 136970240
},
{
"epoch": 0.04,
"learning_rate": 0.00048408224674022064,
"loss": 3.3974,
"theoretical_loss": 4.572668903302886,
"tokens_seen": 137035776
},
{
"epoch": 0.04,
"learning_rate": 0.0004840722166499499,
"loss": 3.184,
"theoretical_loss": 4.572379247459969,
"tokens_seen": 137101312
},
{
"epoch": 0.04,
"learning_rate": 0.00048406218655967906,
"loss": 3.4432,
"theoretical_loss": 4.57208976879014,
"tokens_seen": 137166848
},
{
"epoch": 0.04,
"learning_rate": 0.00048405215646940824,
"loss": 3.5294,
"theoretical_loss": 4.571800467100456,
"tokens_seen": 137232384
},
{
"epoch": 0.04,
"learning_rate": 0.0004840421263791374,
"loss": 3.0596,
"theoretical_loss": 4.5715113421982725,
"tokens_seen": 137297920
},
{
"epoch": 0.04,
"learning_rate": 0.00048403209628886665,
"loss": 3.3131,
"theoretical_loss": 4.571222393891253,
"tokens_seen": 137363456
},
{
"epoch": 0.04,
"learning_rate": 0.0004840220661985958,
"loss": 3.51,
"theoretical_loss": 4.570933621987356,
"tokens_seen": 137428992
},
{
"epoch": 0.04,
"learning_rate": 0.000484012036108325,
"loss": 3.4489,
"theoretical_loss": 4.570645026294844,
"tokens_seen": 137494528
},
{
"epoch": 0.04,
"learning_rate": 0.00048400200601805414,
"loss": 3.3655,
"theoretical_loss": 4.570356606622278,
"tokens_seen": 137560064
},
{
"epoch": 0.04,
"objective/train/docs_used": 88907,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2466330528259277,
"objective/train/theoretical_loss": 4.570068362778516,
"objective/train/tokens_used": 158085600,
"theoretical_loss": 4.570068362778516,
"tokens_seen": 137625600
},
{
"epoch": 0.04,
"learning_rate": 0.0004839919759277834,
"loss": 3.4146,
"theoretical_loss": 4.570068362778516,
"tokens_seen": 137625600
},
{
"epoch": 0.04,
"learning_rate": 0.00048398194583751256,
"loss": 3.3105,
"theoretical_loss": 4.569780294572718,
"tokens_seen": 137691136
},
{
"epoch": 0.04,
"learning_rate": 0.00048397191574724174,
"loss": 3.5781,
"theoretical_loss": 4.569492401814339,
"tokens_seen": 137756672
},
{
"epoch": 0.04,
"learning_rate": 0.0004839618856569709,
"loss": 3.047,
"theoretical_loss": 4.569204684313133,
"tokens_seen": 137822208
},
{
"epoch": 0.04,
"learning_rate": 0.0004839518555667001,
"loss": 3.4353,
"theoretical_loss": 4.568917141879149,
"tokens_seen": 137887744
},
{
"epoch": 0.04,
"learning_rate": 0.0004839418254764293,
"loss": 3.1746,
"theoretical_loss": 4.568629774322736,
"tokens_seen": 137953280
},
{
"epoch": 0.04,
"learning_rate": 0.0004839317953861585,
"loss": 3.4604,
"theoretical_loss": 4.568342581454532,
"tokens_seen": 138018816
},
{
"epoch": 0.04,
"learning_rate": 0.00048392176529588765,
"loss": 3.5407,
"theoretical_loss": 4.568055563085476,
"tokens_seen": 138084352
},
{
"epoch": 0.04,
"learning_rate": 0.0004839117352056169,
"loss": 3.2899,
"theoretical_loss": 4.567768719026797,
"tokens_seen": 138149888
},
{
"epoch": 0.04,
"learning_rate": 0.00048390170511534606,
"loss": 3.3983,
"theoretical_loss": 4.567482049090019,
"tokens_seen": 138215424
},
{
"epoch": 0.04,
"learning_rate": 0.00048389167502507524,
"loss": 3.2304,
"theoretical_loss": 4.567195553086961,
"tokens_seen": 138280960
},
{
"epoch": 0.04,
"learning_rate": 0.0004838816449348044,
"loss": 3.3718,
"theoretical_loss": 4.566909230829729,
"tokens_seen": 138346496
},
{
"epoch": 0.04,
"learning_rate": 0.0004838716148445336,
"loss": 3.4189,
"theoretical_loss": 4.566623082130729,
"tokens_seen": 138412032
},
{
"epoch": 0.04,
"learning_rate": 0.0004838615847542628,
"loss": 3.197,
"theoretical_loss": 4.566337106802651,
"tokens_seen": 138477568
},
{
"epoch": 0.04,
"learning_rate": 0.000483851554663992,
"loss": 3.4958,
"theoretical_loss": 4.56605130465848,
"tokens_seen": 138543104
},
{
"epoch": 0.04,
"learning_rate": 0.00048384152457372115,
"loss": 3.245,
"theoretical_loss": 4.565765675511487,
"tokens_seen": 138608640
},
{
"epoch": 0.04,
"learning_rate": 0.0004838314944834504,
"loss": 3.3454,
"theoretical_loss": 4.565480219175237,
"tokens_seen": 138674176
},
{
"epoch": 0.04,
"learning_rate": 0.0004838214643931795,
"loss": 3.5074,
"theoretical_loss": 4.56519493546358,
"tokens_seen": 138739712
},
{
"epoch": 0.04,
"learning_rate": 0.00048381143430290875,
"loss": 3.5971,
"theoretical_loss": 4.56490982419066,
"tokens_seen": 138805248
},
{
"epoch": 0.04,
"learning_rate": 0.00048380140421263793,
"loss": 3.3556,
"theoretical_loss": 4.564624885170902,
"tokens_seen": 138870784
},
{
"epoch": 0.04,
"learning_rate": 0.0004837913741223671,
"loss": 3.4486,
"theoretical_loss": 4.564340118219022,
"tokens_seen": 138936320
},
{
"epoch": 0.04,
"learning_rate": 0.0004837813440320963,
"loss": 3.3375,
"theoretical_loss": 4.56405552315002,
"tokens_seen": 139001856
},
{
"epoch": 0.04,
"learning_rate": 0.00048377131394182547,
"loss": 3.5742,
"theoretical_loss": 4.563771099779187,
"tokens_seen": 139067392
},
{
"epoch": 0.04,
"learning_rate": 0.00048376128385155465,
"loss": 3.402,
"theoretical_loss": 4.563486847922093,
"tokens_seen": 139132928
},
{
"epoch": 0.04,
"learning_rate": 0.0004837512537612839,
"loss": 3.221,
"theoretical_loss": 4.563202767394597,
"tokens_seen": 139198464
},
{
"epoch": 0.04,
"objective/train/docs_used": 90119,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3730666637420654,
"objective/train/theoretical_loss": 4.562918858012843,
"objective/train/tokens_used": 159724000,
"theoretical_loss": 4.562918858012843,
"tokens_seen": 139264000
},
{
"epoch": 0.04,
"learning_rate": 0.000483741223671013,
"loss": 3.459,
"theoretical_loss": 4.562918858012843,
"tokens_seen": 139264000
},
{
"epoch": 0.04,
"learning_rate": 0.00048373119358074225,
"loss": 3.4475,
"theoretical_loss": 4.562635119593255,
"tokens_seen": 139329536
},
{
"epoch": 0.04,
"learning_rate": 0.00048372116349047143,
"loss": 3.523,
"theoretical_loss": 4.562351551952542,
"tokens_seen": 139395072
},
{
"epoch": 0.04,
"learning_rate": 0.0004837111334002006,
"loss": 3.434,
"theoretical_loss": 4.5620681549076965,
"tokens_seen": 139460608
},
{
"epoch": 0.04,
"learning_rate": 0.0004837011033099298,
"loss": 3.4291,
"theoretical_loss": 4.561784928275992,
"tokens_seen": 139526144
},
{
"epoch": 0.04,
"learning_rate": 0.000483691073219659,
"loss": 3.3811,
"theoretical_loss": 4.561501871874984,
"tokens_seen": 139591680
},
{
"epoch": 0.04,
"learning_rate": 0.00048368104312938816,
"loss": 3.3562,
"theoretical_loss": 4.561218985522507,
"tokens_seen": 139657216
},
{
"epoch": 0.04,
"learning_rate": 0.0004836710130391174,
"loss": 3.7033,
"theoretical_loss": 4.560936269036679,
"tokens_seen": 139722752
},
{
"epoch": 0.04,
"learning_rate": 0.0004836609829488465,
"loss": 3.4809,
"theoretical_loss": 4.560653722235895,
"tokens_seen": 139788288
},
{
"epoch": 0.04,
"learning_rate": 0.00048365095285857575,
"loss": 3.6892,
"theoretical_loss": 4.560371344938831,
"tokens_seen": 139853824
},
{
"epoch": 0.04,
"learning_rate": 0.0004836409227683049,
"loss": 3.3982,
"theoretical_loss": 4.560089136964439,
"tokens_seen": 139919360
},
{
"epoch": 0.04,
"learning_rate": 0.0004836308926780341,
"loss": 3.7439,
"theoretical_loss": 4.559807098131953,
"tokens_seen": 139984896
},
{
"epoch": 0.04,
"learning_rate": 0.0004836208625877633,
"loss": 3.2185,
"theoretical_loss": 4.559525228260882,
"tokens_seen": 140050432
},
{
"epoch": 0.04,
"learning_rate": 0.0004836108324974925,
"loss": 3.2183,
"theoretical_loss": 4.559243527171011,
"tokens_seen": 140115968
},
{
"epoch": 0.04,
"learning_rate": 0.00048360080240722166,
"loss": 3.4715,
"theoretical_loss": 4.558961994682403,
"tokens_seen": 140181504
},
{
"epoch": 0.04,
"learning_rate": 0.00048359077231695084,
"loss": 3.5439,
"theoretical_loss": 4.558680630615397,
"tokens_seen": 140247040
},
{
"epoch": 0.04,
"learning_rate": 0.00048358074222668,
"loss": 3.4811,
"theoretical_loss": 4.558399434790607,
"tokens_seen": 140312576
},
{
"epoch": 0.04,
"learning_rate": 0.00048357071213640926,
"loss": 3.4954,
"theoretical_loss": 4.558118407028921,
"tokens_seen": 140378112
},
{
"epoch": 0.04,
"learning_rate": 0.0004835606820461384,
"loss": 3.3851,
"theoretical_loss": 4.557837547151502,
"tokens_seen": 140443648
},
{
"epoch": 0.04,
"learning_rate": 0.0004835506519558676,
"loss": 3.3218,
"theoretical_loss": 4.557556854979786,
"tokens_seen": 140509184
},
{
"epoch": 0.04,
"learning_rate": 0.0004835406218655968,
"loss": 3.4642,
"theoretical_loss": 4.5572763303354815,
"tokens_seen": 140574720
},
{
"epoch": 0.04,
"learning_rate": 0.000483530591775326,
"loss": 3.237,
"theoretical_loss": 4.556995973040574,
"tokens_seen": 140640256
},
{
"epoch": 0.04,
"learning_rate": 0.00048352056168505516,
"loss": 3.4711,
"theoretical_loss": 4.556715782917314,
"tokens_seen": 140705792
},
{
"epoch": 0.04,
"learning_rate": 0.00048351053159478434,
"loss": 3.3196,
"theoretical_loss": 4.556435759788229,
"tokens_seen": 140771328
},
{
"epoch": 0.04,
"learning_rate": 0.0004835005015045135,
"loss": 3.2629,
"theoretical_loss": 4.556155903476114,
"tokens_seen": 140836864
},
{
"epoch": 0.04,
"objective/train/docs_used": 90792,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.9938900470733643,
"objective/train/theoretical_loss": 4.555876213804037,
"objective/train/tokens_used": 161362400,
"theoretical_loss": 4.555876213804037,
"tokens_seen": 140902400
},
{
"epoch": 0.04,
"learning_rate": 0.00048349047141424276,
"loss": 3.6836,
"theoretical_loss": 4.555876213804037,
"tokens_seen": 140902400
},
{
"epoch": 0.04,
"learning_rate": 0.00048348044132397194,
"loss": 3.4205,
"theoretical_loss": 4.555596690595333,
"tokens_seen": 140967936
},
{
"epoch": 0.04,
"learning_rate": 0.0004834704112337011,
"loss": 3.3216,
"theoretical_loss": 4.555317333673611,
"tokens_seen": 141033472
},
{
"epoch": 0.04,
"learning_rate": 0.0004834603811434303,
"loss": 3.247,
"theoretical_loss": 4.555038142862742,
"tokens_seen": 141099008
},
{
"epoch": 0.04,
"learning_rate": 0.0004834503510531595,
"loss": 3.3618,
"theoretical_loss": 4.5547591179868725,
"tokens_seen": 141164544
},
{
"epoch": 0.04,
"learning_rate": 0.0004834403209628887,
"loss": 3.3634,
"theoretical_loss": 4.554480258870409,
"tokens_seen": 141230080
},
{
"epoch": 0.04,
"learning_rate": 0.00048343029087261785,
"loss": 3.4449,
"theoretical_loss": 4.554201565338033,
"tokens_seen": 141295616
},
{
"epoch": 0.04,
"learning_rate": 0.0004834202607823471,
"loss": 3.212,
"theoretical_loss": 4.5539230372146875,
"tokens_seen": 141361152
},
{
"epoch": 0.04,
"learning_rate": 0.00048341023069207626,
"loss": 3.2166,
"theoretical_loss": 4.553644674325584,
"tokens_seen": 141426688
},
{
"epoch": 0.04,
"learning_rate": 0.00048340020060180544,
"loss": 3.4181,
"theoretical_loss": 4.553366476496198,
"tokens_seen": 141492224
},
{
"epoch": 0.04,
"learning_rate": 0.0004833901705115346,
"loss": 3.2078,
"theoretical_loss": 4.553088443552269,
"tokens_seen": 141557760
},
{
"epoch": 0.04,
"learning_rate": 0.0004833801404212638,
"loss": 3.3662,
"theoretical_loss": 4.552810575319806,
"tokens_seen": 141623296
},
{
"epoch": 0.04,
"learning_rate": 0.000483370110330993,
"loss": 3.2753,
"theoretical_loss": 4.552532871625077,
"tokens_seen": 141688832
},
{
"epoch": 0.04,
"learning_rate": 0.0004833600802407222,
"loss": 3.4694,
"theoretical_loss": 4.5522553322946155,
"tokens_seen": 141754368
},
{
"epoch": 0.04,
"learning_rate": 0.00048335005015045135,
"loss": 3.3584,
"theoretical_loss": 4.551977957155217,
"tokens_seen": 141819904
},
{
"epoch": 0.04,
"learning_rate": 0.0004833400200601806,
"loss": 3.3498,
"theoretical_loss": 4.5517007460339425,
"tokens_seen": 141885440
},
{
"epoch": 0.04,
"learning_rate": 0.0004833299899699097,
"loss": 3.4718,
"theoretical_loss": 4.551423698758111,
"tokens_seen": 141950976
},
{
"epoch": 0.04,
"learning_rate": 0.00048331995987963895,
"loss": 3.4992,
"theoretical_loss": 4.551146815155304,
"tokens_seen": 142016512
},
{
"epoch": 0.04,
"learning_rate": 0.00048330992978936813,
"loss": 3.1915,
"theoretical_loss": 4.550870095053366,
"tokens_seen": 142082048
},
{
"epoch": 0.04,
"learning_rate": 0.0004832998996990973,
"loss": 3.4141,
"theoretical_loss": 4.550593538280398,
"tokens_seen": 142147584
},
{
"epoch": 0.04,
"learning_rate": 0.0004832898696088265,
"loss": 3.4559,
"theoretical_loss": 4.550317144664766,
"tokens_seen": 142213120
},
{
"epoch": 0.04,
"learning_rate": 0.00048327983951855567,
"loss": 3.1613,
"theoretical_loss": 4.55004091403509,
"tokens_seen": 142278656
},
{
"epoch": 0.04,
"learning_rate": 0.00048326980942828485,
"loss": 3.4072,
"theoretical_loss": 4.5497648462202545,
"tokens_seen": 142344192
},
{
"epoch": 0.04,
"learning_rate": 0.0004832597793380141,
"loss": 3.4498,
"theoretical_loss": 4.549488941049397,
"tokens_seen": 142409728
},
{
"epoch": 0.04,
"learning_rate": 0.0004832497492477432,
"loss": 3.1324,
"theoretical_loss": 4.549213198351914,
"tokens_seen": 142475264
},
{
"epoch": 0.04,
"objective/train/docs_used": 92175,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.230867624282837,
"objective/train/theoretical_loss": 4.548937617957463,
"objective/train/tokens_used": 163000800,
"theoretical_loss": 4.548937617957463,
"tokens_seen": 142540800
},
{
"epoch": 0.04,
"learning_rate": 0.00048323971915747245,
"loss": 3.3699,
"theoretical_loss": 4.548937617957463,
"tokens_seen": 142540800
},
{
"epoch": 0.04,
"learning_rate": 0.00048322968906720163,
"loss": 3.3031,
"theoretical_loss": 4.548662199695954,
"tokens_seen": 142606336
},
{
"epoch": 0.04,
"learning_rate": 0.0004832196589769308,
"loss": 3.6595,
"theoretical_loss": 4.548386943397556,
"tokens_seen": 142671872
},
{
"epoch": 0.04,
"learning_rate": 0.00048320962888666,
"loss": 3.518,
"theoretical_loss": 4.548111848892693,
"tokens_seen": 142737408
},
{
"epoch": 0.04,
"learning_rate": 0.0004831995987963892,
"loss": 3.1302,
"theoretical_loss": 4.547836916012042,
"tokens_seen": 142802944
},
{
"epoch": 0.04,
"learning_rate": 0.00048318956870611836,
"loss": 3.4835,
"theoretical_loss": 4.547562144586539,
"tokens_seen": 142868480
},
{
"epoch": 0.04,
"learning_rate": 0.0004831795386158476,
"loss": 3.4292,
"theoretical_loss": 4.547287534447372,
"tokens_seen": 142934016
},
{
"epoch": 0.04,
"learning_rate": 0.0004831695085255767,
"loss": 3.3998,
"theoretical_loss": 4.5470130854259825,
"tokens_seen": 142999552
},
{
"epoch": 0.04,
"learning_rate": 0.00048315947843530595,
"loss": 3.5444,
"theoretical_loss": 4.546738797354065,
"tokens_seen": 143065088
},
{
"epoch": 0.04,
"learning_rate": 0.0004831494483450351,
"loss": 3.3672,
"theoretical_loss": 4.546464670063569,
"tokens_seen": 143130624
},
{
"epoch": 0.04,
"learning_rate": 0.0004831394182547643,
"loss": 3.3033,
"theoretical_loss": 4.546190703386695,
"tokens_seen": 143196160
},
{
"epoch": 0.04,
"learning_rate": 0.0004831293881644935,
"loss": 3.2802,
"theoretical_loss": 4.545916897155894,
"tokens_seen": 143261696
},
{
"epoch": 0.04,
"learning_rate": 0.0004831193580742227,
"loss": 3.2078,
"theoretical_loss": 4.54564325120387,
"tokens_seen": 143327232
},
{
"epoch": 0.04,
"learning_rate": 0.00048310932798395186,
"loss": 3.369,
"theoretical_loss": 4.545369765363578,
"tokens_seen": 143392768
},
{
"epoch": 0.04,
"learning_rate": 0.00048309929789368104,
"loss": 3.0522,
"theoretical_loss": 4.545096439468223,
"tokens_seen": 143458304
},
{
"epoch": 0.04,
"learning_rate": 0.0004830892678034102,
"loss": 3.457,
"theoretical_loss": 4.544823273351257,
"tokens_seen": 143523840
},
{
"epoch": 0.04,
"learning_rate": 0.00048307923771313946,
"loss": 3.2289,
"theoretical_loss": 4.544550266846388,
"tokens_seen": 143589376
},
{
"epoch": 0.04,
"learning_rate": 0.0004830692076228686,
"loss": 3.491,
"theoretical_loss": 4.544277419787566,
"tokens_seen": 143654912
},
{
"epoch": 0.04,
"learning_rate": 0.0004830591775325978,
"loss": 3.273,
"theoretical_loss": 4.544004732008993,
"tokens_seen": 143720448
},
{
"epoch": 0.04,
"learning_rate": 0.000483049147442327,
"loss": 3.4575,
"theoretical_loss": 4.543732203345119,
"tokens_seen": 143785984
},
{
"epoch": 0.04,
"learning_rate": 0.0004830391173520562,
"loss": 3.4134,
"theoretical_loss": 4.543459833630639,
"tokens_seen": 143851520
},
{
"epoch": 0.04,
"learning_rate": 0.00048302908726178536,
"loss": 3.1865,
"theoretical_loss": 4.543187622700497,
"tokens_seen": 143917056
},
{
"epoch": 0.04,
"learning_rate": 0.00048301905717151454,
"loss": 3.4901,
"theoretical_loss": 4.542915570389884,
"tokens_seen": 143982592
},
{
"epoch": 0.04,
"learning_rate": 0.0004830090270812437,
"loss": 3.3153,
"theoretical_loss": 4.542643676534234,
"tokens_seen": 144048128
},
{
"epoch": 0.04,
"learning_rate": 0.00048299899699097296,
"loss": 3.4697,
"theoretical_loss": 4.542371940969231,
"tokens_seen": 144113664
},
{
"epoch": 0.04,
"objective/train/docs_used": 92798,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.180187702178955,
"objective/train/theoretical_loss": 4.542100363530799,
"objective/train/tokens_used": 164639200,
"theoretical_loss": 4.542100363530799,
"tokens_seen": 144179200
},
{
"epoch": 0.04,
"learning_rate": 0.0004829889669007021,
"loss": 3.4208,
"theoretical_loss": 4.542100363530799,
"tokens_seen": 144179200
},
{
"epoch": 0.04,
"learning_rate": 0.0004829789368104313,
"loss": 3.4833,
"theoretical_loss": 4.54182894405511,
"tokens_seen": 144244736
},
{
"epoch": 0.04,
"learning_rate": 0.00048296890672016045,
"loss": 3.5622,
"theoretical_loss": 4.5415576823785795,
"tokens_seen": 144310272
},
{
"epoch": 0.04,
"learning_rate": 0.0004829588766298897,
"loss": 3.3664,
"theoretical_loss": 4.541286578337866,
"tokens_seen": 144375808
},
{
"epoch": 0.04,
"learning_rate": 0.00048294884653961887,
"loss": 3.3885,
"theoretical_loss": 4.541015631769872,
"tokens_seen": 144441344
},
{
"epoch": 0.04,
"learning_rate": 0.00048293881644934805,
"loss": 3.4071,
"theoretical_loss": 4.5407448425117405,
"tokens_seen": 144506880
},
{
"epoch": 0.04,
"learning_rate": 0.00048292878635907723,
"loss": 3.2401,
"theoretical_loss": 4.540474210400859,
"tokens_seen": 144572416
},
{
"epoch": 0.04,
"learning_rate": 0.00048291875626880646,
"loss": 3.3437,
"theoretical_loss": 4.540203735274855,
"tokens_seen": 144637952
},
{
"epoch": 0.04,
"learning_rate": 0.0004829087261785356,
"loss": 3.2116,
"theoretical_loss": 4.5399334169716,
"tokens_seen": 144703488
},
{
"epoch": 0.04,
"learning_rate": 0.0004828986960882648,
"loss": 3.6115,
"theoretical_loss": 4.539663255329202,
"tokens_seen": 144769024
},
{
"epoch": 0.04,
"learning_rate": 0.00048288866599799395,
"loss": 3.0955,
"theoretical_loss": 4.539393250186015,
"tokens_seen": 144834560
},
{
"epoch": 0.04,
"learning_rate": 0.0004828786359077232,
"loss": 3.4351,
"theoretical_loss": 4.539123401380625,
"tokens_seen": 144900096
},
{
"epoch": 0.04,
"learning_rate": 0.00048286860581745237,
"loss": 3.5045,
"theoretical_loss": 4.538853708751866,
"tokens_seen": 144965632
},
{
"epoch": 0.04,
"learning_rate": 0.00048285857572718155,
"loss": 3.3471,
"theoretical_loss": 4.538584172138804,
"tokens_seen": 145031168
},
{
"epoch": 0.04,
"learning_rate": 0.00048284854563691073,
"loss": 3.2684,
"theoretical_loss": 4.538314791380748,
"tokens_seen": 145096704
},
{
"epoch": 0.04,
"learning_rate": 0.0004828385155466399,
"loss": 3.2516,
"theoretical_loss": 4.538045566317242,
"tokens_seen": 145162240
},
{
"epoch": 0.04,
"learning_rate": 0.0004828284854563691,
"loss": 3.4594,
"theoretical_loss": 4.537776496788071,
"tokens_seen": 145227776
},
{
"epoch": 0.04,
"learning_rate": 0.00048281845536609833,
"loss": 3.603,
"theoretical_loss": 4.537507582633253,
"tokens_seen": 145293312
},
{
"epoch": 0.04,
"learning_rate": 0.00048280842527582746,
"loss": 3.2846,
"theoretical_loss": 4.537238823693045,
"tokens_seen": 145358848
},
{
"epoch": 0.04,
"learning_rate": 0.0004827983951855567,
"loss": 3.4105,
"theoretical_loss": 4.536970219807939,
"tokens_seen": 145424384
},
{
"epoch": 0.04,
"learning_rate": 0.0004827883650952858,
"loss": 3.3511,
"theoretical_loss": 4.536701770818665,
"tokens_seen": 145489920
},
{
"epoch": 0.04,
"learning_rate": 0.00048277833500501505,
"loss": 3.2664,
"theoretical_loss": 4.536433476566185,
"tokens_seen": 145555456
},
{
"epoch": 0.04,
"learning_rate": 0.00048276830491474423,
"loss": 3.2244,
"theoretical_loss": 4.536165336891699,
"tokens_seen": 145620992
},
{
"epoch": 0.04,
"learning_rate": 0.0004827582748244734,
"loss": 3.5109,
"theoretical_loss": 4.535897351636638,
"tokens_seen": 145686528
},
{
"epoch": 0.04,
"learning_rate": 0.0004827482447342026,
"loss": 3.509,
"theoretical_loss": 4.53562952064267,
"tokens_seen": 145752064
},
{
"epoch": 0.04,
"objective/train/docs_used": 94036,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4322946071624756,
"objective/train/theoretical_loss": 4.535361843751696,
"objective/train/tokens_used": 166277600,
"theoretical_loss": 4.535361843751696,
"tokens_seen": 145817600
},
{
"epoch": 0.04,
"learning_rate": 0.00048273821464393183,
"loss": 3.2137,
"theoretical_loss": 4.535361843751696,
"tokens_seen": 145817600
},
{
"epoch": 0.04,
"learning_rate": 0.000482728184553661,
"loss": 3.0617,
"theoretical_loss": 4.535094320805847,
"tokens_seen": 145883136
},
{
"epoch": 0.04,
"learning_rate": 0.0004827181544633902,
"loss": 3.4118,
"theoretical_loss": 4.534826951647489,
"tokens_seen": 145948672
},
{
"epoch": 0.04,
"learning_rate": 0.0004827081243731194,
"loss": 3.3963,
"theoretical_loss": 4.5345597361192205,
"tokens_seen": 146014208
},
{
"epoch": 0.04,
"learning_rate": 0.00048269809428284856,
"loss": 3.4053,
"theoretical_loss": 4.53429267406387,
"tokens_seen": 146079744
},
{
"epoch": 0.04,
"learning_rate": 0.0004826880641925778,
"loss": 3.1817,
"theoretical_loss": 4.5340257653244995,
"tokens_seen": 146145280
},
{
"epoch": 0.04,
"learning_rate": 0.0004826780341023069,
"loss": 3.6942,
"theoretical_loss": 4.5337590097444,
"tokens_seen": 146210816
},
{
"epoch": 0.04,
"learning_rate": 0.00048266800401203615,
"loss": 3.4861,
"theoretical_loss": 4.533492407167093,
"tokens_seen": 146276352
},
{
"epoch": 0.04,
"learning_rate": 0.0004826579739217653,
"loss": 3.1617,
"theoretical_loss": 4.53322595743633,
"tokens_seen": 146341888
},
{
"epoch": 0.04,
"learning_rate": 0.0004826479438314945,
"loss": 3.1373,
"theoretical_loss": 4.5329596603960916,
"tokens_seen": 146407424
},
{
"epoch": 0.04,
"learning_rate": 0.0004826379137412237,
"loss": 3.1347,
"theoretical_loss": 4.53269351589059,
"tokens_seen": 146472960
},
{
"epoch": 0.04,
"learning_rate": 0.0004826278836509529,
"loss": 3.3526,
"theoretical_loss": 4.532427523764261,
"tokens_seen": 146538496
},
{
"epoch": 0.04,
"learning_rate": 0.00048261785356068206,
"loss": 3.2586,
"theoretical_loss": 4.532161683861773,
"tokens_seen": 146604032
},
{
"epoch": 0.04,
"learning_rate": 0.00048260782347041124,
"loss": 3.3128,
"theoretical_loss": 4.5318959960280205,
"tokens_seen": 146669568
},
{
"epoch": 0.04,
"learning_rate": 0.0004825977933801404,
"loss": 3.39,
"theoretical_loss": 4.531630460108125,
"tokens_seen": 146735104
},
{
"epoch": 0.04,
"learning_rate": 0.00048258776328986966,
"loss": 3.4165,
"theoretical_loss": 4.531365075947434,
"tokens_seen": 146800640
},
{
"epoch": 0.04,
"learning_rate": 0.0004825777331995988,
"loss": 3.2288,
"theoretical_loss": 4.531099843391524,
"tokens_seen": 146866176
},
{
"epoch": 0.04,
"learning_rate": 0.000482567703109328,
"loss": 3.3832,
"theoretical_loss": 4.5308347622861955,
"tokens_seen": 146931712
},
{
"epoch": 0.04,
"learning_rate": 0.0004825576730190572,
"loss": 3.6141,
"theoretical_loss": 4.5305698324774735,
"tokens_seen": 146997248
},
{
"epoch": 0.04,
"learning_rate": 0.0004825476429287864,
"loss": 3.4217,
"theoretical_loss": 4.530305053811611,
"tokens_seen": 147062784
},
{
"epoch": 0.04,
"learning_rate": 0.00048253761283851556,
"loss": 3.2163,
"theoretical_loss": 4.530040426135084,
"tokens_seen": 147128320
},
{
"epoch": 0.04,
"learning_rate": 0.00048252758274824474,
"loss": 3.3166,
"theoretical_loss": 4.529775949294593,
"tokens_seen": 147193856
},
{
"epoch": 0.04,
"learning_rate": 0.0004825175526579739,
"loss": 3.3181,
"theoretical_loss": 4.529511623137061,
"tokens_seen": 147259392
},
{
"epoch": 0.04,
"learning_rate": 0.00048250752256770316,
"loss": 3.2954,
"theoretical_loss": 4.529247447509637,
"tokens_seen": 147324928
},
{
"epoch": 0.04,
"learning_rate": 0.0004824974924774323,
"loss": 3.2769,
"theoretical_loss": 4.528983422259691,
"tokens_seen": 147390464
},
{
"epoch": 0.04,
"objective/train/docs_used": 94726,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4257872104644775,
"objective/train/theoretical_loss": 4.528719547234816,
"objective/train/tokens_used": 167916000,
"theoretical_loss": 4.528719547234816,
"tokens_seen": 147456000
},
{
"epoch": 0.04,
"learning_rate": 0.0004824874623871615,
"loss": 3.3119,
"theoretical_loss": 4.528719547234816,
"tokens_seen": 147456000
},
{
"epoch": 0.04,
"learning_rate": 0.00048247743229689065,
"loss": 3.3536,
"theoretical_loss": 4.528455822282828,
"tokens_seen": 147521536
},
{
"epoch": 0.04,
"learning_rate": 0.0004824674022066199,
"loss": 3.4758,
"theoretical_loss": 4.528192247251763,
"tokens_seen": 147587072
},
{
"epoch": 0.04,
"learning_rate": 0.00048245737211634907,
"loss": 3.1317,
"theoretical_loss": 4.52792882198988,
"tokens_seen": 147652608
},
{
"epoch": 0.04,
"learning_rate": 0.00048244734202607825,
"loss": 3.3368,
"theoretical_loss": 4.527665546345656,
"tokens_seen": 147718144
},
{
"epoch": 0.04,
"learning_rate": 0.00048243731193580743,
"loss": 3.6421,
"theoretical_loss": 4.5274024201677925,
"tokens_seen": 147783680
},
{
"epoch": 0.04,
"learning_rate": 0.00048242728184553666,
"loss": 3.0839,
"theoretical_loss": 4.527139443305209,
"tokens_seen": 147849216
},
{
"epoch": 0.04,
"learning_rate": 0.0004824172517552658,
"loss": 3.4025,
"theoretical_loss": 4.526876615607042,
"tokens_seen": 147914752
},
{
"epoch": 0.04,
"learning_rate": 0.000482407221664995,
"loss": 3.1217,
"theoretical_loss": 4.526613936922654,
"tokens_seen": 147980288
},
{
"epoch": 0.04,
"learning_rate": 0.00048239719157472415,
"loss": 3.3364,
"theoretical_loss": 4.526351407101618,
"tokens_seen": 148045824
},
{
"epoch": 0.04,
"learning_rate": 0.0004823871614844534,
"loss": 3.0166,
"theoretical_loss": 4.526089025993732,
"tokens_seen": 148111360
},
{
"epoch": 0.04,
"learning_rate": 0.00048237713139418257,
"loss": 3.263,
"theoretical_loss": 4.525826793449008,
"tokens_seen": 148176896
},
{
"epoch": 0.04,
"learning_rate": 0.00048236710130391175,
"loss": 3.4591,
"theoretical_loss": 4.525564709317678,
"tokens_seen": 148242432
},
{
"epoch": 0.04,
"learning_rate": 0.00048235707121364093,
"loss": 3.4337,
"theoretical_loss": 4.525302773450187,
"tokens_seen": 148307968
},
{
"epoch": 0.04,
"learning_rate": 0.0004823470411233701,
"loss": 3.3084,
"theoretical_loss": 4.525040985697203,
"tokens_seen": 148373504
},
{
"epoch": 0.04,
"learning_rate": 0.0004823370110330993,
"loss": 3.5317,
"theoretical_loss": 4.524779345909604,
"tokens_seen": 148439040
},
{
"epoch": 0.05,
"learning_rate": 0.00048232698094282853,
"loss": 3.4339,
"theoretical_loss": 4.524517853938489,
"tokens_seen": 148504576
},
{
"epoch": 0.05,
"learning_rate": 0.00048231695085255766,
"loss": 3.1791,
"theoretical_loss": 4.524256509635169,
"tokens_seen": 148570112
},
{
"epoch": 0.05,
"learning_rate": 0.0004823069207622869,
"loss": 3.1341,
"theoretical_loss": 4.523995312851174,
"tokens_seen": 148635648
},
{
"epoch": 0.05,
"learning_rate": 0.000482296890672016,
"loss": 3.2603,
"theoretical_loss": 4.523734263438241,
"tokens_seen": 148701184
},
{
"epoch": 0.05,
"learning_rate": 0.00048228686058174525,
"loss": 3.2534,
"theoretical_loss": 4.52347336124833,
"tokens_seen": 148766720
},
{
"epoch": 0.05,
"learning_rate": 0.00048227683049147443,
"loss": 3.4183,
"theoretical_loss": 4.52321260613361,
"tokens_seen": 148832256
},
{
"epoch": 0.05,
"learning_rate": 0.0004822668004012036,
"loss": 3.3214,
"theoretical_loss": 4.522951997946466,
"tokens_seen": 148897792
},
{
"epoch": 0.05,
"learning_rate": 0.0004822567703109328,
"loss": 3.3674,
"theoretical_loss": 4.522691536539492,
"tokens_seen": 148963328
},
{
"epoch": 0.05,
"learning_rate": 0.00048224674022066203,
"loss": 3.2784,
"theoretical_loss": 4.522431221765498,
"tokens_seen": 149028864
},
{
"epoch": 0.05,
"objective/train/docs_used": 95869,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1724908351898193,
"objective/train/theoretical_loss": 4.522171053477507,
"objective/train/tokens_used": 169554400,
"theoretical_loss": 4.522171053477507,
"tokens_seen": 149094400
},
{
"epoch": 0.05,
"learning_rate": 0.00048223671013039116,
"loss": 3.2823,
"theoretical_loss": 4.522171053477507,
"tokens_seen": 149094400
},
{
"epoch": 0.05,
"learning_rate": 0.0004822266800401204,
"loss": 3.2289,
"theoretical_loss": 4.5219110315287505,
"tokens_seen": 149159936
},
{
"epoch": 0.05,
"learning_rate": 0.0004822166499498495,
"loss": 3.3112,
"theoretical_loss": 4.521651155772675,
"tokens_seen": 149225472
},
{
"epoch": 0.05,
"learning_rate": 0.00048220661985957876,
"loss": 3.5058,
"theoretical_loss": 4.521391426062934,
"tokens_seen": 149291008
},
{
"epoch": 0.05,
"learning_rate": 0.00048219658976930794,
"loss": 3.5517,
"theoretical_loss": 4.521131842253396,
"tokens_seen": 149356544
},
{
"epoch": 0.05,
"learning_rate": 0.0004821865596790371,
"loss": 3.4992,
"theoretical_loss": 4.520872404198139,
"tokens_seen": 149422080
},
{
"epoch": 0.05,
"learning_rate": 0.0004821765295887663,
"loss": 3.4315,
"theoretical_loss": 4.520613111751445,
"tokens_seen": 149487616
},
{
"epoch": 0.05,
"learning_rate": 0.0004821664994984955,
"loss": 3.3585,
"theoretical_loss": 4.520353964767814,
"tokens_seen": 149553152
},
{
"epoch": 0.05,
"learning_rate": 0.00048215646940822466,
"loss": 3.3818,
"theoretical_loss": 4.5200949631019505,
"tokens_seen": 149618688
},
{
"epoch": 0.05,
"learning_rate": 0.0004821464393179539,
"loss": 3.183,
"theoretical_loss": 4.519836106608768,
"tokens_seen": 149684224
},
{
"epoch": 0.05,
"learning_rate": 0.000482136409227683,
"loss": 3.2419,
"theoretical_loss": 4.519577395143388,
"tokens_seen": 149749760
},
{
"epoch": 0.05,
"learning_rate": 0.00048212637913741226,
"loss": 3.139,
"theoretical_loss": 4.519318828561142,
"tokens_seen": 149815296
},
{
"epoch": 0.05,
"learning_rate": 0.0004821163490471414,
"loss": 3.1937,
"theoretical_loss": 4.519060406717565,
"tokens_seen": 149880832
},
{
"epoch": 0.05,
"learning_rate": 0.0004821063189568706,
"loss": 3.1968,
"theoretical_loss": 4.518802129468405,
"tokens_seen": 149946368
},
{
"epoch": 0.05,
"learning_rate": 0.0004820962888665998,
"loss": 3.0867,
"theoretical_loss": 4.51854399666961,
"tokens_seen": 150011904
},
{
"epoch": 0.05,
"learning_rate": 0.000482086258776329,
"loss": 3.1493,
"theoretical_loss": 4.518286008177341,
"tokens_seen": 150077440
},
{
"epoch": 0.05,
"learning_rate": 0.00048207622868605816,
"loss": 3.4273,
"theoretical_loss": 4.51802816384796,
"tokens_seen": 150142976
},
{
"epoch": 0.05,
"learning_rate": 0.0004820661985957874,
"loss": 3.3718,
"theoretical_loss": 4.517770463538038,
"tokens_seen": 150208512
},
{
"epoch": 0.05,
"learning_rate": 0.00048205616850551653,
"loss": 3.4009,
"theoretical_loss": 4.517512907104347,
"tokens_seen": 150274048
},
{
"epoch": 0.05,
"learning_rate": 0.00048204613841524576,
"loss": 3.2267,
"theoretical_loss": 4.517255494403868,
"tokens_seen": 150339584
},
{
"epoch": 0.05,
"learning_rate": 0.0004820361083249749,
"loss": 3.2091,
"theoretical_loss": 4.516998225293785,
"tokens_seen": 150405120
},
{
"epoch": 0.05,
"learning_rate": 0.0004820260782347041,
"loss": 3.4881,
"theoretical_loss": 4.516741099631485,
"tokens_seen": 150470656
},
{
"epoch": 0.05,
"learning_rate": 0.0004820160481444333,
"loss": 3.2292,
"theoretical_loss": 4.51648411727456,
"tokens_seen": 150536192
},
{
"epoch": 0.05,
"learning_rate": 0.0004820060180541625,
"loss": 3.291,
"theoretical_loss": 4.5162272780808035,
"tokens_seen": 150601728
},
{
"epoch": 0.05,
"learning_rate": 0.00048199598796389167,
"loss": 3.1278,
"theoretical_loss": 4.515970581908216,
"tokens_seen": 150667264
},
{
"epoch": 0.05,
"objective/train/docs_used": 96442,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.100712299346924,
"objective/train/theoretical_loss": 4.515714028614996,
"objective/train/tokens_used": 171192800,
"theoretical_loss": 4.515714028614996,
"tokens_seen": 150732800
},
{
"epoch": 0.05,
"learning_rate": 0.00048198595787362085,
"loss": 3.182,
"theoretical_loss": 4.515714028614996,
"tokens_seen": 150732800
},
{
"epoch": 0.05,
"learning_rate": 0.0004819759277833501,
"loss": 3.4292,
"theoretical_loss": 4.515457618059546,
"tokens_seen": 150798336
},
{
"epoch": 0.05,
"learning_rate": 0.00048196589769307927,
"loss": 3.469,
"theoretical_loss": 4.515201350100471,
"tokens_seen": 150863872
},
{
"epoch": 0.05,
"learning_rate": 0.00048195586760280845,
"loss": 3.0502,
"theoretical_loss": 4.514945224596577,
"tokens_seen": 150929408
},
{
"epoch": 0.05,
"learning_rate": 0.00048194583751253763,
"loss": 3.0543,
"theoretical_loss": 4.5146892414068684,
"tokens_seen": 150994944
},
{
"epoch": 0.05,
"learning_rate": 0.00048193580742226686,
"loss": 3.1962,
"theoretical_loss": 4.514433400390554,
"tokens_seen": 151060480
},
{
"epoch": 0.05,
"learning_rate": 0.000481925777331996,
"loss": 3.2645,
"theoretical_loss": 4.514177701407042,
"tokens_seen": 151126016
},
{
"epoch": 0.05,
"learning_rate": 0.0004819157472417252,
"loss": 3.411,
"theoretical_loss": 4.51392214431594,
"tokens_seen": 151191552
},
{
"epoch": 0.05,
"learning_rate": 0.00048190571715145435,
"loss": 3.4078,
"theoretical_loss": 4.513666728977054,
"tokens_seen": 151257088
},
{
"epoch": 0.05,
"learning_rate": 0.0004818956870611836,
"loss": 3.1651,
"theoretical_loss": 4.51341145525039,
"tokens_seen": 151322624
},
{
"epoch": 0.05,
"learning_rate": 0.00048188565697091277,
"loss": 3.0852,
"theoretical_loss": 4.513156322996155,
"tokens_seen": 151388160
},
{
"epoch": 0.05,
"learning_rate": 0.00048187562688064195,
"loss": 3.223,
"theoretical_loss": 4.512901332074751,
"tokens_seen": 151453696
},
{
"epoch": 0.05,
"learning_rate": 0.00048186559679037113,
"loss": 3.2013,
"theoretical_loss": 4.5126464823467805,
"tokens_seen": 151519232
},
{
"epoch": 0.05,
"learning_rate": 0.0004818555667001003,
"loss": 3.4013,
"theoretical_loss": 4.512391773673042,
"tokens_seen": 151584768
},
{
"epoch": 0.05,
"learning_rate": 0.0004818455366098295,
"loss": 3.1806,
"theoretical_loss": 4.5121372059145335,
"tokens_seen": 151650304
},
{
"epoch": 0.05,
"learning_rate": 0.00048183550651955873,
"loss": 3.0253,
"theoretical_loss": 4.511882778932447,
"tokens_seen": 151715840
},
{
"epoch": 0.05,
"learning_rate": 0.00048182547642928786,
"loss": 3.437,
"theoretical_loss": 4.511628492588174,
"tokens_seen": 151781376
},
{
"epoch": 0.05,
"learning_rate": 0.0004818154463390171,
"loss": 3.2479,
"theoretical_loss": 4.5113743467433,
"tokens_seen": 151846912
},
{
"epoch": 0.05,
"learning_rate": 0.0004818054162487462,
"loss": 3.2266,
"theoretical_loss": 4.511120341259608,
"tokens_seen": 151912448
},
{
"epoch": 0.05,
"learning_rate": 0.00048179538615847545,
"loss": 3.065,
"theoretical_loss": 4.510866475999077,
"tokens_seen": 151977984
},
{
"epoch": 0.05,
"learning_rate": 0.00048178535606820463,
"loss": 3.1586,
"theoretical_loss": 4.510612750823878,
"tokens_seen": 152043520
},
{
"epoch": 0.05,
"learning_rate": 0.0004817753259779338,
"loss": 2.9635,
"theoretical_loss": 4.5103591655963795,
"tokens_seen": 152109056
},
{
"epoch": 0.05,
"learning_rate": 0.000481765295887663,
"loss": 3.2094,
"theoretical_loss": 4.510105720179144,
"tokens_seen": 152174592
},
{
"epoch": 0.05,
"learning_rate": 0.00048175526579739223,
"loss": 3.1597,
"theoretical_loss": 4.5098524144349295,
"tokens_seen": 152240128
},
{
"epoch": 0.05,
"learning_rate": 0.00048174523570712136,
"loss": 3.1867,
"theoretical_loss": 4.509599248226683,
"tokens_seen": 152305664
},
{
"epoch": 0.05,
"objective/train/docs_used": 97778,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.6629927158355713,
"objective/train/theoretical_loss": 4.509346221417552,
"objective/train/tokens_used": 172831200,
"theoretical_loss": 4.509346221417552,
"tokens_seen": 152371200
},
{
"epoch": 0.05,
"learning_rate": 0.0004817352056168506,
"loss": 3.2688,
"theoretical_loss": 4.509346221417552,
"tokens_seen": 152371200
},
{
"epoch": 0.05,
"learning_rate": 0.0004817251755265797,
"loss": 3.4886,
"theoretical_loss": 4.509093333870869,
"tokens_seen": 152436736
},
{
"epoch": 0.05,
"learning_rate": 0.00048171514543630896,
"loss": 3.2402,
"theoretical_loss": 4.508840585450166,
"tokens_seen": 152502272
},
{
"epoch": 0.05,
"learning_rate": 0.00048170511534603814,
"loss": 3.3127,
"theoretical_loss": 4.508587976019164,
"tokens_seen": 152567808
},
{
"epoch": 0.05,
"learning_rate": 0.0004816950852557673,
"loss": 3.2319,
"theoretical_loss": 4.508335505441774,
"tokens_seen": 152633344
},
{
"epoch": 0.05,
"learning_rate": 0.0004816850551654965,
"loss": 3.3341,
"theoretical_loss": 4.508083173582105,
"tokens_seen": 152698880
},
{
"epoch": 0.05,
"learning_rate": 0.0004816750250752257,
"loss": 3.0618,
"theoretical_loss": 4.507830980304451,
"tokens_seen": 152764416
},
{
"epoch": 0.05,
"learning_rate": 0.00048166499498495486,
"loss": 3.4654,
"theoretical_loss": 4.5075789254733,
"tokens_seen": 152829952
},
{
"epoch": 0.05,
"learning_rate": 0.0004816549648946841,
"loss": 2.7021,
"theoretical_loss": 4.507327008953329,
"tokens_seen": 152895488
},
{
"epoch": 0.05,
"learning_rate": 0.0004816449348044132,
"loss": 3.5096,
"theoretical_loss": 4.507075230609407,
"tokens_seen": 152961024
},
{
"epoch": 0.05,
"learning_rate": 0.00048163490471414246,
"loss": 3.1766,
"theoretical_loss": 4.506823590306591,
"tokens_seen": 153026560
},
{
"epoch": 0.05,
"learning_rate": 0.0004816248746238716,
"loss": 3.4232,
"theoretical_loss": 4.506572087910127,
"tokens_seen": 153092096
},
{
"epoch": 0.05,
"learning_rate": 0.0004816148445336008,
"loss": 3.3578,
"theoretical_loss": 4.506320723285455,
"tokens_seen": 153157632
},
{
"epoch": 0.05,
"learning_rate": 0.00048160481444333,
"loss": 3.4206,
"theoretical_loss": 4.506069496298198,
"tokens_seen": 153223168
},
{
"epoch": 0.05,
"learning_rate": 0.0004815947843530592,
"loss": 3.5393,
"theoretical_loss": 4.5058184068141705,
"tokens_seen": 153288704
},
{
"epoch": 0.05,
"learning_rate": 0.00048158475426278837,
"loss": 2.9682,
"theoretical_loss": 4.505567454699373,
"tokens_seen": 153354240
},
{
"epoch": 0.05,
"learning_rate": 0.0004815747241725176,
"loss": 3.3476,
"theoretical_loss": 4.505316639819997,
"tokens_seen": 153419776
},
{
"epoch": 0.05,
"learning_rate": 0.00048156469408224673,
"loss": 3.4125,
"theoretical_loss": 4.505065962042418,
"tokens_seen": 153485312
},
{
"epoch": 0.05,
"learning_rate": 0.00048155466399197596,
"loss": 3.0099,
"theoretical_loss": 4.504815421233202,
"tokens_seen": 153550848
},
{
"epoch": 0.05,
"learning_rate": 0.0004815446339017051,
"loss": 3.2099,
"theoretical_loss": 4.504565017259097,
"tokens_seen": 153616384
},
{
"epoch": 0.05,
"learning_rate": 0.0004815346038114343,
"loss": 3.1983,
"theoretical_loss": 4.504314749987044,
"tokens_seen": 153681920
},
{
"epoch": 0.05,
"learning_rate": 0.0004815245737211635,
"loss": 3.4997,
"theoretical_loss": 4.504064619284163,
"tokens_seen": 153747456
},
{
"epoch": 0.05,
"learning_rate": 0.0004815145436308927,
"loss": 3.3542,
"theoretical_loss": 4.503814625017766,
"tokens_seen": 153812992
},
{
"epoch": 0.05,
"learning_rate": 0.00048150451354062187,
"loss": 3.4172,
"theoretical_loss": 4.5035647670553445,
"tokens_seen": 153878528
},
{
"epoch": 0.05,
"learning_rate": 0.00048149448345035105,
"loss": 3.0587,
"theoretical_loss": 4.503315045264581,
"tokens_seen": 153944064
},
{
"epoch": 0.05,
"objective/train/docs_used": 98099,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.236271858215332,
"objective/train/theoretical_loss": 4.503065459513339,
"objective/train/tokens_used": 174469600,
"theoretical_loss": 4.503065459513339,
"tokens_seen": 154009600
},
{
"epoch": 0.05,
"learning_rate": 0.00048148445336008023,
"loss": 3.1071,
"theoretical_loss": 4.503065459513339,
"tokens_seen": 154009600
},
{
"epoch": 0.05,
"learning_rate": 0.00048147442326980947,
"loss": 3.4448,
"theoretical_loss": 4.502816009669665,
"tokens_seen": 154075136
},
{
"epoch": 0.05,
"learning_rate": 0.0004814643931795386,
"loss": 3.301,
"theoretical_loss": 4.502566695601795,
"tokens_seen": 154140672
},
{
"epoch": 0.05,
"learning_rate": 0.00048145436308926783,
"loss": 3.3052,
"theoretical_loss": 4.502317517178142,
"tokens_seen": 154206208
},
{
"epoch": 0.05,
"learning_rate": 0.00048144433299899696,
"loss": 3.1486,
"theoretical_loss": 4.502068474267309,
"tokens_seen": 154271744
},
{
"epoch": 0.05,
"learning_rate": 0.0004814343029087262,
"loss": 2.9085,
"theoretical_loss": 4.501819566738076,
"tokens_seen": 154337280
},
{
"epoch": 0.05,
"learning_rate": 0.00048142427281845537,
"loss": 3.271,
"theoretical_loss": 4.501570794459411,
"tokens_seen": 154402816
},
{
"epoch": 0.05,
"learning_rate": 0.00048141424272818455,
"loss": 3.4126,
"theoretical_loss": 4.501322157300461,
"tokens_seen": 154468352
},
{
"epoch": 0.05,
"learning_rate": 0.00048140421263791373,
"loss": 3.3014,
"theoretical_loss": 4.501073655130554,
"tokens_seen": 154533888
},
{
"epoch": 0.05,
"learning_rate": 0.00048139418254764297,
"loss": 3.3142,
"theoretical_loss": 4.500825287819205,
"tokens_seen": 154599424
},
{
"epoch": 0.05,
"learning_rate": 0.0004813841524573721,
"loss": 3.3548,
"theoretical_loss": 4.500577055236104,
"tokens_seen": 154664960
},
{
"epoch": 0.05,
"learning_rate": 0.00048137412236710133,
"loss": 3.179,
"theoretical_loss": 4.500328957251128,
"tokens_seen": 154730496
},
{
"epoch": 0.05,
"learning_rate": 0.00048136409227683046,
"loss": 3.2137,
"theoretical_loss": 4.500080993734329,
"tokens_seen": 154796032
},
{
"epoch": 0.05,
"learning_rate": 0.0004813540621865597,
"loss": 3.3226,
"theoretical_loss": 4.499833164555944,
"tokens_seen": 154861568
},
{
"epoch": 0.05,
"learning_rate": 0.0004813440320962889,
"loss": 3.3445,
"theoretical_loss": 4.499585469586387,
"tokens_seen": 154927104
},
{
"epoch": 0.05,
"learning_rate": 0.00048133400200601806,
"loss": 3.2798,
"theoretical_loss": 4.499337908696255,
"tokens_seen": 154992640
},
{
"epoch": 0.05,
"learning_rate": 0.00048132397191574724,
"loss": 3.1576,
"theoretical_loss": 4.499090481756321,
"tokens_seen": 155058176
},
{
"epoch": 0.05,
"learning_rate": 0.0004813139418254764,
"loss": 3.3429,
"theoretical_loss": 4.498843188637538,
"tokens_seen": 155123712
},
{
"epoch": 0.05,
"learning_rate": 0.0004813039117352056,
"loss": 3.2675,
"theoretical_loss": 4.498596029211041,
"tokens_seen": 155189248
},
{
"epoch": 0.05,
"learning_rate": 0.00048129388164493483,
"loss": 3.3317,
"theoretical_loss": 4.498349003348137,
"tokens_seen": 155254784
},
{
"epoch": 0.05,
"learning_rate": 0.00048128385155466396,
"loss": 3.3357,
"theoretical_loss": 4.4981021109203185,
"tokens_seen": 155320320
},
{
"epoch": 0.05,
"learning_rate": 0.0004812738214643932,
"loss": 3.4323,
"theoretical_loss": 4.49785535179925,
"tokens_seen": 155385856
},
{
"epoch": 0.05,
"learning_rate": 0.0004812637913741223,
"loss": 3.3122,
"theoretical_loss": 4.497608725856776,
"tokens_seen": 155451392
},
{
"epoch": 0.05,
"learning_rate": 0.00048125376128385156,
"loss": 3.2995,
"theoretical_loss": 4.497362232964919,
"tokens_seen": 155516928
},
{
"epoch": 0.05,
"learning_rate": 0.00048124373119358074,
"loss": 3.3192,
"theoretical_loss": 4.497115872995876,
"tokens_seen": 155582464
},
{
"epoch": 0.05,
"objective/train/docs_used": 99496,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4420876502990723,
"objective/train/theoretical_loss": 4.496869645822022,
"objective/train/tokens_used": 176108000,
"theoretical_loss": 4.496869645822022,
"tokens_seen": 155648000
},
{
"epoch": 0.05,
"learning_rate": 0.0004812337011033099,
"loss": 3.2141,
"theoretical_loss": 4.496869645822022,
"tokens_seen": 155648000
},
{
"epoch": 0.05,
"learning_rate": 0.00048122367101303916,
"loss": 3.2491,
"theoretical_loss": 4.496623551315908,
"tokens_seen": 155713536
},
{
"epoch": 0.05,
"learning_rate": 0.00048121364092276834,
"loss": 3.5781,
"theoretical_loss": 4.496377589350261,
"tokens_seen": 155779072
},
{
"epoch": 0.05,
"learning_rate": 0.0004812036108324975,
"loss": 3.5251,
"theoretical_loss": 4.496131759797984,
"tokens_seen": 155844608
},
{
"epoch": 0.05,
"learning_rate": 0.0004811935807422267,
"loss": 3.1931,
"theoretical_loss": 4.495886062532153,
"tokens_seen": 155910144
},
{
"epoch": 0.05,
"learning_rate": 0.0004811835506519559,
"loss": 3.425,
"theoretical_loss": 4.495640497426023,
"tokens_seen": 155975680
},
{
"epoch": 0.05,
"learning_rate": 0.00048117352056168506,
"loss": 3.1489,
"theoretical_loss": 4.495395064353019,
"tokens_seen": 156041216
},
{
"epoch": 0.05,
"learning_rate": 0.0004811634904714143,
"loss": 3.3909,
"theoretical_loss": 4.4951497631867445,
"tokens_seen": 156106752
},
{
"epoch": 0.05,
"learning_rate": 0.0004811534603811434,
"loss": 3.505,
"theoretical_loss": 4.494904593800973,
"tokens_seen": 156172288
},
{
"epoch": 0.05,
"learning_rate": 0.00048114343029087266,
"loss": 3.4255,
"theoretical_loss": 4.4946595560696565,
"tokens_seen": 156237824
},
{
"epoch": 0.05,
"learning_rate": 0.0004811334002006018,
"loss": 3.2254,
"theoretical_loss": 4.494414649866915,
"tokens_seen": 156303360
},
{
"epoch": 0.05,
"learning_rate": 0.000481123370110331,
"loss": 3.285,
"theoretical_loss": 4.494169875067046,
"tokens_seen": 156368896
},
{
"epoch": 0.05,
"learning_rate": 0.0004811133400200602,
"loss": 3.1795,
"theoretical_loss": 4.493925231544516,
"tokens_seen": 156434432
},
{
"epoch": 0.05,
"learning_rate": 0.0004811033099297894,
"loss": 3.4117,
"theoretical_loss": 4.493680719173968,
"tokens_seen": 156499968
},
{
"epoch": 0.05,
"learning_rate": 0.00048109327983951857,
"loss": 3.1854,
"theoretical_loss": 4.4934363378302145,
"tokens_seen": 156565504
},
{
"epoch": 0.05,
"learning_rate": 0.0004810832497492478,
"loss": 3.1177,
"theoretical_loss": 4.493192087388239,
"tokens_seen": 156631040
},
{
"epoch": 0.05,
"learning_rate": 0.00048107321965897693,
"loss": 3.3501,
"theoretical_loss": 4.4929479677232,
"tokens_seen": 156696576
},
{
"epoch": 0.05,
"learning_rate": 0.00048106318956870616,
"loss": 3.3047,
"theoretical_loss": 4.4927039787104235,
"tokens_seen": 156762112
},
{
"epoch": 0.05,
"learning_rate": 0.0004810531594784353,
"loss": 3.3947,
"theoretical_loss": 4.4924601202254095,
"tokens_seen": 156827648
},
{
"epoch": 0.05,
"learning_rate": 0.0004810431293881645,
"loss": 3.217,
"theoretical_loss": 4.492216392143826,
"tokens_seen": 156893184
},
{
"epoch": 0.05,
"learning_rate": 0.0004810330992978937,
"loss": 3.3287,
"theoretical_loss": 4.491972794341514,
"tokens_seen": 156958720
},
{
"epoch": 0.05,
"learning_rate": 0.0004810230692076229,
"loss": 3.4629,
"theoretical_loss": 4.49172932669448,
"tokens_seen": 157024256
},
{
"epoch": 0.05,
"learning_rate": 0.00048101303911735207,
"loss": 3.2084,
"theoretical_loss": 4.491485989078906,
"tokens_seen": 157089792
},
{
"epoch": 0.05,
"learning_rate": 0.00048100300902708125,
"loss": 3.4562,
"theoretical_loss": 4.491242781371138,
"tokens_seen": 157155328
},
{
"epoch": 0.05,
"learning_rate": 0.00048099297893681043,
"loss": 3.2665,
"theoretical_loss": 4.490999703447697,
"tokens_seen": 157220864
},
{
"epoch": 0.05,
"objective/train/docs_used": 100000,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1658103466033936,
"objective/train/theoretical_loss": 4.4907567551852665,
"objective/train/tokens_used": 177746400,
"theoretical_loss": 4.4907567551852665,
"tokens_seen": 157286400
},
{
"epoch": 0.05,
"learning_rate": 0.00048098294884653967,
"loss": 3.3429,
"theoretical_loss": 4.4907567551852665,
"tokens_seen": 157286400
},
{
"epoch": 0.05,
"learning_rate": 0.0004809729187562688,
"loss": 3.416,
"theoretical_loss": 4.490513936460702,
"tokens_seen": 157351936
},
{
"epoch": 0.05,
"learning_rate": 0.00048096288866599803,
"loss": 3.1769,
"theoretical_loss": 4.490271247151027,
"tokens_seen": 157417472
},
{
"epoch": 0.05,
"learning_rate": 0.00048095285857572716,
"loss": 3.4918,
"theoretical_loss": 4.490028687133432,
"tokens_seen": 157483008
},
{
"epoch": 0.05,
"learning_rate": 0.0004809428284854564,
"loss": 3.1271,
"theoretical_loss": 4.489786256285276,
"tokens_seen": 157548544
},
{
"epoch": 0.05,
"learning_rate": 0.00048093279839518557,
"loss": 3.3183,
"theoretical_loss": 4.489543954484084,
"tokens_seen": 157614080
},
{
"epoch": 0.05,
"learning_rate": 0.00048092276830491475,
"loss": 3.176,
"theoretical_loss": 4.489301781607551,
"tokens_seen": 157679616
},
{
"epoch": 0.05,
"learning_rate": 0.00048091273821464393,
"loss": 3.2113,
"theoretical_loss": 4.489059737533534,
"tokens_seen": 157745152
},
{
"epoch": 0.05,
"learning_rate": 0.00048090270812437317,
"loss": 3.2051,
"theoretical_loss": 4.48881782214006,
"tokens_seen": 157810688
},
{
"epoch": 0.05,
"learning_rate": 0.0004808926780341023,
"loss": 3.2277,
"theoretical_loss": 4.48857603530532,
"tokens_seen": 157876224
},
{
"epoch": 0.05,
"learning_rate": 0.00048088264794383153,
"loss": 3.3081,
"theoretical_loss": 4.488334376907673,
"tokens_seen": 157941760
},
{
"epoch": 0.05,
"learning_rate": 0.00048087261785356066,
"loss": 3.4009,
"theoretical_loss": 4.4880928468256425,
"tokens_seen": 158007296
},
{
"epoch": 0.05,
"learning_rate": 0.0004808625877632899,
"loss": 3.2696,
"theoretical_loss": 4.487851444937916,
"tokens_seen": 158072832
},
{
"epoch": 0.05,
"learning_rate": 0.0004808525576730191,
"loss": 3.3747,
"theoretical_loss": 4.487610171123347,
"tokens_seen": 158138368
},
{
"epoch": 0.05,
"learning_rate": 0.00048084252758274826,
"loss": 3.2603,
"theoretical_loss": 4.487369025260954,
"tokens_seen": 158203904
},
{
"epoch": 0.05,
"learning_rate": 0.00048083249749247744,
"loss": 3.3168,
"theoretical_loss": 4.48712800722992,
"tokens_seen": 158269440
},
{
"epoch": 0.05,
"learning_rate": 0.0004808224674022066,
"loss": 3.0748,
"theoretical_loss": 4.48688711690959,
"tokens_seen": 158334976
},
{
"epoch": 0.05,
"learning_rate": 0.0004808124373119358,
"loss": 3.2276,
"theoretical_loss": 4.486646354179475,
"tokens_seen": 158400512
},
{
"epoch": 0.05,
"learning_rate": 0.00048080240722166503,
"loss": 3.3016,
"theoretical_loss": 4.48640571891925,
"tokens_seen": 158466048
},
{
"epoch": 0.05,
"learning_rate": 0.00048079237713139416,
"loss": 3.2242,
"theoretical_loss": 4.48616521100875,
"tokens_seen": 158531584
},
{
"epoch": 0.05,
"learning_rate": 0.0004807823470411234,
"loss": 3.5084,
"theoretical_loss": 4.485924830327974,
"tokens_seen": 158597120
},
{
"epoch": 0.05,
"learning_rate": 0.0004807723169508525,
"loss": 3.2451,
"theoretical_loss": 4.485684576757087,
"tokens_seen": 158662656
},
{
"epoch": 0.05,
"learning_rate": 0.00048076228686058176,
"loss": 3.5179,
"theoretical_loss": 4.485444450176413,
"tokens_seen": 158728192
},
{
"epoch": 0.05,
"learning_rate": 0.00048075225677031094,
"loss": 3.4267,
"theoretical_loss": 4.485204450466437,
"tokens_seen": 158793728
},
{
"epoch": 0.05,
"learning_rate": 0.0004807422266800401,
"loss": 3.4344,
"theoretical_loss": 4.484964577507808,
"tokens_seen": 158859264
},
{
"epoch": 0.05,
"objective/train/docs_used": 101476,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1423563957214355,
"objective/train/theoretical_loss": 4.484724831181337,
"objective/train/tokens_used": 179384800,
"theoretical_loss": 4.484724831181337,
"tokens_seen": 158924800
},
{
"epoch": 0.05,
"learning_rate": 0.0004807321965897693,
"loss": 3.3062,
"theoretical_loss": 4.484724831181337,
"tokens_seen": 158924800
},
{
"epoch": 0.05,
"learning_rate": 0.00048072216649949854,
"loss": 3.222,
"theoretical_loss": 4.4844852113679945,
"tokens_seen": 158990336
},
{
"epoch": 0.05,
"learning_rate": 0.00048071213640922766,
"loss": 3.2831,
"theoretical_loss": 4.484245717948913,
"tokens_seen": 159055872
},
{
"epoch": 0.05,
"learning_rate": 0.0004807021063189569,
"loss": 3.2262,
"theoretical_loss": 4.484006350805385,
"tokens_seen": 159121408
},
{
"epoch": 0.05,
"learning_rate": 0.000480692076228686,
"loss": 3.3261,
"theoretical_loss": 4.483767109818862,
"tokens_seen": 159186944
},
{
"epoch": 0.05,
"learning_rate": 0.00048068204613841526,
"loss": 3.483,
"theoretical_loss": 4.483527994870958,
"tokens_seen": 159252480
},
{
"epoch": 0.05,
"learning_rate": 0.00048067201604814444,
"loss": 3.2439,
"theoretical_loss": 4.483289005843445,
"tokens_seen": 159318016
},
{
"epoch": 0.05,
"learning_rate": 0.0004806619859578736,
"loss": 3.4449,
"theoretical_loss": 4.483050142618255,
"tokens_seen": 159383552
},
{
"epoch": 0.05,
"learning_rate": 0.0004806519558676028,
"loss": 3.1009,
"theoretical_loss": 4.482811405077482,
"tokens_seen": 159449088
},
{
"epoch": 0.05,
"learning_rate": 0.000480641925777332,
"loss": 3.2611,
"theoretical_loss": 4.482572793103373,
"tokens_seen": 159514624
},
{
"epoch": 0.05,
"learning_rate": 0.00048063189568706117,
"loss": 3.3544,
"theoretical_loss": 4.482334306578339,
"tokens_seen": 159580160
},
{
"epoch": 0.05,
"learning_rate": 0.0004806218655967904,
"loss": 3.1545,
"theoretical_loss": 4.482095945384946,
"tokens_seen": 159645696
},
{
"epoch": 0.05,
"learning_rate": 0.00048061183550651953,
"loss": 3.2616,
"theoretical_loss": 4.481857709405919,
"tokens_seen": 159711232
},
{
"epoch": 0.05,
"learning_rate": 0.00048060180541624877,
"loss": 3.3525,
"theoretical_loss": 4.4816195985241425,
"tokens_seen": 159776768
},
{
"epoch": 0.05,
"learning_rate": 0.00048059177532597795,
"loss": 3.1274,
"theoretical_loss": 4.481381612622657,
"tokens_seen": 159842304
},
{
"epoch": 0.05,
"learning_rate": 0.00048058174523570713,
"loss": 3.5142,
"theoretical_loss": 4.481143751584659,
"tokens_seen": 159907840
},
{
"epoch": 0.05,
"learning_rate": 0.0004805717151454363,
"loss": 3.4438,
"theoretical_loss": 4.480906015293505,
"tokens_seen": 159973376
},
{
"epoch": 0.05,
"learning_rate": 0.0004805616850551655,
"loss": 3.3428,
"theoretical_loss": 4.480668403632706,
"tokens_seen": 160038912
},
{
"epoch": 0.05,
"learning_rate": 0.00048055165496489467,
"loss": 3.4011,
"theoretical_loss": 4.480430916485929,
"tokens_seen": 160104448
},
{
"epoch": 0.05,
"learning_rate": 0.0004805416248746239,
"loss": 3.1368,
"theoretical_loss": 4.480193553736999,
"tokens_seen": 160169984
},
{
"epoch": 0.05,
"learning_rate": 0.00048053159478435303,
"loss": 3.1091,
"theoretical_loss": 4.479956315269897,
"tokens_seen": 160235520
},
{
"epoch": 0.05,
"learning_rate": 0.00048052156469408227,
"loss": 3.4397,
"theoretical_loss": 4.479719200968757,
"tokens_seen": 160301056
},
{
"epoch": 0.05,
"learning_rate": 0.0004805115346038114,
"loss": 3.4909,
"theoretical_loss": 4.479482210717871,
"tokens_seen": 160366592
},
{
"epoch": 0.05,
"learning_rate": 0.00048050150451354063,
"loss": 3.3116,
"theoretical_loss": 4.479245344401685,
"tokens_seen": 160432128
},
{
"epoch": 0.05,
"learning_rate": 0.0004804914744232698,
"loss": 3.2769,
"theoretical_loss": 4.479008601904798,
"tokens_seen": 160497664
},
{
"epoch": 0.05,
"objective/train/docs_used": 102054,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4178307056427,
"objective/train/theoretical_loss": 4.478771983111967,
"objective/train/tokens_used": 181023200,
"theoretical_loss": 4.478771983111967,
"tokens_seen": 160563200
},
{
"epoch": 0.05,
"learning_rate": 0.000480481444332999,
"loss": 3.467,
"theoretical_loss": 4.478771983111967,
"tokens_seen": 160563200
},
{
"epoch": 0.05,
"learning_rate": 0.00048047141424272823,
"loss": 3.4466,
"theoretical_loss": 4.478535487908101,
"tokens_seen": 160628736
},
{
"epoch": 0.05,
"learning_rate": 0.00048046138415245736,
"loss": 3.3249,
"theoretical_loss": 4.478299116178265,
"tokens_seen": 160694272
},
{
"epoch": 0.05,
"learning_rate": 0.0004804513540621866,
"loss": 3.2936,
"theoretical_loss": 4.478062867807674,
"tokens_seen": 160759808
},
{
"epoch": 0.05,
"learning_rate": 0.00048044132397191577,
"loss": 3.466,
"theoretical_loss": 4.4778267426817,
"tokens_seen": 160825344
},
{
"epoch": 0.05,
"learning_rate": 0.00048043129388164495,
"loss": 3.4092,
"theoretical_loss": 4.477590740685867,
"tokens_seen": 160890880
},
{
"epoch": 0.05,
"learning_rate": 0.00048042126379137413,
"loss": 3.2282,
"theoretical_loss": 4.47735486170585,
"tokens_seen": 160956416
},
{
"epoch": 0.05,
"learning_rate": 0.00048041123370110337,
"loss": 3.088,
"theoretical_loss": 4.47711910562748,
"tokens_seen": 161021952
},
{
"epoch": 0.05,
"learning_rate": 0.0004804012036108325,
"loss": 3.1203,
"theoretical_loss": 4.4768834723367394,
"tokens_seen": 161087488
},
{
"epoch": 0.05,
"learning_rate": 0.00048039117352056173,
"loss": 3.5309,
"theoretical_loss": 4.4766479617197605,
"tokens_seen": 161153024
},
{
"epoch": 0.05,
"learning_rate": 0.00048038114343029086,
"loss": 3.1474,
"theoretical_loss": 4.476412573662829,
"tokens_seen": 161218560
},
{
"epoch": 0.05,
"learning_rate": 0.0004803711133400201,
"loss": 3.1828,
"theoretical_loss": 4.4761773080523835,
"tokens_seen": 161284096
},
{
"epoch": 0.05,
"learning_rate": 0.0004803610832497493,
"loss": 3.3428,
"theoretical_loss": 4.475942164775013,
"tokens_seen": 161349632
},
{
"epoch": 0.05,
"learning_rate": 0.00048035105315947846,
"loss": 3.0744,
"theoretical_loss": 4.475707143717455,
"tokens_seen": 161415168
},
{
"epoch": 0.05,
"learning_rate": 0.00048034102306920764,
"loss": 3.1276,
"theoretical_loss": 4.475472244766601,
"tokens_seen": 161480704
},
{
"epoch": 0.05,
"learning_rate": 0.0004803309929789368,
"loss": 3.0242,
"theoretical_loss": 4.475237467809492,
"tokens_seen": 161546240
},
{
"epoch": 0.05,
"learning_rate": 0.000480320962888666,
"loss": 3.4158,
"theoretical_loss": 4.47500281273332,
"tokens_seen": 161611776
},
{
"epoch": 0.05,
"learning_rate": 0.00048031093279839523,
"loss": 3.2473,
"theoretical_loss": 4.474768279425424,
"tokens_seen": 161677312
},
{
"epoch": 0.05,
"learning_rate": 0.00048030090270812436,
"loss": 3.1312,
"theoretical_loss": 4.474533867773299,
"tokens_seen": 161742848
},
{
"epoch": 0.05,
"learning_rate": 0.0004802908726178536,
"loss": 3.3392,
"theoretical_loss": 4.474299577664581,
"tokens_seen": 161808384
},
{
"epoch": 0.05,
"learning_rate": 0.0004802808425275827,
"loss": 3.3211,
"theoretical_loss": 4.474065408987063,
"tokens_seen": 161873920
},
{
"epoch": 0.05,
"learning_rate": 0.00048027081243731196,
"loss": 3.3298,
"theoretical_loss": 4.473831361628682,
"tokens_seen": 161939456
},
{
"epoch": 0.05,
"learning_rate": 0.00048026078234704114,
"loss": 3.3306,
"theoretical_loss": 4.473597435477526,
"tokens_seen": 162004992
},
{
"epoch": 0.05,
"learning_rate": 0.0004802507522567703,
"loss": 3.1812,
"theoretical_loss": 4.473363630421831,
"tokens_seen": 162070528
},
{
"epoch": 0.05,
"learning_rate": 0.0004802407221664995,
"loss": 3.1926,
"theoretical_loss": 4.473129946349982,
"tokens_seen": 162136064
},
{
"epoch": 0.05,
"objective/train/docs_used": 102718,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.263267993927002,
"objective/train/theoretical_loss": 4.472896383150508,
"objective/train/tokens_used": 182661600,
"theoretical_loss": 4.472896383150508,
"tokens_seen": 162201600
},
{
"epoch": 0.05,
"learning_rate": 0.00048023069207622874,
"loss": 3.1614,
"theoretical_loss": 4.472896383150508,
"tokens_seen": 162201600
},
{
"epoch": 0.05,
"learning_rate": 0.00048022066198595786,
"loss": 3.4761,
"theoretical_loss": 4.472662940712091,
"tokens_seen": 162267136
},
{
"epoch": 0.05,
"learning_rate": 0.0004802106318956871,
"loss": 3.1622,
"theoretical_loss": 4.472429618923558,
"tokens_seen": 162332672
},
{
"epoch": 0.05,
"learning_rate": 0.00048020060180541623,
"loss": 3.2604,
"theoretical_loss": 4.472196417673883,
"tokens_seen": 162398208
},
{
"epoch": 0.05,
"learning_rate": 0.00048019057171514546,
"loss": 3.373,
"theoretical_loss": 4.471963336852187,
"tokens_seen": 162463744
},
{
"epoch": 0.05,
"learning_rate": 0.00048018054162487464,
"loss": 3.1342,
"theoretical_loss": 4.471730376347738,
"tokens_seen": 162529280
},
{
"epoch": 0.05,
"learning_rate": 0.0004801705115346038,
"loss": 3.1844,
"theoretical_loss": 4.4714975360499505,
"tokens_seen": 162594816
},
{
"epoch": 0.05,
"learning_rate": 0.000480160481444333,
"loss": 3.4885,
"theoretical_loss": 4.471264815848384,
"tokens_seen": 162660352
},
{
"epoch": 0.05,
"learning_rate": 0.0004801504513540622,
"loss": 3.2073,
"theoretical_loss": 4.471032215632746,
"tokens_seen": 162725888
},
{
"epoch": 0.05,
"learning_rate": 0.00048014042126379137,
"loss": 3.1441,
"theoretical_loss": 4.470799735292889,
"tokens_seen": 162791424
},
{
"epoch": 0.05,
"learning_rate": 0.0004801303911735206,
"loss": 3.3775,
"theoretical_loss": 4.470567374718808,
"tokens_seen": 162856960
},
{
"epoch": 0.05,
"learning_rate": 0.00048012036108324973,
"loss": 3.4188,
"theoretical_loss": 4.470335133800649,
"tokens_seen": 162922496
},
{
"epoch": 0.05,
"learning_rate": 0.00048011033099297897,
"loss": 3.1071,
"theoretical_loss": 4.470103012428696,
"tokens_seen": 162988032
},
{
"epoch": 0.05,
"learning_rate": 0.00048010030090270815,
"loss": 3.3467,
"theoretical_loss": 4.469871010493383,
"tokens_seen": 163053568
},
{
"epoch": 0.05,
"learning_rate": 0.00048009027081243733,
"loss": 3.3758,
"theoretical_loss": 4.469639127885287,
"tokens_seen": 163119104
},
{
"epoch": 0.05,
"learning_rate": 0.0004800802407221665,
"loss": 3.1418,
"theoretical_loss": 4.4694073644951295,
"tokens_seen": 163184640
},
{
"epoch": 0.05,
"learning_rate": 0.0004800702106318957,
"loss": 3.2458,
"theoretical_loss": 4.469175720213771,
"tokens_seen": 163250176
},
{
"epoch": 0.05,
"learning_rate": 0.00048006018054162487,
"loss": 3.3036,
"theoretical_loss": 4.468944194932225,
"tokens_seen": 163315712
},
{
"epoch": 0.05,
"learning_rate": 0.0004800501504513541,
"loss": 3.2241,
"theoretical_loss": 4.468712788541639,
"tokens_seen": 163381248
},
{
"epoch": 0.05,
"learning_rate": 0.00048004012036108323,
"loss": 3.2485,
"theoretical_loss": 4.46848150093331,
"tokens_seen": 163446784
},
{
"epoch": 0.05,
"learning_rate": 0.00048003009027081247,
"loss": 3.286,
"theoretical_loss": 4.468250331998676,
"tokens_seen": 163512320
},
{
"epoch": 0.05,
"learning_rate": 0.0004800200601805416,
"loss": 3.0841,
"theoretical_loss": 4.468019281629316,
"tokens_seen": 163577856
},
{
"epoch": 0.05,
"learning_rate": 0.00048001003009027083,
"loss": 3.1869,
"theoretical_loss": 4.467788349716955,
"tokens_seen": 163643392
},
{
"epoch": 0.05,
"learning_rate": 0.00048,
"loss": 3.2902,
"theoretical_loss": 4.467557536153457,
"tokens_seen": 163708928
},
{
"epoch": 0.05,
"learning_rate": 0.0004799899699097292,
"loss": 3.179,
"theoretical_loss": 4.467326840830829,
"tokens_seen": 163774464
},
{
"debugging/Self-BLEU-5": 0.47376287031010694,
"debugging/distinct-1-grams": 0.7718165351312889,
"debugging/distinct-2-grams": 0.9624982234606337,
"debugging/entropy-1-grams": 5.847463360327344,
"debugging/entropy-2-grams": 6.816946097439353,
"debugging/length": 477.7857142857143,
"debugging/num_segments": 14,
"epoch": 0.05,
"objective/train/docs_used": 104069,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2056386470794678,
"objective/train/theoretical_loss": 4.467096263641219,
"objective/train/tokens_used": 184300000,
"theoretical_loss": 4.467096263641219,
"tokens_seen": 163840000
},
{
"epoch": 0.05,
"learning_rate": 0.0004799799398194584,
"loss": 3.24,
"theoretical_loss": 4.467096263641219,
"tokens_seen": 163840000
},
{
"epoch": 0.05,
"learning_rate": 0.00047996990972918756,
"loss": 3.1111,
"theoretical_loss": 4.466865804476919,
"tokens_seen": 163905536
},
{
"epoch": 0.05,
"learning_rate": 0.00047995987963891674,
"loss": 3.1284,
"theoretical_loss": 4.466635463230359,
"tokens_seen": 163971072
},
{
"epoch": 0.05,
"learning_rate": 0.00047994984954864597,
"loss": 3.3292,
"theoretical_loss": 4.466405239794113,
"tokens_seen": 164036608
},
{
"epoch": 0.05,
"learning_rate": 0.0004799398194583751,
"loss": 3.3509,
"theoretical_loss": 4.466175134060894,
"tokens_seen": 164102144
},
{
"epoch": 0.05,
"learning_rate": 0.00047992978936810433,
"loss": 3.4643,
"theoretical_loss": 4.465945145923554,
"tokens_seen": 164167680
},
{
"epoch": 0.05,
"learning_rate": 0.0004799197592778335,
"loss": 3.0926,
"theoretical_loss": 4.4657152752750875,
"tokens_seen": 164233216
},
{
"epoch": 0.05,
"learning_rate": 0.0004799097291875627,
"loss": 3.2696,
"theoretical_loss": 4.465485522008629,
"tokens_seen": 164298752
},
{
"epoch": 0.05,
"learning_rate": 0.0004798996990972919,
"loss": 3.158,
"theoretical_loss": 4.465255886017452,
"tokens_seen": 164364288
},
{
"epoch": 0.05,
"learning_rate": 0.00047988966900702106,
"loss": 3.3011,
"theoretical_loss": 4.465026367194971,
"tokens_seen": 164429824
},
{
"epoch": 0.05,
"learning_rate": 0.00047987963891675024,
"loss": 3.0533,
"theoretical_loss": 4.464796965434738,
"tokens_seen": 164495360
},
{
"epoch": 0.05,
"learning_rate": 0.0004798696088264795,
"loss": 3.1821,
"theoretical_loss": 4.464567680630443,
"tokens_seen": 164560896
},
{
"epoch": 0.05,
"learning_rate": 0.0004798595787362086,
"loss": 3.2458,
"theoretical_loss": 4.464338512675919,
"tokens_seen": 164626432
},
{
"epoch": 0.05,
"learning_rate": 0.00047984954864593784,
"loss": 3.3066,
"theoretical_loss": 4.464109461465133,
"tokens_seen": 164691968
},
{
"epoch": 0.05,
"learning_rate": 0.00047983951855566696,
"loss": 3.1252,
"theoretical_loss": 4.4638805268921935,
"tokens_seen": 164757504
},
{
"epoch": 0.05,
"learning_rate": 0.0004798294884653962,
"loss": 3.1787,
"theoretical_loss": 4.463651708851346,
"tokens_seen": 164823040
},
{
"epoch": 0.05,
"learning_rate": 0.0004798194583751254,
"loss": 3.209,
"theoretical_loss": 4.463423007236974,
"tokens_seen": 164888576
},
{
"epoch": 0.05,
"learning_rate": 0.00047980942828485456,
"loss": 3.1093,
"theoretical_loss": 4.4631944219436,
"tokens_seen": 164954112
},
{
"epoch": 0.05,
"learning_rate": 0.00047979939819458374,
"loss": 3.1704,
"theoretical_loss": 4.462965952865879,
"tokens_seen": 165019648
},
{
"epoch": 0.05,
"learning_rate": 0.0004797893681043129,
"loss": 3.3571,
"theoretical_loss": 4.46273759989861,
"tokens_seen": 165085184
},
{
"epoch": 0.05,
"learning_rate": 0.0004797793380140421,
"loss": 3.3585,
"theoretical_loss": 4.462509362936723,
"tokens_seen": 165150720
},
{
"epoch": 0.05,
"learning_rate": 0.00047976930792377134,
"loss": 3.4484,
"theoretical_loss": 4.46228124187529,
"tokens_seen": 165216256
},
{
"epoch": 0.05,
"learning_rate": 0.00047975927783350047,
"loss": 3.2982,
"theoretical_loss": 4.462053236609516,
"tokens_seen": 165281792
},
{
"epoch": 0.05,
"learning_rate": 0.0004797492477432297,
"loss": 3.2709,
"theoretical_loss": 4.461825347034742,
"tokens_seen": 165347328
},
{
"epoch": 0.05,
"learning_rate": 0.0004797392176529589,
"loss": 3.5718,
"theoretical_loss": 4.461597573046449,
"tokens_seen": 165412864
},
{
"epoch": 0.05,
"objective/train/docs_used": 105300,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3106820583343506,
"objective/train/theoretical_loss": 4.461369914540247,
"objective/train/tokens_used": 185938400,
"theoretical_loss": 4.461369914540247,
"tokens_seen": 165478400
},
{
"epoch": 0.05,
"learning_rate": 0.00047972918756268807,
"loss": 3.3645,
"theoretical_loss": 4.461369914540247,
"tokens_seen": 165478400
},
{
"epoch": 0.05,
"learning_rate": 0.0004797191574724173,
"loss": 3.322,
"theoretical_loss": 4.4611423714118885,
"tokens_seen": 165543936
},
{
"epoch": 0.05,
"learning_rate": 0.00047970912738214643,
"loss": 3.3285,
"theoretical_loss": 4.460914943557256,
"tokens_seen": 165609472
},
{
"epoch": 0.05,
"learning_rate": 0.00047969909729187566,
"loss": 3.2845,
"theoretical_loss": 4.460687630872371,
"tokens_seen": 165675008
},
{
"epoch": 0.05,
"learning_rate": 0.00047968906720160484,
"loss": 3.2704,
"theoretical_loss": 4.46046043325339,
"tokens_seen": 165740544
},
{
"epoch": 0.05,
"learning_rate": 0.000479679037111334,
"loss": 3.5378,
"theoretical_loss": 4.460233350596599,
"tokens_seen": 165806080
},
{
"epoch": 0.05,
"learning_rate": 0.0004796690070210632,
"loss": 3.2292,
"theoretical_loss": 4.460006382798425,
"tokens_seen": 165871616
},
{
"epoch": 0.05,
"learning_rate": 0.0004796589769307924,
"loss": 3.4241,
"theoretical_loss": 4.459779529755423,
"tokens_seen": 165937152
},
{
"epoch": 0.05,
"learning_rate": 0.00047964894684052157,
"loss": 3.3032,
"theoretical_loss": 4.459552791364288,
"tokens_seen": 166002688
},
{
"epoch": 0.05,
"learning_rate": 0.0004796389167502508,
"loss": 3.0771,
"theoretical_loss": 4.459326167521844,
"tokens_seen": 166068224
},
{
"epoch": 0.05,
"learning_rate": 0.00047962888665997993,
"loss": 3.6786,
"theoretical_loss": 4.4590996581250515,
"tokens_seen": 166133760
},
{
"epoch": 0.05,
"learning_rate": 0.00047961885656970917,
"loss": 3.4545,
"theoretical_loss": 4.458873263071002,
"tokens_seen": 166199296
},
{
"epoch": 0.05,
"learning_rate": 0.00047960882647943835,
"loss": 3.228,
"theoretical_loss": 4.458646982256921,
"tokens_seen": 166264832
},
{
"epoch": 0.05,
"learning_rate": 0.00047959879638916753,
"loss": 3.2841,
"theoretical_loss": 4.458420815580169,
"tokens_seen": 166330368
},
{
"epoch": 0.05,
"learning_rate": 0.0004795887662988967,
"loss": 2.9158,
"theoretical_loss": 4.458194762938234,
"tokens_seen": 166395904
},
{
"epoch": 0.05,
"learning_rate": 0.0004795787362086259,
"loss": 3.1962,
"theoretical_loss": 4.457968824228743,
"tokens_seen": 166461440
},
{
"epoch": 0.05,
"learning_rate": 0.00047956870611835507,
"loss": 3.3127,
"theoretical_loss": 4.457742999349449,
"tokens_seen": 166526976
},
{
"epoch": 0.05,
"learning_rate": 0.0004795586760280843,
"loss": 3.1228,
"theoretical_loss": 4.4575172881982414,
"tokens_seen": 166592512
},
{
"epoch": 0.05,
"learning_rate": 0.00047954864593781343,
"loss": 3.3468,
"theoretical_loss": 4.457291690673139,
"tokens_seen": 166658048
},
{
"epoch": 0.05,
"learning_rate": 0.00047953861584754267,
"loss": 3.1293,
"theoretical_loss": 4.457066206672291,
"tokens_seen": 166723584
},
{
"epoch": 0.05,
"learning_rate": 0.0004795285857572718,
"loss": 3.4924,
"theoretical_loss": 4.456840836093983,
"tokens_seen": 166789120
},
{
"epoch": 0.05,
"learning_rate": 0.00047951855566700103,
"loss": 3.3401,
"theoretical_loss": 4.456615578836625,
"tokens_seen": 166854656
},
{
"epoch": 0.05,
"learning_rate": 0.0004795085255767302,
"loss": 3.3277,
"theoretical_loss": 4.456390434798762,
"tokens_seen": 166920192
},
{
"epoch": 0.05,
"learning_rate": 0.0004794984954864594,
"loss": 3.1098,
"theoretical_loss": 4.45616540387907,
"tokens_seen": 166985728
},
{
"epoch": 0.05,
"learning_rate": 0.0004794884653961886,
"loss": 3.3298,
"theoretical_loss": 4.4559404859763525,
"tokens_seen": 167051264
},
{
"epoch": 0.05,
"objective/train/docs_used": 106148,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4106695652008057,
"objective/train/theoretical_loss": 4.455715680989545,
"objective/train/tokens_used": 187576800,
"theoretical_loss": 4.455715680989545,
"tokens_seen": 167116800
},
{
"epoch": 0.05,
"learning_rate": 0.00047947843530591776,
"loss": 2.9402,
"theoretical_loss": 4.455715680989545,
"tokens_seen": 167116800
},
{
"epoch": 0.05,
"learning_rate": 0.00047946840521564694,
"loss": 3.432,
"theoretical_loss": 4.455490988817713,
"tokens_seen": 167182336
},
{
"epoch": 0.05,
"learning_rate": 0.00047945837512537617,
"loss": 3.1718,
"theoretical_loss": 4.4552664093600525,
"tokens_seen": 167247872
},
{
"epoch": 0.05,
"learning_rate": 0.0004794483450351053,
"loss": 3.1644,
"theoretical_loss": 4.455041942515887,
"tokens_seen": 167313408
},
{
"epoch": 0.05,
"learning_rate": 0.00047943831494483453,
"loss": 3.1954,
"theoretical_loss": 4.454817588184669,
"tokens_seen": 167378944
},
{
"epoch": 0.05,
"learning_rate": 0.0004794282848545637,
"loss": 3.2134,
"theoretical_loss": 4.454593346265984,
"tokens_seen": 167444480
},
{
"epoch": 0.05,
"learning_rate": 0.0004794182547642929,
"loss": 3.4663,
"theoretical_loss": 4.454369216659542,
"tokens_seen": 167510016
},
{
"epoch": 0.05,
"learning_rate": 0.0004794082246740221,
"loss": 3.3214,
"theoretical_loss": 4.454145199265183,
"tokens_seen": 167575552
},
{
"epoch": 0.05,
"learning_rate": 0.00047939819458375126,
"loss": 3.2135,
"theoretical_loss": 4.453921293982877,
"tokens_seen": 167641088
},
{
"epoch": 0.05,
"learning_rate": 0.00047938816449348044,
"loss": 3.3655,
"theoretical_loss": 4.453697500712722,
"tokens_seen": 167706624
},
{
"epoch": 0.05,
"learning_rate": 0.0004793781344032097,
"loss": 3.202,
"theoretical_loss": 4.453473819354942,
"tokens_seen": 167772160
},
{
"epoch": 0.05,
"learning_rate": 0.0004793681043129388,
"loss": 3.2527,
"theoretical_loss": 4.453250249809889,
"tokens_seen": 167837696
},
{
"epoch": 0.05,
"learning_rate": 0.00047935807422266804,
"loss": 3.2188,
"theoretical_loss": 4.453026791978045,
"tokens_seen": 167903232
},
{
"epoch": 0.05,
"learning_rate": 0.00047934804413239716,
"loss": 3.2968,
"theoretical_loss": 4.4528034457600185,
"tokens_seen": 167968768
},
{
"epoch": 0.05,
"learning_rate": 0.0004793380140421264,
"loss": 3.133,
"theoretical_loss": 4.452580211056542,
"tokens_seen": 168034304
},
{
"epoch": 0.05,
"learning_rate": 0.0004793279839518556,
"loss": 3.1103,
"theoretical_loss": 4.452357087768481,
"tokens_seen": 168099840
},
{
"epoch": 0.05,
"learning_rate": 0.00047931795386158476,
"loss": 3.1566,
"theoretical_loss": 4.45213407579682,
"tokens_seen": 168165376
},
{
"epoch": 0.05,
"learning_rate": 0.00047930792377131394,
"loss": 3.3208,
"theoretical_loss": 4.451911175042679,
"tokens_seen": 168230912
},
{
"epoch": 0.05,
"learning_rate": 0.0004792978936810431,
"loss": 3.3578,
"theoretical_loss": 4.451688385407296,
"tokens_seen": 168296448
},
{
"epoch": 0.05,
"learning_rate": 0.0004792878635907723,
"loss": 3.0301,
"theoretical_loss": 4.451465706792041,
"tokens_seen": 168361984
},
{
"epoch": 0.05,
"learning_rate": 0.00047927783350050154,
"loss": 3.3597,
"theoretical_loss": 4.4512431390984055,
"tokens_seen": 168427520
},
{
"epoch": 0.05,
"learning_rate": 0.00047926780341023067,
"loss": 3.2,
"theoretical_loss": 4.451020682228011,
"tokens_seen": 168493056
},
{
"epoch": 0.05,
"learning_rate": 0.0004792577733199599,
"loss": 3.3903,
"theoretical_loss": 4.450798336082601,
"tokens_seen": 168558592
},
{
"epoch": 0.05,
"learning_rate": 0.0004792477432296891,
"loss": 3.5916,
"theoretical_loss": 4.450576100564046,
"tokens_seen": 168624128
},
{
"epoch": 0.05,
"learning_rate": 0.00047923771313941827,
"loss": 3.0493,
"theoretical_loss": 4.450353975574341,
"tokens_seen": 168689664
},
{
"epoch": 0.05,
"objective/train/docs_used": 106755,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.232574701309204,
"objective/train/theoretical_loss": 4.450131961015606,
"objective/train/tokens_used": 189215200,
"theoretical_loss": 4.450131961015606,
"tokens_seen": 168755200
},
{
"epoch": 0.05,
"learning_rate": 0.00047922768304914745,
"loss": 3.3465,
"theoretical_loss": 4.450131961015606,
"tokens_seen": 168755200
},
{
"epoch": 0.05,
"learning_rate": 0.00047921765295887663,
"loss": 3.4464,
"theoretical_loss": 4.449910056790086,
"tokens_seen": 168820736
},
{
"epoch": 0.05,
"learning_rate": 0.0004792076228686058,
"loss": 3.2509,
"theoretical_loss": 4.44968826280015,
"tokens_seen": 168886272
},
{
"epoch": 0.05,
"learning_rate": 0.00047919759277833504,
"loss": 3.4296,
"theoretical_loss": 4.4494665789482895,
"tokens_seen": 168951808
},
{
"epoch": 0.05,
"learning_rate": 0.00047918756268806417,
"loss": 3.2661,
"theoretical_loss": 4.449245005137125,
"tokens_seen": 169017344
},
{
"epoch": 0.05,
"learning_rate": 0.0004791775325977934,
"loss": 3.2495,
"theoretical_loss": 4.449023541269395,
"tokens_seen": 169082880
},
{
"epoch": 0.05,
"learning_rate": 0.00047916750250752253,
"loss": 3.1478,
"theoretical_loss": 4.448802187247966,
"tokens_seen": 169148416
},
{
"epoch": 0.05,
"learning_rate": 0.00047915747241725177,
"loss": 3.4552,
"theoretical_loss": 4.448580942975825,
"tokens_seen": 169213952
},
{
"epoch": 0.05,
"learning_rate": 0.00047914744232698095,
"loss": 3.34,
"theoretical_loss": 4.448359808356084,
"tokens_seen": 169279488
},
{
"epoch": 0.05,
"learning_rate": 0.00047913741223671013,
"loss": 3.2944,
"theoretical_loss": 4.448138783291979,
"tokens_seen": 169345024
},
{
"epoch": 0.05,
"learning_rate": 0.0004791273821464393,
"loss": 3.2205,
"theoretical_loss": 4.447917867686863,
"tokens_seen": 169410560
},
{
"epoch": 0.05,
"learning_rate": 0.00047911735205616855,
"loss": 3.1967,
"theoretical_loss": 4.44769706144422,
"tokens_seen": 169476096
},
{
"epoch": 0.05,
"learning_rate": 0.0004791073219658977,
"loss": 3.0132,
"theoretical_loss": 4.44747636446765,
"tokens_seen": 169541632
},
{
"epoch": 0.05,
"learning_rate": 0.0004790972918756269,
"loss": 3.4747,
"theoretical_loss": 4.447255776660878,
"tokens_seen": 169607168
},
{
"epoch": 0.05,
"learning_rate": 0.00047908726178535604,
"loss": 3.3565,
"theoretical_loss": 4.44703529792775,
"tokens_seen": 169672704
},
{
"epoch": 0.05,
"learning_rate": 0.00047907723169508527,
"loss": 3.0567,
"theoretical_loss": 4.446814928172234,
"tokens_seen": 169738240
},
{
"epoch": 0.05,
"learning_rate": 0.00047906720160481445,
"loss": 3.1763,
"theoretical_loss": 4.446594667298421,
"tokens_seen": 169803776
},
{
"epoch": 0.05,
"learning_rate": 0.00047905717151454363,
"loss": 3.3902,
"theoretical_loss": 4.446374515210521,
"tokens_seen": 169869312
},
{
"epoch": 0.05,
"learning_rate": 0.0004790471414242728,
"loss": 3.4254,
"theoretical_loss": 4.446154471812866,
"tokens_seen": 169934848
},
{
"epoch": 0.05,
"learning_rate": 0.000479037111334002,
"loss": 3.2961,
"theoretical_loss": 4.445934537009911,
"tokens_seen": 170000384
},
{
"epoch": 0.05,
"learning_rate": 0.0004790270812437312,
"loss": 2.8793,
"theoretical_loss": 4.445714710706228,
"tokens_seen": 170065920
},
{
"epoch": 0.05,
"learning_rate": 0.0004790170511534604,
"loss": 2.8962,
"theoretical_loss": 4.445494992806513,
"tokens_seen": 170131456
},
{
"epoch": 0.05,
"learning_rate": 0.00047900702106318954,
"loss": 3.2095,
"theoretical_loss": 4.44527538321558,
"tokens_seen": 170196992
},
{
"epoch": 0.05,
"learning_rate": 0.0004789969909729188,
"loss": 3.4441,
"theoretical_loss": 4.445055881838365,
"tokens_seen": 170262528
},
{
"epoch": 0.05,
"learning_rate": 0.0004789869608826479,
"loss": 3.3936,
"theoretical_loss": 4.444836488579924,
"tokens_seen": 170328064
},
{
"epoch": 0.05,
"objective/train/docs_used": 107830,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1494503021240234,
"objective/train/theoretical_loss": 4.44461720334543,
"objective/train/tokens_used": 190853600,
"theoretical_loss": 4.44461720334543,
"tokens_seen": 170393600
},
{
"epoch": 0.05,
"learning_rate": 0.00047897693079237714,
"loss": 3.2369,
"theoretical_loss": 4.44461720334543,
"tokens_seen": 170393600
},
{
"epoch": 0.05,
"learning_rate": 0.00047896690070210637,
"loss": 3.3457,
"theoretical_loss": 4.444398026040179,
"tokens_seen": 170459136
},
{
"epoch": 0.05,
"learning_rate": 0.0004789568706118355,
"loss": 3.3824,
"theoretical_loss": 4.444178956569585,
"tokens_seen": 170524672
},
{
"epoch": 0.05,
"learning_rate": 0.00047894684052156473,
"loss": 3.4014,
"theoretical_loss": 4.443959994839181,
"tokens_seen": 170590208
},
{
"epoch": 0.05,
"learning_rate": 0.0004789368104312939,
"loss": 3.3058,
"theoretical_loss": 4.44374114075462,
"tokens_seen": 170655744
},
{
"epoch": 0.05,
"learning_rate": 0.0004789267803410231,
"loss": 3.1202,
"theoretical_loss": 4.443522394221671,
"tokens_seen": 170721280
},
{
"epoch": 0.05,
"learning_rate": 0.0004789167502507523,
"loss": 3.3846,
"theoretical_loss": 4.443303755146225,
"tokens_seen": 170786816
},
{
"epoch": 0.05,
"learning_rate": 0.00047890672016048146,
"loss": 3.2755,
"theoretical_loss": 4.443085223434291,
"tokens_seen": 170852352
},
{
"epoch": 0.05,
"learning_rate": 0.00047889669007021064,
"loss": 3.0001,
"theoretical_loss": 4.442866798991993,
"tokens_seen": 170917888
},
{
"epoch": 0.05,
"learning_rate": 0.0004788866599799399,
"loss": 3.5228,
"theoretical_loss": 4.442648481725577,
"tokens_seen": 170983424
},
{
"epoch": 0.05,
"learning_rate": 0.000478876629889669,
"loss": 3.5152,
"theoretical_loss": 4.442430271541404,
"tokens_seen": 171048960
},
{
"epoch": 0.05,
"learning_rate": 0.00047886659979939824,
"loss": 3.0459,
"theoretical_loss": 4.442212168345956,
"tokens_seen": 171114496
},
{
"epoch": 0.05,
"learning_rate": 0.00047885656970912736,
"loss": 3.2609,
"theoretical_loss": 4.4419941720458285,
"tokens_seen": 171180032
},
{
"epoch": 0.05,
"learning_rate": 0.0004788465396188566,
"loss": 3.1621,
"theoretical_loss": 4.441776282547736,
"tokens_seen": 171245568
},
{
"epoch": 0.05,
"learning_rate": 0.0004788365095285858,
"loss": 3.2143,
"theoretical_loss": 4.441558499758511,
"tokens_seen": 171311104
},
{
"epoch": 0.05,
"learning_rate": 0.00047882647943831496,
"loss": 3.3814,
"theoretical_loss": 4.441340823585101,
"tokens_seen": 171376640
},
{
"epoch": 0.05,
"learning_rate": 0.00047881644934804414,
"loss": 3.4502,
"theoretical_loss": 4.441123253934572,
"tokens_seen": 171442176
},
{
"epoch": 0.05,
"learning_rate": 0.0004788064192577733,
"loss": 3.2365,
"theoretical_loss": 4.440905790714105,
"tokens_seen": 171507712
},
{
"epoch": 0.05,
"learning_rate": 0.0004787963891675025,
"loss": 2.9738,
"theoretical_loss": 4.440688433830999,
"tokens_seen": 171573248
},
{
"epoch": 0.05,
"learning_rate": 0.00047878635907723174,
"loss": 3.2246,
"theoretical_loss": 4.440471183192667,
"tokens_seen": 171638784
},
{
"epoch": 0.05,
"learning_rate": 0.00047877632898696087,
"loss": 3.3308,
"theoretical_loss": 4.440254038706639,
"tokens_seen": 171704320
},
{
"epoch": 0.05,
"learning_rate": 0.0004787662988966901,
"loss": 3.1173,
"theoretical_loss": 4.440037000280561,
"tokens_seen": 171769856
},
{
"epoch": 0.05,
"learning_rate": 0.0004787562688064193,
"loss": 3.2074,
"theoretical_loss": 4.439820067822195,
"tokens_seen": 171835392
},
{
"epoch": 0.05,
"learning_rate": 0.00047874623871614847,
"loss": 3.115,
"theoretical_loss": 4.439603241239416,
"tokens_seen": 171900928
},
{
"epoch": 0.05,
"learning_rate": 0.00047873620862587765,
"loss": 3.3423,
"theoretical_loss": 4.439386520440218,
"tokens_seen": 171966464
},
{
"epoch": 0.05,
"objective/train/docs_used": 108471,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.364562749862671,
"objective/train/theoretical_loss": 4.439169905332706,
"objective/train/tokens_used": 192492000,
"theoretical_loss": 4.439169905332706,
"tokens_seen": 172032000
},
{
"epoch": 0.05,
"learning_rate": 0.00047872617853560683,
"loss": 3.1072,
"theoretical_loss": 4.439169905332706,
"tokens_seen": 172032000
},
{
"epoch": 0.05,
"learning_rate": 0.000478716148445336,
"loss": 3.1176,
"theoretical_loss": 4.438953395825102,
"tokens_seen": 172097536
},
{
"epoch": 0.05,
"learning_rate": 0.00047870611835506524,
"loss": 3.3633,
"theoretical_loss": 4.438736991825744,
"tokens_seen": 172163072
},
{
"epoch": 0.05,
"learning_rate": 0.00047869608826479437,
"loss": 3.38,
"theoretical_loss": 4.438520693243079,
"tokens_seen": 172228608
},
{
"epoch": 0.05,
"learning_rate": 0.0004786860581745236,
"loss": 3.0177,
"theoretical_loss": 4.4383044999856756,
"tokens_seen": 172294144
},
{
"epoch": 0.05,
"learning_rate": 0.00047867602808425273,
"loss": 3.1664,
"theoretical_loss": 4.438088411962211,
"tokens_seen": 172359680
},
{
"epoch": 0.05,
"learning_rate": 0.00047866599799398197,
"loss": 3.3626,
"theoretical_loss": 4.437872429081477,
"tokens_seen": 172425216
},
{
"epoch": 0.05,
"learning_rate": 0.00047865596790371115,
"loss": 3.0495,
"theoretical_loss": 4.437656551252381,
"tokens_seen": 172490752
},
{
"epoch": 0.05,
"learning_rate": 0.00047864593781344033,
"loss": 3.2428,
"theoretical_loss": 4.4374407783839445,
"tokens_seen": 172556288
},
{
"epoch": 0.05,
"learning_rate": 0.0004786359077231695,
"loss": 3.2672,
"theoretical_loss": 4.437225110385297,
"tokens_seen": 172621824
},
{
"epoch": 0.05,
"learning_rate": 0.00047862587763289875,
"loss": 3.1621,
"theoretical_loss": 4.4370095471656885,
"tokens_seen": 172687360
},
{
"epoch": 0.05,
"learning_rate": 0.0004786158475426279,
"loss": 3.3215,
"theoretical_loss": 4.436794088634477,
"tokens_seen": 172752896
},
{
"epoch": 0.05,
"learning_rate": 0.0004786058174523571,
"loss": 3.1152,
"theoretical_loss": 4.4365787347011345,
"tokens_seen": 172818432
},
{
"epoch": 0.05,
"learning_rate": 0.00047859578736208624,
"loss": 3.1985,
"theoretical_loss": 4.436363485275246,
"tokens_seen": 172883968
},
{
"epoch": 0.05,
"learning_rate": 0.00047858575727181547,
"loss": 3.3297,
"theoretical_loss": 4.436148340266508,
"tokens_seen": 172949504
},
{
"epoch": 0.05,
"learning_rate": 0.00047857572718154465,
"loss": 3.3167,
"theoretical_loss": 4.435933299584729,
"tokens_seen": 173015040
},
{
"epoch": 0.05,
"learning_rate": 0.00047856569709127383,
"loss": 3.0363,
"theoretical_loss": 4.4357183631398325,
"tokens_seen": 173080576
},
{
"epoch": 0.05,
"learning_rate": 0.000478555667001003,
"loss": 3.1073,
"theoretical_loss": 4.435503530841849,
"tokens_seen": 173146112
},
{
"epoch": 0.05,
"learning_rate": 0.0004785456369107322,
"loss": 3.1795,
"theoretical_loss": 4.435288802600926,
"tokens_seen": 173211648
},
{
"epoch": 0.05,
"learning_rate": 0.0004785356068204614,
"loss": 3.4799,
"theoretical_loss": 4.4350741783273175,
"tokens_seen": 173277184
},
{
"epoch": 0.05,
"learning_rate": 0.0004785255767301906,
"loss": 3.3446,
"theoretical_loss": 4.434859657931392,
"tokens_seen": 173342720
},
{
"epoch": 0.05,
"learning_rate": 0.00047851554663991974,
"loss": 3.1328,
"theoretical_loss": 4.434645241323629,
"tokens_seen": 173408256
},
{
"epoch": 0.05,
"learning_rate": 0.000478505516549649,
"loss": 2.8784,
"theoretical_loss": 4.434430928414617,
"tokens_seen": 173473792
},
{
"epoch": 0.05,
"learning_rate": 0.0004784954864593781,
"loss": 3.2111,
"theoretical_loss": 4.434216719115057,
"tokens_seen": 173539328
},
{
"epoch": 0.05,
"learning_rate": 0.00047848545636910734,
"loss": 3.3716,
"theoretical_loss": 4.43400261333576,
"tokens_seen": 173604864
},
{
"epoch": 0.05,
"objective/train/docs_used": 109645,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0553133487701416,
"objective/train/theoretical_loss": 4.433788610987646,
"objective/train/tokens_used": 194130400,
"theoretical_loss": 4.433788610987646,
"tokens_seen": 173670400
},
{
"epoch": 0.05,
"learning_rate": 0.0004784754262788365,
"loss": 3.2501,
"theoretical_loss": 4.433788610987646,
"tokens_seen": 173670400
},
{
"epoch": 0.05,
"learning_rate": 0.0004784653961885657,
"loss": 3.4322,
"theoretical_loss": 4.433574711981749,
"tokens_seen": 173735936
},
{
"epoch": 0.05,
"learning_rate": 0.0004784553660982949,
"loss": 3.2243,
"theoretical_loss": 4.433360916229209,
"tokens_seen": 173801472
},
{
"epoch": 0.05,
"learning_rate": 0.0004784453360080241,
"loss": 3.2569,
"theoretical_loss": 4.433147223641278,
"tokens_seen": 173867008
},
{
"epoch": 0.05,
"learning_rate": 0.00047843530591775324,
"loss": 3.2601,
"theoretical_loss": 4.432933634129318,
"tokens_seen": 173932544
},
{
"epoch": 0.05,
"learning_rate": 0.0004784252758274825,
"loss": 3.2892,
"theoretical_loss": 4.4327201476047975,
"tokens_seen": 173998080
},
{
"epoch": 0.05,
"learning_rate": 0.0004784152457372116,
"loss": 3.3204,
"theoretical_loss": 4.432506763979299,
"tokens_seen": 174063616
},
{
"epoch": 0.05,
"learning_rate": 0.00047840521564694084,
"loss": 3.3412,
"theoretical_loss": 4.432293483164512,
"tokens_seen": 174129152
},
{
"epoch": 0.05,
"learning_rate": 0.00047839518555667,
"loss": 3.0976,
"theoretical_loss": 4.432080305072233,
"tokens_seen": 174194688
},
{
"epoch": 0.05,
"learning_rate": 0.0004783851554663992,
"loss": 3.1193,
"theoretical_loss": 4.43186722961437,
"tokens_seen": 174260224
},
{
"epoch": 0.05,
"learning_rate": 0.0004783751253761284,
"loss": 3.2963,
"theoretical_loss": 4.431654256702938,
"tokens_seen": 174325760
},
{
"epoch": 0.05,
"learning_rate": 0.00047836509528585756,
"loss": 3.3016,
"theoretical_loss": 4.431441386250063,
"tokens_seen": 174391296
},
{
"epoch": 0.05,
"learning_rate": 0.00047835506519558675,
"loss": 3.2971,
"theoretical_loss": 4.4312286181679745,
"tokens_seen": 174456832
},
{
"epoch": 0.05,
"learning_rate": 0.000478345035105316,
"loss": 3.123,
"theoretical_loss": 4.431015952369016,
"tokens_seen": 174522368
},
{
"epoch": 0.05,
"learning_rate": 0.0004783350050150451,
"loss": 3.4087,
"theoretical_loss": 4.430803388765636,
"tokens_seen": 174587904
},
{
"epoch": 0.05,
"learning_rate": 0.00047832497492477434,
"loss": 3.5089,
"theoretical_loss": 4.430590927270388,
"tokens_seen": 174653440
},
{
"epoch": 0.05,
"learning_rate": 0.00047831494483450347,
"loss": 3.1789,
"theoretical_loss": 4.430378567795938,
"tokens_seen": 174718976
},
{
"epoch": 0.05,
"learning_rate": 0.0004783049147442327,
"loss": 3.3577,
"theoretical_loss": 4.430166310255057,
"tokens_seen": 174784512
},
{
"epoch": 0.05,
"learning_rate": 0.0004782948846539619,
"loss": 3.044,
"theoretical_loss": 4.429954154560624,
"tokens_seen": 174850048
},
{
"epoch": 0.05,
"learning_rate": 0.00047828485456369107,
"loss": 3.467,
"theoretical_loss": 4.429742100625624,
"tokens_seen": 174915584
},
{
"epoch": 0.05,
"learning_rate": 0.00047827482447342025,
"loss": 3.0549,
"theoretical_loss": 4.429530148363151,
"tokens_seen": 174981120
},
{
"epoch": 0.05,
"learning_rate": 0.0004782647943831495,
"loss": 3.323,
"theoretical_loss": 4.429318297686402,
"tokens_seen": 175046656
},
{
"epoch": 0.05,
"learning_rate": 0.0004782547642928786,
"loss": 3.1715,
"theoretical_loss": 4.429106548508685,
"tokens_seen": 175112192
},
{
"epoch": 0.05,
"learning_rate": 0.00047824473420260785,
"loss": 3.3082,
"theoretical_loss": 4.428894900743411,
"tokens_seen": 175177728
},
{
"epoch": 0.05,
"learning_rate": 0.000478234704112337,
"loss": 3.0858,
"theoretical_loss": 4.428683354304098,
"tokens_seen": 175243264
},
{
"epoch": 0.05,
"objective/train/docs_used": 110166,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.895432710647583,
"objective/train/theoretical_loss": 4.428471909104372,
"objective/train/tokens_used": 195768800,
"theoretical_loss": 4.428471909104372,
"tokens_seen": 175308800
},
{
"epoch": 0.05,
"learning_rate": 0.0004782246740220662,
"loss": 3.1331,
"theoretical_loss": 4.428471909104372,
"tokens_seen": 175308800
},
{
"epoch": 0.05,
"learning_rate": 0.00047821464393179544,
"loss": 3.3663,
"theoretical_loss": 4.428260565057964,
"tokens_seen": 175374336
},
{
"epoch": 0.05,
"learning_rate": 0.00047820461384152457,
"loss": 3.3574,
"theoretical_loss": 4.428049322078708,
"tokens_seen": 175439872
},
{
"epoch": 0.05,
"learning_rate": 0.0004781945837512538,
"loss": 2.9226,
"theoretical_loss": 4.427838180080547,
"tokens_seen": 175505408
},
{
"epoch": 0.05,
"learning_rate": 0.00047818455366098293,
"loss": 2.922,
"theoretical_loss": 4.4276271389775275,
"tokens_seen": 175570944
},
{
"epoch": 0.05,
"learning_rate": 0.00047817452357071217,
"loss": 3.5267,
"theoretical_loss": 4.427416198683803,
"tokens_seen": 175636480
},
{
"epoch": 0.05,
"learning_rate": 0.00047816449348044135,
"loss": 3.2307,
"theoretical_loss": 4.427205359113629,
"tokens_seen": 175702016
},
{
"epoch": 0.05,
"learning_rate": 0.00047815446339017053,
"loss": 3.3649,
"theoretical_loss": 4.42699462018137,
"tokens_seen": 175767552
},
{
"epoch": 0.05,
"learning_rate": 0.0004781444332998997,
"loss": 3.0675,
"theoretical_loss": 4.42678398180149,
"tokens_seen": 175833088
},
{
"epoch": 0.05,
"learning_rate": 0.00047813440320962895,
"loss": 3.3193,
"theoretical_loss": 4.426573443888563,
"tokens_seen": 175898624
},
{
"epoch": 0.05,
"learning_rate": 0.0004781243731193581,
"loss": 3.1539,
"theoretical_loss": 4.426363006357263,
"tokens_seen": 175964160
},
{
"epoch": 0.05,
"learning_rate": 0.0004781143430290873,
"loss": 3.1723,
"theoretical_loss": 4.426152669122374,
"tokens_seen": 176029696
},
{
"epoch": 0.05,
"learning_rate": 0.00047810431293881644,
"loss": 3.3604,
"theoretical_loss": 4.425942432098774,
"tokens_seen": 176095232
},
{
"epoch": 0.05,
"learning_rate": 0.00047809428284854567,
"loss": 3.3204,
"theoretical_loss": 4.425732295201455,
"tokens_seen": 176160768
},
{
"epoch": 0.05,
"learning_rate": 0.00047808425275827485,
"loss": 3.0399,
"theoretical_loss": 4.425522258345508,
"tokens_seen": 176226304
},
{
"epoch": 0.05,
"learning_rate": 0.00047807422266800403,
"loss": 3.1259,
"theoretical_loss": 4.425312321446127,
"tokens_seen": 176291840
},
{
"epoch": 0.05,
"learning_rate": 0.0004780641925777332,
"loss": 3.087,
"theoretical_loss": 4.425102484418613,
"tokens_seen": 176357376
},
{
"epoch": 0.05,
"learning_rate": 0.0004780541624874624,
"loss": 3.0982,
"theoretical_loss": 4.424892747178365,
"tokens_seen": 176422912
},
{
"epoch": 0.05,
"learning_rate": 0.0004780441323971916,
"loss": 3.1463,
"theoretical_loss": 4.42468310964089,
"tokens_seen": 176488448
},
{
"epoch": 0.05,
"learning_rate": 0.0004780341023069208,
"loss": 3.0898,
"theoretical_loss": 4.424473571721794,
"tokens_seen": 176553984
},
{
"epoch": 0.05,
"learning_rate": 0.00047802407221664994,
"loss": 2.6555,
"theoretical_loss": 4.42426413333679,
"tokens_seen": 176619520
},
{
"epoch": 0.05,
"learning_rate": 0.0004780140421263792,
"loss": 3.364,
"theoretical_loss": 4.424054794401689,
"tokens_seen": 176685056
},
{
"epoch": 0.05,
"learning_rate": 0.0004780040120361083,
"loss": 3.2562,
"theoretical_loss": 4.423845554832406,
"tokens_seen": 176750592
},
{
"epoch": 0.05,
"learning_rate": 0.00047799398194583754,
"loss": 3.1273,
"theoretical_loss": 4.42363641454496,
"tokens_seen": 176816128
},
{
"epoch": 0.05,
"learning_rate": 0.0004779839518555667,
"loss": 2.7713,
"theoretical_loss": 4.423427373455471,
"tokens_seen": 176881664
},
{
"epoch": 0.05,
"objective/train/docs_used": 110915,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4182002544403076,
"objective/train/theoretical_loss": 4.42321843148016,
"objective/train/tokens_used": 197407200,
"theoretical_loss": 4.42321843148016,
"tokens_seen": 176947200
},
{
"epoch": 0.05,
"learning_rate": 0.0004779739217652959,
"loss": 3.4005,
"theoretical_loss": 4.42321843148016,
"tokens_seen": 176947200
},
{
"epoch": 0.05,
"learning_rate": 0.0004779638916750251,
"loss": 3.0105,
"theoretical_loss": 4.423009588535351,
"tokens_seen": 177012736
},
{
"epoch": 0.05,
"learning_rate": 0.0004779538615847543,
"loss": 3.3242,
"theoretical_loss": 4.422800844537466,
"tokens_seen": 177078272
},
{
"epoch": 0.05,
"learning_rate": 0.00047794383149448344,
"loss": 3.23,
"theoretical_loss": 4.422592199403036,
"tokens_seen": 177143808
},
{
"epoch": 0.05,
"learning_rate": 0.0004779338014042127,
"loss": 3.1387,
"theoretical_loss": 4.422383653048685,
"tokens_seen": 177209344
},
{
"epoch": 0.05,
"learning_rate": 0.0004779237713139418,
"loss": 3.328,
"theoretical_loss": 4.422175205391145,
"tokens_seen": 177274880
},
{
"epoch": 0.05,
"learning_rate": 0.00047791374122367104,
"loss": 3.1981,
"theoretical_loss": 4.421966856347243,
"tokens_seen": 177340416
},
{
"epoch": 0.05,
"learning_rate": 0.0004779037111334002,
"loss": 3.1989,
"theoretical_loss": 4.421758605833912,
"tokens_seen": 177405952
},
{
"epoch": 0.05,
"learning_rate": 0.0004778936810431294,
"loss": 3.0267,
"theoretical_loss": 4.421550453768181,
"tokens_seen": 177471488
},
{
"epoch": 0.05,
"learning_rate": 0.0004778836509528586,
"loss": 3.4539,
"theoretical_loss": 4.421342400067183,
"tokens_seen": 177537024
},
{
"epoch": 0.05,
"learning_rate": 0.00047787362086258776,
"loss": 3.2626,
"theoretical_loss": 4.42113444464815,
"tokens_seen": 177602560
},
{
"epoch": 0.05,
"learning_rate": 0.00047786359077231695,
"loss": 3.4596,
"theoretical_loss": 4.420926587428411,
"tokens_seen": 177668096
},
{
"epoch": 0.05,
"learning_rate": 0.0004778535606820462,
"loss": 3.1495,
"theoretical_loss": 4.420718828325403,
"tokens_seen": 177733632
},
{
"epoch": 0.05,
"learning_rate": 0.0004778435305917753,
"loss": 3.2311,
"theoretical_loss": 4.420511167256656,
"tokens_seen": 177799168
},
{
"epoch": 0.05,
"learning_rate": 0.00047783350050150454,
"loss": 3.2253,
"theoretical_loss": 4.4203036041398,
"tokens_seen": 177864704
},
{
"epoch": 0.05,
"learning_rate": 0.00047782347041123367,
"loss": 3.1878,
"theoretical_loss": 4.420096138892568,
"tokens_seen": 177930240
},
{
"epoch": 0.05,
"learning_rate": 0.0004778134403209629,
"loss": 3.0134,
"theoretical_loss": 4.419888771432789,
"tokens_seen": 177995776
},
{
"epoch": 0.05,
"learning_rate": 0.0004778034102306921,
"loss": 2.9021,
"theoretical_loss": 4.419681501678395,
"tokens_seen": 178061312
},
{
"epoch": 0.05,
"learning_rate": 0.00047779338014042127,
"loss": 3.3523,
"theoretical_loss": 4.419474329547413,
"tokens_seen": 178126848
},
{
"epoch": 0.05,
"learning_rate": 0.00047778335005015045,
"loss": 2.9923,
"theoretical_loss": 4.419267254957971,
"tokens_seen": 178192384
},
{
"epoch": 0.05,
"learning_rate": 0.0004777733199598797,
"loss": 3.1929,
"theoretical_loss": 4.419060277828295,
"tokens_seen": 178257920
},
{
"epoch": 0.05,
"learning_rate": 0.0004777632898696088,
"loss": 3.4043,
"theoretical_loss": 4.41885339807671,
"tokens_seen": 178323456
},
{
"epoch": 0.05,
"learning_rate": 0.00047775325977933805,
"loss": 3.2683,
"theoretical_loss": 4.4186466156216415,
"tokens_seen": 178388992
},
{
"epoch": 0.05,
"learning_rate": 0.0004777432296890672,
"loss": 3.3439,
"theoretical_loss": 4.418439930381609,
"tokens_seen": 178454528
},
{
"epoch": 0.05,
"learning_rate": 0.0004777331995987964,
"loss": 3.3543,
"theoretical_loss": 4.418233342275233,
"tokens_seen": 178520064
},
{
"epoch": 0.05,
"objective/train/docs_used": 112413,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1684422492980957,
"objective/train/theoretical_loss": 4.418026851221231,
"objective/train/tokens_used": 199045600,
"theoretical_loss": 4.418026851221231,
"tokens_seen": 178585600
},
{
"epoch": 0.05,
"learning_rate": 0.0004777231695085256,
"loss": 3.2288,
"theoretical_loss": 4.418026851221231,
"tokens_seen": 178585600
},
{
"epoch": 0.05,
"learning_rate": 0.00047771313941825477,
"loss": 3.31,
"theoretical_loss": 4.4178204571384185,
"tokens_seen": 178651136
},
{
"epoch": 0.05,
"learning_rate": 0.00047770310932798395,
"loss": 3.2361,
"theoretical_loss": 4.41761415994571,
"tokens_seen": 178716672
},
{
"epoch": 0.05,
"learning_rate": 0.00047769307923771313,
"loss": 2.9928,
"theoretical_loss": 4.417407959562116,
"tokens_seen": 178782208
},
{
"epoch": 0.05,
"learning_rate": 0.0004776830491474423,
"loss": 3.21,
"theoretical_loss": 4.417201855906742,
"tokens_seen": 178847744
},
{
"epoch": 0.05,
"learning_rate": 0.00047767301905717155,
"loss": 3.3201,
"theoretical_loss": 4.416995848898797,
"tokens_seen": 178913280
},
{
"epoch": 0.05,
"learning_rate": 0.0004776629889669007,
"loss": 3.2827,
"theoretical_loss": 4.4167899384575815,
"tokens_seen": 178978816
},
{
"epoch": 0.05,
"learning_rate": 0.0004776529588766299,
"loss": 3.3293,
"theoretical_loss": 4.416584124502495,
"tokens_seen": 179044352
},
{
"epoch": 0.05,
"learning_rate": 0.00047764292878635904,
"loss": 3.3263,
"theoretical_loss": 4.416378406953033,
"tokens_seen": 179109888
},
{
"epoch": 0.05,
"learning_rate": 0.0004776328986960883,
"loss": 3.3501,
"theoretical_loss": 4.41617278572879,
"tokens_seen": 179175424
},
{
"epoch": 0.05,
"learning_rate": 0.00047762286860581746,
"loss": 2.9366,
"theoretical_loss": 4.4159672607494524,
"tokens_seen": 179240960
},
{
"epoch": 0.05,
"learning_rate": 0.00047761283851554664,
"loss": 3.0323,
"theoretical_loss": 4.415761831934808,
"tokens_seen": 179306496
},
{
"epoch": 0.05,
"learning_rate": 0.0004776028084252758,
"loss": 2.9697,
"theoretical_loss": 4.415556499204737,
"tokens_seen": 179372032
},
{
"epoch": 0.05,
"learning_rate": 0.00047759277833500505,
"loss": 3.0708,
"theoretical_loss": 4.415351262479216,
"tokens_seen": 179437568
},
{
"epoch": 0.05,
"learning_rate": 0.0004775827482447342,
"loss": 3.3015,
"theoretical_loss": 4.415146121678321,
"tokens_seen": 179503104
},
{
"epoch": 0.05,
"learning_rate": 0.0004775727181544634,
"loss": 3.1837,
"theoretical_loss": 4.414941076722219,
"tokens_seen": 179568640
},
{
"epoch": 0.05,
"learning_rate": 0.00047756268806419254,
"loss": 3.2274,
"theoretical_loss": 4.4147361275311745,
"tokens_seen": 179634176
},
{
"epoch": 0.05,
"learning_rate": 0.0004775526579739218,
"loss": 3.3152,
"theoretical_loss": 4.414531274025548,
"tokens_seen": 179699712
},
{
"epoch": 0.05,
"learning_rate": 0.00047754262788365096,
"loss": 3.1656,
"theoretical_loss": 4.414326516125795,
"tokens_seen": 179765248
},
{
"epoch": 0.05,
"learning_rate": 0.00047753259779338014,
"loss": 3.0617,
"theoretical_loss": 4.414121853752466,
"tokens_seen": 179830784
},
{
"epoch": 0.05,
"learning_rate": 0.0004775225677031093,
"loss": 3.0591,
"theoretical_loss": 4.413917286826205,
"tokens_seen": 179896320
},
{
"epoch": 0.05,
"learning_rate": 0.0004775125376128385,
"loss": 3.4019,
"theoretical_loss": 4.413712815267752,
"tokens_seen": 179961856
},
{
"epoch": 0.05,
"learning_rate": 0.0004775025075225677,
"loss": 3.4241,
"theoretical_loss": 4.413508438997944,
"tokens_seen": 180027392
},
{
"epoch": 0.05,
"learning_rate": 0.0004774924774322969,
"loss": 2.8506,
"theoretical_loss": 4.4133041579377075,
"tokens_seen": 180092928
},
{
"epoch": 0.05,
"learning_rate": 0.00047748244734202605,
"loss": 3.3765,
"theoretical_loss": 4.413099972008068,
"tokens_seen": 180158464
},
{
"epoch": 0.05,
"objective/train/docs_used": 112982,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9989802837371826,
"objective/train/theoretical_loss": 4.412895881130142,
"objective/train/tokens_used": 200684000,
"theoretical_loss": 4.412895881130142,
"tokens_seen": 180224000
},
{
"epoch": 0.05,
"learning_rate": 0.0004774724172517553,
"loss": 3.1511,
"theoretical_loss": 4.412895881130142,
"tokens_seen": 180224000
},
{
"epoch": 0.05,
"learning_rate": 0.0004774623871614845,
"loss": 3.2793,
"theoretical_loss": 4.412691885225141,
"tokens_seen": 180289536
},
{
"epoch": 0.05,
"learning_rate": 0.00047745235707121364,
"loss": 3.1662,
"theoretical_loss": 4.412487984214373,
"tokens_seen": 180355072
},
{
"epoch": 0.05,
"learning_rate": 0.0004774423269809429,
"loss": 2.9821,
"theoretical_loss": 4.412284178019235,
"tokens_seen": 180420608
},
{
"epoch": 0.05,
"learning_rate": 0.000477432296890672,
"loss": 3.2647,
"theoretical_loss": 4.412080466561221,
"tokens_seen": 180486144
},
{
"epoch": 0.05,
"learning_rate": 0.00047742226680040124,
"loss": 3.041,
"theoretical_loss": 4.411876849761917,
"tokens_seen": 180551680
},
{
"epoch": 0.05,
"learning_rate": 0.0004774122367101304,
"loss": 3.229,
"theoretical_loss": 4.411673327543005,
"tokens_seen": 180617216
},
{
"epoch": 0.05,
"learning_rate": 0.0004774022066198596,
"loss": 2.9898,
"theoretical_loss": 4.4114698998262565,
"tokens_seen": 180682752
},
{
"epoch": 0.05,
"learning_rate": 0.0004773921765295888,
"loss": 3.4276,
"theoretical_loss": 4.411266566533539,
"tokens_seen": 180748288
},
{
"epoch": 0.05,
"learning_rate": 0.00047738214643931797,
"loss": 3.1725,
"theoretical_loss": 4.41106332758681,
"tokens_seen": 180813824
},
{
"epoch": 0.05,
"learning_rate": 0.00047737211634904715,
"loss": 3.1015,
"theoretical_loss": 4.41086018290812,
"tokens_seen": 180879360
},
{
"epoch": 0.05,
"learning_rate": 0.0004773620862587764,
"loss": 3.1636,
"theoretical_loss": 4.410657132419617,
"tokens_seen": 180944896
},
{
"epoch": 0.05,
"learning_rate": 0.0004773520561685055,
"loss": 3.2115,
"theoretical_loss": 4.410454176043537,
"tokens_seen": 181010432
},
{
"epoch": 0.05,
"learning_rate": 0.00047734202607823474,
"loss": 3.1097,
"theoretical_loss": 4.410251313702208,
"tokens_seen": 181075968
},
{
"epoch": 0.05,
"learning_rate": 0.00047733199598796387,
"loss": 3.322,
"theoretical_loss": 4.410048545318052,
"tokens_seen": 181141504
},
{
"epoch": 0.05,
"learning_rate": 0.0004773219658976931,
"loss": 3.1962,
"theoretical_loss": 4.409845870813582,
"tokens_seen": 181207040
},
{
"epoch": 0.05,
"learning_rate": 0.0004773119358074223,
"loss": 3.3378,
"theoretical_loss": 4.409643290111404,
"tokens_seen": 181272576
},
{
"epoch": 0.05,
"learning_rate": 0.00047730190571715147,
"loss": 3.0122,
"theoretical_loss": 4.409440803134215,
"tokens_seen": 181338112
},
{
"epoch": 0.05,
"learning_rate": 0.00047729187562688065,
"loss": 3.3294,
"theoretical_loss": 4.409238409804804,
"tokens_seen": 181403648
},
{
"epoch": 0.05,
"learning_rate": 0.0004772818455366099,
"loss": 3.3616,
"theoretical_loss": 4.409036110046051,
"tokens_seen": 181469184
},
{
"epoch": 0.06,
"learning_rate": 0.000477271815446339,
"loss": 2.9033,
"theoretical_loss": 4.408833903780926,
"tokens_seen": 181534720
},
{
"epoch": 0.06,
"learning_rate": 0.00047726178535606825,
"loss": 3.1769,
"theoretical_loss": 4.408631790932494,
"tokens_seen": 181600256
},
{
"epoch": 0.06,
"learning_rate": 0.0004772517552657974,
"loss": 3.2627,
"theoretical_loss": 4.408429771423909,
"tokens_seen": 181665792
},
{
"epoch": 0.06,
"learning_rate": 0.0004772417251755266,
"loss": 3.2006,
"theoretical_loss": 4.408227845178414,
"tokens_seen": 181731328
},
{
"epoch": 0.06,
"learning_rate": 0.0004772316950852558,
"loss": 3.3776,
"theoretical_loss": 4.408026012119344,
"tokens_seen": 181796864
},
{
"epoch": 0.06,
"objective/train/docs_used": 114216,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.247993230819702,
"objective/train/theoretical_loss": 4.407824272170128,
"objective/train/tokens_used": 202322400,
"theoretical_loss": 4.407824272170128,
"tokens_seen": 181862400
},
{
"epoch": 0.06,
"learning_rate": 0.00047722166499498497,
"loss": 3.1762,
"theoretical_loss": 4.407824272170128,
"tokens_seen": 181862400
},
{
"epoch": 0.06,
"learning_rate": 0.00047721163490471415,
"loss": 3.2095,
"theoretical_loss": 4.407622625254279,
"tokens_seen": 181927936
},
{
"epoch": 0.06,
"learning_rate": 0.00047720160481444333,
"loss": 3.2396,
"theoretical_loss": 4.407421071295406,
"tokens_seen": 181993472
},
{
"epoch": 0.06,
"learning_rate": 0.0004771915747241725,
"loss": 3.2236,
"theoretical_loss": 4.407219610217206,
"tokens_seen": 182059008
},
{
"epoch": 0.06,
"learning_rate": 0.00047718154463390175,
"loss": 3.2076,
"theoretical_loss": 4.407018241943467,
"tokens_seen": 182124544
},
{
"epoch": 0.06,
"learning_rate": 0.0004771715145436309,
"loss": 3.3943,
"theoretical_loss": 4.406816966398064,
"tokens_seen": 182190080
},
{
"epoch": 0.06,
"learning_rate": 0.0004771614844533601,
"loss": 3.0633,
"theoretical_loss": 4.406615783504965,
"tokens_seen": 182255616
},
{
"epoch": 0.06,
"learning_rate": 0.00047715145436308924,
"loss": 3.3245,
"theoretical_loss": 4.4064146931882275,
"tokens_seen": 182321152
},
{
"epoch": 0.06,
"learning_rate": 0.0004771414242728185,
"loss": 3.2816,
"theoretical_loss": 4.406213695371996,
"tokens_seen": 182386688
},
{
"epoch": 0.06,
"learning_rate": 0.00047713139418254766,
"loss": 3.0674,
"theoretical_loss": 4.406012789980506,
"tokens_seen": 182452224
},
{
"epoch": 0.06,
"learning_rate": 0.00047712136409227684,
"loss": 3.2889,
"theoretical_loss": 4.405811976938084,
"tokens_seen": 182517760
},
{
"epoch": 0.06,
"learning_rate": 0.000477111334002006,
"loss": 2.9432,
"theoretical_loss": 4.405611256169143,
"tokens_seen": 182583296
},
{
"epoch": 0.06,
"learning_rate": 0.00047710130391173525,
"loss": 3.1513,
"theoretical_loss": 4.405410627598185,
"tokens_seen": 182648832
},
{
"epoch": 0.06,
"learning_rate": 0.0004770912738214644,
"loss": 3.2537,
"theoretical_loss": 4.405210091149802,
"tokens_seen": 182714368
},
{
"epoch": 0.06,
"learning_rate": 0.0004770812437311936,
"loss": 3.2721,
"theoretical_loss": 4.405009646748674,
"tokens_seen": 182779904
},
{
"epoch": 0.06,
"learning_rate": 0.00047707121364092274,
"loss": 3.2594,
"theoretical_loss": 4.404809294319572,
"tokens_seen": 182845440
},
{
"epoch": 0.06,
"learning_rate": 0.000477061183550652,
"loss": 3.0547,
"theoretical_loss": 4.40460903378735,
"tokens_seen": 182910976
},
{
"epoch": 0.06,
"learning_rate": 0.00047705115346038116,
"loss": 3.238,
"theoretical_loss": 4.404408865076955,
"tokens_seen": 182976512
},
{
"epoch": 0.06,
"learning_rate": 0.00047704112337011034,
"loss": 3.0375,
"theoretical_loss": 4.404208788113422,
"tokens_seen": 183042048
},
{
"epoch": 0.06,
"learning_rate": 0.0004770310932798395,
"loss": 2.9986,
"theoretical_loss": 4.404008802821871,
"tokens_seen": 183107584
},
{
"epoch": 0.06,
"learning_rate": 0.0004770210631895687,
"loss": 3.4418,
"theoretical_loss": 4.4038089091275125,
"tokens_seen": 183173120
},
{
"epoch": 0.06,
"learning_rate": 0.0004770110330992979,
"loss": 3.1524,
"theoretical_loss": 4.403609106955645,
"tokens_seen": 183238656
},
{
"epoch": 0.06,
"learning_rate": 0.0004770010030090271,
"loss": 3.4428,
"theoretical_loss": 4.403409396231651,
"tokens_seen": 183304192
},
{
"epoch": 0.06,
"learning_rate": 0.00047699097291875625,
"loss": 3.097,
"theoretical_loss": 4.403209776881004,
"tokens_seen": 183369728
},
{
"epoch": 0.06,
"learning_rate": 0.0004769809428284855,
"loss": 3.1748,
"theoretical_loss": 4.403010248829265,
"tokens_seen": 183435264
},
{
"epoch": 0.06,
"objective/train/docs_used": 114775,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7343997955322266,
"objective/train/theoretical_loss": 4.4028108120020795,
"objective/train/tokens_used": 203960800,
"theoretical_loss": 4.4028108120020795,
"tokens_seen": 183500800
},
{
"epoch": 0.06,
"learning_rate": 0.00047697091273821466,
"loss": 3.364,
"theoretical_loss": 4.4028108120020795,
"tokens_seen": 183500800
},
{
"epoch": 0.06,
"learning_rate": 0.00047696088264794384,
"loss": 3.5379,
"theoretical_loss": 4.402611466325182,
"tokens_seen": 183566336
},
{
"epoch": 0.06,
"learning_rate": 0.000476950852557673,
"loss": 3.1437,
"theoretical_loss": 4.4024122117243945,
"tokens_seen": 183631872
},
{
"epoch": 0.06,
"learning_rate": 0.0004769408224674022,
"loss": 3.0713,
"theoretical_loss": 4.402213048125624,
"tokens_seen": 183697408
},
{
"epoch": 0.06,
"learning_rate": 0.0004769307923771314,
"loss": 3.1808,
"theoretical_loss": 4.4020139754548655,
"tokens_seen": 183762944
},
{
"epoch": 0.06,
"learning_rate": 0.0004769207622868606,
"loss": 3.5212,
"theoretical_loss": 4.401814993638199,
"tokens_seen": 183828480
},
{
"epoch": 0.06,
"learning_rate": 0.00047691073219658975,
"loss": 3.1228,
"theoretical_loss": 4.4016161026017935,
"tokens_seen": 183894016
},
{
"epoch": 0.06,
"learning_rate": 0.000476900702106319,
"loss": 3.2918,
"theoretical_loss": 4.401417302271902,
"tokens_seen": 183959552
},
{
"epoch": 0.06,
"learning_rate": 0.0004768906720160481,
"loss": 3.3633,
"theoretical_loss": 4.401218592574865,
"tokens_seen": 184025088
},
{
"epoch": 0.06,
"learning_rate": 0.00047688064192577735,
"loss": 3.3739,
"theoretical_loss": 4.401019973437108,
"tokens_seen": 184090624
},
{
"epoch": 0.06,
"learning_rate": 0.00047687061183550653,
"loss": 3.3149,
"theoretical_loss": 4.400821444785143,
"tokens_seen": 184156160
},
{
"epoch": 0.06,
"learning_rate": 0.0004768605817452357,
"loss": 3.3195,
"theoretical_loss": 4.400623006545567,
"tokens_seen": 184221696
},
{
"epoch": 0.06,
"learning_rate": 0.0004768505516549649,
"loss": 3.2683,
"theoretical_loss": 4.400424658645065,
"tokens_seen": 184287232
},
{
"epoch": 0.06,
"learning_rate": 0.00047684052156469407,
"loss": 2.8183,
"theoretical_loss": 4.400226401010404,
"tokens_seen": 184352768
},
{
"epoch": 0.06,
"learning_rate": 0.00047683049147442325,
"loss": 3.4246,
"theoretical_loss": 4.40002823356844,
"tokens_seen": 184418304
},
{
"epoch": 0.06,
"learning_rate": 0.0004768204613841525,
"loss": 3.3673,
"theoretical_loss": 4.39983015624611,
"tokens_seen": 184483840
},
{
"epoch": 0.06,
"learning_rate": 0.0004768104312938816,
"loss": 3.029,
"theoretical_loss": 4.39963216897044,
"tokens_seen": 184549376
},
{
"epoch": 0.06,
"learning_rate": 0.00047680040120361085,
"loss": 3.2208,
"theoretical_loss": 4.3994342716685395,
"tokens_seen": 184614912
},
{
"epoch": 0.06,
"learning_rate": 0.00047679037111334003,
"loss": 3.184,
"theoretical_loss": 4.399236464267602,
"tokens_seen": 184680448
},
{
"epoch": 0.06,
"learning_rate": 0.0004767803410230692,
"loss": 3.2012,
"theoretical_loss": 4.399038746694908,
"tokens_seen": 184745984
},
{
"epoch": 0.06,
"learning_rate": 0.0004767703109327984,
"loss": 3.4012,
"theoretical_loss": 4.398841118877819,
"tokens_seen": 184811520
},
{
"epoch": 0.06,
"learning_rate": 0.0004767602808425276,
"loss": 3.0397,
"theoretical_loss": 4.398643580743785,
"tokens_seen": 184877056
},
{
"epoch": 0.06,
"learning_rate": 0.00047675025075225676,
"loss": 3.3388,
"theoretical_loss": 4.398446132220338,
"tokens_seen": 184942592
},
{
"epoch": 0.06,
"learning_rate": 0.000476740220661986,
"loss": 3.0237,
"theoretical_loss": 4.3982487732350934,
"tokens_seen": 185008128
},
{
"epoch": 0.06,
"learning_rate": 0.0004767301905717151,
"loss": 3.1185,
"theoretical_loss": 4.398051503715753,
"tokens_seen": 185073664
},
{
"epoch": 0.06,
"objective/train/docs_used": 115939,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.676588296890259,
"objective/train/theoretical_loss": 4.397854323590102,
"objective/train/tokens_used": 205599200,
"theoretical_loss": 4.397854323590102,
"tokens_seen": 185139200
},
{
"epoch": 0.06,
"learning_rate": 0.00047672016048144435,
"loss": 3.2948,
"theoretical_loss": 4.397854323590102,
"tokens_seen": 185139200
},
{
"epoch": 0.06,
"learning_rate": 0.00047671013039117353,
"loss": 3.2664,
"theoretical_loss": 4.397657232786008,
"tokens_seen": 185204736
},
{
"epoch": 0.06,
"learning_rate": 0.0004767001003009027,
"loss": 2.8907,
"theoretical_loss": 4.397460231231424,
"tokens_seen": 185270272
},
{
"epoch": 0.06,
"learning_rate": 0.00047669007021063195,
"loss": 3.2102,
"theoretical_loss": 4.397263318854384,
"tokens_seen": 185335808
},
{
"epoch": 0.06,
"learning_rate": 0.0004766800401203611,
"loss": 3.0437,
"theoretical_loss": 4.39706649558301,
"tokens_seen": 185401344
},
{
"epoch": 0.06,
"learning_rate": 0.0004766700100300903,
"loss": 3.2762,
"theoretical_loss": 4.396869761345503,
"tokens_seen": 185466880
},
{
"epoch": 0.06,
"learning_rate": 0.00047665997993981944,
"loss": 3.2785,
"theoretical_loss": 4.396673116070147,
"tokens_seen": 185532416
},
{
"epoch": 0.06,
"learning_rate": 0.0004766499498495487,
"loss": 3.169,
"theoretical_loss": 4.396476559685315,
"tokens_seen": 185597952
},
{
"epoch": 0.06,
"learning_rate": 0.00047663991975927786,
"loss": 2.8861,
"theoretical_loss": 4.396280092119455,
"tokens_seen": 185663488
},
{
"epoch": 0.06,
"learning_rate": 0.00047662988966900704,
"loss": 3.6408,
"theoretical_loss": 4.3960837133011035,
"tokens_seen": 185729024
},
{
"epoch": 0.06,
"learning_rate": 0.0004766198595787362,
"loss": 3.29,
"theoretical_loss": 4.395887423158877,
"tokens_seen": 185794560
},
{
"epoch": 0.06,
"learning_rate": 0.00047660982948846545,
"loss": 3.2726,
"theoretical_loss": 4.395691221621476,
"tokens_seen": 185860096
},
{
"epoch": 0.06,
"learning_rate": 0.0004765997993981946,
"loss": 3.2641,
"theoretical_loss": 4.395495108617682,
"tokens_seen": 185925632
},
{
"epoch": 0.06,
"learning_rate": 0.0004765897693079238,
"loss": 3.3623,
"theoretical_loss": 4.39529908407636,
"tokens_seen": 185991168
},
{
"epoch": 0.06,
"learning_rate": 0.00047657973921765294,
"loss": 3.0954,
"theoretical_loss": 4.3951031479264575,
"tokens_seen": 186056704
},
{
"epoch": 0.06,
"learning_rate": 0.0004765697091273822,
"loss": 3.4236,
"theoretical_loss": 4.394907300097002,
"tokens_seen": 186122240
},
{
"epoch": 0.06,
"learning_rate": 0.00047655967903711136,
"loss": 3.328,
"theoretical_loss": 4.394711540517106,
"tokens_seen": 186187776
},
{
"epoch": 0.06,
"learning_rate": 0.00047654964894684054,
"loss": 3.3462,
"theoretical_loss": 4.39451586911596,
"tokens_seen": 186253312
},
{
"epoch": 0.06,
"learning_rate": 0.0004765396188565697,
"loss": 3.3489,
"theoretical_loss": 4.39432028582284,
"tokens_seen": 186318848
},
{
"epoch": 0.06,
"learning_rate": 0.0004765295887662989,
"loss": 3.2442,
"theoretical_loss": 4.394124790567101,
"tokens_seen": 186384384
},
{
"epoch": 0.06,
"learning_rate": 0.0004765195586760281,
"loss": 3.6337,
"theoretical_loss": 4.3939293832781825,
"tokens_seen": 186449920
},
{
"epoch": 0.06,
"learning_rate": 0.0004765095285857573,
"loss": 3.2979,
"theoretical_loss": 4.393734063885599,
"tokens_seen": 186515456
},
{
"epoch": 0.06,
"learning_rate": 0.00047649949849548645,
"loss": 3.1169,
"theoretical_loss": 4.3935388323189555,
"tokens_seen": 186580992
},
{
"epoch": 0.06,
"learning_rate": 0.0004764894684052157,
"loss": 3.2682,
"theoretical_loss": 4.39334368850793,
"tokens_seen": 186646528
},
{
"epoch": 0.06,
"learning_rate": 0.00047647943831494486,
"loss": 3.1131,
"theoretical_loss": 4.3931486323822835,
"tokens_seen": 186712064
},
{
"epoch": 0.06,
"objective/train/docs_used": 116562,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.736578941345215,
"objective/train/theoretical_loss": 4.392953663871862,
"objective/train/tokens_used": 207237600,
"theoretical_loss": 4.392953663871862,
"tokens_seen": 186777600
},
{
"epoch": 0.06,
"learning_rate": 0.00047646940822467404,
"loss": 3.3584,
"theoretical_loss": 4.392953663871862,
"tokens_seen": 186777600
},
{
"epoch": 0.06,
"learning_rate": 0.0004764593781344032,
"loss": 3.162,
"theoretical_loss": 4.392758782906586,
"tokens_seen": 186843136
},
{
"epoch": 0.06,
"learning_rate": 0.0004764493480441324,
"loss": 3.062,
"theoretical_loss": 4.392563989416462,
"tokens_seen": 186908672
},
{
"epoch": 0.06,
"learning_rate": 0.0004764393179538616,
"loss": 3.478,
"theoretical_loss": 4.392369283331574,
"tokens_seen": 186974208
},
{
"epoch": 0.06,
"learning_rate": 0.0004764292878635908,
"loss": 3.2393,
"theoretical_loss": 4.392174664582085,
"tokens_seen": 187039744
},
{
"epoch": 0.06,
"learning_rate": 0.00047641925777331995,
"loss": 3.1537,
"theoretical_loss": 4.391980133098244,
"tokens_seen": 187105280
},
{
"epoch": 0.06,
"learning_rate": 0.0004764092276830492,
"loss": 3.2428,
"theoretical_loss": 4.391785688810373,
"tokens_seen": 187170816
},
{
"epoch": 0.06,
"learning_rate": 0.0004763991975927783,
"loss": 3.2938,
"theoretical_loss": 4.391591331648879,
"tokens_seen": 187236352
},
{
"epoch": 0.06,
"learning_rate": 0.00047638916750250755,
"loss": 3.4311,
"theoretical_loss": 4.391397061544247,
"tokens_seen": 187301888
},
{
"epoch": 0.06,
"learning_rate": 0.00047637913741223673,
"loss": 3.0581,
"theoretical_loss": 4.391202878427042,
"tokens_seen": 187367424
},
{
"epoch": 0.06,
"learning_rate": 0.0004763691073219659,
"loss": 3.5038,
"theoretical_loss": 4.3910087822279085,
"tokens_seen": 187432960
},
{
"epoch": 0.06,
"learning_rate": 0.0004763590772316951,
"loss": 3.2592,
"theoretical_loss": 4.390814772877571,
"tokens_seen": 187498496
},
{
"epoch": 0.06,
"learning_rate": 0.00047634904714142427,
"loss": 3.0196,
"theoretical_loss": 4.390620850306832,
"tokens_seen": 187564032
},
{
"epoch": 0.06,
"learning_rate": 0.00047633901705115345,
"loss": 3.2439,
"theoretical_loss": 4.390427014446575,
"tokens_seen": 187629568
},
{
"epoch": 0.06,
"learning_rate": 0.0004763289869608827,
"loss": 3.3844,
"theoretical_loss": 4.390233265227764,
"tokens_seen": 187695104
},
{
"epoch": 0.06,
"learning_rate": 0.0004763189568706118,
"loss": 3.1343,
"theoretical_loss": 4.390039602581437,
"tokens_seen": 187760640
},
{
"epoch": 0.06,
"learning_rate": 0.00047630892678034105,
"loss": 2.9602,
"theoretical_loss": 4.389846026438715,
"tokens_seen": 187826176
},
{
"epoch": 0.06,
"learning_rate": 0.00047629889669007023,
"loss": 3.2549,
"theoretical_loss": 4.3896525367307975,
"tokens_seen": 187891712
},
{
"epoch": 0.06,
"learning_rate": 0.0004762888665997994,
"loss": 2.9531,
"theoretical_loss": 4.389459133388962,
"tokens_seen": 187957248
},
{
"epoch": 0.06,
"learning_rate": 0.0004762788365095286,
"loss": 3.1851,
"theoretical_loss": 4.3892658163445635,
"tokens_seen": 188022784
},
{
"epoch": 0.06,
"learning_rate": 0.0004762688064192578,
"loss": 3.2693,
"theoretical_loss": 4.389072585529037,
"tokens_seen": 188088320
},
{
"epoch": 0.06,
"learning_rate": 0.00047625877632898696,
"loss": 3.1727,
"theoretical_loss": 4.388879440873897,
"tokens_seen": 188153856
},
{
"epoch": 0.06,
"learning_rate": 0.0004762487462387162,
"loss": 3.3616,
"theoretical_loss": 4.388686382310732,
"tokens_seen": 188219392
},
{
"epoch": 0.06,
"learning_rate": 0.0004762387161484453,
"loss": 3.176,
"theoretical_loss": 4.388493409771213,
"tokens_seen": 188284928
},
{
"epoch": 0.06,
"learning_rate": 0.00047622868605817455,
"loss": 3.135,
"theoretical_loss": 4.388300523187087,
"tokens_seen": 188350464
},
{
"epoch": 0.06,
"objective/train/docs_used": 117858,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.447967290878296,
"objective/train/theoretical_loss": 4.3881077224901786,
"objective/train/tokens_used": 208876000,
"theoretical_loss": 4.3881077224901786,
"tokens_seen": 188416000
},
{
"epoch": 0.06,
"learning_rate": 0.0004762186559679037,
"loss": 3.4454,
"theoretical_loss": 4.3881077224901786,
"tokens_seen": 188416000
},
{
"epoch": 0.06,
"learning_rate": 0.0004762086258776329,
"loss": 3.1037,
"theoretical_loss": 4.38791500761239,
"tokens_seen": 188481536
},
{
"epoch": 0.06,
"learning_rate": 0.0004761985957873621,
"loss": 3.3634,
"theoretical_loss": 4.387722378485703,
"tokens_seen": 188547072
},
{
"epoch": 0.06,
"learning_rate": 0.0004761885656970913,
"loss": 3.1468,
"theoretical_loss": 4.3875298350421765,
"tokens_seen": 188612608
},
{
"epoch": 0.06,
"learning_rate": 0.00047617853560682046,
"loss": 3.3193,
"theoretical_loss": 4.387337377213943,
"tokens_seen": 188678144
},
{
"epoch": 0.06,
"learning_rate": 0.00047616850551654964,
"loss": 3.2997,
"theoretical_loss": 4.387145004933218,
"tokens_seen": 188743680
},
{
"epoch": 0.06,
"learning_rate": 0.0004761584754262788,
"loss": 3.0325,
"theoretical_loss": 4.38695271813229,
"tokens_seen": 188809216
},
{
"epoch": 0.06,
"learning_rate": 0.00047614844533600806,
"loss": 3.0921,
"theoretical_loss": 4.386760516743526,
"tokens_seen": 188874752
},
{
"epoch": 0.06,
"learning_rate": 0.0004761384152457372,
"loss": 3.3926,
"theoretical_loss": 4.38656840069937,
"tokens_seen": 188940288
},
{
"epoch": 0.06,
"learning_rate": 0.0004761283851554664,
"loss": 3.1398,
"theoretical_loss": 4.386376369932344,
"tokens_seen": 189005824
},
{
"epoch": 0.06,
"learning_rate": 0.0004761183550651956,
"loss": 3.0158,
"theoretical_loss": 4.386184424375044,
"tokens_seen": 189071360
},
{
"epoch": 0.06,
"learning_rate": 0.0004761083249749248,
"loss": 3.4574,
"theoretical_loss": 4.385992563960145,
"tokens_seen": 189136896
},
{
"epoch": 0.06,
"learning_rate": 0.00047609829488465396,
"loss": 3.1658,
"theoretical_loss": 4.385800788620397,
"tokens_seen": 189202432
},
{
"epoch": 0.06,
"learning_rate": 0.00047608826479438314,
"loss": 3.1425,
"theoretical_loss": 4.385609098288628,
"tokens_seen": 189267968
},
{
"epoch": 0.06,
"learning_rate": 0.0004760782347041123,
"loss": 3.239,
"theoretical_loss": 4.385417492897741,
"tokens_seen": 189333504
},
{
"epoch": 0.06,
"learning_rate": 0.00047606820461384156,
"loss": 3.3172,
"theoretical_loss": 4.385225972380715,
"tokens_seen": 189399040
},
{
"epoch": 0.06,
"learning_rate": 0.0004760581745235707,
"loss": 3.2791,
"theoretical_loss": 4.385034536670606,
"tokens_seen": 189464576
},
{
"epoch": 0.06,
"learning_rate": 0.0004760481444332999,
"loss": 3.33,
"theoretical_loss": 4.384843185700544,
"tokens_seen": 189530112
},
{
"epoch": 0.06,
"learning_rate": 0.00047603811434302905,
"loss": 3.3761,
"theoretical_loss": 4.384651919403739,
"tokens_seen": 189595648
},
{
"epoch": 0.06,
"learning_rate": 0.0004760280842527583,
"loss": 3.2921,
"theoretical_loss": 4.384460737713471,
"tokens_seen": 189661184
},
{
"epoch": 0.06,
"learning_rate": 0.00047601805416248746,
"loss": 3.4782,
"theoretical_loss": 4.384269640563101,
"tokens_seen": 189726720
},
{
"epoch": 0.06,
"learning_rate": 0.00047600802407221665,
"loss": 3.1232,
"theoretical_loss": 4.384078627886062,
"tokens_seen": 189792256
},
{
"epoch": 0.06,
"learning_rate": 0.0004759979939819458,
"loss": 3.2606,
"theoretical_loss": 4.383887699615863,
"tokens_seen": 189857792
},
{
"epoch": 0.06,
"learning_rate": 0.00047598796389167506,
"loss": 3.0839,
"theoretical_loss": 4.38369685568609,
"tokens_seen": 189923328
},
{
"epoch": 0.06,
"learning_rate": 0.0004759779338014042,
"loss": 2.9678,
"theoretical_loss": 4.383506096030401,
"tokens_seen": 189988864
},
{
"epoch": 0.06,
"objective/train/docs_used": 118609,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9445102214813232,
"objective/train/theoretical_loss": 4.383315420582533,
"objective/train/tokens_used": 210514400,
"theoretical_loss": 4.383315420582533,
"tokens_seen": 190054400
},
{
"epoch": 0.06,
"learning_rate": 0.0004759679037111334,
"loss": 3.2935,
"theoretical_loss": 4.383315420582533,
"tokens_seen": 190054400
},
{
"epoch": 0.06,
"learning_rate": 0.0004759578736208626,
"loss": 3.2186,
"theoretical_loss": 4.383124829276294,
"tokens_seen": 190119936
},
{
"epoch": 0.06,
"learning_rate": 0.0004759478435305918,
"loss": 3.2593,
"theoretical_loss": 4.38293432204557,
"tokens_seen": 190185472
},
{
"epoch": 0.06,
"learning_rate": 0.000475937813440321,
"loss": 3.1491,
"theoretical_loss": 4.382743898824321,
"tokens_seen": 190251008
},
{
"epoch": 0.06,
"learning_rate": 0.00047592778335005015,
"loss": 3.1876,
"theoretical_loss": 4.3825535595465785,
"tokens_seen": 190316544
},
{
"epoch": 0.06,
"learning_rate": 0.0004759177532597794,
"loss": 3.3843,
"theoretical_loss": 4.382363304146453,
"tokens_seen": 190382080
},
{
"epoch": 0.06,
"learning_rate": 0.0004759077231695085,
"loss": 3.2096,
"theoretical_loss": 4.382173132558126,
"tokens_seen": 190447616
},
{
"epoch": 0.06,
"learning_rate": 0.00047589769307923775,
"loss": 3.0433,
"theoretical_loss": 4.381983044715856,
"tokens_seen": 190513152
},
{
"epoch": 0.06,
"learning_rate": 0.00047588766298896693,
"loss": 3.2308,
"theoretical_loss": 4.381793040553973,
"tokens_seen": 190578688
},
{
"epoch": 0.06,
"learning_rate": 0.0004758776328986961,
"loss": 3.478,
"theoretical_loss": 4.381603120006883,
"tokens_seen": 190644224
},
{
"epoch": 0.06,
"learning_rate": 0.0004758676028084253,
"loss": 3.4116,
"theoretical_loss": 4.381413283009065,
"tokens_seen": 190709760
},
{
"epoch": 0.06,
"learning_rate": 0.00047585757271815447,
"loss": 3.0949,
"theoretical_loss": 4.381223529495073,
"tokens_seen": 190775296
},
{
"epoch": 0.06,
"learning_rate": 0.00047584754262788365,
"loss": 3.1533,
"theoretical_loss": 4.381033859399532,
"tokens_seen": 190840832
},
{
"epoch": 0.06,
"learning_rate": 0.0004758375125376129,
"loss": 3.2304,
"theoretical_loss": 4.380844272657145,
"tokens_seen": 190906368
},
{
"epoch": 0.06,
"learning_rate": 0.000475827482447342,
"loss": 3.3131,
"theoretical_loss": 4.380654769202683,
"tokens_seen": 190971904
},
{
"epoch": 0.06,
"learning_rate": 0.00047581745235707125,
"loss": 2.9842,
"theoretical_loss": 4.380465348970995,
"tokens_seen": 191037440
},
{
"epoch": 0.06,
"learning_rate": 0.00047580742226680043,
"loss": 3.1313,
"theoretical_loss": 4.380276011897003,
"tokens_seen": 191102976
},
{
"epoch": 0.06,
"learning_rate": 0.0004757973921765296,
"loss": 3.2165,
"theoretical_loss": 4.380086757915698,
"tokens_seen": 191168512
},
{
"epoch": 0.06,
"learning_rate": 0.0004757873620862588,
"loss": 3.2402,
"theoretical_loss": 4.379897586962148,
"tokens_seen": 191234048
},
{
"epoch": 0.06,
"learning_rate": 0.000475777331995988,
"loss": 3.1014,
"theoretical_loss": 4.379708498971494,
"tokens_seen": 191299584
},
{
"epoch": 0.06,
"learning_rate": 0.00047576730190571716,
"loss": 3.2106,
"theoretical_loss": 4.379519493878948,
"tokens_seen": 191365120
},
{
"epoch": 0.06,
"learning_rate": 0.0004757572718154464,
"loss": 3.2548,
"theoretical_loss": 4.379330571619795,
"tokens_seen": 191430656
},
{
"epoch": 0.06,
"learning_rate": 0.0004757472417251755,
"loss": 3.3656,
"theoretical_loss": 4.379141732129394,
"tokens_seen": 191496192
},
{
"epoch": 0.06,
"learning_rate": 0.00047573721163490475,
"loss": 3.4178,
"theoretical_loss": 4.378952975343175,
"tokens_seen": 191561728
},
{
"epoch": 0.06,
"learning_rate": 0.0004757271815446339,
"loss": 3.1327,
"theoretical_loss": 4.378764301196642,
"tokens_seen": 191627264
},
{
"epoch": 0.06,
"objective/train/docs_used": 119339,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7788050174713135,
"objective/train/theoretical_loss": 4.37857570962537,
"objective/train/tokens_used": 212152800,
"theoretical_loss": 4.37857570962537,
"tokens_seen": 191692800
},
{
"epoch": 0.06,
"learning_rate": 0.0004757171514543631,
"loss": 3.0324,
"theoretical_loss": 4.37857570962537,
"tokens_seen": 191692800
},
{
"epoch": 0.06,
"learning_rate": 0.0004757071213640923,
"loss": 3.1974,
"theoretical_loss": 4.378387200565006,
"tokens_seen": 191758336
},
{
"epoch": 0.06,
"learning_rate": 0.0004756970912738215,
"loss": 3.4038,
"theoretical_loss": 4.378198773951272,
"tokens_seen": 191823872
},
{
"epoch": 0.06,
"learning_rate": 0.00047568706118355066,
"loss": 3.0394,
"theoretical_loss": 4.378010429719957,
"tokens_seen": 191889408
},
{
"epoch": 0.06,
"learning_rate": 0.00047567703109327984,
"loss": 2.9882,
"theoretical_loss": 4.377822167806928,
"tokens_seen": 191954944
},
{
"epoch": 0.06,
"learning_rate": 0.000475667001003009,
"loss": 3.0199,
"theoretical_loss": 4.377633988148117,
"tokens_seen": 192020480
},
{
"epoch": 0.06,
"learning_rate": 0.00047565697091273826,
"loss": 3.2158,
"theoretical_loss": 4.377445890679534,
"tokens_seen": 192086016
},
{
"epoch": 0.06,
"learning_rate": 0.0004756469408224674,
"loss": 3.0202,
"theoretical_loss": 4.377257875337257,
"tokens_seen": 192151552
},
{
"epoch": 0.06,
"learning_rate": 0.0004756369107321966,
"loss": 3.1161,
"theoretical_loss": 4.377069942057436,
"tokens_seen": 192217088
},
{
"epoch": 0.06,
"learning_rate": 0.0004756268806419258,
"loss": 3.2261,
"theoretical_loss": 4.376882090776293,
"tokens_seen": 192282624
},
{
"epoch": 0.06,
"learning_rate": 0.000475616850551655,
"loss": 3.3294,
"theoretical_loss": 4.376694321430121,
"tokens_seen": 192348160
},
{
"epoch": 0.06,
"learning_rate": 0.00047560682046138416,
"loss": 3.2889,
"theoretical_loss": 4.376506633955286,
"tokens_seen": 192413696
},
{
"epoch": 0.06,
"learning_rate": 0.00047559679037111334,
"loss": 3.3172,
"theoretical_loss": 4.376319028288219,
"tokens_seen": 192479232
},
{
"epoch": 0.06,
"learning_rate": 0.0004755867602808425,
"loss": 3.426,
"theoretical_loss": 4.37613150436543,
"tokens_seen": 192544768
},
{
"epoch": 0.06,
"learning_rate": 0.00047557673019057176,
"loss": 3.0088,
"theoretical_loss": 4.375944062123496,
"tokens_seen": 192610304
},
{
"epoch": 0.06,
"learning_rate": 0.0004755667001003009,
"loss": 3.4675,
"theoretical_loss": 4.375756701499063,
"tokens_seen": 192675840
},
{
"epoch": 0.06,
"learning_rate": 0.0004755566700100301,
"loss": 3.2118,
"theoretical_loss": 4.3755694224288515,
"tokens_seen": 192741376
},
{
"epoch": 0.06,
"learning_rate": 0.00047554663991975925,
"loss": 3.1241,
"theoretical_loss": 4.375382224849648,
"tokens_seen": 192806912
},
{
"epoch": 0.06,
"learning_rate": 0.0004755366098294885,
"loss": 3.2397,
"theoretical_loss": 4.375195108698316,
"tokens_seen": 192872448
},
{
"epoch": 0.06,
"learning_rate": 0.00047552657973921766,
"loss": 3.6427,
"theoretical_loss": 4.375008073911781,
"tokens_seen": 192937984
},
{
"epoch": 0.06,
"learning_rate": 0.00047551654964894685,
"loss": 3.1663,
"theoretical_loss": 4.374821120427047,
"tokens_seen": 193003520
},
{
"epoch": 0.06,
"learning_rate": 0.00047550651955867603,
"loss": 3.0961,
"theoretical_loss": 4.374634248181182,
"tokens_seen": 193069056
},
{
"epoch": 0.06,
"learning_rate": 0.00047549648946840526,
"loss": 3.5514,
"theoretical_loss": 4.3744474571113265,
"tokens_seen": 193134592
},
{
"epoch": 0.06,
"learning_rate": 0.0004754864593781344,
"loss": 3.0639,
"theoretical_loss": 4.374260747154692,
"tokens_seen": 193200128
},
{
"epoch": 0.06,
"learning_rate": 0.0004754764292878636,
"loss": 3.178,
"theoretical_loss": 4.374074118248559,
"tokens_seen": 193265664
},
{
"epoch": 0.06,
"objective/train/docs_used": 120055,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.8682849407196045,
"objective/train/theoretical_loss": 4.373887570330275,
"objective/train/tokens_used": 213791200,
"theoretical_loss": 4.373887570330275,
"tokens_seen": 193331200
},
{
"epoch": 0.06,
"learning_rate": 0.00047546639919759275,
"loss": 3.3739,
"theoretical_loss": 4.373887570330275,
"tokens_seen": 193331200
},
{
"epoch": 0.06,
"learning_rate": 0.000475456369107322,
"loss": 3.3962,
"theoretical_loss": 4.373701103337263,
"tokens_seen": 193396736
},
{
"epoch": 0.06,
"learning_rate": 0.00047544633901705117,
"loss": 3.1677,
"theoretical_loss": 4.373514717207009,
"tokens_seen": 193462272
},
{
"epoch": 0.06,
"learning_rate": 0.00047543630892678035,
"loss": 3.2994,
"theoretical_loss": 4.373328411877073,
"tokens_seen": 193527808
},
{
"epoch": 0.06,
"learning_rate": 0.00047542627883650953,
"loss": 3.3462,
"theoretical_loss": 4.373142187285083,
"tokens_seen": 193593344
},
{
"epoch": 0.06,
"learning_rate": 0.0004754162487462387,
"loss": 3.3214,
"theoretical_loss": 4.372956043368736,
"tokens_seen": 193658880
},
{
"epoch": 0.06,
"learning_rate": 0.0004754062186559679,
"loss": 3.333,
"theoretical_loss": 4.372769980065797,
"tokens_seen": 193724416
},
{
"epoch": 0.06,
"learning_rate": 0.00047539618856569713,
"loss": 3.3139,
"theoretical_loss": 4.372583997314104,
"tokens_seen": 193789952
},
{
"epoch": 0.06,
"learning_rate": 0.00047538615847542625,
"loss": 3.2031,
"theoretical_loss": 4.372398095051559,
"tokens_seen": 193855488
},
{
"epoch": 0.06,
"learning_rate": 0.0004753761283851555,
"loss": 3.3055,
"theoretical_loss": 4.372212273216136,
"tokens_seen": 193921024
},
{
"epoch": 0.06,
"learning_rate": 0.0004753660982948846,
"loss": 3.3132,
"theoretical_loss": 4.372026531745877,
"tokens_seen": 193986560
},
{
"epoch": 0.06,
"learning_rate": 0.00047535606820461385,
"loss": 3.2799,
"theoretical_loss": 4.371840870578891,
"tokens_seen": 194052096
},
{
"epoch": 0.06,
"learning_rate": 0.00047534603811434303,
"loss": 3.0888,
"theoretical_loss": 4.37165528965336,
"tokens_seen": 194117632
},
{
"epoch": 0.06,
"learning_rate": 0.0004753360080240722,
"loss": 3.441,
"theoretical_loss": 4.371469788907529,
"tokens_seen": 194183168
},
{
"epoch": 0.06,
"learning_rate": 0.0004753259779338014,
"loss": 3.1359,
"theoretical_loss": 4.371284368279714,
"tokens_seen": 194248704
},
{
"epoch": 0.06,
"learning_rate": 0.00047531594784353063,
"loss": 3.282,
"theoretical_loss": 4.3710990277083,
"tokens_seen": 194314240
},
{
"epoch": 0.06,
"learning_rate": 0.00047530591775325976,
"loss": 3.1125,
"theoretical_loss": 4.3709137671317375,
"tokens_seen": 194379776
},
{
"epoch": 0.06,
"learning_rate": 0.000475295887662989,
"loss": 3.1613,
"theoretical_loss": 4.37072858648855,
"tokens_seen": 194445312
},
{
"epoch": 0.06,
"learning_rate": 0.0004752858575727181,
"loss": 3.267,
"theoretical_loss": 4.370543485717322,
"tokens_seen": 194510848
},
{
"epoch": 0.06,
"learning_rate": 0.00047527582748244736,
"loss": 3.2022,
"theoretical_loss": 4.370358464756713,
"tokens_seen": 194576384
},
{
"epoch": 0.06,
"learning_rate": 0.00047526579739217654,
"loss": 3.2746,
"theoretical_loss": 4.370173523545443,
"tokens_seen": 194641920
},
{
"epoch": 0.06,
"learning_rate": 0.0004752557673019057,
"loss": 3.3689,
"theoretical_loss": 4.3699886620223065,
"tokens_seen": 194707456
},
{
"epoch": 0.06,
"learning_rate": 0.0004752457372116349,
"loss": 3.4246,
"theoretical_loss": 4.369803880126162,
"tokens_seen": 194772992
},
{
"epoch": 0.06,
"learning_rate": 0.0004752357071213641,
"loss": 3.0546,
"theoretical_loss": 4.3696191777959354,
"tokens_seen": 194838528
},
{
"epoch": 0.06,
"learning_rate": 0.00047522567703109326,
"loss": 3.254,
"theoretical_loss": 4.369434554970621,
"tokens_seen": 194904064
},
{
"epoch": 0.06,
"objective/train/docs_used": 121265,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3207590579986572,
"objective/train/theoretical_loss": 4.369250011589279,
"objective/train/tokens_used": 215429600,
"theoretical_loss": 4.369250011589279,
"tokens_seen": 194969600
},
{
"epoch": 0.06,
"learning_rate": 0.0004752156469408225,
"loss": 3.1958,
"theoretical_loss": 4.369250011589279,
"tokens_seen": 194969600
},
{
"epoch": 0.06,
"learning_rate": 0.0004752056168505517,
"loss": 3.1648,
"theoretical_loss": 4.369065547591038,
"tokens_seen": 195035136
},
{
"epoch": 0.06,
"learning_rate": 0.00047519558676028086,
"loss": 3.2667,
"theoretical_loss": 4.368881162915095,
"tokens_seen": 195100672
},
{
"epoch": 0.06,
"learning_rate": 0.00047518555667001004,
"loss": 3.3682,
"theoretical_loss": 4.36869685750071,
"tokens_seen": 195166208
},
{
"epoch": 0.06,
"learning_rate": 0.0004751755265797392,
"loss": 3.4655,
"theoretical_loss": 4.3685126312872145,
"tokens_seen": 195231744
},
{
"epoch": 0.06,
"learning_rate": 0.00047516549648946846,
"loss": 3.0632,
"theoretical_loss": 4.368328484214002,
"tokens_seen": 195297280
},
{
"epoch": 0.06,
"learning_rate": 0.0004751554663991976,
"loss": 3.2254,
"theoretical_loss": 4.368144416220538,
"tokens_seen": 195362816
},
{
"epoch": 0.06,
"learning_rate": 0.0004751454363089268,
"loss": 3.26,
"theoretical_loss": 4.3679604272463495,
"tokens_seen": 195428352
},
{
"epoch": 0.06,
"learning_rate": 0.000475135406218656,
"loss": 3.19,
"theoretical_loss": 4.367776517231033,
"tokens_seen": 195493888
},
{
"epoch": 0.06,
"learning_rate": 0.0004751253761283852,
"loss": 3.282,
"theoretical_loss": 4.367592686114252,
"tokens_seen": 195559424
},
{
"epoch": 0.06,
"learning_rate": 0.00047511534603811436,
"loss": 3.1387,
"theoretical_loss": 4.367408933835733,
"tokens_seen": 195624960
},
{
"epoch": 0.06,
"learning_rate": 0.00047510531594784354,
"loss": 3.3084,
"theoretical_loss": 4.367225260335272,
"tokens_seen": 195690496
},
{
"epoch": 0.06,
"learning_rate": 0.0004750952858575727,
"loss": 3.409,
"theoretical_loss": 4.36704166555273,
"tokens_seen": 195756032
},
{
"epoch": 0.06,
"learning_rate": 0.00047508525576730196,
"loss": 3.3344,
"theoretical_loss": 4.366858149428032,
"tokens_seen": 195821568
},
{
"epoch": 0.06,
"learning_rate": 0.0004750752256770311,
"loss": 3.0945,
"theoretical_loss": 4.366674711901173,
"tokens_seen": 195887104
},
{
"epoch": 0.06,
"learning_rate": 0.0004750651955867603,
"loss": 3.1295,
"theoretical_loss": 4.366491352912211,
"tokens_seen": 195952640
},
{
"epoch": 0.06,
"learning_rate": 0.00047505516549648945,
"loss": 3.1857,
"theoretical_loss": 4.366308072401271,
"tokens_seen": 196018176
},
{
"epoch": 0.06,
"learning_rate": 0.0004750451354062187,
"loss": 3.2574,
"theoretical_loss": 4.366124870308541,
"tokens_seen": 196083712
},
{
"epoch": 0.06,
"learning_rate": 0.00047503510531594787,
"loss": 3.1826,
"theoretical_loss": 4.365941746574278,
"tokens_seen": 196149248
},
{
"epoch": 0.06,
"learning_rate": 0.00047502507522567705,
"loss": 3.1245,
"theoretical_loss": 4.3657587011388035,
"tokens_seen": 196214784
},
{
"epoch": 0.06,
"learning_rate": 0.00047501504513540623,
"loss": 3.1448,
"theoretical_loss": 4.365575733942503,
"tokens_seen": 196280320
},
{
"epoch": 0.06,
"learning_rate": 0.00047500501504513546,
"loss": 3.1648,
"theoretical_loss": 4.365392844925829,
"tokens_seen": 196345856
},
{
"epoch": 0.06,
"learning_rate": 0.0004749949849548646,
"loss": 3.2024,
"theoretical_loss": 4.365210034029298,
"tokens_seen": 196411392
},
{
"epoch": 0.06,
"learning_rate": 0.0004749849548645938,
"loss": 3.0918,
"theoretical_loss": 4.365027301193491,
"tokens_seen": 196476928
},
{
"epoch": 0.06,
"learning_rate": 0.00047497492477432295,
"loss": 2.8404,
"theoretical_loss": 4.364844646359056,
"tokens_seen": 196542464
},
{
"epoch": 0.06,
"objective/train/docs_used": 121711,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.7159526348114014,
"objective/train/theoretical_loss": 4.364662069466704,
"objective/train/tokens_used": 217068000,
"theoretical_loss": 4.364662069466704,
"tokens_seen": 196608000
},
{
"epoch": 0.06,
"learning_rate": 0.0004749648946840522,
"loss": 3.3992,
"theoretical_loss": 4.364662069466704,
"tokens_seen": 196608000
},
{
"epoch": 0.06,
"learning_rate": 0.00047495486459378137,
"loss": 3.3754,
"theoretical_loss": 4.364479570457213,
"tokens_seen": 196673536
},
{
"epoch": 0.06,
"learning_rate": 0.00047494483450351055,
"loss": 3.1078,
"theoretical_loss": 4.364297149271423,
"tokens_seen": 196739072
},
{
"epoch": 0.06,
"learning_rate": 0.00047493480441323973,
"loss": 3.2097,
"theoretical_loss": 4.3641148058502415,
"tokens_seen": 196804608
},
{
"epoch": 0.06,
"learning_rate": 0.0004749247743229689,
"loss": 3.0253,
"theoretical_loss": 4.363932540134638,
"tokens_seen": 196870144
},
{
"epoch": 0.06,
"learning_rate": 0.0004749147442326981,
"loss": 3.4151,
"theoretical_loss": 4.363750352065647,
"tokens_seen": 196935680
},
{
"epoch": 0.06,
"learning_rate": 0.00047490471414242733,
"loss": 3.1998,
"theoretical_loss": 4.363568241584368,
"tokens_seen": 197001216
},
{
"epoch": 0.06,
"learning_rate": 0.00047489468405215646,
"loss": 3.2949,
"theoretical_loss": 4.363386208631966,
"tokens_seen": 197066752
},
{
"epoch": 0.06,
"learning_rate": 0.0004748846539618857,
"loss": 3.153,
"theoretical_loss": 4.363204253149667,
"tokens_seen": 197132288
},
{
"epoch": 0.06,
"learning_rate": 0.0004748746238716148,
"loss": 3.3637,
"theoretical_loss": 4.3630223750787644,
"tokens_seen": 197197824
},
{
"epoch": 0.06,
"learning_rate": 0.00047486459378134405,
"loss": 3.0955,
"theoretical_loss": 4.362840574360612,
"tokens_seen": 197263360
},
{
"epoch": 0.06,
"learning_rate": 0.00047485456369107323,
"loss": 3.2066,
"theoretical_loss": 4.362658850936631,
"tokens_seen": 197328896
},
{
"epoch": 0.06,
"learning_rate": 0.0004748445336008024,
"loss": 3.065,
"theoretical_loss": 4.362477204748305,
"tokens_seen": 197394432
},
{
"epoch": 0.06,
"learning_rate": 0.0004748345035105316,
"loss": 3.1457,
"theoretical_loss": 4.362295635737179,
"tokens_seen": 197459968
},
{
"epoch": 0.06,
"learning_rate": 0.00047482447342026083,
"loss": 3.2047,
"theoretical_loss": 4.362114143844867,
"tokens_seen": 197525504
},
{
"epoch": 0.06,
"learning_rate": 0.00047481444332998996,
"loss": 3.2647,
"theoretical_loss": 4.3619327290130405,
"tokens_seen": 197591040
},
{
"epoch": 0.06,
"learning_rate": 0.0004748044132397192,
"loss": 3.3005,
"theoretical_loss": 4.3617513911834385,
"tokens_seen": 197656576
},
{
"epoch": 0.06,
"learning_rate": 0.0004747943831494483,
"loss": 3.1384,
"theoretical_loss": 4.361570130297863,
"tokens_seen": 197722112
},
{
"epoch": 0.06,
"learning_rate": 0.00047478435305917756,
"loss": 3.3974,
"theoretical_loss": 4.3613889462981765,
"tokens_seen": 197787648
},
{
"epoch": 0.06,
"learning_rate": 0.00047477432296890674,
"loss": 3.1913,
"theoretical_loss": 4.361207839126308,
"tokens_seen": 197853184
},
{
"epoch": 0.06,
"learning_rate": 0.0004747642928786359,
"loss": 3.3772,
"theoretical_loss": 4.361026808724247,
"tokens_seen": 197918720
},
{
"epoch": 0.06,
"learning_rate": 0.0004747542627883651,
"loss": 3.3336,
"theoretical_loss": 4.360845855034049,
"tokens_seen": 197984256
},
{
"epoch": 0.06,
"learning_rate": 0.0004747442326980943,
"loss": 3.1099,
"theoretical_loss": 4.360664977997828,
"tokens_seen": 198049792
},
{
"epoch": 0.06,
"learning_rate": 0.00047473420260782346,
"loss": 3.1798,
"theoretical_loss": 4.360484177557766,
"tokens_seen": 198115328
},
{
"epoch": 0.06,
"learning_rate": 0.0004747241725175527,
"loss": 3.0723,
"theoretical_loss": 4.360303453656103,
"tokens_seen": 198180864
},
{
"epoch": 0.06,
"objective/train/docs_used": 122248,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.521057605743408,
"objective/train/theoretical_loss": 4.360122806235145,
"objective/train/tokens_used": 218706400,
"theoretical_loss": 4.360122806235145,
"tokens_seen": 198246400
},
{
"epoch": 0.06,
"learning_rate": 0.0004747141424272818,
"loss": 3.3945,
"theoretical_loss": 4.360122806235145,
"tokens_seen": 198246400
},
{
"epoch": 0.06,
"learning_rate": 0.00047470411233701106,
"loss": 3.0591,
"theoretical_loss": 4.359942235237257,
"tokens_seen": 198311936
},
{
"epoch": 0.06,
"learning_rate": 0.0004746940822467402,
"loss": 2.9216,
"theoretical_loss": 4.359761740604871,
"tokens_seen": 198377472
},
{
"epoch": 0.06,
"learning_rate": 0.0004746840521564694,
"loss": 3.3207,
"theoretical_loss": 4.359581322280479,
"tokens_seen": 198443008
},
{
"epoch": 0.06,
"learning_rate": 0.0004746740220661986,
"loss": 3.3274,
"theoretical_loss": 4.359400980206634,
"tokens_seen": 198508544
},
{
"epoch": 0.06,
"learning_rate": 0.0004746639919759278,
"loss": 3.3515,
"theoretical_loss": 4.359220714325954,
"tokens_seen": 198574080
},
{
"epoch": 0.06,
"learning_rate": 0.00047465396188565696,
"loss": 3.272,
"theoretical_loss": 4.359040524581116,
"tokens_seen": 198639616
},
{
"epoch": 0.06,
"learning_rate": 0.0004746439317953862,
"loss": 3.213,
"theoretical_loss": 4.358860410914861,
"tokens_seen": 198705152
},
{
"epoch": 0.06,
"learning_rate": 0.0004746339017051153,
"loss": 3.1872,
"theoretical_loss": 4.358680373269993,
"tokens_seen": 198770688
},
{
"epoch": 0.06,
"learning_rate": 0.00047462387161484456,
"loss": 2.9749,
"theoretical_loss": 4.358500411589375,
"tokens_seen": 198836224
},
{
"epoch": 0.06,
"learning_rate": 0.0004746138415245737,
"loss": 3.2143,
"theoretical_loss": 4.358320525815934,
"tokens_seen": 198901760
},
{
"epoch": 0.06,
"learning_rate": 0.0004746038114343029,
"loss": 3.2556,
"theoretical_loss": 4.358140715892658,
"tokens_seen": 198967296
},
{
"epoch": 0.06,
"learning_rate": 0.0004745937813440321,
"loss": 3.2029,
"theoretical_loss": 4.357960981762595,
"tokens_seen": 199032832
},
{
"epoch": 0.06,
"learning_rate": 0.0004745837512537613,
"loss": 3.1239,
"theoretical_loss": 4.357781323368857,
"tokens_seen": 199098368
},
{
"epoch": 0.06,
"learning_rate": 0.00047457372116349047,
"loss": 2.9612,
"theoretical_loss": 4.357601740654617,
"tokens_seen": 199163904
},
{
"epoch": 0.06,
"learning_rate": 0.00047456369107321965,
"loss": 3.0499,
"theoretical_loss": 4.357422233563106,
"tokens_seen": 199229440
},
{
"epoch": 0.06,
"learning_rate": 0.00047455366098294883,
"loss": 3.0168,
"theoretical_loss": 4.357242802037623,
"tokens_seen": 199294976
},
{
"epoch": 0.06,
"learning_rate": 0.00047454363089267807,
"loss": 3.276,
"theoretical_loss": 4.35706344602152,
"tokens_seen": 199360512
},
{
"epoch": 0.06,
"learning_rate": 0.0004745336008024072,
"loss": 3.2373,
"theoretical_loss": 4.356884165458217,
"tokens_seen": 199426048
},
{
"epoch": 0.06,
"learning_rate": 0.00047452357071213643,
"loss": 3.1589,
"theoretical_loss": 4.356704960291191,
"tokens_seen": 199491584
},
{
"epoch": 0.06,
"learning_rate": 0.00047451354062186555,
"loss": 3.0203,
"theoretical_loss": 4.35652583046398,
"tokens_seen": 199557120
},
{
"epoch": 0.06,
"learning_rate": 0.0004745035105315948,
"loss": 3.136,
"theoretical_loss": 4.356346775920185,
"tokens_seen": 199622656
},
{
"epoch": 0.06,
"learning_rate": 0.00047449348044132397,
"loss": 3.2817,
"theoretical_loss": 4.356167796603467,
"tokens_seen": 199688192
},
{
"epoch": 0.06,
"learning_rate": 0.00047448345035105315,
"loss": 3.282,
"theoretical_loss": 4.355988892457546,
"tokens_seen": 199753728
},
{
"epoch": 0.06,
"learning_rate": 0.0004744734202607824,
"loss": 3.1863,
"theoretical_loss": 4.355810063426204,
"tokens_seen": 199819264
},
{
"epoch": 0.06,
"objective/train/docs_used": 123553,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0212910175323486,
"objective/train/theoretical_loss": 4.355631309453283,
"objective/train/tokens_used": 220344800,
"theoretical_loss": 4.355631309453283,
"tokens_seen": 199884800
},
{
"epoch": 0.06,
"learning_rate": 0.00047446339017051157,
"loss": 3.171,
"theoretical_loss": 4.355631309453283,
"tokens_seen": 199884800
},
{
"epoch": 0.06,
"learning_rate": 0.00047445336008024075,
"loss": 3.0987,
"theoretical_loss": 4.355452630482685,
"tokens_seen": 199950336
},
{
"epoch": 0.06,
"learning_rate": 0.00047444332998996993,
"loss": 3.1614,
"theoretical_loss": 4.355274026458375,
"tokens_seen": 200015872
},
{
"epoch": 0.06,
"learning_rate": 0.0004744332998996991,
"loss": 3.3795,
"theoretical_loss": 4.355095497324373,
"tokens_seen": 200081408
},
{
"epoch": 0.06,
"learning_rate": 0.0004744232698094283,
"loss": 3.0216,
"theoretical_loss": 4.354917043024765,
"tokens_seen": 200146944
},
{
"epoch": 0.06,
"learning_rate": 0.00047441323971915753,
"loss": 3.3482,
"theoretical_loss": 4.354738663503692,
"tokens_seen": 200212480
},
{
"epoch": 0.06,
"learning_rate": 0.00047440320962888666,
"loss": 3.1463,
"theoretical_loss": 4.354560358705358,
"tokens_seen": 200278016
},
{
"epoch": 0.06,
"learning_rate": 0.0004743931795386159,
"loss": 2.9351,
"theoretical_loss": 4.354382128574027,
"tokens_seen": 200343552
},
{
"epoch": 0.06,
"learning_rate": 0.000474383149448345,
"loss": 3.3023,
"theoretical_loss": 4.35420397305402,
"tokens_seen": 200409088
},
{
"epoch": 0.06,
"learning_rate": 0.00047437311935807425,
"loss": 3.2039,
"theoretical_loss": 4.35402589208972,
"tokens_seen": 200474624
},
{
"epoch": 0.06,
"learning_rate": 0.00047436308926780343,
"loss": 3.1976,
"theoretical_loss": 4.353847885625571,
"tokens_seen": 200540160
},
{
"epoch": 0.06,
"learning_rate": 0.0004743530591775326,
"loss": 3.0562,
"theoretical_loss": 4.353669953606072,
"tokens_seen": 200605696
},
{
"epoch": 0.06,
"learning_rate": 0.0004743430290872618,
"loss": 3.4207,
"theoretical_loss": 4.353492095975787,
"tokens_seen": 200671232
},
{
"epoch": 0.06,
"learning_rate": 0.00047433299899699103,
"loss": 3.0709,
"theoretical_loss": 4.353314312679333,
"tokens_seen": 200736768
},
{
"epoch": 0.06,
"learning_rate": 0.00047432296890672016,
"loss": 3.1709,
"theoretical_loss": 4.353136603661392,
"tokens_seen": 200802304
},
{
"epoch": 0.06,
"learning_rate": 0.0004743129388164494,
"loss": 3.2795,
"theoretical_loss": 4.352958968866704,
"tokens_seen": 200867840
},
{
"epoch": 0.06,
"learning_rate": 0.0004743029087261785,
"loss": 3.3119,
"theoretical_loss": 4.352781408240065,
"tokens_seen": 200933376
},
{
"epoch": 0.06,
"learning_rate": 0.00047429287863590776,
"loss": 3.0623,
"theoretical_loss": 4.352603921726334,
"tokens_seen": 200998912
},
{
"epoch": 0.06,
"learning_rate": 0.00047428284854563694,
"loss": 3.4305,
"theoretical_loss": 4.352426509270425,
"tokens_seen": 201064448
},
{
"epoch": 0.06,
"learning_rate": 0.0004742728184553661,
"loss": 3.1922,
"theoretical_loss": 4.352249170817315,
"tokens_seen": 201129984
},
{
"epoch": 0.06,
"learning_rate": 0.0004742627883650953,
"loss": 3.3135,
"theoretical_loss": 4.352071906312037,
"tokens_seen": 201195520
},
{
"epoch": 0.06,
"learning_rate": 0.0004742527582748245,
"loss": 3.1888,
"theoretical_loss": 4.351894715699684,
"tokens_seen": 201261056
},
{
"epoch": 0.06,
"learning_rate": 0.00047424272818455366,
"loss": 3.4327,
"theoretical_loss": 4.351717598925406,
"tokens_seen": 201326592
},
{
"epoch": 0.06,
"learning_rate": 0.0004742326980942829,
"loss": 3.3805,
"theoretical_loss": 4.351540555934414,
"tokens_seen": 201392128
},
{
"epoch": 0.06,
"learning_rate": 0.000474222668004012,
"loss": 3.5804,
"theoretical_loss": 4.351363586671976,
"tokens_seen": 201457664
},
{
"epoch": 0.06,
"objective/train/docs_used": 124253,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.035176992416382,
"objective/train/theoretical_loss": 4.351186691083417,
"objective/train/tokens_used": 221983200,
"theoretical_loss": 4.351186691083417,
"tokens_seen": 201523200
},
{
"epoch": 0.06,
"learning_rate": 0.00047421263791374126,
"loss": 3.126,
"theoretical_loss": 4.351186691083417,
"tokens_seen": 201523200
},
{
"epoch": 0.06,
"learning_rate": 0.0004742026078234704,
"loss": 3.1474,
"theoretical_loss": 4.351009869114124,
"tokens_seen": 201588736
},
{
"epoch": 0.06,
"learning_rate": 0.0004741925777331996,
"loss": 3.2053,
"theoretical_loss": 4.350833120709539,
"tokens_seen": 201654272
},
{
"epoch": 0.06,
"learning_rate": 0.0004741825476429288,
"loss": 3.2585,
"theoretical_loss": 4.350656445815164,
"tokens_seen": 201719808
},
{
"epoch": 0.06,
"learning_rate": 0.000474172517552658,
"loss": 3.1713,
"theoretical_loss": 4.350479844376557,
"tokens_seen": 201785344
},
{
"epoch": 0.06,
"learning_rate": 0.00047416248746238716,
"loss": 3.3766,
"theoretical_loss": 4.350303316339337,
"tokens_seen": 201850880
},
{
"epoch": 0.06,
"learning_rate": 0.0004741524573721164,
"loss": 3.2092,
"theoretical_loss": 4.350126861649178,
"tokens_seen": 201916416
},
{
"epoch": 0.06,
"learning_rate": 0.0004741424272818455,
"loss": 3.1106,
"theoretical_loss": 4.349950480251813,
"tokens_seen": 201981952
},
{
"epoch": 0.06,
"learning_rate": 0.00047413239719157476,
"loss": 3.0031,
"theoretical_loss": 4.349774172093033,
"tokens_seen": 202047488
},
{
"epoch": 0.06,
"learning_rate": 0.0004741223671013039,
"loss": 3.4637,
"theoretical_loss": 4.349597937118687,
"tokens_seen": 202113024
},
{
"epoch": 0.06,
"learning_rate": 0.0004741123370110331,
"loss": 3.169,
"theoretical_loss": 4.3494217752746795,
"tokens_seen": 202178560
},
{
"epoch": 0.06,
"learning_rate": 0.0004741023069207623,
"loss": 3.2752,
"theoretical_loss": 4.349245686506976,
"tokens_seen": 202244096
},
{
"epoch": 0.06,
"learning_rate": 0.0004740922768304915,
"loss": 3.3951,
"theoretical_loss": 4.349069670761597,
"tokens_seen": 202309632
},
{
"epoch": 0.06,
"learning_rate": 0.00047408224674022067,
"loss": 3.12,
"theoretical_loss": 4.348893727984619,
"tokens_seen": 202375168
},
{
"epoch": 0.06,
"learning_rate": 0.00047407221664994985,
"loss": 3.0903,
"theoretical_loss": 4.348717858122178,
"tokens_seen": 202440704
},
{
"epoch": 0.06,
"learning_rate": 0.00047406218655967903,
"loss": 3.1313,
"theoretical_loss": 4.348542061120469,
"tokens_seen": 202506240
},
{
"epoch": 0.06,
"learning_rate": 0.00047405215646940827,
"loss": 3.2476,
"theoretical_loss": 4.348366336925739,
"tokens_seen": 202571776
},
{
"epoch": 0.06,
"learning_rate": 0.0004740421263791374,
"loss": 3.0185,
"theoretical_loss": 4.3481906854842975,
"tokens_seen": 202637312
},
{
"epoch": 0.06,
"learning_rate": 0.00047403209628886663,
"loss": 3.1723,
"theoretical_loss": 4.348015106742507,
"tokens_seen": 202702848
},
{
"epoch": 0.06,
"learning_rate": 0.00047402206619859575,
"loss": 3.1502,
"theoretical_loss": 4.347839600646786,
"tokens_seen": 202768384
},
{
"epoch": 0.06,
"learning_rate": 0.000474012036108325,
"loss": 3.3803,
"theoretical_loss": 4.347664167143615,
"tokens_seen": 202833920
},
{
"epoch": 0.06,
"learning_rate": 0.00047400200601805417,
"loss": 3.1256,
"theoretical_loss": 4.347488806179528,
"tokens_seen": 202899456
},
{
"epoch": 0.06,
"learning_rate": 0.00047399197592778335,
"loss": 3.2037,
"theoretical_loss": 4.347313517701114,
"tokens_seen": 202964992
},
{
"epoch": 0.06,
"learning_rate": 0.00047398194583751253,
"loss": 3.1465,
"theoretical_loss": 4.347138301655021,
"tokens_seen": 203030528
},
{
"epoch": 0.06,
"learning_rate": 0.00047397191574724177,
"loss": 3.4567,
"theoretical_loss": 4.346963157987954,
"tokens_seen": 203096064
},
{
"epoch": 0.06,
"objective/train/docs_used": 125838,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.456173896789551,
"objective/train/theoretical_loss": 4.346788086646671,
"objective/train/tokens_used": 223621600,
"theoretical_loss": 4.346788086646671,
"tokens_seen": 203161600
},
{
"epoch": 0.06,
"learning_rate": 0.0004739618856569709,
"loss": 3.2604,
"theoretical_loss": 4.346788086646671,
"tokens_seen": 203161600
},
{
"epoch": 0.06,
"learning_rate": 0.00047395185556670013,
"loss": 3.2752,
"theoretical_loss": 4.346613087577991,
"tokens_seen": 203227136
},
{
"epoch": 0.06,
"learning_rate": 0.00047394182547642926,
"loss": 3.2155,
"theoretical_loss": 4.346438160728785,
"tokens_seen": 203292672
},
{
"epoch": 0.06,
"learning_rate": 0.0004739317953861585,
"loss": 3.2269,
"theoretical_loss": 4.346263306045983,
"tokens_seen": 203358208
},
{
"epoch": 0.06,
"learning_rate": 0.0004739217652958877,
"loss": 3.1447,
"theoretical_loss": 4.346088523476569,
"tokens_seen": 203423744
},
{
"epoch": 0.06,
"learning_rate": 0.00047391173520561686,
"loss": 3.1881,
"theoretical_loss": 4.345913812967584,
"tokens_seen": 203489280
},
{
"epoch": 0.06,
"learning_rate": 0.00047390170511534604,
"loss": 3.148,
"theoretical_loss": 4.345739174466127,
"tokens_seen": 203554816
},
{
"epoch": 0.06,
"learning_rate": 0.0004738916750250752,
"loss": 3.1253,
"theoretical_loss": 4.345564607919348,
"tokens_seen": 203620352
},
{
"epoch": 0.06,
"learning_rate": 0.0004738816449348044,
"loss": 3.2427,
"theoretical_loss": 4.3453901132744575,
"tokens_seen": 203685888
},
{
"epoch": 0.06,
"learning_rate": 0.00047387161484453363,
"loss": 3.1132,
"theoretical_loss": 4.345215690478719,
"tokens_seen": 203751424
},
{
"epoch": 0.06,
"learning_rate": 0.00047386158475426276,
"loss": 3.0614,
"theoretical_loss": 4.345041339479453,
"tokens_seen": 203816960
},
{
"epoch": 0.06,
"learning_rate": 0.000473851554663992,
"loss": 3.2345,
"theoretical_loss": 4.3448670602240345,
"tokens_seen": 203882496
},
{
"epoch": 0.06,
"learning_rate": 0.0004738415245737211,
"loss": 2.9035,
"theoretical_loss": 4.344692852659895,
"tokens_seen": 203948032
},
{
"epoch": 0.06,
"learning_rate": 0.00047383149448345036,
"loss": 3.2493,
"theoretical_loss": 4.34451871673452,
"tokens_seen": 204013568
},
{
"epoch": 0.06,
"learning_rate": 0.00047382146439317954,
"loss": 3.2788,
"theoretical_loss": 4.344344652395451,
"tokens_seen": 204079104
},
{
"epoch": 0.06,
"learning_rate": 0.0004738114343029087,
"loss": 3.0973,
"theoretical_loss": 4.3441706595902865,
"tokens_seen": 204144640
},
{
"epoch": 0.06,
"learning_rate": 0.0004738014042126379,
"loss": 3.0732,
"theoretical_loss": 4.343996738266677,
"tokens_seen": 204210176
},
{
"epoch": 0.06,
"learning_rate": 0.00047379137412236714,
"loss": 3.2518,
"theoretical_loss": 4.343822888372331,
"tokens_seen": 204275712
},
{
"epoch": 0.06,
"learning_rate": 0.00047378134403209626,
"loss": 3.064,
"theoretical_loss": 4.343649109855009,
"tokens_seen": 204341248
},
{
"epoch": 0.06,
"learning_rate": 0.0004737713139418255,
"loss": 3.3627,
"theoretical_loss": 4.343475402662529,
"tokens_seen": 204406784
},
{
"epoch": 0.06,
"learning_rate": 0.0004737612838515546,
"loss": 3.0398,
"theoretical_loss": 4.343301766742763,
"tokens_seen": 204472320
},
{
"epoch": 0.06,
"learning_rate": 0.00047375125376128386,
"loss": 3.0176,
"theoretical_loss": 4.343128202043638,
"tokens_seen": 204537856
},
{
"epoch": 0.06,
"learning_rate": 0.00047374122367101304,
"loss": 3.257,
"theoretical_loss": 4.342954708513136,
"tokens_seen": 204603392
},
{
"epoch": 0.06,
"learning_rate": 0.0004737311935807422,
"loss": 2.8876,
"theoretical_loss": 4.342781286099291,
"tokens_seen": 204668928
},
{
"epoch": 0.06,
"learning_rate": 0.00047372116349047146,
"loss": 3.3289,
"theoretical_loss": 4.3426079347501965,
"tokens_seen": 204734464
},
{
"epoch": 0.06,
"objective/train/docs_used": 126453,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.122795820236206,
"objective/train/theoretical_loss": 4.342434654413995,
"objective/train/tokens_used": 225260000,
"theoretical_loss": 4.342434654413995,
"tokens_seen": 204800000
},
{
"epoch": 0.06,
"learning_rate": 0.0004737111334002006,
"loss": 3.1005,
"theoretical_loss": 4.342434654413995,
"tokens_seen": 204800000
},
{
"epoch": 0.06,
"learning_rate": 0.0004737011033099298,
"loss": 3.2419,
"theoretical_loss": 4.342261445038888,
"tokens_seen": 204865536
},
{
"epoch": 0.06,
"learning_rate": 0.000473691073219659,
"loss": 3.2928,
"theoretical_loss": 4.342088306573128,
"tokens_seen": 204931072
},
{
"epoch": 0.06,
"learning_rate": 0.0004736810431293882,
"loss": 2.9942,
"theoretical_loss": 4.341915238965026,
"tokens_seen": 204996608
},
{
"epoch": 0.06,
"learning_rate": 0.00047367101303911736,
"loss": 3.3056,
"theoretical_loss": 4.34174224216294,
"tokens_seen": 205062144
},
{
"epoch": 0.06,
"learning_rate": 0.0004736609829488466,
"loss": 3.3729,
"theoretical_loss": 4.34156931611529,
"tokens_seen": 205127680
},
{
"epoch": 0.06,
"learning_rate": 0.0004736509528585757,
"loss": 3.1314,
"theoretical_loss": 4.341396460770547,
"tokens_seen": 205193216
},
{
"epoch": 0.06,
"learning_rate": 0.00047364092276830496,
"loss": 3.0286,
"theoretical_loss": 4.341223676077232,
"tokens_seen": 205258752
},
{
"epoch": 0.06,
"learning_rate": 0.0004736308926780341,
"loss": 3.3591,
"theoretical_loss": 4.341050961983926,
"tokens_seen": 205324288
},
{
"epoch": 0.06,
"learning_rate": 0.0004736208625877633,
"loss": 3.2768,
"theoretical_loss": 4.340878318439261,
"tokens_seen": 205389824
},
{
"epoch": 0.06,
"learning_rate": 0.0004736108324974925,
"loss": 3.1598,
"theoretical_loss": 4.340705745391922,
"tokens_seen": 205455360
},
{
"epoch": 0.06,
"learning_rate": 0.0004736008024072217,
"loss": 3.0908,
"theoretical_loss": 4.3405332427906504,
"tokens_seen": 205520896
},
{
"epoch": 0.06,
"learning_rate": 0.00047359077231695087,
"loss": 3.1934,
"theoretical_loss": 4.340360810584238,
"tokens_seen": 205586432
},
{
"epoch": 0.06,
"learning_rate": 0.00047358074222668005,
"loss": 3.1539,
"theoretical_loss": 4.340188448721532,
"tokens_seen": 205651968
},
{
"epoch": 0.06,
"learning_rate": 0.00047357071213640923,
"loss": 3.2624,
"theoretical_loss": 4.3400161571514335,
"tokens_seen": 205717504
},
{
"epoch": 0.06,
"learning_rate": 0.00047356068204613847,
"loss": 3.2158,
"theoretical_loss": 4.339843935822895,
"tokens_seen": 205783040
},
{
"epoch": 0.06,
"learning_rate": 0.0004735506519558676,
"loss": 3.1277,
"theoretical_loss": 4.339671784684923,
"tokens_seen": 205848576
},
{
"epoch": 0.06,
"learning_rate": 0.00047354062186559683,
"loss": 3.0358,
"theoretical_loss": 4.339499703686579,
"tokens_seen": 205914112
},
{
"epoch": 0.06,
"learning_rate": 0.00047353059177532595,
"loss": 3.2111,
"theoretical_loss": 4.339327692776977,
"tokens_seen": 205979648
},
{
"epoch": 0.06,
"learning_rate": 0.0004735205616850552,
"loss": 3.1453,
"theoretical_loss": 4.339155751905282,
"tokens_seen": 206045184
},
{
"epoch": 0.06,
"learning_rate": 0.00047351053159478437,
"loss": 3.2403,
"theoretical_loss": 4.338983881020713,
"tokens_seen": 206110720
},
{
"epoch": 0.06,
"learning_rate": 0.00047350050150451355,
"loss": 3.3298,
"theoretical_loss": 4.338812080072545,
"tokens_seen": 206176256
},
{
"epoch": 0.06,
"learning_rate": 0.00047349047141424273,
"loss": 3.3233,
"theoretical_loss": 4.338640349010101,
"tokens_seen": 206241792
},
{
"epoch": 0.06,
"learning_rate": 0.00047348044132397197,
"loss": 3.2418,
"theoretical_loss": 4.3384686877827585,
"tokens_seen": 206307328
},
{
"epoch": 0.06,
"learning_rate": 0.0004734704112337011,
"loss": 3.1538,
"theoretical_loss": 4.338297096339951,
"tokens_seen": 206372864
},
{
"epoch": 0.06,
"objective/train/docs_used": 127848,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0600967407226562,
"objective/train/theoretical_loss": 4.33812557463116,
"objective/train/tokens_used": 226898400,
"theoretical_loss": 4.33812557463116,
"tokens_seen": 206438400
},
{
"epoch": 0.06,
"learning_rate": 0.00047346038114343033,
"loss": 3.0882,
"theoretical_loss": 4.33812557463116,
"tokens_seen": 206438400
},
{
"epoch": 0.06,
"learning_rate": 0.00047345035105315946,
"loss": 3.2587,
"theoretical_loss": 4.3379541226059235,
"tokens_seen": 206503936
},
{
"epoch": 0.06,
"learning_rate": 0.0004734403209628887,
"loss": 3.1858,
"theoretical_loss": 4.337782740213827,
"tokens_seen": 206569472
},
{
"epoch": 0.06,
"learning_rate": 0.0004734302908726179,
"loss": 3.1159,
"theoretical_loss": 4.337611427404514,
"tokens_seen": 206635008
},
{
"epoch": 0.06,
"learning_rate": 0.00047342026078234706,
"loss": 3.3164,
"theoretical_loss": 4.337440184127679,
"tokens_seen": 206700544
},
{
"epoch": 0.06,
"learning_rate": 0.00047341023069207624,
"loss": 3.0572,
"theoretical_loss": 4.337269010333065,
"tokens_seen": 206766080
},
{
"epoch": 0.06,
"learning_rate": 0.0004734002006018054,
"loss": 2.8251,
"theoretical_loss": 4.337097905970471,
"tokens_seen": 206831616
},
{
"epoch": 0.06,
"learning_rate": 0.0004733901705115346,
"loss": 3.1861,
"theoretical_loss": 4.336926870989748,
"tokens_seen": 206897152
},
{
"epoch": 0.06,
"learning_rate": 0.00047338014042126383,
"loss": 3.3199,
"theoretical_loss": 4.336755905340797,
"tokens_seen": 206962688
},
{
"epoch": 0.06,
"learning_rate": 0.00047337011033099296,
"loss": 3.151,
"theoretical_loss": 4.336585008973573,
"tokens_seen": 207028224
},
{
"epoch": 0.06,
"learning_rate": 0.0004733600802407222,
"loss": 3.0852,
"theoretical_loss": 4.336414181838082,
"tokens_seen": 207093760
},
{
"epoch": 0.06,
"learning_rate": 0.0004733500501504513,
"loss": 3.1909,
"theoretical_loss": 4.336243423884382,
"tokens_seen": 207159296
},
{
"epoch": 0.06,
"learning_rate": 0.00047334002006018056,
"loss": 3.2911,
"theoretical_loss": 4.336072735062583,
"tokens_seen": 207224832
},
{
"epoch": 0.06,
"learning_rate": 0.00047332998996990974,
"loss": 3.2248,
"theoretical_loss": 4.335902115322847,
"tokens_seen": 207290368
},
{
"epoch": 0.06,
"learning_rate": 0.0004733199598796389,
"loss": 2.9551,
"theoretical_loss": 4.335731564615387,
"tokens_seen": 207355904
},
{
"epoch": 0.06,
"learning_rate": 0.0004733099297893681,
"loss": 3.3696,
"theoretical_loss": 4.335561082890468,
"tokens_seen": 207421440
},
{
"epoch": 0.06,
"learning_rate": 0.00047329989969909734,
"loss": 3.4282,
"theoretical_loss": 4.335390670098407,
"tokens_seen": 207486976
},
{
"epoch": 0.06,
"learning_rate": 0.00047328986960882646,
"loss": 3.2705,
"theoretical_loss": 4.335220326189571,
"tokens_seen": 207552512
},
{
"epoch": 0.06,
"learning_rate": 0.0004732798395185557,
"loss": 3.205,
"theoretical_loss": 4.335050051114379,
"tokens_seen": 207618048
},
{
"epoch": 0.06,
"learning_rate": 0.0004732698094282848,
"loss": 2.9493,
"theoretical_loss": 4.334879844823304,
"tokens_seen": 207683584
},
{
"epoch": 0.06,
"learning_rate": 0.00047325977933801406,
"loss": 3.0003,
"theoretical_loss": 4.334709707266865,
"tokens_seen": 207749120
},
{
"epoch": 0.06,
"learning_rate": 0.00047324974924774324,
"loss": 3.1429,
"theoretical_loss": 4.334539638395636,
"tokens_seen": 207814656
},
{
"epoch": 0.06,
"learning_rate": 0.0004732397191574724,
"loss": 3.2752,
"theoretical_loss": 4.334369638160242,
"tokens_seen": 207880192
},
{
"epoch": 0.06,
"learning_rate": 0.0004732296890672016,
"loss": 3.1693,
"theoretical_loss": 4.334199706511358,
"tokens_seen": 207945728
},
{
"epoch": 0.06,
"learning_rate": 0.0004732196589769308,
"loss": 3.402,
"theoretical_loss": 4.334029843399709,
"tokens_seen": 208011264
},
{
"epoch": 0.06,
"objective/train/docs_used": 128338,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.665210247039795,
"objective/train/theoretical_loss": 4.333860048776074,
"objective/train/tokens_used": 228536800,
"theoretical_loss": 4.333860048776074,
"tokens_seen": 208076800
},
{
"epoch": 0.06,
"learning_rate": 0.00047320962888665997,
"loss": 3.2236,
"theoretical_loss": 4.333860048776074,
"tokens_seen": 208076800
},
{
"epoch": 0.06,
"learning_rate": 0.0004731995987963892,
"loss": 3.167,
"theoretical_loss": 4.33369032259128,
"tokens_seen": 208142336
},
{
"epoch": 0.06,
"learning_rate": 0.00047318956870611833,
"loss": 3.1083,
"theoretical_loss": 4.333520664796206,
"tokens_seen": 208207872
},
{
"epoch": 0.06,
"learning_rate": 0.00047317953861584757,
"loss": 3.1765,
"theoretical_loss": 4.33335107534178,
"tokens_seen": 208273408
},
{
"epoch": 0.06,
"learning_rate": 0.00047316950852557675,
"loss": 3.1977,
"theoretical_loss": 4.333181554178985,
"tokens_seen": 208338944
},
{
"epoch": 0.06,
"learning_rate": 0.00047315947843530593,
"loss": 2.9084,
"theoretical_loss": 4.3330121012588485,
"tokens_seen": 208404480
},
{
"epoch": 0.06,
"learning_rate": 0.0004731494483450351,
"loss": 3.4184,
"theoretical_loss": 4.332842716532454,
"tokens_seen": 208470016
},
{
"epoch": 0.06,
"learning_rate": 0.0004731394182547643,
"loss": 3.2963,
"theoretical_loss": 4.332673399950932,
"tokens_seen": 208535552
},
{
"epoch": 0.06,
"learning_rate": 0.00047312938816449347,
"loss": 3.052,
"theoretical_loss": 4.332504151465464,
"tokens_seen": 208601088
},
{
"epoch": 0.06,
"learning_rate": 0.0004731193580742227,
"loss": 3.0663,
"theoretical_loss": 4.332334971027284,
"tokens_seen": 208666624
},
{
"epoch": 0.06,
"learning_rate": 0.00047310932798395183,
"loss": 3.0481,
"theoretical_loss": 4.332165858587672,
"tokens_seen": 208732160
},
{
"epoch": 0.06,
"learning_rate": 0.00047309929789368107,
"loss": 3.3365,
"theoretical_loss": 4.331996814097963,
"tokens_seen": 208797696
},
{
"epoch": 0.06,
"learning_rate": 0.0004730892678034102,
"loss": 3.1267,
"theoretical_loss": 4.331827837509538,
"tokens_seen": 208863232
},
{
"epoch": 0.06,
"learning_rate": 0.00047307923771313943,
"loss": 3.3878,
"theoretical_loss": 4.331658928773831,
"tokens_seen": 208928768
},
{
"epoch": 0.06,
"learning_rate": 0.0004730692076228686,
"loss": 3.0899,
"theoretical_loss": 4.331490087842324,
"tokens_seen": 208994304
},
{
"epoch": 0.06,
"learning_rate": 0.0004730591775325978,
"loss": 3.2895,
"theoretical_loss": 4.33132131466655,
"tokens_seen": 209059840
},
{
"epoch": 0.06,
"learning_rate": 0.000473049147442327,
"loss": 3.2144,
"theoretical_loss": 4.3311526091980905,
"tokens_seen": 209125376
},
{
"epoch": 0.06,
"learning_rate": 0.00047303911735205615,
"loss": 2.7723,
"theoretical_loss": 4.330983971388578,
"tokens_seen": 209190912
},
{
"epoch": 0.06,
"learning_rate": 0.00047302908726178534,
"loss": 2.9338,
"theoretical_loss": 4.330815401189695,
"tokens_seen": 209256448
},
{
"epoch": 0.06,
"learning_rate": 0.00047301905717151457,
"loss": 3.1443,
"theoretical_loss": 4.330646898553173,
"tokens_seen": 209321984
},
{
"epoch": 0.06,
"learning_rate": 0.0004730090270812437,
"loss": 3.2174,
"theoretical_loss": 4.330478463430792,
"tokens_seen": 209387520
},
{
"epoch": 0.06,
"learning_rate": 0.00047299899699097293,
"loss": 3.2128,
"theoretical_loss": 4.330310095774383,
"tokens_seen": 209453056
},
{
"epoch": 0.06,
"learning_rate": 0.0004729889669007021,
"loss": 3.1469,
"theoretical_loss": 4.330141795535828,
"tokens_seen": 209518592
},
{
"epoch": 0.06,
"learning_rate": 0.0004729789368104313,
"loss": 3.2627,
"theoretical_loss": 4.329973562667053,
"tokens_seen": 209584128
},
{
"epoch": 0.06,
"learning_rate": 0.00047296890672016053,
"loss": 3.3848,
"theoretical_loss": 4.3298053971200385,
"tokens_seen": 209649664
},
{
"epoch": 0.06,
"objective/train/docs_used": 129483,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.797501802444458,
"objective/train/theoretical_loss": 4.329637298846812,
"objective/train/tokens_used": 230175200,
"theoretical_loss": 4.329637298846812,
"tokens_seen": 209715200
},
{
"epoch": 0.06,
"learning_rate": 0.00047295887662988966,
"loss": 3.3015,
"theoretical_loss": 4.329637298846812,
"tokens_seen": 209715200
},
{
"epoch": 0.06,
"learning_rate": 0.0004729488465396189,
"loss": 3.2439,
"theoretical_loss": 4.329469267799451,
"tokens_seen": 209780736
},
{
"epoch": 0.06,
"learning_rate": 0.0004729388164493481,
"loss": 3.2575,
"theoretical_loss": 4.32930130393008,
"tokens_seen": 209846272
},
{
"epoch": 0.06,
"learning_rate": 0.00047292878635907726,
"loss": 3.1579,
"theoretical_loss": 4.329133407190876,
"tokens_seen": 209911808
},
{
"epoch": 0.06,
"learning_rate": 0.00047291875626880644,
"loss": 3.204,
"theoretical_loss": 4.3289655775340625,
"tokens_seen": 209977344
},
{
"epoch": 0.06,
"learning_rate": 0.0004729087261785356,
"loss": 3.1764,
"theoretical_loss": 4.328797814911912,
"tokens_seen": 210042880
},
{
"epoch": 0.06,
"learning_rate": 0.0004728986960882648,
"loss": 3.5043,
"theoretical_loss": 4.328630119276747,
"tokens_seen": 210108416
},
{
"epoch": 0.06,
"learning_rate": 0.00047288866599799403,
"loss": 3.3447,
"theoretical_loss": 4.328462490580938,
"tokens_seen": 210173952
},
{
"epoch": 0.06,
"learning_rate": 0.00047287863590772316,
"loss": 3.1687,
"theoretical_loss": 4.328294928776903,
"tokens_seen": 210239488
},
{
"epoch": 0.06,
"learning_rate": 0.0004728686058174524,
"loss": 3.3405,
"theoretical_loss": 4.328127433817112,
"tokens_seen": 210305024
},
{
"epoch": 0.06,
"learning_rate": 0.0004728585757271815,
"loss": 3.0517,
"theoretical_loss": 4.327960005654081,
"tokens_seen": 210370560
},
{
"epoch": 0.06,
"learning_rate": 0.00047284854563691076,
"loss": 3.3842,
"theoretical_loss": 4.327792644240374,
"tokens_seen": 210436096
},
{
"epoch": 0.06,
"learning_rate": 0.00047283851554663994,
"loss": 3.2548,
"theoretical_loss": 4.327625349528605,
"tokens_seen": 210501632
},
{
"epoch": 0.06,
"learning_rate": 0.0004728284854563691,
"loss": 3.3037,
"theoretical_loss": 4.327458121471436,
"tokens_seen": 210567168
},
{
"epoch": 0.06,
"learning_rate": 0.0004728184553660983,
"loss": 3.1623,
"theoretical_loss": 4.3272909600215765,
"tokens_seen": 210632704
},
{
"epoch": 0.06,
"learning_rate": 0.00047280842527582754,
"loss": 2.8898,
"theoretical_loss": 4.327123865131786,
"tokens_seen": 210698240
},
{
"epoch": 0.06,
"learning_rate": 0.00047279839518555666,
"loss": 3.2277,
"theoretical_loss": 4.326956836754871,
"tokens_seen": 210763776
},
{
"epoch": 0.06,
"learning_rate": 0.0004727883650952859,
"loss": 3.2198,
"theoretical_loss": 4.326789874843685,
"tokens_seen": 210829312
},
{
"epoch": 0.06,
"learning_rate": 0.000472778335005015,
"loss": 3.577,
"theoretical_loss": 4.326622979351132,
"tokens_seen": 210894848
},
{
"epoch": 0.06,
"learning_rate": 0.00047276830491474426,
"loss": 3.1072,
"theoretical_loss": 4.326456150230163,
"tokens_seen": 210960384
},
{
"epoch": 0.06,
"learning_rate": 0.00047275827482447344,
"loss": 3.3812,
"theoretical_loss": 4.326289387433776,
"tokens_seen": 211025920
},
{
"epoch": 0.06,
"learning_rate": 0.0004727482447342026,
"loss": 3.3247,
"theoretical_loss": 4.326122690915017,
"tokens_seen": 211091456
},
{
"epoch": 0.06,
"learning_rate": 0.0004727382146439318,
"loss": 3.1334,
"theoretical_loss": 4.325956060626982,
"tokens_seen": 211156992
},
{
"epoch": 0.06,
"learning_rate": 0.000472728184553661,
"loss": 3.2365,
"theoretical_loss": 4.325789496522812,
"tokens_seen": 211222528
},
{
"epoch": 0.06,
"learning_rate": 0.00047271815446339017,
"loss": 3.3295,
"theoretical_loss": 4.325622998555697,
"tokens_seen": 211288064
},
{
"epoch": 0.06,
"objective/train/docs_used": 130263,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.355026960372925,
"objective/train/theoretical_loss": 4.3254565666788745,
"objective/train/tokens_used": 231813600,
"theoretical_loss": 4.3254565666788745,
"tokens_seen": 211353600
},
{
"epoch": 0.06,
"learning_rate": 0.0004727081243731194,
"loss": 3.3001,
"theoretical_loss": 4.3254565666788745,
"tokens_seen": 211353600
},
{
"epoch": 0.06,
"learning_rate": 0.00047269809428284853,
"loss": 3.2941,
"theoretical_loss": 4.325290200845629,
"tokens_seen": 211419136
},
{
"epoch": 0.06,
"learning_rate": 0.00047268806419257777,
"loss": 3.3354,
"theoretical_loss": 4.3251239010092934,
"tokens_seen": 211484672
},
{
"epoch": 0.06,
"learning_rate": 0.00047267803410230695,
"loss": 3.3392,
"theoretical_loss": 4.324957667123249,
"tokens_seen": 211550208
},
{
"epoch": 0.06,
"learning_rate": 0.00047266800401203613,
"loss": 3.1309,
"theoretical_loss": 4.32479149914092,
"tokens_seen": 211615744
},
{
"epoch": 0.06,
"learning_rate": 0.0004726579739217653,
"loss": 3.2534,
"theoretical_loss": 4.324625397015783,
"tokens_seen": 211681280
},
{
"epoch": 0.06,
"learning_rate": 0.0004726479438314945,
"loss": 3.0925,
"theoretical_loss": 4.3244593607013595,
"tokens_seen": 211746816
},
{
"epoch": 0.06,
"learning_rate": 0.00047263791374122367,
"loss": 3.2174,
"theoretical_loss": 4.324293390151218,
"tokens_seen": 211812352
},
{
"epoch": 0.06,
"learning_rate": 0.0004726278836509529,
"loss": 3.1953,
"theoretical_loss": 4.324127485318975,
"tokens_seen": 211877888
},
{
"epoch": 0.06,
"learning_rate": 0.00047261785356068203,
"loss": 3.0756,
"theoretical_loss": 4.323961646158294,
"tokens_seen": 211943424
},
{
"epoch": 0.06,
"learning_rate": 0.00047260782347041127,
"loss": 3.3479,
"theoretical_loss": 4.323795872622884,
"tokens_seen": 212008960
},
{
"epoch": 0.06,
"learning_rate": 0.0004725977933801404,
"loss": 3.1166,
"theoretical_loss": 4.323630164666502,
"tokens_seen": 212074496
},
{
"epoch": 0.06,
"learning_rate": 0.00047258776328986963,
"loss": 3.3872,
"theoretical_loss": 4.323464522242954,
"tokens_seen": 212140032
},
{
"epoch": 0.06,
"learning_rate": 0.0004725777331995988,
"loss": 3.0378,
"theoretical_loss": 4.323298945306089,
"tokens_seen": 212205568
},
{
"epoch": 0.06,
"learning_rate": 0.000472567703109328,
"loss": 2.9616,
"theoretical_loss": 4.3231334338098035,
"tokens_seen": 212271104
},
{
"epoch": 0.06,
"learning_rate": 0.0004725576730190572,
"loss": 3.1819,
"theoretical_loss": 4.322967987708043,
"tokens_seen": 212336640
},
{
"epoch": 0.06,
"learning_rate": 0.00047254764292878636,
"loss": 3.2994,
"theoretical_loss": 4.322802606954798,
"tokens_seen": 212402176
},
{
"epoch": 0.06,
"learning_rate": 0.00047253761283851554,
"loss": 3.2778,
"theoretical_loss": 4.322637291504106,
"tokens_seen": 212467712
},
{
"epoch": 0.06,
"learning_rate": 0.00047252758274824477,
"loss": 3.3196,
"theoretical_loss": 4.32247204131005,
"tokens_seen": 212533248
},
{
"epoch": 0.06,
"learning_rate": 0.0004725175526579739,
"loss": 3.2477,
"theoretical_loss": 4.322306856326761,
"tokens_seen": 212598784
},
{
"epoch": 0.06,
"learning_rate": 0.00047250752256770313,
"loss": 3.4815,
"theoretical_loss": 4.322141736508415,
"tokens_seen": 212664320
},
{
"epoch": 0.06,
"learning_rate": 0.0004724974924774323,
"loss": 3.3946,
"theoretical_loss": 4.321976681809236,
"tokens_seen": 212729856
},
{
"epoch": 0.06,
"learning_rate": 0.0004724874623871615,
"loss": 3.2502,
"theoretical_loss": 4.321811692183491,
"tokens_seen": 212795392
},
{
"epoch": 0.06,
"learning_rate": 0.0004724774322968907,
"loss": 3.2951,
"theoretical_loss": 4.321646767585497,
"tokens_seen": 212860928
},
{
"epoch": 0.06,
"learning_rate": 0.00047246740220661986,
"loss": 3.0522,
"theoretical_loss": 4.3214819079696145,
"tokens_seen": 212926464
},
{
"epoch": 0.06,
"objective/train/docs_used": 131630,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.6950013637542725,
"objective/train/theoretical_loss": 4.321317113290252,
"objective/train/tokens_used": 233452000,
"theoretical_loss": 4.321317113290252,
"tokens_seen": 212992000
},
{
"epoch": 0.06,
"learning_rate": 0.00047245737211634904,
"loss": 3.3024,
"theoretical_loss": 4.321317113290252,
"tokens_seen": 212992000
},
{
"epoch": 0.06,
"learning_rate": 0.0004724473420260783,
"loss": 3.0431,
"theoretical_loss": 4.321152383501863,
"tokens_seen": 213057536
},
{
"epoch": 0.06,
"learning_rate": 0.0004724373119358074,
"loss": 2.8804,
"theoretical_loss": 4.320987718558945,
"tokens_seen": 213123072
},
{
"epoch": 0.06,
"learning_rate": 0.00047242728184553664,
"loss": 3.1601,
"theoretical_loss": 4.320823118416046,
"tokens_seen": 213188608
},
{
"epoch": 0.06,
"learning_rate": 0.00047241725175526576,
"loss": 3.385,
"theoretical_loss": 4.320658583027755,
"tokens_seen": 213254144
},
{
"epoch": 0.06,
"learning_rate": 0.000472407221664995,
"loss": 3.2701,
"theoretical_loss": 4.32049411234871,
"tokens_seen": 213319680
},
{
"epoch": 0.06,
"learning_rate": 0.0004723971915747242,
"loss": 3.1385,
"theoretical_loss": 4.3203297063335935,
"tokens_seen": 213385216
},
{
"epoch": 0.06,
"learning_rate": 0.00047238716148445336,
"loss": 3.1332,
"theoretical_loss": 4.320165364937134,
"tokens_seen": 213450752
},
{
"epoch": 0.06,
"learning_rate": 0.00047237713139418254,
"loss": 3.3959,
"theoretical_loss": 4.320001088114105,
"tokens_seen": 213516288
},
{
"epoch": 0.06,
"learning_rate": 0.0004723671013039117,
"loss": 3.5302,
"theoretical_loss": 4.319836875819325,
"tokens_seen": 213581824
},
{
"epoch": 0.06,
"learning_rate": 0.0004723570712136409,
"loss": 3.3552,
"theoretical_loss": 4.31967272800766,
"tokens_seen": 213647360
},
{
"epoch": 0.06,
"learning_rate": 0.00047234704112337014,
"loss": 3.2224,
"theoretical_loss": 4.319508644634021,
"tokens_seen": 213712896
},
{
"epoch": 0.06,
"learning_rate": 0.00047233701103309927,
"loss": 3.1956,
"theoretical_loss": 4.319344625653361,
"tokens_seen": 213778432
},
{
"epoch": 0.06,
"learning_rate": 0.0004723269809428285,
"loss": 3.0912,
"theoretical_loss": 4.319180671020684,
"tokens_seen": 213843968
},
{
"epoch": 0.06,
"learning_rate": 0.0004723169508525577,
"loss": 3.22,
"theoretical_loss": 4.319016780691033,
"tokens_seen": 213909504
},
{
"epoch": 0.06,
"learning_rate": 0.00047230692076228686,
"loss": 3.2483,
"theoretical_loss": 4.318852954619501,
"tokens_seen": 213975040
},
{
"epoch": 0.06,
"learning_rate": 0.00047229689067201605,
"loss": 3.19,
"theoretical_loss": 4.318689192761225,
"tokens_seen": 214040576
},
{
"epoch": 0.06,
"learning_rate": 0.0004722868605817452,
"loss": 3.0916,
"theoretical_loss": 4.318525495071385,
"tokens_seen": 214106112
},
{
"epoch": 0.06,
"learning_rate": 0.0004722768304914744,
"loss": 3.5612,
"theoretical_loss": 4.318361861505207,
"tokens_seen": 214171648
},
{
"epoch": 0.06,
"learning_rate": 0.00047226680040120364,
"loss": 3.2789,
"theoretical_loss": 4.318198292017964,
"tokens_seen": 214237184
},
{
"epoch": 0.06,
"learning_rate": 0.00047225677031093277,
"loss": 3.219,
"theoretical_loss": 4.318034786564971,
"tokens_seen": 214302720
},
{
"epoch": 0.06,
"learning_rate": 0.000472246740220662,
"loss": 3.3485,
"theoretical_loss": 4.31787134510159,
"tokens_seen": 214368256
},
{
"epoch": 0.06,
"learning_rate": 0.00047223671013039113,
"loss": 3.2924,
"theoretical_loss": 4.3177079675832255,
"tokens_seen": 214433792
},
{
"epoch": 0.06,
"learning_rate": 0.00047222668004012037,
"loss": 3.2976,
"theoretical_loss": 4.317544653965329,
"tokens_seen": 214499328
},
{
"epoch": 0.07,
"learning_rate": 0.0004722166499498496,
"loss": 3.2069,
"theoretical_loss": 4.3173814042033944,
"tokens_seen": 214564864
},
{
"epoch": 0.07,
"objective/train/docs_used": 132228,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.7650396823883057,
"objective/train/theoretical_loss": 4.317218218252963,
"objective/train/tokens_used": 235090400,
"theoretical_loss": 4.317218218252963,
"tokens_seen": 214630400
},
{
"epoch": 0.07,
"learning_rate": 0.00047220661985957873,
"loss": 3.3818,
"theoretical_loss": 4.317218218252963,
"tokens_seen": 214630400
},
{
"epoch": 0.07,
"learning_rate": 0.00047219658976930797,
"loss": 3.1204,
"theoretical_loss": 4.317055096069618,
"tokens_seen": 214695936
},
{
"epoch": 0.07,
"learning_rate": 0.00047218655967903715,
"loss": 3.2258,
"theoretical_loss": 4.316892037608987,
"tokens_seen": 214761472
},
{
"epoch": 0.07,
"learning_rate": 0.00047217652958876633,
"loss": 2.9512,
"theoretical_loss": 4.316729042826745,
"tokens_seen": 214827008
},
{
"epoch": 0.07,
"learning_rate": 0.0004721664994984955,
"loss": 3.3851,
"theoretical_loss": 4.316566111678609,
"tokens_seen": 214892544
},
{
"epoch": 0.07,
"learning_rate": 0.0004721564694082247,
"loss": 3.307,
"theoretical_loss": 4.316403244120339,
"tokens_seen": 214958080
},
{
"epoch": 0.07,
"learning_rate": 0.00047214643931795387,
"loss": 3.0003,
"theoretical_loss": 4.3162404401077445,
"tokens_seen": 215023616
},
{
"epoch": 0.07,
"learning_rate": 0.0004721364092276831,
"loss": 3.4895,
"theoretical_loss": 4.316077699596671,
"tokens_seen": 215089152
},
{
"epoch": 0.07,
"learning_rate": 0.00047212637913741223,
"loss": 3.1155,
"theoretical_loss": 4.315915022543016,
"tokens_seen": 215154688
},
{
"epoch": 0.07,
"learning_rate": 0.00047211634904714147,
"loss": 3.0629,
"theoretical_loss": 4.315752408902716,
"tokens_seen": 215220224
},
{
"epoch": 0.07,
"learning_rate": 0.0004721063189568706,
"loss": 3.2205,
"theoretical_loss": 4.315589858631755,
"tokens_seen": 215285760
},
{
"epoch": 0.07,
"learning_rate": 0.00047209628886659983,
"loss": 3.2874,
"theoretical_loss": 4.315427371686157,
"tokens_seen": 215351296
},
{
"epoch": 0.07,
"learning_rate": 0.000472086258776329,
"loss": 3.3042,
"theoretical_loss": 4.315264948021994,
"tokens_seen": 215416832
},
{
"epoch": 0.07,
"learning_rate": 0.0004720762286860582,
"loss": 3.3027,
"theoretical_loss": 4.315102587595379,
"tokens_seen": 215482368
},
{
"epoch": 0.07,
"learning_rate": 0.0004720661985957874,
"loss": 3.1877,
"theoretical_loss": 4.31494029036247,
"tokens_seen": 215547904
},
{
"epoch": 0.07,
"learning_rate": 0.00047205616850551656,
"loss": 3.2381,
"theoretical_loss": 4.314778056279468,
"tokens_seen": 215613440
},
{
"epoch": 0.07,
"learning_rate": 0.00047204613841524574,
"loss": 3.0961,
"theoretical_loss": 4.314615885302619,
"tokens_seen": 215678976
},
{
"epoch": 0.07,
"learning_rate": 0.00047203610832497497,
"loss": 3.0676,
"theoretical_loss": 4.314453777388209,
"tokens_seen": 215744512
},
{
"epoch": 0.07,
"learning_rate": 0.0004720260782347041,
"loss": 3.2471,
"theoretical_loss": 4.314291732492573,
"tokens_seen": 215810048
},
{
"epoch": 0.07,
"learning_rate": 0.00047201604814443333,
"loss": 3.2031,
"theoretical_loss": 4.314129750572087,
"tokens_seen": 215875584
},
{
"epoch": 0.07,
"learning_rate": 0.0004720060180541625,
"loss": 2.8863,
"theoretical_loss": 4.3139678315831675,
"tokens_seen": 215941120
},
{
"epoch": 0.07,
"learning_rate": 0.0004719959879638917,
"loss": 3.0952,
"theoretical_loss": 4.313805975482278,
"tokens_seen": 216006656
},
{
"epoch": 0.07,
"learning_rate": 0.0004719859578736209,
"loss": 3.196,
"theoretical_loss": 4.313644182225926,
"tokens_seen": 216072192
},
{
"epoch": 0.07,
"learning_rate": 0.00047197592778335006,
"loss": 2.9889,
"theoretical_loss": 4.313482451770659,
"tokens_seen": 216137728
},
{
"epoch": 0.07,
"learning_rate": 0.00047196589769307924,
"loss": 3.1137,
"theoretical_loss": 4.313320784073069,
"tokens_seen": 216203264
},
{
"epoch": 0.07,
"objective/train/docs_used": 133529,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.224602460861206,
"objective/train/theoretical_loss": 4.3131591790897925,
"objective/train/tokens_used": 236728800,
"theoretical_loss": 4.3131591790897925,
"tokens_seen": 216268800
},
{
"epoch": 0.07,
"learning_rate": 0.0004719558676028085,
"loss": 3.3034,
"theoretical_loss": 4.3131591790897925,
"tokens_seen": 216268800
},
{
"epoch": 0.07,
"learning_rate": 0.0004719458375125376,
"loss": 3.0825,
"theoretical_loss": 4.3129976367775065,
"tokens_seen": 216334336
},
{
"epoch": 0.07,
"learning_rate": 0.00047193580742226684,
"loss": 3.293,
"theoretical_loss": 4.312836157092934,
"tokens_seen": 216399872
},
{
"epoch": 0.07,
"learning_rate": 0.00047192577733199596,
"loss": 3.172,
"theoretical_loss": 4.312674739992839,
"tokens_seen": 216465408
},
{
"epoch": 0.07,
"learning_rate": 0.0004719157472417252,
"loss": 3.128,
"theoretical_loss": 4.31251338543403,
"tokens_seen": 216530944
},
{
"epoch": 0.07,
"learning_rate": 0.0004719057171514544,
"loss": 3.3622,
"theoretical_loss": 4.312352093373354,
"tokens_seen": 216596480
},
{
"epoch": 0.07,
"learning_rate": 0.00047189568706118356,
"loss": 3.3549,
"theoretical_loss": 4.312190863767708,
"tokens_seen": 216662016
},
{
"epoch": 0.07,
"learning_rate": 0.00047188565697091274,
"loss": 3.1291,
"theoretical_loss": 4.312029696574027,
"tokens_seen": 216727552
},
{
"epoch": 0.07,
"learning_rate": 0.0004718756268806419,
"loss": 2.8499,
"theoretical_loss": 4.311868591749287,
"tokens_seen": 216793088
},
{
"epoch": 0.07,
"learning_rate": 0.0004718655967903711,
"loss": 3.0329,
"theoretical_loss": 4.311707549250514,
"tokens_seen": 216858624
},
{
"epoch": 0.07,
"learning_rate": 0.00047185556670010034,
"loss": 3.5888,
"theoretical_loss": 4.311546569034767,
"tokens_seen": 216924160
},
{
"epoch": 0.07,
"learning_rate": 0.00047184553660982947,
"loss": 3.4006,
"theoretical_loss": 4.311385651059155,
"tokens_seen": 216989696
},
{
"epoch": 0.07,
"learning_rate": 0.0004718355065195587,
"loss": 3.2021,
"theoretical_loss": 4.311224795280825,
"tokens_seen": 217055232
},
{
"epoch": 0.07,
"learning_rate": 0.0004718254764292879,
"loss": 3.2574,
"theoretical_loss": 4.3110640016569715,
"tokens_seen": 217120768
},
{
"epoch": 0.07,
"learning_rate": 0.00047181544633901706,
"loss": 3.4922,
"theoretical_loss": 4.310903270144825,
"tokens_seen": 217186304
},
{
"epoch": 0.07,
"learning_rate": 0.00047180541624874625,
"loss": 3.4122,
"theoretical_loss": 4.310742600701664,
"tokens_seen": 217251840
},
{
"epoch": 0.07,
"learning_rate": 0.0004717953861584754,
"loss": 2.913,
"theoretical_loss": 4.310581993284805,
"tokens_seen": 217317376
},
{
"epoch": 0.07,
"learning_rate": 0.0004717853560682046,
"loss": 3.0789,
"theoretical_loss": 4.310421447851609,
"tokens_seen": 217382912
},
{
"epoch": 0.07,
"learning_rate": 0.00047177532597793384,
"loss": 3.1302,
"theoretical_loss": 4.310260964359479,
"tokens_seen": 217448448
},
{
"epoch": 0.07,
"learning_rate": 0.00047176529588766297,
"loss": 3.0638,
"theoretical_loss": 4.310100542765858,
"tokens_seen": 217513984
},
{
"epoch": 0.07,
"learning_rate": 0.0004717552657973922,
"loss": 3.1546,
"theoretical_loss": 4.309940183028236,
"tokens_seen": 217579520
},
{
"epoch": 0.07,
"learning_rate": 0.00047174523570712133,
"loss": 3.3793,
"theoretical_loss": 4.309779885104139,
"tokens_seen": 217645056
},
{
"epoch": 0.07,
"learning_rate": 0.00047173520561685057,
"loss": 3.2103,
"theoretical_loss": 4.309619648951139,
"tokens_seen": 217710592
},
{
"epoch": 0.07,
"learning_rate": 0.00047172517552657975,
"loss": 3.3218,
"theoretical_loss": 4.3094594745268475,
"tokens_seen": 217776128
},
{
"epoch": 0.07,
"learning_rate": 0.00047171514543630893,
"loss": 3.0378,
"theoretical_loss": 4.30929936178892,
"tokens_seen": 217841664
},
{
"epoch": 0.07,
"objective/train/docs_used": 134228,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2407047748565674,
"objective/train/theoretical_loss": 4.309139310695053,
"objective/train/tokens_used": 238367200,
"theoretical_loss": 4.309139310695053,
"tokens_seen": 217907200
},
{
"epoch": 0.07,
"learning_rate": 0.0004717051153460381,
"loss": 3.2923,
"theoretical_loss": 4.309139310695053,
"tokens_seen": 217907200
},
{
"epoch": 0.07,
"learning_rate": 0.00047169508525576735,
"loss": 3.2057,
"theoretical_loss": 4.308979321202983,
"tokens_seen": 217972736
},
{
"epoch": 0.07,
"learning_rate": 0.0004716850551654965,
"loss": 3.2015,
"theoretical_loss": 4.308819393270491,
"tokens_seen": 218038272
},
{
"epoch": 0.07,
"learning_rate": 0.0004716750250752257,
"loss": 3.064,
"theoretical_loss": 4.308659526855396,
"tokens_seen": 218103808
},
{
"epoch": 0.07,
"learning_rate": 0.00047166499498495484,
"loss": 3.0323,
"theoretical_loss": 4.308499721915563,
"tokens_seen": 218169344
},
{
"epoch": 0.07,
"learning_rate": 0.00047165496489468407,
"loss": 3.2744,
"theoretical_loss": 4.308339978408897,
"tokens_seen": 218234880
},
{
"epoch": 0.07,
"learning_rate": 0.00047164493480441325,
"loss": 3.1236,
"theoretical_loss": 4.308180296293341,
"tokens_seen": 218300416
},
{
"epoch": 0.07,
"learning_rate": 0.00047163490471414243,
"loss": 3.117,
"theoretical_loss": 4.308020675526883,
"tokens_seen": 218365952
},
{
"epoch": 0.07,
"learning_rate": 0.0004716248746238716,
"loss": 3.1281,
"theoretical_loss": 4.307861116067554,
"tokens_seen": 218431488
},
{
"epoch": 0.07,
"learning_rate": 0.0004716148445336008,
"loss": 2.9074,
"theoretical_loss": 4.30770161787342,
"tokens_seen": 218497024
},
{
"epoch": 0.07,
"learning_rate": 0.00047160481444333,
"loss": 3.2527,
"theoretical_loss": 4.307542180902594,
"tokens_seen": 218562560
},
{
"epoch": 0.07,
"learning_rate": 0.0004715947843530592,
"loss": 3.0822,
"theoretical_loss": 4.307382805113228,
"tokens_seen": 218628096
},
{
"epoch": 0.07,
"learning_rate": 0.00047158475426278834,
"loss": 3.3932,
"theoretical_loss": 4.307223490463516,
"tokens_seen": 218693632
},
{
"epoch": 0.07,
"learning_rate": 0.0004715747241725176,
"loss": 3.1793,
"theoretical_loss": 4.307064236911692,
"tokens_seen": 218759168
},
{
"epoch": 0.07,
"learning_rate": 0.0004715646940822467,
"loss": 3.3369,
"theoretical_loss": 4.30690504441603,
"tokens_seen": 218824704
},
{
"epoch": 0.07,
"learning_rate": 0.00047155466399197594,
"loss": 3.1318,
"theoretical_loss": 4.306745912934849,
"tokens_seen": 218890240
},
{
"epoch": 0.07,
"learning_rate": 0.0004715446339017051,
"loss": 3.0594,
"theoretical_loss": 4.306586842426504,
"tokens_seen": 218955776
},
{
"epoch": 0.07,
"learning_rate": 0.0004715346038114343,
"loss": 3.3079,
"theoretical_loss": 4.306427832849394,
"tokens_seen": 219021312
},
{
"epoch": 0.07,
"learning_rate": 0.0004715245737211635,
"loss": 3.19,
"theoretical_loss": 4.306268884161959,
"tokens_seen": 219086848
},
{
"epoch": 0.07,
"learning_rate": 0.0004715145436308927,
"loss": 3.3318,
"theoretical_loss": 4.306109996322679,
"tokens_seen": 219152384
},
{
"epoch": 0.07,
"learning_rate": 0.00047150451354062184,
"loss": 3.0093,
"theoretical_loss": 4.305951169290073,
"tokens_seen": 219217920
},
{
"epoch": 0.07,
"learning_rate": 0.0004714944834503511,
"loss": 3.3021,
"theoretical_loss": 4.305792403022703,
"tokens_seen": 219283456
},
{
"epoch": 0.07,
"learning_rate": 0.0004714844533600802,
"loss": 3.0608,
"theoretical_loss": 4.305633697479171,
"tokens_seen": 219348992
},
{
"epoch": 0.07,
"learning_rate": 0.00047147442326980944,
"loss": 2.8014,
"theoretical_loss": 4.305475052618119,
"tokens_seen": 219414528
},
{
"epoch": 0.07,
"learning_rate": 0.0004714643931795387,
"loss": 3.3019,
"theoretical_loss": 4.30531646839823,
"tokens_seen": 219480064
},
{
"epoch": 0.07,
"objective/train/docs_used": 135676,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.066269874572754,
"objective/train/theoretical_loss": 4.305157944778228,
"objective/train/tokens_used": 240005600,
"theoretical_loss": 4.305157944778228,
"tokens_seen": 219545600
},
{
"epoch": 0.07,
"learning_rate": 0.0004714543630892678,
"loss": 3.0659,
"theoretical_loss": 4.305157944778228,
"tokens_seen": 219545600
},
{
"epoch": 0.07,
"learning_rate": 0.00047144433299899704,
"loss": 3.251,
"theoretical_loss": 4.304999481716876,
"tokens_seen": 219611136
},
{
"epoch": 0.07,
"learning_rate": 0.00047143430290872616,
"loss": 3.4636,
"theoretical_loss": 4.304841079172979,
"tokens_seen": 219676672
},
{
"epoch": 0.07,
"learning_rate": 0.0004714242728184554,
"loss": 3.2243,
"theoretical_loss": 4.30468273710538,
"tokens_seen": 219742208
},
{
"epoch": 0.07,
"learning_rate": 0.0004714142427281846,
"loss": 3.2012,
"theoretical_loss": 4.304524455472965,
"tokens_seen": 219807744
},
{
"epoch": 0.07,
"learning_rate": 0.00047140421263791376,
"loss": 2.9528,
"theoretical_loss": 4.304366234234659,
"tokens_seen": 219873280
},
{
"epoch": 0.07,
"learning_rate": 0.00047139418254764294,
"loss": 3.214,
"theoretical_loss": 4.304208073349426,
"tokens_seen": 219938816
},
{
"epoch": 0.07,
"learning_rate": 0.0004713841524573721,
"loss": 3.3243,
"theoretical_loss": 4.304049972776271,
"tokens_seen": 220004352
},
{
"epoch": 0.07,
"learning_rate": 0.0004713741223671013,
"loss": 3.169,
"theoretical_loss": 4.30389193247424,
"tokens_seen": 220069888
},
{
"epoch": 0.07,
"learning_rate": 0.00047136409227683054,
"loss": 3.2084,
"theoretical_loss": 4.303733952402419,
"tokens_seen": 220135424
},
{
"epoch": 0.07,
"learning_rate": 0.00047135406218655967,
"loss": 3.2908,
"theoretical_loss": 4.303576032519931,
"tokens_seen": 220200960
},
{
"epoch": 0.07,
"learning_rate": 0.0004713440320962889,
"loss": 3.415,
"theoretical_loss": 4.303418172785943,
"tokens_seen": 220266496
},
{
"epoch": 0.07,
"learning_rate": 0.0004713340020060181,
"loss": 3.3773,
"theoretical_loss": 4.303260373159659,
"tokens_seen": 220332032
},
{
"epoch": 0.07,
"learning_rate": 0.00047132397191574726,
"loss": 3.2734,
"theoretical_loss": 4.303102633600322,
"tokens_seen": 220397568
},
{
"epoch": 0.07,
"learning_rate": 0.00047131394182547645,
"loss": 2.8756,
"theoretical_loss": 4.30294495406722,
"tokens_seen": 220463104
},
{
"epoch": 0.07,
"learning_rate": 0.00047130391173520563,
"loss": 3.1967,
"theoretical_loss": 4.3027873345196745,
"tokens_seen": 220528640
},
{
"epoch": 0.07,
"learning_rate": 0.0004712938816449348,
"loss": 3.4082,
"theoretical_loss": 4.302629774917049,
"tokens_seen": 220594176
},
{
"epoch": 0.07,
"learning_rate": 0.00047128385155466404,
"loss": 3.1634,
"theoretical_loss": 4.302472275218748,
"tokens_seen": 220659712
},
{
"epoch": 0.07,
"learning_rate": 0.00047127382146439317,
"loss": 3.3901,
"theoretical_loss": 4.302314835384214,
"tokens_seen": 220725248
},
{
"epoch": 0.07,
"learning_rate": 0.0004712637913741224,
"loss": 3.3387,
"theoretical_loss": 4.30215745537293,
"tokens_seen": 220790784
},
{
"epoch": 0.07,
"learning_rate": 0.00047125376128385153,
"loss": 3.2651,
"theoretical_loss": 4.302000135144416,
"tokens_seen": 220856320
},
{
"epoch": 0.07,
"learning_rate": 0.00047124373119358077,
"loss": 3.2082,
"theoretical_loss": 4.301842874658235,
"tokens_seen": 220921856
},
{
"epoch": 0.07,
"learning_rate": 0.00047123370110330995,
"loss": 3.2934,
"theoretical_loss": 4.301685673873987,
"tokens_seen": 220987392
},
{
"epoch": 0.07,
"learning_rate": 0.00047122367101303913,
"loss": 3.1771,
"theoretical_loss": 4.301528532751312,
"tokens_seen": 221052928
},
{
"epoch": 0.07,
"learning_rate": 0.0004712136409227683,
"loss": 3.3551,
"theoretical_loss": 4.301371451249888,
"tokens_seen": 221118464
},
{
"epoch": 0.07,
"objective/train/docs_used": 136371,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.877000093460083,
"objective/train/theoretical_loss": 4.301214429329433,
"objective/train/tokens_used": 241644000,
"theoretical_loss": 4.301214429329433,
"tokens_seen": 221184000
},
{
"epoch": 0.07,
"learning_rate": 0.00047120361083249755,
"loss": 3.4952,
"theoretical_loss": 4.301214429329433,
"tokens_seen": 221184000
},
{
"epoch": 0.07,
"learning_rate": 0.0004711935807422267,
"loss": 3.1406,
"theoretical_loss": 4.301057466949707,
"tokens_seen": 221249536
},
{
"epoch": 0.07,
"learning_rate": 0.0004711835506519559,
"loss": 3.1892,
"theoretical_loss": 4.300900564070504,
"tokens_seen": 221315072
},
{
"epoch": 0.07,
"learning_rate": 0.00047117352056168504,
"loss": 2.9775,
"theoretical_loss": 4.30074372065166,
"tokens_seen": 221380608
},
{
"epoch": 0.07,
"learning_rate": 0.00047116349047141427,
"loss": 3.3187,
"theoretical_loss": 4.300586936653049,
"tokens_seen": 221446144
},
{
"epoch": 0.07,
"learning_rate": 0.00047115346038114345,
"loss": 3.148,
"theoretical_loss": 4.300430212034587,
"tokens_seen": 221511680
},
{
"epoch": 0.07,
"learning_rate": 0.00047114343029087263,
"loss": 2.8728,
"theoretical_loss": 4.300273546756223,
"tokens_seen": 221577216
},
{
"epoch": 0.07,
"learning_rate": 0.0004711334002006018,
"loss": 3.1421,
"theoretical_loss": 4.300116940777951,
"tokens_seen": 221642752
},
{
"epoch": 0.07,
"learning_rate": 0.000471123370110331,
"loss": 3.1462,
"theoretical_loss": 4.299960394059799,
"tokens_seen": 221708288
},
{
"epoch": 0.07,
"learning_rate": 0.0004711133400200602,
"loss": 2.9714,
"theoretical_loss": 4.299803906561835,
"tokens_seen": 221773824
},
{
"epoch": 0.07,
"learning_rate": 0.0004711033099297894,
"loss": 3.278,
"theoretical_loss": 4.29964747824417,
"tokens_seen": 221839360
},
{
"epoch": 0.07,
"learning_rate": 0.00047109327983951854,
"loss": 3.1696,
"theoretical_loss": 4.299491109066947,
"tokens_seen": 221904896
},
{
"epoch": 0.07,
"learning_rate": 0.0004710832497492478,
"loss": 3.1112,
"theoretical_loss": 4.299334798990351,
"tokens_seen": 221970432
},
{
"epoch": 0.07,
"learning_rate": 0.0004710732196589769,
"loss": 3.1501,
"theoretical_loss": 4.2991785479746065,
"tokens_seen": 222035968
},
{
"epoch": 0.07,
"learning_rate": 0.00047106318956870614,
"loss": 2.7828,
"theoretical_loss": 4.299022355979974,
"tokens_seen": 222101504
},
{
"epoch": 0.07,
"learning_rate": 0.0004710531594784353,
"loss": 3.3247,
"theoretical_loss": 4.298866222966755,
"tokens_seen": 222167040
},
{
"epoch": 0.07,
"learning_rate": 0.0004710431293881645,
"loss": 3.1744,
"theoretical_loss": 4.298710148895286,
"tokens_seen": 222232576
},
{
"epoch": 0.07,
"learning_rate": 0.0004710330992978937,
"loss": 3.4915,
"theoretical_loss": 4.298554133725946,
"tokens_seen": 222298112
},
{
"epoch": 0.07,
"learning_rate": 0.0004710230692076229,
"loss": 3.1641,
"theoretical_loss": 4.298398177419149,
"tokens_seen": 222363648
},
{
"epoch": 0.07,
"learning_rate": 0.00047101303911735204,
"loss": 3.2386,
"theoretical_loss": 4.298242279935349,
"tokens_seen": 222429184
},
{
"epoch": 0.07,
"learning_rate": 0.0004710030090270813,
"loss": 3.305,
"theoretical_loss": 4.2980864412350375,
"tokens_seen": 222494720
},
{
"epoch": 0.07,
"learning_rate": 0.0004709929789368104,
"loss": 3.1451,
"theoretical_loss": 4.297930661278745,
"tokens_seen": 222560256
},
{
"epoch": 0.07,
"learning_rate": 0.00047098294884653964,
"loss": 3.031,
"theoretical_loss": 4.297774940027038,
"tokens_seen": 222625792
},
{
"epoch": 0.07,
"learning_rate": 0.0004709729187562688,
"loss": 3.2461,
"theoretical_loss": 4.297619277440523,
"tokens_seen": 222691328
},
{
"epoch": 0.07,
"learning_rate": 0.000470962888665998,
"loss": 3.1183,
"theoretical_loss": 4.297463673479846,
"tokens_seen": 222756864
},
{
"epoch": 0.07,
"objective/train/docs_used": 137650,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3066697120666504,
"objective/train/theoretical_loss": 4.297308128105687,
"objective/train/tokens_used": 243282400,
"theoretical_loss": 4.297308128105687,
"tokens_seen": 222822400
},
{
"epoch": 0.07,
"learning_rate": 0.0004709528585757272,
"loss": 3.3301,
"theoretical_loss": 4.297308128105687,
"tokens_seen": 222822400
},
{
"epoch": 0.07,
"learning_rate": 0.00047094282848545636,
"loss": 3.2327,
"theoretical_loss": 4.297152641278767,
"tokens_seen": 222887936
},
{
"epoch": 0.07,
"learning_rate": 0.00047093279839518555,
"loss": 3.0477,
"theoretical_loss": 4.296997212959842,
"tokens_seen": 222953472
},
{
"epoch": 0.07,
"learning_rate": 0.0004709227683049148,
"loss": 3.4307,
"theoretical_loss": 4.296841843109711,
"tokens_seen": 223019008
},
{
"epoch": 0.07,
"learning_rate": 0.0004709127382146439,
"loss": 3.1878,
"theoretical_loss": 4.296686531689204,
"tokens_seen": 223084544
},
{
"epoch": 0.07,
"learning_rate": 0.00047090270812437314,
"loss": 3.29,
"theoretical_loss": 4.296531278659193,
"tokens_seen": 223150080
},
{
"epoch": 0.07,
"learning_rate": 0.00047089267803410227,
"loss": 3.0173,
"theoretical_loss": 4.296376083980589,
"tokens_seen": 223215616
},
{
"epoch": 0.07,
"learning_rate": 0.0004708826479438315,
"loss": 3.3376,
"theoretical_loss": 4.296220947614337,
"tokens_seen": 223281152
},
{
"epoch": 0.07,
"learning_rate": 0.0004708726178535607,
"loss": 2.9857,
"theoretical_loss": 4.296065869521421,
"tokens_seen": 223346688
},
{
"epoch": 0.07,
"learning_rate": 0.00047086258776328987,
"loss": 3.5665,
"theoretical_loss": 4.295910849662862,
"tokens_seen": 223412224
},
{
"epoch": 0.07,
"learning_rate": 0.00047085255767301905,
"loss": 3.2103,
"theoretical_loss": 4.2957558879997215,
"tokens_seen": 223477760
},
{
"epoch": 0.07,
"learning_rate": 0.0004708425275827483,
"loss": 3.1934,
"theoretical_loss": 4.295600984493093,
"tokens_seen": 223543296
},
{
"epoch": 0.07,
"learning_rate": 0.0004708324974924774,
"loss": 2.9751,
"theoretical_loss": 4.295446139104112,
"tokens_seen": 223608832
},
{
"epoch": 0.07,
"learning_rate": 0.00047082246740220665,
"loss": 3.18,
"theoretical_loss": 4.295291351793951,
"tokens_seen": 223674368
},
{
"epoch": 0.07,
"learning_rate": 0.0004708124373119358,
"loss": 3.1672,
"theoretical_loss": 4.295136622523817,
"tokens_seen": 223739904
},
{
"epoch": 0.07,
"learning_rate": 0.000470802407221665,
"loss": 3.001,
"theoretical_loss": 4.294981951254956,
"tokens_seen": 223805440
},
{
"epoch": 0.07,
"learning_rate": 0.0004707923771313942,
"loss": 3.1423,
"theoretical_loss": 4.294827337948651,
"tokens_seen": 223870976
},
{
"epoch": 0.07,
"learning_rate": 0.00047078234704112337,
"loss": 3.0418,
"theoretical_loss": 4.294672782566224,
"tokens_seen": 223936512
},
{
"epoch": 0.07,
"learning_rate": 0.00047077231695085255,
"loss": 3.1785,
"theoretical_loss": 4.29451828506903,
"tokens_seen": 224002048
},
{
"epoch": 0.07,
"learning_rate": 0.00047076228686058173,
"loss": 3.2824,
"theoretical_loss": 4.294363845418465,
"tokens_seen": 224067584
},
{
"epoch": 0.07,
"learning_rate": 0.0004707522567703109,
"loss": 3.099,
"theoretical_loss": 4.29420946357596,
"tokens_seen": 224133120
},
{
"epoch": 0.07,
"learning_rate": 0.00047074222668004015,
"loss": 3.1006,
"theoretical_loss": 4.294055139502985,
"tokens_seen": 224198656
},
{
"epoch": 0.07,
"learning_rate": 0.0004707321965897693,
"loss": 3.2409,
"theoretical_loss": 4.293900873161043,
"tokens_seen": 224264192
},
{
"epoch": 0.07,
"learning_rate": 0.0004707221664994985,
"loss": 3.0344,
"theoretical_loss": 4.293746664511678,
"tokens_seen": 224329728
},
{
"epoch": 0.07,
"learning_rate": 0.00047071213640922775,
"loss": 2.9979,
"theoretical_loss": 4.293592513516469,
"tokens_seen": 224395264
},
{
"epoch": 0.07,
"objective/train/docs_used": 138458,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.281918525695801,
"objective/train/theoretical_loss": 4.293438420137031,
"objective/train/tokens_used": 244920800,
"theoretical_loss": 4.293438420137031,
"tokens_seen": 224460800
},
{
"epoch": 0.07,
"learning_rate": 0.0004707021063189569,
"loss": 3.1837,
"theoretical_loss": 4.293438420137031,
"tokens_seen": 224460800
},
{
"epoch": 0.07,
"learning_rate": 0.0004706920762286861,
"loss": 3.0455,
"theoretical_loss": 4.293284384335017,
"tokens_seen": 224526336
},
{
"epoch": 0.07,
"learning_rate": 0.00047068204613841524,
"loss": 3.2542,
"theoretical_loss": 4.293130406072118,
"tokens_seen": 224591872
},
{
"epoch": 0.07,
"learning_rate": 0.00047067201604814447,
"loss": 3.2776,
"theoretical_loss": 4.292976485310057,
"tokens_seen": 224657408
},
{
"epoch": 0.07,
"learning_rate": 0.00047066198595787365,
"loss": 3.1208,
"theoretical_loss": 4.2928226220106005,
"tokens_seen": 224722944
},
{
"epoch": 0.07,
"learning_rate": 0.00047065195586760283,
"loss": 3.1642,
"theoretical_loss": 4.292668816135545,
"tokens_seen": 224788480
},
{
"epoch": 0.07,
"learning_rate": 0.000470641925777332,
"loss": 3.3477,
"theoretical_loss": 4.292515067646727,
"tokens_seen": 224854016
},
{
"epoch": 0.07,
"learning_rate": 0.0004706318956870612,
"loss": 3.1525,
"theoretical_loss": 4.29236137650602,
"tokens_seen": 224919552
},
{
"epoch": 0.07,
"learning_rate": 0.0004706218655967904,
"loss": 3.203,
"theoretical_loss": 4.2922077426753305,
"tokens_seen": 224985088
},
{
"epoch": 0.07,
"learning_rate": 0.0004706118355065196,
"loss": 3.304,
"theoretical_loss": 4.292054166116605,
"tokens_seen": 225050624
},
{
"epoch": 0.07,
"learning_rate": 0.00047060180541624874,
"loss": 3.099,
"theoretical_loss": 4.291900646791825,
"tokens_seen": 225116160
},
{
"epoch": 0.07,
"learning_rate": 0.000470591775325978,
"loss": 3.4831,
"theoretical_loss": 4.2917471846630075,
"tokens_seen": 225181696
},
{
"epoch": 0.07,
"learning_rate": 0.0004705817452357071,
"loss": 3.3273,
"theoretical_loss": 4.291593779692207,
"tokens_seen": 225247232
},
{
"epoch": 0.07,
"learning_rate": 0.00047057171514543634,
"loss": 3.0792,
"theoretical_loss": 4.291440431841513,
"tokens_seen": 225312768
},
{
"epoch": 0.07,
"learning_rate": 0.0004705616850551655,
"loss": 3.1787,
"theoretical_loss": 4.291287141073053,
"tokens_seen": 225378304
},
{
"epoch": 0.07,
"learning_rate": 0.0004705516549648947,
"loss": 3.0933,
"theoretical_loss": 4.291133907348989,
"tokens_seen": 225443840
},
{
"epoch": 0.07,
"learning_rate": 0.0004705416248746239,
"loss": 3.2376,
"theoretical_loss": 4.29098073063152,
"tokens_seen": 225509376
},
{
"epoch": 0.07,
"learning_rate": 0.0004705315947843531,
"loss": 3.3523,
"theoretical_loss": 4.29082761088288,
"tokens_seen": 225574912
},
{
"epoch": 0.07,
"learning_rate": 0.00047052156469408224,
"loss": 3.032,
"theoretical_loss": 4.290674548065338,
"tokens_seen": 225640448
},
{
"epoch": 0.07,
"learning_rate": 0.0004705115346038115,
"loss": 3.0997,
"theoretical_loss": 4.290521542141203,
"tokens_seen": 225705984
},
{
"epoch": 0.07,
"learning_rate": 0.0004705015045135406,
"loss": 3.1698,
"theoretical_loss": 4.290368593072817,
"tokens_seen": 225771520
},
{
"epoch": 0.07,
"learning_rate": 0.00047049147442326984,
"loss": 3.2039,
"theoretical_loss": 4.290215700822556,
"tokens_seen": 225837056
},
{
"epoch": 0.07,
"learning_rate": 0.000470481444332999,
"loss": 3.1309,
"theoretical_loss": 4.290062865352837,
"tokens_seen": 225902592
},
{
"epoch": 0.07,
"learning_rate": 0.0004704714142427282,
"loss": 3.1847,
"theoretical_loss": 4.289910086626108,
"tokens_seen": 225968128
},
{
"epoch": 0.07,
"learning_rate": 0.0004704613841524574,
"loss": 2.9217,
"theoretical_loss": 4.289757364604855,
"tokens_seen": 226033664
},
{
"epoch": 0.07,
"objective/train/docs_used": 139641,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.6932528018951416,
"objective/train/theoretical_loss": 4.2896046992515995,
"objective/train/tokens_used": 246559200,
"theoretical_loss": 4.2896046992515995,
"tokens_seen": 226099200
},
{
"epoch": 0.07,
"learning_rate": 0.00047045135406218656,
"loss": 3.5236,
"theoretical_loss": 4.2896046992515995,
"tokens_seen": 226099200
},
{
"epoch": 0.07,
"learning_rate": 0.00047044132397191575,
"loss": 3.1005,
"theoretical_loss": 4.289452090528897,
"tokens_seen": 226164736
},
{
"epoch": 0.07,
"learning_rate": 0.000470431293881645,
"loss": 3.1343,
"theoretical_loss": 4.289299538399341,
"tokens_seen": 226230272
},
{
"epoch": 0.07,
"learning_rate": 0.0004704212637913741,
"loss": 3.3457,
"theoretical_loss": 4.28914704282556,
"tokens_seen": 226295808
},
{
"epoch": 0.07,
"learning_rate": 0.00047041123370110334,
"loss": 3.4157,
"theoretical_loss": 4.288994603770215,
"tokens_seen": 226361344
},
{
"epoch": 0.07,
"learning_rate": 0.00047040120361083247,
"loss": 3.2237,
"theoretical_loss": 4.288842221196007,
"tokens_seen": 226426880
},
{
"epoch": 0.07,
"learning_rate": 0.0004703911735205617,
"loss": 3.2044,
"theoretical_loss": 4.28868989506567,
"tokens_seen": 226492416
},
{
"epoch": 0.07,
"learning_rate": 0.0004703811434302909,
"loss": 3.2856,
"theoretical_loss": 4.288537625341974,
"tokens_seen": 226557952
},
{
"epoch": 0.07,
"learning_rate": 0.00047037111334002007,
"loss": 3.0679,
"theoretical_loss": 4.288385411987722,
"tokens_seen": 226623488
},
{
"epoch": 0.07,
"learning_rate": 0.00047036108324974925,
"loss": 3.4104,
"theoretical_loss": 4.288233254965755,
"tokens_seen": 226689024
},
{
"epoch": 0.07,
"learning_rate": 0.0004703510531594785,
"loss": 3.3151,
"theoretical_loss": 4.2880811542389505,
"tokens_seen": 226754560
},
{
"epoch": 0.07,
"learning_rate": 0.0004703410230692076,
"loss": 3.3593,
"theoretical_loss": 4.287929109770217,
"tokens_seen": 226820096
},
{
"epoch": 0.07,
"learning_rate": 0.00047033099297893685,
"loss": 2.8766,
"theoretical_loss": 4.287777121522501,
"tokens_seen": 226885632
},
{
"epoch": 0.07,
"learning_rate": 0.000470320962888666,
"loss": 3.209,
"theoretical_loss": 4.287625189458781,
"tokens_seen": 226951168
},
{
"epoch": 0.07,
"learning_rate": 0.0004703109327983952,
"loss": 3.1,
"theoretical_loss": 4.287473313542077,
"tokens_seen": 227016704
},
{
"epoch": 0.07,
"learning_rate": 0.0004703009027081244,
"loss": 3.3124,
"theoretical_loss": 4.287321493735438,
"tokens_seen": 227082240
},
{
"epoch": 0.07,
"learning_rate": 0.00047029087261785357,
"loss": 3.1342,
"theoretical_loss": 4.287169730001949,
"tokens_seen": 227147776
},
{
"epoch": 0.07,
"learning_rate": 0.00047028084252758275,
"loss": 3.1754,
"theoretical_loss": 4.287018022304733,
"tokens_seen": 227213312
},
{
"epoch": 0.07,
"learning_rate": 0.00047027081243731193,
"loss": 3.2797,
"theoretical_loss": 4.286866370606943,
"tokens_seen": 227278848
},
{
"epoch": 0.07,
"learning_rate": 0.0004702607823470411,
"loss": 3.3938,
"theoretical_loss": 4.286714774871772,
"tokens_seen": 227344384
},
{
"epoch": 0.07,
"learning_rate": 0.00047025075225677035,
"loss": 2.8743,
"theoretical_loss": 4.286563235062444,
"tokens_seen": 227409920
},
{
"epoch": 0.07,
"learning_rate": 0.0004702407221664995,
"loss": 3.2675,
"theoretical_loss": 4.28641175114222,
"tokens_seen": 227475456
},
{
"epoch": 0.07,
"learning_rate": 0.0004702306920762287,
"loss": 3.5072,
"theoretical_loss": 4.286260323074394,
"tokens_seen": 227540992
},
{
"epoch": 0.07,
"learning_rate": 0.00047022066198595784,
"loss": 3.385,
"theoretical_loss": 4.286108950822296,
"tokens_seen": 227606528
},
{
"epoch": 0.07,
"learning_rate": 0.0004702106318956871,
"loss": 3.2283,
"theoretical_loss": 4.285957634349289,
"tokens_seen": 227672064
},
{
"epoch": 0.07,
"objective/train/docs_used": 140376,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.265362501144409,
"objective/train/theoretical_loss": 4.285806373618774,
"objective/train/tokens_used": 248197600,
"theoretical_loss": 4.285806373618774,
"tokens_seen": 227737600
},
{
"epoch": 0.07,
"learning_rate": 0.00047020060180541626,
"loss": 2.999,
"theoretical_loss": 4.285806373618774,
"tokens_seen": 227737600
},
{
"epoch": 0.07,
"learning_rate": 0.00047019057171514544,
"loss": 3.1608,
"theoretical_loss": 4.285655168594182,
"tokens_seen": 227803136
},
{
"epoch": 0.07,
"learning_rate": 0.0004701805416248746,
"loss": 3.2796,
"theoretical_loss": 4.285504019238982,
"tokens_seen": 227868672
},
{
"epoch": 0.07,
"learning_rate": 0.00047017051153460385,
"loss": 3.4479,
"theoretical_loss": 4.285352925516676,
"tokens_seen": 227934208
},
{
"epoch": 0.07,
"learning_rate": 0.000470160481444333,
"loss": 3.3548,
"theoretical_loss": 4.2852018873908,
"tokens_seen": 227999744
},
{
"epoch": 0.07,
"learning_rate": 0.0004701504513540622,
"loss": 3.2998,
"theoretical_loss": 4.285050904824925,
"tokens_seen": 228065280
},
{
"epoch": 0.07,
"learning_rate": 0.00047014042126379134,
"loss": 3.1287,
"theoretical_loss": 4.284899977782658,
"tokens_seen": 228130816
},
{
"epoch": 0.07,
"learning_rate": 0.0004701303911735206,
"loss": 3.2708,
"theoretical_loss": 4.284749106227636,
"tokens_seen": 228196352
},
{
"epoch": 0.07,
"learning_rate": 0.00047012036108324976,
"loss": 3.1895,
"theoretical_loss": 4.284598290123535,
"tokens_seen": 228261888
},
{
"epoch": 0.07,
"learning_rate": 0.00047011033099297894,
"loss": 3.128,
"theoretical_loss": 4.284447529434061,
"tokens_seen": 228327424
},
{
"epoch": 0.07,
"learning_rate": 0.0004701003009027081,
"loss": 3.2905,
"theoretical_loss": 4.284296824122959,
"tokens_seen": 228392960
},
{
"epoch": 0.07,
"learning_rate": 0.0004700902708124373,
"loss": 3.152,
"theoretical_loss": 4.284146174154003,
"tokens_seen": 228458496
},
{
"epoch": 0.07,
"learning_rate": 0.0004700802407221665,
"loss": 3.4164,
"theoretical_loss": 4.283995579491004,
"tokens_seen": 228524032
},
{
"epoch": 0.07,
"learning_rate": 0.0004700702106318957,
"loss": 2.9542,
"theoretical_loss": 4.283845040097807,
"tokens_seen": 228589568
},
{
"epoch": 0.07,
"learning_rate": 0.00047006018054162484,
"loss": 3.1201,
"theoretical_loss": 4.28369455593829,
"tokens_seen": 228655104
},
{
"epoch": 0.07,
"learning_rate": 0.0004700501504513541,
"loss": 3.3299,
"theoretical_loss": 4.2835441269763646,
"tokens_seen": 228720640
},
{
"epoch": 0.07,
"learning_rate": 0.00047004012036108326,
"loss": 2.9468,
"theoretical_loss": 4.283393753175979,
"tokens_seen": 228786176
},
{
"epoch": 0.07,
"learning_rate": 0.00047003009027081244,
"loss": 3.3356,
"theoretical_loss": 4.283243434501112,
"tokens_seen": 228851712
},
{
"epoch": 0.07,
"learning_rate": 0.0004700200601805416,
"loss": 3.0321,
"theoretical_loss": 4.283093170915778,
"tokens_seen": 228917248
},
{
"epoch": 0.07,
"learning_rate": 0.0004700100300902708,
"loss": 3.143,
"theoretical_loss": 4.282942962384023,
"tokens_seen": 228982784
},
{
"epoch": 0.07,
"learning_rate": 0.00047,
"loss": 3.1754,
"theoretical_loss": 4.282792808869932,
"tokens_seen": 229048320
},
{
"epoch": 0.07,
"learning_rate": 0.0004699899699097292,
"loss": 3.1478,
"theoretical_loss": 4.282642710337618,
"tokens_seen": 229113856
},
{
"epoch": 0.07,
"learning_rate": 0.00046997993981945835,
"loss": 3.336,
"theoretical_loss": 4.28249266675123,
"tokens_seen": 229179392
},
{
"epoch": 0.07,
"learning_rate": 0.0004699699097291876,
"loss": 3.2775,
"theoretical_loss": 4.282342678074951,
"tokens_seen": 229244928
},
{
"epoch": 0.07,
"learning_rate": 0.00046995987963891676,
"loss": 3.2018,
"theoretical_loss": 4.2821927442729955,
"tokens_seen": 229310464
},
{
"epoch": 0.07,
"objective/train/docs_used": 141693,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1491169929504395,
"objective/train/theoretical_loss": 4.282042865309616,
"objective/train/tokens_used": 249836000,
"theoretical_loss": 4.282042865309616,
"tokens_seen": 229376000
},
{
"epoch": 0.07,
"learning_rate": 0.00046994984954864595,
"loss": 3.1859,
"theoretical_loss": 4.282042865309616,
"tokens_seen": 229376000
},
{
"epoch": 0.07,
"learning_rate": 0.0004699398194583752,
"loss": 3.3997,
"theoretical_loss": 4.281893041149093,
"tokens_seen": 229441536
},
{
"epoch": 0.07,
"learning_rate": 0.0004699297893681043,
"loss": 3.1647,
"theoretical_loss": 4.2817432717557455,
"tokens_seen": 229507072
},
{
"epoch": 0.07,
"learning_rate": 0.00046991975927783354,
"loss": 3.2627,
"theoretical_loss": 4.28159355709392,
"tokens_seen": 229572608
},
{
"epoch": 0.07,
"learning_rate": 0.00046990972918756267,
"loss": 3.1173,
"theoretical_loss": 4.281443897128004,
"tokens_seen": 229638144
},
{
"epoch": 0.07,
"learning_rate": 0.0004698996990972919,
"loss": 3.3119,
"theoretical_loss": 4.2812942918224115,
"tokens_seen": 229703680
},
{
"epoch": 0.07,
"learning_rate": 0.0004698896690070211,
"loss": 2.8788,
"theoretical_loss": 4.281144741141593,
"tokens_seen": 229769216
},
{
"epoch": 0.07,
"learning_rate": 0.00046987963891675027,
"loss": 3.2099,
"theoretical_loss": 4.280995245050032,
"tokens_seen": 229834752
},
{
"epoch": 0.07,
"learning_rate": 0.00046986960882647945,
"loss": 3.1133,
"theoretical_loss": 4.2808458035122445,
"tokens_seen": 229900288
},
{
"epoch": 0.07,
"learning_rate": 0.0004698595787362087,
"loss": 3.3331,
"theoretical_loss": 4.2806964164927805,
"tokens_seen": 229965824
},
{
"epoch": 0.07,
"learning_rate": 0.0004698495486459378,
"loss": 3.0517,
"theoretical_loss": 4.280547083956224,
"tokens_seen": 230031360
},
{
"epoch": 0.07,
"learning_rate": 0.00046983951855566705,
"loss": 3.1513,
"theoretical_loss": 4.280397805867188,
"tokens_seen": 230096896
},
{
"epoch": 0.07,
"learning_rate": 0.0004698294884653962,
"loss": 3.1371,
"theoretical_loss": 4.280248582190324,
"tokens_seen": 230162432
},
{
"epoch": 0.07,
"learning_rate": 0.0004698194583751254,
"loss": 3.1632,
"theoretical_loss": 4.280099412890312,
"tokens_seen": 230227968
},
{
"epoch": 0.07,
"learning_rate": 0.0004698094282848546,
"loss": 3.1009,
"theoretical_loss": 4.279950297931869,
"tokens_seen": 230293504
},
{
"epoch": 0.07,
"learning_rate": 0.00046979939819458377,
"loss": 3.414,
"theoretical_loss": 4.27980123727974,
"tokens_seen": 230359040
},
{
"epoch": 0.07,
"learning_rate": 0.00046978936810431295,
"loss": 3.1625,
"theoretical_loss": 4.279652230898709,
"tokens_seen": 230424576
},
{
"epoch": 0.07,
"learning_rate": 0.00046977933801404213,
"loss": 3.2313,
"theoretical_loss": 4.279503278753586,
"tokens_seen": 230490112
},
{
"epoch": 0.07,
"learning_rate": 0.0004697693079237713,
"loss": 3.3858,
"theoretical_loss": 4.27935438080922,
"tokens_seen": 230555648
},
{
"epoch": 0.07,
"learning_rate": 0.00046975927783350055,
"loss": 3.011,
"theoretical_loss": 4.27920553703049,
"tokens_seen": 230621184
},
{
"epoch": 0.07,
"learning_rate": 0.0004697492477432297,
"loss": 2.9844,
"theoretical_loss": 4.279056747382306,
"tokens_seen": 230686720
},
{
"epoch": 0.07,
"learning_rate": 0.0004697392176529589,
"loss": 3.197,
"theoretical_loss": 4.278908011829613,
"tokens_seen": 230752256
},
{
"epoch": 0.07,
"learning_rate": 0.00046972918756268804,
"loss": 3.1954,
"theoretical_loss": 4.27875933033739,
"tokens_seen": 230817792
},
{
"epoch": 0.07,
"learning_rate": 0.0004697191574724173,
"loss": 3.0489,
"theoretical_loss": 4.278610702870646,
"tokens_seen": 230883328
},
{
"epoch": 0.07,
"learning_rate": 0.00046970912738214646,
"loss": 3.2719,
"theoretical_loss": 4.278462129394423,
"tokens_seen": 230948864
},
{
"epoch": 0.07,
"objective/train/docs_used": 142407,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.702136993408203,
"objective/train/theoretical_loss": 4.278313609873795,
"objective/train/tokens_used": 251474400,
"theoretical_loss": 4.278313609873795,
"tokens_seen": 231014400
},
{
"epoch": 0.07,
"learning_rate": 0.00046969909729187564,
"loss": 3.3355,
"theoretical_loss": 4.278313609873795,
"tokens_seen": 231014400
},
{
"epoch": 0.07,
"learning_rate": 0.0004696890672016048,
"loss": 2.9092,
"theoretical_loss": 4.278165144273871,
"tokens_seen": 231079936
},
{
"epoch": 0.07,
"learning_rate": 0.00046967903711133405,
"loss": 3.2064,
"theoretical_loss": 4.27801673255979,
"tokens_seen": 231145472
},
{
"epoch": 0.07,
"learning_rate": 0.0004696690070210632,
"loss": 3.2198,
"theoretical_loss": 4.277868374696725,
"tokens_seen": 231211008
},
{
"epoch": 0.07,
"learning_rate": 0.0004696589769307924,
"loss": 3.1205,
"theoretical_loss": 4.277720070649879,
"tokens_seen": 231276544
},
{
"epoch": 0.07,
"learning_rate": 0.00046964894684052154,
"loss": 3.1119,
"theoretical_loss": 4.277571820384491,
"tokens_seen": 231342080
},
{
"epoch": 0.07,
"learning_rate": 0.0004696389167502508,
"loss": 3.2283,
"theoretical_loss": 4.277423623865829,
"tokens_seen": 231407616
},
{
"epoch": 0.07,
"learning_rate": 0.00046962888665997996,
"loss": 3.036,
"theoretical_loss": 4.277275481059195,
"tokens_seen": 231473152
},
{
"epoch": 0.07,
"learning_rate": 0.00046961885656970914,
"loss": 3.0813,
"theoretical_loss": 4.2771273919299215,
"tokens_seen": 231538688
},
{
"epoch": 0.07,
"learning_rate": 0.0004696088264794383,
"loss": 3.2151,
"theoretical_loss": 4.276979356443377,
"tokens_seen": 231604224
},
{
"epoch": 0.07,
"learning_rate": 0.0004695987963891675,
"loss": 2.992,
"theoretical_loss": 4.276831374564957,
"tokens_seen": 231669760
},
{
"epoch": 0.07,
"learning_rate": 0.0004695887662988967,
"loss": 3.1676,
"theoretical_loss": 4.276683446260093,
"tokens_seen": 231735296
},
{
"epoch": 0.07,
"learning_rate": 0.0004695787362086259,
"loss": 3.0428,
"theoretical_loss": 4.276535571494247,
"tokens_seen": 231800832
},
{
"epoch": 0.07,
"learning_rate": 0.00046956870611835505,
"loss": 2.979,
"theoretical_loss": 4.276387750232913,
"tokens_seen": 231866368
},
{
"epoch": 0.07,
"learning_rate": 0.0004695586760280843,
"loss": 3.0738,
"theoretical_loss": 4.276239982441617,
"tokens_seen": 231931904
},
{
"epoch": 0.07,
"learning_rate": 0.00046954864593781346,
"loss": 3.2279,
"theoretical_loss": 4.276092268085918,
"tokens_seen": 231997440
},
{
"epoch": 0.07,
"learning_rate": 0.00046953861584754264,
"loss": 3.1162,
"theoretical_loss": 4.275944607131406,
"tokens_seen": 232062976
},
{
"epoch": 0.07,
"learning_rate": 0.0004695285857572718,
"loss": 3.0761,
"theoretical_loss": 4.275796999543703,
"tokens_seen": 232128512
},
{
"epoch": 0.07,
"learning_rate": 0.000469518555667001,
"loss": 3.0807,
"theoretical_loss": 4.275649445288461,
"tokens_seen": 232194048
},
{
"epoch": 0.07,
"learning_rate": 0.0004695085255767302,
"loss": 3.1171,
"theoretical_loss": 4.275501944331367,
"tokens_seen": 232259584
},
{
"epoch": 0.07,
"learning_rate": 0.0004694984954864594,
"loss": 3.273,
"theoretical_loss": 4.275354496638139,
"tokens_seen": 232325120
},
{
"epoch": 0.07,
"learning_rate": 0.00046948846539618855,
"loss": 3.2311,
"theoretical_loss": 4.275207102174525,
"tokens_seen": 232390656
},
{
"epoch": 0.07,
"learning_rate": 0.0004694784353059178,
"loss": 3.3267,
"theoretical_loss": 4.275059760906305,
"tokens_seen": 232456192
},
{
"epoch": 0.07,
"learning_rate": 0.0004694684052156469,
"loss": 3.1457,
"theoretical_loss": 4.2749124727992935,
"tokens_seen": 232521728
},
{
"epoch": 0.07,
"learning_rate": 0.00046945837512537615,
"loss": 3.4693,
"theoretical_loss": 4.274765237819333,
"tokens_seen": 232587264
},
{
"epoch": 0.07,
"objective/train/docs_used": 143093,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.266288995742798,
"objective/train/theoretical_loss": 4.274618055932298,
"objective/train/tokens_used": 253112800,
"theoretical_loss": 4.274618055932298,
"tokens_seen": 232652800
},
{
"epoch": 0.07,
"learning_rate": 0.0004694483450351053,
"loss": 3.1205,
"theoretical_loss": 4.274618055932298,
"tokens_seen": 232652800
},
{
"epoch": 0.07,
"learning_rate": 0.0004694383149448345,
"loss": 3.3176,
"theoretical_loss": 4.2744709271040975,
"tokens_seen": 232718336
},
{
"epoch": 0.07,
"learning_rate": 0.0004694282848545637,
"loss": 3.1205,
"theoretical_loss": 4.27432385130067,
"tokens_seen": 232783872
},
{
"epoch": 0.07,
"learning_rate": 0.00046941825476429287,
"loss": 3.1899,
"theoretical_loss": 4.274176828487984,
"tokens_seen": 232849408
},
{
"epoch": 0.07,
"learning_rate": 0.00046940822467402205,
"loss": 3.2353,
"theoretical_loss": 4.2740298586320415,
"tokens_seen": 232914944
},
{
"epoch": 0.07,
"learning_rate": 0.0004693981945837513,
"loss": 3.1096,
"theoretical_loss": 4.273882941698876,
"tokens_seen": 232980480
},
{
"epoch": 0.07,
"learning_rate": 0.0004693881644934804,
"loss": 3.0017,
"theoretical_loss": 4.27373607765455,
"tokens_seen": 233046016
},
{
"epoch": 0.07,
"learning_rate": 0.00046937813440320965,
"loss": 3.0144,
"theoretical_loss": 4.2735892664651605,
"tokens_seen": 233111552
},
{
"epoch": 0.07,
"learning_rate": 0.00046936810431293883,
"loss": 3.2733,
"theoretical_loss": 4.273442508096833,
"tokens_seen": 233177088
},
{
"epoch": 0.07,
"learning_rate": 0.000469358074222668,
"loss": 3.1154,
"theoretical_loss": 4.273295802515726,
"tokens_seen": 233242624
},
{
"epoch": 0.07,
"learning_rate": 0.0004693480441323972,
"loss": 3.1081,
"theoretical_loss": 4.273149149688028,
"tokens_seen": 233308160
},
{
"epoch": 0.07,
"learning_rate": 0.0004693380140421264,
"loss": 3.0391,
"theoretical_loss": 4.27300254957996,
"tokens_seen": 233373696
},
{
"epoch": 0.07,
"learning_rate": 0.00046932798395185555,
"loss": 3.0994,
"theoretical_loss": 4.272856002157772,
"tokens_seen": 233439232
},
{
"epoch": 0.07,
"learning_rate": 0.0004693179538615848,
"loss": 3.0702,
"theoretical_loss": 4.272709507387748,
"tokens_seen": 233504768
},
{
"epoch": 0.07,
"learning_rate": 0.0004693079237713139,
"loss": 3.1512,
"theoretical_loss": 4.2725630652362,
"tokens_seen": 233570304
},
{
"epoch": 0.07,
"learning_rate": 0.00046929789368104315,
"loss": 2.9167,
"theoretical_loss": 4.272416675669473,
"tokens_seen": 233635840
},
{
"epoch": 0.07,
"learning_rate": 0.0004692878635907723,
"loss": 3.353,
"theoretical_loss": 4.272270338653942,
"tokens_seen": 233701376
},
{
"epoch": 0.07,
"learning_rate": 0.0004692778335005015,
"loss": 3.0815,
"theoretical_loss": 4.272124054156014,
"tokens_seen": 233766912
},
{
"epoch": 0.07,
"learning_rate": 0.0004692678034102307,
"loss": 3.1558,
"theoretical_loss": 4.271977822142125,
"tokens_seen": 233832448
},
{
"epoch": 0.07,
"learning_rate": 0.0004692577733199599,
"loss": 3.4225,
"theoretical_loss": 4.271831642578745,
"tokens_seen": 233897984
},
{
"epoch": 0.07,
"learning_rate": 0.00046924774322968906,
"loss": 3.0911,
"theoretical_loss": 4.27168551543237,
"tokens_seen": 233963520
},
{
"epoch": 0.07,
"learning_rate": 0.00046923771313941824,
"loss": 3.1895,
"theoretical_loss": 4.271539440669532,
"tokens_seen": 234029056
},
{
"epoch": 0.07,
"learning_rate": 0.0004692276830491474,
"loss": 3.223,
"theoretical_loss": 4.27139341825679,
"tokens_seen": 234094592
},
{
"epoch": 0.07,
"learning_rate": 0.00046921765295887666,
"loss": 3.4205,
"theoretical_loss": 4.271247448160736,
"tokens_seen": 234160128
},
{
"epoch": 0.07,
"learning_rate": 0.00046920762286860584,
"loss": 3.1894,
"theoretical_loss": 4.27110153034799,
"tokens_seen": 234225664
},
{
"epoch": 0.07,
"objective/train/docs_used": 144323,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.5741994380950928,
"objective/train/theoretical_loss": 4.270955664785207,
"objective/train/tokens_used": 254751200,
"theoretical_loss": 4.270955664785207,
"tokens_seen": 234291200
},
{
"epoch": 0.07,
"learning_rate": 0.000469197592778335,
"loss": 3.2851,
"theoretical_loss": 4.270955664785207,
"tokens_seen": 234291200
},
{
"epoch": 0.07,
"learning_rate": 0.00046918756268806425,
"loss": 3.207,
"theoretical_loss": 4.2708098514390676,
"tokens_seen": 234356736
},
{
"epoch": 0.07,
"learning_rate": 0.0004691775325977934,
"loss": 3.0235,
"theoretical_loss": 4.270664090276286,
"tokens_seen": 234422272
},
{
"epoch": 0.07,
"learning_rate": 0.0004691675025075226,
"loss": 3.2393,
"theoretical_loss": 4.2705183812636065,
"tokens_seen": 234487808
},
{
"epoch": 0.07,
"learning_rate": 0.00046915747241725174,
"loss": 3.1636,
"theoretical_loss": 4.270372724367803,
"tokens_seen": 234553344
},
{
"epoch": 0.07,
"learning_rate": 0.000469147442326981,
"loss": 3.073,
"theoretical_loss": 4.270227119555681,
"tokens_seen": 234618880
},
{
"epoch": 0.07,
"learning_rate": 0.00046913741223671016,
"loss": 3.199,
"theoretical_loss": 4.270081566794076,
"tokens_seen": 234684416
},
{
"epoch": 0.07,
"learning_rate": 0.00046912738214643934,
"loss": 3.0461,
"theoretical_loss": 4.269936066049852,
"tokens_seen": 234749952
},
{
"epoch": 0.07,
"learning_rate": 0.0004691173520561685,
"loss": 3.0538,
"theoretical_loss": 4.269790617289907,
"tokens_seen": 234815488
},
{
"epoch": 0.07,
"learning_rate": 0.0004691073219658977,
"loss": 3.0901,
"theoretical_loss": 4.269645220481166,
"tokens_seen": 234881024
},
{
"epoch": 0.07,
"learning_rate": 0.0004690972918756269,
"loss": 3.1677,
"theoretical_loss": 4.269499875590587,
"tokens_seen": 234946560
},
{
"epoch": 0.07,
"learning_rate": 0.0004690872617853561,
"loss": 3.1924,
"theoretical_loss": 4.269354582585156,
"tokens_seen": 235012096
},
{
"epoch": 0.07,
"learning_rate": 0.00046907723169508525,
"loss": 2.9523,
"theoretical_loss": 4.269209341431889,
"tokens_seen": 235077632
},
{
"epoch": 0.07,
"learning_rate": 0.0004690672016048145,
"loss": 2.9465,
"theoretical_loss": 4.269064152097835,
"tokens_seen": 235143168
},
{
"epoch": 0.07,
"learning_rate": 0.00046905717151454366,
"loss": 3.2225,
"theoretical_loss": 4.26891901455007,
"tokens_seen": 235208704
},
{
"epoch": 0.07,
"learning_rate": 0.00046904714142427284,
"loss": 3.3442,
"theoretical_loss": 4.268773928755701,
"tokens_seen": 235274240
},
{
"epoch": 0.07,
"learning_rate": 0.000469037111334002,
"loss": 2.9913,
"theoretical_loss": 4.268628894681868,
"tokens_seen": 235339776
},
{
"epoch": 0.07,
"learning_rate": 0.0004690270812437312,
"loss": 3.1332,
"theoretical_loss": 4.268483912295735,
"tokens_seen": 235405312
},
{
"epoch": 0.07,
"learning_rate": 0.0004690170511534604,
"loss": 3.1891,
"theoretical_loss": 4.268338981564502,
"tokens_seen": 235470848
},
{
"epoch": 0.07,
"learning_rate": 0.0004690070210631896,
"loss": 3.0551,
"theoretical_loss": 4.268194102455395,
"tokens_seen": 235536384
},
{
"epoch": 0.07,
"learning_rate": 0.00046899699097291875,
"loss": 3.1104,
"theoretical_loss": 4.26804927493567,
"tokens_seen": 235601920
},
{
"epoch": 0.07,
"learning_rate": 0.000468986960882648,
"loss": 3.1895,
"theoretical_loss": 4.267904498972618,
"tokens_seen": 235667456
},
{
"epoch": 0.07,
"learning_rate": 0.0004689769307923771,
"loss": 3.115,
"theoretical_loss": 4.267759774533552,
"tokens_seen": 235732992
},
{
"epoch": 0.07,
"learning_rate": 0.00046896690070210635,
"loss": 3.2606,
"theoretical_loss": 4.267615101585821,
"tokens_seen": 235798528
},
{
"epoch": 0.07,
"learning_rate": 0.00046895687061183553,
"loss": 3.1975,
"theoretical_loss": 4.267470480096801,
"tokens_seen": 235864064
},
{
"epoch": 0.07,
"objective/train/docs_used": 145084,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.02962589263916,
"objective/train/theoretical_loss": 4.267325910033897,
"objective/train/tokens_used": 256389600,
"theoretical_loss": 4.267325910033897,
"tokens_seen": 235929600
},
{
"epoch": 0.07,
"learning_rate": 0.0004689468405215647,
"loss": 3.2423,
"theoretical_loss": 4.267325910033897,
"tokens_seen": 235929600
},
{
"epoch": 0.07,
"learning_rate": 0.0004689368104312939,
"loss": 3.0465,
"theoretical_loss": 4.267181391364547,
"tokens_seen": 235995136
},
{
"epoch": 0.07,
"learning_rate": 0.00046892678034102307,
"loss": 3.0515,
"theoretical_loss": 4.267036924056215,
"tokens_seen": 236060672
},
{
"epoch": 0.07,
"learning_rate": 0.00046891675025075225,
"loss": 3.1894,
"theoretical_loss": 4.266892508076397,
"tokens_seen": 236126208
},
{
"epoch": 0.07,
"learning_rate": 0.0004689067201604815,
"loss": 3.3846,
"theoretical_loss": 4.266748143392617,
"tokens_seen": 236191744
},
{
"epoch": 0.07,
"learning_rate": 0.0004688966900702106,
"loss": 2.8453,
"theoretical_loss": 4.26660382997243,
"tokens_seen": 236257280
},
{
"epoch": 0.07,
"learning_rate": 0.00046888665997993985,
"loss": 3.2638,
"theoretical_loss": 4.26645956778342,
"tokens_seen": 236322816
},
{
"epoch": 0.07,
"learning_rate": 0.00046887662988966903,
"loss": 3.2298,
"theoretical_loss": 4.2663153567932,
"tokens_seen": 236388352
},
{
"epoch": 0.07,
"learning_rate": 0.0004688665997993982,
"loss": 3.4072,
"theoretical_loss": 4.266171196969412,
"tokens_seen": 236453888
},
{
"epoch": 0.07,
"learning_rate": 0.0004688565697091274,
"loss": 3.3305,
"theoretical_loss": 4.2660270882797295,
"tokens_seen": 236519424
},
{
"epoch": 0.07,
"learning_rate": 0.0004688465396188566,
"loss": 3.1997,
"theoretical_loss": 4.265883030691853,
"tokens_seen": 236584960
},
{
"epoch": 0.07,
"learning_rate": 0.00046883650952858575,
"loss": 3.0699,
"theoretical_loss": 4.265739024173515,
"tokens_seen": 236650496
},
{
"epoch": 0.07,
"learning_rate": 0.000468826479438315,
"loss": 3.1803,
"theoretical_loss": 4.265595068692473,
"tokens_seen": 236716032
},
{
"epoch": 0.07,
"learning_rate": 0.0004688164493480441,
"loss": 3.254,
"theoretical_loss": 4.26545116421652,
"tokens_seen": 236781568
},
{
"epoch": 0.07,
"learning_rate": 0.00046880641925777335,
"loss": 2.9222,
"theoretical_loss": 4.265307310713471,
"tokens_seen": 236847104
},
{
"epoch": 0.07,
"learning_rate": 0.0004687963891675025,
"loss": 3.0569,
"theoretical_loss": 4.2651635081511765,
"tokens_seen": 236912640
},
{
"epoch": 0.07,
"learning_rate": 0.0004687863590772317,
"loss": 3.1851,
"theoretical_loss": 4.265019756497512,
"tokens_seen": 236978176
},
{
"epoch": 0.07,
"learning_rate": 0.0004687763289869609,
"loss": 3.0481,
"theoretical_loss": 4.264876055720386,
"tokens_seen": 237043712
},
{
"epoch": 0.07,
"learning_rate": 0.0004687662988966901,
"loss": 3.2352,
"theoretical_loss": 4.264732405787731,
"tokens_seen": 237109248
},
{
"epoch": 0.07,
"learning_rate": 0.00046875626880641926,
"loss": 2.9571,
"theoretical_loss": 4.264588806667513,
"tokens_seen": 237174784
},
{
"epoch": 0.07,
"learning_rate": 0.00046874623871614844,
"loss": 3.07,
"theoretical_loss": 4.264445258327724,
"tokens_seen": 237240320
},
{
"epoch": 0.07,
"learning_rate": 0.0004687362086258776,
"loss": 3.1563,
"theoretical_loss": 4.264301760736389,
"tokens_seen": 237305856
},
{
"epoch": 0.07,
"learning_rate": 0.00046872617853560686,
"loss": 3.3152,
"theoretical_loss": 4.264158313861557,
"tokens_seen": 237371392
},
{
"epoch": 0.07,
"learning_rate": 0.000468716148445336,
"loss": 3.1966,
"theoretical_loss": 4.264014917671309,
"tokens_seen": 237436928
},
{
"epoch": 0.07,
"learning_rate": 0.0004687061183550652,
"loss": 3.0981,
"theoretical_loss": 4.2638715721337554,
"tokens_seen": 237502464
},
{
"epoch": 0.07,
"objective/train/docs_used": 146518,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.969031572341919,
"objective/train/theoretical_loss": 4.263728277217032,
"objective/train/tokens_used": 258028000,
"theoretical_loss": 4.263728277217032,
"tokens_seen": 237568000
},
{
"epoch": 0.07,
"learning_rate": 0.0004686960882647944,
"loss": 2.995,
"theoretical_loss": 4.263728277217032,
"tokens_seen": 237568000
},
{
"epoch": 0.07,
"learning_rate": 0.0004686860581745236,
"loss": 2.7096,
"theoretical_loss": 4.263585032889306,
"tokens_seen": 237633536
},
{
"epoch": 0.07,
"learning_rate": 0.00046867602808425276,
"loss": 3.2779,
"theoretical_loss": 4.263441839118776,
"tokens_seen": 237699072
},
{
"epoch": 0.07,
"learning_rate": 0.00046866599799398194,
"loss": 3.237,
"theoretical_loss": 4.2632986958736625,
"tokens_seen": 237764608
},
{
"epoch": 0.07,
"learning_rate": 0.0004686559679037111,
"loss": 3.1594,
"theoretical_loss": 4.263155603122221,
"tokens_seen": 237830144
},
{
"epoch": 0.07,
"learning_rate": 0.00046864593781344036,
"loss": 3.1277,
"theoretical_loss": 4.263012560832733,
"tokens_seen": 237895680
},
{
"epoch": 0.07,
"learning_rate": 0.0004686359077231695,
"loss": 3.4282,
"theoretical_loss": 4.262869568973508,
"tokens_seen": 237961216
},
{
"epoch": 0.07,
"learning_rate": 0.0004686258776328987,
"loss": 3.1725,
"theoretical_loss": 4.262726627512886,
"tokens_seen": 238026752
},
{
"epoch": 0.07,
"learning_rate": 0.00046861584754262785,
"loss": 3.1387,
"theoretical_loss": 4.262583736419234,
"tokens_seen": 238092288
},
{
"epoch": 0.07,
"learning_rate": 0.0004686058174523571,
"loss": 3.2668,
"theoretical_loss": 4.26244089566095,
"tokens_seen": 238157824
},
{
"epoch": 0.07,
"learning_rate": 0.00046859578736208626,
"loss": 3.023,
"theoretical_loss": 4.262298105206456,
"tokens_seen": 238223360
},
{
"epoch": 0.07,
"learning_rate": 0.00046858575727181545,
"loss": 3.4639,
"theoretical_loss": 4.262155365024207,
"tokens_seen": 238288896
},
{
"epoch": 0.07,
"learning_rate": 0.0004685757271815446,
"loss": 3.1628,
"theoretical_loss": 4.262012675082685,
"tokens_seen": 238354432
},
{
"epoch": 0.07,
"learning_rate": 0.0004685656970912738,
"loss": 3.0802,
"theoretical_loss": 4.261870035350399,
"tokens_seen": 238419968
},
{
"epoch": 0.07,
"learning_rate": 0.000468555667001003,
"loss": 2.9962,
"theoretical_loss": 4.261727445795888,
"tokens_seen": 238485504
},
{
"epoch": 0.07,
"learning_rate": 0.0004685456369107322,
"loss": 3.2431,
"theoretical_loss": 4.26158490638772,
"tokens_seen": 238551040
},
{
"epoch": 0.07,
"learning_rate": 0.00046853560682046135,
"loss": 3.1662,
"theoretical_loss": 4.261442417094488,
"tokens_seen": 238616576
},
{
"epoch": 0.07,
"learning_rate": 0.0004685255767301906,
"loss": 3.1555,
"theoretical_loss": 4.261299977884816,
"tokens_seen": 238682112
},
{
"epoch": 0.07,
"learning_rate": 0.00046851554663991977,
"loss": 2.8705,
"theoretical_loss": 4.2611575887273565,
"tokens_seen": 238747648
},
{
"epoch": 0.07,
"learning_rate": 0.00046850551654964895,
"loss": 2.9495,
"theoretical_loss": 4.261015249590789,
"tokens_seen": 238813184
},
{
"epoch": 0.07,
"learning_rate": 0.00046849548645937813,
"loss": 3.0884,
"theoretical_loss": 4.260872960443822,
"tokens_seen": 238878720
},
{
"epoch": 0.07,
"learning_rate": 0.0004684854563691073,
"loss": 3.192,
"theoretical_loss": 4.260730721255191,
"tokens_seen": 238944256
},
{
"epoch": 0.07,
"learning_rate": 0.0004684754262788365,
"loss": 3.1981,
"theoretical_loss": 4.260588531993662,
"tokens_seen": 239009792
},
{
"epoch": 0.07,
"learning_rate": 0.00046846539618856573,
"loss": 2.9505,
"theoretical_loss": 4.260446392628026,
"tokens_seen": 239075328
},
{
"epoch": 0.07,
"learning_rate": 0.0004684553660982949,
"loss": 3.3252,
"theoretical_loss": 4.2603043031271035,
"tokens_seen": 239140864
},
{
"epoch": 0.07,
"objective/train/docs_used": 147039,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.313131332397461,
"objective/train/theoretical_loss": 4.260162263459744,
"objective/train/tokens_used": 259666400,
"theoretical_loss": 4.260162263459744,
"tokens_seen": 239206400
},
{
"epoch": 0.07,
"learning_rate": 0.0004684453360080241,
"loss": 3.0978,
"theoretical_loss": 4.260162263459744,
"tokens_seen": 239206400
},
{
"epoch": 0.07,
"learning_rate": 0.00046843530591775327,
"loss": 2.9198,
"theoretical_loss": 4.260020273594824,
"tokens_seen": 239271936
},
{
"epoch": 0.07,
"learning_rate": 0.00046842527582748245,
"loss": 3.025,
"theoretical_loss": 4.259878333501247,
"tokens_seen": 239337472
},
{
"epoch": 0.07,
"learning_rate": 0.0004684152457372117,
"loss": 3.0117,
"theoretical_loss": 4.259736443147946,
"tokens_seen": 239403008
},
{
"epoch": 0.07,
"learning_rate": 0.0004684052156469408,
"loss": 3.3059,
"theoretical_loss": 4.259594602503881,
"tokens_seen": 239468544
},
{
"epoch": 0.07,
"learning_rate": 0.00046839518555667005,
"loss": 3.4276,
"theoretical_loss": 4.259452811538041,
"tokens_seen": 239534080
},
{
"epoch": 0.07,
"learning_rate": 0.00046838515546639923,
"loss": 3.3361,
"theoretical_loss": 4.259311070219441,
"tokens_seen": 239599616
},
{
"epoch": 0.07,
"learning_rate": 0.0004683751253761284,
"loss": 3.3203,
"theoretical_loss": 4.259169378517125,
"tokens_seen": 239665152
},
{
"epoch": 0.07,
"learning_rate": 0.0004683650952858576,
"loss": 3.2315,
"theoretical_loss": 4.259027736400165,
"tokens_seen": 239730688
},
{
"epoch": 0.07,
"learning_rate": 0.0004683550651955868,
"loss": 3.014,
"theoretical_loss": 4.258886143837661,
"tokens_seen": 239796224
},
{
"epoch": 0.07,
"learning_rate": 0.00046834503510531595,
"loss": 3.2752,
"theoretical_loss": 4.258744600798739,
"tokens_seen": 239861760
},
{
"epoch": 0.07,
"learning_rate": 0.0004683350050150452,
"loss": 3.4321,
"theoretical_loss": 4.2586031072525525,
"tokens_seen": 239927296
},
{
"epoch": 0.07,
"learning_rate": 0.0004683249749247743,
"loss": 2.9211,
"theoretical_loss": 4.258461663168285,
"tokens_seen": 239992832
},
{
"epoch": 0.07,
"learning_rate": 0.00046831494483450355,
"loss": 3.0035,
"theoretical_loss": 4.258320268515147,
"tokens_seen": 240058368
},
{
"epoch": 0.07,
"learning_rate": 0.0004683049147442327,
"loss": 2.9131,
"theoretical_loss": 4.258178923262376,
"tokens_seen": 240123904
},
{
"epoch": 0.07,
"learning_rate": 0.0004682948846539619,
"loss": 3.3793,
"theoretical_loss": 4.258037627379235,
"tokens_seen": 240189440
},
{
"epoch": 0.07,
"learning_rate": 0.0004682848545636911,
"loss": 3.182,
"theoretical_loss": 4.257896380835018,
"tokens_seen": 240254976
},
{
"epoch": 0.07,
"learning_rate": 0.0004682748244734203,
"loss": 3.0911,
"theoretical_loss": 4.257755183599045,
"tokens_seen": 240320512
},
{
"epoch": 0.07,
"learning_rate": 0.00046826479438314946,
"loss": 3.1443,
"theoretical_loss": 4.257614035640662,
"tokens_seen": 240386048
},
{
"epoch": 0.07,
"learning_rate": 0.00046825476429287864,
"loss": 3.3931,
"theoretical_loss": 4.257472936929246,
"tokens_seen": 240451584
},
{
"epoch": 0.07,
"learning_rate": 0.0004682447342026078,
"loss": 3.2777,
"theoretical_loss": 4.257331887434198,
"tokens_seen": 240517120
},
{
"epoch": 0.07,
"learning_rate": 0.00046823470411233706,
"loss": 3.3189,
"theoretical_loss": 4.257190887124946,
"tokens_seen": 240582656
},
{
"epoch": 0.07,
"learning_rate": 0.0004682246740220662,
"loss": 3.3554,
"theoretical_loss": 4.25704993597095,
"tokens_seen": 240648192
},
{
"epoch": 0.07,
"learning_rate": 0.0004682146439317954,
"loss": 3.1868,
"theoretical_loss": 4.256909033941691,
"tokens_seen": 240713728
},
{
"epoch": 0.07,
"learning_rate": 0.0004682046138415246,
"loss": 3.3277,
"theoretical_loss": 4.256768181006683,
"tokens_seen": 240779264
},
{
"epoch": 0.07,
"objective/train/docs_used": 148118,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0634944438934326,
"objective/train/theoretical_loss": 4.2566273771354615,
"objective/train/tokens_used": 261304800,
"theoretical_loss": 4.2566273771354615,
"tokens_seen": 240844800
},
{
"epoch": 0.07,
"learning_rate": 0.0004681945837512538,
"loss": 3.1663,
"theoretical_loss": 4.2566273771354615,
"tokens_seen": 240844800
},
{
"epoch": 0.07,
"learning_rate": 0.00046818455366098296,
"loss": 3.3725,
"theoretical_loss": 4.256486622297595,
"tokens_seen": 240910336
},
{
"epoch": 0.07,
"learning_rate": 0.00046817452357071214,
"loss": 3.4629,
"theoretical_loss": 4.256345916462674,
"tokens_seen": 240975872
},
{
"epoch": 0.07,
"learning_rate": 0.0004681644934804413,
"loss": 3.216,
"theoretical_loss": 4.256205259600321,
"tokens_seen": 241041408
},
{
"epoch": 0.07,
"learning_rate": 0.00046815446339017056,
"loss": 3.2409,
"theoretical_loss": 4.256064651680182,
"tokens_seen": 241106944
},
{
"epoch": 0.07,
"learning_rate": 0.0004681444332998997,
"loss": 3.3066,
"theoretical_loss": 4.255924092671931,
"tokens_seen": 241172480
},
{
"epoch": 0.07,
"learning_rate": 0.0004681344032096289,
"loss": 3.2428,
"theoretical_loss": 4.255783582545269,
"tokens_seen": 241238016
},
{
"epoch": 0.07,
"learning_rate": 0.00046812437311935805,
"loss": 3.2391,
"theoretical_loss": 4.255643121269924,
"tokens_seen": 241303552
},
{
"epoch": 0.07,
"learning_rate": 0.0004681143430290873,
"loss": 3.1302,
"theoretical_loss": 4.255502708815651,
"tokens_seen": 241369088
},
{
"epoch": 0.07,
"learning_rate": 0.00046810431293881646,
"loss": 3.1376,
"theoretical_loss": 4.255362345152234,
"tokens_seen": 241434624
},
{
"epoch": 0.07,
"learning_rate": 0.00046809428284854565,
"loss": 3.2276,
"theoretical_loss": 4.255222030249479,
"tokens_seen": 241500160
},
{
"epoch": 0.07,
"learning_rate": 0.0004680842527582748,
"loss": 3.1871,
"theoretical_loss": 4.255081764077224,
"tokens_seen": 241565696
},
{
"epoch": 0.07,
"learning_rate": 0.000468074222668004,
"loss": 3.02,
"theoretical_loss": 4.25494154660533,
"tokens_seen": 241631232
},
{
"epoch": 0.07,
"learning_rate": 0.0004680641925777332,
"loss": 3.2228,
"theoretical_loss": 4.254801377803689,
"tokens_seen": 241696768
},
{
"epoch": 0.07,
"learning_rate": 0.0004680541624874624,
"loss": 3.198,
"theoretical_loss": 4.254661257642215,
"tokens_seen": 241762304
},
{
"epoch": 0.07,
"learning_rate": 0.00046804413239719155,
"loss": 3.2133,
"theoretical_loss": 4.254521186090852,
"tokens_seen": 241827840
},
{
"epoch": 0.07,
"learning_rate": 0.0004680341023069208,
"loss": 3.2564,
"theoretical_loss": 4.254381163119568,
"tokens_seen": 241893376
},
{
"epoch": 0.07,
"learning_rate": 0.00046802407221664997,
"loss": 3.1465,
"theoretical_loss": 4.254241188698361,
"tokens_seen": 241958912
},
{
"epoch": 0.07,
"learning_rate": 0.00046801404212637915,
"loss": 3.3137,
"theoretical_loss": 4.2541012627972545,
"tokens_seen": 242024448
},
{
"epoch": 0.07,
"learning_rate": 0.00046800401203610833,
"loss": 3.148,
"theoretical_loss": 4.2539613853862965,
"tokens_seen": 242089984
},
{
"epoch": 0.07,
"learning_rate": 0.0004679939819458375,
"loss": 3.1473,
"theoretical_loss": 4.253821556435565,
"tokens_seen": 242155520
},
{
"epoch": 0.07,
"learning_rate": 0.0004679839518555667,
"loss": 2.9928,
"theoretical_loss": 4.253681775915161,
"tokens_seen": 242221056
},
{
"epoch": 0.07,
"learning_rate": 0.00046797392176529593,
"loss": 3.2938,
"theoretical_loss": 4.253542043795215,
"tokens_seen": 242286592
},
{
"epoch": 0.07,
"learning_rate": 0.00046796389167502505,
"loss": 3.1909,
"theoretical_loss": 4.253402360045882,
"tokens_seen": 242352128
},
{
"epoch": 0.07,
"learning_rate": 0.0004679538615847543,
"loss": 3.3872,
"theoretical_loss": 4.253262724637346,
"tokens_seen": 242417664
},
{
"epoch": 0.07,
"objective/train/docs_used": 148826,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4665045738220215,
"objective/train/theoretical_loss": 4.253123137539814,
"objective/train/tokens_used": 262943200,
"theoretical_loss": 4.253123137539814,
"tokens_seen": 242483200
},
{
"epoch": 0.07,
"learning_rate": 0.0004679438314944834,
"loss": 3.1621,
"theoretical_loss": 4.253123137539814,
"tokens_seen": 242483200
},
{
"epoch": 0.07,
"learning_rate": 0.00046793380140421265,
"loss": 2.9992,
"theoretical_loss": 4.252983598723521,
"tokens_seen": 242548736
},
{
"epoch": 0.07,
"learning_rate": 0.00046792377131394183,
"loss": 3.2264,
"theoretical_loss": 4.25284410815873,
"tokens_seen": 242614272
},
{
"epoch": 0.07,
"learning_rate": 0.000467913741223671,
"loss": 3.0202,
"theoretical_loss": 4.2527046658157275,
"tokens_seen": 242679808
},
{
"epoch": 0.07,
"learning_rate": 0.0004679037111334002,
"loss": 3.2755,
"theoretical_loss": 4.252565271664828,
"tokens_seen": 242745344
},
{
"epoch": 0.07,
"learning_rate": 0.00046789368104312943,
"loss": 3.2883,
"theoretical_loss": 4.252425925676373,
"tokens_seen": 242810880
},
{
"epoch": 0.07,
"learning_rate": 0.00046788365095285856,
"loss": 2.821,
"theoretical_loss": 4.252286627820727,
"tokens_seen": 242876416
},
{
"epoch": 0.07,
"learning_rate": 0.0004678736208625878,
"loss": 3.4768,
"theoretical_loss": 4.252147378068285,
"tokens_seen": 242941952
},
{
"epoch": 0.07,
"learning_rate": 0.0004678635907723169,
"loss": 3.2938,
"theoretical_loss": 4.252008176389465,
"tokens_seen": 243007488
},
{
"epoch": 0.07,
"learning_rate": 0.00046785356068204616,
"loss": 3.0992,
"theoretical_loss": 4.251869022754712,
"tokens_seen": 243073024
},
{
"epoch": 0.07,
"learning_rate": 0.00046784353059177534,
"loss": 3.1465,
"theoretical_loss": 4.251729917134498,
"tokens_seen": 243138560
},
{
"epoch": 0.07,
"learning_rate": 0.0004678335005015045,
"loss": 3.3141,
"theoretical_loss": 4.251590859499322,
"tokens_seen": 243204096
},
{
"epoch": 0.07,
"learning_rate": 0.0004678234704112337,
"loss": 3.0572,
"theoretical_loss": 4.251451849819704,
"tokens_seen": 243269632
},
{
"epoch": 0.07,
"learning_rate": 0.0004678134403209629,
"loss": 3.1641,
"theoretical_loss": 4.251312888066197,
"tokens_seen": 243335168
},
{
"epoch": 0.07,
"learning_rate": 0.00046780341023069206,
"loss": 2.8932,
"theoretical_loss": 4.251173974209375,
"tokens_seen": 243400704
},
{
"epoch": 0.07,
"learning_rate": 0.0004677933801404213,
"loss": 2.9821,
"theoretical_loss": 4.251035108219839,
"tokens_seen": 243466240
},
{
"epoch": 0.07,
"learning_rate": 0.0004677833500501504,
"loss": 3.248,
"theoretical_loss": 4.250896290068218,
"tokens_seen": 243531776
},
{
"epoch": 0.07,
"learning_rate": 0.00046777331995987966,
"loss": 3.1376,
"theoretical_loss": 4.250757519725165,
"tokens_seen": 243597312
},
{
"epoch": 0.07,
"learning_rate": 0.0004677632898696088,
"loss": 3.236,
"theoretical_loss": 4.25061879716136,
"tokens_seen": 243662848
},
{
"epoch": 0.07,
"learning_rate": 0.000467753259779338,
"loss": 2.9136,
"theoretical_loss": 4.250480122347507,
"tokens_seen": 243728384
},
{
"epoch": 0.07,
"learning_rate": 0.0004677432296890672,
"loss": 2.9435,
"theoretical_loss": 4.250341495254337,
"tokens_seen": 243793920
},
{
"epoch": 0.07,
"learning_rate": 0.0004677331995987964,
"loss": 2.9687,
"theoretical_loss": 4.250202915852608,
"tokens_seen": 243859456
},
{
"epoch": 0.07,
"learning_rate": 0.00046772316950852556,
"loss": 3.1614,
"theoretical_loss": 4.250064384113102,
"tokens_seen": 243924992
},
{
"epoch": 0.07,
"learning_rate": 0.0004677131394182548,
"loss": 2.939,
"theoretical_loss": 4.249925900006627,
"tokens_seen": 243990528
},
{
"epoch": 0.07,
"learning_rate": 0.000467703109327984,
"loss": 3.036,
"theoretical_loss": 4.249787463504019,
"tokens_seen": 244056064
},
{
"epoch": 0.07,
"objective/train/docs_used": 150000,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0193285942077637,
"objective/train/theoretical_loss": 4.249649074576134,
"objective/train/tokens_used": 264581600,
"theoretical_loss": 4.249649074576134,
"tokens_seen": 244121600
},
{
"epoch": 0.07,
"learning_rate": 0.00046769307923771316,
"loss": 3.1174,
"theoretical_loss": 4.249649074576134,
"tokens_seen": 244121600
},
{
"epoch": 0.07,
"learning_rate": 0.00046768304914744234,
"loss": 3.278,
"theoretical_loss": 4.249510733193862,
"tokens_seen": 244187136
},
{
"epoch": 0.07,
"learning_rate": 0.0004676730190571715,
"loss": 2.9943,
"theoretical_loss": 4.249372439328111,
"tokens_seen": 244252672
},
{
"epoch": 0.07,
"learning_rate": 0.00046766298896690076,
"loss": 3.3433,
"theoretical_loss": 4.249234192949818,
"tokens_seen": 244318208
},
{
"epoch": 0.07,
"learning_rate": 0.0004676529588766299,
"loss": 3.2905,
"theoretical_loss": 4.249095994029947,
"tokens_seen": 244383744
},
{
"epoch": 0.07,
"learning_rate": 0.0004676429287863591,
"loss": 3.0341,
"theoretical_loss": 4.248957842539484,
"tokens_seen": 244449280
},
{
"epoch": 0.07,
"learning_rate": 0.00046763289869608825,
"loss": 3.3165,
"theoretical_loss": 4.248819738449442,
"tokens_seen": 244514816
},
{
"epoch": 0.07,
"learning_rate": 0.0004676228686058175,
"loss": 3.0861,
"theoretical_loss": 4.2486816817308615,
"tokens_seen": 244580352
},
{
"epoch": 0.07,
"learning_rate": 0.00046761283851554666,
"loss": 3.2345,
"theoretical_loss": 4.248543672354805,
"tokens_seen": 244645888
},
{
"epoch": 0.07,
"learning_rate": 0.00046760280842527585,
"loss": 3.2232,
"theoretical_loss": 4.248405710292364,
"tokens_seen": 244711424
},
{
"epoch": 0.07,
"learning_rate": 0.000467592778335005,
"loss": 3.1468,
"theoretical_loss": 4.248267795514652,
"tokens_seen": 244776960
},
{
"epoch": 0.07,
"learning_rate": 0.0004675827482447342,
"loss": 3.122,
"theoretical_loss": 4.248129927992808,
"tokens_seen": 244842496
},
{
"epoch": 0.07,
"learning_rate": 0.0004675727181544634,
"loss": 3.1999,
"theoretical_loss": 4.247992107698002,
"tokens_seen": 244908032
},
{
"epoch": 0.07,
"learning_rate": 0.0004675626880641926,
"loss": 3.1818,
"theoretical_loss": 4.247854334601421,
"tokens_seen": 244973568
},
{
"epoch": 0.07,
"learning_rate": 0.00046755265797392175,
"loss": 3.2209,
"theoretical_loss": 4.247716608674283,
"tokens_seen": 245039104
},
{
"epoch": 0.07,
"learning_rate": 0.000467542627883651,
"loss": 3.1481,
"theoretical_loss": 4.247578929887829,
"tokens_seen": 245104640
},
{
"epoch": 0.07,
"learning_rate": 0.00046753259779338017,
"loss": 3.0031,
"theoretical_loss": 4.247441298213326,
"tokens_seen": 245170176
},
{
"epoch": 0.07,
"learning_rate": 0.00046752256770310935,
"loss": 2.998,
"theoretical_loss": 4.247303713622067,
"tokens_seen": 245235712
},
{
"epoch": 0.07,
"learning_rate": 0.00046751253761283853,
"loss": 3.1993,
"theoretical_loss": 4.247166176085367,
"tokens_seen": 245301248
},
{
"epoch": 0.07,
"learning_rate": 0.0004675025075225677,
"loss": 3.1806,
"theoretical_loss": 4.247028685574569,
"tokens_seen": 245366784
},
{
"epoch": 0.07,
"learning_rate": 0.0004674924774322969,
"loss": 3.3216,
"theoretical_loss": 4.246891242061041,
"tokens_seen": 245432320
},
{
"epoch": 0.07,
"learning_rate": 0.00046748244734202613,
"loss": 3.0241,
"theoretical_loss": 4.246753845516174,
"tokens_seen": 245497856
},
{
"epoch": 0.07,
"learning_rate": 0.00046747241725175525,
"loss": 3.3789,
"theoretical_loss": 4.246616495911388,
"tokens_seen": 245563392
},
{
"epoch": 0.07,
"learning_rate": 0.0004674623871614845,
"loss": 3.1574,
"theoretical_loss": 4.246479193218123,
"tokens_seen": 245628928
},
{
"epoch": 0.07,
"learning_rate": 0.0004674523570712136,
"loss": 3.1504,
"theoretical_loss": 4.246341937407848,
"tokens_seen": 245694464
},
{
"epoch": 0.07,
"objective/train/docs_used": 150657,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3028979301452637,
"objective/train/theoretical_loss": 4.246204728452055,
"objective/train/tokens_used": 266220000,
"theoretical_loss": 4.246204728452055,
"tokens_seen": 245760000
},
{
"epoch": 0.07,
"learning_rate": 0.00046744232698094285,
"loss": 3.3196,
"theoretical_loss": 4.246204728452055,
"tokens_seen": 245760000
},
{
"epoch": 0.07,
"learning_rate": 0.00046743229689067203,
"loss": 3.2711,
"theoretical_loss": 4.246067566322259,
"tokens_seen": 245825536
},
{
"epoch": 0.07,
"learning_rate": 0.0004674222668004012,
"loss": 3.3492,
"theoretical_loss": 4.245930450990007,
"tokens_seen": 245891072
},
{
"epoch": 0.07,
"learning_rate": 0.0004674122367101304,
"loss": 3.2023,
"theoretical_loss": 4.245793382426861,
"tokens_seen": 245956608
},
{
"epoch": 0.07,
"learning_rate": 0.00046740220661985963,
"loss": 3.0581,
"theoretical_loss": 4.245656360604417,
"tokens_seen": 246022144
},
{
"epoch": 0.07,
"learning_rate": 0.00046739217652958876,
"loss": 3.2156,
"theoretical_loss": 4.24551938549429,
"tokens_seen": 246087680
},
{
"epoch": 0.07,
"learning_rate": 0.000467382146439318,
"loss": 3.1543,
"theoretical_loss": 4.2453824570681205,
"tokens_seen": 246153216
},
{
"epoch": 0.07,
"learning_rate": 0.0004673721163490471,
"loss": 2.9836,
"theoretical_loss": 4.245245575297577,
"tokens_seen": 246218752
},
{
"epoch": 0.07,
"learning_rate": 0.00046736208625877636,
"loss": 3.0194,
"theoretical_loss": 4.2451087401543495,
"tokens_seen": 246284288
},
{
"epoch": 0.07,
"learning_rate": 0.00046735205616850554,
"loss": 3.1133,
"theoretical_loss": 4.244971951610154,
"tokens_seen": 246349824
},
{
"epoch": 0.07,
"learning_rate": 0.0004673420260782347,
"loss": 3.1389,
"theoretical_loss": 4.24483520963673,
"tokens_seen": 246415360
},
{
"epoch": 0.07,
"learning_rate": 0.0004673319959879639,
"loss": 3.3422,
"theoretical_loss": 4.244698514205844,
"tokens_seen": 246480896
},
{
"epoch": 0.07,
"learning_rate": 0.0004673219658976931,
"loss": 3.0803,
"theoretical_loss": 4.244561865289285,
"tokens_seen": 246546432
},
{
"epoch": 0.07,
"learning_rate": 0.00046731193580742226,
"loss": 3.1255,
"theoretical_loss": 4.244425262858867,
"tokens_seen": 246611968
},
{
"epoch": 0.07,
"learning_rate": 0.0004673019057171515,
"loss": 3.152,
"theoretical_loss": 4.2442887068864295,
"tokens_seen": 246677504
},
{
"epoch": 0.07,
"learning_rate": 0.0004672918756268806,
"loss": 3.2079,
"theoretical_loss": 4.244152197343835,
"tokens_seen": 246743040
},
{
"epoch": 0.07,
"learning_rate": 0.00046728184553660986,
"loss": 3.4062,
"theoretical_loss": 4.244015734202973,
"tokens_seen": 246808576
},
{
"epoch": 0.07,
"learning_rate": 0.000467271815446339,
"loss": 3.1603,
"theoretical_loss": 4.243879317435755,
"tokens_seen": 246874112
},
{
"epoch": 0.07,
"learning_rate": 0.0004672617853560682,
"loss": 3.1124,
"theoretical_loss": 4.243742947014117,
"tokens_seen": 246939648
},
{
"epoch": 0.07,
"learning_rate": 0.0004672517552657974,
"loss": 2.9212,
"theoretical_loss": 4.243606622910021,
"tokens_seen": 247005184
},
{
"epoch": 0.07,
"learning_rate": 0.0004672417251755266,
"loss": 3.074,
"theoretical_loss": 4.243470345095453,
"tokens_seen": 247070720
},
{
"epoch": 0.07,
"learning_rate": 0.00046723169508525576,
"loss": 3.1613,
"theoretical_loss": 4.2433341135424225,
"tokens_seen": 247136256
},
{
"epoch": 0.07,
"learning_rate": 0.000467221664994985,
"loss": 3.1584,
"theoretical_loss": 4.243197928222964,
"tokens_seen": 247201792
},
{
"epoch": 0.07,
"learning_rate": 0.0004672116349047141,
"loss": 3.2257,
"theoretical_loss": 4.243061789109136,
"tokens_seen": 247267328
},
{
"epoch": 0.07,
"learning_rate": 0.00046720160481444336,
"loss": 3.1837,
"theoretical_loss": 4.242925696173021,
"tokens_seen": 247332864
},
{
"epoch": 0.07,
"objective/train/docs_used": 151862,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.255047082901001,
"objective/train/theoretical_loss": 4.2427896493867285,
"objective/train/tokens_used": 267858400,
"theoretical_loss": 4.2427896493867285,
"tokens_seen": 247398400
},
{
"epoch": 0.07,
"learning_rate": 0.0004671915747241725,
"loss": 2.9149,
"theoretical_loss": 4.2427896493867285,
"tokens_seen": 247398400
},
{
"epoch": 0.07,
"learning_rate": 0.0004671815446339017,
"loss": 2.8987,
"theoretical_loss": 4.242653648722387,
"tokens_seen": 247463936
},
{
"epoch": 0.08,
"learning_rate": 0.0004671715145436309,
"loss": 2.9504,
"theoretical_loss": 4.242517694152154,
"tokens_seen": 247529472
},
{
"epoch": 0.08,
"learning_rate": 0.0004671614844533601,
"loss": 3.1881,
"theoretical_loss": 4.24238178564821,
"tokens_seen": 247595008
},
{
"epoch": 0.08,
"learning_rate": 0.00046715145436308927,
"loss": 2.9594,
"theoretical_loss": 4.242245923182756,
"tokens_seen": 247660544
},
{
"epoch": 0.08,
"learning_rate": 0.00046714142427281845,
"loss": 3.1327,
"theoretical_loss": 4.242110106728022,
"tokens_seen": 247726080
},
{
"epoch": 0.08,
"learning_rate": 0.00046713139418254763,
"loss": 3.0429,
"theoretical_loss": 4.241974336256261,
"tokens_seen": 247791616
},
{
"epoch": 0.08,
"learning_rate": 0.00046712136409227686,
"loss": 3.1163,
"theoretical_loss": 4.241838611739748,
"tokens_seen": 247857152
},
{
"epoch": 0.08,
"learning_rate": 0.000467111334002006,
"loss": 3.2067,
"theoretical_loss": 4.241702933150783,
"tokens_seen": 247922688
},
{
"epoch": 0.08,
"learning_rate": 0.0004671013039117352,
"loss": 3.3482,
"theoretical_loss": 4.241567300461693,
"tokens_seen": 247988224
},
{
"epoch": 0.08,
"learning_rate": 0.00046709127382146435,
"loss": 3.3143,
"theoretical_loss": 4.241431713644823,
"tokens_seen": 248053760
},
{
"epoch": 0.08,
"learning_rate": 0.0004670812437311936,
"loss": 3.2176,
"theoretical_loss": 4.241296172672547,
"tokens_seen": 248119296
},
{
"epoch": 0.08,
"learning_rate": 0.00046707121364092277,
"loss": 3.1752,
"theoretical_loss": 4.24116067751726,
"tokens_seen": 248184832
},
{
"epoch": 0.08,
"learning_rate": 0.00046706118355065195,
"loss": 3.2537,
"theoretical_loss": 4.241025228151383,
"tokens_seen": 248250368
},
{
"epoch": 0.08,
"learning_rate": 0.00046705115346038113,
"loss": 3.4286,
"theoretical_loss": 4.24088982454736,
"tokens_seen": 248315904
},
{
"epoch": 0.08,
"learning_rate": 0.00046704112337011037,
"loss": 3.248,
"theoretical_loss": 4.240754466677659,
"tokens_seen": 248381440
},
{
"epoch": 0.08,
"learning_rate": 0.0004670310932798395,
"loss": 2.9887,
"theoretical_loss": 4.240619154514771,
"tokens_seen": 248446976
},
{
"epoch": 0.08,
"learning_rate": 0.00046702106318956873,
"loss": 3.2964,
"theoretical_loss": 4.240483888031212,
"tokens_seen": 248512512
},
{
"epoch": 0.08,
"learning_rate": 0.00046701103309929786,
"loss": 2.9216,
"theoretical_loss": 4.240348667199521,
"tokens_seen": 248578048
},
{
"epoch": 0.08,
"learning_rate": 0.0004670010030090271,
"loss": 2.9462,
"theoretical_loss": 4.240213491992261,
"tokens_seen": 248643584
},
{
"epoch": 0.08,
"learning_rate": 0.0004669909729187563,
"loss": 3.2691,
"theoretical_loss": 4.240078362382019,
"tokens_seen": 248709120
},
{
"epoch": 0.08,
"learning_rate": 0.00046698094282848545,
"loss": 2.8617,
"theoretical_loss": 4.239943278341404,
"tokens_seen": 248774656
},
{
"epoch": 0.08,
"learning_rate": 0.00046697091273821464,
"loss": 2.9596,
"theoretical_loss": 4.239808239843052,
"tokens_seen": 248840192
},
{
"epoch": 0.08,
"learning_rate": 0.0004669608826479438,
"loss": 2.9612,
"theoretical_loss": 4.239673246859619,
"tokens_seen": 248905728
},
{
"epoch": 0.08,
"learning_rate": 0.00046695085255767305,
"loss": 2.9767,
"theoretical_loss": 4.239538299363788,
"tokens_seen": 248971264
},
{
"epoch": 0.08,
"objective/train/docs_used": 152583,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3142855167388916,
"objective/train/theoretical_loss": 4.239403397328261,
"objective/train/tokens_used": 269496800,
"theoretical_loss": 4.239403397328261,
"tokens_seen": 249036800
},
{
"epoch": 0.08,
"learning_rate": 0.00046694082246740223,
"loss": 3.3554,
"theoretical_loss": 4.239403397328261,
"tokens_seen": 249036800
},
{
"epoch": 0.08,
"learning_rate": 0.0004669307923771314,
"loss": 3.1367,
"theoretical_loss": 4.239268540725769,
"tokens_seen": 249102336
},
{
"epoch": 0.08,
"learning_rate": 0.0004669207622868606,
"loss": 3.0328,
"theoretical_loss": 4.239133729529064,
"tokens_seen": 249167872
},
{
"epoch": 0.08,
"learning_rate": 0.00046691073219658983,
"loss": 3.1881,
"theoretical_loss": 4.2389989637109196,
"tokens_seen": 249233408
},
{
"epoch": 0.08,
"learning_rate": 0.00046690070210631896,
"loss": 3.1403,
"theoretical_loss": 4.2388642432441355,
"tokens_seen": 249298944
},
{
"epoch": 0.08,
"learning_rate": 0.0004668906720160482,
"loss": 3.099,
"theoretical_loss": 4.238729568101535,
"tokens_seen": 249364480
},
{
"epoch": 0.08,
"learning_rate": 0.0004668806419257773,
"loss": 3.2886,
"theoretical_loss": 4.238594938255963,
"tokens_seen": 249430016
},
{
"epoch": 0.08,
"learning_rate": 0.00046687061183550656,
"loss": 2.9976,
"theoretical_loss": 4.2384603536802885,
"tokens_seen": 249495552
},
{
"epoch": 0.08,
"learning_rate": 0.00046686058174523574,
"loss": 3.1615,
"theoretical_loss": 4.238325814347404,
"tokens_seen": 249561088
},
{
"epoch": 0.08,
"learning_rate": 0.0004668505516549649,
"loss": 3.1596,
"theoretical_loss": 4.238191320230227,
"tokens_seen": 249626624
},
{
"epoch": 0.08,
"learning_rate": 0.0004668405215646941,
"loss": 3.2094,
"theoretical_loss": 4.238056871301695,
"tokens_seen": 249692160
},
{
"epoch": 0.08,
"learning_rate": 0.0004668304914744233,
"loss": 3.0536,
"theoretical_loss": 4.237922467534771,
"tokens_seen": 249757696
},
{
"epoch": 0.08,
"learning_rate": 0.00046682046138415246,
"loss": 2.9296,
"theoretical_loss": 4.237788108902441,
"tokens_seen": 249823232
},
{
"epoch": 0.08,
"learning_rate": 0.0004668104312938817,
"loss": 3.278,
"theoretical_loss": 4.237653795377714,
"tokens_seen": 249888768
},
{
"epoch": 0.08,
"learning_rate": 0.0004668004012036108,
"loss": 3.0784,
"theoretical_loss": 4.237519526933622,
"tokens_seen": 249954304
},
{
"epoch": 0.08,
"learning_rate": 0.00046679037111334006,
"loss": 3.3284,
"theoretical_loss": 4.2373853035432205,
"tokens_seen": 250019840
},
{
"epoch": 0.08,
"learning_rate": 0.0004667803410230692,
"loss": 3.1476,
"theoretical_loss": 4.237251125179588,
"tokens_seen": 250085376
},
{
"epoch": 0.08,
"learning_rate": 0.0004667703109327984,
"loss": 3.4263,
"theoretical_loss": 4.237116991815826,
"tokens_seen": 250150912
},
{
"epoch": 0.08,
"learning_rate": 0.0004667602808425276,
"loss": 3.3711,
"theoretical_loss": 4.23698290342506,
"tokens_seen": 250216448
},
{
"epoch": 0.08,
"learning_rate": 0.0004667502507522568,
"loss": 3.1352,
"theoretical_loss": 4.236848859980437,
"tokens_seen": 250281984
},
{
"epoch": 0.08,
"learning_rate": 0.00046674022066198596,
"loss": 3.1904,
"theoretical_loss": 4.23671486145513,
"tokens_seen": 250347520
},
{
"epoch": 0.08,
"learning_rate": 0.0004667301905717152,
"loss": 3.1398,
"theoretical_loss": 4.236580907822331,
"tokens_seen": 250413056
},
{
"epoch": 0.08,
"learning_rate": 0.0004667201604814443,
"loss": 3.2989,
"theoretical_loss": 4.236446999055257,
"tokens_seen": 250478592
},
{
"epoch": 0.08,
"learning_rate": 0.00046671013039117356,
"loss": 3.215,
"theoretical_loss": 4.2363131351271495,
"tokens_seen": 250544128
},
{
"epoch": 0.08,
"learning_rate": 0.0004667001003009027,
"loss": 3.0471,
"theoretical_loss": 4.2361793160112695,
"tokens_seen": 250609664
},
{
"epoch": 0.08,
"objective/train/docs_used": 153242,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2777795791625977,
"objective/train/theoretical_loss": 4.236045541680905,
"objective/train/tokens_used": 271135200,
"theoretical_loss": 4.236045541680905,
"tokens_seen": 250675200
},
{
"epoch": 0.08,
"learning_rate": 0.0004666900702106319,
"loss": 3.1679,
"theoretical_loss": 4.236045541680905,
"tokens_seen": 250675200
},
{
"epoch": 0.08,
"learning_rate": 0.0004666800401203611,
"loss": 3.0121,
"theoretical_loss": 4.235911812109363,
"tokens_seen": 250740736
},
{
"epoch": 0.08,
"learning_rate": 0.0004666700100300903,
"loss": 3.1682,
"theoretical_loss": 4.235778127269976,
"tokens_seen": 250806272
},
{
"epoch": 0.08,
"learning_rate": 0.00046665997993981947,
"loss": 3.236,
"theoretical_loss": 4.235644487136098,
"tokens_seen": 250871808
},
{
"epoch": 0.08,
"learning_rate": 0.00046664994984954865,
"loss": 3.1782,
"theoretical_loss": 4.235510891681108,
"tokens_seen": 250937344
},
{
"epoch": 0.08,
"learning_rate": 0.00046663991975927783,
"loss": 2.8634,
"theoretical_loss": 4.235377340878404,
"tokens_seen": 251002880
},
{
"epoch": 0.08,
"learning_rate": 0.00046662988966900706,
"loss": 3.3324,
"theoretical_loss": 4.23524383470141,
"tokens_seen": 251068416
},
{
"epoch": 0.08,
"learning_rate": 0.0004666198595787362,
"loss": 3.346,
"theoretical_loss": 4.235110373123572,
"tokens_seen": 251133952
},
{
"epoch": 0.08,
"learning_rate": 0.00046660982948846543,
"loss": 3.2462,
"theoretical_loss": 4.2349769561183574,
"tokens_seen": 251199488
},
{
"epoch": 0.08,
"learning_rate": 0.00046659979939819455,
"loss": 3.1068,
"theoretical_loss": 4.2348435836592575,
"tokens_seen": 251265024
},
{
"epoch": 0.08,
"learning_rate": 0.0004665897693079238,
"loss": 3.1695,
"theoretical_loss": 4.234710255719786,
"tokens_seen": 251330560
},
{
"epoch": 0.08,
"learning_rate": 0.00046657973921765297,
"loss": 3.1471,
"theoretical_loss": 4.234576972273481,
"tokens_seen": 251396096
},
{
"epoch": 0.08,
"learning_rate": 0.00046656970912738215,
"loss": 3.0517,
"theoretical_loss": 4.234443733293899,
"tokens_seen": 251461632
},
{
"epoch": 0.08,
"learning_rate": 0.00046655967903711133,
"loss": 3.1296,
"theoretical_loss": 4.234310538754624,
"tokens_seen": 251527168
},
{
"epoch": 0.08,
"learning_rate": 0.00046654964894684057,
"loss": 3.2316,
"theoretical_loss": 4.2341773886292575,
"tokens_seen": 251592704
},
{
"epoch": 0.08,
"learning_rate": 0.0004665396188565697,
"loss": 3.0754,
"theoretical_loss": 4.234044282891429,
"tokens_seen": 251658240
},
{
"epoch": 0.08,
"learning_rate": 0.00046652958876629893,
"loss": 3.4376,
"theoretical_loss": 4.233911221514787,
"tokens_seen": 251723776
},
{
"epoch": 0.08,
"learning_rate": 0.00046651955867602806,
"loss": 3.1209,
"theoretical_loss": 4.233778204473002,
"tokens_seen": 251789312
},
{
"epoch": 0.08,
"learning_rate": 0.0004665095285857573,
"loss": 3.054,
"theoretical_loss": 4.23364523173977,
"tokens_seen": 251854848
},
{
"epoch": 0.08,
"learning_rate": 0.0004664994984954865,
"loss": 3.1497,
"theoretical_loss": 4.233512303288807,
"tokens_seen": 251920384
},
{
"epoch": 0.08,
"learning_rate": 0.00046648946840521565,
"loss": 3.243,
"theoretical_loss": 4.233379419093851,
"tokens_seen": 251985920
},
{
"epoch": 0.08,
"learning_rate": 0.00046647943831494484,
"loss": 2.9777,
"theoretical_loss": 4.233246579128666,
"tokens_seen": 252051456
},
{
"epoch": 0.08,
"learning_rate": 0.000466469408224674,
"loss": 3.0244,
"theoretical_loss": 4.233113783367033,
"tokens_seen": 252116992
},
{
"epoch": 0.08,
"learning_rate": 0.0004664593781344032,
"loss": 2.8659,
"theoretical_loss": 4.232981031782761,
"tokens_seen": 252182528
},
{
"epoch": 0.08,
"learning_rate": 0.00046644934804413243,
"loss": 3.0599,
"theoretical_loss": 4.232848324349677,
"tokens_seen": 252248064
},
{
"epoch": 0.08,
"objective/train/docs_used": 154565,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2938530445098877,
"objective/train/theoretical_loss": 4.232715661041632,
"objective/train/tokens_used": 272773600,
"theoretical_loss": 4.232715661041632,
"tokens_seen": 252313600
},
{
"epoch": 0.08,
"learning_rate": 0.00046643931795386156,
"loss": 3.1579,
"theoretical_loss": 4.232715661041632,
"tokens_seen": 252313600
},
{
"epoch": 0.08,
"learning_rate": 0.0004664292878635908,
"loss": 2.8977,
"theoretical_loss": 4.232583041832499,
"tokens_seen": 252379136
},
{
"epoch": 0.08,
"learning_rate": 0.0004664192577733199,
"loss": 2.9725,
"theoretical_loss": 4.232450466696174,
"tokens_seen": 252444672
},
{
"epoch": 0.08,
"learning_rate": 0.00046640922768304916,
"loss": 3.247,
"theoretical_loss": 4.2323179356065745,
"tokens_seen": 252510208
},
{
"epoch": 0.08,
"learning_rate": 0.00046639919759277834,
"loss": 3.0137,
"theoretical_loss": 4.23218544853764,
"tokens_seen": 252575744
},
{
"epoch": 0.08,
"learning_rate": 0.0004663891675025075,
"loss": 3.1024,
"theoretical_loss": 4.232053005463333,
"tokens_seen": 252641280
},
{
"epoch": 0.08,
"learning_rate": 0.0004663791374122367,
"loss": 3.1656,
"theoretical_loss": 4.231920606357638,
"tokens_seen": 252706816
},
{
"epoch": 0.08,
"learning_rate": 0.00046636910732196594,
"loss": 3.2556,
"theoretical_loss": 4.231788251194559,
"tokens_seen": 252772352
},
{
"epoch": 0.08,
"learning_rate": 0.00046635907723169506,
"loss": 3.2828,
"theoretical_loss": 4.231655939948127,
"tokens_seen": 252837888
},
{
"epoch": 0.08,
"learning_rate": 0.0004663490471414243,
"loss": 3.0292,
"theoretical_loss": 4.231523672592392,
"tokens_seen": 252903424
},
{
"epoch": 0.08,
"learning_rate": 0.0004663390170511534,
"loss": 3.2209,
"theoretical_loss": 4.231391449101425,
"tokens_seen": 252968960
},
{
"epoch": 0.08,
"learning_rate": 0.00046632898696088266,
"loss": 3.3026,
"theoretical_loss": 4.231259269449322,
"tokens_seen": 253034496
},
{
"epoch": 0.08,
"learning_rate": 0.00046631895687061184,
"loss": 3.2373,
"theoretical_loss": 4.231127133610198,
"tokens_seen": 253100032
},
{
"epoch": 0.08,
"learning_rate": 0.000466308926780341,
"loss": 3.1833,
"theoretical_loss": 4.230995041558194,
"tokens_seen": 253165568
},
{
"epoch": 0.08,
"learning_rate": 0.0004662988966900702,
"loss": 3.0795,
"theoretical_loss": 4.230862993267468,
"tokens_seen": 253231104
},
{
"epoch": 0.08,
"learning_rate": 0.0004662888665997994,
"loss": 3.1932,
"theoretical_loss": 4.230730988712205,
"tokens_seen": 253296640
},
{
"epoch": 0.08,
"learning_rate": 0.00046627883650952857,
"loss": 3.2634,
"theoretical_loss": 4.230599027866606,
"tokens_seen": 253362176
},
{
"epoch": 0.08,
"learning_rate": 0.0004662688064192578,
"loss": 3.3007,
"theoretical_loss": 4.2304671107048994,
"tokens_seen": 253427712
},
{
"epoch": 0.08,
"learning_rate": 0.00046625877632898693,
"loss": 3.1446,
"theoretical_loss": 4.2303352372013325,
"tokens_seen": 253493248
},
{
"epoch": 0.08,
"learning_rate": 0.00046624874623871616,
"loss": 3.2354,
"theoretical_loss": 4.230203407330176,
"tokens_seen": 253558784
},
{
"epoch": 0.08,
"learning_rate": 0.00046623871614844535,
"loss": 3.2521,
"theoretical_loss": 4.230071621065721,
"tokens_seen": 253624320
},
{
"epoch": 0.08,
"learning_rate": 0.0004662286860581745,
"loss": 3.3151,
"theoretical_loss": 4.2299398783822815,
"tokens_seen": 253689856
},
{
"epoch": 0.08,
"learning_rate": 0.0004662186559679037,
"loss": 3.0526,
"theoretical_loss": 4.229808179254192,
"tokens_seen": 253755392
},
{
"epoch": 0.08,
"learning_rate": 0.0004662086258776329,
"loss": 3.1744,
"theoretical_loss": 4.22967652365581,
"tokens_seen": 253820928
},
{
"epoch": 0.08,
"learning_rate": 0.0004661985957873621,
"loss": 3.2694,
"theoretical_loss": 4.229544911561513,
"tokens_seen": 253886464
},
{
"epoch": 0.08,
"objective/train/docs_used": 156106,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.7020857334136963,
"objective/train/theoretical_loss": 4.229413342945703,
"objective/train/tokens_used": 274412000,
"theoretical_loss": 4.229413342945703,
"tokens_seen": 253952000
},
{
"epoch": 0.08,
"learning_rate": 0.0004661885656970913,
"loss": 3.1877,
"theoretical_loss": 4.229413342945703,
"tokens_seen": 253952000
},
{
"epoch": 0.08,
"learning_rate": 0.0004661785356068205,
"loss": 3.2587,
"theoretical_loss": 4.229281817782801,
"tokens_seen": 254017536
},
{
"epoch": 0.08,
"learning_rate": 0.00046616850551654967,
"loss": 3.2171,
"theoretical_loss": 4.229150336047251,
"tokens_seen": 254083072
},
{
"epoch": 0.08,
"learning_rate": 0.00046615847542627885,
"loss": 2.7179,
"theoretical_loss": 4.229018897713519,
"tokens_seen": 254148608
},
{
"epoch": 0.08,
"learning_rate": 0.00046614844533600803,
"loss": 3.0186,
"theoretical_loss": 4.22888750275609,
"tokens_seen": 254214144
},
{
"epoch": 0.08,
"learning_rate": 0.00046613841524573727,
"loss": 3.1104,
"theoretical_loss": 4.228756151149475,
"tokens_seen": 254279680
},
{
"epoch": 0.08,
"learning_rate": 0.0004661283851554664,
"loss": 3.2379,
"theoretical_loss": 4.228624842868202,
"tokens_seen": 254345216
},
{
"epoch": 0.08,
"learning_rate": 0.00046611835506519563,
"loss": 3.1056,
"theoretical_loss": 4.228493577886824,
"tokens_seen": 254410752
},
{
"epoch": 0.08,
"learning_rate": 0.00046610832497492475,
"loss": 3.2572,
"theoretical_loss": 4.228362356179913,
"tokens_seen": 254476288
},
{
"epoch": 0.08,
"learning_rate": 0.000466098294884654,
"loss": 3.019,
"theoretical_loss": 4.228231177722063,
"tokens_seen": 254541824
},
{
"epoch": 0.08,
"learning_rate": 0.00046608826479438317,
"loss": 3.1823,
"theoretical_loss": 4.228100042487892,
"tokens_seen": 254607360
},
{
"epoch": 0.08,
"learning_rate": 0.00046607823470411235,
"loss": 3.0422,
"theoretical_loss": 4.227968950452035,
"tokens_seen": 254672896
},
{
"epoch": 0.08,
"learning_rate": 0.00046606820461384153,
"loss": 3.0621,
"theoretical_loss": 4.227837901589153,
"tokens_seen": 254738432
},
{
"epoch": 0.08,
"learning_rate": 0.00046605817452357077,
"loss": 2.9792,
"theoretical_loss": 4.227706895873924,
"tokens_seen": 254803968
},
{
"epoch": 0.08,
"learning_rate": 0.0004660481444332999,
"loss": 3.2174,
"theoretical_loss": 4.227575933281051,
"tokens_seen": 254869504
},
{
"epoch": 0.08,
"learning_rate": 0.00046603811434302913,
"loss": 3.2159,
"theoretical_loss": 4.227445013785257,
"tokens_seen": 254935040
},
{
"epoch": 0.08,
"learning_rate": 0.00046602808425275826,
"loss": 3.2537,
"theoretical_loss": 4.227314137361285,
"tokens_seen": 255000576
},
{
"epoch": 0.08,
"learning_rate": 0.0004660180541624875,
"loss": 3.2481,
"theoretical_loss": 4.227183303983901,
"tokens_seen": 255066112
},
{
"epoch": 0.08,
"learning_rate": 0.0004660080240722167,
"loss": 3.1691,
"theoretical_loss": 4.227052513627893,
"tokens_seen": 255131648
},
{
"epoch": 0.08,
"learning_rate": 0.00046599799398194586,
"loss": 3.1694,
"theoretical_loss": 4.226921766268067,
"tokens_seen": 255197184
},
{
"epoch": 0.08,
"learning_rate": 0.00046598796389167504,
"loss": 3.1596,
"theoretical_loss": 4.226791061879253,
"tokens_seen": 255262720
},
{
"epoch": 0.08,
"learning_rate": 0.0004659779338014042,
"loss": 3.0473,
"theoretical_loss": 4.226660400436302,
"tokens_seen": 255328256
},
{
"epoch": 0.08,
"learning_rate": 0.0004659679037111334,
"loss": 3.1521,
"theoretical_loss": 4.226529781914084,
"tokens_seen": 255393792
},
{
"epoch": 0.08,
"learning_rate": 0.00046595787362086263,
"loss": 3.0574,
"theoretical_loss": 4.226399206287493,
"tokens_seen": 255459328
},
{
"epoch": 0.08,
"learning_rate": 0.00046594784353059176,
"loss": 3.1197,
"theoretical_loss": 4.226268673531442,
"tokens_seen": 255524864
},
{
"epoch": 0.08,
"objective/train/docs_used": 156630,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.661715507507324,
"objective/train/theoretical_loss": 4.226138183620867,
"objective/train/tokens_used": 276050400,
"theoretical_loss": 4.226138183620867,
"tokens_seen": 255590400
},
{
"epoch": 0.08,
"learning_rate": 0.000465937813440321,
"loss": 3.0664,
"theoretical_loss": 4.226138183620867,
"tokens_seen": 255590400
},
{
"epoch": 0.08,
"learning_rate": 0.0004659277833500501,
"loss": 3.0895,
"theoretical_loss": 4.226007736530723,
"tokens_seen": 255655936
},
{
"epoch": 0.08,
"learning_rate": 0.00046591775325977936,
"loss": 2.9574,
"theoretical_loss": 4.225877332235987,
"tokens_seen": 255721472
},
{
"epoch": 0.08,
"learning_rate": 0.00046590772316950854,
"loss": 3.1183,
"theoretical_loss": 4.225746970711657,
"tokens_seen": 255787008
},
{
"epoch": 0.08,
"learning_rate": 0.0004658976930792377,
"loss": 2.9636,
"theoretical_loss": 4.225616651932753,
"tokens_seen": 255852544
},
{
"epoch": 0.08,
"learning_rate": 0.0004658876629889669,
"loss": 3.2771,
"theoretical_loss": 4.225486375874315,
"tokens_seen": 255918080
},
{
"epoch": 0.08,
"learning_rate": 0.00046587763289869614,
"loss": 2.8867,
"theoretical_loss": 4.225356142511402,
"tokens_seen": 255983616
},
{
"epoch": 0.08,
"learning_rate": 0.00046586760280842526,
"loss": 3.2734,
"theoretical_loss": 4.225225951819099,
"tokens_seen": 256049152
},
{
"epoch": 0.08,
"learning_rate": 0.0004658575727181545,
"loss": 3.1851,
"theoretical_loss": 4.225095803772507,
"tokens_seen": 256114688
},
{
"epoch": 0.08,
"learning_rate": 0.0004658475426278836,
"loss": 3.0544,
"theoretical_loss": 4.22496569834675,
"tokens_seen": 256180224
},
{
"epoch": 0.08,
"learning_rate": 0.00046583751253761286,
"loss": 3.1337,
"theoretical_loss": 4.224835635516973,
"tokens_seen": 256245760
},
{
"epoch": 0.08,
"learning_rate": 0.00046582748244734204,
"loss": 2.9687,
"theoretical_loss": 4.224705615258341,
"tokens_seen": 256311296
},
{
"epoch": 0.08,
"learning_rate": 0.0004658174523570712,
"loss": 3.1143,
"theoretical_loss": 4.224575637546041,
"tokens_seen": 256376832
},
{
"epoch": 0.08,
"learning_rate": 0.0004658074222668004,
"loss": 3.1526,
"theoretical_loss": 4.224445702355279,
"tokens_seen": 256442368
},
{
"epoch": 0.08,
"learning_rate": 0.0004657973921765296,
"loss": 3.3006,
"theoretical_loss": 4.2243158096612845,
"tokens_seen": 256507904
},
{
"epoch": 0.08,
"learning_rate": 0.00046578736208625877,
"loss": 3.0952,
"theoretical_loss": 4.224185959439305,
"tokens_seen": 256573440
},
{
"epoch": 0.08,
"learning_rate": 0.000465777331995988,
"loss": 2.8397,
"theoretical_loss": 4.22405615166461,
"tokens_seen": 256638976
},
{
"epoch": 0.08,
"learning_rate": 0.00046576730190571713,
"loss": 3.1773,
"theoretical_loss": 4.22392638631249,
"tokens_seen": 256704512
},
{
"epoch": 0.08,
"learning_rate": 0.00046575727181544636,
"loss": 3.0894,
"theoretical_loss": 4.223796663358255,
"tokens_seen": 256770048
},
{
"epoch": 0.08,
"learning_rate": 0.00046574724172517555,
"loss": 3.044,
"theoretical_loss": 4.223666982777237,
"tokens_seen": 256835584
},
{
"epoch": 0.08,
"learning_rate": 0.0004657372116349047,
"loss": 3.1196,
"theoretical_loss": 4.223537344544788,
"tokens_seen": 256901120
},
{
"epoch": 0.08,
"learning_rate": 0.0004657271815446339,
"loss": 3.3258,
"theoretical_loss": 4.223407748636282,
"tokens_seen": 256966656
},
{
"epoch": 0.08,
"learning_rate": 0.0004657171514543631,
"loss": 3.1867,
"theoretical_loss": 4.22327819502711,
"tokens_seen": 257032192
},
{
"epoch": 0.08,
"learning_rate": 0.00046570712136409227,
"loss": 3.0392,
"theoretical_loss": 4.223148683692687,
"tokens_seen": 257097728
},
{
"epoch": 0.08,
"learning_rate": 0.0004656970912738215,
"loss": 3.1276,
"theoretical_loss": 4.223019214608446,
"tokens_seen": 257163264
},
{
"epoch": 0.08,
"objective/train/docs_used": 157258,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.008852005004883,
"objective/train/theoretical_loss": 4.222889787749845,
"objective/train/tokens_used": 277688800,
"theoretical_loss": 4.222889787749845,
"tokens_seen": 257228800
},
{
"epoch": 0.08,
"learning_rate": 0.00046568706118355063,
"loss": 3.0922,
"theoretical_loss": 4.222889787749845,
"tokens_seen": 257228800
},
{
"epoch": 0.08,
"learning_rate": 0.00046567703109327987,
"loss": 2.8066,
"theoretical_loss": 4.222760403092358,
"tokens_seen": 257294336
},
{
"epoch": 0.08,
"learning_rate": 0.000465667001003009,
"loss": 3.0272,
"theoretical_loss": 4.22263106061148,
"tokens_seen": 257359872
},
{
"epoch": 0.08,
"learning_rate": 0.00046565697091273823,
"loss": 3.0383,
"theoretical_loss": 4.222501760282729,
"tokens_seen": 257425408
},
{
"epoch": 0.08,
"learning_rate": 0.0004656469408224674,
"loss": 2.9278,
"theoretical_loss": 4.22237250208164,
"tokens_seen": 257490944
},
{
"epoch": 0.08,
"learning_rate": 0.0004656369107321966,
"loss": 3.0506,
"theoretical_loss": 4.222243285983772,
"tokens_seen": 257556480
},
{
"epoch": 0.08,
"learning_rate": 0.0004656268806419258,
"loss": 2.754,
"theoretical_loss": 4.222114111964703,
"tokens_seen": 257622016
},
{
"epoch": 0.08,
"learning_rate": 0.00046561685055165495,
"loss": 2.9813,
"theoretical_loss": 4.221984980000029,
"tokens_seen": 257687552
},
{
"epoch": 0.08,
"learning_rate": 0.00046560682046138414,
"loss": 2.94,
"theoretical_loss": 4.2218558900653695,
"tokens_seen": 257753088
},
{
"epoch": 0.08,
"learning_rate": 0.00046559679037111337,
"loss": 2.7958,
"theoretical_loss": 4.221726842136364,
"tokens_seen": 257818624
},
{
"epoch": 0.08,
"learning_rate": 0.0004655867602808425,
"loss": 3.1393,
"theoretical_loss": 4.2215978361886695,
"tokens_seen": 257884160
},
{
"epoch": 0.08,
"learning_rate": 0.00046557673019057173,
"loss": 3.1812,
"theoretical_loss": 4.221468872197967,
"tokens_seen": 257949696
},
{
"epoch": 0.08,
"learning_rate": 0.0004655667001003009,
"loss": 3.066,
"theoretical_loss": 4.221339950139956,
"tokens_seen": 258015232
},
{
"epoch": 0.08,
"learning_rate": 0.0004655566700100301,
"loss": 2.9906,
"theoretical_loss": 4.221211069990357,
"tokens_seen": 258080768
},
{
"epoch": 0.08,
"learning_rate": 0.0004655466399197593,
"loss": 3.1732,
"theoretical_loss": 4.221082231724908,
"tokens_seen": 258146304
},
{
"epoch": 0.08,
"learning_rate": 0.00046553660982948846,
"loss": 3.0643,
"theoretical_loss": 4.22095343531937,
"tokens_seen": 258211840
},
{
"epoch": 0.08,
"learning_rate": 0.00046552657973921764,
"loss": 3.1475,
"theoretical_loss": 4.220824680749525,
"tokens_seen": 258277376
},
{
"epoch": 0.08,
"learning_rate": 0.0004655165496489469,
"loss": 3.227,
"theoretical_loss": 4.220695967991171,
"tokens_seen": 258342912
},
{
"epoch": 0.08,
"learning_rate": 0.000465506519558676,
"loss": 3.1797,
"theoretical_loss": 4.220567297020131,
"tokens_seen": 258408448
},
{
"epoch": 0.08,
"learning_rate": 0.00046549648946840524,
"loss": 3.0443,
"theoretical_loss": 4.220438667812244,
"tokens_seen": 258473984
},
{
"epoch": 0.08,
"learning_rate": 0.00046548645937813436,
"loss": 3.0716,
"theoretical_loss": 4.220310080343373,
"tokens_seen": 258539520
},
{
"epoch": 0.08,
"learning_rate": 0.0004654764292878636,
"loss": 3.2806,
"theoretical_loss": 4.220181534589398,
"tokens_seen": 258605056
},
{
"epoch": 0.08,
"learning_rate": 0.0004654663991975928,
"loss": 3.0484,
"theoretical_loss": 4.22005303052622,
"tokens_seen": 258670592
},
{
"epoch": 0.08,
"learning_rate": 0.00046545636910732196,
"loss": 3.2093,
"theoretical_loss": 4.219924568129759,
"tokens_seen": 258736128
},
{
"epoch": 0.08,
"learning_rate": 0.0004654463390170512,
"loss": 2.9654,
"theoretical_loss": 4.219796147375957,
"tokens_seen": 258801664
},
{
"epoch": 0.08,
"objective/train/docs_used": 158510,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9906694889068604,
"objective/train/theoretical_loss": 4.219667768240775,
"objective/train/tokens_used": 279327200,
"theoretical_loss": 4.219667768240775,
"tokens_seen": 258867200
},
{
"epoch": 0.08,
"learning_rate": 0.0004654363089267803,
"loss": 3.068,
"theoretical_loss": 4.219667768240775,
"tokens_seen": 258867200
},
{
"epoch": 0.08,
"learning_rate": 0.00046542627883650956,
"loss": 3.2658,
"theoretical_loss": 4.219539430700195,
"tokens_seen": 258932736
},
{
"epoch": 0.08,
"learning_rate": 0.00046541624874623874,
"loss": 3.0762,
"theoretical_loss": 4.2194111347302155,
"tokens_seen": 258998272
},
{
"epoch": 0.08,
"learning_rate": 0.0004654062186559679,
"loss": 3.2659,
"theoretical_loss": 4.219282880306859,
"tokens_seen": 259063808
},
{
"epoch": 0.08,
"learning_rate": 0.0004653961885656971,
"loss": 3.0452,
"theoretical_loss": 4.219154667406166,
"tokens_seen": 259129344
},
{
"epoch": 0.08,
"learning_rate": 0.00046538615847542634,
"loss": 3.0682,
"theoretical_loss": 4.219026496004198,
"tokens_seen": 259194880
},
{
"epoch": 0.08,
"learning_rate": 0.00046537612838515546,
"loss": 3.0652,
"theoretical_loss": 4.218898366077035,
"tokens_seen": 259260416
},
{
"epoch": 0.08,
"learning_rate": 0.0004653660982948847,
"loss": 3.2532,
"theoretical_loss": 4.218770277600775,
"tokens_seen": 259325952
},
{
"epoch": 0.08,
"learning_rate": 0.0004653560682046138,
"loss": 2.9702,
"theoretical_loss": 4.218642230551541,
"tokens_seen": 259391488
},
{
"epoch": 0.08,
"learning_rate": 0.00046534603811434306,
"loss": 2.987,
"theoretical_loss": 4.218514224905472,
"tokens_seen": 259457024
},
{
"epoch": 0.08,
"learning_rate": 0.00046533600802407224,
"loss": 3.1381,
"theoretical_loss": 4.218386260638727,
"tokens_seen": 259522560
},
{
"epoch": 0.08,
"learning_rate": 0.0004653259779338014,
"loss": 2.8499,
"theoretical_loss": 4.2182583377274865,
"tokens_seen": 259588096
},
{
"epoch": 0.08,
"learning_rate": 0.0004653159478435306,
"loss": 3.0011,
"theoretical_loss": 4.218130456147948,
"tokens_seen": 259653632
},
{
"epoch": 0.08,
"learning_rate": 0.0004653059177532598,
"loss": 3.0006,
"theoretical_loss": 4.218002615876332,
"tokens_seen": 259719168
},
{
"epoch": 0.08,
"learning_rate": 0.00046529588766298897,
"loss": 2.8484,
"theoretical_loss": 4.217874816888877,
"tokens_seen": 259784704
},
{
"epoch": 0.08,
"learning_rate": 0.0004652858575727182,
"loss": 3.0114,
"theoretical_loss": 4.217747059161839,
"tokens_seen": 259850240
},
{
"epoch": 0.08,
"learning_rate": 0.00046527582748244733,
"loss": 3.1379,
"theoretical_loss": 4.217619342671498,
"tokens_seen": 259915776
},
{
"epoch": 0.08,
"learning_rate": 0.00046526579739217656,
"loss": 3.1485,
"theoretical_loss": 4.2174916673941505,
"tokens_seen": 259981312
},
{
"epoch": 0.08,
"learning_rate": 0.00046525576730190575,
"loss": 3.1037,
"theoretical_loss": 4.217364033306113,
"tokens_seen": 260046848
},
{
"epoch": 0.08,
"learning_rate": 0.0004652457372116349,
"loss": 3.1131,
"theoretical_loss": 4.217236440383724,
"tokens_seen": 260112384
},
{
"epoch": 0.08,
"learning_rate": 0.0004652357071213641,
"loss": 3.1678,
"theoretical_loss": 4.217108888603337,
"tokens_seen": 260177920
},
{
"epoch": 0.08,
"learning_rate": 0.0004652256770310933,
"loss": 3.1065,
"theoretical_loss": 4.21698137794133,
"tokens_seen": 260243456
},
{
"epoch": 0.08,
"learning_rate": 0.00046521564694082247,
"loss": 2.9596,
"theoretical_loss": 4.216853908374097,
"tokens_seen": 260308992
},
{
"epoch": 0.08,
"learning_rate": 0.0004652056168505517,
"loss": 3.008,
"theoretical_loss": 4.216726479878052,
"tokens_seen": 260374528
},
{
"epoch": 0.08,
"learning_rate": 0.00046519558676028083,
"loss": 3.0454,
"theoretical_loss": 4.216599092429631,
"tokens_seen": 260440064
},
{
"epoch": 0.08,
"objective/train/docs_used": 159098,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7527639865875244,
"objective/train/theoretical_loss": 4.216471746005286,
"objective/train/tokens_used": 280965600,
"theoretical_loss": 4.216471746005286,
"tokens_seen": 260505600
},
{
"epoch": 0.08,
"learning_rate": 0.00046518555667001007,
"loss": 3.0212,
"theoretical_loss": 4.216471746005286,
"tokens_seen": 260505600
},
{
"epoch": 0.08,
"learning_rate": 0.0004651755265797392,
"loss": 3.0566,
"theoretical_loss": 4.216344440581491,
"tokens_seen": 260571136
},
{
"epoch": 0.08,
"learning_rate": 0.00046516549648946843,
"loss": 2.9021,
"theoretical_loss": 4.2162171761347365,
"tokens_seen": 260636672
},
{
"epoch": 0.08,
"learning_rate": 0.0004651554663991976,
"loss": 2.9991,
"theoretical_loss": 4.2160899526415365,
"tokens_seen": 260702208
},
{
"epoch": 0.08,
"learning_rate": 0.0004651454363089268,
"loss": 3.0791,
"theoretical_loss": 4.215962770078422,
"tokens_seen": 260767744
},
{
"epoch": 0.08,
"learning_rate": 0.000465135406218656,
"loss": 3.1413,
"theoretical_loss": 4.215835628421942,
"tokens_seen": 260833280
},
{
"epoch": 0.08,
"learning_rate": 0.00046512537612838515,
"loss": 2.9224,
"theoretical_loss": 4.215708527648667,
"tokens_seen": 260898816
},
{
"epoch": 0.08,
"learning_rate": 0.00046511534603811434,
"loss": 3.189,
"theoretical_loss": 4.215581467735187,
"tokens_seen": 260964352
},
{
"epoch": 0.08,
"learning_rate": 0.00046510531594784357,
"loss": 3.3999,
"theoretical_loss": 4.215454448658109,
"tokens_seen": 261029888
},
{
"epoch": 0.08,
"learning_rate": 0.0004650952858575727,
"loss": 2.9878,
"theoretical_loss": 4.215327470394062,
"tokens_seen": 261095424
},
{
"epoch": 0.08,
"learning_rate": 0.00046508525576730193,
"loss": 3.3008,
"theoretical_loss": 4.215200532919691,
"tokens_seen": 261160960
},
{
"epoch": 0.08,
"learning_rate": 0.0004650752256770311,
"loss": 3.126,
"theoretical_loss": 4.215073636211664,
"tokens_seen": 261226496
},
{
"epoch": 0.08,
"learning_rate": 0.0004650651955867603,
"loss": 3.0601,
"theoretical_loss": 4.214946780246666,
"tokens_seen": 261292032
},
{
"epoch": 0.08,
"learning_rate": 0.0004650551654964895,
"loss": 3.2414,
"theoretical_loss": 4.214819965001401,
"tokens_seen": 261357568
},
{
"epoch": 0.08,
"learning_rate": 0.00046504513540621866,
"loss": 3.3807,
"theoretical_loss": 4.214693190452593,
"tokens_seen": 261423104
},
{
"epoch": 0.08,
"learning_rate": 0.00046503510531594784,
"loss": 3.2496,
"theoretical_loss": 4.214566456576984,
"tokens_seen": 261488640
},
{
"epoch": 0.08,
"learning_rate": 0.0004650250752256771,
"loss": 3.0059,
"theoretical_loss": 4.214439763351336,
"tokens_seen": 261554176
},
{
"epoch": 0.08,
"learning_rate": 0.0004650150451354062,
"loss": 3.3606,
"theoretical_loss": 4.214313110752431,
"tokens_seen": 261619712
},
{
"epoch": 0.08,
"learning_rate": 0.00046500501504513544,
"loss": 3.2016,
"theoretical_loss": 4.214186498757069,
"tokens_seen": 261685248
},
{
"epoch": 0.08,
"learning_rate": 0.00046499498495486456,
"loss": 3.238,
"theoretical_loss": 4.214059927342068,
"tokens_seen": 261750784
},
{
"epoch": 0.08,
"learning_rate": 0.0004649849548645938,
"loss": 3.0229,
"theoretical_loss": 4.213933396484267,
"tokens_seen": 261816320
},
{
"epoch": 0.08,
"learning_rate": 0.000464974924774323,
"loss": 3.3204,
"theoretical_loss": 4.213806906160523,
"tokens_seen": 261881856
},
{
"epoch": 0.08,
"learning_rate": 0.00046496489468405216,
"loss": 3.1548,
"theoretical_loss": 4.213680456347712,
"tokens_seen": 261947392
},
{
"epoch": 0.08,
"learning_rate": 0.00046495486459378134,
"loss": 3.0314,
"theoretical_loss": 4.213554047022729,
"tokens_seen": 262012928
},
{
"epoch": 0.08,
"learning_rate": 0.0004649448345035105,
"loss": 3.1052,
"theoretical_loss": 4.213427678162489,
"tokens_seen": 262078464
},
{
"epoch": 0.08,
"objective/train/docs_used": 160106,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2078521251678467,
"objective/train/theoretical_loss": 4.213301349743924,
"objective/train/tokens_used": 282604000,
"theoretical_loss": 4.213301349743924,
"tokens_seen": 262144000
},
{
"epoch": 0.08,
"learning_rate": 0.0004649348044132397,
"loss": 3.285,
"theoretical_loss": 4.213301349743924,
"tokens_seen": 262144000
},
{
"epoch": 0.08,
"learning_rate": 0.00046492477432296894,
"loss": 3.2845,
"theoretical_loss": 4.2131750617439865,
"tokens_seen": 262209536
},
{
"epoch": 0.08,
"learning_rate": 0.00046491474423269807,
"loss": 3.0031,
"theoretical_loss": 4.213048814139647,
"tokens_seen": 262275072
},
{
"epoch": 0.08,
"learning_rate": 0.0004649047141424273,
"loss": 3.0222,
"theoretical_loss": 4.212922606907895,
"tokens_seen": 262340608
},
{
"epoch": 0.08,
"learning_rate": 0.0004648946840521565,
"loss": 3.0202,
"theoretical_loss": 4.21279644002574,
"tokens_seen": 262406144
},
{
"epoch": 0.08,
"learning_rate": 0.00046488465396188566,
"loss": 2.9464,
"theoretical_loss": 4.212670313470209,
"tokens_seen": 262471680
},
{
"epoch": 0.08,
"learning_rate": 0.00046487462387161485,
"loss": 3.3016,
"theoretical_loss": 4.212544227218347,
"tokens_seen": 262537216
},
{
"epoch": 0.08,
"learning_rate": 0.000464864593781344,
"loss": 3.0782,
"theoretical_loss": 4.21241818124722,
"tokens_seen": 262602752
},
{
"epoch": 0.08,
"learning_rate": 0.0004648545636910732,
"loss": 3.1333,
"theoretical_loss": 4.212292175533912,
"tokens_seen": 262668288
},
{
"epoch": 0.08,
"learning_rate": 0.00046484453360080244,
"loss": 2.8875,
"theoretical_loss": 4.212166210055526,
"tokens_seen": 262733824
},
{
"epoch": 0.08,
"learning_rate": 0.00046483450351053157,
"loss": 2.84,
"theoretical_loss": 4.212040284789181,
"tokens_seen": 262799360
},
{
"epoch": 0.08,
"learning_rate": 0.0004648244734202608,
"loss": 3.2869,
"theoretical_loss": 4.211914399712019,
"tokens_seen": 262864896
},
{
"epoch": 0.08,
"learning_rate": 0.00046481444332998993,
"loss": 3.0755,
"theoretical_loss": 4.211788554801198,
"tokens_seen": 262930432
},
{
"epoch": 0.08,
"learning_rate": 0.00046480441323971917,
"loss": 3.207,
"theoretical_loss": 4.211662750033895,
"tokens_seen": 262995968
},
{
"epoch": 0.08,
"learning_rate": 0.00046479438314944835,
"loss": 2.9361,
"theoretical_loss": 4.211536985387307,
"tokens_seen": 263061504
},
{
"epoch": 0.08,
"learning_rate": 0.00046478435305917753,
"loss": 3.1135,
"theoretical_loss": 4.211411260838647,
"tokens_seen": 263127040
},
{
"epoch": 0.08,
"learning_rate": 0.0004647743229689067,
"loss": 3.003,
"theoretical_loss": 4.2112855763651496,
"tokens_seen": 263192576
},
{
"epoch": 0.08,
"learning_rate": 0.00046476429287863595,
"loss": 3.252,
"theoretical_loss": 4.211159931944065,
"tokens_seen": 263258112
},
{
"epoch": 0.08,
"learning_rate": 0.0004647542627883651,
"loss": 3.0456,
"theoretical_loss": 4.211034327552666,
"tokens_seen": 263323648
},
{
"epoch": 0.08,
"learning_rate": 0.0004647442326980943,
"loss": 3.2239,
"theoretical_loss": 4.210908763168239,
"tokens_seen": 263389184
},
{
"epoch": 0.08,
"learning_rate": 0.00046473420260782344,
"loss": 3.3088,
"theoretical_loss": 4.210783238768093,
"tokens_seen": 263454720
},
{
"epoch": 0.08,
"learning_rate": 0.00046472417251755267,
"loss": 3.1186,
"theoretical_loss": 4.210657754329553,
"tokens_seen": 263520256
},
{
"epoch": 0.08,
"learning_rate": 0.0004647141424272819,
"loss": 3.2965,
"theoretical_loss": 4.210532309829965,
"tokens_seen": 263585792
},
{
"epoch": 0.08,
"learning_rate": 0.00046470411233701103,
"loss": 3.2105,
"theoretical_loss": 4.21040690524669,
"tokens_seen": 263651328
},
{
"epoch": 0.08,
"learning_rate": 0.00046469408224674027,
"loss": 2.8521,
"theoretical_loss": 4.21028154055711,
"tokens_seen": 263716864
},
{
"epoch": 0.08,
"objective/train/docs_used": 160649,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5538294315338135,
"objective/train/theoretical_loss": 4.2101562157386265,
"objective/train/tokens_used": 284242400,
"theoretical_loss": 4.2101562157386265,
"tokens_seen": 263782400
},
{
"epoch": 0.08,
"learning_rate": 0.0004646840521564694,
"loss": 2.7667,
"theoretical_loss": 4.2101562157386265,
"tokens_seen": 263782400
},
{
"epoch": 0.08,
"learning_rate": 0.00046467402206619863,
"loss": 2.8686,
"theoretical_loss": 4.210030930768655,
"tokens_seen": 263847936
},
{
"epoch": 0.08,
"learning_rate": 0.0004646639919759278,
"loss": 3.1047,
"theoretical_loss": 4.2099056856246335,
"tokens_seen": 263913472
},
{
"epoch": 0.08,
"learning_rate": 0.000464653961885657,
"loss": 3.2001,
"theoretical_loss": 4.209780480284017,
"tokens_seen": 263979008
},
{
"epoch": 0.08,
"learning_rate": 0.0004646439317953862,
"loss": 3.1007,
"theoretical_loss": 4.209655314724279,
"tokens_seen": 264044544
},
{
"epoch": 0.08,
"learning_rate": 0.00046463390170511535,
"loss": 2.9901,
"theoretical_loss": 4.209530188922911,
"tokens_seen": 264110080
},
{
"epoch": 0.08,
"learning_rate": 0.00046462387161484454,
"loss": 2.8877,
"theoretical_loss": 4.209405102857422,
"tokens_seen": 264175616
},
{
"epoch": 0.08,
"learning_rate": 0.00046461384152457377,
"loss": 3.2396,
"theoretical_loss": 4.209280056505342,
"tokens_seen": 264241152
},
{
"epoch": 0.08,
"learning_rate": 0.0004646038114343029,
"loss": 2.9324,
"theoretical_loss": 4.209155049844217,
"tokens_seen": 264306688
},
{
"epoch": 0.08,
"learning_rate": 0.00046459378134403213,
"loss": 2.7333,
"theoretical_loss": 4.209030082851612,
"tokens_seen": 264372224
},
{
"epoch": 0.08,
"learning_rate": 0.0004645837512537613,
"loss": 2.7765,
"theoretical_loss": 4.208905155505109,
"tokens_seen": 264437760
},
{
"epoch": 0.08,
"learning_rate": 0.0004645737211634905,
"loss": 3.0924,
"theoretical_loss": 4.20878026778231,
"tokens_seen": 264503296
},
{
"epoch": 0.08,
"learning_rate": 0.0004645636910732197,
"loss": 3.1207,
"theoretical_loss": 4.208655419660834,
"tokens_seen": 264568832
},
{
"epoch": 0.08,
"learning_rate": 0.00046455366098294886,
"loss": 3.1763,
"theoretical_loss": 4.208530611118321,
"tokens_seen": 264634368
},
{
"epoch": 0.08,
"learning_rate": 0.00046454363089267804,
"loss": 3.0411,
"theoretical_loss": 4.208405842132423,
"tokens_seen": 264699904
},
{
"epoch": 0.08,
"learning_rate": 0.0004645336008024073,
"loss": 3.1361,
"theoretical_loss": 4.208281112680817,
"tokens_seen": 264765440
},
{
"epoch": 0.08,
"learning_rate": 0.0004645235707121364,
"loss": 3.1252,
"theoretical_loss": 4.208156422741195,
"tokens_seen": 264830976
},
{
"epoch": 0.08,
"learning_rate": 0.00046451354062186564,
"loss": 3.0564,
"theoretical_loss": 4.208031772291265,
"tokens_seen": 264896512
},
{
"epoch": 0.08,
"learning_rate": 0.00046450351053159476,
"loss": 3.4202,
"theoretical_loss": 4.207907161308757,
"tokens_seen": 264962048
},
{
"epoch": 0.08,
"learning_rate": 0.000464493480441324,
"loss": 3.3518,
"theoretical_loss": 4.2077825897714165,
"tokens_seen": 265027584
},
{
"epoch": 0.08,
"learning_rate": 0.0004644834503510532,
"loss": 2.9007,
"theoretical_loss": 4.207658057657008,
"tokens_seen": 265093120
},
{
"epoch": 0.08,
"learning_rate": 0.00046447342026078236,
"loss": 3.2147,
"theoretical_loss": 4.207533564943316,
"tokens_seen": 265158656
},
{
"epoch": 0.08,
"learning_rate": 0.00046446339017051154,
"loss": 3.1103,
"theoretical_loss": 4.207409111608138,
"tokens_seen": 265224192
},
{
"epoch": 0.08,
"learning_rate": 0.0004644533600802407,
"loss": 3.1886,
"theoretical_loss": 4.2072846976292935,
"tokens_seen": 265289728
},
{
"epoch": 0.08,
"learning_rate": 0.0004644433299899699,
"loss": 3.018,
"theoretical_loss": 4.2071603229846195,
"tokens_seen": 265355264
},
{
"epoch": 0.08,
"objective/train/docs_used": 161970,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.322939157485962,
"objective/train/theoretical_loss": 4.20703598765197,
"objective/train/tokens_used": 285880800,
"theoretical_loss": 4.20703598765197,
"tokens_seen": 265420800
},
{
"epoch": 0.08,
"learning_rate": 0.00046443329989969914,
"loss": 3.2719,
"theoretical_loss": 4.20703598765197,
"tokens_seen": 265420800
},
{
"epoch": 0.08,
"learning_rate": 0.00046442326980942827,
"loss": 3.2149,
"theoretical_loss": 4.206911691609217,
"tokens_seen": 265486336
},
{
"epoch": 0.08,
"learning_rate": 0.0004644132397191575,
"loss": 3.2664,
"theoretical_loss": 4.206787434834251,
"tokens_seen": 265551872
},
{
"epoch": 0.08,
"learning_rate": 0.0004644032096288867,
"loss": 2.9888,
"theoretical_loss": 4.20666321730498,
"tokens_seen": 265617408
},
{
"epoch": 0.08,
"learning_rate": 0.00046439317953861586,
"loss": 3.0063,
"theoretical_loss": 4.206539038999329,
"tokens_seen": 265682944
},
{
"epoch": 0.08,
"learning_rate": 0.00046438314944834505,
"loss": 3.1335,
"theoretical_loss": 4.206414899895244,
"tokens_seen": 265748480
},
{
"epoch": 0.08,
"learning_rate": 0.0004643731193580742,
"loss": 2.8322,
"theoretical_loss": 4.206290799970685,
"tokens_seen": 265814016
},
{
"epoch": 0.08,
"learning_rate": 0.0004643630892678034,
"loss": 3.3554,
"theoretical_loss": 4.206166739203632,
"tokens_seen": 265879552
},
{
"epoch": 0.08,
"learning_rate": 0.00046435305917753264,
"loss": 2.9797,
"theoretical_loss": 4.206042717572082,
"tokens_seen": 265945088
},
{
"epoch": 0.08,
"learning_rate": 0.00046434302908726177,
"loss": 3.1981,
"theoretical_loss": 4.20591873505405,
"tokens_seen": 266010624
},
{
"epoch": 0.08,
"learning_rate": 0.000464332998996991,
"loss": 3.1016,
"theoretical_loss": 4.20579479162757,
"tokens_seen": 266076160
},
{
"epoch": 0.08,
"learning_rate": 0.00046432296890672013,
"loss": 3.2206,
"theoretical_loss": 4.205670887270691,
"tokens_seen": 266141696
},
{
"epoch": 0.08,
"learning_rate": 0.00046431293881644937,
"loss": 2.916,
"theoretical_loss": 4.205547021961482,
"tokens_seen": 266207232
},
{
"epoch": 0.08,
"learning_rate": 0.00046430290872617855,
"loss": 3.171,
"theoretical_loss": 4.205423195678029,
"tokens_seen": 266272768
},
{
"epoch": 0.08,
"learning_rate": 0.00046429287863590773,
"loss": 3.2186,
"theoretical_loss": 4.205299408398435,
"tokens_seen": 266338304
},
{
"epoch": 0.08,
"learning_rate": 0.0004642828485456369,
"loss": 2.7666,
"theoretical_loss": 4.2051756601008226,
"tokens_seen": 266403840
},
{
"epoch": 0.08,
"learning_rate": 0.00046427281845536615,
"loss": 2.9658,
"theoretical_loss": 4.20505195076333,
"tokens_seen": 266469376
},
{
"epoch": 0.08,
"learning_rate": 0.0004642627883650953,
"loss": 3.3264,
"theoretical_loss": 4.204928280364115,
"tokens_seen": 266534912
},
{
"epoch": 0.08,
"learning_rate": 0.0004642527582748245,
"loss": 3.4252,
"theoretical_loss": 4.20480464888135,
"tokens_seen": 266600448
},
{
"epoch": 0.08,
"learning_rate": 0.00046424272818455364,
"loss": 2.9859,
"theoretical_loss": 4.204681056293228,
"tokens_seen": 266665984
},
{
"epoch": 0.08,
"learning_rate": 0.00046423269809428287,
"loss": 3.0057,
"theoretical_loss": 4.204557502577957,
"tokens_seen": 266731520
},
{
"epoch": 0.08,
"learning_rate": 0.00046422266800401205,
"loss": 3.0726,
"theoretical_loss": 4.204433987713767,
"tokens_seen": 266797056
},
{
"epoch": 0.08,
"learning_rate": 0.00046421263791374123,
"loss": 3.2009,
"theoretical_loss": 4.2043105116789,
"tokens_seen": 266862592
},
{
"epoch": 0.08,
"learning_rate": 0.0004642026078234704,
"loss": 2.9325,
"theoretical_loss": 4.204187074451617,
"tokens_seen": 266928128
},
{
"epoch": 0.08,
"learning_rate": 0.0004641925777331996,
"loss": 3.144,
"theoretical_loss": 4.204063676010202,
"tokens_seen": 266993664
},
{
"epoch": 0.08,
"objective/train/docs_used": 162606,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0582354068756104,
"objective/train/theoretical_loss": 4.203940316332948,
"objective/train/tokens_used": 287519200,
"theoretical_loss": 4.203940316332948,
"tokens_seen": 267059200
},
{
"epoch": 0.08,
"learning_rate": 0.0004641825476429288,
"loss": 3.183,
"theoretical_loss": 4.203940316332948,
"tokens_seen": 267059200
},
{
"epoch": 0.08,
"learning_rate": 0.000464172517552658,
"loss": 3.2655,
"theoretical_loss": 4.203816995398171,
"tokens_seen": 267124736
},
{
"epoch": 0.08,
"learning_rate": 0.00046416248746238714,
"loss": 3.1179,
"theoretical_loss": 4.203693713184203,
"tokens_seen": 267190272
},
{
"epoch": 0.08,
"learning_rate": 0.0004641524573721164,
"loss": 3.0399,
"theoretical_loss": 4.203570469669392,
"tokens_seen": 267255808
},
{
"epoch": 0.08,
"learning_rate": 0.0004641424272818455,
"loss": 2.9932,
"theoretical_loss": 4.203447264832107,
"tokens_seen": 267321344
},
{
"epoch": 0.08,
"learning_rate": 0.00046413239719157474,
"loss": 3.1497,
"theoretical_loss": 4.203324098650731,
"tokens_seen": 267386880
},
{
"epoch": 0.08,
"learning_rate": 0.0004641223671013039,
"loss": 3.0141,
"theoretical_loss": 4.203200971103666,
"tokens_seen": 267452416
},
{
"epoch": 0.08,
"learning_rate": 0.0004641123370110331,
"loss": 3.1592,
"theoretical_loss": 4.20307788216933,
"tokens_seen": 267517952
},
{
"epoch": 0.08,
"learning_rate": 0.0004641023069207623,
"loss": 3.0264,
"theoretical_loss": 4.202954831826159,
"tokens_seen": 267583488
},
{
"epoch": 0.08,
"learning_rate": 0.0004640922768304915,
"loss": 3.1223,
"theoretical_loss": 4.202831820052609,
"tokens_seen": 267649024
},
{
"epoch": 0.08,
"learning_rate": 0.00046408224674022064,
"loss": 3.1909,
"theoretical_loss": 4.202708846827148,
"tokens_seen": 267714560
},
{
"epoch": 0.08,
"learning_rate": 0.0004640722166499499,
"loss": 3.2318,
"theoretical_loss": 4.202585912128266,
"tokens_seen": 267780096
},
{
"epoch": 0.08,
"learning_rate": 0.000464062186559679,
"loss": 2.8608,
"theoretical_loss": 4.202463015934468,
"tokens_seen": 267845632
},
{
"epoch": 0.08,
"learning_rate": 0.00046405215646940824,
"loss": 3.2008,
"theoretical_loss": 4.202340158224277,
"tokens_seen": 267911168
},
{
"epoch": 0.08,
"learning_rate": 0.0004640421263791374,
"loss": 3.1812,
"theoretical_loss": 4.202217338976231,
"tokens_seen": 267976704
},
{
"epoch": 0.08,
"learning_rate": 0.0004640320962888666,
"loss": 2.8798,
"theoretical_loss": 4.2020945581688895,
"tokens_seen": 268042240
},
{
"epoch": 0.08,
"learning_rate": 0.0004640220661985958,
"loss": 3.1127,
"theoretical_loss": 4.201971815780826,
"tokens_seen": 268107776
},
{
"epoch": 0.08,
"learning_rate": 0.00046401203610832496,
"loss": 3.0995,
"theoretical_loss": 4.201849111790631,
"tokens_seen": 268173312
},
{
"epoch": 0.08,
"learning_rate": 0.00046400200601805414,
"loss": 3.0434,
"theoretical_loss": 4.201726446176915,
"tokens_seen": 268238848
},
{
"epoch": 0.08,
"learning_rate": 0.0004639919759277834,
"loss": 3.05,
"theoretical_loss": 4.201603818918302,
"tokens_seen": 268304384
},
{
"epoch": 0.08,
"learning_rate": 0.0004639819458375125,
"loss": 3.2322,
"theoretical_loss": 4.201481229993435,
"tokens_seen": 268369920
},
{
"epoch": 0.08,
"learning_rate": 0.00046397191574724174,
"loss": 3.2241,
"theoretical_loss": 4.201358679380976,
"tokens_seen": 268435456
},
{
"epoch": 0.08,
"learning_rate": 0.0004639618856569709,
"loss": 3.1414,
"theoretical_loss": 4.201236167059601,
"tokens_seen": 268500992
},
{
"epoch": 0.08,
"learning_rate": 0.0004639518555667001,
"loss": 3.2403,
"theoretical_loss": 4.201113693008002,
"tokens_seen": 268566528
},
{
"epoch": 0.08,
"learning_rate": 0.00046394182547642934,
"loss": 3.0643,
"theoretical_loss": 4.200991257204894,
"tokens_seen": 268632064
},
{
"epoch": 0.08,
"objective/train/docs_used": 163909,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.577915668487549,
"objective/train/theoretical_loss": 4.2008688596290025,
"objective/train/tokens_used": 289157600,
"theoretical_loss": 4.2008688596290025,
"tokens_seen": 268697600
},
{
"epoch": 0.08,
"learning_rate": 0.00046393179538615847,
"loss": 2.8287,
"theoretical_loss": 4.2008688596290025,
"tokens_seen": 268697600
},
{
"epoch": 0.08,
"learning_rate": 0.0004639217652958877,
"loss": 3.1737,
"theoretical_loss": 4.200746500259073,
"tokens_seen": 268763136
},
{
"epoch": 0.08,
"learning_rate": 0.0004639117352056169,
"loss": 2.6194,
"theoretical_loss": 4.200624179073869,
"tokens_seen": 268828672
},
{
"epoch": 0.08,
"learning_rate": 0.00046390170511534606,
"loss": 3.2024,
"theoretical_loss": 4.2005018960521685,
"tokens_seen": 268894208
},
{
"epoch": 0.08,
"learning_rate": 0.00046389167502507525,
"loss": 3.3453,
"theoretical_loss": 4.200379651172769,
"tokens_seen": 268959744
},
{
"epoch": 0.08,
"learning_rate": 0.0004638816449348044,
"loss": 2.8859,
"theoretical_loss": 4.200257444414483,
"tokens_seen": 269025280
},
{
"epoch": 0.08,
"learning_rate": 0.0004638716148445336,
"loss": 3.1556,
"theoretical_loss": 4.200135275756139,
"tokens_seen": 269090816
},
{
"epoch": 0.08,
"learning_rate": 0.00046386158475426284,
"loss": 3.0459,
"theoretical_loss": 4.200013145176587,
"tokens_seen": 269156352
},
{
"epoch": 0.08,
"learning_rate": 0.00046385155466399197,
"loss": 3.088,
"theoretical_loss": 4.199891052654689,
"tokens_seen": 269221888
},
{
"epoch": 0.08,
"learning_rate": 0.0004638415245737212,
"loss": 3.0612,
"theoretical_loss": 4.199768998169326,
"tokens_seen": 269287424
},
{
"epoch": 0.08,
"learning_rate": 0.00046383149448345033,
"loss": 3.0789,
"theoretical_loss": 4.199646981699395,
"tokens_seen": 269352960
},
{
"epoch": 0.08,
"learning_rate": 0.00046382146439317957,
"loss": 3.1749,
"theoretical_loss": 4.199525003223812,
"tokens_seen": 269418496
},
{
"epoch": 0.08,
"learning_rate": 0.00046381143430290875,
"loss": 3.162,
"theoretical_loss": 4.199403062721506,
"tokens_seen": 269484032
},
{
"epoch": 0.08,
"learning_rate": 0.00046380140421263793,
"loss": 3.1445,
"theoretical_loss": 4.199281160171427,
"tokens_seen": 269549568
},
{
"epoch": 0.08,
"learning_rate": 0.0004637913741223671,
"loss": 3.0702,
"theoretical_loss": 4.1991592955525405,
"tokens_seen": 269615104
},
{
"epoch": 0.08,
"learning_rate": 0.00046378134403209635,
"loss": 3.0764,
"theoretical_loss": 4.199037468843825,
"tokens_seen": 269680640
},
{
"epoch": 0.08,
"learning_rate": 0.0004637713139418255,
"loss": 3.1614,
"theoretical_loss": 4.198915680024282,
"tokens_seen": 269746176
},
{
"epoch": 0.08,
"learning_rate": 0.0004637612838515547,
"loss": 2.84,
"theoretical_loss": 4.198793929072925,
"tokens_seen": 269811712
},
{
"epoch": 0.08,
"learning_rate": 0.00046375125376128384,
"loss": 3.1708,
"theoretical_loss": 4.198672215968785,
"tokens_seen": 269877248
},
{
"epoch": 0.08,
"learning_rate": 0.00046374122367101307,
"loss": 3.0235,
"theoretical_loss": 4.198550540690912,
"tokens_seen": 269942784
},
{
"epoch": 0.08,
"learning_rate": 0.00046373119358074225,
"loss": 3.393,
"theoretical_loss": 4.198428903218371,
"tokens_seen": 270008320
},
{
"epoch": 0.08,
"learning_rate": 0.00046372116349047143,
"loss": 3.1363,
"theoretical_loss": 4.198307303530243,
"tokens_seen": 270073856
},
{
"epoch": 0.08,
"learning_rate": 0.0004637111334002006,
"loss": 3.0941,
"theoretical_loss": 4.198185741605628,
"tokens_seen": 270139392
},
{
"epoch": 0.08,
"learning_rate": 0.0004637011033099298,
"loss": 3.0913,
"theoretical_loss": 4.19806421742364,
"tokens_seen": 270204928
},
{
"epoch": 0.08,
"learning_rate": 0.000463691073219659,
"loss": 3.2218,
"theoretical_loss": 4.197942730963412,
"tokens_seen": 270270464
},
{
"epoch": 0.08,
"objective/train/docs_used": 164570,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8737075328826904,
"objective/train/theoretical_loss": 4.19782128220409,
"objective/train/tokens_used": 290796000,
"theoretical_loss": 4.19782128220409,
"tokens_seen": 270336000
},
{
"epoch": 0.08,
"learning_rate": 0.0004636810431293882,
"loss": 2.9242,
"theoretical_loss": 4.19782128220409,
"tokens_seen": 270336000
},
{
"epoch": 0.08,
"learning_rate": 0.00046367101303911734,
"loss": 2.8725,
"theoretical_loss": 4.19769987112484,
"tokens_seen": 270401536
},
{
"epoch": 0.08,
"learning_rate": 0.0004636609829488466,
"loss": 3.3748,
"theoretical_loss": 4.1975784977048445,
"tokens_seen": 270467072
},
{
"epoch": 0.08,
"learning_rate": 0.0004636509528585757,
"loss": 3.0672,
"theoretical_loss": 4.1974571619233,
"tokens_seen": 270532608
},
{
"epoch": 0.08,
"learning_rate": 0.00046364092276830494,
"loss": 3.1778,
"theoretical_loss": 4.197335863759422,
"tokens_seen": 270598144
},
{
"epoch": 0.08,
"learning_rate": 0.0004636308926780341,
"loss": 3.2507,
"theoretical_loss": 4.1972146031924416,
"tokens_seen": 270663680
},
{
"epoch": 0.08,
"learning_rate": 0.0004636208625877633,
"loss": 2.9583,
"theoretical_loss": 4.197093380201606,
"tokens_seen": 270729216
},
{
"epoch": 0.08,
"learning_rate": 0.0004636108324974925,
"loss": 3.1339,
"theoretical_loss": 4.196972194766179,
"tokens_seen": 270794752
},
{
"epoch": 0.08,
"learning_rate": 0.0004636008024072217,
"loss": 3.2131,
"theoretical_loss": 4.196851046865442,
"tokens_seen": 270860288
},
{
"epoch": 0.08,
"learning_rate": 0.00046359077231695084,
"loss": 3.0693,
"theoretical_loss": 4.1967299364786905,
"tokens_seen": 270925824
},
{
"epoch": 0.08,
"learning_rate": 0.0004635807422266801,
"loss": 3.1201,
"theoretical_loss": 4.196608863585239,
"tokens_seen": 270991360
},
{
"epoch": 0.08,
"learning_rate": 0.0004635707121364092,
"loss": 3.0753,
"theoretical_loss": 4.1964878281644165,
"tokens_seen": 271056896
},
{
"epoch": 0.08,
"learning_rate": 0.00046356068204613844,
"loss": 3.1767,
"theoretical_loss": 4.19636683019557,
"tokens_seen": 271122432
},
{
"epoch": 0.08,
"learning_rate": 0.0004635506519558676,
"loss": 3.1568,
"theoretical_loss": 4.196245869658061,
"tokens_seen": 271187968
},
{
"epoch": 0.08,
"learning_rate": 0.0004635406218655968,
"loss": 3.1067,
"theoretical_loss": 4.1961249465312696,
"tokens_seen": 271253504
},
{
"epoch": 0.08,
"learning_rate": 0.000463530591775326,
"loss": 3.1049,
"theoretical_loss": 4.196004060794589,
"tokens_seen": 271319040
},
{
"epoch": 0.08,
"learning_rate": 0.00046352056168505516,
"loss": 3.1558,
"theoretical_loss": 4.195883212427433,
"tokens_seen": 271384576
},
{
"epoch": 0.08,
"learning_rate": 0.00046351053159478434,
"loss": 3.104,
"theoretical_loss": 4.195762401409229,
"tokens_seen": 271450112
},
{
"epoch": 0.08,
"learning_rate": 0.0004635005015045136,
"loss": 2.9589,
"theoretical_loss": 4.19564162771942,
"tokens_seen": 271515648
},
{
"epoch": 0.08,
"learning_rate": 0.0004634904714142427,
"loss": 3.0468,
"theoretical_loss": 4.195520891337466,
"tokens_seen": 271581184
},
{
"epoch": 0.08,
"learning_rate": 0.00046348044132397194,
"loss": 2.9809,
"theoretical_loss": 4.195400192242845,
"tokens_seen": 271646720
},
{
"epoch": 0.08,
"learning_rate": 0.00046347041123370107,
"loss": 3.2978,
"theoretical_loss": 4.19527953041505,
"tokens_seen": 271712256
},
{
"epoch": 0.08,
"learning_rate": 0.0004634603811434303,
"loss": 3.4054,
"theoretical_loss": 4.19515890583359,
"tokens_seen": 271777792
},
{
"epoch": 0.08,
"learning_rate": 0.0004634503510531595,
"loss": 3.1847,
"theoretical_loss": 4.195038318477989,
"tokens_seen": 271843328
},
{
"epoch": 0.08,
"learning_rate": 0.00046344032096288867,
"loss": 3.1394,
"theoretical_loss": 4.194917768327789,
"tokens_seen": 271908864
},
{
"epoch": 0.08,
"objective/train/docs_used": 165961,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7331738471984863,
"objective/train/theoretical_loss": 4.194797255362549,
"objective/train/tokens_used": 292434400,
"theoretical_loss": 4.194797255362549,
"tokens_seen": 271974400
},
{
"epoch": 0.08,
"learning_rate": 0.00046343029087261785,
"loss": 2.8674,
"theoretical_loss": 4.194797255362549,
"tokens_seen": 271974400
},
{
"epoch": 0.08,
"learning_rate": 0.0004634202607823471,
"loss": 3.05,
"theoretical_loss": 4.194676779561841,
"tokens_seen": 272039936
},
{
"epoch": 0.08,
"learning_rate": 0.0004634102306920762,
"loss": 3.0387,
"theoretical_loss": 4.194556340905256,
"tokens_seen": 272105472
},
{
"epoch": 0.08,
"learning_rate": 0.00046340020060180545,
"loss": 3.0849,
"theoretical_loss": 4.194435939372401,
"tokens_seen": 272171008
},
{
"epoch": 0.08,
"learning_rate": 0.00046339017051153457,
"loss": 3.1541,
"theoretical_loss": 4.194315574942896,
"tokens_seen": 272236544
},
{
"epoch": 0.08,
"learning_rate": 0.0004633801404212638,
"loss": 3.2083,
"theoretical_loss": 4.194195247596381,
"tokens_seen": 272302080
},
{
"epoch": 0.08,
"learning_rate": 0.000463370110330993,
"loss": 3.2071,
"theoretical_loss": 4.19407495731251,
"tokens_seen": 272367616
},
{
"epoch": 0.08,
"learning_rate": 0.00046336008024072217,
"loss": 3.2115,
"theoretical_loss": 4.193954704070952,
"tokens_seen": 272433152
},
{
"epoch": 0.08,
"learning_rate": 0.00046335005015045135,
"loss": 3.1099,
"theoretical_loss": 4.193834487851396,
"tokens_seen": 272498688
},
{
"epoch": 0.08,
"learning_rate": 0.00046334002006018053,
"loss": 2.9455,
"theoretical_loss": 4.193714308633542,
"tokens_seen": 272564224
},
{
"epoch": 0.08,
"learning_rate": 0.0004633299899699097,
"loss": 3.2173,
"theoretical_loss": 4.1935941663971095,
"tokens_seen": 272629760
},
{
"epoch": 0.08,
"learning_rate": 0.00046331995987963895,
"loss": 3.0527,
"theoretical_loss": 4.193474061121833,
"tokens_seen": 272695296
},
{
"epoch": 0.08,
"learning_rate": 0.0004633099297893681,
"loss": 3.0777,
"theoretical_loss": 4.193353992787463,
"tokens_seen": 272760832
},
{
"epoch": 0.08,
"learning_rate": 0.0004632998996990973,
"loss": 3.1036,
"theoretical_loss": 4.193233961373766,
"tokens_seen": 272826368
},
{
"epoch": 0.08,
"learning_rate": 0.00046328986960882644,
"loss": 2.9951,
"theoretical_loss": 4.1931139668605235,
"tokens_seen": 272891904
},
{
"epoch": 0.08,
"learning_rate": 0.0004632798395185557,
"loss": 3.0116,
"theoretical_loss": 4.192994009227535,
"tokens_seen": 272957440
},
{
"epoch": 0.08,
"learning_rate": 0.00046326980942828485,
"loss": 2.8518,
"theoretical_loss": 4.192874088454613,
"tokens_seen": 273022976
},
{
"epoch": 0.08,
"learning_rate": 0.00046325977933801404,
"loss": 2.8654,
"theoretical_loss": 4.19275420452159,
"tokens_seen": 273088512
},
{
"epoch": 0.08,
"learning_rate": 0.0004632497492477432,
"loss": 3.0445,
"theoretical_loss": 4.192634357408309,
"tokens_seen": 273154048
},
{
"epoch": 0.08,
"learning_rate": 0.00046323971915747245,
"loss": 2.9804,
"theoretical_loss": 4.192514547094634,
"tokens_seen": 273219584
},
{
"epoch": 0.08,
"learning_rate": 0.0004632296890672016,
"loss": 3.0997,
"theoretical_loss": 4.192394773560441,
"tokens_seen": 273285120
},
{
"epoch": 0.08,
"learning_rate": 0.0004632196589769308,
"loss": 3.0125,
"theoretical_loss": 4.192275036785625,
"tokens_seen": 273350656
},
{
"epoch": 0.08,
"learning_rate": 0.00046320962888666,
"loss": 2.8637,
"theoretical_loss": 4.192155336750094,
"tokens_seen": 273416192
},
{
"epoch": 0.08,
"learning_rate": 0.0004631995987963892,
"loss": 3.1459,
"theoretical_loss": 4.192035673433773,
"tokens_seen": 273481728
},
{
"epoch": 0.08,
"learning_rate": 0.0004631895687061184,
"loss": 3.0333,
"theoretical_loss": 4.191916046816605,
"tokens_seen": 273547264
},
{
"epoch": 0.08,
"objective/train/docs_used": 166474,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.229163885116577,
"objective/train/theoretical_loss": 4.191796456878544,
"objective/train/tokens_used": 294072800,
"theoretical_loss": 4.191796456878544,
"tokens_seen": 273612800
},
{
"epoch": 0.08,
"learning_rate": 0.00046317953861584754,
"loss": 3.3893,
"theoretical_loss": 4.191796456878544,
"tokens_seen": 273612800
},
{
"epoch": 0.08,
"learning_rate": 0.0004631695085255768,
"loss": 3.2798,
"theoretical_loss": 4.191676903599563,
"tokens_seen": 273678336
},
{
"epoch": 0.08,
"learning_rate": 0.0004631594784353059,
"loss": 3.2739,
"theoretical_loss": 4.191557386959651,
"tokens_seen": 273743872
},
{
"epoch": 0.08,
"learning_rate": 0.00046314944834503514,
"loss": 2.8998,
"theoretical_loss": 4.191437906938811,
"tokens_seen": 273809408
},
{
"epoch": 0.08,
"learning_rate": 0.0004631394182547643,
"loss": 3.1002,
"theoretical_loss": 4.191318463517062,
"tokens_seen": 273874944
},
{
"epoch": 0.08,
"learning_rate": 0.0004631293881644935,
"loss": 3.052,
"theoretical_loss": 4.19119905667444,
"tokens_seen": 273940480
},
{
"epoch": 0.08,
"learning_rate": 0.0004631193580742227,
"loss": 3.0649,
"theoretical_loss": 4.191079686390996,
"tokens_seen": 274006016
},
{
"epoch": 0.08,
"learning_rate": 0.0004631093279839519,
"loss": 3.1153,
"theoretical_loss": 4.190960352646796,
"tokens_seen": 274071552
},
{
"epoch": 0.08,
"learning_rate": 0.00046309929789368104,
"loss": 3.0196,
"theoretical_loss": 4.190841055421921,
"tokens_seen": 274137088
},
{
"epoch": 0.08,
"learning_rate": 0.0004630892678034103,
"loss": 3.0416,
"theoretical_loss": 4.19072179469647,
"tokens_seen": 274202624
},
{
"epoch": 0.08,
"learning_rate": 0.0004630792377131394,
"loss": 3.0144,
"theoretical_loss": 4.190602570450556,
"tokens_seen": 274268160
},
{
"epoch": 0.08,
"learning_rate": 0.00046306920762286864,
"loss": 3.1422,
"theoretical_loss": 4.190483382664308,
"tokens_seen": 274333696
},
{
"epoch": 0.08,
"learning_rate": 0.0004630591775325978,
"loss": 3.1745,
"theoretical_loss": 4.19036423131787,
"tokens_seen": 274399232
},
{
"epoch": 0.08,
"learning_rate": 0.000463049147442327,
"loss": 3.0466,
"theoretical_loss": 4.190245116391403,
"tokens_seen": 274464768
},
{
"epoch": 0.08,
"learning_rate": 0.0004630391173520562,
"loss": 2.9194,
"theoretical_loss": 4.190126037865082,
"tokens_seen": 274530304
},
{
"epoch": 0.08,
"learning_rate": 0.00046302908726178536,
"loss": 3.2119,
"theoretical_loss": 4.190006995719098,
"tokens_seen": 274595840
},
{
"epoch": 0.08,
"learning_rate": 0.00046301905717151455,
"loss": 3.1011,
"theoretical_loss": 4.1898879899336565,
"tokens_seen": 274661376
},
{
"epoch": 0.08,
"learning_rate": 0.0004630090270812438,
"loss": 3.0468,
"theoretical_loss": 4.189769020488981,
"tokens_seen": 274726912
},
{
"epoch": 0.08,
"learning_rate": 0.0004629989969909729,
"loss": 3.368,
"theoretical_loss": 4.189650087365309,
"tokens_seen": 274792448
},
{
"epoch": 0.08,
"learning_rate": 0.00046298896690070214,
"loss": 3.0634,
"theoretical_loss": 4.189531190542893,
"tokens_seen": 274857984
},
{
"epoch": 0.08,
"learning_rate": 0.00046297893681043127,
"loss": 3.0413,
"theoretical_loss": 4.189412330002001,
"tokens_seen": 274923520
},
{
"epoch": 0.08,
"learning_rate": 0.0004629689067201605,
"loss": 3.3153,
"theoretical_loss": 4.189293505722918,
"tokens_seen": 274989056
},
{
"epoch": 0.08,
"learning_rate": 0.0004629588766298897,
"loss": 2.9856,
"theoretical_loss": 4.189174717685942,
"tokens_seen": 275054592
},
{
"epoch": 0.08,
"learning_rate": 0.00046294884653961887,
"loss": 3.1424,
"theoretical_loss": 4.189055965871389,
"tokens_seen": 275120128
},
{
"epoch": 0.08,
"learning_rate": 0.00046293881644934805,
"loss": 3.3162,
"theoretical_loss": 4.188937250259587,
"tokens_seen": 275185664
},
{
"epoch": 0.08,
"objective/train/docs_used": 167916,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9187827110290527,
"objective/train/theoretical_loss": 4.188818570830883,
"objective/train/tokens_used": 295711200,
"theoretical_loss": 4.188818570830883,
"tokens_seen": 275251200
},
{
"epoch": 0.08,
"learning_rate": 0.0004629287863590773,
"loss": 3.233,
"theoretical_loss": 4.188818570830883,
"tokens_seen": 275251200
},
{
"epoch": 0.08,
"learning_rate": 0.0004629187562688064,
"loss": 3.1334,
"theoretical_loss": 4.188699927565638,
"tokens_seen": 275316736
},
{
"epoch": 0.08,
"learning_rate": 0.00046290872617853565,
"loss": 3.2221,
"theoretical_loss": 4.188581320444228,
"tokens_seen": 275382272
},
{
"epoch": 0.08,
"learning_rate": 0.00046289869608826477,
"loss": 3.4486,
"theoretical_loss": 4.1884627494470426,
"tokens_seen": 275447808
},
{
"epoch": 0.08,
"learning_rate": 0.000462888665997994,
"loss": 2.7931,
"theoretical_loss": 4.1883442145544905,
"tokens_seen": 275513344
},
{
"epoch": 0.08,
"learning_rate": 0.0004628786359077232,
"loss": 3.002,
"theoretical_loss": 4.188225715746992,
"tokens_seen": 275578880
},
{
"epoch": 0.08,
"learning_rate": 0.00046286860581745237,
"loss": 3.2486,
"theoretical_loss": 4.188107253004986,
"tokens_seen": 275644416
},
{
"epoch": 0.08,
"learning_rate": 0.00046285857572718155,
"loss": 3.2086,
"theoretical_loss": 4.187988826308925,
"tokens_seen": 275709952
},
{
"epoch": 0.08,
"learning_rate": 0.00046284854563691073,
"loss": 2.9635,
"theoretical_loss": 4.187870435639275,
"tokens_seen": 275775488
},
{
"epoch": 0.08,
"learning_rate": 0.0004628385155466399,
"loss": 2.9808,
"theoretical_loss": 4.18775208097652,
"tokens_seen": 275841024
},
{
"epoch": 0.08,
"learning_rate": 0.00046282848545636915,
"loss": 3.2022,
"theoretical_loss": 4.187633762301159,
"tokens_seen": 275906560
},
{
"epoch": 0.08,
"learning_rate": 0.0004628184553660983,
"loss": 3.0849,
"theoretical_loss": 4.187515479593704,
"tokens_seen": 275972096
},
{
"epoch": 0.08,
"learning_rate": 0.0004628084252758275,
"loss": 3.0014,
"theoretical_loss": 4.187397232834683,
"tokens_seen": 276037632
},
{
"epoch": 0.08,
"learning_rate": 0.00046279839518555664,
"loss": 3.1524,
"theoretical_loss": 4.187279022004642,
"tokens_seen": 276103168
},
{
"epoch": 0.08,
"learning_rate": 0.0004627883650952859,
"loss": 3.1932,
"theoretical_loss": 4.1871608470841375,
"tokens_seen": 276168704
},
{
"epoch": 0.08,
"learning_rate": 0.00046277833500501505,
"loss": 2.89,
"theoretical_loss": 4.1870427080537445,
"tokens_seen": 276234240
},
{
"epoch": 0.08,
"learning_rate": 0.00046276830491474424,
"loss": 3.2589,
"theoretical_loss": 4.1869246048940525,
"tokens_seen": 276299776
},
{
"epoch": 0.08,
"learning_rate": 0.0004627582748244734,
"loss": 3.1626,
"theoretical_loss": 4.186806537585666,
"tokens_seen": 276365312
},
{
"epoch": 0.08,
"learning_rate": 0.00046274824473420265,
"loss": 3.1922,
"theoretical_loss": 4.186688506109202,
"tokens_seen": 276430848
},
{
"epoch": 0.08,
"learning_rate": 0.0004627382146439318,
"loss": 2.9536,
"theoretical_loss": 4.186570510445296,
"tokens_seen": 276496384
},
{
"epoch": 0.08,
"learning_rate": 0.000462728184553661,
"loss": 3.3615,
"theoretical_loss": 4.186452550574599,
"tokens_seen": 276561920
},
{
"epoch": 0.08,
"learning_rate": 0.00046271815446339014,
"loss": 3.2877,
"theoretical_loss": 4.186334626477774,
"tokens_seen": 276627456
},
{
"epoch": 0.08,
"learning_rate": 0.0004627081243731194,
"loss": 3.2169,
"theoretical_loss": 4.186216738135501,
"tokens_seen": 276692992
},
{
"epoch": 0.08,
"learning_rate": 0.00046269809428284856,
"loss": 3.0765,
"theoretical_loss": 4.186098885528473,
"tokens_seen": 276758528
},
{
"epoch": 0.08,
"learning_rate": 0.00046268806419257774,
"loss": 3.1812,
"theoretical_loss": 4.185981068637401,
"tokens_seen": 276824064
},
{
"epoch": 0.08,
"objective/train/docs_used": 168545,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.318019390106201,
"objective/train/theoretical_loss": 4.185863287443008,
"objective/train/tokens_used": 297349600,
"theoretical_loss": 4.185863287443008,
"tokens_seen": 276889600
},
{
"epoch": 0.08,
"learning_rate": 0.0004626780341023069,
"loss": 3.1149,
"theoretical_loss": 4.185863287443008,
"tokens_seen": 276889600
},
{
"epoch": 0.08,
"learning_rate": 0.0004626680040120361,
"loss": 3.1434,
"theoretical_loss": 4.185745541926035,
"tokens_seen": 276955136
},
{
"epoch": 0.08,
"learning_rate": 0.0004626579739217653,
"loss": 3.0085,
"theoretical_loss": 4.185627832067237,
"tokens_seen": 277020672
},
{
"epoch": 0.08,
"learning_rate": 0.0004626479438314945,
"loss": 3.1581,
"theoretical_loss": 4.1855101578473795,
"tokens_seen": 277086208
},
{
"epoch": 0.08,
"learning_rate": 0.00046263791374122364,
"loss": 3.2237,
"theoretical_loss": 4.18539251924725,
"tokens_seen": 277151744
},
{
"epoch": 0.08,
"learning_rate": 0.0004626278836509529,
"loss": 3.0262,
"theoretical_loss": 4.185274916247646,
"tokens_seen": 277217280
},
{
"epoch": 0.08,
"learning_rate": 0.000462617853560682,
"loss": 3.3536,
"theoretical_loss": 4.185157348829383,
"tokens_seen": 277282816
},
{
"epoch": 0.08,
"learning_rate": 0.00046260782347041124,
"loss": 3.1312,
"theoretical_loss": 4.185039816973289,
"tokens_seen": 277348352
},
{
"epoch": 0.08,
"learning_rate": 0.0004625977933801404,
"loss": 3.1212,
"theoretical_loss": 4.184922320660207,
"tokens_seen": 277413888
},
{
"epoch": 0.08,
"learning_rate": 0.0004625877632898696,
"loss": 3.0527,
"theoretical_loss": 4.184804859870997,
"tokens_seen": 277479424
},
{
"epoch": 0.08,
"learning_rate": 0.0004625777331995988,
"loss": 3.1812,
"theoretical_loss": 4.184687434586531,
"tokens_seen": 277544960
},
{
"epoch": 0.08,
"learning_rate": 0.000462567703109328,
"loss": 2.9691,
"theoretical_loss": 4.184570044787698,
"tokens_seen": 277610496
},
{
"epoch": 0.08,
"learning_rate": 0.00046255767301905715,
"loss": 3.073,
"theoretical_loss": 4.1844526904554,
"tokens_seen": 277676032
},
{
"epoch": 0.08,
"learning_rate": 0.0004625476429287864,
"loss": 3.0977,
"theoretical_loss": 4.184335371570556,
"tokens_seen": 277741568
},
{
"epoch": 0.08,
"learning_rate": 0.0004625376128385155,
"loss": 3.0928,
"theoretical_loss": 4.184218088114097,
"tokens_seen": 277807104
},
{
"epoch": 0.08,
"learning_rate": 0.00046252758274824475,
"loss": 3.1863,
"theoretical_loss": 4.1841008400669715,
"tokens_seen": 277872640
},
{
"epoch": 0.08,
"learning_rate": 0.0004625175526579739,
"loss": 3.3092,
"theoretical_loss": 4.183983627410142,
"tokens_seen": 277938176
},
{
"epoch": 0.08,
"learning_rate": 0.0004625075225677031,
"loss": 2.8105,
"theoretical_loss": 4.183866450124584,
"tokens_seen": 278003712
},
{
"epoch": 0.08,
"learning_rate": 0.0004624974924774323,
"loss": 3.1829,
"theoretical_loss": 4.18374930819129,
"tokens_seen": 278069248
},
{
"epoch": 0.08,
"learning_rate": 0.00046248746238716147,
"loss": 3.2848,
"theoretical_loss": 4.183632201591264,
"tokens_seen": 278134784
},
{
"epoch": 0.08,
"learning_rate": 0.00046247743229689065,
"loss": 2.9075,
"theoretical_loss": 4.18351513030553,
"tokens_seen": 278200320
},
{
"epoch": 0.08,
"learning_rate": 0.0004624674022066199,
"loss": 2.9081,
"theoretical_loss": 4.1833980943151206,
"tokens_seen": 278265856
},
{
"epoch": 0.08,
"learning_rate": 0.00046245737211634907,
"loss": 3.117,
"theoretical_loss": 4.183281093601087,
"tokens_seen": 278331392
},
{
"epoch": 0.08,
"learning_rate": 0.00046244734202607825,
"loss": 3.1446,
"theoretical_loss": 4.183164128144495,
"tokens_seen": 278396928
},
{
"epoch": 0.08,
"learning_rate": 0.0004624373119358075,
"loss": 3.1212,
"theoretical_loss": 4.183047197926422,
"tokens_seen": 278462464
},
{
"epoch": 0.08,
"objective/train/docs_used": 169743,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.211092233657837,
"objective/train/theoretical_loss": 4.182930302927963,
"objective/train/tokens_used": 298988000,
"theoretical_loss": 4.182930302927963,
"tokens_seen": 278528000
},
{
"epoch": 0.08,
"learning_rate": 0.0004624272818455366,
"loss": 2.9243,
"theoretical_loss": 4.182930302927963,
"tokens_seen": 278528000
},
{
"epoch": 0.08,
"learning_rate": 0.00046241725175526585,
"loss": 3.1498,
"theoretical_loss": 4.182813443130227,
"tokens_seen": 278593536
},
{
"epoch": 0.08,
"learning_rate": 0.000462407221664995,
"loss": 3.1976,
"theoretical_loss": 4.182696618514337,
"tokens_seen": 278659072
},
{
"epoch": 0.08,
"learning_rate": 0.0004623971915747242,
"loss": 2.9944,
"theoretical_loss": 4.18257982906143,
"tokens_seen": 278724608
},
{
"epoch": 0.08,
"learning_rate": 0.0004623871614844534,
"loss": 3.0975,
"theoretical_loss": 4.1824630747526585,
"tokens_seen": 278790144
},
{
"epoch": 0.08,
"learning_rate": 0.00046237713139418257,
"loss": 3.1447,
"theoretical_loss": 4.182346355569189,
"tokens_seen": 278855680
},
{
"epoch": 0.08,
"learning_rate": 0.00046236710130391175,
"loss": 3.0155,
"theoretical_loss": 4.182229671492204,
"tokens_seen": 278921216
},
{
"epoch": 0.08,
"learning_rate": 0.00046235707121364093,
"loss": 3.1009,
"theoretical_loss": 4.1821130225028975,
"tokens_seen": 278986752
},
{
"epoch": 0.08,
"learning_rate": 0.0004623470411233701,
"loss": 3.1349,
"theoretical_loss": 4.1819964085824815,
"tokens_seen": 279052288
},
{
"epoch": 0.08,
"learning_rate": 0.00046233701103309935,
"loss": 3.2392,
"theoretical_loss": 4.181879829712178,
"tokens_seen": 279117824
},
{
"epoch": 0.08,
"learning_rate": 0.0004623269809428285,
"loss": 2.8839,
"theoretical_loss": 4.181763285873231,
"tokens_seen": 279183360
},
{
"epoch": 0.08,
"learning_rate": 0.0004623169508525577,
"loss": 3.2192,
"theoretical_loss": 4.181646777046889,
"tokens_seen": 279248896
},
{
"epoch": 0.08,
"learning_rate": 0.00046230692076228684,
"loss": 3.1266,
"theoretical_loss": 4.181530303214423,
"tokens_seen": 279314432
},
{
"epoch": 0.08,
"learning_rate": 0.0004622968906720161,
"loss": 3.2385,
"theoretical_loss": 4.181413864357115,
"tokens_seen": 279379968
},
{
"epoch": 0.08,
"learning_rate": 0.00046228686058174525,
"loss": 3.0699,
"theoretical_loss": 4.181297460456262,
"tokens_seen": 279445504
},
{
"epoch": 0.08,
"learning_rate": 0.00046227683049147444,
"loss": 3.2547,
"theoretical_loss": 4.181181091493174,
"tokens_seen": 279511040
},
{
"epoch": 0.08,
"learning_rate": 0.0004622668004012036,
"loss": 3.0276,
"theoretical_loss": 4.181064757449178,
"tokens_seen": 279576576
},
{
"epoch": 0.08,
"learning_rate": 0.00046225677031093285,
"loss": 3.1922,
"theoretical_loss": 4.180948458305615,
"tokens_seen": 279642112
},
{
"epoch": 0.08,
"learning_rate": 0.000462246740220662,
"loss": 3.0343,
"theoretical_loss": 4.180832194043836,
"tokens_seen": 279707648
},
{
"epoch": 0.08,
"learning_rate": 0.0004622367101303912,
"loss": 3.1996,
"theoretical_loss": 4.180715964645213,
"tokens_seen": 279773184
},
{
"epoch": 0.08,
"learning_rate": 0.00046222668004012034,
"loss": 3.3317,
"theoretical_loss": 4.180599770091126,
"tokens_seen": 279838720
},
{
"epoch": 0.08,
"learning_rate": 0.0004622166499498496,
"loss": 2.9687,
"theoretical_loss": 4.180483610362975,
"tokens_seen": 279904256
},
{
"epoch": 0.08,
"learning_rate": 0.00046220661985957876,
"loss": 3.158,
"theoretical_loss": 4.18036748544217,
"tokens_seen": 279969792
},
{
"epoch": 0.08,
"learning_rate": 0.00046219658976930794,
"loss": 3.0762,
"theoretical_loss": 4.180251395310137,
"tokens_seen": 280035328
},
{
"epoch": 0.08,
"learning_rate": 0.0004621865596790371,
"loss": 3.1545,
"theoretical_loss": 4.1801353399483165,
"tokens_seen": 280100864
},
{
"epoch": 0.08,
"objective/train/docs_used": 170343,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.5833492279052734,
"objective/train/theoretical_loss": 4.180019319338163,
"objective/train/tokens_used": 300626400,
"theoretical_loss": 4.180019319338163,
"tokens_seen": 280166400
},
{
"epoch": 0.08,
"learning_rate": 0.0004621765295887663,
"loss": 3.2849,
"theoretical_loss": 4.180019319338163,
"tokens_seen": 280166400
},
{
"epoch": 0.08,
"learning_rate": 0.0004621664994984955,
"loss": 2.8302,
"theoretical_loss": 4.179903333461144,
"tokens_seen": 280231936
},
{
"epoch": 0.08,
"learning_rate": 0.0004621564694082247,
"loss": 2.8809,
"theoretical_loss": 4.179787382298744,
"tokens_seen": 280297472
},
{
"epoch": 0.08,
"learning_rate": 0.00046214643931795384,
"loss": 3.2571,
"theoretical_loss": 4.179671465832458,
"tokens_seen": 280363008
},
{
"epoch": 0.08,
"learning_rate": 0.0004621364092276831,
"loss": 2.9144,
"theoretical_loss": 4.179555584043799,
"tokens_seen": 280428544
},
{
"epoch": 0.08,
"learning_rate": 0.0004621263791374122,
"loss": 3.1233,
"theoretical_loss": 4.17943973691429,
"tokens_seen": 280494080
},
{
"epoch": 0.09,
"learning_rate": 0.00046211634904714144,
"loss": 3.0764,
"theoretical_loss": 4.179323924425472,
"tokens_seen": 280559616
},
{
"epoch": 0.09,
"learning_rate": 0.0004621063189568706,
"loss": 3.1949,
"theoretical_loss": 4.179208146558899,
"tokens_seen": 280625152
},
{
"epoch": 0.09,
"learning_rate": 0.0004620962888665998,
"loss": 3.1481,
"theoretical_loss": 4.1790924032961385,
"tokens_seen": 280690688
},
{
"epoch": 0.09,
"learning_rate": 0.000462086258776329,
"loss": 2.9628,
"theoretical_loss": 4.178976694618772,
"tokens_seen": 280756224
},
{
"epoch": 0.09,
"learning_rate": 0.0004620762286860582,
"loss": 3.1512,
"theoretical_loss": 4.178861020508395,
"tokens_seen": 280821760
},
{
"epoch": 0.09,
"learning_rate": 0.00046206619859578735,
"loss": 2.9415,
"theoretical_loss": 4.178745380946619,
"tokens_seen": 280887296
},
{
"epoch": 0.09,
"learning_rate": 0.0004620561685055166,
"loss": 3.1536,
"theoretical_loss": 4.178629775915066,
"tokens_seen": 280952832
},
{
"epoch": 0.09,
"learning_rate": 0.0004620461384152457,
"loss": 3.0706,
"theoretical_loss": 4.178514205395376,
"tokens_seen": 281018368
},
{
"epoch": 0.09,
"learning_rate": 0.00046203610832497495,
"loss": 3.0371,
"theoretical_loss": 4.178398669369201,
"tokens_seen": 281083904
},
{
"epoch": 0.09,
"learning_rate": 0.0004620260782347041,
"loss": 2.8429,
"theoretical_loss": 4.178283167818206,
"tokens_seen": 281149440
},
{
"epoch": 0.09,
"learning_rate": 0.0004620160481444333,
"loss": 3.1415,
"theoretical_loss": 4.178167700724073,
"tokens_seen": 281214976
},
{
"epoch": 0.09,
"learning_rate": 0.0004620060180541625,
"loss": 3.0451,
"theoretical_loss": 4.178052268068494,
"tokens_seen": 281280512
},
{
"epoch": 0.09,
"learning_rate": 0.00046199598796389167,
"loss": 3.0419,
"theoretical_loss": 4.177936869833179,
"tokens_seen": 281346048
},
{
"epoch": 0.09,
"learning_rate": 0.00046198595787362085,
"loss": 2.8053,
"theoretical_loss": 4.17782150599985,
"tokens_seen": 281411584
},
{
"epoch": 0.09,
"learning_rate": 0.0004619759277833501,
"loss": 2.7688,
"theoretical_loss": 4.1777061765502435,
"tokens_seen": 281477120
},
{
"epoch": 0.09,
"learning_rate": 0.0004619658976930792,
"loss": 3.022,
"theoretical_loss": 4.1775908814661085,
"tokens_seen": 281542656
},
{
"epoch": 0.09,
"learning_rate": 0.00046195586760280845,
"loss": 3.3437,
"theoretical_loss": 4.17747562072921,
"tokens_seen": 281608192
},
{
"epoch": 0.09,
"learning_rate": 0.00046194583751253763,
"loss": 2.9559,
"theoretical_loss": 4.177360394321325,
"tokens_seen": 281673728
},
{
"epoch": 0.09,
"learning_rate": 0.0004619358074222668,
"loss": 2.7982,
"theoretical_loss": 4.177245202224246,
"tokens_seen": 281739264
},
{
"epoch": 0.09,
"objective/train/docs_used": 171008,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.037534236907959,
"objective/train/theoretical_loss": 4.17713004441978,
"objective/train/tokens_used": 302264800,
"theoretical_loss": 4.17713004441978,
"tokens_seen": 281804800
},
{
"epoch": 0.09,
"learning_rate": 0.000461925777331996,
"loss": 3.1612,
"theoretical_loss": 4.17713004441978,
"tokens_seen": 281804800
},
{
"epoch": 0.09,
"learning_rate": 0.0004619157472417252,
"loss": 3.2436,
"theoretical_loss": 4.177014920889745,
"tokens_seen": 281870336
},
{
"epoch": 0.09,
"learning_rate": 0.00046190571715145435,
"loss": 3.0578,
"theoretical_loss": 4.176899831615974,
"tokens_seen": 281935872
},
{
"epoch": 0.09,
"learning_rate": 0.0004618956870611836,
"loss": 3.1397,
"theoretical_loss": 4.176784776580316,
"tokens_seen": 282001408
},
{
"epoch": 0.09,
"learning_rate": 0.0004618856569709127,
"loss": 3.2312,
"theoretical_loss": 4.176669755764632,
"tokens_seen": 282066944
},
{
"epoch": 0.09,
"learning_rate": 0.00046187562688064195,
"loss": 3.0992,
"theoretical_loss": 4.176554769150796,
"tokens_seen": 282132480
},
{
"epoch": 0.09,
"learning_rate": 0.0004618655967903711,
"loss": 3.1607,
"theoretical_loss": 4.176439816720697,
"tokens_seen": 282198016
},
{
"epoch": 0.09,
"learning_rate": 0.0004618555667001003,
"loss": 2.9453,
"theoretical_loss": 4.1763248984562376,
"tokens_seen": 282263552
},
{
"epoch": 0.09,
"learning_rate": 0.0004618455366098295,
"loss": 3.2318,
"theoretical_loss": 4.176210014339335,
"tokens_seen": 282329088
},
{
"epoch": 0.09,
"learning_rate": 0.0004618355065195587,
"loss": 3.171,
"theoretical_loss": 4.17609516435192,
"tokens_seen": 282394624
},
{
"epoch": 0.09,
"learning_rate": 0.00046182547642928786,
"loss": 3.0733,
"theoretical_loss": 4.1759803484759335,
"tokens_seen": 282460160
},
{
"epoch": 0.09,
"learning_rate": 0.00046181544633901704,
"loss": 3.0515,
"theoretical_loss": 4.175865566693336,
"tokens_seen": 282525696
},
{
"epoch": 0.09,
"learning_rate": 0.0004618054162487462,
"loss": 3.2269,
"theoretical_loss": 4.175750818986098,
"tokens_seen": 282591232
},
{
"epoch": 0.09,
"learning_rate": 0.00046179538615847545,
"loss": 3.2102,
"theoretical_loss": 4.1756361053362046,
"tokens_seen": 282656768
},
{
"epoch": 0.09,
"learning_rate": 0.0004617853560682046,
"loss": 3.2833,
"theoretical_loss": 4.1755214257256545,
"tokens_seen": 282722304
},
{
"epoch": 0.09,
"learning_rate": 0.0004617753259779338,
"loss": 2.9001,
"theoretical_loss": 4.17540678013646,
"tokens_seen": 282787840
},
{
"epoch": 0.09,
"learning_rate": 0.000461765295887663,
"loss": 2.989,
"theoretical_loss": 4.175292168550648,
"tokens_seen": 282853376
},
{
"epoch": 0.09,
"learning_rate": 0.0004617552657973922,
"loss": 3.2841,
"theoretical_loss": 4.175177590950257,
"tokens_seen": 282918912
},
{
"epoch": 0.09,
"learning_rate": 0.00046174523570712136,
"loss": 2.9466,
"theoretical_loss": 4.175063047317342,
"tokens_seen": 282984448
},
{
"epoch": 0.09,
"learning_rate": 0.00046173520561685054,
"loss": 2.8851,
"theoretical_loss": 4.174948537633968,
"tokens_seen": 283049984
},
{
"epoch": 0.09,
"learning_rate": 0.0004617251755265797,
"loss": 3.0796,
"theoretical_loss": 4.174834061882218,
"tokens_seen": 283115520
},
{
"epoch": 0.09,
"learning_rate": 0.00046171514543630896,
"loss": 3.0094,
"theoretical_loss": 4.1747196200441845,
"tokens_seen": 283181056
},
{
"epoch": 0.09,
"learning_rate": 0.00046170511534603814,
"loss": 3.125,
"theoretical_loss": 4.174605212101977,
"tokens_seen": 283246592
},
{
"epoch": 0.09,
"learning_rate": 0.0004616950852557673,
"loss": 3.1404,
"theoretical_loss": 4.174490838037716,
"tokens_seen": 283312128
},
{
"epoch": 0.09,
"learning_rate": 0.0004616850551654965,
"loss": 3.1135,
"theoretical_loss": 4.174376497833537,
"tokens_seen": 283377664
},
{
"epoch": 0.09,
"objective/train/docs_used": 172282,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8163797855377197,
"objective/train/theoretical_loss": 4.174262191471587,
"objective/train/tokens_used": 303903200,
"theoretical_loss": 4.174262191471587,
"tokens_seen": 283443200
},
{
"epoch": 0.09,
"learning_rate": 0.0004616750250752257,
"loss": 2.8791,
"theoretical_loss": 4.174262191471587,
"tokens_seen": 283443200
},
{
"epoch": 0.09,
"learning_rate": 0.0004616649949849549,
"loss": 2.9439,
"theoretical_loss": 4.17414791893403,
"tokens_seen": 283508736
},
{
"epoch": 0.09,
"learning_rate": 0.00046165496489468404,
"loss": 2.934,
"theoretical_loss": 4.17403368020304,
"tokens_seen": 283574272
},
{
"epoch": 0.09,
"learning_rate": 0.0004616449348044133,
"loss": 3.2267,
"theoretical_loss": 4.173919475260808,
"tokens_seen": 283639808
},
{
"epoch": 0.09,
"learning_rate": 0.0004616349047141424,
"loss": 3.2523,
"theoretical_loss": 4.173805304089536,
"tokens_seen": 283705344
},
{
"epoch": 0.09,
"learning_rate": 0.00046162487462387164,
"loss": 3.0485,
"theoretical_loss": 4.173691166671439,
"tokens_seen": 283770880
},
{
"epoch": 0.09,
"learning_rate": 0.0004616148445336008,
"loss": 3.3109,
"theoretical_loss": 4.173577062988748,
"tokens_seen": 283836416
},
{
"epoch": 0.09,
"learning_rate": 0.00046160481444333,
"loss": 3.2738,
"theoretical_loss": 4.173462993023706,
"tokens_seen": 283901952
},
{
"epoch": 0.09,
"learning_rate": 0.0004615947843530592,
"loss": 2.992,
"theoretical_loss": 4.173348956758568,
"tokens_seen": 283967488
},
{
"epoch": 0.09,
"learning_rate": 0.0004615847542627884,
"loss": 2.9936,
"theoretical_loss": 4.173234954175605,
"tokens_seen": 284033024
},
{
"epoch": 0.09,
"learning_rate": 0.00046157472417251755,
"loss": 3.3821,
"theoretical_loss": 4.173120985257102,
"tokens_seen": 284098560
},
{
"epoch": 0.09,
"learning_rate": 0.0004615646940822468,
"loss": 2.9413,
"theoretical_loss": 4.173007049985352,
"tokens_seen": 284164096
},
{
"epoch": 0.09,
"learning_rate": 0.0004615546639919759,
"loss": 2.9063,
"theoretical_loss": 4.172893148342667,
"tokens_seen": 284229632
},
{
"epoch": 0.09,
"learning_rate": 0.00046154463390170515,
"loss": 3.2355,
"theoretical_loss": 4.172779280311372,
"tokens_seen": 284295168
},
{
"epoch": 0.09,
"learning_rate": 0.0004615346038114343,
"loss": 3.1892,
"theoretical_loss": 4.172665445873801,
"tokens_seen": 284360704
},
{
"epoch": 0.09,
"learning_rate": 0.0004615245737211635,
"loss": 3.0414,
"theoretical_loss": 4.172551645012307,
"tokens_seen": 284426240
},
{
"epoch": 0.09,
"learning_rate": 0.0004615145436308927,
"loss": 2.9639,
"theoretical_loss": 4.1724378777092515,
"tokens_seen": 284491776
},
{
"epoch": 0.09,
"learning_rate": 0.00046150451354062187,
"loss": 3.2605,
"theoretical_loss": 4.172324143947012,
"tokens_seen": 284557312
},
{
"epoch": 0.09,
"learning_rate": 0.00046149448345035105,
"loss": 3.0332,
"theoretical_loss": 4.172210443707979,
"tokens_seen": 284622848
},
{
"epoch": 0.09,
"learning_rate": 0.0004614844533600803,
"loss": 3.0535,
"theoretical_loss": 4.1720967769745565,
"tokens_seen": 284688384
},
{
"epoch": 0.09,
"learning_rate": 0.0004614744232698094,
"loss": 2.8846,
"theoretical_loss": 4.171983143729159,
"tokens_seen": 284753920
},
{
"epoch": 0.09,
"learning_rate": 0.00046146439317953865,
"loss": 3.3499,
"theoretical_loss": 4.1718695439542195,
"tokens_seen": 284819456
},
{
"epoch": 0.09,
"learning_rate": 0.00046145436308926783,
"loss": 3.0726,
"theoretical_loss": 4.17175597763218,
"tokens_seen": 284884992
},
{
"epoch": 0.09,
"learning_rate": 0.000461444332998997,
"loss": 2.7706,
"theoretical_loss": 4.171642444745497,
"tokens_seen": 284950528
},
{
"epoch": 0.09,
"learning_rate": 0.0004614343029087262,
"loss": 3.0167,
"theoretical_loss": 4.1715289452766395,
"tokens_seen": 285016064
},
{
"epoch": 0.09,
"objective/train/docs_used": 172985,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.503798246383667,
"objective/train/theoretical_loss": 4.1714154792080915,
"objective/train/tokens_used": 305541600,
"theoretical_loss": 4.1714154792080915,
"tokens_seen": 285081600
},
{
"epoch": 0.09,
"learning_rate": 0.0004614242728184554,
"loss": 3.1825,
"theoretical_loss": 4.1714154792080915,
"tokens_seen": 285081600
},
{
"epoch": 0.09,
"learning_rate": 0.00046141424272818455,
"loss": 2.9299,
"theoretical_loss": 4.171302046522349,
"tokens_seen": 285147136
},
{
"epoch": 0.09,
"learning_rate": 0.0004614042126379138,
"loss": 2.9893,
"theoretical_loss": 4.171188647201921,
"tokens_seen": 285212672
},
{
"epoch": 0.09,
"learning_rate": 0.0004613941825476429,
"loss": 3.2218,
"theoretical_loss": 4.1710752812293315,
"tokens_seen": 285278208
},
{
"epoch": 0.09,
"learning_rate": 0.00046138415245737215,
"loss": 3.0444,
"theoretical_loss": 4.170961948587115,
"tokens_seen": 285343744
},
{
"epoch": 0.09,
"learning_rate": 0.0004613741223671013,
"loss": 3.0153,
"theoretical_loss": 4.17084864925782,
"tokens_seen": 285409280
},
{
"epoch": 0.09,
"learning_rate": 0.0004613640922768305,
"loss": 3.0765,
"theoretical_loss": 4.1707353832240095,
"tokens_seen": 285474816
},
{
"epoch": 0.09,
"learning_rate": 0.0004613540621865597,
"loss": 3.1904,
"theoretical_loss": 4.170622150468258,
"tokens_seen": 285540352
},
{
"epoch": 0.09,
"learning_rate": 0.0004613440320962889,
"loss": 3.2306,
"theoretical_loss": 4.170508950973154,
"tokens_seen": 285605888
},
{
"epoch": 0.09,
"learning_rate": 0.00046133400200601806,
"loss": 3.3067,
"theoretical_loss": 4.1703957847213,
"tokens_seen": 285671424
},
{
"epoch": 0.09,
"learning_rate": 0.00046132397191574724,
"loss": 3.0881,
"theoretical_loss": 4.170282651695308,
"tokens_seen": 285736960
},
{
"epoch": 0.09,
"learning_rate": 0.0004613139418254764,
"loss": 2.9596,
"theoretical_loss": 4.170169551877808,
"tokens_seen": 285802496
},
{
"epoch": 0.09,
"learning_rate": 0.00046130391173520566,
"loss": 3.2447,
"theoretical_loss": 4.170056485251439,
"tokens_seen": 285868032
},
{
"epoch": 0.09,
"learning_rate": 0.0004612938816449348,
"loss": 3.2723,
"theoretical_loss": 4.169943451798856,
"tokens_seen": 285933568
},
{
"epoch": 0.09,
"learning_rate": 0.000461283851554664,
"loss": 3.0463,
"theoretical_loss": 4.169830451502724,
"tokens_seen": 285999104
},
{
"epoch": 0.09,
"learning_rate": 0.0004612738214643932,
"loss": 2.924,
"theoretical_loss": 4.169717484345725,
"tokens_seen": 286064640
},
{
"epoch": 0.09,
"learning_rate": 0.0004612637913741224,
"loss": 3.0269,
"theoretical_loss": 4.1696045503105506,
"tokens_seen": 286130176
},
{
"epoch": 0.09,
"learning_rate": 0.00046125376128385156,
"loss": 3.1089,
"theoretical_loss": 4.169491649379905,
"tokens_seen": 286195712
},
{
"epoch": 0.09,
"learning_rate": 0.00046124373119358074,
"loss": 3.0891,
"theoretical_loss": 4.169378781536509,
"tokens_seen": 286261248
},
{
"epoch": 0.09,
"learning_rate": 0.0004612337011033099,
"loss": 3.0611,
"theoretical_loss": 4.169265946763095,
"tokens_seen": 286326784
},
{
"epoch": 0.09,
"learning_rate": 0.00046122367101303916,
"loss": 3.0257,
"theoretical_loss": 4.169153145042405,
"tokens_seen": 286392320
},
{
"epoch": 0.09,
"learning_rate": 0.0004612136409227683,
"loss": 3.1311,
"theoretical_loss": 4.169040376357199,
"tokens_seen": 286457856
},
{
"epoch": 0.09,
"learning_rate": 0.0004612036108324975,
"loss": 3.2128,
"theoretical_loss": 4.168927640690246,
"tokens_seen": 286523392
},
{
"epoch": 0.09,
"learning_rate": 0.00046119358074222665,
"loss": 2.8733,
"theoretical_loss": 4.16881493802433,
"tokens_seen": 286588928
},
{
"epoch": 0.09,
"learning_rate": 0.0004611835506519559,
"loss": 2.9867,
"theoretical_loss": 4.168702268342248,
"tokens_seen": 286654464
},
{
"epoch": 0.09,
"objective/train/docs_used": 174327,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1925644874572754,
"objective/train/theoretical_loss": 4.168589631626808,
"objective/train/tokens_used": 307180000,
"theoretical_loss": 4.168589631626808,
"tokens_seen": 286720000
},
{
"epoch": 0.09,
"learning_rate": 0.00046117352056168506,
"loss": 3.043,
"theoretical_loss": 4.168589631626808,
"tokens_seen": 286720000
},
{
"epoch": 0.09,
"learning_rate": 0.00046116349047141425,
"loss": 2.8574,
"theoretical_loss": 4.168477027860833,
"tokens_seen": 286785536
},
{
"epoch": 0.09,
"learning_rate": 0.0004611534603811434,
"loss": 3.0897,
"theoretical_loss": 4.168364457027158,
"tokens_seen": 286851072
},
{
"epoch": 0.09,
"learning_rate": 0.0004611434302908726,
"loss": 3.0685,
"theoretical_loss": 4.168251919108632,
"tokens_seen": 286916608
},
{
"epoch": 0.09,
"learning_rate": 0.0004611334002006018,
"loss": 3.356,
"theoretical_loss": 4.168139414088113,
"tokens_seen": 286982144
},
{
"epoch": 0.09,
"learning_rate": 0.000461123370110331,
"loss": 3.1332,
"theoretical_loss": 4.168026941948478,
"tokens_seen": 287047680
},
{
"epoch": 0.09,
"learning_rate": 0.00046111334002006015,
"loss": 3.1427,
"theoretical_loss": 4.167914502672611,
"tokens_seen": 287113216
},
{
"epoch": 0.09,
"learning_rate": 0.0004611033099297894,
"loss": 3.263,
"theoretical_loss": 4.1678020962434115,
"tokens_seen": 287178752
},
{
"epoch": 0.09,
"learning_rate": 0.00046109327983951857,
"loss": 2.847,
"theoretical_loss": 4.167689722643792,
"tokens_seen": 287244288
},
{
"epoch": 0.09,
"learning_rate": 0.00046108324974924775,
"loss": 3.1922,
"theoretical_loss": 4.1675773818566775,
"tokens_seen": 287309824
},
{
"epoch": 0.09,
"learning_rate": 0.00046107321965897693,
"loss": 3.224,
"theoretical_loss": 4.167465073865006,
"tokens_seen": 287375360
},
{
"epoch": 0.09,
"learning_rate": 0.0004610631895687061,
"loss": 3.2108,
"theoretical_loss": 4.167352798651726,
"tokens_seen": 287440896
},
{
"epoch": 0.09,
"learning_rate": 0.0004610531594784353,
"loss": 2.9368,
"theoretical_loss": 4.167240556199802,
"tokens_seen": 287506432
},
{
"epoch": 0.09,
"learning_rate": 0.0004610431293881645,
"loss": 3.025,
"theoretical_loss": 4.167128346492211,
"tokens_seen": 287571968
},
{
"epoch": 0.09,
"learning_rate": 0.00046103309929789365,
"loss": 3.1749,
"theoretical_loss": 4.16701616951194,
"tokens_seen": 287637504
},
{
"epoch": 0.09,
"learning_rate": 0.0004610230692076229,
"loss": 3.2276,
"theoretical_loss": 4.1669040252419896,
"tokens_seen": 287703040
},
{
"epoch": 0.09,
"learning_rate": 0.000461013039117352,
"loss": 3.1166,
"theoretical_loss": 4.166791913665375,
"tokens_seen": 287768576
},
{
"epoch": 0.09,
"learning_rate": 0.00046100300902708125,
"loss": 3.2476,
"theoretical_loss": 4.166679834765123,
"tokens_seen": 287834112
},
{
"epoch": 0.09,
"learning_rate": 0.00046099297893681043,
"loss": 3.2406,
"theoretical_loss": 4.166567788524272,
"tokens_seen": 287899648
},
{
"epoch": 0.09,
"learning_rate": 0.0004609829488465396,
"loss": 3.1236,
"theoretical_loss": 4.166455774925875,
"tokens_seen": 287965184
},
{
"epoch": 0.09,
"learning_rate": 0.0004609729187562688,
"loss": 2.6893,
"theoretical_loss": 4.166343793952995,
"tokens_seen": 288030720
},
{
"epoch": 0.09,
"learning_rate": 0.00046096288866599803,
"loss": 2.9429,
"theoretical_loss": 4.166231845588712,
"tokens_seen": 288096256
},
{
"epoch": 0.09,
"learning_rate": 0.0004609528585757272,
"loss": 3.1807,
"theoretical_loss": 4.166119929816113,
"tokens_seen": 288161792
},
{
"epoch": 0.09,
"learning_rate": 0.0004609428284854564,
"loss": 3.054,
"theoretical_loss": 4.166008046618303,
"tokens_seen": 288227328
},
{
"epoch": 0.09,
"learning_rate": 0.0004609327983951856,
"loss": 3.1462,
"theoretical_loss": 4.1658961959783944,
"tokens_seen": 288292864
},
{
"epoch": 0.09,
"objective/train/docs_used": 174865,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1643199920654297,
"objective/train/theoretical_loss": 4.165784377879517,
"objective/train/tokens_used": 308818400,
"theoretical_loss": 4.165784377879517,
"tokens_seen": 288358400
},
{
"epoch": 0.09,
"learning_rate": 0.00046092276830491475,
"loss": 3.2366,
"theoretical_loss": 4.165784377879517,
"tokens_seen": 288358400
},
{
"epoch": 0.09,
"learning_rate": 0.000460912738214644,
"loss": 2.9021,
"theoretical_loss": 4.165672592304811,
"tokens_seen": 288423936
},
{
"epoch": 0.09,
"learning_rate": 0.0004609027081243731,
"loss": 3.0573,
"theoretical_loss": 4.165560839237429,
"tokens_seen": 288489472
},
{
"epoch": 0.09,
"learning_rate": 0.00046089267803410235,
"loss": 3.0862,
"theoretical_loss": 4.165449118660536,
"tokens_seen": 288555008
},
{
"epoch": 0.09,
"learning_rate": 0.0004608826479438315,
"loss": 3.149,
"theoretical_loss": 4.16533743055731,
"tokens_seen": 288620544
},
{
"epoch": 0.09,
"learning_rate": 0.0004608726178535607,
"loss": 3.21,
"theoretical_loss": 4.165225774910941,
"tokens_seen": 288686080
},
{
"epoch": 0.09,
"learning_rate": 0.0004608625877632899,
"loss": 2.8823,
"theoretical_loss": 4.165114151704634,
"tokens_seen": 288751616
},
{
"epoch": 0.09,
"learning_rate": 0.0004608525576730191,
"loss": 3.0306,
"theoretical_loss": 4.165002560921601,
"tokens_seen": 288817152
},
{
"epoch": 0.09,
"learning_rate": 0.00046084252758274826,
"loss": 3.2771,
"theoretical_loss": 4.164891002545073,
"tokens_seen": 288882688
},
{
"epoch": 0.09,
"learning_rate": 0.00046083249749247744,
"loss": 2.7558,
"theoretical_loss": 4.16477947655829,
"tokens_seen": 288948224
},
{
"epoch": 0.09,
"learning_rate": 0.0004608224674022066,
"loss": 3.0144,
"theoretical_loss": 4.164667982944504,
"tokens_seen": 289013760
},
{
"epoch": 0.09,
"learning_rate": 0.00046081243731193586,
"loss": 2.9461,
"theoretical_loss": 4.164556521686981,
"tokens_seen": 289079296
},
{
"epoch": 0.09,
"learning_rate": 0.000460802407221665,
"loss": 3.1734,
"theoretical_loss": 4.1644450927689975,
"tokens_seen": 289144832
},
{
"epoch": 0.09,
"learning_rate": 0.0004607923771313942,
"loss": 3.2232,
"theoretical_loss": 4.164333696173846,
"tokens_seen": 289210368
},
{
"epoch": 0.09,
"learning_rate": 0.0004607823470411234,
"loss": 3.0528,
"theoretical_loss": 4.164222331884827,
"tokens_seen": 289275904
},
{
"epoch": 0.09,
"learning_rate": 0.0004607723169508526,
"loss": 3.0869,
"theoretical_loss": 4.164110999885256,
"tokens_seen": 289341440
},
{
"epoch": 0.09,
"learning_rate": 0.00046076228686058176,
"loss": 3.0237,
"theoretical_loss": 4.163999700158462,
"tokens_seen": 289406976
},
{
"epoch": 0.09,
"learning_rate": 0.00046075225677031094,
"loss": 3.0699,
"theoretical_loss": 4.163888432687784,
"tokens_seen": 289472512
},
{
"epoch": 0.09,
"learning_rate": 0.0004607422266800401,
"loss": 2.7908,
"theoretical_loss": 4.163777197456573,
"tokens_seen": 289538048
},
{
"epoch": 0.09,
"learning_rate": 0.00046073219658976936,
"loss": 3.0935,
"theoretical_loss": 4.163665994448197,
"tokens_seen": 289603584
},
{
"epoch": 0.09,
"learning_rate": 0.0004607221664994985,
"loss": 2.9748,
"theoretical_loss": 4.163554823646027,
"tokens_seen": 289669120
},
{
"epoch": 0.09,
"learning_rate": 0.0004607121364092277,
"loss": 3.0475,
"theoretical_loss": 4.163443685033458,
"tokens_seen": 289734656
},
{
"epoch": 0.09,
"learning_rate": 0.00046070210631895685,
"loss": 2.9678,
"theoretical_loss": 4.163332578593889,
"tokens_seen": 289800192
},
{
"epoch": 0.09,
"learning_rate": 0.0004606920762286861,
"loss": 3.0884,
"theoretical_loss": 4.163221504310734,
"tokens_seen": 289865728
},
{
"epoch": 0.09,
"learning_rate": 0.00046068204613841526,
"loss": 3.1743,
"theoretical_loss": 4.1631104621674195,
"tokens_seen": 289931264
},
{
"epoch": 0.09,
"objective/train/docs_used": 176137,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.816805124282837,
"objective/train/theoretical_loss": 4.162999452147384,
"objective/train/tokens_used": 310456800,
"theoretical_loss": 4.162999452147384,
"tokens_seen": 289996800
},
{
"epoch": 0.09,
"learning_rate": 0.00046067201604814445,
"loss": 2.9639,
"theoretical_loss": 4.162999452147384,
"tokens_seen": 289996800
},
{
"epoch": 0.09,
"learning_rate": 0.0004606619859578736,
"loss": 2.9778,
"theoretical_loss": 4.1628884742340775,
"tokens_seen": 290062336
},
{
"epoch": 0.09,
"learning_rate": 0.0004606519558676028,
"loss": 3.0018,
"theoretical_loss": 4.162777528410963,
"tokens_seen": 290127872
},
{
"epoch": 0.09,
"learning_rate": 0.000460641925777332,
"loss": 2.9703,
"theoretical_loss": 4.162666614661518,
"tokens_seen": 290193408
},
{
"epoch": 0.09,
"learning_rate": 0.0004606318956870612,
"loss": 3.2376,
"theoretical_loss": 4.162555732969227,
"tokens_seen": 290258944
},
{
"epoch": 0.09,
"learning_rate": 0.00046062186559679035,
"loss": 3.0843,
"theoretical_loss": 4.162444883317591,
"tokens_seen": 290324480
},
{
"epoch": 0.09,
"learning_rate": 0.0004606118355065196,
"loss": 3.1161,
"theoretical_loss": 4.162334065690123,
"tokens_seen": 290390016
},
{
"epoch": 0.09,
"learning_rate": 0.00046060180541624877,
"loss": 3.0975,
"theoretical_loss": 4.162223280070345,
"tokens_seen": 290455552
},
{
"epoch": 0.09,
"learning_rate": 0.00046059177532597795,
"loss": 3.0539,
"theoretical_loss": 4.1621125264417955,
"tokens_seen": 290521088
},
{
"epoch": 0.09,
"learning_rate": 0.00046058174523570713,
"loss": 3.2598,
"theoretical_loss": 4.162001804788021,
"tokens_seen": 290586624
},
{
"epoch": 0.09,
"learning_rate": 0.0004605717151454363,
"loss": 3.1444,
"theoretical_loss": 4.161891115092583,
"tokens_seen": 290652160
},
{
"epoch": 0.09,
"learning_rate": 0.0004605616850551655,
"loss": 3.164,
"theoretical_loss": 4.161780457339055,
"tokens_seen": 290717696
},
{
"epoch": 0.09,
"learning_rate": 0.0004605516549648947,
"loss": 3.1421,
"theoretical_loss": 4.161669831511022,
"tokens_seen": 290783232
},
{
"epoch": 0.09,
"learning_rate": 0.00046054162487462385,
"loss": 3.167,
"theoretical_loss": 4.16155923759208,
"tokens_seen": 290848768
},
{
"epoch": 0.09,
"learning_rate": 0.0004605315947843531,
"loss": 3.0247,
"theoretical_loss": 4.161448675565838,
"tokens_seen": 290914304
},
{
"epoch": 0.09,
"learning_rate": 0.0004605215646940822,
"loss": 3.2994,
"theoretical_loss": 4.161338145415918,
"tokens_seen": 290979840
},
{
"epoch": 0.09,
"learning_rate": 0.00046051153460381145,
"loss": 2.9411,
"theoretical_loss": 4.161227647125955,
"tokens_seen": 291045376
},
{
"epoch": 0.09,
"learning_rate": 0.00046050150451354063,
"loss": 2.9685,
"theoretical_loss": 4.161117180679591,
"tokens_seen": 291110912
},
{
"epoch": 0.09,
"learning_rate": 0.0004604914744232698,
"loss": 2.8291,
"theoretical_loss": 4.161006746060488,
"tokens_seen": 291176448
},
{
"epoch": 0.09,
"learning_rate": 0.000460481444332999,
"loss": 3.0402,
"theoretical_loss": 4.160896343252311,
"tokens_seen": 291241984
},
{
"epoch": 0.09,
"learning_rate": 0.00046047141424272823,
"loss": 3.0417,
"theoretical_loss": 4.160785972238745,
"tokens_seen": 291307520
},
{
"epoch": 0.09,
"learning_rate": 0.00046046138415245736,
"loss": 3.3467,
"theoretical_loss": 4.160675633003484,
"tokens_seen": 291373056
},
{
"epoch": 0.09,
"learning_rate": 0.0004604513540621866,
"loss": 3.4501,
"theoretical_loss": 4.16056532553023,
"tokens_seen": 291438592
},
{
"epoch": 0.09,
"learning_rate": 0.0004604413239719157,
"loss": 2.9161,
"theoretical_loss": 4.160455049802706,
"tokens_seen": 291504128
},
{
"epoch": 0.09,
"learning_rate": 0.00046043129388164495,
"loss": 3.0043,
"theoretical_loss": 4.1603448058046375,
"tokens_seen": 291569664
},
{
"epoch": 0.09,
"objective/train/docs_used": 176478,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.099493980407715,
"objective/train/theoretical_loss": 4.160234593519768,
"objective/train/tokens_used": 312095200,
"theoretical_loss": 4.160234593519768,
"tokens_seen": 291635200
},
{
"epoch": 0.09,
"learning_rate": 0.00046042126379137414,
"loss": 3.4061,
"theoretical_loss": 4.160234593519768,
"tokens_seen": 291635200
},
{
"epoch": 0.09,
"learning_rate": 0.0004604112337011033,
"loss": 2.9333,
"theoretical_loss": 4.160124412931852,
"tokens_seen": 291700736
},
{
"epoch": 0.09,
"learning_rate": 0.0004604012036108325,
"loss": 3.2184,
"theoretical_loss": 4.160014264024654,
"tokens_seen": 291766272
},
{
"epoch": 0.09,
"learning_rate": 0.0004603911735205617,
"loss": 3.0181,
"theoretical_loss": 4.159904146781952,
"tokens_seen": 291831808
},
{
"epoch": 0.09,
"learning_rate": 0.00046038114343029086,
"loss": 2.9699,
"theoretical_loss": 4.159794061187536,
"tokens_seen": 291897344
},
{
"epoch": 0.09,
"learning_rate": 0.0004603711133400201,
"loss": 3.0901,
"theoretical_loss": 4.1596840072252075,
"tokens_seen": 291962880
},
{
"epoch": 0.09,
"learning_rate": 0.0004603610832497492,
"loss": 3.0625,
"theoretical_loss": 4.159573984878779,
"tokens_seen": 292028416
},
{
"epoch": 0.09,
"learning_rate": 0.00046035105315947846,
"loss": 3.0933,
"theoretical_loss": 4.159463994132079,
"tokens_seen": 292093952
},
{
"epoch": 0.09,
"learning_rate": 0.0004603410230692076,
"loss": 3.1878,
"theoretical_loss": 4.15935403496894,
"tokens_seen": 292159488
},
{
"epoch": 0.09,
"learning_rate": 0.0004603309929789368,
"loss": 3.2494,
"theoretical_loss": 4.159244107373215,
"tokens_seen": 292225024
},
{
"epoch": 0.09,
"learning_rate": 0.000460320962888666,
"loss": 3.1238,
"theoretical_loss": 4.159134211328765,
"tokens_seen": 292290560
},
{
"epoch": 0.09,
"learning_rate": 0.0004603109327983952,
"loss": 3.2343,
"theoretical_loss": 4.159024346819461,
"tokens_seen": 292356096
},
{
"epoch": 0.09,
"learning_rate": 0.00046030090270812436,
"loss": 2.9651,
"theoretical_loss": 4.158914513829189,
"tokens_seen": 292421632
},
{
"epoch": 0.09,
"learning_rate": 0.0004602908726178536,
"loss": 3.1782,
"theoretical_loss": 4.158804712341845,
"tokens_seen": 292487168
},
{
"epoch": 0.09,
"learning_rate": 0.0004602808425275827,
"loss": 3.2679,
"theoretical_loss": 4.158694942341338,
"tokens_seen": 292552704
},
{
"epoch": 0.09,
"learning_rate": 0.00046027081243731196,
"loss": 3.0873,
"theoretical_loss": 4.1585852038115885,
"tokens_seen": 292618240
},
{
"epoch": 0.09,
"learning_rate": 0.0004602607823470411,
"loss": 3.0467,
"theoretical_loss": 4.1584754967365285,
"tokens_seen": 292683776
},
{
"epoch": 0.09,
"learning_rate": 0.0004602507522567703,
"loss": 3.2751,
"theoretical_loss": 4.1583658211001016,
"tokens_seen": 292749312
},
{
"epoch": 0.09,
"learning_rate": 0.0004602407221664995,
"loss": 3.3058,
"theoretical_loss": 4.158256176886264,
"tokens_seen": 292814848
},
{
"epoch": 0.09,
"learning_rate": 0.0004602306920762287,
"loss": 3.0265,
"theoretical_loss": 4.158146564078982,
"tokens_seen": 292880384
},
{
"epoch": 0.09,
"learning_rate": 0.00046022066198595787,
"loss": 3.0768,
"theoretical_loss": 4.158036982662237,
"tokens_seen": 292945920
},
{
"epoch": 0.09,
"learning_rate": 0.00046021063189568705,
"loss": 2.8461,
"theoretical_loss": 4.157927432620018,
"tokens_seen": 293011456
},
{
"epoch": 0.09,
"learning_rate": 0.0004602006018054163,
"loss": 3.1769,
"theoretical_loss": 4.157817913936329,
"tokens_seen": 293076992
},
{
"epoch": 0.09,
"learning_rate": 0.00046019057171514546,
"loss": 2.9548,
"theoretical_loss": 4.157708426595184,
"tokens_seen": 293142528
},
{
"epoch": 0.09,
"learning_rate": 0.00046018054162487465,
"loss": 2.9129,
"theoretical_loss": 4.157598970580608,
"tokens_seen": 293208064
},
{
"epoch": 0.09,
"objective/train/docs_used": 177825,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.218212842941284,
"objective/train/theoretical_loss": 4.157489545876642,
"objective/train/tokens_used": 313733600,
"theoretical_loss": 4.157489545876642,
"tokens_seen": 293273600
},
{
"epoch": 0.09,
"learning_rate": 0.0004601705115346038,
"loss": 3.1663,
"theoretical_loss": 4.157489545876642,
"tokens_seen": 293273600
},
{
"epoch": 0.09,
"learning_rate": 0.000460160481444333,
"loss": 3.228,
"theoretical_loss": 4.157380152467333,
"tokens_seen": 293339136
},
{
"epoch": 0.09,
"learning_rate": 0.0004601504513540622,
"loss": 3.1855,
"theoretical_loss": 4.157270790336742,
"tokens_seen": 293404672
},
{
"epoch": 0.09,
"learning_rate": 0.0004601404212637914,
"loss": 3.1031,
"theoretical_loss": 4.157161459468944,
"tokens_seen": 293470208
},
{
"epoch": 0.09,
"learning_rate": 0.00046013039117352055,
"loss": 3.11,
"theoretical_loss": 4.157052159848023,
"tokens_seen": 293535744
},
{
"epoch": 0.09,
"learning_rate": 0.0004601203610832498,
"loss": 2.8797,
"theoretical_loss": 4.156942891458074,
"tokens_seen": 293601280
},
{
"epoch": 0.09,
"learning_rate": 0.00046011033099297897,
"loss": 3.1523,
"theoretical_loss": 4.156833654283207,
"tokens_seen": 293666816
},
{
"epoch": 0.09,
"learning_rate": 0.00046010030090270815,
"loss": 2.9046,
"theoretical_loss": 4.15672444830754,
"tokens_seen": 293732352
},
{
"epoch": 0.09,
"learning_rate": 0.00046009027081243733,
"loss": 3.1712,
"theoretical_loss": 4.156615273515205,
"tokens_seen": 293797888
},
{
"epoch": 0.09,
"learning_rate": 0.0004600802407221665,
"loss": 3.1373,
"theoretical_loss": 4.156506129890344,
"tokens_seen": 293863424
},
{
"epoch": 0.09,
"learning_rate": 0.0004600702106318957,
"loss": 3.2344,
"theoretical_loss": 4.156397017417111,
"tokens_seen": 293928960
},
{
"epoch": 0.09,
"learning_rate": 0.00046006018054162493,
"loss": 3.2112,
"theoretical_loss": 4.156287936079675,
"tokens_seen": 293994496
},
{
"epoch": 0.09,
"learning_rate": 0.00046005015045135405,
"loss": 3.186,
"theoretical_loss": 4.156178885862209,
"tokens_seen": 294060032
},
{
"epoch": 0.09,
"learning_rate": 0.0004600401203610833,
"loss": 2.9581,
"theoretical_loss": 4.156069866748906,
"tokens_seen": 294125568
},
{
"epoch": 0.09,
"learning_rate": 0.0004600300902708124,
"loss": 3.0633,
"theoretical_loss": 4.155960878723965,
"tokens_seen": 294191104
},
{
"epoch": 0.09,
"learning_rate": 0.00046002006018054165,
"loss": 2.9769,
"theoretical_loss": 4.155851921771598,
"tokens_seen": 294256640
},
{
"epoch": 0.09,
"learning_rate": 0.00046001003009027083,
"loss": 3.154,
"theoretical_loss": 4.155742995876029,
"tokens_seen": 294322176
},
{
"epoch": 0.09,
"learning_rate": 0.00046,
"loss": 3.2509,
"theoretical_loss": 4.155634101021494,
"tokens_seen": 294387712
},
{
"epoch": 0.09,
"learning_rate": 0.0004599899699097292,
"loss": 2.9986,
"theoretical_loss": 4.155525237192238,
"tokens_seen": 294453248
},
{
"epoch": 0.09,
"learning_rate": 0.00045997993981945843,
"loss": 3.0416,
"theoretical_loss": 4.155416404372522,
"tokens_seen": 294518784
},
{
"epoch": 0.09,
"learning_rate": 0.00045996990972918756,
"loss": 3.113,
"theoretical_loss": 4.155307602546614,
"tokens_seen": 294584320
},
{
"epoch": 0.09,
"learning_rate": 0.0004599598796389168,
"loss": 3.1698,
"theoretical_loss": 4.155198831698795,
"tokens_seen": 294649856
},
{
"epoch": 0.09,
"learning_rate": 0.0004599498495486459,
"loss": 3.1415,
"theoretical_loss": 4.155090091813358,
"tokens_seen": 294715392
},
{
"epoch": 0.09,
"learning_rate": 0.00045993981945837515,
"loss": 2.9568,
"theoretical_loss": 4.154981382874608,
"tokens_seen": 294780928
},
{
"epoch": 0.09,
"learning_rate": 0.00045992978936810434,
"loss": 3.0327,
"theoretical_loss": 4.154872704866859,
"tokens_seen": 294846464
},
{
"epoch": 0.09,
"objective/train/docs_used": 179027,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5512094497680664,
"objective/train/theoretical_loss": 4.15476405777444,
"objective/train/tokens_used": 315372000,
"theoretical_loss": 4.15476405777444,
"tokens_seen": 294912000
},
{
"epoch": 0.09,
"learning_rate": 0.0004599197592778335,
"loss": 2.9603,
"theoretical_loss": 4.15476405777444,
"tokens_seen": 294912000
},
{
"epoch": 0.09,
"learning_rate": 0.0004599097291875627,
"loss": 3.1664,
"theoretical_loss": 4.154655441581687,
"tokens_seen": 294977536
},
{
"epoch": 0.09,
"learning_rate": 0.0004598996990972919,
"loss": 2.91,
"theoretical_loss": 4.154546856272952,
"tokens_seen": 295043072
},
{
"epoch": 0.09,
"learning_rate": 0.00045988966900702106,
"loss": 2.9691,
"theoretical_loss": 4.154438301832596,
"tokens_seen": 295108608
},
{
"epoch": 0.09,
"learning_rate": 0.0004598796389167503,
"loss": 3.0933,
"theoretical_loss": 4.154329778244991,
"tokens_seen": 295174144
},
{
"epoch": 0.09,
"learning_rate": 0.0004598696088264794,
"loss": 2.9358,
"theoretical_loss": 4.154221285494521,
"tokens_seen": 295239680
},
{
"epoch": 0.09,
"learning_rate": 0.00045985957873620866,
"loss": 3.2587,
"theoretical_loss": 4.154112823565582,
"tokens_seen": 295305216
},
{
"epoch": 0.09,
"learning_rate": 0.0004598495486459378,
"loss": 3.3291,
"theoretical_loss": 4.15400439244258,
"tokens_seen": 295370752
},
{
"epoch": 0.09,
"learning_rate": 0.000459839518555667,
"loss": 3.3911,
"theoretical_loss": 4.153895992109935,
"tokens_seen": 295436288
},
{
"epoch": 0.09,
"learning_rate": 0.0004598294884653962,
"loss": 2.9428,
"theoretical_loss": 4.153787622552073,
"tokens_seen": 295501824
},
{
"epoch": 0.09,
"learning_rate": 0.0004598194583751254,
"loss": 3.1373,
"theoretical_loss": 4.153679283753439,
"tokens_seen": 295567360
},
{
"epoch": 0.09,
"learning_rate": 0.00045980942828485456,
"loss": 3.1252,
"theoretical_loss": 4.15357097569848,
"tokens_seen": 295632896
},
{
"epoch": 0.09,
"learning_rate": 0.0004597993981945838,
"loss": 3.1504,
"theoretical_loss": 4.153462698371665,
"tokens_seen": 295698432
},
{
"epoch": 0.09,
"learning_rate": 0.0004597893681043129,
"loss": 3.1088,
"theoretical_loss": 4.1533544517574645,
"tokens_seen": 295763968
},
{
"epoch": 0.09,
"learning_rate": 0.00045977933801404216,
"loss": 2.9716,
"theoretical_loss": 4.153246235840367,
"tokens_seen": 295829504
},
{
"epoch": 0.09,
"learning_rate": 0.0004597693079237713,
"loss": 3.2071,
"theoretical_loss": 4.153138050604868,
"tokens_seen": 295895040
},
{
"epoch": 0.09,
"learning_rate": 0.0004597592778335005,
"loss": 3.2361,
"theoretical_loss": 4.153029896035476,
"tokens_seen": 295960576
},
{
"epoch": 0.09,
"learning_rate": 0.0004597492477432297,
"loss": 2.9966,
"theoretical_loss": 4.152921772116712,
"tokens_seen": 296026112
},
{
"epoch": 0.09,
"learning_rate": 0.0004597392176529589,
"loss": 3.1588,
"theoretical_loss": 4.152813678833106,
"tokens_seen": 296091648
},
{
"epoch": 0.09,
"learning_rate": 0.00045972918756268807,
"loss": 3.2001,
"theoretical_loss": 4.152705616169202,
"tokens_seen": 296157184
},
{
"epoch": 0.09,
"learning_rate": 0.00045971915747241725,
"loss": 2.8042,
"theoretical_loss": 4.15259758410955,
"tokens_seen": 296222720
},
{
"epoch": 0.09,
"learning_rate": 0.00045970912738214643,
"loss": 3.135,
"theoretical_loss": 4.152489582638719,
"tokens_seen": 296288256
},
{
"epoch": 0.09,
"learning_rate": 0.00045969909729187566,
"loss": 3.3503,
"theoretical_loss": 4.152381611741281,
"tokens_seen": 296353792
},
{
"epoch": 0.09,
"learning_rate": 0.0004596890672016048,
"loss": 3.0074,
"theoretical_loss": 4.152273671401824,
"tokens_seen": 296419328
},
{
"epoch": 0.09,
"learning_rate": 0.000459679037111334,
"loss": 3.1205,
"theoretical_loss": 4.152165761604948,
"tokens_seen": 296484864
},
{
"epoch": 0.09,
"objective/train/docs_used": 179648,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.256779432296753,
"objective/train/theoretical_loss": 4.152057882335261,
"objective/train/tokens_used": 317010400,
"theoretical_loss": 4.152057882335261,
"tokens_seen": 296550400
},
{
"epoch": 0.09,
"learning_rate": 0.00045966900702106315,
"loss": 3.1838,
"theoretical_loss": 4.152057882335261,
"tokens_seen": 296550400
},
{
"epoch": 0.09,
"learning_rate": 0.0004596589769307924,
"loss": 3.178,
"theoretical_loss": 4.151950033577383,
"tokens_seen": 296615936
},
{
"epoch": 0.09,
"learning_rate": 0.00045964894684052157,
"loss": 3.1013,
"theoretical_loss": 4.151842215315947,
"tokens_seen": 296681472
},
{
"epoch": 0.09,
"learning_rate": 0.00045963891675025075,
"loss": 3.056,
"theoretical_loss": 4.151734427535594,
"tokens_seen": 296747008
},
{
"epoch": 0.09,
"learning_rate": 0.00045962888665997993,
"loss": 3.1372,
"theoretical_loss": 4.151626670220979,
"tokens_seen": 296812544
},
{
"epoch": 0.09,
"learning_rate": 0.00045961885656970917,
"loss": 3.0696,
"theoretical_loss": 4.151518943356768,
"tokens_seen": 296878080
},
{
"epoch": 0.09,
"learning_rate": 0.0004596088264794383,
"loss": 3.0229,
"theoretical_loss": 4.151411246927636,
"tokens_seen": 296943616
},
{
"epoch": 0.09,
"learning_rate": 0.00045959879638916753,
"loss": 2.9934,
"theoretical_loss": 4.15130358091827,
"tokens_seen": 297009152
},
{
"epoch": 0.09,
"learning_rate": 0.00045958876629889666,
"loss": 3.1986,
"theoretical_loss": 4.151195945313369,
"tokens_seen": 297074688
},
{
"epoch": 0.09,
"learning_rate": 0.0004595787362086259,
"loss": 3.0798,
"theoretical_loss": 4.151088340097642,
"tokens_seen": 297140224
},
{
"epoch": 0.09,
"learning_rate": 0.0004595687061183551,
"loss": 3.0475,
"theoretical_loss": 4.15098076525581,
"tokens_seen": 297205760
},
{
"epoch": 0.09,
"learning_rate": 0.00045955867602808425,
"loss": 3.1249,
"theoretical_loss": 4.150873220772604,
"tokens_seen": 297271296
},
{
"epoch": 0.09,
"learning_rate": 0.00045954864593781344,
"loss": 3.3091,
"theoretical_loss": 4.150765706632766,
"tokens_seen": 297336832
},
{
"epoch": 0.09,
"learning_rate": 0.0004595386158475426,
"loss": 3.2366,
"theoretical_loss": 4.1506582228210505,
"tokens_seen": 297402368
},
{
"epoch": 0.09,
"learning_rate": 0.0004595285857572718,
"loss": 2.9709,
"theoretical_loss": 4.150550769322221,
"tokens_seen": 297467904
},
{
"epoch": 0.09,
"learning_rate": 0.00045951855566700103,
"loss": 3.1059,
"theoretical_loss": 4.150443346121054,
"tokens_seen": 297533440
},
{
"epoch": 0.09,
"learning_rate": 0.00045950852557673016,
"loss": 2.9407,
"theoretical_loss": 4.150335953202336,
"tokens_seen": 297598976
},
{
"epoch": 0.09,
"learning_rate": 0.0004594984954864594,
"loss": 3.0818,
"theoretical_loss": 4.150228590550864,
"tokens_seen": 297664512
},
{
"epoch": 0.09,
"learning_rate": 0.0004594884653961885,
"loss": 3.2076,
"theoretical_loss": 4.150121258151447,
"tokens_seen": 297730048
},
{
"epoch": 0.09,
"learning_rate": 0.00045947843530591776,
"loss": 2.9442,
"theoretical_loss": 4.150013955988905,
"tokens_seen": 297795584
},
{
"epoch": 0.09,
"learning_rate": 0.00045946840521564694,
"loss": 3.1901,
"theoretical_loss": 4.149906684048068,
"tokens_seen": 297861120
},
{
"epoch": 0.09,
"learning_rate": 0.0004594583751253761,
"loss": 2.6943,
"theoretical_loss": 4.1497994423137765,
"tokens_seen": 297926656
},
{
"epoch": 0.09,
"learning_rate": 0.00045944834503510536,
"loss": 2.9976,
"theoretical_loss": 4.149692230770884,
"tokens_seen": 297992192
},
{
"epoch": 0.09,
"learning_rate": 0.00045943831494483454,
"loss": 3.1648,
"theoretical_loss": 4.149585049404253,
"tokens_seen": 298057728
},
{
"epoch": 0.09,
"learning_rate": 0.0004594282848545637,
"loss": 3.1697,
"theoretical_loss": 4.149477898198759,
"tokens_seen": 298123264
},
{
"epoch": 0.09,
"objective/train/docs_used": 180307,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1973071098327637,
"objective/train/theoretical_loss": 4.149370777139286,
"objective/train/tokens_used": 318648800,
"theoretical_loss": 4.149370777139286,
"tokens_seen": 298188800
},
{
"epoch": 0.09,
"learning_rate": 0.0004594182547642929,
"loss": 3.0318,
"theoretical_loss": 4.149370777139286,
"tokens_seen": 298188800
},
{
"epoch": 0.09,
"learning_rate": 0.0004594082246740221,
"loss": 3.1242,
"theoretical_loss": 4.14926368621073,
"tokens_seen": 298254336
},
{
"epoch": 0.09,
"learning_rate": 0.00045939819458375126,
"loss": 3.0613,
"theoretical_loss": 4.149156625397998,
"tokens_seen": 298319872
},
{
"epoch": 0.09,
"learning_rate": 0.0004593881644934805,
"loss": 2.8588,
"theoretical_loss": 4.149049594686008,
"tokens_seen": 298385408
},
{
"epoch": 0.09,
"learning_rate": 0.0004593781344032096,
"loss": 2.7734,
"theoretical_loss": 4.1489425940596885,
"tokens_seen": 298450944
},
{
"epoch": 0.09,
"learning_rate": 0.00045936810431293886,
"loss": 2.7681,
"theoretical_loss": 4.148835623503978,
"tokens_seen": 298516480
},
{
"epoch": 0.09,
"learning_rate": 0.000459358074222668,
"loss": 3.1072,
"theoretical_loss": 4.148728683003829,
"tokens_seen": 298582016
},
{
"epoch": 0.09,
"learning_rate": 0.0004593480441323972,
"loss": 3.1108,
"theoretical_loss": 4.1486217725442005,
"tokens_seen": 298647552
},
{
"epoch": 0.09,
"learning_rate": 0.0004593380140421264,
"loss": 2.91,
"theoretical_loss": 4.148514892110065,
"tokens_seen": 298713088
},
{
"epoch": 0.09,
"learning_rate": 0.0004593279839518556,
"loss": 3.2009,
"theoretical_loss": 4.148408041686406,
"tokens_seen": 298778624
},
{
"epoch": 0.09,
"learning_rate": 0.00045931795386158476,
"loss": 3.0153,
"theoretical_loss": 4.148301221258217,
"tokens_seen": 298844160
},
{
"epoch": 0.09,
"learning_rate": 0.000459307923771314,
"loss": 3.2481,
"theoretical_loss": 4.148194430810502,
"tokens_seen": 298909696
},
{
"epoch": 0.09,
"learning_rate": 0.0004592978936810431,
"loss": 3.3113,
"theoretical_loss": 4.148087670328276,
"tokens_seen": 298975232
},
{
"epoch": 0.09,
"learning_rate": 0.00045928786359077236,
"loss": 3.2193,
"theoretical_loss": 4.147980939796565,
"tokens_seen": 299040768
},
{
"epoch": 0.09,
"learning_rate": 0.0004592778335005015,
"loss": 3.0393,
"theoretical_loss": 4.147874239200405,
"tokens_seen": 299106304
},
{
"epoch": 0.09,
"learning_rate": 0.0004592678034102307,
"loss": 2.8427,
"theoretical_loss": 4.147767568524845,
"tokens_seen": 299171840
},
{
"epoch": 0.09,
"learning_rate": 0.0004592577733199599,
"loss": 3.1985,
"theoretical_loss": 4.147660927754942,
"tokens_seen": 299237376
},
{
"epoch": 0.09,
"learning_rate": 0.0004592477432296891,
"loss": 2.9435,
"theoretical_loss": 4.147554316875766,
"tokens_seen": 299302912
},
{
"epoch": 0.09,
"learning_rate": 0.00045923771313941827,
"loss": 3.0539,
"theoretical_loss": 4.147447735872396,
"tokens_seen": 299368448
},
{
"epoch": 0.09,
"learning_rate": 0.00045922768304914745,
"loss": 2.9897,
"theoretical_loss": 4.147341184729921,
"tokens_seen": 299433984
},
{
"epoch": 0.09,
"learning_rate": 0.00045921765295887663,
"loss": 2.9482,
"theoretical_loss": 4.147234663433444,
"tokens_seen": 299499520
},
{
"epoch": 0.09,
"learning_rate": 0.00045920762286860586,
"loss": 2.8237,
"theoretical_loss": 4.147128171968077,
"tokens_seen": 299565056
},
{
"epoch": 0.09,
"learning_rate": 0.000459197592778335,
"loss": 3.1106,
"theoretical_loss": 4.14702171031894,
"tokens_seen": 299630592
},
{
"epoch": 0.09,
"learning_rate": 0.0004591875626880642,
"loss": 2.9476,
"theoretical_loss": 4.146915278471169,
"tokens_seen": 299696128
},
{
"epoch": 0.09,
"learning_rate": 0.00045917753259779335,
"loss": 2.968,
"theoretical_loss": 4.146808876409906,
"tokens_seen": 299761664
},
{
"epoch": 0.09,
"objective/train/docs_used": 181680,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3923745155334473,
"objective/train/theoretical_loss": 4.146702504120305,
"objective/train/tokens_used": 320287200,
"theoretical_loss": 4.146702504120305,
"tokens_seen": 299827200
},
{
"epoch": 0.09,
"learning_rate": 0.0004591675025075226,
"loss": 2.9731,
"theoretical_loss": 4.146702504120305,
"tokens_seen": 299827200
},
{
"epoch": 0.09,
"learning_rate": 0.00045915747241725177,
"loss": 2.7663,
"theoretical_loss": 4.146596161587532,
"tokens_seen": 299892736
},
{
"epoch": 0.09,
"learning_rate": 0.00045914744232698095,
"loss": 2.5711,
"theoretical_loss": 4.146489848796763,
"tokens_seen": 299958272
},
{
"epoch": 0.09,
"learning_rate": 0.00045913741223671013,
"loss": 3.2189,
"theoretical_loss": 4.146383565733184,
"tokens_seen": 300023808
},
{
"epoch": 0.09,
"learning_rate": 0.00045912738214643937,
"loss": 3.0249,
"theoretical_loss": 4.146277312381991,
"tokens_seen": 300089344
},
{
"epoch": 0.09,
"learning_rate": 0.0004591173520561685,
"loss": 2.903,
"theoretical_loss": 4.1461710887283925,
"tokens_seen": 300154880
},
{
"epoch": 0.09,
"learning_rate": 0.00045910732196589773,
"loss": 3.1524,
"theoretical_loss": 4.146064894757606,
"tokens_seen": 300220416
},
{
"epoch": 0.09,
"learning_rate": 0.00045909729187562686,
"loss": 2.9013,
"theoretical_loss": 4.145958730454861,
"tokens_seen": 300285952
},
{
"epoch": 0.09,
"learning_rate": 0.0004590872617853561,
"loss": 2.9616,
"theoretical_loss": 4.145852595805396,
"tokens_seen": 300351488
},
{
"epoch": 0.09,
"learning_rate": 0.0004590772316950853,
"loss": 3.0813,
"theoretical_loss": 4.145746490794461,
"tokens_seen": 300417024
},
{
"epoch": 0.09,
"learning_rate": 0.00045906720160481445,
"loss": 3.1084,
"theoretical_loss": 4.145640415407317,
"tokens_seen": 300482560
},
{
"epoch": 0.09,
"learning_rate": 0.00045905717151454364,
"loss": 2.9712,
"theoretical_loss": 4.145534369629234,
"tokens_seen": 300548096
},
{
"epoch": 0.09,
"learning_rate": 0.0004590471414242728,
"loss": 2.9661,
"theoretical_loss": 4.145428353445494,
"tokens_seen": 300613632
},
{
"epoch": 0.09,
"learning_rate": 0.000459037111334002,
"loss": 3.211,
"theoretical_loss": 4.145322366841389,
"tokens_seen": 300679168
},
{
"epoch": 0.09,
"learning_rate": 0.00045902708124373123,
"loss": 2.9262,
"theoretical_loss": 4.145216409802221,
"tokens_seen": 300744704
},
{
"epoch": 0.09,
"learning_rate": 0.00045901705115346036,
"loss": 2.9333,
"theoretical_loss": 4.145110482313304,
"tokens_seen": 300810240
},
{
"epoch": 0.09,
"learning_rate": 0.0004590070210631896,
"loss": 3.0263,
"theoretical_loss": 4.1450045843599606,
"tokens_seen": 300875776
},
{
"epoch": 0.09,
"learning_rate": 0.0004589969909729187,
"loss": 2.906,
"theoretical_loss": 4.144898715927525,
"tokens_seen": 300941312
},
{
"epoch": 0.09,
"learning_rate": 0.00045898696088264796,
"loss": 2.8445,
"theoretical_loss": 4.144792877001342,
"tokens_seen": 301006848
},
{
"epoch": 0.09,
"learning_rate": 0.00045897693079237714,
"loss": 2.984,
"theoretical_loss": 4.144687067566765,
"tokens_seen": 301072384
},
{
"epoch": 0.09,
"learning_rate": 0.0004589669007021063,
"loss": 2.9933,
"theoretical_loss": 4.144581287609161,
"tokens_seen": 301137920
},
{
"epoch": 0.09,
"learning_rate": 0.0004589568706118355,
"loss": 2.9471,
"theoretical_loss": 4.144475537113905,
"tokens_seen": 301203456
},
{
"epoch": 0.09,
"learning_rate": 0.00045894684052156474,
"loss": 3.0003,
"theoretical_loss": 4.144369816066385,
"tokens_seen": 301268992
},
{
"epoch": 0.09,
"learning_rate": 0.00045893681043129386,
"loss": 3.1696,
"theoretical_loss": 4.144264124451995,
"tokens_seen": 301334528
},
{
"epoch": 0.09,
"learning_rate": 0.0004589267803410231,
"loss": 3.1541,
"theoretical_loss": 4.1441584622561445,
"tokens_seen": 301400064
},
{
"epoch": 0.09,
"objective/train/docs_used": 182385,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2453272342681885,
"objective/train/theoretical_loss": 4.144052829464249,
"objective/train/tokens_used": 321925600,
"theoretical_loss": 4.144052829464249,
"tokens_seen": 301465600
},
{
"epoch": 0.09,
"learning_rate": 0.0004589167502507522,
"loss": 3.212,
"theoretical_loss": 4.144052829464249,
"tokens_seen": 301465600
},
{
"epoch": 0.09,
"learning_rate": 0.00045890672016048146,
"loss": 3.221,
"theoretical_loss": 4.143947226061737,
"tokens_seen": 301531136
},
{
"epoch": 0.09,
"learning_rate": 0.00045889669007021064,
"loss": 3.1414,
"theoretical_loss": 4.143841652034048,
"tokens_seen": 301596672
},
{
"epoch": 0.09,
"learning_rate": 0.0004588866599799398,
"loss": 2.8214,
"theoretical_loss": 4.143736107366629,
"tokens_seen": 301662208
},
{
"epoch": 0.09,
"learning_rate": 0.000458876629889669,
"loss": 2.9196,
"theoretical_loss": 4.14363059204494,
"tokens_seen": 301727744
},
{
"epoch": 0.09,
"learning_rate": 0.0004588665997993982,
"loss": 2.8951,
"theoretical_loss": 4.14352510605445,
"tokens_seen": 301793280
},
{
"epoch": 0.09,
"learning_rate": 0.00045885656970912737,
"loss": 3.2871,
"theoretical_loss": 4.143419649380639,
"tokens_seen": 301858816
},
{
"epoch": 0.09,
"learning_rate": 0.0004588465396188566,
"loss": 3.0498,
"theoretical_loss": 4.143314222008997,
"tokens_seen": 301924352
},
{
"epoch": 0.09,
"learning_rate": 0.00045883650952858573,
"loss": 3.2756,
"theoretical_loss": 4.143208823925024,
"tokens_seen": 301989888
},
{
"epoch": 0.09,
"learning_rate": 0.00045882647943831496,
"loss": 2.7953,
"theoretical_loss": 4.143103455114231,
"tokens_seen": 302055424
},
{
"epoch": 0.09,
"learning_rate": 0.00045881644934804415,
"loss": 2.9904,
"theoretical_loss": 4.142998115562139,
"tokens_seen": 302120960
},
{
"epoch": 0.09,
"learning_rate": 0.0004588064192577733,
"loss": 2.8618,
"theoretical_loss": 4.14289280525428,
"tokens_seen": 302186496
},
{
"epoch": 0.09,
"learning_rate": 0.0004587963891675025,
"loss": 2.8056,
"theoretical_loss": 4.142787524176194,
"tokens_seen": 302252032
},
{
"epoch": 0.09,
"learning_rate": 0.0004587863590772317,
"loss": 2.9704,
"theoretical_loss": 4.142682272313435,
"tokens_seen": 302317568
},
{
"epoch": 0.09,
"learning_rate": 0.00045877632898696087,
"loss": 3.2137,
"theoretical_loss": 4.142577049651563,
"tokens_seen": 302383104
},
{
"epoch": 0.09,
"learning_rate": 0.0004587662988966901,
"loss": 3.2904,
"theoretical_loss": 4.142471856176152,
"tokens_seen": 302448640
},
{
"epoch": 0.09,
"learning_rate": 0.00045875626880641923,
"loss": 2.9358,
"theoretical_loss": 4.142366691872784,
"tokens_seen": 302514176
},
{
"epoch": 0.09,
"learning_rate": 0.00045874623871614847,
"loss": 3.1388,
"theoretical_loss": 4.142261556727052,
"tokens_seen": 302579712
},
{
"epoch": 0.09,
"learning_rate": 0.0004587362086258776,
"loss": 2.8735,
"theoretical_loss": 4.14215645072456,
"tokens_seen": 302645248
},
{
"epoch": 0.09,
"learning_rate": 0.00045872617853560683,
"loss": 2.9464,
"theoretical_loss": 4.14205137385092,
"tokens_seen": 302710784
},
{
"epoch": 0.09,
"learning_rate": 0.000458716148445336,
"loss": 3.0534,
"theoretical_loss": 4.141946326091756,
"tokens_seen": 302776320
},
{
"epoch": 0.09,
"learning_rate": 0.0004587061183550652,
"loss": 3.0235,
"theoretical_loss": 4.141841307432703,
"tokens_seen": 302841856
},
{
"epoch": 0.09,
"learning_rate": 0.0004586960882647944,
"loss": 2.9913,
"theoretical_loss": 4.1417363178594035,
"tokens_seen": 302907392
},
{
"epoch": 0.09,
"learning_rate": 0.00045868605817452355,
"loss": 3.1955,
"theoretical_loss": 4.141631357357513,
"tokens_seen": 302972928
},
{
"epoch": 0.09,
"learning_rate": 0.0004586760280842528,
"loss": 3.1505,
"theoretical_loss": 4.141526425912694,
"tokens_seen": 303038464
},
{
"epoch": 0.09,
"objective/train/docs_used": 183646,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1540284156799316,
"objective/train/theoretical_loss": 4.141421523510623,
"objective/train/tokens_used": 323564000,
"theoretical_loss": 4.141421523510623,
"tokens_seen": 303104000
},
{
"epoch": 0.09,
"learning_rate": 0.00045866599799398197,
"loss": 2.9251,
"theoretical_loss": 4.141421523510623,
"tokens_seen": 303104000
},
{
"epoch": 0.09,
"learning_rate": 0.00045865596790371115,
"loss": 3.11,
"theoretical_loss": 4.141316650136983,
"tokens_seen": 303169536
},
{
"epoch": 0.09,
"learning_rate": 0.00045864593781344033,
"loss": 2.9084,
"theoretical_loss": 4.14121180577747,
"tokens_seen": 303235072
},
{
"epoch": 0.09,
"learning_rate": 0.00045863590772316957,
"loss": 3.2785,
"theoretical_loss": 4.141106990417789,
"tokens_seen": 303300608
},
{
"epoch": 0.09,
"learning_rate": 0.0004586258776328987,
"loss": 2.9644,
"theoretical_loss": 4.141002204043654,
"tokens_seen": 303366144
},
{
"epoch": 0.09,
"learning_rate": 0.00045861584754262793,
"loss": 3.293,
"theoretical_loss": 4.140897446640793,
"tokens_seen": 303431680
},
{
"epoch": 0.09,
"learning_rate": 0.00045860581745235706,
"loss": 2.9656,
"theoretical_loss": 4.1407927181949375,
"tokens_seen": 303497216
},
{
"epoch": 0.09,
"learning_rate": 0.0004585957873620863,
"loss": 2.9074,
"theoretical_loss": 4.140688018691835,
"tokens_seen": 303562752
},
{
"epoch": 0.09,
"learning_rate": 0.0004585857572718155,
"loss": 2.884,
"theoretical_loss": 4.140583348117241,
"tokens_seen": 303628288
},
{
"epoch": 0.09,
"learning_rate": 0.00045857572718154465,
"loss": 3.0019,
"theoretical_loss": 4.140478706456921,
"tokens_seen": 303693824
},
{
"epoch": 0.09,
"learning_rate": 0.00045856569709127384,
"loss": 3.0583,
"theoretical_loss": 4.140374093696651,
"tokens_seen": 303759360
},
{
"epoch": 0.09,
"learning_rate": 0.000458555667001003,
"loss": 3.1779,
"theoretical_loss": 4.1402695098222155,
"tokens_seen": 303824896
},
{
"epoch": 0.09,
"learning_rate": 0.0004585456369107322,
"loss": 2.8,
"theoretical_loss": 4.140164954819412,
"tokens_seen": 303890432
},
{
"epoch": 0.09,
"learning_rate": 0.00045853560682046143,
"loss": 3.1757,
"theoretical_loss": 4.140060428674046,
"tokens_seen": 303955968
},
{
"epoch": 0.09,
"learning_rate": 0.00045852557673019056,
"loss": 3.1018,
"theoretical_loss": 4.139955931371932,
"tokens_seen": 304021504
},
{
"epoch": 0.09,
"learning_rate": 0.0004585155466399198,
"loss": 2.8922,
"theoretical_loss": 4.139851462898897,
"tokens_seen": 304087040
},
{
"epoch": 0.09,
"learning_rate": 0.0004585055165496489,
"loss": 3.0495,
"theoretical_loss": 4.139747023240777,
"tokens_seen": 304152576
},
{
"epoch": 0.09,
"learning_rate": 0.00045849548645937816,
"loss": 3.159,
"theoretical_loss": 4.139642612383418,
"tokens_seen": 304218112
},
{
"epoch": 0.09,
"learning_rate": 0.00045848545636910734,
"loss": 3.3414,
"theoretical_loss": 4.1395382303126755,
"tokens_seen": 304283648
},
{
"epoch": 0.09,
"learning_rate": 0.0004584754262788365,
"loss": 3.0128,
"theoretical_loss": 4.139433877014415,
"tokens_seen": 304349184
},
{
"epoch": 0.09,
"learning_rate": 0.0004584653961885657,
"loss": 3.2096,
"theoretical_loss": 4.139329552474514,
"tokens_seen": 304414720
},
{
"epoch": 0.09,
"learning_rate": 0.00045845536609829494,
"loss": 2.9645,
"theoretical_loss": 4.139225256678857,
"tokens_seen": 304480256
},
{
"epoch": 0.09,
"learning_rate": 0.00045844533600802406,
"loss": 3.2456,
"theoretical_loss": 4.139120989613341,
"tokens_seen": 304545792
},
{
"epoch": 0.09,
"learning_rate": 0.0004584353059177533,
"loss": 3.2019,
"theoretical_loss": 4.1390167512638705,
"tokens_seen": 304611328
},
{
"epoch": 0.09,
"learning_rate": 0.0004584252758274824,
"loss": 3.2112,
"theoretical_loss": 4.138912541616363,
"tokens_seen": 304676864
},
{
"epoch": 0.09,
"objective/train/docs_used": 184075,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.958650827407837,
"objective/train/theoretical_loss": 4.138808360656742,
"objective/train/tokens_used": 325202400,
"theoretical_loss": 4.138808360656742,
"tokens_seen": 304742400
},
{
"epoch": 0.09,
"learning_rate": 0.00045841524573721166,
"loss": 2.9379,
"theoretical_loss": 4.138808360656742,
"tokens_seen": 304742400
},
{
"epoch": 0.09,
"learning_rate": 0.00045840521564694084,
"loss": 3.0492,
"theoretical_loss": 4.138704208370944,
"tokens_seen": 304807936
},
{
"epoch": 0.09,
"learning_rate": 0.00045839518555667,
"loss": 2.9333,
"theoretical_loss": 4.138600084744915,
"tokens_seen": 304873472
},
{
"epoch": 0.09,
"learning_rate": 0.0004583851554663992,
"loss": 3.0129,
"theoretical_loss": 4.1384959897646105,
"tokens_seen": 304939008
},
{
"epoch": 0.09,
"learning_rate": 0.0004583751253761284,
"loss": 3.1081,
"theoretical_loss": 4.138391923415996,
"tokens_seen": 305004544
},
{
"epoch": 0.09,
"learning_rate": 0.00045836509528585757,
"loss": 3.0953,
"theoretical_loss": 4.138287885685045,
"tokens_seen": 305070080
},
{
"epoch": 0.09,
"learning_rate": 0.0004583550651955868,
"loss": 3.0357,
"theoretical_loss": 4.138183876557745,
"tokens_seen": 305135616
},
{
"epoch": 0.09,
"learning_rate": 0.00045834503510531593,
"loss": 3.2149,
"theoretical_loss": 4.1380798960200895,
"tokens_seen": 305201152
},
{
"epoch": 0.09,
"learning_rate": 0.00045833500501504516,
"loss": 3.0338,
"theoretical_loss": 4.137975944058083,
"tokens_seen": 305266688
},
{
"epoch": 0.09,
"learning_rate": 0.00045832497492477435,
"loss": 3.0291,
"theoretical_loss": 4.137872020657742,
"tokens_seen": 305332224
},
{
"epoch": 0.09,
"learning_rate": 0.0004583149448345035,
"loss": 3.1508,
"theoretical_loss": 4.1377681258050885,
"tokens_seen": 305397760
},
{
"epoch": 0.09,
"learning_rate": 0.0004583049147442327,
"loss": 2.9683,
"theoretical_loss": 4.13766425948616,
"tokens_seen": 305463296
},
{
"epoch": 0.09,
"learning_rate": 0.0004582948846539619,
"loss": 3.0612,
"theoretical_loss": 4.137560421686998,
"tokens_seen": 305528832
},
{
"epoch": 0.09,
"learning_rate": 0.00045828485456369107,
"loss": 2.8031,
"theoretical_loss": 4.137456612393658,
"tokens_seen": 305594368
},
{
"epoch": 0.09,
"learning_rate": 0.0004582748244734203,
"loss": 3.2981,
"theoretical_loss": 4.137352831592203,
"tokens_seen": 305659904
},
{
"epoch": 0.09,
"learning_rate": 0.00045826479438314943,
"loss": 3.1212,
"theoretical_loss": 4.137249079268707,
"tokens_seen": 305725440
},
{
"epoch": 0.09,
"learning_rate": 0.00045825476429287867,
"loss": 3.1303,
"theoretical_loss": 4.137145355409253,
"tokens_seen": 305790976
},
{
"epoch": 0.09,
"learning_rate": 0.0004582447342026078,
"loss": 3.0405,
"theoretical_loss": 4.137041659999936,
"tokens_seen": 305856512
},
{
"epoch": 0.09,
"learning_rate": 0.00045823470411233703,
"loss": 3.0228,
"theoretical_loss": 4.136937993026857,
"tokens_seen": 305922048
},
{
"epoch": 0.09,
"learning_rate": 0.0004582246740220662,
"loss": 3.0352,
"theoretical_loss": 4.136834354476129,
"tokens_seen": 305987584
},
{
"epoch": 0.09,
"learning_rate": 0.0004582146439317954,
"loss": 3.0844,
"theoretical_loss": 4.1367307443338746,
"tokens_seen": 306053120
},
{
"epoch": 0.09,
"learning_rate": 0.00045820461384152457,
"loss": 3.0252,
"theoretical_loss": 4.136627162586226,
"tokens_seen": 306118656
},
{
"epoch": 0.09,
"learning_rate": 0.00045819458375125375,
"loss": 2.9533,
"theoretical_loss": 4.136523609219327,
"tokens_seen": 306184192
},
{
"epoch": 0.09,
"learning_rate": 0.00045818455366098294,
"loss": 3.0824,
"theoretical_loss": 4.136420084219327,
"tokens_seen": 306249728
},
{
"epoch": 0.09,
"learning_rate": 0.00045817452357071217,
"loss": 2.8966,
"theoretical_loss": 4.136316587572388,
"tokens_seen": 306315264
},
{
"epoch": 0.09,
"objective/train/docs_used": 185377,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.6815130710601807,
"objective/train/theoretical_loss": 4.136213119264681,
"objective/train/tokens_used": 326840800,
"theoretical_loss": 4.136213119264681,
"tokens_seen": 306380800
},
{
"epoch": 0.09,
"learning_rate": 0.0004581644934804413,
"loss": 3.2741,
"theoretical_loss": 4.136213119264681,
"tokens_seen": 306380800
},
{
"epoch": 0.09,
"learning_rate": 0.00045815446339017053,
"loss": 3.0248,
"theoretical_loss": 4.136109679282388,
"tokens_seen": 306446336
},
{
"epoch": 0.09,
"learning_rate": 0.0004581444332998997,
"loss": 3.086,
"theoretical_loss": 4.136006267611697,
"tokens_seen": 306511872
},
{
"epoch": 0.09,
"learning_rate": 0.0004581344032096289,
"loss": 3.1913,
"theoretical_loss": 4.135902884238812,
"tokens_seen": 306577408
},
{
"epoch": 0.09,
"learning_rate": 0.0004581243731193581,
"loss": 3.2079,
"theoretical_loss": 4.135799529149939,
"tokens_seen": 306642944
},
{
"epoch": 0.09,
"learning_rate": 0.00045811434302908726,
"loss": 3.0883,
"theoretical_loss": 4.1356962023313,
"tokens_seen": 306708480
},
{
"epoch": 0.09,
"learning_rate": 0.00045810431293881644,
"loss": 3.0245,
"theoretical_loss": 4.135592903769124,
"tokens_seen": 306774016
},
{
"epoch": 0.09,
"learning_rate": 0.0004580942828485457,
"loss": 2.9307,
"theoretical_loss": 4.135489633449649,
"tokens_seen": 306839552
},
{
"epoch": 0.09,
"learning_rate": 0.0004580842527582748,
"loss": 2.8345,
"theoretical_loss": 4.135386391359123,
"tokens_seen": 306905088
},
{
"epoch": 0.09,
"learning_rate": 0.00045807422266800404,
"loss": 3.2493,
"theoretical_loss": 4.135283177483807,
"tokens_seen": 306970624
},
{
"epoch": 0.09,
"learning_rate": 0.00045806419257773316,
"loss": 3.0784,
"theoretical_loss": 4.135179991809965,
"tokens_seen": 307036160
},
{
"epoch": 0.09,
"learning_rate": 0.0004580541624874624,
"loss": 2.9564,
"theoretical_loss": 4.135076834323876,
"tokens_seen": 307101696
},
{
"epoch": 0.09,
"learning_rate": 0.0004580441323971916,
"loss": 3.2303,
"theoretical_loss": 4.134973705011828,
"tokens_seen": 307167232
},
{
"epoch": 0.09,
"learning_rate": 0.00045803410230692076,
"loss": 2.8665,
"theoretical_loss": 4.134870603860117,
"tokens_seen": 307232768
},
{
"epoch": 0.09,
"learning_rate": 0.00045802407221664994,
"loss": 3.1065,
"theoretical_loss": 4.134767530855047,
"tokens_seen": 307298304
},
{
"epoch": 0.09,
"learning_rate": 0.0004580140421263791,
"loss": 3.1849,
"theoretical_loss": 4.1346644859829365,
"tokens_seen": 307363840
},
{
"epoch": 0.09,
"learning_rate": 0.0004580040120361083,
"loss": 3.1406,
"theoretical_loss": 4.1345614692301105,
"tokens_seen": 307429376
},
{
"epoch": 0.09,
"learning_rate": 0.00045799398194583754,
"loss": 3.0077,
"theoretical_loss": 4.134458480582902,
"tokens_seen": 307494912
},
{
"epoch": 0.09,
"learning_rate": 0.00045798395185556667,
"loss": 2.9305,
"theoretical_loss": 4.134355520027657,
"tokens_seen": 307560448
},
{
"epoch": 0.09,
"learning_rate": 0.0004579739217652959,
"loss": 3.3013,
"theoretical_loss": 4.134252587550728,
"tokens_seen": 307625984
},
{
"epoch": 0.09,
"learning_rate": 0.0004579638916750251,
"loss": 2.8856,
"theoretical_loss": 4.134149683138481,
"tokens_seen": 307691520
},
{
"epoch": 0.09,
"learning_rate": 0.00045795386158475426,
"loss": 2.9128,
"theoretical_loss": 4.134046806777286,
"tokens_seen": 307757056
},
{
"epoch": 0.09,
"learning_rate": 0.0004579438314944835,
"loss": 3.0849,
"theoretical_loss": 4.133943958453528,
"tokens_seen": 307822592
},
{
"epoch": 0.09,
"learning_rate": 0.0004579338014042126,
"loss": 2.9414,
"theoretical_loss": 4.133841138153597,
"tokens_seen": 307888128
},
{
"epoch": 0.09,
"learning_rate": 0.00045792377131394186,
"loss": 3.2586,
"theoretical_loss": 4.133738345863896,
"tokens_seen": 307953664
},
{
"epoch": 0.09,
"objective/train/docs_used": 186163,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.693579912185669,
"objective/train/theoretical_loss": 4.133635581570836,
"objective/train/tokens_used": 328479200,
"theoretical_loss": 4.133635581570836,
"tokens_seen": 308019200
},
{
"epoch": 0.09,
"learning_rate": 0.00045791374122367104,
"loss": 3.0399,
"theoretical_loss": 4.133635581570836,
"tokens_seen": 308019200
},
{
"epoch": 0.09,
"learning_rate": 0.0004579037111334002,
"loss": 3.0709,
"theoretical_loss": 4.133532845260836,
"tokens_seen": 308084736
},
{
"epoch": 0.09,
"learning_rate": 0.0004578936810431294,
"loss": 3.0057,
"theoretical_loss": 4.133430136920327,
"tokens_seen": 308150272
},
{
"epoch": 0.09,
"learning_rate": 0.0004578836509528586,
"loss": 3.1167,
"theoretical_loss": 4.133327456535749,
"tokens_seen": 308215808
},
{
"epoch": 0.09,
"learning_rate": 0.00045787362086258777,
"loss": 2.8566,
"theoretical_loss": 4.13322480409355,
"tokens_seen": 308281344
},
{
"epoch": 0.09,
"learning_rate": 0.000457863590772317,
"loss": 2.9801,
"theoretical_loss": 4.133122179580189,
"tokens_seen": 308346880
},
{
"epoch": 0.09,
"learning_rate": 0.00045785356068204613,
"loss": 2.9788,
"theoretical_loss": 4.133019582982134,
"tokens_seen": 308412416
},
{
"epoch": 0.09,
"learning_rate": 0.00045784353059177536,
"loss": 3.0067,
"theoretical_loss": 4.1329170142858604,
"tokens_seen": 308477952
},
{
"epoch": 0.09,
"learning_rate": 0.00045783350050150455,
"loss": 3.0007,
"theoretical_loss": 4.132814473477857,
"tokens_seen": 308543488
},
{
"epoch": 0.09,
"learning_rate": 0.0004578234704112337,
"loss": 3.2335,
"theoretical_loss": 4.1327119605446185,
"tokens_seen": 308609024
},
{
"epoch": 0.09,
"learning_rate": 0.0004578134403209629,
"loss": 2.903,
"theoretical_loss": 4.132609475472651,
"tokens_seen": 308674560
},
{
"epoch": 0.09,
"learning_rate": 0.0004578034102306921,
"loss": 3.0954,
"theoretical_loss": 4.132507018248469,
"tokens_seen": 308740096
},
{
"epoch": 0.09,
"learning_rate": 0.00045779338014042127,
"loss": 3.0897,
"theoretical_loss": 4.132404588858597,
"tokens_seen": 308805632
},
{
"epoch": 0.09,
"learning_rate": 0.0004577833500501505,
"loss": 3.0264,
"theoretical_loss": 4.132302187289568,
"tokens_seen": 308871168
},
{
"epoch": 0.09,
"learning_rate": 0.00045777331995987963,
"loss": 3.3141,
"theoretical_loss": 4.132199813527926,
"tokens_seen": 308936704
},
{
"epoch": 0.09,
"learning_rate": 0.00045776328986960887,
"loss": 3.1779,
"theoretical_loss": 4.132097467560223,
"tokens_seen": 309002240
},
{
"epoch": 0.09,
"learning_rate": 0.000457753259779338,
"loss": 3.2129,
"theoretical_loss": 4.1319951493730205,
"tokens_seen": 309067776
},
{
"epoch": 0.09,
"learning_rate": 0.00045774322968906723,
"loss": 3.3985,
"theoretical_loss": 4.131892858952889,
"tokens_seen": 309133312
},
{
"epoch": 0.09,
"learning_rate": 0.0004577331995987964,
"loss": 3.0588,
"theoretical_loss": 4.131790596286409,
"tokens_seen": 309198848
},
{
"epoch": 0.09,
"learning_rate": 0.0004577231695085256,
"loss": 2.8701,
"theoretical_loss": 4.1316883613601725,
"tokens_seen": 309264384
},
{
"epoch": 0.09,
"learning_rate": 0.0004577131394182548,
"loss": 3.0681,
"theoretical_loss": 4.131586154160775,
"tokens_seen": 309329920
},
{
"epoch": 0.09,
"learning_rate": 0.00045770310932798395,
"loss": 3.0305,
"theoretical_loss": 4.131483974674827,
"tokens_seen": 309395456
},
{
"epoch": 0.09,
"learning_rate": 0.00045769307923771314,
"loss": 2.9912,
"theoretical_loss": 4.131381822888946,
"tokens_seen": 309460992
},
{
"epoch": 0.09,
"learning_rate": 0.00045768304914744237,
"loss": 2.9485,
"theoretical_loss": 4.131279698789759,
"tokens_seen": 309526528
},
{
"epoch": 0.09,
"learning_rate": 0.0004576730190571715,
"loss": 3.1318,
"theoretical_loss": 4.1311776023639,
"tokens_seen": 309592064
},
{
"epoch": 0.09,
"objective/train/docs_used": 187126,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0981154441833496,
"objective/train/theoretical_loss": 4.131075533598018,
"objective/train/tokens_used": 330117600,
"theoretical_loss": 4.131075533598018,
"tokens_seen": 309657600
},
{
"epoch": 0.09,
"learning_rate": 0.00045766298896690073,
"loss": 3.0551,
"theoretical_loss": 4.131075533598018,
"tokens_seen": 309657600
},
{
"epoch": 0.09,
"learning_rate": 0.0004576529588766299,
"loss": 3.3009,
"theoretical_loss": 4.130973492478766,
"tokens_seen": 309723136
},
{
"epoch": 0.09,
"learning_rate": 0.0004576429287863591,
"loss": 3.293,
"theoretical_loss": 4.130871478992807,
"tokens_seen": 309788672
},
{
"epoch": 0.09,
"learning_rate": 0.0004576328986960883,
"loss": 3.2951,
"theoretical_loss": 4.130769493126817,
"tokens_seen": 309854208
},
{
"epoch": 0.09,
"learning_rate": 0.00045762286860581746,
"loss": 3.1751,
"theoretical_loss": 4.130667534867476,
"tokens_seen": 309919744
},
{
"epoch": 0.09,
"learning_rate": 0.00045761283851554664,
"loss": 3.2966,
"theoretical_loss": 4.130565604201477,
"tokens_seen": 309985280
},
{
"epoch": 0.09,
"learning_rate": 0.0004576028084252759,
"loss": 2.9467,
"theoretical_loss": 4.130463701115521,
"tokens_seen": 310050816
},
{
"epoch": 0.09,
"learning_rate": 0.000457592778335005,
"loss": 2.7801,
"theoretical_loss": 4.130361825596317,
"tokens_seen": 310116352
},
{
"epoch": 0.09,
"learning_rate": 0.00045758274824473424,
"loss": 3.1128,
"theoretical_loss": 4.130259977630586,
"tokens_seen": 310181888
},
{
"epoch": 0.09,
"learning_rate": 0.00045757271815446336,
"loss": 3.0139,
"theoretical_loss": 4.130158157205056,
"tokens_seen": 310247424
},
{
"epoch": 0.09,
"learning_rate": 0.0004575626880641926,
"loss": 3.1759,
"theoretical_loss": 4.130056364306465,
"tokens_seen": 310312960
},
{
"epoch": 0.09,
"learning_rate": 0.0004575526579739218,
"loss": 3.0092,
"theoretical_loss": 4.129954598921559,
"tokens_seen": 310378496
},
{
"epoch": 0.09,
"learning_rate": 0.00045754262788365096,
"loss": 2.7612,
"theoretical_loss": 4.1298528610370955,
"tokens_seen": 310444032
},
{
"epoch": 0.09,
"learning_rate": 0.00045753259779338014,
"loss": 3.0608,
"theoretical_loss": 4.12975115063984,
"tokens_seen": 310509568
},
{
"epoch": 0.09,
"learning_rate": 0.0004575225677031093,
"loss": 2.6856,
"theoretical_loss": 4.129649467716565,
"tokens_seen": 310575104
},
{
"epoch": 0.09,
"learning_rate": 0.0004575125376128385,
"loss": 3.2868,
"theoretical_loss": 4.1295478122540565,
"tokens_seen": 310640640
},
{
"epoch": 0.09,
"learning_rate": 0.00045750250752256774,
"loss": 3.2094,
"theoretical_loss": 4.1294461842391055,
"tokens_seen": 310706176
},
{
"epoch": 0.09,
"learning_rate": 0.00045749247743229687,
"loss": 2.9271,
"theoretical_loss": 4.129344583658516,
"tokens_seen": 310771712
},
{
"epoch": 0.09,
"learning_rate": 0.0004574824473420261,
"loss": 3.1971,
"theoretical_loss": 4.1292430104990965,
"tokens_seen": 310837248
},
{
"epoch": 0.09,
"learning_rate": 0.0004574724172517553,
"loss": 2.8271,
"theoretical_loss": 4.1291414647476685,
"tokens_seen": 310902784
},
{
"epoch": 0.09,
"learning_rate": 0.00045746238716148446,
"loss": 2.9373,
"theoretical_loss": 4.129039946391062,
"tokens_seen": 310968320
},
{
"epoch": 0.09,
"learning_rate": 0.00045745235707121364,
"loss": 3.2863,
"theoretical_loss": 4.128938455416115,
"tokens_seen": 311033856
},
{
"epoch": 0.09,
"learning_rate": 0.0004574423269809428,
"loss": 3.0255,
"theoretical_loss": 4.128836991809674,
"tokens_seen": 311099392
},
{
"epoch": 0.09,
"learning_rate": 0.000457432296890672,
"loss": 3.1146,
"theoretical_loss": 4.128735555558597,
"tokens_seen": 311164928
},
{
"epoch": 0.09,
"learning_rate": 0.00045742226680040124,
"loss": 2.9283,
"theoretical_loss": 4.128634146649748,
"tokens_seen": 311230464
},
{
"epoch": 0.09,
"objective/train/docs_used": 187875,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4823386669158936,
"objective/train/theoretical_loss": 4.128532765070004,
"objective/train/tokens_used": 331756000,
"theoretical_loss": 4.128532765070004,
"tokens_seen": 311296000
},
{
"epoch": 0.09,
"learning_rate": 0.00045741223671013037,
"loss": 3.1208,
"theoretical_loss": 4.128532765070004,
"tokens_seen": 311296000
},
{
"epoch": 0.09,
"learning_rate": 0.0004574022066198596,
"loss": 2.94,
"theoretical_loss": 4.128431410806247,
"tokens_seen": 311361536
},
{
"epoch": 0.09,
"learning_rate": 0.00045739217652958873,
"loss": 2.9579,
"theoretical_loss": 4.12833008384537,
"tokens_seen": 311427072
},
{
"epoch": 0.09,
"learning_rate": 0.00045738214643931797,
"loss": 3.2517,
"theoretical_loss": 4.128228784174275,
"tokens_seen": 311492608
},
{
"epoch": 0.09,
"learning_rate": 0.00045737211634904715,
"loss": 3.2745,
"theoretical_loss": 4.128127511779873,
"tokens_seen": 311558144
},
{
"epoch": 0.09,
"learning_rate": 0.00045736208625877633,
"loss": 2.9142,
"theoretical_loss": 4.128026266649085,
"tokens_seen": 311623680
},
{
"epoch": 0.09,
"learning_rate": 0.0004573520561685055,
"loss": 3.03,
"theoretical_loss": 4.127925048768839,
"tokens_seen": 311689216
},
{
"epoch": 0.09,
"learning_rate": 0.00045734202607823475,
"loss": 3.068,
"theoretical_loss": 4.127823858126073,
"tokens_seen": 311754752
},
{
"epoch": 0.09,
"learning_rate": 0.00045733199598796387,
"loss": 3.1706,
"theoretical_loss": 4.1277226947077335,
"tokens_seen": 311820288
},
{
"epoch": 0.09,
"learning_rate": 0.0004573219658976931,
"loss": 2.9845,
"theoretical_loss": 4.127621558500778,
"tokens_seen": 311885824
},
{
"epoch": 0.09,
"learning_rate": 0.00045731193580742223,
"loss": 2.9558,
"theoretical_loss": 4.12752044949217,
"tokens_seen": 311951360
},
{
"epoch": 0.09,
"learning_rate": 0.00045730190571715147,
"loss": 3.0432,
"theoretical_loss": 4.127419367668884,
"tokens_seen": 312016896
},
{
"epoch": 0.09,
"learning_rate": 0.00045729187562688065,
"loss": 2.9326,
"theoretical_loss": 4.127318313017904,
"tokens_seen": 312082432
},
{
"epoch": 0.09,
"learning_rate": 0.00045728184553660983,
"loss": 3.057,
"theoretical_loss": 4.12721728552622,
"tokens_seen": 312147968
},
{
"epoch": 0.09,
"learning_rate": 0.000457271815446339,
"loss": 2.9781,
"theoretical_loss": 4.1271162851808345,
"tokens_seen": 312213504
},
{
"epoch": 0.09,
"learning_rate": 0.0004572617853560682,
"loss": 3.2397,
"theoretical_loss": 4.127015311968757,
"tokens_seen": 312279040
},
{
"epoch": 0.09,
"learning_rate": 0.0004572517552657974,
"loss": 3.0724,
"theoretical_loss": 4.126914365877004,
"tokens_seen": 312344576
},
{
"epoch": 0.09,
"learning_rate": 0.0004572417251755266,
"loss": 3.1184,
"theoretical_loss": 4.126813446892607,
"tokens_seen": 312410112
},
{
"epoch": 0.09,
"learning_rate": 0.00045723169508525574,
"loss": 3.0199,
"theoretical_loss": 4.1267125550026,
"tokens_seen": 312475648
},
{
"epoch": 0.09,
"learning_rate": 0.000457221664994985,
"loss": 3.1512,
"theoretical_loss": 4.1266116901940295,
"tokens_seen": 312541184
},
{
"epoch": 0.09,
"learning_rate": 0.0004572116349047141,
"loss": 2.9637,
"theoretical_loss": 4.126510852453949,
"tokens_seen": 312606720
},
{
"epoch": 0.09,
"learning_rate": 0.00045720160481444334,
"loss": 3.1577,
"theoretical_loss": 4.126410041769423,
"tokens_seen": 312672256
},
{
"epoch": 0.09,
"learning_rate": 0.00045719157472417257,
"loss": 2.8805,
"theoretical_loss": 4.126309258127524,
"tokens_seen": 312737792
},
{
"epoch": 0.09,
"learning_rate": 0.0004571815446339017,
"loss": 3.2546,
"theoretical_loss": 4.126208501515331,
"tokens_seen": 312803328
},
{
"epoch": 0.09,
"learning_rate": 0.00045717151454363093,
"loss": 3.0705,
"theoretical_loss": 4.126107771919935,
"tokens_seen": 312868864
},
{
"epoch": 0.09,
"objective/train/docs_used": 188927,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9257431030273438,
"objective/train/theoretical_loss": 4.126007069328436,
"objective/train/tokens_used": 333394400,
"theoretical_loss": 4.126007069328436,
"tokens_seen": 312934400
},
{
"epoch": 0.09,
"learning_rate": 0.0004571614844533601,
"loss": 3.1474,
"theoretical_loss": 4.126007069328436,
"tokens_seen": 312934400
},
{
"epoch": 0.09,
"learning_rate": 0.0004571514543630893,
"loss": 3.0964,
"theoretical_loss": 4.125906393727941,
"tokens_seen": 312999936
},
{
"epoch": 0.09,
"learning_rate": 0.0004571414242728185,
"loss": 3.2524,
"theoretical_loss": 4.125805745105566,
"tokens_seen": 313065472
},
{
"epoch": 0.09,
"learning_rate": 0.00045713139418254766,
"loss": 3.0474,
"theoretical_loss": 4.125705123448437,
"tokens_seen": 313131008
},
{
"epoch": 0.09,
"learning_rate": 0.00045712136409227684,
"loss": 3.2728,
"theoretical_loss": 4.125604528743689,
"tokens_seen": 313196544
},
{
"epoch": 0.09,
"learning_rate": 0.0004571113340020061,
"loss": 3.2944,
"theoretical_loss": 4.125503960978464,
"tokens_seen": 313262080
},
{
"epoch": 0.09,
"learning_rate": 0.0004571013039117352,
"loss": 3.0976,
"theoretical_loss": 4.1254034201399135,
"tokens_seen": 313327616
},
{
"epoch": 0.09,
"learning_rate": 0.00045709127382146444,
"loss": 3.0754,
"theoretical_loss": 4.125302906215199,
"tokens_seen": 313393152
},
{
"epoch": 0.09,
"learning_rate": 0.00045708124373119356,
"loss": 3.1683,
"theoretical_loss": 4.12520241919149,
"tokens_seen": 313458688
},
{
"epoch": 0.1,
"learning_rate": 0.0004570712136409228,
"loss": 3.1685,
"theoretical_loss": 4.125101959055965,
"tokens_seen": 313524224
},
{
"epoch": 0.1,
"learning_rate": 0.000457061183550652,
"loss": 3.3666,
"theoretical_loss": 4.125001525795811,
"tokens_seen": 313589760
},
{
"epoch": 0.1,
"learning_rate": 0.00045705115346038116,
"loss": 3.0122,
"theoretical_loss": 4.124901119398222,
"tokens_seen": 313655296
},
{
"epoch": 0.1,
"learning_rate": 0.00045704112337011034,
"loss": 3.522,
"theoretical_loss": 4.124800739850406,
"tokens_seen": 313720832
},
{
"epoch": 0.1,
"learning_rate": 0.0004570310932798395,
"loss": 2.9341,
"theoretical_loss": 4.124700387139574,
"tokens_seen": 313786368
},
{
"epoch": 0.1,
"learning_rate": 0.0004570210631895687,
"loss": 3.1314,
"theoretical_loss": 4.12460006125295,
"tokens_seen": 313851904
},
{
"epoch": 0.1,
"learning_rate": 0.00045701103309929794,
"loss": 3.1766,
"theoretical_loss": 4.124499762177764,
"tokens_seen": 313917440
},
{
"epoch": 0.1,
"learning_rate": 0.00045700100300902707,
"loss": 3.2033,
"theoretical_loss": 4.124399489901254,
"tokens_seen": 313982976
},
{
"epoch": 0.1,
"learning_rate": 0.0004569909729187563,
"loss": 3.1851,
"theoretical_loss": 4.124299244410672,
"tokens_seen": 314048512
},
{
"epoch": 0.1,
"learning_rate": 0.0004569809428284855,
"loss": 3.3205,
"theoretical_loss": 4.124199025693272,
"tokens_seen": 314114048
},
{
"epoch": 0.1,
"learning_rate": 0.00045697091273821466,
"loss": 2.9844,
"theoretical_loss": 4.124098833736321,
"tokens_seen": 314179584
},
{
"epoch": 0.1,
"learning_rate": 0.00045696088264794384,
"loss": 3.2222,
"theoretical_loss": 4.123998668527094,
"tokens_seen": 314245120
},
{
"epoch": 0.1,
"learning_rate": 0.000456950852557673,
"loss": 3.1528,
"theoretical_loss": 4.123898530052874,
"tokens_seen": 314310656
},
{
"epoch": 0.1,
"learning_rate": 0.0004569408224674022,
"loss": 3.2758,
"theoretical_loss": 4.123798418300953,
"tokens_seen": 314376192
},
{
"epoch": 0.1,
"learning_rate": 0.00045693079237713144,
"loss": 3.1248,
"theoretical_loss": 4.123698333258631,
"tokens_seen": 314441728
},
{
"epoch": 0.1,
"learning_rate": 0.00045692076228686057,
"loss": 3.2705,
"theoretical_loss": 4.123598274913219,
"tokens_seen": 314507264
},
{
"epoch": 0.1,
"objective/train/docs_used": 189547,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8021438121795654,
"objective/train/theoretical_loss": 4.123498243252032,
"objective/train/tokens_used": 335032800,
"theoretical_loss": 4.123498243252032,
"tokens_seen": 314572800
},
{
"epoch": 0.1,
"learning_rate": 0.0004569107321965898,
"loss": 2.9524,
"theoretical_loss": 4.123498243252032,
"tokens_seen": 314572800
},
{
"epoch": 0.1,
"learning_rate": 0.00045690070210631893,
"loss": 2.9818,
"theoretical_loss": 4.1233982382624,
"tokens_seen": 314638336
},
{
"epoch": 0.1,
"learning_rate": 0.00045689067201604817,
"loss": 3.1294,
"theoretical_loss": 4.123298259931657,
"tokens_seen": 314703872
},
{
"epoch": 0.1,
"learning_rate": 0.00045688064192577735,
"loss": 3.2011,
"theoretical_loss": 4.123198308247146,
"tokens_seen": 314769408
},
{
"epoch": 0.1,
"learning_rate": 0.00045687061183550653,
"loss": 3.2102,
"theoretical_loss": 4.123098383196222,
"tokens_seen": 314834944
},
{
"epoch": 0.1,
"learning_rate": 0.0004568605817452357,
"loss": 3.0984,
"theoretical_loss": 4.122998484766244,
"tokens_seen": 314900480
},
{
"epoch": 0.1,
"learning_rate": 0.00045685055165496495,
"loss": 3.0736,
"theoretical_loss": 4.122898612944582,
"tokens_seen": 314966016
},
{
"epoch": 0.1,
"learning_rate": 0.00045684052156469407,
"loss": 3.4115,
"theoretical_loss": 4.122798767718616,
"tokens_seen": 315031552
},
{
"epoch": 0.1,
"learning_rate": 0.0004568304914744233,
"loss": 3.2703,
"theoretical_loss": 4.122698949075732,
"tokens_seen": 315097088
},
{
"epoch": 0.1,
"learning_rate": 0.00045682046138415243,
"loss": 3.3034,
"theoretical_loss": 4.122599157003327,
"tokens_seen": 315162624
},
{
"epoch": 0.1,
"learning_rate": 0.00045681043129388167,
"loss": 3.0698,
"theoretical_loss": 4.1224993914888035,
"tokens_seen": 315228160
},
{
"epoch": 0.1,
"learning_rate": 0.00045680040120361085,
"loss": 3.2926,
"theoretical_loss": 4.122399652519576,
"tokens_seen": 315293696
},
{
"epoch": 0.1,
"learning_rate": 0.00045679037111334003,
"loss": 3.0107,
"theoretical_loss": 4.122299940083065,
"tokens_seen": 315359232
},
{
"epoch": 0.1,
"learning_rate": 0.0004567803410230692,
"loss": 3.1404,
"theoretical_loss": 4.1222002541667,
"tokens_seen": 315424768
},
{
"epoch": 0.1,
"learning_rate": 0.0004567703109327984,
"loss": 3.2998,
"theoretical_loss": 4.122100594757921,
"tokens_seen": 315490304
},
{
"epoch": 0.1,
"learning_rate": 0.0004567602808425276,
"loss": 3.2533,
"theoretical_loss": 4.122000961844175,
"tokens_seen": 315555840
},
{
"epoch": 0.1,
"learning_rate": 0.0004567502507522568,
"loss": 3.2643,
"theoretical_loss": 4.121901355412917,
"tokens_seen": 315621376
},
{
"epoch": 0.1,
"learning_rate": 0.00045674022066198594,
"loss": 3.0887,
"theoretical_loss": 4.121801775451612,
"tokens_seen": 315686912
},
{
"epoch": 0.1,
"learning_rate": 0.0004567301905717152,
"loss": 3.0258,
"theoretical_loss": 4.121702221947732,
"tokens_seen": 315752448
},
{
"epoch": 0.1,
"learning_rate": 0.0004567201604814443,
"loss": 3.0984,
"theoretical_loss": 4.121602694888759,
"tokens_seen": 315817984
},
{
"epoch": 0.1,
"learning_rate": 0.00045671013039117354,
"loss": 3.0961,
"theoretical_loss": 4.121503194262183,
"tokens_seen": 315883520
},
{
"epoch": 0.1,
"learning_rate": 0.0004567001003009027,
"loss": 3.4246,
"theoretical_loss": 4.121403720055502,
"tokens_seen": 315949056
},
{
"epoch": 0.1,
"learning_rate": 0.0004566900702106319,
"loss": 3.1232,
"theoretical_loss": 4.121304272256222,
"tokens_seen": 316014592
},
{
"epoch": 0.1,
"learning_rate": 0.0004566800401203611,
"loss": 3.2672,
"theoretical_loss": 4.121204850851861,
"tokens_seen": 316080128
},
{
"epoch": 0.1,
"learning_rate": 0.0004566700100300903,
"loss": 3.2468,
"theoretical_loss": 4.121105455829939,
"tokens_seen": 316145664
},
{
"epoch": 0.1,
"objective/train/docs_used": 190733,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.28241229057312,
"objective/train/theoretical_loss": 4.121006087177992,
"objective/train/tokens_used": 336671200,
"theoretical_loss": 4.121006087177992,
"tokens_seen": 316211200
},
{
"epoch": 0.1,
"learning_rate": 0.00045665997993981944,
"loss": 3.0047,
"theoretical_loss": 4.121006087177992,
"tokens_seen": 316211200
},
{
"epoch": 0.1,
"learning_rate": 0.0004566499498495487,
"loss": 3.0604,
"theoretical_loss": 4.120906744883559,
"tokens_seen": 316276736
},
{
"epoch": 0.1,
"learning_rate": 0.0004566399197592778,
"loss": 3.1629,
"theoretical_loss": 4.120807428934189,
"tokens_seen": 316342272
},
{
"epoch": 0.1,
"learning_rate": 0.00045662988966900704,
"loss": 3.0635,
"theoretical_loss": 4.120708139317441,
"tokens_seen": 316407808
},
{
"epoch": 0.1,
"learning_rate": 0.0004566198595787362,
"loss": 3.2192,
"theoretical_loss": 4.12060887602088,
"tokens_seen": 316473344
},
{
"epoch": 0.1,
"learning_rate": 0.0004566098294884654,
"loss": 3.1621,
"theoretical_loss": 4.120509639032081,
"tokens_seen": 316538880
},
{
"epoch": 0.1,
"learning_rate": 0.0004565997993981946,
"loss": 2.9827,
"theoretical_loss": 4.120410428338628,
"tokens_seen": 316604416
},
{
"epoch": 0.1,
"learning_rate": 0.00045658976930792376,
"loss": 2.9799,
"theoretical_loss": 4.120311243928111,
"tokens_seen": 316669952
},
{
"epoch": 0.1,
"learning_rate": 0.00045657973921765294,
"loss": 3.195,
"theoretical_loss": 4.120212085788131,
"tokens_seen": 316735488
},
{
"epoch": 0.1,
"learning_rate": 0.0004565697091273822,
"loss": 3.0131,
"theoretical_loss": 4.120112953906296,
"tokens_seen": 316801024
},
{
"epoch": 0.1,
"learning_rate": 0.0004565596790371113,
"loss": 3.2145,
"theoretical_loss": 4.120013848270222,
"tokens_seen": 316866560
},
{
"epoch": 0.1,
"learning_rate": 0.00045654964894684054,
"loss": 3.1744,
"theoretical_loss": 4.119914768867536,
"tokens_seen": 316932096
},
{
"epoch": 0.1,
"learning_rate": 0.00045653961885656967,
"loss": 2.9482,
"theoretical_loss": 4.11981571568587,
"tokens_seen": 316997632
},
{
"epoch": 0.1,
"learning_rate": 0.0004565295887662989,
"loss": 3.1156,
"theoretical_loss": 4.119716688712866,
"tokens_seen": 317063168
},
{
"epoch": 0.1,
"learning_rate": 0.0004565195586760281,
"loss": 3.0127,
"theoretical_loss": 4.119617687936175,
"tokens_seen": 317128704
},
{
"epoch": 0.1,
"learning_rate": 0.00045650952858575727,
"loss": 2.9893,
"theoretical_loss": 4.119518713343455,
"tokens_seen": 317194240
},
{
"epoch": 0.1,
"learning_rate": 0.00045649949849548645,
"loss": 2.9135,
"theoretical_loss": 4.119419764922374,
"tokens_seen": 317259776
},
{
"epoch": 0.1,
"learning_rate": 0.0004564894684052157,
"loss": 2.9428,
"theoretical_loss": 4.119320842660606,
"tokens_seen": 317325312
},
{
"epoch": 0.1,
"learning_rate": 0.0004564794383149448,
"loss": 3.0893,
"theoretical_loss": 4.119221946545836,
"tokens_seen": 317390848
},
{
"epoch": 0.1,
"learning_rate": 0.00045646940822467405,
"loss": 2.954,
"theoretical_loss": 4.119123076565755,
"tokens_seen": 317456384
},
{
"epoch": 0.1,
"learning_rate": 0.00045645937813440317,
"loss": 3.1031,
"theoretical_loss": 4.119024232708064,
"tokens_seen": 317521920
},
{
"epoch": 0.1,
"learning_rate": 0.0004564493480441324,
"loss": 3.1224,
"theoretical_loss": 4.118925414960472,
"tokens_seen": 317587456
},
{
"epoch": 0.1,
"learning_rate": 0.00045643931795386164,
"loss": 3.1016,
"theoretical_loss": 4.118826623310696,
"tokens_seen": 317652992
},
{
"epoch": 0.1,
"learning_rate": 0.00045642928786359077,
"loss": 3.0232,
"theoretical_loss": 4.11872785774646,
"tokens_seen": 317718528
},
{
"epoch": 0.1,
"learning_rate": 0.00045641925777332,
"loss": 3.211,
"theoretical_loss": 4.1186291182555,
"tokens_seen": 317784064
},
{
"epoch": 0.1,
"objective/train/docs_used": 191330,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.303915023803711,
"objective/train/theoretical_loss": 4.118530404825556,
"objective/train/tokens_used": 338309600,
"theoretical_loss": 4.118530404825556,
"tokens_seen": 317849600
},
{
"epoch": 0.1,
"learning_rate": 0.00045640922768304913,
"loss": 3.1621,
"theoretical_loss": 4.118530404825556,
"tokens_seen": 317849600
},
{
"epoch": 0.1,
"learning_rate": 0.00045639919759277837,
"loss": 3.2408,
"theoretical_loss": 4.11843171744438,
"tokens_seen": 317915136
},
{
"epoch": 0.1,
"learning_rate": 0.00045638916750250755,
"loss": 3.0122,
"theoretical_loss": 4.118333056099728,
"tokens_seen": 317980672
},
{
"epoch": 0.1,
"learning_rate": 0.00045637913741223673,
"loss": 3.3026,
"theoretical_loss": 4.11823442077937,
"tokens_seen": 318046208
},
{
"epoch": 0.1,
"learning_rate": 0.0004563691073219659,
"loss": 3.1502,
"theoretical_loss": 4.1181358114710775,
"tokens_seen": 318111744
},
{
"epoch": 0.1,
"learning_rate": 0.00045635907723169515,
"loss": 3.1201,
"theoretical_loss": 4.1180372281626365,
"tokens_seen": 318177280
},
{
"epoch": 0.1,
"learning_rate": 0.00045634904714142427,
"loss": 2.9375,
"theoretical_loss": 4.117938670841838,
"tokens_seen": 318242816
},
{
"epoch": 0.1,
"learning_rate": 0.0004563390170511535,
"loss": 3.2836,
"theoretical_loss": 4.117840139496482,
"tokens_seen": 318308352
},
{
"epoch": 0.1,
"learning_rate": 0.00045632898696088263,
"loss": 3.2809,
"theoretical_loss": 4.117741634114376,
"tokens_seen": 318373888
},
{
"epoch": 0.1,
"learning_rate": 0.00045631895687061187,
"loss": 3.3275,
"theoretical_loss": 4.1176431546833365,
"tokens_seen": 318439424
},
{
"epoch": 0.1,
"learning_rate": 0.00045630892678034105,
"loss": 3.2046,
"theoretical_loss": 4.117544701191187,
"tokens_seen": 318504960
},
{
"epoch": 0.1,
"learning_rate": 0.00045629889669007023,
"loss": 2.7935,
"theoretical_loss": 4.117446273625763,
"tokens_seen": 318570496
},
{
"epoch": 0.1,
"learning_rate": 0.0004562888665997994,
"loss": 3.1972,
"theoretical_loss": 4.117347871974903,
"tokens_seen": 318636032
},
{
"epoch": 0.1,
"learning_rate": 0.0004562788365095286,
"loss": 3.283,
"theoretical_loss": 4.1172494962264565,
"tokens_seen": 318701568
},
{
"epoch": 0.1,
"learning_rate": 0.0004562688064192578,
"loss": 3.0075,
"theoretical_loss": 4.117151146368282,
"tokens_seen": 318767104
},
{
"epoch": 0.1,
"learning_rate": 0.000456258776328987,
"loss": 3.2145,
"theoretical_loss": 4.117052822388243,
"tokens_seen": 318832640
},
{
"epoch": 0.1,
"learning_rate": 0.00045624874623871614,
"loss": 2.6977,
"theoretical_loss": 4.116954524274216,
"tokens_seen": 318898176
},
{
"epoch": 0.1,
"learning_rate": 0.0004562387161484454,
"loss": 3.2573,
"theoretical_loss": 4.11685625201408,
"tokens_seen": 318963712
},
{
"epoch": 0.1,
"learning_rate": 0.0004562286860581745,
"loss": 2.9499,
"theoretical_loss": 4.116758005595727,
"tokens_seen": 319029248
},
{
"epoch": 0.1,
"learning_rate": 0.00045621865596790374,
"loss": 2.9781,
"theoretical_loss": 4.116659785007055,
"tokens_seen": 319094784
},
{
"epoch": 0.1,
"learning_rate": 0.0004562086258776329,
"loss": 3.1879,
"theoretical_loss": 4.116561590235969,
"tokens_seen": 319160320
},
{
"epoch": 0.1,
"learning_rate": 0.0004561985957873621,
"loss": 3.1697,
"theoretical_loss": 4.116463421270385,
"tokens_seen": 319225856
},
{
"epoch": 0.1,
"learning_rate": 0.0004561885656970913,
"loss": 2.9933,
"theoretical_loss": 4.116365278098225,
"tokens_seen": 319291392
},
{
"epoch": 0.1,
"learning_rate": 0.0004561785356068205,
"loss": 3.1033,
"theoretical_loss": 4.116267160707421,
"tokens_seen": 319356928
},
{
"epoch": 0.1,
"learning_rate": 0.00045616850551654964,
"loss": 2.9044,
"theoretical_loss": 4.11616906908591,
"tokens_seen": 319422464
},
{
"epoch": 0.1,
"objective/train/docs_used": 192764,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9580471515655518,
"objective/train/theoretical_loss": 4.11607100322164,
"objective/train/tokens_used": 339948000,
"theoretical_loss": 4.11607100322164,
"tokens_seen": 319488000
},
{
"epoch": 0.1,
"learning_rate": 0.0004561584754262789,
"loss": 2.9887,
"theoretical_loss": 4.11607100322164,
"tokens_seen": 319488000
},
{
"epoch": 0.1,
"learning_rate": 0.000456148445336008,
"loss": 3.0573,
"theoretical_loss": 4.115972963102565,
"tokens_seen": 319553536
},
{
"epoch": 0.1,
"learning_rate": 0.00045613841524573724,
"loss": 3.2024,
"theoretical_loss": 4.11587494871665,
"tokens_seen": 319619072
},
{
"epoch": 0.1,
"learning_rate": 0.0004561283851554664,
"loss": 3.2376,
"theoretical_loss": 4.115776960051864,
"tokens_seen": 319684608
},
{
"epoch": 0.1,
"learning_rate": 0.0004561183550651956,
"loss": 3.0396,
"theoretical_loss": 4.11567899709619,
"tokens_seen": 319750144
},
{
"epoch": 0.1,
"learning_rate": 0.0004561083249749248,
"loss": 2.7815,
"theoretical_loss": 4.115581059837612,
"tokens_seen": 319815680
},
{
"epoch": 0.1,
"learning_rate": 0.00045609829488465396,
"loss": 3.1572,
"theoretical_loss": 4.115483148264127,
"tokens_seen": 319881216
},
{
"epoch": 0.1,
"learning_rate": 0.00045608826479438314,
"loss": 3.0449,
"theoretical_loss": 4.115385262363739,
"tokens_seen": 319946752
},
{
"epoch": 0.1,
"learning_rate": 0.0004560782347041124,
"loss": 3.0447,
"theoretical_loss": 4.1152874021244585,
"tokens_seen": 320012288
},
{
"epoch": 0.1,
"learning_rate": 0.0004560682046138415,
"loss": 2.905,
"theoretical_loss": 4.115189567534307,
"tokens_seen": 320077824
},
{
"epoch": 0.1,
"learning_rate": 0.00045605817452357074,
"loss": 3.0628,
"theoretical_loss": 4.115091758581309,
"tokens_seen": 320143360
},
{
"epoch": 0.1,
"learning_rate": 0.00045604814443329987,
"loss": 2.9603,
"theoretical_loss": 4.114993975253505,
"tokens_seen": 320208896
},
{
"epoch": 0.1,
"learning_rate": 0.0004560381143430291,
"loss": 2.9837,
"theoretical_loss": 4.114896217538935,
"tokens_seen": 320274432
},
{
"epoch": 0.1,
"learning_rate": 0.0004560280842527583,
"loss": 2.8237,
"theoretical_loss": 4.114798485425652,
"tokens_seen": 320339968
},
{
"epoch": 0.1,
"learning_rate": 0.00045601805416248747,
"loss": 3.0052,
"theoretical_loss": 4.114700778901717,
"tokens_seen": 320405504
},
{
"epoch": 0.1,
"learning_rate": 0.00045600802407221665,
"loss": 3.0487,
"theoretical_loss": 4.114603097955197,
"tokens_seen": 320471040
},
{
"epoch": 0.1,
"learning_rate": 0.0004559979939819459,
"loss": 3.2513,
"theoretical_loss": 4.114505442574167,
"tokens_seen": 320536576
},
{
"epoch": 0.1,
"learning_rate": 0.000455987963891675,
"loss": 3.2844,
"theoretical_loss": 4.1144078127467125,
"tokens_seen": 320602112
},
{
"epoch": 0.1,
"learning_rate": 0.00045597793380140425,
"loss": 2.7771,
"theoretical_loss": 4.114310208460924,
"tokens_seen": 320667648
},
{
"epoch": 0.1,
"learning_rate": 0.00045596790371113337,
"loss": 3.1613,
"theoretical_loss": 4.114212629704902,
"tokens_seen": 320733184
},
{
"epoch": 0.1,
"learning_rate": 0.0004559578736208626,
"loss": 2.9396,
"theoretical_loss": 4.114115076466755,
"tokens_seen": 320798720
},
{
"epoch": 0.1,
"learning_rate": 0.0004559478435305918,
"loss": 2.9128,
"theoretical_loss": 4.114017548734598,
"tokens_seen": 320864256
},
{
"epoch": 0.1,
"learning_rate": 0.00045593781344032097,
"loss": 2.9857,
"theoretical_loss": 4.113920046496554,
"tokens_seen": 320929792
},
{
"epoch": 0.1,
"learning_rate": 0.00045592778335005015,
"loss": 3.1028,
"theoretical_loss": 4.113822569740757,
"tokens_seen": 320995328
},
{
"epoch": 0.1,
"learning_rate": 0.00045591775325977933,
"loss": 3.5144,
"theoretical_loss": 4.113725118455344,
"tokens_seen": 321060864
},
{
"epoch": 0.1,
"objective/train/docs_used": 193449,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.532770872116089,
"objective/train/theoretical_loss": 4.113627692628464,
"objective/train/tokens_used": 341586400,
"theoretical_loss": 4.113627692628464,
"tokens_seen": 321126400
},
{
"epoch": 0.1,
"learning_rate": 0.0004559077231695085,
"loss": 3.0643,
"theoretical_loss": 4.113627692628464,
"tokens_seen": 321126400
},
{
"epoch": 0.1,
"learning_rate": 0.00045589769307923775,
"loss": 3.1924,
"theoretical_loss": 4.113530292248273,
"tokens_seen": 321191936
},
{
"epoch": 0.1,
"learning_rate": 0.0004558876629889669,
"loss": 2.9817,
"theoretical_loss": 4.113432917302934,
"tokens_seen": 321257472
},
{
"epoch": 0.1,
"learning_rate": 0.0004558776328986961,
"loss": 2.954,
"theoretical_loss": 4.113335567780618,
"tokens_seen": 321323008
},
{
"epoch": 0.1,
"learning_rate": 0.00045586760280842524,
"loss": 2.9283,
"theoretical_loss": 4.113238243669504,
"tokens_seen": 321388544
},
{
"epoch": 0.1,
"learning_rate": 0.0004558575727181545,
"loss": 3.0588,
"theoretical_loss": 4.113140944957781,
"tokens_seen": 321454080
},
{
"epoch": 0.1,
"learning_rate": 0.00045584754262788365,
"loss": 2.9978,
"theoretical_loss": 4.113043671633641,
"tokens_seen": 321519616
},
{
"epoch": 0.1,
"learning_rate": 0.00045583751253761284,
"loss": 3.1526,
"theoretical_loss": 4.11294642368529,
"tokens_seen": 321585152
},
{
"epoch": 0.1,
"learning_rate": 0.000455827482447342,
"loss": 2.856,
"theoretical_loss": 4.112849201100938,
"tokens_seen": 321650688
},
{
"epoch": 0.1,
"learning_rate": 0.00045581745235707125,
"loss": 3.0283,
"theoretical_loss": 4.1127520038688035,
"tokens_seen": 321716224
},
{
"epoch": 0.1,
"learning_rate": 0.0004558074222668004,
"loss": 3.1706,
"theoretical_loss": 4.112654831977112,
"tokens_seen": 321781760
},
{
"epoch": 0.1,
"learning_rate": 0.0004557973921765296,
"loss": 3.1208,
"theoretical_loss": 4.1125576854141,
"tokens_seen": 321847296
},
{
"epoch": 0.1,
"learning_rate": 0.00045578736208625874,
"loss": 3.2345,
"theoretical_loss": 4.112460564168009,
"tokens_seen": 321912832
},
{
"epoch": 0.1,
"learning_rate": 0.000455777331995988,
"loss": 3.0428,
"theoretical_loss": 4.112363468227088,
"tokens_seen": 321978368
},
{
"epoch": 0.1,
"learning_rate": 0.00045576730190571716,
"loss": 3.3738,
"theoretical_loss": 4.112266397579598,
"tokens_seen": 322043904
},
{
"epoch": 0.1,
"learning_rate": 0.00045575727181544634,
"loss": 3.0012,
"theoretical_loss": 4.112169352213801,
"tokens_seen": 322109440
},
{
"epoch": 0.1,
"learning_rate": 0.0004557472417251755,
"loss": 2.7209,
"theoretical_loss": 4.1120723321179735,
"tokens_seen": 322174976
},
{
"epoch": 0.1,
"learning_rate": 0.0004557372116349047,
"loss": 3.0685,
"theoretical_loss": 4.111975337280397,
"tokens_seen": 322240512
},
{
"epoch": 0.1,
"learning_rate": 0.0004557271815446339,
"loss": 3.2726,
"theoretical_loss": 4.111878367689359,
"tokens_seen": 322306048
},
{
"epoch": 0.1,
"learning_rate": 0.0004557171514543631,
"loss": 2.9605,
"theoretical_loss": 4.1117814233331575,
"tokens_seen": 322371584
},
{
"epoch": 0.1,
"learning_rate": 0.00045570712136409224,
"loss": 3.0389,
"theoretical_loss": 4.111684504200099,
"tokens_seen": 322437120
},
{
"epoch": 0.1,
"learning_rate": 0.0004556970912738215,
"loss": 3.0147,
"theoretical_loss": 4.111587610278494,
"tokens_seen": 322502656
},
{
"epoch": 0.1,
"learning_rate": 0.0004556870611835507,
"loss": 3.126,
"theoretical_loss": 4.111490741556663,
"tokens_seen": 322568192
},
{
"epoch": 0.1,
"learning_rate": 0.00045567703109327984,
"loss": 3.2523,
"theoretical_loss": 4.1113938980229365,
"tokens_seen": 322633728
},
{
"epoch": 0.1,
"learning_rate": 0.0004556670010030091,
"loss": 2.9724,
"theoretical_loss": 4.11129707966565,
"tokens_seen": 322699264
},
{
"epoch": 0.1,
"objective/train/docs_used": 194727,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2888762950897217,
"objective/train/theoretical_loss": 4.111200286473145,
"objective/train/tokens_used": 343224800,
"theoretical_loss": 4.111200286473145,
"tokens_seen": 322764800
},
{
"epoch": 0.1,
"learning_rate": 0.0004556569709127382,
"loss": 3.1618,
"theoretical_loss": 4.111200286473145,
"tokens_seen": 322764800
},
{
"epoch": 0.1,
"learning_rate": 0.00045564694082246744,
"loss": 2.8571,
"theoretical_loss": 4.111103518433776,
"tokens_seen": 322830336
},
{
"epoch": 0.1,
"learning_rate": 0.0004556369107321966,
"loss": 3.1171,
"theoretical_loss": 4.111006775535901,
"tokens_seen": 322895872
},
{
"epoch": 0.1,
"learning_rate": 0.0004556268806419258,
"loss": 3.0964,
"theoretical_loss": 4.110910057767887,
"tokens_seen": 322961408
},
{
"epoch": 0.1,
"learning_rate": 0.000455616850551655,
"loss": 3.2273,
"theoretical_loss": 4.110813365118109,
"tokens_seen": 323026944
},
{
"epoch": 0.1,
"learning_rate": 0.00045560682046138416,
"loss": 3.0086,
"theoretical_loss": 4.110716697574951,
"tokens_seen": 323092480
},
{
"epoch": 0.1,
"learning_rate": 0.00045559679037111334,
"loss": 2.9121,
"theoretical_loss": 4.110620055126802,
"tokens_seen": 323158016
},
{
"epoch": 0.1,
"learning_rate": 0.0004555867602808426,
"loss": 3.1999,
"theoretical_loss": 4.110523437762059,
"tokens_seen": 323223552
},
{
"epoch": 0.1,
"learning_rate": 0.0004555767301905717,
"loss": 3.254,
"theoretical_loss": 4.11042684546913,
"tokens_seen": 323289088
},
{
"epoch": 0.1,
"learning_rate": 0.00045556670010030094,
"loss": 3.0221,
"theoretical_loss": 4.110330278236427,
"tokens_seen": 323354624
},
{
"epoch": 0.1,
"learning_rate": 0.00045555667001003007,
"loss": 2.9684,
"theoretical_loss": 4.110233736052372,
"tokens_seen": 323420160
},
{
"epoch": 0.1,
"learning_rate": 0.0004555466399197593,
"loss": 3.2605,
"theoretical_loss": 4.110137218905393,
"tokens_seen": 323485696
},
{
"epoch": 0.1,
"learning_rate": 0.0004555366098294885,
"loss": 3.1783,
"theoretical_loss": 4.110040726783927,
"tokens_seen": 323551232
},
{
"epoch": 0.1,
"learning_rate": 0.00045552657973921767,
"loss": 3.0009,
"theoretical_loss": 4.109944259676419,
"tokens_seen": 323616768
},
{
"epoch": 0.1,
"learning_rate": 0.00045551654964894685,
"loss": 2.9394,
"theoretical_loss": 4.109847817571319,
"tokens_seen": 323682304
},
{
"epoch": 0.1,
"learning_rate": 0.0004555065195586761,
"loss": 3.0756,
"theoretical_loss": 4.109751400457089,
"tokens_seen": 323747840
},
{
"epoch": 0.1,
"learning_rate": 0.0004554964894684052,
"loss": 3.1542,
"theoretical_loss": 4.109655008322195,
"tokens_seen": 323813376
},
{
"epoch": 0.1,
"learning_rate": 0.00045548645937813445,
"loss": 3.0225,
"theoretical_loss": 4.109558641155112,
"tokens_seen": 323878912
},
{
"epoch": 0.1,
"learning_rate": 0.00045547642928786357,
"loss": 3.1003,
"theoretical_loss": 4.109462298944322,
"tokens_seen": 323944448
},
{
"epoch": 0.1,
"learning_rate": 0.0004554663991975928,
"loss": 3.1359,
"theoretical_loss": 4.109365981678316,
"tokens_seen": 324009984
},
{
"epoch": 0.1,
"learning_rate": 0.000455456369107322,
"loss": 3.0616,
"theoretical_loss": 4.109269689345592,
"tokens_seen": 324075520
},
{
"epoch": 0.1,
"learning_rate": 0.00045544633901705117,
"loss": 3.0413,
"theoretical_loss": 4.109173421934654,
"tokens_seen": 324141056
},
{
"epoch": 0.1,
"learning_rate": 0.00045543630892678035,
"loss": 3.1208,
"theoretical_loss": 4.109077179434016,
"tokens_seen": 324206592
},
{
"epoch": 0.1,
"learning_rate": 0.00045542627883650953,
"loss": 3.1749,
"theoretical_loss": 4.1089809618321995,
"tokens_seen": 324272128
},
{
"epoch": 0.1,
"learning_rate": 0.0004554162487462387,
"loss": 2.6465,
"theoretical_loss": 4.108884769117731,
"tokens_seen": 324337664
},
{
"epoch": 0.1,
"objective/train/docs_used": 195463,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8933980464935303,
"objective/train/theoretical_loss": 4.108788601279149,
"objective/train/tokens_used": 344863200,
"theoretical_loss": 4.108788601279149,
"tokens_seen": 324403200
},
{
"epoch": 0.1,
"learning_rate": 0.00045540621865596795,
"loss": 3.0332,
"theoretical_loss": 4.108788601279149,
"tokens_seen": 324403200
},
{
"epoch": 0.1,
"learning_rate": 0.0004553961885656971,
"loss": 2.9037,
"theoretical_loss": 4.108692458304994,
"tokens_seen": 324468736
},
{
"epoch": 0.1,
"learning_rate": 0.0004553861584754263,
"loss": 3.2058,
"theoretical_loss": 4.108596340183819,
"tokens_seen": 324534272
},
{
"epoch": 0.1,
"learning_rate": 0.00045537612838515544,
"loss": 3.1413,
"theoretical_loss": 4.108500246904184,
"tokens_seen": 324599808
},
{
"epoch": 0.1,
"learning_rate": 0.0004553660982948847,
"loss": 3.1149,
"theoretical_loss": 4.108404178454651,
"tokens_seen": 324665344
},
{
"epoch": 0.1,
"learning_rate": 0.00045535606820461385,
"loss": 2.7925,
"theoretical_loss": 4.1083081348237975,
"tokens_seen": 324730880
},
{
"epoch": 0.1,
"learning_rate": 0.00045534603811434304,
"loss": 3.0427,
"theoretical_loss": 4.108212116000203,
"tokens_seen": 324796416
},
{
"epoch": 0.1,
"learning_rate": 0.0004553360080240722,
"loss": 2.9514,
"theoretical_loss": 4.108116121972457,
"tokens_seen": 324861952
},
{
"epoch": 0.1,
"learning_rate": 0.00045532597793380145,
"loss": 3.0382,
"theoretical_loss": 4.108020152729157,
"tokens_seen": 324927488
},
{
"epoch": 0.1,
"learning_rate": 0.0004553159478435306,
"loss": 3.0567,
"theoretical_loss": 4.107924208258905,
"tokens_seen": 324993024
},
{
"epoch": 0.1,
"learning_rate": 0.0004553059177532598,
"loss": 3.0865,
"theoretical_loss": 4.107828288550314,
"tokens_seen": 325058560
},
{
"epoch": 0.1,
"learning_rate": 0.00045529588766298894,
"loss": 3.1368,
"theoretical_loss": 4.107732393592003,
"tokens_seen": 325124096
},
{
"epoch": 0.1,
"learning_rate": 0.0004552858575727182,
"loss": 2.9862,
"theoretical_loss": 4.107636523372598,
"tokens_seen": 325189632
},
{
"epoch": 0.1,
"learning_rate": 0.00045527582748244736,
"loss": 3.0282,
"theoretical_loss": 4.107540677880734,
"tokens_seen": 325255168
},
{
"epoch": 0.1,
"learning_rate": 0.00045526579739217654,
"loss": 3.1162,
"theoretical_loss": 4.107444857105052,
"tokens_seen": 325320704
},
{
"epoch": 0.1,
"learning_rate": 0.0004552557673019057,
"loss": 3.1163,
"theoretical_loss": 4.107349061034201,
"tokens_seen": 325386240
},
{
"epoch": 0.1,
"learning_rate": 0.0004552457372116349,
"loss": 2.8527,
"theoretical_loss": 4.107253289656838,
"tokens_seen": 325451776
},
{
"epoch": 0.1,
"learning_rate": 0.0004552357071213641,
"loss": 3.0606,
"theoretical_loss": 4.107157542961628,
"tokens_seen": 325517312
},
{
"epoch": 0.1,
"learning_rate": 0.0004552256770310933,
"loss": 3.0527,
"theoretical_loss": 4.10706182093724,
"tokens_seen": 325582848
},
{
"epoch": 0.1,
"learning_rate": 0.00045521564694082244,
"loss": 3.4321,
"theoretical_loss": 4.106966123572356,
"tokens_seen": 325648384
},
{
"epoch": 0.1,
"learning_rate": 0.0004552056168505517,
"loss": 2.9409,
"theoretical_loss": 4.106870450855661,
"tokens_seen": 325713920
},
{
"epoch": 0.1,
"learning_rate": 0.0004551955867602808,
"loss": 3.1752,
"theoretical_loss": 4.106774802775849,
"tokens_seen": 325779456
},
{
"epoch": 0.1,
"learning_rate": 0.00045518555667001004,
"loss": 3.0561,
"theoretical_loss": 4.106679179321622,
"tokens_seen": 325844992
},
{
"epoch": 0.1,
"learning_rate": 0.0004551755265797392,
"loss": 3.091,
"theoretical_loss": 4.106583580481689,
"tokens_seen": 325910528
},
{
"epoch": 0.1,
"learning_rate": 0.0004551654964894684,
"loss": 3.0422,
"theoretical_loss": 4.106488006244767,
"tokens_seen": 325976064
},
{
"epoch": 0.1,
"objective/train/docs_used": 196731,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.885859727859497,
"objective/train/theoretical_loss": 4.106392456599577,
"objective/train/tokens_used": 346501600,
"theoretical_loss": 4.106392456599577,
"tokens_seen": 326041600
},
{
"epoch": 0.1,
"learning_rate": 0.0004551554663991976,
"loss": 2.9643,
"theoretical_loss": 4.106392456599577,
"tokens_seen": 326041600
},
{
"epoch": 0.1,
"learning_rate": 0.0004551454363089268,
"loss": 3.112,
"theoretical_loss": 4.106296931534854,
"tokens_seen": 326107136
},
{
"epoch": 0.1,
"learning_rate": 0.00045513540621865595,
"loss": 2.9576,
"theoretical_loss": 4.106201431039335,
"tokens_seen": 326172672
},
{
"epoch": 0.1,
"learning_rate": 0.0004551253761283852,
"loss": 2.6294,
"theoretical_loss": 4.106105955101766,
"tokens_seen": 326238208
},
{
"epoch": 0.1,
"learning_rate": 0.0004551153460381143,
"loss": 3.1334,
"theoretical_loss": 4.1060105037109,
"tokens_seen": 326303744
},
{
"epoch": 0.1,
"learning_rate": 0.00045510531594784354,
"loss": 2.9134,
"theoretical_loss": 4.105915076855499,
"tokens_seen": 326369280
},
{
"epoch": 0.1,
"learning_rate": 0.0004550952858575727,
"loss": 3.0419,
"theoretical_loss": 4.105819674524332,
"tokens_seen": 326434816
},
{
"epoch": 0.1,
"learning_rate": 0.0004550852557673019,
"loss": 3.009,
"theoretical_loss": 4.105724296706172,
"tokens_seen": 326500352
},
{
"epoch": 0.1,
"learning_rate": 0.0004550752256770311,
"loss": 2.9717,
"theoretical_loss": 4.105628943389805,
"tokens_seen": 326565888
},
{
"epoch": 0.1,
"learning_rate": 0.00045506519558676027,
"loss": 2.9638,
"theoretical_loss": 4.1055336145640196,
"tokens_seen": 326631424
},
{
"epoch": 0.1,
"learning_rate": 0.00045505516549648945,
"loss": 3.3072,
"theoretical_loss": 4.105438310217615,
"tokens_seen": 326696960
},
{
"epoch": 0.1,
"learning_rate": 0.0004550451354062187,
"loss": 3.0272,
"theoretical_loss": 4.105343030339395,
"tokens_seen": 326762496
},
{
"epoch": 0.1,
"learning_rate": 0.0004550351053159478,
"loss": 3.0128,
"theoretical_loss": 4.1052477749181735,
"tokens_seen": 326828032
},
{
"epoch": 0.1,
"learning_rate": 0.00045502507522567705,
"loss": 3.0124,
"theoretical_loss": 4.10515254394277,
"tokens_seen": 326893568
},
{
"epoch": 0.1,
"learning_rate": 0.00045501504513540623,
"loss": 2.8748,
"theoretical_loss": 4.1050573374020125,
"tokens_seen": 326959104
},
{
"epoch": 0.1,
"learning_rate": 0.0004550050150451354,
"loss": 3.0619,
"theoretical_loss": 4.104962155284734,
"tokens_seen": 327024640
},
{
"epoch": 0.1,
"learning_rate": 0.0004549949849548646,
"loss": 2.8882,
"theoretical_loss": 4.104866997579778,
"tokens_seen": 327090176
},
{
"epoch": 0.1,
"learning_rate": 0.00045498495486459377,
"loss": 3.0027,
"theoretical_loss": 4.104771864275993,
"tokens_seen": 327155712
},
{
"epoch": 0.1,
"learning_rate": 0.00045497492477432295,
"loss": 3.1058,
"theoretical_loss": 4.104676755362237,
"tokens_seen": 327221248
},
{
"epoch": 0.1,
"learning_rate": 0.0004549648946840522,
"loss": 3.0508,
"theoretical_loss": 4.104581670827372,
"tokens_seen": 327286784
},
{
"epoch": 0.1,
"learning_rate": 0.00045495486459378137,
"loss": 2.8934,
"theoretical_loss": 4.10448661066027,
"tokens_seen": 327352320
},
{
"epoch": 0.1,
"learning_rate": 0.00045494483450351055,
"loss": 3.1344,
"theoretical_loss": 4.104391574849812,
"tokens_seen": 327417856
},
{
"epoch": 0.1,
"learning_rate": 0.00045493480441323973,
"loss": 3.0873,
"theoretical_loss": 4.10429656338488,
"tokens_seen": 327483392
},
{
"epoch": 0.1,
"learning_rate": 0.0004549247743229689,
"loss": 3.0143,
"theoretical_loss": 4.104201576254369,
"tokens_seen": 327548928
},
{
"epoch": 0.1,
"learning_rate": 0.00045491474423269815,
"loss": 2.9043,
"theoretical_loss": 4.10410661344718,
"tokens_seen": 327614464
},
{
"debugging/Self-BLEU-5": 0.3841532625917073,
"debugging/distinct-1-grams": 0.7832216589580511,
"debugging/distinct-2-grams": 0.9618292669346826,
"debugging/entropy-1-grams": 5.637375639920272,
"debugging/entropy-2-grams": 6.413549369763027,
"debugging/length": 486.6666666666667,
"debugging/num_segments": 9,
"epoch": 0.1,
"objective/train/docs_used": 197513,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.052098035812378,
"objective/train/theoretical_loss": 4.10401167495222,
"objective/train/tokens_used": 348140000,
"theoretical_loss": 4.10401167495222,
"tokens_seen": 327680000
},
{
"epoch": 0.1,
"learning_rate": 0.0004549047141424273,
"loss": 3.1254,
"theoretical_loss": 4.10401167495222,
"tokens_seen": 327680000
},
{
"epoch": 0.1,
"learning_rate": 0.0004548946840521565,
"loss": 3.2898,
"theoretical_loss": 4.103916760758405,
"tokens_seen": 327745536
},
{
"epoch": 0.1,
"learning_rate": 0.00045488465396188564,
"loss": 3.0752,
"theoretical_loss": 4.103821870854656,
"tokens_seen": 327811072
},
{
"epoch": 0.1,
"learning_rate": 0.0004548746238716149,
"loss": 3.0744,
"theoretical_loss": 4.103727005229903,
"tokens_seen": 327876608
},
{
"epoch": 0.1,
"learning_rate": 0.00045486459378134405,
"loss": 3.0116,
"theoretical_loss": 4.103632163873083,
"tokens_seen": 327942144
},
{
"epoch": 0.1,
"learning_rate": 0.00045485456369107324,
"loss": 3.1184,
"theoretical_loss": 4.10353734677314,
"tokens_seen": 328007680
},
{
"epoch": 0.1,
"learning_rate": 0.0004548445336008024,
"loss": 3.1614,
"theoretical_loss": 4.103442553919026,
"tokens_seen": 328073216
},
{
"epoch": 0.1,
"learning_rate": 0.00045483450351053165,
"loss": 2.9932,
"theoretical_loss": 4.1033477852996985,
"tokens_seen": 328138752
},
{
"epoch": 0.1,
"learning_rate": 0.0004548244734202608,
"loss": 3.1685,
"theoretical_loss": 4.103253040904124,
"tokens_seen": 328204288
},
{
"epoch": 0.1,
"learning_rate": 0.00045481444332999,
"loss": 2.8453,
"theoretical_loss": 4.103158320721276,
"tokens_seen": 328269824
},
{
"epoch": 0.1,
"learning_rate": 0.00045480441323971914,
"loss": 3.1451,
"theoretical_loss": 4.103063624740133,
"tokens_seen": 328335360
},
{
"epoch": 0.1,
"learning_rate": 0.0004547943831494484,
"loss": 3.0218,
"theoretical_loss": 4.102968952949684,
"tokens_seen": 328400896
},
{
"epoch": 0.1,
"learning_rate": 0.00045478435305917756,
"loss": 2.9797,
"theoretical_loss": 4.102874305338923,
"tokens_seen": 328466432
},
{
"epoch": 0.1,
"learning_rate": 0.00045477432296890674,
"loss": 3.0406,
"theoretical_loss": 4.102779681896852,
"tokens_seen": 328531968
},
{
"epoch": 0.1,
"learning_rate": 0.0004547642928786359,
"loss": 2.8546,
"theoretical_loss": 4.10268508261248,
"tokens_seen": 328597504
},
{
"epoch": 0.1,
"learning_rate": 0.0004547542627883651,
"loss": 3.0949,
"theoretical_loss": 4.102590507474824,
"tokens_seen": 328663040
},
{
"epoch": 0.1,
"learning_rate": 0.0004547442326980943,
"loss": 3.256,
"theoretical_loss": 4.1024959564729055,
"tokens_seen": 328728576
},
{
"epoch": 0.1,
"learning_rate": 0.0004547342026078235,
"loss": 3.0471,
"theoretical_loss": 4.102401429595758,
"tokens_seen": 328794112
},
{
"epoch": 0.1,
"learning_rate": 0.00045472417251755264,
"loss": 2.9156,
"theoretical_loss": 4.102306926832417,
"tokens_seen": 328859648
},
{
"epoch": 0.1,
"learning_rate": 0.0004547141424272819,
"loss": 3.2931,
"theoretical_loss": 4.102212448171928,
"tokens_seen": 328925184
},
{
"epoch": 0.1,
"learning_rate": 0.000454704112337011,
"loss": 2.982,
"theoretical_loss": 4.1021179936033425,
"tokens_seen": 328990720
},
{
"epoch": 0.1,
"learning_rate": 0.00045469408224674024,
"loss": 3.0106,
"theoretical_loss": 4.102023563115721,
"tokens_seen": 329056256
},
{
"epoch": 0.1,
"learning_rate": 0.0004546840521564694,
"loss": 3.0727,
"theoretical_loss": 4.10192915669813,
"tokens_seen": 329121792
},
{
"epoch": 0.1,
"learning_rate": 0.0004546740220661986,
"loss": 3.1747,
"theoretical_loss": 4.1018347743396415,
"tokens_seen": 329187328
},
{
"epoch": 0.1,
"learning_rate": 0.0004546639919759278,
"loss": 3.1083,
"theoretical_loss": 4.101740416029338,
"tokens_seen": 329252864
},
{
"epoch": 0.1,
"objective/train/docs_used": 198138,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2101240158081055,
"objective/train/theoretical_loss": 4.101646081756305,
"objective/train/tokens_used": 349778400,
"theoretical_loss": 4.101646081756305,
"tokens_seen": 329318400
},
{
"epoch": 0.1,
"learning_rate": 0.000454653961885657,
"loss": 3.0282,
"theoretical_loss": 4.101646081756305,
"tokens_seen": 329318400
},
{
"epoch": 0.1,
"learning_rate": 0.00045464393179538615,
"loss": 2.993,
"theoretical_loss": 4.101551771509641,
"tokens_seen": 329383936
},
{
"epoch": 0.1,
"learning_rate": 0.0004546339017051154,
"loss": 3.2896,
"theoretical_loss": 4.101457485278444,
"tokens_seen": 329449472
},
{
"epoch": 0.1,
"learning_rate": 0.0004546238716148445,
"loss": 3.0261,
"theoretical_loss": 4.101363223051826,
"tokens_seen": 329515008
},
{
"epoch": 0.1,
"learning_rate": 0.00045461384152457374,
"loss": 3.0682,
"theoretical_loss": 4.101268984818901,
"tokens_seen": 329580544
},
{
"epoch": 0.1,
"learning_rate": 0.0004546038114343029,
"loss": 3.1339,
"theoretical_loss": 4.101174770568795,
"tokens_seen": 329646080
},
{
"epoch": 0.1,
"learning_rate": 0.0004545937813440321,
"loss": 3.0888,
"theoretical_loss": 4.1010805802906365,
"tokens_seen": 329711616
},
{
"epoch": 0.1,
"learning_rate": 0.0004545837512537613,
"loss": 3.0345,
"theoretical_loss": 4.100986413973564,
"tokens_seen": 329777152
},
{
"epoch": 0.1,
"learning_rate": 0.00045457372116349047,
"loss": 3.1313,
"theoretical_loss": 4.100892271606721,
"tokens_seen": 329842688
},
{
"epoch": 0.1,
"learning_rate": 0.00045456369107321965,
"loss": 2.8485,
"theoretical_loss": 4.1007981531792606,
"tokens_seen": 329908224
},
{
"epoch": 0.1,
"learning_rate": 0.0004545536609829489,
"loss": 3.19,
"theoretical_loss": 4.100704058680341,
"tokens_seen": 329973760
},
{
"epoch": 0.1,
"learning_rate": 0.000454543630892678,
"loss": 3.017,
"theoretical_loss": 4.1006099880991265,
"tokens_seen": 330039296
},
{
"epoch": 0.1,
"learning_rate": 0.00045453360080240725,
"loss": 2.9604,
"theoretical_loss": 4.100515941424792,
"tokens_seen": 330104832
},
{
"epoch": 0.1,
"learning_rate": 0.00045452357071213643,
"loss": 2.9596,
"theoretical_loss": 4.100421918646517,
"tokens_seen": 330170368
},
{
"epoch": 0.1,
"learning_rate": 0.0004545135406218656,
"loss": 3.1628,
"theoretical_loss": 4.1003279197534885,
"tokens_seen": 330235904
},
{
"epoch": 0.1,
"learning_rate": 0.0004545035105315948,
"loss": 2.9544,
"theoretical_loss": 4.100233944734899,
"tokens_seen": 330301440
},
{
"epoch": 0.1,
"learning_rate": 0.00045449348044132397,
"loss": 3.1729,
"theoretical_loss": 4.100139993579952,
"tokens_seen": 330366976
},
{
"epoch": 0.1,
"learning_rate": 0.00045448345035105315,
"loss": 2.7687,
"theoretical_loss": 4.100046066277853,
"tokens_seen": 330432512
},
{
"epoch": 0.1,
"learning_rate": 0.0004544734202607824,
"loss": 3.2236,
"theoretical_loss": 4.09995216281782,
"tokens_seen": 330498048
},
{
"epoch": 0.1,
"learning_rate": 0.0004544633901705115,
"loss": 3.1535,
"theoretical_loss": 4.0998582831890715,
"tokens_seen": 330563584
},
{
"epoch": 0.1,
"learning_rate": 0.00045445336008024075,
"loss": 3.11,
"theoretical_loss": 4.0997644273808405,
"tokens_seen": 330629120
},
{
"epoch": 0.1,
"learning_rate": 0.0004544433299899699,
"loss": 2.8457,
"theoretical_loss": 4.09967059538236,
"tokens_seen": 330694656
},
{
"epoch": 0.1,
"learning_rate": 0.0004544332998996991,
"loss": 2.9948,
"theoretical_loss": 4.099576787182874,
"tokens_seen": 330760192
},
{
"epoch": 0.1,
"learning_rate": 0.0004544232698094283,
"loss": 3.1013,
"theoretical_loss": 4.099483002771633,
"tokens_seen": 330825728
},
{
"epoch": 0.1,
"learning_rate": 0.0004544132397191575,
"loss": 2.9936,
"theoretical_loss": 4.099389242137894,
"tokens_seen": 330891264
},
{
"epoch": 0.1,
"objective/train/docs_used": 199416,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0712082386016846,
"objective/train/theoretical_loss": 4.099295505270921,
"objective/train/tokens_used": 351416800,
"theoretical_loss": 4.099295505270921,
"tokens_seen": 330956800
},
{
"epoch": 0.1,
"learning_rate": 0.00045440320962888666,
"loss": 3.0119,
"theoretical_loss": 4.099295505270921,
"tokens_seen": 330956800
},
{
"epoch": 0.1,
"learning_rate": 0.00045439317953861584,
"loss": 2.9404,
"theoretical_loss": 4.099201792159985,
"tokens_seen": 331022336
},
{
"epoch": 0.1,
"learning_rate": 0.000454383149448345,
"loss": 2.9773,
"theoretical_loss": 4.099108102794363,
"tokens_seen": 331087872
},
{
"epoch": 0.1,
"learning_rate": 0.00045437311935807425,
"loss": 3.1837,
"theoretical_loss": 4.099014437163342,
"tokens_seen": 331153408
},
{
"epoch": 0.1,
"learning_rate": 0.0004543630892678034,
"loss": 2.9231,
"theoretical_loss": 4.098920795256213,
"tokens_seen": 331218944
},
{
"epoch": 0.1,
"learning_rate": 0.0004543530591775326,
"loss": 2.9996,
"theoretical_loss": 4.098827177062273,
"tokens_seen": 331284480
},
{
"epoch": 0.1,
"learning_rate": 0.0004543430290872618,
"loss": 2.9186,
"theoretical_loss": 4.098733582570831,
"tokens_seen": 331350016
},
{
"epoch": 0.1,
"learning_rate": 0.000454332998996991,
"loss": 3.0911,
"theoretical_loss": 4.098640011771198,
"tokens_seen": 331415552
},
{
"epoch": 0.1,
"learning_rate": 0.00045432296890672016,
"loss": 2.9677,
"theoretical_loss": 4.098546464652693,
"tokens_seen": 331481088
},
{
"epoch": 0.1,
"learning_rate": 0.00045431293881644934,
"loss": 2.8865,
"theoretical_loss": 4.098452941204643,
"tokens_seen": 331546624
},
{
"epoch": 0.1,
"learning_rate": 0.0004543029087261785,
"loss": 2.9476,
"theoretical_loss": 4.098359441416383,
"tokens_seen": 331612160
},
{
"epoch": 0.1,
"learning_rate": 0.00045429287863590776,
"loss": 2.9588,
"theoretical_loss": 4.0982659652772515,
"tokens_seen": 331677696
},
{
"epoch": 0.1,
"learning_rate": 0.0004542828485456369,
"loss": 3.1995,
"theoretical_loss": 4.098172512776597,
"tokens_seen": 331743232
},
{
"epoch": 0.1,
"learning_rate": 0.0004542728184553661,
"loss": 2.9885,
"theoretical_loss": 4.098079083903773,
"tokens_seen": 331808768
},
{
"epoch": 0.1,
"learning_rate": 0.00045426278836509525,
"loss": 3.1498,
"theoretical_loss": 4.097985678648142,
"tokens_seen": 331874304
},
{
"epoch": 0.1,
"learning_rate": 0.0004542527582748245,
"loss": 2.948,
"theoretical_loss": 4.09789229699907,
"tokens_seen": 331939840
},
{
"epoch": 0.1,
"learning_rate": 0.00045424272818455366,
"loss": 3.1034,
"theoretical_loss": 4.097798938945933,
"tokens_seen": 332005376
},
{
"epoch": 0.1,
"learning_rate": 0.00045423269809428284,
"loss": 3.143,
"theoretical_loss": 4.097705604478112,
"tokens_seen": 332070912
},
{
"epoch": 0.1,
"learning_rate": 0.000454222668004012,
"loss": 2.8918,
"theoretical_loss": 4.097612293584998,
"tokens_seen": 332136448
},
{
"epoch": 0.1,
"learning_rate": 0.0004542126379137412,
"loss": 2.817,
"theoretical_loss": 4.0975190062559825,
"tokens_seen": 332201984
},
{
"epoch": 0.1,
"learning_rate": 0.00045420260782347044,
"loss": 2.9891,
"theoretical_loss": 4.097425742480472,
"tokens_seen": 332267520
},
{
"epoch": 0.1,
"learning_rate": 0.0004541925777331996,
"loss": 2.9818,
"theoretical_loss": 4.097332502247873,
"tokens_seen": 332333056
},
{
"epoch": 0.1,
"learning_rate": 0.0004541825476429288,
"loss": 3.0395,
"theoretical_loss": 4.0972392855476025,
"tokens_seen": 332398592
},
{
"epoch": 0.1,
"learning_rate": 0.000454172517552658,
"loss": 3.127,
"theoretical_loss": 4.097146092369084,
"tokens_seen": 332464128
},
{
"epoch": 0.1,
"learning_rate": 0.0004541624874623872,
"loss": 2.8874,
"theoretical_loss": 4.097052922701746,
"tokens_seen": 332529664
},
{
"epoch": 0.1,
"objective/train/docs_used": 200524,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4482038021087646,
"objective/train/theoretical_loss": 4.0969597765350265,
"objective/train/tokens_used": 353055200,
"theoretical_loss": 4.0969597765350265,
"tokens_seen": 332595200
},
{
"epoch": 0.1,
"learning_rate": 0.00045415245737211635,
"loss": 2.9907,
"theoretical_loss": 4.0969597765350265,
"tokens_seen": 332595200
},
{
"epoch": 0.1,
"learning_rate": 0.0004541424272818456,
"loss": 3.0728,
"theoretical_loss": 4.096866653858368,
"tokens_seen": 332660736
},
{
"epoch": 0.1,
"learning_rate": 0.0004541323971915747,
"loss": 2.9973,
"theoretical_loss": 4.09677355466122,
"tokens_seen": 332726272
},
{
"epoch": 0.1,
"learning_rate": 0.00045412236710130395,
"loss": 2.846,
"theoretical_loss": 4.096680478933042,
"tokens_seen": 332791808
},
{
"epoch": 0.1,
"learning_rate": 0.0004541123370110331,
"loss": 2.9862,
"theoretical_loss": 4.0965874266632945,
"tokens_seen": 332857344
},
{
"epoch": 0.1,
"learning_rate": 0.0004541023069207623,
"loss": 3.0369,
"theoretical_loss": 4.09649439784145,
"tokens_seen": 332922880
},
{
"epoch": 0.1,
"learning_rate": 0.0004540922768304915,
"loss": 3.1214,
"theoretical_loss": 4.096401392456988,
"tokens_seen": 332988416
},
{
"epoch": 0.1,
"learning_rate": 0.00045408224674022067,
"loss": 3.1405,
"theoretical_loss": 4.09630841049939,
"tokens_seen": 333053952
},
{
"epoch": 0.1,
"learning_rate": 0.00045407221664994985,
"loss": 2.9794,
"theoretical_loss": 4.096215451958146,
"tokens_seen": 333119488
},
{
"epoch": 0.1,
"learning_rate": 0.0004540621865596791,
"loss": 3.2064,
"theoretical_loss": 4.096122516822757,
"tokens_seen": 333185024
},
{
"epoch": 0.1,
"learning_rate": 0.0004540521564694082,
"loss": 3.0618,
"theoretical_loss": 4.096029605082726,
"tokens_seen": 333250560
},
{
"epoch": 0.1,
"learning_rate": 0.00045404212637913745,
"loss": 3.0452,
"theoretical_loss": 4.095936716727564,
"tokens_seen": 333316096
},
{
"epoch": 0.1,
"learning_rate": 0.00045403209628886663,
"loss": 3.0753,
"theoretical_loss": 4.095843851746791,
"tokens_seen": 333381632
},
{
"epoch": 0.1,
"learning_rate": 0.0004540220661985958,
"loss": 3.0818,
"theoretical_loss": 4.095751010129929,
"tokens_seen": 333447168
},
{
"epoch": 0.1,
"learning_rate": 0.000454012036108325,
"loss": 3.178,
"theoretical_loss": 4.095658191866512,
"tokens_seen": 333512704
},
{
"epoch": 0.1,
"learning_rate": 0.00045400200601805417,
"loss": 3.0465,
"theoretical_loss": 4.0955653969460775,
"tokens_seen": 333578240
},
{
"epoch": 0.1,
"learning_rate": 0.00045399197592778335,
"loss": 2.8333,
"theoretical_loss": 4.095472625358171,
"tokens_seen": 333643776
},
{
"epoch": 0.1,
"learning_rate": 0.0004539819458375126,
"loss": 3.3654,
"theoretical_loss": 4.095379877092343,
"tokens_seen": 333709312
},
{
"epoch": 0.1,
"learning_rate": 0.0004539719157472417,
"loss": 2.9324,
"theoretical_loss": 4.095287152138154,
"tokens_seen": 333774848
},
{
"epoch": 0.1,
"learning_rate": 0.00045396188565697095,
"loss": 3.1113,
"theoretical_loss": 4.0951944504851685,
"tokens_seen": 333840384
},
{
"epoch": 0.1,
"learning_rate": 0.0004539518555667001,
"loss": 3.1439,
"theoretical_loss": 4.095101772122959,
"tokens_seen": 333905920
},
{
"epoch": 0.1,
"learning_rate": 0.0004539418254764293,
"loss": 3.2737,
"theoretical_loss": 4.095009117041102,
"tokens_seen": 333971456
},
{
"epoch": 0.1,
"learning_rate": 0.0004539317953861585,
"loss": 3.0585,
"theoretical_loss": 4.094916485229186,
"tokens_seen": 334036992
},
{
"epoch": 0.1,
"learning_rate": 0.0004539217652958877,
"loss": 2.8934,
"theoretical_loss": 4.094823876676802,
"tokens_seen": 334102528
},
{
"epoch": 0.1,
"learning_rate": 0.00045391173520561686,
"loss": 2.922,
"theoretical_loss": 4.094731291373548,
"tokens_seen": 334168064
},
{
"epoch": 0.1,
"objective/train/docs_used": 201196,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.268803834915161,
"objective/train/theoretical_loss": 4.094638729309031,
"objective/train/tokens_used": 354693600,
"theoretical_loss": 4.094638729309031,
"tokens_seen": 334233600
},
{
"epoch": 0.1,
"learning_rate": 0.00045390170511534604,
"loss": 3.0566,
"theoretical_loss": 4.094638729309031,
"tokens_seen": 334233600
},
{
"epoch": 0.1,
"learning_rate": 0.0004538916750250752,
"loss": 2.9384,
"theoretical_loss": 4.094546190472862,
"tokens_seen": 334299136
},
{
"epoch": 0.1,
"learning_rate": 0.00045388164493480445,
"loss": 2.7438,
"theoretical_loss": 4.09445367485466,
"tokens_seen": 334364672
},
{
"epoch": 0.1,
"learning_rate": 0.0004538716148445336,
"loss": 3.1787,
"theoretical_loss": 4.09436118244405,
"tokens_seen": 334430208
},
{
"epoch": 0.1,
"learning_rate": 0.0004538615847542628,
"loss": 3.0136,
"theoretical_loss": 4.094268713230667,
"tokens_seen": 334495744
},
{
"epoch": 0.1,
"learning_rate": 0.000453851554663992,
"loss": 3.0249,
"theoretical_loss": 4.094176267204148,
"tokens_seen": 334561280
},
{
"epoch": 0.1,
"learning_rate": 0.0004538415245737212,
"loss": 3.3932,
"theoretical_loss": 4.094083844354137,
"tokens_seen": 334626816
},
{
"epoch": 0.1,
"learning_rate": 0.00045383149448345036,
"loss": 3.2013,
"theoretical_loss": 4.093991444670289,
"tokens_seen": 334692352
},
{
"epoch": 0.1,
"learning_rate": 0.00045382146439317954,
"loss": 2.982,
"theoretical_loss": 4.093899068142262,
"tokens_seen": 334757888
},
{
"epoch": 0.1,
"learning_rate": 0.0004538114343029087,
"loss": 3.1016,
"theoretical_loss": 4.093806714759721,
"tokens_seen": 334823424
},
{
"epoch": 0.1,
"learning_rate": 0.00045380140421263796,
"loss": 3.2014,
"theoretical_loss": 4.093714384512337,
"tokens_seen": 334888960
},
{
"epoch": 0.1,
"learning_rate": 0.0004537913741223671,
"loss": 3.0236,
"theoretical_loss": 4.093622077389791,
"tokens_seen": 334954496
},
{
"epoch": 0.1,
"learning_rate": 0.0004537813440320963,
"loss": 2.9487,
"theoretical_loss": 4.093529793381768,
"tokens_seen": 335020032
},
{
"epoch": 0.1,
"learning_rate": 0.00045377131394182545,
"loss": 3.2927,
"theoretical_loss": 4.093437532477958,
"tokens_seen": 335085568
},
{
"epoch": 0.1,
"learning_rate": 0.0004537612838515547,
"loss": 3.111,
"theoretical_loss": 4.093345294668063,
"tokens_seen": 335151104
},
{
"epoch": 0.1,
"learning_rate": 0.00045375125376128386,
"loss": 3.19,
"theoretical_loss": 4.0932530799417846,
"tokens_seen": 335216640
},
{
"epoch": 0.1,
"learning_rate": 0.00045374122367101304,
"loss": 3.2401,
"theoretical_loss": 4.093160888288837,
"tokens_seen": 335282176
},
{
"epoch": 0.1,
"learning_rate": 0.0004537311935807422,
"loss": 3.1176,
"theoretical_loss": 4.0930687196989375,
"tokens_seen": 335347712
},
{
"epoch": 0.1,
"learning_rate": 0.0004537211634904714,
"loss": 3.0444,
"theoretical_loss": 4.0929765741618125,
"tokens_seen": 335413248
},
{
"epoch": 0.1,
"learning_rate": 0.0004537111334002006,
"loss": 3.046,
"theoretical_loss": 4.092884451667191,
"tokens_seen": 335478784
},
{
"epoch": 0.1,
"learning_rate": 0.0004537011033099298,
"loss": 3.1603,
"theoretical_loss": 4.092792352204814,
"tokens_seen": 335544320
},
{
"epoch": 0.1,
"learning_rate": 0.00045369107321965895,
"loss": 3.0195,
"theoretical_loss": 4.092700275764424,
"tokens_seen": 335609856
},
{
"epoch": 0.1,
"learning_rate": 0.0004536810431293882,
"loss": 3.1597,
"theoretical_loss": 4.092608222335774,
"tokens_seen": 335675392
},
{
"epoch": 0.1,
"learning_rate": 0.00045367101303911737,
"loss": 2.9772,
"theoretical_loss": 4.092516191908621,
"tokens_seen": 335740928
},
{
"epoch": 0.1,
"learning_rate": 0.00045366098294884655,
"loss": 2.8325,
"theoretical_loss": 4.09242418447273,
"tokens_seen": 335806464
},
{
"epoch": 0.1,
"objective/train/docs_used": 202460,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.944324254989624,
"objective/train/theoretical_loss": 4.092332200017871,
"objective/train/tokens_used": 356332000,
"theoretical_loss": 4.092332200017871,
"tokens_seen": 335872000
},
{
"epoch": 0.1,
"learning_rate": 0.00045365095285857573,
"loss": 2.9899,
"theoretical_loss": 4.092332200017871,
"tokens_seen": 335872000
},
{
"epoch": 0.1,
"learning_rate": 0.0004536409227683049,
"loss": 3.0494,
"theoretical_loss": 4.092240238533822,
"tokens_seen": 335937536
},
{
"epoch": 0.1,
"learning_rate": 0.0004536308926780341,
"loss": 2.8351,
"theoretical_loss": 4.092148300010367,
"tokens_seen": 336003072
},
{
"epoch": 0.1,
"learning_rate": 0.0004536208625877633,
"loss": 2.9533,
"theoretical_loss": 4.092056384437297,
"tokens_seen": 336068608
},
{
"epoch": 0.1,
"learning_rate": 0.00045361083249749245,
"loss": 2.9323,
"theoretical_loss": 4.091964491804409,
"tokens_seen": 336134144
},
{
"epoch": 0.1,
"learning_rate": 0.0004536008024072217,
"loss": 2.9688,
"theoretical_loss": 4.091872622101506,
"tokens_seen": 336199680
},
{
"epoch": 0.1,
"learning_rate": 0.0004535907723169508,
"loss": 3.0854,
"theoretical_loss": 4.091780775318399,
"tokens_seen": 336265216
},
{
"epoch": 0.1,
"learning_rate": 0.00045358074222668005,
"loss": 2.9489,
"theoretical_loss": 4.091688951444904,
"tokens_seen": 336330752
},
{
"epoch": 0.1,
"learning_rate": 0.00045357071213640923,
"loss": 2.8713,
"theoretical_loss": 4.091597150470845,
"tokens_seen": 336396288
},
{
"epoch": 0.1,
"learning_rate": 0.0004535606820461384,
"loss": 3.1211,
"theoretical_loss": 4.091505372386051,
"tokens_seen": 336461824
},
{
"epoch": 0.1,
"learning_rate": 0.0004535506519558676,
"loss": 3.0486,
"theoretical_loss": 4.091413617180358,
"tokens_seen": 336527360
},
{
"epoch": 0.1,
"learning_rate": 0.00045354062186559683,
"loss": 3.3574,
"theoretical_loss": 4.091321884843609,
"tokens_seen": 336592896
},
{
"epoch": 0.1,
"learning_rate": 0.00045353059177532596,
"loss": 3.184,
"theoretical_loss": 4.091230175365653,
"tokens_seen": 336658432
},
{
"epoch": 0.1,
"learning_rate": 0.0004535205616850552,
"loss": 3.0774,
"theoretical_loss": 4.0911384887363464,
"tokens_seen": 336723968
},
{
"epoch": 0.1,
"learning_rate": 0.0004535105315947843,
"loss": 3.1025,
"theoretical_loss": 4.091046824945551,
"tokens_seen": 336789504
},
{
"epoch": 0.1,
"learning_rate": 0.00045350050150451355,
"loss": 3.087,
"theoretical_loss": 4.0909551839831355,
"tokens_seen": 336855040
},
{
"epoch": 0.1,
"learning_rate": 0.00045349047141424274,
"loss": 3.155,
"theoretical_loss": 4.090863565838974,
"tokens_seen": 336920576
},
{
"epoch": 0.1,
"learning_rate": 0.0004534804413239719,
"loss": 2.8842,
"theoretical_loss": 4.090771970502948,
"tokens_seen": 336986112
},
{
"epoch": 0.1,
"learning_rate": 0.0004534704112337011,
"loss": 2.9869,
"theoretical_loss": 4.090680397964947,
"tokens_seen": 337051648
},
{
"epoch": 0.1,
"learning_rate": 0.0004534603811434303,
"loss": 3.1412,
"theoretical_loss": 4.090588848214865,
"tokens_seen": 337117184
},
{
"epoch": 0.1,
"learning_rate": 0.0004534503510531595,
"loss": 2.9635,
"theoretical_loss": 4.0904973212426015,
"tokens_seen": 337182720
},
{
"epoch": 0.1,
"learning_rate": 0.0004534403209628887,
"loss": 3.0079,
"theoretical_loss": 4.090405817038065,
"tokens_seen": 337248256
},
{
"epoch": 0.1,
"learning_rate": 0.0004534302908726179,
"loss": 2.8153,
"theoretical_loss": 4.090314335591169,
"tokens_seen": 337313792
},
{
"epoch": 0.1,
"learning_rate": 0.00045342026078234706,
"loss": 2.8761,
"theoretical_loss": 4.090222876891834,
"tokens_seen": 337379328
},
{
"epoch": 0.1,
"learning_rate": 0.00045341023069207624,
"loss": 3.2322,
"theoretical_loss": 4.090131440929985,
"tokens_seen": 337444864
},
{
"epoch": 0.1,
"objective/train/docs_used": 203033,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.978555679321289,
"objective/train/theoretical_loss": 4.090040027695556,
"objective/train/tokens_used": 357970400,
"theoretical_loss": 4.090040027695556,
"tokens_seen": 337510400
},
{
"epoch": 0.1,
"learning_rate": 0.0004534002006018054,
"loss": 3.1313,
"theoretical_loss": 4.090040027695556,
"tokens_seen": 337510400
},
{
"epoch": 0.1,
"learning_rate": 0.00045339017051153465,
"loss": 3.1773,
"theoretical_loss": 4.0899486371784874,
"tokens_seen": 337575936
},
{
"epoch": 0.1,
"learning_rate": 0.0004533801404212638,
"loss": 2.9796,
"theoretical_loss": 4.089857269368725,
"tokens_seen": 337641472
},
{
"epoch": 0.1,
"learning_rate": 0.000453370110330993,
"loss": 3.089,
"theoretical_loss": 4.089765924256218,
"tokens_seen": 337707008
},
{
"epoch": 0.1,
"learning_rate": 0.0004533600802407222,
"loss": 2.932,
"theoretical_loss": 4.089674601830929,
"tokens_seen": 337772544
},
{
"epoch": 0.1,
"learning_rate": 0.0004533500501504514,
"loss": 2.9389,
"theoretical_loss": 4.08958330208282,
"tokens_seen": 337838080
},
{
"epoch": 0.1,
"learning_rate": 0.00045334002006018056,
"loss": 3.0489,
"theoretical_loss": 4.089492025001864,
"tokens_seen": 337903616
},
{
"epoch": 0.1,
"learning_rate": 0.00045332998996990974,
"loss": 3.142,
"theoretical_loss": 4.089400770578038,
"tokens_seen": 337969152
},
{
"epoch": 0.1,
"learning_rate": 0.0004533199598796389,
"loss": 2.7704,
"theoretical_loss": 4.089309538801327,
"tokens_seen": 338034688
},
{
"epoch": 0.1,
"learning_rate": 0.00045330992978936816,
"loss": 3.0028,
"theoretical_loss": 4.08921832966172,
"tokens_seen": 338100224
},
{
"epoch": 0.1,
"learning_rate": 0.0004532998996990973,
"loss": 3.2315,
"theoretical_loss": 4.0891271431492155,
"tokens_seen": 338165760
},
{
"epoch": 0.1,
"learning_rate": 0.0004532898696088265,
"loss": 3.1826,
"theoretical_loss": 4.089035979253816,
"tokens_seen": 338231296
},
{
"epoch": 0.1,
"learning_rate": 0.00045327983951855565,
"loss": 3.1003,
"theoretical_loss": 4.08894483796553,
"tokens_seen": 338296832
},
{
"epoch": 0.1,
"learning_rate": 0.0004532698094282849,
"loss": 3.1606,
"theoretical_loss": 4.0888537192743755,
"tokens_seen": 338362368
},
{
"epoch": 0.1,
"learning_rate": 0.00045325977933801406,
"loss": 2.9502,
"theoretical_loss": 4.088762623170373,
"tokens_seen": 338427904
},
{
"epoch": 0.1,
"learning_rate": 0.00045324974924774324,
"loss": 3.2166,
"theoretical_loss": 4.088671549643553,
"tokens_seen": 338493440
},
{
"epoch": 0.1,
"learning_rate": 0.0004532397191574724,
"loss": 2.9103,
"theoretical_loss": 4.088580498683948,
"tokens_seen": 338558976
},
{
"epoch": 0.1,
"learning_rate": 0.0004532296890672016,
"loss": 3.3325,
"theoretical_loss": 4.088489470281601,
"tokens_seen": 338624512
},
{
"epoch": 0.1,
"learning_rate": 0.0004532196589769308,
"loss": 2.8166,
"theoretical_loss": 4.088398464426559,
"tokens_seen": 338690048
},
{
"epoch": 0.1,
"learning_rate": 0.00045320962888666,
"loss": 3.1542,
"theoretical_loss": 4.088307481108876,
"tokens_seen": 338755584
},
{
"epoch": 0.1,
"learning_rate": 0.00045319959879638915,
"loss": 3.0656,
"theoretical_loss": 4.088216520318612,
"tokens_seen": 338821120
},
{
"epoch": 0.1,
"learning_rate": 0.0004531895687061184,
"loss": 2.9526,
"theoretical_loss": 4.0881255820458335,
"tokens_seen": 338886656
},
{
"epoch": 0.1,
"learning_rate": 0.00045317953861584757,
"loss": 3.1781,
"theoretical_loss": 4.088034666280614,
"tokens_seen": 338952192
},
{
"epoch": 0.1,
"learning_rate": 0.00045316950852557675,
"loss": 2.8558,
"theoretical_loss": 4.087943773013032,
"tokens_seen": 339017728
},
{
"epoch": 0.1,
"learning_rate": 0.00045315947843530593,
"loss": 3.1344,
"theoretical_loss": 4.087852902233173,
"tokens_seen": 339083264
},
{
"epoch": 0.1,
"objective/train/docs_used": 204232,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.195103406906128,
"objective/train/theoretical_loss": 4.087762053931129,
"objective/train/tokens_used": 359608800,
"theoretical_loss": 4.087762053931129,
"tokens_seen": 339148800
},
{
"epoch": 0.1,
"learning_rate": 0.0004531494483450351,
"loss": 3.0253,
"theoretical_loss": 4.087762053931129,
"tokens_seen": 339148800
},
{
"epoch": 0.1,
"learning_rate": 0.0004531394182547643,
"loss": 2.7862,
"theoretical_loss": 4.087671228096997,
"tokens_seen": 339214336
},
{
"epoch": 0.1,
"learning_rate": 0.0004531293881644935,
"loss": 3.0036,
"theoretical_loss": 4.087580424720882,
"tokens_seen": 339279872
},
{
"epoch": 0.1,
"learning_rate": 0.00045311935807422265,
"loss": 3.0948,
"theoretical_loss": 4.087489643792894,
"tokens_seen": 339345408
},
{
"epoch": 0.1,
"learning_rate": 0.0004531093279839519,
"loss": 3.1925,
"theoretical_loss": 4.08739888530315,
"tokens_seen": 339410944
},
{
"epoch": 0.1,
"learning_rate": 0.000453099297893681,
"loss": 3.2017,
"theoretical_loss": 4.087308149241774,
"tokens_seen": 339476480
},
{
"epoch": 0.1,
"learning_rate": 0.00045308926780341025,
"loss": 3.164,
"theoretical_loss": 4.087217435598894,
"tokens_seen": 339542016
},
{
"epoch": 0.1,
"learning_rate": 0.00045307923771313943,
"loss": 3.0525,
"theoretical_loss": 4.087126744364646,
"tokens_seen": 339607552
},
{
"epoch": 0.1,
"learning_rate": 0.0004530692076228686,
"loss": 2.9962,
"theoretical_loss": 4.087036075529172,
"tokens_seen": 339673088
},
{
"epoch": 0.1,
"learning_rate": 0.0004530591775325978,
"loss": 3.0656,
"theoretical_loss": 4.086945429082618,
"tokens_seen": 339738624
},
{
"epoch": 0.1,
"learning_rate": 0.00045304914744232703,
"loss": 3.1211,
"theoretical_loss": 4.086854805015141,
"tokens_seen": 339804160
},
{
"epoch": 0.1,
"learning_rate": 0.00045303911735205616,
"loss": 3.3196,
"theoretical_loss": 4.086764203316902,
"tokens_seen": 339869696
},
{
"epoch": 0.1,
"learning_rate": 0.0004530290872617854,
"loss": 2.9489,
"theoretical_loss": 4.086673623978064,
"tokens_seen": 339935232
},
{
"epoch": 0.1,
"learning_rate": 0.0004530190571715145,
"loss": 3.0491,
"theoretical_loss": 4.086583066988802,
"tokens_seen": 340000768
},
{
"epoch": 0.1,
"learning_rate": 0.00045300902708124375,
"loss": 2.9703,
"theoretical_loss": 4.086492532339296,
"tokens_seen": 340066304
},
{
"epoch": 0.1,
"learning_rate": 0.00045299899699097294,
"loss": 2.8628,
"theoretical_loss": 4.0864020200197295,
"tokens_seen": 340131840
},
{
"epoch": 0.1,
"learning_rate": 0.0004529889669007021,
"loss": 3.0217,
"theoretical_loss": 4.086311530020296,
"tokens_seen": 340197376
},
{
"epoch": 0.1,
"learning_rate": 0.0004529789368104313,
"loss": 3.2726,
"theoretical_loss": 4.086221062331192,
"tokens_seen": 340262912
},
{
"epoch": 0.1,
"learning_rate": 0.0004529689067201605,
"loss": 2.7499,
"theoretical_loss": 4.086130616942621,
"tokens_seen": 340328448
},
{
"epoch": 0.1,
"learning_rate": 0.00045295887662988966,
"loss": 3.0251,
"theoretical_loss": 4.086040193844794,
"tokens_seen": 340393984
},
{
"epoch": 0.1,
"learning_rate": 0.0004529488465396189,
"loss": 2.8855,
"theoretical_loss": 4.085949793027927,
"tokens_seen": 340459520
},
{
"epoch": 0.1,
"learning_rate": 0.000452938816449348,
"loss": 3.3201,
"theoretical_loss": 4.0858594144822415,
"tokens_seen": 340525056
},
{
"epoch": 0.1,
"learning_rate": 0.00045292878635907726,
"loss": 3.067,
"theoretical_loss": 4.085769058197968,
"tokens_seen": 340590592
},
{
"epoch": 0.1,
"learning_rate": 0.0004529187562688064,
"loss": 3.0007,
"theoretical_loss": 4.085678724165341,
"tokens_seen": 340656128
},
{
"epoch": 0.1,
"learning_rate": 0.0004529087261785356,
"loss": 2.9862,
"theoretical_loss": 4.0855884123746,
"tokens_seen": 340721664
},
{
"epoch": 0.1,
"objective/train/docs_used": 205032,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1145529747009277,
"objective/train/theoretical_loss": 4.085498122815992,
"objective/train/tokens_used": 361247200,
"theoretical_loss": 4.085498122815992,
"tokens_seen": 340787200
},
{
"epoch": 0.1,
"learning_rate": 0.0004528986960882648,
"loss": 2.8576,
"theoretical_loss": 4.085498122815992,
"tokens_seen": 340787200
},
{
"epoch": 0.1,
"learning_rate": 0.000452888665997994,
"loss": 2.925,
"theoretical_loss": 4.085407855479772,
"tokens_seen": 340852736
},
{
"epoch": 0.1,
"learning_rate": 0.00045287863590772316,
"loss": 3.0175,
"theoretical_loss": 4.085317610356199,
"tokens_seen": 340918272
},
{
"epoch": 0.1,
"learning_rate": 0.0004528686058174524,
"loss": 3.0725,
"theoretical_loss": 4.085227387435538,
"tokens_seen": 340983808
},
{
"epoch": 0.1,
"learning_rate": 0.0004528585757271815,
"loss": 3.0565,
"theoretical_loss": 4.08513718670806,
"tokens_seen": 341049344
},
{
"epoch": 0.1,
"learning_rate": 0.00045284854563691076,
"loss": 2.9061,
"theoretical_loss": 4.085047008164044,
"tokens_seen": 341114880
},
{
"epoch": 0.1,
"learning_rate": 0.0004528385155466399,
"loss": 3.0068,
"theoretical_loss": 4.084956851793773,
"tokens_seen": 341180416
},
{
"epoch": 0.1,
"learning_rate": 0.0004528284854563691,
"loss": 2.9558,
"theoretical_loss": 4.0848667175875395,
"tokens_seen": 341245952
},
{
"epoch": 0.1,
"learning_rate": 0.0004528184553660983,
"loss": 3.1372,
"theoretical_loss": 4.0847766055356365,
"tokens_seen": 341311488
},
{
"epoch": 0.1,
"learning_rate": 0.0004528084252758275,
"loss": 2.8882,
"theoretical_loss": 4.084686515628368,
"tokens_seen": 341377024
},
{
"epoch": 0.1,
"learning_rate": 0.00045279839518555667,
"loss": 3.1112,
"theoretical_loss": 4.0845964478560415,
"tokens_seen": 341442560
},
{
"epoch": 0.1,
"learning_rate": 0.00045278836509528585,
"loss": 3.0737,
"theoretical_loss": 4.084506402208972,
"tokens_seen": 341508096
},
{
"epoch": 0.1,
"learning_rate": 0.00045277833500501503,
"loss": 3.0444,
"theoretical_loss": 4.0844163786774805,
"tokens_seen": 341573632
},
{
"epoch": 0.1,
"learning_rate": 0.00045276830491474426,
"loss": 3.1,
"theoretical_loss": 4.084326377251894,
"tokens_seen": 341639168
},
{
"epoch": 0.1,
"learning_rate": 0.0004527582748244734,
"loss": 2.8045,
"theoretical_loss": 4.084236397922544,
"tokens_seen": 341704704
},
{
"epoch": 0.1,
"learning_rate": 0.0004527482447342026,
"loss": 2.7967,
"theoretical_loss": 4.084146440679769,
"tokens_seen": 341770240
},
{
"epoch": 0.1,
"learning_rate": 0.00045273821464393175,
"loss": 3.1078,
"theoretical_loss": 4.0840565055139155,
"tokens_seen": 341835776
},
{
"epoch": 0.1,
"learning_rate": 0.000452728184553661,
"loss": 3.0273,
"theoretical_loss": 4.083966592415333,
"tokens_seen": 341901312
},
{
"epoch": 0.1,
"learning_rate": 0.00045271815446339017,
"loss": 3.0421,
"theoretical_loss": 4.083876701374381,
"tokens_seen": 341966848
},
{
"epoch": 0.1,
"learning_rate": 0.00045270812437311935,
"loss": 3.1222,
"theoretical_loss": 4.08378683238142,
"tokens_seen": 342032384
},
{
"epoch": 0.1,
"learning_rate": 0.0004526980942828486,
"loss": 3.0408,
"theoretical_loss": 4.08369698542682,
"tokens_seen": 342097920
},
{
"epoch": 0.1,
"learning_rate": 0.00045268806419257777,
"loss": 3.1872,
"theoretical_loss": 4.083607160500957,
"tokens_seen": 342163456
},
{
"epoch": 0.1,
"learning_rate": 0.00045267803410230695,
"loss": 3.1052,
"theoretical_loss": 4.083517357594212,
"tokens_seen": 342228992
},
{
"epoch": 0.1,
"learning_rate": 0.00045266800401203613,
"loss": 2.7251,
"theoretical_loss": 4.0834275766969705,
"tokens_seen": 342294528
},
{
"epoch": 0.1,
"learning_rate": 0.0004526579739217653,
"loss": 3.1931,
"theoretical_loss": 4.083337817799629,
"tokens_seen": 342360064
},
{
"epoch": 0.1,
"objective/train/docs_used": 206109,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.671995162963867,
"objective/train/theoretical_loss": 4.083248080892584,
"objective/train/tokens_used": 362885600,
"theoretical_loss": 4.083248080892584,
"tokens_seen": 342425600
},
{
"epoch": 0.1,
"learning_rate": 0.0004526479438314945,
"loss": 2.9995,
"theoretical_loss": 4.083248080892584,
"tokens_seen": 342425600
},
{
"epoch": 0.1,
"learning_rate": 0.0004526379137412237,
"loss": 2.9213,
"theoretical_loss": 4.083158365966243,
"tokens_seen": 342491136
},
{
"epoch": 0.1,
"learning_rate": 0.00045262788365095285,
"loss": 2.968,
"theoretical_loss": 4.083068673011017,
"tokens_seen": 342556672
},
{
"epoch": 0.1,
"learning_rate": 0.0004526178535606821,
"loss": 3.0172,
"theoretical_loss": 4.082979002017321,
"tokens_seen": 342622208
},
{
"epoch": 0.1,
"learning_rate": 0.0004526078234704112,
"loss": 2.7793,
"theoretical_loss": 4.082889352975583,
"tokens_seen": 342687744
},
{
"epoch": 0.1,
"learning_rate": 0.00045259779338014045,
"loss": 3.0194,
"theoretical_loss": 4.082799725876228,
"tokens_seen": 342753280
},
{
"epoch": 0.1,
"learning_rate": 0.00045258776328986963,
"loss": 2.8469,
"theoretical_loss": 4.082710120709694,
"tokens_seen": 342818816
},
{
"epoch": 0.1,
"learning_rate": 0.0004525777331995988,
"loss": 3.0244,
"theoretical_loss": 4.082620537466422,
"tokens_seen": 342884352
},
{
"epoch": 0.1,
"learning_rate": 0.000452567703109328,
"loss": 3.0037,
"theoretical_loss": 4.08253097613686,
"tokens_seen": 342949888
},
{
"epoch": 0.1,
"learning_rate": 0.00045255767301905723,
"loss": 3.246,
"theoretical_loss": 4.082441436711459,
"tokens_seen": 343015424
},
{
"epoch": 0.1,
"learning_rate": 0.00045254764292878636,
"loss": 3.0855,
"theoretical_loss": 4.082351919180682,
"tokens_seen": 343080960
},
{
"epoch": 0.1,
"learning_rate": 0.0004525376128385156,
"loss": 3.1057,
"theoretical_loss": 4.082262423534992,
"tokens_seen": 343146496
},
{
"epoch": 0.1,
"learning_rate": 0.0004525275827482447,
"loss": 2.8346,
"theoretical_loss": 4.0821729497648604,
"tokens_seen": 343212032
},
{
"epoch": 0.1,
"learning_rate": 0.00045251755265797395,
"loss": 2.8814,
"theoretical_loss": 4.082083497860766,
"tokens_seen": 343277568
},
{
"epoch": 0.1,
"learning_rate": 0.00045250752256770314,
"loss": 2.9733,
"theoretical_loss": 4.0819940678131905,
"tokens_seen": 343343104
},
{
"epoch": 0.1,
"learning_rate": 0.0004524974924774323,
"loss": 2.9041,
"theoretical_loss": 4.081904659612624,
"tokens_seen": 343408640
},
{
"epoch": 0.1,
"learning_rate": 0.0004524874623871615,
"loss": 2.9287,
"theoretical_loss": 4.081815273249561,
"tokens_seen": 343474176
},
{
"epoch": 0.1,
"learning_rate": 0.0004524774322968907,
"loss": 3.0501,
"theoretical_loss": 4.081725908714505,
"tokens_seen": 343539712
},
{
"epoch": 0.1,
"learning_rate": 0.00045246740220661986,
"loss": 3.0915,
"theoretical_loss": 4.081636565997959,
"tokens_seen": 343605248
},
{
"epoch": 0.1,
"learning_rate": 0.0004524573721163491,
"loss": 3.2362,
"theoretical_loss": 4.08154724509044,
"tokens_seen": 343670784
},
{
"epoch": 0.1,
"learning_rate": 0.0004524473420260782,
"loss": 2.734,
"theoretical_loss": 4.081457945982465,
"tokens_seen": 343736320
},
{
"epoch": 0.1,
"learning_rate": 0.00045243731193580746,
"loss": 2.9276,
"theoretical_loss": 4.08136866866456,
"tokens_seen": 343801856
},
{
"epoch": 0.1,
"learning_rate": 0.0004524272818455366,
"loss": 3.1926,
"theoretical_loss": 4.081279413127254,
"tokens_seen": 343867392
},
{
"epoch": 0.1,
"learning_rate": 0.0004524172517552658,
"loss": 3.1345,
"theoretical_loss": 4.0811901793610845,
"tokens_seen": 343932928
},
{
"epoch": 0.1,
"learning_rate": 0.000452407221664995,
"loss": 3.2088,
"theoretical_loss": 4.081100967356594,
"tokens_seen": 343998464
},
{
"epoch": 0.1,
"objective/train/docs_used": 206760,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.804455041885376,
"objective/train/theoretical_loss": 4.081011777104333,
"objective/train/tokens_used": 364524000,
"theoretical_loss": 4.081011777104333,
"tokens_seen": 344064000
},
{
"epoch": 0.1,
"learning_rate": 0.0004523971915747242,
"loss": 3.0332,
"theoretical_loss": 4.081011777104333,
"tokens_seen": 344064000
},
{
"epoch": 0.1,
"learning_rate": 0.00045238716148445336,
"loss": 3.0908,
"theoretical_loss": 4.080922608594854,
"tokens_seen": 344129536
},
{
"epoch": 0.1,
"learning_rate": 0.0004523771313941826,
"loss": 3.0488,
"theoretical_loss": 4.080833461818718,
"tokens_seen": 344195072
},
{
"epoch": 0.1,
"learning_rate": 0.0004523671013039117,
"loss": 3.1212,
"theoretical_loss": 4.08074433676649,
"tokens_seen": 344260608
},
{
"epoch": 0.1,
"learning_rate": 0.00045235707121364096,
"loss": 3.2725,
"theoretical_loss": 4.080655233428745,
"tokens_seen": 344326144
},
{
"epoch": 0.1,
"learning_rate": 0.0004523470411233701,
"loss": 3.1866,
"theoretical_loss": 4.0805661517960585,
"tokens_seen": 344391680
},
{
"epoch": 0.1,
"learning_rate": 0.0004523370110330993,
"loss": 2.8178,
"theoretical_loss": 4.080477091859015,
"tokens_seen": 344457216
},
{
"epoch": 0.1,
"learning_rate": 0.0004523269809428285,
"loss": 3.1468,
"theoretical_loss": 4.080388053608205,
"tokens_seen": 344522752
},
{
"epoch": 0.1,
"learning_rate": 0.0004523169508525577,
"loss": 2.6369,
"theoretical_loss": 4.080299037034224,
"tokens_seen": 344588288
},
{
"epoch": 0.1,
"learning_rate": 0.00045230692076228687,
"loss": 3.295,
"theoretical_loss": 4.080210042127672,
"tokens_seen": 344653824
},
{
"epoch": 0.1,
"learning_rate": 0.00045229689067201605,
"loss": 2.9329,
"theoretical_loss": 4.0801210688791585,
"tokens_seen": 344719360
},
{
"epoch": 0.1,
"learning_rate": 0.00045228686058174523,
"loss": 2.8028,
"theoretical_loss": 4.080032117279296,
"tokens_seen": 344784896
},
{
"epoch": 0.1,
"learning_rate": 0.00045227683049147446,
"loss": 3.0295,
"theoretical_loss": 4.079943187318703,
"tokens_seen": 344850432
},
{
"epoch": 0.1,
"learning_rate": 0.0004522668004012036,
"loss": 2.8921,
"theoretical_loss": 4.079854278988005,
"tokens_seen": 344915968
},
{
"epoch": 0.1,
"learning_rate": 0.0004522567703109328,
"loss": 2.8358,
"theoretical_loss": 4.0797653922778325,
"tokens_seen": 344981504
},
{
"epoch": 0.1,
"learning_rate": 0.00045224674022066195,
"loss": 3.1175,
"theoretical_loss": 4.0796765271788225,
"tokens_seen": 345047040
},
{
"epoch": 0.1,
"learning_rate": 0.0004522367101303912,
"loss": 3.1971,
"theoretical_loss": 4.079587683681617,
"tokens_seen": 345112576
},
{
"epoch": 0.1,
"learning_rate": 0.00045222668004012037,
"loss": 3.2261,
"theoretical_loss": 4.079498861776865,
"tokens_seen": 345178112
},
{
"epoch": 0.1,
"learning_rate": 0.00045221664994984955,
"loss": 3.0631,
"theoretical_loss": 4.07941006145522,
"tokens_seen": 345243648
},
{
"epoch": 0.1,
"learning_rate": 0.00045220661985957873,
"loss": 3.0806,
"theoretical_loss": 4.079321282707342,
"tokens_seen": 345309184
},
{
"epoch": 0.1,
"learning_rate": 0.00045219658976930797,
"loss": 2.8805,
"theoretical_loss": 4.0792325255238975,
"tokens_seen": 345374720
},
{
"epoch": 0.1,
"learning_rate": 0.0004521865596790371,
"loss": 3.0152,
"theoretical_loss": 4.079143789895556,
"tokens_seen": 345440256
},
{
"epoch": 0.1,
"learning_rate": 0.00045217652958876633,
"loss": 2.8819,
"theoretical_loss": 4.079055075812997,
"tokens_seen": 345505792
},
{
"epoch": 0.1,
"learning_rate": 0.00045216649949849546,
"loss": 3.0413,
"theoretical_loss": 4.078966383266904,
"tokens_seen": 345571328
},
{
"epoch": 0.1,
"learning_rate": 0.0004521564694082247,
"loss": 2.9147,
"theoretical_loss": 4.078877712247963,
"tokens_seen": 345636864
},
{
"epoch": 0.1,
"objective/train/docs_used": 207336,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.845684051513672,
"objective/train/theoretical_loss": 4.078789062746872,
"objective/train/tokens_used": 366162400,
"theoretical_loss": 4.078789062746872,
"tokens_seen": 345702400
},
{
"epoch": 0.1,
"learning_rate": 0.00045214643931795387,
"loss": 3.2098,
"theoretical_loss": 4.078789062746872,
"tokens_seen": 345702400
},
{
"epoch": 0.1,
"learning_rate": 0.00045213640922768305,
"loss": 3.0034,
"theoretical_loss": 4.07870043475433,
"tokens_seen": 345767936
},
{
"epoch": 0.1,
"learning_rate": 0.00045212637913741223,
"loss": 2.8799,
"theoretical_loss": 4.078611828261042,
"tokens_seen": 345833472
},
{
"epoch": 0.1,
"learning_rate": 0.0004521163490471414,
"loss": 3.0001,
"theoretical_loss": 4.078523243257723,
"tokens_seen": 345899008
},
{
"epoch": 0.1,
"learning_rate": 0.0004521063189568706,
"loss": 2.9954,
"theoretical_loss": 4.078434679735088,
"tokens_seen": 345964544
},
{
"epoch": 0.1,
"learning_rate": 0.00045209628886659983,
"loss": 2.9793,
"theoretical_loss": 4.078346137683862,
"tokens_seen": 346030080
},
{
"epoch": 0.1,
"learning_rate": 0.00045208625877632896,
"loss": 2.8979,
"theoretical_loss": 4.078257617094776,
"tokens_seen": 346095616
},
{
"epoch": 0.1,
"learning_rate": 0.0004520762286860582,
"loss": 2.9292,
"theoretical_loss": 4.078169117958562,
"tokens_seen": 346161152
},
{
"epoch": 0.1,
"learning_rate": 0.0004520661985957873,
"loss": 2.9788,
"theoretical_loss": 4.078080640265963,
"tokens_seen": 346226688
},
{
"epoch": 0.1,
"learning_rate": 0.00045205616850551656,
"loss": 2.9652,
"theoretical_loss": 4.077992184007725,
"tokens_seen": 346292224
},
{
"epoch": 0.1,
"learning_rate": 0.00045204613841524574,
"loss": 3.0513,
"theoretical_loss": 4.0779037491746,
"tokens_seen": 346357760
},
{
"epoch": 0.1,
"learning_rate": 0.0004520361083249749,
"loss": 3.1309,
"theoretical_loss": 4.077815335757347,
"tokens_seen": 346423296
},
{
"epoch": 0.1,
"learning_rate": 0.0004520260782347041,
"loss": 2.9017,
"theoretical_loss": 4.077726943746729,
"tokens_seen": 346488832
},
{
"epoch": 0.11,
"learning_rate": 0.00045201604814443334,
"loss": 2.973,
"theoretical_loss": 4.077638573133515,
"tokens_seen": 346554368
},
{
"epoch": 0.11,
"learning_rate": 0.00045200601805416246,
"loss": 3.3341,
"theoretical_loss": 4.077550223908482,
"tokens_seen": 346619904
},
{
"epoch": 0.11,
"learning_rate": 0.0004519959879638917,
"loss": 3.0235,
"theoretical_loss": 4.0774618960624105,
"tokens_seen": 346685440
},
{
"epoch": 0.11,
"learning_rate": 0.0004519859578736208,
"loss": 3.1139,
"theoretical_loss": 4.077373589586087,
"tokens_seen": 346750976
},
{
"epoch": 0.11,
"learning_rate": 0.00045197592778335006,
"loss": 3.1671,
"theoretical_loss": 4.077285304470304,
"tokens_seen": 346816512
},
{
"epoch": 0.11,
"learning_rate": 0.00045196589769307924,
"loss": 2.8109,
"theoretical_loss": 4.077197040705859,
"tokens_seen": 346882048
},
{
"epoch": 0.11,
"learning_rate": 0.0004519558676028084,
"loss": 2.9381,
"theoretical_loss": 4.077108798283557,
"tokens_seen": 346947584
},
{
"epoch": 0.11,
"learning_rate": 0.00045194583751253766,
"loss": 3.0504,
"theoretical_loss": 4.077020577194207,
"tokens_seen": 347013120
},
{
"epoch": 0.11,
"learning_rate": 0.0004519358074222668,
"loss": 3.0095,
"theoretical_loss": 4.076932377428625,
"tokens_seen": 347078656
},
{
"epoch": 0.11,
"learning_rate": 0.000451925777331996,
"loss": 2.9958,
"theoretical_loss": 4.076844198977629,
"tokens_seen": 347144192
},
{
"epoch": 0.11,
"learning_rate": 0.0004519157472417252,
"loss": 2.8433,
"theoretical_loss": 4.076756041832049,
"tokens_seen": 347209728
},
{
"epoch": 0.11,
"learning_rate": 0.0004519057171514544,
"loss": 2.9906,
"theoretical_loss": 4.076667905982717,
"tokens_seen": 347275264
},
{
"epoch": 0.11,
"objective/train/docs_used": 208796,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0841803550720215,
"objective/train/theoretical_loss": 4.076579791420469,
"objective/train/tokens_used": 367800800,
"theoretical_loss": 4.076579791420469,
"tokens_seen": 347340800
},
{
"epoch": 0.11,
"learning_rate": 0.00045189568706118356,
"loss": 3.057,
"theoretical_loss": 4.076579791420469,
"tokens_seen": 347340800
},
{
"epoch": 0.11,
"learning_rate": 0.0004518856569709128,
"loss": 2.9187,
"theoretical_loss": 4.0764916981361505,
"tokens_seen": 347406336
},
{
"epoch": 0.11,
"learning_rate": 0.0004518756268806419,
"loss": 2.898,
"theoretical_loss": 4.07640362612061,
"tokens_seen": 347471872
},
{
"epoch": 0.11,
"learning_rate": 0.00045186559679037116,
"loss": 2.9909,
"theoretical_loss": 4.076315575364702,
"tokens_seen": 347537408
},
{
"epoch": 0.11,
"learning_rate": 0.0004518555667001003,
"loss": 2.9392,
"theoretical_loss": 4.076227545859288,
"tokens_seen": 347602944
},
{
"epoch": 0.11,
"learning_rate": 0.0004518455366098295,
"loss": 2.988,
"theoretical_loss": 4.076139537595234,
"tokens_seen": 347668480
},
{
"epoch": 0.11,
"learning_rate": 0.0004518355065195587,
"loss": 3.1748,
"theoretical_loss": 4.076051550563411,
"tokens_seen": 347734016
},
{
"epoch": 0.11,
"learning_rate": 0.0004518254764292879,
"loss": 3.1815,
"theoretical_loss": 4.075963584754698,
"tokens_seen": 347799552
},
{
"epoch": 0.11,
"learning_rate": 0.00045181544633901707,
"loss": 2.9956,
"theoretical_loss": 4.075875640159978,
"tokens_seen": 347865088
},
{
"epoch": 0.11,
"learning_rate": 0.00045180541624874625,
"loss": 2.8136,
"theoretical_loss": 4.075787716770139,
"tokens_seen": 347930624
},
{
"epoch": 0.11,
"learning_rate": 0.00045179538615847543,
"loss": 3.1089,
"theoretical_loss": 4.075699814576076,
"tokens_seen": 347996160
},
{
"epoch": 0.11,
"learning_rate": 0.00045178535606820466,
"loss": 2.8611,
"theoretical_loss": 4.075611933568689,
"tokens_seen": 348061696
},
{
"epoch": 0.11,
"learning_rate": 0.0004517753259779338,
"loss": 3.0451,
"theoretical_loss": 4.075524073738883,
"tokens_seen": 348127232
},
{
"epoch": 0.11,
"learning_rate": 0.000451765295887663,
"loss": 2.7604,
"theoretical_loss": 4.0754362350775715,
"tokens_seen": 348192768
},
{
"epoch": 0.11,
"learning_rate": 0.00045175526579739215,
"loss": 3.1277,
"theoretical_loss": 4.075348417575669,
"tokens_seen": 348258304
},
{
"epoch": 0.11,
"learning_rate": 0.0004517452357071214,
"loss": 3.0278,
"theoretical_loss": 4.075260621224099,
"tokens_seen": 348323840
},
{
"epoch": 0.11,
"learning_rate": 0.00045173520561685057,
"loss": 2.9556,
"theoretical_loss": 4.075172846013789,
"tokens_seen": 348389376
},
{
"epoch": 0.11,
"learning_rate": 0.00045172517552657975,
"loss": 2.7685,
"theoretical_loss": 4.075085091935675,
"tokens_seen": 348454912
},
{
"epoch": 0.11,
"learning_rate": 0.00045171514543630893,
"loss": 3.0181,
"theoretical_loss": 4.074997358980694,
"tokens_seen": 348520448
},
{
"epoch": 0.11,
"learning_rate": 0.00045170511534603817,
"loss": 2.9357,
"theoretical_loss": 4.0749096471397905,
"tokens_seen": 348585984
},
{
"epoch": 0.11,
"learning_rate": 0.0004516950852557673,
"loss": 3.1258,
"theoretical_loss": 4.074821956403917,
"tokens_seen": 348651520
},
{
"epoch": 0.11,
"learning_rate": 0.00045168505516549653,
"loss": 3.1191,
"theoretical_loss": 4.074734286764029,
"tokens_seen": 348717056
},
{
"epoch": 0.11,
"learning_rate": 0.00045167502507522566,
"loss": 3.2274,
"theoretical_loss": 4.074646638211087,
"tokens_seen": 348782592
},
{
"epoch": 0.11,
"learning_rate": 0.0004516649949849549,
"loss": 3.0471,
"theoretical_loss": 4.074559010736059,
"tokens_seen": 348848128
},
{
"epoch": 0.11,
"learning_rate": 0.00045165496489468407,
"loss": 2.7022,
"theoretical_loss": 4.074471404329918,
"tokens_seen": 348913664
},
{
"epoch": 0.11,
"objective/train/docs_used": 209597,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.47709059715271,
"objective/train/theoretical_loss": 4.074383818983643,
"objective/train/tokens_used": 369439200,
"theoretical_loss": 4.074383818983643,
"tokens_seen": 348979200
},
{
"epoch": 0.11,
"learning_rate": 0.00045164493480441325,
"loss": 3.1346,
"theoretical_loss": 4.074383818983643,
"tokens_seen": 348979200
},
{
"epoch": 0.11,
"learning_rate": 0.00045163490471414244,
"loss": 2.8909,
"theoretical_loss": 4.074296254688216,
"tokens_seen": 349044736
},
{
"epoch": 0.11,
"learning_rate": 0.0004516248746238716,
"loss": 2.9947,
"theoretical_loss": 4.074208711434628,
"tokens_seen": 349110272
},
{
"epoch": 0.11,
"learning_rate": 0.0004516148445336008,
"loss": 3.012,
"theoretical_loss": 4.074121189213873,
"tokens_seen": 349175808
},
{
"epoch": 0.11,
"learning_rate": 0.00045160481444333003,
"loss": 3.2239,
"theoretical_loss": 4.074033688016953,
"tokens_seen": 349241344
},
{
"epoch": 0.11,
"learning_rate": 0.00045159478435305916,
"loss": 3.0182,
"theoretical_loss": 4.073946207834872,
"tokens_seen": 349306880
},
{
"epoch": 0.11,
"learning_rate": 0.0004515847542627884,
"loss": 2.9586,
"theoretical_loss": 4.073858748658644,
"tokens_seen": 349372416
},
{
"epoch": 0.11,
"learning_rate": 0.0004515747241725175,
"loss": 3.0998,
"theoretical_loss": 4.073771310479284,
"tokens_seen": 349437952
},
{
"epoch": 0.11,
"learning_rate": 0.00045156469408224676,
"loss": 3.0868,
"theoretical_loss": 4.073683893287816,
"tokens_seen": 349503488
},
{
"epoch": 0.11,
"learning_rate": 0.00045155466399197594,
"loss": 2.9732,
"theoretical_loss": 4.073596497075267,
"tokens_seen": 349569024
},
{
"epoch": 0.11,
"learning_rate": 0.0004515446339017051,
"loss": 2.7333,
"theoretical_loss": 4.073509121832672,
"tokens_seen": 349634560
},
{
"epoch": 0.11,
"learning_rate": 0.0004515346038114343,
"loss": 2.9115,
"theoretical_loss": 4.07342176755107,
"tokens_seen": 349700096
},
{
"epoch": 0.11,
"learning_rate": 0.00045152457372116354,
"loss": 3.1394,
"theoretical_loss": 4.073334434221506,
"tokens_seen": 349765632
},
{
"epoch": 0.11,
"learning_rate": 0.00045151454363089266,
"loss": 3.1035,
"theoretical_loss": 4.073247121835028,
"tokens_seen": 349831168
},
{
"epoch": 0.11,
"learning_rate": 0.0004515045135406219,
"loss": 3.0943,
"theoretical_loss": 4.073159830382695,
"tokens_seen": 349896704
},
{
"epoch": 0.11,
"learning_rate": 0.000451494483450351,
"loss": 3.0074,
"theoretical_loss": 4.073072559855565,
"tokens_seen": 349962240
},
{
"epoch": 0.11,
"learning_rate": 0.00045148445336008026,
"loss": 2.8102,
"theoretical_loss": 4.072985310244707,
"tokens_seen": 350027776
},
{
"epoch": 0.11,
"learning_rate": 0.00045147442326980944,
"loss": 3.119,
"theoretical_loss": 4.072898081541193,
"tokens_seen": 350093312
},
{
"epoch": 0.11,
"learning_rate": 0.0004514643931795386,
"loss": 3.0616,
"theoretical_loss": 4.0728108737361,
"tokens_seen": 350158848
},
{
"epoch": 0.11,
"learning_rate": 0.0004514543630892678,
"loss": 2.9345,
"theoretical_loss": 4.0727236868205114,
"tokens_seen": 350224384
},
{
"epoch": 0.11,
"learning_rate": 0.000451444332998997,
"loss": 2.7941,
"theoretical_loss": 4.072636520785515,
"tokens_seen": 350289920
},
{
"epoch": 0.11,
"learning_rate": 0.00045143430290872617,
"loss": 3.1957,
"theoretical_loss": 4.072549375622207,
"tokens_seen": 350355456
},
{
"epoch": 0.11,
"learning_rate": 0.0004514242728184554,
"loss": 2.6432,
"theoretical_loss": 4.072462251321684,
"tokens_seen": 350420992
},
{
"epoch": 0.11,
"learning_rate": 0.00045141424272818453,
"loss": 3.1534,
"theoretical_loss": 4.072375147875054,
"tokens_seen": 350486528
},
{
"epoch": 0.11,
"learning_rate": 0.00045140421263791376,
"loss": 2.9646,
"theoretical_loss": 4.072288065273426,
"tokens_seen": 350552064
},
{
"epoch": 0.11,
"objective/train/docs_used": 210243,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.462125539779663,
"objective/train/theoretical_loss": 4.0722010035079155,
"objective/train/tokens_used": 371077600,
"theoretical_loss": 4.0722010035079155,
"tokens_seen": 350617600
},
{
"epoch": 0.11,
"learning_rate": 0.00045139418254764294,
"loss": 2.7561,
"theoretical_loss": 4.0722010035079155,
"tokens_seen": 350617600
},
{
"epoch": 0.11,
"learning_rate": 0.0004513841524573721,
"loss": 2.7177,
"theoretical_loss": 4.072113962569645,
"tokens_seen": 350683136
},
{
"epoch": 0.11,
"learning_rate": 0.0004513741223671013,
"loss": 2.6755,
"theoretical_loss": 4.07202694244974,
"tokens_seen": 350748672
},
{
"epoch": 0.11,
"learning_rate": 0.0004513640922768305,
"loss": 2.8439,
"theoretical_loss": 4.071939943139334,
"tokens_seen": 350814208
},
{
"epoch": 0.11,
"learning_rate": 0.00045135406218655967,
"loss": 2.9904,
"theoretical_loss": 4.0718529646295645,
"tokens_seen": 350879744
},
{
"epoch": 0.11,
"learning_rate": 0.0004513440320962889,
"loss": 3.0021,
"theoretical_loss": 4.071766006911574,
"tokens_seen": 350945280
},
{
"epoch": 0.11,
"learning_rate": 0.00045133400200601803,
"loss": 2.9347,
"theoretical_loss": 4.071679069976511,
"tokens_seen": 351010816
},
{
"epoch": 0.11,
"learning_rate": 0.00045132397191574727,
"loss": 2.9926,
"theoretical_loss": 4.07159215381553,
"tokens_seen": 351076352
},
{
"epoch": 0.11,
"learning_rate": 0.0004513139418254764,
"loss": 2.9278,
"theoretical_loss": 4.07150525841979,
"tokens_seen": 351141888
},
{
"epoch": 0.11,
"learning_rate": 0.00045130391173520563,
"loss": 2.9006,
"theoretical_loss": 4.071418383780456,
"tokens_seen": 351207424
},
{
"epoch": 0.11,
"learning_rate": 0.0004512938816449348,
"loss": 2.9848,
"theoretical_loss": 4.071331529888698,
"tokens_seen": 351272960
},
{
"epoch": 0.11,
"learning_rate": 0.000451283851554664,
"loss": 2.7971,
"theoretical_loss": 4.071244696735692,
"tokens_seen": 351338496
},
{
"epoch": 0.11,
"learning_rate": 0.00045127382146439317,
"loss": 2.8099,
"theoretical_loss": 4.071157884312619,
"tokens_seen": 351404032
},
{
"epoch": 0.11,
"learning_rate": 0.00045126379137412235,
"loss": 3.0982,
"theoretical_loss": 4.071071092610664,
"tokens_seen": 351469568
},
{
"epoch": 0.11,
"learning_rate": 0.00045125376128385153,
"loss": 3.0576,
"theoretical_loss": 4.070984321621021,
"tokens_seen": 351535104
},
{
"epoch": 0.11,
"learning_rate": 0.00045124373119358077,
"loss": 2.9455,
"theoretical_loss": 4.070897571334886,
"tokens_seen": 351600640
},
{
"epoch": 0.11,
"learning_rate": 0.0004512337011033099,
"loss": 3.0845,
"theoretical_loss": 4.0708108417434605,
"tokens_seen": 351666176
},
{
"epoch": 0.11,
"learning_rate": 0.00045122367101303913,
"loss": 3.0453,
"theoretical_loss": 4.070724132837954,
"tokens_seen": 351731712
},
{
"epoch": 0.11,
"learning_rate": 0.0004512136409227683,
"loss": 2.903,
"theoretical_loss": 4.07063744460958,
"tokens_seen": 351797248
},
{
"epoch": 0.11,
"learning_rate": 0.0004512036108324975,
"loss": 3.127,
"theoretical_loss": 4.070550777049555,
"tokens_seen": 351862784
},
{
"epoch": 0.11,
"learning_rate": 0.00045119358074222673,
"loss": 3.1144,
"theoretical_loss": 4.070464130149106,
"tokens_seen": 351928320
},
{
"epoch": 0.11,
"learning_rate": 0.00045118355065195586,
"loss": 2.997,
"theoretical_loss": 4.07037750389946,
"tokens_seen": 351993856
},
{
"epoch": 0.11,
"learning_rate": 0.0004511735205616851,
"loss": 3.002,
"theoretical_loss": 4.070290898291852,
"tokens_seen": 352059392
},
{
"epoch": 0.11,
"learning_rate": 0.0004511634904714143,
"loss": 3.0532,
"theoretical_loss": 4.070204313317523,
"tokens_seen": 352124928
},
{
"epoch": 0.11,
"learning_rate": 0.00045115346038114345,
"loss": 3.0537,
"theoretical_loss": 4.07011774896772,
"tokens_seen": 352190464
},
{
"epoch": 0.11,
"objective/train/docs_used": 211560,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.582488775253296,
"objective/train/theoretical_loss": 4.07003120523369,
"objective/train/tokens_used": 372716000,
"theoretical_loss": 4.07003120523369,
"tokens_seen": 352256000
},
{
"epoch": 0.11,
"learning_rate": 0.00045114343029087264,
"loss": 2.691,
"theoretical_loss": 4.07003120523369,
"tokens_seen": 352256000
},
{
"epoch": 0.11,
"learning_rate": 0.0004511334002006018,
"loss": 3.0749,
"theoretical_loss": 4.069944682106691,
"tokens_seen": 352321536
},
{
"epoch": 0.11,
"learning_rate": 0.000451123370110331,
"loss": 3.0175,
"theoretical_loss": 4.0698581795779845,
"tokens_seen": 352387072
},
{
"epoch": 0.11,
"learning_rate": 0.00045111334002006023,
"loss": 3.0017,
"theoretical_loss": 4.069771697638838,
"tokens_seen": 352452608
},
{
"epoch": 0.11,
"learning_rate": 0.00045110330992978936,
"loss": 3.0474,
"theoretical_loss": 4.069685236280522,
"tokens_seen": 352518144
},
{
"epoch": 0.11,
"learning_rate": 0.0004510932798395186,
"loss": 2.9058,
"theoretical_loss": 4.069598795494315,
"tokens_seen": 352583680
},
{
"epoch": 0.11,
"learning_rate": 0.0004510832497492477,
"loss": 2.8735,
"theoretical_loss": 4.0695123752715,
"tokens_seen": 352649216
},
{
"epoch": 0.11,
"learning_rate": 0.00045107321965897696,
"loss": 2.8146,
"theoretical_loss": 4.069425975603364,
"tokens_seen": 352714752
},
{
"epoch": 0.11,
"learning_rate": 0.00045106318956870614,
"loss": 2.9708,
"theoretical_loss": 4.069339596481201,
"tokens_seen": 352780288
},
{
"epoch": 0.11,
"learning_rate": 0.0004510531594784353,
"loss": 2.9669,
"theoretical_loss": 4.0692532378963095,
"tokens_seen": 352845824
},
{
"epoch": 0.11,
"learning_rate": 0.0004510431293881645,
"loss": 2.8697,
"theoretical_loss": 4.069166899839994,
"tokens_seen": 352911360
},
{
"epoch": 0.11,
"learning_rate": 0.00045103309929789374,
"loss": 3.0089,
"theoretical_loss": 4.069080582303563,
"tokens_seen": 352976896
},
{
"epoch": 0.11,
"learning_rate": 0.00045102306920762286,
"loss": 3.1818,
"theoretical_loss": 4.068994285278332,
"tokens_seen": 353042432
},
{
"epoch": 0.11,
"learning_rate": 0.0004510130391173521,
"loss": 3.1709,
"theoretical_loss": 4.068908008755619,
"tokens_seen": 353107968
},
{
"epoch": 0.11,
"learning_rate": 0.0004510030090270812,
"loss": 3.2865,
"theoretical_loss": 4.068821752726751,
"tokens_seen": 353173504
},
{
"epoch": 0.11,
"learning_rate": 0.00045099297893681046,
"loss": 2.9514,
"theoretical_loss": 4.0687355171830575,
"tokens_seen": 353239040
},
{
"epoch": 0.11,
"learning_rate": 0.00045098294884653964,
"loss": 2.7252,
"theoretical_loss": 4.068649302115875,
"tokens_seen": 353304576
},
{
"epoch": 0.11,
"learning_rate": 0.0004509729187562688,
"loss": 3.2343,
"theoretical_loss": 4.068563107516544,
"tokens_seen": 353370112
},
{
"epoch": 0.11,
"learning_rate": 0.000450962888665998,
"loss": 2.8777,
"theoretical_loss": 4.068476933376411,
"tokens_seen": 353435648
},
{
"epoch": 0.11,
"learning_rate": 0.0004509528585757272,
"loss": 2.9855,
"theoretical_loss": 4.068390779686826,
"tokens_seen": 353501184
},
{
"epoch": 0.11,
"learning_rate": 0.00045094282848545637,
"loss": 2.8759,
"theoretical_loss": 4.068304646439148,
"tokens_seen": 353566720
},
{
"epoch": 0.11,
"learning_rate": 0.0004509327983951856,
"loss": 2.7545,
"theoretical_loss": 4.068218533624737,
"tokens_seen": 353632256
},
{
"epoch": 0.11,
"learning_rate": 0.00045092276830491473,
"loss": 3.0406,
"theoretical_loss": 4.068132441234962,
"tokens_seen": 353697792
},
{
"epoch": 0.11,
"learning_rate": 0.00045091273821464396,
"loss": 2.9013,
"theoretical_loss": 4.068046369261194,
"tokens_seen": 353763328
},
{
"epoch": 0.11,
"learning_rate": 0.00045090270812437314,
"loss": 3.1276,
"theoretical_loss": 4.067960317694812,
"tokens_seen": 353828864
},
{
"epoch": 0.11,
"objective/train/docs_used": 212189,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9633750915527344,
"objective/train/theoretical_loss": 4.067874286527197,
"objective/train/tokens_used": 374354400,
"theoretical_loss": 4.067874286527197,
"tokens_seen": 353894400
},
{
"epoch": 0.11,
"learning_rate": 0.0004508926780341023,
"loss": 3.2043,
"theoretical_loss": 4.067874286527197,
"tokens_seen": 353894400
},
{
"epoch": 0.11,
"learning_rate": 0.0004508826479438315,
"loss": 2.9853,
"theoretical_loss": 4.06778827574974,
"tokens_seen": 353959936
},
{
"epoch": 0.11,
"learning_rate": 0.0004508726178535607,
"loss": 3.1729,
"theoretical_loss": 4.067702285353833,
"tokens_seen": 354025472
},
{
"epoch": 0.11,
"learning_rate": 0.00045086258776328987,
"loss": 3.0317,
"theoretical_loss": 4.0676163153308735,
"tokens_seen": 354091008
},
{
"epoch": 0.11,
"learning_rate": 0.0004508525576730191,
"loss": 2.9049,
"theoretical_loss": 4.067530365672267,
"tokens_seen": 354156544
},
{
"epoch": 0.11,
"learning_rate": 0.00045084252758274823,
"loss": 2.7865,
"theoretical_loss": 4.067444436369422,
"tokens_seen": 354222080
},
{
"epoch": 0.11,
"learning_rate": 0.00045083249749247747,
"loss": 3.1346,
"theoretical_loss": 4.067358527413753,
"tokens_seen": 354287616
},
{
"epoch": 0.11,
"learning_rate": 0.0004508224674022066,
"loss": 2.8867,
"theoretical_loss": 4.06727263879668,
"tokens_seen": 354353152
},
{
"epoch": 0.11,
"learning_rate": 0.00045081243731193583,
"loss": 2.9152,
"theoretical_loss": 4.067186770509627,
"tokens_seen": 354418688
},
{
"epoch": 0.11,
"learning_rate": 0.000450802407221665,
"loss": 2.8679,
"theoretical_loss": 4.067100922544023,
"tokens_seen": 354484224
},
{
"epoch": 0.11,
"learning_rate": 0.0004507923771313942,
"loss": 2.7877,
"theoretical_loss": 4.067015094891306,
"tokens_seen": 354549760
},
{
"epoch": 0.11,
"learning_rate": 0.00045078234704112337,
"loss": 2.9082,
"theoretical_loss": 4.066929287542914,
"tokens_seen": 354615296
},
{
"epoch": 0.11,
"learning_rate": 0.00045077231695085255,
"loss": 3.1084,
"theoretical_loss": 4.066843500490292,
"tokens_seen": 354680832
},
{
"epoch": 0.11,
"learning_rate": 0.00045076228686058173,
"loss": 2.8425,
"theoretical_loss": 4.066757733724893,
"tokens_seen": 354746368
},
{
"epoch": 0.11,
"learning_rate": 0.00045075225677031097,
"loss": 3.1105,
"theoretical_loss": 4.066671987238171,
"tokens_seen": 354811904
},
{
"epoch": 0.11,
"learning_rate": 0.0004507422266800401,
"loss": 2.7815,
"theoretical_loss": 4.0665862610215875,
"tokens_seen": 354877440
},
{
"epoch": 0.11,
"learning_rate": 0.00045073219658976933,
"loss": 2.8093,
"theoretical_loss": 4.066500555066609,
"tokens_seen": 354942976
},
{
"epoch": 0.11,
"learning_rate": 0.0004507221664994985,
"loss": 3.1331,
"theoretical_loss": 4.066414869364707,
"tokens_seen": 355008512
},
{
"epoch": 0.11,
"learning_rate": 0.0004507121364092277,
"loss": 3.0502,
"theoretical_loss": 4.066329203907358,
"tokens_seen": 355074048
},
{
"epoch": 0.11,
"learning_rate": 0.0004507021063189569,
"loss": 2.9057,
"theoretical_loss": 4.066243558686043,
"tokens_seen": 355139584
},
{
"epoch": 0.11,
"learning_rate": 0.00045069207622868606,
"loss": 2.8425,
"theoretical_loss": 4.066157933692249,
"tokens_seen": 355205120
},
{
"epoch": 0.11,
"learning_rate": 0.00045068204613841524,
"loss": 2.8221,
"theoretical_loss": 4.066072328917469,
"tokens_seen": 355270656
},
{
"epoch": 0.11,
"learning_rate": 0.0004506720160481445,
"loss": 2.9991,
"theoretical_loss": 4.0659867443532,
"tokens_seen": 355336192
},
{
"epoch": 0.11,
"learning_rate": 0.0004506619859578736,
"loss": 3.0212,
"theoretical_loss": 4.065901179990943,
"tokens_seen": 355401728
},
{
"epoch": 0.11,
"learning_rate": 0.00045065195586760284,
"loss": 2.8883,
"theoretical_loss": 4.065815635822206,
"tokens_seen": 355467264
},
{
"epoch": 0.11,
"objective/train/docs_used": 213521,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.758910894393921,
"objective/train/theoretical_loss": 4.065730111838502,
"objective/train/tokens_used": 375992800,
"theoretical_loss": 4.065730111838502,
"tokens_seen": 355532800
},
{
"epoch": 0.11,
"learning_rate": 0.00045064192577733196,
"loss": 3.0933,
"theoretical_loss": 4.065730111838502,
"tokens_seen": 355532800
},
{
"epoch": 0.11,
"learning_rate": 0.0004506318956870612,
"loss": 3.0034,
"theoretical_loss": 4.0656446080313495,
"tokens_seen": 355598336
},
{
"epoch": 0.11,
"learning_rate": 0.0004506218655967904,
"loss": 2.9333,
"theoretical_loss": 4.065559124392269,
"tokens_seen": 355663872
},
{
"epoch": 0.11,
"learning_rate": 0.00045061183550651956,
"loss": 3.1183,
"theoretical_loss": 4.065473660912791,
"tokens_seen": 355729408
},
{
"epoch": 0.11,
"learning_rate": 0.00045060180541624874,
"loss": 3.0313,
"theoretical_loss": 4.065388217584447,
"tokens_seen": 355794944
},
{
"epoch": 0.11,
"learning_rate": 0.0004505917753259779,
"loss": 2.7215,
"theoretical_loss": 4.065302794398775,
"tokens_seen": 355860480
},
{
"epoch": 0.11,
"learning_rate": 0.0004505817452357071,
"loss": 2.8688,
"theoretical_loss": 4.0652173913473195,
"tokens_seen": 355926016
},
{
"epoch": 0.11,
"learning_rate": 0.00045057171514543634,
"loss": 2.7887,
"theoretical_loss": 4.065132008421628,
"tokens_seen": 355991552
},
{
"epoch": 0.11,
"learning_rate": 0.00045056168505516547,
"loss": 2.9822,
"theoretical_loss": 4.065046645613254,
"tokens_seen": 356057088
},
{
"epoch": 0.11,
"learning_rate": 0.0004505516549648947,
"loss": 3.0194,
"theoretical_loss": 4.0649613029137575,
"tokens_seen": 356122624
},
{
"epoch": 0.11,
"learning_rate": 0.0004505416248746239,
"loss": 2.9109,
"theoretical_loss": 4.0648759803147,
"tokens_seen": 356188160
},
{
"epoch": 0.11,
"learning_rate": 0.00045053159478435306,
"loss": 2.8684,
"theoretical_loss": 4.064790677807652,
"tokens_seen": 356253696
},
{
"epoch": 0.11,
"learning_rate": 0.00045052156469408224,
"loss": 2.7241,
"theoretical_loss": 4.064705395384187,
"tokens_seen": 356319232
},
{
"epoch": 0.11,
"learning_rate": 0.0004505115346038114,
"loss": 2.9499,
"theoretical_loss": 4.064620133035883,
"tokens_seen": 356384768
},
{
"epoch": 0.11,
"learning_rate": 0.0004505015045135406,
"loss": 2.9387,
"theoretical_loss": 4.0645348907543255,
"tokens_seen": 356450304
},
{
"epoch": 0.11,
"learning_rate": 0.00045049147442326984,
"loss": 2.9353,
"theoretical_loss": 4.064449668531103,
"tokens_seen": 356515840
},
{
"epoch": 0.11,
"learning_rate": 0.00045048144433299897,
"loss": 2.7079,
"theoretical_loss": 4.064364466357809,
"tokens_seen": 356581376
},
{
"epoch": 0.11,
"learning_rate": 0.0004504714142427282,
"loss": 2.8404,
"theoretical_loss": 4.064279284226044,
"tokens_seen": 356646912
},
{
"epoch": 0.11,
"learning_rate": 0.00045046138415245733,
"loss": 2.9141,
"theoretical_loss": 4.064194122127411,
"tokens_seen": 356712448
},
{
"epoch": 0.11,
"learning_rate": 0.00045045135406218657,
"loss": 2.9891,
"theoretical_loss": 4.06410898005352,
"tokens_seen": 356777984
},
{
"epoch": 0.11,
"learning_rate": 0.0004504413239719158,
"loss": 2.9367,
"theoretical_loss": 4.064023857995986,
"tokens_seen": 356843520
},
{
"epoch": 0.11,
"learning_rate": 0.00045043129388164493,
"loss": 2.8608,
"theoretical_loss": 4.0639387559464275,
"tokens_seen": 356909056
},
{
"epoch": 0.11,
"learning_rate": 0.00045042126379137416,
"loss": 2.9035,
"theoretical_loss": 4.063853673896469,
"tokens_seen": 356974592
},
{
"epoch": 0.11,
"learning_rate": 0.00045041123370110334,
"loss": 2.9264,
"theoretical_loss": 4.063768611837742,
"tokens_seen": 357040128
},
{
"epoch": 0.11,
"learning_rate": 0.0004504012036108325,
"loss": 2.9756,
"theoretical_loss": 4.063683569761879,
"tokens_seen": 357105664
},
{
"epoch": 0.11,
"objective/train/docs_used": 214188,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7816708087921143,
"objective/train/theoretical_loss": 4.063598547660519,
"objective/train/tokens_used": 377631200,
"theoretical_loss": 4.063598547660519,
"tokens_seen": 357171200
},
{
"epoch": 0.11,
"learning_rate": 0.0004503911735205617,
"loss": 2.8588,
"theoretical_loss": 4.063598547660519,
"tokens_seen": 357171200
},
{
"epoch": 0.11,
"learning_rate": 0.0004503811434302909,
"loss": 2.9712,
"theoretical_loss": 4.06351354552531,
"tokens_seen": 357236736
},
{
"epoch": 0.11,
"learning_rate": 0.00045037111334002007,
"loss": 2.9797,
"theoretical_loss": 4.063428563347898,
"tokens_seen": 357302272
},
{
"epoch": 0.11,
"learning_rate": 0.0004503610832497493,
"loss": 2.8845,
"theoretical_loss": 4.063343601119941,
"tokens_seen": 357367808
},
{
"epoch": 0.11,
"learning_rate": 0.00045035105315947843,
"loss": 2.6153,
"theoretical_loss": 4.063258658833096,
"tokens_seen": 357433344
},
{
"epoch": 0.11,
"learning_rate": 0.00045034102306920767,
"loss": 2.855,
"theoretical_loss": 4.06317373647903,
"tokens_seen": 357498880
},
{
"epoch": 0.11,
"learning_rate": 0.0004503309929789368,
"loss": 2.9601,
"theoretical_loss": 4.0630888340494105,
"tokens_seen": 357564416
},
{
"epoch": 0.11,
"learning_rate": 0.00045032096288866603,
"loss": 3.0078,
"theoretical_loss": 4.063003951535913,
"tokens_seen": 357629952
},
{
"epoch": 0.11,
"learning_rate": 0.0004503109327983952,
"loss": 2.8686,
"theoretical_loss": 4.06291908893022,
"tokens_seen": 357695488
},
{
"epoch": 0.11,
"learning_rate": 0.0004503009027081244,
"loss": 2.9919,
"theoretical_loss": 4.062834246224012,
"tokens_seen": 357761024
},
{
"epoch": 0.11,
"learning_rate": 0.00045029087261785357,
"loss": 2.882,
"theoretical_loss": 4.062749423408981,
"tokens_seen": 357826560
},
{
"epoch": 0.11,
"learning_rate": 0.00045028084252758275,
"loss": 2.8618,
"theoretical_loss": 4.062664620476822,
"tokens_seen": 357892096
},
{
"epoch": 0.11,
"learning_rate": 0.00045027081243731193,
"loss": 2.8418,
"theoretical_loss": 4.062579837419235,
"tokens_seen": 357957632
},
{
"epoch": 0.11,
"learning_rate": 0.00045026078234704117,
"loss": 2.8871,
"theoretical_loss": 4.062495074227923,
"tokens_seen": 358023168
},
{
"epoch": 0.11,
"learning_rate": 0.0004502507522567703,
"loss": 3.189,
"theoretical_loss": 4.062410330894597,
"tokens_seen": 358088704
},
{
"epoch": 0.11,
"learning_rate": 0.00045024072216649953,
"loss": 2.8227,
"theoretical_loss": 4.062325607410972,
"tokens_seen": 358154240
},
{
"epoch": 0.11,
"learning_rate": 0.0004502306920762287,
"loss": 3.106,
"theoretical_loss": 4.062240903768767,
"tokens_seen": 358219776
},
{
"epoch": 0.11,
"learning_rate": 0.0004502206619859579,
"loss": 2.8611,
"theoretical_loss": 4.062156219959705,
"tokens_seen": 358285312
},
{
"epoch": 0.11,
"learning_rate": 0.0004502106318956871,
"loss": 2.7625,
"theoretical_loss": 4.062071555975519,
"tokens_seen": 358350848
},
{
"epoch": 0.11,
"learning_rate": 0.00045020060180541626,
"loss": 2.8433,
"theoretical_loss": 4.061986911807942,
"tokens_seen": 358416384
},
{
"epoch": 0.11,
"learning_rate": 0.00045019057171514544,
"loss": 2.9341,
"theoretical_loss": 4.061902287448714,
"tokens_seen": 358481920
},
{
"epoch": 0.11,
"learning_rate": 0.0004501805416248747,
"loss": 2.9221,
"theoretical_loss": 4.061817682889577,
"tokens_seen": 358547456
},
{
"epoch": 0.11,
"learning_rate": 0.0004501705115346038,
"loss": 3.0211,
"theoretical_loss": 4.061733098122284,
"tokens_seen": 358612992
},
{
"epoch": 0.11,
"learning_rate": 0.00045016048144433304,
"loss": 2.7245,
"theoretical_loss": 4.061648533138587,
"tokens_seen": 358678528
},
{
"epoch": 0.11,
"learning_rate": 0.00045015045135406216,
"loss": 2.7328,
"theoretical_loss": 4.0615639879302465,
"tokens_seen": 358744064
},
{
"epoch": 0.11,
"objective/train/docs_used": 215284,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.302658796310425,
"objective/train/theoretical_loss": 4.061479462489026,
"objective/train/tokens_used": 379269600,
"theoretical_loss": 4.061479462489026,
"tokens_seen": 358809600
},
{
"epoch": 0.11,
"learning_rate": 0.0004501404212637914,
"loss": 2.824,
"theoretical_loss": 4.061479462489026,
"tokens_seen": 358809600
},
{
"epoch": 0.11,
"learning_rate": 0.0004501303911735206,
"loss": 2.6842,
"theoretical_loss": 4.061394956806694,
"tokens_seen": 358875136
},
{
"epoch": 0.11,
"learning_rate": 0.00045012036108324976,
"loss": 2.8211,
"theoretical_loss": 4.061310470875027,
"tokens_seen": 358940672
},
{
"epoch": 0.11,
"learning_rate": 0.00045011033099297894,
"loss": 2.9749,
"theoretical_loss": 4.0612260046858015,
"tokens_seen": 359006208
},
{
"epoch": 0.11,
"learning_rate": 0.0004501003009027081,
"loss": 3.2034,
"theoretical_loss": 4.061141558230803,
"tokens_seen": 359071744
},
{
"epoch": 0.11,
"learning_rate": 0.0004500902708124373,
"loss": 2.9755,
"theoretical_loss": 4.06105713150182,
"tokens_seen": 359137280
},
{
"epoch": 0.11,
"learning_rate": 0.00045008024072216654,
"loss": 2.9855,
"theoretical_loss": 4.060972724490646,
"tokens_seen": 359202816
},
{
"epoch": 0.11,
"learning_rate": 0.00045007021063189567,
"loss": 3.1566,
"theoretical_loss": 4.0608883371890805,
"tokens_seen": 359268352
},
{
"epoch": 0.11,
"learning_rate": 0.0004500601805416249,
"loss": 3.1019,
"theoretical_loss": 4.060803969588925,
"tokens_seen": 359333888
},
{
"epoch": 0.11,
"learning_rate": 0.0004500501504513541,
"loss": 3.2989,
"theoretical_loss": 4.06071962168199,
"tokens_seen": 359399424
},
{
"epoch": 0.11,
"learning_rate": 0.00045004012036108326,
"loss": 2.6658,
"theoretical_loss": 4.060635293460089,
"tokens_seen": 359464960
},
{
"epoch": 0.11,
"learning_rate": 0.00045003009027081244,
"loss": 2.9667,
"theoretical_loss": 4.060550984915039,
"tokens_seen": 359530496
},
{
"epoch": 0.11,
"learning_rate": 0.0004500200601805416,
"loss": 2.8963,
"theoretical_loss": 4.060466696038664,
"tokens_seen": 359596032
},
{
"epoch": 0.11,
"learning_rate": 0.0004500100300902708,
"loss": 2.9883,
"theoretical_loss": 4.060382426822792,
"tokens_seen": 359661568
},
{
"epoch": 0.11,
"learning_rate": 0.00045000000000000004,
"loss": 2.9093,
"theoretical_loss": 4.060298177259256,
"tokens_seen": 359727104
},
{
"epoch": 0.11,
"learning_rate": 0.00044998996990972917,
"loss": 3.025,
"theoretical_loss": 4.060213947339894,
"tokens_seen": 359792640
},
{
"epoch": 0.11,
"learning_rate": 0.0004499799398194584,
"loss": 3.1524,
"theoretical_loss": 4.060129737056547,
"tokens_seen": 359858176
},
{
"epoch": 0.11,
"learning_rate": 0.00044996990972918753,
"loss": 2.7814,
"theoretical_loss": 4.060045546401065,
"tokens_seen": 359923712
},
{
"epoch": 0.11,
"learning_rate": 0.00044995987963891677,
"loss": 3.1582,
"theoretical_loss": 4.0599613753653,
"tokens_seen": 359989248
},
{
"epoch": 0.11,
"learning_rate": 0.00044994984954864595,
"loss": 2.9273,
"theoretical_loss": 4.0598772239411085,
"tokens_seen": 360054784
},
{
"epoch": 0.11,
"learning_rate": 0.00044993981945837513,
"loss": 3.3608,
"theoretical_loss": 4.059793092120353,
"tokens_seen": 360120320
},
{
"epoch": 0.11,
"learning_rate": 0.0004499297893681043,
"loss": 2.8893,
"theoretical_loss": 4.0597089798949,
"tokens_seen": 360185856
},
{
"epoch": 0.11,
"learning_rate": 0.00044991975927783355,
"loss": 2.9264,
"theoretical_loss": 4.059624887256622,
"tokens_seen": 360251392
},
{
"epoch": 0.11,
"learning_rate": 0.00044990972918756267,
"loss": 2.929,
"theoretical_loss": 4.059540814197396,
"tokens_seen": 360316928
},
{
"epoch": 0.11,
"learning_rate": 0.0004498996990972919,
"loss": 3.0329,
"theoretical_loss": 4.059456760709104,
"tokens_seen": 360382464
},
{
"epoch": 0.11,
"objective/train/docs_used": 215704,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.066336154937744,
"objective/train/theoretical_loss": 4.05937272678363,
"objective/train/tokens_used": 380908000,
"theoretical_loss": 4.05937272678363,
"tokens_seen": 360448000
},
{
"epoch": 0.11,
"learning_rate": 0.00044988966900702103,
"loss": 2.8475,
"theoretical_loss": 4.05937272678363,
"tokens_seen": 360448000
},
{
"epoch": 0.11,
"learning_rate": 0.00044987963891675027,
"loss": 2.8076,
"theoretical_loss": 4.059288712412868,
"tokens_seen": 360513536
},
{
"epoch": 0.11,
"learning_rate": 0.00044986960882647945,
"loss": 2.908,
"theoretical_loss": 4.059204717588712,
"tokens_seen": 360579072
},
{
"epoch": 0.11,
"learning_rate": 0.00044985957873620863,
"loss": 3.1984,
"theoretical_loss": 4.059120742303065,
"tokens_seen": 360644608
},
{
"epoch": 0.11,
"learning_rate": 0.0004498495486459378,
"loss": 2.9602,
"theoretical_loss": 4.059036786547831,
"tokens_seen": 360710144
},
{
"epoch": 0.11,
"learning_rate": 0.000449839518555667,
"loss": 3.3413,
"theoretical_loss": 4.058952850314922,
"tokens_seen": 360775680
},
{
"epoch": 0.11,
"learning_rate": 0.0004498294884653962,
"loss": 2.8901,
"theoretical_loss": 4.058868933596252,
"tokens_seen": 360841216
},
{
"epoch": 0.11,
"learning_rate": 0.0004498194583751254,
"loss": 2.9428,
"theoretical_loss": 4.0587850363837426,
"tokens_seen": 360906752
},
{
"epoch": 0.11,
"learning_rate": 0.00044980942828485454,
"loss": 2.9984,
"theoretical_loss": 4.058701158669318,
"tokens_seen": 360972288
},
{
"epoch": 0.11,
"learning_rate": 0.00044979939819458377,
"loss": 3.1567,
"theoretical_loss": 4.058617300444908,
"tokens_seen": 361037824
},
{
"epoch": 0.11,
"learning_rate": 0.0004497893681043129,
"loss": 2.8662,
"theoretical_loss": 4.0585334617024476,
"tokens_seen": 361103360
},
{
"epoch": 0.11,
"learning_rate": 0.00044977933801404213,
"loss": 2.9641,
"theoretical_loss": 4.058449642433875,
"tokens_seen": 361168896
},
{
"epoch": 0.11,
"learning_rate": 0.0004497693079237713,
"loss": 2.7841,
"theoretical_loss": 4.058365842631137,
"tokens_seen": 361234432
},
{
"epoch": 0.11,
"learning_rate": 0.0004497592778335005,
"loss": 2.808,
"theoretical_loss": 4.05828206228618,
"tokens_seen": 361299968
},
{
"epoch": 0.11,
"learning_rate": 0.0004497492477432297,
"loss": 2.8638,
"theoretical_loss": 4.058198301390959,
"tokens_seen": 361365504
},
{
"epoch": 0.11,
"learning_rate": 0.0004497392176529589,
"loss": 2.7483,
"theoretical_loss": 4.058114559937433,
"tokens_seen": 361431040
},
{
"epoch": 0.11,
"learning_rate": 0.00044972918756268804,
"loss": 2.8876,
"theoretical_loss": 4.058030837917565,
"tokens_seen": 361496576
},
{
"epoch": 0.11,
"learning_rate": 0.0004497191574724173,
"loss": 2.8564,
"theoretical_loss": 4.057947135323323,
"tokens_seen": 361562112
},
{
"epoch": 0.11,
"learning_rate": 0.0004497091273821464,
"loss": 3.2562,
"theoretical_loss": 4.057863452146679,
"tokens_seen": 361627648
},
{
"epoch": 0.11,
"learning_rate": 0.00044969909729187564,
"loss": 2.8309,
"theoretical_loss": 4.057779788379613,
"tokens_seen": 361693184
},
{
"epoch": 0.11,
"learning_rate": 0.0004496890672016049,
"loss": 2.8252,
"theoretical_loss": 4.057696144014106,
"tokens_seen": 361758720
},
{
"epoch": 0.11,
"learning_rate": 0.000449679037111334,
"loss": 3.1268,
"theoretical_loss": 4.057612519042145,
"tokens_seen": 361824256
},
{
"epoch": 0.11,
"learning_rate": 0.00044966900702106324,
"loss": 2.7978,
"theoretical_loss": 4.057528913455723,
"tokens_seen": 361889792
},
{
"epoch": 0.11,
"learning_rate": 0.00044965897693079236,
"loss": 2.8384,
"theoretical_loss": 4.057445327246836,
"tokens_seen": 361955328
},
{
"epoch": 0.11,
"learning_rate": 0.0004496489468405216,
"loss": 2.6464,
"theoretical_loss": 4.0573617604074865,
"tokens_seen": 362020864
},
{
"epoch": 0.11,
"objective/train/docs_used": 216313,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2176437377929688,
"objective/train/theoretical_loss": 4.0572782129296785,
"objective/train/tokens_used": 382546400,
"theoretical_loss": 4.0572782129296785,
"tokens_seen": 362086400
},
{
"epoch": 0.11,
"learning_rate": 0.0004496389167502508,
"loss": 3.0372,
"theoretical_loss": 4.0572782129296785,
"tokens_seen": 362086400
},
{
"epoch": 0.11,
"learning_rate": 0.00044962888665997996,
"loss": 2.9452,
"theoretical_loss": 4.057194684805426,
"tokens_seen": 362151936
},
{
"epoch": 0.11,
"learning_rate": 0.00044961885656970914,
"loss": 3.0605,
"theoretical_loss": 4.057111176026742,
"tokens_seen": 362217472
},
{
"epoch": 0.11,
"learning_rate": 0.0004496088264794383,
"loss": 2.9695,
"theoretical_loss": 4.0570276865856485,
"tokens_seen": 362283008
},
{
"epoch": 0.11,
"learning_rate": 0.0004495987963891675,
"loss": 2.8051,
"theoretical_loss": 4.056944216474171,
"tokens_seen": 362348544
},
{
"epoch": 0.11,
"learning_rate": 0.00044958876629889674,
"loss": 3.0178,
"theoretical_loss": 4.056860765684338,
"tokens_seen": 362414080
},
{
"epoch": 0.11,
"learning_rate": 0.00044957873620862587,
"loss": 3.0671,
"theoretical_loss": 4.056777334208186,
"tokens_seen": 362479616
},
{
"epoch": 0.11,
"learning_rate": 0.0004495687061183551,
"loss": 2.9791,
"theoretical_loss": 4.056693922037753,
"tokens_seen": 362545152
},
{
"epoch": 0.11,
"learning_rate": 0.0004495586760280843,
"loss": 2.8048,
"theoretical_loss": 4.056610529165083,
"tokens_seen": 362610688
},
{
"epoch": 0.11,
"learning_rate": 0.00044954864593781346,
"loss": 3.1233,
"theoretical_loss": 4.056527155582226,
"tokens_seen": 362676224
},
{
"epoch": 0.11,
"learning_rate": 0.00044953861584754264,
"loss": 2.8641,
"theoretical_loss": 4.056443801281235,
"tokens_seen": 362741760
},
{
"epoch": 0.11,
"learning_rate": 0.0004495285857572718,
"loss": 2.8788,
"theoretical_loss": 4.056360466254168,
"tokens_seen": 362807296
},
{
"epoch": 0.11,
"learning_rate": 0.000449518555667001,
"loss": 2.911,
"theoretical_loss": 4.056277150493088,
"tokens_seen": 362872832
},
{
"epoch": 0.11,
"learning_rate": 0.00044950852557673024,
"loss": 2.8492,
"theoretical_loss": 4.056193853990063,
"tokens_seen": 362938368
},
{
"epoch": 0.11,
"learning_rate": 0.00044949849548645937,
"loss": 3.0162,
"theoretical_loss": 4.056110576737166,
"tokens_seen": 363003904
},
{
"epoch": 0.11,
"learning_rate": 0.0004494884653961886,
"loss": 3.2349,
"theoretical_loss": 4.056027318726472,
"tokens_seen": 363069440
},
{
"epoch": 0.11,
"learning_rate": 0.00044947843530591773,
"loss": 2.8015,
"theoretical_loss": 4.055944079950065,
"tokens_seen": 363134976
},
{
"epoch": 0.11,
"learning_rate": 0.00044946840521564697,
"loss": 2.9836,
"theoretical_loss": 4.0558608604000295,
"tokens_seen": 363200512
},
{
"epoch": 0.11,
"learning_rate": 0.00044945837512537615,
"loss": 2.956,
"theoretical_loss": 4.055777660068458,
"tokens_seen": 363266048
},
{
"epoch": 0.11,
"learning_rate": 0.00044944834503510533,
"loss": 2.9963,
"theoretical_loss": 4.055694478947445,
"tokens_seen": 363331584
},
{
"epoch": 0.11,
"learning_rate": 0.0004494383149448345,
"loss": 2.8188,
"theoretical_loss": 4.055611317029093,
"tokens_seen": 363397120
},
{
"epoch": 0.11,
"learning_rate": 0.00044942828485456375,
"loss": 2.9132,
"theoretical_loss": 4.055528174305505,
"tokens_seen": 363462656
},
{
"epoch": 0.11,
"learning_rate": 0.00044941825476429287,
"loss": 2.8309,
"theoretical_loss": 4.055445050768792,
"tokens_seen": 363528192
},
{
"epoch": 0.11,
"learning_rate": 0.0004494082246740221,
"loss": 3.0976,
"theoretical_loss": 4.055361946411068,
"tokens_seen": 363593728
},
{
"epoch": 0.11,
"learning_rate": 0.00044939819458375123,
"loss": 2.7774,
"theoretical_loss": 4.055278861224452,
"tokens_seen": 363659264
},
{
"epoch": 0.11,
"objective/train/docs_used": 217479,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.15562105178833,
"objective/train/theoretical_loss": 4.055195795201069,
"objective/train/tokens_used": 384184800,
"theoretical_loss": 4.055195795201069,
"tokens_seen": 363724800
},
{
"epoch": 0.11,
"learning_rate": 0.00044938816449348047,
"loss": 2.9052,
"theoretical_loss": 4.055195795201069,
"tokens_seen": 363724800
},
{
"epoch": 0.11,
"learning_rate": 0.00044937813440320965,
"loss": 3.0289,
"theoretical_loss": 4.055112748333046,
"tokens_seen": 363790336
},
{
"epoch": 0.11,
"learning_rate": 0.00044936810431293883,
"loss": 3.0627,
"theoretical_loss": 4.0550297206125165,
"tokens_seen": 363855872
},
{
"epoch": 0.11,
"learning_rate": 0.000449358074222668,
"loss": 2.8629,
"theoretical_loss": 4.054946712031619,
"tokens_seen": 363921408
},
{
"epoch": 0.11,
"learning_rate": 0.0004493480441323972,
"loss": 3.1226,
"theoretical_loss": 4.054863722582494,
"tokens_seen": 363986944
},
{
"epoch": 0.11,
"learning_rate": 0.0004493380140421264,
"loss": 2.8903,
"theoretical_loss": 4.054780752257291,
"tokens_seen": 364052480
},
{
"epoch": 0.11,
"learning_rate": 0.0004493279839518556,
"loss": 2.7772,
"theoretical_loss": 4.05469780104816,
"tokens_seen": 364118016
},
{
"epoch": 0.11,
"learning_rate": 0.00044931795386158474,
"loss": 2.8789,
"theoretical_loss": 4.054614868947258,
"tokens_seen": 364183552
},
{
"epoch": 0.11,
"learning_rate": 0.00044930792377131397,
"loss": 2.9326,
"theoretical_loss": 4.054531955946745,
"tokens_seen": 364249088
},
{
"epoch": 0.11,
"learning_rate": 0.0004492978936810431,
"loss": 2.9398,
"theoretical_loss": 4.054449062038787,
"tokens_seen": 364314624
},
{
"epoch": 0.11,
"learning_rate": 0.00044928786359077234,
"loss": 3.1035,
"theoretical_loss": 4.054366187215556,
"tokens_seen": 364380160
},
{
"epoch": 0.11,
"learning_rate": 0.0004492778335005015,
"loss": 2.9538,
"theoretical_loss": 4.0542833314692235,
"tokens_seen": 364445696
},
{
"epoch": 0.11,
"learning_rate": 0.0004492678034102307,
"loss": 2.7926,
"theoretical_loss": 4.054200494791972,
"tokens_seen": 364511232
},
{
"epoch": 0.11,
"learning_rate": 0.0004492577733199599,
"loss": 3.1084,
"theoretical_loss": 4.054117677175983,
"tokens_seen": 364576768
},
{
"epoch": 0.11,
"learning_rate": 0.0004492477432296891,
"loss": 2.8922,
"theoretical_loss": 4.054034878613447,
"tokens_seen": 364642304
},
{
"epoch": 0.11,
"learning_rate": 0.00044923771313941824,
"loss": 3.0288,
"theoretical_loss": 4.053952099096556,
"tokens_seen": 364707840
},
{
"epoch": 0.11,
"learning_rate": 0.0004492276830491475,
"loss": 3.0087,
"theoretical_loss": 4.053869338617508,
"tokens_seen": 364773376
},
{
"epoch": 0.11,
"learning_rate": 0.0004492176529588766,
"loss": 2.8053,
"theoretical_loss": 4.053786597168506,
"tokens_seen": 364838912
},
{
"epoch": 0.11,
"learning_rate": 0.00044920762286860584,
"loss": 3.0586,
"theoretical_loss": 4.053703874741757,
"tokens_seen": 364904448
},
{
"epoch": 0.11,
"learning_rate": 0.000449197592778335,
"loss": 2.9923,
"theoretical_loss": 4.053621171329472,
"tokens_seen": 364969984
},
{
"epoch": 0.11,
"learning_rate": 0.0004491875626880642,
"loss": 3.195,
"theoretical_loss": 4.053538486923868,
"tokens_seen": 365035520
},
{
"epoch": 0.11,
"learning_rate": 0.0004491775325977934,
"loss": 3.128,
"theoretical_loss": 4.053455821517165,
"tokens_seen": 365101056
},
{
"epoch": 0.11,
"learning_rate": 0.00044916750250752256,
"loss": 3.0488,
"theoretical_loss": 4.053373175101589,
"tokens_seen": 365166592
},
{
"epoch": 0.11,
"learning_rate": 0.00044915747241725174,
"loss": 3.1723,
"theoretical_loss": 4.053290547669369,
"tokens_seen": 365232128
},
{
"epoch": 0.11,
"learning_rate": 0.000449147442326981,
"loss": 3.107,
"theoretical_loss": 4.05320793921274,
"tokens_seen": 365297664
},
{
"epoch": 0.11,
"objective/train/docs_used": 218231,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4497435092926025,
"objective/train/theoretical_loss": 4.053125349723942,
"objective/train/tokens_used": 385823200,
"theoretical_loss": 4.053125349723942,
"tokens_seen": 365363200
},
{
"epoch": 0.11,
"learning_rate": 0.0004491374122367101,
"loss": 2.7664,
"theoretical_loss": 4.053125349723942,
"tokens_seen": 365363200
},
{
"epoch": 0.11,
"learning_rate": 0.00044912738214643934,
"loss": 3.0456,
"theoretical_loss": 4.053042779195217,
"tokens_seen": 365428736
},
{
"epoch": 0.11,
"learning_rate": 0.00044911735205616847,
"loss": 2.9308,
"theoretical_loss": 4.052960227618815,
"tokens_seen": 365494272
},
{
"epoch": 0.11,
"learning_rate": 0.0004491073219658977,
"loss": 3.0695,
"theoretical_loss": 4.052877694986987,
"tokens_seen": 365559808
},
{
"epoch": 0.11,
"learning_rate": 0.0004490972918756269,
"loss": 3.0834,
"theoretical_loss": 4.052795181291992,
"tokens_seen": 365625344
},
{
"epoch": 0.11,
"learning_rate": 0.00044908726178535607,
"loss": 3.0867,
"theoretical_loss": 4.05271268652609,
"tokens_seen": 365690880
},
{
"epoch": 0.11,
"learning_rate": 0.00044907723169508525,
"loss": 3.1497,
"theoretical_loss": 4.052630210681549,
"tokens_seen": 365756416
},
{
"epoch": 0.11,
"learning_rate": 0.0004490672016048145,
"loss": 2.7121,
"theoretical_loss": 4.052547753750639,
"tokens_seen": 365821952
},
{
"epoch": 0.11,
"learning_rate": 0.0004490571715145436,
"loss": 3.0036,
"theoretical_loss": 4.0524653157256365,
"tokens_seen": 365887488
},
{
"epoch": 0.11,
"learning_rate": 0.00044904714142427284,
"loss": 2.9278,
"theoretical_loss": 4.05238289659882,
"tokens_seen": 365953024
},
{
"epoch": 0.11,
"learning_rate": 0.00044903711133400197,
"loss": 2.7647,
"theoretical_loss": 4.052300496362475,
"tokens_seen": 366018560
},
{
"epoch": 0.11,
"learning_rate": 0.0004490270812437312,
"loss": 2.772,
"theoretical_loss": 4.052218115008892,
"tokens_seen": 366084096
},
{
"epoch": 0.11,
"learning_rate": 0.0004490170511534604,
"loss": 2.921,
"theoretical_loss": 4.052135752530362,
"tokens_seen": 366149632
},
{
"epoch": 0.11,
"learning_rate": 0.00044900702106318957,
"loss": 3.1039,
"theoretical_loss": 4.0520534089191855,
"tokens_seen": 366215168
},
{
"epoch": 0.11,
"learning_rate": 0.00044899699097291875,
"loss": 2.7293,
"theoretical_loss": 4.051971084167663,
"tokens_seen": 366280704
},
{
"epoch": 0.11,
"learning_rate": 0.00044898696088264793,
"loss": 2.949,
"theoretical_loss": 4.051888778268103,
"tokens_seen": 366346240
},
{
"epoch": 0.11,
"learning_rate": 0.0004489769307923771,
"loss": 3.0874,
"theoretical_loss": 4.051806491212817,
"tokens_seen": 366411776
},
{
"epoch": 0.11,
"learning_rate": 0.00044896690070210635,
"loss": 2.9753,
"theoretical_loss": 4.051724222994122,
"tokens_seen": 366477312
},
{
"epoch": 0.11,
"learning_rate": 0.0004489568706118355,
"loss": 3.0817,
"theoretical_loss": 4.051641973604337,
"tokens_seen": 366542848
},
{
"epoch": 0.11,
"learning_rate": 0.0004489468405215647,
"loss": 2.9177,
"theoretical_loss": 4.051559743035789,
"tokens_seen": 366608384
},
{
"epoch": 0.11,
"learning_rate": 0.00044893681043129395,
"loss": 3.1013,
"theoretical_loss": 4.051477531280806,
"tokens_seen": 366673920
},
{
"epoch": 0.11,
"learning_rate": 0.00044892678034102307,
"loss": 2.8753,
"theoretical_loss": 4.051395338331723,
"tokens_seen": 366739456
},
{
"epoch": 0.11,
"learning_rate": 0.0004489167502507523,
"loss": 3.0245,
"theoretical_loss": 4.05131316418088,
"tokens_seen": 366804992
},
{
"epoch": 0.11,
"learning_rate": 0.00044890672016048143,
"loss": 2.9561,
"theoretical_loss": 4.051231008820618,
"tokens_seen": 366870528
},
{
"epoch": 0.11,
"learning_rate": 0.00044889669007021067,
"loss": 3.0036,
"theoretical_loss": 4.051148872243286,
"tokens_seen": 366936064
},
{
"epoch": 0.11,
"objective/train/docs_used": 219467,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0995004177093506,
"objective/train/theoretical_loss": 4.051066754441235,
"objective/train/tokens_used": 387461600,
"theoretical_loss": 4.051066754441235,
"tokens_seen": 367001600
},
{
"epoch": 0.11,
"learning_rate": 0.00044888665997993985,
"loss": 3.0026,
"theoretical_loss": 4.051066754441235,
"tokens_seen": 367001600
},
{
"epoch": 0.11,
"learning_rate": 0.00044887662988966903,
"loss": 3.1344,
"theoretical_loss": 4.050984655406824,
"tokens_seen": 367067136
},
{
"epoch": 0.11,
"learning_rate": 0.0004488665997993982,
"loss": 3.0819,
"theoretical_loss": 4.050902575132412,
"tokens_seen": 367132672
},
{
"epoch": 0.11,
"learning_rate": 0.0004488565697091274,
"loss": 3.0175,
"theoretical_loss": 4.050820513610366,
"tokens_seen": 367198208
},
{
"epoch": 0.11,
"learning_rate": 0.0004488465396188566,
"loss": 2.9302,
"theoretical_loss": 4.050738470833054,
"tokens_seen": 367263744
},
{
"epoch": 0.11,
"learning_rate": 0.0004488365095285858,
"loss": 3.0906,
"theoretical_loss": 4.0506564467928525,
"tokens_seen": 367329280
},
{
"epoch": 0.11,
"learning_rate": 0.00044882647943831494,
"loss": 2.7848,
"theoretical_loss": 4.05057444148214,
"tokens_seen": 367394816
},
{
"epoch": 0.11,
"learning_rate": 0.0004488164493480442,
"loss": 2.9642,
"theoretical_loss": 4.0504924548933,
"tokens_seen": 367460352
},
{
"epoch": 0.11,
"learning_rate": 0.0004488064192577733,
"loss": 2.9148,
"theoretical_loss": 4.05041048701872,
"tokens_seen": 367525888
},
{
"epoch": 0.11,
"learning_rate": 0.00044879638916750254,
"loss": 2.9993,
"theoretical_loss": 4.050328537850792,
"tokens_seen": 367591424
},
{
"epoch": 0.11,
"learning_rate": 0.0004487863590772317,
"loss": 3.2353,
"theoretical_loss": 4.050246607381913,
"tokens_seen": 367656960
},
{
"epoch": 0.11,
"learning_rate": 0.0004487763289869609,
"loss": 3.096,
"theoretical_loss": 4.0501646956044866,
"tokens_seen": 367722496
},
{
"epoch": 0.11,
"learning_rate": 0.0004487662988966901,
"loss": 2.7814,
"theoretical_loss": 4.050082802510915,
"tokens_seen": 367788032
},
{
"epoch": 0.11,
"learning_rate": 0.0004487562688064193,
"loss": 3.0983,
"theoretical_loss": 4.050000928093611,
"tokens_seen": 367853568
},
{
"epoch": 0.11,
"learning_rate": 0.00044874623871614844,
"loss": 3.2659,
"theoretical_loss": 4.049919072344987,
"tokens_seen": 367919104
},
{
"epoch": 0.11,
"learning_rate": 0.0004487362086258777,
"loss": 3.0178,
"theoretical_loss": 4.049837235257463,
"tokens_seen": 367984640
},
{
"epoch": 0.11,
"learning_rate": 0.0004487261785356068,
"loss": 2.9854,
"theoretical_loss": 4.049755416823463,
"tokens_seen": 368050176
},
{
"epoch": 0.11,
"learning_rate": 0.00044871614844533604,
"loss": 2.9565,
"theoretical_loss": 4.049673617035414,
"tokens_seen": 368115712
},
{
"epoch": 0.11,
"learning_rate": 0.0004487061183550652,
"loss": 2.9942,
"theoretical_loss": 4.0495918358857494,
"tokens_seen": 368181248
},
{
"epoch": 0.11,
"learning_rate": 0.0004486960882647944,
"loss": 3.1321,
"theoretical_loss": 4.049510073366903,
"tokens_seen": 368246784
},
{
"epoch": 0.11,
"learning_rate": 0.0004486860581745236,
"loss": 3.0015,
"theoretical_loss": 4.04942832947132,
"tokens_seen": 368312320
},
{
"epoch": 0.11,
"learning_rate": 0.00044867602808425276,
"loss": 2.8571,
"theoretical_loss": 4.049346604191443,
"tokens_seen": 368377856
},
{
"epoch": 0.11,
"learning_rate": 0.00044866599799398194,
"loss": 2.8144,
"theoretical_loss": 4.049264897519723,
"tokens_seen": 368443392
},
{
"epoch": 0.11,
"learning_rate": 0.0004486559679037112,
"loss": 3.0359,
"theoretical_loss": 4.049183209448615,
"tokens_seen": 368508928
},
{
"epoch": 0.11,
"learning_rate": 0.0004486459378134403,
"loss": 3.0111,
"theoretical_loss": 4.049101539970576,
"tokens_seen": 368574464
},
{
"epoch": 0.11,
"objective/train/docs_used": 220098,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9974594116210938,
"objective/train/theoretical_loss": 4.04901988907807,
"objective/train/tokens_used": 389100000,
"theoretical_loss": 4.04901988907807,
"tokens_seen": 368640000
},
{
"epoch": 0.11,
"learning_rate": 0.00044863590772316954,
"loss": 2.9455,
"theoretical_loss": 4.04901988907807,
"tokens_seen": 368640000
},
{
"epoch": 0.11,
"learning_rate": 0.00044862587763289867,
"loss": 3.1135,
"theoretical_loss": 4.048938256763565,
"tokens_seen": 368705536
},
{
"epoch": 0.11,
"learning_rate": 0.0004486158475426279,
"loss": 2.8557,
"theoretical_loss": 4.048856643019532,
"tokens_seen": 368771072
},
{
"epoch": 0.11,
"learning_rate": 0.0004486058174523571,
"loss": 2.893,
"theoretical_loss": 4.048775047838447,
"tokens_seen": 368836608
},
{
"epoch": 0.11,
"learning_rate": 0.00044859578736208627,
"loss": 3.0628,
"theoretical_loss": 4.048693471212792,
"tokens_seen": 368902144
},
{
"epoch": 0.11,
"learning_rate": 0.00044858575727181545,
"loss": 2.9959,
"theoretical_loss": 4.048611913135052,
"tokens_seen": 368967680
},
{
"epoch": 0.11,
"learning_rate": 0.0004485757271815447,
"loss": 3.0398,
"theoretical_loss": 4.048530373597716,
"tokens_seen": 369033216
},
{
"epoch": 0.11,
"learning_rate": 0.0004485656970912738,
"loss": 3.0683,
"theoretical_loss": 4.048448852593277,
"tokens_seen": 369098752
},
{
"epoch": 0.11,
"learning_rate": 0.00044855566700100304,
"loss": 2.8019,
"theoretical_loss": 4.048367350114235,
"tokens_seen": 369164288
},
{
"epoch": 0.11,
"learning_rate": 0.00044854563691073217,
"loss": 3.3078,
"theoretical_loss": 4.0482858661530905,
"tokens_seen": 369229824
},
{
"epoch": 0.11,
"learning_rate": 0.0004485356068204614,
"loss": 2.8911,
"theoretical_loss": 4.048204400702351,
"tokens_seen": 369295360
},
{
"epoch": 0.11,
"learning_rate": 0.0004485255767301906,
"loss": 2.9811,
"theoretical_loss": 4.04812295375453,
"tokens_seen": 369360896
},
{
"epoch": 0.11,
"learning_rate": 0.00044851554663991977,
"loss": 3.0611,
"theoretical_loss": 4.04804152530214,
"tokens_seen": 369426432
},
{
"epoch": 0.11,
"learning_rate": 0.00044850551654964895,
"loss": 2.666,
"theoretical_loss": 4.047960115337702,
"tokens_seen": 369491968
},
{
"epoch": 0.11,
"learning_rate": 0.00044849548645937813,
"loss": 2.9621,
"theoretical_loss": 4.047878723853741,
"tokens_seen": 369557504
},
{
"epoch": 0.11,
"learning_rate": 0.0004484854563691073,
"loss": 3.1056,
"theoretical_loss": 4.047797350842786,
"tokens_seen": 369623040
},
{
"epoch": 0.11,
"learning_rate": 0.00044847542627883655,
"loss": 2.9818,
"theoretical_loss": 4.047715996297369,
"tokens_seen": 369688576
},
{
"epoch": 0.11,
"learning_rate": 0.0004484653961885657,
"loss": 3.1221,
"theoretical_loss": 4.0476346602100275,
"tokens_seen": 369754112
},
{
"epoch": 0.11,
"learning_rate": 0.0004484553660982949,
"loss": 3.1221,
"theoretical_loss": 4.047553342573304,
"tokens_seen": 369819648
},
{
"epoch": 0.11,
"learning_rate": 0.00044844533600802404,
"loss": 2.7862,
"theoretical_loss": 4.047472043379743,
"tokens_seen": 369885184
},
{
"epoch": 0.11,
"learning_rate": 0.00044843530591775327,
"loss": 2.8429,
"theoretical_loss": 4.047390762621897,
"tokens_seen": 369950720
},
{
"epoch": 0.11,
"learning_rate": 0.00044842527582748245,
"loss": 3.0292,
"theoretical_loss": 4.04730950029232,
"tokens_seen": 370016256
},
{
"epoch": 0.11,
"learning_rate": 0.00044841524573721163,
"loss": 2.6929,
"theoretical_loss": 4.047228256383569,
"tokens_seen": 370081792
},
{
"epoch": 0.11,
"learning_rate": 0.0004484052156469408,
"loss": 3.1469,
"theoretical_loss": 4.04714703088821,
"tokens_seen": 370147328
},
{
"epoch": 0.11,
"learning_rate": 0.00044839518555667005,
"loss": 3.1589,
"theoretical_loss": 4.0470658237988095,
"tokens_seen": 370212864
},
{
"epoch": 0.11,
"objective/train/docs_used": 221122,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.944498300552368,
"objective/train/theoretical_loss": 4.04698463510794,
"objective/train/tokens_used": 390738400,
"theoretical_loss": 4.04698463510794,
"tokens_seen": 370278400
},
{
"epoch": 0.11,
"learning_rate": 0.0004483851554663992,
"loss": 2.9615,
"theoretical_loss": 4.04698463510794,
"tokens_seen": 370278400
},
{
"epoch": 0.11,
"learning_rate": 0.0004483751253761284,
"loss": 2.8592,
"theoretical_loss": 4.046903464808177,
"tokens_seen": 370343936
},
{
"epoch": 0.11,
"learning_rate": 0.00044836509528585754,
"loss": 2.9014,
"theoretical_loss": 4.046822312892102,
"tokens_seen": 370409472
},
{
"epoch": 0.11,
"learning_rate": 0.0004483550651955868,
"loss": 2.9322,
"theoretical_loss": 4.046741179352299,
"tokens_seen": 370475008
},
{
"epoch": 0.11,
"learning_rate": 0.00044834503510531596,
"loss": 2.9574,
"theoretical_loss": 4.046660064181359,
"tokens_seen": 370540544
},
{
"epoch": 0.11,
"learning_rate": 0.00044833500501504514,
"loss": 3.1144,
"theoretical_loss": 4.0465789673718735,
"tokens_seen": 370606080
},
{
"epoch": 0.11,
"learning_rate": 0.0004483249749247743,
"loss": 3.0474,
"theoretical_loss": 4.0464978889164405,
"tokens_seen": 370671616
},
{
"epoch": 0.11,
"learning_rate": 0.0004483149448345035,
"loss": 2.9841,
"theoretical_loss": 4.046416828807663,
"tokens_seen": 370737152
},
{
"epoch": 0.11,
"learning_rate": 0.0004483049147442327,
"loss": 2.9314,
"theoretical_loss": 4.046335787038148,
"tokens_seen": 370802688
},
{
"epoch": 0.11,
"learning_rate": 0.0004482948846539619,
"loss": 3.1382,
"theoretical_loss": 4.046254763600503,
"tokens_seen": 370868224
},
{
"epoch": 0.11,
"learning_rate": 0.00044828485456369104,
"loss": 2.9275,
"theoretical_loss": 4.046173758487347,
"tokens_seen": 370933760
},
{
"epoch": 0.11,
"learning_rate": 0.0004482748244734203,
"loss": 3.0695,
"theoretical_loss": 4.046092771691297,
"tokens_seen": 370999296
},
{
"epoch": 0.11,
"learning_rate": 0.0004482647943831494,
"loss": 3.0102,
"theoretical_loss": 4.046011803204976,
"tokens_seen": 371064832
},
{
"epoch": 0.11,
"learning_rate": 0.00044825476429287864,
"loss": 2.9351,
"theoretical_loss": 4.045930853021014,
"tokens_seen": 371130368
},
{
"epoch": 0.11,
"learning_rate": 0.0004482447342026078,
"loss": 3.0805,
"theoretical_loss": 4.045849921132042,
"tokens_seen": 371195904
},
{
"epoch": 0.11,
"learning_rate": 0.000448234704112337,
"loss": 2.8759,
"theoretical_loss": 4.045769007530695,
"tokens_seen": 371261440
},
{
"epoch": 0.11,
"learning_rate": 0.0004482246740220662,
"loss": 3.2063,
"theoretical_loss": 4.045688112209615,
"tokens_seen": 371326976
},
{
"epoch": 0.11,
"learning_rate": 0.0004482146439317954,
"loss": 2.9021,
"theoretical_loss": 4.045607235161447,
"tokens_seen": 371392512
},
{
"epoch": 0.11,
"learning_rate": 0.00044820461384152455,
"loss": 2.8611,
"theoretical_loss": 4.04552637637884,
"tokens_seen": 371458048
},
{
"epoch": 0.11,
"learning_rate": 0.0004481945837512538,
"loss": 3.1244,
"theoretical_loss": 4.045445535854446,
"tokens_seen": 371523584
},
{
"epoch": 0.11,
"learning_rate": 0.00044818455366098296,
"loss": 2.9103,
"theoretical_loss": 4.045364713580923,
"tokens_seen": 371589120
},
{
"epoch": 0.11,
"learning_rate": 0.00044817452357071214,
"loss": 2.9705,
"theoretical_loss": 4.045283909550935,
"tokens_seen": 371654656
},
{
"epoch": 0.11,
"learning_rate": 0.0004481644934804414,
"loss": 3.0269,
"theoretical_loss": 4.045203123757146,
"tokens_seen": 371720192
},
{
"epoch": 0.11,
"learning_rate": 0.0004481544633901705,
"loss": 3.0925,
"theoretical_loss": 4.045122356192227,
"tokens_seen": 371785728
},
{
"epoch": 0.11,
"learning_rate": 0.00044814443329989974,
"loss": 3.0519,
"theoretical_loss": 4.0450416068488515,
"tokens_seen": 371851264
},
{
"epoch": 0.11,
"objective/train/docs_used": 221664,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7094762325286865,
"objective/train/theoretical_loss": 4.044960875719701,
"objective/train/tokens_used": 392376800,
"theoretical_loss": 4.044960875719701,
"tokens_seen": 371916800
},
{
"epoch": 0.11,
"learning_rate": 0.00044813440320962887,
"loss": 2.7159,
"theoretical_loss": 4.044960875719701,
"tokens_seen": 371916800
},
{
"epoch": 0.11,
"learning_rate": 0.0004481243731193581,
"loss": 2.9937,
"theoretical_loss": 4.0448801627974555,
"tokens_seen": 371982336
},
{
"epoch": 0.11,
"learning_rate": 0.0004481143430290873,
"loss": 3.0136,
"theoretical_loss": 4.044799468074804,
"tokens_seen": 372047872
},
{
"epoch": 0.11,
"learning_rate": 0.00044810431293881647,
"loss": 3.2027,
"theoretical_loss": 4.044718791544438,
"tokens_seen": 372113408
},
{
"epoch": 0.11,
"learning_rate": 0.00044809428284854565,
"loss": 3.171,
"theoretical_loss": 4.044638133199052,
"tokens_seen": 372178944
},
{
"epoch": 0.11,
"learning_rate": 0.0004480842527582749,
"loss": 2.9597,
"theoretical_loss": 4.044557493031347,
"tokens_seen": 372244480
},
{
"epoch": 0.11,
"learning_rate": 0.000448074222668004,
"loss": 2.9182,
"theoretical_loss": 4.044476871034027,
"tokens_seen": 372310016
},
{
"epoch": 0.11,
"learning_rate": 0.00044806419257773324,
"loss": 3.0117,
"theoretical_loss": 4.0443962671998,
"tokens_seen": 372375552
},
{
"epoch": 0.11,
"learning_rate": 0.00044805416248746237,
"loss": 2.7842,
"theoretical_loss": 4.044315681521379,
"tokens_seen": 372441088
},
{
"epoch": 0.11,
"learning_rate": 0.0004480441323971916,
"loss": 2.9061,
"theoretical_loss": 4.04423511399148,
"tokens_seen": 372506624
},
{
"epoch": 0.11,
"learning_rate": 0.0004480341023069208,
"loss": 2.7469,
"theoretical_loss": 4.044154564602826,
"tokens_seen": 372572160
},
{
"epoch": 0.11,
"learning_rate": 0.00044802407221664997,
"loss": 2.9115,
"theoretical_loss": 4.044074033348139,
"tokens_seen": 372637696
},
{
"epoch": 0.11,
"learning_rate": 0.00044801404212637915,
"loss": 3.0106,
"theoretical_loss": 4.043993520220151,
"tokens_seen": 372703232
},
{
"epoch": 0.11,
"learning_rate": 0.00044800401203610833,
"loss": 3.2011,
"theoretical_loss": 4.0439130252115945,
"tokens_seen": 372768768
},
{
"epoch": 0.11,
"learning_rate": 0.0004479939819458375,
"loss": 3.1417,
"theoretical_loss": 4.043832548315207,
"tokens_seen": 372834304
},
{
"epoch": 0.11,
"learning_rate": 0.00044798395185556675,
"loss": 3.0064,
"theoretical_loss": 4.043752089523732,
"tokens_seen": 372899840
},
{
"epoch": 0.11,
"learning_rate": 0.0004479739217652959,
"loss": 3.0273,
"theoretical_loss": 4.043671648829913,
"tokens_seen": 372965376
},
{
"epoch": 0.11,
"learning_rate": 0.0004479638916750251,
"loss": 2.785,
"theoretical_loss": 4.043591226226503,
"tokens_seen": 373030912
},
{
"epoch": 0.11,
"learning_rate": 0.00044795386158475424,
"loss": 2.958,
"theoretical_loss": 4.0435108217062545,
"tokens_seen": 373096448
},
{
"epoch": 0.11,
"learning_rate": 0.00044794383149448347,
"loss": 3.1277,
"theoretical_loss": 4.043430435261927,
"tokens_seen": 373161984
},
{
"epoch": 0.11,
"learning_rate": 0.00044793380140421265,
"loss": 3.2189,
"theoretical_loss": 4.043350066886283,
"tokens_seen": 373227520
},
{
"epoch": 0.11,
"learning_rate": 0.00044792377131394183,
"loss": 2.8955,
"theoretical_loss": 4.043269716572091,
"tokens_seen": 373293056
},
{
"epoch": 0.11,
"learning_rate": 0.000447913741223671,
"loss": 2.713,
"theoretical_loss": 4.04318938431212,
"tokens_seen": 373358592
},
{
"epoch": 0.11,
"learning_rate": 0.00044790371113340025,
"loss": 3.1045,
"theoretical_loss": 4.043109070099146,
"tokens_seen": 373424128
},
{
"epoch": 0.11,
"learning_rate": 0.0004478936810431294,
"loss": 3.053,
"theoretical_loss": 4.043028773925949,
"tokens_seen": 373489664
},
{
"epoch": 0.11,
"objective/train/docs_used": 222906,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8282113075256348,
"objective/train/theoretical_loss": 4.042948495785312,
"objective/train/tokens_used": 394015200,
"theoretical_loss": 4.042948495785312,
"tokens_seen": 373555200
},
{
"epoch": 0.11,
"learning_rate": 0.0004478836509528586,
"loss": 2.8803,
"theoretical_loss": 4.042948495785312,
"tokens_seen": 373555200
},
{
"epoch": 0.11,
"learning_rate": 0.00044787362086258774,
"loss": 3.0613,
"theoretical_loss": 4.042868235670023,
"tokens_seen": 373620736
},
{
"epoch": 0.11,
"learning_rate": 0.000447863590772317,
"loss": 2.8243,
"theoretical_loss": 4.042787993572875,
"tokens_seen": 373686272
},
{
"epoch": 0.11,
"learning_rate": 0.00044785356068204616,
"loss": 3.0637,
"theoretical_loss": 4.042707769486663,
"tokens_seen": 373751808
},
{
"epoch": 0.11,
"learning_rate": 0.00044784353059177534,
"loss": 3.0711,
"theoretical_loss": 4.042627563404187,
"tokens_seen": 373817344
},
{
"epoch": 0.11,
"learning_rate": 0.0004478335005015045,
"loss": 2.85,
"theoretical_loss": 4.042547375318251,
"tokens_seen": 373882880
},
{
"epoch": 0.11,
"learning_rate": 0.0004478234704112337,
"loss": 2.9246,
"theoretical_loss": 4.042467205221665,
"tokens_seen": 373948416
},
{
"epoch": 0.11,
"learning_rate": 0.0004478134403209629,
"loss": 3.1001,
"theoretical_loss": 4.042387053107241,
"tokens_seen": 374013952
},
{
"epoch": 0.11,
"learning_rate": 0.0004478034102306921,
"loss": 2.8072,
"theoretical_loss": 4.042306918967795,
"tokens_seen": 374079488
},
{
"epoch": 0.11,
"learning_rate": 0.00044779338014042124,
"loss": 2.8558,
"theoretical_loss": 4.042226802796149,
"tokens_seen": 374145024
},
{
"epoch": 0.11,
"learning_rate": 0.0004477833500501505,
"loss": 2.8778,
"theoretical_loss": 4.042146704585127,
"tokens_seen": 374210560
},
{
"epoch": 0.11,
"learning_rate": 0.0004477733199598796,
"loss": 2.969,
"theoretical_loss": 4.042066624327559,
"tokens_seen": 374276096
},
{
"epoch": 0.11,
"learning_rate": 0.00044776328986960884,
"loss": 3.0689,
"theoretical_loss": 4.041986562016278,
"tokens_seen": 374341632
},
{
"epoch": 0.11,
"learning_rate": 0.000447753259779338,
"loss": 2.9921,
"theoretical_loss": 4.041906517644121,
"tokens_seen": 374407168
},
{
"epoch": 0.11,
"learning_rate": 0.0004477432296890672,
"loss": 2.9593,
"theoretical_loss": 4.0418264912039295,
"tokens_seen": 374472704
},
{
"epoch": 0.11,
"learning_rate": 0.0004477331995987964,
"loss": 2.8642,
"theoretical_loss": 4.04174648268855,
"tokens_seen": 374538240
},
{
"epoch": 0.11,
"learning_rate": 0.0004477231695085256,
"loss": 2.8123,
"theoretical_loss": 4.041666492090831,
"tokens_seen": 374603776
},
{
"epoch": 0.11,
"learning_rate": 0.00044771313941825475,
"loss": 2.7568,
"theoretical_loss": 4.041586519403627,
"tokens_seen": 374669312
},
{
"epoch": 0.11,
"learning_rate": 0.000447703109327984,
"loss": 2.885,
"theoretical_loss": 4.041506564619796,
"tokens_seen": 374734848
},
{
"epoch": 0.11,
"learning_rate": 0.0004476930792377131,
"loss": 2.7418,
"theoretical_loss": 4.0414266277322,
"tokens_seen": 374800384
},
{
"epoch": 0.11,
"learning_rate": 0.00044768304914744234,
"loss": 3.0767,
"theoretical_loss": 4.041346708733704,
"tokens_seen": 374865920
},
{
"epoch": 0.11,
"learning_rate": 0.0004476730190571715,
"loss": 3.0252,
"theoretical_loss": 4.041266807617179,
"tokens_seen": 374931456
},
{
"epoch": 0.11,
"learning_rate": 0.0004476629889669007,
"loss": 2.6955,
"theoretical_loss": 4.041186924375499,
"tokens_seen": 374996992
},
{
"epoch": 0.11,
"learning_rate": 0.0004476529588766299,
"loss": 2.6624,
"theoretical_loss": 4.041107059001544,
"tokens_seen": 375062528
},
{
"epoch": 0.11,
"learning_rate": 0.00044764292878635907,
"loss": 3.0584,
"theoretical_loss": 4.041027211488194,
"tokens_seen": 375128064
},
{
"epoch": 0.11,
"objective/train/docs_used": 223164,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.815418004989624,
"objective/train/theoretical_loss": 4.040947381828337,
"objective/train/tokens_used": 395653600,
"theoretical_loss": 4.040947381828337,
"tokens_seen": 375193600
},
{
"epoch": 0.11,
"learning_rate": 0.00044763289869608825,
"loss": 3.0237,
"theoretical_loss": 4.040947381828337,
"tokens_seen": 375193600
},
{
"epoch": 0.11,
"learning_rate": 0.0004476228686058175,
"loss": 2.8633,
"theoretical_loss": 4.040867570014862,
"tokens_seen": 375259136
},
{
"epoch": 0.11,
"learning_rate": 0.0004476128385155466,
"loss": 3.0381,
"theoretical_loss": 4.040787776040666,
"tokens_seen": 375324672
},
{
"epoch": 0.11,
"learning_rate": 0.00044760280842527585,
"loss": 2.9233,
"theoretical_loss": 4.040707999898647,
"tokens_seen": 375390208
},
{
"epoch": 0.11,
"learning_rate": 0.00044759277833500503,
"loss": 2.9864,
"theoretical_loss": 4.040628241581706,
"tokens_seen": 375455744
},
{
"epoch": 0.11,
"learning_rate": 0.0004475827482447342,
"loss": 2.9743,
"theoretical_loss": 4.0405485010827515,
"tokens_seen": 375521280
},
{
"epoch": 0.11,
"learning_rate": 0.0004475727181544634,
"loss": 3.1656,
"theoretical_loss": 4.040468778394694,
"tokens_seen": 375586816
},
{
"epoch": 0.11,
"learning_rate": 0.00044756268806419257,
"loss": 2.7738,
"theoretical_loss": 4.040389073510449,
"tokens_seen": 375652352
},
{
"epoch": 0.11,
"learning_rate": 0.00044755265797392175,
"loss": 3.043,
"theoretical_loss": 4.040309386422934,
"tokens_seen": 375717888
},
{
"epoch": 0.11,
"learning_rate": 0.000447542627883651,
"loss": 3.0278,
"theoretical_loss": 4.040229717125074,
"tokens_seen": 375783424
},
{
"epoch": 0.11,
"learning_rate": 0.0004475325977933801,
"loss": 3.1116,
"theoretical_loss": 4.0401500656097955,
"tokens_seen": 375848960
},
{
"epoch": 0.11,
"learning_rate": 0.00044752256770310935,
"loss": 3.0037,
"theoretical_loss": 4.040070431870029,
"tokens_seen": 375914496
},
{
"epoch": 0.11,
"learning_rate": 0.0004475125376128385,
"loss": 3.136,
"theoretical_loss": 4.03999081589871,
"tokens_seen": 375980032
},
{
"epoch": 0.11,
"learning_rate": 0.0004475025075225677,
"loss": 3.1902,
"theoretical_loss": 4.039911217688777,
"tokens_seen": 376045568
},
{
"epoch": 0.11,
"learning_rate": 0.0004474924774322969,
"loss": 2.9869,
"theoretical_loss": 4.039831637233176,
"tokens_seen": 376111104
},
{
"epoch": 0.11,
"learning_rate": 0.0004474824473420261,
"loss": 2.8772,
"theoretical_loss": 4.039752074524851,
"tokens_seen": 376176640
},
{
"epoch": 0.11,
"learning_rate": 0.00044747241725175526,
"loss": 2.94,
"theoretical_loss": 4.039672529556755,
"tokens_seen": 376242176
},
{
"epoch": 0.11,
"learning_rate": 0.00044746238716148444,
"loss": 2.9516,
"theoretical_loss": 4.039593002321844,
"tokens_seen": 376307712
},
{
"epoch": 0.11,
"learning_rate": 0.0004474523570712136,
"loss": 2.8799,
"theoretical_loss": 4.039513492813076,
"tokens_seen": 376373248
},
{
"epoch": 0.11,
"learning_rate": 0.00044744232698094285,
"loss": 2.6616,
"theoretical_loss": 4.039434001023415,
"tokens_seen": 376438784
},
{
"epoch": 0.11,
"learning_rate": 0.00044743229689067204,
"loss": 2.8112,
"theoretical_loss": 4.039354526945829,
"tokens_seen": 376504320
},
{
"epoch": 0.11,
"learning_rate": 0.0004474222668004012,
"loss": 2.8584,
"theoretical_loss": 4.039275070573288,
"tokens_seen": 376569856
},
{
"epoch": 0.11,
"learning_rate": 0.00044741223671013045,
"loss": 3.1688,
"theoretical_loss": 4.039195631898768,
"tokens_seen": 376635392
},
{
"epoch": 0.11,
"learning_rate": 0.0004474022066198596,
"loss": 3.0963,
"theoretical_loss": 4.03911621091525,
"tokens_seen": 376700928
},
{
"epoch": 0.11,
"learning_rate": 0.0004473921765295888,
"loss": 3.2786,
"theoretical_loss": 4.039036807615716,
"tokens_seen": 376766464
},
{
"epoch": 0.11,
"objective/train/docs_used": 224842,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6164164543151855,
"objective/train/theoretical_loss": 4.038957421993153,
"objective/train/tokens_used": 397292000,
"theoretical_loss": 4.038957421993153,
"tokens_seen": 376832000
},
{
"epoch": 0.11,
"learning_rate": 0.00044738214643931794,
"loss": 3.0128,
"theoretical_loss": 4.038957421993153,
"tokens_seen": 376832000
},
{
"epoch": 0.11,
"learning_rate": 0.0004473721163490472,
"loss": 2.5977,
"theoretical_loss": 4.038878054040554,
"tokens_seen": 376897536
},
{
"epoch": 0.11,
"learning_rate": 0.00044736208625877636,
"loss": 3.1554,
"theoretical_loss": 4.038798703750913,
"tokens_seen": 376963072
},
{
"epoch": 0.11,
"learning_rate": 0.00044735205616850554,
"loss": 2.8739,
"theoretical_loss": 4.038719371117231,
"tokens_seen": 377028608
},
{
"epoch": 0.11,
"learning_rate": 0.0004473420260782347,
"loss": 2.8454,
"theoretical_loss": 4.038640056132509,
"tokens_seen": 377094144
},
{
"epoch": 0.11,
"learning_rate": 0.0004473319959879639,
"loss": 3.0288,
"theoretical_loss": 4.038560758789758,
"tokens_seen": 377159680
},
{
"epoch": 0.11,
"learning_rate": 0.0004473219658976931,
"loss": 3.0316,
"theoretical_loss": 4.0384814790819865,
"tokens_seen": 377225216
},
{
"epoch": 0.11,
"learning_rate": 0.0004473119358074223,
"loss": 2.9039,
"theoretical_loss": 4.038402217002211,
"tokens_seen": 377290752
},
{
"epoch": 0.11,
"learning_rate": 0.00044730190571715144,
"loss": 2.9524,
"theoretical_loss": 4.038322972543451,
"tokens_seen": 377356288
},
{
"epoch": 0.11,
"learning_rate": 0.0004472918756268807,
"loss": 2.9518,
"theoretical_loss": 4.03824374569873,
"tokens_seen": 377421824
},
{
"epoch": 0.11,
"learning_rate": 0.0004472818455366098,
"loss": 3.0085,
"theoretical_loss": 4.038164536461075,
"tokens_seen": 377487360
},
{
"epoch": 0.11,
"learning_rate": 0.00044727181544633904,
"loss": 3.0309,
"theoretical_loss": 4.038085344823516,
"tokens_seen": 377552896
},
{
"epoch": 0.11,
"learning_rate": 0.0004472617853560682,
"loss": 2.9822,
"theoretical_loss": 4.038006170779091,
"tokens_seen": 377618432
},
{
"epoch": 0.11,
"learning_rate": 0.0004472517552657974,
"loss": 2.8212,
"theoretical_loss": 4.037927014320838,
"tokens_seen": 377683968
},
{
"epoch": 0.11,
"learning_rate": 0.0004472417251755266,
"loss": 2.6607,
"theoretical_loss": 4.037847875441798,
"tokens_seen": 377749504
},
{
"epoch": 0.11,
"learning_rate": 0.0004472316950852558,
"loss": 3.1658,
"theoretical_loss": 4.037768754135023,
"tokens_seen": 377815040
},
{
"epoch": 0.11,
"learning_rate": 0.00044722166499498495,
"loss": 3.0641,
"theoretical_loss": 4.037689650393559,
"tokens_seen": 377880576
},
{
"epoch": 0.11,
"learning_rate": 0.0004472116349047142,
"loss": 3.0602,
"theoretical_loss": 4.037610564210464,
"tokens_seen": 377946112
},
{
"epoch": 0.11,
"learning_rate": 0.0004472016048144433,
"loss": 2.7781,
"theoretical_loss": 4.037531495578797,
"tokens_seen": 378011648
},
{
"epoch": 0.11,
"learning_rate": 0.00044719157472417254,
"loss": 2.911,
"theoretical_loss": 4.03745244449162,
"tokens_seen": 378077184
},
{
"epoch": 0.11,
"learning_rate": 0.0004471815446339017,
"loss": 3.1229,
"theoretical_loss": 4.037373410942001,
"tokens_seen": 378142720
},
{
"epoch": 0.11,
"learning_rate": 0.0004471715145436309,
"loss": 2.8328,
"theoretical_loss": 4.037294394923009,
"tokens_seen": 378208256
},
{
"epoch": 0.11,
"learning_rate": 0.0004471614844533601,
"loss": 3.0088,
"theoretical_loss": 4.037215396427721,
"tokens_seen": 378273792
},
{
"epoch": 0.11,
"learning_rate": 0.00044715145436308927,
"loss": 3.2103,
"theoretical_loss": 4.037136415449214,
"tokens_seen": 378339328
},
{
"epoch": 0.11,
"learning_rate": 0.00044714142427281845,
"loss": 2.897,
"theoretical_loss": 4.037057451980571,
"tokens_seen": 378404864
},
{
"epoch": 0.11,
"objective/train/docs_used": 225570,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.853151321411133,
"objective/train/theoretical_loss": 4.036978506014879,
"objective/train/tokens_used": 398930400,
"theoretical_loss": 4.036978506014879,
"tokens_seen": 378470400
},
{
"epoch": 0.11,
"learning_rate": 0.0004471313941825477,
"loss": 3.0427,
"theoretical_loss": 4.036978506014879,
"tokens_seen": 378470400
},
{
"epoch": 0.11,
"learning_rate": 0.0004471213640922768,
"loss": 2.9945,
"theoretical_loss": 4.036899577545229,
"tokens_seen": 378535936
},
{
"epoch": 0.11,
"learning_rate": 0.00044711133400200605,
"loss": 3.0209,
"theoretical_loss": 4.036820666564714,
"tokens_seen": 378601472
},
{
"epoch": 0.11,
"learning_rate": 0.00044710130391173523,
"loss": 2.7759,
"theoretical_loss": 4.036741773066434,
"tokens_seen": 378667008
},
{
"epoch": 0.11,
"learning_rate": 0.0004470912738214644,
"loss": 3.096,
"theoretical_loss": 4.03666289704349,
"tokens_seen": 378732544
},
{
"epoch": 0.11,
"learning_rate": 0.0004470812437311936,
"loss": 2.958,
"theoretical_loss": 4.036584038488988,
"tokens_seen": 378798080
},
{
"epoch": 0.11,
"learning_rate": 0.00044707121364092277,
"loss": 2.899,
"theoretical_loss": 4.03650519739604,
"tokens_seen": 378863616
},
{
"epoch": 0.11,
"learning_rate": 0.00044706118355065195,
"loss": 3.0908,
"theoretical_loss": 4.036426373757759,
"tokens_seen": 378929152
},
{
"epoch": 0.11,
"learning_rate": 0.0004470511534603812,
"loss": 3.1033,
"theoretical_loss": 4.036347567567262,
"tokens_seen": 378994688
},
{
"epoch": 0.11,
"learning_rate": 0.0004470411233701103,
"loss": 2.6781,
"theoretical_loss": 4.0362687788176705,
"tokens_seen": 379060224
},
{
"epoch": 0.11,
"learning_rate": 0.00044703109327983955,
"loss": 3.2151,
"theoretical_loss": 4.036190007502112,
"tokens_seen": 379125760
},
{
"epoch": 0.11,
"learning_rate": 0.0004470210631895687,
"loss": 2.9318,
"theoretical_loss": 4.036111253613715,
"tokens_seen": 379191296
},
{
"epoch": 0.11,
"learning_rate": 0.0004470110330992979,
"loss": 2.9802,
"theoretical_loss": 4.036032517145614,
"tokens_seen": 379256832
},
{
"epoch": 0.11,
"learning_rate": 0.0004470010030090271,
"loss": 2.8508,
"theoretical_loss": 4.0359537980909455,
"tokens_seen": 379322368
},
{
"epoch": 0.11,
"learning_rate": 0.0004469909729187563,
"loss": 3.1785,
"theoretical_loss": 4.03587509644285,
"tokens_seen": 379387904
},
{
"epoch": 0.11,
"learning_rate": 0.00044698094282848546,
"loss": 3.0789,
"theoretical_loss": 4.0357964121944745,
"tokens_seen": 379453440
},
{
"epoch": 0.12,
"learning_rate": 0.00044697091273821464,
"loss": 2.9847,
"theoretical_loss": 4.035717745338967,
"tokens_seen": 379518976
},
{
"epoch": 0.12,
"learning_rate": 0.0004469608826479438,
"loss": 2.8707,
"theoretical_loss": 4.035639095869481,
"tokens_seen": 379584512
},
{
"epoch": 0.12,
"learning_rate": 0.00044695085255767305,
"loss": 3.1781,
"theoretical_loss": 4.035560463779173,
"tokens_seen": 379650048
},
{
"epoch": 0.12,
"learning_rate": 0.0004469408224674022,
"loss": 3.2176,
"theoretical_loss": 4.0354818490612026,
"tokens_seen": 379715584
},
{
"epoch": 0.12,
"learning_rate": 0.0004469307923771314,
"loss": 2.9979,
"theoretical_loss": 4.035403251708736,
"tokens_seen": 379781120
},
{
"epoch": 0.12,
"learning_rate": 0.0004469207622868606,
"loss": 3.2267,
"theoretical_loss": 4.035324671714942,
"tokens_seen": 379846656
},
{
"epoch": 0.12,
"learning_rate": 0.0004469107321965898,
"loss": 3.1496,
"theoretical_loss": 4.0352461090729905,
"tokens_seen": 379912192
},
{
"epoch": 0.12,
"learning_rate": 0.00044690070210631896,
"loss": 2.9406,
"theoretical_loss": 4.03516756377606,
"tokens_seen": 379977728
},
{
"epoch": 0.12,
"learning_rate": 0.00044689067201604814,
"loss": 2.9065,
"theoretical_loss": 4.035089035817329,
"tokens_seen": 380043264
},
{
"epoch": 0.12,
"objective/train/docs_used": 226798,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8520052433013916,
"objective/train/theoretical_loss": 4.035010525189982,
"objective/train/tokens_used": 400568800,
"theoretical_loss": 4.035010525189982,
"tokens_seen": 380108800
},
{
"epoch": 0.12,
"learning_rate": 0.0004468806419257773,
"loss": 2.7012,
"theoretical_loss": 4.035010525189982,
"tokens_seen": 380108800
},
{
"epoch": 0.12,
"learning_rate": 0.00044687061183550656,
"loss": 3.035,
"theoretical_loss": 4.034932031887207,
"tokens_seen": 380174336
},
{
"epoch": 0.12,
"learning_rate": 0.0004468605817452357,
"loss": 2.9595,
"theoretical_loss": 4.0348535559021945,
"tokens_seen": 380239872
},
{
"epoch": 0.12,
"learning_rate": 0.0004468505516549649,
"loss": 2.9639,
"theoretical_loss": 4.034775097228142,
"tokens_seen": 380305408
},
{
"epoch": 0.12,
"learning_rate": 0.00044684052156469405,
"loss": 3.1619,
"theoretical_loss": 4.034696655858246,
"tokens_seen": 380370944
},
{
"epoch": 0.12,
"learning_rate": 0.0004468304914744233,
"loss": 2.9525,
"theoretical_loss": 4.0346182317857116,
"tokens_seen": 380436480
},
{
"epoch": 0.12,
"learning_rate": 0.00044682046138415246,
"loss": 3.0604,
"theoretical_loss": 4.034539825003746,
"tokens_seen": 380502016
},
{
"epoch": 0.12,
"learning_rate": 0.00044681043129388164,
"loss": 2.8797,
"theoretical_loss": 4.034461435505557,
"tokens_seen": 380567552
},
{
"epoch": 0.12,
"learning_rate": 0.0004468004012036108,
"loss": 2.8979,
"theoretical_loss": 4.034383063284364,
"tokens_seen": 380633088
},
{
"epoch": 0.12,
"learning_rate": 0.00044679037111334,
"loss": 3.0024,
"theoretical_loss": 4.034304708333382,
"tokens_seen": 380698624
},
{
"epoch": 0.12,
"learning_rate": 0.0004467803410230692,
"loss": 3.0039,
"theoretical_loss": 4.0342263706458334,
"tokens_seen": 380764160
},
{
"epoch": 0.12,
"learning_rate": 0.0004467703109327984,
"loss": 2.9339,
"theoretical_loss": 4.034148050214946,
"tokens_seen": 380829696
},
{
"epoch": 0.12,
"learning_rate": 0.00044676028084252755,
"loss": 3.0316,
"theoretical_loss": 4.034069747033948,
"tokens_seen": 380895232
},
{
"epoch": 0.12,
"learning_rate": 0.0004467502507522568,
"loss": 3.1285,
"theoretical_loss": 4.0339914610960745,
"tokens_seen": 380960768
},
{
"epoch": 0.12,
"learning_rate": 0.00044674022066198597,
"loss": 3.0281,
"theoretical_loss": 4.033913192394563,
"tokens_seen": 381026304
},
{
"epoch": 0.12,
"learning_rate": 0.00044673019057171515,
"loss": 2.8929,
"theoretical_loss": 4.033834940922653,
"tokens_seen": 381091840
},
{
"epoch": 0.12,
"learning_rate": 0.00044672016048144433,
"loss": 2.6042,
"theoretical_loss": 4.033756706673593,
"tokens_seen": 381157376
},
{
"epoch": 0.12,
"learning_rate": 0.0004467101303911735,
"loss": 3.0573,
"theoretical_loss": 4.033678489640629,
"tokens_seen": 381222912
},
{
"epoch": 0.12,
"learning_rate": 0.0004467001003009027,
"loss": 2.9445,
"theoretical_loss": 4.033600289817015,
"tokens_seen": 381288448
},
{
"epoch": 0.12,
"learning_rate": 0.0004466900702106319,
"loss": 3.0191,
"theoretical_loss": 4.033522107196007,
"tokens_seen": 381353984
},
{
"epoch": 0.12,
"learning_rate": 0.0004466800401203611,
"loss": 2.9899,
"theoretical_loss": 4.033443941770867,
"tokens_seen": 381419520
},
{
"epoch": 0.12,
"learning_rate": 0.0004466700100300903,
"loss": 2.7601,
"theoretical_loss": 4.033365793534857,
"tokens_seen": 381485056
},
{
"epoch": 0.12,
"learning_rate": 0.00044665997993981947,
"loss": 3.0316,
"theoretical_loss": 4.033287662481247,
"tokens_seen": 381550592
},
{
"epoch": 0.12,
"learning_rate": 0.00044664994984954865,
"loss": 2.9178,
"theoretical_loss": 4.033209548603308,
"tokens_seen": 381616128
},
{
"epoch": 0.12,
"learning_rate": 0.0004466399197592779,
"loss": 3.1146,
"theoretical_loss": 4.033131451894317,
"tokens_seen": 381681664
},
{
"epoch": 0.12,
"objective/train/docs_used": 227456,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.782240867614746,
"objective/train/theoretical_loss": 4.03305337234755,
"objective/train/tokens_used": 402207200,
"theoretical_loss": 4.03305337234755,
"tokens_seen": 381747200
},
{
"epoch": 0.12,
"learning_rate": 0.000446629889669007,
"loss": 2.8048,
"theoretical_loss": 4.03305337234755,
"tokens_seen": 381747200
},
{
"epoch": 0.12,
"learning_rate": 0.00044661985957873625,
"loss": 3.125,
"theoretical_loss": 4.032975309956292,
"tokens_seen": 381812736
},
{
"epoch": 0.12,
"learning_rate": 0.00044660982948846543,
"loss": 3.0035,
"theoretical_loss": 4.0328972647138315,
"tokens_seen": 381878272
},
{
"epoch": 0.12,
"learning_rate": 0.0004465997993981946,
"loss": 2.9588,
"theoretical_loss": 4.032819236613457,
"tokens_seen": 381943808
},
{
"epoch": 0.12,
"learning_rate": 0.0004465897693079238,
"loss": 3.0603,
"theoretical_loss": 4.032741225648463,
"tokens_seen": 382009344
},
{
"epoch": 0.12,
"learning_rate": 0.00044657973921765297,
"loss": 3.2,
"theoretical_loss": 4.03266323181215,
"tokens_seen": 382074880
},
{
"epoch": 0.12,
"learning_rate": 0.00044656970912738215,
"loss": 2.8238,
"theoretical_loss": 4.0325852550978185,
"tokens_seen": 382140416
},
{
"epoch": 0.12,
"learning_rate": 0.0004465596790371114,
"loss": 3.0385,
"theoretical_loss": 4.032507295498774,
"tokens_seen": 382205952
},
{
"epoch": 0.12,
"learning_rate": 0.0004465496489468405,
"loss": 2.787,
"theoretical_loss": 4.032429353008326,
"tokens_seen": 382271488
},
{
"epoch": 0.12,
"learning_rate": 0.00044653961885656975,
"loss": 3.1649,
"theoretical_loss": 4.032351427619789,
"tokens_seen": 382337024
},
{
"epoch": 0.12,
"learning_rate": 0.0004465295887662989,
"loss": 3.0231,
"theoretical_loss": 4.03227351932648,
"tokens_seen": 382402560
},
{
"epoch": 0.12,
"learning_rate": 0.0004465195586760281,
"loss": 3.1729,
"theoretical_loss": 4.032195628121718,
"tokens_seen": 382468096
},
{
"epoch": 0.12,
"learning_rate": 0.0004465095285857573,
"loss": 2.9617,
"theoretical_loss": 4.032117753998829,
"tokens_seen": 382533632
},
{
"epoch": 0.12,
"learning_rate": 0.0004464994984954865,
"loss": 2.8875,
"theoretical_loss": 4.032039896951142,
"tokens_seen": 382599168
},
{
"epoch": 0.12,
"learning_rate": 0.00044648946840521566,
"loss": 2.9509,
"theoretical_loss": 4.031962056971988,
"tokens_seen": 382664704
},
{
"epoch": 0.12,
"learning_rate": 0.00044647943831494484,
"loss": 2.7274,
"theoretical_loss": 4.0318842340547025,
"tokens_seen": 382730240
},
{
"epoch": 0.12,
"learning_rate": 0.000446469408224674,
"loss": 2.8627,
"theoretical_loss": 4.031806428192627,
"tokens_seen": 382795776
},
{
"epoch": 0.12,
"learning_rate": 0.00044645937813440325,
"loss": 2.9433,
"theoretical_loss": 4.031728639379103,
"tokens_seen": 382861312
},
{
"epoch": 0.12,
"learning_rate": 0.0004464493480441324,
"loss": 3.0113,
"theoretical_loss": 4.031650867607478,
"tokens_seen": 382926848
},
{
"epoch": 0.12,
"learning_rate": 0.0004464393179538616,
"loss": 3.0806,
"theoretical_loss": 4.031573112871103,
"tokens_seen": 382992384
},
{
"epoch": 0.12,
"learning_rate": 0.0004464292878635908,
"loss": 2.9037,
"theoretical_loss": 4.031495375163333,
"tokens_seen": 383057920
},
{
"epoch": 0.12,
"learning_rate": 0.00044641925777332,
"loss": 3.1862,
"theoretical_loss": 4.031417654477526,
"tokens_seen": 383123456
},
{
"epoch": 0.12,
"learning_rate": 0.00044640922768304916,
"loss": 2.9205,
"theoretical_loss": 4.031339950807043,
"tokens_seen": 383188992
},
{
"epoch": 0.12,
"learning_rate": 0.00044639919759277834,
"loss": 2.8033,
"theoretical_loss": 4.03126226414525,
"tokens_seen": 383254528
},
{
"epoch": 0.12,
"learning_rate": 0.0004463891675025075,
"loss": 2.9103,
"theoretical_loss": 4.031184594485517,
"tokens_seen": 383320064
},
{
"epoch": 0.12,
"objective/train/docs_used": 228876,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.048848867416382,
"objective/train/theoretical_loss": 4.031106941821218,
"objective/train/tokens_used": 403845600,
"theoretical_loss": 4.031106941821218,
"tokens_seen": 383385600
},
{
"epoch": 0.12,
"learning_rate": 0.00044637913741223676,
"loss": 2.8579,
"theoretical_loss": 4.031106941821218,
"tokens_seen": 383385600
},
{
"epoch": 0.12,
"learning_rate": 0.0004463691073219659,
"loss": 3.0792,
"theoretical_loss": 4.031029306145728,
"tokens_seen": 383451136
},
{
"epoch": 0.12,
"learning_rate": 0.0004463590772316951,
"loss": 3.1489,
"theoretical_loss": 4.030951687452429,
"tokens_seen": 383516672
},
{
"epoch": 0.12,
"learning_rate": 0.00044634904714142425,
"loss": 3.1007,
"theoretical_loss": 4.0308740857347045,
"tokens_seen": 383582208
},
{
"epoch": 0.12,
"learning_rate": 0.0004463390170511535,
"loss": 3.1964,
"theoretical_loss": 4.030796500985943,
"tokens_seen": 383647744
},
{
"epoch": 0.12,
"learning_rate": 0.00044632898696088266,
"loss": 3.0761,
"theoretical_loss": 4.030718933199536,
"tokens_seen": 383713280
},
{
"epoch": 0.12,
"learning_rate": 0.00044631895687061184,
"loss": 2.9627,
"theoretical_loss": 4.030641382368878,
"tokens_seen": 383778816
},
{
"epoch": 0.12,
"learning_rate": 0.000446308926780341,
"loss": 2.9387,
"theoretical_loss": 4.03056384848737,
"tokens_seen": 383844352
},
{
"epoch": 0.12,
"learning_rate": 0.0004462988966900702,
"loss": 2.7182,
"theoretical_loss": 4.030486331548413,
"tokens_seen": 383909888
},
{
"epoch": 0.12,
"learning_rate": 0.0004462888665997994,
"loss": 3.1237,
"theoretical_loss": 4.030408831545414,
"tokens_seen": 383975424
},
{
"epoch": 0.12,
"learning_rate": 0.0004462788365095286,
"loss": 2.9458,
"theoretical_loss": 4.030331348471784,
"tokens_seen": 384040960
},
{
"epoch": 0.12,
"learning_rate": 0.00044626880641925775,
"loss": 3.0304,
"theoretical_loss": 4.030253882320936,
"tokens_seen": 384106496
},
{
"epoch": 0.12,
"learning_rate": 0.000446258776328987,
"loss": 3.0564,
"theoretical_loss": 4.030176433086289,
"tokens_seen": 384172032
},
{
"epoch": 0.12,
"learning_rate": 0.00044624874623871617,
"loss": 3.1166,
"theoretical_loss": 4.030099000761261,
"tokens_seen": 384237568
},
{
"epoch": 0.12,
"learning_rate": 0.00044623871614844535,
"loss": 3.0936,
"theoretical_loss": 4.030021585339281,
"tokens_seen": 384303104
},
{
"epoch": 0.12,
"learning_rate": 0.00044622868605817453,
"loss": 2.9371,
"theoretical_loss": 4.029944186813774,
"tokens_seen": 384368640
},
{
"epoch": 0.12,
"learning_rate": 0.0004462186559679037,
"loss": 3.0269,
"theoretical_loss": 4.029866805178175,
"tokens_seen": 384434176
},
{
"epoch": 0.12,
"learning_rate": 0.0004462086258776329,
"loss": 2.9255,
"theoretical_loss": 4.029789440425917,
"tokens_seen": 384499712
},
{
"epoch": 0.12,
"learning_rate": 0.0004461985957873621,
"loss": 3.0704,
"theoretical_loss": 4.029712092550443,
"tokens_seen": 384565248
},
{
"epoch": 0.12,
"learning_rate": 0.00044618856569709125,
"loss": 2.9137,
"theoretical_loss": 4.029634761545195,
"tokens_seen": 384630784
},
{
"epoch": 0.12,
"learning_rate": 0.0004461785356068205,
"loss": 3.0362,
"theoretical_loss": 4.029557447403618,
"tokens_seen": 384696320
},
{
"epoch": 0.12,
"learning_rate": 0.0004461685055165496,
"loss": 3.0371,
"theoretical_loss": 4.029480150119165,
"tokens_seen": 384761856
},
{
"epoch": 0.12,
"learning_rate": 0.00044615847542627885,
"loss": 2.7249,
"theoretical_loss": 4.029402869685288,
"tokens_seen": 384827392
},
{
"epoch": 0.12,
"learning_rate": 0.00044614844533600803,
"loss": 3.0639,
"theoretical_loss": 4.0293256060954485,
"tokens_seen": 384892928
},
{
"epoch": 0.12,
"learning_rate": 0.0004461384152457372,
"loss": 2.8059,
"theoretical_loss": 4.029248359343105,
"tokens_seen": 384958464
},
{
"epoch": 0.12,
"objective/train/docs_used": 229454,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6626176834106445,
"objective/train/theoretical_loss": 4.029171129421724,
"objective/train/tokens_used": 405484000,
"theoretical_loss": 4.029171129421724,
"tokens_seen": 385024000
},
{
"epoch": 0.12,
"learning_rate": 0.0004461283851554664,
"loss": 3.0008,
"theoretical_loss": 4.029171129421724,
"tokens_seen": 385024000
},
{
"epoch": 0.12,
"learning_rate": 0.00044611835506519563,
"loss": 2.9454,
"theoretical_loss": 4.029093916324774,
"tokens_seen": 385089536
},
{
"epoch": 0.12,
"learning_rate": 0.00044610832497492476,
"loss": 2.9709,
"theoretical_loss": 4.0290167200457265,
"tokens_seen": 385155072
},
{
"epoch": 0.12,
"learning_rate": 0.000446098294884654,
"loss": 2.8698,
"theoretical_loss": 4.02893954057806,
"tokens_seen": 385220608
},
{
"epoch": 0.12,
"learning_rate": 0.0004460882647943831,
"loss": 3.0199,
"theoretical_loss": 4.028862377915252,
"tokens_seen": 385286144
},
{
"epoch": 0.12,
"learning_rate": 0.00044607823470411235,
"loss": 3.2099,
"theoretical_loss": 4.0287852320507875,
"tokens_seen": 385351680
},
{
"epoch": 0.12,
"learning_rate": 0.00044606820461384153,
"loss": 3.0059,
"theoretical_loss": 4.028708102978152,
"tokens_seen": 385417216
},
{
"epoch": 0.12,
"learning_rate": 0.0004460581745235707,
"loss": 2.9014,
"theoretical_loss": 4.028630990690838,
"tokens_seen": 385482752
},
{
"epoch": 0.12,
"learning_rate": 0.0004460481444332999,
"loss": 2.9839,
"theoretical_loss": 4.028553895182339,
"tokens_seen": 385548288
},
{
"epoch": 0.12,
"learning_rate": 0.0004460381143430291,
"loss": 2.9966,
"theoretical_loss": 4.028476816446153,
"tokens_seen": 385613824
},
{
"epoch": 0.12,
"learning_rate": 0.00044602808425275826,
"loss": 2.9376,
"theoretical_loss": 4.028399754475781,
"tokens_seen": 385679360
},
{
"epoch": 0.12,
"learning_rate": 0.0004460180541624875,
"loss": 2.8859,
"theoretical_loss": 4.028322709264729,
"tokens_seen": 385744896
},
{
"epoch": 0.12,
"learning_rate": 0.0004460080240722166,
"loss": 2.9689,
"theoretical_loss": 4.0282456808065055,
"tokens_seen": 385810432
},
{
"epoch": 0.12,
"learning_rate": 0.00044599799398194586,
"loss": 3.1434,
"theoretical_loss": 4.028168669094623,
"tokens_seen": 385875968
},
{
"epoch": 0.12,
"learning_rate": 0.000445987963891675,
"loss": 3.0785,
"theoretical_loss": 4.028091674122598,
"tokens_seen": 385941504
},
{
"epoch": 0.12,
"learning_rate": 0.0004459779338014042,
"loss": 2.9047,
"theoretical_loss": 4.0280146958839484,
"tokens_seen": 386007040
},
{
"epoch": 0.12,
"learning_rate": 0.0004459679037111334,
"loss": 3.2321,
"theoretical_loss": 4.0279377343722,
"tokens_seen": 386072576
},
{
"epoch": 0.12,
"learning_rate": 0.0004459578736208626,
"loss": 2.9964,
"theoretical_loss": 4.027860789580878,
"tokens_seen": 386138112
},
{
"epoch": 0.12,
"learning_rate": 0.00044594784353059176,
"loss": 2.7246,
"theoretical_loss": 4.027783861503513,
"tokens_seen": 386203648
},
{
"epoch": 0.12,
"learning_rate": 0.000445937813440321,
"loss": 2.9984,
"theoretical_loss": 4.0277069501336396,
"tokens_seen": 386269184
},
{
"epoch": 0.12,
"learning_rate": 0.0004459277833500502,
"loss": 2.9427,
"theoretical_loss": 4.027630055464795,
"tokens_seen": 386334720
},
{
"epoch": 0.12,
"learning_rate": 0.00044591775325977936,
"loss": 2.9795,
"theoretical_loss": 4.027553177490521,
"tokens_seen": 386400256
},
{
"epoch": 0.12,
"learning_rate": 0.00044590772316950854,
"loss": 3.1006,
"theoretical_loss": 4.027476316204362,
"tokens_seen": 386465792
},
{
"epoch": 0.12,
"learning_rate": 0.0004458976930792377,
"loss": 3.0388,
"theoretical_loss": 4.027399471599867,
"tokens_seen": 386531328
},
{
"epoch": 0.12,
"learning_rate": 0.00044588766298896696,
"loss": 3.1798,
"theoretical_loss": 4.027322643670587,
"tokens_seen": 386596864
},
{
"epoch": 0.12,
"objective/train/docs_used": 230121,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1686201095581055,
"objective/train/theoretical_loss": 4.027245832410079,
"objective/train/tokens_used": 407122400,
"theoretical_loss": 4.027245832410079,
"tokens_seen": 386662400
},
{
"epoch": 0.12,
"learning_rate": 0.0004458776328986961,
"loss": 2.9233,
"theoretical_loss": 4.027245832410079,
"tokens_seen": 386662400
},
{
"epoch": 0.12,
"learning_rate": 0.0004458676028084253,
"loss": 2.8835,
"theoretical_loss": 4.027169037811901,
"tokens_seen": 386727936
},
{
"epoch": 0.12,
"learning_rate": 0.00044585757271815445,
"loss": 2.8742,
"theoretical_loss": 4.027092259869616,
"tokens_seen": 386793472
},
{
"epoch": 0.12,
"learning_rate": 0.0004458475426278837,
"loss": 2.8106,
"theoretical_loss": 4.027015498576791,
"tokens_seen": 386859008
},
{
"epoch": 0.12,
"learning_rate": 0.00044583751253761286,
"loss": 2.8313,
"theoretical_loss": 4.026938753926997,
"tokens_seen": 386924544
},
{
"epoch": 0.12,
"learning_rate": 0.00044582748244734204,
"loss": 2.9577,
"theoretical_loss": 4.026862025913804,
"tokens_seen": 386990080
},
{
"epoch": 0.12,
"learning_rate": 0.0004458174523570712,
"loss": 3.0805,
"theoretical_loss": 4.026785314530792,
"tokens_seen": 387055616
},
{
"epoch": 0.12,
"learning_rate": 0.0004458074222668004,
"loss": 3.0757,
"theoretical_loss": 4.026708619771541,
"tokens_seen": 387121152
},
{
"epoch": 0.12,
"learning_rate": 0.0004457973921765296,
"loss": 2.9777,
"theoretical_loss": 4.026631941629635,
"tokens_seen": 387186688
},
{
"epoch": 0.12,
"learning_rate": 0.0004457873620862588,
"loss": 3.1366,
"theoretical_loss": 4.026555280098661,
"tokens_seen": 387252224
},
{
"epoch": 0.12,
"learning_rate": 0.00044577733199598795,
"loss": 3.09,
"theoretical_loss": 4.026478635172212,
"tokens_seen": 387317760
},
{
"epoch": 0.12,
"learning_rate": 0.0004457673019057172,
"loss": 2.9166,
"theoretical_loss": 4.026402006843882,
"tokens_seen": 387383296
},
{
"epoch": 0.12,
"learning_rate": 0.00044575727181544637,
"loss": 2.8882,
"theoretical_loss": 4.0263253951072695,
"tokens_seen": 387448832
},
{
"epoch": 0.12,
"learning_rate": 0.00044574724172517555,
"loss": 2.7889,
"theoretical_loss": 4.026248799955976,
"tokens_seen": 387514368
},
{
"epoch": 0.12,
"learning_rate": 0.00044573721163490473,
"loss": 2.8333,
"theoretical_loss": 4.026172221383607,
"tokens_seen": 387579904
},
{
"epoch": 0.12,
"learning_rate": 0.0004457271815446339,
"loss": 2.7393,
"theoretical_loss": 4.026095659383773,
"tokens_seen": 387645440
},
{
"epoch": 0.12,
"learning_rate": 0.0004457171514543631,
"loss": 2.9765,
"theoretical_loss": 4.026019113950086,
"tokens_seen": 387710976
},
{
"epoch": 0.12,
"learning_rate": 0.0004457071213640923,
"loss": 2.9468,
"theoretical_loss": 4.025942585076161,
"tokens_seen": 387776512
},
{
"epoch": 0.12,
"learning_rate": 0.00044569709127382145,
"loss": 2.9553,
"theoretical_loss": 4.025866072755619,
"tokens_seen": 387842048
},
{
"epoch": 0.12,
"learning_rate": 0.0004456870611835507,
"loss": 3.0161,
"theoretical_loss": 4.025789576982083,
"tokens_seen": 387907584
},
{
"epoch": 0.12,
"learning_rate": 0.0004456770310932798,
"loss": 2.9051,
"theoretical_loss": 4.02571309774918,
"tokens_seen": 387973120
},
{
"epoch": 0.12,
"learning_rate": 0.00044566700100300905,
"loss": 2.9076,
"theoretical_loss": 4.025636635050539,
"tokens_seen": 388038656
},
{
"epoch": 0.12,
"learning_rate": 0.00044565697091273823,
"loss": 3.0276,
"theoretical_loss": 4.025560188879796,
"tokens_seen": 388104192
},
{
"epoch": 0.12,
"learning_rate": 0.0004456469408224674,
"loss": 2.911,
"theoretical_loss": 4.025483759230585,
"tokens_seen": 388169728
},
{
"epoch": 0.12,
"learning_rate": 0.0004456369107321966,
"loss": 2.8527,
"theoretical_loss": 4.025407346096551,
"tokens_seen": 388235264
},
{
"epoch": 0.12,
"objective/train/docs_used": 231531,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4107649326324463,
"objective/train/theoretical_loss": 4.025330949471336,
"objective/train/tokens_used": 408760800,
"theoretical_loss": 4.025330949471336,
"tokens_seen": 388300800
},
{
"epoch": 0.12,
"learning_rate": 0.00044562688064192583,
"loss": 2.909,
"theoretical_loss": 4.025330949471336,
"tokens_seen": 388300800
},
{
"epoch": 0.12,
"learning_rate": 0.00044561685055165496,
"loss": 3.0747,
"theoretical_loss": 4.025254569348588,
"tokens_seen": 388366336
},
{
"epoch": 0.12,
"learning_rate": 0.0004456068204613842,
"loss": 3.0523,
"theoretical_loss": 4.025178205721959,
"tokens_seen": 388431872
},
{
"epoch": 0.12,
"learning_rate": 0.0004455967903711133,
"loss": 2.9877,
"theoretical_loss": 4.025101858585104,
"tokens_seen": 388497408
},
{
"epoch": 0.12,
"learning_rate": 0.00044558676028084255,
"loss": 2.9718,
"theoretical_loss": 4.025025527931681,
"tokens_seen": 388562944
},
{
"epoch": 0.12,
"learning_rate": 0.00044557673019057173,
"loss": 3.0657,
"theoretical_loss": 4.0249492137553515,
"tokens_seen": 388628480
},
{
"epoch": 0.12,
"learning_rate": 0.0004455667001003009,
"loss": 2.8732,
"theoretical_loss": 4.024872916049782,
"tokens_seen": 388694016
},
{
"epoch": 0.12,
"learning_rate": 0.0004455566700100301,
"loss": 3.1106,
"theoretical_loss": 4.0247966348086415,
"tokens_seen": 388759552
},
{
"epoch": 0.12,
"learning_rate": 0.0004455466399197593,
"loss": 3.0527,
"theoretical_loss": 4.0247203700256025,
"tokens_seen": 388825088
},
{
"epoch": 0.12,
"learning_rate": 0.00044553660982948846,
"loss": 3.0707,
"theoretical_loss": 4.02464412169434,
"tokens_seen": 388890624
},
{
"epoch": 0.12,
"learning_rate": 0.0004455265797392177,
"loss": 2.846,
"theoretical_loss": 4.024567889808534,
"tokens_seen": 388956160
},
{
"epoch": 0.12,
"learning_rate": 0.0004455165496489468,
"loss": 3.1694,
"theoretical_loss": 4.024491674361868,
"tokens_seen": 389021696
},
{
"epoch": 0.12,
"learning_rate": 0.00044550651955867606,
"loss": 2.8077,
"theoretical_loss": 4.024415475348027,
"tokens_seen": 389087232
},
{
"epoch": 0.12,
"learning_rate": 0.0004454964894684052,
"loss": 2.9521,
"theoretical_loss": 4.024339292760703,
"tokens_seen": 389152768
},
{
"epoch": 0.12,
"learning_rate": 0.0004454864593781344,
"loss": 2.8418,
"theoretical_loss": 4.024263126593588,
"tokens_seen": 389218304
},
{
"epoch": 0.12,
"learning_rate": 0.0004454764292878636,
"loss": 2.946,
"theoretical_loss": 4.0241869768403795,
"tokens_seen": 389283840
},
{
"epoch": 0.12,
"learning_rate": 0.0004454663991975928,
"loss": 2.8275,
"theoretical_loss": 4.024110843494777,
"tokens_seen": 389349376
},
{
"epoch": 0.12,
"learning_rate": 0.00044545636910732196,
"loss": 2.9606,
"theoretical_loss": 4.024034726550486,
"tokens_seen": 389414912
},
{
"epoch": 0.12,
"learning_rate": 0.0004454463390170512,
"loss": 2.7693,
"theoretical_loss": 4.023958626001212,
"tokens_seen": 389480448
},
{
"epoch": 0.12,
"learning_rate": 0.0004454363089267803,
"loss": 3.21,
"theoretical_loss": 4.023882541840667,
"tokens_seen": 389545984
},
{
"epoch": 0.12,
"learning_rate": 0.00044542627883650956,
"loss": 3.2247,
"theoretical_loss": 4.023806474062566,
"tokens_seen": 389611520
},
{
"epoch": 0.12,
"learning_rate": 0.0004454162487462387,
"loss": 2.8222,
"theoretical_loss": 4.023730422660624,
"tokens_seen": 389677056
},
{
"epoch": 0.12,
"learning_rate": 0.0004454062186559679,
"loss": 3.1396,
"theoretical_loss": 4.023654387628565,
"tokens_seen": 389742592
},
{
"epoch": 0.12,
"learning_rate": 0.0004453961885656971,
"loss": 3.0235,
"theoretical_loss": 4.023578368960113,
"tokens_seen": 389808128
},
{
"epoch": 0.12,
"learning_rate": 0.0004453861584754263,
"loss": 2.8784,
"theoretical_loss": 4.023502366648994,
"tokens_seen": 389873664
},
{
"epoch": 0.12,
"objective/train/docs_used": 232298,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4993464946746826,
"objective/train/theoretical_loss": 4.023426380688943,
"objective/train/tokens_used": 410399200,
"theoretical_loss": 4.023426380688943,
"tokens_seen": 389939200
},
{
"epoch": 0.12,
"learning_rate": 0.00044537612838515547,
"loss": 2.7087,
"theoretical_loss": 4.023426380688943,
"tokens_seen": 389939200
},
{
"epoch": 0.12,
"learning_rate": 0.00044536609829488465,
"loss": 2.7692,
"theoretical_loss": 4.023350411073693,
"tokens_seen": 390004736
},
{
"epoch": 0.12,
"learning_rate": 0.00044535606820461383,
"loss": 3.179,
"theoretical_loss": 4.023274457796983,
"tokens_seen": 390070272
},
{
"epoch": 0.12,
"learning_rate": 0.00044534603811434306,
"loss": 3.0874,
"theoretical_loss": 4.023198520852555,
"tokens_seen": 390135808
},
{
"epoch": 0.12,
"learning_rate": 0.0004453360080240722,
"loss": 2.9398,
"theoretical_loss": 4.023122600234155,
"tokens_seen": 390201344
},
{
"epoch": 0.12,
"learning_rate": 0.0004453259779338014,
"loss": 2.7279,
"theoretical_loss": 4.0230466959355295,
"tokens_seen": 390266880
},
{
"epoch": 0.12,
"learning_rate": 0.00044531594784353055,
"loss": 3.0187,
"theoretical_loss": 4.022970807950434,
"tokens_seen": 390332416
},
{
"epoch": 0.12,
"learning_rate": 0.0004453059177532598,
"loss": 3.1262,
"theoretical_loss": 4.022894936272621,
"tokens_seen": 390397952
},
{
"epoch": 0.12,
"learning_rate": 0.00044529588766298897,
"loss": 2.802,
"theoretical_loss": 4.022819080895853,
"tokens_seen": 390463488
},
{
"epoch": 0.12,
"learning_rate": 0.00044528585757271815,
"loss": 3.0968,
"theoretical_loss": 4.02274324181389,
"tokens_seen": 390529024
},
{
"epoch": 0.12,
"learning_rate": 0.00044527582748244733,
"loss": 2.9464,
"theoretical_loss": 4.022667419020499,
"tokens_seen": 390594560
},
{
"epoch": 0.12,
"learning_rate": 0.00044526579739217657,
"loss": 2.8463,
"theoretical_loss": 4.0225916125094505,
"tokens_seen": 390660096
},
{
"epoch": 0.12,
"learning_rate": 0.0004452557673019057,
"loss": 2.6302,
"theoretical_loss": 4.022515822274515,
"tokens_seen": 390725632
},
{
"epoch": 0.12,
"learning_rate": 0.00044524573721163493,
"loss": 2.9971,
"theoretical_loss": 4.0224400483094715,
"tokens_seen": 390791168
},
{
"epoch": 0.12,
"learning_rate": 0.00044523570712136406,
"loss": 2.9961,
"theoretical_loss": 4.022364290608098,
"tokens_seen": 390856704
},
{
"epoch": 0.12,
"learning_rate": 0.0004452256770310933,
"loss": 3.0202,
"theoretical_loss": 4.022288549164178,
"tokens_seen": 390922240
},
{
"epoch": 0.12,
"learning_rate": 0.00044521564694082247,
"loss": 2.9233,
"theoretical_loss": 4.022212823971498,
"tokens_seen": 390987776
},
{
"epoch": 0.12,
"learning_rate": 0.00044520561685055165,
"loss": 3.165,
"theoretical_loss": 4.022137115023848,
"tokens_seen": 391053312
},
{
"epoch": 0.12,
"learning_rate": 0.00044519558676028083,
"loss": 3.1339,
"theoretical_loss": 4.022061422315023,
"tokens_seen": 391118848
},
{
"epoch": 0.12,
"learning_rate": 0.00044518555667001,
"loss": 2.9278,
"theoretical_loss": 4.021985745838817,
"tokens_seen": 391184384
},
{
"epoch": 0.12,
"learning_rate": 0.00044517552657973925,
"loss": 2.7157,
"theoretical_loss": 4.021910085589032,
"tokens_seen": 391249920
},
{
"epoch": 0.12,
"learning_rate": 0.00044516549648946843,
"loss": 2.8932,
"theoretical_loss": 4.021834441559472,
"tokens_seen": 391315456
},
{
"epoch": 0.12,
"learning_rate": 0.0004451554663991976,
"loss": 2.9059,
"theoretical_loss": 4.021758813743942,
"tokens_seen": 391380992
},
{
"epoch": 0.12,
"learning_rate": 0.0004451454363089268,
"loss": 2.9088,
"theoretical_loss": 4.021683202136255,
"tokens_seen": 391446528
},
{
"epoch": 0.12,
"learning_rate": 0.00044513540621865603,
"loss": 2.995,
"theoretical_loss": 4.021607606730223,
"tokens_seen": 391512064
},
{
"epoch": 0.12,
"objective/train/docs_used": 233436,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.01359224319458,
"objective/train/theoretical_loss": 4.0215320275196635,
"objective/train/tokens_used": 412037600,
"theoretical_loss": 4.0215320275196635,
"tokens_seen": 391577600
},
{
"epoch": 0.12,
"learning_rate": 0.00044512537612838516,
"loss": 2.9565,
"theoretical_loss": 4.0215320275196635,
"tokens_seen": 391577600
},
{
"epoch": 0.12,
"learning_rate": 0.0004451153460381144,
"loss": 3.0462,
"theoretical_loss": 4.0214564644983986,
"tokens_seen": 391643136
},
{
"epoch": 0.12,
"learning_rate": 0.0004451053159478435,
"loss": 2.6842,
"theoretical_loss": 4.021380917660251,
"tokens_seen": 391708672
},
{
"epoch": 0.12,
"learning_rate": 0.00044509528585757275,
"loss": 3.0243,
"theoretical_loss": 4.021305386999048,
"tokens_seen": 391774208
},
{
"epoch": 0.12,
"learning_rate": 0.00044508525576730194,
"loss": 2.9867,
"theoretical_loss": 4.021229872508621,
"tokens_seen": 391839744
},
{
"epoch": 0.12,
"learning_rate": 0.0004450752256770311,
"loss": 2.6909,
"theoretical_loss": 4.021154374182803,
"tokens_seen": 391905280
},
{
"epoch": 0.12,
"learning_rate": 0.0004450651955867603,
"loss": 2.7727,
"theoretical_loss": 4.021078892015433,
"tokens_seen": 391970816
},
{
"epoch": 0.12,
"learning_rate": 0.0004450551654964895,
"loss": 3.0262,
"theoretical_loss": 4.021003426000353,
"tokens_seen": 392036352
},
{
"epoch": 0.12,
"learning_rate": 0.00044504513540621866,
"loss": 3.0198,
"theoretical_loss": 4.020927976131404,
"tokens_seen": 392101888
},
{
"epoch": 0.12,
"learning_rate": 0.0004450351053159479,
"loss": 3.085,
"theoretical_loss": 4.0208525424024355,
"tokens_seen": 392167424
},
{
"epoch": 0.12,
"learning_rate": 0.000445025075225677,
"loss": 2.9415,
"theoretical_loss": 4.020777124807299,
"tokens_seen": 392232960
},
{
"epoch": 0.12,
"learning_rate": 0.00044501504513540626,
"loss": 2.8681,
"theoretical_loss": 4.020701723339848,
"tokens_seen": 392298496
},
{
"epoch": 0.12,
"learning_rate": 0.0004450050150451354,
"loss": 3.0426,
"theoretical_loss": 4.020626337993941,
"tokens_seen": 392364032
},
{
"epoch": 0.12,
"learning_rate": 0.0004449949849548646,
"loss": 2.9429,
"theoretical_loss": 4.020550968763439,
"tokens_seen": 392429568
},
{
"epoch": 0.12,
"learning_rate": 0.0004449849548645938,
"loss": 3.0646,
"theoretical_loss": 4.020475615642207,
"tokens_seen": 392495104
},
{
"epoch": 0.12,
"learning_rate": 0.000444974924774323,
"loss": 2.7898,
"theoretical_loss": 4.020400278624112,
"tokens_seen": 392560640
},
{
"epoch": 0.12,
"learning_rate": 0.00044496489468405216,
"loss": 2.9612,
"theoretical_loss": 4.020324957703025,
"tokens_seen": 392626176
},
{
"epoch": 0.12,
"learning_rate": 0.0004449548645937814,
"loss": 2.9467,
"theoretical_loss": 4.020249652872822,
"tokens_seen": 392691712
},
{
"epoch": 0.12,
"learning_rate": 0.0004449448345035105,
"loss": 2.6348,
"theoretical_loss": 4.02017436412738,
"tokens_seen": 392757248
},
{
"epoch": 0.12,
"learning_rate": 0.00044493480441323976,
"loss": 3.0607,
"theoretical_loss": 4.02009909146058,
"tokens_seen": 392822784
},
{
"epoch": 0.12,
"learning_rate": 0.0004449247743229689,
"loss": 2.826,
"theoretical_loss": 4.020023834866308,
"tokens_seen": 392888320
},
{
"epoch": 0.12,
"learning_rate": 0.0004449147442326981,
"loss": 3.0724,
"theoretical_loss": 4.01994859433845,
"tokens_seen": 392953856
},
{
"epoch": 0.12,
"learning_rate": 0.0004449047141424273,
"loss": 2.9021,
"theoretical_loss": 4.019873369870899,
"tokens_seen": 393019392
},
{
"epoch": 0.12,
"learning_rate": 0.0004448946840521565,
"loss": 3.0023,
"theoretical_loss": 4.019798161457548,
"tokens_seen": 393084928
},
{
"epoch": 0.12,
"learning_rate": 0.00044488465396188567,
"loss": 2.9141,
"theoretical_loss": 4.019722969092298,
"tokens_seen": 393150464
},
{
"epoch": 0.12,
"objective/train/docs_used": 234056,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1465675830841064,
"objective/train/theoretical_loss": 4.019647792769048,
"objective/train/tokens_used": 413676000,
"theoretical_loss": 4.019647792769048,
"tokens_seen": 393216000
},
{
"epoch": 0.12,
"learning_rate": 0.00044487462387161485,
"loss": 3.207,
"theoretical_loss": 4.019647792769048,
"tokens_seen": 393216000
},
{
"epoch": 0.12,
"learning_rate": 0.00044486459378134403,
"loss": 2.9752,
"theoretical_loss": 4.019572632481702,
"tokens_seen": 393281536
},
{
"epoch": 0.12,
"learning_rate": 0.00044485456369107326,
"loss": 3.064,
"theoretical_loss": 4.01949748822417,
"tokens_seen": 393347072
},
{
"epoch": 0.12,
"learning_rate": 0.0004448445336008024,
"loss": 2.7322,
"theoretical_loss": 4.0194223599903625,
"tokens_seen": 393412608
},
{
"epoch": 0.12,
"learning_rate": 0.0004448345035105316,
"loss": 3.1126,
"theoretical_loss": 4.019347247774195,
"tokens_seen": 393478144
},
{
"epoch": 0.12,
"learning_rate": 0.00044482447342026075,
"loss": 2.7526,
"theoretical_loss": 4.0192721515695835,
"tokens_seen": 393543680
},
{
"epoch": 0.12,
"learning_rate": 0.00044481444332999,
"loss": 3.1153,
"theoretical_loss": 4.019197071370451,
"tokens_seen": 393609216
},
{
"epoch": 0.12,
"learning_rate": 0.00044480441323971917,
"loss": 3.1239,
"theoretical_loss": 4.019122007170722,
"tokens_seen": 393674752
},
{
"epoch": 0.12,
"learning_rate": 0.00044479438314944835,
"loss": 2.9529,
"theoretical_loss": 4.019046958964324,
"tokens_seen": 393740288
},
{
"epoch": 0.12,
"learning_rate": 0.00044478435305917753,
"loss": 2.9906,
"theoretical_loss": 4.018971926745189,
"tokens_seen": 393805824
},
{
"epoch": 0.12,
"learning_rate": 0.00044477432296890677,
"loss": 2.6571,
"theoretical_loss": 4.01889691050725,
"tokens_seen": 393871360
},
{
"epoch": 0.12,
"learning_rate": 0.0004447642928786359,
"loss": 3.1611,
"theoretical_loss": 4.018821910244446,
"tokens_seen": 393936896
},
{
"epoch": 0.12,
"learning_rate": 0.00044475426278836513,
"loss": 2.9492,
"theoretical_loss": 4.018746925950719,
"tokens_seen": 394002432
},
{
"epoch": 0.12,
"learning_rate": 0.00044474423269809426,
"loss": 3.2417,
"theoretical_loss": 4.018671957620012,
"tokens_seen": 394067968
},
{
"epoch": 0.12,
"learning_rate": 0.0004447342026078235,
"loss": 2.9681,
"theoretical_loss": 4.018597005246273,
"tokens_seen": 394133504
},
{
"epoch": 0.12,
"learning_rate": 0.00044472417251755267,
"loss": 2.7605,
"theoretical_loss": 4.018522068823454,
"tokens_seen": 394199040
},
{
"epoch": 0.12,
"learning_rate": 0.00044471414242728185,
"loss": 3.1142,
"theoretical_loss": 4.01844714834551,
"tokens_seen": 394264576
},
{
"epoch": 0.12,
"learning_rate": 0.00044470411233701103,
"loss": 2.8985,
"theoretical_loss": 4.018372243806397,
"tokens_seen": 394330112
},
{
"epoch": 0.12,
"learning_rate": 0.0004446940822467402,
"loss": 2.8504,
"theoretical_loss": 4.018297355200076,
"tokens_seen": 394395648
},
{
"epoch": 0.12,
"learning_rate": 0.0004446840521564694,
"loss": 2.8258,
"theoretical_loss": 4.018222482520512,
"tokens_seen": 394461184
},
{
"epoch": 0.12,
"learning_rate": 0.00044467402206619863,
"loss": 2.6849,
"theoretical_loss": 4.018147625761673,
"tokens_seen": 394526720
},
{
"epoch": 0.12,
"learning_rate": 0.00044466399197592776,
"loss": 3.0831,
"theoretical_loss": 4.018072784917529,
"tokens_seen": 394592256
},
{
"epoch": 0.12,
"learning_rate": 0.000444653961885657,
"loss": 3.0542,
"theoretical_loss": 4.017997959982055,
"tokens_seen": 394657792
},
{
"epoch": 0.12,
"learning_rate": 0.0004446439317953861,
"loss": 2.948,
"theoretical_loss": 4.017923150949228,
"tokens_seen": 394723328
},
{
"epoch": 0.12,
"learning_rate": 0.00044463390170511536,
"loss": 2.819,
"theoretical_loss": 4.017848357813028,
"tokens_seen": 394788864
},
{
"epoch": 0.12,
"objective/train/docs_used": 235431,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6532654762268066,
"objective/train/theoretical_loss": 4.01777358056744,
"objective/train/tokens_used": 415314400,
"theoretical_loss": 4.01777358056744,
"tokens_seen": 394854400
},
{
"epoch": 0.12,
"learning_rate": 0.00044462387161484454,
"loss": 3.0346,
"theoretical_loss": 4.01777358056744,
"tokens_seen": 394854400
},
{
"epoch": 0.12,
"learning_rate": 0.0004446138415245737,
"loss": 2.8124,
"theoretical_loss": 4.017698819206451,
"tokens_seen": 394919936
},
{
"epoch": 0.12,
"learning_rate": 0.0004446038114343029,
"loss": 2.7088,
"theoretical_loss": 4.017624073724052,
"tokens_seen": 394985472
},
{
"epoch": 0.12,
"learning_rate": 0.00044459378134403214,
"loss": 3.2092,
"theoretical_loss": 4.017549344114236,
"tokens_seen": 395051008
},
{
"epoch": 0.12,
"learning_rate": 0.00044458375125376126,
"loss": 2.8705,
"theoretical_loss": 4.017474630371,
"tokens_seen": 395116544
},
{
"epoch": 0.12,
"learning_rate": 0.0004445737211634905,
"loss": 3.0769,
"theoretical_loss": 4.0173999324883445,
"tokens_seen": 395182080
},
{
"epoch": 0.12,
"learning_rate": 0.0004445636910732196,
"loss": 2.8085,
"theoretical_loss": 4.0173252504602734,
"tokens_seen": 395247616
},
{
"epoch": 0.12,
"learning_rate": 0.00044455366098294886,
"loss": 2.7325,
"theoretical_loss": 4.017250584280794,
"tokens_seen": 395313152
},
{
"epoch": 0.12,
"learning_rate": 0.00044454363089267804,
"loss": 2.9246,
"theoretical_loss": 4.017175933943916,
"tokens_seen": 395378688
},
{
"epoch": 0.12,
"learning_rate": 0.0004445336008024072,
"loss": 2.6805,
"theoretical_loss": 4.0171012994436515,
"tokens_seen": 395444224
},
{
"epoch": 0.12,
"learning_rate": 0.0004445235707121364,
"loss": 2.7515,
"theoretical_loss": 4.017026680774019,
"tokens_seen": 395509760
},
{
"epoch": 0.12,
"learning_rate": 0.0004445135406218656,
"loss": 2.9577,
"theoretical_loss": 4.016952077929037,
"tokens_seen": 395575296
},
{
"epoch": 0.12,
"learning_rate": 0.00044450351053159477,
"loss": 3.033,
"theoretical_loss": 4.016877490902731,
"tokens_seen": 395640832
},
{
"epoch": 0.12,
"learning_rate": 0.000444493480441324,
"loss": 2.7227,
"theoretical_loss": 4.016802919689124,
"tokens_seen": 395706368
},
{
"epoch": 0.12,
"learning_rate": 0.00044448345035105313,
"loss": 2.9002,
"theoretical_loss": 4.0167283642822476,
"tokens_seen": 395771904
},
{
"epoch": 0.12,
"learning_rate": 0.00044447342026078236,
"loss": 3.0708,
"theoretical_loss": 4.016653824676135,
"tokens_seen": 395837440
},
{
"epoch": 0.12,
"learning_rate": 0.0004444633901705115,
"loss": 2.9276,
"theoretical_loss": 4.016579300864821,
"tokens_seen": 395902976
},
{
"epoch": 0.12,
"learning_rate": 0.0004444533600802407,
"loss": 2.8551,
"theoretical_loss": 4.016504792842346,
"tokens_seen": 395968512
},
{
"epoch": 0.12,
"learning_rate": 0.00044444332998996996,
"loss": 2.9801,
"theoretical_loss": 4.016430300602751,
"tokens_seen": 396034048
},
{
"epoch": 0.12,
"learning_rate": 0.0004444332998996991,
"loss": 2.949,
"theoretical_loss": 4.016355824140085,
"tokens_seen": 396099584
},
{
"epoch": 0.12,
"learning_rate": 0.0004444232698094283,
"loss": 2.8753,
"theoretical_loss": 4.016281363448394,
"tokens_seen": 396165120
},
{
"epoch": 0.12,
"learning_rate": 0.0004444132397191575,
"loss": 2.6945,
"theoretical_loss": 4.016206918521731,
"tokens_seen": 396230656
},
{
"epoch": 0.12,
"learning_rate": 0.0004444032096288867,
"loss": 3.2419,
"theoretical_loss": 4.016132489354153,
"tokens_seen": 396296192
},
{
"epoch": 0.12,
"learning_rate": 0.00044439317953861587,
"loss": 3.0525,
"theoretical_loss": 4.016058075939716,
"tokens_seen": 396361728
},
{
"epoch": 0.12,
"learning_rate": 0.00044438314944834505,
"loss": 3.1025,
"theoretical_loss": 4.015983678272484,
"tokens_seen": 396427264
},
{
"epoch": 0.12,
"objective/train/docs_used": 235840,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4883968830108643,
"objective/train/theoretical_loss": 4.015909296346521,
"objective/train/tokens_used": 416952800,
"theoretical_loss": 4.015909296346521,
"tokens_seen": 396492800
},
{
"epoch": 0.12,
"learning_rate": 0.00044437311935807423,
"loss": 2.8268,
"theoretical_loss": 4.015909296346521,
"tokens_seen": 396492800
},
{
"epoch": 0.12,
"learning_rate": 0.00044436308926780346,
"loss": 2.8041,
"theoretical_loss": 4.015834930155897,
"tokens_seen": 396558336
},
{
"epoch": 0.12,
"learning_rate": 0.0004443530591775326,
"loss": 2.8989,
"theoretical_loss": 4.015760579694683,
"tokens_seen": 396623872
},
{
"epoch": 0.12,
"learning_rate": 0.0004443430290872618,
"loss": 3.0333,
"theoretical_loss": 4.015686244956953,
"tokens_seen": 396689408
},
{
"epoch": 0.12,
"learning_rate": 0.00044433299899699095,
"loss": 2.93,
"theoretical_loss": 4.015611925936785,
"tokens_seen": 396754944
},
{
"epoch": 0.12,
"learning_rate": 0.0004443229689067202,
"loss": 2.6344,
"theoretical_loss": 4.01553762262826,
"tokens_seen": 396820480
},
{
"epoch": 0.12,
"learning_rate": 0.00044431293881644937,
"loss": 2.9469,
"theoretical_loss": 4.015463335025465,
"tokens_seen": 396886016
},
{
"epoch": 0.12,
"learning_rate": 0.00044430290872617855,
"loss": 2.9232,
"theoretical_loss": 4.015389063122484,
"tokens_seen": 396951552
},
{
"epoch": 0.12,
"learning_rate": 0.00044429287863590773,
"loss": 2.7601,
"theoretical_loss": 4.015314806913411,
"tokens_seen": 397017088
},
{
"epoch": 0.12,
"learning_rate": 0.00044428284854563697,
"loss": 2.6964,
"theoretical_loss": 4.015240566392338,
"tokens_seen": 397082624
},
{
"epoch": 0.12,
"learning_rate": 0.0004442728184553661,
"loss": 3.0921,
"theoretical_loss": 4.015166341553362,
"tokens_seen": 397148160
},
{
"epoch": 0.12,
"learning_rate": 0.00044426278836509533,
"loss": 2.902,
"theoretical_loss": 4.015092132390585,
"tokens_seen": 397213696
},
{
"epoch": 0.12,
"learning_rate": 0.00044425275827482446,
"loss": 2.7439,
"theoretical_loss": 4.01501793889811,
"tokens_seen": 397279232
},
{
"epoch": 0.12,
"learning_rate": 0.0004442427281845537,
"loss": 2.9831,
"theoretical_loss": 4.014943761070043,
"tokens_seen": 397344768
},
{
"epoch": 0.12,
"learning_rate": 0.00044423269809428287,
"loss": 2.841,
"theoretical_loss": 4.014869598900496,
"tokens_seen": 397410304
},
{
"epoch": 0.12,
"learning_rate": 0.00044422266800401205,
"loss": 3.0263,
"theoretical_loss": 4.01479545238358,
"tokens_seen": 397475840
},
{
"epoch": 0.12,
"learning_rate": 0.00044421263791374123,
"loss": 2.8788,
"theoretical_loss": 4.014721321513412,
"tokens_seen": 397541376
},
{
"epoch": 0.12,
"learning_rate": 0.0004442026078234704,
"loss": 2.9554,
"theoretical_loss": 4.0146472062841125,
"tokens_seen": 397606912
},
{
"epoch": 0.12,
"learning_rate": 0.0004441925777331996,
"loss": 2.8914,
"theoretical_loss": 4.014573106689803,
"tokens_seen": 397672448
},
{
"epoch": 0.12,
"learning_rate": 0.00044418254764292883,
"loss": 2.9146,
"theoretical_loss": 4.0144990227246105,
"tokens_seen": 397737984
},
{
"epoch": 0.12,
"learning_rate": 0.00044417251755265796,
"loss": 3.127,
"theoretical_loss": 4.014424954382663,
"tokens_seen": 397803520
},
{
"epoch": 0.12,
"learning_rate": 0.0004441624874623872,
"loss": 3.1975,
"theoretical_loss": 4.014350901658093,
"tokens_seen": 397869056
},
{
"epoch": 0.12,
"learning_rate": 0.0004441524573721163,
"loss": 2.8541,
"theoretical_loss": 4.014276864545036,
"tokens_seen": 397934592
},
{
"epoch": 0.12,
"learning_rate": 0.00044414242728184556,
"loss": 3.0876,
"theoretical_loss": 4.014202843037632,
"tokens_seen": 398000128
},
{
"epoch": 0.12,
"learning_rate": 0.00044413239719157474,
"loss": 2.8968,
"theoretical_loss": 4.014128837130021,
"tokens_seen": 398065664
},
{
"epoch": 0.12,
"objective/train/docs_used": 236468,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.918165445327759,
"objective/train/theoretical_loss": 4.0140548468163475,
"objective/train/tokens_used": 418591200,
"theoretical_loss": 4.0140548468163475,
"tokens_seen": 398131200
},
{
"epoch": 0.12,
"learning_rate": 0.0004441223671013039,
"loss": 2.8995,
"theoretical_loss": 4.0140548468163475,
"tokens_seen": 398131200
},
{
"epoch": 0.12,
"learning_rate": 0.0004441123370110331,
"loss": 3.0238,
"theoretical_loss": 4.0139808720907615,
"tokens_seen": 398196736
},
{
"epoch": 0.12,
"learning_rate": 0.00044410230692076234,
"loss": 2.8314,
"theoretical_loss": 4.013906912947412,
"tokens_seen": 398262272
},
{
"epoch": 0.12,
"learning_rate": 0.00044409227683049146,
"loss": 2.9981,
"theoretical_loss": 4.013832969380456,
"tokens_seen": 398327808
},
{
"epoch": 0.12,
"learning_rate": 0.0004440822467402207,
"loss": 3.0402,
"theoretical_loss": 4.013759041384049,
"tokens_seen": 398393344
},
{
"epoch": 0.12,
"learning_rate": 0.0004440722166499498,
"loss": 2.6356,
"theoretical_loss": 4.013685128952353,
"tokens_seen": 398458880
},
{
"epoch": 0.12,
"learning_rate": 0.00044406218655967906,
"loss": 3.1591,
"theoretical_loss": 4.013611232079531,
"tokens_seen": 398524416
},
{
"epoch": 0.12,
"learning_rate": 0.00044405215646940824,
"loss": 2.8642,
"theoretical_loss": 4.01353735075975,
"tokens_seen": 398589952
},
{
"epoch": 0.12,
"learning_rate": 0.0004440421263791374,
"loss": 3.0194,
"theoretical_loss": 4.01346348498718,
"tokens_seen": 398655488
},
{
"epoch": 0.12,
"learning_rate": 0.0004440320962888666,
"loss": 3.0007,
"theoretical_loss": 4.013389634755995,
"tokens_seen": 398721024
},
{
"epoch": 0.12,
"learning_rate": 0.0004440220661985958,
"loss": 2.9131,
"theoretical_loss": 4.013315800060371,
"tokens_seen": 398786560
},
{
"epoch": 0.12,
"learning_rate": 0.00044401203610832497,
"loss": 2.7356,
"theoretical_loss": 4.013241980894488,
"tokens_seen": 398852096
},
{
"epoch": 0.12,
"learning_rate": 0.0004440020060180542,
"loss": 2.8804,
"theoretical_loss": 4.013168177252528,
"tokens_seen": 398917632
},
{
"epoch": 0.12,
"learning_rate": 0.00044399197592778333,
"loss": 2.8778,
"theoretical_loss": 4.013094389128677,
"tokens_seen": 398983168
},
{
"epoch": 0.12,
"learning_rate": 0.00044398194583751256,
"loss": 3.0424,
"theoretical_loss": 4.0130206165171245,
"tokens_seen": 399048704
},
{
"epoch": 0.12,
"learning_rate": 0.0004439719157472417,
"loss": 3.0925,
"theoretical_loss": 4.012946859412062,
"tokens_seen": 399114240
},
{
"epoch": 0.12,
"learning_rate": 0.0004439618856569709,
"loss": 2.771,
"theoretical_loss": 4.012873117807685,
"tokens_seen": 399179776
},
{
"epoch": 0.12,
"learning_rate": 0.0004439518555667001,
"loss": 2.7898,
"theoretical_loss": 4.012799391698191,
"tokens_seen": 399245312
},
{
"epoch": 0.12,
"learning_rate": 0.0004439418254764293,
"loss": 2.9576,
"theoretical_loss": 4.012725681077783,
"tokens_seen": 399310848
},
{
"epoch": 0.12,
"learning_rate": 0.00044393179538615847,
"loss": 2.9726,
"theoretical_loss": 4.012651985940666,
"tokens_seen": 399376384
},
{
"epoch": 0.12,
"learning_rate": 0.0004439217652958877,
"loss": 3.0252,
"theoretical_loss": 4.012578306281045,
"tokens_seen": 399441920
},
{
"epoch": 0.12,
"learning_rate": 0.00044391173520561683,
"loss": 2.6768,
"theoretical_loss": 4.012504642093133,
"tokens_seen": 399507456
},
{
"epoch": 0.12,
"learning_rate": 0.00044390170511534607,
"loss": 3.0128,
"theoretical_loss": 4.012430993371144,
"tokens_seen": 399572992
},
{
"epoch": 0.12,
"learning_rate": 0.0004438916750250752,
"loss": 3.0036,
"theoretical_loss": 4.012357360109293,
"tokens_seen": 399638528
},
{
"epoch": 0.12,
"learning_rate": 0.00044388164493480443,
"loss": 2.8824,
"theoretical_loss": 4.012283742301802,
"tokens_seen": 399704064
},
{
"epoch": 0.12,
"objective/train/docs_used": 237759,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.293104410171509,
"objective/train/theoretical_loss": 4.012210139942894,
"objective/train/tokens_used": 420229600,
"theoretical_loss": 4.012210139942894,
"tokens_seen": 399769600
},
{
"epoch": 0.12,
"learning_rate": 0.0004438716148445336,
"loss": 3.0101,
"theoretical_loss": 4.012210139942894,
"tokens_seen": 399769600
},
{
"epoch": 0.12,
"learning_rate": 0.0004438615847542628,
"loss": 2.7273,
"theoretical_loss": 4.012136553026796,
"tokens_seen": 399835136
},
{
"epoch": 0.12,
"learning_rate": 0.00044385155466399197,
"loss": 2.9857,
"theoretical_loss": 4.012062981547736,
"tokens_seen": 399900672
},
{
"epoch": 0.12,
"learning_rate": 0.00044384152457372115,
"loss": 2.9248,
"theoretical_loss": 4.011989425499947,
"tokens_seen": 399966208
},
{
"epoch": 0.12,
"learning_rate": 0.00044383149448345033,
"loss": 2.9246,
"theoretical_loss": 4.0119158848776655,
"tokens_seen": 400031744
},
{
"epoch": 0.12,
"learning_rate": 0.00044382146439317957,
"loss": 2.8758,
"theoretical_loss": 4.011842359675129,
"tokens_seen": 400097280
},
{
"epoch": 0.12,
"learning_rate": 0.0004438114343029087,
"loss": 2.905,
"theoretical_loss": 4.011768849886582,
"tokens_seen": 400162816
},
{
"epoch": 0.12,
"learning_rate": 0.00044380140421263793,
"loss": 2.9052,
"theoretical_loss": 4.011695355506266,
"tokens_seen": 400228352
},
{
"epoch": 0.12,
"learning_rate": 0.0004437913741223671,
"loss": 3.0508,
"theoretical_loss": 4.011621876528431,
"tokens_seen": 400293888
},
{
"epoch": 0.12,
"learning_rate": 0.0004437813440320963,
"loss": 2.9092,
"theoretical_loss": 4.011548412947327,
"tokens_seen": 400359424
},
{
"epoch": 0.12,
"learning_rate": 0.0004437713139418255,
"loss": 2.6617,
"theoretical_loss": 4.011474964757211,
"tokens_seen": 400424960
},
{
"epoch": 0.12,
"learning_rate": 0.00044376128385155466,
"loss": 2.9935,
"theoretical_loss": 4.0114015319523375,
"tokens_seen": 400490496
},
{
"epoch": 0.12,
"learning_rate": 0.00044375125376128384,
"loss": 2.7608,
"theoretical_loss": 4.011328114526968,
"tokens_seen": 400556032
},
{
"epoch": 0.12,
"learning_rate": 0.00044374122367101307,
"loss": 3.0189,
"theoretical_loss": 4.011254712475365,
"tokens_seen": 400621568
},
{
"epoch": 0.12,
"learning_rate": 0.0004437311935807422,
"loss": 2.8798,
"theoretical_loss": 4.011181325791797,
"tokens_seen": 400687104
},
{
"epoch": 0.12,
"learning_rate": 0.00044372116349047143,
"loss": 3.073,
"theoretical_loss": 4.011107954470531,
"tokens_seen": 400752640
},
{
"epoch": 0.12,
"learning_rate": 0.00044371113340020056,
"loss": 3.0002,
"theoretical_loss": 4.011034598505843,
"tokens_seen": 400818176
},
{
"epoch": 0.12,
"learning_rate": 0.0004437011033099298,
"loss": 2.6837,
"theoretical_loss": 4.010961257892005,
"tokens_seen": 400883712
},
{
"epoch": 0.12,
"learning_rate": 0.00044369107321965903,
"loss": 3.0952,
"theoretical_loss": 4.010887932623299,
"tokens_seen": 400949248
},
{
"epoch": 0.12,
"learning_rate": 0.00044368104312938816,
"loss": 3.1133,
"theoretical_loss": 4.010814622694006,
"tokens_seen": 401014784
},
{
"epoch": 0.12,
"learning_rate": 0.0004436710130391174,
"loss": 2.6992,
"theoretical_loss": 4.01074132809841,
"tokens_seen": 401080320
},
{
"epoch": 0.12,
"learning_rate": 0.0004436609829488465,
"loss": 3.0919,
"theoretical_loss": 4.0106680488308,
"tokens_seen": 401145856
},
{
"epoch": 0.12,
"learning_rate": 0.00044365095285857576,
"loss": 2.9937,
"theoretical_loss": 4.010594784885466,
"tokens_seen": 401211392
},
{
"epoch": 0.12,
"learning_rate": 0.00044364092276830494,
"loss": 2.6733,
"theoretical_loss": 4.010521536256705,
"tokens_seen": 401276928
},
{
"epoch": 0.12,
"learning_rate": 0.0004436308926780341,
"loss": 2.933,
"theoretical_loss": 4.01044830293881,
"tokens_seen": 401342464
},
{
"epoch": 0.12,
"objective/train/docs_used": 238664,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.03971004486084,
"objective/train/theoretical_loss": 4.0103750849260855,
"objective/train/tokens_used": 421868000,
"theoretical_loss": 4.0103750849260855,
"tokens_seen": 401408000
},
{
"epoch": 0.12,
"learning_rate": 0.0004436208625877633,
"loss": 2.7744,
"theoretical_loss": 4.0103750849260855,
"tokens_seen": 401408000
},
{
"epoch": 0.12,
"learning_rate": 0.00044361083249749254,
"loss": 2.9641,
"theoretical_loss": 4.010301882212831,
"tokens_seen": 401473536
},
{
"epoch": 0.12,
"learning_rate": 0.00044360080240722166,
"loss": 2.9476,
"theoretical_loss": 4.010228694793356,
"tokens_seen": 401539072
},
{
"epoch": 0.12,
"learning_rate": 0.0004435907723169509,
"loss": 3.1747,
"theoretical_loss": 4.010155522661968,
"tokens_seen": 401604608
},
{
"epoch": 0.12,
"learning_rate": 0.00044358074222668,
"loss": 2.8852,
"theoretical_loss": 4.0100823658129805,
"tokens_seen": 401670144
},
{
"epoch": 0.12,
"learning_rate": 0.00044357071213640926,
"loss": 3.0247,
"theoretical_loss": 4.010009224240708,
"tokens_seen": 401735680
},
{
"epoch": 0.12,
"learning_rate": 0.00044356068204613844,
"loss": 2.9916,
"theoretical_loss": 4.0099360979394705,
"tokens_seen": 401801216
},
{
"epoch": 0.12,
"learning_rate": 0.0004435506519558676,
"loss": 2.9303,
"theoretical_loss": 4.009862986903588,
"tokens_seen": 401866752
},
{
"epoch": 0.12,
"learning_rate": 0.0004435406218655968,
"loss": 2.8707,
"theoretical_loss": 4.009789891127387,
"tokens_seen": 401932288
},
{
"epoch": 0.12,
"learning_rate": 0.000443530591775326,
"loss": 2.9204,
"theoretical_loss": 4.009716810605193,
"tokens_seen": 401997824
},
{
"epoch": 0.12,
"learning_rate": 0.00044352056168505517,
"loss": 2.7006,
"theoretical_loss": 4.009643745331338,
"tokens_seen": 402063360
},
{
"epoch": 0.12,
"learning_rate": 0.0004435105315947844,
"loss": 2.9334,
"theoretical_loss": 4.009570695300155,
"tokens_seen": 402128896
},
{
"epoch": 0.12,
"learning_rate": 0.00044350050150451353,
"loss": 2.986,
"theoretical_loss": 4.009497660505982,
"tokens_seen": 402194432
},
{
"epoch": 0.12,
"learning_rate": 0.00044349047141424276,
"loss": 2.9527,
"theoretical_loss": 4.009424640943157,
"tokens_seen": 402259968
},
{
"epoch": 0.12,
"learning_rate": 0.0004434804413239719,
"loss": 3.0015,
"theoretical_loss": 4.009351636606025,
"tokens_seen": 402325504
},
{
"epoch": 0.12,
"learning_rate": 0.0004434704112337011,
"loss": 2.7869,
"theoretical_loss": 4.00927864748893,
"tokens_seen": 402391040
},
{
"epoch": 0.12,
"learning_rate": 0.0004434603811434303,
"loss": 2.8129,
"theoretical_loss": 4.009205673586222,
"tokens_seen": 402456576
},
{
"epoch": 0.12,
"learning_rate": 0.0004434503510531595,
"loss": 2.8991,
"theoretical_loss": 4.009132714892251,
"tokens_seen": 402522112
},
{
"epoch": 0.12,
"learning_rate": 0.00044344032096288867,
"loss": 2.8425,
"theoretical_loss": 4.009059771401375,
"tokens_seen": 402587648
},
{
"epoch": 0.12,
"learning_rate": 0.0004434302908726179,
"loss": 2.8532,
"theoretical_loss": 4.008986843107949,
"tokens_seen": 402653184
},
{
"epoch": 0.12,
"learning_rate": 0.00044342026078234703,
"loss": 2.798,
"theoretical_loss": 4.008913930006336,
"tokens_seen": 402718720
},
{
"epoch": 0.12,
"learning_rate": 0.00044341023069207627,
"loss": 2.9858,
"theoretical_loss": 4.008841032090898,
"tokens_seen": 402784256
},
{
"epoch": 0.12,
"learning_rate": 0.0004434002006018054,
"loss": 3.017,
"theoretical_loss": 4.008768149356003,
"tokens_seen": 402849792
},
{
"epoch": 0.12,
"learning_rate": 0.00044339017051153463,
"loss": 2.7734,
"theoretical_loss": 4.0086952817960215,
"tokens_seen": 402915328
},
{
"epoch": 0.12,
"learning_rate": 0.0004433801404212638,
"loss": 3.0587,
"theoretical_loss": 4.0086224294053245,
"tokens_seen": 402980864
},
{
"epoch": 0.12,
"objective/train/docs_used": 240054,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.683651924133301,
"objective/train/theoretical_loss": 4.008549592178291,
"objective/train/tokens_used": 423506400,
"theoretical_loss": 4.008549592178291,
"tokens_seen": 403046400
},
{
"epoch": 0.12,
"learning_rate": 0.000443370110330993,
"loss": 2.7831,
"theoretical_loss": 4.008549592178291,
"tokens_seen": 403046400
},
{
"epoch": 0.12,
"learning_rate": 0.00044336008024072217,
"loss": 3.0083,
"theoretical_loss": 4.008476770109296,
"tokens_seen": 403111936
},
{
"epoch": 0.12,
"learning_rate": 0.00044335005015045135,
"loss": 2.8457,
"theoretical_loss": 4.008403963192725,
"tokens_seen": 403177472
},
{
"epoch": 0.12,
"learning_rate": 0.00044334002006018053,
"loss": 3.0476,
"theoretical_loss": 4.008331171422961,
"tokens_seen": 403243008
},
{
"epoch": 0.12,
"learning_rate": 0.00044332998996990977,
"loss": 2.9666,
"theoretical_loss": 4.008258394794392,
"tokens_seen": 403308544
},
{
"epoch": 0.12,
"learning_rate": 0.0004433199598796389,
"loss": 2.9,
"theoretical_loss": 4.0081856333014105,
"tokens_seen": 403374080
},
{
"epoch": 0.12,
"learning_rate": 0.00044330992978936813,
"loss": 2.8943,
"theoretical_loss": 4.008112886938409,
"tokens_seen": 403439616
},
{
"epoch": 0.12,
"learning_rate": 0.0004432998996990973,
"loss": 2.8262,
"theoretical_loss": 4.008040155699783,
"tokens_seen": 403505152
},
{
"epoch": 0.12,
"learning_rate": 0.0004432898696088265,
"loss": 2.9041,
"theoretical_loss": 4.007967439579936,
"tokens_seen": 403570688
},
{
"epoch": 0.12,
"learning_rate": 0.0004432798395185557,
"loss": 3.088,
"theoretical_loss": 4.007894738573267,
"tokens_seen": 403636224
},
{
"epoch": 0.12,
"learning_rate": 0.00044326980942828486,
"loss": 2.7578,
"theoretical_loss": 4.007822052674185,
"tokens_seen": 403701760
},
{
"epoch": 0.12,
"learning_rate": 0.00044325977933801404,
"loss": 2.875,
"theoretical_loss": 4.007749381877097,
"tokens_seen": 403767296
},
{
"epoch": 0.12,
"learning_rate": 0.00044324974924774327,
"loss": 3.1834,
"theoretical_loss": 4.007676726176415,
"tokens_seen": 403832832
},
{
"epoch": 0.12,
"learning_rate": 0.0004432397191574724,
"loss": 2.5301,
"theoretical_loss": 4.0076040855665545,
"tokens_seen": 403898368
},
{
"epoch": 0.12,
"learning_rate": 0.00044322968906720163,
"loss": 2.9509,
"theoretical_loss": 4.007531460041933,
"tokens_seen": 403963904
},
{
"epoch": 0.12,
"learning_rate": 0.00044321965897693076,
"loss": 2.6442,
"theoretical_loss": 4.00745884959697,
"tokens_seen": 404029440
},
{
"epoch": 0.12,
"learning_rate": 0.00044320962888666,
"loss": 3.2275,
"theoretical_loss": 4.00738625422609,
"tokens_seen": 404094976
},
{
"epoch": 0.12,
"learning_rate": 0.0004431995987963892,
"loss": 2.8503,
"theoretical_loss": 4.00731367392372,
"tokens_seen": 404160512
},
{
"epoch": 0.12,
"learning_rate": 0.00044318956870611836,
"loss": 3.0621,
"theoretical_loss": 4.0072411086842905,
"tokens_seen": 404226048
},
{
"epoch": 0.12,
"learning_rate": 0.00044317953861584754,
"loss": 3.1279,
"theoretical_loss": 4.007168558502232,
"tokens_seen": 404291584
},
{
"epoch": 0.12,
"learning_rate": 0.0004431695085255767,
"loss": 2.9978,
"theoretical_loss": 4.007096023371981,
"tokens_seen": 404357120
},
{
"epoch": 0.12,
"learning_rate": 0.0004431594784353059,
"loss": 2.9717,
"theoretical_loss": 4.007023503287976,
"tokens_seen": 404422656
},
{
"epoch": 0.12,
"learning_rate": 0.00044314944834503514,
"loss": 3.0025,
"theoretical_loss": 4.006950998244658,
"tokens_seen": 404488192
},
{
"epoch": 0.12,
"learning_rate": 0.00044313941825476427,
"loss": 2.7928,
"theoretical_loss": 4.006878508236473,
"tokens_seen": 404553728
},
{
"epoch": 0.12,
"learning_rate": 0.0004431293881644935,
"loss": 3.1174,
"theoretical_loss": 4.006806033257866,
"tokens_seen": 404619264
},
{
"epoch": 0.12,
"objective/train/docs_used": 240788,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.887927770614624,
"objective/train/theoretical_loss": 4.006733573303289,
"objective/train/tokens_used": 425144800,
"theoretical_loss": 4.006733573303289,
"tokens_seen": 404684800
},
{
"epoch": 0.12,
"learning_rate": 0.0004431193580742227,
"loss": 2.8302,
"theoretical_loss": 4.006733573303289,
"tokens_seen": 404684800
},
{
"epoch": 0.12,
"learning_rate": 0.00044310932798395186,
"loss": 3.0272,
"theoretical_loss": 4.0066611283671945,
"tokens_seen": 404750336
},
{
"epoch": 0.12,
"learning_rate": 0.00044309929789368104,
"loss": 2.8301,
"theoretical_loss": 4.0065886984440375,
"tokens_seen": 404815872
},
{
"epoch": 0.12,
"learning_rate": 0.0004430892678034102,
"loss": 3.0603,
"theoretical_loss": 4.00651628352828,
"tokens_seen": 404881408
},
{
"epoch": 0.12,
"learning_rate": 0.0004430792377131394,
"loss": 2.8986,
"theoretical_loss": 4.0064438836143825,
"tokens_seen": 404946944
},
{
"epoch": 0.12,
"learning_rate": 0.00044306920762286864,
"loss": 3.0905,
"theoretical_loss": 4.00637149869681,
"tokens_seen": 405012480
},
{
"epoch": 0.12,
"learning_rate": 0.00044305917753259777,
"loss": 3.0358,
"theoretical_loss": 4.0062991287700305,
"tokens_seen": 405078016
},
{
"epoch": 0.12,
"learning_rate": 0.000443049147442327,
"loss": 2.8933,
"theoretical_loss": 4.006226773828514,
"tokens_seen": 405143552
},
{
"epoch": 0.12,
"learning_rate": 0.00044303911735205613,
"loss": 2.9264,
"theoretical_loss": 4.006154433866736,
"tokens_seen": 405209088
},
{
"epoch": 0.12,
"learning_rate": 0.00044302908726178537,
"loss": 3.1129,
"theoretical_loss": 4.0060821088791725,
"tokens_seen": 405274624
},
{
"epoch": 0.12,
"learning_rate": 0.00044301905717151455,
"loss": 2.9723,
"theoretical_loss": 4.006009798860303,
"tokens_seen": 405340160
},
{
"epoch": 0.12,
"learning_rate": 0.00044300902708124373,
"loss": 2.9772,
"theoretical_loss": 4.00593750380461,
"tokens_seen": 405405696
},
{
"epoch": 0.12,
"learning_rate": 0.0004429989969909729,
"loss": 3.0805,
"theoretical_loss": 4.005865223706579,
"tokens_seen": 405471232
},
{
"epoch": 0.12,
"learning_rate": 0.0004429889669007021,
"loss": 3.0728,
"theoretical_loss": 4.005792958560699,
"tokens_seen": 405536768
},
{
"epoch": 0.12,
"learning_rate": 0.00044297893681043127,
"loss": 2.835,
"theoretical_loss": 4.005720708361461,
"tokens_seen": 405602304
},
{
"epoch": 0.12,
"learning_rate": 0.0004429689067201605,
"loss": 3.1059,
"theoretical_loss": 4.00564847310336,
"tokens_seen": 405667840
},
{
"epoch": 0.12,
"learning_rate": 0.00044295887662988963,
"loss": 2.8118,
"theoretical_loss": 4.005576252780892,
"tokens_seen": 405733376
},
{
"epoch": 0.12,
"learning_rate": 0.00044294884653961887,
"loss": 2.815,
"theoretical_loss": 4.005504047388558,
"tokens_seen": 405798912
},
{
"epoch": 0.12,
"learning_rate": 0.0004429388164493481,
"loss": 2.8697,
"theoretical_loss": 4.00543185692086,
"tokens_seen": 405864448
},
{
"epoch": 0.12,
"learning_rate": 0.00044292878635907723,
"loss": 2.9243,
"theoretical_loss": 4.005359681372306,
"tokens_seen": 405929984
},
{
"epoch": 0.12,
"learning_rate": 0.00044291875626880647,
"loss": 2.9472,
"theoretical_loss": 4.005287520737403,
"tokens_seen": 405995520
},
{
"epoch": 0.12,
"learning_rate": 0.0004429087261785356,
"loss": 2.8168,
"theoretical_loss": 4.005215375010662,
"tokens_seen": 406061056
},
{
"epoch": 0.12,
"learning_rate": 0.00044289869608826483,
"loss": 3.1313,
"theoretical_loss": 4.0051432441866,
"tokens_seen": 406126592
},
{
"epoch": 0.12,
"learning_rate": 0.000442888665997994,
"loss": 2.6992,
"theoretical_loss": 4.005071128259734,
"tokens_seen": 406192128
},
{
"epoch": 0.12,
"learning_rate": 0.0004428786359077232,
"loss": 2.9379,
"theoretical_loss": 4.004999027224584,
"tokens_seen": 406257664
},
{
"epoch": 0.12,
"objective/train/docs_used": 241796,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5601613521575928,
"objective/train/theoretical_loss": 4.004926941075674,
"objective/train/tokens_used": 426783200,
"theoretical_loss": 4.004926941075674,
"tokens_seen": 406323200
},
{
"epoch": 0.12,
"learning_rate": 0.00044286860581745237,
"loss": 2.7494,
"theoretical_loss": 4.004926941075674,
"tokens_seen": 406323200
},
{
"epoch": 0.12,
"learning_rate": 0.00044285857572718155,
"loss": 2.8608,
"theoretical_loss": 4.004854869807528,
"tokens_seen": 406388736
},
{
"epoch": 0.12,
"learning_rate": 0.00044284854563691073,
"loss": 3.0428,
"theoretical_loss": 4.004782813414678,
"tokens_seen": 406454272
},
{
"epoch": 0.12,
"learning_rate": 0.00044283851554663997,
"loss": 2.8653,
"theoretical_loss": 4.004710771891654,
"tokens_seen": 406519808
},
{
"epoch": 0.12,
"learning_rate": 0.0004428284854563691,
"loss": 3.0127,
"theoretical_loss": 4.0046387452329935,
"tokens_seen": 406585344
},
{
"epoch": 0.12,
"learning_rate": 0.00044281845536609833,
"loss": 2.8257,
"theoretical_loss": 4.004566733433231,
"tokens_seen": 406650880
},
{
"epoch": 0.12,
"learning_rate": 0.0004428084252758275,
"loss": 2.8753,
"theoretical_loss": 4.00449473648691,
"tokens_seen": 406716416
},
{
"epoch": 0.12,
"learning_rate": 0.0004427983951855567,
"loss": 2.8526,
"theoretical_loss": 4.004422754388573,
"tokens_seen": 406781952
},
{
"epoch": 0.12,
"learning_rate": 0.0004427883650952859,
"loss": 2.9927,
"theoretical_loss": 4.004350787132766,
"tokens_seen": 406847488
},
{
"epoch": 0.12,
"learning_rate": 0.00044277833500501506,
"loss": 2.7786,
"theoretical_loss": 4.004278834714039,
"tokens_seen": 406913024
},
{
"epoch": 0.12,
"learning_rate": 0.00044276830491474424,
"loss": 2.8439,
"theoretical_loss": 4.004206897126943,
"tokens_seen": 406978560
},
{
"epoch": 0.12,
"learning_rate": 0.00044275827482447347,
"loss": 2.8901,
"theoretical_loss": 4.004134974366036,
"tokens_seen": 407044096
},
{
"epoch": 0.12,
"learning_rate": 0.0004427482447342026,
"loss": 3.004,
"theoretical_loss": 4.004063066425874,
"tokens_seen": 407109632
},
{
"epoch": 0.12,
"learning_rate": 0.00044273821464393184,
"loss": 2.8771,
"theoretical_loss": 4.003991173301018,
"tokens_seen": 407175168
},
{
"epoch": 0.12,
"learning_rate": 0.00044272818455366096,
"loss": 2.9344,
"theoretical_loss": 4.003919294986032,
"tokens_seen": 407240704
},
{
"epoch": 0.12,
"learning_rate": 0.0004427181544633902,
"loss": 2.919,
"theoretical_loss": 4.003847431475482,
"tokens_seen": 407306240
},
{
"epoch": 0.12,
"learning_rate": 0.0004427081243731194,
"loss": 2.9829,
"theoretical_loss": 4.003775582763939,
"tokens_seen": 407371776
},
{
"epoch": 0.12,
"learning_rate": 0.00044269809428284856,
"loss": 2.9322,
"theoretical_loss": 4.0037037488459735,
"tokens_seen": 407437312
},
{
"epoch": 0.12,
"learning_rate": 0.00044268806419257774,
"loss": 2.9934,
"theoretical_loss": 4.003631929716162,
"tokens_seen": 407502848
},
{
"epoch": 0.12,
"learning_rate": 0.0004426780341023069,
"loss": 3.0835,
"theoretical_loss": 4.0035601253690825,
"tokens_seen": 407568384
},
{
"epoch": 0.12,
"learning_rate": 0.0004426680040120361,
"loss": 3.2281,
"theoretical_loss": 4.0034883357993145,
"tokens_seen": 407633920
},
{
"epoch": 0.12,
"learning_rate": 0.00044265797392176534,
"loss": 3.0159,
"theoretical_loss": 4.003416561001444,
"tokens_seen": 407699456
},
{
"epoch": 0.12,
"learning_rate": 0.00044264794383149447,
"loss": 3.0396,
"theoretical_loss": 4.003344800970056,
"tokens_seen": 407764992
},
{
"epoch": 0.12,
"learning_rate": 0.0004426379137412237,
"loss": 3.1254,
"theoretical_loss": 4.003273055699741,
"tokens_seen": 407830528
},
{
"epoch": 0.12,
"learning_rate": 0.0004426278836509529,
"loss": 2.7877,
"theoretical_loss": 4.00320132518509,
"tokens_seen": 407896064
},
{
"epoch": 0.12,
"objective/train/docs_used": 242638,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4354865550994873,
"objective/train/theoretical_loss": 4.0031296094207,
"objective/train/tokens_used": 428421600,
"theoretical_loss": 4.0031296094207,
"tokens_seen": 407961600
},
{
"epoch": 0.12,
"learning_rate": 0.00044261785356068206,
"loss": 2.9397,
"theoretical_loss": 4.0031296094207,
"tokens_seen": 407961600
},
{
"epoch": 0.12,
"learning_rate": 0.00044260782347041124,
"loss": 2.9511,
"theoretical_loss": 4.003057908401168,
"tokens_seen": 408027136
},
{
"epoch": 0.12,
"learning_rate": 0.0004425977933801404,
"loss": 2.7822,
"theoretical_loss": 4.002986222121095,
"tokens_seen": 408092672
},
{
"epoch": 0.12,
"learning_rate": 0.0004425877632898696,
"loss": 3.104,
"theoretical_loss": 4.002914550575085,
"tokens_seen": 408158208
},
{
"epoch": 0.12,
"learning_rate": 0.00044257773319959884,
"loss": 3.1355,
"theoretical_loss": 4.002842893757744,
"tokens_seen": 408223744
},
{
"epoch": 0.12,
"learning_rate": 0.00044256770310932797,
"loss": 2.9234,
"theoretical_loss": 4.002771251663681,
"tokens_seen": 408289280
},
{
"epoch": 0.12,
"learning_rate": 0.0004425576730190572,
"loss": 2.8301,
"theoretical_loss": 4.00269962428751,
"tokens_seen": 408354816
},
{
"epoch": 0.12,
"learning_rate": 0.00044254764292878633,
"loss": 3.0186,
"theoretical_loss": 4.0026280116238455,
"tokens_seen": 408420352
},
{
"epoch": 0.12,
"learning_rate": 0.00044253761283851557,
"loss": 3.0234,
"theoretical_loss": 4.002556413667305,
"tokens_seen": 408485888
},
{
"epoch": 0.12,
"learning_rate": 0.00044252758274824475,
"loss": 2.8821,
"theoretical_loss": 4.0024848304125085,
"tokens_seen": 408551424
},
{
"epoch": 0.12,
"learning_rate": 0.00044251755265797393,
"loss": 2.8889,
"theoretical_loss": 4.002413261854081,
"tokens_seen": 408616960
},
{
"epoch": 0.12,
"learning_rate": 0.0004425075225677031,
"loss": 2.8424,
"theoretical_loss": 4.002341707986648,
"tokens_seen": 408682496
},
{
"epoch": 0.12,
"learning_rate": 0.0004424974924774323,
"loss": 3.0701,
"theoretical_loss": 4.00227016880484,
"tokens_seen": 408748032
},
{
"epoch": 0.12,
"learning_rate": 0.00044248746238716147,
"loss": 2.7459,
"theoretical_loss": 4.002198644303288,
"tokens_seen": 408813568
},
{
"epoch": 0.12,
"learning_rate": 0.0004424774322968907,
"loss": 2.9571,
"theoretical_loss": 4.002127134476626,
"tokens_seen": 408879104
},
{
"epoch": 0.12,
"learning_rate": 0.00044246740220661983,
"loss": 2.8098,
"theoretical_loss": 4.002055639319494,
"tokens_seen": 408944640
},
{
"epoch": 0.12,
"learning_rate": 0.00044245737211634907,
"loss": 2.9085,
"theoretical_loss": 4.00198415882653,
"tokens_seen": 409010176
},
{
"epoch": 0.12,
"learning_rate": 0.00044244734202607825,
"loss": 3.1644,
"theoretical_loss": 4.001912692992379,
"tokens_seen": 409075712
},
{
"epoch": 0.12,
"learning_rate": 0.00044243731193580743,
"loss": 3.0164,
"theoretical_loss": 4.001841241811688,
"tokens_seen": 409141248
},
{
"epoch": 0.12,
"learning_rate": 0.0004424272818455366,
"loss": 2.7907,
"theoretical_loss": 4.001769805279103,
"tokens_seen": 409206784
},
{
"epoch": 0.12,
"learning_rate": 0.0004424172517552658,
"loss": 2.9686,
"theoretical_loss": 4.001698383389279,
"tokens_seen": 409272320
},
{
"epoch": 0.12,
"learning_rate": 0.000442407221664995,
"loss": 2.8624,
"theoretical_loss": 4.001626976136868,
"tokens_seen": 409337856
},
{
"epoch": 0.12,
"learning_rate": 0.0004423971915747242,
"loss": 3.0741,
"theoretical_loss": 4.001555583516529,
"tokens_seen": 409403392
},
{
"epoch": 0.12,
"learning_rate": 0.00044238716148445334,
"loss": 3.0561,
"theoretical_loss": 4.001484205522922,
"tokens_seen": 409468928
},
{
"epoch": 0.12,
"learning_rate": 0.00044237713139418257,
"loss": 2.9995,
"theoretical_loss": 4.00141284215071,
"tokens_seen": 409534464
},
{
"epoch": 0.12,
"objective/train/docs_used": 244086,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1447839736938477,
"objective/train/theoretical_loss": 4.001341493394558,
"objective/train/tokens_used": 430060000,
"theoretical_loss": 4.001341493394558,
"tokens_seen": 409600000
},
{
"epoch": 0.12,
"learning_rate": 0.0004423671013039117,
"loss": 2.9803,
"theoretical_loss": 4.001341493394558,
"tokens_seen": 409600000
},
{
"epoch": 0.12,
"learning_rate": 0.00044235707121364093,
"loss": 2.9221,
"theoretical_loss": 4.0012701592491355,
"tokens_seen": 409665536
},
{
"epoch": 0.12,
"learning_rate": 0.0004423470411233701,
"loss": 3.0829,
"theoretical_loss": 4.001198839709113,
"tokens_seen": 409731072
},
{
"epoch": 0.12,
"learning_rate": 0.0004423370110330993,
"loss": 2.9286,
"theoretical_loss": 4.001127534769166,
"tokens_seen": 409796608
},
{
"epoch": 0.12,
"learning_rate": 0.0004423269809428285,
"loss": 3.0952,
"theoretical_loss": 4.001056244423971,
"tokens_seen": 409862144
},
{
"epoch": 0.12,
"learning_rate": 0.0004423169508525577,
"loss": 2.9253,
"theoretical_loss": 4.000984968668206,
"tokens_seen": 409927680
},
{
"epoch": 0.12,
"learning_rate": 0.00044230692076228684,
"loss": 2.7064,
"theoretical_loss": 4.000913707496556,
"tokens_seen": 409993216
},
{
"epoch": 0.12,
"learning_rate": 0.0004422968906720161,
"loss": 2.9732,
"theoretical_loss": 4.000842460903706,
"tokens_seen": 410058752
},
{
"epoch": 0.12,
"learning_rate": 0.0004422868605817452,
"loss": 3.0309,
"theoretical_loss": 4.000771228884344,
"tokens_seen": 410124288
},
{
"epoch": 0.12,
"learning_rate": 0.00044227683049147444,
"loss": 2.8903,
"theoretical_loss": 4.0007000114331595,
"tokens_seen": 410189824
},
{
"epoch": 0.12,
"learning_rate": 0.0004422668004012036,
"loss": 2.9111,
"theoretical_loss": 4.000628808544848,
"tokens_seen": 410255360
},
{
"epoch": 0.12,
"learning_rate": 0.0004422567703109328,
"loss": 3.1261,
"theoretical_loss": 4.000557620214105,
"tokens_seen": 410320896
},
{
"epoch": 0.12,
"learning_rate": 0.000442246740220662,
"loss": 3.0598,
"theoretical_loss": 4.000486446435629,
"tokens_seen": 410386432
},
{
"epoch": 0.12,
"learning_rate": 0.00044223671013039116,
"loss": 2.9045,
"theoretical_loss": 4.000415287204124,
"tokens_seen": 410451968
},
{
"epoch": 0.12,
"learning_rate": 0.00044222668004012034,
"loss": 2.5777,
"theoretical_loss": 4.000344142514294,
"tokens_seen": 410517504
},
{
"epoch": 0.12,
"learning_rate": 0.0004422166499498496,
"loss": 2.847,
"theoretical_loss": 4.0002730123608465,
"tokens_seen": 410583040
},
{
"epoch": 0.12,
"learning_rate": 0.0004422066198595787,
"loss": 2.9707,
"theoretical_loss": 4.000201896738492,
"tokens_seen": 410648576
},
{
"epoch": 0.12,
"learning_rate": 0.00044219658976930794,
"loss": 2.8757,
"theoretical_loss": 4.000130795641944,
"tokens_seen": 410714112
},
{
"epoch": 0.12,
"learning_rate": 0.0004421865596790371,
"loss": 2.6645,
"theoretical_loss": 4.0000597090659165,
"tokens_seen": 410779648
},
{
"epoch": 0.12,
"learning_rate": 0.0004421765295887663,
"loss": 2.6402,
"theoretical_loss": 3.9999886370051305,
"tokens_seen": 410845184
},
{
"epoch": 0.12,
"learning_rate": 0.00044216649949849554,
"loss": 2.8724,
"theoretical_loss": 3.9999175794543063,
"tokens_seen": 410910720
},
{
"epoch": 0.12,
"learning_rate": 0.00044215646940822467,
"loss": 2.8293,
"theoretical_loss": 3.9998465364081683,
"tokens_seen": 410976256
},
{
"epoch": 0.12,
"learning_rate": 0.0004421464393179539,
"loss": 2.8997,
"theoretical_loss": 3.999775507861443,
"tokens_seen": 411041792
},
{
"epoch": 0.12,
"learning_rate": 0.0004421364092276831,
"loss": 2.8315,
"theoretical_loss": 3.999704493808861,
"tokens_seen": 411107328
},
{
"epoch": 0.12,
"learning_rate": 0.00044212637913741226,
"loss": 3.1415,
"theoretical_loss": 3.9996334942451544,
"tokens_seen": 411172864
},
{
"epoch": 0.12,
"objective/train/docs_used": 244630,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0692129135131836,
"objective/train/theoretical_loss": 3.999562509165058,
"objective/train/tokens_used": 431698400,
"theoretical_loss": 3.999562509165058,
"tokens_seen": 411238400
},
{
"epoch": 0.12,
"learning_rate": 0.00044211634904714144,
"loss": 3.1408,
"theoretical_loss": 3.999562509165058,
"tokens_seen": 411238400
},
{
"epoch": 0.12,
"learning_rate": 0.0004421063189568706,
"loss": 2.9589,
"theoretical_loss": 3.9994915385633103,
"tokens_seen": 411303936
},
{
"epoch": 0.12,
"learning_rate": 0.0004420962888665998,
"loss": 3.0151,
"theoretical_loss": 3.9994205824346514,
"tokens_seen": 411369472
},
{
"epoch": 0.12,
"learning_rate": 0.00044208625877632904,
"loss": 3.0807,
"theoretical_loss": 3.9993496407738256,
"tokens_seen": 411435008
},
{
"epoch": 0.12,
"learning_rate": 0.00044207622868605817,
"loss": 2.8219,
"theoretical_loss": 3.9992787135755785,
"tokens_seen": 411500544
},
{
"epoch": 0.12,
"learning_rate": 0.0004420661985957874,
"loss": 2.8975,
"theoretical_loss": 3.999207800834659,
"tokens_seen": 411566080
},
{
"epoch": 0.12,
"learning_rate": 0.00044205616850551653,
"loss": 2.7822,
"theoretical_loss": 3.9991369025458194,
"tokens_seen": 411631616
},
{
"epoch": 0.12,
"learning_rate": 0.00044204613841524577,
"loss": 2.8842,
"theoretical_loss": 3.999066018703814,
"tokens_seen": 411697152
},
{
"epoch": 0.12,
"learning_rate": 0.00044203610832497495,
"loss": 2.8709,
"theoretical_loss": 3.9989951493033997,
"tokens_seen": 411762688
},
{
"epoch": 0.12,
"learning_rate": 0.00044202607823470413,
"loss": 3.0365,
"theoretical_loss": 3.9989242943393357,
"tokens_seen": 411828224
},
{
"epoch": 0.12,
"learning_rate": 0.0004420160481444333,
"loss": 2.9622,
"theoretical_loss": 3.998853453806386,
"tokens_seen": 411893760
},
{
"epoch": 0.12,
"learning_rate": 0.0004420060180541625,
"loss": 2.8304,
"theoretical_loss": 3.998782627699315,
"tokens_seen": 411959296
},
{
"epoch": 0.12,
"learning_rate": 0.00044199598796389167,
"loss": 3.0473,
"theoretical_loss": 3.998711816012891,
"tokens_seen": 412024832
},
{
"epoch": 0.12,
"learning_rate": 0.0004419859578736209,
"loss": 2.8868,
"theoretical_loss": 3.9986410187418855,
"tokens_seen": 412090368
},
{
"epoch": 0.12,
"learning_rate": 0.00044197592778335003,
"loss": 2.9131,
"theoretical_loss": 3.9985702358810715,
"tokens_seen": 412155904
},
{
"epoch": 0.12,
"learning_rate": 0.00044196589769307927,
"loss": 2.8187,
"theoretical_loss": 3.9984994674252254,
"tokens_seen": 412221440
},
{
"epoch": 0.12,
"learning_rate": 0.00044195586760280845,
"loss": 3.0161,
"theoretical_loss": 3.998428713369126,
"tokens_seen": 412286976
},
{
"epoch": 0.12,
"learning_rate": 0.00044194583751253763,
"loss": 2.8992,
"theoretical_loss": 3.9983579737075554,
"tokens_seen": 412352512
},
{
"epoch": 0.12,
"learning_rate": 0.0004419358074222668,
"loss": 2.9643,
"theoretical_loss": 3.9982872484352976,
"tokens_seen": 412418048
},
{
"epoch": 0.12,
"learning_rate": 0.000441925777331996,
"loss": 3.0318,
"theoretical_loss": 3.9982165375471403,
"tokens_seen": 412483584
},
{
"epoch": 0.13,
"learning_rate": 0.0004419157472417252,
"loss": 2.8583,
"theoretical_loss": 3.998145841037873,
"tokens_seen": 412549120
},
{
"epoch": 0.13,
"learning_rate": 0.0004419057171514544,
"loss": 2.955,
"theoretical_loss": 3.9980751589022887,
"tokens_seen": 412614656
},
{
"epoch": 0.13,
"learning_rate": 0.00044189568706118354,
"loss": 3.0281,
"theoretical_loss": 3.9980044911351826,
"tokens_seen": 412680192
},
{
"epoch": 0.13,
"learning_rate": 0.00044188565697091277,
"loss": 2.9292,
"theoretical_loss": 3.9979338377313525,
"tokens_seen": 412745728
},
{
"epoch": 0.13,
"learning_rate": 0.0004418756268806419,
"loss": 2.8697,
"theoretical_loss": 3.997863198685599,
"tokens_seen": 412811264
},
{
"epoch": 0.13,
"objective/train/docs_used": 245807,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.571214437484741,
"objective/train/theoretical_loss": 3.997792573992726,
"objective/train/tokens_used": 433336800,
"theoretical_loss": 3.997792573992726,
"tokens_seen": 412876800
},
{
"epoch": 0.13,
"learning_rate": 0.00044186559679037113,
"loss": 2.7269,
"theoretical_loss": 3.997792573992726,
"tokens_seen": 412876800
},
{
"epoch": 0.13,
"learning_rate": 0.0004418555667001003,
"loss": 3.0854,
"theoretical_loss": 3.997721963647539,
"tokens_seen": 412942336
},
{
"epoch": 0.13,
"learning_rate": 0.0004418455366098295,
"loss": 3.036,
"theoretical_loss": 3.997651367644848,
"tokens_seen": 413007872
},
{
"epoch": 0.13,
"learning_rate": 0.0004418355065195587,
"loss": 3.1661,
"theoretical_loss": 3.997580785979464,
"tokens_seen": 413073408
},
{
"epoch": 0.13,
"learning_rate": 0.0004418254764292879,
"loss": 3.2061,
"theoretical_loss": 3.9975102186462004,
"tokens_seen": 413138944
},
{
"epoch": 0.13,
"learning_rate": 0.00044181544633901704,
"loss": 2.8441,
"theoretical_loss": 3.997439665639875,
"tokens_seen": 413204480
},
{
"epoch": 0.13,
"learning_rate": 0.0004418054162487463,
"loss": 2.9107,
"theoretical_loss": 3.9973691269553076,
"tokens_seen": 413270016
},
{
"epoch": 0.13,
"learning_rate": 0.0004417953861584754,
"loss": 3.2179,
"theoretical_loss": 3.9972986025873203,
"tokens_seen": 413335552
},
{
"epoch": 0.13,
"learning_rate": 0.00044178535606820464,
"loss": 2.8222,
"theoretical_loss": 3.9972280925307384,
"tokens_seen": 413401088
},
{
"epoch": 0.13,
"learning_rate": 0.0004417753259779338,
"loss": 2.9046,
"theoretical_loss": 3.9971575967803887,
"tokens_seen": 413466624
},
{
"epoch": 0.13,
"learning_rate": 0.000441765295887663,
"loss": 2.9597,
"theoretical_loss": 3.9970871153311025,
"tokens_seen": 413532160
},
{
"epoch": 0.13,
"learning_rate": 0.0004417552657973922,
"loss": 2.7283,
"theoretical_loss": 3.9970166481777127,
"tokens_seen": 413597696
},
{
"epoch": 0.13,
"learning_rate": 0.00044174523570712136,
"loss": 2.9719,
"theoretical_loss": 3.996946195315055,
"tokens_seen": 413663232
},
{
"epoch": 0.13,
"learning_rate": 0.00044173520561685054,
"loss": 3.0133,
"theoretical_loss": 3.996875756737968,
"tokens_seen": 413728768
},
{
"epoch": 0.13,
"learning_rate": 0.0004417251755265798,
"loss": 2.9793,
"theoretical_loss": 3.996805332441293,
"tokens_seen": 413794304
},
{
"epoch": 0.13,
"learning_rate": 0.0004417151454363089,
"loss": 2.6928,
"theoretical_loss": 3.996734922419873,
"tokens_seen": 413859840
},
{
"epoch": 0.13,
"learning_rate": 0.00044170511534603814,
"loss": 2.9173,
"theoretical_loss": 3.996664526668555,
"tokens_seen": 413925376
},
{
"epoch": 0.13,
"learning_rate": 0.00044169508525576727,
"loss": 3.0386,
"theoretical_loss": 3.9965941451821894,
"tokens_seen": 413990912
},
{
"epoch": 0.13,
"learning_rate": 0.0004416850551654965,
"loss": 2.9614,
"theoretical_loss": 3.996523777955626,
"tokens_seen": 414056448
},
{
"epoch": 0.13,
"learning_rate": 0.0004416750250752257,
"loss": 3.1306,
"theoretical_loss": 3.996453424983721,
"tokens_seen": 414121984
},
{
"epoch": 0.13,
"learning_rate": 0.00044166499498495487,
"loss": 2.9258,
"theoretical_loss": 3.9963830862613303,
"tokens_seen": 414187520
},
{
"epoch": 0.13,
"learning_rate": 0.00044165496489468405,
"loss": 3.0298,
"theoretical_loss": 3.9963127617833143,
"tokens_seen": 414253056
},
{
"epoch": 0.13,
"learning_rate": 0.0004416449348044133,
"loss": 2.9546,
"theoretical_loss": 3.9962424515445356,
"tokens_seen": 414318592
},
{
"epoch": 0.13,
"learning_rate": 0.0004416349047141424,
"loss": 2.87,
"theoretical_loss": 3.99617215553986,
"tokens_seen": 414384128
},
{
"epoch": 0.13,
"learning_rate": 0.00044162487462387164,
"loss": 3.0131,
"theoretical_loss": 3.9961018737641547,
"tokens_seen": 414449664
},
{
"epoch": 0.13,
"objective/train/docs_used": 246540,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8978941440582275,
"objective/train/theoretical_loss": 3.99603160621229,
"objective/train/tokens_used": 434975200,
"theoretical_loss": 3.99603160621229,
"tokens_seen": 414515200
},
{
"epoch": 0.13,
"learning_rate": 0.00044161484453360077,
"loss": 2.9393,
"theoretical_loss": 3.99603160621229,
"tokens_seen": 414515200
},
{
"epoch": 0.13,
"learning_rate": 0.00044160481444333,
"loss": 2.9509,
"theoretical_loss": 3.9959613528791396,
"tokens_seen": 414580736
},
{
"epoch": 0.13,
"learning_rate": 0.0004415947843530592,
"loss": 2.7901,
"theoretical_loss": 3.995891113759579,
"tokens_seen": 414646272
},
{
"epoch": 0.13,
"learning_rate": 0.00044158475426278837,
"loss": 2.9503,
"theoretical_loss": 3.9958208888484865,
"tokens_seen": 414711808
},
{
"epoch": 0.13,
"learning_rate": 0.00044157472417251755,
"loss": 3.1345,
"theoretical_loss": 3.9957506781407446,
"tokens_seen": 414777344
},
{
"epoch": 0.13,
"learning_rate": 0.00044156469408224673,
"loss": 2.7583,
"theoretical_loss": 3.9956804816312355,
"tokens_seen": 414842880
},
{
"epoch": 0.13,
"learning_rate": 0.0004415546639919759,
"loss": 2.9844,
"theoretical_loss": 3.995610299314846,
"tokens_seen": 414908416
},
{
"epoch": 0.13,
"learning_rate": 0.00044154463390170515,
"loss": 2.9255,
"theoretical_loss": 3.9955401311864662,
"tokens_seen": 414973952
},
{
"epoch": 0.13,
"learning_rate": 0.0004415346038114343,
"loss": 3.0467,
"theoretical_loss": 3.9954699772409867,
"tokens_seen": 415039488
},
{
"epoch": 0.13,
"learning_rate": 0.0004415245737211635,
"loss": 3.0262,
"theoretical_loss": 3.995399837473302,
"tokens_seen": 415105024
},
{
"epoch": 0.13,
"learning_rate": 0.00044151454363089264,
"loss": 2.9524,
"theoretical_loss": 3.9953297118783104,
"tokens_seen": 415170560
},
{
"epoch": 0.13,
"learning_rate": 0.00044150451354062187,
"loss": 3.0535,
"theoretical_loss": 3.99525960045091,
"tokens_seen": 415236096
},
{
"epoch": 0.13,
"learning_rate": 0.00044149448345035105,
"loss": 2.8636,
"theoretical_loss": 3.9951895031860043,
"tokens_seen": 415301632
},
{
"epoch": 0.13,
"learning_rate": 0.00044148445336008023,
"loss": 2.6274,
"theoretical_loss": 3.9951194200784976,
"tokens_seen": 415367168
},
{
"epoch": 0.13,
"learning_rate": 0.0004414744232698094,
"loss": 3.1258,
"theoretical_loss": 3.9950493511232974,
"tokens_seen": 415432704
},
{
"epoch": 0.13,
"learning_rate": 0.00044146439317953865,
"loss": 3.0572,
"theoretical_loss": 3.9949792963153143,
"tokens_seen": 415498240
},
{
"epoch": 0.13,
"learning_rate": 0.0004414543630892678,
"loss": 3.019,
"theoretical_loss": 3.9949092556494614,
"tokens_seen": 415563776
},
{
"epoch": 0.13,
"learning_rate": 0.000441444332998997,
"loss": 2.9243,
"theoretical_loss": 3.994839229120654,
"tokens_seen": 415629312
},
{
"epoch": 0.13,
"learning_rate": 0.0004414343029087262,
"loss": 3.0326,
"theoretical_loss": 3.99476921672381,
"tokens_seen": 415694848
},
{
"epoch": 0.13,
"learning_rate": 0.0004414242728184554,
"loss": 3.0729,
"theoretical_loss": 3.99469921845385,
"tokens_seen": 415760384
},
{
"epoch": 0.13,
"learning_rate": 0.0004414142427281846,
"loss": 2.4204,
"theoretical_loss": 3.994629234305698,
"tokens_seen": 415825920
},
{
"epoch": 0.13,
"learning_rate": 0.00044140421263791374,
"loss": 2.8157,
"theoretical_loss": 3.9945592642742795,
"tokens_seen": 415891456
},
{
"epoch": 0.13,
"learning_rate": 0.00044139418254764297,
"loss": 3.0637,
"theoretical_loss": 3.994489308354523,
"tokens_seen": 415956992
},
{
"epoch": 0.13,
"learning_rate": 0.0004413841524573721,
"loss": 3.0211,
"theoretical_loss": 3.9944193665413605,
"tokens_seen": 416022528
},
{
"epoch": 0.13,
"learning_rate": 0.00044137412236710133,
"loss": 2.8919,
"theoretical_loss": 3.9943494388297256,
"tokens_seen": 416088064
},
{
"epoch": 0.13,
"objective/train/docs_used": 247824,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.898556709289551,
"objective/train/theoretical_loss": 3.994279525214554,
"objective/train/tokens_used": 436613600,
"theoretical_loss": 3.994279525214554,
"tokens_seen": 416153600
},
{
"epoch": 0.13,
"learning_rate": 0.0004413640922768305,
"loss": 3.0149,
"theoretical_loss": 3.994279525214554,
"tokens_seen": 416153600
},
{
"epoch": 0.13,
"learning_rate": 0.0004413540621865597,
"loss": 2.8584,
"theoretical_loss": 3.9942096256907864,
"tokens_seen": 416219136
},
{
"epoch": 0.13,
"learning_rate": 0.0004413440320962889,
"loss": 3.0482,
"theoretical_loss": 3.994139740253363,
"tokens_seen": 416284672
},
{
"epoch": 0.13,
"learning_rate": 0.0004413340020060181,
"loss": 2.8312,
"theoretical_loss": 3.9940698688972285,
"tokens_seen": 416350208
},
{
"epoch": 0.13,
"learning_rate": 0.00044132397191574724,
"loss": 2.9454,
"theoretical_loss": 3.99400001161733,
"tokens_seen": 416415744
},
{
"epoch": 0.13,
"learning_rate": 0.0004413139418254765,
"loss": 3.0826,
"theoretical_loss": 3.9939301684086175,
"tokens_seen": 416481280
},
{
"epoch": 0.13,
"learning_rate": 0.0004413039117352056,
"loss": 2.4992,
"theoretical_loss": 3.9938603392660426,
"tokens_seen": 416546816
},
{
"epoch": 0.13,
"learning_rate": 0.00044129388164493484,
"loss": 2.8932,
"theoretical_loss": 3.9937905241845595,
"tokens_seen": 416612352
},
{
"epoch": 0.13,
"learning_rate": 0.000441283851554664,
"loss": 2.8243,
"theoretical_loss": 3.993720723159127,
"tokens_seen": 416677888
},
{
"epoch": 0.13,
"learning_rate": 0.0004412738214643932,
"loss": 2.9899,
"theoretical_loss": 3.9936509361847037,
"tokens_seen": 416743424
},
{
"epoch": 0.13,
"learning_rate": 0.0004412637913741224,
"loss": 2.5832,
"theoretical_loss": 3.9935811632562532,
"tokens_seen": 416808960
},
{
"epoch": 0.13,
"learning_rate": 0.00044125376128385156,
"loss": 2.8921,
"theoretical_loss": 3.99351140436874,
"tokens_seen": 416874496
},
{
"epoch": 0.13,
"learning_rate": 0.00044124373119358074,
"loss": 2.8575,
"theoretical_loss": 3.9934416595171323,
"tokens_seen": 416940032
},
{
"epoch": 0.13,
"learning_rate": 0.00044123370110331,
"loss": 2.92,
"theoretical_loss": 3.9933719286963996,
"tokens_seen": 417005568
},
{
"epoch": 0.13,
"learning_rate": 0.0004412236710130391,
"loss": 3.0291,
"theoretical_loss": 3.993302211901516,
"tokens_seen": 417071104
},
{
"epoch": 0.13,
"learning_rate": 0.00044121364092276834,
"loss": 3.0544,
"theoretical_loss": 3.9932325091274565,
"tokens_seen": 417136640
},
{
"epoch": 0.13,
"learning_rate": 0.00044120361083249747,
"loss": 2.965,
"theoretical_loss": 3.993162820369199,
"tokens_seen": 417202176
},
{
"epoch": 0.13,
"learning_rate": 0.0004411935807422267,
"loss": 3.0694,
"theoretical_loss": 3.9930931456217245,
"tokens_seen": 417267712
},
{
"epoch": 0.13,
"learning_rate": 0.0004411835506519559,
"loss": 2.9183,
"theoretical_loss": 3.993023484880016,
"tokens_seen": 417333248
},
{
"epoch": 0.13,
"learning_rate": 0.00044117352056168507,
"loss": 2.9882,
"theoretical_loss": 3.9929538381390595,
"tokens_seen": 417398784
},
{
"epoch": 0.13,
"learning_rate": 0.00044116349047141425,
"loss": 2.7475,
"theoretical_loss": 3.9928842053938443,
"tokens_seen": 417464320
},
{
"epoch": 0.13,
"learning_rate": 0.0004411534603811435,
"loss": 3.0143,
"theoretical_loss": 3.9928145866393594,
"tokens_seen": 417529856
},
{
"epoch": 0.13,
"learning_rate": 0.0004411434302908726,
"loss": 3.0615,
"theoretical_loss": 3.9927449818706005,
"tokens_seen": 417595392
},
{
"epoch": 0.13,
"learning_rate": 0.00044113340020060184,
"loss": 3.0332,
"theoretical_loss": 3.992675391082563,
"tokens_seen": 417660928
},
{
"epoch": 0.13,
"learning_rate": 0.00044112337011033097,
"loss": 2.7898,
"theoretical_loss": 3.9926058142702456,
"tokens_seen": 417726464
},
{
"epoch": 0.13,
"objective/train/docs_used": 248559,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1589126586914062,
"objective/train/theoretical_loss": 3.9925362514286498,
"objective/train/tokens_used": 438252000,
"theoretical_loss": 3.9925362514286498,
"tokens_seen": 417792000
},
{
"epoch": 0.13,
"learning_rate": 0.0004411133400200602,
"loss": 2.8932,
"theoretical_loss": 3.9925362514286498,
"tokens_seen": 417792000
},
{
"epoch": 0.13,
"learning_rate": 0.0004411033099297894,
"loss": 3.1841,
"theoretical_loss": 3.992466702552779,
"tokens_seen": 417857536
},
{
"epoch": 0.13,
"learning_rate": 0.00044109327983951857,
"loss": 2.9222,
"theoretical_loss": 3.99239716763764,
"tokens_seen": 417923072
},
{
"epoch": 0.13,
"learning_rate": 0.00044108324974924775,
"loss": 3.1899,
"theoretical_loss": 3.9923276466782425,
"tokens_seen": 417988608
},
{
"epoch": 0.13,
"learning_rate": 0.00044107321965897693,
"loss": 2.9757,
"theoretical_loss": 3.9922581396695973,
"tokens_seen": 418054144
},
{
"epoch": 0.13,
"learning_rate": 0.0004410631895687061,
"loss": 3.003,
"theoretical_loss": 3.992188646606719,
"tokens_seen": 418119680
},
{
"epoch": 0.13,
"learning_rate": 0.00044105315947843535,
"loss": 2.766,
"theoretical_loss": 3.992119167484624,
"tokens_seen": 418185216
},
{
"epoch": 0.13,
"learning_rate": 0.0004410431293881645,
"loss": 2.8755,
"theoretical_loss": 3.992049702298332,
"tokens_seen": 418250752
},
{
"epoch": 0.13,
"learning_rate": 0.0004410330992978937,
"loss": 2.818,
"theoretical_loss": 3.9919802510428646,
"tokens_seen": 418316288
},
{
"epoch": 0.13,
"learning_rate": 0.00044102306920762284,
"loss": 2.9323,
"theoretical_loss": 3.991910813713246,
"tokens_seen": 418381824
},
{
"epoch": 0.13,
"learning_rate": 0.00044101303911735207,
"loss": 2.9617,
"theoretical_loss": 3.991841390304504,
"tokens_seen": 418447360
},
{
"epoch": 0.13,
"learning_rate": 0.00044100300902708125,
"loss": 3.0045,
"theoretical_loss": 3.991771980811667,
"tokens_seen": 418512896
},
{
"epoch": 0.13,
"learning_rate": 0.00044099297893681043,
"loss": 2.9303,
"theoretical_loss": 3.991702585229768,
"tokens_seen": 418578432
},
{
"epoch": 0.13,
"learning_rate": 0.0004409829488465396,
"loss": 3.0384,
"theoretical_loss": 3.9916332035538415,
"tokens_seen": 418643968
},
{
"epoch": 0.13,
"learning_rate": 0.00044097291875626885,
"loss": 2.9644,
"theoretical_loss": 3.9915638357789245,
"tokens_seen": 418709504
},
{
"epoch": 0.13,
"learning_rate": 0.000440962888665998,
"loss": 2.9537,
"theoretical_loss": 3.991494481900056,
"tokens_seen": 418775040
},
{
"epoch": 0.13,
"learning_rate": 0.0004409528585757272,
"loss": 3.1591,
"theoretical_loss": 3.9914251419122797,
"tokens_seen": 418840576
},
{
"epoch": 0.13,
"learning_rate": 0.00044094282848545634,
"loss": 2.9929,
"theoretical_loss": 3.99135581581064,
"tokens_seen": 418906112
},
{
"epoch": 0.13,
"learning_rate": 0.0004409327983951856,
"loss": 2.9587,
"theoretical_loss": 3.991286503590184,
"tokens_seen": 418971648
},
{
"epoch": 0.13,
"learning_rate": 0.00044092276830491476,
"loss": 2.9491,
"theoretical_loss": 3.9912172052459614,
"tokens_seen": 419037184
},
{
"epoch": 0.13,
"learning_rate": 0.00044091273821464394,
"loss": 3.0031,
"theoretical_loss": 3.9911479207730247,
"tokens_seen": 419102720
},
{
"epoch": 0.13,
"learning_rate": 0.0004409027081243731,
"loss": 3.0363,
"theoretical_loss": 3.991078650166429,
"tokens_seen": 419168256
},
{
"epoch": 0.13,
"learning_rate": 0.0004408926780341023,
"loss": 2.9847,
"theoretical_loss": 3.9910093934212325,
"tokens_seen": 419233792
},
{
"epoch": 0.13,
"learning_rate": 0.0004408826479438315,
"loss": 2.8094,
"theoretical_loss": 3.990940150532494,
"tokens_seen": 419299328
},
{
"epoch": 0.13,
"learning_rate": 0.0004408726178535607,
"loss": 2.831,
"theoretical_loss": 3.990870921495277,
"tokens_seen": 419364864
},
{
"epoch": 0.13,
"objective/train/docs_used": 249785,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2360668182373047,
"objective/train/theoretical_loss": 3.990801706304647,
"objective/train/tokens_used": 439890400,
"theoretical_loss": 3.990801706304647,
"tokens_seen": 419430400
},
{
"epoch": 0.13,
"learning_rate": 0.00044086258776328984,
"loss": 3.028,
"theoretical_loss": 3.990801706304647,
"tokens_seen": 419430400
},
{
"epoch": 0.13,
"learning_rate": 0.0004408525576730191,
"loss": 3.0799,
"theoretical_loss": 3.99073250495567,
"tokens_seen": 419495936
},
{
"epoch": 0.13,
"learning_rate": 0.0004408425275827482,
"loss": 3.1414,
"theoretical_loss": 3.990663317443418,
"tokens_seen": 419561472
},
{
"epoch": 0.13,
"learning_rate": 0.00044083249749247744,
"loss": 2.9101,
"theoretical_loss": 3.990594143762962,
"tokens_seen": 419627008
},
{
"epoch": 0.13,
"learning_rate": 0.0004408224674022066,
"loss": 2.9956,
"theoretical_loss": 3.9905249839093786,
"tokens_seen": 419692544
},
{
"epoch": 0.13,
"learning_rate": 0.0004408124373119358,
"loss": 2.7741,
"theoretical_loss": 3.9904558378777453,
"tokens_seen": 419758080
},
{
"epoch": 0.13,
"learning_rate": 0.000440802407221665,
"loss": 2.8678,
"theoretical_loss": 3.990386705663142,
"tokens_seen": 419823616
},
{
"epoch": 0.13,
"learning_rate": 0.0004407923771313942,
"loss": 2.7025,
"theoretical_loss": 3.9903175872606518,
"tokens_seen": 419889152
},
{
"epoch": 0.13,
"learning_rate": 0.00044078234704112335,
"loss": 2.9144,
"theoretical_loss": 3.990248482665359,
"tokens_seen": 419954688
},
{
"epoch": 0.13,
"learning_rate": 0.0004407723169508526,
"loss": 2.9458,
"theoretical_loss": 3.9901793918723527,
"tokens_seen": 420020224
},
{
"epoch": 0.13,
"learning_rate": 0.0004407622868605817,
"loss": 2.8895,
"theoretical_loss": 3.9901103148767225,
"tokens_seen": 420085760
},
{
"epoch": 0.13,
"learning_rate": 0.00044075225677031094,
"loss": 2.9166,
"theoretical_loss": 3.9900412516735617,
"tokens_seen": 420151296
},
{
"epoch": 0.13,
"learning_rate": 0.0004407422266800401,
"loss": 3.0461,
"theoretical_loss": 3.989972202257966,
"tokens_seen": 420216832
},
{
"epoch": 0.13,
"learning_rate": 0.0004407321965897693,
"loss": 3.2388,
"theoretical_loss": 3.989903166625032,
"tokens_seen": 420282368
},
{
"epoch": 0.13,
"learning_rate": 0.0004407221664994985,
"loss": 2.6707,
"theoretical_loss": 3.989834144769861,
"tokens_seen": 420347904
},
{
"epoch": 0.13,
"learning_rate": 0.00044071213640922767,
"loss": 2.9935,
"theoretical_loss": 3.9897651366875557,
"tokens_seen": 420413440
},
{
"epoch": 0.13,
"learning_rate": 0.00044070210631895685,
"loss": 3.0185,
"theoretical_loss": 3.989696142373221,
"tokens_seen": 420478976
},
{
"epoch": 0.13,
"learning_rate": 0.0004406920762286861,
"loss": 2.7856,
"theoretical_loss": 3.9896271618219656,
"tokens_seen": 420544512
},
{
"epoch": 0.13,
"learning_rate": 0.00044068204613841527,
"loss": 2.7666,
"theoretical_loss": 3.9895581950288994,
"tokens_seen": 420610048
},
{
"epoch": 0.13,
"learning_rate": 0.00044067201604814445,
"loss": 3.0835,
"theoretical_loss": 3.9894892419891352,
"tokens_seen": 420675584
},
{
"epoch": 0.13,
"learning_rate": 0.0004406619859578737,
"loss": 2.7458,
"theoretical_loss": 3.989420302697789,
"tokens_seen": 420741120
},
{
"epoch": 0.13,
"learning_rate": 0.0004406519558676028,
"loss": 2.9788,
"theoretical_loss": 3.989351377149978,
"tokens_seen": 420806656
},
{
"epoch": 0.13,
"learning_rate": 0.00044064192577733204,
"loss": 3.0151,
"theoretical_loss": 3.989282465340823,
"tokens_seen": 420872192
},
{
"epoch": 0.13,
"learning_rate": 0.00044063189568706117,
"loss": 2.7167,
"theoretical_loss": 3.9892135672654465,
"tokens_seen": 420937728
},
{
"epoch": 0.13,
"learning_rate": 0.0004406218655967904,
"loss": 2.7923,
"theoretical_loss": 3.9891446829189734,
"tokens_seen": 421003264
},
{
"epoch": 0.13,
"objective/train/docs_used": 250392,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.840327262878418,
"objective/train/theoretical_loss": 3.989075812296533,
"objective/train/tokens_used": 441528800,
"theoretical_loss": 3.989075812296533,
"tokens_seen": 421068800
},
{
"epoch": 0.13,
"learning_rate": 0.0004406118355065196,
"loss": 2.9502,
"theoretical_loss": 3.989075812296533,
"tokens_seen": 421068800
},
{
"epoch": 0.13,
"learning_rate": 0.00044060180541624877,
"loss": 2.9474,
"theoretical_loss": 3.989006955393254,
"tokens_seen": 421134336
},
{
"epoch": 0.13,
"learning_rate": 0.00044059177532597795,
"loss": 2.8847,
"theoretical_loss": 3.98893811220427,
"tokens_seen": 421199872
},
{
"epoch": 0.13,
"learning_rate": 0.00044058174523570713,
"loss": 3.1015,
"theoretical_loss": 3.9888692827247167,
"tokens_seen": 421265408
},
{
"epoch": 0.13,
"learning_rate": 0.0004405717151454363,
"loss": 2.8387,
"theoretical_loss": 3.9888004669497317,
"tokens_seen": 421330944
},
{
"epoch": 0.13,
"learning_rate": 0.00044056168505516555,
"loss": 2.9615,
"theoretical_loss": 3.988731664874454,
"tokens_seen": 421396480
},
{
"epoch": 0.13,
"learning_rate": 0.0004405516549648947,
"loss": 2.9015,
"theoretical_loss": 3.9886628764940277,
"tokens_seen": 421462016
},
{
"epoch": 0.13,
"learning_rate": 0.0004405416248746239,
"loss": 2.8663,
"theoretical_loss": 3.988594101803598,
"tokens_seen": 421527552
},
{
"epoch": 0.13,
"learning_rate": 0.00044053159478435304,
"loss": 2.9738,
"theoretical_loss": 3.9885253407983114,
"tokens_seen": 421593088
},
{
"epoch": 0.13,
"learning_rate": 0.00044052156469408227,
"loss": 2.8473,
"theoretical_loss": 3.9884565934733196,
"tokens_seen": 421658624
},
{
"epoch": 0.13,
"learning_rate": 0.00044051153460381145,
"loss": 3.1179,
"theoretical_loss": 3.9883878598237743,
"tokens_seen": 421724160
},
{
"epoch": 0.13,
"learning_rate": 0.00044050150451354063,
"loss": 2.9468,
"theoretical_loss": 3.9883191398448306,
"tokens_seen": 421789696
},
{
"epoch": 0.13,
"learning_rate": 0.0004404914744232698,
"loss": 3.0772,
"theoretical_loss": 3.9882504335316464,
"tokens_seen": 421855232
},
{
"epoch": 0.13,
"learning_rate": 0.00044048144433299905,
"loss": 2.9365,
"theoretical_loss": 3.988181740879382,
"tokens_seen": 421920768
},
{
"epoch": 0.13,
"learning_rate": 0.0004404714142427282,
"loss": 2.6722,
"theoretical_loss": 3.9881130618831984,
"tokens_seen": 421986304
},
{
"epoch": 0.13,
"learning_rate": 0.0004404613841524574,
"loss": 3.0752,
"theoretical_loss": 3.988044396538263,
"tokens_seen": 422051840
},
{
"epoch": 0.13,
"learning_rate": 0.00044045135406218654,
"loss": 2.6501,
"theoretical_loss": 3.9879757448397406,
"tokens_seen": 422117376
},
{
"epoch": 0.13,
"learning_rate": 0.0004404413239719158,
"loss": 2.9492,
"theoretical_loss": 3.9879071067828034,
"tokens_seen": 422182912
},
{
"epoch": 0.13,
"learning_rate": 0.00044043129388164496,
"loss": 2.8998,
"theoretical_loss": 3.9878384823626227,
"tokens_seen": 422248448
},
{
"epoch": 0.13,
"learning_rate": 0.00044042126379137414,
"loss": 2.898,
"theoretical_loss": 3.987769871574373,
"tokens_seen": 422313984
},
{
"epoch": 0.13,
"learning_rate": 0.0004404112337011033,
"loss": 3.1466,
"theoretical_loss": 3.9877012744132316,
"tokens_seen": 422379520
},
{
"epoch": 0.13,
"learning_rate": 0.0004404012036108325,
"loss": 2.8585,
"theoretical_loss": 3.987632690874379,
"tokens_seen": 422445056
},
{
"epoch": 0.13,
"learning_rate": 0.0004403911735205617,
"loss": 3.0978,
"theoretical_loss": 3.987564120952997,
"tokens_seen": 422510592
},
{
"epoch": 0.13,
"learning_rate": 0.0004403811434302909,
"loss": 2.8917,
"theoretical_loss": 3.98749556464427,
"tokens_seen": 422576128
},
{
"epoch": 0.13,
"learning_rate": 0.00044037111334002004,
"loss": 2.8132,
"theoretical_loss": 3.987427021943385,
"tokens_seen": 422641664
},
{
"epoch": 0.13,
"objective/train/docs_used": 251199,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.547407627105713,
"objective/train/theoretical_loss": 3.987358492845532,
"objective/train/tokens_used": 443167200,
"theoretical_loss": 3.987358492845532,
"tokens_seen": 422707200
},
{
"epoch": 0.13,
"learning_rate": 0.0004403610832497493,
"loss": 2.9726,
"theoretical_loss": 3.987358492845532,
"tokens_seen": 422707200
},
{
"epoch": 0.13,
"learning_rate": 0.0004403510531594784,
"loss": 3.0115,
"theoretical_loss": 3.9872899773459025,
"tokens_seen": 422772736
},
{
"epoch": 0.13,
"learning_rate": 0.00044034102306920764,
"loss": 2.8462,
"theoretical_loss": 3.987221475439691,
"tokens_seen": 422838272
},
{
"epoch": 0.13,
"learning_rate": 0.0004403309929789368,
"loss": 2.8494,
"theoretical_loss": 3.9871529871220943,
"tokens_seen": 422903808
},
{
"epoch": 0.13,
"learning_rate": 0.000440320962888666,
"loss": 2.5373,
"theoretical_loss": 3.9870845123883116,
"tokens_seen": 422969344
},
{
"epoch": 0.13,
"learning_rate": 0.0004403109327983952,
"loss": 3.0218,
"theoretical_loss": 3.987016051233546,
"tokens_seen": 423034880
},
{
"epoch": 0.13,
"learning_rate": 0.0004403009027081244,
"loss": 2.9801,
"theoretical_loss": 3.9869476036529994,
"tokens_seen": 423100416
},
{
"epoch": 0.13,
"learning_rate": 0.00044029087261785355,
"loss": 2.9104,
"theoretical_loss": 3.98687916964188,
"tokens_seen": 423165952
},
{
"epoch": 0.13,
"learning_rate": 0.0004402808425275828,
"loss": 2.9944,
"theoretical_loss": 3.986810749195396,
"tokens_seen": 423231488
},
{
"epoch": 0.13,
"learning_rate": 0.0004402708124373119,
"loss": 2.9082,
"theoretical_loss": 3.9867423423087596,
"tokens_seen": 423297024
},
{
"epoch": 0.13,
"learning_rate": 0.00044026078234704114,
"loss": 2.9377,
"theoretical_loss": 3.9866739489771845,
"tokens_seen": 423362560
},
{
"epoch": 0.13,
"learning_rate": 0.0004402507522567703,
"loss": 3.0668,
"theoretical_loss": 3.986605569195887,
"tokens_seen": 423428096
},
{
"epoch": 0.13,
"learning_rate": 0.0004402407221664995,
"loss": 3.0737,
"theoretical_loss": 3.986537202960085,
"tokens_seen": 423493632
},
{
"epoch": 0.13,
"learning_rate": 0.0004402306920762287,
"loss": 3.0872,
"theoretical_loss": 3.986468850265001,
"tokens_seen": 423559168
},
{
"epoch": 0.13,
"learning_rate": 0.00044022066198595787,
"loss": 3.1557,
"theoretical_loss": 3.986400511105858,
"tokens_seen": 423624704
},
{
"epoch": 0.13,
"learning_rate": 0.00044021063189568705,
"loss": 3.0565,
"theoretical_loss": 3.9863321854778815,
"tokens_seen": 423690240
},
{
"epoch": 0.13,
"learning_rate": 0.0004402006018054163,
"loss": 3.1438,
"theoretical_loss": 3.986263873376301,
"tokens_seen": 423755776
},
{
"epoch": 0.13,
"learning_rate": 0.0004401905717151454,
"loss": 2.8585,
"theoretical_loss": 3.986195574796347,
"tokens_seen": 423821312
},
{
"epoch": 0.13,
"learning_rate": 0.00044018054162487465,
"loss": 2.5956,
"theoretical_loss": 3.9861272897332527,
"tokens_seen": 423886848
},
{
"epoch": 0.13,
"learning_rate": 0.00044017051153460383,
"loss": 2.902,
"theoretical_loss": 3.986059018182254,
"tokens_seen": 423952384
},
{
"epoch": 0.13,
"learning_rate": 0.000440160481444333,
"loss": 2.987,
"theoretical_loss": 3.9859907601385887,
"tokens_seen": 424017920
},
{
"epoch": 0.13,
"learning_rate": 0.0004401504513540622,
"loss": 2.6594,
"theoretical_loss": 3.9859225155974976,
"tokens_seen": 424083456
},
{
"epoch": 0.13,
"learning_rate": 0.00044014042126379137,
"loss": 2.9757,
"theoretical_loss": 3.985854284554223,
"tokens_seen": 424148992
},
{
"epoch": 0.13,
"learning_rate": 0.00044013039117352055,
"loss": 3.1121,
"theoretical_loss": 3.9857860670040113,
"tokens_seen": 424214528
},
{
"epoch": 0.13,
"learning_rate": 0.0004401203610832498,
"loss": 3.074,
"theoretical_loss": 3.98571786294211,
"tokens_seen": 424280064
},
{
"epoch": 0.13,
"objective/train/docs_used": 252247,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.094590187072754,
"objective/train/theoretical_loss": 3.9856496723637687,
"objective/train/tokens_used": 444805600,
"theoretical_loss": 3.9856496723637687,
"tokens_seen": 424345600
},
{
"epoch": 0.13,
"learning_rate": 0.0004401103309929789,
"loss": 3.0113,
"theoretical_loss": 3.9856496723637687,
"tokens_seen": 424345600
},
{
"epoch": 0.13,
"learning_rate": 0.00044010030090270815,
"loss": 3.0729,
"theoretical_loss": 3.985581495264241,
"tokens_seen": 424411136
},
{
"epoch": 0.13,
"learning_rate": 0.0004400902708124373,
"loss": 2.815,
"theoretical_loss": 3.9855133316387805,
"tokens_seen": 424476672
},
{
"epoch": 0.13,
"learning_rate": 0.0004400802407221665,
"loss": 3.0778,
"theoretical_loss": 3.9854451814826457,
"tokens_seen": 424542208
},
{
"epoch": 0.13,
"learning_rate": 0.0004400702106318957,
"loss": 2.8332,
"theoretical_loss": 3.985377044791096,
"tokens_seen": 424607744
},
{
"epoch": 0.13,
"learning_rate": 0.0004400601805416249,
"loss": 2.9363,
"theoretical_loss": 3.985308921559394,
"tokens_seen": 424673280
},
{
"epoch": 0.13,
"learning_rate": 0.00044005015045135406,
"loss": 2.8434,
"theoretical_loss": 3.9852408117828038,
"tokens_seen": 424738816
},
{
"epoch": 0.13,
"learning_rate": 0.00044004012036108324,
"loss": 2.9161,
"theoretical_loss": 3.9851727154565926,
"tokens_seen": 424804352
},
{
"epoch": 0.13,
"learning_rate": 0.0004400300902708124,
"loss": 2.7819,
"theoretical_loss": 3.98510463257603,
"tokens_seen": 424869888
},
{
"epoch": 0.13,
"learning_rate": 0.00044002006018054165,
"loss": 3.1194,
"theoretical_loss": 3.9850365631363873,
"tokens_seen": 424935424
},
{
"epoch": 0.13,
"learning_rate": 0.0004400100300902708,
"loss": 2.7506,
"theoretical_loss": 3.9849685071329386,
"tokens_seen": 425000960
},
{
"epoch": 0.13,
"learning_rate": 0.00044,
"loss": 3.132,
"theoretical_loss": 3.984900464560961,
"tokens_seen": 425066496
},
{
"epoch": 0.13,
"learning_rate": 0.0004399899699097292,
"loss": 2.8485,
"theoretical_loss": 3.984832435415733,
"tokens_seen": 425132032
},
{
"epoch": 0.13,
"learning_rate": 0.0004399799398194584,
"loss": 2.9311,
"theoretical_loss": 3.984764419692537,
"tokens_seen": 425197568
},
{
"epoch": 0.13,
"learning_rate": 0.00043996990972918756,
"loss": 2.9116,
"theoretical_loss": 3.984696417386655,
"tokens_seen": 425263104
},
{
"epoch": 0.13,
"learning_rate": 0.00043995987963891674,
"loss": 2.7839,
"theoretical_loss": 3.984628428493374,
"tokens_seen": 425328640
},
{
"epoch": 0.13,
"learning_rate": 0.0004399498495486459,
"loss": 2.6203,
"theoretical_loss": 3.9845604530079832,
"tokens_seen": 425394176
},
{
"epoch": 0.13,
"learning_rate": 0.00043993981945837516,
"loss": 2.6998,
"theoretical_loss": 3.984492490925772,
"tokens_seen": 425459712
},
{
"epoch": 0.13,
"learning_rate": 0.00043992978936810434,
"loss": 2.9726,
"theoretical_loss": 3.984424542242034,
"tokens_seen": 425525248
},
{
"epoch": 0.13,
"learning_rate": 0.0004399197592778335,
"loss": 2.9377,
"theoretical_loss": 3.9843566069520664,
"tokens_seen": 425590784
},
{
"epoch": 0.13,
"learning_rate": 0.0004399097291875627,
"loss": 2.6459,
"theoretical_loss": 3.9842886850511654,
"tokens_seen": 425656320
},
{
"epoch": 0.13,
"learning_rate": 0.0004398996990972919,
"loss": 2.7193,
"theoretical_loss": 3.984220776534632,
"tokens_seen": 425721856
},
{
"epoch": 0.13,
"learning_rate": 0.0004398896690070211,
"loss": 2.8158,
"theoretical_loss": 3.984152881397769,
"tokens_seen": 425787392
},
{
"epoch": 0.13,
"learning_rate": 0.00043987963891675024,
"loss": 2.9572,
"theoretical_loss": 3.9840849996358814,
"tokens_seen": 425852928
},
{
"epoch": 0.13,
"learning_rate": 0.0004398696088264795,
"loss": 2.6641,
"theoretical_loss": 3.9840171312442765,
"tokens_seen": 425918464
},
{
"epoch": 0.13,
"objective/train/docs_used": 253004,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.064711093902588,
"objective/train/theoretical_loss": 3.9839492762182647,
"objective/train/tokens_used": 446444000,
"theoretical_loss": 3.9839492762182647,
"tokens_seen": 425984000
},
{
"epoch": 0.13,
"learning_rate": 0.0004398595787362086,
"loss": 2.9181,
"theoretical_loss": 3.9839492762182647,
"tokens_seen": 425984000
},
{
"epoch": 0.13,
"learning_rate": 0.00043984954864593784,
"loss": 3.0364,
"theoretical_loss": 3.9838814345531577,
"tokens_seen": 426049536
},
{
"epoch": 0.13,
"learning_rate": 0.000439839518555667,
"loss": 2.7298,
"theoretical_loss": 3.9838136062442704,
"tokens_seen": 426115072
},
{
"epoch": 0.13,
"learning_rate": 0.0004398294884653962,
"loss": 2.9424,
"theoretical_loss": 3.98374579128692,
"tokens_seen": 426180608
},
{
"epoch": 0.13,
"learning_rate": 0.0004398194583751254,
"loss": 2.9639,
"theoretical_loss": 3.9836779896764254,
"tokens_seen": 426246144
},
{
"epoch": 0.13,
"learning_rate": 0.0004398094282848546,
"loss": 2.8545,
"theoretical_loss": 3.9836102014081076,
"tokens_seen": 426311680
},
{
"epoch": 0.13,
"learning_rate": 0.00043979939819458375,
"loss": 2.826,
"theoretical_loss": 3.983542426477292,
"tokens_seen": 426377216
},
{
"epoch": 0.13,
"learning_rate": 0.000439789368104313,
"loss": 2.8631,
"theoretical_loss": 3.9834746648793042,
"tokens_seen": 426442752
},
{
"epoch": 0.13,
"learning_rate": 0.0004397793380140421,
"loss": 2.925,
"theoretical_loss": 3.983406916609473,
"tokens_seen": 426508288
},
{
"epoch": 0.13,
"learning_rate": 0.00043976930792377134,
"loss": 3.0121,
"theoretical_loss": 3.98333918166313,
"tokens_seen": 426573824
},
{
"epoch": 0.13,
"learning_rate": 0.0004397592778335005,
"loss": 2.9686,
"theoretical_loss": 3.983271460035608,
"tokens_seen": 426639360
},
{
"epoch": 0.13,
"learning_rate": 0.0004397492477432297,
"loss": 2.628,
"theoretical_loss": 3.983203751722243,
"tokens_seen": 426704896
},
{
"epoch": 0.13,
"learning_rate": 0.0004397392176529589,
"loss": 2.8485,
"theoretical_loss": 3.983136056718373,
"tokens_seen": 426770432
},
{
"epoch": 0.13,
"learning_rate": 0.00043972918756268807,
"loss": 2.8924,
"theoretical_loss": 3.9830683750193394,
"tokens_seen": 426835968
},
{
"epoch": 0.13,
"learning_rate": 0.00043971915747241725,
"loss": 2.8836,
"theoretical_loss": 3.9830007066204836,
"tokens_seen": 426901504
},
{
"epoch": 0.13,
"learning_rate": 0.0004397091273821465,
"loss": 2.893,
"theoretical_loss": 3.9829330515171524,
"tokens_seen": 426967040
},
{
"epoch": 0.13,
"learning_rate": 0.0004396990972918756,
"loss": 2.8905,
"theoretical_loss": 3.982865409704692,
"tokens_seen": 427032576
},
{
"epoch": 0.13,
"learning_rate": 0.00043968906720160485,
"loss": 2.9351,
"theoretical_loss": 3.9827977811784523,
"tokens_seen": 427098112
},
{
"epoch": 0.13,
"learning_rate": 0.00043967903711133403,
"loss": 2.8235,
"theoretical_loss": 3.9827301659337864,
"tokens_seen": 427163648
},
{
"epoch": 0.13,
"learning_rate": 0.0004396690070210632,
"loss": 3.0057,
"theoretical_loss": 3.9826625639660485,
"tokens_seen": 427229184
},
{
"epoch": 0.13,
"learning_rate": 0.0004396589769307924,
"loss": 2.958,
"theoretical_loss": 3.982594975270595,
"tokens_seen": 427294720
},
{
"epoch": 0.13,
"learning_rate": 0.00043964894684052157,
"loss": 2.8409,
"theoretical_loss": 3.982527399842786,
"tokens_seen": 427360256
},
{
"epoch": 0.13,
"learning_rate": 0.00043963891675025075,
"loss": 2.7254,
"theoretical_loss": 3.9824598376779825,
"tokens_seen": 427425792
},
{
"epoch": 0.13,
"learning_rate": 0.00043962888665998,
"loss": 2.9997,
"theoretical_loss": 3.982392288771549,
"tokens_seen": 427491328
},
{
"epoch": 0.13,
"learning_rate": 0.0004396188565697091,
"loss": 2.7354,
"theoretical_loss": 3.98232475311885,
"tokens_seen": 427556864
},
{
"epoch": 0.13,
"objective/train/docs_used": 254260,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1842780113220215,
"objective/train/theoretical_loss": 3.9822572307152564,
"objective/train/tokens_used": 448082400,
"theoretical_loss": 3.9822572307152564,
"tokens_seen": 427622400
},
{
"epoch": 0.13,
"learning_rate": 0.00043960882647943835,
"loss": 2.9841,
"theoretical_loss": 3.9822572307152564,
"tokens_seen": 427622400
},
{
"epoch": 0.13,
"learning_rate": 0.0004395987963891675,
"loss": 3.3788,
"theoretical_loss": 3.9821897215561375,
"tokens_seen": 427687936
},
{
"epoch": 0.13,
"learning_rate": 0.0004395887662988967,
"loss": 2.8013,
"theoretical_loss": 3.982122225636867,
"tokens_seen": 427753472
},
{
"epoch": 0.13,
"learning_rate": 0.0004395787362086259,
"loss": 3.0004,
"theoretical_loss": 3.98205474295282,
"tokens_seen": 427819008
},
{
"epoch": 0.13,
"learning_rate": 0.0004395687061183551,
"loss": 2.9367,
"theoretical_loss": 3.9819872734993753,
"tokens_seen": 427884544
},
{
"epoch": 0.13,
"learning_rate": 0.00043955867602808426,
"loss": 2.971,
"theoretical_loss": 3.9819198172719124,
"tokens_seen": 427950080
},
{
"epoch": 0.13,
"learning_rate": 0.00043954864593781344,
"loss": 2.8509,
"theoretical_loss": 3.981852374265814,
"tokens_seen": 428015616
},
{
"epoch": 0.13,
"learning_rate": 0.0004395386158475426,
"loss": 2.9302,
"theoretical_loss": 3.9817849444764644,
"tokens_seen": 428081152
},
{
"epoch": 0.13,
"learning_rate": 0.00043952858575727185,
"loss": 2.7305,
"theoretical_loss": 3.981717527899252,
"tokens_seen": 428146688
},
{
"epoch": 0.13,
"learning_rate": 0.000439518555667001,
"loss": 2.7329,
"theoretical_loss": 3.9816501245295646,
"tokens_seen": 428212224
},
{
"epoch": 0.13,
"learning_rate": 0.0004395085255767302,
"loss": 2.95,
"theoretical_loss": 3.9815827343627954,
"tokens_seen": 428277760
},
{
"epoch": 0.13,
"learning_rate": 0.0004394984954864594,
"loss": 2.7096,
"theoretical_loss": 3.981515357394338,
"tokens_seen": 428343296
},
{
"epoch": 0.13,
"learning_rate": 0.0004394884653961886,
"loss": 3.0307,
"theoretical_loss": 3.9814479936195886,
"tokens_seen": 428408832
},
{
"epoch": 0.13,
"learning_rate": 0.00043947843530591776,
"loss": 2.6595,
"theoretical_loss": 3.981380643033945,
"tokens_seen": 428474368
},
{
"epoch": 0.13,
"learning_rate": 0.00043946840521564694,
"loss": 2.6865,
"theoretical_loss": 3.9813133056328103,
"tokens_seen": 428539904
},
{
"epoch": 0.13,
"learning_rate": 0.0004394583751253761,
"loss": 3.0187,
"theoretical_loss": 3.981245981411586,
"tokens_seen": 428605440
},
{
"epoch": 0.13,
"learning_rate": 0.00043944834503510536,
"loss": 2.813,
"theoretical_loss": 3.981178670365679,
"tokens_seen": 428670976
},
{
"epoch": 0.13,
"learning_rate": 0.0004394383149448345,
"loss": 2.8802,
"theoretical_loss": 3.981111372490496,
"tokens_seen": 428736512
},
{
"epoch": 0.13,
"learning_rate": 0.0004394282848545637,
"loss": 2.9085,
"theoretical_loss": 3.9810440877814486,
"tokens_seen": 428802048
},
{
"epoch": 0.13,
"learning_rate": 0.00043941825476429285,
"loss": 2.5354,
"theoretical_loss": 3.9809768162339476,
"tokens_seen": 428867584
},
{
"epoch": 0.13,
"learning_rate": 0.0004394082246740221,
"loss": 2.8229,
"theoretical_loss": 3.9809095578434093,
"tokens_seen": 428933120
},
{
"epoch": 0.13,
"learning_rate": 0.00043939819458375126,
"loss": 3.0868,
"theoretical_loss": 3.9808423126052506,
"tokens_seen": 428998656
},
{
"epoch": 0.13,
"learning_rate": 0.00043938816449348044,
"loss": 2.8457,
"theoretical_loss": 3.9807750805148903,
"tokens_seen": 429064192
},
{
"epoch": 0.13,
"learning_rate": 0.0004393781344032096,
"loss": 2.9069,
"theoretical_loss": 3.9807078615677502,
"tokens_seen": 429129728
},
{
"epoch": 0.13,
"learning_rate": 0.0004393681043129388,
"loss": 3.0153,
"theoretical_loss": 3.980640655759255,
"tokens_seen": 429195264
},
{
"epoch": 0.13,
"objective/train/docs_used": 255066,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0367283821105957,
"objective/train/theoretical_loss": 3.9805734630848306,
"objective/train/tokens_used": 449720800,
"theoretical_loss": 3.9805734630848306,
"tokens_seen": 429260800
},
{
"epoch": 0.13,
"learning_rate": 0.000439358074222668,
"loss": 2.8789,
"theoretical_loss": 3.9805734630848306,
"tokens_seen": 429260800
},
{
"epoch": 0.13,
"learning_rate": 0.0004393480441323972,
"loss": 3.0386,
"theoretical_loss": 3.9805062835399054,
"tokens_seen": 429326336
},
{
"epoch": 0.13,
"learning_rate": 0.00043933801404212635,
"loss": 2.7223,
"theoretical_loss": 3.9804391171199103,
"tokens_seen": 429391872
},
{
"epoch": 0.13,
"learning_rate": 0.0004393279839518556,
"loss": 2.8458,
"theoretical_loss": 3.9803719638202786,
"tokens_seen": 429457408
},
{
"epoch": 0.13,
"learning_rate": 0.00043931795386158477,
"loss": 2.8339,
"theoretical_loss": 3.9803048236364456,
"tokens_seen": 429522944
},
{
"epoch": 0.13,
"learning_rate": 0.00043930792377131395,
"loss": 3.011,
"theoretical_loss": 3.9802376965638495,
"tokens_seen": 429588480
},
{
"epoch": 0.13,
"learning_rate": 0.00043929789368104313,
"loss": 3.0028,
"theoretical_loss": 3.98017058259793,
"tokens_seen": 429654016
},
{
"epoch": 0.13,
"learning_rate": 0.0004392878635907723,
"loss": 2.925,
"theoretical_loss": 3.980103481734129,
"tokens_seen": 429719552
},
{
"epoch": 0.13,
"learning_rate": 0.0004392778335005015,
"loss": 2.9888,
"theoretical_loss": 3.9800363939678918,
"tokens_seen": 429785088
},
{
"epoch": 0.13,
"learning_rate": 0.0004392678034102307,
"loss": 3.0124,
"theoretical_loss": 3.9799693192946646,
"tokens_seen": 429850624
},
{
"epoch": 0.13,
"learning_rate": 0.00043925777331995985,
"loss": 2.8369,
"theoretical_loss": 3.979902257709897,
"tokens_seen": 429916160
},
{
"epoch": 0.13,
"learning_rate": 0.0004392477432296891,
"loss": 3.0397,
"theoretical_loss": 3.97983520920904,
"tokens_seen": 429981696
},
{
"epoch": 0.13,
"learning_rate": 0.0004392377131394182,
"loss": 2.8115,
"theoretical_loss": 3.979768173787548,
"tokens_seen": 430047232
},
{
"epoch": 0.13,
"learning_rate": 0.00043922768304914745,
"loss": 3.0784,
"theoretical_loss": 3.9797011514408758,
"tokens_seen": 430112768
},
{
"epoch": 0.13,
"learning_rate": 0.00043921765295887663,
"loss": 2.6783,
"theoretical_loss": 3.9796341421644827,
"tokens_seen": 430178304
},
{
"epoch": 0.13,
"learning_rate": 0.0004392076228686058,
"loss": 2.7325,
"theoretical_loss": 3.979567145953829,
"tokens_seen": 430243840
},
{
"epoch": 0.13,
"learning_rate": 0.000439197592778335,
"loss": 2.8376,
"theoretical_loss": 3.979500162804377,
"tokens_seen": 430309376
},
{
"epoch": 0.13,
"learning_rate": 0.00043918756268806423,
"loss": 2.8373,
"theoretical_loss": 3.979433192711592,
"tokens_seen": 430374912
},
{
"epoch": 0.13,
"learning_rate": 0.0004391775325977934,
"loss": 3.0308,
"theoretical_loss": 3.9793662356709407,
"tokens_seen": 430440448
},
{
"epoch": 0.13,
"learning_rate": 0.0004391675025075226,
"loss": 2.8456,
"theoretical_loss": 3.979299291677894,
"tokens_seen": 430505984
},
{
"epoch": 0.13,
"learning_rate": 0.00043915747241725177,
"loss": 2.863,
"theoretical_loss": 3.9792323607279223,
"tokens_seen": 430571520
},
{
"epoch": 0.13,
"learning_rate": 0.00043914744232698095,
"loss": 2.9346,
"theoretical_loss": 3.979165442816501,
"tokens_seen": 430637056
},
{
"epoch": 0.13,
"learning_rate": 0.0004391374122367102,
"loss": 3.1467,
"theoretical_loss": 3.9790985379391053,
"tokens_seen": 430702592
},
{
"epoch": 0.13,
"learning_rate": 0.0004391273821464393,
"loss": 2.9044,
"theoretical_loss": 3.9790316460912143,
"tokens_seen": 430768128
},
{
"epoch": 0.13,
"learning_rate": 0.00043911735205616855,
"loss": 2.9496,
"theoretical_loss": 3.978964767268309,
"tokens_seen": 430833664
},
{
"epoch": 0.13,
"objective/train/docs_used": 256338,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.68479585647583,
"objective/train/theoretical_loss": 3.9788979014658716,
"objective/train/tokens_used": 451359200,
"theoretical_loss": 3.9788979014658716,
"tokens_seen": 430899200
},
{
"epoch": 0.13,
"learning_rate": 0.0004391073219658977,
"loss": 2.7706,
"theoretical_loss": 3.9788979014658716,
"tokens_seen": 430899200
},
{
"epoch": 0.13,
"learning_rate": 0.0004390972918756269,
"loss": 2.7733,
"theoretical_loss": 3.978831048679389,
"tokens_seen": 430964736
},
{
"epoch": 0.13,
"learning_rate": 0.0004390872617853561,
"loss": 3.0516,
"theoretical_loss": 3.978764208904347,
"tokens_seen": 431030272
},
{
"epoch": 0.13,
"learning_rate": 0.0004390772316950853,
"loss": 2.9285,
"theoretical_loss": 3.9786973821362377,
"tokens_seen": 431095808
},
{
"epoch": 0.13,
"learning_rate": 0.00043906720160481446,
"loss": 2.9211,
"theoretical_loss": 3.9786305683705505,
"tokens_seen": 431161344
},
{
"epoch": 0.13,
"learning_rate": 0.00043905717151454364,
"loss": 2.9285,
"theoretical_loss": 3.978563767602782,
"tokens_seen": 431226880
},
{
"epoch": 0.13,
"learning_rate": 0.0004390471414242728,
"loss": 2.6945,
"theoretical_loss": 3.978496979828428,
"tokens_seen": 431292416
},
{
"epoch": 0.13,
"learning_rate": 0.00043903711133400205,
"loss": 2.889,
"theoretical_loss": 3.9784302050429874,
"tokens_seen": 431357952
},
{
"epoch": 0.13,
"learning_rate": 0.0004390270812437312,
"loss": 3.1345,
"theoretical_loss": 3.978363443241961,
"tokens_seen": 431423488
},
{
"epoch": 0.13,
"learning_rate": 0.0004390170511534604,
"loss": 2.9283,
"theoretical_loss": 3.9782966944208518,
"tokens_seen": 431489024
},
{
"epoch": 0.13,
"learning_rate": 0.0004390070210631896,
"loss": 3.0917,
"theoretical_loss": 3.9782299585751666,
"tokens_seen": 431554560
},
{
"epoch": 0.13,
"learning_rate": 0.0004389969909729188,
"loss": 2.7283,
"theoretical_loss": 3.9781632357004124,
"tokens_seen": 431620096
},
{
"epoch": 0.13,
"learning_rate": 0.00043898696088264796,
"loss": 2.92,
"theoretical_loss": 3.9780965257920995,
"tokens_seen": 431685632
},
{
"epoch": 0.13,
"learning_rate": 0.00043897693079237714,
"loss": 3.0776,
"theoretical_loss": 3.9780298288457394,
"tokens_seen": 431751168
},
{
"epoch": 0.13,
"learning_rate": 0.0004389669007021063,
"loss": 2.8729,
"theoretical_loss": 3.9779631448568473,
"tokens_seen": 431816704
},
{
"epoch": 0.13,
"learning_rate": 0.00043895687061183556,
"loss": 2.9113,
"theoretical_loss": 3.977896473820941,
"tokens_seen": 431882240
},
{
"epoch": 0.13,
"learning_rate": 0.0004389468405215647,
"loss": 2.8545,
"theoretical_loss": 3.9778298157335374,
"tokens_seen": 431947776
},
{
"epoch": 0.13,
"learning_rate": 0.0004389368104312939,
"loss": 3.0507,
"theoretical_loss": 3.977763170590159,
"tokens_seen": 432013312
},
{
"epoch": 0.13,
"learning_rate": 0.00043892678034102305,
"loss": 2.8307,
"theoretical_loss": 3.9776965383863288,
"tokens_seen": 432078848
},
{
"epoch": 0.13,
"learning_rate": 0.0004389167502507523,
"loss": 2.8236,
"theoretical_loss": 3.9776299191175726,
"tokens_seen": 432144384
},
{
"epoch": 0.13,
"learning_rate": 0.00043890672016048146,
"loss": 2.9224,
"theoretical_loss": 3.977563312779418,
"tokens_seen": 432209920
},
{
"epoch": 0.13,
"learning_rate": 0.00043889669007021064,
"loss": 2.8145,
"theoretical_loss": 3.977496719367396,
"tokens_seen": 432275456
},
{
"epoch": 0.13,
"learning_rate": 0.0004388866599799398,
"loss": 2.8554,
"theoretical_loss": 3.9774301388770374,
"tokens_seen": 432340992
},
{
"epoch": 0.13,
"learning_rate": 0.000438876629889669,
"loss": 3.0139,
"theoretical_loss": 3.9773635713038784,
"tokens_seen": 432406528
},
{
"epoch": 0.13,
"learning_rate": 0.0004388665997993982,
"loss": 2.9057,
"theoretical_loss": 3.9772970166434547,
"tokens_seen": 432472064
},
{
"epoch": 0.13,
"objective/train/docs_used": 257001,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.186760187149048,
"objective/train/theoretical_loss": 3.9772304748913054,
"objective/train/tokens_used": 452997600,
"theoretical_loss": 3.9772304748913054,
"tokens_seen": 432537600
},
{
"epoch": 0.13,
"learning_rate": 0.0004388565697091274,
"loss": 2.946,
"theoretical_loss": 3.9772304748913054,
"tokens_seen": 432537600
},
{
"epoch": 0.13,
"learning_rate": 0.00043884653961885655,
"loss": 2.8407,
"theoretical_loss": 3.977163946042973,
"tokens_seen": 432603136
},
{
"epoch": 0.13,
"learning_rate": 0.0004388365095285858,
"loss": 2.9356,
"theoretical_loss": 3.977097430093999,
"tokens_seen": 432668672
},
{
"epoch": 0.13,
"learning_rate": 0.00043882647943831497,
"loss": 2.9862,
"theoretical_loss": 3.97703092703993,
"tokens_seen": 432734208
},
{
"epoch": 0.13,
"learning_rate": 0.00043881644934804415,
"loss": 2.9973,
"theoretical_loss": 3.9769644368763135,
"tokens_seen": 432799744
},
{
"epoch": 0.13,
"learning_rate": 0.00043880641925777333,
"loss": 3.1139,
"theoretical_loss": 3.9768979595987,
"tokens_seen": 432865280
},
{
"epoch": 0.13,
"learning_rate": 0.0004387963891675025,
"loss": 3.0764,
"theoretical_loss": 3.9768314952026422,
"tokens_seen": 432930816
},
{
"epoch": 0.13,
"learning_rate": 0.0004387863590772317,
"loss": 2.8346,
"theoretical_loss": 3.9767650436836934,
"tokens_seen": 432996352
},
{
"epoch": 0.13,
"learning_rate": 0.0004387763289869609,
"loss": 2.9051,
"theoretical_loss": 3.976698605037411,
"tokens_seen": 433061888
},
{
"epoch": 0.13,
"learning_rate": 0.00043876629889669005,
"loss": 2.9545,
"theoretical_loss": 3.976632179259354,
"tokens_seen": 433127424
},
{
"epoch": 0.13,
"learning_rate": 0.0004387562688064193,
"loss": 3.116,
"theoretical_loss": 3.976565766345083,
"tokens_seen": 433192960
},
{
"epoch": 0.13,
"learning_rate": 0.0004387462387161484,
"loss": 2.8026,
"theoretical_loss": 3.9764993662901613,
"tokens_seen": 433258496
},
{
"epoch": 0.13,
"learning_rate": 0.00043873620862587765,
"loss": 2.9205,
"theoretical_loss": 3.976432979090155,
"tokens_seen": 433324032
},
{
"epoch": 0.13,
"learning_rate": 0.00043872617853560683,
"loss": 2.9889,
"theoretical_loss": 3.9763666047406314,
"tokens_seen": 433389568
},
{
"epoch": 0.13,
"learning_rate": 0.000438716148445336,
"loss": 2.8799,
"theoretical_loss": 3.9763002432371604,
"tokens_seen": 433455104
},
{
"epoch": 0.13,
"learning_rate": 0.0004387061183550652,
"loss": 3.0946,
"theoretical_loss": 3.9762338945753144,
"tokens_seen": 433520640
},
{
"epoch": 0.13,
"learning_rate": 0.00043869608826479443,
"loss": 2.9393,
"theoretical_loss": 3.9761675587506673,
"tokens_seen": 433586176
},
{
"epoch": 0.13,
"learning_rate": 0.00043868605817452356,
"loss": 3.1171,
"theoretical_loss": 3.9761012357587955,
"tokens_seen": 433651712
},
{
"epoch": 0.13,
"learning_rate": 0.0004386760280842528,
"loss": 2.8962,
"theoretical_loss": 3.976034925595278,
"tokens_seen": 433717248
},
{
"epoch": 0.13,
"learning_rate": 0.0004386659979939819,
"loss": 2.908,
"theoretical_loss": 3.975968628255696,
"tokens_seen": 433782784
},
{
"epoch": 0.13,
"learning_rate": 0.00043865596790371115,
"loss": 3.0901,
"theoretical_loss": 3.9759023437356316,
"tokens_seen": 433848320
},
{
"epoch": 0.13,
"learning_rate": 0.00043864593781344033,
"loss": 2.8724,
"theoretical_loss": 3.9758360720306705,
"tokens_seen": 433913856
},
{
"epoch": 0.13,
"learning_rate": 0.0004386359077231695,
"loss": 3.1435,
"theoretical_loss": 3.9757698131364005,
"tokens_seen": 433979392
},
{
"epoch": 0.13,
"learning_rate": 0.0004386258776328987,
"loss": 2.8047,
"theoretical_loss": 3.9757035670484107,
"tokens_seen": 434044928
},
{
"epoch": 0.13,
"learning_rate": 0.0004386158475426279,
"loss": 2.782,
"theoretical_loss": 3.9756373337622932,
"tokens_seen": 434110464
},
{
"epoch": 0.13,
"objective/train/docs_used": 257982,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2712156772613525,
"objective/train/theoretical_loss": 3.975571113273642,
"objective/train/tokens_used": 454636000,
"theoretical_loss": 3.975571113273642,
"tokens_seen": 434176000
},
{
"epoch": 0.13,
"learning_rate": 0.00043860581745235706,
"loss": 2.8521,
"theoretical_loss": 3.975571113273642,
"tokens_seen": 434176000
},
{
"epoch": 0.13,
"learning_rate": 0.0004385957873620863,
"loss": 2.785,
"theoretical_loss": 3.975504905578053,
"tokens_seen": 434241536
},
{
"epoch": 0.13,
"learning_rate": 0.0004385857572718154,
"loss": 2.8655,
"theoretical_loss": 3.975438710671125,
"tokens_seen": 434307072
},
{
"epoch": 0.13,
"learning_rate": 0.00043857572718154466,
"loss": 2.9464,
"theoretical_loss": 3.9753725285484576,
"tokens_seen": 434372608
},
{
"epoch": 0.13,
"learning_rate": 0.0004385656970912738,
"loss": 3.0666,
"theoretical_loss": 3.9753063592056552,
"tokens_seen": 434438144
},
{
"epoch": 0.13,
"learning_rate": 0.000438555667001003,
"loss": 2.9942,
"theoretical_loss": 3.9752402026383207,
"tokens_seen": 434503680
},
{
"epoch": 0.13,
"learning_rate": 0.0004385456369107322,
"loss": 3.035,
"theoretical_loss": 3.9751740588420623,
"tokens_seen": 434569216
},
{
"epoch": 0.13,
"learning_rate": 0.0004385356068204614,
"loss": 2.9534,
"theoretical_loss": 3.9751079278124886,
"tokens_seen": 434634752
},
{
"epoch": 0.13,
"learning_rate": 0.00043852557673019056,
"loss": 3.0763,
"theoretical_loss": 3.9750418095452122,
"tokens_seen": 434700288
},
{
"epoch": 0.13,
"learning_rate": 0.0004385155466399198,
"loss": 2.9295,
"theoretical_loss": 3.9749757040358453,
"tokens_seen": 434765824
},
{
"epoch": 0.13,
"learning_rate": 0.0004385055165496489,
"loss": 2.9565,
"theoretical_loss": 3.9749096112800046,
"tokens_seen": 434831360
},
{
"epoch": 0.13,
"learning_rate": 0.00043849548645937816,
"loss": 2.7425,
"theoretical_loss": 3.974843531273307,
"tokens_seen": 434896896
},
{
"epoch": 0.13,
"learning_rate": 0.0004384854563691073,
"loss": 2.8579,
"theoretical_loss": 3.9747774640113738,
"tokens_seen": 434962432
},
{
"epoch": 0.13,
"learning_rate": 0.0004384754262788365,
"loss": 2.7193,
"theoretical_loss": 3.974711409489826,
"tokens_seen": 435027968
},
{
"epoch": 0.13,
"learning_rate": 0.0004384653961885657,
"loss": 2.8602,
"theoretical_loss": 3.974645367704289,
"tokens_seen": 435093504
},
{
"epoch": 0.13,
"learning_rate": 0.0004384553660982949,
"loss": 2.9948,
"theoretical_loss": 3.9745793386503885,
"tokens_seen": 435159040
},
{
"epoch": 0.13,
"learning_rate": 0.00043844533600802407,
"loss": 2.9632,
"theoretical_loss": 3.974513322323754,
"tokens_seen": 435224576
},
{
"epoch": 0.13,
"learning_rate": 0.00043843530591775325,
"loss": 2.7926,
"theoretical_loss": 3.9744473187200158,
"tokens_seen": 435290112
},
{
"epoch": 0.13,
"learning_rate": 0.0004384252758274825,
"loss": 3.1408,
"theoretical_loss": 3.974381327834807,
"tokens_seen": 435355648
},
{
"epoch": 0.13,
"learning_rate": 0.00043841524573721166,
"loss": 3.0636,
"theoretical_loss": 3.974315349663763,
"tokens_seen": 435421184
},
{
"epoch": 0.13,
"learning_rate": 0.00043840521564694084,
"loss": 2.8573,
"theoretical_loss": 3.9742493842025213,
"tokens_seen": 435486720
},
{
"epoch": 0.13,
"learning_rate": 0.00043839518555667,
"loss": 2.7695,
"theoretical_loss": 3.974183431446721,
"tokens_seen": 435552256
},
{
"epoch": 0.13,
"learning_rate": 0.0004383851554663992,
"loss": 2.8558,
"theoretical_loss": 3.974117491392004,
"tokens_seen": 435617792
},
{
"epoch": 0.13,
"learning_rate": 0.0004383751253761284,
"loss": 2.7744,
"theoretical_loss": 3.9740515640340135,
"tokens_seen": 435683328
},
{
"epoch": 0.13,
"learning_rate": 0.0004383650952858576,
"loss": 3.0929,
"theoretical_loss": 3.9739856493683967,
"tokens_seen": 435748864
},
{
"epoch": 0.13,
"objective/train/docs_used": 258654,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9247915744781494,
"objective/train/theoretical_loss": 3.973919747390801,
"objective/train/tokens_used": 456274400,
"theoretical_loss": 3.973919747390801,
"tokens_seen": 435814400
},
{
"epoch": 0.13,
"learning_rate": 0.00043835506519558675,
"loss": 3.015,
"theoretical_loss": 3.973919747390801,
"tokens_seen": 435814400
},
{
"epoch": 0.13,
"learning_rate": 0.000438345035105316,
"loss": 3.172,
"theoretical_loss": 3.973853858096876,
"tokens_seen": 435879936
},
{
"epoch": 0.13,
"learning_rate": 0.00043833500501504517,
"loss": 2.922,
"theoretical_loss": 3.9737879814822747,
"tokens_seen": 435945472
},
{
"epoch": 0.13,
"learning_rate": 0.00043832497492477435,
"loss": 2.889,
"theoretical_loss": 3.973722117542652,
"tokens_seen": 436011008
},
{
"epoch": 0.13,
"learning_rate": 0.00043831494483450353,
"loss": 2.9686,
"theoretical_loss": 3.9736562662736636,
"tokens_seen": 436076544
},
{
"epoch": 0.13,
"learning_rate": 0.0004383049147442327,
"loss": 2.9584,
"theoretical_loss": 3.9735904276709695,
"tokens_seen": 436142080
},
{
"epoch": 0.13,
"learning_rate": 0.0004382948846539619,
"loss": 2.9236,
"theoretical_loss": 3.9735246017302295,
"tokens_seen": 436207616
},
{
"epoch": 0.13,
"learning_rate": 0.0004382848545636911,
"loss": 3.0206,
"theoretical_loss": 3.9734587884471075,
"tokens_seen": 436273152
},
{
"epoch": 0.13,
"learning_rate": 0.00043827482447342025,
"loss": 2.9689,
"theoretical_loss": 3.9733929878172685,
"tokens_seen": 436338688
},
{
"epoch": 0.13,
"learning_rate": 0.0004382647943831495,
"loss": 2.8583,
"theoretical_loss": 3.9733271998363797,
"tokens_seen": 436404224
},
{
"epoch": 0.13,
"learning_rate": 0.0004382547642928786,
"loss": 3.0176,
"theoretical_loss": 3.9732614245001106,
"tokens_seen": 436469760
},
{
"epoch": 0.13,
"learning_rate": 0.00043824473420260785,
"loss": 2.8963,
"theoretical_loss": 3.9731956618041333,
"tokens_seen": 436535296
},
{
"epoch": 0.13,
"learning_rate": 0.00043823470411233703,
"loss": 2.9904,
"theoretical_loss": 3.9731299117441203,
"tokens_seen": 436600832
},
{
"epoch": 0.13,
"learning_rate": 0.0004382246740220662,
"loss": 3.0376,
"theoretical_loss": 3.9730641743157493,
"tokens_seen": 436666368
},
{
"epoch": 0.13,
"learning_rate": 0.0004382146439317954,
"loss": 2.8604,
"theoretical_loss": 3.972998449514697,
"tokens_seen": 436731904
},
{
"epoch": 0.13,
"learning_rate": 0.00043820461384152463,
"loss": 2.7958,
"theoretical_loss": 3.9729327373366443,
"tokens_seen": 436797440
},
{
"epoch": 0.13,
"learning_rate": 0.00043819458375125376,
"loss": 2.827,
"theoretical_loss": 3.972867037777273,
"tokens_seen": 436862976
},
{
"epoch": 0.13,
"learning_rate": 0.000438184553660983,
"loss": 2.9757,
"theoretical_loss": 3.9728013508322673,
"tokens_seen": 436928512
},
{
"epoch": 0.13,
"learning_rate": 0.0004381745235707121,
"loss": 2.8648,
"theoretical_loss": 3.9727356764973143,
"tokens_seen": 436994048
},
{
"epoch": 0.13,
"learning_rate": 0.00043816449348044135,
"loss": 2.9147,
"theoretical_loss": 3.972670014768102,
"tokens_seen": 437059584
},
{
"epoch": 0.13,
"learning_rate": 0.00043815446339017053,
"loss": 2.9013,
"theoretical_loss": 3.972604365640322,
"tokens_seen": 437125120
},
{
"epoch": 0.13,
"learning_rate": 0.0004381444332998997,
"loss": 2.9928,
"theoretical_loss": 3.972538729109666,
"tokens_seen": 437190656
},
{
"epoch": 0.13,
"learning_rate": 0.0004381344032096289,
"loss": 2.9384,
"theoretical_loss": 3.97247310517183,
"tokens_seen": 437256192
},
{
"epoch": 0.13,
"learning_rate": 0.0004381243731193581,
"loss": 3.2878,
"theoretical_loss": 3.972407493822511,
"tokens_seen": 437321728
},
{
"epoch": 0.13,
"learning_rate": 0.00043811434302908726,
"loss": 2.6546,
"theoretical_loss": 3.972341895057408,
"tokens_seen": 437387264
},
{
"epoch": 0.13,
"objective/train/docs_used": 259916,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.601888418197632,
"objective/train/theoretical_loss": 3.972276308872222,
"objective/train/tokens_used": 457912800,
"theoretical_loss": 3.972276308872222,
"tokens_seen": 437452800
},
{
"epoch": 0.13,
"learning_rate": 0.0004381043129388165,
"loss": 2.7597,
"theoretical_loss": 3.972276308872222,
"tokens_seen": 437452800
},
{
"epoch": 0.13,
"learning_rate": 0.0004380942828485456,
"loss": 2.8085,
"theoretical_loss": 3.972210735262657,
"tokens_seen": 437518336
},
{
"epoch": 0.13,
"learning_rate": 0.00043808425275827486,
"loss": 3.1744,
"theoretical_loss": 3.9721451742244183,
"tokens_seen": 437583872
},
{
"epoch": 0.13,
"learning_rate": 0.000438074222668004,
"loss": 3.0446,
"theoretical_loss": 3.9720796257532136,
"tokens_seen": 437649408
},
{
"epoch": 0.13,
"learning_rate": 0.0004380641925777332,
"loss": 2.8964,
"theoretical_loss": 3.9720140898447527,
"tokens_seen": 437714944
},
{
"epoch": 0.13,
"learning_rate": 0.0004380541624874624,
"loss": 3.0913,
"theoretical_loss": 3.9719485664947483,
"tokens_seen": 437780480
},
{
"epoch": 0.13,
"learning_rate": 0.0004380441323971916,
"loss": 2.9713,
"theoretical_loss": 3.971883055698913,
"tokens_seen": 437846016
},
{
"epoch": 0.13,
"learning_rate": 0.00043803410230692076,
"loss": 3.012,
"theoretical_loss": 3.9718175574529635,
"tokens_seen": 437911552
},
{
"epoch": 0.13,
"learning_rate": 0.00043802407221665,
"loss": 2.8964,
"theoretical_loss": 3.9717520717526176,
"tokens_seen": 437977088
},
{
"epoch": 0.13,
"learning_rate": 0.0004380140421263791,
"loss": 3.1043,
"theoretical_loss": 3.971686598593596,
"tokens_seen": 438042624
},
{
"epoch": 0.13,
"learning_rate": 0.00043800401203610836,
"loss": 2.9459,
"theoretical_loss": 3.971621137971622,
"tokens_seen": 438108160
},
{
"epoch": 0.13,
"learning_rate": 0.0004379939819458375,
"loss": 3.0117,
"theoretical_loss": 3.9715556898824182,
"tokens_seen": 438173696
},
{
"epoch": 0.13,
"learning_rate": 0.0004379839518555667,
"loss": 3.0628,
"theoretical_loss": 3.971490254321713,
"tokens_seen": 438239232
},
{
"epoch": 0.13,
"learning_rate": 0.0004379739217652959,
"loss": 3.005,
"theoretical_loss": 3.9714248312852343,
"tokens_seen": 438304768
},
{
"epoch": 0.13,
"learning_rate": 0.0004379638916750251,
"loss": 3.2229,
"theoretical_loss": 3.971359420768712,
"tokens_seen": 438370304
},
{
"epoch": 0.13,
"learning_rate": 0.00043795386158475427,
"loss": 3.0165,
"theoretical_loss": 3.9712940227678803,
"tokens_seen": 438435840
},
{
"epoch": 0.13,
"learning_rate": 0.00043794383149448345,
"loss": 2.9107,
"theoretical_loss": 3.9712286372784735,
"tokens_seen": 438501376
},
{
"epoch": 0.13,
"learning_rate": 0.00043793380140421263,
"loss": 3.1229,
"theoretical_loss": 3.971163264296229,
"tokens_seen": 438566912
},
{
"epoch": 0.13,
"learning_rate": 0.00043792377131394186,
"loss": 2.9839,
"theoretical_loss": 3.971097903816886,
"tokens_seen": 438632448
},
{
"epoch": 0.13,
"learning_rate": 0.000437913741223671,
"loss": 2.8638,
"theoretical_loss": 3.9710325558361856,
"tokens_seen": 438697984
},
{
"epoch": 0.13,
"learning_rate": 0.0004379037111334002,
"loss": 2.8033,
"theoretical_loss": 3.9709672203498707,
"tokens_seen": 438763520
},
{
"epoch": 0.13,
"learning_rate": 0.00043789368104312935,
"loss": 2.9359,
"theoretical_loss": 3.970901897353688,
"tokens_seen": 438829056
},
{
"epoch": 0.13,
"learning_rate": 0.0004378836509528586,
"loss": 2.8778,
"theoretical_loss": 3.970836586843383,
"tokens_seen": 438894592
},
{
"epoch": 0.13,
"learning_rate": 0.00043787362086258777,
"loss": 3.0624,
"theoretical_loss": 3.9707712888147073,
"tokens_seen": 438960128
},
{
"epoch": 0.13,
"learning_rate": 0.00043786359077231695,
"loss": 2.7369,
"theoretical_loss": 3.9707060032634107,
"tokens_seen": 439025664
},
{
"epoch": 0.13,
"objective/train/docs_used": 260524,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9223694801330566,
"objective/train/theoretical_loss": 3.9706407301852487,
"objective/train/tokens_used": 459551200,
"theoretical_loss": 3.9706407301852487,
"tokens_seen": 439091200
},
{
"epoch": 0.13,
"learning_rate": 0.00043785356068204613,
"loss": 2.7099,
"theoretical_loss": 3.9706407301852487,
"tokens_seen": 439091200
},
{
"epoch": 0.13,
"learning_rate": 0.00043784353059177537,
"loss": 2.9097,
"theoretical_loss": 3.9705754695759756,
"tokens_seen": 439156736
},
{
"epoch": 0.13,
"learning_rate": 0.0004378335005015045,
"loss": 2.7865,
"theoretical_loss": 3.9705102214313506,
"tokens_seen": 439222272
},
{
"epoch": 0.13,
"learning_rate": 0.00043782347041123373,
"loss": 2.9523,
"theoretical_loss": 3.9704449857471324,
"tokens_seen": 439287808
},
{
"epoch": 0.13,
"learning_rate": 0.00043781344032096286,
"loss": 3.0496,
"theoretical_loss": 3.970379762519084,
"tokens_seen": 439353344
},
{
"epoch": 0.13,
"learning_rate": 0.0004378034102306921,
"loss": 2.9437,
"theoretical_loss": 3.9703145517429697,
"tokens_seen": 439418880
},
{
"epoch": 0.13,
"learning_rate": 0.00043779338014042127,
"loss": 3.0335,
"theoretical_loss": 3.9702493534145544,
"tokens_seen": 439484416
},
{
"epoch": 0.13,
"learning_rate": 0.00043778335005015045,
"loss": 3.0759,
"theoretical_loss": 3.9701841675296077,
"tokens_seen": 439549952
},
{
"epoch": 0.13,
"learning_rate": 0.00043777331995987963,
"loss": 3.2439,
"theoretical_loss": 3.970118994083899,
"tokens_seen": 439615488
},
{
"epoch": 0.13,
"learning_rate": 0.0004377632898696088,
"loss": 3.0867,
"theoretical_loss": 3.9700538330732016,
"tokens_seen": 439681024
},
{
"epoch": 0.13,
"learning_rate": 0.000437753259779338,
"loss": 3.1688,
"theoretical_loss": 3.969988684493289,
"tokens_seen": 439746560
},
{
"epoch": 0.13,
"learning_rate": 0.00043774322968906723,
"loss": 3.0956,
"theoretical_loss": 3.9699235483399384,
"tokens_seen": 439812096
},
{
"epoch": 0.13,
"learning_rate": 0.00043773319959879636,
"loss": 2.8337,
"theoretical_loss": 3.969858424608928,
"tokens_seen": 439877632
},
{
"epoch": 0.13,
"learning_rate": 0.0004377231695085256,
"loss": 3.0612,
"theoretical_loss": 3.969793313296039,
"tokens_seen": 439943168
},
{
"epoch": 0.13,
"learning_rate": 0.0004377131394182547,
"loss": 3.1136,
"theoretical_loss": 3.969728214397053,
"tokens_seen": 440008704
},
{
"epoch": 0.13,
"learning_rate": 0.00043770310932798396,
"loss": 2.9982,
"theoretical_loss": 3.969663127907756,
"tokens_seen": 440074240
},
{
"epoch": 0.13,
"learning_rate": 0.00043769307923771314,
"loss": 3.004,
"theoretical_loss": 3.969598053823934,
"tokens_seen": 440139776
},
{
"epoch": 0.13,
"learning_rate": 0.0004376830491474423,
"loss": 2.9279,
"theoretical_loss": 3.969532992141377,
"tokens_seen": 440205312
},
{
"epoch": 0.13,
"learning_rate": 0.00043767301905717155,
"loss": 2.9182,
"theoretical_loss": 3.969467942855875,
"tokens_seen": 440270848
},
{
"epoch": 0.13,
"learning_rate": 0.00043766298896690073,
"loss": 3.1425,
"theoretical_loss": 3.969402905963221,
"tokens_seen": 440336384
},
{
"epoch": 0.13,
"learning_rate": 0.0004376529588766299,
"loss": 3.2916,
"theoretical_loss": 3.9693378814592104,
"tokens_seen": 440401920
},
{
"epoch": 0.13,
"learning_rate": 0.0004376429287863591,
"loss": 3.0597,
"theoretical_loss": 3.96927286933964,
"tokens_seen": 440467456
},
{
"epoch": 0.13,
"learning_rate": 0.0004376328986960883,
"loss": 2.8762,
"theoretical_loss": 3.969207869600309,
"tokens_seen": 440532992
},
{
"epoch": 0.13,
"learning_rate": 0.00043762286860581746,
"loss": 3.1591,
"theoretical_loss": 3.969142882237019,
"tokens_seen": 440598528
},
{
"epoch": 0.13,
"learning_rate": 0.0004376128385155467,
"loss": 3.0577,
"theoretical_loss": 3.9690779072455733,
"tokens_seen": 440664064
},
{
"epoch": 0.13,
"objective/train/docs_used": 261739,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.807461977005005,
"objective/train/theoretical_loss": 3.969012944621776,
"objective/train/tokens_used": 461189600,
"theoretical_loss": 3.969012944621776,
"tokens_seen": 440729600
},
{
"epoch": 0.13,
"learning_rate": 0.0004376028084252758,
"loss": 2.6719,
"theoretical_loss": 3.969012944621776,
"tokens_seen": 440729600
},
{
"epoch": 0.13,
"learning_rate": 0.00043759277833500506,
"loss": 2.9454,
"theoretical_loss": 3.968947994361436,
"tokens_seen": 440795136
},
{
"epoch": 0.13,
"learning_rate": 0.0004375827482447342,
"loss": 2.6968,
"theoretical_loss": 3.9688830564603625,
"tokens_seen": 440860672
},
{
"epoch": 0.13,
"learning_rate": 0.0004375727181544634,
"loss": 2.9212,
"theoretical_loss": 3.9688181309143657,
"tokens_seen": 440926208
},
{
"epoch": 0.13,
"learning_rate": 0.0004375626880641926,
"loss": 2.8568,
"theoretical_loss": 3.9687532177192604,
"tokens_seen": 440991744
},
{
"epoch": 0.13,
"learning_rate": 0.0004375526579739218,
"loss": 3.0616,
"theoretical_loss": 3.9686883168708613,
"tokens_seen": 441057280
},
{
"epoch": 0.13,
"learning_rate": 0.00043754262788365096,
"loss": 2.9143,
"theoretical_loss": 3.968623428364986,
"tokens_seen": 441122816
},
{
"epoch": 0.13,
"learning_rate": 0.0004375325977933802,
"loss": 2.9528,
"theoretical_loss": 3.9685585521974556,
"tokens_seen": 441188352
},
{
"epoch": 0.13,
"learning_rate": 0.0004375225677031093,
"loss": 2.9707,
"theoretical_loss": 3.968493688364089,
"tokens_seen": 441253888
},
{
"epoch": 0.13,
"learning_rate": 0.00043751253761283856,
"loss": 3.0977,
"theoretical_loss": 3.968428836860712,
"tokens_seen": 441319424
},
{
"epoch": 0.13,
"learning_rate": 0.0004375025075225677,
"loss": 2.7931,
"theoretical_loss": 3.96836399768315,
"tokens_seen": 441384960
},
{
"epoch": 0.13,
"learning_rate": 0.0004374924774322969,
"loss": 3.0323,
"theoretical_loss": 3.96829917082723,
"tokens_seen": 441450496
},
{
"epoch": 0.13,
"learning_rate": 0.0004374824473420261,
"loss": 3.1413,
"theoretical_loss": 3.9682343562887823,
"tokens_seen": 441516032
},
{
"epoch": 0.13,
"learning_rate": 0.0004374724172517553,
"loss": 3.2888,
"theoretical_loss": 3.968169554063639,
"tokens_seen": 441581568
},
{
"epoch": 0.13,
"learning_rate": 0.00043746238716148447,
"loss": 3.019,
"theoretical_loss": 3.9681047641476326,
"tokens_seen": 441647104
},
{
"epoch": 0.13,
"learning_rate": 0.00043745235707121365,
"loss": 2.8898,
"theoretical_loss": 3.968039986536601,
"tokens_seen": 441712640
},
{
"epoch": 0.13,
"learning_rate": 0.00043744232698094283,
"loss": 3.1168,
"theoretical_loss": 3.9679752212263795,
"tokens_seen": 441778176
},
{
"epoch": 0.13,
"learning_rate": 0.00043743229689067206,
"loss": 2.8963,
"theoretical_loss": 3.9679104682128106,
"tokens_seen": 441843712
},
{
"epoch": 0.13,
"learning_rate": 0.0004374222668004012,
"loss": 2.9959,
"theoretical_loss": 3.967845727491735,
"tokens_seen": 441909248
},
{
"epoch": 0.13,
"learning_rate": 0.0004374122367101304,
"loss": 2.9724,
"theoretical_loss": 3.9677809990589967,
"tokens_seen": 441974784
},
{
"epoch": 0.13,
"learning_rate": 0.00043740220661985955,
"loss": 2.7693,
"theoretical_loss": 3.967716282910441,
"tokens_seen": 442040320
},
{
"epoch": 0.13,
"learning_rate": 0.0004373921765295888,
"loss": 2.8861,
"theoretical_loss": 3.9676515790419176,
"tokens_seen": 442105856
},
{
"epoch": 0.13,
"learning_rate": 0.00043738214643931797,
"loss": 2.9332,
"theoretical_loss": 3.967586887449275,
"tokens_seen": 442171392
},
{
"epoch": 0.13,
"learning_rate": 0.00043737211634904715,
"loss": 2.9028,
"theoretical_loss": 3.967522208128366,
"tokens_seen": 442236928
},
{
"epoch": 0.13,
"learning_rate": 0.00043736208625877633,
"loss": 3.0509,
"theoretical_loss": 3.967457541075044,
"tokens_seen": 442302464
},
{
"epoch": 0.13,
"objective/train/docs_used": 262441,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8256170749664307,
"objective/train/theoretical_loss": 3.9673928862851655,
"objective/train/tokens_used": 462828000,
"theoretical_loss": 3.9673928862851655,
"tokens_seen": 442368000
},
{
"epoch": 0.13,
"learning_rate": 0.00043735205616850557,
"loss": 2.7654,
"theoretical_loss": 3.9673928862851655,
"tokens_seen": 442368000
},
{
"epoch": 0.13,
"learning_rate": 0.0004373420260782347,
"loss": 2.9835,
"theoretical_loss": 3.9673282437545883,
"tokens_seen": 442433536
},
{
"epoch": 0.13,
"learning_rate": 0.00043733199598796393,
"loss": 3.1724,
"theoretical_loss": 3.9672636134791737,
"tokens_seen": 442499072
},
{
"epoch": 0.13,
"learning_rate": 0.00043732196589769306,
"loss": 2.9039,
"theoretical_loss": 3.967198995454782,
"tokens_seen": 442564608
},
{
"epoch": 0.13,
"learning_rate": 0.0004373119358074223,
"loss": 2.7732,
"theoretical_loss": 3.9671343896772786,
"tokens_seen": 442630144
},
{
"epoch": 0.13,
"learning_rate": 0.00043730190571715147,
"loss": 2.7082,
"theoretical_loss": 3.9670697961425287,
"tokens_seen": 442695680
},
{
"epoch": 0.13,
"learning_rate": 0.00043729187562688065,
"loss": 3.2087,
"theoretical_loss": 3.967005214846401,
"tokens_seen": 442761216
},
{
"epoch": 0.13,
"learning_rate": 0.00043728184553660983,
"loss": 3.0608,
"theoretical_loss": 3.9669406457847654,
"tokens_seen": 442826752
},
{
"epoch": 0.13,
"learning_rate": 0.000437271815446339,
"loss": 2.9636,
"theoretical_loss": 3.9668760889534944,
"tokens_seen": 442892288
},
{
"epoch": 0.13,
"learning_rate": 0.0004372617853560682,
"loss": 3.1686,
"theoretical_loss": 3.9668115443484617,
"tokens_seen": 442957824
},
{
"epoch": 0.13,
"learning_rate": 0.00043725175526579743,
"loss": 2.8636,
"theoretical_loss": 3.966747011965544,
"tokens_seen": 443023360
},
{
"epoch": 0.13,
"learning_rate": 0.00043724172517552656,
"loss": 2.7879,
"theoretical_loss": 3.9666824918006185,
"tokens_seen": 443088896
},
{
"epoch": 0.13,
"learning_rate": 0.0004372316950852558,
"loss": 2.8745,
"theoretical_loss": 3.9666179838495657,
"tokens_seen": 443154432
},
{
"epoch": 0.13,
"learning_rate": 0.0004372216649949849,
"loss": 2.9572,
"theoretical_loss": 3.9665534881082687,
"tokens_seen": 443219968
},
{
"epoch": 0.13,
"learning_rate": 0.00043721163490471416,
"loss": 2.9991,
"theoretical_loss": 3.96648900457261,
"tokens_seen": 443285504
},
{
"epoch": 0.13,
"learning_rate": 0.00043720160481444334,
"loss": 3.0013,
"theoretical_loss": 3.966424533238477,
"tokens_seen": 443351040
},
{
"epoch": 0.13,
"learning_rate": 0.0004371915747241725,
"loss": 2.9459,
"theoretical_loss": 3.9663600741017575,
"tokens_seen": 443416576
},
{
"epoch": 0.13,
"learning_rate": 0.0004371815446339017,
"loss": 2.9219,
"theoretical_loss": 3.966295627158341,
"tokens_seen": 443482112
},
{
"epoch": 0.13,
"learning_rate": 0.00043717151454363093,
"loss": 2.8067,
"theoretical_loss": 3.9662311924041203,
"tokens_seen": 443547648
},
{
"epoch": 0.13,
"learning_rate": 0.00043716148445336006,
"loss": 3.0382,
"theoretical_loss": 3.96616676983499,
"tokens_seen": 443613184
},
{
"epoch": 0.13,
"learning_rate": 0.0004371514543630893,
"loss": 3.0499,
"theoretical_loss": 3.966102359446845,
"tokens_seen": 443678720
},
{
"epoch": 0.13,
"learning_rate": 0.0004371414242728184,
"loss": 2.7958,
"theoretical_loss": 3.9660379612355836,
"tokens_seen": 443744256
},
{
"epoch": 0.13,
"learning_rate": 0.00043713139418254766,
"loss": 2.9751,
"theoretical_loss": 3.9659735751971064,
"tokens_seen": 443809792
},
{
"epoch": 0.13,
"learning_rate": 0.00043712136409227684,
"loss": 2.864,
"theoretical_loss": 3.965909201327315,
"tokens_seen": 443875328
},
{
"epoch": 0.13,
"learning_rate": 0.000437111334002006,
"loss": 2.9858,
"theoretical_loss": 3.9658448396221138,
"tokens_seen": 443940864
},
{
"epoch": 0.13,
"objective/train/docs_used": 263839,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.407069444656372,
"objective/train/theoretical_loss": 3.9657804900774085,
"objective/train/tokens_used": 464466400,
"theoretical_loss": 3.9657804900774085,
"tokens_seen": 444006400
},
{
"epoch": 0.13,
"learning_rate": 0.0004371013039117352,
"loss": 2.6557,
"theoretical_loss": 3.9657804900774085,
"tokens_seen": 444006400
},
{
"epoch": 0.13,
"learning_rate": 0.0004370912738214644,
"loss": 2.9231,
"theoretical_loss": 3.9657161526891076,
"tokens_seen": 444071936
},
{
"epoch": 0.13,
"learning_rate": 0.00043708124373119356,
"loss": 3.0561,
"theoretical_loss": 3.9656518274531205,
"tokens_seen": 444137472
},
{
"epoch": 0.13,
"learning_rate": 0.0004370712136409228,
"loss": 2.9919,
"theoretical_loss": 3.9655875143653594,
"tokens_seen": 444203008
},
{
"epoch": 0.13,
"learning_rate": 0.0004370611835506519,
"loss": 3.0073,
"theoretical_loss": 3.965523213421738,
"tokens_seen": 444268544
},
{
"epoch": 0.13,
"learning_rate": 0.00043705115346038116,
"loss": 2.9746,
"theoretical_loss": 3.965458924618173,
"tokens_seen": 444334080
},
{
"epoch": 0.13,
"learning_rate": 0.0004370411233701103,
"loss": 2.9064,
"theoretical_loss": 3.965394647950581,
"tokens_seen": 444399616
},
{
"epoch": 0.13,
"learning_rate": 0.0004370310932798395,
"loss": 3.0837,
"theoretical_loss": 3.9653303834148828,
"tokens_seen": 444465152
},
{
"epoch": 0.13,
"learning_rate": 0.0004370210631895687,
"loss": 2.8178,
"theoretical_loss": 3.9652661310070005,
"tokens_seen": 444530688
},
{
"epoch": 0.13,
"learning_rate": 0.0004370110330992979,
"loss": 2.8228,
"theoretical_loss": 3.965201890722857,
"tokens_seen": 444596224
},
{
"epoch": 0.13,
"learning_rate": 0.00043700100300902707,
"loss": 2.8377,
"theoretical_loss": 3.9651376625583783,
"tokens_seen": 444661760
},
{
"epoch": 0.13,
"learning_rate": 0.0004369909729187563,
"loss": 3.2179,
"theoretical_loss": 3.965073446509493,
"tokens_seen": 444727296
},
{
"epoch": 0.13,
"learning_rate": 0.00043698094282848543,
"loss": 2.8271,
"theoretical_loss": 3.9650092425721297,
"tokens_seen": 444792832
},
{
"epoch": 0.13,
"learning_rate": 0.00043697091273821467,
"loss": 2.9445,
"theoretical_loss": 3.9649450507422204,
"tokens_seen": 444858368
},
{
"epoch": 0.13,
"learning_rate": 0.0004369608826479438,
"loss": 3.0335,
"theoretical_loss": 3.9648808710156986,
"tokens_seen": 444923904
},
{
"epoch": 0.13,
"learning_rate": 0.00043695085255767303,
"loss": 2.7964,
"theoretical_loss": 3.964816703388501,
"tokens_seen": 444989440
},
{
"epoch": 0.13,
"learning_rate": 0.0004369408224674022,
"loss": 2.9693,
"theoretical_loss": 3.9647525478565644,
"tokens_seen": 445054976
},
{
"epoch": 0.13,
"learning_rate": 0.0004369307923771314,
"loss": 2.8213,
"theoretical_loss": 3.9646884044158277,
"tokens_seen": 445120512
},
{
"epoch": 0.13,
"learning_rate": 0.0004369207622868606,
"loss": 2.8553,
"theoretical_loss": 3.964624273062234,
"tokens_seen": 445186048
},
{
"epoch": 0.13,
"learning_rate": 0.00043691073219658975,
"loss": 3.0907,
"theoretical_loss": 3.9645601537917248,
"tokens_seen": 445251584
},
{
"epoch": 0.13,
"learning_rate": 0.000436900702106319,
"loss": 3.0419,
"theoretical_loss": 3.964496046600247,
"tokens_seen": 445317120
},
{
"epoch": 0.13,
"learning_rate": 0.00043689067201604817,
"loss": 2.8665,
"theoretical_loss": 3.9644319514837467,
"tokens_seen": 445382656
},
{
"epoch": 0.13,
"learning_rate": 0.00043688064192577735,
"loss": 3.0254,
"theoretical_loss": 3.964367868438175,
"tokens_seen": 445448192
},
{
"epoch": 0.14,
"learning_rate": 0.00043687061183550653,
"loss": 2.9733,
"theoretical_loss": 3.9643037974594817,
"tokens_seen": 445513728
},
{
"epoch": 0.14,
"learning_rate": 0.00043686058174523577,
"loss": 3.0529,
"theoretical_loss": 3.96423973854362,
"tokens_seen": 445579264
},
{
"epoch": 0.14,
"objective/train/docs_used": 264540,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0680365562438965,
"objective/train/theoretical_loss": 3.9641756916865463,
"objective/train/tokens_used": 466104800,
"theoretical_loss": 3.9641756916865463,
"tokens_seen": 445644800
},
{
"epoch": 0.14,
"learning_rate": 0.0004368505516549649,
"loss": 2.7958,
"theoretical_loss": 3.9641756916865463,
"tokens_seen": 445644800
},
{
"epoch": 0.14,
"learning_rate": 0.00043684052156469413,
"loss": 2.7212,
"theoretical_loss": 3.964111656884217,
"tokens_seen": 445710336
},
{
"epoch": 0.14,
"learning_rate": 0.00043683049147442326,
"loss": 2.8679,
"theoretical_loss": 3.9640476341325916,
"tokens_seen": 445775872
},
{
"epoch": 0.14,
"learning_rate": 0.0004368204613841525,
"loss": 2.8905,
"theoretical_loss": 3.96398362342763,
"tokens_seen": 445841408
},
{
"epoch": 0.14,
"learning_rate": 0.00043681043129388167,
"loss": 3.0276,
"theoretical_loss": 3.9639196247652966,
"tokens_seen": 445906944
},
{
"epoch": 0.14,
"learning_rate": 0.00043680040120361085,
"loss": 2.7716,
"theoretical_loss": 3.9638556381415553,
"tokens_seen": 445972480
},
{
"epoch": 0.14,
"learning_rate": 0.00043679037111334003,
"loss": 2.97,
"theoretical_loss": 3.9637916635523736,
"tokens_seen": 446038016
},
{
"epoch": 0.14,
"learning_rate": 0.0004367803410230692,
"loss": 3.0207,
"theoretical_loss": 3.9637277009937195,
"tokens_seen": 446103552
},
{
"epoch": 0.14,
"learning_rate": 0.0004367703109327984,
"loss": 3.0506,
"theoretical_loss": 3.9636637504615653,
"tokens_seen": 446169088
},
{
"epoch": 0.14,
"learning_rate": 0.00043676028084252763,
"loss": 2.8433,
"theoretical_loss": 3.963599811951882,
"tokens_seen": 446234624
},
{
"epoch": 0.14,
"learning_rate": 0.00043675025075225676,
"loss": 2.9764,
"theoretical_loss": 3.9635358854606455,
"tokens_seen": 446300160
},
{
"epoch": 0.14,
"learning_rate": 0.000436740220661986,
"loss": 2.8531,
"theoretical_loss": 3.9634719709838317,
"tokens_seen": 446365696
},
{
"epoch": 0.14,
"learning_rate": 0.0004367301905717151,
"loss": 2.985,
"theoretical_loss": 3.963408068517419,
"tokens_seen": 446431232
},
{
"epoch": 0.14,
"learning_rate": 0.00043672016048144436,
"loss": 2.9554,
"theoretical_loss": 3.9633441780573886,
"tokens_seen": 446496768
},
{
"epoch": 0.14,
"learning_rate": 0.00043671013039117354,
"loss": 2.8556,
"theoretical_loss": 3.963280299599722,
"tokens_seen": 446562304
},
{
"epoch": 0.14,
"learning_rate": 0.0004367001003009027,
"loss": 2.864,
"theoretical_loss": 3.9632164331404045,
"tokens_seen": 446627840
},
{
"epoch": 0.14,
"learning_rate": 0.0004366900702106319,
"loss": 2.7874,
"theoretical_loss": 3.9631525786754214,
"tokens_seen": 446693376
},
{
"epoch": 0.14,
"learning_rate": 0.00043668004012036113,
"loss": 2.7899,
"theoretical_loss": 3.9630887362007616,
"tokens_seen": 446758912
},
{
"epoch": 0.14,
"learning_rate": 0.00043667001003009026,
"loss": 3.0638,
"theoretical_loss": 3.9630249057124147,
"tokens_seen": 446824448
},
{
"epoch": 0.14,
"learning_rate": 0.0004366599799398195,
"loss": 2.8952,
"theoretical_loss": 3.9629610872063727,
"tokens_seen": 446889984
},
{
"epoch": 0.14,
"learning_rate": 0.0004366499498495486,
"loss": 2.8926,
"theoretical_loss": 3.9628972806786305,
"tokens_seen": 446955520
},
{
"epoch": 0.14,
"learning_rate": 0.00043663991975927786,
"loss": 2.74,
"theoretical_loss": 3.9628334861251826,
"tokens_seen": 447021056
},
{
"epoch": 0.14,
"learning_rate": 0.00043662988966900704,
"loss": 2.9397,
"theoretical_loss": 3.9627697035420284,
"tokens_seen": 447086592
},
{
"epoch": 0.14,
"learning_rate": 0.0004366198595787362,
"loss": 3.0488,
"theoretical_loss": 3.9627059329251666,
"tokens_seen": 447152128
},
{
"epoch": 0.14,
"learning_rate": 0.0004366098294884654,
"loss": 2.946,
"theoretical_loss": 3.9626421742705986,
"tokens_seen": 447217664
},
{
"epoch": 0.14,
"objective/train/docs_used": 265849,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.205967664718628,
"objective/train/theoretical_loss": 3.9625784275743294,
"objective/train/tokens_used": 467743200,
"theoretical_loss": 3.9625784275743294,
"tokens_seen": 447283200
},
{
"epoch": 0.14,
"learning_rate": 0.0004365997993981946,
"loss": 3.0183,
"theoretical_loss": 3.9625784275743294,
"tokens_seen": 447283200
},
{
"epoch": 0.14,
"learning_rate": 0.00043658976930792376,
"loss": 2.8804,
"theoretical_loss": 3.9625146928323627,
"tokens_seen": 447348736
},
{
"epoch": 0.14,
"learning_rate": 0.000436579739217653,
"loss": 2.8588,
"theoretical_loss": 3.962450970040707,
"tokens_seen": 447414272
},
{
"epoch": 0.14,
"learning_rate": 0.00043656970912738213,
"loss": 2.9412,
"theoretical_loss": 3.9623872591953715,
"tokens_seen": 447479808
},
{
"epoch": 0.14,
"learning_rate": 0.00043655967903711136,
"loss": 2.982,
"theoretical_loss": 3.9623235602923677,
"tokens_seen": 447545344
},
{
"epoch": 0.14,
"learning_rate": 0.0004365496489468405,
"loss": 2.8512,
"theoretical_loss": 3.9622598733277083,
"tokens_seen": 447610880
},
{
"epoch": 0.14,
"learning_rate": 0.0004365396188565697,
"loss": 2.707,
"theoretical_loss": 3.9621961982974088,
"tokens_seen": 447676416
},
{
"epoch": 0.14,
"learning_rate": 0.0004365295887662989,
"loss": 3.0508,
"theoretical_loss": 3.962132535197486,
"tokens_seen": 447741952
},
{
"epoch": 0.14,
"learning_rate": 0.0004365195586760281,
"loss": 2.8024,
"theoretical_loss": 3.9620688840239593,
"tokens_seen": 447807488
},
{
"epoch": 0.14,
"learning_rate": 0.00043650952858575727,
"loss": 3.1377,
"theoretical_loss": 3.9620052447728478,
"tokens_seen": 447873024
},
{
"epoch": 0.14,
"learning_rate": 0.0004364994984954865,
"loss": 2.8089,
"theoretical_loss": 3.9619416174401767,
"tokens_seen": 447938560
},
{
"epoch": 0.14,
"learning_rate": 0.00043648946840521563,
"loss": 3.0379,
"theoretical_loss": 3.9618780020219693,
"tokens_seen": 448004096
},
{
"epoch": 0.14,
"learning_rate": 0.00043647943831494487,
"loss": 2.8921,
"theoretical_loss": 3.961814398514252,
"tokens_seen": 448069632
},
{
"epoch": 0.14,
"learning_rate": 0.000436469408224674,
"loss": 2.8712,
"theoretical_loss": 3.961750806913054,
"tokens_seen": 448135168
},
{
"epoch": 0.14,
"learning_rate": 0.00043645937813440323,
"loss": 2.8566,
"theoretical_loss": 3.961687227214406,
"tokens_seen": 448200704
},
{
"epoch": 0.14,
"learning_rate": 0.0004364493480441324,
"loss": 2.8192,
"theoretical_loss": 3.961623659414338,
"tokens_seen": 448266240
},
{
"epoch": 0.14,
"learning_rate": 0.0004364393179538616,
"loss": 2.774,
"theoretical_loss": 3.961560103508887,
"tokens_seen": 448331776
},
{
"epoch": 0.14,
"learning_rate": 0.00043642928786359077,
"loss": 3.0392,
"theoretical_loss": 3.961496559494087,
"tokens_seen": 448397312
},
{
"epoch": 0.14,
"learning_rate": 0.00043641925777331995,
"loss": 2.9419,
"theoretical_loss": 3.9614330273659775,
"tokens_seen": 448462848
},
{
"epoch": 0.14,
"learning_rate": 0.00043640922768304913,
"loss": 2.841,
"theoretical_loss": 3.9613695071205974,
"tokens_seen": 448528384
},
{
"epoch": 0.14,
"learning_rate": 0.00043639919759277837,
"loss": 2.9117,
"theoretical_loss": 3.9613059987539887,
"tokens_seen": 448593920
},
{
"epoch": 0.14,
"learning_rate": 0.0004363891675025075,
"loss": 2.7283,
"theoretical_loss": 3.9612425022621953,
"tokens_seen": 448659456
},
{
"epoch": 0.14,
"learning_rate": 0.00043637913741223673,
"loss": 3.0715,
"theoretical_loss": 3.9611790176412622,
"tokens_seen": 448724992
},
{
"epoch": 0.14,
"learning_rate": 0.0004363691073219659,
"loss": 2.7813,
"theoretical_loss": 3.961115544887238,
"tokens_seen": 448790528
},
{
"epoch": 0.14,
"learning_rate": 0.0004363590772316951,
"loss": 2.8004,
"theoretical_loss": 3.961052083996171,
"tokens_seen": 448856064
},
{
"epoch": 0.14,
"objective/train/docs_used": 266141,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7323062419891357,
"objective/train/theoretical_loss": 3.960988634964113,
"objective/train/tokens_used": 469381600,
"theoretical_loss": 3.960988634964113,
"tokens_seen": 448921600
},
{
"epoch": 0.14,
"learning_rate": 0.0004363490471414243,
"loss": 2.9443,
"theoretical_loss": 3.960988634964113,
"tokens_seen": 448921600
},
{
"epoch": 0.14,
"learning_rate": 0.00043633901705115346,
"loss": 2.8353,
"theoretical_loss": 3.960925197787116,
"tokens_seen": 448987136
},
{
"epoch": 0.14,
"learning_rate": 0.00043632898696088264,
"loss": 2.7991,
"theoretical_loss": 3.9608617724612367,
"tokens_seen": 449052672
},
{
"epoch": 0.14,
"learning_rate": 0.00043631895687061187,
"loss": 3.0411,
"theoretical_loss": 3.960798358982531,
"tokens_seen": 449118208
},
{
"epoch": 0.14,
"learning_rate": 0.000436308926780341,
"loss": 2.8481,
"theoretical_loss": 3.960734957347058,
"tokens_seen": 449183744
},
{
"epoch": 0.14,
"learning_rate": 0.00043629889669007023,
"loss": 2.9892,
"theoretical_loss": 3.960671567550878,
"tokens_seen": 449249280
},
{
"epoch": 0.14,
"learning_rate": 0.00043628886659979936,
"loss": 3.0781,
"theoretical_loss": 3.9606081895900545,
"tokens_seen": 449314816
},
{
"epoch": 0.14,
"learning_rate": 0.0004362788365095286,
"loss": 2.7987,
"theoretical_loss": 3.9605448234606513,
"tokens_seen": 449380352
},
{
"epoch": 0.14,
"learning_rate": 0.0004362688064192578,
"loss": 3.1004,
"theoretical_loss": 3.9604814691587347,
"tokens_seen": 449445888
},
{
"epoch": 0.14,
"learning_rate": 0.00043625877632898696,
"loss": 2.9069,
"theoretical_loss": 3.9604181266803726,
"tokens_seen": 449511424
},
{
"epoch": 0.14,
"learning_rate": 0.00043624874623871614,
"loss": 3.0532,
"theoretical_loss": 3.960354796021636,
"tokens_seen": 449576960
},
{
"epoch": 0.14,
"learning_rate": 0.0004362387161484453,
"loss": 2.8056,
"theoretical_loss": 3.960291477178596,
"tokens_seen": 449642496
},
{
"epoch": 0.14,
"learning_rate": 0.0004362286860581745,
"loss": 2.8333,
"theoretical_loss": 3.960228170147327,
"tokens_seen": 449708032
},
{
"epoch": 0.14,
"learning_rate": 0.00043621865596790374,
"loss": 3.1691,
"theoretical_loss": 3.9601648749239047,
"tokens_seen": 449773568
},
{
"epoch": 0.14,
"learning_rate": 0.00043620862587763286,
"loss": 2.9981,
"theoretical_loss": 3.960101591504406,
"tokens_seen": 449839104
},
{
"epoch": 0.14,
"learning_rate": 0.0004361985957873621,
"loss": 3.0074,
"theoretical_loss": 3.960038319884912,
"tokens_seen": 449904640
},
{
"epoch": 0.14,
"learning_rate": 0.0004361885656970913,
"loss": 2.7211,
"theoretical_loss": 3.9599750600615025,
"tokens_seen": 449970176
},
{
"epoch": 0.14,
"learning_rate": 0.00043617853560682046,
"loss": 2.7248,
"theoretical_loss": 3.9599118120302608,
"tokens_seen": 450035712
},
{
"epoch": 0.14,
"learning_rate": 0.0004361685055165497,
"loss": 2.7913,
"theoretical_loss": 3.959848575787273,
"tokens_seen": 450101248
},
{
"epoch": 0.14,
"learning_rate": 0.0004361584754262788,
"loss": 2.513,
"theoretical_loss": 3.959785351328626,
"tokens_seen": 450166784
},
{
"epoch": 0.14,
"learning_rate": 0.00043614844533600806,
"loss": 3.0842,
"theoretical_loss": 3.9597221386504073,
"tokens_seen": 450232320
},
{
"epoch": 0.14,
"learning_rate": 0.00043613841524573724,
"loss": 2.9115,
"theoretical_loss": 3.9596589377487086,
"tokens_seen": 450297856
},
{
"epoch": 0.14,
"learning_rate": 0.0004361283851554664,
"loss": 2.9273,
"theoretical_loss": 3.9595957486196225,
"tokens_seen": 450363392
},
{
"epoch": 0.14,
"learning_rate": 0.0004361183550651956,
"loss": 3.0515,
"theoretical_loss": 3.9595325712592433,
"tokens_seen": 450428928
},
{
"epoch": 0.14,
"learning_rate": 0.0004361083249749248,
"loss": 3.0497,
"theoretical_loss": 3.9594694056636675,
"tokens_seen": 450494464
},
{
"epoch": 0.14,
"objective/train/docs_used": 266764,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3204047679901123,
"objective/train/theoretical_loss": 3.9594062518289923,
"objective/train/tokens_used": 471020000,
"theoretical_loss": 3.9594062518289923,
"tokens_seen": 450560000
},
{
"epoch": 0.14,
"learning_rate": 0.00043609829488465397,
"loss": 2.9596,
"theoretical_loss": 3.9594062518289923,
"tokens_seen": 450560000
},
{
"epoch": 0.14,
"learning_rate": 0.0004360882647943832,
"loss": 2.7606,
"theoretical_loss": 3.9593431097513188,
"tokens_seen": 450625536
},
{
"epoch": 0.14,
"learning_rate": 0.00043607823470411233,
"loss": 3.0245,
"theoretical_loss": 3.9592799794267486,
"tokens_seen": 450691072
},
{
"epoch": 0.14,
"learning_rate": 0.00043606820461384156,
"loss": 2.9339,
"theoretical_loss": 3.9592168608513854,
"tokens_seen": 450756608
},
{
"epoch": 0.14,
"learning_rate": 0.0004360581745235707,
"loss": 3.0752,
"theoretical_loss": 3.959153754021334,
"tokens_seen": 450822144
},
{
"epoch": 0.14,
"learning_rate": 0.0004360481444332999,
"loss": 2.9678,
"theoretical_loss": 3.9590906589327037,
"tokens_seen": 450887680
},
{
"epoch": 0.14,
"learning_rate": 0.0004360381143430291,
"loss": 2.8541,
"theoretical_loss": 3.9590275755816022,
"tokens_seen": 450953216
},
{
"epoch": 0.14,
"learning_rate": 0.0004360280842527583,
"loss": 2.8563,
"theoretical_loss": 3.958964503964141,
"tokens_seen": 451018752
},
{
"epoch": 0.14,
"learning_rate": 0.00043601805416248747,
"loss": 3.1018,
"theoretical_loss": 3.9589014440764334,
"tokens_seen": 451084288
},
{
"epoch": 0.14,
"learning_rate": 0.0004360080240722167,
"loss": 2.9896,
"theoretical_loss": 3.9588383959145945,
"tokens_seen": 451149824
},
{
"epoch": 0.14,
"learning_rate": 0.00043599799398194583,
"loss": 2.8416,
"theoretical_loss": 3.9587753594747404,
"tokens_seen": 451215360
},
{
"epoch": 0.14,
"learning_rate": 0.00043598796389167507,
"loss": 2.883,
"theoretical_loss": 3.95871233475299,
"tokens_seen": 451280896
},
{
"epoch": 0.14,
"learning_rate": 0.0004359779338014042,
"loss": 2.9736,
"theoretical_loss": 3.9586493217454635,
"tokens_seen": 451346432
},
{
"epoch": 0.14,
"learning_rate": 0.00043596790371113343,
"loss": 2.8513,
"theoretical_loss": 3.9585863204482834,
"tokens_seen": 451411968
},
{
"epoch": 0.14,
"learning_rate": 0.0004359578736208626,
"loss": 3.0644,
"theoretical_loss": 3.958523330857574,
"tokens_seen": 451477504
},
{
"epoch": 0.14,
"learning_rate": 0.0004359478435305918,
"loss": 2.7629,
"theoretical_loss": 3.9584603529694604,
"tokens_seen": 451543040
},
{
"epoch": 0.14,
"learning_rate": 0.00043593781344032097,
"loss": 3.0606,
"theoretical_loss": 3.9583973867800717,
"tokens_seen": 451608576
},
{
"epoch": 0.14,
"learning_rate": 0.00043592778335005015,
"loss": 2.8071,
"theoretical_loss": 3.958334432285537,
"tokens_seen": 451674112
},
{
"epoch": 0.14,
"learning_rate": 0.00043591775325977933,
"loss": 2.9459,
"theoretical_loss": 3.9582714894819873,
"tokens_seen": 451739648
},
{
"epoch": 0.14,
"learning_rate": 0.00043590772316950857,
"loss": 2.9942,
"theoretical_loss": 3.9582085583655564,
"tokens_seen": 451805184
},
{
"epoch": 0.14,
"learning_rate": 0.0004358976930792377,
"loss": 3.0168,
"theoretical_loss": 3.958145638932379,
"tokens_seen": 451870720
},
{
"epoch": 0.14,
"learning_rate": 0.00043588766298896693,
"loss": 3.036,
"theoretical_loss": 3.9580827311785924,
"tokens_seen": 451936256
},
{
"epoch": 0.14,
"learning_rate": 0.0004358776328986961,
"loss": 2.8268,
"theoretical_loss": 3.9580198351003357,
"tokens_seen": 452001792
},
{
"epoch": 0.14,
"learning_rate": 0.0004358676028084253,
"loss": 2.9403,
"theoretical_loss": 3.9579569506937498,
"tokens_seen": 452067328
},
{
"epoch": 0.14,
"learning_rate": 0.0004358575727181545,
"loss": 2.7007,
"theoretical_loss": 3.9578940779549763,
"tokens_seen": 452132864
},
{
"epoch": 0.14,
"objective/train/docs_used": 268192,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8176727294921875,
"objective/train/theoretical_loss": 3.9578312168801597,
"objective/train/tokens_used": 472658400,
"theoretical_loss": 3.9578312168801597,
"tokens_seen": 452198400
},
{
"epoch": 0.14,
"learning_rate": 0.00043584754262788366,
"loss": 2.9893,
"theoretical_loss": 3.9578312168801597,
"tokens_seen": 452198400
},
{
"epoch": 0.14,
"learning_rate": 0.00043583751253761284,
"loss": 3.0281,
"theoretical_loss": 3.957768367465447,
"tokens_seen": 452263936
},
{
"epoch": 0.14,
"learning_rate": 0.00043582748244734207,
"loss": 2.9604,
"theoretical_loss": 3.9577055297069856,
"tokens_seen": 452329472
},
{
"epoch": 0.14,
"learning_rate": 0.0004358174523570712,
"loss": 2.8216,
"theoretical_loss": 3.957642703600926,
"tokens_seen": 452395008
},
{
"epoch": 0.14,
"learning_rate": 0.00043580742226680043,
"loss": 3.2024,
"theoretical_loss": 3.9575798891434184,
"tokens_seen": 452460544
},
{
"epoch": 0.14,
"learning_rate": 0.00043579739217652956,
"loss": 2.9174,
"theoretical_loss": 3.957517086330618,
"tokens_seen": 452526080
},
{
"epoch": 0.14,
"learning_rate": 0.0004357873620862588,
"loss": 3.0686,
"theoretical_loss": 3.957454295158679,
"tokens_seen": 452591616
},
{
"epoch": 0.14,
"learning_rate": 0.000435777331995988,
"loss": 2.946,
"theoretical_loss": 3.9573915156237587,
"tokens_seen": 452657152
},
{
"epoch": 0.14,
"learning_rate": 0.00043576730190571716,
"loss": 3.0834,
"theoretical_loss": 3.9573287477220167,
"tokens_seen": 452722688
},
{
"epoch": 0.14,
"learning_rate": 0.00043575727181544634,
"loss": 3.2719,
"theoretical_loss": 3.957265991449613,
"tokens_seen": 452788224
},
{
"epoch": 0.14,
"learning_rate": 0.0004357472417251755,
"loss": 2.8808,
"theoretical_loss": 3.957203246802711,
"tokens_seen": 452853760
},
{
"epoch": 0.14,
"learning_rate": 0.0004357372116349047,
"loss": 3.0469,
"theoretical_loss": 3.9571405137774747,
"tokens_seen": 452919296
},
{
"epoch": 0.14,
"learning_rate": 0.00043572718154463394,
"loss": 2.9176,
"theoretical_loss": 3.9570777923700704,
"tokens_seen": 452984832
},
{
"epoch": 0.14,
"learning_rate": 0.00043571715145436306,
"loss": 2.864,
"theoretical_loss": 3.9570150825766657,
"tokens_seen": 453050368
},
{
"epoch": 0.14,
"learning_rate": 0.0004357071213640923,
"loss": 3.0054,
"theoretical_loss": 3.956952384393432,
"tokens_seen": 453115904
},
{
"epoch": 0.14,
"learning_rate": 0.0004356970912738215,
"loss": 2.9216,
"theoretical_loss": 3.956889697816539,
"tokens_seen": 453181440
},
{
"epoch": 0.14,
"learning_rate": 0.00043568706118355066,
"loss": 3.0549,
"theoretical_loss": 3.9568270228421616,
"tokens_seen": 453246976
},
{
"epoch": 0.14,
"learning_rate": 0.00043567703109327984,
"loss": 2.995,
"theoretical_loss": 3.956764359466475,
"tokens_seen": 453312512
},
{
"epoch": 0.14,
"learning_rate": 0.000435667001003009,
"loss": 2.7352,
"theoretical_loss": 3.9567017076856565,
"tokens_seen": 453378048
},
{
"epoch": 0.14,
"learning_rate": 0.0004356569709127382,
"loss": 2.9331,
"theoretical_loss": 3.9566390674958845,
"tokens_seen": 453443584
},
{
"epoch": 0.14,
"learning_rate": 0.00043564694082246744,
"loss": 3.1616,
"theoretical_loss": 3.9565764388933395,
"tokens_seen": 453509120
},
{
"epoch": 0.14,
"learning_rate": 0.00043563691073219657,
"loss": 2.943,
"theoretical_loss": 3.9565138218742053,
"tokens_seen": 453574656
},
{
"epoch": 0.14,
"learning_rate": 0.0004356268806419258,
"loss": 2.9295,
"theoretical_loss": 3.9564512164346652,
"tokens_seen": 453640192
},
{
"epoch": 0.14,
"learning_rate": 0.00043561685055165493,
"loss": 3.067,
"theoretical_loss": 3.956388622570906,
"tokens_seen": 453705728
},
{
"epoch": 0.14,
"learning_rate": 0.00043560682046138417,
"loss": 3.1012,
"theoretical_loss": 3.956326040279116,
"tokens_seen": 453771264
},
{
"epoch": 0.14,
"objective/train/docs_used": 268717,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.054826021194458,
"objective/train/theoretical_loss": 3.956263469555484,
"objective/train/tokens_used": 474296800,
"theoretical_loss": 3.956263469555484,
"tokens_seen": 453836800
},
{
"epoch": 0.14,
"learning_rate": 0.00043559679037111335,
"loss": 2.8934,
"theoretical_loss": 3.956263469555484,
"tokens_seen": 453836800
},
{
"epoch": 0.14,
"learning_rate": 0.00043558676028084253,
"loss": 3.0294,
"theoretical_loss": 3.956200910396202,
"tokens_seen": 453902336
},
{
"epoch": 0.14,
"learning_rate": 0.0004355767301905717,
"loss": 2.7316,
"theoretical_loss": 3.9561383627974642,
"tokens_seen": 453967872
},
{
"epoch": 0.14,
"learning_rate": 0.0004355667001003009,
"loss": 2.82,
"theoretical_loss": 3.956075826755465,
"tokens_seen": 454033408
},
{
"epoch": 0.14,
"learning_rate": 0.00043555667001003007,
"loss": 2.59,
"theoretical_loss": 3.9560133022664026,
"tokens_seen": 454098944
},
{
"epoch": 0.14,
"learning_rate": 0.0004355466399197593,
"loss": 2.8896,
"theoretical_loss": 3.955950789326474,
"tokens_seen": 454164480
},
{
"epoch": 0.14,
"learning_rate": 0.00043553660982948843,
"loss": 2.7608,
"theoretical_loss": 3.955888287931881,
"tokens_seen": 454230016
},
{
"epoch": 0.14,
"learning_rate": 0.00043552657973921767,
"loss": 2.8049,
"theoretical_loss": 3.955825798078826,
"tokens_seen": 454295552
},
{
"epoch": 0.14,
"learning_rate": 0.00043551654964894685,
"loss": 2.7974,
"theoretical_loss": 3.9557633197635127,
"tokens_seen": 454361088
},
{
"epoch": 0.14,
"learning_rate": 0.00043550651955867603,
"loss": 2.8448,
"theoretical_loss": 3.9557008529821474,
"tokens_seen": 454426624
},
{
"epoch": 0.14,
"learning_rate": 0.0004354964894684052,
"loss": 2.9746,
"theoretical_loss": 3.955638397730939,
"tokens_seen": 454492160
},
{
"epoch": 0.14,
"learning_rate": 0.0004354864593781344,
"loss": 2.9851,
"theoretical_loss": 3.955575954006095,
"tokens_seen": 454557696
},
{
"epoch": 0.14,
"learning_rate": 0.0004354764292878636,
"loss": 3.1936,
"theoretical_loss": 3.9555135218038284,
"tokens_seen": 454623232
},
{
"epoch": 0.14,
"learning_rate": 0.0004354663991975928,
"loss": 2.8998,
"theoretical_loss": 3.955451101120352,
"tokens_seen": 454688768
},
{
"epoch": 0.14,
"learning_rate": 0.00043545636910732194,
"loss": 2.8088,
"theoretical_loss": 3.955388691951881,
"tokens_seen": 454754304
},
{
"epoch": 0.14,
"learning_rate": 0.00043544633901705117,
"loss": 2.9158,
"theoretical_loss": 3.9553262942946312,
"tokens_seen": 454819840
},
{
"epoch": 0.14,
"learning_rate": 0.0004354363089267803,
"loss": 3.0216,
"theoretical_loss": 3.9552639081448224,
"tokens_seen": 454885376
},
{
"epoch": 0.14,
"learning_rate": 0.00043542627883650953,
"loss": 2.8275,
"theoretical_loss": 3.9552015334986743,
"tokens_seen": 454950912
},
{
"epoch": 0.14,
"learning_rate": 0.00043541624874623877,
"loss": 2.887,
"theoretical_loss": 3.9551391703524086,
"tokens_seen": 455016448
},
{
"epoch": 0.14,
"learning_rate": 0.0004354062186559679,
"loss": 2.9557,
"theoretical_loss": 3.95507681870225,
"tokens_seen": 455081984
},
{
"epoch": 0.14,
"learning_rate": 0.00043539618856569713,
"loss": 2.8015,
"theoretical_loss": 3.955014478544425,
"tokens_seen": 455147520
},
{
"epoch": 0.14,
"learning_rate": 0.0004353861584754263,
"loss": 3.0108,
"theoretical_loss": 3.954952149875159,
"tokens_seen": 455213056
},
{
"epoch": 0.14,
"learning_rate": 0.0004353761283851555,
"loss": 2.8855,
"theoretical_loss": 3.9548898326906823,
"tokens_seen": 455278592
},
{
"epoch": 0.14,
"learning_rate": 0.0004353660982948847,
"loss": 3.0832,
"theoretical_loss": 3.954827526987226,
"tokens_seen": 455344128
},
{
"epoch": 0.14,
"learning_rate": 0.00043535606820461386,
"loss": 3.0242,
"theoretical_loss": 3.954765232761023,
"tokens_seen": 455409664
},
{
"epoch": 0.14,
"objective/train/docs_used": 269967,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.981569290161133,
"objective/train/theoretical_loss": 3.954702950008308,
"objective/train/tokens_used": 475935200,
"theoretical_loss": 3.954702950008308,
"tokens_seen": 455475200
},
{
"epoch": 0.14,
"learning_rate": 0.00043534603811434304,
"loss": 3.0046,
"theoretical_loss": 3.954702950008308,
"tokens_seen": 455475200
},
{
"epoch": 0.14,
"learning_rate": 0.00043533600802407227,
"loss": 3.0107,
"theoretical_loss": 3.954640678725317,
"tokens_seen": 455540736
},
{
"epoch": 0.14,
"learning_rate": 0.0004353259779338014,
"loss": 2.8819,
"theoretical_loss": 3.9545784189082887,
"tokens_seen": 455606272
},
{
"epoch": 0.14,
"learning_rate": 0.00043531594784353063,
"loss": 2.7767,
"theoretical_loss": 3.9545161705534624,
"tokens_seen": 455671808
},
{
"epoch": 0.14,
"learning_rate": 0.00043530591775325976,
"loss": 3.0823,
"theoretical_loss": 3.9544539336570796,
"tokens_seen": 455737344
},
{
"epoch": 0.14,
"learning_rate": 0.000435295887662989,
"loss": 2.8641,
"theoretical_loss": 3.954391708215385,
"tokens_seen": 455802880
},
{
"epoch": 0.14,
"learning_rate": 0.0004352858575727182,
"loss": 2.9376,
"theoretical_loss": 3.954329494224623,
"tokens_seen": 455868416
},
{
"epoch": 0.14,
"learning_rate": 0.00043527582748244736,
"loss": 2.9881,
"theoretical_loss": 3.9542672916810404,
"tokens_seen": 455933952
},
{
"epoch": 0.14,
"learning_rate": 0.00043526579739217654,
"loss": 2.596,
"theoretical_loss": 3.9542051005808867,
"tokens_seen": 455999488
},
{
"epoch": 0.14,
"learning_rate": 0.0004352557673019057,
"loss": 2.9525,
"theoretical_loss": 3.954142920920412,
"tokens_seen": 456065024
},
{
"epoch": 0.14,
"learning_rate": 0.0004352457372116349,
"loss": 3.0159,
"theoretical_loss": 3.954080752695868,
"tokens_seen": 456130560
},
{
"epoch": 0.14,
"learning_rate": 0.00043523570712136414,
"loss": 3.0215,
"theoretical_loss": 3.95401859590351,
"tokens_seen": 456196096
},
{
"epoch": 0.14,
"learning_rate": 0.00043522567703109326,
"loss": 2.8253,
"theoretical_loss": 3.953956450539594,
"tokens_seen": 456261632
},
{
"epoch": 0.14,
"learning_rate": 0.0004352156469408225,
"loss": 2.9501,
"theoretical_loss": 3.953894316600376,
"tokens_seen": 456327168
},
{
"epoch": 0.14,
"learning_rate": 0.0004352056168505517,
"loss": 2.9843,
"theoretical_loss": 3.953832194082117,
"tokens_seen": 456392704
},
{
"epoch": 0.14,
"learning_rate": 0.00043519558676028086,
"loss": 3.1067,
"theoretical_loss": 3.9537700829810767,
"tokens_seen": 456458240
},
{
"epoch": 0.14,
"learning_rate": 0.00043518555667001004,
"loss": 2.8276,
"theoretical_loss": 3.95370798329352,
"tokens_seen": 456523776
},
{
"epoch": 0.14,
"learning_rate": 0.0004351755265797392,
"loss": 2.8316,
"theoretical_loss": 3.9536458950157094,
"tokens_seen": 456589312
},
{
"epoch": 0.14,
"learning_rate": 0.0004351654964894684,
"loss": 2.6923,
"theoretical_loss": 3.9535838181439127,
"tokens_seen": 456654848
},
{
"epoch": 0.14,
"learning_rate": 0.00043515546639919764,
"loss": 2.7719,
"theoretical_loss": 3.9535217526743978,
"tokens_seen": 456720384
},
{
"epoch": 0.14,
"learning_rate": 0.00043514543630892677,
"loss": 2.9918,
"theoretical_loss": 3.953459698603434,
"tokens_seen": 456785920
},
{
"epoch": 0.14,
"learning_rate": 0.000435135406218656,
"loss": 2.9902,
"theoretical_loss": 3.9533976559272936,
"tokens_seen": 456851456
},
{
"epoch": 0.14,
"learning_rate": 0.00043512537612838513,
"loss": 2.9201,
"theoretical_loss": 3.95333562464225,
"tokens_seen": 456916992
},
{
"epoch": 0.14,
"learning_rate": 0.00043511534603811437,
"loss": 3.0096,
"theoretical_loss": 3.9532736047445787,
"tokens_seen": 456982528
},
{
"epoch": 0.14,
"learning_rate": 0.00043510531594784355,
"loss": 2.8607,
"theoretical_loss": 3.953211596230556,
"tokens_seen": 457048064
},
{
"epoch": 0.14,
"objective/train/docs_used": 270622,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.566558361053467,
"objective/train/theoretical_loss": 3.953149599096461,
"objective/train/tokens_used": 477573600,
"theoretical_loss": 3.953149599096461,
"tokens_seen": 457113600
},
{
"epoch": 0.14,
"learning_rate": 0.00043509528585757273,
"loss": 2.7782,
"theoretical_loss": 3.953149599096461,
"tokens_seen": 457113600
},
{
"epoch": 0.14,
"learning_rate": 0.0004350852557673019,
"loss": 2.9516,
"theoretical_loss": 3.9530876133385737,
"tokens_seen": 457179136
},
{
"epoch": 0.14,
"learning_rate": 0.0004350752256770311,
"loss": 3.0632,
"theoretical_loss": 3.953025638953177,
"tokens_seen": 457244672
},
{
"epoch": 0.14,
"learning_rate": 0.00043506519558676027,
"loss": 3.1882,
"theoretical_loss": 3.9529636759365543,
"tokens_seen": 457310208
},
{
"epoch": 0.14,
"learning_rate": 0.0004350551654964895,
"loss": 2.9347,
"theoretical_loss": 3.9529017242849918,
"tokens_seen": 457375744
},
{
"epoch": 0.14,
"learning_rate": 0.00043504513540621863,
"loss": 2.8628,
"theoretical_loss": 3.952839783994776,
"tokens_seen": 457441280
},
{
"epoch": 0.14,
"learning_rate": 0.00043503510531594787,
"loss": 2.875,
"theoretical_loss": 3.9527778550621973,
"tokens_seen": 457506816
},
{
"epoch": 0.14,
"learning_rate": 0.00043502507522567705,
"loss": 2.9098,
"theoretical_loss": 3.9527159374835463,
"tokens_seen": 457572352
},
{
"epoch": 0.14,
"learning_rate": 0.00043501504513540623,
"loss": 2.9462,
"theoretical_loss": 3.952654031255115,
"tokens_seen": 457637888
},
{
"epoch": 0.14,
"learning_rate": 0.0004350050150451354,
"loss": 3.0357,
"theoretical_loss": 3.9525921363731977,
"tokens_seen": 457703424
},
{
"epoch": 0.14,
"learning_rate": 0.0004349949849548646,
"loss": 2.8875,
"theoretical_loss": 3.9525302528340918,
"tokens_seen": 457768960
},
{
"epoch": 0.14,
"learning_rate": 0.0004349849548645938,
"loss": 2.8557,
"theoretical_loss": 3.9524683806340937,
"tokens_seen": 457834496
},
{
"epoch": 0.14,
"learning_rate": 0.000434974924774323,
"loss": 2.9504,
"theoretical_loss": 3.9524065197695046,
"tokens_seen": 457900032
},
{
"epoch": 0.14,
"learning_rate": 0.00043496489468405214,
"loss": 3.0495,
"theoretical_loss": 3.9523446702366245,
"tokens_seen": 457965568
},
{
"epoch": 0.14,
"learning_rate": 0.00043495486459378137,
"loss": 2.9949,
"theoretical_loss": 3.9522828320317576,
"tokens_seen": 458031104
},
{
"epoch": 0.14,
"learning_rate": 0.0004349448345035105,
"loss": 3.1083,
"theoretical_loss": 3.9522210051512077,
"tokens_seen": 458096640
},
{
"epoch": 0.14,
"learning_rate": 0.00043493480441323973,
"loss": 2.9009,
"theoretical_loss": 3.9521591895912818,
"tokens_seen": 458162176
},
{
"epoch": 0.14,
"learning_rate": 0.0004349247743229689,
"loss": 2.9672,
"theoretical_loss": 3.9520973853482886,
"tokens_seen": 458227712
},
{
"epoch": 0.14,
"learning_rate": 0.0004349147442326981,
"loss": 2.7928,
"theoretical_loss": 3.9520355924185377,
"tokens_seen": 458293248
},
{
"epoch": 0.14,
"learning_rate": 0.0004349047141424273,
"loss": 2.789,
"theoretical_loss": 3.951973810798341,
"tokens_seen": 458358784
},
{
"epoch": 0.14,
"learning_rate": 0.0004348946840521565,
"loss": 2.956,
"theoretical_loss": 3.951912040484012,
"tokens_seen": 458424320
},
{
"epoch": 0.14,
"learning_rate": 0.00043488465396188564,
"loss": 2.9218,
"theoretical_loss": 3.9518502814718652,
"tokens_seen": 458489856
},
{
"epoch": 0.14,
"learning_rate": 0.0004348746238716149,
"loss": 3.0027,
"theoretical_loss": 3.9517885337582186,
"tokens_seen": 458555392
},
{
"epoch": 0.14,
"learning_rate": 0.000434864593781344,
"loss": 2.722,
"theoretical_loss": 3.951726797339391,
"tokens_seen": 458620928
},
{
"epoch": 0.14,
"learning_rate": 0.00043485456369107324,
"loss": 2.9446,
"theoretical_loss": 3.9516650722117017,
"tokens_seen": 458686464
},
{
"epoch": 0.14,
"objective/train/docs_used": 271952,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9281325340270996,
"objective/train/theoretical_loss": 3.9516033583714734,
"objective/train/tokens_used": 479212000,
"theoretical_loss": 3.9516033583714734,
"tokens_seen": 458752000
},
{
"epoch": 0.14,
"learning_rate": 0.0004348445336008024,
"loss": 2.9981,
"theoretical_loss": 3.9516033583714734,
"tokens_seen": 458752000
},
{
"epoch": 0.14,
"learning_rate": 0.0004348345035105316,
"loss": 2.7936,
"theoretical_loss": 3.9515416558150305,
"tokens_seen": 458817536
},
{
"epoch": 0.14,
"learning_rate": 0.0004348244734202608,
"loss": 2.8275,
"theoretical_loss": 3.9514799645386973,
"tokens_seen": 458883072
},
{
"epoch": 0.14,
"learning_rate": 0.00043481444332998996,
"loss": 2.9014,
"theoretical_loss": 3.9514182845388026,
"tokens_seen": 458948608
},
{
"epoch": 0.14,
"learning_rate": 0.00043480441323971914,
"loss": 2.8708,
"theoretical_loss": 3.951356615811674,
"tokens_seen": 459014144
},
{
"epoch": 0.14,
"learning_rate": 0.0004347943831494484,
"loss": 3.1129,
"theoretical_loss": 3.9512949583536434,
"tokens_seen": 459079680
},
{
"epoch": 0.14,
"learning_rate": 0.0004347843530591775,
"loss": 2.603,
"theoretical_loss": 3.951233312161042,
"tokens_seen": 459145216
},
{
"epoch": 0.14,
"learning_rate": 0.00043477432296890674,
"loss": 2.888,
"theoretical_loss": 3.9511716772302057,
"tokens_seen": 459210752
},
{
"epoch": 0.14,
"learning_rate": 0.00043476429287863587,
"loss": 2.6878,
"theoretical_loss": 3.9511100535574686,
"tokens_seen": 459276288
},
{
"epoch": 0.14,
"learning_rate": 0.0004347542627883651,
"loss": 2.896,
"theoretical_loss": 3.9510484411391698,
"tokens_seen": 459341824
},
{
"epoch": 0.14,
"learning_rate": 0.0004347442326980943,
"loss": 2.7514,
"theoretical_loss": 3.9509868399716472,
"tokens_seen": 459407360
},
{
"epoch": 0.14,
"learning_rate": 0.00043473420260782346,
"loss": 2.6764,
"theoretical_loss": 3.9509252500512426,
"tokens_seen": 459472896
},
{
"epoch": 0.14,
"learning_rate": 0.00043472417251755265,
"loss": 2.9212,
"theoretical_loss": 3.950863671374299,
"tokens_seen": 459538432
},
{
"epoch": 0.14,
"learning_rate": 0.0004347141424272819,
"loss": 3.0134,
"theoretical_loss": 3.9508021039371597,
"tokens_seen": 459603968
},
{
"epoch": 0.14,
"learning_rate": 0.000434704112337011,
"loss": 2.9894,
"theoretical_loss": 3.9507405477361726,
"tokens_seen": 459669504
},
{
"epoch": 0.14,
"learning_rate": 0.00043469408224674024,
"loss": 3.1493,
"theoretical_loss": 3.9506790027676844,
"tokens_seen": 459735040
},
{
"epoch": 0.14,
"learning_rate": 0.0004346840521564694,
"loss": 2.9438,
"theoretical_loss": 3.9506174690280442,
"tokens_seen": 459800576
},
{
"epoch": 0.14,
"learning_rate": 0.0004346740220661986,
"loss": 3.0193,
"theoretical_loss": 3.950555946513605,
"tokens_seen": 459866112
},
{
"epoch": 0.14,
"learning_rate": 0.00043466399197592784,
"loss": 3.1924,
"theoretical_loss": 3.9504944352207176,
"tokens_seen": 459931648
},
{
"epoch": 0.14,
"learning_rate": 0.00043465396188565697,
"loss": 2.9121,
"theoretical_loss": 3.9504329351457383,
"tokens_seen": 459997184
},
{
"epoch": 0.14,
"learning_rate": 0.0004346439317953862,
"loss": 2.8726,
"theoretical_loss": 3.9503714462850237,
"tokens_seen": 460062720
},
{
"epoch": 0.14,
"learning_rate": 0.00043463390170511533,
"loss": 2.7697,
"theoretical_loss": 3.9503099686349303,
"tokens_seen": 460128256
},
{
"epoch": 0.14,
"learning_rate": 0.00043462387161484457,
"loss": 2.9298,
"theoretical_loss": 3.950248502191819,
"tokens_seen": 460193792
},
{
"epoch": 0.14,
"learning_rate": 0.00043461384152457375,
"loss": 3.136,
"theoretical_loss": 3.950187046952051,
"tokens_seen": 460259328
},
{
"epoch": 0.14,
"learning_rate": 0.00043460381143430293,
"loss": 2.7644,
"theoretical_loss": 3.9501256029119896,
"tokens_seen": 460324864
},
{
"epoch": 0.14,
"objective/train/docs_used": 272724,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.293632984161377,
"objective/train/theoretical_loss": 3.950064170068,
"objective/train/tokens_used": 480850400,
"theoretical_loss": 3.950064170068,
"tokens_seen": 460390400
},
{
"epoch": 0.14,
"learning_rate": 0.0004345937813440321,
"loss": 2.7199,
"theoretical_loss": 3.950064170068,
"tokens_seen": 460390400
},
{
"epoch": 0.14,
"learning_rate": 0.0004345837512537613,
"loss": 3.0363,
"theoretical_loss": 3.9500027484164484,
"tokens_seen": 460455936
},
{
"epoch": 0.14,
"learning_rate": 0.00043457372116349047,
"loss": 2.9059,
"theoretical_loss": 3.9499413379537027,
"tokens_seen": 460521472
},
{
"epoch": 0.14,
"learning_rate": 0.0004345636910732197,
"loss": 2.9963,
"theoretical_loss": 3.9498799386761334,
"tokens_seen": 460587008
},
{
"epoch": 0.14,
"learning_rate": 0.00043455366098294883,
"loss": 2.8352,
"theoretical_loss": 3.949818550580112,
"tokens_seen": 460652544
},
{
"epoch": 0.14,
"learning_rate": 0.00043454363089267807,
"loss": 3.0915,
"theoretical_loss": 3.9497571736620127,
"tokens_seen": 460718080
},
{
"epoch": 0.14,
"learning_rate": 0.00043453360080240725,
"loss": 2.9781,
"theoretical_loss": 3.949695807918209,
"tokens_seen": 460783616
},
{
"epoch": 0.14,
"learning_rate": 0.00043452357071213643,
"loss": 3.1897,
"theoretical_loss": 3.9496344533450793,
"tokens_seen": 460849152
},
{
"epoch": 0.14,
"learning_rate": 0.0004345135406218656,
"loss": 2.7796,
"theoretical_loss": 3.949573109939001,
"tokens_seen": 460914688
},
{
"epoch": 0.14,
"learning_rate": 0.0004345035105315948,
"loss": 2.9629,
"theoretical_loss": 3.9495117776963538,
"tokens_seen": 460980224
},
{
"epoch": 0.14,
"learning_rate": 0.000434493480441324,
"loss": 2.961,
"theoretical_loss": 3.949450456613521,
"tokens_seen": 461045760
},
{
"epoch": 0.14,
"learning_rate": 0.0004344834503510532,
"loss": 2.9782,
"theoretical_loss": 3.9493891466868845,
"tokens_seen": 461111296
},
{
"epoch": 0.14,
"learning_rate": 0.00043447342026078234,
"loss": 3.1149,
"theoretical_loss": 3.949327847912831,
"tokens_seen": 461176832
},
{
"epoch": 0.14,
"learning_rate": 0.00043446339017051157,
"loss": 2.9404,
"theoretical_loss": 3.9492665602877457,
"tokens_seen": 461242368
},
{
"epoch": 0.14,
"learning_rate": 0.0004344533600802407,
"loss": 2.8125,
"theoretical_loss": 3.9492052838080185,
"tokens_seen": 461307904
},
{
"epoch": 0.14,
"learning_rate": 0.00043444332998996993,
"loss": 2.6222,
"theoretical_loss": 3.9491440184700393,
"tokens_seen": 461373440
},
{
"epoch": 0.14,
"learning_rate": 0.0004344332998996991,
"loss": 2.9793,
"theoretical_loss": 3.9490827642701998,
"tokens_seen": 461438976
},
{
"epoch": 0.14,
"learning_rate": 0.0004344232698094283,
"loss": 2.938,
"theoretical_loss": 3.9490215212048936,
"tokens_seen": 461504512
},
{
"epoch": 0.14,
"learning_rate": 0.0004344132397191575,
"loss": 2.6944,
"theoretical_loss": 3.9489602892705165,
"tokens_seen": 461570048
},
{
"epoch": 0.14,
"learning_rate": 0.0004344032096288867,
"loss": 2.7612,
"theoretical_loss": 3.948899068463465,
"tokens_seen": 461635584
},
{
"epoch": 0.14,
"learning_rate": 0.00043439317953861584,
"loss": 2.8936,
"theoretical_loss": 3.948837858780138,
"tokens_seen": 461701120
},
{
"epoch": 0.14,
"learning_rate": 0.0004343831494483451,
"loss": 2.9781,
"theoretical_loss": 3.948776660216935,
"tokens_seen": 461766656
},
{
"epoch": 0.14,
"learning_rate": 0.0004343731193580742,
"loss": 3.1315,
"theoretical_loss": 3.948715472770259,
"tokens_seen": 461832192
},
{
"epoch": 0.14,
"learning_rate": 0.00043436308926780344,
"loss": 2.9559,
"theoretical_loss": 3.948654296436513,
"tokens_seen": 461897728
},
{
"epoch": 0.14,
"learning_rate": 0.0004343530591775326,
"loss": 3.0797,
"theoretical_loss": 3.9485931312121028,
"tokens_seen": 461963264
},
{
"epoch": 0.14,
"objective/train/docs_used": 273842,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.113386869430542,
"objective/train/theoretical_loss": 3.9485319770934355,
"objective/train/tokens_used": 482488800,
"theoretical_loss": 3.9485319770934355,
"tokens_seen": 462028800
},
{
"epoch": 0.14,
"learning_rate": 0.0004343430290872618,
"loss": 2.9541,
"theoretical_loss": 3.9485319770934355,
"tokens_seen": 462028800
},
{
"epoch": 0.14,
"learning_rate": 0.000434332998996991,
"loss": 3.0204,
"theoretical_loss": 3.948470834076919,
"tokens_seen": 462094336
},
{
"epoch": 0.14,
"learning_rate": 0.00043432296890672016,
"loss": 2.8271,
"theoretical_loss": 3.9484097021589646,
"tokens_seen": 462159872
},
{
"epoch": 0.14,
"learning_rate": 0.00043431293881644934,
"loss": 3.0506,
"theoretical_loss": 3.948348581335984,
"tokens_seen": 462225408
},
{
"epoch": 0.14,
"learning_rate": 0.0004343029087261786,
"loss": 2.939,
"theoretical_loss": 3.94828747160439,
"tokens_seen": 462290944
},
{
"epoch": 0.14,
"learning_rate": 0.0004342928786359077,
"loss": 2.7207,
"theoretical_loss": 3.9482263729605993,
"tokens_seen": 462356480
},
{
"epoch": 0.14,
"learning_rate": 0.00043428284854563694,
"loss": 2.8159,
"theoretical_loss": 3.9481652854010285,
"tokens_seen": 462422016
},
{
"epoch": 0.14,
"learning_rate": 0.00043427281845536607,
"loss": 2.9092,
"theoretical_loss": 3.948104208922096,
"tokens_seen": 462487552
},
{
"epoch": 0.14,
"learning_rate": 0.0004342627883650953,
"loss": 2.7242,
"theoretical_loss": 3.9480431435202226,
"tokens_seen": 462553088
},
{
"epoch": 0.14,
"learning_rate": 0.0004342527582748245,
"loss": 2.8413,
"theoretical_loss": 3.947982089191829,
"tokens_seen": 462618624
},
{
"epoch": 0.14,
"learning_rate": 0.00043424272818455367,
"loss": 2.6571,
"theoretical_loss": 3.9479210459333407,
"tokens_seen": 462684160
},
{
"epoch": 0.14,
"learning_rate": 0.00043423269809428285,
"loss": 2.7971,
"theoretical_loss": 3.9478600137411823,
"tokens_seen": 462749696
},
{
"epoch": 0.14,
"learning_rate": 0.0004342226680040121,
"loss": 2.6765,
"theoretical_loss": 3.94779899261178,
"tokens_seen": 462815232
},
{
"epoch": 0.14,
"learning_rate": 0.0004342126379137412,
"loss": 2.9225,
"theoretical_loss": 3.9477379825415637,
"tokens_seen": 462880768
},
{
"epoch": 0.14,
"learning_rate": 0.00043420260782347044,
"loss": 3.0055,
"theoretical_loss": 3.9476769835269634,
"tokens_seen": 462946304
},
{
"epoch": 0.14,
"learning_rate": 0.00043419257773319957,
"loss": 2.8627,
"theoretical_loss": 3.9476159955644112,
"tokens_seen": 463011840
},
{
"epoch": 0.14,
"learning_rate": 0.0004341825476429288,
"loss": 2.9815,
"theoretical_loss": 3.94755501865034,
"tokens_seen": 463077376
},
{
"epoch": 0.14,
"learning_rate": 0.000434172517552658,
"loss": 2.7327,
"theoretical_loss": 3.947494052781185,
"tokens_seen": 463142912
},
{
"epoch": 0.14,
"learning_rate": 0.00043416248746238717,
"loss": 2.8695,
"theoretical_loss": 3.9474330979533843,
"tokens_seen": 463208448
},
{
"epoch": 0.14,
"learning_rate": 0.00043415245737211635,
"loss": 3.2073,
"theoretical_loss": 3.947372154163375,
"tokens_seen": 463273984
},
{
"epoch": 0.14,
"learning_rate": 0.00043414242728184553,
"loss": 2.9631,
"theoretical_loss": 3.947311221407599,
"tokens_seen": 463339520
},
{
"epoch": 0.14,
"learning_rate": 0.0004341323971915747,
"loss": 2.9367,
"theoretical_loss": 3.947250299682498,
"tokens_seen": 463405056
},
{
"epoch": 0.14,
"learning_rate": 0.00043412236710130395,
"loss": 2.9153,
"theoretical_loss": 3.9471893889845138,
"tokens_seen": 463470592
},
{
"epoch": 0.14,
"learning_rate": 0.0004341123370110331,
"loss": 2.9749,
"theoretical_loss": 3.947128489310093,
"tokens_seen": 463536128
},
{
"epoch": 0.14,
"learning_rate": 0.0004341023069207623,
"loss": 2.8678,
"theoretical_loss": 3.947067600655682,
"tokens_seen": 463601664
},
{
"epoch": 0.14,
"objective/train/docs_used": 274430,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4908626079559326,
"objective/train/theoretical_loss": 3.94700672301773,
"objective/train/tokens_used": 484127200,
"theoretical_loss": 3.94700672301773,
"tokens_seen": 463667200
},
{
"epoch": 0.14,
"learning_rate": 0.00043409227683049144,
"loss": 2.5694,
"theoretical_loss": 3.94700672301773,
"tokens_seen": 463667200
},
{
"epoch": 0.14,
"learning_rate": 0.00043408224674022067,
"loss": 2.717,
"theoretical_loss": 3.946945856392687,
"tokens_seen": 463732736
},
{
"epoch": 0.14,
"learning_rate": 0.00043407221664994985,
"loss": 2.6902,
"theoretical_loss": 3.9468850007770033,
"tokens_seen": 463798272
},
{
"epoch": 0.14,
"learning_rate": 0.00043406218655967903,
"loss": 2.786,
"theoretical_loss": 3.9468241561671342,
"tokens_seen": 463863808
},
{
"epoch": 0.14,
"learning_rate": 0.0004340521564694082,
"loss": 2.4293,
"theoretical_loss": 3.946763322559534,
"tokens_seen": 463929344
},
{
"epoch": 0.14,
"learning_rate": 0.00043404212637913745,
"loss": 2.6931,
"theoretical_loss": 3.946702499950659,
"tokens_seen": 463994880
},
{
"epoch": 0.14,
"learning_rate": 0.0004340320962888666,
"loss": 3.0037,
"theoretical_loss": 3.9466416883369684,
"tokens_seen": 464060416
},
{
"epoch": 0.14,
"learning_rate": 0.0004340220661985958,
"loss": 2.9052,
"theoretical_loss": 3.946580887714921,
"tokens_seen": 464125952
},
{
"epoch": 0.14,
"learning_rate": 0.00043401203610832494,
"loss": 2.8656,
"theoretical_loss": 3.9465200980809803,
"tokens_seen": 464191488
},
{
"epoch": 0.14,
"learning_rate": 0.0004340020060180542,
"loss": 3.08,
"theoretical_loss": 3.946459319431608,
"tokens_seen": 464257024
},
{
"epoch": 0.14,
"learning_rate": 0.00043399197592778336,
"loss": 2.8027,
"theoretical_loss": 3.9463985517632696,
"tokens_seen": 464322560
},
{
"epoch": 0.14,
"learning_rate": 0.00043398194583751254,
"loss": 2.8701,
"theoretical_loss": 3.9463377950724317,
"tokens_seen": 464388096
},
{
"epoch": 0.14,
"learning_rate": 0.0004339719157472417,
"loss": 2.9089,
"theoretical_loss": 3.9462770493555617,
"tokens_seen": 464453632
},
{
"epoch": 0.14,
"learning_rate": 0.0004339618856569709,
"loss": 2.8239,
"theoretical_loss": 3.9462163146091305,
"tokens_seen": 464519168
},
{
"epoch": 0.14,
"learning_rate": 0.0004339518555667001,
"loss": 2.8597,
"theoretical_loss": 3.9461555908296093,
"tokens_seen": 464584704
},
{
"epoch": 0.14,
"learning_rate": 0.0004339418254764293,
"loss": 2.8746,
"theoretical_loss": 3.9460948780134713,
"tokens_seen": 464650240
},
{
"epoch": 0.14,
"learning_rate": 0.0004339317953861585,
"loss": 2.9924,
"theoretical_loss": 3.94603417615719,
"tokens_seen": 464715776
},
{
"epoch": 0.14,
"learning_rate": 0.0004339217652958877,
"loss": 2.8558,
"theoretical_loss": 3.9459734852572437,
"tokens_seen": 464781312
},
{
"epoch": 0.14,
"learning_rate": 0.0004339117352056169,
"loss": 3.0925,
"theoretical_loss": 3.9459128053101082,
"tokens_seen": 464846848
},
{
"epoch": 0.14,
"learning_rate": 0.00043390170511534604,
"loss": 2.9914,
"theoretical_loss": 3.945852136312265,
"tokens_seen": 464912384
},
{
"epoch": 0.14,
"learning_rate": 0.0004338916750250753,
"loss": 3.0237,
"theoretical_loss": 3.945791478260195,
"tokens_seen": 464977920
},
{
"epoch": 0.14,
"learning_rate": 0.0004338816449348044,
"loss": 2.5918,
"theoretical_loss": 3.94573083115038,
"tokens_seen": 465043456
},
{
"epoch": 0.14,
"learning_rate": 0.00043387161484453364,
"loss": 2.6599,
"theoretical_loss": 3.9456701949793054,
"tokens_seen": 465108992
},
{
"epoch": 0.14,
"learning_rate": 0.0004338615847542628,
"loss": 3.0173,
"theoretical_loss": 3.9456095697434566,
"tokens_seen": 465174528
},
{
"epoch": 0.14,
"learning_rate": 0.000433851554663992,
"loss": 2.9841,
"theoretical_loss": 3.945548955439322,
"tokens_seen": 465240064
},
{
"epoch": 0.14,
"objective/train/docs_used": 275809,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.787060260772705,
"objective/train/theoretical_loss": 3.945488352063391,
"objective/train/tokens_used": 485765600,
"theoretical_loss": 3.945488352063391,
"tokens_seen": 465305600
},
{
"epoch": 0.14,
"learning_rate": 0.0004338415245737212,
"loss": 2.9794,
"theoretical_loss": 3.945488352063391,
"tokens_seen": 465305600
},
{
"epoch": 0.14,
"learning_rate": 0.00043383149448345036,
"loss": 2.8995,
"theoretical_loss": 3.945427759612154,
"tokens_seen": 465371136
},
{
"epoch": 0.14,
"learning_rate": 0.00043382146439317954,
"loss": 2.6856,
"theoretical_loss": 3.945367178082104,
"tokens_seen": 465436672
},
{
"epoch": 0.14,
"learning_rate": 0.0004338114343029088,
"loss": 2.8025,
"theoretical_loss": 3.945306607469735,
"tokens_seen": 465502208
},
{
"epoch": 0.14,
"learning_rate": 0.0004338014042126379,
"loss": 2.8469,
"theoretical_loss": 3.945246047771543,
"tokens_seen": 465567744
},
{
"epoch": 0.14,
"learning_rate": 0.00043379137412236714,
"loss": 3.0447,
"theoretical_loss": 3.9451854989840256,
"tokens_seen": 465633280
},
{
"epoch": 0.14,
"learning_rate": 0.00043378134403209627,
"loss": 2.7582,
"theoretical_loss": 3.945124961103682,
"tokens_seen": 465698816
},
{
"epoch": 0.14,
"learning_rate": 0.0004337713139418255,
"loss": 2.9184,
"theoretical_loss": 3.9450644341270116,
"tokens_seen": 465764352
},
{
"epoch": 0.14,
"learning_rate": 0.0004337612838515547,
"loss": 2.9787,
"theoretical_loss": 3.945003918050518,
"tokens_seen": 465829888
},
{
"epoch": 0.14,
"learning_rate": 0.00043375125376128387,
"loss": 2.8831,
"theoretical_loss": 3.9449434128707055,
"tokens_seen": 465895424
},
{
"epoch": 0.14,
"learning_rate": 0.00043374122367101305,
"loss": 2.9134,
"theoretical_loss": 3.9448829185840784,
"tokens_seen": 465960960
},
{
"epoch": 0.14,
"learning_rate": 0.0004337311935807423,
"loss": 2.7942,
"theoretical_loss": 3.9448224351871444,
"tokens_seen": 466026496
},
{
"epoch": 0.14,
"learning_rate": 0.0004337211634904714,
"loss": 3.0798,
"theoretical_loss": 3.944761962676412,
"tokens_seen": 466092032
},
{
"epoch": 0.14,
"learning_rate": 0.00043371113340020064,
"loss": 3.0969,
"theoretical_loss": 3.9447015010483923,
"tokens_seen": 466157568
},
{
"epoch": 0.14,
"learning_rate": 0.00043370110330992977,
"loss": 2.7817,
"theoretical_loss": 3.9446410502995963,
"tokens_seen": 466223104
},
{
"epoch": 0.14,
"learning_rate": 0.000433691073219659,
"loss": 2.8557,
"theoretical_loss": 3.9445806104265384,
"tokens_seen": 466288640
},
{
"epoch": 0.14,
"learning_rate": 0.0004336810431293882,
"loss": 2.8694,
"theoretical_loss": 3.9445201814257334,
"tokens_seen": 466354176
},
{
"epoch": 0.14,
"learning_rate": 0.00043367101303911737,
"loss": 2.973,
"theoretical_loss": 3.9444597632936977,
"tokens_seen": 466419712
},
{
"epoch": 0.14,
"learning_rate": 0.00043366098294884655,
"loss": 2.6732,
"theoretical_loss": 3.9443993560269504,
"tokens_seen": 466485248
},
{
"epoch": 0.14,
"learning_rate": 0.00043365095285857573,
"loss": 2.9185,
"theoretical_loss": 3.9443389596220113,
"tokens_seen": 466550784
},
{
"epoch": 0.14,
"learning_rate": 0.0004336409227683049,
"loss": 2.7752,
"theoretical_loss": 3.944278574075402,
"tokens_seen": 466616320
},
{
"epoch": 0.14,
"learning_rate": 0.00043363089267803415,
"loss": 2.8808,
"theoretical_loss": 3.944218199383645,
"tokens_seen": 466681856
},
{
"epoch": 0.14,
"learning_rate": 0.0004336208625877633,
"loss": 2.8209,
"theoretical_loss": 3.9441578355432663,
"tokens_seen": 466747392
},
{
"epoch": 0.14,
"learning_rate": 0.0004336108324974925,
"loss": 2.7177,
"theoretical_loss": 3.9440974825507915,
"tokens_seen": 466812928
},
{
"epoch": 0.14,
"learning_rate": 0.00043360080240722164,
"loss": 2.7596,
"theoretical_loss": 3.9440371404027488,
"tokens_seen": 466878464
},
{
"epoch": 0.14,
"objective/train/docs_used": 276320,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.076183795928955,
"objective/train/theoretical_loss": 3.943976809095668,
"objective/train/tokens_used": 487404000,
"theoretical_loss": 3.943976809095668,
"tokens_seen": 466944000
},
{
"epoch": 0.14,
"learning_rate": 0.00043359077231695087,
"loss": 3.1843,
"theoretical_loss": 3.943976809095668,
"tokens_seen": 466944000
},
{
"epoch": 0.14,
"learning_rate": 0.00043358074222668005,
"loss": 2.9046,
"theoretical_loss": 3.94391648862608,
"tokens_seen": 467009536
},
{
"epoch": 0.14,
"learning_rate": 0.00043357071213640923,
"loss": 2.9376,
"theoretical_loss": 3.9438561789905178,
"tokens_seen": 467075072
},
{
"epoch": 0.14,
"learning_rate": 0.0004335606820461384,
"loss": 2.887,
"theoretical_loss": 3.9437958801855157,
"tokens_seen": 467140608
},
{
"epoch": 0.14,
"learning_rate": 0.00043355065195586765,
"loss": 2.7006,
"theoretical_loss": 3.9437355922076094,
"tokens_seen": 467206144
},
{
"epoch": 0.14,
"learning_rate": 0.0004335406218655968,
"loss": 2.6553,
"theoretical_loss": 3.9436753150533375,
"tokens_seen": 467271680
},
{
"epoch": 0.14,
"learning_rate": 0.000433530591775326,
"loss": 2.8684,
"theoretical_loss": 3.943615048719238,
"tokens_seen": 467337216
},
{
"epoch": 0.14,
"learning_rate": 0.00043352056168505514,
"loss": 2.8836,
"theoretical_loss": 3.9435547932018524,
"tokens_seen": 467402752
},
{
"epoch": 0.14,
"learning_rate": 0.0004335105315947844,
"loss": 2.9624,
"theoretical_loss": 3.943494548497723,
"tokens_seen": 467468288
},
{
"epoch": 0.14,
"learning_rate": 0.00043350050150451356,
"loss": 2.6615,
"theoretical_loss": 3.943434314603394,
"tokens_seen": 467533824
},
{
"epoch": 0.14,
"learning_rate": 0.00043349047141424274,
"loss": 3.0893,
"theoretical_loss": 3.9433740915154094,
"tokens_seen": 467599360
},
{
"epoch": 0.14,
"learning_rate": 0.0004334804413239719,
"loss": 3.1646,
"theoretical_loss": 3.9433138792303177,
"tokens_seen": 467664896
},
{
"epoch": 0.14,
"learning_rate": 0.0004334704112337011,
"loss": 3.1173,
"theoretical_loss": 3.943253677744668,
"tokens_seen": 467730432
},
{
"epoch": 0.14,
"learning_rate": 0.0004334603811434303,
"loss": 2.7868,
"theoretical_loss": 3.943193487055009,
"tokens_seen": 467795968
},
{
"epoch": 0.14,
"learning_rate": 0.0004334503510531595,
"loss": 2.7069,
"theoretical_loss": 3.9431333071578942,
"tokens_seen": 467861504
},
{
"epoch": 0.14,
"learning_rate": 0.00043344032096288864,
"loss": 2.8466,
"theoretical_loss": 3.9430731380498765,
"tokens_seen": 467927040
},
{
"epoch": 0.14,
"learning_rate": 0.0004334302908726179,
"loss": 2.9696,
"theoretical_loss": 3.94301297972751,
"tokens_seen": 467992576
},
{
"epoch": 0.14,
"learning_rate": 0.000433420260782347,
"loss": 2.6652,
"theoretical_loss": 3.9429528321873537,
"tokens_seen": 468058112
},
{
"epoch": 0.14,
"learning_rate": 0.00043341023069207624,
"loss": 2.9861,
"theoretical_loss": 3.942892695425963,
"tokens_seen": 468123648
},
{
"epoch": 0.14,
"learning_rate": 0.0004334002006018054,
"loss": 2.9029,
"theoretical_loss": 3.9428325694398993,
"tokens_seen": 468189184
},
{
"epoch": 0.14,
"learning_rate": 0.0004333901705115346,
"loss": 2.8075,
"theoretical_loss": 3.942772454225724,
"tokens_seen": 468254720
},
{
"epoch": 0.14,
"learning_rate": 0.0004333801404212638,
"loss": 2.8315,
"theoretical_loss": 3.94271234978,
"tokens_seen": 468320256
},
{
"epoch": 0.14,
"learning_rate": 0.000433370110330993,
"loss": 3.0407,
"theoretical_loss": 3.9426522560992905,
"tokens_seen": 468385792
},
{
"epoch": 0.14,
"learning_rate": 0.00043336008024072215,
"loss": 2.7421,
"theoretical_loss": 3.942592173180163,
"tokens_seen": 468451328
},
{
"epoch": 0.14,
"learning_rate": 0.0004333500501504514,
"loss": 2.9915,
"theoretical_loss": 3.9425321010191854,
"tokens_seen": 468516864
},
{
"epoch": 0.14,
"objective/train/docs_used": 277290,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1820871829986572,
"objective/train/theoretical_loss": 3.942472039612926,
"objective/train/tokens_used": 489042400,
"theoretical_loss": 3.942472039612926,
"tokens_seen": 468582400
},
{
"epoch": 0.14,
"learning_rate": 0.0004333400200601805,
"loss": 2.9808,
"theoretical_loss": 3.942472039612926,
"tokens_seen": 468582400
},
{
"epoch": 0.14,
"learning_rate": 0.00043332998996990974,
"loss": 3.1072,
"theoretical_loss": 3.9424119889579563,
"tokens_seen": 468647936
},
{
"epoch": 0.14,
"learning_rate": 0.0004333199598796389,
"loss": 2.8486,
"theoretical_loss": 3.9423519490508476,
"tokens_seen": 468713472
},
{
"epoch": 0.14,
"learning_rate": 0.0004333099297893681,
"loss": 2.8733,
"theoretical_loss": 3.9422919198881754,
"tokens_seen": 468779008
},
{
"epoch": 0.14,
"learning_rate": 0.0004332998996990973,
"loss": 2.9926,
"theoretical_loss": 3.9422319014665144,
"tokens_seen": 468844544
},
{
"epoch": 0.14,
"learning_rate": 0.00043328986960882647,
"loss": 2.9607,
"theoretical_loss": 3.942171893782442,
"tokens_seen": 468910080
},
{
"epoch": 0.14,
"learning_rate": 0.00043327983951855565,
"loss": 2.7794,
"theoretical_loss": 3.942111896832537,
"tokens_seen": 468975616
},
{
"epoch": 0.14,
"learning_rate": 0.0004332698094282849,
"loss": 2.9306,
"theoretical_loss": 3.9420519106133787,
"tokens_seen": 469041152
},
{
"epoch": 0.14,
"learning_rate": 0.000433259779338014,
"loss": 3.2266,
"theoretical_loss": 3.9419919351215498,
"tokens_seen": 469106688
},
{
"epoch": 0.14,
"learning_rate": 0.00043324974924774325,
"loss": 2.9204,
"theoretical_loss": 3.9419319703536337,
"tokens_seen": 469172224
},
{
"epoch": 0.14,
"learning_rate": 0.00043323971915747243,
"loss": 2.9933,
"theoretical_loss": 3.941872016306215,
"tokens_seen": 469237760
},
{
"epoch": 0.14,
"learning_rate": 0.0004332296890672016,
"loss": 2.8745,
"theoretical_loss": 3.941812072975881,
"tokens_seen": 469303296
},
{
"epoch": 0.14,
"learning_rate": 0.0004332196589769308,
"loss": 2.9306,
"theoretical_loss": 3.9417521403592186,
"tokens_seen": 469368832
},
{
"epoch": 0.14,
"learning_rate": 0.00043320962888665997,
"loss": 2.6249,
"theoretical_loss": 3.9416922184528183,
"tokens_seen": 469434368
},
{
"epoch": 0.14,
"learning_rate": 0.00043319959879638915,
"loss": 2.9339,
"theoretical_loss": 3.94163230725327,
"tokens_seen": 469499904
},
{
"epoch": 0.14,
"learning_rate": 0.0004331895687061184,
"loss": 2.9625,
"theoretical_loss": 3.9415724067571687,
"tokens_seen": 469565440
},
{
"epoch": 0.14,
"learning_rate": 0.00043317953861584757,
"loss": 3.0605,
"theoretical_loss": 3.941512516961107,
"tokens_seen": 469630976
},
{
"epoch": 0.14,
"learning_rate": 0.00043316950852557675,
"loss": 2.9603,
"theoretical_loss": 3.941452637861681,
"tokens_seen": 469696512
},
{
"epoch": 0.14,
"learning_rate": 0.00043315947843530593,
"loss": 2.817,
"theoretical_loss": 3.9413927694554887,
"tokens_seen": 469762048
},
{
"epoch": 0.14,
"learning_rate": 0.0004331494483450351,
"loss": 2.9923,
"theoretical_loss": 3.9413329117391287,
"tokens_seen": 469827584
},
{
"epoch": 0.14,
"learning_rate": 0.00043313941825476435,
"loss": 2.8578,
"theoretical_loss": 3.9412730647092022,
"tokens_seen": 469893120
},
{
"epoch": 0.14,
"learning_rate": 0.0004331293881644935,
"loss": 2.7678,
"theoretical_loss": 3.94121322836231,
"tokens_seen": 469958656
},
{
"epoch": 0.14,
"learning_rate": 0.0004331193580742227,
"loss": 2.7757,
"theoretical_loss": 3.9411534026950568,
"tokens_seen": 470024192
},
{
"epoch": 0.14,
"learning_rate": 0.00043310932798395184,
"loss": 2.8787,
"theoretical_loss": 3.9410935877040476,
"tokens_seen": 470089728
},
{
"epoch": 0.14,
"learning_rate": 0.00043309929789368107,
"loss": 2.9,
"theoretical_loss": 3.9410337833858886,
"tokens_seen": 470155264
},
{
"epoch": 0.14,
"objective/train/docs_used": 277851,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8162131309509277,
"objective/train/theoretical_loss": 3.9409739897371896,
"objective/train/tokens_used": 490680800,
"theoretical_loss": 3.9409739897371896,
"tokens_seen": 470220800
},
{
"epoch": 0.14,
"learning_rate": 0.00043308926780341025,
"loss": 3.1145,
"theoretical_loss": 3.9409739897371896,
"tokens_seen": 470220800
},
{
"epoch": 0.14,
"learning_rate": 0.00043307923771313943,
"loss": 2.9894,
"theoretical_loss": 3.9409142067545586,
"tokens_seen": 470286336
},
{
"epoch": 0.14,
"learning_rate": 0.0004330692076228686,
"loss": 3.0744,
"theoretical_loss": 3.940854434434608,
"tokens_seen": 470351872
},
{
"epoch": 0.14,
"learning_rate": 0.00043305917753259785,
"loss": 2.8549,
"theoretical_loss": 3.940794672773951,
"tokens_seen": 470417408
},
{
"epoch": 0.14,
"learning_rate": 0.000433049147442327,
"loss": 2.7925,
"theoretical_loss": 3.9407349217692023,
"tokens_seen": 470482944
},
{
"epoch": 0.14,
"learning_rate": 0.0004330391173520562,
"loss": 2.9453,
"theoretical_loss": 3.9406751814169763,
"tokens_seen": 470548480
},
{
"epoch": 0.14,
"learning_rate": 0.00043302908726178534,
"loss": 2.8738,
"theoretical_loss": 3.9406154517138923,
"tokens_seen": 470614016
},
{
"epoch": 0.14,
"learning_rate": 0.0004330190571715146,
"loss": 2.7708,
"theoretical_loss": 3.940555732656569,
"tokens_seen": 470679552
},
{
"epoch": 0.14,
"learning_rate": 0.00043300902708124376,
"loss": 3.0653,
"theoretical_loss": 3.940496024241627,
"tokens_seen": 470745088
},
{
"epoch": 0.14,
"learning_rate": 0.00043299899699097294,
"loss": 3.0183,
"theoretical_loss": 3.9404363264656883,
"tokens_seen": 470810624
},
{
"epoch": 0.14,
"learning_rate": 0.0004329889669007021,
"loss": 2.9802,
"theoretical_loss": 3.9403766393253767,
"tokens_seen": 470876160
},
{
"epoch": 0.14,
"learning_rate": 0.0004329789368104313,
"loss": 2.7904,
"theoretical_loss": 3.9403169628173185,
"tokens_seen": 470941696
},
{
"epoch": 0.14,
"learning_rate": 0.0004329689067201605,
"loss": 2.8852,
"theoretical_loss": 3.9402572969381393,
"tokens_seen": 471007232
},
{
"epoch": 0.14,
"learning_rate": 0.0004329588766298897,
"loss": 2.9922,
"theoretical_loss": 3.9401976416844677,
"tokens_seen": 471072768
},
{
"epoch": 0.14,
"learning_rate": 0.00043294884653961884,
"loss": 2.8197,
"theoretical_loss": 3.9401379970529344,
"tokens_seen": 471138304
},
{
"epoch": 0.14,
"learning_rate": 0.0004329388164493481,
"loss": 3.0398,
"theoretical_loss": 3.94007836304017,
"tokens_seen": 471203840
},
{
"epoch": 0.14,
"learning_rate": 0.0004329287863590772,
"loss": 3.0596,
"theoretical_loss": 3.9400187396428086,
"tokens_seen": 471269376
},
{
"epoch": 0.14,
"learning_rate": 0.00043291875626880644,
"loss": 3.0314,
"theoretical_loss": 3.9399591268574836,
"tokens_seen": 471334912
},
{
"epoch": 0.14,
"learning_rate": 0.0004329087261785356,
"loss": 3.0936,
"theoretical_loss": 3.9398995246808317,
"tokens_seen": 471400448
},
{
"epoch": 0.14,
"learning_rate": 0.0004328986960882648,
"loss": 2.6104,
"theoretical_loss": 3.93983993310949,
"tokens_seen": 471465984
},
{
"epoch": 0.14,
"learning_rate": 0.000432888665997994,
"loss": 2.8409,
"theoretical_loss": 3.9397803521400983,
"tokens_seen": 471531520
},
{
"epoch": 0.14,
"learning_rate": 0.0004328786359077232,
"loss": 3.0642,
"theoretical_loss": 3.9397207817692967,
"tokens_seen": 471597056
},
{
"epoch": 0.14,
"learning_rate": 0.00043286860581745235,
"loss": 2.7443,
"theoretical_loss": 3.9396612219937284,
"tokens_seen": 471662592
},
{
"epoch": 0.14,
"learning_rate": 0.0004328585757271816,
"loss": 2.738,
"theoretical_loss": 3.9396016728100363,
"tokens_seen": 471728128
},
{
"epoch": 0.14,
"learning_rate": 0.0004328485456369107,
"loss": 2.8538,
"theoretical_loss": 3.9395421342148653,
"tokens_seen": 471793664
},
{
"epoch": 0.14,
"objective/train/docs_used": 279270,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0141212940216064,
"objective/train/theoretical_loss": 3.939482606204863,
"objective/train/tokens_used": 492319200,
"theoretical_loss": 3.939482606204863,
"tokens_seen": 471859200
},
{
"epoch": 0.14,
"learning_rate": 0.00043283851554663994,
"loss": 2.8997,
"theoretical_loss": 3.939482606204863,
"tokens_seen": 471859200
},
{
"epoch": 0.14,
"learning_rate": 0.0004328284854563691,
"loss": 3.008,
"theoretical_loss": 3.9394230887766772,
"tokens_seen": 471924736
},
{
"epoch": 0.14,
"learning_rate": 0.0004328184553660983,
"loss": 3.0273,
"theoretical_loss": 3.939363581926959,
"tokens_seen": 471990272
},
{
"epoch": 0.14,
"learning_rate": 0.0004328084252758275,
"loss": 3.1478,
"theoretical_loss": 3.939304085652358,
"tokens_seen": 472055808
},
{
"epoch": 0.14,
"learning_rate": 0.00043279839518555667,
"loss": 3.0269,
"theoretical_loss": 3.939244599949528,
"tokens_seen": 472121344
},
{
"epoch": 0.14,
"learning_rate": 0.00043278836509528585,
"loss": 2.7709,
"theoretical_loss": 3.939185124815123,
"tokens_seen": 472186880
},
{
"epoch": 0.14,
"learning_rate": 0.0004327783350050151,
"loss": 2.7383,
"theoretical_loss": 3.9391256602457996,
"tokens_seen": 472252416
},
{
"epoch": 0.14,
"learning_rate": 0.0004327683049147442,
"loss": 2.7578,
"theoretical_loss": 3.9390662062382145,
"tokens_seen": 472317952
},
{
"epoch": 0.14,
"learning_rate": 0.00043275827482447345,
"loss": 2.7545,
"theoretical_loss": 3.9390067627890275,
"tokens_seen": 472383488
},
{
"epoch": 0.14,
"learning_rate": 0.00043274824473420263,
"loss": 2.9205,
"theoretical_loss": 3.938947329894899,
"tokens_seen": 472449024
},
{
"epoch": 0.14,
"learning_rate": 0.0004327382146439318,
"loss": 3.0668,
"theoretical_loss": 3.9388879075524907,
"tokens_seen": 472514560
},
{
"epoch": 0.14,
"learning_rate": 0.000432728184553661,
"loss": 2.7883,
"theoretical_loss": 3.9388284957584663,
"tokens_seen": 472580096
},
{
"epoch": 0.14,
"learning_rate": 0.00043271815446339017,
"loss": 2.7036,
"theoretical_loss": 3.9387690945094906,
"tokens_seen": 472645632
},
{
"epoch": 0.14,
"learning_rate": 0.00043270812437311935,
"loss": 3.0538,
"theoretical_loss": 3.93870970380223,
"tokens_seen": 472711168
},
{
"epoch": 0.14,
"learning_rate": 0.0004326980942828486,
"loss": 2.8628,
"theoretical_loss": 3.938650323633354,
"tokens_seen": 472776704
},
{
"epoch": 0.14,
"learning_rate": 0.0004326880641925777,
"loss": 2.9396,
"theoretical_loss": 3.9385909539995305,
"tokens_seen": 472842240
},
{
"epoch": 0.14,
"learning_rate": 0.00043267803410230695,
"loss": 3.0433,
"theoretical_loss": 3.9385315948974315,
"tokens_seen": 472907776
},
{
"epoch": 0.14,
"learning_rate": 0.0004326680040120361,
"loss": 2.772,
"theoretical_loss": 3.9384722463237294,
"tokens_seen": 472973312
},
{
"epoch": 0.14,
"learning_rate": 0.0004326579739217653,
"loss": 2.9285,
"theoretical_loss": 3.938412908275099,
"tokens_seen": 473038848
},
{
"epoch": 0.14,
"learning_rate": 0.0004326479438314945,
"loss": 2.8074,
"theoretical_loss": 3.938353580748215,
"tokens_seen": 473104384
},
{
"epoch": 0.14,
"learning_rate": 0.0004326379137412237,
"loss": 2.6641,
"theoretical_loss": 3.938294263739755,
"tokens_seen": 473169920
},
{
"epoch": 0.14,
"learning_rate": 0.00043262788365095286,
"loss": 2.8836,
"theoretical_loss": 3.9382349572463973,
"tokens_seen": 473235456
},
{
"epoch": 0.14,
"learning_rate": 0.00043261785356068204,
"loss": 3.0981,
"theoretical_loss": 3.938175661264823,
"tokens_seen": 473300992
},
{
"epoch": 0.14,
"learning_rate": 0.0004326078234704112,
"loss": 3.0504,
"theoretical_loss": 3.9381163757917133,
"tokens_seen": 473366528
},
{
"epoch": 0.14,
"learning_rate": 0.00043259779338014045,
"loss": 3.0151,
"theoretical_loss": 3.938057100823751,
"tokens_seen": 473432064
},
{
"epoch": 0.14,
"objective/train/docs_used": 279967,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.561666250228882,
"objective/train/theoretical_loss": 3.9379978363576216,
"objective/train/tokens_used": 493957600,
"theoretical_loss": 3.9379978363576216,
"tokens_seen": 473497600
},
{
"epoch": 0.14,
"learning_rate": 0.0004325877632898696,
"loss": 2.6292,
"theoretical_loss": 3.9379978363576216,
"tokens_seen": 473497600
},
{
"epoch": 0.14,
"learning_rate": 0.0004325777331995988,
"loss": 3.0368,
"theoretical_loss": 3.937938582390011,
"tokens_seen": 473563136
},
{
"epoch": 0.14,
"learning_rate": 0.000432567703109328,
"loss": 3.079,
"theoretical_loss": 3.9378793389176066,
"tokens_seen": 473628672
},
{
"epoch": 0.14,
"learning_rate": 0.0004325576730190572,
"loss": 2.7806,
"theoretical_loss": 3.9378201059370976,
"tokens_seen": 473694208
},
{
"epoch": 0.14,
"learning_rate": 0.00043254764292878636,
"loss": 2.6743,
"theoretical_loss": 3.9377608834451756,
"tokens_seen": 473759744
},
{
"epoch": 0.14,
"learning_rate": 0.00043253761283851554,
"loss": 3.1192,
"theoretical_loss": 3.9377016714385316,
"tokens_seen": 473825280
},
{
"epoch": 0.14,
"learning_rate": 0.0004325275827482447,
"loss": 2.9094,
"theoretical_loss": 3.9376424699138606,
"tokens_seen": 473890816
},
{
"epoch": 0.14,
"learning_rate": 0.00043251755265797396,
"loss": 2.7598,
"theoretical_loss": 3.937583278867857,
"tokens_seen": 473956352
},
{
"epoch": 0.14,
"learning_rate": 0.0004325075225677031,
"loss": 2.8362,
"theoretical_loss": 3.9375240982972173,
"tokens_seen": 474021888
},
{
"epoch": 0.14,
"learning_rate": 0.0004324974924774323,
"loss": 3.0722,
"theoretical_loss": 3.9374649281986405,
"tokens_seen": 474087424
},
{
"epoch": 0.14,
"learning_rate": 0.00043248746238716145,
"loss": 2.7437,
"theoretical_loss": 3.937405768568826,
"tokens_seen": 474152960
},
{
"epoch": 0.14,
"learning_rate": 0.0004324774322968907,
"loss": 2.9541,
"theoretical_loss": 3.937346619404475,
"tokens_seen": 474218496
},
{
"epoch": 0.14,
"learning_rate": 0.00043246740220661986,
"loss": 2.8422,
"theoretical_loss": 3.93728748070229,
"tokens_seen": 474284032
},
{
"epoch": 0.14,
"learning_rate": 0.00043245737211634904,
"loss": 2.7602,
"theoretical_loss": 3.937228352458976,
"tokens_seen": 474349568
},
{
"epoch": 0.14,
"learning_rate": 0.0004324473420260782,
"loss": 2.9715,
"theoretical_loss": 3.9371692346712375,
"tokens_seen": 474415104
},
{
"epoch": 0.14,
"learning_rate": 0.0004324373119358074,
"loss": 3.0922,
"theoretical_loss": 3.9371101273357825,
"tokens_seen": 474480640
},
{
"epoch": 0.14,
"learning_rate": 0.00043242728184553664,
"loss": 2.8695,
"theoretical_loss": 3.93705103044932,
"tokens_seen": 474546176
},
{
"epoch": 0.14,
"learning_rate": 0.0004324172517552658,
"loss": 2.9216,
"theoretical_loss": 3.9369919440085592,
"tokens_seen": 474611712
},
{
"epoch": 0.14,
"learning_rate": 0.000432407221664995,
"loss": 2.8716,
"theoretical_loss": 3.9369328680102122,
"tokens_seen": 474677248
},
{
"epoch": 0.14,
"learning_rate": 0.0004323971915747242,
"loss": 2.9434,
"theoretical_loss": 3.9368738024509926,
"tokens_seen": 474742784
},
{
"epoch": 0.14,
"learning_rate": 0.0004323871614844534,
"loss": 3.2621,
"theoretical_loss": 3.9368147473276145,
"tokens_seen": 474808320
},
{
"epoch": 0.14,
"learning_rate": 0.00043237713139418255,
"loss": 2.9925,
"theoretical_loss": 3.936755702636794,
"tokens_seen": 474873856
},
{
"epoch": 0.14,
"learning_rate": 0.0004323671013039118,
"loss": 2.744,
"theoretical_loss": 3.9366966683752493,
"tokens_seen": 474939392
},
{
"epoch": 0.14,
"learning_rate": 0.0004323570712136409,
"loss": 2.7103,
"theoretical_loss": 3.936637644539699,
"tokens_seen": 475004928
},
{
"epoch": 0.14,
"learning_rate": 0.00043234704112337014,
"loss": 2.7144,
"theoretical_loss": 3.9365786311268636,
"tokens_seen": 475070464
},
{
"epoch": 0.14,
"objective/train/docs_used": 280356,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8112435340881348,
"objective/train/theoretical_loss": 3.936519628133466,
"objective/train/tokens_used": 495596000,
"theoretical_loss": 3.936519628133466,
"tokens_seen": 475136000
},
{
"epoch": 0.14,
"learning_rate": 0.0004323370110330993,
"loss": 3.1572,
"theoretical_loss": 3.936519628133466,
"tokens_seen": 475136000
},
{
"epoch": 0.14,
"learning_rate": 0.0004323269809428285,
"loss": 2.7392,
"theoretical_loss": 3.9364606355562284,
"tokens_seen": 475201536
},
{
"epoch": 0.14,
"learning_rate": 0.0004323169508525577,
"loss": 2.6772,
"theoretical_loss": 3.9364016533918766,
"tokens_seen": 475267072
},
{
"epoch": 0.14,
"learning_rate": 0.00043230692076228687,
"loss": 3.0261,
"theoretical_loss": 3.9363426816371376,
"tokens_seen": 475332608
},
{
"epoch": 0.14,
"learning_rate": 0.00043229689067201605,
"loss": 2.7741,
"theoretical_loss": 3.9362837202887384,
"tokens_seen": 475398144
},
{
"epoch": 0.14,
"learning_rate": 0.0004322868605817453,
"loss": 2.9289,
"theoretical_loss": 3.9362247693434096,
"tokens_seen": 475463680
},
{
"epoch": 0.14,
"learning_rate": 0.0004322768304914744,
"loss": 2.7103,
"theoretical_loss": 3.9361658287978813,
"tokens_seen": 475529216
},
{
"epoch": 0.14,
"learning_rate": 0.00043226680040120365,
"loss": 2.6407,
"theoretical_loss": 3.9361068986488856,
"tokens_seen": 475594752
},
{
"epoch": 0.14,
"learning_rate": 0.00043225677031093283,
"loss": 3.1158,
"theoretical_loss": 3.936047978893158,
"tokens_seen": 475660288
},
{
"epoch": 0.14,
"learning_rate": 0.000432246740220662,
"loss": 2.9055,
"theoretical_loss": 3.935989069527432,
"tokens_seen": 475725824
},
{
"epoch": 0.14,
"learning_rate": 0.0004322367101303912,
"loss": 2.8514,
"theoretical_loss": 3.9359301705484455,
"tokens_seen": 475791360
},
{
"epoch": 0.14,
"learning_rate": 0.00043222668004012037,
"loss": 2.8603,
"theoretical_loss": 3.935871281952937,
"tokens_seen": 475856896
},
{
"epoch": 0.14,
"learning_rate": 0.00043221664994984955,
"loss": 2.8935,
"theoretical_loss": 3.935812403737646,
"tokens_seen": 475922432
},
{
"epoch": 0.14,
"learning_rate": 0.0004322066198595788,
"loss": 3.0784,
"theoretical_loss": 3.9357535358993134,
"tokens_seen": 475987968
},
{
"epoch": 0.14,
"learning_rate": 0.0004321965897693079,
"loss": 2.7309,
"theoretical_loss": 3.9356946784346825,
"tokens_seen": 476053504
},
{
"epoch": 0.14,
"learning_rate": 0.00043218655967903715,
"loss": 2.8205,
"theoretical_loss": 3.9356358313404973,
"tokens_seen": 476119040
},
{
"epoch": 0.14,
"learning_rate": 0.0004321765295887663,
"loss": 2.8091,
"theoretical_loss": 3.9355769946135037,
"tokens_seen": 476184576
},
{
"epoch": 0.14,
"learning_rate": 0.0004321664994984955,
"loss": 2.7657,
"theoretical_loss": 3.9355181682504483,
"tokens_seen": 476250112
},
{
"epoch": 0.14,
"learning_rate": 0.0004321564694082247,
"loss": 2.8504,
"theoretical_loss": 3.9354593522480803,
"tokens_seen": 476315648
},
{
"epoch": 0.14,
"learning_rate": 0.0004321464393179539,
"loss": 2.9936,
"theoretical_loss": 3.93540054660315,
"tokens_seen": 476381184
},
{
"epoch": 0.14,
"learning_rate": 0.00043213640922768306,
"loss": 2.8743,
"theoretical_loss": 3.935341751312408,
"tokens_seen": 476446720
},
{
"epoch": 0.14,
"learning_rate": 0.00043212637913741224,
"loss": 2.6885,
"theoretical_loss": 3.9352829663726085,
"tokens_seen": 476512256
},
{
"epoch": 0.14,
"learning_rate": 0.0004321163490471414,
"loss": 2.9095,
"theoretical_loss": 3.935224191780505,
"tokens_seen": 476577792
},
{
"epoch": 0.14,
"learning_rate": 0.00043210631895687065,
"loss": 2.8379,
"theoretical_loss": 3.9351654275328536,
"tokens_seen": 476643328
},
{
"epoch": 0.14,
"learning_rate": 0.0004320962888665998,
"loss": 2.8026,
"theoretical_loss": 3.9351066736264126,
"tokens_seen": 476708864
},
{
"epoch": 0.14,
"objective/train/docs_used": 281398,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6347885131835938,
"objective/train/theoretical_loss": 3.93504793005794,
"objective/train/tokens_used": 497234400,
"theoretical_loss": 3.93504793005794,
"tokens_seen": 476774400
},
{
"epoch": 0.14,
"learning_rate": 0.000432086258776329,
"loss": 2.8477,
"theoretical_loss": 3.93504793005794,
"tokens_seen": 476774400
},
{
"epoch": 0.14,
"learning_rate": 0.0004320762286860582,
"loss": 2.8116,
"theoretical_loss": 3.9349891968241963,
"tokens_seen": 476839936
},
{
"epoch": 0.14,
"learning_rate": 0.0004320661985957874,
"loss": 3.0569,
"theoretical_loss": 3.934930473921944,
"tokens_seen": 476905472
},
{
"epoch": 0.14,
"learning_rate": 0.00043205616850551656,
"loss": 2.689,
"theoretical_loss": 3.934871761347945,
"tokens_seen": 476971008
},
{
"epoch": 0.14,
"learning_rate": 0.00043204613841524574,
"loss": 2.8428,
"theoretical_loss": 3.9348130590989654,
"tokens_seen": 477036544
},
{
"epoch": 0.14,
"learning_rate": 0.0004320361083249749,
"loss": 2.9822,
"theoretical_loss": 3.93475436717177,
"tokens_seen": 477102080
},
{
"epoch": 0.14,
"learning_rate": 0.00043202607823470416,
"loss": 2.9795,
"theoretical_loss": 3.9346956855631285,
"tokens_seen": 477167616
},
{
"epoch": 0.14,
"learning_rate": 0.0004320160481444333,
"loss": 3.0974,
"theoretical_loss": 3.934637014269808,
"tokens_seen": 477233152
},
{
"epoch": 0.14,
"learning_rate": 0.0004320060180541625,
"loss": 3.0523,
"theoretical_loss": 3.93457835328858,
"tokens_seen": 477298688
},
{
"epoch": 0.14,
"learning_rate": 0.00043199598796389165,
"loss": 2.8689,
"theoretical_loss": 3.9345197026162158,
"tokens_seen": 477364224
},
{
"epoch": 0.14,
"learning_rate": 0.0004319859578736209,
"loss": 2.9996,
"theoretical_loss": 3.9344610622494893,
"tokens_seen": 477429760
},
{
"epoch": 0.14,
"learning_rate": 0.00043197592778335006,
"loss": 2.9221,
"theoretical_loss": 3.934402432185176,
"tokens_seen": 477495296
},
{
"epoch": 0.14,
"learning_rate": 0.00043196589769307924,
"loss": 2.8884,
"theoretical_loss": 3.9343438124200514,
"tokens_seen": 477560832
},
{
"epoch": 0.14,
"learning_rate": 0.0004319558676028084,
"loss": 2.9777,
"theoretical_loss": 3.9342852029508935,
"tokens_seen": 477626368
},
{
"epoch": 0.14,
"learning_rate": 0.0004319458375125376,
"loss": 2.7155,
"theoretical_loss": 3.9342266037744817,
"tokens_seen": 477691904
},
{
"epoch": 0.14,
"learning_rate": 0.0004319358074222668,
"loss": 3.0188,
"theoretical_loss": 3.934168014887597,
"tokens_seen": 477757440
},
{
"epoch": 0.14,
"learning_rate": 0.000431925777331996,
"loss": 3.0369,
"theoretical_loss": 3.934109436287021,
"tokens_seen": 477822976
},
{
"epoch": 0.14,
"learning_rate": 0.00043191574724172515,
"loss": 2.8311,
"theoretical_loss": 3.934050867969537,
"tokens_seen": 477888512
},
{
"epoch": 0.14,
"learning_rate": 0.0004319057171514544,
"loss": 2.9345,
"theoretical_loss": 3.9339923099319316,
"tokens_seen": 477954048
},
{
"epoch": 0.14,
"learning_rate": 0.00043189568706118357,
"loss": 2.6172,
"theoretical_loss": 3.93393376217099,
"tokens_seen": 478019584
},
{
"epoch": 0.14,
"learning_rate": 0.00043188565697091275,
"loss": 2.9966,
"theoretical_loss": 3.9338752246835,
"tokens_seen": 478085120
},
{
"epoch": 0.14,
"learning_rate": 0.00043187562688064193,
"loss": 2.6355,
"theoretical_loss": 3.933816697466252,
"tokens_seen": 478150656
},
{
"epoch": 0.14,
"learning_rate": 0.0004318655967903711,
"loss": 3.0528,
"theoretical_loss": 3.933758180516036,
"tokens_seen": 478216192
},
{
"epoch": 0.14,
"learning_rate": 0.0004318555667001003,
"loss": 2.8607,
"theoretical_loss": 3.9336996738296444,
"tokens_seen": 478281728
},
{
"epoch": 0.14,
"learning_rate": 0.0004318455366098295,
"loss": 3.0465,
"theoretical_loss": 3.9336411774038713,
"tokens_seen": 478347264
},
{
"epoch": 0.14,
"objective/train/docs_used": 281924,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0119645595550537,
"objective/train/theoretical_loss": 3.9335826912355114,
"objective/train/tokens_used": 498872800,
"theoretical_loss": 3.9335826912355114,
"tokens_seen": 478412800
},
{
"epoch": 0.14,
"learning_rate": 0.00043183550651955865,
"loss": 2.8013,
"theoretical_loss": 3.9335826912355114,
"tokens_seen": 478412800
},
{
"epoch": 0.14,
"learning_rate": 0.0004318254764292879,
"loss": 3.0172,
"theoretical_loss": 3.9335242153213614,
"tokens_seen": 478478336
},
{
"epoch": 0.15,
"learning_rate": 0.000431815446339017,
"loss": 2.9958,
"theoretical_loss": 3.9334657496582195,
"tokens_seen": 478543872
},
{
"epoch": 0.15,
"learning_rate": 0.00043180541624874625,
"loss": 2.9764,
"theoretical_loss": 3.933407294242885,
"tokens_seen": 478609408
},
{
"epoch": 0.15,
"learning_rate": 0.00043179538615847543,
"loss": 2.9699,
"theoretical_loss": 3.9333488490721598,
"tokens_seen": 478674944
},
{
"epoch": 0.15,
"learning_rate": 0.0004317853560682046,
"loss": 2.7685,
"theoretical_loss": 3.933290414142845,
"tokens_seen": 478740480
},
{
"epoch": 0.15,
"learning_rate": 0.0004317753259779338,
"loss": 2.7827,
"theoretical_loss": 3.933231989451744,
"tokens_seen": 478806016
},
{
"epoch": 0.15,
"learning_rate": 0.00043176529588766303,
"loss": 2.8662,
"theoretical_loss": 3.9331735749956627,
"tokens_seen": 478871552
},
{
"epoch": 0.15,
"learning_rate": 0.00043175526579739215,
"loss": 2.8525,
"theoretical_loss": 3.933115170771409,
"tokens_seen": 478937088
},
{
"epoch": 0.15,
"learning_rate": 0.0004317452357071214,
"loss": 2.8977,
"theoretical_loss": 3.9330567767757887,
"tokens_seen": 479002624
},
{
"epoch": 0.15,
"learning_rate": 0.0004317352056168505,
"loss": 2.9396,
"theoretical_loss": 3.932998393005613,
"tokens_seen": 479068160
},
{
"epoch": 0.15,
"learning_rate": 0.00043172517552657975,
"loss": 2.9447,
"theoretical_loss": 3.932940019457692,
"tokens_seen": 479133696
},
{
"epoch": 0.15,
"learning_rate": 0.00043171514543630893,
"loss": 2.8563,
"theoretical_loss": 3.9328816561288384,
"tokens_seen": 479199232
},
{
"epoch": 0.15,
"learning_rate": 0.0004317051153460381,
"loss": 2.8169,
"theoretical_loss": 3.932823303015866,
"tokens_seen": 479264768
},
{
"epoch": 0.15,
"learning_rate": 0.0004316950852557673,
"loss": 2.7705,
"theoretical_loss": 3.93276496011559,
"tokens_seen": 479330304
},
{
"epoch": 0.15,
"learning_rate": 0.0004316850551654965,
"loss": 3.1308,
"theoretical_loss": 3.9327066274248272,
"tokens_seen": 479395840
},
{
"epoch": 0.15,
"learning_rate": 0.0004316750250752257,
"loss": 3.0848,
"theoretical_loss": 3.932648304940395,
"tokens_seen": 479461376
},
{
"epoch": 0.15,
"learning_rate": 0.0004316649949849549,
"loss": 3.0146,
"theoretical_loss": 3.932589992659114,
"tokens_seen": 479526912
},
{
"epoch": 0.15,
"learning_rate": 0.0004316549648946841,
"loss": 2.9719,
"theoretical_loss": 3.9325316905778047,
"tokens_seen": 479592448
},
{
"epoch": 0.15,
"learning_rate": 0.00043164493480441326,
"loss": 2.9354,
"theoretical_loss": 3.932473398693289,
"tokens_seen": 479657984
},
{
"epoch": 0.15,
"learning_rate": 0.00043163490471414244,
"loss": 3.2547,
"theoretical_loss": 3.932415117002391,
"tokens_seen": 479723520
},
{
"epoch": 0.15,
"learning_rate": 0.0004316248746238716,
"loss": 3.0618,
"theoretical_loss": 3.9323568455019364,
"tokens_seen": 479789056
},
{
"epoch": 0.15,
"learning_rate": 0.00043161484453360085,
"loss": 2.9419,
"theoretical_loss": 3.932298584188751,
"tokens_seen": 479854592
},
{
"epoch": 0.15,
"learning_rate": 0.00043160481444333,
"loss": 2.9944,
"theoretical_loss": 3.9322403330596636,
"tokens_seen": 479920128
},
{
"epoch": 0.15,
"learning_rate": 0.0004315947843530592,
"loss": 2.7101,
"theoretical_loss": 3.9321820921115034,
"tokens_seen": 479985664
},
{
"epoch": 0.15,
"objective/train/docs_used": 283309,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8511404991149902,
"objective/train/theoretical_loss": 3.932123861341101,
"objective/train/tokens_used": 500511200,
"theoretical_loss": 3.932123861341101,
"tokens_seen": 480051200
},
{
"epoch": 0.15,
"learning_rate": 0.0004315847542627884,
"loss": 2.7065,
"theoretical_loss": 3.932123861341101,
"tokens_seen": 480051200
},
{
"epoch": 0.15,
"learning_rate": 0.0004315747241725176,
"loss": 2.8497,
"theoretical_loss": 3.932065640745289,
"tokens_seen": 480116736
},
{
"epoch": 0.15,
"learning_rate": 0.00043156469408224676,
"loss": 2.7446,
"theoretical_loss": 3.932007430320901,
"tokens_seen": 480182272
},
{
"epoch": 0.15,
"learning_rate": 0.00043155466399197594,
"loss": 2.8389,
"theoretical_loss": 3.9319492300647725,
"tokens_seen": 480247808
},
{
"epoch": 0.15,
"learning_rate": 0.0004315446339017051,
"loss": 2.9232,
"theoretical_loss": 3.93189103997374,
"tokens_seen": 480313344
},
{
"epoch": 0.15,
"learning_rate": 0.00043153460381143436,
"loss": 2.9103,
"theoretical_loss": 3.9318328600446413,
"tokens_seen": 480378880
},
{
"epoch": 0.15,
"learning_rate": 0.0004315245737211635,
"loss": 2.9765,
"theoretical_loss": 3.931774690274316,
"tokens_seen": 480444416
},
{
"epoch": 0.15,
"learning_rate": 0.0004315145436308927,
"loss": 2.9858,
"theoretical_loss": 3.931716530659604,
"tokens_seen": 480509952
},
{
"epoch": 0.15,
"learning_rate": 0.00043150451354062185,
"loss": 2.9393,
"theoretical_loss": 3.931658381197349,
"tokens_seen": 480575488
},
{
"epoch": 0.15,
"learning_rate": 0.0004314944834503511,
"loss": 2.8405,
"theoretical_loss": 3.9316002418843934,
"tokens_seen": 480641024
},
{
"epoch": 0.15,
"learning_rate": 0.00043148445336008026,
"loss": 2.9599,
"theoretical_loss": 3.931542112717583,
"tokens_seen": 480706560
},
{
"epoch": 0.15,
"learning_rate": 0.00043147442326980944,
"loss": 2.7486,
"theoretical_loss": 3.931483993693764,
"tokens_seen": 480772096
},
{
"epoch": 0.15,
"learning_rate": 0.0004314643931795386,
"loss": 2.998,
"theoretical_loss": 3.9314258848097845,
"tokens_seen": 480837632
},
{
"epoch": 0.15,
"learning_rate": 0.0004314543630892678,
"loss": 2.8837,
"theoretical_loss": 3.9313677860624936,
"tokens_seen": 480903168
},
{
"epoch": 0.15,
"learning_rate": 0.000431444332998997,
"loss": 2.9617,
"theoretical_loss": 3.931309697448742,
"tokens_seen": 480968704
},
{
"epoch": 0.15,
"learning_rate": 0.0004314343029087262,
"loss": 3.1359,
"theoretical_loss": 3.9312516189653817,
"tokens_seen": 481034240
},
{
"epoch": 0.15,
"learning_rate": 0.00043142427281845535,
"loss": 3.0123,
"theoretical_loss": 3.931193550609266,
"tokens_seen": 481099776
},
{
"epoch": 0.15,
"learning_rate": 0.0004314142427281846,
"loss": 2.8058,
"theoretical_loss": 3.9311354923772504,
"tokens_seen": 481165312
},
{
"epoch": 0.15,
"learning_rate": 0.00043140421263791377,
"loss": 2.9216,
"theoretical_loss": 3.9310774442661907,
"tokens_seen": 481230848
},
{
"epoch": 0.15,
"learning_rate": 0.00043139418254764295,
"loss": 2.9538,
"theoretical_loss": 3.931019406272945,
"tokens_seen": 481296384
},
{
"epoch": 0.15,
"learning_rate": 0.00043138415245737213,
"loss": 2.7984,
"theoretical_loss": 3.930961378394372,
"tokens_seen": 481361920
},
{
"epoch": 0.15,
"learning_rate": 0.0004313741223671013,
"loss": 2.9692,
"theoretical_loss": 3.9309033606273327,
"tokens_seen": 481427456
},
{
"epoch": 0.15,
"learning_rate": 0.0004313640922768305,
"loss": 2.8344,
"theoretical_loss": 3.9308453529686886,
"tokens_seen": 481492992
},
{
"epoch": 0.15,
"learning_rate": 0.0004313540621865597,
"loss": 2.9699,
"theoretical_loss": 3.930787355415304,
"tokens_seen": 481558528
},
{
"epoch": 0.15,
"learning_rate": 0.00043134403209628885,
"loss": 2.9992,
"theoretical_loss": 3.930729367964042,
"tokens_seen": 481624064
},
{
"epoch": 0.15,
"objective/train/docs_used": 283520,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7898130416870117,
"objective/train/theoretical_loss": 3.93067139061177,
"objective/train/tokens_used": 502149600,
"theoretical_loss": 3.93067139061177,
"tokens_seen": 481689600
},
{
"epoch": 0.15,
"learning_rate": 0.0004313340020060181,
"loss": 2.9453,
"theoretical_loss": 3.93067139061177,
"tokens_seen": 481689600
},
{
"epoch": 0.15,
"learning_rate": 0.0004313239719157472,
"loss": 2.9551,
"theoretical_loss": 3.930613423355355,
"tokens_seen": 481755136
},
{
"epoch": 0.15,
"learning_rate": 0.00043131394182547645,
"loss": 3.2456,
"theoretical_loss": 3.9305554661916657,
"tokens_seen": 481820672
},
{
"epoch": 0.15,
"learning_rate": 0.00043130391173520563,
"loss": 2.7297,
"theoretical_loss": 3.9304975191175733,
"tokens_seen": 481886208
},
{
"epoch": 0.15,
"learning_rate": 0.0004312938816449348,
"loss": 3.0574,
"theoretical_loss": 3.9304395821299485,
"tokens_seen": 481951744
},
{
"epoch": 0.15,
"learning_rate": 0.000431283851554664,
"loss": 2.6541,
"theoretical_loss": 3.9303816552256654,
"tokens_seen": 482017280
},
{
"epoch": 0.15,
"learning_rate": 0.00043127382146439323,
"loss": 3.0918,
"theoretical_loss": 3.930323738401597,
"tokens_seen": 482082816
},
{
"epoch": 0.15,
"learning_rate": 0.00043126379137412236,
"loss": 3.1726,
"theoretical_loss": 3.930265831654621,
"tokens_seen": 482148352
},
{
"epoch": 0.15,
"learning_rate": 0.0004312537612838516,
"loss": 2.7858,
"theoretical_loss": 3.930207934981614,
"tokens_seen": 482213888
},
{
"epoch": 0.15,
"learning_rate": 0.0004312437311935807,
"loss": 2.9732,
"theoretical_loss": 3.930150048379454,
"tokens_seen": 482279424
},
{
"epoch": 0.15,
"learning_rate": 0.00043123370110330995,
"loss": 2.9883,
"theoretical_loss": 3.930092171845022,
"tokens_seen": 482344960
},
{
"epoch": 0.15,
"learning_rate": 0.00043123370110330995,
"loss": 3.0501,
"theoretical_loss": 3.930034305375199,
"tokens_seen": 482410496
},
{
"epoch": 0.15,
"learning_rate": 0.00043123370110330995,
"loss": 2.9022,
"theoretical_loss": 3.929976448966868,
"tokens_seen": 482476032
},
{
"epoch": 0.15,
"learning_rate": 0.00043122367101303913,
"loss": 2.8998,
"theoretical_loss": 3.929918602616913,
"tokens_seen": 482541568
},
{
"epoch": 0.15,
"learning_rate": 0.0004312136409227683,
"loss": 2.8981,
"theoretical_loss": 3.92986076632222,
"tokens_seen": 482607104
},
{
"epoch": 0.15,
"learning_rate": 0.0004312036108324975,
"loss": 3.1734,
"theoretical_loss": 3.929802940079676,
"tokens_seen": 482672640
},
{
"epoch": 0.15,
"learning_rate": 0.0004311935807422267,
"loss": 2.7899,
"theoretical_loss": 3.9297451238861685,
"tokens_seen": 482738176
},
{
"epoch": 0.15,
"learning_rate": 0.00043118355065195586,
"loss": 3.0353,
"theoretical_loss": 3.9296873177385887,
"tokens_seen": 482803712
},
{
"epoch": 0.15,
"learning_rate": 0.0004311735205616851,
"loss": 2.9585,
"theoretical_loss": 3.929629521633827,
"tokens_seen": 482869248
},
{
"epoch": 0.15,
"learning_rate": 0.0004311634904714142,
"loss": 3.0221,
"theoretical_loss": 3.929571735568776,
"tokens_seen": 482934784
},
{
"epoch": 0.15,
"learning_rate": 0.00043115346038114346,
"loss": 3.0581,
"theoretical_loss": 3.92951395954033,
"tokens_seen": 483000320
},
{
"epoch": 0.15,
"learning_rate": 0.0004311434302908726,
"loss": 2.9762,
"theoretical_loss": 3.9294561935453833,
"tokens_seen": 483065856
},
{
"epoch": 0.15,
"learning_rate": 0.0004311334002006018,
"loss": 3.0789,
"theoretical_loss": 3.9293984375808337,
"tokens_seen": 483131392
},
{
"epoch": 0.15,
"learning_rate": 0.000431123370110331,
"loss": 2.8882,
"theoretical_loss": 3.929340691643579,
"tokens_seen": 483196928
},
{
"epoch": 0.15,
"learning_rate": 0.0004311133400200602,
"loss": 2.7947,
"theoretical_loss": 3.929282955730519,
"tokens_seen": 483262464
},
{
"epoch": 0.15,
"objective/train/docs_used": 284734,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4735991954803467,
"objective/train/theoretical_loss": 3.9292252298385533,
"objective/train/tokens_used": 503788000,
"theoretical_loss": 3.9292252298385533,
"tokens_seen": 483328000
},
{
"epoch": 0.15,
"learning_rate": 0.00043110330992978936,
"loss": 3.1294,
"theoretical_loss": 3.9292252298385533,
"tokens_seen": 483328000
},
{
"epoch": 0.15,
"learning_rate": 0.0004310932798395186,
"loss": 2.921,
"theoretical_loss": 3.929167513964585,
"tokens_seen": 483393536
},
{
"epoch": 0.15,
"learning_rate": 0.0004310832497492477,
"loss": 2.8072,
"theoretical_loss": 3.929109808105518,
"tokens_seen": 483459072
},
{
"epoch": 0.15,
"learning_rate": 0.00043107321965897696,
"loss": 3.0061,
"theoretical_loss": 3.9290521122582573,
"tokens_seen": 483524608
},
{
"epoch": 0.15,
"learning_rate": 0.0004310631895687061,
"loss": 3.1074,
"theoretical_loss": 3.928994426419708,
"tokens_seen": 483590144
},
{
"epoch": 0.15,
"learning_rate": 0.0004310531594784353,
"loss": 2.9239,
"theoretical_loss": 3.928936750586779,
"tokens_seen": 483655680
},
{
"epoch": 0.15,
"learning_rate": 0.0004310431293881645,
"loss": 3.066,
"theoretical_loss": 3.9288790847563795,
"tokens_seen": 483721216
},
{
"epoch": 0.15,
"learning_rate": 0.0004310330992978937,
"loss": 2.9879,
"theoretical_loss": 3.9288214289254193,
"tokens_seen": 483786752
},
{
"epoch": 0.15,
"learning_rate": 0.00043102306920762286,
"loss": 3.0037,
"theoretical_loss": 3.9287637830908104,
"tokens_seen": 483852288
},
{
"epoch": 0.15,
"learning_rate": 0.00043101303911735205,
"loss": 2.8875,
"theoretical_loss": 3.928706147249466,
"tokens_seen": 483917824
},
{
"epoch": 0.15,
"learning_rate": 0.0004310030090270812,
"loss": 2.911,
"theoretical_loss": 3.9286485213983013,
"tokens_seen": 483983360
},
{
"epoch": 0.15,
"learning_rate": 0.00043099297893681046,
"loss": 2.8981,
"theoretical_loss": 3.9285909055342314,
"tokens_seen": 484048896
},
{
"epoch": 0.15,
"learning_rate": 0.0004309829488465396,
"loss": 3.1723,
"theoretical_loss": 3.928533299654174,
"tokens_seen": 484114432
},
{
"epoch": 0.15,
"learning_rate": 0.0004309729187562688,
"loss": 3.0433,
"theoretical_loss": 3.928475703755048,
"tokens_seen": 484179968
},
{
"epoch": 0.15,
"learning_rate": 0.00043096288866599795,
"loss": 2.9147,
"theoretical_loss": 3.9284181178337736,
"tokens_seen": 484245504
},
{
"epoch": 0.15,
"learning_rate": 0.0004309528585757272,
"loss": 2.8724,
"theoretical_loss": 3.9283605418872716,
"tokens_seen": 484311040
},
{
"epoch": 0.15,
"learning_rate": 0.00043094282848545637,
"loss": 3.0486,
"theoretical_loss": 3.9283029759124646,
"tokens_seen": 484376576
},
{
"epoch": 0.15,
"learning_rate": 0.00043093279839518555,
"loss": 3.2149,
"theoretical_loss": 3.9282454199062777,
"tokens_seen": 484442112
},
{
"epoch": 0.15,
"learning_rate": 0.0004309227683049148,
"loss": 2.8778,
"theoretical_loss": 3.928187873865636,
"tokens_seen": 484507648
},
{
"epoch": 0.15,
"learning_rate": 0.00043091273821464397,
"loss": 3.0867,
"theoretical_loss": 3.928130337787467,
"tokens_seen": 484573184
},
{
"epoch": 0.15,
"learning_rate": 0.00043090270812437315,
"loss": 3.0863,
"theoretical_loss": 3.9280728116686974,
"tokens_seen": 484638720
},
{
"epoch": 0.15,
"learning_rate": 0.00043089267803410233,
"loss": 3.154,
"theoretical_loss": 3.9280152955062584,
"tokens_seen": 484704256
},
{
"epoch": 0.15,
"learning_rate": 0.0004308826479438315,
"loss": 2.9944,
"theoretical_loss": 3.92795778929708,
"tokens_seen": 484769792
},
{
"epoch": 0.15,
"learning_rate": 0.0004308726178535607,
"loss": 2.9434,
"theoretical_loss": 3.9279002930380953,
"tokens_seen": 484835328
},
{
"epoch": 0.15,
"learning_rate": 0.0004308625877632899,
"loss": 3.1584,
"theoretical_loss": 3.927842806726237,
"tokens_seen": 484900864
},
{
"epoch": 0.15,
"objective/train/docs_used": 285372,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4254167079925537,
"objective/train/theoretical_loss": 3.927785330358441,
"objective/train/tokens_used": 505426400,
"theoretical_loss": 3.927785330358441,
"tokens_seen": 484966400
},
{
"epoch": 0.15,
"learning_rate": 0.00043085255767301905,
"loss": 2.8477,
"theoretical_loss": 3.927785330358441,
"tokens_seen": 484966400
},
{
"epoch": 0.15,
"learning_rate": 0.0004308425275827483,
"loss": 2.9169,
"theoretical_loss": 3.9277278639316435,
"tokens_seen": 485031936
},
{
"epoch": 0.15,
"learning_rate": 0.0004308324974924774,
"loss": 2.9502,
"theoretical_loss": 3.9276704074427826,
"tokens_seen": 485097472
},
{
"epoch": 0.15,
"learning_rate": 0.00043082246740220665,
"loss": 2.8843,
"theoretical_loss": 3.927612960888797,
"tokens_seen": 485163008
},
{
"epoch": 0.15,
"learning_rate": 0.00043081243731193583,
"loss": 2.7812,
"theoretical_loss": 3.9275555242666265,
"tokens_seen": 485228544
},
{
"epoch": 0.15,
"learning_rate": 0.000430802407221665,
"loss": 3.0061,
"theoretical_loss": 3.9274980975732143,
"tokens_seen": 485294080
},
{
"epoch": 0.15,
"learning_rate": 0.0004307923771313942,
"loss": 2.9978,
"theoretical_loss": 3.927440680805503,
"tokens_seen": 485359616
},
{
"epoch": 0.15,
"learning_rate": 0.00043078234704112343,
"loss": 2.802,
"theoretical_loss": 3.9273832739604373,
"tokens_seen": 485425152
},
{
"epoch": 0.15,
"learning_rate": 0.00043077231695085256,
"loss": 2.8012,
"theoretical_loss": 3.9273258770349626,
"tokens_seen": 485490688
},
{
"epoch": 0.15,
"learning_rate": 0.0004307622868605818,
"loss": 2.9933,
"theoretical_loss": 3.9272684900260266,
"tokens_seen": 485556224
},
{
"epoch": 0.15,
"learning_rate": 0.0004307522567703109,
"loss": 3.0707,
"theoretical_loss": 3.9272111129305785,
"tokens_seen": 485621760
},
{
"epoch": 0.15,
"learning_rate": 0.00043074222668004015,
"loss": 3.0468,
"theoretical_loss": 3.927153745745567,
"tokens_seen": 485687296
},
{
"epoch": 0.15,
"learning_rate": 0.00043073219658976933,
"loss": 3.0784,
"theoretical_loss": 3.9270963884679437,
"tokens_seen": 485752832
},
{
"epoch": 0.15,
"learning_rate": 0.0004307221664994985,
"loss": 2.8892,
"theoretical_loss": 3.9270390410946625,
"tokens_seen": 485818368
},
{
"epoch": 0.15,
"learning_rate": 0.0004307121364092277,
"loss": 2.9743,
"theoretical_loss": 3.9269817036226757,
"tokens_seen": 485883904
},
{
"epoch": 0.15,
"learning_rate": 0.0004307021063189569,
"loss": 2.8246,
"theoretical_loss": 3.92692437604894,
"tokens_seen": 485949440
},
{
"epoch": 0.15,
"learning_rate": 0.00043069207622868606,
"loss": 2.8975,
"theoretical_loss": 3.9268670583704113,
"tokens_seen": 486014976
},
{
"epoch": 0.15,
"learning_rate": 0.0004306820461384153,
"loss": 3.146,
"theoretical_loss": 3.926809750584048,
"tokens_seen": 486080512
},
{
"epoch": 0.15,
"learning_rate": 0.0004306720160481444,
"loss": 2.9466,
"theoretical_loss": 3.9267524526868085,
"tokens_seen": 486146048
},
{
"epoch": 0.15,
"learning_rate": 0.00043066198595787366,
"loss": 3.0278,
"theoretical_loss": 3.9266951646756554,
"tokens_seen": 486211584
},
{
"epoch": 0.15,
"learning_rate": 0.0004306519558676028,
"loss": 2.8523,
"theoretical_loss": 3.9266378865475495,
"tokens_seen": 486277120
},
{
"epoch": 0.15,
"learning_rate": 0.000430641925777332,
"loss": 2.9906,
"theoretical_loss": 3.9265806182994543,
"tokens_seen": 486342656
},
{
"epoch": 0.15,
"learning_rate": 0.0004306318956870612,
"loss": 2.8391,
"theoretical_loss": 3.926523359928335,
"tokens_seen": 486408192
},
{
"epoch": 0.15,
"learning_rate": 0.0004306218655967904,
"loss": 2.8194,
"theoretical_loss": 3.9264661114311576,
"tokens_seen": 486473728
},
{
"epoch": 0.15,
"learning_rate": 0.00043061183550651956,
"loss": 3.0383,
"theoretical_loss": 3.926408872804889,
"tokens_seen": 486539264
},
{
"epoch": 0.15,
"objective/train/docs_used": 285999,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2507073879241943,
"objective/train/theoretical_loss": 3.9263516440464987,
"objective/train/tokens_used": 507064800,
"theoretical_loss": 3.9263516440464987,
"tokens_seen": 486604800
},
{
"epoch": 0.15,
"learning_rate": 0.0004306018054162488,
"loss": 2.9126,
"theoretical_loss": 3.9263516440464987,
"tokens_seen": 486604800
},
{
"epoch": 0.15,
"learning_rate": 0.0004305917753259779,
"loss": 2.7026,
"theoretical_loss": 3.926294425152957,
"tokens_seen": 486670336
},
{
"epoch": 0.15,
"learning_rate": 0.00043058174523570716,
"loss": 2.9969,
"theoretical_loss": 3.926237216121234,
"tokens_seen": 486735872
},
{
"epoch": 0.15,
"learning_rate": 0.0004305717151454363,
"loss": 3.0097,
"theoretical_loss": 3.9261800169483037,
"tokens_seen": 486801408
},
{
"epoch": 0.15,
"learning_rate": 0.0004305616850551655,
"loss": 2.9747,
"theoretical_loss": 3.9261228276311404,
"tokens_seen": 486866944
},
{
"epoch": 0.15,
"learning_rate": 0.0004305516549648947,
"loss": 2.9391,
"theoretical_loss": 3.926065648166719,
"tokens_seen": 486932480
},
{
"epoch": 0.15,
"learning_rate": 0.0004305416248746239,
"loss": 2.761,
"theoretical_loss": 3.926008478552016,
"tokens_seen": 486998016
},
{
"epoch": 0.15,
"learning_rate": 0.00043053159478435306,
"loss": 2.8618,
"theoretical_loss": 3.9259513187840103,
"tokens_seen": 487063552
},
{
"epoch": 0.15,
"learning_rate": 0.00043052156469408225,
"loss": 2.946,
"theoretical_loss": 3.9258941688596805,
"tokens_seen": 487129088
},
{
"epoch": 0.15,
"learning_rate": 0.0004305115346038114,
"loss": 3.0027,
"theoretical_loss": 3.9258370287760087,
"tokens_seen": 487194624
},
{
"epoch": 0.15,
"learning_rate": 0.00043050150451354066,
"loss": 3.0505,
"theoretical_loss": 3.9257798985299757,
"tokens_seen": 487260160
},
{
"epoch": 0.15,
"learning_rate": 0.0004304914744232698,
"loss": 3.0067,
"theoretical_loss": 3.9257227781185655,
"tokens_seen": 487325696
},
{
"epoch": 0.15,
"learning_rate": 0.000430481444332999,
"loss": 2.9225,
"theoretical_loss": 3.9256656675387633,
"tokens_seen": 487391232
},
{
"epoch": 0.15,
"learning_rate": 0.00043047141424272815,
"loss": 2.8414,
"theoretical_loss": 3.925608566787555,
"tokens_seen": 487456768
},
{
"epoch": 0.15,
"learning_rate": 0.0004304613841524574,
"loss": 3.0204,
"theoretical_loss": 3.925551475861928,
"tokens_seen": 487522304
},
{
"epoch": 0.15,
"learning_rate": 0.00043045135406218657,
"loss": 2.9023,
"theoretical_loss": 3.9254943947588705,
"tokens_seen": 487587840
},
{
"epoch": 0.15,
"learning_rate": 0.00043044132397191575,
"loss": 2.8248,
"theoretical_loss": 3.9254373234753737,
"tokens_seen": 487653376
},
{
"epoch": 0.15,
"learning_rate": 0.00043043129388164493,
"loss": 3.0453,
"theoretical_loss": 3.925380262008428,
"tokens_seen": 487718912
},
{
"epoch": 0.15,
"learning_rate": 0.00043042126379137417,
"loss": 3.1579,
"theoretical_loss": 3.9253232103550273,
"tokens_seen": 487784448
},
{
"epoch": 0.15,
"learning_rate": 0.0004304112337011033,
"loss": 2.7528,
"theoretical_loss": 3.925266168512165,
"tokens_seen": 487849984
},
{
"epoch": 0.15,
"learning_rate": 0.00043040120361083253,
"loss": 2.6098,
"theoretical_loss": 3.925209136476836,
"tokens_seen": 487915520
},
{
"epoch": 0.15,
"learning_rate": 0.00043039117352056165,
"loss": 2.9598,
"theoretical_loss": 3.9251521142460377,
"tokens_seen": 487981056
},
{
"epoch": 0.15,
"learning_rate": 0.0004303811434302909,
"loss": 3.0533,
"theoretical_loss": 3.925095101816768,
"tokens_seen": 488046592
},
{
"epoch": 0.15,
"learning_rate": 0.00043037111334002007,
"loss": 2.87,
"theoretical_loss": 3.925038099186027,
"tokens_seen": 488112128
},
{
"epoch": 0.15,
"learning_rate": 0.00043036108324974925,
"loss": 2.9433,
"theoretical_loss": 3.924981106350815,
"tokens_seen": 488177664
},
{
"epoch": 0.15,
"objective/train/docs_used": 287428,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0103299617767334,
"objective/train/theoretical_loss": 3.9249241233081333,
"objective/train/tokens_used": 508703200,
"theoretical_loss": 3.9249241233081333,
"tokens_seen": 488243200
},
{
"epoch": 0.15,
"learning_rate": 0.00043035105315947843,
"loss": 2.7969,
"theoretical_loss": 3.9249241233081333,
"tokens_seen": 488243200
},
{
"epoch": 0.15,
"learning_rate": 0.0004303410230692076,
"loss": 3.0867,
"theoretical_loss": 3.924867150054985,
"tokens_seen": 488308736
},
{
"epoch": 0.15,
"learning_rate": 0.0004303309929789368,
"loss": 2.7417,
"theoretical_loss": 3.9248101865883767,
"tokens_seen": 488374272
},
{
"epoch": 0.15,
"learning_rate": 0.00043032096288866603,
"loss": 2.773,
"theoretical_loss": 3.9247532329053127,
"tokens_seen": 488439808
},
{
"epoch": 0.15,
"learning_rate": 0.00043031093279839516,
"loss": 2.9414,
"theoretical_loss": 3.924696289002801,
"tokens_seen": 488505344
},
{
"epoch": 0.15,
"learning_rate": 0.0004303009027081244,
"loss": 2.9699,
"theoretical_loss": 3.9246393548778498,
"tokens_seen": 488570880
},
{
"epoch": 0.15,
"learning_rate": 0.0004302908726178535,
"loss": 3.0303,
"theoretical_loss": 3.9245824305274697,
"tokens_seen": 488636416
},
{
"epoch": 0.15,
"learning_rate": 0.00043028084252758276,
"loss": 3.1834,
"theoretical_loss": 3.9245255159486714,
"tokens_seen": 488701952
},
{
"epoch": 0.15,
"learning_rate": 0.00043027081243731194,
"loss": 2.9536,
"theoretical_loss": 3.924468611138467,
"tokens_seen": 488767488
},
{
"epoch": 0.15,
"learning_rate": 0.0004302607823470411,
"loss": 2.851,
"theoretical_loss": 3.9244117160938714,
"tokens_seen": 488833024
},
{
"epoch": 0.15,
"learning_rate": 0.0004302507522567703,
"loss": 2.9988,
"theoretical_loss": 3.9243548308118994,
"tokens_seen": 488898560
},
{
"epoch": 0.15,
"learning_rate": 0.00043024072216649953,
"loss": 2.9765,
"theoretical_loss": 3.9242979552895676,
"tokens_seen": 488964096
},
{
"epoch": 0.15,
"learning_rate": 0.00043023069207622866,
"loss": 2.7578,
"theoretical_loss": 3.9242410895238935,
"tokens_seen": 489029632
},
{
"epoch": 0.15,
"learning_rate": 0.0004302206619859579,
"loss": 2.8285,
"theoretical_loss": 3.9241842335118964,
"tokens_seen": 489095168
},
{
"epoch": 0.15,
"learning_rate": 0.000430210631895687,
"loss": 2.7398,
"theoretical_loss": 3.9241273872505973,
"tokens_seen": 489160704
},
{
"epoch": 0.15,
"learning_rate": 0.00043020060180541626,
"loss": 2.9964,
"theoretical_loss": 3.924070550737017,
"tokens_seen": 489226240
},
{
"epoch": 0.15,
"learning_rate": 0.00043019057171514544,
"loss": 2.8711,
"theoretical_loss": 3.9240137239681787,
"tokens_seen": 489291776
},
{
"epoch": 0.15,
"learning_rate": 0.0004301805416248746,
"loss": 3.0602,
"theoretical_loss": 3.9239569069411067,
"tokens_seen": 489357312
},
{
"epoch": 0.15,
"learning_rate": 0.00043017051153460386,
"loss": 2.8675,
"theoretical_loss": 3.923900099652828,
"tokens_seen": 489422848
},
{
"epoch": 0.15,
"learning_rate": 0.000430160481444333,
"loss": 2.9146,
"theoretical_loss": 3.923843302100368,
"tokens_seen": 489488384
},
{
"epoch": 0.15,
"learning_rate": 0.0004301504513540622,
"loss": 3.0138,
"theoretical_loss": 3.9237865142807555,
"tokens_seen": 489553920
},
{
"epoch": 0.15,
"learning_rate": 0.0004301404212637914,
"loss": 2.8976,
"theoretical_loss": 3.92372973619102,
"tokens_seen": 489619456
},
{
"epoch": 0.15,
"learning_rate": 0.0004301303911735206,
"loss": 2.9251,
"theoretical_loss": 3.9236729678281925,
"tokens_seen": 489684992
},
{
"epoch": 0.15,
"learning_rate": 0.00043012036108324976,
"loss": 2.8385,
"theoretical_loss": 3.923616209189306,
"tokens_seen": 489750528
},
{
"epoch": 0.15,
"learning_rate": 0.000430110330992979,
"loss": 2.9459,
"theoretical_loss": 3.9235594602713926,
"tokens_seen": 489816064
},
{
"epoch": 0.15,
"objective/train/docs_used": 288116,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9460251331329346,
"objective/train/theoretical_loss": 3.923502721071488,
"objective/train/tokens_used": 510341600,
"theoretical_loss": 3.923502721071488,
"tokens_seen": 489881600
},
{
"epoch": 0.15,
"learning_rate": 0.0004301003009027081,
"loss": 2.9832,
"theoretical_loss": 3.923502721071488,
"tokens_seen": 489881600
},
{
"epoch": 0.15,
"learning_rate": 0.00043009027081243736,
"loss": 3.1877,
"theoretical_loss": 3.9234459915866275,
"tokens_seen": 489947136
},
{
"epoch": 0.15,
"learning_rate": 0.0004300802407221665,
"loss": 2.753,
"theoretical_loss": 3.923389271813849,
"tokens_seen": 490012672
},
{
"epoch": 0.15,
"learning_rate": 0.0004300702106318957,
"loss": 2.6838,
"theoretical_loss": 3.923332561750192,
"tokens_seen": 490078208
},
{
"epoch": 0.15,
"learning_rate": 0.0004300601805416249,
"loss": 3.1324,
"theoretical_loss": 3.9232758613926952,
"tokens_seen": 490143744
},
{
"epoch": 0.15,
"learning_rate": 0.0004300501504513541,
"loss": 2.8496,
"theoretical_loss": 3.9232191707384,
"tokens_seen": 490209280
},
{
"epoch": 0.15,
"learning_rate": 0.00043004012036108326,
"loss": 2.9621,
"theoretical_loss": 3.9231624897843496,
"tokens_seen": 490274816
},
{
"epoch": 0.15,
"learning_rate": 0.00043003009027081245,
"loss": 3.0133,
"theoretical_loss": 3.9231058185275884,
"tokens_seen": 490340352
},
{
"epoch": 0.15,
"learning_rate": 0.00043002006018054163,
"loss": 2.9931,
"theoretical_loss": 3.9230491569651598,
"tokens_seen": 490405888
},
{
"epoch": 0.15,
"learning_rate": 0.00043001003009027086,
"loss": 3.1411,
"theoretical_loss": 3.922992505094112,
"tokens_seen": 490471424
},
{
"epoch": 0.15,
"learning_rate": 0.00043,
"loss": 2.8188,
"theoretical_loss": 3.9229358629114923,
"tokens_seen": 490536960
},
{
"epoch": 0.15,
"learning_rate": 0.0004299899699097292,
"loss": 2.8421,
"theoretical_loss": 3.922879230414349,
"tokens_seen": 490602496
},
{
"epoch": 0.15,
"learning_rate": 0.00042997993981945835,
"loss": 2.7538,
"theoretical_loss": 3.9228226075997332,
"tokens_seen": 490668032
},
{
"epoch": 0.15,
"learning_rate": 0.0004299699097291876,
"loss": 2.8657,
"theoretical_loss": 3.9227659944646964,
"tokens_seen": 490733568
},
{
"epoch": 0.15,
"learning_rate": 0.00042995987963891677,
"loss": 2.9915,
"theoretical_loss": 3.9227093910062916,
"tokens_seen": 490799104
},
{
"epoch": 0.15,
"learning_rate": 0.00042994984954864595,
"loss": 3.0582,
"theoretical_loss": 3.922652797221573,
"tokens_seen": 490864640
},
{
"epoch": 0.15,
"learning_rate": 0.00042993981945837513,
"loss": 2.6739,
"theoretical_loss": 3.922596213107596,
"tokens_seen": 490930176
},
{
"epoch": 0.15,
"learning_rate": 0.00042992978936810437,
"loss": 3.0469,
"theoretical_loss": 3.922539638661418,
"tokens_seen": 490995712
},
{
"epoch": 0.15,
"learning_rate": 0.0004299197592778335,
"loss": 3.0504,
"theoretical_loss": 3.9224830738800964,
"tokens_seen": 491061248
},
{
"epoch": 0.15,
"learning_rate": 0.00042990972918756273,
"loss": 2.9923,
"theoretical_loss": 3.92242651876069,
"tokens_seen": 491126784
},
{
"epoch": 0.15,
"learning_rate": 0.00042989969909729185,
"loss": 2.8921,
"theoretical_loss": 3.9223699733002615,
"tokens_seen": 491192320
},
{
"epoch": 0.15,
"learning_rate": 0.0004298896690070211,
"loss": 2.944,
"theoretical_loss": 3.922313437495871,
"tokens_seen": 491257856
},
{
"epoch": 0.15,
"learning_rate": 0.00042987963891675027,
"loss": 2.9777,
"theoretical_loss": 3.9222569113445824,
"tokens_seen": 491323392
},
{
"epoch": 0.15,
"learning_rate": 0.00042986960882647945,
"loss": 2.9274,
"theoretical_loss": 3.9222003948434603,
"tokens_seen": 491388928
},
{
"epoch": 0.15,
"learning_rate": 0.00042985957873620863,
"loss": 2.8219,
"theoretical_loss": 3.92214388798957,
"tokens_seen": 491454464
},
{
"debugging/Self-BLEU-5": 0.43646517028931053,
"debugging/distinct-1-grams": 0.8073822333747359,
"debugging/distinct-2-grams": 0.976779594165329,
"debugging/entropy-1-grams": 5.794474064713653,
"debugging/entropy-2-grams": 6.45939179102421,
"debugging/length": 536.5,
"debugging/num_segments": 8,
"epoch": 0.15,
"objective/train/docs_used": 289197,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.5808117389678955,
"objective/train/theoretical_loss": 3.92208739077998,
"objective/train/tokens_used": 511980000,
"theoretical_loss": 3.92208739077998,
"tokens_seen": 491520000
},
{
"epoch": 0.15,
"learning_rate": 0.0004298495486459378,
"loss": 3.1925,
"theoretical_loss": 3.92208739077998,
"tokens_seen": 491520000
},
{
"epoch": 0.15,
"learning_rate": 0.000429839518555667,
"loss": 2.9464,
"theoretical_loss": 3.9220309032117573,
"tokens_seen": 491585536
},
{
"epoch": 0.15,
"learning_rate": 0.00042982948846539623,
"loss": 2.8567,
"theoretical_loss": 3.921974425281972,
"tokens_seen": 491651072
},
{
"epoch": 0.15,
"learning_rate": 0.00042981945837512536,
"loss": 2.9663,
"theoretical_loss": 3.921917956987695,
"tokens_seen": 491716608
},
{
"epoch": 0.15,
"learning_rate": 0.0004298094282848546,
"loss": 2.9208,
"theoretical_loss": 3.921861498325998,
"tokens_seen": 491782144
},
{
"epoch": 0.15,
"learning_rate": 0.0004297993981945837,
"loss": 3.063,
"theoretical_loss": 3.921805049293956,
"tokens_seen": 491847680
},
{
"epoch": 0.15,
"learning_rate": 0.00042978936810431296,
"loss": 2.874,
"theoretical_loss": 3.9217486098886427,
"tokens_seen": 491913216
},
{
"epoch": 0.15,
"learning_rate": 0.00042977933801404214,
"loss": 2.9789,
"theoretical_loss": 3.921692180107134,
"tokens_seen": 491978752
},
{
"epoch": 0.15,
"learning_rate": 0.0004297693079237713,
"loss": 2.8321,
"theoretical_loss": 3.9216357599465073,
"tokens_seen": 492044288
},
{
"epoch": 0.15,
"learning_rate": 0.0004297592778335005,
"loss": 2.8587,
"theoretical_loss": 3.921579349403842,
"tokens_seen": 492109824
},
{
"epoch": 0.15,
"learning_rate": 0.00042974924774322973,
"loss": 2.9965,
"theoretical_loss": 3.921522948476217,
"tokens_seen": 492175360
},
{
"epoch": 0.15,
"learning_rate": 0.00042973921765295886,
"loss": 2.9735,
"theoretical_loss": 3.9214665571607146,
"tokens_seen": 492240896
},
{
"epoch": 0.15,
"learning_rate": 0.0004297291875626881,
"loss": 3.0153,
"theoretical_loss": 3.921410175454416,
"tokens_seen": 492306432
},
{
"epoch": 0.15,
"learning_rate": 0.0004297191574724172,
"loss": 2.8714,
"theoretical_loss": 3.921353803354406,
"tokens_seen": 492371968
},
{
"epoch": 0.15,
"learning_rate": 0.00042970912738214646,
"loss": 2.9131,
"theoretical_loss": 3.9212974408577685,
"tokens_seen": 492437504
},
{
"epoch": 0.15,
"learning_rate": 0.00042969909729187564,
"loss": 2.7599,
"theoretical_loss": 3.921241087961591,
"tokens_seen": 492503040
},
{
"epoch": 0.15,
"learning_rate": 0.0004296890672016048,
"loss": 2.8852,
"theoretical_loss": 3.92118474466296,
"tokens_seen": 492568576
},
{
"epoch": 0.15,
"learning_rate": 0.000429679037111334,
"loss": 2.9555,
"theoretical_loss": 3.9211284109589646,
"tokens_seen": 492634112
},
{
"epoch": 0.15,
"learning_rate": 0.0004296690070210632,
"loss": 2.831,
"theoretical_loss": 3.9210720868466953,
"tokens_seen": 492699648
},
{
"epoch": 0.15,
"learning_rate": 0.00042965897693079236,
"loss": 3.2238,
"theoretical_loss": 3.9210157723232424,
"tokens_seen": 492765184
},
{
"epoch": 0.15,
"learning_rate": 0.0004296489468405216,
"loss": 3.0512,
"theoretical_loss": 3.9209594673857,
"tokens_seen": 492830720
},
{
"epoch": 0.15,
"learning_rate": 0.0004296389167502507,
"loss": 2.9424,
"theoretical_loss": 3.920903172031161,
"tokens_seen": 492896256
},
{
"epoch": 0.15,
"learning_rate": 0.00042962888665997996,
"loss": 2.9471,
"theoretical_loss": 3.9208468862567205,
"tokens_seen": 492961792
},
{
"epoch": 0.15,
"learning_rate": 0.0004296188565697091,
"loss": 2.8456,
"theoretical_loss": 3.9207906100594756,
"tokens_seen": 493027328
},
{
"epoch": 0.15,
"learning_rate": 0.0004296088264794383,
"loss": 2.9321,
"theoretical_loss": 3.920734343436523,
"tokens_seen": 493092864
},
{
"epoch": 0.15,
"objective/train/docs_used": 289918,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.732017993927002,
"objective/train/theoretical_loss": 3.920678086384963,
"objective/train/tokens_used": 513618400,
"theoretical_loss": 3.920678086384963,
"tokens_seen": 493158400
},
{
"epoch": 0.15,
"learning_rate": 0.0004295987963891675,
"loss": 3.0366,
"theoretical_loss": 3.920678086384963,
"tokens_seen": 493158400
},
{
"epoch": 0.15,
"learning_rate": 0.0004295887662988967,
"loss": 3.0784,
"theoretical_loss": 3.9206218389018943,
"tokens_seen": 493223936
},
{
"epoch": 0.15,
"learning_rate": 0.00042957873620862587,
"loss": 2.7669,
"theoretical_loss": 3.9205656009844194,
"tokens_seen": 493289472
},
{
"epoch": 0.15,
"learning_rate": 0.0004295687061183551,
"loss": 3.008,
"theoretical_loss": 3.9205093726296405,
"tokens_seen": 493355008
},
{
"epoch": 0.15,
"learning_rate": 0.00042955867602808423,
"loss": 2.8343,
"theoretical_loss": 3.920453153834662,
"tokens_seen": 493420544
},
{
"epoch": 0.15,
"learning_rate": 0.00042954864593781347,
"loss": 2.6202,
"theoretical_loss": 3.9203969445965887,
"tokens_seen": 493486080
},
{
"epoch": 0.15,
"learning_rate": 0.0004295386158475426,
"loss": 3.15,
"theoretical_loss": 3.9203407449125276,
"tokens_seen": 493551616
},
{
"epoch": 0.15,
"learning_rate": 0.00042952858575727183,
"loss": 2.8284,
"theoretical_loss": 3.9202845547795864,
"tokens_seen": 493617152
},
{
"epoch": 0.15,
"learning_rate": 0.000429518555667001,
"loss": 2.7332,
"theoretical_loss": 3.920228374194874,
"tokens_seen": 493682688
},
{
"epoch": 0.15,
"learning_rate": 0.0004295085255767302,
"loss": 2.9725,
"theoretical_loss": 3.9201722031555004,
"tokens_seen": 493748224
},
{
"epoch": 0.15,
"learning_rate": 0.00042949849548645937,
"loss": 2.9607,
"theoretical_loss": 3.920116041658578,
"tokens_seen": 493813760
},
{
"epoch": 0.15,
"learning_rate": 0.00042948846539618855,
"loss": 2.7384,
"theoretical_loss": 3.9200598897012187,
"tokens_seen": 493879296
},
{
"epoch": 0.15,
"learning_rate": 0.00042947843530591773,
"loss": 2.9409,
"theoretical_loss": 3.9200037472805374,
"tokens_seen": 493944832
},
{
"epoch": 0.15,
"learning_rate": 0.00042946840521564697,
"loss": 2.9368,
"theoretical_loss": 3.9199476143936485,
"tokens_seen": 494010368
},
{
"epoch": 0.15,
"learning_rate": 0.0004294583751253761,
"loss": 2.8874,
"theoretical_loss": 3.919891491037669,
"tokens_seen": 494075904
},
{
"epoch": 0.15,
"learning_rate": 0.00042944834503510533,
"loss": 2.917,
"theoretical_loss": 3.9198353772097168,
"tokens_seen": 494141440
},
{
"epoch": 0.15,
"learning_rate": 0.0004294383149448345,
"loss": 2.8774,
"theoretical_loss": 3.9197792729069114,
"tokens_seen": 494206976
},
{
"epoch": 0.15,
"learning_rate": 0.0004294282848545637,
"loss": 2.9999,
"theoretical_loss": 3.9197231781263726,
"tokens_seen": 494272512
},
{
"epoch": 0.15,
"learning_rate": 0.00042941825476429293,
"loss": 3.0571,
"theoretical_loss": 3.919667092865222,
"tokens_seen": 494338048
},
{
"epoch": 0.15,
"learning_rate": 0.00042940822467402206,
"loss": 2.979,
"theoretical_loss": 3.919611017120583,
"tokens_seen": 494403584
},
{
"epoch": 0.15,
"learning_rate": 0.0004293981945837513,
"loss": 3.0109,
"theoretical_loss": 3.919554950889579,
"tokens_seen": 494469120
},
{
"epoch": 0.15,
"learning_rate": 0.00042938816449348047,
"loss": 2.9818,
"theoretical_loss": 3.9194988941693354,
"tokens_seen": 494534656
},
{
"epoch": 0.15,
"learning_rate": 0.00042937813440320965,
"loss": 2.922,
"theoretical_loss": 3.9194428469569793,
"tokens_seen": 494600192
},
{
"epoch": 0.15,
"learning_rate": 0.00042936810431293883,
"loss": 3.0647,
"theoretical_loss": 3.9193868092496382,
"tokens_seen": 494665728
},
{
"epoch": 0.15,
"learning_rate": 0.000429358074222668,
"loss": 2.971,
"theoretical_loss": 3.919330781044441,
"tokens_seen": 494731264
},
{
"epoch": 0.15,
"objective/train/docs_used": 291263,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1252217292785645,
"objective/train/theoretical_loss": 3.919274762338519,
"objective/train/tokens_used": 515256800,
"theoretical_loss": 3.919274762338519,
"tokens_seen": 494796800
},
{
"epoch": 0.15,
"learning_rate": 0.0004293480441323972,
"loss": 2.8186,
"theoretical_loss": 3.919274762338519,
"tokens_seen": 494796800
},
{
"epoch": 0.15,
"learning_rate": 0.00042933801404212643,
"loss": 3.0019,
"theoretical_loss": 3.9192187531290026,
"tokens_seen": 494862336
},
{
"epoch": 0.15,
"learning_rate": 0.00042932798395185556,
"loss": 2.8151,
"theoretical_loss": 3.9191627534130253,
"tokens_seen": 494927872
},
{
"epoch": 0.15,
"learning_rate": 0.0004293179538615848,
"loss": 2.7985,
"theoretical_loss": 3.9191067631877208,
"tokens_seen": 494993408
},
{
"epoch": 0.15,
"learning_rate": 0.0004293079237713139,
"loss": 3.0301,
"theoretical_loss": 3.919050782450225,
"tokens_seen": 495058944
},
{
"epoch": 0.15,
"learning_rate": 0.00042929789368104316,
"loss": 2.7674,
"theoretical_loss": 3.9189948111976736,
"tokens_seen": 495124480
},
{
"epoch": 0.15,
"learning_rate": 0.00042928786359077234,
"loss": 2.7445,
"theoretical_loss": 3.9189388494272057,
"tokens_seen": 495190016
},
{
"epoch": 0.15,
"learning_rate": 0.0004292778335005015,
"loss": 2.9838,
"theoretical_loss": 3.9188828971359584,
"tokens_seen": 495255552
},
{
"epoch": 0.15,
"learning_rate": 0.0004292678034102307,
"loss": 2.9377,
"theoretical_loss": 3.918826954321074,
"tokens_seen": 495321088
},
{
"epoch": 0.15,
"learning_rate": 0.00042925777331995993,
"loss": 2.9266,
"theoretical_loss": 3.918771020979693,
"tokens_seen": 495386624
},
{
"epoch": 0.15,
"learning_rate": 0.00042924774322968906,
"loss": 2.7544,
"theoretical_loss": 3.9187150971089584,
"tokens_seen": 495452160
},
{
"epoch": 0.15,
"learning_rate": 0.0004292377131394183,
"loss": 3.0238,
"theoretical_loss": 3.918659182706014,
"tokens_seen": 495517696
},
{
"epoch": 0.15,
"learning_rate": 0.0004292276830491474,
"loss": 2.9448,
"theoretical_loss": 3.9186032777680055,
"tokens_seen": 495583232
},
{
"epoch": 0.15,
"learning_rate": 0.00042921765295887666,
"loss": 3.021,
"theoretical_loss": 3.918547382292079,
"tokens_seen": 495648768
},
{
"epoch": 0.15,
"learning_rate": 0.00042920762286860584,
"loss": 2.8781,
"theoretical_loss": 3.918491496275383,
"tokens_seen": 495714304
},
{
"epoch": 0.15,
"learning_rate": 0.000429197592778335,
"loss": 2.764,
"theoretical_loss": 3.9184356197150647,
"tokens_seen": 495779840
},
{
"epoch": 0.15,
"learning_rate": 0.0004291875626880642,
"loss": 2.8178,
"theoretical_loss": 3.918379752608276,
"tokens_seen": 495845376
},
{
"epoch": 0.15,
"learning_rate": 0.0004291775325977934,
"loss": 3.175,
"theoretical_loss": 3.918323894952168,
"tokens_seen": 495910912
},
{
"epoch": 0.15,
"learning_rate": 0.00042916750250752256,
"loss": 3.0742,
"theoretical_loss": 3.9182680467438926,
"tokens_seen": 495976448
},
{
"epoch": 0.15,
"learning_rate": 0.0004291574724172518,
"loss": 3.1731,
"theoretical_loss": 3.9182122079806048,
"tokens_seen": 496041984
},
{
"epoch": 0.15,
"learning_rate": 0.0004291474423269809,
"loss": 2.8936,
"theoretical_loss": 3.918156378659459,
"tokens_seen": 496107520
},
{
"epoch": 0.15,
"learning_rate": 0.00042913741223671016,
"loss": 2.8979,
"theoretical_loss": 3.9181005587776117,
"tokens_seen": 496173056
},
{
"epoch": 0.15,
"learning_rate": 0.0004291273821464393,
"loss": 2.7485,
"theoretical_loss": 3.9180447483322203,
"tokens_seen": 496238592
},
{
"epoch": 0.15,
"learning_rate": 0.0004291173520561685,
"loss": 2.7746,
"theoretical_loss": 3.9179889473204446,
"tokens_seen": 496304128
},
{
"epoch": 0.15,
"learning_rate": 0.0004291073219658977,
"loss": 3.2393,
"theoretical_loss": 3.9179331557394432,
"tokens_seen": 496369664
},
{
"epoch": 0.15,
"objective/train/docs_used": 291574,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5555975437164307,
"objective/train/theoretical_loss": 3.9178773735863786,
"objective/train/tokens_used": 516895200,
"theoretical_loss": 3.9178773735863786,
"tokens_seen": 496435200
},
{
"epoch": 0.15,
"learning_rate": 0.0004290972918756269,
"loss": 2.9531,
"theoretical_loss": 3.9178773735863786,
"tokens_seen": 496435200
},
{
"epoch": 0.15,
"learning_rate": 0.00042908726178535607,
"loss": 3.0201,
"theoretical_loss": 3.917821600858413,
"tokens_seen": 496500736
},
{
"epoch": 0.15,
"learning_rate": 0.0004290772316950853,
"loss": 3.0068,
"theoretical_loss": 3.917765837552709,
"tokens_seen": 496566272
},
{
"epoch": 0.15,
"learning_rate": 0.00042906720160481443,
"loss": 2.9994,
"theoretical_loss": 3.917710083666434,
"tokens_seen": 496631808
},
{
"epoch": 0.15,
"learning_rate": 0.00042905717151454367,
"loss": 2.8756,
"theoretical_loss": 3.917654339196752,
"tokens_seen": 496697344
},
{
"epoch": 0.15,
"learning_rate": 0.0004290471414242728,
"loss": 3.1561,
"theoretical_loss": 3.9175986041408315,
"tokens_seen": 496762880
},
{
"epoch": 0.15,
"learning_rate": 0.00042903711133400203,
"loss": 3.0554,
"theoretical_loss": 3.9175428784958406,
"tokens_seen": 496828416
},
{
"epoch": 0.15,
"learning_rate": 0.0004290270812437312,
"loss": 2.9631,
"theoretical_loss": 3.9174871622589498,
"tokens_seen": 496893952
},
{
"epoch": 0.15,
"learning_rate": 0.0004290170511534604,
"loss": 3.1769,
"theoretical_loss": 3.91743145542733,
"tokens_seen": 496959488
},
{
"epoch": 0.15,
"learning_rate": 0.00042900702106318957,
"loss": 2.9962,
"theoretical_loss": 3.9173757579981534,
"tokens_seen": 497025024
},
{
"epoch": 0.15,
"learning_rate": 0.00042899699097291875,
"loss": 2.9916,
"theoretical_loss": 3.9173200699685937,
"tokens_seen": 497090560
},
{
"epoch": 0.15,
"learning_rate": 0.00042898696088264793,
"loss": 2.9978,
"theoretical_loss": 3.9172643913358254,
"tokens_seen": 497156096
},
{
"epoch": 0.15,
"learning_rate": 0.00042897693079237717,
"loss": 2.8487,
"theoretical_loss": 3.9172087220970244,
"tokens_seen": 497221632
},
{
"epoch": 0.15,
"learning_rate": 0.0004289669007021063,
"loss": 2.8799,
"theoretical_loss": 3.917153062249368,
"tokens_seen": 497287168
},
{
"epoch": 0.15,
"learning_rate": 0.00042895687061183553,
"loss": 2.991,
"theoretical_loss": 3.917097411790036,
"tokens_seen": 497352704
},
{
"epoch": 0.15,
"learning_rate": 0.0004289468405215647,
"loss": 2.7169,
"theoretical_loss": 3.9170417707162057,
"tokens_seen": 497418240
},
{
"epoch": 0.15,
"learning_rate": 0.0004289368104312939,
"loss": 2.988,
"theoretical_loss": 3.9169861390250595,
"tokens_seen": 497483776
},
{
"epoch": 0.15,
"learning_rate": 0.0004289267803410231,
"loss": 3.1274,
"theoretical_loss": 3.9169305167137796,
"tokens_seen": 497549312
},
{
"epoch": 0.15,
"learning_rate": 0.00042891675025075226,
"loss": 2.8603,
"theoretical_loss": 3.916874903779549,
"tokens_seen": 497614848
},
{
"epoch": 0.15,
"learning_rate": 0.00042890672016048144,
"loss": 3.3354,
"theoretical_loss": 3.9168193002195513,
"tokens_seen": 497680384
},
{
"epoch": 0.15,
"learning_rate": 0.00042889669007021067,
"loss": 3.0345,
"theoretical_loss": 3.9167637060309732,
"tokens_seen": 497745920
},
{
"epoch": 0.15,
"learning_rate": 0.0004288866599799398,
"loss": 2.9234,
"theoretical_loss": 3.916708121211002,
"tokens_seen": 497811456
},
{
"epoch": 0.15,
"learning_rate": 0.00042887662988966903,
"loss": 2.9288,
"theoretical_loss": 3.9166525457568246,
"tokens_seen": 497876992
},
{
"epoch": 0.15,
"learning_rate": 0.00042886659979939816,
"loss": 2.7802,
"theoretical_loss": 3.916596979665632,
"tokens_seen": 497942528
},
{
"epoch": 0.15,
"learning_rate": 0.0004288565697091274,
"loss": 3.1781,
"theoretical_loss": 3.9165414229346136,
"tokens_seen": 498008064
},
{
"epoch": 0.15,
"objective/train/docs_used": 293024,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.047236204147339,
"objective/train/theoretical_loss": 3.9164858755609613,
"objective/train/tokens_used": 518533600,
"theoretical_loss": 3.9164858755609613,
"tokens_seen": 498073600
},
{
"epoch": 0.15,
"learning_rate": 0.0004288465396188566,
"loss": 2.9409,
"theoretical_loss": 3.9164858755609613,
"tokens_seen": 498073600
},
{
"epoch": 0.15,
"learning_rate": 0.00042883650952858576,
"loss": 3.1637,
"theoretical_loss": 3.916430337541869,
"tokens_seen": 498139136
},
{
"epoch": 0.15,
"learning_rate": 0.00042882647943831494,
"loss": 3.1612,
"theoretical_loss": 3.9163748088745303,
"tokens_seen": 498204672
},
{
"epoch": 0.15,
"learning_rate": 0.0004288164493480441,
"loss": 3.1534,
"theoretical_loss": 3.9163192895561405,
"tokens_seen": 498270208
},
{
"epoch": 0.15,
"learning_rate": 0.0004288064192577733,
"loss": 3.0411,
"theoretical_loss": 3.916263779583896,
"tokens_seen": 498335744
},
{
"epoch": 0.15,
"learning_rate": 0.00042879638916750254,
"loss": 3.1491,
"theoretical_loss": 3.9162082789549952,
"tokens_seen": 498401280
},
{
"epoch": 0.15,
"learning_rate": 0.00042878635907723166,
"loss": 3.0577,
"theoretical_loss": 3.916152787666638,
"tokens_seen": 498466816
},
{
"epoch": 0.15,
"learning_rate": 0.0004287763289869609,
"loss": 2.9438,
"theoretical_loss": 3.916097305716023,
"tokens_seen": 498532352
},
{
"epoch": 0.15,
"learning_rate": 0.0004287662988966901,
"loss": 2.9937,
"theoretical_loss": 3.9160418331003526,
"tokens_seen": 498597888
},
{
"epoch": 0.15,
"learning_rate": 0.00042875626880641926,
"loss": 2.8485,
"theoretical_loss": 3.91598636981683,
"tokens_seen": 498663424
},
{
"epoch": 0.15,
"learning_rate": 0.00042874623871614844,
"loss": 2.9531,
"theoretical_loss": 3.9159309158626576,
"tokens_seen": 498728960
},
{
"epoch": 0.15,
"learning_rate": 0.0004287362086258776,
"loss": 3.048,
"theoretical_loss": 3.9158754712350414,
"tokens_seen": 498794496
},
{
"epoch": 0.15,
"learning_rate": 0.0004287261785356068,
"loss": 2.8666,
"theoretical_loss": 3.915820035931188,
"tokens_seen": 498860032
},
{
"epoch": 0.15,
"learning_rate": 0.00042871614844533604,
"loss": 3.0475,
"theoretical_loss": 3.9157646099483046,
"tokens_seen": 498925568
},
{
"epoch": 0.15,
"learning_rate": 0.00042870611835506517,
"loss": 2.6643,
"theoretical_loss": 3.9157091932836,
"tokens_seen": 498991104
},
{
"epoch": 0.15,
"learning_rate": 0.0004286960882647944,
"loss": 2.9152,
"theoretical_loss": 3.9156537859342837,
"tokens_seen": 499056640
},
{
"epoch": 0.15,
"learning_rate": 0.00042868605817452353,
"loss": 2.8303,
"theoretical_loss": 3.9155983878975675,
"tokens_seen": 499122176
},
{
"epoch": 0.15,
"learning_rate": 0.00042867602808425276,
"loss": 3.2016,
"theoretical_loss": 3.9155429991706634,
"tokens_seen": 499187712
},
{
"epoch": 0.15,
"learning_rate": 0.000428665997993982,
"loss": 3.1577,
"theoretical_loss": 3.9154876197507846,
"tokens_seen": 499253248
},
{
"epoch": 0.15,
"learning_rate": 0.0004286559679037111,
"loss": 3.148,
"theoretical_loss": 3.915432249635147,
"tokens_seen": 499318784
},
{
"epoch": 0.15,
"learning_rate": 0.00042864593781344036,
"loss": 2.9411,
"theoretical_loss": 3.9153768888209646,
"tokens_seen": 499384320
},
{
"epoch": 0.15,
"learning_rate": 0.0004286359077231695,
"loss": 3.1934,
"theoretical_loss": 3.915321537305456,
"tokens_seen": 499449856
},
{
"epoch": 0.15,
"learning_rate": 0.0004286258776328987,
"loss": 3.0267,
"theoretical_loss": 3.915266195085839,
"tokens_seen": 499515392
},
{
"epoch": 0.15,
"learning_rate": 0.0004286158475426279,
"loss": 2.9095,
"theoretical_loss": 3.9152108621593333,
"tokens_seen": 499580928
},
{
"epoch": 0.15,
"learning_rate": 0.0004286058174523571,
"loss": 3.3386,
"theoretical_loss": 3.91515553852316,
"tokens_seen": 499646464
},
{
"epoch": 0.15,
"objective/train/docs_used": 293514,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7597315311431885,
"objective/train/theoretical_loss": 3.91510022417454,
"objective/train/tokens_used": 520172000,
"theoretical_loss": 3.91510022417454,
"tokens_seen": 499712000
},
{
"epoch": 0.15,
"learning_rate": 0.00042859578736208627,
"loss": 3.0573,
"theoretical_loss": 3.91510022417454,
"tokens_seen": 499712000
},
{
"epoch": 0.15,
"learning_rate": 0.0004285857572718155,
"loss": 3.036,
"theoretical_loss": 3.915044919110697,
"tokens_seen": 499777536
},
{
"epoch": 0.15,
"learning_rate": 0.00042857572718154463,
"loss": 2.9206,
"theoretical_loss": 3.9149896233288555,
"tokens_seen": 499843072
},
{
"epoch": 0.15,
"learning_rate": 0.00042856569709127387,
"loss": 3.0611,
"theoretical_loss": 3.9149343368262404,
"tokens_seen": 499908608
},
{
"epoch": 0.15,
"learning_rate": 0.000428555667001003,
"loss": 2.884,
"theoretical_loss": 3.9148790596000786,
"tokens_seen": 499974144
},
{
"epoch": 0.15,
"learning_rate": 0.00042854563691073223,
"loss": 2.8042,
"theoretical_loss": 3.9148237916475983,
"tokens_seen": 500039680
},
{
"epoch": 0.15,
"learning_rate": 0.0004285356068204614,
"loss": 3.0384,
"theoretical_loss": 3.9147685329660282,
"tokens_seen": 500105216
},
{
"epoch": 0.15,
"learning_rate": 0.0004285255767301906,
"loss": 2.9233,
"theoretical_loss": 3.914713283552599,
"tokens_seen": 500170752
},
{
"epoch": 0.15,
"learning_rate": 0.00042851554663991977,
"loss": 3.1468,
"theoretical_loss": 3.9146580434045415,
"tokens_seen": 500236288
},
{
"epoch": 0.15,
"learning_rate": 0.00042850551654964895,
"loss": 3.0112,
"theoretical_loss": 3.914602812519089,
"tokens_seen": 500301824
},
{
"epoch": 0.15,
"learning_rate": 0.00042849548645937813,
"loss": 2.71,
"theoretical_loss": 3.9145475908934753,
"tokens_seen": 500367360
},
{
"epoch": 0.15,
"learning_rate": 0.00042848545636910737,
"loss": 3.0124,
"theoretical_loss": 3.914492378524934,
"tokens_seen": 500432896
},
{
"epoch": 0.15,
"learning_rate": 0.0004284754262788365,
"loss": 2.7896,
"theoretical_loss": 3.9144371754107032,
"tokens_seen": 500498432
},
{
"epoch": 0.15,
"learning_rate": 0.00042846539618856573,
"loss": 3.3801,
"theoretical_loss": 3.9143819815480194,
"tokens_seen": 500563968
},
{
"epoch": 0.15,
"learning_rate": 0.0004284553660982949,
"loss": 3.0366,
"theoretical_loss": 3.9143267969341218,
"tokens_seen": 500629504
},
{
"epoch": 0.15,
"learning_rate": 0.0004284453360080241,
"loss": 2.8845,
"theoretical_loss": 3.9142716215662494,
"tokens_seen": 500695040
},
{
"epoch": 0.15,
"learning_rate": 0.0004284353059177533,
"loss": 2.9944,
"theoretical_loss": 3.9142164554416428,
"tokens_seen": 500760576
},
{
"epoch": 0.15,
"learning_rate": 0.00042842527582748246,
"loss": 2.8842,
"theoretical_loss": 3.9141612985575454,
"tokens_seen": 500826112
},
{
"epoch": 0.15,
"learning_rate": 0.00042841524573721164,
"loss": 3.0182,
"theoretical_loss": 3.9141061509111994,
"tokens_seen": 500891648
},
{
"epoch": 0.15,
"learning_rate": 0.00042840521564694087,
"loss": 2.9012,
"theoretical_loss": 3.91405101249985,
"tokens_seen": 500957184
},
{
"epoch": 0.15,
"learning_rate": 0.00042839518555667,
"loss": 3.0928,
"theoretical_loss": 3.9139958833207427,
"tokens_seen": 501022720
},
{
"epoch": 0.15,
"learning_rate": 0.00042838515546639923,
"loss": 2.9251,
"theoretical_loss": 3.9139407633711243,
"tokens_seen": 501088256
},
{
"epoch": 0.15,
"learning_rate": 0.00042837512537612836,
"loss": 3.2046,
"theoretical_loss": 3.9138856526482426,
"tokens_seen": 501153792
},
{
"epoch": 0.15,
"learning_rate": 0.0004283650952858576,
"loss": 2.9961,
"theoretical_loss": 3.9138305511493474,
"tokens_seen": 501219328
},
{
"epoch": 0.15,
"learning_rate": 0.0004283550651955868,
"loss": 3.1201,
"theoretical_loss": 3.9137754588716884,
"tokens_seen": 501284864
},
{
"epoch": 0.15,
"objective/train/docs_used": 294722,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9437036514282227,
"objective/train/theoretical_loss": 3.9137203758125176,
"objective/train/tokens_used": 521810400,
"theoretical_loss": 3.9137203758125176,
"tokens_seen": 501350400
},
{
"epoch": 0.15,
"learning_rate": 0.00042834503510531596,
"loss": 2.9287,
"theoretical_loss": 3.9137203758125176,
"tokens_seen": 501350400
},
{
"epoch": 0.15,
"learning_rate": 0.00042833500501504514,
"loss": 3.0112,
"theoretical_loss": 3.913665301969088,
"tokens_seen": 501415936
},
{
"epoch": 0.15,
"learning_rate": 0.0004283249749247743,
"loss": 2.9346,
"theoretical_loss": 3.913610237338653,
"tokens_seen": 501481472
},
{
"epoch": 0.15,
"learning_rate": 0.0004283149448345035,
"loss": 2.9928,
"theoretical_loss": 3.9135551819184675,
"tokens_seen": 501547008
},
{
"epoch": 0.15,
"learning_rate": 0.00042830491474423274,
"loss": 2.9063,
"theoretical_loss": 3.913500135705789,
"tokens_seen": 501612544
},
{
"epoch": 0.15,
"learning_rate": 0.00042829488465396186,
"loss": 2.83,
"theoretical_loss": 3.9134450986978737,
"tokens_seen": 501678080
},
{
"epoch": 0.15,
"learning_rate": 0.0004282848545636911,
"loss": 2.7924,
"theoretical_loss": 3.913390070891981,
"tokens_seen": 501743616
},
{
"epoch": 0.15,
"learning_rate": 0.0004282748244734203,
"loss": 3.0962,
"theoretical_loss": 3.9133350522853703,
"tokens_seen": 501809152
},
{
"epoch": 0.15,
"learning_rate": 0.00042826479438314946,
"loss": 2.7333,
"theoretical_loss": 3.9132800428753027,
"tokens_seen": 501874688
},
{
"epoch": 0.15,
"learning_rate": 0.00042825476429287864,
"loss": 2.9528,
"theoretical_loss": 3.9132250426590405,
"tokens_seen": 501940224
},
{
"epoch": 0.15,
"learning_rate": 0.0004282447342026078,
"loss": 2.9171,
"theoretical_loss": 3.9131700516338475,
"tokens_seen": 502005760
},
{
"epoch": 0.15,
"learning_rate": 0.000428234704112337,
"loss": 3.0443,
"theoretical_loss": 3.9131150697969868,
"tokens_seen": 502071296
},
{
"epoch": 0.15,
"learning_rate": 0.00042822467402206624,
"loss": 3.1714,
"theoretical_loss": 3.913060097145725,
"tokens_seen": 502136832
},
{
"epoch": 0.15,
"learning_rate": 0.00042821464393179537,
"loss": 3.0249,
"theoretical_loss": 3.913005133677329,
"tokens_seen": 502202368
},
{
"epoch": 0.15,
"learning_rate": 0.0004282046138415246,
"loss": 2.9623,
"theoretical_loss": 3.9129501793890666,
"tokens_seen": 502267904
},
{
"epoch": 0.15,
"learning_rate": 0.00042819458375125373,
"loss": 2.9646,
"theoretical_loss": 3.9128952342782073,
"tokens_seen": 502333440
},
{
"epoch": 0.15,
"learning_rate": 0.00042818455366098296,
"loss": 2.9892,
"theoretical_loss": 3.9128402983420214,
"tokens_seen": 502398976
},
{
"epoch": 0.15,
"learning_rate": 0.00042817452357071215,
"loss": 2.7224,
"theoretical_loss": 3.91278537157778,
"tokens_seen": 502464512
},
{
"epoch": 0.15,
"learning_rate": 0.0004281644934804413,
"loss": 2.9322,
"theoretical_loss": 3.912730453982756,
"tokens_seen": 502530048
},
{
"epoch": 0.15,
"learning_rate": 0.0004281544633901705,
"loss": 2.8625,
"theoretical_loss": 3.9126755455542233,
"tokens_seen": 502595584
},
{
"epoch": 0.15,
"learning_rate": 0.0004281444332998997,
"loss": 2.7713,
"theoretical_loss": 3.912620646289457,
"tokens_seen": 502661120
},
{
"epoch": 0.15,
"learning_rate": 0.00042813440320962887,
"loss": 3.0138,
"theoretical_loss": 3.9125657561857334,
"tokens_seen": 502726656
},
{
"epoch": 0.15,
"learning_rate": 0.0004281243731193581,
"loss": 3.0685,
"theoretical_loss": 3.9125108752403293,
"tokens_seen": 502792192
},
{
"epoch": 0.15,
"learning_rate": 0.00042811434302908723,
"loss": 2.96,
"theoretical_loss": 3.9124560034505236,
"tokens_seen": 502857728
},
{
"epoch": 0.15,
"learning_rate": 0.00042810431293881647,
"loss": 2.7711,
"theoretical_loss": 3.9124011408135964,
"tokens_seen": 502923264
},
{
"epoch": 0.15,
"objective/train/docs_used": 295112,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0761241912841797,
"objective/train/theoretical_loss": 3.9123462873268275,
"objective/train/tokens_used": 523448800,
"theoretical_loss": 3.9123462873268275,
"tokens_seen": 502988800
},
{
"epoch": 0.15,
"learning_rate": 0.00042809428284854565,
"loss": 2.8803,
"theoretical_loss": 3.9123462873268275,
"tokens_seen": 502988800
},
{
"epoch": 0.15,
"learning_rate": 0.00042808425275827483,
"loss": 3.0619,
"theoretical_loss": 3.9122914429874998,
"tokens_seen": 503054336
},
{
"epoch": 0.15,
"learning_rate": 0.000428074222668004,
"loss": 2.8147,
"theoretical_loss": 3.912236607792896,
"tokens_seen": 503119872
},
{
"epoch": 0.15,
"learning_rate": 0.0004280641925777332,
"loss": 3.1577,
"theoretical_loss": 3.9121817817403004,
"tokens_seen": 503185408
},
{
"epoch": 0.15,
"learning_rate": 0.0004280541624874624,
"loss": 3.0249,
"theoretical_loss": 3.912126964826999,
"tokens_seen": 503250944
},
{
"epoch": 0.15,
"learning_rate": 0.0004280441323971916,
"loss": 2.7746,
"theoretical_loss": 3.912072157050278,
"tokens_seen": 503316480
},
{
"epoch": 0.15,
"learning_rate": 0.00042803410230692074,
"loss": 3.0252,
"theoretical_loss": 3.9120173584074256,
"tokens_seen": 503382016
},
{
"epoch": 0.15,
"learning_rate": 0.00042802407221664997,
"loss": 2.9913,
"theoretical_loss": 3.91196256889573,
"tokens_seen": 503447552
},
{
"epoch": 0.15,
"learning_rate": 0.0004280140421263791,
"loss": 2.9239,
"theoretical_loss": 3.911907788512482,
"tokens_seen": 503513088
},
{
"epoch": 0.15,
"learning_rate": 0.00042800401203610833,
"loss": 2.9242,
"theoretical_loss": 3.911853017254973,
"tokens_seen": 503578624
},
{
"epoch": 0.15,
"learning_rate": 0.0004279939819458375,
"loss": 3.0287,
"theoretical_loss": 3.9117982551204946,
"tokens_seen": 503644160
},
{
"epoch": 0.15,
"learning_rate": 0.0004279839518555667,
"loss": 3.0197,
"theoretical_loss": 3.911743502106341,
"tokens_seen": 503709696
},
{
"epoch": 0.15,
"learning_rate": 0.0004279739217652959,
"loss": 2.8153,
"theoretical_loss": 3.911688758209807,
"tokens_seen": 503775232
},
{
"epoch": 0.15,
"learning_rate": 0.0004279638916750251,
"loss": 2.93,
"theoretical_loss": 3.9116340234281886,
"tokens_seen": 503840768
},
{
"epoch": 0.15,
"learning_rate": 0.00042795386158475424,
"loss": 3.1179,
"theoretical_loss": 3.911579297758782,
"tokens_seen": 503906304
},
{
"epoch": 0.15,
"learning_rate": 0.0004279438314944835,
"loss": 2.7839,
"theoretical_loss": 3.9115245811988864,
"tokens_seen": 503971840
},
{
"epoch": 0.15,
"learning_rate": 0.0004279338014042126,
"loss": 2.991,
"theoretical_loss": 3.9114698737458005,
"tokens_seen": 504037376
},
{
"epoch": 0.15,
"learning_rate": 0.00042792377131394184,
"loss": 3.0795,
"theoretical_loss": 3.911415175396825,
"tokens_seen": 504102912
},
{
"epoch": 0.15,
"learning_rate": 0.00042791374122367107,
"loss": 2.9079,
"theoretical_loss": 3.911360486149262,
"tokens_seen": 504168448
},
{
"epoch": 0.15,
"learning_rate": 0.0004279037111334002,
"loss": 2.9344,
"theoretical_loss": 3.9113058060004136,
"tokens_seen": 504233984
},
{
"epoch": 0.15,
"learning_rate": 0.00042789368104312943,
"loss": 2.957,
"theoretical_loss": 3.9112511349475843,
"tokens_seen": 504299520
},
{
"epoch": 0.15,
"learning_rate": 0.00042788365095285856,
"loss": 2.9757,
"theoretical_loss": 3.911196472988079,
"tokens_seen": 504365056
},
{
"epoch": 0.15,
"learning_rate": 0.0004278736208625878,
"loss": 3.0768,
"theoretical_loss": 3.9111418201192034,
"tokens_seen": 504430592
},
{
"epoch": 0.15,
"learning_rate": 0.000427863590772317,
"loss": 2.698,
"theoretical_loss": 3.911087176338266,
"tokens_seen": 504496128
},
{
"epoch": 0.15,
"learning_rate": 0.00042785356068204616,
"loss": 3.0198,
"theoretical_loss": 3.911032541642575,
"tokens_seen": 504561664
},
{
"epoch": 0.15,
"objective/train/docs_used": 296352,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.776776075363159,
"objective/train/theoretical_loss": 3.910977916029439,
"objective/train/tokens_used": 525087200,
"theoretical_loss": 3.910977916029439,
"tokens_seen": 504627200
},
{
"epoch": 0.15,
"learning_rate": 0.00042784353059177534,
"loss": 2.8536,
"theoretical_loss": 3.910977916029439,
"tokens_seen": 504627200
},
{
"epoch": 0.15,
"learning_rate": 0.0004278335005015045,
"loss": 3.1104,
"theoretical_loss": 3.9109232994961705,
"tokens_seen": 504692736
},
{
"epoch": 0.15,
"learning_rate": 0.0004278234704112337,
"loss": 3.0926,
"theoretical_loss": 3.91086869204008,
"tokens_seen": 504758272
},
{
"epoch": 0.15,
"learning_rate": 0.00042781344032096294,
"loss": 3.0256,
"theoretical_loss": 3.910814093658482,
"tokens_seen": 504823808
},
{
"epoch": 0.15,
"learning_rate": 0.00042780341023069206,
"loss": 2.9688,
"theoretical_loss": 3.9107595043486905,
"tokens_seen": 504889344
},
{
"epoch": 0.15,
"learning_rate": 0.0004277933801404213,
"loss": 3.0209,
"theoretical_loss": 3.91070492410802,
"tokens_seen": 504954880
},
{
"epoch": 0.15,
"learning_rate": 0.0004277833500501505,
"loss": 2.9499,
"theoretical_loss": 3.9106503529337875,
"tokens_seen": 505020416
},
{
"epoch": 0.15,
"learning_rate": 0.00042777331995987966,
"loss": 3.0091,
"theoretical_loss": 3.9105957908233107,
"tokens_seen": 505085952
},
{
"epoch": 0.15,
"learning_rate": 0.00042776328986960884,
"loss": 2.6414,
"theoretical_loss": 3.910541237773909,
"tokens_seen": 505151488
},
{
"epoch": 0.15,
"learning_rate": 0.000427753259779338,
"loss": 3.005,
"theoretical_loss": 3.910486693782902,
"tokens_seen": 505217024
},
{
"epoch": 0.15,
"learning_rate": 0.0004277432296890672,
"loss": 2.9431,
"theoretical_loss": 3.9104321588476108,
"tokens_seen": 505282560
},
{
"epoch": 0.15,
"learning_rate": 0.00042773319959879644,
"loss": 3.1832,
"theoretical_loss": 3.910377632965357,
"tokens_seen": 505348096
},
{
"epoch": 0.15,
"learning_rate": 0.00042772316950852557,
"loss": 2.9311,
"theoretical_loss": 3.9103231161334655,
"tokens_seen": 505413632
},
{
"epoch": 0.15,
"learning_rate": 0.0004277131394182548,
"loss": 2.8145,
"theoretical_loss": 3.9102686083492593,
"tokens_seen": 505479168
},
{
"epoch": 0.15,
"learning_rate": 0.00042770310932798393,
"loss": 2.8171,
"theoretical_loss": 3.910214109610065,
"tokens_seen": 505544704
},
{
"epoch": 0.15,
"learning_rate": 0.00042769307923771316,
"loss": 2.9648,
"theoretical_loss": 3.910159619913209,
"tokens_seen": 505610240
},
{
"epoch": 0.15,
"learning_rate": 0.00042768304914744235,
"loss": 3.1846,
"theoretical_loss": 3.9101051392560198,
"tokens_seen": 505675776
},
{
"epoch": 0.15,
"learning_rate": 0.00042767301905717153,
"loss": 2.9035,
"theoretical_loss": 3.9100506676358266,
"tokens_seen": 505741312
},
{
"epoch": 0.15,
"learning_rate": 0.0004276629889669007,
"loss": 2.9588,
"theoretical_loss": 3.9099962050499584,
"tokens_seen": 505806848
},
{
"epoch": 0.15,
"learning_rate": 0.0004276529588766299,
"loss": 2.923,
"theoretical_loss": 3.9099417514957473,
"tokens_seen": 505872384
},
{
"epoch": 0.15,
"learning_rate": 0.00042764292878635907,
"loss": 2.8175,
"theoretical_loss": 3.9098873069705267,
"tokens_seen": 505937920
},
{
"epoch": 0.15,
"learning_rate": 0.0004276328986960883,
"loss": 2.9874,
"theoretical_loss": 3.9098328714716284,
"tokens_seen": 506003456
},
{
"epoch": 0.15,
"learning_rate": 0.00042762286860581743,
"loss": 3.1272,
"theoretical_loss": 3.909778444996389,
"tokens_seen": 506068992
},
{
"epoch": 0.15,
"learning_rate": 0.00042761283851554667,
"loss": 3.181,
"theoretical_loss": 3.9097240275421425,
"tokens_seen": 506134528
},
{
"epoch": 0.15,
"learning_rate": 0.00042760280842527585,
"loss": 2.7998,
"theoretical_loss": 3.9096696191062277,
"tokens_seen": 506200064
},
{
"epoch": 0.15,
"objective/train/docs_used": 297118,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.69783616065979,
"objective/train/theoretical_loss": 3.909615219685982,
"objective/train/tokens_used": 526725600,
"theoretical_loss": 3.909615219685982,
"tokens_seen": 506265600
},
{
"epoch": 0.15,
"learning_rate": 0.00042759277833500503,
"loss": 2.9313,
"theoretical_loss": 3.909615219685982,
"tokens_seen": 506265600
},
{
"epoch": 0.15,
"learning_rate": 0.0004275827482447342,
"loss": 2.841,
"theoretical_loss": 3.9095608292787443,
"tokens_seen": 506331136
},
{
"epoch": 0.15,
"learning_rate": 0.0004275727181544634,
"loss": 3.0147,
"theoretical_loss": 3.9095064478818555,
"tokens_seen": 506396672
},
{
"epoch": 0.15,
"learning_rate": 0.0004275626880641926,
"loss": 2.901,
"theoretical_loss": 3.9094520754926574,
"tokens_seen": 506462208
},
{
"epoch": 0.15,
"learning_rate": 0.0004275526579739218,
"loss": 2.8451,
"theoretical_loss": 3.9093977121084915,
"tokens_seen": 506527744
},
{
"epoch": 0.15,
"learning_rate": 0.00042754262788365094,
"loss": 2.9748,
"theoretical_loss": 3.909343357726703,
"tokens_seen": 506593280
},
{
"epoch": 0.15,
"learning_rate": 0.00042753259779338017,
"loss": 2.9838,
"theoretical_loss": 3.9092890123446367,
"tokens_seen": 506658816
},
{
"epoch": 0.15,
"learning_rate": 0.0004275225677031093,
"loss": 2.9351,
"theoretical_loss": 3.9092346759596373,
"tokens_seen": 506724352
},
{
"epoch": 0.15,
"learning_rate": 0.00042751253761283853,
"loss": 2.9645,
"theoretical_loss": 3.9091803485690537,
"tokens_seen": 506789888
},
{
"epoch": 0.15,
"learning_rate": 0.0004275025075225677,
"loss": 3.053,
"theoretical_loss": 3.9091260301702326,
"tokens_seen": 506855424
},
{
"epoch": 0.15,
"learning_rate": 0.0004274924774322969,
"loss": 3.0622,
"theoretical_loss": 3.9090717207605246,
"tokens_seen": 506920960
},
{
"epoch": 0.15,
"learning_rate": 0.0004274824473420261,
"loss": 3.0204,
"theoretical_loss": 3.90901742033728,
"tokens_seen": 506986496
},
{
"epoch": 0.15,
"learning_rate": 0.0004274724172517553,
"loss": 2.7641,
"theoretical_loss": 3.9089631288978497,
"tokens_seen": 507052032
},
{
"epoch": 0.15,
"learning_rate": 0.00042746238716148444,
"loss": 3.313,
"theoretical_loss": 3.9089088464395876,
"tokens_seen": 507117568
},
{
"epoch": 0.15,
"learning_rate": 0.0004274523570712137,
"loss": 2.8969,
"theoretical_loss": 3.908854572959847,
"tokens_seen": 507183104
},
{
"epoch": 0.15,
"learning_rate": 0.0004274423269809428,
"loss": 3.2916,
"theoretical_loss": 3.9088003084559833,
"tokens_seen": 507248640
},
{
"epoch": 0.15,
"learning_rate": 0.00042743229689067204,
"loss": 2.7171,
"theoretical_loss": 3.908746052925352,
"tokens_seen": 507314176
},
{
"epoch": 0.15,
"learning_rate": 0.0004274222668004012,
"loss": 2.7268,
"theoretical_loss": 3.9086918063653115,
"tokens_seen": 507379712
},
{
"epoch": 0.15,
"learning_rate": 0.0004274122367101304,
"loss": 3.1031,
"theoretical_loss": 3.908637568773219,
"tokens_seen": 507445248
},
{
"epoch": 0.15,
"learning_rate": 0.0004274022066198596,
"loss": 2.8652,
"theoretical_loss": 3.9085833401464347,
"tokens_seen": 507510784
},
{
"epoch": 0.15,
"learning_rate": 0.00042739217652958876,
"loss": 2.9007,
"theoretical_loss": 3.908529120482319,
"tokens_seen": 507576320
},
{
"epoch": 0.15,
"learning_rate": 0.00042738214643931794,
"loss": 2.8738,
"theoretical_loss": 3.908474909778234,
"tokens_seen": 507641856
},
{
"epoch": 0.15,
"learning_rate": 0.0004273721163490472,
"loss": 3.0197,
"theoretical_loss": 3.908420708031542,
"tokens_seen": 507707392
},
{
"epoch": 0.15,
"learning_rate": 0.0004273620862587763,
"loss": 2.8471,
"theoretical_loss": 3.9083665152396074,
"tokens_seen": 507772928
},
{
"epoch": 0.15,
"learning_rate": 0.00042735205616850554,
"loss": 2.9021,
"theoretical_loss": 3.908312331399795,
"tokens_seen": 507838464
},
{
"epoch": 0.15,
"objective/train/docs_used": 298304,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.079557418823242,
"objective/train/theoretical_loss": 3.908258156509472,
"objective/train/tokens_used": 528364000,
"theoretical_loss": 3.908258156509472,
"tokens_seen": 507904000
},
{
"epoch": 0.15,
"learning_rate": 0.00042734202607823467,
"loss": 3.1514,
"theoretical_loss": 3.908258156509472,
"tokens_seen": 507904000
},
{
"epoch": 0.15,
"learning_rate": 0.0004273319959879639,
"loss": 2.9263,
"theoretical_loss": 3.9082039905660046,
"tokens_seen": 507969536
},
{
"epoch": 0.15,
"learning_rate": 0.0004273219658976931,
"loss": 2.7937,
"theoretical_loss": 3.9081498335667613,
"tokens_seen": 508035072
},
{
"epoch": 0.15,
"learning_rate": 0.00042731193580742226,
"loss": 3.133,
"theoretical_loss": 3.908095685509112,
"tokens_seen": 508100608
},
{
"epoch": 0.15,
"learning_rate": 0.00042730190571715145,
"loss": 2.9235,
"theoretical_loss": 3.908041546390428,
"tokens_seen": 508166144
},
{
"epoch": 0.15,
"learning_rate": 0.0004272918756268807,
"loss": 2.9784,
"theoretical_loss": 3.9079874162080794,
"tokens_seen": 508231680
},
{
"epoch": 0.15,
"learning_rate": 0.0004272818455366098,
"loss": 3.1766,
"theoretical_loss": 3.9079332949594408,
"tokens_seen": 508297216
},
{
"epoch": 0.15,
"learning_rate": 0.00042727181544633904,
"loss": 3.2091,
"theoretical_loss": 3.907879182641886,
"tokens_seen": 508362752
},
{
"epoch": 0.15,
"learning_rate": 0.00042726178535606817,
"loss": 2.9524,
"theoretical_loss": 3.907825079252789,
"tokens_seen": 508428288
},
{
"epoch": 0.15,
"learning_rate": 0.0004272517552657974,
"loss": 2.9111,
"theoretical_loss": 3.9077709847895266,
"tokens_seen": 508493824
},
{
"epoch": 0.15,
"learning_rate": 0.0004272417251755266,
"loss": 2.9842,
"theoretical_loss": 3.907716899249477,
"tokens_seen": 508559360
},
{
"epoch": 0.15,
"learning_rate": 0.00042723169508525577,
"loss": 3.1073,
"theoretical_loss": 3.9076628226300167,
"tokens_seen": 508624896
},
{
"epoch": 0.15,
"learning_rate": 0.00042722166499498495,
"loss": 2.9411,
"theoretical_loss": 3.9076087549285274,
"tokens_seen": 508690432
},
{
"epoch": 0.15,
"learning_rate": 0.00042721163490471413,
"loss": 3.1312,
"theoretical_loss": 3.907554696142389,
"tokens_seen": 508755968
},
{
"epoch": 0.15,
"learning_rate": 0.0004272016048144433,
"loss": 3.1056,
"theoretical_loss": 3.9075006462689825,
"tokens_seen": 508821504
},
{
"epoch": 0.15,
"learning_rate": 0.00042719157472417255,
"loss": 3.0852,
"theoretical_loss": 3.9074466053056915,
"tokens_seen": 508887040
},
{
"epoch": 0.15,
"learning_rate": 0.0004271815446339017,
"loss": 3.1324,
"theoretical_loss": 3.9073925732498997,
"tokens_seen": 508952576
},
{
"epoch": 0.15,
"learning_rate": 0.0004271715145436309,
"loss": 2.8519,
"theoretical_loss": 3.907338550098993,
"tokens_seen": 509018112
},
{
"epoch": 0.15,
"learning_rate": 0.0004271614844533601,
"loss": 3.0767,
"theoretical_loss": 3.907284535850356,
"tokens_seen": 509083648
},
{
"epoch": 0.15,
"learning_rate": 0.00042715145436308927,
"loss": 3.1534,
"theoretical_loss": 3.9072305305013773,
"tokens_seen": 509149184
},
{
"epoch": 0.15,
"learning_rate": 0.0004271414242728185,
"loss": 2.9208,
"theoretical_loss": 3.9071765340494453,
"tokens_seen": 509214720
},
{
"epoch": 0.15,
"learning_rate": 0.00042713139418254763,
"loss": 3.0471,
"theoretical_loss": 3.9071225464919483,
"tokens_seen": 509280256
},
{
"epoch": 0.15,
"learning_rate": 0.00042712136409227687,
"loss": 2.9369,
"theoretical_loss": 3.907068567826278,
"tokens_seen": 509345792
},
{
"epoch": 0.15,
"learning_rate": 0.00042711133400200605,
"loss": 3.0517,
"theoretical_loss": 3.9070145980498263,
"tokens_seen": 509411328
},
{
"epoch": 0.15,
"learning_rate": 0.00042710130391173523,
"loss": 2.983,
"theoretical_loss": 3.906960637159985,
"tokens_seen": 509476864
},
{
"epoch": 0.15,
"objective/train/docs_used": 298986,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2158470153808594,
"objective/train/theoretical_loss": 3.9069066851541487,
"objective/train/tokens_used": 530002400,
"theoretical_loss": 3.9069066851541487,
"tokens_seen": 509542400
},
{
"epoch": 0.15,
"learning_rate": 0.0004270912738214644,
"loss": 3.0958,
"theoretical_loss": 3.9069066851541487,
"tokens_seen": 509542400
},
{
"epoch": 0.15,
"learning_rate": 0.0004270812437311936,
"loss": 2.8435,
"theoretical_loss": 3.906852742029712,
"tokens_seen": 509607936
},
{
"epoch": 0.15,
"learning_rate": 0.0004270712136409228,
"loss": 3.1957,
"theoretical_loss": 3.906798807784072,
"tokens_seen": 509673472
},
{
"epoch": 0.15,
"learning_rate": 0.000427061183550652,
"loss": 3.0716,
"theoretical_loss": 3.9067448824146243,
"tokens_seen": 509739008
},
{
"epoch": 0.15,
"learning_rate": 0.00042705115346038114,
"loss": 2.9525,
"theoretical_loss": 3.9066909659187683,
"tokens_seen": 509804544
},
{
"epoch": 0.15,
"learning_rate": 0.00042704112337011037,
"loss": 2.8851,
"theoretical_loss": 3.906637058293903,
"tokens_seen": 509870080
},
{
"epoch": 0.15,
"learning_rate": 0.0004270310932798395,
"loss": 3.1575,
"theoretical_loss": 3.906583159537429,
"tokens_seen": 509935616
},
{
"epoch": 0.15,
"learning_rate": 0.00042702106318956873,
"loss": 2.9141,
"theoretical_loss": 3.9065292696467475,
"tokens_seen": 510001152
},
{
"epoch": 0.15,
"learning_rate": 0.0004270110330992979,
"loss": 3.2035,
"theoretical_loss": 3.906475388619262,
"tokens_seen": 510066688
},
{
"epoch": 0.15,
"learning_rate": 0.0004270010030090271,
"loss": 2.7468,
"theoretical_loss": 3.906421516452376,
"tokens_seen": 510132224
},
{
"epoch": 0.15,
"learning_rate": 0.0004269909729187563,
"loss": 3.0385,
"theoretical_loss": 3.9063676531434934,
"tokens_seen": 510197760
},
{
"epoch": 0.15,
"learning_rate": 0.0004269809428284855,
"loss": 2.796,
"theoretical_loss": 3.9063137986900216,
"tokens_seen": 510263296
},
{
"epoch": 0.15,
"learning_rate": 0.00042697091273821464,
"loss": 3.0383,
"theoretical_loss": 3.906259953089367,
"tokens_seen": 510328832
},
{
"epoch": 0.15,
"learning_rate": 0.0004269608826479439,
"loss": 2.6359,
"theoretical_loss": 3.9062061163389377,
"tokens_seen": 510394368
},
{
"epoch": 0.15,
"learning_rate": 0.000426950852557673,
"loss": 2.7839,
"theoretical_loss": 3.906152288436143,
"tokens_seen": 510459904
},
{
"epoch": 0.15,
"learning_rate": 0.00042694082246740224,
"loss": 2.876,
"theoretical_loss": 3.906098469378393,
"tokens_seen": 510525440
},
{
"epoch": 0.15,
"learning_rate": 0.0004269307923771314,
"loss": 2.8037,
"theoretical_loss": 3.9060446591630997,
"tokens_seen": 510590976
},
{
"epoch": 0.15,
"learning_rate": 0.0004269207622868606,
"loss": 2.8069,
"theoretical_loss": 3.905990857787675,
"tokens_seen": 510656512
},
{
"epoch": 0.15,
"learning_rate": 0.0004269107321965898,
"loss": 2.7768,
"theoretical_loss": 3.9059370652495327,
"tokens_seen": 510722048
},
{
"epoch": 0.15,
"learning_rate": 0.00042690070210631896,
"loss": 3.1726,
"theoretical_loss": 3.905883281546088,
"tokens_seen": 510787584
},
{
"epoch": 0.15,
"learning_rate": 0.00042689067201604814,
"loss": 3.0302,
"theoretical_loss": 3.9058295066747553,
"tokens_seen": 510853120
},
{
"epoch": 0.15,
"learning_rate": 0.0004268806419257774,
"loss": 3.133,
"theoretical_loss": 3.9057757406329534,
"tokens_seen": 510918656
},
{
"epoch": 0.15,
"learning_rate": 0.0004268706118355065,
"loss": 3.2004,
"theoretical_loss": 3.9057219834180987,
"tokens_seen": 510984192
},
{
"epoch": 0.15,
"learning_rate": 0.00042686058174523574,
"loss": 2.9723,
"theoretical_loss": 3.9056682350276106,
"tokens_seen": 511049728
},
{
"epoch": 0.15,
"learning_rate": 0.00042685055165496487,
"loss": 3.0879,
"theoretical_loss": 3.90561449545891,
"tokens_seen": 511115264
},
{
"epoch": 0.15,
"objective/train/docs_used": 299996,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.871511697769165,
"objective/train/theoretical_loss": 3.905560764709417,
"objective/train/tokens_used": 531640800,
"theoretical_loss": 3.905560764709417,
"tokens_seen": 511180800
},
{
"epoch": 0.15,
"learning_rate": 0.0004268405215646941,
"loss": 2.808,
"theoretical_loss": 3.905560764709417,
"tokens_seen": 511180800
},
{
"epoch": 0.15,
"learning_rate": 0.0004268304914744233,
"loss": 3.1141,
"theoretical_loss": 3.905507042776554,
"tokens_seen": 511246336
},
{
"epoch": 0.15,
"learning_rate": 0.00042682046138415246,
"loss": 3.089,
"theoretical_loss": 3.905453329657745,
"tokens_seen": 511311872
},
{
"epoch": 0.15,
"learning_rate": 0.00042681043129388165,
"loss": 3.1051,
"theoretical_loss": 3.905399625350414,
"tokens_seen": 511377408
},
{
"epoch": 0.15,
"learning_rate": 0.0004268004012036109,
"loss": 2.9046,
"theoretical_loss": 3.9053459298519875,
"tokens_seen": 511442944
},
{
"epoch": 0.16,
"learning_rate": 0.00042679037111334,
"loss": 2.8459,
"theoretical_loss": 3.905292243159891,
"tokens_seen": 511508480
},
{
"epoch": 0.16,
"learning_rate": 0.00042678034102306924,
"loss": 2.867,
"theoretical_loss": 3.905238565271553,
"tokens_seen": 511574016
},
{
"epoch": 0.16,
"learning_rate": 0.00042677031093279837,
"loss": 2.9912,
"theoretical_loss": 3.9051848961844007,
"tokens_seen": 511639552
},
{
"epoch": 0.16,
"learning_rate": 0.0004267602808425276,
"loss": 3.0169,
"theoretical_loss": 3.9051312358958663,
"tokens_seen": 511705088
},
{
"epoch": 0.16,
"learning_rate": 0.0004267502507522568,
"loss": 2.7919,
"theoretical_loss": 3.9050775844033785,
"tokens_seen": 511770624
},
{
"epoch": 0.16,
"learning_rate": 0.00042674022066198597,
"loss": 3.0171,
"theoretical_loss": 3.905023941704371,
"tokens_seen": 511836160
},
{
"epoch": 0.16,
"learning_rate": 0.00042673019057171515,
"loss": 2.8172,
"theoretical_loss": 3.904970307796276,
"tokens_seen": 511901696
},
{
"epoch": 0.16,
"learning_rate": 0.00042672016048144433,
"loss": 2.803,
"theoretical_loss": 3.9049166826765283,
"tokens_seen": 511967232
},
{
"epoch": 0.16,
"learning_rate": 0.0004267101303911735,
"loss": 3.141,
"theoretical_loss": 3.9048630663425623,
"tokens_seen": 512032768
},
{
"epoch": 0.16,
"learning_rate": 0.00042670010030090275,
"loss": 3.0635,
"theoretical_loss": 3.9048094587918154,
"tokens_seen": 512098304
},
{
"epoch": 0.16,
"learning_rate": 0.0004266900702106319,
"loss": 2.8628,
"theoretical_loss": 3.904755860021724,
"tokens_seen": 512163840
},
{
"epoch": 0.16,
"learning_rate": 0.0004266800401203611,
"loss": 2.7707,
"theoretical_loss": 3.904702270029727,
"tokens_seen": 512229376
},
{
"epoch": 0.16,
"learning_rate": 0.00042667001003009024,
"loss": 2.894,
"theoretical_loss": 3.904648688813264,
"tokens_seen": 512294912
},
{
"epoch": 0.16,
"learning_rate": 0.00042665997993981947,
"loss": 2.9793,
"theoretical_loss": 3.9045951163697756,
"tokens_seen": 512360448
},
{
"epoch": 0.16,
"learning_rate": 0.00042664994984954865,
"loss": 3.0136,
"theoretical_loss": 3.9045415526967036,
"tokens_seen": 512425984
},
{
"epoch": 0.16,
"learning_rate": 0.00042663991975927783,
"loss": 3.1039,
"theoretical_loss": 3.904487997791491,
"tokens_seen": 512491520
},
{
"epoch": 0.16,
"learning_rate": 0.000426629889669007,
"loss": 2.6166,
"theoretical_loss": 3.904434451651581,
"tokens_seen": 512557056
},
{
"epoch": 0.16,
"learning_rate": 0.00042661985957873625,
"loss": 2.8251,
"theoretical_loss": 3.904380914274419,
"tokens_seen": 512622592
},
{
"epoch": 0.16,
"learning_rate": 0.0004266098294884654,
"loss": 2.9569,
"theoretical_loss": 3.9043273856574507,
"tokens_seen": 512688128
},
{
"epoch": 0.16,
"learning_rate": 0.0004265997993981946,
"loss": 2.8449,
"theoretical_loss": 3.9042738657981237,
"tokens_seen": 512753664
},
{
"epoch": 0.16,
"objective/train/docs_used": 300643,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9767351150512695,
"objective/train/theoretical_loss": 3.9042203546938854,
"objective/train/tokens_used": 533279200,
"theoretical_loss": 3.9042203546938854,
"tokens_seen": 512819200
},
{
"epoch": 0.16,
"learning_rate": 0.00042658976930792374,
"loss": 2.9345,
"theoretical_loss": 3.9042203546938854,
"tokens_seen": 512819200
},
{
"epoch": 0.16,
"learning_rate": 0.000426579739217653,
"loss": 2.6117,
"theoretical_loss": 3.9041668523421853,
"tokens_seen": 512884736
},
{
"epoch": 0.16,
"learning_rate": 0.00042656970912738216,
"loss": 2.9333,
"theoretical_loss": 3.9041133587404744,
"tokens_seen": 512950272
},
{
"epoch": 0.16,
"learning_rate": 0.00042655967903711134,
"loss": 3.0242,
"theoretical_loss": 3.9040598738862027,
"tokens_seen": 513015808
},
{
"epoch": 0.16,
"learning_rate": 0.0004265496489468405,
"loss": 3.1475,
"theoretical_loss": 3.904006397776823,
"tokens_seen": 513081344
},
{
"epoch": 0.16,
"learning_rate": 0.0004265396188565697,
"loss": 3.0377,
"theoretical_loss": 3.90395293040979,
"tokens_seen": 513146880
},
{
"epoch": 0.16,
"learning_rate": 0.0004265295887662989,
"loss": 3.0351,
"theoretical_loss": 3.903899471782557,
"tokens_seen": 513212416
},
{
"epoch": 0.16,
"learning_rate": 0.0004265195586760281,
"loss": 3.0313,
"theoretical_loss": 3.90384602189258,
"tokens_seen": 513277952
},
{
"epoch": 0.16,
"learning_rate": 0.00042650952858575724,
"loss": 2.949,
"theoretical_loss": 3.9037925807373153,
"tokens_seen": 513343488
},
{
"epoch": 0.16,
"learning_rate": 0.0004264994984954865,
"loss": 3.159,
"theoretical_loss": 3.9037391483142208,
"tokens_seen": 513409024
},
{
"epoch": 0.16,
"learning_rate": 0.0004264894684052156,
"loss": 3.0819,
"theoretical_loss": 3.9036857246207557,
"tokens_seen": 513474560
},
{
"epoch": 0.16,
"learning_rate": 0.00042647943831494484,
"loss": 2.8282,
"theoretical_loss": 3.9036323096543795,
"tokens_seen": 513540096
},
{
"epoch": 0.16,
"learning_rate": 0.000426469408224674,
"loss": 3.0349,
"theoretical_loss": 3.903578903412553,
"tokens_seen": 513605632
},
{
"epoch": 0.16,
"learning_rate": 0.0004264593781344032,
"loss": 2.7911,
"theoretical_loss": 3.9035255058927394,
"tokens_seen": 513671168
},
{
"epoch": 0.16,
"learning_rate": 0.0004264493480441324,
"loss": 3.1141,
"theoretical_loss": 3.9034721170924005,
"tokens_seen": 513736704
},
{
"epoch": 0.16,
"learning_rate": 0.0004264393179538616,
"loss": 2.9924,
"theoretical_loss": 3.903418737009,
"tokens_seen": 513802240
},
{
"epoch": 0.16,
"learning_rate": 0.00042642928786359075,
"loss": 2.9924,
"theoretical_loss": 3.9033653656400045,
"tokens_seen": 513867776
},
{
"epoch": 0.16,
"learning_rate": 0.00042641925777332,
"loss": 2.8468,
"theoretical_loss": 3.903312002982879,
"tokens_seen": 513933312
},
{
"epoch": 0.16,
"learning_rate": 0.00042640922768304916,
"loss": 2.8712,
"theoretical_loss": 3.9032586490350916,
"tokens_seen": 513998848
},
{
"epoch": 0.16,
"learning_rate": 0.00042639919759277834,
"loss": 3.0535,
"theoretical_loss": 3.9032053037941106,
"tokens_seen": 514064384
},
{
"epoch": 0.16,
"learning_rate": 0.0004263891675025076,
"loss": 3.075,
"theoretical_loss": 3.9031519672574047,
"tokens_seen": 514129920
},
{
"epoch": 0.16,
"learning_rate": 0.0004263791374122367,
"loss": 2.8464,
"theoretical_loss": 3.903098639422445,
"tokens_seen": 514195456
},
{
"epoch": 0.16,
"learning_rate": 0.00042636910732196594,
"loss": 2.9903,
"theoretical_loss": 3.9030453202867035,
"tokens_seen": 514260992
},
{
"epoch": 0.16,
"learning_rate": 0.00042635907723169507,
"loss": 2.8887,
"theoretical_loss": 3.902992009847652,
"tokens_seen": 514326528
},
{
"epoch": 0.16,
"learning_rate": 0.0004263490471414243,
"loss": 2.8372,
"theoretical_loss": 3.9029387081027638,
"tokens_seen": 514392064
},
{
"epoch": 0.16,
"objective/train/docs_used": 301818,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.592280864715576,
"objective/train/theoretical_loss": 3.9028854150495143,
"objective/train/tokens_used": 534917600,
"theoretical_loss": 3.9028854150495143,
"tokens_seen": 514457600
},
{
"epoch": 0.16,
"learning_rate": 0.0004263390170511535,
"loss": 2.7809,
"theoretical_loss": 3.9028854150495143,
"tokens_seen": 514457600
},
{
"epoch": 0.16,
"learning_rate": 0.00042632898696088266,
"loss": 2.8517,
"theoretical_loss": 3.9028321306853795,
"tokens_seen": 514523136
},
{
"epoch": 0.16,
"learning_rate": 0.00042631895687061185,
"loss": 2.9973,
"theoretical_loss": 3.9027788550078357,
"tokens_seen": 514588672
},
{
"epoch": 0.16,
"learning_rate": 0.0004263089267803411,
"loss": 2.901,
"theoretical_loss": 3.902725588014361,
"tokens_seen": 514654208
},
{
"epoch": 0.16,
"learning_rate": 0.0004262988966900702,
"loss": 3.0018,
"theoretical_loss": 3.9026723297024337,
"tokens_seen": 514719744
},
{
"epoch": 0.16,
"learning_rate": 0.00042628886659979944,
"loss": 2.9766,
"theoretical_loss": 3.9026190800695346,
"tokens_seen": 514785280
},
{
"epoch": 0.16,
"learning_rate": 0.00042627883650952857,
"loss": 2.8017,
"theoretical_loss": 3.9025658391131444,
"tokens_seen": 514850816
},
{
"epoch": 0.16,
"learning_rate": 0.0004262688064192578,
"loss": 2.7732,
"theoretical_loss": 3.902512606830745,
"tokens_seen": 514916352
},
{
"epoch": 0.16,
"learning_rate": 0.000426258776328987,
"loss": 2.7042,
"theoretical_loss": 3.9024593832198198,
"tokens_seen": 514981888
},
{
"epoch": 0.16,
"learning_rate": 0.00042624874623871617,
"loss": 2.6781,
"theoretical_loss": 3.902406168277853,
"tokens_seen": 515047424
},
{
"epoch": 0.16,
"learning_rate": 0.00042623871614844535,
"loss": 2.9133,
"theoretical_loss": 3.9023529620023294,
"tokens_seen": 515112960
},
{
"epoch": 0.16,
"learning_rate": 0.00042622868605817453,
"loss": 2.7778,
"theoretical_loss": 3.902299764390736,
"tokens_seen": 515178496
},
{
"epoch": 0.16,
"learning_rate": 0.0004262186559679037,
"loss": 3.1145,
"theoretical_loss": 3.902246575440559,
"tokens_seen": 515244032
},
{
"epoch": 0.16,
"learning_rate": 0.00042620862587763295,
"loss": 2.8735,
"theoretical_loss": 3.902193395149288,
"tokens_seen": 515309568
},
{
"epoch": 0.16,
"learning_rate": 0.0004261985957873621,
"loss": 2.6957,
"theoretical_loss": 3.9021402235144116,
"tokens_seen": 515375104
},
{
"epoch": 0.16,
"learning_rate": 0.0004261885656970913,
"loss": 2.7606,
"theoretical_loss": 3.9020870605334204,
"tokens_seen": 515440640
},
{
"epoch": 0.16,
"learning_rate": 0.00042617853560682044,
"loss": 2.9639,
"theoretical_loss": 3.9020339062038065,
"tokens_seen": 515506176
},
{
"epoch": 0.16,
"learning_rate": 0.00042616850551654967,
"loss": 2.9819,
"theoretical_loss": 3.9019807605230614,
"tokens_seen": 515571712
},
{
"epoch": 0.16,
"learning_rate": 0.00042615847542627885,
"loss": 2.827,
"theoretical_loss": 3.90192762348868,
"tokens_seen": 515637248
},
{
"epoch": 0.16,
"learning_rate": 0.00042614844533600803,
"loss": 3.0162,
"theoretical_loss": 3.9018744950981556,
"tokens_seen": 515702784
},
{
"epoch": 0.16,
"learning_rate": 0.0004261384152457372,
"loss": 3.0663,
"theoretical_loss": 3.901821375348985,
"tokens_seen": 515768320
},
{
"epoch": 0.16,
"learning_rate": 0.00042612838515546645,
"loss": 2.9748,
"theoretical_loss": 3.9017682642386644,
"tokens_seen": 515833856
},
{
"epoch": 0.16,
"learning_rate": 0.0004261183550651956,
"loss": 3.0898,
"theoretical_loss": 3.9017151617646917,
"tokens_seen": 515899392
},
{
"epoch": 0.16,
"learning_rate": 0.0004261083249749248,
"loss": 2.8773,
"theoretical_loss": 3.901662067924566,
"tokens_seen": 515964928
},
{
"epoch": 0.16,
"learning_rate": 0.00042609829488465394,
"loss": 2.8027,
"theoretical_loss": 3.9016089827157865,
"tokens_seen": 516030464
},
{
"epoch": 0.16,
"objective/train/docs_used": 302480,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9364280700683594,
"objective/train/theoretical_loss": 3.9015559061358545,
"objective/train/tokens_used": 536556000,
"theoretical_loss": 3.9015559061358545,
"tokens_seen": 516096000
},
{
"epoch": 0.16,
"learning_rate": 0.0004260882647943832,
"loss": 2.9585,
"theoretical_loss": 3.9015559061358545,
"tokens_seen": 516096000
},
{
"epoch": 0.16,
"learning_rate": 0.00042607823470411236,
"loss": 2.6055,
"theoretical_loss": 3.901502838182272,
"tokens_seen": 516161536
},
{
"epoch": 0.16,
"learning_rate": 0.00042606820461384154,
"loss": 3.1029,
"theoretical_loss": 3.901449778852542,
"tokens_seen": 516227072
},
{
"epoch": 0.16,
"learning_rate": 0.0004260581745235707,
"loss": 3.0751,
"theoretical_loss": 3.9013967281441677,
"tokens_seen": 516292608
},
{
"epoch": 0.16,
"learning_rate": 0.0004260481444332999,
"loss": 3.002,
"theoretical_loss": 3.9013436860546555,
"tokens_seen": 516358144
},
{
"epoch": 0.16,
"learning_rate": 0.0004260381143430291,
"loss": 2.8306,
"theoretical_loss": 3.901290652581511,
"tokens_seen": 516423680
},
{
"epoch": 0.16,
"learning_rate": 0.0004260280842527583,
"loss": 2.888,
"theoretical_loss": 3.901237627722241,
"tokens_seen": 516489216
},
{
"epoch": 0.16,
"learning_rate": 0.00042601805416248744,
"loss": 2.8373,
"theoretical_loss": 3.9011846114743536,
"tokens_seen": 516554752
},
{
"epoch": 0.16,
"learning_rate": 0.0004260080240722167,
"loss": 2.7766,
"theoretical_loss": 3.901131603835359,
"tokens_seen": 516620288
},
{
"epoch": 0.16,
"learning_rate": 0.0004259979939819458,
"loss": 3.003,
"theoretical_loss": 3.9010786048027666,
"tokens_seen": 516685824
},
{
"epoch": 0.16,
"learning_rate": 0.00042598796389167504,
"loss": 2.819,
"theoretical_loss": 3.9010256143740873,
"tokens_seen": 516751360
},
{
"epoch": 0.16,
"learning_rate": 0.0004259779338014042,
"loss": 3.0103,
"theoretical_loss": 3.900972632546835,
"tokens_seen": 516816896
},
{
"epoch": 0.16,
"learning_rate": 0.0004259679037111334,
"loss": 3.0061,
"theoretical_loss": 3.900919659318521,
"tokens_seen": 516882432
},
{
"epoch": 0.16,
"learning_rate": 0.0004259578736208626,
"loss": 3.2129,
"theoretical_loss": 3.9008666946866613,
"tokens_seen": 516947968
},
{
"epoch": 0.16,
"learning_rate": 0.0004259478435305918,
"loss": 3.0836,
"theoretical_loss": 3.900813738648771,
"tokens_seen": 517013504
},
{
"epoch": 0.16,
"learning_rate": 0.00042593781344032095,
"loss": 2.9082,
"theoretical_loss": 3.900760791202366,
"tokens_seen": 517079040
},
{
"epoch": 0.16,
"learning_rate": 0.0004259277833500502,
"loss": 2.9466,
"theoretical_loss": 3.900707852344964,
"tokens_seen": 517144576
},
{
"epoch": 0.16,
"learning_rate": 0.0004259177532597793,
"loss": 3.081,
"theoretical_loss": 3.9006549220740836,
"tokens_seen": 517210112
},
{
"epoch": 0.16,
"learning_rate": 0.00042590772316950854,
"loss": 2.8449,
"theoretical_loss": 3.9006020003872446,
"tokens_seen": 517275648
},
{
"epoch": 0.16,
"learning_rate": 0.0004258976930792377,
"loss": 2.8311,
"theoretical_loss": 3.9005490872819673,
"tokens_seen": 517341184
},
{
"epoch": 0.16,
"learning_rate": 0.0004258876629889669,
"loss": 2.991,
"theoretical_loss": 3.900496182755773,
"tokens_seen": 517406720
},
{
"epoch": 0.16,
"learning_rate": 0.0004258776328986961,
"loss": 2.9178,
"theoretical_loss": 3.9004432868061856,
"tokens_seen": 517472256
},
{
"epoch": 0.16,
"learning_rate": 0.00042586760280842527,
"loss": 2.7577,
"theoretical_loss": 3.9003903994307274,
"tokens_seen": 517537792
},
{
"epoch": 0.16,
"learning_rate": 0.00042585757271815445,
"loss": 2.7745,
"theoretical_loss": 3.9003375206269233,
"tokens_seen": 517603328
},
{
"epoch": 0.16,
"learning_rate": 0.0004258475426278837,
"loss": 2.9678,
"theoretical_loss": 3.9002846503923,
"tokens_seen": 517668864
},
{
"epoch": 0.16,
"objective/train/docs_used": 303675,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.994904041290283,
"objective/train/theoretical_loss": 3.9002317887243834,
"objective/train/tokens_used": 538194400,
"theoretical_loss": 3.9002317887243834,
"tokens_seen": 517734400
},
{
"epoch": 0.16,
"learning_rate": 0.0004258375125376128,
"loss": 2.8722,
"theoretical_loss": 3.9002317887243834,
"tokens_seen": 517734400
},
{
"epoch": 0.16,
"learning_rate": 0.00042582748244734205,
"loss": 2.9304,
"theoretical_loss": 3.9001789356207013,
"tokens_seen": 517799936
},
{
"epoch": 0.16,
"learning_rate": 0.0004258174523570712,
"loss": 2.8775,
"theoretical_loss": 3.900126091078783,
"tokens_seen": 517865472
},
{
"epoch": 0.16,
"learning_rate": 0.0004258074222668004,
"loss": 3.0728,
"theoretical_loss": 3.9000732550961574,
"tokens_seen": 517931008
},
{
"epoch": 0.16,
"learning_rate": 0.0004257973921765296,
"loss": 2.9802,
"theoretical_loss": 3.9000204276703565,
"tokens_seen": 517996544
},
{
"epoch": 0.16,
"learning_rate": 0.00042578736208625877,
"loss": 2.889,
"theoretical_loss": 3.8999676087989115,
"tokens_seen": 518062080
},
{
"epoch": 0.16,
"learning_rate": 0.00042577733199598795,
"loss": 3.0984,
"theoretical_loss": 3.8999147984793554,
"tokens_seen": 518127616
},
{
"epoch": 0.16,
"learning_rate": 0.0004257673019057172,
"loss": 2.8683,
"theoretical_loss": 3.8998619967092223,
"tokens_seen": 518193152
},
{
"epoch": 0.16,
"learning_rate": 0.0004257572718154463,
"loss": 2.7482,
"theoretical_loss": 3.8998092034860474,
"tokens_seen": 518258688
},
{
"epoch": 0.16,
"learning_rate": 0.00042574724172517555,
"loss": 3.1747,
"theoretical_loss": 3.8997564188073657,
"tokens_seen": 518324224
},
{
"epoch": 0.16,
"learning_rate": 0.0004257372116349047,
"loss": 2.7086,
"theoretical_loss": 3.8997036426707155,
"tokens_seen": 518389760
},
{
"epoch": 0.16,
"learning_rate": 0.0004257271815446339,
"loss": 2.925,
"theoretical_loss": 3.8996508750736343,
"tokens_seen": 518455296
},
{
"epoch": 0.16,
"learning_rate": 0.0004257171514543631,
"loss": 2.9987,
"theoretical_loss": 3.89959811601366,
"tokens_seen": 518520832
},
{
"epoch": 0.16,
"learning_rate": 0.0004257071213640923,
"loss": 2.7581,
"theoretical_loss": 3.8995453654883345,
"tokens_seen": 518586368
},
{
"epoch": 0.16,
"learning_rate": 0.00042569709127382145,
"loss": 2.8995,
"theoretical_loss": 3.899492623495198,
"tokens_seen": 518651904
},
{
"epoch": 0.16,
"learning_rate": 0.00042568706118355064,
"loss": 2.8678,
"theoretical_loss": 3.8994398900317924,
"tokens_seen": 518717440
},
{
"epoch": 0.16,
"learning_rate": 0.0004256770310932798,
"loss": 2.7447,
"theoretical_loss": 3.8993871650956606,
"tokens_seen": 518782976
},
{
"epoch": 0.16,
"learning_rate": 0.00042566700100300905,
"loss": 2.8438,
"theoretical_loss": 3.899334448684348,
"tokens_seen": 518848512
},
{
"epoch": 0.16,
"learning_rate": 0.00042565697091273823,
"loss": 2.8911,
"theoretical_loss": 3.8992817407953986,
"tokens_seen": 518914048
},
{
"epoch": 0.16,
"learning_rate": 0.0004256469408224674,
"loss": 2.7872,
"theoretical_loss": 3.8992290414263593,
"tokens_seen": 518979584
},
{
"epoch": 0.16,
"learning_rate": 0.00042563691073219665,
"loss": 2.8124,
"theoretical_loss": 3.8991763505747765,
"tokens_seen": 519045120
},
{
"epoch": 0.16,
"learning_rate": 0.0004256268806419258,
"loss": 2.9019,
"theoretical_loss": 3.8991236682381993,
"tokens_seen": 519110656
},
{
"epoch": 0.16,
"learning_rate": 0.000425616850551655,
"loss": 2.8403,
"theoretical_loss": 3.899070994414176,
"tokens_seen": 519176192
},
{
"epoch": 0.16,
"learning_rate": 0.00042560682046138414,
"loss": 2.9856,
"theoretical_loss": 3.899018329100257,
"tokens_seen": 519241728
},
{
"epoch": 0.16,
"learning_rate": 0.0004255967903711134,
"loss": 3.0539,
"theoretical_loss": 3.8989656722939943,
"tokens_seen": 519307264
},
{
"epoch": 0.16,
"objective/train/docs_used": 304298,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.55254864692688,
"objective/train/theoretical_loss": 3.8989130239929395,
"objective/train/tokens_used": 539832800,
"theoretical_loss": 3.8989130239929395,
"tokens_seen": 519372800
},
{
"epoch": 0.16,
"learning_rate": 0.00042558676028084256,
"loss": 2.7837,
"theoretical_loss": 3.8989130239929395,
"tokens_seen": 519372800
},
{
"epoch": 0.16,
"learning_rate": 0.00042557673019057174,
"loss": 3.1594,
"theoretical_loss": 3.8988603841946463,
"tokens_seen": 519438336
},
{
"epoch": 0.16,
"learning_rate": 0.0004255667001003009,
"loss": 3.1587,
"theoretical_loss": 3.8988077528966687,
"tokens_seen": 519503872
},
{
"epoch": 0.16,
"learning_rate": 0.0004255566700100301,
"loss": 2.894,
"theoretical_loss": 3.898755130096562,
"tokens_seen": 519569408
},
{
"epoch": 0.16,
"learning_rate": 0.0004255466399197593,
"loss": 2.8944,
"theoretical_loss": 3.898702515791882,
"tokens_seen": 519634944
},
{
"epoch": 0.16,
"learning_rate": 0.0004255366098294885,
"loss": 2.966,
"theoretical_loss": 3.8986499099801875,
"tokens_seen": 519700480
},
{
"epoch": 0.16,
"learning_rate": 0.00042552657973921764,
"loss": 3.0096,
"theoretical_loss": 3.898597312659035,
"tokens_seen": 519766016
},
{
"epoch": 0.16,
"learning_rate": 0.0004255165496489469,
"loss": 3.0903,
"theoretical_loss": 3.8985447238259856,
"tokens_seen": 519831552
},
{
"epoch": 0.16,
"learning_rate": 0.000425506519558676,
"loss": 2.9664,
"theoretical_loss": 3.8984921434785984,
"tokens_seen": 519897088
},
{
"epoch": 0.16,
"learning_rate": 0.00042549648946840524,
"loss": 2.7405,
"theoretical_loss": 3.8984395716144347,
"tokens_seen": 519962624
},
{
"epoch": 0.16,
"learning_rate": 0.0004254864593781344,
"loss": 3.0711,
"theoretical_loss": 3.898387008231058,
"tokens_seen": 520028160
},
{
"epoch": 0.16,
"learning_rate": 0.0004254764292878636,
"loss": 3.0678,
"theoretical_loss": 3.8983344533260307,
"tokens_seen": 520093696
},
{
"epoch": 0.16,
"learning_rate": 0.0004254663991975928,
"loss": 2.9319,
"theoretical_loss": 3.8982819068969174,
"tokens_seen": 520159232
},
{
"epoch": 0.16,
"learning_rate": 0.000425456369107322,
"loss": 2.8959,
"theoretical_loss": 3.8982293689412835,
"tokens_seen": 520224768
},
{
"epoch": 0.16,
"learning_rate": 0.00042544633901705115,
"loss": 2.8478,
"theoretical_loss": 3.8981768394566956,
"tokens_seen": 520290304
},
{
"epoch": 0.16,
"learning_rate": 0.0004254363089267804,
"loss": 3.0487,
"theoretical_loss": 3.898124318440721,
"tokens_seen": 520355840
},
{
"epoch": 0.16,
"learning_rate": 0.0004254262788365095,
"loss": 3.0492,
"theoretical_loss": 3.8980718058909276,
"tokens_seen": 520421376
},
{
"epoch": 0.16,
"learning_rate": 0.00042541624874623874,
"loss": 2.9961,
"theoretical_loss": 3.8980193018048857,
"tokens_seen": 520486912
},
{
"epoch": 0.16,
"learning_rate": 0.0004254062186559679,
"loss": 3.0865,
"theoretical_loss": 3.8979668061801656,
"tokens_seen": 520552448
},
{
"epoch": 0.16,
"learning_rate": 0.0004253961885656971,
"loss": 2.9966,
"theoretical_loss": 3.8979143190143377,
"tokens_seen": 520617984
},
{
"epoch": 0.16,
"learning_rate": 0.0004253861584754263,
"loss": 2.7913,
"theoretical_loss": 3.8978618403049747,
"tokens_seen": 520683520
},
{
"epoch": 0.16,
"learning_rate": 0.00042537612838515547,
"loss": 3.017,
"theoretical_loss": 3.8978093700496514,
"tokens_seen": 520749056
},
{
"epoch": 0.16,
"learning_rate": 0.00042536609829488465,
"loss": 2.8749,
"theoretical_loss": 3.897756908245941,
"tokens_seen": 520814592
},
{
"epoch": 0.16,
"learning_rate": 0.0004253560682046139,
"loss": 2.9132,
"theoretical_loss": 3.897704454891419,
"tokens_seen": 520880128
},
{
"epoch": 0.16,
"learning_rate": 0.000425346038114343,
"loss": 2.6791,
"theoretical_loss": 3.897652009983662,
"tokens_seen": 520945664
},
{
"epoch": 0.16,
"objective/train/docs_used": 305033,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.236427068710327,
"objective/train/theoretical_loss": 3.897599573520247,
"objective/train/tokens_used": 541471200,
"theoretical_loss": 3.897599573520247,
"tokens_seen": 521011200
},
{
"epoch": 0.16,
"learning_rate": 0.00042533600802407225,
"loss": 3.0652,
"theoretical_loss": 3.897599573520247,
"tokens_seen": 521011200
},
{
"epoch": 0.16,
"learning_rate": 0.0004253259779338014,
"loss": 2.7909,
"theoretical_loss": 3.897547145498753,
"tokens_seen": 521076736
},
{
"epoch": 0.16,
"learning_rate": 0.0004253159478435306,
"loss": 2.945,
"theoretical_loss": 3.8974947259167596,
"tokens_seen": 521142272
},
{
"epoch": 0.16,
"learning_rate": 0.0004253059177532598,
"loss": 2.9162,
"theoretical_loss": 3.8974423147718467,
"tokens_seen": 521207808
},
{
"epoch": 0.16,
"learning_rate": 0.00042529588766298897,
"loss": 3.0102,
"theoretical_loss": 3.8973899120615956,
"tokens_seen": 521273344
},
{
"epoch": 0.16,
"learning_rate": 0.00042528585757271815,
"loss": 2.8023,
"theoretical_loss": 3.8973375177835896,
"tokens_seen": 521338880
},
{
"epoch": 0.16,
"learning_rate": 0.0004252758274824474,
"loss": 3.0294,
"theoretical_loss": 3.897285131935411,
"tokens_seen": 521404416
},
{
"epoch": 0.16,
"learning_rate": 0.0004252657973921765,
"loss": 3.1221,
"theoretical_loss": 3.8972327545146443,
"tokens_seen": 521469952
},
{
"epoch": 0.16,
"learning_rate": 0.00042525576730190575,
"loss": 2.6076,
"theoretical_loss": 3.897180385518876,
"tokens_seen": 521535488
},
{
"epoch": 0.16,
"learning_rate": 0.0004252457372116349,
"loss": 3.0293,
"theoretical_loss": 3.897128024945692,
"tokens_seen": 521601024
},
{
"epoch": 0.16,
"learning_rate": 0.0004252357071213641,
"loss": 2.9325,
"theoretical_loss": 3.8970756727926785,
"tokens_seen": 521666560
},
{
"epoch": 0.16,
"learning_rate": 0.0004252256770310933,
"loss": 3.0425,
"theoretical_loss": 3.8970233290574257,
"tokens_seen": 521732096
},
{
"epoch": 0.16,
"learning_rate": 0.0004252156469408225,
"loss": 2.8348,
"theoretical_loss": 3.896970993737522,
"tokens_seen": 521797632
},
{
"epoch": 0.16,
"learning_rate": 0.00042520561685055165,
"loss": 3.0828,
"theoretical_loss": 3.896918666830558,
"tokens_seen": 521863168
},
{
"epoch": 0.16,
"learning_rate": 0.00042519558676028084,
"loss": 2.8449,
"theoretical_loss": 3.8968663483341253,
"tokens_seen": 521928704
},
{
"epoch": 0.16,
"learning_rate": 0.00042518555667001,
"loss": 2.8192,
"theoretical_loss": 3.8968140382458154,
"tokens_seen": 521994240
},
{
"epoch": 0.16,
"learning_rate": 0.00042517552657973925,
"loss": 2.9585,
"theoretical_loss": 3.8967617365632217,
"tokens_seen": 522059776
},
{
"epoch": 0.16,
"learning_rate": 0.0004251654964894684,
"loss": 2.8681,
"theoretical_loss": 3.8967094432839398,
"tokens_seen": 522125312
},
{
"epoch": 0.16,
"learning_rate": 0.0004251554663991976,
"loss": 2.9713,
"theoretical_loss": 3.8966571584055645,
"tokens_seen": 522190848
},
{
"epoch": 0.16,
"learning_rate": 0.0004251454363089268,
"loss": 2.908,
"theoretical_loss": 3.896604881925691,
"tokens_seen": 522256384
},
{
"epoch": 0.16,
"learning_rate": 0.000425135406218656,
"loss": 2.7829,
"theoretical_loss": 3.8965526138419184,
"tokens_seen": 522321920
},
{
"epoch": 0.16,
"learning_rate": 0.00042512537612838516,
"loss": 2.8642,
"theoretical_loss": 3.8965003541518435,
"tokens_seen": 522387456
},
{
"epoch": 0.16,
"learning_rate": 0.00042511534603811434,
"loss": 2.9174,
"theoretical_loss": 3.896448102853067,
"tokens_seen": 522452992
},
{
"epoch": 0.16,
"learning_rate": 0.0004251053159478435,
"loss": 2.9457,
"theoretical_loss": 3.8963958599431874,
"tokens_seen": 522518528
},
{
"epoch": 0.16,
"learning_rate": 0.00042509528585757276,
"loss": 2.7995,
"theoretical_loss": 3.8963436254198074,
"tokens_seen": 522584064
},
{
"epoch": 0.16,
"objective/train/docs_used": 306243,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.018648862838745,
"objective/train/theoretical_loss": 3.8962913992805284,
"objective/train/tokens_used": 543109600,
"theoretical_loss": 3.8962913992805284,
"tokens_seen": 522649600
},
{
"epoch": 0.16,
"learning_rate": 0.0004250852557673019,
"loss": 2.9636,
"theoretical_loss": 3.8962913992805284,
"tokens_seen": 522649600
},
{
"epoch": 0.16,
"learning_rate": 0.0004250752256770311,
"loss": 2.87,
"theoretical_loss": 3.896239181522954,
"tokens_seen": 522715136
},
{
"epoch": 0.16,
"learning_rate": 0.00042506519558676024,
"loss": 2.7462,
"theoretical_loss": 3.8961869721446885,
"tokens_seen": 522780672
},
{
"epoch": 0.16,
"learning_rate": 0.0004250551654964895,
"loss": 2.907,
"theoretical_loss": 3.8961347711433376,
"tokens_seen": 522846208
},
{
"epoch": 0.16,
"learning_rate": 0.00042504513540621866,
"loss": 2.8195,
"theoretical_loss": 3.896082578516506,
"tokens_seen": 522911744
},
{
"epoch": 0.16,
"learning_rate": 0.00042503510531594784,
"loss": 2.8101,
"theoretical_loss": 3.8960303942618024,
"tokens_seen": 522977280
},
{
"epoch": 0.16,
"learning_rate": 0.000425025075225677,
"loss": 3.1211,
"theoretical_loss": 3.8959782183768343,
"tokens_seen": 523042816
},
{
"epoch": 0.16,
"learning_rate": 0.0004250150451354062,
"loss": 2.6012,
"theoretical_loss": 3.8959260508592104,
"tokens_seen": 523108352
},
{
"epoch": 0.16,
"learning_rate": 0.0004250050150451354,
"loss": 3.1034,
"theoretical_loss": 3.895873891706542,
"tokens_seen": 523173888
},
{
"epoch": 0.16,
"learning_rate": 0.0004249949849548646,
"loss": 2.7857,
"theoretical_loss": 3.895821740916439,
"tokens_seen": 523239424
},
{
"epoch": 0.16,
"learning_rate": 0.00042498495486459375,
"loss": 2.9521,
"theoretical_loss": 3.8957695984865137,
"tokens_seen": 523304960
},
{
"epoch": 0.16,
"learning_rate": 0.000424974924774323,
"loss": 2.9685,
"theoretical_loss": 3.89571746441438,
"tokens_seen": 523370496
},
{
"epoch": 0.16,
"learning_rate": 0.00042496489468405216,
"loss": 2.8718,
"theoretical_loss": 3.8956653386976514,
"tokens_seen": 523436032
},
{
"epoch": 0.16,
"learning_rate": 0.00042495486459378135,
"loss": 2.9434,
"theoretical_loss": 3.8956132213339423,
"tokens_seen": 523501568
},
{
"epoch": 0.16,
"learning_rate": 0.0004249448345035105,
"loss": 2.9804,
"theoretical_loss": 3.8955611123208698,
"tokens_seen": 523567104
},
{
"epoch": 0.16,
"learning_rate": 0.0004249348044132397,
"loss": 3.0082,
"theoretical_loss": 3.8955090116560505,
"tokens_seen": 523632640
},
{
"epoch": 0.16,
"learning_rate": 0.00042492477432296894,
"loss": 3.0848,
"theoretical_loss": 3.895456919337102,
"tokens_seen": 523698176
},
{
"epoch": 0.16,
"learning_rate": 0.0004249147442326981,
"loss": 2.8986,
"theoretical_loss": 3.8954048353616435,
"tokens_seen": 523763712
},
{
"epoch": 0.16,
"learning_rate": 0.0004249047141424273,
"loss": 2.6751,
"theoretical_loss": 3.8953527597272952,
"tokens_seen": 523829248
},
{
"epoch": 0.16,
"learning_rate": 0.0004248946840521565,
"loss": 3.1502,
"theoretical_loss": 3.895300692431677,
"tokens_seen": 523894784
},
{
"epoch": 0.16,
"learning_rate": 0.00042488465396188567,
"loss": 3.002,
"theoretical_loss": 3.8952486334724123,
"tokens_seen": 523960320
},
{
"epoch": 0.16,
"learning_rate": 0.00042487462387161485,
"loss": 2.8876,
"theoretical_loss": 3.895196582847123,
"tokens_seen": 524025856
},
{
"epoch": 0.16,
"learning_rate": 0.0004248645937813441,
"loss": 3.1295,
"theoretical_loss": 3.8951445405534324,
"tokens_seen": 524091392
},
{
"epoch": 0.16,
"learning_rate": 0.0004248545636910732,
"loss": 3.0447,
"theoretical_loss": 3.8950925065889663,
"tokens_seen": 524156928
},
{
"epoch": 0.16,
"learning_rate": 0.00042484453360080245,
"loss": 2.6436,
"theoretical_loss": 3.8950404809513506,
"tokens_seen": 524222464
},
{
"epoch": 0.16,
"objective/train/docs_used": 306972,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.143721580505371,
"objective/train/theoretical_loss": 3.8949884636382106,
"objective/train/tokens_used": 544748000,
"theoretical_loss": 3.8949884636382106,
"tokens_seen": 524288000
},
{
"epoch": 0.16,
"learning_rate": 0.0004248345035105316,
"loss": 2.9599,
"theoretical_loss": 3.8949884636382106,
"tokens_seen": 524288000
},
{
"epoch": 0.16,
"learning_rate": 0.0004248244734202608,
"loss": 2.937,
"theoretical_loss": 3.8949364546471754,
"tokens_seen": 524353536
},
{
"epoch": 0.16,
"learning_rate": 0.00042481444332999,
"loss": 2.9483,
"theoretical_loss": 3.894884453975873,
"tokens_seen": 524419072
},
{
"epoch": 0.16,
"learning_rate": 0.00042480441323971917,
"loss": 2.9053,
"theoretical_loss": 3.894832461621934,
"tokens_seen": 524484608
},
{
"epoch": 0.16,
"learning_rate": 0.00042479438314944835,
"loss": 2.8771,
"theoretical_loss": 3.894780477582988,
"tokens_seen": 524550144
},
{
"epoch": 0.16,
"learning_rate": 0.0004247843530591776,
"loss": 2.9409,
"theoretical_loss": 3.894728501856667,
"tokens_seen": 524615680
},
{
"epoch": 0.16,
"learning_rate": 0.0004247743229689067,
"loss": 3.1486,
"theoretical_loss": 3.8946765344406034,
"tokens_seen": 524681216
},
{
"epoch": 0.16,
"learning_rate": 0.00042476429287863595,
"loss": 2.5875,
"theoretical_loss": 3.8946245753324304,
"tokens_seen": 524746752
},
{
"epoch": 0.16,
"learning_rate": 0.0004247542627883651,
"loss": 2.9741,
"theoretical_loss": 3.8945726245297836,
"tokens_seen": 524812288
},
{
"epoch": 0.16,
"learning_rate": 0.0004247442326980943,
"loss": 2.9664,
"theoretical_loss": 3.894520682030297,
"tokens_seen": 524877824
},
{
"epoch": 0.16,
"learning_rate": 0.0004247342026078235,
"loss": 2.7809,
"theoretical_loss": 3.894468747831609,
"tokens_seen": 524943360
},
{
"epoch": 0.16,
"learning_rate": 0.0004247241725175527,
"loss": 3.0679,
"theoretical_loss": 3.8944168219313546,
"tokens_seen": 525008896
},
{
"epoch": 0.16,
"learning_rate": 0.00042471414242728186,
"loss": 3.0167,
"theoretical_loss": 3.8943649043271744,
"tokens_seen": 525074432
},
{
"epoch": 0.16,
"learning_rate": 0.00042470411233701104,
"loss": 3.1154,
"theoretical_loss": 3.8943129950167066,
"tokens_seen": 525139968
},
{
"epoch": 0.16,
"learning_rate": 0.0004246940822467402,
"loss": 3.0939,
"theoretical_loss": 3.894261093997592,
"tokens_seen": 525205504
},
{
"epoch": 0.16,
"learning_rate": 0.00042468405215646945,
"loss": 3.0354,
"theoretical_loss": 3.894209201267471,
"tokens_seen": 525271040
},
{
"epoch": 0.16,
"learning_rate": 0.0004246740220661986,
"loss": 3.0519,
"theoretical_loss": 3.8941573168239874,
"tokens_seen": 525336576
},
{
"epoch": 0.16,
"learning_rate": 0.0004246639919759278,
"loss": 2.9625,
"theoretical_loss": 3.8941054406647826,
"tokens_seen": 525402112
},
{
"epoch": 0.16,
"learning_rate": 0.000424653961885657,
"loss": 2.8217,
"theoretical_loss": 3.894053572787502,
"tokens_seen": 525467648
},
{
"epoch": 0.16,
"learning_rate": 0.0004246439317953862,
"loss": 3.094,
"theoretical_loss": 3.89400171318979,
"tokens_seen": 525533184
},
{
"epoch": 0.16,
"learning_rate": 0.00042463390170511536,
"loss": 2.8674,
"theoretical_loss": 3.8939498618692934,
"tokens_seen": 525598720
},
{
"epoch": 0.16,
"learning_rate": 0.00042462387161484454,
"loss": 2.8711,
"theoretical_loss": 3.893898018823659,
"tokens_seen": 525664256
},
{
"epoch": 0.16,
"learning_rate": 0.0004246138415245737,
"loss": 3.068,
"theoretical_loss": 3.8938461840505347,
"tokens_seen": 525729792
},
{
"epoch": 0.16,
"learning_rate": 0.00042460381143430296,
"loss": 3.1631,
"theoretical_loss": 3.8937943575475695,
"tokens_seen": 525795328
},
{
"epoch": 0.16,
"learning_rate": 0.0004245937813440321,
"loss": 2.852,
"theoretical_loss": 3.893742539312414,
"tokens_seen": 525860864
},
{
"epoch": 0.16,
"objective/train/docs_used": 308263,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0648889541625977,
"objective/train/theoretical_loss": 3.8936907293427176,
"objective/train/tokens_used": 546386400,
"theoretical_loss": 3.8936907293427176,
"tokens_seen": 525926400
},
{
"epoch": 0.16,
"learning_rate": 0.0004245837512537613,
"loss": 3.0778,
"theoretical_loss": 3.8936907293427176,
"tokens_seen": 525926400
},
{
"epoch": 0.16,
"learning_rate": 0.00042457372116349044,
"loss": 2.7427,
"theoretical_loss": 3.893638927636134,
"tokens_seen": 525991936
},
{
"epoch": 0.16,
"learning_rate": 0.0004245636910732197,
"loss": 3.0738,
"theoretical_loss": 3.893587134190314,
"tokens_seen": 526057472
},
{
"epoch": 0.16,
"learning_rate": 0.00042455366098294886,
"loss": 2.9898,
"theoretical_loss": 3.8935353490029136,
"tokens_seen": 526123008
},
{
"epoch": 0.16,
"learning_rate": 0.00042454363089267804,
"loss": 2.9087,
"theoretical_loss": 3.893483572071586,
"tokens_seen": 526188544
},
{
"epoch": 0.16,
"learning_rate": 0.0004245336008024072,
"loss": 2.8176,
"theoretical_loss": 3.8934318033939874,
"tokens_seen": 526254080
},
{
"epoch": 0.16,
"learning_rate": 0.0004245235707121364,
"loss": 2.8114,
"theoretical_loss": 3.8933800429677743,
"tokens_seen": 526319616
},
{
"epoch": 0.16,
"learning_rate": 0.0004245135406218656,
"loss": 2.9196,
"theoretical_loss": 3.8933282907906044,
"tokens_seen": 526385152
},
{
"epoch": 0.16,
"learning_rate": 0.0004245035105315948,
"loss": 2.8632,
"theoretical_loss": 3.8932765468601365,
"tokens_seen": 526450688
},
{
"epoch": 0.16,
"learning_rate": 0.00042449348044132395,
"loss": 2.8034,
"theoretical_loss": 3.89322481117403,
"tokens_seen": 526516224
},
{
"epoch": 0.16,
"learning_rate": 0.0004244834503510532,
"loss": 2.7507,
"theoretical_loss": 3.8931730837299456,
"tokens_seen": 526581760
},
{
"epoch": 0.16,
"learning_rate": 0.00042447342026078236,
"loss": 2.879,
"theoretical_loss": 3.893121364525544,
"tokens_seen": 526647296
},
{
"epoch": 0.16,
"learning_rate": 0.00042446339017051155,
"loss": 2.9306,
"theoretical_loss": 3.893069653558488,
"tokens_seen": 526712832
},
{
"epoch": 0.16,
"learning_rate": 0.0004244533600802407,
"loss": 2.749,
"theoretical_loss": 3.893017950826441,
"tokens_seen": 526778368
},
{
"epoch": 0.16,
"learning_rate": 0.0004244433299899699,
"loss": 2.88,
"theoretical_loss": 3.892966256327068,
"tokens_seen": 526843904
},
{
"epoch": 0.16,
"learning_rate": 0.0004244332998996991,
"loss": 3.0049,
"theoretical_loss": 3.892914570058033,
"tokens_seen": 526909440
},
{
"epoch": 0.16,
"learning_rate": 0.0004244232698094283,
"loss": 2.9474,
"theoretical_loss": 3.8928628920170034,
"tokens_seen": 526974976
},
{
"epoch": 0.16,
"learning_rate": 0.00042441323971915745,
"loss": 2.9997,
"theoretical_loss": 3.8928112222016455,
"tokens_seen": 527040512
},
{
"epoch": 0.16,
"learning_rate": 0.0004244032096288867,
"loss": 2.9753,
"theoretical_loss": 3.892759560609627,
"tokens_seen": 527106048
},
{
"epoch": 0.16,
"learning_rate": 0.0004243931795386158,
"loss": 2.9288,
"theoretical_loss": 3.8927079072386186,
"tokens_seen": 527171584
},
{
"epoch": 0.16,
"learning_rate": 0.00042438314944834505,
"loss": 2.8508,
"theoretical_loss": 3.892656262086289,
"tokens_seen": 527237120
},
{
"epoch": 0.16,
"learning_rate": 0.00042437311935807423,
"loss": 2.9228,
"theoretical_loss": 3.8926046251503092,
"tokens_seen": 527302656
},
{
"epoch": 0.16,
"learning_rate": 0.0004243630892678034,
"loss": 2.9325,
"theoretical_loss": 3.8925529964283516,
"tokens_seen": 527368192
},
{
"epoch": 0.16,
"learning_rate": 0.0004243530591775326,
"loss": 2.9051,
"theoretical_loss": 3.892501375918089,
"tokens_seen": 527433728
},
{
"epoch": 0.16,
"learning_rate": 0.0004243430290872618,
"loss": 2.7249,
"theoretical_loss": 3.8924497636171953,
"tokens_seen": 527499264
},
{
"epoch": 0.16,
"objective/train/docs_used": 309026,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9705495834350586,
"objective/train/theoretical_loss": 3.892398159523345,
"objective/train/tokens_used": 548024800,
"theoretical_loss": 3.892398159523345,
"tokens_seen": 527564800
},
{
"epoch": 0.16,
"learning_rate": 0.00042433299899699095,
"loss": 3.0839,
"theoretical_loss": 3.892398159523345,
"tokens_seen": 527564800
},
{
"epoch": 0.16,
"learning_rate": 0.0004243229689067202,
"loss": 2.9342,
"theoretical_loss": 3.892346563634213,
"tokens_seen": 527630336
},
{
"epoch": 0.16,
"learning_rate": 0.0004243129388164493,
"loss": 2.9797,
"theoretical_loss": 3.892294975947477,
"tokens_seen": 527695872
},
{
"epoch": 0.16,
"learning_rate": 0.00042430290872617855,
"loss": 2.8072,
"theoretical_loss": 3.892243396460815,
"tokens_seen": 527761408
},
{
"epoch": 0.16,
"learning_rate": 0.00042429287863590773,
"loss": 3.1797,
"theoretical_loss": 3.8921918251719054,
"tokens_seen": 527826944
},
{
"epoch": 0.16,
"learning_rate": 0.0004242828485456369,
"loss": 2.9326,
"theoretical_loss": 3.892140262078427,
"tokens_seen": 527892480
},
{
"epoch": 0.16,
"learning_rate": 0.0004242728184553661,
"loss": 2.6238,
"theoretical_loss": 3.8920887071780594,
"tokens_seen": 527958016
},
{
"epoch": 0.16,
"learning_rate": 0.0004242627883650953,
"loss": 2.8564,
"theoretical_loss": 3.8920371604684862,
"tokens_seen": 528023552
},
{
"epoch": 0.16,
"learning_rate": 0.00042425275827482446,
"loss": 2.8973,
"theoretical_loss": 3.891985621947388,
"tokens_seen": 528089088
},
{
"epoch": 0.16,
"learning_rate": 0.0004242427281845537,
"loss": 3.1608,
"theoretical_loss": 3.891934091612449,
"tokens_seen": 528154624
},
{
"epoch": 0.16,
"learning_rate": 0.0004242326980942828,
"loss": 2.9455,
"theoretical_loss": 3.891882569461353,
"tokens_seen": 528220160
},
{
"epoch": 0.16,
"learning_rate": 0.00042422266800401206,
"loss": 2.8902,
"theoretical_loss": 3.891831055491785,
"tokens_seen": 528285696
},
{
"epoch": 0.16,
"learning_rate": 0.0004242126379137412,
"loss": 3.0813,
"theoretical_loss": 3.8917795497014316,
"tokens_seen": 528351232
},
{
"epoch": 0.16,
"learning_rate": 0.0004242026078234704,
"loss": 2.9741,
"theoretical_loss": 3.891728052087979,
"tokens_seen": 528416768
},
{
"epoch": 0.16,
"learning_rate": 0.0004241925777331996,
"loss": 2.83,
"theoretical_loss": 3.891676562649116,
"tokens_seen": 528482304
},
{
"epoch": 0.16,
"learning_rate": 0.0004241825476429288,
"loss": 2.7524,
"theoretical_loss": 3.8916250813825313,
"tokens_seen": 528547840
},
{
"epoch": 0.16,
"learning_rate": 0.000424172517552658,
"loss": 2.8614,
"theoretical_loss": 3.891573608285915,
"tokens_seen": 528613376
},
{
"epoch": 0.16,
"learning_rate": 0.0004241624874623872,
"loss": 3.131,
"theoretical_loss": 3.891522143356957,
"tokens_seen": 528678912
},
{
"epoch": 0.16,
"learning_rate": 0.0004241524573721164,
"loss": 2.6944,
"theoretical_loss": 3.89147068659335,
"tokens_seen": 528744448
},
{
"epoch": 0.16,
"learning_rate": 0.00042414242728184556,
"loss": 2.948,
"theoretical_loss": 3.8914192379927863,
"tokens_seen": 528809984
},
{
"epoch": 0.16,
"learning_rate": 0.00042413239719157474,
"loss": 2.9535,
"theoretical_loss": 3.891367797552959,
"tokens_seen": 528875520
},
{
"epoch": 0.16,
"learning_rate": 0.0004241223671013039,
"loss": 2.9196,
"theoretical_loss": 3.891316365271564,
"tokens_seen": 528941056
},
{
"epoch": 0.16,
"learning_rate": 0.00042411233701103316,
"loss": 2.8756,
"theoretical_loss": 3.891264941146295,
"tokens_seen": 529006592
},
{
"epoch": 0.16,
"learning_rate": 0.0004241023069207623,
"loss": 2.8476,
"theoretical_loss": 3.89121352517485,
"tokens_seen": 529072128
},
{
"epoch": 0.16,
"learning_rate": 0.0004240922768304915,
"loss": 2.9317,
"theoretical_loss": 3.8911621173549253,
"tokens_seen": 529137664
},
{
"epoch": 0.16,
"objective/train/docs_used": 310459,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7565929889678955,
"objective/train/theoretical_loss": 3.89111071768422,
"objective/train/tokens_used": 549663200,
"theoretical_loss": 3.89111071768422,
"tokens_seen": 529203200
},
{
"epoch": 0.16,
"learning_rate": 0.00042408224674022065,
"loss": 2.9638,
"theoretical_loss": 3.89111071768422,
"tokens_seen": 529203200
},
{
"epoch": 0.16,
"learning_rate": 0.0004240722166499499,
"loss": 3.0235,
"theoretical_loss": 3.8910593261604323,
"tokens_seen": 529268736
},
{
"epoch": 0.16,
"learning_rate": 0.00042406218655967906,
"loss": 2.9576,
"theoretical_loss": 3.891007942781264,
"tokens_seen": 529334272
},
{
"epoch": 0.16,
"learning_rate": 0.00042405215646940824,
"loss": 2.7206,
"theoretical_loss": 3.8909565675444138,
"tokens_seen": 529399808
},
{
"epoch": 0.16,
"learning_rate": 0.0004240421263791374,
"loss": 2.7056,
"theoretical_loss": 3.8909052004475857,
"tokens_seen": 529465344
},
{
"epoch": 0.16,
"learning_rate": 0.0004240320962888666,
"loss": 2.6441,
"theoretical_loss": 3.8908538414884823,
"tokens_seen": 529530880
},
{
"epoch": 0.16,
"learning_rate": 0.0004240220661985958,
"loss": 2.8619,
"theoretical_loss": 3.8908024906648073,
"tokens_seen": 529596416
},
{
"epoch": 0.16,
"learning_rate": 0.000424012036108325,
"loss": 2.8988,
"theoretical_loss": 3.890751147974265,
"tokens_seen": 529661952
},
{
"epoch": 0.16,
"learning_rate": 0.00042400200601805415,
"loss": 2.9959,
"theoretical_loss": 3.890699813414562,
"tokens_seen": 529727488
},
{
"epoch": 0.16,
"learning_rate": 0.0004239919759277834,
"loss": 2.955,
"theoretical_loss": 3.890648486983405,
"tokens_seen": 529793024
},
{
"epoch": 0.16,
"learning_rate": 0.00042398194583751256,
"loss": 2.841,
"theoretical_loss": 3.8905971686785006,
"tokens_seen": 529858560
},
{
"epoch": 0.16,
"learning_rate": 0.00042397191574724175,
"loss": 2.9333,
"theoretical_loss": 3.8905458584975583,
"tokens_seen": 529924096
},
{
"epoch": 0.16,
"learning_rate": 0.0004239618856569709,
"loss": 2.7768,
"theoretical_loss": 3.890494556438287,
"tokens_seen": 529989632
},
{
"epoch": 0.16,
"learning_rate": 0.0004239518555667001,
"loss": 2.913,
"theoretical_loss": 3.8904432624983976,
"tokens_seen": 530055168
},
{
"epoch": 0.16,
"learning_rate": 0.0004239418254764293,
"loss": 2.9885,
"theoretical_loss": 3.890391976675601,
"tokens_seen": 530120704
},
{
"epoch": 0.16,
"learning_rate": 0.0004239317953861585,
"loss": 2.9349,
"theoretical_loss": 3.89034069896761,
"tokens_seen": 530186240
},
{
"epoch": 0.16,
"learning_rate": 0.00042392176529588765,
"loss": 2.5791,
"theoretical_loss": 3.890289429372137,
"tokens_seen": 530251776
},
{
"epoch": 0.16,
"learning_rate": 0.0004239117352056169,
"loss": 3.0469,
"theoretical_loss": 3.8902381678868974,
"tokens_seen": 530317312
},
{
"epoch": 0.16,
"learning_rate": 0.000423901705115346,
"loss": 2.8172,
"theoretical_loss": 3.890186914509605,
"tokens_seen": 530382848
},
{
"epoch": 0.16,
"learning_rate": 0.00042389167502507525,
"loss": 2.8591,
"theoretical_loss": 3.890135669237976,
"tokens_seen": 530448384
},
{
"epoch": 0.16,
"learning_rate": 0.00042388164493480443,
"loss": 3.1571,
"theoretical_loss": 3.8900844320697274,
"tokens_seen": 530513920
},
{
"epoch": 0.16,
"learning_rate": 0.0004238716148445336,
"loss": 3.1662,
"theoretical_loss": 3.8900332030025773,
"tokens_seen": 530579456
},
{
"epoch": 0.16,
"learning_rate": 0.0004238615847542628,
"loss": 2.862,
"theoretical_loss": 3.8899819820342443,
"tokens_seen": 530644992
},
{
"epoch": 0.16,
"learning_rate": 0.000423851554663992,
"loss": 2.7471,
"theoretical_loss": 3.889930769162448,
"tokens_seen": 530710528
},
{
"epoch": 0.16,
"learning_rate": 0.00042384152457372115,
"loss": 2.9666,
"theoretical_loss": 3.8898795643849087,
"tokens_seen": 530776064
},
{
"epoch": 0.16,
"objective/train/docs_used": 311126,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.5528128147125244,
"objective/train/theoretical_loss": 3.889828367699349,
"objective/train/tokens_used": 551301600,
"theoretical_loss": 3.889828367699349,
"tokens_seen": 530841600
},
{
"epoch": 0.16,
"learning_rate": 0.0004238314944834504,
"loss": 3.1406,
"theoretical_loss": 3.889828367699349,
"tokens_seen": 530841600
},
{
"epoch": 0.16,
"learning_rate": 0.0004238214643931795,
"loss": 2.7622,
"theoretical_loss": 3.88977717910349,
"tokens_seen": 530907136
},
{
"epoch": 0.16,
"learning_rate": 0.00042381143430290875,
"loss": 2.8677,
"theoretical_loss": 3.8897259985950554,
"tokens_seen": 530972672
},
{
"epoch": 0.16,
"learning_rate": 0.00042380140421263793,
"loss": 2.9447,
"theoretical_loss": 3.8896748261717695,
"tokens_seen": 531038208
},
{
"epoch": 0.16,
"learning_rate": 0.0004237913741223671,
"loss": 2.9975,
"theoretical_loss": 3.889623661831359,
"tokens_seen": 531103744
},
{
"epoch": 0.16,
"learning_rate": 0.0004237813440320963,
"loss": 2.8501,
"theoretical_loss": 3.8895725055715475,
"tokens_seen": 531169280
},
{
"epoch": 0.16,
"learning_rate": 0.0004237713139418255,
"loss": 2.8257,
"theoretical_loss": 3.8895213573900635,
"tokens_seen": 531234816
},
{
"epoch": 0.16,
"learning_rate": 0.00042376128385155466,
"loss": 2.8109,
"theoretical_loss": 3.889470217284635,
"tokens_seen": 531300352
},
{
"epoch": 0.16,
"learning_rate": 0.0004237512537612839,
"loss": 2.7859,
"theoretical_loss": 3.8894190852529906,
"tokens_seen": 531365888
},
{
"epoch": 0.16,
"learning_rate": 0.000423741223671013,
"loss": 2.8411,
"theoretical_loss": 3.8893679612928604,
"tokens_seen": 531431424
},
{
"epoch": 0.16,
"learning_rate": 0.00042373119358074226,
"loss": 2.7203,
"theoretical_loss": 3.889316845401974,
"tokens_seen": 531496960
},
{
"epoch": 0.16,
"learning_rate": 0.0004237211634904714,
"loss": 2.5823,
"theoretical_loss": 3.8892657375780644,
"tokens_seen": 531562496
},
{
"epoch": 0.16,
"learning_rate": 0.0004237111334002006,
"loss": 2.8587,
"theoretical_loss": 3.8892146378188643,
"tokens_seen": 531628032
},
{
"epoch": 0.16,
"learning_rate": 0.0004237011033099298,
"loss": 2.8946,
"theoretical_loss": 3.8891635461221057,
"tokens_seen": 531693568
},
{
"epoch": 0.16,
"learning_rate": 0.000423691073219659,
"loss": 2.8974,
"theoretical_loss": 3.8891124624855236,
"tokens_seen": 531759104
},
{
"epoch": 0.16,
"learning_rate": 0.00042368104312938816,
"loss": 2.7539,
"theoretical_loss": 3.889061386906854,
"tokens_seen": 531824640
},
{
"epoch": 0.16,
"learning_rate": 0.0004236710130391174,
"loss": 2.7946,
"theoretical_loss": 3.889010319383832,
"tokens_seen": 531890176
},
{
"epoch": 0.16,
"learning_rate": 0.0004236609829488465,
"loss": 3.0662,
"theoretical_loss": 3.8889592599141958,
"tokens_seen": 531955712
},
{
"epoch": 0.16,
"learning_rate": 0.00042365095285857576,
"loss": 2.9279,
"theoretical_loss": 3.888908208495683,
"tokens_seen": 532021248
},
{
"epoch": 0.16,
"learning_rate": 0.0004236409227683049,
"loss": 2.8842,
"theoretical_loss": 3.8888571651260326,
"tokens_seen": 532086784
},
{
"epoch": 0.16,
"learning_rate": 0.0004236308926780341,
"loss": 2.6855,
"theoretical_loss": 3.888806129802984,
"tokens_seen": 532152320
},
{
"epoch": 0.16,
"learning_rate": 0.0004236208625877633,
"loss": 2.9097,
"theoretical_loss": 3.888755102524279,
"tokens_seen": 532217856
},
{
"epoch": 0.16,
"learning_rate": 0.0004236108324974925,
"loss": 2.718,
"theoretical_loss": 3.8887040832876587,
"tokens_seen": 532283392
},
{
"epoch": 0.16,
"learning_rate": 0.00042360080240722166,
"loss": 3.062,
"theoretical_loss": 3.8886530720908654,
"tokens_seen": 532348928
},
{
"epoch": 0.16,
"learning_rate": 0.00042359077231695085,
"loss": 2.9053,
"theoretical_loss": 3.888602068931643,
"tokens_seen": 532414464
},
{
"epoch": 0.16,
"objective/train/docs_used": 312276,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.722914934158325,
"objective/train/theoretical_loss": 3.888551073807736,
"objective/train/tokens_used": 552940000,
"theoretical_loss": 3.888551073807736,
"tokens_seen": 532480000
},
{
"epoch": 0.16,
"learning_rate": 0.00042358074222668,
"loss": 2.8273,
"theoretical_loss": 3.888551073807736,
"tokens_seen": 532480000
},
{
"epoch": 0.16,
"learning_rate": 0.00042357071213640926,
"loss": 2.8933,
"theoretical_loss": 3.8885000867168893,
"tokens_seen": 532545536
},
{
"epoch": 0.16,
"learning_rate": 0.0004235606820461384,
"loss": 2.7793,
"theoretical_loss": 3.88844910765685,
"tokens_seen": 532611072
},
{
"epoch": 0.16,
"learning_rate": 0.0004235506519558676,
"loss": 2.9064,
"theoretical_loss": 3.888398136625364,
"tokens_seen": 532676608
},
{
"epoch": 0.16,
"learning_rate": 0.00042354062186559675,
"loss": 3.0965,
"theoretical_loss": 3.888347173620181,
"tokens_seen": 532742144
},
{
"epoch": 0.16,
"learning_rate": 0.000423530591775326,
"loss": 2.8523,
"theoretical_loss": 3.8882962186390486,
"tokens_seen": 532807680
},
{
"epoch": 0.16,
"learning_rate": 0.00042352056168505517,
"loss": 3.0964,
"theoretical_loss": 3.8882452716797173,
"tokens_seen": 532873216
},
{
"epoch": 0.16,
"learning_rate": 0.00042351053159478435,
"loss": 2.7862,
"theoretical_loss": 3.8881943327399378,
"tokens_seen": 532938752
},
{
"epoch": 0.16,
"learning_rate": 0.00042350050150451353,
"loss": 3.0715,
"theoretical_loss": 3.8881434018174614,
"tokens_seen": 533004288
},
{
"epoch": 0.16,
"learning_rate": 0.00042349047141424276,
"loss": 3.2114,
"theoretical_loss": 3.8880924789100417,
"tokens_seen": 533069824
},
{
"epoch": 0.16,
"learning_rate": 0.0004234804413239719,
"loss": 3.0219,
"theoretical_loss": 3.888041564015431,
"tokens_seen": 533135360
},
{
"epoch": 0.16,
"learning_rate": 0.00042347041123370113,
"loss": 2.7477,
"theoretical_loss": 3.8879906571313847,
"tokens_seen": 533200896
},
{
"epoch": 0.16,
"learning_rate": 0.00042346038114343025,
"loss": 2.921,
"theoretical_loss": 3.887939758255658,
"tokens_seen": 533266432
},
{
"epoch": 0.16,
"learning_rate": 0.0004234503510531595,
"loss": 2.9896,
"theoretical_loss": 3.887888867386007,
"tokens_seen": 533331968
},
{
"epoch": 0.16,
"learning_rate": 0.00042344032096288867,
"loss": 2.789,
"theoretical_loss": 3.8878379845201883,
"tokens_seen": 533397504
},
{
"epoch": 0.16,
"learning_rate": 0.00042343029087261785,
"loss": 2.8581,
"theoretical_loss": 3.88778710965596,
"tokens_seen": 533463040
},
{
"epoch": 0.16,
"learning_rate": 0.0004234202607823471,
"loss": 3.0571,
"theoretical_loss": 3.887736242791082,
"tokens_seen": 533528576
},
{
"epoch": 0.16,
"learning_rate": 0.0004234102306920762,
"loss": 2.8806,
"theoretical_loss": 3.887685383923314,
"tokens_seen": 533594112
},
{
"epoch": 0.16,
"learning_rate": 0.00042340020060180545,
"loss": 2.9129,
"theoretical_loss": 3.8876345330504156,
"tokens_seen": 533659648
},
{
"epoch": 0.16,
"learning_rate": 0.00042339017051153463,
"loss": 3.0494,
"theoretical_loss": 3.887583690170149,
"tokens_seen": 533725184
},
{
"epoch": 0.16,
"learning_rate": 0.0004233801404212638,
"loss": 2.7722,
"theoretical_loss": 3.887532855280277,
"tokens_seen": 533790720
},
{
"epoch": 0.16,
"learning_rate": 0.000423370110330993,
"loss": 2.6642,
"theoretical_loss": 3.887482028378564,
"tokens_seen": 533856256
},
{
"epoch": 0.16,
"learning_rate": 0.0004233600802407222,
"loss": 2.8821,
"theoretical_loss": 3.8874312094627723,
"tokens_seen": 533921792
},
{
"epoch": 0.16,
"learning_rate": 0.00042335005015045135,
"loss": 2.8685,
"theoretical_loss": 3.8873803985306683,
"tokens_seen": 533987328
},
{
"epoch": 0.16,
"learning_rate": 0.0004233400200601806,
"loss": 2.7004,
"theoretical_loss": 3.8873295955800184,
"tokens_seen": 534052864
},
{
"epoch": 0.16,
"objective/train/docs_used": 312792,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.320737838745117,
"objective/train/theoretical_loss": 3.8872788006085894,
"objective/train/tokens_used": 554578400,
"theoretical_loss": 3.8872788006085894,
"tokens_seen": 534118400
},
{
"epoch": 0.16,
"learning_rate": 0.0004233299899699097,
"loss": 2.8152,
"theoretical_loss": 3.8872788006085894,
"tokens_seen": 534118400
},
{
"epoch": 0.16,
"learning_rate": 0.00042331995987963895,
"loss": 3.1119,
"theoretical_loss": 3.8872280136141484,
"tokens_seen": 534183936
},
{
"epoch": 0.16,
"learning_rate": 0.00042330992978936813,
"loss": 2.8227,
"theoretical_loss": 3.887177234594466,
"tokens_seen": 534249472
},
{
"epoch": 0.16,
"learning_rate": 0.0004232998996990973,
"loss": 2.7367,
"theoretical_loss": 3.88712646354731,
"tokens_seen": 534315008
},
{
"epoch": 0.16,
"learning_rate": 0.0004232898696088265,
"loss": 2.8633,
"theoretical_loss": 3.8870757004704526,
"tokens_seen": 534380544
},
{
"epoch": 0.16,
"learning_rate": 0.0004232798395185557,
"loss": 2.9422,
"theoretical_loss": 3.8870249453616648,
"tokens_seen": 534446080
},
{
"epoch": 0.16,
"learning_rate": 0.00042326980942828486,
"loss": 2.8519,
"theoretical_loss": 3.8869741982187183,
"tokens_seen": 534511616
},
{
"epoch": 0.16,
"learning_rate": 0.0004232597793380141,
"loss": 3.0843,
"theoretical_loss": 3.8869234590393873,
"tokens_seen": 534577152
},
{
"epoch": 0.16,
"learning_rate": 0.0004232497492477432,
"loss": 3.0001,
"theoretical_loss": 3.886872727821446,
"tokens_seen": 534642688
},
{
"epoch": 0.16,
"learning_rate": 0.00042323971915747246,
"loss": 2.8419,
"theoretical_loss": 3.88682200456267,
"tokens_seen": 534708224
},
{
"epoch": 0.16,
"learning_rate": 0.0004232296890672016,
"loss": 2.9222,
"theoretical_loss": 3.886771289260834,
"tokens_seen": 534773760
},
{
"epoch": 0.16,
"learning_rate": 0.0004232196589769308,
"loss": 2.9703,
"theoretical_loss": 3.886720581913716,
"tokens_seen": 534839296
},
{
"epoch": 0.16,
"learning_rate": 0.00042320962888666,
"loss": 3.1677,
"theoretical_loss": 3.886669882519093,
"tokens_seen": 534904832
},
{
"epoch": 0.16,
"learning_rate": 0.0004231995987963892,
"loss": 2.9986,
"theoretical_loss": 3.886619191074744,
"tokens_seen": 534970368
},
{
"epoch": 0.16,
"learning_rate": 0.00042318956870611836,
"loss": 2.9794,
"theoretical_loss": 3.886568507578448,
"tokens_seen": 535035904
},
{
"epoch": 0.16,
"learning_rate": 0.0004231795386158476,
"loss": 2.9556,
"theoretical_loss": 3.8865178320279874,
"tokens_seen": 535101440
},
{
"epoch": 0.16,
"learning_rate": 0.0004231695085255767,
"loss": 3.1085,
"theoretical_loss": 3.886467164421142,
"tokens_seen": 535166976
},
{
"epoch": 0.16,
"learning_rate": 0.00042315947843530596,
"loss": 2.8245,
"theoretical_loss": 3.8864165047556942,
"tokens_seen": 535232512
},
{
"epoch": 0.16,
"learning_rate": 0.0004231494483450351,
"loss": 2.9032,
"theoretical_loss": 3.886365853029427,
"tokens_seen": 535298048
},
{
"epoch": 0.16,
"learning_rate": 0.0004231394182547643,
"loss": 2.9161,
"theoretical_loss": 3.8863152092401254,
"tokens_seen": 535363584
},
{
"epoch": 0.16,
"learning_rate": 0.0004231293881644935,
"loss": 2.9488,
"theoretical_loss": 3.886264573385574,
"tokens_seen": 535429120
},
{
"epoch": 0.16,
"learning_rate": 0.0004231193580742227,
"loss": 2.9453,
"theoretical_loss": 3.886213945463558,
"tokens_seen": 535494656
},
{
"epoch": 0.16,
"learning_rate": 0.00042310932798395186,
"loss": 3.054,
"theoretical_loss": 3.886163325471864,
"tokens_seen": 535560192
},
{
"epoch": 0.16,
"learning_rate": 0.00042309929789368105,
"loss": 2.8614,
"theoretical_loss": 3.8861127134082807,
"tokens_seen": 535625728
},
{
"epoch": 0.16,
"learning_rate": 0.0004230892678034102,
"loss": 3.0729,
"theoretical_loss": 3.8860621092705956,
"tokens_seen": 535691264
},
{
"epoch": 0.16,
"objective/train/docs_used": 314102,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.790719747543335,
"objective/train/theoretical_loss": 3.8860115130565984,
"objective/train/tokens_used": 556216800,
"theoretical_loss": 3.8860115130565984,
"tokens_seen": 535756800
},
{
"epoch": 0.16,
"learning_rate": 0.00042307923771313946,
"loss": 3.0928,
"theoretical_loss": 3.8860115130565984,
"tokens_seen": 535756800
},
{
"epoch": 0.16,
"learning_rate": 0.0004230692076228686,
"loss": 2.9227,
"theoretical_loss": 3.8859609247640794,
"tokens_seen": 535822336
},
{
"epoch": 0.16,
"learning_rate": 0.0004230591775325978,
"loss": 2.7427,
"theoretical_loss": 3.88591034439083,
"tokens_seen": 535887872
},
{
"epoch": 0.16,
"learning_rate": 0.00042304914744232695,
"loss": 3.0429,
"theoretical_loss": 3.885859771934642,
"tokens_seen": 535953408
},
{
"epoch": 0.16,
"learning_rate": 0.0004230391173520562,
"loss": 2.8292,
"theoretical_loss": 3.885809207393308,
"tokens_seen": 536018944
},
{
"epoch": 0.16,
"learning_rate": 0.00042302908726178537,
"loss": 2.798,
"theoretical_loss": 3.8857586507646227,
"tokens_seen": 536084480
},
{
"epoch": 0.16,
"learning_rate": 0.00042301905717151455,
"loss": 2.9397,
"theoretical_loss": 3.88570810204638,
"tokens_seen": 536150016
},
{
"epoch": 0.16,
"learning_rate": 0.00042300902708124373,
"loss": 2.7327,
"theoretical_loss": 3.8856575612363753,
"tokens_seen": 536215552
},
{
"epoch": 0.16,
"learning_rate": 0.00042299899699097297,
"loss": 2.9514,
"theoretical_loss": 3.8856070283324056,
"tokens_seen": 536281088
},
{
"epoch": 0.16,
"learning_rate": 0.0004229889669007021,
"loss": 2.7247,
"theoretical_loss": 3.885556503332268,
"tokens_seen": 536346624
},
{
"epoch": 0.16,
"learning_rate": 0.00042297893681043133,
"loss": 3.0424,
"theoretical_loss": 3.8855059862337606,
"tokens_seen": 536412160
},
{
"epoch": 0.16,
"learning_rate": 0.00042296890672016045,
"loss": 3.0064,
"theoretical_loss": 3.885455477034683,
"tokens_seen": 536477696
},
{
"epoch": 0.16,
"learning_rate": 0.0004229588766298897,
"loss": 2.8565,
"theoretical_loss": 3.8854049757328353,
"tokens_seen": 536543232
},
{
"epoch": 0.16,
"learning_rate": 0.00042294884653961887,
"loss": 2.9559,
"theoretical_loss": 3.8853544823260178,
"tokens_seen": 536608768
},
{
"epoch": 0.16,
"learning_rate": 0.00042293881644934805,
"loss": 2.8287,
"theoretical_loss": 3.8853039968120324,
"tokens_seen": 536674304
},
{
"epoch": 0.16,
"learning_rate": 0.00042292878635907723,
"loss": 2.909,
"theoretical_loss": 3.885253519188682,
"tokens_seen": 536739840
},
{
"epoch": 0.16,
"learning_rate": 0.0004229187562688064,
"loss": 3.0058,
"theoretical_loss": 3.885203049453769,
"tokens_seen": 536805376
},
{
"epoch": 0.16,
"learning_rate": 0.0004229087261785356,
"loss": 2.9078,
"theoretical_loss": 3.8851525876050994,
"tokens_seen": 536870912
},
{
"epoch": 0.16,
"learning_rate": 0.00042289869608826483,
"loss": 2.6801,
"theoretical_loss": 3.885102133640477,
"tokens_seen": 536936448
},
{
"epoch": 0.16,
"learning_rate": 0.00042288866599799396,
"loss": 2.8486,
"theoretical_loss": 3.8850516875577092,
"tokens_seen": 537001984
},
{
"epoch": 0.16,
"learning_rate": 0.0004228786359077232,
"loss": 2.9838,
"theoretical_loss": 3.8850012493546027,
"tokens_seen": 537067520
},
{
"epoch": 0.16,
"learning_rate": 0.0004228686058174523,
"loss": 2.9971,
"theoretical_loss": 3.884950819028965,
"tokens_seen": 537133056
},
{
"epoch": 0.16,
"learning_rate": 0.00042285857572718155,
"loss": 2.7513,
"theoretical_loss": 3.884900396578605,
"tokens_seen": 537198592
},
{
"epoch": 0.16,
"learning_rate": 0.00042284854563691074,
"loss": 2.9549,
"theoretical_loss": 3.8848499820013327,
"tokens_seen": 537264128
},
{
"epoch": 0.16,
"learning_rate": 0.0004228385155466399,
"loss": 2.9562,
"theoretical_loss": 3.884799575294958,
"tokens_seen": 537329664
},
{
"epoch": 0.16,
"objective/train/docs_used": 314596,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.16971755027771,
"objective/train/theoretical_loss": 3.8847491764572926,
"objective/train/tokens_used": 557855200,
"theoretical_loss": 3.8847491764572926,
"tokens_seen": 537395200
},
{
"epoch": 0.16,
"learning_rate": 0.0004228284854563691,
"loss": 2.9893,
"theoretical_loss": 3.8847491764572926,
"tokens_seen": 537395200
},
{
"epoch": 0.16,
"learning_rate": 0.00042281845536609833,
"loss": 2.9383,
"theoretical_loss": 3.884698785486149,
"tokens_seen": 537460736
},
{
"epoch": 0.16,
"learning_rate": 0.00042280842527582746,
"loss": 2.5575,
"theoretical_loss": 3.884648402379341,
"tokens_seen": 537526272
},
{
"epoch": 0.16,
"learning_rate": 0.0004227983951855567,
"loss": 2.6068,
"theoretical_loss": 3.8845980271346816,
"tokens_seen": 537591808
},
{
"epoch": 0.16,
"learning_rate": 0.0004227883650952858,
"loss": 2.8686,
"theoretical_loss": 3.8845476597499857,
"tokens_seen": 537657344
},
{
"epoch": 0.16,
"learning_rate": 0.00042277833500501506,
"loss": 3.0102,
"theoretical_loss": 3.884497300223069,
"tokens_seen": 537722880
},
{
"epoch": 0.16,
"learning_rate": 0.00042276830491474424,
"loss": 2.943,
"theoretical_loss": 3.8844469485517488,
"tokens_seen": 537788416
},
{
"epoch": 0.16,
"learning_rate": 0.0004227582748244734,
"loss": 2.748,
"theoretical_loss": 3.8843966047338423,
"tokens_seen": 537853952
},
{
"epoch": 0.16,
"learning_rate": 0.0004227482447342026,
"loss": 3.0621,
"theoretical_loss": 3.8843462687671684,
"tokens_seen": 537919488
},
{
"epoch": 0.16,
"learning_rate": 0.0004227382146439318,
"loss": 2.7502,
"theoretical_loss": 3.8842959406495456,
"tokens_seen": 537985024
},
{
"epoch": 0.16,
"learning_rate": 0.00042272818455366096,
"loss": 3.0579,
"theoretical_loss": 3.884245620378794,
"tokens_seen": 538050560
},
{
"epoch": 0.16,
"learning_rate": 0.0004227181544633902,
"loss": 2.8916,
"theoretical_loss": 3.8841953079527354,
"tokens_seen": 538116096
},
{
"epoch": 0.16,
"learning_rate": 0.0004227081243731193,
"loss": 2.8935,
"theoretical_loss": 3.8841450033691913,
"tokens_seen": 538181632
},
{
"epoch": 0.16,
"learning_rate": 0.00042269809428284856,
"loss": 2.9873,
"theoretical_loss": 3.8840947066259845,
"tokens_seen": 538247168
},
{
"epoch": 0.16,
"learning_rate": 0.0004226880641925777,
"loss": 2.7005,
"theoretical_loss": 3.884044417720938,
"tokens_seen": 538312704
},
{
"epoch": 0.16,
"learning_rate": 0.0004226780341023069,
"loss": 2.5778,
"theoretical_loss": 3.8839941366518778,
"tokens_seen": 538378240
},
{
"epoch": 0.16,
"learning_rate": 0.00042266800401203616,
"loss": 2.8099,
"theoretical_loss": 3.8839438634166275,
"tokens_seen": 538443776
},
{
"epoch": 0.16,
"learning_rate": 0.0004226579739217653,
"loss": 2.8237,
"theoretical_loss": 3.883893598013014,
"tokens_seen": 538509312
},
{
"epoch": 0.16,
"learning_rate": 0.0004226479438314945,
"loss": 2.7426,
"theoretical_loss": 3.883843340438865,
"tokens_seen": 538574848
},
{
"epoch": 0.16,
"learning_rate": 0.0004226379137412237,
"loss": 2.84,
"theoretical_loss": 3.8837930906920075,
"tokens_seen": 538640384
},
{
"epoch": 0.16,
"learning_rate": 0.0004226278836509529,
"loss": 2.5758,
"theoretical_loss": 3.8837428487702716,
"tokens_seen": 538705920
},
{
"epoch": 0.16,
"learning_rate": 0.00042261785356068206,
"loss": 2.9282,
"theoretical_loss": 3.8836926146714856,
"tokens_seen": 538771456
},
{
"epoch": 0.16,
"learning_rate": 0.00042260782347041125,
"loss": 2.6134,
"theoretical_loss": 3.883642388393481,
"tokens_seen": 538836992
},
{
"epoch": 0.16,
"learning_rate": 0.0004225977933801404,
"loss": 2.821,
"theoretical_loss": 3.8835921699340887,
"tokens_seen": 538902528
},
{
"epoch": 0.16,
"learning_rate": 0.00042258776328986966,
"loss": 2.9375,
"theoretical_loss": 3.8835419592911413,
"tokens_seen": 538968064
},
{
"epoch": 0.16,
"objective/train/docs_used": 315427,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6693382263183594,
"objective/train/theoretical_loss": 3.8834917564624716,
"objective/train/tokens_used": 559493600,
"theoretical_loss": 3.8834917564624716,
"tokens_seen": 539033600
},
{
"epoch": 0.16,
"learning_rate": 0.0004225777331995988,
"loss": 2.8935,
"theoretical_loss": 3.8834917564624716,
"tokens_seen": 539033600
},
{
"epoch": 0.16,
"learning_rate": 0.000422567703109328,
"loss": 2.8414,
"theoretical_loss": 3.883441561445914,
"tokens_seen": 539099136
},
{
"epoch": 0.16,
"learning_rate": 0.00042255767301905715,
"loss": 3.1531,
"theoretical_loss": 3.8833913742393036,
"tokens_seen": 539164672
},
{
"epoch": 0.16,
"learning_rate": 0.0004225476429287864,
"loss": 2.7758,
"theoretical_loss": 3.8833411948404755,
"tokens_seen": 539230208
},
{
"epoch": 0.16,
"learning_rate": 0.00042253761283851557,
"loss": 2.9681,
"theoretical_loss": 3.883291023247267,
"tokens_seen": 539295744
},
{
"epoch": 0.16,
"learning_rate": 0.00042252758274824475,
"loss": 2.8058,
"theoretical_loss": 3.8832408594575147,
"tokens_seen": 539361280
},
{
"epoch": 0.16,
"learning_rate": 0.00042251755265797393,
"loss": 2.8653,
"theoretical_loss": 3.883190703469058,
"tokens_seen": 539426816
},
{
"epoch": 0.16,
"learning_rate": 0.00042250752256770317,
"loss": 2.8425,
"theoretical_loss": 3.883140555279735,
"tokens_seen": 539492352
},
{
"epoch": 0.16,
"learning_rate": 0.0004224974924774323,
"loss": 2.942,
"theoretical_loss": 3.883090414887387,
"tokens_seen": 539557888
},
{
"epoch": 0.16,
"learning_rate": 0.00042248746238716153,
"loss": 3.0772,
"theoretical_loss": 3.8830402822898535,
"tokens_seen": 539623424
},
{
"epoch": 0.16,
"learning_rate": 0.00042247743229689065,
"loss": 2.6957,
"theoretical_loss": 3.8829901574849774,
"tokens_seen": 539688960
},
{
"epoch": 0.16,
"learning_rate": 0.0004224674022066199,
"loss": 2.8871,
"theoretical_loss": 3.8829400404706003,
"tokens_seen": 539754496
},
{
"epoch": 0.16,
"learning_rate": 0.00042245737211634907,
"loss": 2.855,
"theoretical_loss": 3.8828899312445673,
"tokens_seen": 539820032
},
{
"epoch": 0.16,
"learning_rate": 0.00042244734202607825,
"loss": 2.9288,
"theoretical_loss": 3.8828398298047206,
"tokens_seen": 539885568
},
{
"epoch": 0.16,
"learning_rate": 0.00042243731193580743,
"loss": 3.0415,
"theoretical_loss": 3.8827897361489074,
"tokens_seen": 539951104
},
{
"epoch": 0.16,
"learning_rate": 0.0004224272818455366,
"loss": 2.9481,
"theoretical_loss": 3.8827396502749734,
"tokens_seen": 540016640
},
{
"epoch": 0.16,
"learning_rate": 0.0004224172517552658,
"loss": 3.0045,
"theoretical_loss": 3.882689572180764,
"tokens_seen": 540082176
},
{
"epoch": 0.16,
"learning_rate": 0.00042240722166499503,
"loss": 2.6324,
"theoretical_loss": 3.882639501864129,
"tokens_seen": 540147712
},
{
"epoch": 0.16,
"learning_rate": 0.00042239719157472416,
"loss": 2.7887,
"theoretical_loss": 3.8825894393229152,
"tokens_seen": 540213248
},
{
"epoch": 0.16,
"learning_rate": 0.0004223871614844534,
"loss": 2.7621,
"theoretical_loss": 3.882539384554974,
"tokens_seen": 540278784
},
{
"epoch": 0.16,
"learning_rate": 0.0004223771313941825,
"loss": 2.8169,
"theoretical_loss": 3.8824893375581544,
"tokens_seen": 540344320
},
{
"epoch": 0.16,
"learning_rate": 0.00042236710130391176,
"loss": 2.771,
"theoretical_loss": 3.8824392983303078,
"tokens_seen": 540409856
},
{
"epoch": 0.16,
"learning_rate": 0.00042235707121364094,
"loss": 2.7474,
"theoretical_loss": 3.8823892668692865,
"tokens_seen": 540475392
},
{
"epoch": 0.16,
"learning_rate": 0.0004223470411233701,
"loss": 3.0539,
"theoretical_loss": 3.8823392431729435,
"tokens_seen": 540540928
},
{
"epoch": 0.16,
"learning_rate": 0.0004223370110330993,
"loss": 2.9556,
"theoretical_loss": 3.8822892272391325,
"tokens_seen": 540606464
},
{
"epoch": 0.16,
"objective/train/docs_used": 316530,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7001001834869385,
"objective/train/theoretical_loss": 3.882239219065708,
"objective/train/tokens_used": 561132000,
"theoretical_loss": 3.882239219065708,
"tokens_seen": 540672000
},
{
"epoch": 0.16,
"learning_rate": 0.00042232698094282853,
"loss": 2.8921,
"theoretical_loss": 3.882239219065708,
"tokens_seen": 540672000
},
{
"epoch": 0.16,
"learning_rate": 0.00042231695085255766,
"loss": 2.7654,
"theoretical_loss": 3.882189218650525,
"tokens_seen": 540737536
},
{
"epoch": 0.16,
"learning_rate": 0.0004223069207622869,
"loss": 2.7423,
"theoretical_loss": 3.882139225991441,
"tokens_seen": 540803072
},
{
"epoch": 0.16,
"learning_rate": 0.000422296890672016,
"loss": 2.752,
"theoretical_loss": 3.8820892410863124,
"tokens_seen": 540868608
},
{
"epoch": 0.16,
"learning_rate": 0.00042228686058174526,
"loss": 2.8386,
"theoretical_loss": 3.882039263932997,
"tokens_seen": 540934144
},
{
"epoch": 0.16,
"learning_rate": 0.00042227683049147444,
"loss": 2.7466,
"theoretical_loss": 3.8819892945293546,
"tokens_seen": 540999680
},
{
"epoch": 0.16,
"learning_rate": 0.0004222668004012036,
"loss": 3.0137,
"theoretical_loss": 3.8819393328732446,
"tokens_seen": 541065216
},
{
"epoch": 0.16,
"learning_rate": 0.0004222567703109328,
"loss": 2.9363,
"theoretical_loss": 3.8818893789625273,
"tokens_seen": 541130752
},
{
"epoch": 0.16,
"learning_rate": 0.000422246740220662,
"loss": 2.821,
"theoretical_loss": 3.881839432795064,
"tokens_seen": 541196288
},
{
"epoch": 0.16,
"learning_rate": 0.00042223671013039116,
"loss": 2.8432,
"theoretical_loss": 3.881789494368717,
"tokens_seen": 541261824
},
{
"epoch": 0.16,
"learning_rate": 0.0004222266800401204,
"loss": 2.7617,
"theoretical_loss": 3.88173956368135,
"tokens_seen": 541327360
},
{
"epoch": 0.16,
"learning_rate": 0.0004222166499498495,
"loss": 2.9612,
"theoretical_loss": 3.8816896407308263,
"tokens_seen": 541392896
},
{
"epoch": 0.16,
"learning_rate": 0.00042220661985957876,
"loss": 2.7397,
"theoretical_loss": 3.8816397255150115,
"tokens_seen": 541458432
},
{
"epoch": 0.16,
"learning_rate": 0.0004221965897693079,
"loss": 2.9595,
"theoretical_loss": 3.8815898180317703,
"tokens_seen": 541523968
},
{
"epoch": 0.16,
"learning_rate": 0.0004221865596790371,
"loss": 2.9104,
"theoretical_loss": 3.8815399182789703,
"tokens_seen": 541589504
},
{
"epoch": 0.16,
"learning_rate": 0.0004221765295887663,
"loss": 2.7983,
"theoretical_loss": 3.881490026254478,
"tokens_seen": 541655040
},
{
"epoch": 0.16,
"learning_rate": 0.0004221664994984955,
"loss": 2.832,
"theoretical_loss": 3.881440141956162,
"tokens_seen": 541720576
},
{
"epoch": 0.16,
"learning_rate": 0.00042215646940822467,
"loss": 2.8153,
"theoretical_loss": 3.8813902653818912,
"tokens_seen": 541786112
},
{
"epoch": 0.16,
"learning_rate": 0.0004221464393179539,
"loss": 2.8717,
"theoretical_loss": 3.8813403965295357,
"tokens_seen": 541851648
},
{
"epoch": 0.16,
"learning_rate": 0.00042213640922768303,
"loss": 2.8259,
"theoretical_loss": 3.8812905353969667,
"tokens_seen": 541917184
},
{
"epoch": 0.16,
"learning_rate": 0.00042212637913741226,
"loss": 2.8444,
"theoretical_loss": 3.881240681982054,
"tokens_seen": 541982720
},
{
"epoch": 0.16,
"learning_rate": 0.0004221163490471414,
"loss": 2.8232,
"theoretical_loss": 3.8811908362826726,
"tokens_seen": 542048256
},
{
"epoch": 0.16,
"learning_rate": 0.0004221063189568706,
"loss": 2.9688,
"theoretical_loss": 3.881140998296694,
"tokens_seen": 542113792
},
{
"epoch": 0.16,
"learning_rate": 0.0004220962888665998,
"loss": 2.7413,
"theoretical_loss": 3.8810911680219933,
"tokens_seen": 542179328
},
{
"epoch": 0.16,
"learning_rate": 0.000422086258776329,
"loss": 2.762,
"theoretical_loss": 3.881041345456444,
"tokens_seen": 542244864
},
{
"epoch": 0.16,
"objective/train/docs_used": 317270,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.066617727279663,
"objective/train/theoretical_loss": 3.8809915305979237,
"objective/train/tokens_used": 562770400,
"theoretical_loss": 3.8809915305979237,
"tokens_seen": 542310400
},
{
"epoch": 0.16,
"learning_rate": 0.00042207622868605817,
"loss": 2.8901,
"theoretical_loss": 3.8809915305979237,
"tokens_seen": 542310400
},
{
"epoch": 0.16,
"learning_rate": 0.00042206619859578735,
"loss": 2.9818,
"theoretical_loss": 3.880941723444308,
"tokens_seen": 542375936
},
{
"epoch": 0.16,
"learning_rate": 0.00042205616850551653,
"loss": 2.7985,
"theoretical_loss": 3.8808919239934747,
"tokens_seen": 542441472
},
{
"epoch": 0.16,
"learning_rate": 0.00042204613841524577,
"loss": 2.7908,
"theoretical_loss": 3.8808421322433024,
"tokens_seen": 542507008
},
{
"epoch": 0.16,
"learning_rate": 0.0004220361083249749,
"loss": 2.7691,
"theoretical_loss": 3.8807923481916697,
"tokens_seen": 542572544
},
{
"epoch": 0.16,
"learning_rate": 0.00042202607823470413,
"loss": 2.7955,
"theoretical_loss": 3.880742571836457,
"tokens_seen": 542638080
},
{
"epoch": 0.16,
"learning_rate": 0.0004220160481444333,
"loss": 2.6381,
"theoretical_loss": 3.8806928031755454,
"tokens_seen": 542703616
},
{
"epoch": 0.16,
"learning_rate": 0.0004220060180541625,
"loss": 2.9938,
"theoretical_loss": 3.880643042206816,
"tokens_seen": 542769152
},
{
"epoch": 0.16,
"learning_rate": 0.0004219959879638917,
"loss": 2.7158,
"theoretical_loss": 3.880593288928152,
"tokens_seen": 542834688
},
{
"epoch": 0.16,
"learning_rate": 0.00042198595787362085,
"loss": 2.9342,
"theoretical_loss": 3.8805435433374353,
"tokens_seen": 542900224
},
{
"epoch": 0.16,
"learning_rate": 0.00042197592778335004,
"loss": 2.9032,
"theoretical_loss": 3.880493805432552,
"tokens_seen": 542965760
},
{
"epoch": 0.16,
"learning_rate": 0.00042196589769307927,
"loss": 2.7669,
"theoretical_loss": 3.8804440752113862,
"tokens_seen": 543031296
},
{
"epoch": 0.16,
"learning_rate": 0.0004219558676028084,
"loss": 2.9817,
"theoretical_loss": 3.880394352671824,
"tokens_seen": 543096832
},
{
"epoch": 0.16,
"learning_rate": 0.00042194583751253763,
"loss": 2.9951,
"theoretical_loss": 3.880344637811752,
"tokens_seen": 543162368
},
{
"epoch": 0.16,
"learning_rate": 0.00042193580742226676,
"loss": 3.1305,
"theoretical_loss": 3.880294930629058,
"tokens_seen": 543227904
},
{
"epoch": 0.16,
"learning_rate": 0.000421925777331996,
"loss": 2.7982,
"theoretical_loss": 3.8802452311216302,
"tokens_seen": 543293440
},
{
"epoch": 0.16,
"learning_rate": 0.00042191574724172523,
"loss": 2.9459,
"theoretical_loss": 3.880195539287358,
"tokens_seen": 543358976
},
{
"epoch": 0.16,
"learning_rate": 0.00042190571715145436,
"loss": 2.7107,
"theoretical_loss": 3.8801458551241312,
"tokens_seen": 543424512
},
{
"epoch": 0.16,
"learning_rate": 0.0004218956870611836,
"loss": 2.9262,
"theoretical_loss": 3.8800961786298407,
"tokens_seen": 543490048
},
{
"epoch": 0.16,
"learning_rate": 0.0004218856569709127,
"loss": 2.9468,
"theoretical_loss": 3.880046509802378,
"tokens_seen": 543555584
},
{
"epoch": 0.16,
"learning_rate": 0.00042187562688064196,
"loss": 2.8462,
"theoretical_loss": 3.879996848639637,
"tokens_seen": 543621120
},
{
"epoch": 0.16,
"learning_rate": 0.00042186559679037114,
"loss": 2.8692,
"theoretical_loss": 3.8799471951395095,
"tokens_seen": 543686656
},
{
"epoch": 0.16,
"learning_rate": 0.0004218555667001003,
"loss": 2.7989,
"theoretical_loss": 3.8798975492998906,
"tokens_seen": 543752192
},
{
"epoch": 0.16,
"learning_rate": 0.0004218455366098295,
"loss": 2.9146,
"theoretical_loss": 3.8798479111186754,
"tokens_seen": 543817728
},
{
"epoch": 0.16,
"learning_rate": 0.00042183550651955873,
"loss": 2.9114,
"theoretical_loss": 3.879798280593759,
"tokens_seen": 543883264
},
{
"epoch": 0.16,
"objective/train/docs_used": 318285,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9519379138946533,
"objective/train/theoretical_loss": 3.879748657723039,
"objective/train/tokens_used": 564408800,
"theoretical_loss": 3.879748657723039,
"tokens_seen": 543948800
},
{
"epoch": 0.16,
"learning_rate": 0.00042182547642928786,
"loss": 2.8573,
"theoretical_loss": 3.879748657723039,
"tokens_seen": 543948800
},
{
"epoch": 0.16,
"learning_rate": 0.0004218154463390171,
"loss": 2.7129,
"theoretical_loss": 3.8796990425044124,
"tokens_seen": 544014336
},
{
"epoch": 0.16,
"learning_rate": 0.0004218054162487462,
"loss": 2.9859,
"theoretical_loss": 3.879649434935778,
"tokens_seen": 544079872
},
{
"epoch": 0.16,
"learning_rate": 0.00042179538615847546,
"loss": 2.7631,
"theoretical_loss": 3.8795998350150347,
"tokens_seen": 544145408
},
{
"epoch": 0.16,
"learning_rate": 0.00042178535606820464,
"loss": 2.7474,
"theoretical_loss": 3.879550242740083,
"tokens_seen": 544210944
},
{
"epoch": 0.16,
"learning_rate": 0.0004217753259779338,
"loss": 2.88,
"theoretical_loss": 3.8795006581088236,
"tokens_seen": 544276480
},
{
"epoch": 0.16,
"learning_rate": 0.000421765295887663,
"loss": 2.8279,
"theoretical_loss": 3.879451081119158,
"tokens_seen": 544342016
},
{
"epoch": 0.16,
"learning_rate": 0.0004217552657973922,
"loss": 2.8613,
"theoretical_loss": 3.8794015117689886,
"tokens_seen": 544407552
},
{
"epoch": 0.16,
"learning_rate": 0.00042174523570712136,
"loss": 2.9193,
"theoretical_loss": 3.8793519500562192,
"tokens_seen": 544473088
},
{
"epoch": 0.17,
"learning_rate": 0.0004217352056168506,
"loss": 2.8995,
"theoretical_loss": 3.879302395978754,
"tokens_seen": 544538624
},
{
"epoch": 0.17,
"learning_rate": 0.0004217251755265797,
"loss": 2.848,
"theoretical_loss": 3.879252849534498,
"tokens_seen": 544604160
},
{
"epoch": 0.17,
"learning_rate": 0.00042171514543630896,
"loss": 3.0245,
"theoretical_loss": 3.8792033107213566,
"tokens_seen": 544669696
},
{
"epoch": 0.17,
"learning_rate": 0.0004217051153460381,
"loss": 2.7713,
"theoretical_loss": 3.8791537795372366,
"tokens_seen": 544735232
},
{
"epoch": 0.17,
"learning_rate": 0.0004216950852557673,
"loss": 2.86,
"theoretical_loss": 3.879104255980046,
"tokens_seen": 544800768
},
{
"epoch": 0.17,
"learning_rate": 0.0004216850551654965,
"loss": 2.9259,
"theoretical_loss": 3.8790547400476925,
"tokens_seen": 544866304
},
{
"epoch": 0.17,
"learning_rate": 0.0004216750250752257,
"loss": 2.6663,
"theoretical_loss": 3.879005231738086,
"tokens_seen": 544931840
},
{
"epoch": 0.17,
"learning_rate": 0.00042166499498495487,
"loss": 2.875,
"theoretical_loss": 3.878955731049136,
"tokens_seen": 544997376
},
{
"epoch": 0.17,
"learning_rate": 0.0004216549648946841,
"loss": 3.1208,
"theoretical_loss": 3.878906237978753,
"tokens_seen": 545062912
},
{
"epoch": 0.17,
"learning_rate": 0.00042164493480441323,
"loss": 2.6814,
"theoretical_loss": 3.8788567525248494,
"tokens_seen": 545128448
},
{
"epoch": 0.17,
"learning_rate": 0.00042163490471414246,
"loss": 2.947,
"theoretical_loss": 3.878807274685337,
"tokens_seen": 545193984
},
{
"epoch": 0.17,
"learning_rate": 0.0004216248746238716,
"loss": 2.8911,
"theoretical_loss": 3.8787578044581292,
"tokens_seen": 545259520
},
{
"epoch": 0.17,
"learning_rate": 0.0004216148445336008,
"loss": 2.9855,
"theoretical_loss": 3.87870834184114,
"tokens_seen": 545325056
},
{
"epoch": 0.17,
"learning_rate": 0.00042160481444333,
"loss": 3.1216,
"theoretical_loss": 3.8786588868322847,
"tokens_seen": 545390592
},
{
"epoch": 0.17,
"learning_rate": 0.0004215947843530592,
"loss": 2.7415,
"theoretical_loss": 3.8786094394294794,
"tokens_seen": 545456128
},
{
"epoch": 0.17,
"learning_rate": 0.00042158475426278837,
"loss": 2.9065,
"theoretical_loss": 3.8785599996306397,
"tokens_seen": 545521664
},
{
"epoch": 0.17,
"objective/train/docs_used": 318961,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9253644943237305,
"objective/train/theoretical_loss": 3.8785105674336835,
"objective/train/tokens_used": 566047200,
"theoretical_loss": 3.8785105674336835,
"tokens_seen": 545587200
},
{
"epoch": 0.17,
"learning_rate": 0.00042157472417251755,
"loss": 2.9532,
"theoretical_loss": 3.8785105674336835,
"tokens_seen": 545587200
},
{
"epoch": 0.17,
"learning_rate": 0.00042156469408224673,
"loss": 3.0437,
"theoretical_loss": 3.8784611428365285,
"tokens_seen": 545652736
},
{
"epoch": 0.17,
"learning_rate": 0.00042155466399197597,
"loss": 2.7187,
"theoretical_loss": 3.8784117258370943,
"tokens_seen": 545718272
},
{
"epoch": 0.17,
"learning_rate": 0.0004215446339017051,
"loss": 2.7803,
"theoretical_loss": 3.878362316433301,
"tokens_seen": 545783808
},
{
"epoch": 0.17,
"learning_rate": 0.00042153460381143433,
"loss": 2.9175,
"theoretical_loss": 3.878312914623068,
"tokens_seen": 545849344
},
{
"epoch": 0.17,
"learning_rate": 0.0004215245737211635,
"loss": 3.0657,
"theoretical_loss": 3.8782635204043188,
"tokens_seen": 545914880
},
{
"epoch": 0.17,
"learning_rate": 0.0004215145436308927,
"loss": 3.1568,
"theoretical_loss": 3.8782141337749736,
"tokens_seen": 545980416
},
{
"epoch": 0.17,
"learning_rate": 0.0004215045135406219,
"loss": 3.1295,
"theoretical_loss": 3.8781647547329565,
"tokens_seen": 546045952
},
{
"epoch": 0.17,
"learning_rate": 0.00042149448345035105,
"loss": 2.8782,
"theoretical_loss": 3.8781153832761914,
"tokens_seen": 546111488
},
{
"epoch": 0.17,
"learning_rate": 0.00042148445336008024,
"loss": 2.8349,
"theoretical_loss": 3.8780660194026035,
"tokens_seen": 546177024
},
{
"epoch": 0.17,
"learning_rate": 0.00042147442326980947,
"loss": 2.8772,
"theoretical_loss": 3.8780166631101176,
"tokens_seen": 546242560
},
{
"epoch": 0.17,
"learning_rate": 0.0004214643931795386,
"loss": 2.7605,
"theoretical_loss": 3.8779673143966606,
"tokens_seen": 546308096
},
{
"epoch": 0.17,
"learning_rate": 0.00042145436308926783,
"loss": 2.7632,
"theoretical_loss": 3.8779179732601587,
"tokens_seen": 546373632
},
{
"epoch": 0.17,
"learning_rate": 0.00042144433299899696,
"loss": 2.7438,
"theoretical_loss": 3.8778686396985416,
"tokens_seen": 546439168
},
{
"epoch": 0.17,
"learning_rate": 0.0004214343029087262,
"loss": 3.0832,
"theoretical_loss": 3.877819313709736,
"tokens_seen": 546504704
},
{
"epoch": 0.17,
"learning_rate": 0.0004214242728184554,
"loss": 2.8661,
"theoretical_loss": 3.8777699952916738,
"tokens_seen": 546570240
},
{
"epoch": 0.17,
"learning_rate": 0.00042141424272818456,
"loss": 2.748,
"theoretical_loss": 3.8777206844422842,
"tokens_seen": 546635776
},
{
"epoch": 0.17,
"learning_rate": 0.00042140421263791374,
"loss": 2.7212,
"theoretical_loss": 3.877671381159499,
"tokens_seen": 546701312
},
{
"epoch": 0.17,
"learning_rate": 0.0004213941825476429,
"loss": 2.6799,
"theoretical_loss": 3.8776220854412493,
"tokens_seen": 546766848
},
{
"epoch": 0.17,
"learning_rate": 0.0004213841524573721,
"loss": 2.8845,
"theoretical_loss": 3.877572797285469,
"tokens_seen": 546832384
},
{
"epoch": 0.17,
"learning_rate": 0.00042137412236710134,
"loss": 3.0905,
"theoretical_loss": 3.8775235166900917,
"tokens_seen": 546897920
},
{
"epoch": 0.17,
"learning_rate": 0.00042136409227683046,
"loss": 2.9857,
"theoretical_loss": 3.877474243653051,
"tokens_seen": 546963456
},
{
"epoch": 0.17,
"learning_rate": 0.0004213540621865597,
"loss": 2.8874,
"theoretical_loss": 3.8774249781722836,
"tokens_seen": 547028992
},
{
"epoch": 0.17,
"learning_rate": 0.0004213440320962889,
"loss": 2.9957,
"theoretical_loss": 3.877375720245725,
"tokens_seen": 547094528
},
{
"epoch": 0.17,
"learning_rate": 0.00042133400200601806,
"loss": 2.8372,
"theoretical_loss": 3.8773264698713117,
"tokens_seen": 547160064
},
{
"epoch": 0.17,
"objective/train/docs_used": 320353,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.186674118041992,
"objective/train/theoretical_loss": 3.8772772270469824,
"objective/train/tokens_used": 567685600,
"theoretical_loss": 3.8772772270469824,
"tokens_seen": 547225600
},
{
"epoch": 0.17,
"learning_rate": 0.00042132397191574724,
"loss": 3.0207,
"theoretical_loss": 3.8772772270469824,
"tokens_seen": 547225600
},
{
"epoch": 0.17,
"learning_rate": 0.0004213139418254764,
"loss": 2.9139,
"theoretical_loss": 3.877227991770675,
"tokens_seen": 547291136
},
{
"epoch": 0.17,
"learning_rate": 0.0004213039117352056,
"loss": 2.9367,
"theoretical_loss": 3.877178764040329,
"tokens_seen": 547356672
},
{
"epoch": 0.17,
"learning_rate": 0.00042129388164493484,
"loss": 2.9502,
"theoretical_loss": 3.8771295438538846,
"tokens_seen": 547422208
},
{
"epoch": 0.17,
"learning_rate": 0.00042128385155466397,
"loss": 2.9549,
"theoretical_loss": 3.8770803312092834,
"tokens_seen": 547487744
},
{
"epoch": 0.17,
"learning_rate": 0.0004212738214643932,
"loss": 2.8362,
"theoretical_loss": 3.877031126104466,
"tokens_seen": 547553280
},
{
"epoch": 0.17,
"learning_rate": 0.00042126379137412233,
"loss": 2.7986,
"theoretical_loss": 3.8769819285373766,
"tokens_seen": 547618816
},
{
"epoch": 0.17,
"learning_rate": 0.00042125376128385156,
"loss": 2.8222,
"theoretical_loss": 3.876932738505957,
"tokens_seen": 547684352
},
{
"epoch": 0.17,
"learning_rate": 0.00042124373119358075,
"loss": 2.9447,
"theoretical_loss": 3.8768835560081527,
"tokens_seen": 547749888
},
{
"epoch": 0.17,
"learning_rate": 0.0004212337011033099,
"loss": 2.7848,
"theoretical_loss": 3.876834381041908,
"tokens_seen": 547815424
},
{
"epoch": 0.17,
"learning_rate": 0.0004212236710130391,
"loss": 2.8295,
"theoretical_loss": 3.8767852136051686,
"tokens_seen": 547880960
},
{
"epoch": 0.17,
"learning_rate": 0.0004212136409227683,
"loss": 2.7802,
"theoretical_loss": 3.8767360536958817,
"tokens_seen": 547946496
},
{
"epoch": 0.17,
"learning_rate": 0.00042120361083249747,
"loss": 2.8975,
"theoretical_loss": 3.876686901311995,
"tokens_seen": 548012032
},
{
"epoch": 0.17,
"learning_rate": 0.0004211935807422267,
"loss": 2.9494,
"theoretical_loss": 3.8766377564514563,
"tokens_seen": 548077568
},
{
"epoch": 0.17,
"learning_rate": 0.00042118355065195583,
"loss": 2.7224,
"theoretical_loss": 3.8765886191122148,
"tokens_seen": 548143104
},
{
"epoch": 0.17,
"learning_rate": 0.00042117352056168507,
"loss": 2.9374,
"theoretical_loss": 3.87653948929222,
"tokens_seen": 548208640
},
{
"epoch": 0.17,
"learning_rate": 0.0004211634904714143,
"loss": 3.0219,
"theoretical_loss": 3.876490366989423,
"tokens_seen": 548274176
},
{
"epoch": 0.17,
"learning_rate": 0.00042115346038114343,
"loss": 2.974,
"theoretical_loss": 3.876441252201775,
"tokens_seen": 548339712
},
{
"epoch": 0.17,
"learning_rate": 0.00042114343029087266,
"loss": 2.9682,
"theoretical_loss": 3.876392144927229,
"tokens_seen": 548405248
},
{
"epoch": 0.17,
"learning_rate": 0.0004211334002006018,
"loss": 2.8559,
"theoretical_loss": 3.876343045163737,
"tokens_seen": 548470784
},
{
"epoch": 0.17,
"learning_rate": 0.00042112337011033103,
"loss": 3.0035,
"theoretical_loss": 3.8762939529092537,
"tokens_seen": 548536320
},
{
"epoch": 0.17,
"learning_rate": 0.0004211133400200602,
"loss": 2.7531,
"theoretical_loss": 3.8762448681617334,
"tokens_seen": 548601856
},
{
"epoch": 0.17,
"learning_rate": 0.0004211033099297894,
"loss": 2.7791,
"theoretical_loss": 3.876195790919132,
"tokens_seen": 548667392
},
{
"epoch": 0.17,
"learning_rate": 0.00042109327983951857,
"loss": 2.74,
"theoretical_loss": 3.876146721179405,
"tokens_seen": 548732928
},
{
"epoch": 0.17,
"learning_rate": 0.00042108324974924775,
"loss": 2.8013,
"theoretical_loss": 3.8760976589405107,
"tokens_seen": 548798464
},
{
"epoch": 0.17,
"objective/train/docs_used": 320989,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.815619945526123,
"objective/train/theoretical_loss": 3.8760486042004056,
"objective/train/tokens_used": 569324000,
"theoretical_loss": 3.8760486042004056,
"tokens_seen": 548864000
},
{
"epoch": 0.17,
"learning_rate": 0.00042107321965897693,
"loss": 2.6059,
"theoretical_loss": 3.8760486042004056,
"tokens_seen": 548864000
},
{
"epoch": 0.17,
"learning_rate": 0.00042106318956870617,
"loss": 2.8879,
"theoretical_loss": 3.8759995569570496,
"tokens_seen": 548929536
},
{
"epoch": 0.17,
"learning_rate": 0.0004210531594784353,
"loss": 2.746,
"theoretical_loss": 3.8759505172084014,
"tokens_seen": 548995072
},
{
"epoch": 0.17,
"learning_rate": 0.00042104312938816453,
"loss": 2.9174,
"theoretical_loss": 3.875901484952422,
"tokens_seen": 549060608
},
{
"epoch": 0.17,
"learning_rate": 0.0004210330992978937,
"loss": 2.8429,
"theoretical_loss": 3.8758524601870716,
"tokens_seen": 549126144
},
{
"epoch": 0.17,
"learning_rate": 0.0004210230692076229,
"loss": 2.7353,
"theoretical_loss": 3.8758034429103128,
"tokens_seen": 549191680
},
{
"epoch": 0.17,
"learning_rate": 0.0004210130391173521,
"loss": 2.6284,
"theoretical_loss": 3.8757544331201084,
"tokens_seen": 549257216
},
{
"epoch": 0.17,
"learning_rate": 0.00042100300902708125,
"loss": 2.7378,
"theoretical_loss": 3.875705430814421,
"tokens_seen": 549322752
},
{
"epoch": 0.17,
"learning_rate": 0.00042099297893681044,
"loss": 2.729,
"theoretical_loss": 3.8756564359912153,
"tokens_seen": 549388288
},
{
"epoch": 0.17,
"learning_rate": 0.00042098294884653967,
"loss": 2.7258,
"theoretical_loss": 3.8756074486484566,
"tokens_seen": 549453824
},
{
"epoch": 0.17,
"learning_rate": 0.0004209729187562688,
"loss": 2.7642,
"theoretical_loss": 3.8755584687841114,
"tokens_seen": 549519360
},
{
"epoch": 0.17,
"learning_rate": 0.00042096288866599803,
"loss": 2.6261,
"theoretical_loss": 3.8755094963961447,
"tokens_seen": 549584896
},
{
"epoch": 0.17,
"learning_rate": 0.00042095285857572716,
"loss": 2.6199,
"theoretical_loss": 3.875460531482526,
"tokens_seen": 549650432
},
{
"epoch": 0.17,
"learning_rate": 0.0004209428284854564,
"loss": 2.8711,
"theoretical_loss": 3.8754115740412214,
"tokens_seen": 549715968
},
{
"epoch": 0.17,
"learning_rate": 0.0004209327983951856,
"loss": 2.8933,
"theoretical_loss": 3.8753626240702013,
"tokens_seen": 549781504
},
{
"epoch": 0.17,
"learning_rate": 0.00042092276830491476,
"loss": 2.8939,
"theoretical_loss": 3.8753136815674356,
"tokens_seen": 549847040
},
{
"epoch": 0.17,
"learning_rate": 0.00042091273821464394,
"loss": 2.6116,
"theoretical_loss": 3.875264746530894,
"tokens_seen": 549912576
},
{
"epoch": 0.17,
"learning_rate": 0.0004209027081243731,
"loss": 3.0494,
"theoretical_loss": 3.875215818958549,
"tokens_seen": 549978112
},
{
"epoch": 0.17,
"learning_rate": 0.0004208926780341023,
"loss": 2.8988,
"theoretical_loss": 3.875166898848372,
"tokens_seen": 550043648
},
{
"epoch": 0.17,
"learning_rate": 0.00042088264794383154,
"loss": 2.7881,
"theoretical_loss": 3.8751179861983367,
"tokens_seen": 550109184
},
{
"epoch": 0.17,
"learning_rate": 0.00042087261785356066,
"loss": 2.8783,
"theoretical_loss": 3.8750690810064166,
"tokens_seen": 550174720
},
{
"epoch": 0.17,
"learning_rate": 0.0004208625877632899,
"loss": 2.7529,
"theoretical_loss": 3.8750201832705864,
"tokens_seen": 550240256
},
{
"epoch": 0.17,
"learning_rate": 0.0004208525576730191,
"loss": 2.9125,
"theoretical_loss": 3.874971292988821,
"tokens_seen": 550305792
},
{
"epoch": 0.17,
"learning_rate": 0.00042084252758274826,
"loss": 2.791,
"theoretical_loss": 3.8749224101590976,
"tokens_seen": 550371328
},
{
"epoch": 0.17,
"learning_rate": 0.00042083249749247744,
"loss": 3.0134,
"theoretical_loss": 3.8748735347793923,
"tokens_seen": 550436864
},
{
"epoch": 0.17,
"objective/train/docs_used": 322374,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5075249671936035,
"objective/train/theoretical_loss": 3.8748246668476827,
"objective/train/tokens_used": 570962400,
"theoretical_loss": 3.8748246668476827,
"tokens_seen": 550502400
},
{
"epoch": 0.17,
"learning_rate": 0.0004208224674022066,
"loss": 2.8429,
"theoretical_loss": 3.8748246668476827,
"tokens_seen": 550502400
},
{
"epoch": 0.17,
"learning_rate": 0.0004208124373119358,
"loss": 2.8469,
"theoretical_loss": 3.874775806361948,
"tokens_seen": 550567936
},
{
"epoch": 0.17,
"learning_rate": 0.00042080240722166504,
"loss": 2.9299,
"theoretical_loss": 3.8747269533201676,
"tokens_seen": 550633472
},
{
"epoch": 0.17,
"learning_rate": 0.00042079237713139417,
"loss": 2.8438,
"theoretical_loss": 3.874678107720321,
"tokens_seen": 550699008
},
{
"epoch": 0.17,
"learning_rate": 0.0004207823470411234,
"loss": 2.8016,
"theoretical_loss": 3.8746292695603897,
"tokens_seen": 550764544
},
{
"epoch": 0.17,
"learning_rate": 0.00042077231695085253,
"loss": 2.9375,
"theoretical_loss": 3.874580438838355,
"tokens_seen": 550830080
},
{
"epoch": 0.17,
"learning_rate": 0.00042076228686058176,
"loss": 2.9417,
"theoretical_loss": 3.8745316155521996,
"tokens_seen": 550895616
},
{
"epoch": 0.17,
"learning_rate": 0.00042075225677031095,
"loss": 3.0751,
"theoretical_loss": 3.8744827996999067,
"tokens_seen": 550961152
},
{
"epoch": 0.17,
"learning_rate": 0.0004207422266800401,
"loss": 2.6153,
"theoretical_loss": 3.874433991279461,
"tokens_seen": 551026688
},
{
"epoch": 0.17,
"learning_rate": 0.0004207321965897693,
"loss": 2.8295,
"theoretical_loss": 3.874385190288846,
"tokens_seen": 551092224
},
{
"epoch": 0.17,
"learning_rate": 0.0004207221664994985,
"loss": 2.5518,
"theoretical_loss": 3.874336396726049,
"tokens_seen": 551157760
},
{
"epoch": 0.17,
"learning_rate": 0.00042071213640922767,
"loss": 2.8555,
"theoretical_loss": 3.8742876105890547,
"tokens_seen": 551223296
},
{
"epoch": 0.17,
"learning_rate": 0.0004207021063189569,
"loss": 2.9313,
"theoretical_loss": 3.874238831875852,
"tokens_seen": 551288832
},
{
"epoch": 0.17,
"learning_rate": 0.00042069207622868603,
"loss": 2.8664,
"theoretical_loss": 3.8741900605844273,
"tokens_seen": 551354368
},
{
"epoch": 0.17,
"learning_rate": 0.00042068204613841527,
"loss": 2.6718,
"theoretical_loss": 3.8741412967127706,
"tokens_seen": 551419904
},
{
"epoch": 0.17,
"learning_rate": 0.00042067201604814445,
"loss": 2.9157,
"theoretical_loss": 3.8740925402588706,
"tokens_seen": 551485440
},
{
"epoch": 0.17,
"learning_rate": 0.00042066198595787363,
"loss": 2.8645,
"theoretical_loss": 3.8740437912207186,
"tokens_seen": 551550976
},
{
"epoch": 0.17,
"learning_rate": 0.0004206519558676028,
"loss": 2.8228,
"theoretical_loss": 3.873995049596305,
"tokens_seen": 551616512
},
{
"epoch": 0.17,
"learning_rate": 0.000420641925777332,
"loss": 3.0906,
"theoretical_loss": 3.873946315383622,
"tokens_seen": 551682048
},
{
"epoch": 0.17,
"learning_rate": 0.0004206318956870612,
"loss": 2.8373,
"theoretical_loss": 3.873897588580662,
"tokens_seen": 551747584
},
{
"epoch": 0.17,
"learning_rate": 0.0004206218655967904,
"loss": 2.767,
"theoretical_loss": 3.8738488691854194,
"tokens_seen": 551813120
},
{
"epoch": 0.17,
"learning_rate": 0.00042061183550651954,
"loss": 3.1444,
"theoretical_loss": 3.873800157195887,
"tokens_seen": 551878656
},
{
"epoch": 0.17,
"learning_rate": 0.00042060180541624877,
"loss": 2.8976,
"theoretical_loss": 3.873751452610061,
"tokens_seen": 551944192
},
{
"epoch": 0.17,
"learning_rate": 0.0004205917753259779,
"loss": 2.717,
"theoretical_loss": 3.8737027554259367,
"tokens_seen": 552009728
},
{
"epoch": 0.17,
"learning_rate": 0.00042058174523570713,
"loss": 2.8454,
"theoretical_loss": 3.873654065641511,
"tokens_seen": 552075264
},
{
"epoch": 0.17,
"objective/train/docs_used": 322945,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0310089588165283,
"objective/train/theoretical_loss": 3.8736053832547808,
"objective/train/tokens_used": 572600800,
"theoretical_loss": 3.8736053832547808,
"tokens_seen": 552140800
},
{
"epoch": 0.17,
"learning_rate": 0.0004205717151454363,
"loss": 2.8704,
"theoretical_loss": 3.8736053832547808,
"tokens_seen": 552140800
},
{
"epoch": 0.17,
"learning_rate": 0.0004205616850551655,
"loss": 2.6181,
"theoretical_loss": 3.873556708263745,
"tokens_seen": 552206336
},
{
"epoch": 0.17,
"learning_rate": 0.0004205516549648947,
"loss": 2.9212,
"theoretical_loss": 3.873508040666402,
"tokens_seen": 552271872
},
{
"epoch": 0.17,
"learning_rate": 0.0004205416248746239,
"loss": 2.9482,
"theoretical_loss": 3.8734593804607513,
"tokens_seen": 552337408
},
{
"epoch": 0.17,
"learning_rate": 0.00042053159478435304,
"loss": 2.789,
"theoretical_loss": 3.873410727644794,
"tokens_seen": 552402944
},
{
"epoch": 0.17,
"learning_rate": 0.0004205215646940823,
"loss": 3.0142,
"theoretical_loss": 3.8733620822165316,
"tokens_seen": 552468480
},
{
"epoch": 0.17,
"learning_rate": 0.0004205115346038114,
"loss": 2.9676,
"theoretical_loss": 3.8733134441739647,
"tokens_seen": 552534016
},
{
"epoch": 0.17,
"learning_rate": 0.00042050150451354064,
"loss": 2.8231,
"theoretical_loss": 3.8732648135150978,
"tokens_seen": 552599552
},
{
"epoch": 0.17,
"learning_rate": 0.0004204914744232698,
"loss": 2.862,
"theoretical_loss": 3.8732161902379336,
"tokens_seen": 552665088
},
{
"epoch": 0.17,
"learning_rate": 0.000420481444332999,
"loss": 2.9855,
"theoretical_loss": 3.873167574340477,
"tokens_seen": 552730624
},
{
"epoch": 0.17,
"learning_rate": 0.0004204714142427282,
"loss": 2.7015,
"theoretical_loss": 3.873118965820733,
"tokens_seen": 552796160
},
{
"epoch": 0.17,
"learning_rate": 0.00042046138415245736,
"loss": 2.8132,
"theoretical_loss": 3.873070364676707,
"tokens_seen": 552861696
},
{
"epoch": 0.17,
"learning_rate": 0.00042045135406218654,
"loss": 2.8356,
"theoretical_loss": 3.8730217709064063,
"tokens_seen": 552927232
},
{
"epoch": 0.17,
"learning_rate": 0.0004204413239719158,
"loss": 2.9976,
"theoretical_loss": 3.8729731845078383,
"tokens_seen": 552992768
},
{
"epoch": 0.17,
"learning_rate": 0.0004204312938816449,
"loss": 3.1535,
"theoretical_loss": 3.872924605479011,
"tokens_seen": 553058304
},
{
"epoch": 0.17,
"learning_rate": 0.00042042126379137414,
"loss": 2.526,
"theoretical_loss": 3.872876033817934,
"tokens_seen": 553123840
},
{
"epoch": 0.17,
"learning_rate": 0.0004204112337011033,
"loss": 2.8647,
"theoretical_loss": 3.8728274695226164,
"tokens_seen": 553189376
},
{
"epoch": 0.17,
"learning_rate": 0.0004204012036108325,
"loss": 3.1751,
"theoretical_loss": 3.872778912591069,
"tokens_seen": 553254912
},
{
"epoch": 0.17,
"learning_rate": 0.00042039117352056174,
"loss": 2.7206,
"theoretical_loss": 3.872730363021304,
"tokens_seen": 553320448
},
{
"epoch": 0.17,
"learning_rate": 0.00042038114343029086,
"loss": 2.7845,
"theoretical_loss": 3.8726818208113323,
"tokens_seen": 553385984
},
{
"epoch": 0.17,
"learning_rate": 0.0004203711133400201,
"loss": 3.2867,
"theoretical_loss": 3.872633285959167,
"tokens_seen": 553451520
},
{
"epoch": 0.17,
"learning_rate": 0.0004203610832497493,
"loss": 2.8596,
"theoretical_loss": 3.872584758462823,
"tokens_seen": 553517056
},
{
"epoch": 0.17,
"learning_rate": 0.00042035105315947846,
"loss": 2.7029,
"theoretical_loss": 3.872536238320313,
"tokens_seen": 553582592
},
{
"epoch": 0.17,
"learning_rate": 0.00042034102306920764,
"loss": 2.7705,
"theoretical_loss": 3.8724877255296533,
"tokens_seen": 553648128
},
{
"epoch": 0.17,
"learning_rate": 0.0004203309929789368,
"loss": 2.9573,
"theoretical_loss": 3.872439220088859,
"tokens_seen": 553713664
},
{
"epoch": 0.17,
"objective/train/docs_used": 324208,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7510035037994385,
"objective/train/theoretical_loss": 3.8723907219959486,
"objective/train/tokens_used": 574239200,
"theoretical_loss": 3.8723907219959486,
"tokens_seen": 553779200
},
{
"epoch": 0.17,
"learning_rate": 0.000420320962888666,
"loss": 2.7948,
"theoretical_loss": 3.8723907219959486,
"tokens_seen": 553779200
},
{
"epoch": 0.17,
"learning_rate": 0.00042031093279839524,
"loss": 3.0678,
"theoretical_loss": 3.872342231248938,
"tokens_seen": 553844736
},
{
"epoch": 0.17,
"learning_rate": 0.00042030090270812437,
"loss": 2.913,
"theoretical_loss": 3.872293747845846,
"tokens_seen": 553910272
},
{
"epoch": 0.17,
"learning_rate": 0.0004202908726178536,
"loss": 2.8919,
"theoretical_loss": 3.872245271784692,
"tokens_seen": 553975808
},
{
"epoch": 0.17,
"learning_rate": 0.00042028084252758273,
"loss": 2.9291,
"theoretical_loss": 3.872196803063495,
"tokens_seen": 554041344
},
{
"epoch": 0.17,
"learning_rate": 0.00042027081243731196,
"loss": 2.8092,
"theoretical_loss": 3.872148341680276,
"tokens_seen": 554106880
},
{
"epoch": 0.17,
"learning_rate": 0.00042026078234704115,
"loss": 2.8927,
"theoretical_loss": 3.8720998876330572,
"tokens_seen": 554172416
},
{
"epoch": 0.17,
"learning_rate": 0.0004202507522567703,
"loss": 2.9609,
"theoretical_loss": 3.8720514409198596,
"tokens_seen": 554237952
},
{
"epoch": 0.17,
"learning_rate": 0.0004202407221664995,
"loss": 2.6144,
"theoretical_loss": 3.8720030015387064,
"tokens_seen": 554303488
},
{
"epoch": 0.17,
"learning_rate": 0.0004202306920762287,
"loss": 3.0194,
"theoretical_loss": 3.871954569487622,
"tokens_seen": 554369024
},
{
"epoch": 0.17,
"learning_rate": 0.00042022066198595787,
"loss": 2.7512,
"theoretical_loss": 3.8719061447646297,
"tokens_seen": 554434560
},
{
"epoch": 0.17,
"learning_rate": 0.0004202106318956871,
"loss": 2.496,
"theoretical_loss": 3.8718577273677557,
"tokens_seen": 554500096
},
{
"epoch": 0.17,
"learning_rate": 0.00042020060180541623,
"loss": 2.7344,
"theoretical_loss": 3.8718093172950248,
"tokens_seen": 554565632
},
{
"epoch": 0.17,
"learning_rate": 0.00042019057171514547,
"loss": 2.8702,
"theoretical_loss": 3.8717609145444647,
"tokens_seen": 554631168
},
{
"epoch": 0.17,
"learning_rate": 0.00042018054162487465,
"loss": 2.7379,
"theoretical_loss": 3.8717125191141024,
"tokens_seen": 554696704
},
{
"epoch": 0.17,
"learning_rate": 0.00042017051153460383,
"loss": 2.8682,
"theoretical_loss": 3.871664131001967,
"tokens_seen": 554762240
},
{
"epoch": 0.17,
"learning_rate": 0.000420160481444333,
"loss": 2.6698,
"theoretical_loss": 3.8716157502060864,
"tokens_seen": 554827776
},
{
"epoch": 0.17,
"learning_rate": 0.0004201504513540622,
"loss": 3.04,
"theoretical_loss": 3.8715673767244914,
"tokens_seen": 554893312
},
{
"epoch": 0.17,
"learning_rate": 0.0004201404212637914,
"loss": 2.9472,
"theoretical_loss": 3.871519010555211,
"tokens_seen": 554958848
},
{
"epoch": 0.17,
"learning_rate": 0.0004201303911735206,
"loss": 2.8381,
"theoretical_loss": 3.8714706516962787,
"tokens_seen": 555024384
},
{
"epoch": 0.17,
"learning_rate": 0.00042012036108324974,
"loss": 2.975,
"theoretical_loss": 3.871422300145725,
"tokens_seen": 555089920
},
{
"epoch": 0.17,
"learning_rate": 0.00042011033099297897,
"loss": 2.7581,
"theoretical_loss": 3.8713739559015834,
"tokens_seen": 555155456
},
{
"epoch": 0.17,
"learning_rate": 0.0004201003009027081,
"loss": 2.7519,
"theoretical_loss": 3.871325618961887,
"tokens_seen": 555220992
},
{
"epoch": 0.17,
"learning_rate": 0.00042009027081243733,
"loss": 2.8407,
"theoretical_loss": 3.871277289324671,
"tokens_seen": 555286528
},
{
"epoch": 0.17,
"learning_rate": 0.0004200802407221665,
"loss": 3.044,
"theoretical_loss": 3.871228966987969,
"tokens_seen": 555352064
},
{
"epoch": 0.17,
"objective/train/docs_used": 324930,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.34468674659729,
"objective/train/theoretical_loss": 3.871180651949819,
"objective/train/tokens_used": 575877600,
"theoretical_loss": 3.871180651949819,
"tokens_seen": 555417600
},
{
"epoch": 0.17,
"learning_rate": 0.0004200702106318957,
"loss": 2.9225,
"theoretical_loss": 3.871180651949819,
"tokens_seen": 555417600
},
{
"epoch": 0.17,
"learning_rate": 0.0004200601805416249,
"loss": 2.9334,
"theoretical_loss": 3.871132344208256,
"tokens_seen": 555483136
},
{
"epoch": 0.17,
"learning_rate": 0.0004200501504513541,
"loss": 2.6462,
"theoretical_loss": 3.871084043761318,
"tokens_seen": 555548672
},
{
"epoch": 0.17,
"learning_rate": 0.00042004012036108324,
"loss": 3.0531,
"theoretical_loss": 3.8710357506070427,
"tokens_seen": 555614208
},
{
"epoch": 0.17,
"learning_rate": 0.0004200300902708125,
"loss": 3.0704,
"theoretical_loss": 3.87098746474347,
"tokens_seen": 555679744
},
{
"epoch": 0.17,
"learning_rate": 0.0004200200601805416,
"loss": 2.768,
"theoretical_loss": 3.870939186168638,
"tokens_seen": 555745280
},
{
"epoch": 0.17,
"learning_rate": 0.00042001003009027084,
"loss": 2.888,
"theoretical_loss": 3.8708909148805892,
"tokens_seen": 555810816
},
{
"epoch": 0.17,
"learning_rate": 0.00042,
"loss": 2.9139,
"theoretical_loss": 3.8708426508773632,
"tokens_seen": 555876352
},
{
"epoch": 0.17,
"learning_rate": 0.0004199899699097292,
"loss": 2.7703,
"theoretical_loss": 3.870794394157002,
"tokens_seen": 555941888
},
{
"epoch": 0.17,
"learning_rate": 0.0004199799398194584,
"loss": 2.8197,
"theoretical_loss": 3.8707461447175495,
"tokens_seen": 556007424
},
{
"epoch": 0.17,
"learning_rate": 0.00041996990972918756,
"loss": 2.6741,
"theoretical_loss": 3.8706979025570485,
"tokens_seen": 556072960
},
{
"epoch": 0.17,
"learning_rate": 0.00041995987963891674,
"loss": 2.7795,
"theoretical_loss": 3.870649667673543,
"tokens_seen": 556138496
},
{
"epoch": 0.17,
"learning_rate": 0.000419949849548646,
"loss": 2.8733,
"theoretical_loss": 3.870601440065078,
"tokens_seen": 556204032
},
{
"epoch": 0.17,
"learning_rate": 0.0004199398194583751,
"loss": 2.619,
"theoretical_loss": 3.8705532197296995,
"tokens_seen": 556269568
},
{
"epoch": 0.17,
"learning_rate": 0.00041992978936810434,
"loss": 2.6645,
"theoretical_loss": 3.8705050066654536,
"tokens_seen": 556335104
},
{
"epoch": 0.17,
"learning_rate": 0.00041991975927783347,
"loss": 2.9262,
"theoretical_loss": 3.870456800870388,
"tokens_seen": 556400640
},
{
"epoch": 0.17,
"learning_rate": 0.0004199097291875627,
"loss": 2.6335,
"theoretical_loss": 3.8704086023425512,
"tokens_seen": 556466176
},
{
"epoch": 0.17,
"learning_rate": 0.0004198996990972919,
"loss": 2.961,
"theoretical_loss": 3.8703604110799903,
"tokens_seen": 556531712
},
{
"epoch": 0.17,
"learning_rate": 0.00041988966900702106,
"loss": 2.8548,
"theoretical_loss": 3.870312227080756,
"tokens_seen": 556597248
},
{
"epoch": 0.17,
"learning_rate": 0.00041987963891675025,
"loss": 2.7521,
"theoretical_loss": 3.8702640503428984,
"tokens_seen": 556662784
},
{
"epoch": 0.17,
"learning_rate": 0.0004198696088264795,
"loss": 2.8326,
"theoretical_loss": 3.870215880864469,
"tokens_seen": 556728320
},
{
"epoch": 0.17,
"learning_rate": 0.0004198595787362086,
"loss": 2.9862,
"theoretical_loss": 3.8701677186435184,
"tokens_seen": 556793856
},
{
"epoch": 0.17,
"learning_rate": 0.00041984954864593784,
"loss": 2.7987,
"theoretical_loss": 3.8701195636781,
"tokens_seen": 556859392
},
{
"epoch": 0.17,
"learning_rate": 0.00041983951855566697,
"loss": 2.9304,
"theoretical_loss": 3.8700714159662666,
"tokens_seen": 556924928
},
{
"epoch": 0.17,
"learning_rate": 0.0004198294884653962,
"loss": 2.8822,
"theoretical_loss": 3.870023275506073,
"tokens_seen": 556990464
},
{
"epoch": 0.17,
"objective/train/docs_used": 326252,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7411465644836426,
"objective/train/theoretical_loss": 3.869975142295573,
"objective/train/tokens_used": 577516000,
"theoretical_loss": 3.869975142295573,
"tokens_seen": 557056000
},
{
"epoch": 0.17,
"learning_rate": 0.0004198194583751254,
"loss": 2.9452,
"theoretical_loss": 3.869975142295573,
"tokens_seen": 557056000
},
{
"epoch": 0.17,
"learning_rate": 0.00041980942828485457,
"loss": 2.7595,
"theoretical_loss": 3.8699270163328223,
"tokens_seen": 557121536
},
{
"epoch": 0.17,
"learning_rate": 0.00041979939819458375,
"loss": 2.8155,
"theoretical_loss": 3.8698788976158776,
"tokens_seen": 557187072
},
{
"epoch": 0.17,
"learning_rate": 0.00041978936810431293,
"loss": 2.7187,
"theoretical_loss": 3.8698307861427956,
"tokens_seen": 557252608
},
{
"epoch": 0.17,
"learning_rate": 0.0004197793380140421,
"loss": 2.9983,
"theoretical_loss": 3.8697826819116345,
"tokens_seen": 557318144
},
{
"epoch": 0.17,
"learning_rate": 0.00041976930792377135,
"loss": 2.743,
"theoretical_loss": 3.8697345849204523,
"tokens_seen": 557383680
},
{
"epoch": 0.17,
"learning_rate": 0.00041975927783350047,
"loss": 2.8873,
"theoretical_loss": 3.8696864951673087,
"tokens_seen": 557449216
},
{
"epoch": 0.17,
"learning_rate": 0.0004197492477432297,
"loss": 2.7194,
"theoretical_loss": 3.8696384126502634,
"tokens_seen": 557514752
},
{
"epoch": 0.17,
"learning_rate": 0.00041973921765295883,
"loss": 2.9329,
"theoretical_loss": 3.8695903373673772,
"tokens_seen": 557580288
},
{
"epoch": 0.17,
"learning_rate": 0.00041972918756268807,
"loss": 2.825,
"theoretical_loss": 3.869542269316711,
"tokens_seen": 557645824
},
{
"epoch": 0.17,
"learning_rate": 0.00041971915747241725,
"loss": 2.9068,
"theoretical_loss": 3.869494208496329,
"tokens_seen": 557711360
},
{
"epoch": 0.17,
"learning_rate": 0.00041970912738214643,
"loss": 2.8372,
"theoretical_loss": 3.8694461549042924,
"tokens_seen": 557776896
},
{
"epoch": 0.17,
"learning_rate": 0.0004196990972918756,
"loss": 3.0069,
"theoretical_loss": 3.8693981085386655,
"tokens_seen": 557842432
},
{
"epoch": 0.17,
"learning_rate": 0.00041968906720160485,
"loss": 2.7972,
"theoretical_loss": 3.8693500693975125,
"tokens_seen": 557907968
},
{
"epoch": 0.17,
"learning_rate": 0.000419679037111334,
"loss": 2.6788,
"theoretical_loss": 3.8693020374788993,
"tokens_seen": 557973504
},
{
"epoch": 0.17,
"learning_rate": 0.0004196690070210632,
"loss": 2.8561,
"theoretical_loss": 3.8692540127808916,
"tokens_seen": 558039040
},
{
"epoch": 0.17,
"learning_rate": 0.0004196589769307924,
"loss": 2.5665,
"theoretical_loss": 3.8692059953015563,
"tokens_seen": 558104576
},
{
"epoch": 0.17,
"learning_rate": 0.0004196489468405216,
"loss": 2.8662,
"theoretical_loss": 3.86915798503896,
"tokens_seen": 558170112
},
{
"epoch": 0.17,
"learning_rate": 0.0004196389167502508,
"loss": 3.04,
"theoretical_loss": 3.8691099819911723,
"tokens_seen": 558235648
},
{
"epoch": 0.17,
"learning_rate": 0.00041962888665997994,
"loss": 2.6662,
"theoretical_loss": 3.869061986156261,
"tokens_seen": 558301184
},
{
"epoch": 0.17,
"learning_rate": 0.00041961885656970917,
"loss": 2.9142,
"theoretical_loss": 3.8690139975322966,
"tokens_seen": 558366720
},
{
"epoch": 0.17,
"learning_rate": 0.0004196088264794383,
"loss": 2.968,
"theoretical_loss": 3.8689660161173487,
"tokens_seen": 558432256
},
{
"epoch": 0.17,
"learning_rate": 0.00041959879638916753,
"loss": 2.9044,
"theoretical_loss": 3.8689180419094895,
"tokens_seen": 558497792
},
{
"epoch": 0.17,
"learning_rate": 0.0004195887662988967,
"loss": 2.7042,
"theoretical_loss": 3.8688700749067904,
"tokens_seen": 558563328
},
{
"epoch": 0.17,
"learning_rate": 0.0004195787362086259,
"loss": 2.8081,
"theoretical_loss": 3.868822115107324,
"tokens_seen": 558628864
},
{
"epoch": 0.17,
"objective/train/docs_used": 326886,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5210978984832764,
"objective/train/theoretical_loss": 3.8687741625091636,
"objective/train/tokens_used": 579154400,
"theoretical_loss": 3.8687741625091636,
"tokens_seen": 558694400
},
{
"epoch": 0.17,
"learning_rate": 0.0004195687061183551,
"loss": 2.6311,
"theoretical_loss": 3.8687741625091636,
"tokens_seen": 558694400
},
{
"epoch": 0.17,
"learning_rate": 0.0004195586760280843,
"loss": 2.9603,
"theoretical_loss": 3.8687262171103836,
"tokens_seen": 558759936
},
{
"epoch": 0.17,
"learning_rate": 0.00041954864593781344,
"loss": 2.8054,
"theoretical_loss": 3.8686782789090595,
"tokens_seen": 558825472
},
{
"epoch": 0.17,
"learning_rate": 0.0004195386158475427,
"loss": 2.7241,
"theoretical_loss": 3.868630347903266,
"tokens_seen": 558891008
},
{
"epoch": 0.17,
"learning_rate": 0.0004195285857572718,
"loss": 2.7745,
"theoretical_loss": 3.8685824240910804,
"tokens_seen": 558956544
},
{
"epoch": 0.17,
"learning_rate": 0.00041951855566700104,
"loss": 2.7074,
"theoretical_loss": 3.868534507470579,
"tokens_seen": 559022080
},
{
"epoch": 0.17,
"learning_rate": 0.0004195085255767302,
"loss": 2.8136,
"theoretical_loss": 3.8684865980398393,
"tokens_seen": 559087616
},
{
"epoch": 0.17,
"learning_rate": 0.0004194984954864594,
"loss": 2.8338,
"theoretical_loss": 3.868438695796941,
"tokens_seen": 559153152
},
{
"epoch": 0.17,
"learning_rate": 0.0004194884653961886,
"loss": 2.8892,
"theoretical_loss": 3.8683908007399634,
"tokens_seen": 559218688
},
{
"epoch": 0.17,
"learning_rate": 0.00041947843530591776,
"loss": 3.0437,
"theoretical_loss": 3.8683429128669857,
"tokens_seen": 559284224
},
{
"epoch": 0.17,
"learning_rate": 0.00041946840521564694,
"loss": 2.8545,
"theoretical_loss": 3.8682950321760896,
"tokens_seen": 559349760
},
{
"epoch": 0.17,
"learning_rate": 0.0004194583751253762,
"loss": 2.7766,
"theoretical_loss": 3.868247158665356,
"tokens_seen": 559415296
},
{
"epoch": 0.17,
"learning_rate": 0.0004194483450351053,
"loss": 2.9612,
"theoretical_loss": 3.8681992923328674,
"tokens_seen": 559480832
},
{
"epoch": 0.17,
"learning_rate": 0.00041943831494483454,
"loss": 2.658,
"theoretical_loss": 3.868151433176707,
"tokens_seen": 559546368
},
{
"epoch": 0.17,
"learning_rate": 0.00041942828485456367,
"loss": 2.6934,
"theoretical_loss": 3.8681035811949585,
"tokens_seen": 559611904
},
{
"epoch": 0.17,
"learning_rate": 0.0004194182547642929,
"loss": 2.6309,
"theoretical_loss": 3.868055736385706,
"tokens_seen": 559677440
},
{
"epoch": 0.17,
"learning_rate": 0.0004194082246740221,
"loss": 2.9048,
"theoretical_loss": 3.868007898747036,
"tokens_seen": 559742976
},
{
"epoch": 0.17,
"learning_rate": 0.00041939819458375126,
"loss": 2.7095,
"theoretical_loss": 3.867960068277033,
"tokens_seen": 559808512
},
{
"epoch": 0.17,
"learning_rate": 0.00041938816449348045,
"loss": 2.8633,
"theoretical_loss": 3.867912244973784,
"tokens_seen": 559874048
},
{
"epoch": 0.17,
"learning_rate": 0.0004193781344032097,
"loss": 3.0507,
"theoretical_loss": 3.867864428835377,
"tokens_seen": 559939584
},
{
"epoch": 0.17,
"learning_rate": 0.0004193681043129388,
"loss": 2.9638,
"theoretical_loss": 3.8678166198598998,
"tokens_seen": 560005120
},
{
"epoch": 0.17,
"learning_rate": 0.00041935807422266804,
"loss": 2.8753,
"theoretical_loss": 3.8677688180454415,
"tokens_seen": 560070656
},
{
"epoch": 0.17,
"learning_rate": 0.00041934804413239717,
"loss": 2.7619,
"theoretical_loss": 3.867721023390091,
"tokens_seen": 560136192
},
{
"epoch": 0.17,
"learning_rate": 0.0004193380140421264,
"loss": 2.909,
"theoretical_loss": 3.8676732358919406,
"tokens_seen": 560201728
},
{
"epoch": 0.17,
"learning_rate": 0.0004193279839518556,
"loss": 2.7702,
"theoretical_loss": 3.8676254555490788,
"tokens_seen": 560267264
},
{
"epoch": 0.17,
"objective/train/docs_used": 327930,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8123157024383545,
"objective/train/theoretical_loss": 3.8675776823595998,
"objective/train/tokens_used": 580792800,
"theoretical_loss": 3.8675776823595998,
"tokens_seen": 560332800
},
{
"epoch": 0.17,
"learning_rate": 0.00041931795386158477,
"loss": 2.6362,
"theoretical_loss": 3.8675776823595998,
"tokens_seen": 560332800
},
{
"epoch": 0.17,
"learning_rate": 0.00041930792377131395,
"loss": 2.4958,
"theoretical_loss": 3.867529916321595,
"tokens_seen": 560398336
},
{
"epoch": 0.17,
"learning_rate": 0.00041929789368104313,
"loss": 2.8404,
"theoretical_loss": 3.867482157433157,
"tokens_seen": 560463872
},
{
"epoch": 0.17,
"learning_rate": 0.0004192878635907723,
"loss": 2.8702,
"theoretical_loss": 3.8674344056923813,
"tokens_seen": 560529408
},
{
"epoch": 0.17,
"learning_rate": 0.00041927783350050155,
"loss": 2.9506,
"theoretical_loss": 3.8673866610973624,
"tokens_seen": 560594944
},
{
"epoch": 0.17,
"learning_rate": 0.00041926780341023067,
"loss": 2.8685,
"theoretical_loss": 3.8673389236461952,
"tokens_seen": 560660480
},
{
"epoch": 0.17,
"learning_rate": 0.0004192577733199599,
"loss": 2.8917,
"theoretical_loss": 3.867291193336976,
"tokens_seen": 560726016
},
{
"epoch": 0.17,
"learning_rate": 0.00041924774322968904,
"loss": 2.8617,
"theoretical_loss": 3.8672434701678022,
"tokens_seen": 560791552
},
{
"epoch": 0.17,
"learning_rate": 0.00041923771313941827,
"loss": 2.8296,
"theoretical_loss": 3.867195754136771,
"tokens_seen": 560857088
},
{
"epoch": 0.17,
"learning_rate": 0.00041922768304914745,
"loss": 2.8242,
"theoretical_loss": 3.867148045241981,
"tokens_seen": 560922624
},
{
"epoch": 0.17,
"learning_rate": 0.00041921765295887663,
"loss": 2.8429,
"theoretical_loss": 3.867100343481532,
"tokens_seen": 560988160
},
{
"epoch": 0.17,
"learning_rate": 0.0004192076228686058,
"loss": 2.7912,
"theoretical_loss": 3.8670526488535226,
"tokens_seen": 561053696
},
{
"epoch": 0.17,
"learning_rate": 0.00041919759277833505,
"loss": 2.696,
"theoretical_loss": 3.867004961356054,
"tokens_seen": 561119232
},
{
"epoch": 0.17,
"learning_rate": 0.0004191875626880642,
"loss": 3.068,
"theoretical_loss": 3.8669572809872284,
"tokens_seen": 561184768
},
{
"epoch": 0.17,
"learning_rate": 0.0004191775325977934,
"loss": 2.797,
"theoretical_loss": 3.8669096077451464,
"tokens_seen": 561250304
},
{
"epoch": 0.17,
"learning_rate": 0.00041916750250752254,
"loss": 2.7436,
"theoretical_loss": 3.8668619416279117,
"tokens_seen": 561315840
},
{
"epoch": 0.17,
"learning_rate": 0.0004191574724172518,
"loss": 2.8369,
"theoretical_loss": 3.8668142826336274,
"tokens_seen": 561381376
},
{
"epoch": 0.17,
"learning_rate": 0.00041914744232698095,
"loss": 2.7877,
"theoretical_loss": 3.866766630760398,
"tokens_seen": 561446912
},
{
"epoch": 0.17,
"learning_rate": 0.00041913741223671014,
"loss": 2.872,
"theoretical_loss": 3.8667189860063282,
"tokens_seen": 561512448
},
{
"epoch": 0.17,
"learning_rate": 0.0004191273821464393,
"loss": 2.7135,
"theoretical_loss": 3.866671348369524,
"tokens_seen": 561577984
},
{
"epoch": 0.17,
"learning_rate": 0.0004191173520561685,
"loss": 2.8627,
"theoretical_loss": 3.866623717848092,
"tokens_seen": 561643520
},
{
"epoch": 0.17,
"learning_rate": 0.0004191073219658977,
"loss": 2.6897,
"theoretical_loss": 3.8665760944401377,
"tokens_seen": 561709056
},
{
"epoch": 0.17,
"learning_rate": 0.0004190972918756269,
"loss": 3.0401,
"theoretical_loss": 3.8665284781437714,
"tokens_seen": 561774592
},
{
"epoch": 0.17,
"learning_rate": 0.00041908726178535604,
"loss": 2.9005,
"theoretical_loss": 3.8664808689571,
"tokens_seen": 561840128
},
{
"epoch": 0.17,
"learning_rate": 0.0004190772316950853,
"loss": 2.7574,
"theoretical_loss": 3.866433266878234,
"tokens_seen": 561905664
},
{
"epoch": 0.17,
"objective/train/docs_used": 328746,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2547881603240967,
"objective/train/theoretical_loss": 3.8663856719052827,
"objective/train/tokens_used": 582431200,
"theoretical_loss": 3.8663856719052827,
"tokens_seen": 561971200
},
{
"epoch": 0.17,
"learning_rate": 0.0004190672016048144,
"loss": 2.9363,
"theoretical_loss": 3.8663856719052827,
"tokens_seen": 561971200
},
{
"epoch": 0.17,
"learning_rate": 0.00041905717151454364,
"loss": 3.0646,
"theoretical_loss": 3.8663380840363564,
"tokens_seen": 562036736
},
{
"epoch": 0.17,
"learning_rate": 0.0004190471414242728,
"loss": 2.9747,
"theoretical_loss": 3.866290503269567,
"tokens_seen": 562102272
},
{
"epoch": 0.17,
"learning_rate": 0.000419037111334002,
"loss": 2.915,
"theoretical_loss": 3.866242929603027,
"tokens_seen": 562167808
},
{
"epoch": 0.17,
"learning_rate": 0.0004190270812437312,
"loss": 2.8532,
"theoretical_loss": 3.8661953630348496,
"tokens_seen": 562233344
},
{
"epoch": 0.17,
"learning_rate": 0.0004190170511534604,
"loss": 2.8447,
"theoretical_loss": 3.866147803563148,
"tokens_seen": 562298880
},
{
"epoch": 0.17,
"learning_rate": 0.00041900702106318954,
"loss": 3.027,
"theoretical_loss": 3.866100251186036,
"tokens_seen": 562364416
},
{
"epoch": 0.17,
"learning_rate": 0.0004189969909729188,
"loss": 3.0567,
"theoretical_loss": 3.8660527059016294,
"tokens_seen": 562429952
},
{
"epoch": 0.17,
"learning_rate": 0.0004189869608826479,
"loss": 2.8168,
"theoretical_loss": 3.8660051677080434,
"tokens_seen": 562495488
},
{
"epoch": 0.17,
"learning_rate": 0.00041897693079237714,
"loss": 2.8508,
"theoretical_loss": 3.8659576366033956,
"tokens_seen": 562561024
},
{
"epoch": 0.17,
"learning_rate": 0.0004189669007021063,
"loss": 2.6138,
"theoretical_loss": 3.865910112585802,
"tokens_seen": 562626560
},
{
"epoch": 0.17,
"learning_rate": 0.0004189568706118355,
"loss": 2.9875,
"theoretical_loss": 3.8658625956533816,
"tokens_seen": 562692096
},
{
"epoch": 0.17,
"learning_rate": 0.0004189468405215647,
"loss": 2.9688,
"theoretical_loss": 3.8658150858042526,
"tokens_seen": 562757632
},
{
"epoch": 0.17,
"learning_rate": 0.00041893681043129387,
"loss": 3.0128,
"theoretical_loss": 3.8657675830365346,
"tokens_seen": 562823168
},
{
"epoch": 0.17,
"learning_rate": 0.00041892678034102305,
"loss": 2.6135,
"theoretical_loss": 3.865720087348347,
"tokens_seen": 562888704
},
{
"epoch": 0.17,
"learning_rate": 0.0004189167502507523,
"loss": 2.8912,
"theoretical_loss": 3.8656725987378118,
"tokens_seen": 562954240
},
{
"epoch": 0.17,
"learning_rate": 0.00041890672016048146,
"loss": 2.9469,
"theoretical_loss": 3.8656251172030496,
"tokens_seen": 563019776
},
{
"epoch": 0.17,
"learning_rate": 0.00041889669007021065,
"loss": 2.7757,
"theoretical_loss": 3.865577642742183,
"tokens_seen": 563085312
},
{
"epoch": 0.17,
"learning_rate": 0.0004188866599799399,
"loss": 2.9993,
"theoretical_loss": 3.865530175353335,
"tokens_seen": 563150848
},
{
"epoch": 0.17,
"learning_rate": 0.000418876629889669,
"loss": 2.7734,
"theoretical_loss": 3.86548271503463,
"tokens_seen": 563216384
},
{
"epoch": 0.17,
"learning_rate": 0.00041886659979939824,
"loss": 2.6621,
"theoretical_loss": 3.865435261784191,
"tokens_seen": 563281920
},
{
"epoch": 0.17,
"learning_rate": 0.00041885656970912737,
"loss": 3.0169,
"theoretical_loss": 3.865387815600144,
"tokens_seen": 563347456
},
{
"epoch": 0.17,
"learning_rate": 0.0004188465396188566,
"loss": 3.0002,
"theoretical_loss": 3.8653403764806145,
"tokens_seen": 563412992
},
{
"epoch": 0.17,
"learning_rate": 0.0004188365095285858,
"loss": 3.0746,
"theoretical_loss": 3.86529294442373,
"tokens_seen": 563478528
},
{
"epoch": 0.17,
"learning_rate": 0.00041882647943831497,
"loss": 2.8611,
"theoretical_loss": 3.8652455194276163,
"tokens_seen": 563544064
},
{
"epoch": 0.17,
"objective/train/docs_used": 329425,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.942103862762451,
"objective/train/theoretical_loss": 3.8651981014904027,
"objective/train/tokens_used": 584069600,
"theoretical_loss": 3.8651981014904027,
"tokens_seen": 563609600
},
{
"epoch": 0.17,
"learning_rate": 0.00041881644934804415,
"loss": 2.8762,
"theoretical_loss": 3.8651981014904027,
"tokens_seen": 563609600
},
{
"epoch": 0.17,
"learning_rate": 0.00041880641925777333,
"loss": 2.8032,
"theoretical_loss": 3.8651506906102173,
"tokens_seen": 563675136
},
{
"epoch": 0.17,
"learning_rate": 0.0004187963891675025,
"loss": 2.8483,
"theoretical_loss": 3.8651032867851893,
"tokens_seen": 563740672
},
{
"epoch": 0.17,
"learning_rate": 0.00041878635907723175,
"loss": 3.0158,
"theoretical_loss": 3.8650558900134495,
"tokens_seen": 563806208
},
{
"epoch": 0.17,
"learning_rate": 0.0004187763289869609,
"loss": 3.0126,
"theoretical_loss": 3.8650085002931283,
"tokens_seen": 563871744
},
{
"epoch": 0.17,
"learning_rate": 0.0004187662988966901,
"loss": 3.0105,
"theoretical_loss": 3.8649611176223573,
"tokens_seen": 563937280
},
{
"epoch": 0.17,
"learning_rate": 0.00041875626880641924,
"loss": 2.905,
"theoretical_loss": 3.8649137419992687,
"tokens_seen": 564002816
},
{
"epoch": 0.17,
"learning_rate": 0.00041874623871614847,
"loss": 2.7749,
"theoretical_loss": 3.864866373421996,
"tokens_seen": 564068352
},
{
"epoch": 0.17,
"learning_rate": 0.00041873620862587765,
"loss": 2.9877,
"theoretical_loss": 3.8648190118886725,
"tokens_seen": 564133888
},
{
"epoch": 0.17,
"learning_rate": 0.00041872617853560683,
"loss": 2.8715,
"theoretical_loss": 3.8647716573974327,
"tokens_seen": 564199424
},
{
"epoch": 0.17,
"learning_rate": 0.000418716148445336,
"loss": 2.7122,
"theoretical_loss": 3.8647243099464106,
"tokens_seen": 564264960
},
{
"epoch": 0.17,
"learning_rate": 0.00041870611835506525,
"loss": 2.9309,
"theoretical_loss": 3.864676969533744,
"tokens_seen": 564330496
},
{
"epoch": 0.17,
"learning_rate": 0.0004186960882647944,
"loss": 2.7819,
"theoretical_loss": 3.8646296361575687,
"tokens_seen": 564396032
},
{
"epoch": 0.17,
"learning_rate": 0.0004186860581745236,
"loss": 2.7789,
"theoretical_loss": 3.8645823098160212,
"tokens_seen": 564461568
},
{
"epoch": 0.17,
"learning_rate": 0.00041867602808425274,
"loss": 3.1001,
"theoretical_loss": 3.8645349905072406,
"tokens_seen": 564527104
},
{
"epoch": 0.17,
"learning_rate": 0.000418665997993982,
"loss": 2.7691,
"theoretical_loss": 3.8644876782293647,
"tokens_seen": 564592640
},
{
"epoch": 0.17,
"learning_rate": 0.00041865596790371115,
"loss": 3.0858,
"theoretical_loss": 3.864440372980533,
"tokens_seen": 564658176
},
{
"epoch": 0.17,
"learning_rate": 0.00041864593781344034,
"loss": 2.9535,
"theoretical_loss": 3.8643930747588855,
"tokens_seen": 564723712
},
{
"epoch": 0.17,
"learning_rate": 0.0004186359077231695,
"loss": 2.5333,
"theoretical_loss": 3.8643457835625634,
"tokens_seen": 564789248
},
{
"epoch": 0.17,
"learning_rate": 0.0004186258776328987,
"loss": 2.9543,
"theoretical_loss": 3.8642984993897085,
"tokens_seen": 564854784
},
{
"epoch": 0.17,
"learning_rate": 0.0004186158475426279,
"loss": 2.5846,
"theoretical_loss": 3.864251222238462,
"tokens_seen": 564920320
},
{
"epoch": 0.17,
"learning_rate": 0.0004186058174523571,
"loss": 3.0025,
"theoretical_loss": 3.8642039521069678,
"tokens_seen": 564985856
},
{
"epoch": 0.17,
"learning_rate": 0.00041859578736208624,
"loss": 2.9297,
"theoretical_loss": 3.864156688993369,
"tokens_seen": 565051392
},
{
"epoch": 0.17,
"learning_rate": 0.0004185857572718155,
"loss": 2.944,
"theoretical_loss": 3.86410943289581,
"tokens_seen": 565116928
},
{
"epoch": 0.17,
"learning_rate": 0.0004185757271815446,
"loss": 2.6911,
"theoretical_loss": 3.864062183812436,
"tokens_seen": 565182464
},
{
"epoch": 0.17,
"objective/train/docs_used": 329612,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.730563163757324,
"objective/train/theoretical_loss": 3.864014941741392,
"objective/train/tokens_used": 585708000,
"theoretical_loss": 3.864014941741392,
"tokens_seen": 565248000
},
{
"epoch": 0.17,
"learning_rate": 0.00041856569709127384,
"loss": 3.1174,
"theoretical_loss": 3.864014941741392,
"tokens_seen": 565248000
},
{
"epoch": 0.17,
"learning_rate": 0.000418555667001003,
"loss": 3.1198,
"theoretical_loss": 3.8639677066808256,
"tokens_seen": 565313536
},
{
"epoch": 0.17,
"learning_rate": 0.0004185456369107322,
"loss": 2.9859,
"theoretical_loss": 3.863920478628883,
"tokens_seen": 565379072
},
{
"epoch": 0.17,
"learning_rate": 0.0004185356068204614,
"loss": 2.9614,
"theoretical_loss": 3.8638732575837125,
"tokens_seen": 565444608
},
{
"epoch": 0.17,
"learning_rate": 0.0004185255767301906,
"loss": 2.7855,
"theoretical_loss": 3.8638260435434626,
"tokens_seen": 565510144
},
{
"epoch": 0.17,
"learning_rate": 0.00041851554663991974,
"loss": 2.9431,
"theoretical_loss": 3.863778836506283,
"tokens_seen": 565575680
},
{
"epoch": 0.17,
"learning_rate": 0.000418505516549649,
"loss": 2.7765,
"theoretical_loss": 3.8637316364703227,
"tokens_seen": 565641216
},
{
"epoch": 0.17,
"learning_rate": 0.0004184954864593781,
"loss": 2.9082,
"theoretical_loss": 3.8636844434337325,
"tokens_seen": 565706752
},
{
"epoch": 0.17,
"learning_rate": 0.00041848545636910734,
"loss": 3.0147,
"theoretical_loss": 3.863637257394665,
"tokens_seen": 565772288
},
{
"epoch": 0.17,
"learning_rate": 0.0004184754262788365,
"loss": 3.0763,
"theoretical_loss": 3.863590078351271,
"tokens_seen": 565837824
},
{
"epoch": 0.17,
"learning_rate": 0.0004184653961885657,
"loss": 2.9114,
"theoretical_loss": 3.8635429063017037,
"tokens_seen": 565903360
},
{
"epoch": 0.17,
"learning_rate": 0.0004184553660982949,
"loss": 2.7337,
"theoretical_loss": 3.8634957412441167,
"tokens_seen": 565968896
},
{
"epoch": 0.17,
"learning_rate": 0.00041844533600802407,
"loss": 2.9751,
"theoretical_loss": 3.8634485831766634,
"tokens_seen": 566034432
},
{
"epoch": 0.17,
"learning_rate": 0.00041843530591775325,
"loss": 2.9623,
"theoretical_loss": 3.8634014320975005,
"tokens_seen": 566099968
},
{
"epoch": 0.17,
"learning_rate": 0.0004184252758274825,
"loss": 3.0314,
"theoretical_loss": 3.8633542880047815,
"tokens_seen": 566165504
},
{
"epoch": 0.17,
"learning_rate": 0.0004184152457372116,
"loss": 2.9415,
"theoretical_loss": 3.863307150896664,
"tokens_seen": 566231040
},
{
"epoch": 0.17,
"learning_rate": 0.00041840521564694085,
"loss": 2.6393,
"theoretical_loss": 3.863260020771304,
"tokens_seen": 566296576
},
{
"epoch": 0.17,
"learning_rate": 0.00041839518555666997,
"loss": 3.0198,
"theoretical_loss": 3.8632128976268603,
"tokens_seen": 566362112
},
{
"epoch": 0.17,
"learning_rate": 0.0004183851554663992,
"loss": 2.9247,
"theoretical_loss": 3.8631657814614906,
"tokens_seen": 566427648
},
{
"epoch": 0.17,
"learning_rate": 0.0004183751253761284,
"loss": 3.0689,
"theoretical_loss": 3.8631186722733535,
"tokens_seen": 566493184
},
{
"epoch": 0.17,
"learning_rate": 0.00041836509528585757,
"loss": 2.8912,
"theoretical_loss": 3.86307157006061,
"tokens_seen": 566558720
},
{
"epoch": 0.17,
"learning_rate": 0.00041835506519558675,
"loss": 2.8435,
"theoretical_loss": 3.8630244748214198,
"tokens_seen": 566624256
},
{
"epoch": 0.17,
"learning_rate": 0.000418345035105316,
"loss": 2.9153,
"theoretical_loss": 3.862977386553944,
"tokens_seen": 566689792
},
{
"epoch": 0.17,
"learning_rate": 0.0004183350050150451,
"loss": 2.7937,
"theoretical_loss": 3.862930305256345,
"tokens_seen": 566755328
},
{
"epoch": 0.17,
"learning_rate": 0.00041832497492477435,
"loss": 3.0194,
"theoretical_loss": 3.8628832309267844,
"tokens_seen": 566820864
},
{
"epoch": 0.17,
"objective/train/docs_used": 331071,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8123180866241455,
"objective/train/theoretical_loss": 3.8628361635634265,
"objective/train/tokens_used": 587346400,
"theoretical_loss": 3.8628361635634265,
"tokens_seen": 566886400
},
{
"epoch": 0.17,
"learning_rate": 0.0004183149448345035,
"loss": 2.6294,
"theoretical_loss": 3.8628361635634265,
"tokens_seen": 566886400
},
{
"epoch": 0.17,
"learning_rate": 0.0004183049147442327,
"loss": 3.1253,
"theoretical_loss": 3.862789103164435,
"tokens_seen": 566951936
},
{
"epoch": 0.17,
"learning_rate": 0.0004182948846539619,
"loss": 2.8894,
"theoretical_loss": 3.8627420497279745,
"tokens_seen": 567017472
},
{
"epoch": 0.17,
"learning_rate": 0.0004182848545636911,
"loss": 2.9479,
"theoretical_loss": 3.8626950032522096,
"tokens_seen": 567083008
},
{
"epoch": 0.17,
"learning_rate": 0.00041827482447342025,
"loss": 2.9446,
"theoretical_loss": 3.8626479637353075,
"tokens_seen": 567148544
},
{
"epoch": 0.17,
"learning_rate": 0.00041826479438314944,
"loss": 2.9645,
"theoretical_loss": 3.862600931175434,
"tokens_seen": 567214080
},
{
"epoch": 0.17,
"learning_rate": 0.0004182547642928786,
"loss": 2.9893,
"theoretical_loss": 3.8625539055707576,
"tokens_seen": 567279616
},
{
"epoch": 0.17,
"learning_rate": 0.00041824473420260785,
"loss": 2.9431,
"theoretical_loss": 3.862506886919445,
"tokens_seen": 567345152
},
{
"epoch": 0.17,
"learning_rate": 0.000418234704112337,
"loss": 2.8937,
"theoretical_loss": 3.862459875219667,
"tokens_seen": 567410688
},
{
"epoch": 0.17,
"learning_rate": 0.0004182246740220662,
"loss": 3.0324,
"theoretical_loss": 3.862412870469591,
"tokens_seen": 567476224
},
{
"epoch": 0.17,
"learning_rate": 0.0004182146439317954,
"loss": 2.989,
"theoretical_loss": 3.8623658726673886,
"tokens_seen": 567541760
},
{
"epoch": 0.17,
"learning_rate": 0.0004182046138415246,
"loss": 3.095,
"theoretical_loss": 3.86231888181123,
"tokens_seen": 567607296
},
{
"epoch": 0.17,
"learning_rate": 0.00041819458375125376,
"loss": 2.8723,
"theoretical_loss": 3.862271897899287,
"tokens_seen": 567672832
},
{
"epoch": 0.17,
"learning_rate": 0.00041818455366098294,
"loss": 2.847,
"theoretical_loss": 3.862224920929732,
"tokens_seen": 567738368
},
{
"epoch": 0.17,
"learning_rate": 0.0004181745235707121,
"loss": 2.9045,
"theoretical_loss": 3.8621779509007377,
"tokens_seen": 567803904
},
{
"epoch": 0.17,
"learning_rate": 0.00041816449348044136,
"loss": 2.9238,
"theoretical_loss": 3.8621309878104784,
"tokens_seen": 567869440
},
{
"epoch": 0.17,
"learning_rate": 0.00041815446339017054,
"loss": 2.8785,
"theoretical_loss": 3.862084031657128,
"tokens_seen": 567934976
},
{
"epoch": 0.17,
"learning_rate": 0.0004181444332998997,
"loss": 2.7955,
"theoretical_loss": 3.862037082438861,
"tokens_seen": 568000512
},
{
"epoch": 0.17,
"learning_rate": 0.0004181344032096289,
"loss": 2.9675,
"theoretical_loss": 3.8619901401538543,
"tokens_seen": 568066048
},
{
"epoch": 0.17,
"learning_rate": 0.0004181243731193581,
"loss": 2.691,
"theoretical_loss": 3.8619432048002835,
"tokens_seen": 568131584
},
{
"epoch": 0.17,
"learning_rate": 0.0004181143430290873,
"loss": 3.0101,
"theoretical_loss": 3.861896276376326,
"tokens_seen": 568197120
},
{
"epoch": 0.17,
"learning_rate": 0.00041810431293881644,
"loss": 2.9619,
"theoretical_loss": 3.8618493548801602,
"tokens_seen": 568262656
},
{
"epoch": 0.17,
"learning_rate": 0.0004180942828485457,
"loss": 2.9554,
"theoretical_loss": 3.8618024403099636,
"tokens_seen": 568328192
},
{
"epoch": 0.17,
"learning_rate": 0.0004180842527582748,
"loss": 3.0257,
"theoretical_loss": 3.8617555326639157,
"tokens_seen": 568393728
},
{
"epoch": 0.17,
"learning_rate": 0.00041807422266800404,
"loss": 3.0616,
"theoretical_loss": 3.8617086319401968,
"tokens_seen": 568459264
},
{
"epoch": 0.17,
"objective/train/docs_used": 331416,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.896811008453369,
"objective/train/theoretical_loss": 3.861661738136987,
"objective/train/tokens_used": 588984800,
"theoretical_loss": 3.861661738136987,
"tokens_seen": 568524800
},
{
"epoch": 0.17,
"learning_rate": 0.0004180641925777332,
"loss": 2.9509,
"theoretical_loss": 3.861661738136987,
"tokens_seen": 568524800
},
{
"epoch": 0.17,
"learning_rate": 0.0004180541624874624,
"loss": 2.8699,
"theoretical_loss": 3.8616148512524675,
"tokens_seen": 568590336
},
{
"epoch": 0.17,
"learning_rate": 0.0004180441323971916,
"loss": 2.9458,
"theoretical_loss": 3.8615679712848205,
"tokens_seen": 568655872
},
{
"epoch": 0.17,
"learning_rate": 0.0004180341023069208,
"loss": 3.1109,
"theoretical_loss": 3.8615210982322288,
"tokens_seen": 568721408
},
{
"epoch": 0.17,
"learning_rate": 0.00041802407221664994,
"loss": 3.0128,
"theoretical_loss": 3.861474232092876,
"tokens_seen": 568786944
},
{
"epoch": 0.17,
"learning_rate": 0.0004180140421263792,
"loss": 3.1939,
"theoretical_loss": 3.8614273728649446,
"tokens_seen": 568852480
},
{
"epoch": 0.17,
"learning_rate": 0.0004180040120361083,
"loss": 2.831,
"theoretical_loss": 3.8613805205466205,
"tokens_seen": 568918016
},
{
"epoch": 0.17,
"learning_rate": 0.00041799398194583754,
"loss": 2.7662,
"theoretical_loss": 3.8613336751360894,
"tokens_seen": 568983552
},
{
"epoch": 0.17,
"learning_rate": 0.0004179839518555667,
"loss": 2.8514,
"theoretical_loss": 3.861286836631537,
"tokens_seen": 569049088
},
{
"epoch": 0.17,
"learning_rate": 0.0004179739217652959,
"loss": 2.6863,
"theoretical_loss": 3.8612400050311493,
"tokens_seen": 569114624
},
{
"epoch": 0.17,
"learning_rate": 0.0004179638916750251,
"loss": 2.9529,
"theoretical_loss": 3.8611931803331148,
"tokens_seen": 569180160
},
{
"epoch": 0.17,
"learning_rate": 0.00041795386158475427,
"loss": 2.7516,
"theoretical_loss": 3.861146362535621,
"tokens_seen": 569245696
},
{
"epoch": 0.17,
"learning_rate": 0.00041794383149448345,
"loss": 3.1024,
"theoretical_loss": 3.8610995516368565,
"tokens_seen": 569311232
},
{
"epoch": 0.17,
"learning_rate": 0.0004179338014042127,
"loss": 2.8311,
"theoretical_loss": 3.8610527476350116,
"tokens_seen": 569376768
},
{
"epoch": 0.17,
"learning_rate": 0.0004179237713139418,
"loss": 2.7588,
"theoretical_loss": 3.8610059505282757,
"tokens_seen": 569442304
},
{
"epoch": 0.17,
"learning_rate": 0.00041791374122367105,
"loss": 2.7985,
"theoretical_loss": 3.86095916031484,
"tokens_seen": 569507840
},
{
"epoch": 0.17,
"learning_rate": 0.00041790371113340017,
"loss": 2.8401,
"theoretical_loss": 3.860912376992896,
"tokens_seen": 569573376
},
{
"epoch": 0.17,
"learning_rate": 0.0004178936810431294,
"loss": 2.9647,
"theoretical_loss": 3.8608656005606354,
"tokens_seen": 569638912
},
{
"epoch": 0.17,
"learning_rate": 0.0004178836509528586,
"loss": 2.8687,
"theoretical_loss": 3.8608188310162523,
"tokens_seen": 569704448
},
{
"epoch": 0.17,
"learning_rate": 0.00041787362086258777,
"loss": 2.6109,
"theoretical_loss": 3.8607720683579396,
"tokens_seen": 569769984
},
{
"epoch": 0.17,
"learning_rate": 0.00041786359077231695,
"loss": 3.1195,
"theoretical_loss": 3.8607253125838916,
"tokens_seen": 569835520
},
{
"epoch": 0.17,
"learning_rate": 0.0004178535606820462,
"loss": 2.9842,
"theoretical_loss": 3.8606785636923027,
"tokens_seen": 569901056
},
{
"epoch": 0.17,
"learning_rate": 0.0004178435305917753,
"loss": 2.754,
"theoretical_loss": 3.8606318216813693,
"tokens_seen": 569966592
},
{
"epoch": 0.17,
"learning_rate": 0.00041783350050150455,
"loss": 2.779,
"theoretical_loss": 3.8605850865492872,
"tokens_seen": 570032128
},
{
"epoch": 0.17,
"learning_rate": 0.0004178234704112337,
"loss": 3.0349,
"theoretical_loss": 3.860538358294254,
"tokens_seen": 570097664
},
{
"epoch": 0.17,
"objective/train/docs_used": 332819,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.92439341545105,
"objective/train/theoretical_loss": 3.8604916369144666,
"objective/train/tokens_used": 590623200,
"theoretical_loss": 3.8604916369144666,
"tokens_seen": 570163200
},
{
"epoch": 0.17,
"learning_rate": 0.0004178134403209629,
"loss": 3.011,
"theoretical_loss": 3.8604916369144666,
"tokens_seen": 570163200
},
{
"epoch": 0.17,
"learning_rate": 0.0004178034102306921,
"loss": 2.9937,
"theoretical_loss": 3.8604449224081234,
"tokens_seen": 570228736
},
{
"epoch": 0.17,
"learning_rate": 0.0004177933801404213,
"loss": 2.959,
"theoretical_loss": 3.860398214773425,
"tokens_seen": 570294272
},
{
"epoch": 0.17,
"learning_rate": 0.00041778335005015045,
"loss": 3.0047,
"theoretical_loss": 3.8603515140085687,
"tokens_seen": 570359808
},
{
"epoch": 0.17,
"learning_rate": 0.00041777331995987964,
"loss": 2.81,
"theoretical_loss": 3.8603048201117565,
"tokens_seen": 570425344
},
{
"epoch": 0.17,
"learning_rate": 0.0004177632898696088,
"loss": 2.8243,
"theoretical_loss": 3.8602581330811887,
"tokens_seen": 570490880
},
{
"epoch": 0.17,
"learning_rate": 0.00041775325977933805,
"loss": 2.9167,
"theoretical_loss": 3.8602114529150673,
"tokens_seen": 570556416
},
{
"epoch": 0.17,
"learning_rate": 0.0004177432296890672,
"loss": 2.8378,
"theoretical_loss": 3.860164779611595,
"tokens_seen": 570621952
},
{
"epoch": 0.17,
"learning_rate": 0.0004177331995987964,
"loss": 2.6981,
"theoretical_loss": 3.8601181131689737,
"tokens_seen": 570687488
},
{
"epoch": 0.17,
"learning_rate": 0.0004177231695085256,
"loss": 2.8103,
"theoretical_loss": 3.8600714535854093,
"tokens_seen": 570753024
},
{
"epoch": 0.17,
"learning_rate": 0.0004177131394182548,
"loss": 2.8283,
"theoretical_loss": 3.8600248008591045,
"tokens_seen": 570818560
},
{
"epoch": 0.17,
"learning_rate": 0.00041770310932798396,
"loss": 2.7991,
"theoretical_loss": 3.859978154988265,
"tokens_seen": 570884096
},
{
"epoch": 0.17,
"learning_rate": 0.00041769307923771314,
"loss": 2.8655,
"theoretical_loss": 3.8599315159710965,
"tokens_seen": 570949632
},
{
"epoch": 0.17,
"learning_rate": 0.0004176830491474423,
"loss": 2.8578,
"theoretical_loss": 3.8598848838058055,
"tokens_seen": 571015168
},
{
"epoch": 0.17,
"learning_rate": 0.00041767301905717156,
"loss": 2.8153,
"theoretical_loss": 3.8598382584905995,
"tokens_seen": 571080704
},
{
"epoch": 0.17,
"learning_rate": 0.0004176629889669007,
"loss": 3.1572,
"theoretical_loss": 3.859791640023685,
"tokens_seen": 571146240
},
{
"epoch": 0.17,
"learning_rate": 0.0004176529588766299,
"loss": 2.7899,
"theoretical_loss": 3.859745028403273,
"tokens_seen": 571211776
},
{
"epoch": 0.17,
"learning_rate": 0.00041764292878635904,
"loss": 2.7977,
"theoretical_loss": 3.85969842362757,
"tokens_seen": 571277312
},
{
"epoch": 0.17,
"learning_rate": 0.0004176328986960883,
"loss": 2.98,
"theoretical_loss": 3.859651825694787,
"tokens_seen": 571342848
},
{
"epoch": 0.17,
"learning_rate": 0.00041762286860581746,
"loss": 2.8098,
"theoretical_loss": 3.859605234603135,
"tokens_seen": 571408384
},
{
"epoch": 0.17,
"learning_rate": 0.00041761283851554664,
"loss": 2.8236,
"theoretical_loss": 3.8595586503508246,
"tokens_seen": 571473920
},
{
"epoch": 0.17,
"learning_rate": 0.0004176028084252758,
"loss": 3.083,
"theoretical_loss": 3.8595120729360675,
"tokens_seen": 571539456
},
{
"epoch": 0.17,
"learning_rate": 0.000417592778335005,
"loss": 2.711,
"theoretical_loss": 3.8594655023570765,
"tokens_seen": 571604992
},
{
"epoch": 0.17,
"learning_rate": 0.0004175827482447342,
"loss": 2.8807,
"theoretical_loss": 3.859418938612065,
"tokens_seen": 571670528
},
{
"epoch": 0.17,
"learning_rate": 0.0004175727181544634,
"loss": 3.1158,
"theoretical_loss": 3.859372381699247,
"tokens_seen": 571736064
},
{
"epoch": 0.17,
"objective/train/docs_used": 333556,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.679062604904175,
"objective/train/theoretical_loss": 3.859325831616836,
"objective/train/tokens_used": 592261600,
"theoretical_loss": 3.859325831616836,
"tokens_seen": 571801600
},
{
"epoch": 0.17,
"learning_rate": 0.00041756268806419255,
"loss": 2.6757,
"theoretical_loss": 3.859325831616836,
"tokens_seen": 571801600
},
{
"epoch": 0.17,
"learning_rate": 0.0004175526579739218,
"loss": 2.8884,
"theoretical_loss": 3.859279288363048,
"tokens_seen": 571867136
},
{
"epoch": 0.17,
"learning_rate": 0.00041754262788365096,
"loss": 3.0039,
"theoretical_loss": 3.8592327519360996,
"tokens_seen": 571932672
},
{
"epoch": 0.17,
"learning_rate": 0.00041753259779338015,
"loss": 2.6379,
"theoretical_loss": 3.8591862223342064,
"tokens_seen": 571998208
},
{
"epoch": 0.17,
"learning_rate": 0.0004175225677031093,
"loss": 2.8799,
"theoretical_loss": 3.8591396995555853,
"tokens_seen": 572063744
},
{
"epoch": 0.17,
"learning_rate": 0.0004175125376128385,
"loss": 2.9484,
"theoretical_loss": 3.859093183598455,
"tokens_seen": 572129280
},
{
"epoch": 0.17,
"learning_rate": 0.0004175025075225677,
"loss": 2.8176,
"theoretical_loss": 3.859046674461034,
"tokens_seen": 572194816
},
{
"epoch": 0.17,
"learning_rate": 0.0004174924774322969,
"loss": 2.8921,
"theoretical_loss": 3.859000172141541,
"tokens_seen": 572260352
},
{
"epoch": 0.17,
"learning_rate": 0.00041748244734202605,
"loss": 2.8583,
"theoretical_loss": 3.8589536766381967,
"tokens_seen": 572325888
},
{
"epoch": 0.17,
"learning_rate": 0.0004174724172517553,
"loss": 2.9911,
"theoretical_loss": 3.858907187949221,
"tokens_seen": 572391424
},
{
"epoch": 0.17,
"learning_rate": 0.0004174623871614844,
"loss": 2.6848,
"theoretical_loss": 3.8588607060728357,
"tokens_seen": 572456960
},
{
"epoch": 0.17,
"learning_rate": 0.00041745235707121365,
"loss": 2.8667,
"theoretical_loss": 3.858814231007262,
"tokens_seen": 572522496
},
{
"epoch": 0.17,
"learning_rate": 0.00041744232698094283,
"loss": 3.1139,
"theoretical_loss": 3.8587677627507233,
"tokens_seen": 572588032
},
{
"epoch": 0.17,
"learning_rate": 0.000417432296890672,
"loss": 2.8625,
"theoretical_loss": 3.858721301301442,
"tokens_seen": 572653568
},
{
"epoch": 0.17,
"learning_rate": 0.0004174222668004012,
"loss": 2.9168,
"theoretical_loss": 3.8586748466576424,
"tokens_seen": 572719104
},
{
"epoch": 0.17,
"learning_rate": 0.00041741223671013037,
"loss": 3.0097,
"theoretical_loss": 3.8586283988175496,
"tokens_seen": 572784640
},
{
"epoch": 0.17,
"learning_rate": 0.0004174022066198596,
"loss": 2.9648,
"theoretical_loss": 3.8585819577793883,
"tokens_seen": 572850176
},
{
"epoch": 0.17,
"learning_rate": 0.0004173921765295888,
"loss": 2.7055,
"theoretical_loss": 3.8585355235413843,
"tokens_seen": 572915712
},
{
"epoch": 0.17,
"learning_rate": 0.00041738214643931797,
"loss": 2.6896,
"theoretical_loss": 3.858489096101765,
"tokens_seen": 572981248
},
{
"epoch": 0.17,
"learning_rate": 0.00041737211634904715,
"loss": 2.8474,
"theoretical_loss": 3.858442675458756,
"tokens_seen": 573046784
},
{
"epoch": 0.17,
"learning_rate": 0.0004173620862587764,
"loss": 2.8638,
"theoretical_loss": 3.8583962616105874,
"tokens_seen": 573112320
},
{
"epoch": 0.17,
"learning_rate": 0.0004173520561685055,
"loss": 3.0345,
"theoretical_loss": 3.858349854555486,
"tokens_seen": 573177856
},
{
"epoch": 0.17,
"learning_rate": 0.00041734202607823475,
"loss": 2.8815,
"theoretical_loss": 3.858303454291682,
"tokens_seen": 573243392
},
{
"epoch": 0.17,
"learning_rate": 0.0004173319959879639,
"loss": 2.9809,
"theoretical_loss": 3.8582570608174045,
"tokens_seen": 573308928
},
{
"epoch": 0.17,
"learning_rate": 0.0004173219658976931,
"loss": 2.8129,
"theoretical_loss": 3.858210674130885,
"tokens_seen": 573374464
},
{
"epoch": 0.17,
"objective/train/docs_used": 334721,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6601226329803467,
"objective/train/theoretical_loss": 3.858164294230354,
"objective/train/tokens_used": 593900000,
"theoretical_loss": 3.858164294230354,
"tokens_seen": 573440000
},
{
"epoch": 0.17,
"learning_rate": 0.0004173119358074223,
"loss": 3.0888,
"theoretical_loss": 3.858164294230354,
"tokens_seen": 573440000
},
{
"epoch": 0.17,
"learning_rate": 0.0004173019057171515,
"loss": 2.9326,
"theoretical_loss": 3.8581179211140437,
"tokens_seen": 573505536
},
{
"epoch": 0.17,
"learning_rate": 0.00041729187562688065,
"loss": 2.9435,
"theoretical_loss": 3.8580715547801865,
"tokens_seen": 573571072
},
{
"epoch": 0.17,
"learning_rate": 0.00041728184553660984,
"loss": 2.777,
"theoretical_loss": 3.8580251952270164,
"tokens_seen": 573636608
},
{
"epoch": 0.17,
"learning_rate": 0.000417271815446339,
"loss": 2.9294,
"theoretical_loss": 3.8579788424527655,
"tokens_seen": 573702144
},
{
"epoch": 0.17,
"learning_rate": 0.00041726178535606825,
"loss": 3.1094,
"theoretical_loss": 3.8579324964556703,
"tokens_seen": 573767680
},
{
"epoch": 0.17,
"learning_rate": 0.0004172517552657974,
"loss": 3.2043,
"theoretical_loss": 3.857886157233965,
"tokens_seen": 573833216
},
{
"epoch": 0.17,
"learning_rate": 0.0004172417251755266,
"loss": 2.8649,
"theoretical_loss": 3.8578398247858856,
"tokens_seen": 573898752
},
{
"epoch": 0.17,
"learning_rate": 0.0004172316950852558,
"loss": 2.6749,
"theoretical_loss": 3.857793499109669,
"tokens_seen": 573964288
},
{
"epoch": 0.17,
"learning_rate": 0.000417221664994985,
"loss": 2.7361,
"theoretical_loss": 3.8577471802035515,
"tokens_seen": 574029824
},
{
"epoch": 0.17,
"learning_rate": 0.00041721163490471416,
"loss": 2.934,
"theoretical_loss": 3.8577008680657716,
"tokens_seen": 574095360
},
{
"epoch": 0.17,
"learning_rate": 0.00041720160481444334,
"loss": 2.8451,
"theoretical_loss": 3.8576545626945684,
"tokens_seen": 574160896
},
{
"epoch": 0.17,
"learning_rate": 0.0004171915747241725,
"loss": 2.7165,
"theoretical_loss": 3.8576082640881797,
"tokens_seen": 574226432
},
{
"epoch": 0.17,
"learning_rate": 0.00041718154463390176,
"loss": 2.7004,
"theoretical_loss": 3.857561972244846,
"tokens_seen": 574291968
},
{
"epoch": 0.17,
"learning_rate": 0.0004171715145436309,
"loss": 2.9106,
"theoretical_loss": 3.8575156871628082,
"tokens_seen": 574357504
},
{
"epoch": 0.17,
"learning_rate": 0.0004171614844533601,
"loss": 2.984,
"theoretical_loss": 3.8574694088403065,
"tokens_seen": 574423040
},
{
"epoch": 0.17,
"learning_rate": 0.00041715145436308924,
"loss": 2.8751,
"theoretical_loss": 3.8574231372755836,
"tokens_seen": 574488576
},
{
"epoch": 0.17,
"learning_rate": 0.0004171414242728185,
"loss": 2.7598,
"theoretical_loss": 3.8573768724668813,
"tokens_seen": 574554112
},
{
"epoch": 0.17,
"learning_rate": 0.00041713139418254766,
"loss": 2.8516,
"theoretical_loss": 3.8573306144124437,
"tokens_seen": 574619648
},
{
"epoch": 0.17,
"learning_rate": 0.00041712136409227684,
"loss": 2.8133,
"theoretical_loss": 3.857284363110513,
"tokens_seen": 574685184
},
{
"epoch": 0.17,
"learning_rate": 0.000417111334002006,
"loss": 2.7077,
"theoretical_loss": 3.8572381185593354,
"tokens_seen": 574750720
},
{
"epoch": 0.17,
"learning_rate": 0.0004171013039117352,
"loss": 3.0123,
"theoretical_loss": 3.8571918807571546,
"tokens_seen": 574816256
},
{
"epoch": 0.17,
"learning_rate": 0.0004170912738214644,
"loss": 2.8599,
"theoretical_loss": 3.8571456497022165,
"tokens_seen": 574881792
},
{
"epoch": 0.17,
"learning_rate": 0.0004170812437311936,
"loss": 2.9078,
"theoretical_loss": 3.857099425392768,
"tokens_seen": 574947328
},
{
"epoch": 0.17,
"learning_rate": 0.00041707121364092275,
"loss": 2.8265,
"theoretical_loss": 3.8570532078270565,
"tokens_seen": 575012864
},
{
"epoch": 0.17,
"objective/train/docs_used": 335233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9721813201904297,
"objective/train/theoretical_loss": 3.857006997003329,
"objective/train/tokens_used": 595538400,
"theoretical_loss": 3.857006997003329,
"tokens_seen": 575078400
},
{
"epoch": 0.17,
"learning_rate": 0.000417061183550652,
"loss": 2.9865,
"theoretical_loss": 3.857006997003329,
"tokens_seen": 575078400
},
{
"epoch": 0.17,
"learning_rate": 0.00041705115346038116,
"loss": 2.6959,
"theoretical_loss": 3.8569607929198337,
"tokens_seen": 575143936
},
{
"epoch": 0.17,
"learning_rate": 0.00041704112337011035,
"loss": 2.9934,
"theoretical_loss": 3.85691459557482,
"tokens_seen": 575209472
},
{
"epoch": 0.17,
"learning_rate": 0.0004170310932798395,
"loss": 2.6767,
"theoretical_loss": 3.856868404966538,
"tokens_seen": 575275008
},
{
"epoch": 0.17,
"learning_rate": 0.0004170210631895687,
"loss": 2.934,
"theoretical_loss": 3.8568222210932372,
"tokens_seen": 575340544
},
{
"epoch": 0.17,
"learning_rate": 0.0004170110330992979,
"loss": 2.8967,
"theoretical_loss": 3.856776043953169,
"tokens_seen": 575406080
},
{
"epoch": 0.17,
"learning_rate": 0.0004170010030090271,
"loss": 2.8635,
"theoretical_loss": 3.8567298735445847,
"tokens_seen": 575471616
},
{
"epoch": 0.17,
"learning_rate": 0.00041699097291875625,
"loss": 2.6781,
"theoretical_loss": 3.856683709865737,
"tokens_seen": 575537152
},
{
"epoch": 0.17,
"learning_rate": 0.0004169809428284855,
"loss": 2.9516,
"theoretical_loss": 3.8566375529148793,
"tokens_seen": 575602688
},
{
"epoch": 0.17,
"learning_rate": 0.0004169709127382146,
"loss": 2.7865,
"theoretical_loss": 3.856591402690264,
"tokens_seen": 575668224
},
{
"epoch": 0.17,
"learning_rate": 0.00041696088264794385,
"loss": 2.7155,
"theoretical_loss": 3.8565452591901463,
"tokens_seen": 575733760
},
{
"epoch": 0.17,
"learning_rate": 0.00041695085255767303,
"loss": 2.892,
"theoretical_loss": 3.856499122412781,
"tokens_seen": 575799296
},
{
"epoch": 0.17,
"learning_rate": 0.0004169408224674022,
"loss": 3.0308,
"theoretical_loss": 3.8564529923564232,
"tokens_seen": 575864832
},
{
"epoch": 0.17,
"learning_rate": 0.0004169307923771314,
"loss": 2.7946,
"theoretical_loss": 3.856406869019329,
"tokens_seen": 575930368
},
{
"epoch": 0.17,
"learning_rate": 0.0004169207622868606,
"loss": 2.8613,
"theoretical_loss": 3.8563607523997563,
"tokens_seen": 575995904
},
{
"epoch": 0.17,
"learning_rate": 0.00041691073219658975,
"loss": 2.8431,
"theoretical_loss": 3.856314642495961,
"tokens_seen": 576061440
},
{
"epoch": 0.17,
"learning_rate": 0.000416900702106319,
"loss": 2.8555,
"theoretical_loss": 3.8562685393062024,
"tokens_seen": 576126976
},
{
"epoch": 0.17,
"learning_rate": 0.0004168906720160481,
"loss": 2.857,
"theoretical_loss": 3.8562224428287393,
"tokens_seen": 576192512
},
{
"epoch": 0.17,
"learning_rate": 0.00041688064192577735,
"loss": 2.8366,
"theoretical_loss": 3.85617635306183,
"tokens_seen": 576258048
},
{
"epoch": 0.17,
"learning_rate": 0.00041687061183550653,
"loss": 2.831,
"theoretical_loss": 3.856130270003736,
"tokens_seen": 576323584
},
{
"epoch": 0.17,
"learning_rate": 0.0004168605817452357,
"loss": 2.8643,
"theoretical_loss": 3.856084193652718,
"tokens_seen": 576389120
},
{
"epoch": 0.17,
"learning_rate": 0.0004168505516549649,
"loss": 2.6865,
"theoretical_loss": 3.8560381240070365,
"tokens_seen": 576454656
},
{
"epoch": 0.17,
"learning_rate": 0.0004168405215646941,
"loss": 2.8393,
"theoretical_loss": 3.8559920610649536,
"tokens_seen": 576520192
},
{
"epoch": 0.17,
"learning_rate": 0.00041683049147442326,
"loss": 2.7828,
"theoretical_loss": 3.855946004824733,
"tokens_seen": 576585728
},
{
"epoch": 0.17,
"learning_rate": 0.0004168204613841525,
"loss": 2.7994,
"theoretical_loss": 3.8558999552846367,
"tokens_seen": 576651264
},
{
"epoch": 0.17,
"objective/train/docs_used": 335838,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.021301031112671,
"objective/train/theoretical_loss": 3.85585391244293,
"objective/train/tokens_used": 597176800,
"theoretical_loss": 3.85585391244293,
"tokens_seen": 576716800
},
{
"epoch": 0.17,
"learning_rate": 0.0004168104312938816,
"loss": 2.7419,
"theoretical_loss": 3.85585391244293,
"tokens_seen": 576716800
},
{
"epoch": 0.17,
"learning_rate": 0.00041680040120361085,
"loss": 2.8041,
"theoretical_loss": 3.855807876297876,
"tokens_seen": 576782336
},
{
"epoch": 0.17,
"learning_rate": 0.00041679037111334,
"loss": 2.6524,
"theoretical_loss": 3.8557618468477415,
"tokens_seen": 576847872
},
{
"epoch": 0.17,
"learning_rate": 0.0004167803410230692,
"loss": 3.0374,
"theoretical_loss": 3.8557158240907916,
"tokens_seen": 576913408
},
{
"epoch": 0.17,
"learning_rate": 0.0004167703109327984,
"loss": 3.1909,
"theoretical_loss": 3.8556698080252936,
"tokens_seen": 576978944
},
{
"epoch": 0.17,
"learning_rate": 0.0004167602808425276,
"loss": 2.6784,
"theoretical_loss": 3.8556237986495137,
"tokens_seen": 577044480
},
{
"epoch": 0.17,
"learning_rate": 0.00041675025075225676,
"loss": 2.6979,
"theoretical_loss": 3.8555777959617203,
"tokens_seen": 577110016
},
{
"epoch": 0.17,
"learning_rate": 0.000416740220661986,
"loss": 2.9514,
"theoretical_loss": 3.8555317999601817,
"tokens_seen": 577175552
},
{
"epoch": 0.17,
"learning_rate": 0.0004167301905717151,
"loss": 3.0444,
"theoretical_loss": 3.8554858106431675,
"tokens_seen": 577241088
},
{
"epoch": 0.17,
"learning_rate": 0.00041672016048144436,
"loss": 3.0292,
"theoretical_loss": 3.8554398280089472,
"tokens_seen": 577306624
},
{
"epoch": 0.17,
"learning_rate": 0.0004167101303911735,
"loss": 2.7328,
"theoretical_loss": 3.8553938520557915,
"tokens_seen": 577372160
},
{
"epoch": 0.17,
"learning_rate": 0.0004167001003009027,
"loss": 2.765,
"theoretical_loss": 3.8553478827819707,
"tokens_seen": 577437696
},
{
"epoch": 0.18,
"learning_rate": 0.0004166900702106319,
"loss": 2.8839,
"theoretical_loss": 3.8553019201857577,
"tokens_seen": 577503232
},
{
"epoch": 0.18,
"learning_rate": 0.0004166800401203611,
"loss": 2.7088,
"theoretical_loss": 3.8552559642654236,
"tokens_seen": 577568768
},
{
"epoch": 0.18,
"learning_rate": 0.00041667001003009026,
"loss": 2.9248,
"theoretical_loss": 3.8552100150192423,
"tokens_seen": 577634304
},
{
"epoch": 0.18,
"learning_rate": 0.00041665997993981944,
"loss": 2.7546,
"theoretical_loss": 3.855164072445487,
"tokens_seen": 577699840
},
{
"epoch": 0.18,
"learning_rate": 0.0004166499498495487,
"loss": 3.0814,
"theoretical_loss": 3.855118136542433,
"tokens_seen": 577765376
},
{
"epoch": 0.18,
"learning_rate": 0.00041663991975927786,
"loss": 2.6923,
"theoretical_loss": 3.8550722073083543,
"tokens_seen": 577830912
},
{
"epoch": 0.18,
"learning_rate": 0.00041662988966900704,
"loss": 2.7361,
"theoretical_loss": 3.855026284741526,
"tokens_seen": 577896448
},
{
"epoch": 0.18,
"learning_rate": 0.0004166198595787362,
"loss": 2.867,
"theoretical_loss": 3.8549803688402253,
"tokens_seen": 577961984
},
{
"epoch": 0.18,
"learning_rate": 0.0004166098294884654,
"loss": 2.765,
"theoretical_loss": 3.8549344596027293,
"tokens_seen": 578027520
},
{
"epoch": 0.18,
"learning_rate": 0.0004165997993981946,
"loss": 3.1613,
"theoretical_loss": 3.854888557027315,
"tokens_seen": 578093056
},
{
"epoch": 0.18,
"learning_rate": 0.0004165897693079238,
"loss": 2.8802,
"theoretical_loss": 3.8548426611122597,
"tokens_seen": 578158592
},
{
"epoch": 0.18,
"learning_rate": 0.00041657973921765295,
"loss": 2.7569,
"theoretical_loss": 3.8547967718558436,
"tokens_seen": 578224128
},
{
"epoch": 0.18,
"learning_rate": 0.0004165697091273822,
"loss": 2.7319,
"theoretical_loss": 3.854750889256345,
"tokens_seen": 578289664
},
{
"epoch": 0.18,
"objective/train/docs_used": 337192,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4397428035736084,
"objective/train/theoretical_loss": 3.854705013312045,
"objective/train/tokens_used": 598815200,
"theoretical_loss": 3.854705013312045,
"tokens_seen": 578355200
},
{
"epoch": 0.18,
"learning_rate": 0.00041655967903711136,
"loss": 3.0677,
"theoretical_loss": 3.854705013312045,
"tokens_seen": 578355200
},
{
"epoch": 0.18,
"learning_rate": 0.00041654964894684055,
"loss": 2.7335,
"theoretical_loss": 3.8546591440212237,
"tokens_seen": 578420736
},
{
"epoch": 0.18,
"learning_rate": 0.0004165396188565697,
"loss": 2.7197,
"theoretical_loss": 3.8546132813821625,
"tokens_seen": 578486272
},
{
"epoch": 0.18,
"learning_rate": 0.0004165295887662989,
"loss": 2.7768,
"theoretical_loss": 3.8545674253931432,
"tokens_seen": 578551808
},
{
"epoch": 0.18,
"learning_rate": 0.0004165195586760281,
"loss": 3.0142,
"theoretical_loss": 3.8545215760524494,
"tokens_seen": 578617344
},
{
"epoch": 0.18,
"learning_rate": 0.0004165095285857573,
"loss": 2.7591,
"theoretical_loss": 3.8544757333583632,
"tokens_seen": 578682880
},
{
"epoch": 0.18,
"learning_rate": 0.00041649949849548645,
"loss": 2.8577,
"theoretical_loss": 3.854429897309169,
"tokens_seen": 578748416
},
{
"epoch": 0.18,
"learning_rate": 0.0004164894684052157,
"loss": 2.8507,
"theoretical_loss": 3.854384067903151,
"tokens_seen": 578813952
},
{
"epoch": 0.18,
"learning_rate": 0.0004164794383149448,
"loss": 3.0269,
"theoretical_loss": 3.854338245138595,
"tokens_seen": 578879488
},
{
"epoch": 0.18,
"learning_rate": 0.00041646940822467405,
"loss": 3.0333,
"theoretical_loss": 3.8542924290137863,
"tokens_seen": 578945024
},
{
"epoch": 0.18,
"learning_rate": 0.00041645937813440323,
"loss": 3.0334,
"theoretical_loss": 3.8542466195270118,
"tokens_seen": 579010560
},
{
"epoch": 0.18,
"learning_rate": 0.0004164493480441324,
"loss": 2.9967,
"theoretical_loss": 3.8542008166765576,
"tokens_seen": 579076096
},
{
"epoch": 0.18,
"learning_rate": 0.0004164393179538616,
"loss": 2.7751,
"theoretical_loss": 3.8541550204607127,
"tokens_seen": 579141632
},
{
"epoch": 0.18,
"learning_rate": 0.0004164292878635908,
"loss": 2.7921,
"theoretical_loss": 3.8541092308777647,
"tokens_seen": 579207168
},
{
"epoch": 0.18,
"learning_rate": 0.00041641925777331995,
"loss": 2.8874,
"theoretical_loss": 3.854063447926003,
"tokens_seen": 579272704
},
{
"epoch": 0.18,
"learning_rate": 0.0004164092276830492,
"loss": 3.0442,
"theoretical_loss": 3.8540176716037164,
"tokens_seen": 579338240
},
{
"epoch": 0.18,
"learning_rate": 0.0004163991975927783,
"loss": 2.8858,
"theoretical_loss": 3.853971901909196,
"tokens_seen": 579403776
},
{
"epoch": 0.18,
"learning_rate": 0.00041638916750250755,
"loss": 2.8424,
"theoretical_loss": 3.8539261388407327,
"tokens_seen": 579469312
},
{
"epoch": 0.18,
"learning_rate": 0.00041637913741223673,
"loss": 2.9964,
"theoretical_loss": 3.8538803823966172,
"tokens_seen": 579534848
},
{
"epoch": 0.18,
"learning_rate": 0.0004163691073219659,
"loss": 2.843,
"theoretical_loss": 3.853834632575143,
"tokens_seen": 579600384
},
{
"epoch": 0.18,
"learning_rate": 0.0004163590772316951,
"loss": 2.793,
"theoretical_loss": 3.8537888893746013,
"tokens_seen": 579665920
},
{
"epoch": 0.18,
"learning_rate": 0.0004163490471414243,
"loss": 3.0142,
"theoretical_loss": 3.853743152793287,
"tokens_seen": 579731456
},
{
"epoch": 0.18,
"learning_rate": 0.00041633901705115346,
"loss": 2.8166,
"theoretical_loss": 3.8536974228294927,
"tokens_seen": 579796992
},
{
"epoch": 0.18,
"learning_rate": 0.0004163289869608827,
"loss": 2.996,
"theoretical_loss": 3.8536516994815146,
"tokens_seen": 579862528
},
{
"epoch": 0.18,
"learning_rate": 0.0004163189568706118,
"loss": 2.8994,
"theoretical_loss": 3.8536059827476477,
"tokens_seen": 579928064
},
{
"epoch": 0.18,
"objective/train/docs_used": 337878,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1601502895355225,
"objective/train/theoretical_loss": 3.8535602726261864,
"objective/train/tokens_used": 600453600,
"theoretical_loss": 3.8535602726261864,
"tokens_seen": 579993600
},
{
"epoch": 0.18,
"learning_rate": 0.00041630892678034105,
"loss": 2.8386,
"theoretical_loss": 3.8535602726261864,
"tokens_seen": 579993600
},
{
"epoch": 0.18,
"learning_rate": 0.0004162988966900702,
"loss": 2.8647,
"theoretical_loss": 3.853514569115429,
"tokens_seen": 580059136
},
{
"epoch": 0.18,
"learning_rate": 0.0004162888665997994,
"loss": 2.9424,
"theoretical_loss": 3.8534688722136723,
"tokens_seen": 580124672
},
{
"epoch": 0.18,
"learning_rate": 0.0004162788365095286,
"loss": 2.7317,
"theoretical_loss": 3.853423181919214,
"tokens_seen": 580190208
},
{
"epoch": 0.18,
"learning_rate": 0.0004162688064192578,
"loss": 2.8081,
"theoretical_loss": 3.8533774982303526,
"tokens_seen": 580255744
},
{
"epoch": 0.18,
"learning_rate": 0.00041625877632898696,
"loss": 3.023,
"theoretical_loss": 3.853331821145387,
"tokens_seen": 580321280
},
{
"epoch": 0.18,
"learning_rate": 0.0004162487462387162,
"loss": 2.8905,
"theoretical_loss": 3.8532861506626173,
"tokens_seen": 580386816
},
{
"epoch": 0.18,
"learning_rate": 0.0004162387161484453,
"loss": 2.8099,
"theoretical_loss": 3.853240486780344,
"tokens_seen": 580452352
},
{
"epoch": 0.18,
"learning_rate": 0.00041622868605817456,
"loss": 3.0392,
"theoretical_loss": 3.8531948294968674,
"tokens_seen": 580517888
},
{
"epoch": 0.18,
"learning_rate": 0.0004162186559679037,
"loss": 2.6887,
"theoretical_loss": 3.85314917881049,
"tokens_seen": 580583424
},
{
"epoch": 0.18,
"learning_rate": 0.0004162086258776329,
"loss": 2.8255,
"theoretical_loss": 3.853103534719513,
"tokens_seen": 580648960
},
{
"epoch": 0.18,
"learning_rate": 0.0004161985957873621,
"loss": 2.6804,
"theoretical_loss": 3.8530578972222402,
"tokens_seen": 580714496
},
{
"epoch": 0.18,
"learning_rate": 0.0004161885656970913,
"loss": 3.0892,
"theoretical_loss": 3.8530122663169752,
"tokens_seen": 580780032
},
{
"epoch": 0.18,
"learning_rate": 0.00041617853560682046,
"loss": 2.7583,
"theoretical_loss": 3.8529666420020217,
"tokens_seen": 580845568
},
{
"epoch": 0.18,
"learning_rate": 0.00041616850551654964,
"loss": 2.6621,
"theoretical_loss": 3.852921024275684,
"tokens_seen": 580911104
},
{
"epoch": 0.18,
"learning_rate": 0.0004161584754262788,
"loss": 2.7139,
"theoretical_loss": 3.8528754131362684,
"tokens_seen": 580976640
},
{
"epoch": 0.18,
"learning_rate": 0.00041614844533600806,
"loss": 3.1473,
"theoretical_loss": 3.8528298085820802,
"tokens_seen": 581042176
},
{
"epoch": 0.18,
"learning_rate": 0.0004161384152457372,
"loss": 3.281,
"theoretical_loss": 3.8527842106114267,
"tokens_seen": 581107712
},
{
"epoch": 0.18,
"learning_rate": 0.0004161283851554664,
"loss": 2.8712,
"theoretical_loss": 3.852738619222615,
"tokens_seen": 581173248
},
{
"epoch": 0.18,
"learning_rate": 0.00041611835506519555,
"loss": 2.9122,
"theoretical_loss": 3.8526930344139525,
"tokens_seen": 581238784
},
{
"epoch": 0.18,
"learning_rate": 0.0004161083249749248,
"loss": 2.7869,
"theoretical_loss": 3.8526474561837483,
"tokens_seen": 581304320
},
{
"epoch": 0.18,
"learning_rate": 0.00041609829488465397,
"loss": 2.9412,
"theoretical_loss": 3.852601884530311,
"tokens_seen": 581369856
},
{
"epoch": 0.18,
"learning_rate": 0.00041608826479438315,
"loss": 2.6746,
"theoretical_loss": 3.852556319451952,
"tokens_seen": 581435392
},
{
"epoch": 0.18,
"learning_rate": 0.00041607823470411233,
"loss": 3.056,
"theoretical_loss": 3.8525107609469793,
"tokens_seen": 581500928
},
{
"epoch": 0.18,
"learning_rate": 0.00041606820461384156,
"loss": 2.8142,
"theoretical_loss": 3.852465209013705,
"tokens_seen": 581566464
},
{
"epoch": 0.18,
"objective/train/docs_used": 339259,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0680527687072754,
"objective/train/theoretical_loss": 3.8524196636504415,
"objective/train/tokens_used": 602092000,
"theoretical_loss": 3.8524196636504415,
"tokens_seen": 581632000
},
{
"epoch": 0.18,
"learning_rate": 0.0004160581745235707,
"loss": 2.9815,
"theoretical_loss": 3.8524196636504415,
"tokens_seen": 581632000
},
{
"epoch": 0.18,
"learning_rate": 0.0004160481444332999,
"loss": 2.8625,
"theoretical_loss": 3.8523741248555,
"tokens_seen": 581697536
},
{
"epoch": 0.18,
"learning_rate": 0.00041603811434302905,
"loss": 2.9147,
"theoretical_loss": 3.8523285926271935,
"tokens_seen": 581763072
},
{
"epoch": 0.18,
"learning_rate": 0.0004160280842527583,
"loss": 2.8707,
"theoretical_loss": 3.852283066963836,
"tokens_seen": 581828608
},
{
"epoch": 0.18,
"learning_rate": 0.00041601805416248747,
"loss": 3.0122,
"theoretical_loss": 3.8522375478637425,
"tokens_seen": 581894144
},
{
"epoch": 0.18,
"learning_rate": 0.00041600802407221665,
"loss": 2.7454,
"theoretical_loss": 3.852192035325226,
"tokens_seen": 581959680
},
{
"epoch": 0.18,
"learning_rate": 0.00041599799398194583,
"loss": 2.6902,
"theoretical_loss": 3.8521465293466024,
"tokens_seen": 582025216
},
{
"epoch": 0.18,
"learning_rate": 0.000415987963891675,
"loss": 2.7896,
"theoretical_loss": 3.852101029926188,
"tokens_seen": 582090752
},
{
"epoch": 0.18,
"learning_rate": 0.0004159779338014042,
"loss": 2.8443,
"theoretical_loss": 3.8520555370622995,
"tokens_seen": 582156288
},
{
"epoch": 0.18,
"learning_rate": 0.00041596790371113343,
"loss": 2.575,
"theoretical_loss": 3.852010050753254,
"tokens_seen": 582221824
},
{
"epoch": 0.18,
"learning_rate": 0.00041595787362086256,
"loss": 2.7256,
"theoretical_loss": 3.85196457099737,
"tokens_seen": 582287360
},
{
"epoch": 0.18,
"learning_rate": 0.0004159478435305918,
"loss": 2.8148,
"theoretical_loss": 3.8519190977929645,
"tokens_seen": 582352896
},
{
"epoch": 0.18,
"learning_rate": 0.0004159378134403209,
"loss": 2.7325,
"theoretical_loss": 3.851873631138358,
"tokens_seen": 582418432
},
{
"epoch": 0.18,
"learning_rate": 0.00041592778335005015,
"loss": 2.8538,
"theoretical_loss": 3.85182817103187,
"tokens_seen": 582483968
},
{
"epoch": 0.18,
"learning_rate": 0.00041591775325977934,
"loss": 2.8132,
"theoretical_loss": 3.8517827174718198,
"tokens_seen": 582549504
},
{
"epoch": 0.18,
"learning_rate": 0.0004159077231695085,
"loss": 2.6763,
"theoretical_loss": 3.85173727045653,
"tokens_seen": 582615040
},
{
"epoch": 0.18,
"learning_rate": 0.00041589769307923775,
"loss": 2.7962,
"theoretical_loss": 3.851691829984321,
"tokens_seen": 582680576
},
{
"epoch": 0.18,
"learning_rate": 0.00041588766298896693,
"loss": 2.82,
"theoretical_loss": 3.851646396053516,
"tokens_seen": 582746112
},
{
"epoch": 0.18,
"learning_rate": 0.0004158776328986961,
"loss": 2.9067,
"theoretical_loss": 3.851600968662437,
"tokens_seen": 582811648
},
{
"epoch": 0.18,
"learning_rate": 0.0004158676028084253,
"loss": 2.6858,
"theoretical_loss": 3.851555547809408,
"tokens_seen": 582877184
},
{
"epoch": 0.18,
"learning_rate": 0.0004158575727181545,
"loss": 2.7076,
"theoretical_loss": 3.851510133492752,
"tokens_seen": 582942720
},
{
"epoch": 0.18,
"learning_rate": 0.00041584754262788366,
"loss": 2.7853,
"theoretical_loss": 3.8514647257107955,
"tokens_seen": 583008256
},
{
"epoch": 0.18,
"learning_rate": 0.0004158375125376129,
"loss": 3.0245,
"theoretical_loss": 3.8514193244618617,
"tokens_seen": 583073792
},
{
"epoch": 0.18,
"learning_rate": 0.000415827482447342,
"loss": 2.8126,
"theoretical_loss": 3.8513739297442786,
"tokens_seen": 583139328
},
{
"epoch": 0.18,
"learning_rate": 0.00041581745235707126,
"loss": 2.9131,
"theoretical_loss": 3.851328541556372,
"tokens_seen": 583204864
},
{
"epoch": 0.18,
"objective/train/docs_used": 340083,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4392688274383545,
"objective/train/theoretical_loss": 3.851283159896468,
"objective/train/tokens_used": 603730400,
"theoretical_loss": 3.851283159896468,
"tokens_seen": 583270400
},
{
"epoch": 0.18,
"learning_rate": 0.0004158074222668004,
"loss": 2.9399,
"theoretical_loss": 3.851283159896468,
"tokens_seen": 583270400
},
{
"epoch": 0.18,
"learning_rate": 0.0004157973921765296,
"loss": 2.9062,
"theoretical_loss": 3.851237784762896,
"tokens_seen": 583335936
},
{
"epoch": 0.18,
"learning_rate": 0.0004157873620862588,
"loss": 2.7214,
"theoretical_loss": 3.8511924161539826,
"tokens_seen": 583401472
},
{
"epoch": 0.18,
"learning_rate": 0.000415777331995988,
"loss": 2.7425,
"theoretical_loss": 3.8511470540680586,
"tokens_seen": 583467008
},
{
"epoch": 0.18,
"learning_rate": 0.00041576730190571716,
"loss": 2.8637,
"theoretical_loss": 3.8511016985034523,
"tokens_seen": 583532544
},
{
"epoch": 0.18,
"learning_rate": 0.0004157572718154464,
"loss": 2.9176,
"theoretical_loss": 3.851056349458495,
"tokens_seen": 583598080
},
{
"epoch": 0.18,
"learning_rate": 0.0004157472417251755,
"loss": 2.9171,
"theoretical_loss": 3.8510110069315173,
"tokens_seen": 583663616
},
{
"epoch": 0.18,
"learning_rate": 0.00041573721163490476,
"loss": 2.9357,
"theoretical_loss": 3.85096567092085,
"tokens_seen": 583729152
},
{
"epoch": 0.18,
"learning_rate": 0.0004157271815446339,
"loss": 2.7137,
"theoretical_loss": 3.8509203414248256,
"tokens_seen": 583794688
},
{
"epoch": 0.18,
"learning_rate": 0.0004157171514543631,
"loss": 2.7805,
"theoretical_loss": 3.8508750184417764,
"tokens_seen": 583860224
},
{
"epoch": 0.18,
"learning_rate": 0.0004157071213640923,
"loss": 2.8358,
"theoretical_loss": 3.850829701970037,
"tokens_seen": 583925760
},
{
"epoch": 0.18,
"learning_rate": 0.0004156970912738215,
"loss": 2.943,
"theoretical_loss": 3.85078439200794,
"tokens_seen": 583991296
},
{
"epoch": 0.18,
"learning_rate": 0.00041568706118355066,
"loss": 2.6781,
"theoretical_loss": 3.85073908855382,
"tokens_seen": 584056832
},
{
"epoch": 0.18,
"learning_rate": 0.00041567703109327984,
"loss": 2.738,
"theoretical_loss": 3.8506937916060133,
"tokens_seen": 584122368
},
{
"epoch": 0.18,
"learning_rate": 0.000415667001003009,
"loss": 2.8833,
"theoretical_loss": 3.8506485011628544,
"tokens_seen": 584187904
},
{
"epoch": 0.18,
"learning_rate": 0.00041565697091273826,
"loss": 3.0191,
"theoretical_loss": 3.8506032172226803,
"tokens_seen": 584253440
},
{
"epoch": 0.18,
"learning_rate": 0.0004156469408224674,
"loss": 2.5674,
"theoretical_loss": 3.850557939783828,
"tokens_seen": 584318976
},
{
"epoch": 0.18,
"learning_rate": 0.0004156369107321966,
"loss": 2.6857,
"theoretical_loss": 3.850512668844635,
"tokens_seen": 584384512
},
{
"epoch": 0.18,
"learning_rate": 0.00041562688064192575,
"loss": 2.6629,
"theoretical_loss": 3.8504674044034393,
"tokens_seen": 584450048
},
{
"epoch": 0.18,
"learning_rate": 0.000415616850551655,
"loss": 2.6449,
"theoretical_loss": 3.8504221464585795,
"tokens_seen": 584515584
},
{
"epoch": 0.18,
"learning_rate": 0.00041560682046138417,
"loss": 2.8412,
"theoretical_loss": 3.850376895008396,
"tokens_seen": 584581120
},
{
"epoch": 0.18,
"learning_rate": 0.00041559679037111335,
"loss": 3.0066,
"theoretical_loss": 3.8503316500512286,
"tokens_seen": 584646656
},
{
"epoch": 0.18,
"learning_rate": 0.00041558676028084253,
"loss": 2.8835,
"theoretical_loss": 3.850286411585417,
"tokens_seen": 584712192
},
{
"epoch": 0.18,
"learning_rate": 0.00041557673019057176,
"loss": 2.687,
"theoretical_loss": 3.8502411796093035,
"tokens_seen": 584777728
},
{
"epoch": 0.18,
"learning_rate": 0.0004155667001003009,
"loss": 2.865,
"theoretical_loss": 3.8501959541212294,
"tokens_seen": 584843264
},
{
"epoch": 0.18,
"objective/train/docs_used": 341398,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5760111808776855,
"objective/train/theoretical_loss": 3.850150735119537,
"objective/train/tokens_used": 605368800,
"theoretical_loss": 3.850150735119537,
"tokens_seen": 584908800
},
{
"epoch": 0.18,
"learning_rate": 0.0004155566700100301,
"loss": 2.7967,
"theoretical_loss": 3.850150735119537,
"tokens_seen": 584908800
},
{
"epoch": 0.18,
"learning_rate": 0.00041554663991975925,
"loss": 2.7677,
"theoretical_loss": 3.8501055226025707,
"tokens_seen": 584974336
},
{
"epoch": 0.18,
"learning_rate": 0.0004155366098294885,
"loss": 2.9202,
"theoretical_loss": 3.8500603165686726,
"tokens_seen": 585039872
},
{
"epoch": 0.18,
"learning_rate": 0.00041552657973921767,
"loss": 2.9313,
"theoretical_loss": 3.850015117016188,
"tokens_seen": 585105408
},
{
"epoch": 0.18,
"learning_rate": 0.00041551654964894685,
"loss": 2.7369,
"theoretical_loss": 3.849969923943461,
"tokens_seen": 585170944
},
{
"epoch": 0.18,
"learning_rate": 0.00041550651955867603,
"loss": 2.5288,
"theoretical_loss": 3.8499247373488377,
"tokens_seen": 585236480
},
{
"epoch": 0.18,
"learning_rate": 0.0004154964894684052,
"loss": 2.936,
"theoretical_loss": 3.8498795572306643,
"tokens_seen": 585302016
},
{
"epoch": 0.18,
"learning_rate": 0.0004154864593781344,
"loss": 2.9509,
"theoretical_loss": 3.8498343835872872,
"tokens_seen": 585367552
},
{
"epoch": 0.18,
"learning_rate": 0.00041547642928786363,
"loss": 2.8473,
"theoretical_loss": 3.8497892164170544,
"tokens_seen": 585433088
},
{
"epoch": 0.18,
"learning_rate": 0.00041546639919759276,
"loss": 2.6784,
"theoretical_loss": 3.8497440557183134,
"tokens_seen": 585498624
},
{
"epoch": 0.18,
"learning_rate": 0.000415456369107322,
"loss": 2.5299,
"theoretical_loss": 3.849698901489412,
"tokens_seen": 585564160
},
{
"epoch": 0.18,
"learning_rate": 0.0004154463390170511,
"loss": 2.8405,
"theoretical_loss": 3.849653753728701,
"tokens_seen": 585629696
},
{
"epoch": 0.18,
"learning_rate": 0.00041543630892678035,
"loss": 2.9255,
"theoretical_loss": 3.849608612434529,
"tokens_seen": 585695232
},
{
"epoch": 0.18,
"learning_rate": 0.00041542627883650954,
"loss": 2.8259,
"theoretical_loss": 3.8495634776052463,
"tokens_seen": 585760768
},
{
"epoch": 0.18,
"learning_rate": 0.0004154162487462387,
"loss": 2.873,
"theoretical_loss": 3.849518349239205,
"tokens_seen": 585826304
},
{
"epoch": 0.18,
"learning_rate": 0.0004154062186559679,
"loss": 3.1558,
"theoretical_loss": 3.849473227334755,
"tokens_seen": 585891840
},
{
"epoch": 0.18,
"learning_rate": 0.00041539618856569713,
"loss": 2.99,
"theoretical_loss": 3.8494281118902505,
"tokens_seen": 585957376
},
{
"epoch": 0.18,
"learning_rate": 0.00041538615847542626,
"loss": 2.9531,
"theoretical_loss": 3.849383002904043,
"tokens_seen": 586022912
},
{
"epoch": 0.18,
"learning_rate": 0.0004153761283851555,
"loss": 2.7522,
"theoretical_loss": 3.8493379003744854,
"tokens_seen": 586088448
},
{
"epoch": 0.18,
"learning_rate": 0.0004153660982948846,
"loss": 2.6776,
"theoretical_loss": 3.8492928042999335,
"tokens_seen": 586153984
},
{
"epoch": 0.18,
"learning_rate": 0.00041535606820461386,
"loss": 2.7751,
"theoretical_loss": 3.8492477146787403,
"tokens_seen": 586219520
},
{
"epoch": 0.18,
"learning_rate": 0.00041534603811434304,
"loss": 2.825,
"theoretical_loss": 3.849202631509262,
"tokens_seen": 586285056
},
{
"epoch": 0.18,
"learning_rate": 0.0004153360080240722,
"loss": 2.8614,
"theoretical_loss": 3.849157554789853,
"tokens_seen": 586350592
},
{
"epoch": 0.18,
"learning_rate": 0.0004153259779338014,
"loss": 2.9808,
"theoretical_loss": 3.8491124845188716,
"tokens_seen": 586416128
},
{
"epoch": 0.18,
"learning_rate": 0.0004153159478435306,
"loss": 2.8579,
"theoretical_loss": 3.8490674206946736,
"tokens_seen": 586481664
},
{
"epoch": 0.18,
"objective/train/docs_used": 342071,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.825225353240967,
"objective/train/theoretical_loss": 3.8490223633156173,
"objective/train/tokens_used": 607007200,
"theoretical_loss": 3.8490223633156173,
"tokens_seen": 586547200
},
{
"epoch": 0.18,
"learning_rate": 0.00041530591775325976,
"loss": 2.7571,
"theoretical_loss": 3.8490223633156173,
"tokens_seen": 586547200
},
{
"epoch": 0.18,
"learning_rate": 0.000415295887662989,
"loss": 2.7614,
"theoretical_loss": 3.84897731238006,
"tokens_seen": 586612736
},
{
"epoch": 0.18,
"learning_rate": 0.0004152858575727181,
"loss": 2.8002,
"theoretical_loss": 3.8489322678863624,
"tokens_seen": 586678272
},
{
"epoch": 0.18,
"learning_rate": 0.00041527582748244736,
"loss": 2.8835,
"theoretical_loss": 3.8488872298328816,
"tokens_seen": 586743808
},
{
"epoch": 0.18,
"learning_rate": 0.0004152657973921765,
"loss": 2.8548,
"theoretical_loss": 3.848842198217979,
"tokens_seen": 586809344
},
{
"epoch": 0.18,
"learning_rate": 0.0004152557673019057,
"loss": 2.9912,
"theoretical_loss": 3.8487971730400146,
"tokens_seen": 586874880
},
{
"epoch": 0.18,
"learning_rate": 0.0004152457372116349,
"loss": 2.8354,
"theoretical_loss": 3.84875215429735,
"tokens_seen": 586940416
},
{
"epoch": 0.18,
"learning_rate": 0.0004152357071213641,
"loss": 2.9608,
"theoretical_loss": 3.8487071419883474,
"tokens_seen": 587005952
},
{
"epoch": 0.18,
"learning_rate": 0.00041522567703109327,
"loss": 2.9313,
"theoretical_loss": 3.8486621361113684,
"tokens_seen": 587071488
},
{
"epoch": 0.18,
"learning_rate": 0.0004152156469408225,
"loss": 2.9621,
"theoretical_loss": 3.848617136664777,
"tokens_seen": 587137024
},
{
"epoch": 0.18,
"learning_rate": 0.00041520561685055163,
"loss": 2.762,
"theoretical_loss": 3.848572143646936,
"tokens_seen": 587202560
},
{
"epoch": 0.18,
"learning_rate": 0.00041519558676028086,
"loss": 2.8418,
"theoretical_loss": 3.8485271570562096,
"tokens_seen": 587268096
},
{
"epoch": 0.18,
"learning_rate": 0.00041518555667001,
"loss": 2.9505,
"theoretical_loss": 3.8484821768909634,
"tokens_seen": 587333632
},
{
"epoch": 0.18,
"learning_rate": 0.0004151755265797392,
"loss": 2.7378,
"theoretical_loss": 3.848437203149562,
"tokens_seen": 587399168
},
{
"epoch": 0.18,
"learning_rate": 0.0004151654964894684,
"loss": 2.7264,
"theoretical_loss": 3.8483922358303717,
"tokens_seen": 587464704
},
{
"epoch": 0.18,
"learning_rate": 0.0004151554663991976,
"loss": 2.6414,
"theoretical_loss": 3.8483472749317595,
"tokens_seen": 587530240
},
{
"epoch": 0.18,
"learning_rate": 0.0004151454363089268,
"loss": 2.7191,
"theoretical_loss": 3.8483023204520928,
"tokens_seen": 587595776
},
{
"epoch": 0.18,
"learning_rate": 0.00041513540621865595,
"loss": 2.5959,
"theoretical_loss": 3.848257372389738,
"tokens_seen": 587661312
},
{
"epoch": 0.18,
"learning_rate": 0.0004151253761283852,
"loss": 2.9658,
"theoretical_loss": 3.848212430743065,
"tokens_seen": 587726848
},
{
"epoch": 0.18,
"learning_rate": 0.00041511534603811437,
"loss": 2.544,
"theoretical_loss": 3.8481674955104426,
"tokens_seen": 587792384
},
{
"epoch": 0.18,
"learning_rate": 0.00041510531594784355,
"loss": 2.8289,
"theoretical_loss": 3.848122566690239,
"tokens_seen": 587857920
},
{
"epoch": 0.18,
"learning_rate": 0.00041509528585757273,
"loss": 2.7469,
"theoretical_loss": 3.8480776442808264,
"tokens_seen": 587923456
},
{
"epoch": 0.18,
"learning_rate": 0.00041508525576730196,
"loss": 2.9854,
"theoretical_loss": 3.8480327282805744,
"tokens_seen": 587988992
},
{
"epoch": 0.18,
"learning_rate": 0.0004150752256770311,
"loss": 3.1744,
"theoretical_loss": 3.8479878186878547,
"tokens_seen": 588054528
},
{
"epoch": 0.18,
"learning_rate": 0.0004150651955867603,
"loss": 3.0129,
"theoretical_loss": 3.847942915501039,
"tokens_seen": 588120064
},
{
"epoch": 0.18,
"objective/train/docs_used": 342650,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.821354866027832,
"objective/train/theoretical_loss": 3.8478980187185003,
"objective/train/tokens_used": 608645600,
"theoretical_loss": 3.8478980187185003,
"tokens_seen": 588185600
},
{
"epoch": 0.18,
"learning_rate": 0.00041505516549648945,
"loss": 2.734,
"theoretical_loss": 3.8478980187185003,
"tokens_seen": 588185600
},
{
"epoch": 0.18,
"learning_rate": 0.0004150451354062187,
"loss": 2.91,
"theoretical_loss": 3.8478531283386124,
"tokens_seen": 588251136
},
{
"epoch": 0.18,
"learning_rate": 0.00041503510531594787,
"loss": 2.9527,
"theoretical_loss": 3.847808244359747,
"tokens_seen": 588316672
},
{
"epoch": 0.18,
"learning_rate": 0.00041502507522567705,
"loss": 2.9137,
"theoretical_loss": 3.847763366780281,
"tokens_seen": 588382208
},
{
"epoch": 0.18,
"learning_rate": 0.00041501504513540623,
"loss": 2.8705,
"theoretical_loss": 3.8477184955985875,
"tokens_seen": 588447744
},
{
"epoch": 0.18,
"learning_rate": 0.0004150050150451354,
"loss": 2.7977,
"theoretical_loss": 3.8476736308130426,
"tokens_seen": 588513280
},
{
"epoch": 0.18,
"learning_rate": 0.0004149949849548646,
"loss": 2.9383,
"theoretical_loss": 3.8476287724220226,
"tokens_seen": 588578816
},
{
"epoch": 0.18,
"learning_rate": 0.00041498495486459383,
"loss": 2.7924,
"theoretical_loss": 3.8475839204239044,
"tokens_seen": 588644352
},
{
"epoch": 0.18,
"learning_rate": 0.00041497492477432296,
"loss": 2.854,
"theoretical_loss": 3.847539074817065,
"tokens_seen": 588709888
},
{
"epoch": 0.18,
"learning_rate": 0.0004149648946840522,
"loss": 2.9575,
"theoretical_loss": 3.8474942355998825,
"tokens_seen": 588775424
},
{
"epoch": 0.18,
"learning_rate": 0.0004149548645937813,
"loss": 2.7192,
"theoretical_loss": 3.847449402770735,
"tokens_seen": 588840960
},
{
"epoch": 0.18,
"learning_rate": 0.00041494483450351055,
"loss": 2.7516,
"theoretical_loss": 3.8474045763280023,
"tokens_seen": 588906496
},
{
"epoch": 0.18,
"learning_rate": 0.00041493480441323974,
"loss": 2.9051,
"theoretical_loss": 3.847359756270064,
"tokens_seen": 588972032
},
{
"epoch": 0.18,
"learning_rate": 0.0004149247743229689,
"loss": 2.822,
"theoretical_loss": 3.8473149425953,
"tokens_seen": 589037568
},
{
"epoch": 0.18,
"learning_rate": 0.0004149147442326981,
"loss": 2.9753,
"theoretical_loss": 3.8472701353020913,
"tokens_seen": 589103104
},
{
"epoch": 0.18,
"learning_rate": 0.00041490471414242733,
"loss": 2.5136,
"theoretical_loss": 3.8472253343888196,
"tokens_seen": 589168640
},
{
"epoch": 0.18,
"learning_rate": 0.00041489468405215646,
"loss": 2.9819,
"theoretical_loss": 3.8471805398538663,
"tokens_seen": 589234176
},
{
"epoch": 0.18,
"learning_rate": 0.0004148846539618857,
"loss": 3.0167,
"theoretical_loss": 3.8471357516956153,
"tokens_seen": 589299712
},
{
"epoch": 0.18,
"learning_rate": 0.0004148746238716148,
"loss": 3.0563,
"theoretical_loss": 3.8470909699124487,
"tokens_seen": 589365248
},
{
"epoch": 0.18,
"learning_rate": 0.00041486459378134406,
"loss": 2.8984,
"theoretical_loss": 3.8470461945027505,
"tokens_seen": 589430784
},
{
"epoch": 0.18,
"learning_rate": 0.00041485456369107324,
"loss": 2.9522,
"theoretical_loss": 3.8470014254649056,
"tokens_seen": 589496320
},
{
"epoch": 0.18,
"learning_rate": 0.0004148445336008024,
"loss": 2.7076,
"theoretical_loss": 3.8469566627972993,
"tokens_seen": 589561856
},
{
"epoch": 0.18,
"learning_rate": 0.0004148345035105316,
"loss": 2.6847,
"theoretical_loss": 3.846911906498316,
"tokens_seen": 589627392
},
{
"epoch": 0.18,
"learning_rate": 0.0004148244734202608,
"loss": 2.6315,
"theoretical_loss": 3.8468671565663426,
"tokens_seen": 589692928
},
{
"epoch": 0.18,
"learning_rate": 0.00041481444332998996,
"loss": 2.9967,
"theoretical_loss": 3.8468224129997663,
"tokens_seen": 589758464
},
{
"epoch": 0.18,
"objective/train/docs_used": 343935,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.133723497390747,
"objective/train/theoretical_loss": 3.846777675796974,
"objective/train/tokens_used": 610284000,
"theoretical_loss": 3.846777675796974,
"tokens_seen": 589824000
},
{
"epoch": 0.18,
"learning_rate": 0.0004148044132397192,
"loss": 2.9625,
"theoretical_loss": 3.846777675796974,
"tokens_seen": 589824000
},
{
"epoch": 0.18,
"learning_rate": 0.0004147943831494483,
"loss": 2.7877,
"theoretical_loss": 3.8467329449563534,
"tokens_seen": 589889536
},
{
"epoch": 0.18,
"learning_rate": 0.00041478435305917756,
"loss": 2.6622,
"theoretical_loss": 3.8466882204762936,
"tokens_seen": 589955072
},
{
"epoch": 0.18,
"learning_rate": 0.0004147743229689067,
"loss": 2.7714,
"theoretical_loss": 3.846643502355183,
"tokens_seen": 590020608
},
{
"epoch": 0.18,
"learning_rate": 0.0004147642928786359,
"loss": 2.9137,
"theoretical_loss": 3.846598790591412,
"tokens_seen": 590086144
},
{
"epoch": 0.18,
"learning_rate": 0.0004147542627883651,
"loss": 3.0009,
"theoretical_loss": 3.846554085183371,
"tokens_seen": 590151680
},
{
"epoch": 0.18,
"learning_rate": 0.0004147442326980943,
"loss": 2.5204,
"theoretical_loss": 3.8465093861294504,
"tokens_seen": 590217216
},
{
"epoch": 0.18,
"learning_rate": 0.00041473420260782347,
"loss": 2.6391,
"theoretical_loss": 3.8464646934280413,
"tokens_seen": 590282752
},
{
"epoch": 0.18,
"learning_rate": 0.0004147241725175527,
"loss": 2.7687,
"theoretical_loss": 3.846420007077537,
"tokens_seen": 590348288
},
{
"epoch": 0.18,
"learning_rate": 0.00041471414242728183,
"loss": 2.841,
"theoretical_loss": 3.8463753270763297,
"tokens_seen": 590413824
},
{
"epoch": 0.18,
"learning_rate": 0.00041470411233701106,
"loss": 2.8472,
"theoretical_loss": 3.846330653422812,
"tokens_seen": 590479360
},
{
"epoch": 0.18,
"learning_rate": 0.0004146940822467402,
"loss": 2.7101,
"theoretical_loss": 3.8462859861153778,
"tokens_seen": 590544896
},
{
"epoch": 0.18,
"learning_rate": 0.0004146840521564694,
"loss": 2.6634,
"theoretical_loss": 3.8462413251524215,
"tokens_seen": 590610432
},
{
"epoch": 0.18,
"learning_rate": 0.0004146740220661986,
"loss": 3.0439,
"theoretical_loss": 3.846196670532339,
"tokens_seen": 590675968
},
{
"epoch": 0.18,
"learning_rate": 0.0004146639919759278,
"loss": 3.0207,
"theoretical_loss": 3.8461520222535253,
"tokens_seen": 590741504
},
{
"epoch": 0.18,
"learning_rate": 0.00041465396188565697,
"loss": 2.7321,
"theoretical_loss": 3.8461073803143764,
"tokens_seen": 590807040
},
{
"epoch": 0.18,
"learning_rate": 0.00041464393179538615,
"loss": 2.7399,
"theoretical_loss": 3.846062744713289,
"tokens_seen": 590872576
},
{
"epoch": 0.18,
"learning_rate": 0.00041463390170511533,
"loss": 2.7948,
"theoretical_loss": 3.8460181154486612,
"tokens_seen": 590938112
},
{
"epoch": 0.18,
"learning_rate": 0.00041462387161484457,
"loss": 2.9129,
"theoretical_loss": 3.8459734925188895,
"tokens_seen": 591003648
},
{
"epoch": 0.18,
"learning_rate": 0.0004146138415245737,
"loss": 2.8917,
"theoretical_loss": 3.8459288759223735,
"tokens_seen": 591069184
},
{
"epoch": 0.18,
"learning_rate": 0.00041460381143430293,
"loss": 2.8784,
"theoretical_loss": 3.8458842656575123,
"tokens_seen": 591134720
},
{
"epoch": 0.18,
"learning_rate": 0.0004145937813440321,
"loss": 2.9389,
"theoretical_loss": 3.845839661722705,
"tokens_seen": 591200256
},
{
"epoch": 0.18,
"learning_rate": 0.0004145837512537613,
"loss": 3.1169,
"theoretical_loss": 3.8457950641163516,
"tokens_seen": 591265792
},
{
"epoch": 0.18,
"learning_rate": 0.0004145737211634905,
"loss": 2.958,
"theoretical_loss": 3.845750472836854,
"tokens_seen": 591331328
},
{
"epoch": 0.18,
"learning_rate": 0.00041456369107321965,
"loss": 2.7063,
"theoretical_loss": 3.845705887882612,
"tokens_seen": 591396864
},
{
"epoch": 0.18,
"objective/train/docs_used": 344744,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.702831983566284,
"objective/train/theoretical_loss": 3.845661309252029,
"objective/train/tokens_used": 611922400,
"theoretical_loss": 3.845661309252029,
"tokens_seen": 591462400
},
{
"epoch": 0.18,
"learning_rate": 0.00041455366098294884,
"loss": 2.7622,
"theoretical_loss": 3.845661309252029,
"tokens_seen": 591462400
},
{
"epoch": 0.18,
"learning_rate": 0.00041454363089267807,
"loss": 2.7417,
"theoretical_loss": 3.8456167369435073,
"tokens_seen": 591527936
},
{
"epoch": 0.18,
"learning_rate": 0.0004145336008024072,
"loss": 2.776,
"theoretical_loss": 3.8455721709554496,
"tokens_seen": 591593472
},
{
"epoch": 0.18,
"learning_rate": 0.00041452357071213643,
"loss": 2.8246,
"theoretical_loss": 3.84552761128626,
"tokens_seen": 591659008
},
{
"epoch": 0.18,
"learning_rate": 0.00041451354062186556,
"loss": 3.0347,
"theoretical_loss": 3.8454830579343424,
"tokens_seen": 591724544
},
{
"epoch": 0.18,
"learning_rate": 0.0004145035105315948,
"loss": 2.9839,
"theoretical_loss": 3.845438510898102,
"tokens_seen": 591790080
},
{
"epoch": 0.18,
"learning_rate": 0.000414493480441324,
"loss": 2.7507,
"theoretical_loss": 3.8453939701759436,
"tokens_seen": 591855616
},
{
"epoch": 0.18,
"learning_rate": 0.00041448345035105316,
"loss": 2.8606,
"theoretical_loss": 3.8453494357662747,
"tokens_seen": 591921152
},
{
"epoch": 0.18,
"learning_rate": 0.00041447342026078234,
"loss": 3.0087,
"theoretical_loss": 3.8453049076675008,
"tokens_seen": 591986688
},
{
"epoch": 0.18,
"learning_rate": 0.0004144633901705115,
"loss": 2.8523,
"theoretical_loss": 3.8452603858780288,
"tokens_seen": 592052224
},
{
"epoch": 0.18,
"learning_rate": 0.0004144533600802407,
"loss": 2.7618,
"theoretical_loss": 3.8452158703962676,
"tokens_seen": 592117760
},
{
"epoch": 0.18,
"learning_rate": 0.00041444332998996994,
"loss": 2.8818,
"theoretical_loss": 3.8451713612206246,
"tokens_seen": 592183296
},
{
"epoch": 0.18,
"learning_rate": 0.00041443329989969906,
"loss": 2.8679,
"theoretical_loss": 3.8451268583495084,
"tokens_seen": 592248832
},
{
"epoch": 0.18,
"learning_rate": 0.0004144232698094283,
"loss": 3.0313,
"theoretical_loss": 3.84508236178133,
"tokens_seen": 592314368
},
{
"epoch": 0.18,
"learning_rate": 0.00041441323971915753,
"loss": 2.8172,
"theoretical_loss": 3.8450378715144984,
"tokens_seen": 592379904
},
{
"epoch": 0.18,
"learning_rate": 0.00041440320962888666,
"loss": 2.6686,
"theoretical_loss": 3.844993387547425,
"tokens_seen": 592445440
},
{
"epoch": 0.18,
"learning_rate": 0.0004143931795386159,
"loss": 2.719,
"theoretical_loss": 3.8449489098785197,
"tokens_seen": 592510976
},
{
"epoch": 0.18,
"learning_rate": 0.000414383149448345,
"loss": 2.9581,
"theoretical_loss": 3.8449044385061955,
"tokens_seen": 592576512
},
{
"epoch": 0.18,
"learning_rate": 0.00041437311935807426,
"loss": 2.6012,
"theoretical_loss": 3.8448599734288647,
"tokens_seen": 592642048
},
{
"epoch": 0.18,
"learning_rate": 0.00041436308926780344,
"loss": 2.8978,
"theoretical_loss": 3.8448155146449396,
"tokens_seen": 592707584
},
{
"epoch": 0.18,
"learning_rate": 0.0004143530591775326,
"loss": 3.0502,
"theoretical_loss": 3.844771062152834,
"tokens_seen": 592773120
},
{
"epoch": 0.18,
"learning_rate": 0.0004143430290872618,
"loss": 2.8521,
"theoretical_loss": 3.8447266159509628,
"tokens_seen": 592838656
},
{
"epoch": 0.18,
"learning_rate": 0.000414332998996991,
"loss": 2.7796,
"theoretical_loss": 3.8446821760377397,
"tokens_seen": 592904192
},
{
"epoch": 0.18,
"learning_rate": 0.00041432296890672016,
"loss": 2.7843,
"theoretical_loss": 3.84463774241158,
"tokens_seen": 592969728
},
{
"epoch": 0.18,
"learning_rate": 0.0004143129388164494,
"loss": 2.385,
"theoretical_loss": 3.8445933150708997,
"tokens_seen": 593035264
},
{
"epoch": 0.18,
"objective/train/docs_used": 345976,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.987703561782837,
"objective/train/theoretical_loss": 3.844548894014116,
"objective/train/tokens_used": 613560800,
"theoretical_loss": 3.844548894014116,
"tokens_seen": 593100800
},
{
"epoch": 0.18,
"learning_rate": 0.0004143029087261785,
"loss": 2.7096,
"theoretical_loss": 3.844548894014116,
"tokens_seen": 593100800
},
{
"epoch": 0.18,
"learning_rate": 0.00041429287863590776,
"loss": 2.7678,
"theoretical_loss": 3.844504479239645,
"tokens_seen": 593166336
},
{
"epoch": 0.18,
"learning_rate": 0.0004142828485456369,
"loss": 2.6941,
"theoretical_loss": 3.844460070745905,
"tokens_seen": 593231872
},
{
"epoch": 0.18,
"learning_rate": 0.0004142728184553661,
"loss": 2.6844,
"theoretical_loss": 3.8444156685313127,
"tokens_seen": 593297408
},
{
"epoch": 0.18,
"learning_rate": 0.0004142627883650953,
"loss": 2.9633,
"theoretical_loss": 3.8443712725942882,
"tokens_seen": 593362944
},
{
"epoch": 0.18,
"learning_rate": 0.0004142527582748245,
"loss": 2.8068,
"theoretical_loss": 3.8443268829332498,
"tokens_seen": 593428480
},
{
"epoch": 0.18,
"learning_rate": 0.00041424272818455367,
"loss": 2.9583,
"theoretical_loss": 3.844282499546619,
"tokens_seen": 593494016
},
{
"epoch": 0.18,
"learning_rate": 0.0004142326980942829,
"loss": 2.9592,
"theoretical_loss": 3.844238122432814,
"tokens_seen": 593559552
},
{
"epoch": 0.18,
"learning_rate": 0.00041422266800401203,
"loss": 2.963,
"theoretical_loss": 3.8441937515902573,
"tokens_seen": 593625088
},
{
"epoch": 0.18,
"learning_rate": 0.00041421263791374126,
"loss": 2.7788,
"theoretical_loss": 3.84414938701737,
"tokens_seen": 593690624
},
{
"epoch": 0.18,
"learning_rate": 0.0004142026078234704,
"loss": 2.807,
"theoretical_loss": 3.8441050287125744,
"tokens_seen": 593756160
},
{
"epoch": 0.18,
"learning_rate": 0.0004141925777331996,
"loss": 2.706,
"theoretical_loss": 3.8440606766742933,
"tokens_seen": 593821696
},
{
"epoch": 0.18,
"learning_rate": 0.0004141825476429288,
"loss": 2.9724,
"theoretical_loss": 3.844016330900949,
"tokens_seen": 593887232
},
{
"epoch": 0.18,
"learning_rate": 0.000414172517552658,
"loss": 2.7244,
"theoretical_loss": 3.8439719913909673,
"tokens_seen": 593952768
},
{
"epoch": 0.18,
"learning_rate": 0.00041416248746238717,
"loss": 3.0772,
"theoretical_loss": 3.843927658142771,
"tokens_seen": 594018304
},
{
"epoch": 0.18,
"learning_rate": 0.00041415245737211635,
"loss": 2.6304,
"theoretical_loss": 3.8438833311547853,
"tokens_seen": 594083840
},
{
"epoch": 0.18,
"learning_rate": 0.00041414242728184553,
"loss": 3.0168,
"theoretical_loss": 3.843839010425436,
"tokens_seen": 594149376
},
{
"epoch": 0.18,
"learning_rate": 0.00041413239719157477,
"loss": 2.8647,
"theoretical_loss": 3.8437946959531497,
"tokens_seen": 594214912
},
{
"epoch": 0.18,
"learning_rate": 0.0004141223671013039,
"loss": 2.8631,
"theoretical_loss": 3.8437503877363524,
"tokens_seen": 594280448
},
{
"epoch": 0.18,
"learning_rate": 0.00041411233701103313,
"loss": 2.6276,
"theoretical_loss": 3.8437060857734715,
"tokens_seen": 594345984
},
{
"epoch": 0.18,
"learning_rate": 0.0004141023069207623,
"loss": 3.0555,
"theoretical_loss": 3.843661790062935,
"tokens_seen": 594411520
},
{
"epoch": 0.18,
"learning_rate": 0.0004140922768304915,
"loss": 2.9096,
"theoretical_loss": 3.8436175006031714,
"tokens_seen": 594477056
},
{
"epoch": 0.18,
"learning_rate": 0.0004140822467402207,
"loss": 2.7759,
"theoretical_loss": 3.84357321739261,
"tokens_seen": 594542592
},
{
"epoch": 0.18,
"learning_rate": 0.00041407221664994985,
"loss": 2.7858,
"theoretical_loss": 3.843528940429679,
"tokens_seen": 594608128
},
{
"epoch": 0.18,
"learning_rate": 0.00041406218655967904,
"loss": 3.0237,
"theoretical_loss": 3.8434846697128098,
"tokens_seen": 594673664
},
{
"epoch": 0.18,
"objective/train/docs_used": 346603,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.868394136428833,
"objective/train/theoretical_loss": 3.843440405240433,
"objective/train/tokens_used": 615199200,
"theoretical_loss": 3.843440405240433,
"tokens_seen": 594739200
},
{
"epoch": 0.18,
"learning_rate": 0.00041405215646940827,
"loss": 2.7638,
"theoretical_loss": 3.843440405240433,
"tokens_seen": 594739200
},
{
"epoch": 0.18,
"learning_rate": 0.0004140421263791374,
"loss": 2.8275,
"theoretical_loss": 3.843396147010979,
"tokens_seen": 594804736
},
{
"epoch": 0.18,
"learning_rate": 0.00041403209628886663,
"loss": 3.0026,
"theoretical_loss": 3.843351895022881,
"tokens_seen": 594870272
},
{
"epoch": 0.18,
"learning_rate": 0.00041402206619859576,
"loss": 2.7582,
"theoretical_loss": 3.8433076492745695,
"tokens_seen": 594935808
},
{
"epoch": 0.18,
"learning_rate": 0.000414012036108325,
"loss": 2.8949,
"theoretical_loss": 3.8432634097644787,
"tokens_seen": 595001344
},
{
"epoch": 0.18,
"learning_rate": 0.0004140020060180542,
"loss": 2.8077,
"theoretical_loss": 3.8432191764910426,
"tokens_seen": 595066880
},
{
"epoch": 0.18,
"learning_rate": 0.00041399197592778336,
"loss": 2.7644,
"theoretical_loss": 3.8431749494526937,
"tokens_seen": 595132416
},
{
"epoch": 0.18,
"learning_rate": 0.00041398194583751254,
"loss": 2.7257,
"theoretical_loss": 3.8431307286478678,
"tokens_seen": 595197952
},
{
"epoch": 0.18,
"learning_rate": 0.0004139719157472417,
"loss": 2.8675,
"theoretical_loss": 3.843086514075,
"tokens_seen": 595263488
},
{
"epoch": 0.18,
"learning_rate": 0.0004139618856569709,
"loss": 2.8906,
"theoretical_loss": 3.843042305732526,
"tokens_seen": 595329024
},
{
"epoch": 0.18,
"learning_rate": 0.00041395185556670014,
"loss": 2.7717,
"theoretical_loss": 3.8429981036188816,
"tokens_seen": 595394560
},
{
"epoch": 0.18,
"learning_rate": 0.00041394182547642926,
"loss": 2.8548,
"theoretical_loss": 3.8429539077325043,
"tokens_seen": 595460096
},
{
"epoch": 0.18,
"learning_rate": 0.0004139317953861585,
"loss": 2.9354,
"theoretical_loss": 3.8429097180718315,
"tokens_seen": 595525632
},
{
"epoch": 0.18,
"learning_rate": 0.0004139217652958877,
"loss": 2.9645,
"theoretical_loss": 3.842865534635301,
"tokens_seen": 595591168
},
{
"epoch": 0.18,
"learning_rate": 0.00041391173520561686,
"loss": 2.7224,
"theoretical_loss": 3.842821357421352,
"tokens_seen": 595656704
},
{
"epoch": 0.18,
"learning_rate": 0.00041390170511534604,
"loss": 2.9284,
"theoretical_loss": 3.842777186428423,
"tokens_seen": 595722240
},
{
"epoch": 0.18,
"learning_rate": 0.0004138916750250752,
"loss": 2.5825,
"theoretical_loss": 3.8427330216549533,
"tokens_seen": 595787776
},
{
"epoch": 0.18,
"learning_rate": 0.0004138816449348044,
"loss": 2.8972,
"theoretical_loss": 3.8426888630993847,
"tokens_seen": 595853312
},
{
"epoch": 0.18,
"learning_rate": 0.00041387161484453364,
"loss": 2.4536,
"theoretical_loss": 3.8426447107601565,
"tokens_seen": 595918848
},
{
"epoch": 0.18,
"learning_rate": 0.00041386158475426277,
"loss": 3.055,
"theoretical_loss": 3.8426005646357106,
"tokens_seen": 595984384
},
{
"epoch": 0.18,
"learning_rate": 0.000413851554663992,
"loss": 2.8051,
"theoretical_loss": 3.8425564247244894,
"tokens_seen": 596049920
},
{
"epoch": 0.18,
"learning_rate": 0.00041384152457372113,
"loss": 2.9206,
"theoretical_loss": 3.842512291024935,
"tokens_seen": 596115456
},
{
"epoch": 0.18,
"learning_rate": 0.00041383149448345036,
"loss": 2.9248,
"theoretical_loss": 3.842468163535491,
"tokens_seen": 596180992
},
{
"epoch": 0.18,
"learning_rate": 0.00041382146439317954,
"loss": 2.8616,
"theoretical_loss": 3.842424042254601,
"tokens_seen": 596246528
},
{
"epoch": 0.18,
"learning_rate": 0.0004138114343029087,
"loss": 2.8664,
"theoretical_loss": 3.8423799271807084,
"tokens_seen": 596312064
},
{
"epoch": 0.18,
"objective/train/docs_used": 348035,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6039180755615234,
"objective/train/theoretical_loss": 3.8423358183122582,
"objective/train/tokens_used": 616837600,
"theoretical_loss": 3.8423358183122582,
"tokens_seen": 596377600
},
{
"epoch": 0.18,
"learning_rate": 0.0004138014042126379,
"loss": 2.8114,
"theoretical_loss": 3.8423358183122582,
"tokens_seen": 596377600
},
{
"epoch": 0.18,
"learning_rate": 0.0004137913741223671,
"loss": 2.8918,
"theoretical_loss": 3.8422917156476966,
"tokens_seen": 596443136
},
{
"epoch": 0.18,
"learning_rate": 0.00041378134403209627,
"loss": 2.9722,
"theoretical_loss": 3.8422476191854686,
"tokens_seen": 596508672
},
{
"epoch": 0.18,
"learning_rate": 0.0004137713139418255,
"loss": 2.757,
"theoretical_loss": 3.8422035289240215,
"tokens_seen": 596574208
},
{
"epoch": 0.18,
"learning_rate": 0.00041376128385155463,
"loss": 2.6994,
"theoretical_loss": 3.8421594448618013,
"tokens_seen": 596639744
},
{
"epoch": 0.18,
"learning_rate": 0.00041375125376128387,
"loss": 2.7965,
"theoretical_loss": 3.8421153669972563,
"tokens_seen": 596705280
},
{
"epoch": 0.18,
"learning_rate": 0.00041374122367101305,
"loss": 2.6647,
"theoretical_loss": 3.842071295328835,
"tokens_seen": 596770816
},
{
"epoch": 0.18,
"learning_rate": 0.00041373119358074223,
"loss": 2.7225,
"theoretical_loss": 3.8420272298549847,
"tokens_seen": 596836352
},
{
"epoch": 0.18,
"learning_rate": 0.0004137211634904714,
"loss": 2.8572,
"theoretical_loss": 3.8419831705741556,
"tokens_seen": 596901888
},
{
"epoch": 0.18,
"learning_rate": 0.0004137111334002006,
"loss": 2.7221,
"theoretical_loss": 3.841939117484798,
"tokens_seen": 596967424
},
{
"epoch": 0.18,
"learning_rate": 0.00041370110330992977,
"loss": 2.7622,
"theoretical_loss": 3.841895070585361,
"tokens_seen": 597032960
},
{
"epoch": 0.18,
"learning_rate": 0.000413691073219659,
"loss": 2.8918,
"theoretical_loss": 3.8418510298742965,
"tokens_seen": 597098496
},
{
"epoch": 0.18,
"learning_rate": 0.00041368104312938813,
"loss": 2.869,
"theoretical_loss": 3.8418069953500558,
"tokens_seen": 597164032
},
{
"epoch": 0.18,
"learning_rate": 0.00041367101303911737,
"loss": 2.8955,
"theoretical_loss": 3.8417629670110904,
"tokens_seen": 597229568
},
{
"epoch": 0.18,
"learning_rate": 0.00041366098294884655,
"loss": 2.7212,
"theoretical_loss": 3.841718944855854,
"tokens_seen": 597295104
},
{
"epoch": 0.18,
"learning_rate": 0.00041365095285857573,
"loss": 2.7575,
"theoretical_loss": 3.8416749288827985,
"tokens_seen": 597360640
},
{
"epoch": 0.18,
"learning_rate": 0.00041364092276830497,
"loss": 3.0717,
"theoretical_loss": 3.8416309190903783,
"tokens_seen": 597426176
},
{
"epoch": 0.18,
"learning_rate": 0.0004136308926780341,
"loss": 2.8992,
"theoretical_loss": 3.841586915477048,
"tokens_seen": 597491712
},
{
"epoch": 0.18,
"learning_rate": 0.00041362086258776333,
"loss": 2.7046,
"theoretical_loss": 3.8415429180412617,
"tokens_seen": 597557248
},
{
"epoch": 0.18,
"learning_rate": 0.0004136108324974925,
"loss": 2.8873,
"theoretical_loss": 3.841498926781475,
"tokens_seen": 597622784
},
{
"epoch": 0.18,
"learning_rate": 0.0004136008024072217,
"loss": 2.8448,
"theoretical_loss": 3.8414549416961443,
"tokens_seen": 597688320
},
{
"epoch": 0.18,
"learning_rate": 0.0004135907723169509,
"loss": 2.795,
"theoretical_loss": 3.8414109627837254,
"tokens_seen": 597753856
},
{
"epoch": 0.18,
"learning_rate": 0.00041358074222668005,
"loss": 2.8589,
"theoretical_loss": 3.841366990042676,
"tokens_seen": 597819392
},
{
"epoch": 0.18,
"learning_rate": 0.00041357071213640924,
"loss": 2.636,
"theoretical_loss": 3.8413230234714533,
"tokens_seen": 597884928
},
{
"epoch": 0.18,
"learning_rate": 0.00041356068204613847,
"loss": 2.8705,
"theoretical_loss": 3.8412790630685154,
"tokens_seen": 597950464
},
{
"epoch": 0.18,
"objective/train/docs_used": 348629,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.116250514984131,
"objective/train/theoretical_loss": 3.841235108832321,
"objective/train/tokens_used": 618476000,
"theoretical_loss": 3.841235108832321,
"tokens_seen": 598016000
},
{
"epoch": 0.18,
"learning_rate": 0.0004135506519558676,
"loss": 3.0489,
"theoretical_loss": 3.841235108832321,
"tokens_seen": 598016000
},
{
"epoch": 0.18,
"learning_rate": 0.00041354062186559683,
"loss": 2.7793,
"theoretical_loss": 3.84119116076133,
"tokens_seen": 598081536
},
{
"epoch": 0.18,
"learning_rate": 0.00041353059177532596,
"loss": 2.9613,
"theoretical_loss": 3.841147218854002,
"tokens_seen": 598147072
},
{
"epoch": 0.18,
"learning_rate": 0.0004135205616850552,
"loss": 2.6289,
"theoretical_loss": 3.8411032831087963,
"tokens_seen": 598212608
},
{
"epoch": 0.18,
"learning_rate": 0.0004135105315947844,
"loss": 2.8951,
"theoretical_loss": 3.841059353524175,
"tokens_seen": 598278144
},
{
"epoch": 0.18,
"learning_rate": 0.00041350050150451356,
"loss": 2.8064,
"theoretical_loss": 3.841015430098599,
"tokens_seen": 598343680
},
{
"epoch": 0.18,
"learning_rate": 0.00041349047141424274,
"loss": 2.6171,
"theoretical_loss": 3.8409715128305306,
"tokens_seen": 598409216
},
{
"epoch": 0.18,
"learning_rate": 0.0004134804413239719,
"loss": 2.6367,
"theoretical_loss": 3.8409276017184326,
"tokens_seen": 598474752
},
{
"epoch": 0.18,
"learning_rate": 0.0004134704112337011,
"loss": 2.765,
"theoretical_loss": 3.840883696760768,
"tokens_seen": 598540288
},
{
"epoch": 0.18,
"learning_rate": 0.00041346038114343034,
"loss": 2.6495,
"theoretical_loss": 3.840839797956,
"tokens_seen": 598605824
},
{
"epoch": 0.18,
"learning_rate": 0.00041345035105315946,
"loss": 2.6851,
"theoretical_loss": 3.8407959053025937,
"tokens_seen": 598671360
},
{
"epoch": 0.18,
"learning_rate": 0.0004134403209628887,
"loss": 2.8062,
"theoretical_loss": 3.8407520187990127,
"tokens_seen": 598736896
},
{
"epoch": 0.18,
"learning_rate": 0.0004134302908726179,
"loss": 2.7218,
"theoretical_loss": 3.8407081384437234,
"tokens_seen": 598802432
},
{
"epoch": 0.18,
"learning_rate": 0.00041342026078234706,
"loss": 2.7344,
"theoretical_loss": 3.8406642642351914,
"tokens_seen": 598867968
},
{
"epoch": 0.18,
"learning_rate": 0.00041341023069207624,
"loss": 2.5827,
"theoretical_loss": 3.840620396171883,
"tokens_seen": 598933504
},
{
"epoch": 0.18,
"learning_rate": 0.0004134002006018054,
"loss": 3.0092,
"theoretical_loss": 3.840576534252265,
"tokens_seen": 598999040
},
{
"epoch": 0.18,
"learning_rate": 0.0004133901705115346,
"loss": 2.8179,
"theoretical_loss": 3.8405326784748057,
"tokens_seen": 599064576
},
{
"epoch": 0.18,
"learning_rate": 0.00041338014042126384,
"loss": 2.8732,
"theoretical_loss": 3.840488828837973,
"tokens_seen": 599130112
},
{
"epoch": 0.18,
"learning_rate": 0.00041337011033099297,
"loss": 2.9042,
"theoretical_loss": 3.840444985340235,
"tokens_seen": 599195648
},
{
"epoch": 0.18,
"learning_rate": 0.0004133600802407222,
"loss": 2.9798,
"theoretical_loss": 3.840401147980061,
"tokens_seen": 599261184
},
{
"epoch": 0.18,
"learning_rate": 0.00041335005015045133,
"loss": 2.5213,
"theoretical_loss": 3.840357316755921,
"tokens_seen": 599326720
},
{
"epoch": 0.18,
"learning_rate": 0.00041334002006018056,
"loss": 3.0255,
"theoretical_loss": 3.8403134916662847,
"tokens_seen": 599392256
},
{
"epoch": 0.18,
"learning_rate": 0.00041332998996990975,
"loss": 2.6682,
"theoretical_loss": 3.840269672709624,
"tokens_seen": 599457792
},
{
"epoch": 0.18,
"learning_rate": 0.0004133199598796389,
"loss": 2.9696,
"theoretical_loss": 3.8402258598844097,
"tokens_seen": 599523328
},
{
"epoch": 0.18,
"learning_rate": 0.0004133099297893681,
"loss": 2.8461,
"theoretical_loss": 3.8401820531891135,
"tokens_seen": 599588864
},
{
"epoch": 0.18,
"objective/train/docs_used": 349996,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6275978088378906,
"objective/train/theoretical_loss": 3.840138252622208,
"objective/train/tokens_used": 620114400,
"theoretical_loss": 3.840138252622208,
"tokens_seen": 599654400
},
{
"epoch": 0.18,
"learning_rate": 0.0004132998996990973,
"loss": 2.7282,
"theoretical_loss": 3.840138252622208,
"tokens_seen": 599654400
},
{
"epoch": 0.18,
"learning_rate": 0.00041328986960882647,
"loss": 2.7441,
"theoretical_loss": 3.8400944581821674,
"tokens_seen": 599719936
},
{
"epoch": 0.18,
"learning_rate": 0.0004132798395185557,
"loss": 3.025,
"theoretical_loss": 3.8400506698674635,
"tokens_seen": 599785472
},
{
"epoch": 0.18,
"learning_rate": 0.00041326980942828483,
"loss": 2.7853,
"theoretical_loss": 3.840006887676571,
"tokens_seen": 599851008
},
{
"epoch": 0.18,
"learning_rate": 0.00041325977933801407,
"loss": 2.9446,
"theoretical_loss": 3.839963111607965,
"tokens_seen": 599916544
},
{
"epoch": 0.18,
"learning_rate": 0.00041324974924774325,
"loss": 2.9363,
"theoretical_loss": 3.8399193416601207,
"tokens_seen": 599982080
},
{
"epoch": 0.18,
"learning_rate": 0.00041323971915747243,
"loss": 2.6701,
"theoretical_loss": 3.8398755778315135,
"tokens_seen": 600047616
},
{
"epoch": 0.18,
"learning_rate": 0.0004132296890672016,
"loss": 2.695,
"theoretical_loss": 3.8398318201206196,
"tokens_seen": 600113152
},
{
"epoch": 0.18,
"learning_rate": 0.0004132196589769308,
"loss": 2.9478,
"theoretical_loss": 3.8397880685259165,
"tokens_seen": 600178688
},
{
"epoch": 0.18,
"learning_rate": 0.00041320962888665997,
"loss": 2.9884,
"theoretical_loss": 3.839744323045881,
"tokens_seen": 600244224
},
{
"epoch": 0.18,
"learning_rate": 0.0004131995987963892,
"loss": 2.556,
"theoretical_loss": 3.839700583678991,
"tokens_seen": 600309760
},
{
"epoch": 0.18,
"learning_rate": 0.00041318956870611833,
"loss": 2.7496,
"theoretical_loss": 3.839656850423726,
"tokens_seen": 600375296
},
{
"epoch": 0.18,
"learning_rate": 0.00041317953861584757,
"loss": 2.791,
"theoretical_loss": 3.8396131232785637,
"tokens_seen": 600440832
},
{
"epoch": 0.18,
"learning_rate": 0.0004131695085255767,
"loss": 2.6448,
"theoretical_loss": 3.8395694022419846,
"tokens_seen": 600506368
},
{
"epoch": 0.18,
"learning_rate": 0.00041315947843530593,
"loss": 3.0441,
"theoretical_loss": 3.8395256873124683,
"tokens_seen": 600571904
},
{
"epoch": 0.18,
"learning_rate": 0.0004131494483450351,
"loss": 2.9385,
"theoretical_loss": 3.8394819784884957,
"tokens_seen": 600637440
},
{
"epoch": 0.18,
"learning_rate": 0.0004131394182547643,
"loss": 2.6955,
"theoretical_loss": 3.8394382757685483,
"tokens_seen": 600702976
},
{
"epoch": 0.18,
"learning_rate": 0.0004131293881644935,
"loss": 2.7299,
"theoretical_loss": 3.839394579151107,
"tokens_seen": 600768512
},
{
"epoch": 0.18,
"learning_rate": 0.0004131193580742227,
"loss": 2.7899,
"theoretical_loss": 3.8393508886346552,
"tokens_seen": 600834048
},
{
"epoch": 0.18,
"learning_rate": 0.00041310932798395184,
"loss": 2.7416,
"theoretical_loss": 3.8393072042176755,
"tokens_seen": 600899584
},
{
"epoch": 0.18,
"learning_rate": 0.0004130992978936811,
"loss": 2.7468,
"theoretical_loss": 3.83926352589865,
"tokens_seen": 600965120
},
{
"epoch": 0.18,
"learning_rate": 0.0004130892678034102,
"loss": 2.5742,
"theoretical_loss": 3.839219853676064,
"tokens_seen": 601030656
},
{
"epoch": 0.18,
"learning_rate": 0.00041307923771313944,
"loss": 2.8156,
"theoretical_loss": 3.839176187548402,
"tokens_seen": 601096192
},
{
"epoch": 0.18,
"learning_rate": 0.0004130692076228686,
"loss": 2.8401,
"theoretical_loss": 3.839132527514148,
"tokens_seen": 601161728
},
{
"epoch": 0.18,
"learning_rate": 0.0004130591775325978,
"loss": 2.7502,
"theoretical_loss": 3.8390888735717885,
"tokens_seen": 601227264
},
{
"epoch": 0.18,
"objective/train/docs_used": 350686,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7006852626800537,
"objective/train/theoretical_loss": 3.8390452257198096,
"objective/train/tokens_used": 621752800,
"theoretical_loss": 3.8390452257198096,
"tokens_seen": 601292800
},
{
"epoch": 0.18,
"learning_rate": 0.000413049147442327,
"loss": 2.7745,
"theoretical_loss": 3.8390452257198096,
"tokens_seen": 601292800
},
{
"epoch": 0.18,
"learning_rate": 0.00041303911735205616,
"loss": 3.0797,
"theoretical_loss": 3.839001583956697,
"tokens_seen": 601358336
},
{
"epoch": 0.18,
"learning_rate": 0.00041302908726178534,
"loss": 3.1748,
"theoretical_loss": 3.838957948280939,
"tokens_seen": 601423872
},
{
"epoch": 0.18,
"learning_rate": 0.0004130190571715146,
"loss": 2.9787,
"theoretical_loss": 3.8389143186910224,
"tokens_seen": 601489408
},
{
"epoch": 0.18,
"learning_rate": 0.0004130090270812437,
"loss": 3.0155,
"theoretical_loss": 3.8388706951854354,
"tokens_seen": 601554944
},
{
"epoch": 0.18,
"learning_rate": 0.00041299899699097294,
"loss": 2.7272,
"theoretical_loss": 3.838827077762668,
"tokens_seen": 601620480
},
{
"epoch": 0.18,
"learning_rate": 0.00041298896690070207,
"loss": 2.6408,
"theoretical_loss": 3.8387834664212086,
"tokens_seen": 601686016
},
{
"epoch": 0.18,
"learning_rate": 0.0004129789368104313,
"loss": 2.8119,
"theoretical_loss": 3.8387398611595476,
"tokens_seen": 601751552
},
{
"epoch": 0.18,
"learning_rate": 0.0004129689067201605,
"loss": 2.8147,
"theoretical_loss": 3.8386962619761746,
"tokens_seen": 601817088
},
{
"epoch": 0.18,
"learning_rate": 0.00041295887662988966,
"loss": 2.6513,
"theoretical_loss": 3.838652668869581,
"tokens_seen": 601882624
},
{
"epoch": 0.18,
"learning_rate": 0.00041294884653961884,
"loss": 2.7751,
"theoretical_loss": 3.8386090818382588,
"tokens_seen": 601948160
},
{
"epoch": 0.18,
"learning_rate": 0.0004129388164493481,
"loss": 2.9408,
"theoretical_loss": 3.838565500880699,
"tokens_seen": 602013696
},
{
"epoch": 0.18,
"learning_rate": 0.0004129287863590772,
"loss": 2.5206,
"theoretical_loss": 3.838521925995395,
"tokens_seen": 602079232
},
{
"epoch": 0.18,
"learning_rate": 0.00041291875626880644,
"loss": 2.9071,
"theoretical_loss": 3.83847835718084,
"tokens_seen": 602144768
},
{
"epoch": 0.18,
"learning_rate": 0.0004129087261785356,
"loss": 2.836,
"theoretical_loss": 3.838434794435527,
"tokens_seen": 602210304
},
{
"epoch": 0.18,
"learning_rate": 0.0004128986960882648,
"loss": 2.432,
"theoretical_loss": 3.8383912377579508,
"tokens_seen": 602275840
},
{
"epoch": 0.18,
"learning_rate": 0.00041288866599799404,
"loss": 2.7358,
"theoretical_loss": 3.838347687146605,
"tokens_seen": 602341376
},
{
"epoch": 0.18,
"learning_rate": 0.00041287863590772317,
"loss": 2.8274,
"theoretical_loss": 3.8383041425999864,
"tokens_seen": 602406912
},
{
"epoch": 0.18,
"learning_rate": 0.0004128686058174524,
"loss": 2.7196,
"theoretical_loss": 3.83826060411659,
"tokens_seen": 602472448
},
{
"epoch": 0.18,
"learning_rate": 0.00041285857572718153,
"loss": 2.6672,
"theoretical_loss": 3.8382170716949124,
"tokens_seen": 602537984
},
{
"epoch": 0.18,
"learning_rate": 0.00041284854563691076,
"loss": 2.8981,
"theoretical_loss": 3.83817354533345,
"tokens_seen": 602603520
},
{
"epoch": 0.18,
"learning_rate": 0.00041283851554663995,
"loss": 2.6895,
"theoretical_loss": 3.8381300250307007,
"tokens_seen": 602669056
},
{
"epoch": 0.18,
"learning_rate": 0.0004128284854563691,
"loss": 2.525,
"theoretical_loss": 3.8380865107851623,
"tokens_seen": 602734592
},
{
"epoch": 0.18,
"learning_rate": 0.0004128184553660983,
"loss": 2.618,
"theoretical_loss": 3.8380430025953327,
"tokens_seen": 602800128
},
{
"epoch": 0.18,
"learning_rate": 0.0004128084252758275,
"loss": 2.5809,
"theoretical_loss": 3.837999500459712,
"tokens_seen": 602865664
},
{
"epoch": 0.18,
"objective/train/docs_used": 352047,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8010265827178955,
"objective/train/theoretical_loss": 3.837956004376799,
"objective/train/tokens_used": 623391200,
"theoretical_loss": 3.837956004376799,
"tokens_seen": 602931200
},
{
"epoch": 0.18,
"learning_rate": 0.00041279839518555667,
"loss": 2.56,
"theoretical_loss": 3.837956004376799,
"tokens_seen": 602931200
},
{
"epoch": 0.18,
"learning_rate": 0.0004127883650952859,
"loss": 2.8099,
"theoretical_loss": 3.837912514345094,
"tokens_seen": 602996736
},
{
"epoch": 0.18,
"learning_rate": 0.00041277833500501503,
"loss": 3.0077,
"theoretical_loss": 3.8378690303630973,
"tokens_seen": 603062272
},
{
"epoch": 0.18,
"learning_rate": 0.00041276830491474427,
"loss": 2.8896,
"theoretical_loss": 3.837825552429311,
"tokens_seen": 603127808
},
{
"epoch": 0.18,
"learning_rate": 0.00041275827482447345,
"loss": 2.7193,
"theoretical_loss": 3.837782080542236,
"tokens_seen": 603193344
},
{
"epoch": 0.18,
"learning_rate": 0.00041274824473420263,
"loss": 2.8501,
"theoretical_loss": 3.8377386147003745,
"tokens_seen": 603258880
},
{
"epoch": 0.18,
"learning_rate": 0.0004127382146439318,
"loss": 2.6467,
"theoretical_loss": 3.837695154902229,
"tokens_seen": 603324416
},
{
"epoch": 0.18,
"learning_rate": 0.000412728184553661,
"loss": 2.787,
"theoretical_loss": 3.837651701146304,
"tokens_seen": 603389952
},
{
"epoch": 0.18,
"learning_rate": 0.00041271815446339017,
"loss": 2.7283,
"theoretical_loss": 3.8376082534311022,
"tokens_seen": 603455488
},
{
"epoch": 0.18,
"learning_rate": 0.0004127081243731194,
"loss": 2.8614,
"theoretical_loss": 3.8375648117551284,
"tokens_seen": 603521024
},
{
"epoch": 0.18,
"learning_rate": 0.00041269809428284854,
"loss": 2.7903,
"theoretical_loss": 3.8375213761168867,
"tokens_seen": 603586560
},
{
"epoch": 0.18,
"learning_rate": 0.00041268806419257777,
"loss": 2.729,
"theoretical_loss": 3.837477946514884,
"tokens_seen": 603652096
},
{
"epoch": 0.18,
"learning_rate": 0.0004126780341023069,
"loss": 2.9156,
"theoretical_loss": 3.8374345229476248,
"tokens_seen": 603717632
},
{
"epoch": 0.18,
"learning_rate": 0.00041266800401203613,
"loss": 2.8972,
"theoretical_loss": 3.8373911054136167,
"tokens_seen": 603783168
},
{
"epoch": 0.18,
"learning_rate": 0.0004126579739217653,
"loss": 2.6882,
"theoretical_loss": 3.837347693911366,
"tokens_seen": 603848704
},
{
"epoch": 0.18,
"learning_rate": 0.0004126479438314945,
"loss": 2.9854,
"theoretical_loss": 3.8373042884393813,
"tokens_seen": 603914240
},
{
"epoch": 0.18,
"learning_rate": 0.0004126379137412237,
"loss": 2.725,
"theoretical_loss": 3.837260888996169,
"tokens_seen": 603979776
},
{
"epoch": 0.18,
"learning_rate": 0.0004126278836509529,
"loss": 2.8998,
"theoretical_loss": 3.837217495580239,
"tokens_seen": 604045312
},
{
"epoch": 0.18,
"learning_rate": 0.00041261785356068204,
"loss": 2.9865,
"theoretical_loss": 3.8371741081900996,
"tokens_seen": 604110848
},
{
"epoch": 0.18,
"learning_rate": 0.0004126078234704113,
"loss": 3.0078,
"theoretical_loss": 3.8371307268242614,
"tokens_seen": 604176384
},
{
"epoch": 0.18,
"learning_rate": 0.0004125977933801404,
"loss": 2.9132,
"theoretical_loss": 3.837087351481234,
"tokens_seen": 604241920
},
{
"epoch": 0.18,
"learning_rate": 0.00041258776328986964,
"loss": 2.8328,
"theoretical_loss": 3.8370439821595284,
"tokens_seen": 604307456
},
{
"epoch": 0.18,
"learning_rate": 0.0004125777331995988,
"loss": 2.5516,
"theoretical_loss": 3.8370006188576555,
"tokens_seen": 604372992
},
{
"epoch": 0.18,
"learning_rate": 0.000412567703109328,
"loss": 2.9134,
"theoretical_loss": 3.8369572615741268,
"tokens_seen": 604438528
},
{
"epoch": 0.18,
"learning_rate": 0.0004125576730190572,
"loss": 2.9092,
"theoretical_loss": 3.836913910307456,
"tokens_seen": 604504064
},
{
"epoch": 0.18,
"objective/train/docs_used": 352465,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7846691608428955,
"objective/train/theoretical_loss": 3.836870565056154,
"objective/train/tokens_used": 625029600,
"theoretical_loss": 3.836870565056154,
"tokens_seen": 604569600
},
{
"epoch": 0.18,
"learning_rate": 0.00041254764292878636,
"loss": 2.7988,
"theoretical_loss": 3.836870565056154,
"tokens_seen": 604569600
},
{
"epoch": 0.18,
"learning_rate": 0.00041253761283851554,
"loss": 2.7453,
"theoretical_loss": 3.8368272258187366,
"tokens_seen": 604635136
},
{
"epoch": 0.18,
"learning_rate": 0.0004125275827482448,
"loss": 2.9183,
"theoretical_loss": 3.836783892593715,
"tokens_seen": 604700672
},
{
"epoch": 0.18,
"learning_rate": 0.0004125175526579739,
"loss": 2.8628,
"theoretical_loss": 3.836740565379606,
"tokens_seen": 604766208
},
{
"epoch": 0.18,
"learning_rate": 0.00041250752256770314,
"loss": 2.7834,
"theoretical_loss": 3.836697244174923,
"tokens_seen": 604831744
},
{
"epoch": 0.18,
"learning_rate": 0.00041249749247743227,
"loss": 2.6896,
"theoretical_loss": 3.836653928978182,
"tokens_seen": 604897280
},
{
"epoch": 0.18,
"learning_rate": 0.0004124874623871615,
"loss": 2.9959,
"theoretical_loss": 3.8366106197878995,
"tokens_seen": 604962816
},
{
"epoch": 0.18,
"learning_rate": 0.0004124774322968907,
"loss": 2.9265,
"theoretical_loss": 3.8365673166025913,
"tokens_seen": 605028352
},
{
"epoch": 0.18,
"learning_rate": 0.00041246740220661986,
"loss": 2.7492,
"theoretical_loss": 3.8365240194207746,
"tokens_seen": 605093888
},
{
"epoch": 0.18,
"learning_rate": 0.00041245737211634904,
"loss": 2.8017,
"theoretical_loss": 3.8364807282409674,
"tokens_seen": 605159424
},
{
"epoch": 0.18,
"learning_rate": 0.0004124473420260783,
"loss": 2.8394,
"theoretical_loss": 3.836437443061688,
"tokens_seen": 605224960
},
{
"epoch": 0.18,
"learning_rate": 0.0004124373119358074,
"loss": 2.9353,
"theoretical_loss": 3.8363941638814536,
"tokens_seen": 605290496
},
{
"epoch": 0.18,
"learning_rate": 0.00041242728184553664,
"loss": 2.5652,
"theoretical_loss": 3.8363508906987853,
"tokens_seen": 605356032
},
{
"epoch": 0.18,
"learning_rate": 0.00041241725175526577,
"loss": 2.7886,
"theoretical_loss": 3.8363076235122016,
"tokens_seen": 605421568
},
{
"epoch": 0.18,
"learning_rate": 0.000412407221664995,
"loss": 2.6518,
"theoretical_loss": 3.8362643623202226,
"tokens_seen": 605487104
},
{
"epoch": 0.18,
"learning_rate": 0.0004123971915747242,
"loss": 2.9449,
"theoretical_loss": 3.83622110712137,
"tokens_seen": 605552640
},
{
"epoch": 0.18,
"learning_rate": 0.00041238716148445337,
"loss": 2.871,
"theoretical_loss": 3.8361778579141648,
"tokens_seen": 605618176
},
{
"epoch": 0.18,
"learning_rate": 0.00041237713139418255,
"loss": 2.8066,
"theoretical_loss": 3.8361346146971282,
"tokens_seen": 605683712
},
{
"epoch": 0.18,
"learning_rate": 0.00041236710130391173,
"loss": 3.2092,
"theoretical_loss": 3.836091377468783,
"tokens_seen": 605749248
},
{
"epoch": 0.18,
"learning_rate": 0.0004123570712136409,
"loss": 2.7747,
"theoretical_loss": 3.836048146227652,
"tokens_seen": 605814784
},
{
"epoch": 0.18,
"learning_rate": 0.00041234704112337015,
"loss": 2.6126,
"theoretical_loss": 3.8360049209722584,
"tokens_seen": 605880320
},
{
"epoch": 0.18,
"learning_rate": 0.00041233701103309927,
"loss": 3.015,
"theoretical_loss": 3.8359617017011267,
"tokens_seen": 605945856
},
{
"epoch": 0.18,
"learning_rate": 0.0004123269809428285,
"loss": 2.5441,
"theoretical_loss": 3.835918488412781,
"tokens_seen": 606011392
},
{
"epoch": 0.18,
"learning_rate": 0.00041231695085255763,
"loss": 2.5169,
"theoretical_loss": 3.8358752811057455,
"tokens_seen": 606076928
},
{
"epoch": 0.18,
"learning_rate": 0.00041230692076228687,
"loss": 2.8302,
"theoretical_loss": 3.8358320797785463,
"tokens_seen": 606142464
},
{
"epoch": 0.18,
"objective/train/docs_used": 353634,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.123135566711426,
"objective/train/theoretical_loss": 3.8357888844297094,
"objective/train/tokens_used": 626668000,
"theoretical_loss": 3.8357888844297094,
"tokens_seen": 606208000
},
{
"epoch": 0.18,
"learning_rate": 0.00041229689067201605,
"loss": 2.9484,
"theoretical_loss": 3.8357888844297094,
"tokens_seen": 606208000
},
{
"epoch": 0.18,
"learning_rate": 0.00041228686058174523,
"loss": 2.9537,
"theoretical_loss": 3.8357456950577618,
"tokens_seen": 606273536
},
{
"epoch": 0.18,
"learning_rate": 0.0004122768304914744,
"loss": 2.6897,
"theoretical_loss": 3.8357025116612293,
"tokens_seen": 606339072
},
{
"epoch": 0.18,
"learning_rate": 0.00041226680040120365,
"loss": 2.813,
"theoretical_loss": 3.8356593342386405,
"tokens_seen": 606404608
},
{
"epoch": 0.18,
"learning_rate": 0.0004122567703109328,
"loss": 2.8278,
"theoretical_loss": 3.8356161627885235,
"tokens_seen": 606470144
},
{
"epoch": 0.18,
"learning_rate": 0.000412246740220662,
"loss": 3.0245,
"theoretical_loss": 3.835572997309406,
"tokens_seen": 606535680
},
{
"epoch": 0.18,
"learning_rate": 0.00041223671013039114,
"loss": 2.8529,
"theoretical_loss": 3.835529837799818,
"tokens_seen": 606601216
},
{
"epoch": 0.18,
"learning_rate": 0.0004122266800401204,
"loss": 2.6315,
"theoretical_loss": 3.8354866842582895,
"tokens_seen": 606666752
},
{
"epoch": 0.18,
"learning_rate": 0.00041221664994984955,
"loss": 2.932,
"theoretical_loss": 3.835443536683349,
"tokens_seen": 606732288
},
{
"epoch": 0.18,
"learning_rate": 0.00041220661985957874,
"loss": 2.622,
"theoretical_loss": 3.835400395073529,
"tokens_seen": 606797824
},
{
"epoch": 0.18,
"learning_rate": 0.0004121965897693079,
"loss": 2.9286,
"theoretical_loss": 3.8353572594273597,
"tokens_seen": 606863360
},
{
"epoch": 0.18,
"learning_rate": 0.0004121865596790371,
"loss": 2.9135,
"theoretical_loss": 3.835314129743373,
"tokens_seen": 606928896
},
{
"epoch": 0.18,
"learning_rate": 0.0004121765295887663,
"loss": 2.8349,
"theoretical_loss": 3.835271006020101,
"tokens_seen": 606994432
},
{
"epoch": 0.18,
"learning_rate": 0.0004121664994984955,
"loss": 2.8695,
"theoretical_loss": 3.8352278882560764,
"tokens_seen": 607059968
},
{
"epoch": 0.18,
"learning_rate": 0.0004121564694082247,
"loss": 2.9327,
"theoretical_loss": 3.8351847764498332,
"tokens_seen": 607125504
},
{
"epoch": 0.18,
"learning_rate": 0.0004121464393179539,
"loss": 2.7951,
"theoretical_loss": 3.8351416705999055,
"tokens_seen": 607191040
},
{
"epoch": 0.18,
"learning_rate": 0.0004121364092276831,
"loss": 2.7969,
"theoretical_loss": 3.835098570704826,
"tokens_seen": 607256576
},
{
"epoch": 0.18,
"learning_rate": 0.00041212637913741224,
"loss": 2.9292,
"theoretical_loss": 3.8350554767631304,
"tokens_seen": 607322112
},
{
"epoch": 0.18,
"learning_rate": 0.0004121163490471415,
"loss": 2.9228,
"theoretical_loss": 3.8350123887733543,
"tokens_seen": 607387648
},
{
"epoch": 0.18,
"learning_rate": 0.0004121063189568706,
"loss": 2.6295,
"theoretical_loss": 3.8349693067340334,
"tokens_seen": 607453184
},
{
"epoch": 0.18,
"learning_rate": 0.00041209628886659984,
"loss": 3.0163,
"theoretical_loss": 3.8349262306437044,
"tokens_seen": 607518720
},
{
"epoch": 0.18,
"learning_rate": 0.000412086258776329,
"loss": 3.0652,
"theoretical_loss": 3.8348831605009033,
"tokens_seen": 607584256
},
{
"epoch": 0.18,
"learning_rate": 0.0004120762286860582,
"loss": 2.6424,
"theoretical_loss": 3.8348400963041684,
"tokens_seen": 607649792
},
{
"epoch": 0.18,
"learning_rate": 0.0004120661985957874,
"loss": 2.9793,
"theoretical_loss": 3.8347970380520375,
"tokens_seen": 607715328
},
{
"epoch": 0.18,
"learning_rate": 0.00041205616850551656,
"loss": 2.9131,
"theoretical_loss": 3.834753985743049,
"tokens_seen": 607780864
},
{
"epoch": 0.18,
"objective/train/docs_used": 354263,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.821697950363159,
"objective/train/theoretical_loss": 3.834710939375742,
"objective/train/tokens_used": 628306400,
"theoretical_loss": 3.834710939375742,
"tokens_seen": 607846400
},
{
"epoch": 0.18,
"learning_rate": 0.00041204613841524574,
"loss": 2.8469,
"theoretical_loss": 3.834710939375742,
"tokens_seen": 607846400
},
{
"epoch": 0.18,
"learning_rate": 0.000412036108324975,
"loss": 2.5941,
"theoretical_loss": 3.8346678989486556,
"tokens_seen": 607911936
},
{
"epoch": 0.18,
"learning_rate": 0.0004120260782347041,
"loss": 2.874,
"theoretical_loss": 3.83462486446033,
"tokens_seen": 607977472
},
{
"epoch": 0.18,
"learning_rate": 0.00041201604814443334,
"loss": 2.7832,
"theoretical_loss": 3.8345818359093062,
"tokens_seen": 608043008
},
{
"epoch": 0.18,
"learning_rate": 0.00041200601805416247,
"loss": 2.8339,
"theoretical_loss": 3.8345388132941243,
"tokens_seen": 608108544
},
{
"epoch": 0.18,
"learning_rate": 0.0004119959879638917,
"loss": 2.8034,
"theoretical_loss": 3.834495796613327,
"tokens_seen": 608174080
},
{
"epoch": 0.18,
"learning_rate": 0.0004119859578736209,
"loss": 3.2167,
"theoretical_loss": 3.834452785865455,
"tokens_seen": 608239616
},
{
"epoch": 0.18,
"learning_rate": 0.00041197592778335006,
"loss": 2.6944,
"theoretical_loss": 3.834409781049053,
"tokens_seen": 608305152
},
{
"epoch": 0.18,
"learning_rate": 0.00041196589769307924,
"loss": 2.9677,
"theoretical_loss": 3.834366782162662,
"tokens_seen": 608370688
},
{
"epoch": 0.18,
"learning_rate": 0.0004119558676028085,
"loss": 2.8102,
"theoretical_loss": 3.8343237892048263,
"tokens_seen": 608436224
},
{
"epoch": 0.18,
"learning_rate": 0.0004119458375125376,
"loss": 2.806,
"theoretical_loss": 3.834280802174091,
"tokens_seen": 608501760
},
{
"epoch": 0.18,
"learning_rate": 0.00041193580742226684,
"loss": 2.7141,
"theoretical_loss": 3.8342378210689994,
"tokens_seen": 608567296
},
{
"epoch": 0.18,
"learning_rate": 0.00041192577733199597,
"loss": 2.4579,
"theoretical_loss": 3.834194845888097,
"tokens_seen": 608632832
},
{
"epoch": 0.18,
"learning_rate": 0.0004119157472417252,
"loss": 2.7697,
"theoretical_loss": 3.83415187662993,
"tokens_seen": 608698368
},
{
"epoch": 0.18,
"learning_rate": 0.0004119057171514544,
"loss": 2.8852,
"theoretical_loss": 3.8341089132930444,
"tokens_seen": 608763904
},
{
"epoch": 0.18,
"learning_rate": 0.00041189568706118357,
"loss": 2.9821,
"theoretical_loss": 3.834065955875987,
"tokens_seen": 608829440
},
{
"epoch": 0.18,
"learning_rate": 0.00041188565697091275,
"loss": 2.873,
"theoretical_loss": 3.8340230043773045,
"tokens_seen": 608894976
},
{
"epoch": 0.18,
"learning_rate": 0.00041187562688064193,
"loss": 2.8123,
"theoretical_loss": 3.833980058795545,
"tokens_seen": 608960512
},
{
"epoch": 0.18,
"learning_rate": 0.0004118655967903711,
"loss": 2.8689,
"theoretical_loss": 3.8339371191292573,
"tokens_seen": 609026048
},
{
"epoch": 0.18,
"learning_rate": 0.00041185556670010035,
"loss": 2.7258,
"theoretical_loss": 3.8338941853769892,
"tokens_seen": 609091584
},
{
"epoch": 0.18,
"learning_rate": 0.00041184553660982947,
"loss": 3.06,
"theoretical_loss": 3.8338512575372903,
"tokens_seen": 609157120
},
{
"epoch": 0.18,
"learning_rate": 0.0004118355065195587,
"loss": 2.7505,
"theoretical_loss": 3.8338083356087105,
"tokens_seen": 609222656
},
{
"epoch": 0.18,
"learning_rate": 0.00041182547642928783,
"loss": 2.8264,
"theoretical_loss": 3.8337654195898,
"tokens_seen": 609288192
},
{
"epoch": 0.18,
"learning_rate": 0.00041181544633901707,
"loss": 2.7775,
"theoretical_loss": 3.8337225094791103,
"tokens_seen": 609353728
},
{
"epoch": 0.18,
"learning_rate": 0.00041180541624874625,
"loss": 2.9664,
"theoretical_loss": 3.8336796052751914,
"tokens_seen": 609419264
},
{
"epoch": 0.18,
"objective/train/docs_used": 355721,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.350303888320923,
"objective/train/theoretical_loss": 3.8336367069765958,
"objective/train/tokens_used": 629944800,
"theoretical_loss": 3.8336367069765958,
"tokens_seen": 609484800
},
{
"epoch": 0.18,
"learning_rate": 0.00041179538615847543,
"loss": 2.6211,
"theoretical_loss": 3.8336367069765958,
"tokens_seen": 609484800
},
{
"epoch": 0.18,
"learning_rate": 0.0004117853560682046,
"loss": 2.8117,
"theoretical_loss": 3.833593814581876,
"tokens_seen": 609550336
},
{
"epoch": 0.18,
"learning_rate": 0.00041177532597793385,
"loss": 2.8192,
"theoretical_loss": 3.8335509280895854,
"tokens_seen": 609615872
},
{
"epoch": 0.18,
"learning_rate": 0.000411765295887663,
"loss": 2.8,
"theoretical_loss": 3.833508047498276,
"tokens_seen": 609681408
},
{
"epoch": 0.18,
"learning_rate": 0.0004117552657973922,
"loss": 3.1224,
"theoretical_loss": 3.833465172806503,
"tokens_seen": 609746944
},
{
"epoch": 0.18,
"learning_rate": 0.00041174523570712134,
"loss": 2.7969,
"theoretical_loss": 3.833422304012819,
"tokens_seen": 609812480
},
{
"epoch": 0.18,
"learning_rate": 0.0004117352056168506,
"loss": 2.7491,
"theoretical_loss": 3.833379441115781,
"tokens_seen": 609878016
},
{
"epoch": 0.18,
"learning_rate": 0.00041172517552657975,
"loss": 2.943,
"theoretical_loss": 3.833336584113943,
"tokens_seen": 609943552
},
{
"epoch": 0.18,
"learning_rate": 0.00041171514543630894,
"loss": 2.9704,
"theoretical_loss": 3.8332937330058616,
"tokens_seen": 610009088
},
{
"epoch": 0.18,
"learning_rate": 0.0004117051153460381,
"loss": 2.7496,
"theoretical_loss": 3.8332508877900935,
"tokens_seen": 610074624
},
{
"epoch": 0.18,
"learning_rate": 0.0004116950852557673,
"loss": 2.9887,
"theoretical_loss": 3.8332080484651945,
"tokens_seen": 610140160
},
{
"epoch": 0.18,
"learning_rate": 0.0004116850551654965,
"loss": 2.7619,
"theoretical_loss": 3.8331652150297226,
"tokens_seen": 610205696
},
{
"epoch": 0.18,
"learning_rate": 0.0004116750250752257,
"loss": 2.9078,
"theoretical_loss": 3.833122387482236,
"tokens_seen": 610271232
},
{
"epoch": 0.18,
"learning_rate": 0.00041166499498495484,
"loss": 2.7804,
"theoretical_loss": 3.8330795658212926,
"tokens_seen": 610336768
},
{
"epoch": 0.18,
"learning_rate": 0.0004116549648946841,
"loss": 2.8807,
"theoretical_loss": 3.833036750045452,
"tokens_seen": 610402304
},
{
"epoch": 0.18,
"learning_rate": 0.0004116449348044132,
"loss": 2.8797,
"theoretical_loss": 3.832993940153273,
"tokens_seen": 610467840
},
{
"epoch": 0.19,
"learning_rate": 0.00041163490471414244,
"loss": 2.7678,
"theoretical_loss": 3.8329511361433166,
"tokens_seen": 610533376
},
{
"epoch": 0.19,
"learning_rate": 0.0004116248746238716,
"loss": 2.8182,
"theoretical_loss": 3.832908338014142,
"tokens_seen": 610598912
},
{
"epoch": 0.19,
"learning_rate": 0.0004116148445336008,
"loss": 2.6335,
"theoretical_loss": 3.832865545764311,
"tokens_seen": 610664448
},
{
"epoch": 0.19,
"learning_rate": 0.00041160481444333,
"loss": 2.7076,
"theoretical_loss": 3.8328227593923856,
"tokens_seen": 610729984
},
{
"epoch": 0.19,
"learning_rate": 0.0004115947843530592,
"loss": 2.9658,
"theoretical_loss": 3.832779978896926,
"tokens_seen": 610795520
},
{
"epoch": 0.19,
"learning_rate": 0.00041158475426278834,
"loss": 2.7874,
"theoretical_loss": 3.832737204276497,
"tokens_seen": 610861056
},
{
"epoch": 0.19,
"learning_rate": 0.0004115747241725176,
"loss": 2.7251,
"theoretical_loss": 3.8326944355296595,
"tokens_seen": 610926592
},
{
"epoch": 0.19,
"learning_rate": 0.0004115646940822467,
"loss": 2.8695,
"theoretical_loss": 3.832651672654978,
"tokens_seen": 610992128
},
{
"epoch": 0.19,
"learning_rate": 0.00041155466399197594,
"loss": 2.7779,
"theoretical_loss": 3.832608915651017,
"tokens_seen": 611057664
},
{
"epoch": 0.19,
"objective/train/docs_used": 356348,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.85497784614563,
"objective/train/theoretical_loss": 3.83256616451634,
"objective/train/tokens_used": 631583200,
"theoretical_loss": 3.83256616451634,
"tokens_seen": 611123200
},
{
"epoch": 0.19,
"learning_rate": 0.0004115446339017051,
"loss": 2.8653,
"theoretical_loss": 3.83256616451634,
"tokens_seen": 611123200
},
{
"epoch": 0.19,
"learning_rate": 0.0004115346038114343,
"loss": 2.6779,
"theoretical_loss": 3.8325234192495126,
"tokens_seen": 611188736
},
{
"epoch": 0.19,
"learning_rate": 0.0004115245737211635,
"loss": 2.9147,
"theoretical_loss": 3.832480679849101,
"tokens_seen": 611254272
},
{
"epoch": 0.19,
"learning_rate": 0.00041151454363089267,
"loss": 2.9427,
"theoretical_loss": 3.8324379463136697,
"tokens_seen": 611319808
},
{
"epoch": 0.19,
"learning_rate": 0.00041150451354062185,
"loss": 2.9044,
"theoretical_loss": 3.8323952186417864,
"tokens_seen": 611385344
},
{
"epoch": 0.19,
"learning_rate": 0.0004114944834503511,
"loss": 2.9848,
"theoretical_loss": 3.832352496832018,
"tokens_seen": 611450880
},
{
"epoch": 0.19,
"learning_rate": 0.0004114844533600802,
"loss": 2.8493,
"theoretical_loss": 3.832309780882932,
"tokens_seen": 611516416
},
{
"epoch": 0.19,
"learning_rate": 0.00041147442326980944,
"loss": 2.6212,
"theoretical_loss": 3.8322670707930957,
"tokens_seen": 611581952
},
{
"epoch": 0.19,
"learning_rate": 0.00041146439317953857,
"loss": 2.9301,
"theoretical_loss": 3.832224366561079,
"tokens_seen": 611647488
},
{
"epoch": 0.19,
"learning_rate": 0.0004114543630892678,
"loss": 2.6062,
"theoretical_loss": 3.83218166818545,
"tokens_seen": 611713024
},
{
"epoch": 0.19,
"learning_rate": 0.000411444332998997,
"loss": 2.7474,
"theoretical_loss": 3.8321389756647783,
"tokens_seen": 611778560
},
{
"epoch": 0.19,
"learning_rate": 0.00041143430290872617,
"loss": 2.9489,
"theoretical_loss": 3.8320962889976347,
"tokens_seen": 611844096
},
{
"epoch": 0.19,
"learning_rate": 0.00041142427281845535,
"loss": 2.9554,
"theoretical_loss": 3.832053608182589,
"tokens_seen": 611909632
},
{
"epoch": 0.19,
"learning_rate": 0.0004114142427281846,
"loss": 2.6481,
"theoretical_loss": 3.8320109332182124,
"tokens_seen": 611975168
},
{
"epoch": 0.19,
"learning_rate": 0.00041140421263791377,
"loss": 2.9068,
"theoretical_loss": 3.831968264103077,
"tokens_seen": 612040704
},
{
"epoch": 0.19,
"learning_rate": 0.00041139418254764295,
"loss": 2.779,
"theoretical_loss": 3.831925600835754,
"tokens_seen": 612106240
},
{
"epoch": 0.19,
"learning_rate": 0.00041138415245737213,
"loss": 2.7237,
"theoretical_loss": 3.831882943414817,
"tokens_seen": 612171776
},
{
"epoch": 0.19,
"learning_rate": 0.0004113741223671013,
"loss": 3.0293,
"theoretical_loss": 3.8318402918388386,
"tokens_seen": 612237312
},
{
"epoch": 0.19,
"learning_rate": 0.00041136409227683055,
"loss": 2.7863,
"theoretical_loss": 3.8317976461063923,
"tokens_seen": 612302848
},
{
"epoch": 0.19,
"learning_rate": 0.00041135406218655967,
"loss": 2.7599,
"theoretical_loss": 3.831755006216052,
"tokens_seen": 612368384
},
{
"epoch": 0.19,
"learning_rate": 0.0004113440320962889,
"loss": 2.9175,
"theoretical_loss": 3.8317123721663924,
"tokens_seen": 612433920
},
{
"epoch": 0.19,
"learning_rate": 0.00041133400200601803,
"loss": 2.4928,
"theoretical_loss": 3.831669743955989,
"tokens_seen": 612499456
},
{
"epoch": 0.19,
"learning_rate": 0.00041132397191574727,
"loss": 2.8791,
"theoretical_loss": 3.8316271215834172,
"tokens_seen": 612564992
},
{
"epoch": 0.19,
"learning_rate": 0.00041131394182547645,
"loss": 2.7784,
"theoretical_loss": 3.8315845050472532,
"tokens_seen": 612630528
},
{
"epoch": 0.19,
"learning_rate": 0.00041130391173520563,
"loss": 2.7013,
"theoretical_loss": 3.831541894346073,
"tokens_seen": 612696064
},
{
"epoch": 0.19,
"objective/train/docs_used": 356825,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0833096504211426,
"objective/train/theoretical_loss": 3.8314992894784536,
"objective/train/tokens_used": 633221600,
"theoretical_loss": 3.8314992894784536,
"tokens_seen": 612761600
},
{
"epoch": 0.19,
"learning_rate": 0.0004112938816449348,
"loss": 2.7547,
"theoretical_loss": 3.8314992894784536,
"tokens_seen": 612761600
},
{
"epoch": 0.19,
"learning_rate": 0.00041128385155466405,
"loss": 3.0618,
"theoretical_loss": 3.831456690442974,
"tokens_seen": 612827136
},
{
"epoch": 0.19,
"learning_rate": 0.0004112738214643932,
"loss": 3.0582,
"theoretical_loss": 3.8314140972382105,
"tokens_seen": 612892672
},
{
"epoch": 0.19,
"learning_rate": 0.0004112637913741224,
"loss": 2.9548,
"theoretical_loss": 3.831371509862743,
"tokens_seen": 612958208
},
{
"epoch": 0.19,
"learning_rate": 0.00041125376128385154,
"loss": 3.0347,
"theoretical_loss": 3.8313289283151497,
"tokens_seen": 613023744
},
{
"epoch": 0.19,
"learning_rate": 0.0004112437311935808,
"loss": 3.1745,
"theoretical_loss": 3.831286352594011,
"tokens_seen": 613089280
},
{
"epoch": 0.19,
"learning_rate": 0.00041123370110330995,
"loss": 2.7535,
"theoretical_loss": 3.831243782697906,
"tokens_seen": 613154816
},
{
"epoch": 0.19,
"learning_rate": 0.00041122367101303914,
"loss": 2.8579,
"theoretical_loss": 3.8312012186254156,
"tokens_seen": 613220352
},
{
"epoch": 0.19,
"learning_rate": 0.0004112136409227683,
"loss": 2.7409,
"theoretical_loss": 3.831158660375122,
"tokens_seen": 613285888
},
{
"epoch": 0.19,
"learning_rate": 0.0004112036108324975,
"loss": 2.9041,
"theoretical_loss": 3.831116107945605,
"tokens_seen": 613351424
},
{
"epoch": 0.19,
"learning_rate": 0.0004111935807422267,
"loss": 2.8949,
"theoretical_loss": 3.8310735613354483,
"tokens_seen": 613416960
},
{
"epoch": 0.19,
"learning_rate": 0.0004111835506519559,
"loss": 2.9461,
"theoretical_loss": 3.831031020543233,
"tokens_seen": 613482496
},
{
"epoch": 0.19,
"learning_rate": 0.00041117352056168504,
"loss": 2.8993,
"theoretical_loss": 3.830988485567543,
"tokens_seen": 613548032
},
{
"epoch": 0.19,
"learning_rate": 0.0004111634904714143,
"loss": 2.7299,
"theoretical_loss": 3.8309459564069615,
"tokens_seen": 613613568
},
{
"epoch": 0.19,
"learning_rate": 0.0004111534603811434,
"loss": 2.9761,
"theoretical_loss": 3.830903433060073,
"tokens_seen": 613679104
},
{
"epoch": 0.19,
"learning_rate": 0.00041114343029087264,
"loss": 3.0147,
"theoretical_loss": 3.8308609155254616,
"tokens_seen": 613744640
},
{
"epoch": 0.19,
"learning_rate": 0.0004111334002006018,
"loss": 2.8001,
"theoretical_loss": 3.830818403801713,
"tokens_seen": 613810176
},
{
"epoch": 0.19,
"learning_rate": 0.000411123370110331,
"loss": 3.06,
"theoretical_loss": 3.8307758978874116,
"tokens_seen": 613875712
},
{
"epoch": 0.19,
"learning_rate": 0.0004111133400200602,
"loss": 2.9081,
"theoretical_loss": 3.830733397781144,
"tokens_seen": 613941248
},
{
"epoch": 0.19,
"learning_rate": 0.0004111033099297894,
"loss": 2.8054,
"theoretical_loss": 3.830690903481497,
"tokens_seen": 614006784
},
{
"epoch": 0.19,
"learning_rate": 0.00041109327983951854,
"loss": 2.8168,
"theoretical_loss": 3.830648414987057,
"tokens_seen": 614072320
},
{
"epoch": 0.19,
"learning_rate": 0.0004110832497492478,
"loss": 3.0625,
"theoretical_loss": 3.8306059322964128,
"tokens_seen": 614137856
},
{
"epoch": 0.19,
"learning_rate": 0.0004110732196589769,
"loss": 2.8266,
"theoretical_loss": 3.830563455408151,
"tokens_seen": 614203392
},
{
"epoch": 0.19,
"learning_rate": 0.00041106318956870614,
"loss": 2.9967,
"theoretical_loss": 3.8305209843208603,
"tokens_seen": 614268928
},
{
"epoch": 0.19,
"learning_rate": 0.0004110531594784353,
"loss": 2.8098,
"theoretical_loss": 3.8304785190331305,
"tokens_seen": 614334464
},
{
"epoch": 0.19,
"objective/train/docs_used": 357491,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.030144453048706,
"objective/train/theoretical_loss": 3.830436059543551,
"objective/train/tokens_used": 634860000,
"theoretical_loss": 3.830436059543551,
"tokens_seen": 614400000
},
{
"epoch": 0.19,
"learning_rate": 0.0004110431293881645,
"loss": 2.9081,
"theoretical_loss": 3.830436059543551,
"tokens_seen": 614400000
},
{
"epoch": 0.19,
"learning_rate": 0.0004110330992978937,
"loss": 2.8645,
"theoretical_loss": 3.830393605850711,
"tokens_seen": 614465536
},
{
"epoch": 0.19,
"learning_rate": 0.00041102306920762287,
"loss": 2.6907,
"theoretical_loss": 3.8303511579532015,
"tokens_seen": 614531072
},
{
"epoch": 0.19,
"learning_rate": 0.00041101303911735205,
"loss": 2.8431,
"theoretical_loss": 3.8303087158496134,
"tokens_seen": 614596608
},
{
"epoch": 0.19,
"learning_rate": 0.0004110030090270813,
"loss": 2.9891,
"theoretical_loss": 3.8302662795385385,
"tokens_seen": 614662144
},
{
"epoch": 0.19,
"learning_rate": 0.0004109929789368104,
"loss": 2.9626,
"theoretical_loss": 3.8302238490185676,
"tokens_seen": 614727680
},
{
"epoch": 0.19,
"learning_rate": 0.00041098294884653965,
"loss": 2.6998,
"theoretical_loss": 3.830181424288295,
"tokens_seen": 614793216
},
{
"epoch": 0.19,
"learning_rate": 0.00041097291875626877,
"loss": 2.7542,
"theoretical_loss": 3.830139005346312,
"tokens_seen": 614858752
},
{
"epoch": 0.19,
"learning_rate": 0.000410962888665998,
"loss": 2.7335,
"theoretical_loss": 3.830096592191212,
"tokens_seen": 614924288
},
{
"epoch": 0.19,
"learning_rate": 0.0004109528585757272,
"loss": 2.7658,
"theoretical_loss": 3.8300541848215905,
"tokens_seen": 614989824
},
{
"epoch": 0.19,
"learning_rate": 0.00041094282848545637,
"loss": 2.7634,
"theoretical_loss": 3.8300117832360403,
"tokens_seen": 615055360
},
{
"epoch": 0.19,
"learning_rate": 0.00041093279839518555,
"loss": 2.8337,
"theoretical_loss": 3.8299693874331573,
"tokens_seen": 615120896
},
{
"epoch": 0.19,
"learning_rate": 0.0004109227683049148,
"loss": 2.8861,
"theoretical_loss": 3.829926997411537,
"tokens_seen": 615186432
},
{
"epoch": 0.19,
"learning_rate": 0.0004109127382146439,
"loss": 2.9835,
"theoretical_loss": 3.8298846131697744,
"tokens_seen": 615251968
},
{
"epoch": 0.19,
"learning_rate": 0.00041090270812437315,
"loss": 2.8007,
"theoretical_loss": 3.829842234706466,
"tokens_seen": 615317504
},
{
"epoch": 0.19,
"learning_rate": 0.0004108926780341023,
"loss": 2.9777,
"theoretical_loss": 3.8297998620202094,
"tokens_seen": 615383040
},
{
"epoch": 0.19,
"learning_rate": 0.0004108826479438315,
"loss": 2.7049,
"theoretical_loss": 3.829757495109602,
"tokens_seen": 615448576
},
{
"epoch": 0.19,
"learning_rate": 0.0004108726178535607,
"loss": 2.7635,
"theoretical_loss": 3.8297151339732407,
"tokens_seen": 615514112
},
{
"epoch": 0.19,
"learning_rate": 0.00041086258776328987,
"loss": 2.9916,
"theoretical_loss": 3.8296727786097247,
"tokens_seen": 615579648
},
{
"epoch": 0.19,
"learning_rate": 0.00041085255767301905,
"loss": 2.9294,
"theoretical_loss": 3.8296304290176524,
"tokens_seen": 615645184
},
{
"epoch": 0.19,
"learning_rate": 0.00041084252758274823,
"loss": 2.9406,
"theoretical_loss": 3.829588085195623,
"tokens_seen": 615710720
},
{
"epoch": 0.19,
"learning_rate": 0.0004108324974924774,
"loss": 3.1951,
"theoretical_loss": 3.8295457471422365,
"tokens_seen": 615776256
},
{
"epoch": 0.19,
"learning_rate": 0.00041082246740220665,
"loss": 2.8901,
"theoretical_loss": 3.8295034148560934,
"tokens_seen": 615841792
},
{
"epoch": 0.19,
"learning_rate": 0.0004108124373119358,
"loss": 2.7115,
"theoretical_loss": 3.8294610883357945,
"tokens_seen": 615907328
},
{
"epoch": 0.19,
"learning_rate": 0.000410802407221665,
"loss": 2.8682,
"theoretical_loss": 3.8294187675799405,
"tokens_seen": 615972864
},
{
"epoch": 0.19,
"objective/train/docs_used": 358741,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9842801094055176,
"objective/train/theoretical_loss": 3.829376452587134,
"objective/train/tokens_used": 636498400,
"theoretical_loss": 3.829376452587134,
"tokens_seen": 616038400
},
{
"epoch": 0.19,
"learning_rate": 0.0004107923771313942,
"loss": 2.756,
"theoretical_loss": 3.829376452587134,
"tokens_seen": 616038400
},
{
"epoch": 0.19,
"learning_rate": 0.0004107823470411234,
"loss": 2.7453,
"theoretical_loss": 3.8293341433559767,
"tokens_seen": 616103936
},
{
"epoch": 0.19,
"learning_rate": 0.00041077231695085256,
"loss": 2.7221,
"theoretical_loss": 3.8292918398850713,
"tokens_seen": 616169472
},
{
"epoch": 0.19,
"learning_rate": 0.00041076228686058174,
"loss": 3.0387,
"theoretical_loss": 3.829249542173021,
"tokens_seen": 616235008
},
{
"epoch": 0.19,
"learning_rate": 0.0004107522567703109,
"loss": 2.8791,
"theoretical_loss": 3.82920725021843,
"tokens_seen": 616300544
},
{
"epoch": 0.19,
"learning_rate": 0.00041074222668004015,
"loss": 2.8977,
"theoretical_loss": 3.8291649640199017,
"tokens_seen": 616366080
},
{
"epoch": 0.19,
"learning_rate": 0.0004107321965897693,
"loss": 2.974,
"theoretical_loss": 3.829122683576042,
"tokens_seen": 616431616
},
{
"epoch": 0.19,
"learning_rate": 0.0004107221664994985,
"loss": 3.0914,
"theoretical_loss": 3.8290804088854546,
"tokens_seen": 616497152
},
{
"epoch": 0.19,
"learning_rate": 0.00041071213640922764,
"loss": 2.7757,
"theoretical_loss": 3.8290381399467464,
"tokens_seen": 616562688
},
{
"epoch": 0.19,
"learning_rate": 0.0004107021063189569,
"loss": 2.9125,
"theoretical_loss": 3.8289958767585226,
"tokens_seen": 616628224
},
{
"epoch": 0.19,
"learning_rate": 0.00041069207622868606,
"loss": 2.76,
"theoretical_loss": 3.8289536193193907,
"tokens_seen": 616693760
},
{
"epoch": 0.19,
"learning_rate": 0.00041068204613841524,
"loss": 2.7955,
"theoretical_loss": 3.828911367627957,
"tokens_seen": 616759296
},
{
"epoch": 0.19,
"learning_rate": 0.0004106720160481444,
"loss": 2.8873,
"theoretical_loss": 3.8288691216828297,
"tokens_seen": 616824832
},
{
"epoch": 0.19,
"learning_rate": 0.0004106619859578736,
"loss": 2.6994,
"theoretical_loss": 3.828826881482617,
"tokens_seen": 616890368
},
{
"epoch": 0.19,
"learning_rate": 0.00041065195586760284,
"loss": 2.8009,
"theoretical_loss": 3.8287846470259264,
"tokens_seen": 616955904
},
{
"epoch": 0.19,
"learning_rate": 0.000410641925777332,
"loss": 2.7532,
"theoretical_loss": 3.828742418311368,
"tokens_seen": 617021440
},
{
"epoch": 0.19,
"learning_rate": 0.0004106318956870612,
"loss": 2.768,
"theoretical_loss": 3.828700195337551,
"tokens_seen": 617086976
},
{
"epoch": 0.19,
"learning_rate": 0.0004106218655967904,
"loss": 2.9357,
"theoretical_loss": 3.8286579781030863,
"tokens_seen": 617152512
},
{
"epoch": 0.19,
"learning_rate": 0.0004106118355065196,
"loss": 2.9161,
"theoretical_loss": 3.8286157666065828,
"tokens_seen": 617218048
},
{
"epoch": 0.19,
"learning_rate": 0.00041060180541624874,
"loss": 2.932,
"theoretical_loss": 3.828573560846652,
"tokens_seen": 617283584
},
{
"epoch": 0.19,
"learning_rate": 0.000410591775325978,
"loss": 2.9844,
"theoretical_loss": 3.8285313608219056,
"tokens_seen": 617349120
},
{
"epoch": 0.19,
"learning_rate": 0.0004105817452357071,
"loss": 2.7526,
"theoretical_loss": 3.8284891665309564,
"tokens_seen": 617414656
},
{
"epoch": 0.19,
"learning_rate": 0.00041057171514543634,
"loss": 2.7736,
"theoretical_loss": 3.8284469779724155,
"tokens_seen": 617480192
},
{
"epoch": 0.19,
"learning_rate": 0.0004105616850551655,
"loss": 2.8821,
"theoretical_loss": 3.8284047951448965,
"tokens_seen": 617545728
},
{
"epoch": 0.19,
"learning_rate": 0.0004105516549648947,
"loss": 2.8562,
"theoretical_loss": 3.828362618047013,
"tokens_seen": 617611264
},
{
"epoch": 0.19,
"objective/train/docs_used": 359404,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.935211181640625,
"objective/train/theoretical_loss": 3.828320446677378,
"objective/train/tokens_used": 638136800,
"theoretical_loss": 3.828320446677378,
"tokens_seen": 617676800
},
{
"epoch": 0.19,
"learning_rate": 0.0004105416248746239,
"loss": 2.5811,
"theoretical_loss": 3.828320446677378,
"tokens_seen": 617676800
},
{
"epoch": 0.19,
"learning_rate": 0.00041053159478435307,
"loss": 2.8761,
"theoretical_loss": 3.8282782810346063,
"tokens_seen": 617742336
},
{
"epoch": 0.19,
"learning_rate": 0.00041052156469408225,
"loss": 3.19,
"theoretical_loss": 3.8282361211173135,
"tokens_seen": 617807872
},
{
"epoch": 0.19,
"learning_rate": 0.0004105115346038115,
"loss": 2.6977,
"theoretical_loss": 3.8281939669241143,
"tokens_seen": 617873408
},
{
"epoch": 0.19,
"learning_rate": 0.0004105015045135406,
"loss": 3.0995,
"theoretical_loss": 3.8281518184536245,
"tokens_seen": 617938944
},
{
"epoch": 0.19,
"learning_rate": 0.00041049147442326985,
"loss": 2.9678,
"theoretical_loss": 3.82810967570446,
"tokens_seen": 618004480
},
{
"epoch": 0.19,
"learning_rate": 0.00041048144433299897,
"loss": 2.8602,
"theoretical_loss": 3.8280675386752385,
"tokens_seen": 618070016
},
{
"epoch": 0.19,
"learning_rate": 0.0004104714142427282,
"loss": 2.9506,
"theoretical_loss": 3.828025407364577,
"tokens_seen": 618135552
},
{
"epoch": 0.19,
"learning_rate": 0.0004104613841524574,
"loss": 2.7122,
"theoretical_loss": 3.827983281771093,
"tokens_seen": 618201088
},
{
"epoch": 0.19,
"learning_rate": 0.00041045135406218657,
"loss": 2.6051,
"theoretical_loss": 3.8279411618934045,
"tokens_seen": 618266624
},
{
"epoch": 0.19,
"learning_rate": 0.00041044132397191575,
"loss": 3.041,
"theoretical_loss": 3.8278990477301305,
"tokens_seen": 618332160
},
{
"epoch": 0.19,
"learning_rate": 0.000410431293881645,
"loss": 2.7328,
"theoretical_loss": 3.82785693927989,
"tokens_seen": 618397696
},
{
"epoch": 0.19,
"learning_rate": 0.0004104212637913741,
"loss": 2.7715,
"theoretical_loss": 3.827814836541303,
"tokens_seen": 618463232
},
{
"epoch": 0.19,
"learning_rate": 0.00041041123370110335,
"loss": 2.9858,
"theoretical_loss": 3.82777273951299,
"tokens_seen": 618528768
},
{
"epoch": 0.19,
"learning_rate": 0.0004104012036108325,
"loss": 2.8936,
"theoretical_loss": 3.8277306481935707,
"tokens_seen": 618594304
},
{
"epoch": 0.19,
"learning_rate": 0.0004103911735205617,
"loss": 2.9234,
"theoretical_loss": 3.8276885625816663,
"tokens_seen": 618659840
},
{
"epoch": 0.19,
"learning_rate": 0.0004103811434302909,
"loss": 2.7475,
"theoretical_loss": 3.827646482675899,
"tokens_seen": 618725376
},
{
"epoch": 0.19,
"learning_rate": 0.00041037111334002007,
"loss": 3.0746,
"theoretical_loss": 3.8276044084748904,
"tokens_seen": 618790912
},
{
"epoch": 0.19,
"learning_rate": 0.00041036108324974925,
"loss": 2.7544,
"theoretical_loss": 3.8275623399772636,
"tokens_seen": 618856448
},
{
"epoch": 0.19,
"learning_rate": 0.00041035105315947844,
"loss": 2.4142,
"theoretical_loss": 3.827520277181641,
"tokens_seen": 618921984
},
{
"epoch": 0.19,
"learning_rate": 0.0004103410230692076,
"loss": 2.9755,
"theoretical_loss": 3.8274782200866464,
"tokens_seen": 618987520
},
{
"epoch": 0.19,
"learning_rate": 0.00041033099297893685,
"loss": 2.8222,
"theoretical_loss": 3.8274361686909035,
"tokens_seen": 619053056
},
{
"epoch": 0.19,
"learning_rate": 0.000410320962888666,
"loss": 2.9724,
"theoretical_loss": 3.827394122993037,
"tokens_seen": 619118592
},
{
"epoch": 0.19,
"learning_rate": 0.0004103109327983952,
"loss": 2.8992,
"theoretical_loss": 3.827352082991672,
"tokens_seen": 619184128
},
{
"epoch": 0.19,
"learning_rate": 0.0004103009027081244,
"loss": 2.8078,
"theoretical_loss": 3.8273100486854332,
"tokens_seen": 619249664
},
{
"epoch": 0.19,
"objective/train/docs_used": 360794,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5488815307617188,
"objective/train/theoretical_loss": 3.827268020072948,
"objective/train/tokens_used": 639775200,
"theoretical_loss": 3.827268020072948,
"tokens_seen": 619315200
},
{
"epoch": 0.19,
"learning_rate": 0.0004102908726178536,
"loss": 2.8808,
"theoretical_loss": 3.827268020072948,
"tokens_seen": 619315200
},
{
"epoch": 0.19,
"learning_rate": 0.00041028084252758276,
"loss": 2.8708,
"theoretical_loss": 3.8272259971528406,
"tokens_seen": 619380736
},
{
"epoch": 0.19,
"learning_rate": 0.00041027081243731194,
"loss": 2.7716,
"theoretical_loss": 3.8271839799237397,
"tokens_seen": 619446272
},
{
"epoch": 0.19,
"learning_rate": 0.0004102607823470411,
"loss": 2.7464,
"theoretical_loss": 3.8271419683842716,
"tokens_seen": 619511808
},
{
"epoch": 0.19,
"learning_rate": 0.00041025075225677035,
"loss": 2.6849,
"theoretical_loss": 3.8270999625330644,
"tokens_seen": 619577344
},
{
"epoch": 0.19,
"learning_rate": 0.0004102407221664995,
"loss": 2.963,
"theoretical_loss": 3.8270579623687464,
"tokens_seen": 619642880
},
{
"epoch": 0.19,
"learning_rate": 0.0004102306920762287,
"loss": 2.789,
"theoretical_loss": 3.8270159678899462,
"tokens_seen": 619708416
},
{
"epoch": 0.19,
"learning_rate": 0.00041022066198595784,
"loss": 2.7791,
"theoretical_loss": 3.826973979095293,
"tokens_seen": 619773952
},
{
"epoch": 0.19,
"learning_rate": 0.0004102106318956871,
"loss": 2.9164,
"theoretical_loss": 3.8269319959834167,
"tokens_seen": 619839488
},
{
"epoch": 0.19,
"learning_rate": 0.00041020060180541626,
"loss": 2.74,
"theoretical_loss": 3.8268900185529473,
"tokens_seen": 619905024
},
{
"epoch": 0.19,
"learning_rate": 0.00041019057171514544,
"loss": 2.7933,
"theoretical_loss": 3.8268480468025157,
"tokens_seen": 619970560
},
{
"epoch": 0.19,
"learning_rate": 0.0004101805416248746,
"loss": 2.7509,
"theoretical_loss": 3.826806080730753,
"tokens_seen": 620036096
},
{
"epoch": 0.19,
"learning_rate": 0.0004101705115346038,
"loss": 2.683,
"theoretical_loss": 3.82676412033629,
"tokens_seen": 620101632
},
{
"epoch": 0.19,
"learning_rate": 0.000410160481444333,
"loss": 2.4988,
"theoretical_loss": 3.8267221656177606,
"tokens_seen": 620167168
},
{
"epoch": 0.19,
"learning_rate": 0.0004101504513540622,
"loss": 2.9297,
"theoretical_loss": 3.826680216573795,
"tokens_seen": 620232704
},
{
"epoch": 0.19,
"learning_rate": 0.00041014042126379135,
"loss": 2.8455,
"theoretical_loss": 3.8266382732030277,
"tokens_seen": 620298240
},
{
"epoch": 0.19,
"learning_rate": 0.0004101303911735206,
"loss": 2.5712,
"theoretical_loss": 3.8265963355040915,
"tokens_seen": 620363776
},
{
"epoch": 0.19,
"learning_rate": 0.00041012036108324976,
"loss": 2.7332,
"theoretical_loss": 3.8265544034756216,
"tokens_seen": 620429312
},
{
"epoch": 0.19,
"learning_rate": 0.00041011033099297894,
"loss": 2.8088,
"theoretical_loss": 3.8265124771162506,
"tokens_seen": 620494848
},
{
"epoch": 0.19,
"learning_rate": 0.0004101003009027081,
"loss": 2.8373,
"theoretical_loss": 3.8264705564246153,
"tokens_seen": 620560384
},
{
"epoch": 0.19,
"learning_rate": 0.0004100902708124373,
"loss": 2.8893,
"theoretical_loss": 3.8264286413993496,
"tokens_seen": 620625920
},
{
"epoch": 0.19,
"learning_rate": 0.0004100802407221665,
"loss": 2.7777,
"theoretical_loss": 3.82638673203909,
"tokens_seen": 620691456
},
{
"epoch": 0.19,
"learning_rate": 0.0004100702106318957,
"loss": 2.8751,
"theoretical_loss": 3.8263448283424726,
"tokens_seen": 620756992
},
{
"epoch": 0.19,
"learning_rate": 0.00041006018054162485,
"loss": 2.9087,
"theoretical_loss": 3.8263029303081346,
"tokens_seen": 620822528
},
{
"epoch": 0.19,
"learning_rate": 0.0004100501504513541,
"loss": 2.9091,
"theoretical_loss": 3.826261037934713,
"tokens_seen": 620888064
},
{
"epoch": 0.19,
"objective/train/docs_used": 361341,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.479189395904541,
"objective/train/theoretical_loss": 3.8262191512208448,
"objective/train/tokens_used": 641413600,
"theoretical_loss": 3.8262191512208448,
"tokens_seen": 620953600
},
{
"epoch": 0.19,
"learning_rate": 0.0004100401203610832,
"loss": 2.7642,
"theoretical_loss": 3.8262191512208448,
"tokens_seen": 620953600
},
{
"epoch": 0.19,
"learning_rate": 0.00041003009027081245,
"loss": 3.0473,
"theoretical_loss": 3.8261772701651697,
"tokens_seen": 621019136
},
{
"epoch": 0.19,
"learning_rate": 0.00041002006018054163,
"loss": 2.8635,
"theoretical_loss": 3.8261353947663252,
"tokens_seen": 621084672
},
{
"epoch": 0.19,
"learning_rate": 0.0004100100300902708,
"loss": 2.5395,
"theoretical_loss": 3.8260935250229515,
"tokens_seen": 621150208
},
{
"epoch": 0.19,
"learning_rate": 0.00041,
"loss": 2.8946,
"theoretical_loss": 3.826051660933687,
"tokens_seen": 621215744
},
{
"epoch": 0.19,
"learning_rate": 0.00040998996990972917,
"loss": 2.883,
"theoretical_loss": 3.8260098024971727,
"tokens_seen": 621281280
},
{
"epoch": 0.19,
"learning_rate": 0.00040997993981945835,
"loss": 2.9444,
"theoretical_loss": 3.825967949712049,
"tokens_seen": 621346816
},
{
"epoch": 0.19,
"learning_rate": 0.0004099699097291876,
"loss": 2.9486,
"theoretical_loss": 3.8259261025769566,
"tokens_seen": 621412352
},
{
"epoch": 0.19,
"learning_rate": 0.0004099598796389167,
"loss": 3.0868,
"theoretical_loss": 3.8258842610905375,
"tokens_seen": 621477888
},
{
"epoch": 0.19,
"learning_rate": 0.00040994984954864595,
"loss": 2.8341,
"theoretical_loss": 3.825842425251434,
"tokens_seen": 621543424
},
{
"epoch": 0.19,
"learning_rate": 0.00040993981945837513,
"loss": 2.9232,
"theoretical_loss": 3.8258005950582867,
"tokens_seen": 621608960
},
{
"epoch": 0.19,
"learning_rate": 0.0004099297893681043,
"loss": 3.089,
"theoretical_loss": 3.8257587705097404,
"tokens_seen": 621674496
},
{
"epoch": 0.19,
"learning_rate": 0.0004099197592778335,
"loss": 2.8746,
"theoretical_loss": 3.825716951604438,
"tokens_seen": 621740032
},
{
"epoch": 0.19,
"learning_rate": 0.0004099097291875627,
"loss": 2.9131,
"theoretical_loss": 3.8256751383410235,
"tokens_seen": 621805568
},
{
"epoch": 0.19,
"learning_rate": 0.0004098996990972919,
"loss": 2.8811,
"theoretical_loss": 3.82563333071814,
"tokens_seen": 621871104
},
{
"epoch": 0.19,
"learning_rate": 0.0004098896690070211,
"loss": 2.8428,
"theoretical_loss": 3.825591528734434,
"tokens_seen": 621936640
},
{
"epoch": 0.19,
"learning_rate": 0.0004098796389167503,
"loss": 2.8625,
"theoretical_loss": 3.82554973238855,
"tokens_seen": 622002176
},
{
"epoch": 0.19,
"learning_rate": 0.00040986960882647945,
"loss": 2.6699,
"theoretical_loss": 3.8255079416791338,
"tokens_seen": 622067712
},
{
"epoch": 0.19,
"learning_rate": 0.00040985957873620864,
"loss": 2.7263,
"theoretical_loss": 3.825466156604831,
"tokens_seen": 622133248
},
{
"epoch": 0.19,
"learning_rate": 0.0004098495486459378,
"loss": 2.9442,
"theoretical_loss": 3.8254243771642895,
"tokens_seen": 622198784
},
{
"epoch": 0.19,
"learning_rate": 0.00040983951855566705,
"loss": 2.7449,
"theoretical_loss": 3.8253826033561555,
"tokens_seen": 622264320
},
{
"epoch": 0.19,
"learning_rate": 0.0004098294884653962,
"loss": 2.8408,
"theoretical_loss": 3.8253408351790767,
"tokens_seen": 622329856
},
{
"epoch": 0.19,
"learning_rate": 0.0004098194583751254,
"loss": 2.6127,
"theoretical_loss": 3.8252990726317018,
"tokens_seen": 622395392
},
{
"epoch": 0.19,
"learning_rate": 0.0004098094282848546,
"loss": 3.1112,
"theoretical_loss": 3.825257315712678,
"tokens_seen": 622460928
},
{
"epoch": 0.19,
"learning_rate": 0.0004097993981945838,
"loss": 2.7346,
"theoretical_loss": 3.825215564420656,
"tokens_seen": 622526464
},
{
"epoch": 0.19,
"objective/train/docs_used": 362042,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1056301593780518,
"objective/train/theoretical_loss": 3.8251738187542843,
"objective/train/tokens_used": 643052000,
"theoretical_loss": 3.8251738187542843,
"tokens_seen": 622592000
},
{
"epoch": 0.19,
"learning_rate": 0.00040978936810431296,
"loss": 2.9863,
"theoretical_loss": 3.8251738187542843,
"tokens_seen": 622592000
},
{
"epoch": 0.19,
"learning_rate": 0.00040977933801404214,
"loss": 2.7314,
"theoretical_loss": 3.8251320787122127,
"tokens_seen": 622657536
},
{
"epoch": 0.19,
"learning_rate": 0.0004097693079237713,
"loss": 2.7175,
"theoretical_loss": 3.8250903442930917,
"tokens_seen": 622723072
},
{
"epoch": 0.19,
"learning_rate": 0.00040975927783350055,
"loss": 2.7428,
"theoretical_loss": 3.8250486154955725,
"tokens_seen": 622788608
},
{
"epoch": 0.19,
"learning_rate": 0.0004097492477432297,
"loss": 2.9722,
"theoretical_loss": 3.8250068923183065,
"tokens_seen": 622854144
},
{
"epoch": 0.19,
"learning_rate": 0.0004097392176529589,
"loss": 2.815,
"theoretical_loss": 3.824965174759945,
"tokens_seen": 622919680
},
{
"epoch": 0.19,
"learning_rate": 0.00040972918756268804,
"loss": 2.7863,
"theoretical_loss": 3.8249234628191404,
"tokens_seen": 622985216
},
{
"epoch": 0.19,
"learning_rate": 0.0004097191574724173,
"loss": 2.8445,
"theoretical_loss": 3.8248817564945465,
"tokens_seen": 623050752
},
{
"epoch": 0.19,
"learning_rate": 0.00040970912738214646,
"loss": 2.9068,
"theoretical_loss": 3.8248400557848146,
"tokens_seen": 623116288
},
{
"epoch": 0.19,
"learning_rate": 0.00040969909729187564,
"loss": 2.7792,
"theoretical_loss": 3.8247983606885994,
"tokens_seen": 623181824
},
{
"epoch": 0.19,
"learning_rate": 0.0004096890672016048,
"loss": 2.8544,
"theoretical_loss": 3.824756671204555,
"tokens_seen": 623247360
},
{
"epoch": 0.19,
"learning_rate": 0.000409679037111334,
"loss": 2.6811,
"theoretical_loss": 3.824714987331336,
"tokens_seen": 623312896
},
{
"epoch": 0.19,
"learning_rate": 0.0004096690070210632,
"loss": 3.0151,
"theoretical_loss": 3.824673309067598,
"tokens_seen": 623378432
},
{
"epoch": 0.19,
"learning_rate": 0.0004096589769307924,
"loss": 2.8711,
"theoretical_loss": 3.8246316364119948,
"tokens_seen": 623443968
},
{
"epoch": 0.19,
"learning_rate": 0.00040964894684052155,
"loss": 2.8549,
"theoretical_loss": 3.824589969363184,
"tokens_seen": 623509504
},
{
"epoch": 0.19,
"learning_rate": 0.0004096389167502508,
"loss": 2.7348,
"theoretical_loss": 3.8245483079198213,
"tokens_seen": 623575040
},
{
"epoch": 0.19,
"learning_rate": 0.00040962888665997996,
"loss": 2.5999,
"theoretical_loss": 3.824506652080564,
"tokens_seen": 623640576
},
{
"epoch": 0.19,
"learning_rate": 0.00040961885656970914,
"loss": 2.8808,
"theoretical_loss": 3.824465001844069,
"tokens_seen": 623706112
},
{
"epoch": 0.19,
"learning_rate": 0.0004096088264794383,
"loss": 3.0079,
"theoretical_loss": 3.8244233572089943,
"tokens_seen": 623771648
},
{
"epoch": 0.19,
"learning_rate": 0.0004095987963891675,
"loss": 2.985,
"theoretical_loss": 3.824381718173999,
"tokens_seen": 623837184
},
{
"epoch": 0.19,
"learning_rate": 0.0004095887662988967,
"loss": 3.093,
"theoretical_loss": 3.8243400847377407,
"tokens_seen": 623902720
},
{
"epoch": 0.19,
"learning_rate": 0.0004095787362086259,
"loss": 2.8866,
"theoretical_loss": 3.824298456898879,
"tokens_seen": 623968256
},
{
"epoch": 0.19,
"learning_rate": 0.00040956870611835505,
"loss": 2.8234,
"theoretical_loss": 3.8242568346560737,
"tokens_seen": 624033792
},
{
"epoch": 0.19,
"learning_rate": 0.0004095586760280843,
"loss": 2.8825,
"theoretical_loss": 3.8242152180079847,
"tokens_seen": 624099328
},
{
"epoch": 0.19,
"learning_rate": 0.0004095486459378134,
"loss": 2.6858,
"theoretical_loss": 3.8241736069532726,
"tokens_seen": 624164864
},
{
"epoch": 0.19,
"objective/train/docs_used": 363016,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2174317836761475,
"objective/train/theoretical_loss": 3.8241320014905993,
"objective/train/tokens_used": 644690400,
"theoretical_loss": 3.8241320014905993,
"tokens_seen": 624230400
},
{
"epoch": 0.19,
"learning_rate": 0.00040953861584754265,
"loss": 2.76,
"theoretical_loss": 3.8241320014905993,
"tokens_seen": 624230400
},
{
"epoch": 0.19,
"learning_rate": 0.00040952858575727183,
"loss": 2.7017,
"theoretical_loss": 3.824090401618625,
"tokens_seen": 624295936
},
{
"epoch": 0.19,
"learning_rate": 0.000409518555667001,
"loss": 3.1082,
"theoretical_loss": 3.8240488073360126,
"tokens_seen": 624361472
},
{
"epoch": 0.19,
"learning_rate": 0.0004095085255767302,
"loss": 2.6126,
"theoretical_loss": 3.8240072186414245,
"tokens_seen": 624427008
},
{
"epoch": 0.19,
"learning_rate": 0.00040949849548645937,
"loss": 3.0417,
"theoretical_loss": 3.8239656355335234,
"tokens_seen": 624492544
},
{
"epoch": 0.19,
"learning_rate": 0.00040948846539618855,
"loss": 3.148,
"theoretical_loss": 3.8239240580109723,
"tokens_seen": 624558080
},
{
"epoch": 0.19,
"learning_rate": 0.0004094784353059178,
"loss": 2.7648,
"theoretical_loss": 3.8238824860724354,
"tokens_seen": 624623616
},
{
"epoch": 0.19,
"learning_rate": 0.0004094684052156469,
"loss": 2.9155,
"theoretical_loss": 3.8238409197165772,
"tokens_seen": 624689152
},
{
"epoch": 0.19,
"learning_rate": 0.00040945837512537615,
"loss": 2.7267,
"theoretical_loss": 3.823799358942062,
"tokens_seen": 624754688
},
{
"epoch": 0.19,
"learning_rate": 0.00040944834503510533,
"loss": 2.978,
"theoretical_loss": 3.8237578037475553,
"tokens_seen": 624820224
},
{
"epoch": 0.19,
"learning_rate": 0.0004094383149448345,
"loss": 2.7433,
"theoretical_loss": 3.823716254131723,
"tokens_seen": 624885760
},
{
"epoch": 0.19,
"learning_rate": 0.0004094282848545637,
"loss": 2.8681,
"theoretical_loss": 3.823674710093231,
"tokens_seen": 624951296
},
{
"epoch": 0.19,
"learning_rate": 0.0004094182547642929,
"loss": 3.0491,
"theoretical_loss": 3.8236331716307452,
"tokens_seen": 625016832
},
{
"epoch": 0.19,
"learning_rate": 0.00040940822467402206,
"loss": 2.9378,
"theoretical_loss": 3.8235916387429336,
"tokens_seen": 625082368
},
{
"epoch": 0.19,
"learning_rate": 0.0004093981945837513,
"loss": 2.57,
"theoretical_loss": 3.823550111428463,
"tokens_seen": 625147904
},
{
"epoch": 0.19,
"learning_rate": 0.0004093881644934804,
"loss": 2.7938,
"theoretical_loss": 3.8235085896860017,
"tokens_seen": 625213440
},
{
"epoch": 0.19,
"learning_rate": 0.00040937813440320965,
"loss": 3.0367,
"theoretical_loss": 3.8234670735142187,
"tokens_seen": 625278976
},
{
"epoch": 0.19,
"learning_rate": 0.0004093681043129388,
"loss": 2.8386,
"theoretical_loss": 3.8234255629117815,
"tokens_seen": 625344512
},
{
"epoch": 0.19,
"learning_rate": 0.000409358074222668,
"loss": 2.6205,
"theoretical_loss": 3.8233840578773606,
"tokens_seen": 625410048
},
{
"epoch": 0.19,
"learning_rate": 0.0004093480441323972,
"loss": 2.9303,
"theoretical_loss": 3.8233425584096254,
"tokens_seen": 625475584
},
{
"epoch": 0.19,
"learning_rate": 0.0004093380140421264,
"loss": 2.8171,
"theoretical_loss": 3.8233010645072456,
"tokens_seen": 625541120
},
{
"epoch": 0.19,
"learning_rate": 0.00040932798395185556,
"loss": 2.9981,
"theoretical_loss": 3.8232595761688932,
"tokens_seen": 625606656
},
{
"epoch": 0.19,
"learning_rate": 0.0004093179538615848,
"loss": 2.697,
"theoretical_loss": 3.8232180933932383,
"tokens_seen": 625672192
},
{
"epoch": 0.19,
"learning_rate": 0.0004093079237713139,
"loss": 2.6989,
"theoretical_loss": 3.8231766161789524,
"tokens_seen": 625737728
},
{
"epoch": 0.19,
"learning_rate": 0.00040929789368104316,
"loss": 2.7926,
"theoretical_loss": 3.8231351445247084,
"tokens_seen": 625803264
},
{
"epoch": 0.19,
"objective/train/docs_used": 363672,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9359257221221924,
"objective/train/theoretical_loss": 3.8230936784291787,
"objective/train/tokens_used": 646328800,
"theoretical_loss": 3.8230936784291787,
"tokens_seen": 625868800
},
{
"epoch": 0.19,
"learning_rate": 0.0004092878635907723,
"loss": 2.8781,
"theoretical_loss": 3.8230936784291787,
"tokens_seen": 625868800
},
{
"epoch": 0.19,
"learning_rate": 0.0004092778335005015,
"loss": 2.9496,
"theoretical_loss": 3.8230522178910356,
"tokens_seen": 625934336
},
{
"epoch": 0.19,
"learning_rate": 0.0004092678034102307,
"loss": 2.9525,
"theoretical_loss": 3.823010762908953,
"tokens_seen": 625999872
},
{
"epoch": 0.19,
"learning_rate": 0.0004092577733199599,
"loss": 2.7951,
"theoretical_loss": 3.8229693134816043,
"tokens_seen": 626065408
},
{
"epoch": 0.19,
"learning_rate": 0.00040924774322968906,
"loss": 2.9869,
"theoretical_loss": 3.822927869607664,
"tokens_seen": 626130944
},
{
"epoch": 0.19,
"learning_rate": 0.00040923771313941824,
"loss": 2.7103,
"theoretical_loss": 3.8228864312858084,
"tokens_seen": 626196480
},
{
"epoch": 0.19,
"learning_rate": 0.0004092276830491474,
"loss": 2.9631,
"theoretical_loss": 3.822844998514711,
"tokens_seen": 626262016
},
{
"epoch": 0.19,
"learning_rate": 0.00040921765295887666,
"loss": 3.0385,
"theoretical_loss": 3.822803571293048,
"tokens_seen": 626327552
},
{
"epoch": 0.19,
"learning_rate": 0.0004092076228686058,
"loss": 2.9119,
"theoretical_loss": 3.8227621496194946,
"tokens_seen": 626393088
},
{
"epoch": 0.19,
"learning_rate": 0.000409197592778335,
"loss": 2.7128,
"theoretical_loss": 3.8227207334927296,
"tokens_seen": 626458624
},
{
"epoch": 0.19,
"learning_rate": 0.00040918756268806415,
"loss": 2.9564,
"theoretical_loss": 3.822679322911429,
"tokens_seen": 626524160
},
{
"epoch": 0.19,
"learning_rate": 0.0004091775325977934,
"loss": 2.8966,
"theoretical_loss": 3.8226379178742693,
"tokens_seen": 626589696
},
{
"epoch": 0.19,
"learning_rate": 0.00040916750250752257,
"loss": 3.1937,
"theoretical_loss": 3.82259651837993,
"tokens_seen": 626655232
},
{
"epoch": 0.19,
"learning_rate": 0.00040915747241725175,
"loss": 2.8648,
"theoretical_loss": 3.822555124427089,
"tokens_seen": 626720768
},
{
"epoch": 0.19,
"learning_rate": 0.000409147442326981,
"loss": 2.7456,
"theoretical_loss": 3.8225137360144252,
"tokens_seen": 626786304
},
{
"epoch": 0.19,
"learning_rate": 0.00040913741223671016,
"loss": 2.8642,
"theoretical_loss": 3.8224723531406175,
"tokens_seen": 626851840
},
{
"epoch": 0.19,
"learning_rate": 0.00040912738214643934,
"loss": 3.0406,
"theoretical_loss": 3.8224309758043464,
"tokens_seen": 626917376
},
{
"epoch": 0.19,
"learning_rate": 0.0004091173520561685,
"loss": 2.8214,
"theoretical_loss": 3.8223896040042913,
"tokens_seen": 626982912
},
{
"epoch": 0.19,
"learning_rate": 0.0004091073219658977,
"loss": 3.0053,
"theoretical_loss": 3.822348237739134,
"tokens_seen": 627048448
},
{
"epoch": 0.19,
"learning_rate": 0.0004090972918756269,
"loss": 3.0134,
"theoretical_loss": 3.8223068770075552,
"tokens_seen": 627113984
},
{
"epoch": 0.19,
"learning_rate": 0.0004090872617853561,
"loss": 2.8663,
"theoretical_loss": 3.822265521808236,
"tokens_seen": 627179520
},
{
"epoch": 0.19,
"learning_rate": 0.00040907723169508525,
"loss": 3.1141,
"theoretical_loss": 3.822224172139859,
"tokens_seen": 627245056
},
{
"epoch": 0.19,
"learning_rate": 0.0004090672016048145,
"loss": 2.8602,
"theoretical_loss": 3.8221828280011065,
"tokens_seen": 627310592
},
{
"epoch": 0.19,
"learning_rate": 0.0004090571715145436,
"loss": 2.9019,
"theoretical_loss": 3.8221414893906616,
"tokens_seen": 627376128
},
{
"epoch": 0.19,
"learning_rate": 0.00040904714142427285,
"loss": 2.9343,
"theoretical_loss": 3.822100156307208,
"tokens_seen": 627441664
},
{
"epoch": 0.19,
"objective/train/docs_used": 364826,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7832794189453125,
"objective/train/theoretical_loss": 3.8220588287494284,
"objective/train/tokens_used": 647967200,
"theoretical_loss": 3.8220588287494284,
"tokens_seen": 627507200
},
{
"epoch": 0.19,
"learning_rate": 0.00040903711133400203,
"loss": 2.8941,
"theoretical_loss": 3.8220588287494284,
"tokens_seen": 627507200
},
{
"epoch": 0.19,
"learning_rate": 0.0004090270812437312,
"loss": 3.0035,
"theoretical_loss": 3.8220175067160085,
"tokens_seen": 627572736
},
{
"epoch": 0.19,
"learning_rate": 0.0004090170511534604,
"loss": 2.9709,
"theoretical_loss": 3.821976190205632,
"tokens_seen": 627638272
},
{
"epoch": 0.19,
"learning_rate": 0.00040900702106318957,
"loss": 2.8026,
"theoretical_loss": 3.821934879216985,
"tokens_seen": 627703808
},
{
"epoch": 0.19,
"learning_rate": 0.00040899699097291875,
"loss": 2.8184,
"theoretical_loss": 3.8218935737487527,
"tokens_seen": 627769344
},
{
"epoch": 0.19,
"learning_rate": 0.000408986960882648,
"loss": 2.7926,
"theoretical_loss": 3.821852273799621,
"tokens_seen": 627834880
},
{
"epoch": 0.19,
"learning_rate": 0.0004089769307923771,
"loss": 2.8435,
"theoretical_loss": 3.8218109793682773,
"tokens_seen": 627900416
},
{
"epoch": 0.19,
"learning_rate": 0.00040896690070210635,
"loss": 2.7916,
"theoretical_loss": 3.821769690453407,
"tokens_seen": 627965952
},
{
"epoch": 0.19,
"learning_rate": 0.00040895687061183553,
"loss": 3.0103,
"theoretical_loss": 3.8217284070536994,
"tokens_seen": 628031488
},
{
"epoch": 0.19,
"learning_rate": 0.0004089468405215647,
"loss": 2.9298,
"theoretical_loss": 3.8216871291678416,
"tokens_seen": 628097024
},
{
"epoch": 0.19,
"learning_rate": 0.0004089368104312939,
"loss": 2.8847,
"theoretical_loss": 3.8216458567945217,
"tokens_seen": 628162560
},
{
"epoch": 0.19,
"learning_rate": 0.0004089267803410231,
"loss": 2.7306,
"theoretical_loss": 3.821604589932429,
"tokens_seen": 628228096
},
{
"epoch": 0.19,
"learning_rate": 0.00040891675025075226,
"loss": 3.0658,
"theoretical_loss": 3.821563328580253,
"tokens_seen": 628293632
},
{
"epoch": 0.19,
"learning_rate": 0.0004089067201604815,
"loss": 2.9798,
"theoretical_loss": 3.8215220727366823,
"tokens_seen": 628359168
},
{
"epoch": 0.19,
"learning_rate": 0.0004088966900702106,
"loss": 2.7955,
"theoretical_loss": 3.8214808224004075,
"tokens_seen": 628424704
},
{
"epoch": 0.19,
"learning_rate": 0.00040888665997993985,
"loss": 2.8123,
"theoretical_loss": 3.82143957757012,
"tokens_seen": 628490240
},
{
"epoch": 0.19,
"learning_rate": 0.000408876629889669,
"loss": 2.7577,
"theoretical_loss": 3.82139833824451,
"tokens_seen": 628555776
},
{
"epoch": 0.19,
"learning_rate": 0.0004088665997993982,
"loss": 2.6647,
"theoretical_loss": 3.8213571044222694,
"tokens_seen": 628621312
},
{
"epoch": 0.19,
"learning_rate": 0.0004088565697091274,
"loss": 2.8947,
"theoretical_loss": 3.82131587610209,
"tokens_seen": 628686848
},
{
"epoch": 0.19,
"learning_rate": 0.0004088465396188566,
"loss": 2.7107,
"theoretical_loss": 3.8212746532826642,
"tokens_seen": 628752384
},
{
"epoch": 0.19,
"learning_rate": 0.00040883650952858576,
"loss": 2.9835,
"theoretical_loss": 3.821233435962685,
"tokens_seen": 628817920
},
{
"epoch": 0.19,
"learning_rate": 0.000408826479438315,
"loss": 2.8205,
"theoretical_loss": 3.8211922241408454,
"tokens_seen": 628883456
},
{
"epoch": 0.19,
"learning_rate": 0.0004088164493480441,
"loss": 2.8666,
"theoretical_loss": 3.821151017815839,
"tokens_seen": 628948992
},
{
"epoch": 0.19,
"learning_rate": 0.00040880641925777336,
"loss": 2.8553,
"theoretical_loss": 3.8211098169863607,
"tokens_seen": 629014528
},
{
"epoch": 0.19,
"learning_rate": 0.0004087963891675025,
"loss": 2.689,
"theoretical_loss": 3.8210686216511043,
"tokens_seen": 629080064
},
{
"epoch": 0.19,
"objective/train/docs_used": 365375,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6356372833251953,
"objective/train/theoretical_loss": 3.8210274318087656,
"objective/train/tokens_used": 649605600,
"theoretical_loss": 3.8210274318087656,
"tokens_seen": 629145600
},
{
"epoch": 0.19,
"learning_rate": 0.0004087863590772317,
"loss": 2.6537,
"theoretical_loss": 3.8210274318087656,
"tokens_seen": 629145600
},
{
"epoch": 0.19,
"learning_rate": 0.0004087763289869609,
"loss": 2.9341,
"theoretical_loss": 3.8209862474580394,
"tokens_seen": 629211136
},
{
"epoch": 0.19,
"learning_rate": 0.0004087662988966901,
"loss": 2.9428,
"theoretical_loss": 3.8209450685976227,
"tokens_seen": 629276672
},
{
"epoch": 0.19,
"learning_rate": 0.00040875626880641926,
"loss": 2.7736,
"theoretical_loss": 3.820903895226211,
"tokens_seen": 629342208
},
{
"epoch": 0.19,
"learning_rate": 0.00040874623871614844,
"loss": 2.9722,
"theoretical_loss": 3.820862727342501,
"tokens_seen": 629407744
},
{
"epoch": 0.19,
"learning_rate": 0.0004087362086258776,
"loss": 3.2209,
"theoretical_loss": 3.8208215649451907,
"tokens_seen": 629473280
},
{
"epoch": 0.19,
"learning_rate": 0.00040872617853560686,
"loss": 3.0297,
"theoretical_loss": 3.8207804080329772,
"tokens_seen": 629538816
},
{
"epoch": 0.19,
"learning_rate": 0.000408716148445336,
"loss": 2.4412,
"theoretical_loss": 3.8207392566045595,
"tokens_seen": 629604352
},
{
"epoch": 0.19,
"learning_rate": 0.0004087061183550652,
"loss": 2.7719,
"theoretical_loss": 3.8206981106586357,
"tokens_seen": 629669888
},
{
"epoch": 0.19,
"learning_rate": 0.00040869608826479435,
"loss": 2.9634,
"theoretical_loss": 3.820656970193905,
"tokens_seen": 629735424
},
{
"epoch": 0.19,
"learning_rate": 0.0004086860581745236,
"loss": 3.0284,
"theoretical_loss": 3.8206158352090664,
"tokens_seen": 629800960
},
{
"epoch": 0.19,
"learning_rate": 0.00040867602808425277,
"loss": 2.9308,
"theoretical_loss": 3.820574705702821,
"tokens_seen": 629866496
},
{
"epoch": 0.19,
"learning_rate": 0.00040866599799398195,
"loss": 2.8229,
"theoretical_loss": 3.8205335816738684,
"tokens_seen": 629932032
},
{
"epoch": 0.19,
"learning_rate": 0.00040865596790371113,
"loss": 2.858,
"theoretical_loss": 3.8204924631209094,
"tokens_seen": 629997568
},
{
"epoch": 0.19,
"learning_rate": 0.00040864593781344036,
"loss": 2.7671,
"theoretical_loss": 3.820451350042646,
"tokens_seen": 630063104
},
{
"epoch": 0.19,
"learning_rate": 0.0004086359077231695,
"loss": 2.662,
"theoretical_loss": 3.820410242437779,
"tokens_seen": 630128640
},
{
"epoch": 0.19,
"learning_rate": 0.0004086258776328987,
"loss": 3.0353,
"theoretical_loss": 3.8203691403050106,
"tokens_seen": 630194176
},
{
"epoch": 0.19,
"learning_rate": 0.00040861584754262785,
"loss": 2.8499,
"theoretical_loss": 3.8203280436430447,
"tokens_seen": 630259712
},
{
"epoch": 0.19,
"learning_rate": 0.0004086058174523571,
"loss": 2.6793,
"theoretical_loss": 3.8202869524505827,
"tokens_seen": 630325248
},
{
"epoch": 0.19,
"learning_rate": 0.00040859578736208627,
"loss": 2.764,
"theoretical_loss": 3.8202458667263297,
"tokens_seen": 630390784
},
{
"epoch": 0.19,
"learning_rate": 0.00040858575727181545,
"loss": 3.1935,
"theoretical_loss": 3.8202047864689885,
"tokens_seen": 630456320
},
{
"epoch": 0.19,
"learning_rate": 0.00040857572718154463,
"loss": 3.0614,
"theoretical_loss": 3.8201637116772638,
"tokens_seen": 630521856
},
{
"epoch": 0.19,
"learning_rate": 0.0004085656970912738,
"loss": 2.7295,
"theoretical_loss": 3.820122642349861,
"tokens_seen": 630587392
},
{
"epoch": 0.19,
"learning_rate": 0.000408555667001003,
"loss": 2.9255,
"theoretical_loss": 3.8200815784854845,
"tokens_seen": 630652928
},
{
"epoch": 0.19,
"learning_rate": 0.00040854563691073223,
"loss": 2.8853,
"theoretical_loss": 3.8200405200828405,
"tokens_seen": 630718464
},
{
"epoch": 0.19,
"objective/train/docs_used": 366612,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8384101390838623,
"objective/train/theoretical_loss": 3.8199994671406348,
"objective/train/tokens_used": 651244000,
"theoretical_loss": 3.8199994671406348,
"tokens_seen": 630784000
},
{
"epoch": 0.19,
"learning_rate": 0.00040853560682046136,
"loss": 2.7613,
"theoretical_loss": 3.8199994671406348,
"tokens_seen": 630784000
},
{
"epoch": 0.19,
"learning_rate": 0.0004085255767301906,
"loss": 2.9038,
"theoretical_loss": 3.819958419657574,
"tokens_seen": 630849536
},
{
"epoch": 0.19,
"learning_rate": 0.0004085155466399197,
"loss": 2.9814,
"theoretical_loss": 3.8199173776323665,
"tokens_seen": 630915072
},
{
"epoch": 0.19,
"learning_rate": 0.00040850551654964895,
"loss": 2.8793,
"theoretical_loss": 3.819876341063718,
"tokens_seen": 630980608
},
{
"epoch": 0.19,
"learning_rate": 0.00040849548645937814,
"loss": 2.7772,
"theoretical_loss": 3.8198353099503373,
"tokens_seen": 631046144
},
{
"epoch": 0.19,
"learning_rate": 0.0004084854563691073,
"loss": 2.9659,
"theoretical_loss": 3.8197942842909325,
"tokens_seen": 631111680
},
{
"epoch": 0.19,
"learning_rate": 0.0004084754262788365,
"loss": 3.0307,
"theoretical_loss": 3.8197532640842127,
"tokens_seen": 631177216
},
{
"epoch": 0.19,
"learning_rate": 0.00040846539618856573,
"loss": 3.0699,
"theoretical_loss": 3.8197122493288864,
"tokens_seen": 631242752
},
{
"epoch": 0.19,
"learning_rate": 0.00040845536609829486,
"loss": 2.6885,
"theoretical_loss": 3.819671240023664,
"tokens_seen": 631308288
},
{
"epoch": 0.19,
"learning_rate": 0.0004084453360080241,
"loss": 2.8712,
"theoretical_loss": 3.8196302361672556,
"tokens_seen": 631373824
},
{
"epoch": 0.19,
"learning_rate": 0.0004084353059177532,
"loss": 2.7389,
"theoretical_loss": 3.8195892377583713,
"tokens_seen": 631439360
},
{
"epoch": 0.19,
"learning_rate": 0.00040842527582748246,
"loss": 2.9399,
"theoretical_loss": 3.8195482447957225,
"tokens_seen": 631504896
},
{
"epoch": 0.19,
"learning_rate": 0.00040841524573721164,
"loss": 2.932,
"theoretical_loss": 3.819507257278021,
"tokens_seen": 631570432
},
{
"epoch": 0.19,
"learning_rate": 0.0004084052156469408,
"loss": 2.8941,
"theoretical_loss": 3.8194662752039776,
"tokens_seen": 631635968
},
{
"epoch": 0.19,
"learning_rate": 0.00040839518555667005,
"loss": 2.5435,
"theoretical_loss": 3.8194252985723054,
"tokens_seen": 631701504
},
{
"epoch": 0.19,
"learning_rate": 0.0004083851554663992,
"loss": 2.7923,
"theoretical_loss": 3.8193843273817167,
"tokens_seen": 631767040
},
{
"epoch": 0.19,
"learning_rate": 0.0004083751253761284,
"loss": 2.8786,
"theoretical_loss": 3.819343361630925,
"tokens_seen": 631832576
},
{
"epoch": 0.19,
"learning_rate": 0.0004083650952858576,
"loss": 2.6148,
"theoretical_loss": 3.8193024013186445,
"tokens_seen": 631898112
},
{
"epoch": 0.19,
"learning_rate": 0.0004083550651955868,
"loss": 3.0115,
"theoretical_loss": 3.819261446443588,
"tokens_seen": 631963648
},
{
"epoch": 0.19,
"learning_rate": 0.00040834503510531596,
"loss": 2.8173,
"theoretical_loss": 3.819220497004471,
"tokens_seen": 632029184
},
{
"epoch": 0.19,
"learning_rate": 0.0004083350050150452,
"loss": 2.8355,
"theoretical_loss": 3.8191795530000077,
"tokens_seen": 632094720
},
{
"epoch": 0.19,
"learning_rate": 0.0004083249749247743,
"loss": 3.0112,
"theoretical_loss": 3.8191386144289137,
"tokens_seen": 632160256
},
{
"epoch": 0.19,
"learning_rate": 0.00040831494483450356,
"loss": 2.8569,
"theoretical_loss": 3.8190976812899056,
"tokens_seen": 632225792
},
{
"epoch": 0.19,
"learning_rate": 0.0004083049147442327,
"loss": 3.1,
"theoretical_loss": 3.819056753581699,
"tokens_seen": 632291328
},
{
"epoch": 0.19,
"learning_rate": 0.0004082948846539619,
"loss": 2.8298,
"theoretical_loss": 3.8190158313030103,
"tokens_seen": 632356864
},
{
"epoch": 0.19,
"objective/train/docs_used": 367323,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4437427520751953,
"objective/train/theoretical_loss": 3.818974914452557,
"objective/train/tokens_used": 652882400,
"theoretical_loss": 3.818974914452557,
"tokens_seen": 632422400
},
{
"epoch": 0.19,
"learning_rate": 0.0004082848545636911,
"loss": 3.038,
"theoretical_loss": 3.818974914452557,
"tokens_seen": 632422400
},
{
"epoch": 0.19,
"learning_rate": 0.0004082748244734203,
"loss": 2.9644,
"theoretical_loss": 3.8189340030290566,
"tokens_seen": 632487936
},
{
"epoch": 0.19,
"learning_rate": 0.00040826479438314946,
"loss": 2.883,
"theoretical_loss": 3.8188930970312267,
"tokens_seen": 632553472
},
{
"epoch": 0.19,
"learning_rate": 0.00040825476429287864,
"loss": 2.8525,
"theoretical_loss": 3.8188521964577866,
"tokens_seen": 632619008
},
{
"epoch": 0.19,
"learning_rate": 0.0004082447342026078,
"loss": 3.0628,
"theoretical_loss": 3.8188113013074547,
"tokens_seen": 632684544
},
{
"epoch": 0.19,
"learning_rate": 0.00040823470411233706,
"loss": 2.7063,
"theoretical_loss": 3.8187704115789503,
"tokens_seen": 632750080
},
{
"epoch": 0.19,
"learning_rate": 0.0004082246740220662,
"loss": 2.9827,
"theoretical_loss": 3.8187295272709934,
"tokens_seen": 632815616
},
{
"epoch": 0.19,
"learning_rate": 0.0004082146439317954,
"loss": 2.714,
"theoretical_loss": 3.8186886483823033,
"tokens_seen": 632881152
},
{
"epoch": 0.19,
"learning_rate": 0.00040820461384152455,
"loss": 3.1194,
"theoretical_loss": 3.818647774911601,
"tokens_seen": 632946688
},
{
"epoch": 0.19,
"learning_rate": 0.0004081945837512538,
"loss": 2.8226,
"theoretical_loss": 3.8186069068576085,
"tokens_seen": 633012224
},
{
"epoch": 0.19,
"learning_rate": 0.00040818455366098297,
"loss": 2.7235,
"theoretical_loss": 3.818566044219046,
"tokens_seen": 633077760
},
{
"epoch": 0.19,
"learning_rate": 0.00040817452357071215,
"loss": 2.9174,
"theoretical_loss": 3.818525186994636,
"tokens_seen": 633143296
},
{
"epoch": 0.19,
"learning_rate": 0.00040816449348044133,
"loss": 2.9005,
"theoretical_loss": 3.8184843351831006,
"tokens_seen": 633208832
},
{
"epoch": 0.19,
"learning_rate": 0.00040815446339017056,
"loss": 2.8442,
"theoretical_loss": 3.8184434887831626,
"tokens_seen": 633274368
},
{
"epoch": 0.19,
"learning_rate": 0.0004081444332998997,
"loss": 2.8236,
"theoretical_loss": 3.8184026477935458,
"tokens_seen": 633339904
},
{
"epoch": 0.19,
"learning_rate": 0.0004081344032096289,
"loss": 2.5517,
"theoretical_loss": 3.8183618122129737,
"tokens_seen": 633405440
},
{
"epoch": 0.19,
"learning_rate": 0.00040812437311935805,
"loss": 3.1487,
"theoretical_loss": 3.818320982040169,
"tokens_seen": 633470976
},
{
"epoch": 0.19,
"learning_rate": 0.0004081143430290873,
"loss": 2.7316,
"theoretical_loss": 3.8182801572738585,
"tokens_seen": 633536512
},
{
"epoch": 0.19,
"learning_rate": 0.00040810431293881647,
"loss": 3.1412,
"theoretical_loss": 3.8182393379127655,
"tokens_seen": 633602048
},
{
"epoch": 0.19,
"learning_rate": 0.00040809428284854565,
"loss": 2.8875,
"theoretical_loss": 3.8181985239556155,
"tokens_seen": 633667584
},
{
"epoch": 0.19,
"learning_rate": 0.00040808425275827483,
"loss": 3.026,
"theoretical_loss": 3.8181577154011346,
"tokens_seen": 633733120
},
{
"epoch": 0.19,
"learning_rate": 0.000408074222668004,
"loss": 2.8875,
"theoretical_loss": 3.8181169122480494,
"tokens_seen": 633798656
},
{
"epoch": 0.19,
"learning_rate": 0.0004080641925777332,
"loss": 2.882,
"theoretical_loss": 3.818076114495086,
"tokens_seen": 633864192
},
{
"epoch": 0.19,
"learning_rate": 0.00040805416248746243,
"loss": 2.8508,
"theoretical_loss": 3.8180353221409717,
"tokens_seen": 633929728
},
{
"epoch": 0.19,
"learning_rate": 0.00040804413239719156,
"loss": 2.892,
"theoretical_loss": 3.817994535184434,
"tokens_seen": 633995264
},
{
"epoch": 0.19,
"objective/train/docs_used": 367883,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8706178665161133,
"objective/train/theoretical_loss": 3.8179537536242005,
"objective/train/tokens_used": 654520800,
"theoretical_loss": 3.8179537536242005,
"tokens_seen": 634060800
},
{
"epoch": 0.19,
"learning_rate": 0.0004080341023069208,
"loss": 2.9069,
"theoretical_loss": 3.8179537536242005,
"tokens_seen": 634060800
},
{
"epoch": 0.19,
"learning_rate": 0.0004080240722166499,
"loss": 2.9371,
"theoretical_loss": 3.8179129774590006,
"tokens_seen": 634126336
},
{
"epoch": 0.19,
"learning_rate": 0.00040801404212637915,
"loss": 2.7872,
"theoretical_loss": 3.8178722066875626,
"tokens_seen": 634191872
},
{
"epoch": 0.19,
"learning_rate": 0.00040800401203610834,
"loss": 2.8084,
"theoretical_loss": 3.8178314413086154,
"tokens_seen": 634257408
},
{
"epoch": 0.19,
"learning_rate": 0.0004079939819458375,
"loss": 2.8125,
"theoretical_loss": 3.817790681320889,
"tokens_seen": 634322944
},
{
"epoch": 0.19,
"learning_rate": 0.0004079839518555667,
"loss": 2.9918,
"theoretical_loss": 3.817749926723114,
"tokens_seen": 634388480
},
{
"epoch": 0.19,
"learning_rate": 0.00040797392176529593,
"loss": 3.0913,
"theoretical_loss": 3.8177091775140197,
"tokens_seen": 634454016
},
{
"epoch": 0.19,
"learning_rate": 0.00040796389167502506,
"loss": 3.0007,
"theoretical_loss": 3.8176684336923383,
"tokens_seen": 634519552
},
{
"epoch": 0.19,
"learning_rate": 0.0004079538615847543,
"loss": 2.9475,
"theoretical_loss": 3.8176276952568005,
"tokens_seen": 634585088
},
{
"epoch": 0.19,
"learning_rate": 0.0004079438314944834,
"loss": 2.7219,
"theoretical_loss": 3.8175869622061382,
"tokens_seen": 634650624
},
{
"epoch": 0.19,
"learning_rate": 0.00040793380140421266,
"loss": 2.8263,
"theoretical_loss": 3.8175462345390843,
"tokens_seen": 634716160
},
{
"epoch": 0.19,
"learning_rate": 0.00040792377131394184,
"loss": 2.8433,
"theoretical_loss": 3.817505512254371,
"tokens_seen": 634781696
},
{
"epoch": 0.19,
"learning_rate": 0.000407913741223671,
"loss": 2.7156,
"theoretical_loss": 3.8174647953507312,
"tokens_seen": 634847232
},
{
"epoch": 0.19,
"learning_rate": 0.0004079037111334002,
"loss": 2.9948,
"theoretical_loss": 3.817424083826899,
"tokens_seen": 634912768
},
{
"epoch": 0.19,
"learning_rate": 0.0004078936810431294,
"loss": 2.8635,
"theoretical_loss": 3.817383377681608,
"tokens_seen": 634978304
},
{
"epoch": 0.19,
"learning_rate": 0.00040788365095285856,
"loss": 2.8047,
"theoretical_loss": 3.8173426769135927,
"tokens_seen": 635043840
},
{
"epoch": 0.19,
"learning_rate": 0.0004078736208625878,
"loss": 2.8993,
"theoretical_loss": 3.8173019815215885,
"tokens_seen": 635109376
},
{
"epoch": 0.19,
"learning_rate": 0.0004078635907723169,
"loss": 2.8109,
"theoretical_loss": 3.8172612915043294,
"tokens_seen": 635174912
},
{
"epoch": 0.19,
"learning_rate": 0.00040785356068204616,
"loss": 2.8459,
"theoretical_loss": 3.8172206068605523,
"tokens_seen": 635240448
},
{
"epoch": 0.19,
"learning_rate": 0.0004078435305917753,
"loss": 2.9693,
"theoretical_loss": 3.8171799275889926,
"tokens_seen": 635305984
},
{
"epoch": 0.19,
"learning_rate": 0.0004078335005015045,
"loss": 2.8409,
"theoretical_loss": 3.8171392536883877,
"tokens_seen": 635371520
},
{
"epoch": 0.19,
"learning_rate": 0.0004078234704112337,
"loss": 2.6479,
"theoretical_loss": 3.8170985851574732,
"tokens_seen": 635437056
},
{
"epoch": 0.19,
"learning_rate": 0.0004078134403209629,
"loss": 2.7721,
"theoretical_loss": 3.8170579219949876,
"tokens_seen": 635502592
},
{
"epoch": 0.19,
"learning_rate": 0.00040780341023069207,
"loss": 2.959,
"theoretical_loss": 3.8170172641996682,
"tokens_seen": 635568128
},
{
"epoch": 0.19,
"learning_rate": 0.0004077933801404213,
"loss": 2.8975,
"theoretical_loss": 3.816976611770254,
"tokens_seen": 635633664
},
{
"epoch": 0.19,
"objective/train/docs_used": 369288,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.124577522277832,
"objective/train/theoretical_loss": 3.8169359647054835,
"objective/train/tokens_used": 656159200,
"theoretical_loss": 3.8169359647054835,
"tokens_seen": 635699200
},
{
"epoch": 0.19,
"learning_rate": 0.00040778335005015043,
"loss": 2.8436,
"theoretical_loss": 3.8169359647054835,
"tokens_seen": 635699200
},
{
"epoch": 0.19,
"learning_rate": 0.00040777331995987966,
"loss": 2.9538,
"theoretical_loss": 3.8168953230040947,
"tokens_seen": 635764736
},
{
"epoch": 0.19,
"learning_rate": 0.0004077632898696088,
"loss": 2.9674,
"theoretical_loss": 3.8168546866648283,
"tokens_seen": 635830272
},
{
"epoch": 0.19,
"learning_rate": 0.000407753259779338,
"loss": 3.0346,
"theoretical_loss": 3.816814055686424,
"tokens_seen": 635895808
},
{
"epoch": 0.19,
"learning_rate": 0.0004077432296890672,
"loss": 2.922,
"theoretical_loss": 3.816773430067622,
"tokens_seen": 635961344
},
{
"epoch": 0.19,
"learning_rate": 0.0004077331995987964,
"loss": 2.7976,
"theoretical_loss": 3.816732809807163,
"tokens_seen": 636026880
},
{
"epoch": 0.19,
"learning_rate": 0.00040772316950852557,
"loss": 2.849,
"theoretical_loss": 3.816692194903789,
"tokens_seen": 636092416
},
{
"epoch": 0.19,
"learning_rate": 0.00040771313941825475,
"loss": 2.9071,
"theoretical_loss": 3.816651585356241,
"tokens_seen": 636157952
},
{
"epoch": 0.19,
"learning_rate": 0.00040770310932798393,
"loss": 2.7778,
"theoretical_loss": 3.816610981163261,
"tokens_seen": 636223488
},
{
"epoch": 0.19,
"learning_rate": 0.00040769307923771317,
"loss": 2.7555,
"theoretical_loss": 3.816570382323592,
"tokens_seen": 636289024
},
{
"epoch": 0.19,
"learning_rate": 0.0004076830491474423,
"loss": 2.969,
"theoretical_loss": 3.8165297888359766,
"tokens_seen": 636354560
},
{
"epoch": 0.19,
"learning_rate": 0.00040767301905717153,
"loss": 2.778,
"theoretical_loss": 3.8164892006991584,
"tokens_seen": 636420096
},
{
"epoch": 0.19,
"learning_rate": 0.0004076629889669007,
"loss": 3.004,
"theoretical_loss": 3.816448617911881,
"tokens_seen": 636485632
},
{
"epoch": 0.19,
"learning_rate": 0.0004076529588766299,
"loss": 2.911,
"theoretical_loss": 3.816408040472889,
"tokens_seen": 636551168
},
{
"epoch": 0.19,
"learning_rate": 0.0004076429287863591,
"loss": 2.8413,
"theoretical_loss": 3.816367468380926,
"tokens_seen": 636616704
},
{
"epoch": 0.19,
"learning_rate": 0.00040763289869608825,
"loss": 2.788,
"theoretical_loss": 3.8163269016347385,
"tokens_seen": 636682240
},
{
"epoch": 0.19,
"learning_rate": 0.0004076228686058175,
"loss": 3.115,
"theoretical_loss": 3.8162863402330713,
"tokens_seen": 636747776
},
{
"epoch": 0.19,
"learning_rate": 0.00040761283851554667,
"loss": 2.6516,
"theoretical_loss": 3.81624578417467,
"tokens_seen": 636813312
},
{
"epoch": 0.19,
"learning_rate": 0.00040760280842527585,
"loss": 2.9217,
"theoretical_loss": 3.816205233458281,
"tokens_seen": 636878848
},
{
"epoch": 0.19,
"learning_rate": 0.00040759277833500503,
"loss": 2.7958,
"theoretical_loss": 3.816164688082652,
"tokens_seen": 636944384
},
{
"epoch": 0.19,
"learning_rate": 0.0004075827482447342,
"loss": 2.5185,
"theoretical_loss": 3.816124148046529,
"tokens_seen": 637009920
},
{
"epoch": 0.19,
"learning_rate": 0.0004075727181544634,
"loss": 3.0614,
"theoretical_loss": 3.81608361334866,
"tokens_seen": 637075456
},
{
"epoch": 0.19,
"learning_rate": 0.00040756268806419263,
"loss": 2.9743,
"theoretical_loss": 3.8160430839877937,
"tokens_seen": 637140992
},
{
"epoch": 0.19,
"learning_rate": 0.00040755265797392176,
"loss": 2.926,
"theoretical_loss": 3.8160025599626777,
"tokens_seen": 637206528
},
{
"epoch": 0.19,
"learning_rate": 0.000407542627883651,
"loss": 2.7422,
"theoretical_loss": 3.815962041272061,
"tokens_seen": 637272064
},
{
"epoch": 0.19,
"objective/train/docs_used": 369983,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8315579891204834,
"objective/train/theoretical_loss": 3.815921527914693,
"objective/train/tokens_used": 657797600,
"theoretical_loss": 3.815921527914693,
"tokens_seen": 637337600
},
{
"epoch": 0.19,
"learning_rate": 0.0004075325977933801,
"loss": 2.9318,
"theoretical_loss": 3.815921527914693,
"tokens_seen": 637337600
},
{
"epoch": 0.19,
"learning_rate": 0.00040752256770310935,
"loss": 2.6521,
"theoretical_loss": 3.815881019889324,
"tokens_seen": 637403136
},
{
"epoch": 0.19,
"learning_rate": 0.00040751253761283854,
"loss": 2.8127,
"theoretical_loss": 3.8158405171947027,
"tokens_seen": 637468672
},
{
"epoch": 0.19,
"learning_rate": 0.0004075025075225677,
"loss": 3.033,
"theoretical_loss": 3.8158000198295814,
"tokens_seen": 637534208
},
{
"epoch": 0.19,
"learning_rate": 0.0004074924774322969,
"loss": 3.1116,
"theoretical_loss": 3.81575952779271,
"tokens_seen": 637599744
},
{
"epoch": 0.19,
"learning_rate": 0.00040748244734202613,
"loss": 2.9631,
"theoretical_loss": 3.81571904108284,
"tokens_seen": 637665280
},
{
"epoch": 0.19,
"learning_rate": 0.00040747241725175526,
"loss": 2.7844,
"theoretical_loss": 3.815678559698723,
"tokens_seen": 637730816
},
{
"epoch": 0.19,
"learning_rate": 0.0004074623871614845,
"loss": 2.7158,
"theoretical_loss": 3.8156380836391124,
"tokens_seen": 637796352
},
{
"epoch": 0.19,
"learning_rate": 0.0004074523570712136,
"loss": 2.7155,
"theoretical_loss": 3.8155976129027596,
"tokens_seen": 637861888
},
{
"epoch": 0.19,
"learning_rate": 0.00040744232698094286,
"loss": 3.0124,
"theoretical_loss": 3.815557147488418,
"tokens_seen": 637927424
},
{
"epoch": 0.19,
"learning_rate": 0.00040743229689067204,
"loss": 2.8726,
"theoretical_loss": 3.8155166873948416,
"tokens_seen": 637992960
},
{
"epoch": 0.19,
"learning_rate": 0.0004074222668004012,
"loss": 3.0371,
"theoretical_loss": 3.815476232620784,
"tokens_seen": 638058496
},
{
"epoch": 0.19,
"learning_rate": 0.0004074122367101304,
"loss": 2.7492,
"theoretical_loss": 3.8154357831649994,
"tokens_seen": 638124032
},
{
"epoch": 0.19,
"learning_rate": 0.0004074022066198596,
"loss": 2.8076,
"theoretical_loss": 3.8153953390262427,
"tokens_seen": 638189568
},
{
"epoch": 0.19,
"learning_rate": 0.00040739217652958876,
"loss": 2.772,
"theoretical_loss": 3.815354900203269,
"tokens_seen": 638255104
},
{
"epoch": 0.19,
"learning_rate": 0.000407382146439318,
"loss": 2.7375,
"theoretical_loss": 3.8153144666948338,
"tokens_seen": 638320640
},
{
"epoch": 0.19,
"learning_rate": 0.0004073721163490471,
"loss": 2.8696,
"theoretical_loss": 3.8152740384996937,
"tokens_seen": 638386176
},
{
"epoch": 0.19,
"learning_rate": 0.00040736208625877636,
"loss": 2.7808,
"theoretical_loss": 3.815233615616605,
"tokens_seen": 638451712
},
{
"epoch": 0.19,
"learning_rate": 0.0004073520561685055,
"loss": 2.7748,
"theoretical_loss": 3.815193198044324,
"tokens_seen": 638517248
},
{
"epoch": 0.19,
"learning_rate": 0.0004073420260782347,
"loss": 2.7993,
"theoretical_loss": 3.8151527857816077,
"tokens_seen": 638582784
},
{
"epoch": 0.19,
"learning_rate": 0.0004073319959879639,
"loss": 2.6915,
"theoretical_loss": 3.815112378827215,
"tokens_seen": 638648320
},
{
"epoch": 0.19,
"learning_rate": 0.0004073219658976931,
"loss": 3.0531,
"theoretical_loss": 3.815071977179903,
"tokens_seen": 638713856
},
{
"epoch": 0.19,
"learning_rate": 0.00040731193580742227,
"loss": 2.9624,
"theoretical_loss": 3.815031580838431,
"tokens_seen": 638779392
},
{
"epoch": 0.19,
"learning_rate": 0.0004073019057171515,
"loss": 3.0607,
"theoretical_loss": 3.814991189801557,
"tokens_seen": 638844928
},
{
"epoch": 0.19,
"learning_rate": 0.00040729187562688063,
"loss": 2.6183,
"theoretical_loss": 3.8149508040680415,
"tokens_seen": 638910464
},
{
"epoch": 0.19,
"objective/train/docs_used": 370882,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.928964853286743,
"objective/train/theoretical_loss": 3.8149104236366433,
"objective/train/tokens_used": 659436000,
"theoretical_loss": 3.8149104236366433,
"tokens_seen": 638976000
},
{
"epoch": 0.19,
"learning_rate": 0.00040728184553660986,
"loss": 2.8016,
"theoretical_loss": 3.8149104236366433,
"tokens_seen": 638976000
},
{
"epoch": 0.19,
"learning_rate": 0.000407271815446339,
"loss": 2.7656,
"theoretical_loss": 3.8148700485061235,
"tokens_seen": 639041536
},
{
"epoch": 0.19,
"learning_rate": 0.0004072617853560682,
"loss": 2.8697,
"theoretical_loss": 3.814829678675242,
"tokens_seen": 639107072
},
{
"epoch": 0.19,
"learning_rate": 0.0004072517552657974,
"loss": 2.848,
"theoretical_loss": 3.81478931414276,
"tokens_seen": 639172608
},
{
"epoch": 0.19,
"learning_rate": 0.0004072417251755266,
"loss": 2.922,
"theoretical_loss": 3.814748954907439,
"tokens_seen": 639238144
},
{
"epoch": 0.19,
"learning_rate": 0.00040723169508525577,
"loss": 3.021,
"theoretical_loss": 3.814708600968041,
"tokens_seen": 639303680
},
{
"epoch": 0.19,
"learning_rate": 0.00040722166499498495,
"loss": 2.9457,
"theoretical_loss": 3.8146682523233286,
"tokens_seen": 639369216
},
{
"epoch": 0.19,
"learning_rate": 0.00040721163490471413,
"loss": 2.967,
"theoretical_loss": 3.8146279089720636,
"tokens_seen": 639434752
},
{
"epoch": 0.19,
"learning_rate": 0.00040720160481444337,
"loss": 3.1371,
"theoretical_loss": 3.8145875709130093,
"tokens_seen": 639500288
},
{
"epoch": 0.19,
"learning_rate": 0.0004071915747241725,
"loss": 2.9063,
"theoretical_loss": 3.8145472381449297,
"tokens_seen": 639565824
},
{
"epoch": 0.19,
"learning_rate": 0.00040718154463390173,
"loss": 2.898,
"theoretical_loss": 3.814506910666589,
"tokens_seen": 639631360
},
{
"epoch": 0.19,
"learning_rate": 0.0004071715145436309,
"loss": 2.8287,
"theoretical_loss": 3.814466588476751,
"tokens_seen": 639696896
},
{
"epoch": 0.19,
"learning_rate": 0.0004071614844533601,
"loss": 2.9777,
"theoretical_loss": 3.81442627157418,
"tokens_seen": 639762432
},
{
"epoch": 0.19,
"learning_rate": 0.00040715145436308927,
"loss": 2.9235,
"theoretical_loss": 3.814385959957643,
"tokens_seen": 639827968
},
{
"epoch": 0.19,
"learning_rate": 0.00040714142427281845,
"loss": 2.8904,
"theoretical_loss": 3.8143456536259035,
"tokens_seen": 639893504
},
{
"epoch": 0.19,
"learning_rate": 0.00040713139418254763,
"loss": 2.8318,
"theoretical_loss": 3.814305352577729,
"tokens_seen": 639959040
},
{
"epoch": 0.19,
"learning_rate": 0.00040712136409227687,
"loss": 3.0163,
"theoretical_loss": 3.8142650568118848,
"tokens_seen": 640024576
},
{
"epoch": 0.19,
"learning_rate": 0.000407111334002006,
"loss": 2.8352,
"theoretical_loss": 3.8142247663271385,
"tokens_seen": 640090112
},
{
"epoch": 0.19,
"learning_rate": 0.00040710130391173523,
"loss": 2.7951,
"theoretical_loss": 3.814184481122257,
"tokens_seen": 640155648
},
{
"epoch": 0.19,
"learning_rate": 0.00040709127382146436,
"loss": 2.72,
"theoretical_loss": 3.8141442011960085,
"tokens_seen": 640221184
},
{
"epoch": 0.19,
"learning_rate": 0.0004070812437311936,
"loss": 2.7045,
"theoretical_loss": 3.814103926547161,
"tokens_seen": 640286720
},
{
"epoch": 0.19,
"learning_rate": 0.0004070712136409228,
"loss": 3.0761,
"theoretical_loss": 3.8140636571744824,
"tokens_seen": 640352256
},
{
"epoch": 0.19,
"learning_rate": 0.00040706118355065196,
"loss": 2.9509,
"theoretical_loss": 3.8140233930767424,
"tokens_seen": 640417792
},
{
"epoch": 0.19,
"learning_rate": 0.00040705115346038114,
"loss": 3.1579,
"theoretical_loss": 3.8139831342527093,
"tokens_seen": 640483328
},
{
"epoch": 0.19,
"learning_rate": 0.0004070411233701103,
"loss": 2.7903,
"theoretical_loss": 3.8139428807011537,
"tokens_seen": 640548864
},
{
"epoch": 0.19,
"objective/train/docs_used": 371543,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0356507301330566,
"objective/train/theoretical_loss": 3.8139026324208456,
"objective/train/tokens_used": 661074400,
"theoretical_loss": 3.8139026324208456,
"tokens_seen": 640614400
},
{
"epoch": 0.19,
"learning_rate": 0.0004070310932798395,
"loss": 2.8759,
"theoretical_loss": 3.8139026324208456,
"tokens_seen": 640614400
},
{
"epoch": 0.19,
"learning_rate": 0.00040702106318956874,
"loss": 2.6784,
"theoretical_loss": 3.8138623894105557,
"tokens_seen": 640679936
},
{
"epoch": 0.19,
"learning_rate": 0.00040701103309929786,
"loss": 2.9655,
"theoretical_loss": 3.8138221516690547,
"tokens_seen": 640745472
},
{
"epoch": 0.19,
"learning_rate": 0.0004070010030090271,
"loss": 2.9818,
"theoretical_loss": 3.813781919195114,
"tokens_seen": 640811008
},
{
"epoch": 0.19,
"learning_rate": 0.0004069909729187563,
"loss": 2.8611,
"theoretical_loss": 3.813741691987506,
"tokens_seen": 640876544
},
{
"epoch": 0.19,
"learning_rate": 0.00040698094282848546,
"loss": 2.7041,
"theoretical_loss": 3.813701470045001,
"tokens_seen": 640942080
},
{
"epoch": 0.19,
"learning_rate": 0.00040697091273821464,
"loss": 2.7796,
"theoretical_loss": 3.8136612533663747,
"tokens_seen": 641007616
},
{
"epoch": 0.19,
"learning_rate": 0.0004069608826479438,
"loss": 2.6885,
"theoretical_loss": 3.8136210419503977,
"tokens_seen": 641073152
},
{
"epoch": 0.19,
"learning_rate": 0.000406950852557673,
"loss": 3.197,
"theoretical_loss": 3.813580835795844,
"tokens_seen": 641138688
},
{
"epoch": 0.19,
"learning_rate": 0.00040694082246740224,
"loss": 3.0585,
"theoretical_loss": 3.8135406349014884,
"tokens_seen": 641204224
},
{
"epoch": 0.19,
"learning_rate": 0.00040693079237713137,
"loss": 3.0909,
"theoretical_loss": 3.8135004392661047,
"tokens_seen": 641269760
},
{
"epoch": 0.19,
"learning_rate": 0.0004069207622868606,
"loss": 2.9761,
"theoretical_loss": 3.8134602488884664,
"tokens_seen": 641335296
},
{
"epoch": 0.19,
"learning_rate": 0.00040691073219658973,
"loss": 2.8035,
"theoretical_loss": 3.8134200637673503,
"tokens_seen": 641400832
},
{
"epoch": 0.19,
"learning_rate": 0.00040690070210631896,
"loss": 3.0324,
"theoretical_loss": 3.813379883901531,
"tokens_seen": 641466368
},
{
"epoch": 0.19,
"learning_rate": 0.0004068906720160482,
"loss": 2.9224,
"theoretical_loss": 3.8133397092897843,
"tokens_seen": 641531904
},
{
"epoch": 0.19,
"learning_rate": 0.0004068806419257773,
"loss": 2.6653,
"theoretical_loss": 3.8132995399308873,
"tokens_seen": 641597440
},
{
"epoch": 0.19,
"learning_rate": 0.00040687061183550656,
"loss": 2.8266,
"theoretical_loss": 3.813259375823616,
"tokens_seen": 641662976
},
{
"epoch": 0.19,
"learning_rate": 0.0004068605817452357,
"loss": 2.8396,
"theoretical_loss": 3.8132192169667483,
"tokens_seen": 641728512
},
{
"epoch": 0.19,
"learning_rate": 0.0004068505516549649,
"loss": 3.1059,
"theoretical_loss": 3.813179063359061,
"tokens_seen": 641794048
},
{
"epoch": 0.19,
"learning_rate": 0.0004068405215646941,
"loss": 2.8498,
"theoretical_loss": 3.813138914999332,
"tokens_seen": 641859584
},
{
"epoch": 0.19,
"learning_rate": 0.0004068304914744233,
"loss": 2.9668,
"theoretical_loss": 3.8130987718863407,
"tokens_seen": 641925120
},
{
"epoch": 0.19,
"learning_rate": 0.00040682046138415247,
"loss": 2.7385,
"theoretical_loss": 3.813058634018865,
"tokens_seen": 641990656
},
{
"epoch": 0.19,
"learning_rate": 0.0004068104312938817,
"loss": 2.9445,
"theoretical_loss": 3.8130185013956837,
"tokens_seen": 642056192
},
{
"epoch": 0.19,
"learning_rate": 0.00040680040120361083,
"loss": 3.2271,
"theoretical_loss": 3.812978374015578,
"tokens_seen": 642121728
},
{
"epoch": 0.19,
"learning_rate": 0.00040679037111334006,
"loss": 3.0666,
"theoretical_loss": 3.8129382518773265,
"tokens_seen": 642187264
},
{
"epoch": 0.19,
"objective/train/docs_used": 373013,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.800529718399048,
"objective/train/theoretical_loss": 3.8128981349797098,
"objective/train/tokens_used": 662712800,
"theoretical_loss": 3.8128981349797098,
"tokens_seen": 642252800
},
{
"epoch": 0.19,
"learning_rate": 0.0004067803410230692,
"loss": 2.8899,
"theoretical_loss": 3.8128981349797098,
"tokens_seen": 642252800
},
{
"epoch": 0.19,
"learning_rate": 0.0004067703109327984,
"loss": 2.6683,
"theoretical_loss": 3.8128580233215095,
"tokens_seen": 642318336
},
{
"epoch": 0.19,
"learning_rate": 0.0004067602808425276,
"loss": 2.9476,
"theoretical_loss": 3.8128179169015057,
"tokens_seen": 642383872
},
{
"epoch": 0.19,
"learning_rate": 0.0004067502507522568,
"loss": 3.0143,
"theoretical_loss": 3.8127778157184817,
"tokens_seen": 642449408
},
{
"epoch": 0.19,
"learning_rate": 0.00040674022066198597,
"loss": 2.8827,
"theoretical_loss": 3.8127377197712176,
"tokens_seen": 642514944
},
{
"epoch": 0.19,
"learning_rate": 0.00040673019057171515,
"loss": 2.9197,
"theoretical_loss": 3.8126976290584973,
"tokens_seen": 642580480
},
{
"epoch": 0.19,
"learning_rate": 0.00040672016048144433,
"loss": 2.6803,
"theoretical_loss": 3.812657543579103,
"tokens_seen": 642646016
},
{
"epoch": 0.19,
"learning_rate": 0.00040671013039117357,
"loss": 2.8742,
"theoretical_loss": 3.812617463331818,
"tokens_seen": 642711552
},
{
"epoch": 0.19,
"learning_rate": 0.0004067001003009027,
"loss": 2.8807,
"theoretical_loss": 3.812577388315426,
"tokens_seen": 642777088
},
{
"epoch": 0.19,
"learning_rate": 0.00040669007021063193,
"loss": 3.1897,
"theoretical_loss": 3.8125373185287117,
"tokens_seen": 642842624
},
{
"epoch": 0.19,
"learning_rate": 0.0004066800401203611,
"loss": 2.9624,
"theoretical_loss": 3.812497253970459,
"tokens_seen": 642908160
},
{
"epoch": 0.19,
"learning_rate": 0.0004066700100300903,
"loss": 2.9739,
"theoretical_loss": 3.8124571946394528,
"tokens_seen": 642973696
},
{
"epoch": 0.19,
"learning_rate": 0.00040665997993981947,
"loss": 2.8501,
"theoretical_loss": 3.812417140534478,
"tokens_seen": 643039232
},
{
"epoch": 0.19,
"learning_rate": 0.00040664994984954865,
"loss": 2.9424,
"theoretical_loss": 3.8123770916543216,
"tokens_seen": 643104768
},
{
"epoch": 0.19,
"learning_rate": 0.00040663991975927783,
"loss": 2.8588,
"theoretical_loss": 3.8123370479977687,
"tokens_seen": 643170304
},
{
"epoch": 0.19,
"learning_rate": 0.00040662988966900707,
"loss": 3.0731,
"theoretical_loss": 3.812297009563606,
"tokens_seen": 643235840
},
{
"epoch": 0.19,
"learning_rate": 0.0004066198595787362,
"loss": 3.135,
"theoretical_loss": 3.8122569763506204,
"tokens_seen": 643301376
},
{
"epoch": 0.19,
"learning_rate": 0.00040660982948846543,
"loss": 2.7244,
"theoretical_loss": 3.8122169483576,
"tokens_seen": 643366912
},
{
"epoch": 0.19,
"learning_rate": 0.00040659979939819456,
"loss": 2.6626,
"theoretical_loss": 3.812176925583331,
"tokens_seen": 643432448
},
{
"epoch": 0.19,
"learning_rate": 0.0004065897693079238,
"loss": 2.9799,
"theoretical_loss": 3.812136908026603,
"tokens_seen": 643497984
},
{
"epoch": 0.2,
"learning_rate": 0.000406579739217653,
"loss": 2.9877,
"theoretical_loss": 3.8120968956862034,
"tokens_seen": 643563520
},
{
"epoch": 0.2,
"learning_rate": 0.00040656970912738216,
"loss": 2.865,
"theoretical_loss": 3.8120568885609223,
"tokens_seen": 643629056
},
{
"epoch": 0.2,
"learning_rate": 0.00040655967903711134,
"loss": 2.8761,
"theoretical_loss": 3.8120168866495483,
"tokens_seen": 643694592
},
{
"epoch": 0.2,
"learning_rate": 0.0004065496489468405,
"loss": 2.7782,
"theoretical_loss": 3.8119768899508713,
"tokens_seen": 643760128
},
{
"epoch": 0.2,
"learning_rate": 0.0004065396188565697,
"loss": 2.9398,
"theoretical_loss": 3.8119368984636814,
"tokens_seen": 643825664
},
{
"epoch": 0.2,
"objective/train/docs_used": 373533,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.063260078430176,
"objective/train/theoretical_loss": 3.8118969121867696,
"objective/train/tokens_used": 664351200,
"theoretical_loss": 3.8118969121867696,
"tokens_seen": 643891200
},
{
"epoch": 0.2,
"learning_rate": 0.00040652958876629894,
"loss": 2.9474,
"theoretical_loss": 3.8118969121867696,
"tokens_seen": 643891200
},
{
"epoch": 0.2,
"learning_rate": 0.00040651955867602806,
"loss": 3.0836,
"theoretical_loss": 3.8118569311189265,
"tokens_seen": 643956736
},
{
"epoch": 0.2,
"learning_rate": 0.0004065095285857573,
"loss": 2.8049,
"theoretical_loss": 3.8118169552589434,
"tokens_seen": 644022272
},
{
"epoch": 0.2,
"learning_rate": 0.0004064994984954865,
"loss": 2.8489,
"theoretical_loss": 3.811776984605612,
"tokens_seen": 644087808
},
{
"epoch": 0.2,
"learning_rate": 0.00040648946840521566,
"loss": 2.953,
"theoretical_loss": 3.811737019157725,
"tokens_seen": 644153344
},
{
"epoch": 0.2,
"learning_rate": 0.00040647943831494484,
"loss": 3.0545,
"theoretical_loss": 3.811697058914075,
"tokens_seen": 644218880
},
{
"epoch": 0.2,
"learning_rate": 0.000406469408224674,
"loss": 2.8362,
"theoretical_loss": 3.8116571038734546,
"tokens_seen": 644284416
},
{
"epoch": 0.2,
"learning_rate": 0.0004064593781344032,
"loss": 2.8523,
"theoretical_loss": 3.811617154034657,
"tokens_seen": 644349952
},
{
"epoch": 0.2,
"learning_rate": 0.00040644934804413244,
"loss": 2.8711,
"theoretical_loss": 3.811577209396477,
"tokens_seen": 644415488
},
{
"epoch": 0.2,
"learning_rate": 0.00040643931795386157,
"loss": 2.8579,
"theoretical_loss": 3.8115372699577073,
"tokens_seen": 644481024
},
{
"epoch": 0.2,
"learning_rate": 0.0004064292878635908,
"loss": 3.2256,
"theoretical_loss": 3.8114973357171436,
"tokens_seen": 644546560
},
{
"epoch": 0.2,
"learning_rate": 0.00040641925777331993,
"loss": 2.8823,
"theoretical_loss": 3.8114574066735805,
"tokens_seen": 644612096
},
{
"epoch": 0.2,
"learning_rate": 0.00040640922768304916,
"loss": 2.837,
"theoretical_loss": 3.8114174828258136,
"tokens_seen": 644677632
},
{
"epoch": 0.2,
"learning_rate": 0.00040639919759277834,
"loss": 3.0595,
"theoretical_loss": 3.8113775641726386,
"tokens_seen": 644743168
},
{
"epoch": 0.2,
"learning_rate": 0.0004063891675025075,
"loss": 2.956,
"theoretical_loss": 3.8113376507128516,
"tokens_seen": 644808704
},
{
"epoch": 0.2,
"learning_rate": 0.0004063791374122367,
"loss": 2.7663,
"theoretical_loss": 3.81129774244525,
"tokens_seen": 644874240
},
{
"epoch": 0.2,
"learning_rate": 0.0004063691073219659,
"loss": 2.9184,
"theoretical_loss": 3.811257839368629,
"tokens_seen": 644939776
},
{
"epoch": 0.2,
"learning_rate": 0.00040635907723169507,
"loss": 2.7577,
"theoretical_loss": 3.8112179414817877,
"tokens_seen": 645005312
},
{
"epoch": 0.2,
"learning_rate": 0.0004063490471414243,
"loss": 2.9307,
"theoretical_loss": 3.811178048783524,
"tokens_seen": 645070848
},
{
"epoch": 0.2,
"learning_rate": 0.00040633901705115343,
"loss": 2.9966,
"theoretical_loss": 3.8111381612726345,
"tokens_seen": 645136384
},
{
"epoch": 0.2,
"learning_rate": 0.00040632898696088267,
"loss": 3.0551,
"theoretical_loss": 3.8110982789479193,
"tokens_seen": 645201920
},
{
"epoch": 0.2,
"learning_rate": 0.00040631895687061185,
"loss": 2.8291,
"theoretical_loss": 3.811058401808177,
"tokens_seen": 645267456
},
{
"epoch": 0.2,
"learning_rate": 0.00040630892678034103,
"loss": 2.9193,
"theoretical_loss": 3.8110185298522063,
"tokens_seen": 645332992
},
{
"epoch": 0.2,
"learning_rate": 0.0004062988966900702,
"loss": 2.9016,
"theoretical_loss": 3.8109786630788083,
"tokens_seen": 645398528
},
{
"epoch": 0.2,
"learning_rate": 0.0004062888665997994,
"loss": 3.1083,
"theoretical_loss": 3.8109388014867824,
"tokens_seen": 645464064
},
{
"epoch": 0.2,
"objective/train/docs_used": 374553,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.623807430267334,
"objective/train/theoretical_loss": 3.8108989450749293,
"objective/train/tokens_used": 665989600,
"theoretical_loss": 3.8108989450749293,
"tokens_seen": 645529600
},
{
"epoch": 0.2,
"learning_rate": 0.00040627883650952857,
"loss": 2.8018,
"theoretical_loss": 3.8108989450749293,
"tokens_seen": 645529600
},
{
"epoch": 0.2,
"learning_rate": 0.0004062688064192578,
"loss": 2.8829,
"theoretical_loss": 3.81085909384205,
"tokens_seen": 645595136
},
{
"epoch": 0.2,
"learning_rate": 0.00040625877632898693,
"loss": 2.9422,
"theoretical_loss": 3.810819247786946,
"tokens_seen": 645660672
},
{
"epoch": 0.2,
"learning_rate": 0.00040624874623871617,
"loss": 2.9232,
"theoretical_loss": 3.810779406908419,
"tokens_seen": 645726208
},
{
"epoch": 0.2,
"learning_rate": 0.0004062387161484453,
"loss": 2.9929,
"theoretical_loss": 3.8107395712052714,
"tokens_seen": 645791744
},
{
"epoch": 0.2,
"learning_rate": 0.00040622868605817453,
"loss": 2.9012,
"theoretical_loss": 3.810699740676306,
"tokens_seen": 645857280
},
{
"epoch": 0.2,
"learning_rate": 0.0004062186559679037,
"loss": 2.7108,
"theoretical_loss": 3.8106599153203256,
"tokens_seen": 645922816
},
{
"epoch": 0.2,
"learning_rate": 0.0004062086258776329,
"loss": 2.9696,
"theoretical_loss": 3.810620095136133,
"tokens_seen": 645988352
},
{
"epoch": 0.2,
"learning_rate": 0.0004061985957873621,
"loss": 2.7225,
"theoretical_loss": 3.810580280122533,
"tokens_seen": 646053888
},
{
"epoch": 0.2,
"learning_rate": 0.00040618856569709126,
"loss": 3.0348,
"theoretical_loss": 3.8105404702783288,
"tokens_seen": 646119424
},
{
"epoch": 0.2,
"learning_rate": 0.00040617853560682044,
"loss": 3.0608,
"theoretical_loss": 3.8105006656023264,
"tokens_seen": 646184960
},
{
"epoch": 0.2,
"learning_rate": 0.00040616850551654967,
"loss": 2.9151,
"theoretical_loss": 3.810460866093329,
"tokens_seen": 646250496
},
{
"epoch": 0.2,
"learning_rate": 0.0004061584754262788,
"loss": 2.8537,
"theoretical_loss": 3.8104210717501434,
"tokens_seen": 646316032
},
{
"epoch": 0.2,
"learning_rate": 0.00040614844533600804,
"loss": 2.6747,
"theoretical_loss": 3.8103812825715746,
"tokens_seen": 646381568
},
{
"epoch": 0.2,
"learning_rate": 0.00040613841524573727,
"loss": 2.6813,
"theoretical_loss": 3.8103414985564292,
"tokens_seen": 646447104
},
{
"epoch": 0.2,
"learning_rate": 0.0004061283851554664,
"loss": 2.951,
"theoretical_loss": 3.8103017197035145,
"tokens_seen": 646512640
},
{
"epoch": 0.2,
"learning_rate": 0.00040611835506519563,
"loss": 2.8612,
"theoretical_loss": 3.810261946011636,
"tokens_seen": 646578176
},
{
"epoch": 0.2,
"learning_rate": 0.00040610832497492476,
"loss": 2.9858,
"theoretical_loss": 3.8102221774796012,
"tokens_seen": 646643712
},
{
"epoch": 0.2,
"learning_rate": 0.000406098294884654,
"loss": 2.88,
"theoretical_loss": 3.810182414106219,
"tokens_seen": 646709248
},
{
"epoch": 0.2,
"learning_rate": 0.0004060882647943832,
"loss": 2.7479,
"theoretical_loss": 3.810142655890297,
"tokens_seen": 646774784
},
{
"epoch": 0.2,
"learning_rate": 0.00040607823470411236,
"loss": 3.0342,
"theoretical_loss": 3.8101029028306437,
"tokens_seen": 646840320
},
{
"epoch": 0.2,
"learning_rate": 0.00040606820461384154,
"loss": 3.1329,
"theoretical_loss": 3.810063154926068,
"tokens_seen": 646905856
},
{
"epoch": 0.2,
"learning_rate": 0.0004060581745235707,
"loss": 2.7757,
"theoretical_loss": 3.8100234121753798,
"tokens_seen": 646971392
},
{
"epoch": 0.2,
"learning_rate": 0.0004060481444332999,
"loss": 2.9786,
"theoretical_loss": 3.8099836745773885,
"tokens_seen": 647036928
},
{
"epoch": 0.2,
"learning_rate": 0.00040603811434302914,
"loss": 2.7087,
"theoretical_loss": 3.8099439421309036,
"tokens_seen": 647102464
},
{
"epoch": 0.2,
"objective/train/docs_used": 375146,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.229652166366577,
"objective/train/theoretical_loss": 3.8099042148347366,
"objective/train/tokens_used": 667628000,
"theoretical_loss": 3.8099042148347366,
"tokens_seen": 647168000
},
{
"epoch": 0.2,
"learning_rate": 0.00040602808425275826,
"loss": 2.9167,
"theoretical_loss": 3.8099042148347366,
"tokens_seen": 647168000
},
{
"epoch": 0.2,
"learning_rate": 0.0004060180541624875,
"loss": 2.4942,
"theoretical_loss": 3.809864492687698,
"tokens_seen": 647233536
},
{
"epoch": 0.2,
"learning_rate": 0.0004060080240722167,
"loss": 2.9195,
"theoretical_loss": 3.8098247756885995,
"tokens_seen": 647299072
},
{
"epoch": 0.2,
"learning_rate": 0.00040599799398194586,
"loss": 3.1232,
"theoretical_loss": 3.8097850638362516,
"tokens_seen": 647364608
},
{
"epoch": 0.2,
"learning_rate": 0.00040598796389167504,
"loss": 2.9194,
"theoretical_loss": 3.8097453571294686,
"tokens_seen": 647430144
},
{
"epoch": 0.2,
"learning_rate": 0.0004059779338014042,
"loss": 2.7781,
"theoretical_loss": 3.8097056555670608,
"tokens_seen": 647495680
},
{
"epoch": 0.2,
"learning_rate": 0.0004059679037111334,
"loss": 2.9331,
"theoretical_loss": 3.809665959147843,
"tokens_seen": 647561216
},
{
"epoch": 0.2,
"learning_rate": 0.00040595787362086264,
"loss": 3.0463,
"theoretical_loss": 3.8096262678706267,
"tokens_seen": 647626752
},
{
"epoch": 0.2,
"learning_rate": 0.00040594784353059177,
"loss": 2.7725,
"theoretical_loss": 3.8095865817342274,
"tokens_seen": 647692288
},
{
"epoch": 0.2,
"learning_rate": 0.000405937813440321,
"loss": 3.0406,
"theoretical_loss": 3.809546900737457,
"tokens_seen": 647757824
},
{
"epoch": 0.2,
"learning_rate": 0.00040592778335005013,
"loss": 2.9272,
"theoretical_loss": 3.809507224879132,
"tokens_seen": 647823360
},
{
"epoch": 0.2,
"learning_rate": 0.00040591775325977936,
"loss": 3.1577,
"theoretical_loss": 3.809467554158067,
"tokens_seen": 647888896
},
{
"epoch": 0.2,
"learning_rate": 0.00040590772316950854,
"loss": 2.8766,
"theoretical_loss": 3.809427888573077,
"tokens_seen": 647954432
},
{
"epoch": 0.2,
"learning_rate": 0.0004058976930792377,
"loss": 2.7179,
"theoretical_loss": 3.8093882281229767,
"tokens_seen": 648019968
},
{
"epoch": 0.2,
"learning_rate": 0.0004058876629889669,
"loss": 2.9549,
"theoretical_loss": 3.8093485728065835,
"tokens_seen": 648085504
},
{
"epoch": 0.2,
"learning_rate": 0.0004058776328986961,
"loss": 2.9438,
"theoretical_loss": 3.809308922622713,
"tokens_seen": 648151040
},
{
"epoch": 0.2,
"learning_rate": 0.00040586760280842527,
"loss": 3.0333,
"theoretical_loss": 3.8092692775701824,
"tokens_seen": 648216576
},
{
"epoch": 0.2,
"learning_rate": 0.0004058575727181545,
"loss": 2.7464,
"theoretical_loss": 3.8092296376478094,
"tokens_seen": 648282112
},
{
"epoch": 0.2,
"learning_rate": 0.00040584754262788363,
"loss": 3.149,
"theoretical_loss": 3.8091900028544106,
"tokens_seen": 648347648
},
{
"epoch": 0.2,
"learning_rate": 0.00040583751253761287,
"loss": 2.8448,
"theoretical_loss": 3.809150373188805,
"tokens_seen": 648413184
},
{
"epoch": 0.2,
"learning_rate": 0.00040582748244734205,
"loss": 2.9585,
"theoretical_loss": 3.80911074864981,
"tokens_seen": 648478720
},
{
"epoch": 0.2,
"learning_rate": 0.00040581745235707123,
"loss": 2.8377,
"theoretical_loss": 3.8090711292362456,
"tokens_seen": 648544256
},
{
"epoch": 0.2,
"learning_rate": 0.0004058074222668004,
"loss": 2.8158,
"theoretical_loss": 3.80903151494693,
"tokens_seen": 648609792
},
{
"epoch": 0.2,
"learning_rate": 0.0004057973921765296,
"loss": 3.0696,
"theoretical_loss": 3.808991905780683,
"tokens_seen": 648675328
},
{
"epoch": 0.2,
"learning_rate": 0.00040578736208625877,
"loss": 2.875,
"theoretical_loss": 3.8089523017363254,
"tokens_seen": 648740864
},
{
"epoch": 0.2,
"objective/train/docs_used": 375860,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.106677770614624,
"objective/train/theoretical_loss": 3.8089127028126764,
"objective/train/tokens_used": 669266400,
"theoretical_loss": 3.8089127028126764,
"tokens_seen": 648806400
},
{
"epoch": 0.2,
"learning_rate": 0.000405777331995988,
"loss": 2.8354,
"theoretical_loss": 3.8089127028126764,
"tokens_seen": 648806400
},
{
"epoch": 0.2,
"learning_rate": 0.00040576730190571713,
"loss": 2.6935,
"theoretical_loss": 3.8088731090085575,
"tokens_seen": 648871936
},
{
"epoch": 0.2,
"learning_rate": 0.00040575727181544637,
"loss": 2.7503,
"theoretical_loss": 3.8088335203227897,
"tokens_seen": 648937472
},
{
"epoch": 0.2,
"learning_rate": 0.0004057472417251755,
"loss": 2.9742,
"theoretical_loss": 3.8087939367541943,
"tokens_seen": 649003008
},
{
"epoch": 0.2,
"learning_rate": 0.00040573721163490473,
"loss": 3.0332,
"theoretical_loss": 3.8087543583015933,
"tokens_seen": 649068544
},
{
"epoch": 0.2,
"learning_rate": 0.0004057271815446339,
"loss": 3.0165,
"theoretical_loss": 3.8087147849638097,
"tokens_seen": 649134080
},
{
"epoch": 0.2,
"learning_rate": 0.0004057171514543631,
"loss": 3.0003,
"theoretical_loss": 3.808675216739665,
"tokens_seen": 649199616
},
{
"epoch": 0.2,
"learning_rate": 0.0004057071213640923,
"loss": 3.0969,
"theoretical_loss": 3.808635653627983,
"tokens_seen": 649265152
},
{
"epoch": 0.2,
"learning_rate": 0.00040569709127382146,
"loss": 2.8748,
"theoretical_loss": 3.8085960956275873,
"tokens_seen": 649330688
},
{
"epoch": 0.2,
"learning_rate": 0.00040568706118355064,
"loss": 2.8892,
"theoretical_loss": 3.8085565427373016,
"tokens_seen": 649396224
},
{
"epoch": 0.2,
"learning_rate": 0.0004056770310932799,
"loss": 2.6728,
"theoretical_loss": 3.8085169949559505,
"tokens_seen": 649461760
},
{
"epoch": 0.2,
"learning_rate": 0.000405667001003009,
"loss": 2.8494,
"theoretical_loss": 3.808477452282358,
"tokens_seen": 649527296
},
{
"epoch": 0.2,
"learning_rate": 0.00040565697091273824,
"loss": 3.0504,
"theoretical_loss": 3.8084379147153493,
"tokens_seen": 649592832
},
{
"epoch": 0.2,
"learning_rate": 0.0004056469408224674,
"loss": 2.8025,
"theoretical_loss": 3.808398382253751,
"tokens_seen": 649658368
},
{
"epoch": 0.2,
"learning_rate": 0.0004056369107321966,
"loss": 3.1074,
"theoretical_loss": 3.8083588548963867,
"tokens_seen": 649723904
},
{
"epoch": 0.2,
"learning_rate": 0.0004056268806419258,
"loss": 2.8261,
"theoretical_loss": 3.8083193326420846,
"tokens_seen": 649789440
},
{
"epoch": 0.2,
"learning_rate": 0.00040561685055165496,
"loss": 2.7663,
"theoretical_loss": 3.808279815489671,
"tokens_seen": 649854976
},
{
"epoch": 0.2,
"learning_rate": 0.00040560682046138414,
"loss": 2.811,
"theoretical_loss": 3.8082403034379713,
"tokens_seen": 649920512
},
{
"epoch": 0.2,
"learning_rate": 0.0004055967903711134,
"loss": 2.9304,
"theoretical_loss": 3.808200796485815,
"tokens_seen": 649986048
},
{
"epoch": 0.2,
"learning_rate": 0.0004055867602808425,
"loss": 3.0942,
"theoretical_loss": 3.8081612946320287,
"tokens_seen": 650051584
},
{
"epoch": 0.2,
"learning_rate": 0.00040557673019057174,
"loss": 2.9475,
"theoretical_loss": 3.8081217978754407,
"tokens_seen": 650117120
},
{
"epoch": 0.2,
"learning_rate": 0.00040556670010030087,
"loss": 2.9979,
"theoretical_loss": 3.80808230621488,
"tokens_seen": 650182656
},
{
"epoch": 0.2,
"learning_rate": 0.0004055566700100301,
"loss": 2.8176,
"theoretical_loss": 3.8080428196491747,
"tokens_seen": 650248192
},
{
"epoch": 0.2,
"learning_rate": 0.0004055466399197593,
"loss": 2.9571,
"theoretical_loss": 3.808003338177154,
"tokens_seen": 650313728
},
{
"epoch": 0.2,
"learning_rate": 0.00040553660982948846,
"loss": 2.8646,
"theoretical_loss": 3.8079638617976492,
"tokens_seen": 650379264
},
{
"epoch": 0.2,
"objective/train/docs_used": 377374,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.111595392227173,
"objective/train/theoretical_loss": 3.8079243905094886,
"objective/train/tokens_used": 670904800,
"theoretical_loss": 3.8079243905094886,
"tokens_seen": 650444800
},
{
"epoch": 0.2,
"learning_rate": 0.00040552657973921764,
"loss": 2.8842,
"theoretical_loss": 3.8079243905094886,
"tokens_seen": 650444800
},
{
"epoch": 0.2,
"learning_rate": 0.0004055165496489469,
"loss": 2.8977,
"theoretical_loss": 3.8078849243115043,
"tokens_seen": 650510336
},
{
"epoch": 0.2,
"learning_rate": 0.000405506519558676,
"loss": 2.8774,
"theoretical_loss": 3.8078454632025256,
"tokens_seen": 650575872
},
{
"epoch": 0.2,
"learning_rate": 0.00040549648946840524,
"loss": 2.9541,
"theoretical_loss": 3.8078060071813846,
"tokens_seen": 650641408
},
{
"epoch": 0.2,
"learning_rate": 0.00040548645937813437,
"loss": 2.9887,
"theoretical_loss": 3.8077665562469134,
"tokens_seen": 650706944
},
{
"epoch": 0.2,
"learning_rate": 0.0004054764292878636,
"loss": 2.8827,
"theoretical_loss": 3.807727110397942,
"tokens_seen": 650772480
},
{
"epoch": 0.2,
"learning_rate": 0.0004054663991975928,
"loss": 2.8771,
"theoretical_loss": 3.8076876696333057,
"tokens_seen": 650838016
},
{
"epoch": 0.2,
"learning_rate": 0.00040545636910732197,
"loss": 3.034,
"theoretical_loss": 3.807648233951835,
"tokens_seen": 650903552
},
{
"epoch": 0.2,
"learning_rate": 0.00040544633901705115,
"loss": 3.0342,
"theoretical_loss": 3.8076088033523643,
"tokens_seen": 650969088
},
{
"epoch": 0.2,
"learning_rate": 0.00040543630892678033,
"loss": 2.779,
"theoretical_loss": 3.807569377833726,
"tokens_seen": 651034624
},
{
"epoch": 0.2,
"learning_rate": 0.0004054262788365095,
"loss": 3.1769,
"theoretical_loss": 3.8075299573947556,
"tokens_seen": 651100160
},
{
"epoch": 0.2,
"learning_rate": 0.00040541624874623874,
"loss": 2.9482,
"theoretical_loss": 3.8074905420342864,
"tokens_seen": 651165696
},
{
"epoch": 0.2,
"learning_rate": 0.00040540621865596787,
"loss": 2.8476,
"theoretical_loss": 3.8074511317511535,
"tokens_seen": 651231232
},
{
"epoch": 0.2,
"learning_rate": 0.0004053961885656971,
"loss": 2.8999,
"theoretical_loss": 3.807411726544191,
"tokens_seen": 651296768
},
{
"epoch": 0.2,
"learning_rate": 0.0004053861584754263,
"loss": 3.0428,
"theoretical_loss": 3.8073723264122363,
"tokens_seen": 651362304
},
{
"epoch": 0.2,
"learning_rate": 0.00040537612838515547,
"loss": 2.8553,
"theoretical_loss": 3.8073329313541233,
"tokens_seen": 651427840
},
{
"epoch": 0.2,
"learning_rate": 0.0004053660982948847,
"loss": 2.792,
"theoretical_loss": 3.80729354136869,
"tokens_seen": 651493376
},
{
"epoch": 0.2,
"learning_rate": 0.00040535606820461383,
"loss": 2.9315,
"theoretical_loss": 3.807254156454772,
"tokens_seen": 651558912
},
{
"epoch": 0.2,
"learning_rate": 0.00040534603811434307,
"loss": 3.0491,
"theoretical_loss": 3.807214776611206,
"tokens_seen": 651624448
},
{
"epoch": 0.2,
"learning_rate": 0.00040533600802407225,
"loss": 3.1541,
"theoretical_loss": 3.8071754018368305,
"tokens_seen": 651689984
},
{
"epoch": 0.2,
"learning_rate": 0.00040532597793380143,
"loss": 3.0628,
"theoretical_loss": 3.807136032130483,
"tokens_seen": 651755520
},
{
"epoch": 0.2,
"learning_rate": 0.0004053159478435306,
"loss": 2.8765,
"theoretical_loss": 3.8070966674910007,
"tokens_seen": 651821056
},
{
"epoch": 0.2,
"learning_rate": 0.0004053059177532598,
"loss": 2.9814,
"theoretical_loss": 3.8070573079172227,
"tokens_seen": 651886592
},
{
"epoch": 0.2,
"learning_rate": 0.00040529588766298897,
"loss": 2.9982,
"theoretical_loss": 3.807017953407989,
"tokens_seen": 651952128
},
{
"epoch": 0.2,
"learning_rate": 0.0004052858575727182,
"loss": 2.793,
"theoretical_loss": 3.806978603962137,
"tokens_seen": 652017664
},
{
"epoch": 0.2,
"objective/train/docs_used": 377873,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8976755142211914,
"objective/train/theoretical_loss": 3.8069392595785083,
"objective/train/tokens_used": 672543200,
"theoretical_loss": 3.8069392595785083,
"tokens_seen": 652083200
},
{
"epoch": 0.2,
"learning_rate": 0.00040527582748244733,
"loss": 2.9065,
"theoretical_loss": 3.8069392595785083,
"tokens_seen": 652083200
},
{
"epoch": 0.2,
"learning_rate": 0.00040526579739217657,
"loss": 2.8254,
"theoretical_loss": 3.806899920255942,
"tokens_seen": 652148736
},
{
"epoch": 0.2,
"learning_rate": 0.0004052557673019057,
"loss": 2.8496,
"theoretical_loss": 3.8068605859932783,
"tokens_seen": 652214272
},
{
"epoch": 0.2,
"learning_rate": 0.00040524573721163493,
"loss": 3.0224,
"theoretical_loss": 3.806821256789359,
"tokens_seen": 652279808
},
{
"epoch": 0.2,
"learning_rate": 0.0004052357071213641,
"loss": 2.8427,
"theoretical_loss": 3.806781932643024,
"tokens_seen": 652345344
},
{
"epoch": 0.2,
"learning_rate": 0.0004052256770310933,
"loss": 2.9331,
"theoretical_loss": 3.8067426135531157,
"tokens_seen": 652410880
},
{
"epoch": 0.2,
"learning_rate": 0.0004052156469408225,
"loss": 2.758,
"theoretical_loss": 3.8067032995184764,
"tokens_seen": 652476416
},
{
"epoch": 0.2,
"learning_rate": 0.00040520561685055166,
"loss": 3.0669,
"theoretical_loss": 3.8066639905379485,
"tokens_seen": 652541952
},
{
"epoch": 0.2,
"learning_rate": 0.00040519558676028084,
"loss": 3.0724,
"theoretical_loss": 3.806624686610374,
"tokens_seen": 652607488
},
{
"epoch": 0.2,
"learning_rate": 0.0004051855566700101,
"loss": 2.8816,
"theoretical_loss": 3.8065853877345965,
"tokens_seen": 652673024
},
{
"epoch": 0.2,
"learning_rate": 0.0004051755265797392,
"loss": 2.8926,
"theoretical_loss": 3.8065460939094597,
"tokens_seen": 652738560
},
{
"epoch": 0.2,
"learning_rate": 0.00040516549648946844,
"loss": 2.8759,
"theoretical_loss": 3.8065068051338073,
"tokens_seen": 652804096
},
{
"epoch": 0.2,
"learning_rate": 0.0004051554663991976,
"loss": 2.7865,
"theoretical_loss": 3.806467521406484,
"tokens_seen": 652869632
},
{
"epoch": 0.2,
"learning_rate": 0.0004051454363089268,
"loss": 2.9804,
"theoretical_loss": 3.806428242726333,
"tokens_seen": 652935168
},
{
"epoch": 0.2,
"learning_rate": 0.000405135406218656,
"loss": 3.0628,
"theoretical_loss": 3.8063889690922013,
"tokens_seen": 653000704
},
{
"epoch": 0.2,
"learning_rate": 0.00040512537612838516,
"loss": 2.6469,
"theoretical_loss": 3.806349700502933,
"tokens_seen": 653066240
},
{
"epoch": 0.2,
"learning_rate": 0.00040511534603811434,
"loss": 2.8764,
"theoretical_loss": 3.806310436957374,
"tokens_seen": 653131776
},
{
"epoch": 0.2,
"learning_rate": 0.0004051053159478436,
"loss": 2.9757,
"theoretical_loss": 3.8062711784543715,
"tokens_seen": 653197312
},
{
"epoch": 0.2,
"learning_rate": 0.0004050952858575727,
"loss": 2.5247,
"theoretical_loss": 3.8062319249927707,
"tokens_seen": 653262848
},
{
"epoch": 0.2,
"learning_rate": 0.00040508525576730194,
"loss": 2.9353,
"theoretical_loss": 3.8061926765714196,
"tokens_seen": 653328384
},
{
"epoch": 0.2,
"learning_rate": 0.00040507522567703107,
"loss": 2.9652,
"theoretical_loss": 3.806153433189165,
"tokens_seen": 653393920
},
{
"epoch": 0.2,
"learning_rate": 0.0004050651955867603,
"loss": 3.0484,
"theoretical_loss": 3.806114194844855,
"tokens_seen": 653459456
},
{
"epoch": 0.2,
"learning_rate": 0.0004050551654964895,
"loss": 2.9003,
"theoretical_loss": 3.8060749615373375,
"tokens_seen": 653524992
},
{
"epoch": 0.2,
"learning_rate": 0.00040504513540621866,
"loss": 2.933,
"theoretical_loss": 3.8060357332654604,
"tokens_seen": 653590528
},
{
"epoch": 0.2,
"learning_rate": 0.00040503510531594784,
"loss": 2.7105,
"theoretical_loss": 3.8059965100280735,
"tokens_seen": 653656064
},
{
"epoch": 0.2,
"objective/train/docs_used": 379015,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.882786989212036,
"objective/train/theoretical_loss": 3.8059572918240248,
"objective/train/tokens_used": 674181600,
"theoretical_loss": 3.8059572918240248,
"tokens_seen": 653721600
},
{
"epoch": 0.2,
"learning_rate": 0.0004050250752256771,
"loss": 2.7231,
"theoretical_loss": 3.8059572918240248,
"tokens_seen": 653721600
},
{
"epoch": 0.2,
"learning_rate": 0.0004050150451354062,
"loss": 2.8368,
"theoretical_loss": 3.8059180786521654,
"tokens_seen": 653787136
},
{
"epoch": 0.2,
"learning_rate": 0.00040500501504513544,
"loss": 2.9828,
"theoretical_loss": 3.805878870511344,
"tokens_seen": 653852672
},
{
"epoch": 0.2,
"learning_rate": 0.00040499498495486457,
"loss": 2.8881,
"theoretical_loss": 3.805839667400412,
"tokens_seen": 653918208
},
{
"epoch": 0.2,
"learning_rate": 0.0004049849548645938,
"loss": 2.8925,
"theoretical_loss": 3.8058004693182195,
"tokens_seen": 653983744
},
{
"epoch": 0.2,
"learning_rate": 0.000404974924774323,
"loss": 2.9944,
"theoretical_loss": 3.8057612762636173,
"tokens_seen": 654049280
},
{
"epoch": 0.2,
"learning_rate": 0.00040496489468405217,
"loss": 2.7396,
"theoretical_loss": 3.8057220882354574,
"tokens_seen": 654114816
},
{
"epoch": 0.2,
"learning_rate": 0.00040495486459378135,
"loss": 2.8268,
"theoretical_loss": 3.8056829052325916,
"tokens_seen": 654180352
},
{
"epoch": 0.2,
"learning_rate": 0.00040494483450351053,
"loss": 2.9557,
"theoretical_loss": 3.805643727253872,
"tokens_seen": 654245888
},
{
"epoch": 0.2,
"learning_rate": 0.0004049348044132397,
"loss": 2.5606,
"theoretical_loss": 3.8056045542981516,
"tokens_seen": 654311424
},
{
"epoch": 0.2,
"learning_rate": 0.00040492477432296894,
"loss": 2.9476,
"theoretical_loss": 3.805565386364283,
"tokens_seen": 654376960
},
{
"epoch": 0.2,
"learning_rate": 0.00040491474423269807,
"loss": 2.8194,
"theoretical_loss": 3.8055262234511193,
"tokens_seen": 654442496
},
{
"epoch": 0.2,
"learning_rate": 0.0004049047141424273,
"loss": 2.7053,
"theoretical_loss": 3.8054870655575153,
"tokens_seen": 654508032
},
{
"epoch": 0.2,
"learning_rate": 0.00040489468405215643,
"loss": 2.7497,
"theoretical_loss": 3.805447912682324,
"tokens_seen": 654573568
},
{
"epoch": 0.2,
"learning_rate": 0.00040488465396188567,
"loss": 2.5947,
"theoretical_loss": 3.8054087648244,
"tokens_seen": 654639104
},
{
"epoch": 0.2,
"learning_rate": 0.00040487462387161485,
"loss": 3.0872,
"theoretical_loss": 3.8053696219825994,
"tokens_seen": 654704640
},
{
"epoch": 0.2,
"learning_rate": 0.00040486459378134403,
"loss": 2.738,
"theoretical_loss": 3.805330484155776,
"tokens_seen": 654770176
},
{
"epoch": 0.2,
"learning_rate": 0.0004048545636910732,
"loss": 2.5799,
"theoretical_loss": 3.8052913513427864,
"tokens_seen": 654835712
},
{
"epoch": 0.2,
"learning_rate": 0.00040484453360080245,
"loss": 3.0017,
"theoretical_loss": 3.8052522235424853,
"tokens_seen": 654901248
},
{
"epoch": 0.2,
"learning_rate": 0.0004048345035105316,
"loss": 2.7451,
"theoretical_loss": 3.805213100753731,
"tokens_seen": 654966784
},
{
"epoch": 0.2,
"learning_rate": 0.0004048244734202608,
"loss": 2.7297,
"theoretical_loss": 3.8051739829753792,
"tokens_seen": 655032320
},
{
"epoch": 0.2,
"learning_rate": 0.00040481444332998994,
"loss": 2.9863,
"theoretical_loss": 3.8051348702062864,
"tokens_seen": 655097856
},
{
"epoch": 0.2,
"learning_rate": 0.00040480441323971917,
"loss": 2.7582,
"theoretical_loss": 3.805095762445311,
"tokens_seen": 655163392
},
{
"epoch": 0.2,
"learning_rate": 0.00040479438314944835,
"loss": 3.0981,
"theoretical_loss": 3.805056659691311,
"tokens_seen": 655228928
},
{
"epoch": 0.2,
"learning_rate": 0.00040478435305917753,
"loss": 2.759,
"theoretical_loss": 3.805017561943144,
"tokens_seen": 655294464
},
{
"debugging/Self-BLEU-5": 0.36561994328618,
"debugging/distinct-1-grams": 0.7722597331948876,
"debugging/distinct-2-grams": 0.8492548173712738,
"debugging/entropy-1-grams": 5.826808035476899,
"debugging/entropy-2-grams": 6.590880961095115,
"debugging/length": 495.27272727272725,
"debugging/num_segments": 11,
"epoch": 0.2,
"objective/train/docs_used": 379813,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7408783435821533,
"objective/train/theoretical_loss": 3.804978469199669,
"objective/train/tokens_used": 675820000,
"theoretical_loss": 3.804978469199669,
"tokens_seen": 655360000
},
{
"epoch": 0.2,
"learning_rate": 0.0004047743229689067,
"loss": 2.8606,
"theoretical_loss": 3.804978469199669,
"tokens_seen": 655360000
},
{
"epoch": 0.2,
"learning_rate": 0.0004047642928786359,
"loss": 2.9838,
"theoretical_loss": 3.804939381459745,
"tokens_seen": 655425536
},
{
"epoch": 0.2,
"learning_rate": 0.0004047542627883651,
"loss": 2.8588,
"theoretical_loss": 3.8049002987222313,
"tokens_seen": 655491072
},
{
"epoch": 0.2,
"learning_rate": 0.0004047442326980943,
"loss": 2.91,
"theoretical_loss": 3.8048612209859876,
"tokens_seen": 655556608
},
{
"epoch": 0.2,
"learning_rate": 0.00040473420260782344,
"loss": 2.7598,
"theoretical_loss": 3.8048221482498743,
"tokens_seen": 655622144
},
{
"epoch": 0.2,
"learning_rate": 0.0004047241725175527,
"loss": 2.7256,
"theoretical_loss": 3.8047830805127516,
"tokens_seen": 655687680
},
{
"epoch": 0.2,
"learning_rate": 0.0004047141424272818,
"loss": 2.8889,
"theoretical_loss": 3.8047440177734804,
"tokens_seen": 655753216
},
{
"epoch": 0.2,
"learning_rate": 0.00040470411233701104,
"loss": 2.9605,
"theoretical_loss": 3.8047049600309224,
"tokens_seen": 655818752
},
{
"epoch": 0.2,
"learning_rate": 0.0004046940822467402,
"loss": 3.0472,
"theoretical_loss": 3.804665907283938,
"tokens_seen": 655884288
},
{
"epoch": 0.2,
"learning_rate": 0.0004046840521564694,
"loss": 2.8666,
"theoretical_loss": 3.804626859531391,
"tokens_seen": 655949824
},
{
"epoch": 0.2,
"learning_rate": 0.0004046740220661986,
"loss": 2.9352,
"theoretical_loss": 3.804587816772142,
"tokens_seen": 656015360
},
{
"epoch": 0.2,
"learning_rate": 0.0004046639919759278,
"loss": 2.6207,
"theoretical_loss": 3.8045487790050547,
"tokens_seen": 656080896
},
{
"epoch": 0.2,
"learning_rate": 0.000404653961885657,
"loss": 2.9404,
"theoretical_loss": 3.8045097462289923,
"tokens_seen": 656146432
},
{
"epoch": 0.2,
"learning_rate": 0.0004046439317953862,
"loss": 2.7605,
"theoretical_loss": 3.804470718442818,
"tokens_seen": 656211968
},
{
"epoch": 0.2,
"learning_rate": 0.00040463390170511536,
"loss": 3.0043,
"theoretical_loss": 3.8044316956453956,
"tokens_seen": 656277504
},
{
"epoch": 0.2,
"learning_rate": 0.00040462387161484454,
"loss": 2.8137,
"theoretical_loss": 3.8043926778355894,
"tokens_seen": 656343040
},
{
"epoch": 0.2,
"learning_rate": 0.0004046138415245738,
"loss": 2.9791,
"theoretical_loss": 3.804353665012264,
"tokens_seen": 656408576
},
{
"epoch": 0.2,
"learning_rate": 0.0004046038114343029,
"loss": 3.004,
"theoretical_loss": 3.8043146571742845,
"tokens_seen": 656474112
},
{
"epoch": 0.2,
"learning_rate": 0.00040459378134403214,
"loss": 2.7808,
"theoretical_loss": 3.8042756543205156,
"tokens_seen": 656539648
},
{
"epoch": 0.2,
"learning_rate": 0.00040458375125376127,
"loss": 3.0046,
"theoretical_loss": 3.8042366564498242,
"tokens_seen": 656605184
},
{
"epoch": 0.2,
"learning_rate": 0.0004045737211634905,
"loss": 2.829,
"theoretical_loss": 3.8041976635610752,
"tokens_seen": 656670720
},
{
"epoch": 0.2,
"learning_rate": 0.0004045636910732197,
"loss": 3.0518,
"theoretical_loss": 3.8041586756531363,
"tokens_seen": 656736256
},
{
"epoch": 0.2,
"learning_rate": 0.00040455366098294886,
"loss": 3.0584,
"theoretical_loss": 3.804119692724873,
"tokens_seen": 656801792
},
{
"epoch": 0.2,
"learning_rate": 0.00040454363089267804,
"loss": 3.0102,
"theoretical_loss": 3.804080714775153,
"tokens_seen": 656867328
},
{
"epoch": 0.2,
"learning_rate": 0.0004045336008024073,
"loss": 2.9845,
"theoretical_loss": 3.8040417418028443,
"tokens_seen": 656932864
},
{
"epoch": 0.2,
"objective/train/docs_used": 381022,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.044368267059326,
"objective/train/theoretical_loss": 3.8040027738068143,
"objective/train/tokens_used": 677458400,
"theoretical_loss": 3.8040027738068143,
"tokens_seen": 656998400
},
{
"epoch": 0.2,
"learning_rate": 0.0004045235707121364,
"loss": 2.9637,
"theoretical_loss": 3.8040027738068143,
"tokens_seen": 656998400
},
{
"epoch": 0.2,
"learning_rate": 0.00040451354062186564,
"loss": 2.8752,
"theoretical_loss": 3.8039638107859317,
"tokens_seen": 657063936
},
{
"epoch": 0.2,
"learning_rate": 0.00040450351053159477,
"loss": 3.2028,
"theoretical_loss": 3.803924852739065,
"tokens_seen": 657129472
},
{
"epoch": 0.2,
"learning_rate": 0.000404493480441324,
"loss": 2.9586,
"theoretical_loss": 3.803885899665083,
"tokens_seen": 657195008
},
{
"epoch": 0.2,
"learning_rate": 0.0004044834503510532,
"loss": 3.049,
"theoretical_loss": 3.8038469515628557,
"tokens_seen": 657260544
},
{
"epoch": 0.2,
"learning_rate": 0.00040447342026078237,
"loss": 2.7938,
"theoretical_loss": 3.803808008431252,
"tokens_seen": 657326080
},
{
"epoch": 0.2,
"learning_rate": 0.00040446339017051155,
"loss": 2.7174,
"theoretical_loss": 3.8037690702691433,
"tokens_seen": 657391616
},
{
"epoch": 0.2,
"learning_rate": 0.00040445336008024073,
"loss": 2.754,
"theoretical_loss": 3.8037301370753993,
"tokens_seen": 657457152
},
{
"epoch": 0.2,
"learning_rate": 0.0004044433299899699,
"loss": 3.0205,
"theoretical_loss": 3.803691208848891,
"tokens_seen": 657522688
},
{
"epoch": 0.2,
"learning_rate": 0.00040443329989969915,
"loss": 2.9776,
"theoretical_loss": 3.8036522855884893,
"tokens_seen": 657588224
},
{
"epoch": 0.2,
"learning_rate": 0.00040442326980942827,
"loss": 2.5956,
"theoretical_loss": 3.803613367293067,
"tokens_seen": 657653760
},
{
"epoch": 0.2,
"learning_rate": 0.0004044132397191575,
"loss": 2.9704,
"theoretical_loss": 3.803574453961495,
"tokens_seen": 657719296
},
{
"epoch": 0.2,
"learning_rate": 0.00040440320962888663,
"loss": 3.0798,
"theoretical_loss": 3.803535545592646,
"tokens_seen": 657784832
},
{
"epoch": 0.2,
"learning_rate": 0.00040439317953861587,
"loss": 2.8397,
"theoretical_loss": 3.803496642185393,
"tokens_seen": 657850368
},
{
"epoch": 0.2,
"learning_rate": 0.00040438314944834505,
"loss": 2.7924,
"theoretical_loss": 3.8034577437386083,
"tokens_seen": 657915904
},
{
"epoch": 0.2,
"learning_rate": 0.00040437311935807423,
"loss": 2.82,
"theoretical_loss": 3.8034188502511665,
"tokens_seen": 657981440
},
{
"epoch": 0.2,
"learning_rate": 0.0004043630892678034,
"loss": 2.6594,
"theoretical_loss": 3.803379961721941,
"tokens_seen": 658046976
},
{
"epoch": 0.2,
"learning_rate": 0.00040435305917753265,
"loss": 2.835,
"theoretical_loss": 3.8033410781498054,
"tokens_seen": 658112512
},
{
"epoch": 0.2,
"learning_rate": 0.0004043430290872618,
"loss": 2.9879,
"theoretical_loss": 3.803302199533635,
"tokens_seen": 658178048
},
{
"epoch": 0.2,
"learning_rate": 0.000404332998996991,
"loss": 3.0102,
"theoretical_loss": 3.8032633258723045,
"tokens_seen": 658243584
},
{
"epoch": 0.2,
"learning_rate": 0.00040432296890672014,
"loss": 2.7771,
"theoretical_loss": 3.8032244571646894,
"tokens_seen": 658309120
},
{
"epoch": 0.2,
"learning_rate": 0.00040431293881644937,
"loss": 2.8471,
"theoretical_loss": 3.8031855934096654,
"tokens_seen": 658374656
},
{
"epoch": 0.2,
"learning_rate": 0.00040430290872617855,
"loss": 2.8971,
"theoretical_loss": 3.8031467346061083,
"tokens_seen": 658440192
},
{
"epoch": 0.2,
"learning_rate": 0.00040429287863590773,
"loss": 2.9073,
"theoretical_loss": 3.8031078807528944,
"tokens_seen": 658505728
},
{
"epoch": 0.2,
"learning_rate": 0.0004042828485456369,
"loss": 2.7519,
"theoretical_loss": 3.803069031848901,
"tokens_seen": 658571264
},
{
"epoch": 0.2,
"objective/train/docs_used": 381585,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.120889663696289,
"objective/train/theoretical_loss": 3.803030187893005,
"objective/train/tokens_used": 679096800,
"theoretical_loss": 3.803030187893005,
"tokens_seen": 658636800
},
{
"epoch": 0.2,
"learning_rate": 0.0004042728184553661,
"loss": 2.7675,
"theoretical_loss": 3.803030187893005,
"tokens_seen": 658636800
},
{
"epoch": 0.2,
"learning_rate": 0.0004042627883650953,
"loss": 2.805,
"theoretical_loss": 3.802991348884084,
"tokens_seen": 658702336
},
{
"epoch": 0.2,
"learning_rate": 0.0004042527582748245,
"loss": 2.8179,
"theoretical_loss": 3.8029525148210146,
"tokens_seen": 658767872
},
{
"epoch": 0.2,
"learning_rate": 0.00040424272818455364,
"loss": 3.0148,
"theoretical_loss": 3.802913685702677,
"tokens_seen": 658833408
},
{
"epoch": 0.2,
"learning_rate": 0.0004042326980942829,
"loss": 2.8772,
"theoretical_loss": 3.8028748615279495,
"tokens_seen": 658898944
},
{
"epoch": 0.2,
"learning_rate": 0.000404222668004012,
"loss": 3.0401,
"theoretical_loss": 3.80283604229571,
"tokens_seen": 658964480
},
{
"epoch": 0.2,
"learning_rate": 0.00040421263791374124,
"loss": 2.5957,
"theoretical_loss": 3.802797228004839,
"tokens_seen": 659030016
},
{
"epoch": 0.2,
"learning_rate": 0.0004042026078234704,
"loss": 2.8999,
"theoretical_loss": 3.802758418654215,
"tokens_seen": 659095552
},
{
"epoch": 0.2,
"learning_rate": 0.0004041925777331996,
"loss": 2.931,
"theoretical_loss": 3.8027196142427195,
"tokens_seen": 659161088
},
{
"epoch": 0.2,
"learning_rate": 0.0004041825476429288,
"loss": 2.8085,
"theoretical_loss": 3.802680814769232,
"tokens_seen": 659226624
},
{
"epoch": 0.2,
"learning_rate": 0.000404172517552658,
"loss": 2.7624,
"theoretical_loss": 3.802642020232633,
"tokens_seen": 659292160
},
{
"epoch": 0.2,
"learning_rate": 0.00040416248746238714,
"loss": 2.8777,
"theoretical_loss": 3.8026032306318047,
"tokens_seen": 659357696
},
{
"epoch": 0.2,
"learning_rate": 0.0004041524573721164,
"loss": 3.0243,
"theoretical_loss": 3.8025644459656283,
"tokens_seen": 659423232
},
{
"epoch": 0.2,
"learning_rate": 0.0004041424272818455,
"loss": 2.8105,
"theoretical_loss": 3.802525666232985,
"tokens_seen": 659488768
},
{
"epoch": 0.2,
"learning_rate": 0.00040413239719157474,
"loss": 2.7355,
"theoretical_loss": 3.802486891432758,
"tokens_seen": 659554304
},
{
"epoch": 0.2,
"learning_rate": 0.0004041223671013039,
"loss": 2.9355,
"theoretical_loss": 3.802448121563829,
"tokens_seen": 659619840
},
{
"epoch": 0.2,
"learning_rate": 0.0004041123370110331,
"loss": 2.7167,
"theoretical_loss": 3.802409356625082,
"tokens_seen": 659685376
},
{
"epoch": 0.2,
"learning_rate": 0.0004041023069207623,
"loss": 2.9943,
"theoretical_loss": 3.8023705966153996,
"tokens_seen": 659750912
},
{
"epoch": 0.2,
"learning_rate": 0.00040409227683049147,
"loss": 2.7669,
"theoretical_loss": 3.802331841533666,
"tokens_seen": 659816448
},
{
"epoch": 0.2,
"learning_rate": 0.00040408224674022065,
"loss": 2.9835,
"theoretical_loss": 3.8022930913787647,
"tokens_seen": 659881984
},
{
"epoch": 0.2,
"learning_rate": 0.0004040722166499499,
"loss": 2.847,
"theoretical_loss": 3.8022543461495806,
"tokens_seen": 659947520
},
{
"epoch": 0.2,
"learning_rate": 0.000404062186559679,
"loss": 2.982,
"theoretical_loss": 3.8022156058449985,
"tokens_seen": 660013056
},
{
"epoch": 0.2,
"learning_rate": 0.00040405215646940824,
"loss": 2.9005,
"theoretical_loss": 3.802176870463904,
"tokens_seen": 660078592
},
{
"epoch": 0.2,
"learning_rate": 0.00040404212637913737,
"loss": 2.7324,
"theoretical_loss": 3.802138140005182,
"tokens_seen": 660144128
},
{
"epoch": 0.2,
"learning_rate": 0.0004040320962888666,
"loss": 2.912,
"theoretical_loss": 3.8020994144677185,
"tokens_seen": 660209664
},
{
"epoch": 0.2,
"objective/train/docs_used": 382875,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6055595874786377,
"objective/train/theoretical_loss": 3.8020606938504,
"objective/train/tokens_used": 680735200,
"theoretical_loss": 3.8020606938504,
"tokens_seen": 660275200
},
{
"epoch": 0.2,
"learning_rate": 0.0004040220661985958,
"loss": 2.9253,
"theoretical_loss": 3.8020606938504,
"tokens_seen": 660275200
},
{
"epoch": 0.2,
"learning_rate": 0.00040401203610832497,
"loss": 3.18,
"theoretical_loss": 3.8020219781521125,
"tokens_seen": 660340736
},
{
"epoch": 0.2,
"learning_rate": 0.00040400200601805415,
"loss": 3.0083,
"theoretical_loss": 3.8019832673717437,
"tokens_seen": 660406272
},
{
"epoch": 0.2,
"learning_rate": 0.0004039919759277834,
"loss": 2.6783,
"theoretical_loss": 3.801944561508181,
"tokens_seen": 660471808
},
{
"epoch": 0.2,
"learning_rate": 0.0004039819458375125,
"loss": 2.9787,
"theoretical_loss": 3.801905860560311,
"tokens_seen": 660537344
},
{
"epoch": 0.2,
"learning_rate": 0.00040397191574724175,
"loss": 2.9549,
"theoretical_loss": 3.8018671645270237,
"tokens_seen": 660602880
},
{
"epoch": 0.2,
"learning_rate": 0.0004039618856569709,
"loss": 2.8998,
"theoretical_loss": 3.8018284734072054,
"tokens_seen": 660668416
},
{
"epoch": 0.2,
"learning_rate": 0.0004039518555667001,
"loss": 2.953,
"theoretical_loss": 3.8017897871997466,
"tokens_seen": 660733952
},
{
"epoch": 0.2,
"learning_rate": 0.0004039418254764293,
"loss": 2.8189,
"theoretical_loss": 3.801751105903536,
"tokens_seen": 660799488
},
{
"epoch": 0.2,
"learning_rate": 0.00040393179538615847,
"loss": 2.91,
"theoretical_loss": 3.8017124295174622,
"tokens_seen": 660865024
},
{
"epoch": 0.2,
"learning_rate": 0.00040392176529588765,
"loss": 2.864,
"theoretical_loss": 3.8016737580404163,
"tokens_seen": 660930560
},
{
"epoch": 0.2,
"learning_rate": 0.00040391173520561683,
"loss": 2.773,
"theoretical_loss": 3.8016350914712884,
"tokens_seen": 660996096
},
{
"epoch": 0.2,
"learning_rate": 0.00040390170511534607,
"loss": 2.6822,
"theoretical_loss": 3.801596429808968,
"tokens_seen": 661061632
},
{
"epoch": 0.2,
"learning_rate": 0.00040389167502507525,
"loss": 2.7872,
"theoretical_loss": 3.8015577730523473,
"tokens_seen": 661127168
},
{
"epoch": 0.2,
"learning_rate": 0.00040388164493480443,
"loss": 2.9026,
"theoretical_loss": 3.801519121200317,
"tokens_seen": 661192704
},
{
"epoch": 0.2,
"learning_rate": 0.0004038716148445336,
"loss": 2.9067,
"theoretical_loss": 3.801480474251769,
"tokens_seen": 661258240
},
{
"epoch": 0.2,
"learning_rate": 0.00040386158475426285,
"loss": 2.8634,
"theoretical_loss": 3.801441832205595,
"tokens_seen": 661323776
},
{
"epoch": 0.2,
"learning_rate": 0.000403851554663992,
"loss": 2.8866,
"theoretical_loss": 3.801403195060688,
"tokens_seen": 661389312
},
{
"epoch": 0.2,
"learning_rate": 0.0004038415245737212,
"loss": 2.7794,
"theoretical_loss": 3.801364562815941,
"tokens_seen": 661454848
},
{
"epoch": 0.2,
"learning_rate": 0.00040383149448345034,
"loss": 2.8431,
"theoretical_loss": 3.8013259354702456,
"tokens_seen": 661520384
},
{
"epoch": 0.2,
"learning_rate": 0.00040382146439317957,
"loss": 2.9924,
"theoretical_loss": 3.801287313022497,
"tokens_seen": 661585920
},
{
"epoch": 0.2,
"learning_rate": 0.00040381143430290875,
"loss": 2.7929,
"theoretical_loss": 3.801248695471588,
"tokens_seen": 661651456
},
{
"epoch": 0.2,
"learning_rate": 0.00040380140421263794,
"loss": 3.0859,
"theoretical_loss": 3.8012100828164135,
"tokens_seen": 661716992
},
{
"epoch": 0.2,
"learning_rate": 0.0004037913741223671,
"loss": 2.7847,
"theoretical_loss": 3.801171475055867,
"tokens_seen": 661782528
},
{
"epoch": 0.2,
"learning_rate": 0.0004037813440320963,
"loss": 2.8057,
"theoretical_loss": 3.8011328721888447,
"tokens_seen": 661848064
},
{
"epoch": 0.2,
"objective/train/docs_used": 383560,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1271653175354004,
"objective/train/theoretical_loss": 3.8010942742142415,
"objective/train/tokens_used": 682373600,
"theoretical_loss": 3.8010942742142415,
"tokens_seen": 661913600
},
{
"epoch": 0.2,
"learning_rate": 0.0004037713139418255,
"loss": 2.7091,
"theoretical_loss": 3.8010942742142415,
"tokens_seen": 661913600
},
{
"epoch": 0.2,
"learning_rate": 0.0004037612838515547,
"loss": 2.8806,
"theoretical_loss": 3.8010556811309524,
"tokens_seen": 661979136
},
{
"epoch": 0.2,
"learning_rate": 0.00040375125376128384,
"loss": 2.941,
"theoretical_loss": 3.801017092937874,
"tokens_seen": 662044672
},
{
"epoch": 0.2,
"learning_rate": 0.0004037412236710131,
"loss": 2.9441,
"theoretical_loss": 3.800978509633903,
"tokens_seen": 662110208
},
{
"epoch": 0.2,
"learning_rate": 0.0004037311935807422,
"loss": 2.7173,
"theoretical_loss": 3.8009399312179353,
"tokens_seen": 662175744
},
{
"epoch": 0.2,
"learning_rate": 0.00040372116349047144,
"loss": 2.8,
"theoretical_loss": 3.800901357688868,
"tokens_seen": 662241280
},
{
"epoch": 0.2,
"learning_rate": 0.0004037111334002006,
"loss": 2.9218,
"theoretical_loss": 3.800862789045599,
"tokens_seen": 662306816
},
{
"epoch": 0.2,
"learning_rate": 0.0004037011033099298,
"loss": 2.7922,
"theoretical_loss": 3.800824225287026,
"tokens_seen": 662372352
},
{
"epoch": 0.2,
"learning_rate": 0.000403691073219659,
"loss": 2.9749,
"theoretical_loss": 3.8007856664120476,
"tokens_seen": 662437888
},
{
"epoch": 0.2,
"learning_rate": 0.0004036810431293882,
"loss": 2.8713,
"theoretical_loss": 3.800747112419561,
"tokens_seen": 662503424
},
{
"epoch": 0.2,
"learning_rate": 0.00040367101303911734,
"loss": 2.9567,
"theoretical_loss": 3.8007085633084663,
"tokens_seen": 662568960
},
{
"epoch": 0.2,
"learning_rate": 0.0004036609829488466,
"loss": 2.9524,
"theoretical_loss": 3.8006700190776623,
"tokens_seen": 662634496
},
{
"epoch": 0.2,
"learning_rate": 0.0004036509528585757,
"loss": 2.9928,
"theoretical_loss": 3.8006314797260488,
"tokens_seen": 662700032
},
{
"epoch": 0.2,
"learning_rate": 0.00040364092276830494,
"loss": 2.9063,
"theoretical_loss": 3.800592945252525,
"tokens_seen": 662765568
},
{
"epoch": 0.2,
"learning_rate": 0.0004036308926780341,
"loss": 2.9209,
"theoretical_loss": 3.8005544156559923,
"tokens_seen": 662831104
},
{
"epoch": 0.2,
"learning_rate": 0.0004036208625877633,
"loss": 2.8781,
"theoretical_loss": 3.80051589093535,
"tokens_seen": 662896640
},
{
"epoch": 0.2,
"learning_rate": 0.0004036108324974925,
"loss": 2.7279,
"theoretical_loss": 3.8004773710895003,
"tokens_seen": 662962176
},
{
"epoch": 0.2,
"learning_rate": 0.00040360080240722167,
"loss": 2.896,
"theoretical_loss": 3.800438856117344,
"tokens_seen": 663027712
},
{
"epoch": 0.2,
"learning_rate": 0.00040359077231695085,
"loss": 2.6029,
"theoretical_loss": 3.800400346017783,
"tokens_seen": 663093248
},
{
"epoch": 0.2,
"learning_rate": 0.0004035807422266801,
"loss": 2.9801,
"theoretical_loss": 3.8003618407897193,
"tokens_seen": 663158784
},
{
"epoch": 0.2,
"learning_rate": 0.0004035707121364092,
"loss": 3.0723,
"theoretical_loss": 3.800323340432055,
"tokens_seen": 663224320
},
{
"epoch": 0.2,
"learning_rate": 0.00040356068204613844,
"loss": 2.7635,
"theoretical_loss": 3.8002848449436932,
"tokens_seen": 663289856
},
{
"epoch": 0.2,
"learning_rate": 0.00040355065195586757,
"loss": 2.8274,
"theoretical_loss": 3.800246354323538,
"tokens_seen": 663355392
},
{
"epoch": 0.2,
"learning_rate": 0.0004035406218655968,
"loss": 2.9454,
"theoretical_loss": 3.8002078685704914,
"tokens_seen": 663420928
},
{
"epoch": 0.2,
"learning_rate": 0.000403530591775326,
"loss": 2.9828,
"theoretical_loss": 3.8001693876834572,
"tokens_seen": 663486464
},
{
"epoch": 0.2,
"objective/train/docs_used": 384103,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.467216730117798,
"objective/train/theoretical_loss": 3.8001309116613413,
"objective/train/tokens_used": 684012000,
"theoretical_loss": 3.8001309116613413,
"tokens_seen": 663552000
},
{
"epoch": 0.2,
"learning_rate": 0.00040352056168505517,
"loss": 2.9179,
"theoretical_loss": 3.8001309116613413,
"tokens_seen": 663552000
},
{
"epoch": 0.2,
"learning_rate": 0.00040351053159478435,
"loss": 2.7601,
"theoretical_loss": 3.8000924405030467,
"tokens_seen": 663617536
},
{
"epoch": 0.2,
"learning_rate": 0.0004035005015045136,
"loss": 2.8075,
"theoretical_loss": 3.8000539742074784,
"tokens_seen": 663683072
},
{
"epoch": 0.2,
"learning_rate": 0.0004034904714142427,
"loss": 2.9031,
"theoretical_loss": 3.8000155127735424,
"tokens_seen": 663748608
},
{
"epoch": 0.2,
"learning_rate": 0.00040348044132397195,
"loss": 2.7523,
"theoretical_loss": 3.7999770562001443,
"tokens_seen": 663814144
},
{
"epoch": 0.2,
"learning_rate": 0.0004034704112337011,
"loss": 2.8653,
"theoretical_loss": 3.799938604486189,
"tokens_seen": 663879680
},
{
"epoch": 0.2,
"learning_rate": 0.0004034603811434303,
"loss": 2.7593,
"theoretical_loss": 3.799900157630584,
"tokens_seen": 663945216
},
{
"epoch": 0.2,
"learning_rate": 0.0004034503510531595,
"loss": 2.903,
"theoretical_loss": 3.799861715632236,
"tokens_seen": 664010752
},
{
"epoch": 0.2,
"learning_rate": 0.00040344032096288867,
"loss": 3.0078,
"theoretical_loss": 3.7998232784900514,
"tokens_seen": 664076288
},
{
"epoch": 0.2,
"learning_rate": 0.00040343029087261785,
"loss": 2.588,
"theoretical_loss": 3.7997848462029378,
"tokens_seen": 664141824
},
{
"epoch": 0.2,
"learning_rate": 0.00040342026078234703,
"loss": 2.7286,
"theoretical_loss": 3.7997464187698027,
"tokens_seen": 664207360
},
{
"epoch": 0.2,
"learning_rate": 0.0004034102306920762,
"loss": 2.8097,
"theoretical_loss": 3.799707996189555,
"tokens_seen": 664272896
},
{
"epoch": 0.2,
"learning_rate": 0.00040340020060180545,
"loss": 2.8803,
"theoretical_loss": 3.7996695784611023,
"tokens_seen": 664338432
},
{
"epoch": 0.2,
"learning_rate": 0.0004033901705115346,
"loss": 2.8301,
"theoretical_loss": 3.799631165583354,
"tokens_seen": 664403968
},
{
"epoch": 0.2,
"learning_rate": 0.0004033801404212638,
"loss": 2.8001,
"theoretical_loss": 3.7995927575552186,
"tokens_seen": 664469504
},
{
"epoch": 0.2,
"learning_rate": 0.000403370110330993,
"loss": 2.9368,
"theoretical_loss": 3.7995543543756063,
"tokens_seen": 664535040
},
{
"epoch": 0.2,
"learning_rate": 0.0004033600802407222,
"loss": 2.9599,
"theoretical_loss": 3.799515956043427,
"tokens_seen": 664600576
},
{
"epoch": 0.2,
"learning_rate": 0.00040335005015045136,
"loss": 2.7997,
"theoretical_loss": 3.7994775625575903,
"tokens_seen": 664666112
},
{
"epoch": 0.2,
"learning_rate": 0.00040334002006018054,
"loss": 2.8467,
"theoretical_loss": 3.799439173917007,
"tokens_seen": 664731648
},
{
"epoch": 0.2,
"learning_rate": 0.0004033299899699097,
"loss": 2.9554,
"theoretical_loss": 3.799400790120588,
"tokens_seen": 664797184
},
{
"epoch": 0.2,
"learning_rate": 0.00040331995987963895,
"loss": 2.8535,
"theoretical_loss": 3.799362411167245,
"tokens_seen": 664862720
},
{
"epoch": 0.2,
"learning_rate": 0.0004033099297893681,
"loss": 3.0903,
"theoretical_loss": 3.7993240370558894,
"tokens_seen": 664928256
},
{
"epoch": 0.2,
"learning_rate": 0.0004032998996990973,
"loss": 2.8286,
"theoretical_loss": 3.7992856677854334,
"tokens_seen": 664993792
},
{
"epoch": 0.2,
"learning_rate": 0.00040328986960882644,
"loss": 2.9313,
"theoretical_loss": 3.7992473033547887,
"tokens_seen": 665059328
},
{
"epoch": 0.2,
"learning_rate": 0.0004032798395185557,
"loss": 3.1494,
"theoretical_loss": 3.799208943762868,
"tokens_seen": 665124864
},
{
"epoch": 0.2,
"objective/train/docs_used": 385247,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1371493339538574,
"objective/train/theoretical_loss": 3.799170589008585,
"objective/train/tokens_used": 685650400,
"theoretical_loss": 3.799170589008585,
"tokens_seen": 665190400
},
{
"epoch": 0.2,
"learning_rate": 0.00040326980942828486,
"loss": 3.0874,
"theoretical_loss": 3.799170589008585,
"tokens_seen": 665190400
},
{
"epoch": 0.2,
"learning_rate": 0.00040325977933801404,
"loss": 2.7269,
"theoretical_loss": 3.799132239090852,
"tokens_seen": 665255936
},
{
"epoch": 0.2,
"learning_rate": 0.0004032497492477432,
"loss": 2.8093,
"theoretical_loss": 3.7990938940085845,
"tokens_seen": 665321472
},
{
"epoch": 0.2,
"learning_rate": 0.0004032397191574724,
"loss": 2.9118,
"theoretical_loss": 3.799055553760695,
"tokens_seen": 665387008
},
{
"epoch": 0.2,
"learning_rate": 0.0004032296890672016,
"loss": 2.8174,
"theoretical_loss": 3.7990172183460995,
"tokens_seen": 665452544
},
{
"epoch": 0.2,
"learning_rate": 0.0004032196589769308,
"loss": 2.8863,
"theoretical_loss": 3.7989788877637105,
"tokens_seen": 665518080
},
{
"epoch": 0.2,
"learning_rate": 0.00040320962888665995,
"loss": 2.8796,
"theoretical_loss": 3.7989405620124455,
"tokens_seen": 665583616
},
{
"epoch": 0.2,
"learning_rate": 0.0004031995987963892,
"loss": 2.6664,
"theoretical_loss": 3.7989022410912185,
"tokens_seen": 665649152
},
{
"epoch": 0.2,
"learning_rate": 0.00040318956870611836,
"loss": 2.7397,
"theoretical_loss": 3.7988639249989458,
"tokens_seen": 665714688
},
{
"epoch": 0.2,
"learning_rate": 0.00040317953861584754,
"loss": 2.8563,
"theoretical_loss": 3.7988256137345435,
"tokens_seen": 665780224
},
{
"epoch": 0.2,
"learning_rate": 0.0004031695085255767,
"loss": 2.9544,
"theoretical_loss": 3.7987873072969283,
"tokens_seen": 665845760
},
{
"epoch": 0.2,
"learning_rate": 0.0004031594784353059,
"loss": 3.0253,
"theoretical_loss": 3.798749005685017,
"tokens_seen": 665911296
},
{
"epoch": 0.2,
"learning_rate": 0.00040314944834503514,
"loss": 2.8732,
"theoretical_loss": 3.798710708897727,
"tokens_seen": 665976832
},
{
"epoch": 0.2,
"learning_rate": 0.0004031394182547643,
"loss": 2.7689,
"theoretical_loss": 3.7986724169339756,
"tokens_seen": 666042368
},
{
"epoch": 0.2,
"learning_rate": 0.0004031293881644935,
"loss": 2.8396,
"theoretical_loss": 3.798634129792681,
"tokens_seen": 666107904
},
{
"epoch": 0.2,
"learning_rate": 0.0004031193580742227,
"loss": 2.9368,
"theoretical_loss": 3.798595847472761,
"tokens_seen": 666173440
},
{
"epoch": 0.2,
"learning_rate": 0.00040310932798395187,
"loss": 2.6594,
"theoretical_loss": 3.798557569973135,
"tokens_seen": 666238976
},
{
"epoch": 0.2,
"learning_rate": 0.00040309929789368105,
"loss": 2.682,
"theoretical_loss": 3.7985192972927218,
"tokens_seen": 666304512
},
{
"epoch": 0.2,
"learning_rate": 0.0004030892678034103,
"loss": 2.6207,
"theoretical_loss": 3.79848102943044,
"tokens_seen": 666370048
},
{
"epoch": 0.2,
"learning_rate": 0.0004030792377131394,
"loss": 2.677,
"theoretical_loss": 3.79844276638521,
"tokens_seen": 666435584
},
{
"epoch": 0.2,
"learning_rate": 0.00040306920762286864,
"loss": 2.9129,
"theoretical_loss": 3.7984045081559517,
"tokens_seen": 666501120
},
{
"epoch": 0.2,
"learning_rate": 0.00040305917753259777,
"loss": 2.7053,
"theoretical_loss": 3.7983662547415853,
"tokens_seen": 666566656
},
{
"epoch": 0.2,
"learning_rate": 0.000403049147442327,
"loss": 2.7032,
"theoretical_loss": 3.798328006141032,
"tokens_seen": 666632192
},
{
"epoch": 0.2,
"learning_rate": 0.0004030391173520562,
"loss": 2.8943,
"theoretical_loss": 3.798289762353212,
"tokens_seen": 666697728
},
{
"epoch": 0.2,
"learning_rate": 0.00040302908726178537,
"loss": 2.9658,
"theoretical_loss": 3.798251523377048,
"tokens_seen": 666763264
},
{
"epoch": 0.2,
"objective/train/docs_used": 385855,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6441280841827393,
"objective/train/theoretical_loss": 3.79821328921146,
"objective/train/tokens_used": 687288800,
"theoretical_loss": 3.79821328921146,
"tokens_seen": 666828800
},
{
"epoch": 0.2,
"learning_rate": 0.00040301905717151455,
"loss": 2.741,
"theoretical_loss": 3.79821328921146,
"tokens_seen": 666828800
},
{
"epoch": 0.2,
"learning_rate": 0.0004030090270812438,
"loss": 2.7029,
"theoretical_loss": 3.798175059855372,
"tokens_seen": 666894336
},
{
"epoch": 0.2,
"learning_rate": 0.0004029989969909729,
"loss": 2.8905,
"theoretical_loss": 3.7981368353077056,
"tokens_seen": 666959872
},
{
"epoch": 0.2,
"learning_rate": 0.00040298896690070215,
"loss": 2.8304,
"theoretical_loss": 3.7980986155673833,
"tokens_seen": 667025408
},
{
"epoch": 0.2,
"learning_rate": 0.0004029789368104313,
"loss": 3.083,
"theoretical_loss": 3.7980604006333287,
"tokens_seen": 667090944
},
{
"epoch": 0.2,
"learning_rate": 0.0004029689067201605,
"loss": 2.6557,
"theoretical_loss": 3.7980221905044655,
"tokens_seen": 667156480
},
{
"epoch": 0.2,
"learning_rate": 0.0004029588766298897,
"loss": 2.9013,
"theoretical_loss": 3.797983985179717,
"tokens_seen": 667222016
},
{
"epoch": 0.2,
"learning_rate": 0.00040294884653961887,
"loss": 2.932,
"theoretical_loss": 3.797945784658008,
"tokens_seen": 667287552
},
{
"epoch": 0.2,
"learning_rate": 0.00040293881644934805,
"loss": 2.8594,
"theoretical_loss": 3.797907588938262,
"tokens_seen": 667353088
},
{
"epoch": 0.2,
"learning_rate": 0.00040292878635907723,
"loss": 2.805,
"theoretical_loss": 3.7978693980194054,
"tokens_seen": 667418624
},
{
"epoch": 0.2,
"learning_rate": 0.0004029187562688064,
"loss": 2.9822,
"theoretical_loss": 3.797831211900362,
"tokens_seen": 667484160
},
{
"epoch": 0.2,
"learning_rate": 0.00040290872617853565,
"loss": 2.9129,
"theoretical_loss": 3.7977930305800585,
"tokens_seen": 667549696
},
{
"epoch": 0.2,
"learning_rate": 0.0004028986960882648,
"loss": 2.7209,
"theoretical_loss": 3.7977548540574206,
"tokens_seen": 667615232
},
{
"epoch": 0.2,
"learning_rate": 0.000402888665997994,
"loss": 2.8038,
"theoretical_loss": 3.7977166823313735,
"tokens_seen": 667680768
},
{
"epoch": 0.2,
"learning_rate": 0.0004028786359077232,
"loss": 2.7264,
"theoretical_loss": 3.7976785154008454,
"tokens_seen": 667746304
},
{
"epoch": 0.2,
"learning_rate": 0.0004028686058174524,
"loss": 2.7102,
"theoretical_loss": 3.797640353264762,
"tokens_seen": 667811840
},
{
"epoch": 0.2,
"learning_rate": 0.00040285857572718156,
"loss": 2.8847,
"theoretical_loss": 3.797602195922052,
"tokens_seen": 667877376
},
{
"epoch": 0.2,
"learning_rate": 0.00040284854563691074,
"loss": 3.1612,
"theoretical_loss": 3.7975640433716413,
"tokens_seen": 667942912
},
{
"epoch": 0.2,
"learning_rate": 0.0004028385155466399,
"loss": 2.8909,
"theoretical_loss": 3.797525895612459,
"tokens_seen": 668008448
},
{
"epoch": 0.2,
"learning_rate": 0.00040282848545636915,
"loss": 2.7694,
"theoretical_loss": 3.797487752643434,
"tokens_seen": 668073984
},
{
"epoch": 0.2,
"learning_rate": 0.0004028184553660983,
"loss": 2.6918,
"theoretical_loss": 3.7974496144634937,
"tokens_seen": 668139520
},
{
"epoch": 0.2,
"learning_rate": 0.0004028084252758275,
"loss": 2.7482,
"theoretical_loss": 3.797411481071568,
"tokens_seen": 668205056
},
{
"epoch": 0.2,
"learning_rate": 0.00040279839518555664,
"loss": 2.8803,
"theoretical_loss": 3.797373352466586,
"tokens_seen": 668270592
},
{
"epoch": 0.2,
"learning_rate": 0.0004027883650952859,
"loss": 2.6074,
"theoretical_loss": 3.7973352286474773,
"tokens_seen": 668336128
},
{
"epoch": 0.2,
"learning_rate": 0.00040277833500501506,
"loss": 3.0028,
"theoretical_loss": 3.7972971096131722,
"tokens_seen": 668401664
},
{
"epoch": 0.2,
"objective/train/docs_used": 387212,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1807034015655518,
"objective/train/theoretical_loss": 3.7972589953626006,
"objective/train/tokens_used": 688927200,
"theoretical_loss": 3.7972589953626006,
"tokens_seen": 668467200
},
{
"epoch": 0.2,
"learning_rate": 0.00040276830491474424,
"loss": 2.9601,
"theoretical_loss": 3.7972589953626006,
"tokens_seen": 668467200
},
{
"epoch": 0.2,
"learning_rate": 0.0004027582748244734,
"loss": 3.1079,
"theoretical_loss": 3.797220885894694,
"tokens_seen": 668532736
},
{
"epoch": 0.2,
"learning_rate": 0.0004027482447342026,
"loss": 2.6287,
"theoretical_loss": 3.7971827812083836,
"tokens_seen": 668598272
},
{
"epoch": 0.2,
"learning_rate": 0.0004027382146439318,
"loss": 2.7253,
"theoretical_loss": 3.7971446813026,
"tokens_seen": 668663808
},
{
"epoch": 0.2,
"learning_rate": 0.000402728184553661,
"loss": 2.8358,
"theoretical_loss": 3.7971065861762754,
"tokens_seen": 668729344
},
{
"epoch": 0.2,
"learning_rate": 0.00040271815446339015,
"loss": 2.9321,
"theoretical_loss": 3.7970684958283423,
"tokens_seen": 668794880
},
{
"epoch": 0.2,
"learning_rate": 0.0004027081243731194,
"loss": 2.6072,
"theoretical_loss": 3.797030410257733,
"tokens_seen": 668860416
},
{
"epoch": 0.2,
"learning_rate": 0.00040269809428284856,
"loss": 2.8204,
"theoretical_loss": 3.7969923294633796,
"tokens_seen": 668925952
},
{
"epoch": 0.2,
"learning_rate": 0.00040268806419257774,
"loss": 2.8889,
"theoretical_loss": 3.796954253444216,
"tokens_seen": 668991488
},
{
"epoch": 0.2,
"learning_rate": 0.0004026780341023069,
"loss": 2.8494,
"theoretical_loss": 3.7969161821991757,
"tokens_seen": 669057024
},
{
"epoch": 0.2,
"learning_rate": 0.0004026680040120361,
"loss": 2.8473,
"theoretical_loss": 3.7968781157271922,
"tokens_seen": 669122560
},
{
"epoch": 0.2,
"learning_rate": 0.0004026579739217653,
"loss": 2.9897,
"theoretical_loss": 3.7968400540272,
"tokens_seen": 669188096
},
{
"epoch": 0.2,
"learning_rate": 0.0004026479438314945,
"loss": 2.9307,
"theoretical_loss": 3.796801997098134,
"tokens_seen": 669253632
},
{
"epoch": 0.2,
"learning_rate": 0.00040263791374122365,
"loss": 2.8904,
"theoretical_loss": 3.7967639449389283,
"tokens_seen": 669319168
},
{
"epoch": 0.2,
"learning_rate": 0.0004026278836509529,
"loss": 2.7926,
"theoretical_loss": 3.7967258975485185,
"tokens_seen": 669384704
},
{
"epoch": 0.2,
"learning_rate": 0.000402617853560682,
"loss": 2.7744,
"theoretical_loss": 3.79668785492584,
"tokens_seen": 669450240
},
{
"epoch": 0.2,
"learning_rate": 0.00040260782347041125,
"loss": 3.0617,
"theoretical_loss": 3.796649817069828,
"tokens_seen": 669515776
},
{
"epoch": 0.2,
"learning_rate": 0.00040259779338014043,
"loss": 2.7322,
"theoretical_loss": 3.7966117839794205,
"tokens_seen": 669581312
},
{
"epoch": 0.2,
"learning_rate": 0.0004025877632898696,
"loss": 2.8147,
"theoretical_loss": 3.796573755653553,
"tokens_seen": 669646848
},
{
"epoch": 0.2,
"learning_rate": 0.0004025777331995988,
"loss": 2.7954,
"theoretical_loss": 3.7965357320911624,
"tokens_seen": 669712384
},
{
"epoch": 0.2,
"learning_rate": 0.00040256770310932797,
"loss": 2.7876,
"theoretical_loss": 3.7964977132911866,
"tokens_seen": 669777920
},
{
"epoch": 0.2,
"learning_rate": 0.00040255767301905715,
"loss": 3.0953,
"theoretical_loss": 3.796459699252562,
"tokens_seen": 669843456
},
{
"epoch": 0.2,
"learning_rate": 0.0004025476429287864,
"loss": 2.9212,
"theoretical_loss": 3.7964216899742276,
"tokens_seen": 669908992
},
{
"epoch": 0.2,
"learning_rate": 0.0004025376128385155,
"loss": 2.7308,
"theoretical_loss": 3.7963836854551216,
"tokens_seen": 669974528
},
{
"epoch": 0.2,
"learning_rate": 0.00040252758274824475,
"loss": 2.6901,
"theoretical_loss": 3.796345685694182,
"tokens_seen": 670040064
},
{
"epoch": 0.2,
"objective/train/docs_used": 387914,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.922666311264038,
"objective/train/theoretical_loss": 3.796307690690348,
"objective/train/tokens_used": 690565600,
"theoretical_loss": 3.796307690690348,
"tokens_seen": 670105600
},
{
"epoch": 0.2,
"learning_rate": 0.00040251755265797393,
"loss": 2.9185,
"theoretical_loss": 3.796307690690348,
"tokens_seen": 670105600
},
{
"epoch": 0.2,
"learning_rate": 0.0004025075225677031,
"loss": 3.1152,
"theoretical_loss": 3.7962697004425596,
"tokens_seen": 670171136
},
{
"epoch": 0.2,
"learning_rate": 0.0004024974924774323,
"loss": 2.9066,
"theoretical_loss": 3.796231714949756,
"tokens_seen": 670236672
},
{
"epoch": 0.2,
"learning_rate": 0.0004024874623871615,
"loss": 2.6382,
"theoretical_loss": 3.7961937342108767,
"tokens_seen": 670302208
},
{
"epoch": 0.2,
"learning_rate": 0.00040247743229689066,
"loss": 2.765,
"theoretical_loss": 3.796155758224863,
"tokens_seen": 670367744
},
{
"epoch": 0.2,
"learning_rate": 0.0004024674022066199,
"loss": 3.0396,
"theoretical_loss": 3.796117786990655,
"tokens_seen": 670433280
},
{
"epoch": 0.2,
"learning_rate": 0.000402457372116349,
"loss": 2.6992,
"theoretical_loss": 3.7960798205071935,
"tokens_seen": 670498816
},
{
"epoch": 0.2,
"learning_rate": 0.00040244734202607825,
"loss": 2.7139,
"theoretical_loss": 3.79604185877342,
"tokens_seen": 670564352
},
{
"epoch": 0.2,
"learning_rate": 0.0004024373119358074,
"loss": 3.0082,
"theoretical_loss": 3.7960039017882767,
"tokens_seen": 670629888
},
{
"epoch": 0.2,
"learning_rate": 0.0004024272818455366,
"loss": 2.6133,
"theoretical_loss": 3.7959659495507045,
"tokens_seen": 670695424
},
{
"epoch": 0.2,
"learning_rate": 0.0004024172517552658,
"loss": 2.8012,
"theoretical_loss": 3.7959280020596475,
"tokens_seen": 670760960
},
{
"epoch": 0.2,
"learning_rate": 0.000402407221664995,
"loss": 2.8379,
"theoretical_loss": 3.795890059314047,
"tokens_seen": 670826496
},
{
"epoch": 0.2,
"learning_rate": 0.0004023971915747242,
"loss": 3.1657,
"theoretical_loss": 3.795852121312846,
"tokens_seen": 670892032
},
{
"epoch": 0.2,
"learning_rate": 0.0004023871614844534,
"loss": 2.8205,
"theoretical_loss": 3.7958141880549885,
"tokens_seen": 670957568
},
{
"epoch": 0.2,
"learning_rate": 0.0004023771313941826,
"loss": 2.7751,
"theoretical_loss": 3.7957762595394184,
"tokens_seen": 671023104
},
{
"epoch": 0.2,
"learning_rate": 0.00040236710130391176,
"loss": 2.7387,
"theoretical_loss": 3.79573833576508,
"tokens_seen": 671088640
},
{
"epoch": 0.2,
"learning_rate": 0.00040235707121364094,
"loss": 2.8437,
"theoretical_loss": 3.795700416730916,
"tokens_seen": 671154176
},
{
"epoch": 0.2,
"learning_rate": 0.0004023470411233701,
"loss": 2.8998,
"theoretical_loss": 3.7956625024358726,
"tokens_seen": 671219712
},
{
"epoch": 0.2,
"learning_rate": 0.00040233701103309935,
"loss": 2.8319,
"theoretical_loss": 3.795624592878895,
"tokens_seen": 671285248
},
{
"epoch": 0.2,
"learning_rate": 0.0004023269809428285,
"loss": 2.8291,
"theoretical_loss": 3.795586688058928,
"tokens_seen": 671350784
},
{
"epoch": 0.2,
"learning_rate": 0.0004023169508525577,
"loss": 2.9348,
"theoretical_loss": 3.7955487879749175,
"tokens_seen": 671416320
},
{
"epoch": 0.2,
"learning_rate": 0.00040230692076228684,
"loss": 2.9264,
"theoretical_loss": 3.795510892625809,
"tokens_seen": 671481856
},
{
"epoch": 0.2,
"learning_rate": 0.0004022968906720161,
"loss": 3.0286,
"theoretical_loss": 3.7954730020105503,
"tokens_seen": 671547392
},
{
"epoch": 0.2,
"learning_rate": 0.00040228686058174526,
"loss": 2.9309,
"theoretical_loss": 3.795435116128087,
"tokens_seen": 671612928
},
{
"epoch": 0.2,
"learning_rate": 0.00040227683049147444,
"loss": 2.7353,
"theoretical_loss": 3.795397234977367,
"tokens_seen": 671678464
},
{
"epoch": 0.2,
"objective/train/docs_used": 389368,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1840548515319824,
"objective/train/theoretical_loss": 3.795359358557337,
"objective/train/tokens_used": 692204000,
"theoretical_loss": 3.795359358557337,
"tokens_seen": 671744000
},
{
"epoch": 0.2,
"learning_rate": 0.0004022668004012036,
"loss": 2.5556,
"theoretical_loss": 3.795359358557337,
"tokens_seen": 671744000
},
{
"epoch": 0.2,
"learning_rate": 0.0004022567703109328,
"loss": 2.8124,
"theoretical_loss": 3.7953214868669454,
"tokens_seen": 671809536
},
{
"epoch": 0.2,
"learning_rate": 0.000402246740220662,
"loss": 2.8381,
"theoretical_loss": 3.79528361990514,
"tokens_seen": 671875072
},
{
"epoch": 0.2,
"learning_rate": 0.0004022367101303912,
"loss": 2.7541,
"theoretical_loss": 3.7952457576708687,
"tokens_seen": 671940608
},
{
"epoch": 0.2,
"learning_rate": 0.00040222668004012035,
"loss": 2.8611,
"theoretical_loss": 3.7952079001630814,
"tokens_seen": 672006144
},
{
"epoch": 0.2,
"learning_rate": 0.0004022166499498496,
"loss": 2.8544,
"theoretical_loss": 3.7951700473807266,
"tokens_seen": 672071680
},
{
"epoch": 0.2,
"learning_rate": 0.00040220661985957876,
"loss": 2.8732,
"theoretical_loss": 3.795132199322754,
"tokens_seen": 672137216
},
{
"epoch": 0.2,
"learning_rate": 0.00040219658976930794,
"loss": 2.65,
"theoretical_loss": 3.795094355988113,
"tokens_seen": 672202752
},
{
"epoch": 0.2,
"learning_rate": 0.0004021865596790371,
"loss": 2.8004,
"theoretical_loss": 3.795056517375754,
"tokens_seen": 672268288
},
{
"epoch": 0.2,
"learning_rate": 0.0004021765295887663,
"loss": 2.9555,
"theoretical_loss": 3.795018683484628,
"tokens_seen": 672333824
},
{
"epoch": 0.2,
"learning_rate": 0.0004021664994984955,
"loss": 2.6817,
"theoretical_loss": 3.7949808543136845,
"tokens_seen": 672399360
},
{
"epoch": 0.2,
"learning_rate": 0.0004021564694082247,
"loss": 2.8005,
"theoretical_loss": 3.7949430298618756,
"tokens_seen": 672464896
},
{
"epoch": 0.2,
"learning_rate": 0.00040214643931795385,
"loss": 2.9211,
"theoretical_loss": 3.794905210128152,
"tokens_seen": 672530432
},
{
"epoch": 0.2,
"learning_rate": 0.0004021364092276831,
"loss": 2.8006,
"theoretical_loss": 3.794867395111467,
"tokens_seen": 672595968
},
{
"epoch": 0.2,
"learning_rate": 0.0004021263791374122,
"loss": 2.8702,
"theoretical_loss": 3.7948295848107714,
"tokens_seen": 672661504
},
{
"epoch": 0.2,
"learning_rate": 0.00040211634904714145,
"loss": 2.6242,
"theoretical_loss": 3.794791779225018,
"tokens_seen": 672727040
},
{
"epoch": 0.2,
"learning_rate": 0.00040210631895687063,
"loss": 2.9197,
"theoretical_loss": 3.7947539783531594,
"tokens_seen": 672792576
},
{
"epoch": 0.2,
"learning_rate": 0.0004020962888665998,
"loss": 2.6995,
"theoretical_loss": 3.7947161821941497,
"tokens_seen": 672858112
},
{
"epoch": 0.2,
"learning_rate": 0.000402086258776329,
"loss": 2.7163,
"theoretical_loss": 3.794678390746941,
"tokens_seen": 672923648
},
{
"epoch": 0.2,
"learning_rate": 0.00040207622868605817,
"loss": 2.8487,
"theoretical_loss": 3.794640604010488,
"tokens_seen": 672989184
},
{
"epoch": 0.2,
"learning_rate": 0.00040206619859578735,
"loss": 2.883,
"theoretical_loss": 3.7946028219837453,
"tokens_seen": 673054720
},
{
"epoch": 0.2,
"learning_rate": 0.0004020561685055166,
"loss": 2.95,
"theoretical_loss": 3.7945650446656662,
"tokens_seen": 673120256
},
{
"epoch": 0.2,
"learning_rate": 0.0004020461384152457,
"loss": 2.8632,
"theoretical_loss": 3.794527272055206,
"tokens_seen": 673185792
},
{
"epoch": 0.2,
"learning_rate": 0.00040203610832497495,
"loss": 2.9832,
"theoretical_loss": 3.79448950415132,
"tokens_seen": 673251328
},
{
"epoch": 0.2,
"learning_rate": 0.00040202607823470413,
"loss": 2.8424,
"theoretical_loss": 3.7944517409529634,
"tokens_seen": 673316864
},
{
"epoch": 0.2,
"objective/train/docs_used": 390091,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.549846887588501,
"objective/train/theoretical_loss": 3.7944139824590923,
"objective/train/tokens_used": 693842400,
"theoretical_loss": 3.7944139824590923,
"tokens_seen": 673382400
},
{
"epoch": 0.2,
"learning_rate": 0.0004020160481444333,
"loss": 2.6955,
"theoretical_loss": 3.7944139824590923,
"tokens_seen": 673382400
},
{
"epoch": 0.2,
"learning_rate": 0.0004020060180541625,
"loss": 2.8817,
"theoretical_loss": 3.794376228668663,
"tokens_seen": 673447936
},
{
"epoch": 0.2,
"learning_rate": 0.0004019959879638917,
"loss": 3.0468,
"theoretical_loss": 3.794338479580632,
"tokens_seen": 673513472
},
{
"epoch": 0.2,
"learning_rate": 0.00040198595787362086,
"loss": 2.7082,
"theoretical_loss": 3.7943007351939553,
"tokens_seen": 673579008
},
{
"epoch": 0.2,
"learning_rate": 0.0004019759277833501,
"loss": 2.7375,
"theoretical_loss": 3.794262995507591,
"tokens_seen": 673644544
},
{
"epoch": 0.2,
"learning_rate": 0.0004019658976930792,
"loss": 2.7516,
"theoretical_loss": 3.794225260520496,
"tokens_seen": 673710080
},
{
"epoch": 0.2,
"learning_rate": 0.00040195586760280845,
"loss": 2.9577,
"theoretical_loss": 3.794187530231628,
"tokens_seen": 673775616
},
{
"epoch": 0.2,
"learning_rate": 0.0004019458375125376,
"loss": 2.7577,
"theoretical_loss": 3.7941498046399458,
"tokens_seen": 673841152
},
{
"epoch": 0.2,
"learning_rate": 0.0004019358074222668,
"loss": 2.9868,
"theoretical_loss": 3.794112083744407,
"tokens_seen": 673906688
},
{
"epoch": 0.2,
"learning_rate": 0.000401925777331996,
"loss": 2.8115,
"theoretical_loss": 3.7940743675439714,
"tokens_seen": 673972224
},
{
"epoch": 0.2,
"learning_rate": 0.0004019157472417252,
"loss": 2.8863,
"theoretical_loss": 3.7940366560375978,
"tokens_seen": 674037760
},
{
"epoch": 0.2,
"learning_rate": 0.00040190571715145436,
"loss": 2.697,
"theoretical_loss": 3.793998949224245,
"tokens_seen": 674103296
},
{
"epoch": 0.2,
"learning_rate": 0.0004018956870611836,
"loss": 2.6846,
"theoretical_loss": 3.793961247102874,
"tokens_seen": 674168832
},
{
"epoch": 0.2,
"learning_rate": 0.0004018856569709127,
"loss": 2.5518,
"theoretical_loss": 3.793923549672444,
"tokens_seen": 674234368
},
{
"epoch": 0.2,
"learning_rate": 0.00040187562688064196,
"loss": 2.8618,
"theoretical_loss": 3.793885856931916,
"tokens_seen": 674299904
},
{
"epoch": 0.2,
"learning_rate": 0.0004018655967903711,
"loss": 2.659,
"theoretical_loss": 3.7938481688802503,
"tokens_seen": 674365440
},
{
"epoch": 0.2,
"learning_rate": 0.0004018555667001003,
"loss": 2.8492,
"theoretical_loss": 3.793810485516409,
"tokens_seen": 674430976
},
{
"epoch": 0.2,
"learning_rate": 0.0004018455366098295,
"loss": 2.6916,
"theoretical_loss": 3.793772806839352,
"tokens_seen": 674496512
},
{
"epoch": 0.2,
"learning_rate": 0.0004018355065195587,
"loss": 2.8757,
"theoretical_loss": 3.793735132848042,
"tokens_seen": 674562048
},
{
"epoch": 0.2,
"learning_rate": 0.00040182547642928786,
"loss": 2.9437,
"theoretical_loss": 3.7936974635414416,
"tokens_seen": 674627584
},
{
"epoch": 0.2,
"learning_rate": 0.00040181544633901704,
"loss": 2.877,
"theoretical_loss": 3.7936597989185126,
"tokens_seen": 674693120
},
{
"epoch": 0.2,
"learning_rate": 0.0004018054162487462,
"loss": 2.9069,
"theoretical_loss": 3.793622138978218,
"tokens_seen": 674758656
},
{
"epoch": 0.2,
"learning_rate": 0.00040179538615847546,
"loss": 2.8269,
"theoretical_loss": 3.7935844837195205,
"tokens_seen": 674824192
},
{
"epoch": 0.2,
"learning_rate": 0.0004017853560682046,
"loss": 2.6411,
"theoretical_loss": 3.7935468331413844,
"tokens_seen": 674889728
},
{
"epoch": 0.2,
"learning_rate": 0.0004017753259779338,
"loss": 2.9967,
"theoretical_loss": 3.7935091872427726,
"tokens_seen": 674955264
},
{
"epoch": 0.2,
"objective/train/docs_used": 391518,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.722494125366211,
"objective/train/theoretical_loss": 3.79347154602265,
"objective/train/tokens_used": 695480800,
"theoretical_loss": 3.79347154602265,
"tokens_seen": 675020800
},
{
"epoch": 0.2,
"learning_rate": 0.00040176529588766295,
"loss": 2.9325,
"theoretical_loss": 3.79347154602265,
"tokens_seen": 675020800
},
{
"epoch": 0.2,
"learning_rate": 0.0004017552657973922,
"loss": 3.0138,
"theoretical_loss": 3.7934339094799805,
"tokens_seen": 675086336
},
{
"epoch": 0.2,
"learning_rate": 0.00040174523570712137,
"loss": 2.7254,
"theoretical_loss": 3.7933962776137284,
"tokens_seen": 675151872
},
{
"epoch": 0.2,
"learning_rate": 0.00040173520561685055,
"loss": 2.6646,
"theoretical_loss": 3.79335865042286,
"tokens_seen": 675217408
},
{
"epoch": 0.2,
"learning_rate": 0.00040172517552657973,
"loss": 2.8017,
"theoretical_loss": 3.79332102790634,
"tokens_seen": 675282944
},
{
"epoch": 0.2,
"learning_rate": 0.00040171514543630896,
"loss": 2.9383,
"theoretical_loss": 3.7932834100631343,
"tokens_seen": 675348480
},
{
"epoch": 0.2,
"learning_rate": 0.0004017051153460381,
"loss": 2.7704,
"theoretical_loss": 3.793245796892209,
"tokens_seen": 675414016
},
{
"epoch": 0.2,
"learning_rate": 0.0004016950852557673,
"loss": 3.0444,
"theoretical_loss": 3.7932081883925304,
"tokens_seen": 675479552
},
{
"epoch": 0.2,
"learning_rate": 0.00040168505516549645,
"loss": 2.8548,
"theoretical_loss": 3.7931705845630654,
"tokens_seen": 675545088
},
{
"epoch": 0.2,
"learning_rate": 0.0004016750250752257,
"loss": 3.0835,
"theoretical_loss": 3.7931329854027807,
"tokens_seen": 675610624
},
{
"epoch": 0.2,
"learning_rate": 0.00040166499498495487,
"loss": 2.8866,
"theoretical_loss": 3.7930953909106444,
"tokens_seen": 675676160
},
{
"epoch": 0.2,
"learning_rate": 0.00040165496489468405,
"loss": 2.6754,
"theoretical_loss": 3.7930578010856233,
"tokens_seen": 675741696
},
{
"epoch": 0.2,
"learning_rate": 0.0004016449348044133,
"loss": 2.659,
"theoretical_loss": 3.793020215926686,
"tokens_seen": 675807232
},
{
"epoch": 0.2,
"learning_rate": 0.0004016349047141424,
"loss": 2.7333,
"theoretical_loss": 3.7929826354328013,
"tokens_seen": 675872768
},
{
"epoch": 0.2,
"learning_rate": 0.00040162487462387165,
"loss": 2.783,
"theoretical_loss": 3.7929450596029373,
"tokens_seen": 675938304
},
{
"epoch": 0.2,
"learning_rate": 0.00040161484453360083,
"loss": 2.9364,
"theoretical_loss": 3.792907488436063,
"tokens_seen": 676003840
},
{
"epoch": 0.2,
"learning_rate": 0.00040160481444333,
"loss": 2.77,
"theoretical_loss": 3.7928699219311484,
"tokens_seen": 676069376
},
{
"epoch": 0.2,
"learning_rate": 0.0004015947843530592,
"loss": 3.1056,
"theoretical_loss": 3.7928323600871625,
"tokens_seen": 676134912
},
{
"epoch": 0.2,
"learning_rate": 0.00040158475426278837,
"loss": 2.7051,
"theoretical_loss": 3.7927948029030754,
"tokens_seen": 676200448
},
{
"epoch": 0.2,
"learning_rate": 0.00040157472417251755,
"loss": 2.9152,
"theoretical_loss": 3.792757250377858,
"tokens_seen": 676265984
},
{
"epoch": 0.2,
"learning_rate": 0.0004015646940822468,
"loss": 2.958,
"theoretical_loss": 3.792719702510481,
"tokens_seen": 676331520
},
{
"epoch": 0.2,
"learning_rate": 0.0004015546639919759,
"loss": 2.819,
"theoretical_loss": 3.7926821592999147,
"tokens_seen": 676397056
},
{
"epoch": 0.2,
"learning_rate": 0.00040154463390170515,
"loss": 3.0959,
"theoretical_loss": 3.79264462074513,
"tokens_seen": 676462592
},
{
"epoch": 0.21,
"learning_rate": 0.00040153460381143433,
"loss": 2.9016,
"theoretical_loss": 3.7926070868451003,
"tokens_seen": 676528128
},
{
"epoch": 0.21,
"learning_rate": 0.0004015245737211635,
"loss": 2.9593,
"theoretical_loss": 3.7925695575987963,
"tokens_seen": 676593664
},
{
"epoch": 0.21,
"objective/train/docs_used": 392138,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0610926151275635,
"objective/train/theoretical_loss": 3.792532033005191,
"objective/train/tokens_used": 697119200,
"theoretical_loss": 3.792532033005191,
"tokens_seen": 676659200
},
{
"epoch": 0.21,
"learning_rate": 0.0004015145436308927,
"loss": 2.8106,
"theoretical_loss": 3.792532033005191,
"tokens_seen": 676659200
},
{
"epoch": 0.21,
"learning_rate": 0.0004015045135406219,
"loss": 3.0789,
"theoretical_loss": 3.7924945130632564,
"tokens_seen": 676724736
},
{
"epoch": 0.21,
"learning_rate": 0.00040149448345035106,
"loss": 3.0028,
"theoretical_loss": 3.7924569977719655,
"tokens_seen": 676790272
},
{
"epoch": 0.21,
"learning_rate": 0.0004014844533600803,
"loss": 2.6171,
"theoretical_loss": 3.792419487130292,
"tokens_seen": 676855808
},
{
"epoch": 0.21,
"learning_rate": 0.0004014744232698094,
"loss": 2.96,
"theoretical_loss": 3.792381981137209,
"tokens_seen": 676921344
},
{
"epoch": 0.21,
"learning_rate": 0.00040146439317953865,
"loss": 2.859,
"theoretical_loss": 3.7923444797916916,
"tokens_seen": 676986880
},
{
"epoch": 0.21,
"learning_rate": 0.0004014543630892678,
"loss": 3.0644,
"theoretical_loss": 3.7923069830927125,
"tokens_seen": 677052416
},
{
"epoch": 0.21,
"learning_rate": 0.000401444332998997,
"loss": 2.5961,
"theoretical_loss": 3.7922694910392476,
"tokens_seen": 677117952
},
{
"epoch": 0.21,
"learning_rate": 0.0004014343029087262,
"loss": 3.0954,
"theoretical_loss": 3.792232003630271,
"tokens_seen": 677183488
},
{
"epoch": 0.21,
"learning_rate": 0.0004014242728184554,
"loss": 2.8145,
"theoretical_loss": 3.7921945208647583,
"tokens_seen": 677249024
},
{
"epoch": 0.21,
"learning_rate": 0.00040141424272818456,
"loss": 2.8047,
"theoretical_loss": 3.792157042741685,
"tokens_seen": 677314560
},
{
"epoch": 0.21,
"learning_rate": 0.0004014042126379138,
"loss": 2.8107,
"theoretical_loss": 3.792119569260027,
"tokens_seen": 677380096
},
{
"epoch": 0.21,
"learning_rate": 0.0004013941825476429,
"loss": 2.7587,
"theoretical_loss": 3.79208210041876,
"tokens_seen": 677445632
},
{
"epoch": 0.21,
"learning_rate": 0.00040138415245737216,
"loss": 2.7552,
"theoretical_loss": 3.792044636216862,
"tokens_seen": 677511168
},
{
"epoch": 0.21,
"learning_rate": 0.0004013741223671013,
"loss": 2.9734,
"theoretical_loss": 3.7920071766533088,
"tokens_seen": 677576704
},
{
"epoch": 0.21,
"learning_rate": 0.0004013640922768305,
"loss": 2.972,
"theoretical_loss": 3.791969721727077,
"tokens_seen": 677642240
},
{
"epoch": 0.21,
"learning_rate": 0.0004013540621865597,
"loss": 2.86,
"theoretical_loss": 3.7919322714371457,
"tokens_seen": 677707776
},
{
"epoch": 0.21,
"learning_rate": 0.0004013440320962889,
"loss": 2.8107,
"theoretical_loss": 3.7918948257824914,
"tokens_seen": 677773312
},
{
"epoch": 0.21,
"learning_rate": 0.00040133400200601806,
"loss": 3.0305,
"theoretical_loss": 3.7918573847620927,
"tokens_seen": 677838848
},
{
"epoch": 0.21,
"learning_rate": 0.00040132397191574724,
"loss": 2.8221,
"theoretical_loss": 3.791819948374928,
"tokens_seen": 677904384
},
{
"epoch": 0.21,
"learning_rate": 0.0004013139418254764,
"loss": 2.7825,
"theoretical_loss": 3.791782516619977,
"tokens_seen": 677969920
},
{
"epoch": 0.21,
"learning_rate": 0.00040130391173520566,
"loss": 2.8065,
"theoretical_loss": 3.7917450894962177,
"tokens_seen": 678035456
},
{
"epoch": 0.21,
"learning_rate": 0.0004012938816449348,
"loss": 2.7811,
"theoretical_loss": 3.7917076670026297,
"tokens_seen": 678100992
},
{
"epoch": 0.21,
"learning_rate": 0.000401283851554664,
"loss": 2.7044,
"theoretical_loss": 3.7916702491381935,
"tokens_seen": 678166528
},
{
"epoch": 0.21,
"learning_rate": 0.00040127382146439315,
"loss": 2.8996,
"theoretical_loss": 3.791632835901889,
"tokens_seen": 678232064
},
{
"epoch": 0.21,
"objective/train/docs_used": 393619,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6656699180603027,
"objective/train/theoretical_loss": 3.7915954272926955,
"objective/train/tokens_used": 698757600,
"theoretical_loss": 3.7915954272926955,
"tokens_seen": 678297600
},
{
"epoch": 0.21,
"learning_rate": 0.0004012637913741224,
"loss": 3.056,
"theoretical_loss": 3.7915954272926955,
"tokens_seen": 678297600
},
{
"epoch": 0.21,
"learning_rate": 0.00040125376128385157,
"loss": 2.7484,
"theoretical_loss": 3.7915580233095953,
"tokens_seen": 678363136
},
{
"epoch": 0.21,
"learning_rate": 0.00040124373119358075,
"loss": 2.8545,
"theoretical_loss": 3.791520623951569,
"tokens_seen": 678428672
},
{
"epoch": 0.21,
"learning_rate": 0.00040123370110330993,
"loss": 3.0677,
"theoretical_loss": 3.7914832292175973,
"tokens_seen": 678494208
},
{
"epoch": 0.21,
"learning_rate": 0.00040122367101303916,
"loss": 2.9819,
"theoretical_loss": 3.7914458391066628,
"tokens_seen": 678559744
},
{
"epoch": 0.21,
"learning_rate": 0.0004012136409227683,
"loss": 3.0088,
"theoretical_loss": 3.7914084536177475,
"tokens_seen": 678625280
},
{
"epoch": 0.21,
"learning_rate": 0.0004012036108324975,
"loss": 3.0062,
"theoretical_loss": 3.7913710727498335,
"tokens_seen": 678690816
},
{
"epoch": 0.21,
"learning_rate": 0.00040119358074222665,
"loss": 2.7612,
"theoretical_loss": 3.791333696501903,
"tokens_seen": 678756352
},
{
"epoch": 0.21,
"learning_rate": 0.0004011835506519559,
"loss": 2.8295,
"theoretical_loss": 3.7912963248729397,
"tokens_seen": 678821888
},
{
"epoch": 0.21,
"learning_rate": 0.00040117352056168507,
"loss": 2.9354,
"theoretical_loss": 3.7912589578619267,
"tokens_seen": 678887424
},
{
"epoch": 0.21,
"learning_rate": 0.00040116349047141425,
"loss": 2.6009,
"theoretical_loss": 3.7912215954678476,
"tokens_seen": 678952960
},
{
"epoch": 0.21,
"learning_rate": 0.00040115346038114343,
"loss": 2.7383,
"theoretical_loss": 3.791184237689687,
"tokens_seen": 679018496
},
{
"epoch": 0.21,
"learning_rate": 0.0004011434302908726,
"loss": 2.7452,
"theoretical_loss": 3.791146884526428,
"tokens_seen": 679084032
},
{
"epoch": 0.21,
"learning_rate": 0.0004011334002006018,
"loss": 2.8184,
"theoretical_loss": 3.791109535977056,
"tokens_seen": 679149568
},
{
"epoch": 0.21,
"learning_rate": 0.00040112337011033103,
"loss": 2.8995,
"theoretical_loss": 3.7910721920405566,
"tokens_seen": 679215104
},
{
"epoch": 0.21,
"learning_rate": 0.00040111334002006016,
"loss": 2.7905,
"theoretical_loss": 3.7910348527159137,
"tokens_seen": 679280640
},
{
"epoch": 0.21,
"learning_rate": 0.0004011033099297894,
"loss": 2.7154,
"theoretical_loss": 3.790997518002113,
"tokens_seen": 679346176
},
{
"epoch": 0.21,
"learning_rate": 0.0004010932798395185,
"loss": 2.7819,
"theoretical_loss": 3.790960187898142,
"tokens_seen": 679411712
},
{
"epoch": 0.21,
"learning_rate": 0.00040108324974924775,
"loss": 2.6947,
"theoretical_loss": 3.7909228624029856,
"tokens_seen": 679477248
},
{
"epoch": 0.21,
"learning_rate": 0.00040107321965897693,
"loss": 2.7342,
"theoretical_loss": 3.79088554151563,
"tokens_seen": 679542784
},
{
"epoch": 0.21,
"learning_rate": 0.0004010631895687061,
"loss": 2.7346,
"theoretical_loss": 3.790848225235063,
"tokens_seen": 679608320
},
{
"epoch": 0.21,
"learning_rate": 0.0004010531594784353,
"loss": 2.8089,
"theoretical_loss": 3.7908109135602714,
"tokens_seen": 679673856
},
{
"epoch": 0.21,
"learning_rate": 0.00040104312938816453,
"loss": 3.038,
"theoretical_loss": 3.790773606490243,
"tokens_seen": 679739392
},
{
"epoch": 0.21,
"learning_rate": 0.00040103309929789366,
"loss": 2.8085,
"theoretical_loss": 3.7907363040239654,
"tokens_seen": 679804928
},
{
"epoch": 0.21,
"learning_rate": 0.0004010230692076229,
"loss": 2.7163,
"theoretical_loss": 3.790699006160427,
"tokens_seen": 679870464
},
{
"epoch": 0.21,
"objective/train/docs_used": 393966,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.271401882171631,
"objective/train/theoretical_loss": 3.7906617128986153,
"objective/train/tokens_used": 700396000,
"theoretical_loss": 3.7906617128986153,
"tokens_seen": 679936000
},
{
"epoch": 0.21,
"learning_rate": 0.000401013039117352,
"loss": 2.6118,
"theoretical_loss": 3.7906617128986153,
"tokens_seen": 679936000
},
{
"epoch": 0.21,
"learning_rate": 0.00040100300902708126,
"loss": 2.6319,
"theoretical_loss": 3.79062442423752,
"tokens_seen": 680001536
},
{
"epoch": 0.21,
"learning_rate": 0.00040099297893681044,
"loss": 3.0146,
"theoretical_loss": 3.7905871401761306,
"tokens_seen": 680067072
},
{
"epoch": 0.21,
"learning_rate": 0.0004009829488465396,
"loss": 2.6899,
"theoretical_loss": 3.790549860713435,
"tokens_seen": 680132608
},
{
"epoch": 0.21,
"learning_rate": 0.0004009729187562688,
"loss": 2.8361,
"theoretical_loss": 3.7905125858484245,
"tokens_seen": 680198144
},
{
"epoch": 0.21,
"learning_rate": 0.000400962888665998,
"loss": 3.0412,
"theoretical_loss": 3.7904753155800885,
"tokens_seen": 680263680
},
{
"epoch": 0.21,
"learning_rate": 0.00040095285857572716,
"loss": 2.7734,
"theoretical_loss": 3.7904380499074177,
"tokens_seen": 680329216
},
{
"epoch": 0.21,
"learning_rate": 0.0004009428284854564,
"loss": 2.8019,
"theoretical_loss": 3.7904007888294027,
"tokens_seen": 680394752
},
{
"epoch": 0.21,
"learning_rate": 0.0004009327983951855,
"loss": 2.8287,
"theoretical_loss": 3.790363532345034,
"tokens_seen": 680460288
},
{
"epoch": 0.21,
"learning_rate": 0.00040092276830491476,
"loss": 2.7719,
"theoretical_loss": 3.790326280453303,
"tokens_seen": 680525824
},
{
"epoch": 0.21,
"learning_rate": 0.0004009127382146439,
"loss": 2.5285,
"theoretical_loss": 3.790289033153202,
"tokens_seen": 680591360
},
{
"epoch": 0.21,
"learning_rate": 0.0004009027081243731,
"loss": 2.8977,
"theoretical_loss": 3.7902517904437225,
"tokens_seen": 680656896
},
{
"epoch": 0.21,
"learning_rate": 0.00040089267803410236,
"loss": 2.8517,
"theoretical_loss": 3.7902145523238575,
"tokens_seen": 680722432
},
{
"epoch": 0.21,
"learning_rate": 0.0004008826479438315,
"loss": 2.7916,
"theoretical_loss": 3.790177318792599,
"tokens_seen": 680787968
},
{
"epoch": 0.21,
"learning_rate": 0.0004008726178535607,
"loss": 2.8179,
"theoretical_loss": 3.79014008984894,
"tokens_seen": 680853504
},
{
"epoch": 0.21,
"learning_rate": 0.0004008625877632899,
"loss": 3.0886,
"theoretical_loss": 3.790102865491874,
"tokens_seen": 680919040
},
{
"epoch": 0.21,
"learning_rate": 0.0004008525576730191,
"loss": 2.4511,
"theoretical_loss": 3.7900656457203934,
"tokens_seen": 680984576
},
{
"epoch": 0.21,
"learning_rate": 0.00040084252758274826,
"loss": 2.7924,
"theoretical_loss": 3.7900284305334937,
"tokens_seen": 681050112
},
{
"epoch": 0.21,
"learning_rate": 0.00040083249749247744,
"loss": 2.9634,
"theoretical_loss": 3.7899912199301684,
"tokens_seen": 681115648
},
{
"epoch": 0.21,
"learning_rate": 0.0004008224674022066,
"loss": 2.807,
"theoretical_loss": 3.789954013909412,
"tokens_seen": 681181184
},
{
"epoch": 0.21,
"learning_rate": 0.00040081243731193586,
"loss": 2.8455,
"theoretical_loss": 3.78991681247022,
"tokens_seen": 681246720
},
{
"epoch": 0.21,
"learning_rate": 0.000400802407221665,
"loss": 2.8272,
"theoretical_loss": 3.7898796156115866,
"tokens_seen": 681312256
},
{
"epoch": 0.21,
"learning_rate": 0.0004007923771313942,
"loss": 2.4782,
"theoretical_loss": 3.7898424233325074,
"tokens_seen": 681377792
},
{
"epoch": 0.21,
"learning_rate": 0.00040078234704112335,
"loss": 2.7965,
"theoretical_loss": 3.7898052356319787,
"tokens_seen": 681443328
},
{
"epoch": 0.21,
"learning_rate": 0.0004007723169508526,
"loss": 2.8131,
"theoretical_loss": 3.7897680525089967,
"tokens_seen": 681508864
},
{
"epoch": 0.21,
"objective/train/docs_used": 395160,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6314914226531982,
"objective/train/theoretical_loss": 3.789730873962557,
"objective/train/tokens_used": 702034400,
"theoretical_loss": 3.789730873962557,
"tokens_seen": 681574400
},
{
"epoch": 0.21,
"learning_rate": 0.00040076228686058177,
"loss": 2.6275,
"theoretical_loss": 3.789730873962557,
"tokens_seen": 681574400
},
{
"epoch": 0.21,
"learning_rate": 0.00040075225677031095,
"loss": 2.8045,
"theoretical_loss": 3.789693699991657,
"tokens_seen": 681639936
},
{
"epoch": 0.21,
"learning_rate": 0.00040074222668004013,
"loss": 2.851,
"theoretical_loss": 3.7896565305952934,
"tokens_seen": 681705472
},
{
"epoch": 0.21,
"learning_rate": 0.00040073219658976936,
"loss": 3.0511,
"theoretical_loss": 3.789619365772464,
"tokens_seen": 681771008
},
{
"epoch": 0.21,
"learning_rate": 0.0004007221664994985,
"loss": 2.8067,
"theoretical_loss": 3.7895822055221657,
"tokens_seen": 681836544
},
{
"epoch": 0.21,
"learning_rate": 0.0004007121364092277,
"loss": 2.9501,
"theoretical_loss": 3.7895450498433982,
"tokens_seen": 681902080
},
{
"epoch": 0.21,
"learning_rate": 0.00040070210631895685,
"loss": 2.9843,
"theoretical_loss": 3.789507898735158,
"tokens_seen": 681967616
},
{
"epoch": 0.21,
"learning_rate": 0.0004006920762286861,
"loss": 3.0171,
"theoretical_loss": 3.7894707521964444,
"tokens_seen": 682033152
},
{
"epoch": 0.21,
"learning_rate": 0.00040068204613841527,
"loss": 2.5498,
"theoretical_loss": 3.789433610226256,
"tokens_seen": 682098688
},
{
"epoch": 0.21,
"learning_rate": 0.00040067201604814445,
"loss": 2.8422,
"theoretical_loss": 3.789396472823593,
"tokens_seen": 682164224
},
{
"epoch": 0.21,
"learning_rate": 0.00040066198595787363,
"loss": 2.897,
"theoretical_loss": 3.7893593399874548,
"tokens_seen": 682229760
},
{
"epoch": 0.21,
"learning_rate": 0.0004006519558676028,
"loss": 2.6587,
"theoretical_loss": 3.78932221171684,
"tokens_seen": 682295296
},
{
"epoch": 0.21,
"learning_rate": 0.000400641925777332,
"loss": 2.9829,
"theoretical_loss": 3.7892850880107503,
"tokens_seen": 682360832
},
{
"epoch": 0.21,
"learning_rate": 0.00040063189568706123,
"loss": 2.79,
"theoretical_loss": 3.789247968868186,
"tokens_seen": 682426368
},
{
"epoch": 0.21,
"learning_rate": 0.00040062186559679036,
"loss": 2.9686,
"theoretical_loss": 3.7892108542881475,
"tokens_seen": 682491904
},
{
"epoch": 0.21,
"learning_rate": 0.0004006118355065196,
"loss": 2.7489,
"theoretical_loss": 3.7891737442696356,
"tokens_seen": 682557440
},
{
"epoch": 0.21,
"learning_rate": 0.0004006018054162487,
"loss": 2.8805,
"theoretical_loss": 3.7891366388116525,
"tokens_seen": 682622976
},
{
"epoch": 0.21,
"learning_rate": 0.00040059177532597795,
"loss": 2.8078,
"theoretical_loss": 3.7890995379131995,
"tokens_seen": 682688512
},
{
"epoch": 0.21,
"learning_rate": 0.00040058174523570713,
"loss": 2.9302,
"theoretical_loss": 3.7890624415732796,
"tokens_seen": 682754048
},
{
"epoch": 0.21,
"learning_rate": 0.0004005717151454363,
"loss": 2.9213,
"theoretical_loss": 3.789025349790894,
"tokens_seen": 682819584
},
{
"epoch": 0.21,
"learning_rate": 0.0004005616850551655,
"loss": 2.9542,
"theoretical_loss": 3.7889882625650464,
"tokens_seen": 682885120
},
{
"epoch": 0.21,
"learning_rate": 0.00040055165496489473,
"loss": 2.5394,
"theoretical_loss": 3.788951179894739,
"tokens_seen": 682950656
},
{
"epoch": 0.21,
"learning_rate": 0.00040054162487462386,
"loss": 2.8057,
"theoretical_loss": 3.788914101778976,
"tokens_seen": 683016192
},
{
"epoch": 0.21,
"learning_rate": 0.0004005315947843531,
"loss": 3.0619,
"theoretical_loss": 3.7888770282167608,
"tokens_seen": 683081728
},
{
"epoch": 0.21,
"learning_rate": 0.0004005215646940822,
"loss": 2.8595,
"theoretical_loss": 3.788839959207097,
"tokens_seen": 683147264
},
{
"epoch": 0.21,
"objective/train/docs_used": 395993,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0149126052856445,
"objective/train/theoretical_loss": 3.788802894748989,
"objective/train/tokens_used": 703672800,
"theoretical_loss": 3.788802894748989,
"tokens_seen": 683212800
},
{
"epoch": 0.21,
"learning_rate": 0.00040051153460381146,
"loss": 2.9622,
"theoretical_loss": 3.788802894748989,
"tokens_seen": 683212800
},
{
"epoch": 0.21,
"learning_rate": 0.00040050150451354064,
"loss": 2.9391,
"theoretical_loss": 3.788765834841442,
"tokens_seen": 683278336
},
{
"epoch": 0.21,
"learning_rate": 0.0004004914744232698,
"loss": 2.8744,
"theoretical_loss": 3.78872877948346,
"tokens_seen": 683343872
},
{
"epoch": 0.21,
"learning_rate": 0.000400481444332999,
"loss": 2.6533,
"theoretical_loss": 3.788691728674049,
"tokens_seen": 683409408
},
{
"epoch": 0.21,
"learning_rate": 0.0004004714142427282,
"loss": 2.8744,
"theoretical_loss": 3.788654682412215,
"tokens_seen": 683474944
},
{
"epoch": 0.21,
"learning_rate": 0.00040046138415245736,
"loss": 2.8672,
"theoretical_loss": 3.7886176406969625,
"tokens_seen": 683540480
},
{
"epoch": 0.21,
"learning_rate": 0.0004004513540621866,
"loss": 2.7217,
"theoretical_loss": 3.788580603527298,
"tokens_seen": 683606016
},
{
"epoch": 0.21,
"learning_rate": 0.0004004413239719157,
"loss": 2.8089,
"theoretical_loss": 3.7885435709022284,
"tokens_seen": 683671552
},
{
"epoch": 0.21,
"learning_rate": 0.00040043129388164496,
"loss": 2.7463,
"theoretical_loss": 3.788506542820761,
"tokens_seen": 683737088
},
{
"epoch": 0.21,
"learning_rate": 0.0004004212637913741,
"loss": 2.6237,
"theoretical_loss": 3.788469519281902,
"tokens_seen": 683802624
},
{
"epoch": 0.21,
"learning_rate": 0.0004004112337011033,
"loss": 2.8902,
"theoretical_loss": 3.7884325002846593,
"tokens_seen": 683868160
},
{
"epoch": 0.21,
"learning_rate": 0.0004004012036108325,
"loss": 2.8095,
"theoretical_loss": 3.7883954858280404,
"tokens_seen": 683933696
},
{
"epoch": 0.21,
"learning_rate": 0.0004003911735205617,
"loss": 2.9476,
"theoretical_loss": 3.788358475911054,
"tokens_seen": 683999232
},
{
"epoch": 0.21,
"learning_rate": 0.00040038114343029087,
"loss": 2.8453,
"theoretical_loss": 3.788321470532707,
"tokens_seen": 684064768
},
{
"epoch": 0.21,
"learning_rate": 0.0004003711133400201,
"loss": 2.8872,
"theoretical_loss": 3.7882844696920097,
"tokens_seen": 684130304
},
{
"epoch": 0.21,
"learning_rate": 0.00040036108324974923,
"loss": 2.8312,
"theoretical_loss": 3.78824747338797,
"tokens_seen": 684195840
},
{
"epoch": 0.21,
"learning_rate": 0.00040035105315947846,
"loss": 2.6786,
"theoretical_loss": 3.788210481619598,
"tokens_seen": 684261376
},
{
"epoch": 0.21,
"learning_rate": 0.0004003410230692076,
"loss": 2.8161,
"theoretical_loss": 3.7881734943859025,
"tokens_seen": 684326912
},
{
"epoch": 0.21,
"learning_rate": 0.0004003309929789368,
"loss": 2.8005,
"theoretical_loss": 3.788136511685894,
"tokens_seen": 684392448
},
{
"epoch": 0.21,
"learning_rate": 0.000400320962888666,
"loss": 2.8623,
"theoretical_loss": 3.788099533518582,
"tokens_seen": 684457984
},
{
"epoch": 0.21,
"learning_rate": 0.0004003109327983952,
"loss": 2.9891,
"theoretical_loss": 3.7880625598829782,
"tokens_seen": 684523520
},
{
"epoch": 0.21,
"learning_rate": 0.00040030090270812437,
"loss": 2.7396,
"theoretical_loss": 3.7880255907780924,
"tokens_seen": 684589056
},
{
"epoch": 0.21,
"learning_rate": 0.00040029087261785355,
"loss": 2.7279,
"theoretical_loss": 3.787988626202937,
"tokens_seen": 684654592
},
{
"epoch": 0.21,
"learning_rate": 0.00040028084252758273,
"loss": 2.7536,
"theoretical_loss": 3.7879516661565216,
"tokens_seen": 684720128
},
{
"epoch": 0.21,
"learning_rate": 0.00040027081243731197,
"loss": 3.0539,
"theoretical_loss": 3.7879147106378595,
"tokens_seen": 684785664
},
{
"epoch": 0.21,
"objective/train/docs_used": 397274,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.094200372695923,
"objective/train/theoretical_loss": 3.787877759645963,
"objective/train/tokens_used": 705311200,
"theoretical_loss": 3.787877759645963,
"tokens_seen": 684851200
},
{
"epoch": 0.21,
"learning_rate": 0.0004002607823470411,
"loss": 2.7796,
"theoretical_loss": 3.787877759645963,
"tokens_seen": 684851200
},
{
"epoch": 0.21,
"learning_rate": 0.00040025075225677033,
"loss": 2.486,
"theoretical_loss": 3.7878408131798427,
"tokens_seen": 684916736
},
{
"epoch": 0.21,
"learning_rate": 0.00040024072216649946,
"loss": 2.8741,
"theoretical_loss": 3.787803871238513,
"tokens_seen": 684982272
},
{
"epoch": 0.21,
"learning_rate": 0.0004002306920762287,
"loss": 2.8656,
"theoretical_loss": 3.7877669338209863,
"tokens_seen": 685047808
},
{
"epoch": 0.21,
"learning_rate": 0.00040022066198595787,
"loss": 2.8117,
"theoretical_loss": 3.787730000926276,
"tokens_seen": 685113344
},
{
"epoch": 0.21,
"learning_rate": 0.00040021063189568705,
"loss": 2.5069,
"theoretical_loss": 3.787693072553396,
"tokens_seen": 685178880
},
{
"epoch": 0.21,
"learning_rate": 0.00040020060180541623,
"loss": 2.7255,
"theoretical_loss": 3.78765614870136,
"tokens_seen": 685244416
},
{
"epoch": 0.21,
"learning_rate": 0.00040019057171514547,
"loss": 3.0465,
"theoretical_loss": 3.7876192293691817,
"tokens_seen": 685309952
},
{
"epoch": 0.21,
"learning_rate": 0.0004001805416248746,
"loss": 2.8863,
"theoretical_loss": 3.787582314555877,
"tokens_seen": 685375488
},
{
"epoch": 0.21,
"learning_rate": 0.00040017051153460383,
"loss": 3.0695,
"theoretical_loss": 3.7875454042604595,
"tokens_seen": 685441024
},
{
"epoch": 0.21,
"learning_rate": 0.00040016048144433296,
"loss": 2.7426,
"theoretical_loss": 3.787508498481945,
"tokens_seen": 685506560
},
{
"epoch": 0.21,
"learning_rate": 0.0004001504513540622,
"loss": 2.9573,
"theoretical_loss": 3.7874715972193487,
"tokens_seen": 685572096
},
{
"epoch": 0.21,
"learning_rate": 0.00040014042126379143,
"loss": 2.9465,
"theoretical_loss": 3.7874347004716866,
"tokens_seen": 685637632
},
{
"epoch": 0.21,
"learning_rate": 0.00040013039117352056,
"loss": 3.0187,
"theoretical_loss": 3.787397808237975,
"tokens_seen": 685703168
},
{
"epoch": 0.21,
"learning_rate": 0.0004001203610832498,
"loss": 2.6881,
"theoretical_loss": 3.78736092051723,
"tokens_seen": 685768704
},
{
"epoch": 0.21,
"learning_rate": 0.0004001103309929789,
"loss": 2.9121,
"theoretical_loss": 3.7873240373084687,
"tokens_seen": 685834240
},
{
"epoch": 0.21,
"learning_rate": 0.00040010030090270815,
"loss": 2.9413,
"theoretical_loss": 3.7872871586107077,
"tokens_seen": 685899776
},
{
"epoch": 0.21,
"learning_rate": 0.00040009027081243733,
"loss": 2.9002,
"theoretical_loss": 3.7872502844229645,
"tokens_seen": 685965312
},
{
"epoch": 0.21,
"learning_rate": 0.0004000802407221665,
"loss": 2.9066,
"theoretical_loss": 3.787213414744257,
"tokens_seen": 686030848
},
{
"epoch": 0.21,
"learning_rate": 0.0004000702106318957,
"loss": 2.6328,
"theoretical_loss": 3.787176549573603,
"tokens_seen": 686096384
},
{
"epoch": 0.21,
"learning_rate": 0.00040006018054162493,
"loss": 2.7862,
"theoretical_loss": 3.7871396889100204,
"tokens_seen": 686161920
},
{
"epoch": 0.21,
"learning_rate": 0.00040005015045135406,
"loss": 2.9665,
"theoretical_loss": 3.7871028327525287,
"tokens_seen": 686227456
},
{
"epoch": 0.21,
"learning_rate": 0.0004000401203610833,
"loss": 2.731,
"theoretical_loss": 3.7870659811001457,
"tokens_seen": 686292992
},
{
"epoch": 0.21,
"learning_rate": 0.0004000300902708124,
"loss": 2.7807,
"theoretical_loss": 3.787029133951891,
"tokens_seen": 686358528
},
{
"epoch": 0.21,
"learning_rate": 0.00040002006018054166,
"loss": 2.9232,
"theoretical_loss": 3.7869922913067846,
"tokens_seen": 686424064
},
{
"epoch": 0.21,
"objective/train/docs_used": 397954,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1701653003692627,
"objective/train/theoretical_loss": 3.786955453163846,
"objective/train/tokens_used": 706949600,
"theoretical_loss": 3.786955453163846,
"tokens_seen": 686489600
},
{
"epoch": 0.21,
"learning_rate": 0.00040001003009027084,
"loss": 2.7742,
"theoretical_loss": 3.786955453163846,
"tokens_seen": 686489600
},
{
"epoch": 0.21,
"learning_rate": 0.0004,
"loss": 2.6134,
"theoretical_loss": 3.7869186195220945,
"tokens_seen": 686555136
},
{
"epoch": 0.21,
"learning_rate": 0.0003999899699097292,
"loss": 2.9942,
"theoretical_loss": 3.786881790380552,
"tokens_seen": 686620672
},
{
"epoch": 0.21,
"learning_rate": 0.0003999799398194584,
"loss": 2.7173,
"theoretical_loss": 3.7868449657382377,
"tokens_seen": 686686208
},
{
"epoch": 0.21,
"learning_rate": 0.00039996990972918756,
"loss": 2.6543,
"theoretical_loss": 3.786808145594174,
"tokens_seen": 686751744
},
{
"epoch": 0.21,
"learning_rate": 0.0003999598796389168,
"loss": 2.6488,
"theoretical_loss": 3.7867713299473813,
"tokens_seen": 686817280
},
{
"epoch": 0.21,
"learning_rate": 0.0003999498495486459,
"loss": 3.0095,
"theoretical_loss": 3.786734518796882,
"tokens_seen": 686882816
},
{
"epoch": 0.21,
"learning_rate": 0.00039993981945837516,
"loss": 2.7554,
"theoretical_loss": 3.786697712141697,
"tokens_seen": 686948352
},
{
"epoch": 0.21,
"learning_rate": 0.0003999297893681043,
"loss": 2.8115,
"theoretical_loss": 3.7866609099808493,
"tokens_seen": 687013888
},
{
"epoch": 0.21,
"learning_rate": 0.0003999197592778335,
"loss": 2.8301,
"theoretical_loss": 3.7866241123133615,
"tokens_seen": 687079424
},
{
"epoch": 0.21,
"learning_rate": 0.0003999097291875627,
"loss": 2.8115,
"theoretical_loss": 3.786587319138256,
"tokens_seen": 687144960
},
{
"epoch": 0.21,
"learning_rate": 0.0003998996990972919,
"loss": 2.7402,
"theoretical_loss": 3.786550530454557,
"tokens_seen": 687210496
},
{
"epoch": 0.21,
"learning_rate": 0.00039988966900702107,
"loss": 2.7467,
"theoretical_loss": 3.786513746261287,
"tokens_seen": 687276032
},
{
"epoch": 0.21,
"learning_rate": 0.0003998796389167503,
"loss": 2.7556,
"theoretical_loss": 3.78647696655747,
"tokens_seen": 687341568
},
{
"epoch": 0.21,
"learning_rate": 0.00039986960882647943,
"loss": 2.8232,
"theoretical_loss": 3.7864401913421295,
"tokens_seen": 687407104
},
{
"epoch": 0.21,
"learning_rate": 0.00039985957873620866,
"loss": 2.8257,
"theoretical_loss": 3.7864034206142914,
"tokens_seen": 687472640
},
{
"epoch": 0.21,
"learning_rate": 0.0003998495486459378,
"loss": 2.7636,
"theoretical_loss": 3.786366654372979,
"tokens_seen": 687538176
},
{
"epoch": 0.21,
"learning_rate": 0.000399839518555667,
"loss": 2.5943,
"theoretical_loss": 3.7863298926172186,
"tokens_seen": 687603712
},
{
"epoch": 0.21,
"learning_rate": 0.0003998294884653962,
"loss": 2.9188,
"theoretical_loss": 3.786293135346034,
"tokens_seen": 687669248
},
{
"epoch": 0.21,
"learning_rate": 0.0003998194583751254,
"loss": 2.9494,
"theoretical_loss": 3.786256382558452,
"tokens_seen": 687734784
},
{
"epoch": 0.21,
"learning_rate": 0.00039980942828485457,
"loss": 3.0331,
"theoretical_loss": 3.7862196342534977,
"tokens_seen": 687800320
},
{
"epoch": 0.21,
"learning_rate": 0.00039979939819458375,
"loss": 2.5892,
"theoretical_loss": 3.786182890430198,
"tokens_seen": 687865856
},
{
"epoch": 0.21,
"learning_rate": 0.00039978936810431293,
"loss": 2.7573,
"theoretical_loss": 3.7861461510875793,
"tokens_seen": 687931392
},
{
"epoch": 0.21,
"learning_rate": 0.00039977933801404217,
"loss": 2.6233,
"theoretical_loss": 3.786109416224668,
"tokens_seen": 687996928
},
{
"epoch": 0.21,
"learning_rate": 0.0003997693079237713,
"loss": 2.7465,
"theoretical_loss": 3.786072685840492,
"tokens_seen": 688062464
},
{
"epoch": 0.21,
"objective/train/docs_used": 398552,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.040328025817871,
"objective/train/theoretical_loss": 3.7860359599340776,
"objective/train/tokens_used": 708588000,
"theoretical_loss": 3.7860359599340776,
"tokens_seen": 688128000
},
{
"epoch": 0.21,
"learning_rate": 0.00039975927783350053,
"loss": 2.843,
"theoretical_loss": 3.7860359599340776,
"tokens_seen": 688128000
},
{
"epoch": 0.21,
"learning_rate": 0.00039974924774322966,
"loss": 2.7199,
"theoretical_loss": 3.7859992385044534,
"tokens_seen": 688193536
},
{
"epoch": 0.21,
"learning_rate": 0.0003997392176529589,
"loss": 3.043,
"theoretical_loss": 3.7859625215506476,
"tokens_seen": 688259072
},
{
"epoch": 0.21,
"learning_rate": 0.00039972918756268807,
"loss": 2.8198,
"theoretical_loss": 3.7859258090716876,
"tokens_seen": 688324608
},
{
"epoch": 0.21,
"learning_rate": 0.00039971915747241725,
"loss": 2.9757,
"theoretical_loss": 3.785889101066603,
"tokens_seen": 688390144
},
{
"epoch": 0.21,
"learning_rate": 0.00039970912738214643,
"loss": 2.8251,
"theoretical_loss": 3.785852397534423,
"tokens_seen": 688455680
},
{
"epoch": 0.21,
"learning_rate": 0.00039969909729187567,
"loss": 2.8572,
"theoretical_loss": 3.7858156984741758,
"tokens_seen": 688521216
},
{
"epoch": 0.21,
"learning_rate": 0.0003996890672016048,
"loss": 2.8729,
"theoretical_loss": 3.7857790038848913,
"tokens_seen": 688586752
},
{
"epoch": 0.21,
"learning_rate": 0.00039967903711133403,
"loss": 2.8814,
"theoretical_loss": 3.7857423137655997,
"tokens_seen": 688652288
},
{
"epoch": 0.21,
"learning_rate": 0.00039966900702106316,
"loss": 2.7398,
"theoretical_loss": 3.785705628115331,
"tokens_seen": 688717824
},
{
"epoch": 0.21,
"learning_rate": 0.0003996589769307924,
"loss": 2.7404,
"theoretical_loss": 3.7856689469331153,
"tokens_seen": 688783360
},
{
"epoch": 0.21,
"learning_rate": 0.0003996489468405216,
"loss": 2.7664,
"theoretical_loss": 3.785632270217984,
"tokens_seen": 688848896
},
{
"epoch": 0.21,
"learning_rate": 0.00039963891675025076,
"loss": 2.6706,
"theoretical_loss": 3.7855955979689684,
"tokens_seen": 688914432
},
{
"epoch": 0.21,
"learning_rate": 0.00039962888665997994,
"loss": 2.9331,
"theoretical_loss": 3.7855589301850987,
"tokens_seen": 688979968
},
{
"epoch": 0.21,
"learning_rate": 0.0003996188565697091,
"loss": 2.941,
"theoretical_loss": 3.7855222668654074,
"tokens_seen": 689045504
},
{
"epoch": 0.21,
"learning_rate": 0.0003996088264794383,
"loss": 2.852,
"theoretical_loss": 3.7854856080089268,
"tokens_seen": 689111040
},
{
"epoch": 0.21,
"learning_rate": 0.00039959879638916754,
"loss": 2.824,
"theoretical_loss": 3.7854489536146883,
"tokens_seen": 689176576
},
{
"epoch": 0.21,
"learning_rate": 0.00039958876629889666,
"loss": 2.6997,
"theoretical_loss": 3.785412303681725,
"tokens_seen": 689242112
},
{
"epoch": 0.21,
"learning_rate": 0.0003995787362086259,
"loss": 2.8966,
"theoretical_loss": 3.7853756582090705,
"tokens_seen": 689307648
},
{
"epoch": 0.21,
"learning_rate": 0.0003995687061183551,
"loss": 2.7355,
"theoretical_loss": 3.7853390171957564,
"tokens_seen": 689373184
},
{
"epoch": 0.21,
"learning_rate": 0.00039955867602808426,
"loss": 3.0045,
"theoretical_loss": 3.7853023806408173,
"tokens_seen": 689438720
},
{
"epoch": 0.21,
"learning_rate": 0.00039954864593781344,
"loss": 2.8267,
"theoretical_loss": 3.785265748543287,
"tokens_seen": 689504256
},
{
"epoch": 0.21,
"learning_rate": 0.0003995386158475426,
"loss": 2.7749,
"theoretical_loss": 3.7852291209021987,
"tokens_seen": 689569792
},
{
"epoch": 0.21,
"learning_rate": 0.0003995285857572718,
"loss": 3.0758,
"theoretical_loss": 3.785192497716588,
"tokens_seen": 689635328
},
{
"epoch": 0.21,
"learning_rate": 0.00039951855566700104,
"loss": 2.7149,
"theoretical_loss": 3.7851558789854884,
"tokens_seen": 689700864
},
{
"epoch": 0.21,
"objective/train/docs_used": 399765,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.638550281524658,
"objective/train/theoretical_loss": 3.785119264707936,
"objective/train/tokens_used": 710226400,
"theoretical_loss": 3.785119264707936,
"tokens_seen": 689766400
},
{
"epoch": 0.21,
"learning_rate": 0.00039950852557673017,
"loss": 2.9134,
"theoretical_loss": 3.785119264707936,
"tokens_seen": 689766400
},
{
"epoch": 0.21,
"learning_rate": 0.0003994984954864594,
"loss": 2.9561,
"theoretical_loss": 3.7850826548829652,
"tokens_seen": 689831936
},
{
"epoch": 0.21,
"learning_rate": 0.00039948846539618853,
"loss": 2.9141,
"theoretical_loss": 3.7850460495096123,
"tokens_seen": 689897472
},
{
"epoch": 0.21,
"learning_rate": 0.00039947843530591776,
"loss": 2.8904,
"theoretical_loss": 3.785009448586913,
"tokens_seen": 689963008
},
{
"epoch": 0.21,
"learning_rate": 0.00039946840521564694,
"loss": 3.0049,
"theoretical_loss": 3.784972852113903,
"tokens_seen": 690028544
},
{
"epoch": 0.21,
"learning_rate": 0.0003994583751253761,
"loss": 2.4608,
"theoretical_loss": 3.78493626008962,
"tokens_seen": 690094080
},
{
"epoch": 0.21,
"learning_rate": 0.0003994483450351053,
"loss": 2.7052,
"theoretical_loss": 3.7848996725130992,
"tokens_seen": 690159616
},
{
"epoch": 0.21,
"learning_rate": 0.0003994383149448345,
"loss": 2.8319,
"theoretical_loss": 3.7848630893833786,
"tokens_seen": 690225152
},
{
"epoch": 0.21,
"learning_rate": 0.00039942828485456367,
"loss": 2.9955,
"theoretical_loss": 3.784826510699496,
"tokens_seen": 690290688
},
{
"epoch": 0.21,
"learning_rate": 0.0003994182547642929,
"loss": 2.8431,
"theoretical_loss": 3.7847899364604887,
"tokens_seen": 690356224
},
{
"epoch": 0.21,
"learning_rate": 0.00039940822467402203,
"loss": 2.8027,
"theoretical_loss": 3.7847533666653943,
"tokens_seen": 690421760
},
{
"epoch": 0.21,
"learning_rate": 0.00039939819458375127,
"loss": 2.7308,
"theoretical_loss": 3.7847168013132513,
"tokens_seen": 690487296
},
{
"epoch": 0.21,
"learning_rate": 0.0003993881644934805,
"loss": 2.848,
"theoretical_loss": 3.7846802404030986,
"tokens_seen": 690552832
},
{
"epoch": 0.21,
"learning_rate": 0.00039937813440320963,
"loss": 2.6873,
"theoretical_loss": 3.7846436839339743,
"tokens_seen": 690618368
},
{
"epoch": 0.21,
"learning_rate": 0.00039936810431293886,
"loss": 2.6324,
"theoretical_loss": 3.7846071319049193,
"tokens_seen": 690683904
},
{
"epoch": 0.21,
"learning_rate": 0.000399358074222668,
"loss": 3.0506,
"theoretical_loss": 3.7845705843149715,
"tokens_seen": 690749440
},
{
"epoch": 0.21,
"learning_rate": 0.0003993480441323972,
"loss": 2.7609,
"theoretical_loss": 3.7845340411631714,
"tokens_seen": 690814976
},
{
"epoch": 0.21,
"learning_rate": 0.0003993380140421264,
"loss": 2.7692,
"theoretical_loss": 3.7844975024485583,
"tokens_seen": 690880512
},
{
"epoch": 0.21,
"learning_rate": 0.0003993279839518556,
"loss": 2.9604,
"theoretical_loss": 3.7844609681701735,
"tokens_seen": 690946048
},
{
"epoch": 0.21,
"learning_rate": 0.00039931795386158477,
"loss": 2.7427,
"theoretical_loss": 3.7844244383270578,
"tokens_seen": 691011584
},
{
"epoch": 0.21,
"learning_rate": 0.00039930792377131395,
"loss": 2.9083,
"theoretical_loss": 3.7843879129182514,
"tokens_seen": 691077120
},
{
"epoch": 0.21,
"learning_rate": 0.00039929789368104313,
"loss": 3.0211,
"theoretical_loss": 3.7843513919427965,
"tokens_seen": 691142656
},
{
"epoch": 0.21,
"learning_rate": 0.00039928786359077237,
"loss": 2.7665,
"theoretical_loss": 3.7843148753997333,
"tokens_seen": 691208192
},
{
"epoch": 0.21,
"learning_rate": 0.0003992778335005015,
"loss": 2.9086,
"theoretical_loss": 3.784278363288105,
"tokens_seen": 691273728
},
{
"epoch": 0.21,
"learning_rate": 0.00039926780341023073,
"loss": 3.0024,
"theoretical_loss": 3.7842418556069535,
"tokens_seen": 691339264
},
{
"epoch": 0.21,
"objective/train/docs_used": 400468,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.848792314529419,
"objective/train/theoretical_loss": 3.784205352355321,
"objective/train/tokens_used": 711864800,
"theoretical_loss": 3.784205352355321,
"tokens_seen": 691404800
},
{
"epoch": 0.21,
"learning_rate": 0.00039925777331995986,
"loss": 2.6394,
"theoretical_loss": 3.784205352355321,
"tokens_seen": 691404800
},
{
"epoch": 0.21,
"learning_rate": 0.0003992477432296891,
"loss": 2.5358,
"theoretical_loss": 3.7841688535322504,
"tokens_seen": 691470336
},
{
"epoch": 0.21,
"learning_rate": 0.00039923771313941827,
"loss": 2.8016,
"theoretical_loss": 3.7841323591367844,
"tokens_seen": 691535872
},
{
"epoch": 0.21,
"learning_rate": 0.00039922768304914745,
"loss": 2.9141,
"theoretical_loss": 3.7840958691679667,
"tokens_seen": 691601408
},
{
"epoch": 0.21,
"learning_rate": 0.00039921765295887663,
"loss": 2.7175,
"theoretical_loss": 3.7840593836248413,
"tokens_seen": 691666944
},
{
"epoch": 0.21,
"learning_rate": 0.00039920762286860587,
"loss": 3.0346,
"theoretical_loss": 3.784022902506452,
"tokens_seen": 691732480
},
{
"epoch": 0.21,
"learning_rate": 0.000399197592778335,
"loss": 2.8777,
"theoretical_loss": 3.7839864258118427,
"tokens_seen": 691798016
},
{
"epoch": 0.21,
"learning_rate": 0.00039918756268806423,
"loss": 3.0047,
"theoretical_loss": 3.7839499535400583,
"tokens_seen": 691863552
},
{
"epoch": 0.21,
"learning_rate": 0.00039917753259779336,
"loss": 2.8748,
"theoretical_loss": 3.7839134856901437,
"tokens_seen": 691929088
},
{
"epoch": 0.21,
"learning_rate": 0.0003991675025075226,
"loss": 2.7448,
"theoretical_loss": 3.7838770222611435,
"tokens_seen": 691994624
},
{
"epoch": 0.21,
"learning_rate": 0.0003991574724172518,
"loss": 2.8214,
"theoretical_loss": 3.7838405632521037,
"tokens_seen": 692060160
},
{
"epoch": 0.21,
"learning_rate": 0.00039914744232698096,
"loss": 2.797,
"theoretical_loss": 3.7838041086620704,
"tokens_seen": 692125696
},
{
"epoch": 0.21,
"learning_rate": 0.00039913741223671014,
"loss": 2.813,
"theoretical_loss": 3.7837676584900883,
"tokens_seen": 692191232
},
{
"epoch": 0.21,
"learning_rate": 0.0003991273821464393,
"loss": 2.8504,
"theoretical_loss": 3.7837312127352054,
"tokens_seen": 692256768
},
{
"epoch": 0.21,
"learning_rate": 0.0003991173520561685,
"loss": 2.7765,
"theoretical_loss": 3.7836947713964673,
"tokens_seen": 692322304
},
{
"epoch": 0.21,
"learning_rate": 0.00039910732196589774,
"loss": 2.8783,
"theoretical_loss": 3.783658334472921,
"tokens_seen": 692387840
},
{
"epoch": 0.21,
"learning_rate": 0.00039909729187562686,
"loss": 2.8618,
"theoretical_loss": 3.783621901963614,
"tokens_seen": 692453376
},
{
"epoch": 0.21,
"learning_rate": 0.0003990872617853561,
"loss": 2.753,
"theoretical_loss": 3.7835854738675936,
"tokens_seen": 692518912
},
{
"epoch": 0.21,
"learning_rate": 0.0003990772316950853,
"loss": 2.9642,
"theoretical_loss": 3.7835490501839075,
"tokens_seen": 692584448
},
{
"epoch": 0.21,
"learning_rate": 0.00039906720160481446,
"loss": 2.7674,
"theoretical_loss": 3.7835126309116047,
"tokens_seen": 692649984
},
{
"epoch": 0.21,
"learning_rate": 0.00039905717151454364,
"loss": 2.6269,
"theoretical_loss": 3.7834762160497326,
"tokens_seen": 692715520
},
{
"epoch": 0.21,
"learning_rate": 0.0003990471414242728,
"loss": 3.066,
"theoretical_loss": 3.7834398055973404,
"tokens_seen": 692781056
},
{
"epoch": 0.21,
"learning_rate": 0.000399037111334002,
"loss": 2.742,
"theoretical_loss": 3.7834033995534764,
"tokens_seen": 692846592
},
{
"epoch": 0.21,
"learning_rate": 0.00039902708124373124,
"loss": 2.8775,
"theoretical_loss": 3.7833669979171916,
"tokens_seen": 692912128
},
{
"epoch": 0.21,
"learning_rate": 0.00039901705115346037,
"loss": 2.9756,
"theoretical_loss": 3.7833306006875334,
"tokens_seen": 692977664
},
{
"epoch": 0.21,
"objective/train/docs_used": 401917,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.927727460861206,
"objective/train/theoretical_loss": 3.7832942078635527,
"objective/train/tokens_used": 713503200,
"theoretical_loss": 3.7832942078635527,
"tokens_seen": 693043200
},
{
"epoch": 0.21,
"learning_rate": 0.0003990070210631896,
"loss": 2.9337,
"theoretical_loss": 3.7832942078635527,
"tokens_seen": 693043200
},
{
"epoch": 0.21,
"learning_rate": 0.00039899699097291873,
"loss": 2.9401,
"theoretical_loss": 3.7832578194443007,
"tokens_seen": 693108736
},
{
"epoch": 0.21,
"learning_rate": 0.00039898696088264796,
"loss": 2.9241,
"theoretical_loss": 3.7832214354288265,
"tokens_seen": 693174272
},
{
"epoch": 0.21,
"learning_rate": 0.00039897693079237714,
"loss": 2.8671,
"theoretical_loss": 3.783185055816181,
"tokens_seen": 693239808
},
{
"epoch": 0.21,
"learning_rate": 0.0003989669007021063,
"loss": 3.1308,
"theoretical_loss": 3.7831486806054158,
"tokens_seen": 693305344
},
{
"epoch": 0.21,
"learning_rate": 0.0003989568706118355,
"loss": 2.3691,
"theoretical_loss": 3.783112309795582,
"tokens_seen": 693370880
},
{
"epoch": 0.21,
"learning_rate": 0.0003989468405215647,
"loss": 2.6914,
"theoretical_loss": 3.783075943385731,
"tokens_seen": 693436416
},
{
"epoch": 0.21,
"learning_rate": 0.00039893681043129387,
"loss": 2.6842,
"theoretical_loss": 3.783039581374915,
"tokens_seen": 693501952
},
{
"epoch": 0.21,
"learning_rate": 0.0003989267803410231,
"loss": 2.8924,
"theoretical_loss": 3.7830032237621865,
"tokens_seen": 693567488
},
{
"epoch": 0.21,
"learning_rate": 0.00039891675025075223,
"loss": 2.9746,
"theoretical_loss": 3.7829668705465975,
"tokens_seen": 693633024
},
{
"epoch": 0.21,
"learning_rate": 0.00039890672016048147,
"loss": 2.8291,
"theoretical_loss": 3.7829305217272013,
"tokens_seen": 693698560
},
{
"epoch": 0.21,
"learning_rate": 0.00039889669007021065,
"loss": 2.7735,
"theoretical_loss": 3.7828941773030507,
"tokens_seen": 693764096
},
{
"epoch": 0.21,
"learning_rate": 0.00039888665997993983,
"loss": 2.8931,
"theoretical_loss": 3.782857837273199,
"tokens_seen": 693829632
},
{
"epoch": 0.21,
"learning_rate": 0.000398876629889669,
"loss": 2.7464,
"theoretical_loss": 3.782821501636701,
"tokens_seen": 693895168
},
{
"epoch": 0.21,
"learning_rate": 0.0003988665997993982,
"loss": 2.6741,
"theoretical_loss": 3.782785170392609,
"tokens_seen": 693960704
},
{
"epoch": 0.21,
"learning_rate": 0.00039885656970912737,
"loss": 2.6329,
"theoretical_loss": 3.782748843539978,
"tokens_seen": 694026240
},
{
"epoch": 0.21,
"learning_rate": 0.0003988465396188566,
"loss": 3.0756,
"theoretical_loss": 3.7827125210778636,
"tokens_seen": 694091776
},
{
"epoch": 0.21,
"learning_rate": 0.00039883650952858573,
"loss": 2.7004,
"theoretical_loss": 3.7826762030053196,
"tokens_seen": 694157312
},
{
"epoch": 0.21,
"learning_rate": 0.00039882647943831497,
"loss": 2.7124,
"theoretical_loss": 3.7826398893214006,
"tokens_seen": 694222848
},
{
"epoch": 0.21,
"learning_rate": 0.0003988164493480441,
"loss": 2.955,
"theoretical_loss": 3.782603580025164,
"tokens_seen": 694288384
},
{
"epoch": 0.21,
"learning_rate": 0.00039880641925777333,
"loss": 3.0134,
"theoretical_loss": 3.782567275115663,
"tokens_seen": 694353920
},
{
"epoch": 0.21,
"learning_rate": 0.0003987963891675025,
"loss": 2.8704,
"theoretical_loss": 3.782530974591956,
"tokens_seen": 694419456
},
{
"epoch": 0.21,
"learning_rate": 0.0003987863590772317,
"loss": 2.845,
"theoretical_loss": 3.782494678453098,
"tokens_seen": 694484992
},
{
"epoch": 0.21,
"learning_rate": 0.0003987763289869609,
"loss": 2.642,
"theoretical_loss": 3.7824583866981465,
"tokens_seen": 694550528
},
{
"epoch": 0.21,
"learning_rate": 0.00039876629889669006,
"loss": 2.9498,
"theoretical_loss": 3.7824220993261575,
"tokens_seen": 694616064
},
{
"epoch": 0.21,
"objective/train/docs_used": 402587,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.756427526473999,
"objective/train/theoretical_loss": 3.782385816336189,
"objective/train/tokens_used": 715141600,
"theoretical_loss": 3.782385816336189,
"tokens_seen": 694681600
},
{
"epoch": 0.21,
"learning_rate": 0.00039875626880641924,
"loss": 2.8557,
"theoretical_loss": 3.782385816336189,
"tokens_seen": 694681600
},
{
"epoch": 0.21,
"learning_rate": 0.00039874623871614847,
"loss": 2.8144,
"theoretical_loss": 3.7823495377272978,
"tokens_seen": 694747136
},
{
"epoch": 0.21,
"learning_rate": 0.0003987362086258776,
"loss": 3.1867,
"theoretical_loss": 3.7823132634985424,
"tokens_seen": 694812672
},
{
"epoch": 0.21,
"learning_rate": 0.00039872617853560683,
"loss": 3.0388,
"theoretical_loss": 3.7822769936489804,
"tokens_seen": 694878208
},
{
"epoch": 0.21,
"learning_rate": 0.000398716148445336,
"loss": 2.9194,
"theoretical_loss": 3.78224072817767,
"tokens_seen": 694943744
},
{
"epoch": 0.21,
"learning_rate": 0.0003987061183550652,
"loss": 3.0826,
"theoretical_loss": 3.7822044670836705,
"tokens_seen": 695009280
},
{
"epoch": 0.21,
"learning_rate": 0.0003986960882647944,
"loss": 2.8542,
"theoretical_loss": 3.782168210366041,
"tokens_seen": 695074816
},
{
"epoch": 0.21,
"learning_rate": 0.00039868605817452356,
"loss": 2.8592,
"theoretical_loss": 3.78213195802384,
"tokens_seen": 695140352
},
{
"epoch": 0.21,
"learning_rate": 0.00039867602808425274,
"loss": 2.7006,
"theoretical_loss": 3.7820957100561277,
"tokens_seen": 695205888
},
{
"epoch": 0.21,
"learning_rate": 0.000398665997993982,
"loss": 2.9181,
"theoretical_loss": 3.782059466461963,
"tokens_seen": 695271424
},
{
"epoch": 0.21,
"learning_rate": 0.0003986559679037111,
"loss": 2.8843,
"theoretical_loss": 3.782023227240407,
"tokens_seen": 695336960
},
{
"epoch": 0.21,
"learning_rate": 0.00039864593781344034,
"loss": 2.7748,
"theoretical_loss": 3.78198699239052,
"tokens_seen": 695402496
},
{
"epoch": 0.21,
"learning_rate": 0.0003986359077231695,
"loss": 2.7837,
"theoretical_loss": 3.781950761911362,
"tokens_seen": 695468032
},
{
"epoch": 0.21,
"learning_rate": 0.0003986258776328987,
"loss": 2.8656,
"theoretical_loss": 3.7819145358019948,
"tokens_seen": 695533568
},
{
"epoch": 0.21,
"learning_rate": 0.00039861584754262794,
"loss": 3.1468,
"theoretical_loss": 3.7818783140614793,
"tokens_seen": 695599104
},
{
"epoch": 0.21,
"learning_rate": 0.00039860581745235706,
"loss": 2.7224,
"theoretical_loss": 3.7818420966888775,
"tokens_seen": 695664640
},
{
"epoch": 0.21,
"learning_rate": 0.0003985957873620863,
"loss": 2.9024,
"theoretical_loss": 3.7818058836832504,
"tokens_seen": 695730176
},
{
"epoch": 0.21,
"learning_rate": 0.0003985857572718155,
"loss": 2.6554,
"theoretical_loss": 3.7817696750436607,
"tokens_seen": 695795712
},
{
"epoch": 0.21,
"learning_rate": 0.00039857572718154466,
"loss": 2.817,
"theoretical_loss": 3.781733470769171,
"tokens_seen": 695861248
},
{
"epoch": 0.21,
"learning_rate": 0.00039856569709127384,
"loss": 2.5786,
"theoretical_loss": 3.7816972708588446,
"tokens_seen": 695926784
},
{
"epoch": 0.21,
"learning_rate": 0.000398555667001003,
"loss": 3.0242,
"theoretical_loss": 3.7816610753117432,
"tokens_seen": 695992320
},
{
"epoch": 0.21,
"learning_rate": 0.0003985456369107322,
"loss": 2.8675,
"theoretical_loss": 3.7816248841269307,
"tokens_seen": 696057856
},
{
"epoch": 0.21,
"learning_rate": 0.00039853560682046144,
"loss": 2.6814,
"theoretical_loss": 3.7815886973034707,
"tokens_seen": 696123392
},
{
"epoch": 0.21,
"learning_rate": 0.00039852557673019057,
"loss": 2.6553,
"theoretical_loss": 3.781552514840427,
"tokens_seen": 696188928
},
{
"epoch": 0.21,
"learning_rate": 0.0003985155466399198,
"loss": 2.8206,
"theoretical_loss": 3.781516336736864,
"tokens_seen": 696254464
},
{
"epoch": 0.21,
"objective/train/docs_used": 403173,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1346781253814697,
"objective/train/theoretical_loss": 3.781480162991847,
"objective/train/tokens_used": 716780000,
"theoretical_loss": 3.781480162991847,
"tokens_seen": 696320000
},
{
"epoch": 0.21,
"learning_rate": 0.00039850551654964893,
"loss": 2.6752,
"theoretical_loss": 3.781480162991847,
"tokens_seen": 696320000
},
{
"epoch": 0.21,
"learning_rate": 0.00039849548645937816,
"loss": 2.8116,
"theoretical_loss": 3.7814439936044386,
"tokens_seen": 696385536
},
{
"epoch": 0.21,
"learning_rate": 0.00039848545636910734,
"loss": 2.8614,
"theoretical_loss": 3.781407828573706,
"tokens_seen": 696451072
},
{
"epoch": 0.21,
"learning_rate": 0.0003984754262788365,
"loss": 2.7697,
"theoretical_loss": 3.7813716678987133,
"tokens_seen": 696516608
},
{
"epoch": 0.21,
"learning_rate": 0.0003984653961885657,
"loss": 2.6746,
"theoretical_loss": 3.781335511578526,
"tokens_seen": 696582144
},
{
"epoch": 0.21,
"learning_rate": 0.0003984553660982949,
"loss": 2.8066,
"theoretical_loss": 3.781299359612211,
"tokens_seen": 696647680
},
{
"epoch": 0.21,
"learning_rate": 0.00039844533600802407,
"loss": 2.6988,
"theoretical_loss": 3.781263211998834,
"tokens_seen": 696713216
},
{
"epoch": 0.21,
"learning_rate": 0.0003984353059177533,
"loss": 2.6711,
"theoretical_loss": 3.7812270687374614,
"tokens_seen": 696778752
},
{
"epoch": 0.21,
"learning_rate": 0.00039842527582748243,
"loss": 2.7738,
"theoretical_loss": 3.7811909298271607,
"tokens_seen": 696844288
},
{
"epoch": 0.21,
"learning_rate": 0.00039841524573721167,
"loss": 2.6851,
"theoretical_loss": 3.7811547952669975,
"tokens_seen": 696909824
},
{
"epoch": 0.21,
"learning_rate": 0.00039840521564694085,
"loss": 2.9013,
"theoretical_loss": 3.78111866505604,
"tokens_seen": 696975360
},
{
"epoch": 0.21,
"learning_rate": 0.00039839518555667003,
"loss": 2.6651,
"theoretical_loss": 3.7810825391933562,
"tokens_seen": 697040896
},
{
"epoch": 0.21,
"learning_rate": 0.0003983851554663992,
"loss": 2.8834,
"theoretical_loss": 3.781046417678014,
"tokens_seen": 697106432
},
{
"epoch": 0.21,
"learning_rate": 0.0003983751253761284,
"loss": 2.7809,
"theoretical_loss": 3.781010300509081,
"tokens_seen": 697171968
},
{
"epoch": 0.21,
"learning_rate": 0.00039836509528585757,
"loss": 2.7983,
"theoretical_loss": 3.7809741876856258,
"tokens_seen": 697237504
},
{
"epoch": 0.21,
"learning_rate": 0.0003983550651955868,
"loss": 2.6574,
"theoretical_loss": 3.780938079206718,
"tokens_seen": 697303040
},
{
"epoch": 0.21,
"learning_rate": 0.00039834503510531593,
"loss": 2.7961,
"theoretical_loss": 3.7809019750714254,
"tokens_seen": 697368576
},
{
"epoch": 0.21,
"learning_rate": 0.00039833500501504517,
"loss": 2.7297,
"theoretical_loss": 3.7808658752788187,
"tokens_seen": 697434112
},
{
"epoch": 0.21,
"learning_rate": 0.0003983249749247743,
"loss": 2.9025,
"theoretical_loss": 3.7808297798279664,
"tokens_seen": 697499648
},
{
"epoch": 0.21,
"learning_rate": 0.00039831494483450353,
"loss": 2.6076,
"theoretical_loss": 3.7807936887179396,
"tokens_seen": 697565184
},
{
"epoch": 0.21,
"learning_rate": 0.0003983049147442327,
"loss": 3.0499,
"theoretical_loss": 3.780757601947807,
"tokens_seen": 697630720
},
{
"epoch": 0.21,
"learning_rate": 0.0003982948846539619,
"loss": 2.6712,
"theoretical_loss": 3.780721519516641,
"tokens_seen": 697696256
},
{
"epoch": 0.21,
"learning_rate": 0.0003982848545636911,
"loss": 2.5923,
"theoretical_loss": 3.7806854414235103,
"tokens_seen": 697761792
},
{
"epoch": 0.21,
"learning_rate": 0.00039827482447342026,
"loss": 2.6223,
"theoretical_loss": 3.780649367667488,
"tokens_seen": 697827328
},
{
"epoch": 0.21,
"learning_rate": 0.00039826479438314944,
"loss": 2.6224,
"theoretical_loss": 3.780613298247644,
"tokens_seen": 697892864
},
{
"epoch": 0.21,
"objective/train/docs_used": 404489,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.381112575531006,
"objective/train/theoretical_loss": 3.7805772331630516,
"objective/train/tokens_used": 718418400,
"theoretical_loss": 3.7805772331630516,
"tokens_seen": 697958400
},
{
"epoch": 0.21,
"learning_rate": 0.00039825476429287867,
"loss": 2.574,
"theoretical_loss": 3.7805772331630516,
"tokens_seen": 697958400
},
{
"epoch": 0.21,
"learning_rate": 0.0003982447342026078,
"loss": 2.6631,
"theoretical_loss": 3.7805411724127804,
"tokens_seen": 698023936
},
{
"epoch": 0.21,
"learning_rate": 0.00039823470411233703,
"loss": 2.6501,
"theoretical_loss": 3.780505115995904,
"tokens_seen": 698089472
},
{
"epoch": 0.21,
"learning_rate": 0.0003982246740220662,
"loss": 2.6789,
"theoretical_loss": 3.7804690639114957,
"tokens_seen": 698155008
},
{
"epoch": 0.21,
"learning_rate": 0.0003982146439317954,
"loss": 2.7016,
"theoretical_loss": 3.780433016158627,
"tokens_seen": 698220544
},
{
"epoch": 0.21,
"learning_rate": 0.0003982046138415246,
"loss": 2.7175,
"theoretical_loss": 3.780396972736371,
"tokens_seen": 698286080
},
{
"epoch": 0.21,
"learning_rate": 0.00039819458375125376,
"loss": 2.9371,
"theoretical_loss": 3.780360933643802,
"tokens_seen": 698351616
},
{
"epoch": 0.21,
"learning_rate": 0.00039818455366098294,
"loss": 2.9238,
"theoretical_loss": 3.780324898879993,
"tokens_seen": 698417152
},
{
"epoch": 0.21,
"learning_rate": 0.0003981745235707122,
"loss": 2.7378,
"theoretical_loss": 3.780288868444018,
"tokens_seen": 698482688
},
{
"epoch": 0.21,
"learning_rate": 0.0003981644934804413,
"loss": 2.5624,
"theoretical_loss": 3.7802528423349515,
"tokens_seen": 698548224
},
{
"epoch": 0.21,
"learning_rate": 0.00039815446339017054,
"loss": 2.5656,
"theoretical_loss": 3.7802168205518676,
"tokens_seen": 698613760
},
{
"epoch": 0.21,
"learning_rate": 0.00039814443329989966,
"loss": 2.7549,
"theoretical_loss": 3.7801808030938417,
"tokens_seen": 698679296
},
{
"epoch": 0.21,
"learning_rate": 0.0003981344032096289,
"loss": 2.8578,
"theoretical_loss": 3.780144789959948,
"tokens_seen": 698744832
},
{
"epoch": 0.21,
"learning_rate": 0.0003981243731193581,
"loss": 2.8182,
"theoretical_loss": 3.780108781149263,
"tokens_seen": 698810368
},
{
"epoch": 0.21,
"learning_rate": 0.00039811434302908726,
"loss": 2.6359,
"theoretical_loss": 3.7800727766608615,
"tokens_seen": 698875904
},
{
"epoch": 0.21,
"learning_rate": 0.00039810431293881644,
"loss": 2.7953,
"theoretical_loss": 3.7800367764938194,
"tokens_seen": 698941440
},
{
"epoch": 0.21,
"learning_rate": 0.0003980942828485457,
"loss": 2.632,
"theoretical_loss": 3.7800007806472133,
"tokens_seen": 699006976
},
{
"epoch": 0.21,
"learning_rate": 0.0003980842527582748,
"loss": 2.8145,
"theoretical_loss": 3.77996478912012,
"tokens_seen": 699072512
},
{
"epoch": 0.21,
"learning_rate": 0.00039807422266800404,
"loss": 2.72,
"theoretical_loss": 3.7799288019116153,
"tokens_seen": 699138048
},
{
"epoch": 0.21,
"learning_rate": 0.00039806419257773317,
"loss": 2.8302,
"theoretical_loss": 3.7798928190207772,
"tokens_seen": 699203584
},
{
"epoch": 0.21,
"learning_rate": 0.0003980541624874624,
"loss": 2.9457,
"theoretical_loss": 3.7798568404466826,
"tokens_seen": 699269120
},
{
"epoch": 0.21,
"learning_rate": 0.0003980441323971916,
"loss": 2.7315,
"theoretical_loss": 3.779820866188409,
"tokens_seen": 699334656
},
{
"epoch": 0.21,
"learning_rate": 0.00039803410230692077,
"loss": 2.529,
"theoretical_loss": 3.779784896245035,
"tokens_seen": 699400192
},
{
"epoch": 0.21,
"learning_rate": 0.00039802407221664995,
"loss": 2.6745,
"theoretical_loss": 3.7797489306156384,
"tokens_seen": 699465728
},
{
"epoch": 0.21,
"learning_rate": 0.00039801404212637913,
"loss": 2.6829,
"theoretical_loss": 3.7797129692992977,
"tokens_seen": 699531264
},
{
"epoch": 0.21,
"objective/train/docs_used": 405172,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.945302963256836,
"objective/train/theoretical_loss": 3.779677012295091,
"objective/train/tokens_used": 720056800,
"theoretical_loss": 3.779677012295091,
"tokens_seen": 699596800
},
{
"epoch": 0.21,
"learning_rate": 0.0003980040120361083,
"loss": 2.6299,
"theoretical_loss": 3.779677012295091,
"tokens_seen": 699596800
},
{
"epoch": 0.21,
"learning_rate": 0.00039799398194583754,
"loss": 2.8623,
"theoretical_loss": 3.7796410596020986,
"tokens_seen": 699662336
},
{
"epoch": 0.21,
"learning_rate": 0.00039798395185556667,
"loss": 2.9632,
"theoretical_loss": 3.779605111219399,
"tokens_seen": 699727872
},
{
"epoch": 0.21,
"learning_rate": 0.0003979739217652959,
"loss": 2.9065,
"theoretical_loss": 3.7795691671460725,
"tokens_seen": 699793408
},
{
"epoch": 0.21,
"learning_rate": 0.00039796389167502503,
"loss": 2.8761,
"theoretical_loss": 3.779533227381198,
"tokens_seen": 699858944
},
{
"epoch": 0.21,
"learning_rate": 0.00039795386158475427,
"loss": 2.7551,
"theoretical_loss": 3.7794972919238563,
"tokens_seen": 699924480
},
{
"epoch": 0.21,
"learning_rate": 0.00039794383149448345,
"loss": 2.6173,
"theoretical_loss": 3.779461360773128,
"tokens_seen": 699990016
},
{
"epoch": 0.21,
"learning_rate": 0.00039793380140421263,
"loss": 2.5548,
"theoretical_loss": 3.7794254339280937,
"tokens_seen": 700055552
},
{
"epoch": 0.21,
"learning_rate": 0.0003979237713139418,
"loss": 2.8365,
"theoretical_loss": 3.7793895113878344,
"tokens_seen": 700121088
},
{
"epoch": 0.21,
"learning_rate": 0.00039791374122367105,
"loss": 2.8706,
"theoretical_loss": 3.7793535931514315,
"tokens_seen": 700186624
},
{
"epoch": 0.21,
"learning_rate": 0.0003979037111334002,
"loss": 2.7541,
"theoretical_loss": 3.779317679217966,
"tokens_seen": 700252160
},
{
"epoch": 0.21,
"learning_rate": 0.0003978936810431294,
"loss": 2.7556,
"theoretical_loss": 3.779281769586521,
"tokens_seen": 700317696
},
{
"epoch": 0.21,
"learning_rate": 0.0003978836509528586,
"loss": 2.9774,
"theoretical_loss": 3.7792458642561777,
"tokens_seen": 700383232
},
{
"epoch": 0.21,
"learning_rate": 0.00039787362086258777,
"loss": 2.9304,
"theoretical_loss": 3.7792099632260188,
"tokens_seen": 700448768
},
{
"epoch": 0.21,
"learning_rate": 0.000397863590772317,
"loss": 2.7946,
"theoretical_loss": 3.779174066495127,
"tokens_seen": 700514304
},
{
"epoch": 0.21,
"learning_rate": 0.00039785356068204613,
"loss": 2.8988,
"theoretical_loss": 3.7791381740625853,
"tokens_seen": 700579840
},
{
"epoch": 0.21,
"learning_rate": 0.00039784353059177537,
"loss": 2.784,
"theoretical_loss": 3.779102285927477,
"tokens_seen": 700645376
},
{
"epoch": 0.21,
"learning_rate": 0.0003978335005015045,
"loss": 2.811,
"theoretical_loss": 3.779066402088886,
"tokens_seen": 700710912
},
{
"epoch": 0.21,
"learning_rate": 0.00039782347041123373,
"loss": 2.92,
"theoretical_loss": 3.779030522545896,
"tokens_seen": 700776448
},
{
"epoch": 0.21,
"learning_rate": 0.0003978134403209629,
"loss": 2.8849,
"theoretical_loss": 3.7789946472975906,
"tokens_seen": 700841984
},
{
"epoch": 0.21,
"learning_rate": 0.0003978034102306921,
"loss": 3.0046,
"theoretical_loss": 3.7789587763430545,
"tokens_seen": 700907520
},
{
"epoch": 0.21,
"learning_rate": 0.0003977933801404213,
"loss": 2.8613,
"theoretical_loss": 3.7789229096813726,
"tokens_seen": 700973056
},
{
"epoch": 0.21,
"learning_rate": 0.00039778335005015046,
"loss": 2.7804,
"theoretical_loss": 3.77888704731163,
"tokens_seen": 701038592
},
{
"epoch": 0.21,
"learning_rate": 0.00039777331995987964,
"loss": 2.7396,
"theoretical_loss": 3.7788511892329115,
"tokens_seen": 701104128
},
{
"epoch": 0.21,
"learning_rate": 0.00039776328986960887,
"loss": 3.0226,
"theoretical_loss": 3.7788153354443024,
"tokens_seen": 701169664
},
{
"epoch": 0.21,
"objective/train/docs_used": 406046,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9453184604644775,
"objective/train/theoretical_loss": 3.7787794859448898,
"objective/train/tokens_used": 721695200,
"theoretical_loss": 3.7787794859448898,
"tokens_seen": 701235200
},
{
"epoch": 0.21,
"learning_rate": 0.000397753259779338,
"loss": 2.8136,
"theoretical_loss": 3.7787794859448898,
"tokens_seen": 701235200
},
{
"epoch": 0.21,
"learning_rate": 0.00039774322968906723,
"loss": 2.6325,
"theoretical_loss": 3.7787436407337585,
"tokens_seen": 701300736
},
{
"epoch": 0.21,
"learning_rate": 0.0003977331995987964,
"loss": 2.7707,
"theoretical_loss": 3.7787077998099954,
"tokens_seen": 701366272
},
{
"epoch": 0.21,
"learning_rate": 0.0003977231695085256,
"loss": 2.8152,
"theoretical_loss": 3.778671963172687,
"tokens_seen": 701431808
},
{
"epoch": 0.21,
"learning_rate": 0.0003977131394182548,
"loss": 3.0481,
"theoretical_loss": 3.77863613082092,
"tokens_seen": 701497344
},
{
"epoch": 0.21,
"learning_rate": 0.00039770310932798396,
"loss": 2.8833,
"theoretical_loss": 3.778600302753783,
"tokens_seen": 701562880
},
{
"epoch": 0.21,
"learning_rate": 0.00039769307923771314,
"loss": 2.7699,
"theoretical_loss": 3.7785644789703614,
"tokens_seen": 701628416
},
{
"epoch": 0.21,
"learning_rate": 0.0003976830491474424,
"loss": 3.107,
"theoretical_loss": 3.7785286594697443,
"tokens_seen": 701693952
},
{
"epoch": 0.21,
"learning_rate": 0.0003976730190571715,
"loss": 2.8004,
"theoretical_loss": 3.7784928442510193,
"tokens_seen": 701759488
},
{
"epoch": 0.21,
"learning_rate": 0.00039766298896690074,
"loss": 2.6622,
"theoretical_loss": 3.778457033313275,
"tokens_seen": 701825024
},
{
"epoch": 0.21,
"learning_rate": 0.00039765295887662987,
"loss": 2.9171,
"theoretical_loss": 3.7784212266556008,
"tokens_seen": 701890560
},
{
"epoch": 0.21,
"learning_rate": 0.0003976429287863591,
"loss": 2.7891,
"theoretical_loss": 3.778385424277084,
"tokens_seen": 701956096
},
{
"epoch": 0.21,
"learning_rate": 0.0003976328986960883,
"loss": 2.709,
"theoretical_loss": 3.778349626176814,
"tokens_seen": 702021632
},
{
"epoch": 0.21,
"learning_rate": 0.00039762286860581746,
"loss": 2.9798,
"theoretical_loss": 3.7783138323538807,
"tokens_seen": 702087168
},
{
"epoch": 0.21,
"learning_rate": 0.00039761283851554664,
"loss": 2.5199,
"theoretical_loss": 3.778278042807375,
"tokens_seen": 702152704
},
{
"epoch": 0.21,
"learning_rate": 0.0003976028084252759,
"loss": 2.8978,
"theoretical_loss": 3.7782422575363848,
"tokens_seen": 702218240
},
{
"epoch": 0.21,
"learning_rate": 0.000397592778335005,
"loss": 2.9115,
"theoretical_loss": 3.778206476540001,
"tokens_seen": 702283776
},
{
"epoch": 0.21,
"learning_rate": 0.00039758274824473424,
"loss": 2.9291,
"theoretical_loss": 3.7781706998173155,
"tokens_seen": 702349312
},
{
"epoch": 0.21,
"learning_rate": 0.00039757271815446337,
"loss": 2.8982,
"theoretical_loss": 3.778134927367417,
"tokens_seen": 702414848
},
{
"epoch": 0.21,
"learning_rate": 0.0003975626880641926,
"loss": 2.6145,
"theoretical_loss": 3.778099159189399,
"tokens_seen": 702480384
},
{
"epoch": 0.21,
"learning_rate": 0.0003975526579739218,
"loss": 2.7798,
"theoretical_loss": 3.7780633952823512,
"tokens_seen": 702545920
},
{
"epoch": 0.21,
"learning_rate": 0.00039754262788365097,
"loss": 2.666,
"theoretical_loss": 3.7780276356453655,
"tokens_seen": 702611456
},
{
"epoch": 0.21,
"learning_rate": 0.00039753259779338015,
"loss": 2.7912,
"theoretical_loss": 3.7779918802775345,
"tokens_seen": 702676992
},
{
"epoch": 0.21,
"learning_rate": 0.00039752256770310933,
"loss": 2.6411,
"theoretical_loss": 3.7779561291779498,
"tokens_seen": 702742528
},
{
"epoch": 0.21,
"learning_rate": 0.0003975125376128385,
"loss": 2.9421,
"theoretical_loss": 3.777920382345704,
"tokens_seen": 702808064
},
{
"epoch": 0.21,
"objective/train/docs_used": 406800,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.962623357772827,
"objective/train/theoretical_loss": 3.7778846397798898,
"objective/train/tokens_used": 723333600,
"theoretical_loss": 3.7778846397798898,
"tokens_seen": 702873600
},
{
"epoch": 0.21,
"learning_rate": 0.00039750250752256774,
"loss": 2.9134,
"theoretical_loss": 3.7778846397798898,
"tokens_seen": 702873600
},
{
"epoch": 0.21,
"learning_rate": 0.00039749247743229687,
"loss": 2.9659,
"theoretical_loss": 3.777848901479601,
"tokens_seen": 702939136
},
{
"epoch": 0.21,
"learning_rate": 0.0003974824473420261,
"loss": 2.7156,
"theoretical_loss": 3.77781316744393,
"tokens_seen": 703004672
},
{
"epoch": 0.21,
"learning_rate": 0.00039747241725175523,
"loss": 2.7695,
"theoretical_loss": 3.777777437671971,
"tokens_seen": 703070208
},
{
"epoch": 0.21,
"learning_rate": 0.00039746238716148447,
"loss": 2.8003,
"theoretical_loss": 3.777741712162818,
"tokens_seen": 703135744
},
{
"epoch": 0.21,
"learning_rate": 0.00039745235707121365,
"loss": 2.7796,
"theoretical_loss": 3.777705990915565,
"tokens_seen": 703201280
},
{
"epoch": 0.21,
"learning_rate": 0.00039744232698094283,
"loss": 2.7109,
"theoretical_loss": 3.7776702739293055,
"tokens_seen": 703266816
},
{
"epoch": 0.21,
"learning_rate": 0.000397432296890672,
"loss": 2.9559,
"theoretical_loss": 3.777634561203136,
"tokens_seen": 703332352
},
{
"epoch": 0.21,
"learning_rate": 0.00039742226680040125,
"loss": 2.6793,
"theoretical_loss": 3.7775988527361504,
"tokens_seen": 703397888
},
{
"epoch": 0.21,
"learning_rate": 0.0003974122367101304,
"loss": 2.9123,
"theoretical_loss": 3.7775631485274435,
"tokens_seen": 703463424
},
{
"epoch": 0.21,
"learning_rate": 0.0003974022066198596,
"loss": 2.9313,
"theoretical_loss": 3.7775274485761123,
"tokens_seen": 703528960
},
{
"epoch": 0.21,
"learning_rate": 0.00039739217652958874,
"loss": 2.9136,
"theoretical_loss": 3.7774917528812515,
"tokens_seen": 703594496
},
{
"epoch": 0.21,
"learning_rate": 0.00039738214643931797,
"loss": 2.9252,
"theoretical_loss": 3.7774560614419577,
"tokens_seen": 703660032
},
{
"epoch": 0.21,
"learning_rate": 0.00039737211634904715,
"loss": 2.8489,
"theoretical_loss": 3.777420374257327,
"tokens_seen": 703725568
},
{
"epoch": 0.21,
"learning_rate": 0.00039736208625877633,
"loss": 2.7307,
"theoretical_loss": 3.7773846913264566,
"tokens_seen": 703791104
},
{
"epoch": 0.21,
"learning_rate": 0.0003973520561685055,
"loss": 2.9462,
"theoretical_loss": 3.777349012648443,
"tokens_seen": 703856640
},
{
"epoch": 0.21,
"learning_rate": 0.0003973420260782347,
"loss": 2.8402,
"theoretical_loss": 3.777313338222383,
"tokens_seen": 703922176
},
{
"epoch": 0.21,
"learning_rate": 0.0003973319959879639,
"loss": 2.8882,
"theoretical_loss": 3.7772776680473745,
"tokens_seen": 703987712
},
{
"epoch": 0.21,
"learning_rate": 0.0003973219658976931,
"loss": 2.6995,
"theoretical_loss": 3.7772420021225157,
"tokens_seen": 704053248
},
{
"epoch": 0.21,
"learning_rate": 0.00039731193580742224,
"loss": 2.8313,
"theoretical_loss": 3.777206340446904,
"tokens_seen": 704118784
},
{
"epoch": 0.21,
"learning_rate": 0.0003973019057171515,
"loss": 2.7769,
"theoretical_loss": 3.7771706830196385,
"tokens_seen": 704184320
},
{
"epoch": 0.21,
"learning_rate": 0.0003972918756268806,
"loss": 2.7716,
"theoretical_loss": 3.777135029839817,
"tokens_seen": 704249856
},
{
"epoch": 0.21,
"learning_rate": 0.00039728184553660984,
"loss": 2.9452,
"theoretical_loss": 3.777099380906538,
"tokens_seen": 704315392
},
{
"epoch": 0.21,
"learning_rate": 0.000397271815446339,
"loss": 2.7828,
"theoretical_loss": 3.777063736218902,
"tokens_seen": 704380928
},
{
"epoch": 0.21,
"learning_rate": 0.0003972617853560682,
"loss": 2.5791,
"theoretical_loss": 3.7770280957760076,
"tokens_seen": 704446464
},
{
"epoch": 0.21,
"objective/train/docs_used": 408057,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.633457899093628,
"objective/train/theoretical_loss": 3.7769924595769546,
"objective/train/tokens_used": 724972000,
"theoretical_loss": 3.7769924595769546,
"tokens_seen": 704512000
},
{
"epoch": 0.21,
"learning_rate": 0.0003972517552657974,
"loss": 2.5964,
"theoretical_loss": 3.7769924595769546,
"tokens_seen": 704512000
},
{
"epoch": 0.21,
"learning_rate": 0.0003972417251755266,
"loss": 2.6763,
"theoretical_loss": 3.776956827620843,
"tokens_seen": 704577536
},
{
"epoch": 0.21,
"learning_rate": 0.00039723169508525574,
"loss": 2.8623,
"theoretical_loss": 3.7769211999067727,
"tokens_seen": 704643072
},
{
"epoch": 0.21,
"learning_rate": 0.000397221664994985,
"loss": 2.8363,
"theoretical_loss": 3.7768855764338447,
"tokens_seen": 704708608
},
{
"epoch": 0.21,
"learning_rate": 0.0003972116349047141,
"loss": 2.98,
"theoretical_loss": 3.776849957201159,
"tokens_seen": 704774144
},
{
"epoch": 0.21,
"learning_rate": 0.00039720160481444334,
"loss": 2.9744,
"theoretical_loss": 3.776814342207818,
"tokens_seen": 704839680
},
{
"epoch": 0.21,
"learning_rate": 0.0003971915747241725,
"loss": 2.5489,
"theoretical_loss": 3.7767787314529215,
"tokens_seen": 704905216
},
{
"epoch": 0.21,
"learning_rate": 0.0003971815446339017,
"loss": 2.8402,
"theoretical_loss": 3.7767431249355727,
"tokens_seen": 704970752
},
{
"epoch": 0.21,
"learning_rate": 0.0003971715145436309,
"loss": 2.7301,
"theoretical_loss": 3.7767075226548723,
"tokens_seen": 705036288
},
{
"epoch": 0.21,
"learning_rate": 0.00039716148445336007,
"loss": 2.9275,
"theoretical_loss": 3.7766719246099227,
"tokens_seen": 705101824
},
{
"epoch": 0.21,
"learning_rate": 0.00039715145436308925,
"loss": 2.8887,
"theoretical_loss": 3.7766363307998265,
"tokens_seen": 705167360
},
{
"epoch": 0.21,
"learning_rate": 0.0003971414242728185,
"loss": 2.7245,
"theoretical_loss": 3.7766007412236866,
"tokens_seen": 705232896
},
{
"epoch": 0.21,
"learning_rate": 0.00039713139418254766,
"loss": 3.1424,
"theoretical_loss": 3.7765651558806055,
"tokens_seen": 705298432
},
{
"epoch": 0.21,
"learning_rate": 0.00039712136409227684,
"loss": 2.7296,
"theoretical_loss": 3.776529574769687,
"tokens_seen": 705363968
},
{
"epoch": 0.21,
"learning_rate": 0.0003971113340020061,
"loss": 2.8791,
"theoretical_loss": 3.776493997890034,
"tokens_seen": 705429504
},
{
"epoch": 0.21,
"learning_rate": 0.0003971013039117352,
"loss": 2.6762,
"theoretical_loss": 3.776458425240751,
"tokens_seen": 705495040
},
{
"epoch": 0.21,
"learning_rate": 0.00039709127382146444,
"loss": 2.9455,
"theoretical_loss": 3.776422856820941,
"tokens_seen": 705560576
},
{
"epoch": 0.21,
"learning_rate": 0.00039708124373119357,
"loss": 2.9965,
"theoretical_loss": 3.7763872926297095,
"tokens_seen": 705626112
},
{
"epoch": 0.21,
"learning_rate": 0.0003970712136409228,
"loss": 2.9135,
"theoretical_loss": 3.7763517326661606,
"tokens_seen": 705691648
},
{
"epoch": 0.21,
"learning_rate": 0.000397061183550652,
"loss": 2.5661,
"theoretical_loss": 3.7763161769293996,
"tokens_seen": 705757184
},
{
"epoch": 0.21,
"learning_rate": 0.00039705115346038117,
"loss": 2.6929,
"theoretical_loss": 3.776280625418531,
"tokens_seen": 705822720
},
{
"epoch": 0.21,
"learning_rate": 0.00039704112337011035,
"loss": 2.9297,
"theoretical_loss": 3.77624507813266,
"tokens_seen": 705888256
},
{
"epoch": 0.21,
"learning_rate": 0.00039703109327983953,
"loss": 2.7122,
"theoretical_loss": 3.7762095350708935,
"tokens_seen": 705953792
},
{
"epoch": 0.21,
"learning_rate": 0.0003970210631895687,
"loss": 2.8914,
"theoretical_loss": 3.7761739962323366,
"tokens_seen": 706019328
},
{
"epoch": 0.21,
"learning_rate": 0.00039701103309929794,
"loss": 2.7338,
"theoretical_loss": 3.776138461616096,
"tokens_seen": 706084864
},
{
"epoch": 0.21,
"objective/train/docs_used": 408602,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8542611598968506,
"objective/train/theoretical_loss": 3.7761029312212777,
"objective/train/tokens_used": 726610400,
"theoretical_loss": 3.7761029312212777,
"tokens_seen": 706150400
},
{
"epoch": 0.21,
"learning_rate": 0.00039700100300902707,
"loss": 2.6113,
"theoretical_loss": 3.7761029312212777,
"tokens_seen": 706150400
},
{
"epoch": 0.21,
"learning_rate": 0.0003969909729187563,
"loss": 2.7997,
"theoretical_loss": 3.776067405046989,
"tokens_seen": 706215936
},
{
"epoch": 0.21,
"learning_rate": 0.00039698094282848543,
"loss": 2.7896,
"theoretical_loss": 3.7760318830923367,
"tokens_seen": 706281472
},
{
"epoch": 0.21,
"learning_rate": 0.00039697091273821467,
"loss": 3.0169,
"theoretical_loss": 3.775996365356428,
"tokens_seen": 706347008
},
{
"epoch": 0.21,
"learning_rate": 0.00039696088264794385,
"loss": 2.9092,
"theoretical_loss": 3.775960851838371,
"tokens_seen": 706412544
},
{
"epoch": 0.21,
"learning_rate": 0.00039695085255767303,
"loss": 2.8064,
"theoretical_loss": 3.7759253425372736,
"tokens_seen": 706478080
},
{
"epoch": 0.21,
"learning_rate": 0.0003969408224674022,
"loss": 2.785,
"theoretical_loss": 3.775889837452243,
"tokens_seen": 706543616
},
{
"epoch": 0.21,
"learning_rate": 0.00039693079237713145,
"loss": 2.9194,
"theoretical_loss": 3.7758543365823884,
"tokens_seen": 706609152
},
{
"epoch": 0.21,
"learning_rate": 0.0003969207622868606,
"loss": 2.8403,
"theoretical_loss": 3.775818839926819,
"tokens_seen": 706674688
},
{
"epoch": 0.21,
"learning_rate": 0.0003969107321965898,
"loss": 2.9938,
"theoretical_loss": 3.7757833474846425,
"tokens_seen": 706740224
},
{
"epoch": 0.21,
"learning_rate": 0.00039690070210631894,
"loss": 2.8191,
"theoretical_loss": 3.775747859254969,
"tokens_seen": 706805760
},
{
"epoch": 0.21,
"learning_rate": 0.00039689067201604817,
"loss": 2.6823,
"theoretical_loss": 3.7757123752369077,
"tokens_seen": 706871296
},
{
"epoch": 0.21,
"learning_rate": 0.00039688064192577735,
"loss": 2.7617,
"theoretical_loss": 3.775676895429569,
"tokens_seen": 706936832
},
{
"epoch": 0.21,
"learning_rate": 0.00039687061183550653,
"loss": 2.8066,
"theoretical_loss": 3.775641419832062,
"tokens_seen": 707002368
},
{
"epoch": 0.21,
"learning_rate": 0.0003968605817452357,
"loss": 3.0952,
"theoretical_loss": 3.775605948443497,
"tokens_seen": 707067904
},
{
"epoch": 0.21,
"learning_rate": 0.0003968505516549649,
"loss": 2.7829,
"theoretical_loss": 3.7755704812629856,
"tokens_seen": 707133440
},
{
"epoch": 0.21,
"learning_rate": 0.0003968405215646941,
"loss": 2.9364,
"theoretical_loss": 3.775535018289638,
"tokens_seen": 707198976
},
{
"epoch": 0.21,
"learning_rate": 0.0003968304914744233,
"loss": 2.8138,
"theoretical_loss": 3.7754995595225656,
"tokens_seen": 707264512
},
{
"epoch": 0.21,
"learning_rate": 0.00039682046138415244,
"loss": 2.8969,
"theoretical_loss": 3.7754641049608795,
"tokens_seen": 707330048
},
{
"epoch": 0.21,
"learning_rate": 0.0003968104312938817,
"loss": 2.8343,
"theoretical_loss": 3.775428654603691,
"tokens_seen": 707395584
},
{
"epoch": 0.21,
"learning_rate": 0.0003968004012036108,
"loss": 2.9969,
"theoretical_loss": 3.7753932084501134,
"tokens_seen": 707461120
},
{
"epoch": 0.21,
"learning_rate": 0.00039679037111334004,
"loss": 2.6357,
"theoretical_loss": 3.7753577664992575,
"tokens_seen": 707526656
},
{
"epoch": 0.21,
"learning_rate": 0.0003967803410230692,
"loss": 2.7252,
"theoretical_loss": 3.775322328750237,
"tokens_seen": 707592192
},
{
"epoch": 0.21,
"learning_rate": 0.0003967703109327984,
"loss": 2.9087,
"theoretical_loss": 3.7752868952021634,
"tokens_seen": 707657728
},
{
"epoch": 0.21,
"learning_rate": 0.0003967602808425276,
"loss": 2.911,
"theoretical_loss": 3.7752514658541503,
"tokens_seen": 707723264
},
{
"epoch": 0.21,
"objective/train/docs_used": 409727,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.213099956512451,
"objective/train/theoretical_loss": 3.7752160407053115,
"objective/train/tokens_used": 728248800,
"theoretical_loss": 3.7752160407053115,
"tokens_seen": 707788800
},
{
"epoch": 0.21,
"learning_rate": 0.0003967502507522568,
"loss": 2.6394,
"theoretical_loss": 3.7752160407053115,
"tokens_seen": 707788800
},
{
"epoch": 0.21,
"learning_rate": 0.00039674022066198594,
"loss": 3.1145,
"theoretical_loss": 3.7751806197547593,
"tokens_seen": 707854336
},
{
"epoch": 0.21,
"learning_rate": 0.0003967301905717152,
"loss": 2.9301,
"theoretical_loss": 3.775145203001609,
"tokens_seen": 707919872
},
{
"epoch": 0.21,
"learning_rate": 0.0003967201604814443,
"loss": 2.926,
"theoretical_loss": 3.775109790444974,
"tokens_seen": 707985408
},
{
"epoch": 0.21,
"learning_rate": 0.00039671013039117354,
"loss": 2.9302,
"theoretical_loss": 3.775074382083968,
"tokens_seen": 708050944
},
{
"epoch": 0.21,
"learning_rate": 0.0003967001003009027,
"loss": 2.8918,
"theoretical_loss": 3.775038977917707,
"tokens_seen": 708116480
},
{
"epoch": 0.21,
"learning_rate": 0.0003966900702106319,
"loss": 2.9403,
"theoretical_loss": 3.775003577945305,
"tokens_seen": 708182016
},
{
"epoch": 0.21,
"learning_rate": 0.0003966800401203611,
"loss": 2.8373,
"theoretical_loss": 3.7749681821658774,
"tokens_seen": 708247552
},
{
"epoch": 0.21,
"learning_rate": 0.00039667001003009027,
"loss": 2.8604,
"theoretical_loss": 3.7749327905785393,
"tokens_seen": 708313088
},
{
"epoch": 0.21,
"learning_rate": 0.00039665997993981945,
"loss": 3.1589,
"theoretical_loss": 3.774897403182407,
"tokens_seen": 708378624
},
{
"epoch": 0.21,
"learning_rate": 0.0003966499498495487,
"loss": 2.9429,
"theoretical_loss": 3.774862019976596,
"tokens_seen": 708444160
},
{
"epoch": 0.21,
"learning_rate": 0.0003966399197592778,
"loss": 2.7548,
"theoretical_loss": 3.7748266409602236,
"tokens_seen": 708509696
},
{
"epoch": 0.21,
"learning_rate": 0.00039662988966900704,
"loss": 2.9207,
"theoretical_loss": 3.7747912661324046,
"tokens_seen": 708575232
},
{
"epoch": 0.21,
"learning_rate": 0.00039661985957873617,
"loss": 2.8398,
"theoretical_loss": 3.774755895492257,
"tokens_seen": 708640768
},
{
"epoch": 0.21,
"learning_rate": 0.0003966098294884654,
"loss": 2.7411,
"theoretical_loss": 3.7747205290388974,
"tokens_seen": 708706304
},
{
"epoch": 0.21,
"learning_rate": 0.0003965997993981946,
"loss": 3.0009,
"theoretical_loss": 3.7746851667714436,
"tokens_seen": 708771840
},
{
"epoch": 0.21,
"learning_rate": 0.00039658976930792377,
"loss": 3.0983,
"theoretical_loss": 3.774649808689013,
"tokens_seen": 708837376
},
{
"epoch": 0.21,
"learning_rate": 0.00039657973921765295,
"loss": 2.7061,
"theoretical_loss": 3.774614454790722,
"tokens_seen": 708902912
},
{
"epoch": 0.21,
"learning_rate": 0.0003965697091273822,
"loss": 2.7755,
"theoretical_loss": 3.7745791050756914,
"tokens_seen": 708968448
},
{
"epoch": 0.21,
"learning_rate": 0.0003965596790371113,
"loss": 2.8913,
"theoretical_loss": 3.774543759543037,
"tokens_seen": 709033984
},
{
"epoch": 0.21,
"learning_rate": 0.00039654964894684055,
"loss": 2.7875,
"theoretical_loss": 3.7745084181918793,
"tokens_seen": 709099520
},
{
"epoch": 0.21,
"learning_rate": 0.0003965396188565697,
"loss": 2.5608,
"theoretical_loss": 3.7744730810213367,
"tokens_seen": 709165056
},
{
"epoch": 0.21,
"learning_rate": 0.0003965295887662989,
"loss": 2.765,
"theoretical_loss": 3.774437748030528,
"tokens_seen": 709230592
},
{
"epoch": 0.21,
"learning_rate": 0.0003965195586760281,
"loss": 2.6853,
"theoretical_loss": 3.774402419218573,
"tokens_seen": 709296128
},
{
"epoch": 0.21,
"learning_rate": 0.00039650952858575727,
"loss": 3.0681,
"theoretical_loss": 3.7743670945845915,
"tokens_seen": 709361664
},
{
"epoch": 0.21,
"objective/train/docs_used": 410190,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8154327869415283,
"objective/train/theoretical_loss": 3.774331774127703,
"objective/train/tokens_used": 729887200,
"theoretical_loss": 3.774331774127703,
"tokens_seen": 709427200
},
{
"epoch": 0.21,
"learning_rate": 0.00039649949849548645,
"loss": 2.8535,
"theoretical_loss": 3.774331774127703,
"tokens_seen": 709427200
},
{
"epoch": 0.21,
"learning_rate": 0.00039648946840521563,
"loss": 2.9096,
"theoretical_loss": 3.7742964578470284,
"tokens_seen": 709492736
},
{
"epoch": 0.22,
"learning_rate": 0.0003964794383149448,
"loss": 3.0042,
"theoretical_loss": 3.7742611457416877,
"tokens_seen": 709558272
},
{
"epoch": 0.22,
"learning_rate": 0.00039646940822467405,
"loss": 3.0254,
"theoretical_loss": 3.7742258378108016,
"tokens_seen": 709623808
},
{
"epoch": 0.22,
"learning_rate": 0.0003964593781344032,
"loss": 2.6893,
"theoretical_loss": 3.774190534053492,
"tokens_seen": 709689344
},
{
"epoch": 0.22,
"learning_rate": 0.0003964493480441324,
"loss": 2.8084,
"theoretical_loss": 3.7741552344688793,
"tokens_seen": 709754880
},
{
"epoch": 0.22,
"learning_rate": 0.0003964393179538616,
"loss": 2.8832,
"theoretical_loss": 3.774119939056086,
"tokens_seen": 709820416
},
{
"epoch": 0.22,
"learning_rate": 0.0003964292878635908,
"loss": 3.1153,
"theoretical_loss": 3.7740846478142327,
"tokens_seen": 709885952
},
{
"epoch": 0.22,
"learning_rate": 0.00039641925777331996,
"loss": 2.8624,
"theoretical_loss": 3.774049360742443,
"tokens_seen": 709951488
},
{
"epoch": 0.22,
"learning_rate": 0.00039640922768304914,
"loss": 2.8042,
"theoretical_loss": 3.774014077839838,
"tokens_seen": 710017024
},
{
"epoch": 0.22,
"learning_rate": 0.0003963991975927783,
"loss": 3.1453,
"theoretical_loss": 3.7739787991055413,
"tokens_seen": 710082560
},
{
"epoch": 0.22,
"learning_rate": 0.00039638916750250755,
"loss": 3.0274,
"theoretical_loss": 3.773943524538675,
"tokens_seen": 710148096
},
{
"epoch": 0.22,
"learning_rate": 0.00039637913741223673,
"loss": 2.8395,
"theoretical_loss": 3.773908254138363,
"tokens_seen": 710213632
},
{
"epoch": 0.22,
"learning_rate": 0.0003963691073219659,
"loss": 3.0454,
"theoretical_loss": 3.7738729879037285,
"tokens_seen": 710279168
},
{
"epoch": 0.22,
"learning_rate": 0.0003963590772316951,
"loss": 3.0626,
"theoretical_loss": 3.7738377258338947,
"tokens_seen": 710344704
},
{
"epoch": 0.22,
"learning_rate": 0.0003963490471414243,
"loss": 2.6557,
"theoretical_loss": 3.773802467927986,
"tokens_seen": 710410240
},
{
"epoch": 0.22,
"learning_rate": 0.0003963390170511535,
"loss": 2.8816,
"theoretical_loss": 3.7737672141851273,
"tokens_seen": 710475776
},
{
"epoch": 0.22,
"learning_rate": 0.00039632898696088264,
"loss": 2.9243,
"theoretical_loss": 3.7737319646044423,
"tokens_seen": 710541312
},
{
"epoch": 0.22,
"learning_rate": 0.0003963189568706119,
"loss": 2.8569,
"theoretical_loss": 3.7736967191850557,
"tokens_seen": 710606848
},
{
"epoch": 0.22,
"learning_rate": 0.000396308926780341,
"loss": 2.7212,
"theoretical_loss": 3.773661477926093,
"tokens_seen": 710672384
},
{
"epoch": 0.22,
"learning_rate": 0.00039629889669007024,
"loss": 2.6957,
"theoretical_loss": 3.773626240826679,
"tokens_seen": 710737920
},
{
"epoch": 0.22,
"learning_rate": 0.0003962888665997994,
"loss": 2.8251,
"theoretical_loss": 3.7735910078859396,
"tokens_seen": 710803456
},
{
"epoch": 0.22,
"learning_rate": 0.0003962788365095286,
"loss": 2.9622,
"theoretical_loss": 3.7735557791030008,
"tokens_seen": 710868992
},
{
"epoch": 0.22,
"learning_rate": 0.0003962688064192578,
"loss": 3.0848,
"theoretical_loss": 3.773520554476989,
"tokens_seen": 710934528
},
{
"epoch": 0.22,
"learning_rate": 0.000396258776328987,
"loss": 2.9885,
"theoretical_loss": 3.7734853340070296,
"tokens_seen": 711000064
},
{
"epoch": 0.22,
"objective/train/docs_used": 411414,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0206305980682373,
"objective/train/theoretical_loss": 3.7734501176922493,
"objective/train/tokens_used": 731525600,
"theoretical_loss": 3.7734501176922493,
"tokens_seen": 711065600
},
{
"epoch": 0.22,
"learning_rate": 0.00039624874623871614,
"loss": 2.9379,
"theoretical_loss": 3.7734501176922493,
"tokens_seen": 711065600
},
{
"epoch": 0.22,
"learning_rate": 0.0003962387161484454,
"loss": 2.8041,
"theoretical_loss": 3.7734149055317756,
"tokens_seen": 711131136
},
{
"epoch": 0.22,
"learning_rate": 0.0003962286860581745,
"loss": 2.8005,
"theoretical_loss": 3.773379697524736,
"tokens_seen": 711196672
},
{
"epoch": 0.22,
"learning_rate": 0.00039621865596790374,
"loss": 3.0014,
"theoretical_loss": 3.773344493670257,
"tokens_seen": 711262208
},
{
"epoch": 0.22,
"learning_rate": 0.0003962086258776329,
"loss": 2.9836,
"theoretical_loss": 3.7733092939674666,
"tokens_seen": 711327744
},
{
"epoch": 0.22,
"learning_rate": 0.0003961985957873621,
"loss": 2.7207,
"theoretical_loss": 3.7732740984154924,
"tokens_seen": 711393280
},
{
"epoch": 0.22,
"learning_rate": 0.0003961885656970913,
"loss": 2.8312,
"theoretical_loss": 3.7732389070134635,
"tokens_seen": 711458816
},
{
"epoch": 0.22,
"learning_rate": 0.00039617853560682047,
"loss": 2.6389,
"theoretical_loss": 3.7732037197605077,
"tokens_seen": 711524352
},
{
"epoch": 0.22,
"learning_rate": 0.00039616850551654965,
"loss": 3.0179,
"theoretical_loss": 3.7731685366557537,
"tokens_seen": 711589888
},
{
"epoch": 0.22,
"learning_rate": 0.0003961584754262789,
"loss": 2.6945,
"theoretical_loss": 3.773133357698331,
"tokens_seen": 711655424
},
{
"epoch": 0.22,
"learning_rate": 0.000396148445336008,
"loss": 2.7646,
"theoretical_loss": 3.7730981828873684,
"tokens_seen": 711720960
},
{
"epoch": 0.22,
"learning_rate": 0.00039613841524573724,
"loss": 2.9562,
"theoretical_loss": 3.7730630122219955,
"tokens_seen": 711786496
},
{
"epoch": 0.22,
"learning_rate": 0.00039612838515546637,
"loss": 2.9768,
"theoretical_loss": 3.773027845701342,
"tokens_seen": 711852032
},
{
"epoch": 0.22,
"learning_rate": 0.0003961183550651956,
"loss": 2.8906,
"theoretical_loss": 3.7729926833245386,
"tokens_seen": 711917568
},
{
"epoch": 0.22,
"learning_rate": 0.0003961083249749248,
"loss": 3.0006,
"theoretical_loss": 3.772957525090715,
"tokens_seen": 711983104
},
{
"epoch": 0.22,
"learning_rate": 0.00039609829488465397,
"loss": 2.87,
"theoretical_loss": 3.7729223709990016,
"tokens_seen": 712048640
},
{
"epoch": 0.22,
"learning_rate": 0.00039608826479438315,
"loss": 2.6317,
"theoretical_loss": 3.7728872210485296,
"tokens_seen": 712114176
},
{
"epoch": 0.22,
"learning_rate": 0.0003960782347041124,
"loss": 2.9067,
"theoretical_loss": 3.77285207523843,
"tokens_seen": 712179712
},
{
"epoch": 0.22,
"learning_rate": 0.0003960682046138415,
"loss": 2.8358,
"theoretical_loss": 3.7728169335678343,
"tokens_seen": 712245248
},
{
"epoch": 0.22,
"learning_rate": 0.00039605817452357075,
"loss": 2.9915,
"theoretical_loss": 3.772781796035874,
"tokens_seen": 712310784
},
{
"epoch": 0.22,
"learning_rate": 0.0003960481444332999,
"loss": 2.6196,
"theoretical_loss": 3.772746662641681,
"tokens_seen": 712376320
},
{
"epoch": 0.22,
"learning_rate": 0.0003960381143430291,
"loss": 2.9639,
"theoretical_loss": 3.7727115333843875,
"tokens_seen": 712441856
},
{
"epoch": 0.22,
"learning_rate": 0.0003960280842527583,
"loss": 2.8469,
"theoretical_loss": 3.772676408263126,
"tokens_seen": 712507392
},
{
"epoch": 0.22,
"learning_rate": 0.00039601805416248747,
"loss": 2.904,
"theoretical_loss": 3.772641287277029,
"tokens_seen": 712572928
},
{
"epoch": 0.22,
"learning_rate": 0.00039600802407221665,
"loss": 2.9758,
"theoretical_loss": 3.772606170425229,
"tokens_seen": 712638464
},
{
"epoch": 0.22,
"objective/train/docs_used": 412148,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9750280380249023,
"objective/train/theoretical_loss": 3.77257105770686,
"objective/train/tokens_used": 733164000,
"theoretical_loss": 3.77257105770686,
"tokens_seen": 712704000
},
{
"epoch": 0.22,
"learning_rate": 0.00039599799398194583,
"loss": 2.9183,
"theoretical_loss": 3.77257105770686,
"tokens_seen": 712704000
},
{
"epoch": 0.22,
"learning_rate": 0.000395987963891675,
"loss": 2.9977,
"theoretical_loss": 3.772535949121055,
"tokens_seen": 712769536
},
{
"epoch": 0.22,
"learning_rate": 0.00039597793380140425,
"loss": 2.7235,
"theoretical_loss": 3.772500844666948,
"tokens_seen": 712835072
},
{
"epoch": 0.22,
"learning_rate": 0.0003959679037111334,
"loss": 2.7663,
"theoretical_loss": 3.7724657443436724,
"tokens_seen": 712900608
},
{
"epoch": 0.22,
"learning_rate": 0.0003959578736208626,
"loss": 3.1449,
"theoretical_loss": 3.7724306481503636,
"tokens_seen": 712966144
},
{
"epoch": 0.22,
"learning_rate": 0.0003959478435305918,
"loss": 2.7778,
"theoretical_loss": 3.7723955560861544,
"tokens_seen": 713031680
},
{
"epoch": 0.22,
"learning_rate": 0.000395937813440321,
"loss": 2.7314,
"theoretical_loss": 3.7723604681501812,
"tokens_seen": 713097216
},
{
"epoch": 0.22,
"learning_rate": 0.00039592778335005016,
"loss": 2.9023,
"theoretical_loss": 3.7723253843415785,
"tokens_seen": 713162752
},
{
"epoch": 0.22,
"learning_rate": 0.00039591775325977934,
"loss": 2.9319,
"theoretical_loss": 3.7722903046594807,
"tokens_seen": 713228288
},
{
"epoch": 0.22,
"learning_rate": 0.0003959077231695085,
"loss": 2.987,
"theoretical_loss": 3.772255229103025,
"tokens_seen": 713293824
},
{
"epoch": 0.22,
"learning_rate": 0.00039589769307923775,
"loss": 2.7198,
"theoretical_loss": 3.772220157671345,
"tokens_seen": 713359360
},
{
"epoch": 0.22,
"learning_rate": 0.0003958876629889669,
"loss": 2.7888,
"theoretical_loss": 3.7721850903635787,
"tokens_seen": 713424896
},
{
"epoch": 0.22,
"learning_rate": 0.0003958776328986961,
"loss": 2.7427,
"theoretical_loss": 3.7721500271788617,
"tokens_seen": 713490432
},
{
"epoch": 0.22,
"learning_rate": 0.00039586760280842524,
"loss": 2.7095,
"theoretical_loss": 3.772114968116331,
"tokens_seen": 713555968
},
{
"epoch": 0.22,
"learning_rate": 0.0003958575727181545,
"loss": 3.0662,
"theoretical_loss": 3.772079913175123,
"tokens_seen": 713621504
},
{
"epoch": 0.22,
"learning_rate": 0.00039584754262788366,
"loss": 2.8181,
"theoretical_loss": 3.7720448623543743,
"tokens_seen": 713687040
},
{
"epoch": 0.22,
"learning_rate": 0.00039583751253761284,
"loss": 3.064,
"theoretical_loss": 3.7720098156532234,
"tokens_seen": 713752576
},
{
"epoch": 0.22,
"learning_rate": 0.000395827482447342,
"loss": 2.7751,
"theoretical_loss": 3.771974773070807,
"tokens_seen": 713818112
},
{
"epoch": 0.22,
"learning_rate": 0.0003958174523570712,
"loss": 2.6367,
"theoretical_loss": 3.7719397346062635,
"tokens_seen": 713883648
},
{
"epoch": 0.22,
"learning_rate": 0.0003958074222668004,
"loss": 2.8795,
"theoretical_loss": 3.771904700258731,
"tokens_seen": 713949184
},
{
"epoch": 0.22,
"learning_rate": 0.0003957973921765296,
"loss": 2.6557,
"theoretical_loss": 3.771869670027348,
"tokens_seen": 714014720
},
{
"epoch": 0.22,
"learning_rate": 0.00039578736208625875,
"loss": 2.9415,
"theoretical_loss": 3.7718346439112524,
"tokens_seen": 714080256
},
{
"epoch": 0.22,
"learning_rate": 0.000395777331995988,
"loss": 2.7826,
"theoretical_loss": 3.771799621909584,
"tokens_seen": 714145792
},
{
"epoch": 0.22,
"learning_rate": 0.00039576730190571716,
"loss": 2.9316,
"theoretical_loss": 3.771764604021482,
"tokens_seen": 714211328
},
{
"epoch": 0.22,
"learning_rate": 0.00039575727181544634,
"loss": 2.7793,
"theoretical_loss": 3.771729590246085,
"tokens_seen": 714276864
},
{
"epoch": 0.22,
"objective/train/docs_used": 413348,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.090303421020508,
"objective/train/theoretical_loss": 3.7716945805825337,
"objective/train/tokens_used": 734802400,
"theoretical_loss": 3.7716945805825337,
"tokens_seen": 714342400
},
{
"epoch": 0.22,
"learning_rate": 0.0003957472417251755,
"loss": 3.0001,
"theoretical_loss": 3.7716945805825337,
"tokens_seen": 714342400
},
{
"epoch": 0.22,
"learning_rate": 0.0003957372116349047,
"loss": 3.0458,
"theoretical_loss": 3.7716595750299673,
"tokens_seen": 714407936
},
{
"epoch": 0.22,
"learning_rate": 0.0003957271815446339,
"loss": 2.7011,
"theoretical_loss": 3.771624573587526,
"tokens_seen": 714473472
},
{
"epoch": 0.22,
"learning_rate": 0.0003957171514543631,
"loss": 2.8247,
"theoretical_loss": 3.771589576254351,
"tokens_seen": 714539008
},
{
"epoch": 0.22,
"learning_rate": 0.00039570712136409225,
"loss": 2.7707,
"theoretical_loss": 3.7715545830295825,
"tokens_seen": 714604544
},
{
"epoch": 0.22,
"learning_rate": 0.0003956970912738215,
"loss": 2.7799,
"theoretical_loss": 3.7715195939123616,
"tokens_seen": 714670080
},
{
"epoch": 0.22,
"learning_rate": 0.0003956870611835506,
"loss": 3.0558,
"theoretical_loss": 3.7714846089018295,
"tokens_seen": 714735616
},
{
"epoch": 0.22,
"learning_rate": 0.00039567703109327985,
"loss": 2.7486,
"theoretical_loss": 3.7714496279971277,
"tokens_seen": 714801152
},
{
"epoch": 0.22,
"learning_rate": 0.00039566700100300903,
"loss": 2.9199,
"theoretical_loss": 3.7714146511973983,
"tokens_seen": 714866688
},
{
"epoch": 0.22,
"learning_rate": 0.0003956569709127382,
"loss": 2.8739,
"theoretical_loss": 3.7713796785017824,
"tokens_seen": 714932224
},
{
"epoch": 0.22,
"learning_rate": 0.0003956469408224674,
"loss": 2.7182,
"theoretical_loss": 3.7713447099094233,
"tokens_seen": 714997760
},
{
"epoch": 0.22,
"learning_rate": 0.00039563691073219657,
"loss": 2.7735,
"theoretical_loss": 3.7713097454194635,
"tokens_seen": 715063296
},
{
"epoch": 0.22,
"learning_rate": 0.0003956268806419258,
"loss": 2.8128,
"theoretical_loss": 3.7712747850310446,
"tokens_seen": 715128832
},
{
"epoch": 0.22,
"learning_rate": 0.000395616850551655,
"loss": 2.7627,
"theoretical_loss": 3.7712398287433113,
"tokens_seen": 715194368
},
{
"epoch": 0.22,
"learning_rate": 0.00039560682046138417,
"loss": 2.9056,
"theoretical_loss": 3.771204876555406,
"tokens_seen": 715259904
},
{
"epoch": 0.22,
"learning_rate": 0.00039559679037111335,
"loss": 2.9789,
"theoretical_loss": 3.771169928466472,
"tokens_seen": 715325440
},
{
"epoch": 0.22,
"learning_rate": 0.0003955867602808426,
"loss": 2.6652,
"theoretical_loss": 3.7711349844756543,
"tokens_seen": 715390976
},
{
"epoch": 0.22,
"learning_rate": 0.0003955767301905717,
"loss": 2.7324,
"theoretical_loss": 3.771100044582096,
"tokens_seen": 715456512
},
{
"epoch": 0.22,
"learning_rate": 0.00039556670010030095,
"loss": 2.9166,
"theoretical_loss": 3.771065108784941,
"tokens_seen": 715522048
},
{
"epoch": 0.22,
"learning_rate": 0.0003955566700100301,
"loss": 2.7082,
"theoretical_loss": 3.771030177083335,
"tokens_seen": 715587584
},
{
"epoch": 0.22,
"learning_rate": 0.0003955466399197593,
"loss": 2.8659,
"theoretical_loss": 3.7709952494764227,
"tokens_seen": 715653120
},
{
"epoch": 0.22,
"learning_rate": 0.0003955366098294885,
"loss": 2.8094,
"theoretical_loss": 3.7709603259633484,
"tokens_seen": 715718656
},
{
"epoch": 0.22,
"learning_rate": 0.00039552657973921767,
"loss": 3.0041,
"theoretical_loss": 3.7709254065432583,
"tokens_seen": 715784192
},
{
"epoch": 0.22,
"learning_rate": 0.00039551654964894685,
"loss": 2.7955,
"theoretical_loss": 3.770890491215298,
"tokens_seen": 715849728
},
{
"epoch": 0.22,
"learning_rate": 0.00039550651955867603,
"loss": 3.1147,
"theoretical_loss": 3.7708555799786128,
"tokens_seen": 715915264
},
{
"epoch": 0.22,
"objective/train/docs_used": 413910,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.962822198867798,
"objective/train/theoretical_loss": 3.7708206728323495,
"objective/train/tokens_used": 736440800,
"theoretical_loss": 3.7708206728323495,
"tokens_seen": 715980800
},
{
"epoch": 0.22,
"learning_rate": 0.0003954964894684052,
"loss": 3.0961,
"theoretical_loss": 3.7708206728323495,
"tokens_seen": 715980800
},
{
"epoch": 0.22,
"learning_rate": 0.00039548645937813445,
"loss": 2.8606,
"theoretical_loss": 3.770785769775654,
"tokens_seen": 716046336
},
{
"epoch": 0.22,
"learning_rate": 0.0003954764292878636,
"loss": 2.7818,
"theoretical_loss": 3.7707508708076727,
"tokens_seen": 716111872
},
{
"epoch": 0.22,
"learning_rate": 0.0003954663991975928,
"loss": 2.6957,
"theoretical_loss": 3.770715975927553,
"tokens_seen": 716177408
},
{
"epoch": 0.22,
"learning_rate": 0.000395456369107322,
"loss": 2.8518,
"theoretical_loss": 3.7706810851344423,
"tokens_seen": 716242944
},
{
"epoch": 0.22,
"learning_rate": 0.0003954463390170512,
"loss": 2.6311,
"theoretical_loss": 3.7706461984274875,
"tokens_seen": 716308480
},
{
"epoch": 0.22,
"learning_rate": 0.00039543630892678036,
"loss": 2.9358,
"theoretical_loss": 3.7706113158058363,
"tokens_seen": 716374016
},
{
"epoch": 0.22,
"learning_rate": 0.00039542627883650954,
"loss": 2.958,
"theoretical_loss": 3.7705764372686374,
"tokens_seen": 716439552
},
{
"epoch": 0.22,
"learning_rate": 0.0003954162487462387,
"loss": 3.054,
"theoretical_loss": 3.770541562815038,
"tokens_seen": 716505088
},
{
"epoch": 0.22,
"learning_rate": 0.00039540621865596795,
"loss": 2.9831,
"theoretical_loss": 3.770506692444187,
"tokens_seen": 716570624
},
{
"epoch": 0.22,
"learning_rate": 0.0003953961885656971,
"loss": 2.8916,
"theoretical_loss": 3.7704718261552332,
"tokens_seen": 716636160
},
{
"epoch": 0.22,
"learning_rate": 0.0003953861584754263,
"loss": 2.8749,
"theoretical_loss": 3.770436963947325,
"tokens_seen": 716701696
},
{
"epoch": 0.22,
"learning_rate": 0.00039537612838515544,
"loss": 2.8681,
"theoretical_loss": 3.7704021058196124,
"tokens_seen": 716767232
},
{
"epoch": 0.22,
"learning_rate": 0.0003953660982948847,
"loss": 2.9874,
"theoretical_loss": 3.770367251771245,
"tokens_seen": 716832768
},
{
"epoch": 0.22,
"learning_rate": 0.00039535606820461386,
"loss": 2.9044,
"theoretical_loss": 3.7703324018013706,
"tokens_seen": 716898304
},
{
"epoch": 0.22,
"learning_rate": 0.00039534603811434304,
"loss": 2.9409,
"theoretical_loss": 3.7702975559091416,
"tokens_seen": 716963840
},
{
"epoch": 0.22,
"learning_rate": 0.0003953360080240722,
"loss": 2.7377,
"theoretical_loss": 3.7702627140937075,
"tokens_seen": 717029376
},
{
"epoch": 0.22,
"learning_rate": 0.0003953259779338014,
"loss": 2.9035,
"theoretical_loss": 3.7702278763542174,
"tokens_seen": 717094912
},
{
"epoch": 0.22,
"learning_rate": 0.0003953159478435306,
"loss": 2.7807,
"theoretical_loss": 3.770193042689824,
"tokens_seen": 717160448
},
{
"epoch": 0.22,
"learning_rate": 0.0003953059177532598,
"loss": 2.792,
"theoretical_loss": 3.770158213099677,
"tokens_seen": 717225984
},
{
"epoch": 0.22,
"learning_rate": 0.00039529588766298895,
"loss": 2.9959,
"theoretical_loss": 3.7701233875829283,
"tokens_seen": 717291520
},
{
"epoch": 0.22,
"learning_rate": 0.0003952858575727182,
"loss": 2.8529,
"theoretical_loss": 3.7700885661387287,
"tokens_seen": 717357056
},
{
"epoch": 0.22,
"learning_rate": 0.00039527582748244736,
"loss": 2.7021,
"theoretical_loss": 3.770053748766231,
"tokens_seen": 717422592
},
{
"epoch": 0.22,
"learning_rate": 0.00039526579739217654,
"loss": 3.0528,
"theoretical_loss": 3.770018935464586,
"tokens_seen": 717488128
},
{
"epoch": 0.22,
"learning_rate": 0.0003952557673019057,
"loss": 2.9439,
"theoretical_loss": 3.7699841262329477,
"tokens_seen": 717553664
},
{
"epoch": 0.22,
"objective/train/docs_used": 414484,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9759528636932373,
"objective/train/theoretical_loss": 3.7699493210704667,
"objective/train/tokens_used": 738079200,
"theoretical_loss": 3.7699493210704667,
"tokens_seen": 717619200
},
{
"epoch": 0.22,
"learning_rate": 0.0003952457372116349,
"loss": 2.7352,
"theoretical_loss": 3.7699493210704667,
"tokens_seen": 717619200
},
{
"epoch": 0.22,
"learning_rate": 0.0003952357071213641,
"loss": 2.9144,
"theoretical_loss": 3.7699145199762967,
"tokens_seen": 717684736
},
{
"epoch": 0.22,
"learning_rate": 0.0003952256770310933,
"loss": 2.7381,
"theoretical_loss": 3.7698797229495904,
"tokens_seen": 717750272
},
{
"epoch": 0.22,
"learning_rate": 0.00039521564694082245,
"loss": 2.5926,
"theoretical_loss": 3.7698449299895014,
"tokens_seen": 717815808
},
{
"epoch": 0.22,
"learning_rate": 0.0003952056168505517,
"loss": 2.8309,
"theoretical_loss": 3.769810141095183,
"tokens_seen": 717881344
},
{
"epoch": 0.22,
"learning_rate": 0.0003951955867602808,
"loss": 2.8086,
"theoretical_loss": 3.76977535626579,
"tokens_seen": 717946880
},
{
"epoch": 0.22,
"learning_rate": 0.00039518555667001005,
"loss": 2.7767,
"theoretical_loss": 3.769740575500475,
"tokens_seen": 718012416
},
{
"epoch": 0.22,
"learning_rate": 0.00039517552657973923,
"loss": 2.9595,
"theoretical_loss": 3.769705798798393,
"tokens_seen": 718077952
},
{
"epoch": 0.22,
"learning_rate": 0.0003951654964894684,
"loss": 2.6771,
"theoretical_loss": 3.769671026158698,
"tokens_seen": 718143488
},
{
"epoch": 0.22,
"learning_rate": 0.0003951554663991976,
"loss": 2.86,
"theoretical_loss": 3.769636257580545,
"tokens_seen": 718209024
},
{
"epoch": 0.22,
"learning_rate": 0.00039514543630892677,
"loss": 3.0575,
"theoretical_loss": 3.76960149306309,
"tokens_seen": 718274560
},
{
"epoch": 0.22,
"learning_rate": 0.00039513540621865595,
"loss": 2.756,
"theoretical_loss": 3.769566732605487,
"tokens_seen": 718340096
},
{
"epoch": 0.22,
"learning_rate": 0.0003951253761283852,
"loss": 2.6987,
"theoretical_loss": 3.769531976206893,
"tokens_seen": 718405632
},
{
"epoch": 0.22,
"learning_rate": 0.0003951153460381143,
"loss": 2.7111,
"theoretical_loss": 3.7694972238664617,
"tokens_seen": 718471168
},
{
"epoch": 0.22,
"learning_rate": 0.00039510531594784355,
"loss": 3.0548,
"theoretical_loss": 3.7694624755833512,
"tokens_seen": 718536704
},
{
"epoch": 0.22,
"learning_rate": 0.00039509528585757273,
"loss": 2.964,
"theoretical_loss": 3.769427731356717,
"tokens_seen": 718602240
},
{
"epoch": 0.22,
"learning_rate": 0.0003950852557673019,
"loss": 2.9854,
"theoretical_loss": 3.7693929911857156,
"tokens_seen": 718667776
},
{
"epoch": 0.22,
"learning_rate": 0.0003950752256770311,
"loss": 2.7423,
"theoretical_loss": 3.7693582550695037,
"tokens_seen": 718733312
},
{
"epoch": 0.22,
"learning_rate": 0.0003950651955867603,
"loss": 3.0143,
"theoretical_loss": 3.7693235230072384,
"tokens_seen": 718798848
},
{
"epoch": 0.22,
"learning_rate": 0.00039505516549648946,
"loss": 2.7951,
"theoretical_loss": 3.7692887949980776,
"tokens_seen": 718864384
},
{
"epoch": 0.22,
"learning_rate": 0.0003950451354062187,
"loss": 2.7409,
"theoretical_loss": 3.7692540710411784,
"tokens_seen": 718929920
},
{
"epoch": 0.22,
"learning_rate": 0.0003950351053159478,
"loss": 2.6655,
"theoretical_loss": 3.769219351135699,
"tokens_seen": 718995456
},
{
"epoch": 0.22,
"learning_rate": 0.00039502507522567705,
"loss": 2.9093,
"theoretical_loss": 3.769184635280797,
"tokens_seen": 719060992
},
{
"epoch": 0.22,
"learning_rate": 0.0003950150451354062,
"loss": 2.8913,
"theoretical_loss": 3.7691499234756303,
"tokens_seen": 719126528
},
{
"epoch": 0.22,
"learning_rate": 0.0003950050150451354,
"loss": 3.0383,
"theoretical_loss": 3.7691152157193586,
"tokens_seen": 719192064
},
{
"epoch": 0.22,
"objective/train/docs_used": 415542,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8314599990844727,
"objective/train/theoretical_loss": 3.76908051201114,
"objective/train/tokens_used": 739717600,
"theoretical_loss": 3.76908051201114,
"tokens_seen": 719257600
},
{
"epoch": 0.22,
"learning_rate": 0.0003949949849548646,
"loss": 2.867,
"theoretical_loss": 3.76908051201114,
"tokens_seen": 719257600
},
{
"epoch": 0.22,
"learning_rate": 0.0003949849548645938,
"loss": 2.8603,
"theoretical_loss": 3.769045812350134,
"tokens_seen": 719323136
},
{
"epoch": 0.22,
"learning_rate": 0.00039497492477432296,
"loss": 2.8516,
"theoretical_loss": 3.7690111167355,
"tokens_seen": 719388672
},
{
"epoch": 0.22,
"learning_rate": 0.0003949648946840522,
"loss": 2.9873,
"theoretical_loss": 3.768976425166397,
"tokens_seen": 719454208
},
{
"epoch": 0.22,
"learning_rate": 0.0003949548645937813,
"loss": 2.7485,
"theoretical_loss": 3.7689417376419856,
"tokens_seen": 719519744
},
{
"epoch": 0.22,
"learning_rate": 0.00039494483450351056,
"loss": 3.003,
"theoretical_loss": 3.768907054161425,
"tokens_seen": 719585280
},
{
"epoch": 0.22,
"learning_rate": 0.0003949348044132397,
"loss": 2.6711,
"theoretical_loss": 3.768872374723876,
"tokens_seen": 719650816
},
{
"epoch": 0.22,
"learning_rate": 0.0003949247743229689,
"loss": 2.8642,
"theoretical_loss": 3.7688376993284995,
"tokens_seen": 719716352
},
{
"epoch": 0.22,
"learning_rate": 0.0003949147442326981,
"loss": 2.5387,
"theoretical_loss": 3.768803027974456,
"tokens_seen": 719781888
},
{
"epoch": 0.22,
"learning_rate": 0.0003949047141424273,
"loss": 3.0536,
"theoretical_loss": 3.768768360660907,
"tokens_seen": 719847424
},
{
"epoch": 0.22,
"learning_rate": 0.0003948946840521565,
"loss": 3.0176,
"theoretical_loss": 3.768733697387013,
"tokens_seen": 719912960
},
{
"epoch": 0.22,
"learning_rate": 0.00039488465396188564,
"loss": 2.6639,
"theoretical_loss": 3.7686990381519365,
"tokens_seen": 719978496
},
{
"epoch": 0.22,
"learning_rate": 0.0003948746238716149,
"loss": 3.0455,
"theoretical_loss": 3.7686643829548387,
"tokens_seen": 720044032
},
{
"epoch": 0.22,
"learning_rate": 0.00039486459378134406,
"loss": 2.8563,
"theoretical_loss": 3.768629731794882,
"tokens_seen": 720109568
},
{
"epoch": 0.22,
"learning_rate": 0.00039485456369107324,
"loss": 2.6986,
"theoretical_loss": 3.7685950846712286,
"tokens_seen": 720175104
},
{
"epoch": 0.22,
"learning_rate": 0.0003948445336008024,
"loss": 2.9714,
"theoretical_loss": 3.768560441583041,
"tokens_seen": 720240640
},
{
"epoch": 0.22,
"learning_rate": 0.0003948345035105316,
"loss": 2.7563,
"theoretical_loss": 3.7685258025294823,
"tokens_seen": 720306176
},
{
"epoch": 0.22,
"learning_rate": 0.0003948244734202608,
"loss": 2.8306,
"theoretical_loss": 3.768491167509716,
"tokens_seen": 720371712
},
{
"epoch": 0.22,
"learning_rate": 0.00039481444332999,
"loss": 2.9877,
"theoretical_loss": 3.768456536522905,
"tokens_seen": 720437248
},
{
"epoch": 0.22,
"learning_rate": 0.00039480441323971915,
"loss": 2.9809,
"theoretical_loss": 3.768421909568212,
"tokens_seen": 720502784
},
{
"epoch": 0.22,
"learning_rate": 0.0003947943831494484,
"loss": 2.7219,
"theoretical_loss": 3.768387286644802,
"tokens_seen": 720568320
},
{
"epoch": 0.22,
"learning_rate": 0.00039478435305917756,
"loss": 2.879,
"theoretical_loss": 3.768352667751839,
"tokens_seen": 720633856
},
{
"epoch": 0.22,
"learning_rate": 0.00039477432296890674,
"loss": 2.7271,
"theoretical_loss": 3.768318052888487,
"tokens_seen": 720699392
},
{
"epoch": 0.22,
"learning_rate": 0.0003947642928786359,
"loss": 3.0329,
"theoretical_loss": 3.768283442053911,
"tokens_seen": 720764928
},
{
"epoch": 0.22,
"learning_rate": 0.0003947542627883651,
"loss": 2.8737,
"theoretical_loss": 3.7682488352472756,
"tokens_seen": 720830464
},
{
"epoch": 0.22,
"objective/train/docs_used": 416165,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1811347007751465,
"objective/train/theoretical_loss": 3.7682142324677455,
"objective/train/tokens_used": 741356000,
"theoretical_loss": 3.7682142324677455,
"tokens_seen": 720896000
},
{
"epoch": 0.22,
"learning_rate": 0.0003947442326980943,
"loss": 2.8187,
"theoretical_loss": 3.7682142324677455,
"tokens_seen": 720896000
},
{
"epoch": 0.22,
"learning_rate": 0.0003947342026078235,
"loss": 2.873,
"theoretical_loss": 3.7681796337144866,
"tokens_seen": 720961536
},
{
"epoch": 0.22,
"learning_rate": 0.00039472417251755265,
"loss": 2.9994,
"theoretical_loss": 3.768145038986664,
"tokens_seen": 721027072
},
{
"epoch": 0.22,
"learning_rate": 0.0003947141424272819,
"loss": 2.8722,
"theoretical_loss": 3.768110448283444,
"tokens_seen": 721092608
},
{
"epoch": 0.22,
"learning_rate": 0.000394704112337011,
"loss": 2.7781,
"theoretical_loss": 3.768075861603993,
"tokens_seen": 721158144
},
{
"epoch": 0.22,
"learning_rate": 0.00039469408224674025,
"loss": 2.9471,
"theoretical_loss": 3.7680412789474764,
"tokens_seen": 721223680
},
{
"epoch": 0.22,
"learning_rate": 0.00039468405215646943,
"loss": 3.0068,
"theoretical_loss": 3.7680067003130615,
"tokens_seen": 721289216
},
{
"epoch": 0.22,
"learning_rate": 0.0003946740220661986,
"loss": 2.9134,
"theoretical_loss": 3.767972125699915,
"tokens_seen": 721354752
},
{
"epoch": 0.22,
"learning_rate": 0.0003946639919759278,
"loss": 2.8378,
"theoretical_loss": 3.767937555107203,
"tokens_seen": 721420288
},
{
"epoch": 0.22,
"learning_rate": 0.00039465396188565697,
"loss": 2.723,
"theoretical_loss": 3.767902988534094,
"tokens_seen": 721485824
},
{
"epoch": 0.22,
"learning_rate": 0.00039464393179538615,
"loss": 2.8849,
"theoretical_loss": 3.7678684259797555,
"tokens_seen": 721551360
},
{
"epoch": 0.22,
"learning_rate": 0.0003946339017051154,
"loss": 2.8551,
"theoretical_loss": 3.767833867443355,
"tokens_seen": 721616896
},
{
"epoch": 0.22,
"learning_rate": 0.0003946238716148445,
"loss": 2.8338,
"theoretical_loss": 3.767799312924061,
"tokens_seen": 721682432
},
{
"epoch": 0.22,
"learning_rate": 0.00039461384152457375,
"loss": 2.616,
"theoretical_loss": 3.7677647624210406,
"tokens_seen": 721747968
},
{
"epoch": 0.22,
"learning_rate": 0.00039460381143430293,
"loss": 2.6381,
"theoretical_loss": 3.7677302159334642,
"tokens_seen": 721813504
},
{
"epoch": 0.22,
"learning_rate": 0.0003945937813440321,
"loss": 2.6573,
"theoretical_loss": 3.7676956734604987,
"tokens_seen": 721879040
},
{
"epoch": 0.22,
"learning_rate": 0.0003945837512537613,
"loss": 2.8246,
"theoretical_loss": 3.7676611350013145,
"tokens_seen": 721944576
},
{
"epoch": 0.22,
"learning_rate": 0.0003945737211634905,
"loss": 2.7629,
"theoretical_loss": 3.7676266005550803,
"tokens_seen": 722010112
},
{
"epoch": 0.22,
"learning_rate": 0.00039456369107321966,
"loss": 2.6521,
"theoretical_loss": 3.7675920701209664,
"tokens_seen": 722075648
},
{
"epoch": 0.22,
"learning_rate": 0.0003945536609829489,
"loss": 2.9961,
"theoretical_loss": 3.7675575436981417,
"tokens_seen": 722141184
},
{
"epoch": 0.22,
"learning_rate": 0.000394543630892678,
"loss": 2.8083,
"theoretical_loss": 3.767523021285777,
"tokens_seen": 722206720
},
{
"epoch": 0.22,
"learning_rate": 0.00039453360080240725,
"loss": 2.8144,
"theoretical_loss": 3.7674885028830416,
"tokens_seen": 722272256
},
{
"epoch": 0.22,
"learning_rate": 0.0003945235707121364,
"loss": 2.7149,
"theoretical_loss": 3.767453988489107,
"tokens_seen": 722337792
},
{
"epoch": 0.22,
"learning_rate": 0.0003945135406218656,
"loss": 2.5914,
"theoretical_loss": 3.767419478103144,
"tokens_seen": 722403328
},
{
"epoch": 0.22,
"learning_rate": 0.0003945035105315948,
"loss": 2.6925,
"theoretical_loss": 3.7673849717243226,
"tokens_seen": 722468864
},
{
"epoch": 0.22,
"objective/train/docs_used": 417606,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7808585166931152,
"objective/train/theoretical_loss": 3.7673504693518156,
"objective/train/tokens_used": 742994400,
"theoretical_loss": 3.7673504693518156,
"tokens_seen": 722534400
},
{
"epoch": 0.22,
"learning_rate": 0.000394493480441324,
"loss": 2.7173,
"theoretical_loss": 3.7673504693518156,
"tokens_seen": 722534400
},
{
"epoch": 0.22,
"learning_rate": 0.00039448345035105316,
"loss": 2.7914,
"theoretical_loss": 3.767315970984793,
"tokens_seen": 722599936
},
{
"epoch": 0.22,
"learning_rate": 0.0003944734202607824,
"loss": 2.9235,
"theoretical_loss": 3.7672814766224274,
"tokens_seen": 722665472
},
{
"epoch": 0.22,
"learning_rate": 0.0003944633901705115,
"loss": 2.9004,
"theoretical_loss": 3.767246986263891,
"tokens_seen": 722731008
},
{
"epoch": 0.22,
"learning_rate": 0.00039445336008024076,
"loss": 2.9311,
"theoretical_loss": 3.7672124999083554,
"tokens_seen": 722796544
},
{
"epoch": 0.22,
"learning_rate": 0.0003944433299899699,
"loss": 2.6554,
"theoretical_loss": 3.767178017554994,
"tokens_seen": 722862080
},
{
"epoch": 0.22,
"learning_rate": 0.0003944332998996991,
"loss": 2.6188,
"theoretical_loss": 3.7671435392029786,
"tokens_seen": 722927616
},
{
"epoch": 0.22,
"learning_rate": 0.0003944232698094283,
"loss": 2.9299,
"theoretical_loss": 3.767109064851483,
"tokens_seen": 722993152
},
{
"epoch": 0.22,
"learning_rate": 0.0003944132397191575,
"loss": 2.7601,
"theoretical_loss": 3.7670745944996797,
"tokens_seen": 723058688
},
{
"epoch": 0.22,
"learning_rate": 0.00039440320962888666,
"loss": 2.7038,
"theoretical_loss": 3.7670401281467427,
"tokens_seen": 723124224
},
{
"epoch": 0.22,
"learning_rate": 0.00039439317953861584,
"loss": 2.9328,
"theoretical_loss": 3.767005665791846,
"tokens_seen": 723189760
},
{
"epoch": 0.22,
"learning_rate": 0.000394383149448345,
"loss": 2.7828,
"theoretical_loss": 3.766971207434163,
"tokens_seen": 723255296
},
{
"epoch": 0.22,
"learning_rate": 0.00039437311935807426,
"loss": 2.795,
"theoretical_loss": 3.766936753072869,
"tokens_seen": 723320832
},
{
"epoch": 0.22,
"learning_rate": 0.0003943630892678034,
"loss": 2.6048,
"theoretical_loss": 3.7669023027071367,
"tokens_seen": 723386368
},
{
"epoch": 0.22,
"learning_rate": 0.0003943530591775326,
"loss": 3.0816,
"theoretical_loss": 3.766867856336142,
"tokens_seen": 723451904
},
{
"epoch": 0.22,
"learning_rate": 0.00039434302908726175,
"loss": 3.0234,
"theoretical_loss": 3.76683341395906,
"tokens_seen": 723517440
},
{
"epoch": 0.22,
"learning_rate": 0.000394332998996991,
"loss": 2.8156,
"theoretical_loss": 3.7667989755750657,
"tokens_seen": 723582976
},
{
"epoch": 0.22,
"learning_rate": 0.00039432296890672017,
"loss": 2.795,
"theoretical_loss": 3.766764541183334,
"tokens_seen": 723648512
},
{
"epoch": 0.22,
"learning_rate": 0.00039431293881644935,
"loss": 2.8163,
"theoretical_loss": 3.7667301107830413,
"tokens_seen": 723714048
},
{
"epoch": 0.22,
"learning_rate": 0.00039430290872617853,
"loss": 2.7497,
"theoretical_loss": 3.766695684373363,
"tokens_seen": 723779584
},
{
"epoch": 0.22,
"learning_rate": 0.00039429287863590776,
"loss": 2.8145,
"theoretical_loss": 3.766661261953476,
"tokens_seen": 723845120
},
{
"epoch": 0.22,
"learning_rate": 0.0003942828485456369,
"loss": 3.0573,
"theoretical_loss": 3.7666268435225563,
"tokens_seen": 723910656
},
{
"epoch": 0.22,
"learning_rate": 0.0003942728184553661,
"loss": 2.8575,
"theoretical_loss": 3.7665924290797808,
"tokens_seen": 723976192
},
{
"epoch": 0.22,
"learning_rate": 0.00039426278836509525,
"loss": 2.9934,
"theoretical_loss": 3.766558018624326,
"tokens_seen": 724041728
},
{
"epoch": 0.22,
"learning_rate": 0.0003942527582748245,
"loss": 3.0244,
"theoretical_loss": 3.7665236121553693,
"tokens_seen": 724107264
},
{
"epoch": 0.22,
"objective/train/docs_used": 418215,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7131829261779785,
"objective/train/theoretical_loss": 3.7664892096720886,
"objective/train/tokens_used": 744632800,
"theoretical_loss": 3.7664892096720886,
"tokens_seen": 724172800
},
{
"epoch": 0.22,
"learning_rate": 0.00039424272818455367,
"loss": 3.0867,
"theoretical_loss": 3.7664892096720886,
"tokens_seen": 724172800
},
{
"epoch": 0.22,
"learning_rate": 0.00039423269809428285,
"loss": 2.7673,
"theoretical_loss": 3.766454811173661,
"tokens_seen": 724238336
},
{
"epoch": 0.22,
"learning_rate": 0.00039422266800401203,
"loss": 2.8963,
"theoretical_loss": 3.7664204166592645,
"tokens_seen": 724303872
},
{
"epoch": 0.22,
"learning_rate": 0.0003942126379137412,
"loss": 2.6024,
"theoretical_loss": 3.7663860261280773,
"tokens_seen": 724369408
},
{
"epoch": 0.22,
"learning_rate": 0.0003942026078234704,
"loss": 2.7605,
"theoretical_loss": 3.7663516395792778,
"tokens_seen": 724434944
},
{
"epoch": 0.22,
"learning_rate": 0.00039419257773319963,
"loss": 2.926,
"theoretical_loss": 3.7663172570120445,
"tokens_seen": 724500480
},
{
"epoch": 0.22,
"learning_rate": 0.00039418254764292876,
"loss": 2.6386,
"theoretical_loss": 3.7662828784255566,
"tokens_seen": 724566016
},
{
"epoch": 0.22,
"learning_rate": 0.000394172517552658,
"loss": 2.7594,
"theoretical_loss": 3.766248503818993,
"tokens_seen": 724631552
},
{
"epoch": 0.22,
"learning_rate": 0.0003941624874623871,
"loss": 2.82,
"theoretical_loss": 3.766214133191533,
"tokens_seen": 724697088
},
{
"epoch": 0.22,
"learning_rate": 0.00039415245737211635,
"loss": 2.6962,
"theoretical_loss": 3.7661797665423564,
"tokens_seen": 724762624
},
{
"epoch": 0.22,
"learning_rate": 0.0003941424272818456,
"loss": 2.9067,
"theoretical_loss": 3.766145403870643,
"tokens_seen": 724828160
},
{
"epoch": 0.22,
"learning_rate": 0.0003941323971915747,
"loss": 2.6875,
"theoretical_loss": 3.766111045175573,
"tokens_seen": 724893696
},
{
"epoch": 0.22,
"learning_rate": 0.00039412236710130395,
"loss": 2.8427,
"theoretical_loss": 3.7660766904563268,
"tokens_seen": 724959232
},
{
"epoch": 0.22,
"learning_rate": 0.00039411233701103313,
"loss": 2.8951,
"theoretical_loss": 3.7660423397120844,
"tokens_seen": 725024768
},
{
"epoch": 0.22,
"learning_rate": 0.0003941023069207623,
"loss": 2.7557,
"theoretical_loss": 3.7660079929420274,
"tokens_seen": 725090304
},
{
"epoch": 0.22,
"learning_rate": 0.0003940922768304915,
"loss": 3.0849,
"theoretical_loss": 3.765973650145336,
"tokens_seen": 725155840
},
{
"epoch": 0.22,
"learning_rate": 0.0003940822467402207,
"loss": 2.9737,
"theoretical_loss": 3.7659393113211923,
"tokens_seen": 725221376
},
{
"epoch": 0.22,
"learning_rate": 0.00039407221664994986,
"loss": 2.9243,
"theoretical_loss": 3.7659049764687778,
"tokens_seen": 725286912
},
{
"epoch": 0.22,
"learning_rate": 0.0003940621865596791,
"loss": 2.7752,
"theoretical_loss": 3.765870645587274,
"tokens_seen": 725352448
},
{
"epoch": 0.22,
"learning_rate": 0.0003940521564694082,
"loss": 2.7664,
"theoretical_loss": 3.7658363186758628,
"tokens_seen": 725417984
},
{
"epoch": 0.22,
"learning_rate": 0.00039404212637913745,
"loss": 2.8264,
"theoretical_loss": 3.765801995733727,
"tokens_seen": 725483520
},
{
"epoch": 0.22,
"learning_rate": 0.0003940320962888666,
"loss": 2.8892,
"theoretical_loss": 3.7657676767600488,
"tokens_seen": 725549056
},
{
"epoch": 0.22,
"learning_rate": 0.0003940220661985958,
"loss": 2.4909,
"theoretical_loss": 3.7657333617540107,
"tokens_seen": 725614592
},
{
"epoch": 0.22,
"learning_rate": 0.000394012036108325,
"loss": 2.6818,
"theoretical_loss": 3.7656990507147965,
"tokens_seen": 725680128
},
{
"epoch": 0.22,
"learning_rate": 0.0003940020060180542,
"loss": 2.7511,
"theoretical_loss": 3.765664743641589,
"tokens_seen": 725745664
},
{
"epoch": 0.22,
"objective/train/docs_used": 419550,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.189182996749878,
"objective/train/theoretical_loss": 3.765630440533571,
"objective/train/tokens_used": 746271200,
"theoretical_loss": 3.765630440533571,
"tokens_seen": 725811200
},
{
"epoch": 0.22,
"learning_rate": 0.00039399197592778336,
"loss": 2.8836,
"theoretical_loss": 3.765630440533571,
"tokens_seen": 725811200
},
{
"epoch": 0.22,
"learning_rate": 0.0003939819458375126,
"loss": 2.8831,
"theoretical_loss": 3.7655961413899277,
"tokens_seen": 725876736
},
{
"epoch": 0.22,
"learning_rate": 0.0003939719157472417,
"loss": 2.8661,
"theoretical_loss": 3.765561846209842,
"tokens_seen": 725942272
},
{
"epoch": 0.22,
"learning_rate": 0.00039396188565697096,
"loss": 2.7025,
"theoretical_loss": 3.7655275549924987,
"tokens_seen": 726007808
},
{
"epoch": 0.22,
"learning_rate": 0.0003939518555667001,
"loss": 2.6265,
"theoretical_loss": 3.7654932677370816,
"tokens_seen": 726073344
},
{
"epoch": 0.22,
"learning_rate": 0.0003939418254764293,
"loss": 2.8111,
"theoretical_loss": 3.765458984442776,
"tokens_seen": 726138880
},
{
"epoch": 0.22,
"learning_rate": 0.0003939317953861585,
"loss": 2.7477,
"theoretical_loss": 3.7654247051087664,
"tokens_seen": 726204416
},
{
"epoch": 0.22,
"learning_rate": 0.0003939217652958877,
"loss": 2.9544,
"theoretical_loss": 3.765390429734238,
"tokens_seen": 726269952
},
{
"epoch": 0.22,
"learning_rate": 0.00039391173520561686,
"loss": 2.8087,
"theoretical_loss": 3.765356158318377,
"tokens_seen": 726335488
},
{
"epoch": 0.22,
"learning_rate": 0.00039390170511534604,
"loss": 2.879,
"theoretical_loss": 3.765321890860368,
"tokens_seen": 726401024
},
{
"epoch": 0.22,
"learning_rate": 0.0003938916750250752,
"loss": 2.8933,
"theoretical_loss": 3.7652876273593976,
"tokens_seen": 726466560
},
{
"epoch": 0.22,
"learning_rate": 0.00039388164493480446,
"loss": 3.0621,
"theoretical_loss": 3.765253367814652,
"tokens_seen": 726532096
},
{
"epoch": 0.22,
"learning_rate": 0.0003938716148445336,
"loss": 2.8105,
"theoretical_loss": 3.765219112225317,
"tokens_seen": 726597632
},
{
"epoch": 0.22,
"learning_rate": 0.0003938615847542628,
"loss": 2.8648,
"theoretical_loss": 3.7651848605905798,
"tokens_seen": 726663168
},
{
"epoch": 0.22,
"learning_rate": 0.00039385155466399195,
"loss": 3.0099,
"theoretical_loss": 3.7651506129096273,
"tokens_seen": 726728704
},
{
"epoch": 0.22,
"learning_rate": 0.0003938415245737212,
"loss": 2.718,
"theoretical_loss": 3.765116369181646,
"tokens_seen": 726794240
},
{
"epoch": 0.22,
"learning_rate": 0.00039383149448345037,
"loss": 2.5032,
"theoretical_loss": 3.7650821294058234,
"tokens_seen": 726859776
},
{
"epoch": 0.22,
"learning_rate": 0.00039382146439317955,
"loss": 2.6841,
"theoretical_loss": 3.7650478935813476,
"tokens_seen": 726925312
},
{
"epoch": 0.22,
"learning_rate": 0.00039381143430290873,
"loss": 2.6923,
"theoretical_loss": 3.7650136617074064,
"tokens_seen": 726990848
},
{
"epoch": 0.22,
"learning_rate": 0.00039380140421263796,
"loss": 2.8133,
"theoretical_loss": 3.7649794337831874,
"tokens_seen": 727056384
},
{
"epoch": 0.22,
"learning_rate": 0.0003937913741223671,
"loss": 2.7503,
"theoretical_loss": 3.7649452098078795,
"tokens_seen": 727121920
},
{
"epoch": 0.22,
"learning_rate": 0.0003937813440320963,
"loss": 2.8101,
"theoretical_loss": 3.7649109897806703,
"tokens_seen": 727187456
},
{
"epoch": 0.22,
"learning_rate": 0.00039377131394182545,
"loss": 2.8639,
"theoretical_loss": 3.7648767737007494,
"tokens_seen": 727252992
},
{
"epoch": 0.22,
"learning_rate": 0.0003937612838515547,
"loss": 2.6803,
"theoretical_loss": 3.7648425615673053,
"tokens_seen": 727318528
},
{
"epoch": 0.22,
"learning_rate": 0.00039375125376128387,
"loss": 2.9581,
"theoretical_loss": 3.7648083533795282,
"tokens_seen": 727384064
},
{
"epoch": 0.22,
"objective/train/docs_used": 420047,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.144989252090454,
"objective/train/theoretical_loss": 3.7647741491366067,
"objective/train/tokens_used": 747909600,
"theoretical_loss": 3.7647741491366067,
"tokens_seen": 727449600
},
{
"epoch": 0.22,
"learning_rate": 0.00039374122367101305,
"loss": 2.9528,
"theoretical_loss": 3.7647741491366067,
"tokens_seen": 727449600
},
{
"epoch": 0.22,
"learning_rate": 0.00039373119358074223,
"loss": 2.954,
"theoretical_loss": 3.764739948837731,
"tokens_seen": 727515136
},
{
"epoch": 0.22,
"learning_rate": 0.0003937211634904714,
"loss": 2.7636,
"theoretical_loss": 3.764705752482091,
"tokens_seen": 727580672
},
{
"epoch": 0.22,
"learning_rate": 0.0003937111334002006,
"loss": 2.8125,
"theoretical_loss": 3.7646715600688765,
"tokens_seen": 727646208
},
{
"epoch": 0.22,
"learning_rate": 0.00039370110330992983,
"loss": 3.04,
"theoretical_loss": 3.7646373715972787,
"tokens_seen": 727711744
},
{
"epoch": 0.22,
"learning_rate": 0.00039369107321965896,
"loss": 2.7841,
"theoretical_loss": 3.7646031870664878,
"tokens_seen": 727777280
},
{
"epoch": 0.22,
"learning_rate": 0.0003936810431293882,
"loss": 2.8055,
"theoretical_loss": 3.764569006475695,
"tokens_seen": 727842816
},
{
"epoch": 0.22,
"learning_rate": 0.0003936710130391173,
"loss": 2.8315,
"theoretical_loss": 3.7645348298240915,
"tokens_seen": 727908352
},
{
"epoch": 0.22,
"learning_rate": 0.00039366098294884655,
"loss": 2.9572,
"theoretical_loss": 3.7645006571108683,
"tokens_seen": 727973888
},
{
"epoch": 0.22,
"learning_rate": 0.00039365095285857573,
"loss": 2.7404,
"theoretical_loss": 3.7644664883352172,
"tokens_seen": 728039424
},
{
"epoch": 0.22,
"learning_rate": 0.0003936409227683049,
"loss": 2.8005,
"theoretical_loss": 3.764432323496331,
"tokens_seen": 728104960
},
{
"epoch": 0.22,
"learning_rate": 0.0003936308926780341,
"loss": 2.9156,
"theoretical_loss": 3.764398162593401,
"tokens_seen": 728170496
},
{
"epoch": 0.22,
"learning_rate": 0.00039362086258776333,
"loss": 2.8162,
"theoretical_loss": 3.764364005625619,
"tokens_seen": 728236032
},
{
"epoch": 0.22,
"learning_rate": 0.00039361083249749246,
"loss": 2.8836,
"theoretical_loss": 3.764329852592179,
"tokens_seen": 728301568
},
{
"epoch": 0.22,
"learning_rate": 0.0003936008024072217,
"loss": 2.8037,
"theoretical_loss": 3.764295703492273,
"tokens_seen": 728367104
},
{
"epoch": 0.22,
"learning_rate": 0.0003935907723169508,
"loss": 2.7596,
"theoretical_loss": 3.764261558325094,
"tokens_seen": 728432640
},
{
"epoch": 0.22,
"learning_rate": 0.00039358074222668006,
"loss": 2.6471,
"theoretical_loss": 3.764227417089836,
"tokens_seen": 728498176
},
{
"epoch": 0.22,
"learning_rate": 0.00039357071213640924,
"loss": 2.7925,
"theoretical_loss": 3.764193279785692,
"tokens_seen": 728563712
},
{
"epoch": 0.22,
"learning_rate": 0.0003935606820461384,
"loss": 2.7859,
"theoretical_loss": 3.7641591464118562,
"tokens_seen": 728629248
},
{
"epoch": 0.22,
"learning_rate": 0.0003935506519558676,
"loss": 2.9756,
"theoretical_loss": 3.7641250169675224,
"tokens_seen": 728694784
},
{
"epoch": 0.22,
"learning_rate": 0.0003935406218655968,
"loss": 2.8518,
"theoretical_loss": 3.7640908914518847,
"tokens_seen": 728760320
},
{
"epoch": 0.22,
"learning_rate": 0.00039353059177532596,
"loss": 2.7467,
"theoretical_loss": 3.764056769864138,
"tokens_seen": 728825856
},
{
"epoch": 0.22,
"learning_rate": 0.0003935205616850552,
"loss": 2.8301,
"theoretical_loss": 3.7640226522034768,
"tokens_seen": 728891392
},
{
"epoch": 0.22,
"learning_rate": 0.0003935105315947843,
"loss": 2.9041,
"theoretical_loss": 3.7639885384690963,
"tokens_seen": 728956928
},
{
"epoch": 0.22,
"learning_rate": 0.00039350050150451356,
"loss": 2.6885,
"theoretical_loss": 3.763954428660192,
"tokens_seen": 729022464
},
{
"epoch": 0.22,
"objective/train/docs_used": 421294,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8405401706695557,
"objective/train/theoretical_loss": 3.7639203227759586,
"objective/train/tokens_used": 749548000,
"theoretical_loss": 3.7639203227759586,
"tokens_seen": 729088000
},
{
"epoch": 0.22,
"learning_rate": 0.0003934904714142427,
"loss": 2.914,
"theoretical_loss": 3.7639203227759586,
"tokens_seen": 729088000
},
{
"epoch": 0.22,
"learning_rate": 0.0003934804413239719,
"loss": 2.6164,
"theoretical_loss": 3.763886220815592,
"tokens_seen": 729153536
},
{
"epoch": 0.22,
"learning_rate": 0.0003934704112337011,
"loss": 2.7629,
"theoretical_loss": 3.763852122778289,
"tokens_seen": 729219072
},
{
"epoch": 0.22,
"learning_rate": 0.0003934603811434303,
"loss": 2.8913,
"theoretical_loss": 3.7638180286632443,
"tokens_seen": 729284608
},
{
"epoch": 0.22,
"learning_rate": 0.00039345035105315946,
"loss": 2.5813,
"theoretical_loss": 3.763783938469656,
"tokens_seen": 729350144
},
{
"epoch": 0.22,
"learning_rate": 0.0003934403209628887,
"loss": 2.8871,
"theoretical_loss": 3.763749852196719,
"tokens_seen": 729415680
},
{
"epoch": 0.22,
"learning_rate": 0.00039343029087261783,
"loss": 2.7964,
"theoretical_loss": 3.763715769843632,
"tokens_seen": 729481216
},
{
"epoch": 0.22,
"learning_rate": 0.00039342026078234706,
"loss": 2.6867,
"theoretical_loss": 3.763681691409591,
"tokens_seen": 729546752
},
{
"epoch": 0.22,
"learning_rate": 0.0003934102306920762,
"loss": 3.0095,
"theoretical_loss": 3.7636476168937936,
"tokens_seen": 729612288
},
{
"epoch": 0.22,
"learning_rate": 0.0003934002006018054,
"loss": 2.9734,
"theoretical_loss": 3.7636135462954368,
"tokens_seen": 729677824
},
{
"epoch": 0.22,
"learning_rate": 0.00039339017051153466,
"loss": 2.8591,
"theoretical_loss": 3.76357947961372,
"tokens_seen": 729743360
},
{
"epoch": 0.22,
"learning_rate": 0.0003933801404212638,
"loss": 2.854,
"theoretical_loss": 3.76354541684784,
"tokens_seen": 729808896
},
{
"epoch": 0.22,
"learning_rate": 0.000393370110330993,
"loss": 3.0612,
"theoretical_loss": 3.763511357996995,
"tokens_seen": 729874432
},
{
"epoch": 0.22,
"learning_rate": 0.00039336008024072215,
"loss": 2.5706,
"theoretical_loss": 3.7634773030603847,
"tokens_seen": 729939968
},
{
"epoch": 0.22,
"learning_rate": 0.0003933500501504514,
"loss": 2.7152,
"theoretical_loss": 3.7634432520372068,
"tokens_seen": 730005504
},
{
"epoch": 0.22,
"learning_rate": 0.00039334002006018057,
"loss": 2.6651,
"theoretical_loss": 3.7634092049266608,
"tokens_seen": 730071040
},
{
"epoch": 0.22,
"learning_rate": 0.00039332998996990975,
"loss": 2.5543,
"theoretical_loss": 3.763375161727946,
"tokens_seen": 730136576
},
{
"epoch": 0.22,
"learning_rate": 0.00039331995987963893,
"loss": 2.7306,
"theoretical_loss": 3.7633411224402615,
"tokens_seen": 730202112
},
{
"epoch": 0.22,
"learning_rate": 0.00039330992978936816,
"loss": 2.6186,
"theoretical_loss": 3.7633070870628074,
"tokens_seen": 730267648
},
{
"epoch": 0.22,
"learning_rate": 0.0003932998996990973,
"loss": 2.7816,
"theoretical_loss": 3.7632730555947833,
"tokens_seen": 730333184
},
{
"epoch": 0.22,
"learning_rate": 0.0003932898696088265,
"loss": 2.7457,
"theoretical_loss": 3.76323902803539,
"tokens_seen": 730398720
},
{
"epoch": 0.22,
"learning_rate": 0.00039327983951855565,
"loss": 3.0602,
"theoretical_loss": 3.763205004383827,
"tokens_seen": 730464256
},
{
"epoch": 0.22,
"learning_rate": 0.0003932698094282849,
"loss": 2.8779,
"theoretical_loss": 3.7631709846392964,
"tokens_seen": 730529792
},
{
"epoch": 0.22,
"learning_rate": 0.00039325977933801407,
"loss": 2.6766,
"theoretical_loss": 3.763136968800998,
"tokens_seen": 730595328
},
{
"epoch": 0.22,
"learning_rate": 0.00039324974924774325,
"loss": 2.9635,
"theoretical_loss": 3.7631029568681327,
"tokens_seen": 730660864
},
{
"epoch": 0.22,
"objective/train/docs_used": 422072,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6152868270874023,
"objective/train/theoretical_loss": 3.7630689488399027,
"objective/train/tokens_used": 751186400,
"theoretical_loss": 3.7630689488399027,
"tokens_seen": 730726400
},
{
"epoch": 0.22,
"learning_rate": 0.00039323971915747243,
"loss": 2.8188,
"theoretical_loss": 3.7630689488399027,
"tokens_seen": 730726400
},
{
"epoch": 0.22,
"learning_rate": 0.0003932296890672016,
"loss": 2.6186,
"theoretical_loss": 3.763034944715509,
"tokens_seen": 730791936
},
{
"epoch": 0.22,
"learning_rate": 0.0003932196589769308,
"loss": 2.8907,
"theoretical_loss": 3.763000944494154,
"tokens_seen": 730857472
},
{
"epoch": 0.22,
"learning_rate": 0.00039320962888666003,
"loss": 2.6449,
"theoretical_loss": 3.7629669481750394,
"tokens_seen": 730923008
},
{
"epoch": 0.22,
"learning_rate": 0.00039319959879638916,
"loss": 2.7861,
"theoretical_loss": 3.7629329557573676,
"tokens_seen": 730988544
},
{
"epoch": 0.22,
"learning_rate": 0.0003931895687061184,
"loss": 2.8843,
"theoretical_loss": 3.762898967240341,
"tokens_seen": 731054080
},
{
"epoch": 0.22,
"learning_rate": 0.0003931795386158475,
"loss": 2.7126,
"theoretical_loss": 3.762864982623163,
"tokens_seen": 731119616
},
{
"epoch": 0.22,
"learning_rate": 0.00039316950852557675,
"loss": 3.0089,
"theoretical_loss": 3.762831001905036,
"tokens_seen": 731185152
},
{
"epoch": 0.22,
"learning_rate": 0.00039315947843530593,
"loss": 2.85,
"theoretical_loss": 3.7627970250851632,
"tokens_seen": 731250688
},
{
"epoch": 0.22,
"learning_rate": 0.0003931494483450351,
"loss": 3.021,
"theoretical_loss": 3.7627630521627484,
"tokens_seen": 731316224
},
{
"epoch": 0.22,
"learning_rate": 0.0003931394182547643,
"loss": 2.5956,
"theoretical_loss": 3.7627290831369953,
"tokens_seen": 731381760
},
{
"epoch": 0.22,
"learning_rate": 0.00039312938816449353,
"loss": 2.7156,
"theoretical_loss": 3.762695118007107,
"tokens_seen": 731447296
},
{
"epoch": 0.22,
"learning_rate": 0.00039311935807422266,
"loss": 2.6705,
"theoretical_loss": 3.7626611567722894,
"tokens_seen": 731512832
},
{
"epoch": 0.22,
"learning_rate": 0.0003931093279839519,
"loss": 2.6842,
"theoretical_loss": 3.7626271994317455,
"tokens_seen": 731578368
},
{
"epoch": 0.22,
"learning_rate": 0.000393099297893681,
"loss": 2.8427,
"theoretical_loss": 3.7625932459846805,
"tokens_seen": 731643904
},
{
"epoch": 0.22,
"learning_rate": 0.00039308926780341026,
"loss": 2.7812,
"theoretical_loss": 3.7625592964302994,
"tokens_seen": 731709440
},
{
"epoch": 0.22,
"learning_rate": 0.00039307923771313944,
"loss": 2.8303,
"theoretical_loss": 3.762525350767807,
"tokens_seen": 731774976
},
{
"epoch": 0.22,
"learning_rate": 0.0003930692076228686,
"loss": 2.9574,
"theoretical_loss": 3.762491408996409,
"tokens_seen": 731840512
},
{
"epoch": 0.22,
"learning_rate": 0.0003930591775325978,
"loss": 2.8239,
"theoretical_loss": 3.7624574711153107,
"tokens_seen": 731906048
},
{
"epoch": 0.22,
"learning_rate": 0.000393049147442327,
"loss": 3.0284,
"theoretical_loss": 3.7624235371237176,
"tokens_seen": 731971584
},
{
"epoch": 0.22,
"learning_rate": 0.00039303911735205616,
"loss": 2.7515,
"theoretical_loss": 3.7623896070208365,
"tokens_seen": 732037120
},
{
"epoch": 0.22,
"learning_rate": 0.0003930290872617854,
"loss": 2.8283,
"theoretical_loss": 3.7623556808058733,
"tokens_seen": 732102656
},
{
"epoch": 0.22,
"learning_rate": 0.0003930190571715145,
"loss": 3.0297,
"theoretical_loss": 3.762321758478034,
"tokens_seen": 732168192
},
{
"epoch": 0.22,
"learning_rate": 0.00039300902708124376,
"loss": 3.0333,
"theoretical_loss": 3.762287840036527,
"tokens_seen": 732233728
},
{
"epoch": 0.22,
"learning_rate": 0.0003929989969909729,
"loss": 2.6491,
"theoretical_loss": 3.7622539254805574,
"tokens_seen": 732299264
},
{
"epoch": 0.22,
"objective/train/docs_used": 423458,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7574894428253174,
"objective/train/theoretical_loss": 3.7622200148093334,
"objective/train/tokens_used": 752824800,
"theoretical_loss": 3.7622200148093334,
"tokens_seen": 732364800
},
{
"epoch": 0.22,
"learning_rate": 0.0003929889669007021,
"loss": 2.4834,
"theoretical_loss": 3.7622200148093334,
"tokens_seen": 732364800
},
{
"epoch": 0.22,
"learning_rate": 0.0003929789368104313,
"loss": 3.0277,
"theoretical_loss": 3.7621861080220627,
"tokens_seen": 732430336
},
{
"epoch": 0.22,
"learning_rate": 0.0003929689067201605,
"loss": 2.6675,
"theoretical_loss": 3.7621522051179523,
"tokens_seen": 732495872
},
{
"epoch": 0.22,
"learning_rate": 0.00039295887662988967,
"loss": 2.9897,
"theoretical_loss": 3.7621183060962102,
"tokens_seen": 732561408
},
{
"epoch": 0.22,
"learning_rate": 0.0003929488465396189,
"loss": 2.6445,
"theoretical_loss": 3.762084410956045,
"tokens_seen": 732626944
},
{
"epoch": 0.22,
"learning_rate": 0.00039293881644934803,
"loss": 2.6183,
"theoretical_loss": 3.7620505196966647,
"tokens_seen": 732692480
},
{
"epoch": 0.22,
"learning_rate": 0.00039292878635907726,
"loss": 2.6577,
"theoretical_loss": 3.762016632317278,
"tokens_seen": 732758016
},
{
"epoch": 0.22,
"learning_rate": 0.0003929187562688064,
"loss": 3.1477,
"theoretical_loss": 3.761982748817094,
"tokens_seen": 732823552
},
{
"epoch": 0.22,
"learning_rate": 0.0003929087261785356,
"loss": 2.4979,
"theoretical_loss": 3.7619488691953213,
"tokens_seen": 732889088
},
{
"epoch": 0.22,
"learning_rate": 0.0003928986960882648,
"loss": 2.801,
"theoretical_loss": 3.7619149934511693,
"tokens_seen": 732954624
},
{
"epoch": 0.22,
"learning_rate": 0.000392888665997994,
"loss": 2.6168,
"theoretical_loss": 3.761881121583848,
"tokens_seen": 733020160
},
{
"epoch": 0.22,
"learning_rate": 0.00039287863590772317,
"loss": 2.8622,
"theoretical_loss": 3.761847253592567,
"tokens_seen": 733085696
},
{
"epoch": 0.22,
"learning_rate": 0.00039286860581745235,
"loss": 2.9506,
"theoretical_loss": 3.7618133894765364,
"tokens_seen": 733151232
},
{
"epoch": 0.22,
"learning_rate": 0.00039285857572718153,
"loss": 2.9322,
"theoretical_loss": 3.761779529234966,
"tokens_seen": 733216768
},
{
"epoch": 0.22,
"learning_rate": 0.00039284854563691077,
"loss": 2.9263,
"theoretical_loss": 3.761745672867066,
"tokens_seen": 733282304
},
{
"epoch": 0.22,
"learning_rate": 0.0003928385155466399,
"loss": 2.9353,
"theoretical_loss": 3.7617118203720485,
"tokens_seen": 733347840
},
{
"epoch": 0.22,
"learning_rate": 0.00039282848545636913,
"loss": 2.9427,
"theoretical_loss": 3.7616779717491227,
"tokens_seen": 733413376
},
{
"epoch": 0.22,
"learning_rate": 0.00039281845536609825,
"loss": 2.779,
"theoretical_loss": 3.761644126997501,
"tokens_seen": 733478912
},
{
"epoch": 0.22,
"learning_rate": 0.0003928084252758275,
"loss": 2.9026,
"theoretical_loss": 3.7616102861163947,
"tokens_seen": 733544448
},
{
"epoch": 0.22,
"learning_rate": 0.00039279839518555667,
"loss": 3.0365,
"theoretical_loss": 3.7615764491050143,
"tokens_seen": 733609984
},
{
"epoch": 0.22,
"learning_rate": 0.00039278836509528585,
"loss": 2.9241,
"theoretical_loss": 3.761542615962573,
"tokens_seen": 733675520
},
{
"epoch": 0.22,
"learning_rate": 0.00039277833500501503,
"loss": 2.9019,
"theoretical_loss": 3.7615087866882826,
"tokens_seen": 733741056
},
{
"epoch": 0.22,
"learning_rate": 0.00039276830491474427,
"loss": 2.6419,
"theoretical_loss": 3.7614749612813547,
"tokens_seen": 733806592
},
{
"epoch": 0.22,
"learning_rate": 0.0003927582748244734,
"loss": 2.7379,
"theoretical_loss": 3.761441139741003,
"tokens_seen": 733872128
},
{
"epoch": 0.22,
"learning_rate": 0.00039274824473420263,
"loss": 2.8309,
"theoretical_loss": 3.7614073220664395,
"tokens_seen": 733937664
},
{
"epoch": 0.22,
"objective/train/docs_used": 424116,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.785700798034668,
"objective/train/theoretical_loss": 3.7613735082568764,
"objective/train/tokens_used": 754463200,
"theoretical_loss": 3.7613735082568764,
"tokens_seen": 734003200
},
{
"epoch": 0.22,
"learning_rate": 0.00039273821464393176,
"loss": 2.6518,
"theoretical_loss": 3.7613735082568764,
"tokens_seen": 734003200
},
{
"epoch": 0.22,
"learning_rate": 0.000392728184553661,
"loss": 3.0309,
"theoretical_loss": 3.7613396983115286,
"tokens_seen": 734068736
},
{
"epoch": 0.22,
"learning_rate": 0.0003927181544633902,
"loss": 2.6054,
"theoretical_loss": 3.761305892229609,
"tokens_seen": 734134272
},
{
"epoch": 0.22,
"learning_rate": 0.00039270812437311936,
"loss": 2.7434,
"theoretical_loss": 3.7612720900103307,
"tokens_seen": 734199808
},
{
"epoch": 0.22,
"learning_rate": 0.00039269809428284854,
"loss": 2.7315,
"theoretical_loss": 3.7612382916529086,
"tokens_seen": 734265344
},
{
"epoch": 0.22,
"learning_rate": 0.0003926880641925777,
"loss": 2.8032,
"theoretical_loss": 3.7612044971565557,
"tokens_seen": 734330880
},
{
"epoch": 0.22,
"learning_rate": 0.0003926780341023069,
"loss": 2.5744,
"theoretical_loss": 3.7611707065204874,
"tokens_seen": 734396416
},
{
"epoch": 0.22,
"learning_rate": 0.00039266800401203613,
"loss": 2.5115,
"theoretical_loss": 3.761136919743918,
"tokens_seen": 734461952
},
{
"epoch": 0.22,
"learning_rate": 0.00039265797392176526,
"loss": 2.7293,
"theoretical_loss": 3.761103136826062,
"tokens_seen": 734527488
},
{
"epoch": 0.22,
"learning_rate": 0.0003926479438314945,
"loss": 2.6237,
"theoretical_loss": 3.761069357766135,
"tokens_seen": 734593024
},
{
"epoch": 0.22,
"learning_rate": 0.00039263791374122373,
"loss": 2.9789,
"theoretical_loss": 3.761035582563352,
"tokens_seen": 734658560
},
{
"epoch": 0.22,
"learning_rate": 0.00039262788365095286,
"loss": 2.7604,
"theoretical_loss": 3.7610018112169286,
"tokens_seen": 734724096
},
{
"epoch": 0.22,
"learning_rate": 0.0003926178535606821,
"loss": 2.7513,
"theoretical_loss": 3.760968043726081,
"tokens_seen": 734789632
},
{
"epoch": 0.22,
"learning_rate": 0.0003926078234704112,
"loss": 2.8022,
"theoretical_loss": 3.7609342800900247,
"tokens_seen": 734855168
},
{
"epoch": 0.22,
"learning_rate": 0.00039259779338014046,
"loss": 2.5513,
"theoretical_loss": 3.7609005203079757,
"tokens_seen": 734920704
},
{
"epoch": 0.22,
"learning_rate": 0.00039258776328986964,
"loss": 2.7556,
"theoretical_loss": 3.760866764379151,
"tokens_seen": 734986240
},
{
"epoch": 0.22,
"learning_rate": 0.0003925777331995988,
"loss": 2.7191,
"theoretical_loss": 3.7608330123027667,
"tokens_seen": 735051776
},
{
"epoch": 0.22,
"learning_rate": 0.000392567703109328,
"loss": 2.8348,
"theoretical_loss": 3.7607992640780408,
"tokens_seen": 735117312
},
{
"epoch": 0.22,
"learning_rate": 0.0003925576730190572,
"loss": 2.7799,
"theoretical_loss": 3.760765519704189,
"tokens_seen": 735182848
},
{
"epoch": 0.22,
"learning_rate": 0.00039254764292878636,
"loss": 2.7014,
"theoretical_loss": 3.7607317791804302,
"tokens_seen": 735248384
},
{
"epoch": 0.22,
"learning_rate": 0.0003925376128385156,
"loss": 2.6922,
"theoretical_loss": 3.7606980425059806,
"tokens_seen": 735313920
},
{
"epoch": 0.22,
"learning_rate": 0.0003925275827482447,
"loss": 2.8887,
"theoretical_loss": 3.7606643096800587,
"tokens_seen": 735379456
},
{
"epoch": 0.22,
"learning_rate": 0.00039251755265797396,
"loss": 2.8958,
"theoretical_loss": 3.760630580701883,
"tokens_seen": 735444992
},
{
"epoch": 0.22,
"learning_rate": 0.0003925075225677031,
"loss": 2.5467,
"theoretical_loss": 3.760596855570671,
"tokens_seen": 735510528
},
{
"epoch": 0.22,
"learning_rate": 0.0003924974924774323,
"loss": 2.7178,
"theoretical_loss": 3.7605631342856416,
"tokens_seen": 735576064
},
{
"epoch": 0.22,
"objective/train/docs_used": 424834,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7061710357666016,
"objective/train/theoretical_loss": 3.7605294168460133,
"objective/train/tokens_used": 756101600,
"theoretical_loss": 3.7605294168460133,
"tokens_seen": 735641600
},
{
"epoch": 0.22,
"learning_rate": 0.0003924874623871615,
"loss": 2.9599,
"theoretical_loss": 3.7605294168460133,
"tokens_seen": 735641600
},
{
"epoch": 0.22,
"learning_rate": 0.0003924774322968907,
"loss": 2.9143,
"theoretical_loss": 3.7604957032510056,
"tokens_seen": 735707136
},
{
"epoch": 0.22,
"learning_rate": 0.00039246740220661987,
"loss": 2.8619,
"theoretical_loss": 3.760461993499837,
"tokens_seen": 735772672
},
{
"epoch": 0.22,
"learning_rate": 0.0003924573721163491,
"loss": 2.8088,
"theoretical_loss": 3.7604282875917274,
"tokens_seen": 735838208
},
{
"epoch": 0.22,
"learning_rate": 0.00039244734202607823,
"loss": 3.14,
"theoretical_loss": 3.760394585525896,
"tokens_seen": 735903744
},
{
"epoch": 0.22,
"learning_rate": 0.00039243731193580746,
"loss": 2.6493,
"theoretical_loss": 3.7603608873015633,
"tokens_seen": 735969280
},
{
"epoch": 0.22,
"learning_rate": 0.0003924272818455366,
"loss": 2.7973,
"theoretical_loss": 3.7603271929179494,
"tokens_seen": 736034816
},
{
"epoch": 0.22,
"learning_rate": 0.0003924172517552658,
"loss": 2.8388,
"theoretical_loss": 3.7602935023742736,
"tokens_seen": 736100352
},
{
"epoch": 0.22,
"learning_rate": 0.000392407221664995,
"loss": 2.7299,
"theoretical_loss": 3.760259815669758,
"tokens_seen": 736165888
},
{
"epoch": 0.22,
"learning_rate": 0.0003923971915747242,
"loss": 2.7601,
"theoretical_loss": 3.7602261328036226,
"tokens_seen": 736231424
},
{
"epoch": 0.22,
"learning_rate": 0.00039238716148445337,
"loss": 2.7582,
"theoretical_loss": 3.760192453775088,
"tokens_seen": 736296960
},
{
"epoch": 0.22,
"learning_rate": 0.00039237713139418255,
"loss": 2.8659,
"theoretical_loss": 3.760158778583376,
"tokens_seen": 736362496
},
{
"epoch": 0.22,
"learning_rate": 0.00039236710130391173,
"loss": 2.9821,
"theoretical_loss": 3.7601251072277084,
"tokens_seen": 736428032
},
{
"epoch": 0.22,
"learning_rate": 0.00039235707121364097,
"loss": 2.8787,
"theoretical_loss": 3.760091439707306,
"tokens_seen": 736493568
},
{
"epoch": 0.22,
"learning_rate": 0.0003923470411233701,
"loss": 2.816,
"theoretical_loss": 3.760057776021392,
"tokens_seen": 736559104
},
{
"epoch": 0.22,
"learning_rate": 0.00039233701103309933,
"loss": 2.794,
"theoretical_loss": 3.7600241161691876,
"tokens_seen": 736624640
},
{
"epoch": 0.22,
"learning_rate": 0.00039232698094282846,
"loss": 2.8768,
"theoretical_loss": 3.759990460149915,
"tokens_seen": 736690176
},
{
"epoch": 0.22,
"learning_rate": 0.0003923169508525577,
"loss": 2.7201,
"theoretical_loss": 3.759956807962797,
"tokens_seen": 736755712
},
{
"epoch": 0.22,
"learning_rate": 0.00039230692076228687,
"loss": 3.1597,
"theoretical_loss": 3.759923159607057,
"tokens_seen": 736821248
},
{
"epoch": 0.22,
"learning_rate": 0.00039229689067201605,
"loss": 2.8136,
"theoretical_loss": 3.759889515081918,
"tokens_seen": 736886784
},
{
"epoch": 0.22,
"learning_rate": 0.00039228686058174523,
"loss": 2.8847,
"theoretical_loss": 3.759855874386602,
"tokens_seen": 736952320
},
{
"epoch": 0.22,
"learning_rate": 0.00039227683049147447,
"loss": 2.707,
"theoretical_loss": 3.7598222375203347,
"tokens_seen": 737017856
},
{
"epoch": 0.22,
"learning_rate": 0.0003922668004012036,
"loss": 2.6899,
"theoretical_loss": 3.7597886044823383,
"tokens_seen": 737083392
},
{
"epoch": 0.22,
"learning_rate": 0.00039225677031093283,
"loss": 2.9802,
"theoretical_loss": 3.7597549752718367,
"tokens_seen": 737148928
},
{
"epoch": 0.22,
"learning_rate": 0.00039224674022066196,
"loss": 2.7919,
"theoretical_loss": 3.7597213498880553,
"tokens_seen": 737214464
},
{
"epoch": 0.22,
"objective/train/docs_used": 426204,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4200565814971924,
"objective/train/theoretical_loss": 3.759687728330217,
"objective/train/tokens_used": 757740000,
"theoretical_loss": 3.759687728330217,
"tokens_seen": 737280000
},
{
"epoch": 0.22,
"learning_rate": 0.0003922367101303912,
"loss": 2.6965,
"theoretical_loss": 3.759687728330217,
"tokens_seen": 737280000
},
{
"epoch": 0.22,
"learning_rate": 0.0003922266800401204,
"loss": 2.8828,
"theoretical_loss": 3.759654110597548,
"tokens_seen": 737345536
},
{
"epoch": 0.22,
"learning_rate": 0.00039221664994984956,
"loss": 2.8852,
"theoretical_loss": 3.759620496689272,
"tokens_seen": 737411072
},
{
"epoch": 0.22,
"learning_rate": 0.00039220661985957874,
"loss": 2.7873,
"theoretical_loss": 3.7595868866046143,
"tokens_seen": 737476608
},
{
"epoch": 0.22,
"learning_rate": 0.0003921965897693079,
"loss": 2.9692,
"theoretical_loss": 3.759553280342801,
"tokens_seen": 737542144
},
{
"epoch": 0.22,
"learning_rate": 0.0003921865596790371,
"loss": 2.5655,
"theoretical_loss": 3.7595196779030564,
"tokens_seen": 737607680
},
{
"epoch": 0.22,
"learning_rate": 0.00039217652958876633,
"loss": 2.8685,
"theoretical_loss": 3.759486079284607,
"tokens_seen": 737673216
},
{
"epoch": 0.22,
"learning_rate": 0.00039216649949849546,
"loss": 2.8385,
"theoretical_loss": 3.7594524844866797,
"tokens_seen": 737738752
},
{
"epoch": 0.22,
"learning_rate": 0.0003921564694082247,
"loss": 3.0792,
"theoretical_loss": 3.7594188935084993,
"tokens_seen": 737804288
},
{
"epoch": 0.22,
"learning_rate": 0.0003921464393179539,
"loss": 2.9041,
"theoretical_loss": 3.7593853063492926,
"tokens_seen": 737869824
},
{
"epoch": 0.22,
"learning_rate": 0.00039213640922768306,
"loss": 2.6946,
"theoretical_loss": 3.7593517230082867,
"tokens_seen": 737935360
},
{
"epoch": 0.22,
"learning_rate": 0.00039212637913741224,
"loss": 2.7761,
"theoretical_loss": 3.759318143484708,
"tokens_seen": 738000896
},
{
"epoch": 0.22,
"learning_rate": 0.0003921163490471414,
"loss": 2.3848,
"theoretical_loss": 3.759284567777784,
"tokens_seen": 738066432
},
{
"epoch": 0.22,
"learning_rate": 0.0003921063189568706,
"loss": 2.7913,
"theoretical_loss": 3.7592509958867426,
"tokens_seen": 738131968
},
{
"epoch": 0.22,
"learning_rate": 0.00039209628886659984,
"loss": 2.7902,
"theoretical_loss": 3.75921742781081,
"tokens_seen": 738197504
},
{
"epoch": 0.22,
"learning_rate": 0.00039208625877632896,
"loss": 2.8065,
"theoretical_loss": 3.759183863549215,
"tokens_seen": 738263040
},
{
"epoch": 0.22,
"learning_rate": 0.0003920762286860582,
"loss": 2.9661,
"theoretical_loss": 3.759150303101185,
"tokens_seen": 738328576
},
{
"epoch": 0.22,
"learning_rate": 0.0003920661985957873,
"loss": 2.8999,
"theoretical_loss": 3.759116746465949,
"tokens_seen": 738394112
},
{
"epoch": 0.22,
"learning_rate": 0.00039205616850551656,
"loss": 2.6292,
"theoretical_loss": 3.7590831936427356,
"tokens_seen": 738459648
},
{
"epoch": 0.22,
"learning_rate": 0.00039204613841524574,
"loss": 2.6481,
"theoretical_loss": 3.759049644630773,
"tokens_seen": 738525184
},
{
"epoch": 0.22,
"learning_rate": 0.0003920361083249749,
"loss": 2.7701,
"theoretical_loss": 3.7590160994292896,
"tokens_seen": 738590720
},
{
"epoch": 0.22,
"learning_rate": 0.0003920260782347041,
"loss": 2.8019,
"theoretical_loss": 3.7589825580375154,
"tokens_seen": 738656256
},
{
"epoch": 0.22,
"learning_rate": 0.0003920160481444333,
"loss": 2.8663,
"theoretical_loss": 3.758949020454679,
"tokens_seen": 738721792
},
{
"epoch": 0.22,
"learning_rate": 0.00039200601805416247,
"loss": 2.8756,
"theoretical_loss": 3.758915486680011,
"tokens_seen": 738787328
},
{
"epoch": 0.22,
"learning_rate": 0.0003919959879638917,
"loss": 2.553,
"theoretical_loss": 3.7588819567127407,
"tokens_seen": 738852864
},
{
"epoch": 0.22,
"objective/train/docs_used": 426747,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5556228160858154,
"objective/train/theoretical_loss": 3.7588484305520984,
"objective/train/tokens_used": 759378400,
"theoretical_loss": 3.7588484305520984,
"tokens_seen": 738918400
},
{
"epoch": 0.22,
"learning_rate": 0.00039198595787362083,
"loss": 2.7615,
"theoretical_loss": 3.7588484305520984,
"tokens_seen": 738918400
},
{
"epoch": 0.22,
"learning_rate": 0.00039197592778335007,
"loss": 2.7367,
"theoretical_loss": 3.7588149081973143,
"tokens_seen": 738983936
},
{
"epoch": 0.22,
"learning_rate": 0.00039196589769307925,
"loss": 2.7388,
"theoretical_loss": 3.7587813896476185,
"tokens_seen": 739049472
},
{
"epoch": 0.22,
"learning_rate": 0.00039195586760280843,
"loss": 2.7414,
"theoretical_loss": 3.758747874902242,
"tokens_seen": 739115008
},
{
"epoch": 0.22,
"learning_rate": 0.0003919458375125376,
"loss": 2.4166,
"theoretical_loss": 3.758714363960416,
"tokens_seen": 739180544
},
{
"epoch": 0.22,
"learning_rate": 0.0003919358074222668,
"loss": 2.7885,
"theoretical_loss": 3.758680856821371,
"tokens_seen": 739246080
},
{
"epoch": 0.22,
"learning_rate": 0.00039192577733199597,
"loss": 2.9123,
"theoretical_loss": 3.7586473534843394,
"tokens_seen": 739311616
},
{
"epoch": 0.22,
"learning_rate": 0.0003919157472417252,
"loss": 2.7433,
"theoretical_loss": 3.7586138539485514,
"tokens_seen": 739377152
},
{
"epoch": 0.22,
"learning_rate": 0.00039190571715145433,
"loss": 2.787,
"theoretical_loss": 3.7585803582132407,
"tokens_seen": 739442688
},
{
"epoch": 0.22,
"learning_rate": 0.00039189568706118357,
"loss": 2.7486,
"theoretical_loss": 3.7585468662776376,
"tokens_seen": 739508224
},
{
"epoch": 0.22,
"learning_rate": 0.00039188565697091275,
"loss": 2.9068,
"theoretical_loss": 3.758513378140975,
"tokens_seen": 739573760
},
{
"epoch": 0.22,
"learning_rate": 0.00039187562688064193,
"loss": 2.5403,
"theoretical_loss": 3.7584798938024866,
"tokens_seen": 739639296
},
{
"epoch": 0.22,
"learning_rate": 0.00039186559679037117,
"loss": 2.8348,
"theoretical_loss": 3.7584464132614035,
"tokens_seen": 739704832
},
{
"epoch": 0.22,
"learning_rate": 0.0003918555667001003,
"loss": 2.9462,
"theoretical_loss": 3.758412936516959,
"tokens_seen": 739770368
},
{
"epoch": 0.22,
"learning_rate": 0.00039184553660982953,
"loss": 2.7113,
"theoretical_loss": 3.758379463568387,
"tokens_seen": 739835904
},
{
"epoch": 0.22,
"learning_rate": 0.00039183550651955866,
"loss": 2.7578,
"theoretical_loss": 3.7583459944149205,
"tokens_seen": 739901440
},
{
"epoch": 0.22,
"learning_rate": 0.0003918254764292879,
"loss": 3.0413,
"theoretical_loss": 3.7583125290557926,
"tokens_seen": 739966976
},
{
"epoch": 0.22,
"learning_rate": 0.00039181544633901707,
"loss": 2.8557,
"theoretical_loss": 3.7582790674902378,
"tokens_seen": 740032512
},
{
"epoch": 0.22,
"learning_rate": 0.00039180541624874625,
"loss": 2.9974,
"theoretical_loss": 3.7582456097174903,
"tokens_seen": 740098048
},
{
"epoch": 0.22,
"learning_rate": 0.00039179538615847543,
"loss": 2.8235,
"theoretical_loss": 3.758212155736784,
"tokens_seen": 740163584
},
{
"epoch": 0.22,
"learning_rate": 0.00039178535606820467,
"loss": 2.8946,
"theoretical_loss": 3.758178705547353,
"tokens_seen": 740229120
},
{
"epoch": 0.22,
"learning_rate": 0.0003917753259779338,
"loss": 2.9569,
"theoretical_loss": 3.7581452591484323,
"tokens_seen": 740294656
},
{
"epoch": 0.22,
"learning_rate": 0.00039176529588766303,
"loss": 2.8835,
"theoretical_loss": 3.758111816539258,
"tokens_seen": 740360192
},
{
"epoch": 0.22,
"learning_rate": 0.00039175526579739216,
"loss": 2.8443,
"theoretical_loss": 3.7580783777190643,
"tokens_seen": 740425728
},
{
"epoch": 0.22,
"learning_rate": 0.0003917452357071214,
"loss": 2.7275,
"theoretical_loss": 3.7580449426870866,
"tokens_seen": 740491264
},
{
"epoch": 0.22,
"objective/train/docs_used": 427723,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.891180992126465,
"objective/train/theoretical_loss": 3.75801151144256,
"objective/train/tokens_used": 761016800,
"theoretical_loss": 3.75801151144256,
"tokens_seen": 740556800
},
{
"epoch": 0.22,
"learning_rate": 0.0003917352056168506,
"loss": 2.9421,
"theoretical_loss": 3.75801151144256,
"tokens_seen": 740556800
},
{
"epoch": 0.22,
"learning_rate": 0.00039172517552657976,
"loss": 3.0342,
"theoretical_loss": 3.757978083984722,
"tokens_seen": 740622336
},
{
"epoch": 0.22,
"learning_rate": 0.00039171514543630894,
"loss": 2.9514,
"theoretical_loss": 3.7579446603128064,
"tokens_seen": 740687872
},
{
"epoch": 0.22,
"learning_rate": 0.0003917051153460381,
"loss": 2.6242,
"theoretical_loss": 3.7579112404260515,
"tokens_seen": 740753408
},
{
"epoch": 0.22,
"learning_rate": 0.0003916950852557673,
"loss": 3.0941,
"theoretical_loss": 3.7578778243236934,
"tokens_seen": 740818944
},
{
"epoch": 0.22,
"learning_rate": 0.00039168505516549653,
"loss": 2.6706,
"theoretical_loss": 3.757844412004968,
"tokens_seen": 740884480
},
{
"epoch": 0.22,
"learning_rate": 0.00039167502507522566,
"loss": 2.6818,
"theoretical_loss": 3.7578110034691123,
"tokens_seen": 740950016
},
{
"epoch": 0.22,
"learning_rate": 0.0003916649949849549,
"loss": 2.778,
"theoretical_loss": 3.7577775987153648,
"tokens_seen": 741015552
},
{
"epoch": 0.22,
"learning_rate": 0.0003916549648946841,
"loss": 2.776,
"theoretical_loss": 3.7577441977429613,
"tokens_seen": 741081088
},
{
"epoch": 0.22,
"learning_rate": 0.00039164493480441326,
"loss": 2.7072,
"theoretical_loss": 3.7577108005511404,
"tokens_seen": 741146624
},
{
"epoch": 0.22,
"learning_rate": 0.00039163490471414244,
"loss": 2.9878,
"theoretical_loss": 3.7576774071391403,
"tokens_seen": 741212160
},
{
"epoch": 0.22,
"learning_rate": 0.0003916248746238716,
"loss": 2.9135,
"theoretical_loss": 3.7576440175061974,
"tokens_seen": 741277696
},
{
"epoch": 0.22,
"learning_rate": 0.0003916148445336008,
"loss": 2.837,
"theoretical_loss": 3.757610631651552,
"tokens_seen": 741343232
},
{
"epoch": 0.22,
"learning_rate": 0.00039160481444333004,
"loss": 2.7899,
"theoretical_loss": 3.757577249574441,
"tokens_seen": 741408768
},
{
"epoch": 0.22,
"learning_rate": 0.00039159478435305916,
"loss": 3.1061,
"theoretical_loss": 3.7575438712741036,
"tokens_seen": 741474304
},
{
"epoch": 0.22,
"learning_rate": 0.0003915847542627884,
"loss": 2.6674,
"theoretical_loss": 3.75751049674978,
"tokens_seen": 741539840
},
{
"epoch": 0.22,
"learning_rate": 0.0003915747241725175,
"loss": 2.867,
"theoretical_loss": 3.7574771260007074,
"tokens_seen": 741605376
},
{
"epoch": 0.22,
"learning_rate": 0.00039156469408224676,
"loss": 2.9263,
"theoretical_loss": 3.7574437590261263,
"tokens_seen": 741670912
},
{
"epoch": 0.22,
"learning_rate": 0.00039155466399197594,
"loss": 2.8301,
"theoretical_loss": 3.757410395825276,
"tokens_seen": 741736448
},
{
"epoch": 0.22,
"learning_rate": 0.0003915446339017051,
"loss": 2.9813,
"theoretical_loss": 3.757377036397396,
"tokens_seen": 741801984
},
{
"epoch": 0.22,
"learning_rate": 0.0003915346038114343,
"loss": 2.7947,
"theoretical_loss": 3.7573436807417275,
"tokens_seen": 741867520
},
{
"epoch": 0.22,
"learning_rate": 0.0003915245737211635,
"loss": 2.9957,
"theoretical_loss": 3.7573103288575096,
"tokens_seen": 741933056
},
{
"epoch": 0.22,
"learning_rate": 0.00039151454363089267,
"loss": 2.8792,
"theoretical_loss": 3.757276980743983,
"tokens_seen": 741998592
},
{
"epoch": 0.22,
"learning_rate": 0.0003915045135406219,
"loss": 2.7323,
"theoretical_loss": 3.757243636400389,
"tokens_seen": 742064128
},
{
"epoch": 0.22,
"learning_rate": 0.00039149448345035103,
"loss": 3.0987,
"theoretical_loss": 3.757210295825968,
"tokens_seen": 742129664
},
{
"epoch": 0.22,
"objective/train/docs_used": 429091,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3885855674743652,
"objective/train/theoretical_loss": 3.7571769590199615,
"objective/train/tokens_used": 762655200,
"theoretical_loss": 3.7571769590199615,
"tokens_seen": 742195200
},
{
"epoch": 0.22,
"learning_rate": 0.00039148445336008027,
"loss": 2.5697,
"theoretical_loss": 3.7571769590199615,
"tokens_seen": 742195200
},
{
"epoch": 0.22,
"learning_rate": 0.00039147442326980945,
"loss": 2.8096,
"theoretical_loss": 3.7571436259816107,
"tokens_seen": 742260736
},
{
"epoch": 0.22,
"learning_rate": 0.00039146439317953863,
"loss": 2.7815,
"theoretical_loss": 3.757110296710157,
"tokens_seen": 742326272
},
{
"epoch": 0.22,
"learning_rate": 0.0003914543630892678,
"loss": 2.9203,
"theoretical_loss": 3.7570769712048424,
"tokens_seen": 742391808
},
{
"epoch": 0.22,
"learning_rate": 0.000391444332998997,
"loss": 2.9053,
"theoretical_loss": 3.757043649464909,
"tokens_seen": 742457344
},
{
"epoch": 0.23,
"learning_rate": 0.00039143430290872617,
"loss": 2.8275,
"theoretical_loss": 3.7570103314895986,
"tokens_seen": 742522880
},
{
"epoch": 0.23,
"learning_rate": 0.0003914242728184554,
"loss": 2.7974,
"theoretical_loss": 3.756977017278154,
"tokens_seen": 742588416
},
{
"epoch": 0.23,
"learning_rate": 0.00039141424272818453,
"loss": 2.8394,
"theoretical_loss": 3.7569437068298184,
"tokens_seen": 742653952
},
{
"epoch": 0.23,
"learning_rate": 0.00039140421263791377,
"loss": 2.798,
"theoretical_loss": 3.756910400143833,
"tokens_seen": 742719488
},
{
"epoch": 0.23,
"learning_rate": 0.0003913941825476429,
"loss": 3.0578,
"theoretical_loss": 3.7568770972194434,
"tokens_seen": 742785024
},
{
"epoch": 0.23,
"learning_rate": 0.00039138415245737213,
"loss": 2.8774,
"theoretical_loss": 3.7568437980558915,
"tokens_seen": 742850560
},
{
"epoch": 0.23,
"learning_rate": 0.0003913741223671013,
"loss": 3.0182,
"theoretical_loss": 3.756810502652421,
"tokens_seen": 742916096
},
{
"epoch": 0.23,
"learning_rate": 0.0003913640922768305,
"loss": 2.9584,
"theoretical_loss": 3.7567772110082753,
"tokens_seen": 742981632
},
{
"epoch": 0.23,
"learning_rate": 0.0003913540621865597,
"loss": 2.8377,
"theoretical_loss": 3.756743923122699,
"tokens_seen": 743047168
},
{
"epoch": 0.23,
"learning_rate": 0.00039134403209628886,
"loss": 2.915,
"theoretical_loss": 3.756710638994936,
"tokens_seen": 743112704
},
{
"epoch": 0.23,
"learning_rate": 0.00039133400200601804,
"loss": 2.9824,
"theoretical_loss": 3.756677358624231,
"tokens_seen": 743178240
},
{
"epoch": 0.23,
"learning_rate": 0.00039132397191574727,
"loss": 3.047,
"theoretical_loss": 3.7566440820098284,
"tokens_seen": 743243776
},
{
"epoch": 0.23,
"learning_rate": 0.0003913139418254764,
"loss": 2.9693,
"theoretical_loss": 3.756610809150973,
"tokens_seen": 743309312
},
{
"epoch": 0.23,
"learning_rate": 0.00039130391173520563,
"loss": 3.0048,
"theoretical_loss": 3.7565775400469104,
"tokens_seen": 743374848
},
{
"epoch": 0.23,
"learning_rate": 0.0003912938816449348,
"loss": 2.7205,
"theoretical_loss": 3.7565442746968856,
"tokens_seen": 743440384
},
{
"epoch": 0.23,
"learning_rate": 0.000391283851554664,
"loss": 2.8748,
"theoretical_loss": 3.7565110131001442,
"tokens_seen": 743505920
},
{
"epoch": 0.23,
"learning_rate": 0.0003912738214643932,
"loss": 2.8347,
"theoretical_loss": 3.756477755255932,
"tokens_seen": 743571456
},
{
"epoch": 0.23,
"learning_rate": 0.00039126379137412236,
"loss": 2.5357,
"theoretical_loss": 3.7564445011634944,
"tokens_seen": 743636992
},
{
"epoch": 0.23,
"learning_rate": 0.00039125376128385154,
"loss": 3.0181,
"theoretical_loss": 3.7564112508220777,
"tokens_seen": 743702528
},
{
"epoch": 0.23,
"learning_rate": 0.0003912437311935808,
"loss": 2.9297,
"theoretical_loss": 3.756378004230929,
"tokens_seen": 743768064
},
{
"epoch": 0.23,
"objective/train/docs_used": 429832,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0402815341949463,
"objective/train/theoretical_loss": 3.756344761389295,
"objective/train/tokens_used": 764293600,
"theoretical_loss": 3.756344761389295,
"tokens_seen": 743833600
},
{
"epoch": 0.23,
"learning_rate": 0.0003912337011033099,
"loss": 2.8495,
"theoretical_loss": 3.756344761389295,
"tokens_seen": 743833600
},
{
"epoch": 0.23,
"learning_rate": 0.00039122367101303914,
"loss": 2.9931,
"theoretical_loss": 3.7563115222964214,
"tokens_seen": 743899136
},
{
"epoch": 0.23,
"learning_rate": 0.00039121364092276826,
"loss": 2.8322,
"theoretical_loss": 3.7562782869515567,
"tokens_seen": 743964672
},
{
"epoch": 0.23,
"learning_rate": 0.0003912036108324975,
"loss": 2.8582,
"theoretical_loss": 3.756245055353947,
"tokens_seen": 744030208
},
{
"epoch": 0.23,
"learning_rate": 0.0003911935807422267,
"loss": 2.7416,
"theoretical_loss": 3.75621182750284,
"tokens_seen": 744095744
},
{
"epoch": 0.23,
"learning_rate": 0.00039118355065195586,
"loss": 2.5157,
"theoretical_loss": 3.7561786033974833,
"tokens_seen": 744161280
},
{
"epoch": 0.23,
"learning_rate": 0.00039117352056168504,
"loss": 2.9504,
"theoretical_loss": 3.7561453830371256,
"tokens_seen": 744226816
},
{
"epoch": 0.23,
"learning_rate": 0.0003911634904714143,
"loss": 2.7165,
"theoretical_loss": 3.7561121664210146,
"tokens_seen": 744292352
},
{
"epoch": 0.23,
"learning_rate": 0.0003911534603811434,
"loss": 2.7861,
"theoretical_loss": 3.756078953548398,
"tokens_seen": 744357888
},
{
"epoch": 0.23,
"learning_rate": 0.00039114343029087264,
"loss": 2.9328,
"theoretical_loss": 3.756045744418526,
"tokens_seen": 744423424
},
{
"epoch": 0.23,
"learning_rate": 0.0003911334002006018,
"loss": 2.7074,
"theoretical_loss": 3.7560125390306456,
"tokens_seen": 744488960
},
{
"epoch": 0.23,
"learning_rate": 0.000391123370110331,
"loss": 2.7314,
"theoretical_loss": 3.755979337384007,
"tokens_seen": 744554496
},
{
"epoch": 0.23,
"learning_rate": 0.00039111334002006024,
"loss": 2.8234,
"theoretical_loss": 3.7559461394778584,
"tokens_seen": 744620032
},
{
"epoch": 0.23,
"learning_rate": 0.00039110330992978936,
"loss": 2.9033,
"theoretical_loss": 3.755912945311451,
"tokens_seen": 744685568
},
{
"epoch": 0.23,
"learning_rate": 0.0003910932798395186,
"loss": 2.9737,
"theoretical_loss": 3.755879754884032,
"tokens_seen": 744751104
},
{
"epoch": 0.23,
"learning_rate": 0.00039108324974924773,
"loss": 2.6771,
"theoretical_loss": 3.755846568194853,
"tokens_seen": 744816640
},
{
"epoch": 0.23,
"learning_rate": 0.00039107321965897696,
"loss": 2.7807,
"theoretical_loss": 3.755813385243164,
"tokens_seen": 744882176
},
{
"epoch": 0.23,
"learning_rate": 0.00039106318956870614,
"loss": 2.8837,
"theoretical_loss": 3.7557802060282146,
"tokens_seen": 744947712
},
{
"epoch": 0.23,
"learning_rate": 0.0003910531594784353,
"loss": 2.684,
"theoretical_loss": 3.755747030549256,
"tokens_seen": 745013248
},
{
"epoch": 0.23,
"learning_rate": 0.0003910431293881645,
"loss": 2.754,
"theoretical_loss": 3.7557138588055388,
"tokens_seen": 745078784
},
{
"epoch": 0.23,
"learning_rate": 0.0003910330992978937,
"loss": 2.9002,
"theoretical_loss": 3.7556806907963134,
"tokens_seen": 745144320
},
{
"epoch": 0.23,
"learning_rate": 0.00039102306920762287,
"loss": 2.8028,
"theoretical_loss": 3.7556475265208316,
"tokens_seen": 745209856
},
{
"epoch": 0.23,
"learning_rate": 0.0003910130391173521,
"loss": 2.6639,
"theoretical_loss": 3.7556143659783445,
"tokens_seen": 745275392
},
{
"epoch": 0.23,
"learning_rate": 0.00039100300902708123,
"loss": 2.766,
"theoretical_loss": 3.755581209168104,
"tokens_seen": 745340928
},
{
"epoch": 0.23,
"learning_rate": 0.00039099297893681047,
"loss": 2.9254,
"theoretical_loss": 3.7555480560893617,
"tokens_seen": 745406464
},
{
"epoch": 0.23,
"objective/train/docs_used": 430489,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6736252307891846,
"objective/train/theoretical_loss": 3.7555149067413693,
"objective/train/tokens_used": 765932000,
"theoretical_loss": 3.7555149067413693,
"tokens_seen": 745472000
},
{
"epoch": 0.23,
"learning_rate": 0.00039098294884653965,
"loss": 2.8444,
"theoretical_loss": 3.7555149067413693,
"tokens_seen": 745472000
},
{
"epoch": 0.23,
"learning_rate": 0.00039097291875626883,
"loss": 2.5584,
"theoretical_loss": 3.755481761123379,
"tokens_seen": 745537536
},
{
"epoch": 0.23,
"learning_rate": 0.000390962888665998,
"loss": 2.8561,
"theoretical_loss": 3.7554486192346443,
"tokens_seen": 745603072
},
{
"epoch": 0.23,
"learning_rate": 0.0003909528585757272,
"loss": 2.9953,
"theoretical_loss": 3.7554154810744174,
"tokens_seen": 745668608
},
{
"epoch": 0.23,
"learning_rate": 0.00039094282848545637,
"loss": 2.9008,
"theoretical_loss": 3.7553823466419507,
"tokens_seen": 745734144
},
{
"epoch": 0.23,
"learning_rate": 0.0003909327983951856,
"loss": 2.9929,
"theoretical_loss": 3.755349215936498,
"tokens_seen": 745799680
},
{
"epoch": 0.23,
"learning_rate": 0.00039092276830491473,
"loss": 2.9679,
"theoretical_loss": 3.755316088957312,
"tokens_seen": 745865216
},
{
"epoch": 0.23,
"learning_rate": 0.00039091273821464397,
"loss": 3.1088,
"theoretical_loss": 3.755282965703646,
"tokens_seen": 745930752
},
{
"epoch": 0.23,
"learning_rate": 0.0003909027081243731,
"loss": 2.7632,
"theoretical_loss": 3.755249846174755,
"tokens_seen": 745996288
},
{
"epoch": 0.23,
"learning_rate": 0.00039089267803410233,
"loss": 2.9198,
"theoretical_loss": 3.7552167303698925,
"tokens_seen": 746061824
},
{
"epoch": 0.23,
"learning_rate": 0.0003908826479438315,
"loss": 3.2606,
"theoretical_loss": 3.755183618288312,
"tokens_seen": 746127360
},
{
"epoch": 0.23,
"learning_rate": 0.0003908726178535607,
"loss": 2.6917,
"theoretical_loss": 3.755150509929268,
"tokens_seen": 746192896
},
{
"epoch": 0.23,
"learning_rate": 0.0003908625877632899,
"loss": 3.1855,
"theoretical_loss": 3.7551174052920158,
"tokens_seen": 746258432
},
{
"epoch": 0.23,
"learning_rate": 0.00039085255767301906,
"loss": 2.9023,
"theoretical_loss": 3.75508430437581,
"tokens_seen": 746323968
},
{
"epoch": 0.23,
"learning_rate": 0.00039084252758274824,
"loss": 2.5393,
"theoretical_loss": 3.755051207179905,
"tokens_seen": 746389504
},
{
"epoch": 0.23,
"learning_rate": 0.00039083249749247747,
"loss": 2.7435,
"theoretical_loss": 3.7550181137035574,
"tokens_seen": 746455040
},
{
"epoch": 0.23,
"learning_rate": 0.0003908224674022066,
"loss": 2.7961,
"theoretical_loss": 3.7549850239460216,
"tokens_seen": 746520576
},
{
"epoch": 0.23,
"learning_rate": 0.00039081243731193583,
"loss": 2.6037,
"theoretical_loss": 3.7549519379065535,
"tokens_seen": 746586112
},
{
"epoch": 0.23,
"learning_rate": 0.000390802407221665,
"loss": 2.9674,
"theoretical_loss": 3.7549188555844095,
"tokens_seen": 746651648
},
{
"epoch": 0.23,
"learning_rate": 0.0003907923771313942,
"loss": 2.6577,
"theoretical_loss": 3.7548857769788446,
"tokens_seen": 746717184
},
{
"epoch": 0.23,
"learning_rate": 0.0003907823470411234,
"loss": 2.8449,
"theoretical_loss": 3.7548527020891163,
"tokens_seen": 746782720
},
{
"epoch": 0.23,
"learning_rate": 0.00039077231695085256,
"loss": 2.881,
"theoretical_loss": 3.754819630914481,
"tokens_seen": 746848256
},
{
"epoch": 0.23,
"learning_rate": 0.00039076228686058174,
"loss": 2.9179,
"theoretical_loss": 3.7547865634541946,
"tokens_seen": 746913792
},
{
"epoch": 0.23,
"learning_rate": 0.000390752256770311,
"loss": 2.9315,
"theoretical_loss": 3.7547534997075145,
"tokens_seen": 746979328
},
{
"epoch": 0.23,
"learning_rate": 0.0003907422266800401,
"loss": 3.12,
"theoretical_loss": 3.754720439673698,
"tokens_seen": 747044864
},
{
"epoch": 0.23,
"objective/train/docs_used": 431396,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.491447687149048,
"objective/train/theoretical_loss": 3.754687383352003,
"objective/train/tokens_used": 767570400,
"theoretical_loss": 3.754687383352003,
"tokens_seen": 747110400
},
{
"epoch": 0.23,
"learning_rate": 0.00039073219658976934,
"loss": 2.8623,
"theoretical_loss": 3.754687383352003,
"tokens_seen": 747110400
},
{
"epoch": 0.23,
"learning_rate": 0.00039072216649949846,
"loss": 2.9521,
"theoretical_loss": 3.754654330741687,
"tokens_seen": 747175936
},
{
"epoch": 0.23,
"learning_rate": 0.0003907121364092277,
"loss": 3.0817,
"theoretical_loss": 3.754621281842007,
"tokens_seen": 747241472
},
{
"epoch": 0.23,
"learning_rate": 0.0003907021063189569,
"loss": 2.6782,
"theoretical_loss": 3.7545882366522214,
"tokens_seen": 747307008
},
{
"epoch": 0.23,
"learning_rate": 0.00039069207622868606,
"loss": 2.8759,
"theoretical_loss": 3.7545551951715885,
"tokens_seen": 747372544
},
{
"epoch": 0.23,
"learning_rate": 0.00039068204613841524,
"loss": 2.9082,
"theoretical_loss": 3.754522157399367,
"tokens_seen": 747438080
},
{
"epoch": 0.23,
"learning_rate": 0.0003906720160481445,
"loss": 2.8501,
"theoretical_loss": 3.7544891233348157,
"tokens_seen": 747503616
},
{
"epoch": 0.23,
"learning_rate": 0.0003906619859578736,
"loss": 2.9944,
"theoretical_loss": 3.7544560929771924,
"tokens_seen": 747569152
},
{
"epoch": 0.23,
"learning_rate": 0.00039065195586760284,
"loss": 2.7861,
"theoretical_loss": 3.7544230663257583,
"tokens_seen": 747634688
},
{
"epoch": 0.23,
"learning_rate": 0.00039064192577733197,
"loss": 2.7932,
"theoretical_loss": 3.7543900433797703,
"tokens_seen": 747700224
},
{
"epoch": 0.23,
"learning_rate": 0.0003906318956870612,
"loss": 2.9967,
"theoretical_loss": 3.75435702413849,
"tokens_seen": 747765760
},
{
"epoch": 0.23,
"learning_rate": 0.0003906218655967904,
"loss": 2.6927,
"theoretical_loss": 3.7543240086011753,
"tokens_seen": 747831296
},
{
"epoch": 0.23,
"learning_rate": 0.00039061183550651957,
"loss": 2.9451,
"theoretical_loss": 3.7542909967670877,
"tokens_seen": 747896832
},
{
"epoch": 0.23,
"learning_rate": 0.00039060180541624875,
"loss": 2.8362,
"theoretical_loss": 3.754257988635487,
"tokens_seen": 747962368
},
{
"epoch": 0.23,
"learning_rate": 0.00039059177532597793,
"loss": 3.012,
"theoretical_loss": 3.754224984205633,
"tokens_seen": 748027904
},
{
"epoch": 0.23,
"learning_rate": 0.0003905817452357071,
"loss": 2.889,
"theoretical_loss": 3.7541919834767867,
"tokens_seen": 748093440
},
{
"epoch": 0.23,
"learning_rate": 0.00039057171514543634,
"loss": 2.9712,
"theoretical_loss": 3.754158986448209,
"tokens_seen": 748158976
},
{
"epoch": 0.23,
"learning_rate": 0.00039056168505516547,
"loss": 2.8266,
"theoretical_loss": 3.7541259931191604,
"tokens_seen": 748224512
},
{
"epoch": 0.23,
"learning_rate": 0.0003905516549648947,
"loss": 2.8058,
"theoretical_loss": 3.7540930034889035,
"tokens_seen": 748290048
},
{
"epoch": 0.23,
"learning_rate": 0.00039054162487462383,
"loss": 2.96,
"theoretical_loss": 3.7540600175566983,
"tokens_seen": 748355584
},
{
"epoch": 0.23,
"learning_rate": 0.00039053159478435307,
"loss": 2.8323,
"theoretical_loss": 3.7540270353218066,
"tokens_seen": 748421120
},
{
"epoch": 0.23,
"learning_rate": 0.00039052156469408225,
"loss": 3.0316,
"theoretical_loss": 3.7539940567834913,
"tokens_seen": 748486656
},
{
"epoch": 0.23,
"learning_rate": 0.00039051153460381143,
"loss": 2.7533,
"theoretical_loss": 3.753961081941013,
"tokens_seen": 748552192
},
{
"epoch": 0.23,
"learning_rate": 0.0003905015045135406,
"loss": 2.818,
"theoretical_loss": 3.7539281107936358,
"tokens_seen": 748617728
},
{
"epoch": 0.23,
"learning_rate": 0.00039049147442326985,
"loss": 2.718,
"theoretical_loss": 3.7538951433406202,
"tokens_seen": 748683264
},
{
"epoch": 0.23,
"objective/train/docs_used": 432599,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9539995193481445,
"objective/train/theoretical_loss": 3.7538621795812306,
"objective/train/tokens_used": 769208800,
"theoretical_loss": 3.7538621795812306,
"tokens_seen": 748748800
},
{
"epoch": 0.23,
"learning_rate": 0.000390481444332999,
"loss": 2.9239,
"theoretical_loss": 3.7538621795812306,
"tokens_seen": 748748800
},
{
"epoch": 0.23,
"learning_rate": 0.0003904714142427282,
"loss": 2.7473,
"theoretical_loss": 3.7538292195147287,
"tokens_seen": 748814336
},
{
"epoch": 0.23,
"learning_rate": 0.00039046138415245734,
"loss": 2.7974,
"theoretical_loss": 3.753796263140379,
"tokens_seen": 748879872
},
{
"epoch": 0.23,
"learning_rate": 0.00039045135406218657,
"loss": 2.9379,
"theoretical_loss": 3.753763310457444,
"tokens_seen": 748945408
},
{
"epoch": 0.23,
"learning_rate": 0.00039044132397191575,
"loss": 2.6428,
"theoretical_loss": 3.7537303614651867,
"tokens_seen": 749010944
},
{
"epoch": 0.23,
"learning_rate": 0.00039043129388164493,
"loss": 2.8572,
"theoretical_loss": 3.753697416162872,
"tokens_seen": 749076480
},
{
"epoch": 0.23,
"learning_rate": 0.0003904212637913741,
"loss": 2.868,
"theoretical_loss": 3.753664474549763,
"tokens_seen": 749142016
},
{
"epoch": 0.23,
"learning_rate": 0.0003904112337011033,
"loss": 2.9766,
"theoretical_loss": 3.753631536625125,
"tokens_seen": 749207552
},
{
"epoch": 0.23,
"learning_rate": 0.0003904012036108325,
"loss": 3.168,
"theoretical_loss": 3.7535986023882213,
"tokens_seen": 749273088
},
{
"epoch": 0.23,
"learning_rate": 0.0003903911735205617,
"loss": 2.792,
"theoretical_loss": 3.753565671838317,
"tokens_seen": 749338624
},
{
"epoch": 0.23,
"learning_rate": 0.0003903811434302909,
"loss": 2.9154,
"theoretical_loss": 3.753532744974677,
"tokens_seen": 749404160
},
{
"epoch": 0.23,
"learning_rate": 0.0003903711133400201,
"loss": 2.8761,
"theoretical_loss": 3.753499821796566,
"tokens_seen": 749469696
},
{
"epoch": 0.23,
"learning_rate": 0.00039036108324974926,
"loss": 2.9695,
"theoretical_loss": 3.7534669023032494,
"tokens_seen": 749535232
},
{
"epoch": 0.23,
"learning_rate": 0.00039035105315947844,
"loss": 2.991,
"theoretical_loss": 3.7534339864939934,
"tokens_seen": 749600768
},
{
"epoch": 0.23,
"learning_rate": 0.00039034102306920767,
"loss": 2.8697,
"theoretical_loss": 3.7534010743680626,
"tokens_seen": 749666304
},
{
"epoch": 0.23,
"learning_rate": 0.0003903309929789368,
"loss": 3.077,
"theoretical_loss": 3.7533681659247233,
"tokens_seen": 749731840
},
{
"epoch": 0.23,
"learning_rate": 0.00039032096288866603,
"loss": 2.8097,
"theoretical_loss": 3.7533352611632416,
"tokens_seen": 749797376
},
{
"epoch": 0.23,
"learning_rate": 0.0003903109327983952,
"loss": 2.7494,
"theoretical_loss": 3.753302360082884,
"tokens_seen": 749862912
},
{
"epoch": 0.23,
"learning_rate": 0.0003903009027081244,
"loss": 2.9132,
"theoretical_loss": 3.7532694626829164,
"tokens_seen": 749928448
},
{
"epoch": 0.23,
"learning_rate": 0.0003902908726178536,
"loss": 2.8225,
"theoretical_loss": 3.753236568962606,
"tokens_seen": 749993984
},
{
"epoch": 0.23,
"learning_rate": 0.00039028084252758276,
"loss": 3.188,
"theoretical_loss": 3.7532036789212206,
"tokens_seen": 750059520
},
{
"epoch": 0.23,
"learning_rate": 0.00039027081243731194,
"loss": 3.0369,
"theoretical_loss": 3.7531707925580253,
"tokens_seen": 750125056
},
{
"epoch": 0.23,
"learning_rate": 0.0003902607823470412,
"loss": 3.0019,
"theoretical_loss": 3.7531379098722897,
"tokens_seen": 750190592
},
{
"epoch": 0.23,
"learning_rate": 0.0003902507522567703,
"loss": 2.7219,
"theoretical_loss": 3.7531050308632796,
"tokens_seen": 750256128
},
{
"epoch": 0.23,
"learning_rate": 0.00039024072216649954,
"loss": 2.8947,
"theoretical_loss": 3.7530721555302637,
"tokens_seen": 750321664
},
{
"epoch": 0.23,
"objective/train/docs_used": 433353,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.757408380508423,
"objective/train/theoretical_loss": 3.7530392838725097,
"objective/train/tokens_used": 770847200,
"theoretical_loss": 3.7530392838725097,
"tokens_seen": 750387200
},
{
"epoch": 0.23,
"learning_rate": 0.00039023069207622866,
"loss": 3.0001,
"theoretical_loss": 3.7530392838725097,
"tokens_seen": 750387200
},
{
"epoch": 0.23,
"learning_rate": 0.0003902206619859579,
"loss": 2.794,
"theoretical_loss": 3.7530064158892857,
"tokens_seen": 750452736
},
{
"epoch": 0.23,
"learning_rate": 0.0003902106318956871,
"loss": 2.8246,
"theoretical_loss": 3.7529735515798603,
"tokens_seen": 750518272
},
{
"epoch": 0.23,
"learning_rate": 0.00039020060180541626,
"loss": 2.8506,
"theoretical_loss": 3.752940690943502,
"tokens_seen": 750583808
},
{
"epoch": 0.23,
"learning_rate": 0.00039019057171514544,
"loss": 2.9174,
"theoretical_loss": 3.7529078339794806,
"tokens_seen": 750649344
},
{
"epoch": 0.23,
"learning_rate": 0.0003901805416248747,
"loss": 2.8396,
"theoretical_loss": 3.7528749806870634,
"tokens_seen": 750714880
},
{
"epoch": 0.23,
"learning_rate": 0.0003901705115346038,
"loss": 3.0644,
"theoretical_loss": 3.7528421310655204,
"tokens_seen": 750780416
},
{
"epoch": 0.23,
"learning_rate": 0.00039016048144433304,
"loss": 2.973,
"theoretical_loss": 3.7528092851141213,
"tokens_seen": 750845952
},
{
"epoch": 0.23,
"learning_rate": 0.00039015045135406217,
"loss": 3.081,
"theoretical_loss": 3.7527764428321357,
"tokens_seen": 750911488
},
{
"epoch": 0.23,
"learning_rate": 0.0003901404212637914,
"loss": 2.7496,
"theoretical_loss": 3.752743604218834,
"tokens_seen": 750977024
},
{
"epoch": 0.23,
"learning_rate": 0.0003901303911735206,
"loss": 2.9144,
"theoretical_loss": 3.7527107692734845,
"tokens_seen": 751042560
},
{
"epoch": 0.23,
"learning_rate": 0.00039012036108324977,
"loss": 3.0587,
"theoretical_loss": 3.752677937995359,
"tokens_seen": 751108096
},
{
"epoch": 0.23,
"learning_rate": 0.00039011033099297895,
"loss": 2.9057,
"theoretical_loss": 3.752645110383728,
"tokens_seen": 751173632
},
{
"epoch": 0.23,
"learning_rate": 0.00039010030090270813,
"loss": 2.7963,
"theoretical_loss": 3.752612286437862,
"tokens_seen": 751239168
},
{
"epoch": 0.23,
"learning_rate": 0.0003900902708124373,
"loss": 2.982,
"theoretical_loss": 3.752579466157031,
"tokens_seen": 751304704
},
{
"epoch": 0.23,
"learning_rate": 0.00039008024072216654,
"loss": 2.7883,
"theoretical_loss": 3.7525466495405073,
"tokens_seen": 751370240
},
{
"epoch": 0.23,
"learning_rate": 0.00039007021063189567,
"loss": 2.8758,
"theoretical_loss": 3.7525138365875623,
"tokens_seen": 751435776
},
{
"epoch": 0.23,
"learning_rate": 0.0003900601805416249,
"loss": 2.8594,
"theoretical_loss": 3.752481027297466,
"tokens_seen": 751501312
},
{
"epoch": 0.23,
"learning_rate": 0.00039005015045135403,
"loss": 2.9585,
"theoretical_loss": 3.7524482216694923,
"tokens_seen": 751566848
},
{
"epoch": 0.23,
"learning_rate": 0.00039004012036108327,
"loss": 2.7236,
"theoretical_loss": 3.7524154197029116,
"tokens_seen": 751632384
},
{
"epoch": 0.23,
"learning_rate": 0.00039003009027081245,
"loss": 2.8638,
"theoretical_loss": 3.752382621396997,
"tokens_seen": 751697920
},
{
"epoch": 0.23,
"learning_rate": 0.00039002006018054163,
"loss": 2.8205,
"theoretical_loss": 3.7523498267510202,
"tokens_seen": 751763456
},
{
"epoch": 0.23,
"learning_rate": 0.0003900100300902708,
"loss": 2.813,
"theoretical_loss": 3.7523170357642535,
"tokens_seen": 751828992
},
{
"epoch": 0.23,
"learning_rate": 0.00039000000000000005,
"loss": 2.9686,
"theoretical_loss": 3.752284248435971,
"tokens_seen": 751894528
},
{
"epoch": 0.23,
"learning_rate": 0.0003899899699097292,
"loss": 2.8649,
"theoretical_loss": 3.752251464765444,
"tokens_seen": 751960064
},
{
"epoch": 0.23,
"objective/train/docs_used": 434593,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3101871013641357,
"objective/train/theoretical_loss": 3.7522186847519476,
"objective/train/tokens_used": 772485600,
"theoretical_loss": 3.7522186847519476,
"tokens_seen": 752025600
},
{
"epoch": 0.23,
"learning_rate": 0.0003899799398194584,
"loss": 3.2136,
"theoretical_loss": 3.7522186847519476,
"tokens_seen": 752025600
},
{
"epoch": 0.23,
"learning_rate": 0.00038996990972918754,
"loss": 2.9204,
"theoretical_loss": 3.752185908394754,
"tokens_seen": 752091136
},
{
"epoch": 0.23,
"learning_rate": 0.00038995987963891677,
"loss": 2.9649,
"theoretical_loss": 3.7521531356931366,
"tokens_seen": 752156672
},
{
"epoch": 0.23,
"learning_rate": 0.00038994984954864595,
"loss": 2.7295,
"theoretical_loss": 3.75212036664637,
"tokens_seen": 752222208
},
{
"epoch": 0.23,
"learning_rate": 0.00038993981945837513,
"loss": 3.1142,
"theoretical_loss": 3.7520876012537276,
"tokens_seen": 752287744
},
{
"epoch": 0.23,
"learning_rate": 0.0003899297893681043,
"loss": 3.072,
"theoretical_loss": 3.7520548395144844,
"tokens_seen": 752353280
},
{
"epoch": 0.23,
"learning_rate": 0.0003899197592778335,
"loss": 2.9184,
"theoretical_loss": 3.7520220814279144,
"tokens_seen": 752418816
},
{
"epoch": 0.23,
"learning_rate": 0.0003899097291875627,
"loss": 2.7563,
"theoretical_loss": 3.751989326993292,
"tokens_seen": 752484352
},
{
"epoch": 0.23,
"learning_rate": 0.0003898996990972919,
"loss": 2.795,
"theoretical_loss": 3.751956576209892,
"tokens_seen": 752549888
},
{
"epoch": 0.23,
"learning_rate": 0.00038988966900702104,
"loss": 2.8861,
"theoretical_loss": 3.7519238290769907,
"tokens_seen": 752615424
},
{
"epoch": 0.23,
"learning_rate": 0.0003898796389167503,
"loss": 2.8211,
"theoretical_loss": 3.751891085593862,
"tokens_seen": 752680960
},
{
"epoch": 0.23,
"learning_rate": 0.0003898696088264794,
"loss": 2.5466,
"theoretical_loss": 3.751858345759782,
"tokens_seen": 752746496
},
{
"epoch": 0.23,
"learning_rate": 0.00038985957873620864,
"loss": 2.9317,
"theoretical_loss": 3.751825609574026,
"tokens_seen": 752812032
},
{
"epoch": 0.23,
"learning_rate": 0.0003898495486459378,
"loss": 3.0198,
"theoretical_loss": 3.7517928770358706,
"tokens_seen": 752877568
},
{
"epoch": 0.23,
"learning_rate": 0.000389839518555667,
"loss": 2.7687,
"theoretical_loss": 3.7517601481445917,
"tokens_seen": 752943104
},
{
"epoch": 0.23,
"learning_rate": 0.0003898294884653962,
"loss": 3.0448,
"theoretical_loss": 3.7517274228994646,
"tokens_seen": 753008640
},
{
"epoch": 0.23,
"learning_rate": 0.0003898194583751254,
"loss": 3.0726,
"theoretical_loss": 3.7516947012997672,
"tokens_seen": 753074176
},
{
"epoch": 0.23,
"learning_rate": 0.00038980942828485454,
"loss": 2.9167,
"theoretical_loss": 3.7516619833447757,
"tokens_seen": 753139712
},
{
"epoch": 0.23,
"learning_rate": 0.0003897993981945838,
"loss": 3.0128,
"theoretical_loss": 3.7516292690337667,
"tokens_seen": 753205248
},
{
"epoch": 0.23,
"learning_rate": 0.0003897893681043129,
"loss": 2.8765,
"theoretical_loss": 3.7515965583660176,
"tokens_seen": 753270784
},
{
"epoch": 0.23,
"learning_rate": 0.00038977933801404214,
"loss": 2.6383,
"theoretical_loss": 3.751563851340806,
"tokens_seen": 753336320
},
{
"epoch": 0.23,
"learning_rate": 0.0003897693079237713,
"loss": 2.801,
"theoretical_loss": 3.7515311479574094,
"tokens_seen": 753401856
},
{
"epoch": 0.23,
"learning_rate": 0.0003897592778335005,
"loss": 2.8801,
"theoretical_loss": 3.7514984482151053,
"tokens_seen": 753467392
},
{
"epoch": 0.23,
"learning_rate": 0.0003897492477432297,
"loss": 3.096,
"theoretical_loss": 3.7514657521131713,
"tokens_seen": 753532928
},
{
"epoch": 0.23,
"learning_rate": 0.00038973921765295886,
"loss": 2.8696,
"theoretical_loss": 3.7514330596508865,
"tokens_seen": 753598464
},
{
"epoch": 0.23,
"objective/train/docs_used": 435245,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3033926486968994,
"objective/train/theoretical_loss": 3.751400370827529,
"objective/train/tokens_used": 774124000,
"theoretical_loss": 3.751400370827529,
"tokens_seen": 753664000
},
{
"epoch": 0.23,
"learning_rate": 0.00038972918756268805,
"loss": 2.933,
"theoretical_loss": 3.751400370827529,
"tokens_seen": 753664000
},
{
"epoch": 0.23,
"learning_rate": 0.0003897191574724173,
"loss": 2.8697,
"theoretical_loss": 3.7513676856423768,
"tokens_seen": 753729536
},
{
"epoch": 0.23,
"learning_rate": 0.0003897091273821464,
"loss": 2.8624,
"theoretical_loss": 3.751335004094709,
"tokens_seen": 753795072
},
{
"epoch": 0.23,
"learning_rate": 0.00038969909729187564,
"loss": 2.8837,
"theoretical_loss": 3.751302326183805,
"tokens_seen": 753860608
},
{
"epoch": 0.23,
"learning_rate": 0.00038968906720160477,
"loss": 3.0994,
"theoretical_loss": 3.7512696519089435,
"tokens_seen": 753926144
},
{
"epoch": 0.23,
"learning_rate": 0.000389679037111334,
"loss": 2.9184,
"theoretical_loss": 3.751236981269404,
"tokens_seen": 753991680
},
{
"epoch": 0.23,
"learning_rate": 0.0003896690070210632,
"loss": 3.0235,
"theoretical_loss": 3.7512043142644664,
"tokens_seen": 754057216
},
{
"epoch": 0.23,
"learning_rate": 0.00038965897693079237,
"loss": 2.8313,
"theoretical_loss": 3.7511716508934105,
"tokens_seen": 754122752
},
{
"epoch": 0.23,
"learning_rate": 0.00038964894684052155,
"loss": 3.1002,
"theoretical_loss": 3.7511389911555155,
"tokens_seen": 754188288
},
{
"epoch": 0.23,
"learning_rate": 0.0003896389167502508,
"loss": 3.2198,
"theoretical_loss": 3.7511063350500624,
"tokens_seen": 754253824
},
{
"epoch": 0.23,
"learning_rate": 0.00038962888665997997,
"loss": 3.0033,
"theoretical_loss": 3.7510736825763322,
"tokens_seen": 754319360
},
{
"epoch": 0.23,
"learning_rate": 0.00038961885656970915,
"loss": 2.6073,
"theoretical_loss": 3.751041033733604,
"tokens_seen": 754384896
},
{
"epoch": 0.23,
"learning_rate": 0.00038960882647943833,
"loss": 2.6935,
"theoretical_loss": 3.75100838852116,
"tokens_seen": 754450432
},
{
"epoch": 0.23,
"learning_rate": 0.0003895987963891675,
"loss": 3.0088,
"theoretical_loss": 3.7509757469382805,
"tokens_seen": 754515968
},
{
"epoch": 0.23,
"learning_rate": 0.00038958876629889674,
"loss": 2.797,
"theoretical_loss": 3.750943108984247,
"tokens_seen": 754581504
},
{
"epoch": 0.23,
"learning_rate": 0.00038957873620862587,
"loss": 3.004,
"theoretical_loss": 3.7509104746583413,
"tokens_seen": 754647040
},
{
"epoch": 0.23,
"learning_rate": 0.0003895687061183551,
"loss": 2.8447,
"theoretical_loss": 3.7508778439598442,
"tokens_seen": 754712576
},
{
"epoch": 0.23,
"learning_rate": 0.00038955867602808423,
"loss": 2.8994,
"theoretical_loss": 3.750845216888038,
"tokens_seen": 754778112
},
{
"epoch": 0.23,
"learning_rate": 0.00038954864593781347,
"loss": 2.9306,
"theoretical_loss": 3.7508125934422054,
"tokens_seen": 754843648
},
{
"epoch": 0.23,
"learning_rate": 0.00038953861584754265,
"loss": 2.8417,
"theoretical_loss": 3.7507799736216274,
"tokens_seen": 754909184
},
{
"epoch": 0.23,
"learning_rate": 0.00038952858575727183,
"loss": 2.9297,
"theoretical_loss": 3.7507473574255874,
"tokens_seen": 754974720
},
{
"epoch": 0.23,
"learning_rate": 0.000389518555667001,
"loss": 2.7477,
"theoretical_loss": 3.7507147448533678,
"tokens_seen": 755040256
},
{
"epoch": 0.23,
"learning_rate": 0.00038950852557673025,
"loss": 2.8212,
"theoretical_loss": 3.7506821359042517,
"tokens_seen": 755105792
},
{
"epoch": 0.23,
"learning_rate": 0.0003894984954864594,
"loss": 2.8769,
"theoretical_loss": 3.7506495305775225,
"tokens_seen": 755171328
},
{
"epoch": 0.23,
"learning_rate": 0.0003894884653961886,
"loss": 2.9594,
"theoretical_loss": 3.7506169288724625,
"tokens_seen": 755236864
},
{
"epoch": 0.23,
"objective/train/docs_used": 436337,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6367499828338623,
"objective/train/theoretical_loss": 3.7505843307883557,
"objective/train/tokens_used": 775762400,
"theoretical_loss": 3.7505843307883557,
"tokens_seen": 755302400
},
{
"epoch": 0.23,
"learning_rate": 0.00038947843530591774,
"loss": 2.7764,
"theoretical_loss": 3.7505843307883557,
"tokens_seen": 755302400
},
{
"epoch": 0.23,
"learning_rate": 0.00038946840521564697,
"loss": 2.834,
"theoretical_loss": 3.750551736324486,
"tokens_seen": 755367936
},
{
"epoch": 0.23,
"learning_rate": 0.00038945837512537615,
"loss": 2.86,
"theoretical_loss": 3.750519145480137,
"tokens_seen": 755433472
},
{
"epoch": 0.23,
"learning_rate": 0.00038944834503510533,
"loss": 2.7089,
"theoretical_loss": 3.750486558254593,
"tokens_seen": 755499008
},
{
"epoch": 0.23,
"learning_rate": 0.0003894383149448345,
"loss": 2.7661,
"theoretical_loss": 3.7504539746471384,
"tokens_seen": 755564544
},
{
"epoch": 0.23,
"learning_rate": 0.0003894282848545637,
"loss": 3.2421,
"theoretical_loss": 3.7504213946570575,
"tokens_seen": 755630080
},
{
"epoch": 0.23,
"learning_rate": 0.0003894182547642929,
"loss": 2.7973,
"theoretical_loss": 3.7503888182836356,
"tokens_seen": 755695616
},
{
"epoch": 0.23,
"learning_rate": 0.0003894082246740221,
"loss": 2.8901,
"theoretical_loss": 3.7503562455261568,
"tokens_seen": 755761152
},
{
"epoch": 0.23,
"learning_rate": 0.00038939819458375124,
"loss": 3.0572,
"theoretical_loss": 3.750323676383906,
"tokens_seen": 755826688
},
{
"epoch": 0.23,
"learning_rate": 0.0003893881644934805,
"loss": 2.8714,
"theoretical_loss": 3.7502911108561694,
"tokens_seen": 755892224
},
{
"epoch": 0.23,
"learning_rate": 0.0003893781344032096,
"loss": 2.637,
"theoretical_loss": 3.7502585489422327,
"tokens_seen": 755957760
},
{
"epoch": 0.23,
"learning_rate": 0.00038936810431293884,
"loss": 2.9113,
"theoretical_loss": 3.750225990641381,
"tokens_seen": 756023296
},
{
"epoch": 0.23,
"learning_rate": 0.000389358074222668,
"loss": 2.8942,
"theoretical_loss": 3.7501934359529,
"tokens_seen": 756088832
},
{
"epoch": 0.23,
"learning_rate": 0.0003893480441323972,
"loss": 2.9899,
"theoretical_loss": 3.7501608848760766,
"tokens_seen": 756154368
},
{
"epoch": 0.23,
"learning_rate": 0.0003893380140421264,
"loss": 2.9712,
"theoretical_loss": 3.750128337410197,
"tokens_seen": 756219904
},
{
"epoch": 0.23,
"learning_rate": 0.0003893279839518556,
"loss": 2.8745,
"theoretical_loss": 3.7500957935545474,
"tokens_seen": 756285440
},
{
"epoch": 0.23,
"learning_rate": 0.00038931795386158474,
"loss": 2.9926,
"theoretical_loss": 3.750063253308414,
"tokens_seen": 756350976
},
{
"epoch": 0.23,
"learning_rate": 0.000389307923771314,
"loss": 2.7558,
"theoretical_loss": 3.750030716671086,
"tokens_seen": 756416512
},
{
"epoch": 0.23,
"learning_rate": 0.0003892978936810431,
"loss": 2.7521,
"theoretical_loss": 3.7499981836418477,
"tokens_seen": 756482048
},
{
"epoch": 0.23,
"learning_rate": 0.00038928786359077234,
"loss": 2.8834,
"theoretical_loss": 3.7499656542199884,
"tokens_seen": 756547584
},
{
"epoch": 0.23,
"learning_rate": 0.0003892778335005015,
"loss": 3.0121,
"theoretical_loss": 3.7499331284047948,
"tokens_seen": 756613120
},
{
"epoch": 0.23,
"learning_rate": 0.0003892678034102307,
"loss": 3.0472,
"theoretical_loss": 3.749900606195555,
"tokens_seen": 756678656
},
{
"epoch": 0.23,
"learning_rate": 0.0003892577733199599,
"loss": 2.8345,
"theoretical_loss": 3.749868087591557,
"tokens_seen": 756744192
},
{
"epoch": 0.23,
"learning_rate": 0.00038924774322968906,
"loss": 3.0445,
"theoretical_loss": 3.7498355725920884,
"tokens_seen": 756809728
},
{
"epoch": 0.23,
"learning_rate": 0.00038923771313941825,
"loss": 2.8067,
"theoretical_loss": 3.7498030611964386,
"tokens_seen": 756875264
},
{
"epoch": 0.23,
"objective/train/docs_used": 436904,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.364593267440796,
"objective/train/theoretical_loss": 3.749770553403895,
"objective/train/tokens_used": 777400800,
"theoretical_loss": 3.749770553403895,
"tokens_seen": 756940800
},
{
"epoch": 0.23,
"learning_rate": 0.0003892276830491475,
"loss": 2.9628,
"theoretical_loss": 3.749770553403895,
"tokens_seen": 756940800
},
{
"epoch": 0.23,
"learning_rate": 0.0003892176529588766,
"loss": 3.083,
"theoretical_loss": 3.7497380492137475,
"tokens_seen": 757006336
},
{
"epoch": 0.23,
"learning_rate": 0.00038920762286860584,
"loss": 3.0409,
"theoretical_loss": 3.7497055486252844,
"tokens_seen": 757071872
},
{
"epoch": 0.23,
"learning_rate": 0.00038919759277833497,
"loss": 2.9215,
"theoretical_loss": 3.749673051637795,
"tokens_seen": 757137408
},
{
"epoch": 0.23,
"learning_rate": 0.0003891875626880642,
"loss": 2.878,
"theoretical_loss": 3.7496405582505687,
"tokens_seen": 757202944
},
{
"epoch": 0.23,
"learning_rate": 0.0003891775325977934,
"loss": 2.7803,
"theoretical_loss": 3.7496080684628956,
"tokens_seen": 757268480
},
{
"epoch": 0.23,
"learning_rate": 0.00038916750250752257,
"loss": 2.901,
"theoretical_loss": 3.7495755822740646,
"tokens_seen": 757334016
},
{
"epoch": 0.23,
"learning_rate": 0.00038915747241725175,
"loss": 2.9094,
"theoretical_loss": 3.749543099683366,
"tokens_seen": 757399552
},
{
"epoch": 0.23,
"learning_rate": 0.000389147442326981,
"loss": 2.6775,
"theoretical_loss": 3.7495106206900903,
"tokens_seen": 757465088
},
{
"epoch": 0.23,
"learning_rate": 0.0003891374122367101,
"loss": 2.8156,
"theoretical_loss": 3.7494781452935273,
"tokens_seen": 757530624
},
{
"epoch": 0.23,
"learning_rate": 0.00038912738214643935,
"loss": 2.8003,
"theoretical_loss": 3.749445673492968,
"tokens_seen": 757596160
},
{
"epoch": 0.23,
"learning_rate": 0.0003891173520561685,
"loss": 2.9171,
"theoretical_loss": 3.7494132052877034,
"tokens_seen": 757661696
},
{
"epoch": 0.23,
"learning_rate": 0.0003891073219658977,
"loss": 2.5722,
"theoretical_loss": 3.749380740677024,
"tokens_seen": 757727232
},
{
"epoch": 0.23,
"learning_rate": 0.0003890972918756269,
"loss": 2.8653,
"theoretical_loss": 3.7493482796602216,
"tokens_seen": 757792768
},
{
"epoch": 0.23,
"learning_rate": 0.00038908726178535607,
"loss": 3.0111,
"theoretical_loss": 3.749315822236587,
"tokens_seen": 757858304
},
{
"epoch": 0.23,
"learning_rate": 0.00038907723169508525,
"loss": 2.8182,
"theoretical_loss": 3.749283368405412,
"tokens_seen": 757923840
},
{
"epoch": 0.23,
"learning_rate": 0.00038906720160481443,
"loss": 2.8502,
"theoretical_loss": 3.7492509181659885,
"tokens_seen": 757989376
},
{
"epoch": 0.23,
"learning_rate": 0.0003890571715145436,
"loss": 2.8867,
"theoretical_loss": 3.749218471517608,
"tokens_seen": 758054912
},
{
"epoch": 0.23,
"learning_rate": 0.00038904714142427285,
"loss": 2.7202,
"theoretical_loss": 3.749186028459564,
"tokens_seen": 758120448
},
{
"epoch": 0.23,
"learning_rate": 0.000389037111334002,
"loss": 2.8137,
"theoretical_loss": 3.749153588991147,
"tokens_seen": 758185984
},
{
"epoch": 0.23,
"learning_rate": 0.0003890270812437312,
"loss": 2.9705,
"theoretical_loss": 3.7491211531116515,
"tokens_seen": 758251520
},
{
"epoch": 0.23,
"learning_rate": 0.0003890170511534604,
"loss": 2.7548,
"theoretical_loss": 3.749088720820369,
"tokens_seen": 758317056
},
{
"epoch": 0.23,
"learning_rate": 0.0003890070210631896,
"loss": 2.7693,
"theoretical_loss": 3.749056292116593,
"tokens_seen": 758382592
},
{
"epoch": 0.23,
"learning_rate": 0.00038899699097291876,
"loss": 2.939,
"theoretical_loss": 3.7490238669996163,
"tokens_seen": 758448128
},
{
"epoch": 0.23,
"learning_rate": 0.00038898696088264794,
"loss": 2.8689,
"theoretical_loss": 3.748991445468733,
"tokens_seen": 758513664
},
{
"epoch": 0.23,
"objective/train/docs_used": 438157,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9970943927764893,
"objective/train/theoretical_loss": 3.7489590275232367,
"objective/train/tokens_used": 779039200,
"theoretical_loss": 3.7489590275232367,
"tokens_seen": 758579200
},
{
"epoch": 0.23,
"learning_rate": 0.0003889769307923771,
"loss": 3.1026,
"theoretical_loss": 3.7489590275232367,
"tokens_seen": 758579200
},
{
"epoch": 0.23,
"learning_rate": 0.00038896690070210635,
"loss": 2.8217,
"theoretical_loss": 3.7489266131624204,
"tokens_seen": 758644736
},
{
"epoch": 0.23,
"learning_rate": 0.0003889568706118355,
"loss": 2.8756,
"theoretical_loss": 3.7488942023855785,
"tokens_seen": 758710272
},
{
"epoch": 0.23,
"learning_rate": 0.0003889468405215647,
"loss": 3.1058,
"theoretical_loss": 3.7488617951920054,
"tokens_seen": 758775808
},
{
"epoch": 0.23,
"learning_rate": 0.00038893681043129384,
"loss": 2.8331,
"theoretical_loss": 3.748829391580995,
"tokens_seen": 758841344
},
{
"epoch": 0.23,
"learning_rate": 0.0003889267803410231,
"loss": 2.9086,
"theoretical_loss": 3.748796991551843,
"tokens_seen": 758906880
},
{
"epoch": 0.23,
"learning_rate": 0.00038891675025075226,
"loss": 2.8446,
"theoretical_loss": 3.7487645951038426,
"tokens_seen": 758972416
},
{
"epoch": 0.23,
"learning_rate": 0.00038890672016048144,
"loss": 2.5735,
"theoretical_loss": 3.7487322022362903,
"tokens_seen": 759037952
},
{
"epoch": 0.23,
"learning_rate": 0.0003888966900702106,
"loss": 2.7167,
"theoretical_loss": 3.7486998129484803,
"tokens_seen": 759103488
},
{
"epoch": 0.23,
"learning_rate": 0.0003888866599799398,
"loss": 2.8274,
"theoretical_loss": 3.7486674272397082,
"tokens_seen": 759169024
},
{
"epoch": 0.23,
"learning_rate": 0.00038887662988966904,
"loss": 2.8973,
"theoretical_loss": 3.7486350451092703,
"tokens_seen": 759234560
},
{
"epoch": 0.23,
"learning_rate": 0.0003888665997993982,
"loss": 2.9813,
"theoretical_loss": 3.7486026665564616,
"tokens_seen": 759300096
},
{
"epoch": 0.23,
"learning_rate": 0.0003888565697091274,
"loss": 2.634,
"theoretical_loss": 3.7485702915805783,
"tokens_seen": 759365632
},
{
"epoch": 0.23,
"learning_rate": 0.0003888465396188566,
"loss": 2.9067,
"theoretical_loss": 3.7485379201809166,
"tokens_seen": 759431168
},
{
"epoch": 0.23,
"learning_rate": 0.0003888365095285858,
"loss": 2.9773,
"theoretical_loss": 3.7485055523567734,
"tokens_seen": 759496704
},
{
"epoch": 0.23,
"learning_rate": 0.00038882647943831494,
"loss": 2.5512,
"theoretical_loss": 3.748473188107444,
"tokens_seen": 759562240
},
{
"epoch": 0.23,
"learning_rate": 0.0003888164493480442,
"loss": 2.972,
"theoretical_loss": 3.7484408274322267,
"tokens_seen": 759627776
},
{
"epoch": 0.23,
"learning_rate": 0.0003888064192577733,
"loss": 2.7541,
"theoretical_loss": 3.7484084703304177,
"tokens_seen": 759693312
},
{
"epoch": 0.23,
"learning_rate": 0.00038879638916750254,
"loss": 2.8224,
"theoretical_loss": 3.7483761168013143,
"tokens_seen": 759758848
},
{
"epoch": 0.23,
"learning_rate": 0.0003887863590772317,
"loss": 2.8299,
"theoretical_loss": 3.748343766844214,
"tokens_seen": 759824384
},
{
"epoch": 0.23,
"learning_rate": 0.0003887763289869609,
"loss": 2.855,
"theoretical_loss": 3.7483114204584145,
"tokens_seen": 759889920
},
{
"epoch": 0.23,
"learning_rate": 0.0003887662988966901,
"loss": 2.7481,
"theoretical_loss": 3.748279077643213,
"tokens_seen": 759955456
},
{
"epoch": 0.23,
"learning_rate": 0.00038875626880641927,
"loss": 3.0196,
"theoretical_loss": 3.748246738397908,
"tokens_seen": 760020992
},
{
"epoch": 0.23,
"learning_rate": 0.00038874623871614845,
"loss": 2.7769,
"theoretical_loss": 3.748214402721798,
"tokens_seen": 760086528
},
{
"epoch": 0.23,
"learning_rate": 0.0003887362086258777,
"loss": 2.7349,
"theoretical_loss": 3.7481820706141806,
"tokens_seen": 760152064
},
{
"epoch": 0.23,
"objective/train/docs_used": 438864,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4547252655029297,
"objective/train/theoretical_loss": 3.748149742074355,
"objective/train/tokens_used": 780677600,
"theoretical_loss": 3.748149742074355,
"tokens_seen": 760217600
},
{
"epoch": 0.23,
"learning_rate": 0.0003887261785356068,
"loss": 2.7673,
"theoretical_loss": 3.748149742074355,
"tokens_seen": 760217600
},
{
"epoch": 0.23,
"learning_rate": 0.00038871614844533604,
"loss": 2.7436,
"theoretical_loss": 3.7481174171016196,
"tokens_seen": 760283136
},
{
"epoch": 0.23,
"learning_rate": 0.00038870611835506517,
"loss": 2.8185,
"theoretical_loss": 3.7480850956952736,
"tokens_seen": 760348672
},
{
"epoch": 0.23,
"learning_rate": 0.0003886960882647944,
"loss": 2.7709,
"theoretical_loss": 3.7480527778546167,
"tokens_seen": 760414208
},
{
"epoch": 0.23,
"learning_rate": 0.0003886860581745236,
"loss": 2.785,
"theoretical_loss": 3.7480204635789462,
"tokens_seen": 760479744
},
{
"epoch": 0.23,
"learning_rate": 0.00038867602808425277,
"loss": 2.9846,
"theoretical_loss": 3.7479881528675643,
"tokens_seen": 760545280
},
{
"epoch": 0.23,
"learning_rate": 0.00038866599799398195,
"loss": 2.7666,
"theoretical_loss": 3.74795584571977,
"tokens_seen": 760610816
},
{
"epoch": 0.23,
"learning_rate": 0.0003886559679037112,
"loss": 2.9121,
"theoretical_loss": 3.7479235421348616,
"tokens_seen": 760676352
},
{
"epoch": 0.23,
"learning_rate": 0.0003886459378134403,
"loss": 2.9415,
"theoretical_loss": 3.747891242112142,
"tokens_seen": 760741888
},
{
"epoch": 0.23,
"learning_rate": 0.00038863590772316955,
"loss": 2.6811,
"theoretical_loss": 3.747858945650909,
"tokens_seen": 760807424
},
{
"epoch": 0.23,
"learning_rate": 0.0003886258776328987,
"loss": 2.7723,
"theoretical_loss": 3.747826652750465,
"tokens_seen": 760872960
},
{
"epoch": 0.23,
"learning_rate": 0.0003886158475426279,
"loss": 2.7644,
"theoretical_loss": 3.74779436341011,
"tokens_seen": 760938496
},
{
"epoch": 0.23,
"learning_rate": 0.0003886058174523571,
"loss": 2.8101,
"theoretical_loss": 3.7477620776291447,
"tokens_seen": 761004032
},
{
"epoch": 0.23,
"learning_rate": 0.00038859578736208627,
"loss": 2.8455,
"theoretical_loss": 3.7477297954068707,
"tokens_seen": 761069568
},
{
"epoch": 0.23,
"learning_rate": 0.00038858575727181545,
"loss": 2.7138,
"theoretical_loss": 3.747697516742589,
"tokens_seen": 761135104
},
{
"epoch": 0.23,
"learning_rate": 0.00038857572718154463,
"loss": 2.7626,
"theoretical_loss": 3.7476652416356018,
"tokens_seen": 761200640
},
{
"epoch": 0.23,
"learning_rate": 0.0003885656970912738,
"loss": 2.6976,
"theoretical_loss": 3.7476329700852107,
"tokens_seen": 761266176
},
{
"epoch": 0.23,
"learning_rate": 0.00038855566700100305,
"loss": 2.7191,
"theoretical_loss": 3.7476007020907165,
"tokens_seen": 761331712
},
{
"epoch": 0.23,
"learning_rate": 0.0003885456369107322,
"loss": 2.6053,
"theoretical_loss": 3.747568437651423,
"tokens_seen": 761397248
},
{
"epoch": 0.23,
"learning_rate": 0.0003885356068204614,
"loss": 2.8847,
"theoretical_loss": 3.7475361767666318,
"tokens_seen": 761462784
},
{
"epoch": 0.23,
"learning_rate": 0.0003885255767301906,
"loss": 2.7498,
"theoretical_loss": 3.747503919435645,
"tokens_seen": 761528320
},
{
"epoch": 0.23,
"learning_rate": 0.0003885155466399198,
"loss": 2.9592,
"theoretical_loss": 3.747471665657766,
"tokens_seen": 761593856
},
{
"epoch": 0.23,
"learning_rate": 0.00038850551654964896,
"loss": 2.743,
"theoretical_loss": 3.7474394154322974,
"tokens_seen": 761659392
},
{
"epoch": 0.23,
"learning_rate": 0.00038849548645937814,
"loss": 2.668,
"theoretical_loss": 3.747407168758542,
"tokens_seen": 761724928
},
{
"epoch": 0.23,
"learning_rate": 0.0003884854563691073,
"loss": 2.9205,
"theoretical_loss": 3.747374925635804,
"tokens_seen": 761790464
},
{
"epoch": 0.23,
"objective/train/docs_used": 439620,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.54030442237854,
"objective/train/theoretical_loss": 3.7473426860633863,
"objective/train/tokens_used": 782316000,
"theoretical_loss": 3.7473426860633863,
"tokens_seen": 761856000
},
{
"epoch": 0.23,
"learning_rate": 0.00038847542627883655,
"loss": 2.6085,
"theoretical_loss": 3.7473426860633863,
"tokens_seen": 761856000
},
{
"epoch": 0.23,
"learning_rate": 0.0003884653961885657,
"loss": 2.9202,
"theoretical_loss": 3.747310450040593,
"tokens_seen": 761921536
},
{
"epoch": 0.23,
"learning_rate": 0.0003884553660982949,
"loss": 2.8398,
"theoretical_loss": 3.747278217566728,
"tokens_seen": 761987072
},
{
"epoch": 0.23,
"learning_rate": 0.00038844533600802404,
"loss": 2.75,
"theoretical_loss": 3.7472459886410947,
"tokens_seen": 762052608
},
{
"epoch": 0.23,
"learning_rate": 0.0003884353059177533,
"loss": 2.9014,
"theoretical_loss": 3.747213763262998,
"tokens_seen": 762118144
},
{
"epoch": 0.23,
"learning_rate": 0.00038842527582748246,
"loss": 2.9821,
"theoretical_loss": 3.7471815414317424,
"tokens_seen": 762183680
},
{
"epoch": 0.23,
"learning_rate": 0.00038841524573721164,
"loss": 2.8165,
"theoretical_loss": 3.7471493231466324,
"tokens_seen": 762249216
},
{
"epoch": 0.23,
"learning_rate": 0.0003884052156469408,
"loss": 2.5504,
"theoretical_loss": 3.747117108406973,
"tokens_seen": 762314752
},
{
"epoch": 0.23,
"learning_rate": 0.00038839518555667,
"loss": 3.0332,
"theoretical_loss": 3.747084897212069,
"tokens_seen": 762380288
},
{
"epoch": 0.23,
"learning_rate": 0.0003883851554663992,
"loss": 2.905,
"theoretical_loss": 3.747052689561227,
"tokens_seen": 762445824
},
{
"epoch": 0.23,
"learning_rate": 0.0003883751253761284,
"loss": 3.101,
"theoretical_loss": 3.74702048545375,
"tokens_seen": 762511360
},
{
"epoch": 0.23,
"learning_rate": 0.00038836509528585755,
"loss": 2.9868,
"theoretical_loss": 3.746988284888946,
"tokens_seen": 762576896
},
{
"epoch": 0.23,
"learning_rate": 0.0003883550651955868,
"loss": 2.635,
"theoretical_loss": 3.7469560878661197,
"tokens_seen": 762642432
},
{
"epoch": 0.23,
"learning_rate": 0.00038834503510531596,
"loss": 2.8003,
"theoretical_loss": 3.746923894384578,
"tokens_seen": 762707968
},
{
"epoch": 0.23,
"learning_rate": 0.00038833500501504514,
"loss": 3.1069,
"theoretical_loss": 3.7468917044436267,
"tokens_seen": 762773504
},
{
"epoch": 0.23,
"learning_rate": 0.0003883249749247743,
"loss": 2.8493,
"theoretical_loss": 3.7468595180425712,
"tokens_seen": 762839040
},
{
"epoch": 0.23,
"learning_rate": 0.0003883149448345035,
"loss": 2.8202,
"theoretical_loss": 3.74682733518072,
"tokens_seen": 762904576
},
{
"epoch": 0.23,
"learning_rate": 0.0003883049147442327,
"loss": 2.9543,
"theoretical_loss": 3.746795155857379,
"tokens_seen": 762970112
},
{
"epoch": 0.23,
"learning_rate": 0.0003882948846539619,
"loss": 2.8232,
"theoretical_loss": 3.746762980071855,
"tokens_seen": 763035648
},
{
"epoch": 0.23,
"learning_rate": 0.00038828485456369105,
"loss": 2.5409,
"theoretical_loss": 3.7467308078234565,
"tokens_seen": 763101184
},
{
"epoch": 0.23,
"learning_rate": 0.0003882748244734203,
"loss": 2.8501,
"theoretical_loss": 3.7466986391114894,
"tokens_seen": 763166720
},
{
"epoch": 0.23,
"learning_rate": 0.0003882647943831494,
"loss": 2.7522,
"theoretical_loss": 3.746666473935262,
"tokens_seen": 763232256
},
{
"epoch": 0.23,
"learning_rate": 0.00038825476429287865,
"loss": 2.6641,
"theoretical_loss": 3.7466343122940815,
"tokens_seen": 763297792
},
{
"epoch": 0.23,
"learning_rate": 0.00038824473420260783,
"loss": 3.0147,
"theoretical_loss": 3.7466021541872574,
"tokens_seen": 763363328
},
{
"epoch": 0.23,
"learning_rate": 0.000388234704112337,
"loss": 2.7999,
"theoretical_loss": 3.7465699996140964,
"tokens_seen": 763428864
},
{
"epoch": 0.23,
"objective/train/docs_used": 440741,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.76839280128479,
"objective/train/theoretical_loss": 3.746537848573908,
"objective/train/tokens_used": 783954400,
"theoretical_loss": 3.746537848573908,
"tokens_seen": 763494400
},
{
"epoch": 0.23,
"learning_rate": 0.0003882246740220662,
"loss": 2.7056,
"theoretical_loss": 3.746537848573908,
"tokens_seen": 763494400
},
{
"epoch": 0.23,
"learning_rate": 0.00038821464393179537,
"loss": 2.7497,
"theoretical_loss": 3.746505701066001,
"tokens_seen": 763559936
},
{
"epoch": 0.23,
"learning_rate": 0.00038820461384152455,
"loss": 2.7523,
"theoretical_loss": 3.7464735570896823,
"tokens_seen": 763625472
},
{
"epoch": 0.23,
"learning_rate": 0.0003881945837512538,
"loss": 2.8847,
"theoretical_loss": 3.746441416644263,
"tokens_seen": 763691008
},
{
"epoch": 0.23,
"learning_rate": 0.0003881845536609829,
"loss": 2.8288,
"theoretical_loss": 3.7464092797290505,
"tokens_seen": 763756544
},
{
"epoch": 0.23,
"learning_rate": 0.00038817452357071215,
"loss": 2.8737,
"theoretical_loss": 3.746377146343356,
"tokens_seen": 763822080
},
{
"epoch": 0.23,
"learning_rate": 0.00038816449348044133,
"loss": 2.6503,
"theoretical_loss": 3.746345016486488,
"tokens_seen": 763887616
},
{
"epoch": 0.23,
"learning_rate": 0.0003881544633901705,
"loss": 2.7654,
"theoretical_loss": 3.7463128901577565,
"tokens_seen": 763953152
},
{
"epoch": 0.23,
"learning_rate": 0.0003881444332998997,
"loss": 2.7917,
"theoretical_loss": 3.7462807673564713,
"tokens_seen": 764018688
},
{
"epoch": 0.23,
"learning_rate": 0.0003881344032096289,
"loss": 2.8016,
"theoretical_loss": 3.746248648081943,
"tokens_seen": 764084224
},
{
"epoch": 0.23,
"learning_rate": 0.0003881243731193581,
"loss": 2.6645,
"theoretical_loss": 3.7462165323334813,
"tokens_seen": 764149760
},
{
"epoch": 0.23,
"learning_rate": 0.0003881143430290873,
"loss": 3.0916,
"theoretical_loss": 3.746184420110397,
"tokens_seen": 764215296
},
{
"epoch": 0.23,
"learning_rate": 0.00038810431293881647,
"loss": 2.7536,
"theoretical_loss": 3.7461523114120014,
"tokens_seen": 764280832
},
{
"epoch": 0.23,
"learning_rate": 0.00038809428284854565,
"loss": 3.0446,
"theoretical_loss": 3.746120206237605,
"tokens_seen": 764346368
},
{
"epoch": 0.23,
"learning_rate": 0.00038808425275827483,
"loss": 2.968,
"theoretical_loss": 3.7460881045865184,
"tokens_seen": 764411904
},
{
"epoch": 0.23,
"learning_rate": 0.000388074222668004,
"loss": 2.785,
"theoretical_loss": 3.746056006458054,
"tokens_seen": 764477440
},
{
"epoch": 0.23,
"learning_rate": 0.00038806419257773325,
"loss": 2.9627,
"theoretical_loss": 3.7460239118515224,
"tokens_seen": 764542976
},
{
"epoch": 0.23,
"learning_rate": 0.0003880541624874624,
"loss": 2.7222,
"theoretical_loss": 3.745991820766236,
"tokens_seen": 764608512
},
{
"epoch": 0.23,
"learning_rate": 0.0003880441323971916,
"loss": 2.7322,
"theoretical_loss": 3.745959733201506,
"tokens_seen": 764674048
},
{
"epoch": 0.23,
"learning_rate": 0.0003880341023069208,
"loss": 2.8814,
"theoretical_loss": 3.7459276491566453,
"tokens_seen": 764739584
},
{
"epoch": 0.23,
"learning_rate": 0.00038802407221665,
"loss": 2.9449,
"theoretical_loss": 3.7458955686309654,
"tokens_seen": 764805120
},
{
"epoch": 0.23,
"learning_rate": 0.00038801404212637916,
"loss": 2.7037,
"theoretical_loss": 3.74586349162378,
"tokens_seen": 764870656
},
{
"epoch": 0.23,
"learning_rate": 0.00038800401203610834,
"loss": 2.8741,
"theoretical_loss": 3.7458314181344,
"tokens_seen": 764936192
},
{
"epoch": 0.23,
"learning_rate": 0.0003879939819458375,
"loss": 2.9241,
"theoretical_loss": 3.74579934816214,
"tokens_seen": 765001728
},
{
"epoch": 0.23,
"learning_rate": 0.00038798395185556675,
"loss": 2.4488,
"theoretical_loss": 3.7457672817063115,
"tokens_seen": 765067264
},
{
"epoch": 0.23,
"objective/train/docs_used": 441342,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6258957386016846,
"objective/train/theoretical_loss": 3.745735218766229,
"objective/train/tokens_used": 785592800,
"theoretical_loss": 3.745735218766229,
"tokens_seen": 765132800
},
{
"epoch": 0.23,
"learning_rate": 0.0003879739217652959,
"loss": 2.8597,
"theoretical_loss": 3.745735218766229,
"tokens_seen": 765132800
},
{
"epoch": 0.23,
"learning_rate": 0.0003879638916750251,
"loss": 2.8452,
"theoretical_loss": 3.745703159341205,
"tokens_seen": 765198336
},
{
"epoch": 0.23,
"learning_rate": 0.00038795386158475424,
"loss": 2.9551,
"theoretical_loss": 3.745671103430554,
"tokens_seen": 765263872
},
{
"epoch": 0.23,
"learning_rate": 0.0003879438314944835,
"loss": 2.6119,
"theoretical_loss": 3.7456390510335895,
"tokens_seen": 765329408
},
{
"epoch": 0.23,
"learning_rate": 0.00038793380140421266,
"loss": 2.779,
"theoretical_loss": 3.7456070021496255,
"tokens_seen": 765394944
},
{
"epoch": 0.23,
"learning_rate": 0.00038792377131394184,
"loss": 2.9343,
"theoretical_loss": 3.7455749567779764,
"tokens_seen": 765460480
},
{
"epoch": 0.23,
"learning_rate": 0.000387913741223671,
"loss": 2.8971,
"theoretical_loss": 3.745542914917956,
"tokens_seen": 765526016
},
{
"epoch": 0.23,
"learning_rate": 0.0003879037111334002,
"loss": 2.8223,
"theoretical_loss": 3.74551087656888,
"tokens_seen": 765591552
},
{
"epoch": 0.23,
"learning_rate": 0.0003878936810431294,
"loss": 2.7153,
"theoretical_loss": 3.745478841730062,
"tokens_seen": 765657088
},
{
"epoch": 0.23,
"learning_rate": 0.0003878836509528586,
"loss": 2.7312,
"theoretical_loss": 3.7454468104008174,
"tokens_seen": 765722624
},
{
"epoch": 0.23,
"learning_rate": 0.00038787362086258775,
"loss": 2.8997,
"theoretical_loss": 3.7454147825804616,
"tokens_seen": 765788160
},
{
"epoch": 0.23,
"learning_rate": 0.000387863590772317,
"loss": 2.7791,
"theoretical_loss": 3.74538275826831,
"tokens_seen": 765853696
},
{
"epoch": 0.23,
"learning_rate": 0.00038785356068204616,
"loss": 2.9699,
"theoretical_loss": 3.745350737463678,
"tokens_seen": 765919232
},
{
"epoch": 0.23,
"learning_rate": 0.00038784353059177534,
"loss": 2.8551,
"theoretical_loss": 3.7453187201658817,
"tokens_seen": 765984768
},
{
"epoch": 0.23,
"learning_rate": 0.0003878335005015045,
"loss": 2.6744,
"theoretical_loss": 3.7452867063742366,
"tokens_seen": 766050304
},
{
"epoch": 0.23,
"learning_rate": 0.0003878234704112337,
"loss": 2.8144,
"theoretical_loss": 3.7452546960880593,
"tokens_seen": 766115840
},
{
"epoch": 0.23,
"learning_rate": 0.0003878134403209629,
"loss": 2.8965,
"theoretical_loss": 3.745222689306665,
"tokens_seen": 766181376
},
{
"epoch": 0.23,
"learning_rate": 0.0003878034102306921,
"loss": 2.7609,
"theoretical_loss": 3.745190686029372,
"tokens_seen": 766246912
},
{
"epoch": 0.23,
"learning_rate": 0.00038779338014042125,
"loss": 2.6836,
"theoretical_loss": 3.745158686255496,
"tokens_seen": 766312448
},
{
"epoch": 0.23,
"learning_rate": 0.0003877833500501505,
"loss": 2.6351,
"theoretical_loss": 3.7451266899843536,
"tokens_seen": 766377984
},
{
"epoch": 0.23,
"learning_rate": 0.0003877733199598796,
"loss": 3.0103,
"theoretical_loss": 3.7450946972152623,
"tokens_seen": 766443520
},
{
"epoch": 0.23,
"learning_rate": 0.00038776328986960885,
"loss": 2.7607,
"theoretical_loss": 3.7450627079475396,
"tokens_seen": 766509056
},
{
"epoch": 0.23,
"learning_rate": 0.00038775325977933803,
"loss": 2.7748,
"theoretical_loss": 3.745030722180503,
"tokens_seen": 766574592
},
{
"epoch": 0.23,
"learning_rate": 0.0003877432296890672,
"loss": 3.0818,
"theoretical_loss": 3.74499873991347,
"tokens_seen": 766640128
},
{
"epoch": 0.23,
"learning_rate": 0.0003877331995987964,
"loss": 2.9272,
"theoretical_loss": 3.7449667611457578,
"tokens_seen": 766705664
},
{
"epoch": 0.23,
"objective/train/docs_used": 442736,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3047845363616943,
"objective/train/theoretical_loss": 3.744934785876686,
"objective/train/tokens_used": 787231200,
"theoretical_loss": 3.744934785876686,
"tokens_seen": 766771200
},
{
"epoch": 0.23,
"learning_rate": 0.00038772316950852557,
"loss": 2.6242,
"theoretical_loss": 3.744934785876686,
"tokens_seen": 766771200
},
{
"epoch": 0.23,
"learning_rate": 0.00038771313941825475,
"loss": 2.7337,
"theoretical_loss": 3.744902814105571,
"tokens_seen": 766836736
},
{
"epoch": 0.23,
"learning_rate": 0.000387703109327984,
"loss": 3.07,
"theoretical_loss": 3.7448708458317332,
"tokens_seen": 766902272
},
{
"epoch": 0.23,
"learning_rate": 0.0003876930792377131,
"loss": 2.9027,
"theoretical_loss": 3.7448388810544895,
"tokens_seen": 766967808
},
{
"epoch": 0.23,
"learning_rate": 0.00038768304914744235,
"loss": 2.8153,
"theoretical_loss": 3.7448069197731595,
"tokens_seen": 767033344
},
{
"epoch": 0.23,
"learning_rate": 0.00038767301905717153,
"loss": 2.8659,
"theoretical_loss": 3.744774961987063,
"tokens_seen": 767098880
},
{
"epoch": 0.23,
"learning_rate": 0.0003876629889669007,
"loss": 2.7751,
"theoretical_loss": 3.7447430076955177,
"tokens_seen": 767164416
},
{
"epoch": 0.23,
"learning_rate": 0.0003876529588766299,
"loss": 2.6712,
"theoretical_loss": 3.7447110568978434,
"tokens_seen": 767229952
},
{
"epoch": 0.23,
"learning_rate": 0.0003876429287863591,
"loss": 2.6803,
"theoretical_loss": 3.7446791095933603,
"tokens_seen": 767295488
},
{
"epoch": 0.23,
"learning_rate": 0.00038763289869608826,
"loss": 2.7714,
"theoretical_loss": 3.7446471657813882,
"tokens_seen": 767361024
},
{
"epoch": 0.23,
"learning_rate": 0.0003876228686058175,
"loss": 3.091,
"theoretical_loss": 3.744615225461246,
"tokens_seen": 767426560
},
{
"epoch": 0.23,
"learning_rate": 0.0003876128385155466,
"loss": 2.9273,
"theoretical_loss": 3.744583288632255,
"tokens_seen": 767492096
},
{
"epoch": 0.23,
"learning_rate": 0.00038760280842527585,
"loss": 2.6865,
"theoretical_loss": 3.7445513552937353,
"tokens_seen": 767557632
},
{
"epoch": 0.23,
"learning_rate": 0.000387592778335005,
"loss": 2.7196,
"theoretical_loss": 3.7445194254450067,
"tokens_seen": 767623168
},
{
"epoch": 0.23,
"learning_rate": 0.0003875827482447342,
"loss": 2.689,
"theoretical_loss": 3.7444874990853907,
"tokens_seen": 767688704
},
{
"epoch": 0.23,
"learning_rate": 0.0003875727181544634,
"loss": 2.8977,
"theoretical_loss": 3.744455576214208,
"tokens_seen": 767754240
},
{
"epoch": 0.23,
"learning_rate": 0.0003875626880641926,
"loss": 2.7307,
"theoretical_loss": 3.7444236568307803,
"tokens_seen": 767819776
},
{
"epoch": 0.23,
"learning_rate": 0.00038755265797392176,
"loss": 2.7366,
"theoretical_loss": 3.7443917409344274,
"tokens_seen": 767885312
},
{
"epoch": 0.23,
"learning_rate": 0.000387542627883651,
"loss": 2.745,
"theoretical_loss": 3.7443598285244724,
"tokens_seen": 767950848
},
{
"epoch": 0.23,
"learning_rate": 0.0003875325977933801,
"loss": 3.1207,
"theoretical_loss": 3.7443279196002357,
"tokens_seen": 768016384
},
{
"epoch": 0.23,
"learning_rate": 0.00038752256770310936,
"loss": 2.8191,
"theoretical_loss": 3.74429601416104,
"tokens_seen": 768081920
},
{
"epoch": 0.23,
"learning_rate": 0.0003875125376128385,
"loss": 2.8704,
"theoretical_loss": 3.7442641122062073,
"tokens_seen": 768147456
},
{
"epoch": 0.23,
"learning_rate": 0.0003875025075225677,
"loss": 2.8826,
"theoretical_loss": 3.7442322137350597,
"tokens_seen": 768212992
},
{
"epoch": 0.23,
"learning_rate": 0.0003874924774322969,
"loss": 2.8508,
"theoretical_loss": 3.7442003187469193,
"tokens_seen": 768278528
},
{
"epoch": 0.23,
"learning_rate": 0.0003874824473420261,
"loss": 2.8384,
"theoretical_loss": 3.7441684272411093,
"tokens_seen": 768344064
},
{
"epoch": 0.23,
"objective/train/docs_used": 443303,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.697742223739624,
"objective/train/theoretical_loss": 3.744136539216952,
"objective/train/tokens_used": 788869600,
"theoretical_loss": 3.744136539216952,
"tokens_seen": 768409600
},
{
"epoch": 0.23,
"learning_rate": 0.00038747241725175526,
"loss": 2.5811,
"theoretical_loss": 3.744136539216952,
"tokens_seen": 768409600
},
{
"epoch": 0.23,
"learning_rate": 0.00038746238716148444,
"loss": 2.7877,
"theoretical_loss": 3.744104654673771,
"tokens_seen": 768475136
},
{
"epoch": 0.23,
"learning_rate": 0.0003874523570712136,
"loss": 2.5555,
"theoretical_loss": 3.7440727736108887,
"tokens_seen": 768540672
},
{
"epoch": 0.23,
"learning_rate": 0.00038744232698094286,
"loss": 2.7894,
"theoretical_loss": 3.744040896027629,
"tokens_seen": 768606208
},
{
"epoch": 0.23,
"learning_rate": 0.000387432296890672,
"loss": 2.9867,
"theoretical_loss": 3.7440090219233157,
"tokens_seen": 768671744
},
{
"epoch": 0.23,
"learning_rate": 0.0003874222668004012,
"loss": 2.8178,
"theoretical_loss": 3.7439771512972726,
"tokens_seen": 768737280
},
{
"epoch": 0.23,
"learning_rate": 0.00038741223671013035,
"loss": 2.9102,
"theoretical_loss": 3.7439452841488228,
"tokens_seen": 768802816
},
{
"epoch": 0.23,
"learning_rate": 0.0003874022066198596,
"loss": 2.7569,
"theoretical_loss": 3.7439134204772913,
"tokens_seen": 768868352
},
{
"epoch": 0.23,
"learning_rate": 0.00038739217652958876,
"loss": 2.7005,
"theoretical_loss": 3.7438815602820017,
"tokens_seen": 768933888
},
{
"epoch": 0.23,
"learning_rate": 0.00038738214643931795,
"loss": 2.7511,
"theoretical_loss": 3.743849703562279,
"tokens_seen": 768999424
},
{
"epoch": 0.23,
"learning_rate": 0.0003873721163490472,
"loss": 2.7152,
"theoretical_loss": 3.743817850317448,
"tokens_seen": 769064960
},
{
"epoch": 0.23,
"learning_rate": 0.00038736208625877636,
"loss": 2.5626,
"theoretical_loss": 3.7437860005468337,
"tokens_seen": 769130496
},
{
"epoch": 0.23,
"learning_rate": 0.00038735205616850554,
"loss": 2.8514,
"theoretical_loss": 3.7437541542497605,
"tokens_seen": 769196032
},
{
"epoch": 0.23,
"learning_rate": 0.0003873420260782347,
"loss": 2.9543,
"theoretical_loss": 3.7437223114255542,
"tokens_seen": 769261568
},
{
"epoch": 0.23,
"learning_rate": 0.0003873319959879639,
"loss": 2.6382,
"theoretical_loss": 3.74369047207354,
"tokens_seen": 769327104
},
{
"epoch": 0.23,
"learning_rate": 0.0003873219658976931,
"loss": 2.9459,
"theoretical_loss": 3.7436586361930444,
"tokens_seen": 769392640
},
{
"epoch": 0.23,
"learning_rate": 0.0003873119358074223,
"loss": 2.7832,
"theoretical_loss": 3.743626803783392,
"tokens_seen": 769458176
},
{
"epoch": 0.23,
"learning_rate": 0.00038730190571715145,
"loss": 3.03,
"theoretical_loss": 3.7435949748439095,
"tokens_seen": 769523712
},
{
"epoch": 0.23,
"learning_rate": 0.0003872918756268807,
"loss": 2.7303,
"theoretical_loss": 3.743563149373923,
"tokens_seen": 769589248
},
{
"epoch": 0.23,
"learning_rate": 0.0003872818455366098,
"loss": 2.6207,
"theoretical_loss": 3.743531327372759,
"tokens_seen": 769654784
},
{
"epoch": 0.23,
"learning_rate": 0.00038727181544633905,
"loss": 2.7811,
"theoretical_loss": 3.743499508839744,
"tokens_seen": 769720320
},
{
"epoch": 0.23,
"learning_rate": 0.00038726178535606823,
"loss": 2.5147,
"theoretical_loss": 3.7434676937742046,
"tokens_seen": 769785856
},
{
"epoch": 0.23,
"learning_rate": 0.0003872517552657974,
"loss": 2.945,
"theoretical_loss": 3.743435882175468,
"tokens_seen": 769851392
},
{
"epoch": 0.23,
"learning_rate": 0.0003872417251755266,
"loss": 2.9277,
"theoretical_loss": 3.7434040740428616,
"tokens_seen": 769916928
},
{
"epoch": 0.23,
"learning_rate": 0.00038723169508525577,
"loss": 2.8227,
"theoretical_loss": 3.743372269375712,
"tokens_seen": 769982464
},
{
"epoch": 0.23,
"objective/train/docs_used": 444724,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.963942527770996,
"objective/train/theoretical_loss": 3.7433404681733475,
"objective/train/tokens_used": 790508000,
"theoretical_loss": 3.7433404681733475,
"tokens_seen": 770048000
},
{
"epoch": 0.23,
"learning_rate": 0.00038722166499498495,
"loss": 2.5265,
"theoretical_loss": 3.7433404681733475,
"tokens_seen": 770048000
},
{
"epoch": 0.23,
"learning_rate": 0.0003872116349047142,
"loss": 2.7182,
"theoretical_loss": 3.7433086704350957,
"tokens_seen": 770113536
},
{
"epoch": 0.23,
"learning_rate": 0.0003872016048144433,
"loss": 2.9499,
"theoretical_loss": 3.7432768761602837,
"tokens_seen": 770179072
},
{
"epoch": 0.23,
"learning_rate": 0.00038719157472417255,
"loss": 2.6917,
"theoretical_loss": 3.743245085348241,
"tokens_seen": 770244608
},
{
"epoch": 0.23,
"learning_rate": 0.00038718154463390173,
"loss": 2.866,
"theoretical_loss": 3.7432132979982944,
"tokens_seen": 770310144
},
{
"epoch": 0.23,
"learning_rate": 0.0003871715145436309,
"loss": 2.6999,
"theoretical_loss": 3.7431815141097733,
"tokens_seen": 770375680
},
{
"epoch": 0.23,
"learning_rate": 0.0003871614844533601,
"loss": 2.9155,
"theoretical_loss": 3.7431497336820065,
"tokens_seen": 770441216
},
{
"epoch": 0.23,
"learning_rate": 0.0003871514543630893,
"loss": 2.8286,
"theoretical_loss": 3.743117956714322,
"tokens_seen": 770506752
},
{
"epoch": 0.23,
"learning_rate": 0.00038714142427281846,
"loss": 2.6173,
"theoretical_loss": 3.7430861832060494,
"tokens_seen": 770572288
},
{
"epoch": 0.23,
"learning_rate": 0.0003871313941825477,
"loss": 2.5913,
"theoretical_loss": 3.7430544131565178,
"tokens_seen": 770637824
},
{
"epoch": 0.23,
"learning_rate": 0.0003871213640922768,
"loss": 2.5277,
"theoretical_loss": 3.7430226465650573,
"tokens_seen": 770703360
},
{
"epoch": 0.23,
"learning_rate": 0.00038711133400200605,
"loss": 2.947,
"theoretical_loss": 3.7429908834309957,
"tokens_seen": 770768896
},
{
"epoch": 0.23,
"learning_rate": 0.0003871013039117352,
"loss": 2.9524,
"theoretical_loss": 3.742959123753665,
"tokens_seen": 770834432
},
{
"epoch": 0.23,
"learning_rate": 0.0003870912738214644,
"loss": 2.8162,
"theoretical_loss": 3.7429273675323933,
"tokens_seen": 770899968
},
{
"epoch": 0.23,
"learning_rate": 0.0003870812437311936,
"loss": 2.7862,
"theoretical_loss": 3.742895614766512,
"tokens_seen": 770965504
},
{
"epoch": 0.23,
"learning_rate": 0.0003870712136409228,
"loss": 2.809,
"theoretical_loss": 3.7428638654553508,
"tokens_seen": 771031040
},
{
"epoch": 0.23,
"learning_rate": 0.00038706118355065196,
"loss": 2.751,
"theoretical_loss": 3.74283211959824,
"tokens_seen": 771096576
},
{
"epoch": 0.23,
"learning_rate": 0.0003870511534603812,
"loss": 2.8002,
"theoretical_loss": 3.7428003771945115,
"tokens_seen": 771162112
},
{
"epoch": 0.23,
"learning_rate": 0.0003870411233701103,
"loss": 2.6938,
"theoretical_loss": 3.7427686382434953,
"tokens_seen": 771227648
},
{
"epoch": 0.23,
"learning_rate": 0.00038703109327983956,
"loss": 2.9299,
"theoretical_loss": 3.742736902744522,
"tokens_seen": 771293184
},
{
"epoch": 0.23,
"learning_rate": 0.0003870210631895687,
"loss": 2.8176,
"theoretical_loss": 3.742705170696924,
"tokens_seen": 771358720
},
{
"epoch": 0.23,
"learning_rate": 0.0003870110330992979,
"loss": 2.8464,
"theoretical_loss": 3.742673442100032,
"tokens_seen": 771424256
},
{
"epoch": 0.23,
"learning_rate": 0.0003870010030090271,
"loss": 2.7446,
"theoretical_loss": 3.7426417169531776,
"tokens_seen": 771489792
},
{
"epoch": 0.23,
"learning_rate": 0.0003869909729187563,
"loss": 2.8558,
"theoretical_loss": 3.7426099952556937,
"tokens_seen": 771555328
},
{
"epoch": 0.23,
"learning_rate": 0.00038698094282848546,
"loss": 2.7574,
"theoretical_loss": 3.742578277006911,
"tokens_seen": 771620864
},
{
"epoch": 0.23,
"objective/train/docs_used": 445347,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0361459255218506,
"objective/train/theoretical_loss": 3.7425465622061624,
"objective/train/tokens_used": 792146400,
"theoretical_loss": 3.7425465622061624,
"tokens_seen": 771686400
},
{
"epoch": 0.23,
"learning_rate": 0.00038697091273821464,
"loss": 2.8847,
"theoretical_loss": 3.7425465622061624,
"tokens_seen": 771686400
},
{
"epoch": 0.23,
"learning_rate": 0.0003869608826479438,
"loss": 2.9605,
"theoretical_loss": 3.74251485085278,
"tokens_seen": 771751936
},
{
"epoch": 0.23,
"learning_rate": 0.00038695085255767306,
"loss": 2.8552,
"theoretical_loss": 3.7424831429460967,
"tokens_seen": 771817472
},
{
"epoch": 0.23,
"learning_rate": 0.0003869408224674022,
"loss": 2.735,
"theoretical_loss": 3.7424514384854453,
"tokens_seen": 771883008
},
{
"epoch": 0.23,
"learning_rate": 0.0003869307923771314,
"loss": 2.8366,
"theoretical_loss": 3.742419737470158,
"tokens_seen": 771948544
},
{
"epoch": 0.23,
"learning_rate": 0.00038692076228686055,
"loss": 2.8744,
"theoretical_loss": 3.742388039899569,
"tokens_seen": 772014080
},
{
"epoch": 0.23,
"learning_rate": 0.0003869107321965898,
"loss": 2.7767,
"theoretical_loss": 3.742356345773011,
"tokens_seen": 772079616
},
{
"epoch": 0.23,
"learning_rate": 0.00038690070210631896,
"loss": 2.7706,
"theoretical_loss": 3.742324655089817,
"tokens_seen": 772145152
},
{
"epoch": 0.23,
"learning_rate": 0.00038689067201604815,
"loss": 2.5963,
"theoretical_loss": 3.742292967849322,
"tokens_seen": 772210688
},
{
"epoch": 0.23,
"learning_rate": 0.00038688064192577733,
"loss": 3.016,
"theoretical_loss": 3.742261284050859,
"tokens_seen": 772276224
},
{
"epoch": 0.23,
"learning_rate": 0.00038687061183550656,
"loss": 2.6511,
"theoretical_loss": 3.7422296036937617,
"tokens_seen": 772341760
},
{
"epoch": 0.23,
"learning_rate": 0.0003868605817452357,
"loss": 2.5635,
"theoretical_loss": 3.7421979267773655,
"tokens_seen": 772407296
},
{
"epoch": 0.23,
"learning_rate": 0.0003868505516549649,
"loss": 2.6261,
"theoretical_loss": 3.742166253301004,
"tokens_seen": 772472832
},
{
"epoch": 0.23,
"learning_rate": 0.00038684052156469405,
"loss": 2.7541,
"theoretical_loss": 3.7421345832640123,
"tokens_seen": 772538368
},
{
"epoch": 0.23,
"learning_rate": 0.0003868304914744233,
"loss": 3.0098,
"theoretical_loss": 3.7421029166657243,
"tokens_seen": 772603904
},
{
"epoch": 0.23,
"learning_rate": 0.00038682046138415247,
"loss": 2.8456,
"theoretical_loss": 3.7420712535054763,
"tokens_seen": 772669440
},
{
"epoch": 0.23,
"learning_rate": 0.00038681043129388165,
"loss": 2.8107,
"theoretical_loss": 3.742039593782602,
"tokens_seen": 772734976
},
{
"epoch": 0.23,
"learning_rate": 0.00038680040120361083,
"loss": 2.936,
"theoretical_loss": 3.742007937496438,
"tokens_seen": 772800512
},
{
"epoch": 0.23,
"learning_rate": 0.00038679037111334,
"loss": 2.7138,
"theoretical_loss": 3.7419762846463196,
"tokens_seen": 772866048
},
{
"epoch": 0.23,
"learning_rate": 0.0003867803410230692,
"loss": 2.7256,
"theoretical_loss": 3.741944635231582,
"tokens_seen": 772931584
},
{
"epoch": 0.23,
"learning_rate": 0.00038677031093279843,
"loss": 2.9617,
"theoretical_loss": 3.7419129892515617,
"tokens_seen": 772997120
},
{
"epoch": 0.23,
"learning_rate": 0.00038676028084252755,
"loss": 2.8077,
"theoretical_loss": 3.7418813467055942,
"tokens_seen": 773062656
},
{
"epoch": 0.23,
"learning_rate": 0.0003867502507522568,
"loss": 2.8191,
"theoretical_loss": 3.7418497075930164,
"tokens_seen": 773128192
},
{
"epoch": 0.23,
"learning_rate": 0.0003867402206619859,
"loss": 2.7094,
"theoretical_loss": 3.741818071913164,
"tokens_seen": 773193728
},
{
"epoch": 0.23,
"learning_rate": 0.00038673019057171515,
"loss": 2.6635,
"theoretical_loss": 3.7417864396653746,
"tokens_seen": 773259264
},
{
"epoch": 0.23,
"objective/train/docs_used": 446623,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.632535219192505,
"objective/train/theoretical_loss": 3.7417548108489846,
"objective/train/tokens_used": 793784800,
"theoretical_loss": 3.7417548108489846,
"tokens_seen": 773324800
},
{
"epoch": 0.23,
"learning_rate": 0.00038672016048144433,
"loss": 3.0971,
"theoretical_loss": 3.7417548108489846,
"tokens_seen": 773324800
},
{
"epoch": 0.23,
"learning_rate": 0.0003867101303911735,
"loss": 2.8281,
"theoretical_loss": 3.741723185463331,
"tokens_seen": 773390336
},
{
"epoch": 0.23,
"learning_rate": 0.0003867001003009027,
"loss": 2.852,
"theoretical_loss": 3.7416915635077506,
"tokens_seen": 773455872
},
{
"epoch": 0.23,
"learning_rate": 0.00038669007021063193,
"loss": 2.8355,
"theoretical_loss": 3.7416599449815813,
"tokens_seen": 773521408
},
{
"epoch": 0.23,
"learning_rate": 0.00038668004012036106,
"loss": 2.8096,
"theoretical_loss": 3.7416283298841604,
"tokens_seen": 773586944
},
{
"epoch": 0.23,
"learning_rate": 0.0003866700100300903,
"loss": 2.8216,
"theoretical_loss": 3.7415967182148266,
"tokens_seen": 773652480
},
{
"epoch": 0.23,
"learning_rate": 0.0003866599799398194,
"loss": 2.942,
"theoretical_loss": 3.7415651099729166,
"tokens_seen": 773718016
},
{
"epoch": 0.23,
"learning_rate": 0.00038664994984954866,
"loss": 2.8512,
"theoretical_loss": 3.7415335051577685,
"tokens_seen": 773783552
},
{
"epoch": 0.23,
"learning_rate": 0.00038663991975927784,
"loss": 2.7113,
"theoretical_loss": 3.7415019037687216,
"tokens_seen": 773849088
},
{
"epoch": 0.23,
"learning_rate": 0.000386629889669007,
"loss": 2.7419,
"theoretical_loss": 3.7414703058051137,
"tokens_seen": 773914624
},
{
"epoch": 0.23,
"learning_rate": 0.00038661985957873625,
"loss": 2.9951,
"theoretical_loss": 3.7414387112662837,
"tokens_seen": 773980160
},
{
"epoch": 0.23,
"learning_rate": 0.0003866098294884654,
"loss": 2.7494,
"theoretical_loss": 3.74140712015157,
"tokens_seen": 774045696
},
{
"epoch": 0.23,
"learning_rate": 0.0003865997993981946,
"loss": 2.8323,
"theoretical_loss": 3.741375532460313,
"tokens_seen": 774111232
},
{
"epoch": 0.23,
"learning_rate": 0.0003865897693079238,
"loss": 2.9304,
"theoretical_loss": 3.74134394819185,
"tokens_seen": 774176768
},
{
"epoch": 0.23,
"learning_rate": 0.000386579739217653,
"loss": 2.6338,
"theoretical_loss": 3.741312367345522,
"tokens_seen": 774242304
},
{
"epoch": 0.23,
"learning_rate": 0.00038656970912738216,
"loss": 2.7479,
"theoretical_loss": 3.7412807899206677,
"tokens_seen": 774307840
},
{
"epoch": 0.23,
"learning_rate": 0.0003865596790371114,
"loss": 2.788,
"theoretical_loss": 3.7412492159166275,
"tokens_seen": 774373376
},
{
"epoch": 0.23,
"learning_rate": 0.0003865496489468405,
"loss": 3.0283,
"theoretical_loss": 3.741217645332741,
"tokens_seen": 774438912
},
{
"epoch": 0.23,
"learning_rate": 0.00038653961885656976,
"loss": 2.7545,
"theoretical_loss": 3.7411860781683477,
"tokens_seen": 774504448
},
{
"epoch": 0.23,
"learning_rate": 0.0003865295887662989,
"loss": 2.8885,
"theoretical_loss": 3.741154514422789,
"tokens_seen": 774569984
},
{
"epoch": 0.23,
"learning_rate": 0.0003865195586760281,
"loss": 2.9714,
"theoretical_loss": 3.741122954095405,
"tokens_seen": 774635520
},
{
"epoch": 0.23,
"learning_rate": 0.0003865095285857573,
"loss": 2.8065,
"theoretical_loss": 3.7410913971855364,
"tokens_seen": 774701056
},
{
"epoch": 0.23,
"learning_rate": 0.0003864994984954865,
"loss": 2.8563,
"theoretical_loss": 3.7410598436925238,
"tokens_seen": 774766592
},
{
"epoch": 0.23,
"learning_rate": 0.00038648946840521566,
"loss": 2.675,
"theoretical_loss": 3.7410282936157087,
"tokens_seen": 774832128
},
{
"epoch": 0.23,
"learning_rate": 0.00038647943831494484,
"loss": 3.0078,
"theoretical_loss": 3.7409967469544325,
"tokens_seen": 774897664
},
{
"epoch": 0.23,
"objective/train/docs_used": 447277,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8082783222198486,
"objective/train/theoretical_loss": 3.7409652037080363,
"objective/train/tokens_used": 795423200,
"theoretical_loss": 3.7409652037080363,
"tokens_seen": 774963200
},
{
"epoch": 0.23,
"learning_rate": 0.000386469408224674,
"loss": 2.8857,
"theoretical_loss": 3.7409652037080363,
"tokens_seen": 774963200
},
{
"epoch": 0.23,
"learning_rate": 0.00038645937813440326,
"loss": 2.7649,
"theoretical_loss": 3.740933663875861,
"tokens_seen": 775028736
},
{
"epoch": 0.23,
"learning_rate": 0.0003864493480441324,
"loss": 2.8994,
"theoretical_loss": 3.7409021274572494,
"tokens_seen": 775094272
},
{
"epoch": 0.23,
"learning_rate": 0.0003864393179538616,
"loss": 2.8243,
"theoretical_loss": 3.7408705944515432,
"tokens_seen": 775159808
},
{
"epoch": 0.23,
"learning_rate": 0.00038642928786359075,
"loss": 2.7318,
"theoretical_loss": 3.740839064858085,
"tokens_seen": 775225344
},
{
"epoch": 0.23,
"learning_rate": 0.00038641925777332,
"loss": 2.8808,
"theoretical_loss": 3.740807538676216,
"tokens_seen": 775290880
},
{
"epoch": 0.23,
"learning_rate": 0.00038640922768304917,
"loss": 2.9177,
"theoretical_loss": 3.7407760159052796,
"tokens_seen": 775356416
},
{
"epoch": 0.23,
"learning_rate": 0.00038639919759277835,
"loss": 2.7471,
"theoretical_loss": 3.740744496544618,
"tokens_seen": 775421952
},
{
"epoch": 0.23,
"learning_rate": 0.00038638916750250753,
"loss": 2.7278,
"theoretical_loss": 3.7407129805935746,
"tokens_seen": 775487488
},
{
"epoch": 0.24,
"learning_rate": 0.00038637913741223676,
"loss": 2.925,
"theoretical_loss": 3.740681468051492,
"tokens_seen": 775553024
},
{
"epoch": 0.24,
"learning_rate": 0.0003863691073219659,
"loss": 2.7908,
"theoretical_loss": 3.740649958917714,
"tokens_seen": 775618560
},
{
"epoch": 0.24,
"learning_rate": 0.0003863590772316951,
"loss": 2.9113,
"theoretical_loss": 3.7406184531915834,
"tokens_seen": 775684096
},
{
"epoch": 0.24,
"learning_rate": 0.00038634904714142425,
"loss": 2.8416,
"theoretical_loss": 3.7405869508724443,
"tokens_seen": 775749632
},
{
"epoch": 0.24,
"learning_rate": 0.0003863390170511535,
"loss": 2.614,
"theoretical_loss": 3.74055545195964,
"tokens_seen": 775815168
},
{
"epoch": 0.24,
"learning_rate": 0.00038632898696088267,
"loss": 2.9536,
"theoretical_loss": 3.740523956452515,
"tokens_seen": 775880704
},
{
"epoch": 0.24,
"learning_rate": 0.00038631895687061185,
"loss": 2.8925,
"theoretical_loss": 3.7404924643504134,
"tokens_seen": 775946240
},
{
"epoch": 0.24,
"learning_rate": 0.00038630892678034103,
"loss": 2.9184,
"theoretical_loss": 3.740460975652679,
"tokens_seen": 776011776
},
{
"epoch": 0.24,
"learning_rate": 0.0003862988966900702,
"loss": 2.493,
"theoretical_loss": 3.740429490358657,
"tokens_seen": 776077312
},
{
"epoch": 0.24,
"learning_rate": 0.0003862888665997994,
"loss": 2.7677,
"theoretical_loss": 3.7403980084676918,
"tokens_seen": 776142848
},
{
"epoch": 0.24,
"learning_rate": 0.00038627883650952863,
"loss": 2.6157,
"theoretical_loss": 3.740366529979128,
"tokens_seen": 776208384
},
{
"epoch": 0.24,
"learning_rate": 0.00038626880641925775,
"loss": 2.9094,
"theoretical_loss": 3.740335054892311,
"tokens_seen": 776273920
},
{
"epoch": 0.24,
"learning_rate": 0.000386258776328987,
"loss": 2.9608,
"theoretical_loss": 3.740303583206586,
"tokens_seen": 776339456
},
{
"epoch": 0.24,
"learning_rate": 0.0003862487462387161,
"loss": 2.6947,
"theoretical_loss": 3.7402721149212983,
"tokens_seen": 776404992
},
{
"epoch": 0.24,
"learning_rate": 0.00038623871614844535,
"loss": 2.7206,
"theoretical_loss": 3.7402406500357936,
"tokens_seen": 776470528
},
{
"epoch": 0.24,
"learning_rate": 0.00038622868605817453,
"loss": 2.7929,
"theoretical_loss": 3.740209188549418,
"tokens_seen": 776536064
},
{
"epoch": 0.24,
"objective/train/docs_used": 447858,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.078845500946045,
"objective/train/theoretical_loss": 3.740177730461517,
"objective/train/tokens_used": 797061600,
"theoretical_loss": 3.740177730461517,
"tokens_seen": 776601600
},
{
"epoch": 0.24,
"learning_rate": 0.0003862186559679037,
"loss": 3.0116,
"theoretical_loss": 3.740177730461517,
"tokens_seen": 776601600
},
{
"epoch": 0.24,
"learning_rate": 0.0003862086258776329,
"loss": 2.7036,
"theoretical_loss": 3.740146275771437,
"tokens_seen": 776667136
},
{
"epoch": 0.24,
"learning_rate": 0.00038619859578736213,
"loss": 2.5406,
"theoretical_loss": 3.740114824478524,
"tokens_seen": 776732672
},
{
"epoch": 0.24,
"learning_rate": 0.00038618856569709126,
"loss": 2.683,
"theoretical_loss": 3.740083376582125,
"tokens_seen": 776798208
},
{
"epoch": 0.24,
"learning_rate": 0.0003861785356068205,
"loss": 2.5957,
"theoretical_loss": 3.7400519320815864,
"tokens_seen": 776863744
},
{
"epoch": 0.24,
"learning_rate": 0.0003861685055165496,
"loss": 2.646,
"theoretical_loss": 3.740020490976255,
"tokens_seen": 776929280
},
{
"epoch": 0.24,
"learning_rate": 0.00038615847542627886,
"loss": 2.8098,
"theoretical_loss": 3.7399890532654783,
"tokens_seen": 776994816
},
{
"epoch": 0.24,
"learning_rate": 0.00038614844533600804,
"loss": 2.8436,
"theoretical_loss": 3.739957618948603,
"tokens_seen": 777060352
},
{
"epoch": 0.24,
"learning_rate": 0.0003861384152457372,
"loss": 2.8699,
"theoretical_loss": 3.7399261880249766,
"tokens_seen": 777125888
},
{
"epoch": 0.24,
"learning_rate": 0.0003861283851554664,
"loss": 2.9615,
"theoretical_loss": 3.7398947604939474,
"tokens_seen": 777191424
},
{
"epoch": 0.24,
"learning_rate": 0.0003861183550651956,
"loss": 2.7857,
"theoretical_loss": 3.739863336354862,
"tokens_seen": 777256960
},
{
"epoch": 0.24,
"learning_rate": 0.00038610832497492476,
"loss": 2.8816,
"theoretical_loss": 3.7398319156070694,
"tokens_seen": 777322496
},
{
"epoch": 0.24,
"learning_rate": 0.000386098294884654,
"loss": 2.7589,
"theoretical_loss": 3.739800498249917,
"tokens_seen": 777388032
},
{
"epoch": 0.24,
"learning_rate": 0.0003860882647943831,
"loss": 2.7221,
"theoretical_loss": 3.7397690842827536,
"tokens_seen": 777453568
},
{
"epoch": 0.24,
"learning_rate": 0.00038607823470411236,
"loss": 2.9624,
"theoretical_loss": 3.7397376737049277,
"tokens_seen": 777519104
},
{
"epoch": 0.24,
"learning_rate": 0.0003860682046138415,
"loss": 2.8394,
"theoretical_loss": 3.7397062665157876,
"tokens_seen": 777584640
},
{
"epoch": 0.24,
"learning_rate": 0.0003860581745235707,
"loss": 2.8596,
"theoretical_loss": 3.7396748627146827,
"tokens_seen": 777650176
},
{
"epoch": 0.24,
"learning_rate": 0.0003860481444332999,
"loss": 2.7443,
"theoretical_loss": 3.739643462300961,
"tokens_seen": 777715712
},
{
"epoch": 0.24,
"learning_rate": 0.0003860381143430291,
"loss": 2.6521,
"theoretical_loss": 3.739612065273973,
"tokens_seen": 777781248
},
{
"epoch": 0.24,
"learning_rate": 0.00038602808425275826,
"loss": 2.5931,
"theoretical_loss": 3.7395806716330675,
"tokens_seen": 777846784
},
{
"epoch": 0.24,
"learning_rate": 0.0003860180541624875,
"loss": 2.7803,
"theoretical_loss": 3.7395492813775943,
"tokens_seen": 777912320
},
{
"epoch": 0.24,
"learning_rate": 0.0003860080240722166,
"loss": 2.882,
"theoretical_loss": 3.7395178945069025,
"tokens_seen": 777977856
},
{
"epoch": 0.24,
"learning_rate": 0.00038599799398194586,
"loss": 2.7028,
"theoretical_loss": 3.739486511020343,
"tokens_seen": 778043392
},
{
"epoch": 0.24,
"learning_rate": 0.000385987963891675,
"loss": 2.6532,
"theoretical_loss": 3.739455130917265,
"tokens_seen": 778108928
},
{
"epoch": 0.24,
"learning_rate": 0.0003859779338014042,
"loss": 2.6865,
"theoretical_loss": 3.739423754197019,
"tokens_seen": 778174464
},
{
"epoch": 0.24,
"objective/train/docs_used": 448758,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.072265386581421,
"objective/train/theoretical_loss": 3.7393923808589564,
"objective/train/tokens_used": 798700000,
"theoretical_loss": 3.7393923808589564,
"tokens_seen": 778240000
},
{
"epoch": 0.24,
"learning_rate": 0.0003859679037111334,
"loss": 3.0101,
"theoretical_loss": 3.7393923808589564,
"tokens_seen": 778240000
},
{
"epoch": 0.24,
"learning_rate": 0.0003859578736208626,
"loss": 2.7858,
"theoretical_loss": 3.7393610109024267,
"tokens_seen": 778305536
},
{
"epoch": 0.24,
"learning_rate": 0.00038594784353059177,
"loss": 2.6392,
"theoretical_loss": 3.7393296443267814,
"tokens_seen": 778371072
},
{
"epoch": 0.24,
"learning_rate": 0.00038593781344032095,
"loss": 2.8128,
"theoretical_loss": 3.7392982811313713,
"tokens_seen": 778436608
},
{
"epoch": 0.24,
"learning_rate": 0.00038592778335005013,
"loss": 2.9169,
"theoretical_loss": 3.739266921315547,
"tokens_seen": 778502144
},
{
"epoch": 0.24,
"learning_rate": 0.00038591775325977937,
"loss": 2.7289,
"theoretical_loss": 3.739235564878661,
"tokens_seen": 778567680
},
{
"epoch": 0.24,
"learning_rate": 0.0003859077231695085,
"loss": 2.6185,
"theoretical_loss": 3.7392042118200646,
"tokens_seen": 778633216
},
{
"epoch": 0.24,
"learning_rate": 0.00038589769307923773,
"loss": 2.8243,
"theoretical_loss": 3.739172862139109,
"tokens_seen": 778698752
},
{
"epoch": 0.24,
"learning_rate": 0.00038588766298896685,
"loss": 2.7481,
"theoretical_loss": 3.739141515835146,
"tokens_seen": 778764288
},
{
"epoch": 0.24,
"learning_rate": 0.0003858776328986961,
"loss": 2.8098,
"theoretical_loss": 3.7391101729075285,
"tokens_seen": 778829824
},
{
"epoch": 0.24,
"learning_rate": 0.0003858676028084253,
"loss": 2.677,
"theoretical_loss": 3.7390788333556086,
"tokens_seen": 778895360
},
{
"epoch": 0.24,
"learning_rate": 0.00038585757271815445,
"loss": 2.9516,
"theoretical_loss": 3.7390474971787375,
"tokens_seen": 778960896
},
{
"epoch": 0.24,
"learning_rate": 0.0003858475426278837,
"loss": 2.787,
"theoretical_loss": 3.7390161643762694,
"tokens_seen": 779026432
},
{
"epoch": 0.24,
"learning_rate": 0.00038583751253761287,
"loss": 2.8272,
"theoretical_loss": 3.7389848349475567,
"tokens_seen": 779091968
},
{
"epoch": 0.24,
"learning_rate": 0.00038582748244734205,
"loss": 2.7676,
"theoretical_loss": 3.738953508891952,
"tokens_seen": 779157504
},
{
"epoch": 0.24,
"learning_rate": 0.00038581745235707123,
"loss": 2.9695,
"theoretical_loss": 3.7389221862088085,
"tokens_seen": 779223040
},
{
"epoch": 0.24,
"learning_rate": 0.0003858074222668004,
"loss": 2.8802,
"theoretical_loss": 3.7388908668974796,
"tokens_seen": 779288576
},
{
"epoch": 0.24,
"learning_rate": 0.0003857973921765296,
"loss": 2.8759,
"theoretical_loss": 3.738859550957319,
"tokens_seen": 779354112
},
{
"epoch": 0.24,
"learning_rate": 0.00038578736208625883,
"loss": 2.8336,
"theoretical_loss": 3.73882823838768,
"tokens_seen": 779419648
},
{
"epoch": 0.24,
"learning_rate": 0.00038577733199598796,
"loss": 3.049,
"theoretical_loss": 3.7387969291879175,
"tokens_seen": 779485184
},
{
"epoch": 0.24,
"learning_rate": 0.0003857673019057172,
"loss": 3.0444,
"theoretical_loss": 3.7387656233573843,
"tokens_seen": 779550720
},
{
"epoch": 0.24,
"learning_rate": 0.0003857572718154463,
"loss": 2.8627,
"theoretical_loss": 3.7387343208954356,
"tokens_seen": 779616256
},
{
"epoch": 0.24,
"learning_rate": 0.00038574724172517555,
"loss": 2.9205,
"theoretical_loss": 3.7387030218014248,
"tokens_seen": 779681792
},
{
"epoch": 0.24,
"learning_rate": 0.00038573721163490473,
"loss": 2.9955,
"theoretical_loss": 3.738671726074707,
"tokens_seen": 779747328
},
{
"epoch": 0.24,
"learning_rate": 0.0003857271815446339,
"loss": 2.8837,
"theoretical_loss": 3.7386404337146373,
"tokens_seen": 779812864
},
{
"epoch": 0.24,
"objective/train/docs_used": 449318,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2265398502349854,
"objective/train/theoretical_loss": 3.73860914472057,
"objective/train/tokens_used": 800338400,
"theoretical_loss": 3.73860914472057,
"tokens_seen": 779878400
},
{
"epoch": 0.24,
"learning_rate": 0.0003857171514543631,
"loss": 3.16,
"theoretical_loss": 3.73860914472057,
"tokens_seen": 779878400
},
{
"epoch": 0.24,
"learning_rate": 0.00038570712136409233,
"loss": 2.9075,
"theoretical_loss": 3.738577859091861,
"tokens_seen": 779943936
},
{
"epoch": 0.24,
"learning_rate": 0.00038569709127382146,
"loss": 3.0539,
"theoretical_loss": 3.738546576827865,
"tokens_seen": 780009472
},
{
"epoch": 0.24,
"learning_rate": 0.0003856870611835507,
"loss": 2.9479,
"theoretical_loss": 3.738515297927938,
"tokens_seen": 780075008
},
{
"epoch": 0.24,
"learning_rate": 0.0003856770310932798,
"loss": 2.9774,
"theoretical_loss": 3.7384840223914346,
"tokens_seen": 780140544
},
{
"epoch": 0.24,
"learning_rate": 0.00038566700100300906,
"loss": 2.9774,
"theoretical_loss": 3.7384527502177116,
"tokens_seen": 780206080
},
{
"epoch": 0.24,
"learning_rate": 0.00038565697091273824,
"loss": 2.8533,
"theoretical_loss": 3.738421481406125,
"tokens_seen": 780271616
},
{
"epoch": 0.24,
"learning_rate": 0.0003856469408224674,
"loss": 2.8696,
"theoretical_loss": 3.7383902159560307,
"tokens_seen": 780337152
},
{
"epoch": 0.24,
"learning_rate": 0.0003856369107321966,
"loss": 2.8891,
"theoretical_loss": 3.7383589538667845,
"tokens_seen": 780402688
},
{
"epoch": 0.24,
"learning_rate": 0.0003856268806419258,
"loss": 2.9451,
"theoretical_loss": 3.7383276951377438,
"tokens_seen": 780468224
},
{
"epoch": 0.24,
"learning_rate": 0.00038561685055165496,
"loss": 2.924,
"theoretical_loss": 3.7382964397682654,
"tokens_seen": 780533760
},
{
"epoch": 0.24,
"learning_rate": 0.0003856068204613842,
"loss": 2.8117,
"theoretical_loss": 3.7382651877577056,
"tokens_seen": 780599296
},
{
"epoch": 0.24,
"learning_rate": 0.0003855967903711133,
"loss": 2.7474,
"theoretical_loss": 3.7382339391054216,
"tokens_seen": 780664832
},
{
"epoch": 0.24,
"learning_rate": 0.00038558676028084256,
"loss": 2.9307,
"theoretical_loss": 3.738202693810771,
"tokens_seen": 780730368
},
{
"epoch": 0.24,
"learning_rate": 0.0003855767301905717,
"loss": 2.6861,
"theoretical_loss": 3.738171451873111,
"tokens_seen": 780795904
},
{
"epoch": 0.24,
"learning_rate": 0.0003855667001003009,
"loss": 2.6089,
"theoretical_loss": 3.738140213291799,
"tokens_seen": 780861440
},
{
"epoch": 0.24,
"learning_rate": 0.0003855566700100301,
"loss": 3.1311,
"theoretical_loss": 3.7381089780661925,
"tokens_seen": 780926976
},
{
"epoch": 0.24,
"learning_rate": 0.0003855466399197593,
"loss": 2.9175,
"theoretical_loss": 3.7380777461956503,
"tokens_seen": 780992512
},
{
"epoch": 0.24,
"learning_rate": 0.00038553660982948846,
"loss": 2.8237,
"theoretical_loss": 3.73804651767953,
"tokens_seen": 781058048
},
{
"epoch": 0.24,
"learning_rate": 0.0003855265797392177,
"loss": 2.9295,
"theoretical_loss": 3.738015292517191,
"tokens_seen": 781123584
},
{
"epoch": 0.24,
"learning_rate": 0.0003855165496489468,
"loss": 2.5766,
"theoretical_loss": 3.7379840707079897,
"tokens_seen": 781189120
},
{
"epoch": 0.24,
"learning_rate": 0.00038550651955867606,
"loss": 2.9724,
"theoretical_loss": 3.7379528522512864,
"tokens_seen": 781254656
},
{
"epoch": 0.24,
"learning_rate": 0.0003854964894684052,
"loss": 2.9863,
"theoretical_loss": 3.7379216371464388,
"tokens_seen": 781320192
},
{
"epoch": 0.24,
"learning_rate": 0.0003854864593781344,
"loss": 2.8897,
"theoretical_loss": 3.7378904253928074,
"tokens_seen": 781385728
},
{
"epoch": 0.24,
"learning_rate": 0.0003854764292878636,
"loss": 2.8479,
"theoretical_loss": 3.73785921698975,
"tokens_seen": 781451264
},
{
"epoch": 0.24,
"objective/train/docs_used": 450726,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.263371467590332,
"objective/train/theoretical_loss": 3.7378280119366267,
"objective/train/tokens_used": 801976800,
"theoretical_loss": 3.7378280119366267,
"tokens_seen": 781516800
},
{
"epoch": 0.24,
"learning_rate": 0.0003854663991975928,
"loss": 2.8589,
"theoretical_loss": 3.7378280119366267,
"tokens_seen": 781516800
},
{
"epoch": 0.24,
"learning_rate": 0.00038545636910732197,
"loss": 2.9092,
"theoretical_loss": 3.7377968102327968,
"tokens_seen": 781582336
},
{
"epoch": 0.24,
"learning_rate": 0.00038544633901705115,
"loss": 2.8044,
"theoretical_loss": 3.7377656118776206,
"tokens_seen": 781647872
},
{
"epoch": 0.24,
"learning_rate": 0.00038543630892678033,
"loss": 2.5931,
"theoretical_loss": 3.737734416870457,
"tokens_seen": 781713408
},
{
"epoch": 0.24,
"learning_rate": 0.00038542627883650957,
"loss": 3.0313,
"theoretical_loss": 3.737703225210667,
"tokens_seen": 781778944
},
{
"epoch": 0.24,
"learning_rate": 0.0003854162487462387,
"loss": 2.959,
"theoretical_loss": 3.7376720368976093,
"tokens_seen": 781844480
},
{
"epoch": 0.24,
"learning_rate": 0.00038540621865596793,
"loss": 2.7604,
"theoretical_loss": 3.7376408519306463,
"tokens_seen": 781910016
},
{
"epoch": 0.24,
"learning_rate": 0.00038539618856569705,
"loss": 2.8251,
"theoretical_loss": 3.7376096703091375,
"tokens_seen": 781975552
},
{
"epoch": 0.24,
"learning_rate": 0.0003853861584754263,
"loss": 2.7295,
"theoretical_loss": 3.737578492032444,
"tokens_seen": 782041088
},
{
"epoch": 0.24,
"learning_rate": 0.00038537612838515547,
"loss": 2.8775,
"theoretical_loss": 3.7375473170999265,
"tokens_seen": 782106624
},
{
"epoch": 0.24,
"learning_rate": 0.00038536609829488465,
"loss": 2.8229,
"theoretical_loss": 3.7375161455109467,
"tokens_seen": 782172160
},
{
"epoch": 0.24,
"learning_rate": 0.00038535606820461383,
"loss": 2.8542,
"theoretical_loss": 3.737484977264865,
"tokens_seen": 782237696
},
{
"epoch": 0.24,
"learning_rate": 0.00038534603811434307,
"loss": 2.5699,
"theoretical_loss": 3.7374538123610432,
"tokens_seen": 782303232
},
{
"epoch": 0.24,
"learning_rate": 0.0003853360080240722,
"loss": 2.7131,
"theoretical_loss": 3.7374226507988437,
"tokens_seen": 782368768
},
{
"epoch": 0.24,
"learning_rate": 0.00038532597793380143,
"loss": 2.8221,
"theoretical_loss": 3.7373914925776273,
"tokens_seen": 782434304
},
{
"epoch": 0.24,
"learning_rate": 0.00038531594784353056,
"loss": 2.9776,
"theoretical_loss": 3.7373603376967566,
"tokens_seen": 782499840
},
{
"epoch": 0.24,
"learning_rate": 0.0003853059177532598,
"loss": 2.6617,
"theoretical_loss": 3.737329186155593,
"tokens_seen": 782565376
},
{
"epoch": 0.24,
"learning_rate": 0.000385295887662989,
"loss": 2.8275,
"theoretical_loss": 3.7372980379535004,
"tokens_seen": 782630912
},
{
"epoch": 0.24,
"learning_rate": 0.00038528585757271816,
"loss": 2.9258,
"theoretical_loss": 3.73726689308984,
"tokens_seen": 782696448
},
{
"epoch": 0.24,
"learning_rate": 0.00038527582748244734,
"loss": 2.5556,
"theoretical_loss": 3.7372357515639747,
"tokens_seen": 782761984
},
{
"epoch": 0.24,
"learning_rate": 0.0003852657973921765,
"loss": 2.8397,
"theoretical_loss": 3.737204613375268,
"tokens_seen": 782827520
},
{
"epoch": 0.24,
"learning_rate": 0.0003852557673019057,
"loss": 2.9835,
"theoretical_loss": 3.7371734785230823,
"tokens_seen": 782893056
},
{
"epoch": 0.24,
"learning_rate": 0.00038524573721163493,
"loss": 2.7787,
"theoretical_loss": 3.737142347006781,
"tokens_seen": 782958592
},
{
"epoch": 0.24,
"learning_rate": 0.00038523570712136406,
"loss": 2.9816,
"theoretical_loss": 3.737111218825728,
"tokens_seen": 783024128
},
{
"epoch": 0.24,
"learning_rate": 0.0003852256770310933,
"loss": 2.7271,
"theoretical_loss": 3.737080093979286,
"tokens_seen": 783089664
},
{
"epoch": 0.24,
"objective/train/docs_used": 451239,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1663591861724854,
"objective/train/theoretical_loss": 3.7370489724668197,
"objective/train/tokens_used": 803615200,
"theoretical_loss": 3.7370489724668197,
"tokens_seen": 783155200
},
{
"epoch": 0.24,
"learning_rate": 0.0003852156469408225,
"loss": 3.0609,
"theoretical_loss": 3.7370489724668197,
"tokens_seen": 783155200
},
{
"epoch": 0.24,
"learning_rate": 0.00038520561685055166,
"loss": 2.7544,
"theoretical_loss": 3.737017854287692,
"tokens_seen": 783220736
},
{
"epoch": 0.24,
"learning_rate": 0.00038519558676028084,
"loss": 2.8836,
"theoretical_loss": 3.7369867394412677,
"tokens_seen": 783286272
},
{
"epoch": 0.24,
"learning_rate": 0.00038518555667001,
"loss": 2.8657,
"theoretical_loss": 3.7369556279269105,
"tokens_seen": 783351808
},
{
"epoch": 0.24,
"learning_rate": 0.0003851755265797392,
"loss": 2.8652,
"theoretical_loss": 3.736924519743986,
"tokens_seen": 783417344
},
{
"epoch": 0.24,
"learning_rate": 0.00038516549648946844,
"loss": 2.9151,
"theoretical_loss": 3.736893414891858,
"tokens_seen": 783482880
},
{
"epoch": 0.24,
"learning_rate": 0.00038515546639919756,
"loss": 2.8787,
"theoretical_loss": 3.7368623133698913,
"tokens_seen": 783548416
},
{
"epoch": 0.24,
"learning_rate": 0.0003851454363089268,
"loss": 3.0007,
"theoretical_loss": 3.7368312151774505,
"tokens_seen": 783613952
},
{
"epoch": 0.24,
"learning_rate": 0.0003851354062186559,
"loss": 2.6451,
"theoretical_loss": 3.7368001203139016,
"tokens_seen": 783679488
},
{
"epoch": 0.24,
"learning_rate": 0.00038512537612838516,
"loss": 2.8102,
"theoretical_loss": 3.7367690287786095,
"tokens_seen": 783745024
},
{
"epoch": 0.24,
"learning_rate": 0.0003851153460381144,
"loss": 2.9624,
"theoretical_loss": 3.73673794057094,
"tokens_seen": 783810560
},
{
"epoch": 0.24,
"learning_rate": 0.0003851053159478435,
"loss": 2.8873,
"theoretical_loss": 3.736706855690258,
"tokens_seen": 783876096
},
{
"epoch": 0.24,
"learning_rate": 0.00038509528585757276,
"loss": 2.6384,
"theoretical_loss": 3.7366757741359304,
"tokens_seen": 783941632
},
{
"epoch": 0.24,
"learning_rate": 0.0003850852557673019,
"loss": 2.8962,
"theoretical_loss": 3.736644695907322,
"tokens_seen": 784007168
},
{
"epoch": 0.24,
"learning_rate": 0.0003850752256770311,
"loss": 2.7524,
"theoretical_loss": 3.7366136210038006,
"tokens_seen": 784072704
},
{
"epoch": 0.24,
"learning_rate": 0.0003850651955867603,
"loss": 2.8286,
"theoretical_loss": 3.736582549424731,
"tokens_seen": 784138240
},
{
"epoch": 0.24,
"learning_rate": 0.0003850551654964895,
"loss": 2.9882,
"theoretical_loss": 3.73655148116948,
"tokens_seen": 784203776
},
{
"epoch": 0.24,
"learning_rate": 0.00038504513540621866,
"loss": 2.6904,
"theoretical_loss": 3.7365204162374153,
"tokens_seen": 784269312
},
{
"epoch": 0.24,
"learning_rate": 0.0003850351053159479,
"loss": 2.7659,
"theoretical_loss": 3.736489354627903,
"tokens_seen": 784334848
},
{
"epoch": 0.24,
"learning_rate": 0.000385025075225677,
"loss": 2.9343,
"theoretical_loss": 3.7364582963403103,
"tokens_seen": 784400384
},
{
"epoch": 0.24,
"learning_rate": 0.00038501504513540626,
"loss": 2.8505,
"theoretical_loss": 3.7364272413740047,
"tokens_seen": 784465920
},
{
"epoch": 0.24,
"learning_rate": 0.0003850050150451354,
"loss": 2.6217,
"theoretical_loss": 3.736396189728353,
"tokens_seen": 784531456
},
{
"epoch": 0.24,
"learning_rate": 0.0003849949849548646,
"loss": 3.0504,
"theoretical_loss": 3.7363651414027235,
"tokens_seen": 784596992
},
{
"epoch": 0.24,
"learning_rate": 0.0003849849548645938,
"loss": 2.8179,
"theoretical_loss": 3.7363340963964835,
"tokens_seen": 784662528
},
{
"epoch": 0.24,
"learning_rate": 0.000384974924774323,
"loss": 2.8577,
"theoretical_loss": 3.7363030547090013,
"tokens_seen": 784728064
},
{
"epoch": 0.24,
"objective/train/docs_used": 452482,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.605654001235962,
"objective/train/theoretical_loss": 3.736272016339644,
"objective/train/tokens_used": 805253600,
"theoretical_loss": 3.736272016339644,
"tokens_seen": 784793600
},
{
"epoch": 0.24,
"learning_rate": 0.00038496489468405217,
"loss": 2.7124,
"theoretical_loss": 3.736272016339644,
"tokens_seen": 784793600
},
{
"epoch": 0.24,
"learning_rate": 0.00038495486459378135,
"loss": 2.8757,
"theoretical_loss": 3.7362409812877813,
"tokens_seen": 784859136
},
{
"epoch": 0.24,
"learning_rate": 0.00038494483450351053,
"loss": 2.9496,
"theoretical_loss": 3.7362099495527805,
"tokens_seen": 784924672
},
{
"epoch": 0.24,
"learning_rate": 0.00038493480441323977,
"loss": 2.6525,
"theoretical_loss": 3.736178921134011,
"tokens_seen": 784990208
},
{
"epoch": 0.24,
"learning_rate": 0.0003849247743229689,
"loss": 2.4628,
"theoretical_loss": 3.7361478960308414,
"tokens_seen": 785055744
},
{
"epoch": 0.24,
"learning_rate": 0.00038491474423269813,
"loss": 2.9341,
"theoretical_loss": 3.73611687424264,
"tokens_seen": 785121280
},
{
"epoch": 0.24,
"learning_rate": 0.00038490471414242725,
"loss": 2.6181,
"theoretical_loss": 3.736085855768777,
"tokens_seen": 785186816
},
{
"epoch": 0.24,
"learning_rate": 0.0003848946840521565,
"loss": 2.7802,
"theoretical_loss": 3.736054840608621,
"tokens_seen": 785252352
},
{
"epoch": 0.24,
"learning_rate": 0.00038488465396188567,
"loss": 2.8405,
"theoretical_loss": 3.7360238287615415,
"tokens_seen": 785317888
},
{
"epoch": 0.24,
"learning_rate": 0.00038487462387161485,
"loss": 2.9543,
"theoretical_loss": 3.7359928202269086,
"tokens_seen": 785383424
},
{
"epoch": 0.24,
"learning_rate": 0.00038486459378134403,
"loss": 2.8568,
"theoretical_loss": 3.735961815004092,
"tokens_seen": 785448960
},
{
"epoch": 0.24,
"learning_rate": 0.00038485456369107327,
"loss": 2.7492,
"theoretical_loss": 3.735930813092461,
"tokens_seen": 785514496
},
{
"epoch": 0.24,
"learning_rate": 0.0003848445336008024,
"loss": 3.0155,
"theoretical_loss": 3.7358998144913866,
"tokens_seen": 785580032
},
{
"epoch": 0.24,
"learning_rate": 0.00038483450351053163,
"loss": 2.7537,
"theoretical_loss": 3.735868819200239,
"tokens_seen": 785645568
},
{
"epoch": 0.24,
"learning_rate": 0.00038482447342026076,
"loss": 2.8791,
"theoretical_loss": 3.7358378272183885,
"tokens_seen": 785711104
},
{
"epoch": 0.24,
"learning_rate": 0.00038481444332999,
"loss": 2.66,
"theoretical_loss": 3.735806838545206,
"tokens_seen": 785776640
},
{
"epoch": 0.24,
"learning_rate": 0.0003848044132397192,
"loss": 2.9089,
"theoretical_loss": 3.735775853180062,
"tokens_seen": 785842176
},
{
"epoch": 0.24,
"learning_rate": 0.00038479438314944836,
"loss": 2.7553,
"theoretical_loss": 3.735744871122328,
"tokens_seen": 785907712
},
{
"epoch": 0.24,
"learning_rate": 0.00038478435305917754,
"loss": 2.6689,
"theoretical_loss": 3.7357138923713746,
"tokens_seen": 785973248
},
{
"epoch": 0.24,
"learning_rate": 0.0003847743229689067,
"loss": 2.7998,
"theoretical_loss": 3.7356829169265744,
"tokens_seen": 786038784
},
{
"epoch": 0.24,
"learning_rate": 0.0003847642928786359,
"loss": 2.7201,
"theoretical_loss": 3.735651944787297,
"tokens_seen": 786104320
},
{
"epoch": 0.24,
"learning_rate": 0.00038475426278836513,
"loss": 2.8254,
"theoretical_loss": 3.7356209759529158,
"tokens_seen": 786169856
},
{
"epoch": 0.24,
"learning_rate": 0.00038474423269809426,
"loss": 2.9047,
"theoretical_loss": 3.7355900104228024,
"tokens_seen": 786235392
},
{
"epoch": 0.24,
"learning_rate": 0.0003847342026078235,
"loss": 2.9713,
"theoretical_loss": 3.735559048196328,
"tokens_seen": 786300928
},
{
"epoch": 0.24,
"learning_rate": 0.0003847241725175527,
"loss": 2.8486,
"theoretical_loss": 3.735528089272866,
"tokens_seen": 786366464
},
{
"epoch": 0.24,
"objective/train/docs_used": 453141,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.833493232727051,
"objective/train/theoretical_loss": 3.735497133651788,
"objective/train/tokens_used": 806892000,
"theoretical_loss": 3.735497133651788,
"tokens_seen": 786432000
},
{
"epoch": 0.24,
"learning_rate": 0.00038471414242728186,
"loss": 2.8428,
"theoretical_loss": 3.735497133651788,
"tokens_seen": 786432000
},
{
"epoch": 0.24,
"learning_rate": 0.00038470411233701104,
"loss": 2.8402,
"theoretical_loss": 3.735466181332467,
"tokens_seen": 786497536
},
{
"epoch": 0.24,
"learning_rate": 0.0003846940822467402,
"loss": 2.8328,
"theoretical_loss": 3.735435232314275,
"tokens_seen": 786563072
},
{
"epoch": 0.24,
"learning_rate": 0.0003846840521564694,
"loss": 2.8892,
"theoretical_loss": 3.7354042865965864,
"tokens_seen": 786628608
},
{
"epoch": 0.24,
"learning_rate": 0.00038467402206619864,
"loss": 2.8053,
"theoretical_loss": 3.7353733441787726,
"tokens_seen": 786694144
},
{
"epoch": 0.24,
"learning_rate": 0.00038466399197592776,
"loss": 2.7826,
"theoretical_loss": 3.735342405060208,
"tokens_seen": 786759680
},
{
"epoch": 0.24,
"learning_rate": 0.000384653961885657,
"loss": 3.0341,
"theoretical_loss": 3.7353114692402656,
"tokens_seen": 786825216
},
{
"epoch": 0.24,
"learning_rate": 0.0003846439317953861,
"loss": 2.8594,
"theoretical_loss": 3.7352805367183195,
"tokens_seen": 786890752
},
{
"epoch": 0.24,
"learning_rate": 0.00038463390170511536,
"loss": 2.6673,
"theoretical_loss": 3.7352496074937425,
"tokens_seen": 786956288
},
{
"epoch": 0.24,
"learning_rate": 0.00038462387161484454,
"loss": 2.8609,
"theoretical_loss": 3.735218681565909,
"tokens_seen": 787021824
},
{
"epoch": 0.24,
"learning_rate": 0.0003846138415245737,
"loss": 2.6769,
"theoretical_loss": 3.7351877589341944,
"tokens_seen": 787087360
},
{
"epoch": 0.24,
"learning_rate": 0.0003846038114343029,
"loss": 2.8148,
"theoretical_loss": 3.7351568395979706,
"tokens_seen": 787152896
},
{
"epoch": 0.24,
"learning_rate": 0.0003845937813440321,
"loss": 2.8508,
"theoretical_loss": 3.7351259235566143,
"tokens_seen": 787218432
},
{
"epoch": 0.24,
"learning_rate": 0.00038458375125376127,
"loss": 2.8239,
"theoretical_loss": 3.7350950108094985,
"tokens_seen": 787283968
},
{
"epoch": 0.24,
"learning_rate": 0.0003845737211634905,
"loss": 2.4766,
"theoretical_loss": 3.7350641013559986,
"tokens_seen": 787349504
},
{
"epoch": 0.24,
"learning_rate": 0.00038456369107321963,
"loss": 2.852,
"theoretical_loss": 3.73503319519549,
"tokens_seen": 787415040
},
{
"epoch": 0.24,
"learning_rate": 0.00038455366098294886,
"loss": 2.5974,
"theoretical_loss": 3.7350022923273474,
"tokens_seen": 787480576
},
{
"epoch": 0.24,
"learning_rate": 0.00038454363089267805,
"loss": 2.6787,
"theoretical_loss": 3.7349713927509463,
"tokens_seen": 787546112
},
{
"epoch": 0.24,
"learning_rate": 0.00038453360080240723,
"loss": 2.726,
"theoretical_loss": 3.7349404964656614,
"tokens_seen": 787611648
},
{
"epoch": 0.24,
"learning_rate": 0.0003845235707121364,
"loss": 2.7888,
"theoretical_loss": 3.7349096034708698,
"tokens_seen": 787677184
},
{
"epoch": 0.24,
"learning_rate": 0.0003845135406218656,
"loss": 2.9275,
"theoretical_loss": 3.734878713765946,
"tokens_seen": 787742720
},
{
"epoch": 0.24,
"learning_rate": 0.00038450351053159477,
"loss": 2.6911,
"theoretical_loss": 3.734847827350267,
"tokens_seen": 787808256
},
{
"epoch": 0.24,
"learning_rate": 0.000384493480441324,
"loss": 3.1024,
"theoretical_loss": 3.7348169442232084,
"tokens_seen": 787873792
},
{
"epoch": 0.24,
"learning_rate": 0.00038448345035105313,
"loss": 2.8902,
"theoretical_loss": 3.7347860643841466,
"tokens_seen": 787939328
},
{
"epoch": 0.24,
"learning_rate": 0.00038447342026078237,
"loss": 2.4852,
"theoretical_loss": 3.7347551878324587,
"tokens_seen": 788004864
},
{
"epoch": 0.24,
"objective/train/docs_used": 453953,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.858163833618164,
"objective/train/theoretical_loss": 3.73472431456752,
"objective/train/tokens_used": 808530400,
"theoretical_loss": 3.73472431456752,
"tokens_seen": 788070400
},
{
"epoch": 0.24,
"learning_rate": 0.0003844633901705115,
"loss": 2.7977,
"theoretical_loss": 3.73472431456752,
"tokens_seen": 788070400
},
{
"epoch": 0.24,
"learning_rate": 0.00038445336008024073,
"loss": 2.6308,
"theoretical_loss": 3.7346934445887086,
"tokens_seen": 788135936
},
{
"epoch": 0.24,
"learning_rate": 0.0003844433299899699,
"loss": 2.8514,
"theoretical_loss": 3.734662577895401,
"tokens_seen": 788201472
},
{
"epoch": 0.24,
"learning_rate": 0.0003844332998996991,
"loss": 3.0053,
"theoretical_loss": 3.7346317144869747,
"tokens_seen": 788267008
},
{
"epoch": 0.24,
"learning_rate": 0.0003844232698094283,
"loss": 2.6488,
"theoretical_loss": 3.7346008543628066,
"tokens_seen": 788332544
},
{
"epoch": 0.24,
"learning_rate": 0.00038441323971915745,
"loss": 2.6365,
"theoretical_loss": 3.734569997522275,
"tokens_seen": 788398080
},
{
"epoch": 0.24,
"learning_rate": 0.00038440320962888664,
"loss": 2.7907,
"theoretical_loss": 3.734539143964757,
"tokens_seen": 788463616
},
{
"epoch": 0.24,
"learning_rate": 0.00038439317953861587,
"loss": 2.9101,
"theoretical_loss": 3.73450829368963,
"tokens_seen": 788529152
},
{
"epoch": 0.24,
"learning_rate": 0.00038438314944834505,
"loss": 2.7747,
"theoretical_loss": 3.734477446696273,
"tokens_seen": 788594688
},
{
"epoch": 0.24,
"learning_rate": 0.00038437311935807423,
"loss": 2.7988,
"theoretical_loss": 3.7344466029840637,
"tokens_seen": 788660224
},
{
"epoch": 0.24,
"learning_rate": 0.00038436308926780347,
"loss": 2.6645,
"theoretical_loss": 3.7344157625523806,
"tokens_seen": 788725760
},
{
"epoch": 0.24,
"learning_rate": 0.0003843530591775326,
"loss": 2.7731,
"theoretical_loss": 3.7343849254006027,
"tokens_seen": 788791296
},
{
"epoch": 0.24,
"learning_rate": 0.00038434302908726183,
"loss": 2.8733,
"theoretical_loss": 3.7343540915281075,
"tokens_seen": 788856832
},
{
"epoch": 0.24,
"learning_rate": 0.00038433299899699096,
"loss": 2.8864,
"theoretical_loss": 3.734323260934275,
"tokens_seen": 788922368
},
{
"epoch": 0.24,
"learning_rate": 0.0003843229689067202,
"loss": 2.7456,
"theoretical_loss": 3.734292433618484,
"tokens_seen": 788987904
},
{
"epoch": 0.24,
"learning_rate": 0.0003843129388164494,
"loss": 2.9933,
"theoretical_loss": 3.7342616095801136,
"tokens_seen": 789053440
},
{
"epoch": 0.24,
"learning_rate": 0.00038430290872617856,
"loss": 2.8561,
"theoretical_loss": 3.734230788818543,
"tokens_seen": 789118976
},
{
"epoch": 0.24,
"learning_rate": 0.00038429287863590774,
"loss": 2.743,
"theoretical_loss": 3.734199971333152,
"tokens_seen": 789184512
},
{
"epoch": 0.24,
"learning_rate": 0.0003842828485456369,
"loss": 2.8065,
"theoretical_loss": 3.73416915712332,
"tokens_seen": 789250048
},
{
"epoch": 0.24,
"learning_rate": 0.0003842728184553661,
"loss": 2.8362,
"theoretical_loss": 3.734138346188428,
"tokens_seen": 789315584
},
{
"epoch": 0.24,
"learning_rate": 0.00038426278836509533,
"loss": 2.7156,
"theoretical_loss": 3.734107538527854,
"tokens_seen": 789381120
},
{
"epoch": 0.24,
"learning_rate": 0.00038425275827482446,
"loss": 2.8287,
"theoretical_loss": 3.734076734140981,
"tokens_seen": 789446656
},
{
"epoch": 0.24,
"learning_rate": 0.0003842427281845537,
"loss": 2.8407,
"theoretical_loss": 3.7340459330271867,
"tokens_seen": 789512192
},
{
"epoch": 0.24,
"learning_rate": 0.0003842326980942829,
"loss": 2.6617,
"theoretical_loss": 3.7340151351858535,
"tokens_seen": 789577728
},
{
"epoch": 0.24,
"learning_rate": 0.00038422266800401206,
"loss": 2.7942,
"theoretical_loss": 3.7339843406163613,
"tokens_seen": 789643264
},
{
"epoch": 0.24,
"objective/train/docs_used": 455438,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6975836753845215,
"objective/train/theoretical_loss": 3.733953549318091,
"objective/train/tokens_used": 810168800,
"theoretical_loss": 3.733953549318091,
"tokens_seen": 789708800
},
{
"epoch": 0.24,
"learning_rate": 0.00038421263791374124,
"loss": 2.8087,
"theoretical_loss": 3.733953549318091,
"tokens_seen": 789708800
},
{
"epoch": 0.24,
"learning_rate": 0.0003842026078234704,
"loss": 3.0872,
"theoretical_loss": 3.733922761290424,
"tokens_seen": 789774336
},
{
"epoch": 0.24,
"learning_rate": 0.0003841925777331996,
"loss": 2.8199,
"theoretical_loss": 3.733891976532742,
"tokens_seen": 789839872
},
{
"epoch": 0.24,
"learning_rate": 0.00038418254764292884,
"loss": 2.7833,
"theoretical_loss": 3.733861195044425,
"tokens_seen": 789905408
},
{
"epoch": 0.24,
"learning_rate": 0.00038417251755265796,
"loss": 2.5702,
"theoretical_loss": 3.7338304168248566,
"tokens_seen": 789970944
},
{
"epoch": 0.24,
"learning_rate": 0.0003841624874623872,
"loss": 2.6715,
"theoretical_loss": 3.733799641873417,
"tokens_seen": 790036480
},
{
"epoch": 0.24,
"learning_rate": 0.0003841524573721163,
"loss": 2.8999,
"theoretical_loss": 3.733768870189488,
"tokens_seen": 790102016
},
{
"epoch": 0.24,
"learning_rate": 0.00038414242728184556,
"loss": 2.5953,
"theoretical_loss": 3.7337381017724525,
"tokens_seen": 790167552
},
{
"epoch": 0.24,
"learning_rate": 0.00038413239719157474,
"loss": 2.6979,
"theoretical_loss": 3.733707336621692,
"tokens_seen": 790233088
},
{
"epoch": 0.24,
"learning_rate": 0.0003841223671013039,
"loss": 2.9423,
"theoretical_loss": 3.7336765747365903,
"tokens_seen": 790298624
},
{
"epoch": 0.24,
"learning_rate": 0.0003841123370110331,
"loss": 2.9057,
"theoretical_loss": 3.7336458161165287,
"tokens_seen": 790364160
},
{
"epoch": 0.24,
"learning_rate": 0.0003841023069207623,
"loss": 2.7296,
"theoretical_loss": 3.73361506076089,
"tokens_seen": 790429696
},
{
"epoch": 0.24,
"learning_rate": 0.00038409227683049147,
"loss": 2.8154,
"theoretical_loss": 3.7335843086690583,
"tokens_seen": 790495232
},
{
"epoch": 0.24,
"learning_rate": 0.0003840822467402207,
"loss": 2.8818,
"theoretical_loss": 3.7335535598404155,
"tokens_seen": 790560768
},
{
"epoch": 0.24,
"learning_rate": 0.00038407221664994983,
"loss": 2.5063,
"theoretical_loss": 3.733522814274345,
"tokens_seen": 790626304
},
{
"epoch": 0.24,
"learning_rate": 0.00038406218655967907,
"loss": 2.9018,
"theoretical_loss": 3.7334920719702307,
"tokens_seen": 790691840
},
{
"epoch": 0.24,
"learning_rate": 0.00038405215646940825,
"loss": 2.7301,
"theoretical_loss": 3.733461332927456,
"tokens_seen": 790757376
},
{
"epoch": 0.24,
"learning_rate": 0.00038404212637913743,
"loss": 2.6452,
"theoretical_loss": 3.7334305971454045,
"tokens_seen": 790822912
},
{
"epoch": 0.24,
"learning_rate": 0.0003840320962888666,
"loss": 2.7411,
"theoretical_loss": 3.7333998646234603,
"tokens_seen": 790888448
},
{
"epoch": 0.24,
"learning_rate": 0.0003840220661985958,
"loss": 2.8237,
"theoretical_loss": 3.7333691353610075,
"tokens_seen": 790953984
},
{
"epoch": 0.24,
"learning_rate": 0.00038401203610832497,
"loss": 2.9892,
"theoretical_loss": 3.7333384093574304,
"tokens_seen": 791019520
},
{
"epoch": 0.24,
"learning_rate": 0.0003840020060180542,
"loss": 2.9209,
"theoretical_loss": 3.7333076866121138,
"tokens_seen": 791085056
},
{
"epoch": 0.24,
"learning_rate": 0.00038399197592778333,
"loss": 2.6288,
"theoretical_loss": 3.733276967124441,
"tokens_seen": 791150592
},
{
"epoch": 0.24,
"learning_rate": 0.00038398194583751257,
"loss": 2.882,
"theoretical_loss": 3.7332462508937985,
"tokens_seen": 791216128
},
{
"epoch": 0.24,
"learning_rate": 0.0003839719157472417,
"loss": 2.6578,
"theoretical_loss": 3.7332155379195697,
"tokens_seen": 791281664
},
{
"epoch": 0.24,
"objective/train/docs_used": 456184,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.73787260055542,
"objective/train/theoretical_loss": 3.733184828201141,
"objective/train/tokens_used": 811807200,
"theoretical_loss": 3.733184828201141,
"tokens_seen": 791347200
},
{
"epoch": 0.24,
"learning_rate": 0.00038396188565697093,
"loss": 2.8026,
"theoretical_loss": 3.733184828201141,
"tokens_seen": 791347200
},
{
"epoch": 0.24,
"learning_rate": 0.0003839518555667001,
"loss": 2.8852,
"theoretical_loss": 3.733154121737897,
"tokens_seen": 791412736
},
{
"epoch": 0.24,
"learning_rate": 0.0003839418254764293,
"loss": 2.8437,
"theoretical_loss": 3.733123418529223,
"tokens_seen": 791478272
},
{
"epoch": 0.24,
"learning_rate": 0.0003839317953861585,
"loss": 2.7537,
"theoretical_loss": 3.733092718574505,
"tokens_seen": 791543808
},
{
"epoch": 0.24,
"learning_rate": 0.00038392176529588765,
"loss": 2.7147,
"theoretical_loss": 3.7330620218731285,
"tokens_seen": 791609344
},
{
"epoch": 0.24,
"learning_rate": 0.00038391173520561684,
"loss": 2.7914,
"theoretical_loss": 3.7330313284244796,
"tokens_seen": 791674880
},
{
"epoch": 0.24,
"learning_rate": 0.00038390170511534607,
"loss": 2.7979,
"theoretical_loss": 3.7330006382279444,
"tokens_seen": 791740416
},
{
"epoch": 0.24,
"learning_rate": 0.0003838916750250752,
"loss": 2.7431,
"theoretical_loss": 3.732969951282909,
"tokens_seen": 791805952
},
{
"epoch": 0.24,
"learning_rate": 0.00038388164493480443,
"loss": 2.6554,
"theoretical_loss": 3.73293926758876,
"tokens_seen": 791871488
},
{
"epoch": 0.24,
"learning_rate": 0.0003838716148445336,
"loss": 2.852,
"theoretical_loss": 3.732908587144884,
"tokens_seen": 791937024
},
{
"epoch": 0.24,
"learning_rate": 0.0003838615847542628,
"loss": 2.9275,
"theoretical_loss": 3.732877909950668,
"tokens_seen": 792002560
},
{
"epoch": 0.24,
"learning_rate": 0.000383851554663992,
"loss": 2.9101,
"theoretical_loss": 3.7328472360054983,
"tokens_seen": 792068096
},
{
"epoch": 0.24,
"learning_rate": 0.00038384152457372116,
"loss": 2.7506,
"theoretical_loss": 3.7328165653087626,
"tokens_seen": 792133632
},
{
"epoch": 0.24,
"learning_rate": 0.00038383149448345034,
"loss": 2.7916,
"theoretical_loss": 3.7327858978598476,
"tokens_seen": 792199168
},
{
"epoch": 0.24,
"learning_rate": 0.0003838214643931796,
"loss": 2.7201,
"theoretical_loss": 3.7327552336581418,
"tokens_seen": 792264704
},
{
"epoch": 0.24,
"learning_rate": 0.0003838114343029087,
"loss": 2.8005,
"theoretical_loss": 3.732724572703032,
"tokens_seen": 792330240
},
{
"epoch": 0.24,
"learning_rate": 0.00038380140421263794,
"loss": 3.0203,
"theoretical_loss": 3.732693914993906,
"tokens_seen": 792395776
},
{
"epoch": 0.24,
"learning_rate": 0.00038379137412236706,
"loss": 2.7241,
"theoretical_loss": 3.732663260530151,
"tokens_seen": 792461312
},
{
"epoch": 0.24,
"learning_rate": 0.0003837813440320963,
"loss": 2.6356,
"theoretical_loss": 3.732632609311157,
"tokens_seen": 792526848
},
{
"epoch": 0.24,
"learning_rate": 0.0003837713139418255,
"loss": 2.7378,
"theoretical_loss": 3.7326019613363104,
"tokens_seen": 792592384
},
{
"epoch": 0.24,
"learning_rate": 0.00038376128385155466,
"loss": 3.0324,
"theoretical_loss": 3.732571316605001,
"tokens_seen": 792657920
},
{
"epoch": 0.24,
"learning_rate": 0.00038375125376128384,
"loss": 2.7243,
"theoretical_loss": 3.7325406751166166,
"tokens_seen": 792723456
},
{
"epoch": 0.24,
"learning_rate": 0.0003837412236710131,
"loss": 2.5504,
"theoretical_loss": 3.732510036870546,
"tokens_seen": 792788992
},
{
"epoch": 0.24,
"learning_rate": 0.0003837311935807422,
"loss": 2.8917,
"theoretical_loss": 3.7324794018661787,
"tokens_seen": 792854528
},
{
"epoch": 0.24,
"learning_rate": 0.00038372116349047144,
"loss": 2.8771,
"theoretical_loss": 3.732448770102903,
"tokens_seen": 792920064
},
{
"epoch": 0.24,
"objective/train/docs_used": 457386,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9626502990722656,
"objective/train/theoretical_loss": 3.7324181415801094,
"objective/train/tokens_used": 813445600,
"theoretical_loss": 3.7324181415801094,
"tokens_seen": 792985600
},
{
"epoch": 0.24,
"learning_rate": 0.00038371113340020057,
"loss": 2.8763,
"theoretical_loss": 3.7324181415801094,
"tokens_seen": 792985600
},
{
"epoch": 0.24,
"learning_rate": 0.0003837011033099298,
"loss": 2.8423,
"theoretical_loss": 3.7323875162971856,
"tokens_seen": 793051136
},
{
"epoch": 0.24,
"learning_rate": 0.000383691073219659,
"loss": 2.9869,
"theoretical_loss": 3.7323568942535226,
"tokens_seen": 793116672
},
{
"epoch": 0.24,
"learning_rate": 0.00038368104312938816,
"loss": 2.7476,
"theoretical_loss": 3.73232627544851,
"tokens_seen": 793182208
},
{
"epoch": 0.24,
"learning_rate": 0.00038367101303911735,
"loss": 3.1257,
"theoretical_loss": 3.7322956598815367,
"tokens_seen": 793247744
},
{
"epoch": 0.24,
"learning_rate": 0.0003836609829488465,
"loss": 2.7995,
"theoretical_loss": 3.7322650475519934,
"tokens_seen": 793313280
},
{
"epoch": 0.24,
"learning_rate": 0.0003836509528585757,
"loss": 2.6196,
"theoretical_loss": 3.7322344384592707,
"tokens_seen": 793378816
},
{
"epoch": 0.24,
"learning_rate": 0.00038364092276830494,
"loss": 2.7951,
"theoretical_loss": 3.7322038326027593,
"tokens_seen": 793444352
},
{
"epoch": 0.24,
"learning_rate": 0.0003836308926780341,
"loss": 2.585,
"theoretical_loss": 3.7321732299818486,
"tokens_seen": 793509888
},
{
"epoch": 0.24,
"learning_rate": 0.0003836208625877633,
"loss": 2.9008,
"theoretical_loss": 3.73214263059593,
"tokens_seen": 793575424
},
{
"epoch": 0.24,
"learning_rate": 0.0003836108324974925,
"loss": 2.6402,
"theoretical_loss": 3.7321120344443948,
"tokens_seen": 793640960
},
{
"epoch": 0.24,
"learning_rate": 0.00038360080240722167,
"loss": 2.7686,
"theoretical_loss": 3.732081441526634,
"tokens_seen": 793706496
},
{
"epoch": 0.24,
"learning_rate": 0.0003835907723169509,
"loss": 2.7614,
"theoretical_loss": 3.732050851842038,
"tokens_seen": 793772032
},
{
"epoch": 0.24,
"learning_rate": 0.00038358074222668003,
"loss": 2.631,
"theoretical_loss": 3.7320202653899983,
"tokens_seen": 793837568
},
{
"epoch": 0.24,
"learning_rate": 0.00038357071213640927,
"loss": 2.5471,
"theoretical_loss": 3.7319896821699077,
"tokens_seen": 793903104
},
{
"epoch": 0.24,
"learning_rate": 0.00038356068204613845,
"loss": 3.0798,
"theoretical_loss": 3.731959102181157,
"tokens_seen": 793968640
},
{
"epoch": 0.24,
"learning_rate": 0.00038355065195586763,
"loss": 2.726,
"theoretical_loss": 3.7319285254231382,
"tokens_seen": 794034176
},
{
"epoch": 0.24,
"learning_rate": 0.0003835406218655968,
"loss": 2.6926,
"theoretical_loss": 3.7318979518952435,
"tokens_seen": 794099712
},
{
"epoch": 0.24,
"learning_rate": 0.000383530591775326,
"loss": 2.6099,
"theoretical_loss": 3.7318673815968655,
"tokens_seen": 794165248
},
{
"epoch": 0.24,
"learning_rate": 0.00038352056168505517,
"loss": 2.9393,
"theoretical_loss": 3.7318368145273952,
"tokens_seen": 794230784
},
{
"epoch": 0.24,
"learning_rate": 0.0003835105315947844,
"loss": 2.9172,
"theoretical_loss": 3.731806250686227,
"tokens_seen": 794296320
},
{
"epoch": 0.24,
"learning_rate": 0.00038350050150451353,
"loss": 2.7474,
"theoretical_loss": 3.7317756900727526,
"tokens_seen": 794361856
},
{
"epoch": 0.24,
"learning_rate": 0.00038349047141424277,
"loss": 2.6424,
"theoretical_loss": 3.731745132686365,
"tokens_seen": 794427392
},
{
"epoch": 0.24,
"learning_rate": 0.0003834804413239719,
"loss": 2.8595,
"theoretical_loss": 3.731714578526457,
"tokens_seen": 794492928
},
{
"epoch": 0.24,
"learning_rate": 0.00038347041123370113,
"loss": 2.6604,
"theoretical_loss": 3.7316840275924226,
"tokens_seen": 794558464
},
{
"epoch": 0.24,
"objective/train/docs_used": 458050,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9160664081573486,
"objective/train/theoretical_loss": 3.7316534798836543,
"objective/train/tokens_used": 815084000,
"theoretical_loss": 3.7316534798836543,
"tokens_seen": 794624000
},
{
"epoch": 0.24,
"learning_rate": 0.0003834603811434303,
"loss": 2.8156,
"theoretical_loss": 3.7316534798836543,
"tokens_seen": 794624000
},
{
"epoch": 0.24,
"learning_rate": 0.0003834503510531595,
"loss": 2.8904,
"theoretical_loss": 3.731622935399547,
"tokens_seen": 794689536
},
{
"epoch": 0.24,
"learning_rate": 0.0003834403209628887,
"loss": 2.6288,
"theoretical_loss": 3.7315923941394926,
"tokens_seen": 794755072
},
{
"epoch": 0.24,
"learning_rate": 0.00038343029087261786,
"loss": 2.6766,
"theoretical_loss": 3.7315618561028865,
"tokens_seen": 794820608
},
{
"epoch": 0.24,
"learning_rate": 0.00038342026078234704,
"loss": 2.9291,
"theoretical_loss": 3.731531321289121,
"tokens_seen": 794886144
},
{
"epoch": 0.24,
"learning_rate": 0.00038341023069207627,
"loss": 2.7842,
"theoretical_loss": 3.7315007896975927,
"tokens_seen": 794951680
},
{
"epoch": 0.24,
"learning_rate": 0.0003834002006018054,
"loss": 2.4602,
"theoretical_loss": 3.731470261327694,
"tokens_seen": 795017216
},
{
"epoch": 0.24,
"learning_rate": 0.00038339017051153463,
"loss": 2.854,
"theoretical_loss": 3.73143973617882,
"tokens_seen": 795082752
},
{
"epoch": 0.24,
"learning_rate": 0.0003833801404212638,
"loss": 2.8226,
"theoretical_loss": 3.731409214250366,
"tokens_seen": 795148288
},
{
"epoch": 0.24,
"learning_rate": 0.000383370110330993,
"loss": 2.7945,
"theoretical_loss": 3.731378695541726,
"tokens_seen": 795213824
},
{
"epoch": 0.24,
"learning_rate": 0.0003833600802407222,
"loss": 2.6623,
"theoretical_loss": 3.7313481800522954,
"tokens_seen": 795279360
},
{
"epoch": 0.24,
"learning_rate": 0.00038335005015045136,
"loss": 2.6378,
"theoretical_loss": 3.73131766778147,
"tokens_seen": 795344896
},
{
"epoch": 0.24,
"learning_rate": 0.00038334002006018054,
"loss": 2.7031,
"theoretical_loss": 3.7312871587286436,
"tokens_seen": 795410432
},
{
"epoch": 0.24,
"learning_rate": 0.0003833299899699098,
"loss": 2.9365,
"theoretical_loss": 3.7312566528932134,
"tokens_seen": 795475968
},
{
"epoch": 0.24,
"learning_rate": 0.0003833199598796389,
"loss": 2.5402,
"theoretical_loss": 3.7312261502745736,
"tokens_seen": 795541504
},
{
"epoch": 0.24,
"learning_rate": 0.00038330992978936814,
"loss": 2.6777,
"theoretical_loss": 3.731195650872121,
"tokens_seen": 795607040
},
{
"epoch": 0.24,
"learning_rate": 0.00038329989969909726,
"loss": 2.7142,
"theoretical_loss": 3.731165154685252,
"tokens_seen": 795672576
},
{
"epoch": 0.24,
"learning_rate": 0.0003832898696088265,
"loss": 2.5974,
"theoretical_loss": 3.7311346617133614,
"tokens_seen": 795738112
},
{
"epoch": 0.24,
"learning_rate": 0.0003832798395185557,
"loss": 2.5814,
"theoretical_loss": 3.731104171955847,
"tokens_seen": 795803648
},
{
"epoch": 0.24,
"learning_rate": 0.00038326980942828486,
"loss": 3.0703,
"theoretical_loss": 3.731073685412104,
"tokens_seen": 795869184
},
{
"epoch": 0.24,
"learning_rate": 0.00038325977933801404,
"loss": 2.985,
"theoretical_loss": 3.73104320208153,
"tokens_seen": 795934720
},
{
"epoch": 0.24,
"learning_rate": 0.0003832497492477433,
"loss": 2.7374,
"theoretical_loss": 3.7310127219635207,
"tokens_seen": 796000256
},
{
"epoch": 0.24,
"learning_rate": 0.0003832397191574724,
"loss": 2.7459,
"theoretical_loss": 3.730982245057474,
"tokens_seen": 796065792
},
{
"epoch": 0.24,
"learning_rate": 0.00038322968906720164,
"loss": 2.9528,
"theoretical_loss": 3.730951771362787,
"tokens_seen": 796131328
},
{
"epoch": 0.24,
"learning_rate": 0.00038321965897693077,
"loss": 2.7962,
"theoretical_loss": 3.7309213008788573,
"tokens_seen": 796196864
},
{
"epoch": 0.24,
"objective/train/docs_used": 459137,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.604027271270752,
"objective/train/theoretical_loss": 3.7308908336050814,
"objective/train/tokens_used": 816722400,
"theoretical_loss": 3.7308908336050814,
"tokens_seen": 796262400
},
{
"epoch": 0.24,
"learning_rate": 0.00038320962888666,
"loss": 2.7054,
"theoretical_loss": 3.7308908336050814,
"tokens_seen": 796262400
},
{
"epoch": 0.24,
"learning_rate": 0.0003831995987963892,
"loss": 2.6706,
"theoretical_loss": 3.730860369540858,
"tokens_seen": 796327936
},
{
"epoch": 0.24,
"learning_rate": 0.00038318956870611836,
"loss": 2.7243,
"theoretical_loss": 3.7308299086855836,
"tokens_seen": 796393472
},
{
"epoch": 0.24,
"learning_rate": 0.00038317953861584755,
"loss": 2.7006,
"theoretical_loss": 3.7307994510386577,
"tokens_seen": 796459008
},
{
"epoch": 0.24,
"learning_rate": 0.0003831695085255767,
"loss": 2.8493,
"theoretical_loss": 3.730768996599477,
"tokens_seen": 796524544
},
{
"epoch": 0.24,
"learning_rate": 0.0003831594784353059,
"loss": 2.5476,
"theoretical_loss": 3.7307385453674407,
"tokens_seen": 796590080
},
{
"epoch": 0.24,
"learning_rate": 0.00038314944834503514,
"loss": 2.8749,
"theoretical_loss": 3.730708097341947,
"tokens_seen": 796655616
},
{
"epoch": 0.24,
"learning_rate": 0.00038313941825476427,
"loss": 2.7664,
"theoretical_loss": 3.7306776525223944,
"tokens_seen": 796721152
},
{
"epoch": 0.24,
"learning_rate": 0.0003831293881644935,
"loss": 2.6439,
"theoretical_loss": 3.730647210908182,
"tokens_seen": 796786688
},
{
"epoch": 0.24,
"learning_rate": 0.00038311935807422263,
"loss": 2.7646,
"theoretical_loss": 3.730616772498708,
"tokens_seen": 796852224
},
{
"epoch": 0.24,
"learning_rate": 0.00038310932798395187,
"loss": 2.9244,
"theoretical_loss": 3.7305863372933725,
"tokens_seen": 796917760
},
{
"epoch": 0.24,
"learning_rate": 0.00038309929789368105,
"loss": 2.9017,
"theoretical_loss": 3.730555905291574,
"tokens_seen": 796983296
},
{
"epoch": 0.24,
"learning_rate": 0.00038308926780341023,
"loss": 2.8472,
"theoretical_loss": 3.730525476492712,
"tokens_seen": 797048832
},
{
"epoch": 0.24,
"learning_rate": 0.0003830792377131394,
"loss": 2.523,
"theoretical_loss": 3.7304950508961863,
"tokens_seen": 797114368
},
{
"epoch": 0.24,
"learning_rate": 0.00038306920762286865,
"loss": 2.7902,
"theoretical_loss": 3.7304646285013963,
"tokens_seen": 797179904
},
{
"epoch": 0.24,
"learning_rate": 0.0003830591775325978,
"loss": 2.7738,
"theoretical_loss": 3.7304342093077425,
"tokens_seen": 797245440
},
{
"epoch": 0.24,
"learning_rate": 0.000383049147442327,
"loss": 2.6629,
"theoretical_loss": 3.7304037933146246,
"tokens_seen": 797310976
},
{
"epoch": 0.24,
"learning_rate": 0.00038303911735205614,
"loss": 2.9269,
"theoretical_loss": 3.7303733805214425,
"tokens_seen": 797376512
},
{
"epoch": 0.24,
"learning_rate": 0.00038302908726178537,
"loss": 2.8032,
"theoretical_loss": 3.7303429709275973,
"tokens_seen": 797442048
},
{
"epoch": 0.24,
"learning_rate": 0.00038301905717151455,
"loss": 2.8251,
"theoretical_loss": 3.730312564532489,
"tokens_seen": 797507584
},
{
"epoch": 0.24,
"learning_rate": 0.00038300902708124373,
"loss": 2.8853,
"theoretical_loss": 3.7302821613355186,
"tokens_seen": 797573120
},
{
"epoch": 0.24,
"learning_rate": 0.0003829989969909729,
"loss": 2.8428,
"theoretical_loss": 3.7302517613360866,
"tokens_seen": 797638656
},
{
"epoch": 0.24,
"learning_rate": 0.0003829889669007021,
"loss": 2.6407,
"theoretical_loss": 3.7302213645335947,
"tokens_seen": 797704192
},
{
"epoch": 0.24,
"learning_rate": 0.0003829789368104313,
"loss": 2.6374,
"theoretical_loss": 3.7301909709274437,
"tokens_seen": 797769728
},
{
"epoch": 0.24,
"learning_rate": 0.0003829689067201605,
"loss": 2.6257,
"theoretical_loss": 3.730160580517035,
"tokens_seen": 797835264
},
{
"epoch": 0.24,
"objective/train/docs_used": 459607,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.257622003555298,
"objective/train/theoretical_loss": 3.73013019330177,
"objective/train/tokens_used": 818360800,
"theoretical_loss": 3.73013019330177,
"tokens_seen": 797900800
},
{
"epoch": 0.24,
"learning_rate": 0.00038295887662988964,
"loss": 2.8393,
"theoretical_loss": 3.73013019330177,
"tokens_seen": 797900800
},
{
"epoch": 0.24,
"learning_rate": 0.0003829488465396189,
"loss": 2.6401,
"theoretical_loss": 3.7300998092810502,
"tokens_seen": 797966336
},
{
"epoch": 0.24,
"learning_rate": 0.000382938816449348,
"loss": 2.5864,
"theoretical_loss": 3.7300694284542786,
"tokens_seen": 798031872
},
{
"epoch": 0.24,
"learning_rate": 0.00038292878635907724,
"loss": 2.7187,
"theoretical_loss": 3.730039050820855,
"tokens_seen": 798097408
},
{
"epoch": 0.24,
"learning_rate": 0.0003829187562688064,
"loss": 2.6976,
"theoretical_loss": 3.7300086763801836,
"tokens_seen": 798162944
},
{
"epoch": 0.24,
"learning_rate": 0.0003829087261785356,
"loss": 2.8406,
"theoretical_loss": 3.729978305131666,
"tokens_seen": 798228480
},
{
"epoch": 0.24,
"learning_rate": 0.0003828986960882648,
"loss": 2.8614,
"theoretical_loss": 3.729947937074705,
"tokens_seen": 798294016
},
{
"epoch": 0.24,
"learning_rate": 0.000382888665997994,
"loss": 2.6668,
"theoretical_loss": 3.7299175722087026,
"tokens_seen": 798359552
},
{
"epoch": 0.24,
"learning_rate": 0.0003828786359077232,
"loss": 3.0132,
"theoretical_loss": 3.729887210533062,
"tokens_seen": 798425088
},
{
"epoch": 0.24,
"learning_rate": 0.0003828686058174524,
"loss": 2.7879,
"theoretical_loss": 3.7298568520471855,
"tokens_seen": 798490624
},
{
"epoch": 0.24,
"learning_rate": 0.00038285857572718156,
"loss": 2.8579,
"theoretical_loss": 3.7298264967504773,
"tokens_seen": 798556160
},
{
"epoch": 0.24,
"learning_rate": 0.00038284854563691074,
"loss": 2.7117,
"theoretical_loss": 3.72979614464234,
"tokens_seen": 798621696
},
{
"epoch": 0.24,
"learning_rate": 0.00038283851554664,
"loss": 2.8432,
"theoretical_loss": 3.7297657957221784,
"tokens_seen": 798687232
},
{
"epoch": 0.24,
"learning_rate": 0.0003828284854563691,
"loss": 2.7207,
"theoretical_loss": 3.729735449989394,
"tokens_seen": 798752768
},
{
"epoch": 0.24,
"learning_rate": 0.00038281845536609834,
"loss": 2.7617,
"theoretical_loss": 3.729705107443391,
"tokens_seen": 798818304
},
{
"epoch": 0.24,
"learning_rate": 0.00038280842527582746,
"loss": 2.9106,
"theoretical_loss": 3.7296747680835747,
"tokens_seen": 798883840
},
{
"epoch": 0.24,
"learning_rate": 0.0003827983951855567,
"loss": 2.6975,
"theoretical_loss": 3.7296444319093487,
"tokens_seen": 798949376
},
{
"epoch": 0.24,
"learning_rate": 0.0003827883650952859,
"loss": 2.712,
"theoretical_loss": 3.7296140989201163,
"tokens_seen": 799014912
},
{
"epoch": 0.24,
"learning_rate": 0.00038277833500501506,
"loss": 2.9892,
"theoretical_loss": 3.7295837691152824,
"tokens_seen": 799080448
},
{
"epoch": 0.24,
"learning_rate": 0.00038276830491474424,
"loss": 2.765,
"theoretical_loss": 3.729553442494252,
"tokens_seen": 799145984
},
{
"epoch": 0.24,
"learning_rate": 0.0003827582748244735,
"loss": 2.7596,
"theoretical_loss": 3.72952311905643,
"tokens_seen": 799211520
},
{
"epoch": 0.24,
"learning_rate": 0.0003827482447342026,
"loss": 2.7614,
"theoretical_loss": 3.7294927988012203,
"tokens_seen": 799277056
},
{
"epoch": 0.24,
"learning_rate": 0.00038273821464393184,
"loss": 2.7994,
"theoretical_loss": 3.729462481728029,
"tokens_seen": 799342592
},
{
"epoch": 0.24,
"learning_rate": 0.00038272818455366097,
"loss": 2.9987,
"theoretical_loss": 3.7294321678362596,
"tokens_seen": 799408128
},
{
"epoch": 0.24,
"learning_rate": 0.0003827181544633902,
"loss": 2.9538,
"theoretical_loss": 3.72940185712532,
"tokens_seen": 799473664
},
{
"epoch": 0.24,
"objective/train/docs_used": 460885,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3522489070892334,
"objective/train/theoretical_loss": 3.729371549594614,
"objective/train/tokens_used": 819999200,
"theoretical_loss": 3.729371549594614,
"tokens_seen": 799539200
},
{
"epoch": 0.24,
"learning_rate": 0.0003827081243731194,
"loss": 2.5521,
"theoretical_loss": 3.729371549594614,
"tokens_seen": 799539200
},
{
"epoch": 0.24,
"learning_rate": 0.00038269809428284856,
"loss": 2.4308,
"theoretical_loss": 3.729341245243547,
"tokens_seen": 799604736
},
{
"epoch": 0.24,
"learning_rate": 0.00038268806419257775,
"loss": 2.808,
"theoretical_loss": 3.7293109440715266,
"tokens_seen": 799670272
},
{
"epoch": 0.24,
"learning_rate": 0.0003826780341023069,
"loss": 2.7623,
"theoretical_loss": 3.729280646077957,
"tokens_seen": 799735808
},
{
"epoch": 0.24,
"learning_rate": 0.0003826680040120361,
"loss": 2.6722,
"theoretical_loss": 3.7292503512622455,
"tokens_seen": 799801344
},
{
"epoch": 0.24,
"learning_rate": 0.00038265797392176534,
"loss": 2.8779,
"theoretical_loss": 3.729220059623798,
"tokens_seen": 799866880
},
{
"epoch": 0.24,
"learning_rate": 0.00038264794383149447,
"loss": 2.9275,
"theoretical_loss": 3.729189771162021,
"tokens_seen": 799932416
},
{
"epoch": 0.24,
"learning_rate": 0.0003826379137412237,
"loss": 2.9871,
"theoretical_loss": 3.7291594858763215,
"tokens_seen": 799997952
},
{
"epoch": 0.24,
"learning_rate": 0.00038262788365095283,
"loss": 3.1188,
"theoretical_loss": 3.7291292037661057,
"tokens_seen": 800063488
},
{
"epoch": 0.24,
"learning_rate": 0.00038261785356068207,
"loss": 2.8238,
"theoretical_loss": 3.7290989248307813,
"tokens_seen": 800129024
},
{
"epoch": 0.24,
"learning_rate": 0.00038260782347041125,
"loss": 2.8839,
"theoretical_loss": 3.729068649069754,
"tokens_seen": 800194560
},
{
"epoch": 0.24,
"learning_rate": 0.00038259779338014043,
"loss": 3.1508,
"theoretical_loss": 3.729038376482433,
"tokens_seen": 800260096
},
{
"epoch": 0.24,
"learning_rate": 0.0003825877632898696,
"loss": 2.6426,
"theoretical_loss": 3.729008107068225,
"tokens_seen": 800325632
},
{
"epoch": 0.24,
"learning_rate": 0.00038257773319959885,
"loss": 2.7861,
"theoretical_loss": 3.728977840826537,
"tokens_seen": 800391168
},
{
"epoch": 0.24,
"learning_rate": 0.000382567703109328,
"loss": 3.0881,
"theoretical_loss": 3.7289475777567773,
"tokens_seen": 800456704
},
{
"epoch": 0.24,
"learning_rate": 0.0003825576730190572,
"loss": 2.5042,
"theoretical_loss": 3.728917317858354,
"tokens_seen": 800522240
},
{
"epoch": 0.24,
"learning_rate": 0.00038254764292878634,
"loss": 2.7613,
"theoretical_loss": 3.7288870611306746,
"tokens_seen": 800587776
},
{
"epoch": 0.24,
"learning_rate": 0.00038253761283851557,
"loss": 2.8506,
"theoretical_loss": 3.7288568075731474,
"tokens_seen": 800653312
},
{
"epoch": 0.24,
"learning_rate": 0.00038252758274824475,
"loss": 2.7841,
"theoretical_loss": 3.7288265571851817,
"tokens_seen": 800718848
},
{
"epoch": 0.24,
"learning_rate": 0.00038251755265797393,
"loss": 2.844,
"theoretical_loss": 3.728796309966185,
"tokens_seen": 800784384
},
{
"epoch": 0.24,
"learning_rate": 0.0003825075225677031,
"loss": 2.5967,
"theoretical_loss": 3.7287660659155666,
"tokens_seen": 800849920
},
{
"epoch": 0.24,
"learning_rate": 0.0003824974924774323,
"loss": 2.9147,
"theoretical_loss": 3.7287358250327354,
"tokens_seen": 800915456
},
{
"epoch": 0.24,
"learning_rate": 0.0003824874623871615,
"loss": 2.8379,
"theoretical_loss": 3.7287055873171,
"tokens_seen": 800980992
},
{
"epoch": 0.24,
"learning_rate": 0.0003824774322968907,
"loss": 2.9526,
"theoretical_loss": 3.7286753527680703,
"tokens_seen": 801046528
},
{
"epoch": 0.24,
"learning_rate": 0.00038246740220661984,
"loss": 2.6213,
"theoretical_loss": 3.7286451213850547,
"tokens_seen": 801112064
},
{
"epoch": 0.24,
"objective/train/docs_used": 461399,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8124303817749023,
"objective/train/theoretical_loss": 3.7286148931674634,
"objective/train/tokens_used": 821637600,
"theoretical_loss": 3.7286148931674634,
"tokens_seen": 801177600
},
{
"epoch": 0.24,
"learning_rate": 0.0003824573721163491,
"loss": 2.6891,
"theoretical_loss": 3.7286148931674634,
"tokens_seen": 801177600
},
{
"epoch": 0.24,
"learning_rate": 0.0003824473420260782,
"loss": 2.8056,
"theoretical_loss": 3.728584668114706,
"tokens_seen": 801243136
},
{
"epoch": 0.24,
"learning_rate": 0.00038243731193580744,
"loss": 2.7398,
"theoretical_loss": 3.728554446226192,
"tokens_seen": 801308672
},
{
"epoch": 0.24,
"learning_rate": 0.0003824272818455366,
"loss": 2.8626,
"theoretical_loss": 3.7285242275013317,
"tokens_seen": 801374208
},
{
"epoch": 0.24,
"learning_rate": 0.0003824172517552658,
"loss": 2.8589,
"theoretical_loss": 3.728494011939535,
"tokens_seen": 801439744
},
{
"epoch": 0.24,
"learning_rate": 0.000382407221664995,
"loss": 2.9172,
"theoretical_loss": 3.7284637995402123,
"tokens_seen": 801505280
},
{
"epoch": 0.24,
"learning_rate": 0.0003823971915747242,
"loss": 2.7267,
"theoretical_loss": 3.7284335903027745,
"tokens_seen": 801570816
},
{
"epoch": 0.24,
"learning_rate": 0.00038238716148445334,
"loss": 2.8227,
"theoretical_loss": 3.728403384226632,
"tokens_seen": 801636352
},
{
"epoch": 0.24,
"learning_rate": 0.0003823771313941826,
"loss": 2.9309,
"theoretical_loss": 3.728373181311195,
"tokens_seen": 801701888
},
{
"epoch": 0.24,
"learning_rate": 0.0003823671013039117,
"loss": 2.8779,
"theoretical_loss": 3.728342981555875,
"tokens_seen": 801767424
},
{
"epoch": 0.24,
"learning_rate": 0.00038235707121364094,
"loss": 2.9116,
"theoretical_loss": 3.7283127849600826,
"tokens_seen": 801832960
},
{
"epoch": 0.24,
"learning_rate": 0.0003823470411233701,
"loss": 2.6001,
"theoretical_loss": 3.7282825915232296,
"tokens_seen": 801898496
},
{
"epoch": 0.24,
"learning_rate": 0.0003823370110330993,
"loss": 2.9938,
"theoretical_loss": 3.728252401244727,
"tokens_seen": 801964032
},
{
"epoch": 0.24,
"learning_rate": 0.0003823269809428285,
"loss": 2.6885,
"theoretical_loss": 3.7282222141239867,
"tokens_seen": 802029568
},
{
"epoch": 0.24,
"learning_rate": 0.00038231695085255766,
"loss": 2.9653,
"theoretical_loss": 3.7281920301604208,
"tokens_seen": 802095104
},
{
"epoch": 0.24,
"learning_rate": 0.00038230692076228685,
"loss": 2.6412,
"theoretical_loss": 3.72816184935344,
"tokens_seen": 802160640
},
{
"epoch": 0.24,
"learning_rate": 0.0003822968906720161,
"loss": 2.6503,
"theoretical_loss": 3.728131671702457,
"tokens_seen": 802226176
},
{
"epoch": 0.24,
"learning_rate": 0.0003822868605817452,
"loss": 2.7236,
"theoretical_loss": 3.728101497206884,
"tokens_seen": 802291712
},
{
"epoch": 0.24,
"learning_rate": 0.00038227683049147444,
"loss": 2.7188,
"theoretical_loss": 3.728071325866134,
"tokens_seen": 802357248
},
{
"epoch": 0.24,
"learning_rate": 0.00038226680040120357,
"loss": 2.6648,
"theoretical_loss": 3.7280411576796184,
"tokens_seen": 802422784
},
{
"epoch": 0.24,
"learning_rate": 0.0003822567703109328,
"loss": 2.6355,
"theoretical_loss": 3.7280109926467504,
"tokens_seen": 802488320
},
{
"epoch": 0.24,
"learning_rate": 0.000382246740220662,
"loss": 2.7441,
"theoretical_loss": 3.727980830766943,
"tokens_seen": 802553856
},
{
"epoch": 0.24,
"learning_rate": 0.00038223671013039117,
"loss": 2.7275,
"theoretical_loss": 3.7279506720396083,
"tokens_seen": 802619392
},
{
"epoch": 0.24,
"learning_rate": 0.00038222668004012035,
"loss": 2.7602,
"theoretical_loss": 3.7279205164641604,
"tokens_seen": 802684928
},
{
"epoch": 0.24,
"learning_rate": 0.0003822166499498496,
"loss": 2.7808,
"theoretical_loss": 3.7278903640400127,
"tokens_seen": 802750464
},
{
"epoch": 0.24,
"objective/train/docs_used": 462084,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6324245929718018,
"objective/train/theoretical_loss": 3.7278602147665776,
"objective/train/tokens_used": 823276000,
"theoretical_loss": 3.7278602147665776,
"tokens_seen": 802816000
},
{
"epoch": 0.24,
"learning_rate": 0.0003822066198595787,
"loss": 2.6443,
"theoretical_loss": 3.7278602147665776,
"tokens_seen": 802816000
},
{
"epoch": 0.24,
"learning_rate": 0.00038219658976930795,
"loss": 2.851,
"theoretical_loss": 3.7278300686432697,
"tokens_seen": 802881536
},
{
"epoch": 0.24,
"learning_rate": 0.0003821865596790371,
"loss": 2.7112,
"theoretical_loss": 3.7277999256695025,
"tokens_seen": 802947072
},
{
"epoch": 0.24,
"learning_rate": 0.0003821765295887663,
"loss": 2.888,
"theoretical_loss": 3.7277697858446897,
"tokens_seen": 803012608
},
{
"epoch": 0.24,
"learning_rate": 0.0003821664994984955,
"loss": 2.9293,
"theoretical_loss": 3.727739649168245,
"tokens_seen": 803078144
},
{
"epoch": 0.24,
"learning_rate": 0.00038215646940822467,
"loss": 2.6849,
"theoretical_loss": 3.7277095156395843,
"tokens_seen": 803143680
},
{
"epoch": 0.24,
"learning_rate": 0.00038214643931795385,
"loss": 2.6406,
"theoretical_loss": 3.7276793852581203,
"tokens_seen": 803209216
},
{
"epoch": 0.24,
"learning_rate": 0.00038213640922768303,
"loss": 2.6306,
"theoretical_loss": 3.727649258023268,
"tokens_seen": 803274752
},
{
"epoch": 0.24,
"learning_rate": 0.00038212637913741227,
"loss": 2.6947,
"theoretical_loss": 3.7276191339344424,
"tokens_seen": 803340288
},
{
"epoch": 0.24,
"learning_rate": 0.00038211634904714145,
"loss": 2.7612,
"theoretical_loss": 3.727589012991058,
"tokens_seen": 803405824
},
{
"epoch": 0.24,
"learning_rate": 0.00038210631895687063,
"loss": 2.8112,
"theoretical_loss": 3.72755889519253,
"tokens_seen": 803471360
},
{
"epoch": 0.24,
"learning_rate": 0.0003820962888665998,
"loss": 2.9324,
"theoretical_loss": 3.7275287805382744,
"tokens_seen": 803536896
},
{
"epoch": 0.24,
"learning_rate": 0.00038208625877632905,
"loss": 2.6108,
"theoretical_loss": 3.7274986690277045,
"tokens_seen": 803602432
},
{
"epoch": 0.24,
"learning_rate": 0.0003820762286860582,
"loss": 3.0453,
"theoretical_loss": 3.727468560660238,
"tokens_seen": 803667968
},
{
"epoch": 0.24,
"learning_rate": 0.0003820661985957874,
"loss": 2.7045,
"theoretical_loss": 3.727438455435289,
"tokens_seen": 803733504
},
{
"epoch": 0.24,
"learning_rate": 0.00038205616850551654,
"loss": 2.9285,
"theoretical_loss": 3.727408353352274,
"tokens_seen": 803799040
},
{
"epoch": 0.24,
"learning_rate": 0.00038204613841524577,
"loss": 2.7763,
"theoretical_loss": 3.727378254410609,
"tokens_seen": 803864576
},
{
"epoch": 0.24,
"learning_rate": 0.00038203610832497495,
"loss": 2.7055,
"theoretical_loss": 3.72734815860971,
"tokens_seen": 803930112
},
{
"epoch": 0.24,
"learning_rate": 0.00038202607823470413,
"loss": 2.8921,
"theoretical_loss": 3.7273180659489933,
"tokens_seen": 803995648
},
{
"epoch": 0.24,
"learning_rate": 0.0003820160481444333,
"loss": 2.8998,
"theoretical_loss": 3.727287976427875,
"tokens_seen": 804061184
},
{
"epoch": 0.24,
"learning_rate": 0.0003820060180541625,
"loss": 2.883,
"theoretical_loss": 3.727257890045772,
"tokens_seen": 804126720
},
{
"epoch": 0.24,
"learning_rate": 0.0003819959879638917,
"loss": 2.9426,
"theoretical_loss": 3.7272278068021008,
"tokens_seen": 804192256
},
{
"epoch": 0.24,
"learning_rate": 0.0003819859578736209,
"loss": 2.8635,
"theoretical_loss": 3.7271977266962786,
"tokens_seen": 804257792
},
{
"epoch": 0.24,
"learning_rate": 0.00038197592778335004,
"loss": 2.6897,
"theoretical_loss": 3.727167649727723,
"tokens_seen": 804323328
},
{
"epoch": 0.24,
"learning_rate": 0.0003819658976930793,
"loss": 2.825,
"theoretical_loss": 3.7271375758958496,
"tokens_seen": 804388864
},
{
"epoch": 0.24,
"objective/train/docs_used": 463313,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.007526397705078,
"objective/train/theoretical_loss": 3.7271075052000775,
"objective/train/tokens_used": 824914400,
"theoretical_loss": 3.7271075052000775,
"tokens_seen": 804454400
},
{
"epoch": 0.24,
"learning_rate": 0.0003819558676028084,
"loss": 2.7619,
"theoretical_loss": 3.7271075052000775,
"tokens_seen": 804454400
},
{
"epoch": 0.24,
"learning_rate": 0.00038194583751253764,
"loss": 2.9055,
"theoretical_loss": 3.7270774376398226,
"tokens_seen": 804519936
},
{
"epoch": 0.24,
"learning_rate": 0.0003819358074222668,
"loss": 2.9252,
"theoretical_loss": 3.727047373214504,
"tokens_seen": 804585472
},
{
"epoch": 0.24,
"learning_rate": 0.000381925777331996,
"loss": 2.7044,
"theoretical_loss": 3.727017311923539,
"tokens_seen": 804651008
},
{
"epoch": 0.24,
"learning_rate": 0.0003819157472417252,
"loss": 2.6669,
"theoretical_loss": 3.726987253766345,
"tokens_seen": 804716544
},
{
"epoch": 0.24,
"learning_rate": 0.0003819057171514544,
"loss": 2.8518,
"theoretical_loss": 3.7269571987423413,
"tokens_seen": 804782080
},
{
"epoch": 0.24,
"learning_rate": 0.00038189568706118354,
"loss": 2.8163,
"theoretical_loss": 3.726927146850945,
"tokens_seen": 804847616
},
{
"epoch": 0.24,
"learning_rate": 0.0003818856569709128,
"loss": 2.8407,
"theoretical_loss": 3.726897098091576,
"tokens_seen": 804913152
},
{
"epoch": 0.24,
"learning_rate": 0.0003818756268806419,
"loss": 2.6759,
"theoretical_loss": 3.7268670524636507,
"tokens_seen": 804978688
},
{
"epoch": 0.24,
"learning_rate": 0.00038186559679037114,
"loss": 3.0038,
"theoretical_loss": 3.7268370099665904,
"tokens_seen": 805044224
},
{
"epoch": 0.24,
"learning_rate": 0.0003818555667001003,
"loss": 2.7643,
"theoretical_loss": 3.7268069705998124,
"tokens_seen": 805109760
},
{
"epoch": 0.24,
"learning_rate": 0.0003818455366098295,
"loss": 2.6087,
"theoretical_loss": 3.7267769343627357,
"tokens_seen": 805175296
},
{
"epoch": 0.24,
"learning_rate": 0.0003818355065195587,
"loss": 2.881,
"theoretical_loss": 3.726746901254781,
"tokens_seen": 805240832
},
{
"epoch": 0.24,
"learning_rate": 0.00038182547642928786,
"loss": 2.6553,
"theoretical_loss": 3.726716871275366,
"tokens_seen": 805306368
},
{
"epoch": 0.24,
"learning_rate": 0.00038181544633901705,
"loss": 2.9477,
"theoretical_loss": 3.726686844423911,
"tokens_seen": 805371904
},
{
"epoch": 0.24,
"learning_rate": 0.0003818054162487463,
"loss": 3.0997,
"theoretical_loss": 3.726656820699836,
"tokens_seen": 805437440
},
{
"epoch": 0.24,
"learning_rate": 0.0003817953861584754,
"loss": 2.772,
"theoretical_loss": 3.72662680010256,
"tokens_seen": 805502976
},
{
"epoch": 0.24,
"learning_rate": 0.00038178535606820464,
"loss": 3.038,
"theoretical_loss": 3.726596782631504,
"tokens_seen": 805568512
},
{
"epoch": 0.24,
"learning_rate": 0.00038177532597793377,
"loss": 2.936,
"theoretical_loss": 3.7265667682860872,
"tokens_seen": 805634048
},
{
"epoch": 0.24,
"learning_rate": 0.000381765295887663,
"loss": 2.8583,
"theoretical_loss": 3.7265367570657304,
"tokens_seen": 805699584
},
{
"epoch": 0.24,
"learning_rate": 0.0003817552657973922,
"loss": 2.8806,
"theoretical_loss": 3.7265067489698547,
"tokens_seen": 805765120
},
{
"epoch": 0.24,
"learning_rate": 0.00038174523570712137,
"loss": 2.9012,
"theoretical_loss": 3.726476743997879,
"tokens_seen": 805830656
},
{
"epoch": 0.24,
"learning_rate": 0.00038173520561685055,
"loss": 2.795,
"theoretical_loss": 3.7264467421492258,
"tokens_seen": 805896192
},
{
"epoch": 0.24,
"learning_rate": 0.0003817251755265798,
"loss": 2.7376,
"theoretical_loss": 3.7264167434233153,
"tokens_seen": 805961728
},
{
"epoch": 0.24,
"learning_rate": 0.0003817151454363089,
"loss": 2.8773,
"theoretical_loss": 3.7263867478195687,
"tokens_seen": 806027264
},
{
"epoch": 0.24,
"objective/train/docs_used": 464591,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2207796573638916,
"objective/train/theoretical_loss": 3.726356755337407,
"objective/train/tokens_used": 826552800,
"theoretical_loss": 3.726356755337407,
"tokens_seen": 806092800
},
{
"epoch": 0.24,
"learning_rate": 0.00038170511534603815,
"loss": 2.9218,
"theoretical_loss": 3.726356755337407,
"tokens_seen": 806092800
},
{
"epoch": 0.24,
"learning_rate": 0.0003816950852557673,
"loss": 2.7547,
"theoretical_loss": 3.7263267659762516,
"tokens_seen": 806158336
},
{
"epoch": 0.24,
"learning_rate": 0.0003816850551654965,
"loss": 2.8154,
"theoretical_loss": 3.7262967797355246,
"tokens_seen": 806223872
},
{
"epoch": 0.24,
"learning_rate": 0.0003816750250752257,
"loss": 2.6115,
"theoretical_loss": 3.7262667966146474,
"tokens_seen": 806289408
},
{
"epoch": 0.24,
"learning_rate": 0.00038166499498495487,
"loss": 2.5836,
"theoretical_loss": 3.7262368166130413,
"tokens_seen": 806354944
},
{
"epoch": 0.24,
"learning_rate": 0.00038165496489468405,
"loss": 2.853,
"theoretical_loss": 3.7262068397301293,
"tokens_seen": 806420480
},
{
"epoch": 0.24,
"learning_rate": 0.00038164493480441323,
"loss": 3.042,
"theoretical_loss": 3.726176865965333,
"tokens_seen": 806486016
},
{
"epoch": 0.24,
"learning_rate": 0.0003816349047141424,
"loss": 2.8644,
"theoretical_loss": 3.7261468953180747,
"tokens_seen": 806551552
},
{
"epoch": 0.24,
"learning_rate": 0.00038162487462387165,
"loss": 2.733,
"theoretical_loss": 3.7261169277877766,
"tokens_seen": 806617088
},
{
"epoch": 0.24,
"learning_rate": 0.0003816148445336008,
"loss": 2.8458,
"theoretical_loss": 3.7260869633738625,
"tokens_seen": 806682624
},
{
"epoch": 0.24,
"learning_rate": 0.00038160481444333,
"loss": 2.7422,
"theoretical_loss": 3.7260570020757537,
"tokens_seen": 806748160
},
{
"epoch": 0.24,
"learning_rate": 0.00038159478435305914,
"loss": 2.8559,
"theoretical_loss": 3.7260270438928744,
"tokens_seen": 806813696
},
{
"epoch": 0.24,
"learning_rate": 0.0003815847542627884,
"loss": 2.9231,
"theoretical_loss": 3.7259970888246468,
"tokens_seen": 806879232
},
{
"epoch": 0.24,
"learning_rate": 0.00038157472417251756,
"loss": 3.1096,
"theoretical_loss": 3.7259671368704943,
"tokens_seen": 806944768
},
{
"epoch": 0.24,
"learning_rate": 0.00038156469408224674,
"loss": 2.8669,
"theoretical_loss": 3.7259371880298406,
"tokens_seen": 807010304
},
{
"epoch": 0.24,
"learning_rate": 0.0003815546639919759,
"loss": 3.0003,
"theoretical_loss": 3.7259072423021085,
"tokens_seen": 807075840
},
{
"epoch": 0.24,
"learning_rate": 0.00038154463390170515,
"loss": 3.0098,
"theoretical_loss": 3.725877299686723,
"tokens_seen": 807141376
},
{
"epoch": 0.24,
"learning_rate": 0.0003815346038114343,
"loss": 3.0327,
"theoretical_loss": 3.7258473601831072,
"tokens_seen": 807206912
},
{
"epoch": 0.24,
"learning_rate": 0.0003815245737211635,
"loss": 2.7805,
"theoretical_loss": 3.725817423790685,
"tokens_seen": 807272448
},
{
"epoch": 0.24,
"learning_rate": 0.00038151454363089264,
"loss": 2.7118,
"theoretical_loss": 3.72578749050888,
"tokens_seen": 807337984
},
{
"epoch": 0.24,
"learning_rate": 0.0003815045135406219,
"loss": 2.9266,
"theoretical_loss": 3.7257575603371187,
"tokens_seen": 807403520
},
{
"epoch": 0.24,
"learning_rate": 0.00038149448345035106,
"loss": 2.8271,
"theoretical_loss": 3.7257276332748233,
"tokens_seen": 807469056
},
{
"epoch": 0.24,
"learning_rate": 0.00038148445336008024,
"loss": 2.882,
"theoretical_loss": 3.725697709321419,
"tokens_seen": 807534592
},
{
"epoch": 0.24,
"learning_rate": 0.0003814744232698094,
"loss": 2.6269,
"theoretical_loss": 3.725667788476331,
"tokens_seen": 807600128
},
{
"epoch": 0.24,
"learning_rate": 0.0003814643931795386,
"loss": 2.9558,
"theoretical_loss": 3.7256378707389843,
"tokens_seen": 807665664
},
{
"epoch": 0.24,
"objective/train/docs_used": 465345,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9331934452056885,
"objective/train/theoretical_loss": 3.725607956108804,
"objective/train/tokens_used": 828191200,
"theoretical_loss": 3.725607956108804,
"tokens_seen": 807731200
},
{
"epoch": 0.24,
"learning_rate": 0.0003814543630892678,
"loss": 3.0677,
"theoretical_loss": 3.725607956108804,
"tokens_seen": 807731200
},
{
"epoch": 0.24,
"learning_rate": 0.000381444332998997,
"loss": 2.7662,
"theoretical_loss": 3.7255780445852142,
"tokens_seen": 807796736
},
{
"epoch": 0.24,
"learning_rate": 0.00038143430290872614,
"loss": 3.0367,
"theoretical_loss": 3.7255481361676415,
"tokens_seen": 807862272
},
{
"epoch": 0.24,
"learning_rate": 0.0003814242728184554,
"loss": 2.9789,
"theoretical_loss": 3.725518230855511,
"tokens_seen": 807927808
},
{
"epoch": 0.24,
"learning_rate": 0.00038141424272818456,
"loss": 2.8395,
"theoretical_loss": 3.7254883286482485,
"tokens_seen": 807993344
},
{
"epoch": 0.24,
"learning_rate": 0.00038140421263791374,
"loss": 2.7475,
"theoretical_loss": 3.72545842954528,
"tokens_seen": 808058880
},
{
"epoch": 0.24,
"learning_rate": 0.0003813941825476429,
"loss": 2.8677,
"theoretical_loss": 3.7254285335460313,
"tokens_seen": 808124416
},
{
"epoch": 0.24,
"learning_rate": 0.0003813841524573721,
"loss": 2.8224,
"theoretical_loss": 3.725398640649929,
"tokens_seen": 808189952
},
{
"epoch": 0.24,
"learning_rate": 0.00038137412236710134,
"loss": 2.7711,
"theoretical_loss": 3.725368750856398,
"tokens_seen": 808255488
},
{
"epoch": 0.24,
"learning_rate": 0.0003813640922768305,
"loss": 2.5627,
"theoretical_loss": 3.7253388641648666,
"tokens_seen": 808321024
},
{
"epoch": 0.24,
"learning_rate": 0.0003813540621865597,
"loss": 2.7109,
"theoretical_loss": 3.72530898057476,
"tokens_seen": 808386560
},
{
"epoch": 0.24,
"learning_rate": 0.0003813440320962889,
"loss": 2.723,
"theoretical_loss": 3.7252791000855057,
"tokens_seen": 808452096
},
{
"epoch": 0.25,
"learning_rate": 0.00038133400200601806,
"loss": 2.9187,
"theoretical_loss": 3.7252492226965304,
"tokens_seen": 808517632
},
{
"epoch": 0.25,
"learning_rate": 0.00038132397191574725,
"loss": 2.8077,
"theoretical_loss": 3.7252193484072613,
"tokens_seen": 808583168
},
{
"epoch": 0.25,
"learning_rate": 0.0003813139418254765,
"loss": 2.858,
"theoretical_loss": 3.7251894772171257,
"tokens_seen": 808648704
},
{
"epoch": 0.25,
"learning_rate": 0.0003813039117352056,
"loss": 2.7209,
"theoretical_loss": 3.72515960912555,
"tokens_seen": 808714240
},
{
"epoch": 0.25,
"learning_rate": 0.00038129388164493484,
"loss": 2.9029,
"theoretical_loss": 3.725129744131963,
"tokens_seen": 808779776
},
{
"epoch": 0.25,
"learning_rate": 0.00038128385155466397,
"loss": 2.6208,
"theoretical_loss": 3.725099882235792,
"tokens_seen": 808845312
},
{
"epoch": 0.25,
"learning_rate": 0.0003812738214643932,
"loss": 2.9088,
"theoretical_loss": 3.725070023436465,
"tokens_seen": 808910848
},
{
"epoch": 0.25,
"learning_rate": 0.0003812637913741224,
"loss": 2.9519,
"theoretical_loss": 3.7250401677334093,
"tokens_seen": 808976384
},
{
"epoch": 0.25,
"learning_rate": 0.00038125376128385157,
"loss": 2.7065,
"theoretical_loss": 3.7250103151260534,
"tokens_seen": 809041920
},
{
"epoch": 0.25,
"learning_rate": 0.00038124373119358075,
"loss": 2.9344,
"theoretical_loss": 3.7249804656138257,
"tokens_seen": 809107456
},
{
"epoch": 0.25,
"learning_rate": 0.00038123370110331,
"loss": 2.7836,
"theoretical_loss": 3.7249506191961546,
"tokens_seen": 809172992
},
{
"epoch": 0.25,
"learning_rate": 0.0003812236710130391,
"loss": 2.8671,
"theoretical_loss": 3.7249207758724685,
"tokens_seen": 809238528
},
{
"epoch": 0.25,
"learning_rate": 0.00038121364092276835,
"loss": 2.8609,
"theoretical_loss": 3.7248909356421964,
"tokens_seen": 809304064
},
{
"epoch": 0.25,
"objective/train/docs_used": 466409,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.010345935821533,
"objective/train/theoretical_loss": 3.724861098504767,
"objective/train/tokens_used": 829829600,
"theoretical_loss": 3.724861098504767,
"tokens_seen": 809369600
},
{
"epoch": 0.25,
"learning_rate": 0.0003812036108324975,
"loss": 2.8907,
"theoretical_loss": 3.724861098504767,
"tokens_seen": 809369600
},
{
"epoch": 0.25,
"learning_rate": 0.0003811935807422267,
"loss": 2.7471,
"theoretical_loss": 3.7248312644596098,
"tokens_seen": 809435136
},
{
"epoch": 0.25,
"learning_rate": 0.0003811835506519559,
"loss": 3.0079,
"theoretical_loss": 3.724801433506153,
"tokens_seen": 809500672
},
{
"epoch": 0.25,
"learning_rate": 0.00038117352056168507,
"loss": 2.9426,
"theoretical_loss": 3.724771605643827,
"tokens_seen": 809566208
},
{
"epoch": 0.25,
"learning_rate": 0.00038116349047141425,
"loss": 3.0126,
"theoretical_loss": 3.7247417808720606,
"tokens_seen": 809631744
},
{
"epoch": 0.25,
"learning_rate": 0.00038115346038114343,
"loss": 2.5553,
"theoretical_loss": 3.7247119591902837,
"tokens_seen": 809697280
},
{
"epoch": 0.25,
"learning_rate": 0.0003811434302908726,
"loss": 2.8839,
"theoretical_loss": 3.7246821405979267,
"tokens_seen": 809762816
},
{
"epoch": 0.25,
"learning_rate": 0.00038113340020060185,
"loss": 2.8646,
"theoretical_loss": 3.724652325094418,
"tokens_seen": 809828352
},
{
"epoch": 0.25,
"learning_rate": 0.000381123370110331,
"loss": 2.5228,
"theoretical_loss": 3.724622512679189,
"tokens_seen": 809893888
},
{
"epoch": 0.25,
"learning_rate": 0.0003811133400200602,
"loss": 2.6679,
"theoretical_loss": 3.7245927033516697,
"tokens_seen": 809959424
},
{
"epoch": 0.25,
"learning_rate": 0.00038110330992978934,
"loss": 2.7885,
"theoretical_loss": 3.7245628971112907,
"tokens_seen": 810024960
},
{
"epoch": 0.25,
"learning_rate": 0.0003810932798395186,
"loss": 3.0271,
"theoretical_loss": 3.724533093957482,
"tokens_seen": 810090496
},
{
"epoch": 0.25,
"learning_rate": 0.00038108324974924776,
"loss": 2.6992,
"theoretical_loss": 3.7245032938896743,
"tokens_seen": 810156032
},
{
"epoch": 0.25,
"learning_rate": 0.00038107321965897694,
"loss": 2.8992,
"theoretical_loss": 3.724473496907299,
"tokens_seen": 810221568
},
{
"epoch": 0.25,
"learning_rate": 0.0003810631895687061,
"loss": 2.8084,
"theoretical_loss": 3.724443703009787,
"tokens_seen": 810287104
},
{
"epoch": 0.25,
"learning_rate": 0.00038105315947843535,
"loss": 2.9769,
"theoretical_loss": 3.724413912196569,
"tokens_seen": 810352640
},
{
"epoch": 0.25,
"learning_rate": 0.0003810431293881645,
"loss": 2.8847,
"theoretical_loss": 3.7243841244670772,
"tokens_seen": 810418176
},
{
"epoch": 0.25,
"learning_rate": 0.0003810330992978937,
"loss": 3.009,
"theoretical_loss": 3.724354339820742,
"tokens_seen": 810483712
},
{
"epoch": 0.25,
"learning_rate": 0.00038102306920762284,
"loss": 2.6281,
"theoretical_loss": 3.7243245582569955,
"tokens_seen": 810549248
},
{
"epoch": 0.25,
"learning_rate": 0.0003810130391173521,
"loss": 3.0026,
"theoretical_loss": 3.7242947797752697,
"tokens_seen": 810614784
},
{
"epoch": 0.25,
"learning_rate": 0.00038100300902708126,
"loss": 2.8076,
"theoretical_loss": 3.7242650043749963,
"tokens_seen": 810680320
},
{
"epoch": 0.25,
"learning_rate": 0.00038099297893681044,
"loss": 2.9244,
"theoretical_loss": 3.724235232055607,
"tokens_seen": 810745856
},
{
"epoch": 0.25,
"learning_rate": 0.0003809829488465396,
"loss": 2.8115,
"theoretical_loss": 3.7242054628165344,
"tokens_seen": 810811392
},
{
"epoch": 0.25,
"learning_rate": 0.0003809729187562688,
"loss": 2.8842,
"theoretical_loss": 3.724175696657211,
"tokens_seen": 810876928
},
{
"epoch": 0.25,
"learning_rate": 0.000380962888665998,
"loss": 3.0004,
"theoretical_loss": 3.724145933577069,
"tokens_seen": 810942464
},
{
"epoch": 0.25,
"objective/train/docs_used": 466961,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.270719289779663,
"objective/train/theoretical_loss": 3.724116173575542,
"objective/train/tokens_used": 831468000,
"theoretical_loss": 3.724116173575542,
"tokens_seen": 811008000
},
{
"epoch": 0.25,
"learning_rate": 0.0003809528585757272,
"loss": 2.9679,
"theoretical_loss": 3.724116173575542,
"tokens_seen": 811008000
},
{
"epoch": 0.25,
"learning_rate": 0.00038094282848545635,
"loss": 2.6337,
"theoretical_loss": 3.724086416652061,
"tokens_seen": 811073536
},
{
"epoch": 0.25,
"learning_rate": 0.0003809327983951856,
"loss": 2.6586,
"theoretical_loss": 3.7240566628060607,
"tokens_seen": 811139072
},
{
"epoch": 0.25,
"learning_rate": 0.00038092276830491476,
"loss": 2.6643,
"theoretical_loss": 3.7240269120369733,
"tokens_seen": 811204608
},
{
"epoch": 0.25,
"learning_rate": 0.00038091273821464394,
"loss": 2.8427,
"theoretical_loss": 3.7239971643442322,
"tokens_seen": 811270144
},
{
"epoch": 0.25,
"learning_rate": 0.0003809027081243731,
"loss": 2.8732,
"theoretical_loss": 3.7239674197272707,
"tokens_seen": 811335680
},
{
"epoch": 0.25,
"learning_rate": 0.0003808926780341023,
"loss": 2.7359,
"theoretical_loss": 3.7239376781855222,
"tokens_seen": 811401216
},
{
"epoch": 0.25,
"learning_rate": 0.0003808826479438315,
"loss": 2.7734,
"theoretical_loss": 3.723907939718422,
"tokens_seen": 811466752
},
{
"epoch": 0.25,
"learning_rate": 0.0003808726178535607,
"loss": 2.6518,
"theoretical_loss": 3.723878204325402,
"tokens_seen": 811532288
},
{
"epoch": 0.25,
"learning_rate": 0.00038086258776328985,
"loss": 3.0589,
"theoretical_loss": 3.7238484720058973,
"tokens_seen": 811597824
},
{
"epoch": 0.25,
"learning_rate": 0.0003808525576730191,
"loss": 2.8884,
"theoretical_loss": 3.7238187427593417,
"tokens_seen": 811663360
},
{
"epoch": 0.25,
"learning_rate": 0.0003808425275827482,
"loss": 2.8258,
"theoretical_loss": 3.7237890165851693,
"tokens_seen": 811728896
},
{
"epoch": 0.25,
"learning_rate": 0.00038083249749247745,
"loss": 2.8689,
"theoretical_loss": 3.723759293482815,
"tokens_seen": 811794432
},
{
"epoch": 0.25,
"learning_rate": 0.0003808224674022066,
"loss": 2.9286,
"theoretical_loss": 3.7237295734517133,
"tokens_seen": 811859968
},
{
"epoch": 0.25,
"learning_rate": 0.0003808124373119358,
"loss": 2.8474,
"theoretical_loss": 3.7236998564912986,
"tokens_seen": 811925504
},
{
"epoch": 0.25,
"learning_rate": 0.000380802407221665,
"loss": 2.797,
"theoretical_loss": 3.723670142601007,
"tokens_seen": 811991040
},
{
"epoch": 0.25,
"learning_rate": 0.00038079237713139417,
"loss": 2.8285,
"theoretical_loss": 3.7236404317802716,
"tokens_seen": 812056576
},
{
"epoch": 0.25,
"learning_rate": 0.00038078234704112335,
"loss": 2.6872,
"theoretical_loss": 3.7236107240285294,
"tokens_seen": 812122112
},
{
"epoch": 0.25,
"learning_rate": 0.0003807723169508526,
"loss": 2.6742,
"theoretical_loss": 3.723581019345215,
"tokens_seen": 812187648
},
{
"epoch": 0.25,
"learning_rate": 0.0003807622868605817,
"loss": 2.9135,
"theoretical_loss": 3.723551317729764,
"tokens_seen": 812253184
},
{
"epoch": 0.25,
"learning_rate": 0.00038075225677031095,
"loss": 2.9234,
"theoretical_loss": 3.723521619181612,
"tokens_seen": 812318720
},
{
"epoch": 0.25,
"learning_rate": 0.00038074222668004013,
"loss": 2.7277,
"theoretical_loss": 3.7234919237001947,
"tokens_seen": 812384256
},
{
"epoch": 0.25,
"learning_rate": 0.0003807321965897693,
"loss": 2.9399,
"theoretical_loss": 3.723462231284949,
"tokens_seen": 812449792
},
{
"epoch": 0.25,
"learning_rate": 0.0003807221664994985,
"loss": 2.8377,
"theoretical_loss": 3.72343254193531,
"tokens_seen": 812515328
},
{
"epoch": 0.25,
"learning_rate": 0.0003807121364092277,
"loss": 2.76,
"theoretical_loss": 3.7234028556507135,
"tokens_seen": 812580864
},
{
"epoch": 0.25,
"objective/train/docs_used": 468484,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.763483762741089,
"objective/train/theoretical_loss": 3.7233731724305974,
"objective/train/tokens_used": 833106400,
"theoretical_loss": 3.7233731724305974,
"tokens_seen": 812646400
},
{
"epoch": 0.25,
"learning_rate": 0.00038070210631895685,
"loss": 2.9626,
"theoretical_loss": 3.7233731724305974,
"tokens_seen": 812646400
},
{
"epoch": 0.25,
"learning_rate": 0.0003806920762286861,
"loss": 2.6844,
"theoretical_loss": 3.723343492274397,
"tokens_seen": 812711936
},
{
"epoch": 0.25,
"learning_rate": 0.0003806820461384152,
"loss": 2.8921,
"theoretical_loss": 3.72331381518155,
"tokens_seen": 812777472
},
{
"epoch": 0.25,
"learning_rate": 0.00038067201604814445,
"loss": 2.8727,
"theoretical_loss": 3.7232841411514923,
"tokens_seen": 812843008
},
{
"epoch": 0.25,
"learning_rate": 0.0003806619859578736,
"loss": 2.8058,
"theoretical_loss": 3.7232544701836616,
"tokens_seen": 812908544
},
{
"epoch": 0.25,
"learning_rate": 0.0003806519558676028,
"loss": 2.8807,
"theoretical_loss": 3.7232248022774943,
"tokens_seen": 812974080
},
{
"epoch": 0.25,
"learning_rate": 0.000380641925777332,
"loss": 2.9078,
"theoretical_loss": 3.723195137432429,
"tokens_seen": 813039616
},
{
"epoch": 0.25,
"learning_rate": 0.0003806318956870612,
"loss": 2.7565,
"theoretical_loss": 3.7231654756479022,
"tokens_seen": 813105152
},
{
"epoch": 0.25,
"learning_rate": 0.0003806218655967904,
"loss": 2.7272,
"theoretical_loss": 3.723135816923352,
"tokens_seen": 813170688
},
{
"epoch": 0.25,
"learning_rate": 0.00038061183550651954,
"loss": 2.773,
"theoretical_loss": 3.7231061612582153,
"tokens_seen": 813236224
},
{
"epoch": 0.25,
"learning_rate": 0.0003806018054162488,
"loss": 2.7753,
"theoretical_loss": 3.7230765086519306,
"tokens_seen": 813301760
},
{
"epoch": 0.25,
"learning_rate": 0.00038059177532597796,
"loss": 2.6932,
"theoretical_loss": 3.723046859103936,
"tokens_seen": 813367296
},
{
"epoch": 0.25,
"learning_rate": 0.00038058174523570714,
"loss": 3.0367,
"theoretical_loss": 3.7230172126136694,
"tokens_seen": 813432832
},
{
"epoch": 0.25,
"learning_rate": 0.0003805717151454363,
"loss": 2.7694,
"theoretical_loss": 3.7229875691805696,
"tokens_seen": 813498368
},
{
"epoch": 0.25,
"learning_rate": 0.00038056168505516555,
"loss": 3.0093,
"theoretical_loss": 3.7229579288040746,
"tokens_seen": 813563904
},
{
"epoch": 0.25,
"learning_rate": 0.0003805516549648947,
"loss": 2.8105,
"theoretical_loss": 3.7229282914836235,
"tokens_seen": 813629440
},
{
"epoch": 0.25,
"learning_rate": 0.0003805416248746239,
"loss": 2.7174,
"theoretical_loss": 3.7228986572186553,
"tokens_seen": 813694976
},
{
"epoch": 0.25,
"learning_rate": 0.00038053159478435304,
"loss": 2.8305,
"theoretical_loss": 3.722869026008608,
"tokens_seen": 813760512
},
{
"epoch": 0.25,
"learning_rate": 0.0003805215646940823,
"loss": 2.924,
"theoretical_loss": 3.722839397852921,
"tokens_seen": 813826048
},
{
"epoch": 0.25,
"learning_rate": 0.00038051153460381146,
"loss": 2.9905,
"theoretical_loss": 3.7228097727510336,
"tokens_seen": 813891584
},
{
"epoch": 0.25,
"learning_rate": 0.00038050150451354064,
"loss": 2.8801,
"theoretical_loss": 3.722780150702386,
"tokens_seen": 813957120
},
{
"epoch": 0.25,
"learning_rate": 0.0003804914744232698,
"loss": 2.8307,
"theoretical_loss": 3.7227505317064162,
"tokens_seen": 814022656
},
{
"epoch": 0.25,
"learning_rate": 0.000380481444332999,
"loss": 2.8192,
"theoretical_loss": 3.722720915762565,
"tokens_seen": 814088192
},
{
"epoch": 0.25,
"learning_rate": 0.0003804714142427282,
"loss": 2.783,
"theoretical_loss": 3.722691302870272,
"tokens_seen": 814153728
},
{
"epoch": 0.25,
"learning_rate": 0.0003804613841524574,
"loss": 2.6482,
"theoretical_loss": 3.7226616930289773,
"tokens_seen": 814219264
},
{
"epoch": 0.25,
"objective/train/docs_used": 469177,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4081919193267822,
"objective/train/theoretical_loss": 3.7226320862381206,
"objective/train/tokens_used": 834744800,
"theoretical_loss": 3.7226320862381206,
"tokens_seen": 814284800
},
{
"epoch": 0.25,
"learning_rate": 0.00038045135406218655,
"loss": 2.8145,
"theoretical_loss": 3.7226320862381206,
"tokens_seen": 814284800
},
{
"epoch": 0.25,
"learning_rate": 0.0003804413239719158,
"loss": 3.0085,
"theoretical_loss": 3.722602482497143,
"tokens_seen": 814350336
},
{
"epoch": 0.25,
"learning_rate": 0.00038043129388164496,
"loss": 2.8507,
"theoretical_loss": 3.7225728818054833,
"tokens_seen": 814415872
},
{
"epoch": 0.25,
"learning_rate": 0.00038042126379137414,
"loss": 2.6513,
"theoretical_loss": 3.7225432841625836,
"tokens_seen": 814481408
},
{
"epoch": 0.25,
"learning_rate": 0.0003804112337011033,
"loss": 2.7457,
"theoretical_loss": 3.722513689567884,
"tokens_seen": 814546944
},
{
"epoch": 0.25,
"learning_rate": 0.0003804012036108325,
"loss": 2.7664,
"theoretical_loss": 3.722484098020826,
"tokens_seen": 814612480
},
{
"epoch": 0.25,
"learning_rate": 0.0003803911735205617,
"loss": 2.9305,
"theoretical_loss": 3.7224545095208494,
"tokens_seen": 814678016
},
{
"epoch": 0.25,
"learning_rate": 0.0003803811434302909,
"loss": 2.6032,
"theoretical_loss": 3.722424924067396,
"tokens_seen": 814743552
},
{
"epoch": 0.25,
"learning_rate": 0.00038037111334002005,
"loss": 2.6402,
"theoretical_loss": 3.722395341659907,
"tokens_seen": 814809088
},
{
"epoch": 0.25,
"learning_rate": 0.0003803610832497493,
"loss": 2.7893,
"theoretical_loss": 3.722365762297825,
"tokens_seen": 814874624
},
{
"epoch": 0.25,
"learning_rate": 0.0003803510531594784,
"loss": 2.9118,
"theoretical_loss": 3.72233618598059,
"tokens_seen": 814940160
},
{
"epoch": 0.25,
"learning_rate": 0.00038034102306920765,
"loss": 2.8133,
"theoretical_loss": 3.7223066127076443,
"tokens_seen": 815005696
},
{
"epoch": 0.25,
"learning_rate": 0.00038033099297893683,
"loss": 2.6754,
"theoretical_loss": 3.7222770424784297,
"tokens_seen": 815071232
},
{
"epoch": 0.25,
"learning_rate": 0.000380320962888666,
"loss": 2.8255,
"theoretical_loss": 3.722247475292388,
"tokens_seen": 815136768
},
{
"epoch": 0.25,
"learning_rate": 0.0003803109327983952,
"loss": 2.9273,
"theoretical_loss": 3.7222179111489626,
"tokens_seen": 815202304
},
{
"epoch": 0.25,
"learning_rate": 0.00038030090270812437,
"loss": 2.8611,
"theoretical_loss": 3.7221883500475945,
"tokens_seen": 815267840
},
{
"epoch": 0.25,
"learning_rate": 0.00038029087261785355,
"loss": 2.8973,
"theoretical_loss": 3.7221587919877264,
"tokens_seen": 815333376
},
{
"epoch": 0.25,
"learning_rate": 0.0003802808425275828,
"loss": 2.6638,
"theoretical_loss": 3.7221292369688017,
"tokens_seen": 815398912
},
{
"epoch": 0.25,
"learning_rate": 0.0003802708124373119,
"loss": 2.9989,
"theoretical_loss": 3.722099684990263,
"tokens_seen": 815464448
},
{
"epoch": 0.25,
"learning_rate": 0.00038026078234704115,
"loss": 2.6776,
"theoretical_loss": 3.722070136051552,
"tokens_seen": 815529984
},
{
"epoch": 0.25,
"learning_rate": 0.00038025075225677033,
"loss": 2.8841,
"theoretical_loss": 3.7220405901521127,
"tokens_seen": 815595520
},
{
"epoch": 0.25,
"learning_rate": 0.0003802407221664995,
"loss": 2.8372,
"theoretical_loss": 3.7220110472913888,
"tokens_seen": 815661056
},
{
"epoch": 0.25,
"learning_rate": 0.0003802306920762287,
"loss": 2.7543,
"theoretical_loss": 3.721981507468823,
"tokens_seen": 815726592
},
{
"epoch": 0.25,
"learning_rate": 0.0003802206619859579,
"loss": 2.8584,
"theoretical_loss": 3.7219519706838584,
"tokens_seen": 815792128
},
{
"epoch": 0.25,
"learning_rate": 0.00038021063189568705,
"loss": 2.7657,
"theoretical_loss": 3.72192243693594,
"tokens_seen": 815857664
},
{
"epoch": 0.25,
"objective/train/docs_used": 469961,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.183133840560913,
"objective/train/theoretical_loss": 3.7218929062245105,
"objective/train/tokens_used": 836383200,
"theoretical_loss": 3.7218929062245105,
"tokens_seen": 815923200
},
{
"epoch": 0.25,
"learning_rate": 0.0003802006018054163,
"loss": 2.9464,
"theoretical_loss": 3.7218929062245105,
"tokens_seen": 815923200
},
{
"epoch": 0.25,
"learning_rate": 0.0003801905717151454,
"loss": 2.6844,
"theoretical_loss": 3.7218633785490143,
"tokens_seen": 815988736
},
{
"epoch": 0.25,
"learning_rate": 0.00038018054162487465,
"loss": 2.8558,
"theoretical_loss": 3.721833853908895,
"tokens_seen": 816054272
},
{
"epoch": 0.25,
"learning_rate": 0.0003801705115346038,
"loss": 2.9005,
"theoretical_loss": 3.721804332303597,
"tokens_seen": 816119808
},
{
"epoch": 0.25,
"learning_rate": 0.000380160481444333,
"loss": 2.7787,
"theoretical_loss": 3.7217748137325652,
"tokens_seen": 816185344
},
{
"epoch": 0.25,
"learning_rate": 0.0003801504513540622,
"loss": 2.8542,
"theoretical_loss": 3.721745298195244,
"tokens_seen": 816250880
},
{
"epoch": 0.25,
"learning_rate": 0.0003801404212637914,
"loss": 2.7552,
"theoretical_loss": 3.7217157856910776,
"tokens_seen": 816316416
},
{
"epoch": 0.25,
"learning_rate": 0.00038013039117352056,
"loss": 2.7411,
"theoretical_loss": 3.7216862762195113,
"tokens_seen": 816381952
},
{
"epoch": 0.25,
"learning_rate": 0.00038012036108324974,
"loss": 2.9057,
"theoretical_loss": 3.72165676977999,
"tokens_seen": 816447488
},
{
"epoch": 0.25,
"learning_rate": 0.0003801103309929789,
"loss": 2.9397,
"theoretical_loss": 3.721627266371958,
"tokens_seen": 816513024
},
{
"epoch": 0.25,
"learning_rate": 0.00038010030090270816,
"loss": 2.6087,
"theoretical_loss": 3.721597765994862,
"tokens_seen": 816578560
},
{
"epoch": 0.25,
"learning_rate": 0.0003800902708124373,
"loss": 2.9571,
"theoretical_loss": 3.721568268648146,
"tokens_seen": 816644096
},
{
"epoch": 0.25,
"learning_rate": 0.0003800802407221665,
"loss": 2.5315,
"theoretical_loss": 3.7215387743312567,
"tokens_seen": 816709632
},
{
"epoch": 0.25,
"learning_rate": 0.0003800702106318957,
"loss": 2.8646,
"theoretical_loss": 3.7215092830436394,
"tokens_seen": 816775168
},
{
"epoch": 0.25,
"learning_rate": 0.0003800601805416249,
"loss": 2.9061,
"theoretical_loss": 3.72147979478474,
"tokens_seen": 816840704
},
{
"epoch": 0.25,
"learning_rate": 0.00038005015045135406,
"loss": 2.8172,
"theoretical_loss": 3.7214503095540037,
"tokens_seen": 816906240
},
{
"epoch": 0.25,
"learning_rate": 0.00038004012036108324,
"loss": 2.9905,
"theoretical_loss": 3.721420827350878,
"tokens_seen": 816971776
},
{
"epoch": 0.25,
"learning_rate": 0.0003800300902708124,
"loss": 2.7211,
"theoretical_loss": 3.7213913481748087,
"tokens_seen": 817037312
},
{
"epoch": 0.25,
"learning_rate": 0.00038002006018054166,
"loss": 2.772,
"theoretical_loss": 3.7213618720252413,
"tokens_seen": 817102848
},
{
"epoch": 0.25,
"learning_rate": 0.0003800100300902708,
"loss": 3.0289,
"theoretical_loss": 3.7213323989016236,
"tokens_seen": 817168384
},
{
"epoch": 0.25,
"learning_rate": 0.00038,
"loss": 2.8633,
"theoretical_loss": 3.721302928803402,
"tokens_seen": 817233920
},
{
"epoch": 0.25,
"learning_rate": 0.00037998996990972915,
"loss": 2.5601,
"theoretical_loss": 3.7212734617300223,
"tokens_seen": 817299456
},
{
"epoch": 0.25,
"learning_rate": 0.0003799799398194584,
"loss": 2.875,
"theoretical_loss": 3.7212439976809337,
"tokens_seen": 817364992
},
{
"epoch": 0.25,
"learning_rate": 0.00037996990972918756,
"loss": 2.7411,
"theoretical_loss": 3.721214536655581,
"tokens_seen": 817430528
},
{
"epoch": 0.25,
"learning_rate": 0.00037995987963891675,
"loss": 2.8575,
"theoretical_loss": 3.721185078653413,
"tokens_seen": 817496064
},
{
"epoch": 0.25,
"objective/train/docs_used": 471203,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5141611099243164,
"objective/train/theoretical_loss": 3.7211556236738765,
"objective/train/tokens_used": 838021600,
"theoretical_loss": 3.7211556236738765,
"tokens_seen": 817561600
},
{
"epoch": 0.25,
"learning_rate": 0.0003799498495486459,
"loss": 2.5787,
"theoretical_loss": 3.7211556236738765,
"tokens_seen": 817561600
},
{
"epoch": 0.25,
"learning_rate": 0.00037993981945837516,
"loss": 2.8159,
"theoretical_loss": 3.7211261717164197,
"tokens_seen": 817627136
},
{
"epoch": 0.25,
"learning_rate": 0.0003799297893681043,
"loss": 2.8371,
"theoretical_loss": 3.72109672278049,
"tokens_seen": 817692672
},
{
"epoch": 0.25,
"learning_rate": 0.0003799197592778335,
"loss": 2.9,
"theoretical_loss": 3.7210672768655355,
"tokens_seen": 817758208
},
{
"epoch": 0.25,
"learning_rate": 0.00037990972918756265,
"loss": 2.8366,
"theoretical_loss": 3.7210378339710033,
"tokens_seen": 817823744
},
{
"epoch": 0.25,
"learning_rate": 0.0003798996990972919,
"loss": 2.9229,
"theoretical_loss": 3.7210083940963425,
"tokens_seen": 817889280
},
{
"epoch": 0.25,
"learning_rate": 0.00037988966900702107,
"loss": 2.6867,
"theoretical_loss": 3.7209789572410013,
"tokens_seen": 817954816
},
{
"epoch": 0.25,
"learning_rate": 0.00037987963891675025,
"loss": 2.7363,
"theoretical_loss": 3.7209495234044283,
"tokens_seen": 818020352
},
{
"epoch": 0.25,
"learning_rate": 0.0003798696088264795,
"loss": 2.5668,
"theoretical_loss": 3.7209200925860717,
"tokens_seen": 818085888
},
{
"epoch": 0.25,
"learning_rate": 0.0003798595787362086,
"loss": 2.6929,
"theoretical_loss": 3.7208906647853803,
"tokens_seen": 818151424
},
{
"epoch": 0.25,
"learning_rate": 0.00037984954864593785,
"loss": 2.9396,
"theoretical_loss": 3.720861240001803,
"tokens_seen": 818216960
},
{
"epoch": 0.25,
"learning_rate": 0.00037983951855566703,
"loss": 2.7418,
"theoretical_loss": 3.720831818234789,
"tokens_seen": 818282496
},
{
"epoch": 0.25,
"learning_rate": 0.0003798294884653962,
"loss": 2.6825,
"theoretical_loss": 3.7208023994837878,
"tokens_seen": 818348032
},
{
"epoch": 0.25,
"learning_rate": 0.0003798194583751254,
"loss": 2.8422,
"theoretical_loss": 3.7207729837482484,
"tokens_seen": 818413568
},
{
"epoch": 0.25,
"learning_rate": 0.00037980942828485457,
"loss": 2.8084,
"theoretical_loss": 3.7207435710276195,
"tokens_seen": 818479104
},
{
"epoch": 0.25,
"learning_rate": 0.00037979939819458375,
"loss": 2.9062,
"theoretical_loss": 3.720714161321352,
"tokens_seen": 818544640
},
{
"epoch": 0.25,
"learning_rate": 0.000379789368104313,
"loss": 2.7317,
"theoretical_loss": 3.7206847546288957,
"tokens_seen": 818610176
},
{
"epoch": 0.25,
"learning_rate": 0.0003797793380140421,
"loss": 2.8819,
"theoretical_loss": 3.7206553509496993,
"tokens_seen": 818675712
},
{
"epoch": 0.25,
"learning_rate": 0.00037976930792377135,
"loss": 3.007,
"theoretical_loss": 3.720625950283214,
"tokens_seen": 818741248
},
{
"epoch": 0.25,
"learning_rate": 0.00037975927783350053,
"loss": 2.9027,
"theoretical_loss": 3.720596552628889,
"tokens_seen": 818806784
},
{
"epoch": 0.25,
"learning_rate": 0.0003797492477432297,
"loss": 2.816,
"theoretical_loss": 3.720567157986175,
"tokens_seen": 818872320
},
{
"epoch": 0.25,
"learning_rate": 0.0003797392176529589,
"loss": 2.8617,
"theoretical_loss": 3.7205377663545236,
"tokens_seen": 818937856
},
{
"epoch": 0.25,
"learning_rate": 0.0003797291875626881,
"loss": 2.7669,
"theoretical_loss": 3.720508377733384,
"tokens_seen": 819003392
},
{
"epoch": 0.25,
"learning_rate": 0.00037971915747241725,
"loss": 2.7101,
"theoretical_loss": 3.7204789921222075,
"tokens_seen": 819068928
},
{
"epoch": 0.25,
"learning_rate": 0.0003797091273821465,
"loss": 2.9332,
"theoretical_loss": 3.720449609520445,
"tokens_seen": 819134464
},
{
"debugging/Self-BLEU-5": 0.49160234205869136,
"debugging/distinct-1-grams": 0.8070841648877926,
"debugging/distinct-2-grams": 0.9681388503540196,
"debugging/entropy-1-grams": 6.02777455899678,
"debugging/entropy-2-grams": 6.929977811593813,
"debugging/length": 501.2142857142857,
"debugging/num_segments": 14,
"epoch": 0.25,
"objective/train/docs_used": 471797,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7884488105773926,
"objective/train/theoretical_loss": 3.7204202299275475,
"objective/train/tokens_used": 839660000,
"theoretical_loss": 3.7204202299275475,
"tokens_seen": 819200000
},
{
"epoch": 0.25,
"learning_rate": 0.0003796990972918756,
"loss": 2.9633,
"theoretical_loss": 3.7204202299275475,
"tokens_seen": 819200000
},
{
"epoch": 0.25,
"learning_rate": 0.00037968906720160485,
"loss": 2.5588,
"theoretical_loss": 3.7203908533429666,
"tokens_seen": 819265536
},
{
"epoch": 0.25,
"learning_rate": 0.000379679037111334,
"loss": 2.8539,
"theoretical_loss": 3.7203614797661535,
"tokens_seen": 819331072
},
{
"epoch": 0.25,
"learning_rate": 0.0003796690070210632,
"loss": 2.6465,
"theoretical_loss": 3.720332109196559,
"tokens_seen": 819396608
},
{
"epoch": 0.25,
"learning_rate": 0.0003796589769307924,
"loss": 2.8236,
"theoretical_loss": 3.7203027416336356,
"tokens_seen": 819462144
},
{
"epoch": 0.25,
"learning_rate": 0.0003796489468405216,
"loss": 3.0052,
"theoretical_loss": 3.7202733770768344,
"tokens_seen": 819527680
},
{
"epoch": 0.25,
"learning_rate": 0.00037963891675025076,
"loss": 2.8943,
"theoretical_loss": 3.7202440155256076,
"tokens_seen": 819593216
},
{
"epoch": 0.25,
"learning_rate": 0.00037962888665997994,
"loss": 2.5976,
"theoretical_loss": 3.720214656979408,
"tokens_seen": 819658752
},
{
"epoch": 0.25,
"learning_rate": 0.0003796188565697091,
"loss": 2.8677,
"theoretical_loss": 3.720185301437687,
"tokens_seen": 819724288
},
{
"epoch": 0.25,
"learning_rate": 0.00037960882647943836,
"loss": 2.8317,
"theoretical_loss": 3.720155948899897,
"tokens_seen": 819789824
},
{
"epoch": 0.25,
"learning_rate": 0.0003795987963891675,
"loss": 2.8588,
"theoretical_loss": 3.720126599365491,
"tokens_seen": 819855360
},
{
"epoch": 0.25,
"learning_rate": 0.0003795887662988967,
"loss": 2.7674,
"theoretical_loss": 3.720097252833921,
"tokens_seen": 819920896
},
{
"epoch": 0.25,
"learning_rate": 0.0003795787362086259,
"loss": 2.8217,
"theoretical_loss": 3.72006790930464,
"tokens_seen": 819986432
},
{
"epoch": 0.25,
"learning_rate": 0.0003795687061183551,
"loss": 2.8634,
"theoretical_loss": 3.720038568777101,
"tokens_seen": 820051968
},
{
"epoch": 0.25,
"learning_rate": 0.00037955867602808426,
"loss": 2.9685,
"theoretical_loss": 3.7200092312507573,
"tokens_seen": 820117504
},
{
"epoch": 0.25,
"learning_rate": 0.00037954864593781344,
"loss": 2.7761,
"theoretical_loss": 3.719979896725062,
"tokens_seen": 820183040
},
{
"epoch": 0.25,
"learning_rate": 0.0003795386158475426,
"loss": 2.5492,
"theoretical_loss": 3.7199505651994684,
"tokens_seen": 820248576
},
{
"epoch": 0.25,
"learning_rate": 0.00037952858575727186,
"loss": 2.5691,
"theoretical_loss": 3.71992123667343,
"tokens_seen": 820314112
},
{
"epoch": 0.25,
"learning_rate": 0.000379518555667001,
"loss": 2.6376,
"theoretical_loss": 3.7198919111464006,
"tokens_seen": 820379648
},
{
"epoch": 0.25,
"learning_rate": 0.0003795085255767302,
"loss": 2.7935,
"theoretical_loss": 3.7198625886178336,
"tokens_seen": 820445184
},
{
"epoch": 0.25,
"learning_rate": 0.00037949849548645935,
"loss": 2.6882,
"theoretical_loss": 3.719833269087183,
"tokens_seen": 820510720
},
{
"epoch": 0.25,
"learning_rate": 0.0003794884653961886,
"loss": 2.438,
"theoretical_loss": 3.719803952553904,
"tokens_seen": 820576256
},
{
"epoch": 0.25,
"learning_rate": 0.00037947843530591776,
"loss": 2.7553,
"theoretical_loss": 3.719774639017449,
"tokens_seen": 820641792
},
{
"epoch": 0.25,
"learning_rate": 0.00037946840521564695,
"loss": 2.6967,
"theoretical_loss": 3.7197453284772735,
"tokens_seen": 820707328
},
{
"epoch": 0.25,
"learning_rate": 0.0003794583751253761,
"loss": 2.679,
"theoretical_loss": 3.719716020932832,
"tokens_seen": 820772864
},
{
"epoch": 0.25,
"objective/train/docs_used": 473281,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.718562602996826,
"objective/train/theoretical_loss": 3.719686716383579,
"objective/train/tokens_used": 841298400,
"theoretical_loss": 3.719686716383579,
"tokens_seen": 820838400
},
{
"epoch": 0.25,
"learning_rate": 0.00037944834503510536,
"loss": 2.5748,
"theoretical_loss": 3.719686716383579,
"tokens_seen": 820838400
},
{
"epoch": 0.25,
"learning_rate": 0.0003794383149448345,
"loss": 2.7844,
"theoretical_loss": 3.719657414828969,
"tokens_seen": 820903936
},
{
"epoch": 0.25,
"learning_rate": 0.0003794282848545637,
"loss": 2.8819,
"theoretical_loss": 3.7196281162684572,
"tokens_seen": 820969472
},
{
"epoch": 0.25,
"learning_rate": 0.00037941825476429285,
"loss": 2.7177,
"theoretical_loss": 3.7195988207014983,
"tokens_seen": 821035008
},
{
"epoch": 0.25,
"learning_rate": 0.0003794082246740221,
"loss": 2.6298,
"theoretical_loss": 3.7195695281275487,
"tokens_seen": 821100544
},
{
"epoch": 0.25,
"learning_rate": 0.00037939819458375127,
"loss": 2.6639,
"theoretical_loss": 3.7195402385460623,
"tokens_seen": 821166080
},
{
"epoch": 0.25,
"learning_rate": 0.00037938816449348045,
"loss": 2.5637,
"theoretical_loss": 3.7195109519564955,
"tokens_seen": 821231616
},
{
"epoch": 0.25,
"learning_rate": 0.00037937813440320963,
"loss": 2.8666,
"theoretical_loss": 3.719481668358304,
"tokens_seen": 821297152
},
{
"epoch": 0.25,
"learning_rate": 0.0003793681043129388,
"loss": 2.7569,
"theoretical_loss": 3.7194523877509424,
"tokens_seen": 821362688
},
{
"epoch": 0.25,
"learning_rate": 0.000379358074222668,
"loss": 2.7826,
"theoretical_loss": 3.719423110133868,
"tokens_seen": 821428224
},
{
"epoch": 0.25,
"learning_rate": 0.00037934804413239723,
"loss": 2.8492,
"theoretical_loss": 3.719393835506537,
"tokens_seen": 821493760
},
{
"epoch": 0.25,
"learning_rate": 0.00037933801404212635,
"loss": 2.6386,
"theoretical_loss": 3.719364563868404,
"tokens_seen": 821559296
},
{
"epoch": 0.25,
"learning_rate": 0.0003793279839518556,
"loss": 2.8469,
"theoretical_loss": 3.719335295218927,
"tokens_seen": 821624832
},
{
"epoch": 0.25,
"learning_rate": 0.0003793179538615847,
"loss": 2.8802,
"theoretical_loss": 3.719306029557562,
"tokens_seen": 821690368
},
{
"epoch": 0.25,
"learning_rate": 0.00037930792377131395,
"loss": 2.7726,
"theoretical_loss": 3.719276766883765,
"tokens_seen": 821755904
},
{
"epoch": 0.25,
"learning_rate": 0.00037929789368104313,
"loss": 2.7707,
"theoretical_loss": 3.719247507196994,
"tokens_seen": 821821440
},
{
"epoch": 0.25,
"learning_rate": 0.0003792878635907723,
"loss": 2.774,
"theoretical_loss": 3.719218250496705,
"tokens_seen": 821886976
},
{
"epoch": 0.25,
"learning_rate": 0.0003792778335005015,
"loss": 2.6994,
"theoretical_loss": 3.7191889967823557,
"tokens_seen": 821952512
},
{
"epoch": 0.25,
"learning_rate": 0.00037926780341023073,
"loss": 2.8041,
"theoretical_loss": 3.7191597460534025,
"tokens_seen": 822018048
},
{
"epoch": 0.25,
"learning_rate": 0.00037925777331995986,
"loss": 2.8086,
"theoretical_loss": 3.7191304983093034,
"tokens_seen": 822083584
},
{
"epoch": 0.25,
"learning_rate": 0.0003792477432296891,
"loss": 2.7191,
"theoretical_loss": 3.719101253549516,
"tokens_seen": 822149120
},
{
"epoch": 0.25,
"learning_rate": 0.0003792377131394182,
"loss": 2.8849,
"theoretical_loss": 3.719072011773497,
"tokens_seen": 822214656
},
{
"epoch": 0.25,
"learning_rate": 0.00037922768304914746,
"loss": 2.8194,
"theoretical_loss": 3.719042772980705,
"tokens_seen": 822280192
},
{
"epoch": 0.25,
"learning_rate": 0.00037921765295887664,
"loss": 2.8368,
"theoretical_loss": 3.719013537170598,
"tokens_seen": 822345728
},
{
"epoch": 0.25,
"learning_rate": 0.0003792076228686058,
"loss": 2.9321,
"theoretical_loss": 3.718984304342634,
"tokens_seen": 822411264
},
{
"epoch": 0.25,
"objective/train/docs_used": 473825,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7618255615234375,
"objective/train/theoretical_loss": 3.7189550744962707,
"objective/train/tokens_used": 842936800,
"theoretical_loss": 3.7189550744962707,
"tokens_seen": 822476800
},
{
"epoch": 0.25,
"learning_rate": 0.000379197592778335,
"loss": 2.6672,
"theoretical_loss": 3.7189550744962707,
"tokens_seen": 822476800
},
{
"epoch": 0.25,
"learning_rate": 0.0003791875626880642,
"loss": 2.7987,
"theoretical_loss": 3.718925847630967,
"tokens_seen": 822542336
},
{
"epoch": 0.25,
"learning_rate": 0.00037917753259779336,
"loss": 2.8212,
"theoretical_loss": 3.7188966237461814,
"tokens_seen": 822607872
},
{
"epoch": 0.25,
"learning_rate": 0.0003791675025075226,
"loss": 2.8259,
"theoretical_loss": 3.718867402841372,
"tokens_seen": 822673408
},
{
"epoch": 0.25,
"learning_rate": 0.0003791574724172517,
"loss": 2.6386,
"theoretical_loss": 3.718838184915998,
"tokens_seen": 822738944
},
{
"epoch": 0.25,
"learning_rate": 0.00037914744232698096,
"loss": 3.0267,
"theoretical_loss": 3.718808969969518,
"tokens_seen": 822804480
},
{
"epoch": 0.25,
"learning_rate": 0.0003791374122367101,
"loss": 2.5974,
"theoretical_loss": 3.718779758001392,
"tokens_seen": 822870016
},
{
"epoch": 0.25,
"learning_rate": 0.0003791273821464393,
"loss": 2.647,
"theoretical_loss": 3.7187505490110775,
"tokens_seen": 822935552
},
{
"epoch": 0.25,
"learning_rate": 0.00037911735205616856,
"loss": 2.7119,
"theoretical_loss": 3.7187213429980357,
"tokens_seen": 823001088
},
{
"epoch": 0.25,
"learning_rate": 0.0003791073219658977,
"loss": 2.7735,
"theoretical_loss": 3.7186921399617248,
"tokens_seen": 823066624
},
{
"epoch": 0.25,
"learning_rate": 0.0003790972918756269,
"loss": 2.7473,
"theoretical_loss": 3.7186629399016047,
"tokens_seen": 823132160
},
{
"epoch": 0.25,
"learning_rate": 0.0003790872617853561,
"loss": 3.0663,
"theoretical_loss": 3.7186337428171354,
"tokens_seen": 823197696
},
{
"epoch": 0.25,
"learning_rate": 0.0003790772316950853,
"loss": 2.7563,
"theoretical_loss": 3.7186045487077766,
"tokens_seen": 823263232
},
{
"epoch": 0.25,
"learning_rate": 0.00037906720160481446,
"loss": 2.7419,
"theoretical_loss": 3.7185753575729885,
"tokens_seen": 823328768
},
{
"epoch": 0.25,
"learning_rate": 0.00037905717151454364,
"loss": 2.9488,
"theoretical_loss": 3.7185461694122313,
"tokens_seen": 823394304
},
{
"epoch": 0.25,
"learning_rate": 0.0003790471414242728,
"loss": 2.871,
"theoretical_loss": 3.7185169842249652,
"tokens_seen": 823459840
},
{
"epoch": 0.25,
"learning_rate": 0.00037903711133400206,
"loss": 2.9046,
"theoretical_loss": 3.7184878020106504,
"tokens_seen": 823525376
},
{
"epoch": 0.25,
"learning_rate": 0.0003790270812437312,
"loss": 2.7884,
"theoretical_loss": 3.718458622768748,
"tokens_seen": 823590912
},
{
"epoch": 0.25,
"learning_rate": 0.0003790170511534604,
"loss": 2.8426,
"theoretical_loss": 3.718429446498719,
"tokens_seen": 823656448
},
{
"epoch": 0.25,
"learning_rate": 0.00037900702106318955,
"loss": 2.9179,
"theoretical_loss": 3.7184002732000234,
"tokens_seen": 823721984
},
{
"epoch": 0.25,
"learning_rate": 0.0003789969909729188,
"loss": 2.8312,
"theoretical_loss": 3.7183711028721227,
"tokens_seen": 823787520
},
{
"epoch": 0.25,
"learning_rate": 0.00037898696088264796,
"loss": 2.7328,
"theoretical_loss": 3.7183419355144784,
"tokens_seen": 823853056
},
{
"epoch": 0.25,
"learning_rate": 0.00037897693079237715,
"loss": 2.6463,
"theoretical_loss": 3.718312771126551,
"tokens_seen": 823918592
},
{
"epoch": 0.25,
"learning_rate": 0.0003789669007021063,
"loss": 2.6934,
"theoretical_loss": 3.718283609707803,
"tokens_seen": 823984128
},
{
"epoch": 0.25,
"learning_rate": 0.00037895687061183556,
"loss": 2.7448,
"theoretical_loss": 3.718254451257695,
"tokens_seen": 824049664
},
{
"epoch": 0.25,
"objective/train/docs_used": 475209,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.179006338119507,
"objective/train/theoretical_loss": 3.7182252957756896,
"objective/train/tokens_used": 844575200,
"theoretical_loss": 3.7182252957756896,
"tokens_seen": 824115200
},
{
"epoch": 0.25,
"learning_rate": 0.0003789468405215647,
"loss": 2.9396,
"theoretical_loss": 3.7182252957756896,
"tokens_seen": 824115200
},
{
"epoch": 0.25,
"learning_rate": 0.0003789368104312939,
"loss": 2.7847,
"theoretical_loss": 3.718196143261248,
"tokens_seen": 824180736
},
{
"epoch": 0.25,
"learning_rate": 0.00037892678034102305,
"loss": 2.745,
"theoretical_loss": 3.718166993713832,
"tokens_seen": 824246272
},
{
"epoch": 0.25,
"learning_rate": 0.0003789167502507523,
"loss": 2.7957,
"theoretical_loss": 3.7181378471329047,
"tokens_seen": 824311808
},
{
"epoch": 0.25,
"learning_rate": 0.00037890672016048147,
"loss": 2.7108,
"theoretical_loss": 3.7181087035179283,
"tokens_seen": 824377344
},
{
"epoch": 0.25,
"learning_rate": 0.00037889669007021065,
"loss": 2.6135,
"theoretical_loss": 3.718079562868364,
"tokens_seen": 824442880
},
{
"epoch": 0.25,
"learning_rate": 0.00037888665997993983,
"loss": 2.5861,
"theoretical_loss": 3.7180504251836757,
"tokens_seen": 824508416
},
{
"epoch": 0.25,
"learning_rate": 0.000378876629889669,
"loss": 2.8015,
"theoretical_loss": 3.7180212904633256,
"tokens_seen": 824573952
},
{
"epoch": 0.25,
"learning_rate": 0.0003788665997993982,
"loss": 2.827,
"theoretical_loss": 3.7179921587067764,
"tokens_seen": 824639488
},
{
"epoch": 0.25,
"learning_rate": 0.00037885656970912743,
"loss": 2.6574,
"theoretical_loss": 3.7179630299134914,
"tokens_seen": 824705024
},
{
"epoch": 0.25,
"learning_rate": 0.00037884653961885655,
"loss": 2.7249,
"theoretical_loss": 3.7179339040829333,
"tokens_seen": 824770560
},
{
"epoch": 0.25,
"learning_rate": 0.0003788365095285858,
"loss": 2.7057,
"theoretical_loss": 3.7179047812145662,
"tokens_seen": 824836096
},
{
"epoch": 0.25,
"learning_rate": 0.0003788264794383149,
"loss": 2.6668,
"theoretical_loss": 3.7178756613078527,
"tokens_seen": 824901632
},
{
"epoch": 0.25,
"learning_rate": 0.00037881644934804415,
"loss": 2.8663,
"theoretical_loss": 3.717846544362257,
"tokens_seen": 824967168
},
{
"epoch": 0.25,
"learning_rate": 0.00037880641925777333,
"loss": 2.8774,
"theoretical_loss": 3.717817430377242,
"tokens_seen": 825032704
},
{
"epoch": 0.25,
"learning_rate": 0.0003787963891675025,
"loss": 2.7932,
"theoretical_loss": 3.7177883193522723,
"tokens_seen": 825098240
},
{
"epoch": 0.25,
"learning_rate": 0.0003787863590772317,
"loss": 2.8644,
"theoretical_loss": 3.717759211286811,
"tokens_seen": 825163776
},
{
"epoch": 0.25,
"learning_rate": 0.00037877632898696093,
"loss": 2.8038,
"theoretical_loss": 3.7177301061803236,
"tokens_seen": 825229312
},
{
"epoch": 0.25,
"learning_rate": 0.00037876629889669006,
"loss": 2.5448,
"theoretical_loss": 3.717701004032273,
"tokens_seen": 825294848
},
{
"epoch": 0.25,
"learning_rate": 0.0003787562688064193,
"loss": 2.8176,
"theoretical_loss": 3.7176719048421245,
"tokens_seen": 825360384
},
{
"epoch": 0.25,
"learning_rate": 0.0003787462387161484,
"loss": 2.8675,
"theoretical_loss": 3.717642808609342,
"tokens_seen": 825425920
},
{
"epoch": 0.25,
"learning_rate": 0.00037873620862587766,
"loss": 2.7802,
"theoretical_loss": 3.717613715333391,
"tokens_seen": 825491456
},
{
"epoch": 0.25,
"learning_rate": 0.00037872617853560684,
"loss": 2.8902,
"theoretical_loss": 3.7175846250137345,
"tokens_seen": 825556992
},
{
"epoch": 0.25,
"learning_rate": 0.000378716148445336,
"loss": 2.7592,
"theoretical_loss": 3.7175555376498397,
"tokens_seen": 825622528
},
{
"epoch": 0.25,
"learning_rate": 0.0003787061183550652,
"loss": 2.9172,
"theoretical_loss": 3.7175264532411703,
"tokens_seen": 825688064
},
{
"epoch": 0.25,
"objective/train/docs_used": 475848,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.414538860321045,
"objective/train/theoretical_loss": 3.717497371787192,
"objective/train/tokens_used": 846213600,
"theoretical_loss": 3.717497371787192,
"tokens_seen": 825753600
},
{
"epoch": 0.25,
"learning_rate": 0.0003786960882647944,
"loss": 2.5937,
"theoretical_loss": 3.717497371787192,
"tokens_seen": 825753600
},
{
"epoch": 0.25,
"learning_rate": 0.00037868605817452356,
"loss": 2.8341,
"theoretical_loss": 3.7174682932873706,
"tokens_seen": 825819136
},
{
"epoch": 0.25,
"learning_rate": 0.0003786760280842528,
"loss": 2.8919,
"theoretical_loss": 3.7174392177411706,
"tokens_seen": 825884672
},
{
"epoch": 0.25,
"learning_rate": 0.0003786659979939819,
"loss": 2.9754,
"theoretical_loss": 3.717410145148058,
"tokens_seen": 825950208
},
{
"epoch": 0.25,
"learning_rate": 0.00037865596790371116,
"loss": 2.7715,
"theoretical_loss": 3.717381075507499,
"tokens_seen": 826015744
},
{
"epoch": 0.25,
"learning_rate": 0.0003786459378134403,
"loss": 2.9639,
"theoretical_loss": 3.717352008818959,
"tokens_seen": 826081280
},
{
"epoch": 0.25,
"learning_rate": 0.0003786359077231695,
"loss": 2.942,
"theoretical_loss": 3.7173229450819045,
"tokens_seen": 826146816
},
{
"epoch": 0.25,
"learning_rate": 0.0003786258776328987,
"loss": 2.9407,
"theoretical_loss": 3.7172938842958017,
"tokens_seen": 826212352
},
{
"epoch": 0.25,
"learning_rate": 0.0003786158475426279,
"loss": 2.7382,
"theoretical_loss": 3.7172648264601165,
"tokens_seen": 826277888
},
{
"epoch": 0.25,
"learning_rate": 0.00037860581745235706,
"loss": 2.7991,
"theoretical_loss": 3.7172357715743156,
"tokens_seen": 826343424
},
{
"epoch": 0.25,
"learning_rate": 0.0003785957873620863,
"loss": 2.8059,
"theoretical_loss": 3.7172067196378658,
"tokens_seen": 826408960
},
{
"epoch": 0.25,
"learning_rate": 0.0003785857572718154,
"loss": 2.7626,
"theoretical_loss": 3.7171776706502335,
"tokens_seen": 826474496
},
{
"epoch": 0.25,
"learning_rate": 0.00037857572718154466,
"loss": 2.8048,
"theoretical_loss": 3.717148624610886,
"tokens_seen": 826540032
},
{
"epoch": 0.25,
"learning_rate": 0.0003785656970912738,
"loss": 2.8321,
"theoretical_loss": 3.71711958151929,
"tokens_seen": 826605568
},
{
"epoch": 0.25,
"learning_rate": 0.000378555667001003,
"loss": 2.9544,
"theoretical_loss": 3.717090541374913,
"tokens_seen": 826671104
},
{
"epoch": 0.25,
"learning_rate": 0.0003785456369107322,
"loss": 2.8569,
"theoretical_loss": 3.7170615041772224,
"tokens_seen": 826736640
},
{
"epoch": 0.25,
"learning_rate": 0.0003785356068204614,
"loss": 2.8937,
"theoretical_loss": 3.7170324699256847,
"tokens_seen": 826802176
},
{
"epoch": 0.25,
"learning_rate": 0.00037852557673019057,
"loss": 2.6687,
"theoretical_loss": 3.717003438619769,
"tokens_seen": 826867712
},
{
"epoch": 0.25,
"learning_rate": 0.00037851554663991975,
"loss": 2.8527,
"theoretical_loss": 3.7169744102589415,
"tokens_seen": 826933248
},
{
"epoch": 0.25,
"learning_rate": 0.00037850551654964893,
"loss": 3.021,
"theoretical_loss": 3.7169453848426715,
"tokens_seen": 826998784
},
{
"epoch": 0.25,
"learning_rate": 0.00037849548645937816,
"loss": 2.6607,
"theoretical_loss": 3.7169163623704256,
"tokens_seen": 827064320
},
{
"epoch": 0.25,
"learning_rate": 0.0003784854563691073,
"loss": 2.7623,
"theoretical_loss": 3.716887342841673,
"tokens_seen": 827129856
},
{
"epoch": 0.25,
"learning_rate": 0.0003784754262788365,
"loss": 2.9672,
"theoretical_loss": 3.716858326255881,
"tokens_seen": 827195392
},
{
"epoch": 0.25,
"learning_rate": 0.00037846539618856565,
"loss": 2.9033,
"theoretical_loss": 3.7168293126125196,
"tokens_seen": 827260928
},
{
"epoch": 0.25,
"learning_rate": 0.0003784553660982949,
"loss": 2.6257,
"theoretical_loss": 3.7168003019110554,
"tokens_seen": 827326464
},
{
"epoch": 0.25,
"objective/train/docs_used": 477214,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.126943826675415,
"objective/train/theoretical_loss": 3.7167712941509587,
"objective/train/tokens_used": 847852000,
"theoretical_loss": 3.7167712941509587,
"tokens_seen": 827392000
},
{
"epoch": 0.25,
"learning_rate": 0.00037844533600802407,
"loss": 2.929,
"theoretical_loss": 3.7167712941509587,
"tokens_seen": 827392000
},
{
"epoch": 0.25,
"learning_rate": 0.00037843530591775325,
"loss": 3.0456,
"theoretical_loss": 3.7167422893316977,
"tokens_seen": 827457536
},
{
"epoch": 0.25,
"learning_rate": 0.00037842527582748243,
"loss": 2.8873,
"theoretical_loss": 3.716713287452741,
"tokens_seen": 827523072
},
{
"epoch": 0.25,
"learning_rate": 0.00037841524573721167,
"loss": 2.6903,
"theoretical_loss": 3.7166842885135583,
"tokens_seen": 827588608
},
{
"epoch": 0.25,
"learning_rate": 0.0003784052156469408,
"loss": 2.6523,
"theoretical_loss": 3.716655292513619,
"tokens_seen": 827654144
},
{
"epoch": 0.25,
"learning_rate": 0.00037839518555667003,
"loss": 2.8055,
"theoretical_loss": 3.7166262994523915,
"tokens_seen": 827719680
},
{
"epoch": 0.25,
"learning_rate": 0.00037838515546639916,
"loss": 2.7148,
"theoretical_loss": 3.7165973093293463,
"tokens_seen": 827785216
},
{
"epoch": 0.25,
"learning_rate": 0.0003783751253761284,
"loss": 2.7917,
"theoretical_loss": 3.7165683221439525,
"tokens_seen": 827850752
},
{
"epoch": 0.25,
"learning_rate": 0.00037836509528585763,
"loss": 2.8019,
"theoretical_loss": 3.7165393378956804,
"tokens_seen": 827916288
},
{
"epoch": 0.25,
"learning_rate": 0.00037835506519558675,
"loss": 2.8402,
"theoretical_loss": 3.716510356583999,
"tokens_seen": 827981824
},
{
"epoch": 0.25,
"learning_rate": 0.000378345035105316,
"loss": 2.9068,
"theoretical_loss": 3.71648137820838,
"tokens_seen": 828047360
},
{
"epoch": 0.25,
"learning_rate": 0.0003783350050150451,
"loss": 2.5881,
"theoretical_loss": 3.716452402768292,
"tokens_seen": 828112896
},
{
"epoch": 0.25,
"learning_rate": 0.00037832497492477435,
"loss": 2.8217,
"theoretical_loss": 3.716423430263206,
"tokens_seen": 828178432
},
{
"epoch": 0.25,
"learning_rate": 0.00037831494483450353,
"loss": 2.6975,
"theoretical_loss": 3.7163944606925927,
"tokens_seen": 828243968
},
{
"epoch": 0.25,
"learning_rate": 0.0003783049147442327,
"loss": 2.8574,
"theoretical_loss": 3.716365494055922,
"tokens_seen": 828309504
},
{
"epoch": 0.25,
"learning_rate": 0.0003782948846539619,
"loss": 2.7526,
"theoretical_loss": 3.716336530352666,
"tokens_seen": 828375040
},
{
"epoch": 0.25,
"learning_rate": 0.00037828485456369113,
"loss": 2.8619,
"theoretical_loss": 3.716307569582294,
"tokens_seen": 828440576
},
{
"epoch": 0.25,
"learning_rate": 0.00037827482447342026,
"loss": 2.6352,
"theoretical_loss": 3.716278611744278,
"tokens_seen": 828506112
},
{
"epoch": 0.25,
"learning_rate": 0.0003782647943831495,
"loss": 2.5646,
"theoretical_loss": 3.7162496568380887,
"tokens_seen": 828571648
},
{
"epoch": 0.25,
"learning_rate": 0.0003782547642928786,
"loss": 2.8873,
"theoretical_loss": 3.716220704863198,
"tokens_seen": 828637184
},
{
"epoch": 0.25,
"learning_rate": 0.00037824473420260786,
"loss": 2.8708,
"theoretical_loss": 3.716191755819077,
"tokens_seen": 828702720
},
{
"epoch": 0.25,
"learning_rate": 0.00037823470411233704,
"loss": 2.8699,
"theoretical_loss": 3.716162809705197,
"tokens_seen": 828768256
},
{
"epoch": 0.25,
"learning_rate": 0.0003782246740220662,
"loss": 2.5523,
"theoretical_loss": 3.71613386652103,
"tokens_seen": 828833792
},
{
"epoch": 0.25,
"learning_rate": 0.0003782146439317954,
"loss": 2.7889,
"theoretical_loss": 3.7161049262660484,
"tokens_seen": 828899328
},
{
"epoch": 0.25,
"learning_rate": 0.0003782046138415246,
"loss": 2.8731,
"theoretical_loss": 3.7160759889397235,
"tokens_seen": 828964864
},
{
"epoch": 0.25,
"objective/train/docs_used": 478014,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.946833610534668,
"objective/train/theoretical_loss": 3.7160470545415274,
"objective/train/tokens_used": 849490400,
"theoretical_loss": 3.7160470545415274,
"tokens_seen": 829030400
},
{
"epoch": 0.25,
"learning_rate": 0.00037819458375125376,
"loss": 2.7679,
"theoretical_loss": 3.7160470545415274,
"tokens_seen": 829030400
},
{
"epoch": 0.25,
"learning_rate": 0.000378184553660983,
"loss": 2.7083,
"theoretical_loss": 3.7160181230709326,
"tokens_seen": 829095936
},
{
"epoch": 0.25,
"learning_rate": 0.0003781745235707121,
"loss": 2.8322,
"theoretical_loss": 3.7159891945274115,
"tokens_seen": 829161472
},
{
"epoch": 0.25,
"learning_rate": 0.00037816449348044136,
"loss": 2.6655,
"theoretical_loss": 3.715960268910437,
"tokens_seen": 829227008
},
{
"epoch": 0.25,
"learning_rate": 0.0003781544633901705,
"loss": 2.9303,
"theoretical_loss": 3.715931346219481,
"tokens_seen": 829292544
},
{
"epoch": 0.25,
"learning_rate": 0.0003781444332998997,
"loss": 2.8083,
"theoretical_loss": 3.715902426454017,
"tokens_seen": 829358080
},
{
"epoch": 0.25,
"learning_rate": 0.0003781344032096289,
"loss": 2.7229,
"theoretical_loss": 3.715873509613517,
"tokens_seen": 829423616
},
{
"epoch": 0.25,
"learning_rate": 0.0003781243731193581,
"loss": 2.819,
"theoretical_loss": 3.7158445956974555,
"tokens_seen": 829489152
},
{
"epoch": 0.25,
"learning_rate": 0.00037811434302908726,
"loss": 2.8863,
"theoretical_loss": 3.715815684705304,
"tokens_seen": 829554688
},
{
"epoch": 0.25,
"learning_rate": 0.0003781043129388165,
"loss": 2.6033,
"theoretical_loss": 3.715786776636538,
"tokens_seen": 829620224
},
{
"epoch": 0.25,
"learning_rate": 0.0003780942828485456,
"loss": 2.8563,
"theoretical_loss": 3.715757871490629,
"tokens_seen": 829685760
},
{
"epoch": 0.25,
"learning_rate": 0.00037808425275827486,
"loss": 2.8748,
"theoretical_loss": 3.715728969267052,
"tokens_seen": 829751296
},
{
"epoch": 0.25,
"learning_rate": 0.000378074222668004,
"loss": 2.7935,
"theoretical_loss": 3.71570006996528,
"tokens_seen": 829816832
},
{
"epoch": 0.25,
"learning_rate": 0.0003780641925777332,
"loss": 2.5931,
"theoretical_loss": 3.715671173584787,
"tokens_seen": 829882368
},
{
"epoch": 0.25,
"learning_rate": 0.0003780541624874624,
"loss": 2.6617,
"theoretical_loss": 3.7156422801250475,
"tokens_seen": 829947904
},
{
"epoch": 0.25,
"learning_rate": 0.0003780441323971916,
"loss": 2.8459,
"theoretical_loss": 3.7156133895855357,
"tokens_seen": 830013440
},
{
"epoch": 0.25,
"learning_rate": 0.00037803410230692077,
"loss": 2.537,
"theoretical_loss": 3.7155845019657248,
"tokens_seen": 830078976
},
{
"epoch": 0.25,
"learning_rate": 0.00037802407221664995,
"loss": 2.6426,
"theoretical_loss": 3.7155556172650903,
"tokens_seen": 830144512
},
{
"epoch": 0.25,
"learning_rate": 0.00037801404212637913,
"loss": 2.7212,
"theoretical_loss": 3.715526735483107,
"tokens_seen": 830210048
},
{
"epoch": 0.25,
"learning_rate": 0.00037800401203610836,
"loss": 2.9911,
"theoretical_loss": 3.7154978566192485,
"tokens_seen": 830275584
},
{
"epoch": 0.25,
"learning_rate": 0.0003779939819458375,
"loss": 2.8867,
"theoretical_loss": 3.715468980672991,
"tokens_seen": 830341120
},
{
"epoch": 0.25,
"learning_rate": 0.00037798395185556673,
"loss": 2.7502,
"theoretical_loss": 3.715440107643808,
"tokens_seen": 830406656
},
{
"epoch": 0.25,
"learning_rate": 0.00037797392176529585,
"loss": 2.6845,
"theoretical_loss": 3.7154112375311765,
"tokens_seen": 830472192
},
{
"epoch": 0.25,
"learning_rate": 0.0003779638916750251,
"loss": 2.8462,
"theoretical_loss": 3.71538237033457,
"tokens_seen": 830537728
},
{
"epoch": 0.25,
"learning_rate": 0.00037795386158475427,
"loss": 3.0778,
"theoretical_loss": 3.7153535060534644,
"tokens_seen": 830603264
},
{
"epoch": 0.25,
"objective/train/docs_used": 479315,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0639984607696533,
"objective/train/theoretical_loss": 3.7153246446873363,
"objective/train/tokens_used": 851128800,
"theoretical_loss": 3.7153246446873363,
"tokens_seen": 830668800
},
{
"epoch": 0.25,
"learning_rate": 0.00037794383149448345,
"loss": 2.9552,
"theoretical_loss": 3.7153246446873363,
"tokens_seen": 830668800
},
{
"epoch": 0.25,
"learning_rate": 0.00037793380140421263,
"loss": 2.9306,
"theoretical_loss": 3.71529578623566,
"tokens_seen": 830734336
},
{
"epoch": 0.25,
"learning_rate": 0.00037792377131394187,
"loss": 2.6139,
"theoretical_loss": 3.715266930697912,
"tokens_seen": 830799872
},
{
"epoch": 0.25,
"learning_rate": 0.000377913741223671,
"loss": 2.9602,
"theoretical_loss": 3.715238078073568,
"tokens_seen": 830865408
},
{
"epoch": 0.25,
"learning_rate": 0.00037790371113340023,
"loss": 2.6543,
"theoretical_loss": 3.715209228362104,
"tokens_seen": 830930944
},
{
"epoch": 0.25,
"learning_rate": 0.00037789368104312936,
"loss": 2.6834,
"theoretical_loss": 3.715180381562997,
"tokens_seen": 830996480
},
{
"epoch": 0.25,
"learning_rate": 0.0003778836509528586,
"loss": 2.7777,
"theoretical_loss": 3.7151515376757223,
"tokens_seen": 831062016
},
{
"epoch": 0.25,
"learning_rate": 0.0003778736208625878,
"loss": 2.7747,
"theoretical_loss": 3.7151226966997566,
"tokens_seen": 831127552
},
{
"epoch": 0.25,
"learning_rate": 0.00037786359077231695,
"loss": 2.6538,
"theoretical_loss": 3.715093858634577,
"tokens_seen": 831193088
},
{
"epoch": 0.25,
"learning_rate": 0.00037785356068204614,
"loss": 2.7834,
"theoretical_loss": 3.71506502347966,
"tokens_seen": 831258624
},
{
"epoch": 0.25,
"learning_rate": 0.0003778435305917753,
"loss": 2.7525,
"theoretical_loss": 3.715036191234483,
"tokens_seen": 831324160
},
{
"epoch": 0.25,
"learning_rate": 0.0003778335005015045,
"loss": 2.6787,
"theoretical_loss": 3.715007361898522,
"tokens_seen": 831389696
},
{
"epoch": 0.25,
"learning_rate": 0.00037782347041123373,
"loss": 2.8364,
"theoretical_loss": 3.7149785354712552,
"tokens_seen": 831455232
},
{
"epoch": 0.25,
"learning_rate": 0.00037781344032096286,
"loss": 2.6082,
"theoretical_loss": 3.7149497119521593,
"tokens_seen": 831520768
},
{
"epoch": 0.25,
"learning_rate": 0.0003778034102306921,
"loss": 2.6332,
"theoretical_loss": 3.7149208913407117,
"tokens_seen": 831586304
},
{
"epoch": 0.25,
"learning_rate": 0.0003777933801404213,
"loss": 2.6998,
"theoretical_loss": 3.71489207363639,
"tokens_seen": 831651840
},
{
"epoch": 0.25,
"learning_rate": 0.00037778335005015046,
"loss": 2.5125,
"theoretical_loss": 3.714863258838672,
"tokens_seen": 831717376
},
{
"epoch": 0.25,
"learning_rate": 0.00037777331995987964,
"loss": 2.7522,
"theoretical_loss": 3.7148344469470356,
"tokens_seen": 831782912
},
{
"epoch": 0.25,
"learning_rate": 0.0003777632898696088,
"loss": 2.5927,
"theoretical_loss": 3.714805637960959,
"tokens_seen": 831848448
},
{
"epoch": 0.25,
"learning_rate": 0.000377753259779338,
"loss": 2.9185,
"theoretical_loss": 3.71477683187992,
"tokens_seen": 831913984
},
{
"epoch": 0.25,
"learning_rate": 0.00037774322968906724,
"loss": 2.5378,
"theoretical_loss": 3.7147480287033967,
"tokens_seen": 831979520
},
{
"epoch": 0.25,
"learning_rate": 0.00037773319959879636,
"loss": 2.6364,
"theoretical_loss": 3.7147192284308677,
"tokens_seen": 832045056
},
{
"epoch": 0.25,
"learning_rate": 0.0003777231695085256,
"loss": 2.8516,
"theoretical_loss": 3.7146904310618116,
"tokens_seen": 832110592
},
{
"epoch": 0.25,
"learning_rate": 0.0003777131394182547,
"loss": 2.8394,
"theoretical_loss": 3.7146616365957073,
"tokens_seen": 832176128
},
{
"epoch": 0.25,
"learning_rate": 0.00037770310932798396,
"loss": 2.5992,
"theoretical_loss": 3.7146328450320327,
"tokens_seen": 832241664
},
{
"epoch": 0.25,
"objective/train/docs_used": 479893,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3925576210021973,
"objective/train/theoretical_loss": 3.714604056370267,
"objective/train/tokens_used": 852767200,
"theoretical_loss": 3.714604056370267,
"tokens_seen": 832307200
},
{
"epoch": 0.25,
"learning_rate": 0.00037769307923771314,
"loss": 2.6285,
"theoretical_loss": 3.714604056370267,
"tokens_seen": 832307200
},
{
"epoch": 0.25,
"learning_rate": 0.0003776830491474423,
"loss": 2.8159,
"theoretical_loss": 3.71457527060989,
"tokens_seen": 832372736
},
{
"epoch": 0.25,
"learning_rate": 0.0003776730190571715,
"loss": 2.7639,
"theoretical_loss": 3.7145464877503804,
"tokens_seen": 832438272
},
{
"epoch": 0.25,
"learning_rate": 0.0003776629889669007,
"loss": 2.7796,
"theoretical_loss": 3.7145177077912175,
"tokens_seen": 832503808
},
{
"epoch": 0.25,
"learning_rate": 0.00037765295887662987,
"loss": 3.0142,
"theoretical_loss": 3.714488930731881,
"tokens_seen": 832569344
},
{
"epoch": 0.25,
"learning_rate": 0.0003776429287863591,
"loss": 2.7225,
"theoretical_loss": 3.7144601565718496,
"tokens_seen": 832634880
},
{
"epoch": 0.25,
"learning_rate": 0.00037763289869608823,
"loss": 2.8554,
"theoretical_loss": 3.714431385310604,
"tokens_seen": 832700416
},
{
"epoch": 0.25,
"learning_rate": 0.00037762286860581746,
"loss": 2.8259,
"theoretical_loss": 3.7144026169476243,
"tokens_seen": 832765952
},
{
"epoch": 0.25,
"learning_rate": 0.0003776128385155467,
"loss": 2.8776,
"theoretical_loss": 3.714373851482389,
"tokens_seen": 832831488
},
{
"epoch": 0.25,
"learning_rate": 0.0003776028084252758,
"loss": 2.6099,
"theoretical_loss": 3.7143450889143796,
"tokens_seen": 832897024
},
{
"epoch": 0.25,
"learning_rate": 0.00037759277833500506,
"loss": 2.6554,
"theoretical_loss": 3.7143163292430765,
"tokens_seen": 832962560
},
{
"epoch": 0.25,
"learning_rate": 0.0003775827482447342,
"loss": 2.6405,
"theoretical_loss": 3.714287572467959,
"tokens_seen": 833028096
},
{
"epoch": 0.25,
"learning_rate": 0.0003775727181544634,
"loss": 3.0464,
"theoretical_loss": 3.714258818588508,
"tokens_seen": 833093632
},
{
"epoch": 0.25,
"learning_rate": 0.0003775626880641926,
"loss": 2.5907,
"theoretical_loss": 3.7142300676042046,
"tokens_seen": 833159168
},
{
"epoch": 0.25,
"learning_rate": 0.0003775526579739218,
"loss": 2.7689,
"theoretical_loss": 3.7142013195145287,
"tokens_seen": 833224704
},
{
"epoch": 0.25,
"learning_rate": 0.00037754262788365097,
"loss": 2.6223,
"theoretical_loss": 3.7141725743189626,
"tokens_seen": 833290240
},
{
"epoch": 0.25,
"learning_rate": 0.00037753259779338015,
"loss": 2.872,
"theoretical_loss": 3.714143832016987,
"tokens_seen": 833355776
},
{
"epoch": 0.25,
"learning_rate": 0.00037752256770310933,
"loss": 2.5715,
"theoretical_loss": 3.7141150926080817,
"tokens_seen": 833421312
},
{
"epoch": 0.25,
"learning_rate": 0.00037751253761283857,
"loss": 2.611,
"theoretical_loss": 3.7140863560917294,
"tokens_seen": 833486848
},
{
"epoch": 0.25,
"learning_rate": 0.0003775025075225677,
"loss": 2.7251,
"theoretical_loss": 3.7140576224674113,
"tokens_seen": 833552384
},
{
"epoch": 0.25,
"learning_rate": 0.00037749247743229693,
"loss": 2.8024,
"theoretical_loss": 3.714028891734609,
"tokens_seen": 833617920
},
{
"epoch": 0.25,
"learning_rate": 0.00037748244734202605,
"loss": 2.7034,
"theoretical_loss": 3.714000163892804,
"tokens_seen": 833683456
},
{
"epoch": 0.25,
"learning_rate": 0.0003774724172517553,
"loss": 2.6596,
"theoretical_loss": 3.7139714389414786,
"tokens_seen": 833748992
},
{
"epoch": 0.25,
"learning_rate": 0.00037746238716148447,
"loss": 2.5804,
"theoretical_loss": 3.7139427168801147,
"tokens_seen": 833814528
},
{
"epoch": 0.25,
"learning_rate": 0.00037745235707121365,
"loss": 2.9389,
"theoretical_loss": 3.713913997708194,
"tokens_seen": 833880064
},
{
"epoch": 0.25,
"objective/train/docs_used": 480416,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0427303314208984,
"objective/train/theoretical_loss": 3.713885281425199,
"objective/train/tokens_used": 854405600,
"theoretical_loss": 3.713885281425199,
"tokens_seen": 833945600
},
{
"epoch": 0.25,
"learning_rate": 0.00037744232698094283,
"loss": 2.8136,
"theoretical_loss": 3.713885281425199,
"tokens_seen": 833945600
},
{
"epoch": 0.25,
"learning_rate": 0.00037743229689067207,
"loss": 2.8858,
"theoretical_loss": 3.7138565680306117,
"tokens_seen": 834011136
},
{
"epoch": 0.25,
"learning_rate": 0.0003774222668004012,
"loss": 2.8619,
"theoretical_loss": 3.7138278575239156,
"tokens_seen": 834076672
},
{
"epoch": 0.25,
"learning_rate": 0.00037741223671013043,
"loss": 2.7992,
"theoretical_loss": 3.7137991499045926,
"tokens_seen": 834142208
},
{
"epoch": 0.25,
"learning_rate": 0.00037740220661985956,
"loss": 2.6126,
"theoretical_loss": 3.713770445172126,
"tokens_seen": 834207744
},
{
"epoch": 0.25,
"learning_rate": 0.0003773921765295888,
"loss": 2.8539,
"theoretical_loss": 3.7137417433259987,
"tokens_seen": 834273280
},
{
"epoch": 0.25,
"learning_rate": 0.000377382146439318,
"loss": 2.5988,
"theoretical_loss": 3.7137130443656936,
"tokens_seen": 834338816
},
{
"epoch": 0.25,
"learning_rate": 0.00037737211634904715,
"loss": 2.6746,
"theoretical_loss": 3.713684348290693,
"tokens_seen": 834404352
},
{
"epoch": 0.25,
"learning_rate": 0.00037736208625877634,
"loss": 3.0138,
"theoretical_loss": 3.713655655100482,
"tokens_seen": 834469888
},
{
"epoch": 0.25,
"learning_rate": 0.0003773520561685055,
"loss": 2.9321,
"theoretical_loss": 3.713626964794543,
"tokens_seen": 834535424
},
{
"epoch": 0.25,
"learning_rate": 0.0003773420260782347,
"loss": 2.6799,
"theoretical_loss": 3.7135982773723595,
"tokens_seen": 834600960
},
{
"epoch": 0.25,
"learning_rate": 0.00037733199598796393,
"loss": 3.0015,
"theoretical_loss": 3.713569592833416,
"tokens_seen": 834666496
},
{
"epoch": 0.25,
"learning_rate": 0.00037732196589769306,
"loss": 2.7036,
"theoretical_loss": 3.713540911177195,
"tokens_seen": 834732032
},
{
"epoch": 0.25,
"learning_rate": 0.0003773119358074223,
"loss": 2.4857,
"theoretical_loss": 3.713512232403182,
"tokens_seen": 834797568
},
{
"epoch": 0.25,
"learning_rate": 0.0003773019057171515,
"loss": 2.8052,
"theoretical_loss": 3.7134835565108606,
"tokens_seen": 834863104
},
{
"epoch": 0.25,
"learning_rate": 0.00037729187562688066,
"loss": 2.7363,
"theoretical_loss": 3.7134548834997148,
"tokens_seen": 834928640
},
{
"epoch": 0.25,
"learning_rate": 0.00037728184553660984,
"loss": 2.7412,
"theoretical_loss": 3.713426213369229,
"tokens_seen": 834994176
},
{
"epoch": 0.25,
"learning_rate": 0.000377271815446339,
"loss": 2.7004,
"theoretical_loss": 3.7133975461188884,
"tokens_seen": 835059712
},
{
"epoch": 0.25,
"learning_rate": 0.0003772617853560682,
"loss": 2.8129,
"theoretical_loss": 3.7133688817481767,
"tokens_seen": 835125248
},
{
"epoch": 0.25,
"learning_rate": 0.00037725175526579744,
"loss": 2.7683,
"theoretical_loss": 3.7133402202565793,
"tokens_seen": 835190784
},
{
"epoch": 0.25,
"learning_rate": 0.00037724172517552656,
"loss": 2.7813,
"theoretical_loss": 3.713311561643581,
"tokens_seen": 835256320
},
{
"epoch": 0.25,
"learning_rate": 0.0003772316950852558,
"loss": 2.7644,
"theoretical_loss": 3.7132829059086667,
"tokens_seen": 835321856
},
{
"epoch": 0.25,
"learning_rate": 0.0003772216649949849,
"loss": 2.6473,
"theoretical_loss": 3.713254253051322,
"tokens_seen": 835387392
},
{
"epoch": 0.25,
"learning_rate": 0.00037721163490471416,
"loss": 2.7979,
"theoretical_loss": 3.7132256030710318,
"tokens_seen": 835452928
},
{
"epoch": 0.25,
"learning_rate": 0.00037720160481444334,
"loss": 2.8127,
"theoretical_loss": 3.713196955967282,
"tokens_seen": 835518464
},
{
"epoch": 0.25,
"objective/train/docs_used": 481156,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.772949695587158,
"objective/train/theoretical_loss": 3.713168311739558,
"objective/train/tokens_used": 856044000,
"theoretical_loss": 3.713168311739558,
"tokens_seen": 835584000
},
{
"epoch": 0.25,
"learning_rate": 0.0003771915747241725,
"loss": 2.6229,
"theoretical_loss": 3.713168311739558,
"tokens_seen": 835584000
},
{
"epoch": 0.25,
"learning_rate": 0.0003771815446339017,
"loss": 2.8202,
"theoretical_loss": 3.7131396703873456,
"tokens_seen": 835649536
},
{
"epoch": 0.25,
"learning_rate": 0.0003771715145436309,
"loss": 2.6472,
"theoretical_loss": 3.71311103191013,
"tokens_seen": 835715072
},
{
"epoch": 0.25,
"learning_rate": 0.00037716148445336007,
"loss": 2.8891,
"theoretical_loss": 3.713082396307398,
"tokens_seen": 835780608
},
{
"epoch": 0.25,
"learning_rate": 0.0003771514543630893,
"loss": 2.6245,
"theoretical_loss": 3.7130537635786354,
"tokens_seen": 835846144
},
{
"epoch": 0.25,
"learning_rate": 0.00037714142427281843,
"loss": 2.6752,
"theoretical_loss": 3.7130251337233284,
"tokens_seen": 835911680
},
{
"epoch": 0.25,
"learning_rate": 0.00037713139418254766,
"loss": 2.8157,
"theoretical_loss": 3.7129965067409634,
"tokens_seen": 835977216
},
{
"epoch": 0.25,
"learning_rate": 0.00037712136409227685,
"loss": 2.6388,
"theoretical_loss": 3.712967882631027,
"tokens_seen": 836042752
},
{
"epoch": 0.25,
"learning_rate": 0.000377111334002006,
"loss": 2.9241,
"theoretical_loss": 3.7129392613930063,
"tokens_seen": 836108288
},
{
"epoch": 0.25,
"learning_rate": 0.0003771013039117352,
"loss": 2.6929,
"theoretical_loss": 3.712910643026387,
"tokens_seen": 836173824
},
{
"epoch": 0.25,
"learning_rate": 0.0003770912738214644,
"loss": 2.8544,
"theoretical_loss": 3.712882027530657,
"tokens_seen": 836239360
},
{
"epoch": 0.25,
"learning_rate": 0.00037708124373119357,
"loss": 3.0015,
"theoretical_loss": 3.712853414905303,
"tokens_seen": 836304896
},
{
"epoch": 0.25,
"learning_rate": 0.0003770712136409228,
"loss": 2.6845,
"theoretical_loss": 3.712824805149812,
"tokens_seen": 836370432
},
{
"epoch": 0.25,
"learning_rate": 0.00037706118355065193,
"loss": 2.8922,
"theoretical_loss": 3.7127961982636712,
"tokens_seen": 836435968
},
{
"epoch": 0.25,
"learning_rate": 0.00037705115346038117,
"loss": 2.7206,
"theoretical_loss": 3.712767594246369,
"tokens_seen": 836501504
},
{
"epoch": 0.25,
"learning_rate": 0.0003770411233701103,
"loss": 2.8132,
"theoretical_loss": 3.7127389930973917,
"tokens_seen": 836567040
},
{
"epoch": 0.25,
"learning_rate": 0.00037703109327983953,
"loss": 2.8708,
"theoretical_loss": 3.712710394816228,
"tokens_seen": 836632576
},
{
"epoch": 0.25,
"learning_rate": 0.0003770210631895687,
"loss": 2.7132,
"theoretical_loss": 3.7126817994023655,
"tokens_seen": 836698112
},
{
"epoch": 0.25,
"learning_rate": 0.0003770110330992979,
"loss": 2.7194,
"theoretical_loss": 3.7126532068552915,
"tokens_seen": 836763648
},
{
"epoch": 0.25,
"learning_rate": 0.0003770010030090271,
"loss": 2.6615,
"theoretical_loss": 3.7126246171744945,
"tokens_seen": 836829184
},
{
"epoch": 0.25,
"learning_rate": 0.00037699097291875625,
"loss": 2.6387,
"theoretical_loss": 3.712596030359463,
"tokens_seen": 836894720
},
{
"epoch": 0.25,
"learning_rate": 0.00037698094282848544,
"loss": 2.5661,
"theoretical_loss": 3.712567446409685,
"tokens_seen": 836960256
},
{
"epoch": 0.25,
"learning_rate": 0.00037697091273821467,
"loss": 2.7436,
"theoretical_loss": 3.7125388653246496,
"tokens_seen": 837025792
},
{
"epoch": 0.25,
"learning_rate": 0.0003769608826479438,
"loss": 2.7242,
"theoretical_loss": 3.7125102871038447,
"tokens_seen": 837091328
},
{
"epoch": 0.25,
"learning_rate": 0.00037695085255767303,
"loss": 2.8043,
"theoretical_loss": 3.7124817117467597,
"tokens_seen": 837156864
},
{
"epoch": 0.25,
"objective/train/docs_used": 482509,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.509737968444824,
"objective/train/theoretical_loss": 3.712453139252883,
"objective/train/tokens_used": 857682400,
"theoretical_loss": 3.712453139252883,
"tokens_seen": 837222400
},
{
"epoch": 0.25,
"learning_rate": 0.0003769408224674022,
"loss": 2.6631,
"theoretical_loss": 3.712453139252883,
"tokens_seen": 837222400
},
{
"epoch": 0.25,
"learning_rate": 0.0003769307923771314,
"loss": 2.5743,
"theoretical_loss": 3.712424569621703,
"tokens_seen": 837287936
},
{
"epoch": 0.25,
"learning_rate": 0.0003769207622868606,
"loss": 2.7263,
"theoretical_loss": 3.712396002852711,
"tokens_seen": 837353472
},
{
"epoch": 0.25,
"learning_rate": 0.00037691073219658976,
"loss": 2.6958,
"theoretical_loss": 3.712367438945394,
"tokens_seen": 837419008
},
{
"epoch": 0.25,
"learning_rate": 0.00037690070210631894,
"loss": 2.6002,
"theoretical_loss": 3.712338877899242,
"tokens_seen": 837484544
},
{
"epoch": 0.25,
"learning_rate": 0.0003768906720160482,
"loss": 2.7062,
"theoretical_loss": 3.7123103197137457,
"tokens_seen": 837550080
},
{
"epoch": 0.25,
"learning_rate": 0.0003768806419257773,
"loss": 2.5197,
"theoretical_loss": 3.7122817643883934,
"tokens_seen": 837615616
},
{
"epoch": 0.25,
"learning_rate": 0.00037687061183550654,
"loss": 2.519,
"theoretical_loss": 3.7122532119226754,
"tokens_seen": 837681152
},
{
"epoch": 0.25,
"learning_rate": 0.0003768605817452357,
"loss": 2.5725,
"theoretical_loss": 3.712224662316082,
"tokens_seen": 837746688
},
{
"epoch": 0.25,
"learning_rate": 0.0003768505516549649,
"loss": 2.6365,
"theoretical_loss": 3.7121961155681027,
"tokens_seen": 837812224
},
{
"epoch": 0.25,
"learning_rate": 0.00037684052156469413,
"loss": 2.6872,
"theoretical_loss": 3.712167571678228,
"tokens_seen": 837877760
},
{
"epoch": 0.25,
"learning_rate": 0.00037683049147442326,
"loss": 2.6711,
"theoretical_loss": 3.712139030645948,
"tokens_seen": 837943296
},
{
"epoch": 0.25,
"learning_rate": 0.0003768204613841525,
"loss": 2.8883,
"theoretical_loss": 3.7121104924707535,
"tokens_seen": 838008832
},
{
"epoch": 0.25,
"learning_rate": 0.0003768104312938817,
"loss": 2.6613,
"theoretical_loss": 3.712081957152135,
"tokens_seen": 838074368
},
{
"epoch": 0.25,
"learning_rate": 0.00037680040120361086,
"loss": 2.4701,
"theoretical_loss": 3.712053424689583,
"tokens_seen": 838139904
},
{
"epoch": 0.25,
"learning_rate": 0.00037679037111334004,
"loss": 2.6771,
"theoretical_loss": 3.712024895082588,
"tokens_seen": 838205440
},
{
"epoch": 0.25,
"learning_rate": 0.0003767803410230692,
"loss": 2.8247,
"theoretical_loss": 3.711996368330642,
"tokens_seen": 838270976
},
{
"epoch": 0.25,
"learning_rate": 0.0003767703109327984,
"loss": 2.7239,
"theoretical_loss": 3.7119678444332354,
"tokens_seen": 838336512
},
{
"epoch": 0.25,
"learning_rate": 0.00037676028084252764,
"loss": 2.8107,
"theoretical_loss": 3.71193932338986,
"tokens_seen": 838402048
},
{
"epoch": 0.25,
"learning_rate": 0.00037675025075225676,
"loss": 2.8059,
"theoretical_loss": 3.711910805200006,
"tokens_seen": 838467584
},
{
"epoch": 0.25,
"learning_rate": 0.000376740220661986,
"loss": 2.8619,
"theoretical_loss": 3.7118822898631665,
"tokens_seen": 838533120
},
{
"epoch": 0.25,
"learning_rate": 0.0003767301905717151,
"loss": 2.7824,
"theoretical_loss": 3.711853777378832,
"tokens_seen": 838598656
},
{
"epoch": 0.25,
"learning_rate": 0.00037672016048144436,
"loss": 2.7684,
"theoretical_loss": 3.7118252677464945,
"tokens_seen": 838664192
},
{
"epoch": 0.25,
"learning_rate": 0.00037671013039117354,
"loss": 2.8005,
"theoretical_loss": 3.7117967609656457,
"tokens_seen": 838729728
},
{
"epoch": 0.25,
"learning_rate": 0.0003767001003009027,
"loss": 2.7785,
"theoretical_loss": 3.7117682570357786,
"tokens_seen": 838795264
},
{
"epoch": 0.25,
"objective/train/docs_used": 483094,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.447807788848877,
"objective/train/theoretical_loss": 3.7117397559563843,
"objective/train/tokens_used": 859320800,
"theoretical_loss": 3.7117397559563843,
"tokens_seen": 838860800
},
{
"epoch": 0.25,
"learning_rate": 0.0003766900702106319,
"loss": 2.8866,
"theoretical_loss": 3.7117397559563843,
"tokens_seen": 838860800
},
{
"epoch": 0.25,
"learning_rate": 0.0003766800401203611,
"loss": 2.7138,
"theoretical_loss": 3.7117112577269555,
"tokens_seen": 838926336
},
{
"epoch": 0.25,
"learning_rate": 0.00037667001003009027,
"loss": 2.8894,
"theoretical_loss": 3.7116827623469844,
"tokens_seen": 838991872
},
{
"epoch": 0.25,
"learning_rate": 0.0003766599799398195,
"loss": 2.7039,
"theoretical_loss": 3.7116542698159636,
"tokens_seen": 839057408
},
{
"epoch": 0.25,
"learning_rate": 0.00037664994984954863,
"loss": 2.9631,
"theoretical_loss": 3.711625780133386,
"tokens_seen": 839122944
},
{
"epoch": 0.25,
"learning_rate": 0.00037663991975927786,
"loss": 3.0211,
"theoretical_loss": 3.711597293298744,
"tokens_seen": 839188480
},
{
"epoch": 0.25,
"learning_rate": 0.00037662988966900705,
"loss": 2.716,
"theoretical_loss": 3.7115688093115313,
"tokens_seen": 839254016
},
{
"epoch": 0.25,
"learning_rate": 0.0003766198595787362,
"loss": 2.5976,
"theoretical_loss": 3.7115403281712407,
"tokens_seen": 839319552
},
{
"epoch": 0.25,
"learning_rate": 0.0003766098294884654,
"loss": 2.7309,
"theoretical_loss": 3.7115118498773647,
"tokens_seen": 839385088
},
{
"epoch": 0.25,
"learning_rate": 0.0003765997993981946,
"loss": 2.9032,
"theoretical_loss": 3.711483374429397,
"tokens_seen": 839450624
},
{
"epoch": 0.25,
"learning_rate": 0.00037658976930792377,
"loss": 2.7365,
"theoretical_loss": 3.7114549018268317,
"tokens_seen": 839516160
},
{
"epoch": 0.25,
"learning_rate": 0.000376579739217653,
"loss": 2.936,
"theoretical_loss": 3.7114264320691612,
"tokens_seen": 839581696
},
{
"epoch": 0.25,
"learning_rate": 0.00037656970912738213,
"loss": 2.9003,
"theoretical_loss": 3.7113979651558804,
"tokens_seen": 839647232
},
{
"epoch": 0.25,
"learning_rate": 0.00037655967903711137,
"loss": 2.6155,
"theoretical_loss": 3.7113695010864824,
"tokens_seen": 839712768
},
{
"epoch": 0.25,
"learning_rate": 0.0003765496489468405,
"loss": 2.9367,
"theoretical_loss": 3.7113410398604616,
"tokens_seen": 839778304
},
{
"epoch": 0.25,
"learning_rate": 0.00037653961885656973,
"loss": 2.9881,
"theoretical_loss": 3.711312581477311,
"tokens_seen": 839843840
},
{
"epoch": 0.25,
"learning_rate": 0.0003765295887662989,
"loss": 2.652,
"theoretical_loss": 3.7112841259365266,
"tokens_seen": 839909376
},
{
"epoch": 0.25,
"learning_rate": 0.0003765195586760281,
"loss": 2.7591,
"theoretical_loss": 3.7112556732376016,
"tokens_seen": 839974912
},
{
"epoch": 0.25,
"learning_rate": 0.0003765095285857573,
"loss": 2.9137,
"theoretical_loss": 3.711227223380031,
"tokens_seen": 840040448
},
{
"epoch": 0.25,
"learning_rate": 0.00037649949849548645,
"loss": 2.879,
"theoretical_loss": 3.7111987763633083,
"tokens_seen": 840105984
},
{
"epoch": 0.25,
"learning_rate": 0.00037648946840521564,
"loss": 2.7503,
"theoretical_loss": 3.71117033218693,
"tokens_seen": 840171520
},
{
"epoch": 0.25,
"learning_rate": 0.00037647943831494487,
"loss": 2.5582,
"theoretical_loss": 3.71114189085039,
"tokens_seen": 840237056
},
{
"epoch": 0.25,
"learning_rate": 0.000376469408224674,
"loss": 3.0146,
"theoretical_loss": 3.7111134523531835,
"tokens_seen": 840302592
},
{
"epoch": 0.25,
"learning_rate": 0.00037645937813440323,
"loss": 2.7823,
"theoretical_loss": 3.711085016694805,
"tokens_seen": 840368128
},
{
"epoch": 0.25,
"learning_rate": 0.0003764493480441324,
"loss": 2.8959,
"theoretical_loss": 3.7110565838747505,
"tokens_seen": 840433664
},
{
"epoch": 0.25,
"objective/train/docs_used": 483926,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8763413429260254,
"objective/train/theoretical_loss": 3.7110281538925154,
"objective/train/tokens_used": 860959200,
"theoretical_loss": 3.7110281538925154,
"tokens_seen": 840499200
},
{
"epoch": 0.25,
"learning_rate": 0.0003764393179538616,
"loss": 2.6683,
"theoretical_loss": 3.7110281538925154,
"tokens_seen": 840499200
},
{
"epoch": 0.25,
"learning_rate": 0.0003764292878635908,
"loss": 2.6297,
"theoretical_loss": 3.7109997267475947,
"tokens_seen": 840564736
},
{
"epoch": 0.25,
"learning_rate": 0.00037641925777331996,
"loss": 2.6832,
"theoretical_loss": 3.710971302439485,
"tokens_seen": 840630272
},
{
"epoch": 0.25,
"learning_rate": 0.00037640922768304914,
"loss": 2.776,
"theoretical_loss": 3.710942880967681,
"tokens_seen": 840695808
},
{
"epoch": 0.25,
"learning_rate": 0.0003763991975927784,
"loss": 2.871,
"theoretical_loss": 3.710914462331679,
"tokens_seen": 840761344
},
{
"epoch": 0.25,
"learning_rate": 0.0003763891675025075,
"loss": 2.8175,
"theoretical_loss": 3.710886046530975,
"tokens_seen": 840826880
},
{
"epoch": 0.25,
"learning_rate": 0.00037637913741223674,
"loss": 2.9003,
"theoretical_loss": 3.710857633565066,
"tokens_seen": 840892416
},
{
"epoch": 0.25,
"learning_rate": 0.00037636910732196586,
"loss": 2.8229,
"theoretical_loss": 3.7108292234334472,
"tokens_seen": 840957952
},
{
"epoch": 0.25,
"learning_rate": 0.0003763590772316951,
"loss": 2.7856,
"theoretical_loss": 3.710800816135615,
"tokens_seen": 841023488
},
{
"epoch": 0.25,
"learning_rate": 0.0003763490471414243,
"loss": 2.6937,
"theoretical_loss": 3.7107724116710665,
"tokens_seen": 841089024
},
{
"epoch": 0.25,
"learning_rate": 0.00037633901705115346,
"loss": 2.9438,
"theoretical_loss": 3.7107440100392988,
"tokens_seen": 841154560
},
{
"epoch": 0.25,
"learning_rate": 0.00037632898696088264,
"loss": 2.8203,
"theoretical_loss": 3.7107156112398076,
"tokens_seen": 841220096
},
{
"epoch": 0.25,
"learning_rate": 0.0003763189568706119,
"loss": 2.8662,
"theoretical_loss": 3.710687215272091,
"tokens_seen": 841285632
},
{
"epoch": 0.25,
"learning_rate": 0.000376308926780341,
"loss": 2.7946,
"theoretical_loss": 3.7106588221356445,
"tokens_seen": 841351168
},
{
"epoch": 0.25,
"learning_rate": 0.00037629889669007024,
"loss": 2.8416,
"theoretical_loss": 3.7106304318299674,
"tokens_seen": 841416704
},
{
"epoch": 0.25,
"learning_rate": 0.00037628886659979937,
"loss": 2.7334,
"theoretical_loss": 3.7106020443545553,
"tokens_seen": 841482240
},
{
"epoch": 0.26,
"learning_rate": 0.0003762788365095286,
"loss": 2.9068,
"theoretical_loss": 3.7105736597089063,
"tokens_seen": 841547776
},
{
"epoch": 0.26,
"learning_rate": 0.0003762688064192578,
"loss": 2.8162,
"theoretical_loss": 3.7105452778925176,
"tokens_seen": 841613312
},
{
"epoch": 0.26,
"learning_rate": 0.00037625877632898696,
"loss": 2.6369,
"theoretical_loss": 3.7105168989048876,
"tokens_seen": 841678848
},
{
"epoch": 0.26,
"learning_rate": 0.00037624874623871615,
"loss": 3.0235,
"theoretical_loss": 3.7104885227455138,
"tokens_seen": 841744384
},
{
"epoch": 0.26,
"learning_rate": 0.0003762387161484453,
"loss": 2.8327,
"theoretical_loss": 3.710460149413894,
"tokens_seen": 841809920
},
{
"epoch": 0.26,
"learning_rate": 0.0003762286860581745,
"loss": 2.8314,
"theoretical_loss": 3.7104317789095265,
"tokens_seen": 841875456
},
{
"epoch": 0.26,
"learning_rate": 0.00037621865596790374,
"loss": 2.9063,
"theoretical_loss": 3.7104034112319093,
"tokens_seen": 841940992
},
{
"epoch": 0.26,
"learning_rate": 0.00037620862587763287,
"loss": 2.8198,
"theoretical_loss": 3.710375046380541,
"tokens_seen": 842006528
},
{
"epoch": 0.26,
"learning_rate": 0.0003761985957873621,
"loss": 2.8501,
"theoretical_loss": 3.7103466843549207,
"tokens_seen": 842072064
},
{
"epoch": 0.26,
"objective/train/docs_used": 484283,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9245941638946533,
"objective/train/theoretical_loss": 3.710318325154545,
"objective/train/tokens_used": 862597600,
"theoretical_loss": 3.710318325154545,
"tokens_seen": 842137600
},
{
"epoch": 0.26,
"learning_rate": 0.00037618856569709123,
"loss": 3.0961,
"theoretical_loss": 3.710318325154545,
"tokens_seen": 842137600
},
{
"epoch": 0.26,
"learning_rate": 0.00037617853560682047,
"loss": 2.9189,
"theoretical_loss": 3.7102899687789153,
"tokens_seen": 842203136
},
{
"epoch": 0.26,
"learning_rate": 0.00037616850551654965,
"loss": 2.6023,
"theoretical_loss": 3.710261615227528,
"tokens_seen": 842268672
},
{
"epoch": 0.26,
"learning_rate": 0.00037615847542627883,
"loss": 3.0486,
"theoretical_loss": 3.710233264499884,
"tokens_seen": 842334208
},
{
"epoch": 0.26,
"learning_rate": 0.000376148445336008,
"loss": 2.6121,
"theoretical_loss": 3.7102049165954814,
"tokens_seen": 842399744
},
{
"epoch": 0.26,
"learning_rate": 0.00037613841524573725,
"loss": 2.9972,
"theoretical_loss": 3.710176571513819,
"tokens_seen": 842465280
},
{
"epoch": 0.26,
"learning_rate": 0.00037612838515546637,
"loss": 2.9639,
"theoretical_loss": 3.710148229254398,
"tokens_seen": 842530816
},
{
"epoch": 0.26,
"learning_rate": 0.0003761183550651956,
"loss": 2.9467,
"theoretical_loss": 3.710119889816716,
"tokens_seen": 842596352
},
{
"epoch": 0.26,
"learning_rate": 0.0003761083249749248,
"loss": 2.8008,
"theoretical_loss": 3.7100915532002734,
"tokens_seen": 842661888
},
{
"epoch": 0.26,
"learning_rate": 0.00037609829488465397,
"loss": 2.8887,
"theoretical_loss": 3.71006321940457,
"tokens_seen": 842727424
},
{
"epoch": 0.26,
"learning_rate": 0.0003760882647943832,
"loss": 2.7901,
"theoretical_loss": 3.7100348884291057,
"tokens_seen": 842792960
},
{
"epoch": 0.26,
"learning_rate": 0.00037607823470411233,
"loss": 3.0092,
"theoretical_loss": 3.7100065602733805,
"tokens_seen": 842858496
},
{
"epoch": 0.26,
"learning_rate": 0.00037606820461384157,
"loss": 2.6721,
"theoretical_loss": 3.7099782349368944,
"tokens_seen": 842924032
},
{
"epoch": 0.26,
"learning_rate": 0.0003760581745235707,
"loss": 2.6529,
"theoretical_loss": 3.7099499124191473,
"tokens_seen": 842989568
},
{
"epoch": 0.26,
"learning_rate": 0.00037604814443329993,
"loss": 2.8272,
"theoretical_loss": 3.709921592719641,
"tokens_seen": 843055104
},
{
"epoch": 0.26,
"learning_rate": 0.0003760381143430291,
"loss": 2.7953,
"theoretical_loss": 3.7098932758378744,
"tokens_seen": 843120640
},
{
"epoch": 0.26,
"learning_rate": 0.0003760280842527583,
"loss": 2.7496,
"theoretical_loss": 3.7098649617733486,
"tokens_seen": 843186176
},
{
"epoch": 0.26,
"learning_rate": 0.0003760180541624875,
"loss": 2.8355,
"theoretical_loss": 3.709836650525565,
"tokens_seen": 843251712
},
{
"epoch": 0.26,
"learning_rate": 0.00037600802407221665,
"loss": 2.8497,
"theoretical_loss": 3.709808342094024,
"tokens_seen": 843317248
},
{
"epoch": 0.26,
"learning_rate": 0.00037599799398194584,
"loss": 2.7832,
"theoretical_loss": 3.709780036478227,
"tokens_seen": 843382784
},
{
"epoch": 0.26,
"learning_rate": 0.00037598796389167507,
"loss": 2.5483,
"theoretical_loss": 3.7097517336776744,
"tokens_seen": 843448320
},
{
"epoch": 0.26,
"learning_rate": 0.0003759779338014042,
"loss": 2.835,
"theoretical_loss": 3.7097234336918685,
"tokens_seen": 843513856
},
{
"epoch": 0.26,
"learning_rate": 0.00037596790371113343,
"loss": 2.6419,
"theoretical_loss": 3.7096951365203097,
"tokens_seen": 843579392
},
{
"epoch": 0.26,
"learning_rate": 0.0003759578736208626,
"loss": 2.6895,
"theoretical_loss": 3.7096668421625,
"tokens_seen": 843644928
},
{
"epoch": 0.26,
"learning_rate": 0.0003759478435305918,
"loss": 2.9369,
"theoretical_loss": 3.7096385506179415,
"tokens_seen": 843710464
},
{
"epoch": 0.26,
"objective/train/docs_used": 485567,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4992148876190186,
"objective/train/theoretical_loss": 3.7096102618861355,
"objective/train/tokens_used": 864236000,
"theoretical_loss": 3.7096102618861355,
"tokens_seen": 843776000
},
{
"epoch": 0.26,
"learning_rate": 0.000375937813440321,
"loss": 2.6463,
"theoretical_loss": 3.7096102618861355,
"tokens_seen": 843776000
},
{
"epoch": 0.26,
"learning_rate": 0.00037592778335005016,
"loss": 2.5875,
"theoretical_loss": 3.709581975966584,
"tokens_seen": 843841536
},
{
"epoch": 0.26,
"learning_rate": 0.00037591775325977934,
"loss": 2.9699,
"theoretical_loss": 3.709553692858789,
"tokens_seen": 843907072
},
{
"epoch": 0.26,
"learning_rate": 0.0003759077231695086,
"loss": 2.8961,
"theoretical_loss": 3.709525412562253,
"tokens_seen": 843972608
},
{
"epoch": 0.26,
"learning_rate": 0.0003758976930792377,
"loss": 3.119,
"theoretical_loss": 3.709497135076478,
"tokens_seen": 844038144
},
{
"epoch": 0.26,
"learning_rate": 0.00037588766298896694,
"loss": 2.8449,
"theoretical_loss": 3.709468860400966,
"tokens_seen": 844103680
},
{
"epoch": 0.26,
"learning_rate": 0.00037587763289869606,
"loss": 2.7891,
"theoretical_loss": 3.7094405885352204,
"tokens_seen": 844169216
},
{
"epoch": 0.26,
"learning_rate": 0.0003758676028084253,
"loss": 2.9344,
"theoretical_loss": 3.7094123194787434,
"tokens_seen": 844234752
},
{
"epoch": 0.26,
"learning_rate": 0.0003758575727181545,
"loss": 2.7693,
"theoretical_loss": 3.709384053231038,
"tokens_seen": 844300288
},
{
"epoch": 0.26,
"learning_rate": 0.00037584754262788366,
"loss": 2.8961,
"theoretical_loss": 3.7093557897916067,
"tokens_seen": 844365824
},
{
"epoch": 0.26,
"learning_rate": 0.00037583751253761284,
"loss": 2.7053,
"theoretical_loss": 3.7093275291599532,
"tokens_seen": 844431360
},
{
"epoch": 0.26,
"learning_rate": 0.0003758274824473421,
"loss": 2.7831,
"theoretical_loss": 3.7092992713355804,
"tokens_seen": 844496896
},
{
"epoch": 0.26,
"learning_rate": 0.0003758174523570712,
"loss": 2.7193,
"theoretical_loss": 3.7092710163179916,
"tokens_seen": 844562432
},
{
"epoch": 0.26,
"learning_rate": 0.00037580742226680044,
"loss": 2.6069,
"theoretical_loss": 3.7092427641066905,
"tokens_seen": 844627968
},
{
"epoch": 0.26,
"learning_rate": 0.00037579739217652957,
"loss": 2.6507,
"theoretical_loss": 3.7092145147011806,
"tokens_seen": 844693504
},
{
"epoch": 0.26,
"learning_rate": 0.0003757873620862588,
"loss": 2.6215,
"theoretical_loss": 3.709186268100965,
"tokens_seen": 844759040
},
{
"epoch": 0.26,
"learning_rate": 0.000375777331995988,
"loss": 2.6668,
"theoretical_loss": 3.7091580243055478,
"tokens_seen": 844824576
},
{
"epoch": 0.26,
"learning_rate": 0.00037576730190571716,
"loss": 2.6749,
"theoretical_loss": 3.709129783314433,
"tokens_seen": 844890112
},
{
"epoch": 0.26,
"learning_rate": 0.00037575727181544635,
"loss": 2.8027,
"theoretical_loss": 3.709101545127125,
"tokens_seen": 844955648
},
{
"epoch": 0.26,
"learning_rate": 0.0003757472417251755,
"loss": 2.7055,
"theoretical_loss": 3.709073309743128,
"tokens_seen": 845021184
},
{
"epoch": 0.26,
"learning_rate": 0.0003757372116349047,
"loss": 2.9118,
"theoretical_loss": 3.709045077161946,
"tokens_seen": 845086720
},
{
"epoch": 0.26,
"learning_rate": 0.00037572718154463394,
"loss": 2.5867,
"theoretical_loss": 3.7090168473830833,
"tokens_seen": 845152256
},
{
"epoch": 0.26,
"learning_rate": 0.00037571715145436307,
"loss": 2.8607,
"theoretical_loss": 3.708988620406045,
"tokens_seen": 845217792
},
{
"epoch": 0.26,
"learning_rate": 0.0003757071213640923,
"loss": 2.7773,
"theoretical_loss": 3.708960396230335,
"tokens_seen": 845283328
},
{
"epoch": 0.26,
"learning_rate": 0.00037569709127382143,
"loss": 2.8783,
"theoretical_loss": 3.7089321748554593,
"tokens_seen": 845348864
},
{
"epoch": 0.26,
"objective/train/docs_used": 486205,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9482920169830322,
"objective/train/theoretical_loss": 3.7089039562809223,
"objective/train/tokens_used": 865874400,
"theoretical_loss": 3.7089039562809223,
"tokens_seen": 845414400
},
{
"epoch": 0.26,
"learning_rate": 0.00037568706118355067,
"loss": 2.8699,
"theoretical_loss": 3.7089039562809223,
"tokens_seen": 845414400
},
{
"epoch": 0.26,
"learning_rate": 0.00037567703109327985,
"loss": 2.5942,
"theoretical_loss": 3.7088757405062283,
"tokens_seen": 845479936
},
{
"epoch": 0.26,
"learning_rate": 0.00037566700100300903,
"loss": 2.8531,
"theoretical_loss": 3.7088475275308834,
"tokens_seen": 845545472
},
{
"epoch": 0.26,
"learning_rate": 0.0003756569709127382,
"loss": 2.7736,
"theoretical_loss": 3.708819317354393,
"tokens_seen": 845611008
},
{
"epoch": 0.26,
"learning_rate": 0.00037564694082246745,
"loss": 2.7642,
"theoretical_loss": 3.708791109976262,
"tokens_seen": 845676544
},
{
"epoch": 0.26,
"learning_rate": 0.0003756369107321966,
"loss": 2.6166,
"theoretical_loss": 3.7087629053959965,
"tokens_seen": 845742080
},
{
"epoch": 0.26,
"learning_rate": 0.0003756268806419258,
"loss": 2.8616,
"theoretical_loss": 3.7087347036131018,
"tokens_seen": 845807616
},
{
"epoch": 0.26,
"learning_rate": 0.00037561685055165494,
"loss": 2.6069,
"theoretical_loss": 3.7087065046270844,
"tokens_seen": 845873152
},
{
"epoch": 0.26,
"learning_rate": 0.00037560682046138417,
"loss": 2.895,
"theoretical_loss": 3.708678308437449,
"tokens_seen": 845938688
},
{
"epoch": 0.26,
"learning_rate": 0.00037559679037111335,
"loss": 2.601,
"theoretical_loss": 3.708650115043703,
"tokens_seen": 846004224
},
{
"epoch": 0.26,
"learning_rate": 0.00037558676028084253,
"loss": 2.7854,
"theoretical_loss": 3.7086219244453518,
"tokens_seen": 846069760
},
{
"epoch": 0.26,
"learning_rate": 0.0003755767301905717,
"loss": 2.7391,
"theoretical_loss": 3.7085937366419017,
"tokens_seen": 846135296
},
{
"epoch": 0.26,
"learning_rate": 0.0003755667001003009,
"loss": 2.9038,
"theoretical_loss": 3.7085655516328604,
"tokens_seen": 846200832
},
{
"epoch": 0.26,
"learning_rate": 0.0003755566700100301,
"loss": 2.8498,
"theoretical_loss": 3.7085373694177326,
"tokens_seen": 846266368
},
{
"epoch": 0.26,
"learning_rate": 0.0003755466399197593,
"loss": 2.8441,
"theoretical_loss": 3.7085091899960263,
"tokens_seen": 846331904
},
{
"epoch": 0.26,
"learning_rate": 0.00037553660982948844,
"loss": 2.777,
"theoretical_loss": 3.7084810133672486,
"tokens_seen": 846397440
},
{
"epoch": 0.26,
"learning_rate": 0.0003755265797392177,
"loss": 2.8675,
"theoretical_loss": 3.708452839530905,
"tokens_seen": 846462976
},
{
"epoch": 0.26,
"learning_rate": 0.0003755165496489468,
"loss": 2.8405,
"theoretical_loss": 3.7084246684865043,
"tokens_seen": 846528512
},
{
"epoch": 0.26,
"learning_rate": 0.00037550651955867604,
"loss": 2.8271,
"theoretical_loss": 3.708396500233552,
"tokens_seen": 846594048
},
{
"epoch": 0.26,
"learning_rate": 0.0003754964894684052,
"loss": 2.7956,
"theoretical_loss": 3.708368334771557,
"tokens_seen": 846659584
},
{
"epoch": 0.26,
"learning_rate": 0.0003754864593781344,
"loss": 2.7386,
"theoretical_loss": 3.7083401721000255,
"tokens_seen": 846725120
},
{
"epoch": 0.26,
"learning_rate": 0.0003754764292878636,
"loss": 2.6701,
"theoretical_loss": 3.708312012218466,
"tokens_seen": 846790656
},
{
"epoch": 0.26,
"learning_rate": 0.0003754663991975928,
"loss": 2.8169,
"theoretical_loss": 3.708283855126386,
"tokens_seen": 846856192
},
{
"epoch": 0.26,
"learning_rate": 0.00037545636910732194,
"loss": 2.9556,
"theoretical_loss": 3.708255700823293,
"tokens_seen": 846921728
},
{
"epoch": 0.26,
"learning_rate": 0.0003754463390170512,
"loss": 2.778,
"theoretical_loss": 3.7082275493086954,
"tokens_seen": 846987264
},
{
"epoch": 0.26,
"objective/train/docs_used": 487507,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2727932929992676,
"objective/train/theoretical_loss": 3.708199400582101,
"objective/train/tokens_used": 867512800,
"theoretical_loss": 3.708199400582101,
"tokens_seen": 847052800
},
{
"epoch": 0.26,
"learning_rate": 0.0003754363089267803,
"loss": 2.7312,
"theoretical_loss": 3.708199400582101,
"tokens_seen": 847052800
},
{
"epoch": 0.26,
"learning_rate": 0.00037542627883650954,
"loss": 2.8009,
"theoretical_loss": 3.708171254643018,
"tokens_seen": 847118336
},
{
"epoch": 0.26,
"learning_rate": 0.0003754162487462387,
"loss": 2.8064,
"theoretical_loss": 3.708143111490955,
"tokens_seen": 847183872
},
{
"epoch": 0.26,
"learning_rate": 0.0003754062186559679,
"loss": 2.5236,
"theoretical_loss": 3.70811497112542,
"tokens_seen": 847249408
},
{
"epoch": 0.26,
"learning_rate": 0.0003753961885656971,
"loss": 2.4188,
"theoretical_loss": 3.708086833545922,
"tokens_seen": 847314944
},
{
"epoch": 0.26,
"learning_rate": 0.00037538615847542626,
"loss": 2.7878,
"theoretical_loss": 3.70805869875197,
"tokens_seen": 847380480
},
{
"epoch": 0.26,
"learning_rate": 0.00037537612838515544,
"loss": 2.7722,
"theoretical_loss": 3.708030566743072,
"tokens_seen": 847446016
},
{
"epoch": 0.26,
"learning_rate": 0.0003753660982948847,
"loss": 2.8349,
"theoretical_loss": 3.7080024375187373,
"tokens_seen": 847511552
},
{
"epoch": 0.26,
"learning_rate": 0.00037535606820461386,
"loss": 2.8763,
"theoretical_loss": 3.7079743110784755,
"tokens_seen": 847577088
},
{
"epoch": 0.26,
"learning_rate": 0.00037534603811434304,
"loss": 2.6812,
"theoretical_loss": 3.707946187421795,
"tokens_seen": 847642624
},
{
"epoch": 0.26,
"learning_rate": 0.0003753360080240723,
"loss": 2.8974,
"theoretical_loss": 3.707918066548206,
"tokens_seen": 847708160
},
{
"epoch": 0.26,
"learning_rate": 0.0003753259779338014,
"loss": 2.8834,
"theoretical_loss": 3.707889948457217,
"tokens_seen": 847773696
},
{
"epoch": 0.26,
"learning_rate": 0.00037531594784353064,
"loss": 2.9305,
"theoretical_loss": 3.7078618331483386,
"tokens_seen": 847839232
},
{
"epoch": 0.26,
"learning_rate": 0.00037530591775325977,
"loss": 2.7361,
"theoretical_loss": 3.7078337206210796,
"tokens_seen": 847904768
},
{
"epoch": 0.26,
"learning_rate": 0.000375295887662989,
"loss": 2.6933,
"theoretical_loss": 3.70780561087495,
"tokens_seen": 847970304
},
{
"epoch": 0.26,
"learning_rate": 0.0003752858575727182,
"loss": 2.8247,
"theoretical_loss": 3.7077775039094605,
"tokens_seen": 848035840
},
{
"epoch": 0.26,
"learning_rate": 0.00037527582748244736,
"loss": 2.9398,
"theoretical_loss": 3.70774939972412,
"tokens_seen": 848101376
},
{
"epoch": 0.26,
"learning_rate": 0.00037526579739217655,
"loss": 2.7715,
"theoretical_loss": 3.7077212983184404,
"tokens_seen": 848166912
},
{
"epoch": 0.26,
"learning_rate": 0.0003752557673019057,
"loss": 2.8193,
"theoretical_loss": 3.70769319969193,
"tokens_seen": 848232448
},
{
"epoch": 0.26,
"learning_rate": 0.0003752457372116349,
"loss": 2.6643,
"theoretical_loss": 3.707665103844101,
"tokens_seen": 848297984
},
{
"epoch": 0.26,
"learning_rate": 0.00037523570712136414,
"loss": 2.7414,
"theoretical_loss": 3.7076370107744627,
"tokens_seen": 848363520
},
{
"epoch": 0.26,
"learning_rate": 0.00037522567703109327,
"loss": 2.6838,
"theoretical_loss": 3.7076089204825267,
"tokens_seen": 848429056
},
{
"epoch": 0.26,
"learning_rate": 0.0003752156469408225,
"loss": 2.9158,
"theoretical_loss": 3.7075808329678033,
"tokens_seen": 848494592
},
{
"epoch": 0.26,
"learning_rate": 0.00037520561685055163,
"loss": 2.7536,
"theoretical_loss": 3.707552748229803,
"tokens_seen": 848560128
},
{
"epoch": 0.26,
"learning_rate": 0.00037519558676028087,
"loss": 2.7188,
"theoretical_loss": 3.7075246662680383,
"tokens_seen": 848625664
},
{
"epoch": 0.26,
"objective/train/docs_used": 487917,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6695802211761475,
"objective/train/theoretical_loss": 3.7074965870820193,
"objective/train/tokens_used": 869151200,
"theoretical_loss": 3.7074965870820193,
"tokens_seen": 848691200
},
{
"epoch": 0.26,
"learning_rate": 0.00037518555667001005,
"loss": 2.8457,
"theoretical_loss": 3.7074965870820193,
"tokens_seen": 848691200
},
{
"epoch": 0.26,
"learning_rate": 0.00037517552657973923,
"loss": 2.8154,
"theoretical_loss": 3.7074685106712577,
"tokens_seen": 848756736
},
{
"epoch": 0.26,
"learning_rate": 0.0003751654964894684,
"loss": 2.8332,
"theoretical_loss": 3.7074404370352645,
"tokens_seen": 848822272
},
{
"epoch": 0.26,
"learning_rate": 0.00037515546639919765,
"loss": 2.7252,
"theoretical_loss": 3.707412366173552,
"tokens_seen": 848887808
},
{
"epoch": 0.26,
"learning_rate": 0.0003751454363089268,
"loss": 2.9011,
"theoretical_loss": 3.707384298085631,
"tokens_seen": 848953344
},
{
"epoch": 0.26,
"learning_rate": 0.000375135406218656,
"loss": 2.9282,
"theoretical_loss": 3.7073562327710143,
"tokens_seen": 849018880
},
{
"epoch": 0.26,
"learning_rate": 0.00037512537612838514,
"loss": 2.8749,
"theoretical_loss": 3.7073281702292133,
"tokens_seen": 849084416
},
{
"epoch": 0.26,
"learning_rate": 0.00037511534603811437,
"loss": 2.6977,
"theoretical_loss": 3.70730011045974,
"tokens_seen": 849149952
},
{
"epoch": 0.26,
"learning_rate": 0.00037510531594784355,
"loss": 2.8463,
"theoretical_loss": 3.707272053462107,
"tokens_seen": 849215488
},
{
"epoch": 0.26,
"learning_rate": 0.00037509528585757273,
"loss": 2.8365,
"theoretical_loss": 3.7072439992358257,
"tokens_seen": 849281024
},
{
"epoch": 0.26,
"learning_rate": 0.0003750852557673019,
"loss": 2.8203,
"theoretical_loss": 3.7072159477804094,
"tokens_seen": 849346560
},
{
"epoch": 0.26,
"learning_rate": 0.0003750752256770311,
"loss": 3.1436,
"theoretical_loss": 3.707187899095371,
"tokens_seen": 849412096
},
{
"epoch": 0.26,
"learning_rate": 0.0003750651955867603,
"loss": 2.5124,
"theoretical_loss": 3.707159853180222,
"tokens_seen": 849477632
},
{
"epoch": 0.26,
"learning_rate": 0.0003750551654964895,
"loss": 2.7687,
"theoretical_loss": 3.7071318100344754,
"tokens_seen": 849543168
},
{
"epoch": 0.26,
"learning_rate": 0.00037504513540621864,
"loss": 2.8021,
"theoretical_loss": 3.707103769657645,
"tokens_seen": 849608704
},
{
"epoch": 0.26,
"learning_rate": 0.0003750351053159479,
"loss": 2.6576,
"theoretical_loss": 3.707075732049244,
"tokens_seen": 849674240
},
{
"epoch": 0.26,
"learning_rate": 0.000375025075225677,
"loss": 2.9572,
"theoretical_loss": 3.7070476972087842,
"tokens_seen": 849739776
},
{
"epoch": 0.26,
"learning_rate": 0.00037501504513540624,
"loss": 2.7768,
"theoretical_loss": 3.7070196651357796,
"tokens_seen": 849805312
},
{
"epoch": 0.26,
"learning_rate": 0.0003750050150451354,
"loss": 2.9411,
"theoretical_loss": 3.706991635829744,
"tokens_seen": 849870848
},
{
"epoch": 0.26,
"learning_rate": 0.0003749949849548646,
"loss": 2.7345,
"theoretical_loss": 3.70696360929019,
"tokens_seen": 849936384
},
{
"epoch": 0.26,
"learning_rate": 0.0003749849548645938,
"loss": 2.7635,
"theoretical_loss": 3.706935585516632,
"tokens_seen": 850001920
},
{
"epoch": 0.26,
"learning_rate": 0.000374974924774323,
"loss": 2.6752,
"theoretical_loss": 3.7069075645085845,
"tokens_seen": 850067456
},
{
"epoch": 0.26,
"learning_rate": 0.00037496489468405214,
"loss": 2.6164,
"theoretical_loss": 3.7068795462655597,
"tokens_seen": 850132992
},
{
"epoch": 0.26,
"learning_rate": 0.0003749548645937814,
"loss": 2.8422,
"theoretical_loss": 3.7068515307870733,
"tokens_seen": 850198528
},
{
"epoch": 0.26,
"learning_rate": 0.0003749448345035105,
"loss": 2.9901,
"theoretical_loss": 3.706823518072638,
"tokens_seen": 850264064
},
{
"epoch": 0.26,
"objective/train/docs_used": 489068,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7429158687591553,
"objective/train/theoretical_loss": 3.706795508121769,
"objective/train/tokens_used": 870789600,
"theoretical_loss": 3.706795508121769,
"tokens_seen": 850329600
},
{
"epoch": 0.26,
"learning_rate": 0.00037493480441323974,
"loss": 2.677,
"theoretical_loss": 3.706795508121769,
"tokens_seen": 850329600
},
{
"epoch": 0.26,
"learning_rate": 0.0003749247743229689,
"loss": 2.6939,
"theoretical_loss": 3.7067675009339807,
"tokens_seen": 850395136
},
{
"epoch": 0.26,
"learning_rate": 0.0003749147442326981,
"loss": 2.691,
"theoretical_loss": 3.7067394965087868,
"tokens_seen": 850460672
},
{
"epoch": 0.26,
"learning_rate": 0.0003749047141424273,
"loss": 2.6482,
"theoretical_loss": 3.7067114948457025,
"tokens_seen": 850526208
},
{
"epoch": 0.26,
"learning_rate": 0.00037489468405215646,
"loss": 3.0822,
"theoretical_loss": 3.706683495944243,
"tokens_seen": 850591744
},
{
"epoch": 0.26,
"learning_rate": 0.00037488465396188564,
"loss": 2.6257,
"theoretical_loss": 3.7066554998039227,
"tokens_seen": 850657280
},
{
"epoch": 0.26,
"learning_rate": 0.0003748746238716149,
"loss": 2.8582,
"theoretical_loss": 3.7066275064242564,
"tokens_seen": 850722816
},
{
"epoch": 0.26,
"learning_rate": 0.000374864593781344,
"loss": 2.5449,
"theoretical_loss": 3.70659951580476,
"tokens_seen": 850788352
},
{
"epoch": 0.26,
"learning_rate": 0.00037485456369107324,
"loss": 2.9432,
"theoretical_loss": 3.706571527944948,
"tokens_seen": 850853888
},
{
"epoch": 0.26,
"learning_rate": 0.00037484453360080237,
"loss": 2.8962,
"theoretical_loss": 3.7065435428443365,
"tokens_seen": 850919424
},
{
"epoch": 0.26,
"learning_rate": 0.0003748345035105316,
"loss": 3.0175,
"theoretical_loss": 3.70651556050244,
"tokens_seen": 850984960
},
{
"epoch": 0.26,
"learning_rate": 0.0003748244734202608,
"loss": 2.7384,
"theoretical_loss": 3.706487580918775,
"tokens_seen": 851050496
},
{
"epoch": 0.26,
"learning_rate": 0.00037481444332998997,
"loss": 2.6731,
"theoretical_loss": 3.7064596040928572,
"tokens_seen": 851116032
},
{
"epoch": 0.26,
"learning_rate": 0.00037480441323971915,
"loss": 2.9541,
"theoretical_loss": 3.706431630024202,
"tokens_seen": 851181568
},
{
"epoch": 0.26,
"learning_rate": 0.0003747943831494484,
"loss": 2.7875,
"theoretical_loss": 3.7064036587123255,
"tokens_seen": 851247104
},
{
"epoch": 0.26,
"learning_rate": 0.0003747843530591775,
"loss": 2.7249,
"theoretical_loss": 3.7063756901567437,
"tokens_seen": 851312640
},
{
"epoch": 0.26,
"learning_rate": 0.00037477432296890675,
"loss": 2.9977,
"theoretical_loss": 3.7063477243569736,
"tokens_seen": 851378176
},
{
"epoch": 0.26,
"learning_rate": 0.00037476429287863587,
"loss": 2.654,
"theoretical_loss": 3.706319761312531,
"tokens_seen": 851443712
},
{
"epoch": 0.26,
"learning_rate": 0.0003747542627883651,
"loss": 2.764,
"theoretical_loss": 3.7062918010229327,
"tokens_seen": 851509248
},
{
"epoch": 0.26,
"learning_rate": 0.0003747442326980943,
"loss": 2.7853,
"theoretical_loss": 3.7062638434876947,
"tokens_seen": 851574784
},
{
"epoch": 0.26,
"learning_rate": 0.00037473420260782347,
"loss": 2.7226,
"theoretical_loss": 3.706235888706334,
"tokens_seen": 851640320
},
{
"epoch": 0.26,
"learning_rate": 0.00037472417251755265,
"loss": 2.7892,
"theoretical_loss": 3.706207936678368,
"tokens_seen": 851705856
},
{
"epoch": 0.26,
"learning_rate": 0.00037471414242728183,
"loss": 2.9169,
"theoretical_loss": 3.7061799874033126,
"tokens_seen": 851771392
},
{
"epoch": 0.26,
"learning_rate": 0.000374704112337011,
"loss": 2.7041,
"theoretical_loss": 3.706152040880686,
"tokens_seen": 851836928
},
{
"epoch": 0.26,
"learning_rate": 0.00037469408224674025,
"loss": 2.5983,
"theoretical_loss": 3.7061240971100045,
"tokens_seen": 851902464
},
{
"epoch": 0.26,
"objective/train/docs_used": 489617,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0048906803131104,
"objective/train/theoretical_loss": 3.7060961560907857,
"objective/train/tokens_used": 872428000,
"theoretical_loss": 3.7060961560907857,
"tokens_seen": 851968000
},
{
"epoch": 0.26,
"learning_rate": 0.0003746840521564694,
"loss": 2.9298,
"theoretical_loss": 3.7060961560907857,
"tokens_seen": 851968000
},
{
"epoch": 0.26,
"learning_rate": 0.0003746740220661986,
"loss": 2.6606,
"theoretical_loss": 3.7060682178225477,
"tokens_seen": 852033536
},
{
"epoch": 0.26,
"learning_rate": 0.00037466399197592774,
"loss": 2.794,
"theoretical_loss": 3.7060402823048073,
"tokens_seen": 852099072
},
{
"epoch": 0.26,
"learning_rate": 0.000374653961885657,
"loss": 2.7546,
"theoretical_loss": 3.7060123495370827,
"tokens_seen": 852164608
},
{
"epoch": 0.26,
"learning_rate": 0.00037464393179538615,
"loss": 2.8952,
"theoretical_loss": 3.705984419518891,
"tokens_seen": 852230144
},
{
"epoch": 0.26,
"learning_rate": 0.00037463390170511534,
"loss": 2.8843,
"theoretical_loss": 3.7059564922497517,
"tokens_seen": 852295680
},
{
"epoch": 0.26,
"learning_rate": 0.00037462387161484457,
"loss": 2.7003,
"theoretical_loss": 3.7059285677291807,
"tokens_seen": 852361216
},
{
"epoch": 0.26,
"learning_rate": 0.00037461384152457375,
"loss": 2.826,
"theoretical_loss": 3.705900645956698,
"tokens_seen": 852426752
},
{
"epoch": 0.26,
"learning_rate": 0.00037460381143430293,
"loss": 2.7706,
"theoretical_loss": 3.705872726931821,
"tokens_seen": 852492288
},
{
"epoch": 0.26,
"learning_rate": 0.0003745937813440321,
"loss": 2.845,
"theoretical_loss": 3.7058448106540682,
"tokens_seen": 852557824
},
{
"epoch": 0.26,
"learning_rate": 0.0003745837512537613,
"loss": 2.7092,
"theoretical_loss": 3.7058168971229586,
"tokens_seen": 852623360
},
{
"epoch": 0.26,
"learning_rate": 0.0003745737211634905,
"loss": 2.7973,
"theoretical_loss": 3.7057889863380105,
"tokens_seen": 852688896
},
{
"epoch": 0.26,
"learning_rate": 0.0003745636910732197,
"loss": 2.8317,
"theoretical_loss": 3.7057610782987433,
"tokens_seen": 852754432
},
{
"epoch": 0.26,
"learning_rate": 0.00037455366098294884,
"loss": 2.5929,
"theoretical_loss": 3.7057331730046745,
"tokens_seen": 852819968
},
{
"epoch": 0.26,
"learning_rate": 0.0003745436308926781,
"loss": 2.6961,
"theoretical_loss": 3.705705270455325,
"tokens_seen": 852885504
},
{
"epoch": 0.26,
"learning_rate": 0.0003745336008024072,
"loss": 2.7982,
"theoretical_loss": 3.7056773706502124,
"tokens_seen": 852951040
},
{
"epoch": 0.26,
"learning_rate": 0.00037452357071213644,
"loss": 2.7643,
"theoretical_loss": 3.7056494735888563,
"tokens_seen": 853016576
},
{
"epoch": 0.26,
"learning_rate": 0.0003745135406218656,
"loss": 2.8991,
"theoretical_loss": 3.705621579270777,
"tokens_seen": 853082112
},
{
"epoch": 0.26,
"learning_rate": 0.0003745035105315948,
"loss": 2.9144,
"theoretical_loss": 3.705593687695493,
"tokens_seen": 853147648
},
{
"epoch": 0.26,
"learning_rate": 0.000374493480441324,
"loss": 2.8401,
"theoretical_loss": 3.7055657988625246,
"tokens_seen": 853213184
},
{
"epoch": 0.26,
"learning_rate": 0.0003744834503510532,
"loss": 2.7668,
"theoretical_loss": 3.705537912771391,
"tokens_seen": 853278720
},
{
"epoch": 0.26,
"learning_rate": 0.00037447342026078234,
"loss": 2.8167,
"theoretical_loss": 3.7055100294216126,
"tokens_seen": 853344256
},
{
"epoch": 0.26,
"learning_rate": 0.0003744633901705116,
"loss": 2.5555,
"theoretical_loss": 3.7054821488127097,
"tokens_seen": 853409792
},
{
"epoch": 0.26,
"learning_rate": 0.0003744533600802407,
"loss": 2.4235,
"theoretical_loss": 3.705454270944201,
"tokens_seen": 853475328
},
{
"epoch": 0.26,
"learning_rate": 0.00037444332998996994,
"loss": 2.9843,
"theoretical_loss": 3.7054263958156084,
"tokens_seen": 853540864
},
{
"epoch": 0.26,
"objective/train/docs_used": 491051,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.861933469772339,
"objective/train/theoretical_loss": 3.7053985234264513,
"objective/train/tokens_used": 874066400,
"theoretical_loss": 3.7053985234264513,
"tokens_seen": 853606400
},
{
"epoch": 0.26,
"learning_rate": 0.0003744332998996991,
"loss": 2.6979,
"theoretical_loss": 3.7053985234264513,
"tokens_seen": 853606400
},
{
"epoch": 0.26,
"learning_rate": 0.0003744232698094283,
"loss": 3.0682,
"theoretical_loss": 3.7053706537762503,
"tokens_seen": 853671936
},
{
"epoch": 0.26,
"learning_rate": 0.0003744132397191575,
"loss": 2.8266,
"theoretical_loss": 3.7053427868645263,
"tokens_seen": 853737472
},
{
"epoch": 0.26,
"learning_rate": 0.00037440320962888666,
"loss": 2.6731,
"theoretical_loss": 3.7053149226908,
"tokens_seen": 853803008
},
{
"epoch": 0.26,
"learning_rate": 0.00037439317953861585,
"loss": 2.8001,
"theoretical_loss": 3.7052870612545927,
"tokens_seen": 853868544
},
{
"epoch": 0.26,
"learning_rate": 0.0003743831494483451,
"loss": 2.8354,
"theoretical_loss": 3.705259202555424,
"tokens_seen": 853934080
},
{
"epoch": 0.26,
"learning_rate": 0.0003743731193580742,
"loss": 2.9043,
"theoretical_loss": 3.705231346592816,
"tokens_seen": 853999616
},
{
"epoch": 0.26,
"learning_rate": 0.00037436308926780344,
"loss": 2.6877,
"theoretical_loss": 3.70520349336629,
"tokens_seen": 854065152
},
{
"epoch": 0.26,
"learning_rate": 0.00037435305917753257,
"loss": 2.7725,
"theoretical_loss": 3.705175642875367,
"tokens_seen": 854130688
},
{
"epoch": 0.26,
"learning_rate": 0.0003743430290872618,
"loss": 2.5796,
"theoretical_loss": 3.7051477951195677,
"tokens_seen": 854196224
},
{
"epoch": 0.26,
"learning_rate": 0.000374332998996991,
"loss": 2.8358,
"theoretical_loss": 3.7051199500984158,
"tokens_seen": 854261760
},
{
"epoch": 0.26,
"learning_rate": 0.00037432296890672017,
"loss": 2.8851,
"theoretical_loss": 3.705092107811431,
"tokens_seen": 854327296
},
{
"epoch": 0.26,
"learning_rate": 0.00037431293881644935,
"loss": 2.7564,
"theoretical_loss": 3.705064268258136,
"tokens_seen": 854392832
},
{
"epoch": 0.26,
"learning_rate": 0.0003743029087261786,
"loss": 2.9807,
"theoretical_loss": 3.705036431438052,
"tokens_seen": 854458368
},
{
"epoch": 0.26,
"learning_rate": 0.0003742928786359077,
"loss": 2.8497,
"theoretical_loss": 3.7050085973507025,
"tokens_seen": 854523904
},
{
"epoch": 0.26,
"learning_rate": 0.00037428284854563695,
"loss": 2.9148,
"theoretical_loss": 3.704980765995608,
"tokens_seen": 854589440
},
{
"epoch": 0.26,
"learning_rate": 0.00037427281845536607,
"loss": 3.006,
"theoretical_loss": 3.7049529373722923,
"tokens_seen": 854654976
},
{
"epoch": 0.26,
"learning_rate": 0.0003742627883650953,
"loss": 2.8407,
"theoretical_loss": 3.704925111480277,
"tokens_seen": 854720512
},
{
"epoch": 0.26,
"learning_rate": 0.0003742527582748245,
"loss": 2.9128,
"theoretical_loss": 3.704897288319084,
"tokens_seen": 854786048
},
{
"epoch": 0.26,
"learning_rate": 0.00037424272818455367,
"loss": 2.7484,
"theoretical_loss": 3.704869467888237,
"tokens_seen": 854851584
},
{
"epoch": 0.26,
"learning_rate": 0.00037423269809428285,
"loss": 2.97,
"theoretical_loss": 3.7048416501872588,
"tokens_seen": 854917120
},
{
"epoch": 0.26,
"learning_rate": 0.00037422266800401203,
"loss": 2.7888,
"theoretical_loss": 3.7048138352156723,
"tokens_seen": 854982656
},
{
"epoch": 0.26,
"learning_rate": 0.0003742126379137412,
"loss": 2.9088,
"theoretical_loss": 3.7047860229729994,
"tokens_seen": 855048192
},
{
"epoch": 0.26,
"learning_rate": 0.00037420260782347045,
"loss": 2.9126,
"theoretical_loss": 3.7047582134587644,
"tokens_seen": 855113728
},
{
"epoch": 0.26,
"learning_rate": 0.0003741925777331996,
"loss": 2.7399,
"theoretical_loss": 3.70473040667249,
"tokens_seen": 855179264
},
{
"epoch": 0.26,
"objective/train/docs_used": 491675,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.609562635421753,
"objective/train/theoretical_loss": 3.7047026026137,
"objective/train/tokens_used": 875704800,
"theoretical_loss": 3.7047026026137,
"tokens_seen": 855244800
},
{
"epoch": 0.26,
"learning_rate": 0.0003741825476429288,
"loss": 3.0614,
"theoretical_loss": 3.7047026026137,
"tokens_seen": 855244800
},
{
"epoch": 0.26,
"learning_rate": 0.00037417251755265794,
"loss": 2.724,
"theoretical_loss": 3.7046748012819175,
"tokens_seen": 855310336
},
{
"epoch": 0.26,
"learning_rate": 0.0003741624874623872,
"loss": 2.8514,
"theoretical_loss": 3.7046470026766665,
"tokens_seen": 855375872
},
{
"epoch": 0.26,
"learning_rate": 0.00037415245737211635,
"loss": 2.8871,
"theoretical_loss": 3.70461920679747,
"tokens_seen": 855441408
},
{
"epoch": 0.26,
"learning_rate": 0.00037414242728184554,
"loss": 2.9418,
"theoretical_loss": 3.704591413643853,
"tokens_seen": 855506944
},
{
"epoch": 0.26,
"learning_rate": 0.0003741323971915747,
"loss": 3.0173,
"theoretical_loss": 3.7045636232153383,
"tokens_seen": 855572480
},
{
"epoch": 0.26,
"learning_rate": 0.00037412236710130395,
"loss": 2.6986,
"theoretical_loss": 3.704535835511451,
"tokens_seen": 855638016
},
{
"epoch": 0.26,
"learning_rate": 0.0003741123370110331,
"loss": 2.619,
"theoretical_loss": 3.7045080505317145,
"tokens_seen": 855703552
},
{
"epoch": 0.26,
"learning_rate": 0.0003741023069207623,
"loss": 2.7184,
"theoretical_loss": 3.7044802682756535,
"tokens_seen": 855769088
},
{
"epoch": 0.26,
"learning_rate": 0.00037409227683049144,
"loss": 2.8852,
"theoretical_loss": 3.7044524887427928,
"tokens_seen": 855834624
},
{
"epoch": 0.26,
"learning_rate": 0.0003740822467402207,
"loss": 2.7811,
"theoretical_loss": 3.7044247119326568,
"tokens_seen": 855900160
},
{
"epoch": 0.26,
"learning_rate": 0.00037407221664994986,
"loss": 2.6483,
"theoretical_loss": 3.7043969378447694,
"tokens_seen": 855965696
},
{
"epoch": 0.26,
"learning_rate": 0.00037406218655967904,
"loss": 2.5415,
"theoretical_loss": 3.7043691664786564,
"tokens_seen": 856031232
},
{
"epoch": 0.26,
"learning_rate": 0.0003740521564694082,
"loss": 2.9868,
"theoretical_loss": 3.7043413978338426,
"tokens_seen": 856096768
},
{
"epoch": 0.26,
"learning_rate": 0.0003740421263791374,
"loss": 2.7854,
"theoretical_loss": 3.7043136319098524,
"tokens_seen": 856162304
},
{
"epoch": 0.26,
"learning_rate": 0.0003740320962888666,
"loss": 3.0667,
"theoretical_loss": 3.7042858687062115,
"tokens_seen": 856227840
},
{
"epoch": 0.26,
"learning_rate": 0.0003740220661985958,
"loss": 2.764,
"theoretical_loss": 3.7042581082224446,
"tokens_seen": 856293376
},
{
"epoch": 0.26,
"learning_rate": 0.00037401203610832494,
"loss": 2.7086,
"theoretical_loss": 3.7042303504580785,
"tokens_seen": 856358912
},
{
"epoch": 0.26,
"learning_rate": 0.0003740020060180542,
"loss": 2.9189,
"theoretical_loss": 3.704202595412637,
"tokens_seen": 856424448
},
{
"epoch": 0.26,
"learning_rate": 0.00037399197592778336,
"loss": 2.7851,
"theoretical_loss": 3.7041748430856467,
"tokens_seen": 856489984
},
{
"epoch": 0.26,
"learning_rate": 0.00037398194583751254,
"loss": 2.711,
"theoretical_loss": 3.7041470934766334,
"tokens_seen": 856555520
},
{
"epoch": 0.26,
"learning_rate": 0.0003739719157472417,
"loss": 3.1173,
"theoretical_loss": 3.7041193465851228,
"tokens_seen": 856621056
},
{
"epoch": 0.26,
"learning_rate": 0.0003739618856569709,
"loss": 2.6248,
"theoretical_loss": 3.704091602410641,
"tokens_seen": 856686592
},
{
"epoch": 0.26,
"learning_rate": 0.0003739518555667001,
"loss": 2.7416,
"theoretical_loss": 3.704063860952714,
"tokens_seen": 856752128
},
{
"epoch": 0.26,
"learning_rate": 0.0003739418254764293,
"loss": 2.9514,
"theoretical_loss": 3.704036122210868,
"tokens_seen": 856817664
},
{
"epoch": 0.26,
"objective/train/docs_used": 492132,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.999392032623291,
"objective/train/theoretical_loss": 3.704008386184629,
"objective/train/tokens_used": 877343200,
"theoretical_loss": 3.704008386184629,
"tokens_seen": 856883200
},
{
"epoch": 0.26,
"learning_rate": 0.00037393179538615845,
"loss": 2.8936,
"theoretical_loss": 3.704008386184629,
"tokens_seen": 856883200
},
{
"epoch": 0.26,
"learning_rate": 0.0003739217652958877,
"loss": 2.915,
"theoretical_loss": 3.7039806528735246,
"tokens_seen": 856948736
},
{
"epoch": 0.26,
"learning_rate": 0.0003739117352056168,
"loss": 2.7055,
"theoretical_loss": 3.7039529222770806,
"tokens_seen": 857014272
},
{
"epoch": 0.26,
"learning_rate": 0.00037390170511534605,
"loss": 2.8961,
"theoretical_loss": 3.7039251943948237,
"tokens_seen": 857079808
},
{
"epoch": 0.26,
"learning_rate": 0.0003738916750250752,
"loss": 2.8075,
"theoretical_loss": 3.703897469226281,
"tokens_seen": 857145344
},
{
"epoch": 0.26,
"learning_rate": 0.0003738816449348044,
"loss": 2.8155,
"theoretical_loss": 3.7038697467709794,
"tokens_seen": 857210880
},
{
"epoch": 0.26,
"learning_rate": 0.00037387161484453364,
"loss": 2.9075,
"theoretical_loss": 3.703842027028445,
"tokens_seen": 857276416
},
{
"epoch": 0.26,
"learning_rate": 0.00037386158475426277,
"loss": 2.8364,
"theoretical_loss": 3.703814309998207,
"tokens_seen": 857341952
},
{
"epoch": 0.26,
"learning_rate": 0.000373851554663992,
"loss": 2.8838,
"theoretical_loss": 3.703786595679791,
"tokens_seen": 857407488
},
{
"epoch": 0.26,
"learning_rate": 0.0003738415245737212,
"loss": 2.6412,
"theoretical_loss": 3.703758884072725,
"tokens_seen": 857473024
},
{
"epoch": 0.26,
"learning_rate": 0.00037383149448345037,
"loss": 2.9262,
"theoretical_loss": 3.7037311751765367,
"tokens_seen": 857538560
},
{
"epoch": 0.26,
"learning_rate": 0.00037382146439317955,
"loss": 2.82,
"theoretical_loss": 3.7037034689907538,
"tokens_seen": 857604096
},
{
"epoch": 0.26,
"learning_rate": 0.0003738114343029088,
"loss": 2.8519,
"theoretical_loss": 3.7036757655149044,
"tokens_seen": 857669632
},
{
"epoch": 0.26,
"learning_rate": 0.0003738014042126379,
"loss": 2.9878,
"theoretical_loss": 3.7036480647485153,
"tokens_seen": 857735168
},
{
"epoch": 0.26,
"learning_rate": 0.00037379137412236715,
"loss": 2.8849,
"theoretical_loss": 3.7036203666911147,
"tokens_seen": 857800704
},
{
"epoch": 0.26,
"learning_rate": 0.00037378134403209627,
"loss": 2.8394,
"theoretical_loss": 3.703592671342232,
"tokens_seen": 857866240
},
{
"epoch": 0.26,
"learning_rate": 0.0003737713139418255,
"loss": 2.9581,
"theoretical_loss": 3.703564978701394,
"tokens_seen": 857931776
},
{
"epoch": 0.26,
"learning_rate": 0.0003737612838515547,
"loss": 2.7661,
"theoretical_loss": 3.70353728876813,
"tokens_seen": 857997312
},
{
"epoch": 0.26,
"learning_rate": 0.00037375125376128387,
"loss": 2.9811,
"theoretical_loss": 3.703509601541968,
"tokens_seen": 858062848
},
{
"epoch": 0.26,
"learning_rate": 0.00037374122367101305,
"loss": 2.7672,
"theoretical_loss": 3.703481917022437,
"tokens_seen": 858128384
},
{
"epoch": 0.26,
"learning_rate": 0.00037373119358074223,
"loss": 2.8545,
"theoretical_loss": 3.703454235209066,
"tokens_seen": 858193920
},
{
"epoch": 0.26,
"learning_rate": 0.0003737211634904714,
"loss": 3.1598,
"theoretical_loss": 3.7034265561013826,
"tokens_seen": 858259456
},
{
"epoch": 0.26,
"learning_rate": 0.00037371113340020065,
"loss": 2.7928,
"theoretical_loss": 3.7033988796989172,
"tokens_seen": 858324992
},
{
"epoch": 0.26,
"learning_rate": 0.0003737011033099298,
"loss": 2.6854,
"theoretical_loss": 3.703371206001198,
"tokens_seen": 858390528
},
{
"epoch": 0.26,
"learning_rate": 0.000373691073219659,
"loss": 2.9603,
"theoretical_loss": 3.703343535007754,
"tokens_seen": 858456064
},
{
"epoch": 0.26,
"objective/train/docs_used": 493247,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8336102962493896,
"objective/train/theoretical_loss": 3.7033158667181154,
"objective/train/tokens_used": 878981600,
"theoretical_loss": 3.7033158667181154,
"tokens_seen": 858521600
},
{
"epoch": 0.26,
"learning_rate": 0.00037368104312938814,
"loss": 2.8954,
"theoretical_loss": 3.7033158667181154,
"tokens_seen": 858521600
},
{
"epoch": 0.26,
"learning_rate": 0.0003736710130391174,
"loss": 2.8266,
"theoretical_loss": 3.7032882011318113,
"tokens_seen": 858587136
},
{
"epoch": 0.26,
"learning_rate": 0.00037366098294884655,
"loss": 2.8543,
"theoretical_loss": 3.703260538248371,
"tokens_seen": 858652672
},
{
"epoch": 0.26,
"learning_rate": 0.00037365095285857574,
"loss": 2.8304,
"theoretical_loss": 3.7032328780673245,
"tokens_seen": 858718208
},
{
"epoch": 0.26,
"learning_rate": 0.0003736409227683049,
"loss": 3.0557,
"theoretical_loss": 3.7032052205882007,
"tokens_seen": 858783744
},
{
"epoch": 0.26,
"learning_rate": 0.00037363089267803415,
"loss": 2.8675,
"theoretical_loss": 3.703177565810531,
"tokens_seen": 858849280
},
{
"epoch": 0.26,
"learning_rate": 0.0003736208625877633,
"loss": 2.9398,
"theoretical_loss": 3.7031499137338444,
"tokens_seen": 858914816
},
{
"epoch": 0.26,
"learning_rate": 0.0003736108324974925,
"loss": 2.8792,
"theoretical_loss": 3.7031222643576713,
"tokens_seen": 858980352
},
{
"epoch": 0.26,
"learning_rate": 0.00037360080240722164,
"loss": 2.9061,
"theoretical_loss": 3.7030946176815425,
"tokens_seen": 859045888
},
{
"epoch": 0.26,
"learning_rate": 0.0003735907723169509,
"loss": 2.8015,
"theoretical_loss": 3.703066973704987,
"tokens_seen": 859111424
},
{
"epoch": 0.26,
"learning_rate": 0.00037358074222668006,
"loss": 2.6859,
"theoretical_loss": 3.7030393324275366,
"tokens_seen": 859176960
},
{
"epoch": 0.26,
"learning_rate": 0.00037357071213640924,
"loss": 2.7758,
"theoretical_loss": 3.7030116938487208,
"tokens_seen": 859242496
},
{
"epoch": 0.26,
"learning_rate": 0.0003735606820461384,
"loss": 2.7745,
"theoretical_loss": 3.702984057968072,
"tokens_seen": 859308032
},
{
"epoch": 0.26,
"learning_rate": 0.0003735506519558676,
"loss": 2.8328,
"theoretical_loss": 3.702956424785119,
"tokens_seen": 859373568
},
{
"epoch": 0.26,
"learning_rate": 0.0003735406218655968,
"loss": 2.6465,
"theoretical_loss": 3.7029287942993943,
"tokens_seen": 859439104
},
{
"epoch": 0.26,
"learning_rate": 0.000373530591775326,
"loss": 2.9517,
"theoretical_loss": 3.7029011665104283,
"tokens_seen": 859504640
},
{
"epoch": 0.26,
"learning_rate": 0.00037352056168505514,
"loss": 3.0575,
"theoretical_loss": 3.7028735414177527,
"tokens_seen": 859570176
},
{
"epoch": 0.26,
"learning_rate": 0.0003735105315947844,
"loss": 2.9161,
"theoretical_loss": 3.7028459190208984,
"tokens_seen": 859635712
},
{
"epoch": 0.26,
"learning_rate": 0.00037350050150451356,
"loss": 2.7755,
"theoretical_loss": 3.7028182993193965,
"tokens_seen": 859701248
},
{
"epoch": 0.26,
"learning_rate": 0.00037349047141424274,
"loss": 2.6293,
"theoretical_loss": 3.70279068231278,
"tokens_seen": 859766784
},
{
"epoch": 0.26,
"learning_rate": 0.0003734804413239719,
"loss": 3.12,
"theoretical_loss": 3.7027630680005785,
"tokens_seen": 859832320
},
{
"epoch": 0.26,
"learning_rate": 0.0003734704112337011,
"loss": 2.9265,
"theoretical_loss": 3.702735456382325,
"tokens_seen": 859897856
},
{
"epoch": 0.26,
"learning_rate": 0.0003734603811434303,
"loss": 2.9915,
"theoretical_loss": 3.7027078474575514,
"tokens_seen": 859963392
},
{
"epoch": 0.26,
"learning_rate": 0.0003734503510531595,
"loss": 2.9319,
"theoretical_loss": 3.7026802412257895,
"tokens_seen": 860028928
},
{
"epoch": 0.26,
"learning_rate": 0.00037344032096288865,
"loss": 2.7569,
"theoretical_loss": 3.7026526376865716,
"tokens_seen": 860094464
},
{
"epoch": 0.26,
"objective/train/docs_used": 493898,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8726143836975098,
"objective/train/theoretical_loss": 3.7026250368394296,
"objective/train/tokens_used": 880620000,
"theoretical_loss": 3.7026250368394296,
"tokens_seen": 860160000
},
{
"epoch": 0.26,
"learning_rate": 0.0003734302908726179,
"loss": 2.6986,
"theoretical_loss": 3.7026250368394296,
"tokens_seen": 860160000
},
{
"epoch": 0.26,
"learning_rate": 0.000373420260782347,
"loss": 2.8695,
"theoretical_loss": 3.7025974386838962,
"tokens_seen": 860225536
},
{
"epoch": 0.26,
"learning_rate": 0.00037341023069207625,
"loss": 2.9149,
"theoretical_loss": 3.7025698432195044,
"tokens_seen": 860291072
},
{
"epoch": 0.26,
"learning_rate": 0.0003734002006018054,
"loss": 2.6152,
"theoretical_loss": 3.7025422504457857,
"tokens_seen": 860356608
},
{
"epoch": 0.26,
"learning_rate": 0.0003733901705115346,
"loss": 2.9686,
"theoretical_loss": 3.702514660362273,
"tokens_seen": 860422144
},
{
"epoch": 0.26,
"learning_rate": 0.0003733801404212638,
"loss": 2.951,
"theoretical_loss": 3.7024870729684998,
"tokens_seen": 860487680
},
{
"epoch": 0.26,
"learning_rate": 0.00037337011033099297,
"loss": 2.9274,
"theoretical_loss": 3.7024594882639987,
"tokens_seen": 860553216
},
{
"epoch": 0.26,
"learning_rate": 0.00037336008024072215,
"loss": 2.7648,
"theoretical_loss": 3.702431906248303,
"tokens_seen": 860618752
},
{
"epoch": 0.26,
"learning_rate": 0.0003733500501504514,
"loss": 2.7993,
"theoretical_loss": 3.702404326920945,
"tokens_seen": 860684288
},
{
"epoch": 0.26,
"learning_rate": 0.0003733400200601805,
"loss": 2.6367,
"theoretical_loss": 3.7023767502814593,
"tokens_seen": 860749824
},
{
"epoch": 0.26,
"learning_rate": 0.00037332998996990975,
"loss": 2.9093,
"theoretical_loss": 3.7023491763293785,
"tokens_seen": 860815360
},
{
"epoch": 0.26,
"learning_rate": 0.00037331995987963893,
"loss": 2.9705,
"theoretical_loss": 3.7023216050642365,
"tokens_seen": 860880896
},
{
"epoch": 0.26,
"learning_rate": 0.0003733099297893681,
"loss": 2.9332,
"theoretical_loss": 3.7022940364855668,
"tokens_seen": 860946432
},
{
"epoch": 0.26,
"learning_rate": 0.0003732998996990973,
"loss": 3.0288,
"theoretical_loss": 3.702266470592903,
"tokens_seen": 861011968
},
{
"epoch": 0.26,
"learning_rate": 0.0003732898696088265,
"loss": 2.6812,
"theoretical_loss": 3.7022389073857793,
"tokens_seen": 861077504
},
{
"epoch": 0.26,
"learning_rate": 0.00037327983951855565,
"loss": 2.9152,
"theoretical_loss": 3.70221134686373,
"tokens_seen": 861143040
},
{
"epoch": 0.26,
"learning_rate": 0.0003732698094282849,
"loss": 3.0165,
"theoretical_loss": 3.7021837890262876,
"tokens_seen": 861208576
},
{
"epoch": 0.26,
"learning_rate": 0.000373259779338014,
"loss": 2.827,
"theoretical_loss": 3.7021562338729885,
"tokens_seen": 861274112
},
{
"epoch": 0.26,
"learning_rate": 0.00037324974924774325,
"loss": 2.7391,
"theoretical_loss": 3.702128681403366,
"tokens_seen": 861339648
},
{
"epoch": 0.26,
"learning_rate": 0.0003732397191574724,
"loss": 2.7554,
"theoretical_loss": 3.702101131616954,
"tokens_seen": 861405184
},
{
"epoch": 0.26,
"learning_rate": 0.0003732296890672016,
"loss": 2.8663,
"theoretical_loss": 3.702073584513288,
"tokens_seen": 861470720
},
{
"epoch": 0.26,
"learning_rate": 0.0003732196589769308,
"loss": 2.7666,
"theoretical_loss": 3.7020460400919024,
"tokens_seen": 861536256
},
{
"epoch": 0.26,
"learning_rate": 0.00037320962888666,
"loss": 2.7293,
"theoretical_loss": 3.702018498352332,
"tokens_seen": 861601792
},
{
"epoch": 0.26,
"learning_rate": 0.00037319959879638916,
"loss": 2.8819,
"theoretical_loss": 3.701990959294112,
"tokens_seen": 861667328
},
{
"epoch": 0.26,
"learning_rate": 0.00037318956870611834,
"loss": 2.6619,
"theoretical_loss": 3.7019634229167773,
"tokens_seen": 861732864
},
{
"epoch": 0.26,
"objective/train/docs_used": 495406,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8202531337738037,
"objective/train/theoretical_loss": 3.701935889219863,
"objective/train/tokens_used": 882258400,
"theoretical_loss": 3.701935889219863,
"tokens_seen": 861798400
},
{
"epoch": 0.26,
"learning_rate": 0.0003731795386158475,
"loss": 2.6781,
"theoretical_loss": 3.701935889219863,
"tokens_seen": 861798400
},
{
"epoch": 0.26,
"learning_rate": 0.00037316950852557675,
"loss": 2.7052,
"theoretical_loss": 3.7019083582029038,
"tokens_seen": 861863936
},
{
"epoch": 0.26,
"learning_rate": 0.0003731594784353059,
"loss": 3.0729,
"theoretical_loss": 3.701880829865436,
"tokens_seen": 861929472
},
{
"epoch": 0.26,
"learning_rate": 0.0003731494483450351,
"loss": 2.8602,
"theoretical_loss": 3.7018533042069945,
"tokens_seen": 861995008
},
{
"epoch": 0.26,
"learning_rate": 0.0003731394182547643,
"loss": 2.8795,
"theoretical_loss": 3.7018257812271154,
"tokens_seen": 862060544
},
{
"epoch": 0.26,
"learning_rate": 0.0003731293881644935,
"loss": 2.9575,
"theoretical_loss": 3.701798260925335,
"tokens_seen": 862126080
},
{
"epoch": 0.26,
"learning_rate": 0.0003731193580742227,
"loss": 2.7236,
"theoretical_loss": 3.7017707433011875,
"tokens_seen": 862191616
},
{
"epoch": 0.26,
"learning_rate": 0.00037310932798395184,
"loss": 2.6756,
"theoretical_loss": 3.7017432283542098,
"tokens_seen": 862257152
},
{
"epoch": 0.26,
"learning_rate": 0.0003730992978936811,
"loss": 3.0776,
"theoretical_loss": 3.701715716083938,
"tokens_seen": 862322688
},
{
"epoch": 0.26,
"learning_rate": 0.00037308926780341026,
"loss": 2.8599,
"theoretical_loss": 3.7016882064899086,
"tokens_seen": 862388224
},
{
"epoch": 0.26,
"learning_rate": 0.00037307923771313944,
"loss": 2.9194,
"theoretical_loss": 3.701660699571657,
"tokens_seen": 862453760
},
{
"epoch": 0.26,
"learning_rate": 0.0003730692076228686,
"loss": 2.665,
"theoretical_loss": 3.701633195328721,
"tokens_seen": 862519296
},
{
"epoch": 0.26,
"learning_rate": 0.0003730591775325978,
"loss": 2.7973,
"theoretical_loss": 3.701605693760636,
"tokens_seen": 862584832
},
{
"epoch": 0.26,
"learning_rate": 0.000373049147442327,
"loss": 2.9113,
"theoretical_loss": 3.701578194866939,
"tokens_seen": 862650368
},
{
"epoch": 0.26,
"learning_rate": 0.0003730391173520562,
"loss": 2.8849,
"theoretical_loss": 3.701550698647167,
"tokens_seen": 862715904
},
{
"epoch": 0.26,
"learning_rate": 0.00037302908726178534,
"loss": 2.8669,
"theoretical_loss": 3.701523205100856,
"tokens_seen": 862781440
},
{
"epoch": 0.26,
"learning_rate": 0.0003730190571715146,
"loss": 2.8948,
"theoretical_loss": 3.7014957142275446,
"tokens_seen": 862846976
},
{
"epoch": 0.26,
"learning_rate": 0.00037300902708124376,
"loss": 2.8338,
"theoretical_loss": 3.701468226026769,
"tokens_seen": 862912512
},
{
"epoch": 0.26,
"learning_rate": 0.00037299899699097294,
"loss": 2.7343,
"theoretical_loss": 3.7014407404980663,
"tokens_seen": 862978048
},
{
"epoch": 0.26,
"learning_rate": 0.0003729889669007021,
"loss": 2.7257,
"theoretical_loss": 3.7014132576409744,
"tokens_seen": 863043584
},
{
"epoch": 0.26,
"learning_rate": 0.0003729789368104313,
"loss": 2.8995,
"theoretical_loss": 3.7013857774550303,
"tokens_seen": 863109120
},
{
"epoch": 0.26,
"learning_rate": 0.0003729689067201605,
"loss": 2.8747,
"theoretical_loss": 3.7013582999397716,
"tokens_seen": 863174656
},
{
"epoch": 0.26,
"learning_rate": 0.0003729588766298897,
"loss": 2.7776,
"theoretical_loss": 3.701330825094736,
"tokens_seen": 863240192
},
{
"epoch": 0.26,
"learning_rate": 0.00037294884653961885,
"loss": 2.8394,
"theoretical_loss": 3.701303352919462,
"tokens_seen": 863305728
},
{
"epoch": 0.26,
"learning_rate": 0.0003729388164493481,
"loss": 2.7051,
"theoretical_loss": 3.701275883413487,
"tokens_seen": 863371264
},
{
"epoch": 0.26,
"objective/train/docs_used": 496136,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.844367504119873,
"objective/train/theoretical_loss": 3.7012484165763486,
"objective/train/tokens_used": 883896800,
"theoretical_loss": 3.7012484165763486,
"tokens_seen": 863436800
},
{
"epoch": 0.26,
"learning_rate": 0.0003729287863590772,
"loss": 2.6533,
"theoretical_loss": 3.7012484165763486,
"tokens_seen": 863436800
},
{
"epoch": 0.26,
"learning_rate": 0.00037291875626880645,
"loss": 3.013,
"theoretical_loss": 3.701220952407586,
"tokens_seen": 863502336
},
{
"epoch": 0.26,
"learning_rate": 0.0003729087261785356,
"loss": 2.9574,
"theoretical_loss": 3.7011934909067365,
"tokens_seen": 863567872
},
{
"epoch": 0.26,
"learning_rate": 0.0003728986960882648,
"loss": 2.893,
"theoretical_loss": 3.7011660320733393,
"tokens_seen": 863633408
},
{
"epoch": 0.26,
"learning_rate": 0.000372888665997994,
"loss": 2.8463,
"theoretical_loss": 3.701138575906932,
"tokens_seen": 863698944
},
{
"epoch": 0.26,
"learning_rate": 0.00037287863590772317,
"loss": 2.9812,
"theoretical_loss": 3.7011111224070543,
"tokens_seen": 863764480
},
{
"epoch": 0.26,
"learning_rate": 0.00037286860581745235,
"loss": 2.9189,
"theoretical_loss": 3.701083671573244,
"tokens_seen": 863830016
},
{
"epoch": 0.26,
"learning_rate": 0.0003728585757271816,
"loss": 2.68,
"theoretical_loss": 3.7010562234050406,
"tokens_seen": 863895552
},
{
"epoch": 0.26,
"learning_rate": 0.0003728485456369107,
"loss": 2.8874,
"theoretical_loss": 3.701028777901983,
"tokens_seen": 863961088
},
{
"epoch": 0.26,
"learning_rate": 0.00037283851554663995,
"loss": 2.6729,
"theoretical_loss": 3.7010013350636095,
"tokens_seen": 864026624
},
{
"epoch": 0.26,
"learning_rate": 0.00037282848545636913,
"loss": 2.9088,
"theoretical_loss": 3.7009738948894597,
"tokens_seen": 864092160
},
{
"epoch": 0.26,
"learning_rate": 0.0003728184553660983,
"loss": 2.6118,
"theoretical_loss": 3.7009464573790742,
"tokens_seen": 864157696
},
{
"epoch": 0.26,
"learning_rate": 0.0003728084252758275,
"loss": 2.8119,
"theoretical_loss": 3.7009190225319903,
"tokens_seen": 864223232
},
{
"epoch": 0.26,
"learning_rate": 0.0003727983951855567,
"loss": 3.0315,
"theoretical_loss": 3.7008915903477497,
"tokens_seen": 864288768
},
{
"epoch": 0.26,
"learning_rate": 0.00037278836509528585,
"loss": 3.0733,
"theoretical_loss": 3.70086416082589,
"tokens_seen": 864354304
},
{
"epoch": 0.26,
"learning_rate": 0.0003727783350050151,
"loss": 2.8104,
"theoretical_loss": 3.7008367339659523,
"tokens_seen": 864419840
},
{
"epoch": 0.26,
"learning_rate": 0.0003727683049147442,
"loss": 2.9714,
"theoretical_loss": 3.700809309767476,
"tokens_seen": 864485376
},
{
"epoch": 0.26,
"learning_rate": 0.00037275827482447345,
"loss": 2.8635,
"theoretical_loss": 3.7007818882300016,
"tokens_seen": 864550912
},
{
"epoch": 0.26,
"learning_rate": 0.0003727482447342026,
"loss": 2.7127,
"theoretical_loss": 3.7007544693530683,
"tokens_seen": 864616448
},
{
"epoch": 0.26,
"learning_rate": 0.0003727382146439318,
"loss": 2.7181,
"theoretical_loss": 3.7007270531362173,
"tokens_seen": 864681984
},
{
"epoch": 0.26,
"learning_rate": 0.000372728184553661,
"loss": 2.9238,
"theoretical_loss": 3.700699639578988,
"tokens_seen": 864747520
},
{
"epoch": 0.26,
"learning_rate": 0.0003727181544633902,
"loss": 2.6806,
"theoretical_loss": 3.700672228680922,
"tokens_seen": 864813056
},
{
"epoch": 0.26,
"learning_rate": 0.00037270812437311936,
"loss": 2.5866,
"theoretical_loss": 3.700644820441559,
"tokens_seen": 864878592
},
{
"epoch": 0.26,
"learning_rate": 0.00037269809428284854,
"loss": 2.8575,
"theoretical_loss": 3.70061741486044,
"tokens_seen": 864944128
},
{
"epoch": 0.26,
"learning_rate": 0.0003726880641925777,
"loss": 2.916,
"theoretical_loss": 3.7005900119371056,
"tokens_seen": 865009664
},
{
"epoch": 0.26,
"objective/train/docs_used": 497267,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8608226776123047,
"objective/train/theoretical_loss": 3.7005626116710966,
"objective/train/tokens_used": 885535200,
"theoretical_loss": 3.7005626116710966,
"tokens_seen": 865075200
},
{
"epoch": 0.26,
"learning_rate": 0.00037267803410230696,
"loss": 2.9605,
"theoretical_loss": 3.7005626116710966,
"tokens_seen": 865075200
},
{
"epoch": 0.26,
"learning_rate": 0.0003726680040120361,
"loss": 3.1571,
"theoretical_loss": 3.700535214061955,
"tokens_seen": 865140736
},
{
"epoch": 0.26,
"learning_rate": 0.0003726579739217653,
"loss": 2.8383,
"theoretical_loss": 3.7005078191092204,
"tokens_seen": 865206272
},
{
"epoch": 0.26,
"learning_rate": 0.0003726479438314945,
"loss": 3.0967,
"theoretical_loss": 3.7004804268124354,
"tokens_seen": 865271808
},
{
"epoch": 0.26,
"learning_rate": 0.0003726379137412237,
"loss": 2.8194,
"theoretical_loss": 3.700453037171141,
"tokens_seen": 865337344
},
{
"epoch": 0.26,
"learning_rate": 0.00037262788365095286,
"loss": 2.8713,
"theoretical_loss": 3.700425650184878,
"tokens_seen": 865402880
},
{
"epoch": 0.26,
"learning_rate": 0.00037261785356068204,
"loss": 3.1292,
"theoretical_loss": 3.7003982658531887,
"tokens_seen": 865468416
},
{
"epoch": 0.26,
"learning_rate": 0.0003726078234704112,
"loss": 2.7138,
"theoretical_loss": 3.700370884175615,
"tokens_seen": 865533952
},
{
"epoch": 0.26,
"learning_rate": 0.00037259779338014046,
"loss": 3.0194,
"theoretical_loss": 3.7003435051516975,
"tokens_seen": 865599488
},
{
"epoch": 0.26,
"learning_rate": 0.0003725877632898696,
"loss": 2.703,
"theoretical_loss": 3.7003161287809796,
"tokens_seen": 865665024
},
{
"epoch": 0.26,
"learning_rate": 0.0003725777331995988,
"loss": 2.9966,
"theoretical_loss": 3.700288755063003,
"tokens_seen": 865730560
},
{
"epoch": 0.26,
"learning_rate": 0.00037256770310932795,
"loss": 2.7864,
"theoretical_loss": 3.7002613839973097,
"tokens_seen": 865796096
},
{
"epoch": 0.26,
"learning_rate": 0.0003725576730190572,
"loss": 2.8818,
"theoretical_loss": 3.7002340155834412,
"tokens_seen": 865861632
},
{
"epoch": 0.26,
"learning_rate": 0.00037254764292878636,
"loss": 2.8897,
"theoretical_loss": 3.7002066498209416,
"tokens_seen": 865927168
},
{
"epoch": 0.26,
"learning_rate": 0.00037253761283851554,
"loss": 2.9211,
"theoretical_loss": 3.7001792867093517,
"tokens_seen": 865992704
},
{
"epoch": 0.26,
"learning_rate": 0.0003725275827482447,
"loss": 2.9654,
"theoretical_loss": 3.700151926248215,
"tokens_seen": 866058240
},
{
"epoch": 0.26,
"learning_rate": 0.00037251755265797396,
"loss": 2.6239,
"theoretical_loss": 3.700124568437074,
"tokens_seen": 866123776
},
{
"epoch": 0.26,
"learning_rate": 0.0003725075225677031,
"loss": 2.9168,
"theoretical_loss": 3.7000972132754715,
"tokens_seen": 866189312
},
{
"epoch": 0.26,
"learning_rate": 0.0003724974924774323,
"loss": 2.765,
"theoretical_loss": 3.7000698607629507,
"tokens_seen": 866254848
},
{
"epoch": 0.26,
"learning_rate": 0.00037248746238716145,
"loss": 2.9265,
"theoretical_loss": 3.7000425108990553,
"tokens_seen": 866320384
},
{
"epoch": 0.26,
"learning_rate": 0.0003724774322968907,
"loss": 2.6633,
"theoretical_loss": 3.700015163683327,
"tokens_seen": 866385920
},
{
"epoch": 0.26,
"learning_rate": 0.00037246740220661987,
"loss": 2.8371,
"theoretical_loss": 3.69998781911531,
"tokens_seen": 866451456
},
{
"epoch": 0.26,
"learning_rate": 0.00037245737211634905,
"loss": 2.6589,
"theoretical_loss": 3.6999604771945473,
"tokens_seen": 866516992
},
{
"epoch": 0.26,
"learning_rate": 0.00037244734202607823,
"loss": 2.9843,
"theoretical_loss": 3.699933137920583,
"tokens_seen": 866582528
},
{
"epoch": 0.26,
"learning_rate": 0.0003724373119358074,
"loss": 2.8056,
"theoretical_loss": 3.6999058012929606,
"tokens_seen": 866648064
},
{
"epoch": 0.26,
"objective/train/docs_used": 497986,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8182787895202637,
"objective/train/theoretical_loss": 3.6998784673112235,
"objective/train/tokens_used": 887173600,
"theoretical_loss": 3.6998784673112235,
"tokens_seen": 866713600
},
{
"epoch": 0.26,
"learning_rate": 0.0003724272818455366,
"loss": 2.7399,
"theoretical_loss": 3.6998784673112235,
"tokens_seen": 866713600
},
{
"epoch": 0.26,
"learning_rate": 0.0003724172517552658,
"loss": 2.7555,
"theoretical_loss": 3.6998511359749164,
"tokens_seen": 866779136
},
{
"epoch": 0.26,
"learning_rate": 0.00037240722166499495,
"loss": 3.024,
"theoretical_loss": 3.6998238072835816,
"tokens_seen": 866844672
},
{
"epoch": 0.26,
"learning_rate": 0.0003723971915747242,
"loss": 2.8323,
"theoretical_loss": 3.699796481236765,
"tokens_seen": 866910208
},
{
"epoch": 0.26,
"learning_rate": 0.0003723871614844533,
"loss": 2.8729,
"theoretical_loss": 3.6997691578340097,
"tokens_seen": 866975744
},
{
"epoch": 0.26,
"learning_rate": 0.00037237713139418255,
"loss": 2.6956,
"theoretical_loss": 3.6997418370748605,
"tokens_seen": 867041280
},
{
"epoch": 0.26,
"learning_rate": 0.0003723671013039118,
"loss": 2.9541,
"theoretical_loss": 3.699714518958862,
"tokens_seen": 867106816
},
{
"epoch": 0.26,
"learning_rate": 0.0003723570712136409,
"loss": 3.1246,
"theoretical_loss": 3.699687203485558,
"tokens_seen": 867172352
},
{
"epoch": 0.26,
"learning_rate": 0.00037234704112337015,
"loss": 2.9496,
"theoretical_loss": 3.699659890654494,
"tokens_seen": 867237888
},
{
"epoch": 0.26,
"learning_rate": 0.00037233701103309933,
"loss": 2.9057,
"theoretical_loss": 3.699632580465215,
"tokens_seen": 867303424
},
{
"epoch": 0.26,
"learning_rate": 0.0003723269809428285,
"loss": 2.7664,
"theoretical_loss": 3.699605272917264,
"tokens_seen": 867368960
},
{
"epoch": 0.26,
"learning_rate": 0.0003723169508525577,
"loss": 2.8459,
"theoretical_loss": 3.699577968010188,
"tokens_seen": 867434496
},
{
"epoch": 0.26,
"learning_rate": 0.0003723069207622869,
"loss": 2.7134,
"theoretical_loss": 3.6995506657435313,
"tokens_seen": 867500032
},
{
"epoch": 0.26,
"learning_rate": 0.00037229689067201605,
"loss": 2.8287,
"theoretical_loss": 3.699523366116839,
"tokens_seen": 867565568
},
{
"epoch": 0.26,
"learning_rate": 0.0003722868605817453,
"loss": 2.8463,
"theoretical_loss": 3.6994960691296574,
"tokens_seen": 867631104
},
{
"epoch": 0.26,
"learning_rate": 0.0003722768304914744,
"loss": 2.858,
"theoretical_loss": 3.6994687747815305,
"tokens_seen": 867696640
},
{
"epoch": 0.26,
"learning_rate": 0.00037226680040120365,
"loss": 2.6604,
"theoretical_loss": 3.699441483072005,
"tokens_seen": 867762176
},
{
"epoch": 0.26,
"learning_rate": 0.0003722567703109328,
"loss": 2.9352,
"theoretical_loss": 3.6994141940006253,
"tokens_seen": 867827712
},
{
"epoch": 0.26,
"learning_rate": 0.000372246740220662,
"loss": 2.9085,
"theoretical_loss": 3.6993869075669386,
"tokens_seen": 867893248
},
{
"epoch": 0.26,
"learning_rate": 0.0003722367101303912,
"loss": 2.4464,
"theoretical_loss": 3.6993596237704907,
"tokens_seen": 867958784
},
{
"epoch": 0.26,
"learning_rate": 0.0003722266800401204,
"loss": 2.9246,
"theoretical_loss": 3.699332342610827,
"tokens_seen": 868024320
},
{
"epoch": 0.26,
"learning_rate": 0.00037221664994984956,
"loss": 3.1173,
"theoretical_loss": 3.6993050640874934,
"tokens_seen": 868089856
},
{
"epoch": 0.26,
"learning_rate": 0.00037220661985957874,
"loss": 2.7362,
"theoretical_loss": 3.6992777882000363,
"tokens_seen": 868155392
},
{
"epoch": 0.26,
"learning_rate": 0.0003721965897693079,
"loss": 2.9022,
"theoretical_loss": 3.6992505149480026,
"tokens_seen": 868220928
},
{
"epoch": 0.26,
"learning_rate": 0.00037218655967903716,
"loss": 2.9048,
"theoretical_loss": 3.6992232443309385,
"tokens_seen": 868286464
},
{
"epoch": 0.26,
"objective/train/docs_used": 499247,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.538700819015503,
"objective/train/theoretical_loss": 3.69919597634839,
"objective/train/tokens_used": 888812000,
"theoretical_loss": 3.69919597634839,
"tokens_seen": 868352000
},
{
"epoch": 0.26,
"learning_rate": 0.0003721765295887663,
"loss": 2.7579,
"theoretical_loss": 3.69919597634839,
"tokens_seen": 868352000
},
{
"epoch": 0.26,
"learning_rate": 0.0003721664994984955,
"loss": 2.8086,
"theoretical_loss": 3.699168710999905,
"tokens_seen": 868417536
},
{
"epoch": 0.26,
"learning_rate": 0.0003721564694082247,
"loss": 3.0749,
"theoretical_loss": 3.6991414482850296,
"tokens_seen": 868483072
},
{
"epoch": 0.26,
"learning_rate": 0.0003721464393179539,
"loss": 2.8366,
"theoretical_loss": 3.69911418820331,
"tokens_seen": 868548608
},
{
"epoch": 0.26,
"learning_rate": 0.00037213640922768306,
"loss": 2.7217,
"theoretical_loss": 3.6990869307542944,
"tokens_seen": 868614144
},
{
"epoch": 0.26,
"learning_rate": 0.00037212637913741224,
"loss": 2.7085,
"theoretical_loss": 3.699059675937529,
"tokens_seen": 868679680
},
{
"epoch": 0.26,
"learning_rate": 0.0003721163490471414,
"loss": 2.7312,
"theoretical_loss": 3.6990324237525622,
"tokens_seen": 868745216
},
{
"epoch": 0.26,
"learning_rate": 0.00037210631895687066,
"loss": 3.1576,
"theoretical_loss": 3.6990051741989403,
"tokens_seen": 868810752
},
{
"epoch": 0.26,
"learning_rate": 0.0003720962888665998,
"loss": 2.825,
"theoretical_loss": 3.698977927276211,
"tokens_seen": 868876288
},
{
"epoch": 0.26,
"learning_rate": 0.000372086258776329,
"loss": 2.8702,
"theoretical_loss": 3.6989506829839227,
"tokens_seen": 868941824
},
{
"epoch": 0.26,
"learning_rate": 0.00037207622868605815,
"loss": 2.7311,
"theoretical_loss": 3.6989234413216217,
"tokens_seen": 869007360
},
{
"epoch": 0.26,
"learning_rate": 0.0003720661985957874,
"loss": 2.8122,
"theoretical_loss": 3.6988962022888563,
"tokens_seen": 869072896
},
{
"epoch": 0.26,
"learning_rate": 0.00037205616850551656,
"loss": 2.6112,
"theoretical_loss": 3.698868965885175,
"tokens_seen": 869138432
},
{
"epoch": 0.26,
"learning_rate": 0.00037204613841524575,
"loss": 2.7756,
"theoretical_loss": 3.6988417321101252,
"tokens_seen": 869203968
},
{
"epoch": 0.26,
"learning_rate": 0.0003720361083249749,
"loss": 2.7713,
"theoretical_loss": 3.6988145009632563,
"tokens_seen": 869269504
},
{
"epoch": 0.26,
"learning_rate": 0.00037202607823470416,
"loss": 2.8174,
"theoretical_loss": 3.6987872724441146,
"tokens_seen": 869335040
},
{
"epoch": 0.26,
"learning_rate": 0.0003720160481444333,
"loss": 2.8216,
"theoretical_loss": 3.698760046552249,
"tokens_seen": 869400576
},
{
"epoch": 0.26,
"learning_rate": 0.0003720060180541625,
"loss": 2.7106,
"theoretical_loss": 3.6987328232872096,
"tokens_seen": 869466112
},
{
"epoch": 0.26,
"learning_rate": 0.00037199598796389165,
"loss": 3.0358,
"theoretical_loss": 3.698705602648543,
"tokens_seen": 869531648
},
{
"epoch": 0.26,
"learning_rate": 0.0003719859578736209,
"loss": 2.7823,
"theoretical_loss": 3.698678384635799,
"tokens_seen": 869597184
},
{
"epoch": 0.26,
"learning_rate": 0.00037197592778335007,
"loss": 2.7946,
"theoretical_loss": 3.698651169248526,
"tokens_seen": 869662720
},
{
"epoch": 0.26,
"learning_rate": 0.00037196589769307925,
"loss": 2.7262,
"theoretical_loss": 3.698623956486273,
"tokens_seen": 869728256
},
{
"epoch": 0.26,
"learning_rate": 0.00037195586760280843,
"loss": 2.8652,
"theoretical_loss": 3.6985967463485894,
"tokens_seen": 869793792
},
{
"epoch": 0.26,
"learning_rate": 0.0003719458375125376,
"loss": 2.7269,
"theoretical_loss": 3.6985695388350233,
"tokens_seen": 869859328
},
{
"epoch": 0.26,
"learning_rate": 0.0003719358074222668,
"loss": 2.6208,
"theoretical_loss": 3.698542333945125,
"tokens_seen": 869924864
},
{
"epoch": 0.26,
"objective/train/docs_used": 499921,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.791407346725464,
"objective/train/theoretical_loss": 3.6985151316784437,
"objective/train/tokens_used": 890450400,
"theoretical_loss": 3.6985151316784437,
"tokens_seen": 869990400
},
{
"epoch": 0.26,
"learning_rate": 0.000371925777331996,
"loss": 3.009,
"theoretical_loss": 3.6985151316784437,
"tokens_seen": 869990400
},
{
"epoch": 0.26,
"learning_rate": 0.00037191574724172515,
"loss": 2.68,
"theoretical_loss": 3.6984879320345287,
"tokens_seen": 870055936
},
{
"epoch": 0.26,
"learning_rate": 0.0003719057171514544,
"loss": 2.7942,
"theoretical_loss": 3.6984607350129295,
"tokens_seen": 870121472
},
{
"epoch": 0.26,
"learning_rate": 0.0003718956870611835,
"loss": 2.93,
"theoretical_loss": 3.6984335406131956,
"tokens_seen": 870187008
},
{
"epoch": 0.26,
"learning_rate": 0.00037188565697091275,
"loss": 2.7328,
"theoretical_loss": 3.6984063488348777,
"tokens_seen": 870252544
},
{
"epoch": 0.26,
"learning_rate": 0.00037187562688064193,
"loss": 2.9684,
"theoretical_loss": 3.698379159677525,
"tokens_seen": 870318080
},
{
"epoch": 0.26,
"learning_rate": 0.0003718655967903711,
"loss": 2.8294,
"theoretical_loss": 3.6983519731406878,
"tokens_seen": 870383616
},
{
"epoch": 0.26,
"learning_rate": 0.0003718555667001003,
"loss": 2.827,
"theoretical_loss": 3.6983247892239155,
"tokens_seen": 870449152
},
{
"epoch": 0.26,
"learning_rate": 0.00037184553660982953,
"loss": 2.6839,
"theoretical_loss": 3.6982976079267598,
"tokens_seen": 870514688
},
{
"epoch": 0.26,
"learning_rate": 0.00037183550651955866,
"loss": 2.8553,
"theoretical_loss": 3.69827042924877,
"tokens_seen": 870580224
},
{
"epoch": 0.26,
"learning_rate": 0.0003718254764292879,
"loss": 2.7846,
"theoretical_loss": 3.6982432531894966,
"tokens_seen": 870645760
},
{
"epoch": 0.26,
"learning_rate": 0.000371815446339017,
"loss": 3.101,
"theoretical_loss": 3.69821607974849,
"tokens_seen": 870711296
},
{
"epoch": 0.26,
"learning_rate": 0.00037180541624874625,
"loss": 2.8129,
"theoretical_loss": 3.698188908925302,
"tokens_seen": 870776832
},
{
"epoch": 0.26,
"learning_rate": 0.00037179538615847544,
"loss": 2.6254,
"theoretical_loss": 3.698161740719483,
"tokens_seen": 870842368
},
{
"epoch": 0.26,
"learning_rate": 0.0003717853560682046,
"loss": 2.9442,
"theoretical_loss": 3.6981345751305827,
"tokens_seen": 870907904
},
{
"epoch": 0.26,
"learning_rate": 0.0003717753259779338,
"loss": 2.9083,
"theoretical_loss": 3.698107412158154,
"tokens_seen": 870973440
},
{
"epoch": 0.26,
"learning_rate": 0.000371765295887663,
"loss": 2.7982,
"theoretical_loss": 3.6980802518017466,
"tokens_seen": 871038976
},
{
"epoch": 0.26,
"learning_rate": 0.00037175526579739216,
"loss": 2.8036,
"theoretical_loss": 3.6980530940609126,
"tokens_seen": 871104512
},
{
"epoch": 0.26,
"learning_rate": 0.0003717452357071214,
"loss": 2.999,
"theoretical_loss": 3.6980259389352033,
"tokens_seen": 871170048
},
{
"epoch": 0.26,
"learning_rate": 0.0003717352056168505,
"loss": 2.9263,
"theoretical_loss": 3.6979987864241695,
"tokens_seen": 871235584
},
{
"epoch": 0.26,
"learning_rate": 0.00037172517552657976,
"loss": 2.7857,
"theoretical_loss": 3.697971636527364,
"tokens_seen": 871301120
},
{
"epoch": 0.26,
"learning_rate": 0.0003717151454363089,
"loss": 2.6934,
"theoretical_loss": 3.697944489244337,
"tokens_seen": 871366656
},
{
"epoch": 0.26,
"learning_rate": 0.0003717051153460381,
"loss": 2.9488,
"theoretical_loss": 3.697917344574641,
"tokens_seen": 871432192
},
{
"epoch": 0.26,
"learning_rate": 0.0003716950852557673,
"loss": 2.7705,
"theoretical_loss": 3.6978902025178284,
"tokens_seen": 871497728
},
{
"epoch": 0.26,
"learning_rate": 0.0003716850551654965,
"loss": 2.8756,
"theoretical_loss": 3.697863063073451,
"tokens_seen": 871563264
},
{
"epoch": 0.26,
"objective/train/docs_used": 501092,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5174808502197266,
"objective/train/theoretical_loss": 3.6978359262410603,
"objective/train/tokens_used": 892088800,
"theoretical_loss": 3.6978359262410603,
"tokens_seen": 871628800
},
{
"epoch": 0.26,
"learning_rate": 0.00037167502507522566,
"loss": 2.7388,
"theoretical_loss": 3.6978359262410603,
"tokens_seen": 871628800
},
{
"epoch": 0.26,
"learning_rate": 0.0003716649949849549,
"loss": 2.9399,
"theoretical_loss": 3.6978087920202087,
"tokens_seen": 871694336
},
{
"epoch": 0.26,
"learning_rate": 0.000371654964894684,
"loss": 2.7911,
"theoretical_loss": 3.6977816604104494,
"tokens_seen": 871759872
},
{
"epoch": 0.26,
"learning_rate": 0.00037164493480441326,
"loss": 2.7918,
"theoretical_loss": 3.6977545314113343,
"tokens_seen": 871825408
},
{
"epoch": 0.26,
"learning_rate": 0.0003716349047141424,
"loss": 2.8859,
"theoretical_loss": 3.697727405022416,
"tokens_seen": 871890944
},
{
"epoch": 0.26,
"learning_rate": 0.0003716248746238716,
"loss": 2.8506,
"theoretical_loss": 3.6977002812432476,
"tokens_seen": 871956480
},
{
"epoch": 0.26,
"learning_rate": 0.00037161484453360086,
"loss": 2.6407,
"theoretical_loss": 3.6976731600733808,
"tokens_seen": 872022016
},
{
"epoch": 0.26,
"learning_rate": 0.00037160481444333,
"loss": 2.8717,
"theoretical_loss": 3.6976460415123693,
"tokens_seen": 872087552
},
{
"epoch": 0.26,
"learning_rate": 0.0003715947843530592,
"loss": 2.8034,
"theoretical_loss": 3.6976189255597665,
"tokens_seen": 872153088
},
{
"epoch": 0.26,
"learning_rate": 0.00037158475426278835,
"loss": 2.7559,
"theoretical_loss": 3.697591812215125,
"tokens_seen": 872218624
},
{
"epoch": 0.26,
"learning_rate": 0.0003715747241725176,
"loss": 2.9023,
"theoretical_loss": 3.697564701477998,
"tokens_seen": 872284160
},
{
"epoch": 0.26,
"learning_rate": 0.00037156469408224676,
"loss": 2.7032,
"theoretical_loss": 3.6975375933479393,
"tokens_seen": 872349696
},
{
"epoch": 0.26,
"learning_rate": 0.00037155466399197595,
"loss": 2.8198,
"theoretical_loss": 3.6975104878245015,
"tokens_seen": 872415232
},
{
"epoch": 0.26,
"learning_rate": 0.0003715446339017051,
"loss": 2.684,
"theoretical_loss": 3.6974833849072395,
"tokens_seen": 872480768
},
{
"epoch": 0.26,
"learning_rate": 0.00037153460381143436,
"loss": 2.8333,
"theoretical_loss": 3.6974562845957055,
"tokens_seen": 872546304
},
{
"epoch": 0.26,
"learning_rate": 0.0003715245737211635,
"loss": 2.6457,
"theoretical_loss": 3.697429186889454,
"tokens_seen": 872611840
},
{
"epoch": 0.26,
"learning_rate": 0.0003715145436308927,
"loss": 2.7519,
"theoretical_loss": 3.6974020917880397,
"tokens_seen": 872677376
},
{
"epoch": 0.26,
"learning_rate": 0.00037150451354062185,
"loss": 2.6511,
"theoretical_loss": 3.697374999291015,
"tokens_seen": 872742912
},
{
"epoch": 0.26,
"learning_rate": 0.0003714944834503511,
"loss": 2.723,
"theoretical_loss": 3.6973479093979353,
"tokens_seen": 872808448
},
{
"epoch": 0.26,
"learning_rate": 0.00037148445336008027,
"loss": 2.9641,
"theoretical_loss": 3.697320822108354,
"tokens_seen": 872873984
},
{
"epoch": 0.26,
"learning_rate": 0.00037147442326980945,
"loss": 2.7983,
"theoretical_loss": 3.697293737421826,
"tokens_seen": 872939520
},
{
"epoch": 0.26,
"learning_rate": 0.00037146439317953863,
"loss": 2.7666,
"theoretical_loss": 3.697266655337906,
"tokens_seen": 873005056
},
{
"epoch": 0.26,
"learning_rate": 0.0003714543630892678,
"loss": 3.0111,
"theoretical_loss": 3.6972395758561483,
"tokens_seen": 873070592
},
{
"epoch": 0.26,
"learning_rate": 0.000371444332998997,
"loss": 2.8086,
"theoretical_loss": 3.6972124989761066,
"tokens_seen": 873136128
},
{
"epoch": 0.26,
"learning_rate": 0.00037143430290872623,
"loss": 2.9129,
"theoretical_loss": 3.697185424697337,
"tokens_seen": 873201664
},
{
"epoch": 0.26,
"objective/train/docs_used": 501627,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7002902030944824,
"objective/train/theoretical_loss": 3.6971583530193937,
"objective/train/tokens_used": 893727200,
"theoretical_loss": 3.6971583530193937,
"tokens_seen": 873267200
},
{
"epoch": 0.26,
"learning_rate": 0.00037142427281845535,
"loss": 2.8948,
"theoretical_loss": 3.6971583530193937,
"tokens_seen": 873267200
},
{
"epoch": 0.26,
"learning_rate": 0.0003714142427281846,
"loss": 2.9773,
"theoretical_loss": 3.697131283941832,
"tokens_seen": 873332736
},
{
"epoch": 0.26,
"learning_rate": 0.0003714042126379137,
"loss": 2.8083,
"theoretical_loss": 3.697104217464207,
"tokens_seen": 873398272
},
{
"epoch": 0.26,
"learning_rate": 0.00037139418254764295,
"loss": 3.0307,
"theoretical_loss": 3.697077153586074,
"tokens_seen": 873463808
},
{
"epoch": 0.26,
"learning_rate": 0.00037138415245737213,
"loss": 2.6658,
"theoretical_loss": 3.697050092306988,
"tokens_seen": 873529344
},
{
"epoch": 0.26,
"learning_rate": 0.0003713741223671013,
"loss": 2.8165,
"theoretical_loss": 3.697023033626504,
"tokens_seen": 873594880
},
{
"epoch": 0.26,
"learning_rate": 0.0003713640922768305,
"loss": 3.0105,
"theoretical_loss": 3.696995977544179,
"tokens_seen": 873660416
},
{
"epoch": 0.26,
"learning_rate": 0.00037135406218655973,
"loss": 2.6068,
"theoretical_loss": 3.6969689240595676,
"tokens_seen": 873725952
},
{
"epoch": 0.26,
"learning_rate": 0.00037134403209628886,
"loss": 2.8436,
"theoretical_loss": 3.6969418731722254,
"tokens_seen": 873791488
},
{
"epoch": 0.26,
"learning_rate": 0.0003713340020060181,
"loss": 2.9607,
"theoretical_loss": 3.696914824881709,
"tokens_seen": 873857024
},
{
"epoch": 0.26,
"learning_rate": 0.0003713239719157472,
"loss": 2.9046,
"theoretical_loss": 3.6968877791875743,
"tokens_seen": 873922560
},
{
"epoch": 0.26,
"learning_rate": 0.00037131394182547645,
"loss": 2.8498,
"theoretical_loss": 3.6968607360893766,
"tokens_seen": 873988096
},
{
"epoch": 0.26,
"learning_rate": 0.00037130391173520564,
"loss": 3.0017,
"theoretical_loss": 3.6968336955866725,
"tokens_seen": 874053632
},
{
"epoch": 0.26,
"learning_rate": 0.0003712938816449348,
"loss": 3.0806,
"theoretical_loss": 3.6968066576790193,
"tokens_seen": 874119168
},
{
"epoch": 0.26,
"learning_rate": 0.000371283851554664,
"loss": 2.8098,
"theoretical_loss": 3.6967796223659724,
"tokens_seen": 874184704
},
{
"epoch": 0.26,
"learning_rate": 0.0003712738214643932,
"loss": 2.6266,
"theoretical_loss": 3.6967525896470876,
"tokens_seen": 874250240
},
{
"epoch": 0.26,
"learning_rate": 0.00037126379137412236,
"loss": 2.7467,
"theoretical_loss": 3.6967255595219233,
"tokens_seen": 874315776
},
{
"epoch": 0.26,
"learning_rate": 0.0003712537612838516,
"loss": 2.9334,
"theoretical_loss": 3.6966985319900356,
"tokens_seen": 874381312
},
{
"epoch": 0.26,
"learning_rate": 0.0003712437311935807,
"loss": 2.7791,
"theoretical_loss": 3.6966715070509806,
"tokens_seen": 874446848
},
{
"epoch": 0.27,
"learning_rate": 0.00037123370110330996,
"loss": 2.9757,
"theoretical_loss": 3.696644484704316,
"tokens_seen": 874512384
},
{
"epoch": 0.27,
"learning_rate": 0.0003712236710130391,
"loss": 2.7997,
"theoretical_loss": 3.6966174649495986,
"tokens_seen": 874577920
},
{
"epoch": 0.27,
"learning_rate": 0.0003712136409227683,
"loss": 2.8874,
"theoretical_loss": 3.696590447786386,
"tokens_seen": 874643456
},
{
"epoch": 0.27,
"learning_rate": 0.0003712036108324975,
"loss": 2.7927,
"theoretical_loss": 3.6965634332142345,
"tokens_seen": 874708992
},
{
"epoch": 0.27,
"learning_rate": 0.0003711935807422267,
"loss": 2.7128,
"theoretical_loss": 3.696536421232703,
"tokens_seen": 874774528
},
{
"epoch": 0.27,
"learning_rate": 0.00037118355065195586,
"loss": 2.9539,
"theoretical_loss": 3.696509411841348,
"tokens_seen": 874840064
},
{
"epoch": 0.27,
"objective/train/docs_used": 502963,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7756283283233643,
"objective/train/theoretical_loss": 3.6964824050397276,
"objective/train/tokens_used": 895365600,
"theoretical_loss": 3.6964824050397276,
"tokens_seen": 874905600
},
{
"epoch": 0.27,
"learning_rate": 0.0003711735205616851,
"loss": 2.8658,
"theoretical_loss": 3.6964824050397276,
"tokens_seen": 874905600
},
{
"epoch": 0.27,
"learning_rate": 0.0003711634904714142,
"loss": 2.7761,
"theoretical_loss": 3.696455400827398,
"tokens_seen": 874971136
},
{
"epoch": 0.27,
"learning_rate": 0.00037115346038114346,
"loss": 2.9083,
"theoretical_loss": 3.6964283992039197,
"tokens_seen": 875036672
},
{
"epoch": 0.27,
"learning_rate": 0.0003711434302908726,
"loss": 2.7506,
"theoretical_loss": 3.6964014001688486,
"tokens_seen": 875102208
},
{
"epoch": 0.27,
"learning_rate": 0.0003711334002006018,
"loss": 2.9002,
"theoretical_loss": 3.6963744037217436,
"tokens_seen": 875167744
},
{
"epoch": 0.27,
"learning_rate": 0.000371123370110331,
"loss": 2.9138,
"theoretical_loss": 3.6963474098621623,
"tokens_seen": 875233280
},
{
"epoch": 0.27,
"learning_rate": 0.0003711133400200602,
"loss": 2.832,
"theoretical_loss": 3.6963204185896643,
"tokens_seen": 875298816
},
{
"epoch": 0.27,
"learning_rate": 0.00037110330992978937,
"loss": 2.7469,
"theoretical_loss": 3.6962934299038057,
"tokens_seen": 875364352
},
{
"epoch": 0.27,
"learning_rate": 0.00037109327983951855,
"loss": 2.7685,
"theoretical_loss": 3.6962664438041473,
"tokens_seen": 875429888
},
{
"epoch": 0.27,
"learning_rate": 0.00037108324974924773,
"loss": 2.889,
"theoretical_loss": 3.696239460290246,
"tokens_seen": 875495424
},
{
"epoch": 0.27,
"learning_rate": 0.00037107321965897696,
"loss": 2.5318,
"theoretical_loss": 3.6962124793616615,
"tokens_seen": 875560960
},
{
"epoch": 0.27,
"learning_rate": 0.0003710631895687061,
"loss": 2.6106,
"theoretical_loss": 3.6961855010179527,
"tokens_seen": 875626496
},
{
"epoch": 0.27,
"learning_rate": 0.0003710531594784353,
"loss": 2.7615,
"theoretical_loss": 3.696158525258678,
"tokens_seen": 875692032
},
{
"epoch": 0.27,
"learning_rate": 0.00037104312938816445,
"loss": 2.8598,
"theoretical_loss": 3.696131552083396,
"tokens_seen": 875757568
},
{
"epoch": 0.27,
"learning_rate": 0.0003710330992978937,
"loss": 2.9162,
"theoretical_loss": 3.696104581491667,
"tokens_seen": 875823104
},
{
"epoch": 0.27,
"learning_rate": 0.00037102306920762287,
"loss": 2.6105,
"theoretical_loss": 3.696077613483049,
"tokens_seen": 875888640
},
{
"epoch": 0.27,
"learning_rate": 0.00037101303911735205,
"loss": 2.7057,
"theoretical_loss": 3.6960506480571027,
"tokens_seen": 875954176
},
{
"epoch": 0.27,
"learning_rate": 0.00037100300902708123,
"loss": 2.9171,
"theoretical_loss": 3.6960236852133868,
"tokens_seen": 876019712
},
{
"epoch": 0.27,
"learning_rate": 0.00037099297893681047,
"loss": 2.9705,
"theoretical_loss": 3.6959967249514607,
"tokens_seen": 876085248
},
{
"epoch": 0.27,
"learning_rate": 0.0003709829488465396,
"loss": 2.8636,
"theoretical_loss": 3.6959697672708844,
"tokens_seen": 876150784
},
{
"epoch": 0.27,
"learning_rate": 0.00037097291875626883,
"loss": 2.8293,
"theoretical_loss": 3.695942812171218,
"tokens_seen": 876216320
},
{
"epoch": 0.27,
"learning_rate": 0.00037096288866599796,
"loss": 2.8739,
"theoretical_loss": 3.6959158596520205,
"tokens_seen": 876281856
},
{
"epoch": 0.27,
"learning_rate": 0.0003709528585757272,
"loss": 2.7225,
"theoretical_loss": 3.6958889097128527,
"tokens_seen": 876347392
},
{
"epoch": 0.27,
"learning_rate": 0.0003709428284854564,
"loss": 2.8191,
"theoretical_loss": 3.6958619623532742,
"tokens_seen": 876412928
},
{
"epoch": 0.27,
"learning_rate": 0.00037093279839518555,
"loss": 2.8665,
"theoretical_loss": 3.6958350175728456,
"tokens_seen": 876478464
},
{
"epoch": 0.27,
"objective/train/docs_used": 503455,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.034245491027832,
"objective/train/theoretical_loss": 3.695808075371127,
"objective/train/tokens_used": 897004000,
"theoretical_loss": 3.695808075371127,
"tokens_seen": 876544000
},
{
"epoch": 0.27,
"learning_rate": 0.00037092276830491474,
"loss": 2.9443,
"theoretical_loss": 3.695808075371127,
"tokens_seen": 876544000
},
{
"epoch": 0.27,
"learning_rate": 0.0003709127382146439,
"loss": 2.8702,
"theoretical_loss": 3.695781135747679,
"tokens_seen": 876609536
},
{
"epoch": 0.27,
"learning_rate": 0.0003709027081243731,
"loss": 2.9946,
"theoretical_loss": 3.695754198702062,
"tokens_seen": 876675072
},
{
"epoch": 0.27,
"learning_rate": 0.00037089267803410233,
"loss": 2.7969,
"theoretical_loss": 3.6957272642338372,
"tokens_seen": 876740608
},
{
"epoch": 0.27,
"learning_rate": 0.00037088264794383146,
"loss": 2.7994,
"theoretical_loss": 3.695700332342564,
"tokens_seen": 876806144
},
{
"epoch": 0.27,
"learning_rate": 0.0003708726178535607,
"loss": 3.0413,
"theoretical_loss": 3.6956734030278047,
"tokens_seen": 876871680
},
{
"epoch": 0.27,
"learning_rate": 0.00037086258776328993,
"loss": 2.8023,
"theoretical_loss": 3.6956464762891192,
"tokens_seen": 876937216
},
{
"epoch": 0.27,
"learning_rate": 0.00037085255767301906,
"loss": 2.7356,
"theoretical_loss": 3.69561955212607,
"tokens_seen": 877002752
},
{
"epoch": 0.27,
"learning_rate": 0.0003708425275827483,
"loss": 2.9497,
"theoretical_loss": 3.6955926305382167,
"tokens_seen": 877068288
},
{
"epoch": 0.27,
"learning_rate": 0.0003708324974924774,
"loss": 2.863,
"theoretical_loss": 3.6955657115251213,
"tokens_seen": 877133824
},
{
"epoch": 0.27,
"learning_rate": 0.00037082246740220665,
"loss": 2.7944,
"theoretical_loss": 3.695538795086345,
"tokens_seen": 877199360
},
{
"epoch": 0.27,
"learning_rate": 0.00037081243731193584,
"loss": 2.5839,
"theoretical_loss": 3.69551188122145,
"tokens_seen": 877264896
},
{
"epoch": 0.27,
"learning_rate": 0.000370802407221665,
"loss": 2.9412,
"theoretical_loss": 3.695484969929997,
"tokens_seen": 877330432
},
{
"epoch": 0.27,
"learning_rate": 0.0003707923771313942,
"loss": 2.6793,
"theoretical_loss": 3.695458061211548,
"tokens_seen": 877395968
},
{
"epoch": 0.27,
"learning_rate": 0.0003707823470411234,
"loss": 2.9084,
"theoretical_loss": 3.6954311550656658,
"tokens_seen": 877461504
},
{
"epoch": 0.27,
"learning_rate": 0.00037077231695085256,
"loss": 2.8463,
"theoretical_loss": 3.6954042514919108,
"tokens_seen": 877527040
},
{
"epoch": 0.27,
"learning_rate": 0.0003707622868605818,
"loss": 2.7138,
"theoretical_loss": 3.6953773504898457,
"tokens_seen": 877592576
},
{
"epoch": 0.27,
"learning_rate": 0.0003707522567703109,
"loss": 2.8347,
"theoretical_loss": 3.6953504520590332,
"tokens_seen": 877658112
},
{
"epoch": 0.27,
"learning_rate": 0.00037074222668004016,
"loss": 2.5882,
"theoretical_loss": 3.695323556199035,
"tokens_seen": 877723648
},
{
"epoch": 0.27,
"learning_rate": 0.0003707321965897693,
"loss": 3.1172,
"theoretical_loss": 3.6952966629094135,
"tokens_seen": 877789184
},
{
"epoch": 0.27,
"learning_rate": 0.0003707221664994985,
"loss": 2.9794,
"theoretical_loss": 3.695269772189731,
"tokens_seen": 877854720
},
{
"epoch": 0.27,
"learning_rate": 0.0003707121364092277,
"loss": 2.82,
"theoretical_loss": 3.6952428840395504,
"tokens_seen": 877920256
},
{
"epoch": 0.27,
"learning_rate": 0.0003707021063189569,
"loss": 2.8615,
"theoretical_loss": 3.6952159984584343,
"tokens_seen": 877985792
},
{
"epoch": 0.27,
"learning_rate": 0.00037069207622868606,
"loss": 2.9351,
"theoretical_loss": 3.6951891154459453,
"tokens_seen": 878051328
},
{
"epoch": 0.27,
"learning_rate": 0.0003706820461384153,
"loss": 2.572,
"theoretical_loss": 3.6951622350016473,
"tokens_seen": 878116864
},
{
"epoch": 0.27,
"objective/train/docs_used": 504719,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.43080472946167,
"objective/train/theoretical_loss": 3.6951353571251015,
"objective/train/tokens_used": 898642400,
"theoretical_loss": 3.6951353571251015,
"tokens_seen": 878182400
},
{
"epoch": 0.27,
"learning_rate": 0.0003706720160481444,
"loss": 2.7282,
"theoretical_loss": 3.6951353571251015,
"tokens_seen": 878182400
},
{
"epoch": 0.27,
"learning_rate": 0.00037066198595787366,
"loss": 2.7001,
"theoretical_loss": 3.695108481815873,
"tokens_seen": 878247936
},
{
"epoch": 0.27,
"learning_rate": 0.0003706519558676028,
"loss": 2.7422,
"theoretical_loss": 3.6950816090735232,
"tokens_seen": 878313472
},
{
"epoch": 0.27,
"learning_rate": 0.000370641925777332,
"loss": 2.8634,
"theoretical_loss": 3.6950547388976167,
"tokens_seen": 878379008
},
{
"epoch": 0.27,
"learning_rate": 0.0003706318956870612,
"loss": 2.8655,
"theoretical_loss": 3.6950278712877163,
"tokens_seen": 878444544
},
{
"epoch": 0.27,
"learning_rate": 0.0003706218655967904,
"loss": 2.8279,
"theoretical_loss": 3.6950010062433862,
"tokens_seen": 878510080
},
{
"epoch": 0.27,
"learning_rate": 0.00037061183550651957,
"loss": 2.6889,
"theoretical_loss": 3.694974143764189,
"tokens_seen": 878575616
},
{
"epoch": 0.27,
"learning_rate": 0.00037060180541624875,
"loss": 2.8813,
"theoretical_loss": 3.6949472838496895,
"tokens_seen": 878641152
},
{
"epoch": 0.27,
"learning_rate": 0.00037059177532597793,
"loss": 2.739,
"theoretical_loss": 3.6949204264994515,
"tokens_seen": 878706688
},
{
"epoch": 0.27,
"learning_rate": 0.00037058174523570716,
"loss": 2.7992,
"theoretical_loss": 3.6948935717130382,
"tokens_seen": 878772224
},
{
"epoch": 0.27,
"learning_rate": 0.0003705717151454363,
"loss": 2.818,
"theoretical_loss": 3.6948667194900144,
"tokens_seen": 878837760
},
{
"epoch": 0.27,
"learning_rate": 0.0003705616850551655,
"loss": 2.9304,
"theoretical_loss": 3.6948398698299436,
"tokens_seen": 878903296
},
{
"epoch": 0.27,
"learning_rate": 0.00037055165496489465,
"loss": 2.7381,
"theoretical_loss": 3.694813022732391,
"tokens_seen": 878968832
},
{
"epoch": 0.27,
"learning_rate": 0.0003705416248746239,
"loss": 2.7307,
"theoretical_loss": 3.69478617819692,
"tokens_seen": 879034368
},
{
"epoch": 0.27,
"learning_rate": 0.00037053159478435307,
"loss": 2.782,
"theoretical_loss": 3.694759336223096,
"tokens_seen": 879099904
},
{
"epoch": 0.27,
"learning_rate": 0.00037052156469408225,
"loss": 3.0477,
"theoretical_loss": 3.694732496810483,
"tokens_seen": 879165440
},
{
"epoch": 0.27,
"learning_rate": 0.00037051153460381143,
"loss": 3.1684,
"theoretical_loss": 3.694705659958646,
"tokens_seen": 879230976
},
{
"epoch": 0.27,
"learning_rate": 0.00037050150451354067,
"loss": 2.8218,
"theoretical_loss": 3.6946788256671494,
"tokens_seen": 879296512
},
{
"epoch": 0.27,
"learning_rate": 0.0003704914744232698,
"loss": 2.8404,
"theoretical_loss": 3.694651993935559,
"tokens_seen": 879362048
},
{
"epoch": 0.27,
"learning_rate": 0.00037048144433299903,
"loss": 2.8719,
"theoretical_loss": 3.6946251647634396,
"tokens_seen": 879427584
},
{
"epoch": 0.27,
"learning_rate": 0.00037047141424272816,
"loss": 2.7368,
"theoretical_loss": 3.6945983381503558,
"tokens_seen": 879493120
},
{
"epoch": 0.27,
"learning_rate": 0.0003704613841524574,
"loss": 3.0134,
"theoretical_loss": 3.6945715140958733,
"tokens_seen": 879558656
},
{
"epoch": 0.27,
"learning_rate": 0.0003704513540621866,
"loss": 2.9259,
"theoretical_loss": 3.694544692599557,
"tokens_seen": 879624192
},
{
"epoch": 0.27,
"learning_rate": 0.00037044132397191575,
"loss": 2.9133,
"theoretical_loss": 3.6945178736609723,
"tokens_seen": 879689728
},
{
"epoch": 0.27,
"learning_rate": 0.00037043129388164494,
"loss": 2.7737,
"theoretical_loss": 3.6944910572796856,
"tokens_seen": 879755264
},
{
"epoch": 0.27,
"objective/train/docs_used": 505361,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5627317428588867,
"objective/train/theoretical_loss": 3.694464243455262,
"objective/train/tokens_used": 900280800,
"theoretical_loss": 3.694464243455262,
"tokens_seen": 879820800
},
{
"epoch": 0.27,
"learning_rate": 0.0003704212637913741,
"loss": 2.7141,
"theoretical_loss": 3.694464243455262,
"tokens_seen": 879820800
},
{
"epoch": 0.27,
"learning_rate": 0.0003704112337011033,
"loss": 2.8957,
"theoretical_loss": 3.694437432187268,
"tokens_seen": 879886336
},
{
"epoch": 0.27,
"learning_rate": 0.00037040120361083253,
"loss": 2.959,
"theoretical_loss": 3.694410623475268,
"tokens_seen": 879951872
},
{
"epoch": 0.27,
"learning_rate": 0.00037039117352056166,
"loss": 2.7843,
"theoretical_loss": 3.694383817318829,
"tokens_seen": 880017408
},
{
"epoch": 0.27,
"learning_rate": 0.0003703811434302909,
"loss": 2.4929,
"theoretical_loss": 3.694357013717517,
"tokens_seen": 880082944
},
{
"epoch": 0.27,
"learning_rate": 0.0003703711133400201,
"loss": 3.0005,
"theoretical_loss": 3.6943302126708986,
"tokens_seen": 880148480
},
{
"epoch": 0.27,
"learning_rate": 0.00037036108324974926,
"loss": 2.8931,
"theoretical_loss": 3.6943034141785396,
"tokens_seen": 880214016
},
{
"epoch": 0.27,
"learning_rate": 0.00037035105315947844,
"loss": 2.8677,
"theoretical_loss": 3.694276618240006,
"tokens_seen": 880279552
},
{
"epoch": 0.27,
"learning_rate": 0.0003703410230692076,
"loss": 2.7247,
"theoretical_loss": 3.6942498248548654,
"tokens_seen": 880345088
},
{
"epoch": 0.27,
"learning_rate": 0.0003703309929789368,
"loss": 2.7371,
"theoretical_loss": 3.6942230340226834,
"tokens_seen": 880410624
},
{
"epoch": 0.27,
"learning_rate": 0.00037032096288866604,
"loss": 2.9029,
"theoretical_loss": 3.6941962457430275,
"tokens_seen": 880476160
},
{
"epoch": 0.27,
"learning_rate": 0.00037031093279839516,
"loss": 3.002,
"theoretical_loss": 3.6941694600154644,
"tokens_seen": 880541696
},
{
"epoch": 0.27,
"learning_rate": 0.0003703009027081244,
"loss": 2.918,
"theoretical_loss": 3.6941426768395607,
"tokens_seen": 880607232
},
{
"epoch": 0.27,
"learning_rate": 0.0003702908726178535,
"loss": 2.7209,
"theoretical_loss": 3.6941158962148837,
"tokens_seen": 880672768
},
{
"epoch": 0.27,
"learning_rate": 0.00037028084252758276,
"loss": 2.8952,
"theoretical_loss": 3.694089118141,
"tokens_seen": 880738304
},
{
"epoch": 0.27,
"learning_rate": 0.00037027081243731194,
"loss": 2.9075,
"theoretical_loss": 3.694062342617478,
"tokens_seen": 880803840
},
{
"epoch": 0.27,
"learning_rate": 0.0003702607823470411,
"loss": 3.0009,
"theoretical_loss": 3.6940355696438845,
"tokens_seen": 880869376
},
{
"epoch": 0.27,
"learning_rate": 0.0003702507522567703,
"loss": 2.8192,
"theoretical_loss": 3.6940087992197865,
"tokens_seen": 880934912
},
{
"epoch": 0.27,
"learning_rate": 0.0003702407221664995,
"loss": 2.843,
"theoretical_loss": 3.693982031344752,
"tokens_seen": 881000448
},
{
"epoch": 0.27,
"learning_rate": 0.00037023069207622867,
"loss": 2.8356,
"theoretical_loss": 3.6939552660183486,
"tokens_seen": 881065984
},
{
"epoch": 0.27,
"learning_rate": 0.0003702206619859579,
"loss": 2.9324,
"theoretical_loss": 3.6939285032401443,
"tokens_seen": 881131520
},
{
"epoch": 0.27,
"learning_rate": 0.00037021063189568703,
"loss": 2.8722,
"theoretical_loss": 3.6939017430097074,
"tokens_seen": 881197056
},
{
"epoch": 0.27,
"learning_rate": 0.00037020060180541626,
"loss": 2.9037,
"theoretical_loss": 3.6938749853266044,
"tokens_seen": 881262592
},
{
"epoch": 0.27,
"learning_rate": 0.00037019057171514544,
"loss": 3.1357,
"theoretical_loss": 3.6938482301904054,
"tokens_seen": 881328128
},
{
"epoch": 0.27,
"learning_rate": 0.0003701805416248746,
"loss": 2.691,
"theoretical_loss": 3.693821477600677,
"tokens_seen": 881393664
},
{
"epoch": 0.27,
"objective/train/docs_used": 505856,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0436346530914307,
"objective/train/theoretical_loss": 3.693794727556988,
"objective/train/tokens_used": 901919200,
"theoretical_loss": 3.693794727556988,
"tokens_seen": 881459200
},
{
"epoch": 0.27,
"learning_rate": 0.0003701705115346038,
"loss": 2.9207,
"theoretical_loss": 3.693794727556988,
"tokens_seen": 881459200
},
{
"epoch": 0.27,
"learning_rate": 0.000370160481444333,
"loss": 2.7694,
"theoretical_loss": 3.693767980058907,
"tokens_seen": 881524736
},
{
"epoch": 0.27,
"learning_rate": 0.00037015045135406217,
"loss": 2.8024,
"theoretical_loss": 3.693741235106003,
"tokens_seen": 881590272
},
{
"epoch": 0.27,
"learning_rate": 0.0003701404212637914,
"loss": 2.6004,
"theoretical_loss": 3.6937144926978434,
"tokens_seen": 881655808
},
{
"epoch": 0.27,
"learning_rate": 0.00037013039117352053,
"loss": 2.9457,
"theoretical_loss": 3.693687752833998,
"tokens_seen": 881721344
},
{
"epoch": 0.27,
"learning_rate": 0.00037012036108324977,
"loss": 2.9344,
"theoretical_loss": 3.693661015514035,
"tokens_seen": 881786880
},
{
"epoch": 0.27,
"learning_rate": 0.00037011033099297895,
"loss": 2.9552,
"theoretical_loss": 3.693634280737524,
"tokens_seen": 881852416
},
{
"epoch": 0.27,
"learning_rate": 0.00037010030090270813,
"loss": 2.7691,
"theoretical_loss": 3.6936075485040334,
"tokens_seen": 881917952
},
{
"epoch": 0.27,
"learning_rate": 0.00037009027081243736,
"loss": 2.7163,
"theoretical_loss": 3.6935808188131327,
"tokens_seen": 881983488
},
{
"epoch": 0.27,
"learning_rate": 0.0003700802407221665,
"loss": 2.8381,
"theoretical_loss": 3.6935540916643914,
"tokens_seen": 882049024
},
{
"epoch": 0.27,
"learning_rate": 0.0003700702106318957,
"loss": 2.8436,
"theoretical_loss": 3.6935273670573787,
"tokens_seen": 882114560
},
{
"epoch": 0.27,
"learning_rate": 0.00037006018054162485,
"loss": 2.9531,
"theoretical_loss": 3.693500644991664,
"tokens_seen": 882180096
},
{
"epoch": 0.27,
"learning_rate": 0.0003700501504513541,
"loss": 2.7172,
"theoretical_loss": 3.693473925466816,
"tokens_seen": 882245632
},
{
"epoch": 0.27,
"learning_rate": 0.00037004012036108327,
"loss": 3.2923,
"theoretical_loss": 3.693447208482406,
"tokens_seen": 882311168
},
{
"epoch": 0.27,
"learning_rate": 0.00037003009027081245,
"loss": 2.6717,
"theoretical_loss": 3.693420494038003,
"tokens_seen": 882376704
},
{
"epoch": 0.27,
"learning_rate": 0.00037002006018054163,
"loss": 3.0844,
"theoretical_loss": 3.6933937821331764,
"tokens_seen": 882442240
},
{
"epoch": 0.27,
"learning_rate": 0.00037001003009027087,
"loss": 2.8243,
"theoretical_loss": 3.693367072767497,
"tokens_seen": 882507776
},
{
"epoch": 0.27,
"learning_rate": 0.00037,
"loss": 2.8318,
"theoretical_loss": 3.6933403659405344,
"tokens_seen": 882573312
},
{
"epoch": 0.27,
"learning_rate": 0.00036998996990972923,
"loss": 2.7628,
"theoretical_loss": 3.6933136616518594,
"tokens_seen": 882638848
},
{
"epoch": 0.27,
"learning_rate": 0.00036997993981945836,
"loss": 2.8754,
"theoretical_loss": 3.693286959901042,
"tokens_seen": 882704384
},
{
"epoch": 0.27,
"learning_rate": 0.0003699699097291876,
"loss": 2.8869,
"theoretical_loss": 3.6932602606876523,
"tokens_seen": 882769920
},
{
"epoch": 0.27,
"learning_rate": 0.0003699598796389168,
"loss": 2.763,
"theoretical_loss": 3.693233564011261,
"tokens_seen": 882835456
},
{
"epoch": 0.27,
"learning_rate": 0.00036994984954864595,
"loss": 3.0164,
"theoretical_loss": 3.693206869871439,
"tokens_seen": 882900992
},
{
"epoch": 0.27,
"learning_rate": 0.00036993981945837514,
"loss": 2.9286,
"theoretical_loss": 3.6931801782677565,
"tokens_seen": 882966528
},
{
"epoch": 0.27,
"learning_rate": 0.0003699297893681043,
"loss": 2.7663,
"theoretical_loss": 3.693153489199785,
"tokens_seen": 883032064
},
{
"epoch": 0.27,
"objective/train/docs_used": 506939,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7277796268463135,
"objective/train/theoretical_loss": 3.693126802667095,
"objective/train/tokens_used": 903557600,
"theoretical_loss": 3.693126802667095,
"tokens_seen": 883097600
},
{
"epoch": 0.27,
"learning_rate": 0.0003699197592778335,
"loss": 2.809,
"theoretical_loss": 3.693126802667095,
"tokens_seen": 883097600
},
{
"epoch": 0.27,
"learning_rate": 0.00036990972918756273,
"loss": 2.8276,
"theoretical_loss": 3.693100118669258,
"tokens_seen": 883163136
},
{
"epoch": 0.27,
"learning_rate": 0.00036989969909729186,
"loss": 2.8909,
"theoretical_loss": 3.693073437205844,
"tokens_seen": 883228672
},
{
"epoch": 0.27,
"learning_rate": 0.0003698896690070211,
"loss": 2.845,
"theoretical_loss": 3.6930467582764255,
"tokens_seen": 883294208
},
{
"epoch": 0.27,
"learning_rate": 0.0003698796389167503,
"loss": 2.696,
"theoretical_loss": 3.6930200818805736,
"tokens_seen": 883359744
},
{
"epoch": 0.27,
"learning_rate": 0.00036986960882647946,
"loss": 2.9337,
"theoretical_loss": 3.6929934080178595,
"tokens_seen": 883425280
},
{
"epoch": 0.27,
"learning_rate": 0.00036985957873620864,
"loss": 2.9079,
"theoretical_loss": 3.692966736687855,
"tokens_seen": 883490816
},
{
"epoch": 0.27,
"learning_rate": 0.0003698495486459378,
"loss": 3.0945,
"theoretical_loss": 3.692940067890131,
"tokens_seen": 883556352
},
{
"epoch": 0.27,
"learning_rate": 0.000369839518555667,
"loss": 2.949,
"theoretical_loss": 3.6929134016242604,
"tokens_seen": 883621888
},
{
"epoch": 0.27,
"learning_rate": 0.00036982948846539624,
"loss": 2.858,
"theoretical_loss": 3.6928867378898147,
"tokens_seen": 883687424
},
{
"epoch": 0.27,
"learning_rate": 0.00036981945837512536,
"loss": 2.9654,
"theoretical_loss": 3.6928600766863653,
"tokens_seen": 883752960
},
{
"epoch": 0.27,
"learning_rate": 0.0003698094282848546,
"loss": 2.814,
"theoretical_loss": 3.6928334180134854,
"tokens_seen": 883818496
},
{
"epoch": 0.27,
"learning_rate": 0.0003697993981945837,
"loss": 2.7366,
"theoretical_loss": 3.692806761870746,
"tokens_seen": 883884032
},
{
"epoch": 0.27,
"learning_rate": 0.00036978936810431296,
"loss": 2.9304,
"theoretical_loss": 3.69278010825772,
"tokens_seen": 883949568
},
{
"epoch": 0.27,
"learning_rate": 0.00036977933801404214,
"loss": 3.3269,
"theoretical_loss": 3.6927534571739797,
"tokens_seen": 884015104
},
{
"epoch": 0.27,
"learning_rate": 0.0003697693079237713,
"loss": 2.8818,
"theoretical_loss": 3.692726808619098,
"tokens_seen": 884080640
},
{
"epoch": 0.27,
"learning_rate": 0.0003697592778335005,
"loss": 2.7432,
"theoretical_loss": 3.6927001625926463,
"tokens_seen": 884146176
},
{
"epoch": 0.27,
"learning_rate": 0.0003697492477432297,
"loss": 3.0138,
"theoretical_loss": 3.692673519094199,
"tokens_seen": 884211712
},
{
"epoch": 0.27,
"learning_rate": 0.00036973921765295887,
"loss": 2.8948,
"theoretical_loss": 3.6926468781233277,
"tokens_seen": 884277248
},
{
"epoch": 0.27,
"learning_rate": 0.0003697291875626881,
"loss": 2.7694,
"theoretical_loss": 3.6926202396796057,
"tokens_seen": 884342784
},
{
"epoch": 0.27,
"learning_rate": 0.00036971915747241723,
"loss": 2.8063,
"theoretical_loss": 3.692593603762606,
"tokens_seen": 884408320
},
{
"epoch": 0.27,
"learning_rate": 0.00036970912738214646,
"loss": 2.99,
"theoretical_loss": 3.692566970371902,
"tokens_seen": 884473856
},
{
"epoch": 0.27,
"learning_rate": 0.00036969909729187565,
"loss": 2.7824,
"theoretical_loss": 3.692540339507066,
"tokens_seen": 884539392
},
{
"epoch": 0.27,
"learning_rate": 0.0003696890672016048,
"loss": 2.8356,
"theoretical_loss": 3.692513711167673,
"tokens_seen": 884604928
},
{
"epoch": 0.27,
"learning_rate": 0.000369679037111334,
"loss": 2.7873,
"theoretical_loss": 3.692487085353295,
"tokens_seen": 884670464
},
{
"epoch": 0.27,
"objective/train/docs_used": 507654,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7764241695404053,
"objective/train/theoretical_loss": 3.692460462063506,
"objective/train/tokens_used": 905196000,
"theoretical_loss": 3.692460462063506,
"tokens_seen": 884736000
},
{
"epoch": 0.27,
"learning_rate": 0.0003696690070210632,
"loss": 2.7896,
"theoretical_loss": 3.692460462063506,
"tokens_seen": 884736000
},
{
"epoch": 0.27,
"learning_rate": 0.00036965897693079237,
"loss": 2.9791,
"theoretical_loss": 3.6924338412978797,
"tokens_seen": 884801536
},
{
"epoch": 0.27,
"learning_rate": 0.0003696489468405216,
"loss": 2.9888,
"theoretical_loss": 3.6924072230559895,
"tokens_seen": 884867072
},
{
"epoch": 0.27,
"learning_rate": 0.00036963891675025073,
"loss": 2.7863,
"theoretical_loss": 3.6923806073374097,
"tokens_seen": 884932608
},
{
"epoch": 0.27,
"learning_rate": 0.00036962888665997997,
"loss": 3.2259,
"theoretical_loss": 3.692353994141714,
"tokens_seen": 884998144
},
{
"epoch": 0.27,
"learning_rate": 0.0003696188565697091,
"loss": 2.8733,
"theoretical_loss": 3.6923273834684767,
"tokens_seen": 885063680
},
{
"epoch": 0.27,
"learning_rate": 0.00036960882647943833,
"loss": 2.8413,
"theoretical_loss": 3.6923007753172716,
"tokens_seen": 885129216
},
{
"epoch": 0.27,
"learning_rate": 0.0003695987963891675,
"loss": 2.9133,
"theoretical_loss": 3.692274169687673,
"tokens_seen": 885194752
},
{
"epoch": 0.27,
"learning_rate": 0.0003695887662988967,
"loss": 2.7258,
"theoretical_loss": 3.692247566579256,
"tokens_seen": 885260288
},
{
"epoch": 0.27,
"learning_rate": 0.00036957873620862587,
"loss": 2.7134,
"theoretical_loss": 3.692220965991594,
"tokens_seen": 885325824
},
{
"epoch": 0.27,
"learning_rate": 0.00036956870611835505,
"loss": 2.9428,
"theoretical_loss": 3.6921943679242624,
"tokens_seen": 885391360
},
{
"epoch": 0.27,
"learning_rate": 0.00036955867602808424,
"loss": 2.9112,
"theoretical_loss": 3.692167772376836,
"tokens_seen": 885456896
},
{
"epoch": 0.27,
"learning_rate": 0.00036954864593781347,
"loss": 2.8838,
"theoretical_loss": 3.692141179348888,
"tokens_seen": 885522432
},
{
"epoch": 0.27,
"learning_rate": 0.0003695386158475426,
"loss": 2.7144,
"theoretical_loss": 3.6921145888399955,
"tokens_seen": 885587968
},
{
"epoch": 0.27,
"learning_rate": 0.00036952858575727183,
"loss": 2.9291,
"theoretical_loss": 3.692088000849732,
"tokens_seen": 885653504
},
{
"epoch": 0.27,
"learning_rate": 0.000369518555667001,
"loss": 2.6712,
"theoretical_loss": 3.692061415377673,
"tokens_seen": 885719040
},
{
"epoch": 0.27,
"learning_rate": 0.0003695085255767302,
"loss": 2.8139,
"theoretical_loss": 3.692034832423394,
"tokens_seen": 885784576
},
{
"epoch": 0.27,
"learning_rate": 0.0003694984954864594,
"loss": 2.9603,
"theoretical_loss": 3.6920082519864694,
"tokens_seen": 885850112
},
{
"epoch": 0.27,
"learning_rate": 0.00036948846539618856,
"loss": 3.0491,
"theoretical_loss": 3.6919816740664757,
"tokens_seen": 885915648
},
{
"epoch": 0.27,
"learning_rate": 0.00036947843530591774,
"loss": 3.1018,
"theoretical_loss": 3.691955098662988,
"tokens_seen": 885981184
},
{
"epoch": 0.27,
"learning_rate": 0.000369468405215647,
"loss": 2.7082,
"theoretical_loss": 3.6919285257755816,
"tokens_seen": 886046720
},
{
"epoch": 0.27,
"learning_rate": 0.0003694583751253761,
"loss": 3.1097,
"theoretical_loss": 3.6919019554038326,
"tokens_seen": 886112256
},
{
"epoch": 0.27,
"learning_rate": 0.00036944834503510534,
"loss": 2.9962,
"theoretical_loss": 3.6918753875473165,
"tokens_seen": 886177792
},
{
"epoch": 0.27,
"learning_rate": 0.00036943831494483446,
"loss": 2.8691,
"theoretical_loss": 3.6918488222056096,
"tokens_seen": 886243328
},
{
"epoch": 0.27,
"learning_rate": 0.0003694282848545637,
"loss": 2.8452,
"theoretical_loss": 3.691822259378287,
"tokens_seen": 886308864
},
{
"epoch": 0.27,
"objective/train/docs_used": 509172,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.591435432434082,
"objective/train/theoretical_loss": 3.6917956990649263,
"objective/train/tokens_used": 906834400,
"theoretical_loss": 3.6917956990649263,
"tokens_seen": 886374400
},
{
"epoch": 0.27,
"learning_rate": 0.0003694182547642929,
"loss": 2.8168,
"theoretical_loss": 3.6917956990649263,
"tokens_seen": 886374400
},
{
"epoch": 0.27,
"learning_rate": 0.00036940822467402206,
"loss": 2.9163,
"theoretical_loss": 3.6917691412651026,
"tokens_seen": 886439936
},
{
"epoch": 0.27,
"learning_rate": 0.00036939819458375124,
"loss": 2.7836,
"theoretical_loss": 3.6917425859783926,
"tokens_seen": 886505472
},
{
"epoch": 0.27,
"learning_rate": 0.0003693881644934805,
"loss": 2.7905,
"theoretical_loss": 3.691716033204373,
"tokens_seen": 886571008
},
{
"epoch": 0.27,
"learning_rate": 0.0003693781344032096,
"loss": 3.0825,
"theoretical_loss": 3.6916894829426194,
"tokens_seen": 886636544
},
{
"epoch": 0.27,
"learning_rate": 0.00036936810431293884,
"loss": 2.7517,
"theoretical_loss": 3.6916629351927095,
"tokens_seen": 886702080
},
{
"epoch": 0.27,
"learning_rate": 0.000369358074222668,
"loss": 2.8642,
"theoretical_loss": 3.69163638995422,
"tokens_seen": 886767616
},
{
"epoch": 0.27,
"learning_rate": 0.0003693480441323972,
"loss": 2.838,
"theoretical_loss": 3.6916098472267267,
"tokens_seen": 886833152
},
{
"epoch": 0.27,
"learning_rate": 0.00036933801404212644,
"loss": 2.7702,
"theoretical_loss": 3.691583307009807,
"tokens_seen": 886898688
},
{
"epoch": 0.27,
"learning_rate": 0.00036932798395185556,
"loss": 2.836,
"theoretical_loss": 3.6915567693030384,
"tokens_seen": 886964224
},
{
"epoch": 0.27,
"learning_rate": 0.0003693179538615848,
"loss": 2.8052,
"theoretical_loss": 3.691530234105998,
"tokens_seen": 887029760
},
{
"epoch": 0.27,
"learning_rate": 0.0003693079237713139,
"loss": 3.0104,
"theoretical_loss": 3.6915037014182626,
"tokens_seen": 887095296
},
{
"epoch": 0.27,
"learning_rate": 0.00036929789368104316,
"loss": 2.7261,
"theoretical_loss": 3.69147717123941,
"tokens_seen": 887160832
},
{
"epoch": 0.27,
"learning_rate": 0.00036928786359077234,
"loss": 2.9,
"theoretical_loss": 3.6914506435690173,
"tokens_seen": 887226368
},
{
"epoch": 0.27,
"learning_rate": 0.0003692778335005015,
"loss": 3.1903,
"theoretical_loss": 3.691424118406662,
"tokens_seen": 887291904
},
{
"epoch": 0.27,
"learning_rate": 0.0003692678034102307,
"loss": 2.7584,
"theoretical_loss": 3.691397595751922,
"tokens_seen": 887357440
},
{
"epoch": 0.27,
"learning_rate": 0.0003692577733199599,
"loss": 2.8061,
"theoretical_loss": 3.6913710756043754,
"tokens_seen": 887422976
},
{
"epoch": 0.27,
"learning_rate": 0.00036924774322968907,
"loss": 2.9605,
"theoretical_loss": 3.6913445579635997,
"tokens_seen": 887488512
},
{
"epoch": 0.27,
"learning_rate": 0.0003692377131394183,
"loss": 2.9533,
"theoretical_loss": 3.691318042829172,
"tokens_seen": 887554048
},
{
"epoch": 0.27,
"learning_rate": 0.00036922768304914743,
"loss": 2.7638,
"theoretical_loss": 3.691291530200672,
"tokens_seen": 887619584
},
{
"epoch": 0.27,
"learning_rate": 0.00036921765295887666,
"loss": 2.8661,
"theoretical_loss": 3.691265020077677,
"tokens_seen": 887685120
},
{
"epoch": 0.27,
"learning_rate": 0.00036920762286860585,
"loss": 2.709,
"theoretical_loss": 3.6912385124597655,
"tokens_seen": 887750656
},
{
"epoch": 0.27,
"learning_rate": 0.000369197592778335,
"loss": 2.6437,
"theoretical_loss": 3.6912120073465156,
"tokens_seen": 887816192
},
{
"epoch": 0.27,
"learning_rate": 0.0003691875626880642,
"loss": 2.7857,
"theoretical_loss": 3.6911855047375055,
"tokens_seen": 887881728
},
{
"epoch": 0.27,
"learning_rate": 0.0003691775325977934,
"loss": 2.6542,
"theoretical_loss": 3.691159004632315,
"tokens_seen": 887947264
},
{
"epoch": 0.27,
"objective/train/docs_used": 509747,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.058424949645996,
"objective/train/theoretical_loss": 3.691132507030521,
"objective/train/tokens_used": 908472800,
"theoretical_loss": 3.691132507030521,
"tokens_seen": 888012800
},
{
"epoch": 0.27,
"learning_rate": 0.00036916750250752257,
"loss": 2.6919,
"theoretical_loss": 3.691132507030521,
"tokens_seen": 888012800
},
{
"epoch": 0.27,
"learning_rate": 0.0003691574724172518,
"loss": 3.1712,
"theoretical_loss": 3.691106011931704,
"tokens_seen": 888078336
},
{
"epoch": 0.27,
"learning_rate": 0.00036914744232698093,
"loss": 2.7696,
"theoretical_loss": 3.6910795193354415,
"tokens_seen": 888143872
},
{
"epoch": 0.27,
"learning_rate": 0.00036913741223671017,
"loss": 2.8276,
"theoretical_loss": 3.6910530292413135,
"tokens_seen": 888209408
},
{
"epoch": 0.27,
"learning_rate": 0.0003691273821464393,
"loss": 2.9819,
"theoretical_loss": 3.691026541648899,
"tokens_seen": 888274944
},
{
"epoch": 0.27,
"learning_rate": 0.00036911735205616853,
"loss": 2.7368,
"theoretical_loss": 3.691000056557777,
"tokens_seen": 888340480
},
{
"epoch": 0.27,
"learning_rate": 0.0003691073219658977,
"loss": 2.889,
"theoretical_loss": 3.6909735739675265,
"tokens_seen": 888406016
},
{
"epoch": 0.27,
"learning_rate": 0.0003690972918756269,
"loss": 3.1258,
"theoretical_loss": 3.690947093877727,
"tokens_seen": 888471552
},
{
"epoch": 0.27,
"learning_rate": 0.0003690872617853561,
"loss": 2.8085,
"theoretical_loss": 3.690920616287958,
"tokens_seen": 888537088
},
{
"epoch": 0.27,
"learning_rate": 0.00036907723169508525,
"loss": 2.8923,
"theoretical_loss": 3.6908941411977993,
"tokens_seen": 888602624
},
{
"epoch": 0.27,
"learning_rate": 0.00036906720160481444,
"loss": 2.6124,
"theoretical_loss": 3.6908676686068307,
"tokens_seen": 888668160
},
{
"epoch": 0.27,
"learning_rate": 0.00036905717151454367,
"loss": 2.7562,
"theoretical_loss": 3.690841198514632,
"tokens_seen": 888733696
},
{
"epoch": 0.27,
"learning_rate": 0.0003690471414242728,
"loss": 2.9326,
"theoretical_loss": 3.6908147309207826,
"tokens_seen": 888799232
},
{
"epoch": 0.27,
"learning_rate": 0.00036903711133400203,
"loss": 2.9735,
"theoretical_loss": 3.690788265824863,
"tokens_seen": 888864768
},
{
"epoch": 0.27,
"learning_rate": 0.0003690270812437312,
"loss": 2.7222,
"theoretical_loss": 3.6907618032264526,
"tokens_seen": 888930304
},
{
"epoch": 0.27,
"learning_rate": 0.0003690170511534604,
"loss": 2.7085,
"theoretical_loss": 3.6907353431251324,
"tokens_seen": 888995840
},
{
"epoch": 0.27,
"learning_rate": 0.0003690070210631896,
"loss": 2.8201,
"theoretical_loss": 3.6907088855204826,
"tokens_seen": 889061376
},
{
"epoch": 0.27,
"learning_rate": 0.00036899699097291876,
"loss": 2.8149,
"theoretical_loss": 3.6906824304120835,
"tokens_seen": 889126912
},
{
"epoch": 0.27,
"learning_rate": 0.00036898696088264794,
"loss": 2.7414,
"theoretical_loss": 3.6906559777995156,
"tokens_seen": 889192448
},
{
"epoch": 0.27,
"learning_rate": 0.0003689769307923772,
"loss": 2.9731,
"theoretical_loss": 3.690629527682359,
"tokens_seen": 889257984
},
{
"epoch": 0.27,
"learning_rate": 0.0003689669007021063,
"loss": 2.9211,
"theoretical_loss": 3.690603080060195,
"tokens_seen": 889323520
},
{
"epoch": 0.27,
"learning_rate": 0.00036895687061183554,
"loss": 2.8549,
"theoretical_loss": 3.690576634932605,
"tokens_seen": 889389056
},
{
"epoch": 0.27,
"learning_rate": 0.00036894684052156466,
"loss": 2.7699,
"theoretical_loss": 3.6905501922991686,
"tokens_seen": 889454592
},
{
"epoch": 0.27,
"learning_rate": 0.0003689368104312939,
"loss": 2.8798,
"theoretical_loss": 3.690523752159468,
"tokens_seen": 889520128
},
{
"epoch": 0.27,
"learning_rate": 0.0003689267803410231,
"loss": 2.8154,
"theoretical_loss": 3.690497314513083,
"tokens_seen": 889585664
},
{
"epoch": 0.27,
"objective/train/docs_used": 511200,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.968623161315918,
"objective/train/theoretical_loss": 3.690470879359596,
"objective/train/tokens_used": 910111200,
"theoretical_loss": 3.690470879359596,
"tokens_seen": 889651200
},
{
"epoch": 0.27,
"learning_rate": 0.00036891675025075226,
"loss": 2.8038,
"theoretical_loss": 3.690470879359596,
"tokens_seen": 889651200
},
{
"epoch": 0.27,
"learning_rate": 0.00036890672016048144,
"loss": 2.6796,
"theoretical_loss": 3.6904444466985877,
"tokens_seen": 889716736
},
{
"epoch": 0.27,
"learning_rate": 0.0003688966900702107,
"loss": 2.8343,
"theoretical_loss": 3.69041801652964,
"tokens_seen": 889782272
},
{
"epoch": 0.27,
"learning_rate": 0.0003688866599799398,
"loss": 2.8876,
"theoretical_loss": 3.6903915888523344,
"tokens_seen": 889847808
},
{
"epoch": 0.27,
"learning_rate": 0.00036887662988966904,
"loss": 2.7446,
"theoretical_loss": 3.690365163666252,
"tokens_seen": 889913344
},
{
"epoch": 0.27,
"learning_rate": 0.00036886659979939817,
"loss": 2.7522,
"theoretical_loss": 3.690338740970975,
"tokens_seen": 889978880
},
{
"epoch": 0.27,
"learning_rate": 0.0003688565697091274,
"loss": 2.8593,
"theoretical_loss": 3.6903123207660844,
"tokens_seen": 890044416
},
{
"epoch": 0.27,
"learning_rate": 0.0003688465396188566,
"loss": 3.2997,
"theoretical_loss": 3.6902859030511634,
"tokens_seen": 890109952
},
{
"epoch": 0.27,
"learning_rate": 0.00036883650952858576,
"loss": 2.8372,
"theoretical_loss": 3.6902594878257933,
"tokens_seen": 890175488
},
{
"epoch": 0.27,
"learning_rate": 0.00036882647943831494,
"loss": 2.8746,
"theoretical_loss": 3.6902330750895564,
"tokens_seen": 890241024
},
{
"epoch": 0.27,
"learning_rate": 0.0003688164493480441,
"loss": 2.6497,
"theoretical_loss": 3.6902066648420346,
"tokens_seen": 890306560
},
{
"epoch": 0.27,
"learning_rate": 0.0003688064192577733,
"loss": 2.9106,
"theoretical_loss": 3.690180257082811,
"tokens_seen": 890372096
},
{
"epoch": 0.27,
"learning_rate": 0.00036879638916750254,
"loss": 2.9607,
"theoretical_loss": 3.690153851811467,
"tokens_seen": 890437632
},
{
"epoch": 0.27,
"learning_rate": 0.00036878635907723167,
"loss": 2.6183,
"theoretical_loss": 3.690127449027586,
"tokens_seen": 890503168
},
{
"epoch": 0.27,
"learning_rate": 0.0003687763289869609,
"loss": 2.8109,
"theoretical_loss": 3.6901010487307504,
"tokens_seen": 890568704
},
{
"epoch": 0.27,
"learning_rate": 0.00036876629889669003,
"loss": 2.871,
"theoretical_loss": 3.6900746509205424,
"tokens_seen": 890634240
},
{
"epoch": 0.27,
"learning_rate": 0.00036875626880641927,
"loss": 2.8275,
"theoretical_loss": 3.6900482555965457,
"tokens_seen": 890699776
},
{
"epoch": 0.27,
"learning_rate": 0.00036874623871614845,
"loss": 2.7973,
"theoretical_loss": 3.6900218627583428,
"tokens_seen": 890765312
},
{
"epoch": 0.27,
"learning_rate": 0.00036873620862587763,
"loss": 2.619,
"theoretical_loss": 3.689995472405516,
"tokens_seen": 890830848
},
{
"epoch": 0.27,
"learning_rate": 0.0003687261785356068,
"loss": 2.9688,
"theoretical_loss": 3.6899690845376503,
"tokens_seen": 890896384
},
{
"epoch": 0.27,
"learning_rate": 0.00036871614844533605,
"loss": 2.7628,
"theoretical_loss": 3.689942699154327,
"tokens_seen": 890961920
},
{
"epoch": 0.27,
"learning_rate": 0.00036870611835506517,
"loss": 2.8281,
"theoretical_loss": 3.6899163162551307,
"tokens_seen": 891027456
},
{
"epoch": 0.27,
"learning_rate": 0.0003686960882647944,
"loss": 2.8418,
"theoretical_loss": 3.689889935839645,
"tokens_seen": 891092992
},
{
"epoch": 0.27,
"learning_rate": 0.00036868605817452353,
"loss": 2.6142,
"theoretical_loss": 3.6898635579074517,
"tokens_seen": 891158528
},
{
"epoch": 0.27,
"learning_rate": 0.00036867602808425277,
"loss": 2.8782,
"theoretical_loss": 3.6898371824581364,
"tokens_seen": 891224064
},
{
"epoch": 0.27,
"objective/train/docs_used": 511885,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8643696308135986,
"objective/train/theoretical_loss": 3.6898108094912816,
"objective/train/tokens_used": 911749600,
"theoretical_loss": 3.6898108094912816,
"tokens_seen": 891289600
},
{
"epoch": 0.27,
"learning_rate": 0.00036866599799398195,
"loss": 2.8802,
"theoretical_loss": 3.6898108094912816,
"tokens_seen": 891289600
},
{
"epoch": 0.27,
"learning_rate": 0.00036865596790371113,
"loss": 2.7226,
"theoretical_loss": 3.6897844390064716,
"tokens_seen": 891355136
},
{
"epoch": 0.27,
"learning_rate": 0.0003686459378134403,
"loss": 2.7362,
"theoretical_loss": 3.6897580710032907,
"tokens_seen": 891420672
},
{
"epoch": 0.27,
"learning_rate": 0.0003686359077231695,
"loss": 2.7629,
"theoretical_loss": 3.6897317054813223,
"tokens_seen": 891486208
},
{
"epoch": 0.27,
"learning_rate": 0.0003686258776328987,
"loss": 2.864,
"theoretical_loss": 3.689705342440151,
"tokens_seen": 891551744
},
{
"epoch": 0.27,
"learning_rate": 0.0003686158475426279,
"loss": 2.7218,
"theoretical_loss": 3.6896789818793607,
"tokens_seen": 891617280
},
{
"epoch": 0.27,
"learning_rate": 0.0003686058174523571,
"loss": 2.7221,
"theoretical_loss": 3.689652623798536,
"tokens_seen": 891682816
},
{
"epoch": 0.27,
"learning_rate": 0.0003685957873620863,
"loss": 2.5585,
"theoretical_loss": 3.689626268197261,
"tokens_seen": 891748352
},
{
"epoch": 0.27,
"learning_rate": 0.00036858575727181545,
"loss": 2.94,
"theoretical_loss": 3.689599915075121,
"tokens_seen": 891813888
},
{
"epoch": 0.27,
"learning_rate": 0.00036857572718154464,
"loss": 3.0991,
"theoretical_loss": 3.6895735644317,
"tokens_seen": 891879424
},
{
"epoch": 0.27,
"learning_rate": 0.00036856569709127387,
"loss": 2.7652,
"theoretical_loss": 3.6895472162665826,
"tokens_seen": 891944960
},
{
"epoch": 0.27,
"learning_rate": 0.000368555667001003,
"loss": 2.9931,
"theoretical_loss": 3.6895208705793534,
"tokens_seen": 892010496
},
{
"epoch": 0.27,
"learning_rate": 0.00036854563691073223,
"loss": 2.805,
"theoretical_loss": 3.6894945273695985,
"tokens_seen": 892076032
},
{
"epoch": 0.27,
"learning_rate": 0.0003685356068204614,
"loss": 2.903,
"theoretical_loss": 3.6894681866369026,
"tokens_seen": 892141568
},
{
"epoch": 0.27,
"learning_rate": 0.0003685255767301906,
"loss": 2.9022,
"theoretical_loss": 3.6894418483808504,
"tokens_seen": 892207104
},
{
"epoch": 0.27,
"learning_rate": 0.0003685155466399198,
"loss": 2.9142,
"theoretical_loss": 3.689415512601027,
"tokens_seen": 892272640
},
{
"epoch": 0.27,
"learning_rate": 0.00036850551654964896,
"loss": 3.0771,
"theoretical_loss": 3.6893891792970184,
"tokens_seen": 892338176
},
{
"epoch": 0.27,
"learning_rate": 0.00036849548645937814,
"loss": 2.9371,
"theoretical_loss": 3.6893628484684093,
"tokens_seen": 892403712
},
{
"epoch": 0.27,
"learning_rate": 0.0003684854563691074,
"loss": 2.8764,
"theoretical_loss": 3.689336520114786,
"tokens_seen": 892469248
},
{
"epoch": 0.27,
"learning_rate": 0.0003684754262788365,
"loss": 2.7767,
"theoretical_loss": 3.689310194235733,
"tokens_seen": 892534784
},
{
"epoch": 0.27,
"learning_rate": 0.00036846539618856574,
"loss": 2.9238,
"theoretical_loss": 3.6892838708308378,
"tokens_seen": 892600320
},
{
"epoch": 0.27,
"learning_rate": 0.00036845536609829486,
"loss": 2.7216,
"theoretical_loss": 3.6892575498996845,
"tokens_seen": 892665856
},
{
"epoch": 0.27,
"learning_rate": 0.0003684453360080241,
"loss": 2.5306,
"theoretical_loss": 3.6892312314418607,
"tokens_seen": 892731392
},
{
"epoch": 0.27,
"learning_rate": 0.0003684353059177533,
"loss": 2.8682,
"theoretical_loss": 3.689204915456951,
"tokens_seen": 892796928
},
{
"epoch": 0.27,
"learning_rate": 0.00036842527582748246,
"loss": 2.5644,
"theoretical_loss": 3.689178601944542,
"tokens_seen": 892862464
},
{
"epoch": 0.27,
"objective/train/docs_used": 512575,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.900513172149658,
"objective/train/theoretical_loss": 3.68915229090422,
"objective/train/tokens_used": 913388000,
"theoretical_loss": 3.68915229090422,
"tokens_seen": 892928000
},
{
"epoch": 0.27,
"learning_rate": 0.00036841524573721164,
"loss": 2.9675,
"theoretical_loss": 3.68915229090422,
"tokens_seen": 892928000
},
{
"epoch": 0.27,
"learning_rate": 0.0003684052156469409,
"loss": 2.7512,
"theoretical_loss": 3.6891259823355718,
"tokens_seen": 892993536
},
{
"epoch": 0.27,
"learning_rate": 0.00036839518555667,
"loss": 2.7546,
"theoretical_loss": 3.689099676238183,
"tokens_seen": 893059072
},
{
"epoch": 0.27,
"learning_rate": 0.00036838515546639924,
"loss": 2.9604,
"theoretical_loss": 3.6890733726116407,
"tokens_seen": 893124608
},
{
"epoch": 0.27,
"learning_rate": 0.00036837512537612837,
"loss": 2.8949,
"theoretical_loss": 3.6890470714555317,
"tokens_seen": 893190144
},
{
"epoch": 0.27,
"learning_rate": 0.0003683650952858576,
"loss": 2.3633,
"theoretical_loss": 3.6890207727694415,
"tokens_seen": 893255680
},
{
"epoch": 0.27,
"learning_rate": 0.0003683550651955868,
"loss": 2.8377,
"theoretical_loss": 3.688994476552959,
"tokens_seen": 893321216
},
{
"epoch": 0.27,
"learning_rate": 0.00036834503510531596,
"loss": 2.7521,
"theoretical_loss": 3.6889681828056693,
"tokens_seen": 893386752
},
{
"epoch": 0.27,
"learning_rate": 0.00036833500501504514,
"loss": 2.7764,
"theoretical_loss": 3.6889418915271603,
"tokens_seen": 893452288
},
{
"epoch": 0.27,
"learning_rate": 0.0003683249749247743,
"loss": 2.8684,
"theoretical_loss": 3.688915602717019,
"tokens_seen": 893517824
},
{
"epoch": 0.27,
"learning_rate": 0.0003683149448345035,
"loss": 2.8537,
"theoretical_loss": 3.688889316374833,
"tokens_seen": 893583360
},
{
"epoch": 0.27,
"learning_rate": 0.00036830491474423274,
"loss": 2.7952,
"theoretical_loss": 3.688863032500189,
"tokens_seen": 893648896
},
{
"epoch": 0.27,
"learning_rate": 0.00036829488465396187,
"loss": 2.6483,
"theoretical_loss": 3.6888367510926745,
"tokens_seen": 893714432
},
{
"epoch": 0.27,
"learning_rate": 0.0003682848545636911,
"loss": 2.788,
"theoretical_loss": 3.6888104721518777,
"tokens_seen": 893779968
},
{
"epoch": 0.27,
"learning_rate": 0.00036827482447342023,
"loss": 2.9144,
"theoretical_loss": 3.6887841956773855,
"tokens_seen": 893845504
},
{
"epoch": 0.27,
"learning_rate": 0.00036826479438314947,
"loss": 2.636,
"theoretical_loss": 3.688757921668786,
"tokens_seen": 893911040
},
{
"epoch": 0.27,
"learning_rate": 0.00036825476429287865,
"loss": 2.5928,
"theoretical_loss": 3.688731650125667,
"tokens_seen": 893976576
},
{
"epoch": 0.27,
"learning_rate": 0.00036824473420260783,
"loss": 2.7913,
"theoretical_loss": 3.6887053810476162,
"tokens_seen": 894042112
},
{
"epoch": 0.27,
"learning_rate": 0.000368234704112337,
"loss": 2.7643,
"theoretical_loss": 3.6886791144342217,
"tokens_seen": 894107648
},
{
"epoch": 0.27,
"learning_rate": 0.00036822467402206625,
"loss": 2.8219,
"theoretical_loss": 3.688652850285072,
"tokens_seen": 894173184
},
{
"epoch": 0.27,
"learning_rate": 0.00036821464393179537,
"loss": 2.5441,
"theoretical_loss": 3.688626588599755,
"tokens_seen": 894238720
},
{
"epoch": 0.27,
"learning_rate": 0.0003682046138415246,
"loss": 3.017,
"theoretical_loss": 3.6886003293778584,
"tokens_seen": 894304256
},
{
"epoch": 0.27,
"learning_rate": 0.00036819458375125373,
"loss": 2.878,
"theoretical_loss": 3.688574072618972,
"tokens_seen": 894369792
},
{
"epoch": 0.27,
"learning_rate": 0.00036818455366098297,
"loss": 2.7478,
"theoretical_loss": 3.6885478183226836,
"tokens_seen": 894435328
},
{
"epoch": 0.27,
"learning_rate": 0.00036817452357071215,
"loss": 2.8723,
"theoretical_loss": 3.688521566488582,
"tokens_seen": 894500864
},
{
"epoch": 0.27,
"objective/train/docs_used": 513630,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.845224618911743,
"objective/train/theoretical_loss": 3.6884953171162556,
"objective/train/tokens_used": 915026400,
"theoretical_loss": 3.6884953171162556,
"tokens_seen": 894566400
},
{
"epoch": 0.27,
"learning_rate": 0.00036816449348044133,
"loss": 3.0862,
"theoretical_loss": 3.6884953171162556,
"tokens_seen": 894566400
},
{
"epoch": 0.27,
"learning_rate": 0.0003681544633901705,
"loss": 2.8544,
"theoretical_loss": 3.688469070205293,
"tokens_seen": 894631936
},
{
"epoch": 0.27,
"learning_rate": 0.0003681444332998997,
"loss": 2.8469,
"theoretical_loss": 3.688442825755284,
"tokens_seen": 894697472
},
{
"epoch": 0.27,
"learning_rate": 0.0003681344032096289,
"loss": 2.7415,
"theoretical_loss": 3.688416583765817,
"tokens_seen": 894763008
},
{
"epoch": 0.27,
"learning_rate": 0.0003681243731193581,
"loss": 2.7954,
"theoretical_loss": 3.688390344236481,
"tokens_seen": 894828544
},
{
"epoch": 0.27,
"learning_rate": 0.00036811434302908724,
"loss": 2.486,
"theoretical_loss": 3.688364107166866,
"tokens_seen": 894894080
},
{
"epoch": 0.27,
"learning_rate": 0.0003681043129388165,
"loss": 2.3997,
"theoretical_loss": 3.6883378725565605,
"tokens_seen": 894959616
},
{
"epoch": 0.27,
"learning_rate": 0.0003680942828485456,
"loss": 2.6915,
"theoretical_loss": 3.6883116404051544,
"tokens_seen": 895025152
},
{
"epoch": 0.27,
"learning_rate": 0.00036808425275827484,
"loss": 2.4102,
"theoretical_loss": 3.6882854107122363,
"tokens_seen": 895090688
},
{
"epoch": 0.27,
"learning_rate": 0.000368074222668004,
"loss": 2.9228,
"theoretical_loss": 3.6882591834773972,
"tokens_seen": 895156224
},
{
"epoch": 0.27,
"learning_rate": 0.0003680641925777332,
"loss": 2.7083,
"theoretical_loss": 3.688232958700226,
"tokens_seen": 895221760
},
{
"epoch": 0.27,
"learning_rate": 0.0003680541624874624,
"loss": 2.8069,
"theoretical_loss": 3.688206736380313,
"tokens_seen": 895287296
},
{
"epoch": 0.27,
"learning_rate": 0.0003680441323971916,
"loss": 2.6752,
"theoretical_loss": 3.688180516517247,
"tokens_seen": 895352832
},
{
"epoch": 0.27,
"learning_rate": 0.00036803410230692074,
"loss": 2.7655,
"theoretical_loss": 3.6881542991106193,
"tokens_seen": 895418368
},
{
"epoch": 0.27,
"learning_rate": 0.00036802407221665,
"loss": 2.9023,
"theoretical_loss": 3.6881280841600192,
"tokens_seen": 895483904
},
{
"epoch": 0.27,
"learning_rate": 0.0003680140421263791,
"loss": 2.7868,
"theoretical_loss": 3.6881018716650376,
"tokens_seen": 895549440
},
{
"epoch": 0.27,
"learning_rate": 0.00036800401203610834,
"loss": 2.8197,
"theoretical_loss": 3.688075661625264,
"tokens_seen": 895614976
},
{
"epoch": 0.27,
"learning_rate": 0.0003679939819458375,
"loss": 2.9352,
"theoretical_loss": 3.6880494540402893,
"tokens_seen": 895680512
},
{
"epoch": 0.27,
"learning_rate": 0.0003679839518555667,
"loss": 3.0127,
"theoretical_loss": 3.688023248909704,
"tokens_seen": 895746048
},
{
"epoch": 0.27,
"learning_rate": 0.0003679739217652959,
"loss": 2.8546,
"theoretical_loss": 3.6879970462330975,
"tokens_seen": 895811584
},
{
"epoch": 0.27,
"learning_rate": 0.00036796389167502506,
"loss": 2.7313,
"theoretical_loss": 3.687970846010063,
"tokens_seen": 895877120
},
{
"epoch": 0.27,
"learning_rate": 0.00036795386158475424,
"loss": 2.885,
"theoretical_loss": 3.6879446482401885,
"tokens_seen": 895942656
},
{
"epoch": 0.27,
"learning_rate": 0.0003679438314944835,
"loss": 2.6131,
"theoretical_loss": 3.6879184529230673,
"tokens_seen": 896008192
},
{
"epoch": 0.27,
"learning_rate": 0.0003679338014042126,
"loss": 2.7754,
"theoretical_loss": 3.687892260058289,
"tokens_seen": 896073728
},
{
"epoch": 0.27,
"learning_rate": 0.00036792377131394184,
"loss": 3.0305,
"theoretical_loss": 3.6878660696454446,
"tokens_seen": 896139264
},
{
"epoch": 0.27,
"objective/train/docs_used": 514378,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.814505100250244,
"objective/train/theoretical_loss": 3.6878398816841265,
"objective/train/tokens_used": 916664800,
"theoretical_loss": 3.6878398816841265,
"tokens_seen": 896204800
},
{
"epoch": 0.27,
"learning_rate": 0.00036791374122367097,
"loss": 2.8531,
"theoretical_loss": 3.6878398816841265,
"tokens_seen": 896204800
},
{
"epoch": 0.27,
"learning_rate": 0.0003679037111334002,
"loss": 2.7559,
"theoretical_loss": 3.6878136961739245,
"tokens_seen": 896270336
},
{
"epoch": 0.27,
"learning_rate": 0.0003678936810431294,
"loss": 2.8415,
"theoretical_loss": 3.687787513114431,
"tokens_seen": 896335872
},
{
"epoch": 0.27,
"learning_rate": 0.00036788365095285857,
"loss": 2.9554,
"theoretical_loss": 3.6877613325052367,
"tokens_seen": 896401408
},
{
"epoch": 0.27,
"learning_rate": 0.00036787362086258775,
"loss": 2.6277,
"theoretical_loss": 3.687735154345934,
"tokens_seen": 896466944
},
{
"epoch": 0.27,
"learning_rate": 0.000367863590772317,
"loss": 2.577,
"theoretical_loss": 3.687708978636114,
"tokens_seen": 896532480
},
{
"epoch": 0.27,
"learning_rate": 0.00036785356068204616,
"loss": 2.9656,
"theoretical_loss": 3.6876828053753687,
"tokens_seen": 896598016
},
{
"epoch": 0.27,
"learning_rate": 0.00036784353059177535,
"loss": 2.7697,
"theoretical_loss": 3.68765663456329,
"tokens_seen": 896663552
},
{
"epoch": 0.27,
"learning_rate": 0.0003678335005015045,
"loss": 2.681,
"theoretical_loss": 3.6876304661994697,
"tokens_seen": 896729088
},
{
"epoch": 0.27,
"learning_rate": 0.0003678234704112337,
"loss": 2.9727,
"theoretical_loss": 3.6876043002834997,
"tokens_seen": 896794624
},
{
"epoch": 0.27,
"learning_rate": 0.00036781344032096294,
"loss": 2.6717,
"theoretical_loss": 3.687578136814973,
"tokens_seen": 896860160
},
{
"epoch": 0.27,
"learning_rate": 0.00036780341023069207,
"loss": 2.9856,
"theoretical_loss": 3.6875519757934807,
"tokens_seen": 896925696
},
{
"epoch": 0.27,
"learning_rate": 0.0003677933801404213,
"loss": 2.697,
"theoretical_loss": 3.6875258172186163,
"tokens_seen": 896991232
},
{
"epoch": 0.27,
"learning_rate": 0.00036778335005015043,
"loss": 2.7576,
"theoretical_loss": 3.687499661089971,
"tokens_seen": 897056768
},
{
"epoch": 0.27,
"learning_rate": 0.00036777331995987967,
"loss": 2.646,
"theoretical_loss": 3.6874735074071383,
"tokens_seen": 897122304
},
{
"epoch": 0.27,
"learning_rate": 0.00036776328986960885,
"loss": 2.536,
"theoretical_loss": 3.687447356169711,
"tokens_seen": 897187840
},
{
"epoch": 0.27,
"learning_rate": 0.00036775325977933803,
"loss": 2.6163,
"theoretical_loss": 3.687421207377281,
"tokens_seen": 897253376
},
{
"epoch": 0.27,
"learning_rate": 0.0003677432296890672,
"loss": 2.5614,
"theoretical_loss": 3.6873950610294415,
"tokens_seen": 897318912
},
{
"epoch": 0.27,
"learning_rate": 0.00036773319959879645,
"loss": 2.99,
"theoretical_loss": 3.687368917125786,
"tokens_seen": 897384448
},
{
"epoch": 0.27,
"learning_rate": 0.00036772316950852557,
"loss": 3.1565,
"theoretical_loss": 3.6873427756659067,
"tokens_seen": 897449984
},
{
"epoch": 0.27,
"learning_rate": 0.0003677131394182548,
"loss": 2.5717,
"theoretical_loss": 3.6873166366493972,
"tokens_seen": 897515520
},
{
"epoch": 0.27,
"learning_rate": 0.00036770310932798393,
"loss": 2.8417,
"theoretical_loss": 3.687290500075851,
"tokens_seen": 897581056
},
{
"epoch": 0.27,
"learning_rate": 0.00036769307923771317,
"loss": 2.9928,
"theoretical_loss": 3.687264365944861,
"tokens_seen": 897646592
},
{
"epoch": 0.27,
"learning_rate": 0.00036768304914744235,
"loss": 2.7528,
"theoretical_loss": 3.6872382342560206,
"tokens_seen": 897712128
},
{
"epoch": 0.27,
"learning_rate": 0.00036767301905717153,
"loss": 2.7993,
"theoretical_loss": 3.687212105008923,
"tokens_seen": 897777664
},
{
"epoch": 0.27,
"objective/train/docs_used": 515835,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.866816759109497,
"objective/train/theoretical_loss": 3.6871859782031624,
"objective/train/tokens_used": 918303200,
"theoretical_loss": 3.6871859782031624,
"tokens_seen": 897843200
},
{
"epoch": 0.27,
"learning_rate": 0.0003676629889669007,
"loss": 2.9239,
"theoretical_loss": 3.6871859782031624,
"tokens_seen": 897843200
},
{
"epoch": 0.27,
"learning_rate": 0.0003676529588766299,
"loss": 2.6791,
"theoretical_loss": 3.687159853838333,
"tokens_seen": 897908736
},
{
"epoch": 0.27,
"learning_rate": 0.0003676429287863591,
"loss": 2.7364,
"theoretical_loss": 3.6871337319140274,
"tokens_seen": 897974272
},
{
"epoch": 0.27,
"learning_rate": 0.0003676328986960883,
"loss": 2.8648,
"theoretical_loss": 3.6871076124298403,
"tokens_seen": 898039808
},
{
"epoch": 0.27,
"learning_rate": 0.00036762286860581744,
"loss": 2.8609,
"theoretical_loss": 3.6870814953853657,
"tokens_seen": 898105344
},
{
"epoch": 0.27,
"learning_rate": 0.0003676128385155467,
"loss": 2.8412,
"theoretical_loss": 3.6870553807801976,
"tokens_seen": 898170880
},
{
"epoch": 0.27,
"learning_rate": 0.0003676028084252758,
"loss": 2.5308,
"theoretical_loss": 3.68702926861393,
"tokens_seen": 898236416
},
{
"epoch": 0.27,
"learning_rate": 0.00036759277833500504,
"loss": 2.9146,
"theoretical_loss": 3.687003158886158,
"tokens_seen": 898301952
},
{
"epoch": 0.27,
"learning_rate": 0.0003675827482447342,
"loss": 2.731,
"theoretical_loss": 3.6869770515964744,
"tokens_seen": 898367488
},
{
"epoch": 0.27,
"learning_rate": 0.0003675727181544634,
"loss": 2.8668,
"theoretical_loss": 3.6869509467444757,
"tokens_seen": 898433024
},
{
"epoch": 0.27,
"learning_rate": 0.0003675626880641926,
"loss": 2.8757,
"theoretical_loss": 3.686924844329755,
"tokens_seen": 898498560
},
{
"epoch": 0.27,
"learning_rate": 0.0003675526579739218,
"loss": 2.7673,
"theoretical_loss": 3.686898744351907,
"tokens_seen": 898564096
},
{
"epoch": 0.27,
"learning_rate": 0.00036754262788365094,
"loss": 3.1246,
"theoretical_loss": 3.6868726468105275,
"tokens_seen": 898629632
},
{
"epoch": 0.27,
"learning_rate": 0.0003675325977933802,
"loss": 2.7947,
"theoretical_loss": 3.6868465517052105,
"tokens_seen": 898695168
},
{
"epoch": 0.27,
"learning_rate": 0.0003675225677031093,
"loss": 2.7493,
"theoretical_loss": 3.686820459035552,
"tokens_seen": 898760704
},
{
"epoch": 0.27,
"learning_rate": 0.00036751253761283854,
"loss": 2.7911,
"theoretical_loss": 3.686794368801146,
"tokens_seen": 898826240
},
{
"epoch": 0.27,
"learning_rate": 0.0003675025075225677,
"loss": 2.8207,
"theoretical_loss": 3.686768281001588,
"tokens_seen": 898891776
},
{
"epoch": 0.27,
"learning_rate": 0.0003674924774322969,
"loss": 2.7579,
"theoretical_loss": 3.6867421956364734,
"tokens_seen": 898957312
},
{
"epoch": 0.27,
"learning_rate": 0.0003674824473420261,
"loss": 2.5761,
"theoretical_loss": 3.6867161127053976,
"tokens_seen": 899022848
},
{
"epoch": 0.27,
"learning_rate": 0.00036747241725175526,
"loss": 2.8826,
"theoretical_loss": 3.686690032207956,
"tokens_seen": 899088384
},
{
"epoch": 0.27,
"learning_rate": 0.00036746238716148444,
"loss": 2.8425,
"theoretical_loss": 3.6866639541437447,
"tokens_seen": 899153920
},
{
"epoch": 0.27,
"learning_rate": 0.0003674523570712137,
"loss": 2.6685,
"theoretical_loss": 3.686637878512358,
"tokens_seen": 899219456
},
{
"epoch": 0.27,
"learning_rate": 0.0003674423269809428,
"loss": 2.7721,
"theoretical_loss": 3.686611805313393,
"tokens_seen": 899284992
},
{
"epoch": 0.27,
"learning_rate": 0.00036743229689067204,
"loss": 2.8463,
"theoretical_loss": 3.686585734546445,
"tokens_seen": 899350528
},
{
"epoch": 0.27,
"learning_rate": 0.00036742226680040117,
"loss": 2.8633,
"theoretical_loss": 3.68655966621111,
"tokens_seen": 899416064
},
{
"epoch": 0.27,
"objective/train/docs_used": 516375,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.457921266555786,
"objective/train/theoretical_loss": 3.6865336003069844,
"objective/train/tokens_used": 919941600,
"theoretical_loss": 3.6865336003069844,
"tokens_seen": 899481600
},
{
"epoch": 0.27,
"learning_rate": 0.0003674122367101304,
"loss": 2.7374,
"theoretical_loss": 3.6865336003069844,
"tokens_seen": 899481600
},
{
"epoch": 0.27,
"learning_rate": 0.0003674022066198596,
"loss": 2.5723,
"theoretical_loss": 3.6865075368336635,
"tokens_seen": 899547136
},
{
"epoch": 0.27,
"learning_rate": 0.00036739217652958877,
"loss": 2.7103,
"theoretical_loss": 3.6864814757907443,
"tokens_seen": 899612672
},
{
"epoch": 0.27,
"learning_rate": 0.00036738214643931795,
"loss": 2.6715,
"theoretical_loss": 3.686455417177823,
"tokens_seen": 899678208
},
{
"epoch": 0.27,
"learning_rate": 0.0003673721163490472,
"loss": 2.787,
"theoretical_loss": 3.6864293609944956,
"tokens_seen": 899743744
},
{
"epoch": 0.27,
"learning_rate": 0.0003673620862587763,
"loss": 2.78,
"theoretical_loss": 3.6864033072403593,
"tokens_seen": 899809280
},
{
"epoch": 0.27,
"learning_rate": 0.00036735205616850555,
"loss": 2.6552,
"theoretical_loss": 3.6863772559150103,
"tokens_seen": 899874816
},
{
"epoch": 0.27,
"learning_rate": 0.00036734202607823467,
"loss": 2.6897,
"theoretical_loss": 3.6863512070180455,
"tokens_seen": 899940352
},
{
"epoch": 0.27,
"learning_rate": 0.0003673319959879639,
"loss": 2.7224,
"theoretical_loss": 3.6863251605490612,
"tokens_seen": 900005888
},
{
"epoch": 0.27,
"learning_rate": 0.0003673219658976931,
"loss": 2.9014,
"theoretical_loss": 3.686299116507655,
"tokens_seen": 900071424
},
{
"epoch": 0.27,
"learning_rate": 0.00036731193580742227,
"loss": 2.8336,
"theoretical_loss": 3.686273074893424,
"tokens_seen": 900136960
},
{
"epoch": 0.27,
"learning_rate": 0.00036730190571715145,
"loss": 2.4775,
"theoretical_loss": 3.6862470357059642,
"tokens_seen": 900202496
},
{
"epoch": 0.27,
"learning_rate": 0.00036729187562688063,
"loss": 2.6902,
"theoretical_loss": 3.6862209989448744,
"tokens_seen": 900268032
},
{
"epoch": 0.27,
"learning_rate": 0.0003672818455366098,
"loss": 2.8416,
"theoretical_loss": 3.6861949646097507,
"tokens_seen": 900333568
},
{
"epoch": 0.27,
"learning_rate": 0.00036727181544633905,
"loss": 2.9189,
"theoretical_loss": 3.686168932700191,
"tokens_seen": 900399104
},
{
"epoch": 0.27,
"learning_rate": 0.0003672617853560682,
"loss": 3.0051,
"theoretical_loss": 3.6861429032157926,
"tokens_seen": 900464640
},
{
"epoch": 0.27,
"learning_rate": 0.0003672517552657974,
"loss": 2.9879,
"theoretical_loss": 3.686116876156153,
"tokens_seen": 900530176
},
{
"epoch": 0.27,
"learning_rate": 0.00036724172517552654,
"loss": 2.5967,
"theoretical_loss": 3.6860908515208703,
"tokens_seen": 900595712
},
{
"epoch": 0.27,
"learning_rate": 0.00036723169508525577,
"loss": 2.6481,
"theoretical_loss": 3.686064829309542,
"tokens_seen": 900661248
},
{
"epoch": 0.27,
"learning_rate": 0.00036722166499498495,
"loss": 2.7597,
"theoretical_loss": 3.686038809521766,
"tokens_seen": 900726784
},
{
"epoch": 0.27,
"learning_rate": 0.00036721163490471414,
"loss": 2.8096,
"theoretical_loss": 3.6860127921571406,
"tokens_seen": 900792320
},
{
"epoch": 0.27,
"learning_rate": 0.0003672016048144433,
"loss": 2.8267,
"theoretical_loss": 3.685986777215263,
"tokens_seen": 900857856
},
{
"epoch": 0.27,
"learning_rate": 0.00036719157472417255,
"loss": 3.0114,
"theoretical_loss": 3.685960764695732,
"tokens_seen": 900923392
},
{
"epoch": 0.27,
"learning_rate": 0.0003671815446339017,
"loss": 2.7245,
"theoretical_loss": 3.6859347545981462,
"tokens_seen": 900988928
},
{
"epoch": 0.27,
"learning_rate": 0.0003671715145436309,
"loss": 2.9083,
"theoretical_loss": 3.6859087469221032,
"tokens_seen": 901054464
},
{
"epoch": 0.27,
"objective/train/docs_used": 517599,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.637302875518799,
"objective/train/theoretical_loss": 3.685882741667202,
"objective/train/tokens_used": 921580000,
"theoretical_loss": 3.685882741667202,
"tokens_seen": 901120000
},
{
"epoch": 0.27,
"learning_rate": 0.00036716148445336004,
"loss": 2.7865,
"theoretical_loss": 3.685882741667202,
"tokens_seen": 901120000
},
{
"epoch": 0.27,
"learning_rate": 0.0003671514543630893,
"loss": 2.8568,
"theoretical_loss": 3.6858567388330408,
"tokens_seen": 901185536
},
{
"epoch": 0.27,
"learning_rate": 0.00036714142427281846,
"loss": 2.6407,
"theoretical_loss": 3.6858307384192184,
"tokens_seen": 901251072
},
{
"epoch": 0.27,
"learning_rate": 0.00036713139418254764,
"loss": 2.8436,
"theoretical_loss": 3.6858047404253336,
"tokens_seen": 901316608
},
{
"epoch": 0.27,
"learning_rate": 0.0003671213640922768,
"loss": 2.6432,
"theoretical_loss": 3.6857787448509853,
"tokens_seen": 901382144
},
{
"epoch": 0.27,
"learning_rate": 0.000367111334002006,
"loss": 2.936,
"theoretical_loss": 3.685752751695772,
"tokens_seen": 901447680
},
{
"epoch": 0.27,
"learning_rate": 0.00036710130391173524,
"loss": 2.7945,
"theoretical_loss": 3.685726760959293,
"tokens_seen": 901513216
},
{
"epoch": 0.27,
"learning_rate": 0.0003670912738214644,
"loss": 2.9851,
"theoretical_loss": 3.6857007726411477,
"tokens_seen": 901578752
},
{
"epoch": 0.27,
"learning_rate": 0.0003670812437311936,
"loss": 2.5452,
"theoretical_loss": 3.685674786740935,
"tokens_seen": 901644288
},
{
"epoch": 0.27,
"learning_rate": 0.0003670712136409228,
"loss": 2.7208,
"theoretical_loss": 3.6856488032582546,
"tokens_seen": 901709824
},
{
"epoch": 0.27,
"learning_rate": 0.000367061183550652,
"loss": 2.7073,
"theoretical_loss": 3.6856228221927054,
"tokens_seen": 901775360
},
{
"epoch": 0.27,
"learning_rate": 0.00036705115346038114,
"loss": 2.8649,
"theoretical_loss": 3.6855968435438866,
"tokens_seen": 901840896
},
{
"epoch": 0.27,
"learning_rate": 0.0003670411233701104,
"loss": 2.5949,
"theoretical_loss": 3.685570867311399,
"tokens_seen": 901906432
},
{
"epoch": 0.27,
"learning_rate": 0.0003670310932798395,
"loss": 2.7568,
"theoretical_loss": 3.685544893494841,
"tokens_seen": 901971968
},
{
"epoch": 0.27,
"learning_rate": 0.00036702106318956874,
"loss": 2.8821,
"theoretical_loss": 3.6855189220938134,
"tokens_seen": 902037504
},
{
"epoch": 0.27,
"learning_rate": 0.0003670110330992979,
"loss": 2.8801,
"theoretical_loss": 3.6854929531079157,
"tokens_seen": 902103040
},
{
"epoch": 0.27,
"learning_rate": 0.0003670010030090271,
"loss": 2.6808,
"theoretical_loss": 3.6854669865367473,
"tokens_seen": 902168576
},
{
"epoch": 0.27,
"learning_rate": 0.0003669909729187563,
"loss": 3.1388,
"theoretical_loss": 3.6854410223799094,
"tokens_seen": 902234112
},
{
"epoch": 0.27,
"learning_rate": 0.00036698094282848546,
"loss": 2.5709,
"theoretical_loss": 3.685415060637002,
"tokens_seen": 902299648
},
{
"epoch": 0.27,
"learning_rate": 0.00036697091273821464,
"loss": 2.7695,
"theoretical_loss": 3.6853891013076243,
"tokens_seen": 902365184
},
{
"epoch": 0.27,
"learning_rate": 0.0003669608826479439,
"loss": 2.8013,
"theoretical_loss": 3.6853631443913772,
"tokens_seen": 902430720
},
{
"epoch": 0.27,
"learning_rate": 0.000366950852557673,
"loss": 2.8871,
"theoretical_loss": 3.6853371898878615,
"tokens_seen": 902496256
},
{
"epoch": 0.27,
"learning_rate": 0.00036694082246740224,
"loss": 2.7994,
"theoretical_loss": 3.685311237796678,
"tokens_seen": 902561792
},
{
"epoch": 0.27,
"learning_rate": 0.00036693079237713137,
"loss": 2.6384,
"theoretical_loss": 3.6852852881174263,
"tokens_seen": 902627328
},
{
"epoch": 0.27,
"learning_rate": 0.0003669207622868606,
"loss": 2.7996,
"theoretical_loss": 3.685259340849708,
"tokens_seen": 902692864
},
{
"epoch": 0.27,
"objective/train/docs_used": 518319,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.869065523147583,
"objective/train/theoretical_loss": 3.685233395993124,
"objective/train/tokens_used": 923218400,
"theoretical_loss": 3.685233395993124,
"tokens_seen": 902758400
},
{
"epoch": 0.27,
"learning_rate": 0.0003669107321965898,
"loss": 2.9435,
"theoretical_loss": 3.685233395993124,
"tokens_seen": 902758400
},
{
"epoch": 0.27,
"learning_rate": 0.00036690070210631897,
"loss": 2.6153,
"theoretical_loss": 3.6852074535472745,
"tokens_seen": 902823936
},
{
"epoch": 0.27,
"learning_rate": 0.00036689067201604815,
"loss": 2.7406,
"theoretical_loss": 3.685181513511761,
"tokens_seen": 902889472
},
{
"epoch": 0.27,
"learning_rate": 0.0003668806419257774,
"loss": 2.7407,
"theoretical_loss": 3.685155575886185,
"tokens_seen": 902955008
},
{
"epoch": 0.27,
"learning_rate": 0.0003668706118355065,
"loss": 2.7482,
"theoretical_loss": 3.6851296406701466,
"tokens_seen": 903020544
},
{
"epoch": 0.27,
"learning_rate": 0.00036686058174523575,
"loss": 3.0303,
"theoretical_loss": 3.6851037078632483,
"tokens_seen": 903086080
},
{
"epoch": 0.27,
"learning_rate": 0.00036685055165496487,
"loss": 2.7484,
"theoretical_loss": 3.685077777465091,
"tokens_seen": 903151616
},
{
"epoch": 0.27,
"learning_rate": 0.0003668405215646941,
"loss": 2.6951,
"theoretical_loss": 3.685051849475276,
"tokens_seen": 903217152
},
{
"epoch": 0.27,
"learning_rate": 0.0003668304914744233,
"loss": 2.7608,
"theoretical_loss": 3.6850259238934053,
"tokens_seen": 903282688
},
{
"epoch": 0.27,
"learning_rate": 0.00036682046138415247,
"loss": 2.646,
"theoretical_loss": 3.6850000007190804,
"tokens_seen": 903348224
},
{
"epoch": 0.27,
"learning_rate": 0.00036681043129388165,
"loss": 2.7623,
"theoretical_loss": 3.684974079951903,
"tokens_seen": 903413760
},
{
"epoch": 0.27,
"learning_rate": 0.00036680040120361083,
"loss": 2.9223,
"theoretical_loss": 3.684948161591475,
"tokens_seen": 903479296
},
{
"epoch": 0.27,
"learning_rate": 0.00036679037111334,
"loss": 2.8219,
"theoretical_loss": 3.6849222456373987,
"tokens_seen": 903544832
},
{
"epoch": 0.27,
"learning_rate": 0.00036678034102306925,
"loss": 2.7417,
"theoretical_loss": 3.684896332089276,
"tokens_seen": 903610368
},
{
"epoch": 0.27,
"learning_rate": 0.0003667703109327984,
"loss": 2.9203,
"theoretical_loss": 3.6848704209467087,
"tokens_seen": 903675904
},
{
"epoch": 0.27,
"learning_rate": 0.0003667602808425276,
"loss": 2.6219,
"theoretical_loss": 3.6848445122093,
"tokens_seen": 903741440
},
{
"epoch": 0.27,
"learning_rate": 0.00036675025075225674,
"loss": 2.9307,
"theoretical_loss": 3.684818605876651,
"tokens_seen": 903806976
},
{
"epoch": 0.27,
"learning_rate": 0.000366740220661986,
"loss": 2.5648,
"theoretical_loss": 3.6847927019483646,
"tokens_seen": 903872512
},
{
"epoch": 0.27,
"learning_rate": 0.00036673019057171515,
"loss": 2.8643,
"theoretical_loss": 3.6847668004240437,
"tokens_seen": 903938048
},
{
"epoch": 0.27,
"learning_rate": 0.00036672016048144434,
"loss": 2.9814,
"theoretical_loss": 3.6847409013032912,
"tokens_seen": 904003584
},
{
"epoch": 0.27,
"learning_rate": 0.0003667101303911735,
"loss": 2.8465,
"theoretical_loss": 3.684715004585709,
"tokens_seen": 904069120
},
{
"epoch": 0.27,
"learning_rate": 0.00036670010030090275,
"loss": 2.634,
"theoretical_loss": 3.6846891102709005,
"tokens_seen": 904134656
},
{
"epoch": 0.27,
"learning_rate": 0.0003666900702106319,
"loss": 2.6441,
"theoretical_loss": 3.684663218358468,
"tokens_seen": 904200192
},
{
"epoch": 0.27,
"learning_rate": 0.0003666800401203611,
"loss": 2.6918,
"theoretical_loss": 3.684637328848016,
"tokens_seen": 904265728
},
{
"epoch": 0.27,
"learning_rate": 0.00036667001003009024,
"loss": 2.8855,
"theoretical_loss": 3.684611441739145,
"tokens_seen": 904331264
},
{
"epoch": 0.27,
"objective/train/docs_used": 519704,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6285362243652344,
"objective/train/theoretical_loss": 3.684585557031461,
"objective/train/tokens_used": 924856800,
"theoretical_loss": 3.684585557031461,
"tokens_seen": 904396800
},
{
"epoch": 0.27,
"learning_rate": 0.0003666599799398195,
"loss": 2.5311,
"theoretical_loss": 3.684585557031461,
"tokens_seen": 904396800
},
{
"epoch": 0.27,
"learning_rate": 0.00036664994984954866,
"loss": 2.8557,
"theoretical_loss": 3.684559674724566,
"tokens_seen": 904462336
},
{
"epoch": 0.27,
"learning_rate": 0.00036663991975927784,
"loss": 2.7816,
"theoretical_loss": 3.684533794818063,
"tokens_seen": 904527872
},
{
"epoch": 0.27,
"learning_rate": 0.000366629889669007,
"loss": 2.6324,
"theoretical_loss": 3.6845079173115565,
"tokens_seen": 904593408
},
{
"epoch": 0.27,
"learning_rate": 0.0003666198595787362,
"loss": 2.798,
"theoretical_loss": 3.684482042204649,
"tokens_seen": 904658944
},
{
"epoch": 0.27,
"learning_rate": 0.0003666098294884654,
"loss": 2.5304,
"theoretical_loss": 3.6844561694969453,
"tokens_seen": 904724480
},
{
"epoch": 0.27,
"learning_rate": 0.0003665997993981946,
"loss": 2.9537,
"theoretical_loss": 3.6844302991880484,
"tokens_seen": 904790016
},
{
"epoch": 0.27,
"learning_rate": 0.00036658976930792374,
"loss": 2.8049,
"theoretical_loss": 3.684404431277562,
"tokens_seen": 904855552
},
{
"epoch": 0.27,
"learning_rate": 0.000366579739217653,
"loss": 2.6469,
"theoretical_loss": 3.6843785657650905,
"tokens_seen": 904921088
},
{
"epoch": 0.27,
"learning_rate": 0.00036656970912738216,
"loss": 2.7804,
"theoretical_loss": 3.684352702650238,
"tokens_seen": 904986624
},
{
"epoch": 0.27,
"learning_rate": 0.00036655967903711134,
"loss": 2.9802,
"theoretical_loss": 3.684326841932609,
"tokens_seen": 905052160
},
{
"epoch": 0.27,
"learning_rate": 0.0003665496489468405,
"loss": 2.7824,
"theoretical_loss": 3.684300983611807,
"tokens_seen": 905117696
},
{
"epoch": 0.27,
"learning_rate": 0.0003665396188565697,
"loss": 2.5405,
"theoretical_loss": 3.684275127687436,
"tokens_seen": 905183232
},
{
"epoch": 0.27,
"learning_rate": 0.0003665295887662989,
"loss": 2.9494,
"theoretical_loss": 3.684249274159102,
"tokens_seen": 905248768
},
{
"epoch": 0.27,
"learning_rate": 0.0003665195586760281,
"loss": 2.7371,
"theoretical_loss": 3.684223423026408,
"tokens_seen": 905314304
},
{
"epoch": 0.27,
"learning_rate": 0.00036650952858575725,
"loss": 2.3495,
"theoretical_loss": 3.6841975742889588,
"tokens_seen": 905379840
},
{
"epoch": 0.27,
"learning_rate": 0.0003664994984954865,
"loss": 2.7101,
"theoretical_loss": 3.68417172794636,
"tokens_seen": 905445376
},
{
"epoch": 0.27,
"learning_rate": 0.0003664894684052156,
"loss": 2.6259,
"theoretical_loss": 3.684145883998216,
"tokens_seen": 905510912
},
{
"epoch": 0.27,
"learning_rate": 0.00036647943831494484,
"loss": 2.713,
"theoretical_loss": 3.684120042444132,
"tokens_seen": 905576448
},
{
"epoch": 0.27,
"learning_rate": 0.000366469408224674,
"loss": 2.6741,
"theoretical_loss": 3.6840942032837116,
"tokens_seen": 905641984
},
{
"epoch": 0.27,
"learning_rate": 0.0003664593781344032,
"loss": 2.8036,
"theoretical_loss": 3.6840683665165614,
"tokens_seen": 905707520
},
{
"epoch": 0.27,
"learning_rate": 0.0003664493480441324,
"loss": 2.4032,
"theoretical_loss": 3.6840425321422856,
"tokens_seen": 905773056
},
{
"epoch": 0.27,
"learning_rate": 0.00036643931795386157,
"loss": 2.8506,
"theoretical_loss": 3.6840167001604907,
"tokens_seen": 905838592
},
{
"epoch": 0.27,
"learning_rate": 0.00036642928786359075,
"loss": 2.6629,
"theoretical_loss": 3.683990870570781,
"tokens_seen": 905904128
},
{
"epoch": 0.27,
"learning_rate": 0.00036641925777332,
"loss": 2.6987,
"theoretical_loss": 3.683965043372762,
"tokens_seen": 905969664
},
{
"epoch": 0.27,
"objective/train/docs_used": 520483,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.922074317932129,
"objective/train/theoretical_loss": 3.6839392185660396,
"objective/train/tokens_used": 926495200,
"theoretical_loss": 3.6839392185660396,
"tokens_seen": 906035200
},
{
"epoch": 0.27,
"learning_rate": 0.0003664092276830491,
"loss": 2.9621,
"theoretical_loss": 3.6839392185660396,
"tokens_seen": 906035200
},
{
"epoch": 0.27,
"learning_rate": 0.00036639919759277835,
"loss": 2.8046,
"theoretical_loss": 3.6839133961502197,
"tokens_seen": 906100736
},
{
"epoch": 0.27,
"learning_rate": 0.00036638916750250753,
"loss": 2.638,
"theoretical_loss": 3.6838875761249072,
"tokens_seen": 906166272
},
{
"epoch": 0.27,
"learning_rate": 0.0003663791374122367,
"loss": 3.0013,
"theoretical_loss": 3.6838617584897086,
"tokens_seen": 906231808
},
{
"epoch": 0.27,
"learning_rate": 0.0003663691073219659,
"loss": 2.4993,
"theoretical_loss": 3.68383594324423,
"tokens_seen": 906297344
},
{
"epoch": 0.27,
"learning_rate": 0.00036635907723169507,
"loss": 2.6938,
"theoretical_loss": 3.6838101303880766,
"tokens_seen": 906362880
},
{
"epoch": 0.27,
"learning_rate": 0.0003663490471414243,
"loss": 2.5562,
"theoretical_loss": 3.6837843199208558,
"tokens_seen": 906428416
},
{
"epoch": 0.27,
"learning_rate": 0.0003663390170511535,
"loss": 2.8366,
"theoretical_loss": 3.6837585118421723,
"tokens_seen": 906493952
},
{
"epoch": 0.27,
"learning_rate": 0.00036632898696088267,
"loss": 2.7984,
"theoretical_loss": 3.683732706151633,
"tokens_seen": 906559488
},
{
"epoch": 0.27,
"learning_rate": 0.00036631895687061185,
"loss": 2.9263,
"theoretical_loss": 3.6837069028488454,
"tokens_seen": 906625024
},
{
"epoch": 0.27,
"learning_rate": 0.00036630892678034103,
"loss": 2.7543,
"theoretical_loss": 3.683681101933414,
"tokens_seen": 906690560
},
{
"epoch": 0.27,
"learning_rate": 0.0003662988966900702,
"loss": 2.5262,
"theoretical_loss": 3.683655303404947,
"tokens_seen": 906756096
},
{
"epoch": 0.27,
"learning_rate": 0.00036628886659979945,
"loss": 2.5008,
"theoretical_loss": 3.68362950726305,
"tokens_seen": 906821632
},
{
"epoch": 0.27,
"learning_rate": 0.0003662788365095286,
"loss": 2.9098,
"theoretical_loss": 3.6836037135073307,
"tokens_seen": 906887168
},
{
"epoch": 0.27,
"learning_rate": 0.0003662688064192578,
"loss": 2.9096,
"theoretical_loss": 3.6835779221373954,
"tokens_seen": 906952704
},
{
"epoch": 0.27,
"learning_rate": 0.00036625877632898694,
"loss": 2.7425,
"theoretical_loss": 3.683552133152851,
"tokens_seen": 907018240
},
{
"epoch": 0.27,
"learning_rate": 0.0003662487462387162,
"loss": 2.7461,
"theoretical_loss": 3.683526346553305,
"tokens_seen": 907083776
},
{
"epoch": 0.27,
"learning_rate": 0.00036623871614844535,
"loss": 2.6051,
"theoretical_loss": 3.683500562338364,
"tokens_seen": 907149312
},
{
"epoch": 0.27,
"learning_rate": 0.00036622868605817454,
"loss": 2.5585,
"theoretical_loss": 3.6834747805076353,
"tokens_seen": 907214848
},
{
"epoch": 0.27,
"learning_rate": 0.0003662186559679037,
"loss": 2.7987,
"theoretical_loss": 3.6834490010607266,
"tokens_seen": 907280384
},
{
"epoch": 0.27,
"learning_rate": 0.00036620862587763295,
"loss": 2.5842,
"theoretical_loss": 3.683423223997245,
"tokens_seen": 907345920
},
{
"epoch": 0.27,
"learning_rate": 0.0003661985957873621,
"loss": 2.8675,
"theoretical_loss": 3.6833974493167982,
"tokens_seen": 907411456
},
{
"epoch": 0.27,
"learning_rate": 0.0003661885656970913,
"loss": 2.6299,
"theoretical_loss": 3.683371677018994,
"tokens_seen": 907476992
},
{
"epoch": 0.28,
"learning_rate": 0.00036617853560682044,
"loss": 2.5256,
"theoretical_loss": 3.6833459071034396,
"tokens_seen": 907542528
},
{
"epoch": 0.28,
"learning_rate": 0.0003661685055165497,
"loss": 2.6402,
"theoretical_loss": 3.6833201395697435,
"tokens_seen": 907608064
},
{
"epoch": 0.28,
"objective/train/docs_used": 521740,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3810319900512695,
"objective/train/theoretical_loss": 3.6832943744175126,
"objective/train/tokens_used": 928133600,
"theoretical_loss": 3.6832943744175126,
"tokens_seen": 907673600
},
{
"epoch": 0.28,
"learning_rate": 0.00036615847542627886,
"loss": 2.7852,
"theoretical_loss": 3.6832943744175126,
"tokens_seen": 907673600
},
{
"epoch": 0.28,
"learning_rate": 0.00036614844533600804,
"loss": 2.8433,
"theoretical_loss": 3.6832686116463558,
"tokens_seen": 907739136
},
{
"epoch": 0.28,
"learning_rate": 0.0003661384152457372,
"loss": 2.7605,
"theoretical_loss": 3.683242851255881,
"tokens_seen": 907804672
},
{
"epoch": 0.28,
"learning_rate": 0.0003661283851554664,
"loss": 2.7733,
"theoretical_loss": 3.6832170932456956,
"tokens_seen": 907870208
},
{
"epoch": 0.28,
"learning_rate": 0.0003661183550651956,
"loss": 2.5039,
"theoretical_loss": 3.6831913376154084,
"tokens_seen": 907935744
},
{
"epoch": 0.28,
"learning_rate": 0.0003661083249749248,
"loss": 2.6721,
"theoretical_loss": 3.683165584364628,
"tokens_seen": 908001280
},
{
"epoch": 0.28,
"learning_rate": 0.00036609829488465394,
"loss": 2.8251,
"theoretical_loss": 3.6831398334929624,
"tokens_seen": 908066816
},
{
"epoch": 0.28,
"learning_rate": 0.0003660882647943832,
"loss": 2.7343,
"theoretical_loss": 3.6831140850000206,
"tokens_seen": 908132352
},
{
"epoch": 0.28,
"learning_rate": 0.00036607823470411236,
"loss": 2.6125,
"theoretical_loss": 3.683088338885411,
"tokens_seen": 908197888
},
{
"epoch": 0.28,
"learning_rate": 0.00036606820461384154,
"loss": 2.8327,
"theoretical_loss": 3.6830625951487423,
"tokens_seen": 908263424
},
{
"epoch": 0.28,
"learning_rate": 0.0003660581745235707,
"loss": 2.7923,
"theoretical_loss": 3.6830368537896225,
"tokens_seen": 908328960
},
{
"epoch": 0.28,
"learning_rate": 0.0003660481444332999,
"loss": 2.877,
"theoretical_loss": 3.6830111148076625,
"tokens_seen": 908394496
},
{
"epoch": 0.28,
"learning_rate": 0.0003660381143430291,
"loss": 2.5799,
"theoretical_loss": 3.6829853782024697,
"tokens_seen": 908460032
},
{
"epoch": 0.28,
"learning_rate": 0.0003660280842527583,
"loss": 2.5612,
"theoretical_loss": 3.6829596439736534,
"tokens_seen": 908525568
},
{
"epoch": 0.28,
"learning_rate": 0.00036601805416248745,
"loss": 2.5835,
"theoretical_loss": 3.6829339121208227,
"tokens_seen": 908591104
},
{
"epoch": 0.28,
"learning_rate": 0.0003660080240722167,
"loss": 2.5849,
"theoretical_loss": 3.6829081826435877,
"tokens_seen": 908656640
},
{
"epoch": 0.28,
"learning_rate": 0.0003659979939819458,
"loss": 2.6949,
"theoretical_loss": 3.682882455541557,
"tokens_seen": 908722176
},
{
"epoch": 0.28,
"learning_rate": 0.00036598796389167504,
"loss": 2.5956,
"theoretical_loss": 3.6828567308143403,
"tokens_seen": 908787712
},
{
"epoch": 0.28,
"learning_rate": 0.0003659779338014042,
"loss": 2.9759,
"theoretical_loss": 3.682831008461547,
"tokens_seen": 908853248
},
{
"epoch": 0.28,
"learning_rate": 0.0003659679037111334,
"loss": 2.6955,
"theoretical_loss": 3.682805288482787,
"tokens_seen": 908918784
},
{
"epoch": 0.28,
"learning_rate": 0.0003659578736208626,
"loss": 2.6967,
"theoretical_loss": 3.6827795708776705,
"tokens_seen": 908984320
},
{
"epoch": 0.28,
"learning_rate": 0.00036594784353059177,
"loss": 2.7206,
"theoretical_loss": 3.6827538556458057,
"tokens_seen": 909049856
},
{
"epoch": 0.28,
"learning_rate": 0.00036593781344032095,
"loss": 2.8277,
"theoretical_loss": 3.6827281427868046,
"tokens_seen": 909115392
},
{
"epoch": 0.28,
"learning_rate": 0.0003659277833500502,
"loss": 2.6394,
"theoretical_loss": 3.6827024323002755,
"tokens_seen": 909180928
},
{
"epoch": 0.28,
"learning_rate": 0.0003659177532597793,
"loss": 2.7085,
"theoretical_loss": 3.6826767241858294,
"tokens_seen": 909246464
},
{
"epoch": 0.28,
"objective/train/docs_used": 522301,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.564457893371582,
"objective/train/theoretical_loss": 3.6826510184430763,
"objective/train/tokens_used": 929772000,
"theoretical_loss": 3.6826510184430763,
"tokens_seen": 909312000
},
{
"epoch": 0.28,
"learning_rate": 0.00036590772316950855,
"loss": 2.7025,
"theoretical_loss": 3.6826510184430763,
"tokens_seen": 909312000
},
{
"epoch": 0.28,
"learning_rate": 0.00036589769307923773,
"loss": 2.9312,
"theoretical_loss": 3.6826253150716264,
"tokens_seen": 909377536
},
{
"epoch": 0.28,
"learning_rate": 0.0003658876629889669,
"loss": 2.9742,
"theoretical_loss": 3.6825996140710906,
"tokens_seen": 909443072
},
{
"epoch": 0.28,
"learning_rate": 0.0003658776328986961,
"loss": 2.7432,
"theoretical_loss": 3.682573915441078,
"tokens_seen": 909508608
},
{
"epoch": 0.28,
"learning_rate": 0.00036586760280842527,
"loss": 2.8466,
"theoretical_loss": 3.682548219181201,
"tokens_seen": 909574144
},
{
"epoch": 0.28,
"learning_rate": 0.00036585757271815445,
"loss": 3.1277,
"theoretical_loss": 3.682522525291069,
"tokens_seen": 909639680
},
{
"epoch": 0.28,
"learning_rate": 0.0003658475426278837,
"loss": 2.7698,
"theoretical_loss": 3.6824968337702932,
"tokens_seen": 909705216
},
{
"epoch": 0.28,
"learning_rate": 0.0003658375125376128,
"loss": 2.9429,
"theoretical_loss": 3.6824711446184843,
"tokens_seen": 909770752
},
{
"epoch": 0.28,
"learning_rate": 0.00036582748244734205,
"loss": 2.9036,
"theoretical_loss": 3.6824454578352537,
"tokens_seen": 909836288
},
{
"epoch": 0.28,
"learning_rate": 0.0003658174523570712,
"loss": 2.9695,
"theoretical_loss": 3.6824197734202118,
"tokens_seen": 909901824
},
{
"epoch": 0.28,
"learning_rate": 0.0003658074222668004,
"loss": 2.9616,
"theoretical_loss": 3.6823940913729696,
"tokens_seen": 909967360
},
{
"epoch": 0.28,
"learning_rate": 0.0003657973921765296,
"loss": 2.7555,
"theoretical_loss": 3.6823684116931386,
"tokens_seen": 910032896
},
{
"epoch": 0.28,
"learning_rate": 0.0003657873620862588,
"loss": 2.7536,
"theoretical_loss": 3.682342734380331,
"tokens_seen": 910098432
},
{
"epoch": 0.28,
"learning_rate": 0.00036577733199598796,
"loss": 2.9484,
"theoretical_loss": 3.6823170594341565,
"tokens_seen": 910163968
},
{
"epoch": 0.28,
"learning_rate": 0.00036576730190571714,
"loss": 2.7641,
"theoretical_loss": 3.682291386854228,
"tokens_seen": 910229504
},
{
"epoch": 0.28,
"learning_rate": 0.0003657572718154463,
"loss": 2.8987,
"theoretical_loss": 3.6822657166401562,
"tokens_seen": 910295040
},
{
"epoch": 0.28,
"learning_rate": 0.00036574724172517555,
"loss": 2.6981,
"theoretical_loss": 3.682240048791553,
"tokens_seen": 910360576
},
{
"epoch": 0.28,
"learning_rate": 0.0003657372116349047,
"loss": 2.6351,
"theoretical_loss": 3.6822143833080307,
"tokens_seen": 910426112
},
{
"epoch": 0.28,
"learning_rate": 0.0003657271815446339,
"loss": 2.8937,
"theoretical_loss": 3.6821887201892007,
"tokens_seen": 910491648
},
{
"epoch": 0.28,
"learning_rate": 0.0003657171514543631,
"loss": 2.8801,
"theoretical_loss": 3.682163059434674,
"tokens_seen": 910557184
},
{
"epoch": 0.28,
"learning_rate": 0.0003657071213640923,
"loss": 2.7551,
"theoretical_loss": 3.6821374010440646,
"tokens_seen": 910622720
},
{
"epoch": 0.28,
"learning_rate": 0.00036569709127382146,
"loss": 2.8167,
"theoretical_loss": 3.6821117450169836,
"tokens_seen": 910688256
},
{
"epoch": 0.28,
"learning_rate": 0.00036568706118355064,
"loss": 2.8311,
"theoretical_loss": 3.682086091353043,
"tokens_seen": 910753792
},
{
"epoch": 0.28,
"learning_rate": 0.0003656770310932798,
"loss": 2.7046,
"theoretical_loss": 3.6820604400518553,
"tokens_seen": 910819328
},
{
"epoch": 0.28,
"learning_rate": 0.00036566700100300906,
"loss": 2.7093,
"theoretical_loss": 3.6820347911130327,
"tokens_seen": 910884864
},
{
"epoch": 0.28,
"objective/train/docs_used": 523753,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4224841594696045,
"objective/train/theoretical_loss": 3.682009144536188,
"objective/train/tokens_used": 931410400,
"theoretical_loss": 3.682009144536188,
"tokens_seen": 910950400
},
{
"epoch": 0.28,
"learning_rate": 0.0003656569709127382,
"loss": 2.4905,
"theoretical_loss": 3.682009144536188,
"tokens_seen": 910950400
},
{
"epoch": 0.28,
"learning_rate": 0.0003656469408224674,
"loss": 2.7217,
"theoretical_loss": 3.6819835003209342,
"tokens_seen": 911015936
},
{
"epoch": 0.28,
"learning_rate": 0.00036563691073219655,
"loss": 2.7902,
"theoretical_loss": 3.681957858466883,
"tokens_seen": 911081472
},
{
"epoch": 0.28,
"learning_rate": 0.0003656268806419258,
"loss": 2.7788,
"theoretical_loss": 3.681932218973648,
"tokens_seen": 911147008
},
{
"epoch": 0.28,
"learning_rate": 0.00036561685055165496,
"loss": 2.9581,
"theoretical_loss": 3.681906581840842,
"tokens_seen": 911212544
},
{
"epoch": 0.28,
"learning_rate": 0.00036560682046138414,
"loss": 2.5835,
"theoretical_loss": 3.681880947068077,
"tokens_seen": 911278080
},
{
"epoch": 0.28,
"learning_rate": 0.0003655967903711134,
"loss": 2.7265,
"theoretical_loss": 3.681855314654967,
"tokens_seen": 911343616
},
{
"epoch": 0.28,
"learning_rate": 0.00036558676028084256,
"loss": 2.9533,
"theoretical_loss": 3.681829684601125,
"tokens_seen": 911409152
},
{
"epoch": 0.28,
"learning_rate": 0.00036557673019057174,
"loss": 2.5712,
"theoretical_loss": 3.681804056906164,
"tokens_seen": 911474688
},
{
"epoch": 0.28,
"learning_rate": 0.0003655667001003009,
"loss": 2.7122,
"theoretical_loss": 3.6817784315696978,
"tokens_seen": 911540224
},
{
"epoch": 0.28,
"learning_rate": 0.0003655566700100301,
"loss": 2.6819,
"theoretical_loss": 3.6817528085913382,
"tokens_seen": 911605760
},
{
"epoch": 0.28,
"learning_rate": 0.0003655466399197593,
"loss": 2.7965,
"theoretical_loss": 3.681727187970701,
"tokens_seen": 911671296
},
{
"epoch": 0.28,
"learning_rate": 0.0003655366098294885,
"loss": 2.8174,
"theoretical_loss": 3.6817015697073985,
"tokens_seen": 911736832
},
{
"epoch": 0.28,
"learning_rate": 0.00036552657973921765,
"loss": 2.8962,
"theoretical_loss": 3.6816759538010446,
"tokens_seen": 911802368
},
{
"epoch": 0.28,
"learning_rate": 0.0003655165496489469,
"loss": 2.775,
"theoretical_loss": 3.6816503402512533,
"tokens_seen": 911867904
},
{
"epoch": 0.28,
"learning_rate": 0.000365506519558676,
"loss": 2.9705,
"theoretical_loss": 3.6816247290576376,
"tokens_seen": 911933440
},
{
"epoch": 0.28,
"learning_rate": 0.00036549648946840525,
"loss": 2.6689,
"theoretical_loss": 3.681599120219812,
"tokens_seen": 911998976
},
{
"epoch": 0.28,
"learning_rate": 0.0003654864593781344,
"loss": 2.8548,
"theoretical_loss": 3.681573513737391,
"tokens_seen": 912064512
},
{
"epoch": 0.28,
"learning_rate": 0.0003654764292878636,
"loss": 2.6712,
"theoretical_loss": 3.6815479096099883,
"tokens_seen": 912130048
},
{
"epoch": 0.28,
"learning_rate": 0.0003654663991975928,
"loss": 2.5822,
"theoretical_loss": 3.681522307837218,
"tokens_seen": 912195584
},
{
"epoch": 0.28,
"learning_rate": 0.00036545636910732197,
"loss": 2.8407,
"theoretical_loss": 3.6814967084186945,
"tokens_seen": 912261120
},
{
"epoch": 0.28,
"learning_rate": 0.00036544633901705115,
"loss": 2.5391,
"theoretical_loss": 3.681471111354033,
"tokens_seen": 912326656
},
{
"epoch": 0.28,
"learning_rate": 0.0003654363089267804,
"loss": 2.6647,
"theoretical_loss": 3.681445516642847,
"tokens_seen": 912392192
},
{
"epoch": 0.28,
"learning_rate": 0.0003654262788365095,
"loss": 2.6346,
"theoretical_loss": 3.681419924284751,
"tokens_seen": 912457728
},
{
"epoch": 0.28,
"learning_rate": 0.00036541624874623875,
"loss": 2.5153,
"theoretical_loss": 3.68139433427936,
"tokens_seen": 912523264
},
{
"epoch": 0.28,
"objective/train/docs_used": 524575,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.88395094871521,
"objective/train/theoretical_loss": 3.681368746626289,
"objective/train/tokens_used": 933048800,
"theoretical_loss": 3.681368746626289,
"tokens_seen": 912588800
},
{
"epoch": 0.28,
"learning_rate": 0.00036540621865596793,
"loss": 2.7257,
"theoretical_loss": 3.681368746626289,
"tokens_seen": 912588800
},
{
"epoch": 0.28,
"learning_rate": 0.0003653961885656971,
"loss": 2.633,
"theoretical_loss": 3.6813431613251533,
"tokens_seen": 912654336
},
{
"epoch": 0.28,
"learning_rate": 0.0003653861584754263,
"loss": 2.4453,
"theoretical_loss": 3.6813175783755665,
"tokens_seen": 912719872
},
{
"epoch": 0.28,
"learning_rate": 0.00036537612838515547,
"loss": 2.5774,
"theoretical_loss": 3.6812919977771448,
"tokens_seen": 912785408
},
{
"epoch": 0.28,
"learning_rate": 0.00036536609829488465,
"loss": 2.9087,
"theoretical_loss": 3.6812664195295026,
"tokens_seen": 912850944
},
{
"epoch": 0.28,
"learning_rate": 0.0003653560682046139,
"loss": 2.6798,
"theoretical_loss": 3.6812408436322555,
"tokens_seen": 912916480
},
{
"epoch": 0.28,
"learning_rate": 0.000365346038114343,
"loss": 2.7959,
"theoretical_loss": 3.681215270085019,
"tokens_seen": 912982016
},
{
"epoch": 0.28,
"learning_rate": 0.00036533600802407225,
"loss": 2.8895,
"theoretical_loss": 3.681189698887408,
"tokens_seen": 913047552
},
{
"epoch": 0.28,
"learning_rate": 0.0003653259779338014,
"loss": 2.6949,
"theoretical_loss": 3.681164130039039,
"tokens_seen": 913113088
},
{
"epoch": 0.28,
"learning_rate": 0.0003653159478435306,
"loss": 2.6985,
"theoretical_loss": 3.681138563539526,
"tokens_seen": 913178624
},
{
"epoch": 0.28,
"learning_rate": 0.0003653059177532598,
"loss": 2.8195,
"theoretical_loss": 3.6811129993884864,
"tokens_seen": 913244160
},
{
"epoch": 0.28,
"learning_rate": 0.000365295887662989,
"loss": 2.6316,
"theoretical_loss": 3.6810874375855342,
"tokens_seen": 913309696
},
{
"epoch": 0.28,
"learning_rate": 0.00036528585757271816,
"loss": 2.8729,
"theoretical_loss": 3.6810618781302864,
"tokens_seen": 913375232
},
{
"epoch": 0.28,
"learning_rate": 0.00036527582748244734,
"loss": 2.6401,
"theoretical_loss": 3.681036321022359,
"tokens_seen": 913440768
},
{
"epoch": 0.28,
"learning_rate": 0.0003652657973921765,
"loss": 2.6803,
"theoretical_loss": 3.681010766261368,
"tokens_seen": 913506304
},
{
"epoch": 0.28,
"learning_rate": 0.00036525576730190575,
"loss": 2.7925,
"theoretical_loss": 3.680985213846929,
"tokens_seen": 913571840
},
{
"epoch": 0.28,
"learning_rate": 0.0003652457372116349,
"loss": 2.5781,
"theoretical_loss": 3.6809596637786584,
"tokens_seen": 913637376
},
{
"epoch": 0.28,
"learning_rate": 0.0003652357071213641,
"loss": 2.6535,
"theoretical_loss": 3.6809341160561724,
"tokens_seen": 913702912
},
{
"epoch": 0.28,
"learning_rate": 0.0003652256770310933,
"loss": 2.7715,
"theoretical_loss": 3.6809085706790876,
"tokens_seen": 913768448
},
{
"epoch": 0.28,
"learning_rate": 0.0003652156469408225,
"loss": 2.835,
"theoretical_loss": 3.6808830276470212,
"tokens_seen": 913833984
},
{
"epoch": 0.28,
"learning_rate": 0.00036520561685055166,
"loss": 2.9342,
"theoretical_loss": 3.6808574869595887,
"tokens_seen": 913899520
},
{
"epoch": 0.28,
"learning_rate": 0.00036519558676028084,
"loss": 2.7387,
"theoretical_loss": 3.6808319486164067,
"tokens_seen": 913965056
},
{
"epoch": 0.28,
"learning_rate": 0.00036518555667001,
"loss": 2.7792,
"theoretical_loss": 3.680806412617093,
"tokens_seen": 914030592
},
{
"epoch": 0.28,
"learning_rate": 0.00036517552657973926,
"loss": 2.9226,
"theoretical_loss": 3.6807808789612633,
"tokens_seen": 914096128
},
{
"epoch": 0.28,
"learning_rate": 0.0003651654964894684,
"loss": 2.7356,
"theoretical_loss": 3.680755347648536,
"tokens_seen": 914161664
},
{
"epoch": 0.28,
"objective/train/docs_used": 525722,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.794320821762085,
"objective/train/theoretical_loss": 3.680729818678526,
"objective/train/tokens_used": 934687200,
"theoretical_loss": 3.680729818678526,
"tokens_seen": 914227200
},
{
"epoch": 0.28,
"learning_rate": 0.0003651554663991976,
"loss": 2.8242,
"theoretical_loss": 3.680729818678526,
"tokens_seen": 914227200
},
{
"epoch": 0.28,
"learning_rate": 0.00036514543630892675,
"loss": 2.6443,
"theoretical_loss": 3.6807042920508533,
"tokens_seen": 914292736
},
{
"epoch": 0.28,
"learning_rate": 0.000365135406218656,
"loss": 2.8175,
"theoretical_loss": 3.6806787677651323,
"tokens_seen": 914358272
},
{
"epoch": 0.28,
"learning_rate": 0.00036512537612838516,
"loss": 2.8376,
"theoretical_loss": 3.6806532458209817,
"tokens_seen": 914423808
},
{
"epoch": 0.28,
"learning_rate": 0.00036511534603811434,
"loss": 2.4964,
"theoretical_loss": 3.680627726218019,
"tokens_seen": 914489344
},
{
"epoch": 0.28,
"learning_rate": 0.0003651053159478435,
"loss": 2.8057,
"theoretical_loss": 3.680602208955861,
"tokens_seen": 914554880
},
{
"epoch": 0.28,
"learning_rate": 0.00036509528585757276,
"loss": 2.733,
"theoretical_loss": 3.6805766940341256,
"tokens_seen": 914620416
},
{
"epoch": 0.28,
"learning_rate": 0.0003650852557673019,
"loss": 2.6344,
"theoretical_loss": 3.6805511814524308,
"tokens_seen": 914685952
},
{
"epoch": 0.28,
"learning_rate": 0.0003650752256770311,
"loss": 2.766,
"theoretical_loss": 3.6805256712103938,
"tokens_seen": 914751488
},
{
"epoch": 0.28,
"learning_rate": 0.00036506519558676025,
"loss": 2.8122,
"theoretical_loss": 3.6805001633076326,
"tokens_seen": 914817024
},
{
"epoch": 0.28,
"learning_rate": 0.0003650551654964895,
"loss": 3.0508,
"theoretical_loss": 3.6804746577437655,
"tokens_seen": 914882560
},
{
"epoch": 0.28,
"learning_rate": 0.00036504513540621867,
"loss": 2.8123,
"theoretical_loss": 3.68044915451841,
"tokens_seen": 914948096
},
{
"epoch": 0.28,
"learning_rate": 0.00036503510531594785,
"loss": 2.6405,
"theoretical_loss": 3.680423653631184,
"tokens_seen": 915013632
},
{
"epoch": 0.28,
"learning_rate": 0.00036502507522567703,
"loss": 2.9047,
"theoretical_loss": 3.680398155081707,
"tokens_seen": 915079168
},
{
"epoch": 0.28,
"learning_rate": 0.0003650150451354062,
"loss": 2.8159,
"theoretical_loss": 3.680372658869596,
"tokens_seen": 915144704
},
{
"epoch": 0.28,
"learning_rate": 0.0003650050150451354,
"loss": 2.7696,
"theoretical_loss": 3.68034716499447,
"tokens_seen": 915210240
},
{
"epoch": 0.28,
"learning_rate": 0.0003649949849548646,
"loss": 2.6392,
"theoretical_loss": 3.680321673455947,
"tokens_seen": 915275776
},
{
"epoch": 0.28,
"learning_rate": 0.00036498495486459375,
"loss": 2.861,
"theoretical_loss": 3.6802961842536464,
"tokens_seen": 915341312
},
{
"epoch": 0.28,
"learning_rate": 0.000364974924774323,
"loss": 2.9131,
"theoretical_loss": 3.6802706973871855,
"tokens_seen": 915406848
},
{
"epoch": 0.28,
"learning_rate": 0.0003649648946840521,
"loss": 2.5935,
"theoretical_loss": 3.6802452128561844,
"tokens_seen": 915472384
},
{
"epoch": 0.28,
"learning_rate": 0.00036495486459378135,
"loss": 2.7192,
"theoretical_loss": 3.680219730660261,
"tokens_seen": 915537920
},
{
"epoch": 0.28,
"learning_rate": 0.00036494483450351053,
"loss": 2.6388,
"theoretical_loss": 3.6801942507990346,
"tokens_seen": 915603456
},
{
"epoch": 0.28,
"learning_rate": 0.0003649348044132397,
"loss": 2.6408,
"theoretical_loss": 3.680168773272124,
"tokens_seen": 915668992
},
{
"epoch": 0.28,
"learning_rate": 0.0003649247743229689,
"loss": 2.9475,
"theoretical_loss": 3.680143298079149,
"tokens_seen": 915734528
},
{
"epoch": 0.28,
"learning_rate": 0.00036491474423269813,
"loss": 2.5988,
"theoretical_loss": 3.680117825219728,
"tokens_seen": 915800064
},
{
"epoch": 0.28,
"objective/train/docs_used": 526330,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.590721845626831,
"objective/train/theoretical_loss": 3.6800923546934805,
"objective/train/tokens_used": 936325600,
"theoretical_loss": 3.6800923546934805,
"tokens_seen": 915865600
},
{
"epoch": 0.28,
"learning_rate": 0.00036490471414242726,
"loss": 2.8309,
"theoretical_loss": 3.6800923546934805,
"tokens_seen": 915865600
},
{
"epoch": 0.28,
"learning_rate": 0.0003648946840521565,
"loss": 2.7859,
"theoretical_loss": 3.680066886500026,
"tokens_seen": 915931136
},
{
"epoch": 0.28,
"learning_rate": 0.0003648846539618856,
"loss": 2.9457,
"theoretical_loss": 3.6800414206389833,
"tokens_seen": 915996672
},
{
"epoch": 0.28,
"learning_rate": 0.00036487462387161485,
"loss": 2.725,
"theoretical_loss": 3.6800159571099735,
"tokens_seen": 916062208
},
{
"epoch": 0.28,
"learning_rate": 0.0003648645937813441,
"loss": 2.8414,
"theoretical_loss": 3.679990495912614,
"tokens_seen": 916127744
},
{
"epoch": 0.28,
"learning_rate": 0.0003648545636910732,
"loss": 2.8541,
"theoretical_loss": 3.679965037046527,
"tokens_seen": 916193280
},
{
"epoch": 0.28,
"learning_rate": 0.00036484453360080245,
"loss": 2.9698,
"theoretical_loss": 3.67993958051133,
"tokens_seen": 916258816
},
{
"epoch": 0.28,
"learning_rate": 0.0003648345035105316,
"loss": 2.8537,
"theoretical_loss": 3.6799141263066444,
"tokens_seen": 916324352
},
{
"epoch": 0.28,
"learning_rate": 0.0003648244734202608,
"loss": 2.8458,
"theoretical_loss": 3.67988867443209,
"tokens_seen": 916389888
},
{
"epoch": 0.28,
"learning_rate": 0.00036481444332999,
"loss": 2.7542,
"theoretical_loss": 3.6798632248872862,
"tokens_seen": 916455424
},
{
"epoch": 0.28,
"learning_rate": 0.0003648044132397192,
"loss": 2.8072,
"theoretical_loss": 3.679837777671854,
"tokens_seen": 916520960
},
{
"epoch": 0.28,
"learning_rate": 0.00036479438314944836,
"loss": 2.7023,
"theoretical_loss": 3.6798123327854126,
"tokens_seen": 916586496
},
{
"epoch": 0.28,
"learning_rate": 0.00036478435305917754,
"loss": 3.0474,
"theoretical_loss": 3.679786890227583,
"tokens_seen": 916652032
},
{
"epoch": 0.28,
"learning_rate": 0.0003647743229689067,
"loss": 2.6315,
"theoretical_loss": 3.6797614499979865,
"tokens_seen": 916717568
},
{
"epoch": 0.28,
"learning_rate": 0.00036476429287863595,
"loss": 2.541,
"theoretical_loss": 3.6797360120962423,
"tokens_seen": 916783104
},
{
"epoch": 0.28,
"learning_rate": 0.0003647542627883651,
"loss": 2.6416,
"theoretical_loss": 3.679710576521971,
"tokens_seen": 916848640
},
{
"epoch": 0.28,
"learning_rate": 0.0003647442326980943,
"loss": 2.5852,
"theoretical_loss": 3.679685143274794,
"tokens_seen": 916914176
},
{
"epoch": 0.28,
"learning_rate": 0.0003647342026078235,
"loss": 2.7403,
"theoretical_loss": 3.679659712354332,
"tokens_seen": 916979712
},
{
"epoch": 0.28,
"learning_rate": 0.0003647241725175527,
"loss": 2.9413,
"theoretical_loss": 3.6796342837602056,
"tokens_seen": 917045248
},
{
"epoch": 0.28,
"learning_rate": 0.00036471414242728186,
"loss": 2.9572,
"theoretical_loss": 3.6796088574920356,
"tokens_seen": 917110784
},
{
"epoch": 0.28,
"learning_rate": 0.00036470411233701104,
"loss": 2.9489,
"theoretical_loss": 3.6795834335494444,
"tokens_seen": 917176320
},
{
"epoch": 0.28,
"learning_rate": 0.0003646940822467402,
"loss": 2.7406,
"theoretical_loss": 3.679558011932051,
"tokens_seen": 917241856
},
{
"epoch": 0.28,
"learning_rate": 0.00036468405215646946,
"loss": 2.58,
"theoretical_loss": 3.679532592639478,
"tokens_seen": 917307392
},
{
"epoch": 0.28,
"learning_rate": 0.0003646740220661986,
"loss": 2.8286,
"theoretical_loss": 3.6795071756713464,
"tokens_seen": 917372928
},
{
"epoch": 0.28,
"learning_rate": 0.0003646639919759278,
"loss": 2.6719,
"theoretical_loss": 3.6794817610272776,
"tokens_seen": 917438464
},
{
"epoch": 0.28,
"objective/train/docs_used": 527026,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6163148880004883,
"objective/train/theoretical_loss": 3.6794563487068936,
"objective/train/tokens_used": 937964000,
"theoretical_loss": 3.6794563487068936,
"tokens_seen": 917504000
},
{
"epoch": 0.28,
"learning_rate": 0.00036465396188565695,
"loss": 2.7765,
"theoretical_loss": 3.6794563487068936,
"tokens_seen": 917504000
},
{
"epoch": 0.28,
"learning_rate": 0.0003646439317953862,
"loss": 2.7137,
"theoretical_loss": 3.679430938709815,
"tokens_seen": 917569536
},
{
"epoch": 0.28,
"learning_rate": 0.00036463390170511536,
"loss": 2.7848,
"theoretical_loss": 3.6794055310356644,
"tokens_seen": 917635072
},
{
"epoch": 0.28,
"learning_rate": 0.00036462387161484454,
"loss": 2.7791,
"theoretical_loss": 3.6793801256840633,
"tokens_seen": 917700608
},
{
"epoch": 0.28,
"learning_rate": 0.0003646138415245737,
"loss": 2.7892,
"theoretical_loss": 3.679354722654633,
"tokens_seen": 917766144
},
{
"epoch": 0.28,
"learning_rate": 0.00036460381143430296,
"loss": 2.7536,
"theoretical_loss": 3.6793293219469962,
"tokens_seen": 917831680
},
{
"epoch": 0.28,
"learning_rate": 0.0003645937813440321,
"loss": 2.6171,
"theoretical_loss": 3.6793039235607745,
"tokens_seen": 917897216
},
{
"epoch": 0.28,
"learning_rate": 0.0003645837512537613,
"loss": 2.7994,
"theoretical_loss": 3.6792785274955904,
"tokens_seen": 917962752
},
{
"epoch": 0.28,
"learning_rate": 0.00036457372116349045,
"loss": 2.9293,
"theoretical_loss": 3.6792531337510654,
"tokens_seen": 918028288
},
{
"epoch": 0.28,
"learning_rate": 0.0003645636910732197,
"loss": 2.69,
"theoretical_loss": 3.6792277423268223,
"tokens_seen": 918093824
},
{
"epoch": 0.28,
"learning_rate": 0.00036455366098294887,
"loss": 2.6624,
"theoretical_loss": 3.6792023532224833,
"tokens_seen": 918159360
},
{
"epoch": 0.28,
"learning_rate": 0.00036454363089267805,
"loss": 2.8763,
"theoretical_loss": 3.6791769664376712,
"tokens_seen": 918224896
},
{
"epoch": 0.28,
"learning_rate": 0.00036453360080240723,
"loss": 2.9066,
"theoretical_loss": 3.6791515819720084,
"tokens_seen": 918290432
},
{
"epoch": 0.28,
"learning_rate": 0.0003645235707121364,
"loss": 2.58,
"theoretical_loss": 3.679126199825118,
"tokens_seen": 918355968
},
{
"epoch": 0.28,
"learning_rate": 0.0003645135406218656,
"loss": 2.627,
"theoretical_loss": 3.6791008199966218,
"tokens_seen": 918421504
},
{
"epoch": 0.28,
"learning_rate": 0.0003645035105315948,
"loss": 2.7983,
"theoretical_loss": 3.679075442486143,
"tokens_seen": 918487040
},
{
"epoch": 0.28,
"learning_rate": 0.00036449348044132395,
"loss": 2.8331,
"theoretical_loss": 3.6790500672933044,
"tokens_seen": 918552576
},
{
"epoch": 0.28,
"learning_rate": 0.0003644834503510532,
"loss": 2.7077,
"theoretical_loss": 3.6790246944177296,
"tokens_seen": 918618112
},
{
"epoch": 0.28,
"learning_rate": 0.0003644734202607823,
"loss": 2.7229,
"theoretical_loss": 3.678999323859041,
"tokens_seen": 918683648
},
{
"epoch": 0.28,
"learning_rate": 0.00036446339017051155,
"loss": 2.7356,
"theoretical_loss": 3.678973955616862,
"tokens_seen": 918749184
},
{
"epoch": 0.28,
"learning_rate": 0.00036445336008024073,
"loss": 2.7679,
"theoretical_loss": 3.6789485896908163,
"tokens_seen": 918814720
},
{
"epoch": 0.28,
"learning_rate": 0.0003644433299899699,
"loss": 2.4938,
"theoretical_loss": 3.678923226080527,
"tokens_seen": 918880256
},
{
"epoch": 0.28,
"learning_rate": 0.0003644332998996991,
"loss": 2.8542,
"theoretical_loss": 3.6788978647856174,
"tokens_seen": 918945792
},
{
"epoch": 0.28,
"learning_rate": 0.00036442326980942833,
"loss": 2.7272,
"theoretical_loss": 3.678872505805711,
"tokens_seen": 919011328
},
{
"epoch": 0.28,
"learning_rate": 0.00036441323971915746,
"loss": 2.8887,
"theoretical_loss": 3.678847149140431,
"tokens_seen": 919076864
},
{
"epoch": 0.28,
"objective/train/docs_used": 528220,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8550822734832764,
"objective/train/theoretical_loss": 3.678821794789402,
"objective/train/tokens_used": 939602400,
"theoretical_loss": 3.678821794789402,
"tokens_seen": 919142400
},
{
"epoch": 0.28,
"learning_rate": 0.0003644032096288867,
"loss": 2.9183,
"theoretical_loss": 3.678821794789402,
"tokens_seen": 919142400
},
{
"epoch": 0.28,
"learning_rate": 0.0003643931795386158,
"loss": 2.9376,
"theoretical_loss": 3.6787964427522475,
"tokens_seen": 919207936
},
{
"epoch": 0.28,
"learning_rate": 0.00036438314944834505,
"loss": 2.9797,
"theoretical_loss": 3.678771093028591,
"tokens_seen": 919273472
},
{
"epoch": 0.28,
"learning_rate": 0.00036437311935807424,
"loss": 2.8196,
"theoretical_loss": 3.678745745618057,
"tokens_seen": 919339008
},
{
"epoch": 0.28,
"learning_rate": 0.0003643630892678034,
"loss": 2.6328,
"theoretical_loss": 3.6787204005202687,
"tokens_seen": 919404544
},
{
"epoch": 0.28,
"learning_rate": 0.0003643530591775326,
"loss": 2.8675,
"theoretical_loss": 3.678695057734852,
"tokens_seen": 919470080
},
{
"epoch": 0.28,
"learning_rate": 0.0003643430290872618,
"loss": 2.6186,
"theoretical_loss": 3.678669717261429,
"tokens_seen": 919535616
},
{
"epoch": 0.28,
"learning_rate": 0.00036433299899699096,
"loss": 2.7025,
"theoretical_loss": 3.6786443790996257,
"tokens_seen": 919601152
},
{
"epoch": 0.28,
"learning_rate": 0.0003643229689067202,
"loss": 2.6529,
"theoretical_loss": 3.6786190432490655,
"tokens_seen": 919666688
},
{
"epoch": 0.28,
"learning_rate": 0.0003643129388164493,
"loss": 2.7316,
"theoretical_loss": 3.6785937097093733,
"tokens_seen": 919732224
},
{
"epoch": 0.28,
"learning_rate": 0.00036430290872617856,
"loss": 2.6018,
"theoretical_loss": 3.6785683784801737,
"tokens_seen": 919797760
},
{
"epoch": 0.28,
"learning_rate": 0.0003642928786359077,
"loss": 2.6559,
"theoretical_loss": 3.678543049561091,
"tokens_seen": 919863296
},
{
"epoch": 0.28,
"learning_rate": 0.0003642828485456369,
"loss": 2.8535,
"theoretical_loss": 3.6785177229517507,
"tokens_seen": 919928832
},
{
"epoch": 0.28,
"learning_rate": 0.0003642728184553661,
"loss": 2.5945,
"theoretical_loss": 3.678492398651777,
"tokens_seen": 919994368
},
{
"epoch": 0.28,
"learning_rate": 0.0003642627883650953,
"loss": 2.5138,
"theoretical_loss": 3.678467076660795,
"tokens_seen": 920059904
},
{
"epoch": 0.28,
"learning_rate": 0.00036425275827482446,
"loss": 2.7882,
"theoretical_loss": 3.6784417569784296,
"tokens_seen": 920125440
},
{
"epoch": 0.28,
"learning_rate": 0.0003642427281845537,
"loss": 2.7772,
"theoretical_loss": 3.6784164396043066,
"tokens_seen": 920190976
},
{
"epoch": 0.28,
"learning_rate": 0.0003642326980942828,
"loss": 2.6346,
"theoretical_loss": 3.6783911245380505,
"tokens_seen": 920256512
},
{
"epoch": 0.28,
"learning_rate": 0.00036422266800401206,
"loss": 2.7973,
"theoretical_loss": 3.678365811779287,
"tokens_seen": 920322048
},
{
"epoch": 0.28,
"learning_rate": 0.0003642126379137412,
"loss": 2.8444,
"theoretical_loss": 3.6783405013276402,
"tokens_seen": 920387584
},
{
"epoch": 0.28,
"learning_rate": 0.0003642026078234704,
"loss": 2.6365,
"theoretical_loss": 3.678315193182738,
"tokens_seen": 920453120
},
{
"epoch": 0.28,
"learning_rate": 0.0003641925777331996,
"loss": 2.6578,
"theoretical_loss": 3.678289887344204,
"tokens_seen": 920518656
},
{
"epoch": 0.28,
"learning_rate": 0.0003641825476429288,
"loss": 2.8046,
"theoretical_loss": 3.6782645838116643,
"tokens_seen": 920584192
},
{
"epoch": 0.28,
"learning_rate": 0.00036417251755265797,
"loss": 2.6981,
"theoretical_loss": 3.6782392825847445,
"tokens_seen": 920649728
},
{
"epoch": 0.28,
"learning_rate": 0.00036416248746238715,
"loss": 2.7028,
"theoretical_loss": 3.6782139836630714,
"tokens_seen": 920715264
},
{
"epoch": 0.28,
"objective/train/docs_used": 528886,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8424010276794434,
"objective/train/theoretical_loss": 3.6781886870462692,
"objective/train/tokens_used": 941240800,
"theoretical_loss": 3.6781886870462692,
"tokens_seen": 920780800
},
{
"epoch": 0.28,
"learning_rate": 0.00036415245737211633,
"loss": 2.6055,
"theoretical_loss": 3.6781886870462692,
"tokens_seen": 920780800
},
{
"epoch": 0.28,
"learning_rate": 0.00036414242728184556,
"loss": 2.9502,
"theoretical_loss": 3.678163392733966,
"tokens_seen": 920846336
},
{
"epoch": 0.28,
"learning_rate": 0.0003641323971915747,
"loss": 2.988,
"theoretical_loss": 3.6781381007257856,
"tokens_seen": 920911872
},
{
"epoch": 0.28,
"learning_rate": 0.0003641223671013039,
"loss": 2.8284,
"theoretical_loss": 3.6781128110213555,
"tokens_seen": 920977408
},
{
"epoch": 0.28,
"learning_rate": 0.00036411233701103316,
"loss": 2.8981,
"theoretical_loss": 3.6780875236203023,
"tokens_seen": 921042944
},
{
"epoch": 0.28,
"learning_rate": 0.0003641023069207623,
"loss": 2.8117,
"theoretical_loss": 3.6780622385222514,
"tokens_seen": 921108480
},
{
"epoch": 0.28,
"learning_rate": 0.0003640922768304915,
"loss": 2.6993,
"theoretical_loss": 3.6780369557268298,
"tokens_seen": 921174016
},
{
"epoch": 0.28,
"learning_rate": 0.00036408224674022065,
"loss": 2.6418,
"theoretical_loss": 3.6780116752336633,
"tokens_seen": 921239552
},
{
"epoch": 0.28,
"learning_rate": 0.0003640722166499499,
"loss": 2.697,
"theoretical_loss": 3.6779863970423796,
"tokens_seen": 921305088
},
{
"epoch": 0.28,
"learning_rate": 0.00036406218655967907,
"loss": 2.9485,
"theoretical_loss": 3.677961121152604,
"tokens_seen": 921370624
},
{
"epoch": 0.28,
"learning_rate": 0.00036405215646940825,
"loss": 2.9066,
"theoretical_loss": 3.677935847563965,
"tokens_seen": 921436160
},
{
"epoch": 0.28,
"learning_rate": 0.00036404212637913743,
"loss": 2.858,
"theoretical_loss": 3.677910576276088,
"tokens_seen": 921501696
},
{
"epoch": 0.28,
"learning_rate": 0.0003640320962888666,
"loss": 2.5784,
"theoretical_loss": 3.6778853072886006,
"tokens_seen": 921567232
},
{
"epoch": 0.28,
"learning_rate": 0.0003640220661985958,
"loss": 2.8853,
"theoretical_loss": 3.6778600406011295,
"tokens_seen": 921632768
},
{
"epoch": 0.28,
"learning_rate": 0.000364012036108325,
"loss": 2.8332,
"theoretical_loss": 3.6778347762133023,
"tokens_seen": 921698304
},
{
"epoch": 0.28,
"learning_rate": 0.00036400200601805415,
"loss": 2.6464,
"theoretical_loss": 3.677809514124746,
"tokens_seen": 921763840
},
{
"epoch": 0.28,
"learning_rate": 0.0003639919759277834,
"loss": 2.9439,
"theoretical_loss": 3.677784254335087,
"tokens_seen": 921829376
},
{
"epoch": 0.28,
"learning_rate": 0.0003639819458375125,
"loss": 2.7961,
"theoretical_loss": 3.677758996843955,
"tokens_seen": 921894912
},
{
"epoch": 0.28,
"learning_rate": 0.00036397191574724175,
"loss": 2.905,
"theoretical_loss": 3.6777337416509743,
"tokens_seen": 921960448
},
{
"epoch": 0.28,
"learning_rate": 0.00036396188565697093,
"loss": 2.9192,
"theoretical_loss": 3.6777084887557754,
"tokens_seen": 922025984
},
{
"epoch": 0.28,
"learning_rate": 0.0003639518555667001,
"loss": 2.7735,
"theoretical_loss": 3.677683238157984,
"tokens_seen": 922091520
},
{
"epoch": 0.28,
"learning_rate": 0.0003639418254764293,
"loss": 2.886,
"theoretical_loss": 3.677657989857229,
"tokens_seen": 922157056
},
{
"epoch": 0.28,
"learning_rate": 0.00036393179538615853,
"loss": 2.7154,
"theoretical_loss": 3.6776327438531373,
"tokens_seen": 922222592
},
{
"epoch": 0.28,
"learning_rate": 0.00036392176529588766,
"loss": 2.9479,
"theoretical_loss": 3.677607500145337,
"tokens_seen": 922288128
},
{
"epoch": 0.28,
"learning_rate": 0.0003639117352056169,
"loss": 2.5411,
"theoretical_loss": 3.6775822587334566,
"tokens_seen": 922353664
},
{
"epoch": 0.28,
"objective/train/docs_used": 530064,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0876314640045166,
"objective/train/theoretical_loss": 3.677557019617124,
"objective/train/tokens_used": 942879200,
"theoretical_loss": 3.677557019617124,
"tokens_seen": 922419200
},
{
"epoch": 0.28,
"learning_rate": 0.000363901705115346,
"loss": 2.8363,
"theoretical_loss": 3.677557019617124,
"tokens_seen": 922419200
},
{
"epoch": 0.28,
"learning_rate": 0.00036389167502507525,
"loss": 2.7629,
"theoretical_loss": 3.6775317827959673,
"tokens_seen": 922484736
},
{
"epoch": 0.28,
"learning_rate": 0.00036388164493480444,
"loss": 2.5874,
"theoretical_loss": 3.6775065482696148,
"tokens_seen": 922550272
},
{
"epoch": 0.28,
"learning_rate": 0.0003638716148445336,
"loss": 2.656,
"theoretical_loss": 3.677481316037694,
"tokens_seen": 922615808
},
{
"epoch": 0.28,
"learning_rate": 0.0003638615847542628,
"loss": 2.764,
"theoretical_loss": 3.6774560860998347,
"tokens_seen": 922681344
},
{
"epoch": 0.28,
"learning_rate": 0.000363851554663992,
"loss": 2.7882,
"theoretical_loss": 3.6774308584556645,
"tokens_seen": 922746880
},
{
"epoch": 0.28,
"learning_rate": 0.00036384152457372116,
"loss": 2.7684,
"theoretical_loss": 3.6774056331048124,
"tokens_seen": 922812416
},
{
"epoch": 0.28,
"learning_rate": 0.0003638314944834504,
"loss": 2.6818,
"theoretical_loss": 3.677380410046907,
"tokens_seen": 922877952
},
{
"epoch": 0.28,
"learning_rate": 0.0003638214643931795,
"loss": 2.7185,
"theoretical_loss": 3.6773551892815766,
"tokens_seen": 922943488
},
{
"epoch": 0.28,
"learning_rate": 0.00036381143430290876,
"loss": 2.7851,
"theoretical_loss": 3.677329970808451,
"tokens_seen": 923009024
},
{
"epoch": 0.28,
"learning_rate": 0.0003638014042126379,
"loss": 2.6912,
"theoretical_loss": 3.6773047546271584,
"tokens_seen": 923074560
},
{
"epoch": 0.28,
"learning_rate": 0.0003637913741223671,
"loss": 2.6723,
"theoretical_loss": 3.677279540737328,
"tokens_seen": 923140096
},
{
"epoch": 0.28,
"learning_rate": 0.0003637813440320963,
"loss": 2.7705,
"theoretical_loss": 3.677254329138589,
"tokens_seen": 923205632
},
{
"epoch": 0.28,
"learning_rate": 0.0003637713139418255,
"loss": 2.7884,
"theoretical_loss": 3.67722911983057,
"tokens_seen": 923271168
},
{
"epoch": 0.28,
"learning_rate": 0.00036376128385155466,
"loss": 2.7033,
"theoretical_loss": 3.6772039128129013,
"tokens_seen": 923336704
},
{
"epoch": 0.28,
"learning_rate": 0.0003637512537612839,
"loss": 2.673,
"theoretical_loss": 3.6771787080852123,
"tokens_seen": 923402240
},
{
"epoch": 0.28,
"learning_rate": 0.000363741223671013,
"loss": 2.6294,
"theoretical_loss": 3.677153505647132,
"tokens_seen": 923467776
},
{
"epoch": 0.28,
"learning_rate": 0.00036373119358074226,
"loss": 2.7855,
"theoretical_loss": 3.6771283054982886,
"tokens_seen": 923533312
},
{
"epoch": 0.28,
"learning_rate": 0.0003637211634904714,
"loss": 2.5843,
"theoretical_loss": 3.677103107638314,
"tokens_seen": 923598848
},
{
"epoch": 0.28,
"learning_rate": 0.0003637111334002006,
"loss": 2.6338,
"theoretical_loss": 3.677077912066837,
"tokens_seen": 923664384
},
{
"epoch": 0.28,
"learning_rate": 0.0003637011033099298,
"loss": 2.6755,
"theoretical_loss": 3.677052718783487,
"tokens_seen": 923729920
},
{
"epoch": 0.28,
"learning_rate": 0.000363691073219659,
"loss": 2.6902,
"theoretical_loss": 3.6770275277878945,
"tokens_seen": 923795456
},
{
"epoch": 0.28,
"learning_rate": 0.00036368104312938817,
"loss": 2.942,
"theoretical_loss": 3.677002339079689,
"tokens_seen": 923860992
},
{
"epoch": 0.28,
"learning_rate": 0.00036367101303911735,
"loss": 2.5199,
"theoretical_loss": 3.676977152658501,
"tokens_seen": 923926528
},
{
"epoch": 0.28,
"learning_rate": 0.00036366098294884653,
"loss": 2.7769,
"theoretical_loss": 3.6769519685239604,
"tokens_seen": 923992064
},
{
"epoch": 0.28,
"objective/train/docs_used": 530632,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.187617063522339,
"objective/train/theoretical_loss": 3.676926786675698,
"objective/train/tokens_used": 944517600,
"theoretical_loss": 3.676926786675698,
"tokens_seen": 924057600
},
{
"epoch": 0.28,
"learning_rate": 0.00036365095285857576,
"loss": 2.6898,
"theoretical_loss": 3.676926786675698,
"tokens_seen": 924057600
},
{
"epoch": 0.28,
"learning_rate": 0.0003636409227683049,
"loss": 2.7456,
"theoretical_loss": 3.6769016071133427,
"tokens_seen": 924123136
},
{
"epoch": 0.28,
"learning_rate": 0.0003636308926780341,
"loss": 2.6839,
"theoretical_loss": 3.6768764298365264,
"tokens_seen": 924188672
},
{
"epoch": 0.28,
"learning_rate": 0.00036362086258776325,
"loss": 2.892,
"theoretical_loss": 3.6768512548448786,
"tokens_seen": 924254208
},
{
"epoch": 0.28,
"learning_rate": 0.0003636108324974925,
"loss": 2.4548,
"theoretical_loss": 3.6768260821380307,
"tokens_seen": 924319744
},
{
"epoch": 0.28,
"learning_rate": 0.00036360080240722167,
"loss": 2.5309,
"theoretical_loss": 3.6768009117156124,
"tokens_seen": 924385280
},
{
"epoch": 0.28,
"learning_rate": 0.00036359077231695085,
"loss": 2.7414,
"theoretical_loss": 3.6767757435772555,
"tokens_seen": 924450816
},
{
"epoch": 0.28,
"learning_rate": 0.00036358074222668003,
"loss": 2.6921,
"theoretical_loss": 3.67675057772259,
"tokens_seen": 924516352
},
{
"epoch": 0.28,
"learning_rate": 0.00036357071213640927,
"loss": 2.9353,
"theoretical_loss": 3.6767254141512478,
"tokens_seen": 924581888
},
{
"epoch": 0.28,
"learning_rate": 0.0003635606820461384,
"loss": 2.7932,
"theoretical_loss": 3.6767002528628585,
"tokens_seen": 924647424
},
{
"epoch": 0.28,
"learning_rate": 0.00036355065195586763,
"loss": 2.8327,
"theoretical_loss": 3.676675093857054,
"tokens_seen": 924712960
},
{
"epoch": 0.28,
"learning_rate": 0.00036354062186559676,
"loss": 2.7044,
"theoretical_loss": 3.676649937133466,
"tokens_seen": 924778496
},
{
"epoch": 0.28,
"learning_rate": 0.000363530591775326,
"loss": 2.8093,
"theoretical_loss": 3.6766247826917247,
"tokens_seen": 924844032
},
{
"epoch": 0.28,
"learning_rate": 0.00036352056168505517,
"loss": 2.6249,
"theoretical_loss": 3.676599630531462,
"tokens_seen": 924909568
},
{
"epoch": 0.28,
"learning_rate": 0.00036351053159478435,
"loss": 2.8067,
"theoretical_loss": 3.6765744806523095,
"tokens_seen": 924975104
},
{
"epoch": 0.28,
"learning_rate": 0.00036350050150451353,
"loss": 2.7558,
"theoretical_loss": 3.6765493330538987,
"tokens_seen": 925040640
},
{
"epoch": 0.28,
"learning_rate": 0.0003634904714142427,
"loss": 2.6869,
"theoretical_loss": 3.6765241877358603,
"tokens_seen": 925106176
},
{
"epoch": 0.28,
"learning_rate": 0.0003634804413239719,
"loss": 3.0124,
"theoretical_loss": 3.6764990446978274,
"tokens_seen": 925171712
},
{
"epoch": 0.28,
"learning_rate": 0.00036347041123370113,
"loss": 2.6412,
"theoretical_loss": 3.6764739039394305,
"tokens_seen": 925237248
},
{
"epoch": 0.28,
"learning_rate": 0.00036346038114343026,
"loss": 2.8575,
"theoretical_loss": 3.676448765460303,
"tokens_seen": 925302784
},
{
"epoch": 0.28,
"learning_rate": 0.0003634503510531595,
"loss": 2.6551,
"theoretical_loss": 3.676423629260075,
"tokens_seen": 925368320
},
{
"epoch": 0.28,
"learning_rate": 0.0003634403209628886,
"loss": 2.966,
"theoretical_loss": 3.67639849533838,
"tokens_seen": 925433856
},
{
"epoch": 0.28,
"learning_rate": 0.00036343029087261786,
"loss": 2.5963,
"theoretical_loss": 3.676373363694849,
"tokens_seen": 925499392
},
{
"epoch": 0.28,
"learning_rate": 0.00036342026078234704,
"loss": 2.807,
"theoretical_loss": 3.676348234329115,
"tokens_seen": 925564928
},
{
"epoch": 0.28,
"learning_rate": 0.0003634102306920762,
"loss": 2.8617,
"theoretical_loss": 3.67632310724081,
"tokens_seen": 925630464
},
{
"epoch": 0.28,
"objective/train/docs_used": 532017,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.651747703552246,
"objective/train/theoretical_loss": 3.6762979824295665,
"objective/train/tokens_used": 946156000,
"theoretical_loss": 3.6762979824295665,
"tokens_seen": 925696000
},
{
"epoch": 0.28,
"learning_rate": 0.0003634002006018054,
"loss": 2.5877,
"theoretical_loss": 3.6762979824295665,
"tokens_seen": 925696000
},
{
"epoch": 0.28,
"learning_rate": 0.00036339017051153464,
"loss": 2.582,
"theoretical_loss": 3.6762728598950174,
"tokens_seen": 925761536
},
{
"epoch": 0.28,
"learning_rate": 0.00036338014042126376,
"loss": 2.7674,
"theoretical_loss": 3.676247739636794,
"tokens_seen": 925827072
},
{
"epoch": 0.28,
"learning_rate": 0.000363370110330993,
"loss": 2.4674,
"theoretical_loss": 3.6762226216545297,
"tokens_seen": 925892608
},
{
"epoch": 0.28,
"learning_rate": 0.0003633600802407222,
"loss": 2.9286,
"theoretical_loss": 3.676197505947858,
"tokens_seen": 925958144
},
{
"epoch": 0.28,
"learning_rate": 0.00036335005015045136,
"loss": 2.6354,
"theoretical_loss": 3.67617239251641,
"tokens_seen": 926023680
},
{
"epoch": 0.28,
"learning_rate": 0.0003633400200601806,
"loss": 2.5666,
"theoretical_loss": 3.67614728135982,
"tokens_seen": 926089216
},
{
"epoch": 0.28,
"learning_rate": 0.0003633299899699097,
"loss": 2.6142,
"theoretical_loss": 3.6761221724777204,
"tokens_seen": 926154752
},
{
"epoch": 0.28,
"learning_rate": 0.00036331995987963896,
"loss": 2.6912,
"theoretical_loss": 3.676097065869744,
"tokens_seen": 926220288
},
{
"epoch": 0.28,
"learning_rate": 0.0003633099297893681,
"loss": 2.9076,
"theoretical_loss": 3.6760719615355244,
"tokens_seen": 926285824
},
{
"epoch": 0.28,
"learning_rate": 0.0003632998996990973,
"loss": 2.9941,
"theoretical_loss": 3.6760468594746953,
"tokens_seen": 926351360
},
{
"epoch": 0.28,
"learning_rate": 0.0003632898696088265,
"loss": 2.8999,
"theoretical_loss": 3.676021759686889,
"tokens_seen": 926416896
},
{
"epoch": 0.28,
"learning_rate": 0.0003632798395185557,
"loss": 2.7575,
"theoretical_loss": 3.6759966621717393,
"tokens_seen": 926482432
},
{
"epoch": 0.28,
"learning_rate": 0.00036326980942828486,
"loss": 2.815,
"theoretical_loss": 3.6759715669288795,
"tokens_seen": 926547968
},
{
"epoch": 0.28,
"learning_rate": 0.0003632597793380141,
"loss": 2.5852,
"theoretical_loss": 3.6759464739579437,
"tokens_seen": 926613504
},
{
"epoch": 0.28,
"learning_rate": 0.0003632497492477432,
"loss": 2.5959,
"theoretical_loss": 3.6759213832585647,
"tokens_seen": 926679040
},
{
"epoch": 0.28,
"learning_rate": 0.00036323971915747246,
"loss": 2.8012,
"theoretical_loss": 3.675896294830377,
"tokens_seen": 926744576
},
{
"epoch": 0.28,
"learning_rate": 0.0003632296890672016,
"loss": 2.6884,
"theoretical_loss": 3.6758712086730148,
"tokens_seen": 926810112
},
{
"epoch": 0.28,
"learning_rate": 0.0003632196589769308,
"loss": 2.8487,
"theoretical_loss": 3.6758461247861107,
"tokens_seen": 926875648
},
{
"epoch": 0.28,
"learning_rate": 0.00036320962888666,
"loss": 2.8187,
"theoretical_loss": 3.6758210431692993,
"tokens_seen": 926941184
},
{
"epoch": 0.28,
"learning_rate": 0.0003631995987963892,
"loss": 2.9032,
"theoretical_loss": 3.6757959638222153,
"tokens_seen": 927006720
},
{
"epoch": 0.28,
"learning_rate": 0.00036318956870611837,
"loss": 2.7618,
"theoretical_loss": 3.675770886744492,
"tokens_seen": 927072256
},
{
"epoch": 0.28,
"learning_rate": 0.00036317953861584755,
"loss": 2.7611,
"theoretical_loss": 3.675745811935764,
"tokens_seen": 927137792
},
{
"epoch": 0.28,
"learning_rate": 0.00036316950852557673,
"loss": 2.8511,
"theoretical_loss": 3.6757207393956657,
"tokens_seen": 927203328
},
{
"epoch": 0.28,
"learning_rate": 0.00036315947843530596,
"loss": 2.9554,
"theoretical_loss": 3.675695669123832,
"tokens_seen": 927268864
},
{
"epoch": 0.28,
"objective/train/docs_used": 532753,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1987879276275635,
"objective/train/theoretical_loss": 3.6756706011198963,
"objective/train/tokens_used": 947794400,
"theoretical_loss": 3.6756706011198963,
"tokens_seen": 927334400
},
{
"epoch": 0.28,
"learning_rate": 0.0003631494483450351,
"loss": 2.8745,
"theoretical_loss": 3.6756706011198963,
"tokens_seen": 927334400
},
{
"epoch": 0.28,
"learning_rate": 0.0003631394182547643,
"loss": 2.6481,
"theoretical_loss": 3.6756455353834934,
"tokens_seen": 927399936
},
{
"epoch": 0.28,
"learning_rate": 0.00036312938816449345,
"loss": 2.7057,
"theoretical_loss": 3.675620471914259,
"tokens_seen": 927465472
},
{
"epoch": 0.28,
"learning_rate": 0.0003631193580742227,
"loss": 2.7894,
"theoretical_loss": 3.6755954107118267,
"tokens_seen": 927531008
},
{
"epoch": 0.28,
"learning_rate": 0.00036310932798395187,
"loss": 2.7338,
"theoretical_loss": 3.675570351775832,
"tokens_seen": 927596544
},
{
"epoch": 0.28,
"learning_rate": 0.00036309929789368105,
"loss": 2.8098,
"theoretical_loss": 3.675545295105909,
"tokens_seen": 927662080
},
{
"epoch": 0.28,
"learning_rate": 0.00036308926780341023,
"loss": 2.7548,
"theoretical_loss": 3.6755202407016943,
"tokens_seen": 927727616
},
{
"epoch": 0.28,
"learning_rate": 0.00036307923771313947,
"loss": 2.8226,
"theoretical_loss": 3.675495188562822,
"tokens_seen": 927793152
},
{
"epoch": 0.28,
"learning_rate": 0.0003630692076228686,
"loss": 2.7664,
"theoretical_loss": 3.675470138688927,
"tokens_seen": 927858688
},
{
"epoch": 0.28,
"learning_rate": 0.00036305917753259783,
"loss": 2.8094,
"theoretical_loss": 3.6754450910796446,
"tokens_seen": 927924224
},
{
"epoch": 0.28,
"learning_rate": 0.00036304914744232696,
"loss": 2.6965,
"theoretical_loss": 3.6754200457346107,
"tokens_seen": 927989760
},
{
"epoch": 0.28,
"learning_rate": 0.0003630391173520562,
"loss": 2.6007,
"theoretical_loss": 3.6753950026534605,
"tokens_seen": 928055296
},
{
"epoch": 0.28,
"learning_rate": 0.00036302908726178537,
"loss": 2.6115,
"theoretical_loss": 3.6753699618358295,
"tokens_seen": 928120832
},
{
"epoch": 0.28,
"learning_rate": 0.00036301905717151455,
"loss": 2.7133,
"theoretical_loss": 3.675344923281353,
"tokens_seen": 928186368
},
{
"epoch": 0.28,
"learning_rate": 0.00036300902708124374,
"loss": 2.7399,
"theoretical_loss": 3.6753198869896675,
"tokens_seen": 928251904
},
{
"epoch": 0.28,
"learning_rate": 0.0003629989969909729,
"loss": 2.7635,
"theoretical_loss": 3.675294852960408,
"tokens_seen": 928317440
},
{
"epoch": 0.28,
"learning_rate": 0.0003629889669007021,
"loss": 2.7423,
"theoretical_loss": 3.6752698211932104,
"tokens_seen": 928382976
},
{
"epoch": 0.28,
"learning_rate": 0.00036297893681043133,
"loss": 2.7513,
"theoretical_loss": 3.675244791687711,
"tokens_seen": 928448512
},
{
"epoch": 0.28,
"learning_rate": 0.00036296890672016046,
"loss": 2.7648,
"theoretical_loss": 3.675219764443546,
"tokens_seen": 928514048
},
{
"epoch": 0.28,
"learning_rate": 0.0003629588766298897,
"loss": 2.8612,
"theoretical_loss": 3.6751947394603506,
"tokens_seen": 928579584
},
{
"epoch": 0.28,
"learning_rate": 0.0003629488465396188,
"loss": 2.9186,
"theoretical_loss": 3.675169716737762,
"tokens_seen": 928645120
},
{
"epoch": 0.28,
"learning_rate": 0.00036293881644934806,
"loss": 2.7467,
"theoretical_loss": 3.6751446962754155,
"tokens_seen": 928710656
},
{
"epoch": 0.28,
"learning_rate": 0.00036292878635907724,
"loss": 2.5148,
"theoretical_loss": 3.6751196780729485,
"tokens_seen": 928776192
},
{
"epoch": 0.28,
"learning_rate": 0.0003629187562688064,
"loss": 2.6044,
"theoretical_loss": 3.675094662129997,
"tokens_seen": 928841728
},
{
"epoch": 0.28,
"learning_rate": 0.0003629087261785356,
"loss": 2.8017,
"theoretical_loss": 3.6750696484461978,
"tokens_seen": 928907264
},
{
"epoch": 0.28,
"objective/train/docs_used": 534166,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8485100269317627,
"objective/train/theoretical_loss": 3.6750446370211867,
"objective/train/tokens_used": 949432800,
"theoretical_loss": 3.6750446370211867,
"tokens_seen": 928972800
},
{
"epoch": 0.28,
"learning_rate": 0.00036289869608826484,
"loss": 2.7026,
"theoretical_loss": 3.6750446370211867,
"tokens_seen": 928972800
},
{
"epoch": 0.28,
"learning_rate": 0.00036288866599799396,
"loss": 2.6785,
"theoretical_loss": 3.675019627854601,
"tokens_seen": 929038336
},
{
"epoch": 0.28,
"learning_rate": 0.0003628786359077232,
"loss": 2.755,
"theoretical_loss": 3.6749946209460775,
"tokens_seen": 929103872
},
{
"epoch": 0.28,
"learning_rate": 0.0003628686058174523,
"loss": 2.797,
"theoretical_loss": 3.6749696162952525,
"tokens_seen": 929169408
},
{
"epoch": 0.28,
"learning_rate": 0.00036285857572718156,
"loss": 2.54,
"theoretical_loss": 3.674944613901764,
"tokens_seen": 929234944
},
{
"epoch": 0.28,
"learning_rate": 0.00036284854563691074,
"loss": 2.6215,
"theoretical_loss": 3.674919613765248,
"tokens_seen": 929300480
},
{
"epoch": 0.28,
"learning_rate": 0.0003628385155466399,
"loss": 2.6767,
"theoretical_loss": 3.674894615885342,
"tokens_seen": 929366016
},
{
"epoch": 0.28,
"learning_rate": 0.0003628284854563691,
"loss": 2.4901,
"theoretical_loss": 3.6748696202616835,
"tokens_seen": 929431552
},
{
"epoch": 0.28,
"learning_rate": 0.0003628184553660983,
"loss": 2.8441,
"theoretical_loss": 3.6748446268939094,
"tokens_seen": 929497088
},
{
"epoch": 0.28,
"learning_rate": 0.00036280842527582747,
"loss": 2.5351,
"theoretical_loss": 3.674819635781658,
"tokens_seen": 929562624
},
{
"epoch": 0.28,
"learning_rate": 0.0003627983951855567,
"loss": 2.7771,
"theoretical_loss": 3.674794646924565,
"tokens_seen": 929628160
},
{
"epoch": 0.28,
"learning_rate": 0.00036278836509528583,
"loss": 2.9465,
"theoretical_loss": 3.674769660322269,
"tokens_seen": 929693696
},
{
"epoch": 0.28,
"learning_rate": 0.00036277833500501506,
"loss": 2.5454,
"theoretical_loss": 3.6747446759744076,
"tokens_seen": 929759232
},
{
"epoch": 0.28,
"learning_rate": 0.00036276830491474424,
"loss": 2.8287,
"theoretical_loss": 3.6747196938806184,
"tokens_seen": 929824768
},
{
"epoch": 0.28,
"learning_rate": 0.0003627582748244734,
"loss": 2.8148,
"theoretical_loss": 3.6746947140405393,
"tokens_seen": 929890304
},
{
"epoch": 0.28,
"learning_rate": 0.0003627482447342026,
"loss": 2.9313,
"theoretical_loss": 3.6746697364538083,
"tokens_seen": 929955840
},
{
"epoch": 0.28,
"learning_rate": 0.0003627382146439318,
"loss": 2.6544,
"theoretical_loss": 3.6746447611200628,
"tokens_seen": 930021376
},
{
"epoch": 0.28,
"learning_rate": 0.00036272818455366097,
"loss": 2.8565,
"theoretical_loss": 3.6746197880389415,
"tokens_seen": 930086912
},
{
"epoch": 0.28,
"learning_rate": 0.0003627181544633902,
"loss": 2.6709,
"theoretical_loss": 3.674594817210082,
"tokens_seen": 930152448
},
{
"epoch": 0.28,
"learning_rate": 0.00036270812437311933,
"loss": 2.69,
"theoretical_loss": 3.674569848633123,
"tokens_seen": 930217984
},
{
"epoch": 0.28,
"learning_rate": 0.00036269809428284857,
"loss": 2.7869,
"theoretical_loss": 3.674544882307702,
"tokens_seen": 930283520
},
{
"epoch": 0.28,
"learning_rate": 0.0003626880641925777,
"loss": 2.7294,
"theoretical_loss": 3.6745199182334582,
"tokens_seen": 930349056
},
{
"epoch": 0.28,
"learning_rate": 0.00036267803410230693,
"loss": 2.7548,
"theoretical_loss": 3.6744949564100295,
"tokens_seen": 930414592
},
{
"epoch": 0.28,
"learning_rate": 0.0003626680040120361,
"loss": 2.6677,
"theoretical_loss": 3.674469996837055,
"tokens_seen": 930480128
},
{
"epoch": 0.28,
"learning_rate": 0.0003626579739217653,
"loss": 3.0318,
"theoretical_loss": 3.674445039514173,
"tokens_seen": 930545664
},
{
"epoch": 0.28,
"objective/train/docs_used": 534638,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4507763385772705,
"objective/train/theoretical_loss": 3.6744200844410217,
"objective/train/tokens_used": 951071200,
"theoretical_loss": 3.6744200844410217,
"tokens_seen": 930611200
},
{
"epoch": 0.28,
"learning_rate": 0.00036264794383149447,
"loss": 2.6932,
"theoretical_loss": 3.6744200844410217,
"tokens_seen": 930611200
},
{
"epoch": 0.28,
"learning_rate": 0.00036263791374122365,
"loss": 2.9109,
"theoretical_loss": 3.674395131617241,
"tokens_seen": 930676736
},
{
"epoch": 0.28,
"learning_rate": 0.00036262788365095283,
"loss": 2.946,
"theoretical_loss": 3.674370181042469,
"tokens_seen": 930742272
},
{
"epoch": 0.28,
"learning_rate": 0.00036261785356068207,
"loss": 2.5546,
"theoretical_loss": 3.6743452327163446,
"tokens_seen": 930807808
},
{
"epoch": 0.28,
"learning_rate": 0.00036260782347041125,
"loss": 2.5891,
"theoretical_loss": 3.6743202866385074,
"tokens_seen": 930873344
},
{
"epoch": 0.28,
"learning_rate": 0.00036259779338014043,
"loss": 2.7027,
"theoretical_loss": 3.6742953428085965,
"tokens_seen": 930938880
},
{
"epoch": 0.28,
"learning_rate": 0.00036258776328986967,
"loss": 2.5516,
"theoretical_loss": 3.6742704012262504,
"tokens_seen": 931004416
},
{
"epoch": 0.28,
"learning_rate": 0.0003625777331995988,
"loss": 2.866,
"theoretical_loss": 3.6742454618911085,
"tokens_seen": 931069952
},
{
"epoch": 0.28,
"learning_rate": 0.00036256770310932803,
"loss": 2.7333,
"theoretical_loss": 3.674220524802811,
"tokens_seen": 931135488
},
{
"epoch": 0.28,
"learning_rate": 0.00036255767301905716,
"loss": 2.7304,
"theoretical_loss": 3.6741955899609966,
"tokens_seen": 931201024
},
{
"epoch": 0.28,
"learning_rate": 0.0003625476429287864,
"loss": 2.6393,
"theoretical_loss": 3.674170657365305,
"tokens_seen": 931266560
},
{
"epoch": 0.28,
"learning_rate": 0.0003625376128385156,
"loss": 2.8481,
"theoretical_loss": 3.674145727015376,
"tokens_seen": 931332096
},
{
"epoch": 0.28,
"learning_rate": 0.00036252758274824475,
"loss": 2.797,
"theoretical_loss": 3.6741207989108493,
"tokens_seen": 931397632
},
{
"epoch": 0.28,
"learning_rate": 0.00036251755265797394,
"loss": 2.6922,
"theoretical_loss": 3.674095873051365,
"tokens_seen": 931463168
},
{
"epoch": 0.28,
"learning_rate": 0.0003625075225677031,
"loss": 2.5694,
"theoretical_loss": 3.6740709494365618,
"tokens_seen": 931528704
},
{
"epoch": 0.28,
"learning_rate": 0.0003624974924774323,
"loss": 2.3978,
"theoretical_loss": 3.6740460280660807,
"tokens_seen": 931594240
},
{
"epoch": 0.28,
"learning_rate": 0.00036248746238716153,
"loss": 2.7417,
"theoretical_loss": 3.6740211089395616,
"tokens_seen": 931659776
},
{
"epoch": 0.28,
"learning_rate": 0.00036247743229689066,
"loss": 2.7392,
"theoretical_loss": 3.673996192056644,
"tokens_seen": 931725312
},
{
"epoch": 0.28,
"learning_rate": 0.0003624674022066199,
"loss": 2.4658,
"theoretical_loss": 3.6739712774169693,
"tokens_seen": 931790848
},
{
"epoch": 0.28,
"learning_rate": 0.000362457372116349,
"loss": 2.9612,
"theoretical_loss": 3.673946365020177,
"tokens_seen": 931856384
},
{
"epoch": 0.28,
"learning_rate": 0.00036244734202607826,
"loss": 2.8027,
"theoretical_loss": 3.6739214548659076,
"tokens_seen": 931921920
},
{
"epoch": 0.28,
"learning_rate": 0.00036243731193580744,
"loss": 2.5507,
"theoretical_loss": 3.6738965469538014,
"tokens_seen": 931987456
},
{
"epoch": 0.28,
"learning_rate": 0.0003624272818455366,
"loss": 2.8148,
"theoretical_loss": 3.6738716412834984,
"tokens_seen": 932052992
},
{
"epoch": 0.28,
"learning_rate": 0.0003624172517552658,
"loss": 2.5767,
"theoretical_loss": 3.6738467378546407,
"tokens_seen": 932118528
},
{
"epoch": 0.28,
"learning_rate": 0.00036240722166499504,
"loss": 2.5937,
"theoretical_loss": 3.6738218366668685,
"tokens_seen": 932184064
},
{
"epoch": 0.28,
"objective/train/docs_used": 535899,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.974701404571533,
"objective/train/theoretical_loss": 3.673796937719821,
"objective/train/tokens_used": 952709600,
"theoretical_loss": 3.673796937719821,
"tokens_seen": 932249600
},
{
"epoch": 0.28,
"learning_rate": 0.00036239719157472416,
"loss": 2.7614,
"theoretical_loss": 3.673796937719821,
"tokens_seen": 932249600
},
{
"epoch": 0.28,
"learning_rate": 0.0003623871614844534,
"loss": 2.652,
"theoretical_loss": 3.673772041013142,
"tokens_seen": 932315136
},
{
"epoch": 0.28,
"learning_rate": 0.0003623771313941825,
"loss": 2.7489,
"theoretical_loss": 3.6737471465464697,
"tokens_seen": 932380672
},
{
"epoch": 0.28,
"learning_rate": 0.00036236710130391176,
"loss": 2.9869,
"theoretical_loss": 3.673722254319446,
"tokens_seen": 932446208
},
{
"epoch": 0.28,
"learning_rate": 0.00036235707121364094,
"loss": 2.9501,
"theoretical_loss": 3.673697364331713,
"tokens_seen": 932511744
},
{
"epoch": 0.28,
"learning_rate": 0.0003623470411233701,
"loss": 2.7449,
"theoretical_loss": 3.6736724765829107,
"tokens_seen": 932577280
},
{
"epoch": 0.28,
"learning_rate": 0.0003623370110330993,
"loss": 3.0826,
"theoretical_loss": 3.6736475910726814,
"tokens_seen": 932642816
},
{
"epoch": 0.28,
"learning_rate": 0.0003623269809428285,
"loss": 2.532,
"theoretical_loss": 3.673622707800665,
"tokens_seen": 932708352
},
{
"epoch": 0.28,
"learning_rate": 0.00036231695085255767,
"loss": 2.7205,
"theoretical_loss": 3.6735978267665046,
"tokens_seen": 932773888
},
{
"epoch": 0.28,
"learning_rate": 0.0003623069207622869,
"loss": 2.7443,
"theoretical_loss": 3.6735729479698405,
"tokens_seen": 932839424
},
{
"epoch": 0.28,
"learning_rate": 0.00036229689067201603,
"loss": 2.648,
"theoretical_loss": 3.6735480714103153,
"tokens_seen": 932904960
},
{
"epoch": 0.28,
"learning_rate": 0.00036228686058174526,
"loss": 2.504,
"theoretical_loss": 3.6735231970875697,
"tokens_seen": 932970496
},
{
"epoch": 0.28,
"learning_rate": 0.00036227683049147444,
"loss": 2.7656,
"theoretical_loss": 3.6734983250012467,
"tokens_seen": 933036032
},
{
"epoch": 0.28,
"learning_rate": 0.0003622668004012036,
"loss": 2.6825,
"theoretical_loss": 3.6734734551509867,
"tokens_seen": 933101568
},
{
"epoch": 0.28,
"learning_rate": 0.0003622567703109328,
"loss": 2.8309,
"theoretical_loss": 3.6734485875364324,
"tokens_seen": 933167104
},
{
"epoch": 0.28,
"learning_rate": 0.000362246740220662,
"loss": 2.6824,
"theoretical_loss": 3.673423722157226,
"tokens_seen": 933232640
},
{
"epoch": 0.28,
"learning_rate": 0.00036223671013039117,
"loss": 2.7006,
"theoretical_loss": 3.6733988590130098,
"tokens_seen": 933298176
},
{
"epoch": 0.28,
"learning_rate": 0.0003622266800401204,
"loss": 2.583,
"theoretical_loss": 3.673373998103425,
"tokens_seen": 933363712
},
{
"epoch": 0.28,
"learning_rate": 0.00036221664994984953,
"loss": 2.68,
"theoretical_loss": 3.6733491394281153,
"tokens_seen": 933429248
},
{
"epoch": 0.28,
"learning_rate": 0.00036220661985957877,
"loss": 2.6334,
"theoretical_loss": 3.673324282986721,
"tokens_seen": 933494784
},
{
"epoch": 0.28,
"learning_rate": 0.0003621965897693079,
"loss": 2.9355,
"theoretical_loss": 3.6732994287788867,
"tokens_seen": 933560320
},
{
"epoch": 0.28,
"learning_rate": 0.00036218655967903713,
"loss": 2.7466,
"theoretical_loss": 3.673274576804254,
"tokens_seen": 933625856
},
{
"epoch": 0.28,
"learning_rate": 0.0003621765295887663,
"loss": 2.8013,
"theoretical_loss": 3.6732497270624656,
"tokens_seen": 933691392
},
{
"epoch": 0.28,
"learning_rate": 0.0003621664994984955,
"loss": 2.7259,
"theoretical_loss": 3.673224879553163,
"tokens_seen": 933756928
},
{
"epoch": 0.28,
"learning_rate": 0.00036215646940822467,
"loss": 2.5396,
"theoretical_loss": 3.6732000342759914,
"tokens_seen": 933822464
},
{
"epoch": 0.28,
"objective/train/docs_used": 536516,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.540604591369629,
"objective/train/theoretical_loss": 3.6731751912305914,
"objective/train/tokens_used": 954348000,
"theoretical_loss": 3.6731751912305914,
"tokens_seen": 933888000
},
{
"epoch": 0.28,
"learning_rate": 0.00036214643931795385,
"loss": 2.6151,
"theoretical_loss": 3.6731751912305914,
"tokens_seen": 933888000
},
{
"epoch": 0.28,
"learning_rate": 0.00036213640922768303,
"loss": 2.8447,
"theoretical_loss": 3.6731503504166074,
"tokens_seen": 933953536
},
{
"epoch": 0.28,
"learning_rate": 0.00036212637913741227,
"loss": 2.7343,
"theoretical_loss": 3.673125511833682,
"tokens_seen": 934019072
},
{
"epoch": 0.28,
"learning_rate": 0.0003621163490471414,
"loss": 2.9437,
"theoretical_loss": 3.6731006754814572,
"tokens_seen": 934084608
},
{
"epoch": 0.28,
"learning_rate": 0.00036210631895687063,
"loss": 2.8175,
"theoretical_loss": 3.673075841359578,
"tokens_seen": 934150144
},
{
"epoch": 0.28,
"learning_rate": 0.0003620962888665998,
"loss": 2.6252,
"theoretical_loss": 3.6730510094676863,
"tokens_seen": 934215680
},
{
"epoch": 0.28,
"learning_rate": 0.000362086258776329,
"loss": 2.7038,
"theoretical_loss": 3.6730261798054262,
"tokens_seen": 934281216
},
{
"epoch": 0.28,
"learning_rate": 0.0003620762286860582,
"loss": 2.6538,
"theoretical_loss": 3.6730013523724407,
"tokens_seen": 934346752
},
{
"epoch": 0.28,
"learning_rate": 0.00036206619859578736,
"loss": 2.8343,
"theoretical_loss": 3.6729765271683736,
"tokens_seen": 934412288
},
{
"epoch": 0.28,
"learning_rate": 0.00036205616850551654,
"loss": 2.6312,
"theoretical_loss": 3.6729517041928683,
"tokens_seen": 934477824
},
{
"epoch": 0.28,
"learning_rate": 0.0003620461384152458,
"loss": 2.9034,
"theoretical_loss": 3.6729268834455686,
"tokens_seen": 934543360
},
{
"epoch": 0.28,
"learning_rate": 0.0003620361083249749,
"loss": 2.8968,
"theoretical_loss": 3.6729020649261184,
"tokens_seen": 934608896
},
{
"epoch": 0.28,
"learning_rate": 0.00036202607823470414,
"loss": 3.0807,
"theoretical_loss": 3.672877248634161,
"tokens_seen": 934674432
},
{
"epoch": 0.28,
"learning_rate": 0.00036201604814443326,
"loss": 2.777,
"theoretical_loss": 3.6728524345693403,
"tokens_seen": 934739968
},
{
"epoch": 0.28,
"learning_rate": 0.0003620060180541625,
"loss": 2.7155,
"theoretical_loss": 3.672827622731301,
"tokens_seen": 934805504
},
{
"epoch": 0.28,
"learning_rate": 0.0003619959879638917,
"loss": 2.9603,
"theoretical_loss": 3.672802813119687,
"tokens_seen": 934871040
},
{
"epoch": 0.28,
"learning_rate": 0.00036198595787362086,
"loss": 2.7647,
"theoretical_loss": 3.672778005734142,
"tokens_seen": 934936576
},
{
"epoch": 0.28,
"learning_rate": 0.00036197592778335004,
"loss": 2.9068,
"theoretical_loss": 3.672753200574311,
"tokens_seen": 935002112
},
{
"epoch": 0.28,
"learning_rate": 0.0003619658976930792,
"loss": 2.9055,
"theoretical_loss": 3.672728397639837,
"tokens_seen": 935067648
},
{
"epoch": 0.28,
"learning_rate": 0.0003619558676028084,
"loss": 2.6746,
"theoretical_loss": 3.672703596930366,
"tokens_seen": 935133184
},
{
"epoch": 0.28,
"learning_rate": 0.00036194583751253764,
"loss": 2.9933,
"theoretical_loss": 3.672678798445541,
"tokens_seen": 935198720
},
{
"epoch": 0.28,
"learning_rate": 0.00036193580742226677,
"loss": 2.7178,
"theoretical_loss": 3.6726540021850083,
"tokens_seen": 935264256
},
{
"epoch": 0.28,
"learning_rate": 0.000361925777331996,
"loss": 2.9509,
"theoretical_loss": 3.672629208148411,
"tokens_seen": 935329792
},
{
"epoch": 0.28,
"learning_rate": 0.0003619157472417252,
"loss": 2.8702,
"theoretical_loss": 3.6726044163353944,
"tokens_seen": 935395328
},
{
"epoch": 0.28,
"learning_rate": 0.00036190571715145436,
"loss": 2.8079,
"theoretical_loss": 3.6725796267456032,
"tokens_seen": 935460864
},
{
"epoch": 0.28,
"objective/train/docs_used": 537462,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.976658821105957,
"objective/train/theoretical_loss": 3.672554839378683,
"objective/train/tokens_used": 955986400,
"theoretical_loss": 3.672554839378683,
"tokens_seen": 935526400
},
{
"epoch": 0.28,
"learning_rate": 0.00036189568706118354,
"loss": 2.7469,
"theoretical_loss": 3.672554839378683,
"tokens_seen": 935526400
},
{
"epoch": 0.28,
"learning_rate": 0.0003618856569709127,
"loss": 2.7961,
"theoretical_loss": 3.6725300542342776,
"tokens_seen": 935591936
},
{
"epoch": 0.28,
"learning_rate": 0.0003618756268806419,
"loss": 3.0477,
"theoretical_loss": 3.6725052713120325,
"tokens_seen": 935657472
},
{
"epoch": 0.28,
"learning_rate": 0.00036186559679037114,
"loss": 2.7097,
"theoretical_loss": 3.6724804906115933,
"tokens_seen": 935723008
},
{
"epoch": 0.28,
"learning_rate": 0.0003618555667001003,
"loss": 2.7282,
"theoretical_loss": 3.672455712132605,
"tokens_seen": 935788544
},
{
"epoch": 0.28,
"learning_rate": 0.0003618455366098295,
"loss": 3.0653,
"theoretical_loss": 3.672430935874713,
"tokens_seen": 935854080
},
{
"epoch": 0.28,
"learning_rate": 0.0003618355065195587,
"loss": 2.7904,
"theoretical_loss": 3.6724061618375625,
"tokens_seen": 935919616
},
{
"epoch": 0.28,
"learning_rate": 0.00036182547642928787,
"loss": 2.9641,
"theoretical_loss": 3.6723813900207984,
"tokens_seen": 935985152
},
{
"epoch": 0.28,
"learning_rate": 0.0003618154463390171,
"loss": 2.583,
"theoretical_loss": 3.6723566204240674,
"tokens_seen": 936050688
},
{
"epoch": 0.28,
"learning_rate": 0.00036180541624874623,
"loss": 2.7073,
"theoretical_loss": 3.672331853047014,
"tokens_seen": 936116224
},
{
"epoch": 0.28,
"learning_rate": 0.00036179538615847546,
"loss": 2.6654,
"theoretical_loss": 3.672307087889285,
"tokens_seen": 936181760
},
{
"epoch": 0.28,
"learning_rate": 0.00036178535606820464,
"loss": 2.5865,
"theoretical_loss": 3.6722823249505256,
"tokens_seen": 936247296
},
{
"epoch": 0.28,
"learning_rate": 0.0003617753259779338,
"loss": 3.1394,
"theoretical_loss": 3.6722575642303816,
"tokens_seen": 936312832
},
{
"epoch": 0.28,
"learning_rate": 0.000361765295887663,
"loss": 2.7203,
"theoretical_loss": 3.672232805728499,
"tokens_seen": 936378368
},
{
"epoch": 0.28,
"learning_rate": 0.0003617552657973922,
"loss": 2.6865,
"theoretical_loss": 3.6722080494445244,
"tokens_seen": 936443904
},
{
"epoch": 0.28,
"learning_rate": 0.00036174523570712137,
"loss": 2.6965,
"theoretical_loss": 3.672183295378103,
"tokens_seen": 936509440
},
{
"epoch": 0.28,
"learning_rate": 0.0003617352056168506,
"loss": 2.9529,
"theoretical_loss": 3.672158543528882,
"tokens_seen": 936574976
},
{
"epoch": 0.28,
"learning_rate": 0.00036172517552657973,
"loss": 2.7092,
"theoretical_loss": 3.6721337938965064,
"tokens_seen": 936640512
},
{
"epoch": 0.28,
"learning_rate": 0.00036171514543630897,
"loss": 2.7811,
"theoretical_loss": 3.672109046480624,
"tokens_seen": 936706048
},
{
"epoch": 0.28,
"learning_rate": 0.0003617051153460381,
"loss": 2.8966,
"theoretical_loss": 3.67208430128088,
"tokens_seen": 936771584
},
{
"epoch": 0.28,
"learning_rate": 0.00036169508525576733,
"loss": 2.5755,
"theoretical_loss": 3.672059558296922,
"tokens_seen": 936837120
},
{
"epoch": 0.28,
"learning_rate": 0.0003616850551654965,
"loss": 2.5704,
"theoretical_loss": 3.672034817528396,
"tokens_seen": 936902656
},
{
"epoch": 0.28,
"learning_rate": 0.0003616750250752257,
"loss": 2.7997,
"theoretical_loss": 3.6720100789749486,
"tokens_seen": 936968192
},
{
"epoch": 0.28,
"learning_rate": 0.00036166499498495487,
"loss": 2.8214,
"theoretical_loss": 3.6719853426362263,
"tokens_seen": 937033728
},
{
"epoch": 0.28,
"learning_rate": 0.00036165496489468405,
"loss": 2.5532,
"theoretical_loss": 3.6719606085118772,
"tokens_seen": 937099264
},
{
"epoch": 0.28,
"objective/train/docs_used": 538123,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.850224256515503,
"objective/train/theoretical_loss": 3.671935876601547,
"objective/train/tokens_used": 957624800,
"theoretical_loss": 3.671935876601547,
"tokens_seen": 937164800
},
{
"epoch": 0.28,
"learning_rate": 0.00036164493480441323,
"loss": 2.8184,
"theoretical_loss": 3.671935876601547,
"tokens_seen": 937164800
},
{
"epoch": 0.28,
"learning_rate": 0.00036163490471414247,
"loss": 2.7056,
"theoretical_loss": 3.671911146904883,
"tokens_seen": 937230336
},
{
"epoch": 0.28,
"learning_rate": 0.0003616248746238716,
"loss": 2.8177,
"theoretical_loss": 3.671886419421533,
"tokens_seen": 937295872
},
{
"epoch": 0.28,
"learning_rate": 0.00036161484453360083,
"loss": 2.5827,
"theoretical_loss": 3.6718616941511426,
"tokens_seen": 937361408
},
{
"epoch": 0.28,
"learning_rate": 0.00036160481444333,
"loss": 2.8639,
"theoretical_loss": 3.6718369710933607,
"tokens_seen": 937426944
},
{
"epoch": 0.28,
"learning_rate": 0.0003615947843530592,
"loss": 2.6248,
"theoretical_loss": 3.671812250247834,
"tokens_seen": 937492480
},
{
"epoch": 0.28,
"learning_rate": 0.0003615847542627884,
"loss": 2.8862,
"theoretical_loss": 3.6717875316142097,
"tokens_seen": 937558016
},
{
"epoch": 0.28,
"learning_rate": 0.00036157472417251756,
"loss": 2.7416,
"theoretical_loss": 3.6717628151921353,
"tokens_seen": 937623552
},
{
"epoch": 0.28,
"learning_rate": 0.00036156469408224674,
"loss": 2.6448,
"theoretical_loss": 3.6717381009812593,
"tokens_seen": 937689088
},
{
"epoch": 0.28,
"learning_rate": 0.000361554663991976,
"loss": 2.5303,
"theoretical_loss": 3.671713388981228,
"tokens_seen": 937754624
},
{
"epoch": 0.28,
"learning_rate": 0.0003615446339017051,
"loss": 3.0531,
"theoretical_loss": 3.67168867919169,
"tokens_seen": 937820160
},
{
"epoch": 0.28,
"learning_rate": 0.00036153460381143434,
"loss": 2.6743,
"theoretical_loss": 3.671663971612293,
"tokens_seen": 937885696
},
{
"epoch": 0.28,
"learning_rate": 0.00036152457372116346,
"loss": 2.5429,
"theoretical_loss": 3.6716392662426847,
"tokens_seen": 937951232
},
{
"epoch": 0.28,
"learning_rate": 0.0003615145436308927,
"loss": 2.7924,
"theoretical_loss": 3.671614563082513,
"tokens_seen": 938016768
},
{
"epoch": 0.28,
"learning_rate": 0.0003615045135406219,
"loss": 2.9896,
"theoretical_loss": 3.6715898621314262,
"tokens_seen": 938082304
},
{
"epoch": 0.28,
"learning_rate": 0.00036149448345035106,
"loss": 2.952,
"theoretical_loss": 3.6715651633890722,
"tokens_seen": 938147840
},
{
"epoch": 0.28,
"learning_rate": 0.00036148445336008024,
"loss": 3.0603,
"theoretical_loss": 3.6715404668550997,
"tokens_seen": 938213376
},
{
"epoch": 0.28,
"learning_rate": 0.0003614744232698094,
"loss": 2.8517,
"theoretical_loss": 3.671515772529157,
"tokens_seen": 938278912
},
{
"epoch": 0.28,
"learning_rate": 0.0003614643931795386,
"loss": 2.7601,
"theoretical_loss": 3.671491080410892,
"tokens_seen": 938344448
},
{
"epoch": 0.28,
"learning_rate": 0.00036145436308926784,
"loss": 2.8342,
"theoretical_loss": 3.671466390499953,
"tokens_seen": 938409984
},
{
"epoch": 0.28,
"learning_rate": 0.00036144433299899697,
"loss": 2.9919,
"theoretical_loss": 3.6714417027959887,
"tokens_seen": 938475520
},
{
"epoch": 0.28,
"learning_rate": 0.0003614343029087262,
"loss": 2.6369,
"theoretical_loss": 3.6714170172986487,
"tokens_seen": 938541056
},
{
"epoch": 0.28,
"learning_rate": 0.0003614242728184554,
"loss": 2.8895,
"theoretical_loss": 3.6713923340075802,
"tokens_seen": 938606592
},
{
"epoch": 0.28,
"learning_rate": 0.00036141424272818456,
"loss": 2.8678,
"theoretical_loss": 3.6713676529224335,
"tokens_seen": 938672128
},
{
"epoch": 0.28,
"learning_rate": 0.00036140421263791374,
"loss": 2.7087,
"theoretical_loss": 3.671342974042856,
"tokens_seen": 938737664
},
{
"epoch": 0.28,
"objective/train/docs_used": 539283,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.31764817237854,
"objective/train/theoretical_loss": 3.6713182973684972,
"objective/train/tokens_used": 959263200,
"theoretical_loss": 3.6713182973684972,
"tokens_seen": 938803200
},
{
"epoch": 0.28,
"learning_rate": 0.0003613941825476429,
"loss": 2.6159,
"theoretical_loss": 3.6713182973684972,
"tokens_seen": 938803200
},
{
"epoch": 0.28,
"learning_rate": 0.0003613841524573721,
"loss": 2.7508,
"theoretical_loss": 3.6712936228990065,
"tokens_seen": 938868736
},
{
"epoch": 0.28,
"learning_rate": 0.00036137412236710134,
"loss": 2.5819,
"theoretical_loss": 3.6712689506340332,
"tokens_seen": 938934272
},
{
"epoch": 0.28,
"learning_rate": 0.00036136409227683047,
"loss": 2.8337,
"theoretical_loss": 3.6712442805732257,
"tokens_seen": 938999808
},
{
"epoch": 0.28,
"learning_rate": 0.0003613540621865597,
"loss": 2.8358,
"theoretical_loss": 3.671219612716234,
"tokens_seen": 939065344
},
{
"epoch": 0.28,
"learning_rate": 0.00036134403209628883,
"loss": 3.0175,
"theoretical_loss": 3.6711949470627063,
"tokens_seen": 939130880
},
{
"epoch": 0.28,
"learning_rate": 0.00036133400200601807,
"loss": 2.7812,
"theoretical_loss": 3.6711702836122937,
"tokens_seen": 939196416
},
{
"epoch": 0.28,
"learning_rate": 0.00036132397191574725,
"loss": 2.9098,
"theoretical_loss": 3.671145622364644,
"tokens_seen": 939261952
},
{
"epoch": 0.28,
"learning_rate": 0.00036131394182547643,
"loss": 2.6346,
"theoretical_loss": 3.671120963319408,
"tokens_seen": 939327488
},
{
"epoch": 0.28,
"learning_rate": 0.0003613039117352056,
"loss": 2.7648,
"theoretical_loss": 3.671096306476235,
"tokens_seen": 939393024
},
{
"epoch": 0.28,
"learning_rate": 0.00036129388164493484,
"loss": 2.7961,
"theoretical_loss": 3.6710716518347746,
"tokens_seen": 939458560
},
{
"epoch": 0.28,
"learning_rate": 0.00036128385155466397,
"loss": 2.8081,
"theoretical_loss": 3.671046999394677,
"tokens_seen": 939524096
},
{
"epoch": 0.28,
"learning_rate": 0.0003612738214643932,
"loss": 2.8393,
"theoretical_loss": 3.6710223491555913,
"tokens_seen": 939589632
},
{
"epoch": 0.28,
"learning_rate": 0.00036126379137412233,
"loss": 2.8474,
"theoretical_loss": 3.6709977011171686,
"tokens_seen": 939655168
},
{
"epoch": 0.28,
"learning_rate": 0.00036125376128385157,
"loss": 2.9569,
"theoretical_loss": 3.6709730552790583,
"tokens_seen": 939720704
},
{
"epoch": 0.28,
"learning_rate": 0.00036124373119358075,
"loss": 2.8802,
"theoretical_loss": 3.670948411640911,
"tokens_seen": 939786240
},
{
"epoch": 0.28,
"learning_rate": 0.00036123370110330993,
"loss": 2.6987,
"theoretical_loss": 3.670923770202376,
"tokens_seen": 939851776
},
{
"epoch": 0.28,
"learning_rate": 0.0003612236710130391,
"loss": 2.71,
"theoretical_loss": 3.6708991309631043,
"tokens_seen": 939917312
},
{
"epoch": 0.28,
"learning_rate": 0.0003612136409227683,
"loss": 3.0099,
"theoretical_loss": 3.6708744939227462,
"tokens_seen": 939982848
},
{
"epoch": 0.28,
"learning_rate": 0.0003612036108324975,
"loss": 2.6308,
"theoretical_loss": 3.6708498590809526,
"tokens_seen": 940048384
},
{
"epoch": 0.28,
"learning_rate": 0.0003611935807422267,
"loss": 3.0827,
"theoretical_loss": 3.6708252264373735,
"tokens_seen": 940113920
},
{
"epoch": 0.28,
"learning_rate": 0.00036118355065195584,
"loss": 2.9748,
"theoretical_loss": 3.6708005959916594,
"tokens_seen": 940179456
},
{
"epoch": 0.28,
"learning_rate": 0.00036117352056168507,
"loss": 3.0523,
"theoretical_loss": 3.6707759677434613,
"tokens_seen": 940244992
},
{
"epoch": 0.28,
"learning_rate": 0.0003611634904714142,
"loss": 2.8766,
"theoretical_loss": 3.67075134169243,
"tokens_seen": 940310528
},
{
"epoch": 0.28,
"learning_rate": 0.00036115346038114343,
"loss": 2.8059,
"theoretical_loss": 3.6707267178382166,
"tokens_seen": 940376064
},
{
"epoch": 0.28,
"objective/train/docs_used": 539932,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.662252187728882,
"objective/train/theoretical_loss": 3.6707020961804715,
"objective/train/tokens_used": 960901600,
"theoretical_loss": 3.6707020961804715,
"tokens_seen": 940441600
},
{
"epoch": 0.28,
"learning_rate": 0.0003611434302908726,
"loss": 2.7905,
"theoretical_loss": 3.6707020961804715,
"tokens_seen": 940441600
},
{
"epoch": 0.29,
"learning_rate": 0.0003611334002006018,
"loss": 2.607,
"theoretical_loss": 3.670677476718846,
"tokens_seen": 940507136
},
{
"epoch": 0.29,
"learning_rate": 0.000361123370110331,
"loss": 2.5543,
"theoretical_loss": 3.6706528594529915,
"tokens_seen": 940572672
},
{
"epoch": 0.29,
"learning_rate": 0.0003611133400200602,
"loss": 2.68,
"theoretical_loss": 3.6706282443825584,
"tokens_seen": 940638208
},
{
"epoch": 0.29,
"learning_rate": 0.0003611033099297894,
"loss": 3.0164,
"theoretical_loss": 3.6706036315071993,
"tokens_seen": 940703744
},
{
"epoch": 0.29,
"learning_rate": 0.0003610932798395186,
"loss": 2.6452,
"theoretical_loss": 3.670579020826564,
"tokens_seen": 940769280
},
{
"epoch": 0.29,
"learning_rate": 0.00036108324974924776,
"loss": 2.8592,
"theoretical_loss": 3.670554412340305,
"tokens_seen": 940834816
},
{
"epoch": 0.29,
"learning_rate": 0.00036107321965897694,
"loss": 2.6365,
"theoretical_loss": 3.670529806048073,
"tokens_seen": 940900352
},
{
"epoch": 0.29,
"learning_rate": 0.0003610631895687062,
"loss": 2.7318,
"theoretical_loss": 3.6705052019495206,
"tokens_seen": 940965888
},
{
"epoch": 0.29,
"learning_rate": 0.0003610531594784353,
"loss": 2.6669,
"theoretical_loss": 3.6704806000442987,
"tokens_seen": 941031424
},
{
"epoch": 0.29,
"learning_rate": 0.00036104312938816454,
"loss": 2.8778,
"theoretical_loss": 3.6704560003320594,
"tokens_seen": 941096960
},
{
"epoch": 0.29,
"learning_rate": 0.00036103309929789366,
"loss": 2.7689,
"theoretical_loss": 3.6704314028124543,
"tokens_seen": 941162496
},
{
"epoch": 0.29,
"learning_rate": 0.0003610230692076229,
"loss": 2.5239,
"theoretical_loss": 3.6704068074851355,
"tokens_seen": 941228032
},
{
"epoch": 0.29,
"learning_rate": 0.0003610130391173521,
"loss": 3.0063,
"theoretical_loss": 3.670382214349755,
"tokens_seen": 941293568
},
{
"epoch": 0.29,
"learning_rate": 0.00036100300902708126,
"loss": 2.7359,
"theoretical_loss": 3.6703576234059643,
"tokens_seen": 941359104
},
{
"epoch": 0.29,
"learning_rate": 0.00036099297893681044,
"loss": 2.5217,
"theoretical_loss": 3.6703330346534155,
"tokens_seen": 941424640
},
{
"epoch": 0.29,
"learning_rate": 0.0003609829488465396,
"loss": 2.7516,
"theoretical_loss": 3.6703084480917623,
"tokens_seen": 941490176
},
{
"epoch": 0.29,
"learning_rate": 0.0003609729187562688,
"loss": 2.6525,
"theoretical_loss": 3.670283863720656,
"tokens_seen": 941555712
},
{
"epoch": 0.29,
"learning_rate": 0.00036096288866599804,
"loss": 3.0291,
"theoretical_loss": 3.6702592815397486,
"tokens_seen": 941621248
},
{
"epoch": 0.29,
"learning_rate": 0.00036095285857572717,
"loss": 2.8075,
"theoretical_loss": 3.6702347015486927,
"tokens_seen": 941686784
},
{
"epoch": 0.29,
"learning_rate": 0.0003609428284854564,
"loss": 2.8183,
"theoretical_loss": 3.670210123747142,
"tokens_seen": 941752320
},
{
"epoch": 0.29,
"learning_rate": 0.0003609327983951856,
"loss": 2.8686,
"theoretical_loss": 3.6701855481347474,
"tokens_seen": 941817856
},
{
"epoch": 0.29,
"learning_rate": 0.00036092276830491476,
"loss": 2.6911,
"theoretical_loss": 3.6701609747111625,
"tokens_seen": 941883392
},
{
"epoch": 0.29,
"learning_rate": 0.00036091273821464394,
"loss": 2.6227,
"theoretical_loss": 3.6701364034760404,
"tokens_seen": 941948928
},
{
"epoch": 0.29,
"learning_rate": 0.0003609027081243731,
"loss": 2.6772,
"theoretical_loss": 3.6701118344290333,
"tokens_seen": 942014464
},
{
"epoch": 0.29,
"objective/train/docs_used": 541311,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6685099601745605,
"objective/train/theoretical_loss": 3.6700872675697944,
"objective/train/tokens_used": 962540000,
"theoretical_loss": 3.6700872675697944,
"tokens_seen": 942080000
},
{
"epoch": 0.29,
"learning_rate": 0.0003608926780341023,
"loss": 2.6392,
"theoretical_loss": 3.6700872675697944,
"tokens_seen": 942080000
},
{
"epoch": 0.29,
"learning_rate": 0.00036088264794383154,
"loss": 2.8054,
"theoretical_loss": 3.670062702897977,
"tokens_seen": 942145536
},
{
"epoch": 0.29,
"learning_rate": 0.00036087261785356067,
"loss": 2.8148,
"theoretical_loss": 3.670038140413234,
"tokens_seen": 942211072
},
{
"epoch": 0.29,
"learning_rate": 0.0003608625877632899,
"loss": 2.6291,
"theoretical_loss": 3.6700135801152176,
"tokens_seen": 942276608
},
{
"epoch": 0.29,
"learning_rate": 0.00036085255767301903,
"loss": 2.7428,
"theoretical_loss": 3.6699890220035822,
"tokens_seen": 942342144
},
{
"epoch": 0.29,
"learning_rate": 0.00036084252758274827,
"loss": 2.9915,
"theoretical_loss": 3.6699644660779818,
"tokens_seen": 942407680
},
{
"epoch": 0.29,
"learning_rate": 0.00036083249749247745,
"loss": 2.7635,
"theoretical_loss": 3.669939912338068,
"tokens_seen": 942473216
},
{
"epoch": 0.29,
"learning_rate": 0.00036082246740220663,
"loss": 2.5771,
"theoretical_loss": 3.669915360783495,
"tokens_seen": 942538752
},
{
"epoch": 0.29,
"learning_rate": 0.0003608124373119358,
"loss": 2.6393,
"theoretical_loss": 3.6698908114139175,
"tokens_seen": 942604288
},
{
"epoch": 0.29,
"learning_rate": 0.00036080240722166505,
"loss": 2.6948,
"theoretical_loss": 3.6698662642289874,
"tokens_seen": 942669824
},
{
"epoch": 0.29,
"learning_rate": 0.00036079237713139417,
"loss": 2.6442,
"theoretical_loss": 3.6698417192283594,
"tokens_seen": 942735360
},
{
"epoch": 0.29,
"learning_rate": 0.0003607823470411234,
"loss": 2.7101,
"theoretical_loss": 3.669817176411687,
"tokens_seen": 942800896
},
{
"epoch": 0.29,
"learning_rate": 0.00036077231695085253,
"loss": 2.7209,
"theoretical_loss": 3.6697926357786246,
"tokens_seen": 942866432
},
{
"epoch": 0.29,
"learning_rate": 0.00036076228686058177,
"loss": 2.9209,
"theoretical_loss": 3.6697680973288254,
"tokens_seen": 942931968
},
{
"epoch": 0.29,
"learning_rate": 0.00036075225677031095,
"loss": 2.5517,
"theoretical_loss": 3.669743561061944,
"tokens_seen": 942997504
},
{
"epoch": 0.29,
"learning_rate": 0.00036074222668004013,
"loss": 2.9826,
"theoretical_loss": 3.6697190269776345,
"tokens_seen": 943063040
},
{
"epoch": 0.29,
"learning_rate": 0.0003607321965897693,
"loss": 2.6366,
"theoretical_loss": 3.669694495075551,
"tokens_seen": 943128576
},
{
"epoch": 0.29,
"learning_rate": 0.0003607221664994985,
"loss": 3.046,
"theoretical_loss": 3.6696699653553475,
"tokens_seen": 943194112
},
{
"epoch": 0.29,
"learning_rate": 0.0003607121364092277,
"loss": 2.615,
"theoretical_loss": 3.669645437816678,
"tokens_seen": 943259648
},
{
"epoch": 0.29,
"learning_rate": 0.0003607021063189569,
"loss": 2.812,
"theoretical_loss": 3.6696209124591985,
"tokens_seen": 943325184
},
{
"epoch": 0.29,
"learning_rate": 0.00036069207622868604,
"loss": 2.8309,
"theoretical_loss": 3.669596389282562,
"tokens_seen": 943390720
},
{
"epoch": 0.29,
"learning_rate": 0.00036068204613841527,
"loss": 2.6949,
"theoretical_loss": 3.6695718682864236,
"tokens_seen": 943456256
},
{
"epoch": 0.29,
"learning_rate": 0.0003606720160481444,
"loss": 2.7781,
"theoretical_loss": 3.6695473494704385,
"tokens_seen": 943521792
},
{
"epoch": 0.29,
"learning_rate": 0.00036066198595787364,
"loss": 2.9506,
"theoretical_loss": 3.669522832834261,
"tokens_seen": 943587328
},
{
"epoch": 0.29,
"learning_rate": 0.0003606519558676028,
"loss": 2.5744,
"theoretical_loss": 3.6694983183775447,
"tokens_seen": 943652864
},
{
"epoch": 0.29,
"objective/train/docs_used": 542104,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.430126905441284,
"objective/train/theoretical_loss": 3.6694738060999468,
"objective/train/tokens_used": 964178400,
"theoretical_loss": 3.6694738060999468,
"tokens_seen": 943718400
},
{
"epoch": 0.29,
"learning_rate": 0.000360641925777332,
"loss": 2.6168,
"theoretical_loss": 3.6694738060999468,
"tokens_seen": 943718400
},
{
"epoch": 0.29,
"learning_rate": 0.0003606318956870612,
"loss": 2.7455,
"theoretical_loss": 3.6694492960011207,
"tokens_seen": 943783936
},
{
"epoch": 0.29,
"learning_rate": 0.0003606218655967904,
"loss": 2.7966,
"theoretical_loss": 3.6694247880807223,
"tokens_seen": 943849472
},
{
"epoch": 0.29,
"learning_rate": 0.00036061183550651954,
"loss": 2.5594,
"theoretical_loss": 3.6694002823384064,
"tokens_seen": 943915008
},
{
"epoch": 0.29,
"learning_rate": 0.0003606018054162488,
"loss": 2.9438,
"theoretical_loss": 3.669375778773828,
"tokens_seen": 943980544
},
{
"epoch": 0.29,
"learning_rate": 0.0003605917753259779,
"loss": 2.8496,
"theoretical_loss": 3.6693512773866424,
"tokens_seen": 944046080
},
{
"epoch": 0.29,
"learning_rate": 0.00036058174523570714,
"loss": 2.7438,
"theoretical_loss": 3.6693267781765058,
"tokens_seen": 944111616
},
{
"epoch": 0.29,
"learning_rate": 0.0003605717151454363,
"loss": 2.9613,
"theoretical_loss": 3.6693022811430724,
"tokens_seen": 944177152
},
{
"epoch": 0.29,
"learning_rate": 0.0003605616850551655,
"loss": 2.4904,
"theoretical_loss": 3.669277786285999,
"tokens_seen": 944242688
},
{
"epoch": 0.29,
"learning_rate": 0.0003605516549648947,
"loss": 2.7179,
"theoretical_loss": 3.6692532936049407,
"tokens_seen": 944308224
},
{
"epoch": 0.29,
"learning_rate": 0.00036054162487462386,
"loss": 2.7821,
"theoretical_loss": 3.6692288030995526,
"tokens_seen": 944373760
},
{
"epoch": 0.29,
"learning_rate": 0.00036053159478435304,
"loss": 2.7714,
"theoretical_loss": 3.669204314769491,
"tokens_seen": 944439296
},
{
"epoch": 0.29,
"learning_rate": 0.0003605215646940823,
"loss": 2.6083,
"theoretical_loss": 3.6691798286144124,
"tokens_seen": 944504832
},
{
"epoch": 0.29,
"learning_rate": 0.0003605115346038114,
"loss": 2.5801,
"theoretical_loss": 3.669155344633971,
"tokens_seen": 944570368
},
{
"epoch": 0.29,
"learning_rate": 0.00036050150451354064,
"loss": 2.6406,
"theoretical_loss": 3.6691308628278247,
"tokens_seen": 944635904
},
{
"epoch": 0.29,
"learning_rate": 0.00036049147442326977,
"loss": 3.143,
"theoretical_loss": 3.6691063831956283,
"tokens_seen": 944701440
},
{
"epoch": 0.29,
"learning_rate": 0.000360481444332999,
"loss": 2.8056,
"theoretical_loss": 3.669081905737039,
"tokens_seen": 944766976
},
{
"epoch": 0.29,
"learning_rate": 0.0003604714142427282,
"loss": 2.7397,
"theoretical_loss": 3.669057430451712,
"tokens_seen": 944832512
},
{
"epoch": 0.29,
"learning_rate": 0.00036046138415245737,
"loss": 2.8207,
"theoretical_loss": 3.6690329573393043,
"tokens_seen": 944898048
},
{
"epoch": 0.29,
"learning_rate": 0.00036045135406218655,
"loss": 2.6611,
"theoretical_loss": 3.669008486399472,
"tokens_seen": 944963584
},
{
"epoch": 0.29,
"learning_rate": 0.0003604413239719158,
"loss": 2.6967,
"theoretical_loss": 3.668984017631872,
"tokens_seen": 945029120
},
{
"epoch": 0.29,
"learning_rate": 0.0003604312938816449,
"loss": 2.6119,
"theoretical_loss": 3.66895955103616,
"tokens_seen": 945094656
},
{
"epoch": 0.29,
"learning_rate": 0.00036042126379137414,
"loss": 2.6816,
"theoretical_loss": 3.6689350866119934,
"tokens_seen": 945160192
},
{
"epoch": 0.29,
"learning_rate": 0.00036041123370110327,
"loss": 2.8118,
"theoretical_loss": 3.668910624359029,
"tokens_seen": 945225728
},
{
"epoch": 0.29,
"learning_rate": 0.0003604012036108325,
"loss": 2.8631,
"theoretical_loss": 3.668886164276923,
"tokens_seen": 945291264
},
{
"epoch": 0.29,
"objective/train/docs_used": 543471,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2914235591888428,
"objective/train/theoretical_loss": 3.668861706365332,
"objective/train/tokens_used": 965816800,
"theoretical_loss": 3.668861706365332,
"tokens_seen": 945356800
},
{
"epoch": 0.29,
"learning_rate": 0.0003603911735205617,
"loss": 2.805,
"theoretical_loss": 3.668861706365332,
"tokens_seen": 945356800
},
{
"epoch": 0.29,
"learning_rate": 0.00036038114343029087,
"loss": 2.843,
"theoretical_loss": 3.668837250623914,
"tokens_seen": 945422336
},
{
"epoch": 0.29,
"learning_rate": 0.00036037111334002005,
"loss": 2.679,
"theoretical_loss": 3.6688127970523254,
"tokens_seen": 945487872
},
{
"epoch": 0.29,
"learning_rate": 0.00036036108324974923,
"loss": 2.7933,
"theoretical_loss": 3.6687883456502233,
"tokens_seen": 945553408
},
{
"epoch": 0.29,
"learning_rate": 0.00036035105315947847,
"loss": 2.9906,
"theoretical_loss": 3.668763896417265,
"tokens_seen": 945618944
},
{
"epoch": 0.29,
"learning_rate": 0.00036034102306920765,
"loss": 2.6576,
"theoretical_loss": 3.668739449353108,
"tokens_seen": 945684480
},
{
"epoch": 0.29,
"learning_rate": 0.00036033099297893683,
"loss": 2.8351,
"theoretical_loss": 3.6687150044574093,
"tokens_seen": 945750016
},
{
"epoch": 0.29,
"learning_rate": 0.000360320962888666,
"loss": 2.7163,
"theoretical_loss": 3.6686905617298264,
"tokens_seen": 945815552
},
{
"epoch": 0.29,
"learning_rate": 0.00036031093279839525,
"loss": 2.8029,
"theoretical_loss": 3.6686661211700167,
"tokens_seen": 945881088
},
{
"epoch": 0.29,
"learning_rate": 0.00036030090270812437,
"loss": 2.7711,
"theoretical_loss": 3.668641682777638,
"tokens_seen": 945946624
},
{
"epoch": 0.29,
"learning_rate": 0.0003602908726178536,
"loss": 2.7675,
"theoretical_loss": 3.6686172465523477,
"tokens_seen": 946012160
},
{
"epoch": 0.29,
"learning_rate": 0.00036028084252758273,
"loss": 2.8669,
"theoretical_loss": 3.668592812493803,
"tokens_seen": 946077696
},
{
"epoch": 0.29,
"learning_rate": 0.00036027081243731197,
"loss": 2.8675,
"theoretical_loss": 3.668568380601663,
"tokens_seen": 946143232
},
{
"epoch": 0.29,
"learning_rate": 0.00036026078234704115,
"loss": 2.7275,
"theoretical_loss": 3.668543950875585,
"tokens_seen": 946208768
},
{
"epoch": 0.29,
"learning_rate": 0.00036025075225677033,
"loss": 2.6035,
"theoretical_loss": 3.6685195233152266,
"tokens_seen": 946274304
},
{
"epoch": 0.29,
"learning_rate": 0.0003602407221664995,
"loss": 2.6121,
"theoretical_loss": 3.668495097920246,
"tokens_seen": 946339840
},
{
"epoch": 0.29,
"learning_rate": 0.0003602306920762287,
"loss": 2.5615,
"theoretical_loss": 3.668470674690302,
"tokens_seen": 946405376
},
{
"epoch": 0.29,
"learning_rate": 0.0003602206619859579,
"loss": 2.6635,
"theoretical_loss": 3.668446253625051,
"tokens_seen": 946470912
},
{
"epoch": 0.29,
"learning_rate": 0.0003602106318956871,
"loss": 2.7366,
"theoretical_loss": 3.6684218347241533,
"tokens_seen": 946536448
},
{
"epoch": 0.29,
"learning_rate": 0.00036020060180541624,
"loss": 2.716,
"theoretical_loss": 3.6683974179872667,
"tokens_seen": 946601984
},
{
"epoch": 0.29,
"learning_rate": 0.0003601905717151455,
"loss": 2.6551,
"theoretical_loss": 3.6683730034140485,
"tokens_seen": 946667520
},
{
"epoch": 0.29,
"learning_rate": 0.0003601805416248746,
"loss": 2.7032,
"theoretical_loss": 3.668348591004159,
"tokens_seen": 946733056
},
{
"epoch": 0.29,
"learning_rate": 0.00036017051153460384,
"loss": 2.6094,
"theoretical_loss": 3.668324180757255,
"tokens_seen": 946798592
},
{
"epoch": 0.29,
"learning_rate": 0.000360160481444333,
"loss": 2.6728,
"theoretical_loss": 3.6682997726729965,
"tokens_seen": 946864128
},
{
"epoch": 0.29,
"learning_rate": 0.0003601504513540622,
"loss": 2.6664,
"theoretical_loss": 3.6682753667510415,
"tokens_seen": 946929664
},
{
"epoch": 0.29,
"objective/train/docs_used": 544140,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.884384870529175,
"objective/train/theoretical_loss": 3.668250962991049,
"objective/train/tokens_used": 967455200,
"theoretical_loss": 3.668250962991049,
"tokens_seen": 946995200
},
{
"epoch": 0.29,
"learning_rate": 0.0003601404212637914,
"loss": 2.7751,
"theoretical_loss": 3.668250962991049,
"tokens_seen": 946995200
},
{
"epoch": 0.29,
"learning_rate": 0.0003601303911735206,
"loss": 2.7952,
"theoretical_loss": 3.668226561392678,
"tokens_seen": 947060736
},
{
"epoch": 0.29,
"learning_rate": 0.00036012036108324974,
"loss": 2.8799,
"theoretical_loss": 3.668202161955587,
"tokens_seen": 947126272
},
{
"epoch": 0.29,
"learning_rate": 0.000360110330992979,
"loss": 2.8656,
"theoretical_loss": 3.6681777646794362,
"tokens_seen": 947191808
},
{
"epoch": 0.29,
"learning_rate": 0.0003601003009027081,
"loss": 2.7117,
"theoretical_loss": 3.668153369563883,
"tokens_seen": 947257344
},
{
"epoch": 0.29,
"learning_rate": 0.00036009027081243734,
"loss": 2.8989,
"theoretical_loss": 3.668128976608588,
"tokens_seen": 947322880
},
{
"epoch": 0.29,
"learning_rate": 0.0003600802407221665,
"loss": 2.7868,
"theoretical_loss": 3.6681045858132104,
"tokens_seen": 947388416
},
{
"epoch": 0.29,
"learning_rate": 0.0003600702106318957,
"loss": 2.805,
"theoretical_loss": 3.6680801971774084,
"tokens_seen": 947453952
},
{
"epoch": 0.29,
"learning_rate": 0.0003600601805416249,
"loss": 2.607,
"theoretical_loss": 3.668055810700843,
"tokens_seen": 947519488
},
{
"epoch": 0.29,
"learning_rate": 0.00036005015045135406,
"loss": 2.7592,
"theoretical_loss": 3.6680314263831724,
"tokens_seen": 947585024
},
{
"epoch": 0.29,
"learning_rate": 0.00036004012036108324,
"loss": 2.5931,
"theoretical_loss": 3.6680070442240567,
"tokens_seen": 947650560
},
{
"epoch": 0.29,
"learning_rate": 0.0003600300902708125,
"loss": 2.5053,
"theoretical_loss": 3.667982664223156,
"tokens_seen": 947716096
},
{
"epoch": 0.29,
"learning_rate": 0.0003600200601805416,
"loss": 2.8014,
"theoretical_loss": 3.667958286380129,
"tokens_seen": 947781632
},
{
"epoch": 0.29,
"learning_rate": 0.00036001003009027084,
"loss": 2.6762,
"theoretical_loss": 3.6679339106946363,
"tokens_seen": 947847168
},
{
"epoch": 0.29,
"learning_rate": 0.00035999999999999997,
"loss": 2.614,
"theoretical_loss": 3.667909537166338,
"tokens_seen": 947912704
},
{
"epoch": 0.29,
"learning_rate": 0.0003599899699097292,
"loss": 2.6162,
"theoretical_loss": 3.667885165794893,
"tokens_seen": 947978240
},
{
"epoch": 0.29,
"learning_rate": 0.0003599799398194584,
"loss": 2.6903,
"theoretical_loss": 3.667860796579962,
"tokens_seen": 948043776
},
{
"epoch": 0.29,
"learning_rate": 0.00035996990972918757,
"loss": 3.1263,
"theoretical_loss": 3.6678364295212056,
"tokens_seen": 948109312
},
{
"epoch": 0.29,
"learning_rate": 0.00035995987963891675,
"loss": 2.9297,
"theoretical_loss": 3.6678120646182837,
"tokens_seen": 948174848
},
{
"epoch": 0.29,
"learning_rate": 0.000359949849548646,
"loss": 2.7986,
"theoretical_loss": 3.6677877018708562,
"tokens_seen": 948240384
},
{
"epoch": 0.29,
"learning_rate": 0.0003599398194583751,
"loss": 2.7316,
"theoretical_loss": 3.6677633412785835,
"tokens_seen": 948305920
},
{
"epoch": 0.29,
"learning_rate": 0.00035992978936810434,
"loss": 2.8413,
"theoretical_loss": 3.667738982841126,
"tokens_seen": 948371456
},
{
"epoch": 0.29,
"learning_rate": 0.00035991975927783347,
"loss": 2.7668,
"theoretical_loss": 3.667714626558145,
"tokens_seen": 948436992
},
{
"epoch": 0.29,
"learning_rate": 0.0003599097291875627,
"loss": 2.7968,
"theoretical_loss": 3.6676902724293003,
"tokens_seen": 948502528
},
{
"epoch": 0.29,
"learning_rate": 0.0003598996990972919,
"loss": 2.7865,
"theoretical_loss": 3.6676659204542528,
"tokens_seen": 948568064
},
{
"epoch": 0.29,
"objective/train/docs_used": 544543,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.719106674194336,
"objective/train/theoretical_loss": 3.6676415706326635,
"objective/train/tokens_used": 969093600,
"theoretical_loss": 3.6676415706326635,
"tokens_seen": 948633600
},
{
"epoch": 0.29,
"learning_rate": 0.00035988966900702107,
"loss": 2.7794,
"theoretical_loss": 3.6676415706326635,
"tokens_seen": 948633600
},
{
"epoch": 0.29,
"learning_rate": 0.00035987963891675025,
"loss": 2.8178,
"theoretical_loss": 3.6676172229641923,
"tokens_seen": 948699136
},
{
"epoch": 0.29,
"learning_rate": 0.00035986960882647943,
"loss": 2.5496,
"theoretical_loss": 3.667592877448501,
"tokens_seen": 948764672
},
{
"epoch": 0.29,
"learning_rate": 0.0003598595787362086,
"loss": 2.7673,
"theoretical_loss": 3.66756853408525,
"tokens_seen": 948830208
},
{
"epoch": 0.29,
"learning_rate": 0.00035984954864593785,
"loss": 2.979,
"theoretical_loss": 3.667544192874101,
"tokens_seen": 948895744
},
{
"epoch": 0.29,
"learning_rate": 0.000359839518555667,
"loss": 2.658,
"theoretical_loss": 3.6675198538147153,
"tokens_seen": 948961280
},
{
"epoch": 0.29,
"learning_rate": 0.0003598294884653962,
"loss": 2.9247,
"theoretical_loss": 3.6674955169067527,
"tokens_seen": 949026816
},
{
"epoch": 0.29,
"learning_rate": 0.00035981945837512534,
"loss": 2.6721,
"theoretical_loss": 3.6674711821498756,
"tokens_seen": 949092352
},
{
"epoch": 0.29,
"learning_rate": 0.00035980942828485457,
"loss": 2.8078,
"theoretical_loss": 3.6674468495437456,
"tokens_seen": 949157888
},
{
"epoch": 0.29,
"learning_rate": 0.00035979939819458375,
"loss": 2.7741,
"theoretical_loss": 3.667422519088023,
"tokens_seen": 949223424
},
{
"epoch": 0.29,
"learning_rate": 0.00035978936810431293,
"loss": 2.5236,
"theoretical_loss": 3.66739819078237,
"tokens_seen": 949288960
},
{
"epoch": 0.29,
"learning_rate": 0.0003597793380140421,
"loss": 2.8453,
"theoretical_loss": 3.667373864626448,
"tokens_seen": 949354496
},
{
"epoch": 0.29,
"learning_rate": 0.00035976930792377135,
"loss": 2.8485,
"theoretical_loss": 3.6673495406199197,
"tokens_seen": 949420032
},
{
"epoch": 0.29,
"learning_rate": 0.0003597592778335005,
"loss": 2.963,
"theoretical_loss": 3.667325218762445,
"tokens_seen": 949485568
},
{
"epoch": 0.29,
"learning_rate": 0.0003597492477432297,
"loss": 2.8668,
"theoretical_loss": 3.667300899053687,
"tokens_seen": 949551104
},
{
"epoch": 0.29,
"learning_rate": 0.00035973921765295884,
"loss": 2.8555,
"theoretical_loss": 3.667276581493307,
"tokens_seen": 949616640
},
{
"epoch": 0.29,
"learning_rate": 0.0003597291875626881,
"loss": 2.8316,
"theoretical_loss": 3.6672522660809674,
"tokens_seen": 949682176
},
{
"epoch": 0.29,
"learning_rate": 0.00035971915747241726,
"loss": 2.9972,
"theoretical_loss": 3.6672279528163303,
"tokens_seen": 949747712
},
{
"epoch": 0.29,
"learning_rate": 0.00035970912738214644,
"loss": 2.8298,
"theoretical_loss": 3.667203641699057,
"tokens_seen": 949813248
},
{
"epoch": 0.29,
"learning_rate": 0.0003596990972918756,
"loss": 2.5397,
"theoretical_loss": 3.6671793327288107,
"tokens_seen": 949878784
},
{
"epoch": 0.29,
"learning_rate": 0.0003596890672016048,
"loss": 2.6681,
"theoretical_loss": 3.667155025905253,
"tokens_seen": 949944320
},
{
"epoch": 0.29,
"learning_rate": 0.000359679037111334,
"loss": 2.9004,
"theoretical_loss": 3.6671307212280473,
"tokens_seen": 950009856
},
{
"epoch": 0.29,
"learning_rate": 0.0003596690070210632,
"loss": 2.5837,
"theoretical_loss": 3.6671064186968545,
"tokens_seen": 950075392
},
{
"epoch": 0.29,
"learning_rate": 0.00035965897693079234,
"loss": 2.8557,
"theoretical_loss": 3.6670821183113373,
"tokens_seen": 950140928
},
{
"epoch": 0.29,
"learning_rate": 0.0003596489468405216,
"loss": 2.9066,
"theoretical_loss": 3.6670578200711597,
"tokens_seen": 950206464
},
{
"epoch": 0.29,
"objective/train/docs_used": 545942,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7416210174560547,
"objective/train/theoretical_loss": 3.667033523975983,
"objective/train/tokens_used": 970732000,
"theoretical_loss": 3.667033523975983,
"tokens_seen": 950272000
},
{
"epoch": 0.29,
"learning_rate": 0.00035963891675025076,
"loss": 2.848,
"theoretical_loss": 3.667033523975983,
"tokens_seen": 950272000
},
{
"epoch": 0.29,
"learning_rate": 0.00035962888665997994,
"loss": 2.8121,
"theoretical_loss": 3.667009230025471,
"tokens_seen": 950337536
},
{
"epoch": 0.29,
"learning_rate": 0.0003596188565697091,
"loss": 2.9414,
"theoretical_loss": 3.6669849382192847,
"tokens_seen": 950403072
},
{
"epoch": 0.29,
"learning_rate": 0.0003596088264794383,
"loss": 2.689,
"theoretical_loss": 3.666960648557089,
"tokens_seen": 950468608
},
{
"epoch": 0.29,
"learning_rate": 0.00035959879638916754,
"loss": 2.7776,
"theoretical_loss": 3.6669363610385464,
"tokens_seen": 950534144
},
{
"epoch": 0.29,
"learning_rate": 0.0003595887662988967,
"loss": 2.8579,
"theoretical_loss": 3.6669120756633187,
"tokens_seen": 950599680
},
{
"epoch": 0.29,
"learning_rate": 0.0003595787362086259,
"loss": 2.7987,
"theoretical_loss": 3.66688779243107,
"tokens_seen": 950665216
},
{
"epoch": 0.29,
"learning_rate": 0.0003595687061183551,
"loss": 2.7164,
"theoretical_loss": 3.6668635113414636,
"tokens_seen": 950730752
},
{
"epoch": 0.29,
"learning_rate": 0.00035955867602808426,
"loss": 2.9697,
"theoretical_loss": 3.666839232394163,
"tokens_seen": 950796288
},
{
"epoch": 0.29,
"learning_rate": 0.00035954864593781344,
"loss": 2.7071,
"theoretical_loss": 3.6668149555888307,
"tokens_seen": 950861824
},
{
"epoch": 0.29,
"learning_rate": 0.0003595386158475427,
"loss": 2.7276,
"theoretical_loss": 3.6667906809251303,
"tokens_seen": 950927360
},
{
"epoch": 0.29,
"learning_rate": 0.0003595285857572718,
"loss": 2.5539,
"theoretical_loss": 3.666766408402726,
"tokens_seen": 950992896
},
{
"epoch": 0.29,
"learning_rate": 0.00035951855566700104,
"loss": 2.6754,
"theoretical_loss": 3.666742138021281,
"tokens_seen": 951058432
},
{
"epoch": 0.29,
"learning_rate": 0.00035950852557673017,
"loss": 2.8099,
"theoretical_loss": 3.6667178697804577,
"tokens_seen": 951123968
},
{
"epoch": 0.29,
"learning_rate": 0.0003594984954864594,
"loss": 2.7383,
"theoretical_loss": 3.6666936036799216,
"tokens_seen": 951189504
},
{
"epoch": 0.29,
"learning_rate": 0.0003594884653961886,
"loss": 2.8711,
"theoretical_loss": 3.666669339719336,
"tokens_seen": 951255040
},
{
"epoch": 0.29,
"learning_rate": 0.00035947843530591777,
"loss": 2.9524,
"theoretical_loss": 3.666645077898364,
"tokens_seen": 951320576
},
{
"epoch": 0.29,
"learning_rate": 0.00035946840521564695,
"loss": 2.5983,
"theoretical_loss": 3.666620818216671,
"tokens_seen": 951386112
},
{
"epoch": 0.29,
"learning_rate": 0.0003594583751253762,
"loss": 2.9286,
"theoretical_loss": 3.66659656067392,
"tokens_seen": 951451648
},
{
"epoch": 0.29,
"learning_rate": 0.0003594483450351053,
"loss": 2.8591,
"theoretical_loss": 3.666572305269775,
"tokens_seen": 951517184
},
{
"epoch": 0.29,
"learning_rate": 0.00035943831494483454,
"loss": 2.6527,
"theoretical_loss": 3.6665480520039004,
"tokens_seen": 951582720
},
{
"epoch": 0.29,
"learning_rate": 0.00035942828485456367,
"loss": 3.0843,
"theoretical_loss": 3.666523800875961,
"tokens_seen": 951648256
},
{
"epoch": 0.29,
"learning_rate": 0.0003594182547642929,
"loss": 2.8544,
"theoretical_loss": 3.66649955188562,
"tokens_seen": 951713792
},
{
"epoch": 0.29,
"learning_rate": 0.0003594082246740221,
"loss": 2.9678,
"theoretical_loss": 3.6664753050325425,
"tokens_seen": 951779328
},
{
"epoch": 0.29,
"learning_rate": 0.00035939819458375127,
"loss": 2.6741,
"theoretical_loss": 3.666451060316393,
"tokens_seen": 951844864
},
{
"epoch": 0.29,
"objective/train/docs_used": 546639,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.579766035079956,
"objective/train/theoretical_loss": 3.6664268177368364,
"objective/train/tokens_used": 972370400,
"theoretical_loss": 3.6664268177368364,
"tokens_seen": 951910400
},
{
"epoch": 0.29,
"learning_rate": 0.00035938816449348045,
"loss": 2.8223,
"theoretical_loss": 3.6664268177368364,
"tokens_seen": 951910400
},
{
"epoch": 0.29,
"learning_rate": 0.00035937813440320963,
"loss": 2.6345,
"theoretical_loss": 3.666402577293537,
"tokens_seen": 951975936
},
{
"epoch": 0.29,
"learning_rate": 0.0003593681043129388,
"loss": 2.7627,
"theoretical_loss": 3.666378338986159,
"tokens_seen": 952041472
},
{
"epoch": 0.29,
"learning_rate": 0.00035935807422266805,
"loss": 2.7344,
"theoretical_loss": 3.666354102814368,
"tokens_seen": 952107008
},
{
"epoch": 0.29,
"learning_rate": 0.0003593480441323972,
"loss": 2.785,
"theoretical_loss": 3.6663298687778276,
"tokens_seen": 952172544
},
{
"epoch": 0.29,
"learning_rate": 0.0003593380140421264,
"loss": 2.6277,
"theoretical_loss": 3.6663056368762046,
"tokens_seen": 952238080
},
{
"epoch": 0.29,
"learning_rate": 0.00035932798395185554,
"loss": 2.6655,
"theoretical_loss": 3.6662814071091625,
"tokens_seen": 952303616
},
{
"epoch": 0.29,
"learning_rate": 0.00035931795386158477,
"loss": 2.7485,
"theoretical_loss": 3.666257179476367,
"tokens_seen": 952369152
},
{
"epoch": 0.29,
"learning_rate": 0.00035930792377131395,
"loss": 2.7019,
"theoretical_loss": 3.6662329539774827,
"tokens_seen": 952434688
},
{
"epoch": 0.29,
"learning_rate": 0.00035929789368104313,
"loss": 2.6907,
"theoretical_loss": 3.6662087306121762,
"tokens_seen": 952500224
},
{
"epoch": 0.29,
"learning_rate": 0.0003592878635907723,
"loss": 2.8721,
"theoretical_loss": 3.6661845093801118,
"tokens_seen": 952565760
},
{
"epoch": 0.29,
"learning_rate": 0.00035927783350050155,
"loss": 2.4393,
"theoretical_loss": 3.6661602902809545,
"tokens_seen": 952631296
},
{
"epoch": 0.29,
"learning_rate": 0.0003592678034102307,
"loss": 2.5003,
"theoretical_loss": 3.6661360733143704,
"tokens_seen": 952696832
},
{
"epoch": 0.29,
"learning_rate": 0.0003592577733199599,
"loss": 2.9256,
"theoretical_loss": 3.6661118584800247,
"tokens_seen": 952762368
},
{
"epoch": 0.29,
"learning_rate": 0.00035924774322968904,
"loss": 2.5951,
"theoretical_loss": 3.6660876457775835,
"tokens_seen": 952827904
},
{
"epoch": 0.29,
"learning_rate": 0.0003592377131394183,
"loss": 2.9128,
"theoretical_loss": 3.6660634352067123,
"tokens_seen": 952893440
},
{
"epoch": 0.29,
"learning_rate": 0.00035922768304914746,
"loss": 2.572,
"theoretical_loss": 3.6660392267670767,
"tokens_seen": 952958976
},
{
"epoch": 0.29,
"learning_rate": 0.00035921765295887664,
"loss": 2.6336,
"theoretical_loss": 3.6660150204583424,
"tokens_seen": 953024512
},
{
"epoch": 0.29,
"learning_rate": 0.0003592076228686058,
"loss": 2.5672,
"theoretical_loss": 3.6659908162801758,
"tokens_seen": 953090048
},
{
"epoch": 0.29,
"learning_rate": 0.000359197592778335,
"loss": 2.782,
"theoretical_loss": 3.665966614232243,
"tokens_seen": 953155584
},
{
"epoch": 0.29,
"learning_rate": 0.0003591875626880642,
"loss": 2.8939,
"theoretical_loss": 3.665942414314209,
"tokens_seen": 953221120
},
{
"epoch": 0.29,
"learning_rate": 0.0003591775325977934,
"loss": 2.7267,
"theoretical_loss": 3.665918216525741,
"tokens_seen": 953286656
},
{
"epoch": 0.29,
"learning_rate": 0.00035916750250752254,
"loss": 2.9992,
"theoretical_loss": 3.6658940208665047,
"tokens_seen": 953352192
},
{
"epoch": 0.29,
"learning_rate": 0.0003591574724172518,
"loss": 2.5708,
"theoretical_loss": 3.6658698273361665,
"tokens_seen": 953417728
},
{
"epoch": 0.29,
"learning_rate": 0.00035914744232698096,
"loss": 2.8824,
"theoretical_loss": 3.6658456359343927,
"tokens_seen": 953483264
},
{
"epoch": 0.29,
"objective/train/docs_used": 547886,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.0660064220428467,
"objective/train/theoretical_loss": 3.66582144666085,
"objective/train/tokens_used": 974008800,
"theoretical_loss": 3.66582144666085,
"tokens_seen": 953548800
},
{
"epoch": 0.29,
"learning_rate": 0.00035913741223671014,
"loss": 2.42,
"theoretical_loss": 3.66582144666085,
"tokens_seen": 953548800
},
{
"epoch": 0.29,
"learning_rate": 0.0003591273821464393,
"loss": 2.5187,
"theoretical_loss": 3.6657972595152053,
"tokens_seen": 953614336
},
{
"epoch": 0.29,
"learning_rate": 0.0003591173520561685,
"loss": 2.7639,
"theoretical_loss": 3.6657730744971238,
"tokens_seen": 953679872
},
{
"epoch": 0.29,
"learning_rate": 0.0003591073219658977,
"loss": 2.7651,
"theoretical_loss": 3.6657488916062735,
"tokens_seen": 953745408
},
{
"epoch": 0.29,
"learning_rate": 0.0003590972918756269,
"loss": 2.8471,
"theoretical_loss": 3.665724710842321,
"tokens_seen": 953810944
},
{
"epoch": 0.29,
"learning_rate": 0.00035908726178535605,
"loss": 2.6443,
"theoretical_loss": 3.665700532204932,
"tokens_seen": 953876480
},
{
"epoch": 0.29,
"learning_rate": 0.0003590772316950853,
"loss": 2.6903,
"theoretical_loss": 3.6656763556937744,
"tokens_seen": 953942016
},
{
"epoch": 0.29,
"learning_rate": 0.0003590672016048144,
"loss": 2.3794,
"theoretical_loss": 3.6656521813085154,
"tokens_seen": 954007552
},
{
"epoch": 0.29,
"learning_rate": 0.00035905717151454364,
"loss": 2.7081,
"theoretical_loss": 3.6656280090488216,
"tokens_seen": 954073088
},
{
"epoch": 0.29,
"learning_rate": 0.0003590471414242728,
"loss": 2.7642,
"theoretical_loss": 3.66560383891436,
"tokens_seen": 954138624
},
{
"epoch": 0.29,
"learning_rate": 0.000359037111334002,
"loss": 2.7121,
"theoretical_loss": 3.6655796709047976,
"tokens_seen": 954204160
},
{
"epoch": 0.29,
"learning_rate": 0.0003590270812437312,
"loss": 2.5816,
"theoretical_loss": 3.6655555050198023,
"tokens_seen": 954269696
},
{
"epoch": 0.29,
"learning_rate": 0.00035901705115346037,
"loss": 2.6968,
"theoretical_loss": 3.665531341259041,
"tokens_seen": 954335232
},
{
"epoch": 0.29,
"learning_rate": 0.00035900702106318955,
"loss": 2.735,
"theoretical_loss": 3.665507179622182,
"tokens_seen": 954400768
},
{
"epoch": 0.29,
"learning_rate": 0.0003589969909729188,
"loss": 2.7587,
"theoretical_loss": 3.6654830201088906,
"tokens_seen": 954466304
},
{
"epoch": 0.29,
"learning_rate": 0.0003589869608826479,
"loss": 2.771,
"theoretical_loss": 3.665458862718837,
"tokens_seen": 954531840
},
{
"epoch": 0.29,
"learning_rate": 0.00035897693079237715,
"loss": 2.8383,
"theoretical_loss": 3.6654347074516873,
"tokens_seen": 954597376
},
{
"epoch": 0.29,
"learning_rate": 0.00035896690070210633,
"loss": 2.735,
"theoretical_loss": 3.6654105543071096,
"tokens_seen": 954662912
},
{
"epoch": 0.29,
"learning_rate": 0.0003589568706118355,
"loss": 2.7604,
"theoretical_loss": 3.6653864032847716,
"tokens_seen": 954728448
},
{
"epoch": 0.29,
"learning_rate": 0.0003589468405215647,
"loss": 2.7873,
"theoretical_loss": 3.665362254384341,
"tokens_seen": 954793984
},
{
"epoch": 0.29,
"learning_rate": 0.00035893681043129387,
"loss": 2.8608,
"theoretical_loss": 3.6653381076054865,
"tokens_seen": 954859520
},
{
"epoch": 0.29,
"learning_rate": 0.00035892678034102305,
"loss": 2.7854,
"theoretical_loss": 3.6653139629478755,
"tokens_seen": 954925056
},
{
"epoch": 0.29,
"learning_rate": 0.0003589167502507523,
"loss": 2.5955,
"theoretical_loss": 3.6652898204111763,
"tokens_seen": 954990592
},
{
"epoch": 0.29,
"learning_rate": 0.0003589067201604814,
"loss": 2.6941,
"theoretical_loss": 3.6652656799950565,
"tokens_seen": 955056128
},
{
"epoch": 0.29,
"learning_rate": 0.00035889669007021065,
"loss": 2.9047,
"theoretical_loss": 3.665241541699185,
"tokens_seen": 955121664
},
{
"epoch": 0.29,
"objective/train/docs_used": 548540,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.787440299987793,
"objective/train/theoretical_loss": 3.6652174055232303,
"objective/train/tokens_used": 975647200,
"theoretical_loss": 3.6652174055232303,
"tokens_seen": 955187200
},
{
"epoch": 0.29,
"learning_rate": 0.0003588866599799398,
"loss": 2.8282,
"theoretical_loss": 3.6652174055232303,
"tokens_seen": 955187200
},
{
"epoch": 0.29,
"learning_rate": 0.000358876629889669,
"loss": 2.6453,
"theoretical_loss": 3.66519327146686,
"tokens_seen": 955252736
},
{
"epoch": 0.29,
"learning_rate": 0.0003588665997993982,
"loss": 2.9431,
"theoretical_loss": 3.665169139529743,
"tokens_seen": 955318272
},
{
"epoch": 0.29,
"learning_rate": 0.0003588565697091274,
"loss": 2.9877,
"theoretical_loss": 3.665145009711548,
"tokens_seen": 955383808
},
{
"epoch": 0.29,
"learning_rate": 0.0003588465396188566,
"loss": 2.7069,
"theoretical_loss": 3.6651208820119434,
"tokens_seen": 955449344
},
{
"epoch": 0.29,
"learning_rate": 0.00035883650952858574,
"loss": 2.7804,
"theoretical_loss": 3.6650967564305983,
"tokens_seen": 955514880
},
{
"epoch": 0.29,
"learning_rate": 0.00035882647943831497,
"loss": 2.7952,
"theoretical_loss": 3.6650726329671803,
"tokens_seen": 955580416
},
{
"epoch": 0.29,
"learning_rate": 0.00035881644934804415,
"loss": 2.5644,
"theoretical_loss": 3.6650485116213596,
"tokens_seen": 955645952
},
{
"epoch": 0.29,
"learning_rate": 0.00035880641925777333,
"loss": 2.8745,
"theoretical_loss": 3.6650243923928043,
"tokens_seen": 955711488
},
{
"epoch": 0.29,
"learning_rate": 0.0003587963891675025,
"loss": 2.7769,
"theoretical_loss": 3.6650002752811837,
"tokens_seen": 955777024
},
{
"epoch": 0.29,
"learning_rate": 0.00035878635907723175,
"loss": 2.8761,
"theoretical_loss": 3.664976160286167,
"tokens_seen": 955842560
},
{
"epoch": 0.29,
"learning_rate": 0.0003587763289869609,
"loss": 2.7607,
"theoretical_loss": 3.664952047407423,
"tokens_seen": 955908096
},
{
"epoch": 0.29,
"learning_rate": 0.0003587662988966901,
"loss": 2.7965,
"theoretical_loss": 3.664927936644621,
"tokens_seen": 955973632
},
{
"epoch": 0.29,
"learning_rate": 0.00035875626880641924,
"loss": 2.6413,
"theoretical_loss": 3.66490382799743,
"tokens_seen": 956039168
},
{
"epoch": 0.29,
"learning_rate": 0.0003587462387161485,
"loss": 2.9122,
"theoretical_loss": 3.66487972146552,
"tokens_seen": 956104704
},
{
"epoch": 0.29,
"learning_rate": 0.00035873620862587766,
"loss": 2.8972,
"theoretical_loss": 3.66485561704856,
"tokens_seen": 956170240
},
{
"epoch": 0.29,
"learning_rate": 0.00035872617853560684,
"loss": 2.6023,
"theoretical_loss": 3.66483151474622,
"tokens_seen": 956235776
},
{
"epoch": 0.29,
"learning_rate": 0.000358716148445336,
"loss": 2.8244,
"theoretical_loss": 3.664807414558169,
"tokens_seen": 956301312
},
{
"epoch": 0.29,
"learning_rate": 0.0003587061183550652,
"loss": 2.7275,
"theoretical_loss": 3.664783316484076,
"tokens_seen": 956366848
},
{
"epoch": 0.29,
"learning_rate": 0.0003586960882647944,
"loss": 2.527,
"theoretical_loss": 3.664759220523613,
"tokens_seen": 956432384
},
{
"epoch": 0.29,
"learning_rate": 0.0003586860581745236,
"loss": 2.8982,
"theoretical_loss": 3.664735126676448,
"tokens_seen": 956497920
},
{
"epoch": 0.29,
"learning_rate": 0.00035867602808425274,
"loss": 2.6432,
"theoretical_loss": 3.6647110349422505,
"tokens_seen": 956563456
},
{
"epoch": 0.29,
"learning_rate": 0.000358665997993982,
"loss": 2.5959,
"theoretical_loss": 3.6646869453206925,
"tokens_seen": 956628992
},
{
"epoch": 0.29,
"learning_rate": 0.00035865596790371116,
"loss": 2.6861,
"theoretical_loss": 3.6646628578114417,
"tokens_seen": 956694528
},
{
"epoch": 0.29,
"learning_rate": 0.00035864593781344034,
"loss": 2.84,
"theoretical_loss": 3.66463877241417,
"tokens_seen": 956760064
},
{
"epoch": 0.29,
"objective/train/docs_used": 549880,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7895476818084717,
"objective/train/theoretical_loss": 3.664614689128546,
"objective/train/tokens_used": 977285600,
"theoretical_loss": 3.664614689128546,
"tokens_seen": 956825600
},
{
"epoch": 0.29,
"learning_rate": 0.0003586359077231695,
"loss": 2.6545,
"theoretical_loss": 3.664614689128546,
"tokens_seen": 956825600
},
{
"epoch": 0.29,
"learning_rate": 0.0003586258776328987,
"loss": 2.7404,
"theoretical_loss": 3.6645906079542416,
"tokens_seen": 956891136
},
{
"epoch": 0.29,
"learning_rate": 0.0003586158475426279,
"loss": 2.6504,
"theoretical_loss": 3.6645665288909255,
"tokens_seen": 956956672
},
{
"epoch": 0.29,
"learning_rate": 0.0003586058174523571,
"loss": 2.7437,
"theoretical_loss": 3.66454245193827,
"tokens_seen": 957022208
},
{
"epoch": 0.29,
"learning_rate": 0.00035859578736208625,
"loss": 2.5396,
"theoretical_loss": 3.664518377095944,
"tokens_seen": 957087744
},
{
"epoch": 0.29,
"learning_rate": 0.0003585857572718155,
"loss": 2.9188,
"theoretical_loss": 3.6644943043636182,
"tokens_seen": 957153280
},
{
"epoch": 0.29,
"learning_rate": 0.0003585757271815446,
"loss": 2.7328,
"theoretical_loss": 3.6644702337409636,
"tokens_seen": 957218816
},
{
"epoch": 0.29,
"learning_rate": 0.00035856569709127384,
"loss": 2.8699,
"theoretical_loss": 3.664446165227651,
"tokens_seen": 957284352
},
{
"epoch": 0.29,
"learning_rate": 0.000358555667001003,
"loss": 2.5744,
"theoretical_loss": 3.6644220988233513,
"tokens_seen": 957349888
},
{
"epoch": 0.29,
"learning_rate": 0.0003585456369107322,
"loss": 2.8642,
"theoretical_loss": 3.6643980345277347,
"tokens_seen": 957415424
},
{
"epoch": 0.29,
"learning_rate": 0.0003585356068204614,
"loss": 2.6768,
"theoretical_loss": 3.6643739723404725,
"tokens_seen": 957480960
},
{
"epoch": 0.29,
"learning_rate": 0.00035852557673019057,
"loss": 3.0698,
"theoretical_loss": 3.664349912261236,
"tokens_seen": 957546496
},
{
"epoch": 0.29,
"learning_rate": 0.00035851554663991975,
"loss": 2.6459,
"theoretical_loss": 3.6643258542896957,
"tokens_seen": 957612032
},
{
"epoch": 0.29,
"learning_rate": 0.000358505516549649,
"loss": 2.9687,
"theoretical_loss": 3.664301798425523,
"tokens_seen": 957677568
},
{
"epoch": 0.29,
"learning_rate": 0.0003584954864593781,
"loss": 2.8791,
"theoretical_loss": 3.664277744668389,
"tokens_seen": 957743104
},
{
"epoch": 0.29,
"learning_rate": 0.00035848545636910735,
"loss": 2.969,
"theoretical_loss": 3.664253693017965,
"tokens_seen": 957808640
},
{
"epoch": 0.29,
"learning_rate": 0.00035847542627883653,
"loss": 2.8335,
"theoretical_loss": 3.6642296434739228,
"tokens_seen": 957874176
},
{
"epoch": 0.29,
"learning_rate": 0.0003584653961885657,
"loss": 2.6202,
"theoretical_loss": 3.664205596035933,
"tokens_seen": 957939712
},
{
"epoch": 0.29,
"learning_rate": 0.0003584553660982949,
"loss": 2.5163,
"theoretical_loss": 3.6641815507036677,
"tokens_seen": 958005248
},
{
"epoch": 0.29,
"learning_rate": 0.00035844533600802407,
"loss": 2.6266,
"theoretical_loss": 3.6641575074767982,
"tokens_seen": 958070784
},
{
"epoch": 0.29,
"learning_rate": 0.00035843530591775325,
"loss": 2.8701,
"theoretical_loss": 3.664133466354997,
"tokens_seen": 958136320
},
{
"epoch": 0.29,
"learning_rate": 0.0003584252758274825,
"loss": 2.6422,
"theoretical_loss": 3.664109427337934,
"tokens_seen": 958201856
},
{
"epoch": 0.29,
"learning_rate": 0.0003584152457372116,
"loss": 2.607,
"theoretical_loss": 3.664085390425283,
"tokens_seen": 958267392
},
{
"epoch": 0.29,
"learning_rate": 0.00035840521564694085,
"loss": 2.9064,
"theoretical_loss": 3.6640613556167145,
"tokens_seen": 958332928
},
{
"epoch": 0.29,
"learning_rate": 0.00035839518555667,
"loss": 2.7914,
"theoretical_loss": 3.664037322911901,
"tokens_seen": 958398464
},
{
"epoch": 0.29,
"objective/train/docs_used": 550706,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1120011806488037,
"objective/train/theoretical_loss": 3.664013292310515,
"objective/train/tokens_used": 978924000,
"theoretical_loss": 3.664013292310515,
"tokens_seen": 958464000
},
{
"epoch": 0.29,
"learning_rate": 0.0003583851554663992,
"loss": 2.8079,
"theoretical_loss": 3.664013292310515,
"tokens_seen": 958464000
},
{
"epoch": 0.29,
"learning_rate": 0.0003583751253761284,
"loss": 2.6562,
"theoretical_loss": 3.6639892638122276,
"tokens_seen": 958529536
},
{
"epoch": 0.29,
"learning_rate": 0.0003583650952858576,
"loss": 2.8738,
"theoretical_loss": 3.6639652374167113,
"tokens_seen": 958595072
},
{
"epoch": 0.29,
"learning_rate": 0.00035835506519558676,
"loss": 2.6755,
"theoretical_loss": 3.663941213123639,
"tokens_seen": 958660608
},
{
"epoch": 0.29,
"learning_rate": 0.00035834503510531594,
"loss": 2.7487,
"theoretical_loss": 3.6639171909326818,
"tokens_seen": 958726144
},
{
"epoch": 0.29,
"learning_rate": 0.0003583350050150451,
"loss": 2.7114,
"theoretical_loss": 3.6638931708435134,
"tokens_seen": 958791680
},
{
"epoch": 0.29,
"learning_rate": 0.00035832497492477435,
"loss": 2.6538,
"theoretical_loss": 3.6638691528558054,
"tokens_seen": 958857216
},
{
"epoch": 0.29,
"learning_rate": 0.0003583149448345035,
"loss": 2.7177,
"theoretical_loss": 3.663845136969231,
"tokens_seen": 958922752
},
{
"epoch": 0.29,
"learning_rate": 0.0003583049147442327,
"loss": 2.655,
"theoretical_loss": 3.6638211231834616,
"tokens_seen": 958988288
},
{
"epoch": 0.29,
"learning_rate": 0.0003582948846539619,
"loss": 2.6276,
"theoretical_loss": 3.6637971114981713,
"tokens_seen": 959053824
},
{
"epoch": 0.29,
"learning_rate": 0.0003582848545636911,
"loss": 2.7065,
"theoretical_loss": 3.6637731019130317,
"tokens_seen": 959119360
},
{
"epoch": 0.29,
"learning_rate": 0.00035827482447342026,
"loss": 2.6441,
"theoretical_loss": 3.6637490944277165,
"tokens_seen": 959184896
},
{
"epoch": 0.29,
"learning_rate": 0.00035826479438314944,
"loss": 2.9371,
"theoretical_loss": 3.663725089041898,
"tokens_seen": 959250432
},
{
"epoch": 0.29,
"learning_rate": 0.0003582547642928786,
"loss": 2.7021,
"theoretical_loss": 3.6637010857552497,
"tokens_seen": 959315968
},
{
"epoch": 0.29,
"learning_rate": 0.00035824473420260786,
"loss": 2.6329,
"theoretical_loss": 3.6636770845674445,
"tokens_seen": 959381504
},
{
"epoch": 0.29,
"learning_rate": 0.000358234704112337,
"loss": 2.9901,
"theoretical_loss": 3.6636530854781553,
"tokens_seen": 959447040
},
{
"epoch": 0.29,
"learning_rate": 0.0003582246740220662,
"loss": 2.5648,
"theoretical_loss": 3.6636290884870553,
"tokens_seen": 959512576
},
{
"epoch": 0.29,
"learning_rate": 0.00035821464393179535,
"loss": 2.9225,
"theoretical_loss": 3.663605093593818,
"tokens_seen": 959578112
},
{
"epoch": 0.29,
"learning_rate": 0.0003582046138415246,
"loss": 2.8341,
"theoretical_loss": 3.6635811007981163,
"tokens_seen": 959643648
},
{
"epoch": 0.29,
"learning_rate": 0.00035819458375125376,
"loss": 2.7224,
"theoretical_loss": 3.6635571100996245,
"tokens_seen": 959709184
},
{
"epoch": 0.29,
"learning_rate": 0.00035818455366098294,
"loss": 2.7777,
"theoretical_loss": 3.6635331214980154,
"tokens_seen": 959774720
},
{
"epoch": 0.29,
"learning_rate": 0.0003581745235707121,
"loss": 2.6866,
"theoretical_loss": 3.663509134992962,
"tokens_seen": 959840256
},
{
"epoch": 0.29,
"learning_rate": 0.00035816449348044136,
"loss": 2.7674,
"theoretical_loss": 3.6634851505841395,
"tokens_seen": 959905792
},
{
"epoch": 0.29,
"learning_rate": 0.0003581544633901705,
"loss": 2.7841,
"theoretical_loss": 3.66346116827122,
"tokens_seen": 959971328
},
{
"epoch": 0.29,
"learning_rate": 0.0003581444332998997,
"loss": 2.7275,
"theoretical_loss": 3.6634371880538783,
"tokens_seen": 960036864
},
{
"epoch": 0.29,
"objective/train/docs_used": 551132,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.612628221511841,
"objective/train/theoretical_loss": 3.6634132099317886,
"objective/train/tokens_used": 980562400,
"theoretical_loss": 3.6634132099317886,
"tokens_seen": 960102400
},
{
"epoch": 0.29,
"learning_rate": 0.00035813440320962885,
"loss": 2.79,
"theoretical_loss": 3.6634132099317886,
"tokens_seen": 960102400
},
{
"epoch": 0.29,
"learning_rate": 0.0003581243731193581,
"loss": 2.6352,
"theoretical_loss": 3.6633892339046232,
"tokens_seen": 960167936
},
{
"epoch": 0.29,
"learning_rate": 0.00035811434302908727,
"loss": 2.6469,
"theoretical_loss": 3.663365259972058,
"tokens_seen": 960233472
},
{
"epoch": 0.29,
"learning_rate": 0.00035810431293881645,
"loss": 2.5599,
"theoretical_loss": 3.6633412881337653,
"tokens_seen": 960299008
},
{
"epoch": 0.29,
"learning_rate": 0.0003580942828485457,
"loss": 2.7929,
"theoretical_loss": 3.6633173183894208,
"tokens_seen": 960364544
},
{
"epoch": 0.29,
"learning_rate": 0.0003580842527582748,
"loss": 2.9718,
"theoretical_loss": 3.6632933507386976,
"tokens_seen": 960430080
},
{
"epoch": 0.29,
"learning_rate": 0.00035807422266800404,
"loss": 2.7959,
"theoretical_loss": 3.6632693851812705,
"tokens_seen": 960495616
},
{
"epoch": 0.29,
"learning_rate": 0.0003580641925777332,
"loss": 2.5113,
"theoretical_loss": 3.663245421716814,
"tokens_seen": 960561152
},
{
"epoch": 0.29,
"learning_rate": 0.0003580541624874624,
"loss": 2.6789,
"theoretical_loss": 3.663221460345002,
"tokens_seen": 960626688
},
{
"epoch": 0.29,
"learning_rate": 0.0003580441323971916,
"loss": 2.7452,
"theoretical_loss": 3.6631975010655093,
"tokens_seen": 960692224
},
{
"epoch": 0.29,
"learning_rate": 0.00035803410230692077,
"loss": 2.8472,
"theoretical_loss": 3.6631735438780106,
"tokens_seen": 960757760
},
{
"epoch": 0.29,
"learning_rate": 0.00035802407221664995,
"loss": 2.7504,
"theoretical_loss": 3.663149588782181,
"tokens_seen": 960823296
},
{
"epoch": 0.29,
"learning_rate": 0.0003580140421263792,
"loss": 2.5795,
"theoretical_loss": 3.6631256357776945,
"tokens_seen": 960888832
},
{
"epoch": 0.29,
"learning_rate": 0.0003580040120361083,
"loss": 2.8623,
"theoretical_loss": 3.6631016848642255,
"tokens_seen": 960954368
},
{
"epoch": 0.29,
"learning_rate": 0.00035799398194583755,
"loss": 2.5766,
"theoretical_loss": 3.6630777360414504,
"tokens_seen": 961019904
},
{
"epoch": 0.29,
"learning_rate": 0.00035798395185556673,
"loss": 2.7627,
"theoretical_loss": 3.6630537893090422,
"tokens_seen": 961085440
},
{
"epoch": 0.29,
"learning_rate": 0.0003579739217652959,
"loss": 2.8586,
"theoretical_loss": 3.663029844666678,
"tokens_seen": 961150976
},
{
"epoch": 0.29,
"learning_rate": 0.0003579638916750251,
"loss": 2.7952,
"theoretical_loss": 3.663005902114031,
"tokens_seen": 961216512
},
{
"epoch": 0.29,
"learning_rate": 0.00035795386158475427,
"loss": 2.6944,
"theoretical_loss": 3.6629819616507775,
"tokens_seen": 961282048
},
{
"epoch": 0.29,
"learning_rate": 0.00035794383149448345,
"loss": 2.6976,
"theoretical_loss": 3.662958023276593,
"tokens_seen": 961347584
},
{
"epoch": 0.29,
"learning_rate": 0.0003579338014042127,
"loss": 3.0912,
"theoretical_loss": 3.662934086991151,
"tokens_seen": 961413120
},
{
"epoch": 0.29,
"learning_rate": 0.0003579237713139418,
"loss": 2.852,
"theoretical_loss": 3.6629101527941295,
"tokens_seen": 961478656
},
{
"epoch": 0.29,
"learning_rate": 0.00035791374122367105,
"loss": 2.8252,
"theoretical_loss": 3.662886220685202,
"tokens_seen": 961544192
},
{
"epoch": 0.29,
"learning_rate": 0.0003579037111334002,
"loss": 2.7743,
"theoretical_loss": 3.6628622906640445,
"tokens_seen": 961609728
},
{
"epoch": 0.29,
"learning_rate": 0.0003578936810431294,
"loss": 2.7297,
"theoretical_loss": 3.6628383627303327,
"tokens_seen": 961675264
},
{
"epoch": 0.29,
"objective/train/docs_used": 552317,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.984802722930908,
"objective/train/theoretical_loss": 3.6628144368837425,
"objective/train/tokens_used": 982200800,
"theoretical_loss": 3.6628144368837425,
"tokens_seen": 961740800
},
{
"epoch": 0.29,
"learning_rate": 0.0003578836509528586,
"loss": 2.971,
"theoretical_loss": 3.6628144368837425,
"tokens_seen": 961740800
},
{
"epoch": 0.29,
"learning_rate": 0.0003578736208625878,
"loss": 2.6001,
"theoretical_loss": 3.662790513123949,
"tokens_seen": 961806336
},
{
"epoch": 0.29,
"learning_rate": 0.00035786359077231696,
"loss": 2.8465,
"theoretical_loss": 3.662766591450629,
"tokens_seen": 961871872
},
{
"epoch": 0.29,
"learning_rate": 0.00035785356068204614,
"loss": 2.8041,
"theoretical_loss": 3.662742671863458,
"tokens_seen": 961937408
},
{
"epoch": 0.29,
"learning_rate": 0.0003578435305917753,
"loss": 2.6517,
"theoretical_loss": 3.6627187543621114,
"tokens_seen": 962002944
},
{
"epoch": 0.29,
"learning_rate": 0.00035783350050150455,
"loss": 2.9378,
"theoretical_loss": 3.6626948389462655,
"tokens_seen": 962068480
},
{
"epoch": 0.29,
"learning_rate": 0.0003578234704112337,
"loss": 3.0396,
"theoretical_loss": 3.662670925615597,
"tokens_seen": 962134016
},
{
"epoch": 0.29,
"learning_rate": 0.0003578134403209629,
"loss": 2.9046,
"theoretical_loss": 3.6626470143697816,
"tokens_seen": 962199552
},
{
"epoch": 0.29,
"learning_rate": 0.0003578034102306921,
"loss": 2.8725,
"theoretical_loss": 3.662623105208495,
"tokens_seen": 962265088
},
{
"epoch": 0.29,
"learning_rate": 0.0003577933801404213,
"loss": 2.5775,
"theoretical_loss": 3.6625991981314145,
"tokens_seen": 962330624
},
{
"epoch": 0.29,
"learning_rate": 0.00035778335005015046,
"loss": 2.8424,
"theoretical_loss": 3.6625752931382163,
"tokens_seen": 962396160
},
{
"epoch": 0.29,
"learning_rate": 0.00035777331995987964,
"loss": 2.9177,
"theoretical_loss": 3.6625513902285767,
"tokens_seen": 962461696
},
{
"epoch": 0.29,
"learning_rate": 0.0003577632898696088,
"loss": 2.6825,
"theoretical_loss": 3.6625274894021724,
"tokens_seen": 962527232
},
{
"epoch": 0.29,
"learning_rate": 0.00035775325977933806,
"loss": 2.5715,
"theoretical_loss": 3.6625035906586794,
"tokens_seen": 962592768
},
{
"epoch": 0.29,
"learning_rate": 0.0003577432296890672,
"loss": 2.8327,
"theoretical_loss": 3.662479693997775,
"tokens_seen": 962658304
},
{
"epoch": 0.29,
"learning_rate": 0.0003577331995987964,
"loss": 2.8361,
"theoretical_loss": 3.662455799419136,
"tokens_seen": 962723840
},
{
"epoch": 0.29,
"learning_rate": 0.00035772316950852555,
"loss": 2.7667,
"theoretical_loss": 3.6624319069224387,
"tokens_seen": 962789376
},
{
"epoch": 0.29,
"learning_rate": 0.0003577131394182548,
"loss": 2.8045,
"theoretical_loss": 3.66240801650736,
"tokens_seen": 962854912
},
{
"epoch": 0.29,
"learning_rate": 0.00035770310932798396,
"loss": 2.7772,
"theoretical_loss": 3.662384128173578,
"tokens_seen": 962920448
},
{
"epoch": 0.29,
"learning_rate": 0.00035769307923771314,
"loss": 2.7664,
"theoretical_loss": 3.6623602419207684,
"tokens_seen": 962985984
},
{
"epoch": 0.29,
"learning_rate": 0.0003576830491474423,
"loss": 2.7752,
"theoretical_loss": 3.662336357748609,
"tokens_seen": 963051520
},
{
"epoch": 0.29,
"learning_rate": 0.00035767301905717156,
"loss": 2.7263,
"theoretical_loss": 3.662312475656777,
"tokens_seen": 963117056
},
{
"epoch": 0.29,
"learning_rate": 0.0003576629889669007,
"loss": 2.9977,
"theoretical_loss": 3.6622885956449496,
"tokens_seen": 963182592
},
{
"epoch": 0.29,
"learning_rate": 0.0003576529588766299,
"loss": 2.5862,
"theoretical_loss": 3.6622647177128034,
"tokens_seen": 963248128
},
{
"epoch": 0.29,
"learning_rate": 0.00035764292878635905,
"loss": 2.8692,
"theoretical_loss": 3.662240841860017,
"tokens_seen": 963313664
},
{
"epoch": 0.29,
"objective/train/docs_used": 552873,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.86100172996521,
"objective/train/theoretical_loss": 3.662216968086267,
"objective/train/tokens_used": 983839200,
"theoretical_loss": 3.662216968086267,
"tokens_seen": 963379200
},
{
"epoch": 0.29,
"learning_rate": 0.0003576328986960883,
"loss": 3.0285,
"theoretical_loss": 3.662216968086267,
"tokens_seen": 963379200
},
{
"epoch": 0.29,
"learning_rate": 0.00035762286860581747,
"loss": 3.0706,
"theoretical_loss": 3.6621930963912317,
"tokens_seen": 963444736
},
{
"epoch": 0.29,
"learning_rate": 0.00035761283851554665,
"loss": 2.8019,
"theoretical_loss": 3.662169226774588,
"tokens_seen": 963510272
},
{
"epoch": 0.29,
"learning_rate": 0.00035760280842527583,
"loss": 2.6545,
"theoretical_loss": 3.662145359236014,
"tokens_seen": 963575808
},
{
"epoch": 0.29,
"learning_rate": 0.000357592778335005,
"loss": 2.7602,
"theoretical_loss": 3.6621214937751874,
"tokens_seen": 963641344
},
{
"epoch": 0.29,
"learning_rate": 0.0003575827482447342,
"loss": 2.5556,
"theoretical_loss": 3.662097630391786,
"tokens_seen": 963706880
},
{
"epoch": 0.29,
"learning_rate": 0.0003575727181544634,
"loss": 2.8026,
"theoretical_loss": 3.662073769085488,
"tokens_seen": 963772416
},
{
"epoch": 0.29,
"learning_rate": 0.00035756268806419255,
"loss": 2.8048,
"theoretical_loss": 3.6620499098559707,
"tokens_seen": 963837952
},
{
"epoch": 0.29,
"learning_rate": 0.0003575526579739218,
"loss": 2.6082,
"theoretical_loss": 3.6620260527029127,
"tokens_seen": 963903488
},
{
"epoch": 0.29,
"learning_rate": 0.0003575426278836509,
"loss": 2.7888,
"theoretical_loss": 3.662002197625992,
"tokens_seen": 963969024
},
{
"epoch": 0.29,
"learning_rate": 0.00035753259779338015,
"loss": 2.5317,
"theoretical_loss": 3.661978344624887,
"tokens_seen": 964034560
},
{
"epoch": 0.29,
"learning_rate": 0.00035752256770310933,
"loss": 2.8318,
"theoretical_loss": 3.6619544936992754,
"tokens_seen": 964100096
},
{
"epoch": 0.29,
"learning_rate": 0.0003575125376128385,
"loss": 2.7315,
"theoretical_loss": 3.661930644848836,
"tokens_seen": 964165632
},
{
"epoch": 0.29,
"learning_rate": 0.0003575025075225677,
"loss": 2.7713,
"theoretical_loss": 3.661906798073247,
"tokens_seen": 964231168
},
{
"epoch": 0.29,
"learning_rate": 0.00035749247743229693,
"loss": 2.8709,
"theoretical_loss": 3.6618829533721873,
"tokens_seen": 964296704
},
{
"epoch": 0.29,
"learning_rate": 0.00035748244734202606,
"loss": 2.6504,
"theoretical_loss": 3.661859110745335,
"tokens_seen": 964362240
},
{
"epoch": 0.29,
"learning_rate": 0.0003574724172517553,
"loss": 2.4641,
"theoretical_loss": 3.661835270192369,
"tokens_seen": 964427776
},
{
"epoch": 0.29,
"learning_rate": 0.0003574623871614844,
"loss": 2.9216,
"theoretical_loss": 3.6618114317129677,
"tokens_seen": 964493312
},
{
"epoch": 0.29,
"learning_rate": 0.00035745235707121365,
"loss": 2.7216,
"theoretical_loss": 3.6617875953068095,
"tokens_seen": 964558848
},
{
"epoch": 0.29,
"learning_rate": 0.00035744232698094283,
"loss": 2.6892,
"theoretical_loss": 3.6617637609735745,
"tokens_seen": 964624384
},
{
"epoch": 0.29,
"learning_rate": 0.000357432296890672,
"loss": 2.6432,
"theoretical_loss": 3.661739928712941,
"tokens_seen": 964689920
},
{
"epoch": 0.29,
"learning_rate": 0.0003574222668004012,
"loss": 2.8089,
"theoretical_loss": 3.661716098524588,
"tokens_seen": 964755456
},
{
"epoch": 0.29,
"learning_rate": 0.0003574122367101304,
"loss": 2.8221,
"theoretical_loss": 3.661692270408194,
"tokens_seen": 964820992
},
{
"epoch": 0.29,
"learning_rate": 0.00035740220661985956,
"loss": 2.5763,
"theoretical_loss": 3.661668444363438,
"tokens_seen": 964886528
},
{
"epoch": 0.29,
"learning_rate": 0.0003573921765295888,
"loss": 2.5806,
"theoretical_loss": 3.6616446203900006,
"tokens_seen": 964952064
},
{
"epoch": 0.29,
"objective/train/docs_used": 554313,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1256487369537354,
"objective/train/theoretical_loss": 3.6616207984875606,
"objective/train/tokens_used": 985477600,
"theoretical_loss": 3.6616207984875606,
"tokens_seen": 965017600
},
{
"epoch": 0.29,
"learning_rate": 0.0003573821464393179,
"loss": 2.7724,
"theoretical_loss": 3.6616207984875606,
"tokens_seen": 965017600
},
{
"epoch": 0.29,
"learning_rate": 0.00035737211634904716,
"loss": 2.7063,
"theoretical_loss": 3.6615969786557967,
"tokens_seen": 965083136
},
{
"epoch": 0.29,
"learning_rate": 0.0003573620862587763,
"loss": 2.8369,
"theoretical_loss": 3.6615731608943882,
"tokens_seen": 965148672
},
{
"epoch": 0.29,
"learning_rate": 0.0003573520561685055,
"loss": 3.0536,
"theoretical_loss": 3.6615493452030154,
"tokens_seen": 965214208
},
{
"epoch": 0.29,
"learning_rate": 0.00035734202607823475,
"loss": 2.5745,
"theoretical_loss": 3.661525531581357,
"tokens_seen": 965279744
},
{
"epoch": 0.29,
"learning_rate": 0.0003573319959879639,
"loss": 2.6382,
"theoretical_loss": 3.6615017200290936,
"tokens_seen": 965345280
},
{
"epoch": 0.29,
"learning_rate": 0.0003573219658976931,
"loss": 2.5736,
"theoretical_loss": 3.661477910545904,
"tokens_seen": 965410816
},
{
"epoch": 0.29,
"learning_rate": 0.0003573119358074223,
"loss": 2.6105,
"theoretical_loss": 3.6614541031314687,
"tokens_seen": 965476352
},
{
"epoch": 0.29,
"learning_rate": 0.0003573019057171515,
"loss": 2.5517,
"theoretical_loss": 3.6614302977854667,
"tokens_seen": 965541888
},
{
"epoch": 0.29,
"learning_rate": 0.00035729187562688066,
"loss": 3.0012,
"theoretical_loss": 3.6614064945075784,
"tokens_seen": 965607424
},
{
"epoch": 0.29,
"learning_rate": 0.00035728184553660984,
"loss": 2.7646,
"theoretical_loss": 3.661382693297484,
"tokens_seen": 965672960
},
{
"epoch": 0.29,
"learning_rate": 0.000357271815446339,
"loss": 2.4853,
"theoretical_loss": 3.661358894154864,
"tokens_seen": 965738496
},
{
"epoch": 0.29,
"learning_rate": 0.00035726178535606826,
"loss": 2.5964,
"theoretical_loss": 3.661335097079397,
"tokens_seen": 965804032
},
{
"epoch": 0.29,
"learning_rate": 0.0003572517552657974,
"loss": 2.6459,
"theoretical_loss": 3.6613113020707644,
"tokens_seen": 965869568
},
{
"epoch": 0.29,
"learning_rate": 0.0003572417251755266,
"loss": 2.7305,
"theoretical_loss": 3.6612875091286465,
"tokens_seen": 965935104
},
{
"epoch": 0.29,
"learning_rate": 0.00035723169508525575,
"loss": 2.6997,
"theoretical_loss": 3.661263718252723,
"tokens_seen": 966000640
},
{
"epoch": 0.29,
"learning_rate": 0.000357221664994985,
"loss": 2.8212,
"theoretical_loss": 3.661239929442674,
"tokens_seen": 966066176
},
{
"epoch": 0.29,
"learning_rate": 0.00035721163490471416,
"loss": 2.584,
"theoretical_loss": 3.6612161426981817,
"tokens_seen": 966131712
},
{
"epoch": 0.29,
"learning_rate": 0.00035720160481444334,
"loss": 2.694,
"theoretical_loss": 3.6611923580189245,
"tokens_seen": 966197248
},
{
"epoch": 0.29,
"learning_rate": 0.0003571915747241725,
"loss": 2.8091,
"theoretical_loss": 3.661168575404585,
"tokens_seen": 966262784
},
{
"epoch": 0.29,
"learning_rate": 0.00035718154463390176,
"loss": 2.6221,
"theoretical_loss": 3.6611447948548426,
"tokens_seen": 966328320
},
{
"epoch": 0.29,
"learning_rate": 0.0003571715145436309,
"loss": 2.636,
"theoretical_loss": 3.6611210163693784,
"tokens_seen": 966393856
},
{
"epoch": 0.29,
"learning_rate": 0.0003571614844533601,
"loss": 2.721,
"theoretical_loss": 3.661097239947873,
"tokens_seen": 966459392
},
{
"epoch": 0.29,
"learning_rate": 0.00035715145436308925,
"loss": 2.6895,
"theoretical_loss": 3.661073465590008,
"tokens_seen": 966524928
},
{
"epoch": 0.29,
"learning_rate": 0.0003571414242728185,
"loss": 2.5593,
"theoretical_loss": 3.6610496932954635,
"tokens_seen": 966590464
},
{
"epoch": 0.29,
"objective/train/docs_used": 554946,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4580020904541016,
"objective/train/theoretical_loss": 3.6610259230639217,
"objective/train/tokens_used": 987116000,
"theoretical_loss": 3.6610259230639217,
"tokens_seen": 966656000
},
{
"epoch": 0.29,
"learning_rate": 0.00035713139418254767,
"loss": 2.6539,
"theoretical_loss": 3.6610259230639217,
"tokens_seen": 966656000
},
{
"epoch": 0.29,
"learning_rate": 0.00035712136409227685,
"loss": 2.6217,
"theoretical_loss": 3.661002154895062,
"tokens_seen": 966721536
},
{
"epoch": 0.29,
"learning_rate": 0.00035711133400200603,
"loss": 2.6341,
"theoretical_loss": 3.6609783887885676,
"tokens_seen": 966787072
},
{
"epoch": 0.29,
"learning_rate": 0.0003571013039117352,
"loss": 2.4883,
"theoretical_loss": 3.6609546247441185,
"tokens_seen": 966852608
},
{
"epoch": 0.29,
"learning_rate": 0.0003570912738214644,
"loss": 2.6018,
"theoretical_loss": 3.660930862761396,
"tokens_seen": 966918144
},
{
"epoch": 0.29,
"learning_rate": 0.0003570812437311936,
"loss": 2.7759,
"theoretical_loss": 3.660907102840082,
"tokens_seen": 966983680
},
{
"epoch": 0.29,
"learning_rate": 0.00035707121364092275,
"loss": 3.0174,
"theoretical_loss": 3.660883344979858,
"tokens_seen": 967049216
},
{
"epoch": 0.29,
"learning_rate": 0.000357061183550652,
"loss": 2.6203,
"theoretical_loss": 3.6608595891804057,
"tokens_seen": 967114752
},
{
"epoch": 0.29,
"learning_rate": 0.0003570511534603811,
"loss": 2.9698,
"theoretical_loss": 3.660835835441406,
"tokens_seen": 967180288
},
{
"epoch": 0.29,
"learning_rate": 0.00035704112337011035,
"loss": 2.7082,
"theoretical_loss": 3.660812083762541,
"tokens_seen": 967245824
},
{
"epoch": 0.29,
"learning_rate": 0.00035703109327983953,
"loss": 2.7959,
"theoretical_loss": 3.660788334143492,
"tokens_seen": 967311360
},
{
"epoch": 0.29,
"learning_rate": 0.0003570210631895687,
"loss": 2.8344,
"theoretical_loss": 3.6607645865839418,
"tokens_seen": 967376896
},
{
"epoch": 0.29,
"learning_rate": 0.0003570110330992979,
"loss": 2.7334,
"theoretical_loss": 3.6607408410835713,
"tokens_seen": 967442432
},
{
"epoch": 0.29,
"learning_rate": 0.00035700100300902713,
"loss": 2.5863,
"theoretical_loss": 3.6607170976420638,
"tokens_seen": 967507968
},
{
"epoch": 0.29,
"learning_rate": 0.00035699097291875626,
"loss": 2.5569,
"theoretical_loss": 3.6606933562590998,
"tokens_seen": 967573504
},
{
"epoch": 0.29,
"learning_rate": 0.0003569809428284855,
"loss": 2.827,
"theoretical_loss": 3.660669616934362,
"tokens_seen": 967639040
},
{
"epoch": 0.29,
"learning_rate": 0.0003569709127382146,
"loss": 2.7125,
"theoretical_loss": 3.660645879667533,
"tokens_seen": 967704576
},
{
"epoch": 0.29,
"learning_rate": 0.00035696088264794385,
"loss": 2.5385,
"theoretical_loss": 3.6606221444582943,
"tokens_seen": 967770112
},
{
"epoch": 0.29,
"learning_rate": 0.00035695085255767303,
"loss": 2.8582,
"theoretical_loss": 3.6605984113063292,
"tokens_seen": 967835648
},
{
"epoch": 0.29,
"learning_rate": 0.0003569408224674022,
"loss": 2.4556,
"theoretical_loss": 3.6605746802113197,
"tokens_seen": 967901184
},
{
"epoch": 0.29,
"learning_rate": 0.0003569307923771314,
"loss": 2.8685,
"theoretical_loss": 3.6605509511729473,
"tokens_seen": 967966720
},
{
"epoch": 0.29,
"learning_rate": 0.0003569207622868606,
"loss": 2.9663,
"theoretical_loss": 3.660527224190896,
"tokens_seen": 968032256
},
{
"epoch": 0.29,
"learning_rate": 0.00035691073219658976,
"loss": 2.548,
"theoretical_loss": 3.6605034992648475,
"tokens_seen": 968097792
},
{
"epoch": 0.29,
"learning_rate": 0.000356900702106319,
"loss": 2.7812,
"theoretical_loss": 3.6604797763944843,
"tokens_seen": 968163328
},
{
"epoch": 0.29,
"learning_rate": 0.0003568906720160481,
"loss": 2.6745,
"theoretical_loss": 3.6604560555794903,
"tokens_seen": 968228864
},
{
"epoch": 0.29,
"objective/train/docs_used": 556500,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.069166660308838,
"objective/train/theoretical_loss": 3.6604323368195475,
"objective/train/tokens_used": 988754400,
"theoretical_loss": 3.6604323368195475,
"tokens_seen": 968294400
},
{
"epoch": 0.29,
"learning_rate": 0.00035688064192577736,
"loss": 2.7138,
"theoretical_loss": 3.6604323368195475,
"tokens_seen": 968294400
},
{
"epoch": 0.29,
"learning_rate": 0.0003568706118355065,
"loss": 2.7409,
"theoretical_loss": 3.660408620114339,
"tokens_seen": 968359936
},
{
"epoch": 0.29,
"learning_rate": 0.0003568605817452357,
"loss": 2.6894,
"theoretical_loss": 3.6603849054635473,
"tokens_seen": 968425472
},
{
"epoch": 0.29,
"learning_rate": 0.0003568505516549649,
"loss": 2.6524,
"theoretical_loss": 3.660361192866856,
"tokens_seen": 968491008
},
{
"epoch": 0.29,
"learning_rate": 0.0003568405215646941,
"loss": 2.9258,
"theoretical_loss": 3.660337482323948,
"tokens_seen": 968556544
},
{
"epoch": 0.29,
"learning_rate": 0.00035683049147442326,
"loss": 2.6823,
"theoretical_loss": 3.660313773834506,
"tokens_seen": 968622080
},
{
"epoch": 0.29,
"learning_rate": 0.0003568204613841525,
"loss": 2.9387,
"theoretical_loss": 3.660290067398214,
"tokens_seen": 968687616
},
{
"epoch": 0.29,
"learning_rate": 0.0003568104312938816,
"loss": 2.8882,
"theoretical_loss": 3.660266363014755,
"tokens_seen": 968753152
},
{
"epoch": 0.29,
"learning_rate": 0.00035680040120361086,
"loss": 2.5292,
"theoretical_loss": 3.6602426606838128,
"tokens_seen": 968818688
},
{
"epoch": 0.29,
"learning_rate": 0.00035679037111334,
"loss": 2.7123,
"theoretical_loss": 3.6602189604050697,
"tokens_seen": 968884224
},
{
"epoch": 0.29,
"learning_rate": 0.0003567803410230692,
"loss": 2.3463,
"theoretical_loss": 3.6601952621782106,
"tokens_seen": 968949760
},
{
"epoch": 0.29,
"learning_rate": 0.0003567703109327984,
"loss": 2.7714,
"theoretical_loss": 3.660171566002918,
"tokens_seen": 969015296
},
{
"epoch": 0.29,
"learning_rate": 0.0003567602808425276,
"loss": 2.7549,
"theoretical_loss": 3.660147871878876,
"tokens_seen": 969080832
},
{
"epoch": 0.29,
"learning_rate": 0.00035675025075225677,
"loss": 2.8196,
"theoretical_loss": 3.660124179805769,
"tokens_seen": 969146368
},
{
"epoch": 0.29,
"learning_rate": 0.00035674022066198595,
"loss": 2.9975,
"theoretical_loss": 3.6601004897832796,
"tokens_seen": 969211904
},
{
"epoch": 0.29,
"learning_rate": 0.00035673019057171513,
"loss": 2.5688,
"theoretical_loss": 3.6600768018110923,
"tokens_seen": 969277440
},
{
"epoch": 0.29,
"learning_rate": 0.00035672016048144436,
"loss": 2.5278,
"theoretical_loss": 3.660053115888891,
"tokens_seen": 969342976
},
{
"epoch": 0.29,
"learning_rate": 0.0003567101303911735,
"loss": 2.8858,
"theoretical_loss": 3.6600294320163593,
"tokens_seen": 969408512
},
{
"epoch": 0.29,
"learning_rate": 0.0003567001003009027,
"loss": 2.6697,
"theoretical_loss": 3.660005750193182,
"tokens_seen": 969474048
},
{
"epoch": 0.29,
"learning_rate": 0.00035669007021063185,
"loss": 2.9448,
"theoretical_loss": 3.6599820704190433,
"tokens_seen": 969539584
},
{
"epoch": 0.29,
"learning_rate": 0.0003566800401203611,
"loss": 2.8043,
"theoretical_loss": 3.6599583926936265,
"tokens_seen": 969605120
},
{
"epoch": 0.29,
"learning_rate": 0.00035667001003009027,
"loss": 2.8298,
"theoretical_loss": 3.6599347170166165,
"tokens_seen": 969670656
},
{
"epoch": 0.29,
"learning_rate": 0.00035665997993981945,
"loss": 2.6407,
"theoretical_loss": 3.659911043387698,
"tokens_seen": 969736192
},
{
"epoch": 0.29,
"learning_rate": 0.00035664994984954863,
"loss": 2.5983,
"theoretical_loss": 3.6598873718065548,
"tokens_seen": 969801728
},
{
"epoch": 0.29,
"learning_rate": 0.00035663991975927787,
"loss": 2.6588,
"theoretical_loss": 3.6598637022728715,
"tokens_seen": 969867264
},
{
"epoch": 0.29,
"objective/train/docs_used": 557201,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6766343116760254,
"objective/train/theoretical_loss": 3.659840034786333,
"objective/train/tokens_used": 990392800,
"theoretical_loss": 3.659840034786333,
"tokens_seen": 969932800
},
{
"epoch": 0.29,
"learning_rate": 0.000356629889669007,
"loss": 2.6032,
"theoretical_loss": 3.659840034786333,
"tokens_seen": 969932800
},
{
"epoch": 0.29,
"learning_rate": 0.00035661985957873623,
"loss": 2.6817,
"theoretical_loss": 3.6598163693466232,
"tokens_seen": 969998336
},
{
"epoch": 0.29,
"learning_rate": 0.00035660982948846536,
"loss": 2.6757,
"theoretical_loss": 3.6597927059534276,
"tokens_seen": 970063872
},
{
"epoch": 0.29,
"learning_rate": 0.0003565997993981946,
"loss": 2.693,
"theoretical_loss": 3.6597690446064313,
"tokens_seen": 970129408
},
{
"epoch": 0.29,
"learning_rate": 0.0003565897693079238,
"loss": 2.6001,
"theoretical_loss": 3.659745385305318,
"tokens_seen": 970194944
},
{
"epoch": 0.29,
"learning_rate": 0.00035657973921765295,
"loss": 2.8058,
"theoretical_loss": 3.6597217280497736,
"tokens_seen": 970260480
},
{
"epoch": 0.29,
"learning_rate": 0.0003565697091273822,
"loss": 2.9184,
"theoretical_loss": 3.659698072839482,
"tokens_seen": 970326016
},
{
"epoch": 0.29,
"learning_rate": 0.0003565596790371113,
"loss": 2.4896,
"theoretical_loss": 3.6596744196741295,
"tokens_seen": 970391552
},
{
"epoch": 0.29,
"learning_rate": 0.00035654964894684055,
"loss": 2.5986,
"theoretical_loss": 3.6596507685534005,
"tokens_seen": 970457088
},
{
"epoch": 0.29,
"learning_rate": 0.00035653961885656973,
"loss": 2.7272,
"theoretical_loss": 3.6596271194769807,
"tokens_seen": 970522624
},
{
"epoch": 0.29,
"learning_rate": 0.0003565295887662989,
"loss": 2.9751,
"theoretical_loss": 3.659603472444555,
"tokens_seen": 970588160
},
{
"epoch": 0.29,
"learning_rate": 0.0003565195586760281,
"loss": 2.9842,
"theoretical_loss": 3.6595798274558087,
"tokens_seen": 970653696
},
{
"epoch": 0.29,
"learning_rate": 0.00035650952858575733,
"loss": 2.8473,
"theoretical_loss": 3.6595561845104276,
"tokens_seen": 970719232
},
{
"epoch": 0.29,
"learning_rate": 0.00035649949849548646,
"loss": 2.6646,
"theoretical_loss": 3.6595325436080968,
"tokens_seen": 970784768
},
{
"epoch": 0.29,
"learning_rate": 0.0003564894684052157,
"loss": 2.7958,
"theoretical_loss": 3.6595089047485017,
"tokens_seen": 970850304
},
{
"epoch": 0.29,
"learning_rate": 0.0003564794383149448,
"loss": 2.576,
"theoretical_loss": 3.6594852679313283,
"tokens_seen": 970915840
},
{
"epoch": 0.29,
"learning_rate": 0.00035646940822467405,
"loss": 2.6872,
"theoretical_loss": 3.659461633156262,
"tokens_seen": 970981376
},
{
"epoch": 0.29,
"learning_rate": 0.00035645937813440323,
"loss": 2.7364,
"theoretical_loss": 3.6594380004229894,
"tokens_seen": 971046912
},
{
"epoch": 0.29,
"learning_rate": 0.0003564493480441324,
"loss": 2.7545,
"theoretical_loss": 3.659414369731195,
"tokens_seen": 971112448
},
{
"epoch": 0.29,
"learning_rate": 0.0003564393179538616,
"loss": 2.5612,
"theoretical_loss": 3.659390741080565,
"tokens_seen": 971177984
},
{
"epoch": 0.29,
"learning_rate": 0.0003564292878635908,
"loss": 2.5606,
"theoretical_loss": 3.659367114470786,
"tokens_seen": 971243520
},
{
"epoch": 0.29,
"learning_rate": 0.00035641925777331996,
"loss": 2.6111,
"theoretical_loss": 3.659343489901544,
"tokens_seen": 971309056
},
{
"epoch": 0.29,
"learning_rate": 0.0003564092276830492,
"loss": 2.9236,
"theoretical_loss": 3.659319867372525,
"tokens_seen": 971374592
},
{
"epoch": 0.29,
"learning_rate": 0.0003563991975927783,
"loss": 2.8843,
"theoretical_loss": 3.6592962468834145,
"tokens_seen": 971440128
},
{
"epoch": 0.29,
"learning_rate": 0.00035638916750250756,
"loss": 2.7683,
"theoretical_loss": 3.6592726284339,
"tokens_seen": 971505664
},
{
"epoch": 0.29,
"objective/train/docs_used": 558372,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.605837106704712,
"objective/train/theoretical_loss": 3.6592490120236665,
"objective/train/tokens_used": 992031200,
"theoretical_loss": 3.6592490120236665,
"tokens_seen": 971571200
},
{
"epoch": 0.29,
"learning_rate": 0.0003563791374122367,
"loss": 2.6944,
"theoretical_loss": 3.6592490120236665,
"tokens_seen": 971571200
},
{
"epoch": 0.29,
"learning_rate": 0.0003563691073219659,
"loss": 2.7385,
"theoretical_loss": 3.6592253976524014,
"tokens_seen": 971636736
},
{
"epoch": 0.29,
"learning_rate": 0.0003563590772316951,
"loss": 2.6796,
"theoretical_loss": 3.65920178531979,
"tokens_seen": 971702272
},
{
"epoch": 0.29,
"learning_rate": 0.0003563490471414243,
"loss": 2.8781,
"theoretical_loss": 3.65917817502552,
"tokens_seen": 971767808
},
{
"epoch": 0.29,
"learning_rate": 0.00035633901705115346,
"loss": 2.7765,
"theoretical_loss": 3.6591545667692777,
"tokens_seen": 971833344
},
{
"epoch": 0.29,
"learning_rate": 0.0003563289869608827,
"loss": 2.6282,
"theoretical_loss": 3.6591309605507494,
"tokens_seen": 971898880
},
{
"epoch": 0.29,
"learning_rate": 0.0003563189568706118,
"loss": 2.7455,
"theoretical_loss": 3.6591073563696224,
"tokens_seen": 971964416
},
{
"epoch": 0.29,
"learning_rate": 0.00035630892678034106,
"loss": 2.5957,
"theoretical_loss": 3.6590837542255827,
"tokens_seen": 972029952
},
{
"epoch": 0.29,
"learning_rate": 0.0003562988966900702,
"loss": 2.8167,
"theoretical_loss": 3.6590601541183174,
"tokens_seen": 972095488
},
{
"epoch": 0.29,
"learning_rate": 0.0003562888665997994,
"loss": 2.5397,
"theoretical_loss": 3.659036556047514,
"tokens_seen": 972161024
},
{
"epoch": 0.29,
"learning_rate": 0.0003562788365095286,
"loss": 2.891,
"theoretical_loss": 3.659012960012859,
"tokens_seen": 972226560
},
{
"epoch": 0.29,
"learning_rate": 0.0003562688064192578,
"loss": 2.9036,
"theoretical_loss": 3.65898936601404,
"tokens_seen": 972292096
},
{
"epoch": 0.29,
"learning_rate": 0.00035625877632898697,
"loss": 2.6923,
"theoretical_loss": 3.6589657740507437,
"tokens_seen": 972357632
},
{
"epoch": 0.29,
"learning_rate": 0.00035624874623871615,
"loss": 2.8141,
"theoretical_loss": 3.658942184122657,
"tokens_seen": 972423168
},
{
"epoch": 0.29,
"learning_rate": 0.00035623871614844533,
"loss": 3.0178,
"theoretical_loss": 3.6589185962294675,
"tokens_seen": 972488704
},
{
"epoch": 0.29,
"learning_rate": 0.00035622868605817456,
"loss": 2.4921,
"theoretical_loss": 3.658895010370863,
"tokens_seen": 972554240
},
{
"epoch": 0.29,
"learning_rate": 0.0003562186559679037,
"loss": 2.6415,
"theoretical_loss": 3.65887142654653,
"tokens_seen": 972619776
},
{
"epoch": 0.29,
"learning_rate": 0.0003562086258776329,
"loss": 2.4167,
"theoretical_loss": 3.658847844756157,
"tokens_seen": 972685312
},
{
"epoch": 0.29,
"learning_rate": 0.00035619859578736205,
"loss": 2.6767,
"theoretical_loss": 3.6588242649994314,
"tokens_seen": 972750848
},
{
"epoch": 0.29,
"learning_rate": 0.0003561885656970913,
"loss": 3.0011,
"theoretical_loss": 3.65880068727604,
"tokens_seen": 972816384
},
{
"epoch": 0.29,
"learning_rate": 0.00035617853560682047,
"loss": 2.7456,
"theoretical_loss": 3.658777111585671,
"tokens_seen": 972881920
},
{
"epoch": 0.29,
"learning_rate": 0.00035616850551654965,
"loss": 2.7208,
"theoretical_loss": 3.6587535379280123,
"tokens_seen": 972947456
},
{
"epoch": 0.29,
"learning_rate": 0.00035615847542627883,
"loss": 2.8736,
"theoretical_loss": 3.6587299663027517,
"tokens_seen": 973012992
},
{
"epoch": 0.29,
"learning_rate": 0.00035614844533600807,
"loss": 2.6401,
"theoretical_loss": 3.6587063967095776,
"tokens_seen": 973078528
},
{
"epoch": 0.29,
"learning_rate": 0.0003561384152457372,
"loss": 2.868,
"theoretical_loss": 3.6586828291481766,
"tokens_seen": 973144064
},
{
"epoch": 0.29,
"objective/train/docs_used": 558840,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0834107398986816,
"objective/train/theoretical_loss": 3.6586592636182376,
"objective/train/tokens_used": 993669600,
"theoretical_loss": 3.6586592636182376,
"tokens_seen": 973209600
},
{
"epoch": 0.29,
"learning_rate": 0.00035612838515546643,
"loss": 2.7695,
"theoretical_loss": 3.6586592636182376,
"tokens_seen": 973209600
},
{
"epoch": 0.29,
"learning_rate": 0.00035611835506519556,
"loss": 2.6722,
"theoretical_loss": 3.658635700119449,
"tokens_seen": 973275136
},
{
"epoch": 0.29,
"learning_rate": 0.0003561083249749248,
"loss": 2.6688,
"theoretical_loss": 3.658612138651498,
"tokens_seen": 973340672
},
{
"epoch": 0.29,
"learning_rate": 0.00035609829488465397,
"loss": 2.7194,
"theoretical_loss": 3.6585885792140744,
"tokens_seen": 973406208
},
{
"epoch": 0.29,
"learning_rate": 0.00035608826479438315,
"loss": 2.5585,
"theoretical_loss": 3.658565021806865,
"tokens_seen": 973471744
},
{
"epoch": 0.3,
"learning_rate": 0.00035607823470411233,
"loss": 2.7061,
"theoretical_loss": 3.658541466429559,
"tokens_seen": 973537280
},
{
"epoch": 0.3,
"learning_rate": 0.0003560682046138415,
"loss": 2.9302,
"theoretical_loss": 3.6585179130818446,
"tokens_seen": 973602816
},
{
"epoch": 0.3,
"learning_rate": 0.0003560581745235707,
"loss": 2.6372,
"theoretical_loss": 3.6584943617634105,
"tokens_seen": 973668352
},
{
"epoch": 0.3,
"learning_rate": 0.00035604814443329993,
"loss": 2.5818,
"theoretical_loss": 3.6584708124739453,
"tokens_seen": 973733888
},
{
"epoch": 0.3,
"learning_rate": 0.00035603811434302906,
"loss": 2.6986,
"theoretical_loss": 3.658447265213137,
"tokens_seen": 973799424
},
{
"epoch": 0.3,
"learning_rate": 0.0003560280842527583,
"loss": 2.6995,
"theoretical_loss": 3.658423719980675,
"tokens_seen": 973864960
},
{
"epoch": 0.3,
"learning_rate": 0.0003560180541624874,
"loss": 2.6033,
"theoretical_loss": 3.658400176776248,
"tokens_seen": 973930496
},
{
"epoch": 0.3,
"learning_rate": 0.00035600802407221666,
"loss": 2.9474,
"theoretical_loss": 3.658376635599545,
"tokens_seen": 973996032
},
{
"epoch": 0.3,
"learning_rate": 0.00035599799398194584,
"loss": 3.0066,
"theoretical_loss": 3.658353096450255,
"tokens_seen": 974061568
},
{
"epoch": 0.3,
"learning_rate": 0.000355987963891675,
"loss": 2.8698,
"theoretical_loss": 3.6583295593280667,
"tokens_seen": 974127104
},
{
"epoch": 0.3,
"learning_rate": 0.0003559779338014042,
"loss": 2.804,
"theoretical_loss": 3.658306024232669,
"tokens_seen": 974192640
},
{
"epoch": 0.3,
"learning_rate": 0.00035596790371113344,
"loss": 2.8447,
"theoretical_loss": 3.6582824911637513,
"tokens_seen": 974258176
},
{
"epoch": 0.3,
"learning_rate": 0.00035595787362086256,
"loss": 2.9167,
"theoretical_loss": 3.658258960121003,
"tokens_seen": 974323712
},
{
"epoch": 0.3,
"learning_rate": 0.0003559478435305918,
"loss": 2.6992,
"theoretical_loss": 3.6582354311041128,
"tokens_seen": 974389248
},
{
"epoch": 0.3,
"learning_rate": 0.0003559378134403209,
"loss": 2.8208,
"theoretical_loss": 3.658211904112771,
"tokens_seen": 974454784
},
{
"epoch": 0.3,
"learning_rate": 0.00035592778335005016,
"loss": 2.8196,
"theoretical_loss": 3.658188379146666,
"tokens_seen": 974520320
},
{
"epoch": 0.3,
"learning_rate": 0.00035591775325977934,
"loss": 2.8196,
"theoretical_loss": 3.658164856205488,
"tokens_seen": 974585856
},
{
"epoch": 0.3,
"learning_rate": 0.0003559077231695085,
"loss": 2.8959,
"theoretical_loss": 3.658141335288926,
"tokens_seen": 974651392
},
{
"epoch": 0.3,
"learning_rate": 0.0003558976930792377,
"loss": 2.5979,
"theoretical_loss": 3.6581178163966706,
"tokens_seen": 974716928
},
{
"epoch": 0.3,
"learning_rate": 0.0003558876629889669,
"loss": 2.7449,
"theoretical_loss": 3.6580942995284103,
"tokens_seen": 974782464
},
{
"epoch": 0.3,
"objective/train/docs_used": 560118,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9865477085113525,
"objective/train/theoretical_loss": 3.6580707846838356,
"objective/train/tokens_used": 995308000,
"theoretical_loss": 3.6580707846838356,
"tokens_seen": 974848000
},
{
"epoch": 0.3,
"learning_rate": 0.00035587763289869607,
"loss": 2.8332,
"theoretical_loss": 3.6580707846838356,
"tokens_seen": 974848000
},
{
"epoch": 0.3,
"learning_rate": 0.0003558676028084253,
"loss": 2.767,
"theoretical_loss": 3.6580472718626353,
"tokens_seen": 974913536
},
{
"epoch": 0.3,
"learning_rate": 0.00035585757271815443,
"loss": 2.7179,
"theoretical_loss": 3.6580237610645012,
"tokens_seen": 974979072
},
{
"epoch": 0.3,
"learning_rate": 0.00035584754262788366,
"loss": 2.7694,
"theoretical_loss": 3.6580002522891215,
"tokens_seen": 975044608
},
{
"epoch": 0.3,
"learning_rate": 0.0003558375125376129,
"loss": 2.6339,
"theoretical_loss": 3.657976745536187,
"tokens_seen": 975110144
},
{
"epoch": 0.3,
"learning_rate": 0.000355827482447342,
"loss": 2.7374,
"theoretical_loss": 3.657953240805388,
"tokens_seen": 975175680
},
{
"epoch": 0.3,
"learning_rate": 0.00035581745235707126,
"loss": 2.7385,
"theoretical_loss": 3.6579297380964144,
"tokens_seen": 975241216
},
{
"epoch": 0.3,
"learning_rate": 0.0003558074222668004,
"loss": 2.7796,
"theoretical_loss": 3.657906237408956,
"tokens_seen": 975306752
},
{
"epoch": 0.3,
"learning_rate": 0.0003557973921765296,
"loss": 2.7574,
"theoretical_loss": 3.6578827387427038,
"tokens_seen": 975372288
},
{
"epoch": 0.3,
"learning_rate": 0.0003557873620862588,
"loss": 2.6918,
"theoretical_loss": 3.6578592420973477,
"tokens_seen": 975437824
},
{
"epoch": 0.3,
"learning_rate": 0.000355777331995988,
"loss": 2.7729,
"theoretical_loss": 3.6578357474725784,
"tokens_seen": 975503360
},
{
"epoch": 0.3,
"learning_rate": 0.00035576730190571717,
"loss": 2.4314,
"theoretical_loss": 3.6578122548680865,
"tokens_seen": 975568896
},
{
"epoch": 0.3,
"learning_rate": 0.00035575727181544635,
"loss": 2.6777,
"theoretical_loss": 3.657788764283562,
"tokens_seen": 975634432
},
{
"epoch": 0.3,
"learning_rate": 0.00035574724172517553,
"loss": 2.7037,
"theoretical_loss": 3.6577652757186963,
"tokens_seen": 975699968
},
{
"epoch": 0.3,
"learning_rate": 0.00035573721163490476,
"loss": 2.4771,
"theoretical_loss": 3.65774178917318,
"tokens_seen": 975765504
},
{
"epoch": 0.3,
"learning_rate": 0.0003557271815446339,
"loss": 2.9538,
"theoretical_loss": 3.657718304646703,
"tokens_seen": 975831040
},
{
"epoch": 0.3,
"learning_rate": 0.0003557171514543631,
"loss": 2.7745,
"theoretical_loss": 3.657694822138957,
"tokens_seen": 975896576
},
{
"epoch": 0.3,
"learning_rate": 0.00035570712136409225,
"loss": 2.7002,
"theoretical_loss": 3.657671341649633,
"tokens_seen": 975962112
},
{
"epoch": 0.3,
"learning_rate": 0.0003556970912738215,
"loss": 2.704,
"theoretical_loss": 3.657647863178422,
"tokens_seen": 976027648
},
{
"epoch": 0.3,
"learning_rate": 0.00035568706118355067,
"loss": 2.631,
"theoretical_loss": 3.657624386725014,
"tokens_seen": 976093184
},
{
"epoch": 0.3,
"learning_rate": 0.00035567703109327985,
"loss": 2.6218,
"theoretical_loss": 3.6576009122891016,
"tokens_seen": 976158720
},
{
"epoch": 0.3,
"learning_rate": 0.00035566700100300903,
"loss": 2.8102,
"theoretical_loss": 3.657577439870375,
"tokens_seen": 976224256
},
{
"epoch": 0.3,
"learning_rate": 0.00035565697091273827,
"loss": 2.6767,
"theoretical_loss": 3.657553969468526,
"tokens_seen": 976289792
},
{
"epoch": 0.3,
"learning_rate": 0.0003556469408224674,
"loss": 2.7494,
"theoretical_loss": 3.6575305010832455,
"tokens_seen": 976355328
},
{
"epoch": 0.3,
"learning_rate": 0.00035563691073219663,
"loss": 2.7781,
"theoretical_loss": 3.657507034714225,
"tokens_seen": 976420864
},
{
"epoch": 0.3,
"objective/train/docs_used": 560789,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0597338676452637,
"objective/train/theoretical_loss": 3.6574835703611566,
"objective/train/tokens_used": 996946400,
"theoretical_loss": 3.6574835703611566,
"tokens_seen": 976486400
},
{
"epoch": 0.3,
"learning_rate": 0.00035562688064192576,
"loss": 2.9362,
"theoretical_loss": 3.6574835703611566,
"tokens_seen": 976486400
},
{
"epoch": 0.3,
"learning_rate": 0.000355616850551655,
"loss": 2.771,
"theoretical_loss": 3.657460108023731,
"tokens_seen": 976551936
},
{
"epoch": 0.3,
"learning_rate": 0.00035560682046138417,
"loss": 2.7002,
"theoretical_loss": 3.6574366477016396,
"tokens_seen": 976617472
},
{
"epoch": 0.3,
"learning_rate": 0.00035559679037111335,
"loss": 2.7904,
"theoretical_loss": 3.6574131893945747,
"tokens_seen": 976683008
},
{
"epoch": 0.3,
"learning_rate": 0.00035558676028084253,
"loss": 2.7346,
"theoretical_loss": 3.657389733102228,
"tokens_seen": 976748544
},
{
"epoch": 0.3,
"learning_rate": 0.0003555767301905717,
"loss": 2.8859,
"theoretical_loss": 3.6573662788242913,
"tokens_seen": 976814080
},
{
"epoch": 0.3,
"learning_rate": 0.0003555667001003009,
"loss": 2.7506,
"theoretical_loss": 3.6573428265604564,
"tokens_seen": 976879616
},
{
"epoch": 0.3,
"learning_rate": 0.00035555667001003013,
"loss": 2.711,
"theoretical_loss": 3.657319376310415,
"tokens_seen": 976945152
},
{
"epoch": 0.3,
"learning_rate": 0.00035554663991975926,
"loss": 2.9277,
"theoretical_loss": 3.6572959280738595,
"tokens_seen": 977010688
},
{
"epoch": 0.3,
"learning_rate": 0.0003555366098294885,
"loss": 2.7666,
"theoretical_loss": 3.6572724818504816,
"tokens_seen": 977076224
},
{
"epoch": 0.3,
"learning_rate": 0.0003555265797392176,
"loss": 2.5392,
"theoretical_loss": 3.657249037639974,
"tokens_seen": 977141760
},
{
"epoch": 0.3,
"learning_rate": 0.00035551654964894686,
"loss": 2.6823,
"theoretical_loss": 3.6572255954420276,
"tokens_seen": 977207296
},
{
"epoch": 0.3,
"learning_rate": 0.00035550651955867604,
"loss": 2.8229,
"theoretical_loss": 3.6572021552563365,
"tokens_seen": 977272832
},
{
"epoch": 0.3,
"learning_rate": 0.0003554964894684052,
"loss": 2.7096,
"theoretical_loss": 3.6571787170825925,
"tokens_seen": 977338368
},
{
"epoch": 0.3,
"learning_rate": 0.0003554864593781344,
"loss": 2.5198,
"theoretical_loss": 3.657155280920487,
"tokens_seen": 977403904
},
{
"epoch": 0.3,
"learning_rate": 0.00035547642928786364,
"loss": 2.9726,
"theoretical_loss": 3.657131846769713,
"tokens_seen": 977469440
},
{
"epoch": 0.3,
"learning_rate": 0.00035546639919759276,
"loss": 2.8303,
"theoretical_loss": 3.6571084146299633,
"tokens_seen": 977534976
},
{
"epoch": 0.3,
"learning_rate": 0.000355456369107322,
"loss": 2.8391,
"theoretical_loss": 3.657084984500931,
"tokens_seen": 977600512
},
{
"epoch": 0.3,
"learning_rate": 0.0003554463390170511,
"loss": 2.7676,
"theoretical_loss": 3.6570615563823075,
"tokens_seen": 977666048
},
{
"epoch": 0.3,
"learning_rate": 0.00035543630892678036,
"loss": 2.706,
"theoretical_loss": 3.6570381302737864,
"tokens_seen": 977731584
},
{
"epoch": 0.3,
"learning_rate": 0.00035542627883650954,
"loss": 2.752,
"theoretical_loss": 3.657014706175061,
"tokens_seen": 977797120
},
{
"epoch": 0.3,
"learning_rate": 0.0003554162487462387,
"loss": 2.954,
"theoretical_loss": 3.656991284085823,
"tokens_seen": 977862656
},
{
"epoch": 0.3,
"learning_rate": 0.0003554062186559679,
"loss": 2.6019,
"theoretical_loss": 3.6569678640057655,
"tokens_seen": 977928192
},
{
"epoch": 0.3,
"learning_rate": 0.0003553961885656971,
"loss": 2.6752,
"theoretical_loss": 3.656944445934583,
"tokens_seen": 977993728
},
{
"epoch": 0.3,
"learning_rate": 0.00035538615847542627,
"loss": 2.8835,
"theoretical_loss": 3.6569210298719668,
"tokens_seen": 978059264
},
{
"epoch": 0.3,
"objective/train/docs_used": 561934,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.036074638366699,
"objective/train/theoretical_loss": 3.656897615817611,
"objective/train/tokens_used": 998584800,
"theoretical_loss": 3.656897615817611,
"tokens_seen": 978124800
},
{
"epoch": 0.3,
"learning_rate": 0.0003553761283851555,
"loss": 2.8248,
"theoretical_loss": 3.656897615817611,
"tokens_seen": 978124800
},
{
"epoch": 0.3,
"learning_rate": 0.00035536609829488463,
"loss": 2.8754,
"theoretical_loss": 3.656874203771208,
"tokens_seen": 978190336
},
{
"epoch": 0.3,
"learning_rate": 0.00035535606820461386,
"loss": 2.7397,
"theoretical_loss": 3.6568507937324526,
"tokens_seen": 978255872
},
{
"epoch": 0.3,
"learning_rate": 0.00035534603811434304,
"loss": 2.6859,
"theoretical_loss": 3.6568273857010363,
"tokens_seen": 978321408
},
{
"epoch": 0.3,
"learning_rate": 0.0003553360080240722,
"loss": 2.8051,
"theoretical_loss": 3.6568039796766545,
"tokens_seen": 978386944
},
{
"epoch": 0.3,
"learning_rate": 0.0003553259779338014,
"loss": 2.731,
"theoretical_loss": 3.656780575658999,
"tokens_seen": 978452480
},
{
"epoch": 0.3,
"learning_rate": 0.0003553159478435306,
"loss": 3.0446,
"theoretical_loss": 3.6567571736477644,
"tokens_seen": 978518016
},
{
"epoch": 0.3,
"learning_rate": 0.00035530591775325977,
"loss": 2.5783,
"theoretical_loss": 3.6567337736426433,
"tokens_seen": 978583552
},
{
"epoch": 0.3,
"learning_rate": 0.000355295887662989,
"loss": 2.7872,
"theoretical_loss": 3.656710375643331,
"tokens_seen": 978649088
},
{
"epoch": 0.3,
"learning_rate": 0.00035528585757271813,
"loss": 2.8888,
"theoretical_loss": 3.6566869796495194,
"tokens_seen": 978714624
},
{
"epoch": 0.3,
"learning_rate": 0.00035527582748244737,
"loss": 2.7153,
"theoretical_loss": 3.6566635856609038,
"tokens_seen": 978780160
},
{
"epoch": 0.3,
"learning_rate": 0.0003552657973921765,
"loss": 2.835,
"theoretical_loss": 3.6566401936771773,
"tokens_seen": 978845696
},
{
"epoch": 0.3,
"learning_rate": 0.00035525576730190573,
"loss": 2.7991,
"theoretical_loss": 3.6566168036980344,
"tokens_seen": 978911232
},
{
"epoch": 0.3,
"learning_rate": 0.0003552457372116349,
"loss": 2.7741,
"theoretical_loss": 3.6565934157231688,
"tokens_seen": 978976768
},
{
"epoch": 0.3,
"learning_rate": 0.0003552357071213641,
"loss": 2.9521,
"theoretical_loss": 3.6565700297522743,
"tokens_seen": 979042304
},
{
"epoch": 0.3,
"learning_rate": 0.00035522567703109327,
"loss": 2.6532,
"theoretical_loss": 3.6565466457850455,
"tokens_seen": 979107840
},
{
"epoch": 0.3,
"learning_rate": 0.00035521564694082245,
"loss": 2.6748,
"theoretical_loss": 3.656523263821177,
"tokens_seen": 979173376
},
{
"epoch": 0.3,
"learning_rate": 0.00035520561685055163,
"loss": 2.9522,
"theoretical_loss": 3.656499883860362,
"tokens_seen": 979238912
},
{
"epoch": 0.3,
"learning_rate": 0.00035519558676028087,
"loss": 2.7171,
"theoretical_loss": 3.6564765059022957,
"tokens_seen": 979304448
},
{
"epoch": 0.3,
"learning_rate": 0.00035518555667001,
"loss": 2.9185,
"theoretical_loss": 3.6564531299466725,
"tokens_seen": 979369984
},
{
"epoch": 0.3,
"learning_rate": 0.00035517552657973923,
"loss": 2.7674,
"theoretical_loss": 3.6564297559931864,
"tokens_seen": 979435520
},
{
"epoch": 0.3,
"learning_rate": 0.0003551654964894684,
"loss": 2.6529,
"theoretical_loss": 3.656406384041533,
"tokens_seen": 979501056
},
{
"epoch": 0.3,
"learning_rate": 0.0003551554663991976,
"loss": 2.7675,
"theoretical_loss": 3.6563830140914058,
"tokens_seen": 979566592
},
{
"epoch": 0.3,
"learning_rate": 0.0003551454363089268,
"loss": 2.9393,
"theoretical_loss": 3.6563596461424996,
"tokens_seen": 979632128
},
{
"epoch": 0.3,
"learning_rate": 0.00035513540621865596,
"loss": 2.828,
"theoretical_loss": 3.65633628019451,
"tokens_seen": 979697664
},
{
"epoch": 0.3,
"objective/train/docs_used": 562605,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.608224630355835,
"objective/train/theoretical_loss": 3.6563129162471313,
"objective/train/tokens_used": 1000223200,
"theoretical_loss": 3.6563129162471313,
"tokens_seen": 979763200
},
{
"epoch": 0.3,
"learning_rate": 0.00035512537612838514,
"loss": 2.796,
"theoretical_loss": 3.6563129162471313,
"tokens_seen": 979763200
},
{
"epoch": 0.3,
"learning_rate": 0.00035511534603811437,
"loss": 2.6893,
"theoretical_loss": 3.6562895543000584,
"tokens_seen": 979828736
},
{
"epoch": 0.3,
"learning_rate": 0.0003551053159478435,
"loss": 2.851,
"theoretical_loss": 3.656266194352986,
"tokens_seen": 979894272
},
{
"epoch": 0.3,
"learning_rate": 0.00035509528585757273,
"loss": 2.5674,
"theoretical_loss": 3.65624283640561,
"tokens_seen": 979959808
},
{
"epoch": 0.3,
"learning_rate": 0.0003550852557673019,
"loss": 2.6627,
"theoretical_loss": 3.6562194804576245,
"tokens_seen": 980025344
},
{
"epoch": 0.3,
"learning_rate": 0.0003550752256770311,
"loss": 2.6832,
"theoretical_loss": 3.6561961265087253,
"tokens_seen": 980090880
},
{
"epoch": 0.3,
"learning_rate": 0.00035506519558676033,
"loss": 2.7568,
"theoretical_loss": 3.6561727745586072,
"tokens_seen": 980156416
},
{
"epoch": 0.3,
"learning_rate": 0.00035505516549648946,
"loss": 2.5421,
"theoretical_loss": 3.6561494246069666,
"tokens_seen": 980221952
},
{
"epoch": 0.3,
"learning_rate": 0.0003550451354062187,
"loss": 2.5529,
"theoretical_loss": 3.6561260766534973,
"tokens_seen": 980287488
},
{
"epoch": 0.3,
"learning_rate": 0.0003550351053159478,
"loss": 2.7084,
"theoretical_loss": 3.6561027306978957,
"tokens_seen": 980353024
},
{
"epoch": 0.3,
"learning_rate": 0.00035502507522567706,
"loss": 2.9732,
"theoretical_loss": 3.656079386739857,
"tokens_seen": 980418560
},
{
"epoch": 0.3,
"learning_rate": 0.00035501504513540624,
"loss": 2.8606,
"theoretical_loss": 3.656056044779077,
"tokens_seen": 980484096
},
{
"epoch": 0.3,
"learning_rate": 0.0003550050150451354,
"loss": 2.7026,
"theoretical_loss": 3.656032704815251,
"tokens_seen": 980549632
},
{
"epoch": 0.3,
"learning_rate": 0.0003549949849548646,
"loss": 2.7225,
"theoretical_loss": 3.656009366848075,
"tokens_seen": 980615168
},
{
"epoch": 0.3,
"learning_rate": 0.00035498495486459384,
"loss": 2.8108,
"theoretical_loss": 3.6559860308772447,
"tokens_seen": 980680704
},
{
"epoch": 0.3,
"learning_rate": 0.00035497492477432296,
"loss": 2.5544,
"theoretical_loss": 3.6559626969024563,
"tokens_seen": 980746240
},
{
"epoch": 0.3,
"learning_rate": 0.0003549648946840522,
"loss": 2.7182,
"theoretical_loss": 3.655939364923405,
"tokens_seen": 980811776
},
{
"epoch": 0.3,
"learning_rate": 0.0003549548645937813,
"loss": 2.8927,
"theoretical_loss": 3.6559160349397866,
"tokens_seen": 980877312
},
{
"epoch": 0.3,
"learning_rate": 0.00035494483450351056,
"loss": 2.8708,
"theoretical_loss": 3.655892706951298,
"tokens_seen": 980942848
},
{
"epoch": 0.3,
"learning_rate": 0.00035493480441323974,
"loss": 2.8085,
"theoretical_loss": 3.6558693809576353,
"tokens_seen": 981008384
},
{
"epoch": 0.3,
"learning_rate": 0.0003549247743229689,
"loss": 2.7466,
"theoretical_loss": 3.6558460569584943,
"tokens_seen": 981073920
},
{
"epoch": 0.3,
"learning_rate": 0.0003549147442326981,
"loss": 2.6276,
"theoretical_loss": 3.6558227349535706,
"tokens_seen": 981139456
},
{
"epoch": 0.3,
"learning_rate": 0.0003549047141424273,
"loss": 2.8119,
"theoretical_loss": 3.655799414942561,
"tokens_seen": 981204992
},
{
"epoch": 0.3,
"learning_rate": 0.00035489468405215647,
"loss": 2.8536,
"theoretical_loss": 3.655776096925163,
"tokens_seen": 981270528
},
{
"epoch": 0.3,
"learning_rate": 0.0003548846539618857,
"loss": 2.7093,
"theoretical_loss": 3.6557527809010715,
"tokens_seen": 981336064
},
{
"epoch": 0.3,
"objective/train/docs_used": 563033,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6633009910583496,
"objective/train/theoretical_loss": 3.6557294668699836,
"objective/train/tokens_used": 1001861600,
"theoretical_loss": 3.6557294668699836,
"tokens_seen": 981401600
},
{
"epoch": 0.3,
"learning_rate": 0.00035487462387161483,
"loss": 2.5274,
"theoretical_loss": 3.6557294668699836,
"tokens_seen": 981401600
},
{
"epoch": 0.3,
"learning_rate": 0.00035486459378134406,
"loss": 3.0037,
"theoretical_loss": 3.655706154831596,
"tokens_seen": 981467136
},
{
"epoch": 0.3,
"learning_rate": 0.00035485456369107324,
"loss": 2.6504,
"theoretical_loss": 3.6556828447856047,
"tokens_seen": 981532672
},
{
"epoch": 0.3,
"learning_rate": 0.0003548445336008024,
"loss": 2.5678,
"theoretical_loss": 3.6556595367317075,
"tokens_seen": 981598208
},
{
"epoch": 0.3,
"learning_rate": 0.0003548345035105316,
"loss": 2.8837,
"theoretical_loss": 3.6556362306696,
"tokens_seen": 981663744
},
{
"epoch": 0.3,
"learning_rate": 0.0003548244734202608,
"loss": 2.8442,
"theoretical_loss": 3.6556129265989803,
"tokens_seen": 981729280
},
{
"epoch": 0.3,
"learning_rate": 0.00035481444332998997,
"loss": 2.8878,
"theoretical_loss": 3.6555896245195436,
"tokens_seen": 981794816
},
{
"epoch": 0.3,
"learning_rate": 0.0003548044132397192,
"loss": 2.8574,
"theoretical_loss": 3.655566324430989,
"tokens_seen": 981860352
},
{
"epoch": 0.3,
"learning_rate": 0.00035479438314944833,
"loss": 2.5928,
"theoretical_loss": 3.6555430263330115,
"tokens_seen": 981925888
},
{
"epoch": 0.3,
"learning_rate": 0.00035478435305917757,
"loss": 2.7681,
"theoretical_loss": 3.6555197302253095,
"tokens_seen": 981991424
},
{
"epoch": 0.3,
"learning_rate": 0.0003547743229689067,
"loss": 2.7337,
"theoretical_loss": 3.6554964361075797,
"tokens_seen": 982056960
},
{
"epoch": 0.3,
"learning_rate": 0.00035476429287863593,
"loss": 2.9776,
"theoretical_loss": 3.6554731439795187,
"tokens_seen": 982122496
},
{
"epoch": 0.3,
"learning_rate": 0.0003547542627883651,
"loss": 2.8791,
"theoretical_loss": 3.6554498538408255,
"tokens_seen": 982188032
},
{
"epoch": 0.3,
"learning_rate": 0.0003547442326980943,
"loss": 2.6715,
"theoretical_loss": 3.655426565691196,
"tokens_seen": 982253568
},
{
"epoch": 0.3,
"learning_rate": 0.00035473420260782347,
"loss": 2.8106,
"theoretical_loss": 3.6554032795303284,
"tokens_seen": 982319104
},
{
"epoch": 0.3,
"learning_rate": 0.00035472417251755265,
"loss": 2.687,
"theoretical_loss": 3.6553799953579196,
"tokens_seen": 982384640
},
{
"epoch": 0.3,
"learning_rate": 0.00035471414242728183,
"loss": 2.7234,
"theoretical_loss": 3.655356713173667,
"tokens_seen": 982450176
},
{
"epoch": 0.3,
"learning_rate": 0.00035470411233701107,
"loss": 2.762,
"theoretical_loss": 3.6553334329772693,
"tokens_seen": 982515712
},
{
"epoch": 0.3,
"learning_rate": 0.0003546940822467402,
"loss": 2.8332,
"theoretical_loss": 3.6553101547684235,
"tokens_seen": 982581248
},
{
"epoch": 0.3,
"learning_rate": 0.00035468405215646943,
"loss": 3.0038,
"theoretical_loss": 3.6552868785468275,
"tokens_seen": 982646784
},
{
"epoch": 0.3,
"learning_rate": 0.0003546740220661986,
"loss": 2.7208,
"theoretical_loss": 3.6552636043121787,
"tokens_seen": 982712320
},
{
"epoch": 0.3,
"learning_rate": 0.0003546639919759278,
"loss": 2.7522,
"theoretical_loss": 3.655240332064176,
"tokens_seen": 982777856
},
{
"epoch": 0.3,
"learning_rate": 0.000354653961885657,
"loss": 2.7516,
"theoretical_loss": 3.6552170618025164,
"tokens_seen": 982843392
},
{
"epoch": 0.3,
"learning_rate": 0.00035464393179538616,
"loss": 2.7366,
"theoretical_loss": 3.655193793526898,
"tokens_seen": 982908928
},
{
"epoch": 0.3,
"learning_rate": 0.00035463390170511534,
"loss": 2.9126,
"theoretical_loss": 3.6551705272370194,
"tokens_seen": 982974464
},
{
"debugging/Self-BLEU-5": 0.4599865317671292,
"debugging/distinct-1-grams": 0.7753828778379399,
"debugging/distinct-2-grams": 0.9579668873607778,
"debugging/entropy-1-grams": 5.851307016100259,
"debugging/entropy-2-grams": 6.818568697998407,
"debugging/length": 480.5,
"debugging/num_segments": 14,
"epoch": 0.3,
"objective/train/docs_used": 564216,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.978569746017456,
"objective/train/theoretical_loss": 3.6551472629325787,
"objective/train/tokens_used": 1003500000,
"theoretical_loss": 3.6551472629325787,
"tokens_seen": 983040000
},
{
"epoch": 0.3,
"learning_rate": 0.00035462387161484457,
"loss": 2.7497,
"theoretical_loss": 3.6551472629325787,
"tokens_seen": 983040000
},
{
"epoch": 0.3,
"learning_rate": 0.0003546138415245737,
"loss": 2.7414,
"theoretical_loss": 3.6551240006132737,
"tokens_seen": 983105536
},
{
"epoch": 0.3,
"learning_rate": 0.00035460381143430293,
"loss": 2.6714,
"theoretical_loss": 3.655100740278803,
"tokens_seen": 983171072
},
{
"epoch": 0.3,
"learning_rate": 0.00035459378134403206,
"loss": 2.4484,
"theoretical_loss": 3.655077481928865,
"tokens_seen": 983236608
},
{
"epoch": 0.3,
"learning_rate": 0.0003545837512537613,
"loss": 2.6524,
"theoretical_loss": 3.655054225563158,
"tokens_seen": 983302144
},
{
"epoch": 0.3,
"learning_rate": 0.0003545737211634905,
"loss": 2.9037,
"theoretical_loss": 3.6550309711813807,
"tokens_seen": 983367680
},
{
"epoch": 0.3,
"learning_rate": 0.00035456369107321966,
"loss": 3.0571,
"theoretical_loss": 3.655007718783231,
"tokens_seen": 983433216
},
{
"epoch": 0.3,
"learning_rate": 0.00035455366098294884,
"loss": 2.8883,
"theoretical_loss": 3.6549844683684083,
"tokens_seen": 983498752
},
{
"epoch": 0.3,
"learning_rate": 0.000354543630892678,
"loss": 2.8876,
"theoretical_loss": 3.654961219936611,
"tokens_seen": 983564288
},
{
"epoch": 0.3,
"learning_rate": 0.0003545336008024072,
"loss": 2.7674,
"theoretical_loss": 3.6549379734875376,
"tokens_seen": 983629824
},
{
"epoch": 0.3,
"learning_rate": 0.00035452357071213644,
"loss": 2.7639,
"theoretical_loss": 3.654914729020887,
"tokens_seen": 983695360
},
{
"epoch": 0.3,
"learning_rate": 0.00035451354062186556,
"loss": 2.4901,
"theoretical_loss": 3.654891486536359,
"tokens_seen": 983760896
},
{
"epoch": 0.3,
"learning_rate": 0.0003545035105315948,
"loss": 2.7361,
"theoretical_loss": 3.6548682460336512,
"tokens_seen": 983826432
},
{
"epoch": 0.3,
"learning_rate": 0.000354493480441324,
"loss": 2.6819,
"theoretical_loss": 3.6548450075124634,
"tokens_seen": 983891968
},
{
"epoch": 0.3,
"learning_rate": 0.00035448345035105316,
"loss": 2.7766,
"theoretical_loss": 3.6548217709724944,
"tokens_seen": 983957504
},
{
"epoch": 0.3,
"learning_rate": 0.00035448345035105316,
"loss": 2.9617,
"theoretical_loss": 3.6547985364134434,
"tokens_seen": 984023040
},
{
"epoch": 0.3,
"learning_rate": 0.00035447342026078234,
"loss": 2.9049,
"theoretical_loss": 3.6547753038350095,
"tokens_seen": 984088576
},
{
"epoch": 0.3,
"learning_rate": 0.0003544633901705115,
"loss": 2.735,
"theoretical_loss": 3.6547520732368923,
"tokens_seen": 984154112
},
{
"epoch": 0.3,
"learning_rate": 0.0003544533600802407,
"loss": 2.5976,
"theoretical_loss": 3.6547288446187913,
"tokens_seen": 984219648
},
{
"epoch": 0.3,
"learning_rate": 0.00035444332998996994,
"loss": 2.8728,
"theoretical_loss": 3.6547056179804054,
"tokens_seen": 984285184
},
{
"epoch": 0.3,
"learning_rate": 0.00035443329989969907,
"loss": 2.5684,
"theoretical_loss": 3.654682393321434,
"tokens_seen": 984350720
},
{
"epoch": 0.3,
"learning_rate": 0.0003544232698094283,
"loss": 2.7744,
"theoretical_loss": 3.654659170641577,
"tokens_seen": 984416256
},
{
"epoch": 0.3,
"learning_rate": 0.00035441323971915743,
"loss": 2.7469,
"theoretical_loss": 3.654635949940534,
"tokens_seen": 984481792
},
{
"epoch": 0.3,
"learning_rate": 0.00035440320962888667,
"loss": 2.9534,
"theoretical_loss": 3.6546127312180046,
"tokens_seen": 984547328
},
{
"epoch": 0.3,
"learning_rate": 0.00035439317953861585,
"loss": 2.7831,
"theoretical_loss": 3.654589514473688,
"tokens_seen": 984612864
},
{
"epoch": 0.3,
"objective/train/docs_used": 564602,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5193448066711426,
"objective/train/theoretical_loss": 3.654566299707285,
"objective/train/tokens_used": 1005138400,
"theoretical_loss": 3.654566299707285,
"tokens_seen": 984678400
},
{
"epoch": 0.3,
"learning_rate": 0.00035438314944834503,
"loss": 2.692,
"theoretical_loss": 3.654566299707285,
"tokens_seen": 984678400
},
{
"epoch": 0.3,
"learning_rate": 0.0003543731193580742,
"loss": 2.7013,
"theoretical_loss": 3.6545430869184954,
"tokens_seen": 984743936
},
{
"epoch": 0.3,
"learning_rate": 0.00035436308926780344,
"loss": 3.0013,
"theoretical_loss": 3.6545198761070186,
"tokens_seen": 984809472
},
{
"epoch": 0.3,
"learning_rate": 0.0003543530591775326,
"loss": 2.9815,
"theoretical_loss": 3.6544966672725545,
"tokens_seen": 984875008
},
{
"epoch": 0.3,
"learning_rate": 0.0003543430290872618,
"loss": 2.9508,
"theoretical_loss": 3.654473460414803,
"tokens_seen": 984940544
},
{
"epoch": 0.3,
"learning_rate": 0.000354332998996991,
"loss": 2.8524,
"theoretical_loss": 3.6544502555334653,
"tokens_seen": 985006080
},
{
"epoch": 0.3,
"learning_rate": 0.00035432296890672017,
"loss": 2.7937,
"theoretical_loss": 3.6544270526282405,
"tokens_seen": 985071616
},
{
"epoch": 0.3,
"learning_rate": 0.0003543129388164494,
"loss": 2.7983,
"theoretical_loss": 3.6544038516988295,
"tokens_seen": 985137152
},
{
"epoch": 0.3,
"learning_rate": 0.00035430290872617853,
"loss": 2.724,
"theoretical_loss": 3.6543806527449325,
"tokens_seen": 985202688
},
{
"epoch": 0.3,
"learning_rate": 0.00035429287863590777,
"loss": 2.8615,
"theoretical_loss": 3.6543574557662497,
"tokens_seen": 985268224
},
{
"epoch": 0.3,
"learning_rate": 0.0003542828485456369,
"loss": 2.8071,
"theoretical_loss": 3.6543342607624814,
"tokens_seen": 985333760
},
{
"epoch": 0.3,
"learning_rate": 0.00035427281845536613,
"loss": 2.6446,
"theoretical_loss": 3.654311067733329,
"tokens_seen": 985399296
},
{
"epoch": 0.3,
"learning_rate": 0.0003542627883650953,
"loss": 2.8115,
"theoretical_loss": 3.654287876678492,
"tokens_seen": 985464832
},
{
"epoch": 0.3,
"learning_rate": 0.0003542527582748245,
"loss": 2.6024,
"theoretical_loss": 3.6542646875976716,
"tokens_seen": 985530368
},
{
"epoch": 0.3,
"learning_rate": 0.00035424272818455367,
"loss": 2.61,
"theoretical_loss": 3.654241500490569,
"tokens_seen": 985595904
},
{
"epoch": 0.3,
"learning_rate": 0.00035423269809428285,
"loss": 2.7067,
"theoretical_loss": 3.654218315356884,
"tokens_seen": 985661440
},
{
"epoch": 0.3,
"learning_rate": 0.00035422266800401203,
"loss": 2.9262,
"theoretical_loss": 3.6541951321963175,
"tokens_seen": 985726976
},
{
"epoch": 0.3,
"learning_rate": 0.00035421263791374127,
"loss": 2.8278,
"theoretical_loss": 3.6541719510085713,
"tokens_seen": 985792512
},
{
"epoch": 0.3,
"learning_rate": 0.0003542026078234704,
"loss": 2.6213,
"theoretical_loss": 3.6541487717933463,
"tokens_seen": 985858048
},
{
"epoch": 0.3,
"learning_rate": 0.00035419257773319963,
"loss": 2.9176,
"theoretical_loss": 3.654125594550343,
"tokens_seen": 985923584
},
{
"epoch": 0.3,
"learning_rate": 0.0003541825476429288,
"loss": 2.8411,
"theoretical_loss": 3.654102419279262,
"tokens_seen": 985989120
},
{
"epoch": 0.3,
"learning_rate": 0.000354172517552658,
"loss": 2.8451,
"theoretical_loss": 3.6540792459798057,
"tokens_seen": 986054656
},
{
"epoch": 0.3,
"learning_rate": 0.0003541624874623872,
"loss": 2.6535,
"theoretical_loss": 3.6540560746516744,
"tokens_seen": 986120192
},
{
"epoch": 0.3,
"learning_rate": 0.00035415245737211636,
"loss": 3.0508,
"theoretical_loss": 3.65403290529457,
"tokens_seen": 986185728
},
{
"epoch": 0.3,
"learning_rate": 0.00035414242728184554,
"loss": 2.8576,
"theoretical_loss": 3.6540097379081935,
"tokens_seen": 986251264
},
{
"epoch": 0.3,
"objective/train/docs_used": 565772,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8129312992095947,
"objective/train/theoretical_loss": 3.653986572492247,
"objective/train/tokens_used": 1006776800,
"theoretical_loss": 3.653986572492247,
"tokens_seen": 986316800
},
{
"epoch": 0.3,
"learning_rate": 0.00035413239719157477,
"loss": 2.9933,
"theoretical_loss": 3.653986572492247,
"tokens_seen": 986316800
},
{
"epoch": 0.3,
"learning_rate": 0.0003541223671013039,
"loss": 2.6191,
"theoretical_loss": 3.653963409046431,
"tokens_seen": 986382336
},
{
"epoch": 0.3,
"learning_rate": 0.00035411233701103314,
"loss": 2.8362,
"theoretical_loss": 3.653940247570448,
"tokens_seen": 986447872
},
{
"epoch": 0.3,
"learning_rate": 0.00035410230692076226,
"loss": 2.7837,
"theoretical_loss": 3.6539170880639986,
"tokens_seen": 986513408
},
{
"epoch": 0.3,
"learning_rate": 0.0003540922768304915,
"loss": 2.8254,
"theoretical_loss": 3.6538939305267855,
"tokens_seen": 986578944
},
{
"epoch": 0.3,
"learning_rate": 0.0003540822467402207,
"loss": 2.5447,
"theoretical_loss": 3.65387077495851,
"tokens_seen": 986644480
},
{
"epoch": 0.3,
"learning_rate": 0.00035407221664994986,
"loss": 2.9583,
"theoretical_loss": 3.653847621358874,
"tokens_seen": 986710016
},
{
"epoch": 0.3,
"learning_rate": 0.00035406218655967904,
"loss": 2.9756,
"theoretical_loss": 3.6538244697275792,
"tokens_seen": 986775552
},
{
"epoch": 0.3,
"learning_rate": 0.0003540521564694082,
"loss": 2.5865,
"theoretical_loss": 3.653801320064328,
"tokens_seen": 986841088
},
{
"epoch": 0.3,
"learning_rate": 0.0003540421263791374,
"loss": 2.727,
"theoretical_loss": 3.6537781723688223,
"tokens_seen": 986906624
},
{
"epoch": 0.3,
"learning_rate": 0.00035403209628886664,
"loss": 2.9969,
"theoretical_loss": 3.6537550266407637,
"tokens_seen": 986972160
},
{
"epoch": 0.3,
"learning_rate": 0.00035402206619859577,
"loss": 2.6281,
"theoretical_loss": 3.6537318828798546,
"tokens_seen": 987037696
},
{
"epoch": 0.3,
"learning_rate": 0.000354012036108325,
"loss": 2.908,
"theoretical_loss": 3.6537087410857976,
"tokens_seen": 987103232
},
{
"epoch": 0.3,
"learning_rate": 0.0003540020060180542,
"loss": 2.8281,
"theoretical_loss": 3.6536856012582946,
"tokens_seen": 987168768
},
{
"epoch": 0.3,
"learning_rate": 0.00035399197592778336,
"loss": 2.8686,
"theoretical_loss": 3.653662463397048,
"tokens_seen": 987234304
},
{
"epoch": 0.3,
"learning_rate": 0.00035398194583751254,
"loss": 2.6817,
"theoretical_loss": 3.65363932750176,
"tokens_seen": 987299840
},
{
"epoch": 0.3,
"learning_rate": 0.0003539719157472417,
"loss": 2.9323,
"theoretical_loss": 3.653616193572134,
"tokens_seen": 987365376
},
{
"epoch": 0.3,
"learning_rate": 0.0003539618856569709,
"loss": 2.8283,
"theoretical_loss": 3.6535930616078716,
"tokens_seen": 987430912
},
{
"epoch": 0.3,
"learning_rate": 0.00035395185556670014,
"loss": 2.7609,
"theoretical_loss": 3.6535699316086756,
"tokens_seen": 987496448
},
{
"epoch": 0.3,
"learning_rate": 0.00035394182547642927,
"loss": 3.0859,
"theoretical_loss": 3.6535468035742484,
"tokens_seen": 987561984
},
{
"epoch": 0.3,
"learning_rate": 0.0003539317953861585,
"loss": 2.7932,
"theoretical_loss": 3.6535236775042934,
"tokens_seen": 987627520
},
{
"epoch": 0.3,
"learning_rate": 0.00035392176529588763,
"loss": 2.9264,
"theoretical_loss": 3.6535005533985134,
"tokens_seen": 987693056
},
{
"epoch": 0.3,
"learning_rate": 0.00035391173520561687,
"loss": 2.8345,
"theoretical_loss": 3.6534774312566105,
"tokens_seen": 987758592
},
{
"epoch": 0.3,
"learning_rate": 0.00035390170511534605,
"loss": 2.5918,
"theoretical_loss": 3.6534543110782884,
"tokens_seen": 987824128
},
{
"epoch": 0.3,
"learning_rate": 0.00035389167502507523,
"loss": 2.6456,
"theoretical_loss": 3.6534311928632497,
"tokens_seen": 987889664
},
{
"epoch": 0.3,
"objective/train/docs_used": 566398,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7780752182006836,
"objective/train/theoretical_loss": 3.653408076611197,
"objective/train/tokens_used": 1008415200,
"theoretical_loss": 3.653408076611197,
"tokens_seen": 987955200
},
{
"epoch": 0.3,
"learning_rate": 0.0003538816449348044,
"loss": 2.5945,
"theoretical_loss": 3.653408076611197,
"tokens_seen": 987955200
},
{
"epoch": 0.3,
"learning_rate": 0.00035387161484453364,
"loss": 2.7342,
"theoretical_loss": 3.6533849623218346,
"tokens_seen": 988020736
},
{
"epoch": 0.3,
"learning_rate": 0.00035386158475426277,
"loss": 2.9909,
"theoretical_loss": 3.653361849994865,
"tokens_seen": 988086272
},
{
"epoch": 0.3,
"learning_rate": 0.000353851554663992,
"loss": 2.7667,
"theoretical_loss": 3.6533387396299912,
"tokens_seen": 988151808
},
{
"epoch": 0.3,
"learning_rate": 0.00035384152457372113,
"loss": 2.8075,
"theoretical_loss": 3.653315631226917,
"tokens_seen": 988217344
},
{
"epoch": 0.3,
"learning_rate": 0.00035383149448345037,
"loss": 2.6409,
"theoretical_loss": 3.6532925247853454,
"tokens_seen": 988282880
},
{
"epoch": 0.3,
"learning_rate": 0.00035382146439317955,
"loss": 2.8614,
"theoretical_loss": 3.65326942030498,
"tokens_seen": 988348416
},
{
"epoch": 0.3,
"learning_rate": 0.00035381143430290873,
"loss": 2.8925,
"theoretical_loss": 3.6532463177855248,
"tokens_seen": 988413952
},
{
"epoch": 0.3,
"learning_rate": 0.0003538014042126379,
"loss": 2.7223,
"theoretical_loss": 3.6532232172266825,
"tokens_seen": 988479488
},
{
"epoch": 0.3,
"learning_rate": 0.0003537913741223671,
"loss": 2.6875,
"theoretical_loss": 3.653200118628157,
"tokens_seen": 988545024
},
{
"epoch": 0.3,
"learning_rate": 0.0003537813440320963,
"loss": 2.6271,
"theoretical_loss": 3.653177021989653,
"tokens_seen": 988610560
},
{
"epoch": 0.3,
"learning_rate": 0.0003537713139418255,
"loss": 2.6394,
"theoretical_loss": 3.653153927310873,
"tokens_seen": 988676096
},
{
"epoch": 0.3,
"learning_rate": 0.00035376128385155464,
"loss": 2.9597,
"theoretical_loss": 3.6531308345915208,
"tokens_seen": 988741632
},
{
"epoch": 0.3,
"learning_rate": 0.00035375125376128387,
"loss": 2.7696,
"theoretical_loss": 3.653107743831301,
"tokens_seen": 988807168
},
{
"epoch": 0.3,
"learning_rate": 0.000353741223671013,
"loss": 2.5583,
"theoretical_loss": 3.6530846550299176,
"tokens_seen": 988872704
},
{
"epoch": 0.3,
"learning_rate": 0.00035373119358074223,
"loss": 2.7835,
"theoretical_loss": 3.6530615681870744,
"tokens_seen": 988938240
},
{
"epoch": 0.3,
"learning_rate": 0.0003537211634904714,
"loss": 2.6557,
"theoretical_loss": 3.653038483302475,
"tokens_seen": 989003776
},
{
"epoch": 0.3,
"learning_rate": 0.0003537111334002006,
"loss": 2.9332,
"theoretical_loss": 3.653015400375824,
"tokens_seen": 989069312
},
{
"epoch": 0.3,
"learning_rate": 0.0003537011033099298,
"loss": 2.8454,
"theoretical_loss": 3.652992319406826,
"tokens_seen": 989134848
},
{
"epoch": 0.3,
"learning_rate": 0.000353691073219659,
"loss": 2.733,
"theoretical_loss": 3.652969240395185,
"tokens_seen": 989200384
},
{
"epoch": 0.3,
"learning_rate": 0.00035368104312938814,
"loss": 2.6363,
"theoretical_loss": 3.652946163340605,
"tokens_seen": 989265920
},
{
"epoch": 0.3,
"learning_rate": 0.0003536710130391174,
"loss": 2.9971,
"theoretical_loss": 3.6529230882427903,
"tokens_seen": 989331456
},
{
"epoch": 0.3,
"learning_rate": 0.0003536609829488465,
"loss": 2.67,
"theoretical_loss": 3.652900015101446,
"tokens_seen": 989396992
},
{
"epoch": 0.3,
"learning_rate": 0.00035365095285857574,
"loss": 2.6334,
"theoretical_loss": 3.6528769439162767,
"tokens_seen": 989462528
},
{
"epoch": 0.3,
"learning_rate": 0.0003536409227683049,
"loss": 2.7806,
"theoretical_loss": 3.6528538746869863,
"tokens_seen": 989528064
},
{
"epoch": 0.3,
"objective/train/docs_used": 567574,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.520167112350464,
"objective/train/theoretical_loss": 3.65283080741328,
"objective/train/tokens_used": 1010053600,
"theoretical_loss": 3.65283080741328,
"tokens_seen": 989593600
},
{
"epoch": 0.3,
"learning_rate": 0.0003536308926780341,
"loss": 2.7074,
"theoretical_loss": 3.65283080741328,
"tokens_seen": 989593600
},
{
"epoch": 0.3,
"learning_rate": 0.0003536208625877633,
"loss": 2.8679,
"theoretical_loss": 3.652807742094862,
"tokens_seen": 989659136
},
{
"epoch": 0.3,
"learning_rate": 0.00035361083249749246,
"loss": 2.6061,
"theoretical_loss": 3.652784678731438,
"tokens_seen": 989724672
},
{
"epoch": 0.3,
"learning_rate": 0.0003536008024072217,
"loss": 2.6135,
"theoretical_loss": 3.652761617322712,
"tokens_seen": 989790208
},
{
"epoch": 0.3,
"learning_rate": 0.0003535907723169509,
"loss": 2.7568,
"theoretical_loss": 3.65273855786839,
"tokens_seen": 989855744
},
{
"epoch": 0.3,
"learning_rate": 0.00035358074222668006,
"loss": 2.8058,
"theoretical_loss": 3.652715500368175,
"tokens_seen": 989921280
},
{
"epoch": 0.3,
"learning_rate": 0.00035357071213640924,
"loss": 2.8859,
"theoretical_loss": 3.652692444821774,
"tokens_seen": 989986816
},
{
"epoch": 0.3,
"learning_rate": 0.0003535606820461384,
"loss": 2.7766,
"theoretical_loss": 3.6526693912288914,
"tokens_seen": 990052352
},
{
"epoch": 0.3,
"learning_rate": 0.0003535506519558676,
"loss": 2.7321,
"theoretical_loss": 3.652646339589232,
"tokens_seen": 990117888
},
{
"epoch": 0.3,
"learning_rate": 0.00035354062186559684,
"loss": 2.9107,
"theoretical_loss": 3.652623289902502,
"tokens_seen": 990183424
},
{
"epoch": 0.3,
"learning_rate": 0.00035353059177532597,
"loss": 2.9097,
"theoretical_loss": 3.6526002421684063,
"tokens_seen": 990248960
},
{
"epoch": 0.3,
"learning_rate": 0.0003535205616850552,
"loss": 2.9184,
"theoretical_loss": 3.6525771963866496,
"tokens_seen": 990314496
},
{
"epoch": 0.3,
"learning_rate": 0.0003535105315947844,
"loss": 2.779,
"theoretical_loss": 3.652554152556938,
"tokens_seen": 990380032
},
{
"epoch": 0.3,
"learning_rate": 0.00035350050150451356,
"loss": 2.9134,
"theoretical_loss": 3.652531110678977,
"tokens_seen": 990445568
},
{
"epoch": 0.3,
"learning_rate": 0.00035349047141424274,
"loss": 2.9167,
"theoretical_loss": 3.6525080707524715,
"tokens_seen": 990511104
},
{
"epoch": 0.3,
"learning_rate": 0.0003534804413239719,
"loss": 2.709,
"theoretical_loss": 3.652485032777128,
"tokens_seen": 990576640
},
{
"epoch": 0.3,
"learning_rate": 0.0003534704112337011,
"loss": 2.7339,
"theoretical_loss": 3.652461996752652,
"tokens_seen": 990642176
},
{
"epoch": 0.3,
"learning_rate": 0.00035346038114343034,
"loss": 2.9431,
"theoretical_loss": 3.6524389626787492,
"tokens_seen": 990707712
},
{
"epoch": 0.3,
"learning_rate": 0.00035345035105315947,
"loss": 2.611,
"theoretical_loss": 3.652415930555125,
"tokens_seen": 990773248
},
{
"epoch": 0.3,
"learning_rate": 0.0003534403209628887,
"loss": 2.7398,
"theoretical_loss": 3.6523929003814857,
"tokens_seen": 990838784
},
{
"epoch": 0.3,
"learning_rate": 0.00035343029087261783,
"loss": 2.7389,
"theoretical_loss": 3.652369872157537,
"tokens_seen": 990904320
},
{
"epoch": 0.3,
"learning_rate": 0.00035342026078234707,
"loss": 2.7379,
"theoretical_loss": 3.6523468458829855,
"tokens_seen": 990969856
},
{
"epoch": 0.3,
"learning_rate": 0.00035341023069207625,
"loss": 2.909,
"theoretical_loss": 3.6523238215575367,
"tokens_seen": 991035392
},
{
"epoch": 0.3,
"learning_rate": 0.00035340020060180543,
"loss": 2.8185,
"theoretical_loss": 3.6523007991808965,
"tokens_seen": 991100928
},
{
"epoch": 0.3,
"learning_rate": 0.0003533901705115346,
"loss": 2.7694,
"theoretical_loss": 3.6522777787527714,
"tokens_seen": 991166464
},
{
"epoch": 0.3,
"objective/train/docs_used": 568180,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.9635374546051025,
"objective/train/theoretical_loss": 3.6522547602728683,
"objective/train/tokens_used": 1011692000,
"theoretical_loss": 3.6522547602728683,
"tokens_seen": 991232000
},
{
"epoch": 0.3,
"learning_rate": 0.00035338014042126384,
"loss": 2.493,
"theoretical_loss": 3.6522547602728683,
"tokens_seen": 991232000
},
{
"epoch": 0.3,
"learning_rate": 0.00035337011033099297,
"loss": 2.5553,
"theoretical_loss": 3.652231743740893,
"tokens_seen": 991297536
},
{
"epoch": 0.3,
"learning_rate": 0.0003533600802407222,
"loss": 2.8313,
"theoretical_loss": 3.652208729156551,
"tokens_seen": 991363072
},
{
"epoch": 0.3,
"learning_rate": 0.00035335005015045133,
"loss": 2.718,
"theoretical_loss": 3.6521857165195506,
"tokens_seen": 991428608
},
{
"epoch": 0.3,
"learning_rate": 0.00035334002006018057,
"loss": 2.784,
"theoretical_loss": 3.652162705829597,
"tokens_seen": 991494144
},
{
"epoch": 0.3,
"learning_rate": 0.00035332998996990975,
"loss": 2.7913,
"theoretical_loss": 3.652139697086397,
"tokens_seen": 991559680
},
{
"epoch": 0.3,
"learning_rate": 0.00035331995987963893,
"loss": 2.7558,
"theoretical_loss": 3.652116690289658,
"tokens_seen": 991625216
},
{
"epoch": 0.3,
"learning_rate": 0.0003533099297893681,
"loss": 2.7334,
"theoretical_loss": 3.652093685439085,
"tokens_seen": 991690752
},
{
"epoch": 0.3,
"learning_rate": 0.0003532998996990973,
"loss": 2.6956,
"theoretical_loss": 3.6520706825343865,
"tokens_seen": 991756288
},
{
"epoch": 0.3,
"learning_rate": 0.0003532898696088265,
"loss": 2.6809,
"theoretical_loss": 3.6520476815752687,
"tokens_seen": 991821824
},
{
"epoch": 0.3,
"learning_rate": 0.0003532798395185557,
"loss": 2.8384,
"theoretical_loss": 3.6520246825614384,
"tokens_seen": 991887360
},
{
"epoch": 0.3,
"learning_rate": 0.00035326980942828484,
"loss": 2.7101,
"theoretical_loss": 3.652001685492603,
"tokens_seen": 991952896
},
{
"epoch": 0.3,
"learning_rate": 0.00035325977933801407,
"loss": 2.8614,
"theoretical_loss": 3.651978690368469,
"tokens_seen": 992018432
},
{
"epoch": 0.3,
"learning_rate": 0.0003532497492477432,
"loss": 2.7498,
"theoretical_loss": 3.6519556971887437,
"tokens_seen": 992083968
},
{
"epoch": 0.3,
"learning_rate": 0.00035323971915747243,
"loss": 2.8793,
"theoretical_loss": 3.651932705953134,
"tokens_seen": 992149504
},
{
"epoch": 0.3,
"learning_rate": 0.0003532296890672016,
"loss": 2.6344,
"theoretical_loss": 3.651909716661348,
"tokens_seen": 992215040
},
{
"epoch": 0.3,
"learning_rate": 0.0003532196589769308,
"loss": 2.8564,
"theoretical_loss": 3.651886729313092,
"tokens_seen": 992280576
},
{
"epoch": 0.3,
"learning_rate": 0.00035320962888666,
"loss": 2.8906,
"theoretical_loss": 3.6518637439080734,
"tokens_seen": 992346112
},
{
"epoch": 0.3,
"learning_rate": 0.0003531995987963892,
"loss": 2.8745,
"theoretical_loss": 3.651840760446,
"tokens_seen": 992411648
},
{
"epoch": 0.3,
"learning_rate": 0.00035318956870611834,
"loss": 2.7317,
"theoretical_loss": 3.6518177789265795,
"tokens_seen": 992477184
},
{
"epoch": 0.3,
"learning_rate": 0.0003531795386158476,
"loss": 2.695,
"theoretical_loss": 3.651794799349519,
"tokens_seen": 992542720
},
{
"epoch": 0.3,
"learning_rate": 0.0003531695085255767,
"loss": 2.7629,
"theoretical_loss": 3.651771821714526,
"tokens_seen": 992608256
},
{
"epoch": 0.3,
"learning_rate": 0.00035315947843530594,
"loss": 2.8136,
"theoretical_loss": 3.6517488460213086,
"tokens_seen": 992673792
},
{
"epoch": 0.3,
"learning_rate": 0.0003531494483450351,
"loss": 2.9297,
"theoretical_loss": 3.6517258722695747,
"tokens_seen": 992739328
},
{
"epoch": 0.3,
"learning_rate": 0.0003531394182547643,
"loss": 2.849,
"theoretical_loss": 3.651702900459031,
"tokens_seen": 992804864
},
{
"epoch": 0.3,
"objective/train/docs_used": 569645,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.724067211151123,
"objective/train/theoretical_loss": 3.6516799305893866,
"objective/train/tokens_used": 1013330400,
"theoretical_loss": 3.6516799305893866,
"tokens_seen": 992870400
},
{
"epoch": 0.3,
"learning_rate": 0.0003531293881644935,
"loss": 2.6757,
"theoretical_loss": 3.6516799305893866,
"tokens_seen": 992870400
},
{
"epoch": 0.3,
"learning_rate": 0.00035311935807422266,
"loss": 2.8368,
"theoretical_loss": 3.651656962660349,
"tokens_seen": 992935936
},
{
"epoch": 0.3,
"learning_rate": 0.00035310932798395184,
"loss": 2.568,
"theoretical_loss": 3.6516339966716256,
"tokens_seen": 993001472
},
{
"epoch": 0.3,
"learning_rate": 0.0003530992978936811,
"loss": 2.8462,
"theoretical_loss": 3.6516110326229247,
"tokens_seen": 993067008
},
{
"epoch": 0.3,
"learning_rate": 0.0003530892678034102,
"loss": 2.7974,
"theoretical_loss": 3.651588070513955,
"tokens_seen": 993132544
},
{
"epoch": 0.3,
"learning_rate": 0.00035307923771313944,
"loss": 2.7571,
"theoretical_loss": 3.651565110344424,
"tokens_seen": 993198080
},
{
"epoch": 0.3,
"learning_rate": 0.00035306920762286857,
"loss": 2.669,
"theoretical_loss": 3.6515421521140405,
"tokens_seen": 993263616
},
{
"epoch": 0.3,
"learning_rate": 0.0003530591775325978,
"loss": 2.9598,
"theoretical_loss": 3.6515191958225124,
"tokens_seen": 993329152
},
{
"epoch": 0.3,
"learning_rate": 0.000353049147442327,
"loss": 2.6423,
"theoretical_loss": 3.651496241469548,
"tokens_seen": 993394688
},
{
"epoch": 0.3,
"learning_rate": 0.00035303911735205617,
"loss": 2.8564,
"theoretical_loss": 3.6514732890548567,
"tokens_seen": 993460224
},
{
"epoch": 0.3,
"learning_rate": 0.00035302908726178535,
"loss": 2.9506,
"theoretical_loss": 3.6514503385781456,
"tokens_seen": 993525760
},
{
"epoch": 0.3,
"learning_rate": 0.0003530190571715146,
"loss": 2.8827,
"theoretical_loss": 3.6514273900391236,
"tokens_seen": 993591296
},
{
"epoch": 0.3,
"learning_rate": 0.0003530090270812437,
"loss": 2.7779,
"theoretical_loss": 3.6514044434374995,
"tokens_seen": 993656832
},
{
"epoch": 0.3,
"learning_rate": 0.00035299899699097294,
"loss": 2.7959,
"theoretical_loss": 3.6513814987729822,
"tokens_seen": 993722368
},
{
"epoch": 0.3,
"learning_rate": 0.00035298896690070207,
"loss": 2.8546,
"theoretical_loss": 3.65135855604528,
"tokens_seen": 993787904
},
{
"epoch": 0.3,
"learning_rate": 0.0003529789368104313,
"loss": 2.7358,
"theoretical_loss": 3.6513356152541023,
"tokens_seen": 993853440
},
{
"epoch": 0.3,
"learning_rate": 0.0003529689067201605,
"loss": 2.7453,
"theoretical_loss": 3.651312676399158,
"tokens_seen": 993918976
},
{
"epoch": 0.3,
"learning_rate": 0.00035295887662988967,
"loss": 2.957,
"theoretical_loss": 3.6512897394801547,
"tokens_seen": 993984512
},
{
"epoch": 0.3,
"learning_rate": 0.00035294884653961885,
"loss": 2.8961,
"theoretical_loss": 3.651266804496803,
"tokens_seen": 994050048
},
{
"epoch": 0.3,
"learning_rate": 0.00035293881644934803,
"loss": 2.9044,
"theoretical_loss": 3.651243871448811,
"tokens_seen": 994115584
},
{
"epoch": 0.3,
"learning_rate": 0.0003529287863590772,
"loss": 2.7187,
"theoretical_loss": 3.651220940335888,
"tokens_seen": 994181120
},
{
"epoch": 0.3,
"learning_rate": 0.00035291875626880645,
"loss": 2.8454,
"theoretical_loss": 3.651198011157743,
"tokens_seen": 994246656
},
{
"epoch": 0.3,
"learning_rate": 0.0003529087261785356,
"loss": 2.6628,
"theoretical_loss": 3.6511750839140866,
"tokens_seen": 994312192
},
{
"epoch": 0.3,
"learning_rate": 0.0003528986960882648,
"loss": 2.9214,
"theoretical_loss": 3.651152158604626,
"tokens_seen": 994377728
},
{
"epoch": 0.3,
"learning_rate": 0.00035288866599799394,
"loss": 2.8779,
"theoretical_loss": 3.651129235229072,
"tokens_seen": 994443264
},
{
"epoch": 0.3,
"objective/train/docs_used": 570064,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9148354530334473,
"objective/train/theoretical_loss": 3.6511063137871336,
"objective/train/tokens_used": 1014968800,
"theoretical_loss": 3.6511063137871336,
"tokens_seen": 994508800
},
{
"epoch": 0.3,
"learning_rate": 0.00035287863590772317,
"loss": 2.9757,
"theoretical_loss": 3.6511063137871336,
"tokens_seen": 994508800
},
{
"epoch": 0.3,
"learning_rate": 0.00035286860581745235,
"loss": 2.9214,
"theoretical_loss": 3.65108339427852,
"tokens_seen": 994574336
},
{
"epoch": 0.3,
"learning_rate": 0.00035285857572718153,
"loss": 2.7073,
"theoretical_loss": 3.6510604767029413,
"tokens_seen": 994639872
},
{
"epoch": 0.3,
"learning_rate": 0.00035284854563691077,
"loss": 3.113,
"theoretical_loss": 3.6510375610601065,
"tokens_seen": 994705408
},
{
"epoch": 0.3,
"learning_rate": 0.00035283851554663995,
"loss": 2.7075,
"theoretical_loss": 3.6510146473497262,
"tokens_seen": 994770944
},
{
"epoch": 0.3,
"learning_rate": 0.00035282848545636913,
"loss": 2.7323,
"theoretical_loss": 3.6509917355715094,
"tokens_seen": 994836480
},
{
"epoch": 0.3,
"learning_rate": 0.0003528184553660983,
"loss": 2.9125,
"theoretical_loss": 3.6509688257251662,
"tokens_seen": 994902016
},
{
"epoch": 0.3,
"learning_rate": 0.0003528084252758275,
"loss": 3.0109,
"theoretical_loss": 3.650945917810406,
"tokens_seen": 994967552
},
{
"epoch": 0.3,
"learning_rate": 0.0003527983951855567,
"loss": 2.5705,
"theoretical_loss": 3.650923011826939,
"tokens_seen": 995033088
},
{
"epoch": 0.3,
"learning_rate": 0.0003527883650952859,
"loss": 2.8056,
"theoretical_loss": 3.6509001077744756,
"tokens_seen": 995098624
},
{
"epoch": 0.3,
"learning_rate": 0.00035277833500501504,
"loss": 2.9234,
"theoretical_loss": 3.6508772056527263,
"tokens_seen": 995164160
},
{
"epoch": 0.3,
"learning_rate": 0.00035276830491474427,
"loss": 2.7495,
"theoretical_loss": 3.6508543054613996,
"tokens_seen": 995229696
},
{
"epoch": 0.3,
"learning_rate": 0.0003527582748244734,
"loss": 2.9357,
"theoretical_loss": 3.650831407200206,
"tokens_seen": 995295232
},
{
"epoch": 0.3,
"learning_rate": 0.00035274824473420263,
"loss": 2.693,
"theoretical_loss": 3.650808510868858,
"tokens_seen": 995360768
},
{
"epoch": 0.3,
"learning_rate": 0.0003527382146439318,
"loss": 2.8199,
"theoretical_loss": 3.650785616467063,
"tokens_seen": 995426304
},
{
"epoch": 0.3,
"learning_rate": 0.000352728184553661,
"loss": 2.7761,
"theoretical_loss": 3.6507627239945326,
"tokens_seen": 995491840
},
{
"epoch": 0.3,
"learning_rate": 0.0003527181544633902,
"loss": 2.851,
"theoretical_loss": 3.6507398334509773,
"tokens_seen": 995557376
},
{
"epoch": 0.3,
"learning_rate": 0.0003527081243731194,
"loss": 2.8116,
"theoretical_loss": 3.650716944836107,
"tokens_seen": 995622912
},
{
"epoch": 0.3,
"learning_rate": 0.00035269809428284854,
"loss": 2.8554,
"theoretical_loss": 3.650694058149634,
"tokens_seen": 995688448
},
{
"epoch": 0.3,
"learning_rate": 0.0003526880641925778,
"loss": 2.613,
"theoretical_loss": 3.6506711733912667,
"tokens_seen": 995753984
},
{
"epoch": 0.3,
"learning_rate": 0.0003526780341023069,
"loss": 2.7508,
"theoretical_loss": 3.650648290560717,
"tokens_seen": 995819520
},
{
"epoch": 0.3,
"learning_rate": 0.00035266800401203614,
"loss": 2.8561,
"theoretical_loss": 3.6506254096576956,
"tokens_seen": 995885056
},
{
"epoch": 0.3,
"learning_rate": 0.0003526579739217653,
"loss": 2.7976,
"theoretical_loss": 3.6506025306819128,
"tokens_seen": 995950592
},
{
"epoch": 0.3,
"learning_rate": 0.0003526479438314945,
"loss": 2.6007,
"theoretical_loss": 3.65057965363308,
"tokens_seen": 996016128
},
{
"epoch": 0.3,
"learning_rate": 0.0003526379137412237,
"loss": 2.8354,
"theoretical_loss": 3.650556778510908,
"tokens_seen": 996081664
},
{
"epoch": 0.3,
"objective/train/docs_used": 571551,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8192026615142822,
"objective/train/theoretical_loss": 3.6505339053151076,
"objective/train/tokens_used": 1016607200,
"theoretical_loss": 3.6505339053151076,
"tokens_seen": 996147200
},
{
"epoch": 0.3,
"learning_rate": 0.00035262788365095286,
"loss": 2.7932,
"theoretical_loss": 3.6505339053151076,
"tokens_seen": 996147200
},
{
"epoch": 0.3,
"learning_rate": 0.00035261785356068204,
"loss": 3.0392,
"theoretical_loss": 3.65051103404539,
"tokens_seen": 996212736
},
{
"epoch": 0.3,
"learning_rate": 0.0003526078234704113,
"loss": 2.745,
"theoretical_loss": 3.650488164701466,
"tokens_seen": 996278272
},
{
"epoch": 0.3,
"learning_rate": 0.0003525977933801404,
"loss": 2.7019,
"theoretical_loss": 3.6504652972830476,
"tokens_seen": 996343808
},
{
"epoch": 0.3,
"learning_rate": 0.00035258776328986964,
"loss": 2.757,
"theoretical_loss": 3.6504424317898447,
"tokens_seen": 996409344
},
{
"epoch": 0.3,
"learning_rate": 0.00035257773319959877,
"loss": 2.6365,
"theoretical_loss": 3.6504195682215697,
"tokens_seen": 996474880
},
{
"epoch": 0.3,
"learning_rate": 0.000352567703109328,
"loss": 2.8054,
"theoretical_loss": 3.650396706577934,
"tokens_seen": 996540416
},
{
"epoch": 0.3,
"learning_rate": 0.0003525576730190572,
"loss": 2.8481,
"theoretical_loss": 3.6503738468586486,
"tokens_seen": 996605952
},
{
"epoch": 0.3,
"learning_rate": 0.00035254764292878637,
"loss": 2.8039,
"theoretical_loss": 3.6503509890634254,
"tokens_seen": 996671488
},
{
"epoch": 0.3,
"learning_rate": 0.00035253761283851555,
"loss": 2.9588,
"theoretical_loss": 3.650328133191975,
"tokens_seen": 996737024
},
{
"epoch": 0.3,
"learning_rate": 0.0003525275827482448,
"loss": 2.5477,
"theoretical_loss": 3.65030527924401,
"tokens_seen": 996802560
},
{
"epoch": 0.3,
"learning_rate": 0.0003525175526579739,
"loss": 2.7206,
"theoretical_loss": 3.650282427219242,
"tokens_seen": 996868096
},
{
"epoch": 0.3,
"learning_rate": 0.00035250752256770314,
"loss": 2.7316,
"theoretical_loss": 3.650259577117382,
"tokens_seen": 996933632
},
{
"epoch": 0.3,
"learning_rate": 0.00035249749247743227,
"loss": 2.6952,
"theoretical_loss": 3.6502367289381423,
"tokens_seen": 996999168
},
{
"epoch": 0.3,
"learning_rate": 0.0003524874623871615,
"loss": 2.9592,
"theoretical_loss": 3.650213882681235,
"tokens_seen": 997064704
},
{
"epoch": 0.3,
"learning_rate": 0.0003524774322968907,
"loss": 2.7477,
"theoretical_loss": 3.650191038346372,
"tokens_seen": 997130240
},
{
"epoch": 0.3,
"learning_rate": 0.00035246740220661987,
"loss": 2.7021,
"theoretical_loss": 3.6501681959332646,
"tokens_seen": 997195776
},
{
"epoch": 0.3,
"learning_rate": 0.00035245737211634905,
"loss": 2.836,
"theoretical_loss": 3.6501453554416257,
"tokens_seen": 997261312
},
{
"epoch": 0.3,
"learning_rate": 0.00035244734202607823,
"loss": 2.834,
"theoretical_loss": 3.650122516871167,
"tokens_seen": 997326848
},
{
"epoch": 0.3,
"learning_rate": 0.0003524373119358074,
"loss": 2.5699,
"theoretical_loss": 3.650099680221601,
"tokens_seen": 997392384
},
{
"epoch": 0.3,
"learning_rate": 0.00035242728184553665,
"loss": 2.7655,
"theoretical_loss": 3.650076845492639,
"tokens_seen": 997457920
},
{
"epoch": 0.3,
"learning_rate": 0.0003524172517552658,
"loss": 2.6262,
"theoretical_loss": 3.650054012683994,
"tokens_seen": 997523456
},
{
"epoch": 0.3,
"learning_rate": 0.000352407221664995,
"loss": 2.7289,
"theoretical_loss": 3.6500311817953786,
"tokens_seen": 997588992
},
{
"epoch": 0.3,
"learning_rate": 0.00035239719157472414,
"loss": 2.8868,
"theoretical_loss": 3.6500083528265046,
"tokens_seen": 997654528
},
{
"epoch": 0.3,
"learning_rate": 0.00035238716148445337,
"loss": 2.7043,
"theoretical_loss": 3.6499855257770855,
"tokens_seen": 997720064
},
{
"epoch": 0.3,
"objective/train/docs_used": 572250,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.730179786682129,
"objective/train/theoretical_loss": 3.6499627006468325,
"objective/train/tokens_used": 1018245600,
"theoretical_loss": 3.6499627006468325,
"tokens_seen": 997785600
},
{
"epoch": 0.3,
"learning_rate": 0.00035237713139418255,
"loss": 2.9821,
"theoretical_loss": 3.6499627006468325,
"tokens_seen": 997785600
},
{
"epoch": 0.3,
"learning_rate": 0.00035236710130391173,
"loss": 2.5968,
"theoretical_loss": 3.6499398774354592,
"tokens_seen": 997851136
},
{
"epoch": 0.3,
"learning_rate": 0.0003523570712136409,
"loss": 2.6288,
"theoretical_loss": 3.6499170561426784,
"tokens_seen": 997916672
},
{
"epoch": 0.3,
"learning_rate": 0.00035234704112337015,
"loss": 2.4961,
"theoretical_loss": 3.6498942367682017,
"tokens_seen": 997982208
},
{
"epoch": 0.3,
"learning_rate": 0.0003523370110330993,
"loss": 2.7138,
"theoretical_loss": 3.6498714193117427,
"tokens_seen": 998047744
},
{
"epoch": 0.3,
"learning_rate": 0.0003523269809428285,
"loss": 2.6841,
"theoretical_loss": 3.6498486037730142,
"tokens_seen": 998113280
},
{
"epoch": 0.3,
"learning_rate": 0.00035231695085255764,
"loss": 2.6308,
"theoretical_loss": 3.6498257901517293,
"tokens_seen": 998178816
},
{
"epoch": 0.3,
"learning_rate": 0.0003523069207622869,
"loss": 2.7265,
"theoretical_loss": 3.6498029784476005,
"tokens_seen": 998244352
},
{
"epoch": 0.3,
"learning_rate": 0.00035229689067201606,
"loss": 2.4037,
"theoretical_loss": 3.6497801686603415,
"tokens_seen": 998309888
},
{
"epoch": 0.3,
"learning_rate": 0.00035228686058174524,
"loss": 3.0811,
"theoretical_loss": 3.649757360789664,
"tokens_seen": 998375424
},
{
"epoch": 0.3,
"learning_rate": 0.0003522768304914744,
"loss": 2.7667,
"theoretical_loss": 3.649734554835283,
"tokens_seen": 998440960
},
{
"epoch": 0.3,
"learning_rate": 0.0003522668004012036,
"loss": 2.9096,
"theoretical_loss": 3.649711750796911,
"tokens_seen": 998506496
},
{
"epoch": 0.3,
"learning_rate": 0.0003522567703109328,
"loss": 2.6938,
"theoretical_loss": 3.6496889486742607,
"tokens_seen": 998572032
},
{
"epoch": 0.3,
"learning_rate": 0.000352246740220662,
"loss": 2.9711,
"theoretical_loss": 3.6496661484670465,
"tokens_seen": 998637568
},
{
"epoch": 0.3,
"learning_rate": 0.00035223671013039114,
"loss": 2.6404,
"theoretical_loss": 3.6496433501749808,
"tokens_seen": 998703104
},
{
"epoch": 0.3,
"learning_rate": 0.0003522266800401204,
"loss": 2.7666,
"theoretical_loss": 3.649620553797777,
"tokens_seen": 998768640
},
{
"epoch": 0.3,
"learning_rate": 0.00035221664994984956,
"loss": 2.7017,
"theoretical_loss": 3.6495977593351503,
"tokens_seen": 998834176
},
{
"epoch": 0.3,
"learning_rate": 0.00035220661985957874,
"loss": 2.6534,
"theoretical_loss": 3.6495749667868123,
"tokens_seen": 998899712
},
{
"epoch": 0.3,
"learning_rate": 0.0003521965897693079,
"loss": 2.7827,
"theoretical_loss": 3.649552176152478,
"tokens_seen": 998965248
},
{
"epoch": 0.3,
"learning_rate": 0.0003521865596790371,
"loss": 2.7338,
"theoretical_loss": 3.64952938743186,
"tokens_seen": 999030784
},
{
"epoch": 0.3,
"learning_rate": 0.0003521765295887663,
"loss": 2.5273,
"theoretical_loss": 3.6495066006246732,
"tokens_seen": 999096320
},
{
"epoch": 0.3,
"learning_rate": 0.0003521664994984955,
"loss": 2.8272,
"theoretical_loss": 3.6494838157306306,
"tokens_seen": 999161856
},
{
"epoch": 0.3,
"learning_rate": 0.00035215646940822465,
"loss": 2.7084,
"theoretical_loss": 3.649461032749447,
"tokens_seen": 999227392
},
{
"epoch": 0.3,
"learning_rate": 0.0003521464393179539,
"loss": 2.5987,
"theoretical_loss": 3.6494382516808357,
"tokens_seen": 999292928
},
{
"epoch": 0.3,
"learning_rate": 0.000352136409227683,
"loss": 2.7496,
"theoretical_loss": 3.6494154725245105,
"tokens_seen": 999358464
},
{
"epoch": 0.3,
"objective/train/docs_used": 573370,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.015373468399048,
"objective/train/theoretical_loss": 3.649392695280186,
"objective/train/tokens_used": 1019884000,
"theoretical_loss": 3.649392695280186,
"tokens_seen": 999424000
},
{
"epoch": 0.3,
"learning_rate": 0.00035212637913741224,
"loss": 2.6249,
"theoretical_loss": 3.649392695280186,
"tokens_seen": 999424000
},
{
"epoch": 0.3,
"learning_rate": 0.0003521163490471414,
"loss": 2.8725,
"theoretical_loss": 3.649369919947577,
"tokens_seen": 999489536
},
{
"epoch": 0.3,
"learning_rate": 0.0003521063189568706,
"loss": 2.5105,
"theoretical_loss": 3.649347146526396,
"tokens_seen": 999555072
},
{
"epoch": 0.3,
"learning_rate": 0.00035209628886659984,
"loss": 2.5379,
"theoretical_loss": 3.6493243750163584,
"tokens_seen": 999620608
},
{
"epoch": 0.3,
"learning_rate": 0.00035208625877632897,
"loss": 2.7168,
"theoretical_loss": 3.649301605417178,
"tokens_seen": 999686144
},
{
"epoch": 0.3,
"learning_rate": 0.0003520762286860582,
"loss": 2.472,
"theoretical_loss": 3.64927883772857,
"tokens_seen": 999751680
},
{
"epoch": 0.3,
"learning_rate": 0.0003520661985957874,
"loss": 2.8682,
"theoretical_loss": 3.6492560719502483,
"tokens_seen": 999817216
},
{
"epoch": 0.3,
"learning_rate": 0.00035205616850551657,
"loss": 2.9424,
"theoretical_loss": 3.6492333080819277,
"tokens_seen": 999882752
},
{
"epoch": 0.3,
"learning_rate": 0.00035204613841524575,
"loss": 2.8286,
"theoretical_loss": 3.649210546123322,
"tokens_seen": 999948288
},
{
"epoch": 0.3,
"learning_rate": 0.000352036108324975,
"loss": 2.9333,
"theoretical_loss": 3.649187786074147,
"tokens_seen": 1000013824
},
{
"epoch": 0.3,
"learning_rate": 0.0003520260782347041,
"loss": 2.6922,
"theoretical_loss": 3.6491650279341163,
"tokens_seen": 1000079360
},
{
"epoch": 0.3,
"learning_rate": 0.00035201604814443334,
"loss": 2.9712,
"theoretical_loss": 3.649142271702946,
"tokens_seen": 1000144896
},
{
"epoch": 0.3,
"learning_rate": 0.00035200601805416247,
"loss": 2.7429,
"theoretical_loss": 3.6491195173803495,
"tokens_seen": 1000210432
},
{
"epoch": 0.3,
"learning_rate": 0.0003519959879638917,
"loss": 2.8951,
"theoretical_loss": 3.6490967649660426,
"tokens_seen": 1000275968
},
{
"epoch": 0.3,
"learning_rate": 0.0003519859578736209,
"loss": 2.6331,
"theoretical_loss": 3.64907401445974,
"tokens_seen": 1000341504
},
{
"epoch": 0.3,
"learning_rate": 0.00035197592778335007,
"loss": 2.9098,
"theoretical_loss": 3.6490512658611562,
"tokens_seen": 1000407040
},
{
"epoch": 0.3,
"learning_rate": 0.00035196589769307925,
"loss": 2.4394,
"theoretical_loss": 3.6490285191700074,
"tokens_seen": 1000472576
},
{
"epoch": 0.3,
"learning_rate": 0.00035195586760280843,
"loss": 2.6232,
"theoretical_loss": 3.6490057743860076,
"tokens_seen": 1000538112
},
{
"epoch": 0.3,
"learning_rate": 0.0003519458375125376,
"loss": 2.723,
"theoretical_loss": 3.6489830315088727,
"tokens_seen": 1000603648
},
{
"epoch": 0.3,
"learning_rate": 0.00035193580742226685,
"loss": 2.8309,
"theoretical_loss": 3.6489602905383176,
"tokens_seen": 1000669184
},
{
"epoch": 0.3,
"learning_rate": 0.000351925777331996,
"loss": 2.7596,
"theoretical_loss": 3.6489375514740576,
"tokens_seen": 1000734720
},
{
"epoch": 0.3,
"learning_rate": 0.0003519157472417252,
"loss": 2.7676,
"theoretical_loss": 3.648914814315809,
"tokens_seen": 1000800256
},
{
"epoch": 0.3,
"learning_rate": 0.00035190571715145434,
"loss": 2.7898,
"theoretical_loss": 3.6488920790632857,
"tokens_seen": 1000865792
},
{
"epoch": 0.3,
"learning_rate": 0.00035189568706118357,
"loss": 2.3045,
"theoretical_loss": 3.6488693457162045,
"tokens_seen": 1000931328
},
{
"epoch": 0.3,
"learning_rate": 0.00035188565697091275,
"loss": 2.6878,
"theoretical_loss": 3.64884661427428,
"tokens_seen": 1000996864
},
{
"epoch": 0.3,
"objective/train/docs_used": 574030,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.797346830368042,
"objective/train/theoretical_loss": 3.6488238847372285,
"objective/train/tokens_used": 1021522400,
"theoretical_loss": 3.6488238847372285,
"tokens_seen": 1001062400
},
{
"epoch": 0.3,
"learning_rate": 0.00035187562688064193,
"loss": 2.6267,
"theoretical_loss": 3.6488238847372285,
"tokens_seen": 1001062400
},
{
"epoch": 0.3,
"learning_rate": 0.0003518655967903711,
"loss": 2.6166,
"theoretical_loss": 3.6488011571047654,
"tokens_seen": 1001127936
},
{
"epoch": 0.3,
"learning_rate": 0.00035185556670010035,
"loss": 2.9248,
"theoretical_loss": 3.6487784313766065,
"tokens_seen": 1001193472
},
{
"epoch": 0.3,
"learning_rate": 0.0003518455366098295,
"loss": 2.8819,
"theoretical_loss": 3.6487557075524677,
"tokens_seen": 1001259008
},
{
"epoch": 0.3,
"learning_rate": 0.0003518355065195587,
"loss": 2.7652,
"theoretical_loss": 3.6487329856320647,
"tokens_seen": 1001324544
},
{
"epoch": 0.3,
"learning_rate": 0.00035182547642928784,
"loss": 2.7026,
"theoretical_loss": 3.6487102656151134,
"tokens_seen": 1001390080
},
{
"epoch": 0.3,
"learning_rate": 0.0003518154463390171,
"loss": 2.5374,
"theoretical_loss": 3.6486875475013303,
"tokens_seen": 1001455616
},
{
"epoch": 0.3,
"learning_rate": 0.00035180541624874626,
"loss": 2.8382,
"theoretical_loss": 3.6486648312904304,
"tokens_seen": 1001521152
},
{
"epoch": 0.3,
"learning_rate": 0.00035179538615847544,
"loss": 2.8238,
"theoretical_loss": 3.648642116982131,
"tokens_seen": 1001586688
},
{
"epoch": 0.3,
"learning_rate": 0.0003517853560682046,
"loss": 2.9434,
"theoretical_loss": 3.6486194045761473,
"tokens_seen": 1001652224
},
{
"epoch": 0.3,
"learning_rate": 0.0003517753259779338,
"loss": 2.477,
"theoretical_loss": 3.6485966940721957,
"tokens_seen": 1001717760
},
{
"epoch": 0.3,
"learning_rate": 0.000351765295887663,
"loss": 2.8723,
"theoretical_loss": 3.6485739854699935,
"tokens_seen": 1001783296
},
{
"epoch": 0.3,
"learning_rate": 0.0003517552657973922,
"loss": 2.7882,
"theoretical_loss": 3.648551278769256,
"tokens_seen": 1001848832
},
{
"epoch": 0.3,
"learning_rate": 0.00035174523570712134,
"loss": 2.8,
"theoretical_loss": 3.6485285739696995,
"tokens_seen": 1001914368
},
{
"epoch": 0.3,
"learning_rate": 0.0003517352056168506,
"loss": 2.8476,
"theoretical_loss": 3.648505871071041,
"tokens_seen": 1001979904
},
{
"epoch": 0.3,
"learning_rate": 0.00035172517552657976,
"loss": 2.8294,
"theoretical_loss": 3.648483170072997,
"tokens_seen": 1002045440
},
{
"epoch": 0.3,
"learning_rate": 0.00035171514543630894,
"loss": 2.8956,
"theoretical_loss": 3.6484604709752846,
"tokens_seen": 1002110976
},
{
"epoch": 0.3,
"learning_rate": 0.0003517051153460381,
"loss": 2.845,
"theoretical_loss": 3.6484377737776192,
"tokens_seen": 1002176512
},
{
"epoch": 0.3,
"learning_rate": 0.0003516950852557673,
"loss": 2.5499,
"theoretical_loss": 3.6484150784797187,
"tokens_seen": 1002242048
},
{
"epoch": 0.3,
"learning_rate": 0.0003516850551654965,
"loss": 2.7895,
"theoretical_loss": 3.6483923850812987,
"tokens_seen": 1002307584
},
{
"epoch": 0.3,
"learning_rate": 0.0003516750250752257,
"loss": 2.5873,
"theoretical_loss": 3.6483696935820777,
"tokens_seen": 1002373120
},
{
"epoch": 0.3,
"learning_rate": 0.00035166499498495485,
"loss": 2.9668,
"theoretical_loss": 3.6483470039817707,
"tokens_seen": 1002438656
},
{
"epoch": 0.3,
"learning_rate": 0.0003516549648946841,
"loss": 2.7931,
"theoretical_loss": 3.648324316280096,
"tokens_seen": 1002504192
},
{
"epoch": 0.3,
"learning_rate": 0.0003516449348044132,
"loss": 2.8564,
"theoretical_loss": 3.6483016304767704,
"tokens_seen": 1002569728
},
{
"epoch": 0.3,
"learning_rate": 0.00035163490471414244,
"loss": 2.701,
"theoretical_loss": 3.6482789465715104,
"tokens_seen": 1002635264
},
{
"epoch": 0.3,
"objective/train/docs_used": 574442,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.728538990020752,
"objective/train/theoretical_loss": 3.6482562645640337,
"objective/train/tokens_used": 1023160800,
"theoretical_loss": 3.6482562645640337,
"tokens_seen": 1002700800
},
{
"epoch": 0.3,
"learning_rate": 0.0003516248746238716,
"loss": 2.8289,
"theoretical_loss": 3.6482562645640337,
"tokens_seen": 1002700800
},
{
"epoch": 0.3,
"learning_rate": 0.0003516148445336008,
"loss": 2.7044,
"theoretical_loss": 3.648233584454057,
"tokens_seen": 1002766336
},
{
"epoch": 0.3,
"learning_rate": 0.00035160481444333,
"loss": 2.778,
"theoretical_loss": 3.6482109062412986,
"tokens_seen": 1002831872
},
{
"epoch": 0.3,
"learning_rate": 0.00035159478435305917,
"loss": 2.7129,
"theoretical_loss": 3.6481882299254744,
"tokens_seen": 1002897408
},
{
"epoch": 0.3,
"learning_rate": 0.00035158475426278835,
"loss": 2.6938,
"theoretical_loss": 3.648165555506303,
"tokens_seen": 1002962944
},
{
"epoch": 0.3,
"learning_rate": 0.0003515747241725176,
"loss": 3.0319,
"theoretical_loss": 3.648142882983501,
"tokens_seen": 1003028480
},
{
"epoch": 0.3,
"learning_rate": 0.0003515646940822467,
"loss": 2.8901,
"theoretical_loss": 3.6481202123567864,
"tokens_seen": 1003094016
},
{
"epoch": 0.3,
"learning_rate": 0.00035155466399197595,
"loss": 2.6829,
"theoretical_loss": 3.6480975436258767,
"tokens_seen": 1003159552
},
{
"epoch": 0.3,
"learning_rate": 0.00035154463390170513,
"loss": 2.7388,
"theoretical_loss": 3.6480748767904894,
"tokens_seen": 1003225088
},
{
"epoch": 0.3,
"learning_rate": 0.0003515346038114343,
"loss": 2.7612,
"theoretical_loss": 3.6480522118503425,
"tokens_seen": 1003290624
},
{
"epoch": 0.3,
"learning_rate": 0.0003515245737211635,
"loss": 2.6815,
"theoretical_loss": 3.6480295488051535,
"tokens_seen": 1003356160
},
{
"epoch": 0.3,
"learning_rate": 0.00035151454363089267,
"loss": 2.6679,
"theoretical_loss": 3.64800688765464,
"tokens_seen": 1003421696
},
{
"epoch": 0.3,
"learning_rate": 0.00035150451354062185,
"loss": 2.9232,
"theoretical_loss": 3.64798422839852,
"tokens_seen": 1003487232
},
{
"epoch": 0.3,
"learning_rate": 0.0003514944834503511,
"loss": 2.5723,
"theoretical_loss": 3.647961571036512,
"tokens_seen": 1003552768
},
{
"epoch": 0.3,
"learning_rate": 0.0003514844533600802,
"loss": 2.8263,
"theoretical_loss": 3.647938915568333,
"tokens_seen": 1003618304
},
{
"epoch": 0.3,
"learning_rate": 0.00035147442326980945,
"loss": 2.7832,
"theoretical_loss": 3.647916261993702,
"tokens_seen": 1003683840
},
{
"epoch": 0.3,
"learning_rate": 0.0003514643931795386,
"loss": 2.5526,
"theoretical_loss": 3.6478936103123365,
"tokens_seen": 1003749376
},
{
"epoch": 0.3,
"learning_rate": 0.0003514543630892678,
"loss": 2.79,
"theoretical_loss": 3.6478709605239548,
"tokens_seen": 1003814912
},
{
"epoch": 0.3,
"learning_rate": 0.000351444332998997,
"loss": 2.8623,
"theoretical_loss": 3.647848312628275,
"tokens_seen": 1003880448
},
{
"epoch": 0.3,
"learning_rate": 0.0003514343029087262,
"loss": 2.9593,
"theoretical_loss": 3.6478256666250157,
"tokens_seen": 1003945984
},
{
"epoch": 0.3,
"learning_rate": 0.00035142427281845536,
"loss": 2.8865,
"theoretical_loss": 3.6478030225138953,
"tokens_seen": 1004011520
},
{
"epoch": 0.3,
"learning_rate": 0.00035141424272818454,
"loss": 2.9961,
"theoretical_loss": 3.647780380294632,
"tokens_seen": 1004077056
},
{
"epoch": 0.3,
"learning_rate": 0.0003514042126379137,
"loss": 2.7263,
"theoretical_loss": 3.647757739966944,
"tokens_seen": 1004142592
},
{
"epoch": 0.3,
"learning_rate": 0.00035139418254764295,
"loss": 2.8292,
"theoretical_loss": 3.6477351015305506,
"tokens_seen": 1004208128
},
{
"epoch": 0.3,
"learning_rate": 0.0003513841524573721,
"loss": 2.9099,
"theoretical_loss": 3.6477124649851698,
"tokens_seen": 1004273664
},
{
"epoch": 0.3,
"objective/train/docs_used": 575683,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8427352905273438,
"objective/train/theoretical_loss": 3.64768983033052,
"objective/train/tokens_used": 1024799200,
"theoretical_loss": 3.64768983033052,
"tokens_seen": 1004339200
},
{
"epoch": 0.3,
"learning_rate": 0.0003513741223671013,
"loss": 2.8187,
"theoretical_loss": 3.64768983033052,
"tokens_seen": 1004339200
},
{
"epoch": 0.3,
"learning_rate": 0.0003513640922768305,
"loss": 2.8167,
"theoretical_loss": 3.647667197566321,
"tokens_seen": 1004404736
},
{
"epoch": 0.3,
"learning_rate": 0.0003513540621865597,
"loss": 2.6978,
"theoretical_loss": 3.6476445666922905,
"tokens_seen": 1004470272
},
{
"epoch": 0.3,
"learning_rate": 0.0003513440320962889,
"loss": 2.806,
"theoretical_loss": 3.6476219377081476,
"tokens_seen": 1004535808
},
{
"epoch": 0.3,
"learning_rate": 0.00035133400200601804,
"loss": 2.8574,
"theoretical_loss": 3.6475993106136118,
"tokens_seen": 1004601344
},
{
"epoch": 0.3,
"learning_rate": 0.0003513239719157473,
"loss": 2.6449,
"theoretical_loss": 3.647576685408401,
"tokens_seen": 1004666880
},
{
"epoch": 0.3,
"learning_rate": 0.00035131394182547646,
"loss": 2.5899,
"theoretical_loss": 3.6475540620922353,
"tokens_seen": 1004732416
},
{
"epoch": 0.3,
"learning_rate": 0.00035130391173520564,
"loss": 2.6667,
"theoretical_loss": 3.6475314406648334,
"tokens_seen": 1004797952
},
{
"epoch": 0.3,
"learning_rate": 0.0003512938816449348,
"loss": 2.7024,
"theoretical_loss": 3.647508821125914,
"tokens_seen": 1004863488
},
{
"epoch": 0.3,
"learning_rate": 0.000351283851554664,
"loss": 2.734,
"theoretical_loss": 3.6474862034751965,
"tokens_seen": 1004929024
},
{
"epoch": 0.3,
"learning_rate": 0.0003512738214643932,
"loss": 2.6087,
"theoretical_loss": 3.6474635877124,
"tokens_seen": 1004994560
},
{
"epoch": 0.3,
"learning_rate": 0.0003512637913741224,
"loss": 2.8027,
"theoretical_loss": 3.6474409738372446,
"tokens_seen": 1005060096
},
{
"epoch": 0.3,
"learning_rate": 0.00035125376128385154,
"loss": 2.5738,
"theoretical_loss": 3.647418361849449,
"tokens_seen": 1005125632
},
{
"epoch": 0.3,
"learning_rate": 0.0003512437311935808,
"loss": 2.9155,
"theoretical_loss": 3.647395751748733,
"tokens_seen": 1005191168
},
{
"epoch": 0.3,
"learning_rate": 0.00035123370110330996,
"loss": 2.8431,
"theoretical_loss": 3.6473731435348156,
"tokens_seen": 1005256704
},
{
"epoch": 0.3,
"learning_rate": 0.00035122367101303914,
"loss": 2.8715,
"theoretical_loss": 3.6473505372074166,
"tokens_seen": 1005322240
},
{
"epoch": 0.3,
"learning_rate": 0.0003512136409227683,
"loss": 2.9025,
"theoretical_loss": 3.647327932766256,
"tokens_seen": 1005387776
},
{
"epoch": 0.3,
"learning_rate": 0.0003512036108324975,
"loss": 2.8852,
"theoretical_loss": 3.647305330211053,
"tokens_seen": 1005453312
},
{
"epoch": 0.3,
"learning_rate": 0.0003511935807422267,
"loss": 2.6091,
"theoretical_loss": 3.6472827295415273,
"tokens_seen": 1005518848
},
{
"epoch": 0.3,
"learning_rate": 0.0003511835506519559,
"loss": 2.805,
"theoretical_loss": 3.647260130757399,
"tokens_seen": 1005584384
},
{
"epoch": 0.3,
"learning_rate": 0.00035117352056168505,
"loss": 2.8735,
"theoretical_loss": 3.647237533858388,
"tokens_seen": 1005649920
},
{
"epoch": 0.3,
"learning_rate": 0.0003511634904714143,
"loss": 2.8247,
"theoretical_loss": 3.647214938844214,
"tokens_seen": 1005715456
},
{
"epoch": 0.3,
"learning_rate": 0.0003511534603811434,
"loss": 2.831,
"theoretical_loss": 3.647192345714597,
"tokens_seen": 1005780992
},
{
"epoch": 0.3,
"learning_rate": 0.00035114343029087264,
"loss": 2.7901,
"theoretical_loss": 3.6471697544692567,
"tokens_seen": 1005846528
},
{
"epoch": 0.3,
"learning_rate": 0.0003511334002006018,
"loss": 2.5861,
"theoretical_loss": 3.6471471651079135,
"tokens_seen": 1005912064
},
{
"epoch": 0.3,
"objective/train/docs_used": 576356,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0992517471313477,
"objective/train/theoretical_loss": 3.6471245776302883,
"objective/train/tokens_used": 1026437600,
"theoretical_loss": 3.6471245776302883,
"tokens_seen": 1005977600
},
{
"epoch": 0.3,
"learning_rate": 0.000351123370110331,
"loss": 2.7293,
"theoretical_loss": 3.6471245776302883,
"tokens_seen": 1005977600
},
{
"epoch": 0.3,
"learning_rate": 0.0003511133400200602,
"loss": 2.7258,
"theoretical_loss": 3.6471019920361,
"tokens_seen": 1006043136
},
{
"epoch": 0.3,
"learning_rate": 0.00035110330992978937,
"loss": 2.7864,
"theoretical_loss": 3.6470794083250695,
"tokens_seen": 1006108672
},
{
"epoch": 0.3,
"learning_rate": 0.00035109327983951855,
"loss": 2.6791,
"theoretical_loss": 3.6470568264969176,
"tokens_seen": 1006174208
},
{
"epoch": 0.3,
"learning_rate": 0.0003510832497492478,
"loss": 2.8272,
"theoretical_loss": 3.647034246551364,
"tokens_seen": 1006239744
},
{
"epoch": 0.3,
"learning_rate": 0.0003510732196589769,
"loss": 2.8252,
"theoretical_loss": 3.6470116684881297,
"tokens_seen": 1006305280
},
{
"epoch": 0.3,
"learning_rate": 0.00035106318956870615,
"loss": 2.7312,
"theoretical_loss": 3.646989092306934,
"tokens_seen": 1006370816
},
{
"epoch": 0.3,
"learning_rate": 0.00035105315947843533,
"loss": 2.6442,
"theoretical_loss": 3.646966518007499,
"tokens_seen": 1006436352
},
{
"epoch": 0.31,
"learning_rate": 0.0003510431293881645,
"loss": 2.6596,
"theoretical_loss": 3.6469439455895447,
"tokens_seen": 1006501888
},
{
"epoch": 0.31,
"learning_rate": 0.0003510330992978937,
"loss": 2.5505,
"theoretical_loss": 3.646921375052792,
"tokens_seen": 1006567424
},
{
"epoch": 0.31,
"learning_rate": 0.00035102306920762287,
"loss": 2.8254,
"theoretical_loss": 3.646898806396961,
"tokens_seen": 1006632960
},
{
"epoch": 0.31,
"learning_rate": 0.00035101303911735205,
"loss": 2.8271,
"theoretical_loss": 3.646876239621773,
"tokens_seen": 1006698496
},
{
"epoch": 0.31,
"learning_rate": 0.0003510030090270813,
"loss": 2.4623,
"theoretical_loss": 3.646853674726949,
"tokens_seen": 1006764032
},
{
"epoch": 0.31,
"learning_rate": 0.0003509929789368104,
"loss": 2.7809,
"theoretical_loss": 3.64683111171221,
"tokens_seen": 1006829568
},
{
"epoch": 0.31,
"learning_rate": 0.00035098294884653965,
"loss": 2.8044,
"theoretical_loss": 3.6468085505772763,
"tokens_seen": 1006895104
},
{
"epoch": 0.31,
"learning_rate": 0.0003509729187562688,
"loss": 2.8757,
"theoretical_loss": 3.6467859913218694,
"tokens_seen": 1006960640
},
{
"epoch": 0.31,
"learning_rate": 0.000350962888665998,
"loss": 2.7019,
"theoretical_loss": 3.646763433945711,
"tokens_seen": 1007026176
},
{
"epoch": 0.31,
"learning_rate": 0.0003509528585757272,
"loss": 2.8487,
"theoretical_loss": 3.6467408784485214,
"tokens_seen": 1007091712
},
{
"epoch": 0.31,
"learning_rate": 0.0003509428284854564,
"loss": 2.7668,
"theoretical_loss": 3.6467183248300215,
"tokens_seen": 1007157248
},
{
"epoch": 0.31,
"learning_rate": 0.00035093279839518556,
"loss": 2.648,
"theoretical_loss": 3.6466957730899336,
"tokens_seen": 1007222784
},
{
"epoch": 0.31,
"learning_rate": 0.00035092276830491474,
"loss": 2.9223,
"theoretical_loss": 3.646673223227979,
"tokens_seen": 1007288320
},
{
"epoch": 0.31,
"learning_rate": 0.0003509127382146439,
"loss": 2.8834,
"theoretical_loss": 3.6466506752438788,
"tokens_seen": 1007353856
},
{
"epoch": 0.31,
"learning_rate": 0.00035090270812437315,
"loss": 2.6454,
"theoretical_loss": 3.646628129137354,
"tokens_seen": 1007419392
},
{
"epoch": 0.31,
"learning_rate": 0.0003508926780341023,
"loss": 2.6594,
"theoretical_loss": 3.6466055849081265,
"tokens_seen": 1007484928
},
{
"epoch": 0.31,
"learning_rate": 0.0003508826479438315,
"loss": 2.8483,
"theoretical_loss": 3.6465830425559185,
"tokens_seen": 1007550464
},
{
"epoch": 0.31,
"objective/train/docs_used": 577861,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6302406787872314,
"objective/train/theoretical_loss": 3.6465605020804506,
"objective/train/tokens_used": 1028076000,
"theoretical_loss": 3.6465605020804506,
"tokens_seen": 1007616000
},
{
"epoch": 0.31,
"learning_rate": 0.0003508726178535607,
"loss": 2.8136,
"theoretical_loss": 3.6465605020804506,
"tokens_seen": 1007616000
},
{
"epoch": 0.31,
"learning_rate": 0.0003508625877632899,
"loss": 2.9103,
"theoretical_loss": 3.646537963481445,
"tokens_seen": 1007681536
},
{
"epoch": 0.31,
"learning_rate": 0.00035085255767301906,
"loss": 3.0026,
"theoretical_loss": 3.646515426758624,
"tokens_seen": 1007747072
},
{
"epoch": 0.31,
"learning_rate": 0.00035084252758274824,
"loss": 2.4689,
"theoretical_loss": 3.646492891911708,
"tokens_seen": 1007812608
},
{
"epoch": 0.31,
"learning_rate": 0.0003508324974924774,
"loss": 2.8969,
"theoretical_loss": 3.6464703589404204,
"tokens_seen": 1007878144
},
{
"epoch": 0.31,
"learning_rate": 0.00035082246740220666,
"loss": 2.6418,
"theoretical_loss": 3.646447827844482,
"tokens_seen": 1007943680
},
{
"epoch": 0.31,
"learning_rate": 0.0003508124373119358,
"loss": 2.9756,
"theoretical_loss": 3.646425298623616,
"tokens_seen": 1008009216
},
{
"epoch": 0.31,
"learning_rate": 0.000350802407221665,
"loss": 2.6738,
"theoretical_loss": 3.6464027712775433,
"tokens_seen": 1008074752
},
{
"epoch": 0.31,
"learning_rate": 0.00035079237713139415,
"loss": 2.8822,
"theoretical_loss": 3.646380245805986,
"tokens_seen": 1008140288
},
{
"epoch": 0.31,
"learning_rate": 0.0003507823470411234,
"loss": 2.7789,
"theoretical_loss": 3.6463577222086676,
"tokens_seen": 1008205824
},
{
"epoch": 0.31,
"learning_rate": 0.00035077231695085256,
"loss": 2.6163,
"theoretical_loss": 3.6463352004853093,
"tokens_seen": 1008271360
},
{
"epoch": 0.31,
"learning_rate": 0.00035076228686058174,
"loss": 2.7014,
"theoretical_loss": 3.646312680635633,
"tokens_seen": 1008336896
},
{
"epoch": 0.31,
"learning_rate": 0.0003507522567703109,
"loss": 2.8553,
"theoretical_loss": 3.646290162659363,
"tokens_seen": 1008402432
},
{
"epoch": 0.31,
"learning_rate": 0.00035074222668004016,
"loss": 2.7895,
"theoretical_loss": 3.646267646556219,
"tokens_seen": 1008467968
},
{
"epoch": 0.31,
"learning_rate": 0.0003507321965897693,
"loss": 2.6309,
"theoretical_loss": 3.6462451323259257,
"tokens_seen": 1008533504
},
{
"epoch": 0.31,
"learning_rate": 0.0003507221664994985,
"loss": 2.7109,
"theoretical_loss": 3.6462226199682046,
"tokens_seen": 1008599040
},
{
"epoch": 0.31,
"learning_rate": 0.00035071213640922765,
"loss": 2.9442,
"theoretical_loss": 3.646200109482778,
"tokens_seen": 1008664576
},
{
"epoch": 0.31,
"learning_rate": 0.0003507021063189569,
"loss": 2.534,
"theoretical_loss": 3.646177600869369,
"tokens_seen": 1008730112
},
{
"epoch": 0.31,
"learning_rate": 0.00035069207622868607,
"loss": 2.7153,
"theoretical_loss": 3.646155094127701,
"tokens_seen": 1008795648
},
{
"epoch": 0.31,
"learning_rate": 0.00035068204613841525,
"loss": 2.8011,
"theoretical_loss": 3.646132589257496,
"tokens_seen": 1008861184
},
{
"epoch": 0.31,
"learning_rate": 0.00035067201604814443,
"loss": 2.7132,
"theoretical_loss": 3.6461100862584765,
"tokens_seen": 1008926720
},
{
"epoch": 0.31,
"learning_rate": 0.0003506619859578736,
"loss": 2.737,
"theoretical_loss": 3.646087585130366,
"tokens_seen": 1008992256
},
{
"epoch": 0.31,
"learning_rate": 0.0003506519558676028,
"loss": 2.8703,
"theoretical_loss": 3.6460650858728876,
"tokens_seen": 1009057792
},
{
"epoch": 0.31,
"learning_rate": 0.000350641925777332,
"loss": 2.6688,
"theoretical_loss": 3.6460425884857637,
"tokens_seen": 1009123328
},
{
"epoch": 0.31,
"learning_rate": 0.00035063189568706115,
"loss": 2.8116,
"theoretical_loss": 3.6460200929687177,
"tokens_seen": 1009188864
},
{
"epoch": 0.31,
"objective/train/docs_used": 578450,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1611180305480957,
"objective/train/theoretical_loss": 3.6459975993214724,
"objective/train/tokens_used": 1029714400,
"theoretical_loss": 3.6459975993214724,
"tokens_seen": 1009254400
},
{
"epoch": 0.31,
"learning_rate": 0.0003506218655967904,
"loss": 2.7406,
"theoretical_loss": 3.6459975993214724,
"tokens_seen": 1009254400
},
{
"epoch": 0.31,
"learning_rate": 0.0003506118355065195,
"loss": 3.0198,
"theoretical_loss": 3.6459751075437516,
"tokens_seen": 1009319936
},
{
"epoch": 0.31,
"learning_rate": 0.00035060180541624875,
"loss": 2.7561,
"theoretical_loss": 3.645952617635278,
"tokens_seen": 1009385472
},
{
"epoch": 0.31,
"learning_rate": 0.000350591775325978,
"loss": 2.5017,
"theoretical_loss": 3.645930129595775,
"tokens_seen": 1009451008
},
{
"epoch": 0.31,
"learning_rate": 0.0003505817452357071,
"loss": 2.7587,
"theoretical_loss": 3.645907643424966,
"tokens_seen": 1009516544
},
{
"epoch": 0.31,
"learning_rate": 0.00035057171514543635,
"loss": 2.5996,
"theoretical_loss": 3.6458851591225754,
"tokens_seen": 1009582080
},
{
"epoch": 0.31,
"learning_rate": 0.00035056168505516553,
"loss": 2.9329,
"theoretical_loss": 3.645862676688325,
"tokens_seen": 1009647616
},
{
"epoch": 0.31,
"learning_rate": 0.0003505516549648947,
"loss": 2.6225,
"theoretical_loss": 3.645840196121939,
"tokens_seen": 1009713152
},
{
"epoch": 0.31,
"learning_rate": 0.0003505416248746239,
"loss": 2.7193,
"theoretical_loss": 3.6458177174231414,
"tokens_seen": 1009778688
},
{
"epoch": 0.31,
"learning_rate": 0.00035053159478435307,
"loss": 2.738,
"theoretical_loss": 3.6457952405916556,
"tokens_seen": 1009844224
},
{
"epoch": 0.31,
"learning_rate": 0.00035052156469408225,
"loss": 2.6249,
"theoretical_loss": 3.6457727656272043,
"tokens_seen": 1009909760
},
{
"epoch": 0.31,
"learning_rate": 0.0003505115346038115,
"loss": 2.8273,
"theoretical_loss": 3.645750292529513,
"tokens_seen": 1009975296
},
{
"epoch": 0.31,
"learning_rate": 0.0003505015045135406,
"loss": 2.8875,
"theoretical_loss": 3.6457278212983053,
"tokens_seen": 1010040832
},
{
"epoch": 0.31,
"learning_rate": 0.00035049147442326985,
"loss": 2.8674,
"theoretical_loss": 3.645705351933304,
"tokens_seen": 1010106368
},
{
"epoch": 0.31,
"learning_rate": 0.000350481444332999,
"loss": 2.8731,
"theoretical_loss": 3.6456828844342337,
"tokens_seen": 1010171904
},
{
"epoch": 0.31,
"learning_rate": 0.0003504714142427282,
"loss": 2.7254,
"theoretical_loss": 3.6456604188008184,
"tokens_seen": 1010237440
},
{
"epoch": 0.31,
"learning_rate": 0.0003504613841524574,
"loss": 2.8503,
"theoretical_loss": 3.6456379550327815,
"tokens_seen": 1010302976
},
{
"epoch": 0.31,
"learning_rate": 0.0003504513540621866,
"loss": 2.6326,
"theoretical_loss": 3.6456154931298483,
"tokens_seen": 1010368512
},
{
"epoch": 0.31,
"learning_rate": 0.00035044132397191576,
"loss": 2.8711,
"theoretical_loss": 3.6455930330917425,
"tokens_seen": 1010434048
},
{
"epoch": 0.31,
"learning_rate": 0.00035043129388164494,
"loss": 2.5508,
"theoretical_loss": 3.645570574918188,
"tokens_seen": 1010499584
},
{
"epoch": 0.31,
"learning_rate": 0.0003504212637913741,
"loss": 2.8305,
"theoretical_loss": 3.645548118608909,
"tokens_seen": 1010565120
},
{
"epoch": 0.31,
"learning_rate": 0.00035041123370110335,
"loss": 2.7541,
"theoretical_loss": 3.6455256641636304,
"tokens_seen": 1010630656
},
{
"epoch": 0.31,
"learning_rate": 0.0003504012036108325,
"loss": 2.5253,
"theoretical_loss": 3.6455032115820765,
"tokens_seen": 1010696192
},
{
"epoch": 0.31,
"learning_rate": 0.0003503911735205617,
"loss": 2.6133,
"theoretical_loss": 3.645480760863972,
"tokens_seen": 1010761728
},
{
"epoch": 0.31,
"learning_rate": 0.0003503811434302909,
"loss": 2.9394,
"theoretical_loss": 3.6454583120090405,
"tokens_seen": 1010827264
},
{
"epoch": 0.31,
"objective/train/docs_used": 579792,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7755496501922607,
"objective/train/theoretical_loss": 3.6454358650170073,
"objective/train/tokens_used": 1031352800,
"theoretical_loss": 3.6454358650170073,
"tokens_seen": 1010892800
},
{
"epoch": 0.31,
"learning_rate": 0.0003503711133400201,
"loss": 2.6469,
"theoretical_loss": 3.6454358650170073,
"tokens_seen": 1010892800
},
{
"epoch": 0.31,
"learning_rate": 0.00035036108324974926,
"loss": 2.9115,
"theoretical_loss": 3.645413419887597,
"tokens_seen": 1010958336
},
{
"epoch": 0.31,
"learning_rate": 0.00035035105315947844,
"loss": 2.7752,
"theoretical_loss": 3.6453909766205346,
"tokens_seen": 1011023872
},
{
"epoch": 0.31,
"learning_rate": 0.0003503410230692076,
"loss": 2.6336,
"theoretical_loss": 3.6453685352155443,
"tokens_seen": 1011089408
},
{
"epoch": 0.31,
"learning_rate": 0.00035033099297893686,
"loss": 2.7308,
"theoretical_loss": 3.645346095672352,
"tokens_seen": 1011154944
},
{
"epoch": 0.31,
"learning_rate": 0.000350320962888666,
"loss": 2.7355,
"theoretical_loss": 3.6453236579906805,
"tokens_seen": 1011220480
},
{
"epoch": 0.31,
"learning_rate": 0.0003503109327983952,
"loss": 2.8988,
"theoretical_loss": 3.6453012221702563,
"tokens_seen": 1011286016
},
{
"epoch": 0.31,
"learning_rate": 0.00035030090270812435,
"loss": 2.5644,
"theoretical_loss": 3.6452787882108044,
"tokens_seen": 1011351552
},
{
"epoch": 0.31,
"learning_rate": 0.0003502908726178536,
"loss": 2.5655,
"theoretical_loss": 3.64525635611205,
"tokens_seen": 1011417088
},
{
"epoch": 0.31,
"learning_rate": 0.00035028084252758276,
"loss": 2.6341,
"theoretical_loss": 3.645233925873717,
"tokens_seen": 1011482624
},
{
"epoch": 0.31,
"learning_rate": 0.00035027081243731194,
"loss": 2.821,
"theoretical_loss": 3.6452114974955316,
"tokens_seen": 1011548160
},
{
"epoch": 0.31,
"learning_rate": 0.0003502607823470411,
"loss": 2.6058,
"theoretical_loss": 3.645189070977219,
"tokens_seen": 1011613696
},
{
"epoch": 0.31,
"learning_rate": 0.00035025075225677036,
"loss": 2.6163,
"theoretical_loss": 3.645166646318504,
"tokens_seen": 1011679232
},
{
"epoch": 0.31,
"learning_rate": 0.0003502407221664995,
"loss": 2.9207,
"theoretical_loss": 3.6451442235191127,
"tokens_seen": 1011744768
},
{
"epoch": 0.31,
"learning_rate": 0.0003502306920762287,
"loss": 2.7379,
"theoretical_loss": 3.64512180257877,
"tokens_seen": 1011810304
},
{
"epoch": 0.31,
"learning_rate": 0.00035022066198595785,
"loss": 2.8262,
"theoretical_loss": 3.645099383497201,
"tokens_seen": 1011875840
},
{
"epoch": 0.31,
"learning_rate": 0.0003502106318956871,
"loss": 2.823,
"theoretical_loss": 3.6450769662741322,
"tokens_seen": 1011941376
},
{
"epoch": 0.31,
"learning_rate": 0.00035020060180541627,
"loss": 2.7166,
"theoretical_loss": 3.6450545509092884,
"tokens_seen": 1012006912
},
{
"epoch": 0.31,
"learning_rate": 0.00035019057171514545,
"loss": 2.7444,
"theoretical_loss": 3.6450321374023957,
"tokens_seen": 1012072448
},
{
"epoch": 0.31,
"learning_rate": 0.00035018054162487463,
"loss": 2.8552,
"theoretical_loss": 3.6450097257531793,
"tokens_seen": 1012137984
},
{
"epoch": 0.31,
"learning_rate": 0.0003501705115346038,
"loss": 2.8329,
"theoretical_loss": 3.6449873159613655,
"tokens_seen": 1012203520
},
{
"epoch": 0.31,
"learning_rate": 0.000350160481444333,
"loss": 2.9205,
"theoretical_loss": 3.64496490802668,
"tokens_seen": 1012269056
},
{
"epoch": 0.31,
"learning_rate": 0.0003501504513540622,
"loss": 3.0081,
"theoretical_loss": 3.644942501948848,
"tokens_seen": 1012334592
},
{
"epoch": 0.31,
"learning_rate": 0.00035014042126379135,
"loss": 2.447,
"theoretical_loss": 3.6449200977275966,
"tokens_seen": 1012400128
},
{
"epoch": 0.31,
"learning_rate": 0.0003501303911735206,
"loss": 2.895,
"theoretical_loss": 3.6448976953626513,
"tokens_seen": 1012465664
},
{
"epoch": 0.31,
"objective/train/docs_used": 580179,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.598353147506714,
"objective/train/theoretical_loss": 3.6448752948537377,
"objective/train/tokens_used": 1032991200,
"theoretical_loss": 3.6448752948537377,
"tokens_seen": 1012531200
},
{
"epoch": 0.31,
"learning_rate": 0.0003501203610832497,
"loss": 2.6433,
"theoretical_loss": 3.6448752948537377,
"tokens_seen": 1012531200
},
{
"epoch": 0.31,
"learning_rate": 0.00035011033099297895,
"loss": 2.7563,
"theoretical_loss": 3.6448528962005824,
"tokens_seen": 1012596736
},
{
"epoch": 0.31,
"learning_rate": 0.00035010030090270813,
"loss": 2.7673,
"theoretical_loss": 3.6448304994029113,
"tokens_seen": 1012662272
},
{
"epoch": 0.31,
"learning_rate": 0.0003500902708124373,
"loss": 2.8887,
"theoretical_loss": 3.644808104460451,
"tokens_seen": 1012727808
},
{
"epoch": 0.31,
"learning_rate": 0.0003500802407221665,
"loss": 2.8419,
"theoretical_loss": 3.6447857113729274,
"tokens_seen": 1012793344
},
{
"epoch": 0.31,
"learning_rate": 0.00035007021063189573,
"loss": 2.9237,
"theoretical_loss": 3.6447633201400667,
"tokens_seen": 1012858880
},
{
"epoch": 0.31,
"learning_rate": 0.00035006018054162486,
"loss": 2.5599,
"theoretical_loss": 3.6447409307615963,
"tokens_seen": 1012924416
},
{
"epoch": 0.31,
"learning_rate": 0.0003500501504513541,
"loss": 2.764,
"theoretical_loss": 3.644718543237241,
"tokens_seen": 1012989952
},
{
"epoch": 0.31,
"learning_rate": 0.0003500401203610832,
"loss": 2.9066,
"theoretical_loss": 3.644696157566729,
"tokens_seen": 1013055488
},
{
"epoch": 0.31,
"learning_rate": 0.00035003009027081245,
"loss": 2.5506,
"theoretical_loss": 3.644673773749786,
"tokens_seen": 1013121024
},
{
"epoch": 0.31,
"learning_rate": 0.00035002006018054163,
"loss": 2.7998,
"theoretical_loss": 3.644651391786139,
"tokens_seen": 1013186560
},
{
"epoch": 0.31,
"learning_rate": 0.0003500100300902708,
"loss": 2.7608,
"theoretical_loss": 3.6446290116755145,
"tokens_seen": 1013252096
},
{
"epoch": 0.31,
"learning_rate": 0.00035,
"loss": 2.7788,
"theoretical_loss": 3.644606633417639,
"tokens_seen": 1013317632
},
{
"epoch": 0.31,
"learning_rate": 0.0003499899699097292,
"loss": 2.8235,
"theoretical_loss": 3.6445842570122395,
"tokens_seen": 1013383168
},
{
"epoch": 0.31,
"learning_rate": 0.00034997993981945836,
"loss": 2.7038,
"theoretical_loss": 3.644561882459043,
"tokens_seen": 1013448704
},
{
"epoch": 0.31,
"learning_rate": 0.0003499699097291876,
"loss": 2.7552,
"theoretical_loss": 3.6445395097577764,
"tokens_seen": 1013514240
},
{
"epoch": 0.31,
"learning_rate": 0.0003499598796389167,
"loss": 2.5695,
"theoretical_loss": 3.6445171389081663,
"tokens_seen": 1013579776
},
{
"epoch": 0.31,
"learning_rate": 0.00034994984954864596,
"loss": 2.6678,
"theoretical_loss": 3.64449476990994,
"tokens_seen": 1013645312
},
{
"epoch": 0.31,
"learning_rate": 0.0003499398194583751,
"loss": 2.9301,
"theoretical_loss": 3.6444724027628252,
"tokens_seen": 1013710848
},
{
"epoch": 0.31,
"learning_rate": 0.0003499297893681043,
"loss": 2.6588,
"theoretical_loss": 3.644450037466548,
"tokens_seen": 1013776384
},
{
"epoch": 0.31,
"learning_rate": 0.0003499197592778335,
"loss": 2.8683,
"theoretical_loss": 3.6444276740208363,
"tokens_seen": 1013841920
},
{
"epoch": 0.31,
"learning_rate": 0.0003499097291875627,
"loss": 2.6186,
"theoretical_loss": 3.6444053124254174,
"tokens_seen": 1013907456
},
{
"epoch": 0.31,
"learning_rate": 0.00034989969909729186,
"loss": 2.6965,
"theoretical_loss": 3.644382952680018,
"tokens_seen": 1013972992
},
{
"epoch": 0.31,
"learning_rate": 0.0003498896690070211,
"loss": 2.6817,
"theoretical_loss": 3.6443605947843665,
"tokens_seen": 1014038528
},
{
"epoch": 0.31,
"learning_rate": 0.0003498796389167502,
"loss": 2.6881,
"theoretical_loss": 3.644338238738189,
"tokens_seen": 1014104064
},
{
"epoch": 0.31,
"objective/train/docs_used": 581599,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7992031574249268,
"objective/train/theoretical_loss": 3.6443158845412142,
"objective/train/tokens_used": 1034629600,
"theoretical_loss": 3.6443158845412142,
"tokens_seen": 1014169600
},
{
"epoch": 0.31,
"learning_rate": 0.00034986960882647946,
"loss": 2.7364,
"theoretical_loss": 3.6443158845412142,
"tokens_seen": 1014169600
},
{
"epoch": 0.31,
"learning_rate": 0.0003498595787362086,
"loss": 2.6882,
"theoretical_loss": 3.6442935321931693,
"tokens_seen": 1014235136
},
{
"epoch": 0.31,
"learning_rate": 0.0003498495486459378,
"loss": 2.5854,
"theoretical_loss": 3.6442711816937816,
"tokens_seen": 1014300672
},
{
"epoch": 0.31,
"learning_rate": 0.00034983951855566706,
"loss": 2.6338,
"theoretical_loss": 3.644248833042779,
"tokens_seen": 1014366208
},
{
"epoch": 0.31,
"learning_rate": 0.0003498294884653962,
"loss": 2.6233,
"theoretical_loss": 3.644226486239889,
"tokens_seen": 1014431744
},
{
"epoch": 0.31,
"learning_rate": 0.0003498194583751254,
"loss": 2.7558,
"theoretical_loss": 3.64420414128484,
"tokens_seen": 1014497280
},
{
"epoch": 0.31,
"learning_rate": 0.00034980942828485455,
"loss": 2.7184,
"theoretical_loss": 3.6441817981773594,
"tokens_seen": 1014562816
},
{
"epoch": 0.31,
"learning_rate": 0.0003497993981945838,
"loss": 2.8104,
"theoretical_loss": 3.6441594569171754,
"tokens_seen": 1014628352
},
{
"epoch": 0.31,
"learning_rate": 0.00034978936810431296,
"loss": 2.6841,
"theoretical_loss": 3.6441371175040156,
"tokens_seen": 1014693888
},
{
"epoch": 0.31,
"learning_rate": 0.00034977933801404214,
"loss": 2.7573,
"theoretical_loss": 3.644114779937608,
"tokens_seen": 1014759424
},
{
"epoch": 0.31,
"learning_rate": 0.0003497693079237713,
"loss": 2.7306,
"theoretical_loss": 3.644092444217681,
"tokens_seen": 1014824960
},
{
"epoch": 0.31,
"learning_rate": 0.00034975927783350056,
"loss": 2.8343,
"theoretical_loss": 3.6440701103439626,
"tokens_seen": 1014890496
},
{
"epoch": 0.31,
"learning_rate": 0.0003497492477432297,
"loss": 2.7811,
"theoretical_loss": 3.644047778316181,
"tokens_seen": 1014956032
},
{
"epoch": 0.31,
"learning_rate": 0.0003497392176529589,
"loss": 2.6442,
"theoretical_loss": 3.644025448134064,
"tokens_seen": 1015021568
},
{
"epoch": 0.31,
"learning_rate": 0.00034972918756268805,
"loss": 2.7827,
"theoretical_loss": 3.644003119797341,
"tokens_seen": 1015087104
},
{
"epoch": 0.31,
"learning_rate": 0.0003497191574724173,
"loss": 2.823,
"theoretical_loss": 3.6439807933057393,
"tokens_seen": 1015152640
},
{
"epoch": 0.31,
"learning_rate": 0.00034970912738214647,
"loss": 2.6504,
"theoretical_loss": 3.643958468658988,
"tokens_seen": 1015218176
},
{
"epoch": 0.31,
"learning_rate": 0.00034969909729187565,
"loss": 2.7059,
"theoretical_loss": 3.6439361458568147,
"tokens_seen": 1015283712
},
{
"epoch": 0.31,
"learning_rate": 0.00034968906720160483,
"loss": 2.6369,
"theoretical_loss": 3.6439138248989495,
"tokens_seen": 1015349248
},
{
"epoch": 0.31,
"learning_rate": 0.000349679037111334,
"loss": 2.5631,
"theoretical_loss": 3.6438915057851196,
"tokens_seen": 1015414784
},
{
"epoch": 0.31,
"learning_rate": 0.0003496690070210632,
"loss": 2.8507,
"theoretical_loss": 3.643869188515054,
"tokens_seen": 1015480320
},
{
"epoch": 0.31,
"learning_rate": 0.0003496589769307924,
"loss": 2.951,
"theoretical_loss": 3.6438468730884814,
"tokens_seen": 1015545856
},
{
"epoch": 0.31,
"learning_rate": 0.00034964894684052155,
"loss": 2.8248,
"theoretical_loss": 3.643824559505131,
"tokens_seen": 1015611392
},
{
"epoch": 0.31,
"learning_rate": 0.0003496389167502508,
"loss": 2.5432,
"theoretical_loss": 3.6438022477647314,
"tokens_seen": 1015676928
},
{
"epoch": 0.31,
"learning_rate": 0.0003496288866599799,
"loss": 2.7165,
"theoretical_loss": 3.643779937867011,
"tokens_seen": 1015742464
},
{
"epoch": 0.31,
"objective/train/docs_used": 582330,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5798532962799072,
"objective/train/theoretical_loss": 3.6437576298116996,
"objective/train/tokens_used": 1036268000,
"theoretical_loss": 3.6437576298116996,
"tokens_seen": 1015808000
},
{
"epoch": 0.31,
"learning_rate": 0.00034961885656970915,
"loss": 2.691,
"theoretical_loss": 3.6437576298116996,
"tokens_seen": 1015808000
},
{
"epoch": 0.31,
"learning_rate": 0.00034960882647943833,
"loss": 2.6215,
"theoretical_loss": 3.6437353235985253,
"tokens_seen": 1015873536
},
{
"epoch": 0.31,
"learning_rate": 0.0003495987963891675,
"loss": 2.6331,
"theoretical_loss": 3.6437130192272176,
"tokens_seen": 1015939072
},
{
"epoch": 0.31,
"learning_rate": 0.0003495887662988967,
"loss": 2.8445,
"theoretical_loss": 3.6436907166975057,
"tokens_seen": 1016004608
},
{
"epoch": 0.31,
"learning_rate": 0.00034957873620862593,
"loss": 2.5605,
"theoretical_loss": 3.6436684160091186,
"tokens_seen": 1016070144
},
{
"epoch": 0.31,
"learning_rate": 0.00034956870611835506,
"loss": 2.5688,
"theoretical_loss": 3.6436461171617864,
"tokens_seen": 1016135680
},
{
"epoch": 0.31,
"learning_rate": 0.0003495586760280843,
"loss": 2.6761,
"theoretical_loss": 3.643623820155237,
"tokens_seen": 1016201216
},
{
"epoch": 0.31,
"learning_rate": 0.0003495486459378134,
"loss": 2.7407,
"theoretical_loss": 3.6436015249892,
"tokens_seen": 1016266752
},
{
"epoch": 0.31,
"learning_rate": 0.00034953861584754265,
"loss": 2.6692,
"theoretical_loss": 3.6435792316634057,
"tokens_seen": 1016332288
},
{
"epoch": 0.31,
"learning_rate": 0.00034952858575727183,
"loss": 2.7898,
"theoretical_loss": 3.6435569401775822,
"tokens_seen": 1016397824
},
{
"epoch": 0.31,
"learning_rate": 0.000349518555667001,
"loss": 2.7532,
"theoretical_loss": 3.6435346505314605,
"tokens_seen": 1016463360
},
{
"epoch": 0.31,
"learning_rate": 0.0003495085255767302,
"loss": 2.6596,
"theoretical_loss": 3.643512362724769,
"tokens_seen": 1016528896
},
{
"epoch": 0.31,
"learning_rate": 0.0003494984954864594,
"loss": 2.7528,
"theoretical_loss": 3.6434900767572382,
"tokens_seen": 1016594432
},
{
"epoch": 0.31,
"learning_rate": 0.00034948846539618856,
"loss": 2.719,
"theoretical_loss": 3.6434677926285977,
"tokens_seen": 1016659968
},
{
"epoch": 0.31,
"learning_rate": 0.0003494784353059178,
"loss": 2.9299,
"theoretical_loss": 3.6434455103385766,
"tokens_seen": 1016725504
},
{
"epoch": 0.31,
"learning_rate": 0.0003494684052156469,
"loss": 2.8396,
"theoretical_loss": 3.6434232298869045,
"tokens_seen": 1016791040
},
{
"epoch": 0.31,
"learning_rate": 0.00034945837512537616,
"loss": 2.8253,
"theoretical_loss": 3.6434009512733123,
"tokens_seen": 1016856576
},
{
"epoch": 0.31,
"learning_rate": 0.0003494483450351053,
"loss": 2.7118,
"theoretical_loss": 3.643378674497529,
"tokens_seen": 1016922112
},
{
"epoch": 0.31,
"learning_rate": 0.0003494383149448345,
"loss": 2.6008,
"theoretical_loss": 3.643356399559286,
"tokens_seen": 1016987648
},
{
"epoch": 0.31,
"learning_rate": 0.0003494282848545637,
"loss": 2.7631,
"theoretical_loss": 3.6433341264583117,
"tokens_seen": 1017053184
},
{
"epoch": 0.31,
"learning_rate": 0.0003494182547642929,
"loss": 2.6783,
"theoretical_loss": 3.6433118551943364,
"tokens_seen": 1017118720
},
{
"epoch": 0.31,
"learning_rate": 0.00034940822467402206,
"loss": 2.61,
"theoretical_loss": 3.643289585767091,
"tokens_seen": 1017184256
},
{
"epoch": 0.31,
"learning_rate": 0.0003493981945837513,
"loss": 2.9815,
"theoretical_loss": 3.6432673181763056,
"tokens_seen": 1017249792
},
{
"epoch": 0.31,
"learning_rate": 0.0003493881644934804,
"loss": 2.8301,
"theoretical_loss": 3.64324505242171,
"tokens_seen": 1017315328
},
{
"epoch": 0.31,
"learning_rate": 0.00034937813440320966,
"loss": 2.4589,
"theoretical_loss": 3.643222788503034,
"tokens_seen": 1017380864
},
{
"epoch": 0.31,
"objective/train/docs_used": 582938,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.8724844455718994,
"objective/train/theoretical_loss": 3.64320052642001,
"objective/train/tokens_used": 1037906400,
"theoretical_loss": 3.64320052642001,
"tokens_seen": 1017446400
},
{
"epoch": 0.31,
"learning_rate": 0.0003493681043129388,
"loss": 2.5999,
"theoretical_loss": 3.64320052642001,
"tokens_seen": 1017446400
},
{
"epoch": 0.31,
"learning_rate": 0.000349358074222668,
"loss": 2.5611,
"theoretical_loss": 3.643178266172366,
"tokens_seen": 1017511936
},
{
"epoch": 0.31,
"learning_rate": 0.0003493480441323972,
"loss": 2.7362,
"theoretical_loss": 3.643156007759834,
"tokens_seen": 1017577472
},
{
"epoch": 0.31,
"learning_rate": 0.0003493380140421264,
"loss": 2.7779,
"theoretical_loss": 3.6431337511821438,
"tokens_seen": 1017643008
},
{
"epoch": 0.31,
"learning_rate": 0.00034932798395185557,
"loss": 2.7461,
"theoretical_loss": 3.643111496439027,
"tokens_seen": 1017708544
},
{
"epoch": 0.31,
"learning_rate": 0.00034931795386158475,
"loss": 2.7905,
"theoretical_loss": 3.6430892435302136,
"tokens_seen": 1017774080
},
{
"epoch": 0.31,
"learning_rate": 0.00034930792377131393,
"loss": 2.8245,
"theoretical_loss": 3.6430669924554335,
"tokens_seen": 1017839616
},
{
"epoch": 0.31,
"learning_rate": 0.00034929789368104316,
"loss": 2.8408,
"theoretical_loss": 3.6430447432144186,
"tokens_seen": 1017905152
},
{
"epoch": 0.31,
"learning_rate": 0.0003492878635907723,
"loss": 2.7008,
"theoretical_loss": 3.6430224958068997,
"tokens_seen": 1017970688
},
{
"epoch": 0.31,
"learning_rate": 0.0003492778335005015,
"loss": 2.7693,
"theoretical_loss": 3.643000250232607,
"tokens_seen": 1018036224
},
{
"epoch": 0.31,
"learning_rate": 0.00034926780341023065,
"loss": 2.939,
"theoretical_loss": 3.642978006491272,
"tokens_seen": 1018101760
},
{
"epoch": 0.31,
"learning_rate": 0.0003492577733199599,
"loss": 2.7411,
"theoretical_loss": 3.6429557645826254,
"tokens_seen": 1018167296
},
{
"epoch": 0.31,
"learning_rate": 0.00034924774322968907,
"loss": 2.8778,
"theoretical_loss": 3.6429335245063985,
"tokens_seen": 1018232832
},
{
"epoch": 0.31,
"learning_rate": 0.00034923771313941825,
"loss": 2.8842,
"theoretical_loss": 3.642911286262322,
"tokens_seen": 1018298368
},
{
"epoch": 0.31,
"learning_rate": 0.00034922768304914743,
"loss": 2.5595,
"theoretical_loss": 3.642889049850128,
"tokens_seen": 1018363904
},
{
"epoch": 0.31,
"learning_rate": 0.00034921765295887667,
"loss": 2.6854,
"theoretical_loss": 3.6428668152695467,
"tokens_seen": 1018429440
},
{
"epoch": 0.31,
"learning_rate": 0.0003492076228686058,
"loss": 2.5677,
"theoretical_loss": 3.6428445825203095,
"tokens_seen": 1018494976
},
{
"epoch": 0.31,
"learning_rate": 0.00034919759277833503,
"loss": 2.6218,
"theoretical_loss": 3.642822351602148,
"tokens_seen": 1018560512
},
{
"epoch": 0.31,
"learning_rate": 0.00034918756268806416,
"loss": 2.7141,
"theoretical_loss": 3.6428001225147937,
"tokens_seen": 1018626048
},
{
"epoch": 0.31,
"learning_rate": 0.0003491775325977934,
"loss": 2.656,
"theoretical_loss": 3.642777895257978,
"tokens_seen": 1018691584
},
{
"epoch": 0.31,
"learning_rate": 0.00034916750250752257,
"loss": 2.6184,
"theoretical_loss": 3.6427556698314323,
"tokens_seen": 1018757120
},
{
"epoch": 0.31,
"learning_rate": 0.00034915747241725175,
"loss": 2.5637,
"theoretical_loss": 3.6427334462348884,
"tokens_seen": 1018822656
},
{
"epoch": 0.31,
"learning_rate": 0.00034914744232698093,
"loss": 2.6612,
"theoretical_loss": 3.6427112244680773,
"tokens_seen": 1018888192
},
{
"epoch": 0.31,
"learning_rate": 0.0003491374122367101,
"loss": 2.254,
"theoretical_loss": 3.642689004530731,
"tokens_seen": 1018953728
},
{
"epoch": 0.31,
"learning_rate": 0.0003491273821464393,
"loss": 2.7307,
"theoretical_loss": 3.642666786422582,
"tokens_seen": 1019019264
},
{
"epoch": 0.31,
"objective/train/docs_used": 584109,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8709795475006104,
"objective/train/theoretical_loss": 3.6426445701433607,
"objective/train/tokens_used": 1039544800,
"theoretical_loss": 3.6426445701433607,
"tokens_seen": 1019084800
},
{
"epoch": 0.31,
"learning_rate": 0.00034911735205616853,
"loss": 2.7546,
"theoretical_loss": 3.6426445701433607,
"tokens_seen": 1019084800
},
{
"epoch": 0.31,
"learning_rate": 0.00034910732196589766,
"loss": 2.6998,
"theoretical_loss": 3.6426223556928,
"tokens_seen": 1019150336
},
{
"epoch": 0.31,
"learning_rate": 0.0003490972918756269,
"loss": 2.7981,
"theoretical_loss": 3.6426001430706316,
"tokens_seen": 1019215872
},
{
"epoch": 0.31,
"learning_rate": 0.00034908726178535613,
"loss": 2.7559,
"theoretical_loss": 3.642577932276587,
"tokens_seen": 1019281408
},
{
"epoch": 0.31,
"learning_rate": 0.00034907723169508526,
"loss": 2.4275,
"theoretical_loss": 3.642555723310399,
"tokens_seen": 1019346944
},
{
"epoch": 0.31,
"learning_rate": 0.0003490672016048145,
"loss": 2.6416,
"theoretical_loss": 3.642533516171799,
"tokens_seen": 1019412480
},
{
"epoch": 0.31,
"learning_rate": 0.0003490571715145436,
"loss": 2.8139,
"theoretical_loss": 3.6425113108605194,
"tokens_seen": 1019478016
},
{
"epoch": 0.31,
"learning_rate": 0.00034904714142427285,
"loss": 2.5846,
"theoretical_loss": 3.642489107376292,
"tokens_seen": 1019543552
},
{
"epoch": 0.31,
"learning_rate": 0.00034903711133400203,
"loss": 2.7267,
"theoretical_loss": 3.6424669057188495,
"tokens_seen": 1019609088
},
{
"epoch": 0.31,
"learning_rate": 0.0003490270812437312,
"loss": 2.8604,
"theoretical_loss": 3.6424447058879243,
"tokens_seen": 1019674624
},
{
"epoch": 0.31,
"learning_rate": 0.0003490170511534604,
"loss": 2.6626,
"theoretical_loss": 3.642422507883248,
"tokens_seen": 1019740160
},
{
"epoch": 0.31,
"learning_rate": 0.0003490070210631896,
"loss": 2.6812,
"theoretical_loss": 3.642400311704554,
"tokens_seen": 1019805696
},
{
"epoch": 0.31,
"learning_rate": 0.00034899699097291876,
"loss": 2.6994,
"theoretical_loss": 3.6423781173515746,
"tokens_seen": 1019871232
},
{
"epoch": 0.31,
"learning_rate": 0.000348986960882648,
"loss": 2.819,
"theoretical_loss": 3.6423559248240416,
"tokens_seen": 1019936768
},
{
"epoch": 0.31,
"learning_rate": 0.0003489769307923771,
"loss": 2.7941,
"theoretical_loss": 3.6423337341216877,
"tokens_seen": 1020002304
},
{
"epoch": 0.31,
"learning_rate": 0.00034896690070210636,
"loss": 2.7381,
"theoretical_loss": 3.6423115452442465,
"tokens_seen": 1020067840
},
{
"epoch": 0.31,
"learning_rate": 0.0003489568706118355,
"loss": 2.5756,
"theoretical_loss": 3.6422893581914497,
"tokens_seen": 1020133376
},
{
"epoch": 0.31,
"learning_rate": 0.0003489468405215647,
"loss": 2.7073,
"theoretical_loss": 3.64226717296303,
"tokens_seen": 1020198912
},
{
"epoch": 0.31,
"learning_rate": 0.0003489368104312939,
"loss": 2.7763,
"theoretical_loss": 3.642244989558721,
"tokens_seen": 1020264448
},
{
"epoch": 0.31,
"learning_rate": 0.0003489267803410231,
"loss": 2.9399,
"theoretical_loss": 3.6422228079782553,
"tokens_seen": 1020329984
},
{
"epoch": 0.31,
"learning_rate": 0.00034891675025075226,
"loss": 2.8106,
"theoretical_loss": 3.6422006282213655,
"tokens_seen": 1020395520
},
{
"epoch": 0.31,
"learning_rate": 0.0003489067201604815,
"loss": 2.5296,
"theoretical_loss": 3.642178450287785,
"tokens_seen": 1020461056
},
{
"epoch": 0.31,
"learning_rate": 0.0003488966900702106,
"loss": 2.467,
"theoretical_loss": 3.642156274177246,
"tokens_seen": 1020526592
},
{
"epoch": 0.31,
"learning_rate": 0.00034888665997993986,
"loss": 2.6134,
"theoretical_loss": 3.642134099889483,
"tokens_seen": 1020592128
},
{
"epoch": 0.31,
"learning_rate": 0.000348876629889669,
"loss": 2.6419,
"theoretical_loss": 3.642111927424228,
"tokens_seen": 1020657664
},
{
"epoch": 0.31,
"objective/train/docs_used": 584789,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9844183921813965,
"objective/train/theoretical_loss": 3.642089756781214,
"objective/train/tokens_used": 1041183200,
"theoretical_loss": 3.642089756781214,
"tokens_seen": 1020723200
},
{
"epoch": 0.31,
"learning_rate": 0.0003488665997993982,
"loss": 2.814,
"theoretical_loss": 3.642089756781214,
"tokens_seen": 1020723200
},
{
"epoch": 0.31,
"learning_rate": 0.0003488565697091274,
"loss": 2.7047,
"theoretical_loss": 3.642067587960175,
"tokens_seen": 1020788736
},
{
"epoch": 0.31,
"learning_rate": 0.0003488465396188566,
"loss": 2.7754,
"theoretical_loss": 3.6420454209608444,
"tokens_seen": 1020854272
},
{
"epoch": 0.31,
"learning_rate": 0.00034883650952858577,
"loss": 2.6558,
"theoretical_loss": 3.6420232557829553,
"tokens_seen": 1020919808
},
{
"epoch": 0.31,
"learning_rate": 0.00034882647943831495,
"loss": 2.7028,
"theoretical_loss": 3.6420010924262405,
"tokens_seen": 1020985344
},
{
"epoch": 0.31,
"learning_rate": 0.00034881644934804413,
"loss": 2.707,
"theoretical_loss": 3.6419789308904345,
"tokens_seen": 1021050880
},
{
"epoch": 0.31,
"learning_rate": 0.00034880641925777336,
"loss": 2.5738,
"theoretical_loss": 3.6419567711752703,
"tokens_seen": 1021116416
},
{
"epoch": 0.31,
"learning_rate": 0.0003487963891675025,
"loss": 2.6711,
"theoretical_loss": 3.6419346132804815,
"tokens_seen": 1021181952
},
{
"epoch": 0.31,
"learning_rate": 0.0003487863590772317,
"loss": 2.4474,
"theoretical_loss": 3.641912457205802,
"tokens_seen": 1021247488
},
{
"epoch": 0.31,
"learning_rate": 0.00034877632898696085,
"loss": 2.7671,
"theoretical_loss": 3.6418903029509653,
"tokens_seen": 1021313024
},
{
"epoch": 0.31,
"learning_rate": 0.0003487662988966901,
"loss": 2.6037,
"theoretical_loss": 3.6418681505157053,
"tokens_seen": 1021378560
},
{
"epoch": 0.31,
"learning_rate": 0.00034875626880641927,
"loss": 2.8539,
"theoretical_loss": 3.641845999899755,
"tokens_seen": 1021444096
},
{
"epoch": 0.31,
"learning_rate": 0.00034874623871614845,
"loss": 2.6638,
"theoretical_loss": 3.64182385110285,
"tokens_seen": 1021509632
},
{
"epoch": 0.31,
"learning_rate": 0.00034873620862587763,
"loss": 2.8181,
"theoretical_loss": 3.6418017041247226,
"tokens_seen": 1021575168
},
{
"epoch": 0.31,
"learning_rate": 0.00034872617853560687,
"loss": 2.7189,
"theoretical_loss": 3.6417795589651076,
"tokens_seen": 1021640704
},
{
"epoch": 0.31,
"learning_rate": 0.000348716148445336,
"loss": 2.7544,
"theoretical_loss": 3.6417574156237387,
"tokens_seen": 1021706240
},
{
"epoch": 0.31,
"learning_rate": 0.00034870611835506523,
"loss": 2.6909,
"theoretical_loss": 3.64173527410035,
"tokens_seen": 1021771776
},
{
"epoch": 0.31,
"learning_rate": 0.00034869608826479436,
"loss": 2.7898,
"theoretical_loss": 3.641713134394676,
"tokens_seen": 1021837312
},
{
"epoch": 0.31,
"learning_rate": 0.0003486860581745236,
"loss": 2.6752,
"theoretical_loss": 3.6416909965064512,
"tokens_seen": 1021902848
},
{
"epoch": 0.31,
"learning_rate": 0.00034867602808425277,
"loss": 2.7605,
"theoretical_loss": 3.6416688604354084,
"tokens_seen": 1021968384
},
{
"epoch": 0.31,
"learning_rate": 0.00034866599799398195,
"loss": 2.8171,
"theoretical_loss": 3.6416467261812837,
"tokens_seen": 1022033920
},
{
"epoch": 0.31,
"learning_rate": 0.00034865596790371113,
"loss": 2.6379,
"theoretical_loss": 3.64162459374381,
"tokens_seen": 1022099456
},
{
"epoch": 0.31,
"learning_rate": 0.0003486459378134403,
"loss": 2.6961,
"theoretical_loss": 3.641602463122723,
"tokens_seen": 1022164992
},
{
"epoch": 0.31,
"learning_rate": 0.0003486359077231695,
"loss": 2.6299,
"theoretical_loss": 3.641580334317756,
"tokens_seen": 1022230528
},
{
"epoch": 0.31,
"learning_rate": 0.00034862587763289873,
"loss": 2.8477,
"theoretical_loss": 3.6415582073286448,
"tokens_seen": 1022296064
},
{
"epoch": 0.31,
"objective/train/docs_used": 586155,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7868759632110596,
"objective/train/theoretical_loss": 3.6415360821551226,
"objective/train/tokens_used": 1042821600,
"theoretical_loss": 3.6415360821551226,
"tokens_seen": 1022361600
},
{
"epoch": 0.31,
"learning_rate": 0.00034861584754262786,
"loss": 2.6661,
"theoretical_loss": 3.6415360821551226,
"tokens_seen": 1022361600
},
{
"epoch": 0.31,
"learning_rate": 0.0003486058174523571,
"loss": 2.9048,
"theoretical_loss": 3.641513958796925,
"tokens_seen": 1022427136
},
{
"epoch": 0.31,
"learning_rate": 0.0003485957873620862,
"loss": 2.4171,
"theoretical_loss": 3.6414918372537866,
"tokens_seen": 1022492672
},
{
"epoch": 0.31,
"learning_rate": 0.00034858575727181546,
"loss": 2.7446,
"theoretical_loss": 3.6414697175254425,
"tokens_seen": 1022558208
},
{
"epoch": 0.31,
"learning_rate": 0.00034857572718154464,
"loss": 2.7082,
"theoretical_loss": 3.6414475996116265,
"tokens_seen": 1022623744
},
{
"epoch": 0.31,
"learning_rate": 0.0003485656970912738,
"loss": 2.7745,
"theoretical_loss": 3.641425483512074,
"tokens_seen": 1022689280
},
{
"epoch": 0.31,
"learning_rate": 0.000348555667001003,
"loss": 2.7055,
"theoretical_loss": 3.6414033692265204,
"tokens_seen": 1022754816
},
{
"epoch": 0.31,
"learning_rate": 0.00034854563691073223,
"loss": 2.7589,
"theoretical_loss": 3.6413812567547,
"tokens_seen": 1022820352
},
{
"epoch": 0.31,
"learning_rate": 0.00034853560682046136,
"loss": 2.6822,
"theoretical_loss": 3.641359146096348,
"tokens_seen": 1022885888
},
{
"epoch": 0.31,
"learning_rate": 0.0003485255767301906,
"loss": 2.6696,
"theoretical_loss": 3.6413370372512,
"tokens_seen": 1022951424
},
{
"epoch": 0.31,
"learning_rate": 0.0003485155466399197,
"loss": 2.6209,
"theoretical_loss": 3.6413149302189907,
"tokens_seen": 1023016960
},
{
"epoch": 0.31,
"learning_rate": 0.00034850551654964896,
"loss": 2.6994,
"theoretical_loss": 3.6412928249994554,
"tokens_seen": 1023082496
},
{
"epoch": 0.31,
"learning_rate": 0.00034849548645937814,
"loss": 2.7902,
"theoretical_loss": 3.6412707215923295,
"tokens_seen": 1023148032
},
{
"epoch": 0.31,
"learning_rate": 0.0003484854563691073,
"loss": 2.5874,
"theoretical_loss": 3.6412486199973477,
"tokens_seen": 1023213568
},
{
"epoch": 0.31,
"learning_rate": 0.0003484754262788365,
"loss": 2.5051,
"theoretical_loss": 3.641226520214246,
"tokens_seen": 1023279104
},
{
"epoch": 0.31,
"learning_rate": 0.0003484653961885657,
"loss": 2.6376,
"theoretical_loss": 3.64120442224276,
"tokens_seen": 1023344640
},
{
"epoch": 0.31,
"learning_rate": 0.00034845536609829486,
"loss": 2.5984,
"theoretical_loss": 3.6411823260826246,
"tokens_seen": 1023410176
},
{
"epoch": 0.31,
"learning_rate": 0.0003484453360080241,
"loss": 2.7222,
"theoretical_loss": 3.6411602317335756,
"tokens_seen": 1023475712
},
{
"epoch": 0.31,
"learning_rate": 0.0003484353059177532,
"loss": 2.4275,
"theoretical_loss": 3.6411381391953492,
"tokens_seen": 1023541248
},
{
"epoch": 0.31,
"learning_rate": 0.00034842527582748246,
"loss": 2.4543,
"theoretical_loss": 3.64111604846768,
"tokens_seen": 1023606784
},
{
"epoch": 0.31,
"learning_rate": 0.00034841524573721164,
"loss": 2.5276,
"theoretical_loss": 3.6410939595503047,
"tokens_seen": 1023672320
},
{
"epoch": 0.31,
"learning_rate": 0.0003484052156469408,
"loss": 2.4026,
"theoretical_loss": 3.641071872442958,
"tokens_seen": 1023737856
},
{
"epoch": 0.31,
"learning_rate": 0.00034839518555667,
"loss": 2.5717,
"theoretical_loss": 3.641049787145376,
"tokens_seen": 1023803392
},
{
"epoch": 0.31,
"learning_rate": 0.0003483851554663992,
"loss": 2.7724,
"theoretical_loss": 3.641027703657296,
"tokens_seen": 1023868928
},
{
"epoch": 0.31,
"learning_rate": 0.00034837512537612837,
"loss": 2.8711,
"theoretical_loss": 3.6410056219784526,
"tokens_seen": 1023934464
},
{
"epoch": 0.31,
"objective/train/docs_used": 586672,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.843909502029419,
"objective/train/theoretical_loss": 3.6409835421085814,
"objective/train/tokens_used": 1044460000,
"theoretical_loss": 3.6409835421085814,
"tokens_seen": 1024000000
},
{
"epoch": 0.31,
"learning_rate": 0.0003483650952858576,
"loss": 2.7892,
"theoretical_loss": 3.6409835421085814,
"tokens_seen": 1024000000
},
{
"epoch": 0.31,
"learning_rate": 0.00034835506519558673,
"loss": 2.8256,
"theoretical_loss": 3.640961464047419,
"tokens_seen": 1024065536
},
{
"epoch": 0.31,
"learning_rate": 0.00034834503510531597,
"loss": 2.7922,
"theoretical_loss": 3.640939387794702,
"tokens_seen": 1024131072
},
{
"epoch": 0.31,
"learning_rate": 0.00034833500501504515,
"loss": 2.6516,
"theoretical_loss": 3.640917313350166,
"tokens_seen": 1024196608
},
{
"epoch": 0.31,
"learning_rate": 0.00034832497492477433,
"loss": 2.4996,
"theoretical_loss": 3.6408952407135473,
"tokens_seen": 1024262144
},
{
"epoch": 0.31,
"learning_rate": 0.00034831494483450356,
"loss": 2.6662,
"theoretical_loss": 3.640873169884582,
"tokens_seen": 1024327680
},
{
"epoch": 0.31,
"learning_rate": 0.0003483049147442327,
"loss": 2.7156,
"theoretical_loss": 3.6408511008630073,
"tokens_seen": 1024393216
},
{
"epoch": 0.31,
"learning_rate": 0.0003482948846539619,
"loss": 2.4559,
"theoretical_loss": 3.6408290336485583,
"tokens_seen": 1024458752
},
{
"epoch": 0.31,
"learning_rate": 0.00034828485456369105,
"loss": 2.3368,
"theoretical_loss": 3.6408069682409723,
"tokens_seen": 1024524288
},
{
"epoch": 0.31,
"learning_rate": 0.0003482748244734203,
"loss": 2.6698,
"theoretical_loss": 3.640784904639985,
"tokens_seen": 1024589824
},
{
"epoch": 0.31,
"learning_rate": 0.00034826479438314947,
"loss": 2.7541,
"theoretical_loss": 3.640762842845334,
"tokens_seen": 1024655360
},
{
"epoch": 0.31,
"learning_rate": 0.00034825476429287865,
"loss": 2.7042,
"theoretical_loss": 3.6407407828567555,
"tokens_seen": 1024720896
},
{
"epoch": 0.31,
"learning_rate": 0.00034824473420260783,
"loss": 2.8388,
"theoretical_loss": 3.6407187246739854,
"tokens_seen": 1024786432
},
{
"epoch": 0.31,
"learning_rate": 0.00034823470411233707,
"loss": 2.7813,
"theoretical_loss": 3.640696668296761,
"tokens_seen": 1024851968
},
{
"epoch": 0.31,
"learning_rate": 0.0003482246740220662,
"loss": 2.766,
"theoretical_loss": 3.6406746137248196,
"tokens_seen": 1024917504
},
{
"epoch": 0.31,
"learning_rate": 0.00034821464393179543,
"loss": 2.5837,
"theoretical_loss": 3.6406525609578972,
"tokens_seen": 1024983040
},
{
"epoch": 0.31,
"learning_rate": 0.00034820461384152456,
"loss": 2.8018,
"theoretical_loss": 3.6406305099957312,
"tokens_seen": 1025048576
},
{
"epoch": 0.31,
"learning_rate": 0.0003481945837512538,
"loss": 2.927,
"theoretical_loss": 3.6406084608380582,
"tokens_seen": 1025114112
},
{
"epoch": 0.31,
"learning_rate": 0.00034818455366098297,
"loss": 2.7055,
"theoretical_loss": 3.6405864134846153,
"tokens_seen": 1025179648
},
{
"epoch": 0.31,
"learning_rate": 0.00034817452357071215,
"loss": 2.9929,
"theoretical_loss": 3.6405643679351396,
"tokens_seen": 1025245184
},
{
"epoch": 0.31,
"learning_rate": 0.00034816449348044133,
"loss": 2.4391,
"theoretical_loss": 3.6405423241893677,
"tokens_seen": 1025310720
},
{
"epoch": 0.31,
"learning_rate": 0.0003481544633901705,
"loss": 2.7304,
"theoretical_loss": 3.6405202822470377,
"tokens_seen": 1025376256
},
{
"epoch": 0.31,
"learning_rate": 0.0003481444332998997,
"loss": 2.6821,
"theoretical_loss": 3.6404982421078858,
"tokens_seen": 1025441792
},
{
"epoch": 0.31,
"learning_rate": 0.00034813440320962893,
"loss": 2.6489,
"theoretical_loss": 3.6404762037716503,
"tokens_seen": 1025507328
},
{
"epoch": 0.31,
"learning_rate": 0.00034812437311935806,
"loss": 2.8529,
"theoretical_loss": 3.640454167238068,
"tokens_seen": 1025572864
},
{
"epoch": 0.31,
"objective/train/docs_used": 587795,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.533445119857788,
"objective/train/theoretical_loss": 3.6404321325068754,
"objective/train/tokens_used": 1046098400,
"theoretical_loss": 3.6404321325068754,
"tokens_seen": 1025638400
},
{
"epoch": 0.31,
"learning_rate": 0.0003481143430290873,
"loss": 2.5025,
"theoretical_loss": 3.6404321325068754,
"tokens_seen": 1025638400
},
{
"epoch": 0.31,
"learning_rate": 0.0003481043129388164,
"loss": 2.9398,
"theoretical_loss": 3.6404100995778115,
"tokens_seen": 1025703936
},
{
"epoch": 0.31,
"learning_rate": 0.00034809428284854566,
"loss": 2.6935,
"theoretical_loss": 3.640388068450613,
"tokens_seen": 1025769472
},
{
"epoch": 0.31,
"learning_rate": 0.00034808425275827484,
"loss": 2.7919,
"theoretical_loss": 3.640366039125017,
"tokens_seen": 1025835008
},
{
"epoch": 0.31,
"learning_rate": 0.000348074222668004,
"loss": 2.6769,
"theoretical_loss": 3.640344011600762,
"tokens_seen": 1025900544
},
{
"epoch": 0.31,
"learning_rate": 0.0003480641925777332,
"loss": 2.884,
"theoretical_loss": 3.640321985877585,
"tokens_seen": 1025966080
},
{
"epoch": 0.31,
"learning_rate": 0.00034805416248746243,
"loss": 2.6561,
"theoretical_loss": 3.6402999619552245,
"tokens_seen": 1026031616
},
{
"epoch": 0.31,
"learning_rate": 0.00034804413239719156,
"loss": 2.6806,
"theoretical_loss": 3.640277939833417,
"tokens_seen": 1026097152
},
{
"epoch": 0.31,
"learning_rate": 0.0003480341023069208,
"loss": 2.5546,
"theoretical_loss": 3.640255919511901,
"tokens_seen": 1026162688
},
{
"epoch": 0.31,
"learning_rate": 0.0003480240722166499,
"loss": 2.8009,
"theoretical_loss": 3.640233900990415,
"tokens_seen": 1026228224
},
{
"epoch": 0.31,
"learning_rate": 0.00034801404212637916,
"loss": 2.7283,
"theoretical_loss": 3.6402118842686955,
"tokens_seen": 1026293760
},
{
"epoch": 0.31,
"learning_rate": 0.00034800401203610834,
"loss": 2.4986,
"theoretical_loss": 3.6401898693464814,
"tokens_seen": 1026359296
},
{
"epoch": 0.31,
"learning_rate": 0.0003479939819458375,
"loss": 2.8072,
"theoretical_loss": 3.6401678562235107,
"tokens_seen": 1026424832
},
{
"epoch": 0.31,
"learning_rate": 0.0003479839518555667,
"loss": 2.7235,
"theoretical_loss": 3.6401458448995214,
"tokens_seen": 1026490368
},
{
"epoch": 0.31,
"learning_rate": 0.0003479739217652959,
"loss": 2.6855,
"theoretical_loss": 3.6401238353742515,
"tokens_seen": 1026555904
},
{
"epoch": 0.31,
"learning_rate": 0.00034796389167502506,
"loss": 2.5712,
"theoretical_loss": 3.6401018276474395,
"tokens_seen": 1026621440
},
{
"epoch": 0.31,
"learning_rate": 0.0003479538615847543,
"loss": 2.9763,
"theoretical_loss": 3.6400798217188233,
"tokens_seen": 1026686976
},
{
"epoch": 0.31,
"learning_rate": 0.00034794383149448343,
"loss": 2.7002,
"theoretical_loss": 3.6400578175881417,
"tokens_seen": 1026752512
},
{
"epoch": 0.31,
"learning_rate": 0.00034793380140421266,
"loss": 2.509,
"theoretical_loss": 3.640035815255132,
"tokens_seen": 1026818048
},
{
"epoch": 0.31,
"learning_rate": 0.00034792377131394184,
"loss": 2.5622,
"theoretical_loss": 3.640013814719534,
"tokens_seen": 1026883584
},
{
"epoch": 0.31,
"learning_rate": 0.000347913741223671,
"loss": 2.7858,
"theoretical_loss": 3.6399918159810847,
"tokens_seen": 1026949120
},
{
"epoch": 0.31,
"learning_rate": 0.0003479037111334002,
"loss": 2.5976,
"theoretical_loss": 3.639969819039524,
"tokens_seen": 1027014656
},
{
"epoch": 0.31,
"learning_rate": 0.0003478936810431294,
"loss": 2.7576,
"theoretical_loss": 3.63994782389459,
"tokens_seen": 1027080192
},
{
"epoch": 0.31,
"learning_rate": 0.00034788365095285857,
"loss": 2.8097,
"theoretical_loss": 3.639925830546021,
"tokens_seen": 1027145728
},
{
"epoch": 0.31,
"learning_rate": 0.0003478736208625878,
"loss": 2.7113,
"theoretical_loss": 3.639903838993556,
"tokens_seen": 1027211264
},
{
"epoch": 0.31,
"objective/train/docs_used": 588595,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.8003904819488525,
"objective/train/theoretical_loss": 3.6398818492369336,
"objective/train/tokens_used": 1047736800,
"theoretical_loss": 3.6398818492369336,
"tokens_seen": 1027276800
},
{
"epoch": 0.31,
"learning_rate": 0.00034786359077231693,
"loss": 2.3251,
"theoretical_loss": 3.6398818492369336,
"tokens_seen": 1027276800
},
{
"epoch": 0.31,
"learning_rate": 0.00034785356068204617,
"loss": 2.7813,
"theoretical_loss": 3.639859861275893,
"tokens_seen": 1027342336
},
{
"epoch": 0.31,
"learning_rate": 0.0003478435305917753,
"loss": 2.8883,
"theoretical_loss": 3.639837875110172,
"tokens_seen": 1027407872
},
{
"epoch": 0.31,
"learning_rate": 0.00034783350050150453,
"loss": 2.7609,
"theoretical_loss": 3.6398158907395106,
"tokens_seen": 1027473408
},
{
"epoch": 0.31,
"learning_rate": 0.0003478234704112337,
"loss": 2.8853,
"theoretical_loss": 3.6397939081636475,
"tokens_seen": 1027538944
},
{
"epoch": 0.31,
"learning_rate": 0.0003478134403209629,
"loss": 2.9025,
"theoretical_loss": 3.6397719273823212,
"tokens_seen": 1027604480
},
{
"epoch": 0.31,
"learning_rate": 0.00034780341023069207,
"loss": 2.6823,
"theoretical_loss": 3.6397499483952718,
"tokens_seen": 1027670016
},
{
"epoch": 0.31,
"learning_rate": 0.00034779338014042125,
"loss": 2.7099,
"theoretical_loss": 3.639727971202237,
"tokens_seen": 1027735552
},
{
"epoch": 0.31,
"learning_rate": 0.00034778335005015043,
"loss": 2.7471,
"theoretical_loss": 3.6397059958029576,
"tokens_seen": 1027801088
},
{
"epoch": 0.31,
"learning_rate": 0.00034777331995987967,
"loss": 2.7257,
"theoretical_loss": 3.639684022197171,
"tokens_seen": 1027866624
},
{
"epoch": 0.31,
"learning_rate": 0.0003477632898696088,
"loss": 2.6031,
"theoretical_loss": 3.639662050384618,
"tokens_seen": 1027932160
},
{
"epoch": 0.31,
"learning_rate": 0.00034775325977933803,
"loss": 2.5162,
"theoretical_loss": 3.6396400803650377,
"tokens_seen": 1027997696
},
{
"epoch": 0.31,
"learning_rate": 0.0003477432296890672,
"loss": 2.6341,
"theoretical_loss": 3.639618112138169,
"tokens_seen": 1028063232
},
{
"epoch": 0.31,
"learning_rate": 0.0003477331995987964,
"loss": 2.6251,
"theoretical_loss": 3.639596145703752,
"tokens_seen": 1028128768
},
{
"epoch": 0.31,
"learning_rate": 0.0003477231695085256,
"loss": 2.6078,
"theoretical_loss": 3.6395741810615254,
"tokens_seen": 1028194304
},
{
"epoch": 0.31,
"learning_rate": 0.00034771313941825476,
"loss": 2.8629,
"theoretical_loss": 3.6395522182112288,
"tokens_seen": 1028259840
},
{
"epoch": 0.31,
"learning_rate": 0.00034770310932798394,
"loss": 2.4741,
"theoretical_loss": 3.6395302571526025,
"tokens_seen": 1028325376
},
{
"epoch": 0.31,
"learning_rate": 0.00034769307923771317,
"loss": 2.8197,
"theoretical_loss": 3.639508297885386,
"tokens_seen": 1028390912
},
{
"epoch": 0.31,
"learning_rate": 0.0003476830491474423,
"loss": 2.5807,
"theoretical_loss": 3.6394863404093183,
"tokens_seen": 1028456448
},
{
"epoch": 0.31,
"learning_rate": 0.00034767301905717153,
"loss": 2.6001,
"theoretical_loss": 3.6394643847241404,
"tokens_seen": 1028521984
},
{
"epoch": 0.31,
"learning_rate": 0.00034766298896690066,
"loss": 2.8089,
"theoretical_loss": 3.6394424308295914,
"tokens_seen": 1028587520
},
{
"epoch": 0.31,
"learning_rate": 0.0003476529588766299,
"loss": 2.7217,
"theoretical_loss": 3.6394204787254107,
"tokens_seen": 1028653056
},
{
"epoch": 0.31,
"learning_rate": 0.0003476429287863591,
"loss": 2.6235,
"theoretical_loss": 3.6393985284113395,
"tokens_seen": 1028718592
},
{
"epoch": 0.31,
"learning_rate": 0.00034763289869608826,
"loss": 2.7296,
"theoretical_loss": 3.6393765798871165,
"tokens_seen": 1028784128
},
{
"epoch": 0.31,
"learning_rate": 0.00034762286860581744,
"loss": 2.795,
"theoretical_loss": 3.6393546331524824,
"tokens_seen": 1028849664
},
{
"epoch": 0.31,
"objective/train/docs_used": 589927,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3315000534057617,
"objective/train/theoretical_loss": 3.639332688207178,
"objective/train/tokens_used": 1049375200,
"theoretical_loss": 3.639332688207178,
"tokens_seen": 1028915200
},
{
"epoch": 0.31,
"learning_rate": 0.0003476128385155466,
"loss": 2.4643,
"theoretical_loss": 3.639332688207178,
"tokens_seen": 1028915200
},
{
"epoch": 0.31,
"learning_rate": 0.0003476028084252758,
"loss": 2.6326,
"theoretical_loss": 3.639310745050942,
"tokens_seen": 1028980736
},
{
"epoch": 0.31,
"learning_rate": 0.00034759277833500504,
"loss": 2.7926,
"theoretical_loss": 3.639288803683515,
"tokens_seen": 1029046272
},
{
"epoch": 0.31,
"learning_rate": 0.0003475827482447342,
"loss": 2.749,
"theoretical_loss": 3.639266864104638,
"tokens_seen": 1029111808
},
{
"epoch": 0.31,
"learning_rate": 0.0003475727181544634,
"loss": 2.7039,
"theoretical_loss": 3.639244926314051,
"tokens_seen": 1029177344
},
{
"epoch": 0.31,
"learning_rate": 0.00034756268806419263,
"loss": 2.7149,
"theoretical_loss": 3.6392229903114943,
"tokens_seen": 1029242880
},
{
"epoch": 0.31,
"learning_rate": 0.00034755265797392176,
"loss": 2.9641,
"theoretical_loss": 3.6392010560967085,
"tokens_seen": 1029308416
},
{
"epoch": 0.31,
"learning_rate": 0.000347542627883651,
"loss": 2.625,
"theoretical_loss": 3.6391791236694333,
"tokens_seen": 1029373952
},
{
"epoch": 0.31,
"learning_rate": 0.0003475325977933801,
"loss": 2.7873,
"theoretical_loss": 3.63915719302941,
"tokens_seen": 1029439488
},
{
"epoch": 0.31,
"learning_rate": 0.00034752256770310936,
"loss": 2.7307,
"theoretical_loss": 3.6391352641763794,
"tokens_seen": 1029505024
},
{
"epoch": 0.31,
"learning_rate": 0.00034751253761283854,
"loss": 2.4786,
"theoretical_loss": 3.6391133371100817,
"tokens_seen": 1029570560
},
{
"epoch": 0.31,
"learning_rate": 0.0003475025075225677,
"loss": 2.8477,
"theoretical_loss": 3.6390914118302575,
"tokens_seen": 1029636096
},
{
"epoch": 0.31,
"learning_rate": 0.0003474924774322969,
"loss": 2.7595,
"theoretical_loss": 3.639069488336648,
"tokens_seen": 1029701632
},
{
"epoch": 0.31,
"learning_rate": 0.0003474824473420261,
"loss": 2.4965,
"theoretical_loss": 3.6390475666289936,
"tokens_seen": 1029767168
},
{
"epoch": 0.31,
"learning_rate": 0.00034747241725175527,
"loss": 2.607,
"theoretical_loss": 3.639025646707035,
"tokens_seen": 1029832704
},
{
"epoch": 0.31,
"learning_rate": 0.0003474623871614845,
"loss": 2.7025,
"theoretical_loss": 3.639003728570514,
"tokens_seen": 1029898240
},
{
"epoch": 0.31,
"learning_rate": 0.00034745235707121363,
"loss": 2.6771,
"theoretical_loss": 3.638981812219171,
"tokens_seen": 1029963776
},
{
"epoch": 0.31,
"learning_rate": 0.00034744232698094286,
"loss": 2.7801,
"theoretical_loss": 3.6389598976527466,
"tokens_seen": 1030029312
},
{
"epoch": 0.31,
"learning_rate": 0.00034743229689067204,
"loss": 2.6717,
"theoretical_loss": 3.6389379848709824,
"tokens_seen": 1030094848
},
{
"epoch": 0.31,
"learning_rate": 0.0003474222668004012,
"loss": 2.4826,
"theoretical_loss": 3.6389160738736193,
"tokens_seen": 1030160384
},
{
"epoch": 0.31,
"learning_rate": 0.0003474122367101304,
"loss": 2.8287,
"theoretical_loss": 3.638894164660399,
"tokens_seen": 1030225920
},
{
"epoch": 0.31,
"learning_rate": 0.0003474022066198596,
"loss": 2.6661,
"theoretical_loss": 3.6388722572310623,
"tokens_seen": 1030291456
},
{
"epoch": 0.31,
"learning_rate": 0.00034739217652958877,
"loss": 2.6881,
"theoretical_loss": 3.6388503515853503,
"tokens_seen": 1030356992
},
{
"epoch": 0.31,
"learning_rate": 0.000347382146439318,
"loss": 2.6919,
"theoretical_loss": 3.6388284477230046,
"tokens_seen": 1030422528
},
{
"epoch": 0.31,
"learning_rate": 0.00034737211634904713,
"loss": 2.7771,
"theoretical_loss": 3.6388065456437673,
"tokens_seen": 1030488064
},
{
"epoch": 0.31,
"objective/train/docs_used": 590720,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.85727596282959,
"objective/train/theoretical_loss": 3.6387846453473784,
"objective/train/tokens_used": 1051013600,
"theoretical_loss": 3.6387846453473784,
"tokens_seen": 1030553600
},
{
"epoch": 0.31,
"learning_rate": 0.00034736208625877637,
"loss": 2.7996,
"theoretical_loss": 3.6387846453473784,
"tokens_seen": 1030553600
},
{
"epoch": 0.31,
"learning_rate": 0.0003473520561685055,
"loss": 2.6883,
"theoretical_loss": 3.6387627468335806,
"tokens_seen": 1030619136
},
{
"epoch": 0.31,
"learning_rate": 0.00034734202607823473,
"loss": 2.8209,
"theoretical_loss": 3.638740850102115,
"tokens_seen": 1030684672
},
{
"epoch": 0.31,
"learning_rate": 0.0003473319959879639,
"loss": 2.9573,
"theoretical_loss": 3.6387189551527226,
"tokens_seen": 1030750208
},
{
"epoch": 0.31,
"learning_rate": 0.0003473219658976931,
"loss": 2.6337,
"theoretical_loss": 3.6386970619851464,
"tokens_seen": 1030815744
},
{
"epoch": 0.31,
"learning_rate": 0.00034731193580742227,
"loss": 2.8236,
"theoretical_loss": 3.6386751705991274,
"tokens_seen": 1030881280
},
{
"epoch": 0.31,
"learning_rate": 0.00034730190571715145,
"loss": 2.9283,
"theoretical_loss": 3.6386532809944074,
"tokens_seen": 1030946816
},
{
"epoch": 0.31,
"learning_rate": 0.00034729187562688063,
"loss": 2.7536,
"theoretical_loss": 3.638631393170728,
"tokens_seen": 1031012352
},
{
"epoch": 0.31,
"learning_rate": 0.00034728184553660987,
"loss": 2.4854,
"theoretical_loss": 3.638609507127832,
"tokens_seen": 1031077888
},
{
"epoch": 0.31,
"learning_rate": 0.000347271815446339,
"loss": 2.774,
"theoretical_loss": 3.6385876228654603,
"tokens_seen": 1031143424
},
{
"epoch": 0.31,
"learning_rate": 0.00034726178535606823,
"loss": 2.7051,
"theoretical_loss": 3.6385657403833553,
"tokens_seen": 1031208960
},
{
"epoch": 0.31,
"learning_rate": 0.0003472517552657974,
"loss": 2.7163,
"theoretical_loss": 3.6385438596812594,
"tokens_seen": 1031274496
},
{
"epoch": 0.31,
"learning_rate": 0.0003472417251755266,
"loss": 2.7764,
"theoretical_loss": 3.638521980758914,
"tokens_seen": 1031340032
},
{
"epoch": 0.31,
"learning_rate": 0.0003472316950852558,
"loss": 2.5904,
"theoretical_loss": 3.638500103616062,
"tokens_seen": 1031405568
},
{
"epoch": 0.31,
"learning_rate": 0.00034722166499498496,
"loss": 2.7907,
"theoretical_loss": 3.638478228252445,
"tokens_seen": 1031471104
},
{
"epoch": 0.31,
"learning_rate": 0.00034721163490471414,
"loss": 2.6719,
"theoretical_loss": 3.6384563546678055,
"tokens_seen": 1031536640
},
{
"epoch": 0.31,
"learning_rate": 0.00034720160481444337,
"loss": 2.8774,
"theoretical_loss": 3.6384344828618858,
"tokens_seen": 1031602176
},
{
"epoch": 0.31,
"learning_rate": 0.0003471915747241725,
"loss": 2.5041,
"theoretical_loss": 3.6384126128344283,
"tokens_seen": 1031667712
},
{
"epoch": 0.31,
"learning_rate": 0.00034718154463390173,
"loss": 2.9542,
"theoretical_loss": 3.6383907445851755,
"tokens_seen": 1031733248
},
{
"epoch": 0.31,
"learning_rate": 0.00034717151454363086,
"loss": 2.5853,
"theoretical_loss": 3.6383688781138694,
"tokens_seen": 1031798784
},
{
"epoch": 0.31,
"learning_rate": 0.0003471614844533601,
"loss": 2.8232,
"theoretical_loss": 3.638347013420254,
"tokens_seen": 1031864320
},
{
"epoch": 0.31,
"learning_rate": 0.0003471514543630893,
"loss": 2.6672,
"theoretical_loss": 3.6383251505040697,
"tokens_seen": 1031929856
},
{
"epoch": 0.31,
"learning_rate": 0.00034714142427281846,
"loss": 2.6854,
"theoretical_loss": 3.6383032893650604,
"tokens_seen": 1031995392
},
{
"epoch": 0.31,
"learning_rate": 0.00034713139418254764,
"loss": 2.6707,
"theoretical_loss": 3.6382814300029693,
"tokens_seen": 1032060928
},
{
"epoch": 0.31,
"learning_rate": 0.0003471213640922768,
"loss": 2.662,
"theoretical_loss": 3.638259572417538,
"tokens_seen": 1032126464
},
{
"epoch": 0.31,
"objective/train/docs_used": 591991,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9173693656921387,
"objective/train/theoretical_loss": 3.6382377166085096,
"objective/train/tokens_used": 1052652000,
"theoretical_loss": 3.6382377166085096,
"tokens_seen": 1032192000
},
{
"epoch": 0.31,
"learning_rate": 0.000347111334002006,
"loss": 2.7598,
"theoretical_loss": 3.6382377166085096,
"tokens_seen": 1032192000
},
{
"epoch": 0.31,
"learning_rate": 0.00034710130391173524,
"loss": 2.4278,
"theoretical_loss": 3.6382158625756276,
"tokens_seen": 1032257536
},
{
"epoch": 0.31,
"learning_rate": 0.00034709127382146436,
"loss": 2.7269,
"theoretical_loss": 3.6381940103186343,
"tokens_seen": 1032323072
},
{
"epoch": 0.31,
"learning_rate": 0.0003470812437311936,
"loss": 2.7161,
"theoretical_loss": 3.638172159837273,
"tokens_seen": 1032388608
},
{
"epoch": 0.31,
"learning_rate": 0.0003470712136409228,
"loss": 2.6777,
"theoretical_loss": 3.6381503111312865,
"tokens_seen": 1032454144
},
{
"epoch": 0.31,
"learning_rate": 0.00034706118355065196,
"loss": 2.8306,
"theoretical_loss": 3.6381284642004177,
"tokens_seen": 1032519680
},
{
"epoch": 0.31,
"learning_rate": 0.00034705115346038114,
"loss": 2.6086,
"theoretical_loss": 3.63810661904441,
"tokens_seen": 1032585216
},
{
"epoch": 0.31,
"learning_rate": 0.0003470411233701103,
"loss": 2.7023,
"theoretical_loss": 3.638084775663007,
"tokens_seen": 1032650752
},
{
"epoch": 0.31,
"learning_rate": 0.0003470310932798395,
"loss": 2.7408,
"theoretical_loss": 3.6380629340559505,
"tokens_seen": 1032716288
},
{
"epoch": 0.31,
"learning_rate": 0.00034702106318956874,
"loss": 2.7956,
"theoretical_loss": 3.6380410942229853,
"tokens_seen": 1032781824
},
{
"epoch": 0.31,
"learning_rate": 0.00034701103309929787,
"loss": 2.7543,
"theoretical_loss": 3.6380192561638545,
"tokens_seen": 1032847360
},
{
"epoch": 0.31,
"learning_rate": 0.0003470010030090271,
"loss": 2.4988,
"theoretical_loss": 3.6379974198783005,
"tokens_seen": 1032912896
},
{
"epoch": 0.31,
"learning_rate": 0.00034699097291875623,
"loss": 2.7449,
"theoretical_loss": 3.6379755853660676,
"tokens_seen": 1032978432
},
{
"epoch": 0.31,
"learning_rate": 0.00034698094282848547,
"loss": 2.9559,
"theoretical_loss": 3.637953752626899,
"tokens_seen": 1033043968
},
{
"epoch": 0.31,
"learning_rate": 0.00034697091273821465,
"loss": 2.7226,
"theoretical_loss": 3.6379319216605386,
"tokens_seen": 1033109504
},
{
"epoch": 0.31,
"learning_rate": 0.00034696088264794383,
"loss": 2.4807,
"theoretical_loss": 3.6379100924667296,
"tokens_seen": 1033175040
},
{
"epoch": 0.31,
"learning_rate": 0.000346950852557673,
"loss": 2.6215,
"theoretical_loss": 3.637888265045216,
"tokens_seen": 1033240576
},
{
"epoch": 0.31,
"learning_rate": 0.00034694082246740224,
"loss": 2.848,
"theoretical_loss": 3.637866439395741,
"tokens_seen": 1033306112
},
{
"epoch": 0.31,
"learning_rate": 0.00034693079237713137,
"loss": 2.7529,
"theoretical_loss": 3.6378446155180484,
"tokens_seen": 1033371648
},
{
"epoch": 0.31,
"learning_rate": 0.0003469207622868606,
"loss": 2.9887,
"theoretical_loss": 3.6378227934118827,
"tokens_seen": 1033437184
},
{
"epoch": 0.31,
"learning_rate": 0.00034691073219658973,
"loss": 2.8938,
"theoretical_loss": 3.6378009730769874,
"tokens_seen": 1033502720
},
{
"epoch": 0.31,
"learning_rate": 0.00034690070210631897,
"loss": 2.9539,
"theoretical_loss": 3.6377791545131064,
"tokens_seen": 1033568256
},
{
"epoch": 0.31,
"learning_rate": 0.00034689067201604815,
"loss": 2.7607,
"theoretical_loss": 3.6377573377199837,
"tokens_seen": 1033633792
},
{
"epoch": 0.31,
"learning_rate": 0.00034688064192577733,
"loss": 2.5311,
"theoretical_loss": 3.6377355226973633,
"tokens_seen": 1033699328
},
{
"epoch": 0.31,
"learning_rate": 0.0003468706118355065,
"loss": 2.6857,
"theoretical_loss": 3.6377137094449887,
"tokens_seen": 1033764864
},
{
"epoch": 0.31,
"objective/train/docs_used": 592462,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.183917760848999,
"objective/train/theoretical_loss": 3.6376918979626054,
"objective/train/tokens_used": 1054290400,
"theoretical_loss": 3.6376918979626054,
"tokens_seen": 1033830400
},
{
"epoch": 0.31,
"learning_rate": 0.0003468605817452357,
"loss": 2.8308,
"theoretical_loss": 3.6376918979626054,
"tokens_seen": 1033830400
},
{
"epoch": 0.31,
"learning_rate": 0.0003468505516549649,
"loss": 2.6942,
"theoretical_loss": 3.6376700882499566,
"tokens_seen": 1033895936
},
{
"epoch": 0.31,
"learning_rate": 0.0003468405215646941,
"loss": 2.7195,
"theoretical_loss": 3.637648280306786,
"tokens_seen": 1033961472
},
{
"epoch": 0.31,
"learning_rate": 0.0003468304914744233,
"loss": 2.6013,
"theoretical_loss": 3.6376264741328397,
"tokens_seen": 1034027008
},
{
"epoch": 0.31,
"learning_rate": 0.00034682046138415247,
"loss": 2.538,
"theoretical_loss": 3.6376046697278603,
"tokens_seen": 1034092544
},
{
"epoch": 0.31,
"learning_rate": 0.00034681043129388165,
"loss": 2.5627,
"theoretical_loss": 3.6375828670915933,
"tokens_seen": 1034158080
},
{
"epoch": 0.31,
"learning_rate": 0.00034680040120361083,
"loss": 2.6186,
"theoretical_loss": 3.6375610662237827,
"tokens_seen": 1034223616
},
{
"epoch": 0.31,
"learning_rate": 0.00034679037111334007,
"loss": 2.7466,
"theoretical_loss": 3.637539267124173,
"tokens_seen": 1034289152
},
{
"epoch": 0.31,
"learning_rate": 0.0003467803410230692,
"loss": 2.7698,
"theoretical_loss": 3.637517469792508,
"tokens_seen": 1034354688
},
{
"epoch": 0.31,
"learning_rate": 0.00034677031093279843,
"loss": 2.9972,
"theoretical_loss": 3.6374956742285343,
"tokens_seen": 1034420224
},
{
"epoch": 0.31,
"learning_rate": 0.0003467602808425276,
"loss": 2.8979,
"theoretical_loss": 3.637473880431995,
"tokens_seen": 1034485760
},
{
"epoch": 0.31,
"learning_rate": 0.0003467502507522568,
"loss": 2.7346,
"theoretical_loss": 3.6374520884026347,
"tokens_seen": 1034551296
},
{
"epoch": 0.31,
"learning_rate": 0.000346740220661986,
"loss": 2.7542,
"theoretical_loss": 3.637430298140199,
"tokens_seen": 1034616832
},
{
"epoch": 0.31,
"learning_rate": 0.00034673019057171516,
"loss": 2.8469,
"theoretical_loss": 3.6374085096444326,
"tokens_seen": 1034682368
},
{
"epoch": 0.31,
"learning_rate": 0.00034672016048144434,
"loss": 2.8009,
"theoretical_loss": 3.63738672291508,
"tokens_seen": 1034747904
},
{
"epoch": 0.31,
"learning_rate": 0.00034671013039117357,
"loss": 2.7117,
"theoretical_loss": 3.6373649379518866,
"tokens_seen": 1034813440
},
{
"epoch": 0.31,
"learning_rate": 0.0003467001003009027,
"loss": 2.915,
"theoretical_loss": 3.637343154754597,
"tokens_seen": 1034878976
},
{
"epoch": 0.31,
"learning_rate": 0.00034669007021063193,
"loss": 2.7028,
"theoretical_loss": 3.637321373322956,
"tokens_seen": 1034944512
},
{
"epoch": 0.31,
"learning_rate": 0.00034668004012036106,
"loss": 2.8819,
"theoretical_loss": 3.6372995936567096,
"tokens_seen": 1035010048
},
{
"epoch": 0.31,
"learning_rate": 0.0003466700100300903,
"loss": 2.9058,
"theoretical_loss": 3.637277815755602,
"tokens_seen": 1035075584
},
{
"epoch": 0.31,
"learning_rate": 0.0003466599799398195,
"loss": 2.6997,
"theoretical_loss": 3.6372560396193787,
"tokens_seen": 1035141120
},
{
"epoch": 0.31,
"learning_rate": 0.00034664994984954866,
"loss": 2.5127,
"theoretical_loss": 3.6372342652477854,
"tokens_seen": 1035206656
},
{
"epoch": 0.31,
"learning_rate": 0.00034663991975927784,
"loss": 2.6031,
"theoretical_loss": 3.637212492640567,
"tokens_seen": 1035272192
},
{
"epoch": 0.31,
"learning_rate": 0.000346629889669007,
"loss": 2.7444,
"theoretical_loss": 3.6371907217974684,
"tokens_seen": 1035337728
},
{
"epoch": 0.31,
"learning_rate": 0.0003466198595787362,
"loss": 2.7672,
"theoretical_loss": 3.637168952718236,
"tokens_seen": 1035403264
},
{
"epoch": 0.31,
"objective/train/docs_used": 593828,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.567089319229126,
"objective/train/theoretical_loss": 3.6371471854026147,
"objective/train/tokens_used": 1055928800,
"theoretical_loss": 3.6371471854026147,
"tokens_seen": 1035468800
},
{
"epoch": 0.31,
"learning_rate": 0.00034660982948846544,
"loss": 2.7029,
"theoretical_loss": 3.6371471854026147,
"tokens_seen": 1035468800
},
{
"epoch": 0.31,
"learning_rate": 0.00034659979939819456,
"loss": 2.5621,
"theoretical_loss": 3.63712541985035,
"tokens_seen": 1035534336
},
{
"epoch": 0.31,
"learning_rate": 0.0003465897693079238,
"loss": 2.7115,
"theoretical_loss": 3.6371036560611874,
"tokens_seen": 1035599872
},
{
"epoch": 0.31,
"learning_rate": 0.000346579739217653,
"loss": 2.8134,
"theoretical_loss": 3.637081894034873,
"tokens_seen": 1035665408
},
{
"epoch": 0.31,
"learning_rate": 0.00034656970912738216,
"loss": 2.7556,
"theoretical_loss": 3.6370601337711514,
"tokens_seen": 1035730944
},
{
"epoch": 0.31,
"learning_rate": 0.00034655967903711134,
"loss": 2.7241,
"theoretical_loss": 3.63703837526977,
"tokens_seen": 1035796480
},
{
"epoch": 0.31,
"learning_rate": 0.0003465496489468405,
"loss": 2.7257,
"theoretical_loss": 3.637016618530473,
"tokens_seen": 1035862016
},
{
"epoch": 0.31,
"learning_rate": 0.0003465396188565697,
"loss": 2.5786,
"theoretical_loss": 3.6369948635530074,
"tokens_seen": 1035927552
},
{
"epoch": 0.31,
"learning_rate": 0.00034652958876629894,
"loss": 2.8216,
"theoretical_loss": 3.6369731103371183,
"tokens_seen": 1035993088
},
{
"epoch": 0.31,
"learning_rate": 0.00034651955867602807,
"loss": 2.7028,
"theoretical_loss": 3.636951358882552,
"tokens_seen": 1036058624
},
{
"epoch": 0.31,
"learning_rate": 0.0003465095285857573,
"loss": 2.8427,
"theoretical_loss": 3.6369296091890537,
"tokens_seen": 1036124160
},
{
"epoch": 0.31,
"learning_rate": 0.00034649949849548643,
"loss": 2.7081,
"theoretical_loss": 3.636907861256371,
"tokens_seen": 1036189696
},
{
"epoch": 0.31,
"learning_rate": 0.00034648946840521567,
"loss": 2.5669,
"theoretical_loss": 3.6368861150842484,
"tokens_seen": 1036255232
},
{
"epoch": 0.31,
"learning_rate": 0.00034647943831494485,
"loss": 2.8287,
"theoretical_loss": 3.6368643706724333,
"tokens_seen": 1036320768
},
{
"epoch": 0.31,
"learning_rate": 0.00034646940822467403,
"loss": 2.7646,
"theoretical_loss": 3.636842628020671,
"tokens_seen": 1036386304
},
{
"epoch": 0.31,
"learning_rate": 0.0003464593781344032,
"loss": 2.7109,
"theoretical_loss": 3.6368208871287084,
"tokens_seen": 1036451840
},
{
"epoch": 0.31,
"learning_rate": 0.00034644934804413244,
"loss": 2.5991,
"theoretical_loss": 3.6367991479962916,
"tokens_seen": 1036517376
},
{
"epoch": 0.31,
"learning_rate": 0.00034643931795386157,
"loss": 3.0318,
"theoretical_loss": 3.6367774106231665,
"tokens_seen": 1036582912
},
{
"epoch": 0.31,
"learning_rate": 0.0003464292878635908,
"loss": 2.8121,
"theoretical_loss": 3.6367556750090797,
"tokens_seen": 1036648448
},
{
"epoch": 0.31,
"learning_rate": 0.00034641925777331993,
"loss": 2.9779,
"theoretical_loss": 3.6367339411537785,
"tokens_seen": 1036713984
},
{
"epoch": 0.31,
"learning_rate": 0.00034640922768304917,
"loss": 2.8549,
"theoretical_loss": 3.6367122090570088,
"tokens_seen": 1036779520
},
{
"epoch": 0.31,
"learning_rate": 0.00034639919759277835,
"loss": 2.7136,
"theoretical_loss": 3.6366904787185166,
"tokens_seen": 1036845056
},
{
"epoch": 0.31,
"learning_rate": 0.00034638916750250753,
"loss": 2.9807,
"theoretical_loss": 3.6366687501380492,
"tokens_seen": 1036910592
},
{
"epoch": 0.31,
"learning_rate": 0.0003463791374122367,
"loss": 2.8208,
"theoretical_loss": 3.636647023315353,
"tokens_seen": 1036976128
},
{
"epoch": 0.31,
"learning_rate": 0.0003463691073219659,
"loss": 2.7295,
"theoretical_loss": 3.6366252982501757,
"tokens_seen": 1037041664
},
{
"epoch": 0.31,
"objective/train/docs_used": 594513,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.783609628677368,
"objective/train/theoretical_loss": 3.6366035749422627,
"objective/train/tokens_used": 1057567200,
"theoretical_loss": 3.6366035749422627,
"tokens_seen": 1037107200
},
{
"epoch": 0.31,
"learning_rate": 0.0003463590772316951,
"loss": 2.7807,
"theoretical_loss": 3.6366035749422627,
"tokens_seen": 1037107200
},
{
"epoch": 0.31,
"learning_rate": 0.0003463490471414243,
"loss": 2.3821,
"theoretical_loss": 3.6365818533913608,
"tokens_seen": 1037172736
},
{
"epoch": 0.31,
"learning_rate": 0.00034633901705115344,
"loss": 2.9194,
"theoretical_loss": 3.6365601335972184,
"tokens_seen": 1037238272
},
{
"epoch": 0.31,
"learning_rate": 0.00034632898696088267,
"loss": 2.8947,
"theoretical_loss": 3.636538415559581,
"tokens_seen": 1037303808
},
{
"epoch": 0.31,
"learning_rate": 0.0003463189568706118,
"loss": 2.7321,
"theoretical_loss": 3.636516699278196,
"tokens_seen": 1037369344
},
{
"epoch": 0.31,
"learning_rate": 0.00034630892678034103,
"loss": 2.6896,
"theoretical_loss": 3.6364949847528107,
"tokens_seen": 1037434880
},
{
"epoch": 0.31,
"learning_rate": 0.0003462988966900702,
"loss": 2.6695,
"theoretical_loss": 3.636473271983172,
"tokens_seen": 1037500416
},
{
"epoch": 0.31,
"learning_rate": 0.0003462888665997994,
"loss": 2.6639,
"theoretical_loss": 3.636451560969027,
"tokens_seen": 1037565952
},
{
"epoch": 0.31,
"learning_rate": 0.0003462788365095286,
"loss": 2.8041,
"theoretical_loss": 3.636429851710123,
"tokens_seen": 1037631488
},
{
"epoch": 0.31,
"learning_rate": 0.0003462688064192578,
"loss": 2.6749,
"theoretical_loss": 3.6364081442062073,
"tokens_seen": 1037697024
},
{
"epoch": 0.31,
"learning_rate": 0.00034625877632898694,
"loss": 2.7544,
"theoretical_loss": 3.636386438457027,
"tokens_seen": 1037762560
},
{
"epoch": 0.31,
"learning_rate": 0.0003462487462387162,
"loss": 3.019,
"theoretical_loss": 3.6363647344623296,
"tokens_seen": 1037828096
},
{
"epoch": 0.31,
"learning_rate": 0.0003462387161484453,
"loss": 2.8214,
"theoretical_loss": 3.636343032221862,
"tokens_seen": 1037893632
},
{
"epoch": 0.31,
"learning_rate": 0.00034622868605817454,
"loss": 2.884,
"theoretical_loss": 3.6363213317353726,
"tokens_seen": 1037959168
},
{
"epoch": 0.31,
"learning_rate": 0.0003462186559679037,
"loss": 2.7941,
"theoretical_loss": 3.6362996330026087,
"tokens_seen": 1038024704
},
{
"epoch": 0.31,
"learning_rate": 0.0003462086258776329,
"loss": 2.6661,
"theoretical_loss": 3.6362779360233173,
"tokens_seen": 1038090240
},
{
"epoch": 0.31,
"learning_rate": 0.0003461985957873621,
"loss": 2.5838,
"theoretical_loss": 3.636256240797246,
"tokens_seen": 1038155776
},
{
"epoch": 0.31,
"learning_rate": 0.00034618856569709126,
"loss": 2.7363,
"theoretical_loss": 3.6362345473241433,
"tokens_seen": 1038221312
},
{
"epoch": 0.31,
"learning_rate": 0.00034617853560682044,
"loss": 2.8479,
"theoretical_loss": 3.636212855603756,
"tokens_seen": 1038286848
},
{
"epoch": 0.31,
"learning_rate": 0.0003461685055165497,
"loss": 2.7412,
"theoretical_loss": 3.6361911656358323,
"tokens_seen": 1038352384
},
{
"epoch": 0.31,
"learning_rate": 0.0003461584754262788,
"loss": 2.9473,
"theoretical_loss": 3.63616947742012,
"tokens_seen": 1038417920
},
{
"epoch": 0.31,
"learning_rate": 0.00034614844533600804,
"loss": 2.6147,
"theoretical_loss": 3.6361477909563673,
"tokens_seen": 1038483456
},
{
"epoch": 0.31,
"learning_rate": 0.00034613841524573717,
"loss": 2.6479,
"theoretical_loss": 3.6361261062443218,
"tokens_seen": 1038548992
},
{
"epoch": 0.31,
"learning_rate": 0.0003461283851554664,
"loss": 2.6593,
"theoretical_loss": 3.636104423283731,
"tokens_seen": 1038614528
},
{
"epoch": 0.31,
"learning_rate": 0.0003461183550651956,
"loss": 2.8537,
"theoretical_loss": 3.6360827420743442,
"tokens_seen": 1038680064
},
{
"epoch": 0.31,
"objective/train/docs_used": 595196,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.021951675415039,
"objective/train/theoretical_loss": 3.6360610626159087,
"objective/train/tokens_used": 1059205600,
"theoretical_loss": 3.6360610626159087,
"tokens_seen": 1038745600
},
{
"epoch": 0.31,
"learning_rate": 0.00034610832497492476,
"loss": 2.7348,
"theoretical_loss": 3.6360610626159087,
"tokens_seen": 1038745600
},
{
"epoch": 0.31,
"learning_rate": 0.00034609829488465395,
"loss": 2.7342,
"theoretical_loss": 3.6360393849081722,
"tokens_seen": 1038811136
},
{
"epoch": 0.31,
"learning_rate": 0.0003460882647943832,
"loss": 2.4902,
"theoretical_loss": 3.636017708950883,
"tokens_seen": 1038876672
},
{
"epoch": 0.31,
"learning_rate": 0.00034607823470411236,
"loss": 2.7407,
"theoretical_loss": 3.6359960347437905,
"tokens_seen": 1038942208
},
{
"epoch": 0.31,
"learning_rate": 0.00034606820461384154,
"loss": 2.6393,
"theoretical_loss": 3.6359743622866416,
"tokens_seen": 1039007744
},
{
"epoch": 0.31,
"learning_rate": 0.0003460581745235707,
"loss": 2.5211,
"theoretical_loss": 3.6359526915791855,
"tokens_seen": 1039073280
},
{
"epoch": 0.31,
"learning_rate": 0.0003460481444332999,
"loss": 2.5356,
"theoretical_loss": 3.63593102262117,
"tokens_seen": 1039138816
},
{
"epoch": 0.31,
"learning_rate": 0.00034603811434302914,
"loss": 2.6141,
"theoretical_loss": 3.6359093554123447,
"tokens_seen": 1039204352
},
{
"epoch": 0.31,
"learning_rate": 0.00034602808425275827,
"loss": 2.8159,
"theoretical_loss": 3.6358876899524564,
"tokens_seen": 1039269888
},
{
"epoch": 0.31,
"learning_rate": 0.0003460180541624875,
"loss": 2.6592,
"theoretical_loss": 3.635866026241255,
"tokens_seen": 1039335424
},
{
"epoch": 0.31,
"learning_rate": 0.00034600802407221663,
"loss": 2.6715,
"theoretical_loss": 3.6358443642784883,
"tokens_seen": 1039400960
},
{
"epoch": 0.31,
"learning_rate": 0.00034599799398194587,
"loss": 2.7017,
"theoretical_loss": 3.6358227040639055,
"tokens_seen": 1039466496
},
{
"epoch": 0.32,
"learning_rate": 0.00034598796389167505,
"loss": 2.8213,
"theoretical_loss": 3.635801045597255,
"tokens_seen": 1039532032
},
{
"epoch": 0.32,
"learning_rate": 0.00034597793380140423,
"loss": 2.6238,
"theoretical_loss": 3.6357793888782863,
"tokens_seen": 1039597568
},
{
"epoch": 0.32,
"learning_rate": 0.0003459679037111334,
"loss": 2.5763,
"theoretical_loss": 3.635757733906747,
"tokens_seen": 1039663104
},
{
"epoch": 0.32,
"learning_rate": 0.00034595787362086264,
"loss": 2.7467,
"theoretical_loss": 3.6357360806823866,
"tokens_seen": 1039728640
},
{
"epoch": 0.32,
"learning_rate": 0.00034594784353059177,
"loss": 2.8878,
"theoretical_loss": 3.635714429204954,
"tokens_seen": 1039794176
},
{
"epoch": 0.32,
"learning_rate": 0.000345937813440321,
"loss": 2.8138,
"theoretical_loss": 3.6356927794741978,
"tokens_seen": 1039859712
},
{
"epoch": 0.32,
"learning_rate": 0.00034592778335005013,
"loss": 2.79,
"theoretical_loss": 3.6356711314898678,
"tokens_seen": 1039925248
},
{
"epoch": 0.32,
"learning_rate": 0.00034591775325977937,
"loss": 2.7172,
"theoretical_loss": 3.6356494852517125,
"tokens_seen": 1039990784
},
{
"epoch": 0.32,
"learning_rate": 0.00034590772316950855,
"loss": 2.7562,
"theoretical_loss": 3.635627840759481,
"tokens_seen": 1040056320
},
{
"epoch": 0.32,
"learning_rate": 0.00034589769307923773,
"loss": 2.7936,
"theoretical_loss": 3.6356061980129235,
"tokens_seen": 1040121856
},
{
"epoch": 0.32,
"learning_rate": 0.0003458876629889669,
"loss": 2.801,
"theoretical_loss": 3.6355845570117875,
"tokens_seen": 1040187392
},
{
"epoch": 0.32,
"learning_rate": 0.0003458776328986961,
"loss": 2.6287,
"theoretical_loss": 3.6355629177558235,
"tokens_seen": 1040252928
},
{
"epoch": 0.32,
"learning_rate": 0.0003458676028084253,
"loss": 2.8525,
"theoretical_loss": 3.63554128024478,
"tokens_seen": 1040318464
},
{
"epoch": 0.32,
"objective/train/docs_used": 596146,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8540592193603516,
"objective/train/theoretical_loss": 3.635519644478408,
"objective/train/tokens_used": 1060844000,
"theoretical_loss": 3.635519644478408,
"tokens_seen": 1040384000
},
{
"epoch": 0.32,
"learning_rate": 0.0003458575727181545,
"loss": 2.7168,
"theoretical_loss": 3.635519644478408,
"tokens_seen": 1040384000
},
{
"epoch": 0.32,
"learning_rate": 0.00034584754262788364,
"loss": 2.7366,
"theoretical_loss": 3.6354980104564545,
"tokens_seen": 1040449536
},
{
"epoch": 0.32,
"learning_rate": 0.00034583751253761287,
"loss": 2.5537,
"theoretical_loss": 3.635476378178671,
"tokens_seen": 1040515072
},
{
"epoch": 0.32,
"learning_rate": 0.000345827482447342,
"loss": 2.9031,
"theoretical_loss": 3.6354547476448067,
"tokens_seen": 1040580608
},
{
"epoch": 0.32,
"learning_rate": 0.00034581745235707123,
"loss": 2.7119,
"theoretical_loss": 3.63543311885461,
"tokens_seen": 1040646144
},
{
"epoch": 0.32,
"learning_rate": 0.0003458074222668004,
"loss": 2.717,
"theoretical_loss": 3.6354114918078317,
"tokens_seen": 1040711680
},
{
"epoch": 0.32,
"learning_rate": 0.0003457973921765296,
"loss": 2.6363,
"theoretical_loss": 3.6353898665042212,
"tokens_seen": 1040777216
},
{
"epoch": 0.32,
"learning_rate": 0.0003457873620862588,
"loss": 2.7108,
"theoretical_loss": 3.6353682429435286,
"tokens_seen": 1040842752
},
{
"epoch": 0.32,
"learning_rate": 0.000345777331995988,
"loss": 2.6199,
"theoretical_loss": 3.635346621125503,
"tokens_seen": 1040908288
},
{
"epoch": 0.32,
"learning_rate": 0.00034576730190571714,
"loss": 2.7043,
"theoretical_loss": 3.6353250010498943,
"tokens_seen": 1040973824
},
{
"epoch": 0.32,
"learning_rate": 0.0003457572718154464,
"loss": 2.8504,
"theoretical_loss": 3.635303382716453,
"tokens_seen": 1041039360
},
{
"epoch": 0.32,
"learning_rate": 0.0003457472417251755,
"loss": 2.8313,
"theoretical_loss": 3.635281766124929,
"tokens_seen": 1041104896
},
{
"epoch": 0.32,
"learning_rate": 0.00034573721163490474,
"loss": 2.6828,
"theoretical_loss": 3.6352601512750713,
"tokens_seen": 1041170432
},
{
"epoch": 0.32,
"learning_rate": 0.0003457271815446339,
"loss": 2.5189,
"theoretical_loss": 3.6352385381666315,
"tokens_seen": 1041235968
},
{
"epoch": 0.32,
"learning_rate": 0.0003457171514543631,
"loss": 2.7186,
"theoretical_loss": 3.635216926799358,
"tokens_seen": 1041301504
},
{
"epoch": 0.32,
"learning_rate": 0.0003457071213640923,
"loss": 2.7577,
"theoretical_loss": 3.6351953171730025,
"tokens_seen": 1041367040
},
{
"epoch": 0.32,
"learning_rate": 0.00034569709127382146,
"loss": 2.664,
"theoretical_loss": 3.635173709287314,
"tokens_seen": 1041432576
},
{
"epoch": 0.32,
"learning_rate": 0.00034568706118355064,
"loss": 2.5825,
"theoretical_loss": 3.635152103142044,
"tokens_seen": 1041498112
},
{
"epoch": 0.32,
"learning_rate": 0.0003456770310932799,
"loss": 2.461,
"theoretical_loss": 3.6351304987369417,
"tokens_seen": 1041563648
},
{
"epoch": 0.32,
"learning_rate": 0.000345667001003009,
"loss": 2.5191,
"theoretical_loss": 3.635108896071759,
"tokens_seen": 1041629184
},
{
"epoch": 0.32,
"learning_rate": 0.00034565697091273824,
"loss": 2.7986,
"theoretical_loss": 3.635087295146244,
"tokens_seen": 1041694720
},
{
"epoch": 0.32,
"learning_rate": 0.00034564694082246737,
"loss": 2.8264,
"theoretical_loss": 3.6350656959601486,
"tokens_seen": 1041760256
},
{
"epoch": 0.32,
"learning_rate": 0.0003456369107321966,
"loss": 2.6925,
"theoretical_loss": 3.635044098513223,
"tokens_seen": 1041825792
},
{
"epoch": 0.32,
"learning_rate": 0.0003456268806419258,
"loss": 2.6474,
"theoretical_loss": 3.6350225028052185,
"tokens_seen": 1041891328
},
{
"epoch": 0.32,
"learning_rate": 0.00034561685055165496,
"loss": 2.6016,
"theoretical_loss": 3.6350009088358846,
"tokens_seen": 1041956864
},
{
"epoch": 0.32,
"objective/train/docs_used": 596810,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1325435638427734,
"objective/train/theoretical_loss": 3.634979316604973,
"objective/train/tokens_used": 1062482400,
"theoretical_loss": 3.634979316604973,
"tokens_seen": 1042022400
},
{
"epoch": 0.32,
"learning_rate": 0.00034560682046138415,
"loss": 2.5139,
"theoretical_loss": 3.634979316604973,
"tokens_seen": 1042022400
},
{
"epoch": 0.32,
"learning_rate": 0.0003455967903711134,
"loss": 2.631,
"theoretical_loss": 3.6349577261122334,
"tokens_seen": 1042087936
},
{
"epoch": 0.32,
"learning_rate": 0.0003455867602808425,
"loss": 2.6916,
"theoretical_loss": 3.6349361373574176,
"tokens_seen": 1042153472
},
{
"epoch": 0.32,
"learning_rate": 0.00034557673019057174,
"loss": 2.6816,
"theoretical_loss": 3.6349145503402758,
"tokens_seen": 1042219008
},
{
"epoch": 0.32,
"learning_rate": 0.00034556670010030087,
"loss": 2.91,
"theoretical_loss": 3.6348929650605584,
"tokens_seen": 1042284544
},
{
"epoch": 0.32,
"learning_rate": 0.0003455566700100301,
"loss": 2.6672,
"theoretical_loss": 3.6348713815180176,
"tokens_seen": 1042350080
},
{
"epoch": 0.32,
"learning_rate": 0.0003455466399197593,
"loss": 2.7211,
"theoretical_loss": 3.634849799712404,
"tokens_seen": 1042415616
},
{
"epoch": 0.32,
"learning_rate": 0.00034553660982948847,
"loss": 2.6108,
"theoretical_loss": 3.6348282196434676,
"tokens_seen": 1042481152
},
{
"epoch": 0.32,
"learning_rate": 0.00034552657973921765,
"loss": 2.5731,
"theoretical_loss": 3.63480664131096,
"tokens_seen": 1042546688
},
{
"epoch": 0.32,
"learning_rate": 0.00034551654964894683,
"loss": 2.7859,
"theoretical_loss": 3.6347850647146336,
"tokens_seen": 1042612224
},
{
"epoch": 0.32,
"learning_rate": 0.000345506519558676,
"loss": 2.5827,
"theoretical_loss": 3.6347634898542385,
"tokens_seen": 1042677760
},
{
"epoch": 0.32,
"learning_rate": 0.00034549648946840525,
"loss": 2.8312,
"theoretical_loss": 3.634741916729525,
"tokens_seen": 1042743296
},
{
"epoch": 0.32,
"learning_rate": 0.0003454864593781344,
"loss": 2.7075,
"theoretical_loss": 3.6347203453402464,
"tokens_seen": 1042808832
},
{
"epoch": 0.32,
"learning_rate": 0.0003454764292878636,
"loss": 2.8898,
"theoretical_loss": 3.634698775686153,
"tokens_seen": 1042874368
},
{
"epoch": 0.32,
"learning_rate": 0.00034546639919759274,
"loss": 2.5798,
"theoretical_loss": 3.6346772077669955,
"tokens_seen": 1042939904
},
{
"epoch": 0.32,
"learning_rate": 0.00034545636910732197,
"loss": 2.9125,
"theoretical_loss": 3.634655641582527,
"tokens_seen": 1043005440
},
{
"epoch": 0.32,
"learning_rate": 0.00034544633901705115,
"loss": 2.7016,
"theoretical_loss": 3.6346340771324974,
"tokens_seen": 1043070976
},
{
"epoch": 0.32,
"learning_rate": 0.00034543630892678033,
"loss": 2.7667,
"theoretical_loss": 3.6346125144166592,
"tokens_seen": 1043136512
},
{
"epoch": 0.32,
"learning_rate": 0.0003454262788365095,
"loss": 2.8322,
"theoretical_loss": 3.634590953434764,
"tokens_seen": 1043202048
},
{
"epoch": 0.32,
"learning_rate": 0.00034541624874623875,
"loss": 2.6207,
"theoretical_loss": 3.6345693941865624,
"tokens_seen": 1043267584
},
{
"epoch": 0.32,
"learning_rate": 0.0003454062186559679,
"loss": 2.9534,
"theoretical_loss": 3.634547836671808,
"tokens_seen": 1043333120
},
{
"epoch": 0.32,
"learning_rate": 0.0003453961885656971,
"loss": 2.7372,
"theoretical_loss": 3.6345262808902508,
"tokens_seen": 1043398656
},
{
"epoch": 0.32,
"learning_rate": 0.00034538615847542624,
"loss": 2.7226,
"theoretical_loss": 3.6345047268416435,
"tokens_seen": 1043464192
},
{
"epoch": 0.32,
"learning_rate": 0.0003453761283851555,
"loss": 2.789,
"theoretical_loss": 3.6344831745257373,
"tokens_seen": 1043529728
},
{
"epoch": 0.32,
"learning_rate": 0.00034536609829488466,
"loss": 2.8811,
"theoretical_loss": 3.6344616239422844,
"tokens_seen": 1043595264
},
{
"epoch": 0.32,
"objective/train/docs_used": 598197,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.708169937133789,
"objective/train/theoretical_loss": 3.634440075091037,
"objective/train/tokens_used": 1064120800,
"theoretical_loss": 3.634440075091037,
"tokens_seen": 1043660800
},
{
"epoch": 0.32,
"learning_rate": 0.00034535606820461384,
"loss": 2.6356,
"theoretical_loss": 3.634440075091037,
"tokens_seen": 1043660800
},
{
"epoch": 0.32,
"learning_rate": 0.000345346038114343,
"loss": 2.833,
"theoretical_loss": 3.6344185279717474,
"tokens_seen": 1043726336
},
{
"epoch": 0.32,
"learning_rate": 0.0003453360080240722,
"loss": 2.7528,
"theoretical_loss": 3.634396982584167,
"tokens_seen": 1043791872
},
{
"epoch": 0.32,
"learning_rate": 0.00034532597793380143,
"loss": 2.8282,
"theoretical_loss": 3.634375438928048,
"tokens_seen": 1043857408
},
{
"epoch": 0.32,
"learning_rate": 0.0003453159478435306,
"loss": 2.5418,
"theoretical_loss": 3.6343538970031433,
"tokens_seen": 1043922944
},
{
"epoch": 0.32,
"learning_rate": 0.0003453059177532598,
"loss": 2.7797,
"theoretical_loss": 3.634332356809204,
"tokens_seen": 1043988480
},
{
"epoch": 0.32,
"learning_rate": 0.000345295887662989,
"loss": 2.8315,
"theoretical_loss": 3.6343108183459827,
"tokens_seen": 1044054016
},
{
"epoch": 0.32,
"learning_rate": 0.0003452858575727182,
"loss": 2.6344,
"theoretical_loss": 3.634289281613232,
"tokens_seen": 1044119552
},
{
"epoch": 0.32,
"learning_rate": 0.00034527582748244734,
"loss": 3.0662,
"theoretical_loss": 3.6342677466107043,
"tokens_seen": 1044185088
},
{
"epoch": 0.32,
"learning_rate": 0.0003452657973921766,
"loss": 2.6573,
"theoretical_loss": 3.634246213338152,
"tokens_seen": 1044250624
},
{
"epoch": 0.32,
"learning_rate": 0.0003452557673019057,
"loss": 2.8743,
"theoretical_loss": 3.6342246817953274,
"tokens_seen": 1044316160
},
{
"epoch": 0.32,
"learning_rate": 0.00034524573721163494,
"loss": 2.8355,
"theoretical_loss": 3.634203151981983,
"tokens_seen": 1044381696
},
{
"epoch": 0.32,
"learning_rate": 0.0003452357071213641,
"loss": 2.7479,
"theoretical_loss": 3.6341816238978715,
"tokens_seen": 1044447232
},
{
"epoch": 0.32,
"learning_rate": 0.0003452256770310933,
"loss": 2.7237,
"theoretical_loss": 3.634160097542745,
"tokens_seen": 1044512768
},
{
"epoch": 0.32,
"learning_rate": 0.0003452156469408225,
"loss": 2.7335,
"theoretical_loss": 3.634138572916357,
"tokens_seen": 1044578304
},
{
"epoch": 0.32,
"learning_rate": 0.00034520561685055166,
"loss": 2.8916,
"theoretical_loss": 3.6341170500184603,
"tokens_seen": 1044643840
},
{
"epoch": 0.32,
"learning_rate": 0.00034519558676028084,
"loss": 3.0492,
"theoretical_loss": 3.6340955288488064,
"tokens_seen": 1044709376
},
{
"epoch": 0.32,
"learning_rate": 0.0003451855566700101,
"loss": 2.7161,
"theoretical_loss": 3.6340740094071493,
"tokens_seen": 1044774912
},
{
"epoch": 0.32,
"learning_rate": 0.0003451755265797392,
"loss": 2.7836,
"theoretical_loss": 3.6340524916932413,
"tokens_seen": 1044840448
},
{
"epoch": 0.32,
"learning_rate": 0.00034516549648946844,
"loss": 2.885,
"theoretical_loss": 3.6340309757068354,
"tokens_seen": 1044905984
},
{
"epoch": 0.32,
"learning_rate": 0.00034515546639919757,
"loss": 3.0083,
"theoretical_loss": 3.634009461447685,
"tokens_seen": 1044971520
},
{
"epoch": 0.32,
"learning_rate": 0.0003451454363089268,
"loss": 2.8429,
"theoretical_loss": 3.6339879489155433,
"tokens_seen": 1045037056
},
{
"epoch": 0.32,
"learning_rate": 0.000345135406218656,
"loss": 2.3434,
"theoretical_loss": 3.6339664381101624,
"tokens_seen": 1045102592
},
{
"epoch": 0.32,
"learning_rate": 0.00034512537612838517,
"loss": 2.727,
"theoretical_loss": 3.633944929031296,
"tokens_seen": 1045168128
},
{
"epoch": 0.32,
"learning_rate": 0.00034511534603811435,
"loss": 2.7834,
"theoretical_loss": 3.6339234216786975,
"tokens_seen": 1045233664
},
{
"epoch": 0.32,
"objective/train/docs_used": 598880,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.650927782058716,
"objective/train/theoretical_loss": 3.6339019160521198,
"objective/train/tokens_used": 1065759200,
"theoretical_loss": 3.6339019160521198,
"tokens_seen": 1045299200
},
{
"epoch": 0.32,
"learning_rate": 0.0003451053159478436,
"loss": 2.6322,
"theoretical_loss": 3.6339019160521198,
"tokens_seen": 1045299200
},
{
"epoch": 0.32,
"learning_rate": 0.0003450952858575727,
"loss": 2.602,
"theoretical_loss": 3.633880412151316,
"tokens_seen": 1045364736
},
{
"epoch": 0.32,
"learning_rate": 0.00034508525576730194,
"loss": 2.7525,
"theoretical_loss": 3.6338589099760394,
"tokens_seen": 1045430272
},
{
"epoch": 0.32,
"learning_rate": 0.00034507522567703107,
"loss": 2.6995,
"theoretical_loss": 3.633837409526044,
"tokens_seen": 1045495808
},
{
"epoch": 0.32,
"learning_rate": 0.0003450651955867603,
"loss": 2.9153,
"theoretical_loss": 3.6338159108010832,
"tokens_seen": 1045561344
},
{
"epoch": 0.32,
"learning_rate": 0.0003450551654964895,
"loss": 2.7391,
"theoretical_loss": 3.63379441380091,
"tokens_seen": 1045626880
},
{
"epoch": 0.32,
"learning_rate": 0.00034504513540621867,
"loss": 2.9231,
"theoretical_loss": 3.6337729185252776,
"tokens_seen": 1045692416
},
{
"epoch": 0.32,
"learning_rate": 0.00034503510531594785,
"loss": 2.6171,
"theoretical_loss": 3.6337514249739407,
"tokens_seen": 1045757952
},
{
"epoch": 0.32,
"learning_rate": 0.00034502507522567703,
"loss": 2.6923,
"theoretical_loss": 3.633729933146652,
"tokens_seen": 1045823488
},
{
"epoch": 0.32,
"learning_rate": 0.0003450150451354062,
"loss": 2.8199,
"theoretical_loss": 3.6337084430431656,
"tokens_seen": 1045889024
},
{
"epoch": 0.32,
"learning_rate": 0.00034500501504513545,
"loss": 3.0133,
"theoretical_loss": 3.633686954663235,
"tokens_seen": 1045954560
},
{
"epoch": 0.32,
"learning_rate": 0.0003449949849548646,
"loss": 2.7952,
"theoretical_loss": 3.6336654680066145,
"tokens_seen": 1046020096
},
{
"epoch": 0.32,
"learning_rate": 0.0003449849548645938,
"loss": 2.7813,
"theoretical_loss": 3.6336439830730574,
"tokens_seen": 1046085632
},
{
"epoch": 0.32,
"learning_rate": 0.00034497492477432294,
"loss": 2.7422,
"theoretical_loss": 3.6336224998623177,
"tokens_seen": 1046151168
},
{
"epoch": 0.32,
"learning_rate": 0.00034496489468405217,
"loss": 3.0118,
"theoretical_loss": 3.6336010183741494,
"tokens_seen": 1046216704
},
{
"epoch": 0.32,
"learning_rate": 0.00034495486459378135,
"loss": 2.6331,
"theoretical_loss": 3.6335795386083065,
"tokens_seen": 1046282240
},
{
"epoch": 0.32,
"learning_rate": 0.00034494483450351053,
"loss": 2.8236,
"theoretical_loss": 3.6335580605645434,
"tokens_seen": 1046347776
},
{
"epoch": 0.32,
"learning_rate": 0.0003449348044132397,
"loss": 2.7138,
"theoretical_loss": 3.633536584242613,
"tokens_seen": 1046413312
},
{
"epoch": 0.32,
"learning_rate": 0.00034492477432296895,
"loss": 2.7371,
"theoretical_loss": 3.6335151096422713,
"tokens_seen": 1046478848
},
{
"epoch": 0.32,
"learning_rate": 0.0003449147442326981,
"loss": 2.7032,
"theoretical_loss": 3.6334936367632706,
"tokens_seen": 1046544384
},
{
"epoch": 0.32,
"learning_rate": 0.0003449047141424273,
"loss": 2.7739,
"theoretical_loss": 3.633472165605366,
"tokens_seen": 1046609920
},
{
"epoch": 0.32,
"learning_rate": 0.00034489468405215644,
"loss": 2.765,
"theoretical_loss": 3.633450696168312,
"tokens_seen": 1046675456
},
{
"epoch": 0.32,
"learning_rate": 0.0003448846539618857,
"loss": 2.4997,
"theoretical_loss": 3.6334292284518632,
"tokens_seen": 1046740992
},
{
"epoch": 0.32,
"learning_rate": 0.00034487462387161486,
"loss": 2.6,
"theoretical_loss": 3.6334077624557732,
"tokens_seen": 1046806528
},
{
"epoch": 0.32,
"learning_rate": 0.00034486459378134404,
"loss": 2.6627,
"theoretical_loss": 3.633386298179797,
"tokens_seen": 1046872064
},
{
"epoch": 0.32,
"objective/train/docs_used": 599993,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.623795509338379,
"objective/train/theoretical_loss": 3.6333648356236883,
"objective/train/tokens_used": 1067397600,
"theoretical_loss": 3.6333648356236883,
"tokens_seen": 1046937600
},
{
"epoch": 0.32,
"learning_rate": 0.0003448545636910732,
"loss": 2.6726,
"theoretical_loss": 3.6333648356236883,
"tokens_seen": 1046937600
},
{
"epoch": 0.32,
"learning_rate": 0.0003448445336008024,
"loss": 2.6566,
"theoretical_loss": 3.6333433747872026,
"tokens_seen": 1047003136
},
{
"epoch": 0.32,
"learning_rate": 0.0003448345035105316,
"loss": 2.6772,
"theoretical_loss": 3.633321915670094,
"tokens_seen": 1047068672
},
{
"epoch": 0.32,
"learning_rate": 0.0003448244734202608,
"loss": 2.5991,
"theoretical_loss": 3.6333004582721173,
"tokens_seen": 1047134208
},
{
"epoch": 0.32,
"learning_rate": 0.00034481444332998994,
"loss": 2.6313,
"theoretical_loss": 3.633279002593027,
"tokens_seen": 1047199744
},
{
"epoch": 0.32,
"learning_rate": 0.0003448044132397192,
"loss": 2.7027,
"theoretical_loss": 3.6332575486325784,
"tokens_seen": 1047265280
},
{
"epoch": 0.32,
"learning_rate": 0.00034479438314944836,
"loss": 2.857,
"theoretical_loss": 3.6332360963905255,
"tokens_seen": 1047330816
},
{
"epoch": 0.32,
"learning_rate": 0.00034478435305917754,
"loss": 2.8275,
"theoretical_loss": 3.633214645866624,
"tokens_seen": 1047396352
},
{
"epoch": 0.32,
"learning_rate": 0.0003447743229689067,
"loss": 2.731,
"theoretical_loss": 3.6331931970606277,
"tokens_seen": 1047461888
},
{
"epoch": 0.32,
"learning_rate": 0.0003447642928786359,
"loss": 2.8738,
"theoretical_loss": 3.6331717499722926,
"tokens_seen": 1047527424
},
{
"epoch": 0.32,
"learning_rate": 0.0003447542627883651,
"loss": 2.5055,
"theoretical_loss": 3.633150304601373,
"tokens_seen": 1047592960
},
{
"epoch": 0.32,
"learning_rate": 0.0003447442326980943,
"loss": 2.8259,
"theoretical_loss": 3.6331288609476244,
"tokens_seen": 1047658496
},
{
"epoch": 0.32,
"learning_rate": 0.00034473420260782345,
"loss": 2.4348,
"theoretical_loss": 3.633107419010802,
"tokens_seen": 1047724032
},
{
"epoch": 0.32,
"learning_rate": 0.0003447241725175527,
"loss": 2.7631,
"theoretical_loss": 3.6330859787906604,
"tokens_seen": 1047789568
},
{
"epoch": 0.32,
"learning_rate": 0.0003447141424272818,
"loss": 2.5878,
"theoretical_loss": 3.6330645402869552,
"tokens_seen": 1047855104
},
{
"epoch": 0.32,
"learning_rate": 0.00034470411233701104,
"loss": 2.821,
"theoretical_loss": 3.6330431034994417,
"tokens_seen": 1047920640
},
{
"epoch": 0.32,
"learning_rate": 0.0003446940822467402,
"loss": 2.8055,
"theoretical_loss": 3.6330216684278747,
"tokens_seen": 1047986176
},
{
"epoch": 0.32,
"learning_rate": 0.0003446840521564694,
"loss": 2.7781,
"theoretical_loss": 3.63300023507201,
"tokens_seen": 1048051712
},
{
"epoch": 0.32,
"learning_rate": 0.0003446740220661986,
"loss": 2.9707,
"theoretical_loss": 3.6329788034316026,
"tokens_seen": 1048117248
},
{
"epoch": 0.32,
"learning_rate": 0.00034466399197592777,
"loss": 2.6958,
"theoretical_loss": 3.6329573735064087,
"tokens_seen": 1048182784
},
{
"epoch": 0.32,
"learning_rate": 0.00034465396188565695,
"loss": 2.819,
"theoretical_loss": 3.6329359452961834,
"tokens_seen": 1048248320
},
{
"epoch": 0.32,
"learning_rate": 0.0003446439317953862,
"loss": 2.5474,
"theoretical_loss": 3.6329145188006815,
"tokens_seen": 1048313856
},
{
"epoch": 0.32,
"learning_rate": 0.0003446339017051153,
"loss": 2.5532,
"theoretical_loss": 3.6328930940196598,
"tokens_seen": 1048379392
},
{
"epoch": 0.32,
"learning_rate": 0.00034462387161484455,
"loss": 2.7918,
"theoretical_loss": 3.6328716709528734,
"tokens_seen": 1048444928
},
{
"epoch": 0.32,
"learning_rate": 0.00034461384152457373,
"loss": 2.6842,
"theoretical_loss": 3.632850249600078,
"tokens_seen": 1048510464
},
{
"epoch": 0.32,
"objective/train/docs_used": 601088,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8884835243225098,
"objective/train/theoretical_loss": 3.632828829961029,
"objective/train/tokens_used": 1069036000,
"theoretical_loss": 3.632828829961029,
"tokens_seen": 1048576000
},
{
"epoch": 0.32,
"learning_rate": 0.0003446038114343029,
"loss": 2.7337,
"theoretical_loss": 3.632828829961029,
"tokens_seen": 1048576000
},
{
"epoch": 0.32,
"learning_rate": 0.00034459378134403214,
"loss": 2.5878,
"theoretical_loss": 3.632807412035483,
"tokens_seen": 1048641536
},
{
"epoch": 0.32,
"learning_rate": 0.00034458375125376127,
"loss": 2.7271,
"theoretical_loss": 3.6327859958231956,
"tokens_seen": 1048707072
},
{
"epoch": 0.32,
"learning_rate": 0.0003445737211634905,
"loss": 2.6982,
"theoretical_loss": 3.632764581323922,
"tokens_seen": 1048772608
},
{
"epoch": 0.32,
"learning_rate": 0.0003445636910732197,
"loss": 2.7898,
"theoretical_loss": 3.632743168537419,
"tokens_seen": 1048838144
},
{
"epoch": 0.32,
"learning_rate": 0.00034455366098294887,
"loss": 2.7673,
"theoretical_loss": 3.6327217574634423,
"tokens_seen": 1048903680
},
{
"epoch": 0.32,
"learning_rate": 0.00034454363089267805,
"loss": 2.8375,
"theoretical_loss": 3.632700348101748,
"tokens_seen": 1048969216
},
{
"epoch": 0.32,
"learning_rate": 0.00034453360080240723,
"loss": 2.7455,
"theoretical_loss": 3.632678940452092,
"tokens_seen": 1049034752
},
{
"epoch": 0.32,
"learning_rate": 0.0003445235707121364,
"loss": 2.6849,
"theoretical_loss": 3.6326575345142307,
"tokens_seen": 1049100288
},
{
"epoch": 0.32,
"learning_rate": 0.00034451354062186565,
"loss": 2.7952,
"theoretical_loss": 3.63263613028792,
"tokens_seen": 1049165824
},
{
"epoch": 0.32,
"learning_rate": 0.0003445035105315948,
"loss": 2.802,
"theoretical_loss": 3.632614727772917,
"tokens_seen": 1049231360
},
{
"epoch": 0.32,
"learning_rate": 0.000344493480441324,
"loss": 2.8472,
"theoretical_loss": 3.6325933269689763,
"tokens_seen": 1049296896
},
{
"epoch": 0.32,
"learning_rate": 0.00034448345035105314,
"loss": 2.7377,
"theoretical_loss": 3.6325719278758557,
"tokens_seen": 1049362432
},
{
"epoch": 0.32,
"learning_rate": 0.00034447342026078237,
"loss": 2.8441,
"theoretical_loss": 3.6325505304933117,
"tokens_seen": 1049427968
},
{
"epoch": 0.32,
"learning_rate": 0.00034446339017051155,
"loss": 2.8109,
"theoretical_loss": 3.6325291348211,
"tokens_seen": 1049493504
},
{
"epoch": 0.32,
"learning_rate": 0.00034445336008024073,
"loss": 2.4558,
"theoretical_loss": 3.632507740858977,
"tokens_seen": 1049559040
},
{
"epoch": 0.32,
"learning_rate": 0.0003444433299899699,
"loss": 2.865,
"theoretical_loss": 3.6324863486067,
"tokens_seen": 1049624576
},
{
"epoch": 0.32,
"learning_rate": 0.00034443329989969915,
"loss": 2.7005,
"theoretical_loss": 3.632464958064025,
"tokens_seen": 1049690112
},
{
"epoch": 0.32,
"learning_rate": 0.0003444232698094283,
"loss": 2.4789,
"theoretical_loss": 3.6324435692307087,
"tokens_seen": 1049755648
},
{
"epoch": 0.32,
"learning_rate": 0.0003444132397191575,
"loss": 2.6313,
"theoretical_loss": 3.6324221821065077,
"tokens_seen": 1049821184
},
{
"epoch": 0.32,
"learning_rate": 0.00034440320962888664,
"loss": 3.0732,
"theoretical_loss": 3.6324007966911798,
"tokens_seen": 1049886720
},
{
"epoch": 0.32,
"learning_rate": 0.0003443931795386159,
"loss": 2.8584,
"theoretical_loss": 3.6323794129844806,
"tokens_seen": 1049952256
},
{
"epoch": 0.32,
"learning_rate": 0.00034438314944834506,
"loss": 2.8858,
"theoretical_loss": 3.6323580309861674,
"tokens_seen": 1050017792
},
{
"epoch": 0.32,
"learning_rate": 0.00034437311935807424,
"loss": 2.6757,
"theoretical_loss": 3.632336650695997,
"tokens_seen": 1050083328
},
{
"epoch": 0.32,
"learning_rate": 0.0003443630892678034,
"loss": 2.617,
"theoretical_loss": 3.6323152721137264,
"tokens_seen": 1050148864
},
{
"epoch": 0.32,
"objective/train/docs_used": 601638,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.944180727005005,
"objective/train/theoretical_loss": 3.6322938952391124,
"objective/train/tokens_used": 1070674400,
"theoretical_loss": 3.6322938952391124,
"tokens_seen": 1050214400
},
{
"epoch": 0.32,
"learning_rate": 0.0003443530591775326,
"loss": 2.789,
"theoretical_loss": 3.6322938952391124,
"tokens_seen": 1050214400
},
{
"epoch": 0.32,
"learning_rate": 0.0003443430290872618,
"loss": 2.8658,
"theoretical_loss": 3.632272520071912,
"tokens_seen": 1050279936
},
{
"epoch": 0.32,
"learning_rate": 0.000344332998996991,
"loss": 2.707,
"theoretical_loss": 3.632251146611883,
"tokens_seen": 1050345472
},
{
"epoch": 0.32,
"learning_rate": 0.00034432296890672014,
"loss": 2.6993,
"theoretical_loss": 3.632229774858782,
"tokens_seen": 1050411008
},
{
"epoch": 0.32,
"learning_rate": 0.0003443129388164494,
"loss": 2.898,
"theoretical_loss": 3.6322084048123666,
"tokens_seen": 1050476544
},
{
"epoch": 0.32,
"learning_rate": 0.00034430290872617856,
"loss": 2.662,
"theoretical_loss": 3.632187036472393,
"tokens_seen": 1050542080
},
{
"epoch": 0.32,
"learning_rate": 0.00034429287863590774,
"loss": 2.9507,
"theoretical_loss": 3.63216566983862,
"tokens_seen": 1050607616
},
{
"epoch": 0.32,
"learning_rate": 0.0003442828485456369,
"loss": 2.7941,
"theoretical_loss": 3.632144304910804,
"tokens_seen": 1050673152
},
{
"epoch": 0.32,
"learning_rate": 0.0003442728184553661,
"loss": 2.5593,
"theoretical_loss": 3.6321229416887024,
"tokens_seen": 1050738688
},
{
"epoch": 0.32,
"learning_rate": 0.0003442627883650953,
"loss": 2.6101,
"theoretical_loss": 3.6321015801720726,
"tokens_seen": 1050804224
},
{
"epoch": 0.32,
"learning_rate": 0.0003442527582748245,
"loss": 2.6657,
"theoretical_loss": 3.632080220360673,
"tokens_seen": 1050869760
},
{
"epoch": 0.32,
"learning_rate": 0.00034424272818455365,
"loss": 2.8881,
"theoretical_loss": 3.63205886225426,
"tokens_seen": 1050935296
},
{
"epoch": 0.32,
"learning_rate": 0.0003442326980942829,
"loss": 2.624,
"theoretical_loss": 3.6320375058525913,
"tokens_seen": 1051000832
},
{
"epoch": 0.32,
"learning_rate": 0.000344222668004012,
"loss": 2.7263,
"theoretical_loss": 3.632016151155425,
"tokens_seen": 1051066368
},
{
"epoch": 0.32,
"learning_rate": 0.00034421263791374124,
"loss": 2.6556,
"theoretical_loss": 3.631994798162519,
"tokens_seen": 1051131904
},
{
"epoch": 0.32,
"learning_rate": 0.0003442026078234704,
"loss": 2.8008,
"theoretical_loss": 3.631973446873631,
"tokens_seen": 1051197440
},
{
"epoch": 0.32,
"learning_rate": 0.0003441925777331996,
"loss": 2.7473,
"theoretical_loss": 3.6319520972885178,
"tokens_seen": 1051262976
},
{
"epoch": 0.32,
"learning_rate": 0.0003441825476429288,
"loss": 2.9124,
"theoretical_loss": 3.6319307494069384,
"tokens_seen": 1051328512
},
{
"epoch": 0.32,
"learning_rate": 0.00034417251755265797,
"loss": 2.6377,
"theoretical_loss": 3.63190940322865,
"tokens_seen": 1051394048
},
{
"epoch": 0.32,
"learning_rate": 0.00034416248746238715,
"loss": 2.7996,
"theoretical_loss": 3.631888058753411,
"tokens_seen": 1051459584
},
{
"epoch": 0.32,
"learning_rate": 0.0003441524573721164,
"loss": 2.7602,
"theoretical_loss": 3.6318667159809794,
"tokens_seen": 1051525120
},
{
"epoch": 0.32,
"learning_rate": 0.0003441424272818455,
"loss": 2.9705,
"theoretical_loss": 3.631845374911113,
"tokens_seen": 1051590656
},
{
"epoch": 0.32,
"learning_rate": 0.00034413239719157475,
"loss": 2.8155,
"theoretical_loss": 3.63182403554357,
"tokens_seen": 1051656192
},
{
"epoch": 0.32,
"learning_rate": 0.00034412236710130393,
"loss": 2.7239,
"theoretical_loss": 3.6318026978781077,
"tokens_seen": 1051721728
},
{
"epoch": 0.32,
"learning_rate": 0.0003441123370110331,
"loss": 2.896,
"theoretical_loss": 3.631781361914485,
"tokens_seen": 1051787264
},
{
"epoch": 0.32,
"objective/train/docs_used": 602388,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.871016025543213,
"objective/train/theoretical_loss": 3.631760027652461,
"objective/train/tokens_used": 1072312800,
"theoretical_loss": 3.631760027652461,
"tokens_seen": 1051852800
},
{
"epoch": 0.32,
"learning_rate": 0.0003441023069207623,
"loss": 2.7952,
"theoretical_loss": 3.631760027652461,
"tokens_seen": 1051852800
},
{
"epoch": 0.32,
"learning_rate": 0.00034409227683049147,
"loss": 2.8177,
"theoretical_loss": 3.631738695091793,
"tokens_seen": 1051918336
},
{
"epoch": 0.32,
"learning_rate": 0.00034408224674022065,
"loss": 2.7177,
"theoretical_loss": 3.631717364232239,
"tokens_seen": 1051983872
},
{
"epoch": 0.32,
"learning_rate": 0.0003440722166499499,
"loss": 2.7326,
"theoretical_loss": 3.6316960350735585,
"tokens_seen": 1052049408
},
{
"epoch": 0.32,
"learning_rate": 0.000344062186559679,
"loss": 3.0416,
"theoretical_loss": 3.631674707615509,
"tokens_seen": 1052114944
},
{
"epoch": 0.32,
"learning_rate": 0.00034405215646940825,
"loss": 2.8153,
"theoretical_loss": 3.6316533818578494,
"tokens_seen": 1052180480
},
{
"epoch": 0.32,
"learning_rate": 0.0003440421263791374,
"loss": 2.5938,
"theoretical_loss": 3.6316320578003385,
"tokens_seen": 1052246016
},
{
"epoch": 0.32,
"learning_rate": 0.0003440320962888666,
"loss": 2.7816,
"theoretical_loss": 3.631610735442734,
"tokens_seen": 1052311552
},
{
"epoch": 0.32,
"learning_rate": 0.0003440220661985958,
"loss": 2.8944,
"theoretical_loss": 3.631589414784795,
"tokens_seen": 1052377088
},
{
"epoch": 0.32,
"learning_rate": 0.000344012036108325,
"loss": 2.6378,
"theoretical_loss": 3.631568095826281,
"tokens_seen": 1052442624
},
{
"epoch": 0.32,
"learning_rate": 0.00034400200601805416,
"loss": 2.9245,
"theoretical_loss": 3.631546778566949,
"tokens_seen": 1052508160
},
{
"epoch": 0.32,
"learning_rate": 0.00034399197592778334,
"loss": 2.7578,
"theoretical_loss": 3.631525463006559,
"tokens_seen": 1052573696
},
{
"epoch": 0.32,
"learning_rate": 0.0003439819458375125,
"loss": 2.7361,
"theoretical_loss": 3.6315041491448703,
"tokens_seen": 1052639232
},
{
"epoch": 0.32,
"learning_rate": 0.00034397191574724175,
"loss": 2.8612,
"theoretical_loss": 3.6314828369816405,
"tokens_seen": 1052704768
},
{
"epoch": 0.32,
"learning_rate": 0.0003439618856569709,
"loss": 2.5115,
"theoretical_loss": 3.6314615265166292,
"tokens_seen": 1052770304
},
{
"epoch": 0.32,
"learning_rate": 0.0003439518555667001,
"loss": 2.7186,
"theoretical_loss": 3.6314402177495957,
"tokens_seen": 1052835840
},
{
"epoch": 0.32,
"learning_rate": 0.0003439418254764293,
"loss": 2.6994,
"theoretical_loss": 3.6314189106802974,
"tokens_seen": 1052901376
},
{
"epoch": 0.32,
"learning_rate": 0.0003439317953861585,
"loss": 2.6783,
"theoretical_loss": 3.6313976053084955,
"tokens_seen": 1052966912
},
{
"epoch": 0.32,
"learning_rate": 0.00034392176529588766,
"loss": 2.8136,
"theoretical_loss": 3.6313763016339484,
"tokens_seen": 1053032448
},
{
"epoch": 0.32,
"learning_rate": 0.00034391173520561684,
"loss": 2.7269,
"theoretical_loss": 3.6313549996564145,
"tokens_seen": 1053097984
},
{
"epoch": 0.32,
"learning_rate": 0.000343901705115346,
"loss": 2.4639,
"theoretical_loss": 3.631333699375654,
"tokens_seen": 1053163520
},
{
"epoch": 0.32,
"learning_rate": 0.00034389167502507526,
"loss": 2.5983,
"theoretical_loss": 3.6313124007914253,
"tokens_seen": 1053229056
},
{
"epoch": 0.32,
"learning_rate": 0.0003438816449348044,
"loss": 2.576,
"theoretical_loss": 3.6312911039034885,
"tokens_seen": 1053294592
},
{
"epoch": 0.32,
"learning_rate": 0.0003438716148445336,
"loss": 2.8648,
"theoretical_loss": 3.6312698087116027,
"tokens_seen": 1053360128
},
{
"epoch": 0.32,
"learning_rate": 0.00034386158475426275,
"loss": 2.8372,
"theoretical_loss": 3.6312485152155274,
"tokens_seen": 1053425664
},
{
"epoch": 0.32,
"objective/train/docs_used": 603451,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2299208641052246,
"objective/train/theoretical_loss": 3.6312272234150216,
"objective/train/tokens_used": 1073951200,
"theoretical_loss": 3.6312272234150216,
"tokens_seen": 1053491200
},
{
"epoch": 0.32,
"learning_rate": 0.000343851554663992,
"loss": 3.0082,
"theoretical_loss": 3.6312272234150216,
"tokens_seen": 1053491200
},
{
"epoch": 0.32,
"learning_rate": 0.0003438415245737212,
"loss": 2.622,
"theoretical_loss": 3.6312059333098454,
"tokens_seen": 1053556736
},
{
"epoch": 0.32,
"learning_rate": 0.00034383149448345034,
"loss": 2.5149,
"theoretical_loss": 3.631184644899758,
"tokens_seen": 1053622272
},
{
"epoch": 0.32,
"learning_rate": 0.0003438214643931796,
"loss": 2.8208,
"theoretical_loss": 3.6311633581845193,
"tokens_seen": 1053687808
},
{
"epoch": 0.32,
"learning_rate": 0.00034381143430290876,
"loss": 2.6723,
"theoretical_loss": 3.631142073163889,
"tokens_seen": 1053753344
},
{
"epoch": 0.32,
"learning_rate": 0.00034380140421263794,
"loss": 2.7109,
"theoretical_loss": 3.631120789837626,
"tokens_seen": 1053818880
},
{
"epoch": 0.32,
"learning_rate": 0.0003437913741223671,
"loss": 2.9199,
"theoretical_loss": 3.6310995082054913,
"tokens_seen": 1053884416
},
{
"epoch": 0.32,
"learning_rate": 0.0003437813440320963,
"loss": 2.7179,
"theoretical_loss": 3.631078228267244,
"tokens_seen": 1053949952
},
{
"epoch": 0.32,
"learning_rate": 0.0003437713139418255,
"loss": 2.4402,
"theoretical_loss": 3.6310569500226437,
"tokens_seen": 1054015488
},
{
"epoch": 0.32,
"learning_rate": 0.0003437612838515547,
"loss": 2.6728,
"theoretical_loss": 3.631035673471451,
"tokens_seen": 1054081024
},
{
"epoch": 0.32,
"learning_rate": 0.00034375125376128385,
"loss": 2.5186,
"theoretical_loss": 3.6310143986134253,
"tokens_seen": 1054146560
},
{
"epoch": 0.32,
"learning_rate": 0.0003437412236710131,
"loss": 2.7742,
"theoretical_loss": 3.6309931254483274,
"tokens_seen": 1054212096
},
{
"epoch": 0.32,
"learning_rate": 0.0003437311935807422,
"loss": 2.7916,
"theoretical_loss": 3.6309718539759164,
"tokens_seen": 1054277632
},
{
"epoch": 0.32,
"learning_rate": 0.00034372116349047144,
"loss": 2.6565,
"theoretical_loss": 3.630950584195953,
"tokens_seen": 1054343168
},
{
"epoch": 0.32,
"learning_rate": 0.0003437111334002006,
"loss": 2.8457,
"theoretical_loss": 3.6309293161081966,
"tokens_seen": 1054408704
},
{
"epoch": 0.32,
"learning_rate": 0.0003437011033099298,
"loss": 2.6985,
"theoretical_loss": 3.6309080497124087,
"tokens_seen": 1054474240
},
{
"epoch": 0.32,
"learning_rate": 0.000343691073219659,
"loss": 2.5254,
"theoretical_loss": 3.6308867850083484,
"tokens_seen": 1054539776
},
{
"epoch": 0.32,
"learning_rate": 0.00034368104312938817,
"loss": 2.7334,
"theoretical_loss": 3.6308655219957764,
"tokens_seen": 1054605312
},
{
"epoch": 0.32,
"learning_rate": 0.00034367101303911735,
"loss": 2.7792,
"theoretical_loss": 3.6308442606744533,
"tokens_seen": 1054670848
},
{
"epoch": 0.32,
"learning_rate": 0.0003436609829488466,
"loss": 2.6795,
"theoretical_loss": 3.6308230010441394,
"tokens_seen": 1054736384
},
{
"epoch": 0.32,
"learning_rate": 0.0003436509528585757,
"loss": 2.8233,
"theoretical_loss": 3.6308017431045947,
"tokens_seen": 1054801920
},
{
"epoch": 0.32,
"learning_rate": 0.00034364092276830495,
"loss": 2.7253,
"theoretical_loss": 3.6307804868555804,
"tokens_seen": 1054867456
},
{
"epoch": 0.32,
"learning_rate": 0.00034363089267803413,
"loss": 2.706,
"theoretical_loss": 3.6307592322968563,
"tokens_seen": 1054932992
},
{
"epoch": 0.32,
"learning_rate": 0.0003436208625877633,
"loss": 2.6864,
"theoretical_loss": 3.6307379794281838,
"tokens_seen": 1054998528
},
{
"epoch": 0.32,
"learning_rate": 0.0003436108324974925,
"loss": 2.9054,
"theoretical_loss": 3.6307167282493227,
"tokens_seen": 1055064064
},
{
"epoch": 0.32,
"objective/train/docs_used": 603451,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7547433376312256,
"objective/train/theoretical_loss": 3.630695478760034,
"objective/train/tokens_used": 1075589600,
"theoretical_loss": 3.630695478760034,
"tokens_seen": 1055129600
},
{
"epoch": 0.32,
"learning_rate": 0.00034360080240722167,
"loss": 2.9874,
"theoretical_loss": 3.630695478760034,
"tokens_seen": 1055129600
},
{
"epoch": 0.32,
"learning_rate": 0.00034359077231695085,
"loss": 2.8133,
"theoretical_loss": 3.6306742309600786,
"tokens_seen": 1055195136
},
{
"epoch": 0.32,
"learning_rate": 0.0003435807422266801,
"loss": 2.8307,
"theoretical_loss": 3.6306529848492177,
"tokens_seen": 1055260672
},
{
"epoch": 0.32,
"learning_rate": 0.0003435707121364092,
"loss": 2.7493,
"theoretical_loss": 3.6306317404272113,
"tokens_seen": 1055326208
},
{
"epoch": 0.32,
"learning_rate": 0.00034356068204613845,
"loss": 2.9727,
"theoretical_loss": 3.6306104976938203,
"tokens_seen": 1055391744
},
{
"epoch": 0.32,
"learning_rate": 0.0003435506519558676,
"loss": 2.8607,
"theoretical_loss": 3.6305892566488067,
"tokens_seen": 1055457280
},
{
"epoch": 0.32,
"learning_rate": 0.0003435406218655968,
"loss": 2.842,
"theoretical_loss": 3.6305680172919303,
"tokens_seen": 1055522816
},
{
"epoch": 0.32,
"learning_rate": 0.000343530591775326,
"loss": 2.7256,
"theoretical_loss": 3.6305467796229527,
"tokens_seen": 1055588352
},
{
"epoch": 0.32,
"learning_rate": 0.0003435205616850552,
"loss": 2.7126,
"theoretical_loss": 3.6305255436416353,
"tokens_seen": 1055653888
},
{
"epoch": 0.32,
"learning_rate": 0.00034351053159478436,
"loss": 2.8311,
"theoretical_loss": 3.6305043093477383,
"tokens_seen": 1055719424
},
{
"epoch": 0.32,
"learning_rate": 0.00034350050150451354,
"loss": 2.7722,
"theoretical_loss": 3.630483076741023,
"tokens_seen": 1055784960
},
{
"epoch": 0.32,
"learning_rate": 0.0003434904714142427,
"loss": 2.9931,
"theoretical_loss": 3.6304618458212516,
"tokens_seen": 1055850496
},
{
"epoch": 0.32,
"learning_rate": 0.00034348044132397195,
"loss": 2.9067,
"theoretical_loss": 3.630440616588185,
"tokens_seen": 1055916032
},
{
"epoch": 0.32,
"learning_rate": 0.0003434704112337011,
"loss": 2.8135,
"theoretical_loss": 3.630419389041584,
"tokens_seen": 1055981568
},
{
"epoch": 0.32,
"learning_rate": 0.0003434603811434303,
"loss": 2.7987,
"theoretical_loss": 3.6303981631812103,
"tokens_seen": 1056047104
},
{
"epoch": 0.32,
"learning_rate": 0.0003434503510531595,
"loss": 3.1312,
"theoretical_loss": 3.6303769390068257,
"tokens_seen": 1056112640
},
{
"epoch": 0.32,
"learning_rate": 0.0003434403209628887,
"loss": 2.7903,
"theoretical_loss": 3.6303557165181912,
"tokens_seen": 1056178176
},
{
"epoch": 0.32,
"learning_rate": 0.00034343029087261786,
"loss": 3.1406,
"theoretical_loss": 3.630334495715068,
"tokens_seen": 1056243712
},
{
"epoch": 0.32,
"learning_rate": 0.00034342026078234704,
"loss": 2.6818,
"theoretical_loss": 3.630313276597218,
"tokens_seen": 1056309248
},
{
"epoch": 0.32,
"learning_rate": 0.0003434102306920762,
"loss": 2.8386,
"theoretical_loss": 3.630292059164404,
"tokens_seen": 1056374784
},
{
"epoch": 0.32,
"learning_rate": 0.00034340020060180546,
"loss": 2.738,
"theoretical_loss": 3.630270843416385,
"tokens_seen": 1056440320
},
{
"epoch": 0.32,
"learning_rate": 0.0003433901705115346,
"loss": 2.8997,
"theoretical_loss": 3.6302496293529254,
"tokens_seen": 1056505856
},
{
"epoch": 0.32,
"learning_rate": 0.0003433801404212638,
"loss": 2.5948,
"theoretical_loss": 3.6302284169737855,
"tokens_seen": 1056571392
},
{
"epoch": 0.32,
"learning_rate": 0.00034337011033099295,
"loss": 2.983,
"theoretical_loss": 3.630207206278727,
"tokens_seen": 1056636928
},
{
"epoch": 0.32,
"learning_rate": 0.0003433600802407222,
"loss": 2.752,
"theoretical_loss": 3.630185997267513,
"tokens_seen": 1056702464
},
{
"epoch": 0.32,
"objective/train/docs_used": 603451,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9996352195739746,
"objective/train/theoretical_loss": 3.6301647899399043,
"objective/train/tokens_used": 1077228000,
"theoretical_loss": 3.6301647899399043,
"tokens_seen": 1056768000
},
{
"epoch": 0.32,
"learning_rate": 0.00034335005015045136,
"loss": 2.9905,
"theoretical_loss": 3.6301647899399043,
"tokens_seen": 1056768000
},
{
"epoch": 0.32,
"learning_rate": 0.00034334002006018054,
"loss": 2.6675,
"theoretical_loss": 3.630143584295663,
"tokens_seen": 1056833536
},
{
"epoch": 0.32,
"learning_rate": 0.0003433299899699097,
"loss": 3.0026,
"theoretical_loss": 3.6301223803345515,
"tokens_seen": 1056899072
},
{
"epoch": 0.32,
"learning_rate": 0.0003433199598796389,
"loss": 2.9595,
"theoretical_loss": 3.6301011780563313,
"tokens_seen": 1056964608
},
{
"epoch": 0.32,
"learning_rate": 0.0003433099297893681,
"loss": 3.0226,
"theoretical_loss": 3.630079977460765,
"tokens_seen": 1057030144
},
{
"epoch": 0.32,
"learning_rate": 0.0003432998996990973,
"loss": 2.6762,
"theoretical_loss": 3.630058778547615,
"tokens_seen": 1057095680
},
{
"epoch": 0.32,
"learning_rate": 0.00034328986960882645,
"loss": 2.8085,
"theoretical_loss": 3.630037581316642,
"tokens_seen": 1057161216
},
{
"epoch": 0.32,
"learning_rate": 0.0003432798395185557,
"loss": 2.825,
"theoretical_loss": 3.63001638576761,
"tokens_seen": 1057226752
},
{
"epoch": 0.32,
"learning_rate": 0.00034326980942828487,
"loss": 2.8789,
"theoretical_loss": 3.6299951919002806,
"tokens_seen": 1057292288
},
{
"epoch": 0.32,
"learning_rate": 0.00034325977933801405,
"loss": 2.6158,
"theoretical_loss": 3.6299739997144163,
"tokens_seen": 1057357824
},
{
"epoch": 0.32,
"learning_rate": 0.00034324974924774323,
"loss": 2.8906,
"theoretical_loss": 3.6299528092097795,
"tokens_seen": 1057423360
},
{
"epoch": 0.32,
"learning_rate": 0.0003432397191574724,
"loss": 2.8765,
"theoretical_loss": 3.629931620386132,
"tokens_seen": 1057488896
},
{
"epoch": 0.32,
"learning_rate": 0.0003432296890672016,
"loss": 2.5226,
"theoretical_loss": 3.6299104332432366,
"tokens_seen": 1057554432
},
{
"epoch": 0.32,
"learning_rate": 0.0003432196589769308,
"loss": 2.6325,
"theoretical_loss": 3.6298892477808566,
"tokens_seen": 1057619968
},
{
"epoch": 0.32,
"learning_rate": 0.00034320962888665995,
"loss": 2.7525,
"theoretical_loss": 3.6298680639987535,
"tokens_seen": 1057685504
},
{
"epoch": 0.32,
"learning_rate": 0.0003431995987963892,
"loss": 2.6333,
"theoretical_loss": 3.6298468818966905,
"tokens_seen": 1057751040
},
{
"epoch": 0.32,
"learning_rate": 0.0003431895687061183,
"loss": 2.816,
"theoretical_loss": 3.6298257014744304,
"tokens_seen": 1057816576
},
{
"epoch": 0.32,
"learning_rate": 0.00034317953861584755,
"loss": 2.8446,
"theoretical_loss": 3.6298045227317357,
"tokens_seen": 1057882112
},
{
"epoch": 0.32,
"learning_rate": 0.00034316950852557673,
"loss": 2.781,
"theoretical_loss": 3.6297833456683692,
"tokens_seen": 1057947648
},
{
"epoch": 0.32,
"learning_rate": 0.0003431594784353059,
"loss": 2.9066,
"theoretical_loss": 3.629762170284094,
"tokens_seen": 1058013184
},
{
"epoch": 0.32,
"learning_rate": 0.0003431494483450351,
"loss": 2.6623,
"theoretical_loss": 3.629740996578672,
"tokens_seen": 1058078720
},
{
"epoch": 0.32,
"learning_rate": 0.00034313941825476433,
"loss": 2.9219,
"theoretical_loss": 3.6297198245518674,
"tokens_seen": 1058144256
},
{
"epoch": 0.32,
"learning_rate": 0.00034312938816449345,
"loss": 2.5825,
"theoretical_loss": 3.629698654203442,
"tokens_seen": 1058209792
},
{
"epoch": 0.32,
"learning_rate": 0.0003431193580742227,
"loss": 2.732,
"theoretical_loss": 3.6296774855331604,
"tokens_seen": 1058275328
},
{
"epoch": 0.32,
"learning_rate": 0.0003431093279839518,
"loss": 2.7374,
"theoretical_loss": 3.6296563185407837,
"tokens_seen": 1058340864
},
{
"epoch": 0.32,
"objective/train/docs_used": 604503,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9277987480163574,
"objective/train/theoretical_loss": 3.6296351532260767,
"objective/train/tokens_used": 1078866400,
"theoretical_loss": 3.6296351532260767,
"tokens_seen": 1058406400
},
{
"epoch": 0.32,
"learning_rate": 0.00034309929789368105,
"loss": 2.8834,
"theoretical_loss": 3.6296351532260767,
"tokens_seen": 1058406400
},
{
"epoch": 0.32,
"learning_rate": 0.0003430892678034103,
"loss": 2.5189,
"theoretical_loss": 3.629613989588801,
"tokens_seen": 1058471936
},
{
"epoch": 0.32,
"learning_rate": 0.0003430792377131394,
"loss": 2.7527,
"theoretical_loss": 3.6295928276287217,
"tokens_seen": 1058537472
},
{
"epoch": 0.32,
"learning_rate": 0.00034306920762286865,
"loss": 2.6368,
"theoretical_loss": 3.6295716673456004,
"tokens_seen": 1058603008
},
{
"epoch": 0.32,
"learning_rate": 0.0003430591775325978,
"loss": 2.9654,
"theoretical_loss": 3.629550508739201,
"tokens_seen": 1058668544
},
{
"epoch": 0.32,
"learning_rate": 0.000343049147442327,
"loss": 2.8913,
"theoretical_loss": 3.6295293518092873,
"tokens_seen": 1058734080
},
{
"epoch": 0.32,
"learning_rate": 0.0003430391173520562,
"loss": 2.6603,
"theoretical_loss": 3.629508196555622,
"tokens_seen": 1058799616
},
{
"epoch": 0.32,
"learning_rate": 0.0003430290872617854,
"loss": 2.7419,
"theoretical_loss": 3.629487042977969,
"tokens_seen": 1058865152
},
{
"epoch": 0.32,
"learning_rate": 0.00034301905717151456,
"loss": 2.6629,
"theoretical_loss": 3.6294658910760913,
"tokens_seen": 1058930688
},
{
"epoch": 0.32,
"learning_rate": 0.00034300902708124374,
"loss": 2.8847,
"theoretical_loss": 3.6294447408497525,
"tokens_seen": 1058996224
},
{
"epoch": 0.32,
"learning_rate": 0.0003429989969909729,
"loss": 3.1126,
"theoretical_loss": 3.6294235922987177,
"tokens_seen": 1059061760
},
{
"epoch": 0.32,
"learning_rate": 0.00034298896690070215,
"loss": 3.1108,
"theoretical_loss": 3.6294024454227483,
"tokens_seen": 1059127296
},
{
"epoch": 0.32,
"learning_rate": 0.0003429789368104313,
"loss": 2.8361,
"theoretical_loss": 3.6293813002216093,
"tokens_seen": 1059192832
},
{
"epoch": 0.32,
"learning_rate": 0.0003429689067201605,
"loss": 2.8625,
"theoretical_loss": 3.629360156695064,
"tokens_seen": 1059258368
},
{
"epoch": 0.32,
"learning_rate": 0.0003429588766298897,
"loss": 2.7045,
"theoretical_loss": 3.629339014842876,
"tokens_seen": 1059323904
},
{
"epoch": 0.32,
"learning_rate": 0.0003429488465396189,
"loss": 2.8442,
"theoretical_loss": 3.6293178746648103,
"tokens_seen": 1059389440
},
{
"epoch": 0.32,
"learning_rate": 0.00034293881644934806,
"loss": 2.7541,
"theoretical_loss": 3.6292967361606294,
"tokens_seen": 1059454976
},
{
"epoch": 0.32,
"learning_rate": 0.00034292878635907724,
"loss": 2.7767,
"theoretical_loss": 3.6292755993300982,
"tokens_seen": 1059520512
},
{
"epoch": 0.32,
"learning_rate": 0.0003429187562688064,
"loss": 3.0202,
"theoretical_loss": 3.62925446417298,
"tokens_seen": 1059586048
},
{
"epoch": 0.32,
"learning_rate": 0.00034290872617853566,
"loss": 2.9265,
"theoretical_loss": 3.629233330689039,
"tokens_seen": 1059651584
},
{
"epoch": 0.32,
"learning_rate": 0.0003428986960882648,
"loss": 2.9686,
"theoretical_loss": 3.6292121988780393,
"tokens_seen": 1059717120
},
{
"epoch": 0.32,
"learning_rate": 0.000342888665997994,
"loss": 2.8697,
"theoretical_loss": 3.6291910687397455,
"tokens_seen": 1059782656
},
{
"epoch": 0.32,
"learning_rate": 0.00034287863590772315,
"loss": 2.7533,
"theoretical_loss": 3.629169940273921,
"tokens_seen": 1059848192
},
{
"epoch": 0.32,
"learning_rate": 0.0003428686058174524,
"loss": 2.6972,
"theoretical_loss": 3.6291488134803305,
"tokens_seen": 1059913728
},
{
"epoch": 0.32,
"learning_rate": 0.00034285857572718156,
"loss": 2.8273,
"theoretical_loss": 3.629127688358738,
"tokens_seen": 1059979264
},
{
"epoch": 0.32,
"objective/train/docs_used": 605021,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0615365505218506,
"objective/train/theoretical_loss": 3.6291065649089083,
"objective/train/tokens_used": 1080504800,
"theoretical_loss": 3.6291065649089083,
"tokens_seen": 1060044800
},
{
"epoch": 0.32,
"learning_rate": 0.00034284854563691074,
"loss": 2.9174,
"theoretical_loss": 3.6291065649089083,
"tokens_seen": 1060044800
},
{
"epoch": 0.32,
"learning_rate": 0.0003428385155466399,
"loss": 3.004,
"theoretical_loss": 3.629085443130605,
"tokens_seen": 1060110336
},
{
"epoch": 0.32,
"learning_rate": 0.0003428284854563691,
"loss": 2.7801,
"theoretical_loss": 3.6290643230235924,
"tokens_seen": 1060175872
},
{
"epoch": 0.32,
"learning_rate": 0.0003428184553660983,
"loss": 2.8507,
"theoretical_loss": 3.6290432045876364,
"tokens_seen": 1060241408
},
{
"epoch": 0.32,
"learning_rate": 0.0003428084252758275,
"loss": 2.7978,
"theoretical_loss": 3.6290220878225004,
"tokens_seen": 1060306944
},
{
"epoch": 0.32,
"learning_rate": 0.00034279839518555665,
"loss": 2.7992,
"theoretical_loss": 3.629000972727949,
"tokens_seen": 1060372480
},
{
"epoch": 0.32,
"learning_rate": 0.0003427883650952859,
"loss": 2.7377,
"theoretical_loss": 3.6289798593037466,
"tokens_seen": 1060438016
},
{
"epoch": 0.32,
"learning_rate": 0.00034277833500501507,
"loss": 2.9946,
"theoretical_loss": 3.6289587475496585,
"tokens_seen": 1060503552
},
{
"epoch": 0.32,
"learning_rate": 0.00034276830491474425,
"loss": 3.0868,
"theoretical_loss": 3.628937637465449,
"tokens_seen": 1060569088
},
{
"epoch": 0.32,
"learning_rate": 0.00034275827482447343,
"loss": 2.7121,
"theoretical_loss": 3.628916529050883,
"tokens_seen": 1060634624
},
{
"epoch": 0.32,
"learning_rate": 0.0003427482447342026,
"loss": 2.8935,
"theoretical_loss": 3.628895422305725,
"tokens_seen": 1060700160
},
{
"epoch": 0.32,
"learning_rate": 0.0003427382146439318,
"loss": 2.9228,
"theoretical_loss": 3.6288743172297404,
"tokens_seen": 1060765696
},
{
"epoch": 0.32,
"learning_rate": 0.000342728184553661,
"loss": 2.9529,
"theoretical_loss": 3.628853213822693,
"tokens_seen": 1060831232
},
{
"epoch": 0.32,
"learning_rate": 0.00034271815446339015,
"loss": 2.8673,
"theoretical_loss": 3.6288321120843494,
"tokens_seen": 1060896768
},
{
"epoch": 0.32,
"learning_rate": 0.0003427081243731194,
"loss": 2.646,
"theoretical_loss": 3.6288110120144728,
"tokens_seen": 1060962304
},
{
"epoch": 0.32,
"learning_rate": 0.0003426980942828485,
"loss": 2.9534,
"theoretical_loss": 3.6287899136128297,
"tokens_seen": 1061027840
},
{
"epoch": 0.32,
"learning_rate": 0.00034268806419257775,
"loss": 2.8805,
"theoretical_loss": 3.6287688168791843,
"tokens_seen": 1061093376
},
{
"epoch": 0.32,
"learning_rate": 0.00034267803410230693,
"loss": 2.7306,
"theoretical_loss": 3.6287477218133017,
"tokens_seen": 1061158912
},
{
"epoch": 0.32,
"learning_rate": 0.0003426680040120361,
"loss": 2.7265,
"theoretical_loss": 3.6287266284149475,
"tokens_seen": 1061224448
},
{
"epoch": 0.32,
"learning_rate": 0.0003426579739217653,
"loss": 2.7336,
"theoretical_loss": 3.6287055366838867,
"tokens_seen": 1061289984
},
{
"epoch": 0.32,
"learning_rate": 0.00034264794383149453,
"loss": 2.6095,
"theoretical_loss": 3.628684446619885,
"tokens_seen": 1061355520
},
{
"epoch": 0.32,
"learning_rate": 0.00034263791374122366,
"loss": 2.5724,
"theoretical_loss": 3.628663358222707,
"tokens_seen": 1061421056
},
{
"epoch": 0.32,
"learning_rate": 0.0003426278836509529,
"loss": 2.6912,
"theoretical_loss": 3.6286422714921187,
"tokens_seen": 1061486592
},
{
"epoch": 0.32,
"learning_rate": 0.000342617853560682,
"loss": 3.1256,
"theoretical_loss": 3.6286211864278846,
"tokens_seen": 1061552128
},
{
"epoch": 0.32,
"learning_rate": 0.00034260782347041125,
"loss": 2.7972,
"theoretical_loss": 3.628600103029771,
"tokens_seen": 1061617664
},
{
"epoch": 0.32,
"objective/train/docs_used": 606310,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.676234245300293,
"objective/train/theoretical_loss": 3.6285790212975435,
"objective/train/tokens_used": 1082143200,
"theoretical_loss": 3.6285790212975435,
"tokens_seen": 1061683200
},
{
"epoch": 0.32,
"learning_rate": 0.00034259779338014043,
"loss": 2.7698,
"theoretical_loss": 3.6285790212975435,
"tokens_seen": 1061683200
},
{
"epoch": 0.32,
"learning_rate": 0.0003425877632898696,
"loss": 2.6778,
"theoretical_loss": 3.6285579412309668,
"tokens_seen": 1061748736
},
{
"epoch": 0.32,
"learning_rate": 0.0003425777331995988,
"loss": 2.7597,
"theoretical_loss": 3.628536862829807,
"tokens_seen": 1061814272
},
{
"epoch": 0.32,
"learning_rate": 0.000342567703109328,
"loss": 2.8761,
"theoretical_loss": 3.62851578609383,
"tokens_seen": 1061879808
},
{
"epoch": 0.32,
"learning_rate": 0.00034255767301905716,
"loss": 2.6526,
"theoretical_loss": 3.6284947110228014,
"tokens_seen": 1061945344
},
{
"epoch": 0.32,
"learning_rate": 0.0003425476429287864,
"loss": 2.9541,
"theoretical_loss": 3.6284736376164872,
"tokens_seen": 1062010880
},
{
"epoch": 0.32,
"learning_rate": 0.0003425376128385155,
"loss": 2.9868,
"theoretical_loss": 3.6284525658746523,
"tokens_seen": 1062076416
},
{
"epoch": 0.32,
"learning_rate": 0.00034252758274824476,
"loss": 2.6102,
"theoretical_loss": 3.628431495797063,
"tokens_seen": 1062141952
},
{
"epoch": 0.32,
"learning_rate": 0.0003425175526579739,
"loss": 3.0963,
"theoretical_loss": 3.6284104273834847,
"tokens_seen": 1062207488
},
{
"epoch": 0.32,
"learning_rate": 0.0003425075225677031,
"loss": 2.7167,
"theoretical_loss": 3.628389360633684,
"tokens_seen": 1062273024
},
{
"epoch": 0.32,
"learning_rate": 0.0003424974924774323,
"loss": 2.7428,
"theoretical_loss": 3.6283682955474275,
"tokens_seen": 1062338560
},
{
"epoch": 0.32,
"learning_rate": 0.0003424874623871615,
"loss": 2.9964,
"theoretical_loss": 3.62834723212448,
"tokens_seen": 1062404096
},
{
"epoch": 0.32,
"learning_rate": 0.00034247743229689066,
"loss": 2.7255,
"theoretical_loss": 3.628326170364608,
"tokens_seen": 1062469632
},
{
"epoch": 0.32,
"learning_rate": 0.0003424674022066199,
"loss": 2.8482,
"theoretical_loss": 3.6283051102675774,
"tokens_seen": 1062535168
},
{
"epoch": 0.32,
"learning_rate": 0.000342457372116349,
"loss": 2.8758,
"theoretical_loss": 3.6282840518331545,
"tokens_seen": 1062600704
},
{
"epoch": 0.32,
"learning_rate": 0.00034244734202607826,
"loss": 2.8415,
"theoretical_loss": 3.628262995061106,
"tokens_seen": 1062666240
},
{
"epoch": 0.32,
"learning_rate": 0.0003424373119358074,
"loss": 2.8792,
"theoretical_loss": 3.628241939951198,
"tokens_seen": 1062731776
},
{
"epoch": 0.32,
"learning_rate": 0.0003424272818455366,
"loss": 2.8654,
"theoretical_loss": 3.628220886503196,
"tokens_seen": 1062797312
},
{
"epoch": 0.32,
"learning_rate": 0.0003424172517552658,
"loss": 2.9266,
"theoretical_loss": 3.628199834716867,
"tokens_seen": 1062862848
},
{
"epoch": 0.32,
"learning_rate": 0.000342407221664995,
"loss": 2.7639,
"theoretical_loss": 3.6281787845919777,
"tokens_seen": 1062928384
},
{
"epoch": 0.32,
"learning_rate": 0.00034239719157472416,
"loss": 2.8376,
"theoretical_loss": 3.628157736128294,
"tokens_seen": 1062993920
},
{
"epoch": 0.32,
"learning_rate": 0.00034238716148445335,
"loss": 2.7866,
"theoretical_loss": 3.6281366893255824,
"tokens_seen": 1063059456
},
{
"epoch": 0.32,
"learning_rate": 0.0003423771313941825,
"loss": 2.88,
"theoretical_loss": 3.6281156441836098,
"tokens_seen": 1063124992
},
{
"epoch": 0.32,
"learning_rate": 0.00034236710130391176,
"loss": 3.0008,
"theoretical_loss": 3.628094600702143,
"tokens_seen": 1063190528
},
{
"epoch": 0.32,
"learning_rate": 0.0003423570712136409,
"loss": 2.8461,
"theoretical_loss": 3.6280735588809474,
"tokens_seen": 1063256064
},
{
"epoch": 0.32,
"objective/train/docs_used": 606853,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7467336654663086,
"objective/train/theoretical_loss": 3.628052518719791,
"objective/train/tokens_used": 1083781600,
"theoretical_loss": 3.628052518719791,
"tokens_seen": 1063321600
},
{
"epoch": 0.32,
"learning_rate": 0.0003423470411233701,
"loss": 2.9097,
"theoretical_loss": 3.628052518719791,
"tokens_seen": 1063321600
},
{
"epoch": 0.32,
"learning_rate": 0.0003423370110330993,
"loss": 2.9078,
"theoretical_loss": 3.62803148021844,
"tokens_seen": 1063387136
},
{
"epoch": 0.32,
"learning_rate": 0.0003423269809428285,
"loss": 3.0148,
"theoretical_loss": 3.628010443376662,
"tokens_seen": 1063452672
},
{
"epoch": 0.32,
"learning_rate": 0.0003423169508525577,
"loss": 2.8218,
"theoretical_loss": 3.6279894081942223,
"tokens_seen": 1063518208
},
{
"epoch": 0.32,
"learning_rate": 0.00034230692076228685,
"loss": 2.8516,
"theoretical_loss": 3.6279683746708886,
"tokens_seen": 1063583744
},
{
"epoch": 0.32,
"learning_rate": 0.0003422968906720161,
"loss": 3.0888,
"theoretical_loss": 3.627947342806428,
"tokens_seen": 1063649280
},
{
"epoch": 0.32,
"learning_rate": 0.00034228686058174527,
"loss": 3.0361,
"theoretical_loss": 3.627926312600607,
"tokens_seen": 1063714816
},
{
"epoch": 0.32,
"learning_rate": 0.00034227683049147445,
"loss": 2.777,
"theoretical_loss": 3.6279052840531936,
"tokens_seen": 1063780352
},
{
"epoch": 0.32,
"learning_rate": 0.00034226680040120363,
"loss": 2.5332,
"theoretical_loss": 3.627884257163953,
"tokens_seen": 1063845888
},
{
"epoch": 0.32,
"learning_rate": 0.0003422567703109328,
"loss": 3.1086,
"theoretical_loss": 3.6278632319326545,
"tokens_seen": 1063911424
},
{
"epoch": 0.32,
"learning_rate": 0.000342246740220662,
"loss": 2.9834,
"theoretical_loss": 3.6278422083590636,
"tokens_seen": 1063976960
},
{
"epoch": 0.32,
"learning_rate": 0.0003422367101303912,
"loss": 2.7646,
"theoretical_loss": 3.627821186442948,
"tokens_seen": 1064042496
},
{
"epoch": 0.32,
"learning_rate": 0.00034222668004012035,
"loss": 2.7916,
"theoretical_loss": 3.627800166184076,
"tokens_seen": 1064108032
},
{
"epoch": 0.32,
"learning_rate": 0.0003422166499498496,
"loss": 3.0408,
"theoretical_loss": 3.6277791475822134,
"tokens_seen": 1064173568
},
{
"epoch": 0.32,
"learning_rate": 0.0003422066198595787,
"loss": 3.1587,
"theoretical_loss": 3.627758130637128,
"tokens_seen": 1064239104
},
{
"epoch": 0.32,
"learning_rate": 0.00034219658976930795,
"loss": 2.9047,
"theoretical_loss": 3.6277371153485873,
"tokens_seen": 1064304640
},
{
"epoch": 0.32,
"learning_rate": 0.00034218655967903713,
"loss": 2.8556,
"theoretical_loss": 3.6277161017163593,
"tokens_seen": 1064370176
},
{
"epoch": 0.32,
"learning_rate": 0.0003421765295887663,
"loss": 2.7572,
"theoretical_loss": 3.62769508974021,
"tokens_seen": 1064435712
},
{
"epoch": 0.32,
"learning_rate": 0.0003421664994984955,
"loss": 2.9897,
"theoretical_loss": 3.6276740794199087,
"tokens_seen": 1064501248
},
{
"epoch": 0.32,
"learning_rate": 0.00034215646940822473,
"loss": 2.8272,
"theoretical_loss": 3.6276530707552217,
"tokens_seen": 1064566784
},
{
"epoch": 0.32,
"learning_rate": 0.00034214643931795386,
"loss": 3.1123,
"theoretical_loss": 3.627632063745917,
"tokens_seen": 1064632320
},
{
"epoch": 0.32,
"learning_rate": 0.0003421364092276831,
"loss": 3.0017,
"theoretical_loss": 3.6276110583917625,
"tokens_seen": 1064697856
},
{
"epoch": 0.32,
"learning_rate": 0.0003421263791374122,
"loss": 3.1134,
"theoretical_loss": 3.6275900546925257,
"tokens_seen": 1064763392
},
{
"epoch": 0.32,
"learning_rate": 0.00034211634904714145,
"loss": 3.2282,
"theoretical_loss": 3.6275690526479747,
"tokens_seen": 1064828928
},
{
"epoch": 0.32,
"learning_rate": 0.00034210631895687063,
"loss": 2.8405,
"theoretical_loss": 3.627548052257877,
"tokens_seen": 1064894464
},
{
"epoch": 0.32,
"objective/train/docs_used": 607789,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.023940086364746,
"objective/train/theoretical_loss": 3.6275270535220008,
"objective/train/tokens_used": 1085420000,
"theoretical_loss": 3.6275270535220008,
"tokens_seen": 1064960000
},
{
"epoch": 0.32,
"learning_rate": 0.0003420962888665998,
"loss": 2.8383,
"theoretical_loss": 3.6275270535220008,
"tokens_seen": 1064960000
},
{
"epoch": 0.32,
"learning_rate": 0.000342086258776329,
"loss": 2.7306,
"theoretical_loss": 3.627506056440113,
"tokens_seen": 1065025536
},
{
"epoch": 0.32,
"learning_rate": 0.0003420762286860582,
"loss": 2.9994,
"theoretical_loss": 3.627485061011983,
"tokens_seen": 1065091072
},
{
"epoch": 0.32,
"learning_rate": 0.00034206619859578736,
"loss": 2.8171,
"theoretical_loss": 3.6274640672373772,
"tokens_seen": 1065156608
},
{
"epoch": 0.32,
"learning_rate": 0.0003420561685055166,
"loss": 2.9446,
"theoretical_loss": 3.6274430751160653,
"tokens_seen": 1065222144
},
{
"epoch": 0.32,
"learning_rate": 0.0003420461384152457,
"loss": 2.7752,
"theoretical_loss": 3.6274220846478142,
"tokens_seen": 1065287680
},
{
"epoch": 0.32,
"learning_rate": 0.00034203610832497496,
"loss": 3.0448,
"theoretical_loss": 3.627401095832393,
"tokens_seen": 1065353216
},
{
"epoch": 0.32,
"learning_rate": 0.0003420260782347041,
"loss": 2.9602,
"theoretical_loss": 3.627380108669569,
"tokens_seen": 1065418752
},
{
"epoch": 0.32,
"learning_rate": 0.0003420160481444333,
"loss": 2.7389,
"theoretical_loss": 3.6273591231591107,
"tokens_seen": 1065484288
},
{
"epoch": 0.32,
"learning_rate": 0.0003420060180541625,
"loss": 2.8129,
"theoretical_loss": 3.627338139300787,
"tokens_seen": 1065549824
},
{
"epoch": 0.32,
"learning_rate": 0.0003419959879638917,
"loss": 2.6285,
"theoretical_loss": 3.627317157094365,
"tokens_seen": 1065615360
},
{
"epoch": 0.32,
"learning_rate": 0.00034198595787362086,
"loss": 2.7106,
"theoretical_loss": 3.6272961765396143,
"tokens_seen": 1065680896
},
{
"epoch": 0.32,
"learning_rate": 0.0003419759277833501,
"loss": 2.8329,
"theoretical_loss": 3.627275197636303,
"tokens_seen": 1065746432
},
{
"epoch": 0.32,
"learning_rate": 0.0003419658976930792,
"loss": 2.87,
"theoretical_loss": 3.627254220384199,
"tokens_seen": 1065811968
},
{
"epoch": 0.32,
"learning_rate": 0.00034195586760280846,
"loss": 2.8234,
"theoretical_loss": 3.6272332447830715,
"tokens_seen": 1065877504
},
{
"epoch": 0.32,
"learning_rate": 0.0003419458375125376,
"loss": 2.9306,
"theoretical_loss": 3.6272122708326884,
"tokens_seen": 1065943040
},
{
"epoch": 0.32,
"learning_rate": 0.0003419358074222668,
"loss": 2.8147,
"theoretical_loss": 3.6271912985328187,
"tokens_seen": 1066008576
},
{
"epoch": 0.32,
"learning_rate": 0.000341925777331996,
"loss": 2.6402,
"theoretical_loss": 3.6271703278832312,
"tokens_seen": 1066074112
},
{
"epoch": 0.32,
"learning_rate": 0.0003419157472417252,
"loss": 2.8876,
"theoretical_loss": 3.6271493588836945,
"tokens_seen": 1066139648
},
{
"epoch": 0.32,
"learning_rate": 0.00034190571715145436,
"loss": 2.9706,
"theoretical_loss": 3.6271283915339776,
"tokens_seen": 1066205184
},
{
"epoch": 0.32,
"learning_rate": 0.00034189568706118355,
"loss": 2.8062,
"theoretical_loss": 3.6271074258338487,
"tokens_seen": 1066270720
},
{
"epoch": 0.32,
"learning_rate": 0.0003418856569709127,
"loss": 3.1038,
"theoretical_loss": 3.6270864617830765,
"tokens_seen": 1066336256
},
{
"epoch": 0.32,
"learning_rate": 0.00034187562688064196,
"loss": 2.9067,
"theoretical_loss": 3.627065499381431,
"tokens_seen": 1066401792
},
{
"epoch": 0.32,
"learning_rate": 0.0003418655967903711,
"loss": 2.9924,
"theoretical_loss": 3.62704453862868,
"tokens_seen": 1066467328
},
{
"epoch": 0.32,
"learning_rate": 0.0003418555667001003,
"loss": 2.6878,
"theoretical_loss": 3.627023579524593,
"tokens_seen": 1066532864
},
{
"epoch": 0.32,
"objective/train/docs_used": 608475,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1068174839019775,
"objective/train/theoretical_loss": 3.627002622068939,
"objective/train/tokens_used": 1087058400,
"theoretical_loss": 3.627002622068939,
"tokens_seen": 1066598400
},
{
"epoch": 0.32,
"learning_rate": 0.00034184553660982945,
"loss": 2.9949,
"theoretical_loss": 3.627002622068939,
"tokens_seen": 1066598400
},
{
"epoch": 0.32,
"learning_rate": 0.0003418355065195587,
"loss": 2.8873,
"theoretical_loss": 3.6269816662614875,
"tokens_seen": 1066663936
},
{
"epoch": 0.32,
"learning_rate": 0.00034182547642928787,
"loss": 2.9303,
"theoretical_loss": 3.626960712102006,
"tokens_seen": 1066729472
},
{
"epoch": 0.32,
"learning_rate": 0.00034181544633901705,
"loss": 2.8761,
"theoretical_loss": 3.6269397595902664,
"tokens_seen": 1066795008
},
{
"epoch": 0.32,
"learning_rate": 0.00034180541624874623,
"loss": 2.9039,
"theoretical_loss": 3.6269188087260353,
"tokens_seen": 1066860544
},
{
"epoch": 0.32,
"learning_rate": 0.00034179538615847547,
"loss": 2.9286,
"theoretical_loss": 3.6268978595090835,
"tokens_seen": 1066926080
},
{
"epoch": 0.32,
"learning_rate": 0.0003417853560682046,
"loss": 3.1704,
"theoretical_loss": 3.6268769119391795,
"tokens_seen": 1066991616
},
{
"epoch": 0.32,
"learning_rate": 0.00034177532597793383,
"loss": 2.8796,
"theoretical_loss": 3.6268559660160933,
"tokens_seen": 1067057152
},
{
"epoch": 0.32,
"learning_rate": 0.00034176529588766295,
"loss": 2.8998,
"theoretical_loss": 3.626835021739594,
"tokens_seen": 1067122688
},
{
"epoch": 0.32,
"learning_rate": 0.0003417552657973922,
"loss": 2.9396,
"theoretical_loss": 3.626814079109451,
"tokens_seen": 1067188224
},
{
"epoch": 0.32,
"learning_rate": 0.00034174523570712137,
"loss": 2.9509,
"theoretical_loss": 3.6267931381254335,
"tokens_seen": 1067253760
},
{
"epoch": 0.32,
"learning_rate": 0.00034173520561685055,
"loss": 2.9999,
"theoretical_loss": 3.626772198787312,
"tokens_seen": 1067319296
},
{
"epoch": 0.32,
"learning_rate": 0.00034172517552657973,
"loss": 2.7196,
"theoretical_loss": 3.6267512610948547,
"tokens_seen": 1067384832
},
{
"epoch": 0.32,
"learning_rate": 0.0003417151454363089,
"loss": 2.7615,
"theoretical_loss": 3.6267303250478324,
"tokens_seen": 1067450368
},
{
"epoch": 0.32,
"learning_rate": 0.0003417051153460381,
"loss": 2.9704,
"theoretical_loss": 3.626709390646014,
"tokens_seen": 1067515904
},
{
"epoch": 0.32,
"learning_rate": 0.00034169508525576733,
"loss": 3.0049,
"theoretical_loss": 3.6266884578891703,
"tokens_seen": 1067581440
},
{
"epoch": 0.32,
"learning_rate": 0.00034168505516549646,
"loss": 2.6171,
"theoretical_loss": 3.6266675267770703,
"tokens_seen": 1067646976
},
{
"epoch": 0.32,
"learning_rate": 0.0003416750250752257,
"loss": 2.9664,
"theoretical_loss": 3.6266465973094832,
"tokens_seen": 1067712512
},
{
"epoch": 0.32,
"learning_rate": 0.0003416649949849548,
"loss": 2.7568,
"theoretical_loss": 3.62662566948618,
"tokens_seen": 1067778048
},
{
"epoch": 0.32,
"learning_rate": 0.00034165496489468406,
"loss": 3.0979,
"theoretical_loss": 3.62660474330693,
"tokens_seen": 1067843584
},
{
"epoch": 0.32,
"learning_rate": 0.00034164493480441324,
"loss": 2.8891,
"theoretical_loss": 3.626583818771504,
"tokens_seen": 1067909120
},
{
"epoch": 0.32,
"learning_rate": 0.0003416349047141424,
"loss": 3.0236,
"theoretical_loss": 3.6265628958796707,
"tokens_seen": 1067974656
},
{
"epoch": 0.32,
"learning_rate": 0.0003416248746238716,
"loss": 2.8587,
"theoretical_loss": 3.626541974631201,
"tokens_seen": 1068040192
},
{
"epoch": 0.32,
"learning_rate": 0.00034161484453360083,
"loss": 3.0196,
"theoretical_loss": 3.626521055025864,
"tokens_seen": 1068105728
},
{
"epoch": 0.32,
"learning_rate": 0.00034160481444332996,
"loss": 2.9928,
"theoretical_loss": 3.6265001370634318,
"tokens_seen": 1068171264
},
{
"epoch": 0.32,
"objective/train/docs_used": 609657,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8975472450256348,
"objective/train/theoretical_loss": 3.626479220743673,
"objective/train/tokens_used": 1088696800,
"theoretical_loss": 3.626479220743673,
"tokens_seen": 1068236800
},
{
"epoch": 0.32,
"learning_rate": 0.0003415947843530592,
"loss": 2.842,
"theoretical_loss": 3.626479220743673,
"tokens_seen": 1068236800
},
{
"epoch": 0.32,
"learning_rate": 0.0003415847542627884,
"loss": 2.926,
"theoretical_loss": 3.6264583060663584,
"tokens_seen": 1068302336
},
{
"epoch": 0.32,
"learning_rate": 0.00034157472417251756,
"loss": 2.877,
"theoretical_loss": 3.6264373930312575,
"tokens_seen": 1068367872
},
{
"epoch": 0.32,
"learning_rate": 0.0003415646940822468,
"loss": 2.9445,
"theoretical_loss": 3.6264164816381417,
"tokens_seen": 1068433408
},
{
"epoch": 0.32,
"learning_rate": 0.0003415546639919759,
"loss": 2.9531,
"theoretical_loss": 3.626395571886781,
"tokens_seen": 1068498944
},
{
"epoch": 0.32,
"learning_rate": 0.00034154463390170516,
"loss": 2.7465,
"theoretical_loss": 3.626374663776946,
"tokens_seen": 1068564480
},
{
"epoch": 0.32,
"learning_rate": 0.0003415346038114343,
"loss": 3.0081,
"theoretical_loss": 3.6263537573084066,
"tokens_seen": 1068630016
},
{
"epoch": 0.32,
"learning_rate": 0.0003415245737211635,
"loss": 2.9763,
"theoretical_loss": 3.6263328524809335,
"tokens_seen": 1068695552
},
{
"epoch": 0.32,
"learning_rate": 0.0003415145436308927,
"loss": 2.899,
"theoretical_loss": 3.6263119492942977,
"tokens_seen": 1068761088
},
{
"epoch": 0.32,
"learning_rate": 0.0003415045135406219,
"loss": 2.9819,
"theoretical_loss": 3.6262910477482695,
"tokens_seen": 1068826624
},
{
"epoch": 0.32,
"learning_rate": 0.00034149448345035106,
"loss": 2.7849,
"theoretical_loss": 3.62627014784262,
"tokens_seen": 1068892160
},
{
"epoch": 0.32,
"learning_rate": 0.0003414844533600803,
"loss": 2.8588,
"theoretical_loss": 3.6262492495771186,
"tokens_seen": 1068957696
},
{
"epoch": 0.32,
"learning_rate": 0.0003414744232698094,
"loss": 2.8491,
"theoretical_loss": 3.6262283529515376,
"tokens_seen": 1069023232
},
{
"epoch": 0.32,
"learning_rate": 0.00034146439317953866,
"loss": 2.7678,
"theoretical_loss": 3.6262074579656467,
"tokens_seen": 1069088768
},
{
"epoch": 0.32,
"learning_rate": 0.0003414543630892678,
"loss": 2.9926,
"theoretical_loss": 3.626186564619217,
"tokens_seen": 1069154304
},
{
"epoch": 0.32,
"learning_rate": 0.000341444332998997,
"loss": 2.6953,
"theoretical_loss": 3.6261656729120197,
"tokens_seen": 1069219840
},
{
"epoch": 0.32,
"learning_rate": 0.0003414343029087262,
"loss": 2.903,
"theoretical_loss": 3.6261447828438254,
"tokens_seen": 1069285376
},
{
"epoch": 0.32,
"learning_rate": 0.0003414242728184554,
"loss": 2.8572,
"theoretical_loss": 3.6261238944144054,
"tokens_seen": 1069350912
},
{
"epoch": 0.32,
"learning_rate": 0.00034141424272818456,
"loss": 2.8379,
"theoretical_loss": 3.6261030076235303,
"tokens_seen": 1069416448
},
{
"epoch": 0.32,
"learning_rate": 0.00034140421263791375,
"loss": 2.8818,
"theoretical_loss": 3.6260821224709714,
"tokens_seen": 1069481984
},
{
"epoch": 0.32,
"learning_rate": 0.00034139418254764293,
"loss": 2.8425,
"theoretical_loss": 3.6260612389565,
"tokens_seen": 1069547520
},
{
"epoch": 0.32,
"learning_rate": 0.00034138415245737216,
"loss": 2.8203,
"theoretical_loss": 3.626040357079887,
"tokens_seen": 1069613056
},
{
"epoch": 0.32,
"learning_rate": 0.0003413741223671013,
"loss": 2.871,
"theoretical_loss": 3.626019476840904,
"tokens_seen": 1069678592
},
{
"epoch": 0.32,
"learning_rate": 0.0003413640922768305,
"loss": 2.9195,
"theoretical_loss": 3.6259985982393217,
"tokens_seen": 1069744128
},
{
"epoch": 0.32,
"learning_rate": 0.00034135406218655965,
"loss": 2.9318,
"theoretical_loss": 3.6259777212749116,
"tokens_seen": 1069809664
},
{
"epoch": 0.32,
"objective/train/docs_used": 610408,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7998862266540527,
"objective/train/theoretical_loss": 3.625956845947445,
"objective/train/tokens_used": 1090335200,
"theoretical_loss": 3.625956845947445,
"tokens_seen": 1069875200
},
{
"epoch": 0.32,
"learning_rate": 0.0003413440320962889,
"loss": 2.8033,
"theoretical_loss": 3.625956845947445,
"tokens_seen": 1069875200
},
{
"epoch": 0.32,
"learning_rate": 0.00034133400200601807,
"loss": 2.6016,
"theoretical_loss": 3.625935972256694,
"tokens_seen": 1069940736
},
{
"epoch": 0.32,
"learning_rate": 0.00034132397191574725,
"loss": 2.5468,
"theoretical_loss": 3.6259151002024286,
"tokens_seen": 1070006272
},
{
"epoch": 0.32,
"learning_rate": 0.00034131394182547643,
"loss": 2.8853,
"theoretical_loss": 3.6258942297844214,
"tokens_seen": 1070071808
},
{
"epoch": 0.32,
"learning_rate": 0.00034130391173520567,
"loss": 2.614,
"theoretical_loss": 3.6258733610024434,
"tokens_seen": 1070137344
},
{
"epoch": 0.32,
"learning_rate": 0.0003412938816449348,
"loss": 3.1269,
"theoretical_loss": 3.6258524938562666,
"tokens_seen": 1070202880
},
{
"epoch": 0.32,
"learning_rate": 0.00034128385155466403,
"loss": 2.6439,
"theoretical_loss": 3.6258316283456624,
"tokens_seen": 1070268416
},
{
"epoch": 0.32,
"learning_rate": 0.00034127382146439315,
"loss": 2.4498,
"theoretical_loss": 3.625810764470402,
"tokens_seen": 1070333952
},
{
"epoch": 0.32,
"learning_rate": 0.0003412637913741224,
"loss": 2.7657,
"theoretical_loss": 3.6257899022302578,
"tokens_seen": 1070399488
},
{
"epoch": 0.32,
"learning_rate": 0.00034125376128385157,
"loss": 2.9138,
"theoretical_loss": 3.6257690416250012,
"tokens_seen": 1070465024
},
{
"epoch": 0.32,
"learning_rate": 0.00034124373119358075,
"loss": 2.8376,
"theoretical_loss": 3.6257481826544047,
"tokens_seen": 1070530560
},
{
"epoch": 0.32,
"learning_rate": 0.00034123370110330993,
"loss": 2.8292,
"theoretical_loss": 3.6257273253182394,
"tokens_seen": 1070596096
},
{
"epoch": 0.32,
"learning_rate": 0.0003412236710130391,
"loss": 2.8643,
"theoretical_loss": 3.6257064696162766,
"tokens_seen": 1070661632
},
{
"epoch": 0.32,
"learning_rate": 0.0003412136409227683,
"loss": 2.8705,
"theoretical_loss": 3.62568561554829,
"tokens_seen": 1070727168
},
{
"epoch": 0.32,
"learning_rate": 0.00034120361083249753,
"loss": 2.9985,
"theoretical_loss": 3.6256647631140497,
"tokens_seen": 1070792704
},
{
"epoch": 0.32,
"learning_rate": 0.00034119358074222666,
"loss": 2.5494,
"theoretical_loss": 3.625643912313329,
"tokens_seen": 1070858240
},
{
"epoch": 0.32,
"learning_rate": 0.0003411835506519559,
"loss": 2.6166,
"theoretical_loss": 3.6256230631458997,
"tokens_seen": 1070923776
},
{
"epoch": 0.32,
"learning_rate": 0.000341173520561685,
"loss": 2.8007,
"theoretical_loss": 3.6256022156115337,
"tokens_seen": 1070989312
},
{
"epoch": 0.32,
"learning_rate": 0.00034116349047141426,
"loss": 2.747,
"theoretical_loss": 3.6255813697100034,
"tokens_seen": 1071054848
},
{
"epoch": 0.32,
"learning_rate": 0.00034115346038114344,
"loss": 2.846,
"theoretical_loss": 3.625560525441081,
"tokens_seen": 1071120384
},
{
"epoch": 0.32,
"learning_rate": 0.0003411434302908726,
"loss": 2.9109,
"theoretical_loss": 3.625539682804538,
"tokens_seen": 1071185920
},
{
"epoch": 0.32,
"learning_rate": 0.0003411334002006018,
"loss": 2.7876,
"theoretical_loss": 3.6255188418001483,
"tokens_seen": 1071251456
},
{
"epoch": 0.32,
"learning_rate": 0.00034112337011033103,
"loss": 3.1141,
"theoretical_loss": 3.6254980024276824,
"tokens_seen": 1071316992
},
{
"epoch": 0.32,
"learning_rate": 0.00034111334002006016,
"loss": 2.9731,
"theoretical_loss": 3.625477164686914,
"tokens_seen": 1071382528
},
{
"epoch": 0.32,
"learning_rate": 0.0003411033099297894,
"loss": 2.757,
"theoretical_loss": 3.625456328577615,
"tokens_seen": 1071448064
},
{
"epoch": 0.32,
"objective/train/docs_used": 611059,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1763269901275635,
"objective/train/theoretical_loss": 3.6254354940995586,
"objective/train/tokens_used": 1091973600,
"theoretical_loss": 3.6254354940995586,
"tokens_seen": 1071513600
},
{
"epoch": 0.32,
"learning_rate": 0.0003410932798395185,
"loss": 2.9363,
"theoretical_loss": 3.6254354940995586,
"tokens_seen": 1071513600
},
{
"epoch": 0.32,
"learning_rate": 0.00034108324974924776,
"loss": 2.8348,
"theoretical_loss": 3.625414661252516,
"tokens_seen": 1071579136
},
{
"epoch": 0.32,
"learning_rate": 0.00034107321965897694,
"loss": 2.5485,
"theoretical_loss": 3.625393830036261,
"tokens_seen": 1071644672
},
{
"epoch": 0.32,
"learning_rate": 0.0003410631895687061,
"loss": 2.7068,
"theoretical_loss": 3.6253730004505655,
"tokens_seen": 1071710208
},
{
"epoch": 0.32,
"learning_rate": 0.0003410531594784353,
"loss": 2.6271,
"theoretical_loss": 3.6253521724952025,
"tokens_seen": 1071775744
},
{
"epoch": 0.32,
"learning_rate": 0.0003410431293881645,
"loss": 2.7909,
"theoretical_loss": 3.6253313461699452,
"tokens_seen": 1071841280
},
{
"epoch": 0.32,
"learning_rate": 0.00034103309929789366,
"loss": 2.7227,
"theoretical_loss": 3.6253105214745647,
"tokens_seen": 1071906816
},
{
"epoch": 0.32,
"learning_rate": 0.0003410230692076229,
"loss": 2.7313,
"theoretical_loss": 3.625289698408836,
"tokens_seen": 1071972352
},
{
"epoch": 0.32,
"learning_rate": 0.000341013039117352,
"loss": 2.9958,
"theoretical_loss": 3.6252688769725303,
"tokens_seen": 1072037888
},
{
"epoch": 0.32,
"learning_rate": 0.00034100300902708126,
"loss": 2.6061,
"theoretical_loss": 3.625248057165421,
"tokens_seen": 1072103424
},
{
"epoch": 0.32,
"learning_rate": 0.00034099297893681044,
"loss": 2.8496,
"theoretical_loss": 3.6252272389872813,
"tokens_seen": 1072168960
},
{
"epoch": 0.32,
"learning_rate": 0.0003409829488465396,
"loss": 3.1983,
"theoretical_loss": 3.625206422437884,
"tokens_seen": 1072234496
},
{
"epoch": 0.32,
"learning_rate": 0.0003409729187562688,
"loss": 2.8964,
"theoretical_loss": 3.625185607517002,
"tokens_seen": 1072300032
},
{
"epoch": 0.32,
"learning_rate": 0.000340962888665998,
"loss": 2.8173,
"theoretical_loss": 3.6251647942244087,
"tokens_seen": 1072365568
},
{
"epoch": 0.32,
"learning_rate": 0.00034095285857572717,
"loss": 2.7593,
"theoretical_loss": 3.625143982559877,
"tokens_seen": 1072431104
},
{
"epoch": 0.32,
"learning_rate": 0.0003409428284854564,
"loss": 2.6857,
"theoretical_loss": 3.62512317252318,
"tokens_seen": 1072496640
},
{
"epoch": 0.33,
"learning_rate": 0.00034093279839518553,
"loss": 2.8435,
"theoretical_loss": 3.625102364114091,
"tokens_seen": 1072562176
},
{
"epoch": 0.33,
"learning_rate": 0.00034093279839518553,
"loss": 2.886,
"theoretical_loss": 3.6250815573323836,
"tokens_seen": 1072627712
},
{
"epoch": 0.33,
"learning_rate": 0.00034092276830491477,
"loss": 2.9396,
"theoretical_loss": 3.6250607521778306,
"tokens_seen": 1072693248
},
{
"epoch": 0.33,
"learning_rate": 0.0003409127382146439,
"loss": 2.5766,
"theoretical_loss": 3.6250399486502056,
"tokens_seen": 1072758784
},
{
"epoch": 0.33,
"learning_rate": 0.00034090270812437313,
"loss": 2.981,
"theoretical_loss": 3.6250191467492816,
"tokens_seen": 1072824320
},
{
"epoch": 0.33,
"learning_rate": 0.0003408926780341023,
"loss": 2.9541,
"theoretical_loss": 3.6249983464748325,
"tokens_seen": 1072889856
},
{
"epoch": 0.33,
"learning_rate": 0.0003408826479438315,
"loss": 2.7073,
"theoretical_loss": 3.624977547826632,
"tokens_seen": 1072955392
},
{
"epoch": 0.33,
"learning_rate": 0.00034087261785356067,
"loss": 2.9152,
"theoretical_loss": 3.6249567508044533,
"tokens_seen": 1073020928
},
{
"epoch": 0.33,
"learning_rate": 0.00034086258776328985,
"loss": 2.7047,
"theoretical_loss": 3.62493595540807,
"tokens_seen": 1073086464
},
{
"epoch": 0.33,
"objective/train/docs_used": 612309,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5029196739196777,
"objective/train/theoretical_loss": 3.6249151616372552,
"objective/train/tokens_used": 1093612000,
"theoretical_loss": 3.6249151616372552,
"tokens_seen": 1073152000
},
{
"epoch": 0.33,
"learning_rate": 0.00034085255767301903,
"loss": 2.6938,
"theoretical_loss": 3.6249151616372552,
"tokens_seen": 1073152000
},
{
"epoch": 0.33,
"learning_rate": 0.00034084252758274827,
"loss": 2.877,
"theoretical_loss": 3.6248943694917832,
"tokens_seen": 1073217536
},
{
"epoch": 0.33,
"learning_rate": 0.00034083249749247745,
"loss": 2.7699,
"theoretical_loss": 3.624873578971428,
"tokens_seen": 1073283072
},
{
"epoch": 0.33,
"learning_rate": 0.00034082246740220663,
"loss": 2.754,
"theoretical_loss": 3.6248527900759626,
"tokens_seen": 1073348608
},
{
"epoch": 0.33,
"learning_rate": 0.00034081243731193587,
"loss": 2.9606,
"theoretical_loss": 3.6248320028051615,
"tokens_seen": 1073414144
},
{
"epoch": 0.33,
"learning_rate": 0.000340802407221665,
"loss": 3.0962,
"theoretical_loss": 3.624811217158798,
"tokens_seen": 1073479680
},
{
"epoch": 0.33,
"learning_rate": 0.00034079237713139423,
"loss": 3.0965,
"theoretical_loss": 3.6247904331366465,
"tokens_seen": 1073545216
},
{
"epoch": 0.33,
"learning_rate": 0.00034078234704112335,
"loss": 2.8786,
"theoretical_loss": 3.6247696507384806,
"tokens_seen": 1073610752
},
{
"epoch": 0.33,
"learning_rate": 0.0003407723169508526,
"loss": 2.7591,
"theoretical_loss": 3.6247488699640735,
"tokens_seen": 1073676288
},
{
"epoch": 0.33,
"learning_rate": 0.00034076228686058177,
"loss": 2.8041,
"theoretical_loss": 3.6247280908132007,
"tokens_seen": 1073741824
},
{
"epoch": 0.33,
"learning_rate": 0.00034075225677031095,
"loss": 3.1,
"theoretical_loss": 3.6247073132856364,
"tokens_seen": 1073807360
},
{
"epoch": 0.33,
"learning_rate": 0.00034074222668004013,
"loss": 2.7407,
"theoretical_loss": 3.624686537381153,
"tokens_seen": 1073872896
},
{
"epoch": 0.33,
"learning_rate": 0.0003407321965897693,
"loss": 2.8795,
"theoretical_loss": 3.624665763099526,
"tokens_seen": 1073938432
},
{
"epoch": 0.33,
"learning_rate": 0.0003407221664994985,
"loss": 2.7453,
"theoretical_loss": 3.6246449904405287,
"tokens_seen": 1074003968
},
{
"epoch": 0.33,
"learning_rate": 0.00034071213640922773,
"loss": 2.7588,
"theoretical_loss": 3.6246242194039366,
"tokens_seen": 1074069504
},
{
"epoch": 0.33,
"learning_rate": 0.00034070210631895686,
"loss": 2.8376,
"theoretical_loss": 3.6246034499895234,
"tokens_seen": 1074135040
},
{
"epoch": 0.33,
"learning_rate": 0.0003406920762286861,
"loss": 3.0039,
"theoretical_loss": 3.6245826821970626,
"tokens_seen": 1074200576
},
{
"epoch": 0.33,
"learning_rate": 0.0003406820461384152,
"loss": 2.8109,
"theoretical_loss": 3.6245619160263303,
"tokens_seen": 1074266112
},
{
"epoch": 0.33,
"learning_rate": 0.00034067201604814446,
"loss": 2.5549,
"theoretical_loss": 3.624541151477099,
"tokens_seen": 1074331648
},
{
"epoch": 0.33,
"learning_rate": 0.00034066198595787364,
"loss": 2.9192,
"theoretical_loss": 3.6245203885491453,
"tokens_seen": 1074397184
},
{
"epoch": 0.33,
"learning_rate": 0.0003406519558676028,
"loss": 2.9521,
"theoretical_loss": 3.6244996272422414,
"tokens_seen": 1074462720
},
{
"epoch": 0.33,
"learning_rate": 0.000340641925777332,
"loss": 2.7991,
"theoretical_loss": 3.624478867556164,
"tokens_seen": 1074528256
},
{
"epoch": 0.33,
"learning_rate": 0.00034063189568706123,
"loss": 2.9081,
"theoretical_loss": 3.6244581094906865,
"tokens_seen": 1074593792
},
{
"epoch": 0.33,
"learning_rate": 0.00034062186559679036,
"loss": 2.6443,
"theoretical_loss": 3.6244373530455842,
"tokens_seen": 1074659328
},
{
"epoch": 0.33,
"learning_rate": 0.0003406118355065196,
"loss": 2.908,
"theoretical_loss": 3.624416598220631,
"tokens_seen": 1074724864
},
{
"epoch": 0.33,
"objective/train/docs_used": 613082,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.583402156829834,
"objective/train/theoretical_loss": 3.624395845015602,
"objective/train/tokens_used": 1095250400,
"theoretical_loss": 3.624395845015602,
"tokens_seen": 1074790400
},
{
"epoch": 0.33,
"learning_rate": 0.0003406018054162487,
"loss": 2.711,
"theoretical_loss": 3.624395845015602,
"tokens_seen": 1074790400
},
{
"epoch": 0.33,
"learning_rate": 0.00034059177532597796,
"loss": 2.949,
"theoretical_loss": 3.624375093430273,
"tokens_seen": 1074855936
},
{
"epoch": 0.33,
"learning_rate": 0.00034058174523570714,
"loss": 2.9351,
"theoretical_loss": 3.624354343464417,
"tokens_seen": 1074921472
},
{
"epoch": 0.33,
"learning_rate": 0.0003405717151454363,
"loss": 2.9132,
"theoretical_loss": 3.62433359511781,
"tokens_seen": 1074987008
},
{
"epoch": 0.33,
"learning_rate": 0.0003405616850551655,
"loss": 2.7919,
"theoretical_loss": 3.6243128483902276,
"tokens_seen": 1075052544
},
{
"epoch": 0.33,
"learning_rate": 0.0003405516549648947,
"loss": 2.9655,
"theoretical_loss": 3.6242921032814435,
"tokens_seen": 1075118080
},
{
"epoch": 0.33,
"learning_rate": 0.00034054162487462386,
"loss": 2.8059,
"theoretical_loss": 3.6242713597912335,
"tokens_seen": 1075183616
},
{
"epoch": 0.33,
"learning_rate": 0.0003405315947843531,
"loss": 2.9168,
"theoretical_loss": 3.6242506179193716,
"tokens_seen": 1075249152
},
{
"epoch": 0.33,
"learning_rate": 0.0003405215646940822,
"loss": 2.8958,
"theoretical_loss": 3.624229877665634,
"tokens_seen": 1075314688
},
{
"epoch": 0.33,
"learning_rate": 0.00034051153460381146,
"loss": 2.6823,
"theoretical_loss": 3.624209139029796,
"tokens_seen": 1075380224
},
{
"epoch": 0.33,
"learning_rate": 0.00034050150451354064,
"loss": 2.876,
"theoretical_loss": 3.624188402011632,
"tokens_seen": 1075445760
},
{
"epoch": 0.33,
"learning_rate": 0.0003404914744232698,
"loss": 2.8416,
"theoretical_loss": 3.624167666610918,
"tokens_seen": 1075511296
},
{
"epoch": 0.33,
"learning_rate": 0.000340481444332999,
"loss": 2.9374,
"theoretical_loss": 3.6241469328274283,
"tokens_seen": 1075576832
},
{
"epoch": 0.33,
"learning_rate": 0.0003404714142427282,
"loss": 2.7441,
"theoretical_loss": 3.6241262006609394,
"tokens_seen": 1075642368
},
{
"epoch": 0.33,
"learning_rate": 0.00034046138415245737,
"loss": 2.805,
"theoretical_loss": 3.6241054701112256,
"tokens_seen": 1075707904
},
{
"epoch": 0.33,
"learning_rate": 0.0003404513540621866,
"loss": 2.8264,
"theoretical_loss": 3.6240847411780637,
"tokens_seen": 1075773440
},
{
"epoch": 0.33,
"learning_rate": 0.00034044132397191573,
"loss": 2.8649,
"theoretical_loss": 3.6240640138612275,
"tokens_seen": 1075838976
},
{
"epoch": 0.33,
"learning_rate": 0.00034043129388164497,
"loss": 2.7754,
"theoretical_loss": 3.6240432881604936,
"tokens_seen": 1075904512
},
{
"epoch": 0.33,
"learning_rate": 0.0003404212637913741,
"loss": 2.6649,
"theoretical_loss": 3.6240225640756374,
"tokens_seen": 1075970048
},
{
"epoch": 0.33,
"learning_rate": 0.00034041123370110333,
"loss": 2.6012,
"theoretical_loss": 3.624001841606434,
"tokens_seen": 1076035584
},
{
"epoch": 0.33,
"learning_rate": 0.0003404012036108325,
"loss": 3.0185,
"theoretical_loss": 3.62398112075266,
"tokens_seen": 1076101120
},
{
"epoch": 0.33,
"learning_rate": 0.0003403911735205617,
"loss": 2.657,
"theoretical_loss": 3.6239604015140907,
"tokens_seen": 1076166656
},
{
"epoch": 0.33,
"learning_rate": 0.00034038114343029087,
"loss": 2.8257,
"theoretical_loss": 3.6239396838905016,
"tokens_seen": 1076232192
},
{
"epoch": 0.33,
"learning_rate": 0.00034037111334002005,
"loss": 2.8672,
"theoretical_loss": 3.623918967881668,
"tokens_seen": 1076297728
},
{
"epoch": 0.33,
"learning_rate": 0.00034036108324974923,
"loss": 2.8801,
"theoretical_loss": 3.6238982534873667,
"tokens_seen": 1076363264
},
{
"epoch": 0.33,
"objective/train/docs_used": 614240,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9492299556732178,
"objective/train/theoretical_loss": 3.6238775407073733,
"objective/train/tokens_used": 1096888800,
"theoretical_loss": 3.6238775407073733,
"tokens_seen": 1076428800
},
{
"epoch": 0.33,
"learning_rate": 0.00034035105315947847,
"loss": 2.881,
"theoretical_loss": 3.6238775407073733,
"tokens_seen": 1076428800
},
{
"epoch": 0.33,
"learning_rate": 0.0003403410230692076,
"loss": 2.6599,
"theoretical_loss": 3.623856829541464,
"tokens_seen": 1076494336
},
{
"epoch": 0.33,
"learning_rate": 0.00034033099297893683,
"loss": 2.9966,
"theoretical_loss": 3.623836119989414,
"tokens_seen": 1076559872
},
{
"epoch": 0.33,
"learning_rate": 0.000340320962888666,
"loss": 3.0381,
"theoretical_loss": 3.6238154120509996,
"tokens_seen": 1076625408
},
{
"epoch": 0.33,
"learning_rate": 0.0003403109327983952,
"loss": 2.6261,
"theoretical_loss": 3.623794705725997,
"tokens_seen": 1076690944
},
{
"epoch": 0.33,
"learning_rate": 0.0003403009027081244,
"loss": 2.8024,
"theoretical_loss": 3.6237740010141826,
"tokens_seen": 1076756480
},
{
"epoch": 0.33,
"learning_rate": 0.00034029087261785356,
"loss": 2.9619,
"theoretical_loss": 3.6237532979153317,
"tokens_seen": 1076822016
},
{
"epoch": 0.33,
"learning_rate": 0.00034028084252758274,
"loss": 2.8307,
"theoretical_loss": 3.6237325964292215,
"tokens_seen": 1076887552
},
{
"epoch": 0.33,
"learning_rate": 0.00034027081243731197,
"loss": 2.8706,
"theoretical_loss": 3.6237118965556276,
"tokens_seen": 1076953088
},
{
"epoch": 0.33,
"learning_rate": 0.0003402607823470411,
"loss": 2.8094,
"theoretical_loss": 3.6236911982943267,
"tokens_seen": 1077018624
},
{
"epoch": 0.33,
"learning_rate": 0.00034025075225677033,
"loss": 2.9059,
"theoretical_loss": 3.623670501645094,
"tokens_seen": 1077084160
},
{
"epoch": 0.33,
"learning_rate": 0.00034024072216649946,
"loss": 2.7788,
"theoretical_loss": 3.623649806607707,
"tokens_seen": 1077149696
},
{
"epoch": 0.33,
"learning_rate": 0.0003402306920762287,
"loss": 2.7341,
"theoretical_loss": 3.6236291131819427,
"tokens_seen": 1077215232
},
{
"epoch": 0.33,
"learning_rate": 0.0003402206619859579,
"loss": 2.7524,
"theoretical_loss": 3.623608421367576,
"tokens_seen": 1077280768
},
{
"epoch": 0.33,
"learning_rate": 0.00034021063189568706,
"loss": 3.0065,
"theoretical_loss": 3.623587731164384,
"tokens_seen": 1077346304
},
{
"epoch": 0.33,
"learning_rate": 0.00034020060180541624,
"loss": 2.9128,
"theoretical_loss": 3.6235670425721436,
"tokens_seen": 1077411840
},
{
"epoch": 0.33,
"learning_rate": 0.0003401905717151454,
"loss": 2.8977,
"theoretical_loss": 3.623546355590631,
"tokens_seen": 1077477376
},
{
"epoch": 0.33,
"learning_rate": 0.0003401805416248746,
"loss": 2.8233,
"theoretical_loss": 3.623525670219623,
"tokens_seen": 1077542912
},
{
"epoch": 0.33,
"learning_rate": 0.00034017051153460384,
"loss": 2.9469,
"theoretical_loss": 3.6235049864588964,
"tokens_seen": 1077608448
},
{
"epoch": 0.33,
"learning_rate": 0.00034016048144433296,
"loss": 2.8139,
"theoretical_loss": 3.623484304308228,
"tokens_seen": 1077673984
},
{
"epoch": 0.33,
"learning_rate": 0.0003401504513540622,
"loss": 2.9174,
"theoretical_loss": 3.6234636237673934,
"tokens_seen": 1077739520
},
{
"epoch": 0.33,
"learning_rate": 0.0003401404212637914,
"loss": 2.7571,
"theoretical_loss": 3.623442944836171,
"tokens_seen": 1077805056
},
{
"epoch": 0.33,
"learning_rate": 0.00034013039117352056,
"loss": 2.8011,
"theoretical_loss": 3.623422267514337,
"tokens_seen": 1077870592
},
{
"epoch": 0.33,
"learning_rate": 0.00034012036108324974,
"loss": 2.8221,
"theoretical_loss": 3.6234015918016684,
"tokens_seen": 1077936128
},
{
"epoch": 0.33,
"learning_rate": 0.0003401103309929789,
"loss": 2.8289,
"theoretical_loss": 3.6233809176979417,
"tokens_seen": 1078001664
},
{
"epoch": 0.33,
"objective/train/docs_used": 615029,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.288437843322754,
"objective/train/theoretical_loss": 3.6233602452029348,
"objective/train/tokens_used": 1098527200,
"theoretical_loss": 3.6233602452029348,
"tokens_seen": 1078067200
},
{
"epoch": 0.33,
"learning_rate": 0.0003401003009027081,
"loss": 2.7948,
"theoretical_loss": 3.6233602452029348,
"tokens_seen": 1078067200
},
{
"epoch": 0.33,
"learning_rate": 0.00034009027081243734,
"loss": 2.7555,
"theoretical_loss": 3.6233395743164243,
"tokens_seen": 1078132736
},
{
"epoch": 0.33,
"learning_rate": 0.0003400802407221665,
"loss": 3.1052,
"theoretical_loss": 3.623318905038187,
"tokens_seen": 1078198272
},
{
"epoch": 0.33,
"learning_rate": 0.0003400702106318957,
"loss": 2.9238,
"theoretical_loss": 3.6232982373679996,
"tokens_seen": 1078263808
},
{
"epoch": 0.33,
"learning_rate": 0.0003400601805416249,
"loss": 2.6923,
"theoretical_loss": 3.6232775713056404,
"tokens_seen": 1078329344
},
{
"epoch": 0.33,
"learning_rate": 0.00034005015045135406,
"loss": 2.8902,
"theoretical_loss": 3.623256906850886,
"tokens_seen": 1078394880
},
{
"epoch": 0.33,
"learning_rate": 0.0003400401203610833,
"loss": 2.9232,
"theoretical_loss": 3.6232362440035137,
"tokens_seen": 1078460416
},
{
"epoch": 0.33,
"learning_rate": 0.0003400300902708124,
"loss": 2.8736,
"theoretical_loss": 3.623215582763301,
"tokens_seen": 1078525952
},
{
"epoch": 0.33,
"learning_rate": 0.00034002006018054166,
"loss": 3.0412,
"theoretical_loss": 3.623194923130025,
"tokens_seen": 1078591488
},
{
"epoch": 0.33,
"learning_rate": 0.00034001003009027084,
"loss": 2.7477,
"theoretical_loss": 3.6231742651034633,
"tokens_seen": 1078657024
},
{
"epoch": 0.33,
"learning_rate": 0.00034,
"loss": 2.9183,
"theoretical_loss": 3.623153608683393,
"tokens_seen": 1078722560
},
{
"epoch": 0.33,
"learning_rate": 0.0003399899699097292,
"loss": 3.075,
"theoretical_loss": 3.623132953869592,
"tokens_seen": 1078788096
},
{
"epoch": 0.33,
"learning_rate": 0.0003399799398194584,
"loss": 2.8524,
"theoretical_loss": 3.6231123006618375,
"tokens_seen": 1078853632
},
{
"epoch": 0.33,
"learning_rate": 0.00033996990972918757,
"loss": 2.5494,
"theoretical_loss": 3.6230916490599077,
"tokens_seen": 1078919168
},
{
"epoch": 0.33,
"learning_rate": 0.0003399598796389168,
"loss": 2.8883,
"theoretical_loss": 3.623070999063579,
"tokens_seen": 1078984704
},
{
"epoch": 0.33,
"learning_rate": 0.00033994984954864593,
"loss": 2.699,
"theoretical_loss": 3.6230503506726306,
"tokens_seen": 1079050240
},
{
"epoch": 0.33,
"learning_rate": 0.00033993981945837517,
"loss": 2.7665,
"theoretical_loss": 3.623029703886839,
"tokens_seen": 1079115776
},
{
"epoch": 0.33,
"learning_rate": 0.0003399297893681043,
"loss": 2.3742,
"theoretical_loss": 3.623009058705982,
"tokens_seen": 1079181312
},
{
"epoch": 0.33,
"learning_rate": 0.00033991975927783353,
"loss": 2.6969,
"theoretical_loss": 3.622988415129838,
"tokens_seen": 1079246848
},
{
"epoch": 0.33,
"learning_rate": 0.0003399097291875627,
"loss": 2.8683,
"theoretical_loss": 3.6229677731581846,
"tokens_seen": 1079312384
},
{
"epoch": 0.33,
"learning_rate": 0.0003398996990972919,
"loss": 2.7275,
"theoretical_loss": 3.6229471327907996,
"tokens_seen": 1079377920
},
{
"epoch": 0.33,
"learning_rate": 0.00033988966900702107,
"loss": 2.6161,
"theoretical_loss": 3.6229264940274613,
"tokens_seen": 1079443456
},
{
"epoch": 0.33,
"learning_rate": 0.00033987963891675025,
"loss": 2.8862,
"theoretical_loss": 3.6229058568679466,
"tokens_seen": 1079508992
},
{
"epoch": 0.33,
"learning_rate": 0.00033986960882647943,
"loss": 2.5894,
"theoretical_loss": 3.6228852213120346,
"tokens_seen": 1079574528
},
{
"epoch": 0.33,
"learning_rate": 0.00033985957873620867,
"loss": 2.7033,
"theoretical_loss": 3.622864587359503,
"tokens_seen": 1079640064
},
{
"epoch": 0.33,
"objective/train/docs_used": 616160,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7225987911224365,
"objective/train/theoretical_loss": 3.62284395501013,
"objective/train/tokens_used": 1100165600,
"theoretical_loss": 3.62284395501013,
"tokens_seen": 1079705600
},
{
"epoch": 0.33,
"learning_rate": 0.0003398495486459378,
"loss": 2.7249,
"theoretical_loss": 3.62284395501013,
"tokens_seen": 1079705600
},
{
"epoch": 0.33,
"learning_rate": 0.00033983951855566703,
"loss": 2.7454,
"theoretical_loss": 3.622823324263694,
"tokens_seen": 1079771136
},
{
"epoch": 0.33,
"learning_rate": 0.0003398294884653962,
"loss": 2.7357,
"theoretical_loss": 3.6228026951199723,
"tokens_seen": 1079836672
},
{
"epoch": 0.33,
"learning_rate": 0.0003398194583751254,
"loss": 2.9365,
"theoretical_loss": 3.6227820675787443,
"tokens_seen": 1079902208
},
{
"epoch": 0.33,
"learning_rate": 0.0003398094282848546,
"loss": 2.8477,
"theoretical_loss": 3.6227614416397866,
"tokens_seen": 1079967744
},
{
"epoch": 0.33,
"learning_rate": 0.00033979939819458376,
"loss": 2.8644,
"theoretical_loss": 3.6227408173028794,
"tokens_seen": 1080033280
},
{
"epoch": 0.33,
"learning_rate": 0.00033978936810431294,
"loss": 2.8412,
"theoretical_loss": 3.6227201945678003,
"tokens_seen": 1080098816
},
{
"epoch": 0.33,
"learning_rate": 0.00033977933801404217,
"loss": 2.8175,
"theoretical_loss": 3.6226995734343275,
"tokens_seen": 1080164352
},
{
"epoch": 0.33,
"learning_rate": 0.0003397693079237713,
"loss": 2.783,
"theoretical_loss": 3.6226789539022395,
"tokens_seen": 1080229888
},
{
"epoch": 0.33,
"learning_rate": 0.00033975927783350053,
"loss": 3.0381,
"theoretical_loss": 3.6226583359713147,
"tokens_seen": 1080295424
},
{
"epoch": 0.33,
"learning_rate": 0.00033974924774322966,
"loss": 2.7598,
"theoretical_loss": 3.622637719641332,
"tokens_seen": 1080360960
},
{
"epoch": 0.33,
"learning_rate": 0.0003397392176529589,
"loss": 3.104,
"theoretical_loss": 3.6226171049120697,
"tokens_seen": 1080426496
},
{
"epoch": 0.33,
"learning_rate": 0.0003397291875626881,
"loss": 2.7519,
"theoretical_loss": 3.6225964917833067,
"tokens_seen": 1080492032
},
{
"epoch": 0.33,
"learning_rate": 0.00033971915747241726,
"loss": 2.9322,
"theoretical_loss": 3.6225758802548214,
"tokens_seen": 1080557568
},
{
"epoch": 0.33,
"learning_rate": 0.00033970912738214644,
"loss": 2.6213,
"theoretical_loss": 3.622555270326393,
"tokens_seen": 1080623104
},
{
"epoch": 0.33,
"learning_rate": 0.0003396990972918756,
"loss": 2.7981,
"theoretical_loss": 3.6225346619977996,
"tokens_seen": 1080688640
},
{
"epoch": 0.33,
"learning_rate": 0.0003396890672016048,
"loss": 2.9475,
"theoretical_loss": 3.6225140552688204,
"tokens_seen": 1080754176
},
{
"epoch": 0.33,
"learning_rate": 0.00033967903711133404,
"loss": 2.758,
"theoretical_loss": 3.6224934501392343,
"tokens_seen": 1080819712
},
{
"epoch": 0.33,
"learning_rate": 0.00033966900702106316,
"loss": 2.8701,
"theoretical_loss": 3.62247284660882,
"tokens_seen": 1080885248
},
{
"epoch": 0.33,
"learning_rate": 0.0003396589769307924,
"loss": 2.7938,
"theoretical_loss": 3.6224522446773566,
"tokens_seen": 1080950784
},
{
"epoch": 0.33,
"learning_rate": 0.0003396489468405216,
"loss": 2.8561,
"theoretical_loss": 3.6224316443446227,
"tokens_seen": 1081016320
},
{
"epoch": 0.33,
"learning_rate": 0.00033963891675025076,
"loss": 2.818,
"theoretical_loss": 3.6224110456103977,
"tokens_seen": 1081081856
},
{
"epoch": 0.33,
"learning_rate": 0.00033962888665997994,
"loss": 2.9428,
"theoretical_loss": 3.622390448474461,
"tokens_seen": 1081147392
},
{
"epoch": 0.33,
"learning_rate": 0.0003396188565697091,
"loss": 2.9299,
"theoretical_loss": 3.6223698529365906,
"tokens_seen": 1081212928
},
{
"epoch": 0.33,
"learning_rate": 0.0003396088264794383,
"loss": 3.0431,
"theoretical_loss": 3.622349258996567,
"tokens_seen": 1081278464
},
{
"epoch": 0.33,
"objective/train/docs_used": 616786,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.226166248321533,
"objective/train/theoretical_loss": 3.6223286666541683,
"objective/train/tokens_used": 1101804000,
"theoretical_loss": 3.6223286666541683,
"tokens_seen": 1081344000
},
{
"epoch": 0.33,
"learning_rate": 0.00033959879638916754,
"loss": 3.0893,
"theoretical_loss": 3.6223286666541683,
"tokens_seen": 1081344000
},
{
"epoch": 0.33,
"learning_rate": 0.00033958876629889667,
"loss": 2.7527,
"theoretical_loss": 3.6223080759091744,
"tokens_seen": 1081409536
},
{
"epoch": 0.33,
"learning_rate": 0.0003395787362086259,
"loss": 2.816,
"theoretical_loss": 3.6222874867613646,
"tokens_seen": 1081475072
},
{
"epoch": 0.33,
"learning_rate": 0.00033956870611835503,
"loss": 3.0062,
"theoretical_loss": 3.6222668992105174,
"tokens_seen": 1081540608
},
{
"epoch": 0.33,
"learning_rate": 0.00033955867602808426,
"loss": 2.8417,
"theoretical_loss": 3.6222463132564133,
"tokens_seen": 1081606144
},
{
"epoch": 0.33,
"learning_rate": 0.00033954864593781345,
"loss": 2.9386,
"theoretical_loss": 3.6222257288988313,
"tokens_seen": 1081671680
},
{
"epoch": 0.33,
"learning_rate": 0.0003395386158475426,
"loss": 2.6028,
"theoretical_loss": 3.62220514613755,
"tokens_seen": 1081737216
},
{
"epoch": 0.33,
"learning_rate": 0.0003395285857572718,
"loss": 2.7195,
"theoretical_loss": 3.6221845649723505,
"tokens_seen": 1081802752
},
{
"epoch": 0.33,
"learning_rate": 0.00033951855566700104,
"loss": 2.7131,
"theoretical_loss": 3.6221639854030117,
"tokens_seen": 1081868288
},
{
"epoch": 0.33,
"learning_rate": 0.00033950852557673017,
"loss": 2.7788,
"theoretical_loss": 3.6221434074293124,
"tokens_seen": 1081933824
},
{
"epoch": 0.33,
"learning_rate": 0.0003394984954864594,
"loss": 2.7707,
"theoretical_loss": 3.6221228310510334,
"tokens_seen": 1081999360
},
{
"epoch": 0.33,
"learning_rate": 0.00033948846539618853,
"loss": 2.7362,
"theoretical_loss": 3.6221022562679535,
"tokens_seen": 1082064896
},
{
"epoch": 0.33,
"learning_rate": 0.00033947843530591777,
"loss": 2.904,
"theoretical_loss": 3.6220816830798532,
"tokens_seen": 1082130432
},
{
"epoch": 0.33,
"learning_rate": 0.00033946840521564695,
"loss": 2.8422,
"theoretical_loss": 3.6220611114865116,
"tokens_seen": 1082195968
},
{
"epoch": 0.33,
"learning_rate": 0.00033945837512537613,
"loss": 2.9708,
"theoretical_loss": 3.622040541487709,
"tokens_seen": 1082261504
},
{
"epoch": 0.33,
"learning_rate": 0.0003394483450351053,
"loss": 2.6929,
"theoretical_loss": 3.6220199730832245,
"tokens_seen": 1082327040
},
{
"epoch": 0.33,
"learning_rate": 0.0003394383149448345,
"loss": 2.9388,
"theoretical_loss": 3.621999406272839,
"tokens_seen": 1082392576
},
{
"epoch": 0.33,
"learning_rate": 0.0003394282848545637,
"loss": 2.8105,
"theoretical_loss": 3.6219788410563316,
"tokens_seen": 1082458112
},
{
"epoch": 0.33,
"learning_rate": 0.0003394182547642929,
"loss": 2.7379,
"theoretical_loss": 3.6219582774334835,
"tokens_seen": 1082523648
},
{
"epoch": 0.33,
"learning_rate": 0.00033940822467402204,
"loss": 2.9088,
"theoretical_loss": 3.6219377154040733,
"tokens_seen": 1082589184
},
{
"epoch": 0.33,
"learning_rate": 0.00033939819458375127,
"loss": 2.8761,
"theoretical_loss": 3.621917154967882,
"tokens_seen": 1082654720
},
{
"epoch": 0.33,
"learning_rate": 0.0003393881644934804,
"loss": 2.7074,
"theoretical_loss": 3.6218965961246887,
"tokens_seen": 1082720256
},
{
"epoch": 0.33,
"learning_rate": 0.00033937813440320963,
"loss": 2.6584,
"theoretical_loss": 3.621876038874275,
"tokens_seen": 1082785792
},
{
"epoch": 0.33,
"learning_rate": 0.0003393681043129388,
"loss": 2.8082,
"theoretical_loss": 3.6218554832164203,
"tokens_seen": 1082851328
},
{
"epoch": 0.33,
"learning_rate": 0.000339358074222668,
"loss": 2.7318,
"theoretical_loss": 3.621834929150905,
"tokens_seen": 1082916864
},
{
"epoch": 0.33,
"objective/train/docs_used": 617451,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.848808526992798,
"objective/train/theoretical_loss": 3.6218143766775093,
"objective/train/tokens_used": 1103442400,
"theoretical_loss": 3.6218143766775093,
"tokens_seen": 1082982400
},
{
"epoch": 0.33,
"learning_rate": 0.0003393480441323972,
"loss": 2.8078,
"theoretical_loss": 3.6218143766775093,
"tokens_seen": 1082982400
},
{
"epoch": 0.33,
"learning_rate": 0.0003393380140421264,
"loss": 2.8168,
"theoretical_loss": 3.621793825796013,
"tokens_seen": 1083047936
},
{
"epoch": 0.33,
"learning_rate": 0.0003393279839518556,
"loss": 2.7581,
"theoretical_loss": 3.6217732765061976,
"tokens_seen": 1083113472
},
{
"epoch": 0.33,
"learning_rate": 0.0003393179538615848,
"loss": 2.6706,
"theoretical_loss": 3.621752728807843,
"tokens_seen": 1083179008
},
{
"epoch": 0.33,
"learning_rate": 0.00033930792377131396,
"loss": 3.0084,
"theoretical_loss": 3.62173218270073,
"tokens_seen": 1083244544
},
{
"epoch": 0.33,
"learning_rate": 0.00033929789368104314,
"loss": 2.7497,
"theoretical_loss": 3.621711638184638,
"tokens_seen": 1083310080
},
{
"epoch": 0.33,
"learning_rate": 0.00033928786359077237,
"loss": 2.8653,
"theoretical_loss": 3.6216910952593486,
"tokens_seen": 1083375616
},
{
"epoch": 0.33,
"learning_rate": 0.0003392778335005015,
"loss": 2.7895,
"theoretical_loss": 3.621670553924642,
"tokens_seen": 1083441152
},
{
"epoch": 0.33,
"learning_rate": 0.00033926780341023073,
"loss": 3.0124,
"theoretical_loss": 3.6216500141802994,
"tokens_seen": 1083506688
},
{
"epoch": 0.33,
"learning_rate": 0.00033925777331995986,
"loss": 2.8966,
"theoretical_loss": 3.6216294760261007,
"tokens_seen": 1083572224
},
{
"epoch": 0.33,
"learning_rate": 0.0003392477432296891,
"loss": 2.8246,
"theoretical_loss": 3.621608939461827,
"tokens_seen": 1083637760
},
{
"epoch": 0.33,
"learning_rate": 0.0003392377131394183,
"loss": 3.065,
"theoretical_loss": 3.6215884044872593,
"tokens_seen": 1083703296
},
{
"epoch": 0.33,
"learning_rate": 0.00033922768304914746,
"loss": 2.9036,
"theoretical_loss": 3.621567871102178,
"tokens_seen": 1083768832
},
{
"epoch": 0.33,
"learning_rate": 0.00033921765295887664,
"loss": 2.8076,
"theoretical_loss": 3.6215473393063644,
"tokens_seen": 1083834368
},
{
"epoch": 0.33,
"learning_rate": 0.0003392076228686058,
"loss": 2.6857,
"theoretical_loss": 3.621526809099599,
"tokens_seen": 1083899904
},
{
"epoch": 0.33,
"learning_rate": 0.000339197592778335,
"loss": 2.4906,
"theoretical_loss": 3.621506280481663,
"tokens_seen": 1083965440
},
{
"epoch": 0.33,
"learning_rate": 0.00033918756268806424,
"loss": 2.7165,
"theoretical_loss": 3.621485753452337,
"tokens_seen": 1084030976
},
{
"epoch": 0.33,
"learning_rate": 0.00033917753259779336,
"loss": 2.8244,
"theoretical_loss": 3.6214652280114024,
"tokens_seen": 1084096512
},
{
"epoch": 0.33,
"learning_rate": 0.0003391675025075226,
"loss": 2.973,
"theoretical_loss": 3.62144470415864,
"tokens_seen": 1084162048
},
{
"epoch": 0.33,
"learning_rate": 0.0003391574724172518,
"loss": 3.0287,
"theoretical_loss": 3.6214241818938318,
"tokens_seen": 1084227584
},
{
"epoch": 0.33,
"learning_rate": 0.00033914744232698096,
"loss": 2.6882,
"theoretical_loss": 3.621403661216758,
"tokens_seen": 1084293120
},
{
"epoch": 0.33,
"learning_rate": 0.00033913741223671014,
"loss": 3.0493,
"theoretical_loss": 3.6213831421272005,
"tokens_seen": 1084358656
},
{
"epoch": 0.33,
"learning_rate": 0.0003391273821464393,
"loss": 2.9557,
"theoretical_loss": 3.6213626246249393,
"tokens_seen": 1084424192
},
{
"epoch": 0.33,
"learning_rate": 0.0003391173520561685,
"loss": 2.7769,
"theoretical_loss": 3.621342108709757,
"tokens_seen": 1084489728
},
{
"epoch": 0.33,
"learning_rate": 0.00033910732196589774,
"loss": 2.8582,
"theoretical_loss": 3.6213215943814347,
"tokens_seen": 1084555264
},
{
"epoch": 0.33,
"objective/train/docs_used": 618647,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2976839542388916,
"objective/train/theoretical_loss": 3.621301081639753,
"objective/train/tokens_used": 1105080800,
"theoretical_loss": 3.621301081639753,
"tokens_seen": 1084620800
},
{
"epoch": 0.33,
"learning_rate": 0.00033909729187562687,
"loss": 2.8717,
"theoretical_loss": 3.621301081639753,
"tokens_seen": 1084620800
},
{
"epoch": 0.33,
"learning_rate": 0.0003390872617853561,
"loss": 2.5493,
"theoretical_loss": 3.621280570484494,
"tokens_seen": 1084686336
},
{
"epoch": 0.33,
"learning_rate": 0.00033907723169508523,
"loss": 2.6963,
"theoretical_loss": 3.6212600609154393,
"tokens_seen": 1084751872
},
{
"epoch": 0.33,
"learning_rate": 0.00033906720160481446,
"loss": 2.7886,
"theoretical_loss": 3.62123955293237,
"tokens_seen": 1084817408
},
{
"epoch": 0.33,
"learning_rate": 0.00033905717151454365,
"loss": 2.6911,
"theoretical_loss": 3.621219046535068,
"tokens_seen": 1084882944
},
{
"epoch": 0.33,
"learning_rate": 0.00033904714142427283,
"loss": 2.6735,
"theoretical_loss": 3.6211985417233157,
"tokens_seen": 1084948480
},
{
"epoch": 0.33,
"learning_rate": 0.000339037111334002,
"loss": 2.7847,
"theoretical_loss": 3.6211780384968923,
"tokens_seen": 1085014016
},
{
"epoch": 0.33,
"learning_rate": 0.00033902708124373124,
"loss": 2.7761,
"theoretical_loss": 3.6211575368555815,
"tokens_seen": 1085079552
},
{
"epoch": 0.33,
"learning_rate": 0.00033901705115346037,
"loss": 2.7384,
"theoretical_loss": 3.6211370367991647,
"tokens_seen": 1085145088
},
{
"epoch": 0.33,
"learning_rate": 0.0003390070210631896,
"loss": 2.5371,
"theoretical_loss": 3.6211165383274233,
"tokens_seen": 1085210624
},
{
"epoch": 0.33,
"learning_rate": 0.00033899699097291873,
"loss": 2.8146,
"theoretical_loss": 3.621096041440139,
"tokens_seen": 1085276160
},
{
"epoch": 0.33,
"learning_rate": 0.00033898696088264797,
"loss": 2.7703,
"theoretical_loss": 3.621075546137094,
"tokens_seen": 1085341696
},
{
"epoch": 0.33,
"learning_rate": 0.00033897693079237715,
"loss": 3.0453,
"theoretical_loss": 3.621055052418071,
"tokens_seen": 1085407232
},
{
"epoch": 0.33,
"learning_rate": 0.00033896690070210633,
"loss": 2.7395,
"theoretical_loss": 3.62103456028285,
"tokens_seen": 1085472768
},
{
"epoch": 0.33,
"learning_rate": 0.0003389568706118355,
"loss": 2.5529,
"theoretical_loss": 3.6210140697312143,
"tokens_seen": 1085538304
},
{
"epoch": 0.33,
"learning_rate": 0.0003389468405215647,
"loss": 2.7368,
"theoretical_loss": 3.6209935807629465,
"tokens_seen": 1085603840
},
{
"epoch": 0.33,
"learning_rate": 0.0003389368104312939,
"loss": 3.0337,
"theoretical_loss": 3.620973093377827,
"tokens_seen": 1085669376
},
{
"epoch": 0.33,
"learning_rate": 0.0003389267803410231,
"loss": 2.6518,
"theoretical_loss": 3.620952607575639,
"tokens_seen": 1085734912
},
{
"epoch": 0.33,
"learning_rate": 0.00033891675025075224,
"loss": 2.7324,
"theoretical_loss": 3.6209321233561647,
"tokens_seen": 1085800448
},
{
"epoch": 0.33,
"learning_rate": 0.00033890672016048147,
"loss": 2.7577,
"theoretical_loss": 3.6209116407191857,
"tokens_seen": 1085865984
},
{
"epoch": 0.33,
"learning_rate": 0.0003388966900702106,
"loss": 2.7245,
"theoretical_loss": 3.620891159664485,
"tokens_seen": 1085931520
},
{
"epoch": 0.33,
"learning_rate": 0.00033888665997993983,
"loss": 3.1003,
"theoretical_loss": 3.620870680191844,
"tokens_seen": 1085997056
},
{
"epoch": 0.33,
"learning_rate": 0.000338876629889669,
"loss": 2.7088,
"theoretical_loss": 3.620850202301046,
"tokens_seen": 1086062592
},
{
"epoch": 0.33,
"learning_rate": 0.0003388665997993982,
"loss": 2.8462,
"theoretical_loss": 3.620829725991873,
"tokens_seen": 1086128128
},
{
"epoch": 0.33,
"learning_rate": 0.0003388565697091274,
"loss": 2.7234,
"theoretical_loss": 3.620809251264107,
"tokens_seen": 1086193664
},
{
"epoch": 0.33,
"objective/train/docs_used": 619458,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.794792890548706,
"objective/train/theoretical_loss": 3.6207887781175305,
"objective/train/tokens_used": 1106719200,
"theoretical_loss": 3.6207887781175305,
"tokens_seen": 1086259200
},
{
"epoch": 0.33,
"learning_rate": 0.0003388465396188566,
"loss": 2.728,
"theoretical_loss": 3.6207887781175305,
"tokens_seen": 1086259200
},
{
"epoch": 0.33,
"learning_rate": 0.00033883650952858574,
"loss": 2.7347,
"theoretical_loss": 3.620768306551927,
"tokens_seen": 1086324736
},
{
"epoch": 0.33,
"learning_rate": 0.000338826479438315,
"loss": 2.9595,
"theoretical_loss": 3.620747836567078,
"tokens_seen": 1086390272
},
{
"epoch": 0.33,
"learning_rate": 0.0003388164493480441,
"loss": 2.6803,
"theoretical_loss": 3.6207273681627665,
"tokens_seen": 1086455808
},
{
"epoch": 0.33,
"learning_rate": 0.00033880641925777334,
"loss": 2.6531,
"theoretical_loss": 3.620706901338775,
"tokens_seen": 1086521344
},
{
"epoch": 0.33,
"learning_rate": 0.0003387963891675025,
"loss": 2.8872,
"theoretical_loss": 3.6206864360948865,
"tokens_seen": 1086586880
},
{
"epoch": 0.33,
"learning_rate": 0.0003387863590772317,
"loss": 2.7253,
"theoretical_loss": 3.620665972430883,
"tokens_seen": 1086652416
},
{
"epoch": 0.33,
"learning_rate": 0.0003387763289869609,
"loss": 2.7798,
"theoretical_loss": 3.620645510346548,
"tokens_seen": 1086717952
},
{
"epoch": 0.33,
"learning_rate": 0.00033876629889669006,
"loss": 2.7156,
"theoretical_loss": 3.620625049841664,
"tokens_seen": 1086783488
},
{
"epoch": 0.33,
"learning_rate": 0.00033875626880641924,
"loss": 2.7868,
"theoretical_loss": 3.620604590916014,
"tokens_seen": 1086849024
},
{
"epoch": 0.33,
"learning_rate": 0.0003387462387161485,
"loss": 2.825,
"theoretical_loss": 3.6205841335693805,
"tokens_seen": 1086914560
},
{
"epoch": 0.33,
"learning_rate": 0.0003387362086258776,
"loss": 2.8742,
"theoretical_loss": 3.620563677801547,
"tokens_seen": 1086980096
},
{
"epoch": 0.33,
"learning_rate": 0.00033872617853560684,
"loss": 2.7112,
"theoretical_loss": 3.6205432236122963,
"tokens_seen": 1087045632
},
{
"epoch": 0.33,
"learning_rate": 0.00033871614844533597,
"loss": 2.7712,
"theoretical_loss": 3.620522771001411,
"tokens_seen": 1087111168
},
{
"epoch": 0.33,
"learning_rate": 0.0003387061183550652,
"loss": 2.9341,
"theoretical_loss": 3.6205023199686743,
"tokens_seen": 1087176704
},
{
"epoch": 0.33,
"learning_rate": 0.0003386960882647944,
"loss": 2.8971,
"theoretical_loss": 3.6204818705138697,
"tokens_seen": 1087242240
},
{
"epoch": 0.33,
"learning_rate": 0.00033868605817452356,
"loss": 2.9336,
"theoretical_loss": 3.6204614226367804,
"tokens_seen": 1087307776
},
{
"epoch": 0.33,
"learning_rate": 0.00033867602808425275,
"loss": 2.9889,
"theoretical_loss": 3.620440976337189,
"tokens_seen": 1087373312
},
{
"epoch": 0.33,
"learning_rate": 0.000338665997993982,
"loss": 2.6655,
"theoretical_loss": 3.6204205316148794,
"tokens_seen": 1087438848
},
{
"epoch": 0.33,
"learning_rate": 0.0003386559679037111,
"loss": 2.7717,
"theoretical_loss": 3.6204000884696343,
"tokens_seen": 1087504384
},
{
"epoch": 0.33,
"learning_rate": 0.00033864593781344034,
"loss": 2.6457,
"theoretical_loss": 3.6203796469012373,
"tokens_seen": 1087569920
},
{
"epoch": 0.33,
"learning_rate": 0.00033863590772316947,
"loss": 2.8224,
"theoretical_loss": 3.6203592069094714,
"tokens_seen": 1087635456
},
{
"epoch": 0.33,
"learning_rate": 0.0003386258776328987,
"loss": 2.9446,
"theoretical_loss": 3.620338768494121,
"tokens_seen": 1087700992
},
{
"epoch": 0.33,
"learning_rate": 0.0003386158475426279,
"loss": 2.891,
"theoretical_loss": 3.6203183316549685,
"tokens_seen": 1087766528
},
{
"epoch": 0.33,
"learning_rate": 0.00033860581745235707,
"loss": 2.5772,
"theoretical_loss": 3.620297896391798,
"tokens_seen": 1087832064
},
{
"epoch": 0.33,
"objective/train/docs_used": 620460,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9795472621917725,
"objective/train/theoretical_loss": 3.6202774627043923,
"objective/train/tokens_used": 1108357600,
"theoretical_loss": 3.6202774627043923,
"tokens_seen": 1087897600
},
{
"epoch": 0.33,
"learning_rate": 0.00033859578736208625,
"loss": 2.9624,
"theoretical_loss": 3.6202774627043923,
"tokens_seen": 1087897600
},
{
"epoch": 0.33,
"learning_rate": 0.00033858575727181543,
"loss": 2.8154,
"theoretical_loss": 3.620257030592536,
"tokens_seen": 1087963136
},
{
"epoch": 0.33,
"learning_rate": 0.00033857572718154467,
"loss": 2.6792,
"theoretical_loss": 3.6202366000560118,
"tokens_seen": 1088028672
},
{
"epoch": 0.33,
"learning_rate": 0.00033856569709127385,
"loss": 2.6527,
"theoretical_loss": 3.6202161710946044,
"tokens_seen": 1088094208
},
{
"epoch": 0.33,
"learning_rate": 0.00033855566700100303,
"loss": 2.9154,
"theoretical_loss": 3.620195743708096,
"tokens_seen": 1088159744
},
{
"epoch": 0.33,
"learning_rate": 0.0003385456369107322,
"loss": 2.804,
"theoretical_loss": 3.6201753178962717,
"tokens_seen": 1088225280
},
{
"epoch": 0.33,
"learning_rate": 0.00033853560682046144,
"loss": 2.8334,
"theoretical_loss": 3.620154893658915,
"tokens_seen": 1088290816
},
{
"epoch": 0.33,
"learning_rate": 0.00033852557673019057,
"loss": 2.7449,
"theoretical_loss": 3.620134470995809,
"tokens_seen": 1088356352
},
{
"epoch": 0.33,
"learning_rate": 0.0003385155466399198,
"loss": 2.7031,
"theoretical_loss": 3.620114049906739,
"tokens_seen": 1088421888
},
{
"epoch": 0.33,
"learning_rate": 0.00033850551654964893,
"loss": 3.0577,
"theoretical_loss": 3.6200936303914872,
"tokens_seen": 1088487424
},
{
"epoch": 0.33,
"learning_rate": 0.00033849548645937817,
"loss": 2.4825,
"theoretical_loss": 3.6200732124498387,
"tokens_seen": 1088552960
},
{
"epoch": 0.33,
"learning_rate": 0.00033848545636910735,
"loss": 2.8966,
"theoretical_loss": 3.6200527960815774,
"tokens_seen": 1088618496
},
{
"epoch": 0.33,
"learning_rate": 0.00033847542627883653,
"loss": 2.6848,
"theoretical_loss": 3.620032381286487,
"tokens_seen": 1088684032
},
{
"epoch": 0.33,
"learning_rate": 0.0003384653961885657,
"loss": 2.9237,
"theoretical_loss": 3.6200119680643517,
"tokens_seen": 1088749568
},
{
"epoch": 0.33,
"learning_rate": 0.0003384553660982949,
"loss": 2.6756,
"theoretical_loss": 3.6199915564149556,
"tokens_seen": 1088815104
},
{
"epoch": 0.33,
"learning_rate": 0.0003384453360080241,
"loss": 2.5468,
"theoretical_loss": 3.6199711463380835,
"tokens_seen": 1088880640
},
{
"epoch": 0.33,
"learning_rate": 0.0003384353059177533,
"loss": 2.8179,
"theoretical_loss": 3.6199507378335185,
"tokens_seen": 1088946176
},
{
"epoch": 0.33,
"learning_rate": 0.00033842527582748244,
"loss": 2.8959,
"theoretical_loss": 3.6199303309010458,
"tokens_seen": 1089011712
},
{
"epoch": 0.33,
"learning_rate": 0.00033841524573721167,
"loss": 2.5635,
"theoretical_loss": 3.6199099255404494,
"tokens_seen": 1089077248
},
{
"epoch": 0.33,
"learning_rate": 0.0003384052156469408,
"loss": 2.8662,
"theoretical_loss": 3.6198895217515132,
"tokens_seen": 1089142784
},
{
"epoch": 0.33,
"learning_rate": 0.00033839518555667003,
"loss": 2.8684,
"theoretical_loss": 3.6198691195340222,
"tokens_seen": 1089208320
},
{
"epoch": 0.33,
"learning_rate": 0.0003383851554663992,
"loss": 2.7928,
"theoretical_loss": 3.6198487188877606,
"tokens_seen": 1089273856
},
{
"epoch": 0.33,
"learning_rate": 0.0003383751253761284,
"loss": 2.6275,
"theoretical_loss": 3.619828319812513,
"tokens_seen": 1089339392
},
{
"epoch": 0.33,
"learning_rate": 0.0003383650952858576,
"loss": 2.8432,
"theoretical_loss": 3.6198079223080635,
"tokens_seen": 1089404928
},
{
"epoch": 0.33,
"learning_rate": 0.0003383550651955868,
"loss": 2.9882,
"theoretical_loss": 3.619787526374197,
"tokens_seen": 1089470464
},
{
"epoch": 0.33,
"objective/train/docs_used": 620976,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6567726135253906,
"objective/train/theoretical_loss": 3.6197671320106983,
"objective/train/tokens_used": 1109996000,
"theoretical_loss": 3.6197671320106983,
"tokens_seen": 1089536000
},
{
"epoch": 0.33,
"learning_rate": 0.00033834503510531594,
"loss": 2.7722,
"theoretical_loss": 3.6197671320106983,
"tokens_seen": 1089536000
},
{
"epoch": 0.33,
"learning_rate": 0.0003383350050150452,
"loss": 2.8941,
"theoretical_loss": 3.6197467392173515,
"tokens_seen": 1089601536
},
{
"epoch": 0.33,
"learning_rate": 0.0003383249749247743,
"loss": 2.7155,
"theoretical_loss": 3.619726347993942,
"tokens_seen": 1089667072
},
{
"epoch": 0.33,
"learning_rate": 0.00033831494483450354,
"loss": 2.7363,
"theoretical_loss": 3.619705958340254,
"tokens_seen": 1089732608
},
{
"epoch": 0.33,
"learning_rate": 0.0003383049147442327,
"loss": 2.6805,
"theoretical_loss": 3.619685570256072,
"tokens_seen": 1089798144
},
{
"epoch": 0.33,
"learning_rate": 0.0003382948846539619,
"loss": 2.8967,
"theoretical_loss": 3.6196651837411817,
"tokens_seen": 1089863680
},
{
"epoch": 0.33,
"learning_rate": 0.0003382848545636911,
"loss": 3.0483,
"theoretical_loss": 3.6196447987953677,
"tokens_seen": 1089929216
},
{
"epoch": 0.33,
"learning_rate": 0.00033827482447342026,
"loss": 3.1468,
"theoretical_loss": 3.619624415418414,
"tokens_seen": 1089994752
},
{
"epoch": 0.33,
"learning_rate": 0.00033826479438314944,
"loss": 3.0014,
"theoretical_loss": 3.6196040336101065,
"tokens_seen": 1090060288
},
{
"epoch": 0.33,
"learning_rate": 0.0003382547642928787,
"loss": 2.7614,
"theoretical_loss": 3.6195836533702295,
"tokens_seen": 1090125824
},
{
"epoch": 0.33,
"learning_rate": 0.0003382447342026078,
"loss": 2.7542,
"theoretical_loss": 3.619563274698569,
"tokens_seen": 1090191360
},
{
"epoch": 0.33,
"learning_rate": 0.00033823470411233704,
"loss": 2.7535,
"theoretical_loss": 3.619542897594909,
"tokens_seen": 1090256896
},
{
"epoch": 0.33,
"learning_rate": 0.00033822467402206617,
"loss": 2.8806,
"theoretical_loss": 3.619522522059035,
"tokens_seen": 1090322432
},
{
"epoch": 0.33,
"learning_rate": 0.0003382146439317954,
"loss": 2.7364,
"theoretical_loss": 3.619502148090733,
"tokens_seen": 1090387968
},
{
"epoch": 0.33,
"learning_rate": 0.0003382046138415246,
"loss": 2.8314,
"theoretical_loss": 3.6194817756897866,
"tokens_seen": 1090453504
},
{
"epoch": 0.33,
"learning_rate": 0.00033819458375125376,
"loss": 2.9392,
"theoretical_loss": 3.6194614048559828,
"tokens_seen": 1090519040
},
{
"epoch": 0.33,
"learning_rate": 0.00033818455366098295,
"loss": 2.9713,
"theoretical_loss": 3.619441035589105,
"tokens_seen": 1090584576
},
{
"epoch": 0.33,
"learning_rate": 0.0003381745235707122,
"loss": 2.8976,
"theoretical_loss": 3.6194206678889396,
"tokens_seen": 1090650112
},
{
"epoch": 0.33,
"learning_rate": 0.0003381644934804413,
"loss": 2.6781,
"theoretical_loss": 3.619400301755272,
"tokens_seen": 1090715648
},
{
"epoch": 0.33,
"learning_rate": 0.00033815446339017054,
"loss": 2.7229,
"theoretical_loss": 3.6193799371878876,
"tokens_seen": 1090781184
},
{
"epoch": 0.33,
"learning_rate": 0.00033814443329989967,
"loss": 2.7748,
"theoretical_loss": 3.619359574186572,
"tokens_seen": 1090846720
},
{
"epoch": 0.33,
"learning_rate": 0.0003381344032096289,
"loss": 2.7824,
"theoretical_loss": 3.6193392127511097,
"tokens_seen": 1090912256
},
{
"epoch": 0.33,
"learning_rate": 0.0003381243731193581,
"loss": 2.7185,
"theoretical_loss": 3.619318852881287,
"tokens_seen": 1090977792
},
{
"epoch": 0.33,
"learning_rate": 0.00033811434302908727,
"loss": 2.6123,
"theoretical_loss": 3.61929849457689,
"tokens_seen": 1091043328
},
{
"epoch": 0.33,
"learning_rate": 0.00033810431293881645,
"loss": 2.9183,
"theoretical_loss": 3.619278137837703,
"tokens_seen": 1091108864
},
{
"epoch": 0.33,
"objective/train/docs_used": 621707,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7134923934936523,
"objective/train/theoretical_loss": 3.619257782663513,
"objective/train/tokens_used": 1111634400,
"theoretical_loss": 3.619257782663513,
"tokens_seen": 1091174400
},
{
"epoch": 0.33,
"learning_rate": 0.00033809428284854563,
"loss": 2.858,
"theoretical_loss": 3.619257782663513,
"tokens_seen": 1091174400
},
{
"epoch": 0.33,
"learning_rate": 0.0003380842527582748,
"loss": 2.5795,
"theoretical_loss": 3.6192374290541047,
"tokens_seen": 1091239936
},
{
"epoch": 0.33,
"learning_rate": 0.00033807422266800405,
"loss": 2.7247,
"theoretical_loss": 3.619217077009264,
"tokens_seen": 1091305472
},
{
"epoch": 0.33,
"learning_rate": 0.0003380641925777332,
"loss": 2.9602,
"theoretical_loss": 3.619196726528777,
"tokens_seen": 1091371008
},
{
"epoch": 0.33,
"learning_rate": 0.0003380541624874624,
"loss": 2.8758,
"theoretical_loss": 3.6191763776124297,
"tokens_seen": 1091436544
},
{
"epoch": 0.33,
"learning_rate": 0.00033804413239719154,
"loss": 2.8556,
"theoretical_loss": 3.619156030260008,
"tokens_seen": 1091502080
},
{
"epoch": 0.33,
"learning_rate": 0.00033803410230692077,
"loss": 2.901,
"theoretical_loss": 3.619135684471297,
"tokens_seen": 1091567616
},
{
"epoch": 0.33,
"learning_rate": 0.00033802407221664995,
"loss": 2.9111,
"theoretical_loss": 3.6191153402460836,
"tokens_seen": 1091633152
},
{
"epoch": 0.33,
"learning_rate": 0.00033801404212637913,
"loss": 2.8593,
"theoretical_loss": 3.6190949975841527,
"tokens_seen": 1091698688
},
{
"epoch": 0.33,
"learning_rate": 0.0003380040120361083,
"loss": 2.7776,
"theoretical_loss": 3.6190746564852914,
"tokens_seen": 1091764224
},
{
"epoch": 0.33,
"learning_rate": 0.00033799398194583755,
"loss": 2.7304,
"theoretical_loss": 3.6190543169492857,
"tokens_seen": 1091829760
},
{
"epoch": 0.33,
"learning_rate": 0.0003379839518555667,
"loss": 2.9581,
"theoretical_loss": 3.6190339789759207,
"tokens_seen": 1091895296
},
{
"epoch": 0.33,
"learning_rate": 0.0003379739217652959,
"loss": 2.7959,
"theoretical_loss": 3.6190136425649837,
"tokens_seen": 1091960832
},
{
"epoch": 0.33,
"learning_rate": 0.00033796389167502504,
"loss": 2.8648,
"theoretical_loss": 3.6189933077162606,
"tokens_seen": 1092026368
},
{
"epoch": 0.33,
"learning_rate": 0.0003379538615847543,
"loss": 2.8459,
"theoretical_loss": 3.6189729744295374,
"tokens_seen": 1092091904
},
{
"epoch": 0.33,
"learning_rate": 0.00033794383149448346,
"loss": 2.9552,
"theoretical_loss": 3.618952642704601,
"tokens_seen": 1092157440
},
{
"epoch": 0.33,
"learning_rate": 0.00033793380140421264,
"loss": 2.8336,
"theoretical_loss": 3.6189323125412365,
"tokens_seen": 1092222976
},
{
"epoch": 0.33,
"learning_rate": 0.0003379237713139418,
"loss": 2.9111,
"theoretical_loss": 3.618911983939231,
"tokens_seen": 1092288512
},
{
"epoch": 0.33,
"learning_rate": 0.000337913741223671,
"loss": 2.7222,
"theoretical_loss": 3.618891656898372,
"tokens_seen": 1092354048
},
{
"epoch": 0.33,
"learning_rate": 0.0003379037111334002,
"loss": 3.0484,
"theoretical_loss": 3.618871331418444,
"tokens_seen": 1092419584
},
{
"epoch": 0.33,
"learning_rate": 0.0003378936810431294,
"loss": 2.8383,
"theoretical_loss": 3.6188510074992344,
"tokens_seen": 1092485120
},
{
"epoch": 0.33,
"learning_rate": 0.00033788365095285854,
"loss": 2.8092,
"theoretical_loss": 3.61883068514053,
"tokens_seen": 1092550656
},
{
"epoch": 0.33,
"learning_rate": 0.0003378736208625878,
"loss": 2.782,
"theoretical_loss": 3.6188103643421172,
"tokens_seen": 1092616192
},
{
"epoch": 0.33,
"learning_rate": 0.0003378635907723169,
"loss": 2.6424,
"theoretical_loss": 3.6187900451037827,
"tokens_seen": 1092681728
},
{
"epoch": 0.33,
"learning_rate": 0.00033785356068204614,
"loss": 2.9171,
"theoretical_loss": 3.618769727425313,
"tokens_seen": 1092747264
},
{
"epoch": 0.33,
"objective/train/docs_used": 622698,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.208833932876587,
"objective/train/theoretical_loss": 3.618749411306495,
"objective/train/tokens_used": 1113272800,
"theoretical_loss": 3.618749411306495,
"tokens_seen": 1092812800
},
{
"epoch": 0.33,
"learning_rate": 0.0003378435305917753,
"loss": 2.9182,
"theoretical_loss": 3.618749411306495,
"tokens_seen": 1092812800
},
{
"epoch": 0.33,
"learning_rate": 0.0003378335005015045,
"loss": 2.7654,
"theoretical_loss": 3.618729096747115,
"tokens_seen": 1092878336
},
{
"epoch": 0.33,
"learning_rate": 0.00033782347041123374,
"loss": 2.8012,
"theoretical_loss": 3.6187087837469605,
"tokens_seen": 1092943872
},
{
"epoch": 0.33,
"learning_rate": 0.0003378134403209629,
"loss": 2.6379,
"theoretical_loss": 3.6186884723058172,
"tokens_seen": 1093009408
},
{
"epoch": 0.33,
"learning_rate": 0.0003378034102306921,
"loss": 2.7481,
"theoretical_loss": 3.6186681624234733,
"tokens_seen": 1093074944
},
{
"epoch": 0.33,
"learning_rate": 0.0003377933801404213,
"loss": 2.8545,
"theoretical_loss": 3.6186478540997156,
"tokens_seen": 1093140480
},
{
"epoch": 0.33,
"learning_rate": 0.00033778335005015046,
"loss": 2.8514,
"theoretical_loss": 3.61862754733433,
"tokens_seen": 1093206016
},
{
"epoch": 0.33,
"learning_rate": 0.00033777331995987964,
"loss": 2.7944,
"theoretical_loss": 3.618607242127104,
"tokens_seen": 1093271552
},
{
"epoch": 0.33,
"learning_rate": 0.0003377632898696089,
"loss": 2.9109,
"theoretical_loss": 3.6185869384778253,
"tokens_seen": 1093337088
},
{
"epoch": 0.33,
"learning_rate": 0.000337753259779338,
"loss": 2.8934,
"theoretical_loss": 3.61856663638628,
"tokens_seen": 1093402624
},
{
"epoch": 0.33,
"learning_rate": 0.00033774322968906724,
"loss": 2.8559,
"theoretical_loss": 3.6185463358522565,
"tokens_seen": 1093468160
},
{
"epoch": 0.33,
"learning_rate": 0.00033773319959879637,
"loss": 2.8513,
"theoretical_loss": 3.61852603687554,
"tokens_seen": 1093533696
},
{
"epoch": 0.33,
"learning_rate": 0.0003377231695085256,
"loss": 3.069,
"theoretical_loss": 3.61850573945592,
"tokens_seen": 1093599232
},
{
"epoch": 0.33,
"learning_rate": 0.0003377131394182548,
"loss": 3.034,
"theoretical_loss": 3.618485443593182,
"tokens_seen": 1093664768
},
{
"epoch": 0.33,
"learning_rate": 0.00033770310932798396,
"loss": 2.882,
"theoretical_loss": 3.618465149287114,
"tokens_seen": 1093730304
},
{
"epoch": 0.33,
"learning_rate": 0.00033769307923771315,
"loss": 2.6874,
"theoretical_loss": 3.6184448565375034,
"tokens_seen": 1093795840
},
{
"epoch": 0.33,
"learning_rate": 0.0003376830491474424,
"loss": 2.6402,
"theoretical_loss": 3.6184245653441374,
"tokens_seen": 1093861376
},
{
"epoch": 0.33,
"learning_rate": 0.0003376730190571715,
"loss": 2.8453,
"theoretical_loss": 3.6184042757068036,
"tokens_seen": 1093926912
},
{
"epoch": 0.33,
"learning_rate": 0.00033766298896690074,
"loss": 2.8933,
"theoretical_loss": 3.6183839876252892,
"tokens_seen": 1093992448
},
{
"epoch": 0.33,
"learning_rate": 0.00033765295887662987,
"loss": 2.7304,
"theoretical_loss": 3.618363701099382,
"tokens_seen": 1094057984
},
{
"epoch": 0.33,
"learning_rate": 0.0003376429287863591,
"loss": 2.6913,
"theoretical_loss": 3.6183434161288694,
"tokens_seen": 1094123520
},
{
"epoch": 0.33,
"learning_rate": 0.0003376328986960883,
"loss": 2.8347,
"theoretical_loss": 3.618323132713539,
"tokens_seen": 1094189056
},
{
"epoch": 0.33,
"learning_rate": 0.00033762286860581747,
"loss": 2.7156,
"theoretical_loss": 3.6183028508531785,
"tokens_seen": 1094254592
},
{
"epoch": 0.33,
"learning_rate": 0.00033761283851554665,
"loss": 2.6125,
"theoretical_loss": 3.618282570547575,
"tokens_seen": 1094320128
},
{
"epoch": 0.33,
"learning_rate": 0.00033760280842527583,
"loss": 2.8333,
"theoretical_loss": 3.6182622917965173,
"tokens_seen": 1094385664
},
{
"epoch": 0.33,
"objective/train/docs_used": 623416,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.117913246154785,
"objective/train/theoretical_loss": 3.618242014599793,
"objective/train/tokens_used": 1114911200,
"theoretical_loss": 3.618242014599793,
"tokens_seen": 1094451200
},
{
"epoch": 0.33,
"learning_rate": 0.000337592778335005,
"loss": 2.9244,
"theoretical_loss": 3.618242014599793,
"tokens_seen": 1094451200
},
{
"epoch": 0.33,
"learning_rate": 0.00033758274824473425,
"loss": 2.5804,
"theoretical_loss": 3.618221738957189,
"tokens_seen": 1094516736
},
{
"epoch": 0.33,
"learning_rate": 0.0003375727181544634,
"loss": 2.8608,
"theoretical_loss": 3.618201464868493,
"tokens_seen": 1094582272
},
{
"epoch": 0.33,
"learning_rate": 0.0003375626880641926,
"loss": 2.7937,
"theoretical_loss": 3.6181811923334943,
"tokens_seen": 1094647808
},
{
"epoch": 0.33,
"learning_rate": 0.00033755265797392174,
"loss": 2.7945,
"theoretical_loss": 3.61816092135198,
"tokens_seen": 1094713344
},
{
"epoch": 0.33,
"learning_rate": 0.00033754262788365097,
"loss": 2.8165,
"theoretical_loss": 3.6181406519237376,
"tokens_seen": 1094778880
},
{
"epoch": 0.33,
"learning_rate": 0.00033753259779338015,
"loss": 2.5732,
"theoretical_loss": 3.618120384048556,
"tokens_seen": 1094844416
},
{
"epoch": 0.33,
"learning_rate": 0.00033752256770310933,
"loss": 2.7943,
"theoretical_loss": 3.618100117726223,
"tokens_seen": 1094909952
},
{
"epoch": 0.33,
"learning_rate": 0.0003375125376128385,
"loss": 3.0198,
"theoretical_loss": 3.6180798529565266,
"tokens_seen": 1094975488
},
{
"epoch": 0.33,
"learning_rate": 0.00033750250752256775,
"loss": 2.7794,
"theoretical_loss": 3.6180595897392545,
"tokens_seen": 1095041024
},
{
"epoch": 0.33,
"learning_rate": 0.0003374924774322969,
"loss": 2.6935,
"theoretical_loss": 3.6180393280741954,
"tokens_seen": 1095106560
},
{
"epoch": 0.33,
"learning_rate": 0.0003374824473420261,
"loss": 2.8761,
"theoretical_loss": 3.6180190679611375,
"tokens_seen": 1095172096
},
{
"epoch": 0.33,
"learning_rate": 0.00033747241725175524,
"loss": 2.741,
"theoretical_loss": 3.617998809399869,
"tokens_seen": 1095237632
},
{
"epoch": 0.33,
"learning_rate": 0.0003374623871614845,
"loss": 2.732,
"theoretical_loss": 3.6179785523901775,
"tokens_seen": 1095303168
},
{
"epoch": 0.33,
"learning_rate": 0.00033745235707121366,
"loss": 2.6805,
"theoretical_loss": 3.6179582969318527,
"tokens_seen": 1095368704
},
{
"epoch": 0.33,
"learning_rate": 0.00033744232698094284,
"loss": 2.6215,
"theoretical_loss": 3.617938043024682,
"tokens_seen": 1095434240
},
{
"epoch": 0.33,
"learning_rate": 0.000337432296890672,
"loss": 2.8945,
"theoretical_loss": 3.6179177906684536,
"tokens_seen": 1095499776
},
{
"epoch": 0.33,
"learning_rate": 0.0003374222668004012,
"loss": 2.7473,
"theoretical_loss": 3.617897539862957,
"tokens_seen": 1095565312
},
{
"epoch": 0.33,
"learning_rate": 0.0003374122367101304,
"loss": 2.9123,
"theoretical_loss": 3.6178772906079804,
"tokens_seen": 1095630848
},
{
"epoch": 0.33,
"learning_rate": 0.0003374022066198596,
"loss": 2.9442,
"theoretical_loss": 3.617857042903312,
"tokens_seen": 1095696384
},
{
"epoch": 0.33,
"learning_rate": 0.00033739217652958874,
"loss": 2.7532,
"theoretical_loss": 3.6178367967487404,
"tokens_seen": 1095761920
},
{
"epoch": 0.33,
"learning_rate": 0.000337382146439318,
"loss": 2.8294,
"theoretical_loss": 3.617816552144054,
"tokens_seen": 1095827456
},
{
"epoch": 0.33,
"learning_rate": 0.0003373721163490471,
"loss": 2.8021,
"theoretical_loss": 3.6177963090890417,
"tokens_seen": 1095892992
},
{
"epoch": 0.33,
"learning_rate": 0.00033736208625877634,
"loss": 2.7619,
"theoretical_loss": 3.6177760675834927,
"tokens_seen": 1095958528
},
{
"epoch": 0.33,
"learning_rate": 0.0003373520561685055,
"loss": 2.8762,
"theoretical_loss": 3.617755827627195,
"tokens_seen": 1096024064
},
{
"epoch": 0.33,
"objective/train/docs_used": 624708,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9976794719696045,
"objective/train/theoretical_loss": 3.6177355892199383,
"objective/train/tokens_used": 1116549600,
"theoretical_loss": 3.6177355892199383,
"tokens_seen": 1096089600
},
{
"epoch": 0.33,
"learning_rate": 0.0003373420260782347,
"loss": 3.0007,
"theoretical_loss": 3.6177355892199383,
"tokens_seen": 1096089600
},
{
"epoch": 0.33,
"learning_rate": 0.0003373319959879639,
"loss": 2.7359,
"theoretical_loss": 3.6177153523615107,
"tokens_seen": 1096155136
},
{
"epoch": 0.33,
"learning_rate": 0.0003373219658976931,
"loss": 2.8883,
"theoretical_loss": 3.617695117051701,
"tokens_seen": 1096220672
},
{
"epoch": 0.33,
"learning_rate": 0.00033731193580742225,
"loss": 2.6286,
"theoretical_loss": 3.6176748832902987,
"tokens_seen": 1096286208
},
{
"epoch": 0.33,
"learning_rate": 0.0003373019057171515,
"loss": 2.8231,
"theoretical_loss": 3.6176546510770926,
"tokens_seen": 1096351744
},
{
"epoch": 0.33,
"learning_rate": 0.0003372918756268806,
"loss": 2.9542,
"theoretical_loss": 3.6176344204118713,
"tokens_seen": 1096417280
},
{
"epoch": 0.33,
"learning_rate": 0.00033728184553660984,
"loss": 2.9017,
"theoretical_loss": 3.617614191294424,
"tokens_seen": 1096482816
},
{
"epoch": 0.33,
"learning_rate": 0.000337271815446339,
"loss": 2.887,
"theoretical_loss": 3.6175939637245405,
"tokens_seen": 1096548352
},
{
"epoch": 0.33,
"learning_rate": 0.0003372617853560682,
"loss": 2.8257,
"theoretical_loss": 3.6175737377020085,
"tokens_seen": 1096613888
},
{
"epoch": 0.33,
"learning_rate": 0.0003372517552657974,
"loss": 2.7134,
"theoretical_loss": 3.6175535132266186,
"tokens_seen": 1096679424
},
{
"epoch": 0.33,
"learning_rate": 0.00033724172517552657,
"loss": 2.8109,
"theoretical_loss": 3.6175332902981587,
"tokens_seen": 1096744960
},
{
"epoch": 0.33,
"learning_rate": 0.00033723169508525575,
"loss": 2.9008,
"theoretical_loss": 3.6175130689164194,
"tokens_seen": 1096810496
},
{
"epoch": 0.33,
"learning_rate": 0.000337221664994985,
"loss": 2.7659,
"theoretical_loss": 3.6174928490811893,
"tokens_seen": 1096876032
},
{
"epoch": 0.33,
"learning_rate": 0.0003372116349047141,
"loss": 2.814,
"theoretical_loss": 3.617472630792258,
"tokens_seen": 1096941568
},
{
"epoch": 0.33,
"learning_rate": 0.00033720160481444335,
"loss": 2.9074,
"theoretical_loss": 3.617452414049414,
"tokens_seen": 1097007104
},
{
"epoch": 0.33,
"learning_rate": 0.0003371915747241725,
"loss": 2.7868,
"theoretical_loss": 3.617432198852448,
"tokens_seen": 1097072640
},
{
"epoch": 0.33,
"learning_rate": 0.0003371815446339017,
"loss": 2.7929,
"theoretical_loss": 3.6174119852011484,
"tokens_seen": 1097138176
},
{
"epoch": 0.33,
"learning_rate": 0.0003371715145436309,
"loss": 2.993,
"theoretical_loss": 3.6173917730953056,
"tokens_seen": 1097203712
},
{
"epoch": 0.33,
"learning_rate": 0.00033716148445336007,
"loss": 2.8783,
"theoretical_loss": 3.617371562534708,
"tokens_seen": 1097269248
},
{
"epoch": 0.33,
"learning_rate": 0.00033715145436308925,
"loss": 2.8414,
"theoretical_loss": 3.6173513535191466,
"tokens_seen": 1097334784
},
{
"epoch": 0.33,
"learning_rate": 0.0003371414242728185,
"loss": 2.7673,
"theoretical_loss": 3.61733114604841,
"tokens_seen": 1097400320
},
{
"epoch": 0.33,
"learning_rate": 0.0003371313941825476,
"loss": 2.7075,
"theoretical_loss": 3.6173109401222883,
"tokens_seen": 1097465856
},
{
"epoch": 0.33,
"learning_rate": 0.00033712136409227685,
"loss": 2.6105,
"theoretical_loss": 3.6172907357405704,
"tokens_seen": 1097531392
},
{
"epoch": 0.33,
"learning_rate": 0.000337111334002006,
"loss": 2.7759,
"theoretical_loss": 3.617270532903047,
"tokens_seen": 1097596928
},
{
"epoch": 0.33,
"learning_rate": 0.0003371013039117352,
"loss": 2.8803,
"theoretical_loss": 3.617250331609508,
"tokens_seen": 1097662464
},
{
"epoch": 0.33,
"objective/train/docs_used": 625328,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8064966201782227,
"objective/train/theoretical_loss": 3.617230131859743,
"objective/train/tokens_used": 1118188000,
"theoretical_loss": 3.617230131859743,
"tokens_seen": 1097728000
},
{
"epoch": 0.33,
"learning_rate": 0.0003370912738214644,
"loss": 2.855,
"theoretical_loss": 3.617230131859743,
"tokens_seen": 1097728000
},
{
"epoch": 0.33,
"learning_rate": 0.0003370812437311936,
"loss": 2.9841,
"theoretical_loss": 3.617209933653541,
"tokens_seen": 1097793536
},
{
"epoch": 0.33,
"learning_rate": 0.0003370712136409228,
"loss": 2.7245,
"theoretical_loss": 3.6171897369906922,
"tokens_seen": 1097859072
},
{
"epoch": 0.33,
"learning_rate": 0.00033706118355065194,
"loss": 2.8872,
"theoretical_loss": 3.6171695418709877,
"tokens_seen": 1097924608
},
{
"epoch": 0.33,
"learning_rate": 0.00033705115346038117,
"loss": 2.7532,
"theoretical_loss": 3.617149348294217,
"tokens_seen": 1097990144
},
{
"epoch": 0.33,
"learning_rate": 0.00033704112337011035,
"loss": 2.5497,
"theoretical_loss": 3.6171291562601695,
"tokens_seen": 1098055680
},
{
"epoch": 0.33,
"learning_rate": 0.00033703109327983953,
"loss": 2.7846,
"theoretical_loss": 3.6171089657686357,
"tokens_seen": 1098121216
},
{
"epoch": 0.33,
"learning_rate": 0.0003370210631895687,
"loss": 2.9733,
"theoretical_loss": 3.617088776819406,
"tokens_seen": 1098186752
},
{
"epoch": 0.33,
"learning_rate": 0.00033701103309929795,
"loss": 2.8992,
"theoretical_loss": 3.61706858941227,
"tokens_seen": 1098252288
},
{
"epoch": 0.33,
"learning_rate": 0.0003370010030090271,
"loss": 2.7328,
"theoretical_loss": 3.6170484035470176,
"tokens_seen": 1098317824
},
{
"epoch": 0.33,
"learning_rate": 0.0003369909729187563,
"loss": 2.8205,
"theoretical_loss": 3.61702821922344,
"tokens_seen": 1098383360
},
{
"epoch": 0.33,
"learning_rate": 0.00033698094282848544,
"loss": 3.192,
"theoretical_loss": 3.6170080364413275,
"tokens_seen": 1098448896
},
{
"epoch": 0.33,
"learning_rate": 0.0003369709127382147,
"loss": 2.6897,
"theoretical_loss": 3.6169878552004695,
"tokens_seen": 1098514432
},
{
"epoch": 0.33,
"learning_rate": 0.00033696088264794386,
"loss": 2.9493,
"theoretical_loss": 3.616967675500657,
"tokens_seen": 1098579968
},
{
"epoch": 0.33,
"learning_rate": 0.00033695085255767304,
"loss": 2.8288,
"theoretical_loss": 3.6169474973416804,
"tokens_seen": 1098645504
},
{
"epoch": 0.33,
"learning_rate": 0.0003369408224674022,
"loss": 2.502,
"theoretical_loss": 3.6169273207233292,
"tokens_seen": 1098711040
},
{
"epoch": 0.33,
"learning_rate": 0.0003369307923771314,
"loss": 2.7993,
"theoretical_loss": 3.6169071456453956,
"tokens_seen": 1098776576
},
{
"epoch": 0.33,
"learning_rate": 0.0003369207622868606,
"loss": 2.8948,
"theoretical_loss": 3.6168869721076686,
"tokens_seen": 1098842112
},
{
"epoch": 0.33,
"learning_rate": 0.0003369107321965898,
"loss": 2.8771,
"theoretical_loss": 3.61686680010994,
"tokens_seen": 1098907648
},
{
"epoch": 0.33,
"learning_rate": 0.00033690070210631894,
"loss": 2.7655,
"theoretical_loss": 3.616846629651999,
"tokens_seen": 1098973184
},
{
"epoch": 0.33,
"learning_rate": 0.0003368906720160482,
"loss": 2.6609,
"theoretical_loss": 3.6168264607336376,
"tokens_seen": 1099038720
},
{
"epoch": 0.33,
"learning_rate": 0.0003368806419257773,
"loss": 2.7443,
"theoretical_loss": 3.616806293354646,
"tokens_seen": 1099104256
},
{
"epoch": 0.33,
"learning_rate": 0.00033687061183550654,
"loss": 2.7576,
"theoretical_loss": 3.6167861275148145,
"tokens_seen": 1099169792
},
{
"epoch": 0.33,
"learning_rate": 0.0003368605817452357,
"loss": 3.0501,
"theoretical_loss": 3.6167659632139344,
"tokens_seen": 1099235328
},
{
"epoch": 0.33,
"learning_rate": 0.0003368505516549649,
"loss": 2.7541,
"theoretical_loss": 3.6167458004517963,
"tokens_seen": 1099300864
},
{
"epoch": 0.33,
"objective/train/docs_used": 626712,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4663491249084473,
"objective/train/theoretical_loss": 3.616725639228191,
"objective/train/tokens_used": 1119826400,
"theoretical_loss": 3.616725639228191,
"tokens_seen": 1099366400
},
{
"epoch": 0.33,
"learning_rate": 0.0003368405215646941,
"loss": 2.7476,
"theoretical_loss": 3.616725639228191,
"tokens_seen": 1099366400
},
{
"epoch": 0.33,
"learning_rate": 0.0003368304914744233,
"loss": 2.8463,
"theoretical_loss": 3.61670547954291,
"tokens_seen": 1099431936
},
{
"epoch": 0.33,
"learning_rate": 0.00033682046138415245,
"loss": 2.9191,
"theoretical_loss": 3.6166853213957433,
"tokens_seen": 1099497472
},
{
"epoch": 0.33,
"learning_rate": 0.0003368104312938817,
"loss": 2.8337,
"theoretical_loss": 3.616665164786482,
"tokens_seen": 1099563008
},
{
"epoch": 0.33,
"learning_rate": 0.0003368004012036108,
"loss": 2.7328,
"theoretical_loss": 3.616645009714918,
"tokens_seen": 1099628544
},
{
"epoch": 0.33,
"learning_rate": 0.00033679037111334004,
"loss": 2.8719,
"theoretical_loss": 3.6166248561808416,
"tokens_seen": 1099694080
},
{
"epoch": 0.33,
"learning_rate": 0.0003367803410230692,
"loss": 2.9544,
"theoretical_loss": 3.616604704184044,
"tokens_seen": 1099759616
},
{
"epoch": 0.33,
"learning_rate": 0.0003367703109327984,
"loss": 2.7704,
"theoretical_loss": 3.616584553724316,
"tokens_seen": 1099825152
},
{
"epoch": 0.33,
"learning_rate": 0.0003367602808425276,
"loss": 2.7187,
"theoretical_loss": 3.6165644048014505,
"tokens_seen": 1099890688
},
{
"epoch": 0.33,
"learning_rate": 0.00033675025075225677,
"loss": 2.8217,
"theoretical_loss": 3.616544257415236,
"tokens_seen": 1099956224
},
{
"epoch": 0.33,
"learning_rate": 0.00033674022066198595,
"loss": 2.8959,
"theoretical_loss": 3.6165241115654663,
"tokens_seen": 1100021760
},
{
"epoch": 0.33,
"learning_rate": 0.0003367301905717152,
"loss": 2.596,
"theoretical_loss": 3.616503967251931,
"tokens_seen": 1100087296
},
{
"epoch": 0.33,
"learning_rate": 0.0003367201604814443,
"loss": 2.6467,
"theoretical_loss": 3.616483824474422,
"tokens_seen": 1100152832
},
{
"epoch": 0.33,
"learning_rate": 0.00033671013039117355,
"loss": 2.7106,
"theoretical_loss": 3.6164636832327313,
"tokens_seen": 1100218368
},
{
"epoch": 0.33,
"learning_rate": 0.00033670010030090273,
"loss": 2.7007,
"theoretical_loss": 3.616443543526649,
"tokens_seen": 1100283904
},
{
"epoch": 0.33,
"learning_rate": 0.0003366900702106319,
"loss": 2.9253,
"theoretical_loss": 3.6164234053559676,
"tokens_seen": 1100349440
},
{
"epoch": 0.33,
"learning_rate": 0.0003366800401203611,
"loss": 2.9289,
"theoretical_loss": 3.616403268720479,
"tokens_seen": 1100414976
},
{
"epoch": 0.33,
"learning_rate": 0.00033667001003009027,
"loss": 2.7677,
"theoretical_loss": 3.616383133619973,
"tokens_seen": 1100480512
},
{
"epoch": 0.33,
"learning_rate": 0.00033665997993981945,
"loss": 2.8159,
"theoretical_loss": 3.6163630000542435,
"tokens_seen": 1100546048
},
{
"epoch": 0.33,
"learning_rate": 0.0003366499498495487,
"loss": 2.848,
"theoretical_loss": 3.6163428680230796,
"tokens_seen": 1100611584
},
{
"epoch": 0.33,
"learning_rate": 0.0003366399197592778,
"loss": 2.6666,
"theoretical_loss": 3.616322737526275,
"tokens_seen": 1100677120
},
{
"epoch": 0.33,
"learning_rate": 0.00033662988966900705,
"loss": 2.8327,
"theoretical_loss": 3.6163026085636205,
"tokens_seen": 1100742656
},
{
"epoch": 0.33,
"learning_rate": 0.0003366198595787362,
"loss": 2.9125,
"theoretical_loss": 3.6162824811349084,
"tokens_seen": 1100808192
},
{
"epoch": 0.33,
"learning_rate": 0.0003366098294884654,
"loss": 3.047,
"theoretical_loss": 3.61626235523993,
"tokens_seen": 1100873728
},
{
"epoch": 0.33,
"learning_rate": 0.0003365997993981946,
"loss": 2.8014,
"theoretical_loss": 3.616242230878477,
"tokens_seen": 1100939264
},
{
"epoch": 0.33,
"objective/train/docs_used": 627263,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6512184143066406,
"objective/train/theoretical_loss": 3.6162221080503416,
"objective/train/tokens_used": 1121464800,
"theoretical_loss": 3.6162221080503416,
"tokens_seen": 1101004800
},
{
"epoch": 0.33,
"learning_rate": 0.0003365897693079238,
"loss": 2.651,
"theoretical_loss": 3.6162221080503416,
"tokens_seen": 1101004800
},
{
"epoch": 0.33,
"learning_rate": 0.00033657973921765295,
"loss": 2.7943,
"theoretical_loss": 3.6162019867553155,
"tokens_seen": 1101070336
},
{
"epoch": 0.33,
"learning_rate": 0.00033656970912738214,
"loss": 2.7048,
"theoretical_loss": 3.616181866993191,
"tokens_seen": 1101135872
},
{
"epoch": 0.33,
"learning_rate": 0.0003365596790371113,
"loss": 2.9445,
"theoretical_loss": 3.6161617487637603,
"tokens_seen": 1101201408
},
{
"epoch": 0.33,
"learning_rate": 0.00033654964894684055,
"loss": 2.8262,
"theoretical_loss": 3.6161416320668147,
"tokens_seen": 1101266944
},
{
"epoch": 0.33,
"learning_rate": 0.0003365396188565697,
"loss": 2.7006,
"theoretical_loss": 3.6161215169021466,
"tokens_seen": 1101332480
},
{
"epoch": 0.33,
"learning_rate": 0.0003365295887662989,
"loss": 2.8283,
"theoretical_loss": 3.616101403269548,
"tokens_seen": 1101398016
},
{
"epoch": 0.33,
"learning_rate": 0.0003365195586760281,
"loss": 2.952,
"theoretical_loss": 3.616081291168811,
"tokens_seen": 1101463552
},
{
"epoch": 0.33,
"learning_rate": 0.0003365095285857573,
"loss": 2.7475,
"theoretical_loss": 3.6160611805997283,
"tokens_seen": 1101529088
},
{
"epoch": 0.33,
"learning_rate": 0.00033649949849548646,
"loss": 3.0166,
"theoretical_loss": 3.616041071562092,
"tokens_seen": 1101594624
},
{
"epoch": 0.33,
"learning_rate": 0.00033648946840521564,
"loss": 2.9459,
"theoretical_loss": 3.616020964055694,
"tokens_seen": 1101660160
},
{
"epoch": 0.33,
"learning_rate": 0.0003364794383149448,
"loss": 2.8278,
"theoretical_loss": 3.616000858080327,
"tokens_seen": 1101725696
},
{
"epoch": 0.33,
"learning_rate": 0.00033646940822467406,
"loss": 2.6301,
"theoretical_loss": 3.615980753635783,
"tokens_seen": 1101791232
},
{
"epoch": 0.33,
"learning_rate": 0.0003364593781344032,
"loss": 2.8888,
"theoretical_loss": 3.615960650721855,
"tokens_seen": 1101856768
},
{
"epoch": 0.33,
"learning_rate": 0.0003364493480441324,
"loss": 2.6239,
"theoretical_loss": 3.6159405493383345,
"tokens_seen": 1101922304
},
{
"epoch": 0.33,
"learning_rate": 0.00033643931795386154,
"loss": 2.8704,
"theoretical_loss": 3.6159204494850146,
"tokens_seen": 1101987840
},
{
"epoch": 0.33,
"learning_rate": 0.0003364292878635908,
"loss": 2.7586,
"theoretical_loss": 3.615900351161688,
"tokens_seen": 1102053376
},
{
"epoch": 0.33,
"learning_rate": 0.00033641925777331996,
"loss": 2.4564,
"theoretical_loss": 3.6158802543681468,
"tokens_seen": 1102118912
},
{
"epoch": 0.33,
"learning_rate": 0.00033640922768304914,
"loss": 2.7289,
"theoretical_loss": 3.615860159104184,
"tokens_seen": 1102184448
},
{
"epoch": 0.33,
"learning_rate": 0.0003363991975927783,
"loss": 2.6576,
"theoretical_loss": 3.615840065369592,
"tokens_seen": 1102249984
},
{
"epoch": 0.33,
"learning_rate": 0.0003363891675025075,
"loss": 2.7203,
"theoretical_loss": 3.6158199731641636,
"tokens_seen": 1102315520
},
{
"epoch": 0.33,
"learning_rate": 0.0003363791374122367,
"loss": 2.7681,
"theoretical_loss": 3.6157998824876914,
"tokens_seen": 1102381056
},
{
"epoch": 0.33,
"learning_rate": 0.0003363691073219659,
"loss": 2.6521,
"theoretical_loss": 3.6157797933399682,
"tokens_seen": 1102446592
},
{
"epoch": 0.33,
"learning_rate": 0.00033635907723169505,
"loss": 2.7282,
"theoretical_loss": 3.6157597057207873,
"tokens_seen": 1102512128
},
{
"epoch": 0.33,
"learning_rate": 0.0003363490471414243,
"loss": 2.9267,
"theoretical_loss": 3.6157396196299403,
"tokens_seen": 1102577664
},
{
"epoch": 0.33,
"objective/train/docs_used": 628470,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.904186725616455,
"objective/train/theoretical_loss": 3.615719535067221,
"objective/train/tokens_used": 1123103200,
"theoretical_loss": 3.615719535067221,
"tokens_seen": 1102643200
},
{
"epoch": 0.33,
"learning_rate": 0.00033633901705115346,
"loss": 2.5501,
"theoretical_loss": 3.615719535067221,
"tokens_seen": 1102643200
},
{
"epoch": 0.33,
"learning_rate": 0.00033632898696088265,
"loss": 2.8633,
"theoretical_loss": 3.6156994520324224,
"tokens_seen": 1102708736
},
{
"epoch": 0.33,
"learning_rate": 0.0003363189568706119,
"loss": 2.855,
"theoretical_loss": 3.6156793705253376,
"tokens_seen": 1102774272
},
{
"epoch": 0.33,
"learning_rate": 0.000336308926780341,
"loss": 2.6297,
"theoretical_loss": 3.615659290545759,
"tokens_seen": 1102839808
},
{
"epoch": 0.33,
"learning_rate": 0.00033629889669007024,
"loss": 2.8356,
"theoretical_loss": 3.6156392120934795,
"tokens_seen": 1102905344
},
{
"epoch": 0.33,
"learning_rate": 0.0003362888665997994,
"loss": 2.8024,
"theoretical_loss": 3.6156191351682936,
"tokens_seen": 1102970880
},
{
"epoch": 0.33,
"learning_rate": 0.0003362788365095286,
"loss": 2.6984,
"theoretical_loss": 3.615599059769993,
"tokens_seen": 1103036416
},
{
"epoch": 0.33,
"learning_rate": 0.0003362688064192578,
"loss": 2.5766,
"theoretical_loss": 3.615578985898371,
"tokens_seen": 1103101952
},
{
"epoch": 0.33,
"learning_rate": 0.00033625877632898697,
"loss": 2.6332,
"theoretical_loss": 3.615558913553221,
"tokens_seen": 1103167488
},
{
"epoch": 0.33,
"learning_rate": 0.00033624874623871615,
"loss": 2.9194,
"theoretical_loss": 3.615538842734337,
"tokens_seen": 1103233024
},
{
"epoch": 0.33,
"learning_rate": 0.0003362387161484454,
"loss": 2.7642,
"theoretical_loss": 3.615518773441511,
"tokens_seen": 1103298560
},
{
"epoch": 0.33,
"learning_rate": 0.0003362286860581745,
"loss": 2.9489,
"theoretical_loss": 3.615498705674537,
"tokens_seen": 1103364096
},
{
"epoch": 0.33,
"learning_rate": 0.00033621865596790375,
"loss": 2.7583,
"theoretical_loss": 3.615478639433209,
"tokens_seen": 1103429632
},
{
"epoch": 0.33,
"learning_rate": 0.00033620862587763293,
"loss": 2.7024,
"theoretical_loss": 3.6154585747173194,
"tokens_seen": 1103495168
},
{
"epoch": 0.33,
"learning_rate": 0.0003361985957873621,
"loss": 2.8871,
"theoretical_loss": 3.615438511526662,
"tokens_seen": 1103560704
},
{
"epoch": 0.33,
"learning_rate": 0.0003361885656970913,
"loss": 2.9806,
"theoretical_loss": 3.6154184498610307,
"tokens_seen": 1103626240
},
{
"epoch": 0.33,
"learning_rate": 0.00033617853560682047,
"loss": 2.7416,
"theoretical_loss": 3.6153983897202187,
"tokens_seen": 1103691776
},
{
"epoch": 0.33,
"learning_rate": 0.00033616850551654965,
"loss": 2.7067,
"theoretical_loss": 3.6153783311040186,
"tokens_seen": 1103757312
},
{
"epoch": 0.33,
"learning_rate": 0.0003361584754262789,
"loss": 2.9121,
"theoretical_loss": 3.615358274012226,
"tokens_seen": 1103822848
},
{
"epoch": 0.33,
"learning_rate": 0.000336148445336008,
"loss": 2.9651,
"theoretical_loss": 3.6153382184446325,
"tokens_seen": 1103888384
},
{
"epoch": 0.33,
"learning_rate": 0.00033613841524573725,
"loss": 2.8923,
"theoretical_loss": 3.6153181644010335,
"tokens_seen": 1103953920
},
{
"epoch": 0.33,
"learning_rate": 0.0003361283851554664,
"loss": 2.8935,
"theoretical_loss": 3.6152981118812217,
"tokens_seen": 1104019456
},
{
"epoch": 0.33,
"learning_rate": 0.0003361183550651956,
"loss": 2.6723,
"theoretical_loss": 3.6152780608849913,
"tokens_seen": 1104084992
},
{
"epoch": 0.33,
"learning_rate": 0.0003361083249749248,
"loss": 2.7965,
"theoretical_loss": 3.615258011412136,
"tokens_seen": 1104150528
},
{
"epoch": 0.33,
"learning_rate": 0.000336098294884654,
"loss": 2.6593,
"theoretical_loss": 3.6152379634624494,
"tokens_seen": 1104216064
},
{
"epoch": 0.33,
"objective/train/docs_used": 629051,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5870096683502197,
"objective/train/theoretical_loss": 3.615217917035726,
"objective/train/tokens_used": 1124741600,
"theoretical_loss": 3.615217917035726,
"tokens_seen": 1104281600
},
{
"epoch": 0.33,
"learning_rate": 0.00033608826479438316,
"loss": 2.7457,
"theoretical_loss": 3.615217917035726,
"tokens_seen": 1104281600
},
{
"epoch": 0.33,
"learning_rate": 0.00033607823470411234,
"loss": 2.6401,
"theoretical_loss": 3.615197872131759,
"tokens_seen": 1104347136
},
{
"epoch": 0.33,
"learning_rate": 0.0003360682046138415,
"loss": 2.8359,
"theoretical_loss": 3.615177828750343,
"tokens_seen": 1104412672
},
{
"epoch": 0.33,
"learning_rate": 0.00033605817452357075,
"loss": 3.0754,
"theoretical_loss": 3.6151577868912717,
"tokens_seen": 1104478208
},
{
"epoch": 0.33,
"learning_rate": 0.0003360481444332999,
"loss": 2.7979,
"theoretical_loss": 3.615137746554339,
"tokens_seen": 1104543744
},
{
"epoch": 0.33,
"learning_rate": 0.0003360381143430291,
"loss": 3.0203,
"theoretical_loss": 3.6151177077393397,
"tokens_seen": 1104609280
},
{
"epoch": 0.33,
"learning_rate": 0.0003360280842527583,
"loss": 2.6359,
"theoretical_loss": 3.615097670446067,
"tokens_seen": 1104674816
},
{
"epoch": 0.33,
"learning_rate": 0.0003360180541624875,
"loss": 2.9658,
"theoretical_loss": 3.615077634674316,
"tokens_seen": 1104740352
},
{
"epoch": 0.33,
"learning_rate": 0.00033600802407221666,
"loss": 2.6826,
"theoretical_loss": 3.6150576004238797,
"tokens_seen": 1104805888
},
{
"epoch": 0.33,
"learning_rate": 0.00033599799398194584,
"loss": 3.1063,
"theoretical_loss": 3.6150375676945536,
"tokens_seen": 1104871424
},
{
"epoch": 0.33,
"learning_rate": 0.000335987963891675,
"loss": 2.9796,
"theoretical_loss": 3.6150175364861314,
"tokens_seen": 1104936960
},
{
"epoch": 0.33,
"learning_rate": 0.00033597793380140426,
"loss": 2.7002,
"theoretical_loss": 3.6149975067984075,
"tokens_seen": 1105002496
},
{
"epoch": 0.33,
"learning_rate": 0.0003359679037111334,
"loss": 3.1258,
"theoretical_loss": 3.614977478631176,
"tokens_seen": 1105068032
},
{
"epoch": 0.33,
"learning_rate": 0.0003359578736208626,
"loss": 2.6779,
"theoretical_loss": 3.6149574519842322,
"tokens_seen": 1105133568
},
{
"epoch": 0.33,
"learning_rate": 0.00033594784353059174,
"loss": 2.8053,
"theoretical_loss": 3.6149374268573697,
"tokens_seen": 1105199104
},
{
"epoch": 0.33,
"learning_rate": 0.000335937813440321,
"loss": 2.7281,
"theoretical_loss": 3.6149174032503835,
"tokens_seen": 1105264640
},
{
"epoch": 0.33,
"learning_rate": 0.00033592778335005016,
"loss": 2.9565,
"theoretical_loss": 3.6148973811630674,
"tokens_seen": 1105330176
},
{
"epoch": 0.33,
"learning_rate": 0.00033591775325977934,
"loss": 2.9339,
"theoretical_loss": 3.6148773605952167,
"tokens_seen": 1105395712
},
{
"epoch": 0.33,
"learning_rate": 0.0003359077231695085,
"loss": 2.7226,
"theoretical_loss": 3.6148573415466254,
"tokens_seen": 1105461248
},
{
"epoch": 0.34,
"learning_rate": 0.0003358976930792377,
"loss": 2.8038,
"theoretical_loss": 3.6148373240170892,
"tokens_seen": 1105526784
},
{
"epoch": 0.34,
"learning_rate": 0.0003358876629889669,
"loss": 2.8803,
"theoretical_loss": 3.614817308006401,
"tokens_seen": 1105592320
},
{
"epoch": 0.34,
"learning_rate": 0.0003358776328986961,
"loss": 2.867,
"theoretical_loss": 3.614797293514358,
"tokens_seen": 1105657856
},
{
"epoch": 0.34,
"learning_rate": 0.00033586760280842525,
"loss": 2.9638,
"theoretical_loss": 3.614777280540753,
"tokens_seen": 1105723392
},
{
"epoch": 0.34,
"learning_rate": 0.0003358575727181545,
"loss": 2.6823,
"theoretical_loss": 3.6147572690853815,
"tokens_seen": 1105788928
},
{
"epoch": 0.34,
"learning_rate": 0.00033584754262788366,
"loss": 2.875,
"theoretical_loss": 3.614737259148038,
"tokens_seen": 1105854464
},
{
"epoch": 0.34,
"objective/train/docs_used": 630354,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0831401348114014,
"objective/train/theoretical_loss": 3.614717250728518,
"objective/train/tokens_used": 1126380000,
"theoretical_loss": 3.614717250728518,
"tokens_seen": 1105920000
},
{
"epoch": 0.34,
"learning_rate": 0.00033583751253761285,
"loss": 2.9851,
"theoretical_loss": 3.614717250728518,
"tokens_seen": 1105920000
},
{
"epoch": 0.34,
"learning_rate": 0.000335827482447342,
"loss": 2.9,
"theoretical_loss": 3.6146972438266163,
"tokens_seen": 1105985536
},
{
"epoch": 0.34,
"learning_rate": 0.0003358174523570712,
"loss": 2.6426,
"theoretical_loss": 3.6146772384421273,
"tokens_seen": 1106051072
},
{
"epoch": 0.34,
"learning_rate": 0.0003358074222668004,
"loss": 2.8515,
"theoretical_loss": 3.6146572345748464,
"tokens_seen": 1106116608
},
{
"epoch": 0.34,
"learning_rate": 0.0003357973921765296,
"loss": 2.475,
"theoretical_loss": 3.614637232224569,
"tokens_seen": 1106182144
},
{
"epoch": 0.34,
"learning_rate": 0.00033578736208625875,
"loss": 2.7845,
"theoretical_loss": 3.6146172313910894,
"tokens_seen": 1106247680
},
{
"epoch": 0.34,
"learning_rate": 0.000335777331995988,
"loss": 3.1055,
"theoretical_loss": 3.6145972320742037,
"tokens_seen": 1106313216
},
{
"epoch": 0.34,
"learning_rate": 0.0003357673019057171,
"loss": 2.62,
"theoretical_loss": 3.614577234273706,
"tokens_seen": 1106378752
},
{
"epoch": 0.34,
"learning_rate": 0.00033575727181544635,
"loss": 2.8343,
"theoretical_loss": 3.6145572379893927,
"tokens_seen": 1106444288
},
{
"epoch": 0.34,
"learning_rate": 0.00033574724172517553,
"loss": 2.6114,
"theoretical_loss": 3.614537243221058,
"tokens_seen": 1106509824
},
{
"epoch": 0.34,
"learning_rate": 0.0003357372116349047,
"loss": 2.7002,
"theoretical_loss": 3.6145172499684977,
"tokens_seen": 1106575360
},
{
"epoch": 0.34,
"learning_rate": 0.0003357271815446339,
"loss": 2.8165,
"theoretical_loss": 3.6144972582315065,
"tokens_seen": 1106640896
},
{
"epoch": 0.34,
"learning_rate": 0.00033571715145436313,
"loss": 2.8835,
"theoretical_loss": 3.614477268009881,
"tokens_seen": 1106706432
},
{
"epoch": 0.34,
"learning_rate": 0.00033570712136409225,
"loss": 2.9088,
"theoretical_loss": 3.6144572793034158,
"tokens_seen": 1106771968
},
{
"epoch": 0.34,
"learning_rate": 0.0003356970912738215,
"loss": 2.8632,
"theoretical_loss": 3.6144372921119063,
"tokens_seen": 1106837504
},
{
"epoch": 0.34,
"learning_rate": 0.0003356870611835506,
"loss": 2.8433,
"theoretical_loss": 3.614417306435148,
"tokens_seen": 1106903040
},
{
"epoch": 0.34,
"learning_rate": 0.00033567703109327985,
"loss": 2.7411,
"theoretical_loss": 3.614397322272937,
"tokens_seen": 1106968576
},
{
"epoch": 0.34,
"learning_rate": 0.00033566700100300903,
"loss": 2.5221,
"theoretical_loss": 3.6143773396250687,
"tokens_seen": 1107034112
},
{
"epoch": 0.34,
"learning_rate": 0.0003356569709127382,
"loss": 2.7519,
"theoretical_loss": 3.614357358491338,
"tokens_seen": 1107099648
},
{
"epoch": 0.34,
"learning_rate": 0.0003356469408224674,
"loss": 2.5795,
"theoretical_loss": 3.614337378871541,
"tokens_seen": 1107165184
},
{
"epoch": 0.34,
"learning_rate": 0.0003356369107321966,
"loss": 2.7536,
"theoretical_loss": 3.6143174007654735,
"tokens_seen": 1107230720
},
{
"epoch": 0.34,
"learning_rate": 0.00033562688064192576,
"loss": 2.7769,
"theoretical_loss": 3.614297424172931,
"tokens_seen": 1107296256
},
{
"epoch": 0.34,
"learning_rate": 0.000335616850551655,
"loss": 2.8017,
"theoretical_loss": 3.6142774490937093,
"tokens_seen": 1107361792
},
{
"epoch": 0.34,
"learning_rate": 0.0003356068204613841,
"loss": 2.5502,
"theoretical_loss": 3.6142574755276047,
"tokens_seen": 1107427328
},
{
"epoch": 0.34,
"learning_rate": 0.00033559679037111336,
"loss": 2.999,
"theoretical_loss": 3.6142375034744125,
"tokens_seen": 1107492864
},
{
"epoch": 0.34,
"objective/train/docs_used": 630918,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6021676063537598,
"objective/train/theoretical_loss": 3.614217532933929,
"objective/train/tokens_used": 1128018400,
"theoretical_loss": 3.614217532933929,
"tokens_seen": 1107558400
},
{
"epoch": 0.34,
"learning_rate": 0.0003355867602808425,
"loss": 2.6942,
"theoretical_loss": 3.614217532933929,
"tokens_seen": 1107558400
},
{
"epoch": 0.34,
"learning_rate": 0.0003355767301905717,
"loss": 2.8617,
"theoretical_loss": 3.6141975639059494,
"tokens_seen": 1107623936
},
{
"epoch": 0.34,
"learning_rate": 0.00033556670010030095,
"loss": 2.7636,
"theoretical_loss": 3.6141775963902703,
"tokens_seen": 1107689472
},
{
"epoch": 0.34,
"learning_rate": 0.0003355566700100301,
"loss": 2.6072,
"theoretical_loss": 3.6141576303866874,
"tokens_seen": 1107755008
},
{
"epoch": 0.34,
"learning_rate": 0.0003355466399197593,
"loss": 2.8879,
"theoretical_loss": 3.6141376658949973,
"tokens_seen": 1107820544
},
{
"epoch": 0.34,
"learning_rate": 0.0003355366098294885,
"loss": 2.9558,
"theoretical_loss": 3.614117702914995,
"tokens_seen": 1107886080
},
{
"epoch": 0.34,
"learning_rate": 0.0003355265797392177,
"loss": 2.6605,
"theoretical_loss": 3.614097741446478,
"tokens_seen": 1107951616
},
{
"epoch": 0.34,
"learning_rate": 0.00033551654964894686,
"loss": 2.8668,
"theoretical_loss": 3.6140777814892413,
"tokens_seen": 1108017152
},
{
"epoch": 0.34,
"learning_rate": 0.00033550651955867604,
"loss": 2.9977,
"theoretical_loss": 3.6140578230430815,
"tokens_seen": 1108082688
},
{
"epoch": 0.34,
"learning_rate": 0.0003354964894684052,
"loss": 2.9509,
"theoretical_loss": 3.6140378661077954,
"tokens_seen": 1108148224
},
{
"epoch": 0.34,
"learning_rate": 0.00033548645937813446,
"loss": 2.9658,
"theoretical_loss": 3.6140179106831787,
"tokens_seen": 1108213760
},
{
"epoch": 0.34,
"learning_rate": 0.0003354764292878636,
"loss": 2.7394,
"theoretical_loss": 3.6139979567690275,
"tokens_seen": 1108279296
},
{
"epoch": 0.34,
"learning_rate": 0.0003354663991975928,
"loss": 2.7803,
"theoretical_loss": 3.613978004365139,
"tokens_seen": 1108344832
},
{
"epoch": 0.34,
"learning_rate": 0.00033545636910732195,
"loss": 2.9092,
"theoretical_loss": 3.6139580534713085,
"tokens_seen": 1108410368
},
{
"epoch": 0.34,
"learning_rate": 0.0003354463390170512,
"loss": 2.7155,
"theoretical_loss": 3.613938104087333,
"tokens_seen": 1108475904
},
{
"epoch": 0.34,
"learning_rate": 0.00033543630892678036,
"loss": 2.8691,
"theoretical_loss": 3.6139181562130096,
"tokens_seen": 1108541440
},
{
"epoch": 0.34,
"learning_rate": 0.00033542627883650954,
"loss": 2.9269,
"theoretical_loss": 3.6138982098481334,
"tokens_seen": 1108606976
},
{
"epoch": 0.34,
"learning_rate": 0.0003354162487462387,
"loss": 2.8834,
"theoretical_loss": 3.6138782649925023,
"tokens_seen": 1108672512
},
{
"epoch": 0.34,
"learning_rate": 0.0003354062186559679,
"loss": 2.7947,
"theoretical_loss": 3.613858321645912,
"tokens_seen": 1108738048
},
{
"epoch": 0.34,
"learning_rate": 0.0003353961885656971,
"loss": 2.8029,
"theoretical_loss": 3.61383837980816,
"tokens_seen": 1108803584
},
{
"epoch": 0.34,
"learning_rate": 0.0003353861584754263,
"loss": 2.8418,
"theoretical_loss": 3.613818439479042,
"tokens_seen": 1108869120
},
{
"epoch": 0.34,
"learning_rate": 0.00033537612838515545,
"loss": 2.7181,
"theoretical_loss": 3.6137985006583557,
"tokens_seen": 1108934656
},
{
"epoch": 0.34,
"learning_rate": 0.0003353660982948847,
"loss": 2.6323,
"theoretical_loss": 3.613778563345897,
"tokens_seen": 1109000192
},
{
"epoch": 0.34,
"learning_rate": 0.00033535606820461386,
"loss": 2.9105,
"theoretical_loss": 3.613758627541463,
"tokens_seen": 1109065728
},
{
"epoch": 0.34,
"learning_rate": 0.00033534603811434305,
"loss": 2.7222,
"theoretical_loss": 3.6137386932448505,
"tokens_seen": 1109131264
},
{
"epoch": 0.34,
"objective/train/docs_used": 632203,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.511586904525757,
"objective/train/theoretical_loss": 3.613718760455857,
"objective/train/tokens_used": 1129656800,
"theoretical_loss": 3.613718760455857,
"tokens_seen": 1109196800
},
{
"epoch": 0.34,
"learning_rate": 0.0003353360080240722,
"loss": 2.7363,
"theoretical_loss": 3.613718760455857,
"tokens_seen": 1109196800
},
{
"epoch": 0.34,
"learning_rate": 0.0003353259779338014,
"loss": 2.7138,
"theoretical_loss": 3.613698829174278,
"tokens_seen": 1109262336
},
{
"epoch": 0.34,
"learning_rate": 0.0003353159478435306,
"loss": 2.4828,
"theoretical_loss": 3.613678899399912,
"tokens_seen": 1109327872
},
{
"epoch": 0.34,
"learning_rate": 0.0003353059177532598,
"loss": 2.9716,
"theoretical_loss": 3.6136589711325557,
"tokens_seen": 1109393408
},
{
"epoch": 0.34,
"learning_rate": 0.00033529588766298895,
"loss": 2.7271,
"theoretical_loss": 3.613639044372005,
"tokens_seen": 1109458944
},
{
"epoch": 0.34,
"learning_rate": 0.0003352858575727182,
"loss": 2.7787,
"theoretical_loss": 3.613619119118058,
"tokens_seen": 1109524480
},
{
"epoch": 0.34,
"learning_rate": 0.0003352758274824473,
"loss": 2.9427,
"theoretical_loss": 3.6135991953705116,
"tokens_seen": 1109590016
},
{
"epoch": 0.34,
"learning_rate": 0.00033526579739217655,
"loss": 2.7428,
"theoretical_loss": 3.6135792731291625,
"tokens_seen": 1109655552
},
{
"epoch": 0.34,
"learning_rate": 0.00033525576730190573,
"loss": 3.0903,
"theoretical_loss": 3.6135593523938088,
"tokens_seen": 1109721088
},
{
"epoch": 0.34,
"learning_rate": 0.0003352457372116349,
"loss": 2.8525,
"theoretical_loss": 3.613539433164247,
"tokens_seen": 1109786624
},
{
"epoch": 0.34,
"learning_rate": 0.0003352357071213641,
"loss": 2.7676,
"theoretical_loss": 3.613519515440274,
"tokens_seen": 1109852160
},
{
"epoch": 0.34,
"learning_rate": 0.00033522567703109333,
"loss": 3.0649,
"theoretical_loss": 3.6134995992216883,
"tokens_seen": 1109917696
},
{
"epoch": 0.34,
"learning_rate": 0.00033521564694082245,
"loss": 2.7937,
"theoretical_loss": 3.6134796845082864,
"tokens_seen": 1109983232
},
{
"epoch": 0.34,
"learning_rate": 0.0003352056168505517,
"loss": 3.045,
"theoretical_loss": 3.613459771299866,
"tokens_seen": 1110048768
},
{
"epoch": 0.34,
"learning_rate": 0.0003351955867602808,
"loss": 2.7311,
"theoretical_loss": 3.613439859596224,
"tokens_seen": 1110114304
},
{
"epoch": 0.34,
"learning_rate": 0.00033518555667001005,
"loss": 2.7102,
"theoretical_loss": 3.613419949397159,
"tokens_seen": 1110179840
},
{
"epoch": 0.34,
"learning_rate": 0.00033517552657973923,
"loss": 2.7269,
"theoretical_loss": 3.6134000407024676,
"tokens_seen": 1110245376
},
{
"epoch": 0.34,
"learning_rate": 0.0003351654964894684,
"loss": 2.8204,
"theoretical_loss": 3.6133801335119475,
"tokens_seen": 1110310912
},
{
"epoch": 0.34,
"learning_rate": 0.0003351554663991976,
"loss": 2.628,
"theoretical_loss": 3.613360227825396,
"tokens_seen": 1110376448
},
{
"epoch": 0.34,
"learning_rate": 0.0003351454363089268,
"loss": 3.0326,
"theoretical_loss": 3.6133403236426114,
"tokens_seen": 1110441984
},
{
"epoch": 0.34,
"learning_rate": 0.00033513540621865596,
"loss": 2.8738,
"theoretical_loss": 3.6133204209633907,
"tokens_seen": 1110507520
},
{
"epoch": 0.34,
"learning_rate": 0.0003351253761283852,
"loss": 2.7193,
"theoretical_loss": 3.613300519787532,
"tokens_seen": 1110573056
},
{
"epoch": 0.34,
"learning_rate": 0.0003351153460381143,
"loss": 2.9122,
"theoretical_loss": 3.613280620114833,
"tokens_seen": 1110638592
},
{
"epoch": 0.34,
"learning_rate": 0.00033510531594784356,
"loss": 2.8921,
"theoretical_loss": 3.6132607219450916,
"tokens_seen": 1110704128
},
{
"epoch": 0.34,
"learning_rate": 0.0003350952858575727,
"loss": 2.9127,
"theoretical_loss": 3.6132408252781048,
"tokens_seen": 1110769664
},
{
"epoch": 0.34,
"objective/train/docs_used": 632721,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1957690715789795,
"objective/train/theoretical_loss": 3.6132209301136715,
"objective/train/tokens_used": 1131295200,
"theoretical_loss": 3.6132209301136715,
"tokens_seen": 1110835200
},
{
"epoch": 0.34,
"learning_rate": 0.0003350852557673019,
"loss": 2.7819,
"theoretical_loss": 3.6132209301136715,
"tokens_seen": 1110835200
},
{
"epoch": 0.34,
"learning_rate": 0.0003350752256770311,
"loss": 2.8398,
"theoretical_loss": 3.6132010364515894,
"tokens_seen": 1110900736
},
{
"epoch": 0.34,
"learning_rate": 0.0003350651955867603,
"loss": 2.703,
"theoretical_loss": 3.613181144291656,
"tokens_seen": 1110966272
},
{
"epoch": 0.34,
"learning_rate": 0.00033505516549648946,
"loss": 2.7306,
"theoretical_loss": 3.6131612536336695,
"tokens_seen": 1111031808
},
{
"epoch": 0.34,
"learning_rate": 0.0003350451354062187,
"loss": 2.6332,
"theoretical_loss": 3.6131413644774275,
"tokens_seen": 1111097344
},
{
"epoch": 0.34,
"learning_rate": 0.0003350351053159478,
"loss": 2.7469,
"theoretical_loss": 3.613121476822729,
"tokens_seen": 1111162880
},
{
"epoch": 0.34,
"learning_rate": 0.00033502507522567706,
"loss": 2.6241,
"theoretical_loss": 3.613101590669371,
"tokens_seen": 1111228416
},
{
"epoch": 0.34,
"learning_rate": 0.0003350150451354062,
"loss": 2.774,
"theoretical_loss": 3.6130817060171525,
"tokens_seen": 1111293952
},
{
"epoch": 0.34,
"learning_rate": 0.0003350050150451354,
"loss": 2.5481,
"theoretical_loss": 3.6130618228658715,
"tokens_seen": 1111359488
},
{
"epoch": 0.34,
"learning_rate": 0.0003349949849548646,
"loss": 2.6742,
"theoretical_loss": 3.6130419412153256,
"tokens_seen": 1111425024
},
{
"epoch": 0.34,
"learning_rate": 0.0003349849548645938,
"loss": 2.9338,
"theoretical_loss": 3.6130220610653137,
"tokens_seen": 1111490560
},
{
"epoch": 0.34,
"learning_rate": 0.00033497492477432296,
"loss": 2.8657,
"theoretical_loss": 3.613002182415634,
"tokens_seen": 1111556096
},
{
"epoch": 0.34,
"learning_rate": 0.00033496489468405215,
"loss": 2.7368,
"theoretical_loss": 3.6129823052660845,
"tokens_seen": 1111621632
},
{
"epoch": 0.34,
"learning_rate": 0.0003349548645937813,
"loss": 2.8201,
"theoretical_loss": 3.6129624296164637,
"tokens_seen": 1111687168
},
{
"epoch": 0.34,
"learning_rate": 0.00033494483450351056,
"loss": 2.8364,
"theoretical_loss": 3.61294255546657,
"tokens_seen": 1111752704
},
{
"epoch": 0.34,
"learning_rate": 0.0003349348044132397,
"loss": 2.7586,
"theoretical_loss": 3.612922682816202,
"tokens_seen": 1111818240
},
{
"epoch": 0.34,
"learning_rate": 0.0003349247743229689,
"loss": 2.6124,
"theoretical_loss": 3.6129028116651583,
"tokens_seen": 1111883776
},
{
"epoch": 0.34,
"learning_rate": 0.00033491474423269805,
"loss": 2.9045,
"theoretical_loss": 3.612882942013237,
"tokens_seen": 1111949312
},
{
"epoch": 0.34,
"learning_rate": 0.0003349047141424273,
"loss": 2.7054,
"theoretical_loss": 3.612863073860237,
"tokens_seen": 1112014848
},
{
"epoch": 0.34,
"learning_rate": 0.00033489468405215647,
"loss": 2.6627,
"theoretical_loss": 3.6128432072059566,
"tokens_seen": 1112080384
},
{
"epoch": 0.34,
"learning_rate": 0.00033488465396188565,
"loss": 2.878,
"theoretical_loss": 3.6128233420501945,
"tokens_seen": 1112145920
},
{
"epoch": 0.34,
"learning_rate": 0.00033487462387161483,
"loss": 2.6196,
"theoretical_loss": 3.61280347839275,
"tokens_seen": 1112211456
},
{
"epoch": 0.34,
"learning_rate": 0.00033486459378134406,
"loss": 2.7715,
"theoretical_loss": 3.6127836162334206,
"tokens_seen": 1112276992
},
{
"epoch": 0.34,
"learning_rate": 0.0003348545636910732,
"loss": 2.8168,
"theoretical_loss": 3.612763755572006,
"tokens_seen": 1112342528
},
{
"epoch": 0.34,
"learning_rate": 0.0003348445336008024,
"loss": 2.7374,
"theoretical_loss": 3.6127438964083045,
"tokens_seen": 1112408064
},
{
"epoch": 0.34,
"objective/train/docs_used": 634025,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.472653388977051,
"objective/train/theoretical_loss": 3.6127240387421153,
"objective/train/tokens_used": 1132933600,
"theoretical_loss": 3.6127240387421153,
"tokens_seen": 1112473600
},
{
"epoch": 0.34,
"learning_rate": 0.0003348345035105316,
"loss": 2.6992,
"theoretical_loss": 3.6127240387421153,
"tokens_seen": 1112473600
},
{
"epoch": 0.34,
"learning_rate": 0.0003348244734202608,
"loss": 2.6468,
"theoretical_loss": 3.6127041825732373,
"tokens_seen": 1112539136
},
{
"epoch": 0.34,
"learning_rate": 0.00033481444332999,
"loss": 2.714,
"theoretical_loss": 3.612684327901469,
"tokens_seen": 1112604672
},
{
"epoch": 0.34,
"learning_rate": 0.00033480441323971915,
"loss": 3.0287,
"theoretical_loss": 3.6126644747266097,
"tokens_seen": 1112670208
},
{
"epoch": 0.34,
"learning_rate": 0.0003347943831494484,
"loss": 3.0317,
"theoretical_loss": 3.6126446230484577,
"tokens_seen": 1112735744
},
{
"epoch": 0.34,
"learning_rate": 0.0003347843530591775,
"loss": 2.7996,
"theoretical_loss": 3.612624772866813,
"tokens_seen": 1112801280
},
{
"epoch": 0.34,
"learning_rate": 0.00033477432296890675,
"loss": 2.968,
"theoretical_loss": 3.6126049241814746,
"tokens_seen": 1112866816
},
{
"epoch": 0.34,
"learning_rate": 0.00033476429287863593,
"loss": 2.8833,
"theoretical_loss": 3.6125850769922407,
"tokens_seen": 1112932352
},
{
"epoch": 0.34,
"learning_rate": 0.0003347542627883651,
"loss": 3.053,
"theoretical_loss": 3.612565231298911,
"tokens_seen": 1112997888
},
{
"epoch": 0.34,
"learning_rate": 0.0003347442326980943,
"loss": 2.8087,
"theoretical_loss": 3.612545387101285,
"tokens_seen": 1113063424
},
{
"epoch": 0.34,
"learning_rate": 0.00033473420260782353,
"loss": 2.6987,
"theoretical_loss": 3.6125255443991615,
"tokens_seen": 1113128960
},
{
"epoch": 0.34,
"learning_rate": 0.00033472417251755265,
"loss": 2.9778,
"theoretical_loss": 3.61250570319234,
"tokens_seen": 1113194496
},
{
"epoch": 0.34,
"learning_rate": 0.0003347141424272819,
"loss": 2.8556,
"theoretical_loss": 3.6124858634806194,
"tokens_seen": 1113260032
},
{
"epoch": 0.34,
"learning_rate": 0.000334704112337011,
"loss": 2.6581,
"theoretical_loss": 3.612466025263799,
"tokens_seen": 1113325568
},
{
"epoch": 0.34,
"learning_rate": 0.00033469408224674025,
"loss": 2.971,
"theoretical_loss": 3.612446188541679,
"tokens_seen": 1113391104
},
{
"epoch": 0.34,
"learning_rate": 0.00033468405215646943,
"loss": 2.9822,
"theoretical_loss": 3.612426353314058,
"tokens_seen": 1113456640
},
{
"epoch": 0.34,
"learning_rate": 0.0003346740220661986,
"loss": 2.883,
"theoretical_loss": 3.6124065195807358,
"tokens_seen": 1113522176
},
{
"epoch": 0.34,
"learning_rate": 0.0003346639919759278,
"loss": 2.9221,
"theoretical_loss": 3.6123866873415116,
"tokens_seen": 1113587712
},
{
"epoch": 0.34,
"learning_rate": 0.000334653961885657,
"loss": 2.7282,
"theoretical_loss": 3.612366856596185,
"tokens_seen": 1113653248
},
{
"epoch": 0.34,
"learning_rate": 0.00033464393179538616,
"loss": 2.6266,
"theoretical_loss": 3.612347027344556,
"tokens_seen": 1113718784
},
{
"epoch": 0.34,
"learning_rate": 0.0003346339017051154,
"loss": 3.0552,
"theoretical_loss": 3.6123271995864235,
"tokens_seen": 1113784320
},
{
"epoch": 0.34,
"learning_rate": 0.0003346238716148445,
"loss": 2.8499,
"theoretical_loss": 3.6123073733215882,
"tokens_seen": 1113849856
},
{
"epoch": 0.34,
"learning_rate": 0.00033461384152457376,
"loss": 2.7885,
"theoretical_loss": 3.6122875485498485,
"tokens_seen": 1113915392
},
{
"epoch": 0.34,
"learning_rate": 0.0003346038114343029,
"loss": 2.7566,
"theoretical_loss": 3.612267725271005,
"tokens_seen": 1113980928
},
{
"epoch": 0.34,
"learning_rate": 0.0003345937813440321,
"loss": 2.7659,
"theoretical_loss": 3.6122479034848576,
"tokens_seen": 1114046464
},
{
"epoch": 0.34,
"objective/train/docs_used": 634689,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7385952472686768,
"objective/train/theoretical_loss": 3.612228083191205,
"objective/train/tokens_used": 1134572000,
"theoretical_loss": 3.612228083191205,
"tokens_seen": 1114112000
},
{
"epoch": 0.34,
"learning_rate": 0.0003345837512537613,
"loss": 2.6877,
"theoretical_loss": 3.612228083191205,
"tokens_seen": 1114112000
},
{
"epoch": 0.34,
"learning_rate": 0.0003345737211634905,
"loss": 2.804,
"theoretical_loss": 3.6122082643898485,
"tokens_seen": 1114177536
},
{
"epoch": 0.34,
"learning_rate": 0.00033456369107321966,
"loss": 2.8803,
"theoretical_loss": 3.6121884470805865,
"tokens_seen": 1114243072
},
{
"epoch": 0.34,
"learning_rate": 0.0003345536609829489,
"loss": 3.01,
"theoretical_loss": 3.61216863126322,
"tokens_seen": 1114308608
},
{
"epoch": 0.34,
"learning_rate": 0.000334543630892678,
"loss": 2.5688,
"theoretical_loss": 3.612148816937548,
"tokens_seen": 1114374144
},
{
"epoch": 0.34,
"learning_rate": 0.00033453360080240726,
"loss": 2.8065,
"theoretical_loss": 3.6121290041033722,
"tokens_seen": 1114439680
},
{
"epoch": 0.34,
"learning_rate": 0.0003345235707121364,
"loss": 2.813,
"theoretical_loss": 3.6121091927604905,
"tokens_seen": 1114505216
},
{
"epoch": 0.34,
"learning_rate": 0.0003345135406218656,
"loss": 2.823,
"theoretical_loss": 3.612089382908705,
"tokens_seen": 1114570752
},
{
"epoch": 0.34,
"learning_rate": 0.0003345035105315948,
"loss": 2.8424,
"theoretical_loss": 3.6120695745478137,
"tokens_seen": 1114636288
},
{
"epoch": 0.34,
"learning_rate": 0.000334493480441324,
"loss": 2.6189,
"theoretical_loss": 3.612049767677618,
"tokens_seen": 1114701824
},
{
"epoch": 0.34,
"learning_rate": 0.00033448345035105316,
"loss": 2.9676,
"theoretical_loss": 3.6120299622979184,
"tokens_seen": 1114767360
},
{
"epoch": 0.34,
"learning_rate": 0.00033447342026078235,
"loss": 2.7569,
"theoretical_loss": 3.6120101584085145,
"tokens_seen": 1114832896
},
{
"epoch": 0.34,
"learning_rate": 0.0003344633901705115,
"loss": 2.9089,
"theoretical_loss": 3.6119903560092066,
"tokens_seen": 1114898432
},
{
"epoch": 0.34,
"learning_rate": 0.00033445336008024076,
"loss": 2.9464,
"theoretical_loss": 3.611970555099795,
"tokens_seen": 1114963968
},
{
"epoch": 0.34,
"learning_rate": 0.0003344433299899699,
"loss": 2.7356,
"theoretical_loss": 3.6119507556800805,
"tokens_seen": 1115029504
},
{
"epoch": 0.34,
"learning_rate": 0.0003344332998996991,
"loss": 2.4431,
"theoretical_loss": 3.611930957749863,
"tokens_seen": 1115095040
},
{
"epoch": 0.34,
"learning_rate": 0.00033442326980942825,
"loss": 2.7838,
"theoretical_loss": 3.611911161308943,
"tokens_seen": 1115160576
},
{
"epoch": 0.34,
"learning_rate": 0.0003344132397191575,
"loss": 2.8567,
"theoretical_loss": 3.6118913663571206,
"tokens_seen": 1115226112
},
{
"epoch": 0.34,
"learning_rate": 0.00033440320962888667,
"loss": 2.8515,
"theoretical_loss": 3.6118715728941972,
"tokens_seen": 1115291648
},
{
"epoch": 0.34,
"learning_rate": 0.00033439317953861585,
"loss": 2.7556,
"theoretical_loss": 3.611851780919973,
"tokens_seen": 1115357184
},
{
"epoch": 0.34,
"learning_rate": 0.00033438314944834503,
"loss": 2.834,
"theoretical_loss": 3.611831990434248,
"tokens_seen": 1115422720
},
{
"epoch": 0.34,
"learning_rate": 0.00033437311935807427,
"loss": 2.7653,
"theoretical_loss": 3.611812201436823,
"tokens_seen": 1115488256
},
{
"epoch": 0.34,
"learning_rate": 0.0003343630892678034,
"loss": 2.9198,
"theoretical_loss": 3.611792413927499,
"tokens_seen": 1115553792
},
{
"epoch": 0.34,
"learning_rate": 0.00033435305917753263,
"loss": 2.8045,
"theoretical_loss": 3.6117726279060767,
"tokens_seen": 1115619328
},
{
"epoch": 0.34,
"learning_rate": 0.00033434302908726175,
"loss": 2.976,
"theoretical_loss": 3.611752843372356,
"tokens_seen": 1115684864
},
{
"epoch": 0.34,
"objective/train/docs_used": 636068,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5940232276916504,
"objective/train/theoretical_loss": 3.611733060326139,
"objective/train/tokens_used": 1136210400,
"theoretical_loss": 3.611733060326139,
"tokens_seen": 1115750400
},
{
"epoch": 0.34,
"learning_rate": 0.000334332998996991,
"loss": 2.6831,
"theoretical_loss": 3.611733060326139,
"tokens_seen": 1115750400
},
{
"epoch": 0.34,
"learning_rate": 0.00033432296890672017,
"loss": 2.9727,
"theoretical_loss": 3.6117132787672257,
"tokens_seen": 1115815936
},
{
"epoch": 0.34,
"learning_rate": 0.00033431293881644935,
"loss": 2.8706,
"theoretical_loss": 3.611693498695417,
"tokens_seen": 1115881472
},
{
"epoch": 0.34,
"learning_rate": 0.00033430290872617853,
"loss": 2.6261,
"theoretical_loss": 3.6116737201105136,
"tokens_seen": 1115947008
},
{
"epoch": 0.34,
"learning_rate": 0.0003342928786359077,
"loss": 2.789,
"theoretical_loss": 3.6116539430123167,
"tokens_seen": 1116012544
},
{
"epoch": 0.34,
"learning_rate": 0.0003342828485456369,
"loss": 2.8224,
"theoretical_loss": 3.611634167400627,
"tokens_seen": 1116078080
},
{
"epoch": 0.34,
"learning_rate": 0.00033427281845536613,
"loss": 2.8912,
"theoretical_loss": 3.6116143932752465,
"tokens_seen": 1116143616
},
{
"epoch": 0.34,
"learning_rate": 0.00033426278836509526,
"loss": 3.0675,
"theoretical_loss": 3.611594620635975,
"tokens_seen": 1116209152
},
{
"epoch": 0.34,
"learning_rate": 0.0003342527582748245,
"loss": 2.8242,
"theoretical_loss": 3.6115748494826136,
"tokens_seen": 1116274688
},
{
"epoch": 0.34,
"learning_rate": 0.0003342427281845536,
"loss": 2.6982,
"theoretical_loss": 3.6115550798149645,
"tokens_seen": 1116340224
},
{
"epoch": 0.34,
"learning_rate": 0.00033423269809428285,
"loss": 3.0192,
"theoretical_loss": 3.6115353116328275,
"tokens_seen": 1116405760
},
{
"epoch": 0.34,
"learning_rate": 0.00033422266800401204,
"loss": 2.9734,
"theoretical_loss": 3.6115155449360046,
"tokens_seen": 1116471296
},
{
"epoch": 0.34,
"learning_rate": 0.0003342126379137412,
"loss": 2.7111,
"theoretical_loss": 3.611495779724297,
"tokens_seen": 1116536832
},
{
"epoch": 0.34,
"learning_rate": 0.0003342026078234704,
"loss": 2.7085,
"theoretical_loss": 3.6114760159975052,
"tokens_seen": 1116602368
},
{
"epoch": 0.34,
"learning_rate": 0.00033419257773319963,
"loss": 2.8904,
"theoretical_loss": 3.6114562537554318,
"tokens_seen": 1116667904
},
{
"epoch": 0.34,
"learning_rate": 0.00033418254764292876,
"loss": 2.9897,
"theoretical_loss": 3.611436492997877,
"tokens_seen": 1116733440
},
{
"epoch": 0.34,
"learning_rate": 0.000334172517552658,
"loss": 2.6401,
"theoretical_loss": 3.611416733724643,
"tokens_seen": 1116798976
},
{
"epoch": 0.34,
"learning_rate": 0.0003341624874623871,
"loss": 2.9192,
"theoretical_loss": 3.6113969759355307,
"tokens_seen": 1116864512
},
{
"epoch": 0.34,
"learning_rate": 0.00033415245737211636,
"loss": 2.8175,
"theoretical_loss": 3.6113772196303415,
"tokens_seen": 1116930048
},
{
"epoch": 0.34,
"learning_rate": 0.00033414242728184554,
"loss": 2.7708,
"theoretical_loss": 3.6113574648088766,
"tokens_seen": 1116995584
},
{
"epoch": 0.34,
"learning_rate": 0.0003341323971915747,
"loss": 2.8277,
"theoretical_loss": 3.611337711470939,
"tokens_seen": 1117061120
},
{
"epoch": 0.34,
"learning_rate": 0.0003341223671013039,
"loss": 2.9322,
"theoretical_loss": 3.6113179596163283,
"tokens_seen": 1117126656
},
{
"epoch": 0.34,
"learning_rate": 0.0003341123370110331,
"loss": 2.7288,
"theoretical_loss": 3.611298209244848,
"tokens_seen": 1117192192
},
{
"epoch": 0.34,
"learning_rate": 0.00033410230692076226,
"loss": 2.8705,
"theoretical_loss": 3.611278460356298,
"tokens_seen": 1117257728
},
{
"epoch": 0.34,
"learning_rate": 0.0003340922768304915,
"loss": 2.8835,
"theoretical_loss": 3.611258712950481,
"tokens_seen": 1117323264
},
{
"epoch": 0.34,
"objective/train/docs_used": 636687,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.152287721633911,
"objective/train/theoretical_loss": 3.611238967027199,
"objective/train/tokens_used": 1137848800,
"theoretical_loss": 3.611238967027199,
"tokens_seen": 1117388800
},
{
"epoch": 0.34,
"learning_rate": 0.0003340822467402207,
"loss": 2.5869,
"theoretical_loss": 3.611238967027199,
"tokens_seen": 1117388800
},
{
"epoch": 0.34,
"learning_rate": 0.00033407221664994986,
"loss": 2.687,
"theoretical_loss": 3.6112192225862527,
"tokens_seen": 1117454336
},
{
"epoch": 0.34,
"learning_rate": 0.0003340621865596791,
"loss": 2.7419,
"theoretical_loss": 3.6111994796274445,
"tokens_seen": 1117519872
},
{
"epoch": 0.34,
"learning_rate": 0.0003340521564694082,
"loss": 2.7018,
"theoretical_loss": 3.611179738150576,
"tokens_seen": 1117585408
},
{
"epoch": 0.34,
"learning_rate": 0.00033404212637913746,
"loss": 2.7725,
"theoretical_loss": 3.6111599981554496,
"tokens_seen": 1117650944
},
{
"epoch": 0.34,
"learning_rate": 0.0003340320962888666,
"loss": 2.9505,
"theoretical_loss": 3.611140259641867,
"tokens_seen": 1117716480
},
{
"epoch": 0.34,
"learning_rate": 0.0003340220661985958,
"loss": 2.9446,
"theoretical_loss": 3.61112052260963,
"tokens_seen": 1117782016
},
{
"epoch": 0.34,
"learning_rate": 0.000334012036108325,
"loss": 2.6528,
"theoretical_loss": 3.6111007870585405,
"tokens_seen": 1117847552
},
{
"epoch": 0.34,
"learning_rate": 0.0003340020060180542,
"loss": 2.8242,
"theoretical_loss": 3.6110810529884003,
"tokens_seen": 1117913088
},
{
"epoch": 0.34,
"learning_rate": 0.00033399197592778336,
"loss": 2.6812,
"theoretical_loss": 3.6110613203990116,
"tokens_seen": 1117978624
},
{
"epoch": 0.34,
"learning_rate": 0.00033398194583751255,
"loss": 2.857,
"theoretical_loss": 3.6110415892901773,
"tokens_seen": 1118044160
},
{
"epoch": 0.34,
"learning_rate": 0.0003339719157472417,
"loss": 2.8018,
"theoretical_loss": 3.6110218596616988,
"tokens_seen": 1118109696
},
{
"epoch": 0.34,
"learning_rate": 0.00033396188565697096,
"loss": 2.6773,
"theoretical_loss": 3.611002131513378,
"tokens_seen": 1118175232
},
{
"epoch": 0.34,
"learning_rate": 0.0003339518555667001,
"loss": 2.6636,
"theoretical_loss": 3.610982404845018,
"tokens_seen": 1118240768
},
{
"epoch": 0.34,
"learning_rate": 0.0003339418254764293,
"loss": 2.9554,
"theoretical_loss": 3.6109626796564203,
"tokens_seen": 1118306304
},
{
"epoch": 0.34,
"learning_rate": 0.00033393179538615845,
"loss": 2.7346,
"theoretical_loss": 3.6109429559473876,
"tokens_seen": 1118371840
},
{
"epoch": 0.34,
"learning_rate": 0.0003339217652958877,
"loss": 2.6578,
"theoretical_loss": 3.6109232337177217,
"tokens_seen": 1118437376
},
{
"epoch": 0.34,
"learning_rate": 0.00033391173520561687,
"loss": 2.7436,
"theoretical_loss": 3.6109035129672256,
"tokens_seen": 1118502912
},
{
"epoch": 0.34,
"learning_rate": 0.00033390170511534605,
"loss": 2.7963,
"theoretical_loss": 3.6108837936957014,
"tokens_seen": 1118568448
},
{
"epoch": 0.34,
"learning_rate": 0.00033389167502507523,
"loss": 2.8952,
"theoretical_loss": 3.6108640759029518,
"tokens_seen": 1118633984
},
{
"epoch": 0.34,
"learning_rate": 0.00033388164493480447,
"loss": 2.9821,
"theoretical_loss": 3.6108443595887785,
"tokens_seen": 1118699520
},
{
"epoch": 0.34,
"learning_rate": 0.0003338716148445336,
"loss": 2.7333,
"theoretical_loss": 3.6108246447529853,
"tokens_seen": 1118765056
},
{
"epoch": 0.34,
"learning_rate": 0.00033386158475426283,
"loss": 2.7342,
"theoretical_loss": 3.610804931395373,
"tokens_seen": 1118830592
},
{
"epoch": 0.34,
"learning_rate": 0.00033385155466399195,
"loss": 2.8163,
"theoretical_loss": 3.6107852195157464,
"tokens_seen": 1118896128
},
{
"epoch": 0.34,
"learning_rate": 0.0003338415245737212,
"loss": 2.6105,
"theoretical_loss": 3.6107655091139064,
"tokens_seen": 1118961664
},
{
"epoch": 0.34,
"objective/train/docs_used": 638066,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.664874792098999,
"objective/train/theoretical_loss": 3.6107458001896564,
"objective/train/tokens_used": 1139487200,
"theoretical_loss": 3.6107458001896564,
"tokens_seen": 1119027200
},
{
"epoch": 0.34,
"learning_rate": 0.00033383149448345037,
"loss": 2.7851,
"theoretical_loss": 3.6107458001896564,
"tokens_seen": 1119027200
},
{
"epoch": 0.34,
"learning_rate": 0.00033382146439317955,
"loss": 2.815,
"theoretical_loss": 3.6107260927427984,
"tokens_seen": 1119092736
},
{
"epoch": 0.34,
"learning_rate": 0.00033381143430290873,
"loss": 2.6369,
"theoretical_loss": 3.610706386773136,
"tokens_seen": 1119158272
},
{
"epoch": 0.34,
"learning_rate": 0.0003338014042126379,
"loss": 2.7499,
"theoretical_loss": 3.610686682280472,
"tokens_seen": 1119223808
},
{
"epoch": 0.34,
"learning_rate": 0.0003337913741223671,
"loss": 2.7677,
"theoretical_loss": 3.6106669792646087,
"tokens_seen": 1119289344
},
{
"epoch": 0.34,
"learning_rate": 0.00033378134403209633,
"loss": 2.6087,
"theoretical_loss": 3.610647277725349,
"tokens_seen": 1119354880
},
{
"epoch": 0.34,
"learning_rate": 0.00033377131394182546,
"loss": 2.7966,
"theoretical_loss": 3.6106275776624956,
"tokens_seen": 1119420416
},
{
"epoch": 0.34,
"learning_rate": 0.0003337612838515547,
"loss": 2.8476,
"theoretical_loss": 3.610607879075853,
"tokens_seen": 1119485952
},
{
"epoch": 0.34,
"learning_rate": 0.0003337512537612838,
"loss": 2.8756,
"theoretical_loss": 3.6105881819652215,
"tokens_seen": 1119551488
},
{
"epoch": 0.34,
"learning_rate": 0.00033374122367101306,
"loss": 2.9109,
"theoretical_loss": 3.6105684863304064,
"tokens_seen": 1119617024
},
{
"epoch": 0.34,
"learning_rate": 0.00033373119358074224,
"loss": 2.8659,
"theoretical_loss": 3.6105487921712096,
"tokens_seen": 1119682560
},
{
"epoch": 0.34,
"learning_rate": 0.0003337211634904714,
"loss": 2.9085,
"theoretical_loss": 3.610529099487435,
"tokens_seen": 1119748096
},
{
"epoch": 0.34,
"learning_rate": 0.0003337111334002006,
"loss": 2.9344,
"theoretical_loss": 3.6105094082788844,
"tokens_seen": 1119813632
},
{
"epoch": 0.34,
"learning_rate": 0.00033370110330992983,
"loss": 2.72,
"theoretical_loss": 3.610489718545362,
"tokens_seen": 1119879168
},
{
"epoch": 0.34,
"learning_rate": 0.00033369107321965896,
"loss": 2.7302,
"theoretical_loss": 3.6104700302866712,
"tokens_seen": 1119944704
},
{
"epoch": 0.34,
"learning_rate": 0.0003336810431293882,
"loss": 2.6103,
"theoretical_loss": 3.610450343502614,
"tokens_seen": 1120010240
},
{
"epoch": 0.34,
"learning_rate": 0.0003336710130391173,
"loss": 2.9552,
"theoretical_loss": 3.6104306581929952,
"tokens_seen": 1120075776
},
{
"epoch": 0.34,
"learning_rate": 0.00033366098294884656,
"loss": 2.6791,
"theoretical_loss": 3.6104109743576176,
"tokens_seen": 1120141312
},
{
"epoch": 0.34,
"learning_rate": 0.00033365095285857574,
"loss": 2.6207,
"theoretical_loss": 3.610391291996284,
"tokens_seen": 1120206848
},
{
"epoch": 0.34,
"learning_rate": 0.0003336409227683049,
"loss": 2.8663,
"theoretical_loss": 3.610371611108798,
"tokens_seen": 1120272384
},
{
"epoch": 0.34,
"learning_rate": 0.0003336308926780341,
"loss": 3.2142,
"theoretical_loss": 3.6103519316949626,
"tokens_seen": 1120337920
},
{
"epoch": 0.34,
"learning_rate": 0.0003336208625877633,
"loss": 2.7828,
"theoretical_loss": 3.6103322537545828,
"tokens_seen": 1120403456
},
{
"epoch": 0.34,
"learning_rate": 0.00033361083249749246,
"loss": 2.8228,
"theoretical_loss": 3.610312577287461,
"tokens_seen": 1120468992
},
{
"epoch": 0.34,
"learning_rate": 0.0003336008024072217,
"loss": 2.832,
"theoretical_loss": 3.6102929022934003,
"tokens_seen": 1120534528
},
{
"epoch": 0.34,
"learning_rate": 0.0003335907723169508,
"loss": 2.7825,
"theoretical_loss": 3.610273228772205,
"tokens_seen": 1120600064
},
{
"epoch": 0.34,
"objective/train/docs_used": 638756,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.506070852279663,
"objective/train/theoretical_loss": 3.610253556723679,
"objective/train/tokens_used": 1141125600,
"theoretical_loss": 3.610253556723679,
"tokens_seen": 1120665600
},
{
"epoch": 0.34,
"learning_rate": 0.00033358074222668006,
"loss": 2.8011,
"theoretical_loss": 3.610253556723679,
"tokens_seen": 1120665600
},
{
"epoch": 0.34,
"learning_rate": 0.00033357071213640924,
"loss": 2.7042,
"theoretical_loss": 3.610233886147625,
"tokens_seen": 1120731136
},
{
"epoch": 0.34,
"learning_rate": 0.0003335606820461384,
"loss": 2.8496,
"theoretical_loss": 3.6102142170438474,
"tokens_seen": 1120796672
},
{
"epoch": 0.34,
"learning_rate": 0.0003335506519558676,
"loss": 2.5992,
"theoretical_loss": 3.61019454941215,
"tokens_seen": 1120862208
},
{
"epoch": 0.34,
"learning_rate": 0.0003335406218655968,
"loss": 2.6919,
"theoretical_loss": 3.610174883252336,
"tokens_seen": 1120927744
},
{
"epoch": 0.34,
"learning_rate": 0.00033353059177532597,
"loss": 2.8277,
"theoretical_loss": 3.610155218564209,
"tokens_seen": 1120993280
},
{
"epoch": 0.34,
"learning_rate": 0.0003335205616850552,
"loss": 2.6509,
"theoretical_loss": 3.610135555347574,
"tokens_seen": 1121058816
},
{
"epoch": 0.34,
"learning_rate": 0.00033351053159478433,
"loss": 2.6488,
"theoretical_loss": 3.610115893602234,
"tokens_seen": 1121124352
},
{
"epoch": 0.34,
"learning_rate": 0.00033350050150451356,
"loss": 2.4415,
"theoretical_loss": 3.610096233327993,
"tokens_seen": 1121189888
},
{
"epoch": 0.34,
"learning_rate": 0.0003334904714142427,
"loss": 2.8967,
"theoretical_loss": 3.6100765745246552,
"tokens_seen": 1121255424
},
{
"epoch": 0.34,
"learning_rate": 0.0003334804413239719,
"loss": 2.9756,
"theoretical_loss": 3.6100569171920247,
"tokens_seen": 1121320960
},
{
"epoch": 0.34,
"learning_rate": 0.0003334704112337011,
"loss": 2.7517,
"theoretical_loss": 3.6100372613299045,
"tokens_seen": 1121386496
},
{
"epoch": 0.34,
"learning_rate": 0.0003334603811434303,
"loss": 2.9856,
"theoretical_loss": 3.6100176069381003,
"tokens_seen": 1121452032
},
{
"epoch": 0.34,
"learning_rate": 0.00033345035105315947,
"loss": 2.8534,
"theoretical_loss": 3.609997954016415,
"tokens_seen": 1121517568
},
{
"epoch": 0.34,
"learning_rate": 0.00033344032096288865,
"loss": 2.8062,
"theoretical_loss": 3.6099783025646532,
"tokens_seen": 1121583104
},
{
"epoch": 0.34,
"learning_rate": 0.00033343029087261783,
"loss": 2.7804,
"theoretical_loss": 3.609958652582619,
"tokens_seen": 1121648640
},
{
"epoch": 0.34,
"learning_rate": 0.00033342026078234707,
"loss": 2.7629,
"theoretical_loss": 3.6099390040701165,
"tokens_seen": 1121714176
},
{
"epoch": 0.34,
"learning_rate": 0.0003334102306920762,
"loss": 2.6065,
"theoretical_loss": 3.60991935702695,
"tokens_seen": 1121779712
},
{
"epoch": 0.34,
"learning_rate": 0.00033340020060180543,
"loss": 2.6849,
"theoretical_loss": 3.6098997114529237,
"tokens_seen": 1121845248
},
{
"epoch": 0.34,
"learning_rate": 0.0003333901705115346,
"loss": 2.5853,
"theoretical_loss": 3.609880067347842,
"tokens_seen": 1121910784
},
{
"epoch": 0.34,
"learning_rate": 0.0003333801404212638,
"loss": 2.829,
"theoretical_loss": 3.60986042471151,
"tokens_seen": 1121976320
},
{
"epoch": 0.34,
"learning_rate": 0.000333370110330993,
"loss": 2.8643,
"theoretical_loss": 3.6098407835437305,
"tokens_seen": 1122041856
},
{
"epoch": 0.34,
"learning_rate": 0.00033336008024072215,
"loss": 2.6548,
"theoretical_loss": 3.609821143844309,
"tokens_seen": 1122107392
},
{
"epoch": 0.34,
"learning_rate": 0.00033335005015045134,
"loss": 2.5597,
"theoretical_loss": 3.6098015056130506,
"tokens_seen": 1122172928
},
{
"epoch": 0.34,
"learning_rate": 0.00033334002006018057,
"loss": 2.7587,
"theoretical_loss": 3.6097818688497583,
"tokens_seen": 1122238464
},
{
"epoch": 0.34,
"objective/train/docs_used": 639983,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1040778160095215,
"objective/train/theoretical_loss": 3.609762233554238,
"objective/train/tokens_used": 1142764000,
"theoretical_loss": 3.609762233554238,
"tokens_seen": 1122304000
},
{
"epoch": 0.34,
"learning_rate": 0.00033332998996990975,
"loss": 2.9086,
"theoretical_loss": 3.609762233554238,
"tokens_seen": 1122304000
},
{
"epoch": 0.34,
"learning_rate": 0.00033331995987963893,
"loss": 2.8331,
"theoretical_loss": 3.6097425997262933,
"tokens_seen": 1122369536
},
{
"epoch": 0.34,
"learning_rate": 0.0003333099297893681,
"loss": 2.7353,
"theoretical_loss": 3.6097229673657294,
"tokens_seen": 1122435072
},
{
"epoch": 0.34,
"learning_rate": 0.0003332998996990973,
"loss": 2.728,
"theoretical_loss": 3.6097033364723505,
"tokens_seen": 1122500608
},
{
"epoch": 0.34,
"learning_rate": 0.00033328986960882653,
"loss": 2.8081,
"theoretical_loss": 3.6096837070459618,
"tokens_seen": 1122566144
},
{
"epoch": 0.34,
"learning_rate": 0.00033327983951855566,
"loss": 2.8237,
"theoretical_loss": 3.609664079086368,
"tokens_seen": 1122631680
},
{
"epoch": 0.34,
"learning_rate": 0.0003332698094282849,
"loss": 2.7421,
"theoretical_loss": 3.6096444525933737,
"tokens_seen": 1122697216
},
{
"epoch": 0.34,
"learning_rate": 0.000333259779338014,
"loss": 2.7613,
"theoretical_loss": 3.6096248275667833,
"tokens_seen": 1122762752
},
{
"epoch": 0.34,
"learning_rate": 0.00033324974924774326,
"loss": 2.6767,
"theoretical_loss": 3.6096052040064026,
"tokens_seen": 1122828288
},
{
"epoch": 0.34,
"learning_rate": 0.00033323971915747244,
"loss": 2.6183,
"theoretical_loss": 3.609585581912036,
"tokens_seen": 1122893824
},
{
"epoch": 0.34,
"learning_rate": 0.0003332296890672016,
"loss": 2.832,
"theoretical_loss": 3.6095659612834883,
"tokens_seen": 1122959360
},
{
"epoch": 0.34,
"learning_rate": 0.0003332196589769308,
"loss": 2.7969,
"theoretical_loss": 3.6095463421205642,
"tokens_seen": 1123024896
},
{
"epoch": 0.34,
"learning_rate": 0.00033320962888666003,
"loss": 2.8418,
"theoretical_loss": 3.609526724423069,
"tokens_seen": 1123090432
},
{
"epoch": 0.34,
"learning_rate": 0.00033319959879638916,
"loss": 2.6971,
"theoretical_loss": 3.609507108190808,
"tokens_seen": 1123155968
},
{
"epoch": 0.34,
"learning_rate": 0.0003331895687061184,
"loss": 2.7487,
"theoretical_loss": 3.609487493423586,
"tokens_seen": 1123221504
},
{
"epoch": 0.34,
"learning_rate": 0.0003331795386158475,
"loss": 3.2122,
"theoretical_loss": 3.609467880121209,
"tokens_seen": 1123287040
},
{
"epoch": 0.34,
"learning_rate": 0.00033316950852557676,
"loss": 2.7219,
"theoretical_loss": 3.6094482682834803,
"tokens_seen": 1123352576
},
{
"epoch": 0.34,
"learning_rate": 0.00033315947843530594,
"loss": 2.7099,
"theoretical_loss": 3.6094286579102066,
"tokens_seen": 1123418112
},
{
"epoch": 0.34,
"learning_rate": 0.0003331494483450351,
"loss": 2.7935,
"theoretical_loss": 3.6094090490011927,
"tokens_seen": 1123483648
},
{
"epoch": 0.34,
"learning_rate": 0.0003331394182547643,
"loss": 2.5765,
"theoretical_loss": 3.6093894415562437,
"tokens_seen": 1123549184
},
{
"epoch": 0.34,
"learning_rate": 0.0003331293881644935,
"loss": 2.8254,
"theoretical_loss": 3.6093698355751647,
"tokens_seen": 1123614720
},
{
"epoch": 0.34,
"learning_rate": 0.00033311935807422266,
"loss": 2.7878,
"theoretical_loss": 3.609350231057762,
"tokens_seen": 1123680256
},
{
"epoch": 0.34,
"learning_rate": 0.0003331093279839519,
"loss": 2.5508,
"theoretical_loss": 3.60933062800384,
"tokens_seen": 1123745792
},
{
"epoch": 0.34,
"learning_rate": 0.000333099297893681,
"loss": 2.7028,
"theoretical_loss": 3.609311026413204,
"tokens_seen": 1123811328
},
{
"epoch": 0.34,
"learning_rate": 0.00033308926780341026,
"loss": 2.8765,
"theoretical_loss": 3.609291426285661,
"tokens_seen": 1123876864
},
{
"epoch": 0.34,
"objective/train/docs_used": 640598,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1615970134735107,
"objective/train/theoretical_loss": 3.609271827621014,
"objective/train/tokens_used": 1144402400,
"theoretical_loss": 3.609271827621014,
"tokens_seen": 1123942400
},
{
"epoch": 0.34,
"learning_rate": 0.00033307923771313944,
"loss": 2.8542,
"theoretical_loss": 3.609271827621014,
"tokens_seen": 1123942400
},
{
"epoch": 0.34,
"learning_rate": 0.0003330692076228686,
"loss": 2.6216,
"theoretical_loss": 3.609252230419071,
"tokens_seen": 1124007936
},
{
"epoch": 0.34,
"learning_rate": 0.0003330591775325978,
"loss": 2.7696,
"theoretical_loss": 3.6092326346796355,
"tokens_seen": 1124073472
},
{
"epoch": 0.34,
"learning_rate": 0.000333049147442327,
"loss": 2.8678,
"theoretical_loss": 3.6092130404025147,
"tokens_seen": 1124139008
},
{
"epoch": 0.34,
"learning_rate": 0.00033303911735205617,
"loss": 2.9717,
"theoretical_loss": 3.6091934475875136,
"tokens_seen": 1124204544
},
{
"epoch": 0.34,
"learning_rate": 0.0003330290872617854,
"loss": 2.798,
"theoretical_loss": 3.6091738562344378,
"tokens_seen": 1124270080
},
{
"epoch": 0.34,
"learning_rate": 0.00033301905717151453,
"loss": 2.7759,
"theoretical_loss": 3.6091542663430927,
"tokens_seen": 1124335616
},
{
"epoch": 0.34,
"learning_rate": 0.00033300902708124376,
"loss": 2.8993,
"theoretical_loss": 3.6091346779132847,
"tokens_seen": 1124401152
},
{
"epoch": 0.34,
"learning_rate": 0.0003329989969909729,
"loss": 2.7992,
"theoretical_loss": 3.6091150909448193,
"tokens_seen": 1124466688
},
{
"epoch": 0.34,
"learning_rate": 0.0003329889669007021,
"loss": 2.7327,
"theoretical_loss": 3.6090955054375025,
"tokens_seen": 1124532224
},
{
"epoch": 0.34,
"learning_rate": 0.0003329789368104313,
"loss": 2.6884,
"theoretical_loss": 3.6090759213911396,
"tokens_seen": 1124597760
},
{
"epoch": 0.34,
"learning_rate": 0.0003329689067201605,
"loss": 2.8175,
"theoretical_loss": 3.6090563388055372,
"tokens_seen": 1124663296
},
{
"epoch": 0.34,
"learning_rate": 0.00033295887662988967,
"loss": 2.7456,
"theoretical_loss": 3.6090367576805007,
"tokens_seen": 1124728832
},
{
"epoch": 0.34,
"learning_rate": 0.00033294884653961885,
"loss": 2.6801,
"theoretical_loss": 3.6090171780158364,
"tokens_seen": 1124794368
},
{
"epoch": 0.34,
"learning_rate": 0.00033293881644934803,
"loss": 2.6192,
"theoretical_loss": 3.60899759981135,
"tokens_seen": 1124859904
},
{
"epoch": 0.34,
"learning_rate": 0.00033292878635907727,
"loss": 2.8495,
"theoretical_loss": 3.6089780230668476,
"tokens_seen": 1124925440
},
{
"epoch": 0.34,
"learning_rate": 0.0003329187562688064,
"loss": 2.7159,
"theoretical_loss": 3.6089584477821357,
"tokens_seen": 1124990976
},
{
"epoch": 0.34,
"learning_rate": 0.00033290872617853563,
"loss": 2.8856,
"theoretical_loss": 3.60893887395702,
"tokens_seen": 1125056512
},
{
"epoch": 0.34,
"learning_rate": 0.0003328986960882648,
"loss": 2.5557,
"theoretical_loss": 3.608919301591307,
"tokens_seen": 1125122048
},
{
"epoch": 0.34,
"learning_rate": 0.000332888665997994,
"loss": 2.8093,
"theoretical_loss": 3.608899730684802,
"tokens_seen": 1125187584
},
{
"epoch": 0.34,
"learning_rate": 0.0003328786359077232,
"loss": 2.8042,
"theoretical_loss": 3.6088801612373125,
"tokens_seen": 1125253120
},
{
"epoch": 0.34,
"learning_rate": 0.00033286860581745235,
"loss": 2.7134,
"theoretical_loss": 3.608860593248644,
"tokens_seen": 1125318656
},
{
"epoch": 0.34,
"learning_rate": 0.00033285857572718154,
"loss": 2.7081,
"theoretical_loss": 3.608841026718603,
"tokens_seen": 1125384192
},
{
"epoch": 0.34,
"learning_rate": 0.00033284854563691077,
"loss": 2.559,
"theoretical_loss": 3.608821461646995,
"tokens_seen": 1125449728
},
{
"epoch": 0.34,
"learning_rate": 0.0003328385155466399,
"loss": 2.854,
"theoretical_loss": 3.6088018980336285,
"tokens_seen": 1125515264
},
{
"epoch": 0.34,
"objective/train/docs_used": 641290,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.566493511199951,
"objective/train/theoretical_loss": 3.608782335878308,
"objective/train/tokens_used": 1146040800,
"theoretical_loss": 3.608782335878308,
"tokens_seen": 1125580800
},
{
"epoch": 0.34,
"learning_rate": 0.00033282848545636913,
"loss": 2.513,
"theoretical_loss": 3.608782335878308,
"tokens_seen": 1125580800
},
{
"epoch": 0.34,
"learning_rate": 0.00033281845536609826,
"loss": 2.7491,
"theoretical_loss": 3.6087627751808404,
"tokens_seen": 1125646336
},
{
"epoch": 0.34,
"learning_rate": 0.0003328084252758275,
"loss": 2.8207,
"theoretical_loss": 3.608743215941032,
"tokens_seen": 1125711872
},
{
"epoch": 0.34,
"learning_rate": 0.0003327983951855567,
"loss": 2.5722,
"theoretical_loss": 3.6087236581586906,
"tokens_seen": 1125777408
},
{
"epoch": 0.34,
"learning_rate": 0.00033278836509528586,
"loss": 2.7429,
"theoretical_loss": 3.608704101833621,
"tokens_seen": 1125842944
},
{
"epoch": 0.34,
"learning_rate": 0.00033277833500501504,
"loss": 2.7817,
"theoretical_loss": 3.608684546965631,
"tokens_seen": 1125908480
},
{
"epoch": 0.34,
"learning_rate": 0.0003327683049147442,
"loss": 2.591,
"theoretical_loss": 3.6086649935545267,
"tokens_seen": 1125974016
},
{
"epoch": 0.34,
"learning_rate": 0.0003327582748244734,
"loss": 2.6098,
"theoretical_loss": 3.608645441600115,
"tokens_seen": 1126039552
},
{
"epoch": 0.34,
"learning_rate": 0.00033274824473420264,
"loss": 2.4764,
"theoretical_loss": 3.6086258911022018,
"tokens_seen": 1126105088
},
{
"epoch": 0.34,
"learning_rate": 0.00033273821464393176,
"loss": 2.741,
"theoretical_loss": 3.6086063420605954,
"tokens_seen": 1126170624
},
{
"epoch": 0.34,
"learning_rate": 0.000332728184553661,
"loss": 2.9118,
"theoretical_loss": 3.6085867944751016,
"tokens_seen": 1126236160
},
{
"epoch": 0.34,
"learning_rate": 0.0003327181544633902,
"loss": 2.5905,
"theoretical_loss": 3.608567248345527,
"tokens_seen": 1126301696
},
{
"epoch": 0.34,
"learning_rate": 0.00033270812437311936,
"loss": 2.6218,
"theoretical_loss": 3.6085477036716793,
"tokens_seen": 1126367232
},
{
"epoch": 0.34,
"learning_rate": 0.00033269809428284854,
"loss": 2.7802,
"theoretical_loss": 3.6085281604533646,
"tokens_seen": 1126432768
},
{
"epoch": 0.34,
"learning_rate": 0.0003326880641925777,
"loss": 2.7848,
"theoretical_loss": 3.60850861869039,
"tokens_seen": 1126498304
},
{
"epoch": 0.34,
"learning_rate": 0.0003326780341023069,
"loss": 2.4742,
"theoretical_loss": 3.608489078382563,
"tokens_seen": 1126563840
},
{
"epoch": 0.34,
"learning_rate": 0.00033266800401203614,
"loss": 2.6443,
"theoretical_loss": 3.60846953952969,
"tokens_seen": 1126629376
},
{
"epoch": 0.34,
"learning_rate": 0.00033265797392176527,
"loss": 2.7489,
"theoretical_loss": 3.6084500021315784,
"tokens_seen": 1126694912
},
{
"epoch": 0.34,
"learning_rate": 0.0003326479438314945,
"loss": 2.6452,
"theoretical_loss": 3.608430466188035,
"tokens_seen": 1126760448
},
{
"epoch": 0.34,
"learning_rate": 0.00033263791374122363,
"loss": 2.9796,
"theoretical_loss": 3.6084109316988666,
"tokens_seen": 1126825984
},
{
"epoch": 0.34,
"learning_rate": 0.00033262788365095286,
"loss": 2.8909,
"theoretical_loss": 3.6083913986638816,
"tokens_seen": 1126891520
},
{
"epoch": 0.34,
"learning_rate": 0.00033261785356068205,
"loss": 2.8066,
"theoretical_loss": 3.6083718670828864,
"tokens_seen": 1126957056
},
{
"epoch": 0.34,
"learning_rate": 0.0003326078234704112,
"loss": 2.7769,
"theoretical_loss": 3.6083523369556874,
"tokens_seen": 1127022592
},
{
"epoch": 0.34,
"learning_rate": 0.0003325977933801404,
"loss": 2.984,
"theoretical_loss": 3.608332808282093,
"tokens_seen": 1127088128
},
{
"epoch": 0.34,
"learning_rate": 0.00033258776328986964,
"loss": 2.7793,
"theoretical_loss": 3.6083132810619105,
"tokens_seen": 1127153664
},
{
"epoch": 0.34,
"objective/train/docs_used": 642585,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5387659072875977,
"objective/train/theoretical_loss": 3.6082937552949463,
"objective/train/tokens_used": 1147679200,
"theoretical_loss": 3.6082937552949463,
"tokens_seen": 1127219200
},
{
"epoch": 0.34,
"learning_rate": 0.0003325777331995988,
"loss": 2.7054,
"theoretical_loss": 3.6082937552949463,
"tokens_seen": 1127219200
},
{
"epoch": 0.34,
"learning_rate": 0.000332567703109328,
"loss": 2.8608,
"theoretical_loss": 3.608274230981009,
"tokens_seen": 1127284736
},
{
"epoch": 0.34,
"learning_rate": 0.0003325576730190572,
"loss": 2.8157,
"theoretical_loss": 3.608254708119905,
"tokens_seen": 1127350272
},
{
"epoch": 0.34,
"learning_rate": 0.00033254764292878637,
"loss": 2.8771,
"theoretical_loss": 3.6082351867114424,
"tokens_seen": 1127415808
},
{
"epoch": 0.34,
"learning_rate": 0.0003325376128385156,
"loss": 2.9729,
"theoretical_loss": 3.608215666755428,
"tokens_seen": 1127481344
},
{
"epoch": 0.34,
"learning_rate": 0.00033252758274824473,
"loss": 2.7865,
"theoretical_loss": 3.60819614825167,
"tokens_seen": 1127546880
},
{
"epoch": 0.34,
"learning_rate": 0.00033251755265797396,
"loss": 2.7095,
"theoretical_loss": 3.6081766311999757,
"tokens_seen": 1127612416
},
{
"epoch": 0.34,
"learning_rate": 0.0003325075225677031,
"loss": 2.6237,
"theoretical_loss": 3.608157115600153,
"tokens_seen": 1127677952
},
{
"epoch": 0.34,
"learning_rate": 0.00033249749247743233,
"loss": 2.6591,
"theoretical_loss": 3.6081376014520083,
"tokens_seen": 1127743488
},
{
"epoch": 0.34,
"learning_rate": 0.0003324874623871615,
"loss": 2.6671,
"theoretical_loss": 3.608118088755351,
"tokens_seen": 1127809024
},
{
"epoch": 0.34,
"learning_rate": 0.0003324774322968907,
"loss": 2.6051,
"theoretical_loss": 3.6080985775099874,
"tokens_seen": 1127874560
},
{
"epoch": 0.34,
"learning_rate": 0.00033246740220661987,
"loss": 2.4694,
"theoretical_loss": 3.6080790677157255,
"tokens_seen": 1127940096
},
{
"epoch": 0.34,
"learning_rate": 0.00033245737211634905,
"loss": 2.8934,
"theoretical_loss": 3.608059559372374,
"tokens_seen": 1128005632
},
{
"epoch": 0.34,
"learning_rate": 0.00033244734202607823,
"loss": 2.6493,
"theoretical_loss": 3.60804005247974,
"tokens_seen": 1128071168
},
{
"epoch": 0.34,
"learning_rate": 0.00033243731193580747,
"loss": 2.7096,
"theoretical_loss": 3.6080205470376314,
"tokens_seen": 1128136704
},
{
"epoch": 0.34,
"learning_rate": 0.0003324272818455366,
"loss": 2.5269,
"theoretical_loss": 3.608001043045856,
"tokens_seen": 1128202240
},
{
"epoch": 0.34,
"learning_rate": 0.00033241725175526583,
"loss": 2.718,
"theoretical_loss": 3.607981540504222,
"tokens_seen": 1128267776
},
{
"epoch": 0.34,
"learning_rate": 0.000332407221664995,
"loss": 2.8509,
"theoretical_loss": 3.607962039412537,
"tokens_seen": 1128333312
},
{
"epoch": 0.34,
"learning_rate": 0.0003323971915747242,
"loss": 2.6053,
"theoretical_loss": 3.607942539770609,
"tokens_seen": 1128398848
},
{
"epoch": 0.34,
"learning_rate": 0.0003323871614844534,
"loss": 2.605,
"theoretical_loss": 3.607923041578247,
"tokens_seen": 1128464384
},
{
"epoch": 0.34,
"learning_rate": 0.00033237713139418255,
"loss": 2.9847,
"theoretical_loss": 3.6079035448352577,
"tokens_seen": 1128529920
},
{
"epoch": 0.34,
"learning_rate": 0.00033236710130391174,
"loss": 2.7313,
"theoretical_loss": 3.6078840495414495,
"tokens_seen": 1128595456
},
{
"epoch": 0.34,
"learning_rate": 0.00033235707121364097,
"loss": 2.7448,
"theoretical_loss": 3.6078645556966316,
"tokens_seen": 1128660992
},
{
"epoch": 0.34,
"learning_rate": 0.0003323470411233701,
"loss": 2.7622,
"theoretical_loss": 3.607845063300611,
"tokens_seen": 1128726528
},
{
"epoch": 0.34,
"learning_rate": 0.00033233701103309933,
"loss": 2.7321,
"theoretical_loss": 3.6078255723531965,
"tokens_seen": 1128792064
},
{
"epoch": 0.34,
"objective/train/docs_used": 643131,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.757744550704956,
"objective/train/theoretical_loss": 3.6078060828541956,
"objective/train/tokens_used": 1149317600,
"theoretical_loss": 3.6078060828541956,
"tokens_seen": 1128857600
},
{
"epoch": 0.34,
"learning_rate": 0.00033232698094282846,
"loss": 2.869,
"theoretical_loss": 3.6078060828541956,
"tokens_seen": 1128857600
},
{
"epoch": 0.34,
"learning_rate": 0.0003323169508525577,
"loss": 2.7306,
"theoretical_loss": 3.607786594803418,
"tokens_seen": 1128923136
},
{
"epoch": 0.34,
"learning_rate": 0.0003323069207622869,
"loss": 2.7989,
"theoretical_loss": 3.6077671082006706,
"tokens_seen": 1128988672
},
{
"epoch": 0.34,
"learning_rate": 0.00033229689067201606,
"loss": 2.8331,
"theoretical_loss": 3.6077476230457624,
"tokens_seen": 1129054208
},
{
"epoch": 0.34,
"learning_rate": 0.00033228686058174524,
"loss": 2.8449,
"theoretical_loss": 3.6077281393385023,
"tokens_seen": 1129119744
},
{
"epoch": 0.34,
"learning_rate": 0.0003322768304914744,
"loss": 2.6927,
"theoretical_loss": 3.6077086570786974,
"tokens_seen": 1129185280
},
{
"epoch": 0.34,
"learning_rate": 0.0003322668004012036,
"loss": 2.6204,
"theoretical_loss": 3.6076891762661574,
"tokens_seen": 1129250816
},
{
"epoch": 0.34,
"learning_rate": 0.00033225677031093284,
"loss": 2.4774,
"theoretical_loss": 3.6076696969006905,
"tokens_seen": 1129316352
},
{
"epoch": 0.34,
"learning_rate": 0.00033224674022066196,
"loss": 2.9947,
"theoretical_loss": 3.6076502189821045,
"tokens_seen": 1129381888
},
{
"epoch": 0.34,
"learning_rate": 0.0003322367101303912,
"loss": 2.7695,
"theoretical_loss": 3.6076307425102088,
"tokens_seen": 1129447424
},
{
"epoch": 0.34,
"learning_rate": 0.0003322266800401204,
"loss": 2.6874,
"theoretical_loss": 3.6076112674848115,
"tokens_seen": 1129512960
},
{
"epoch": 0.34,
"learning_rate": 0.00033221664994984956,
"loss": 2.8529,
"theoretical_loss": 3.6075917939057223,
"tokens_seen": 1129578496
},
{
"epoch": 0.34,
"learning_rate": 0.00033220661985957874,
"loss": 2.8303,
"theoretical_loss": 3.6075723217727487,
"tokens_seen": 1129644032
},
{
"epoch": 0.34,
"learning_rate": 0.0003321965897693079,
"loss": 2.7575,
"theoretical_loss": 3.6075528510857,
"tokens_seen": 1129709568
},
{
"epoch": 0.34,
"learning_rate": 0.0003321865596790371,
"loss": 2.6134,
"theoretical_loss": 3.6075333818443847,
"tokens_seen": 1129775104
},
{
"epoch": 0.34,
"learning_rate": 0.00033217652958876634,
"loss": 3.0722,
"theoretical_loss": 3.6075139140486114,
"tokens_seen": 1129840640
},
{
"epoch": 0.34,
"learning_rate": 0.00033216649949849547,
"loss": 2.6285,
"theoretical_loss": 3.6074944476981896,
"tokens_seen": 1129906176
},
{
"epoch": 0.34,
"learning_rate": 0.0003321564694082247,
"loss": 2.7835,
"theoretical_loss": 3.6074749827929278,
"tokens_seen": 1129971712
},
{
"epoch": 0.34,
"learning_rate": 0.00033214643931795383,
"loss": 2.7994,
"theoretical_loss": 3.6074555193326345,
"tokens_seen": 1130037248
},
{
"epoch": 0.34,
"learning_rate": 0.00033213640922768306,
"loss": 2.6659,
"theoretical_loss": 3.6074360573171194,
"tokens_seen": 1130102784
},
{
"epoch": 0.34,
"learning_rate": 0.00033212637913741225,
"loss": 2.9294,
"theoretical_loss": 3.607416596746191,
"tokens_seen": 1130168320
},
{
"epoch": 0.34,
"learning_rate": 0.0003321163490471414,
"loss": 2.8299,
"theoretical_loss": 3.6073971376196585,
"tokens_seen": 1130233856
},
{
"epoch": 0.34,
"learning_rate": 0.0003321063189568706,
"loss": 2.9268,
"theoretical_loss": 3.607377679937331,
"tokens_seen": 1130299392
},
{
"epoch": 0.34,
"learning_rate": 0.00033209628886659984,
"loss": 2.6992,
"theoretical_loss": 3.6073582236990176,
"tokens_seen": 1130364928
},
{
"epoch": 0.34,
"learning_rate": 0.00033208625877632897,
"loss": 2.8375,
"theoretical_loss": 3.607338768904527,
"tokens_seen": 1130430464
},
{
"epoch": 0.34,
"objective/train/docs_used": 644333,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7336442470550537,
"objective/train/theoretical_loss": 3.607319315553669,
"objective/train/tokens_used": 1150956000,
"theoretical_loss": 3.607319315553669,
"tokens_seen": 1130496000
},
{
"epoch": 0.34,
"learning_rate": 0.0003320762286860582,
"loss": 2.6878,
"theoretical_loss": 3.607319315553669,
"tokens_seen": 1130496000
},
{
"epoch": 0.34,
"learning_rate": 0.00033206619859578733,
"loss": 2.59,
"theoretical_loss": 3.607299863646252,
"tokens_seen": 1130561536
},
{
"epoch": 0.34,
"learning_rate": 0.00033205616850551657,
"loss": 2.6529,
"theoretical_loss": 3.607280413182086,
"tokens_seen": 1130627072
},
{
"epoch": 0.34,
"learning_rate": 0.00033204613841524575,
"loss": 2.7768,
"theoretical_loss": 3.60726096416098,
"tokens_seen": 1130692608
},
{
"epoch": 0.34,
"learning_rate": 0.00033203610832497493,
"loss": 2.6649,
"theoretical_loss": 3.607241516582743,
"tokens_seen": 1130758144
},
{
"epoch": 0.34,
"learning_rate": 0.0003320260782347041,
"loss": 2.6553,
"theoretical_loss": 3.607222070447185,
"tokens_seen": 1130823680
},
{
"epoch": 0.34,
"learning_rate": 0.0003320160481444333,
"loss": 3.0156,
"theoretical_loss": 3.607202625754115,
"tokens_seen": 1130889216
},
{
"epoch": 0.34,
"learning_rate": 0.0003320060180541625,
"loss": 2.9669,
"theoretical_loss": 3.6071831825033422,
"tokens_seen": 1130954752
},
{
"epoch": 0.34,
"learning_rate": 0.0003319959879638917,
"loss": 2.5195,
"theoretical_loss": 3.607163740694676,
"tokens_seen": 1131020288
},
{
"epoch": 0.34,
"learning_rate": 0.00033198595787362084,
"loss": 2.749,
"theoretical_loss": 3.607144300327927,
"tokens_seen": 1131085824
},
{
"epoch": 0.34,
"learning_rate": 0.00033197592778335007,
"loss": 2.8035,
"theoretical_loss": 3.6071248614029026,
"tokens_seen": 1131151360
},
{
"epoch": 0.34,
"learning_rate": 0.0003319658976930792,
"loss": 2.677,
"theoretical_loss": 3.607105423919415,
"tokens_seen": 1131216896
},
{
"epoch": 0.34,
"learning_rate": 0.00033195586760280843,
"loss": 2.8358,
"theoretical_loss": 3.607085987877271,
"tokens_seen": 1131282432
},
{
"epoch": 0.34,
"learning_rate": 0.0003319458375125376,
"loss": 2.8257,
"theoretical_loss": 3.607066553276283,
"tokens_seen": 1131347968
},
{
"epoch": 0.34,
"learning_rate": 0.0003319358074222668,
"loss": 3.0617,
"theoretical_loss": 3.607047120116259,
"tokens_seen": 1131413504
},
{
"epoch": 0.34,
"learning_rate": 0.000331925777331996,
"loss": 2.7982,
"theoretical_loss": 3.6070276883970083,
"tokens_seen": 1131479040
},
{
"epoch": 0.34,
"learning_rate": 0.0003319157472417252,
"loss": 2.6445,
"theoretical_loss": 3.607008258118342,
"tokens_seen": 1131544576
},
{
"epoch": 0.34,
"learning_rate": 0.00033190571715145434,
"loss": 2.8996,
"theoretical_loss": 3.606988829280069,
"tokens_seen": 1131610112
},
{
"epoch": 0.34,
"learning_rate": 0.0003318956870611836,
"loss": 2.7423,
"theoretical_loss": 3.606969401881999,
"tokens_seen": 1131675648
},
{
"epoch": 0.34,
"learning_rate": 0.0003318856569709127,
"loss": 2.5126,
"theoretical_loss": 3.6069499759239427,
"tokens_seen": 1131741184
},
{
"epoch": 0.34,
"learning_rate": 0.00033187562688064194,
"loss": 2.73,
"theoretical_loss": 3.6069305514057097,
"tokens_seen": 1131806720
},
{
"epoch": 0.34,
"learning_rate": 0.0003318655967903711,
"loss": 2.886,
"theoretical_loss": 3.606911128327109,
"tokens_seen": 1131872256
},
{
"epoch": 0.34,
"learning_rate": 0.0003318555667001003,
"loss": 2.711,
"theoretical_loss": 3.606891706687952,
"tokens_seen": 1131937792
},
{
"epoch": 0.34,
"learning_rate": 0.0003318455366098295,
"loss": 2.65,
"theoretical_loss": 3.6068722864880476,
"tokens_seen": 1132003328
},
{
"epoch": 0.34,
"learning_rate": 0.00033183550651955866,
"loss": 2.9195,
"theoretical_loss": 3.6068528677272065,
"tokens_seen": 1132068864
},
{
"epoch": 0.34,
"objective/train/docs_used": 644979,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6290628910064697,
"objective/train/theoretical_loss": 3.6068334504052375,
"objective/train/tokens_used": 1152594400,
"theoretical_loss": 3.6068334504052375,
"tokens_seen": 1132134400
},
{
"epoch": 0.34,
"learning_rate": 0.0003318254764292879,
"loss": 2.6003,
"theoretical_loss": 3.6068334504052375,
"tokens_seen": 1132134400
},
{
"epoch": 0.34,
"learning_rate": 0.0003318154463390171,
"loss": 2.7656,
"theoretical_loss": 3.6068140345219524,
"tokens_seen": 1132199936
},
{
"epoch": 0.34,
"learning_rate": 0.00033180541624874626,
"loss": 2.7474,
"theoretical_loss": 3.6067946200771606,
"tokens_seen": 1132265472
},
{
"epoch": 0.34,
"learning_rate": 0.00033179538615847544,
"loss": 2.7913,
"theoretical_loss": 3.6067752070706725,
"tokens_seen": 1132331008
},
{
"epoch": 0.34,
"learning_rate": 0.0003317853560682046,
"loss": 2.6928,
"theoretical_loss": 3.6067557955022975,
"tokens_seen": 1132396544
},
{
"epoch": 0.34,
"learning_rate": 0.0003317753259779338,
"loss": 2.6615,
"theoretical_loss": 3.606736385371846,
"tokens_seen": 1132462080
},
{
"epoch": 0.34,
"learning_rate": 0.00033176529588766304,
"loss": 2.5865,
"theoretical_loss": 3.6067169766791296,
"tokens_seen": 1132527616
},
{
"epoch": 0.34,
"learning_rate": 0.00033175526579739216,
"loss": 2.8606,
"theoretical_loss": 3.6066975694239574,
"tokens_seen": 1132593152
},
{
"epoch": 0.34,
"learning_rate": 0.0003317452357071214,
"loss": 2.508,
"theoretical_loss": 3.6066781636061402,
"tokens_seen": 1132658688
},
{
"epoch": 0.34,
"learning_rate": 0.0003317352056168506,
"loss": 2.6269,
"theoretical_loss": 3.606658759225488,
"tokens_seen": 1132724224
},
{
"epoch": 0.34,
"learning_rate": 0.00033172517552657976,
"loss": 2.8208,
"theoretical_loss": 3.6066393562818115,
"tokens_seen": 1132789760
},
{
"epoch": 0.34,
"learning_rate": 0.00033171514543630894,
"loss": 2.7786,
"theoretical_loss": 3.6066199547749216,
"tokens_seen": 1132855296
},
{
"epoch": 0.34,
"learning_rate": 0.0003317051153460381,
"loss": 2.7886,
"theoretical_loss": 3.606600554704628,
"tokens_seen": 1132920832
},
{
"epoch": 0.34,
"learning_rate": 0.0003316950852557673,
"loss": 2.4415,
"theoretical_loss": 3.6065811560707415,
"tokens_seen": 1132986368
},
{
"epoch": 0.34,
"learning_rate": 0.00033168505516549654,
"loss": 2.8795,
"theoretical_loss": 3.6065617588730725,
"tokens_seen": 1133051904
},
{
"epoch": 0.34,
"learning_rate": 0.00033167502507522567,
"loss": 2.9156,
"theoretical_loss": 3.606542363111432,
"tokens_seen": 1133117440
},
{
"epoch": 0.34,
"learning_rate": 0.0003316649949849549,
"loss": 2.8252,
"theoretical_loss": 3.6065229687856304,
"tokens_seen": 1133182976
},
{
"epoch": 0.34,
"learning_rate": 0.00033165496489468403,
"loss": 2.7102,
"theoretical_loss": 3.606503575895479,
"tokens_seen": 1133248512
},
{
"epoch": 0.34,
"learning_rate": 0.00033164493480441326,
"loss": 2.6816,
"theoretical_loss": 3.6064841844407876,
"tokens_seen": 1133314048
},
{
"epoch": 0.34,
"learning_rate": 0.00033163490471414245,
"loss": 2.9253,
"theoretical_loss": 3.6064647944213672,
"tokens_seen": 1133379584
},
{
"epoch": 0.34,
"learning_rate": 0.0003316248746238716,
"loss": 2.7161,
"theoretical_loss": 3.6064454058370288,
"tokens_seen": 1133445120
},
{
"epoch": 0.34,
"learning_rate": 0.0003316148445336008,
"loss": 2.7453,
"theoretical_loss": 3.6064260186875834,
"tokens_seen": 1133510656
},
{
"epoch": 0.34,
"learning_rate": 0.00033160481444333004,
"loss": 2.7935,
"theoretical_loss": 3.606406632972841,
"tokens_seen": 1133576192
},
{
"epoch": 0.34,
"learning_rate": 0.00033159478435305917,
"loss": 2.8471,
"theoretical_loss": 3.606387248692614,
"tokens_seen": 1133641728
},
{
"epoch": 0.34,
"learning_rate": 0.0003315847542627884,
"loss": 2.6168,
"theoretical_loss": 3.6063678658467113,
"tokens_seen": 1133707264
},
{
"epoch": 0.34,
"objective/train/docs_used": 646384,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.9743421077728271,
"objective/train/theoretical_loss": 3.6063484844349456,
"objective/train/tokens_used": 1154232800,
"theoretical_loss": 3.6063484844349456,
"tokens_seen": 1133772800
},
{
"epoch": 0.34,
"learning_rate": 0.00033157472417251753,
"loss": 2.5171,
"theoretical_loss": 3.6063484844349456,
"tokens_seen": 1133772800
},
{
"epoch": 0.34,
"learning_rate": 0.00033156469408224677,
"loss": 2.7615,
"theoretical_loss": 3.606329104457127,
"tokens_seen": 1133838336
},
{
"epoch": 0.34,
"learning_rate": 0.00033155466399197595,
"loss": 2.7525,
"theoretical_loss": 3.606309725913067,
"tokens_seen": 1133903872
},
{
"epoch": 0.34,
"learning_rate": 0.00033154463390170513,
"loss": 2.9148,
"theoretical_loss": 3.6062903488025766,
"tokens_seen": 1133969408
},
{
"epoch": 0.34,
"learning_rate": 0.0003315346038114343,
"loss": 2.646,
"theoretical_loss": 3.6062709731254663,
"tokens_seen": 1134034944
},
{
"epoch": 0.34,
"learning_rate": 0.0003315245737211635,
"loss": 3.1244,
"theoretical_loss": 3.6062515988815482,
"tokens_seen": 1134100480
},
{
"epoch": 0.34,
"learning_rate": 0.0003315145436308927,
"loss": 2.6945,
"theoretical_loss": 3.6062322260706328,
"tokens_seen": 1134166016
},
{
"epoch": 0.34,
"learning_rate": 0.0003315045135406219,
"loss": 2.7218,
"theoretical_loss": 3.606212854692531,
"tokens_seen": 1134231552
},
{
"epoch": 0.34,
"learning_rate": 0.00033149448345035104,
"loss": 2.6449,
"theoretical_loss": 3.606193484747055,
"tokens_seen": 1134297088
},
{
"epoch": 0.34,
"learning_rate": 0.00033148445336008027,
"loss": 2.5771,
"theoretical_loss": 3.606174116234016,
"tokens_seen": 1134362624
},
{
"epoch": 0.34,
"learning_rate": 0.0003314744232698094,
"loss": 2.7395,
"theoretical_loss": 3.606154749153225,
"tokens_seen": 1134428160
},
{
"epoch": 0.34,
"learning_rate": 0.00033146439317953863,
"loss": 2.632,
"theoretical_loss": 3.6061353835044927,
"tokens_seen": 1134493696
},
{
"epoch": 0.34,
"learning_rate": 0.0003314543630892678,
"loss": 2.7776,
"theoretical_loss": 3.606116019287631,
"tokens_seen": 1134559232
},
{
"epoch": 0.34,
"learning_rate": 0.000331444332998997,
"loss": 2.8511,
"theoretical_loss": 3.6060966565024524,
"tokens_seen": 1134624768
},
{
"epoch": 0.34,
"learning_rate": 0.0003314343029087262,
"loss": 2.9081,
"theoretical_loss": 3.6060772951487667,
"tokens_seen": 1134690304
},
{
"epoch": 0.34,
"learning_rate": 0.0003314242728184554,
"loss": 2.6425,
"theoretical_loss": 3.606057935226386,
"tokens_seen": 1134755840
},
{
"epoch": 0.34,
"learning_rate": 0.00033141424272818454,
"loss": 2.6831,
"theoretical_loss": 3.6060385767351226,
"tokens_seen": 1134821376
},
{
"epoch": 0.34,
"learning_rate": 0.0003314042126379138,
"loss": 2.3672,
"theoretical_loss": 3.6060192196747867,
"tokens_seen": 1134886912
},
{
"epoch": 0.34,
"learning_rate": 0.0003313941825476429,
"loss": 2.6881,
"theoretical_loss": 3.6059998640451907,
"tokens_seen": 1134952448
},
{
"epoch": 0.34,
"learning_rate": 0.00033138415245737214,
"loss": 2.6506,
"theoretical_loss": 3.605980509846147,
"tokens_seen": 1135017984
},
{
"epoch": 0.34,
"learning_rate": 0.0003313741223671013,
"loss": 2.8889,
"theoretical_loss": 3.6059611570774655,
"tokens_seen": 1135083520
},
{
"epoch": 0.34,
"learning_rate": 0.0003313640922768305,
"loss": 2.8921,
"theoretical_loss": 3.6059418057389596,
"tokens_seen": 1135149056
},
{
"epoch": 0.34,
"learning_rate": 0.0003313540621865597,
"loss": 2.6728,
"theoretical_loss": 3.6059224558304397,
"tokens_seen": 1135214592
},
{
"epoch": 0.34,
"learning_rate": 0.00033134403209628886,
"loss": 2.8802,
"theoretical_loss": 3.6059031073517183,
"tokens_seen": 1135280128
},
{
"epoch": 0.34,
"learning_rate": 0.00033133400200601804,
"loss": 2.7168,
"theoretical_loss": 3.6058837603026075,
"tokens_seen": 1135345664
},
{
"epoch": 0.34,
"objective/train/docs_used": 647116,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1141538619995117,
"objective/train/theoretical_loss": 3.6058644146829186,
"objective/train/tokens_used": 1155871200,
"theoretical_loss": 3.6058644146829186,
"tokens_seen": 1135411200
},
{
"epoch": 0.34,
"learning_rate": 0.0003313239719157473,
"loss": 2.8572,
"theoretical_loss": 3.6058644146829186,
"tokens_seen": 1135411200
},
{
"epoch": 0.34,
"learning_rate": 0.0003313139418254764,
"loss": 2.8062,
"theoretical_loss": 3.605845070492464,
"tokens_seen": 1135476736
},
{
"epoch": 0.34,
"learning_rate": 0.00033130391173520564,
"loss": 2.6768,
"theoretical_loss": 3.6058257277310544,
"tokens_seen": 1135542272
},
{
"epoch": 0.34,
"learning_rate": 0.00033129388164493477,
"loss": 2.9178,
"theoretical_loss": 3.6058063863985033,
"tokens_seen": 1135607808
},
{
"epoch": 0.34,
"learning_rate": 0.000331283851554664,
"loss": 2.8738,
"theoretical_loss": 3.605787046494622,
"tokens_seen": 1135673344
},
{
"epoch": 0.34,
"learning_rate": 0.0003312738214643932,
"loss": 2.9863,
"theoretical_loss": 3.605767708019222,
"tokens_seen": 1135738880
},
{
"epoch": 0.34,
"learning_rate": 0.00033126379137412236,
"loss": 2.752,
"theoretical_loss": 3.6057483709721168,
"tokens_seen": 1135804416
},
{
"epoch": 0.34,
"learning_rate": 0.00033125376128385155,
"loss": 2.7478,
"theoretical_loss": 3.605729035353117,
"tokens_seen": 1135869952
},
{
"epoch": 0.34,
"learning_rate": 0.0003312437311935808,
"loss": 2.6542,
"theoretical_loss": 3.605709701162036,
"tokens_seen": 1135935488
},
{
"epoch": 0.34,
"learning_rate": 0.0003312337011033099,
"loss": 2.7706,
"theoretical_loss": 3.6056903683986854,
"tokens_seen": 1136001024
},
{
"epoch": 0.34,
"learning_rate": 0.00033122367101303914,
"loss": 2.6313,
"theoretical_loss": 3.6056710370628773,
"tokens_seen": 1136066560
},
{
"epoch": 0.34,
"learning_rate": 0.00033121364092276827,
"loss": 2.4377,
"theoretical_loss": 3.605651707154424,
"tokens_seen": 1136132096
},
{
"epoch": 0.34,
"learning_rate": 0.0003312036108324975,
"loss": 2.6523,
"theoretical_loss": 3.6056323786731372,
"tokens_seen": 1136197632
},
{
"epoch": 0.34,
"learning_rate": 0.0003311935807422267,
"loss": 2.8853,
"theoretical_loss": 3.6056130516188305,
"tokens_seen": 1136263168
},
{
"epoch": 0.34,
"learning_rate": 0.00033118355065195587,
"loss": 2.96,
"theoretical_loss": 3.605593725991316,
"tokens_seen": 1136328704
},
{
"epoch": 0.34,
"learning_rate": 0.00033117352056168505,
"loss": 2.7856,
"theoretical_loss": 3.605574401790405,
"tokens_seen": 1136394240
},
{
"epoch": 0.34,
"learning_rate": 0.00033116349047141423,
"loss": 2.6421,
"theoretical_loss": 3.605555079015911,
"tokens_seen": 1136459776
},
{
"epoch": 0.34,
"learning_rate": 0.0003311534603811434,
"loss": 2.9207,
"theoretical_loss": 3.6055357576676457,
"tokens_seen": 1136525312
},
{
"epoch": 0.34,
"learning_rate": 0.00033114343029087265,
"loss": 2.8436,
"theoretical_loss": 3.605516437745422,
"tokens_seen": 1136590848
},
{
"epoch": 0.34,
"learning_rate": 0.00033113340020060177,
"loss": 2.7711,
"theoretical_loss": 3.6054971192490526,
"tokens_seen": 1136656384
},
{
"epoch": 0.34,
"learning_rate": 0.000331123370110331,
"loss": 2.8246,
"theoretical_loss": 3.60547780217835,
"tokens_seen": 1136721920
},
{
"epoch": 0.34,
"learning_rate": 0.00033111334002006013,
"loss": 2.6141,
"theoretical_loss": 3.6054584865331263,
"tokens_seen": 1136787456
},
{
"epoch": 0.34,
"learning_rate": 0.00033110330992978937,
"loss": 2.8557,
"theoretical_loss": 3.605439172313195,
"tokens_seen": 1136852992
},
{
"epoch": 0.34,
"learning_rate": 0.00033109327983951855,
"loss": 2.8189,
"theoretical_loss": 3.605419859518368,
"tokens_seen": 1136918528
},
{
"epoch": 0.34,
"learning_rate": 0.00033108324974924773,
"loss": 2.7353,
"theoretical_loss": 3.6054005481484586,
"tokens_seen": 1136984064
},
{
"epoch": 0.34,
"objective/train/docs_used": 648555,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1323835849761963,
"objective/train/theoretical_loss": 3.605381238203279,
"objective/train/tokens_used": 1157509600,
"theoretical_loss": 3.605381238203279,
"tokens_seen": 1137049600
},
{
"epoch": 0.34,
"learning_rate": 0.00033107321965897697,
"loss": 2.5809,
"theoretical_loss": 3.605381238203279,
"tokens_seen": 1137049600
},
{
"epoch": 0.34,
"learning_rate": 0.00033106318956870615,
"loss": 2.7712,
"theoretical_loss": 3.6053619296826422,
"tokens_seen": 1137115136
},
{
"epoch": 0.34,
"learning_rate": 0.00033105315947843533,
"loss": 2.7117,
"theoretical_loss": 3.6053426225863614,
"tokens_seen": 1137180672
},
{
"epoch": 0.34,
"learning_rate": 0.0003310431293881645,
"loss": 2.9508,
"theoretical_loss": 3.605323316914249,
"tokens_seen": 1137246208
},
{
"epoch": 0.34,
"learning_rate": 0.0003310330992978937,
"loss": 2.9234,
"theoretical_loss": 3.6053040126661173,
"tokens_seen": 1137311744
},
{
"epoch": 0.34,
"learning_rate": 0.0003310230692076229,
"loss": 2.3999,
"theoretical_loss": 3.6052847098417806,
"tokens_seen": 1137377280
},
{
"epoch": 0.34,
"learning_rate": 0.0003310130391173521,
"loss": 2.9373,
"theoretical_loss": 3.605265408441051,
"tokens_seen": 1137442816
},
{
"epoch": 0.34,
"learning_rate": 0.00033100300902708124,
"loss": 2.8885,
"theoretical_loss": 3.6052461084637413,
"tokens_seen": 1137508352
},
{
"epoch": 0.34,
"learning_rate": 0.00033099297893681047,
"loss": 2.8577,
"theoretical_loss": 3.6052268099096656,
"tokens_seen": 1137573888
},
{
"epoch": 0.34,
"learning_rate": 0.0003309829488465396,
"loss": 2.8555,
"theoretical_loss": 3.605207512778636,
"tokens_seen": 1137639424
},
{
"epoch": 0.34,
"learning_rate": 0.00033097291875626883,
"loss": 2.8957,
"theoretical_loss": 3.6051882170704657,
"tokens_seen": 1137704960
},
{
"epoch": 0.34,
"learning_rate": 0.000330962888665998,
"loss": 2.5257,
"theoretical_loss": 3.605168922784968,
"tokens_seen": 1137770496
},
{
"epoch": 0.34,
"learning_rate": 0.0003309528585757272,
"loss": 2.5782,
"theoretical_loss": 3.605149629921956,
"tokens_seen": 1137836032
},
{
"epoch": 0.34,
"learning_rate": 0.0003309428284854564,
"loss": 2.7627,
"theoretical_loss": 3.605130338481243,
"tokens_seen": 1137901568
},
{
"epoch": 0.34,
"learning_rate": 0.0003309327983951856,
"loss": 2.8151,
"theoretical_loss": 3.6051110484626423,
"tokens_seen": 1137967104
},
{
"epoch": 0.34,
"learning_rate": 0.00033092276830491474,
"loss": 2.7584,
"theoretical_loss": 3.605091759865967,
"tokens_seen": 1138032640
},
{
"epoch": 0.34,
"learning_rate": 0.000330912738214644,
"loss": 2.774,
"theoretical_loss": 3.60507247269103,
"tokens_seen": 1138098176
},
{
"epoch": 0.34,
"learning_rate": 0.0003309027081243731,
"loss": 2.817,
"theoretical_loss": 3.6050531869376456,
"tokens_seen": 1138163712
},
{
"epoch": 0.34,
"learning_rate": 0.00033089267803410234,
"loss": 2.8403,
"theoretical_loss": 3.605033902605627,
"tokens_seen": 1138229248
},
{
"epoch": 0.34,
"learning_rate": 0.0003308826479438315,
"loss": 2.8191,
"theoretical_loss": 3.6050146196947868,
"tokens_seen": 1138294784
},
{
"epoch": 0.34,
"learning_rate": 0.0003308726178535607,
"loss": 2.8957,
"theoretical_loss": 3.6049953382049393,
"tokens_seen": 1138360320
},
{
"epoch": 0.34,
"learning_rate": 0.0003308625877632899,
"loss": 2.8585,
"theoretical_loss": 3.6049760581358976,
"tokens_seen": 1138425856
},
{
"epoch": 0.34,
"learning_rate": 0.00033085255767301906,
"loss": 2.778,
"theoretical_loss": 3.604956779487475,
"tokens_seen": 1138491392
},
{
"epoch": 0.35,
"learning_rate": 0.00033084252758274824,
"loss": 2.8685,
"theoretical_loss": 3.604937502259485,
"tokens_seen": 1138556928
},
{
"epoch": 0.35,
"learning_rate": 0.0003308324974924775,
"loss": 2.6466,
"theoretical_loss": 3.6049182264517423,
"tokens_seen": 1138622464
},
{
"epoch": 0.35,
"objective/train/docs_used": 649005,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6947977542877197,
"objective/train/theoretical_loss": 3.60489895206406,
"objective/train/tokens_used": 1159148000,
"theoretical_loss": 3.60489895206406,
"tokens_seen": 1138688000
},
{
"epoch": 0.35,
"learning_rate": 0.0003308224674022066,
"loss": 2.7302,
"theoretical_loss": 3.60489895206406,
"tokens_seen": 1138688000
},
{
"epoch": 0.35,
"learning_rate": 0.00033081243731193584,
"loss": 2.7559,
"theoretical_loss": 3.6048796790962507,
"tokens_seen": 1138753536
},
{
"epoch": 0.35,
"learning_rate": 0.00033080240722166497,
"loss": 2.8367,
"theoretical_loss": 3.604860407548129,
"tokens_seen": 1138819072
},
{
"epoch": 0.35,
"learning_rate": 0.0003307923771313942,
"loss": 2.7693,
"theoretical_loss": 3.604841137419509,
"tokens_seen": 1138884608
},
{
"epoch": 0.35,
"learning_rate": 0.0003307823470411234,
"loss": 2.7063,
"theoretical_loss": 3.6048218687102036,
"tokens_seen": 1138950144
},
{
"epoch": 0.35,
"learning_rate": 0.00033077231695085256,
"loss": 2.6783,
"theoretical_loss": 3.6048026014200274,
"tokens_seen": 1139015680
},
{
"epoch": 0.35,
"learning_rate": 0.00033076228686058175,
"loss": 2.8529,
"theoretical_loss": 3.6047833355487935,
"tokens_seen": 1139081216
},
{
"epoch": 0.35,
"learning_rate": 0.000330752256770311,
"loss": 2.872,
"theoretical_loss": 3.6047640710963162,
"tokens_seen": 1139146752
},
{
"epoch": 0.35,
"learning_rate": 0.0003307422266800401,
"loss": 2.8379,
"theoretical_loss": 3.604744808062409,
"tokens_seen": 1139212288
},
{
"epoch": 0.35,
"learning_rate": 0.00033073219658976934,
"loss": 2.6487,
"theoretical_loss": 3.604725546446887,
"tokens_seen": 1139277824
},
{
"epoch": 0.35,
"learning_rate": 0.00033072216649949847,
"loss": 2.666,
"theoretical_loss": 3.604706286249563,
"tokens_seen": 1139343360
},
{
"epoch": 0.35,
"learning_rate": 0.0003307121364092277,
"loss": 2.7396,
"theoretical_loss": 3.604687027470251,
"tokens_seen": 1139408896
},
{
"epoch": 0.35,
"learning_rate": 0.0003307021063189569,
"loss": 2.7975,
"theoretical_loss": 3.604667770108766,
"tokens_seen": 1139474432
},
{
"epoch": 0.35,
"learning_rate": 0.00033069207622868607,
"loss": 2.5601,
"theoretical_loss": 3.6046485141649214,
"tokens_seen": 1139539968
},
{
"epoch": 0.35,
"learning_rate": 0.00033068204613841525,
"loss": 2.9034,
"theoretical_loss": 3.604629259638531,
"tokens_seen": 1139605504
},
{
"epoch": 0.35,
"learning_rate": 0.00033067201604814443,
"loss": 2.836,
"theoretical_loss": 3.60461000652941,
"tokens_seen": 1139671040
},
{
"epoch": 0.35,
"learning_rate": 0.0003306619859578736,
"loss": 2.8252,
"theoretical_loss": 3.6045907548373712,
"tokens_seen": 1139736576
},
{
"epoch": 0.35,
"learning_rate": 0.00033065195586760285,
"loss": 2.8731,
"theoretical_loss": 3.60457150456223,
"tokens_seen": 1139802112
},
{
"epoch": 0.35,
"learning_rate": 0.00033064192577733197,
"loss": 2.6517,
"theoretical_loss": 3.6045522557038003,
"tokens_seen": 1139867648
},
{
"epoch": 0.35,
"learning_rate": 0.0003306318956870612,
"loss": 2.8269,
"theoretical_loss": 3.604533008261896,
"tokens_seen": 1139933184
},
{
"epoch": 0.35,
"learning_rate": 0.00033062186559679034,
"loss": 2.7872,
"theoretical_loss": 3.6045137622363326,
"tokens_seen": 1139998720
},
{
"epoch": 0.35,
"learning_rate": 0.00033061183550651957,
"loss": 2.7574,
"theoretical_loss": 3.6044945176269225,
"tokens_seen": 1140064256
},
{
"epoch": 0.35,
"learning_rate": 0.00033060180541624875,
"loss": 2.8454,
"theoretical_loss": 3.604475274433482,
"tokens_seen": 1140129792
},
{
"epoch": 0.35,
"learning_rate": 0.00033059177532597793,
"loss": 2.5575,
"theoretical_loss": 3.6044560326558246,
"tokens_seen": 1140195328
},
{
"epoch": 0.35,
"learning_rate": 0.0003305817452357071,
"loss": 2.7423,
"theoretical_loss": 3.6044367922937646,
"tokens_seen": 1140260864
},
{
"epoch": 0.35,
"objective/train/docs_used": 649993,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.981397867202759,
"objective/train/theoretical_loss": 3.604417553347117,
"objective/train/tokens_used": 1160786400,
"theoretical_loss": 3.604417553347117,
"tokens_seen": 1140326400
},
{
"epoch": 0.35,
"learning_rate": 0.00033057171514543635,
"loss": 2.8374,
"theoretical_loss": 3.604417553347117,
"tokens_seen": 1140326400
},
{
"epoch": 0.35,
"learning_rate": 0.0003305616850551655,
"loss": 2.8061,
"theoretical_loss": 3.604398315815697,
"tokens_seen": 1140391936
},
{
"epoch": 0.35,
"learning_rate": 0.0003305516549648947,
"loss": 2.8699,
"theoretical_loss": 3.6043790796993176,
"tokens_seen": 1140457472
},
{
"epoch": 0.35,
"learning_rate": 0.00033054162487462384,
"loss": 2.5853,
"theoretical_loss": 3.604359844997794,
"tokens_seen": 1140523008
},
{
"epoch": 0.35,
"learning_rate": 0.0003305315947843531,
"loss": 2.7451,
"theoretical_loss": 3.604340611710941,
"tokens_seen": 1140588544
},
{
"epoch": 0.35,
"learning_rate": 0.00033052156469408225,
"loss": 2.7422,
"theoretical_loss": 3.604321379838574,
"tokens_seen": 1140654080
},
{
"epoch": 0.35,
"learning_rate": 0.00033051153460381144,
"loss": 2.7241,
"theoretical_loss": 3.6043021493805067,
"tokens_seen": 1140719616
},
{
"epoch": 0.35,
"learning_rate": 0.0003305015045135406,
"loss": 2.5897,
"theoretical_loss": 3.604282920336554,
"tokens_seen": 1140785152
},
{
"epoch": 0.35,
"learning_rate": 0.0003304914744232698,
"loss": 2.5632,
"theoretical_loss": 3.604263692706531,
"tokens_seen": 1140850688
},
{
"epoch": 0.35,
"learning_rate": 0.000330481444332999,
"loss": 2.4954,
"theoretical_loss": 3.6042444664902518,
"tokens_seen": 1140916224
},
{
"epoch": 0.35,
"learning_rate": 0.0003304714142427282,
"loss": 2.8019,
"theoretical_loss": 3.604225241687532,
"tokens_seen": 1140981760
},
{
"epoch": 0.35,
"learning_rate": 0.00033046138415245734,
"loss": 2.5568,
"theoretical_loss": 3.6042060182981865,
"tokens_seen": 1141047296
},
{
"epoch": 0.35,
"learning_rate": 0.0003304513540621866,
"loss": 2.651,
"theoretical_loss": 3.60418679632203,
"tokens_seen": 1141112832
},
{
"epoch": 0.35,
"learning_rate": 0.0003304413239719157,
"loss": 2.7954,
"theoretical_loss": 3.6041675757588774,
"tokens_seen": 1141178368
},
{
"epoch": 0.35,
"learning_rate": 0.00033043129388164494,
"loss": 2.6274,
"theoretical_loss": 3.604148356608544,
"tokens_seen": 1141243904
},
{
"epoch": 0.35,
"learning_rate": 0.0003304212637913741,
"loss": 2.9704,
"theoretical_loss": 3.604129138870844,
"tokens_seen": 1141309440
},
{
"epoch": 0.35,
"learning_rate": 0.0003304112337011033,
"loss": 2.8942,
"theoretical_loss": 3.604109922545594,
"tokens_seen": 1141374976
},
{
"epoch": 0.35,
"learning_rate": 0.0003304012036108325,
"loss": 2.7095,
"theoretical_loss": 3.6040907076326074,
"tokens_seen": 1141440512
},
{
"epoch": 0.35,
"learning_rate": 0.0003303911735205617,
"loss": 2.941,
"theoretical_loss": 3.6040714941317002,
"tokens_seen": 1141506048
},
{
"epoch": 0.35,
"learning_rate": 0.00033038114343029084,
"loss": 2.6138,
"theoretical_loss": 3.6040522820426877,
"tokens_seen": 1141571584
},
{
"epoch": 0.35,
"learning_rate": 0.0003303711133400201,
"loss": 2.8044,
"theoretical_loss": 3.6040330713653845,
"tokens_seen": 1141637120
},
{
"epoch": 0.35,
"learning_rate": 0.0003303610832497492,
"loss": 2.6448,
"theoretical_loss": 3.604013862099607,
"tokens_seen": 1141702656
},
{
"epoch": 0.35,
"learning_rate": 0.00033035105315947844,
"loss": 2.7507,
"theoretical_loss": 3.6039946542451684,
"tokens_seen": 1141768192
},
{
"epoch": 0.35,
"learning_rate": 0.0003303410230692076,
"loss": 2.749,
"theoretical_loss": 3.6039754478018864,
"tokens_seen": 1141833728
},
{
"epoch": 0.35,
"learning_rate": 0.0003303309929789368,
"loss": 2.665,
"theoretical_loss": 3.6039562427695744,
"tokens_seen": 1141899264
},
{
"epoch": 0.35,
"objective/train/docs_used": 650682,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1637423038482666,
"objective/train/theoretical_loss": 3.603937039148049,
"objective/train/tokens_used": 1162424800,
"theoretical_loss": 3.603937039148049,
"tokens_seen": 1141964800
},
{
"epoch": 0.35,
"learning_rate": 0.00033032096288866604,
"loss": 2.6127,
"theoretical_loss": 3.603937039148049,
"tokens_seen": 1141964800
},
{
"epoch": 0.35,
"learning_rate": 0.00033031093279839517,
"loss": 2.698,
"theoretical_loss": 3.603917836937125,
"tokens_seen": 1142030336
},
{
"epoch": 0.35,
"learning_rate": 0.0003303009027081244,
"loss": 2.6779,
"theoretical_loss": 3.603898636136618,
"tokens_seen": 1142095872
},
{
"epoch": 0.35,
"learning_rate": 0.0003302908726178536,
"loss": 2.7799,
"theoretical_loss": 3.603879436746344,
"tokens_seen": 1142161408
},
{
"epoch": 0.35,
"learning_rate": 0.00033028084252758276,
"loss": 2.6615,
"theoretical_loss": 3.6038602387661176,
"tokens_seen": 1142226944
},
{
"epoch": 0.35,
"learning_rate": 0.00033027081243731195,
"loss": 2.7079,
"theoretical_loss": 3.603841042195755,
"tokens_seen": 1142292480
},
{
"epoch": 0.35,
"learning_rate": 0.0003302607823470412,
"loss": 2.7778,
"theoretical_loss": 3.6038218470350714,
"tokens_seen": 1142358016
},
{
"epoch": 0.35,
"learning_rate": 0.0003302507522567703,
"loss": 2.7254,
"theoretical_loss": 3.603802653283883,
"tokens_seen": 1142423552
},
{
"epoch": 0.35,
"learning_rate": 0.00033024072216649954,
"loss": 2.6211,
"theoretical_loss": 3.6037834609420045,
"tokens_seen": 1142489088
},
{
"epoch": 0.35,
"learning_rate": 0.00033023069207622867,
"loss": 2.9376,
"theoretical_loss": 3.6037642700092523,
"tokens_seen": 1142554624
},
{
"epoch": 0.35,
"learning_rate": 0.0003302206619859579,
"loss": 2.8105,
"theoretical_loss": 3.603745080485442,
"tokens_seen": 1142620160
},
{
"epoch": 0.35,
"learning_rate": 0.0003302106318956871,
"loss": 2.6391,
"theoretical_loss": 3.603725892370389,
"tokens_seen": 1142685696
},
{
"epoch": 0.35,
"learning_rate": 0.00033020060180541627,
"loss": 2.6696,
"theoretical_loss": 3.60370670566391,
"tokens_seen": 1142751232
},
{
"epoch": 0.35,
"learning_rate": 0.00033019057171514545,
"loss": 2.6671,
"theoretical_loss": 3.6036875203658196,
"tokens_seen": 1142816768
},
{
"epoch": 0.35,
"learning_rate": 0.00033018054162487463,
"loss": 2.6701,
"theoretical_loss": 3.6036683364759345,
"tokens_seen": 1142882304
},
{
"epoch": 0.35,
"learning_rate": 0.0003301705115346038,
"loss": 2.4225,
"theoretical_loss": 3.6036491539940707,
"tokens_seen": 1142947840
},
{
"epoch": 0.35,
"learning_rate": 0.00033016048144433305,
"loss": 2.4601,
"theoretical_loss": 3.6036299729200434,
"tokens_seen": 1143013376
},
{
"epoch": 0.35,
"learning_rate": 0.0003301504513540622,
"loss": 2.5929,
"theoretical_loss": 3.603610793253669,
"tokens_seen": 1143078912
},
{
"epoch": 0.35,
"learning_rate": 0.0003301404212637914,
"loss": 2.7939,
"theoretical_loss": 3.6035916149947638,
"tokens_seen": 1143144448
},
{
"epoch": 0.35,
"learning_rate": 0.00033013039117352054,
"loss": 2.6977,
"theoretical_loss": 3.6035724381431433,
"tokens_seen": 1143209984
},
{
"epoch": 0.35,
"learning_rate": 0.00033012036108324977,
"loss": 3.0647,
"theoretical_loss": 3.6035532626986235,
"tokens_seen": 1143275520
},
{
"epoch": 0.35,
"learning_rate": 0.00033011033099297895,
"loss": 2.7457,
"theoretical_loss": 3.6035340886610214,
"tokens_seen": 1143341056
},
{
"epoch": 0.35,
"learning_rate": 0.00033010030090270813,
"loss": 2.7155,
"theoretical_loss": 3.6035149160301523,
"tokens_seen": 1143406592
},
{
"epoch": 0.35,
"learning_rate": 0.0003300902708124373,
"loss": 2.839,
"theoretical_loss": 3.6034957448058327,
"tokens_seen": 1143472128
},
{
"epoch": 0.35,
"learning_rate": 0.00033008024072216655,
"loss": 2.8907,
"theoretical_loss": 3.6034765749878783,
"tokens_seen": 1143537664
},
{
"epoch": 0.35,
"objective/train/docs_used": 652177,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.300645589828491,
"objective/train/theoretical_loss": 3.603457406576106,
"objective/train/tokens_used": 1164063200,
"theoretical_loss": 3.603457406576106,
"tokens_seen": 1143603200
},
{
"epoch": 0.35,
"learning_rate": 0.0003300702106318957,
"loss": 2.7365,
"theoretical_loss": 3.603457406576106,
"tokens_seen": 1143603200
},
{
"epoch": 0.35,
"learning_rate": 0.0003300601805416249,
"loss": 2.8359,
"theoretical_loss": 3.603438239570332,
"tokens_seen": 1143668736
},
{
"epoch": 0.35,
"learning_rate": 0.00033005015045135404,
"loss": 2.6647,
"theoretical_loss": 3.6034190739703718,
"tokens_seen": 1143734272
},
{
"epoch": 0.35,
"learning_rate": 0.0003300401203610833,
"loss": 2.8436,
"theoretical_loss": 3.6033999097760434,
"tokens_seen": 1143799808
},
{
"epoch": 0.35,
"learning_rate": 0.00033003009027081245,
"loss": 2.678,
"theoretical_loss": 3.6033807469871615,
"tokens_seen": 1143865344
},
{
"epoch": 0.35,
"learning_rate": 0.00033002006018054164,
"loss": 2.8835,
"theoretical_loss": 3.603361585603543,
"tokens_seen": 1143930880
},
{
"epoch": 0.35,
"learning_rate": 0.0003300100300902708,
"loss": 2.673,
"theoretical_loss": 3.603342425625005,
"tokens_seen": 1143996416
},
{
"epoch": 0.35,
"learning_rate": 0.00033,
"loss": 2.8912,
"theoretical_loss": 3.603323267051363,
"tokens_seen": 1144061952
},
{
"epoch": 0.35,
"learning_rate": 0.0003299899699097292,
"loss": 2.5714,
"theoretical_loss": 3.6033041098824343,
"tokens_seen": 1144127488
},
{
"epoch": 0.35,
"learning_rate": 0.0003299799398194584,
"loss": 2.6295,
"theoretical_loss": 3.603284954118035,
"tokens_seen": 1144193024
},
{
"epoch": 0.35,
"learning_rate": 0.00032996990972918754,
"loss": 3.0462,
"theoretical_loss": 3.6032657997579824,
"tokens_seen": 1144258560
},
{
"epoch": 0.35,
"learning_rate": 0.0003299598796389168,
"loss": 2.7383,
"theoretical_loss": 3.6032466468020923,
"tokens_seen": 1144324096
},
{
"epoch": 0.35,
"learning_rate": 0.0003299498495486459,
"loss": 2.5814,
"theoretical_loss": 3.6032274952501817,
"tokens_seen": 1144389632
},
{
"epoch": 0.35,
"learning_rate": 0.00032993981945837514,
"loss": 2.923,
"theoretical_loss": 3.603208345102067,
"tokens_seen": 1144455168
},
{
"epoch": 0.35,
"learning_rate": 0.0003299297893681043,
"loss": 2.5904,
"theoretical_loss": 3.6031891963575653,
"tokens_seen": 1144520704
},
{
"epoch": 0.35,
"learning_rate": 0.0003299197592778335,
"loss": 2.5105,
"theoretical_loss": 3.6031700490164926,
"tokens_seen": 1144586240
},
{
"epoch": 0.35,
"learning_rate": 0.0003299097291875627,
"loss": 2.7215,
"theoretical_loss": 3.6031509030786673,
"tokens_seen": 1144651776
},
{
"epoch": 0.35,
"learning_rate": 0.0003298996990972919,
"loss": 2.8319,
"theoretical_loss": 3.6031317585439044,
"tokens_seen": 1144717312
},
{
"epoch": 0.35,
"learning_rate": 0.00032988966900702104,
"loss": 2.8828,
"theoretical_loss": 3.603112615412022,
"tokens_seen": 1144782848
},
{
"epoch": 0.35,
"learning_rate": 0.0003298796389167503,
"loss": 2.7023,
"theoretical_loss": 3.603093473682837,
"tokens_seen": 1144848384
},
{
"epoch": 0.35,
"learning_rate": 0.0003298696088264794,
"loss": 2.8012,
"theoretical_loss": 3.6030743333561652,
"tokens_seen": 1144913920
},
{
"epoch": 0.35,
"learning_rate": 0.00032985957873620864,
"loss": 2.6341,
"theoretical_loss": 3.6030551944318248,
"tokens_seen": 1144979456
},
{
"epoch": 0.35,
"learning_rate": 0.0003298495486459378,
"loss": 2.6463,
"theoretical_loss": 3.603036056909632,
"tokens_seen": 1145044992
},
{
"epoch": 0.35,
"learning_rate": 0.000329839518555667,
"loss": 2.7908,
"theoretical_loss": 3.6030169207894045,
"tokens_seen": 1145110528
},
{
"epoch": 0.35,
"learning_rate": 0.0003298294884653962,
"loss": 2.8535,
"theoretical_loss": 3.6029977860709588,
"tokens_seen": 1145176064
},
{
"epoch": 0.35,
"objective/train/docs_used": 652765,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7057530879974365,
"objective/train/theoretical_loss": 3.6029786527541123,
"objective/train/tokens_used": 1165701600,
"theoretical_loss": 3.6029786527541123,
"tokens_seen": 1145241600
},
{
"epoch": 0.35,
"learning_rate": 0.00032981945837512537,
"loss": 2.8447,
"theoretical_loss": 3.6029786527541123,
"tokens_seen": 1145241600
},
{
"epoch": 0.35,
"learning_rate": 0.00032980942828485455,
"loss": 2.7223,
"theoretical_loss": 3.6029595208386818,
"tokens_seen": 1145307136
},
{
"epoch": 0.35,
"learning_rate": 0.0003297993981945838,
"loss": 2.7242,
"theoretical_loss": 3.6029403903244845,
"tokens_seen": 1145372672
},
{
"epoch": 0.35,
"learning_rate": 0.0003297893681043129,
"loss": 2.7495,
"theoretical_loss": 3.602921261211338,
"tokens_seen": 1145438208
},
{
"epoch": 0.35,
"learning_rate": 0.00032977933801404215,
"loss": 2.8466,
"theoretical_loss": 3.6029021334990596,
"tokens_seen": 1145503744
},
{
"epoch": 0.35,
"learning_rate": 0.0003297693079237713,
"loss": 2.6977,
"theoretical_loss": 3.602883007187466,
"tokens_seen": 1145569280
},
{
"epoch": 0.35,
"learning_rate": 0.0003297592778335005,
"loss": 2.7659,
"theoretical_loss": 3.6028638822763748,
"tokens_seen": 1145634816
},
{
"epoch": 0.35,
"learning_rate": 0.0003297492477432297,
"loss": 2.7938,
"theoretical_loss": 3.6028447587656034,
"tokens_seen": 1145700352
},
{
"epoch": 0.35,
"learning_rate": 0.00032973921765295887,
"loss": 2.5563,
"theoretical_loss": 3.60282563665497,
"tokens_seen": 1145765888
},
{
"epoch": 0.35,
"learning_rate": 0.00032972918756268805,
"loss": 2.6649,
"theoretical_loss": 3.60280651594429,
"tokens_seen": 1145831424
},
{
"epoch": 0.35,
"learning_rate": 0.0003297191574724173,
"loss": 2.618,
"theoretical_loss": 3.6027873966333823,
"tokens_seen": 1145896960
},
{
"epoch": 0.35,
"learning_rate": 0.0003297091273821464,
"loss": 2.5961,
"theoretical_loss": 3.6027682787220643,
"tokens_seen": 1145962496
},
{
"epoch": 0.35,
"learning_rate": 0.00032969909729187565,
"loss": 2.7279,
"theoretical_loss": 3.6027491622101535,
"tokens_seen": 1146028032
},
{
"epoch": 0.35,
"learning_rate": 0.0003296890672016048,
"loss": 2.7788,
"theoretical_loss": 3.602730047097467,
"tokens_seen": 1146093568
},
{
"epoch": 0.35,
"learning_rate": 0.000329679037111334,
"loss": 2.8394,
"theoretical_loss": 3.6027109333838228,
"tokens_seen": 1146159104
},
{
"epoch": 0.35,
"learning_rate": 0.0003296690070210632,
"loss": 2.6347,
"theoretical_loss": 3.602691821069038,
"tokens_seen": 1146224640
},
{
"epoch": 0.35,
"learning_rate": 0.0003296589769307924,
"loss": 2.7234,
"theoretical_loss": 3.6026727101529312,
"tokens_seen": 1146290176
},
{
"epoch": 0.35,
"learning_rate": 0.00032964894684052155,
"loss": 2.8187,
"theoretical_loss": 3.602653600635319,
"tokens_seen": 1146355712
},
{
"epoch": 0.35,
"learning_rate": 0.00032963891675025074,
"loss": 2.8054,
"theoretical_loss": 3.6026344925160196,
"tokens_seen": 1146421248
},
{
"epoch": 0.35,
"learning_rate": 0.0003296288866599799,
"loss": 2.5507,
"theoretical_loss": 3.602615385794851,
"tokens_seen": 1146486784
},
{
"epoch": 0.35,
"learning_rate": 0.00032961885656970915,
"loss": 2.5361,
"theoretical_loss": 3.6025962804716305,
"tokens_seen": 1146552320
},
{
"epoch": 0.35,
"learning_rate": 0.0003296088264794383,
"loss": 2.6872,
"theoretical_loss": 3.6025771765461765,
"tokens_seen": 1146617856
},
{
"epoch": 0.35,
"learning_rate": 0.0003295987963891675,
"loss": 2.8442,
"theoretical_loss": 3.6025580740183063,
"tokens_seen": 1146683392
},
{
"epoch": 0.35,
"learning_rate": 0.0003295887662988967,
"loss": 2.7728,
"theoretical_loss": 3.602538972887838,
"tokens_seen": 1146748928
},
{
"epoch": 0.35,
"learning_rate": 0.0003295787362086259,
"loss": 2.7887,
"theoretical_loss": 3.6025198731545895,
"tokens_seen": 1146814464
},
{
"debugging/Self-BLEU-5": 0.293406318468253,
"debugging/distinct-1-grams": 0.8098961360359294,
"debugging/distinct-2-grams": 0.9639839252742478,
"debugging/entropy-1-grams": 5.203556100794769,
"debugging/entropy-2-grams": 5.919939470517111,
"debugging/length": 412.8333333333333,
"debugging/num_segments": 6,
"epoch": 0.35,
"objective/train/docs_used": 652965,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6444215774536133,
"objective/train/theoretical_loss": 3.602500774818379,
"objective/train/tokens_used": 1167340000,
"theoretical_loss": 3.602500774818379,
"tokens_seen": 1146880000
},
{
"epoch": 0.35,
"learning_rate": 0.0003295687061183551,
"loss": 2.4763,
"theoretical_loss": 3.602500774818379,
"tokens_seen": 1146880000
},
{
"epoch": 0.35,
"learning_rate": 0.00032955867602808424,
"loss": 2.9422,
"theoretical_loss": 3.602481677879024,
"tokens_seen": 1146945536
},
{
"epoch": 0.35,
"learning_rate": 0.0003295486459378135,
"loss": 2.7119,
"theoretical_loss": 3.6024625823363428,
"tokens_seen": 1147011072
},
{
"epoch": 0.35,
"learning_rate": 0.00032953861584754265,
"loss": 3.1455,
"theoretical_loss": 3.602443488190154,
"tokens_seen": 1147076608
},
{
"epoch": 0.35,
"learning_rate": 0.00032952858575727184,
"loss": 2.6335,
"theoretical_loss": 3.602424395440275,
"tokens_seen": 1147142144
},
{
"epoch": 0.35,
"learning_rate": 0.000329518555667001,
"loss": 2.6786,
"theoretical_loss": 3.602405304086524,
"tokens_seen": 1147207680
},
{
"epoch": 0.35,
"learning_rate": 0.0003295085255767302,
"loss": 2.9652,
"theoretical_loss": 3.6023862141287193,
"tokens_seen": 1147273216
},
{
"epoch": 0.35,
"learning_rate": 0.0003294984954864594,
"loss": 3.0246,
"theoretical_loss": 3.6023671255666785,
"tokens_seen": 1147338752
},
{
"epoch": 0.35,
"learning_rate": 0.0003294884653961886,
"loss": 2.8174,
"theoretical_loss": 3.602348038400221,
"tokens_seen": 1147404288
},
{
"epoch": 0.35,
"learning_rate": 0.00032947843530591774,
"loss": 2.8229,
"theoretical_loss": 3.6023289526291644,
"tokens_seen": 1147469824
},
{
"epoch": 0.35,
"learning_rate": 0.000329468405215647,
"loss": 2.7311,
"theoretical_loss": 3.602309868253327,
"tokens_seen": 1147535360
},
{
"epoch": 0.35,
"learning_rate": 0.0003294583751253761,
"loss": 2.6722,
"theoretical_loss": 3.602290785272527,
"tokens_seen": 1147600896
},
{
"epoch": 0.35,
"learning_rate": 0.00032944834503510534,
"loss": 2.7077,
"theoretical_loss": 3.6022717036865832,
"tokens_seen": 1147666432
},
{
"epoch": 0.35,
"learning_rate": 0.0003294383149448345,
"loss": 2.6058,
"theoretical_loss": 3.6022526234953136,
"tokens_seen": 1147731968
},
{
"epoch": 0.35,
"learning_rate": 0.0003294282848545637,
"loss": 3.1203,
"theoretical_loss": 3.602233544698537,
"tokens_seen": 1147797504
},
{
"epoch": 0.35,
"learning_rate": 0.0003294182547642929,
"loss": 2.831,
"theoretical_loss": 3.6022144672960716,
"tokens_seen": 1147863040
},
{
"epoch": 0.35,
"learning_rate": 0.0003294082246740221,
"loss": 2.8054,
"theoretical_loss": 3.602195391287736,
"tokens_seen": 1147928576
},
{
"epoch": 0.35,
"learning_rate": 0.00032939819458375124,
"loss": 2.5191,
"theoretical_loss": 3.602176316673349,
"tokens_seen": 1147994112
},
{
"epoch": 0.35,
"learning_rate": 0.0003293881644934805,
"loss": 2.7671,
"theoretical_loss": 3.6021572434527283,
"tokens_seen": 1148059648
},
{
"epoch": 0.35,
"learning_rate": 0.0003293781344032096,
"loss": 2.6454,
"theoretical_loss": 3.6021381716256933,
"tokens_seen": 1148125184
},
{
"epoch": 0.35,
"learning_rate": 0.00032936810431293884,
"loss": 2.9713,
"theoretical_loss": 3.6021191011920624,
"tokens_seen": 1148190720
},
{
"epoch": 0.35,
"learning_rate": 0.000329358074222668,
"loss": 2.4402,
"theoretical_loss": 3.6021000321516543,
"tokens_seen": 1148256256
},
{
"epoch": 0.35,
"learning_rate": 0.0003293480441323972,
"loss": 2.8817,
"theoretical_loss": 3.602080964504288,
"tokens_seen": 1148321792
},
{
"epoch": 0.35,
"learning_rate": 0.0003293380140421264,
"loss": 2.6756,
"theoretical_loss": 3.602061898249781,
"tokens_seen": 1148387328
},
{
"epoch": 0.35,
"learning_rate": 0.00032932798395185557,
"loss": 2.9151,
"theoretical_loss": 3.602042833387954,
"tokens_seen": 1148452864
},
{
"epoch": 0.35,
"objective/train/docs_used": 654364,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.505870819091797,
"objective/train/theoretical_loss": 3.602023769918625,
"objective/train/tokens_used": 1168978400,
"theoretical_loss": 3.602023769918625,
"tokens_seen": 1148518400
},
{
"epoch": 0.35,
"learning_rate": 0.00032931795386158475,
"loss": 2.6687,
"theoretical_loss": 3.602023769918625,
"tokens_seen": 1148518400
},
{
"epoch": 0.35,
"learning_rate": 0.000329307923771314,
"loss": 2.8978,
"theoretical_loss": 3.602004707841612,
"tokens_seen": 1148583936
},
{
"epoch": 0.35,
"learning_rate": 0.0003292978936810431,
"loss": 2.7754,
"theoretical_loss": 3.6019856471567344,
"tokens_seen": 1148649472
},
{
"epoch": 0.35,
"learning_rate": 0.00032928786359077235,
"loss": 2.8878,
"theoretical_loss": 3.601966587863812,
"tokens_seen": 1148715008
},
{
"epoch": 0.35,
"learning_rate": 0.0003292778335005015,
"loss": 2.7819,
"theoretical_loss": 3.601947529962662,
"tokens_seen": 1148780544
},
{
"epoch": 0.35,
"learning_rate": 0.0003292678034102307,
"loss": 2.6523,
"theoretical_loss": 3.6019284734531056,
"tokens_seen": 1148846080
},
{
"epoch": 0.35,
"learning_rate": 0.0003292577733199599,
"loss": 2.7522,
"theoretical_loss": 3.6019094183349596,
"tokens_seen": 1148911616
},
{
"epoch": 0.35,
"learning_rate": 0.00032924774322968907,
"loss": 2.9312,
"theoretical_loss": 3.6018903646080447,
"tokens_seen": 1148977152
},
{
"epoch": 0.35,
"learning_rate": 0.00032923771313941825,
"loss": 2.7561,
"theoretical_loss": 3.601871312272179,
"tokens_seen": 1149042688
},
{
"epoch": 0.35,
"learning_rate": 0.0003292276830491475,
"loss": 2.5657,
"theoretical_loss": 3.601852261327182,
"tokens_seen": 1149108224
},
{
"epoch": 0.35,
"learning_rate": 0.0003292176529588766,
"loss": 2.7096,
"theoretical_loss": 3.601833211772873,
"tokens_seen": 1149173760
},
{
"epoch": 0.35,
"learning_rate": 0.00032920762286860585,
"loss": 2.7541,
"theoretical_loss": 3.601814163609071,
"tokens_seen": 1149239296
},
{
"epoch": 0.35,
"learning_rate": 0.000329197592778335,
"loss": 2.7013,
"theoretical_loss": 3.6017951168355946,
"tokens_seen": 1149304832
},
{
"epoch": 0.35,
"learning_rate": 0.0003291875626880642,
"loss": 2.7963,
"theoretical_loss": 3.6017760714522637,
"tokens_seen": 1149370368
},
{
"epoch": 0.35,
"learning_rate": 0.0003291775325977934,
"loss": 2.667,
"theoretical_loss": 3.601757027458898,
"tokens_seen": 1149435904
},
{
"epoch": 0.35,
"learning_rate": 0.0003291675025075226,
"loss": 2.6304,
"theoretical_loss": 3.6017379848553164,
"tokens_seen": 1149501440
},
{
"epoch": 0.35,
"learning_rate": 0.00032915747241725175,
"loss": 2.7482,
"theoretical_loss": 3.6017189436413375,
"tokens_seen": 1149566976
},
{
"epoch": 0.35,
"learning_rate": 0.00032914744232698094,
"loss": 2.9272,
"theoretical_loss": 3.6016999038167814,
"tokens_seen": 1149632512
},
{
"epoch": 0.35,
"learning_rate": 0.0003291374122367101,
"loss": 2.9623,
"theoretical_loss": 3.601680865381468,
"tokens_seen": 1149698048
},
{
"epoch": 0.35,
"learning_rate": 0.00032912738214643935,
"loss": 2.7524,
"theoretical_loss": 3.6016618283352164,
"tokens_seen": 1149763584
},
{
"epoch": 0.35,
"learning_rate": 0.0003291173520561685,
"loss": 2.802,
"theoretical_loss": 3.601642792677845,
"tokens_seen": 1149829120
},
{
"epoch": 0.35,
"learning_rate": 0.0003291073219658977,
"loss": 2.6867,
"theoretical_loss": 3.6016237584091755,
"tokens_seen": 1149894656
},
{
"epoch": 0.35,
"learning_rate": 0.0003290972918756269,
"loss": 2.7663,
"theoretical_loss": 3.6016047255290253,
"tokens_seen": 1149960192
},
{
"epoch": 0.35,
"learning_rate": 0.0003290872617853561,
"loss": 2.7962,
"theoretical_loss": 3.601585694037215,
"tokens_seen": 1150025728
},
{
"epoch": 0.35,
"learning_rate": 0.00032907723169508526,
"loss": 3.0419,
"theoretical_loss": 3.601566663933564,
"tokens_seen": 1150091264
},
{
"epoch": 0.35,
"objective/train/docs_used": 654951,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.669060230255127,
"objective/train/theoretical_loss": 3.601547635217892,
"objective/train/tokens_used": 1170616800,
"theoretical_loss": 3.601547635217892,
"tokens_seen": 1150156800
},
{
"epoch": 0.35,
"learning_rate": 0.00032906720160481444,
"loss": 2.6868,
"theoretical_loss": 3.601547635217892,
"tokens_seen": 1150156800
},
{
"epoch": 0.35,
"learning_rate": 0.0003290571715145436,
"loss": 2.577,
"theoretical_loss": 3.6015286078900193,
"tokens_seen": 1150222336
},
{
"epoch": 0.35,
"learning_rate": 0.00032904714142427286,
"loss": 2.8514,
"theoretical_loss": 3.601509581949765,
"tokens_seen": 1150287872
},
{
"epoch": 0.35,
"learning_rate": 0.000329037111334002,
"loss": 2.5885,
"theoretical_loss": 3.6014905573969482,
"tokens_seen": 1150353408
},
{
"epoch": 0.35,
"learning_rate": 0.0003290270812437312,
"loss": 2.9213,
"theoretical_loss": 3.60147153423139,
"tokens_seen": 1150418944
},
{
"epoch": 0.35,
"learning_rate": 0.00032901705115346034,
"loss": 2.7093,
"theoretical_loss": 3.6014525124529095,
"tokens_seen": 1150484480
},
{
"epoch": 0.35,
"learning_rate": 0.0003290070210631896,
"loss": 2.7337,
"theoretical_loss": 3.601433492061326,
"tokens_seen": 1150550016
},
{
"epoch": 0.35,
"learning_rate": 0.00032899699097291876,
"loss": 2.6119,
"theoretical_loss": 3.6014144730564612,
"tokens_seen": 1150615552
},
{
"epoch": 0.35,
"learning_rate": 0.00032898696088264794,
"loss": 2.831,
"theoretical_loss": 3.6013954554381336,
"tokens_seen": 1150681088
},
{
"epoch": 0.35,
"learning_rate": 0.0003289769307923771,
"loss": 2.7987,
"theoretical_loss": 3.601376439206163,
"tokens_seen": 1150746624
},
{
"epoch": 0.35,
"learning_rate": 0.0003289669007021063,
"loss": 2.6857,
"theoretical_loss": 3.60135742436037,
"tokens_seen": 1150812160
},
{
"epoch": 0.35,
"learning_rate": 0.0003289568706118355,
"loss": 2.9048,
"theoretical_loss": 3.6013384109005746,
"tokens_seen": 1150877696
},
{
"epoch": 0.35,
"learning_rate": 0.0003289468405215647,
"loss": 2.7318,
"theoretical_loss": 3.601319398826597,
"tokens_seen": 1150943232
},
{
"epoch": 0.35,
"learning_rate": 0.00032893681043129385,
"loss": 2.5255,
"theoretical_loss": 3.601300388138257,
"tokens_seen": 1151008768
},
{
"epoch": 0.35,
"learning_rate": 0.0003289267803410231,
"loss": 2.6708,
"theoretical_loss": 3.601281378835374,
"tokens_seen": 1151074304
},
{
"epoch": 0.35,
"learning_rate": 0.00032891675025075226,
"loss": 2.6849,
"theoretical_loss": 3.6012623709177696,
"tokens_seen": 1151139840
},
{
"epoch": 0.35,
"learning_rate": 0.00032890672016048145,
"loss": 2.7282,
"theoretical_loss": 3.6012433643852626,
"tokens_seen": 1151205376
},
{
"epoch": 0.35,
"learning_rate": 0.0003288966900702106,
"loss": 2.5884,
"theoretical_loss": 3.6012243592376745,
"tokens_seen": 1151270912
},
{
"epoch": 0.35,
"learning_rate": 0.0003288866599799398,
"loss": 2.5943,
"theoretical_loss": 3.601205355474825,
"tokens_seen": 1151336448
},
{
"epoch": 0.35,
"learning_rate": 0.000328876629889669,
"loss": 2.8828,
"theoretical_loss": 3.6011863530965345,
"tokens_seen": 1151401984
},
{
"epoch": 0.35,
"learning_rate": 0.0003288665997993982,
"loss": 2.5207,
"theoretical_loss": 3.601167352102623,
"tokens_seen": 1151467520
},
{
"epoch": 0.35,
"learning_rate": 0.00032885656970912735,
"loss": 2.6787,
"theoretical_loss": 3.6011483524929107,
"tokens_seen": 1151533056
},
{
"epoch": 0.35,
"learning_rate": 0.0003288465396188566,
"loss": 2.7742,
"theoretical_loss": 3.601129354267219,
"tokens_seen": 1151598592
},
{
"epoch": 0.35,
"learning_rate": 0.0003288365095285857,
"loss": 2.9551,
"theoretical_loss": 3.6011103574253673,
"tokens_seen": 1151664128
},
{
"epoch": 0.35,
"learning_rate": 0.00032882647943831495,
"loss": 2.6633,
"theoretical_loss": 3.6010913619671765,
"tokens_seen": 1151729664
},
{
"epoch": 0.35,
"objective/train/docs_used": 656079,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.636674642562866,
"objective/train/theoretical_loss": 3.601072367892467,
"objective/train/tokens_used": 1172255200,
"theoretical_loss": 3.601072367892467,
"tokens_seen": 1151795200
},
{
"epoch": 0.35,
"learning_rate": 0.0003288164493480442,
"loss": 2.8479,
"theoretical_loss": 3.601072367892467,
"tokens_seen": 1151795200
},
{
"epoch": 0.35,
"learning_rate": 0.0003288064192577733,
"loss": 2.7806,
"theoretical_loss": 3.6010533752010594,
"tokens_seen": 1151860736
},
{
"epoch": 0.35,
"learning_rate": 0.00032879638916750255,
"loss": 2.6194,
"theoretical_loss": 3.6010343838927743,
"tokens_seen": 1151926272
},
{
"epoch": 0.35,
"learning_rate": 0.0003287863590772317,
"loss": 2.7566,
"theoretical_loss": 3.6010153939674323,
"tokens_seen": 1151991808
},
{
"epoch": 0.35,
"learning_rate": 0.0003287763289869609,
"loss": 2.9479,
"theoretical_loss": 3.600996405424854,
"tokens_seen": 1152057344
},
{
"epoch": 0.35,
"learning_rate": 0.0003287662988966901,
"loss": 2.6796,
"theoretical_loss": 3.6009774182648595,
"tokens_seen": 1152122880
},
{
"epoch": 0.35,
"learning_rate": 0.00032875626880641927,
"loss": 2.8101,
"theoretical_loss": 3.6009584324872703,
"tokens_seen": 1152188416
},
{
"epoch": 0.35,
"learning_rate": 0.00032874623871614845,
"loss": 2.6331,
"theoretical_loss": 3.600939448091907,
"tokens_seen": 1152253952
},
{
"epoch": 0.35,
"learning_rate": 0.0003287362086258777,
"loss": 2.674,
"theoretical_loss": 3.60092046507859,
"tokens_seen": 1152319488
},
{
"epoch": 0.35,
"learning_rate": 0.0003287261785356068,
"loss": 2.7408,
"theoretical_loss": 3.6009014834471404,
"tokens_seen": 1152385024
},
{
"epoch": 0.35,
"learning_rate": 0.00032871614844533605,
"loss": 2.7257,
"theoretical_loss": 3.6008825031973792,
"tokens_seen": 1152450560
},
{
"epoch": 0.35,
"learning_rate": 0.0003287061183550652,
"loss": 2.7157,
"theoretical_loss": 3.6008635243291263,
"tokens_seen": 1152516096
},
{
"epoch": 0.35,
"learning_rate": 0.0003286960882647944,
"loss": 2.7719,
"theoretical_loss": 3.600844546842204,
"tokens_seen": 1152581632
},
{
"epoch": 0.35,
"learning_rate": 0.0003286860581745236,
"loss": 2.9268,
"theoretical_loss": 3.6008255707364323,
"tokens_seen": 1152647168
},
{
"epoch": 0.35,
"learning_rate": 0.0003286760280842528,
"loss": 2.723,
"theoretical_loss": 3.600806596011632,
"tokens_seen": 1152712704
},
{
"epoch": 0.35,
"learning_rate": 0.00032866599799398195,
"loss": 2.8554,
"theoretical_loss": 3.6007876226676245,
"tokens_seen": 1152778240
},
{
"epoch": 0.35,
"learning_rate": 0.00032865596790371114,
"loss": 2.8454,
"theoretical_loss": 3.6007686507042314,
"tokens_seen": 1152843776
},
{
"epoch": 0.35,
"learning_rate": 0.0003286459378134403,
"loss": 2.7729,
"theoretical_loss": 3.6007496801212726,
"tokens_seen": 1152909312
},
{
"epoch": 0.35,
"learning_rate": 0.00032863590772316955,
"loss": 2.6791,
"theoretical_loss": 3.60073071091857,
"tokens_seen": 1152974848
},
{
"epoch": 0.35,
"learning_rate": 0.0003286258776328987,
"loss": 2.6747,
"theoretical_loss": 3.6007117430959443,
"tokens_seen": 1153040384
},
{
"epoch": 0.35,
"learning_rate": 0.0003286158475426279,
"loss": 2.7216,
"theoretical_loss": 3.6006927766532173,
"tokens_seen": 1153105920
},
{
"epoch": 0.35,
"learning_rate": 0.0003286058174523571,
"loss": 2.8416,
"theoretical_loss": 3.6006738115902097,
"tokens_seen": 1153171456
},
{
"epoch": 0.35,
"learning_rate": 0.0003285957873620863,
"loss": 2.8568,
"theoretical_loss": 3.600654847906742,
"tokens_seen": 1153236992
},
{
"epoch": 0.35,
"learning_rate": 0.00032858575727181546,
"loss": 2.6365,
"theoretical_loss": 3.600635885602637,
"tokens_seen": 1153302528
},
{
"epoch": 0.35,
"learning_rate": 0.00032857572718154464,
"loss": 2.8981,
"theoretical_loss": 3.6006169246777153,
"tokens_seen": 1153368064
},
{
"epoch": 0.35,
"objective/train/docs_used": 656588,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7284483909606934,
"objective/train/theoretical_loss": 3.6005979651317976,
"objective/train/tokens_used": 1173893600,
"theoretical_loss": 3.6005979651317976,
"tokens_seen": 1153433600
},
{
"epoch": 0.35,
"learning_rate": 0.0003285656970912738,
"loss": 2.8495,
"theoretical_loss": 3.6005979651317976,
"tokens_seen": 1153433600
},
{
"epoch": 0.35,
"learning_rate": 0.00032855566700100306,
"loss": 2.8056,
"theoretical_loss": 3.6005790069647063,
"tokens_seen": 1153499136
},
{
"epoch": 0.35,
"learning_rate": 0.0003285456369107322,
"loss": 2.8479,
"theoretical_loss": 3.600560050176262,
"tokens_seen": 1153564672
},
{
"epoch": 0.35,
"learning_rate": 0.0003285356068204614,
"loss": 2.9824,
"theoretical_loss": 3.6005410947662866,
"tokens_seen": 1153630208
},
{
"epoch": 0.35,
"learning_rate": 0.00032852557673019054,
"loss": 2.9775,
"theoretical_loss": 3.6005221407346015,
"tokens_seen": 1153695744
},
{
"epoch": 0.35,
"learning_rate": 0.0003285155466399198,
"loss": 2.7088,
"theoretical_loss": 3.600503188081028,
"tokens_seen": 1153761280
},
{
"epoch": 0.35,
"learning_rate": 0.00032850551654964896,
"loss": 2.7298,
"theoretical_loss": 3.6004842368053875,
"tokens_seen": 1153826816
},
{
"epoch": 0.35,
"learning_rate": 0.00032849548645937814,
"loss": 2.8762,
"theoretical_loss": 3.600465286907502,
"tokens_seen": 1153892352
},
{
"epoch": 0.35,
"learning_rate": 0.0003284854563691073,
"loss": 2.7066,
"theoretical_loss": 3.6004463383871927,
"tokens_seen": 1153957888
},
{
"epoch": 0.35,
"learning_rate": 0.0003284754262788365,
"loss": 2.8963,
"theoretical_loss": 3.600427391244282,
"tokens_seen": 1154023424
},
{
"epoch": 0.35,
"learning_rate": 0.0003284653961885657,
"loss": 2.7236,
"theoretical_loss": 3.6004084454785907,
"tokens_seen": 1154088960
},
{
"epoch": 0.35,
"learning_rate": 0.0003284553660982949,
"loss": 2.7272,
"theoretical_loss": 3.600389501089941,
"tokens_seen": 1154154496
},
{
"epoch": 0.35,
"learning_rate": 0.00032844533600802405,
"loss": 2.8383,
"theoretical_loss": 3.600370558078154,
"tokens_seen": 1154220032
},
{
"epoch": 0.35,
"learning_rate": 0.0003284353059177533,
"loss": 2.8055,
"theoretical_loss": 3.600351616443052,
"tokens_seen": 1154285568
},
{
"epoch": 0.35,
"learning_rate": 0.00032842527582748246,
"loss": 2.6421,
"theoretical_loss": 3.6003326761844567,
"tokens_seen": 1154351104
},
{
"epoch": 0.35,
"learning_rate": 0.00032841524573721165,
"loss": 2.8575,
"theoretical_loss": 3.60031373730219,
"tokens_seen": 1154416640
},
{
"epoch": 0.35,
"learning_rate": 0.0003284052156469408,
"loss": 2.7099,
"theoretical_loss": 3.6002947997960737,
"tokens_seen": 1154482176
},
{
"epoch": 0.35,
"learning_rate": 0.00032839518555667,
"loss": 2.5372,
"theoretical_loss": 3.6002758636659293,
"tokens_seen": 1154547712
},
{
"epoch": 0.35,
"learning_rate": 0.0003283851554663992,
"loss": 2.8912,
"theoretical_loss": 3.6002569289115796,
"tokens_seen": 1154613248
},
{
"epoch": 0.35,
"learning_rate": 0.0003283751253761284,
"loss": 2.8892,
"theoretical_loss": 3.6002379955328454,
"tokens_seen": 1154678784
},
{
"epoch": 0.35,
"learning_rate": 0.00032836509528585755,
"loss": 2.5542,
"theoretical_loss": 3.6002190635295497,
"tokens_seen": 1154744320
},
{
"epoch": 0.35,
"learning_rate": 0.0003283550651955868,
"loss": 2.7433,
"theoretical_loss": 3.600200132901514,
"tokens_seen": 1154809856
},
{
"epoch": 0.35,
"learning_rate": 0.0003283450351053159,
"loss": 2.8713,
"theoretical_loss": 3.6001812036485608,
"tokens_seen": 1154875392
},
{
"epoch": 0.35,
"learning_rate": 0.00032833500501504515,
"loss": 2.6652,
"theoretical_loss": 3.600162275770512,
"tokens_seen": 1154940928
},
{
"epoch": 0.35,
"learning_rate": 0.00032832497492477433,
"loss": 2.775,
"theoretical_loss": 3.6001433492671895,
"tokens_seen": 1155006464
},
{
"epoch": 0.35,
"objective/train/docs_used": 657761,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6598703861236572,
"objective/train/theoretical_loss": 3.600124424138415,
"objective/train/tokens_used": 1175532000,
"theoretical_loss": 3.600124424138415,
"tokens_seen": 1155072000
},
{
"epoch": 0.35,
"learning_rate": 0.0003283149448345035,
"loss": 2.6972,
"theoretical_loss": 3.600124424138415,
"tokens_seen": 1155072000
},
{
"epoch": 0.35,
"learning_rate": 0.0003283049147442327,
"loss": 2.8308,
"theoretical_loss": 3.600105500384012,
"tokens_seen": 1155137536
},
{
"epoch": 0.35,
"learning_rate": 0.0003282948846539619,
"loss": 2.8366,
"theoretical_loss": 3.6000865780038023,
"tokens_seen": 1155203072
},
{
"epoch": 0.35,
"learning_rate": 0.00032828485456369105,
"loss": 2.492,
"theoretical_loss": 3.600067656997607,
"tokens_seen": 1155268608
},
{
"epoch": 0.35,
"learning_rate": 0.0003282748244734203,
"loss": 3.0027,
"theoretical_loss": 3.60004873736525,
"tokens_seen": 1155334144
},
{
"epoch": 0.35,
"learning_rate": 0.0003282647943831494,
"loss": 2.8409,
"theoretical_loss": 3.6000298191065525,
"tokens_seen": 1155399680
},
{
"epoch": 0.35,
"learning_rate": 0.00032825476429287865,
"loss": 2.7812,
"theoretical_loss": 3.600010902221337,
"tokens_seen": 1155465216
},
{
"epoch": 0.35,
"learning_rate": 0.00032824473420260783,
"loss": 2.7435,
"theoretical_loss": 3.5999919867094268,
"tokens_seen": 1155530752
},
{
"epoch": 0.35,
"learning_rate": 0.000328234704112337,
"loss": 2.5861,
"theoretical_loss": 3.599973072570643,
"tokens_seen": 1155596288
},
{
"epoch": 0.35,
"learning_rate": 0.0003282246740220662,
"loss": 2.8676,
"theoretical_loss": 3.5999541598048097,
"tokens_seen": 1155661824
},
{
"epoch": 0.35,
"learning_rate": 0.0003282146439317954,
"loss": 2.7459,
"theoretical_loss": 3.5999352484117475,
"tokens_seen": 1155727360
},
{
"epoch": 0.35,
"learning_rate": 0.00032820461384152456,
"loss": 2.7908,
"theoretical_loss": 3.59991633839128,
"tokens_seen": 1155792896
},
{
"epoch": 0.35,
"learning_rate": 0.0003281945837512538,
"loss": 2.6817,
"theoretical_loss": 3.59989742974323,
"tokens_seen": 1155858432
},
{
"epoch": 0.35,
"learning_rate": 0.0003281845536609829,
"loss": 2.7517,
"theoretical_loss": 3.599878522467419,
"tokens_seen": 1155923968
},
{
"epoch": 0.35,
"learning_rate": 0.00032817452357071215,
"loss": 2.8414,
"theoretical_loss": 3.5998596165636707,
"tokens_seen": 1155989504
},
{
"epoch": 0.35,
"learning_rate": 0.0003281644934804413,
"loss": 2.6255,
"theoretical_loss": 3.599840712031807,
"tokens_seen": 1156055040
},
{
"epoch": 0.35,
"learning_rate": 0.0003281544633901705,
"loss": 2.5766,
"theoretical_loss": 3.599821808871652,
"tokens_seen": 1156120576
},
{
"epoch": 0.35,
"learning_rate": 0.0003281444332998997,
"loss": 2.7953,
"theoretical_loss": 3.5998029070830264,
"tokens_seen": 1156186112
},
{
"epoch": 0.35,
"learning_rate": 0.0003281344032096289,
"loss": 2.7626,
"theoretical_loss": 3.5997840066657543,
"tokens_seen": 1156251648
},
{
"epoch": 0.35,
"learning_rate": 0.00032812437311935806,
"loss": 2.5991,
"theoretical_loss": 3.5997651076196577,
"tokens_seen": 1156317184
},
{
"epoch": 0.35,
"learning_rate": 0.0003281143430290873,
"loss": 2.6398,
"theoretical_loss": 3.59974620994456,
"tokens_seen": 1156382720
},
{
"epoch": 0.35,
"learning_rate": 0.0003281043129388164,
"loss": 2.6518,
"theoretical_loss": 3.599727313640284,
"tokens_seen": 1156448256
},
{
"epoch": 0.35,
"learning_rate": 0.00032809428284854566,
"loss": 2.6225,
"theoretical_loss": 3.599708418706653,
"tokens_seen": 1156513792
},
{
"epoch": 0.35,
"learning_rate": 0.0003280842527582748,
"loss": 2.6727,
"theoretical_loss": 3.599689525143489,
"tokens_seen": 1156579328
},
{
"epoch": 0.35,
"learning_rate": 0.000328074222668004,
"loss": 2.5532,
"theoretical_loss": 3.599670632950615,
"tokens_seen": 1156644864
},
{
"epoch": 0.35,
"objective/train/docs_used": 658495,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1720292568206787,
"objective/train/theoretical_loss": 3.599651742127855,
"objective/train/tokens_used": 1177170400,
"theoretical_loss": 3.599651742127855,
"tokens_seen": 1156710400
},
{
"epoch": 0.35,
"learning_rate": 0.00032806419257773326,
"loss": 2.8054,
"theoretical_loss": 3.599651742127855,
"tokens_seen": 1156710400
},
{
"epoch": 0.35,
"learning_rate": 0.0003280541624874624,
"loss": 2.5312,
"theoretical_loss": 3.5996328526750307,
"tokens_seen": 1156775936
},
{
"epoch": 0.35,
"learning_rate": 0.0003280441323971916,
"loss": 2.8534,
"theoretical_loss": 3.5996139645919656,
"tokens_seen": 1156841472
},
{
"epoch": 0.35,
"learning_rate": 0.00032803410230692074,
"loss": 2.9213,
"theoretical_loss": 3.5995950778784835,
"tokens_seen": 1156907008
},
{
"epoch": 0.35,
"learning_rate": 0.00032802407221665,
"loss": 2.8143,
"theoretical_loss": 3.599576192534407,
"tokens_seen": 1156972544
},
{
"epoch": 0.35,
"learning_rate": 0.00032801404212637916,
"loss": 2.7532,
"theoretical_loss": 3.5995573085595587,
"tokens_seen": 1157038080
},
{
"epoch": 0.35,
"learning_rate": 0.00032800401203610834,
"loss": 2.6884,
"theoretical_loss": 3.599538425953763,
"tokens_seen": 1157103616
},
{
"epoch": 0.35,
"learning_rate": 0.0003279939819458375,
"loss": 2.7899,
"theoretical_loss": 3.5995195447168418,
"tokens_seen": 1157169152
},
{
"epoch": 0.35,
"learning_rate": 0.0003279839518555667,
"loss": 2.6852,
"theoretical_loss": 3.5995006648486196,
"tokens_seen": 1157234688
},
{
"epoch": 0.35,
"learning_rate": 0.0003279739217652959,
"loss": 2.7334,
"theoretical_loss": 3.5994817863489184,
"tokens_seen": 1157300224
},
{
"epoch": 0.35,
"learning_rate": 0.0003279638916750251,
"loss": 2.7452,
"theoretical_loss": 3.5994629092175625,
"tokens_seen": 1157365760
},
{
"epoch": 0.35,
"learning_rate": 0.00032795386158475425,
"loss": 2.8794,
"theoretical_loss": 3.5994440334543754,
"tokens_seen": 1157431296
},
{
"epoch": 0.35,
"learning_rate": 0.0003279438314944835,
"loss": 2.7654,
"theoretical_loss": 3.5994251590591793,
"tokens_seen": 1157496832
},
{
"epoch": 0.35,
"learning_rate": 0.00032793380140421266,
"loss": 2.9068,
"theoretical_loss": 3.5994062860317984,
"tokens_seen": 1157562368
},
{
"epoch": 0.35,
"learning_rate": 0.00032792377131394185,
"loss": 2.6298,
"theoretical_loss": 3.5993874143720563,
"tokens_seen": 1157627904
},
{
"epoch": 0.35,
"learning_rate": 0.000327913741223671,
"loss": 2.653,
"theoretical_loss": 3.599368544079776,
"tokens_seen": 1157693440
},
{
"epoch": 0.35,
"learning_rate": 0.0003279037111334002,
"loss": 2.8192,
"theoretical_loss": 3.5993496751547815,
"tokens_seen": 1157758976
},
{
"epoch": 0.35,
"learning_rate": 0.0003278936810431294,
"loss": 2.7284,
"theoretical_loss": 3.599330807596896,
"tokens_seen": 1157824512
},
{
"epoch": 0.35,
"learning_rate": 0.0003278836509528586,
"loss": 2.7489,
"theoretical_loss": 3.5993119414059427,
"tokens_seen": 1157890048
},
{
"epoch": 0.35,
"learning_rate": 0.00032787362086258775,
"loss": 2.84,
"theoretical_loss": 3.5992930765817466,
"tokens_seen": 1157955584
},
{
"epoch": 0.35,
"learning_rate": 0.000327863590772317,
"loss": 2.9812,
"theoretical_loss": 3.5992742131241293,
"tokens_seen": 1158021120
},
{
"epoch": 0.35,
"learning_rate": 0.0003278535606820461,
"loss": 2.7413,
"theoretical_loss": 3.599255351032917,
"tokens_seen": 1158086656
},
{
"epoch": 0.35,
"learning_rate": 0.00032784353059177535,
"loss": 2.8853,
"theoretical_loss": 3.5992364903079306,
"tokens_seen": 1158152192
},
{
"epoch": 0.35,
"learning_rate": 0.00032783350050150453,
"loss": 2.5142,
"theoretical_loss": 3.599217630948996,
"tokens_seen": 1158217728
},
{
"epoch": 0.35,
"learning_rate": 0.0003278234704112337,
"loss": 2.7628,
"theoretical_loss": 3.599198772955936,
"tokens_seen": 1158283264
},
{
"epoch": 0.35,
"objective/train/docs_used": 659118,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.80985426902771,
"objective/train/theoretical_loss": 3.5991799163285743,
"objective/train/tokens_used": 1178808800,
"theoretical_loss": 3.5991799163285743,
"tokens_seen": 1158348800
},
{
"epoch": 0.35,
"learning_rate": 0.0003278134403209629,
"loss": 2.7534,
"theoretical_loss": 3.5991799163285743,
"tokens_seen": 1158348800
},
{
"epoch": 0.35,
"learning_rate": 0.00032780341023069213,
"loss": 2.5129,
"theoretical_loss": 3.5991610610667353,
"tokens_seen": 1158414336
},
{
"epoch": 0.35,
"learning_rate": 0.00032779338014042125,
"loss": 2.6657,
"theoretical_loss": 3.599142207170243,
"tokens_seen": 1158479872
},
{
"epoch": 0.35,
"learning_rate": 0.0003277833500501505,
"loss": 2.8232,
"theoretical_loss": 3.5991233546389205,
"tokens_seen": 1158545408
},
{
"epoch": 0.35,
"learning_rate": 0.0003277733199598796,
"loss": 2.6716,
"theoretical_loss": 3.599104503472592,
"tokens_seen": 1158610944
},
{
"epoch": 0.35,
"learning_rate": 0.00032776328986960885,
"loss": 2.6798,
"theoretical_loss": 3.599085653671082,
"tokens_seen": 1158676480
},
{
"epoch": 0.35,
"learning_rate": 0.00032775325977933803,
"loss": 2.4626,
"theoretical_loss": 3.599066805234214,
"tokens_seen": 1158742016
},
{
"epoch": 0.35,
"learning_rate": 0.0003277432296890672,
"loss": 2.6438,
"theoretical_loss": 3.5990479581618118,
"tokens_seen": 1158807552
},
{
"epoch": 0.35,
"learning_rate": 0.0003277331995987964,
"loss": 2.9801,
"theoretical_loss": 3.5990291124537004,
"tokens_seen": 1158873088
},
{
"epoch": 0.35,
"learning_rate": 0.0003277231695085256,
"loss": 2.8312,
"theoretical_loss": 3.5990102681097027,
"tokens_seen": 1158938624
},
{
"epoch": 0.35,
"learning_rate": 0.00032771313941825476,
"loss": 2.7652,
"theoretical_loss": 3.5989914251296438,
"tokens_seen": 1159004160
},
{
"epoch": 0.35,
"learning_rate": 0.000327703109327984,
"loss": 2.6901,
"theoretical_loss": 3.5989725835133473,
"tokens_seen": 1159069696
},
{
"epoch": 0.35,
"learning_rate": 0.0003276930792377131,
"loss": 2.825,
"theoretical_loss": 3.5989537432606378,
"tokens_seen": 1159135232
},
{
"epoch": 0.35,
"learning_rate": 0.00032768304914744235,
"loss": 2.5402,
"theoretical_loss": 3.5989349043713394,
"tokens_seen": 1159200768
},
{
"epoch": 0.35,
"learning_rate": 0.0003276730190571715,
"loss": 2.7923,
"theoretical_loss": 3.598916066845276,
"tokens_seen": 1159266304
},
{
"epoch": 0.35,
"learning_rate": 0.0003276629889669007,
"loss": 2.7972,
"theoretical_loss": 3.5988972306822724,
"tokens_seen": 1159331840
},
{
"epoch": 0.35,
"learning_rate": 0.0003276529588766299,
"loss": 2.8042,
"theoretical_loss": 3.5988783958821524,
"tokens_seen": 1159397376
},
{
"epoch": 0.35,
"learning_rate": 0.0003276429287863591,
"loss": 2.7863,
"theoretical_loss": 3.598859562444741,
"tokens_seen": 1159462912
},
{
"epoch": 0.35,
"learning_rate": 0.00032763289869608826,
"loss": 2.5554,
"theoretical_loss": 3.5988407303698624,
"tokens_seen": 1159528448
},
{
"epoch": 0.35,
"learning_rate": 0.0003276228686058175,
"loss": 2.4522,
"theoretical_loss": 3.5988218996573407,
"tokens_seen": 1159593984
},
{
"epoch": 0.35,
"learning_rate": 0.0003276128385155466,
"loss": 2.8459,
"theoretical_loss": 3.5988030703070004,
"tokens_seen": 1159659520
},
{
"epoch": 0.35,
"learning_rate": 0.00032760280842527586,
"loss": 2.6811,
"theoretical_loss": 3.5987842423186662,
"tokens_seen": 1159725056
},
{
"epoch": 0.35,
"learning_rate": 0.000327592778335005,
"loss": 2.8846,
"theoretical_loss": 3.5987654156921627,
"tokens_seen": 1159790592
},
{
"epoch": 0.35,
"learning_rate": 0.0003275827482447342,
"loss": 2.6492,
"theoretical_loss": 3.5987465904273144,
"tokens_seen": 1159856128
},
{
"epoch": 0.35,
"learning_rate": 0.0003275727181544634,
"loss": 2.6449,
"theoretical_loss": 3.598727766523945,
"tokens_seen": 1159921664
},
{
"epoch": 0.35,
"objective/train/docs_used": 660623,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9213621616363525,
"objective/train/theoretical_loss": 3.5987089439818805,
"objective/train/tokens_used": 1180447200,
"theoretical_loss": 3.5987089439818805,
"tokens_seen": 1159987200
},
{
"epoch": 0.35,
"learning_rate": 0.0003275626880641926,
"loss": 2.7226,
"theoretical_loss": 3.5987089439818805,
"tokens_seen": 1159987200
},
{
"epoch": 0.35,
"learning_rate": 0.00032755265797392176,
"loss": 2.8748,
"theoretical_loss": 3.5986901228009454,
"tokens_seen": 1160052736
},
{
"epoch": 0.35,
"learning_rate": 0.00032754262788365094,
"loss": 2.8368,
"theoretical_loss": 3.5986713029809634,
"tokens_seen": 1160118272
},
{
"epoch": 0.35,
"learning_rate": 0.0003275325977933801,
"loss": 3.0101,
"theoretical_loss": 3.59865248452176,
"tokens_seen": 1160183808
},
{
"epoch": 0.35,
"learning_rate": 0.00032752256770310936,
"loss": 2.597,
"theoretical_loss": 3.59863366742316,
"tokens_seen": 1160249344
},
{
"epoch": 0.35,
"learning_rate": 0.0003275125376128385,
"loss": 2.4453,
"theoretical_loss": 3.598614851684987,
"tokens_seen": 1160314880
},
{
"epoch": 0.35,
"learning_rate": 0.0003275025075225677,
"loss": 2.9713,
"theoretical_loss": 3.5985960373070673,
"tokens_seen": 1160380416
},
{
"epoch": 0.35,
"learning_rate": 0.00032749247743229685,
"loss": 2.5849,
"theoretical_loss": 3.598577224289225,
"tokens_seen": 1160445952
},
{
"epoch": 0.35,
"learning_rate": 0.0003274824473420261,
"loss": 2.75,
"theoretical_loss": 3.5985584126312853,
"tokens_seen": 1160511488
},
{
"epoch": 0.35,
"learning_rate": 0.00032747241725175527,
"loss": 2.9672,
"theoretical_loss": 3.598539602333073,
"tokens_seen": 1160577024
},
{
"epoch": 0.35,
"learning_rate": 0.00032746238716148445,
"loss": 3.0223,
"theoretical_loss": 3.598520793394413,
"tokens_seen": 1160642560
},
{
"epoch": 0.35,
"learning_rate": 0.00032745235707121363,
"loss": 2.8772,
"theoretical_loss": 3.59850198581513,
"tokens_seen": 1160708096
},
{
"epoch": 0.35,
"learning_rate": 0.00032744232698094286,
"loss": 2.6363,
"theoretical_loss": 3.59848317959505,
"tokens_seen": 1160773632
},
{
"epoch": 0.35,
"learning_rate": 0.000327432296890672,
"loss": 2.8532,
"theoretical_loss": 3.5984643747339966,
"tokens_seen": 1160839168
},
{
"epoch": 0.35,
"learning_rate": 0.0003274222668004012,
"loss": 2.9029,
"theoretical_loss": 3.5984455712317955,
"tokens_seen": 1160904704
},
{
"epoch": 0.35,
"learning_rate": 0.00032741223671013035,
"loss": 2.9571,
"theoretical_loss": 3.598426769088272,
"tokens_seen": 1160970240
},
{
"epoch": 0.35,
"learning_rate": 0.0003274022066198596,
"loss": 2.7601,
"theoretical_loss": 3.5984079683032517,
"tokens_seen": 1161035776
},
{
"epoch": 0.35,
"learning_rate": 0.00032739217652958877,
"loss": 2.6786,
"theoretical_loss": 3.5983891688765586,
"tokens_seen": 1161101312
},
{
"epoch": 0.35,
"learning_rate": 0.00032738214643931795,
"loss": 2.7311,
"theoretical_loss": 3.598370370808019,
"tokens_seen": 1161166848
},
{
"epoch": 0.35,
"learning_rate": 0.00032737211634904713,
"loss": 2.7082,
"theoretical_loss": 3.598351574097457,
"tokens_seen": 1161232384
},
{
"epoch": 0.35,
"learning_rate": 0.0003273620862587763,
"loss": 2.6171,
"theoretical_loss": 3.598332778744699,
"tokens_seen": 1161297920
},
{
"epoch": 0.35,
"learning_rate": 0.0003273520561685055,
"loss": 2.8257,
"theoretical_loss": 3.59831398474957,
"tokens_seen": 1161363456
},
{
"epoch": 0.35,
"learning_rate": 0.00032734202607823473,
"loss": 2.4946,
"theoretical_loss": 3.5982951921118946,
"tokens_seen": 1161428992
},
{
"epoch": 0.35,
"learning_rate": 0.00032733199598796386,
"loss": 2.6251,
"theoretical_loss": 3.598276400831499,
"tokens_seen": 1161494528
},
{
"epoch": 0.35,
"learning_rate": 0.0003273219658976931,
"loss": 2.5994,
"theoretical_loss": 3.5982576109082083,
"tokens_seen": 1161560064
},
{
"epoch": 0.35,
"objective/train/docs_used": 661405,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.898366928100586,
"objective/train/theoretical_loss": 3.598238822341848,
"objective/train/tokens_used": 1182085600,
"theoretical_loss": 3.598238822341848,
"tokens_seen": 1161625600
},
{
"epoch": 0.35,
"learning_rate": 0.00032731193580742233,
"loss": 2.9395,
"theoretical_loss": 3.598238822341848,
"tokens_seen": 1161625600
},
{
"epoch": 0.35,
"learning_rate": 0.00032730190571715145,
"loss": 2.7984,
"theoretical_loss": 3.5982200351322433,
"tokens_seen": 1161691136
},
{
"epoch": 0.35,
"learning_rate": 0.0003272918756268807,
"loss": 2.7132,
"theoretical_loss": 3.59820124927922,
"tokens_seen": 1161756672
},
{
"epoch": 0.35,
"learning_rate": 0.0003272818455366098,
"loss": 2.9602,
"theoretical_loss": 3.598182464782604,
"tokens_seen": 1161822208
},
{
"epoch": 0.35,
"learning_rate": 0.00032727181544633905,
"loss": 2.6402,
"theoretical_loss": 3.5981636816422196,
"tokens_seen": 1161887744
},
{
"epoch": 0.35,
"learning_rate": 0.00032726178535606823,
"loss": 2.6814,
"theoretical_loss": 3.5981448998578935,
"tokens_seen": 1161953280
},
{
"epoch": 0.35,
"learning_rate": 0.0003272517552657974,
"loss": 2.5315,
"theoretical_loss": 3.598126119429451,
"tokens_seen": 1162018816
},
{
"epoch": 0.35,
"learning_rate": 0.0003272417251755266,
"loss": 2.6051,
"theoretical_loss": 3.5981073403567176,
"tokens_seen": 1162084352
},
{
"epoch": 0.35,
"learning_rate": 0.0003272316950852558,
"loss": 2.7406,
"theoretical_loss": 3.5980885626395196,
"tokens_seen": 1162149888
},
{
"epoch": 0.35,
"learning_rate": 0.00032722166499498496,
"loss": 2.9172,
"theoretical_loss": 3.598069786277682,
"tokens_seen": 1162215424
},
{
"epoch": 0.35,
"learning_rate": 0.0003272116349047142,
"loss": 2.4802,
"theoretical_loss": 3.598051011271031,
"tokens_seen": 1162280960
},
{
"epoch": 0.35,
"learning_rate": 0.0003272016048144433,
"loss": 2.8406,
"theoretical_loss": 3.5980322376193916,
"tokens_seen": 1162346496
},
{
"epoch": 0.35,
"learning_rate": 0.00032719157472417256,
"loss": 2.8724,
"theoretical_loss": 3.598013465322591,
"tokens_seen": 1162412032
},
{
"epoch": 0.35,
"learning_rate": 0.0003271815446339017,
"loss": 2.7169,
"theoretical_loss": 3.5979946943804535,
"tokens_seen": 1162477568
},
{
"epoch": 0.35,
"learning_rate": 0.0003271715145436309,
"loss": 2.7276,
"theoretical_loss": 3.5979759247928067,
"tokens_seen": 1162543104
},
{
"epoch": 0.35,
"learning_rate": 0.0003271614844533601,
"loss": 2.6143,
"theoretical_loss": 3.597957156559475,
"tokens_seen": 1162608640
},
{
"epoch": 0.35,
"learning_rate": 0.0003271514543630893,
"loss": 2.9037,
"theoretical_loss": 3.5979383896802846,
"tokens_seen": 1162674176
},
{
"epoch": 0.35,
"learning_rate": 0.00032714142427281846,
"loss": 2.6418,
"theoretical_loss": 3.597919624155062,
"tokens_seen": 1162739712
},
{
"epoch": 0.35,
"learning_rate": 0.0003271313941825477,
"loss": 2.5921,
"theoretical_loss": 3.597900859983633,
"tokens_seen": 1162805248
},
{
"epoch": 0.35,
"learning_rate": 0.0003271213640922768,
"loss": 2.4427,
"theoretical_loss": 3.5978820971658236,
"tokens_seen": 1162870784
},
{
"epoch": 0.35,
"learning_rate": 0.00032711133400200606,
"loss": 2.8283,
"theoretical_loss": 3.5978633357014598,
"tokens_seen": 1162936320
},
{
"epoch": 0.35,
"learning_rate": 0.0003271013039117352,
"loss": 2.8178,
"theoretical_loss": 3.5978445755903685,
"tokens_seen": 1163001856
},
{
"epoch": 0.35,
"learning_rate": 0.0003270912738214644,
"loss": 2.7422,
"theoretical_loss": 3.5978258168323745,
"tokens_seen": 1163067392
},
{
"epoch": 0.35,
"learning_rate": 0.0003270812437311936,
"loss": 2.7597,
"theoretical_loss": 3.5978070594273044,
"tokens_seen": 1163132928
},
{
"epoch": 0.35,
"learning_rate": 0.0003270712136409228,
"loss": 3.0362,
"theoretical_loss": 3.597788303374985,
"tokens_seen": 1163198464
},
{
"epoch": 0.35,
"objective/train/docs_used": 662532,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9417991638183594,
"objective/train/theoretical_loss": 3.5977695486752426,
"objective/train/tokens_used": 1183724000,
"theoretical_loss": 3.5977695486752426,
"tokens_seen": 1163264000
},
{
"epoch": 0.35,
"learning_rate": 0.00032706118355065196,
"loss": 2.7781,
"theoretical_loss": 3.5977695486752426,
"tokens_seen": 1163264000
},
{
"epoch": 0.35,
"learning_rate": 0.00032705115346038114,
"loss": 2.6273,
"theoretical_loss": 3.597750795327902,
"tokens_seen": 1163329536
},
{
"epoch": 0.35,
"learning_rate": 0.0003270411233701103,
"loss": 2.5328,
"theoretical_loss": 3.5977320433327913,
"tokens_seen": 1163395072
},
{
"epoch": 0.35,
"learning_rate": 0.00032703109327983956,
"loss": 2.6372,
"theoretical_loss": 3.597713292689736,
"tokens_seen": 1163460608
},
{
"epoch": 0.35,
"learning_rate": 0.0003270210631895687,
"loss": 2.8203,
"theoretical_loss": 3.5976945433985623,
"tokens_seen": 1163526144
},
{
"epoch": 0.35,
"learning_rate": 0.0003270110330992979,
"loss": 2.9471,
"theoretical_loss": 3.5976757954590965,
"tokens_seen": 1163591680
},
{
"epoch": 0.35,
"learning_rate": 0.00032700100300902705,
"loss": 2.9006,
"theoretical_loss": 3.597657048871166,
"tokens_seen": 1163657216
},
{
"epoch": 0.35,
"learning_rate": 0.0003269909729187563,
"loss": 2.5695,
"theoretical_loss": 3.5976383036345965,
"tokens_seen": 1163722752
},
{
"epoch": 0.35,
"learning_rate": 0.00032698094282848547,
"loss": 2.4612,
"theoretical_loss": 3.597619559749214,
"tokens_seen": 1163788288
},
{
"epoch": 0.35,
"learning_rate": 0.00032697091273821465,
"loss": 2.5603,
"theoretical_loss": 3.5976008172148464,
"tokens_seen": 1163853824
},
{
"epoch": 0.35,
"learning_rate": 0.00032696088264794383,
"loss": 2.8392,
"theoretical_loss": 3.597582076031319,
"tokens_seen": 1163919360
},
{
"epoch": 0.35,
"learning_rate": 0.00032695085255767306,
"loss": 2.6305,
"theoretical_loss": 3.5975633361984594,
"tokens_seen": 1163984896
},
{
"epoch": 0.35,
"learning_rate": 0.0003269408224674022,
"loss": 2.8959,
"theoretical_loss": 3.5975445977160927,
"tokens_seen": 1164050432
},
{
"epoch": 0.35,
"learning_rate": 0.0003269307923771314,
"loss": 2.7856,
"theoretical_loss": 3.597525860584047,
"tokens_seen": 1164115968
},
{
"epoch": 0.35,
"learning_rate": 0.00032692076228686055,
"loss": 2.8837,
"theoretical_loss": 3.597507124802149,
"tokens_seen": 1164181504
},
{
"epoch": 0.35,
"learning_rate": 0.0003269107321965898,
"loss": 2.6621,
"theoretical_loss": 3.5974883903702244,
"tokens_seen": 1164247040
},
{
"epoch": 0.35,
"learning_rate": 0.00032690070210631897,
"loss": 2.5934,
"theoretical_loss": 3.5974696572881006,
"tokens_seen": 1164312576
},
{
"epoch": 0.35,
"learning_rate": 0.00032689067201604815,
"loss": 2.7571,
"theoretical_loss": 3.597450925555604,
"tokens_seen": 1164378112
},
{
"epoch": 0.35,
"learning_rate": 0.00032688064192577733,
"loss": 2.7714,
"theoretical_loss": 3.597432195172562,
"tokens_seen": 1164443648
},
{
"epoch": 0.35,
"learning_rate": 0.0003268706118355065,
"loss": 2.7628,
"theoretical_loss": 3.597413466138801,
"tokens_seen": 1164509184
},
{
"epoch": 0.35,
"learning_rate": 0.0003268605817452357,
"loss": 2.6072,
"theoretical_loss": 3.597394738454148,
"tokens_seen": 1164574720
},
{
"epoch": 0.35,
"learning_rate": 0.00032685055165496493,
"loss": 2.5946,
"theoretical_loss": 3.59737601211843,
"tokens_seen": 1164640256
},
{
"epoch": 0.35,
"learning_rate": 0.00032684052156469406,
"loss": 2.5734,
"theoretical_loss": 3.597357287131474,
"tokens_seen": 1164705792
},
{
"epoch": 0.35,
"learning_rate": 0.0003268304914744233,
"loss": 2.8016,
"theoretical_loss": 3.597338563493106,
"tokens_seen": 1164771328
},
{
"epoch": 0.35,
"learning_rate": 0.0003268204613841524,
"loss": 2.6846,
"theoretical_loss": 3.5973198412031544,
"tokens_seen": 1164836864
},
{
"epoch": 0.35,
"objective/train/docs_used": 663185,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.48357892036438,
"objective/train/theoretical_loss": 3.5973011202614456,
"objective/train/tokens_used": 1185362400,
"theoretical_loss": 3.5973011202614456,
"tokens_seen": 1164902400
},
{
"epoch": 0.35,
"learning_rate": 0.00032681043129388165,
"loss": 2.8072,
"theoretical_loss": 3.5973011202614456,
"tokens_seen": 1164902400
},
{
"epoch": 0.35,
"learning_rate": 0.00032680040120361084,
"loss": 2.9716,
"theoretical_loss": 3.5972824006678064,
"tokens_seen": 1164967936
},
{
"epoch": 0.35,
"learning_rate": 0.00032679037111334,
"loss": 2.6925,
"theoretical_loss": 3.5972636824220645,
"tokens_seen": 1165033472
},
{
"epoch": 0.35,
"learning_rate": 0.0003267803410230692,
"loss": 2.8444,
"theoretical_loss": 3.5972449655240473,
"tokens_seen": 1165099008
},
{
"epoch": 0.35,
"learning_rate": 0.00032677031093279843,
"loss": 2.6155,
"theoretical_loss": 3.5972262499735805,
"tokens_seen": 1165164544
},
{
"epoch": 0.35,
"learning_rate": 0.00032676028084252756,
"loss": 2.7625,
"theoretical_loss": 3.597207535770493,
"tokens_seen": 1165230080
},
{
"epoch": 0.35,
"learning_rate": 0.0003267502507522568,
"loss": 2.6869,
"theoretical_loss": 3.5971888229146103,
"tokens_seen": 1165295616
},
{
"epoch": 0.35,
"learning_rate": 0.0003267402206619859,
"loss": 2.7357,
"theoretical_loss": 3.5971701114057613,
"tokens_seen": 1165361152
},
{
"epoch": 0.35,
"learning_rate": 0.00032673019057171516,
"loss": 2.9526,
"theoretical_loss": 3.5971514012437718,
"tokens_seen": 1165426688
},
{
"epoch": 0.35,
"learning_rate": 0.00032672016048144434,
"loss": 2.6874,
"theoretical_loss": 3.5971326924284703,
"tokens_seen": 1165492224
},
{
"epoch": 0.35,
"learning_rate": 0.0003267101303911735,
"loss": 2.8473,
"theoretical_loss": 3.5971139849596843,
"tokens_seen": 1165557760
},
{
"epoch": 0.35,
"learning_rate": 0.0003267001003009027,
"loss": 2.8707,
"theoretical_loss": 3.59709527883724,
"tokens_seen": 1165623296
},
{
"epoch": 0.35,
"learning_rate": 0.0003266900702106319,
"loss": 2.7017,
"theoretical_loss": 3.5970765740609654,
"tokens_seen": 1165688832
},
{
"epoch": 0.35,
"learning_rate": 0.00032668004012036106,
"loss": 2.6754,
"theoretical_loss": 3.5970578706306884,
"tokens_seen": 1165754368
},
{
"epoch": 0.35,
"learning_rate": 0.0003266700100300903,
"loss": 2.6808,
"theoretical_loss": 3.597039168546236,
"tokens_seen": 1165819904
},
{
"epoch": 0.35,
"learning_rate": 0.0003266599799398194,
"loss": 2.6475,
"theoretical_loss": 3.597020467807436,
"tokens_seen": 1165885440
},
{
"epoch": 0.35,
"learning_rate": 0.00032664994984954866,
"loss": 2.7745,
"theoretical_loss": 3.597001768414115,
"tokens_seen": 1165950976
},
{
"epoch": 0.35,
"learning_rate": 0.00032663991975927784,
"loss": 2.6524,
"theoretical_loss": 3.596983070366102,
"tokens_seen": 1166016512
},
{
"epoch": 0.35,
"learning_rate": 0.000326629889669007,
"loss": 2.7328,
"theoretical_loss": 3.5969643736632237,
"tokens_seen": 1166082048
},
{
"epoch": 0.35,
"learning_rate": 0.0003266198595787362,
"loss": 2.7223,
"theoretical_loss": 3.5969456783053078,
"tokens_seen": 1166147584
},
{
"epoch": 0.35,
"learning_rate": 0.0003266098294884654,
"loss": 2.752,
"theoretical_loss": 3.5969269842921823,
"tokens_seen": 1166213120
},
{
"epoch": 0.35,
"learning_rate": 0.00032659979939819457,
"loss": 2.8258,
"theoretical_loss": 3.5969082916236745,
"tokens_seen": 1166278656
},
{
"epoch": 0.35,
"learning_rate": 0.0003265897693079238,
"loss": 2.9704,
"theoretical_loss": 3.5968896002996127,
"tokens_seen": 1166344192
},
{
"epoch": 0.35,
"learning_rate": 0.00032657973921765293,
"loss": 2.5373,
"theoretical_loss": 3.596870910319824,
"tokens_seen": 1166409728
},
{
"epoch": 0.35,
"learning_rate": 0.00032656970912738216,
"loss": 2.516,
"theoretical_loss": 3.596852221684137,
"tokens_seen": 1166475264
},
{
"epoch": 0.35,
"objective/train/docs_used": 664585,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.543100118637085,
"objective/train/theoretical_loss": 3.596833534392379,
"objective/train/tokens_used": 1187000800,
"theoretical_loss": 3.596833534392379,
"tokens_seen": 1166540800
},
{
"epoch": 0.35,
"learning_rate": 0.00032655967903711135,
"loss": 2.7175,
"theoretical_loss": 3.596833534392379,
"tokens_seen": 1166540800
},
{
"epoch": 0.35,
"learning_rate": 0.0003265496489468405,
"loss": 2.9905,
"theoretical_loss": 3.596814848444378,
"tokens_seen": 1166606336
},
{
"epoch": 0.35,
"learning_rate": 0.00032653961885656976,
"loss": 2.3393,
"theoretical_loss": 3.5967961638399615,
"tokens_seen": 1166671872
},
{
"epoch": 0.35,
"learning_rate": 0.0003265295887662989,
"loss": 2.7836,
"theoretical_loss": 3.596777480578958,
"tokens_seen": 1166737408
},
{
"epoch": 0.35,
"learning_rate": 0.0003265195586760281,
"loss": 2.8507,
"theoretical_loss": 3.596758798661195,
"tokens_seen": 1166802944
},
{
"epoch": 0.35,
"learning_rate": 0.00032650952858575725,
"loss": 2.7257,
"theoretical_loss": 3.5967401180865006,
"tokens_seen": 1166868480
},
{
"epoch": 0.35,
"learning_rate": 0.0003264994984954865,
"loss": 2.8779,
"theoretical_loss": 3.5967214388547033,
"tokens_seen": 1166934016
},
{
"epoch": 0.35,
"learning_rate": 0.00032648946840521567,
"loss": 2.8145,
"theoretical_loss": 3.5967027609656306,
"tokens_seen": 1166999552
},
{
"epoch": 0.35,
"learning_rate": 0.00032647943831494485,
"loss": 2.5983,
"theoretical_loss": 3.5966840844191106,
"tokens_seen": 1167065088
},
{
"epoch": 0.35,
"learning_rate": 0.00032646940822467403,
"loss": 2.7798,
"theoretical_loss": 3.596665409214972,
"tokens_seen": 1167130624
},
{
"epoch": 0.35,
"learning_rate": 0.00032645937813440326,
"loss": 2.8694,
"theoretical_loss": 3.5966467353530422,
"tokens_seen": 1167196160
},
{
"epoch": 0.35,
"learning_rate": 0.0003264493480441324,
"loss": 2.5823,
"theoretical_loss": 3.5966280628331493,
"tokens_seen": 1167261696
},
{
"epoch": 0.35,
"learning_rate": 0.0003264393179538616,
"loss": 2.8105,
"theoretical_loss": 3.5966093916551225,
"tokens_seen": 1167327232
},
{
"epoch": 0.35,
"learning_rate": 0.00032642928786359075,
"loss": 2.6563,
"theoretical_loss": 3.5965907218187896,
"tokens_seen": 1167392768
},
{
"epoch": 0.35,
"learning_rate": 0.00032641925777332,
"loss": 2.9677,
"theoretical_loss": 3.5965720533239782,
"tokens_seen": 1167458304
},
{
"epoch": 0.35,
"learning_rate": 0.00032640922768304917,
"loss": 2.7158,
"theoretical_loss": 3.596553386170518,
"tokens_seen": 1167523840
},
{
"epoch": 0.35,
"learning_rate": 0.00032639919759277835,
"loss": 2.6701,
"theoretical_loss": 3.5965347203582354,
"tokens_seen": 1167589376
},
{
"epoch": 0.35,
"learning_rate": 0.00032638916750250753,
"loss": 2.9209,
"theoretical_loss": 3.59651605588696,
"tokens_seen": 1167654912
},
{
"epoch": 0.35,
"learning_rate": 0.0003263791374122367,
"loss": 2.8191,
"theoretical_loss": 3.596497392756521,
"tokens_seen": 1167720448
},
{
"epoch": 0.35,
"learning_rate": 0.0003263691073219659,
"loss": 2.5613,
"theoretical_loss": 3.596478730966745,
"tokens_seen": 1167785984
},
{
"epoch": 0.35,
"learning_rate": 0.00032635907723169513,
"loss": 2.6782,
"theoretical_loss": 3.5964600705174616,
"tokens_seen": 1167851520
},
{
"epoch": 0.35,
"learning_rate": 0.00032634904714142426,
"loss": 2.7618,
"theoretical_loss": 3.596441411408499,
"tokens_seen": 1167917056
},
{
"epoch": 0.35,
"learning_rate": 0.0003263390170511535,
"loss": 2.8667,
"theoretical_loss": 3.5964227536396858,
"tokens_seen": 1167982592
},
{
"epoch": 0.35,
"learning_rate": 0.0003263289869608826,
"loss": 2.7474,
"theoretical_loss": 3.5964040972108506,
"tokens_seen": 1168048128
},
{
"epoch": 0.35,
"learning_rate": 0.00032631895687061185,
"loss": 2.5867,
"theoretical_loss": 3.5963854421218215,
"tokens_seen": 1168113664
},
{
"epoch": 0.35,
"objective/train/docs_used": 665171,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.364827871322632,
"objective/train/theoretical_loss": 3.596366788372428,
"objective/train/tokens_used": 1188639200,
"theoretical_loss": 3.596366788372428,
"tokens_seen": 1168179200
},
{
"epoch": 0.35,
"learning_rate": 0.00032630892678034104,
"loss": 2.5715,
"theoretical_loss": 3.596366788372428,
"tokens_seen": 1168179200
},
{
"epoch": 0.35,
"learning_rate": 0.0003262988966900702,
"loss": 2.6271,
"theoretical_loss": 3.596348135962498,
"tokens_seen": 1168244736
},
{
"epoch": 0.35,
"learning_rate": 0.0003262888665997994,
"loss": 2.7179,
"theoretical_loss": 3.5963294848918608,
"tokens_seen": 1168310272
},
{
"epoch": 0.35,
"learning_rate": 0.00032627883650952863,
"loss": 2.6429,
"theoretical_loss": 3.5963108351603443,
"tokens_seen": 1168375808
},
{
"epoch": 0.35,
"learning_rate": 0.00032626880641925776,
"loss": 2.7291,
"theoretical_loss": 3.596292186767778,
"tokens_seen": 1168441344
},
{
"epoch": 0.35,
"learning_rate": 0.000326258776328987,
"loss": 2.4907,
"theoretical_loss": 3.596273539713991,
"tokens_seen": 1168506880
},
{
"epoch": 0.35,
"learning_rate": 0.0003262487462387161,
"loss": 2.6258,
"theoretical_loss": 3.5962548939988106,
"tokens_seen": 1168572416
},
{
"epoch": 0.35,
"learning_rate": 0.00032623871614844536,
"loss": 2.7656,
"theoretical_loss": 3.5962362496220672,
"tokens_seen": 1168637952
},
{
"epoch": 0.35,
"learning_rate": 0.00032622868605817454,
"loss": 2.5593,
"theoretical_loss": 3.5962176065835885,
"tokens_seen": 1168703488
},
{
"epoch": 0.35,
"learning_rate": 0.0003262186559679037,
"loss": 2.7993,
"theoretical_loss": 3.5961989648832042,
"tokens_seen": 1168769024
},
{
"epoch": 0.35,
"learning_rate": 0.0003262086258776329,
"loss": 2.7814,
"theoretical_loss": 3.596180324520743,
"tokens_seen": 1168834560
},
{
"epoch": 0.35,
"learning_rate": 0.0003261985957873621,
"loss": 2.6933,
"theoretical_loss": 3.5961616854960337,
"tokens_seen": 1168900096
},
{
"epoch": 0.35,
"learning_rate": 0.00032618856569709126,
"loss": 2.9056,
"theoretical_loss": 3.596143047808906,
"tokens_seen": 1168965632
},
{
"epoch": 0.35,
"learning_rate": 0.0003261785356068205,
"loss": 2.7665,
"theoretical_loss": 3.5961244114591877,
"tokens_seen": 1169031168
},
{
"epoch": 0.35,
"learning_rate": 0.0003261685055165496,
"loss": 2.8794,
"theoretical_loss": 3.596105776446709,
"tokens_seen": 1169096704
},
{
"epoch": 0.35,
"learning_rate": 0.00032615847542627886,
"loss": 3.016,
"theoretical_loss": 3.596087142771298,
"tokens_seen": 1169162240
},
{
"epoch": 0.35,
"learning_rate": 0.00032614844533600804,
"loss": 2.7644,
"theoretical_loss": 3.5960685104327847,
"tokens_seen": 1169227776
},
{
"epoch": 0.35,
"learning_rate": 0.0003261384152457372,
"loss": 2.8329,
"theoretical_loss": 3.596049879430998,
"tokens_seen": 1169293312
},
{
"epoch": 0.35,
"learning_rate": 0.0003261283851554664,
"loss": 2.664,
"theoretical_loss": 3.596031249765767,
"tokens_seen": 1169358848
},
{
"epoch": 0.35,
"learning_rate": 0.0003261183550651956,
"loss": 2.5904,
"theoretical_loss": 3.5960126214369206,
"tokens_seen": 1169424384
},
{
"epoch": 0.35,
"learning_rate": 0.00032610832497492477,
"loss": 2.3949,
"theoretical_loss": 3.5959939944442887,
"tokens_seen": 1169489920
},
{
"epoch": 0.35,
"learning_rate": 0.000326098294884654,
"loss": 2.6785,
"theoretical_loss": 3.5959753687877005,
"tokens_seen": 1169555456
},
{
"epoch": 0.35,
"learning_rate": 0.00032608826479438313,
"loss": 2.6023,
"theoretical_loss": 3.5959567444669847,
"tokens_seen": 1169620992
},
{
"epoch": 0.35,
"learning_rate": 0.00032607823470411236,
"loss": 2.7648,
"theoretical_loss": 3.595938121481971,
"tokens_seen": 1169686528
},
{
"epoch": 0.35,
"learning_rate": 0.0003260682046138415,
"loss": 2.9621,
"theoretical_loss": 3.5959194998324886,
"tokens_seen": 1169752064
},
{
"epoch": 0.35,
"objective/train/docs_used": 666471,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.916159152984619,
"objective/train/theoretical_loss": 3.595900879518368,
"objective/train/tokens_used": 1190277600,
"theoretical_loss": 3.595900879518368,
"tokens_seen": 1169817600
},
{
"epoch": 0.35,
"learning_rate": 0.0003260581745235707,
"loss": 2.8866,
"theoretical_loss": 3.595900879518368,
"tokens_seen": 1169817600
},
{
"epoch": 0.35,
"learning_rate": 0.0003260481444332999,
"loss": 2.7242,
"theoretical_loss": 3.5958822605394367,
"tokens_seen": 1169883136
},
{
"epoch": 0.35,
"learning_rate": 0.0003260381143430291,
"loss": 2.8964,
"theoretical_loss": 3.5958636428955257,
"tokens_seen": 1169948672
},
{
"epoch": 0.35,
"learning_rate": 0.00032602808425275827,
"loss": 3.0386,
"theoretical_loss": 3.595845026586464,
"tokens_seen": 1170014208
},
{
"epoch": 0.35,
"learning_rate": 0.00032601805416248745,
"loss": 2.8981,
"theoretical_loss": 3.595826411612081,
"tokens_seen": 1170079744
},
{
"epoch": 0.35,
"learning_rate": 0.00032600802407221663,
"loss": 2.7267,
"theoretical_loss": 3.5958077979722063,
"tokens_seen": 1170145280
},
{
"epoch": 0.35,
"learning_rate": 0.00032599799398194587,
"loss": 2.6425,
"theoretical_loss": 3.59578918566667,
"tokens_seen": 1170210816
},
{
"epoch": 0.35,
"learning_rate": 0.000325987963891675,
"loss": 2.8463,
"theoretical_loss": 3.595770574695301,
"tokens_seen": 1170276352
},
{
"epoch": 0.35,
"learning_rate": 0.00032597793380140423,
"loss": 2.7973,
"theoretical_loss": 3.5957519650579295,
"tokens_seen": 1170341888
},
{
"epoch": 0.35,
"learning_rate": 0.0003259679037111334,
"loss": 2.8039,
"theoretical_loss": 3.5957333567543848,
"tokens_seen": 1170407424
},
{
"epoch": 0.35,
"learning_rate": 0.0003259578736208626,
"loss": 2.7256,
"theoretical_loss": 3.595714749784497,
"tokens_seen": 1170472960
},
{
"epoch": 0.35,
"learning_rate": 0.0003259478435305918,
"loss": 2.7232,
"theoretical_loss": 3.5956961441480955,
"tokens_seen": 1170538496
},
{
"epoch": 0.35,
"learning_rate": 0.00032593781344032095,
"loss": 2.7851,
"theoretical_loss": 3.5956775398450103,
"tokens_seen": 1170604032
},
{
"epoch": 0.35,
"learning_rate": 0.00032592778335005014,
"loss": 2.6992,
"theoretical_loss": 3.595658936875071,
"tokens_seen": 1170669568
},
{
"epoch": 0.35,
"learning_rate": 0.00032591775325977937,
"loss": 2.5875,
"theoretical_loss": 3.595640335238108,
"tokens_seen": 1170735104
},
{
"epoch": 0.35,
"learning_rate": 0.0003259077231695085,
"loss": 2.4026,
"theoretical_loss": 3.5956217349339505,
"tokens_seen": 1170800640
},
{
"epoch": 0.35,
"learning_rate": 0.00032589769307923773,
"loss": 2.9415,
"theoretical_loss": 3.5956031359624285,
"tokens_seen": 1170866176
},
{
"epoch": 0.35,
"learning_rate": 0.00032588766298896686,
"loss": 2.6081,
"theoretical_loss": 3.595584538323372,
"tokens_seen": 1170931712
},
{
"epoch": 0.35,
"learning_rate": 0.0003258776328986961,
"loss": 2.6203,
"theoretical_loss": 3.595565942016612,
"tokens_seen": 1170997248
},
{
"epoch": 0.35,
"learning_rate": 0.0003258676028084253,
"loss": 2.8058,
"theoretical_loss": 3.5955473470419763,
"tokens_seen": 1171062784
},
{
"epoch": 0.35,
"learning_rate": 0.00032585757271815446,
"loss": 2.6251,
"theoretical_loss": 3.5955287533992975,
"tokens_seen": 1171128320
},
{
"epoch": 0.35,
"learning_rate": 0.00032584754262788364,
"loss": 2.936,
"theoretical_loss": 3.5955101610884035,
"tokens_seen": 1171193856
},
{
"epoch": 0.35,
"learning_rate": 0.0003258375125376128,
"loss": 2.8934,
"theoretical_loss": 3.5954915701091257,
"tokens_seen": 1171259392
},
{
"epoch": 0.35,
"learning_rate": 0.000325827482447342,
"loss": 2.4635,
"theoretical_loss": 3.5954729804612935,
"tokens_seen": 1171324928
},
{
"epoch": 0.35,
"learning_rate": 0.00032581745235707124,
"loss": 2.8988,
"theoretical_loss": 3.5954543921447377,
"tokens_seen": 1171390464
},
{
"epoch": 0.35,
"objective/train/docs_used": 666886,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.071739912033081,
"objective/train/theoretical_loss": 3.595435805159288,
"objective/train/tokens_used": 1191916000,
"theoretical_loss": 3.595435805159288,
"tokens_seen": 1171456000
},
{
"epoch": 0.35,
"learning_rate": 0.0003258074222668004,
"loss": 2.7488,
"theoretical_loss": 3.595435805159288,
"tokens_seen": 1171456000
},
{
"epoch": 0.36,
"learning_rate": 0.0003257973921765296,
"loss": 2.569,
"theoretical_loss": 3.5954172195047747,
"tokens_seen": 1171521536
},
{
"epoch": 0.36,
"learning_rate": 0.00032578736208625883,
"loss": 2.894,
"theoretical_loss": 3.5953986351810285,
"tokens_seen": 1171587072
},
{
"epoch": 0.36,
"learning_rate": 0.00032577733199598796,
"loss": 2.6486,
"theoretical_loss": 3.5953800521878794,
"tokens_seen": 1171652608
},
{
"epoch": 0.36,
"learning_rate": 0.0003257673019057172,
"loss": 2.7822,
"theoretical_loss": 3.595361470525157,
"tokens_seen": 1171718144
},
{
"epoch": 0.36,
"learning_rate": 0.0003257572718154463,
"loss": 2.8212,
"theoretical_loss": 3.5953428901926925,
"tokens_seen": 1171783680
},
{
"epoch": 0.36,
"learning_rate": 0.00032574724172517556,
"loss": 2.6602,
"theoretical_loss": 3.595324311190316,
"tokens_seen": 1171849216
},
{
"epoch": 0.36,
"learning_rate": 0.00032573721163490474,
"loss": 2.903,
"theoretical_loss": 3.595305733517858,
"tokens_seen": 1171914752
},
{
"epoch": 0.36,
"learning_rate": 0.0003257271815446339,
"loss": 2.6185,
"theoretical_loss": 3.5952871571751492,
"tokens_seen": 1171980288
},
{
"epoch": 0.36,
"learning_rate": 0.0003257171514543631,
"loss": 2.7524,
"theoretical_loss": 3.5952685821620194,
"tokens_seen": 1172045824
},
{
"epoch": 0.36,
"learning_rate": 0.0003257071213640923,
"loss": 2.7078,
"theoretical_loss": 3.5952500084782995,
"tokens_seen": 1172111360
},
{
"epoch": 0.36,
"learning_rate": 0.00032569709127382146,
"loss": 2.6379,
"theoretical_loss": 3.59523143612382,
"tokens_seen": 1172176896
},
{
"epoch": 0.36,
"learning_rate": 0.0003256870611835507,
"loss": 2.814,
"theoretical_loss": 3.5952128650984116,
"tokens_seen": 1172242432
},
{
"epoch": 0.36,
"learning_rate": 0.0003256770310932798,
"loss": 2.5858,
"theoretical_loss": 3.595194295401905,
"tokens_seen": 1172307968
},
{
"epoch": 0.36,
"learning_rate": 0.00032566700100300906,
"loss": 2.7107,
"theoretical_loss": 3.59517572703413,
"tokens_seen": 1172373504
},
{
"epoch": 0.36,
"learning_rate": 0.00032565697091273824,
"loss": 2.5871,
"theoretical_loss": 3.595157159994918,
"tokens_seen": 1172439040
},
{
"epoch": 0.36,
"learning_rate": 0.0003256469408224674,
"loss": 2.7446,
"theoretical_loss": 3.5951385942840997,
"tokens_seen": 1172504576
},
{
"epoch": 0.36,
"learning_rate": 0.0003256369107321966,
"loss": 2.9401,
"theoretical_loss": 3.595120029901506,
"tokens_seen": 1172570112
},
{
"epoch": 0.36,
"learning_rate": 0.0003256268806419258,
"loss": 2.9301,
"theoretical_loss": 3.595101466846966,
"tokens_seen": 1172635648
},
{
"epoch": 0.36,
"learning_rate": 0.00032561685055165497,
"loss": 2.7135,
"theoretical_loss": 3.595082905120313,
"tokens_seen": 1172701184
},
{
"epoch": 0.36,
"learning_rate": 0.0003256068204613842,
"loss": 2.8924,
"theoretical_loss": 3.595064344721376,
"tokens_seen": 1172766720
},
{
"epoch": 0.36,
"learning_rate": 0.00032559679037111333,
"loss": 2.6339,
"theoretical_loss": 3.5950457856499867,
"tokens_seen": 1172832256
},
{
"epoch": 0.36,
"learning_rate": 0.00032558676028084256,
"loss": 2.6502,
"theoretical_loss": 3.595027227905976,
"tokens_seen": 1172897792
},
{
"epoch": 0.36,
"learning_rate": 0.0003255767301905717,
"loss": 2.7711,
"theoretical_loss": 3.5950086714891736,
"tokens_seen": 1172963328
},
{
"epoch": 0.36,
"learning_rate": 0.0003255667001003009,
"loss": 3.0462,
"theoretical_loss": 3.594990116399412,
"tokens_seen": 1173028864
},
{
"epoch": 0.36,
"objective/train/docs_used": 667905,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8657987117767334,
"objective/train/theoretical_loss": 3.594971562636521,
"objective/train/tokens_used": 1193554400,
"theoretical_loss": 3.594971562636521,
"tokens_seen": 1173094400
},
{
"epoch": 0.36,
"learning_rate": 0.0003255566700100301,
"loss": 2.8427,
"theoretical_loss": 3.594971562636521,
"tokens_seen": 1173094400
},
{
"epoch": 0.36,
"learning_rate": 0.0003255466399197593,
"loss": 2.6936,
"theoretical_loss": 3.5949530102003324,
"tokens_seen": 1173159936
},
{
"epoch": 0.36,
"learning_rate": 0.00032553660982948847,
"loss": 2.5937,
"theoretical_loss": 3.594934459090677,
"tokens_seen": 1173225472
},
{
"epoch": 0.36,
"learning_rate": 0.00032552657973921765,
"loss": 2.6374,
"theoretical_loss": 3.5949159093073857,
"tokens_seen": 1173291008
},
{
"epoch": 0.36,
"learning_rate": 0.00032551654964894683,
"loss": 2.821,
"theoretical_loss": 3.5948973608502897,
"tokens_seen": 1173356544
},
{
"epoch": 0.36,
"learning_rate": 0.00032550651955867607,
"loss": 2.7654,
"theoretical_loss": 3.5948788137192196,
"tokens_seen": 1173422080
},
{
"epoch": 0.36,
"learning_rate": 0.0003254964894684052,
"loss": 2.9424,
"theoretical_loss": 3.594860267914008,
"tokens_seen": 1173487616
},
{
"epoch": 0.36,
"learning_rate": 0.00032548645937813443,
"loss": 2.7051,
"theoretical_loss": 3.5948417234344845,
"tokens_seen": 1173553152
},
{
"epoch": 0.36,
"learning_rate": 0.0003254764292878636,
"loss": 2.7042,
"theoretical_loss": 3.594823180280481,
"tokens_seen": 1173618688
},
{
"epoch": 0.36,
"learning_rate": 0.0003254663991975928,
"loss": 2.6254,
"theoretical_loss": 3.5948046384518286,
"tokens_seen": 1173684224
},
{
"epoch": 0.36,
"learning_rate": 0.000325456369107322,
"loss": 2.7023,
"theoretical_loss": 3.594786097948359,
"tokens_seen": 1173749760
},
{
"epoch": 0.36,
"learning_rate": 0.00032544633901705115,
"loss": 2.569,
"theoretical_loss": 3.5947675587699024,
"tokens_seen": 1173815296
},
{
"epoch": 0.36,
"learning_rate": 0.00032543630892678034,
"loss": 2.6629,
"theoretical_loss": 3.594749020916292,
"tokens_seen": 1173880832
},
{
"epoch": 0.36,
"learning_rate": 0.00032542627883650957,
"loss": 3.0275,
"theoretical_loss": 3.594730484387357,
"tokens_seen": 1173946368
},
{
"epoch": 0.36,
"learning_rate": 0.0003254162487462387,
"loss": 3.0199,
"theoretical_loss": 3.5947119491829307,
"tokens_seen": 1174011904
},
{
"epoch": 0.36,
"learning_rate": 0.00032540621865596793,
"loss": 2.7053,
"theoretical_loss": 3.594693415302843,
"tokens_seen": 1174077440
},
{
"epoch": 0.36,
"learning_rate": 0.00032539618856569706,
"loss": 2.6854,
"theoretical_loss": 3.5946748827469266,
"tokens_seen": 1174142976
},
{
"epoch": 0.36,
"learning_rate": 0.0003253861584754263,
"loss": 2.6369,
"theoretical_loss": 3.5946563515150123,
"tokens_seen": 1174208512
},
{
"epoch": 0.36,
"learning_rate": 0.0003253761283851555,
"loss": 3.0562,
"theoretical_loss": 3.5946378216069315,
"tokens_seen": 1174274048
},
{
"epoch": 0.36,
"learning_rate": 0.00032536609829488466,
"loss": 2.7145,
"theoretical_loss": 3.5946192930225163,
"tokens_seen": 1174339584
},
{
"epoch": 0.36,
"learning_rate": 0.00032535606820461384,
"loss": 2.9609,
"theoretical_loss": 3.594600765761598,
"tokens_seen": 1174405120
},
{
"epoch": 0.36,
"learning_rate": 0.000325346038114343,
"loss": 2.6532,
"theoretical_loss": 3.5945822398240077,
"tokens_seen": 1174470656
},
{
"epoch": 0.36,
"learning_rate": 0.0003253360080240722,
"loss": 2.8418,
"theoretical_loss": 3.5945637152095786,
"tokens_seen": 1174536192
},
{
"epoch": 0.36,
"learning_rate": 0.00032532597793380144,
"loss": 2.7498,
"theoretical_loss": 3.59454519191814,
"tokens_seen": 1174601728
},
{
"epoch": 0.36,
"learning_rate": 0.00032531594784353056,
"loss": 2.6862,
"theoretical_loss": 3.594526669949526,
"tokens_seen": 1174667264
},
{
"epoch": 0.36,
"objective/train/docs_used": 668590,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.402711868286133,
"objective/train/theoretical_loss": 3.594508149303567,
"objective/train/tokens_used": 1195192800,
"theoretical_loss": 3.594508149303567,
"tokens_seen": 1174732800
},
{
"epoch": 0.36,
"learning_rate": 0.0003253059177532598,
"loss": 2.5384,
"theoretical_loss": 3.594508149303567,
"tokens_seen": 1174732800
},
{
"epoch": 0.36,
"learning_rate": 0.000325295887662989,
"loss": 2.7567,
"theoretical_loss": 3.594489629980095,
"tokens_seen": 1174798336
},
{
"epoch": 0.36,
"learning_rate": 0.00032528585757271816,
"loss": 2.7332,
"theoretical_loss": 3.594471111978942,
"tokens_seen": 1174863872
},
{
"epoch": 0.36,
"learning_rate": 0.00032527582748244734,
"loss": 2.7846,
"theoretical_loss": 3.5944525952999395,
"tokens_seen": 1174929408
},
{
"epoch": 0.36,
"learning_rate": 0.0003252657973921765,
"loss": 2.8344,
"theoretical_loss": 3.5944340799429195,
"tokens_seen": 1174994944
},
{
"epoch": 0.36,
"learning_rate": 0.0003252557673019057,
"loss": 2.7638,
"theoretical_loss": 3.594415565907714,
"tokens_seen": 1175060480
},
{
"epoch": 0.36,
"learning_rate": 0.00032524573721163494,
"loss": 2.7933,
"theoretical_loss": 3.5943970531941556,
"tokens_seen": 1175126016
},
{
"epoch": 0.36,
"learning_rate": 0.00032523570712136407,
"loss": 2.7216,
"theoretical_loss": 3.5943785418020746,
"tokens_seen": 1175191552
},
{
"epoch": 0.36,
"learning_rate": 0.0003252256770310933,
"loss": 2.7536,
"theoretical_loss": 3.594360031731304,
"tokens_seen": 1175257088
},
{
"epoch": 0.36,
"learning_rate": 0.00032521564694082243,
"loss": 2.5867,
"theoretical_loss": 3.5943415229816758,
"tokens_seen": 1175322624
},
{
"epoch": 0.36,
"learning_rate": 0.00032520561685055166,
"loss": 2.7909,
"theoretical_loss": 3.594323015553022,
"tokens_seen": 1175388160
},
{
"epoch": 0.36,
"learning_rate": 0.00032519558676028084,
"loss": 2.777,
"theoretical_loss": 3.5943045094451747,
"tokens_seen": 1175453696
},
{
"epoch": 0.36,
"learning_rate": 0.00032518555667001,
"loss": 2.7844,
"theoretical_loss": 3.594286004657966,
"tokens_seen": 1175519232
},
{
"epoch": 0.36,
"learning_rate": 0.0003251755265797392,
"loss": 2.9169,
"theoretical_loss": 3.5942675011912275,
"tokens_seen": 1175584768
},
{
"epoch": 0.36,
"learning_rate": 0.00032516549648946844,
"loss": 2.6493,
"theoretical_loss": 3.594248999044792,
"tokens_seen": 1175650304
},
{
"epoch": 0.36,
"learning_rate": 0.00032515546639919757,
"loss": 2.7996,
"theoretical_loss": 3.594230498218492,
"tokens_seen": 1175715840
},
{
"epoch": 0.36,
"learning_rate": 0.0003251454363089268,
"loss": 2.6557,
"theoretical_loss": 3.594211998712159,
"tokens_seen": 1175781376
},
{
"epoch": 0.36,
"learning_rate": 0.00032513540621865593,
"loss": 2.7386,
"theoretical_loss": 3.5941935005256256,
"tokens_seen": 1175846912
},
{
"epoch": 0.36,
"learning_rate": 0.00032512537612838517,
"loss": 2.52,
"theoretical_loss": 3.594175003658724,
"tokens_seen": 1175912448
},
{
"epoch": 0.36,
"learning_rate": 0.00032511534603811435,
"loss": 2.6669,
"theoretical_loss": 3.5941565081112863,
"tokens_seen": 1175977984
},
{
"epoch": 0.36,
"learning_rate": 0.00032510531594784353,
"loss": 2.6655,
"theoretical_loss": 3.5941380138831454,
"tokens_seen": 1176043520
},
{
"epoch": 0.36,
"learning_rate": 0.0003250952858575727,
"loss": 2.7348,
"theoretical_loss": 3.5941195209741332,
"tokens_seen": 1176109056
},
{
"epoch": 0.36,
"learning_rate": 0.0003250852557673019,
"loss": 2.6709,
"theoretical_loss": 3.5941010293840825,
"tokens_seen": 1176174592
},
{
"epoch": 0.36,
"learning_rate": 0.00032507522567703107,
"loss": 2.9324,
"theoretical_loss": 3.5940825391128257,
"tokens_seen": 1176240128
},
{
"epoch": 0.36,
"learning_rate": 0.0003250651955867603,
"loss": 2.7949,
"theoretical_loss": 3.5940640501601946,
"tokens_seen": 1176305664
},
{
"epoch": 0.36,
"objective/train/docs_used": 669752,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6799681186676025,
"objective/train/theoretical_loss": 3.5940455625260226,
"objective/train/tokens_used": 1196831200,
"theoretical_loss": 3.5940455625260226,
"tokens_seen": 1176371200
},
{
"epoch": 0.36,
"learning_rate": 0.0003250551654964895,
"loss": 2.719,
"theoretical_loss": 3.5940455625260226,
"tokens_seen": 1176371200
},
{
"epoch": 0.36,
"learning_rate": 0.00032504513540621867,
"loss": 2.8102,
"theoretical_loss": 3.594027076210142,
"tokens_seen": 1176436736
},
{
"epoch": 0.36,
"learning_rate": 0.00032503510531594785,
"loss": 2.8918,
"theoretical_loss": 3.5940085912123845,
"tokens_seen": 1176502272
},
{
"epoch": 0.36,
"learning_rate": 0.00032502507522567703,
"loss": 2.6584,
"theoretical_loss": 3.593990107532584,
"tokens_seen": 1176567808
},
{
"epoch": 0.36,
"learning_rate": 0.00032501504513540627,
"loss": 2.8405,
"theoretical_loss": 3.593971625170572,
"tokens_seen": 1176633344
},
{
"epoch": 0.36,
"learning_rate": 0.0003250050150451354,
"loss": 2.8718,
"theoretical_loss": 3.593953144126182,
"tokens_seen": 1176698880
},
{
"epoch": 0.36,
"learning_rate": 0.00032499498495486463,
"loss": 2.7597,
"theoretical_loss": 3.5939346643992467,
"tokens_seen": 1176764416
},
{
"epoch": 0.36,
"learning_rate": 0.0003249849548645938,
"loss": 2.8567,
"theoretical_loss": 3.5939161859895985,
"tokens_seen": 1176829952
},
{
"epoch": 0.36,
"learning_rate": 0.000324974924774323,
"loss": 2.8985,
"theoretical_loss": 3.59389770889707,
"tokens_seen": 1176895488
},
{
"epoch": 0.36,
"learning_rate": 0.0003249648946840522,
"loss": 2.8225,
"theoretical_loss": 3.593879233121494,
"tokens_seen": 1176961024
},
{
"epoch": 0.36,
"learning_rate": 0.00032495486459378135,
"loss": 2.6239,
"theoretical_loss": 3.5938607586627036,
"tokens_seen": 1177026560
},
{
"epoch": 0.36,
"learning_rate": 0.00032494483450351054,
"loss": 2.5093,
"theoretical_loss": 3.5938422855205316,
"tokens_seen": 1177092096
},
{
"epoch": 0.36,
"learning_rate": 0.00032493480441323977,
"loss": 2.5778,
"theoretical_loss": 3.5938238136948106,
"tokens_seen": 1177157632
},
{
"epoch": 0.36,
"learning_rate": 0.0003249247743229689,
"loss": 2.7336,
"theoretical_loss": 3.5938053431853736,
"tokens_seen": 1177223168
},
{
"epoch": 0.36,
"learning_rate": 0.00032491474423269813,
"loss": 2.6829,
"theoretical_loss": 3.5937868739920535,
"tokens_seen": 1177288704
},
{
"epoch": 0.36,
"learning_rate": 0.00032490471414242726,
"loss": 2.6898,
"theoretical_loss": 3.593768406114684,
"tokens_seen": 1177354240
},
{
"epoch": 0.36,
"learning_rate": 0.0003248946840521565,
"loss": 2.8576,
"theoretical_loss": 3.593749939553097,
"tokens_seen": 1177419776
},
{
"epoch": 0.36,
"learning_rate": 0.0003248846539618857,
"loss": 2.9735,
"theoretical_loss": 3.593731474307126,
"tokens_seen": 1177485312
},
{
"epoch": 0.36,
"learning_rate": 0.00032487462387161486,
"loss": 2.6283,
"theoretical_loss": 3.5937130103766046,
"tokens_seen": 1177550848
},
{
"epoch": 0.36,
"learning_rate": 0.00032486459378134404,
"loss": 2.8618,
"theoretical_loss": 3.5936945477613644,
"tokens_seen": 1177616384
},
{
"epoch": 0.36,
"learning_rate": 0.0003248545636910732,
"loss": 2.8002,
"theoretical_loss": 3.59367608646124,
"tokens_seen": 1177681920
},
{
"epoch": 0.36,
"learning_rate": 0.0003248445336008024,
"loss": 2.7042,
"theoretical_loss": 3.593657626476064,
"tokens_seen": 1177747456
},
{
"epoch": 0.36,
"learning_rate": 0.00032483450351053164,
"loss": 2.795,
"theoretical_loss": 3.5936391678056694,
"tokens_seen": 1177812992
},
{
"epoch": 0.36,
"learning_rate": 0.00032482447342026076,
"loss": 2.7721,
"theoretical_loss": 3.5936207104498896,
"tokens_seen": 1177878528
},
{
"epoch": 0.36,
"learning_rate": 0.00032481444332999,
"loss": 2.6581,
"theoretical_loss": 3.5936022544085575,
"tokens_seen": 1177944064
},
{
"epoch": 0.36,
"objective/train/docs_used": 670392,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.102618455886841,
"objective/train/theoretical_loss": 3.593583799681507,
"objective/train/tokens_used": 1198469600,
"theoretical_loss": 3.593583799681507,
"tokens_seen": 1178009600
},
{
"epoch": 0.36,
"learning_rate": 0.0003248044132397192,
"loss": 2.7469,
"theoretical_loss": 3.593583799681507,
"tokens_seen": 1178009600
},
{
"epoch": 0.36,
"learning_rate": 0.00032479438314944836,
"loss": 2.7262,
"theoretical_loss": 3.593565346268571,
"tokens_seen": 1178075136
},
{
"epoch": 0.36,
"learning_rate": 0.00032478435305917754,
"loss": 2.7464,
"theoretical_loss": 3.593546894169583,
"tokens_seen": 1178140672
},
{
"epoch": 0.36,
"learning_rate": 0.0003247743229689067,
"loss": 2.7882,
"theoretical_loss": 3.5935284433843764,
"tokens_seen": 1178206208
},
{
"epoch": 0.36,
"learning_rate": 0.0003247642928786359,
"loss": 2.5463,
"theoretical_loss": 3.593509993912784,
"tokens_seen": 1178271744
},
{
"epoch": 0.36,
"learning_rate": 0.00032475426278836514,
"loss": 2.5567,
"theoretical_loss": 3.593491545754641,
"tokens_seen": 1178337280
},
{
"epoch": 0.36,
"learning_rate": 0.00032474423269809427,
"loss": 2.778,
"theoretical_loss": 3.5934730989097776,
"tokens_seen": 1178402816
},
{
"epoch": 0.36,
"learning_rate": 0.0003247342026078235,
"loss": 2.8068,
"theoretical_loss": 3.5934546533780303,
"tokens_seen": 1178468352
},
{
"epoch": 0.36,
"learning_rate": 0.00032472417251755263,
"loss": 2.7575,
"theoretical_loss": 3.593436209159232,
"tokens_seen": 1178533888
},
{
"epoch": 0.36,
"learning_rate": 0.00032471414242728186,
"loss": 2.7711,
"theoretical_loss": 3.5934177662532147,
"tokens_seen": 1178599424
},
{
"epoch": 0.36,
"learning_rate": 0.00032470411233701104,
"loss": 2.602,
"theoretical_loss": 3.5933993246598135,
"tokens_seen": 1178664960
},
{
"epoch": 0.36,
"learning_rate": 0.0003246940822467402,
"loss": 2.5463,
"theoretical_loss": 3.5933808843788615,
"tokens_seen": 1178730496
},
{
"epoch": 0.36,
"learning_rate": 0.0003246840521564694,
"loss": 2.6522,
"theoretical_loss": 3.5933624454101922,
"tokens_seen": 1178796032
},
{
"epoch": 0.36,
"learning_rate": 0.00032467402206619864,
"loss": 2.4387,
"theoretical_loss": 3.5933440077536396,
"tokens_seen": 1178861568
},
{
"epoch": 0.36,
"learning_rate": 0.00032466399197592777,
"loss": 2.7109,
"theoretical_loss": 3.5933255714090375,
"tokens_seen": 1178927104
},
{
"epoch": 0.36,
"learning_rate": 0.000324653961885657,
"loss": 2.7608,
"theoretical_loss": 3.593307136376219,
"tokens_seen": 1178992640
},
{
"epoch": 0.36,
"learning_rate": 0.00032464393179538613,
"loss": 2.7884,
"theoretical_loss": 3.5932887026550184,
"tokens_seen": 1179058176
},
{
"epoch": 0.36,
"learning_rate": 0.00032463390170511537,
"loss": 3.0072,
"theoretical_loss": 3.5932702702452692,
"tokens_seen": 1179123712
},
{
"epoch": 0.36,
"learning_rate": 0.00032462387161484455,
"loss": 2.6411,
"theoretical_loss": 3.5932518391468053,
"tokens_seen": 1179189248
},
{
"epoch": 0.36,
"learning_rate": 0.00032461384152457373,
"loss": 2.8625,
"theoretical_loss": 3.5932334093594607,
"tokens_seen": 1179254784
},
{
"epoch": 0.36,
"learning_rate": 0.0003246038114343029,
"loss": 2.7317,
"theoretical_loss": 3.5932149808830687,
"tokens_seen": 1179320320
},
{
"epoch": 0.36,
"learning_rate": 0.0003245937813440321,
"loss": 2.6374,
"theoretical_loss": 3.593196553717464,
"tokens_seen": 1179385856
},
{
"epoch": 0.36,
"learning_rate": 0.00032458375125376127,
"loss": 2.8173,
"theoretical_loss": 3.5931781278624806,
"tokens_seen": 1179451392
},
{
"epoch": 0.36,
"learning_rate": 0.0003245737211634905,
"loss": 2.7988,
"theoretical_loss": 3.5931597033179514,
"tokens_seen": 1179516928
},
{
"epoch": 0.36,
"learning_rate": 0.00032456369107321963,
"loss": 3.0459,
"theoretical_loss": 3.593141280083711,
"tokens_seen": 1179582464
},
{
"epoch": 0.36,
"objective/train/docs_used": 671618,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9856789112091064,
"objective/train/theoretical_loss": 3.5931228581595938,
"objective/train/tokens_used": 1200108000,
"theoretical_loss": 3.5931228581595938,
"tokens_seen": 1179648000
},
{
"epoch": 0.36,
"learning_rate": 0.00032455366098294887,
"loss": 2.6388,
"theoretical_loss": 3.5931228581595938,
"tokens_seen": 1179648000
},
{
"epoch": 0.36,
"learning_rate": 0.000324543630892678,
"loss": 2.7625,
"theoretical_loss": 3.593104437545433,
"tokens_seen": 1179713536
},
{
"epoch": 0.36,
"learning_rate": 0.00032453360080240723,
"loss": 2.8977,
"theoretical_loss": 3.5930860182410638,
"tokens_seen": 1179779072
},
{
"epoch": 0.36,
"learning_rate": 0.0003245235707121364,
"loss": 2.6929,
"theoretical_loss": 3.5930676002463198,
"tokens_seen": 1179844608
},
{
"epoch": 0.36,
"learning_rate": 0.0003245135406218656,
"loss": 2.4501,
"theoretical_loss": 3.5930491835610345,
"tokens_seen": 1179910144
},
{
"epoch": 0.36,
"learning_rate": 0.0003245035105315948,
"loss": 2.5436,
"theoretical_loss": 3.593030768185043,
"tokens_seen": 1179975680
},
{
"epoch": 0.36,
"learning_rate": 0.000324493480441324,
"loss": 2.8713,
"theoretical_loss": 3.593012354118179,
"tokens_seen": 1180041216
},
{
"epoch": 0.36,
"learning_rate": 0.00032448345035105314,
"loss": 3.1031,
"theoretical_loss": 3.5929939413602767,
"tokens_seen": 1180106752
},
{
"epoch": 0.36,
"learning_rate": 0.0003244734202607824,
"loss": 2.9392,
"theoretical_loss": 3.592975529911171,
"tokens_seen": 1180172288
},
{
"epoch": 0.36,
"learning_rate": 0.0003244633901705115,
"loss": 2.8918,
"theoretical_loss": 3.5929571197706958,
"tokens_seen": 1180237824
},
{
"epoch": 0.36,
"learning_rate": 0.00032445336008024074,
"loss": 2.9253,
"theoretical_loss": 3.592938710938685,
"tokens_seen": 1180303360
},
{
"epoch": 0.36,
"learning_rate": 0.0003244433299899699,
"loss": 2.6136,
"theoretical_loss": 3.5929203034149735,
"tokens_seen": 1180368896
},
{
"epoch": 0.36,
"learning_rate": 0.0003244332998996991,
"loss": 2.8332,
"theoretical_loss": 3.5929018971993956,
"tokens_seen": 1180434432
},
{
"epoch": 0.36,
"learning_rate": 0.0003244232698094283,
"loss": 2.6955,
"theoretical_loss": 3.5928834922917856,
"tokens_seen": 1180499968
},
{
"epoch": 0.36,
"learning_rate": 0.00032441323971915746,
"loss": 2.8181,
"theoretical_loss": 3.592865088691978,
"tokens_seen": 1180565504
},
{
"epoch": 0.36,
"learning_rate": 0.00032440320962888664,
"loss": 2.9094,
"theoretical_loss": 3.592846686399807,
"tokens_seen": 1180631040
},
{
"epoch": 0.36,
"learning_rate": 0.0003243931795386159,
"loss": 2.7064,
"theoretical_loss": 3.592828285415108,
"tokens_seen": 1180696576
},
{
"epoch": 0.36,
"learning_rate": 0.000324383149448345,
"loss": 2.6319,
"theoretical_loss": 3.5928098857377146,
"tokens_seen": 1180762112
},
{
"epoch": 0.36,
"learning_rate": 0.00032437311935807424,
"loss": 2.6637,
"theoretical_loss": 3.592791487367462,
"tokens_seen": 1180827648
},
{
"epoch": 0.36,
"learning_rate": 0.00032436308926780337,
"loss": 2.6428,
"theoretical_loss": 3.592773090304184,
"tokens_seen": 1180893184
},
{
"epoch": 0.36,
"learning_rate": 0.0003243530591775326,
"loss": 2.9756,
"theoretical_loss": 3.592754694547716,
"tokens_seen": 1180958720
},
{
"epoch": 0.36,
"learning_rate": 0.0003243430290872618,
"loss": 2.8454,
"theoretical_loss": 3.5927363000978927,
"tokens_seen": 1181024256
},
{
"epoch": 0.36,
"learning_rate": 0.00032433299899699096,
"loss": 2.687,
"theoretical_loss": 3.592717906954548,
"tokens_seen": 1181089792
},
{
"epoch": 0.36,
"learning_rate": 0.0003243229689067202,
"loss": 2.7435,
"theoretical_loss": 3.592699515117517,
"tokens_seen": 1181155328
},
{
"epoch": 0.36,
"learning_rate": 0.0003243129388164494,
"loss": 2.5767,
"theoretical_loss": 3.592681124586635,
"tokens_seen": 1181220864
},
{
"epoch": 0.36,
"objective/train/docs_used": 672229,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7187225818634033,
"objective/train/theoretical_loss": 3.592662735361736,
"objective/train/tokens_used": 1201746400,
"theoretical_loss": 3.592662735361736,
"tokens_seen": 1181286400
},
{
"epoch": 0.36,
"learning_rate": 0.00032430290872617856,
"loss": 2.6995,
"theoretical_loss": 3.592662735361736,
"tokens_seen": 1181286400
},
{
"epoch": 0.36,
"learning_rate": 0.00032429287863590774,
"loss": 2.8963,
"theoretical_loss": 3.5926443474426555,
"tokens_seen": 1181351936
},
{
"epoch": 0.36,
"learning_rate": 0.0003242828485456369,
"loss": 2.7124,
"theoretical_loss": 3.5926259608292277,
"tokens_seen": 1181417472
},
{
"epoch": 0.36,
"learning_rate": 0.0003242728184553661,
"loss": 2.8587,
"theoretical_loss": 3.592607575521288,
"tokens_seen": 1181483008
},
{
"epoch": 0.36,
"learning_rate": 0.00032426278836509534,
"loss": 2.7514,
"theoretical_loss": 3.592589191518671,
"tokens_seen": 1181548544
},
{
"epoch": 0.36,
"learning_rate": 0.00032425275827482447,
"loss": 2.5642,
"theoretical_loss": 3.5925708088212116,
"tokens_seen": 1181614080
},
{
"epoch": 0.36,
"learning_rate": 0.0003242427281845537,
"loss": 2.5439,
"theoretical_loss": 3.592552427428745,
"tokens_seen": 1181679616
},
{
"epoch": 0.36,
"learning_rate": 0.00032423269809428283,
"loss": 2.6284,
"theoretical_loss": 3.592534047341106,
"tokens_seen": 1181745152
},
{
"epoch": 0.36,
"learning_rate": 0.00032422266800401206,
"loss": 2.6546,
"theoretical_loss": 3.5925156685581294,
"tokens_seen": 1181810688
},
{
"epoch": 0.36,
"learning_rate": 0.00032421263791374125,
"loss": 2.6512,
"theoretical_loss": 3.592497291079651,
"tokens_seen": 1181876224
},
{
"epoch": 0.36,
"learning_rate": 0.0003242026078234704,
"loss": 2.7279,
"theoretical_loss": 3.592478914905505,
"tokens_seen": 1181941760
},
{
"epoch": 0.36,
"learning_rate": 0.0003241925777331996,
"loss": 2.655,
"theoretical_loss": 3.592460540035527,
"tokens_seen": 1182007296
},
{
"epoch": 0.36,
"learning_rate": 0.00032418254764292884,
"loss": 2.7598,
"theoretical_loss": 3.592442166469552,
"tokens_seen": 1182072832
},
{
"epoch": 0.36,
"learning_rate": 0.00032417251755265797,
"loss": 2.95,
"theoretical_loss": 3.5924237942074155,
"tokens_seen": 1182138368
},
{
"epoch": 0.36,
"learning_rate": 0.0003241624874623872,
"loss": 2.7133,
"theoretical_loss": 3.5924054232489517,
"tokens_seen": 1182203904
},
{
"epoch": 0.36,
"learning_rate": 0.00032415245737211633,
"loss": 2.7422,
"theoretical_loss": 3.5923870535939972,
"tokens_seen": 1182269440
},
{
"epoch": 0.36,
"learning_rate": 0.00032414242728184557,
"loss": 2.6309,
"theoretical_loss": 3.592368685242387,
"tokens_seen": 1182334976
},
{
"epoch": 0.36,
"learning_rate": 0.00032413239719157475,
"loss": 2.594,
"theoretical_loss": 3.5923503181939553,
"tokens_seen": 1182400512
},
{
"epoch": 0.36,
"learning_rate": 0.00032412236710130393,
"loss": 2.8306,
"theoretical_loss": 3.5923319524485384,
"tokens_seen": 1182466048
},
{
"epoch": 0.36,
"learning_rate": 0.0003241123370110331,
"loss": 2.8296,
"theoretical_loss": 3.592313588005971,
"tokens_seen": 1182531584
},
{
"epoch": 0.36,
"learning_rate": 0.0003241023069207623,
"loss": 2.7662,
"theoretical_loss": 3.5922952248660893,
"tokens_seen": 1182597120
},
{
"epoch": 0.36,
"learning_rate": 0.00032409227683049147,
"loss": 2.8765,
"theoretical_loss": 3.592276863028728,
"tokens_seen": 1182662656
},
{
"epoch": 0.36,
"learning_rate": 0.0003240822467402207,
"loss": 2.8068,
"theoretical_loss": 3.5922585024937232,
"tokens_seen": 1182728192
},
{
"epoch": 0.36,
"learning_rate": 0.00032407221664994984,
"loss": 2.7935,
"theoretical_loss": 3.5922401432609092,
"tokens_seen": 1182793728
},
{
"epoch": 0.36,
"learning_rate": 0.00032406218655967907,
"loss": 2.6531,
"theoretical_loss": 3.592221785330123,
"tokens_seen": 1182859264
},
{
"epoch": 0.36,
"objective/train/docs_used": 673021,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.705904245376587,
"objective/train/theoretical_loss": 3.5922034287011995,
"objective/train/tokens_used": 1203384800,
"theoretical_loss": 3.5922034287011995,
"tokens_seen": 1182924800
},
{
"epoch": 0.36,
"learning_rate": 0.0003240521564694082,
"loss": 2.8498,
"theoretical_loss": 3.5922034287011995,
"tokens_seen": 1182924800
},
{
"epoch": 0.36,
"learning_rate": 0.00032404212637913743,
"loss": 2.72,
"theoretical_loss": 3.5921850733739733,
"tokens_seen": 1182990336
},
{
"epoch": 0.36,
"learning_rate": 0.0003240320962888666,
"loss": 2.7635,
"theoretical_loss": 3.592166719348282,
"tokens_seen": 1183055872
},
{
"epoch": 0.36,
"learning_rate": 0.0003240220661985958,
"loss": 2.6158,
"theoretical_loss": 3.5921483666239595,
"tokens_seen": 1183121408
},
{
"epoch": 0.36,
"learning_rate": 0.000324012036108325,
"loss": 2.7275,
"theoretical_loss": 3.592130015200842,
"tokens_seen": 1183186944
},
{
"epoch": 0.36,
"learning_rate": 0.0003240020060180542,
"loss": 2.6102,
"theoretical_loss": 3.592111665078765,
"tokens_seen": 1183252480
},
{
"epoch": 0.36,
"learning_rate": 0.00032399197592778334,
"loss": 2.662,
"theoretical_loss": 3.592093316257565,
"tokens_seen": 1183318016
},
{
"epoch": 0.36,
"learning_rate": 0.0003239819458375126,
"loss": 2.724,
"theoretical_loss": 3.5920749687370765,
"tokens_seen": 1183383552
},
{
"epoch": 0.36,
"learning_rate": 0.0003239719157472417,
"loss": 2.7788,
"theoretical_loss": 3.5920566225171364,
"tokens_seen": 1183449088
},
{
"epoch": 0.36,
"learning_rate": 0.00032396188565697094,
"loss": 2.6336,
"theoretical_loss": 3.5920382775975797,
"tokens_seen": 1183514624
},
{
"epoch": 0.36,
"learning_rate": 0.0003239518555667001,
"loss": 2.9366,
"theoretical_loss": 3.592019933978243,
"tokens_seen": 1183580160
},
{
"epoch": 0.36,
"learning_rate": 0.0003239418254764293,
"loss": 2.7448,
"theoretical_loss": 3.592001591658961,
"tokens_seen": 1183645696
},
{
"epoch": 0.36,
"learning_rate": 0.0003239317953861585,
"loss": 2.7088,
"theoretical_loss": 3.5919832506395712,
"tokens_seen": 1183711232
},
{
"epoch": 0.36,
"learning_rate": 0.00032392176529588766,
"loss": 2.3999,
"theoretical_loss": 3.591964910919908,
"tokens_seen": 1183776768
},
{
"epoch": 0.36,
"learning_rate": 0.00032391173520561684,
"loss": 2.7176,
"theoretical_loss": 3.5919465724998076,
"tokens_seen": 1183842304
},
{
"epoch": 0.36,
"learning_rate": 0.0003239017051153461,
"loss": 2.7519,
"theoretical_loss": 3.591928235379107,
"tokens_seen": 1183907840
},
{
"epoch": 0.36,
"learning_rate": 0.0003238916750250752,
"loss": 2.8226,
"theoretical_loss": 3.5919098995576415,
"tokens_seen": 1183973376
},
{
"epoch": 0.36,
"learning_rate": 0.00032388164493480444,
"loss": 2.7163,
"theoretical_loss": 3.5918915650352474,
"tokens_seen": 1184038912
},
{
"epoch": 0.36,
"learning_rate": 0.00032387161484453357,
"loss": 2.921,
"theoretical_loss": 3.59187323181176,
"tokens_seen": 1184104448
},
{
"epoch": 0.36,
"learning_rate": 0.0003238615847542628,
"loss": 2.5301,
"theoretical_loss": 3.591854899887016,
"tokens_seen": 1184169984
},
{
"epoch": 0.36,
"learning_rate": 0.000323851554663992,
"loss": 2.7581,
"theoretical_loss": 3.5918365692608516,
"tokens_seen": 1184235520
},
{
"epoch": 0.36,
"learning_rate": 0.00032384152457372116,
"loss": 2.5786,
"theoretical_loss": 3.5918182399331036,
"tokens_seen": 1184301056
},
{
"epoch": 0.36,
"learning_rate": 0.00032383149448345034,
"loss": 2.5258,
"theoretical_loss": 3.5917999119036064,
"tokens_seen": 1184366592
},
{
"epoch": 0.36,
"learning_rate": 0.0003238214643931796,
"loss": 2.5422,
"theoretical_loss": 3.5917815851721975,
"tokens_seen": 1184432128
},
{
"epoch": 0.36,
"learning_rate": 0.0003238114343029087,
"loss": 2.9155,
"theoretical_loss": 3.5917632597387126,
"tokens_seen": 1184497664
},
{
"epoch": 0.36,
"objective/train/docs_used": 674157,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0029098987579346,
"objective/train/theoretical_loss": 3.591744935602989,
"objective/train/tokens_used": 1205023200,
"theoretical_loss": 3.591744935602989,
"tokens_seen": 1184563200
},
{
"epoch": 0.36,
"learning_rate": 0.00032380140421263794,
"loss": 2.7951,
"theoretical_loss": 3.591744935602989,
"tokens_seen": 1184563200
},
{
"epoch": 0.36,
"learning_rate": 0.00032379137412236707,
"loss": 2.5513,
"theoretical_loss": 3.5917266127648615,
"tokens_seen": 1184628736
},
{
"epoch": 0.36,
"learning_rate": 0.0003237813440320963,
"loss": 2.7087,
"theoretical_loss": 3.591708291224168,
"tokens_seen": 1184694272
},
{
"epoch": 0.36,
"learning_rate": 0.0003237713139418255,
"loss": 2.8639,
"theoretical_loss": 3.5916899709807435,
"tokens_seen": 1184759808
},
{
"epoch": 0.36,
"learning_rate": 0.00032376128385155467,
"loss": 2.9445,
"theoretical_loss": 3.5916716520344254,
"tokens_seen": 1184825344
},
{
"epoch": 0.36,
"learning_rate": 0.00032375125376128385,
"loss": 2.9902,
"theoretical_loss": 3.591653334385049,
"tokens_seen": 1184890880
},
{
"epoch": 0.36,
"learning_rate": 0.00032374122367101303,
"loss": 2.8478,
"theoretical_loss": 3.591635018032452,
"tokens_seen": 1184956416
},
{
"epoch": 0.36,
"learning_rate": 0.0003237311935807422,
"loss": 2.6566,
"theoretical_loss": 3.59161670297647,
"tokens_seen": 1185021952
},
{
"epoch": 0.36,
"learning_rate": 0.00032372116349047145,
"loss": 2.8089,
"theoretical_loss": 3.59159838921694,
"tokens_seen": 1185087488
},
{
"epoch": 0.36,
"learning_rate": 0.00032371113340020057,
"loss": 2.7585,
"theoretical_loss": 3.591580076753698,
"tokens_seen": 1185153024
},
{
"epoch": 0.36,
"learning_rate": 0.0003237011033099298,
"loss": 2.7187,
"theoretical_loss": 3.5915617655865812,
"tokens_seen": 1185218560
},
{
"epoch": 0.36,
"learning_rate": 0.00032369107321965893,
"loss": 2.4654,
"theoretical_loss": 3.591543455715426,
"tokens_seen": 1185284096
},
{
"epoch": 0.36,
"learning_rate": 0.00032368104312938817,
"loss": 2.8456,
"theoretical_loss": 3.591525147140069,
"tokens_seen": 1185349632
},
{
"epoch": 0.36,
"learning_rate": 0.00032367101303911735,
"loss": 2.7388,
"theoretical_loss": 3.591506839860347,
"tokens_seen": 1185415168
},
{
"epoch": 0.36,
"learning_rate": 0.00032366098294884653,
"loss": 2.8047,
"theoretical_loss": 3.5914885338760962,
"tokens_seen": 1185480704
},
{
"epoch": 0.36,
"learning_rate": 0.0003236509528585757,
"loss": 2.6592,
"theoretical_loss": 3.591470229187154,
"tokens_seen": 1185546240
},
{
"epoch": 0.36,
"learning_rate": 0.00032364092276830495,
"loss": 2.8388,
"theoretical_loss": 3.5914519257933564,
"tokens_seen": 1185611776
},
{
"epoch": 0.36,
"learning_rate": 0.0003236308926780341,
"loss": 2.7451,
"theoretical_loss": 3.5914336236945408,
"tokens_seen": 1185677312
},
{
"epoch": 0.36,
"learning_rate": 0.0003236208625877633,
"loss": 2.6856,
"theoretical_loss": 3.5914153228905437,
"tokens_seen": 1185742848
},
{
"epoch": 0.36,
"learning_rate": 0.00032361083249749244,
"loss": 2.7718,
"theoretical_loss": 3.5913970233812016,
"tokens_seen": 1185808384
},
{
"epoch": 0.36,
"learning_rate": 0.0003236008024072217,
"loss": 2.6986,
"theoretical_loss": 3.5913787251663525,
"tokens_seen": 1185873920
},
{
"epoch": 0.36,
"learning_rate": 0.00032359077231695085,
"loss": 2.9689,
"theoretical_loss": 3.5913604282458325,
"tokens_seen": 1185939456
},
{
"epoch": 0.36,
"learning_rate": 0.00032358074222668004,
"loss": 2.8069,
"theoretical_loss": 3.591342132619478,
"tokens_seen": 1186004992
},
{
"epoch": 0.36,
"learning_rate": 0.00032357071213640927,
"loss": 2.5974,
"theoretical_loss": 3.5913238382871273,
"tokens_seen": 1186070528
},
{
"epoch": 0.36,
"learning_rate": 0.0003235606820461384,
"loss": 2.7063,
"theoretical_loss": 3.5913055452486162,
"tokens_seen": 1186136064
},
{
"epoch": 0.36,
"objective/train/docs_used": 674881,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.36728572845459,
"objective/train/theoretical_loss": 3.5912872535037828,
"objective/train/tokens_used": 1206661600,
"theoretical_loss": 3.5912872535037828,
"tokens_seen": 1186201600
},
{
"epoch": 0.36,
"learning_rate": 0.00032355065195586763,
"loss": 2.789,
"theoretical_loss": 3.5912872535037828,
"tokens_seen": 1186201600
},
{
"epoch": 0.36,
"learning_rate": 0.0003235406218655968,
"loss": 2.4429,
"theoretical_loss": 3.591268963052463,
"tokens_seen": 1186267136
},
{
"epoch": 0.36,
"learning_rate": 0.000323530591775326,
"loss": 2.7282,
"theoretical_loss": 3.591250673894494,
"tokens_seen": 1186332672
},
{
"epoch": 0.36,
"learning_rate": 0.0003235205616850552,
"loss": 2.718,
"theoretical_loss": 3.5912323860297137,
"tokens_seen": 1186398208
},
{
"epoch": 0.36,
"learning_rate": 0.0003235105315947844,
"loss": 2.6947,
"theoretical_loss": 3.591214099457959,
"tokens_seen": 1186463744
},
{
"epoch": 0.36,
"learning_rate": 0.00032350050150451354,
"loss": 2.7273,
"theoretical_loss": 3.5911958141790667,
"tokens_seen": 1186529280
},
{
"epoch": 0.36,
"learning_rate": 0.0003234904714142428,
"loss": 2.7379,
"theoretical_loss": 3.5911775301928746,
"tokens_seen": 1186594816
},
{
"epoch": 0.36,
"learning_rate": 0.0003234804413239719,
"loss": 2.5378,
"theoretical_loss": 3.5911592474992187,
"tokens_seen": 1186660352
},
{
"epoch": 0.36,
"learning_rate": 0.00032347041123370114,
"loss": 2.7344,
"theoretical_loss": 3.591140966097938,
"tokens_seen": 1186725888
},
{
"epoch": 0.36,
"learning_rate": 0.0003234603811434303,
"loss": 2.8929,
"theoretical_loss": 3.591122685988868,
"tokens_seen": 1186791424
},
{
"epoch": 0.36,
"learning_rate": 0.0003234503510531595,
"loss": 2.7838,
"theoretical_loss": 3.5911044071718474,
"tokens_seen": 1186856960
},
{
"epoch": 0.36,
"learning_rate": 0.0003234403209628887,
"loss": 2.832,
"theoretical_loss": 3.5910861296467127,
"tokens_seen": 1186922496
},
{
"epoch": 0.36,
"learning_rate": 0.00032343029087261786,
"loss": 2.6891,
"theoretical_loss": 3.5910678534133016,
"tokens_seen": 1186988032
},
{
"epoch": 0.36,
"learning_rate": 0.00032342026078234704,
"loss": 2.7365,
"theoretical_loss": 3.591049578471452,
"tokens_seen": 1187053568
},
{
"epoch": 0.36,
"learning_rate": 0.0003234102306920763,
"loss": 2.7601,
"theoretical_loss": 3.591031304821,
"tokens_seen": 1187119104
},
{
"epoch": 0.36,
"learning_rate": 0.0003234002006018054,
"loss": 2.6613,
"theoretical_loss": 3.5910130324617846,
"tokens_seen": 1187184640
},
{
"epoch": 0.36,
"learning_rate": 0.00032339017051153464,
"loss": 2.7578,
"theoretical_loss": 3.5909947613936417,
"tokens_seen": 1187250176
},
{
"epoch": 0.36,
"learning_rate": 0.00032338014042126377,
"loss": 2.4274,
"theoretical_loss": 3.59097649161641,
"tokens_seen": 1187315712
},
{
"epoch": 0.36,
"learning_rate": 0.000323370110330993,
"loss": 2.6348,
"theoretical_loss": 3.5909582231299266,
"tokens_seen": 1187381248
},
{
"epoch": 0.36,
"learning_rate": 0.0003233600802407222,
"loss": 2.6339,
"theoretical_loss": 3.5909399559340294,
"tokens_seen": 1187446784
},
{
"epoch": 0.36,
"learning_rate": 0.00032335005015045136,
"loss": 2.5628,
"theoretical_loss": 3.590921690028556,
"tokens_seen": 1187512320
},
{
"epoch": 0.36,
"learning_rate": 0.00032334002006018054,
"loss": 2.7724,
"theoretical_loss": 3.5909034254133436,
"tokens_seen": 1187577856
},
{
"epoch": 0.36,
"learning_rate": 0.0003233299899699098,
"loss": 2.8222,
"theoretical_loss": 3.59088516208823,
"tokens_seen": 1187643392
},
{
"epoch": 0.36,
"learning_rate": 0.0003233199598796389,
"loss": 2.6923,
"theoretical_loss": 3.590866900053053,
"tokens_seen": 1187708928
},
{
"epoch": 0.36,
"learning_rate": 0.00032330992978936814,
"loss": 2.7877,
"theoretical_loss": 3.5908486393076506,
"tokens_seen": 1187774464
},
{
"epoch": 0.36,
"objective/train/docs_used": 676005,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7910947799682617,
"objective/train/theoretical_loss": 3.5908303798518597,
"objective/train/tokens_used": 1208300000,
"theoretical_loss": 3.5908303798518597,
"tokens_seen": 1187840000
},
{
"epoch": 0.36,
"learning_rate": 0.00032329989969909727,
"loss": 2.7337,
"theoretical_loss": 3.5908303798518597,
"tokens_seen": 1187840000
},
{
"epoch": 0.36,
"learning_rate": 0.0003232898696088265,
"loss": 2.8501,
"theoretical_loss": 3.590812121685519,
"tokens_seen": 1187905536
},
{
"epoch": 0.36,
"learning_rate": 0.0003232798395185557,
"loss": 2.581,
"theoretical_loss": 3.590793864808466,
"tokens_seen": 1187971072
},
{
"epoch": 0.36,
"learning_rate": 0.00032326980942828487,
"loss": 2.8114,
"theoretical_loss": 3.590775609220538,
"tokens_seen": 1188036608
},
{
"epoch": 0.36,
"learning_rate": 0.00032325977933801405,
"loss": 2.7663,
"theoretical_loss": 3.590757354921574,
"tokens_seen": 1188102144
},
{
"epoch": 0.36,
"learning_rate": 0.00032324974924774323,
"loss": 2.6171,
"theoretical_loss": 3.590739101911411,
"tokens_seen": 1188167680
},
{
"epoch": 0.36,
"learning_rate": 0.0003232397191574724,
"loss": 2.6536,
"theoretical_loss": 3.5907208501898866,
"tokens_seen": 1188233216
},
{
"epoch": 0.36,
"learning_rate": 0.00032322968906720165,
"loss": 2.6616,
"theoretical_loss": 3.59070259975684,
"tokens_seen": 1188298752
},
{
"epoch": 0.36,
"learning_rate": 0.00032321965897693077,
"loss": 2.8405,
"theoretical_loss": 3.5906843506121087,
"tokens_seen": 1188364288
},
{
"epoch": 0.36,
"learning_rate": 0.00032320962888666,
"loss": 2.6566,
"theoretical_loss": 3.59066610275553,
"tokens_seen": 1188429824
},
{
"epoch": 0.36,
"learning_rate": 0.00032319959879638913,
"loss": 2.7427,
"theoretical_loss": 3.5906478561869424,
"tokens_seen": 1188495360
},
{
"epoch": 0.36,
"learning_rate": 0.00032318956870611837,
"loss": 3.0077,
"theoretical_loss": 3.590629610906184,
"tokens_seen": 1188560896
},
{
"epoch": 0.36,
"learning_rate": 0.00032317953861584755,
"loss": 2.8552,
"theoretical_loss": 3.5906113669130937,
"tokens_seen": 1188626432
},
{
"epoch": 0.36,
"learning_rate": 0.00032316950852557673,
"loss": 2.7814,
"theoretical_loss": 3.5905931242075084,
"tokens_seen": 1188691968
},
{
"epoch": 0.36,
"learning_rate": 0.0003231594784353059,
"loss": 2.965,
"theoretical_loss": 3.5905748827892667,
"tokens_seen": 1188757504
},
{
"epoch": 0.36,
"learning_rate": 0.00032314944834503515,
"loss": 2.6988,
"theoretical_loss": 3.5905566426582074,
"tokens_seen": 1188823040
},
{
"epoch": 0.36,
"learning_rate": 0.0003231394182547643,
"loss": 2.7156,
"theoretical_loss": 3.5905384038141674,
"tokens_seen": 1188888576
},
{
"epoch": 0.36,
"learning_rate": 0.0003231293881644935,
"loss": 2.8122,
"theoretical_loss": 3.5905201662569866,
"tokens_seen": 1188954112
},
{
"epoch": 0.36,
"learning_rate": 0.00032311935807422264,
"loss": 2.7288,
"theoretical_loss": 3.5905019299865018,
"tokens_seen": 1189019648
},
{
"epoch": 0.36,
"learning_rate": 0.0003231093279839519,
"loss": 2.9086,
"theoretical_loss": 3.590483695002552,
"tokens_seen": 1189085184
},
{
"epoch": 0.36,
"learning_rate": 0.00032309929789368105,
"loss": 2.4989,
"theoretical_loss": 3.5904654613049756,
"tokens_seen": 1189150720
},
{
"epoch": 0.36,
"learning_rate": 0.00032308926780341024,
"loss": 2.673,
"theoretical_loss": 3.5904472288936105,
"tokens_seen": 1189216256
},
{
"epoch": 0.36,
"learning_rate": 0.0003230792377131394,
"loss": 2.9794,
"theoretical_loss": 3.5904289977682957,
"tokens_seen": 1189281792
},
{
"epoch": 0.36,
"learning_rate": 0.0003230692076228686,
"loss": 2.619,
"theoretical_loss": 3.590410767928869,
"tokens_seen": 1189347328
},
{
"epoch": 0.36,
"learning_rate": 0.0003230591775325978,
"loss": 2.9585,
"theoretical_loss": 3.5903925393751694,
"tokens_seen": 1189412864
},
{
"epoch": 0.36,
"objective/train/docs_used": 676669,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9600603580474854,
"objective/train/theoretical_loss": 3.590374312107035,
"objective/train/tokens_used": 1209938400,
"theoretical_loss": 3.590374312107035,
"tokens_seen": 1189478400
},
{
"epoch": 0.36,
"learning_rate": 0.000323049147442327,
"loss": 2.756,
"theoretical_loss": 3.590374312107035,
"tokens_seen": 1189478400
},
{
"epoch": 0.36,
"learning_rate": 0.00032303911735205614,
"loss": 2.7584,
"theoretical_loss": 3.590356086124305,
"tokens_seen": 1189543936
},
{
"epoch": 0.36,
"learning_rate": 0.0003230290872617854,
"loss": 2.6635,
"theoretical_loss": 3.590337861426817,
"tokens_seen": 1189609472
},
{
"epoch": 0.36,
"learning_rate": 0.0003230190571715145,
"loss": 2.7539,
"theoretical_loss": 3.59031963801441,
"tokens_seen": 1189675008
},
{
"epoch": 0.36,
"learning_rate": 0.00032300902708124374,
"loss": 2.6752,
"theoretical_loss": 3.5903014158869224,
"tokens_seen": 1189740544
},
{
"epoch": 0.36,
"learning_rate": 0.0003229989969909729,
"loss": 2.7851,
"theoretical_loss": 3.5902831950441936,
"tokens_seen": 1189806080
},
{
"epoch": 0.36,
"learning_rate": 0.0003229889669007021,
"loss": 2.5704,
"theoretical_loss": 3.590264975486061,
"tokens_seen": 1189871616
},
{
"epoch": 0.36,
"learning_rate": 0.0003229789368104313,
"loss": 2.9446,
"theoretical_loss": 3.5902467572123644,
"tokens_seen": 1189937152
},
{
"epoch": 0.36,
"learning_rate": 0.0003229689067201605,
"loss": 2.7123,
"theoretical_loss": 3.5902285402229426,
"tokens_seen": 1190002688
},
{
"epoch": 0.36,
"learning_rate": 0.00032295887662988964,
"loss": 2.7531,
"theoretical_loss": 3.590210324517633,
"tokens_seen": 1190068224
},
{
"epoch": 0.36,
"learning_rate": 0.0003229488465396189,
"loss": 2.8656,
"theoretical_loss": 3.5901921100962753,
"tokens_seen": 1190133760
},
{
"epoch": 0.36,
"learning_rate": 0.000322938816449348,
"loss": 2.7605,
"theoretical_loss": 3.590173896958709,
"tokens_seen": 1190199296
},
{
"epoch": 0.36,
"learning_rate": 0.00032292878635907724,
"loss": 2.9852,
"theoretical_loss": 3.5901556851047713,
"tokens_seen": 1190264832
},
{
"epoch": 0.36,
"learning_rate": 0.0003229187562688064,
"loss": 2.7213,
"theoretical_loss": 3.590137474534302,
"tokens_seen": 1190330368
},
{
"epoch": 0.36,
"learning_rate": 0.0003229087261785356,
"loss": 2.5884,
"theoretical_loss": 3.5901192652471403,
"tokens_seen": 1190395904
},
{
"epoch": 0.36,
"learning_rate": 0.0003228986960882648,
"loss": 2.7389,
"theoretical_loss": 3.5901010572431247,
"tokens_seen": 1190461440
},
{
"epoch": 0.36,
"learning_rate": 0.00032288866599799397,
"loss": 2.7639,
"theoretical_loss": 3.5900828505220943,
"tokens_seen": 1190526976
},
{
"epoch": 0.36,
"learning_rate": 0.00032287863590772315,
"loss": 2.6307,
"theoretical_loss": 3.5900646450838876,
"tokens_seen": 1190592512
},
{
"epoch": 0.36,
"learning_rate": 0.0003228686058174524,
"loss": 2.4467,
"theoretical_loss": 3.590046440928344,
"tokens_seen": 1190658048
},
{
"epoch": 0.36,
"learning_rate": 0.0003228585757271815,
"loss": 2.9293,
"theoretical_loss": 3.590028238055303,
"tokens_seen": 1190723584
},
{
"epoch": 0.36,
"learning_rate": 0.00032284854563691074,
"loss": 2.5403,
"theoretical_loss": 3.5900100364646024,
"tokens_seen": 1190789120
},
{
"epoch": 0.36,
"learning_rate": 0.0003228385155466399,
"loss": 2.7504,
"theoretical_loss": 3.5899918361560825,
"tokens_seen": 1190854656
},
{
"epoch": 0.36,
"learning_rate": 0.0003228284854563691,
"loss": 2.6713,
"theoretical_loss": 3.5899736371295825,
"tokens_seen": 1190920192
},
{
"epoch": 0.36,
"learning_rate": 0.00032281845536609834,
"loss": 2.515,
"theoretical_loss": 3.5899554393849398,
"tokens_seen": 1190985728
},
{
"epoch": 0.36,
"learning_rate": 0.00032280842527582747,
"loss": 2.5318,
"theoretical_loss": 3.589937242921996,
"tokens_seen": 1191051264
},
{
"epoch": 0.36,
"objective/train/docs_used": 677774,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.19030499458313,
"objective/train/theoretical_loss": 3.5899190477405885,
"objective/train/tokens_used": 1211576800,
"theoretical_loss": 3.5899190477405885,
"tokens_seen": 1191116800
},
{
"epoch": 0.36,
"learning_rate": 0.0003227983951855567,
"loss": 2.4512,
"theoretical_loss": 3.5899190477405885,
"tokens_seen": 1191116800
},
{
"epoch": 0.36,
"learning_rate": 0.0003227883650952859,
"loss": 2.7765,
"theoretical_loss": 3.5899008538405575,
"tokens_seen": 1191182336
},
{
"epoch": 0.36,
"learning_rate": 0.00032277833500501507,
"loss": 2.8868,
"theoretical_loss": 3.5898826612217416,
"tokens_seen": 1191247872
},
{
"epoch": 0.36,
"learning_rate": 0.00032276830491474425,
"loss": 2.6325,
"theoretical_loss": 3.589864469883981,
"tokens_seen": 1191313408
},
{
"epoch": 0.36,
"learning_rate": 0.00032275827482447343,
"loss": 2.7816,
"theoretical_loss": 3.589846279827114,
"tokens_seen": 1191378944
},
{
"epoch": 0.36,
"learning_rate": 0.0003227482447342026,
"loss": 2.7742,
"theoretical_loss": 3.5898280910509808,
"tokens_seen": 1191444480
},
{
"epoch": 0.36,
"learning_rate": 0.00032273821464393185,
"loss": 2.4518,
"theoretical_loss": 3.5898099035554196,
"tokens_seen": 1191510016
},
{
"epoch": 0.36,
"learning_rate": 0.00032272818455366097,
"loss": 2.8739,
"theoretical_loss": 3.5897917173402716,
"tokens_seen": 1191575552
},
{
"epoch": 0.36,
"learning_rate": 0.0003227181544633902,
"loss": 2.4489,
"theoretical_loss": 3.589773532405375,
"tokens_seen": 1191641088
},
{
"epoch": 0.36,
"learning_rate": 0.00032270812437311933,
"loss": 2.754,
"theoretical_loss": 3.589755348750569,
"tokens_seen": 1191706624
},
{
"epoch": 0.36,
"learning_rate": 0.00032269809428284857,
"loss": 2.8626,
"theoretical_loss": 3.5897371663756945,
"tokens_seen": 1191772160
},
{
"epoch": 0.36,
"learning_rate": 0.00032268806419257775,
"loss": 2.9359,
"theoretical_loss": 3.58971898528059,
"tokens_seen": 1191837696
},
{
"epoch": 0.36,
"learning_rate": 0.00032267803410230693,
"loss": 2.7581,
"theoretical_loss": 3.5897008054650947,
"tokens_seen": 1191903232
},
{
"epoch": 0.36,
"learning_rate": 0.0003226680040120361,
"loss": 2.8072,
"theoretical_loss": 3.589682626929049,
"tokens_seen": 1191968768
},
{
"epoch": 0.36,
"learning_rate": 0.00032265797392176535,
"loss": 2.4849,
"theoretical_loss": 3.5896644496722927,
"tokens_seen": 1192034304
},
{
"epoch": 0.36,
"learning_rate": 0.0003226479438314945,
"loss": 2.7475,
"theoretical_loss": 3.5896462736946644,
"tokens_seen": 1192099840
},
{
"epoch": 0.36,
"learning_rate": 0.0003226379137412237,
"loss": 2.8035,
"theoretical_loss": 3.5896280989960045,
"tokens_seen": 1192165376
},
{
"epoch": 0.36,
"learning_rate": 0.00032262788365095284,
"loss": 2.4847,
"theoretical_loss": 3.589609925576153,
"tokens_seen": 1192230912
},
{
"epoch": 0.36,
"learning_rate": 0.0003226178535606821,
"loss": 2.5835,
"theoretical_loss": 3.5895917534349486,
"tokens_seen": 1192296448
},
{
"epoch": 0.36,
"learning_rate": 0.00032260782347041125,
"loss": 2.6635,
"theoretical_loss": 3.589573582572232,
"tokens_seen": 1192361984
},
{
"epoch": 0.36,
"learning_rate": 0.00032259779338014044,
"loss": 2.8255,
"theoretical_loss": 3.589555412987843,
"tokens_seen": 1192427520
},
{
"epoch": 0.36,
"learning_rate": 0.0003225877632898696,
"loss": 2.7527,
"theoretical_loss": 3.5895372446816203,
"tokens_seen": 1192493056
},
{
"epoch": 0.36,
"learning_rate": 0.0003225777331995988,
"loss": 2.6693,
"theoretical_loss": 3.5895190776534047,
"tokens_seen": 1192558592
},
{
"epoch": 0.36,
"learning_rate": 0.000322567703109328,
"loss": 3.0173,
"theoretical_loss": 3.5895009119030368,
"tokens_seen": 1192624128
},
{
"epoch": 0.36,
"learning_rate": 0.0003225576730190572,
"loss": 2.8358,
"theoretical_loss": 3.5894827474303543,
"tokens_seen": 1192689664
},
{
"epoch": 0.36,
"objective/train/docs_used": 678420,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.657104015350342,
"objective/train/theoretical_loss": 3.5894645842351993,
"objective/train/tokens_used": 1213215200,
"theoretical_loss": 3.5894645842351993,
"tokens_seen": 1192755200
},
{
"epoch": 0.36,
"learning_rate": 0.00032254764292878634,
"loss": 2.6667,
"theoretical_loss": 3.5894645842351993,
"tokens_seen": 1192755200
},
{
"epoch": 0.36,
"learning_rate": 0.0003225376128385156,
"loss": 2.9124,
"theoretical_loss": 3.589446422317411,
"tokens_seen": 1192820736
},
{
"epoch": 0.36,
"learning_rate": 0.0003225275827482447,
"loss": 2.7861,
"theoretical_loss": 3.5894282616768285,
"tokens_seen": 1192886272
},
{
"epoch": 0.36,
"learning_rate": 0.00032251755265797394,
"loss": 2.8851,
"theoretical_loss": 3.5894101023132934,
"tokens_seen": 1192951808
},
{
"epoch": 0.36,
"learning_rate": 0.0003225075225677031,
"loss": 2.6693,
"theoretical_loss": 3.5893919442266444,
"tokens_seen": 1193017344
},
{
"epoch": 0.36,
"learning_rate": 0.0003224974924774323,
"loss": 2.6664,
"theoretical_loss": 3.5893737874167226,
"tokens_seen": 1193082880
},
{
"epoch": 0.36,
"learning_rate": 0.0003224874623871615,
"loss": 2.5494,
"theoretical_loss": 3.5893556318833673,
"tokens_seen": 1193148416
},
{
"epoch": 0.36,
"learning_rate": 0.0003224774322968907,
"loss": 2.7376,
"theoretical_loss": 3.5893374776264197,
"tokens_seen": 1193213952
},
{
"epoch": 0.36,
"learning_rate": 0.00032246740220661984,
"loss": 2.5839,
"theoretical_loss": 3.589319324645719,
"tokens_seen": 1193279488
},
{
"epoch": 0.36,
"learning_rate": 0.0003224573721163491,
"loss": 2.8386,
"theoretical_loss": 3.589301172941105,
"tokens_seen": 1193345024
},
{
"epoch": 0.36,
"learning_rate": 0.0003224473420260782,
"loss": 2.6211,
"theoretical_loss": 3.5892830225124186,
"tokens_seen": 1193410560
},
{
"epoch": 0.36,
"learning_rate": 0.00032243731193580744,
"loss": 2.6716,
"theoretical_loss": 3.5892648733595007,
"tokens_seen": 1193476096
},
{
"epoch": 0.36,
"learning_rate": 0.0003224272818455366,
"loss": 2.6551,
"theoretical_loss": 3.589246725482191,
"tokens_seen": 1193541632
},
{
"epoch": 0.36,
"learning_rate": 0.0003224172517552658,
"loss": 2.6357,
"theoretical_loss": 3.589228578880329,
"tokens_seen": 1193607168
},
{
"epoch": 0.36,
"learning_rate": 0.000322407221664995,
"loss": 2.4128,
"theoretical_loss": 3.589210433553756,
"tokens_seen": 1193672704
},
{
"epoch": 0.36,
"learning_rate": 0.00032239719157472417,
"loss": 2.7972,
"theoretical_loss": 3.5891922895023125,
"tokens_seen": 1193738240
},
{
"epoch": 0.36,
"learning_rate": 0.00032238716148445335,
"loss": 2.7134,
"theoretical_loss": 3.5891741467258385,
"tokens_seen": 1193803776
},
{
"epoch": 0.36,
"learning_rate": 0.0003223771313941826,
"loss": 2.5095,
"theoretical_loss": 3.589156005224174,
"tokens_seen": 1193869312
},
{
"epoch": 0.36,
"learning_rate": 0.0003223671013039117,
"loss": 2.6976,
"theoretical_loss": 3.58913786499716,
"tokens_seen": 1193934848
},
{
"epoch": 0.36,
"learning_rate": 0.00032235707121364095,
"loss": 2.7431,
"theoretical_loss": 3.589119726044637,
"tokens_seen": 1194000384
},
{
"epoch": 0.36,
"learning_rate": 0.0003223470411233701,
"loss": 2.8417,
"theoretical_loss": 3.5891015883664457,
"tokens_seen": 1194065920
},
{
"epoch": 0.36,
"learning_rate": 0.0003223370110330993,
"loss": 2.5904,
"theoretical_loss": 3.5890834519624257,
"tokens_seen": 1194131456
},
{
"epoch": 0.36,
"learning_rate": 0.0003223269809428285,
"loss": 2.4864,
"theoretical_loss": 3.5890653168324187,
"tokens_seen": 1194196992
},
{
"epoch": 0.36,
"learning_rate": 0.00032231695085255767,
"loss": 2.5469,
"theoretical_loss": 3.5890471829762647,
"tokens_seen": 1194262528
},
{
"epoch": 0.36,
"learning_rate": 0.00032230692076228685,
"loss": 2.7724,
"theoretical_loss": 3.5890290503938047,
"tokens_seen": 1194328064
},
{
"epoch": 0.36,
"objective/train/docs_used": 679445,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4121041297912598,
"objective/train/theoretical_loss": 3.5890109190848785,
"objective/train/tokens_used": 1214853600,
"theoretical_loss": 3.5890109190848785,
"tokens_seen": 1194393600
},
{
"epoch": 0.36,
"learning_rate": 0.0003222968906720161,
"loss": 2.7183,
"theoretical_loss": 3.5890109190848785,
"tokens_seen": 1194393600
},
{
"epoch": 0.36,
"learning_rate": 0.0003222868605817452,
"loss": 2.6152,
"theoretical_loss": 3.5889927890493274,
"tokens_seen": 1194459136
},
{
"epoch": 0.36,
"learning_rate": 0.00032227683049147445,
"loss": 2.4245,
"theoretical_loss": 3.5889746602869925,
"tokens_seen": 1194524672
},
{
"epoch": 0.36,
"learning_rate": 0.0003222668004012036,
"loss": 2.8445,
"theoretical_loss": 3.588956532797714,
"tokens_seen": 1194590208
},
{
"epoch": 0.36,
"learning_rate": 0.0003222567703109328,
"loss": 2.7595,
"theoretical_loss": 3.5889384065813323,
"tokens_seen": 1194655744
},
{
"epoch": 0.36,
"learning_rate": 0.000322246740220662,
"loss": 2.9345,
"theoretical_loss": 3.588920281637689,
"tokens_seen": 1194721280
},
{
"epoch": 0.36,
"learning_rate": 0.00032223671013039117,
"loss": 2.6782,
"theoretical_loss": 3.5889021579666247,
"tokens_seen": 1194786816
},
{
"epoch": 0.36,
"learning_rate": 0.00032222668004012035,
"loss": 2.6425,
"theoretical_loss": 3.58888403556798,
"tokens_seen": 1194852352
},
{
"epoch": 0.36,
"learning_rate": 0.00032221664994984953,
"loss": 2.7263,
"theoretical_loss": 3.5888659144415964,
"tokens_seen": 1194917888
},
{
"epoch": 0.36,
"learning_rate": 0.0003222066198595787,
"loss": 2.6193,
"theoretical_loss": 3.5888477945873136,
"tokens_seen": 1194983424
},
{
"epoch": 0.36,
"learning_rate": 0.00032219658976930795,
"loss": 2.5633,
"theoretical_loss": 3.5888296760049743,
"tokens_seen": 1195048960
},
{
"epoch": 0.36,
"learning_rate": 0.0003221865596790371,
"loss": 2.8446,
"theoretical_loss": 3.5888115586944176,
"tokens_seen": 1195114496
},
{
"epoch": 0.36,
"learning_rate": 0.0003221765295887663,
"loss": 2.7141,
"theoretical_loss": 3.5887934426554855,
"tokens_seen": 1195180032
},
{
"epoch": 0.36,
"learning_rate": 0.0003221664994984955,
"loss": 2.5448,
"theoretical_loss": 3.588775327888019,
"tokens_seen": 1195245568
},
{
"epoch": 0.36,
"learning_rate": 0.0003221564694082247,
"loss": 2.6758,
"theoretical_loss": 3.588757214391859,
"tokens_seen": 1195311104
},
{
"epoch": 0.36,
"learning_rate": 0.00032214643931795386,
"loss": 2.7138,
"theoretical_loss": 3.588739102166847,
"tokens_seen": 1195376640
},
{
"epoch": 0.36,
"learning_rate": 0.00032213640922768304,
"loss": 2.7448,
"theoretical_loss": 3.5887209912128233,
"tokens_seen": 1195442176
},
{
"epoch": 0.36,
"learning_rate": 0.0003221263791374122,
"loss": 2.9608,
"theoretical_loss": 3.5887028815296294,
"tokens_seen": 1195507712
},
{
"epoch": 0.36,
"learning_rate": 0.00032211634904714145,
"loss": 2.7007,
"theoretical_loss": 3.588684773117107,
"tokens_seen": 1195573248
},
{
"epoch": 0.36,
"learning_rate": 0.0003221063189568706,
"loss": 2.7629,
"theoretical_loss": 3.5886666659750963,
"tokens_seen": 1195638784
},
{
"epoch": 0.36,
"learning_rate": 0.0003220962888665998,
"loss": 2.7929,
"theoretical_loss": 3.5886485601034397,
"tokens_seen": 1195704320
},
{
"epoch": 0.36,
"learning_rate": 0.00032208625877632894,
"loss": 2.619,
"theoretical_loss": 3.5886304555019772,
"tokens_seen": 1195769856
},
{
"epoch": 0.36,
"learning_rate": 0.0003220762286860582,
"loss": 2.8329,
"theoretical_loss": 3.588612352170551,
"tokens_seen": 1195835392
},
{
"epoch": 0.36,
"learning_rate": 0.0003220661985957874,
"loss": 2.8715,
"theoretical_loss": 3.5885942501090025,
"tokens_seen": 1195900928
},
{
"epoch": 0.36,
"learning_rate": 0.00032205616850551654,
"loss": 2.5133,
"theoretical_loss": 3.588576149317172,
"tokens_seen": 1195966464
},
{
"epoch": 0.36,
"objective/train/docs_used": 680084,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8646416664123535,
"objective/train/theoretical_loss": 3.588558049794902,
"objective/train/tokens_used": 1216492000,
"theoretical_loss": 3.588558049794902,
"tokens_seen": 1196032000
},
{
"epoch": 0.36,
"learning_rate": 0.0003220461384152458,
"loss": 2.75,
"theoretical_loss": 3.588558049794902,
"tokens_seen": 1196032000
},
{
"epoch": 0.36,
"learning_rate": 0.0003220361083249749,
"loss": 2.5503,
"theoretical_loss": 3.588539951542033,
"tokens_seen": 1196097536
},
{
"epoch": 0.36,
"learning_rate": 0.00032202607823470414,
"loss": 2.7361,
"theoretical_loss": 3.588521854558407,
"tokens_seen": 1196163072
},
{
"epoch": 0.36,
"learning_rate": 0.0003220160481444333,
"loss": 2.9557,
"theoretical_loss": 3.5885037588438657,
"tokens_seen": 1196228608
},
{
"epoch": 0.36,
"learning_rate": 0.0003220060180541625,
"loss": 2.8562,
"theoretical_loss": 3.5884856643982497,
"tokens_seen": 1196294144
},
{
"epoch": 0.36,
"learning_rate": 0.0003219959879638917,
"loss": 2.7628,
"theoretical_loss": 3.5884675712214014,
"tokens_seen": 1196359680
},
{
"epoch": 0.36,
"learning_rate": 0.0003219859578736209,
"loss": 2.7625,
"theoretical_loss": 3.588449479313162,
"tokens_seen": 1196425216
},
{
"epoch": 0.36,
"learning_rate": 0.00032197592778335004,
"loss": 2.681,
"theoretical_loss": 3.5884313886733725,
"tokens_seen": 1196490752
},
{
"epoch": 0.36,
"learning_rate": 0.0003219658976930793,
"loss": 2.6811,
"theoretical_loss": 3.5884132993018754,
"tokens_seen": 1196556288
},
{
"epoch": 0.36,
"learning_rate": 0.0003219558676028084,
"loss": 3.0184,
"theoretical_loss": 3.588395211198512,
"tokens_seen": 1196621824
},
{
"epoch": 0.36,
"learning_rate": 0.00032194583751253764,
"loss": 2.7562,
"theoretical_loss": 3.5883771243631237,
"tokens_seen": 1196687360
},
{
"epoch": 0.36,
"learning_rate": 0.0003219358074222668,
"loss": 2.8586,
"theoretical_loss": 3.588359038795552,
"tokens_seen": 1196752896
},
{
"epoch": 0.36,
"learning_rate": 0.000321925777331996,
"loss": 2.8663,
"theoretical_loss": 3.5883409544956395,
"tokens_seen": 1196818432
},
{
"epoch": 0.36,
"learning_rate": 0.0003219157472417252,
"loss": 2.5794,
"theoretical_loss": 3.5883228714632276,
"tokens_seen": 1196883968
},
{
"epoch": 0.36,
"learning_rate": 0.00032190571715145437,
"loss": 2.6597,
"theoretical_loss": 3.5883047896981575,
"tokens_seen": 1196949504
},
{
"epoch": 0.36,
"learning_rate": 0.00032189568706118355,
"loss": 2.7205,
"theoretical_loss": 3.588286709200271,
"tokens_seen": 1197015040
},
{
"epoch": 0.36,
"learning_rate": 0.0003218856569709128,
"loss": 2.3492,
"theoretical_loss": 3.588268629969411,
"tokens_seen": 1197080576
},
{
"epoch": 0.36,
"learning_rate": 0.0003218756268806419,
"loss": 2.9369,
"theoretical_loss": 3.5882505520054186,
"tokens_seen": 1197146112
},
{
"epoch": 0.36,
"learning_rate": 0.00032186559679037115,
"loss": 2.6266,
"theoretical_loss": 3.5882324753081356,
"tokens_seen": 1197211648
},
{
"epoch": 0.36,
"learning_rate": 0.0003218555667001003,
"loss": 2.8349,
"theoretical_loss": 3.588214399877404,
"tokens_seen": 1197277184
},
{
"epoch": 0.36,
"learning_rate": 0.0003218455366098295,
"loss": 3.0148,
"theoretical_loss": 3.588196325713066,
"tokens_seen": 1197342720
},
{
"epoch": 0.36,
"learning_rate": 0.0003218355065195587,
"loss": 2.6358,
"theoretical_loss": 3.588178252814963,
"tokens_seen": 1197408256
},
{
"epoch": 0.36,
"learning_rate": 0.00032182547642928787,
"loss": 2.8407,
"theoretical_loss": 3.5881601811829373,
"tokens_seen": 1197473792
},
{
"epoch": 0.36,
"learning_rate": 0.00032181544633901705,
"loss": 2.8879,
"theoretical_loss": 3.5881421108168317,
"tokens_seen": 1197539328
},
{
"epoch": 0.36,
"learning_rate": 0.0003218054162487463,
"loss": 2.6884,
"theoretical_loss": 3.588124041716487,
"tokens_seen": 1197604864
},
{
"epoch": 0.36,
"objective/train/docs_used": 681502,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6860108375549316,
"objective/train/theoretical_loss": 3.5881059738817456,
"objective/train/tokens_used": 1218130400,
"theoretical_loss": 3.5881059738817456,
"tokens_seen": 1197670400
},
{
"epoch": 0.36,
"learning_rate": 0.0003217953861584754,
"loss": 2.7064,
"theoretical_loss": 3.5881059738817456,
"tokens_seen": 1197670400
},
{
"epoch": 0.36,
"learning_rate": 0.00032178535606820465,
"loss": 2.7804,
"theoretical_loss": 3.58808790731245,
"tokens_seen": 1197735936
},
{
"epoch": 0.36,
"learning_rate": 0.0003217753259779338,
"loss": 2.7077,
"theoretical_loss": 3.588069842008442,
"tokens_seen": 1197801472
},
{
"epoch": 0.36,
"learning_rate": 0.000321765295887663,
"loss": 2.6148,
"theoretical_loss": 3.5880517779695644,
"tokens_seen": 1197867008
},
{
"epoch": 0.36,
"learning_rate": 0.0003217552657973922,
"loss": 2.8128,
"theoretical_loss": 3.5880337151956585,
"tokens_seen": 1197932544
},
{
"epoch": 0.36,
"learning_rate": 0.00032174523570712137,
"loss": 2.5728,
"theoretical_loss": 3.588015653686567,
"tokens_seen": 1197998080
},
{
"epoch": 0.36,
"learning_rate": 0.00032173520561685055,
"loss": 2.7221,
"theoretical_loss": 3.5879975934421324,
"tokens_seen": 1198063616
},
{
"epoch": 0.36,
"learning_rate": 0.00032172517552657974,
"loss": 2.5329,
"theoretical_loss": 3.587979534462196,
"tokens_seen": 1198129152
},
{
"epoch": 0.36,
"learning_rate": 0.0003217151454363089,
"loss": 2.7608,
"theoretical_loss": 3.5879614767466013,
"tokens_seen": 1198194688
},
{
"epoch": 0.36,
"learning_rate": 0.00032170511534603815,
"loss": 2.4442,
"theoretical_loss": 3.58794342029519,
"tokens_seen": 1198260224
},
{
"epoch": 0.36,
"learning_rate": 0.0003216950852557673,
"loss": 2.8497,
"theoretical_loss": 3.5879253651078047,
"tokens_seen": 1198325760
},
{
"epoch": 0.36,
"learning_rate": 0.0003216850551654965,
"loss": 2.6229,
"theoretical_loss": 3.5879073111842876,
"tokens_seen": 1198391296
},
{
"epoch": 0.36,
"learning_rate": 0.0003216750250752257,
"loss": 2.6468,
"theoretical_loss": 3.5878892585244806,
"tokens_seen": 1198456832
},
{
"epoch": 0.36,
"learning_rate": 0.0003216649949849549,
"loss": 2.7774,
"theoretical_loss": 3.5878712071282273,
"tokens_seen": 1198522368
},
{
"epoch": 0.36,
"learning_rate": 0.00032165496489468406,
"loss": 2.552,
"theoretical_loss": 3.5878531569953696,
"tokens_seen": 1198587904
},
{
"epoch": 0.36,
"learning_rate": 0.00032164493480441324,
"loss": 2.7501,
"theoretical_loss": 3.58783510812575,
"tokens_seen": 1198653440
},
{
"epoch": 0.36,
"learning_rate": 0.0003216349047141424,
"loss": 2.8143,
"theoretical_loss": 3.5878170605192103,
"tokens_seen": 1198718976
},
{
"epoch": 0.36,
"learning_rate": 0.00032162487462387165,
"loss": 2.6673,
"theoretical_loss": 3.5877990141755944,
"tokens_seen": 1198784512
},
{
"epoch": 0.36,
"learning_rate": 0.0003216148445336008,
"loss": 2.8903,
"theoretical_loss": 3.5877809690947444,
"tokens_seen": 1198850048
},
{
"epoch": 0.36,
"learning_rate": 0.00032160481444333,
"loss": 2.4771,
"theoretical_loss": 3.5877629252765026,
"tokens_seen": 1198915584
},
{
"epoch": 0.36,
"learning_rate": 0.00032159478435305914,
"loss": 2.5476,
"theoretical_loss": 3.587744882720712,
"tokens_seen": 1198981120
},
{
"epoch": 0.36,
"learning_rate": 0.0003215847542627884,
"loss": 2.6365,
"theoretical_loss": 3.587726841427215,
"tokens_seen": 1199046656
},
{
"epoch": 0.36,
"learning_rate": 0.00032157472417251756,
"loss": 2.8072,
"theoretical_loss": 3.5877088013958547,
"tokens_seen": 1199112192
},
{
"epoch": 0.36,
"learning_rate": 0.00032156469408224674,
"loss": 2.8503,
"theoretical_loss": 3.5876907626264734,
"tokens_seen": 1199177728
},
{
"epoch": 0.36,
"learning_rate": 0.0003215546639919759,
"loss": 2.4098,
"theoretical_loss": 3.5876727251189138,
"tokens_seen": 1199243264
},
{
"epoch": 0.36,
"objective/train/docs_used": 682215,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9197044372558594,
"objective/train/theoretical_loss": 3.5876546888730187,
"objective/train/tokens_used": 1219768800,
"theoretical_loss": 3.5876546888730187,
"tokens_seen": 1199308800
},
{
"epoch": 0.36,
"learning_rate": 0.0003215446339017051,
"loss": 2.7925,
"theoretical_loss": 3.5876546888730187,
"tokens_seen": 1199308800
},
{
"epoch": 0.36,
"learning_rate": 0.0003215346038114343,
"loss": 2.9501,
"theoretical_loss": 3.5876366538886315,
"tokens_seen": 1199374336
},
{
"epoch": 0.36,
"learning_rate": 0.0003215245737211635,
"loss": 2.5173,
"theoretical_loss": 3.5876186201655944,
"tokens_seen": 1199439872
},
{
"epoch": 0.36,
"learning_rate": 0.00032151454363089265,
"loss": 2.9603,
"theoretical_loss": 3.5876005877037507,
"tokens_seen": 1199505408
},
{
"epoch": 0.36,
"learning_rate": 0.0003215045135406219,
"loss": 2.4422,
"theoretical_loss": 3.587582556502943,
"tokens_seen": 1199570944
},
{
"epoch": 0.36,
"learning_rate": 0.00032149448345035106,
"loss": 2.659,
"theoretical_loss": 3.5875645265630145,
"tokens_seen": 1199636480
},
{
"epoch": 0.36,
"learning_rate": 0.00032148445336008024,
"loss": 2.6048,
"theoretical_loss": 3.5875464978838076,
"tokens_seen": 1199702016
},
{
"epoch": 0.36,
"learning_rate": 0.0003214744232698094,
"loss": 2.669,
"theoretical_loss": 3.587528470465166,
"tokens_seen": 1199767552
},
{
"epoch": 0.36,
"learning_rate": 0.0003214643931795386,
"loss": 2.5343,
"theoretical_loss": 3.5875104443069326,
"tokens_seen": 1199833088
},
{
"epoch": 0.36,
"learning_rate": 0.0003214543630892678,
"loss": 2.7642,
"theoretical_loss": 3.5874924194089504,
"tokens_seen": 1199898624
},
{
"epoch": 0.36,
"learning_rate": 0.000321444332998997,
"loss": 2.6452,
"theoretical_loss": 3.5874743957710615,
"tokens_seen": 1199964160
},
{
"epoch": 0.36,
"learning_rate": 0.00032143430290872615,
"loss": 2.5025,
"theoretical_loss": 3.5874563733931106,
"tokens_seen": 1200029696
},
{
"epoch": 0.36,
"learning_rate": 0.0003214242728184554,
"loss": 2.7499,
"theoretical_loss": 3.5874383522749396,
"tokens_seen": 1200095232
},
{
"epoch": 0.36,
"learning_rate": 0.0003214142427281845,
"loss": 2.9532,
"theoretical_loss": 3.587420332416392,
"tokens_seen": 1200160768
},
{
"epoch": 0.36,
"learning_rate": 0.00032140421263791375,
"loss": 2.6131,
"theoretical_loss": 3.5874023138173117,
"tokens_seen": 1200226304
},
{
"epoch": 0.36,
"learning_rate": 0.00032139418254764293,
"loss": 2.652,
"theoretical_loss": 3.5873842964775404,
"tokens_seen": 1200291840
},
{
"epoch": 0.36,
"learning_rate": 0.0003213841524573721,
"loss": 2.7076,
"theoretical_loss": 3.587366280396923,
"tokens_seen": 1200357376
},
{
"epoch": 0.36,
"learning_rate": 0.0003213741223671013,
"loss": 2.4923,
"theoretical_loss": 3.5873482655753013,
"tokens_seen": 1200422912
},
{
"epoch": 0.36,
"learning_rate": 0.0003213640922768305,
"loss": 2.9235,
"theoretical_loss": 3.5873302520125194,
"tokens_seen": 1200488448
},
{
"epoch": 0.36,
"learning_rate": 0.00032135406218655965,
"loss": 2.7501,
"theoretical_loss": 3.587312239708421,
"tokens_seen": 1200553984
},
{
"epoch": 0.36,
"learning_rate": 0.0003213440320962889,
"loss": 2.6007,
"theoretical_loss": 3.587294228662848,
"tokens_seen": 1200619520
},
{
"epoch": 0.36,
"learning_rate": 0.000321334002006018,
"loss": 2.5353,
"theoretical_loss": 3.5872762188756453,
"tokens_seen": 1200685056
},
{
"epoch": 0.36,
"learning_rate": 0.00032132397191574725,
"loss": 2.5841,
"theoretical_loss": 3.587258210346656,
"tokens_seen": 1200750592
},
{
"epoch": 0.36,
"learning_rate": 0.0003213139418254765,
"loss": 2.5549,
"theoretical_loss": 3.5872402030757224,
"tokens_seen": 1200816128
},
{
"epoch": 0.36,
"learning_rate": 0.0003213039117352056,
"loss": 2.8397,
"theoretical_loss": 3.5872221970626894,
"tokens_seen": 1200881664
},
{
"epoch": 0.36,
"objective/train/docs_used": 683386,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.876089572906494,
"objective/train/theoretical_loss": 3.5872041923074,
"objective/train/tokens_used": 1221407200,
"theoretical_loss": 3.5872041923074,
"tokens_seen": 1200947200
},
{
"epoch": 0.36,
"learning_rate": 0.00032129388164493485,
"loss": 2.7921,
"theoretical_loss": 3.5872041923074,
"tokens_seen": 1200947200
},
{
"epoch": 0.36,
"learning_rate": 0.000321283851554664,
"loss": 2.7806,
"theoretical_loss": 3.587186188809697,
"tokens_seen": 1201012736
},
{
"epoch": 0.36,
"learning_rate": 0.0003212738214643932,
"loss": 2.3056,
"theoretical_loss": 3.587168186569425,
"tokens_seen": 1201078272
},
{
"epoch": 0.36,
"learning_rate": 0.0003212637913741224,
"loss": 2.5618,
"theoretical_loss": 3.5871501855864274,
"tokens_seen": 1201143808
},
{
"epoch": 0.36,
"learning_rate": 0.0003212537612838516,
"loss": 2.6997,
"theoretical_loss": 3.587132185860547,
"tokens_seen": 1201209344
},
{
"epoch": 0.36,
"learning_rate": 0.00032124373119358075,
"loss": 3.0025,
"theoretical_loss": 3.587114187391628,
"tokens_seen": 1201274880
},
{
"epoch": 0.36,
"learning_rate": 0.00032123370110330994,
"loss": 2.6524,
"theoretical_loss": 3.587096190179514,
"tokens_seen": 1201340416
},
{
"epoch": 0.36,
"learning_rate": 0.0003212236710130391,
"loss": 2.5902,
"theoretical_loss": 3.5870781942240493,
"tokens_seen": 1201405952
},
{
"epoch": 0.36,
"learning_rate": 0.00032121364092276835,
"loss": 2.74,
"theoretical_loss": 3.5870601995250766,
"tokens_seen": 1201471488
},
{
"epoch": 0.36,
"learning_rate": 0.0003212036108324975,
"loss": 2.7347,
"theoretical_loss": 3.58704220608244,
"tokens_seen": 1201537024
},
{
"epoch": 0.36,
"learning_rate": 0.0003211935807422267,
"loss": 2.7035,
"theoretical_loss": 3.587024213895983,
"tokens_seen": 1201602560
},
{
"epoch": 0.36,
"learning_rate": 0.0003211835506519559,
"loss": 2.8717,
"theoretical_loss": 3.58700622296555,
"tokens_seen": 1201668096
},
{
"epoch": 0.36,
"learning_rate": 0.0003211735205616851,
"loss": 2.5479,
"theoretical_loss": 3.5869882332909846,
"tokens_seen": 1201733632
},
{
"epoch": 0.36,
"learning_rate": 0.00032116349047141426,
"loss": 2.7331,
"theoretical_loss": 3.5869702448721306,
"tokens_seen": 1201799168
},
{
"epoch": 0.36,
"learning_rate": 0.00032115346038114344,
"loss": 2.6757,
"theoretical_loss": 3.5869522577088317,
"tokens_seen": 1201864704
},
{
"epoch": 0.36,
"learning_rate": 0.0003211434302908726,
"loss": 2.9767,
"theoretical_loss": 3.5869342718009323,
"tokens_seen": 1201930240
},
{
"epoch": 0.36,
"learning_rate": 0.00032113340020060185,
"loss": 2.5559,
"theoretical_loss": 3.586916287148276,
"tokens_seen": 1201995776
},
{
"epoch": 0.36,
"learning_rate": 0.000321123370110331,
"loss": 2.7281,
"theoretical_loss": 3.5868983037507065,
"tokens_seen": 1202061312
},
{
"epoch": 0.36,
"learning_rate": 0.0003211133400200602,
"loss": 2.6022,
"theoretical_loss": 3.586880321608068,
"tokens_seen": 1202126848
},
{
"epoch": 0.36,
"learning_rate": 0.00032110330992978934,
"loss": 2.9402,
"theoretical_loss": 3.586862340720205,
"tokens_seen": 1202192384
},
{
"epoch": 0.36,
"learning_rate": 0.0003210932798395186,
"loss": 2.7419,
"theoretical_loss": 3.5868443610869605,
"tokens_seen": 1202257920
},
{
"epoch": 0.36,
"learning_rate": 0.00032108324974924776,
"loss": 2.842,
"theoretical_loss": 3.58682638270818,
"tokens_seen": 1202323456
},
{
"epoch": 0.36,
"learning_rate": 0.00032107321965897694,
"loss": 2.7231,
"theoretical_loss": 3.5868084055837066,
"tokens_seen": 1202388992
},
{
"epoch": 0.36,
"learning_rate": 0.0003210631895687061,
"loss": 2.8937,
"theoretical_loss": 3.5867904297133846,
"tokens_seen": 1202454528
},
{
"epoch": 0.36,
"learning_rate": 0.0003210531594784353,
"loss": 2.803,
"theoretical_loss": 3.5867724550970577,
"tokens_seen": 1202520064
},
{
"epoch": 0.36,
"objective/train/docs_used": 683937,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8448095321655273,
"objective/train/theoretical_loss": 3.5867544817345713,
"objective/train/tokens_used": 1223045600,
"theoretical_loss": 3.5867544817345713,
"tokens_seen": 1202585600
},
{
"epoch": 0.36,
"learning_rate": 0.0003210431293881645,
"loss": 2.7715,
"theoretical_loss": 3.5867544817345713,
"tokens_seen": 1202585600
},
{
"epoch": 0.36,
"learning_rate": 0.0003210330992978937,
"loss": 2.8258,
"theoretical_loss": 3.5867365096257684,
"tokens_seen": 1202651136
},
{
"epoch": 0.36,
"learning_rate": 0.00032102306920762285,
"loss": 2.7905,
"theoretical_loss": 3.5867185387704943,
"tokens_seen": 1202716672
},
{
"epoch": 0.36,
"learning_rate": 0.0003210130391173521,
"loss": 2.6062,
"theoretical_loss": 3.586700569168592,
"tokens_seen": 1202782208
},
{
"epoch": 0.36,
"learning_rate": 0.00032100300902708126,
"loss": 2.5707,
"theoretical_loss": 3.5866826008199073,
"tokens_seen": 1202847744
},
{
"epoch": 0.36,
"learning_rate": 0.00032099297893681044,
"loss": 2.8216,
"theoretical_loss": 3.5866646337242836,
"tokens_seen": 1202913280
},
{
"epoch": 0.36,
"learning_rate": 0.0003209829488465396,
"loss": 2.8189,
"theoretical_loss": 3.586646667881565,
"tokens_seen": 1202978816
},
{
"epoch": 0.36,
"learning_rate": 0.0003209729187562688,
"loss": 2.8828,
"theoretical_loss": 3.586628703291596,
"tokens_seen": 1203044352
},
{
"epoch": 0.36,
"learning_rate": 0.000320962888665998,
"loss": 2.8984,
"theoretical_loss": 3.5866107399542217,
"tokens_seen": 1203109888
},
{
"epoch": 0.36,
"learning_rate": 0.0003209528585757272,
"loss": 2.637,
"theoretical_loss": 3.5865927778692868,
"tokens_seen": 1203175424
},
{
"epoch": 0.36,
"learning_rate": 0.00032094282848545635,
"loss": 2.7072,
"theoretical_loss": 3.5865748170366345,
"tokens_seen": 1203240960
},
{
"epoch": 0.36,
"learning_rate": 0.0003209327983951856,
"loss": 3.0161,
"theoretical_loss": 3.5865568574561095,
"tokens_seen": 1203306496
},
{
"epoch": 0.36,
"learning_rate": 0.0003209227683049147,
"loss": 2.8436,
"theoretical_loss": 3.5865388991275573,
"tokens_seen": 1203372032
},
{
"epoch": 0.36,
"learning_rate": 0.00032091273821464395,
"loss": 2.7572,
"theoretical_loss": 3.5865209420508215,
"tokens_seen": 1203437568
},
{
"epoch": 0.36,
"learning_rate": 0.00032090270812437313,
"loss": 2.7479,
"theoretical_loss": 3.5865029862257476,
"tokens_seen": 1203503104
},
{
"epoch": 0.36,
"learning_rate": 0.0003208926780341023,
"loss": 2.6151,
"theoretical_loss": 3.5864850316521792,
"tokens_seen": 1203568640
},
{
"epoch": 0.36,
"learning_rate": 0.0003208826479438315,
"loss": 2.726,
"theoretical_loss": 3.586467078329961,
"tokens_seen": 1203634176
},
{
"epoch": 0.36,
"learning_rate": 0.0003208726178535607,
"loss": 2.6799,
"theoretical_loss": 3.586449126258939,
"tokens_seen": 1203699712
},
{
"epoch": 0.36,
"learning_rate": 0.00032086258776328985,
"loss": 2.8371,
"theoretical_loss": 3.586431175438956,
"tokens_seen": 1203765248
},
{
"epoch": 0.36,
"learning_rate": 0.0003208525576730191,
"loss": 2.6901,
"theoretical_loss": 3.5864132258698582,
"tokens_seen": 1203830784
},
{
"epoch": 0.36,
"learning_rate": 0.0003208425275827482,
"loss": 2.5776,
"theoretical_loss": 3.58639527755149,
"tokens_seen": 1203896320
},
{
"epoch": 0.36,
"learning_rate": 0.00032083249749247745,
"loss": 2.8478,
"theoretical_loss": 3.5863773304836952,
"tokens_seen": 1203961856
},
{
"epoch": 0.36,
"learning_rate": 0.00032082246740220663,
"loss": 2.8211,
"theoretical_loss": 3.58635938466632,
"tokens_seen": 1204027392
},
{
"epoch": 0.36,
"learning_rate": 0.0003208124373119358,
"loss": 2.7746,
"theoretical_loss": 3.586341440099208,
"tokens_seen": 1204092928
},
{
"epoch": 0.36,
"learning_rate": 0.000320802407221665,
"loss": 2.7835,
"theoretical_loss": 3.586323496782205,
"tokens_seen": 1204158464
},
{
"epoch": 0.36,
"objective/train/docs_used": 684175,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0094797611236572,
"objective/train/theoretical_loss": 3.5863055547151554,
"objective/train/tokens_used": 1224684000,
"theoretical_loss": 3.5863055547151554,
"tokens_seen": 1204224000
},
{
"epoch": 0.36,
"learning_rate": 0.0003207923771313942,
"loss": 2.868,
"theoretical_loss": 3.5863055547151554,
"tokens_seen": 1204224000
},
{
"epoch": 0.36,
"learning_rate": 0.00032078234704112336,
"loss": 2.6753,
"theoretical_loss": 3.586287613897904,
"tokens_seen": 1204289536
},
{
"epoch": 0.36,
"learning_rate": 0.0003207723169508526,
"loss": 2.8803,
"theoretical_loss": 3.586269674330296,
"tokens_seen": 1204355072
},
{
"epoch": 0.36,
"learning_rate": 0.0003207622868605817,
"loss": 2.8664,
"theoretical_loss": 3.5862517360121773,
"tokens_seen": 1204420608
},
{
"epoch": 0.36,
"learning_rate": 0.00032075225677031095,
"loss": 2.6754,
"theoretical_loss": 3.5862337989433906,
"tokens_seen": 1204486144
},
{
"epoch": 0.37,
"learning_rate": 0.0003207422266800401,
"loss": 3.0604,
"theoretical_loss": 3.5862158631237833,
"tokens_seen": 1204551680
},
{
"epoch": 0.37,
"learning_rate": 0.0003207321965897693,
"loss": 2.9499,
"theoretical_loss": 3.5861979285531986,
"tokens_seen": 1204617216
},
{
"epoch": 0.37,
"learning_rate": 0.0003207221664994985,
"loss": 2.6037,
"theoretical_loss": 3.5861799952314826,
"tokens_seen": 1204682752
},
{
"epoch": 0.37,
"learning_rate": 0.0003207121364092277,
"loss": 2.8882,
"theoretical_loss": 3.58616206315848,
"tokens_seen": 1204748288
},
{
"epoch": 0.37,
"learning_rate": 0.00032070210631895686,
"loss": 2.7013,
"theoretical_loss": 3.586144132334036,
"tokens_seen": 1204813824
},
{
"epoch": 0.37,
"learning_rate": 0.0003206920762286861,
"loss": 2.7021,
"theoretical_loss": 3.5861262027579963,
"tokens_seen": 1204879360
},
{
"epoch": 0.37,
"learning_rate": 0.0003206820461384152,
"loss": 2.754,
"theoretical_loss": 3.5861082744302055,
"tokens_seen": 1204944896
},
{
"epoch": 0.37,
"learning_rate": 0.00032067201604814446,
"loss": 2.9552,
"theoretical_loss": 3.586090347350509,
"tokens_seen": 1205010432
},
{
"epoch": 0.37,
"learning_rate": 0.0003206619859578736,
"loss": 3.0368,
"theoretical_loss": 3.586072421518752,
"tokens_seen": 1205075968
},
{
"epoch": 0.37,
"learning_rate": 0.0003206519558676028,
"loss": 2.8423,
"theoretical_loss": 3.5860544969347794,
"tokens_seen": 1205141504
},
{
"epoch": 0.37,
"learning_rate": 0.000320641925777332,
"loss": 2.6757,
"theoretical_loss": 3.586036573598437,
"tokens_seen": 1205207040
},
{
"epoch": 0.37,
"learning_rate": 0.0003206318956870612,
"loss": 2.7942,
"theoretical_loss": 3.58601865150957,
"tokens_seen": 1205272576
},
{
"epoch": 0.37,
"learning_rate": 0.00032062186559679036,
"loss": 2.7478,
"theoretical_loss": 3.586000730668024,
"tokens_seen": 1205338112
},
{
"epoch": 0.37,
"learning_rate": 0.00032061183550651954,
"loss": 2.7694,
"theoretical_loss": 3.585982811073644,
"tokens_seen": 1205403648
},
{
"epoch": 0.37,
"learning_rate": 0.0003206018054162487,
"loss": 2.9152,
"theoretical_loss": 3.5859648927262753,
"tokens_seen": 1205469184
},
{
"epoch": 0.37,
"learning_rate": 0.00032059177532597796,
"loss": 2.6544,
"theoretical_loss": 3.5859469756257636,
"tokens_seen": 1205534720
},
{
"epoch": 0.37,
"learning_rate": 0.0003205817452357071,
"loss": 2.6135,
"theoretical_loss": 3.5859290597719546,
"tokens_seen": 1205600256
},
{
"epoch": 0.37,
"learning_rate": 0.0003205717151454363,
"loss": 2.8581,
"theoretical_loss": 3.585911145164693,
"tokens_seen": 1205665792
},
{
"epoch": 0.37,
"learning_rate": 0.0003205616850551655,
"loss": 2.6743,
"theoretical_loss": 3.5858932318038255,
"tokens_seen": 1205731328
},
{
"epoch": 0.37,
"learning_rate": 0.0003205516549648947,
"loss": 2.7588,
"theoretical_loss": 3.585875319689196,
"tokens_seen": 1205796864
},
{
"epoch": 0.37,
"objective/train/docs_used": 684684,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1852867603302,
"objective/train/theoretical_loss": 3.585857408820652,
"objective/train/tokens_used": 1226322400,
"theoretical_loss": 3.585857408820652,
"tokens_seen": 1205862400
},
{
"epoch": 0.37,
"learning_rate": 0.0003205416248746239,
"loss": 2.8134,
"theoretical_loss": 3.585857408820652,
"tokens_seen": 1205862400
},
{
"epoch": 0.37,
"learning_rate": 0.00032053159478435305,
"loss": 2.7803,
"theoretical_loss": 3.5858394991980376,
"tokens_seen": 1205927936
},
{
"epoch": 0.37,
"learning_rate": 0.0003205215646940823,
"loss": 2.7456,
"theoretical_loss": 3.585821590821199,
"tokens_seen": 1205993472
},
{
"epoch": 0.37,
"learning_rate": 0.00032051153460381146,
"loss": 2.7311,
"theoretical_loss": 3.5858036836899814,
"tokens_seen": 1206059008
},
{
"epoch": 0.37,
"learning_rate": 0.00032050150451354064,
"loss": 2.8993,
"theoretical_loss": 3.5857857778042312,
"tokens_seen": 1206124544
},
{
"epoch": 0.37,
"learning_rate": 0.0003204914744232698,
"loss": 2.8772,
"theoretical_loss": 3.5857678731637943,
"tokens_seen": 1206190080
},
{
"epoch": 0.37,
"learning_rate": 0.000320481444332999,
"loss": 2.9819,
"theoretical_loss": 3.585749969768515,
"tokens_seen": 1206255616
},
{
"epoch": 0.37,
"learning_rate": 0.0003204714142427282,
"loss": 2.7326,
"theoretical_loss": 3.585732067618241,
"tokens_seen": 1206321152
},
{
"epoch": 0.37,
"learning_rate": 0.0003204613841524574,
"loss": 2.8371,
"theoretical_loss": 3.5857141667128163,
"tokens_seen": 1206386688
},
{
"epoch": 0.37,
"learning_rate": 0.00032045135406218655,
"loss": 2.739,
"theoretical_loss": 3.5856962670520884,
"tokens_seen": 1206452224
},
{
"epoch": 0.37,
"learning_rate": 0.0003204413239719158,
"loss": 2.8868,
"theoretical_loss": 3.5856783686359015,
"tokens_seen": 1206517760
},
{
"epoch": 0.37,
"learning_rate": 0.0003204312938816449,
"loss": 2.8174,
"theoretical_loss": 3.5856604714641023,
"tokens_seen": 1206583296
},
{
"epoch": 0.37,
"learning_rate": 0.00032042126379137415,
"loss": 2.7849,
"theoretical_loss": 3.5856425755365366,
"tokens_seen": 1206648832
},
{
"epoch": 0.37,
"learning_rate": 0.00032041123370110333,
"loss": 2.7228,
"theoretical_loss": 3.585624680853051,
"tokens_seen": 1206714368
},
{
"epoch": 0.37,
"learning_rate": 0.0003204012036108325,
"loss": 2.8239,
"theoretical_loss": 3.5856067874134903,
"tokens_seen": 1206779904
},
{
"epoch": 0.37,
"learning_rate": 0.0003203911735205617,
"loss": 2.8136,
"theoretical_loss": 3.5855888952177013,
"tokens_seen": 1206845440
},
{
"epoch": 0.37,
"learning_rate": 0.0003203811434302909,
"loss": 2.8028,
"theoretical_loss": 3.585571004265529,
"tokens_seen": 1206910976
},
{
"epoch": 0.37,
"learning_rate": 0.00032037111334002005,
"loss": 2.7477,
"theoretical_loss": 3.585553114556821,
"tokens_seen": 1206976512
},
{
"epoch": 0.37,
"learning_rate": 0.0003203610832497493,
"loss": 2.6941,
"theoretical_loss": 3.585535226091422,
"tokens_seen": 1207042048
},
{
"epoch": 0.37,
"learning_rate": 0.0003203510531594784,
"loss": 3.0069,
"theoretical_loss": 3.585517338869179,
"tokens_seen": 1207107584
},
{
"epoch": 0.37,
"learning_rate": 0.00032034102306920765,
"loss": 2.6754,
"theoretical_loss": 3.5854994528899375,
"tokens_seen": 1207173120
},
{
"epoch": 0.37,
"learning_rate": 0.00032033099297893683,
"loss": 2.6892,
"theoretical_loss": 3.5854815681535435,
"tokens_seen": 1207238656
},
{
"epoch": 0.37,
"learning_rate": 0.000320320962888666,
"loss": 2.9404,
"theoretical_loss": 3.585463684659844,
"tokens_seen": 1207304192
},
{
"epoch": 0.37,
"learning_rate": 0.0003203109327983952,
"loss": 2.7756,
"theoretical_loss": 3.5854458024086853,
"tokens_seen": 1207369728
},
{
"epoch": 0.37,
"learning_rate": 0.0003203009027081244,
"loss": 2.8733,
"theoretical_loss": 3.585427921399912,
"tokens_seen": 1207435264
},
{
"epoch": 0.37,
"objective/train/docs_used": 686035,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.943772315979004,
"objective/train/theoretical_loss": 3.585410041633372,
"objective/train/tokens_used": 1227960800,
"theoretical_loss": 3.585410041633372,
"tokens_seen": 1207500800
},
{
"epoch": 0.37,
"learning_rate": 0.00032029087261785356,
"loss": 2.8615,
"theoretical_loss": 3.585410041633372,
"tokens_seen": 1207500800
},
{
"epoch": 0.37,
"learning_rate": 0.0003202808425275828,
"loss": 2.7489,
"theoretical_loss": 3.585392163108911,
"tokens_seen": 1207566336
},
{
"epoch": 0.37,
"learning_rate": 0.0003202708124373119,
"loss": 2.6875,
"theoretical_loss": 3.5853742858263753,
"tokens_seen": 1207631872
},
{
"epoch": 0.37,
"learning_rate": 0.00032026078234704115,
"loss": 3.0138,
"theoretical_loss": 3.5853564097856117,
"tokens_seen": 1207697408
},
{
"epoch": 0.37,
"learning_rate": 0.0003202507522567703,
"loss": 2.7469,
"theoretical_loss": 3.585338534986466,
"tokens_seen": 1207762944
},
{
"epoch": 0.37,
"learning_rate": 0.0003202407221664995,
"loss": 2.7398,
"theoretical_loss": 3.585320661428784,
"tokens_seen": 1207828480
},
{
"epoch": 0.37,
"learning_rate": 0.0003202306920762287,
"loss": 2.943,
"theoretical_loss": 3.585302789112413,
"tokens_seen": 1207894016
},
{
"epoch": 0.37,
"learning_rate": 0.0003202206619859579,
"loss": 2.9809,
"theoretical_loss": 3.5852849180372006,
"tokens_seen": 1207959552
},
{
"epoch": 0.37,
"learning_rate": 0.00032021063189568706,
"loss": 2.885,
"theoretical_loss": 3.585267048202991,
"tokens_seen": 1208025088
},
{
"epoch": 0.37,
"learning_rate": 0.0003202006018054163,
"loss": 2.8194,
"theoretical_loss": 3.5852491796096317,
"tokens_seen": 1208090624
},
{
"epoch": 0.37,
"learning_rate": 0.0003201905717151454,
"loss": 2.6666,
"theoretical_loss": 3.5852313122569694,
"tokens_seen": 1208156160
},
{
"epoch": 0.37,
"learning_rate": 0.00032018054162487466,
"loss": 2.9067,
"theoretical_loss": 3.5852134461448504,
"tokens_seen": 1208221696
},
{
"epoch": 0.37,
"learning_rate": 0.0003201705115346038,
"loss": 2.7803,
"theoretical_loss": 3.5851955812731218,
"tokens_seen": 1208287232
},
{
"epoch": 0.37,
"learning_rate": 0.000320160481444333,
"loss": 2.6831,
"theoretical_loss": 3.585177717641629,
"tokens_seen": 1208352768
},
{
"epoch": 0.37,
"learning_rate": 0.0003201504513540622,
"loss": 2.7725,
"theoretical_loss": 3.58515985525022,
"tokens_seen": 1208418304
},
{
"epoch": 0.37,
"learning_rate": 0.0003201404212637914,
"loss": 2.717,
"theoretical_loss": 3.585141994098741,
"tokens_seen": 1208483840
},
{
"epoch": 0.37,
"learning_rate": 0.00032013039117352056,
"loss": 2.5351,
"theoretical_loss": 3.585124134187038,
"tokens_seen": 1208549376
},
{
"epoch": 0.37,
"learning_rate": 0.00032012036108324974,
"loss": 2.8223,
"theoretical_loss": 3.585106275514959,
"tokens_seen": 1208614912
},
{
"epoch": 0.37,
"learning_rate": 0.0003201103309929789,
"loss": 2.8615,
"theoretical_loss": 3.58508841808235,
"tokens_seen": 1208680448
},
{
"epoch": 0.37,
"learning_rate": 0.00032010030090270816,
"loss": 2.9506,
"theoretical_loss": 3.5850705618890575,
"tokens_seen": 1208745984
},
{
"epoch": 0.37,
"learning_rate": 0.0003200902708124373,
"loss": 2.764,
"theoretical_loss": 3.5850527069349285,
"tokens_seen": 1208811520
},
{
"epoch": 0.37,
"learning_rate": 0.0003200802407221665,
"loss": 2.9071,
"theoretical_loss": 3.5850348532198106,
"tokens_seen": 1208877056
},
{
"epoch": 0.37,
"learning_rate": 0.00032007021063189565,
"loss": 2.5657,
"theoretical_loss": 3.58501700074355,
"tokens_seen": 1208942592
},
{
"epoch": 0.37,
"learning_rate": 0.0003200601805416249,
"loss": 3.0396,
"theoretical_loss": 3.5849991495059927,
"tokens_seen": 1209008128
},
{
"epoch": 0.37,
"learning_rate": 0.00032005015045135407,
"loss": 2.5806,
"theoretical_loss": 3.584981299506987,
"tokens_seen": 1209073664
},
{
"epoch": 0.37,
"objective/train/docs_used": 686549,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.866199493408203,
"objective/train/theoretical_loss": 3.58496345074638,
"objective/train/tokens_used": 1229599200,
"theoretical_loss": 3.58496345074638,
"tokens_seen": 1209139200
},
{
"epoch": 0.37,
"learning_rate": 0.00032004012036108325,
"loss": 2.7346,
"theoretical_loss": 3.58496345074638,
"tokens_seen": 1209139200
},
{
"epoch": 0.37,
"learning_rate": 0.00032003009027081243,
"loss": 2.6968,
"theoretical_loss": 3.5849456032240172,
"tokens_seen": 1209204736
},
{
"epoch": 0.37,
"learning_rate": 0.00032002006018054166,
"loss": 2.8378,
"theoretical_loss": 3.584927756939747,
"tokens_seen": 1209270272
},
{
"epoch": 0.37,
"learning_rate": 0.0003200100300902708,
"loss": 2.9177,
"theoretical_loss": 3.5849099118934156,
"tokens_seen": 1209335808
},
{
"epoch": 0.37,
"learning_rate": 0.00032,
"loss": 2.9126,
"theoretical_loss": 3.58489206808487,
"tokens_seen": 1209401344
},
{
"epoch": 0.37,
"learning_rate": 0.00031998996990972915,
"loss": 2.8929,
"theoretical_loss": 3.5848742255139583,
"tokens_seen": 1209466880
},
{
"epoch": 0.37,
"learning_rate": 0.0003199799398194584,
"loss": 2.857,
"theoretical_loss": 3.5848563841805268,
"tokens_seen": 1209532416
},
{
"epoch": 0.37,
"learning_rate": 0.00031996990972918757,
"loss": 2.6283,
"theoretical_loss": 3.5848385440844224,
"tokens_seen": 1209597952
},
{
"epoch": 0.37,
"learning_rate": 0.00031995987963891675,
"loss": 2.7608,
"theoretical_loss": 3.584820705225493,
"tokens_seen": 1209663488
},
{
"epoch": 0.37,
"learning_rate": 0.00031994984954864593,
"loss": 2.7299,
"theoretical_loss": 3.584802867603585,
"tokens_seen": 1209729024
},
{
"epoch": 0.37,
"learning_rate": 0.0003199398194583751,
"loss": 2.7904,
"theoretical_loss": 3.584785031218546,
"tokens_seen": 1209794560
},
{
"epoch": 0.37,
"learning_rate": 0.0003199297893681043,
"loss": 2.7092,
"theoretical_loss": 3.584767196070223,
"tokens_seen": 1209860096
},
{
"epoch": 0.37,
"learning_rate": 0.00031991975927783353,
"loss": 2.8205,
"theoretical_loss": 3.5847493621584636,
"tokens_seen": 1209925632
},
{
"epoch": 0.37,
"learning_rate": 0.00031990972918756266,
"loss": 2.6762,
"theoretical_loss": 3.5847315294831152,
"tokens_seen": 1209991168
},
{
"epoch": 0.37,
"learning_rate": 0.0003198996990972919,
"loss": 2.8353,
"theoretical_loss": 3.584713698044025,
"tokens_seen": 1210056704
},
{
"epoch": 0.37,
"learning_rate": 0.000319889669007021,
"loss": 2.8299,
"theoretical_loss": 3.58469586784104,
"tokens_seen": 1210122240
},
{
"epoch": 0.37,
"learning_rate": 0.00031987963891675025,
"loss": 2.4616,
"theoretical_loss": 3.5846780388740074,
"tokens_seen": 1210187776
},
{
"epoch": 0.37,
"learning_rate": 0.00031986960882647943,
"loss": 2.6834,
"theoretical_loss": 3.5846602111427757,
"tokens_seen": 1210253312
},
{
"epoch": 0.37,
"learning_rate": 0.0003198595787362086,
"loss": 2.84,
"theoretical_loss": 3.5846423846471915,
"tokens_seen": 1210318848
},
{
"epoch": 0.37,
"learning_rate": 0.0003198495486459378,
"loss": 2.8026,
"theoretical_loss": 3.5846245593871027,
"tokens_seen": 1210384384
},
{
"epoch": 0.37,
"learning_rate": 0.00031983951855566703,
"loss": 2.8356,
"theoretical_loss": 3.584606735362356,
"tokens_seen": 1210449920
},
{
"epoch": 0.37,
"learning_rate": 0.00031982948846539616,
"loss": 3.0488,
"theoretical_loss": 3.5845889125727997,
"tokens_seen": 1210515456
},
{
"epoch": 0.37,
"learning_rate": 0.0003198194583751254,
"loss": 2.8902,
"theoretical_loss": 3.5845710910182804,
"tokens_seen": 1210580992
},
{
"epoch": 0.37,
"learning_rate": 0.0003198094282848546,
"loss": 2.6176,
"theoretical_loss": 3.584553270698647,
"tokens_seen": 1210646528
},
{
"epoch": 0.37,
"learning_rate": 0.00031979939819458376,
"loss": 2.9413,
"theoretical_loss": 3.584535451613746,
"tokens_seen": 1210712064
},
{
"epoch": 0.37,
"objective/train/docs_used": 687794,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6891303062438965,
"objective/train/theoretical_loss": 3.5845176337634257,
"objective/train/tokens_used": 1231237600,
"theoretical_loss": 3.5845176337634257,
"tokens_seen": 1210777600
},
{
"epoch": 0.37,
"learning_rate": 0.000319789368104313,
"loss": 2.8759,
"theoretical_loss": 3.5845176337634257,
"tokens_seen": 1210777600
},
{
"epoch": 0.37,
"learning_rate": 0.0003197793380140421,
"loss": 2.7465,
"theoretical_loss": 3.5844998171475333,
"tokens_seen": 1210843136
},
{
"epoch": 0.37,
"learning_rate": 0.00031976930792377135,
"loss": 2.8036,
"theoretical_loss": 3.5844820017659167,
"tokens_seen": 1210908672
},
{
"epoch": 0.37,
"learning_rate": 0.0003197592778335005,
"loss": 2.8724,
"theoretical_loss": 3.5844641876184236,
"tokens_seen": 1210974208
},
{
"epoch": 0.37,
"learning_rate": 0.0003197492477432297,
"loss": 2.9114,
"theoretical_loss": 3.5844463747049016,
"tokens_seen": 1211039744
},
{
"epoch": 0.37,
"learning_rate": 0.0003197392176529589,
"loss": 2.7357,
"theoretical_loss": 3.5844285630251984,
"tokens_seen": 1211105280
},
{
"epoch": 0.37,
"learning_rate": 0.0003197291875626881,
"loss": 2.5608,
"theoretical_loss": 3.5844107525791618,
"tokens_seen": 1211170816
},
{
"epoch": 0.37,
"learning_rate": 0.00031971915747241726,
"loss": 2.8363,
"theoretical_loss": 3.58439294336664,
"tokens_seen": 1211236352
},
{
"epoch": 0.37,
"learning_rate": 0.0003197091273821465,
"loss": 2.9067,
"theoretical_loss": 3.5843751353874804,
"tokens_seen": 1211301888
},
{
"epoch": 0.37,
"learning_rate": 0.0003196990972918756,
"loss": 2.6231,
"theoretical_loss": 3.584357328641531,
"tokens_seen": 1211367424
},
{
"epoch": 0.37,
"learning_rate": 0.00031968906720160486,
"loss": 2.8998,
"theoretical_loss": 3.58433952312864,
"tokens_seen": 1211432960
},
{
"epoch": 0.37,
"learning_rate": 0.000319679037111334,
"loss": 2.7272,
"theoretical_loss": 3.5843217188486545,
"tokens_seen": 1211498496
},
{
"epoch": 0.37,
"learning_rate": 0.0003196690070210632,
"loss": 2.9637,
"theoretical_loss": 3.584303915801423,
"tokens_seen": 1211564032
},
{
"epoch": 0.37,
"learning_rate": 0.0003196589769307924,
"loss": 2.8202,
"theoretical_loss": 3.584286113986794,
"tokens_seen": 1211629568
},
{
"epoch": 0.37,
"learning_rate": 0.0003196489468405216,
"loss": 2.8463,
"theoretical_loss": 3.584268313404614,
"tokens_seen": 1211695104
},
{
"epoch": 0.37,
"learning_rate": 0.00031963891675025076,
"loss": 2.7979,
"theoretical_loss": 3.584250514054732,
"tokens_seen": 1211760640
},
{
"epoch": 0.37,
"learning_rate": 0.00031962888665997994,
"loss": 2.6365,
"theoretical_loss": 3.584232715936997,
"tokens_seen": 1211826176
},
{
"epoch": 0.37,
"learning_rate": 0.0003196188565697091,
"loss": 2.8412,
"theoretical_loss": 3.584214919051255,
"tokens_seen": 1211891712
},
{
"epoch": 0.37,
"learning_rate": 0.00031960882647943836,
"loss": 2.7844,
"theoretical_loss": 3.584197123397356,
"tokens_seen": 1211957248
},
{
"epoch": 0.37,
"learning_rate": 0.0003195987963891675,
"loss": 2.6693,
"theoretical_loss": 3.5841793289751465,
"tokens_seen": 1212022784
},
{
"epoch": 0.37,
"learning_rate": 0.0003195887662988967,
"loss": 2.8021,
"theoretical_loss": 3.5841615357844763,
"tokens_seen": 1212088320
},
{
"epoch": 0.37,
"learning_rate": 0.00031957873620862585,
"loss": 2.6975,
"theoretical_loss": 3.584143743825192,
"tokens_seen": 1212153856
},
{
"epoch": 0.37,
"learning_rate": 0.0003195687061183551,
"loss": 3.0082,
"theoretical_loss": 3.5841259530971428,
"tokens_seen": 1212219392
},
{
"epoch": 0.37,
"learning_rate": 0.00031955867602808427,
"loss": 2.8563,
"theoretical_loss": 3.584108163600177,
"tokens_seen": 1212284928
},
{
"epoch": 0.37,
"learning_rate": 0.00031954864593781345,
"loss": 2.8012,
"theoretical_loss": 3.584090375334142,
"tokens_seen": 1212350464
},
{
"epoch": 0.37,
"objective/train/docs_used": 688453,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.202460765838623,
"objective/train/theoretical_loss": 3.5840725882988873,
"objective/train/tokens_used": 1232876000,
"theoretical_loss": 3.5840725882988873,
"tokens_seen": 1212416000
},
{
"epoch": 0.37,
"learning_rate": 0.00031953861584754263,
"loss": 2.9825,
"theoretical_loss": 3.5840725882988873,
"tokens_seen": 1212416000
},
{
"epoch": 0.37,
"learning_rate": 0.00031952858575727186,
"loss": 2.8735,
"theoretical_loss": 3.5840548024942604,
"tokens_seen": 1212481536
},
{
"epoch": 0.37,
"learning_rate": 0.000319518555667001,
"loss": 2.7155,
"theoretical_loss": 3.5840370179201093,
"tokens_seen": 1212547072
},
{
"epoch": 0.37,
"learning_rate": 0.0003195085255767302,
"loss": 2.7543,
"theoretical_loss": 3.5840192345762834,
"tokens_seen": 1212612608
},
{
"epoch": 0.37,
"learning_rate": 0.00031949849548645935,
"loss": 2.8648,
"theoretical_loss": 3.5840014524626307,
"tokens_seen": 1212678144
},
{
"epoch": 0.37,
"learning_rate": 0.0003194884653961886,
"loss": 2.8293,
"theoretical_loss": 3.5839836715789994,
"tokens_seen": 1212743680
},
{
"epoch": 0.37,
"learning_rate": 0.00031947843530591777,
"loss": 2.829,
"theoretical_loss": 3.5839658919252377,
"tokens_seen": 1212809216
},
{
"epoch": 0.37,
"learning_rate": 0.00031946840521564695,
"loss": 2.6923,
"theoretical_loss": 3.5839481135011946,
"tokens_seen": 1212874752
},
{
"epoch": 0.37,
"learning_rate": 0.00031945837512537613,
"loss": 2.4879,
"theoretical_loss": 3.583930336306719,
"tokens_seen": 1212940288
},
{
"epoch": 0.37,
"learning_rate": 0.0003194483450351053,
"loss": 2.8539,
"theoretical_loss": 3.583912560341658,
"tokens_seen": 1213005824
},
{
"epoch": 0.37,
"learning_rate": 0.0003194383149448345,
"loss": 2.6715,
"theoretical_loss": 3.583894785605862,
"tokens_seen": 1213071360
},
{
"epoch": 0.37,
"learning_rate": 0.00031942828485456373,
"loss": 2.7903,
"theoretical_loss": 3.583877012099178,
"tokens_seen": 1213136896
},
{
"epoch": 0.37,
"learning_rate": 0.00031941825476429286,
"loss": 2.624,
"theoretical_loss": 3.5838592398214555,
"tokens_seen": 1213202432
},
{
"epoch": 0.37,
"learning_rate": 0.0003194082246740221,
"loss": 2.7636,
"theoretical_loss": 3.583841468772543,
"tokens_seen": 1213267968
},
{
"epoch": 0.37,
"learning_rate": 0.0003193981945837512,
"loss": 2.7333,
"theoretical_loss": 3.5838236989522887,
"tokens_seen": 1213333504
},
{
"epoch": 0.37,
"learning_rate": 0.00031938816449348045,
"loss": 2.8011,
"theoretical_loss": 3.5838059303605423,
"tokens_seen": 1213399040
},
{
"epoch": 0.37,
"learning_rate": 0.00031937813440320964,
"loss": 2.8364,
"theoretical_loss": 3.583788162997151,
"tokens_seen": 1213464576
},
{
"epoch": 0.37,
"learning_rate": 0.0003193681043129388,
"loss": 2.925,
"theoretical_loss": 3.5837703968619654,
"tokens_seen": 1213530112
},
{
"epoch": 0.37,
"learning_rate": 0.000319358074222668,
"loss": 2.6972,
"theoretical_loss": 3.583752631954833,
"tokens_seen": 1213595648
},
{
"epoch": 0.37,
"learning_rate": 0.00031934804413239723,
"loss": 2.8524,
"theoretical_loss": 3.5837348682756027,
"tokens_seen": 1213661184
},
{
"epoch": 0.37,
"learning_rate": 0.00031933801404212636,
"loss": 2.6672,
"theoretical_loss": 3.5837171058241237,
"tokens_seen": 1213726720
},
{
"epoch": 0.37,
"learning_rate": 0.0003193279839518556,
"loss": 2.8383,
"theoretical_loss": 3.5836993446002445,
"tokens_seen": 1213792256
},
{
"epoch": 0.37,
"learning_rate": 0.0003193179538615847,
"loss": 2.5754,
"theoretical_loss": 3.5836815846038146,
"tokens_seen": 1213857792
},
{
"epoch": 0.37,
"learning_rate": 0.00031930792377131396,
"loss": 2.6234,
"theoretical_loss": 3.5836638258346825,
"tokens_seen": 1213923328
},
{
"epoch": 0.37,
"learning_rate": 0.00031929789368104314,
"loss": 2.7275,
"theoretical_loss": 3.5836460682926967,
"tokens_seen": 1213988864
},
{
"epoch": 0.37,
"objective/train/docs_used": 689723,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7307770252227783,
"objective/train/theoretical_loss": 3.583628311977707,
"objective/train/tokens_used": 1234514400,
"theoretical_loss": 3.583628311977707,
"tokens_seen": 1214054400
},
{
"epoch": 0.37,
"learning_rate": 0.0003192878635907723,
"loss": 2.6881,
"theoretical_loss": 3.583628311977707,
"tokens_seen": 1214054400
},
{
"epoch": 0.37,
"learning_rate": 0.0003192778335005015,
"loss": 2.8562,
"theoretical_loss": 3.583610556889562,
"tokens_seen": 1214119936
},
{
"epoch": 0.37,
"learning_rate": 0.0003192678034102307,
"loss": 2.9461,
"theoretical_loss": 3.5835928030281106,
"tokens_seen": 1214185472
},
{
"epoch": 0.37,
"learning_rate": 0.00031925777331995986,
"loss": 2.7858,
"theoretical_loss": 3.583575050393202,
"tokens_seen": 1214251008
},
{
"epoch": 0.37,
"learning_rate": 0.0003192477432296891,
"loss": 2.7801,
"theoretical_loss": 3.583557298984685,
"tokens_seen": 1214316544
},
{
"epoch": 0.37,
"learning_rate": 0.0003192377131394182,
"loss": 2.5697,
"theoretical_loss": 3.583539548802409,
"tokens_seen": 1214382080
},
{
"epoch": 0.37,
"learning_rate": 0.00031922768304914746,
"loss": 2.7894,
"theoretical_loss": 3.583521799846223,
"tokens_seen": 1214447616
},
{
"epoch": 0.37,
"learning_rate": 0.0003192176529588766,
"loss": 2.6564,
"theoretical_loss": 3.5835040521159764,
"tokens_seen": 1214513152
},
{
"epoch": 0.37,
"learning_rate": 0.0003192076228686058,
"loss": 2.8306,
"theoretical_loss": 3.583486305611518,
"tokens_seen": 1214578688
},
{
"epoch": 0.37,
"learning_rate": 0.000319197592778335,
"loss": 2.8219,
"theoretical_loss": 3.5834685603326975,
"tokens_seen": 1214644224
},
{
"epoch": 0.37,
"learning_rate": 0.0003191875626880642,
"loss": 2.577,
"theoretical_loss": 3.5834508162793637,
"tokens_seen": 1214709760
},
{
"epoch": 0.37,
"learning_rate": 0.00031917753259779337,
"loss": 2.7776,
"theoretical_loss": 3.583433073451366,
"tokens_seen": 1214775296
},
{
"epoch": 0.37,
"learning_rate": 0.0003191675025075226,
"loss": 2.667,
"theoretical_loss": 3.5834153318485535,
"tokens_seen": 1214840832
},
{
"epoch": 0.37,
"learning_rate": 0.00031915747241725173,
"loss": 2.7734,
"theoretical_loss": 3.5833975914707756,
"tokens_seen": 1214906368
},
{
"epoch": 0.37,
"learning_rate": 0.00031914744232698096,
"loss": 2.8999,
"theoretical_loss": 3.583379852317882,
"tokens_seen": 1214971904
},
{
"epoch": 0.37,
"learning_rate": 0.0003191374122367101,
"loss": 2.7922,
"theoretical_loss": 3.5833621143897214,
"tokens_seen": 1215037440
},
{
"epoch": 0.37,
"learning_rate": 0.0003191273821464393,
"loss": 2.8252,
"theoretical_loss": 3.583344377686144,
"tokens_seen": 1215102976
},
{
"epoch": 0.37,
"learning_rate": 0.0003191173520561685,
"loss": 2.5096,
"theoretical_loss": 3.5833266422069983,
"tokens_seen": 1215168512
},
{
"epoch": 0.37,
"learning_rate": 0.0003191073219658977,
"loss": 2.6976,
"theoretical_loss": 3.5833089079521345,
"tokens_seen": 1215234048
},
{
"epoch": 0.37,
"learning_rate": 0.00031909729187562687,
"loss": 2.7347,
"theoretical_loss": 3.583291174921402,
"tokens_seen": 1215299584
},
{
"epoch": 0.37,
"learning_rate": 0.00031908726178535605,
"loss": 2.4873,
"theoretical_loss": 3.5832734431146496,
"tokens_seen": 1215365120
},
{
"epoch": 0.37,
"learning_rate": 0.00031907723169508523,
"loss": 2.7415,
"theoretical_loss": 3.5832557125317273,
"tokens_seen": 1215430656
},
{
"epoch": 0.37,
"learning_rate": 0.00031906720160481447,
"loss": 2.7511,
"theoretical_loss": 3.5832379831724848,
"tokens_seen": 1215496192
},
{
"epoch": 0.37,
"learning_rate": 0.00031905717151454365,
"loss": 2.4586,
"theoretical_loss": 3.583220255036772,
"tokens_seen": 1215561728
},
{
"epoch": 0.37,
"learning_rate": 0.00031904714142427283,
"loss": 2.8427,
"theoretical_loss": 3.5832025281244375,
"tokens_seen": 1215627264
},
{
"epoch": 0.37,
"objective/train/docs_used": 690230,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9391164779663086,
"objective/train/theoretical_loss": 3.5831848024353317,
"objective/train/tokens_used": 1236152800,
"theoretical_loss": 3.5831848024353317,
"tokens_seen": 1215692800
},
{
"epoch": 0.37,
"learning_rate": 0.00031903711133400206,
"loss": 2.668,
"theoretical_loss": 3.5831848024353317,
"tokens_seen": 1215692800
},
{
"epoch": 0.37,
"learning_rate": 0.0003190270812437312,
"loss": 2.7712,
"theoretical_loss": 3.5831670779693034,
"tokens_seen": 1215758336
},
{
"epoch": 0.37,
"learning_rate": 0.0003190170511534604,
"loss": 2.8082,
"theoretical_loss": 3.5831493547262037,
"tokens_seen": 1215823872
},
{
"epoch": 0.37,
"learning_rate": 0.00031900702106318955,
"loss": 2.7769,
"theoretical_loss": 3.5831316327058813,
"tokens_seen": 1215889408
},
{
"epoch": 0.37,
"learning_rate": 0.0003189969909729188,
"loss": 2.6279,
"theoretical_loss": 3.5831139119081863,
"tokens_seen": 1215954944
},
{
"epoch": 0.37,
"learning_rate": 0.00031898696088264797,
"loss": 2.7789,
"theoretical_loss": 3.583096192332968,
"tokens_seen": 1216020480
},
{
"epoch": 0.37,
"learning_rate": 0.00031897693079237715,
"loss": 2.9196,
"theoretical_loss": 3.583078473980077,
"tokens_seen": 1216086016
},
{
"epoch": 0.37,
"learning_rate": 0.00031896690070210633,
"loss": 2.7921,
"theoretical_loss": 3.5830607568493624,
"tokens_seen": 1216151552
},
{
"epoch": 0.37,
"learning_rate": 0.0003189568706118355,
"loss": 2.6885,
"theoretical_loss": 3.583043040940674,
"tokens_seen": 1216217088
},
{
"epoch": 0.37,
"learning_rate": 0.0003189468405215647,
"loss": 2.9383,
"theoretical_loss": 3.5830253262538623,
"tokens_seen": 1216282624
},
{
"epoch": 0.37,
"learning_rate": 0.00031893681043129393,
"loss": 2.6309,
"theoretical_loss": 3.5830076127887764,
"tokens_seen": 1216348160
},
{
"epoch": 0.37,
"learning_rate": 0.00031892678034102306,
"loss": 2.692,
"theoretical_loss": 3.582989900545267,
"tokens_seen": 1216413696
},
{
"epoch": 0.37,
"learning_rate": 0.0003189167502507523,
"loss": 2.8183,
"theoretical_loss": 3.5829721895231836,
"tokens_seen": 1216479232
},
{
"epoch": 0.37,
"learning_rate": 0.0003189067201604814,
"loss": 2.7339,
"theoretical_loss": 3.5829544797223765,
"tokens_seen": 1216544768
},
{
"epoch": 0.37,
"learning_rate": 0.00031889669007021065,
"loss": 2.849,
"theoretical_loss": 3.582936771142695,
"tokens_seen": 1216610304
},
{
"epoch": 0.37,
"learning_rate": 0.00031888665997993984,
"loss": 2.7686,
"theoretical_loss": 3.58291906378399,
"tokens_seen": 1216675840
},
{
"epoch": 0.37,
"learning_rate": 0.000318876629889669,
"loss": 2.7499,
"theoretical_loss": 3.582901357646111,
"tokens_seen": 1216741376
},
{
"epoch": 0.37,
"learning_rate": 0.0003188665997993982,
"loss": 2.9151,
"theoretical_loss": 3.582883652728908,
"tokens_seen": 1216806912
},
{
"epoch": 0.37,
"learning_rate": 0.00031885656970912743,
"loss": 2.6998,
"theoretical_loss": 3.5828659490322323,
"tokens_seen": 1216872448
},
{
"epoch": 0.37,
"learning_rate": 0.00031884653961885656,
"loss": 2.5388,
"theoretical_loss": 3.5828482465559324,
"tokens_seen": 1216937984
},
{
"epoch": 0.37,
"learning_rate": 0.0003188365095285858,
"loss": 2.8868,
"theoretical_loss": 3.582830545299859,
"tokens_seen": 1217003520
},
{
"epoch": 0.37,
"learning_rate": 0.0003188264794383149,
"loss": 2.7887,
"theoretical_loss": 3.582812845263863,
"tokens_seen": 1217069056
},
{
"epoch": 0.37,
"learning_rate": 0.00031881644934804416,
"loss": 2.6942,
"theoretical_loss": 3.5827951464477934,
"tokens_seen": 1217134592
},
{
"epoch": 0.37,
"learning_rate": 0.00031880641925777334,
"loss": 2.777,
"theoretical_loss": 3.582777448851502,
"tokens_seen": 1217200128
},
{
"epoch": 0.37,
"learning_rate": 0.0003187963891675025,
"loss": 2.5946,
"theoretical_loss": 3.5827597524748374,
"tokens_seen": 1217265664
},
{
"epoch": 0.37,
"objective/train/docs_used": 691368,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.031107187271118,
"objective/train/theoretical_loss": 3.5827420573176507,
"objective/train/tokens_used": 1237791200,
"theoretical_loss": 3.5827420573176507,
"tokens_seen": 1217331200
},
{
"epoch": 0.37,
"learning_rate": 0.0003187863590772317,
"loss": 2.9001,
"theoretical_loss": 3.5827420573176507,
"tokens_seen": 1217331200
},
{
"epoch": 0.37,
"learning_rate": 0.0003187763289869609,
"loss": 2.8563,
"theoretical_loss": 3.582724363379792,
"tokens_seen": 1217396736
},
{
"epoch": 0.37,
"learning_rate": 0.00031876629889669006,
"loss": 2.7018,
"theoretical_loss": 3.5827066706611124,
"tokens_seen": 1217462272
},
{
"epoch": 0.37,
"learning_rate": 0.0003187562688064193,
"loss": 2.5914,
"theoretical_loss": 3.5826889791614613,
"tokens_seen": 1217527808
},
{
"epoch": 0.37,
"learning_rate": 0.0003187462387161484,
"loss": 2.7329,
"theoretical_loss": 3.5826712888806895,
"tokens_seen": 1217593344
},
{
"epoch": 0.37,
"learning_rate": 0.00031873620862587766,
"loss": 2.5309,
"theoretical_loss": 3.5826535998186477,
"tokens_seen": 1217658880
},
{
"epoch": 0.37,
"learning_rate": 0.0003187261785356068,
"loss": 2.7608,
"theoretical_loss": 3.582635911975186,
"tokens_seen": 1217724416
},
{
"epoch": 0.37,
"learning_rate": 0.000318716148445336,
"loss": 2.5912,
"theoretical_loss": 3.5826182253501546,
"tokens_seen": 1217789952
},
{
"epoch": 0.37,
"learning_rate": 0.0003187061183550652,
"loss": 2.6982,
"theoretical_loss": 3.582600539943405,
"tokens_seen": 1217855488
},
{
"epoch": 0.37,
"learning_rate": 0.0003186960882647944,
"loss": 2.6921,
"theoretical_loss": 3.5825828557547865,
"tokens_seen": 1217921024
},
{
"epoch": 0.37,
"learning_rate": 0.00031868605817452357,
"loss": 2.7541,
"theoretical_loss": 3.5825651727841503,
"tokens_seen": 1217986560
},
{
"epoch": 0.37,
"learning_rate": 0.0003186760280842528,
"loss": 2.8919,
"theoretical_loss": 3.582547491031347,
"tokens_seen": 1218052096
},
{
"epoch": 0.37,
"learning_rate": 0.00031866599799398193,
"loss": 2.8268,
"theoretical_loss": 3.582529810496227,
"tokens_seen": 1218117632
},
{
"epoch": 0.37,
"learning_rate": 0.00031865596790371116,
"loss": 2.934,
"theoretical_loss": 3.5825121311786416,
"tokens_seen": 1218183168
},
{
"epoch": 0.37,
"learning_rate": 0.0003186459378134403,
"loss": 2.7611,
"theoretical_loss": 3.58249445307844,
"tokens_seen": 1218248704
},
{
"epoch": 0.37,
"learning_rate": 0.0003186359077231695,
"loss": 2.6069,
"theoretical_loss": 3.5824767761954748,
"tokens_seen": 1218314240
},
{
"epoch": 0.37,
"learning_rate": 0.0003186258776328987,
"loss": 2.556,
"theoretical_loss": 3.5824591005295954,
"tokens_seen": 1218379776
},
{
"epoch": 0.37,
"learning_rate": 0.0003186158475426279,
"loss": 2.7165,
"theoretical_loss": 3.5824414260806527,
"tokens_seen": 1218445312
},
{
"epoch": 0.37,
"learning_rate": 0.00031860581745235707,
"loss": 2.9247,
"theoretical_loss": 3.5824237528484977,
"tokens_seen": 1218510848
},
{
"epoch": 0.37,
"learning_rate": 0.00031859578736208625,
"loss": 2.7907,
"theoretical_loss": 3.582406080832981,
"tokens_seen": 1218576384
},
{
"epoch": 0.37,
"learning_rate": 0.00031858575727181543,
"loss": 2.8492,
"theoretical_loss": 3.5823884100339543,
"tokens_seen": 1218641920
},
{
"epoch": 0.37,
"learning_rate": 0.00031857572718154467,
"loss": 2.7635,
"theoretical_loss": 3.582370740451267,
"tokens_seen": 1218707456
},
{
"epoch": 0.37,
"learning_rate": 0.0003185656970912738,
"loss": 2.766,
"theoretical_loss": 3.582353072084771,
"tokens_seen": 1218772992
},
{
"epoch": 0.37,
"learning_rate": 0.00031855566700100303,
"loss": 2.5094,
"theoretical_loss": 3.582335404934317,
"tokens_seen": 1218838528
},
{
"epoch": 0.37,
"learning_rate": 0.0003185456369107322,
"loss": 2.5667,
"theoretical_loss": 3.5823177389997554,
"tokens_seen": 1218904064
},
{
"epoch": 0.37,
"objective/train/docs_used": 692060,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.841186285018921,
"objective/train/theoretical_loss": 3.5823000742809374,
"objective/train/tokens_used": 1239429600,
"theoretical_loss": 3.5823000742809374,
"tokens_seen": 1218969600
},
{
"epoch": 0.37,
"learning_rate": 0.0003185356068204614,
"loss": 2.863,
"theoretical_loss": 3.5823000742809374,
"tokens_seen": 1218969600
},
{
"epoch": 0.37,
"learning_rate": 0.00031852557673019057,
"loss": 2.8497,
"theoretical_loss": 3.5822824107777143,
"tokens_seen": 1219035136
},
{
"epoch": 0.37,
"learning_rate": 0.00031851554663991975,
"loss": 2.7888,
"theoretical_loss": 3.582264748489937,
"tokens_seen": 1219100672
},
{
"epoch": 0.37,
"learning_rate": 0.00031850551654964893,
"loss": 2.6429,
"theoretical_loss": 3.582247087417457,
"tokens_seen": 1219166208
},
{
"epoch": 0.37,
"learning_rate": 0.00031849548645937817,
"loss": 2.606,
"theoretical_loss": 3.582229427560124,
"tokens_seen": 1219231744
},
{
"epoch": 0.37,
"learning_rate": 0.0003184854563691073,
"loss": 2.6575,
"theoretical_loss": 3.5822117689177904,
"tokens_seen": 1219297280
},
{
"epoch": 0.37,
"learning_rate": 0.00031847542627883653,
"loss": 2.5711,
"theoretical_loss": 3.582194111490307,
"tokens_seen": 1219362816
},
{
"epoch": 0.37,
"learning_rate": 0.00031846539618856566,
"loss": 2.7252,
"theoretical_loss": 3.582176455277524,
"tokens_seen": 1219428352
},
{
"epoch": 0.37,
"learning_rate": 0.0003184553660982949,
"loss": 2.9585,
"theoretical_loss": 3.582158800279294,
"tokens_seen": 1219493888
},
{
"epoch": 0.37,
"learning_rate": 0.0003184453360080241,
"loss": 2.6115,
"theoretical_loss": 3.5821411464954673,
"tokens_seen": 1219559424
},
{
"epoch": 0.37,
"learning_rate": 0.00031843530591775326,
"loss": 2.7457,
"theoretical_loss": 3.5821234939258955,
"tokens_seen": 1219624960
},
{
"epoch": 0.37,
"learning_rate": 0.00031842527582748244,
"loss": 2.8974,
"theoretical_loss": 3.5821058425704293,
"tokens_seen": 1219690496
},
{
"epoch": 0.37,
"learning_rate": 0.0003184152457372116,
"loss": 2.8218,
"theoretical_loss": 3.5820881924289205,
"tokens_seen": 1219756032
},
{
"epoch": 0.37,
"learning_rate": 0.0003184052156469408,
"loss": 2.6197,
"theoretical_loss": 3.5820705435012203,
"tokens_seen": 1219821568
},
{
"epoch": 0.37,
"learning_rate": 0.00031839518555667004,
"loss": 2.7979,
"theoretical_loss": 3.58205289578718,
"tokens_seen": 1219887104
},
{
"epoch": 0.37,
"learning_rate": 0.00031838515546639916,
"loss": 2.6614,
"theoretical_loss": 3.5820352492866507,
"tokens_seen": 1219952640
},
{
"epoch": 0.37,
"learning_rate": 0.0003183751253761284,
"loss": 2.8157,
"theoretical_loss": 3.5820176039994838,
"tokens_seen": 1220018176
},
{
"epoch": 0.37,
"learning_rate": 0.0003183650952858576,
"loss": 2.7071,
"theoretical_loss": 3.5819999599255317,
"tokens_seen": 1220083712
},
{
"epoch": 0.37,
"learning_rate": 0.00031835506519558676,
"loss": 2.8135,
"theoretical_loss": 3.581982317064644,
"tokens_seen": 1220149248
},
{
"epoch": 0.37,
"learning_rate": 0.00031834503510531594,
"loss": 2.7885,
"theoretical_loss": 3.581964675416674,
"tokens_seen": 1220214784
},
{
"epoch": 0.37,
"learning_rate": 0.0003183350050150451,
"loss": 2.5861,
"theoretical_loss": 3.5819470349814715,
"tokens_seen": 1220280320
},
{
"epoch": 0.37,
"learning_rate": 0.0003183249749247743,
"loss": 2.5333,
"theoretical_loss": 3.5819293957588894,
"tokens_seen": 1220345856
},
{
"epoch": 0.37,
"learning_rate": 0.00031831494483450354,
"loss": 2.5696,
"theoretical_loss": 3.5819117577487782,
"tokens_seen": 1220411392
},
{
"epoch": 0.37,
"learning_rate": 0.0003183049147442327,
"loss": 2.9596,
"theoretical_loss": 3.5818941209509902,
"tokens_seen": 1220476928
},
{
"epoch": 0.37,
"learning_rate": 0.0003182948846539619,
"loss": 2.7481,
"theoretical_loss": 3.5818764853653766,
"tokens_seen": 1220542464
},
{
"epoch": 0.37,
"objective/train/docs_used": 693613,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.470189332962036,
"objective/train/theoretical_loss": 3.581858850991789,
"objective/train/tokens_used": 1241068000,
"theoretical_loss": 3.581858850991789,
"tokens_seen": 1220608000
},
{
"epoch": 0.37,
"learning_rate": 0.0003182848545636911,
"loss": 2.5984,
"theoretical_loss": 3.581858850991789,
"tokens_seen": 1220608000
},
{
"epoch": 0.37,
"learning_rate": 0.0003182848545636911,
"loss": 2.8454,
"theoretical_loss": 3.581841217830079,
"tokens_seen": 1220673536
},
{
"epoch": 0.37,
"learning_rate": 0.00031827482447342026,
"loss": 2.6263,
"theoretical_loss": 3.581823585880099,
"tokens_seen": 1220739072
},
{
"epoch": 0.37,
"learning_rate": 0.0003182647943831495,
"loss": 2.8596,
"theoretical_loss": 3.5818059551416996,
"tokens_seen": 1220804608
},
{
"epoch": 0.37,
"learning_rate": 0.0003182547642928786,
"loss": 2.6334,
"theoretical_loss": 3.581788325614733,
"tokens_seen": 1220870144
},
{
"epoch": 0.37,
"learning_rate": 0.00031824473420260786,
"loss": 2.8718,
"theoretical_loss": 3.581770697299051,
"tokens_seen": 1220935680
},
{
"epoch": 0.37,
"learning_rate": 0.000318234704112337,
"loss": 2.9277,
"theoretical_loss": 3.5817530701945053,
"tokens_seen": 1221001216
},
{
"epoch": 0.37,
"learning_rate": 0.0003182246740220662,
"loss": 2.5156,
"theoretical_loss": 3.5817354443009473,
"tokens_seen": 1221066752
},
{
"epoch": 0.37,
"learning_rate": 0.0003182146439317954,
"loss": 2.8322,
"theoretical_loss": 3.5817178196182295,
"tokens_seen": 1221132288
},
{
"epoch": 0.37,
"learning_rate": 0.0003182046138415246,
"loss": 3.0646,
"theoretical_loss": 3.581700196146204,
"tokens_seen": 1221197824
},
{
"epoch": 0.37,
"learning_rate": 0.00031819458375125377,
"loss": 2.6113,
"theoretical_loss": 3.5816825738847213,
"tokens_seen": 1221263360
},
{
"epoch": 0.37,
"learning_rate": 0.000318184553660983,
"loss": 2.7832,
"theoretical_loss": 3.581664952833634,
"tokens_seen": 1221328896
},
{
"epoch": 0.37,
"learning_rate": 0.00031817452357071213,
"loss": 2.6397,
"theoretical_loss": 3.581647332992795,
"tokens_seen": 1221394432
},
{
"epoch": 0.37,
"learning_rate": 0.00031816449348044136,
"loss": 2.8125,
"theoretical_loss": 3.5816297143620544,
"tokens_seen": 1221459968
},
{
"epoch": 0.37,
"learning_rate": 0.0003181544633901705,
"loss": 2.4128,
"theoretical_loss": 3.581612096941266,
"tokens_seen": 1221525504
},
{
"epoch": 0.37,
"learning_rate": 0.0003181444332998997,
"loss": 2.8308,
"theoretical_loss": 3.5815944807302804,
"tokens_seen": 1221591040
},
{
"epoch": 0.37,
"learning_rate": 0.0003181344032096289,
"loss": 2.7176,
"theoretical_loss": 3.58157686572895,
"tokens_seen": 1221656576
},
{
"epoch": 0.37,
"learning_rate": 0.0003181243731193581,
"loss": 2.6636,
"theoretical_loss": 3.5815592519371275,
"tokens_seen": 1221722112
},
{
"epoch": 0.37,
"learning_rate": 0.00031811434302908727,
"loss": 2.5769,
"theoretical_loss": 3.581541639354664,
"tokens_seen": 1221787648
},
{
"epoch": 0.37,
"learning_rate": 0.00031810431293881645,
"loss": 2.7536,
"theoretical_loss": 3.5815240279814127,
"tokens_seen": 1221853184
},
{
"epoch": 0.37,
"learning_rate": 0.00031809428284854563,
"loss": 2.7448,
"theoretical_loss": 3.5815064178172245,
"tokens_seen": 1221918720
},
{
"epoch": 0.37,
"learning_rate": 0.00031808425275827487,
"loss": 3.0262,
"theoretical_loss": 3.5814888088619528,
"tokens_seen": 1221984256
},
{
"epoch": 0.37,
"learning_rate": 0.000318074222668004,
"loss": 2.6619,
"theoretical_loss": 3.5814712011154484,
"tokens_seen": 1222049792
},
{
"epoch": 0.37,
"learning_rate": 0.00031806419257773323,
"loss": 2.6895,
"theoretical_loss": 3.5814535945775647,
"tokens_seen": 1222115328
},
{
"epoch": 0.37,
"learning_rate": 0.0003180541624874624,
"loss": 2.8971,
"theoretical_loss": 3.5814359892481535,
"tokens_seen": 1222180864
},
{
"epoch": 0.37,
"objective/train/docs_used": 694027,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0180294513702393,
"objective/train/theoretical_loss": 3.5814183851270673,
"objective/train/tokens_used": 1242706400,
"theoretical_loss": 3.5814183851270673,
"tokens_seen": 1222246400
},
{
"epoch": 0.37,
"learning_rate": 0.0003180441323971916,
"loss": 2.8262,
"theoretical_loss": 3.5814183851270673,
"tokens_seen": 1222246400
},
{
"epoch": 0.37,
"learning_rate": 0.00031803410230692077,
"loss": 2.7565,
"theoretical_loss": 3.581400782214158,
"tokens_seen": 1222311936
},
{
"epoch": 0.37,
"learning_rate": 0.00031802407221664995,
"loss": 2.7304,
"theoretical_loss": 3.5813831805092775,
"tokens_seen": 1222377472
},
{
"epoch": 0.37,
"learning_rate": 0.00031801404212637913,
"loss": 2.7989,
"theoretical_loss": 3.581365580012279,
"tokens_seen": 1222443008
},
{
"epoch": 0.37,
"learning_rate": 0.00031800401203610837,
"loss": 2.7945,
"theoretical_loss": 3.581347980723015,
"tokens_seen": 1222508544
},
{
"epoch": 0.37,
"learning_rate": 0.0003179939819458375,
"loss": 2.7958,
"theoretical_loss": 3.5813303826413367,
"tokens_seen": 1222574080
},
{
"epoch": 0.37,
"learning_rate": 0.00031798395185556673,
"loss": 2.799,
"theoretical_loss": 3.5813127857670977,
"tokens_seen": 1222639616
},
{
"epoch": 0.37,
"learning_rate": 0.00031797392176529586,
"loss": 2.816,
"theoretical_loss": 3.5812951901001497,
"tokens_seen": 1222705152
},
{
"epoch": 0.37,
"learning_rate": 0.0003179638916750251,
"loss": 2.8781,
"theoretical_loss": 3.581277595640346,
"tokens_seen": 1222770688
},
{
"epoch": 0.37,
"learning_rate": 0.0003179538615847543,
"loss": 2.7954,
"theoretical_loss": 3.581260002387538,
"tokens_seen": 1222836224
},
{
"epoch": 0.37,
"learning_rate": 0.00031794383149448346,
"loss": 2.6826,
"theoretical_loss": 3.581242410341579,
"tokens_seen": 1222901760
},
{
"epoch": 0.37,
"learning_rate": 0.00031793380140421264,
"loss": 2.8594,
"theoretical_loss": 3.5812248195023213,
"tokens_seen": 1222967296
},
{
"epoch": 0.37,
"learning_rate": 0.0003179237713139418,
"loss": 2.8571,
"theoretical_loss": 3.5812072298696176,
"tokens_seen": 1223032832
},
{
"epoch": 0.37,
"learning_rate": 0.000317913741223671,
"loss": 2.905,
"theoretical_loss": 3.58118964144332,
"tokens_seen": 1223098368
},
{
"epoch": 0.37,
"learning_rate": 0.00031790371113340024,
"loss": 2.7261,
"theoretical_loss": 3.581172054223282,
"tokens_seen": 1223163904
},
{
"epoch": 0.37,
"learning_rate": 0.00031789368104312936,
"loss": 2.5692,
"theoretical_loss": 3.5811544682093555,
"tokens_seen": 1223229440
},
{
"epoch": 0.37,
"learning_rate": 0.0003178836509528586,
"loss": 2.7388,
"theoretical_loss": 3.5811368834013937,
"tokens_seen": 1223294976
},
{
"epoch": 0.37,
"learning_rate": 0.0003178736208625878,
"loss": 2.7383,
"theoretical_loss": 3.5811192997992487,
"tokens_seen": 1223360512
},
{
"epoch": 0.37,
"learning_rate": 0.00031786359077231696,
"loss": 2.683,
"theoretical_loss": 3.581101717402773,
"tokens_seen": 1223426048
},
{
"epoch": 0.37,
"learning_rate": 0.00031785356068204614,
"loss": 2.8045,
"theoretical_loss": 3.581084136211821,
"tokens_seen": 1223491584
},
{
"epoch": 0.37,
"learning_rate": 0.0003178435305917753,
"loss": 2.5617,
"theoretical_loss": 3.5810665562262436,
"tokens_seen": 1223557120
},
{
"epoch": 0.37,
"learning_rate": 0.0003178335005015045,
"loss": 2.8582,
"theoretical_loss": 3.581048977445895,
"tokens_seen": 1223622656
},
{
"epoch": 0.37,
"learning_rate": 0.00031782347041123374,
"loss": 2.675,
"theoretical_loss": 3.581031399870627,
"tokens_seen": 1223688192
},
{
"epoch": 0.37,
"learning_rate": 0.00031781344032096287,
"loss": 2.6946,
"theoretical_loss": 3.5810138235002933,
"tokens_seen": 1223753728
},
{
"epoch": 0.37,
"learning_rate": 0.0003178034102306921,
"loss": 2.8771,
"theoretical_loss": 3.580996248334746,
"tokens_seen": 1223819264
},
{
"epoch": 0.37,
"objective/train/docs_used": 695222,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7954161167144775,
"objective/train/theoretical_loss": 3.580978674373838,
"objective/train/tokens_used": 1244344800,
"theoretical_loss": 3.580978674373838,
"tokens_seen": 1223884800
},
{
"epoch": 0.37,
"learning_rate": 0.00031779338014042123,
"loss": 2.789,
"theoretical_loss": 3.580978674373838,
"tokens_seen": 1223884800
},
{
"epoch": 0.37,
"learning_rate": 0.00031778335005015046,
"loss": 2.5778,
"theoretical_loss": 3.580961101617423,
"tokens_seen": 1223950336
},
{
"epoch": 0.37,
"learning_rate": 0.00031777331995987964,
"loss": 2.7388,
"theoretical_loss": 3.580943530065354,
"tokens_seen": 1224015872
},
{
"epoch": 0.37,
"learning_rate": 0.0003177632898696088,
"loss": 2.8557,
"theoretical_loss": 3.580925959717483,
"tokens_seen": 1224081408
},
{
"epoch": 0.37,
"learning_rate": 0.000317753259779338,
"loss": 2.7463,
"theoretical_loss": 3.580908390573663,
"tokens_seen": 1224146944
},
{
"epoch": 0.37,
"learning_rate": 0.0003177432296890672,
"loss": 2.6134,
"theoretical_loss": 3.580890822633749,
"tokens_seen": 1224212480
},
{
"epoch": 0.37,
"learning_rate": 0.00031773319959879637,
"loss": 2.8383,
"theoretical_loss": 3.5808732558975915,
"tokens_seen": 1224278016
},
{
"epoch": 0.37,
"learning_rate": 0.0003177231695085256,
"loss": 2.7193,
"theoretical_loss": 3.580855690365045,
"tokens_seen": 1224343552
},
{
"epoch": 0.37,
"learning_rate": 0.00031771313941825473,
"loss": 2.5693,
"theoretical_loss": 3.580838126035962,
"tokens_seen": 1224409088
},
{
"epoch": 0.37,
"learning_rate": 0.00031770310932798397,
"loss": 2.6565,
"theoretical_loss": 3.580820562910197,
"tokens_seen": 1224474624
},
{
"epoch": 0.37,
"learning_rate": 0.00031769307923771315,
"loss": 2.8056,
"theoretical_loss": 3.5808030009876015,
"tokens_seen": 1224540160
},
{
"epoch": 0.37,
"learning_rate": 0.00031768304914744233,
"loss": 2.7544,
"theoretical_loss": 3.580785440268029,
"tokens_seen": 1224605696
},
{
"epoch": 0.37,
"learning_rate": 0.0003176730190571715,
"loss": 2.6557,
"theoretical_loss": 3.5807678807513335,
"tokens_seen": 1224671232
},
{
"epoch": 0.37,
"learning_rate": 0.0003176629889669007,
"loss": 2.6111,
"theoretical_loss": 3.5807503224373676,
"tokens_seen": 1224736768
},
{
"epoch": 0.37,
"learning_rate": 0.00031765295887662987,
"loss": 2.9226,
"theoretical_loss": 3.580732765325985,
"tokens_seen": 1224802304
},
{
"epoch": 0.37,
"learning_rate": 0.0003176429287863591,
"loss": 2.8189,
"theoretical_loss": 3.5807152094170385,
"tokens_seen": 1224867840
},
{
"epoch": 0.37,
"learning_rate": 0.00031763289869608823,
"loss": 2.9356,
"theoretical_loss": 3.5806976547103817,
"tokens_seen": 1224933376
},
{
"epoch": 0.37,
"learning_rate": 0.00031762286860581747,
"loss": 2.8249,
"theoretical_loss": 3.5806801012058678,
"tokens_seen": 1224998912
},
{
"epoch": 0.37,
"learning_rate": 0.0003176128385155466,
"loss": 2.8726,
"theoretical_loss": 3.5806625489033506,
"tokens_seen": 1225064448
},
{
"epoch": 0.37,
"learning_rate": 0.00031760280842527583,
"loss": 2.727,
"theoretical_loss": 3.580644997802683,
"tokens_seen": 1225129984
},
{
"epoch": 0.37,
"learning_rate": 0.000317592778335005,
"loss": 2.6864,
"theoretical_loss": 3.580627447903719,
"tokens_seen": 1225195520
},
{
"epoch": 0.37,
"learning_rate": 0.0003175827482447342,
"loss": 2.7201,
"theoretical_loss": 3.580609899206311,
"tokens_seen": 1225261056
},
{
"epoch": 0.37,
"learning_rate": 0.0003175727181544634,
"loss": 2.5987,
"theoretical_loss": 3.5805923517103135,
"tokens_seen": 1225326592
},
{
"epoch": 0.37,
"learning_rate": 0.0003175626880641926,
"loss": 2.6345,
"theoretical_loss": 3.58057480541558,
"tokens_seen": 1225392128
},
{
"epoch": 0.37,
"learning_rate": 0.0003175526579739218,
"loss": 2.7863,
"theoretical_loss": 3.580557260321963,
"tokens_seen": 1225457664
},
{
"epoch": 0.37,
"objective/train/docs_used": 695595,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.116090774536133,
"objective/train/theoretical_loss": 3.5805397164293167,
"objective/train/tokens_used": 1245983200,
"theoretical_loss": 3.5805397164293167,
"tokens_seen": 1225523200
},
{
"epoch": 0.37,
"learning_rate": 0.00031754262788365097,
"loss": 2.5485,
"theoretical_loss": 3.5805397164293167,
"tokens_seen": 1225523200
},
{
"epoch": 0.37,
"learning_rate": 0.00031753259779338015,
"loss": 2.7816,
"theoretical_loss": 3.580522173737495,
"tokens_seen": 1225588736
},
{
"epoch": 0.37,
"learning_rate": 0.00031752256770310933,
"loss": 2.6114,
"theoretical_loss": 3.580504632246351,
"tokens_seen": 1225654272
},
{
"epoch": 0.37,
"learning_rate": 0.00031751253761283857,
"loss": 2.6292,
"theoretical_loss": 3.5804870919557388,
"tokens_seen": 1225719808
},
{
"epoch": 0.37,
"learning_rate": 0.0003175025075225677,
"loss": 2.7409,
"theoretical_loss": 3.5804695528655115,
"tokens_seen": 1225785344
},
{
"epoch": 0.37,
"learning_rate": 0.00031749247743229693,
"loss": 2.8125,
"theoretical_loss": 3.5804520149755232,
"tokens_seen": 1225850880
},
{
"epoch": 0.37,
"learning_rate": 0.00031748244734202606,
"loss": 2.6946,
"theoretical_loss": 3.5804344782856274,
"tokens_seen": 1225916416
},
{
"epoch": 0.37,
"learning_rate": 0.0003174724172517553,
"loss": 2.5816,
"theoretical_loss": 3.580416942795678,
"tokens_seen": 1225981952
},
{
"epoch": 0.37,
"learning_rate": 0.0003174623871614845,
"loss": 2.8434,
"theoretical_loss": 3.5803994085055284,
"tokens_seen": 1226047488
},
{
"epoch": 0.37,
"learning_rate": 0.00031745235707121366,
"loss": 2.6919,
"theoretical_loss": 3.580381875415033,
"tokens_seen": 1226113024
},
{
"epoch": 0.37,
"learning_rate": 0.00031744232698094284,
"loss": 2.8662,
"theoretical_loss": 3.5803643435240446,
"tokens_seen": 1226178560
},
{
"epoch": 0.37,
"learning_rate": 0.000317432296890672,
"loss": 2.5723,
"theoretical_loss": 3.580346812832418,
"tokens_seen": 1226244096
},
{
"epoch": 0.37,
"learning_rate": 0.0003174222668004012,
"loss": 2.6612,
"theoretical_loss": 3.5803292833400064,
"tokens_seen": 1226309632
},
{
"epoch": 0.37,
"learning_rate": 0.00031741223671013044,
"loss": 2.8795,
"theoretical_loss": 3.5803117550466643,
"tokens_seen": 1226375168
},
{
"epoch": 0.37,
"learning_rate": 0.00031740220661985956,
"loss": 2.7488,
"theoretical_loss": 3.580294227952246,
"tokens_seen": 1226440704
},
{
"epoch": 0.37,
"learning_rate": 0.0003173921765295888,
"loss": 2.6022,
"theoretical_loss": 3.5802767020566035,
"tokens_seen": 1226506240
},
{
"epoch": 0.37,
"learning_rate": 0.000317382146439318,
"loss": 2.7923,
"theoretical_loss": 3.5802591773595926,
"tokens_seen": 1226571776
},
{
"epoch": 0.37,
"learning_rate": 0.00031737211634904716,
"loss": 2.6165,
"theoretical_loss": 3.5802416538610666,
"tokens_seen": 1226637312
},
{
"epoch": 0.37,
"learning_rate": 0.00031736208625877634,
"loss": 2.9011,
"theoretical_loss": 3.58022413156088,
"tokens_seen": 1226702848
},
{
"epoch": 0.37,
"learning_rate": 0.0003173520561685055,
"loss": 2.6587,
"theoretical_loss": 3.580206610458886,
"tokens_seen": 1226768384
},
{
"epoch": 0.37,
"learning_rate": 0.0003173420260782347,
"loss": 2.6878,
"theoretical_loss": 3.5801890905549394,
"tokens_seen": 1226833920
},
{
"epoch": 0.37,
"learning_rate": 0.00031733199598796394,
"loss": 2.6823,
"theoretical_loss": 3.5801715718488936,
"tokens_seen": 1226899456
},
{
"epoch": 0.37,
"learning_rate": 0.00031732196589769307,
"loss": 2.7846,
"theoretical_loss": 3.5801540543406034,
"tokens_seen": 1226964992
},
{
"epoch": 0.37,
"learning_rate": 0.0003173119358074223,
"loss": 2.7766,
"theoretical_loss": 3.580136538029923,
"tokens_seen": 1227030528
},
{
"epoch": 0.37,
"learning_rate": 0.00031730190571715143,
"loss": 2.7606,
"theoretical_loss": 3.5801190229167057,
"tokens_seen": 1227096064
},
{
"epoch": 0.37,
"objective/train/docs_used": 696474,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2769222259521484,
"objective/train/theoretical_loss": 3.580101509000806,
"objective/train/tokens_used": 1247621600,
"theoretical_loss": 3.580101509000806,
"tokens_seen": 1227161600
},
{
"epoch": 0.37,
"learning_rate": 0.00031729187562688066,
"loss": 2.8791,
"theoretical_loss": 3.580101509000806,
"tokens_seen": 1227161600
},
{
"epoch": 0.37,
"learning_rate": 0.00031728184553660984,
"loss": 2.8868,
"theoretical_loss": 3.5800839962820787,
"tokens_seen": 1227227136
},
{
"epoch": 0.37,
"learning_rate": 0.000317271815446339,
"loss": 2.5709,
"theoretical_loss": 3.580066484760378,
"tokens_seen": 1227292672
},
{
"epoch": 0.37,
"learning_rate": 0.0003172617853560682,
"loss": 2.5781,
"theoretical_loss": 3.5800489744355573,
"tokens_seen": 1227358208
},
{
"epoch": 0.37,
"learning_rate": 0.0003172517552657974,
"loss": 2.718,
"theoretical_loss": 3.5800314653074716,
"tokens_seen": 1227423744
},
{
"epoch": 0.37,
"learning_rate": 0.00031724172517552657,
"loss": 2.8943,
"theoretical_loss": 3.580013957375975,
"tokens_seen": 1227489280
},
{
"epoch": 0.37,
"learning_rate": 0.0003172316950852558,
"loss": 2.5928,
"theoretical_loss": 3.5799964506409214,
"tokens_seen": 1227554816
},
{
"epoch": 0.37,
"learning_rate": 0.00031722166499498493,
"loss": 2.6286,
"theoretical_loss": 3.5799789451021664,
"tokens_seen": 1227620352
},
{
"epoch": 0.37,
"learning_rate": 0.00031721163490471417,
"loss": 2.5952,
"theoretical_loss": 3.579961440759564,
"tokens_seen": 1227685888
},
{
"epoch": 0.37,
"learning_rate": 0.00031720160481444335,
"loss": 2.7073,
"theoretical_loss": 3.579943937612967,
"tokens_seen": 1227751424
},
{
"epoch": 0.37,
"learning_rate": 0.00031719157472417253,
"loss": 2.8598,
"theoretical_loss": 3.579926435662232,
"tokens_seen": 1227816960
},
{
"epoch": 0.37,
"learning_rate": 0.0003171815446339017,
"loss": 2.8607,
"theoretical_loss": 3.5799089349072126,
"tokens_seen": 1227882496
},
{
"epoch": 0.37,
"learning_rate": 0.0003171715145436309,
"loss": 2.6763,
"theoretical_loss": 3.579891435347763,
"tokens_seen": 1227948032
},
{
"epoch": 0.37,
"learning_rate": 0.00031716148445336007,
"loss": 2.6742,
"theoretical_loss": 3.5798739369837382,
"tokens_seen": 1228013568
},
{
"epoch": 0.37,
"learning_rate": 0.0003171514543630893,
"loss": 2.8094,
"theoretical_loss": 3.5798564398149924,
"tokens_seen": 1228079104
},
{
"epoch": 0.37,
"learning_rate": 0.00031714142427281843,
"loss": 2.8235,
"theoretical_loss": 3.5798389438413807,
"tokens_seen": 1228144640
},
{
"epoch": 0.37,
"learning_rate": 0.00031713139418254767,
"loss": 2.8293,
"theoretical_loss": 3.5798214490627567,
"tokens_seen": 1228210176
},
{
"epoch": 0.37,
"learning_rate": 0.0003171213640922768,
"loss": 2.7927,
"theoretical_loss": 3.5798039554789765,
"tokens_seen": 1228275712
},
{
"epoch": 0.37,
"learning_rate": 0.00031711133400200603,
"loss": 2.8553,
"theoretical_loss": 3.579786463089893,
"tokens_seen": 1228341248
},
{
"epoch": 0.37,
"learning_rate": 0.0003171013039117352,
"loss": 2.7281,
"theoretical_loss": 3.5797689718953625,
"tokens_seen": 1228406784
},
{
"epoch": 0.37,
"learning_rate": 0.0003170912738214644,
"loss": 2.8916,
"theoretical_loss": 3.5797514818952383,
"tokens_seen": 1228472320
},
{
"epoch": 0.37,
"learning_rate": 0.0003170812437311936,
"loss": 2.8023,
"theoretical_loss": 3.579733993089377,
"tokens_seen": 1228537856
},
{
"epoch": 0.37,
"learning_rate": 0.0003170712136409228,
"loss": 2.5461,
"theoretical_loss": 3.579716505477631,
"tokens_seen": 1228603392
},
{
"epoch": 0.37,
"learning_rate": 0.00031706118355065194,
"loss": 2.7576,
"theoretical_loss": 3.5796990190598565,
"tokens_seen": 1228668928
},
{
"epoch": 0.37,
"learning_rate": 0.0003170511534603812,
"loss": 2.8951,
"theoretical_loss": 3.5796815338359087,
"tokens_seen": 1228734464
},
{
"epoch": 0.37,
"objective/train/docs_used": 696846,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.509882688522339,
"objective/train/theoretical_loss": 3.5796640498056407,
"objective/train/tokens_used": 1249260000,
"theoretical_loss": 3.5796640498056407,
"tokens_seen": 1228800000
},
{
"epoch": 0.37,
"learning_rate": 0.0003170411233701103,
"loss": 2.8441,
"theoretical_loss": 3.5796640498056407,
"tokens_seen": 1228800000
},
{
"epoch": 0.37,
"learning_rate": 0.00031703109327983954,
"loss": 2.4928,
"theoretical_loss": 3.5796465669689095,
"tokens_seen": 1228865536
},
{
"epoch": 0.37,
"learning_rate": 0.0003170210631895687,
"loss": 2.7226,
"theoretical_loss": 3.579629085325568,
"tokens_seen": 1228931072
},
{
"epoch": 0.37,
"learning_rate": 0.0003170110330992979,
"loss": 2.8586,
"theoretical_loss": 3.5796116048754727,
"tokens_seen": 1228996608
},
{
"epoch": 0.37,
"learning_rate": 0.0003170010030090271,
"loss": 2.6202,
"theoretical_loss": 3.5795941256184776,
"tokens_seen": 1229062144
},
{
"epoch": 0.37,
"learning_rate": 0.00031699097291875626,
"loss": 2.7964,
"theoretical_loss": 3.579576647554438,
"tokens_seen": 1229127680
},
{
"epoch": 0.37,
"learning_rate": 0.00031698094282848544,
"loss": 2.4914,
"theoretical_loss": 3.5795591706832086,
"tokens_seen": 1229193216
},
{
"epoch": 0.37,
"learning_rate": 0.0003169709127382147,
"loss": 2.7169,
"theoretical_loss": 3.5795416950046444,
"tokens_seen": 1229258752
},
{
"epoch": 0.37,
"learning_rate": 0.0003169608826479438,
"loss": 2.619,
"theoretical_loss": 3.579524220518601,
"tokens_seen": 1229324288
},
{
"epoch": 0.37,
"learning_rate": 0.00031695085255767304,
"loss": 2.5838,
"theoretical_loss": 3.579506747224933,
"tokens_seen": 1229389824
},
{
"epoch": 0.37,
"learning_rate": 0.00031694082246740217,
"loss": 2.6459,
"theoretical_loss": 3.5794892751234952,
"tokens_seen": 1229455360
},
{
"epoch": 0.37,
"learning_rate": 0.0003169307923771314,
"loss": 2.6864,
"theoretical_loss": 3.579471804214143,
"tokens_seen": 1229520896
},
{
"epoch": 0.37,
"learning_rate": 0.0003169207622868606,
"loss": 2.896,
"theoretical_loss": 3.579454334496732,
"tokens_seen": 1229586432
},
{
"epoch": 0.37,
"learning_rate": 0.00031691073219658976,
"loss": 2.6729,
"theoretical_loss": 3.579436865971117,
"tokens_seen": 1229651968
},
{
"epoch": 0.37,
"learning_rate": 0.00031690070210631894,
"loss": 2.896,
"theoretical_loss": 3.5794193986371527,
"tokens_seen": 1229717504
},
{
"epoch": 0.37,
"learning_rate": 0.0003168906720160482,
"loss": 2.541,
"theoretical_loss": 3.579401932494695,
"tokens_seen": 1229783040
},
{
"epoch": 0.37,
"learning_rate": 0.0003168806419257773,
"loss": 2.787,
"theoretical_loss": 3.579384467543599,
"tokens_seen": 1229848576
},
{
"epoch": 0.37,
"learning_rate": 0.00031687061183550654,
"loss": 2.7853,
"theoretical_loss": 3.5793670037837195,
"tokens_seen": 1229914112
},
{
"epoch": 0.37,
"learning_rate": 0.00031686058174523567,
"loss": 2.9784,
"theoretical_loss": 3.5793495412149126,
"tokens_seen": 1229979648
},
{
"epoch": 0.37,
"learning_rate": 0.0003168505516549649,
"loss": 2.813,
"theoretical_loss": 3.579332079837033,
"tokens_seen": 1230045184
},
{
"epoch": 0.37,
"learning_rate": 0.0003168405215646941,
"loss": 2.6062,
"theoretical_loss": 3.579314619649936,
"tokens_seen": 1230110720
},
{
"epoch": 0.37,
"learning_rate": 0.00031683049147442327,
"loss": 2.7495,
"theoretical_loss": 3.5792971606534767,
"tokens_seen": 1230176256
},
{
"epoch": 0.37,
"learning_rate": 0.00031682046138415245,
"loss": 2.7231,
"theoretical_loss": 3.5792797028475114,
"tokens_seen": 1230241792
},
{
"epoch": 0.37,
"learning_rate": 0.00031681043129388163,
"loss": 2.8999,
"theoretical_loss": 3.579262246231895,
"tokens_seen": 1230307328
},
{
"epoch": 0.37,
"learning_rate": 0.00031680040120361086,
"loss": 2.8195,
"theoretical_loss": 3.5792447908064826,
"tokens_seen": 1230372864
},
{
"epoch": 0.37,
"objective/train/docs_used": 698095,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0689139366149902,
"objective/train/theoretical_loss": 3.5792273365711305,
"objective/train/tokens_used": 1250898400,
"theoretical_loss": 3.5792273365711305,
"tokens_seen": 1230438400
},
{
"epoch": 0.37,
"learning_rate": 0.00031679037111334004,
"loss": 2.7956,
"theoretical_loss": 3.5792273365711305,
"tokens_seen": 1230438400
},
{
"epoch": 0.37,
"learning_rate": 0.0003167803410230692,
"loss": 2.8374,
"theoretical_loss": 3.5792098835256936,
"tokens_seen": 1230503936
},
{
"epoch": 0.37,
"learning_rate": 0.0003167703109327984,
"loss": 2.7872,
"theoretical_loss": 3.5791924316700277,
"tokens_seen": 1230569472
},
{
"epoch": 0.37,
"learning_rate": 0.0003167602808425276,
"loss": 2.7968,
"theoretical_loss": 3.5791749810039875,
"tokens_seen": 1230635008
},
{
"epoch": 0.37,
"learning_rate": 0.00031675025075225677,
"loss": 2.8191,
"theoretical_loss": 3.5791575315274295,
"tokens_seen": 1230700544
},
{
"epoch": 0.37,
"learning_rate": 0.000316740220661986,
"loss": 2.9371,
"theoretical_loss": 3.579140083240209,
"tokens_seen": 1230766080
},
{
"epoch": 0.37,
"learning_rate": 0.00031673019057171513,
"loss": 2.6996,
"theoretical_loss": 3.5791226361421815,
"tokens_seen": 1230831616
},
{
"epoch": 0.37,
"learning_rate": 0.00031672016048144437,
"loss": 2.7995,
"theoretical_loss": 3.579105190233203,
"tokens_seen": 1230897152
},
{
"epoch": 0.37,
"learning_rate": 0.00031671013039117355,
"loss": 2.6821,
"theoretical_loss": 3.579087745513129,
"tokens_seen": 1230962688
},
{
"epoch": 0.37,
"learning_rate": 0.00031670010030090273,
"loss": 2.7865,
"theoretical_loss": 3.579070301981815,
"tokens_seen": 1231028224
},
{
"epoch": 0.37,
"learning_rate": 0.0003166900702106319,
"loss": 2.7781,
"theoretical_loss": 3.5790528596391167,
"tokens_seen": 1231093760
},
{
"epoch": 0.37,
"learning_rate": 0.0003166800401203611,
"loss": 2.5713,
"theoretical_loss": 3.5790354184848896,
"tokens_seen": 1231159296
},
{
"epoch": 0.37,
"learning_rate": 0.00031667001003009027,
"loss": 2.871,
"theoretical_loss": 3.5790179785189906,
"tokens_seen": 1231224832
},
{
"epoch": 0.37,
"learning_rate": 0.0003166599799398195,
"loss": 2.7824,
"theoretical_loss": 3.579000539741274,
"tokens_seen": 1231290368
},
{
"epoch": 0.37,
"learning_rate": 0.00031664994984954863,
"loss": 2.8687,
"theoretical_loss": 3.5789831021515965,
"tokens_seen": 1231355904
},
{
"epoch": 0.37,
"learning_rate": 0.00031663991975927787,
"loss": 2.5368,
"theoretical_loss": 3.5789656657498137,
"tokens_seen": 1231421440
},
{
"epoch": 0.37,
"learning_rate": 0.000316629889669007,
"loss": 2.7274,
"theoretical_loss": 3.578948230535782,
"tokens_seen": 1231486976
},
{
"epoch": 0.37,
"learning_rate": 0.00031661985957873623,
"loss": 2.6497,
"theoretical_loss": 3.5789307965093564,
"tokens_seen": 1231552512
},
{
"epoch": 0.37,
"learning_rate": 0.0003166098294884654,
"loss": 2.7224,
"theoretical_loss": 3.5789133636703934,
"tokens_seen": 1231618048
},
{
"epoch": 0.37,
"learning_rate": 0.0003165997993981946,
"loss": 2.8495,
"theoretical_loss": 3.5788959320187486,
"tokens_seen": 1231683584
},
{
"epoch": 0.37,
"learning_rate": 0.0003165897693079238,
"loss": 2.4992,
"theoretical_loss": 3.5788785015542786,
"tokens_seen": 1231749120
},
{
"epoch": 0.37,
"learning_rate": 0.000316579739217653,
"loss": 2.892,
"theoretical_loss": 3.578861072276838,
"tokens_seen": 1231814656
},
{
"epoch": 0.37,
"learning_rate": 0.00031656970912738214,
"loss": 2.6577,
"theoretical_loss": 3.5788436441862843,
"tokens_seen": 1231880192
},
{
"epoch": 0.37,
"learning_rate": 0.0003165596790371114,
"loss": 2.8404,
"theoretical_loss": 3.578826217282473,
"tokens_seen": 1231945728
},
{
"epoch": 0.37,
"learning_rate": 0.0003165496489468405,
"loss": 2.7093,
"theoretical_loss": 3.57880879156526,
"tokens_seen": 1232011264
},
{
"epoch": 0.37,
"objective/train/docs_used": 698724,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7369630336761475,
"objective/train/theoretical_loss": 3.5787913670345013,
"objective/train/tokens_used": 1252536800,
"theoretical_loss": 3.5787913670345013,
"tokens_seen": 1232076800
},
{
"epoch": 0.37,
"learning_rate": 0.00031653961885656974,
"loss": 2.727,
"theoretical_loss": 3.5787913670345013,
"tokens_seen": 1232076800
},
{
"epoch": 0.37,
"learning_rate": 0.0003165295887662989,
"loss": 2.6632,
"theoretical_loss": 3.578773943690054,
"tokens_seen": 1232142336
},
{
"epoch": 0.37,
"learning_rate": 0.0003165195586760281,
"loss": 2.8287,
"theoretical_loss": 3.5787565215317727,
"tokens_seen": 1232207872
},
{
"epoch": 0.37,
"learning_rate": 0.0003165095285857573,
"loss": 2.9034,
"theoretical_loss": 3.578739100559515,
"tokens_seen": 1232273408
},
{
"epoch": 0.37,
"learning_rate": 0.00031649949849548646,
"loss": 2.5421,
"theoretical_loss": 3.578721680773136,
"tokens_seen": 1232338944
},
{
"epoch": 0.37,
"learning_rate": 0.00031648946840521564,
"loss": 2.7615,
"theoretical_loss": 3.5787042621724927,
"tokens_seen": 1232404480
},
{
"epoch": 0.37,
"learning_rate": 0.0003164794383149449,
"loss": 2.6775,
"theoretical_loss": 3.5786868447574403,
"tokens_seen": 1232470016
},
{
"epoch": 0.37,
"learning_rate": 0.000316469408224674,
"loss": 2.743,
"theoretical_loss": 3.5786694285278364,
"tokens_seen": 1232535552
},
{
"epoch": 0.37,
"learning_rate": 0.00031645937813440324,
"loss": 2.7367,
"theoretical_loss": 3.578652013483537,
"tokens_seen": 1232601088
},
{
"epoch": 0.37,
"learning_rate": 0.00031644934804413237,
"loss": 2.9572,
"theoretical_loss": 3.5786345996243973,
"tokens_seen": 1232666624
},
{
"epoch": 0.37,
"learning_rate": 0.0003164393179538616,
"loss": 2.449,
"theoretical_loss": 3.5786171869502743,
"tokens_seen": 1232732160
},
{
"epoch": 0.37,
"learning_rate": 0.0003164292878635908,
"loss": 2.8238,
"theoretical_loss": 3.578599775461025,
"tokens_seen": 1232797696
},
{
"epoch": 0.37,
"learning_rate": 0.00031641925777331996,
"loss": 2.6135,
"theoretical_loss": 3.578582365156505,
"tokens_seen": 1232863232
},
{
"epoch": 0.37,
"learning_rate": 0.00031640922768304914,
"loss": 2.5416,
"theoretical_loss": 3.5785649560365718,
"tokens_seen": 1232928768
},
{
"epoch": 0.37,
"learning_rate": 0.0003163991975927784,
"loss": 2.729,
"theoretical_loss": 3.5785475481010796,
"tokens_seen": 1232994304
},
{
"epoch": 0.37,
"learning_rate": 0.0003163891675025075,
"loss": 2.8401,
"theoretical_loss": 3.5785301413498876,
"tokens_seen": 1233059840
},
{
"epoch": 0.37,
"learning_rate": 0.00031637913741223674,
"loss": 2.8976,
"theoretical_loss": 3.57851273578285,
"tokens_seen": 1233125376
},
{
"epoch": 0.37,
"learning_rate": 0.00031636910732196587,
"loss": 2.9508,
"theoretical_loss": 3.578495331399825,
"tokens_seen": 1233190912
},
{
"epoch": 0.37,
"learning_rate": 0.0003163590772316951,
"loss": 2.7645,
"theoretical_loss": 3.5784779282006682,
"tokens_seen": 1233256448
},
{
"epoch": 0.37,
"learning_rate": 0.0003163490471414243,
"loss": 2.6163,
"theoretical_loss": 3.5784605261852365,
"tokens_seen": 1233321984
},
{
"epoch": 0.37,
"learning_rate": 0.00031633901705115347,
"loss": 2.8523,
"theoretical_loss": 3.5784431253533864,
"tokens_seen": 1233387520
},
{
"epoch": 0.37,
"learning_rate": 0.00031632898696088265,
"loss": 2.8331,
"theoretical_loss": 3.5784257257049745,
"tokens_seen": 1233453056
},
{
"epoch": 0.37,
"learning_rate": 0.00031631895687061183,
"loss": 2.6979,
"theoretical_loss": 3.578408327239857,
"tokens_seen": 1233518592
},
{
"epoch": 0.37,
"learning_rate": 0.000316308926780341,
"loss": 2.7807,
"theoretical_loss": 3.5783909299578918,
"tokens_seen": 1233584128
},
{
"epoch": 0.37,
"learning_rate": 0.00031629889669007024,
"loss": 2.7772,
"theoretical_loss": 3.578373533858935,
"tokens_seen": 1233649664
},
{
"epoch": 0.37,
"objective/train/docs_used": 699861,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.794823408126831,
"objective/train/theoretical_loss": 3.578356138942842,
"objective/train/tokens_used": 1254175200,
"theoretical_loss": 3.578356138942842,
"tokens_seen": 1233715200
},
{
"epoch": 0.37,
"learning_rate": 0.00031628886659979937,
"loss": 2.8246,
"theoretical_loss": 3.578356138942842,
"tokens_seen": 1233715200
},
{
"epoch": 0.37,
"learning_rate": 0.0003162788365095286,
"loss": 2.7831,
"theoretical_loss": 3.5783387452094715,
"tokens_seen": 1233780736
},
{
"epoch": 0.37,
"learning_rate": 0.00031626880641925773,
"loss": 2.6493,
"theoretical_loss": 3.578321352658679,
"tokens_seen": 1233846272
},
{
"epoch": 0.37,
"learning_rate": 0.00031625877632898697,
"loss": 2.628,
"theoretical_loss": 3.578303961290322,
"tokens_seen": 1233911808
},
{
"epoch": 0.37,
"learning_rate": 0.00031624874623871615,
"loss": 2.6145,
"theoretical_loss": 3.5782865711042575,
"tokens_seen": 1233977344
},
{
"epoch": 0.37,
"learning_rate": 0.00031623871614844533,
"loss": 2.7008,
"theoretical_loss": 3.5782691821003407,
"tokens_seen": 1234042880
},
{
"epoch": 0.37,
"learning_rate": 0.0003162286860581745,
"loss": 2.899,
"theoretical_loss": 3.578251794278431,
"tokens_seen": 1234108416
},
{
"epoch": 0.37,
"learning_rate": 0.00031621865596790375,
"loss": 2.7755,
"theoretical_loss": 3.578234407638383,
"tokens_seen": 1234173952
},
{
"epoch": 0.37,
"learning_rate": 0.0003162086258776329,
"loss": 2.7027,
"theoretical_loss": 3.5782170221800556,
"tokens_seen": 1234239488
},
{
"epoch": 0.37,
"learning_rate": 0.0003161985957873621,
"loss": 2.5809,
"theoretical_loss": 3.5781996379033036,
"tokens_seen": 1234305024
},
{
"epoch": 0.37,
"learning_rate": 0.00031618856569709124,
"loss": 2.8074,
"theoretical_loss": 3.5781822548079854,
"tokens_seen": 1234370560
},
{
"epoch": 0.37,
"learning_rate": 0.00031617853560682047,
"loss": 2.5818,
"theoretical_loss": 3.578164872893958,
"tokens_seen": 1234436096
},
{
"epoch": 0.37,
"learning_rate": 0.00031616850551654965,
"loss": 2.6525,
"theoretical_loss": 3.5781474921610776,
"tokens_seen": 1234501632
},
{
"epoch": 0.37,
"learning_rate": 0.00031615847542627883,
"loss": 2.8084,
"theoretical_loss": 3.578130112609202,
"tokens_seen": 1234567168
},
{
"epoch": 0.37,
"learning_rate": 0.000316148445336008,
"loss": 2.2807,
"theoretical_loss": 3.5781127342381884,
"tokens_seen": 1234632704
},
{
"epoch": 0.37,
"learning_rate": 0.0003161384152457372,
"loss": 2.6845,
"theoretical_loss": 3.578095357047893,
"tokens_seen": 1234698240
},
{
"epoch": 0.37,
"learning_rate": 0.0003161283851554664,
"loss": 2.5949,
"theoretical_loss": 3.5780779810381733,
"tokens_seen": 1234763776
},
{
"epoch": 0.37,
"learning_rate": 0.0003161183550651956,
"loss": 3.0014,
"theoretical_loss": 3.578060606208887,
"tokens_seen": 1234829312
},
{
"epoch": 0.37,
"learning_rate": 0.00031610832497492474,
"loss": 2.7142,
"theoretical_loss": 3.5780432325598905,
"tokens_seen": 1234894848
},
{
"epoch": 0.37,
"learning_rate": 0.000316098294884654,
"loss": 2.7802,
"theoretical_loss": 3.5780258600910413,
"tokens_seen": 1234960384
},
{
"epoch": 0.37,
"learning_rate": 0.0003160882647943831,
"loss": 2.8679,
"theoretical_loss": 3.5780084888021966,
"tokens_seen": 1235025920
},
{
"epoch": 0.37,
"learning_rate": 0.00031607823470411234,
"loss": 2.8715,
"theoretical_loss": 3.5779911186932134,
"tokens_seen": 1235091456
},
{
"epoch": 0.37,
"learning_rate": 0.0003160682046138415,
"loss": 2.7672,
"theoretical_loss": 3.5779737497639497,
"tokens_seen": 1235156992
},
{
"epoch": 0.37,
"learning_rate": 0.0003160581745235707,
"loss": 2.7153,
"theoretical_loss": 3.5779563820142624,
"tokens_seen": 1235222528
},
{
"epoch": 0.37,
"learning_rate": 0.00031604814443329994,
"loss": 2.702,
"theoretical_loss": 3.577939015444008,
"tokens_seen": 1235288064
},
{
"epoch": 0.37,
"objective/train/docs_used": 700550,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0249288082122803,
"objective/train/theoretical_loss": 3.577921650053045,
"objective/train/tokens_used": 1255813600,
"theoretical_loss": 3.577921650053045,
"tokens_seen": 1235353600
},
{
"epoch": 0.37,
"learning_rate": 0.0003160381143430291,
"loss": 3.0282,
"theoretical_loss": 3.577921650053045,
"tokens_seen": 1235353600
},
{
"epoch": 0.37,
"learning_rate": 0.0003160280842527583,
"loss": 2.7869,
"theoretical_loss": 3.57790428584123,
"tokens_seen": 1235419136
},
{
"epoch": 0.37,
"learning_rate": 0.0003160180541624875,
"loss": 2.6236,
"theoretical_loss": 3.577886922808421,
"tokens_seen": 1235484672
},
{
"epoch": 0.37,
"learning_rate": 0.00031600802407221666,
"loss": 2.4895,
"theoretical_loss": 3.5778695609544755,
"tokens_seen": 1235550208
},
{
"epoch": 0.37,
"learning_rate": 0.00031599799398194584,
"loss": 2.7629,
"theoretical_loss": 3.57785220027925,
"tokens_seen": 1235615744
},
{
"epoch": 0.37,
"learning_rate": 0.0003159879638916751,
"loss": 2.8776,
"theoretical_loss": 3.577834840782603,
"tokens_seen": 1235681280
},
{
"epoch": 0.37,
"learning_rate": 0.0003159779338014042,
"loss": 2.8141,
"theoretical_loss": 3.5778174824643907,
"tokens_seen": 1235746816
},
{
"epoch": 0.37,
"learning_rate": 0.00031596790371113344,
"loss": 2.8081,
"theoretical_loss": 3.577800125324472,
"tokens_seen": 1235812352
},
{
"epoch": 0.37,
"learning_rate": 0.00031595787362086257,
"loss": 2.7631,
"theoretical_loss": 3.577782769362704,
"tokens_seen": 1235877888
},
{
"epoch": 0.37,
"learning_rate": 0.0003159478435305918,
"loss": 2.8438,
"theoretical_loss": 3.577765414578944,
"tokens_seen": 1235943424
},
{
"epoch": 0.37,
"learning_rate": 0.000315937813440321,
"loss": 2.6265,
"theoretical_loss": 3.5777480609730494,
"tokens_seen": 1236008960
},
{
"epoch": 0.37,
"learning_rate": 0.00031592778335005016,
"loss": 2.733,
"theoretical_loss": 3.577730708544878,
"tokens_seen": 1236074496
},
{
"epoch": 0.37,
"learning_rate": 0.00031591775325977934,
"loss": 2.8012,
"theoretical_loss": 3.577713357294288,
"tokens_seen": 1236140032
},
{
"epoch": 0.37,
"learning_rate": 0.0003159077231695086,
"loss": 2.8408,
"theoretical_loss": 3.5776960072211366,
"tokens_seen": 1236205568
},
{
"epoch": 0.37,
"learning_rate": 0.0003158976930792377,
"loss": 2.8117,
"theoretical_loss": 3.577678658325281,
"tokens_seen": 1236271104
},
{
"epoch": 0.37,
"learning_rate": 0.00031588766298896694,
"loss": 2.6825,
"theoretical_loss": 3.5776613106065804,
"tokens_seen": 1236336640
},
{
"epoch": 0.37,
"learning_rate": 0.00031587763289869607,
"loss": 2.7114,
"theoretical_loss": 3.5776439640648907,
"tokens_seen": 1236402176
},
{
"epoch": 0.37,
"learning_rate": 0.0003158676028084253,
"loss": 2.4975,
"theoretical_loss": 3.5776266187000707,
"tokens_seen": 1236467712
},
{
"epoch": 0.37,
"learning_rate": 0.0003158575727181545,
"loss": 2.7708,
"theoretical_loss": 3.5776092745119783,
"tokens_seen": 1236533248
},
{
"epoch": 0.37,
"learning_rate": 0.00031584754262788367,
"loss": 2.8647,
"theoretical_loss": 3.5775919315004705,
"tokens_seen": 1236598784
},
{
"epoch": 0.37,
"learning_rate": 0.00031583751253761285,
"loss": 2.8909,
"theoretical_loss": 3.5775745896654056,
"tokens_seen": 1236664320
},
{
"epoch": 0.37,
"learning_rate": 0.00031582748244734203,
"loss": 2.7242,
"theoretical_loss": 3.5775572490066425,
"tokens_seen": 1236729856
},
{
"epoch": 0.37,
"learning_rate": 0.0003158174523570712,
"loss": 2.6398,
"theoretical_loss": 3.577539909524037,
"tokens_seen": 1236795392
},
{
"epoch": 0.37,
"learning_rate": 0.00031580742226680044,
"loss": 2.7851,
"theoretical_loss": 3.5775225712174485,
"tokens_seen": 1236860928
},
{
"epoch": 0.37,
"learning_rate": 0.00031579739217652957,
"loss": 2.7963,
"theoretical_loss": 3.5775052340867344,
"tokens_seen": 1236926464
},
{
"epoch": 0.37,
"objective/train/docs_used": 701918,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.480823040008545,
"objective/train/theoretical_loss": 3.577487898131753,
"objective/train/tokens_used": 1257452000,
"theoretical_loss": 3.577487898131753,
"tokens_seen": 1236992000
},
{
"epoch": 0.37,
"learning_rate": 0.0003157873620862588,
"loss": 2.7963,
"theoretical_loss": 3.577487898131753,
"tokens_seen": 1236992000
},
{
"epoch": 0.37,
"learning_rate": 0.00031577733199598793,
"loss": 2.5642,
"theoretical_loss": 3.577470563352362,
"tokens_seen": 1237057536
},
{
"epoch": 0.37,
"learning_rate": 0.00031576730190571717,
"loss": 2.696,
"theoretical_loss": 3.5774532297484196,
"tokens_seen": 1237123072
},
{
"epoch": 0.37,
"learning_rate": 0.00031575727181544635,
"loss": 2.6577,
"theoretical_loss": 3.5774358973197833,
"tokens_seen": 1237188608
},
{
"epoch": 0.37,
"learning_rate": 0.00031574724172517553,
"loss": 2.7681,
"theoretical_loss": 3.577418566066312,
"tokens_seen": 1237254144
},
{
"epoch": 0.37,
"learning_rate": 0.0003157372116349047,
"loss": 2.6433,
"theoretical_loss": 3.577401235987863,
"tokens_seen": 1237319680
},
{
"epoch": 0.37,
"learning_rate": 0.00031572718154463395,
"loss": 2.6367,
"theoretical_loss": 3.577383907084295,
"tokens_seen": 1237385216
},
{
"epoch": 0.37,
"learning_rate": 0.0003157171514543631,
"loss": 2.9513,
"theoretical_loss": 3.5773665793554654,
"tokens_seen": 1237450752
},
{
"epoch": 0.38,
"learning_rate": 0.0003157071213640923,
"loss": 2.4045,
"theoretical_loss": 3.577349252801233,
"tokens_seen": 1237516288
},
{
"epoch": 0.38,
"learning_rate": 0.00031569709127382144,
"loss": 2.7357,
"theoretical_loss": 3.577331927421456,
"tokens_seen": 1237581824
},
{
"epoch": 0.38,
"learning_rate": 0.00031568706118355067,
"loss": 2.7114,
"theoretical_loss": 3.5773146032159926,
"tokens_seen": 1237647360
},
{
"epoch": 0.38,
"learning_rate": 0.00031567703109327985,
"loss": 2.62,
"theoretical_loss": 3.5772972801847005,
"tokens_seen": 1237712896
},
{
"epoch": 0.38,
"learning_rate": 0.00031566700100300903,
"loss": 2.5568,
"theoretical_loss": 3.5772799583274386,
"tokens_seen": 1237778432
},
{
"epoch": 0.38,
"learning_rate": 0.0003156569709127382,
"loss": 2.6671,
"theoretical_loss": 3.577262637644065,
"tokens_seen": 1237843968
},
{
"epoch": 0.38,
"learning_rate": 0.0003156469408224674,
"loss": 2.9327,
"theoretical_loss": 3.5772453181344375,
"tokens_seen": 1237909504
},
{
"epoch": 0.38,
"learning_rate": 0.0003156369107321966,
"loss": 2.7319,
"theoretical_loss": 3.5772279997984144,
"tokens_seen": 1237975040
},
{
"epoch": 0.38,
"learning_rate": 0.0003156268806419258,
"loss": 2.6504,
"theoretical_loss": 3.5772106826358554,
"tokens_seen": 1238040576
},
{
"epoch": 0.38,
"learning_rate": 0.00031561685055165494,
"loss": 2.8083,
"theoretical_loss": 3.5771933666466174,
"tokens_seen": 1238106112
},
{
"epoch": 0.38,
"learning_rate": 0.0003156068204613842,
"loss": 2.6122,
"theoretical_loss": 3.5771760518305595,
"tokens_seen": 1238171648
},
{
"epoch": 0.38,
"learning_rate": 0.0003155967903711133,
"loss": 2.7898,
"theoretical_loss": 3.57715873818754,
"tokens_seen": 1238237184
},
{
"epoch": 0.38,
"learning_rate": 0.00031558676028084254,
"loss": 2.8134,
"theoretical_loss": 3.577141425717417,
"tokens_seen": 1238302720
},
{
"epoch": 0.38,
"learning_rate": 0.0003155767301905717,
"loss": 2.7531,
"theoretical_loss": 3.5771241144200494,
"tokens_seen": 1238368256
},
{
"epoch": 0.38,
"learning_rate": 0.0003155667001003009,
"loss": 2.5578,
"theoretical_loss": 3.577106804295296,
"tokens_seen": 1238433792
},
{
"epoch": 0.38,
"learning_rate": 0.0003155566700100301,
"loss": 2.7643,
"theoretical_loss": 3.5770894953430146,
"tokens_seen": 1238499328
},
{
"epoch": 0.38,
"learning_rate": 0.0003155466399197593,
"loss": 2.8641,
"theoretical_loss": 3.577072187563064,
"tokens_seen": 1238564864
},
{
"epoch": 0.38,
"objective/train/docs_used": 702757,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4627716541290283,
"objective/train/theoretical_loss": 3.577054880955303,
"objective/train/tokens_used": 1259090400,
"theoretical_loss": 3.577054880955303,
"tokens_seen": 1238630400
},
{
"epoch": 0.38,
"learning_rate": 0.00031553660982948844,
"loss": 2.625,
"theoretical_loss": 3.577054880955303,
"tokens_seen": 1238630400
},
{
"epoch": 0.38,
"learning_rate": 0.0003155265797392177,
"loss": 2.8843,
"theoretical_loss": 3.57703757551959,
"tokens_seen": 1238695936
},
{
"epoch": 0.38,
"learning_rate": 0.0003155165496489468,
"loss": 2.5492,
"theoretical_loss": 3.5770202712557833,
"tokens_seen": 1238761472
},
{
"epoch": 0.38,
"learning_rate": 0.00031550651955867604,
"loss": 2.7016,
"theoretical_loss": 3.5770029681637423,
"tokens_seen": 1238827008
},
{
"epoch": 0.38,
"learning_rate": 0.0003154964894684052,
"loss": 2.7995,
"theoretical_loss": 3.576985666243325,
"tokens_seen": 1238892544
},
{
"epoch": 0.38,
"learning_rate": 0.0003154864593781344,
"loss": 3.0331,
"theoretical_loss": 3.5769683654943907,
"tokens_seen": 1238958080
},
{
"epoch": 0.38,
"learning_rate": 0.0003154764292878636,
"loss": 2.6721,
"theoretical_loss": 3.5769510659167976,
"tokens_seen": 1239023616
},
{
"epoch": 0.38,
"learning_rate": 0.00031546639919759277,
"loss": 2.6853,
"theoretical_loss": 3.576933767510405,
"tokens_seen": 1239089152
},
{
"epoch": 0.38,
"learning_rate": 0.00031545636910732195,
"loss": 2.8014,
"theoretical_loss": 3.5769164702750706,
"tokens_seen": 1239154688
},
{
"epoch": 0.38,
"learning_rate": 0.0003154463390170512,
"loss": 2.7085,
"theoretical_loss": 3.576899174210655,
"tokens_seen": 1239220224
},
{
"epoch": 0.38,
"learning_rate": 0.0003154363089267803,
"loss": 2.7305,
"theoretical_loss": 3.576881879317015,
"tokens_seen": 1239285760
},
{
"epoch": 0.38,
"learning_rate": 0.00031542627883650954,
"loss": 2.8276,
"theoretical_loss": 3.57686458559401,
"tokens_seen": 1239351296
},
{
"epoch": 0.38,
"learning_rate": 0.0003154162487462387,
"loss": 2.5395,
"theoretical_loss": 3.5768472930415003,
"tokens_seen": 1239416832
},
{
"epoch": 0.38,
"learning_rate": 0.0003154062186559679,
"loss": 2.8954,
"theoretical_loss": 3.576830001659343,
"tokens_seen": 1239482368
},
{
"epoch": 0.38,
"learning_rate": 0.0003153961885656971,
"loss": 2.5172,
"theoretical_loss": 3.576812711447398,
"tokens_seen": 1239547904
},
{
"epoch": 0.38,
"learning_rate": 0.00031538615847542627,
"loss": 2.8914,
"theoretical_loss": 3.5767954224055236,
"tokens_seen": 1239613440
},
{
"epoch": 0.38,
"learning_rate": 0.00031537612838515545,
"loss": 2.7154,
"theoretical_loss": 3.57677813453358,
"tokens_seen": 1239678976
},
{
"epoch": 0.38,
"learning_rate": 0.0003153660982948847,
"loss": 2.8692,
"theoretical_loss": 3.576760847831425,
"tokens_seen": 1239744512
},
{
"epoch": 0.38,
"learning_rate": 0.0003153560682046138,
"loss": 2.6618,
"theoretical_loss": 3.576743562298917,
"tokens_seen": 1239810048
},
{
"epoch": 0.38,
"learning_rate": 0.00031534603811434305,
"loss": 2.469,
"theoretical_loss": 3.576726277935917,
"tokens_seen": 1239875584
},
{
"epoch": 0.38,
"learning_rate": 0.0003153360080240722,
"loss": 2.8153,
"theoretical_loss": 3.5767089947422828,
"tokens_seen": 1239941120
},
{
"epoch": 0.38,
"learning_rate": 0.0003153259779338014,
"loss": 2.7576,
"theoretical_loss": 3.5766917127178735,
"tokens_seen": 1240006656
},
{
"epoch": 0.38,
"learning_rate": 0.0003153159478435306,
"loss": 2.8019,
"theoretical_loss": 3.576674431862549,
"tokens_seen": 1240072192
},
{
"epoch": 0.38,
"learning_rate": 0.00031530591775325977,
"loss": 2.7301,
"theoretical_loss": 3.5766571521761676,
"tokens_seen": 1240137728
},
{
"epoch": 0.38,
"learning_rate": 0.000315295887662989,
"loss": 2.7119,
"theoretical_loss": 3.5766398736585883,
"tokens_seen": 1240203264
},
{
"epoch": 0.38,
"objective/train/docs_used": 703958,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8278799057006836,
"objective/train/theoretical_loss": 3.5766225963096714,
"objective/train/tokens_used": 1260728800,
"theoretical_loss": 3.5766225963096714,
"tokens_seen": 1240268800
},
{
"epoch": 0.38,
"learning_rate": 0.00031528585757271813,
"loss": 2.7822,
"theoretical_loss": 3.5766225963096714,
"tokens_seen": 1240268800
},
{
"epoch": 0.38,
"learning_rate": 0.00031527582748244737,
"loss": 2.7243,
"theoretical_loss": 3.576605320129275,
"tokens_seen": 1240334336
},
{
"epoch": 0.38,
"learning_rate": 0.00031526579739217655,
"loss": 2.8629,
"theoretical_loss": 3.576588045117259,
"tokens_seen": 1240399872
},
{
"epoch": 0.38,
"learning_rate": 0.00031525576730190573,
"loss": 2.7275,
"theoretical_loss": 3.576570771273482,
"tokens_seen": 1240465408
},
{
"epoch": 0.38,
"learning_rate": 0.0003152457372116349,
"loss": 2.8858,
"theoretical_loss": 3.5765534985978045,
"tokens_seen": 1240530944
},
{
"epoch": 0.38,
"learning_rate": 0.00031523570712136415,
"loss": 2.8537,
"theoretical_loss": 3.5765362270900845,
"tokens_seen": 1240596480
},
{
"epoch": 0.38,
"learning_rate": 0.0003152256770310933,
"loss": 2.8613,
"theoretical_loss": 3.576518956750182,
"tokens_seen": 1240662016
},
{
"epoch": 0.38,
"learning_rate": 0.0003152156469408225,
"loss": 2.6706,
"theoretical_loss": 3.576501687577956,
"tokens_seen": 1240727552
},
{
"epoch": 0.38,
"learning_rate": 0.00031520561685055164,
"loss": 2.9122,
"theoretical_loss": 3.5764844195732666,
"tokens_seen": 1240793088
},
{
"epoch": 0.38,
"learning_rate": 0.00031519558676028087,
"loss": 2.6755,
"theoretical_loss": 3.576467152735972,
"tokens_seen": 1240858624
},
{
"epoch": 0.38,
"learning_rate": 0.00031518555667001005,
"loss": 2.695,
"theoretical_loss": 3.5764498870659325,
"tokens_seen": 1240924160
},
{
"epoch": 0.38,
"learning_rate": 0.00031517552657973924,
"loss": 2.7172,
"theoretical_loss": 3.576432622563008,
"tokens_seen": 1240989696
},
{
"epoch": 0.38,
"learning_rate": 0.0003151654964894684,
"loss": 2.7416,
"theoretical_loss": 3.576415359227057,
"tokens_seen": 1241055232
},
{
"epoch": 0.38,
"learning_rate": 0.0003151554663991976,
"loss": 2.7809,
"theoretical_loss": 3.5763980970579388,
"tokens_seen": 1241120768
},
{
"epoch": 0.38,
"learning_rate": 0.0003151454363089268,
"loss": 2.7322,
"theoretical_loss": 3.576380836055514,
"tokens_seen": 1241186304
},
{
"epoch": 0.38,
"learning_rate": 0.000315135406218656,
"loss": 2.6492,
"theoretical_loss": 3.5763635762196415,
"tokens_seen": 1241251840
},
{
"epoch": 0.38,
"learning_rate": 0.00031512537612838514,
"loss": 2.8375,
"theoretical_loss": 3.576346317550181,
"tokens_seen": 1241317376
},
{
"epoch": 0.38,
"learning_rate": 0.0003151153460381144,
"loss": 3.04,
"theoretical_loss": 3.5763290600469926,
"tokens_seen": 1241382912
},
{
"epoch": 0.38,
"learning_rate": 0.0003151053159478435,
"loss": 2.3818,
"theoretical_loss": 3.5763118037099346,
"tokens_seen": 1241448448
},
{
"epoch": 0.38,
"learning_rate": 0.00031509528585757274,
"loss": 2.9598,
"theoretical_loss": 3.5762945485388675,
"tokens_seen": 1241513984
},
{
"epoch": 0.38,
"learning_rate": 0.0003150852557673019,
"loss": 2.7182,
"theoretical_loss": 3.576277294533652,
"tokens_seen": 1241579520
},
{
"epoch": 0.38,
"learning_rate": 0.0003150752256770311,
"loss": 2.9666,
"theoretical_loss": 3.5762600416941455,
"tokens_seen": 1241645056
},
{
"epoch": 0.38,
"learning_rate": 0.0003150651955867603,
"loss": 2.7804,
"theoretical_loss": 3.57624279002021,
"tokens_seen": 1241710592
},
{
"epoch": 0.38,
"learning_rate": 0.0003150551654964895,
"loss": 2.7183,
"theoretical_loss": 3.5762255395117037,
"tokens_seen": 1241776128
},
{
"epoch": 0.38,
"learning_rate": 0.00031504513540621864,
"loss": 2.749,
"theoretical_loss": 3.576208290168487,
"tokens_seen": 1241841664
},
{
"epoch": 0.38,
"objective/train/docs_used": 704662,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.979909896850586,
"objective/train/theoretical_loss": 3.5761910419904193,
"objective/train/tokens_used": 1262367200,
"theoretical_loss": 3.5761910419904193,
"tokens_seen": 1241907200
},
{
"epoch": 0.38,
"learning_rate": 0.0003150351053159479,
"loss": 2.8272,
"theoretical_loss": 3.5761910419904193,
"tokens_seen": 1241907200
},
{
"epoch": 0.38,
"learning_rate": 0.000315025075225677,
"loss": 2.5991,
"theoretical_loss": 3.576173794977361,
"tokens_seen": 1241972736
},
{
"epoch": 0.38,
"learning_rate": 0.00031501504513540624,
"loss": 2.7972,
"theoretical_loss": 3.5761565491291716,
"tokens_seen": 1242038272
},
{
"epoch": 0.38,
"learning_rate": 0.0003150050150451354,
"loss": 2.627,
"theoretical_loss": 3.576139304445711,
"tokens_seen": 1242103808
},
{
"epoch": 0.38,
"learning_rate": 0.0003149949849548646,
"loss": 2.7313,
"theoretical_loss": 3.5761220609268385,
"tokens_seen": 1242169344
},
{
"epoch": 0.38,
"learning_rate": 0.0003149849548645938,
"loss": 2.6177,
"theoretical_loss": 3.5761048185724156,
"tokens_seen": 1242234880
},
{
"epoch": 0.38,
"learning_rate": 0.00031497492477432297,
"loss": 2.7754,
"theoretical_loss": 3.5760875773823004,
"tokens_seen": 1242300416
},
{
"epoch": 0.38,
"learning_rate": 0.00031496489468405215,
"loss": 2.5347,
"theoretical_loss": 3.5760703373563545,
"tokens_seen": 1242365952
},
{
"epoch": 0.38,
"learning_rate": 0.0003149548645937814,
"loss": 2.7643,
"theoretical_loss": 3.5760530984944365,
"tokens_seen": 1242431488
},
{
"epoch": 0.38,
"learning_rate": 0.0003149448345035105,
"loss": 2.9243,
"theoretical_loss": 3.5760358607964076,
"tokens_seen": 1242497024
},
{
"epoch": 0.38,
"learning_rate": 0.00031493480441323974,
"loss": 2.5989,
"theoretical_loss": 3.576018624262127,
"tokens_seen": 1242562560
},
{
"epoch": 0.38,
"learning_rate": 0.0003149247743229689,
"loss": 2.5527,
"theoretical_loss": 3.5760013888914544,
"tokens_seen": 1242628096
},
{
"epoch": 0.38,
"learning_rate": 0.0003149147442326981,
"loss": 2.7858,
"theoretical_loss": 3.5759841546842512,
"tokens_seen": 1242693632
},
{
"epoch": 0.38,
"learning_rate": 0.0003149047141424273,
"loss": 2.7876,
"theoretical_loss": 3.575966921640376,
"tokens_seen": 1242759168
},
{
"epoch": 0.38,
"learning_rate": 0.00031489468405215647,
"loss": 2.5762,
"theoretical_loss": 3.5759496897596907,
"tokens_seen": 1242824704
},
{
"epoch": 0.38,
"learning_rate": 0.00031488465396188565,
"loss": 2.6986,
"theoretical_loss": 3.575932459042054,
"tokens_seen": 1242890240
},
{
"epoch": 0.38,
"learning_rate": 0.0003148746238716149,
"loss": 2.8653,
"theoretical_loss": 3.5759152294873267,
"tokens_seen": 1242955776
},
{
"epoch": 0.38,
"learning_rate": 0.000314864593781344,
"loss": 2.712,
"theoretical_loss": 3.5758980010953687,
"tokens_seen": 1243021312
},
{
"epoch": 0.38,
"learning_rate": 0.00031485456369107325,
"loss": 2.7925,
"theoretical_loss": 3.5758807738660403,
"tokens_seen": 1243086848
},
{
"epoch": 0.38,
"learning_rate": 0.0003148445336008024,
"loss": 2.6113,
"theoretical_loss": 3.5758635477992025,
"tokens_seen": 1243152384
},
{
"epoch": 0.38,
"learning_rate": 0.0003148345035105316,
"loss": 2.8295,
"theoretical_loss": 3.575846322894715,
"tokens_seen": 1243217920
},
{
"epoch": 0.38,
"learning_rate": 0.0003148244734202608,
"loss": 2.6815,
"theoretical_loss": 3.575829099152437,
"tokens_seen": 1243283456
},
{
"epoch": 0.38,
"learning_rate": 0.00031481444332998997,
"loss": 2.7788,
"theoretical_loss": 3.5758118765722307,
"tokens_seen": 1243348992
},
{
"epoch": 0.38,
"learning_rate": 0.00031480441323971915,
"loss": 2.8384,
"theoretical_loss": 3.5757946551539552,
"tokens_seen": 1243414528
},
{
"epoch": 0.38,
"learning_rate": 0.00031479438314944833,
"loss": 2.6062,
"theoretical_loss": 3.5757774348974714,
"tokens_seen": 1243480064
},
{
"epoch": 0.38,
"objective/train/docs_used": 705833,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.820856809616089,
"objective/train/theoretical_loss": 3.5757602158026396,
"objective/train/tokens_used": 1264005600,
"theoretical_loss": 3.5757602158026396,
"tokens_seen": 1243545600
},
{
"epoch": 0.38,
"learning_rate": 0.0003147843530591775,
"loss": 2.8557,
"theoretical_loss": 3.5757602158026396,
"tokens_seen": 1243545600
},
{
"epoch": 0.38,
"learning_rate": 0.00031477432296890675,
"loss": 2.6229,
"theoretical_loss": 3.5757429978693205,
"tokens_seen": 1243611136
},
{
"epoch": 0.38,
"learning_rate": 0.0003147642928786359,
"loss": 2.6138,
"theoretical_loss": 3.5757257810973737,
"tokens_seen": 1243676672
},
{
"epoch": 0.38,
"learning_rate": 0.0003147542627883651,
"loss": 2.6794,
"theoretical_loss": 3.5757085654866607,
"tokens_seen": 1243742208
},
{
"epoch": 0.38,
"learning_rate": 0.0003147442326980943,
"loss": 2.6147,
"theoretical_loss": 3.575691351037041,
"tokens_seen": 1243807744
},
{
"epoch": 0.38,
"learning_rate": 0.0003147342026078235,
"loss": 2.736,
"theoretical_loss": 3.5756741377483765,
"tokens_seen": 1243873280
},
{
"epoch": 0.38,
"learning_rate": 0.00031472417251755266,
"loss": 2.63,
"theoretical_loss": 3.5756569256205264,
"tokens_seen": 1243938816
},
{
"epoch": 0.38,
"learning_rate": 0.00031471414242728184,
"loss": 2.7831,
"theoretical_loss": 3.5756397146533514,
"tokens_seen": 1244004352
},
{
"epoch": 0.38,
"learning_rate": 0.000314704112337011,
"loss": 2.6247,
"theoretical_loss": 3.575622504846713,
"tokens_seen": 1244069888
},
{
"epoch": 0.38,
"learning_rate": 0.00031469408224674025,
"loss": 2.7036,
"theoretical_loss": 3.5756052962004707,
"tokens_seen": 1244135424
},
{
"epoch": 0.38,
"learning_rate": 0.0003146840521564694,
"loss": 2.4536,
"theoretical_loss": 3.575588088714486,
"tokens_seen": 1244200960
},
{
"epoch": 0.38,
"learning_rate": 0.0003146740220661986,
"loss": 2.5715,
"theoretical_loss": 3.57557088238862,
"tokens_seen": 1244266496
},
{
"epoch": 0.38,
"learning_rate": 0.00031466399197592774,
"loss": 2.7476,
"theoretical_loss": 3.5755536772227314,
"tokens_seen": 1244332032
},
{
"epoch": 0.38,
"learning_rate": 0.000314653961885657,
"loss": 2.6566,
"theoretical_loss": 3.5755364732166828,
"tokens_seen": 1244397568
},
{
"epoch": 0.38,
"learning_rate": 0.00031464393179538616,
"loss": 2.7241,
"theoretical_loss": 3.5755192703703345,
"tokens_seen": 1244463104
},
{
"epoch": 0.38,
"learning_rate": 0.00031463390170511534,
"loss": 2.7724,
"theoretical_loss": 3.5755020686835466,
"tokens_seen": 1244528640
},
{
"epoch": 0.38,
"learning_rate": 0.0003146238716148445,
"loss": 2.8264,
"theoretical_loss": 3.5754848681561806,
"tokens_seen": 1244594176
},
{
"epoch": 0.38,
"learning_rate": 0.0003146138415245737,
"loss": 2.5038,
"theoretical_loss": 3.575467668788097,
"tokens_seen": 1244659712
},
{
"epoch": 0.38,
"learning_rate": 0.0003146038114343029,
"loss": 2.7311,
"theoretical_loss": 3.575450470579157,
"tokens_seen": 1244725248
},
{
"epoch": 0.38,
"learning_rate": 0.0003145937813440321,
"loss": 2.8227,
"theoretical_loss": 3.575433273529221,
"tokens_seen": 1244790784
},
{
"epoch": 0.38,
"learning_rate": 0.00031458375125376125,
"loss": 2.6541,
"theoretical_loss": 3.5754160776381494,
"tokens_seen": 1244856320
},
{
"epoch": 0.38,
"learning_rate": 0.0003145737211634905,
"loss": 2.8847,
"theoretical_loss": 3.5753988829058043,
"tokens_seen": 1244921856
},
{
"epoch": 0.38,
"learning_rate": 0.0003145636910732197,
"loss": 2.7964,
"theoretical_loss": 3.575381689332046,
"tokens_seen": 1244987392
},
{
"epoch": 0.38,
"learning_rate": 0.00031455366098294884,
"loss": 2.6054,
"theoretical_loss": 3.5753644969167353,
"tokens_seen": 1245052928
},
{
"epoch": 0.38,
"learning_rate": 0.0003145436308926781,
"loss": 2.5047,
"theoretical_loss": 3.575347305659734,
"tokens_seen": 1245118464
},
{
"epoch": 0.38,
"objective/train/docs_used": 706471,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1011083126068115,
"objective/train/theoretical_loss": 3.5753301155609014,
"objective/train/tokens_used": 1265644000,
"theoretical_loss": 3.5753301155609014,
"tokens_seen": 1245184000
},
{
"epoch": 0.38,
"learning_rate": 0.0003145336008024072,
"loss": 2.753,
"theoretical_loss": 3.5753301155609014,
"tokens_seen": 1245184000
},
{
"epoch": 0.38,
"learning_rate": 0.00031452357071213644,
"loss": 2.8365,
"theoretical_loss": 3.5753129266201005,
"tokens_seen": 1245249536
},
{
"epoch": 0.38,
"learning_rate": 0.0003145135406218656,
"loss": 2.8212,
"theoretical_loss": 3.575295738837191,
"tokens_seen": 1245315072
},
{
"epoch": 0.38,
"learning_rate": 0.0003145035105315948,
"loss": 2.621,
"theoretical_loss": 3.5752785522120347,
"tokens_seen": 1245380608
},
{
"epoch": 0.38,
"learning_rate": 0.000314493480441324,
"loss": 2.6573,
"theoretical_loss": 3.5752613667444924,
"tokens_seen": 1245446144
},
{
"epoch": 0.38,
"learning_rate": 0.00031448345035105317,
"loss": 2.7172,
"theoretical_loss": 3.5752441824344254,
"tokens_seen": 1245511680
},
{
"epoch": 0.38,
"learning_rate": 0.00031447342026078235,
"loss": 2.6991,
"theoretical_loss": 3.5752269992816945,
"tokens_seen": 1245577216
},
{
"epoch": 0.38,
"learning_rate": 0.0003144633901705116,
"loss": 2.4365,
"theoretical_loss": 3.5752098172861606,
"tokens_seen": 1245642752
},
{
"epoch": 0.38,
"learning_rate": 0.0003144533600802407,
"loss": 2.7597,
"theoretical_loss": 3.575192636447686,
"tokens_seen": 1245708288
},
{
"epoch": 0.38,
"learning_rate": 0.00031444332998996994,
"loss": 2.7879,
"theoretical_loss": 3.5751754567661314,
"tokens_seen": 1245773824
},
{
"epoch": 0.38,
"learning_rate": 0.0003144332998996991,
"loss": 2.7083,
"theoretical_loss": 3.575158278241357,
"tokens_seen": 1245839360
},
{
"epoch": 0.38,
"learning_rate": 0.0003144232698094283,
"loss": 2.4857,
"theoretical_loss": 3.5751411008732257,
"tokens_seen": 1245904896
},
{
"epoch": 0.38,
"learning_rate": 0.0003144132397191575,
"loss": 2.6923,
"theoretical_loss": 3.575123924661598,
"tokens_seen": 1245970432
},
{
"epoch": 0.38,
"learning_rate": 0.00031440320962888667,
"loss": 2.8867,
"theoretical_loss": 3.575106749606335,
"tokens_seen": 1246035968
},
{
"epoch": 0.38,
"learning_rate": 0.00031439317953861585,
"loss": 2.6329,
"theoretical_loss": 3.575089575707299,
"tokens_seen": 1246101504
},
{
"epoch": 0.38,
"learning_rate": 0.0003143831494483451,
"loss": 3.0614,
"theoretical_loss": 3.5750724029643495,
"tokens_seen": 1246167040
},
{
"epoch": 0.38,
"learning_rate": 0.0003143731193580742,
"loss": 2.6759,
"theoretical_loss": 3.57505523137735,
"tokens_seen": 1246232576
},
{
"epoch": 0.38,
"learning_rate": 0.00031436308926780345,
"loss": 2.6356,
"theoretical_loss": 3.5750380609461607,
"tokens_seen": 1246298112
},
{
"epoch": 0.38,
"learning_rate": 0.0003143530591775326,
"loss": 2.6982,
"theoretical_loss": 3.5750208916706434,
"tokens_seen": 1246363648
},
{
"epoch": 0.38,
"learning_rate": 0.0003143430290872618,
"loss": 2.7405,
"theoretical_loss": 3.575003723550659,
"tokens_seen": 1246429184
},
{
"epoch": 0.38,
"learning_rate": 0.000314332998996991,
"loss": 2.7048,
"theoretical_loss": 3.5749865565860697,
"tokens_seen": 1246494720
},
{
"epoch": 0.38,
"learning_rate": 0.00031432296890672017,
"loss": 2.7758,
"theoretical_loss": 3.5749693907767366,
"tokens_seen": 1246560256
},
{
"epoch": 0.38,
"learning_rate": 0.00031431293881644935,
"loss": 2.6691,
"theoretical_loss": 3.5749522261225217,
"tokens_seen": 1246625792
},
{
"epoch": 0.38,
"learning_rate": 0.00031430290872617853,
"loss": 2.8084,
"theoretical_loss": 3.574935062623286,
"tokens_seen": 1246691328
},
{
"epoch": 0.38,
"learning_rate": 0.0003142928786359077,
"loss": 2.8783,
"theoretical_loss": 3.5749179002788916,
"tokens_seen": 1246756864
},
{
"epoch": 0.38,
"objective/train/docs_used": 707914,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1946613788604736,
"objective/train/theoretical_loss": 3.574900739089199,
"objective/train/tokens_used": 1267282400,
"theoretical_loss": 3.574900739089199,
"tokens_seen": 1246822400
},
{
"epoch": 0.38,
"learning_rate": 0.00031428284854563695,
"loss": 2.6586,
"theoretical_loss": 3.574900739089199,
"tokens_seen": 1246822400
},
{
"epoch": 0.38,
"learning_rate": 0.0003142728184553661,
"loss": 2.6369,
"theoretical_loss": 3.574883579054071,
"tokens_seen": 1246887936
},
{
"epoch": 0.38,
"learning_rate": 0.0003142627883650953,
"loss": 2.9502,
"theoretical_loss": 3.574866420173369,
"tokens_seen": 1246953472
},
{
"epoch": 0.38,
"learning_rate": 0.0003142527582748245,
"loss": 2.63,
"theoretical_loss": 3.5748492624469543,
"tokens_seen": 1247019008
},
{
"epoch": 0.38,
"learning_rate": 0.0003142427281845537,
"loss": 2.7253,
"theoretical_loss": 3.5748321058746892,
"tokens_seen": 1247084544
},
{
"epoch": 0.38,
"learning_rate": 0.00031423269809428286,
"loss": 2.7969,
"theoretical_loss": 3.5748149504564344,
"tokens_seen": 1247150080
},
{
"epoch": 0.38,
"learning_rate": 0.00031422266800401204,
"loss": 2.8245,
"theoretical_loss": 3.574797796192053,
"tokens_seen": 1247215616
},
{
"epoch": 0.38,
"learning_rate": 0.0003142126379137412,
"loss": 2.6331,
"theoretical_loss": 3.5747806430814055,
"tokens_seen": 1247281152
},
{
"epoch": 0.38,
"learning_rate": 0.00031420260782347045,
"loss": 2.7339,
"theoretical_loss": 3.5747634911243544,
"tokens_seen": 1247346688
},
{
"epoch": 0.38,
"learning_rate": 0.0003141925777331996,
"loss": 2.8285,
"theoretical_loss": 3.574746340320761,
"tokens_seen": 1247412224
},
{
"epoch": 0.38,
"learning_rate": 0.0003141825476429288,
"loss": 2.5232,
"theoretical_loss": 3.574729190670488,
"tokens_seen": 1247477760
},
{
"epoch": 0.38,
"learning_rate": 0.00031417251755265794,
"loss": 2.5242,
"theoretical_loss": 3.574712042173396,
"tokens_seen": 1247543296
},
{
"epoch": 0.38,
"learning_rate": 0.0003141624874623872,
"loss": 2.6886,
"theoretical_loss": 3.5746948948293484,
"tokens_seen": 1247608832
},
{
"epoch": 0.38,
"learning_rate": 0.00031415245737211636,
"loss": 2.815,
"theoretical_loss": 3.5746777486382064,
"tokens_seen": 1247674368
},
{
"epoch": 0.38,
"learning_rate": 0.00031414242728184554,
"loss": 2.6398,
"theoretical_loss": 3.574660603599831,
"tokens_seen": 1247739904
},
{
"epoch": 0.38,
"learning_rate": 0.0003141323971915747,
"loss": 2.7142,
"theoretical_loss": 3.574643459714086,
"tokens_seen": 1247805440
},
{
"epoch": 0.38,
"learning_rate": 0.0003141223671013039,
"loss": 2.8863,
"theoretical_loss": 3.574626316980832,
"tokens_seen": 1247870976
},
{
"epoch": 0.38,
"learning_rate": 0.0003141123370110331,
"loss": 2.668,
"theoretical_loss": 3.574609175399931,
"tokens_seen": 1247936512
},
{
"epoch": 0.38,
"learning_rate": 0.0003141023069207623,
"loss": 2.6066,
"theoretical_loss": 3.5745920349712463,
"tokens_seen": 1248002048
},
{
"epoch": 0.38,
"learning_rate": 0.00031409227683049145,
"loss": 2.6791,
"theoretical_loss": 3.574574895694638,
"tokens_seen": 1248067584
},
{
"epoch": 0.38,
"learning_rate": 0.0003140822467402207,
"loss": 2.7663,
"theoretical_loss": 3.57455775756997,
"tokens_seen": 1248133120
},
{
"epoch": 0.38,
"learning_rate": 0.00031407221664994986,
"loss": 2.6538,
"theoretical_loss": 3.5745406205971033,
"tokens_seen": 1248198656
},
{
"epoch": 0.38,
"learning_rate": 0.00031406218655967904,
"loss": 2.5664,
"theoretical_loss": 3.5745234847759004,
"tokens_seen": 1248264192
},
{
"epoch": 0.38,
"learning_rate": 0.0003140521564694082,
"loss": 2.9206,
"theoretical_loss": 3.5745063501062235,
"tokens_seen": 1248329728
},
{
"epoch": 0.38,
"learning_rate": 0.0003140421263791374,
"loss": 2.782,
"theoretical_loss": 3.574489216587935,
"tokens_seen": 1248395264
},
{
"epoch": 0.38,
"objective/train/docs_used": 708590,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7215492725372314,
"objective/train/theoretical_loss": 3.574472084220896,
"objective/train/tokens_used": 1268920800,
"theoretical_loss": 3.574472084220896,
"tokens_seen": 1248460800
},
{
"epoch": 0.38,
"learning_rate": 0.0003140320962888666,
"loss": 2.5714,
"theoretical_loss": 3.574472084220896,
"tokens_seen": 1248460800
},
{
"epoch": 0.38,
"learning_rate": 0.0003140220661985958,
"loss": 2.9589,
"theoretical_loss": 3.5744549530049703,
"tokens_seen": 1248526336
},
{
"epoch": 0.38,
"learning_rate": 0.00031401203610832495,
"loss": 2.5998,
"theoretical_loss": 3.574437822940019,
"tokens_seen": 1248591872
},
{
"epoch": 0.38,
"learning_rate": 0.0003140020060180542,
"loss": 2.9073,
"theoretical_loss": 3.574420694025905,
"tokens_seen": 1248657408
},
{
"epoch": 0.38,
"learning_rate": 0.0003139919759277833,
"loss": 2.6512,
"theoretical_loss": 3.5744035662624896,
"tokens_seen": 1248722944
},
{
"epoch": 0.38,
"learning_rate": 0.00031398194583751255,
"loss": 2.6902,
"theoretical_loss": 3.5743864396496363,
"tokens_seen": 1248788480
},
{
"epoch": 0.38,
"learning_rate": 0.00031397191574724173,
"loss": 2.6186,
"theoretical_loss": 3.574369314187207,
"tokens_seen": 1248854016
},
{
"epoch": 0.38,
"learning_rate": 0.0003139618856569709,
"loss": 2.6886,
"theoretical_loss": 3.5743521898750643,
"tokens_seen": 1248919552
},
{
"epoch": 0.38,
"learning_rate": 0.0003139518555667001,
"loss": 2.8654,
"theoretical_loss": 3.5743350667130698,
"tokens_seen": 1248985088
},
{
"epoch": 0.38,
"learning_rate": 0.0003139418254764293,
"loss": 2.6405,
"theoretical_loss": 3.574317944701086,
"tokens_seen": 1249050624
},
{
"epoch": 0.38,
"learning_rate": 0.00031393179538615845,
"loss": 2.8042,
"theoretical_loss": 3.5743008238389766,
"tokens_seen": 1249116160
},
{
"epoch": 0.38,
"learning_rate": 0.0003139217652958877,
"loss": 2.5316,
"theoretical_loss": 3.5742837041266027,
"tokens_seen": 1249181696
},
{
"epoch": 0.38,
"learning_rate": 0.0003139117352056168,
"loss": 2.6635,
"theoretical_loss": 3.5742665855638274,
"tokens_seen": 1249247232
},
{
"epoch": 0.38,
"learning_rate": 0.00031390170511534605,
"loss": 2.8068,
"theoretical_loss": 3.5742494681505126,
"tokens_seen": 1249312768
},
{
"epoch": 0.38,
"learning_rate": 0.00031389167502507523,
"loss": 2.6061,
"theoretical_loss": 3.574232351886522,
"tokens_seen": 1249378304
},
{
"epoch": 0.38,
"learning_rate": 0.0003138816449348044,
"loss": 2.7094,
"theoretical_loss": 3.5742152367717166,
"tokens_seen": 1249443840
},
{
"epoch": 0.38,
"learning_rate": 0.0003138716148445336,
"loss": 2.9424,
"theoretical_loss": 3.5741981228059605,
"tokens_seen": 1249509376
},
{
"epoch": 0.38,
"learning_rate": 0.0003138615847542628,
"loss": 2.82,
"theoretical_loss": 3.5741810099891156,
"tokens_seen": 1249574912
},
{
"epoch": 0.38,
"learning_rate": 0.00031385155466399196,
"loss": 2.6973,
"theoretical_loss": 3.5741638983210438,
"tokens_seen": 1249640448
},
{
"epoch": 0.38,
"learning_rate": 0.0003138415245737212,
"loss": 2.7727,
"theoretical_loss": 3.574146787801609,
"tokens_seen": 1249705984
},
{
"epoch": 0.38,
"learning_rate": 0.0003138314944834503,
"loss": 2.7466,
"theoretical_loss": 3.5741296784306735,
"tokens_seen": 1249771520
},
{
"epoch": 0.38,
"learning_rate": 0.00031382146439317955,
"loss": 2.6041,
"theoretical_loss": 3.5741125702081,
"tokens_seen": 1249837056
},
{
"epoch": 0.38,
"learning_rate": 0.00031381143430290873,
"loss": 2.7379,
"theoretical_loss": 3.57409546313375,
"tokens_seen": 1249902592
},
{
"epoch": 0.38,
"learning_rate": 0.0003138014042126379,
"loss": 2.6596,
"theoretical_loss": 3.5740783572074877,
"tokens_seen": 1249968128
},
{
"epoch": 0.38,
"learning_rate": 0.00031379137412236715,
"loss": 2.6012,
"theoretical_loss": 3.5740612524291757,
"tokens_seen": 1250033664
},
{
"epoch": 0.38,
"objective/train/docs_used": 709757,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5650100708007812,
"objective/train/theoretical_loss": 3.5740441487986767,
"objective/train/tokens_used": 1270559200,
"theoretical_loss": 3.5740441487986767,
"tokens_seen": 1250099200
},
{
"epoch": 0.38,
"learning_rate": 0.0003137813440320963,
"loss": 2.6186,
"theoretical_loss": 3.5740441487986767,
"tokens_seen": 1250099200
},
{
"epoch": 0.38,
"learning_rate": 0.0003137713139418255,
"loss": 2.837,
"theoretical_loss": 3.574027046315853,
"tokens_seen": 1250164736
},
{
"epoch": 0.38,
"learning_rate": 0.0003137612838515547,
"loss": 2.8225,
"theoretical_loss": 3.574009944980568,
"tokens_seen": 1250230272
},
{
"epoch": 0.38,
"learning_rate": 0.0003137512537612839,
"loss": 2.7208,
"theoretical_loss": 3.573992844792684,
"tokens_seen": 1250295808
},
{
"epoch": 0.38,
"learning_rate": 0.00031374122367101306,
"loss": 2.8093,
"theoretical_loss": 3.5739757457520644,
"tokens_seen": 1250361344
},
{
"epoch": 0.38,
"learning_rate": 0.00031373119358074224,
"loss": 2.6792,
"theoretical_loss": 3.5739586478585714,
"tokens_seen": 1250426880
},
{
"epoch": 0.38,
"learning_rate": 0.0003137211634904714,
"loss": 2.5133,
"theoretical_loss": 3.5739415511120693,
"tokens_seen": 1250492416
},
{
"epoch": 0.38,
"learning_rate": 0.00031371113340020065,
"loss": 2.939,
"theoretical_loss": 3.573924455512419,
"tokens_seen": 1250557952
},
{
"epoch": 0.38,
"learning_rate": 0.0003137011033099298,
"loss": 2.9356,
"theoretical_loss": 3.5739073610594856,
"tokens_seen": 1250623488
},
{
"epoch": 0.38,
"learning_rate": 0.000313691073219659,
"loss": 2.7497,
"theoretical_loss": 3.5738902677531312,
"tokens_seen": 1250689024
},
{
"epoch": 0.38,
"learning_rate": 0.00031368104312938814,
"loss": 2.55,
"theoretical_loss": 3.573873175593218,
"tokens_seen": 1250754560
},
{
"epoch": 0.38,
"learning_rate": 0.0003136710130391174,
"loss": 2.7093,
"theoretical_loss": 3.5738560845796106,
"tokens_seen": 1250820096
},
{
"epoch": 0.38,
"learning_rate": 0.00031366098294884656,
"loss": 2.7528,
"theoretical_loss": 3.5738389947121707,
"tokens_seen": 1250885632
},
{
"epoch": 0.38,
"learning_rate": 0.00031365095285857574,
"loss": 2.8143,
"theoretical_loss": 3.573821905990762,
"tokens_seen": 1250951168
},
{
"epoch": 0.38,
"learning_rate": 0.0003136409227683049,
"loss": 2.7563,
"theoretical_loss": 3.573804818415248,
"tokens_seen": 1251016704
},
{
"epoch": 0.38,
"learning_rate": 0.0003136308926780341,
"loss": 2.547,
"theoretical_loss": 3.573787731985491,
"tokens_seen": 1251082240
},
{
"epoch": 0.38,
"learning_rate": 0.0003136208625877633,
"loss": 2.837,
"theoretical_loss": 3.5737706467013552,
"tokens_seen": 1251147776
},
{
"epoch": 0.38,
"learning_rate": 0.0003136108324974925,
"loss": 2.7738,
"theoretical_loss": 3.5737535625627026,
"tokens_seen": 1251213312
},
{
"epoch": 0.38,
"learning_rate": 0.00031360080240722165,
"loss": 2.8921,
"theoretical_loss": 3.573736479569397,
"tokens_seen": 1251278848
},
{
"epoch": 0.38,
"learning_rate": 0.0003135907723169509,
"loss": 2.6497,
"theoretical_loss": 3.5737193977213018,
"tokens_seen": 1251344384
},
{
"epoch": 0.38,
"learning_rate": 0.00031358074222668006,
"loss": 2.672,
"theoretical_loss": 3.57370231701828,
"tokens_seen": 1251409920
},
{
"epoch": 0.38,
"learning_rate": 0.00031357071213640924,
"loss": 2.56,
"theoretical_loss": 3.573685237460195,
"tokens_seen": 1251475456
},
{
"epoch": 0.38,
"learning_rate": 0.0003135606820461384,
"loss": 2.9082,
"theoretical_loss": 3.5736681590469104,
"tokens_seen": 1251540992
},
{
"epoch": 0.38,
"learning_rate": 0.0003135506519558676,
"loss": 2.6807,
"theoretical_loss": 3.5736510817782885,
"tokens_seen": 1251606528
},
{
"epoch": 0.38,
"learning_rate": 0.0003135406218655968,
"loss": 2.6469,
"theoretical_loss": 3.5736340056541933,
"tokens_seen": 1251672064
},
{
"epoch": 0.38,
"objective/train/docs_used": 710221,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.267810821533203,
"objective/train/theoretical_loss": 3.5736169306744885,
"objective/train/tokens_used": 1272197600,
"theoretical_loss": 3.5736169306744885,
"tokens_seen": 1251737600
},
{
"epoch": 0.38,
"learning_rate": 0.000313530591775326,
"loss": 2.6684,
"theoretical_loss": 3.5736169306744885,
"tokens_seen": 1251737600
},
{
"epoch": 0.38,
"learning_rate": 0.00031352056168505515,
"loss": 2.6632,
"theoretical_loss": 3.5735998568390372,
"tokens_seen": 1251803136
},
{
"epoch": 0.38,
"learning_rate": 0.0003135105315947844,
"loss": 2.5829,
"theoretical_loss": 3.573582784147703,
"tokens_seen": 1251868672
},
{
"epoch": 0.38,
"learning_rate": 0.0003135005015045135,
"loss": 2.6931,
"theoretical_loss": 3.573565712600349,
"tokens_seen": 1251934208
},
{
"epoch": 0.38,
"learning_rate": 0.00031349047141424275,
"loss": 3.0187,
"theoretical_loss": 3.5735486421968385,
"tokens_seen": 1251999744
},
{
"epoch": 0.38,
"learning_rate": 0.00031348044132397193,
"loss": 2.4757,
"theoretical_loss": 3.5735315729370356,
"tokens_seen": 1252065280
},
{
"epoch": 0.38,
"learning_rate": 0.0003134704112337011,
"loss": 3.0258,
"theoretical_loss": 3.5735145048208032,
"tokens_seen": 1252130816
},
{
"epoch": 0.38,
"learning_rate": 0.0003134603811434303,
"loss": 2.819,
"theoretical_loss": 3.5734974378480056,
"tokens_seen": 1252196352
},
{
"epoch": 0.38,
"learning_rate": 0.0003134503510531595,
"loss": 2.8586,
"theoretical_loss": 3.5734803720185058,
"tokens_seen": 1252261888
},
{
"epoch": 0.38,
"learning_rate": 0.00031344032096288865,
"loss": 2.5257,
"theoretical_loss": 3.573463307332168,
"tokens_seen": 1252327424
},
{
"epoch": 0.38,
"learning_rate": 0.0003134302908726179,
"loss": 2.7047,
"theoretical_loss": 3.5734462437888546,
"tokens_seen": 1252392960
},
{
"epoch": 0.38,
"learning_rate": 0.000313420260782347,
"loss": 2.6683,
"theoretical_loss": 3.57342918138843,
"tokens_seen": 1252458496
},
{
"epoch": 0.38,
"learning_rate": 0.00031341023069207625,
"loss": 2.6904,
"theoretical_loss": 3.573412120130758,
"tokens_seen": 1252524032
},
{
"epoch": 0.38,
"learning_rate": 0.00031340020060180543,
"loss": 2.8445,
"theoretical_loss": 3.5733950600157023,
"tokens_seen": 1252589568
},
{
"epoch": 0.38,
"learning_rate": 0.0003133901705115346,
"loss": 2.6077,
"theoretical_loss": 3.573378001043126,
"tokens_seen": 1252655104
},
{
"epoch": 0.38,
"learning_rate": 0.0003133801404212638,
"loss": 2.6902,
"theoretical_loss": 3.5733609432128937,
"tokens_seen": 1252720640
},
{
"epoch": 0.38,
"learning_rate": 0.000313370110330993,
"loss": 2.8203,
"theoretical_loss": 3.573343886524868,
"tokens_seen": 1252786176
},
{
"epoch": 0.38,
"learning_rate": 0.00031336008024072216,
"loss": 2.8622,
"theoretical_loss": 3.5733268309789135,
"tokens_seen": 1252851712
},
{
"epoch": 0.38,
"learning_rate": 0.0003133500501504514,
"loss": 2.6238,
"theoretical_loss": 3.5733097765748942,
"tokens_seen": 1252917248
},
{
"epoch": 0.38,
"learning_rate": 0.0003133400200601805,
"loss": 2.8019,
"theoretical_loss": 3.5732927233126732,
"tokens_seen": 1252982784
},
{
"epoch": 0.38,
"learning_rate": 0.00031332998996990975,
"loss": 2.5795,
"theoretical_loss": 3.573275671192115,
"tokens_seen": 1253048320
},
{
"epoch": 0.38,
"learning_rate": 0.0003133199598796389,
"loss": 2.7717,
"theoretical_loss": 3.5732586202130827,
"tokens_seen": 1253113856
},
{
"epoch": 0.38,
"learning_rate": 0.0003133099297893681,
"loss": 2.6299,
"theoretical_loss": 3.5732415703754405,
"tokens_seen": 1253179392
},
{
"epoch": 0.38,
"learning_rate": 0.0003132998996990973,
"loss": 2.7553,
"theoretical_loss": 3.573224521679053,
"tokens_seen": 1253244928
},
{
"epoch": 0.38,
"learning_rate": 0.0003132898696088265,
"loss": 2.6711,
"theoretical_loss": 3.5732074741237834,
"tokens_seen": 1253310464
},
{
"epoch": 0.38,
"objective/train/docs_used": 710893,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5935490131378174,
"objective/train/theoretical_loss": 3.5731904277094957,
"objective/train/tokens_used": 1273836000,
"theoretical_loss": 3.5731904277094957,
"tokens_seen": 1253376000
},
{
"epoch": 0.38,
"learning_rate": 0.00031327983951855566,
"loss": 2.6374,
"theoretical_loss": 3.5731904277094957,
"tokens_seen": 1253376000
},
{
"epoch": 0.38,
"learning_rate": 0.0003132698094282849,
"loss": 2.5807,
"theoretical_loss": 3.573173382436054,
"tokens_seen": 1253441536
},
{
"epoch": 0.38,
"learning_rate": 0.000313259779338014,
"loss": 2.8188,
"theoretical_loss": 3.5731563383033227,
"tokens_seen": 1253507072
},
{
"epoch": 0.38,
"learning_rate": 0.00031324974924774326,
"loss": 2.9162,
"theoretical_loss": 3.573139295311165,
"tokens_seen": 1253572608
},
{
"epoch": 0.38,
"learning_rate": 0.0003132397191574724,
"loss": 2.6715,
"theoretical_loss": 3.573122253459446,
"tokens_seen": 1253638144
},
{
"epoch": 0.38,
"learning_rate": 0.0003132296890672016,
"loss": 2.5742,
"theoretical_loss": 3.5731052127480285,
"tokens_seen": 1253703680
},
{
"epoch": 0.38,
"learning_rate": 0.0003132196589769308,
"loss": 2.782,
"theoretical_loss": 3.573088173176778,
"tokens_seen": 1253769216
},
{
"epoch": 0.38,
"learning_rate": 0.00031320962888666,
"loss": 2.8723,
"theoretical_loss": 3.5730711347455575,
"tokens_seen": 1253834752
},
{
"epoch": 0.38,
"learning_rate": 0.00031319959879638916,
"loss": 2.5261,
"theoretical_loss": 3.5730540974542313,
"tokens_seen": 1253900288
},
{
"epoch": 0.38,
"learning_rate": 0.00031318956870611834,
"loss": 2.4189,
"theoretical_loss": 3.573037061302664,
"tokens_seen": 1253965824
},
{
"epoch": 0.38,
"learning_rate": 0.0003131795386158475,
"loss": 2.516,
"theoretical_loss": 3.5730200262907195,
"tokens_seen": 1254031360
},
{
"epoch": 0.38,
"learning_rate": 0.00031316950852557676,
"loss": 2.7851,
"theoretical_loss": 3.5730029924182625,
"tokens_seen": 1254096896
},
{
"epoch": 0.38,
"learning_rate": 0.0003131594784353059,
"loss": 2.8999,
"theoretical_loss": 3.5729859596851563,
"tokens_seen": 1254162432
},
{
"epoch": 0.38,
"learning_rate": 0.0003131494483450351,
"loss": 2.809,
"theoretical_loss": 3.5729689280912664,
"tokens_seen": 1254227968
},
{
"epoch": 0.38,
"learning_rate": 0.00031313941825476425,
"loss": 2.5909,
"theoretical_loss": 3.5729518976364556,
"tokens_seen": 1254293504
},
{
"epoch": 0.38,
"learning_rate": 0.0003131293881644935,
"loss": 2.7303,
"theoretical_loss": 3.5729348683205897,
"tokens_seen": 1254359040
},
{
"epoch": 0.38,
"learning_rate": 0.00031311935807422267,
"loss": 2.8692,
"theoretical_loss": 3.572917840143532,
"tokens_seen": 1254424576
},
{
"epoch": 0.38,
"learning_rate": 0.00031310932798395185,
"loss": 2.8758,
"theoretical_loss": 3.5729008131051474,
"tokens_seen": 1254490112
},
{
"epoch": 0.38,
"learning_rate": 0.00031309929789368103,
"loss": 2.806,
"theoretical_loss": 3.5728837872053,
"tokens_seen": 1254555648
},
{
"epoch": 0.38,
"learning_rate": 0.00031308926780341026,
"loss": 2.6893,
"theoretical_loss": 3.572866762443854,
"tokens_seen": 1254621184
},
{
"epoch": 0.38,
"learning_rate": 0.0003130792377131394,
"loss": 2.654,
"theoretical_loss": 3.5728497388206746,
"tokens_seen": 1254686720
},
{
"epoch": 0.38,
"learning_rate": 0.0003130692076228686,
"loss": 2.5998,
"theoretical_loss": 3.5728327163356255,
"tokens_seen": 1254752256
},
{
"epoch": 0.38,
"learning_rate": 0.0003130591775325978,
"loss": 3.0761,
"theoretical_loss": 3.572815694988572,
"tokens_seen": 1254817792
},
{
"epoch": 0.38,
"learning_rate": 0.000313049147442327,
"loss": 2.989,
"theoretical_loss": 3.572798674779377,
"tokens_seen": 1254883328
},
{
"epoch": 0.38,
"learning_rate": 0.0003130391173520562,
"loss": 2.8244,
"theoretical_loss": 3.5727816557079066,
"tokens_seen": 1254948864
},
{
"epoch": 0.38,
"objective/train/docs_used": 712142,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.188223123550415,
"objective/train/theoretical_loss": 3.572764637774024,
"objective/train/tokens_used": 1275474400,
"theoretical_loss": 3.572764637774024,
"tokens_seen": 1255014400
},
{
"epoch": 0.38,
"learning_rate": 0.00031302908726178535,
"loss": 2.5956,
"theoretical_loss": 3.572764637774024,
"tokens_seen": 1255014400
},
{
"epoch": 0.38,
"learning_rate": 0.0003130190571715146,
"loss": 2.8055,
"theoretical_loss": 3.5727476209775952,
"tokens_seen": 1255079936
},
{
"epoch": 0.38,
"learning_rate": 0.0003130090270812437,
"loss": 2.9003,
"theoretical_loss": 3.572730605318484,
"tokens_seen": 1255145472
},
{
"epoch": 0.38,
"learning_rate": 0.00031299899699097295,
"loss": 2.9744,
"theoretical_loss": 3.572713590796555,
"tokens_seen": 1255211008
},
{
"epoch": 0.38,
"learning_rate": 0.00031298896690070213,
"loss": 2.8192,
"theoretical_loss": 3.5726965774116737,
"tokens_seen": 1255276544
},
{
"epoch": 0.38,
"learning_rate": 0.0003129789368104313,
"loss": 2.7114,
"theoretical_loss": 3.5726795651637033,
"tokens_seen": 1255342080
},
{
"epoch": 0.38,
"learning_rate": 0.0003129689067201605,
"loss": 2.7261,
"theoretical_loss": 3.572662554052509,
"tokens_seen": 1255407616
},
{
"epoch": 0.38,
"learning_rate": 0.0003129588766298897,
"loss": 2.8431,
"theoretical_loss": 3.5726455440779565,
"tokens_seen": 1255473152
},
{
"epoch": 0.38,
"learning_rate": 0.00031294884653961885,
"loss": 2.7116,
"theoretical_loss": 3.572628535239909,
"tokens_seen": 1255538688
},
{
"epoch": 0.38,
"learning_rate": 0.0003129388164493481,
"loss": 2.9527,
"theoretical_loss": 3.572611527538232,
"tokens_seen": 1255604224
},
{
"epoch": 0.38,
"learning_rate": 0.0003129287863590772,
"loss": 2.4099,
"theoretical_loss": 3.5725945209727907,
"tokens_seen": 1255669760
},
{
"epoch": 0.38,
"learning_rate": 0.00031291875626880645,
"loss": 2.7246,
"theoretical_loss": 3.5725775155434496,
"tokens_seen": 1255735296
},
{
"epoch": 0.38,
"learning_rate": 0.00031290872617853563,
"loss": 2.864,
"theoretical_loss": 3.5725605112500727,
"tokens_seen": 1255800832
},
{
"epoch": 0.38,
"learning_rate": 0.0003128986960882648,
"loss": 2.6202,
"theoretical_loss": 3.572543508092526,
"tokens_seen": 1255866368
},
{
"epoch": 0.38,
"learning_rate": 0.000312888665997994,
"loss": 2.8477,
"theoretical_loss": 3.5725265060706732,
"tokens_seen": 1255931904
},
{
"epoch": 0.38,
"learning_rate": 0.0003128786359077232,
"loss": 2.535,
"theoretical_loss": 3.5725095051843803,
"tokens_seen": 1255997440
},
{
"epoch": 0.38,
"learning_rate": 0.00031286860581745236,
"loss": 2.7614,
"theoretical_loss": 3.5724925054335115,
"tokens_seen": 1256062976
},
{
"epoch": 0.38,
"learning_rate": 0.0003128585757271816,
"loss": 2.7501,
"theoretical_loss": 3.5724755068179324,
"tokens_seen": 1256128512
},
{
"epoch": 0.38,
"learning_rate": 0.0003128485456369107,
"loss": 2.6084,
"theoretical_loss": 3.5724585093375074,
"tokens_seen": 1256194048
},
{
"epoch": 0.38,
"learning_rate": 0.00031283851554663995,
"loss": 2.8664,
"theoretical_loss": 3.5724415129921017,
"tokens_seen": 1256259584
},
{
"epoch": 0.38,
"learning_rate": 0.0003128284854563691,
"loss": 2.5761,
"theoretical_loss": 3.5724245177815797,
"tokens_seen": 1256325120
},
{
"epoch": 0.38,
"learning_rate": 0.0003128184553660983,
"loss": 2.4724,
"theoretical_loss": 3.5724075237058077,
"tokens_seen": 1256390656
},
{
"epoch": 0.38,
"learning_rate": 0.0003128084252758275,
"loss": 2.9419,
"theoretical_loss": 3.5723905307646495,
"tokens_seen": 1256456192
},
{
"epoch": 0.38,
"learning_rate": 0.0003127983951855567,
"loss": 2.6834,
"theoretical_loss": 3.5723735389579705,
"tokens_seen": 1256521728
},
{
"epoch": 0.38,
"learning_rate": 0.00031278836509528586,
"loss": 2.9014,
"theoretical_loss": 3.5723565482856365,
"tokens_seen": 1256587264
},
{
"epoch": 0.38,
"objective/train/docs_used": 712613,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0056488513946533,
"objective/train/theoretical_loss": 3.5723395587475117,
"objective/train/tokens_used": 1277112800,
"theoretical_loss": 3.5723395587475117,
"tokens_seen": 1256652800
},
{
"epoch": 0.38,
"learning_rate": 0.0003127783350050151,
"loss": 2.7558,
"theoretical_loss": 3.5723395587475117,
"tokens_seen": 1256652800
},
{
"epoch": 0.38,
"learning_rate": 0.0003127683049147442,
"loss": 2.7145,
"theoretical_loss": 3.572322570343462,
"tokens_seen": 1256718336
},
{
"epoch": 0.38,
"learning_rate": 0.00031275827482447346,
"loss": 2.6827,
"theoretical_loss": 3.572305583073352,
"tokens_seen": 1256783872
},
{
"epoch": 0.38,
"learning_rate": 0.0003127482447342026,
"loss": 2.716,
"theoretical_loss": 3.5722885969370473,
"tokens_seen": 1256849408
},
{
"epoch": 0.38,
"learning_rate": 0.0003127382146439318,
"loss": 2.699,
"theoretical_loss": 3.5722716119344122,
"tokens_seen": 1256914944
},
{
"epoch": 0.38,
"learning_rate": 0.000312728184553661,
"loss": 2.5228,
"theoretical_loss": 3.5722546280653136,
"tokens_seen": 1256980480
},
{
"epoch": 0.38,
"learning_rate": 0.0003127181544633902,
"loss": 2.649,
"theoretical_loss": 3.572237645329615,
"tokens_seen": 1257046016
},
{
"epoch": 0.38,
"learning_rate": 0.00031270812437311936,
"loss": 2.7383,
"theoretical_loss": 3.572220663727183,
"tokens_seen": 1257111552
},
{
"epoch": 0.38,
"learning_rate": 0.00031269809428284854,
"loss": 2.5695,
"theoretical_loss": 3.5722036832578823,
"tokens_seen": 1257177088
},
{
"epoch": 0.38,
"learning_rate": 0.0003126880641925777,
"loss": 2.668,
"theoretical_loss": 3.572186703921578,
"tokens_seen": 1257242624
},
{
"epoch": 0.38,
"learning_rate": 0.00031267803410230696,
"loss": 2.7599,
"theoretical_loss": 3.572169725718136,
"tokens_seen": 1257308160
},
{
"epoch": 0.38,
"learning_rate": 0.0003126680040120361,
"loss": 2.779,
"theoretical_loss": 3.5721527486474214,
"tokens_seen": 1257373696
},
{
"epoch": 0.38,
"learning_rate": 0.0003126579739217653,
"loss": 2.7647,
"theoretical_loss": 3.5721357727092995,
"tokens_seen": 1257439232
},
{
"epoch": 0.38,
"learning_rate": 0.00031264794383149445,
"loss": 2.7534,
"theoretical_loss": 3.5721187979036353,
"tokens_seen": 1257504768
},
{
"epoch": 0.38,
"learning_rate": 0.0003126379137412237,
"loss": 2.8479,
"theoretical_loss": 3.5721018242302955,
"tokens_seen": 1257570304
},
{
"epoch": 0.38,
"learning_rate": 0.00031262788365095287,
"loss": 2.6903,
"theoretical_loss": 3.572084851689145,
"tokens_seen": 1257635840
},
{
"epoch": 0.38,
"learning_rate": 0.00031261785356068205,
"loss": 2.8858,
"theoretical_loss": 3.5720678802800485,
"tokens_seen": 1257701376
},
{
"epoch": 0.38,
"learning_rate": 0.00031260782347041123,
"loss": 2.6162,
"theoretical_loss": 3.572050910002872,
"tokens_seen": 1257766912
},
{
"epoch": 0.38,
"learning_rate": 0.00031259779338014046,
"loss": 2.6587,
"theoretical_loss": 3.572033940857482,
"tokens_seen": 1257832448
},
{
"epoch": 0.38,
"learning_rate": 0.0003125877632898696,
"loss": 2.6784,
"theoretical_loss": 3.572016972843743,
"tokens_seen": 1257897984
},
{
"epoch": 0.38,
"learning_rate": 0.0003125777331995988,
"loss": 2.7368,
"theoretical_loss": 3.5720000059615202,
"tokens_seen": 1257963520
},
{
"epoch": 0.38,
"learning_rate": 0.00031256770310932795,
"loss": 2.8653,
"theoretical_loss": 3.5719830402106805,
"tokens_seen": 1258029056
},
{
"epoch": 0.38,
"learning_rate": 0.0003125576730190572,
"loss": 2.6969,
"theoretical_loss": 3.5719660755910883,
"tokens_seen": 1258094592
},
{
"epoch": 0.38,
"learning_rate": 0.00031254764292878637,
"loss": 2.7406,
"theoretical_loss": 3.57194911210261,
"tokens_seen": 1258160128
},
{
"epoch": 0.38,
"learning_rate": 0.00031253761283851555,
"loss": 2.9451,
"theoretical_loss": 3.571932149745111,
"tokens_seen": 1258225664
},
{
"epoch": 0.38,
"objective/train/docs_used": 713934,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.485351800918579,
"objective/train/theoretical_loss": 3.571915188518457,
"objective/train/tokens_used": 1278751200,
"theoretical_loss": 3.571915188518457,
"tokens_seen": 1258291200
},
{
"epoch": 0.38,
"learning_rate": 0.00031252758274824473,
"loss": 2.6269,
"theoretical_loss": 3.571915188518457,
"tokens_seen": 1258291200
},
{
"epoch": 0.38,
"learning_rate": 0.0003125175526579739,
"loss": 2.8277,
"theoretical_loss": 3.571898228422514,
"tokens_seen": 1258356736
},
{
"epoch": 0.38,
"learning_rate": 0.0003125075225677031,
"loss": 2.7538,
"theoretical_loss": 3.571881269457147,
"tokens_seen": 1258422272
},
{
"epoch": 0.38,
"learning_rate": 0.00031249749247743233,
"loss": 2.6033,
"theoretical_loss": 3.571864311622223,
"tokens_seen": 1258487808
},
{
"epoch": 0.38,
"learning_rate": 0.00031248746238716146,
"loss": 2.7369,
"theoretical_loss": 3.5718473549176064,
"tokens_seen": 1258553344
},
{
"epoch": 0.38,
"learning_rate": 0.0003124774322968907,
"loss": 2.8659,
"theoretical_loss": 3.5718303993431637,
"tokens_seen": 1258618880
},
{
"epoch": 0.38,
"learning_rate": 0.0003124674022066198,
"loss": 2.8519,
"theoretical_loss": 3.571813444898761,
"tokens_seen": 1258684416
},
{
"epoch": 0.38,
"learning_rate": 0.00031245737211634905,
"loss": 2.9317,
"theoretical_loss": 3.5717964915842635,
"tokens_seen": 1258749952
},
{
"epoch": 0.38,
"learning_rate": 0.00031244734202607823,
"loss": 2.744,
"theoretical_loss": 3.5717795393995377,
"tokens_seen": 1258815488
},
{
"epoch": 0.38,
"learning_rate": 0.0003124373119358074,
"loss": 2.9309,
"theoretical_loss": 3.5717625883444493,
"tokens_seen": 1258881024
},
{
"epoch": 0.38,
"learning_rate": 0.0003124272818455366,
"loss": 2.7585,
"theoretical_loss": 3.5717456384188635,
"tokens_seen": 1258946560
},
{
"epoch": 0.38,
"learning_rate": 0.00031241725175526583,
"loss": 2.8885,
"theoretical_loss": 3.5717286896226472,
"tokens_seen": 1259012096
},
{
"epoch": 0.38,
"learning_rate": 0.00031240722166499496,
"loss": 2.7084,
"theoretical_loss": 3.5717117419556663,
"tokens_seen": 1259077632
},
{
"epoch": 0.38,
"learning_rate": 0.0003123971915747242,
"loss": 2.5253,
"theoretical_loss": 3.5716947954177867,
"tokens_seen": 1259143168
},
{
"epoch": 0.38,
"learning_rate": 0.0003123871614844533,
"loss": 2.7139,
"theoretical_loss": 3.5716778500088737,
"tokens_seen": 1259208704
},
{
"epoch": 0.38,
"learning_rate": 0.00031237713139418256,
"loss": 2.8604,
"theoretical_loss": 3.5716609057287942,
"tokens_seen": 1259274240
},
{
"epoch": 0.38,
"learning_rate": 0.00031236710130391174,
"loss": 2.5723,
"theoretical_loss": 3.571643962577414,
"tokens_seen": 1259339776
},
{
"epoch": 0.38,
"learning_rate": 0.0003123570712136409,
"loss": 2.7314,
"theoretical_loss": 3.5716270205545992,
"tokens_seen": 1259405312
},
{
"epoch": 0.38,
"learning_rate": 0.0003123470411233701,
"loss": 2.9133,
"theoretical_loss": 3.571610079660215,
"tokens_seen": 1259470848
},
{
"epoch": 0.38,
"learning_rate": 0.0003123370110330993,
"loss": 2.7556,
"theoretical_loss": 3.571593139894129,
"tokens_seen": 1259536384
},
{
"epoch": 0.38,
"learning_rate": 0.00031232698094282846,
"loss": 2.9155,
"theoretical_loss": 3.571576201256207,
"tokens_seen": 1259601920
},
{
"epoch": 0.38,
"learning_rate": 0.0003123169508525577,
"loss": 2.7953,
"theoretical_loss": 3.571559263746315,
"tokens_seen": 1259667456
},
{
"epoch": 0.38,
"learning_rate": 0.0003123069207622869,
"loss": 2.8103,
"theoretical_loss": 3.5715423273643188,
"tokens_seen": 1259732992
},
{
"epoch": 0.38,
"learning_rate": 0.00031229689067201606,
"loss": 2.8602,
"theoretical_loss": 3.5715253921100847,
"tokens_seen": 1259798528
},
{
"epoch": 0.38,
"learning_rate": 0.0003122868605817453,
"loss": 2.6346,
"theoretical_loss": 3.5715084579834793,
"tokens_seen": 1259864064
},
{
"epoch": 0.38,
"objective/train/docs_used": 714588,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.705703020095825,
"objective/train/theoretical_loss": 3.571491524984369,
"objective/train/tokens_used": 1280389600,
"theoretical_loss": 3.571491524984369,
"tokens_seen": 1259929600
},
{
"epoch": 0.38,
"learning_rate": 0.0003122768304914744,
"loss": 2.836,
"theoretical_loss": 3.571491524984369,
"tokens_seen": 1259929600
},
{
"epoch": 0.38,
"learning_rate": 0.00031226680040120366,
"loss": 2.8379,
"theoretical_loss": 3.5714745931126193,
"tokens_seen": 1259995136
},
{
"epoch": 0.38,
"learning_rate": 0.0003122567703109328,
"loss": 2.9507,
"theoretical_loss": 3.5714576623680974,
"tokens_seen": 1260060672
},
{
"epoch": 0.38,
"learning_rate": 0.000312246740220662,
"loss": 2.5956,
"theoretical_loss": 3.571440732750669,
"tokens_seen": 1260126208
},
{
"epoch": 0.38,
"learning_rate": 0.0003122367101303912,
"loss": 2.7284,
"theoretical_loss": 3.571423804260201,
"tokens_seen": 1260191744
},
{
"epoch": 0.38,
"learning_rate": 0.0003122266800401204,
"loss": 2.7424,
"theoretical_loss": 3.5714068768965594,
"tokens_seen": 1260257280
},
{
"epoch": 0.38,
"learning_rate": 0.00031221664994984956,
"loss": 2.7198,
"theoretical_loss": 3.5713899506596105,
"tokens_seen": 1260322816
},
{
"epoch": 0.38,
"learning_rate": 0.00031220661985957874,
"loss": 2.4531,
"theoretical_loss": 3.571373025549221,
"tokens_seen": 1260388352
},
{
"epoch": 0.38,
"learning_rate": 0.0003121965897693079,
"loss": 2.7087,
"theoretical_loss": 3.5713561015652573,
"tokens_seen": 1260453888
},
{
"epoch": 0.38,
"learning_rate": 0.00031218655967903716,
"loss": 2.7467,
"theoretical_loss": 3.5713391787075857,
"tokens_seen": 1260519424
},
{
"epoch": 0.38,
"learning_rate": 0.0003121765295887663,
"loss": 2.7961,
"theoretical_loss": 3.571322256976073,
"tokens_seen": 1260584960
},
{
"epoch": 0.38,
"learning_rate": 0.0003121664994984955,
"loss": 2.5677,
"theoretical_loss": 3.571305336370586,
"tokens_seen": 1260650496
},
{
"epoch": 0.38,
"learning_rate": 0.00031215646940822465,
"loss": 2.5382,
"theoretical_loss": 3.57128841689099,
"tokens_seen": 1260716032
},
{
"epoch": 0.38,
"learning_rate": 0.0003121464393179539,
"loss": 2.9111,
"theoretical_loss": 3.5712714985371528,
"tokens_seen": 1260781568
},
{
"epoch": 0.38,
"learning_rate": 0.00031213640922768307,
"loss": 2.8048,
"theoretical_loss": 3.5712545813089402,
"tokens_seen": 1260847104
},
{
"epoch": 0.38,
"learning_rate": 0.00031212637913741225,
"loss": 2.8088,
"theoretical_loss": 3.57123766520622,
"tokens_seen": 1260912640
},
{
"epoch": 0.38,
"learning_rate": 0.00031211634904714143,
"loss": 2.552,
"theoretical_loss": 3.571220750228857,
"tokens_seen": 1260978176
},
{
"epoch": 0.38,
"learning_rate": 0.00031210631895687066,
"loss": 2.494,
"theoretical_loss": 3.571203836376719,
"tokens_seen": 1261043712
},
{
"epoch": 0.38,
"learning_rate": 0.0003120962888665998,
"loss": 2.8721,
"theoretical_loss": 3.5711869236496727,
"tokens_seen": 1261109248
},
{
"epoch": 0.38,
"learning_rate": 0.000312086258776329,
"loss": 2.6454,
"theoretical_loss": 3.5711700120475847,
"tokens_seen": 1261174784
},
{
"epoch": 0.38,
"learning_rate": 0.00031207622868605815,
"loss": 2.676,
"theoretical_loss": 3.5711531015703217,
"tokens_seen": 1261240320
},
{
"epoch": 0.38,
"learning_rate": 0.0003120661985957874,
"loss": 2.7527,
"theoretical_loss": 3.5711361922177502,
"tokens_seen": 1261305856
},
{
"epoch": 0.38,
"learning_rate": 0.00031205616850551657,
"loss": 2.9068,
"theoretical_loss": 3.571119283989737,
"tokens_seen": 1261371392
},
{
"epoch": 0.38,
"learning_rate": 0.00031204613841524575,
"loss": 2.7535,
"theoretical_loss": 3.5711023768861487,
"tokens_seen": 1261436928
},
{
"epoch": 0.38,
"learning_rate": 0.00031203610832497493,
"loss": 2.6763,
"theoretical_loss": 3.571085470906853,
"tokens_seen": 1261502464
},
{
"epoch": 0.38,
"objective/train/docs_used": 715954,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.431766986846924,
"objective/train/theoretical_loss": 3.571068566051716,
"objective/train/tokens_used": 1282028000,
"theoretical_loss": 3.571068566051716,
"tokens_seen": 1261568000
},
{
"epoch": 0.38,
"learning_rate": 0.0003120260782347041,
"loss": 2.7248,
"theoretical_loss": 3.571068566051716,
"tokens_seen": 1261568000
},
{
"epoch": 0.38,
"learning_rate": 0.0003120160481444333,
"loss": 2.6752,
"theoretical_loss": 3.571051662320605,
"tokens_seen": 1261633536
},
{
"epoch": 0.38,
"learning_rate": 0.00031200601805416253,
"loss": 2.8189,
"theoretical_loss": 3.5710347597133865,
"tokens_seen": 1261699072
},
{
"epoch": 0.38,
"learning_rate": 0.00031199598796389166,
"loss": 2.7487,
"theoretical_loss": 3.5710178582299275,
"tokens_seen": 1261764608
},
{
"epoch": 0.38,
"learning_rate": 0.0003119859578736209,
"loss": 2.5613,
"theoretical_loss": 3.571000957870094,
"tokens_seen": 1261830144
},
{
"epoch": 0.38,
"learning_rate": 0.00031197592778335,
"loss": 2.5675,
"theoretical_loss": 3.570984058633755,
"tokens_seen": 1261895680
},
{
"epoch": 0.38,
"learning_rate": 0.00031196589769307925,
"loss": 2.8396,
"theoretical_loss": 3.5709671605207762,
"tokens_seen": 1261961216
},
{
"epoch": 0.38,
"learning_rate": 0.00031195586760280843,
"loss": 2.7539,
"theoretical_loss": 3.5709502635310244,
"tokens_seen": 1262026752
},
{
"epoch": 0.38,
"learning_rate": 0.0003119458375125376,
"loss": 2.941,
"theoretical_loss": 3.5709333676643675,
"tokens_seen": 1262092288
},
{
"epoch": 0.38,
"learning_rate": 0.0003119358074222668,
"loss": 2.7996,
"theoretical_loss": 3.570916472920671,
"tokens_seen": 1262157824
},
{
"epoch": 0.38,
"learning_rate": 0.00031192577733199603,
"loss": 2.522,
"theoretical_loss": 3.570899579299804,
"tokens_seen": 1262223360
},
{
"epoch": 0.38,
"learning_rate": 0.00031191574724172516,
"loss": 2.7553,
"theoretical_loss": 3.570882686801632,
"tokens_seen": 1262288896
},
{
"epoch": 0.38,
"learning_rate": 0.0003119057171514544,
"loss": 2.7409,
"theoretical_loss": 3.570865795426023,
"tokens_seen": 1262354432
},
{
"epoch": 0.38,
"learning_rate": 0.0003118956870611835,
"loss": 2.6033,
"theoretical_loss": 3.570848905172843,
"tokens_seen": 1262419968
},
{
"epoch": 0.38,
"learning_rate": 0.00031188565697091276,
"loss": 2.617,
"theoretical_loss": 3.570832016041961,
"tokens_seen": 1262485504
},
{
"epoch": 0.38,
"learning_rate": 0.00031187562688064194,
"loss": 2.4,
"theoretical_loss": 3.5708151280332423,
"tokens_seen": 1262551040
},
{
"epoch": 0.38,
"learning_rate": 0.0003118655967903711,
"loss": 2.8599,
"theoretical_loss": 3.5707982411465555,
"tokens_seen": 1262616576
},
{
"epoch": 0.38,
"learning_rate": 0.0003118555667001003,
"loss": 2.6878,
"theoretical_loss": 3.570781355381767,
"tokens_seen": 1262682112
},
{
"epoch": 0.38,
"learning_rate": 0.0003118455366098295,
"loss": 2.6888,
"theoretical_loss": 3.570764470738744,
"tokens_seen": 1262747648
},
{
"epoch": 0.38,
"learning_rate": 0.00031183550651955866,
"loss": 2.7425,
"theoretical_loss": 3.5707475872173546,
"tokens_seen": 1262813184
},
{
"epoch": 0.38,
"learning_rate": 0.0003118254764292879,
"loss": 2.95,
"theoretical_loss": 3.570730704817465,
"tokens_seen": 1262878720
},
{
"epoch": 0.38,
"learning_rate": 0.000311815446339017,
"loss": 2.6572,
"theoretical_loss": 3.5707138235389433,
"tokens_seen": 1262944256
},
{
"epoch": 0.38,
"learning_rate": 0.00031180541624874626,
"loss": 2.4727,
"theoretical_loss": 3.570696943381656,
"tokens_seen": 1263009792
},
{
"epoch": 0.38,
"learning_rate": 0.0003117953861584754,
"loss": 2.583,
"theoretical_loss": 3.5706800643454715,
"tokens_seen": 1263075328
},
{
"epoch": 0.38,
"learning_rate": 0.0003117853560682046,
"loss": 2.6069,
"theoretical_loss": 3.5706631864302567,
"tokens_seen": 1263140864
},
{
"epoch": 0.38,
"objective/train/docs_used": 716702,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3566877841949463,
"objective/train/theoretical_loss": 3.5706463096358787,
"objective/train/tokens_used": 1283666400,
"theoretical_loss": 3.5706463096358787,
"tokens_seen": 1263206400
},
{
"epoch": 0.38,
"learning_rate": 0.0003117753259779338,
"loss": 2.7248,
"theoretical_loss": 3.5706463096358787,
"tokens_seen": 1263206400
},
{
"epoch": 0.38,
"learning_rate": 0.000311765295887663,
"loss": 2.5914,
"theoretical_loss": 3.570629433962206,
"tokens_seen": 1263271936
},
{
"epoch": 0.38,
"learning_rate": 0.00031175526579739217,
"loss": 2.5451,
"theoretical_loss": 3.5706125594091045,
"tokens_seen": 1263337472
},
{
"epoch": 0.38,
"learning_rate": 0.0003117452357071214,
"loss": 2.7604,
"theoretical_loss": 3.5705956859764427,
"tokens_seen": 1263403008
},
{
"epoch": 0.38,
"learning_rate": 0.00031173520561685053,
"loss": 2.7612,
"theoretical_loss": 3.5705788136640875,
"tokens_seen": 1263468544
},
{
"epoch": 0.38,
"learning_rate": 0.00031172517552657976,
"loss": 2.6592,
"theoretical_loss": 3.5705619424719073,
"tokens_seen": 1263534080
},
{
"epoch": 0.38,
"learning_rate": 0.0003117151454363089,
"loss": 2.7355,
"theoretical_loss": 3.5705450723997685,
"tokens_seen": 1263599616
},
{
"epoch": 0.38,
"learning_rate": 0.0003117051153460381,
"loss": 2.7262,
"theoretical_loss": 3.5705282034475396,
"tokens_seen": 1263665152
},
{
"epoch": 0.38,
"learning_rate": 0.0003116950852557673,
"loss": 2.6035,
"theoretical_loss": 3.5705113356150875,
"tokens_seen": 1263730688
},
{
"epoch": 0.38,
"learning_rate": 0.0003116850551654965,
"loss": 2.8007,
"theoretical_loss": 3.5704944689022806,
"tokens_seen": 1263796224
},
{
"epoch": 0.38,
"learning_rate": 0.00031167502507522567,
"loss": 2.5633,
"theoretical_loss": 3.570477603308986,
"tokens_seen": 1263861760
},
{
"epoch": 0.38,
"learning_rate": 0.00031166499498495485,
"loss": 2.692,
"theoretical_loss": 3.570460738835071,
"tokens_seen": 1263927296
},
{
"epoch": 0.38,
"learning_rate": 0.00031165496489468403,
"loss": 2.7517,
"theoretical_loss": 3.5704438754804038,
"tokens_seen": 1263992832
},
{
"epoch": 0.38,
"learning_rate": 0.00031164493480441327,
"loss": 2.6612,
"theoretical_loss": 3.570427013244852,
"tokens_seen": 1264058368
},
{
"epoch": 0.38,
"learning_rate": 0.0003116349047141424,
"loss": 2.6474,
"theoretical_loss": 3.570410152128283,
"tokens_seen": 1264123904
},
{
"epoch": 0.38,
"learning_rate": 0.00031162487462387163,
"loss": 2.6163,
"theoretical_loss": 3.570393292130565,
"tokens_seen": 1264189440
},
{
"epoch": 0.38,
"learning_rate": 0.0003116148445336008,
"loss": 2.6046,
"theoretical_loss": 3.5703764332515657,
"tokens_seen": 1264254976
},
{
"epoch": 0.38,
"learning_rate": 0.00031160481444333,
"loss": 2.7036,
"theoretical_loss": 3.5703595754911523,
"tokens_seen": 1264320512
},
{
"epoch": 0.38,
"learning_rate": 0.00031159478435305917,
"loss": 2.7901,
"theoretical_loss": 3.5703427188491936,
"tokens_seen": 1264386048
},
{
"epoch": 0.38,
"learning_rate": 0.00031158475426278835,
"loss": 2.5361,
"theoretical_loss": 3.570325863325557,
"tokens_seen": 1264451584
},
{
"epoch": 0.38,
"learning_rate": 0.00031157472417251753,
"loss": 2.8387,
"theoretical_loss": 3.5703090089201095,
"tokens_seen": 1264517120
},
{
"epoch": 0.38,
"learning_rate": 0.00031156469408224677,
"loss": 2.7913,
"theoretical_loss": 3.5702921556327203,
"tokens_seen": 1264582656
},
{
"epoch": 0.38,
"learning_rate": 0.00031155466399197595,
"loss": 2.7061,
"theoretical_loss": 3.5702753034632564,
"tokens_seen": 1264648192
},
{
"epoch": 0.38,
"learning_rate": 0.00031154463390170513,
"loss": 2.6524,
"theoretical_loss": 3.5702584524115863,
"tokens_seen": 1264713728
},
{
"epoch": 0.38,
"learning_rate": 0.0003115346038114343,
"loss": 2.7961,
"theoretical_loss": 3.5702416024775774,
"tokens_seen": 1264779264
},
{
"epoch": 0.38,
"objective/train/docs_used": 717941,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.782820701599121,
"objective/train/theoretical_loss": 3.5702247536610976,
"objective/train/tokens_used": 1285304800,
"theoretical_loss": 3.5702247536610976,
"tokens_seen": 1264844800
},
{
"epoch": 0.38,
"learning_rate": 0.0003115245737211635,
"loss": 2.6629,
"theoretical_loss": 3.5702247536610976,
"tokens_seen": 1264844800
},
{
"epoch": 0.38,
"learning_rate": 0.00031151454363089273,
"loss": 2.6638,
"theoretical_loss": 3.570207905962016,
"tokens_seen": 1264910336
},
{
"epoch": 0.38,
"learning_rate": 0.00031150451354062186,
"loss": 2.9176,
"theoretical_loss": 3.570191059380199,
"tokens_seen": 1264975872
},
{
"epoch": 0.38,
"learning_rate": 0.0003114944834503511,
"loss": 2.7268,
"theoretical_loss": 3.570174213915516,
"tokens_seen": 1265041408
},
{
"epoch": 0.38,
"learning_rate": 0.0003114844533600802,
"loss": 2.8269,
"theoretical_loss": 3.5701573695678346,
"tokens_seen": 1265106944
},
{
"epoch": 0.38,
"learning_rate": 0.00031147442326980945,
"loss": 2.6878,
"theoretical_loss": 3.5701405263370223,
"tokens_seen": 1265172480
},
{
"epoch": 0.38,
"learning_rate": 0.00031146439317953863,
"loss": 2.5203,
"theoretical_loss": 3.5701236842229482,
"tokens_seen": 1265238016
},
{
"epoch": 0.38,
"learning_rate": 0.0003114543630892678,
"loss": 2.6806,
"theoretical_loss": 3.5701068432254797,
"tokens_seen": 1265303552
},
{
"epoch": 0.38,
"learning_rate": 0.000311444332998997,
"loss": 2.9634,
"theoretical_loss": 3.570090003344485,
"tokens_seen": 1265369088
},
{
"epoch": 0.38,
"learning_rate": 0.00031143430290872623,
"loss": 2.8153,
"theoretical_loss": 3.5700731645798323,
"tokens_seen": 1265434624
},
{
"epoch": 0.38,
"learning_rate": 0.00031142427281845536,
"loss": 2.7336,
"theoretical_loss": 3.57005632693139,
"tokens_seen": 1265500160
},
{
"epoch": 0.38,
"learning_rate": 0.0003114142427281846,
"loss": 2.8735,
"theoretical_loss": 3.570039490399026,
"tokens_seen": 1265565696
},
{
"epoch": 0.38,
"learning_rate": 0.0003114042126379137,
"loss": 2.773,
"theoretical_loss": 3.5700226549826093,
"tokens_seen": 1265631232
},
{
"epoch": 0.38,
"learning_rate": 0.00031139418254764296,
"loss": 2.7537,
"theoretical_loss": 3.570005820682007,
"tokens_seen": 1265696768
},
{
"epoch": 0.38,
"learning_rate": 0.00031138415245737214,
"loss": 2.6522,
"theoretical_loss": 3.569988987497088,
"tokens_seen": 1265762304
},
{
"epoch": 0.38,
"learning_rate": 0.0003113741223671013,
"loss": 2.8977,
"theoretical_loss": 3.569972155427721,
"tokens_seen": 1265827840
},
{
"epoch": 0.38,
"learning_rate": 0.0003113640922768305,
"loss": 2.8844,
"theoretical_loss": 3.5699553244737734,
"tokens_seen": 1265893376
},
{
"epoch": 0.38,
"learning_rate": 0.0003113540621865597,
"loss": 2.8688,
"theoretical_loss": 3.5699384946351147,
"tokens_seen": 1265958912
},
{
"epoch": 0.38,
"learning_rate": 0.00031134403209628886,
"loss": 2.5489,
"theoretical_loss": 3.569921665911612,
"tokens_seen": 1266024448
},
{
"epoch": 0.38,
"learning_rate": 0.0003113340020060181,
"loss": 2.7754,
"theoretical_loss": 3.569904838303134,
"tokens_seen": 1266089984
},
{
"epoch": 0.38,
"learning_rate": 0.0003113239719157472,
"loss": 2.8463,
"theoretical_loss": 3.56988801180955,
"tokens_seen": 1266155520
},
{
"epoch": 0.38,
"learning_rate": 0.00031131394182547646,
"loss": 2.5223,
"theoretical_loss": 3.5698711864307278,
"tokens_seen": 1266221056
},
{
"epoch": 0.38,
"learning_rate": 0.0003113039117352056,
"loss": 2.7505,
"theoretical_loss": 3.5698543621665353,
"tokens_seen": 1266286592
},
{
"epoch": 0.38,
"learning_rate": 0.0003112938816449348,
"loss": 2.7874,
"theoretical_loss": 3.5698375390168424,
"tokens_seen": 1266352128
},
{
"epoch": 0.38,
"learning_rate": 0.000311283851554664,
"loss": 2.6072,
"theoretical_loss": 3.569820716981516,
"tokens_seen": 1266417664
},
{
"epoch": 0.38,
"objective/train/docs_used": 718481,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8580243587493896,
"objective/train/theoretical_loss": 3.5698038960604257,
"objective/train/tokens_used": 1286943200,
"theoretical_loss": 3.5698038960604257,
"tokens_seen": 1266483200
},
{
"epoch": 0.38,
"learning_rate": 0.0003112738214643932,
"loss": 2.7003,
"theoretical_loss": 3.5698038960604257,
"tokens_seen": 1266483200
},
{
"epoch": 0.38,
"learning_rate": 0.00031126379137412237,
"loss": 2.7578,
"theoretical_loss": 3.56978707625344,
"tokens_seen": 1266548736
},
{
"epoch": 0.38,
"learning_rate": 0.0003112537612838516,
"loss": 3.0204,
"theoretical_loss": 3.5697702575604264,
"tokens_seen": 1266614272
},
{
"epoch": 0.38,
"learning_rate": 0.00031124373119358073,
"loss": 2.8372,
"theoretical_loss": 3.5697534399812545,
"tokens_seen": 1266679808
},
{
"epoch": 0.38,
"learning_rate": 0.00031123370110330996,
"loss": 2.5878,
"theoretical_loss": 3.5697366235157935,
"tokens_seen": 1266745344
},
{
"epoch": 0.38,
"learning_rate": 0.0003112236710130391,
"loss": 2.9305,
"theoretical_loss": 3.56971980816391,
"tokens_seen": 1266810880
},
{
"epoch": 0.38,
"learning_rate": 0.0003112136409227683,
"loss": 2.6326,
"theoretical_loss": 3.5697029939254747,
"tokens_seen": 1266876416
},
{
"epoch": 0.38,
"learning_rate": 0.0003112036108324975,
"loss": 2.7421,
"theoretical_loss": 3.5696861808003555,
"tokens_seen": 1266941952
},
{
"epoch": 0.38,
"learning_rate": 0.0003111935807422267,
"loss": 2.7792,
"theoretical_loss": 3.5696693687884204,
"tokens_seen": 1267007488
},
{
"epoch": 0.38,
"learning_rate": 0.00031118355065195587,
"loss": 2.7509,
"theoretical_loss": 3.5696525578895395,
"tokens_seen": 1267073024
},
{
"epoch": 0.38,
"learning_rate": 0.00031117352056168505,
"loss": 2.8954,
"theoretical_loss": 3.5696357481035808,
"tokens_seen": 1267138560
},
{
"epoch": 0.38,
"learning_rate": 0.00031116349047141423,
"loss": 2.706,
"theoretical_loss": 3.5696189394304128,
"tokens_seen": 1267204096
},
{
"epoch": 0.38,
"learning_rate": 0.00031115346038114347,
"loss": 2.8657,
"theoretical_loss": 3.5696021318699045,
"tokens_seen": 1267269632
},
{
"epoch": 0.38,
"learning_rate": 0.0003111434302908726,
"loss": 2.8849,
"theoretical_loss": 3.569585325421925,
"tokens_seen": 1267335168
},
{
"epoch": 0.38,
"learning_rate": 0.00031113340020060183,
"loss": 2.5871,
"theoretical_loss": 3.5695685200863427,
"tokens_seen": 1267400704
},
{
"epoch": 0.38,
"learning_rate": 0.000311123370110331,
"loss": 2.686,
"theoretical_loss": 3.5695517158630268,
"tokens_seen": 1267466240
},
{
"epoch": 0.38,
"learning_rate": 0.0003111133400200602,
"loss": 2.9244,
"theoretical_loss": 3.569534912751846,
"tokens_seen": 1267531776
},
{
"epoch": 0.38,
"learning_rate": 0.00031110330992978937,
"loss": 2.7764,
"theoretical_loss": 3.5695181107526697,
"tokens_seen": 1267597312
},
{
"epoch": 0.38,
"learning_rate": 0.00031109327983951855,
"loss": 2.6351,
"theoretical_loss": 3.569501309865366,
"tokens_seen": 1267662848
},
{
"epoch": 0.38,
"learning_rate": 0.00031108324974924773,
"loss": 2.947,
"theoretical_loss": 3.5694845100898043,
"tokens_seen": 1267728384
},
{
"epoch": 0.38,
"learning_rate": 0.00031107321965897697,
"loss": 2.6625,
"theoretical_loss": 3.5694677114258533,
"tokens_seen": 1267793920
},
{
"epoch": 0.38,
"learning_rate": 0.0003110631895687061,
"loss": 2.7761,
"theoretical_loss": 3.569450913873382,
"tokens_seen": 1267859456
},
{
"epoch": 0.38,
"learning_rate": 0.00031105315947843533,
"loss": 2.7426,
"theoretical_loss": 3.5694341174322606,
"tokens_seen": 1267924992
},
{
"epoch": 0.38,
"learning_rate": 0.00031104312938816446,
"loss": 2.6185,
"theoretical_loss": 3.5694173221023564,
"tokens_seen": 1267990528
},
{
"epoch": 0.38,
"learning_rate": 0.0003110330992978937,
"loss": 2.5305,
"theoretical_loss": 3.5694005278835395,
"tokens_seen": 1268056064
},
{
"epoch": 0.38,
"objective/train/docs_used": 719716,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7100017070770264,
"objective/train/theoretical_loss": 3.5693837347756783,
"objective/train/tokens_used": 1288581600,
"theoretical_loss": 3.5693837347756783,
"tokens_seen": 1268121600
},
{
"epoch": 0.38,
"learning_rate": 0.0003110230692076229,
"loss": 2.5275,
"theoretical_loss": 3.5693837347756783,
"tokens_seen": 1268121600
},
{
"epoch": 0.38,
"learning_rate": 0.00031101303911735206,
"loss": 2.5816,
"theoretical_loss": 3.5693669427786423,
"tokens_seen": 1268187136
},
{
"epoch": 0.38,
"learning_rate": 0.00031100300902708124,
"loss": 2.7494,
"theoretical_loss": 3.5693501518923005,
"tokens_seen": 1268252672
},
{
"epoch": 0.38,
"learning_rate": 0.0003109929789368104,
"loss": 2.6569,
"theoretical_loss": 3.569333362116523,
"tokens_seen": 1268318208
},
{
"epoch": 0.38,
"learning_rate": 0.0003109829488465396,
"loss": 2.906,
"theoretical_loss": 3.5693165734511774,
"tokens_seen": 1268383744
},
{
"epoch": 0.38,
"learning_rate": 0.00031097291875626883,
"loss": 2.8283,
"theoretical_loss": 3.569299785896134,
"tokens_seen": 1268449280
},
{
"epoch": 0.38,
"learning_rate": 0.00031096288866599796,
"loss": 2.5976,
"theoretical_loss": 3.5692829994512607,
"tokens_seen": 1268514816
},
{
"epoch": 0.38,
"learning_rate": 0.0003109528585757272,
"loss": 2.7735,
"theoretical_loss": 3.5692662141164284,
"tokens_seen": 1268580352
},
{
"epoch": 0.38,
"learning_rate": 0.0003109428284854564,
"loss": 2.7579,
"theoretical_loss": 3.5692494298915056,
"tokens_seen": 1268645888
},
{
"epoch": 0.38,
"learning_rate": 0.00031093279839518556,
"loss": 2.651,
"theoretical_loss": 3.5692326467763618,
"tokens_seen": 1268711424
},
{
"epoch": 0.38,
"learning_rate": 0.00031092276830491474,
"loss": 2.6555,
"theoretical_loss": 3.5692158647708654,
"tokens_seen": 1268776960
},
{
"epoch": 0.38,
"learning_rate": 0.0003109127382146439,
"loss": 2.7182,
"theoretical_loss": 3.569199083874887,
"tokens_seen": 1268842496
},
{
"epoch": 0.38,
"learning_rate": 0.0003109027081243731,
"loss": 2.7742,
"theoretical_loss": 3.5691823040882946,
"tokens_seen": 1268908032
},
{
"epoch": 0.38,
"learning_rate": 0.00031089267803410234,
"loss": 2.7789,
"theoretical_loss": 3.569165525410959,
"tokens_seen": 1268973568
},
{
"epoch": 0.38,
"learning_rate": 0.00031088264794383147,
"loss": 2.5376,
"theoretical_loss": 3.569148747842749,
"tokens_seen": 1269039104
},
{
"epoch": 0.38,
"learning_rate": 0.0003108726178535607,
"loss": 2.6317,
"theoretical_loss": 3.569131971383533,
"tokens_seen": 1269104640
},
{
"epoch": 0.38,
"learning_rate": 0.00031086258776328983,
"loss": 2.8159,
"theoretical_loss": 3.5691151960331826,
"tokens_seen": 1269170176
},
{
"epoch": 0.38,
"learning_rate": 0.00031085255767301906,
"loss": 2.5943,
"theoretical_loss": 3.5690984217915656,
"tokens_seen": 1269235712
},
{
"epoch": 0.38,
"learning_rate": 0.00031084252758274824,
"loss": 2.8706,
"theoretical_loss": 3.5690816486585515,
"tokens_seen": 1269301248
},
{
"epoch": 0.38,
"learning_rate": 0.0003108324974924774,
"loss": 2.7675,
"theoretical_loss": 3.56906487663401,
"tokens_seen": 1269366784
},
{
"epoch": 0.38,
"learning_rate": 0.0003108224674022066,
"loss": 2.8154,
"theoretical_loss": 3.5690481057178114,
"tokens_seen": 1269432320
},
{
"epoch": 0.38,
"learning_rate": 0.0003108124373119358,
"loss": 2.896,
"theoretical_loss": 3.5690313359098242,
"tokens_seen": 1269497856
},
{
"epoch": 0.38,
"learning_rate": 0.000310802407221665,
"loss": 2.9032,
"theoretical_loss": 3.5690145672099187,
"tokens_seen": 1269563392
},
{
"epoch": 0.38,
"learning_rate": 0.0003107923771313942,
"loss": 3.0774,
"theoretical_loss": 3.5689977996179643,
"tokens_seen": 1269628928
},
{
"epoch": 0.38,
"learning_rate": 0.0003107823470411234,
"loss": 2.6425,
"theoretical_loss": 3.56898103313383,
"tokens_seen": 1269694464
},
{
"epoch": 0.38,
"objective/train/docs_used": 720440,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.099595785140991,
"objective/train/theoretical_loss": 3.5689642677573863,
"objective/train/tokens_used": 1290220000,
"theoretical_loss": 3.5689642677573863,
"tokens_seen": 1269760000
},
{
"epoch": 0.38,
"learning_rate": 0.00031077231695085257,
"loss": 2.7608,
"theoretical_loss": 3.5689642677573863,
"tokens_seen": 1269760000
},
{
"epoch": 0.38,
"learning_rate": 0.0003107622868605818,
"loss": 2.7135,
"theoretical_loss": 3.568947503488503,
"tokens_seen": 1269825536
},
{
"epoch": 0.38,
"learning_rate": 0.00031075225677031093,
"loss": 2.9044,
"theoretical_loss": 3.5689307403270485,
"tokens_seen": 1269891072
},
{
"epoch": 0.38,
"learning_rate": 0.00031074222668004016,
"loss": 2.8508,
"theoretical_loss": 3.568913978272893,
"tokens_seen": 1269956608
},
{
"epoch": 0.38,
"learning_rate": 0.0003107321965897693,
"loss": 2.8343,
"theoretical_loss": 3.5688972173259073,
"tokens_seen": 1270022144
},
{
"epoch": 0.38,
"learning_rate": 0.0003107221664994985,
"loss": 2.7869,
"theoretical_loss": 3.5688804574859603,
"tokens_seen": 1270087680
},
{
"epoch": 0.38,
"learning_rate": 0.0003107121364092277,
"loss": 2.711,
"theoretical_loss": 3.5688636987529216,
"tokens_seen": 1270153216
},
{
"epoch": 0.38,
"learning_rate": 0.0003107021063189569,
"loss": 2.778,
"theoretical_loss": 3.568846941126661,
"tokens_seen": 1270218752
},
{
"epoch": 0.38,
"learning_rate": 0.00031069207622868607,
"loss": 2.5944,
"theoretical_loss": 3.5688301846070485,
"tokens_seen": 1270284288
},
{
"epoch": 0.38,
"learning_rate": 0.00031068204613841525,
"loss": 2.8229,
"theoretical_loss": 3.568813429193954,
"tokens_seen": 1270349824
},
{
"epoch": 0.38,
"learning_rate": 0.00031067201604814443,
"loss": 2.8504,
"theoretical_loss": 3.5687966748872473,
"tokens_seen": 1270415360
},
{
"epoch": 0.38,
"learning_rate": 0.00031066198595787367,
"loss": 2.5962,
"theoretical_loss": 3.568779921686798,
"tokens_seen": 1270480896
},
{
"epoch": 0.39,
"learning_rate": 0.0003106519558676028,
"loss": 2.8539,
"theoretical_loss": 3.5687631695924766,
"tokens_seen": 1270546432
},
{
"epoch": 0.39,
"learning_rate": 0.00031064192577733203,
"loss": 2.9034,
"theoretical_loss": 3.568746418604153,
"tokens_seen": 1270611968
},
{
"epoch": 0.39,
"learning_rate": 0.0003106318956870612,
"loss": 2.6145,
"theoretical_loss": 3.5687296687216965,
"tokens_seen": 1270677504
},
{
"epoch": 0.39,
"learning_rate": 0.0003106218655967904,
"loss": 2.7747,
"theoretical_loss": 3.568712919944977,
"tokens_seen": 1270743040
},
{
"epoch": 0.39,
"learning_rate": 0.00031061183550651957,
"loss": 2.6721,
"theoretical_loss": 3.5686961722738655,
"tokens_seen": 1270808576
},
{
"epoch": 0.39,
"learning_rate": 0.00031060180541624875,
"loss": 2.5613,
"theoretical_loss": 3.568679425708231,
"tokens_seen": 1270874112
},
{
"epoch": 0.39,
"learning_rate": 0.00031059177532597793,
"loss": 2.8381,
"theoretical_loss": 3.568662680247944,
"tokens_seen": 1270939648
},
{
"epoch": 0.39,
"learning_rate": 0.00031058174523570717,
"loss": 2.8601,
"theoretical_loss": 3.5686459358928744,
"tokens_seen": 1271005184
},
{
"epoch": 0.39,
"learning_rate": 0.0003105717151454363,
"loss": 2.8052,
"theoretical_loss": 3.568629192642892,
"tokens_seen": 1271070720
},
{
"epoch": 0.39,
"learning_rate": 0.00031056168505516553,
"loss": 2.6909,
"theoretical_loss": 3.5686124504978673,
"tokens_seen": 1271136256
},
{
"epoch": 0.39,
"learning_rate": 0.00031055165496489466,
"loss": 2.7748,
"theoretical_loss": 3.5685957094576706,
"tokens_seen": 1271201792
},
{
"epoch": 0.39,
"learning_rate": 0.0003105416248746239,
"loss": 2.8056,
"theoretical_loss": 3.5685789695221715,
"tokens_seen": 1271267328
},
{
"epoch": 0.39,
"learning_rate": 0.0003105315947843531,
"loss": 2.7376,
"theoretical_loss": 3.5685622306912403,
"tokens_seen": 1271332864
},
{
"epoch": 0.39,
"objective/train/docs_used": 721295,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.745565414428711,
"objective/train/theoretical_loss": 3.5685454929647475,
"objective/train/tokens_used": 1291858400,
"theoretical_loss": 3.5685454929647475,
"tokens_seen": 1271398400
},
{
"epoch": 0.39,
"learning_rate": 0.00031052156469408226,
"loss": 2.7062,
"theoretical_loss": 3.5685454929647475,
"tokens_seen": 1271398400
},
{
"epoch": 0.39,
"learning_rate": 0.00031051153460381144,
"loss": 2.6028,
"theoretical_loss": 3.5685287563425634,
"tokens_seen": 1271463936
},
{
"epoch": 0.39,
"learning_rate": 0.0003105015045135406,
"loss": 2.5033,
"theoretical_loss": 3.5685120208245578,
"tokens_seen": 1271529472
},
{
"epoch": 0.39,
"learning_rate": 0.0003104914744232698,
"loss": 2.6158,
"theoretical_loss": 3.568495286410601,
"tokens_seen": 1271595008
},
{
"epoch": 0.39,
"learning_rate": 0.00031048144433299904,
"loss": 2.4951,
"theoretical_loss": 3.5684785531005634,
"tokens_seen": 1271660544
},
{
"epoch": 0.39,
"learning_rate": 0.00031047141424272816,
"loss": 2.5493,
"theoretical_loss": 3.5684618208943153,
"tokens_seen": 1271726080
},
{
"epoch": 0.39,
"learning_rate": 0.0003104613841524574,
"loss": 2.6043,
"theoretical_loss": 3.5684450897917266,
"tokens_seen": 1271791616
},
{
"epoch": 0.39,
"learning_rate": 0.0003104513540621866,
"loss": 2.7523,
"theoretical_loss": 3.5684283597926685,
"tokens_seen": 1271857152
},
{
"epoch": 0.39,
"learning_rate": 0.00031044132397191576,
"loss": 2.679,
"theoretical_loss": 3.5684116308970104,
"tokens_seen": 1271922688
},
{
"epoch": 0.39,
"learning_rate": 0.00031043129388164494,
"loss": 2.783,
"theoretical_loss": 3.5683949031046236,
"tokens_seen": 1271988224
},
{
"epoch": 0.39,
"learning_rate": 0.0003104212637913741,
"loss": 2.6209,
"theoretical_loss": 3.5683781764153775,
"tokens_seen": 1272053760
},
{
"epoch": 0.39,
"learning_rate": 0.0003104112337011033,
"loss": 2.8761,
"theoretical_loss": 3.5683614508291432,
"tokens_seen": 1272119296
},
{
"epoch": 0.39,
"learning_rate": 0.00031040120361083254,
"loss": 2.8764,
"theoretical_loss": 3.5683447263457913,
"tokens_seen": 1272184832
},
{
"epoch": 0.39,
"learning_rate": 0.00031039117352056167,
"loss": 2.8194,
"theoretical_loss": 3.568328002965192,
"tokens_seen": 1272250368
},
{
"epoch": 0.39,
"learning_rate": 0.0003103811434302909,
"loss": 2.7318,
"theoretical_loss": 3.5683112806872157,
"tokens_seen": 1272315904
},
{
"epoch": 0.39,
"learning_rate": 0.00031037111334002003,
"loss": 2.9822,
"theoretical_loss": 3.568294559511733,
"tokens_seen": 1272381440
},
{
"epoch": 0.39,
"learning_rate": 0.00031036108324974926,
"loss": 2.7693,
"theoretical_loss": 3.568277839438614,
"tokens_seen": 1272446976
},
{
"epoch": 0.39,
"learning_rate": 0.00031035105315947844,
"loss": 2.6971,
"theoretical_loss": 3.5682611204677297,
"tokens_seen": 1272512512
},
{
"epoch": 0.39,
"learning_rate": 0.0003103410230692076,
"loss": 2.7853,
"theoretical_loss": 3.568244402598951,
"tokens_seen": 1272578048
},
{
"epoch": 0.39,
"learning_rate": 0.0003103309929789368,
"loss": 2.5278,
"theoretical_loss": 3.5682276858321478,
"tokens_seen": 1272643584
},
{
"epoch": 0.39,
"learning_rate": 0.000310320962888666,
"loss": 2.8012,
"theoretical_loss": 3.5682109701671916,
"tokens_seen": 1272709120
},
{
"epoch": 0.39,
"learning_rate": 0.00031031093279839517,
"loss": 2.8747,
"theoretical_loss": 3.568194255603952,
"tokens_seen": 1272774656
},
{
"epoch": 0.39,
"learning_rate": 0.0003103009027081244,
"loss": 2.6363,
"theoretical_loss": 3.5681775421423003,
"tokens_seen": 1272840192
},
{
"epoch": 0.39,
"learning_rate": 0.00031029087261785353,
"loss": 2.8111,
"theoretical_loss": 3.5681608297821072,
"tokens_seen": 1272905728
},
{
"epoch": 0.39,
"learning_rate": 0.00031028084252758277,
"loss": 2.4614,
"theoretical_loss": 3.568144118523243,
"tokens_seen": 1272971264
},
{
"epoch": 0.39,
"objective/train/docs_used": 722001,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.550626039505005,
"objective/train/theoretical_loss": 3.5681274083655787,
"objective/train/tokens_used": 1293496800,
"theoretical_loss": 3.5681274083655787,
"tokens_seen": 1273036800
},
{
"epoch": 0.39,
"learning_rate": 0.00031027081243731195,
"loss": 2.7761,
"theoretical_loss": 3.5681274083655787,
"tokens_seen": 1273036800
},
{
"epoch": 0.39,
"learning_rate": 0.00031026078234704113,
"loss": 2.912,
"theoretical_loss": 3.5681106993089853,
"tokens_seen": 1273102336
},
{
"epoch": 0.39,
"learning_rate": 0.0003102507522567703,
"loss": 2.6889,
"theoretical_loss": 3.568093991353333,
"tokens_seen": 1273167872
},
{
"epoch": 0.39,
"learning_rate": 0.0003102407221664995,
"loss": 2.6532,
"theoretical_loss": 3.568077284498493,
"tokens_seen": 1273233408
},
{
"epoch": 0.39,
"learning_rate": 0.00031023069207622867,
"loss": 2.8478,
"theoretical_loss": 3.568060578744336,
"tokens_seen": 1273298944
},
{
"epoch": 0.39,
"learning_rate": 0.0003102206619859579,
"loss": 2.6443,
"theoretical_loss": 3.5680438740907334,
"tokens_seen": 1273364480
},
{
"epoch": 0.39,
"learning_rate": 0.00031021063189568703,
"loss": 2.7186,
"theoretical_loss": 3.5680271705375546,
"tokens_seen": 1273430016
},
{
"epoch": 0.39,
"learning_rate": 0.00031020060180541627,
"loss": 2.68,
"theoretical_loss": 3.5680104680846725,
"tokens_seen": 1273495552
},
{
"epoch": 0.39,
"learning_rate": 0.0003101905717151454,
"loss": 2.6118,
"theoretical_loss": 3.5679937667319557,
"tokens_seen": 1273561088
},
{
"epoch": 0.39,
"learning_rate": 0.00031018054162487463,
"loss": 2.6951,
"theoretical_loss": 3.567977066479277,
"tokens_seen": 1273626624
},
{
"epoch": 0.39,
"learning_rate": 0.0003101705115346038,
"loss": 2.6006,
"theoretical_loss": 3.567960367326507,
"tokens_seen": 1273692160
},
{
"epoch": 0.39,
"learning_rate": 0.000310160481444333,
"loss": 2.4322,
"theoretical_loss": 3.5679436692735154,
"tokens_seen": 1273757696
},
{
"epoch": 0.39,
"learning_rate": 0.0003101504513540622,
"loss": 2.696,
"theoretical_loss": 3.567926972320175,
"tokens_seen": 1273823232
},
{
"epoch": 0.39,
"learning_rate": 0.0003101404212637914,
"loss": 2.7984,
"theoretical_loss": 3.5679102764663555,
"tokens_seen": 1273888768
},
{
"epoch": 0.39,
"learning_rate": 0.00031013039117352054,
"loss": 2.7746,
"theoretical_loss": 3.567893581711929,
"tokens_seen": 1273954304
},
{
"epoch": 0.39,
"learning_rate": 0.00031012036108324977,
"loss": 2.8089,
"theoretical_loss": 3.567876888056765,
"tokens_seen": 1274019840
},
{
"epoch": 0.39,
"learning_rate": 0.0003101103309929789,
"loss": 2.7885,
"theoretical_loss": 3.567860195500736,
"tokens_seen": 1274085376
},
{
"epoch": 0.39,
"learning_rate": 0.00031010030090270813,
"loss": 2.6409,
"theoretical_loss": 3.5678435040437124,
"tokens_seen": 1274150912
},
{
"epoch": 0.39,
"learning_rate": 0.0003100902708124373,
"loss": 2.8328,
"theoretical_loss": 3.567826813685566,
"tokens_seen": 1274216448
},
{
"epoch": 0.39,
"learning_rate": 0.0003100802407221665,
"loss": 2.8305,
"theoretical_loss": 3.5678101244261677,
"tokens_seen": 1274281984
},
{
"epoch": 0.39,
"learning_rate": 0.0003100702106318957,
"loss": 2.7481,
"theoretical_loss": 3.5677934362653874,
"tokens_seen": 1274347520
},
{
"epoch": 0.39,
"learning_rate": 0.00031006018054162486,
"loss": 2.7745,
"theoretical_loss": 3.567776749203098,
"tokens_seen": 1274413056
},
{
"epoch": 0.39,
"learning_rate": 0.0003100501504513541,
"loss": 2.6743,
"theoretical_loss": 3.56776006323917,
"tokens_seen": 1274478592
},
{
"epoch": 0.39,
"learning_rate": 0.0003100401203610833,
"loss": 2.9037,
"theoretical_loss": 3.567743378373475,
"tokens_seen": 1274544128
},
{
"epoch": 0.39,
"learning_rate": 0.00031003009027081246,
"loss": 2.6933,
"theoretical_loss": 3.5677266946058834,
"tokens_seen": 1274609664
},
{
"epoch": 0.39,
"objective/train/docs_used": 723183,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8249852657318115,
"objective/train/theoretical_loss": 3.5677100119362675,
"objective/train/tokens_used": 1295135200,
"theoretical_loss": 3.5677100119362675,
"tokens_seen": 1274675200
},
{
"epoch": 0.39,
"learning_rate": 0.00031002006018054164,
"loss": 2.707,
"theoretical_loss": 3.5677100119362675,
"tokens_seen": 1274675200
},
{
"epoch": 0.39,
"learning_rate": 0.0003100100300902708,
"loss": 2.9391,
"theoretical_loss": 3.5676933303644978,
"tokens_seen": 1274740736
},
{
"epoch": 0.39,
"learning_rate": 0.00031,
"loss": 2.9349,
"theoretical_loss": 3.567676649890446,
"tokens_seen": 1274806272
},
{
"epoch": 0.39,
"learning_rate": 0.00030998996990972924,
"loss": 2.7078,
"theoretical_loss": 3.5676599705139833,
"tokens_seen": 1274871808
},
{
"epoch": 0.39,
"learning_rate": 0.00030997993981945836,
"loss": 2.9643,
"theoretical_loss": 3.5676432922349814,
"tokens_seen": 1274937344
},
{
"epoch": 0.39,
"learning_rate": 0.0003099699097291876,
"loss": 2.8009,
"theoretical_loss": 3.567626615053311,
"tokens_seen": 1275002880
},
{
"epoch": 0.39,
"learning_rate": 0.0003099598796389168,
"loss": 2.8868,
"theoretical_loss": 3.567609938968844,
"tokens_seen": 1275068416
},
{
"epoch": 0.39,
"learning_rate": 0.00030994984954864596,
"loss": 2.8576,
"theoretical_loss": 3.567593263981452,
"tokens_seen": 1275133952
},
{
"epoch": 0.39,
"learning_rate": 0.00030993981945837514,
"loss": 2.6296,
"theoretical_loss": 3.5675765900910057,
"tokens_seen": 1275199488
},
{
"epoch": 0.39,
"learning_rate": 0.0003099297893681043,
"loss": 2.584,
"theoretical_loss": 3.5675599172973778,
"tokens_seen": 1275265024
},
{
"epoch": 0.39,
"learning_rate": 0.0003099197592778335,
"loss": 2.6457,
"theoretical_loss": 3.5675432456004383,
"tokens_seen": 1275330560
},
{
"epoch": 0.39,
"learning_rate": 0.00030990972918756274,
"loss": 2.7585,
"theoretical_loss": 3.5675265750000595,
"tokens_seen": 1275396096
},
{
"epoch": 0.39,
"learning_rate": 0.00030989969909729187,
"loss": 2.81,
"theoretical_loss": 3.567509905496113,
"tokens_seen": 1275461632
},
{
"epoch": 0.39,
"learning_rate": 0.0003098896690070211,
"loss": 2.3885,
"theoretical_loss": 3.5674932370884704,
"tokens_seen": 1275527168
},
{
"epoch": 0.39,
"learning_rate": 0.00030987963891675023,
"loss": 2.601,
"theoretical_loss": 3.5674765697770034,
"tokens_seen": 1275592704
},
{
"epoch": 0.39,
"learning_rate": 0.00030986960882647946,
"loss": 2.7631,
"theoretical_loss": 3.5674599035615824,
"tokens_seen": 1275658240
},
{
"epoch": 0.39,
"learning_rate": 0.00030985957873620864,
"loss": 2.7953,
"theoretical_loss": 3.5674432384420807,
"tokens_seen": 1275723776
},
{
"epoch": 0.39,
"learning_rate": 0.0003098495486459378,
"loss": 2.6715,
"theoretical_loss": 3.567426574418369,
"tokens_seen": 1275789312
},
{
"epoch": 0.39,
"learning_rate": 0.000309839518555667,
"loss": 2.7551,
"theoretical_loss": 3.5674099114903193,
"tokens_seen": 1275854848
},
{
"epoch": 0.39,
"learning_rate": 0.0003098294884653962,
"loss": 2.6867,
"theoretical_loss": 3.567393249657803,
"tokens_seen": 1275920384
},
{
"epoch": 0.39,
"learning_rate": 0.00030981945837512537,
"loss": 2.7303,
"theoretical_loss": 3.567376588920692,
"tokens_seen": 1275985920
},
{
"epoch": 0.39,
"learning_rate": 0.0003098094282848546,
"loss": 2.9016,
"theoretical_loss": 3.567359929278858,
"tokens_seen": 1276051456
},
{
"epoch": 0.39,
"learning_rate": 0.00030979939819458373,
"loss": 2.7891,
"theoretical_loss": 3.567343270732172,
"tokens_seen": 1276116992
},
{
"epoch": 0.39,
"learning_rate": 0.00030978936810431297,
"loss": 2.6907,
"theoretical_loss": 3.5673266132805073,
"tokens_seen": 1276182528
},
{
"epoch": 0.39,
"learning_rate": 0.00030977933801404215,
"loss": 2.8597,
"theoretical_loss": 3.5673099569237348,
"tokens_seen": 1276248064
},
{
"epoch": 0.39,
"objective/train/docs_used": 723876,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.591486692428589,
"objective/train/theoretical_loss": 3.567293301661726,
"objective/train/tokens_used": 1296773600,
"theoretical_loss": 3.567293301661726,
"tokens_seen": 1276313600
},
{
"epoch": 0.39,
"learning_rate": 0.00030976930792377133,
"loss": 2.79,
"theoretical_loss": 3.567293301661726,
"tokens_seen": 1276313600
},
{
"epoch": 0.39,
"learning_rate": 0.0003097592778335005,
"loss": 2.6179,
"theoretical_loss": 3.567276647494354,
"tokens_seen": 1276379136
},
{
"epoch": 0.39,
"learning_rate": 0.0003097492477432297,
"loss": 2.7678,
"theoretical_loss": 3.567259994421489,
"tokens_seen": 1276444672
},
{
"epoch": 0.39,
"learning_rate": 0.00030973921765295887,
"loss": 2.7048,
"theoretical_loss": 3.5672433424430037,
"tokens_seen": 1276510208
},
{
"epoch": 0.39,
"learning_rate": 0.0003097291875626881,
"loss": 2.6649,
"theoretical_loss": 3.56722669155877,
"tokens_seen": 1276575744
},
{
"epoch": 0.39,
"learning_rate": 0.00030971915747241723,
"loss": 2.9094,
"theoretical_loss": 3.56721004176866,
"tokens_seen": 1276641280
},
{
"epoch": 0.39,
"learning_rate": 0.00030970912738214647,
"loss": 3.0008,
"theoretical_loss": 3.5671933930725457,
"tokens_seen": 1276706816
},
{
"epoch": 0.39,
"learning_rate": 0.0003096990972918756,
"loss": 2.6808,
"theoretical_loss": 3.5671767454702983,
"tokens_seen": 1276772352
},
{
"epoch": 0.39,
"learning_rate": 0.00030968906720160483,
"loss": 2.7928,
"theoretical_loss": 3.56716009896179,
"tokens_seen": 1276837888
},
{
"epoch": 0.39,
"learning_rate": 0.000309679037111334,
"loss": 2.8783,
"theoretical_loss": 3.567143453546894,
"tokens_seen": 1276903424
},
{
"epoch": 0.39,
"learning_rate": 0.0003096690070210632,
"loss": 2.7838,
"theoretical_loss": 3.567126809225481,
"tokens_seen": 1276968960
},
{
"epoch": 0.39,
"learning_rate": 0.0003096589769307924,
"loss": 2.8027,
"theoretical_loss": 3.5671101659974234,
"tokens_seen": 1277034496
},
{
"epoch": 0.39,
"learning_rate": 0.0003096489468405216,
"loss": 2.5963,
"theoretical_loss": 3.567093523862593,
"tokens_seen": 1277100032
},
{
"epoch": 0.39,
"learning_rate": 0.00030963891675025074,
"loss": 2.6907,
"theoretical_loss": 3.567076882820863,
"tokens_seen": 1277165568
},
{
"epoch": 0.39,
"learning_rate": 0.00030962888665997997,
"loss": 2.8332,
"theoretical_loss": 3.5670602428721043,
"tokens_seen": 1277231104
},
{
"epoch": 0.39,
"learning_rate": 0.0003096188565697091,
"loss": 2.9968,
"theoretical_loss": 3.56704360401619,
"tokens_seen": 1277296640
},
{
"epoch": 0.39,
"learning_rate": 0.00030960882647943833,
"loss": 2.6943,
"theoretical_loss": 3.5670269662529916,
"tokens_seen": 1277362176
},
{
"epoch": 0.39,
"learning_rate": 0.0003095987963891675,
"loss": 2.7755,
"theoretical_loss": 3.5670103295823816,
"tokens_seen": 1277427712
},
{
"epoch": 0.39,
"learning_rate": 0.0003095887662988967,
"loss": 2.7823,
"theoretical_loss": 3.5669936940042315,
"tokens_seen": 1277493248
},
{
"epoch": 0.39,
"learning_rate": 0.0003095787362086259,
"loss": 2.758,
"theoretical_loss": 3.5669770595184143,
"tokens_seen": 1277558784
},
{
"epoch": 0.39,
"learning_rate": 0.00030956870611835506,
"loss": 2.7704,
"theoretical_loss": 3.566960426124802,
"tokens_seen": 1277624320
},
{
"epoch": 0.39,
"learning_rate": 0.00030955867602808424,
"loss": 2.9628,
"theoretical_loss": 3.566943793823267,
"tokens_seen": 1277689856
},
{
"epoch": 0.39,
"learning_rate": 0.0003095486459378135,
"loss": 2.7132,
"theoretical_loss": 3.5669271626136814,
"tokens_seen": 1277755392
},
{
"epoch": 0.39,
"learning_rate": 0.0003095386158475426,
"loss": 3.0201,
"theoretical_loss": 3.5669105324959176,
"tokens_seen": 1277820928
},
{
"epoch": 0.39,
"learning_rate": 0.00030952858575727184,
"loss": 2.7589,
"theoretical_loss": 3.566893903469848,
"tokens_seen": 1277886464
},
{
"epoch": 0.39,
"objective/train/docs_used": 725220,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5600483417510986,
"objective/train/theoretical_loss": 3.566877275535345,
"objective/train/tokens_used": 1298412000,
"theoretical_loss": 3.566877275535345,
"tokens_seen": 1277952000
},
{
"epoch": 0.39,
"learning_rate": 0.00030951855566700096,
"loss": 2.6912,
"theoretical_loss": 3.566877275535345,
"tokens_seen": 1277952000
},
{
"epoch": 0.39,
"learning_rate": 0.0003095085255767302,
"loss": 2.6797,
"theoretical_loss": 3.5668606486922805,
"tokens_seen": 1278017536
},
{
"epoch": 0.39,
"learning_rate": 0.0003094984954864594,
"loss": 2.9684,
"theoretical_loss": 3.566844022940528,
"tokens_seen": 1278083072
},
{
"epoch": 0.39,
"learning_rate": 0.00030948846539618856,
"loss": 2.4837,
"theoretical_loss": 3.5668273982799583,
"tokens_seen": 1278148608
},
{
"epoch": 0.39,
"learning_rate": 0.00030947843530591774,
"loss": 2.7275,
"theoretical_loss": 3.5668107747104454,
"tokens_seen": 1278214144
},
{
"epoch": 0.39,
"learning_rate": 0.000309468405215647,
"loss": 2.4473,
"theoretical_loss": 3.5667941522318607,
"tokens_seen": 1278279680
},
{
"epoch": 0.39,
"learning_rate": 0.0003094583751253761,
"loss": 2.7427,
"theoretical_loss": 3.566777530844077,
"tokens_seen": 1278345216
},
{
"epoch": 0.39,
"learning_rate": 0.00030944834503510534,
"loss": 2.8094,
"theoretical_loss": 3.5667609105469666,
"tokens_seen": 1278410752
},
{
"epoch": 0.39,
"learning_rate": 0.00030943831494483447,
"loss": 2.7084,
"theoretical_loss": 3.5667442913404024,
"tokens_seen": 1278476288
},
{
"epoch": 0.39,
"learning_rate": 0.0003094282848545637,
"loss": 2.8972,
"theoretical_loss": 3.5667276732242574,
"tokens_seen": 1278541824
},
{
"epoch": 0.39,
"learning_rate": 0.0003094182547642929,
"loss": 2.4249,
"theoretical_loss": 3.566711056198403,
"tokens_seen": 1278607360
},
{
"epoch": 0.39,
"learning_rate": 0.00030940822467402207,
"loss": 2.7347,
"theoretical_loss": 3.5666944402627125,
"tokens_seen": 1278672896
},
{
"epoch": 0.39,
"learning_rate": 0.00030939819458375125,
"loss": 2.9499,
"theoretical_loss": 3.5666778254170586,
"tokens_seen": 1278738432
},
{
"epoch": 0.39,
"learning_rate": 0.00030938816449348043,
"loss": 2.5048,
"theoretical_loss": 3.5666612116613137,
"tokens_seen": 1278803968
},
{
"epoch": 0.39,
"learning_rate": 0.0003093781344032096,
"loss": 2.7884,
"theoretical_loss": 3.5666445989953504,
"tokens_seen": 1278869504
},
{
"epoch": 0.39,
"learning_rate": 0.00030936810431293884,
"loss": 2.7442,
"theoretical_loss": 3.566627987419041,
"tokens_seen": 1278935040
},
{
"epoch": 0.39,
"learning_rate": 0.00030935807422266797,
"loss": 2.692,
"theoretical_loss": 3.5666113769322596,
"tokens_seen": 1279000576
},
{
"epoch": 0.39,
"learning_rate": 0.0003093480441323972,
"loss": 2.7968,
"theoretical_loss": 3.5665947675348777,
"tokens_seen": 1279066112
},
{
"epoch": 0.39,
"learning_rate": 0.00030933801404212633,
"loss": 2.8812,
"theoretical_loss": 3.5665781592267676,
"tokens_seen": 1279131648
},
{
"epoch": 0.39,
"learning_rate": 0.00030932798395185557,
"loss": 2.7018,
"theoretical_loss": 3.5665615520078036,
"tokens_seen": 1279197184
},
{
"epoch": 0.39,
"learning_rate": 0.00030931795386158475,
"loss": 2.6316,
"theoretical_loss": 3.5665449458778573,
"tokens_seen": 1279262720
},
{
"epoch": 0.39,
"learning_rate": 0.00030930792377131393,
"loss": 2.9534,
"theoretical_loss": 3.566528340836802,
"tokens_seen": 1279328256
},
{
"epoch": 0.39,
"learning_rate": 0.00030929789368104317,
"loss": 2.8629,
"theoretical_loss": 3.56651173688451,
"tokens_seen": 1279393792
},
{
"epoch": 0.39,
"learning_rate": 0.00030928786359077235,
"loss": 2.8241,
"theoretical_loss": 3.5664951340208546,
"tokens_seen": 1279459328
},
{
"epoch": 0.39,
"learning_rate": 0.00030927783350050153,
"loss": 2.9834,
"theoretical_loss": 3.566478532245709,
"tokens_seen": 1279524864
},
{
"epoch": 0.39,
"objective/train/docs_used": 725568,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5770699977874756,
"objective/train/theoretical_loss": 3.566461931558945,
"objective/train/tokens_used": 1300050400,
"theoretical_loss": 3.566461931558945,
"tokens_seen": 1279590400
},
{
"epoch": 0.39,
"learning_rate": 0.0003092678034102307,
"loss": 2.8702,
"theoretical_loss": 3.566461931558945,
"tokens_seen": 1279590400
},
{
"epoch": 0.39,
"learning_rate": 0.0003092577733199599,
"loss": 2.6505,
"theoretical_loss": 3.566445331960437,
"tokens_seen": 1279655936
},
{
"epoch": 0.39,
"learning_rate": 0.00030924774322968907,
"loss": 2.6086,
"theoretical_loss": 3.5664287334500564,
"tokens_seen": 1279721472
},
{
"epoch": 0.39,
"learning_rate": 0.0003092377131394183,
"loss": 2.7104,
"theoretical_loss": 3.5664121360276777,
"tokens_seen": 1279787008
},
{
"epoch": 0.39,
"learning_rate": 0.00030922768304914743,
"loss": 2.6944,
"theoretical_loss": 3.566395539693172,
"tokens_seen": 1279852544
},
{
"epoch": 0.39,
"learning_rate": 0.00030921765295887667,
"loss": 2.6119,
"theoretical_loss": 3.566378944446414,
"tokens_seen": 1279918080
},
{
"epoch": 0.39,
"learning_rate": 0.0003092076228686058,
"loss": 3.0136,
"theoretical_loss": 3.5663623502872763,
"tokens_seen": 1279983616
},
{
"epoch": 0.39,
"learning_rate": 0.00030919759277833503,
"loss": 2.6903,
"theoretical_loss": 3.5663457572156307,
"tokens_seen": 1280049152
},
{
"epoch": 0.39,
"learning_rate": 0.0003091875626880642,
"loss": 2.9385,
"theoretical_loss": 3.5663291652313522,
"tokens_seen": 1280114688
},
{
"epoch": 0.39,
"learning_rate": 0.0003091775325977934,
"loss": 2.5997,
"theoretical_loss": 3.5663125743343125,
"tokens_seen": 1280180224
},
{
"epoch": 0.39,
"learning_rate": 0.0003091675025075226,
"loss": 2.8462,
"theoretical_loss": 3.566295984524386,
"tokens_seen": 1280245760
},
{
"epoch": 0.39,
"learning_rate": 0.0003091574724172518,
"loss": 2.618,
"theoretical_loss": 3.566279395801444,
"tokens_seen": 1280311296
},
{
"epoch": 0.39,
"learning_rate": 0.00030914744232698094,
"loss": 2.7617,
"theoretical_loss": 3.5662628081653605,
"tokens_seen": 1280376832
},
{
"epoch": 0.39,
"learning_rate": 0.00030913741223671017,
"loss": 2.8456,
"theoretical_loss": 3.566246221616009,
"tokens_seen": 1280442368
},
{
"epoch": 0.39,
"learning_rate": 0.0003091273821464393,
"loss": 2.8847,
"theoretical_loss": 3.566229636153263,
"tokens_seen": 1280507904
},
{
"epoch": 0.39,
"learning_rate": 0.00030911735205616853,
"loss": 2.7655,
"theoretical_loss": 3.5662130517769945,
"tokens_seen": 1280573440
},
{
"epoch": 0.39,
"learning_rate": 0.0003091073219658977,
"loss": 2.7001,
"theoretical_loss": 3.5661964684870777,
"tokens_seen": 1280638976
},
{
"epoch": 0.39,
"learning_rate": 0.0003090972918756269,
"loss": 2.9887,
"theoretical_loss": 3.5661798862833853,
"tokens_seen": 1280704512
},
{
"epoch": 0.39,
"learning_rate": 0.0003090872617853561,
"loss": 2.7539,
"theoretical_loss": 3.5661633051657917,
"tokens_seen": 1280770048
},
{
"epoch": 0.39,
"learning_rate": 0.00030907723169508526,
"loss": 2.9286,
"theoretical_loss": 3.5661467251341685,
"tokens_seen": 1280835584
},
{
"epoch": 0.39,
"learning_rate": 0.00030906720160481444,
"loss": 2.8043,
"theoretical_loss": 3.5661301461883896,
"tokens_seen": 1280901120
},
{
"epoch": 0.39,
"learning_rate": 0.0003090571715145437,
"loss": 2.7779,
"theoretical_loss": 3.566113568328329,
"tokens_seen": 1280966656
},
{
"epoch": 0.39,
"learning_rate": 0.0003090471414242728,
"loss": 2.5705,
"theoretical_loss": 3.5660969915538594,
"tokens_seen": 1281032192
},
{
"epoch": 0.39,
"learning_rate": 0.00030903711133400204,
"loss": 2.9114,
"theoretical_loss": 3.566080415864855,
"tokens_seen": 1281097728
},
{
"epoch": 0.39,
"learning_rate": 0.00030902708124373116,
"loss": 2.9009,
"theoretical_loss": 3.566063841261188,
"tokens_seen": 1281163264
},
{
"epoch": 0.39,
"objective/train/docs_used": 726917,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.055298328399658,
"objective/train/theoretical_loss": 3.566047267742733,
"objective/train/tokens_used": 1301688800,
"theoretical_loss": 3.566047267742733,
"tokens_seen": 1281228800
},
{
"epoch": 0.39,
"learning_rate": 0.0003090170511534604,
"loss": 2.8336,
"theoretical_loss": 3.566047267742733,
"tokens_seen": 1281228800
},
{
"epoch": 0.39,
"learning_rate": 0.0003090070210631896,
"loss": 2.5814,
"theoretical_loss": 3.5660306953093626,
"tokens_seen": 1281294336
},
{
"epoch": 0.39,
"learning_rate": 0.00030899699097291876,
"loss": 2.9455,
"theoretical_loss": 3.56601412396095,
"tokens_seen": 1281359872
},
{
"epoch": 0.39,
"learning_rate": 0.00030898696088264794,
"loss": 2.6698,
"theoretical_loss": 3.56599755369737,
"tokens_seen": 1281425408
},
{
"epoch": 0.39,
"learning_rate": 0.0003089769307923772,
"loss": 2.6922,
"theoretical_loss": 3.565980984518495,
"tokens_seen": 1281490944
},
{
"epoch": 0.39,
"learning_rate": 0.0003089669007021063,
"loss": 2.744,
"theoretical_loss": 3.5659644164241984,
"tokens_seen": 1281556480
},
{
"epoch": 0.39,
"learning_rate": 0.00030895687061183554,
"loss": 2.6802,
"theoretical_loss": 3.565947849414355,
"tokens_seen": 1281622016
},
{
"epoch": 0.39,
"learning_rate": 0.00030894684052156467,
"loss": 2.719,
"theoretical_loss": 3.565931283488837,
"tokens_seen": 1281687552
},
{
"epoch": 0.39,
"learning_rate": 0.0003089368104312939,
"loss": 2.9657,
"theoretical_loss": 3.5659147186475186,
"tokens_seen": 1281753088
},
{
"epoch": 0.39,
"learning_rate": 0.0003089267803410231,
"loss": 2.6404,
"theoretical_loss": 3.5658981548902737,
"tokens_seen": 1281818624
},
{
"epoch": 0.39,
"learning_rate": 0.00030891675025075227,
"loss": 2.7408,
"theoretical_loss": 3.565881592216975,
"tokens_seen": 1281884160
},
{
"epoch": 0.39,
"learning_rate": 0.00030890672016048145,
"loss": 2.8528,
"theoretical_loss": 3.5658650306274975,
"tokens_seen": 1281949696
},
{
"epoch": 0.39,
"learning_rate": 0.00030889669007021063,
"loss": 2.6371,
"theoretical_loss": 3.565848470121714,
"tokens_seen": 1282015232
},
{
"epoch": 0.39,
"learning_rate": 0.0003088866599799398,
"loss": 2.7485,
"theoretical_loss": 3.5658319106994982,
"tokens_seen": 1282080768
},
{
"epoch": 0.39,
"learning_rate": 0.00030887662988966904,
"loss": 2.8788,
"theoretical_loss": 3.5658153523607234,
"tokens_seen": 1282146304
},
{
"epoch": 0.39,
"learning_rate": 0.00030886659979939817,
"loss": 2.6034,
"theoretical_loss": 3.5657987951052648,
"tokens_seen": 1282211840
},
{
"epoch": 0.39,
"learning_rate": 0.0003088565697091274,
"loss": 2.8554,
"theoretical_loss": 3.5657822389329947,
"tokens_seen": 1282277376
},
{
"epoch": 0.39,
"learning_rate": 0.00030884653961885653,
"loss": 2.7123,
"theoretical_loss": 3.5657656838437877,
"tokens_seen": 1282342912
},
{
"epoch": 0.39,
"learning_rate": 0.00030883650952858577,
"loss": 2.6225,
"theoretical_loss": 3.565749129837517,
"tokens_seen": 1282408448
},
{
"epoch": 0.39,
"learning_rate": 0.00030882647943831495,
"loss": 2.6894,
"theoretical_loss": 3.565732576914057,
"tokens_seen": 1282473984
},
{
"epoch": 0.39,
"learning_rate": 0.00030881644934804413,
"loss": 2.6719,
"theoretical_loss": 3.5657160250732813,
"tokens_seen": 1282539520
},
{
"epoch": 0.39,
"learning_rate": 0.0003088064192577733,
"loss": 2.7548,
"theoretical_loss": 3.5656994743150636,
"tokens_seen": 1282605056
},
{
"epoch": 0.39,
"learning_rate": 0.00030879638916750255,
"loss": 2.772,
"theoretical_loss": 3.565682924639278,
"tokens_seen": 1282670592
},
{
"epoch": 0.39,
"learning_rate": 0.0003087863590772317,
"loss": 2.6619,
"theoretical_loss": 3.5656663760457987,
"tokens_seen": 1282736128
},
{
"epoch": 0.39,
"learning_rate": 0.0003087763289869609,
"loss": 2.6643,
"theoretical_loss": 3.565649828534499,
"tokens_seen": 1282801664
},
{
"epoch": 0.39,
"objective/train/docs_used": 727506,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9273886680603027,
"objective/train/theoretical_loss": 3.565633282105253,
"objective/train/tokens_used": 1303327200,
"theoretical_loss": 3.565633282105253,
"tokens_seen": 1282867200
},
{
"epoch": 0.39,
"learning_rate": 0.00030876629889669004,
"loss": 2.7058,
"theoretical_loss": 3.565633282105253,
"tokens_seen": 1282867200
},
{
"epoch": 0.39,
"learning_rate": 0.00030875626880641927,
"loss": 2.6803,
"theoretical_loss": 3.565616736757935,
"tokens_seen": 1282932736
},
{
"epoch": 0.39,
"learning_rate": 0.00030874623871614845,
"loss": 2.4302,
"theoretical_loss": 3.565600192492419,
"tokens_seen": 1282998272
},
{
"epoch": 0.39,
"learning_rate": 0.00030873620862587763,
"loss": 2.7642,
"theoretical_loss": 3.5655836493085786,
"tokens_seen": 1283063808
},
{
"epoch": 0.39,
"learning_rate": 0.0003087261785356068,
"loss": 2.5602,
"theoretical_loss": 3.565567107206288,
"tokens_seen": 1283129344
},
{
"epoch": 0.39,
"learning_rate": 0.000308716148445336,
"loss": 2.944,
"theoretical_loss": 3.5655505661854217,
"tokens_seen": 1283194880
},
{
"epoch": 0.39,
"learning_rate": 0.0003087061183550652,
"loss": 2.9148,
"theoretical_loss": 3.565534026245853,
"tokens_seen": 1283260416
},
{
"epoch": 0.39,
"learning_rate": 0.0003086960882647944,
"loss": 2.5494,
"theoretical_loss": 3.5655174873874564,
"tokens_seen": 1283325952
},
{
"epoch": 0.39,
"learning_rate": 0.00030868605817452354,
"loss": 2.7837,
"theoretical_loss": 3.565500949610106,
"tokens_seen": 1283391488
},
{
"epoch": 0.39,
"learning_rate": 0.0003086760280842528,
"loss": 2.8258,
"theoretical_loss": 3.5654844129136762,
"tokens_seen": 1283457024
},
{
"epoch": 0.39,
"learning_rate": 0.0003086659979939819,
"loss": 2.7493,
"theoretical_loss": 3.565467877298041,
"tokens_seen": 1283522560
},
{
"epoch": 0.39,
"learning_rate": 0.00030865596790371114,
"loss": 2.7915,
"theoretical_loss": 3.5654513427630743,
"tokens_seen": 1283588096
},
{
"epoch": 0.39,
"learning_rate": 0.0003086459378134403,
"loss": 2.7985,
"theoretical_loss": 3.56543480930865,
"tokens_seen": 1283653632
},
{
"epoch": 0.39,
"learning_rate": 0.0003086359077231695,
"loss": 2.5707,
"theoretical_loss": 3.5654182769346434,
"tokens_seen": 1283719168
},
{
"epoch": 0.39,
"learning_rate": 0.0003086258776328987,
"loss": 2.8751,
"theoretical_loss": 3.565401745640928,
"tokens_seen": 1283784704
},
{
"epoch": 0.39,
"learning_rate": 0.0003086158475426279,
"loss": 2.7324,
"theoretical_loss": 3.565385215427378,
"tokens_seen": 1283850240
},
{
"epoch": 0.39,
"learning_rate": 0.00030860581745235704,
"loss": 2.5143,
"theoretical_loss": 3.5653686862938687,
"tokens_seen": 1283915776
},
{
"epoch": 0.39,
"learning_rate": 0.0003085957873620863,
"loss": 2.7379,
"theoretical_loss": 3.565352158240273,
"tokens_seen": 1283981312
},
{
"epoch": 0.39,
"learning_rate": 0.0003085857572718154,
"loss": 2.9366,
"theoretical_loss": 3.565335631266466,
"tokens_seen": 1284046848
},
{
"epoch": 0.39,
"learning_rate": 0.00030857572718154464,
"loss": 2.692,
"theoretical_loss": 3.565319105372322,
"tokens_seen": 1284112384
},
{
"epoch": 0.39,
"learning_rate": 0.0003085656970912738,
"loss": 2.6389,
"theoretical_loss": 3.565302580557715,
"tokens_seen": 1284177920
},
{
"epoch": 0.39,
"learning_rate": 0.000308555667001003,
"loss": 2.8127,
"theoretical_loss": 3.5652860568225204,
"tokens_seen": 1284243456
},
{
"epoch": 0.39,
"learning_rate": 0.00030854563691073224,
"loss": 2.9062,
"theoretical_loss": 3.565269534166611,
"tokens_seen": 1284308992
},
{
"epoch": 0.39,
"learning_rate": 0.00030853560682046137,
"loss": 2.6976,
"theoretical_loss": 3.5652530125898627,
"tokens_seen": 1284374528
},
{
"epoch": 0.39,
"learning_rate": 0.0003085255767301906,
"loss": 2.8234,
"theoretical_loss": 3.565236492092149,
"tokens_seen": 1284440064
},
{
"epoch": 0.39,
"objective/train/docs_used": 728764,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.980675458908081,
"objective/train/theoretical_loss": 3.5652199726733453,
"objective/train/tokens_used": 1304965600,
"theoretical_loss": 3.5652199726733453,
"tokens_seen": 1284505600
},
{
"epoch": 0.39,
"learning_rate": 0.0003085155466399198,
"loss": 2.8691,
"theoretical_loss": 3.5652199726733453,
"tokens_seen": 1284505600
},
{
"epoch": 0.39,
"learning_rate": 0.00030850551654964896,
"loss": 2.6539,
"theoretical_loss": 3.565203454333325,
"tokens_seen": 1284571136
},
{
"epoch": 0.39,
"learning_rate": 0.00030849548645937814,
"loss": 2.8747,
"theoretical_loss": 3.5651869370719633,
"tokens_seen": 1284636672
},
{
"epoch": 0.39,
"learning_rate": 0.0003084854563691074,
"loss": 2.7207,
"theoretical_loss": 3.565170420889135,
"tokens_seen": 1284702208
},
{
"epoch": 0.39,
"learning_rate": 0.0003084754262788365,
"loss": 2.7163,
"theoretical_loss": 3.565153905784714,
"tokens_seen": 1284767744
},
{
"epoch": 0.39,
"learning_rate": 0.00030846539618856574,
"loss": 2.9998,
"theoretical_loss": 3.565137391758575,
"tokens_seen": 1284833280
},
{
"epoch": 0.39,
"learning_rate": 0.00030845536609829487,
"loss": 2.676,
"theoretical_loss": 3.565120878810593,
"tokens_seen": 1284898816
},
{
"epoch": 0.39,
"learning_rate": 0.0003084453360080241,
"loss": 2.6041,
"theoretical_loss": 3.565104366940642,
"tokens_seen": 1284964352
},
{
"epoch": 0.39,
"learning_rate": 0.0003084353059177533,
"loss": 2.7746,
"theoretical_loss": 3.565087856148598,
"tokens_seen": 1285029888
},
{
"epoch": 0.39,
"learning_rate": 0.00030842527582748247,
"loss": 2.7575,
"theoretical_loss": 3.5650713464343338,
"tokens_seen": 1285095424
},
{
"epoch": 0.39,
"learning_rate": 0.00030841524573721165,
"loss": 2.5431,
"theoretical_loss": 3.5650548377977254,
"tokens_seen": 1285160960
},
{
"epoch": 0.39,
"learning_rate": 0.00030840521564694083,
"loss": 2.6105,
"theoretical_loss": 3.5650383302386466,
"tokens_seen": 1285226496
},
{
"epoch": 0.39,
"learning_rate": 0.00030839518555667,
"loss": 2.7934,
"theoretical_loss": 3.5650218237569735,
"tokens_seen": 1285292032
},
{
"epoch": 0.39,
"learning_rate": 0.00030838515546639924,
"loss": 2.7544,
"theoretical_loss": 3.565005318352579,
"tokens_seen": 1285357568
},
{
"epoch": 0.39,
"learning_rate": 0.00030837512537612837,
"loss": 2.5653,
"theoretical_loss": 3.5649888140253396,
"tokens_seen": 1285423104
},
{
"epoch": 0.39,
"learning_rate": 0.0003083650952858576,
"loss": 2.6105,
"theoretical_loss": 3.564972310775129,
"tokens_seen": 1285488640
},
{
"epoch": 0.39,
"learning_rate": 0.00030835506519558673,
"loss": 2.7116,
"theoretical_loss": 3.5649558086018223,
"tokens_seen": 1285554176
},
{
"epoch": 0.39,
"learning_rate": 0.00030834503510531597,
"loss": 2.8415,
"theoretical_loss": 3.564939307505295,
"tokens_seen": 1285619712
},
{
"epoch": 0.39,
"learning_rate": 0.00030833500501504515,
"loss": 2.7409,
"theoretical_loss": 3.564922807485421,
"tokens_seen": 1285685248
},
{
"epoch": 0.39,
"learning_rate": 0.00030832497492477433,
"loss": 2.7997,
"theoretical_loss": 3.564906308542075,
"tokens_seen": 1285750784
},
{
"epoch": 0.39,
"learning_rate": 0.0003083149448345035,
"loss": 2.8154,
"theoretical_loss": 3.5648898106751328,
"tokens_seen": 1285816320
},
{
"epoch": 0.39,
"learning_rate": 0.00030830491474423275,
"loss": 2.8888,
"theoretical_loss": 3.564873313884469,
"tokens_seen": 1285881856
},
{
"epoch": 0.39,
"learning_rate": 0.0003082948846539619,
"loss": 2.8754,
"theoretical_loss": 3.5648568181699583,
"tokens_seen": 1285947392
},
{
"epoch": 0.39,
"learning_rate": 0.0003082848545636911,
"loss": 2.6699,
"theoretical_loss": 3.5648403235314756,
"tokens_seen": 1286012928
},
{
"epoch": 0.39,
"learning_rate": 0.00030827482447342024,
"loss": 2.8704,
"theoretical_loss": 3.564823829968897,
"tokens_seen": 1286078464
},
{
"epoch": 0.39,
"objective/train/docs_used": 729512,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.317129135131836,
"objective/train/theoretical_loss": 3.5648073374820957,
"objective/train/tokens_used": 1306604000,
"theoretical_loss": 3.5648073374820957,
"tokens_seen": 1286144000
},
{
"epoch": 0.39,
"learning_rate": 0.00030826479438314947,
"loss": 2.7962,
"theoretical_loss": 3.5648073374820957,
"tokens_seen": 1286144000
},
{
"epoch": 0.39,
"learning_rate": 0.00030825476429287865,
"loss": 2.9322,
"theoretical_loss": 3.564790846070948,
"tokens_seen": 1286209536
},
{
"epoch": 0.39,
"learning_rate": 0.00030824473420260783,
"loss": 2.8003,
"theoretical_loss": 3.5647743557353286,
"tokens_seen": 1286275072
},
{
"epoch": 0.39,
"learning_rate": 0.000308234704112337,
"loss": 2.757,
"theoretical_loss": 3.5647578664751123,
"tokens_seen": 1286340608
},
{
"epoch": 0.39,
"learning_rate": 0.0003082246740220662,
"loss": 2.5395,
"theoretical_loss": 3.564741378290175,
"tokens_seen": 1286406144
},
{
"epoch": 0.39,
"learning_rate": 0.0003082146439317954,
"loss": 2.9932,
"theoretical_loss": 3.564724891180391,
"tokens_seen": 1286471680
},
{
"epoch": 0.39,
"learning_rate": 0.0003082046138415246,
"loss": 2.6274,
"theoretical_loss": 3.564708405145635,
"tokens_seen": 1286537216
},
{
"epoch": 0.39,
"learning_rate": 0.00030819458375125374,
"loss": 2.6687,
"theoretical_loss": 3.5646919201857834,
"tokens_seen": 1286602752
},
{
"epoch": 0.39,
"learning_rate": 0.000308184553660983,
"loss": 2.8443,
"theoretical_loss": 3.5646754363007105,
"tokens_seen": 1286668288
},
{
"epoch": 0.39,
"learning_rate": 0.0003081745235707121,
"loss": 2.7352,
"theoretical_loss": 3.564658953490292,
"tokens_seen": 1286733824
},
{
"epoch": 0.39,
"learning_rate": 0.00030816449348044134,
"loss": 2.7762,
"theoretical_loss": 3.5646424717544027,
"tokens_seen": 1286799360
},
{
"epoch": 0.39,
"learning_rate": 0.0003081544633901705,
"loss": 2.8203,
"theoretical_loss": 3.5646259910929183,
"tokens_seen": 1286864896
},
{
"epoch": 0.39,
"learning_rate": 0.0003081444332998997,
"loss": 2.6607,
"theoretical_loss": 3.5646095115057137,
"tokens_seen": 1286930432
},
{
"epoch": 0.39,
"learning_rate": 0.0003081344032096289,
"loss": 2.5533,
"theoretical_loss": 3.564593032992664,
"tokens_seen": 1286995968
},
{
"epoch": 0.39,
"learning_rate": 0.0003081243731193581,
"loss": 2.7993,
"theoretical_loss": 3.5645765555536446,
"tokens_seen": 1287061504
},
{
"epoch": 0.39,
"learning_rate": 0.00030811434302908724,
"loss": 2.8256,
"theoretical_loss": 3.5645600791885315,
"tokens_seen": 1287127040
},
{
"epoch": 0.39,
"learning_rate": 0.0003081043129388165,
"loss": 2.7019,
"theoretical_loss": 3.564543603897199,
"tokens_seen": 1287192576
},
{
"epoch": 0.39,
"learning_rate": 0.0003080942828485456,
"loss": 2.7151,
"theoretical_loss": 3.564527129679523,
"tokens_seen": 1287258112
},
{
"epoch": 0.39,
"learning_rate": 0.00030808425275827484,
"loss": 2.8256,
"theoretical_loss": 3.564510656535379,
"tokens_seen": 1287323648
},
{
"epoch": 0.39,
"learning_rate": 0.000308074222668004,
"loss": 2.6186,
"theoretical_loss": 3.5644941844646416,
"tokens_seen": 1287389184
},
{
"epoch": 0.39,
"learning_rate": 0.0003080641925777332,
"loss": 2.8066,
"theoretical_loss": 3.5644777134671877,
"tokens_seen": 1287454720
},
{
"epoch": 0.39,
"learning_rate": 0.0003080541624874624,
"loss": 2.7874,
"theoretical_loss": 3.5644612435428913,
"tokens_seen": 1287520256
},
{
"epoch": 0.39,
"learning_rate": 0.00030804413239719157,
"loss": 2.8107,
"theoretical_loss": 3.564444774691628,
"tokens_seen": 1287585792
},
{
"epoch": 0.39,
"learning_rate": 0.00030803410230692075,
"loss": 3.0598,
"theoretical_loss": 3.5644283069132743,
"tokens_seen": 1287651328
},
{
"epoch": 0.39,
"learning_rate": 0.00030802407221665,
"loss": 2.5032,
"theoretical_loss": 3.564411840207705,
"tokens_seen": 1287716864
},
{
"epoch": 0.39,
"objective/train/docs_used": 730476,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.809786796569824,
"objective/train/theoretical_loss": 3.564395374574796,
"objective/train/tokens_used": 1308242400,
"theoretical_loss": 3.564395374574796,
"tokens_seen": 1287782400
},
{
"epoch": 0.39,
"learning_rate": 0.0003080140421263791,
"loss": 2.7636,
"theoretical_loss": 3.564395374574796,
"tokens_seen": 1287782400
},
{
"epoch": 0.39,
"learning_rate": 0.00030800401203610834,
"loss": 2.8776,
"theoretical_loss": 3.564378910014422,
"tokens_seen": 1287847936
},
{
"epoch": 0.39,
"learning_rate": 0.0003079939819458375,
"loss": 2.551,
"theoretical_loss": 3.5643624465264594,
"tokens_seen": 1287913472
},
{
"epoch": 0.39,
"learning_rate": 0.0003079839518555667,
"loss": 2.7014,
"theoretical_loss": 3.5643459841107834,
"tokens_seen": 1287979008
},
{
"epoch": 0.39,
"learning_rate": 0.0003079739217652959,
"loss": 2.5949,
"theoretical_loss": 3.5643295227672698,
"tokens_seen": 1288044544
},
{
"epoch": 0.39,
"learning_rate": 0.00030796389167502507,
"loss": 2.8958,
"theoretical_loss": 3.564313062495794,
"tokens_seen": 1288110080
},
{
"epoch": 0.39,
"learning_rate": 0.00030795386158475425,
"loss": 3.0505,
"theoretical_loss": 3.564296603296232,
"tokens_seen": 1288175616
},
{
"epoch": 0.39,
"learning_rate": 0.0003079438314944835,
"loss": 2.9146,
"theoretical_loss": 3.564280145168459,
"tokens_seen": 1288241152
},
{
"epoch": 0.39,
"learning_rate": 0.0003079338014042126,
"loss": 2.8062,
"theoretical_loss": 3.5642636881123515,
"tokens_seen": 1288306688
},
{
"epoch": 0.39,
"learning_rate": 0.00030792377131394185,
"loss": 2.7606,
"theoretical_loss": 3.564247232127784,
"tokens_seen": 1288372224
},
{
"epoch": 0.39,
"learning_rate": 0.000307913741223671,
"loss": 2.628,
"theoretical_loss": 3.5642307772146333,
"tokens_seen": 1288437760
},
{
"epoch": 0.39,
"learning_rate": 0.0003079037111334002,
"loss": 2.6312,
"theoretical_loss": 3.5642143233727746,
"tokens_seen": 1288503296
},
{
"epoch": 0.39,
"learning_rate": 0.0003078936810431294,
"loss": 2.7153,
"theoretical_loss": 3.5641978706020834,
"tokens_seen": 1288568832
},
{
"epoch": 0.39,
"learning_rate": 0.00030788365095285857,
"loss": 2.7593,
"theoretical_loss": 3.5641814189024363,
"tokens_seen": 1288634368
},
{
"epoch": 0.39,
"learning_rate": 0.00030787362086258775,
"loss": 2.7391,
"theoretical_loss": 3.5641649682737087,
"tokens_seen": 1288699904
},
{
"epoch": 0.39,
"learning_rate": 0.00030786359077231693,
"loss": 2.7398,
"theoretical_loss": 3.5641485187157764,
"tokens_seen": 1288765440
},
{
"epoch": 0.39,
"learning_rate": 0.0003078535606820461,
"loss": 2.8419,
"theoretical_loss": 3.5641320702285153,
"tokens_seen": 1288830976
},
{
"epoch": 0.39,
"learning_rate": 0.00030784353059177535,
"loss": 2.8168,
"theoretical_loss": 3.5641156228118014,
"tokens_seen": 1288896512
},
{
"epoch": 0.39,
"learning_rate": 0.0003078335005015045,
"loss": 2.5466,
"theoretical_loss": 3.5640991764655103,
"tokens_seen": 1288962048
},
{
"epoch": 0.39,
"learning_rate": 0.0003078234704112337,
"loss": 2.8578,
"theoretical_loss": 3.564082731189518,
"tokens_seen": 1289027584
},
{
"epoch": 0.39,
"learning_rate": 0.0003078134403209629,
"loss": 2.9294,
"theoretical_loss": 3.5640662869837003,
"tokens_seen": 1289093120
},
{
"epoch": 0.39,
"learning_rate": 0.0003078034102306921,
"loss": 2.6838,
"theoretical_loss": 3.5640498438479336,
"tokens_seen": 1289158656
},
{
"epoch": 0.39,
"learning_rate": 0.0003077933801404213,
"loss": 2.7992,
"theoretical_loss": 3.564033401782094,
"tokens_seen": 1289224192
},
{
"epoch": 0.39,
"learning_rate": 0.00030778335005015044,
"loss": 2.7989,
"theoretical_loss": 3.5640169607860566,
"tokens_seen": 1289289728
},
{
"epoch": 0.39,
"learning_rate": 0.00030777331995987967,
"loss": 2.6282,
"theoretical_loss": 3.564000520859698,
"tokens_seen": 1289355264
},
{
"epoch": 0.39,
"objective/train/docs_used": 731120,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.566955089569092,
"objective/train/theoretical_loss": 3.5639840820028947,
"objective/train/tokens_used": 1309880800,
"theoretical_loss": 3.5639840820028947,
"tokens_seen": 1289420800
},
{
"epoch": 0.39,
"learning_rate": 0.00030776328986960885,
"loss": 2.6542,
"theoretical_loss": 3.5639840820028947,
"tokens_seen": 1289420800
},
{
"epoch": 0.39,
"learning_rate": 0.00030775325977933803,
"loss": 2.7215,
"theoretical_loss": 3.5639676442155217,
"tokens_seen": 1289486336
},
{
"epoch": 0.39,
"learning_rate": 0.0003077432296890672,
"loss": 2.6158,
"theoretical_loss": 3.563951207497456,
"tokens_seen": 1289551872
},
{
"epoch": 0.39,
"learning_rate": 0.0003077331995987964,
"loss": 2.6254,
"theoretical_loss": 3.563934771848573,
"tokens_seen": 1289617408
},
{
"epoch": 0.39,
"learning_rate": 0.0003077231695085256,
"loss": 2.7296,
"theoretical_loss": 3.5639183372687495,
"tokens_seen": 1289682944
},
{
"epoch": 0.39,
"learning_rate": 0.0003077131394182548,
"loss": 3.0182,
"theoretical_loss": 3.5639019037578614,
"tokens_seen": 1289748480
},
{
"epoch": 0.39,
"learning_rate": 0.00030770310932798394,
"loss": 2.7658,
"theoretical_loss": 3.5638854713157846,
"tokens_seen": 1289814016
},
{
"epoch": 0.39,
"learning_rate": 0.0003076930792377132,
"loss": 2.6954,
"theoretical_loss": 3.5638690399423956,
"tokens_seen": 1289879552
},
{
"epoch": 0.39,
"learning_rate": 0.0003076830491474423,
"loss": 2.7811,
"theoretical_loss": 3.5638526096375704,
"tokens_seen": 1289945088
},
{
"epoch": 0.39,
"learning_rate": 0.00030767301905717154,
"loss": 2.5791,
"theoretical_loss": 3.563836180401185,
"tokens_seen": 1290010624
},
{
"epoch": 0.39,
"learning_rate": 0.0003076629889669007,
"loss": 2.8416,
"theoretical_loss": 3.5638197522331163,
"tokens_seen": 1290076160
},
{
"epoch": 0.39,
"learning_rate": 0.0003076529588766299,
"loss": 2.7342,
"theoretical_loss": 3.56380332513324,
"tokens_seen": 1290141696
},
{
"epoch": 0.39,
"learning_rate": 0.0003076429287863591,
"loss": 2.6018,
"theoretical_loss": 3.5637868991014328,
"tokens_seen": 1290207232
},
{
"epoch": 0.39,
"learning_rate": 0.0003076328986960883,
"loss": 2.8256,
"theoretical_loss": 3.563770474137571,
"tokens_seen": 1290272768
},
{
"epoch": 0.39,
"learning_rate": 0.00030762286860581744,
"loss": 2.7164,
"theoretical_loss": 3.56375405024153,
"tokens_seen": 1290338304
},
{
"epoch": 0.39,
"learning_rate": 0.0003076128385155467,
"loss": 2.8796,
"theoretical_loss": 3.5637376274131873,
"tokens_seen": 1290403840
},
{
"epoch": 0.39,
"learning_rate": 0.0003076028084252758,
"loss": 2.8144,
"theoretical_loss": 3.563721205652419,
"tokens_seen": 1290469376
},
{
"epoch": 0.39,
"learning_rate": 0.00030759277833500504,
"loss": 2.7399,
"theoretical_loss": 3.563704784959101,
"tokens_seen": 1290534912
},
{
"epoch": 0.39,
"learning_rate": 0.0003075827482447342,
"loss": 2.6592,
"theoretical_loss": 3.5636883653331104,
"tokens_seen": 1290600448
},
{
"epoch": 0.39,
"learning_rate": 0.0003075727181544634,
"loss": 2.7876,
"theoretical_loss": 3.563671946774323,
"tokens_seen": 1290665984
},
{
"epoch": 0.39,
"learning_rate": 0.0003075626880641926,
"loss": 2.7977,
"theoretical_loss": 3.5636555292826158,
"tokens_seen": 1290731520
},
{
"epoch": 0.39,
"learning_rate": 0.00030755265797392177,
"loss": 2.6674,
"theoretical_loss": 3.563639112857865,
"tokens_seen": 1290797056
},
{
"epoch": 0.39,
"learning_rate": 0.00030754262788365095,
"loss": 2.7472,
"theoretical_loss": 3.563622697499947,
"tokens_seen": 1290862592
},
{
"epoch": 0.39,
"learning_rate": 0.0003075325977933802,
"loss": 2.7049,
"theoretical_loss": 3.5636062832087383,
"tokens_seen": 1290928128
},
{
"epoch": 0.39,
"learning_rate": 0.0003075225677031093,
"loss": 2.8728,
"theoretical_loss": 3.5635898699841158,
"tokens_seen": 1290993664
},
{
"epoch": 0.39,
"objective/train/docs_used": 732356,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9922869205474854,
"objective/train/theoretical_loss": 3.5635734578259557,
"objective/train/tokens_used": 1311519200,
"theoretical_loss": 3.5635734578259557,
"tokens_seen": 1291059200
},
{
"epoch": 0.39,
"learning_rate": 0.00030751253761283854,
"loss": 2.905,
"theoretical_loss": 3.5635734578259557,
"tokens_seen": 1291059200
},
{
"epoch": 0.39,
"learning_rate": 0.0003075025075225677,
"loss": 2.8317,
"theoretical_loss": 3.5635570467341346,
"tokens_seen": 1291124736
},
{
"epoch": 0.39,
"learning_rate": 0.0003074924774322969,
"loss": 2.806,
"theoretical_loss": 3.563540636708529,
"tokens_seen": 1291190272
},
{
"epoch": 0.39,
"learning_rate": 0.0003074824473420261,
"loss": 2.9096,
"theoretical_loss": 3.5635242277490162,
"tokens_seen": 1291255808
},
{
"epoch": 0.39,
"learning_rate": 0.00030747241725175527,
"loss": 2.862,
"theoretical_loss": 3.563507819855472,
"tokens_seen": 1291321344
},
{
"epoch": 0.39,
"learning_rate": 0.00030746238716148445,
"loss": 2.7015,
"theoretical_loss": 3.5634914130277737,
"tokens_seen": 1291386880
},
{
"epoch": 0.39,
"learning_rate": 0.0003074523570712137,
"loss": 2.8505,
"theoretical_loss": 3.5634750072657972,
"tokens_seen": 1291452416
},
{
"epoch": 0.39,
"learning_rate": 0.0003074423269809428,
"loss": 2.5852,
"theoretical_loss": 3.56345860256942,
"tokens_seen": 1291517952
},
{
"epoch": 0.39,
"learning_rate": 0.00030743229689067205,
"loss": 2.6636,
"theoretical_loss": 3.5634421989385183,
"tokens_seen": 1291583488
},
{
"epoch": 0.39,
"learning_rate": 0.0003074222668004012,
"loss": 2.686,
"theoretical_loss": 3.5634257963729694,
"tokens_seen": 1291649024
},
{
"epoch": 0.39,
"learning_rate": 0.0003074122367101304,
"loss": 2.8305,
"theoretical_loss": 3.5634093948726493,
"tokens_seen": 1291714560
},
{
"epoch": 0.39,
"learning_rate": 0.0003074022066198596,
"loss": 2.9239,
"theoretical_loss": 3.5633929944374354,
"tokens_seen": 1291780096
},
{
"epoch": 0.39,
"learning_rate": 0.00030739217652958877,
"loss": 2.7754,
"theoretical_loss": 3.5633765950672043,
"tokens_seen": 1291845632
},
{
"epoch": 0.39,
"learning_rate": 0.00030738214643931795,
"loss": 2.6991,
"theoretical_loss": 3.563360196761833,
"tokens_seen": 1291911168
},
{
"epoch": 0.39,
"learning_rate": 0.00030737211634904713,
"loss": 2.7849,
"theoretical_loss": 3.5633437995211974,
"tokens_seen": 1291976704
},
{
"epoch": 0.39,
"learning_rate": 0.0003073620862587763,
"loss": 2.7706,
"theoretical_loss": 3.5633274033451756,
"tokens_seen": 1292042240
},
{
"epoch": 0.39,
"learning_rate": 0.00030735205616850555,
"loss": 3.082,
"theoretical_loss": 3.563311008233644,
"tokens_seen": 1292107776
},
{
"epoch": 0.39,
"learning_rate": 0.0003073420260782347,
"loss": 2.7744,
"theoretical_loss": 3.5632946141864794,
"tokens_seen": 1292173312
},
{
"epoch": 0.39,
"learning_rate": 0.0003073319959879639,
"loss": 2.8576,
"theoretical_loss": 3.563278221203559,
"tokens_seen": 1292238848
},
{
"epoch": 0.39,
"learning_rate": 0.0003073219658976931,
"loss": 2.6724,
"theoretical_loss": 3.5632618292847593,
"tokens_seen": 1292304384
},
{
"epoch": 0.39,
"learning_rate": 0.0003073119358074223,
"loss": 2.9095,
"theoretical_loss": 3.5632454384299574,
"tokens_seen": 1292369920
},
{
"epoch": 0.39,
"learning_rate": 0.00030730190571715146,
"loss": 2.7594,
"theoretical_loss": 3.563229048639031,
"tokens_seen": 1292435456
},
{
"epoch": 0.39,
"learning_rate": 0.00030729187562688064,
"loss": 2.7651,
"theoretical_loss": 3.563212659911856,
"tokens_seen": 1292500992
},
{
"epoch": 0.39,
"learning_rate": 0.0003072818455366098,
"loss": 2.7926,
"theoretical_loss": 3.5631962722483097,
"tokens_seen": 1292566528
},
{
"epoch": 0.39,
"learning_rate": 0.00030727181544633905,
"loss": 3.0679,
"theoretical_loss": 3.56317988564827,
"tokens_seen": 1292632064
},
{
"epoch": 0.39,
"objective/train/docs_used": 733109,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.573679208755493,
"objective/train/theoretical_loss": 3.5631635001116138,
"objective/train/tokens_used": 1313157600,
"theoretical_loss": 3.5631635001116138,
"tokens_seen": 1292697600
},
{
"epoch": 0.39,
"learning_rate": 0.0003072617853560682,
"loss": 2.6614,
"theoretical_loss": 3.5631635001116138,
"tokens_seen": 1292697600
},
{
"epoch": 0.39,
"learning_rate": 0.0003072517552657974,
"loss": 2.5882,
"theoretical_loss": 3.5631471156382166,
"tokens_seen": 1292763136
},
{
"epoch": 0.39,
"learning_rate": 0.00030724172517552654,
"loss": 2.7539,
"theoretical_loss": 3.5631307322279575,
"tokens_seen": 1292828672
},
{
"epoch": 0.39,
"learning_rate": 0.0003072316950852558,
"loss": 2.6011,
"theoretical_loss": 3.5631143498807125,
"tokens_seen": 1292894208
},
{
"epoch": 0.39,
"learning_rate": 0.00030722166499498496,
"loss": 2.8452,
"theoretical_loss": 3.5630979685963586,
"tokens_seen": 1292959744
},
{
"epoch": 0.39,
"learning_rate": 0.00030721163490471414,
"loss": 2.8432,
"theoretical_loss": 3.563081588374774,
"tokens_seen": 1293025280
},
{
"epoch": 0.39,
"learning_rate": 0.0003072016048144433,
"loss": 2.8147,
"theoretical_loss": 3.563065209215835,
"tokens_seen": 1293090816
},
{
"epoch": 0.39,
"learning_rate": 0.0003071915747241725,
"loss": 2.7311,
"theoretical_loss": 3.5630488311194193,
"tokens_seen": 1293156352
},
{
"epoch": 0.39,
"learning_rate": 0.0003071815446339017,
"loss": 2.6771,
"theoretical_loss": 3.5630324540854037,
"tokens_seen": 1293221888
},
{
"epoch": 0.39,
"learning_rate": 0.0003071715145436309,
"loss": 2.8248,
"theoretical_loss": 3.563016078113666,
"tokens_seen": 1293287424
},
{
"epoch": 0.39,
"learning_rate": 0.00030716148445336005,
"loss": 2.6336,
"theoretical_loss": 3.562999703204083,
"tokens_seen": 1293352960
},
{
"epoch": 0.39,
"learning_rate": 0.0003071514543630893,
"loss": 2.8651,
"theoretical_loss": 3.5629833293565323,
"tokens_seen": 1293418496
},
{
"epoch": 0.39,
"learning_rate": 0.00030714142427281846,
"loss": 2.7568,
"theoretical_loss": 3.5629669565708904,
"tokens_seen": 1293484032
},
{
"epoch": 0.39,
"learning_rate": 0.00030713139418254764,
"loss": 2.9525,
"theoretical_loss": 3.562950584847036,
"tokens_seen": 1293549568
},
{
"epoch": 0.39,
"learning_rate": 0.0003071213640922768,
"loss": 2.8553,
"theoretical_loss": 3.5629342141848452,
"tokens_seen": 1293615104
},
{
"epoch": 0.39,
"learning_rate": 0.000307111334002006,
"loss": 2.8396,
"theoretical_loss": 3.562917844584197,
"tokens_seen": 1293680640
},
{
"epoch": 0.39,
"learning_rate": 0.0003071013039117352,
"loss": 2.7222,
"theoretical_loss": 3.5629014760449667,
"tokens_seen": 1293746176
},
{
"epoch": 0.39,
"learning_rate": 0.0003070912738214644,
"loss": 2.465,
"theoretical_loss": 3.562885108567033,
"tokens_seen": 1293811712
},
{
"epoch": 0.39,
"learning_rate": 0.00030708124373119355,
"loss": 2.7737,
"theoretical_loss": 3.562868742150273,
"tokens_seen": 1293877248
},
{
"epoch": 0.39,
"learning_rate": 0.0003070712136409228,
"loss": 2.5588,
"theoretical_loss": 3.5628523767945643,
"tokens_seen": 1293942784
},
{
"epoch": 0.39,
"learning_rate": 0.0003070611835506519,
"loss": 2.7376,
"theoretical_loss": 3.5628360124997847,
"tokens_seen": 1294008320
},
{
"epoch": 0.39,
"learning_rate": 0.00030705115346038115,
"loss": 2.9586,
"theoretical_loss": 3.562819649265811,
"tokens_seen": 1294073856
},
{
"epoch": 0.39,
"learning_rate": 0.0003070411233701104,
"loss": 2.8583,
"theoretical_loss": 3.562803287092521,
"tokens_seen": 1294139392
},
{
"epoch": 0.39,
"learning_rate": 0.0003070310932798395,
"loss": 2.9683,
"theoretical_loss": 3.5627869259797924,
"tokens_seen": 1294204928
},
{
"epoch": 0.39,
"learning_rate": 0.00030702106318956874,
"loss": 2.8382,
"theoretical_loss": 3.562770565927502,
"tokens_seen": 1294270464
},
{
"epoch": 0.39,
"objective/train/docs_used": 734261,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9057581424713135,
"objective/train/theoretical_loss": 3.5627542069355282,
"objective/train/tokens_used": 1314796000,
"theoretical_loss": 3.5627542069355282,
"tokens_seen": 1294336000
},
{
"epoch": 0.39,
"learning_rate": 0.0003070110330992979,
"loss": 2.7205,
"theoretical_loss": 3.5627542069355282,
"tokens_seen": 1294336000
},
{
"epoch": 0.39,
"learning_rate": 0.0003070010030090271,
"loss": 2.7091,
"theoretical_loss": 3.562737849003749,
"tokens_seen": 1294401536
},
{
"epoch": 0.39,
"learning_rate": 0.0003069909729187563,
"loss": 2.7958,
"theoretical_loss": 3.5627214921320407,
"tokens_seen": 1294467072
},
{
"epoch": 0.39,
"learning_rate": 0.00030698094282848547,
"loss": 2.732,
"theoretical_loss": 3.5627051363202824,
"tokens_seen": 1294532608
},
{
"epoch": 0.39,
"learning_rate": 0.00030697091273821465,
"loss": 2.6597,
"theoretical_loss": 3.56268878156835,
"tokens_seen": 1294598144
},
{
"epoch": 0.39,
"learning_rate": 0.0003069608826479439,
"loss": 2.7751,
"theoretical_loss": 3.562672427876123,
"tokens_seen": 1294663680
},
{
"epoch": 0.39,
"learning_rate": 0.000306950852557673,
"loss": 2.9601,
"theoretical_loss": 3.5626560752434777,
"tokens_seen": 1294729216
},
{
"epoch": 0.39,
"learning_rate": 0.00030694082246740225,
"loss": 2.9184,
"theoretical_loss": 3.5626397236702934,
"tokens_seen": 1294794752
},
{
"epoch": 0.39,
"learning_rate": 0.0003069307923771314,
"loss": 2.7393,
"theoretical_loss": 3.5626233731564456,
"tokens_seen": 1294860288
},
{
"epoch": 0.39,
"learning_rate": 0.0003069207622868606,
"loss": 2.8166,
"theoretical_loss": 3.562607023701814,
"tokens_seen": 1294925824
},
{
"epoch": 0.39,
"learning_rate": 0.0003069107321965898,
"loss": 2.7968,
"theoretical_loss": 3.562590675306276,
"tokens_seen": 1294991360
},
{
"epoch": 0.39,
"learning_rate": 0.00030690070210631897,
"loss": 2.484,
"theoretical_loss": 3.562574327969708,
"tokens_seen": 1295056896
},
{
"epoch": 0.39,
"learning_rate": 0.00030689067201604815,
"loss": 2.9644,
"theoretical_loss": 3.56255798169199,
"tokens_seen": 1295122432
},
{
"epoch": 0.39,
"learning_rate": 0.00030688064192577733,
"loss": 2.6026,
"theoretical_loss": 3.5625416364729983,
"tokens_seen": 1295187968
},
{
"epoch": 0.39,
"learning_rate": 0.0003068706118355065,
"loss": 2.7475,
"theoretical_loss": 3.562525292312611,
"tokens_seen": 1295253504
},
{
"epoch": 0.39,
"learning_rate": 0.00030686058174523575,
"loss": 2.7043,
"theoretical_loss": 3.5625089492107063,
"tokens_seen": 1295319040
},
{
"epoch": 0.39,
"learning_rate": 0.0003068505516549649,
"loss": 2.5498,
"theoretical_loss": 3.562492607167162,
"tokens_seen": 1295384576
},
{
"epoch": 0.39,
"learning_rate": 0.0003068405215646941,
"loss": 2.6363,
"theoretical_loss": 3.5624762661818563,
"tokens_seen": 1295450112
},
{
"epoch": 0.39,
"learning_rate": 0.0003068304914744233,
"loss": 2.7985,
"theoretical_loss": 3.5624599262546663,
"tokens_seen": 1295515648
},
{
"epoch": 0.39,
"learning_rate": 0.0003068204613841525,
"loss": 2.8171,
"theoretical_loss": 3.562443587385471,
"tokens_seen": 1295581184
},
{
"epoch": 0.39,
"learning_rate": 0.00030681043129388166,
"loss": 2.9013,
"theoretical_loss": 3.5624272495741476,
"tokens_seen": 1295646720
},
{
"epoch": 0.39,
"learning_rate": 0.00030680040120361084,
"loss": 2.8031,
"theoretical_loss": 3.5624109128205745,
"tokens_seen": 1295712256
},
{
"epoch": 0.39,
"learning_rate": 0.00030679037111334,
"loss": 2.7623,
"theoretical_loss": 3.5623945771246297,
"tokens_seen": 1295777792
},
{
"epoch": 0.39,
"learning_rate": 0.00030678034102306925,
"loss": 2.5916,
"theoretical_loss": 3.562378242486191,
"tokens_seen": 1295843328
},
{
"epoch": 0.39,
"learning_rate": 0.0003067703109327984,
"loss": 2.8308,
"theoretical_loss": 3.562361908905137,
"tokens_seen": 1295908864
},
{
"epoch": 0.39,
"objective/train/docs_used": 734748,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.128178596496582,
"objective/train/theoretical_loss": 3.562345576381345,
"objective/train/tokens_used": 1316434400,
"theoretical_loss": 3.562345576381345,
"tokens_seen": 1295974400
},
{
"epoch": 0.39,
"learning_rate": 0.0003067602808425276,
"loss": 2.7001,
"theoretical_loss": 3.562345576381345,
"tokens_seen": 1295974400
},
{
"epoch": 0.39,
"learning_rate": 0.00030675025075225674,
"loss": 2.8961,
"theoretical_loss": 3.562329244914694,
"tokens_seen": 1296039936
},
{
"epoch": 0.39,
"learning_rate": 0.000306740220661986,
"loss": 2.6752,
"theoretical_loss": 3.5623129145050614,
"tokens_seen": 1296105472
},
{
"epoch": 0.39,
"learning_rate": 0.00030673019057171516,
"loss": 2.8726,
"theoretical_loss": 3.5622965851523256,
"tokens_seen": 1296171008
},
{
"epoch": 0.39,
"learning_rate": 0.00030672016048144434,
"loss": 2.7992,
"theoretical_loss": 3.5622802568563645,
"tokens_seen": 1296236544
},
{
"epoch": 0.39,
"learning_rate": 0.0003067101303911735,
"loss": 2.5911,
"theoretical_loss": 3.562263929617057,
"tokens_seen": 1296302080
},
{
"epoch": 0.39,
"learning_rate": 0.0003067001003009027,
"loss": 2.922,
"theoretical_loss": 3.5622476034342805,
"tokens_seen": 1296367616
},
{
"epoch": 0.39,
"learning_rate": 0.0003066900702106319,
"loss": 2.8528,
"theoretical_loss": 3.5622312783079133,
"tokens_seen": 1296433152
},
{
"epoch": 0.39,
"learning_rate": 0.0003066800401203611,
"loss": 2.6952,
"theoretical_loss": 3.5622149542378345,
"tokens_seen": 1296498688
},
{
"epoch": 0.39,
"learning_rate": 0.00030667001003009025,
"loss": 2.5966,
"theoretical_loss": 3.5621986312239216,
"tokens_seen": 1296564224
},
{
"epoch": 0.39,
"learning_rate": 0.0003066599799398195,
"loss": 2.7084,
"theoretical_loss": 3.562182309266053,
"tokens_seen": 1296629760
},
{
"epoch": 0.39,
"learning_rate": 0.00030664994984954866,
"loss": 2.7577,
"theoretical_loss": 3.5621659883641072,
"tokens_seen": 1296695296
},
{
"epoch": 0.39,
"learning_rate": 0.00030663991975927784,
"loss": 2.8606,
"theoretical_loss": 3.5621496685179626,
"tokens_seen": 1296760832
},
{
"epoch": 0.39,
"learning_rate": 0.000306629889669007,
"loss": 2.767,
"theoretical_loss": 3.562133349727497,
"tokens_seen": 1296826368
},
{
"epoch": 0.39,
"learning_rate": 0.0003066198595787362,
"loss": 2.7084,
"theoretical_loss": 3.5621170319925897,
"tokens_seen": 1296891904
},
{
"epoch": 0.39,
"learning_rate": 0.0003066098294884654,
"loss": 2.6593,
"theoretical_loss": 3.5621007153131172,
"tokens_seen": 1296957440
},
{
"epoch": 0.39,
"learning_rate": 0.0003065997993981946,
"loss": 2.8421,
"theoretical_loss": 3.5620843996889606,
"tokens_seen": 1297022976
},
{
"epoch": 0.39,
"learning_rate": 0.00030658976930792375,
"loss": 2.8013,
"theoretical_loss": 3.5620680851199964,
"tokens_seen": 1297088512
},
{
"epoch": 0.39,
"learning_rate": 0.000306579739217653,
"loss": 2.7017,
"theoretical_loss": 3.5620517716061038,
"tokens_seen": 1297154048
},
{
"epoch": 0.39,
"learning_rate": 0.0003065697091273821,
"loss": 2.8159,
"theoretical_loss": 3.5620354591471606,
"tokens_seen": 1297219584
},
{
"epoch": 0.39,
"learning_rate": 0.00030655967903711135,
"loss": 2.6206,
"theoretical_loss": 3.5620191477430465,
"tokens_seen": 1297285120
},
{
"epoch": 0.39,
"learning_rate": 0.00030654964894684053,
"loss": 2.8547,
"theoretical_loss": 3.5620028373936385,
"tokens_seen": 1297350656
},
{
"epoch": 0.39,
"learning_rate": 0.0003065396188565697,
"loss": 2.7819,
"theoretical_loss": 3.5619865280988163,
"tokens_seen": 1297416192
},
{
"epoch": 0.39,
"learning_rate": 0.0003065295887662989,
"loss": 2.5183,
"theoretical_loss": 3.5619702198584577,
"tokens_seen": 1297481728
},
{
"epoch": 0.39,
"learning_rate": 0.0003065195586760281,
"loss": 2.8041,
"theoretical_loss": 3.561953912672442,
"tokens_seen": 1297547264
},
{
"epoch": 0.39,
"objective/train/docs_used": 735602,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7405169010162354,
"objective/train/theoretical_loss": 3.5619376065406474,
"objective/train/tokens_used": 1318072800,
"theoretical_loss": 3.5619376065406474,
"tokens_seen": 1297612800
},
{
"epoch": 0.39,
"learning_rate": 0.00030650952858575725,
"loss": 2.6918,
"theoretical_loss": 3.5619376065406474,
"tokens_seen": 1297612800
},
{
"epoch": 0.39,
"learning_rate": 0.0003064994984954865,
"loss": 2.8901,
"theoretical_loss": 3.5619213014629523,
"tokens_seen": 1297678336
},
{
"epoch": 0.39,
"learning_rate": 0.0003064894684052156,
"loss": 2.8438,
"theoretical_loss": 3.5619049974392354,
"tokens_seen": 1297743872
},
{
"epoch": 0.39,
"learning_rate": 0.00030647943831494485,
"loss": 2.8376,
"theoretical_loss": 3.5618886944693755,
"tokens_seen": 1297809408
},
{
"epoch": 0.39,
"learning_rate": 0.00030646940822467403,
"loss": 2.9617,
"theoretical_loss": 3.5618723925532514,
"tokens_seen": 1297874944
},
{
"epoch": 0.39,
"learning_rate": 0.0003064593781344032,
"loss": 2.5918,
"theoretical_loss": 3.561856091690742,
"tokens_seen": 1297940480
},
{
"epoch": 0.39,
"learning_rate": 0.0003064493480441324,
"loss": 2.8507,
"theoretical_loss": 3.5618397918817246,
"tokens_seen": 1298006016
},
{
"epoch": 0.39,
"learning_rate": 0.0003064393179538616,
"loss": 2.7798,
"theoretical_loss": 3.56182349312608,
"tokens_seen": 1298071552
},
{
"epoch": 0.39,
"learning_rate": 0.00030642928786359076,
"loss": 2.8298,
"theoretical_loss": 3.561807195423685,
"tokens_seen": 1298137088
},
{
"epoch": 0.39,
"learning_rate": 0.00030641925777332,
"loss": 2.8258,
"theoretical_loss": 3.5617908987744205,
"tokens_seen": 1298202624
},
{
"epoch": 0.39,
"learning_rate": 0.0003064092276830491,
"loss": 2.9716,
"theoretical_loss": 3.561774603178163,
"tokens_seen": 1298268160
},
{
"epoch": 0.39,
"learning_rate": 0.00030639919759277835,
"loss": 2.8265,
"theoretical_loss": 3.561758308634793,
"tokens_seen": 1298333696
},
{
"epoch": 0.39,
"learning_rate": 0.0003063891675025075,
"loss": 2.7583,
"theoretical_loss": 3.5617420151441888,
"tokens_seen": 1298399232
},
{
"epoch": 0.39,
"learning_rate": 0.0003063791374122367,
"loss": 2.7069,
"theoretical_loss": 3.5617257227062287,
"tokens_seen": 1298464768
},
{
"epoch": 0.39,
"learning_rate": 0.0003063691073219659,
"loss": 2.8428,
"theoretical_loss": 3.561709431320792,
"tokens_seen": 1298530304
},
{
"epoch": 0.39,
"learning_rate": 0.0003063590772316951,
"loss": 2.9255,
"theoretical_loss": 3.5616931409877584,
"tokens_seen": 1298595840
},
{
"epoch": 0.39,
"learning_rate": 0.00030634904714142426,
"loss": 2.7537,
"theoretical_loss": 3.5616768517070057,
"tokens_seen": 1298661376
},
{
"epoch": 0.39,
"learning_rate": 0.0003063390170511535,
"loss": 2.8364,
"theoretical_loss": 3.5616605634784126,
"tokens_seen": 1298726912
},
{
"epoch": 0.39,
"learning_rate": 0.0003063289869608826,
"loss": 2.9181,
"theoretical_loss": 3.561644276301859,
"tokens_seen": 1298792448
},
{
"epoch": 0.39,
"learning_rate": 0.00030631895687061186,
"loss": 2.5776,
"theoretical_loss": 3.5616279901772234,
"tokens_seen": 1298857984
},
{
"epoch": 0.39,
"learning_rate": 0.000306308926780341,
"loss": 2.9224,
"theoretical_loss": 3.5616117051043856,
"tokens_seen": 1298923520
},
{
"epoch": 0.39,
"learning_rate": 0.0003062988966900702,
"loss": 2.8632,
"theoretical_loss": 3.561595421083223,
"tokens_seen": 1298989056
},
{
"epoch": 0.39,
"learning_rate": 0.00030628886659979945,
"loss": 2.9397,
"theoretical_loss": 3.5615791381136157,
"tokens_seen": 1299054592
},
{
"epoch": 0.39,
"learning_rate": 0.0003062788365095286,
"loss": 2.9669,
"theoretical_loss": 3.561562856195443,
"tokens_seen": 1299120128
},
{
"epoch": 0.39,
"learning_rate": 0.0003062688064192578,
"loss": 2.6766,
"theoretical_loss": 3.5615465753285833,
"tokens_seen": 1299185664
},
{
"epoch": 0.39,
"objective/train/docs_used": 736089,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8279144763946533,
"objective/train/theoretical_loss": 3.561530295512916,
"objective/train/tokens_used": 1319711200,
"theoretical_loss": 3.561530295512916,
"tokens_seen": 1299251200
},
{
"epoch": 0.39,
"learning_rate": 0.00030625877632898694,
"loss": 2.8143,
"theoretical_loss": 3.561530295512916,
"tokens_seen": 1299251200
},
{
"epoch": 0.39,
"learning_rate": 0.0003062487462387162,
"loss": 2.7302,
"theoretical_loss": 3.56151401674832,
"tokens_seen": 1299316736
},
{
"epoch": 0.39,
"learning_rate": 0.00030623871614844536,
"loss": 2.8944,
"theoretical_loss": 3.5614977390346745,
"tokens_seen": 1299382272
},
{
"epoch": 0.39,
"learning_rate": 0.00030622868605817454,
"loss": 2.7262,
"theoretical_loss": 3.5614814623718587,
"tokens_seen": 1299447808
},
{
"epoch": 0.39,
"learning_rate": 0.0003062186559679037,
"loss": 2.7755,
"theoretical_loss": 3.5614651867597518,
"tokens_seen": 1299513344
},
{
"epoch": 0.39,
"learning_rate": 0.0003062086258776329,
"loss": 2.5148,
"theoretical_loss": 3.5614489121982333,
"tokens_seen": 1299578880
},
{
"epoch": 0.39,
"learning_rate": 0.0003061985957873621,
"loss": 2.9674,
"theoretical_loss": 3.5614326386871817,
"tokens_seen": 1299644416
},
{
"epoch": 0.39,
"learning_rate": 0.0003061885656970913,
"loss": 2.5579,
"theoretical_loss": 3.561416366226477,
"tokens_seen": 1299709952
},
{
"epoch": 0.39,
"learning_rate": 0.00030617853560682045,
"loss": 2.9011,
"theoretical_loss": 3.5614000948159976,
"tokens_seen": 1299775488
},
{
"epoch": 0.39,
"learning_rate": 0.0003061685055165497,
"loss": 2.9407,
"theoretical_loss": 3.561383824455624,
"tokens_seen": 1299841024
},
{
"epoch": 0.39,
"learning_rate": 0.00030615847542627886,
"loss": 2.866,
"theoretical_loss": 3.5613675551452344,
"tokens_seen": 1299906560
},
{
"epoch": 0.39,
"learning_rate": 0.00030614844533600804,
"loss": 2.6619,
"theoretical_loss": 3.561351286884708,
"tokens_seen": 1299972096
},
{
"epoch": 0.39,
"learning_rate": 0.0003061384152457372,
"loss": 2.6623,
"theoretical_loss": 3.5613350196739253,
"tokens_seen": 1300037632
},
{
"epoch": 0.39,
"learning_rate": 0.0003061283851554664,
"loss": 2.619,
"theoretical_loss": 3.561318753512764,
"tokens_seen": 1300103168
},
{
"epoch": 0.39,
"learning_rate": 0.0003061183550651956,
"loss": 3.0017,
"theoretical_loss": 3.561302488401105,
"tokens_seen": 1300168704
},
{
"epoch": 0.39,
"learning_rate": 0.0003061083249749248,
"loss": 2.8441,
"theoretical_loss": 3.5612862243388275,
"tokens_seen": 1300234240
},
{
"epoch": 0.39,
"learning_rate": 0.00030609829488465395,
"loss": 2.7424,
"theoretical_loss": 3.5612699613258094,
"tokens_seen": 1300299776
},
{
"epoch": 0.39,
"learning_rate": 0.0003060882647943832,
"loss": 2.8959,
"theoretical_loss": 3.561253699361932,
"tokens_seen": 1300365312
},
{
"epoch": 0.39,
"learning_rate": 0.0003060782347041123,
"loss": 2.9004,
"theoretical_loss": 3.561237438447074,
"tokens_seen": 1300430848
},
{
"epoch": 0.39,
"learning_rate": 0.00030606820461384155,
"loss": 2.6521,
"theoretical_loss": 3.561221178581114,
"tokens_seen": 1300496384
},
{
"epoch": 0.39,
"learning_rate": 0.00030605817452357073,
"loss": 2.7307,
"theoretical_loss": 3.5612049197639335,
"tokens_seen": 1300561920
},
{
"epoch": 0.39,
"learning_rate": 0.0003060481444332999,
"loss": 2.9231,
"theoretical_loss": 3.56118866199541,
"tokens_seen": 1300627456
},
{
"epoch": 0.39,
"learning_rate": 0.0003060381143430291,
"loss": 2.8617,
"theoretical_loss": 3.5611724052754243,
"tokens_seen": 1300692992
},
{
"epoch": 0.39,
"learning_rate": 0.0003060280842527583,
"loss": 2.5448,
"theoretical_loss": 3.5611561496038555,
"tokens_seen": 1300758528
},
{
"epoch": 0.39,
"learning_rate": 0.00030601805416248745,
"loss": 2.9459,
"theoretical_loss": 3.5611398949805833,
"tokens_seen": 1300824064
},
{
"epoch": 0.39,
"objective/train/docs_used": 737292,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.041644811630249,
"objective/train/theoretical_loss": 3.5611236414054868,
"objective/train/tokens_used": 1321349600,
"theoretical_loss": 3.5611236414054868,
"tokens_seen": 1300889600
},
{
"epoch": 0.39,
"learning_rate": 0.0003060080240722167,
"loss": 3.0103,
"theoretical_loss": 3.5611236414054868,
"tokens_seen": 1300889600
},
{
"epoch": 0.39,
"learning_rate": 0.0003059979939819458,
"loss": 2.8587,
"theoretical_loss": 3.561107388878446,
"tokens_seen": 1300955136
},
{
"epoch": 0.39,
"learning_rate": 0.00030598796389167505,
"loss": 2.6303,
"theoretical_loss": 3.561091137399341,
"tokens_seen": 1301020672
},
{
"epoch": 0.39,
"learning_rate": 0.00030597793380140423,
"loss": 2.7545,
"theoretical_loss": 3.5610748869680506,
"tokens_seen": 1301086208
},
{
"epoch": 0.39,
"learning_rate": 0.0003059679037111334,
"loss": 2.8775,
"theoretical_loss": 3.5610586375844546,
"tokens_seen": 1301151744
},
{
"epoch": 0.39,
"learning_rate": 0.0003059578736208626,
"loss": 2.9042,
"theoretical_loss": 3.561042389248433,
"tokens_seen": 1301217280
},
{
"epoch": 0.39,
"learning_rate": 0.0003059478435305918,
"loss": 2.9674,
"theoretical_loss": 3.561026141959866,
"tokens_seen": 1301282816
},
{
"epoch": 0.39,
"learning_rate": 0.00030593781344032096,
"loss": 2.8209,
"theoretical_loss": 3.5610098957186325,
"tokens_seen": 1301348352
},
{
"epoch": 0.39,
"learning_rate": 0.0003059277833500502,
"loss": 2.6546,
"theoretical_loss": 3.5609936505246123,
"tokens_seen": 1301413888
},
{
"epoch": 0.39,
"learning_rate": 0.0003059177532597793,
"loss": 2.7486,
"theoretical_loss": 3.5609774063776856,
"tokens_seen": 1301479424
},
{
"epoch": 0.39,
"learning_rate": 0.00030590772316950855,
"loss": 2.6918,
"theoretical_loss": 3.5609611632777316,
"tokens_seen": 1301544960
},
{
"epoch": 0.39,
"learning_rate": 0.0003058976930792377,
"loss": 2.8942,
"theoretical_loss": 3.560944921224631,
"tokens_seen": 1301610496
},
{
"epoch": 0.39,
"learning_rate": 0.0003058876629889669,
"loss": 2.5243,
"theoretical_loss": 3.5609286802182627,
"tokens_seen": 1301676032
},
{
"epoch": 0.39,
"learning_rate": 0.0003058776328986961,
"loss": 2.6214,
"theoretical_loss": 3.560912440258507,
"tokens_seen": 1301741568
},
{
"epoch": 0.39,
"learning_rate": 0.0003058676028084253,
"loss": 2.7839,
"theoretical_loss": 3.560896201345243,
"tokens_seen": 1301807104
},
{
"epoch": 0.39,
"learning_rate": 0.00030585757271815446,
"loss": 2.8346,
"theoretical_loss": 3.5608799634783526,
"tokens_seen": 1301872640
},
{
"epoch": 0.39,
"learning_rate": 0.0003058475426278837,
"loss": 2.7421,
"theoretical_loss": 3.560863726657714,
"tokens_seen": 1301938176
},
{
"epoch": 0.39,
"learning_rate": 0.0003058375125376128,
"loss": 2.6593,
"theoretical_loss": 3.560847490883207,
"tokens_seen": 1302003712
},
{
"epoch": 0.39,
"learning_rate": 0.00030582748244734206,
"loss": 2.7504,
"theoretical_loss": 3.560831256154713,
"tokens_seen": 1302069248
},
{
"epoch": 0.39,
"learning_rate": 0.0003058174523570712,
"loss": 2.7264,
"theoretical_loss": 3.5608150224721102,
"tokens_seen": 1302134784
},
{
"epoch": 0.39,
"learning_rate": 0.0003058074222668004,
"loss": 2.7769,
"theoretical_loss": 3.5607987898352795,
"tokens_seen": 1302200320
},
{
"epoch": 0.39,
"learning_rate": 0.0003057973921765296,
"loss": 2.7097,
"theoretical_loss": 3.5607825582441013,
"tokens_seen": 1302265856
},
{
"epoch": 0.39,
"learning_rate": 0.0003057873620862588,
"loss": 2.8082,
"theoretical_loss": 3.5607663276984547,
"tokens_seen": 1302331392
},
{
"epoch": 0.39,
"learning_rate": 0.00030577733199598796,
"loss": 2.8026,
"theoretical_loss": 3.5607500981982207,
"tokens_seen": 1302396928
},
{
"epoch": 0.39,
"learning_rate": 0.00030576730190571714,
"loss": 2.7792,
"theoretical_loss": 3.560733869743278,
"tokens_seen": 1302462464
},
{
"epoch": 0.39,
"objective/train/docs_used": 737760,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.601710557937622,
"objective/train/theoretical_loss": 3.5607176423335085,
"objective/train/tokens_used": 1322988000,
"theoretical_loss": 3.5607176423335085,
"tokens_seen": 1302528000
},
{
"epoch": 0.39,
"learning_rate": 0.0003057572718154463,
"loss": 2.9437,
"theoretical_loss": 3.5607176423335085,
"tokens_seen": 1302528000
},
{
"epoch": 0.39,
"learning_rate": 0.00030574724172517556,
"loss": 2.8546,
"theoretical_loss": 3.560701415968791,
"tokens_seen": 1302593536
},
{
"epoch": 0.39,
"learning_rate": 0.0003057372116349047,
"loss": 2.7948,
"theoretical_loss": 3.560685190649006,
"tokens_seen": 1302659072
},
{
"epoch": 0.39,
"learning_rate": 0.0003057271815446339,
"loss": 2.9573,
"theoretical_loss": 3.560668966374034,
"tokens_seen": 1302724608
},
{
"epoch": 0.39,
"learning_rate": 0.00030571715145436305,
"loss": 2.838,
"theoretical_loss": 3.560652743143754,
"tokens_seen": 1302790144
},
{
"epoch": 0.39,
"learning_rate": 0.0003057071213640923,
"loss": 2.7137,
"theoretical_loss": 3.560636520958047,
"tokens_seen": 1302855680
},
{
"epoch": 0.39,
"learning_rate": 0.00030569709127382147,
"loss": 2.9206,
"theoretical_loss": 3.5606202998167937,
"tokens_seen": 1302921216
},
{
"epoch": 0.39,
"learning_rate": 0.00030568706118355065,
"loss": 2.939,
"theoretical_loss": 3.560604079719874,
"tokens_seen": 1302986752
},
{
"epoch": 0.39,
"learning_rate": 0.00030567703109327983,
"loss": 2.901,
"theoretical_loss": 3.5605878606671677,
"tokens_seen": 1303052288
},
{
"epoch": 0.39,
"learning_rate": 0.00030566700100300906,
"loss": 3.0031,
"theoretical_loss": 3.560571642658555,
"tokens_seen": 1303117824
},
{
"epoch": 0.39,
"learning_rate": 0.0003056569709127382,
"loss": 2.6807,
"theoretical_loss": 3.5605554256939165,
"tokens_seen": 1303183360
},
{
"epoch": 0.39,
"learning_rate": 0.0003056469408224674,
"loss": 3.0052,
"theoretical_loss": 3.560539209773133,
"tokens_seen": 1303248896
},
{
"epoch": 0.39,
"learning_rate": 0.00030563691073219655,
"loss": 2.7501,
"theoretical_loss": 3.560522994896084,
"tokens_seen": 1303314432
},
{
"epoch": 0.39,
"learning_rate": 0.0003056268806419258,
"loss": 2.8752,
"theoretical_loss": 3.5605067810626503,
"tokens_seen": 1303379968
},
{
"epoch": 0.39,
"learning_rate": 0.00030561685055165497,
"loss": 3.0023,
"theoretical_loss": 3.560490568272712,
"tokens_seen": 1303445504
},
{
"epoch": 0.4,
"learning_rate": 0.00030560682046138415,
"loss": 2.9135,
"theoretical_loss": 3.5604743565261496,
"tokens_seen": 1303511040
},
{
"epoch": 0.4,
"learning_rate": 0.00030559679037111333,
"loss": 2.8064,
"theoretical_loss": 3.5604581458228437,
"tokens_seen": 1303576576
},
{
"epoch": 0.4,
"learning_rate": 0.0003055867602808425,
"loss": 2.7952,
"theoretical_loss": 3.5604419361626745,
"tokens_seen": 1303642112
},
{
"epoch": 0.4,
"learning_rate": 0.0003055767301905717,
"loss": 2.7594,
"theoretical_loss": 3.560425727545522,
"tokens_seen": 1303707648
},
{
"epoch": 0.4,
"learning_rate": 0.00030556670010030093,
"loss": 2.8994,
"theoretical_loss": 3.5604095199712678,
"tokens_seen": 1303773184
},
{
"epoch": 0.4,
"learning_rate": 0.00030555667001003006,
"loss": 2.8126,
"theoretical_loss": 3.560393313439791,
"tokens_seen": 1303838720
},
{
"epoch": 0.4,
"learning_rate": 0.0003055466399197593,
"loss": 2.7057,
"theoretical_loss": 3.5603771079509734,
"tokens_seen": 1303904256
},
{
"epoch": 0.4,
"learning_rate": 0.0003055366098294885,
"loss": 2.7114,
"theoretical_loss": 3.5603609035046944,
"tokens_seen": 1303969792
},
{
"epoch": 0.4,
"learning_rate": 0.00030552657973921765,
"loss": 2.934,
"theoretical_loss": 3.5603447001008357,
"tokens_seen": 1304035328
},
{
"epoch": 0.4,
"learning_rate": 0.0003055165496489469,
"loss": 2.6079,
"theoretical_loss": 3.5603284977392766,
"tokens_seen": 1304100864
},
{
"epoch": 0.4,
"objective/train/docs_used": 739145,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.761068344116211,
"objective/train/theoretical_loss": 3.560312296419899,
"objective/train/tokens_used": 1324626400,
"theoretical_loss": 3.560312296419899,
"tokens_seen": 1304166400
},
{
"epoch": 0.4,
"learning_rate": 0.000305506519558676,
"loss": 2.7059,
"theoretical_loss": 3.560312296419899,
"tokens_seen": 1304166400
},
{
"epoch": 0.4,
"learning_rate": 0.00030549648946840525,
"loss": 2.8934,
"theoretical_loss": 3.560296096142582,
"tokens_seen": 1304231936
},
{
"epoch": 0.4,
"learning_rate": 0.00030548645937813443,
"loss": 2.9735,
"theoretical_loss": 3.5602798969072076,
"tokens_seen": 1304297472
},
{
"epoch": 0.4,
"learning_rate": 0.0003054764292878636,
"loss": 2.5604,
"theoretical_loss": 3.5602636987136558,
"tokens_seen": 1304363008
},
{
"epoch": 0.4,
"learning_rate": 0.0003054663991975928,
"loss": 2.9606,
"theoretical_loss": 3.5602475015618067,
"tokens_seen": 1304428544
},
{
"epoch": 0.4,
"learning_rate": 0.000305456369107322,
"loss": 2.7042,
"theoretical_loss": 3.560231305451542,
"tokens_seen": 1304494080
},
{
"epoch": 0.4,
"learning_rate": 0.00030544633901705116,
"loss": 2.8524,
"theoretical_loss": 3.5602151103827424,
"tokens_seen": 1304559616
},
{
"epoch": 0.4,
"learning_rate": 0.0003054363089267804,
"loss": 2.8532,
"theoretical_loss": 3.5601989163552883,
"tokens_seen": 1304625152
},
{
"epoch": 0.4,
"learning_rate": 0.0003054262788365095,
"loss": 2.5383,
"theoretical_loss": 3.56018272336906,
"tokens_seen": 1304690688
},
{
"epoch": 0.4,
"learning_rate": 0.00030541624874623875,
"loss": 2.8742,
"theoretical_loss": 3.5601665314239384,
"tokens_seen": 1304756224
},
{
"epoch": 0.4,
"learning_rate": 0.0003054062186559679,
"loss": 2.7907,
"theoretical_loss": 3.5601503405198045,
"tokens_seen": 1304821760
},
{
"epoch": 0.4,
"learning_rate": 0.0003053961885656971,
"loss": 2.7021,
"theoretical_loss": 3.560134150656539,
"tokens_seen": 1304887296
},
{
"epoch": 0.4,
"learning_rate": 0.0003053861584754263,
"loss": 2.6907,
"theoretical_loss": 3.560117961834023,
"tokens_seen": 1304952832
},
{
"epoch": 0.4,
"learning_rate": 0.0003053761283851555,
"loss": 2.8921,
"theoretical_loss": 3.5601017740521366,
"tokens_seen": 1305018368
},
{
"epoch": 0.4,
"learning_rate": 0.00030536609829488466,
"loss": 2.7612,
"theoretical_loss": 3.5600855873107617,
"tokens_seen": 1305083904
},
{
"epoch": 0.4,
"learning_rate": 0.0003053560682046139,
"loss": 2.5995,
"theoretical_loss": 3.5600694016097787,
"tokens_seen": 1305149440
},
{
"epoch": 0.4,
"learning_rate": 0.000305346038114343,
"loss": 2.8446,
"theoretical_loss": 3.5600532169490684,
"tokens_seen": 1305214976
},
{
"epoch": 0.4,
"learning_rate": 0.00030533600802407226,
"loss": 2.7949,
"theoretical_loss": 3.560037033328512,
"tokens_seen": 1305280512
},
{
"epoch": 0.4,
"learning_rate": 0.0003053259779338014,
"loss": 2.8021,
"theoretical_loss": 3.560020850747989,
"tokens_seen": 1305346048
},
{
"epoch": 0.4,
"learning_rate": 0.0003053159478435306,
"loss": 2.8784,
"theoretical_loss": 3.560004669207383,
"tokens_seen": 1305411584
},
{
"epoch": 0.4,
"learning_rate": 0.0003053059177532598,
"loss": 2.8172,
"theoretical_loss": 3.5599884887065727,
"tokens_seen": 1305477120
},
{
"epoch": 0.4,
"learning_rate": 0.000305295887662989,
"loss": 2.6127,
"theoretical_loss": 3.55997230924544,
"tokens_seen": 1305542656
},
{
"epoch": 0.4,
"learning_rate": 0.00030528585757271816,
"loss": 2.7187,
"theoretical_loss": 3.5599561308238656,
"tokens_seen": 1305608192
},
{
"epoch": 0.4,
"learning_rate": 0.00030527582748244734,
"loss": 2.6612,
"theoretical_loss": 3.559939953441731,
"tokens_seen": 1305673728
},
{
"epoch": 0.4,
"learning_rate": 0.0003052657973921765,
"loss": 2.7634,
"theoretical_loss": 3.559923777098917,
"tokens_seen": 1305739264
},
{
"epoch": 0.4,
"objective/train/docs_used": 739729,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7067713737487793,
"objective/train/theoretical_loss": 3.5599076017953046,
"objective/train/tokens_used": 1326264800,
"theoretical_loss": 3.5599076017953046,
"tokens_seen": 1305804800
},
{
"epoch": 0.4,
"learning_rate": 0.00030525576730190576,
"loss": 2.8072,
"theoretical_loss": 3.5599076017953046,
"tokens_seen": 1305804800
},
{
"epoch": 0.4,
"learning_rate": 0.0003052457372116349,
"loss": 2.8243,
"theoretical_loss": 3.5598914275307747,
"tokens_seen": 1305870336
},
{
"epoch": 0.4,
"learning_rate": 0.0003052357071213641,
"loss": 2.8374,
"theoretical_loss": 3.5598752543052092,
"tokens_seen": 1305935872
},
{
"epoch": 0.4,
"learning_rate": 0.00030522567703109325,
"loss": 2.8183,
"theoretical_loss": 3.5598590821184883,
"tokens_seen": 1306001408
},
{
"epoch": 0.4,
"learning_rate": 0.0003052156469408225,
"loss": 2.9477,
"theoretical_loss": 3.5598429109704934,
"tokens_seen": 1306066944
},
{
"epoch": 0.4,
"learning_rate": 0.00030520561685055167,
"loss": 2.8132,
"theoretical_loss": 3.559826740861106,
"tokens_seen": 1306132480
},
{
"epoch": 0.4,
"learning_rate": 0.00030519558676028085,
"loss": 2.6226,
"theoretical_loss": 3.5598105717902073,
"tokens_seen": 1306198016
},
{
"epoch": 0.4,
"learning_rate": 0.00030518555667001003,
"loss": 2.8515,
"theoretical_loss": 3.559794403757678,
"tokens_seen": 1306263552
},
{
"epoch": 0.4,
"learning_rate": 0.00030517552657973926,
"loss": 2.6981,
"theoretical_loss": 3.5597782367633997,
"tokens_seen": 1306329088
},
{
"epoch": 0.4,
"learning_rate": 0.0003051654964894684,
"loss": 2.7981,
"theoretical_loss": 3.5597620708072535,
"tokens_seen": 1306394624
},
{
"epoch": 0.4,
"learning_rate": 0.0003051554663991976,
"loss": 2.9088,
"theoretical_loss": 3.5597459058891205,
"tokens_seen": 1306460160
},
{
"epoch": 0.4,
"learning_rate": 0.00030514543630892675,
"loss": 2.7643,
"theoretical_loss": 3.5597297420088827,
"tokens_seen": 1306525696
},
{
"epoch": 0.4,
"learning_rate": 0.000305135406218656,
"loss": 2.6535,
"theoretical_loss": 3.559713579166421,
"tokens_seen": 1306591232
},
{
"epoch": 0.4,
"learning_rate": 0.00030512537612838517,
"loss": 2.6875,
"theoretical_loss": 3.5596974173616163,
"tokens_seen": 1306656768
},
{
"epoch": 0.4,
"learning_rate": 0.00030511534603811435,
"loss": 2.8503,
"theoretical_loss": 3.55968125659435,
"tokens_seen": 1306722304
},
{
"epoch": 0.4,
"learning_rate": 0.00030510531594784353,
"loss": 2.9737,
"theoretical_loss": 3.559665096864504,
"tokens_seen": 1306787840
},
{
"epoch": 0.4,
"learning_rate": 0.0003050952858575727,
"loss": 2.6009,
"theoretical_loss": 3.5596489381719594,
"tokens_seen": 1306853376
},
{
"epoch": 0.4,
"learning_rate": 0.0003050852557673019,
"loss": 2.7417,
"theoretical_loss": 3.5596327805165977,
"tokens_seen": 1306918912
},
{
"epoch": 0.4,
"learning_rate": 0.00030507522567703113,
"loss": 2.8652,
"theoretical_loss": 3.5596166238983002,
"tokens_seen": 1306984448
},
{
"epoch": 0.4,
"learning_rate": 0.00030506519558676026,
"loss": 2.6451,
"theoretical_loss": 3.5596004683169484,
"tokens_seen": 1307049984
},
{
"epoch": 0.4,
"learning_rate": 0.0003050551654964895,
"loss": 3.0152,
"theoretical_loss": 3.5595843137724237,
"tokens_seen": 1307115520
},
{
"epoch": 0.4,
"learning_rate": 0.0003050451354062186,
"loss": 2.6977,
"theoretical_loss": 3.5595681602646074,
"tokens_seen": 1307181056
},
{
"epoch": 0.4,
"learning_rate": 0.00030503510531594785,
"loss": 2.9991,
"theoretical_loss": 3.559552007793382,
"tokens_seen": 1307246592
},
{
"epoch": 0.4,
"learning_rate": 0.00030502507522567703,
"loss": 2.6067,
"theoretical_loss": 3.5595358563586275,
"tokens_seen": 1307312128
},
{
"epoch": 0.4,
"learning_rate": 0.0003050150451354062,
"loss": 2.9821,
"theoretical_loss": 3.559519705960226,
"tokens_seen": 1307377664
},
{
"epoch": 0.4,
"objective/train/docs_used": 740313,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.487309694290161,
"objective/train/theoretical_loss": 3.55950355659806,
"objective/train/tokens_used": 1327903200,
"theoretical_loss": 3.55950355659806,
"tokens_seen": 1307443200
},
{
"epoch": 0.4,
"learning_rate": 0.0003050050150451354,
"loss": 2.681,
"theoretical_loss": 3.55950355659806,
"tokens_seen": 1307443200
},
{
"epoch": 0.4,
"learning_rate": 0.00030499498495486463,
"loss": 2.7291,
"theoretical_loss": 3.55948740827201,
"tokens_seen": 1307508736
},
{
"epoch": 0.4,
"learning_rate": 0.00030498495486459376,
"loss": 2.9009,
"theoretical_loss": 3.5594712609819577,
"tokens_seen": 1307574272
},
{
"epoch": 0.4,
"learning_rate": 0.000304974924774323,
"loss": 2.7318,
"theoretical_loss": 3.5594551147277853,
"tokens_seen": 1307639808
},
{
"epoch": 0.4,
"learning_rate": 0.0003049648946840521,
"loss": 2.6473,
"theoretical_loss": 3.5594389695093738,
"tokens_seen": 1307705344
},
{
"epoch": 0.4,
"learning_rate": 0.00030495486459378136,
"loss": 2.7645,
"theoretical_loss": 3.559422825326605,
"tokens_seen": 1307770880
},
{
"epoch": 0.4,
"learning_rate": 0.00030494483450351054,
"loss": 2.9166,
"theoretical_loss": 3.559406682179361,
"tokens_seen": 1307836416
},
{
"epoch": 0.4,
"learning_rate": 0.0003049348044132397,
"loss": 2.6138,
"theoretical_loss": 3.5593905400675228,
"tokens_seen": 1307901952
},
{
"epoch": 0.4,
"learning_rate": 0.0003049247743229689,
"loss": 2.648,
"theoretical_loss": 3.5593743989909727,
"tokens_seen": 1307967488
},
{
"epoch": 0.4,
"learning_rate": 0.0003049147442326981,
"loss": 2.782,
"theoretical_loss": 3.559358258949592,
"tokens_seen": 1308033024
},
{
"epoch": 0.4,
"learning_rate": 0.00030490471414242726,
"loss": 2.7467,
"theoretical_loss": 3.5593421199432633,
"tokens_seen": 1308098560
},
{
"epoch": 0.4,
"learning_rate": 0.0003048946840521565,
"loss": 2.5688,
"theoretical_loss": 3.5593259819718677,
"tokens_seen": 1308164096
},
{
"epoch": 0.4,
"learning_rate": 0.0003048846539618856,
"loss": 2.7468,
"theoretical_loss": 3.5593098450352865,
"tokens_seen": 1308229632
},
{
"epoch": 0.4,
"learning_rate": 0.00030487462387161486,
"loss": 2.7893,
"theoretical_loss": 3.559293709133402,
"tokens_seen": 1308295168
},
{
"epoch": 0.4,
"learning_rate": 0.000304864593781344,
"loss": 2.7845,
"theoretical_loss": 3.5592775742660967,
"tokens_seen": 1308360704
},
{
"epoch": 0.4,
"learning_rate": 0.0003048545636910732,
"loss": 2.565,
"theoretical_loss": 3.5592614404332514,
"tokens_seen": 1308426240
},
{
"epoch": 0.4,
"learning_rate": 0.0003048445336008024,
"loss": 2.8389,
"theoretical_loss": 3.5592453076347486,
"tokens_seen": 1308491776
},
{
"epoch": 0.4,
"learning_rate": 0.0003048345035105316,
"loss": 2.711,
"theoretical_loss": 3.55922917587047,
"tokens_seen": 1308557312
},
{
"epoch": 0.4,
"learning_rate": 0.00030482447342026076,
"loss": 2.7905,
"theoretical_loss": 3.5592130451402975,
"tokens_seen": 1308622848
},
{
"epoch": 0.4,
"learning_rate": 0.00030481444332999,
"loss": 2.8934,
"theoretical_loss": 3.5591969154441125,
"tokens_seen": 1308688384
},
{
"epoch": 0.4,
"learning_rate": 0.0003048044132397192,
"loss": 2.7367,
"theoretical_loss": 3.5591807867817984,
"tokens_seen": 1308753920
},
{
"epoch": 0.4,
"learning_rate": 0.00030479438314944836,
"loss": 2.9623,
"theoretical_loss": 3.559164659153235,
"tokens_seen": 1308819456
},
{
"epoch": 0.4,
"learning_rate": 0.00030478435305917754,
"loss": 2.8138,
"theoretical_loss": 3.5591485325583068,
"tokens_seen": 1308884992
},
{
"epoch": 0.4,
"learning_rate": 0.0003047743229689067,
"loss": 2.8908,
"theoretical_loss": 3.559132406996894,
"tokens_seen": 1308950528
},
{
"epoch": 0.4,
"learning_rate": 0.00030476429287863596,
"loss": 2.9264,
"theoretical_loss": 3.559116282468879,
"tokens_seen": 1309016064
},
{
"epoch": 0.4,
"objective/train/docs_used": 741446,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6592326164245605,
"objective/train/theoretical_loss": 3.559100158974144,
"objective/train/tokens_used": 1329541600,
"theoretical_loss": 3.559100158974144,
"tokens_seen": 1309081600
},
{
"epoch": 0.4,
"learning_rate": 0.0003047542627883651,
"loss": 2.6899,
"theoretical_loss": 3.559100158974144,
"tokens_seen": 1309081600
},
{
"epoch": 0.4,
"learning_rate": 0.0003047442326980943,
"loss": 2.7265,
"theoretical_loss": 3.559084036512571,
"tokens_seen": 1309147136
},
{
"epoch": 0.4,
"learning_rate": 0.00030473420260782345,
"loss": 2.7168,
"theoretical_loss": 3.5590679150840425,
"tokens_seen": 1309212672
},
{
"epoch": 0.4,
"learning_rate": 0.0003047241725175527,
"loss": 2.7503,
"theoretical_loss": 3.5590517946884406,
"tokens_seen": 1309278208
},
{
"epoch": 0.4,
"learning_rate": 0.00030471414242728187,
"loss": 2.6965,
"theoretical_loss": 3.559035675325646,
"tokens_seen": 1309343744
},
{
"epoch": 0.4,
"learning_rate": 0.00030470411233701105,
"loss": 2.7733,
"theoretical_loss": 3.5590195569955423,
"tokens_seen": 1309409280
},
{
"epoch": 0.4,
"learning_rate": 0.00030469408224674023,
"loss": 2.7626,
"theoretical_loss": 3.5590034396980115,
"tokens_seen": 1309474816
},
{
"epoch": 0.4,
"learning_rate": 0.00030468405215646946,
"loss": 2.8583,
"theoretical_loss": 3.5589873234329357,
"tokens_seen": 1309540352
},
{
"epoch": 0.4,
"learning_rate": 0.0003046740220661986,
"loss": 2.7252,
"theoretical_loss": 3.5589712082001963,
"tokens_seen": 1309605888
},
{
"epoch": 0.4,
"learning_rate": 0.0003046639919759278,
"loss": 3.0912,
"theoretical_loss": 3.5589550939996766,
"tokens_seen": 1309671424
},
{
"epoch": 0.4,
"learning_rate": 0.00030465396188565695,
"loss": 2.7501,
"theoretical_loss": 3.5589389808312584,
"tokens_seen": 1309736960
},
{
"epoch": 0.4,
"learning_rate": 0.0003046439317953862,
"loss": 2.8526,
"theoretical_loss": 3.5589228686948235,
"tokens_seen": 1309802496
},
{
"epoch": 0.4,
"learning_rate": 0.00030463390170511537,
"loss": 2.9377,
"theoretical_loss": 3.5589067575902553,
"tokens_seen": 1309868032
},
{
"epoch": 0.4,
"learning_rate": 0.00030462387161484455,
"loss": 2.6637,
"theoretical_loss": 3.558890647517435,
"tokens_seen": 1309933568
},
{
"epoch": 0.4,
"learning_rate": 0.00030461384152457373,
"loss": 2.6966,
"theoretical_loss": 3.5588745384762452,
"tokens_seen": 1309999104
},
{
"epoch": 0.4,
"learning_rate": 0.0003046038114343029,
"loss": 2.7003,
"theoretical_loss": 3.558858430466569,
"tokens_seen": 1310064640
},
{
"epoch": 0.4,
"learning_rate": 0.0003045937813440321,
"loss": 2.8847,
"theoretical_loss": 3.558842323488287,
"tokens_seen": 1310130176
},
{
"epoch": 0.4,
"learning_rate": 0.00030458375125376133,
"loss": 2.8235,
"theoretical_loss": 3.5588262175412835,
"tokens_seen": 1310195712
},
{
"epoch": 0.4,
"learning_rate": 0.00030457372116349046,
"loss": 2.8005,
"theoretical_loss": 3.55881011262544,
"tokens_seen": 1310261248
},
{
"epoch": 0.4,
"learning_rate": 0.0003045636910732197,
"loss": 2.7929,
"theoretical_loss": 3.558794008740639,
"tokens_seen": 1310326784
},
{
"epoch": 0.4,
"learning_rate": 0.0003045536609829488,
"loss": 2.7022,
"theoretical_loss": 3.558777905886763,
"tokens_seen": 1310392320
},
{
"epoch": 0.4,
"learning_rate": 0.00030454363089267805,
"loss": 2.8238,
"theoretical_loss": 3.5587618040636944,
"tokens_seen": 1310457856
},
{
"epoch": 0.4,
"learning_rate": 0.00030453360080240723,
"loss": 2.9138,
"theoretical_loss": 3.558745703271315,
"tokens_seen": 1310523392
},
{
"epoch": 0.4,
"learning_rate": 0.0003045235707121364,
"loss": 2.8405,
"theoretical_loss": 3.5587296035095086,
"tokens_seen": 1310588928
},
{
"epoch": 0.4,
"learning_rate": 0.0003045135406218656,
"loss": 2.8933,
"theoretical_loss": 3.558713504778156,
"tokens_seen": 1310654464
},
{
"debugging/Self-BLEU-5": 0.23052991959830066,
"debugging/distinct-1-grams": 0.8218563883058809,
"debugging/distinct-2-grams": 0.9709930029461279,
"debugging/entropy-1-grams": 5.44784936020325,
"debugging/entropy-2-grams": 5.931401419421702,
"debugging/length": 343.125,
"debugging/num_segments": 8,
"epoch": 0.4,
"objective/train/docs_used": 742066,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2548415660858154,
"objective/train/theoretical_loss": 3.558697407077142,
"objective/train/tokens_used": 1331180000,
"theoretical_loss": 3.558697407077142,
"tokens_seen": 1310720000
},
{
"epoch": 0.4,
"learning_rate": 0.00030450351053159483,
"loss": 2.6755,
"theoretical_loss": 3.558697407077142,
"tokens_seen": 1310720000
},
{
"epoch": 0.4,
"learning_rate": 0.00030449348044132396,
"loss": 2.7624,
"theoretical_loss": 3.5586813104063477,
"tokens_seen": 1310785536
},
{
"epoch": 0.4,
"learning_rate": 0.0003044834503510532,
"loss": 2.8186,
"theoretical_loss": 3.5586652147656554,
"tokens_seen": 1310851072
},
{
"epoch": 0.4,
"learning_rate": 0.0003044734202607823,
"loss": 3.0225,
"theoretical_loss": 3.558649120154948,
"tokens_seen": 1310916608
},
{
"epoch": 0.4,
"learning_rate": 0.00030446339017051156,
"loss": 2.9216,
"theoretical_loss": 3.5586330265741086,
"tokens_seen": 1310982144
},
{
"epoch": 0.4,
"learning_rate": 0.00030445336008024074,
"loss": 2.9001,
"theoretical_loss": 3.5586169340230196,
"tokens_seen": 1311047680
},
{
"epoch": 0.4,
"learning_rate": 0.0003044433299899699,
"loss": 2.6475,
"theoretical_loss": 3.5586008425015634,
"tokens_seen": 1311113216
},
{
"epoch": 0.4,
"learning_rate": 0.0003044332998996991,
"loss": 2.7204,
"theoretical_loss": 3.5585847520096228,
"tokens_seen": 1311178752
},
{
"epoch": 0.4,
"learning_rate": 0.0003044232698094283,
"loss": 3.1465,
"theoretical_loss": 3.5585686625470805,
"tokens_seen": 1311244288
},
{
"epoch": 0.4,
"learning_rate": 0.00030441323971915746,
"loss": 2.7757,
"theoretical_loss": 3.558552574113819,
"tokens_seen": 1311309824
},
{
"epoch": 0.4,
"learning_rate": 0.0003044032096288867,
"loss": 2.9264,
"theoretical_loss": 3.558536486709721,
"tokens_seen": 1311375360
},
{
"epoch": 0.4,
"learning_rate": 0.0003043931795386158,
"loss": 2.5978,
"theoretical_loss": 3.55852040033467,
"tokens_seen": 1311440896
},
{
"epoch": 0.4,
"learning_rate": 0.00030438314944834506,
"loss": 2.8787,
"theoretical_loss": 3.5585043149885474,
"tokens_seen": 1311506432
},
{
"epoch": 0.4,
"learning_rate": 0.0003043731193580742,
"loss": 2.8009,
"theoretical_loss": 3.5584882306712373,
"tokens_seen": 1311571968
},
{
"epoch": 0.4,
"learning_rate": 0.0003043630892678034,
"loss": 2.8077,
"theoretical_loss": 3.5584721473826217,
"tokens_seen": 1311637504
},
{
"epoch": 0.4,
"learning_rate": 0.0003043530591775326,
"loss": 2.9237,
"theoretical_loss": 3.5584560651225834,
"tokens_seen": 1311703040
},
{
"epoch": 0.4,
"learning_rate": 0.0003043430290872618,
"loss": 2.7887,
"theoretical_loss": 3.558439983891006,
"tokens_seen": 1311768576
},
{
"epoch": 0.4,
"learning_rate": 0.00030433299899699097,
"loss": 2.9066,
"theoretical_loss": 3.558423903687772,
"tokens_seen": 1311834112
},
{
"epoch": 0.4,
"learning_rate": 0.0003043229689067202,
"loss": 2.5477,
"theoretical_loss": 3.5584078245127637,
"tokens_seen": 1311899648
},
{
"epoch": 0.4,
"learning_rate": 0.00030431293881644933,
"loss": 2.7465,
"theoretical_loss": 3.5583917463658645,
"tokens_seen": 1311965184
},
{
"epoch": 0.4,
"learning_rate": 0.00030430290872617856,
"loss": 2.8087,
"theoretical_loss": 3.558375669246957,
"tokens_seen": 1312030720
},
{
"epoch": 0.4,
"learning_rate": 0.0003042928786359077,
"loss": 2.5034,
"theoretical_loss": 3.5583595931559246,
"tokens_seen": 1312096256
},
{
"epoch": 0.4,
"learning_rate": 0.0003042828485456369,
"loss": 2.9027,
"theoretical_loss": 3.5583435180926495,
"tokens_seen": 1312161792
},
{
"epoch": 0.4,
"learning_rate": 0.0003042728184553661,
"loss": 2.888,
"theoretical_loss": 3.558327444057016,
"tokens_seen": 1312227328
},
{
"epoch": 0.4,
"learning_rate": 0.0003042627883650953,
"loss": 2.8126,
"theoretical_loss": 3.558311371048906,
"tokens_seen": 1312292864
},
{
"epoch": 0.4,
"objective/train/docs_used": 743478,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8261537551879883,
"objective/train/theoretical_loss": 3.5582952990682024,
"objective/train/tokens_used": 1332818400,
"theoretical_loss": 3.5582952990682024,
"tokens_seen": 1312358400
},
{
"epoch": 0.4,
"learning_rate": 0.00030425275827482447,
"loss": 2.8145,
"theoretical_loss": 3.5582952990682024,
"tokens_seen": 1312358400
},
{
"epoch": 0.4,
"learning_rate": 0.00030424272818455365,
"loss": 2.6854,
"theoretical_loss": 3.5582792281147886,
"tokens_seen": 1312423936
},
{
"epoch": 0.4,
"learning_rate": 0.00030423269809428283,
"loss": 2.8365,
"theoretical_loss": 3.5582631581885478,
"tokens_seen": 1312489472
},
{
"epoch": 0.4,
"learning_rate": 0.00030422266800401207,
"loss": 2.7367,
"theoretical_loss": 3.5582470892893627,
"tokens_seen": 1312555008
},
{
"epoch": 0.4,
"learning_rate": 0.0003042126379137412,
"loss": 2.8553,
"theoretical_loss": 3.558231021417117,
"tokens_seen": 1312620544
},
{
"epoch": 0.4,
"learning_rate": 0.00030420260782347043,
"loss": 2.7497,
"theoretical_loss": 3.5582149545716932,
"tokens_seen": 1312686080
},
{
"epoch": 0.4,
"learning_rate": 0.0003041925777331996,
"loss": 2.904,
"theoretical_loss": 3.5581988887529743,
"tokens_seen": 1312751616
},
{
"epoch": 0.4,
"learning_rate": 0.0003041825476429288,
"loss": 2.4611,
"theoretical_loss": 3.558182823960844,
"tokens_seen": 1312817152
},
{
"epoch": 0.4,
"learning_rate": 0.00030417251755265797,
"loss": 2.7449,
"theoretical_loss": 3.558166760195185,
"tokens_seen": 1312882688
},
{
"epoch": 0.4,
"learning_rate": 0.00030416248746238715,
"loss": 2.747,
"theoretical_loss": 3.558150697455881,
"tokens_seen": 1312948224
},
{
"epoch": 0.4,
"learning_rate": 0.00030415245737211633,
"loss": 2.7183,
"theoretical_loss": 3.5581346357428147,
"tokens_seen": 1313013760
},
{
"epoch": 0.4,
"learning_rate": 0.00030414242728184557,
"loss": 2.9587,
"theoretical_loss": 3.5581185750558695,
"tokens_seen": 1313079296
},
{
"epoch": 0.4,
"learning_rate": 0.0003041323971915747,
"loss": 2.8059,
"theoretical_loss": 3.5581025153949284,
"tokens_seen": 1313144832
},
{
"epoch": 0.4,
"learning_rate": 0.00030412236710130393,
"loss": 2.8095,
"theoretical_loss": 3.558086456759875,
"tokens_seen": 1313210368
},
{
"epoch": 0.4,
"learning_rate": 0.00030411233701103306,
"loss": 2.7875,
"theoretical_loss": 3.5580703991505924,
"tokens_seen": 1313275904
},
{
"epoch": 0.4,
"learning_rate": 0.0003041023069207623,
"loss": 2.542,
"theoretical_loss": 3.5580543425669644,
"tokens_seen": 1313341440
},
{
"epoch": 0.4,
"learning_rate": 0.0003040922768304915,
"loss": 2.7569,
"theoretical_loss": 3.5580382870088734,
"tokens_seen": 1313406976
},
{
"epoch": 0.4,
"learning_rate": 0.00030408224674022066,
"loss": 2.6222,
"theoretical_loss": 3.5580222324762034,
"tokens_seen": 1313472512
},
{
"epoch": 0.4,
"learning_rate": 0.00030407221664994984,
"loss": 2.6309,
"theoretical_loss": 3.5580061789688378,
"tokens_seen": 1313538048
},
{
"epoch": 0.4,
"learning_rate": 0.000304062186559679,
"loss": 2.7657,
"theoretical_loss": 3.557990126486659,
"tokens_seen": 1313603584
},
{
"epoch": 0.4,
"learning_rate": 0.00030405215646940825,
"loss": 2.4174,
"theoretical_loss": 3.5579740750295517,
"tokens_seen": 1313669120
},
{
"epoch": 0.4,
"learning_rate": 0.00030404212637913743,
"loss": 2.6734,
"theoretical_loss": 3.5579580245973985,
"tokens_seen": 1313734656
},
{
"epoch": 0.4,
"learning_rate": 0.0003040320962888666,
"loss": 2.6854,
"theoretical_loss": 3.557941975190083,
"tokens_seen": 1313800192
},
{
"epoch": 0.4,
"learning_rate": 0.0003040220661985958,
"loss": 2.7697,
"theoretical_loss": 3.5579259268074885,
"tokens_seen": 1313865728
},
{
"epoch": 0.4,
"learning_rate": 0.00030401203610832503,
"loss": 2.516,
"theoretical_loss": 3.557909879449499,
"tokens_seen": 1313931264
},
{
"epoch": 0.4,
"objective/train/docs_used": 744202,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7266733646392822,
"objective/train/theoretical_loss": 3.5578938331159975,
"objective/train/tokens_used": 1334456800,
"theoretical_loss": 3.5578938331159975,
"tokens_seen": 1313996800
},
{
"epoch": 0.4,
"learning_rate": 0.00030400200601805416,
"loss": 2.8459,
"theoretical_loss": 3.5578938331159975,
"tokens_seen": 1313996800
},
{
"epoch": 0.4,
"learning_rate": 0.0003039919759277834,
"loss": 2.8271,
"theoretical_loss": 3.5578777878068677,
"tokens_seen": 1314062336
},
{
"epoch": 0.4,
"learning_rate": 0.0003039819458375125,
"loss": 2.8784,
"theoretical_loss": 3.5578617435219932,
"tokens_seen": 1314127872
},
{
"epoch": 0.4,
"learning_rate": 0.00030397191574724176,
"loss": 2.6871,
"theoretical_loss": 3.557845700261257,
"tokens_seen": 1314193408
},
{
"epoch": 0.4,
"learning_rate": 0.00030396188565697094,
"loss": 2.8582,
"theoretical_loss": 3.557829658024543,
"tokens_seen": 1314258944
},
{
"epoch": 0.4,
"learning_rate": 0.0003039518555667001,
"loss": 2.7413,
"theoretical_loss": 3.5578136168117354,
"tokens_seen": 1314324480
},
{
"epoch": 0.4,
"learning_rate": 0.0003039418254764293,
"loss": 2.8306,
"theoretical_loss": 3.5577975766227166,
"tokens_seen": 1314390016
},
{
"epoch": 0.4,
"learning_rate": 0.0003039317953861585,
"loss": 2.6711,
"theoretical_loss": 3.5577815374573714,
"tokens_seen": 1314455552
},
{
"epoch": 0.4,
"learning_rate": 0.00030392176529588766,
"loss": 2.7744,
"theoretical_loss": 3.5577654993155825,
"tokens_seen": 1314521088
},
{
"epoch": 0.4,
"learning_rate": 0.0003039117352056169,
"loss": 2.6377,
"theoretical_loss": 3.5577494621972345,
"tokens_seen": 1314586624
},
{
"epoch": 0.4,
"learning_rate": 0.000303901705115346,
"loss": 2.8283,
"theoretical_loss": 3.55773342610221,
"tokens_seen": 1314652160
},
{
"epoch": 0.4,
"learning_rate": 0.00030389167502507526,
"loss": 2.4819,
"theoretical_loss": 3.557717391030393,
"tokens_seen": 1314717696
},
{
"epoch": 0.4,
"learning_rate": 0.0003038816449348044,
"loss": 2.504,
"theoretical_loss": 3.557701356981668,
"tokens_seen": 1314783232
},
{
"epoch": 0.4,
"learning_rate": 0.0003038716148445336,
"loss": 2.8657,
"theoretical_loss": 3.5576853239559183,
"tokens_seen": 1314848768
},
{
"epoch": 0.4,
"learning_rate": 0.0003038615847542628,
"loss": 2.6859,
"theoretical_loss": 3.557669291953027,
"tokens_seen": 1314914304
},
{
"epoch": 0.4,
"learning_rate": 0.000303851554663992,
"loss": 2.6601,
"theoretical_loss": 3.5576532609728786,
"tokens_seen": 1314979840
},
{
"epoch": 0.4,
"learning_rate": 0.00030384152457372117,
"loss": 2.6658,
"theoretical_loss": 3.557637231015357,
"tokens_seen": 1315045376
},
{
"epoch": 0.4,
"learning_rate": 0.0003038314944834504,
"loss": 2.8509,
"theoretical_loss": 3.5576212020803446,
"tokens_seen": 1315110912
},
{
"epoch": 0.4,
"learning_rate": 0.00030382146439317953,
"loss": 2.6027,
"theoretical_loss": 3.557605174167727,
"tokens_seen": 1315176448
},
{
"epoch": 0.4,
"learning_rate": 0.00030381143430290876,
"loss": 2.6963,
"theoretical_loss": 3.5575891472773877,
"tokens_seen": 1315241984
},
{
"epoch": 0.4,
"learning_rate": 0.0003038014042126379,
"loss": 2.6856,
"theoretical_loss": 3.5575731214092094,
"tokens_seen": 1315307520
},
{
"epoch": 0.4,
"learning_rate": 0.0003037913741223671,
"loss": 2.6077,
"theoretical_loss": 3.5575570965630776,
"tokens_seen": 1315373056
},
{
"epoch": 0.4,
"learning_rate": 0.0003037813440320963,
"loss": 2.7655,
"theoretical_loss": 3.557541072738875,
"tokens_seen": 1315438592
},
{
"epoch": 0.4,
"learning_rate": 0.0003037713139418255,
"loss": 2.5562,
"theoretical_loss": 3.557525049936486,
"tokens_seen": 1315504128
},
{
"epoch": 0.4,
"learning_rate": 0.00030376128385155467,
"loss": 2.7195,
"theoretical_loss": 3.5575090281557946,
"tokens_seen": 1315569664
},
{
"epoch": 0.4,
"objective/train/docs_used": 745555,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8993453979492188,
"objective/train/theoretical_loss": 3.557493007396684,
"objective/train/tokens_used": 1336095200,
"theoretical_loss": 3.557493007396684,
"tokens_seen": 1315635200
},
{
"epoch": 0.4,
"learning_rate": 0.00030375125376128385,
"loss": 2.7302,
"theoretical_loss": 3.557493007396684,
"tokens_seen": 1315635200
},
{
"epoch": 0.4,
"learning_rate": 0.00030374122367101303,
"loss": 2.6425,
"theoretical_loss": 3.5574769876590397,
"tokens_seen": 1315700736
},
{
"epoch": 0.4,
"learning_rate": 0.00030373119358074227,
"loss": 2.8214,
"theoretical_loss": 3.5574609689427446,
"tokens_seen": 1315766272
},
{
"epoch": 0.4,
"learning_rate": 0.0003037211634904714,
"loss": 2.7795,
"theoretical_loss": 3.5574449512476822,
"tokens_seen": 1315831808
},
{
"epoch": 0.4,
"learning_rate": 0.00030371113340020063,
"loss": 2.7352,
"theoretical_loss": 3.5574289345737378,
"tokens_seen": 1315897344
},
{
"epoch": 0.4,
"learning_rate": 0.0003037011033099298,
"loss": 2.7978,
"theoretical_loss": 3.557412918920795,
"tokens_seen": 1315962880
},
{
"epoch": 0.4,
"learning_rate": 0.000303691073219659,
"loss": 2.5843,
"theoretical_loss": 3.5573969042887374,
"tokens_seen": 1316028416
},
{
"epoch": 0.4,
"learning_rate": 0.00030368104312938817,
"loss": 2.8484,
"theoretical_loss": 3.55738089067745,
"tokens_seen": 1316093952
},
{
"epoch": 0.4,
"learning_rate": 0.00030367101303911735,
"loss": 2.4635,
"theoretical_loss": 3.557364878086816,
"tokens_seen": 1316159488
},
{
"epoch": 0.4,
"learning_rate": 0.00030366098294884653,
"loss": 2.6768,
"theoretical_loss": 3.55734886651672,
"tokens_seen": 1316225024
},
{
"epoch": 0.4,
"learning_rate": 0.00030365095285857577,
"loss": 2.7476,
"theoretical_loss": 3.5573328559670463,
"tokens_seen": 1316290560
},
{
"epoch": 0.4,
"learning_rate": 0.0003036409227683049,
"loss": 2.8013,
"theoretical_loss": 3.5573168464376783,
"tokens_seen": 1316356096
},
{
"epoch": 0.4,
"learning_rate": 0.00030363089267803413,
"loss": 2.8145,
"theoretical_loss": 3.557300837928501,
"tokens_seen": 1316421632
},
{
"epoch": 0.4,
"learning_rate": 0.00030362086258776326,
"loss": 2.671,
"theoretical_loss": 3.5572848304393982,
"tokens_seen": 1316487168
},
{
"epoch": 0.4,
"learning_rate": 0.0003036108324974925,
"loss": 2.5876,
"theoretical_loss": 3.5572688239702543,
"tokens_seen": 1316552704
},
{
"epoch": 0.4,
"learning_rate": 0.0003036008024072217,
"loss": 2.947,
"theoretical_loss": 3.5572528185209538,
"tokens_seen": 1316618240
},
{
"epoch": 0.4,
"learning_rate": 0.00030359077231695086,
"loss": 2.8099,
"theoretical_loss": 3.55723681409138,
"tokens_seen": 1316683776
},
{
"epoch": 0.4,
"learning_rate": 0.00030358074222668004,
"loss": 2.6691,
"theoretical_loss": 3.5572208106814185,
"tokens_seen": 1316749312
},
{
"epoch": 0.4,
"learning_rate": 0.0003035707121364092,
"loss": 2.7003,
"theoretical_loss": 3.557204808290953,
"tokens_seen": 1316814848
},
{
"epoch": 0.4,
"learning_rate": 0.0003035606820461384,
"loss": 2.6908,
"theoretical_loss": 3.557188806919867,
"tokens_seen": 1316880384
},
{
"epoch": 0.4,
"learning_rate": 0.00030355065195586763,
"loss": 2.6948,
"theoretical_loss": 3.5571728065680457,
"tokens_seen": 1316945920
},
{
"epoch": 0.4,
"learning_rate": 0.00030354062186559676,
"loss": 2.7362,
"theoretical_loss": 3.5571568072353736,
"tokens_seen": 1317011456
},
{
"epoch": 0.4,
"learning_rate": 0.000303530591775326,
"loss": 2.6517,
"theoretical_loss": 3.5571408089217353,
"tokens_seen": 1317076992
},
{
"epoch": 0.4,
"learning_rate": 0.0003035205616850552,
"loss": 2.7791,
"theoretical_loss": 3.557124811627014,
"tokens_seen": 1317142528
},
{
"epoch": 0.4,
"learning_rate": 0.00030351053159478436,
"loss": 2.8295,
"theoretical_loss": 3.5571088153510955,
"tokens_seen": 1317208064
},
{
"epoch": 0.4,
"objective/train/docs_used": 746236,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6158335208892822,
"objective/train/theoretical_loss": 3.557092820093863,
"objective/train/tokens_used": 1337733600,
"theoretical_loss": 3.557092820093863,
"tokens_seen": 1317273600
},
{
"epoch": 0.4,
"learning_rate": 0.00030350050150451354,
"loss": 2.4992,
"theoretical_loss": 3.557092820093863,
"tokens_seen": 1317273600
},
{
"epoch": 0.4,
"learning_rate": 0.0003034904714142427,
"loss": 2.7621,
"theoretical_loss": 3.5570768258552015,
"tokens_seen": 1317339136
},
{
"epoch": 0.4,
"learning_rate": 0.0003034804413239719,
"loss": 2.9166,
"theoretical_loss": 3.5570608326349955,
"tokens_seen": 1317404672
},
{
"epoch": 0.4,
"learning_rate": 0.00030347041123370114,
"loss": 2.6463,
"theoretical_loss": 3.5570448404331296,
"tokens_seen": 1317470208
},
{
"epoch": 0.4,
"learning_rate": 0.00030346038114343026,
"loss": 2.7451,
"theoretical_loss": 3.5570288492494884,
"tokens_seen": 1317535744
},
{
"epoch": 0.4,
"learning_rate": 0.0003034503510531595,
"loss": 2.7115,
"theoretical_loss": 3.5570128590839563,
"tokens_seen": 1317601280
},
{
"epoch": 0.4,
"learning_rate": 0.0003034403209628886,
"loss": 2.6365,
"theoretical_loss": 3.5569968699364174,
"tokens_seen": 1317666816
},
{
"epoch": 0.4,
"learning_rate": 0.00030343029087261786,
"loss": 2.4589,
"theoretical_loss": 3.5569808818067568,
"tokens_seen": 1317732352
},
{
"epoch": 0.4,
"learning_rate": 0.00030342026078234704,
"loss": 2.5788,
"theoretical_loss": 3.5569648946948584,
"tokens_seen": 1317797888
},
{
"epoch": 0.4,
"learning_rate": 0.0003034102306920762,
"loss": 2.6506,
"theoretical_loss": 3.556948908600608,
"tokens_seen": 1317863424
},
{
"epoch": 0.4,
"learning_rate": 0.0003034002006018054,
"loss": 2.6908,
"theoretical_loss": 3.556932923523889,
"tokens_seen": 1317928960
},
{
"epoch": 0.4,
"learning_rate": 0.0003033901705115346,
"loss": 2.9564,
"theoretical_loss": 3.556916939464587,
"tokens_seen": 1317994496
},
{
"epoch": 0.4,
"learning_rate": 0.00030338014042126377,
"loss": 2.7427,
"theoretical_loss": 3.556900956422586,
"tokens_seen": 1318060032
},
{
"epoch": 0.4,
"learning_rate": 0.000303370110330993,
"loss": 2.73,
"theoretical_loss": 3.5568849743977706,
"tokens_seen": 1318125568
},
{
"epoch": 0.4,
"learning_rate": 0.00030336008024072213,
"loss": 2.7546,
"theoretical_loss": 3.5568689933900264,
"tokens_seen": 1318191104
},
{
"epoch": 0.4,
"learning_rate": 0.00030335005015045137,
"loss": 2.5373,
"theoretical_loss": 3.5568530133992375,
"tokens_seen": 1318256640
},
{
"epoch": 0.4,
"learning_rate": 0.00030334002006018055,
"loss": 2.7849,
"theoretical_loss": 3.556837034425288,
"tokens_seen": 1318322176
},
{
"epoch": 0.4,
"learning_rate": 0.00030332998996990973,
"loss": 2.6679,
"theoretical_loss": 3.556821056468064,
"tokens_seen": 1318387712
},
{
"epoch": 0.4,
"learning_rate": 0.0003033199598796389,
"loss": 2.8725,
"theoretical_loss": 3.5568050795274493,
"tokens_seen": 1318453248
},
{
"epoch": 0.4,
"learning_rate": 0.0003033099297893681,
"loss": 2.6282,
"theoretical_loss": 3.556789103603329,
"tokens_seen": 1318518784
},
{
"epoch": 0.4,
"learning_rate": 0.0003032998996990973,
"loss": 2.8236,
"theoretical_loss": 3.5567731286955873,
"tokens_seen": 1318584320
},
{
"epoch": 0.4,
"learning_rate": 0.0003032898696088265,
"loss": 2.6777,
"theoretical_loss": 3.55675715480411,
"tokens_seen": 1318649856
},
{
"epoch": 0.4,
"learning_rate": 0.0003032798395185557,
"loss": 2.5345,
"theoretical_loss": 3.5567411819287815,
"tokens_seen": 1318715392
},
{
"epoch": 0.4,
"learning_rate": 0.00030326980942828487,
"loss": 2.7409,
"theoretical_loss": 3.556725210069487,
"tokens_seen": 1318780928
},
{
"epoch": 0.4,
"learning_rate": 0.00030325977933801405,
"loss": 2.8074,
"theoretical_loss": 3.5567092392261106,
"tokens_seen": 1318846464
},
{
"epoch": 0.4,
"objective/train/docs_used": 747487,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.791353464126587,
"objective/train/theoretical_loss": 3.5566932693985382,
"objective/train/tokens_used": 1339372000,
"theoretical_loss": 3.5566932693985382,
"tokens_seen": 1318912000
},
{
"epoch": 0.4,
"learning_rate": 0.00030324974924774323,
"loss": 2.7433,
"theoretical_loss": 3.5566932693985382,
"tokens_seen": 1318912000
},
{
"epoch": 0.4,
"learning_rate": 0.00030323971915747247,
"loss": 2.7284,
"theoretical_loss": 3.556677300586654,
"tokens_seen": 1318977536
},
{
"epoch": 0.4,
"learning_rate": 0.0003032296890672016,
"loss": 2.9265,
"theoretical_loss": 3.556661332790343,
"tokens_seen": 1319043072
},
{
"epoch": 0.4,
"learning_rate": 0.00030321965897693083,
"loss": 2.8113,
"theoretical_loss": 3.55664536600949,
"tokens_seen": 1319108608
},
{
"epoch": 0.4,
"learning_rate": 0.00030320962888666,
"loss": 2.8172,
"theoretical_loss": 3.556629400243981,
"tokens_seen": 1319174144
},
{
"epoch": 0.4,
"learning_rate": 0.0003031995987963892,
"loss": 2.8449,
"theoretical_loss": 3.5566134354937,
"tokens_seen": 1319239680
},
{
"epoch": 0.4,
"learning_rate": 0.00030318956870611837,
"loss": 2.7369,
"theoretical_loss": 3.5565974717585322,
"tokens_seen": 1319305216
},
{
"epoch": 0.4,
"learning_rate": 0.00030317953861584755,
"loss": 2.9892,
"theoretical_loss": 3.556581509038363,
"tokens_seen": 1319370752
},
{
"epoch": 0.4,
"learning_rate": 0.00030316950852557673,
"loss": 2.9343,
"theoretical_loss": 3.5565655473330766,
"tokens_seen": 1319436288
},
{
"epoch": 0.4,
"learning_rate": 0.00030315947843530597,
"loss": 2.6272,
"theoretical_loss": 3.556549586642559,
"tokens_seen": 1319501824
},
{
"epoch": 0.4,
"learning_rate": 0.0003031494483450351,
"loss": 2.6829,
"theoretical_loss": 3.5565336269666954,
"tokens_seen": 1319567360
},
{
"epoch": 0.4,
"learning_rate": 0.00030313941825476433,
"loss": 2.8001,
"theoretical_loss": 3.5565176683053696,
"tokens_seen": 1319632896
},
{
"epoch": 0.4,
"learning_rate": 0.00030312938816449346,
"loss": 2.8549,
"theoretical_loss": 3.5565017106584684,
"tokens_seen": 1319698432
},
{
"epoch": 0.4,
"learning_rate": 0.0003031193580742227,
"loss": 2.91,
"theoretical_loss": 3.556485754025876,
"tokens_seen": 1319763968
},
{
"epoch": 0.4,
"learning_rate": 0.0003031093279839519,
"loss": 2.6955,
"theoretical_loss": 3.556469798407477,
"tokens_seen": 1319829504
},
{
"epoch": 0.4,
"learning_rate": 0.00030309929789368106,
"loss": 2.7869,
"theoretical_loss": 3.556453843803158,
"tokens_seen": 1319895040
},
{
"epoch": 0.4,
"learning_rate": 0.00030308926780341024,
"loss": 2.5101,
"theoretical_loss": 3.556437890212803,
"tokens_seen": 1319960576
},
{
"epoch": 0.4,
"learning_rate": 0.0003030792377131394,
"loss": 2.664,
"theoretical_loss": 3.5564219376362978,
"tokens_seen": 1320026112
},
{
"epoch": 0.4,
"learning_rate": 0.0003030692076228686,
"loss": 2.8819,
"theoretical_loss": 3.5564059860735275,
"tokens_seen": 1320091648
},
{
"epoch": 0.4,
"learning_rate": 0.00030305917753259783,
"loss": 2.754,
"theoretical_loss": 3.5563900355243776,
"tokens_seen": 1320157184
},
{
"epoch": 0.4,
"learning_rate": 0.00030304914744232696,
"loss": 2.7626,
"theoretical_loss": 3.5563740859887334,
"tokens_seen": 1320222720
},
{
"epoch": 0.4,
"learning_rate": 0.0003030391173520562,
"loss": 2.6749,
"theoretical_loss": 3.5563581374664794,
"tokens_seen": 1320288256
},
{
"epoch": 0.4,
"learning_rate": 0.0003030290872617854,
"loss": 2.639,
"theoretical_loss": 3.556342189957502,
"tokens_seen": 1320353792
},
{
"epoch": 0.4,
"learning_rate": 0.00030301905717151456,
"loss": 2.522,
"theoretical_loss": 3.556326243461686,
"tokens_seen": 1320419328
},
{
"epoch": 0.4,
"learning_rate": 0.00030300902708124374,
"loss": 2.5591,
"theoretical_loss": 3.556310297978916,
"tokens_seen": 1320484864
},
{
"epoch": 0.4,
"objective/train/docs_used": 748180,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4906086921691895,
"objective/train/theoretical_loss": 3.556294353509079,
"objective/train/tokens_used": 1341010400,
"theoretical_loss": 3.556294353509079,
"tokens_seen": 1320550400
},
{
"epoch": 0.4,
"learning_rate": 0.0003029989969909729,
"loss": 2.6062,
"theoretical_loss": 3.556294353509079,
"tokens_seen": 1320550400
},
{
"epoch": 0.4,
"learning_rate": 0.0003029889669007021,
"loss": 2.7192,
"theoretical_loss": 3.5562784100520592,
"tokens_seen": 1320615936
},
{
"epoch": 0.4,
"learning_rate": 0.00030297893681043134,
"loss": 2.8824,
"theoretical_loss": 3.556262467607742,
"tokens_seen": 1320681472
},
{
"epoch": 0.4,
"learning_rate": 0.00030296890672016046,
"loss": 2.7495,
"theoretical_loss": 3.5562465261760137,
"tokens_seen": 1320747008
},
{
"epoch": 0.4,
"learning_rate": 0.0003029588766298897,
"loss": 2.8335,
"theoretical_loss": 3.556230585756759,
"tokens_seen": 1320812544
},
{
"epoch": 0.4,
"learning_rate": 0.0003029488465396188,
"loss": 2.6312,
"theoretical_loss": 3.556214646349864,
"tokens_seen": 1320878080
},
{
"epoch": 0.4,
"learning_rate": 0.00030293881644934806,
"loss": 2.5628,
"theoretical_loss": 3.5561987079552133,
"tokens_seen": 1320943616
},
{
"epoch": 0.4,
"learning_rate": 0.00030292878635907724,
"loss": 2.6942,
"theoretical_loss": 3.5561827705726925,
"tokens_seen": 1321009152
},
{
"epoch": 0.4,
"learning_rate": 0.0003029187562688064,
"loss": 2.4939,
"theoretical_loss": 3.5561668342021884,
"tokens_seen": 1321074688
},
{
"epoch": 0.4,
"learning_rate": 0.0003029087261785356,
"loss": 2.891,
"theoretical_loss": 3.556150898843585,
"tokens_seen": 1321140224
},
{
"epoch": 0.4,
"learning_rate": 0.0003028986960882648,
"loss": 2.7296,
"theoretical_loss": 3.556134964496769,
"tokens_seen": 1321205760
},
{
"epoch": 0.4,
"learning_rate": 0.00030288866599799397,
"loss": 2.697,
"theoretical_loss": 3.556119031161624,
"tokens_seen": 1321271296
},
{
"epoch": 0.4,
"learning_rate": 0.0003028786359077232,
"loss": 2.8246,
"theoretical_loss": 3.5561030988380384,
"tokens_seen": 1321336832
},
{
"epoch": 0.4,
"learning_rate": 0.00030286860581745233,
"loss": 2.4882,
"theoretical_loss": 3.556087167525896,
"tokens_seen": 1321402368
},
{
"epoch": 0.4,
"learning_rate": 0.00030285857572718157,
"loss": 2.7594,
"theoretical_loss": 3.5560712372250833,
"tokens_seen": 1321467904
},
{
"epoch": 0.4,
"learning_rate": 0.00030284854563691075,
"loss": 2.7012,
"theoretical_loss": 3.556055307935485,
"tokens_seen": 1321533440
},
{
"epoch": 0.4,
"learning_rate": 0.00030283851554663993,
"loss": 2.9405,
"theoretical_loss": 3.556039379656988,
"tokens_seen": 1321598976
},
{
"epoch": 0.4,
"learning_rate": 0.0003028284854563691,
"loss": 2.6646,
"theoretical_loss": 3.5560234523894763,
"tokens_seen": 1321664512
},
{
"epoch": 0.4,
"learning_rate": 0.0003028184553660983,
"loss": 2.6296,
"theoretical_loss": 3.556007526132837,
"tokens_seen": 1321730048
},
{
"epoch": 0.4,
"learning_rate": 0.00030280842527582747,
"loss": 2.7891,
"theoretical_loss": 3.555991600886956,
"tokens_seen": 1321795584
},
{
"epoch": 0.4,
"learning_rate": 0.0003027983951855567,
"loss": 2.6449,
"theoretical_loss": 3.555975676651718,
"tokens_seen": 1321861120
},
{
"epoch": 0.4,
"learning_rate": 0.00030278836509528583,
"loss": 2.801,
"theoretical_loss": 3.555959753427009,
"tokens_seen": 1321926656
},
{
"epoch": 0.4,
"learning_rate": 0.00030277833500501507,
"loss": 2.656,
"theoretical_loss": 3.5559438312127147,
"tokens_seen": 1321992192
},
{
"epoch": 0.4,
"learning_rate": 0.0003027683049147442,
"loss": 2.7042,
"theoretical_loss": 3.555927910008722,
"tokens_seen": 1322057728
},
{
"epoch": 0.4,
"learning_rate": 0.00030275827482447343,
"loss": 2.7992,
"theoretical_loss": 3.5559119898149154,
"tokens_seen": 1322123264
},
{
"epoch": 0.4,
"objective/train/docs_used": 749233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.753972291946411,
"objective/train/theoretical_loss": 3.5558960706311815,
"objective/train/tokens_used": 1342648800,
"theoretical_loss": 3.5558960706311815,
"tokens_seen": 1322188800
},
{
"epoch": 0.4,
"learning_rate": 0.0003027482447342026,
"loss": 2.7698,
"theoretical_loss": 3.5558960706311815,
"tokens_seen": 1322188800
},
{
"epoch": 0.4,
"learning_rate": 0.0003027382146439318,
"loss": 2.526,
"theoretical_loss": 3.555880152457406,
"tokens_seen": 1322254336
},
{
"epoch": 0.4,
"learning_rate": 0.000302728184553661,
"loss": 2.5035,
"theoretical_loss": 3.555864235293474,
"tokens_seen": 1322319872
},
{
"epoch": 0.4,
"learning_rate": 0.0003027181544633902,
"loss": 2.792,
"theoretical_loss": 3.5558483191392734,
"tokens_seen": 1322385408
},
{
"epoch": 0.4,
"learning_rate": 0.00030270812437311934,
"loss": 2.8171,
"theoretical_loss": 3.5558324039946876,
"tokens_seen": 1322450944
},
{
"epoch": 0.4,
"learning_rate": 0.00030269809428284857,
"loss": 2.6753,
"theoretical_loss": 3.555816489859604,
"tokens_seen": 1322516480
},
{
"epoch": 0.4,
"learning_rate": 0.0003026880641925777,
"loss": 2.5654,
"theoretical_loss": 3.5558005767339083,
"tokens_seen": 1322582016
},
{
"epoch": 0.4,
"learning_rate": 0.00030267803410230693,
"loss": 2.4443,
"theoretical_loss": 3.5557846646174864,
"tokens_seen": 1322647552
},
{
"epoch": 0.4,
"learning_rate": 0.0003026680040120361,
"loss": 2.8719,
"theoretical_loss": 3.5557687535102245,
"tokens_seen": 1322713088
},
{
"epoch": 0.4,
"learning_rate": 0.0003026579739217653,
"loss": 2.6304,
"theoretical_loss": 3.5557528434120083,
"tokens_seen": 1322778624
},
{
"epoch": 0.4,
"learning_rate": 0.0003026479438314945,
"loss": 2.8115,
"theoretical_loss": 3.5557369343227236,
"tokens_seen": 1322844160
},
{
"epoch": 0.4,
"learning_rate": 0.00030263791374122366,
"loss": 2.9638,
"theoretical_loss": 3.5557210262422574,
"tokens_seen": 1322909696
},
{
"epoch": 0.4,
"learning_rate": 0.00030262788365095284,
"loss": 2.7404,
"theoretical_loss": 3.5557051191704945,
"tokens_seen": 1322975232
},
{
"epoch": 0.4,
"learning_rate": 0.0003026178535606821,
"loss": 2.631,
"theoretical_loss": 3.555689213107322,
"tokens_seen": 1323040768
},
{
"epoch": 0.4,
"learning_rate": 0.0003026078234704112,
"loss": 2.7413,
"theoretical_loss": 3.5556733080526257,
"tokens_seen": 1323106304
},
{
"epoch": 0.4,
"learning_rate": 0.00030259779338014044,
"loss": 2.6749,
"theoretical_loss": 3.5556574040062916,
"tokens_seen": 1323171840
},
{
"epoch": 0.4,
"learning_rate": 0.00030258776328986956,
"loss": 2.6144,
"theoretical_loss": 3.555641500968205,
"tokens_seen": 1323237376
},
{
"epoch": 0.4,
"learning_rate": 0.0003025777331995988,
"loss": 2.7005,
"theoretical_loss": 3.5556255989382537,
"tokens_seen": 1323302912
},
{
"epoch": 0.4,
"learning_rate": 0.000302567703109328,
"loss": 2.4197,
"theoretical_loss": 3.5556096979163234,
"tokens_seen": 1323368448
},
{
"epoch": 0.4,
"learning_rate": 0.00030255767301905716,
"loss": 2.757,
"theoretical_loss": 3.555593797902299,
"tokens_seen": 1323433984
},
{
"epoch": 0.4,
"learning_rate": 0.0003025476429287864,
"loss": 2.9548,
"theoretical_loss": 3.5555778988960682,
"tokens_seen": 1323499520
},
{
"epoch": 0.4,
"learning_rate": 0.0003025376128385156,
"loss": 2.8712,
"theoretical_loss": 3.555562000897517,
"tokens_seen": 1323565056
},
{
"epoch": 0.4,
"learning_rate": 0.00030252758274824476,
"loss": 2.752,
"theoretical_loss": 3.5555461039065306,
"tokens_seen": 1323630592
},
{
"epoch": 0.4,
"learning_rate": 0.00030251755265797394,
"loss": 2.877,
"theoretical_loss": 3.5555302079229962,
"tokens_seen": 1323696128
},
{
"epoch": 0.4,
"learning_rate": 0.0003025075225677031,
"loss": 2.7961,
"theoretical_loss": 3.5555143129467996,
"tokens_seen": 1323761664
},
{
"epoch": 0.4,
"objective/train/docs_used": 749806,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8539295196533203,
"objective/train/theoretical_loss": 3.555498418977828,
"objective/train/tokens_used": 1344287200,
"theoretical_loss": 3.555498418977828,
"tokens_seen": 1323827200
},
{
"epoch": 0.4,
"learning_rate": 0.0003024974924774323,
"loss": 2.5934,
"theoretical_loss": 3.555498418977828,
"tokens_seen": 1323827200
},
{
"epoch": 0.4,
"learning_rate": 0.00030248746238716154,
"loss": 2.6769,
"theoretical_loss": 3.555482526015967,
"tokens_seen": 1323892736
},
{
"epoch": 0.4,
"learning_rate": 0.00030247743229689066,
"loss": 2.7546,
"theoretical_loss": 3.5554666340611023,
"tokens_seen": 1323958272
},
{
"epoch": 0.4,
"learning_rate": 0.0003024674022066199,
"loss": 2.895,
"theoretical_loss": 3.5554507431131217,
"tokens_seen": 1324023808
},
{
"epoch": 0.4,
"learning_rate": 0.00030245737211634903,
"loss": 2.7721,
"theoretical_loss": 3.555434853171911,
"tokens_seen": 1324089344
},
{
"epoch": 0.4,
"learning_rate": 0.00030244734202607826,
"loss": 2.6975,
"theoretical_loss": 3.555418964237356,
"tokens_seen": 1324154880
},
{
"epoch": 0.4,
"learning_rate": 0.00030243731193580744,
"loss": 2.8426,
"theoretical_loss": 3.5554030763093434,
"tokens_seen": 1324220416
},
{
"epoch": 0.4,
"learning_rate": 0.0003024272818455366,
"loss": 2.765,
"theoretical_loss": 3.55538718938776,
"tokens_seen": 1324285952
},
{
"epoch": 0.4,
"learning_rate": 0.0003024172517552658,
"loss": 2.7178,
"theoretical_loss": 3.555371303472492,
"tokens_seen": 1324351488
},
{
"epoch": 0.4,
"learning_rate": 0.000302407221664995,
"loss": 2.8108,
"theoretical_loss": 3.555355418563426,
"tokens_seen": 1324417024
},
{
"epoch": 0.4,
"learning_rate": 0.00030239719157472417,
"loss": 2.6449,
"theoretical_loss": 3.5553395346604484,
"tokens_seen": 1324482560
},
{
"epoch": 0.4,
"learning_rate": 0.0003023871614844534,
"loss": 2.5653,
"theoretical_loss": 3.5553236517634454,
"tokens_seen": 1324548096
},
{
"epoch": 0.4,
"learning_rate": 0.00030237713139418253,
"loss": 2.7724,
"theoretical_loss": 3.5553077698723037,
"tokens_seen": 1324613632
},
{
"epoch": 0.4,
"learning_rate": 0.00030236710130391177,
"loss": 2.7362,
"theoretical_loss": 3.5552918889869103,
"tokens_seen": 1324679168
},
{
"epoch": 0.4,
"learning_rate": 0.00030235707121364095,
"loss": 2.7352,
"theoretical_loss": 3.5552760091071507,
"tokens_seen": 1324744704
},
{
"epoch": 0.4,
"learning_rate": 0.00030234704112337013,
"loss": 2.8404,
"theoretical_loss": 3.555260130232913,
"tokens_seen": 1324810240
},
{
"epoch": 0.4,
"learning_rate": 0.0003023370110330993,
"loss": 2.5112,
"theoretical_loss": 3.5552442523640826,
"tokens_seen": 1324875776
},
{
"epoch": 0.4,
"learning_rate": 0.0003023269809428285,
"loss": 2.753,
"theoretical_loss": 3.5552283755005463,
"tokens_seen": 1324941312
},
{
"epoch": 0.4,
"learning_rate": 0.00030231695085255767,
"loss": 2.7132,
"theoretical_loss": 3.5552124996421908,
"tokens_seen": 1325006848
},
{
"epoch": 0.4,
"learning_rate": 0.0003023069207622869,
"loss": 2.6915,
"theoretical_loss": 3.5551966247889033,
"tokens_seen": 1325072384
},
{
"epoch": 0.4,
"learning_rate": 0.00030229689067201603,
"loss": 2.8898,
"theoretical_loss": 3.5551807509405697,
"tokens_seen": 1325137920
},
{
"epoch": 0.4,
"learning_rate": 0.00030228686058174527,
"loss": 2.5311,
"theoretical_loss": 3.5551648780970773,
"tokens_seen": 1325203456
},
{
"epoch": 0.4,
"learning_rate": 0.0003022768304914744,
"loss": 2.795,
"theoretical_loss": 3.555149006258312,
"tokens_seen": 1325268992
},
{
"epoch": 0.4,
"learning_rate": 0.00030226680040120363,
"loss": 2.8633,
"theoretical_loss": 3.5551331354241613,
"tokens_seen": 1325334528
},
{
"epoch": 0.4,
"learning_rate": 0.0003022567703109328,
"loss": 2.6868,
"theoretical_loss": 3.5551172655945114,
"tokens_seen": 1325400064
},
{
"epoch": 0.4,
"objective/train/docs_used": 750658,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7877793312072754,
"objective/train/theoretical_loss": 3.55510139676925,
"objective/train/tokens_used": 1345925600,
"theoretical_loss": 3.55510139676925,
"tokens_seen": 1325465600
},
{
"epoch": 0.4,
"learning_rate": 0.000302246740220662,
"loss": 2.7724,
"theoretical_loss": 3.55510139676925,
"tokens_seen": 1325465600
},
{
"epoch": 0.4,
"learning_rate": 0.0003022367101303912,
"loss": 2.7237,
"theoretical_loss": 3.555085528948262,
"tokens_seen": 1325531136
},
{
"epoch": 0.4,
"learning_rate": 0.0003022266800401204,
"loss": 2.7723,
"theoretical_loss": 3.5550696621314364,
"tokens_seen": 1325596672
},
{
"epoch": 0.4,
"learning_rate": 0.00030221664994984954,
"loss": 2.8545,
"theoretical_loss": 3.555053796318659,
"tokens_seen": 1325662208
},
{
"epoch": 0.4,
"learning_rate": 0.00030220661985957877,
"loss": 2.5729,
"theoretical_loss": 3.5550379315098164,
"tokens_seen": 1325727744
},
{
"epoch": 0.4,
"learning_rate": 0.0003021965897693079,
"loss": 2.7035,
"theoretical_loss": 3.5550220677047952,
"tokens_seen": 1325793280
},
{
"epoch": 0.4,
"learning_rate": 0.00030218655967903713,
"loss": 2.8236,
"theoretical_loss": 3.555006204903483,
"tokens_seen": 1325858816
},
{
"epoch": 0.4,
"learning_rate": 0.0003021765295887663,
"loss": 2.7369,
"theoretical_loss": 3.554990343105767,
"tokens_seen": 1325924352
},
{
"epoch": 0.4,
"learning_rate": 0.0003021664994984955,
"loss": 2.8159,
"theoretical_loss": 3.554974482311533,
"tokens_seen": 1325989888
},
{
"epoch": 0.4,
"learning_rate": 0.0003021564694082247,
"loss": 2.6895,
"theoretical_loss": 3.554958622520669,
"tokens_seen": 1326055424
},
{
"epoch": 0.4,
"learning_rate": 0.00030214643931795386,
"loss": 2.6766,
"theoretical_loss": 3.5549427637330604,
"tokens_seen": 1326120960
},
{
"epoch": 0.4,
"learning_rate": 0.00030213640922768304,
"loss": 2.5689,
"theoretical_loss": 3.554926905948596,
"tokens_seen": 1326186496
},
{
"epoch": 0.4,
"learning_rate": 0.0003021263791374123,
"loss": 2.8289,
"theoretical_loss": 3.5549110491671616,
"tokens_seen": 1326252032
},
{
"epoch": 0.4,
"learning_rate": 0.0003021163490471414,
"loss": 2.879,
"theoretical_loss": 3.554895193388645,
"tokens_seen": 1326317568
},
{
"epoch": 0.4,
"learning_rate": 0.00030210631895687064,
"loss": 2.5766,
"theoretical_loss": 3.5548793386129325,
"tokens_seen": 1326383104
},
{
"epoch": 0.4,
"learning_rate": 0.00030209628886659976,
"loss": 2.6821,
"theoretical_loss": 3.5548634848399114,
"tokens_seen": 1326448640
},
{
"epoch": 0.4,
"learning_rate": 0.000302086258776329,
"loss": 2.7302,
"theoretical_loss": 3.5548476320694684,
"tokens_seen": 1326514176
},
{
"epoch": 0.4,
"learning_rate": 0.0003020762286860582,
"loss": 2.7115,
"theoretical_loss": 3.554831780301491,
"tokens_seen": 1326579712
},
{
"epoch": 0.4,
"learning_rate": 0.00030206619859578736,
"loss": 2.8077,
"theoretical_loss": 3.554815929535867,
"tokens_seen": 1326645248
},
{
"epoch": 0.4,
"learning_rate": 0.00030205616850551654,
"loss": 2.5985,
"theoretical_loss": 3.554800079772482,
"tokens_seen": 1326710784
},
{
"epoch": 0.4,
"learning_rate": 0.0003020461384152458,
"loss": 2.5804,
"theoretical_loss": 3.554784231011224,
"tokens_seen": 1326776320
},
{
"epoch": 0.4,
"learning_rate": 0.0003020361083249749,
"loss": 2.7809,
"theoretical_loss": 3.55476838325198,
"tokens_seen": 1326841856
},
{
"epoch": 0.4,
"learning_rate": 0.00030202607823470414,
"loss": 2.8554,
"theoretical_loss": 3.5547525364946373,
"tokens_seen": 1326907392
},
{
"epoch": 0.4,
"learning_rate": 0.00030201604814443327,
"loss": 2.7362,
"theoretical_loss": 3.554736690739083,
"tokens_seen": 1326972928
},
{
"epoch": 0.4,
"learning_rate": 0.0003020060180541625,
"loss": 2.9231,
"theoretical_loss": 3.5547208459852033,
"tokens_seen": 1327038464
},
{
"epoch": 0.4,
"objective/train/docs_used": 751457,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.970331907272339,
"objective/train/theoretical_loss": 3.5547050022328874,
"objective/train/tokens_used": 1347564000,
"theoretical_loss": 3.5547050022328874,
"tokens_seen": 1327104000
},
{
"epoch": 0.4,
"learning_rate": 0.0003019959879638917,
"loss": 2.8879,
"theoretical_loss": 3.5547050022328874,
"tokens_seen": 1327104000
},
{
"epoch": 0.4,
"learning_rate": 0.00030198595787362087,
"loss": 2.7053,
"theoretical_loss": 3.554689159482021,
"tokens_seen": 1327169536
},
{
"epoch": 0.4,
"learning_rate": 0.00030197592778335005,
"loss": 2.7052,
"theoretical_loss": 3.554673317732492,
"tokens_seen": 1327235072
},
{
"epoch": 0.4,
"learning_rate": 0.00030196589769307923,
"loss": 2.7315,
"theoretical_loss": 3.5546574769841874,
"tokens_seen": 1327300608
},
{
"epoch": 0.4,
"learning_rate": 0.0003019558676028084,
"loss": 2.8051,
"theoretical_loss": 3.554641637236995,
"tokens_seen": 1327366144
},
{
"epoch": 0.4,
"learning_rate": 0.00030194583751253764,
"loss": 2.6287,
"theoretical_loss": 3.554625798490801,
"tokens_seen": 1327431680
},
{
"epoch": 0.4,
"learning_rate": 0.00030193580742226677,
"loss": 2.725,
"theoretical_loss": 3.554609960745494,
"tokens_seen": 1327497216
},
{
"epoch": 0.4,
"learning_rate": 0.000301925777331996,
"loss": 2.9346,
"theoretical_loss": 3.5545941240009604,
"tokens_seen": 1327562752
},
{
"epoch": 0.4,
"learning_rate": 0.00030191574724172513,
"loss": 2.9416,
"theoretical_loss": 3.5545782882570878,
"tokens_seen": 1327628288
},
{
"epoch": 0.4,
"learning_rate": 0.00030190571715145437,
"loss": 2.7084,
"theoretical_loss": 3.554562453513764,
"tokens_seen": 1327693824
},
{
"epoch": 0.4,
"learning_rate": 0.00030189568706118355,
"loss": 2.6133,
"theoretical_loss": 3.554546619770876,
"tokens_seen": 1327759360
},
{
"epoch": 0.4,
"learning_rate": 0.00030188565697091273,
"loss": 2.7368,
"theoretical_loss": 3.5545307870283116,
"tokens_seen": 1327824896
},
{
"epoch": 0.4,
"learning_rate": 0.0003018756268806419,
"loss": 2.8332,
"theoretical_loss": 3.554514955285958,
"tokens_seen": 1327890432
},
{
"epoch": 0.4,
"learning_rate": 0.00030186559679037115,
"loss": 2.651,
"theoretical_loss": 3.5544991245437023,
"tokens_seen": 1327955968
},
{
"epoch": 0.4,
"learning_rate": 0.0003018555667001003,
"loss": 2.6852,
"theoretical_loss": 3.5544832948014324,
"tokens_seen": 1328021504
},
{
"epoch": 0.4,
"learning_rate": 0.0003018455366098295,
"loss": 2.7483,
"theoretical_loss": 3.5544674660590356,
"tokens_seen": 1328087040
},
{
"epoch": 0.4,
"learning_rate": 0.00030183550651955864,
"loss": 2.6111,
"theoretical_loss": 3.5544516383164,
"tokens_seen": 1328152576
},
{
"epoch": 0.4,
"learning_rate": 0.00030182547642928787,
"loss": 2.7985,
"theoretical_loss": 3.554435811573412,
"tokens_seen": 1328218112
},
{
"epoch": 0.4,
"learning_rate": 0.00030181544633901705,
"loss": 2.6271,
"theoretical_loss": 3.55441998582996,
"tokens_seen": 1328283648
},
{
"epoch": 0.4,
"learning_rate": 0.00030180541624874623,
"loss": 2.4844,
"theoretical_loss": 3.554404161085931,
"tokens_seen": 1328349184
},
{
"epoch": 0.4,
"learning_rate": 0.00030179538615847547,
"loss": 2.8319,
"theoretical_loss": 3.554388337341213,
"tokens_seen": 1328414720
},
{
"epoch": 0.4,
"learning_rate": 0.0003017853560682046,
"loss": 2.8188,
"theoretical_loss": 3.554372514595694,
"tokens_seen": 1328480256
},
{
"epoch": 0.4,
"learning_rate": 0.00030177532597793383,
"loss": 2.812,
"theoretical_loss": 3.554356692849261,
"tokens_seen": 1328545792
},
{
"epoch": 0.4,
"learning_rate": 0.000301765295887663,
"loss": 2.6832,
"theoretical_loss": 3.5543408721018013,
"tokens_seen": 1328611328
},
{
"epoch": 0.4,
"learning_rate": 0.0003017552657973922,
"loss": 2.6711,
"theoretical_loss": 3.5543250523532035,
"tokens_seen": 1328676864
},
{
"epoch": 0.4,
"objective/train/docs_used": 752138,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7791147232055664,
"objective/train/theoretical_loss": 3.5543092336033544,
"objective/train/tokens_used": 1349202400,
"theoretical_loss": 3.5543092336033544,
"tokens_seen": 1328742400
},
{
"epoch": 0.4,
"learning_rate": 0.0003017452357071214,
"loss": 2.6527,
"theoretical_loss": 3.5543092336033544,
"tokens_seen": 1328742400
},
{
"epoch": 0.4,
"learning_rate": 0.0003017352056168506,
"loss": 2.7393,
"theoretical_loss": 3.5542934158521424,
"tokens_seen": 1328807936
},
{
"epoch": 0.4,
"learning_rate": 0.00030172517552657974,
"loss": 2.4577,
"theoretical_loss": 3.554277599099455,
"tokens_seen": 1328873472
},
{
"epoch": 0.4,
"learning_rate": 0.00030171514543630897,
"loss": 2.8202,
"theoretical_loss": 3.5542617833451793,
"tokens_seen": 1328939008
},
{
"epoch": 0.4,
"learning_rate": 0.0003017051153460381,
"loss": 2.6443,
"theoretical_loss": 3.554245968589204,
"tokens_seen": 1329004544
},
{
"epoch": 0.4,
"learning_rate": 0.00030169508525576733,
"loss": 2.5624,
"theoretical_loss": 3.554230154831416,
"tokens_seen": 1329070080
},
{
"epoch": 0.4,
"learning_rate": 0.0003016850551654965,
"loss": 2.7081,
"theoretical_loss": 3.5542143420717043,
"tokens_seen": 1329135616
},
{
"epoch": 0.4,
"learning_rate": 0.0003016750250752257,
"loss": 2.6364,
"theoretical_loss": 3.554198530309955,
"tokens_seen": 1329201152
},
{
"epoch": 0.4,
"learning_rate": 0.0003016649949849549,
"loss": 2.7581,
"theoretical_loss": 3.5541827195460574,
"tokens_seen": 1329266688
},
{
"epoch": 0.4,
"learning_rate": 0.00030165496489468406,
"loss": 2.593,
"theoretical_loss": 3.554166909779898,
"tokens_seen": 1329332224
},
{
"epoch": 0.4,
"learning_rate": 0.00030164493480441324,
"loss": 2.757,
"theoretical_loss": 3.5541511010113664,
"tokens_seen": 1329397760
},
{
"epoch": 0.4,
"learning_rate": 0.0003016349047141425,
"loss": 2.8298,
"theoretical_loss": 3.554135293240349,
"tokens_seen": 1329463296
},
{
"epoch": 0.4,
"learning_rate": 0.0003016248746238716,
"loss": 2.5892,
"theoretical_loss": 3.554119486466734,
"tokens_seen": 1329528832
},
{
"epoch": 0.4,
"learning_rate": 0.00030161484453360084,
"loss": 2.7757,
"theoretical_loss": 3.55410368069041,
"tokens_seen": 1329594368
},
{
"epoch": 0.4,
"learning_rate": 0.00030160481444332996,
"loss": 2.6751,
"theoretical_loss": 3.554087875911264,
"tokens_seen": 1329659904
},
{
"epoch": 0.4,
"learning_rate": 0.0003015947843530592,
"loss": 2.5996,
"theoretical_loss": 3.554072072129184,
"tokens_seen": 1329725440
},
{
"epoch": 0.4,
"learning_rate": 0.0003015847542627884,
"loss": 2.8936,
"theoretical_loss": 3.554056269344059,
"tokens_seen": 1329790976
},
{
"epoch": 0.4,
"learning_rate": 0.00030157472417251756,
"loss": 2.6458,
"theoretical_loss": 3.554040467555776,
"tokens_seen": 1329856512
},
{
"epoch": 0.4,
"learning_rate": 0.00030156469408224674,
"loss": 2.8319,
"theoretical_loss": 3.5540246667642235,
"tokens_seen": 1329922048
},
{
"epoch": 0.4,
"learning_rate": 0.000301554663991976,
"loss": 2.9577,
"theoretical_loss": 3.554008866969289,
"tokens_seen": 1329987584
},
{
"epoch": 0.4,
"learning_rate": 0.0003015446339017051,
"loss": 2.642,
"theoretical_loss": 3.553993068170861,
"tokens_seen": 1330053120
},
{
"epoch": 0.4,
"learning_rate": 0.00030153460381143434,
"loss": 2.6487,
"theoretical_loss": 3.5539772703688275,
"tokens_seen": 1330118656
},
{
"epoch": 0.4,
"learning_rate": 0.00030152457372116347,
"loss": 3.1157,
"theoretical_loss": 3.553961473563076,
"tokens_seen": 1330184192
},
{
"epoch": 0.4,
"learning_rate": 0.0003015145436308927,
"loss": 2.7066,
"theoretical_loss": 3.5539456777534957,
"tokens_seen": 1330249728
},
{
"epoch": 0.4,
"learning_rate": 0.0003015045135406219,
"loss": 2.4438,
"theoretical_loss": 3.5539298829399737,
"tokens_seen": 1330315264
},
{
"epoch": 0.4,
"objective/train/docs_used": 753163,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.783708333969116,
"objective/train/theoretical_loss": 3.553914089122399,
"objective/train/tokens_used": 1350840800,
"theoretical_loss": 3.553914089122399,
"tokens_seen": 1330380800
},
{
"epoch": 0.4,
"learning_rate": 0.00030149448345035107,
"loss": 2.7517,
"theoretical_loss": 3.553914089122399,
"tokens_seen": 1330380800
},
{
"epoch": 0.4,
"learning_rate": 0.00030148445336008025,
"loss": 2.7072,
"theoretical_loss": 3.5538982963006585,
"tokens_seen": 1330446336
},
{
"epoch": 0.4,
"learning_rate": 0.00030147442326980943,
"loss": 2.296,
"theoretical_loss": 3.5538825044746414,
"tokens_seen": 1330511872
},
{
"epoch": 0.4,
"learning_rate": 0.0003014643931795386,
"loss": 2.7889,
"theoretical_loss": 3.5538667136442355,
"tokens_seen": 1330577408
},
{
"epoch": 0.4,
"learning_rate": 0.00030145436308926784,
"loss": 2.7001,
"theoretical_loss": 3.553850923809329,
"tokens_seen": 1330642944
},
{
"epoch": 0.4,
"learning_rate": 0.00030144433299899697,
"loss": 2.5116,
"theoretical_loss": 3.5538351349698107,
"tokens_seen": 1330708480
},
{
"epoch": 0.4,
"learning_rate": 0.0003014343029087262,
"loss": 2.592,
"theoretical_loss": 3.553819347125568,
"tokens_seen": 1330774016
},
{
"epoch": 0.4,
"learning_rate": 0.00030142427281845533,
"loss": 2.88,
"theoretical_loss": 3.5538035602764895,
"tokens_seen": 1330839552
},
{
"epoch": 0.4,
"learning_rate": 0.00030141424272818457,
"loss": 2.595,
"theoretical_loss": 3.5537877744224633,
"tokens_seen": 1330905088
},
{
"epoch": 0.4,
"learning_rate": 0.00030140421263791375,
"loss": 2.9143,
"theoretical_loss": 3.553771989563378,
"tokens_seen": 1330970624
},
{
"epoch": 0.4,
"learning_rate": 0.00030139418254764293,
"loss": 2.8218,
"theoretical_loss": 3.5537562056991217,
"tokens_seen": 1331036160
},
{
"epoch": 0.4,
"learning_rate": 0.0003013841524573721,
"loss": 2.6219,
"theoretical_loss": 3.5537404228295824,
"tokens_seen": 1331101696
},
{
"epoch": 0.4,
"learning_rate": 0.00030137412236710135,
"loss": 3.1023,
"theoretical_loss": 3.5537246409546492,
"tokens_seen": 1331167232
},
{
"epoch": 0.4,
"learning_rate": 0.0003013640922768305,
"loss": 2.6421,
"theoretical_loss": 3.5537088600742104,
"tokens_seen": 1331232768
},
{
"epoch": 0.4,
"learning_rate": 0.0003013540621865597,
"loss": 2.5835,
"theoretical_loss": 3.5536930801881534,
"tokens_seen": 1331298304
},
{
"epoch": 0.4,
"learning_rate": 0.00030134403209628884,
"loss": 2.7319,
"theoretical_loss": 3.5536773012963674,
"tokens_seen": 1331363840
},
{
"epoch": 0.4,
"learning_rate": 0.00030133400200601807,
"loss": 2.424,
"theoretical_loss": 3.553661523398741,
"tokens_seen": 1331429376
},
{
"epoch": 0.4,
"learning_rate": 0.00030132397191574725,
"loss": 2.8519,
"theoretical_loss": 3.553645746495162,
"tokens_seen": 1331494912
},
{
"epoch": 0.4,
"learning_rate": 0.00030131394182547643,
"loss": 2.8427,
"theoretical_loss": 3.553629970585519,
"tokens_seen": 1331560448
},
{
"epoch": 0.4,
"learning_rate": 0.0003013039117352056,
"loss": 2.8392,
"theoretical_loss": 3.5536141956697005,
"tokens_seen": 1331625984
},
{
"epoch": 0.4,
"learning_rate": 0.0003012938816449348,
"loss": 2.6119,
"theoretical_loss": 3.5535984217475955,
"tokens_seen": 1331691520
},
{
"epoch": 0.4,
"learning_rate": 0.000301283851554664,
"loss": 2.4941,
"theoretical_loss": 3.5535826488190922,
"tokens_seen": 1331757056
},
{
"epoch": 0.4,
"learning_rate": 0.0003012738214643932,
"loss": 2.7494,
"theoretical_loss": 3.5535668768840782,
"tokens_seen": 1331822592
},
{
"epoch": 0.4,
"learning_rate": 0.00030126379137412234,
"loss": 2.8029,
"theoretical_loss": 3.553551105942444,
"tokens_seen": 1331888128
},
{
"epoch": 0.4,
"learning_rate": 0.0003012537612838516,
"loss": 2.8463,
"theoretical_loss": 3.5535353359940762,
"tokens_seen": 1331953664
},
{
"epoch": 0.4,
"objective/train/docs_used": 753870,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5567798614501953,
"objective/train/theoretical_loss": 3.5535195670388644,
"objective/train/tokens_used": 1352479200,
"theoretical_loss": 3.5535195670388644,
"tokens_seen": 1332019200
},
{
"epoch": 0.4,
"learning_rate": 0.0003012437311935807,
"loss": 2.7643,
"theoretical_loss": 3.5535195670388644,
"tokens_seen": 1332019200
},
{
"epoch": 0.4,
"learning_rate": 0.00030123370110330994,
"loss": 2.8575,
"theoretical_loss": 3.553503799076697,
"tokens_seen": 1332084736
},
{
"epoch": 0.4,
"learning_rate": 0.0003012236710130391,
"loss": 2.8268,
"theoretical_loss": 3.5534880321074627,
"tokens_seen": 1332150272
},
{
"epoch": 0.4,
"learning_rate": 0.0003012136409227683,
"loss": 2.8019,
"theoretical_loss": 3.55347226613105,
"tokens_seen": 1332215808
},
{
"epoch": 0.4,
"learning_rate": 0.0003012036108324975,
"loss": 2.6464,
"theoretical_loss": 3.5534565011473473,
"tokens_seen": 1332281344
},
{
"epoch": 0.4,
"learning_rate": 0.0003011935807422267,
"loss": 2.639,
"theoretical_loss": 3.5534407371562438,
"tokens_seen": 1332346880
},
{
"epoch": 0.4,
"learning_rate": 0.00030118355065195584,
"loss": 2.5222,
"theoretical_loss": 3.553424974157628,
"tokens_seen": 1332412416
},
{
"epoch": 0.4,
"learning_rate": 0.0003011735205616851,
"loss": 2.745,
"theoretical_loss": 3.5534092121513883,
"tokens_seen": 1332477952
},
{
"epoch": 0.4,
"learning_rate": 0.0003011634904714142,
"loss": 2.6648,
"theoretical_loss": 3.5533934511374135,
"tokens_seen": 1332543488
},
{
"epoch": 0.4,
"learning_rate": 0.00030115346038114344,
"loss": 2.6752,
"theoretical_loss": 3.553377691115593,
"tokens_seen": 1332609024
},
{
"epoch": 0.4,
"learning_rate": 0.0003011434302908726,
"loss": 2.6614,
"theoretical_loss": 3.5533619320858145,
"tokens_seen": 1332674560
},
{
"epoch": 0.4,
"learning_rate": 0.0003011334002006018,
"loss": 2.7651,
"theoretical_loss": 3.5533461740479675,
"tokens_seen": 1332740096
},
{
"epoch": 0.4,
"learning_rate": 0.000301123370110331,
"loss": 2.6632,
"theoretical_loss": 3.553330417001941,
"tokens_seen": 1332805632
},
{
"epoch": 0.4,
"learning_rate": 0.00030111334002006016,
"loss": 2.62,
"theoretical_loss": 3.553314660947623,
"tokens_seen": 1332871168
},
{
"epoch": 0.4,
"learning_rate": 0.00030110330992978935,
"loss": 2.6843,
"theoretical_loss": 3.553298905884903,
"tokens_seen": 1332936704
},
{
"epoch": 0.4,
"learning_rate": 0.0003010932798395186,
"loss": 2.8737,
"theoretical_loss": 3.5532831518136696,
"tokens_seen": 1333002240
},
{
"epoch": 0.4,
"learning_rate": 0.0003010832497492477,
"loss": 2.8016,
"theoretical_loss": 3.5532673987338113,
"tokens_seen": 1333067776
},
{
"epoch": 0.4,
"learning_rate": 0.00030107321965897694,
"loss": 2.8507,
"theoretical_loss": 3.5532516466452173,
"tokens_seen": 1333133312
},
{
"epoch": 0.4,
"learning_rate": 0.00030106318956870607,
"loss": 2.5341,
"theoretical_loss": 3.553235895547777,
"tokens_seen": 1333198848
},
{
"epoch": 0.4,
"learning_rate": 0.0003010531594784353,
"loss": 2.7813,
"theoretical_loss": 3.5532201454413785,
"tokens_seen": 1333264384
},
{
"epoch": 0.4,
"learning_rate": 0.00030104312938816454,
"loss": 2.684,
"theoretical_loss": 3.5532043963259112,
"tokens_seen": 1333329920
},
{
"epoch": 0.4,
"learning_rate": 0.00030103309929789367,
"loss": 2.9155,
"theoretical_loss": 3.553188648201264,
"tokens_seen": 1333395456
},
{
"epoch": 0.4,
"learning_rate": 0.0003010230692076229,
"loss": 2.824,
"theoretical_loss": 3.5531729010673256,
"tokens_seen": 1333460992
},
{
"epoch": 0.4,
"learning_rate": 0.0003010130391173521,
"loss": 2.8636,
"theoretical_loss": 3.5531571549239853,
"tokens_seen": 1333526528
},
{
"epoch": 0.4,
"learning_rate": 0.00030100300902708127,
"loss": 2.7233,
"theoretical_loss": 3.5531414097711322,
"tokens_seen": 1333592064
},
{
"epoch": 0.4,
"objective/train/docs_used": 755212,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2996084690093994,
"objective/train/theoretical_loss": 3.553125665608655,
"objective/train/tokens_used": 1354117600,
"theoretical_loss": 3.553125665608655,
"tokens_seen": 1333657600
},
{
"epoch": 0.4,
"learning_rate": 0.00030099297893681045,
"loss": 2.6401,
"theoretical_loss": 3.553125665608655,
"tokens_seen": 1333657600
},
{
"epoch": 0.4,
"learning_rate": 0.00030098294884653963,
"loss": 2.8207,
"theoretical_loss": 3.553109922436443,
"tokens_seen": 1333723136
},
{
"epoch": 0.4,
"learning_rate": 0.0003009729187562688,
"loss": 2.7657,
"theoretical_loss": 3.553094180254385,
"tokens_seen": 1333788672
},
{
"epoch": 0.4,
"learning_rate": 0.00030096288866599804,
"loss": 2.575,
"theoretical_loss": 3.5530784390623698,
"tokens_seen": 1333854208
},
{
"epoch": 0.4,
"learning_rate": 0.00030095285857572717,
"loss": 2.6178,
"theoretical_loss": 3.5530626988602876,
"tokens_seen": 1333919744
},
{
"epoch": 0.4,
"learning_rate": 0.0003009428284854564,
"loss": 2.4539,
"theoretical_loss": 3.5530469596480265,
"tokens_seen": 1333985280
},
{
"epoch": 0.4,
"learning_rate": 0.00030093279839518553,
"loss": 2.7162,
"theoretical_loss": 3.553031221425476,
"tokens_seen": 1334050816
},
{
"epoch": 0.4,
"learning_rate": 0.00030092276830491477,
"loss": 2.7339,
"theoretical_loss": 3.553015484192525,
"tokens_seen": 1334116352
},
{
"epoch": 0.4,
"learning_rate": 0.00030091273821464395,
"loss": 2.6798,
"theoretical_loss": 3.5529997479490634,
"tokens_seen": 1334181888
},
{
"epoch": 0.4,
"learning_rate": 0.00030090270812437313,
"loss": 2.6608,
"theoretical_loss": 3.552984012694979,
"tokens_seen": 1334247424
},
{
"epoch": 0.4,
"learning_rate": 0.0003008926780341023,
"loss": 2.499,
"theoretical_loss": 3.5529682784301624,
"tokens_seen": 1334312960
},
{
"epoch": 0.4,
"learning_rate": 0.00030088264794383155,
"loss": 2.6213,
"theoretical_loss": 3.5529525451545023,
"tokens_seen": 1334378496
},
{
"epoch": 0.4,
"learning_rate": 0.0003008726178535607,
"loss": 2.8959,
"theoretical_loss": 3.552936812867888,
"tokens_seen": 1334444032
},
{
"epoch": 0.4,
"learning_rate": 0.0003008625877632899,
"loss": 2.6688,
"theoretical_loss": 3.552921081570209,
"tokens_seen": 1334509568
},
{
"epoch": 0.4,
"learning_rate": 0.00030085255767301904,
"loss": 2.6146,
"theoretical_loss": 3.5529053512613533,
"tokens_seen": 1334575104
},
{
"epoch": 0.4,
"learning_rate": 0.00030084252758274827,
"loss": 2.5269,
"theoretical_loss": 3.552889621941212,
"tokens_seen": 1334640640
},
{
"epoch": 0.4,
"learning_rate": 0.00030083249749247745,
"loss": 2.6377,
"theoretical_loss": 3.5528738936096724,
"tokens_seen": 1334706176
},
{
"epoch": 0.4,
"learning_rate": 0.00030082246740220663,
"loss": 2.5796,
"theoretical_loss": 3.552858166266626,
"tokens_seen": 1334771712
},
{
"epoch": 0.4,
"learning_rate": 0.0003008124373119358,
"loss": 2.8968,
"theoretical_loss": 3.552842439911961,
"tokens_seen": 1334837248
},
{
"epoch": 0.4,
"learning_rate": 0.000300802407221665,
"loss": 2.7097,
"theoretical_loss": 3.5528267145455668,
"tokens_seen": 1334902784
},
{
"epoch": 0.4,
"learning_rate": 0.0003007923771313942,
"loss": 2.6791,
"theoretical_loss": 3.552810990167333,
"tokens_seen": 1334968320
},
{
"epoch": 0.4,
"learning_rate": 0.0003007823470411234,
"loss": 2.8787,
"theoretical_loss": 3.5527952667771485,
"tokens_seen": 1335033856
},
{
"epoch": 0.4,
"learning_rate": 0.00030077231695085254,
"loss": 2.7212,
"theoretical_loss": 3.5527795443749035,
"tokens_seen": 1335099392
},
{
"epoch": 0.4,
"learning_rate": 0.0003007622868605818,
"loss": 3.0692,
"theoretical_loss": 3.5527638229604865,
"tokens_seen": 1335164928
},
{
"epoch": 0.4,
"learning_rate": 0.0003007522567703109,
"loss": 2.8623,
"theoretical_loss": 3.552748102533788,
"tokens_seen": 1335230464
},
{
"epoch": 0.4,
"objective/train/docs_used": 755960,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.947859287261963,
"objective/train/theoretical_loss": 3.552732383094696,
"objective/train/tokens_used": 1355756000,
"theoretical_loss": 3.552732383094696,
"tokens_seen": 1335296000
},
{
"epoch": 0.4,
"learning_rate": 0.00030074222668004014,
"loss": 2.6327,
"theoretical_loss": 3.552732383094696,
"tokens_seen": 1335296000
},
{
"epoch": 0.4,
"learning_rate": 0.0003007321965897693,
"loss": 2.5274,
"theoretical_loss": 3.5527166646431025,
"tokens_seen": 1335361536
},
{
"epoch": 0.4,
"learning_rate": 0.0003007221664994985,
"loss": 2.5458,
"theoretical_loss": 3.552700947178894,
"tokens_seen": 1335427072
},
{
"epoch": 0.4,
"learning_rate": 0.0003007121364092277,
"loss": 2.7934,
"theoretical_loss": 3.5526852307019623,
"tokens_seen": 1335492608
},
{
"epoch": 0.4,
"learning_rate": 0.0003007021063189569,
"loss": 2.5115,
"theoretical_loss": 3.5526695152121963,
"tokens_seen": 1335558144
},
{
"epoch": 0.4,
"learning_rate": 0.00030069207622868604,
"loss": 2.5152,
"theoretical_loss": 3.5526538007094848,
"tokens_seen": 1335623680
},
{
"epoch": 0.4,
"learning_rate": 0.0003006820461384153,
"loss": 2.6331,
"theoretical_loss": 3.552638087193718,
"tokens_seen": 1335689216
},
{
"epoch": 0.4,
"learning_rate": 0.0003006720160481444,
"loss": 2.5758,
"theoretical_loss": 3.552622374664786,
"tokens_seen": 1335754752
},
{
"epoch": 0.4,
"learning_rate": 0.00030066198595787364,
"loss": 2.6963,
"theoretical_loss": 3.5526066631225772,
"tokens_seen": 1335820288
},
{
"epoch": 0.4,
"learning_rate": 0.0003006519558676028,
"loss": 2.5305,
"theoretical_loss": 3.5525909525669817,
"tokens_seen": 1335885824
},
{
"epoch": 0.4,
"learning_rate": 0.000300641925777332,
"loss": 2.6756,
"theoretical_loss": 3.5525752429978894,
"tokens_seen": 1335951360
},
{
"epoch": 0.4,
"learning_rate": 0.0003006318956870612,
"loss": 2.6985,
"theoretical_loss": 3.552559534415191,
"tokens_seen": 1336016896
},
{
"epoch": 0.4,
"learning_rate": 0.00030062186559679036,
"loss": 2.6552,
"theoretical_loss": 3.552543826818774,
"tokens_seen": 1336082432
},
{
"epoch": 0.4,
"learning_rate": 0.00030061183550651955,
"loss": 2.4122,
"theoretical_loss": 3.552528120208529,
"tokens_seen": 1336147968
},
{
"epoch": 0.4,
"learning_rate": 0.0003006018054162488,
"loss": 2.8085,
"theoretical_loss": 3.5525124145843465,
"tokens_seen": 1336213504
},
{
"epoch": 0.4,
"learning_rate": 0.0003005917753259779,
"loss": 2.6274,
"theoretical_loss": 3.5524967099461158,
"tokens_seen": 1336279040
},
{
"epoch": 0.4,
"learning_rate": 0.00030058174523570714,
"loss": 2.7776,
"theoretical_loss": 3.5524810062937258,
"tokens_seen": 1336344576
},
{
"epoch": 0.4,
"learning_rate": 0.00030057171514543627,
"loss": 2.6432,
"theoretical_loss": 3.5524653036270673,
"tokens_seen": 1336410112
},
{
"epoch": 0.4,
"learning_rate": 0.0003005616850551655,
"loss": 2.8035,
"theoretical_loss": 3.55244960194603,
"tokens_seen": 1336475648
},
{
"epoch": 0.41,
"learning_rate": 0.0003005516549648947,
"loss": 2.4267,
"theoretical_loss": 3.5524339012505033,
"tokens_seen": 1336541184
},
{
"epoch": 0.41,
"learning_rate": 0.00030054162487462387,
"loss": 2.4338,
"theoretical_loss": 3.552418201540377,
"tokens_seen": 1336606720
},
{
"epoch": 0.41,
"learning_rate": 0.00030053159478435305,
"loss": 2.5759,
"theoretical_loss": 3.5524025028155415,
"tokens_seen": 1336672256
},
{
"epoch": 0.41,
"learning_rate": 0.0003005215646940823,
"loss": 2.8307,
"theoretical_loss": 3.5523868050758862,
"tokens_seen": 1336737792
},
{
"epoch": 0.41,
"learning_rate": 0.0003005115346038114,
"loss": 2.3744,
"theoretical_loss": 3.552371108321301,
"tokens_seen": 1336803328
},
{
"epoch": 0.41,
"learning_rate": 0.00030050150451354065,
"loss": 2.7009,
"theoretical_loss": 3.552355412551676,
"tokens_seen": 1336868864
},
{
"epoch": 0.41,
"objective/train/docs_used": 757325,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9396798610687256,
"objective/train/theoretical_loss": 3.5523397177669005,
"objective/train/tokens_used": 1357394400,
"theoretical_loss": 3.5523397177669005,
"tokens_seen": 1336934400
},
{
"epoch": 0.41,
"learning_rate": 0.0003004914744232698,
"loss": 2.8785,
"theoretical_loss": 3.5523397177669005,
"tokens_seen": 1336934400
},
{
"epoch": 0.41,
"learning_rate": 0.000300481444332999,
"loss": 2.776,
"theoretical_loss": 3.5523240239668654,
"tokens_seen": 1336999936
},
{
"epoch": 0.41,
"learning_rate": 0.0003004714142427282,
"loss": 2.8396,
"theoretical_loss": 3.55230833115146,
"tokens_seen": 1337065472
},
{
"epoch": 0.41,
"learning_rate": 0.00030046138415245737,
"loss": 2.5761,
"theoretical_loss": 3.5522926393205743,
"tokens_seen": 1337131008
},
{
"epoch": 0.41,
"learning_rate": 0.00030045135406218655,
"loss": 2.7026,
"theoretical_loss": 3.5522769484740984,
"tokens_seen": 1337196544
},
{
"epoch": 0.41,
"learning_rate": 0.00030044132397191573,
"loss": 2.6464,
"theoretical_loss": 3.5522612586119227,
"tokens_seen": 1337262080
},
{
"epoch": 0.41,
"learning_rate": 0.0003004312938816449,
"loss": 2.6629,
"theoretical_loss": 3.5522455697339366,
"tokens_seen": 1337327616
},
{
"epoch": 0.41,
"learning_rate": 0.00030042126379137415,
"loss": 2.8503,
"theoretical_loss": 3.55222988184003,
"tokens_seen": 1337393152
},
{
"epoch": 0.41,
"learning_rate": 0.0003004112337011033,
"loss": 2.7219,
"theoretical_loss": 3.552214194930094,
"tokens_seen": 1337458688
},
{
"epoch": 0.41,
"learning_rate": 0.0003004012036108325,
"loss": 2.7211,
"theoretical_loss": 3.552198509004018,
"tokens_seen": 1337524224
},
{
"epoch": 0.41,
"learning_rate": 0.0003003911735205617,
"loss": 2.9427,
"theoretical_loss": 3.5521828240616915,
"tokens_seen": 1337589760
},
{
"epoch": 0.41,
"learning_rate": 0.0003003811434302909,
"loss": 2.9471,
"theoretical_loss": 3.5521671401030055,
"tokens_seen": 1337655296
},
{
"epoch": 0.41,
"learning_rate": 0.00030037111334002006,
"loss": 2.8086,
"theoretical_loss": 3.5521514571278496,
"tokens_seen": 1337720832
},
{
"epoch": 0.41,
"learning_rate": 0.00030036108324974924,
"loss": 2.7219,
"theoretical_loss": 3.552135775136115,
"tokens_seen": 1337786368
},
{
"epoch": 0.41,
"learning_rate": 0.0003003510531594784,
"loss": 2.5232,
"theoretical_loss": 3.5521200941276905,
"tokens_seen": 1337851904
},
{
"epoch": 0.41,
"learning_rate": 0.00030034102306920765,
"loss": 2.8912,
"theoretical_loss": 3.552104414102467,
"tokens_seen": 1337917440
},
{
"epoch": 0.41,
"learning_rate": 0.0003003309929789368,
"loss": 2.7018,
"theoretical_loss": 3.552088735060334,
"tokens_seen": 1337982976
},
{
"epoch": 0.41,
"learning_rate": 0.000300320962888666,
"loss": 2.6068,
"theoretical_loss": 3.5520730570011825,
"tokens_seen": 1338048512
},
{
"epoch": 0.41,
"learning_rate": 0.00030031093279839514,
"loss": 2.7305,
"theoretical_loss": 3.5520573799249027,
"tokens_seen": 1338114048
},
{
"epoch": 0.41,
"learning_rate": 0.0003003009027081244,
"loss": 2.8573,
"theoretical_loss": 3.5520417038313843,
"tokens_seen": 1338179584
},
{
"epoch": 0.41,
"learning_rate": 0.0003002908726178536,
"loss": 2.8162,
"theoretical_loss": 3.552026028720518,
"tokens_seen": 1338245120
},
{
"epoch": 0.41,
"learning_rate": 0.00030028084252758274,
"loss": 2.4318,
"theoretical_loss": 3.552010354592194,
"tokens_seen": 1338310656
},
{
"epoch": 0.41,
"learning_rate": 0.000300270812437312,
"loss": 2.6385,
"theoretical_loss": 3.5519946814463026,
"tokens_seen": 1338376192
},
{
"epoch": 0.41,
"learning_rate": 0.0003002607823470411,
"loss": 2.76,
"theoretical_loss": 3.551979009282734,
"tokens_seen": 1338441728
},
{
"epoch": 0.41,
"learning_rate": 0.00030025075225677034,
"loss": 2.8521,
"theoretical_loss": 3.5519633381013787,
"tokens_seen": 1338507264
},
{
"epoch": 0.41,
"objective/train/docs_used": 758060,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0366034507751465,
"objective/train/theoretical_loss": 3.5519476679021267,
"objective/train/tokens_used": 1359032800,
"theoretical_loss": 3.5519476679021267,
"tokens_seen": 1338572800
},
{
"epoch": 0.41,
"learning_rate": 0.0003002407221664995,
"loss": 2.7489,
"theoretical_loss": 3.5519476679021267,
"tokens_seen": 1338572800
},
{
"epoch": 0.41,
"learning_rate": 0.0003002306920762287,
"loss": 2.8657,
"theoretical_loss": 3.551931998684869,
"tokens_seen": 1338638336
},
{
"epoch": 0.41,
"learning_rate": 0.0003002206619859579,
"loss": 2.3944,
"theoretical_loss": 3.5519163304494956,
"tokens_seen": 1338703872
},
{
"epoch": 0.41,
"learning_rate": 0.0003002106318956871,
"loss": 2.7097,
"theoretical_loss": 3.551900663195897,
"tokens_seen": 1338769408
},
{
"epoch": 0.41,
"learning_rate": 0.00030020060180541624,
"loss": 2.5708,
"theoretical_loss": 3.5518849969239636,
"tokens_seen": 1338834944
},
{
"epoch": 0.41,
"learning_rate": 0.0003001905717151455,
"loss": 2.6684,
"theoretical_loss": 3.551869331633586,
"tokens_seen": 1338900480
},
{
"epoch": 0.41,
"learning_rate": 0.0003001805416248746,
"loss": 2.5861,
"theoretical_loss": 3.5518536673246537,
"tokens_seen": 1338966016
},
{
"epoch": 0.41,
"learning_rate": 0.00030017051153460384,
"loss": 2.7405,
"theoretical_loss": 3.5518380039970587,
"tokens_seen": 1339031552
},
{
"epoch": 0.41,
"learning_rate": 0.000300160481444333,
"loss": 2.4739,
"theoretical_loss": 3.5518223416506904,
"tokens_seen": 1339097088
},
{
"epoch": 0.41,
"learning_rate": 0.0003001504513540622,
"loss": 2.8074,
"theoretical_loss": 3.55180668028544,
"tokens_seen": 1339162624
},
{
"epoch": 0.41,
"learning_rate": 0.0003001404212637914,
"loss": 2.8873,
"theoretical_loss": 3.5517910199011977,
"tokens_seen": 1339228160
},
{
"epoch": 0.41,
"learning_rate": 0.00030013039117352056,
"loss": 2.5308,
"theoretical_loss": 3.551775360497854,
"tokens_seen": 1339293696
},
{
"epoch": 0.41,
"learning_rate": 0.00030012036108324975,
"loss": 2.6381,
"theoretical_loss": 3.551759702075299,
"tokens_seen": 1339359232
},
{
"epoch": 0.41,
"learning_rate": 0.000300110330992979,
"loss": 2.8043,
"theoretical_loss": 3.5517440446334243,
"tokens_seen": 1339424768
},
{
"epoch": 0.41,
"learning_rate": 0.0003001003009027081,
"loss": 2.8763,
"theoretical_loss": 3.55172838817212,
"tokens_seen": 1339490304
},
{
"epoch": 0.41,
"learning_rate": 0.00030009027081243734,
"loss": 2.8604,
"theoretical_loss": 3.5517127326912767,
"tokens_seen": 1339555840
},
{
"epoch": 0.41,
"learning_rate": 0.00030008024072216647,
"loss": 2.8951,
"theoretical_loss": 3.5516970781907853,
"tokens_seen": 1339621376
},
{
"epoch": 0.41,
"learning_rate": 0.0003000702106318957,
"loss": 2.5762,
"theoretical_loss": 3.5516814246705355,
"tokens_seen": 1339686912
},
{
"epoch": 0.41,
"learning_rate": 0.0003000601805416249,
"loss": 3.0108,
"theoretical_loss": 3.5516657721304195,
"tokens_seen": 1339752448
},
{
"epoch": 0.41,
"learning_rate": 0.00030005015045135407,
"loss": 2.8557,
"theoretical_loss": 3.5516501205703266,
"tokens_seen": 1339817984
},
{
"epoch": 0.41,
"learning_rate": 0.00030004012036108325,
"loss": 2.6587,
"theoretical_loss": 3.551634469990148,
"tokens_seen": 1339883520
},
{
"epoch": 0.41,
"learning_rate": 0.0003000300902708125,
"loss": 2.4164,
"theoretical_loss": 3.551618820389775,
"tokens_seen": 1339949056
},
{
"epoch": 0.41,
"learning_rate": 0.0003000200601805416,
"loss": 2.9265,
"theoretical_loss": 3.5516031717690977,
"tokens_seen": 1340014592
},
{
"epoch": 0.41,
"learning_rate": 0.00030001003009027085,
"loss": 2.4716,
"theoretical_loss": 3.5515875241280073,
"tokens_seen": 1340080128
},
{
"epoch": 0.41,
"learning_rate": 0.0003,
"loss": 2.7637,
"theoretical_loss": 3.551571877466394,
"tokens_seen": 1340145664
},
{
"epoch": 0.41,
"objective/train/docs_used": 759300,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.826815128326416,
"objective/train/theoretical_loss": 3.551556231784149,
"objective/train/tokens_used": 1360671200,
"theoretical_loss": 3.551556231784149,
"tokens_seen": 1340211200
},
{
"epoch": 0.41,
"learning_rate": 0.0002999899699097292,
"loss": 2.7081,
"theoretical_loss": 3.551556231784149,
"tokens_seen": 1340211200
},
{
"epoch": 0.41,
"learning_rate": 0.0002999799398194584,
"loss": 2.6957,
"theoretical_loss": 3.5515405870811625,
"tokens_seen": 1340276736
},
{
"epoch": 0.41,
"learning_rate": 0.00029996990972918757,
"loss": 2.9563,
"theoretical_loss": 3.551524943357326,
"tokens_seen": 1340342272
},
{
"epoch": 0.41,
"learning_rate": 0.00029995987963891675,
"loss": 2.8668,
"theoretical_loss": 3.5515093006125307,
"tokens_seen": 1340407808
},
{
"epoch": 0.41,
"learning_rate": 0.00029994984954864593,
"loss": 2.5876,
"theoretical_loss": 3.551493658846667,
"tokens_seen": 1340473344
},
{
"epoch": 0.41,
"learning_rate": 0.0002999398194583751,
"loss": 2.8057,
"theoretical_loss": 3.5514780180596253,
"tokens_seen": 1340538880
},
{
"epoch": 0.41,
"learning_rate": 0.00029992978936810435,
"loss": 2.5726,
"theoretical_loss": 3.551462378251297,
"tokens_seen": 1340604416
},
{
"epoch": 0.41,
"learning_rate": 0.0002999197592778335,
"loss": 2.7319,
"theoretical_loss": 3.5514467394215723,
"tokens_seen": 1340669952
},
{
"epoch": 0.41,
"learning_rate": 0.0002999097291875627,
"loss": 2.7271,
"theoretical_loss": 3.551431101570344,
"tokens_seen": 1340735488
},
{
"epoch": 0.41,
"learning_rate": 0.0002998996990972919,
"loss": 2.8348,
"theoretical_loss": 3.551415464697501,
"tokens_seen": 1340801024
},
{
"epoch": 0.41,
"learning_rate": 0.0002998896690070211,
"loss": 2.6795,
"theoretical_loss": 3.5513998288029356,
"tokens_seen": 1340866560
},
{
"epoch": 0.41,
"learning_rate": 0.00029987963891675026,
"loss": 2.7639,
"theoretical_loss": 3.5513841938865376,
"tokens_seen": 1340932096
},
{
"epoch": 0.41,
"learning_rate": 0.00029986960882647944,
"loss": 2.8492,
"theoretical_loss": 3.551368559948199,
"tokens_seen": 1340997632
},
{
"epoch": 0.41,
"learning_rate": 0.0002998595787362086,
"loss": 2.6827,
"theoretical_loss": 3.5513529269878106,
"tokens_seen": 1341063168
},
{
"epoch": 0.41,
"learning_rate": 0.00029984954864593785,
"loss": 2.6076,
"theoretical_loss": 3.551337295005264,
"tokens_seen": 1341128704
},
{
"epoch": 0.41,
"learning_rate": 0.000299839518555667,
"loss": 2.6197,
"theoretical_loss": 3.5513216640004486,
"tokens_seen": 1341194240
},
{
"epoch": 0.41,
"learning_rate": 0.0002998294884653962,
"loss": 2.7823,
"theoretical_loss": 3.551306033973257,
"tokens_seen": 1341259776
},
{
"epoch": 0.41,
"learning_rate": 0.00029981945837512534,
"loss": 2.7048,
"theoretical_loss": 3.5512904049235794,
"tokens_seen": 1341325312
},
{
"epoch": 0.41,
"learning_rate": 0.0002998094282848546,
"loss": 2.4925,
"theoretical_loss": 3.551274776851308,
"tokens_seen": 1341390848
},
{
"epoch": 0.41,
"learning_rate": 0.00029979939819458376,
"loss": 2.7767,
"theoretical_loss": 3.5512591497563326,
"tokens_seen": 1341456384
},
{
"epoch": 0.41,
"learning_rate": 0.00029978936810431294,
"loss": 2.7094,
"theoretical_loss": 3.551243523638545,
"tokens_seen": 1341521920
},
{
"epoch": 0.41,
"learning_rate": 0.0002997793380140421,
"loss": 2.7681,
"theoretical_loss": 3.551227898497836,
"tokens_seen": 1341587456
},
{
"epoch": 0.41,
"learning_rate": 0.0002997693079237713,
"loss": 2.8532,
"theoretical_loss": 3.5512122743340977,
"tokens_seen": 1341652992
},
{
"epoch": 0.41,
"learning_rate": 0.0002997592778335005,
"loss": 2.6845,
"theoretical_loss": 3.5511966511472206,
"tokens_seen": 1341718528
},
{
"epoch": 0.41,
"learning_rate": 0.0002997492477432297,
"loss": 2.8226,
"theoretical_loss": 3.551181028937096,
"tokens_seen": 1341784064
},
{
"epoch": 0.41,
"objective/train/docs_used": 759950,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6778948307037354,
"objective/train/theoretical_loss": 3.551165407703615,
"objective/train/tokens_used": 1362309600,
"theoretical_loss": 3.551165407703615,
"tokens_seen": 1341849600
},
{
"epoch": 0.41,
"learning_rate": 0.00029973921765295885,
"loss": 2.5291,
"theoretical_loss": 3.551165407703615,
"tokens_seen": 1341849600
},
{
"epoch": 0.41,
"learning_rate": 0.0002997291875626881,
"loss": 2.5863,
"theoretical_loss": 3.551149787446669,
"tokens_seen": 1341915136
},
{
"epoch": 0.41,
"learning_rate": 0.00029971915747241726,
"loss": 2.5835,
"theoretical_loss": 3.5511341681661492,
"tokens_seen": 1341980672
},
{
"epoch": 0.41,
"learning_rate": 0.00029970912738214644,
"loss": 2.8931,
"theoretical_loss": 3.551118549861947,
"tokens_seen": 1342046208
},
{
"epoch": 0.41,
"learning_rate": 0.0002996990972918756,
"loss": 2.6234,
"theoretical_loss": 3.5511029325339534,
"tokens_seen": 1342111744
},
{
"epoch": 0.41,
"learning_rate": 0.0002996890672016048,
"loss": 2.6337,
"theoretical_loss": 3.5510873161820604,
"tokens_seen": 1342177280
},
{
"epoch": 0.41,
"learning_rate": 0.000299679037111334,
"loss": 2.7751,
"theoretical_loss": 3.5510717008061587,
"tokens_seen": 1342242816
},
{
"epoch": 0.41,
"learning_rate": 0.0002996690070210632,
"loss": 2.4692,
"theoretical_loss": 3.55105608640614,
"tokens_seen": 1342308352
},
{
"epoch": 0.41,
"learning_rate": 0.00029965897693079235,
"loss": 2.7911,
"theoretical_loss": 3.5510404729818954,
"tokens_seen": 1342373888
},
{
"epoch": 0.41,
"learning_rate": 0.0002996489468405216,
"loss": 2.47,
"theoretical_loss": 3.5510248605333166,
"tokens_seen": 1342439424
},
{
"epoch": 0.41,
"learning_rate": 0.0002996389167502507,
"loss": 2.7274,
"theoretical_loss": 3.5510092490602942,
"tokens_seen": 1342504960
},
{
"epoch": 0.41,
"learning_rate": 0.00029962888665997995,
"loss": 2.7533,
"theoretical_loss": 3.550993638562721,
"tokens_seen": 1342570496
},
{
"epoch": 0.41,
"learning_rate": 0.00029961885656970913,
"loss": 2.7345,
"theoretical_loss": 3.550978029040487,
"tokens_seen": 1342636032
},
{
"epoch": 0.41,
"learning_rate": 0.0002996088264794383,
"loss": 2.681,
"theoretical_loss": 3.5509624204934847,
"tokens_seen": 1342701568
},
{
"epoch": 0.41,
"learning_rate": 0.0002995987963891675,
"loss": 2.6994,
"theoretical_loss": 3.5509468129216053,
"tokens_seen": 1342767104
},
{
"epoch": 0.41,
"learning_rate": 0.00029958876629889667,
"loss": 2.6173,
"theoretical_loss": 3.55093120632474,
"tokens_seen": 1342832640
},
{
"epoch": 0.41,
"learning_rate": 0.00029957873620862585,
"loss": 2.8339,
"theoretical_loss": 3.5509156007027807,
"tokens_seen": 1342898176
},
{
"epoch": 0.41,
"learning_rate": 0.0002995687061183551,
"loss": 2.4827,
"theoretical_loss": 3.5508999960556182,
"tokens_seen": 1342963712
},
{
"epoch": 0.41,
"learning_rate": 0.0002995586760280842,
"loss": 2.7775,
"theoretical_loss": 3.550884392383145,
"tokens_seen": 1343029248
},
{
"epoch": 0.41,
"learning_rate": 0.00029954864593781345,
"loss": 2.7558,
"theoretical_loss": 3.550868789685252,
"tokens_seen": 1343094784
},
{
"epoch": 0.41,
"learning_rate": 0.0002995386158475427,
"loss": 2.7322,
"theoretical_loss": 3.5508531879618315,
"tokens_seen": 1343160320
},
{
"epoch": 0.41,
"learning_rate": 0.0002995285857572718,
"loss": 2.7153,
"theoretical_loss": 3.550837587212774,
"tokens_seen": 1343225856
},
{
"epoch": 0.41,
"learning_rate": 0.00029951855566700105,
"loss": 2.4996,
"theoretical_loss": 3.5508219874379723,
"tokens_seen": 1343291392
},
{
"epoch": 0.41,
"learning_rate": 0.0002995085255767302,
"loss": 2.7312,
"theoretical_loss": 3.5508063886373176,
"tokens_seen": 1343356928
},
{
"epoch": 0.41,
"learning_rate": 0.0002994984954864594,
"loss": 2.7957,
"theoretical_loss": 3.5507907908107006,
"tokens_seen": 1343422464
},
{
"epoch": 0.41,
"objective/train/docs_used": 761102,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.762277364730835,
"objective/train/theoretical_loss": 3.5507751939580148,
"objective/train/tokens_used": 1363948000,
"theoretical_loss": 3.5507751939580148,
"tokens_seen": 1343488000
},
{
"epoch": 0.41,
"learning_rate": 0.0002994884653961886,
"loss": 2.9387,
"theoretical_loss": 3.5507751939580148,
"tokens_seen": 1343488000
},
{
"epoch": 0.41,
"learning_rate": 0.00029947843530591777,
"loss": 2.6678,
"theoretical_loss": 3.55075959807915,
"tokens_seen": 1343553536
},
{
"epoch": 0.41,
"learning_rate": 0.00029946840521564695,
"loss": 2.8529,
"theoretical_loss": 3.550744003173999,
"tokens_seen": 1343619072
},
{
"epoch": 0.41,
"learning_rate": 0.00029945837512537613,
"loss": 2.5847,
"theoretical_loss": 3.5507284092424536,
"tokens_seen": 1343684608
},
{
"epoch": 0.41,
"learning_rate": 0.0002994483450351053,
"loss": 2.6928,
"theoretical_loss": 3.550712816284405,
"tokens_seen": 1343750144
},
{
"epoch": 0.41,
"learning_rate": 0.00029943831494483455,
"loss": 2.7275,
"theoretical_loss": 3.550697224299745,
"tokens_seen": 1343815680
},
{
"epoch": 0.41,
"learning_rate": 0.0002994282848545637,
"loss": 2.8396,
"theoretical_loss": 3.5506816332883657,
"tokens_seen": 1343881216
},
{
"epoch": 0.41,
"learning_rate": 0.0002994182547642929,
"loss": 2.7839,
"theoretical_loss": 3.5506660432501587,
"tokens_seen": 1343946752
},
{
"epoch": 0.41,
"learning_rate": 0.0002994082246740221,
"loss": 2.515,
"theoretical_loss": 3.550650454185016,
"tokens_seen": 1344012288
},
{
"epoch": 0.41,
"learning_rate": 0.0002993981945837513,
"loss": 2.8401,
"theoretical_loss": 3.550634866092829,
"tokens_seen": 1344077824
},
{
"epoch": 0.41,
"learning_rate": 0.00029938816449348046,
"loss": 2.9656,
"theoretical_loss": 3.55061927897349,
"tokens_seen": 1344143360
},
{
"epoch": 0.41,
"learning_rate": 0.00029937813440320964,
"loss": 2.6999,
"theoretical_loss": 3.5506036928268907,
"tokens_seen": 1344208896
},
{
"epoch": 0.41,
"learning_rate": 0.0002993681043129388,
"loss": 2.643,
"theoretical_loss": 3.5505881076529224,
"tokens_seen": 1344274432
},
{
"epoch": 0.41,
"learning_rate": 0.00029935807422266805,
"loss": 2.9362,
"theoretical_loss": 3.550572523451478,
"tokens_seen": 1344339968
},
{
"epoch": 0.41,
"learning_rate": 0.0002993480441323972,
"loss": 2.484,
"theoretical_loss": 3.5505569402224486,
"tokens_seen": 1344405504
},
{
"epoch": 0.41,
"learning_rate": 0.0002993380140421264,
"loss": 2.6879,
"theoretical_loss": 3.550541357965727,
"tokens_seen": 1344471040
},
{
"epoch": 0.41,
"learning_rate": 0.00029932798395185554,
"loss": 2.6689,
"theoretical_loss": 3.550525776681204,
"tokens_seen": 1344536576
},
{
"epoch": 0.41,
"learning_rate": 0.0002993179538615848,
"loss": 2.4986,
"theoretical_loss": 3.5505101963687724,
"tokens_seen": 1344602112
},
{
"epoch": 0.41,
"learning_rate": 0.00029930792377131396,
"loss": 2.7902,
"theoretical_loss": 3.550494617028324,
"tokens_seen": 1344667648
},
{
"epoch": 0.41,
"learning_rate": 0.00029929789368104314,
"loss": 2.6288,
"theoretical_loss": 3.5504790386597507,
"tokens_seen": 1344733184
},
{
"epoch": 0.41,
"learning_rate": 0.0002992878635907723,
"loss": 2.8328,
"theoretical_loss": 3.5504634612629444,
"tokens_seen": 1344798720
},
{
"epoch": 0.41,
"learning_rate": 0.0002992778335005015,
"loss": 2.8677,
"theoretical_loss": 3.5504478848377974,
"tokens_seen": 1344864256
},
{
"epoch": 0.41,
"learning_rate": 0.0002992678034102307,
"loss": 2.7088,
"theoretical_loss": 3.5504323093842016,
"tokens_seen": 1344929792
},
{
"epoch": 0.41,
"learning_rate": 0.0002992577733199599,
"loss": 2.6719,
"theoretical_loss": 3.550416734902049,
"tokens_seen": 1344995328
},
{
"epoch": 0.41,
"learning_rate": 0.00029924774322968905,
"loss": 2.6494,
"theoretical_loss": 3.550401161391232,
"tokens_seen": 1345060864
},
{
"epoch": 0.41,
"objective/train/docs_used": 761644,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.757722854614258,
"objective/train/theoretical_loss": 3.550385588851642,
"objective/train/tokens_used": 1365586400,
"theoretical_loss": 3.550385588851642,
"tokens_seen": 1345126400
},
{
"epoch": 0.41,
"learning_rate": 0.0002992377131394183,
"loss": 2.6141,
"theoretical_loss": 3.550385588851642,
"tokens_seen": 1345126400
},
{
"epoch": 0.41,
"learning_rate": 0.00029922768304914746,
"loss": 2.506,
"theoretical_loss": 3.550370017283172,
"tokens_seen": 1345191936
},
{
"epoch": 0.41,
"learning_rate": 0.00029921765295887664,
"loss": 2.8152,
"theoretical_loss": 3.5503544466857133,
"tokens_seen": 1345257472
},
{
"epoch": 0.41,
"learning_rate": 0.0002992076228686058,
"loss": 2.7255,
"theoretical_loss": 3.5503388770591586,
"tokens_seen": 1345323008
},
{
"epoch": 0.41,
"learning_rate": 0.000299197592778335,
"loss": 2.5884,
"theoretical_loss": 3.5503233084034003,
"tokens_seen": 1345388544
},
{
"epoch": 0.41,
"learning_rate": 0.0002991875626880642,
"loss": 2.6959,
"theoretical_loss": 3.5503077407183294,
"tokens_seen": 1345454080
},
{
"epoch": 0.41,
"learning_rate": 0.0002991775325977934,
"loss": 2.7112,
"theoretical_loss": 3.5502921740038396,
"tokens_seen": 1345519616
},
{
"epoch": 0.41,
"learning_rate": 0.00029916750250752255,
"loss": 2.8435,
"theoretical_loss": 3.5502766082598223,
"tokens_seen": 1345585152
},
{
"epoch": 0.41,
"learning_rate": 0.0002991574724172518,
"loss": 2.788,
"theoretical_loss": 3.55026104348617,
"tokens_seen": 1345650688
},
{
"epoch": 0.41,
"learning_rate": 0.0002991474423269809,
"loss": 2.6025,
"theoretical_loss": 3.5502454796827743,
"tokens_seen": 1345716224
},
{
"epoch": 0.41,
"learning_rate": 0.00029913741223671015,
"loss": 2.6477,
"theoretical_loss": 3.5502299168495286,
"tokens_seen": 1345781760
},
{
"epoch": 0.41,
"learning_rate": 0.00029912738214643933,
"loss": 2.8934,
"theoretical_loss": 3.550214354986324,
"tokens_seen": 1345847296
},
{
"epoch": 0.41,
"learning_rate": 0.0002991173520561685,
"loss": 2.62,
"theoretical_loss": 3.5501987940930535,
"tokens_seen": 1345912832
},
{
"epoch": 0.41,
"learning_rate": 0.0002991073219658977,
"loss": 2.784,
"theoretical_loss": 3.5501832341696096,
"tokens_seen": 1345978368
},
{
"epoch": 0.41,
"learning_rate": 0.00029909729187562687,
"loss": 2.5896,
"theoretical_loss": 3.550167675215884,
"tokens_seen": 1346043904
},
{
"epoch": 0.41,
"learning_rate": 0.00029908726178535605,
"loss": 2.5204,
"theoretical_loss": 3.55015211723177,
"tokens_seen": 1346109440
},
{
"epoch": 0.41,
"learning_rate": 0.0002990772316950853,
"loss": 2.6217,
"theoretical_loss": 3.5501365602171586,
"tokens_seen": 1346174976
},
{
"epoch": 0.41,
"learning_rate": 0.0002990672016048144,
"loss": 2.7421,
"theoretical_loss": 3.5501210041719435,
"tokens_seen": 1346240512
},
{
"epoch": 0.41,
"learning_rate": 0.00029905717151454365,
"loss": 2.6526,
"theoretical_loss": 3.550105449096016,
"tokens_seen": 1346306048
},
{
"epoch": 0.41,
"learning_rate": 0.00029904714142427283,
"loss": 2.6508,
"theoretical_loss": 3.5500898949892696,
"tokens_seen": 1346371584
},
{
"epoch": 0.41,
"learning_rate": 0.000299037111334002,
"loss": 2.6586,
"theoretical_loss": 3.5500743418515963,
"tokens_seen": 1346437120
},
{
"epoch": 0.41,
"learning_rate": 0.0002990270812437312,
"loss": 2.594,
"theoretical_loss": 3.550058789682888,
"tokens_seen": 1346502656
},
{
"epoch": 0.41,
"learning_rate": 0.0002990170511534604,
"loss": 2.5986,
"theoretical_loss": 3.5500432384830383,
"tokens_seen": 1346568192
},
{
"epoch": 0.41,
"learning_rate": 0.00029900702106318956,
"loss": 2.909,
"theoretical_loss": 3.550027688251938,
"tokens_seen": 1346633728
},
{
"epoch": 0.41,
"learning_rate": 0.0002989969909729188,
"loss": 2.5677,
"theoretical_loss": 3.550012138989482,
"tokens_seen": 1346699264
},
{
"epoch": 0.41,
"objective/train/docs_used": 762571,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.025847911834717,
"objective/train/theoretical_loss": 3.5499965906955606,
"objective/train/tokens_used": 1367224800,
"theoretical_loss": 3.5499965906955606,
"tokens_seen": 1346764800
},
{
"epoch": 0.41,
"learning_rate": 0.0002989869608826479,
"loss": 2.8399,
"theoretical_loss": 3.5499965906955606,
"tokens_seen": 1346764800
},
{
"epoch": 0.41,
"learning_rate": 0.00029897693079237715,
"loss": 2.6636,
"theoretical_loss": 3.549981043370068,
"tokens_seen": 1346830336
},
{
"epoch": 0.41,
"learning_rate": 0.0002989669007021063,
"loss": 2.7202,
"theoretical_loss": 3.549965497012895,
"tokens_seen": 1346895872
},
{
"epoch": 0.41,
"learning_rate": 0.0002989568706118355,
"loss": 2.893,
"theoretical_loss": 3.549949951623936,
"tokens_seen": 1346961408
},
{
"epoch": 0.41,
"learning_rate": 0.0002989468405215647,
"loss": 2.8669,
"theoretical_loss": 3.5499344072030823,
"tokens_seen": 1347026944
},
{
"epoch": 0.41,
"learning_rate": 0.0002989368104312939,
"loss": 2.5447,
"theoretical_loss": 3.5499188637502277,
"tokens_seen": 1347092480
},
{
"epoch": 0.41,
"learning_rate": 0.00029892678034102306,
"loss": 2.7483,
"theoretical_loss": 3.5499033212652638,
"tokens_seen": 1347158016
},
{
"epoch": 0.41,
"learning_rate": 0.0002989167502507523,
"loss": 2.8141,
"theoretical_loss": 3.549887779748083,
"tokens_seen": 1347223552
},
{
"epoch": 0.41,
"learning_rate": 0.0002989067201604814,
"loss": 2.8984,
"theoretical_loss": 3.549872239198579,
"tokens_seen": 1347289088
},
{
"epoch": 0.41,
"learning_rate": 0.00029889669007021066,
"loss": 2.7593,
"theoretical_loss": 3.5498566996166443,
"tokens_seen": 1347354624
},
{
"epoch": 0.41,
"learning_rate": 0.0002988866599799398,
"loss": 2.7952,
"theoretical_loss": 3.549841161002171,
"tokens_seen": 1347420160
},
{
"epoch": 0.41,
"learning_rate": 0.000298876629889669,
"loss": 2.7224,
"theoretical_loss": 3.5498256233550523,
"tokens_seen": 1347485696
},
{
"epoch": 0.41,
"learning_rate": 0.0002988665997993982,
"loss": 2.6147,
"theoretical_loss": 3.5498100866751807,
"tokens_seen": 1347551232
},
{
"epoch": 0.41,
"learning_rate": 0.0002988565697091274,
"loss": 2.9044,
"theoretical_loss": 3.549794550962449,
"tokens_seen": 1347616768
},
{
"epoch": 0.41,
"learning_rate": 0.00029884653961885656,
"loss": 2.5725,
"theoretical_loss": 3.5497790162167497,
"tokens_seen": 1347682304
},
{
"epoch": 0.41,
"learning_rate": 0.00029883650952858574,
"loss": 2.4907,
"theoretical_loss": 3.5497634824379762,
"tokens_seen": 1347747840
},
{
"epoch": 0.41,
"learning_rate": 0.0002988264794383149,
"loss": 2.7674,
"theoretical_loss": 3.5497479496260214,
"tokens_seen": 1347813376
},
{
"epoch": 0.41,
"learning_rate": 0.00029881644934804416,
"loss": 2.6976,
"theoretical_loss": 3.549732417780777,
"tokens_seen": 1347878912
},
{
"epoch": 0.41,
"learning_rate": 0.0002988064192577733,
"loss": 2.5191,
"theoretical_loss": 3.5497168869021367,
"tokens_seen": 1347944448
},
{
"epoch": 0.41,
"learning_rate": 0.0002987963891675025,
"loss": 2.9245,
"theoretical_loss": 3.549701356989994,
"tokens_seen": 1348009984
},
{
"epoch": 0.41,
"learning_rate": 0.0002987863590772317,
"loss": 2.807,
"theoretical_loss": 3.54968582804424,
"tokens_seen": 1348075520
},
{
"epoch": 0.41,
"learning_rate": 0.0002987763289869609,
"loss": 2.8442,
"theoretical_loss": 3.549670300064769,
"tokens_seen": 1348141056
},
{
"epoch": 0.41,
"learning_rate": 0.0002987662988966901,
"loss": 2.8152,
"theoretical_loss": 3.549654773051473,
"tokens_seen": 1348206592
},
{
"epoch": 0.41,
"learning_rate": 0.00029875626880641925,
"loss": 2.8223,
"theoretical_loss": 3.549639247004246,
"tokens_seen": 1348272128
},
{
"epoch": 0.41,
"learning_rate": 0.0002987462387161485,
"loss": 3.0077,
"theoretical_loss": 3.54962372192298,
"tokens_seen": 1348337664
},
{
"epoch": 0.41,
"objective/train/docs_used": 763291,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6128878593444824,
"objective/train/theoretical_loss": 3.549608197807568,
"objective/train/tokens_used": 1368863200,
"theoretical_loss": 3.549608197807568,
"tokens_seen": 1348403200
},
{
"epoch": 0.41,
"learning_rate": 0.00029873620862587766,
"loss": 2.6703,
"theoretical_loss": 3.549608197807568,
"tokens_seen": 1348403200
},
{
"epoch": 0.41,
"learning_rate": 0.00029872617853560684,
"loss": 2.7354,
"theoretical_loss": 3.5495926746579034,
"tokens_seen": 1348468736
},
{
"epoch": 0.41,
"learning_rate": 0.000298716148445336,
"loss": 2.813,
"theoretical_loss": 3.5495771524738795,
"tokens_seen": 1348534272
},
{
"epoch": 0.41,
"learning_rate": 0.0002987061183550652,
"loss": 2.9232,
"theoretical_loss": 3.5495616312553886,
"tokens_seen": 1348599808
},
{
"epoch": 0.41,
"learning_rate": 0.0002986960882647944,
"loss": 2.4726,
"theoretical_loss": 3.549546111002324,
"tokens_seen": 1348665344
},
{
"epoch": 0.41,
"learning_rate": 0.0002986860581745236,
"loss": 2.8936,
"theoretical_loss": 3.5495305917145785,
"tokens_seen": 1348730880
},
{
"epoch": 0.41,
"learning_rate": 0.00029867602808425275,
"loss": 2.6277,
"theoretical_loss": 3.5495150733920457,
"tokens_seen": 1348796416
},
{
"epoch": 0.41,
"learning_rate": 0.000298665997993982,
"loss": 2.7266,
"theoretical_loss": 3.549499556034618,
"tokens_seen": 1348861952
},
{
"epoch": 0.41,
"learning_rate": 0.0002986559679037111,
"loss": 2.6077,
"theoretical_loss": 3.5494840396421887,
"tokens_seen": 1348927488
},
{
"epoch": 0.41,
"learning_rate": 0.00029864593781344035,
"loss": 2.6246,
"theoretical_loss": 3.5494685242146513,
"tokens_seen": 1348993024
},
{
"epoch": 0.41,
"learning_rate": 0.00029863590772316953,
"loss": 2.7984,
"theoretical_loss": 3.549453009751899,
"tokens_seen": 1349058560
},
{
"epoch": 0.41,
"learning_rate": 0.0002986258776328987,
"loss": 2.7341,
"theoretical_loss": 3.549437496253824,
"tokens_seen": 1349124096
},
{
"epoch": 0.41,
"learning_rate": 0.0002986158475426279,
"loss": 2.8037,
"theoretical_loss": 3.549421983720321,
"tokens_seen": 1349189632
},
{
"epoch": 0.41,
"learning_rate": 0.00029860581745235707,
"loss": 2.9096,
"theoretical_loss": 3.549406472151281,
"tokens_seen": 1349255168
},
{
"epoch": 0.41,
"learning_rate": 0.00029859578736208625,
"loss": 2.6781,
"theoretical_loss": 3.5493909615465995,
"tokens_seen": 1349320704
},
{
"epoch": 0.41,
"learning_rate": 0.0002985857572718155,
"loss": 2.5995,
"theoretical_loss": 3.549375451906168,
"tokens_seen": 1349386240
},
{
"epoch": 0.41,
"learning_rate": 0.0002985757271815446,
"loss": 2.9521,
"theoretical_loss": 3.549359943229881,
"tokens_seen": 1349451776
},
{
"epoch": 0.41,
"learning_rate": 0.00029856569709127385,
"loss": 2.7715,
"theoretical_loss": 3.5493444355176305,
"tokens_seen": 1349517312
},
{
"epoch": 0.41,
"learning_rate": 0.00029855566700100303,
"loss": 2.8562,
"theoretical_loss": 3.549328928769311,
"tokens_seen": 1349582848
},
{
"epoch": 0.41,
"learning_rate": 0.0002985456369107322,
"loss": 2.5774,
"theoretical_loss": 3.549313422984815,
"tokens_seen": 1349648384
},
{
"epoch": 0.41,
"learning_rate": 0.0002985356068204614,
"loss": 2.5478,
"theoretical_loss": 3.5492979181640356,
"tokens_seen": 1349713920
},
{
"epoch": 0.41,
"learning_rate": 0.0002985255767301906,
"loss": 2.6206,
"theoretical_loss": 3.549282414306867,
"tokens_seen": 1349779456
},
{
"epoch": 0.41,
"learning_rate": 0.00029851554663991976,
"loss": 2.4354,
"theoretical_loss": 3.549266911413201,
"tokens_seen": 1349844992
},
{
"epoch": 0.41,
"learning_rate": 0.000298505516549649,
"loss": 2.6838,
"theoretical_loss": 3.549251409482933,
"tokens_seen": 1349910528
},
{
"epoch": 0.41,
"learning_rate": 0.0002984954864593781,
"loss": 2.6087,
"theoretical_loss": 3.5492359085159553,
"tokens_seen": 1349976064
},
{
"epoch": 0.41,
"objective/train/docs_used": 764652,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3573992252349854,
"objective/train/theoretical_loss": 3.549220408512161,
"objective/train/tokens_used": 1370501600,
"theoretical_loss": 3.549220408512161,
"tokens_seen": 1350041600
},
{
"epoch": 0.41,
"learning_rate": 0.00029848545636910735,
"loss": 2.707,
"theoretical_loss": 3.549220408512161,
"tokens_seen": 1350041600
},
{
"epoch": 0.41,
"learning_rate": 0.0002984754262788365,
"loss": 2.5907,
"theoretical_loss": 3.549204909471444,
"tokens_seen": 1350107136
},
{
"epoch": 0.41,
"learning_rate": 0.0002984653961885657,
"loss": 2.4526,
"theoretical_loss": 3.549189411393697,
"tokens_seen": 1350172672
},
{
"epoch": 0.41,
"learning_rate": 0.0002984553660982949,
"loss": 2.5071,
"theoretical_loss": 3.5491739142788146,
"tokens_seen": 1350238208
},
{
"epoch": 0.41,
"learning_rate": 0.0002984453360080241,
"loss": 2.8169,
"theoretical_loss": 3.549158418126689,
"tokens_seen": 1350303744
},
{
"epoch": 0.41,
"learning_rate": 0.00029843530591775326,
"loss": 2.6122,
"theoretical_loss": 3.5491429229372153,
"tokens_seen": 1350369280
},
{
"epoch": 0.41,
"learning_rate": 0.0002984252758274825,
"loss": 2.7673,
"theoretical_loss": 3.549127428710285,
"tokens_seen": 1350434816
},
{
"epoch": 0.41,
"learning_rate": 0.0002984152457372116,
"loss": 2.6882,
"theoretical_loss": 3.5491119354457927,
"tokens_seen": 1350500352
},
{
"epoch": 0.41,
"learning_rate": 0.00029840521564694086,
"loss": 2.767,
"theoretical_loss": 3.549096443143632,
"tokens_seen": 1350565888
},
{
"epoch": 0.41,
"learning_rate": 0.00029839518555667,
"loss": 2.6705,
"theoretical_loss": 3.5490809518036963,
"tokens_seen": 1350631424
},
{
"epoch": 0.41,
"learning_rate": 0.0002983851554663992,
"loss": 2.6979,
"theoretical_loss": 3.5490654614258785,
"tokens_seen": 1350696960
},
{
"epoch": 0.41,
"learning_rate": 0.0002983751253761284,
"loss": 2.6624,
"theoretical_loss": 3.5490499720100734,
"tokens_seen": 1350762496
},
{
"epoch": 0.41,
"learning_rate": 0.0002983650952858576,
"loss": 2.7254,
"theoretical_loss": 3.5490344835561736,
"tokens_seen": 1350828032
},
{
"epoch": 0.41,
"learning_rate": 0.00029835506519558676,
"loss": 2.4605,
"theoretical_loss": 3.5490189960640732,
"tokens_seen": 1350893568
},
{
"epoch": 0.41,
"learning_rate": 0.00029834503510531594,
"loss": 2.7626,
"theoretical_loss": 3.5490035095336654,
"tokens_seen": 1350959104
},
{
"epoch": 0.41,
"learning_rate": 0.0002983350050150451,
"loss": 2.6751,
"theoretical_loss": 3.5489880239648444,
"tokens_seen": 1351024640
},
{
"epoch": 0.41,
"learning_rate": 0.00029832497492477436,
"loss": 2.6996,
"theoretical_loss": 3.5489725393575036,
"tokens_seen": 1351090176
},
{
"epoch": 0.41,
"learning_rate": 0.0002983149448345035,
"loss": 2.8123,
"theoretical_loss": 3.548957055711536,
"tokens_seen": 1351155712
},
{
"epoch": 0.41,
"learning_rate": 0.0002983049147442327,
"loss": 2.5293,
"theoretical_loss": 3.548941573026836,
"tokens_seen": 1351221248
},
{
"epoch": 0.41,
"learning_rate": 0.00029829488465396185,
"loss": 2.8064,
"theoretical_loss": 3.548926091303297,
"tokens_seen": 1351286784
},
{
"epoch": 0.41,
"learning_rate": 0.0002982848545636911,
"loss": 2.6408,
"theoretical_loss": 3.5489106105408137,
"tokens_seen": 1351352320
},
{
"epoch": 0.41,
"learning_rate": 0.00029827482447342026,
"loss": 2.8422,
"theoretical_loss": 3.548895130739279,
"tokens_seen": 1351417856
},
{
"epoch": 0.41,
"learning_rate": 0.00029826479438314945,
"loss": 2.6055,
"theoretical_loss": 3.548879651898586,
"tokens_seen": 1351483392
},
{
"epoch": 0.41,
"learning_rate": 0.0002982547642928786,
"loss": 2.5632,
"theoretical_loss": 3.5488641740186293,
"tokens_seen": 1351548928
},
{
"epoch": 0.41,
"learning_rate": 0.00029824473420260786,
"loss": 2.7028,
"theoretical_loss": 3.5488486970993027,
"tokens_seen": 1351614464
},
{
"epoch": 0.41,
"objective/train/docs_used": 765372,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.994880437850952,
"objective/train/theoretical_loss": 3.5488332211405,
"objective/train/tokens_used": 1372140000,
"theoretical_loss": 3.5488332211405,
"tokens_seen": 1351680000
},
{
"epoch": 0.41,
"learning_rate": 0.000298234704112337,
"loss": 2.8615,
"theoretical_loss": 3.5488332211405,
"tokens_seen": 1351680000
},
{
"epoch": 0.41,
"learning_rate": 0.0002982246740220662,
"loss": 2.7228,
"theoretical_loss": 3.548817746142115,
"tokens_seen": 1351745536
},
{
"epoch": 0.41,
"learning_rate": 0.00029821464393179535,
"loss": 2.62,
"theoretical_loss": 3.548802272104041,
"tokens_seen": 1351811072
},
{
"epoch": 0.41,
"learning_rate": 0.0002982046138415246,
"loss": 2.6727,
"theoretical_loss": 3.548786799026172,
"tokens_seen": 1351876608
},
{
"epoch": 0.41,
"learning_rate": 0.00029819458375125377,
"loss": 2.7072,
"theoretical_loss": 3.548771326908403,
"tokens_seen": 1351942144
},
{
"epoch": 0.41,
"learning_rate": 0.00029818455366098295,
"loss": 2.6532,
"theoretical_loss": 3.5487558557506267,
"tokens_seen": 1352007680
},
{
"epoch": 0.41,
"learning_rate": 0.00029817452357071213,
"loss": 2.7973,
"theoretical_loss": 3.548740385552737,
"tokens_seen": 1352073216
},
{
"epoch": 0.41,
"learning_rate": 0.0002981644934804413,
"loss": 2.7006,
"theoretical_loss": 3.5487249163146286,
"tokens_seen": 1352138752
},
{
"epoch": 0.41,
"learning_rate": 0.0002981544633901705,
"loss": 2.6004,
"theoretical_loss": 3.5487094480361945,
"tokens_seen": 1352204288
},
{
"epoch": 0.41,
"learning_rate": 0.00029814443329989973,
"loss": 2.7913,
"theoretical_loss": 3.5486939807173297,
"tokens_seen": 1352269824
},
{
"epoch": 0.41,
"learning_rate": 0.00029813440320962885,
"loss": 2.5624,
"theoretical_loss": 3.5486785143579276,
"tokens_seen": 1352335360
},
{
"epoch": 0.41,
"learning_rate": 0.0002981243731193581,
"loss": 2.7529,
"theoretical_loss": 3.548663048957882,
"tokens_seen": 1352400896
},
{
"epoch": 0.41,
"learning_rate": 0.0002981143430290872,
"loss": 2.7455,
"theoretical_loss": 3.548647584517087,
"tokens_seen": 1352466432
},
{
"epoch": 0.41,
"learning_rate": 0.00029810431293881645,
"loss": 2.7899,
"theoretical_loss": 3.5486321210354372,
"tokens_seen": 1352531968
},
{
"epoch": 0.41,
"learning_rate": 0.00029809428284854563,
"loss": 2.8969,
"theoretical_loss": 3.548616658512826,
"tokens_seen": 1352597504
},
{
"epoch": 0.41,
"learning_rate": 0.0002980842527582748,
"loss": 2.5323,
"theoretical_loss": 3.5486011969491478,
"tokens_seen": 1352663040
},
{
"epoch": 0.41,
"learning_rate": 0.000298074222668004,
"loss": 2.4192,
"theoretical_loss": 3.548585736344296,
"tokens_seen": 1352728576
},
{
"epoch": 0.41,
"learning_rate": 0.00029806419257773323,
"loss": 2.5293,
"theoretical_loss": 3.5485702766981655,
"tokens_seen": 1352794112
},
{
"epoch": 0.41,
"learning_rate": 0.00029805416248746236,
"loss": 2.5155,
"theoretical_loss": 3.5485548180106505,
"tokens_seen": 1352859648
},
{
"epoch": 0.41,
"learning_rate": 0.0002980441323971916,
"loss": 2.5142,
"theoretical_loss": 3.5485393602816444,
"tokens_seen": 1352925184
},
{
"epoch": 0.41,
"learning_rate": 0.0002980341023069208,
"loss": 2.5914,
"theoretical_loss": 3.548523903511042,
"tokens_seen": 1352990720
},
{
"epoch": 0.41,
"learning_rate": 0.00029802407221664996,
"loss": 2.8317,
"theoretical_loss": 3.5485084476987367,
"tokens_seen": 1353056256
},
{
"epoch": 0.41,
"learning_rate": 0.0002980140421263792,
"loss": 2.5369,
"theoretical_loss": 3.548492992844623,
"tokens_seen": 1353121792
},
{
"epoch": 0.41,
"learning_rate": 0.0002980040120361083,
"loss": 2.7321,
"theoretical_loss": 3.548477538948596,
"tokens_seen": 1353187328
},
{
"epoch": 0.41,
"learning_rate": 0.00029799398194583755,
"loss": 2.6755,
"theoretical_loss": 3.5484620860105487,
"tokens_seen": 1353252864
},
{
"epoch": 0.41,
"objective/train/docs_used": 766365,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2769768238067627,
"objective/train/theoretical_loss": 3.5484466340303755,
"objective/train/tokens_used": 1373778400,
"theoretical_loss": 3.5484466340303755,
"tokens_seen": 1353318400
},
{
"epoch": 0.41,
"learning_rate": 0.0002979839518555667,
"loss": 2.5893,
"theoretical_loss": 3.5484466340303755,
"tokens_seen": 1353318400
},
{
"epoch": 0.41,
"learning_rate": 0.0002979739217652959,
"loss": 2.7481,
"theoretical_loss": 3.5484311830079713,
"tokens_seen": 1353383936
},
{
"epoch": 0.41,
"learning_rate": 0.0002979638916750251,
"loss": 2.7043,
"theoretical_loss": 3.54841573294323,
"tokens_seen": 1353449472
},
{
"epoch": 0.41,
"learning_rate": 0.0002979538615847543,
"loss": 2.5963,
"theoretical_loss": 3.5484002838360453,
"tokens_seen": 1353515008
},
{
"epoch": 0.41,
"learning_rate": 0.00029794383149448346,
"loss": 2.6533,
"theoretical_loss": 3.5483848356863126,
"tokens_seen": 1353580544
},
{
"epoch": 0.41,
"learning_rate": 0.0002979338014042127,
"loss": 2.6323,
"theoretical_loss": 3.5483693884939256,
"tokens_seen": 1353646080
},
{
"epoch": 0.41,
"learning_rate": 0.0002979237713139418,
"loss": 2.5745,
"theoretical_loss": 3.548353942258778,
"tokens_seen": 1353711616
},
{
"epoch": 0.41,
"learning_rate": 0.00029791374122367106,
"loss": 2.705,
"theoretical_loss": 3.5483384969807656,
"tokens_seen": 1353777152
},
{
"epoch": 0.41,
"learning_rate": 0.0002979037111334002,
"loss": 2.6183,
"theoretical_loss": 3.548323052659782,
"tokens_seen": 1353842688
},
{
"epoch": 0.41,
"learning_rate": 0.0002978936810431294,
"loss": 2.7121,
"theoretical_loss": 3.548307609295721,
"tokens_seen": 1353908224
},
{
"epoch": 0.41,
"learning_rate": 0.0002978836509528586,
"loss": 2.6224,
"theoretical_loss": 3.548292166888478,
"tokens_seen": 1353973760
},
{
"epoch": 0.41,
"learning_rate": 0.0002978736208625878,
"loss": 2.5969,
"theoretical_loss": 3.5482767254379466,
"tokens_seen": 1354039296
},
{
"epoch": 0.41,
"learning_rate": 0.00029786359077231696,
"loss": 2.8643,
"theoretical_loss": 3.5482612849440223,
"tokens_seen": 1354104832
},
{
"epoch": 0.41,
"learning_rate": 0.00029785356068204614,
"loss": 2.7321,
"theoretical_loss": 3.5482458454065977,
"tokens_seen": 1354170368
},
{
"epoch": 0.41,
"learning_rate": 0.0002978435305917753,
"loss": 2.6767,
"theoretical_loss": 3.548230406825569,
"tokens_seen": 1354235904
},
{
"epoch": 0.41,
"learning_rate": 0.00029783350050150456,
"loss": 2.5504,
"theoretical_loss": 3.54821496920083,
"tokens_seen": 1354301440
},
{
"epoch": 0.41,
"learning_rate": 0.0002978234704112337,
"loss": 2.6599,
"theoretical_loss": 3.5481995325322755,
"tokens_seen": 1354366976
},
{
"epoch": 0.41,
"learning_rate": 0.0002978134403209629,
"loss": 2.6439,
"theoretical_loss": 3.5481840968197993,
"tokens_seen": 1354432512
},
{
"epoch": 0.41,
"learning_rate": 0.00029780341023069205,
"loss": 2.7223,
"theoretical_loss": 3.5481686620632966,
"tokens_seen": 1354498048
},
{
"epoch": 0.41,
"learning_rate": 0.0002977933801404213,
"loss": 2.6714,
"theoretical_loss": 3.548153228262662,
"tokens_seen": 1354563584
},
{
"epoch": 0.41,
"learning_rate": 0.00029778335005015046,
"loss": 2.5644,
"theoretical_loss": 3.5481377954177895,
"tokens_seen": 1354629120
},
{
"epoch": 0.41,
"learning_rate": 0.00029777331995987965,
"loss": 2.7552,
"theoretical_loss": 3.548122363528574,
"tokens_seen": 1354694656
},
{
"epoch": 0.41,
"learning_rate": 0.00029776328986960883,
"loss": 2.6056,
"theoretical_loss": 3.5481069325949104,
"tokens_seen": 1354760192
},
{
"epoch": 0.41,
"learning_rate": 0.00029775325977933806,
"loss": 2.5938,
"theoretical_loss": 3.5480915026166926,
"tokens_seen": 1354825728
},
{
"epoch": 0.41,
"learning_rate": 0.0002977432296890672,
"loss": 2.7494,
"theoretical_loss": 3.5480760735938155,
"tokens_seen": 1354891264
},
{
"epoch": 0.41,
"objective/train/docs_used": 767078,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.520714282989502,
"objective/train/theoretical_loss": 3.5480606455261743,
"objective/train/tokens_used": 1375416800,
"theoretical_loss": 3.5480606455261743,
"tokens_seen": 1354956800
},
{
"epoch": 0.41,
"learning_rate": 0.0002977331995987964,
"loss": 2.5214,
"theoretical_loss": 3.5480606455261743,
"tokens_seen": 1354956800
},
{
"epoch": 0.41,
"learning_rate": 0.00029772316950852555,
"loss": 2.8872,
"theoretical_loss": 3.5480452184136633,
"tokens_seen": 1355022336
},
{
"epoch": 0.41,
"learning_rate": 0.0002977131394182548,
"loss": 2.8561,
"theoretical_loss": 3.5480297922561768,
"tokens_seen": 1355087872
},
{
"epoch": 0.41,
"learning_rate": 0.00029770310932798397,
"loss": 2.7376,
"theoretical_loss": 3.54801436705361,
"tokens_seen": 1355153408
},
{
"epoch": 0.41,
"learning_rate": 0.00029769307923771315,
"loss": 2.6976,
"theoretical_loss": 3.547998942805857,
"tokens_seen": 1355218944
},
{
"epoch": 0.41,
"learning_rate": 0.00029768304914744233,
"loss": 2.8986,
"theoretical_loss": 3.547983519512813,
"tokens_seen": 1355284480
},
{
"epoch": 0.41,
"learning_rate": 0.0002976730190571715,
"loss": 2.5225,
"theoretical_loss": 3.547968097174373,
"tokens_seen": 1355350016
},
{
"epoch": 0.41,
"learning_rate": 0.0002976629889669007,
"loss": 2.8603,
"theoretical_loss": 3.547952675790431,
"tokens_seen": 1355415552
},
{
"epoch": 0.41,
"learning_rate": 0.00029765295887662993,
"loss": 2.7317,
"theoretical_loss": 3.547937255360883,
"tokens_seen": 1355481088
},
{
"epoch": 0.41,
"learning_rate": 0.00029764292878635905,
"loss": 2.6488,
"theoretical_loss": 3.5479218358856226,
"tokens_seen": 1355546624
},
{
"epoch": 0.41,
"learning_rate": 0.0002976328986960883,
"loss": 2.6325,
"theoretical_loss": 3.547906417364545,
"tokens_seen": 1355612160
},
{
"epoch": 0.41,
"learning_rate": 0.0002976228686058174,
"loss": 2.6691,
"theoretical_loss": 3.547890999797545,
"tokens_seen": 1355677696
},
{
"epoch": 0.41,
"learning_rate": 0.00029761283851554665,
"loss": 2.6347,
"theoretical_loss": 3.5478755831845175,
"tokens_seen": 1355743232
},
{
"epoch": 0.41,
"learning_rate": 0.00029760280842527583,
"loss": 3.0312,
"theoretical_loss": 3.547860167525357,
"tokens_seen": 1355808768
},
{
"epoch": 0.41,
"learning_rate": 0.000297592778335005,
"loss": 2.7578,
"theoretical_loss": 3.5478447528199597,
"tokens_seen": 1355874304
},
{
"epoch": 0.41,
"learning_rate": 0.0002975827482447342,
"loss": 2.7916,
"theoretical_loss": 3.547829339068219,
"tokens_seen": 1355939840
},
{
"epoch": 0.41,
"learning_rate": 0.00029757271815446343,
"loss": 2.8093,
"theoretical_loss": 3.54781392627003,
"tokens_seen": 1356005376
},
{
"epoch": 0.41,
"learning_rate": 0.00029756268806419256,
"loss": 2.8071,
"theoretical_loss": 3.5477985144252884,
"tokens_seen": 1356070912
},
{
"epoch": 0.41,
"learning_rate": 0.0002975526579739218,
"loss": 2.7454,
"theoretical_loss": 3.547783103533889,
"tokens_seen": 1356136448
},
{
"epoch": 0.41,
"learning_rate": 0.0002975426278836509,
"loss": 2.4953,
"theoretical_loss": 3.547767693595726,
"tokens_seen": 1356201984
},
{
"epoch": 0.41,
"learning_rate": 0.00029753259779338016,
"loss": 2.4831,
"theoretical_loss": 3.547752284610695,
"tokens_seen": 1356267520
},
{
"epoch": 0.41,
"learning_rate": 0.00029752256770310934,
"loss": 2.7992,
"theoretical_loss": 3.547736876578691,
"tokens_seen": 1356333056
},
{
"epoch": 0.41,
"learning_rate": 0.0002975125376128385,
"loss": 2.4733,
"theoretical_loss": 3.547721469499608,
"tokens_seen": 1356398592
},
{
"epoch": 0.41,
"learning_rate": 0.0002975025075225677,
"loss": 2.5303,
"theoretical_loss": 3.547706063373343,
"tokens_seen": 1356464128
},
{
"epoch": 0.41,
"learning_rate": 0.0002974924774322969,
"loss": 2.7047,
"theoretical_loss": 3.54769065819979,
"tokens_seen": 1356529664
},
{
"epoch": 0.41,
"objective/train/docs_used": 768221,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.369994878768921,
"objective/train/theoretical_loss": 3.547675253978843,
"objective/train/tokens_used": 1377055200,
"theoretical_loss": 3.547675253978843,
"tokens_seen": 1356595200
},
{
"epoch": 0.41,
"learning_rate": 0.00029748244734202606,
"loss": 2.5736,
"theoretical_loss": 3.547675253978843,
"tokens_seen": 1356595200
},
{
"epoch": 0.41,
"learning_rate": 0.0002974724172517553,
"loss": 2.5475,
"theoretical_loss": 3.5476598507103994,
"tokens_seen": 1356660736
},
{
"epoch": 0.41,
"learning_rate": 0.0002974623871614844,
"loss": 2.6484,
"theoretical_loss": 3.547644448394352,
"tokens_seen": 1356726272
},
{
"epoch": 0.41,
"learning_rate": 0.00029745235707121366,
"loss": 2.7106,
"theoretical_loss": 3.547629047030597,
"tokens_seen": 1356791808
},
{
"epoch": 0.41,
"learning_rate": 0.0002974423269809428,
"loss": 3.0358,
"theoretical_loss": 3.5476136466190296,
"tokens_seen": 1356857344
},
{
"epoch": 0.41,
"learning_rate": 0.000297432296890672,
"loss": 2.784,
"theoretical_loss": 3.547598247159545,
"tokens_seen": 1356922880
},
{
"epoch": 0.41,
"learning_rate": 0.0002974222668004012,
"loss": 2.4364,
"theoretical_loss": 3.5475828486520378,
"tokens_seen": 1356988416
},
{
"epoch": 0.41,
"learning_rate": 0.0002974122367101304,
"loss": 2.7573,
"theoretical_loss": 3.5475674510964037,
"tokens_seen": 1357053952
},
{
"epoch": 0.41,
"learning_rate": 0.00029740220661985956,
"loss": 2.8778,
"theoretical_loss": 3.547552054492537,
"tokens_seen": 1357119488
},
{
"epoch": 0.41,
"learning_rate": 0.0002973921765295888,
"loss": 2.7092,
"theoretical_loss": 3.5475366588403343,
"tokens_seen": 1357185024
},
{
"epoch": 0.41,
"learning_rate": 0.0002973821464393179,
"loss": 2.8935,
"theoretical_loss": 3.5475212641396903,
"tokens_seen": 1357250560
},
{
"epoch": 0.41,
"learning_rate": 0.00029737211634904716,
"loss": 2.5589,
"theoretical_loss": 3.5475058703905,
"tokens_seen": 1357316096
},
{
"epoch": 0.41,
"learning_rate": 0.0002973620862587763,
"loss": 2.6766,
"theoretical_loss": 3.5474904775926586,
"tokens_seen": 1357381632
},
{
"epoch": 0.41,
"learning_rate": 0.0002973520561685055,
"loss": 2.4717,
"theoretical_loss": 3.5474750857460613,
"tokens_seen": 1357447168
},
{
"epoch": 0.41,
"learning_rate": 0.0002973420260782347,
"loss": 2.5724,
"theoretical_loss": 3.547459694850604,
"tokens_seen": 1357512704
},
{
"epoch": 0.41,
"learning_rate": 0.0002973319959879639,
"loss": 2.4783,
"theoretical_loss": 3.5474443049061812,
"tokens_seen": 1357578240
},
{
"epoch": 0.41,
"learning_rate": 0.00029732196589769307,
"loss": 2.6654,
"theoretical_loss": 3.5474289159126893,
"tokens_seen": 1357643776
},
{
"epoch": 0.41,
"learning_rate": 0.00029731193580742225,
"loss": 2.6991,
"theoretical_loss": 3.5474135278700225,
"tokens_seen": 1357709312
},
{
"epoch": 0.41,
"learning_rate": 0.00029730190571715143,
"loss": 2.6222,
"theoretical_loss": 3.5473981407780766,
"tokens_seen": 1357774848
},
{
"epoch": 0.41,
"learning_rate": 0.00029729187562688067,
"loss": 2.8254,
"theoretical_loss": 3.5473827546367476,
"tokens_seen": 1357840384
},
{
"epoch": 0.41,
"learning_rate": 0.00029728184553660985,
"loss": 2.7546,
"theoretical_loss": 3.54736736944593,
"tokens_seen": 1357905920
},
{
"epoch": 0.41,
"learning_rate": 0.00029727181544633903,
"loss": 2.6518,
"theoretical_loss": 3.5473519852055193,
"tokens_seen": 1357971456
},
{
"epoch": 0.41,
"learning_rate": 0.00029726178535606826,
"loss": 2.7444,
"theoretical_loss": 3.5473366019154113,
"tokens_seen": 1358036992
},
{
"epoch": 0.41,
"learning_rate": 0.0002972517552657974,
"loss": 2.6374,
"theoretical_loss": 3.5473212195755015,
"tokens_seen": 1358102528
},
{
"epoch": 0.41,
"learning_rate": 0.0002972417251755266,
"loss": 2.8248,
"theoretical_loss": 3.547305838185685,
"tokens_seen": 1358168064
},
{
"epoch": 0.41,
"objective/train/docs_used": 768734,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4809162616729736,
"objective/train/theoretical_loss": 3.5472904577458575,
"objective/train/tokens_used": 1378693600,
"theoretical_loss": 3.5472904577458575,
"tokens_seen": 1358233600
},
{
"epoch": 0.41,
"learning_rate": 0.00029723169508525575,
"loss": 2.541,
"theoretical_loss": 3.5472904577458575,
"tokens_seen": 1358233600
},
{
"epoch": 0.41,
"learning_rate": 0.000297221664994985,
"loss": 2.6457,
"theoretical_loss": 3.5472750782559146,
"tokens_seen": 1358299136
},
{
"epoch": 0.41,
"learning_rate": 0.00029721163490471417,
"loss": 2.5745,
"theoretical_loss": 3.5472596997157515,
"tokens_seen": 1358364672
},
{
"epoch": 0.41,
"learning_rate": 0.00029720160481444335,
"loss": 2.6636,
"theoretical_loss": 3.5472443221252634,
"tokens_seen": 1358430208
},
{
"epoch": 0.41,
"learning_rate": 0.00029719157472417253,
"loss": 2.5608,
"theoretical_loss": 3.547228945484347,
"tokens_seen": 1358495744
},
{
"epoch": 0.41,
"learning_rate": 0.0002971815446339017,
"loss": 2.86,
"theoretical_loss": 3.5472135697928966,
"tokens_seen": 1358561280
},
{
"epoch": 0.41,
"learning_rate": 0.0002971715145436309,
"loss": 2.6354,
"theoretical_loss": 3.547198195050809,
"tokens_seen": 1358626816
},
{
"epoch": 0.41,
"learning_rate": 0.00029716148445336013,
"loss": 2.6621,
"theoretical_loss": 3.5471828212579792,
"tokens_seen": 1358692352
},
{
"epoch": 0.41,
"learning_rate": 0.00029715145436308926,
"loss": 2.6802,
"theoretical_loss": 3.547167448414302,
"tokens_seen": 1358757888
},
{
"epoch": 0.41,
"learning_rate": 0.0002971414242728185,
"loss": 2.7873,
"theoretical_loss": 3.5471520765196747,
"tokens_seen": 1358823424
},
{
"epoch": 0.41,
"learning_rate": 0.0002971313941825476,
"loss": 2.5375,
"theoretical_loss": 3.5471367055739913,
"tokens_seen": 1358888960
},
{
"epoch": 0.41,
"learning_rate": 0.00029712136409227685,
"loss": 2.4851,
"theoretical_loss": 3.547121335577148,
"tokens_seen": 1358954496
},
{
"epoch": 0.41,
"learning_rate": 0.00029711133400200603,
"loss": 2.7848,
"theoretical_loss": 3.5471059665290414,
"tokens_seen": 1359020032
},
{
"epoch": 0.41,
"learning_rate": 0.0002971013039117352,
"loss": 2.3968,
"theoretical_loss": 3.547090598429566,
"tokens_seen": 1359085568
},
{
"epoch": 0.41,
"learning_rate": 0.0002970912738214644,
"loss": 2.658,
"theoretical_loss": 3.547075231278618,
"tokens_seen": 1359151104
},
{
"epoch": 0.41,
"learning_rate": 0.00029708124373119363,
"loss": 2.6693,
"theoretical_loss": 3.5470598650760934,
"tokens_seen": 1359216640
},
{
"epoch": 0.41,
"learning_rate": 0.00029707121364092276,
"loss": 2.7014,
"theoretical_loss": 3.5470444998218875,
"tokens_seen": 1359282176
},
{
"epoch": 0.41,
"learning_rate": 0.000297061183550652,
"loss": 2.7696,
"theoretical_loss": 3.5470291355158956,
"tokens_seen": 1359347712
},
{
"epoch": 0.41,
"learning_rate": 0.0002970511534603811,
"loss": 2.7374,
"theoretical_loss": 3.5470137721580146,
"tokens_seen": 1359413248
},
{
"epoch": 0.41,
"learning_rate": 0.00029704112337011036,
"loss": 2.6167,
"theoretical_loss": 3.5469984097481397,
"tokens_seen": 1359478784
},
{
"epoch": 0.41,
"learning_rate": 0.00029703109327983954,
"loss": 2.8433,
"theoretical_loss": 3.546983048286167,
"tokens_seen": 1359544320
},
{
"epoch": 0.41,
"learning_rate": 0.0002970210631895687,
"loss": 2.7589,
"theoretical_loss": 3.5469676877719913,
"tokens_seen": 1359609856
},
{
"epoch": 0.41,
"learning_rate": 0.0002970110330992979,
"loss": 2.8687,
"theoretical_loss": 3.54695232820551,
"tokens_seen": 1359675392
},
{
"epoch": 0.41,
"learning_rate": 0.0002970010030090271,
"loss": 2.7233,
"theoretical_loss": 3.5469369695866177,
"tokens_seen": 1359740928
},
{
"epoch": 0.41,
"learning_rate": 0.00029699097291875626,
"loss": 2.8498,
"theoretical_loss": 3.546921611915211,
"tokens_seen": 1359806464
},
{
"epoch": 0.41,
"objective/train/docs_used": 769315,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7516112327575684,
"objective/train/theoretical_loss": 3.5469062551911854,
"objective/train/tokens_used": 1380332000,
"theoretical_loss": 3.5469062551911854,
"tokens_seen": 1359872000
},
{
"epoch": 0.41,
"learning_rate": 0.0002969809428284855,
"loss": 2.7173,
"theoretical_loss": 3.5469062551911854,
"tokens_seen": 1359872000
},
{
"epoch": 0.41,
"learning_rate": 0.0002969709127382146,
"loss": 2.6821,
"theoretical_loss": 3.5468908994144366,
"tokens_seen": 1359937536
},
{
"epoch": 0.41,
"learning_rate": 0.00029696088264794386,
"loss": 2.87,
"theoretical_loss": 3.546875544584861,
"tokens_seen": 1360003072
},
{
"epoch": 0.41,
"learning_rate": 0.000296950852557673,
"loss": 2.7799,
"theoretical_loss": 3.546860190702355,
"tokens_seen": 1360068608
},
{
"epoch": 0.41,
"learning_rate": 0.0002969408224674022,
"loss": 2.5751,
"theoretical_loss": 3.5468448377668134,
"tokens_seen": 1360134144
},
{
"epoch": 0.41,
"learning_rate": 0.0002969307923771314,
"loss": 2.8268,
"theoretical_loss": 3.5468294857781326,
"tokens_seen": 1360199680
},
{
"epoch": 0.41,
"learning_rate": 0.0002969207622868606,
"loss": 2.6486,
"theoretical_loss": 3.546814134736209,
"tokens_seen": 1360265216
},
{
"epoch": 0.41,
"learning_rate": 0.00029691073219658976,
"loss": 2.7086,
"theoretical_loss": 3.546798784640939,
"tokens_seen": 1360330752
},
{
"epoch": 0.41,
"learning_rate": 0.000296900702106319,
"loss": 2.569,
"theoretical_loss": 3.5467834354922165,
"tokens_seen": 1360396288
},
{
"epoch": 0.41,
"learning_rate": 0.0002968906720160481,
"loss": 2.856,
"theoretical_loss": 3.54676808728994,
"tokens_seen": 1360461824
},
{
"epoch": 0.41,
"learning_rate": 0.00029688064192577736,
"loss": 2.5984,
"theoretical_loss": 3.546752740034004,
"tokens_seen": 1360527360
},
{
"epoch": 0.41,
"learning_rate": 0.0002968706118355065,
"loss": 2.7462,
"theoretical_loss": 3.546737393724305,
"tokens_seen": 1360592896
},
{
"epoch": 0.41,
"learning_rate": 0.0002968605817452357,
"loss": 2.8904,
"theoretical_loss": 3.5467220483607393,
"tokens_seen": 1360658432
},
{
"epoch": 0.41,
"learning_rate": 0.0002968505516549649,
"loss": 2.8539,
"theoretical_loss": 3.546706703943203,
"tokens_seen": 1360723968
},
{
"epoch": 0.41,
"learning_rate": 0.0002968405215646941,
"loss": 2.8026,
"theoretical_loss": 3.546691360471592,
"tokens_seen": 1360789504
},
{
"epoch": 0.41,
"learning_rate": 0.00029683049147442327,
"loss": 2.655,
"theoretical_loss": 3.546676017945803,
"tokens_seen": 1360855040
},
{
"epoch": 0.41,
"learning_rate": 0.00029682046138415245,
"loss": 2.7597,
"theoretical_loss": 3.5466606763657316,
"tokens_seen": 1360920576
},
{
"epoch": 0.41,
"learning_rate": 0.00029681043129388163,
"loss": 2.6457,
"theoretical_loss": 3.5466453357312737,
"tokens_seen": 1360986112
},
{
"epoch": 0.41,
"learning_rate": 0.00029680040120361087,
"loss": 2.8325,
"theoretical_loss": 3.5466299960423258,
"tokens_seen": 1361051648
},
{
"epoch": 0.41,
"learning_rate": 0.00029679037111334,
"loss": 2.6247,
"theoretical_loss": 3.5466146572987842,
"tokens_seen": 1361117184
},
{
"epoch": 0.41,
"learning_rate": 0.00029678034102306923,
"loss": 2.7892,
"theoretical_loss": 3.5465993195005456,
"tokens_seen": 1361182720
},
{
"epoch": 0.41,
"learning_rate": 0.0002967703109327984,
"loss": 3.0455,
"theoretical_loss": 3.5465839826475047,
"tokens_seen": 1361248256
},
{
"epoch": 0.41,
"learning_rate": 0.0002967602808425276,
"loss": 2.5638,
"theoretical_loss": 3.5465686467395594,
"tokens_seen": 1361313792
},
{
"epoch": 0.41,
"learning_rate": 0.00029675025075225677,
"loss": 2.6878,
"theoretical_loss": 3.5465533117766053,
"tokens_seen": 1361379328
},
{
"epoch": 0.41,
"learning_rate": 0.00029674022066198595,
"loss": 2.7423,
"theoretical_loss": 3.5465379777585384,
"tokens_seen": 1361444864
},
{
"epoch": 0.41,
"objective/train/docs_used": 770761,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.751443862915039,
"objective/train/theoretical_loss": 3.546522644685256,
"objective/train/tokens_used": 1381970400,
"theoretical_loss": 3.546522644685256,
"tokens_seen": 1361510400
},
{
"epoch": 0.41,
"learning_rate": 0.00029673019057171513,
"loss": 2.6285,
"theoretical_loss": 3.546522644685256,
"tokens_seen": 1361510400
},
{
"epoch": 0.41,
"learning_rate": 0.00029672016048144437,
"loss": 2.6354,
"theoretical_loss": 3.5465073125566526,
"tokens_seen": 1361575936
},
{
"epoch": 0.41,
"learning_rate": 0.0002967101303911735,
"loss": 2.7158,
"theoretical_loss": 3.5464919813726263,
"tokens_seen": 1361641472
},
{
"epoch": 0.41,
"learning_rate": 0.00029670010030090273,
"loss": 2.6443,
"theoretical_loss": 3.546476651133073,
"tokens_seen": 1361707008
},
{
"epoch": 0.41,
"learning_rate": 0.00029669007021063186,
"loss": 2.8097,
"theoretical_loss": 3.5464613218378886,
"tokens_seen": 1361772544
},
{
"epoch": 0.41,
"learning_rate": 0.0002966800401203611,
"loss": 2.6944,
"theoretical_loss": 3.5464459934869694,
"tokens_seen": 1361838080
},
{
"epoch": 0.41,
"learning_rate": 0.0002966700100300903,
"loss": 2.6159,
"theoretical_loss": 3.5464306660802123,
"tokens_seen": 1361903616
},
{
"epoch": 0.41,
"learning_rate": 0.00029665997993981946,
"loss": 2.7461,
"theoretical_loss": 3.5464153396175138,
"tokens_seen": 1361969152
},
{
"epoch": 0.41,
"learning_rate": 0.00029664994984954864,
"loss": 2.6038,
"theoretical_loss": 3.54640001409877,
"tokens_seen": 1362034688
},
{
"epoch": 0.41,
"learning_rate": 0.0002966399197592778,
"loss": 2.7161,
"theoretical_loss": 3.5463846895238773,
"tokens_seen": 1362100224
},
{
"epoch": 0.41,
"learning_rate": 0.000296629889669007,
"loss": 2.515,
"theoretical_loss": 3.5463693658927324,
"tokens_seen": 1362165760
},
{
"epoch": 0.41,
"learning_rate": 0.00029661985957873623,
"loss": 2.6667,
"theoretical_loss": 3.5463540432052314,
"tokens_seen": 1362231296
},
{
"epoch": 0.41,
"learning_rate": 0.00029660982948846536,
"loss": 2.5314,
"theoretical_loss": 3.546338721461271,
"tokens_seen": 1362296832
},
{
"epoch": 0.41,
"learning_rate": 0.0002965997993981946,
"loss": 2.7056,
"theoretical_loss": 3.5463234006607482,
"tokens_seen": 1362362368
},
{
"epoch": 0.41,
"learning_rate": 0.0002965897693079238,
"loss": 2.4547,
"theoretical_loss": 3.546308080803559,
"tokens_seen": 1362427904
},
{
"epoch": 0.41,
"learning_rate": 0.00029657973921765296,
"loss": 2.457,
"theoretical_loss": 3.5462927618895996,
"tokens_seen": 1362493440
},
{
"epoch": 0.41,
"learning_rate": 0.00029656970912738214,
"loss": 2.6059,
"theoretical_loss": 3.546277443918768,
"tokens_seen": 1362558976
},
{
"epoch": 0.41,
"learning_rate": 0.0002965596790371113,
"loss": 2.5763,
"theoretical_loss": 3.5462621268909587,
"tokens_seen": 1362624512
},
{
"epoch": 0.41,
"learning_rate": 0.0002965496489468405,
"loss": 2.7864,
"theoretical_loss": 3.5462468108060703,
"tokens_seen": 1362690048
},
{
"epoch": 0.41,
"learning_rate": 0.00029653961885656974,
"loss": 2.8341,
"theoretical_loss": 3.5462314956639975,
"tokens_seen": 1362755584
},
{
"epoch": 0.41,
"learning_rate": 0.0002965295887662989,
"loss": 2.6833,
"theoretical_loss": 3.5462161814646387,
"tokens_seen": 1362821120
},
{
"epoch": 0.41,
"learning_rate": 0.0002965195586760281,
"loss": 2.884,
"theoretical_loss": 3.5462008682078894,
"tokens_seen": 1362886656
},
{
"epoch": 0.41,
"learning_rate": 0.0002965095285857573,
"loss": 2.6461,
"theoretical_loss": 3.546185555893647,
"tokens_seen": 1362952192
},
{
"epoch": 0.41,
"learning_rate": 0.00029649949849548646,
"loss": 2.4623,
"theoretical_loss": 3.5461702445218073,
"tokens_seen": 1363017728
},
{
"epoch": 0.41,
"learning_rate": 0.0002964894684052157,
"loss": 2.7135,
"theoretical_loss": 3.5461549340922676,
"tokens_seen": 1363083264
},
{
"epoch": 0.41,
"objective/train/docs_used": 771430,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0539212226867676,
"objective/train/theoretical_loss": 3.5461396246049244,
"objective/train/tokens_used": 1383608800,
"theoretical_loss": 3.5461396246049244,
"tokens_seen": 1363148800
},
{
"epoch": 0.41,
"learning_rate": 0.0002964794383149448,
"loss": 2.8598,
"theoretical_loss": 3.5461396246049244,
"tokens_seen": 1363148800
},
{
"epoch": 0.41,
"learning_rate": 0.00029646940822467406,
"loss": 2.4655,
"theoretical_loss": 3.546124316059675,
"tokens_seen": 1363214336
},
{
"epoch": 0.41,
"learning_rate": 0.0002964593781344032,
"loss": 2.7003,
"theoretical_loss": 3.546109008456415,
"tokens_seen": 1363279872
},
{
"epoch": 0.41,
"learning_rate": 0.0002964493480441324,
"loss": 2.7464,
"theoretical_loss": 3.5460937017950425,
"tokens_seen": 1363345408
},
{
"epoch": 0.41,
"learning_rate": 0.0002964393179538616,
"loss": 2.8109,
"theoretical_loss": 3.546078396075453,
"tokens_seen": 1363410944
},
{
"epoch": 0.41,
"learning_rate": 0.0002964292878635908,
"loss": 2.7751,
"theoretical_loss": 3.546063091297544,
"tokens_seen": 1363476480
},
{
"epoch": 0.41,
"learning_rate": 0.00029641925777331996,
"loss": 2.7234,
"theoretical_loss": 3.5460477874612124,
"tokens_seen": 1363542016
},
{
"epoch": 0.41,
"learning_rate": 0.0002964092276830492,
"loss": 2.8947,
"theoretical_loss": 3.5460324845663544,
"tokens_seen": 1363607552
},
{
"epoch": 0.41,
"learning_rate": 0.0002963991975927783,
"loss": 2.52,
"theoretical_loss": 3.5460171826128675,
"tokens_seen": 1363673088
},
{
"epoch": 0.41,
"learning_rate": 0.00029638916750250756,
"loss": 2.6371,
"theoretical_loss": 3.5460018816006484,
"tokens_seen": 1363738624
},
{
"epoch": 0.41,
"learning_rate": 0.0002963791374122367,
"loss": 2.8417,
"theoretical_loss": 3.5459865815295935,
"tokens_seen": 1363804160
},
{
"epoch": 0.41,
"learning_rate": 0.0002963691073219659,
"loss": 2.8137,
"theoretical_loss": 3.5459712823996004,
"tokens_seen": 1363869696
},
{
"epoch": 0.41,
"learning_rate": 0.0002963590772316951,
"loss": 2.8462,
"theoretical_loss": 3.5459559842105657,
"tokens_seen": 1363935232
},
{
"epoch": 0.41,
"learning_rate": 0.0002963490471414243,
"loss": 2.7617,
"theoretical_loss": 3.5459406869623855,
"tokens_seen": 1364000768
},
{
"epoch": 0.41,
"learning_rate": 0.00029633901705115347,
"loss": 2.6973,
"theoretical_loss": 3.5459253906549586,
"tokens_seen": 1364066304
},
{
"epoch": 0.41,
"learning_rate": 0.00029632898696088265,
"loss": 2.4741,
"theoretical_loss": 3.54591009528818,
"tokens_seen": 1364131840
},
{
"epoch": 0.41,
"learning_rate": 0.00029631895687061183,
"loss": 2.5621,
"theoretical_loss": 3.545894800861948,
"tokens_seen": 1364197376
},
{
"epoch": 0.41,
"learning_rate": 0.00029630892678034107,
"loss": 2.54,
"theoretical_loss": 3.545879507376159,
"tokens_seen": 1364262912
},
{
"epoch": 0.41,
"learning_rate": 0.0002962988966900702,
"loss": 2.6231,
"theoretical_loss": 3.54586421483071,
"tokens_seen": 1364328448
},
{
"epoch": 0.41,
"learning_rate": 0.00029628886659979943,
"loss": 2.6424,
"theoretical_loss": 3.545848923225498,
"tokens_seen": 1364393984
},
{
"epoch": 0.41,
"learning_rate": 0.0002962788365095286,
"loss": 2.5287,
"theoretical_loss": 3.5458336325604205,
"tokens_seen": 1364459520
},
{
"epoch": 0.41,
"learning_rate": 0.0002962688064192578,
"loss": 2.6491,
"theoretical_loss": 3.545818342835374,
"tokens_seen": 1364525056
},
{
"epoch": 0.41,
"learning_rate": 0.00029625877632898697,
"loss": 2.62,
"theoretical_loss": 3.545803054050256,
"tokens_seen": 1364590592
},
{
"epoch": 0.41,
"learning_rate": 0.00029624874623871615,
"loss": 2.7765,
"theoretical_loss": 3.5457877662049633,
"tokens_seen": 1364656128
},
{
"epoch": 0.41,
"learning_rate": 0.00029623871614844533,
"loss": 2.6162,
"theoretical_loss": 3.545772479299393,
"tokens_seen": 1364721664
},
{
"epoch": 0.41,
"objective/train/docs_used": 772731,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.583862781524658,
"objective/train/theoretical_loss": 3.5457571933334417,
"objective/train/tokens_used": 1385247200,
"theoretical_loss": 3.5457571933334417,
"tokens_seen": 1364787200
},
{
"epoch": 0.41,
"learning_rate": 0.00029622868605817457,
"loss": 2.683,
"theoretical_loss": 3.5457571933334417,
"tokens_seen": 1364787200
},
{
"epoch": 0.41,
"learning_rate": 0.0002962186559679037,
"loss": 2.7512,
"theoretical_loss": 3.545741908307008,
"tokens_seen": 1364852736
},
{
"epoch": 0.41,
"learning_rate": 0.00029620862587763293,
"loss": 2.7339,
"theoretical_loss": 3.545726624219988,
"tokens_seen": 1364918272
},
{
"epoch": 0.41,
"learning_rate": 0.00029619859578736206,
"loss": 2.4535,
"theoretical_loss": 3.5457113410722787,
"tokens_seen": 1364983808
},
{
"epoch": 0.41,
"learning_rate": 0.0002961885656970913,
"loss": 2.7388,
"theoretical_loss": 3.5456960588637774,
"tokens_seen": 1365049344
},
{
"epoch": 0.41,
"learning_rate": 0.0002961785356068205,
"loss": 2.7631,
"theoretical_loss": 3.5456807775943817,
"tokens_seen": 1365114880
},
{
"epoch": 0.41,
"learning_rate": 0.00029616850551654966,
"loss": 2.7526,
"theoretical_loss": 3.5456654972639883,
"tokens_seen": 1365180416
},
{
"epoch": 0.41,
"learning_rate": 0.00029615847542627884,
"loss": 2.8129,
"theoretical_loss": 3.5456502178724953,
"tokens_seen": 1365245952
},
{
"epoch": 0.41,
"learning_rate": 0.000296148445336008,
"loss": 2.8538,
"theoretical_loss": 3.5456349394197986,
"tokens_seen": 1365311488
},
{
"epoch": 0.41,
"learning_rate": 0.0002961384152457372,
"loss": 2.7588,
"theoretical_loss": 3.545619661905797,
"tokens_seen": 1365377024
},
{
"epoch": 0.41,
"learning_rate": 0.00029612838515546643,
"loss": 2.5929,
"theoretical_loss": 3.5456043853303862,
"tokens_seen": 1365442560
},
{
"epoch": 0.41,
"learning_rate": 0.00029611835506519556,
"loss": 2.6533,
"theoretical_loss": 3.545589109693465,
"tokens_seen": 1365508096
},
{
"epoch": 0.41,
"learning_rate": 0.0002961083249749248,
"loss": 2.6795,
"theoretical_loss": 3.54557383499493,
"tokens_seen": 1365573632
},
{
"epoch": 0.41,
"learning_rate": 0.000296098294884654,
"loss": 2.6232,
"theoretical_loss": 3.545558561234678,
"tokens_seen": 1365639168
},
{
"epoch": 0.41,
"learning_rate": 0.00029608826479438316,
"loss": 2.7643,
"theoretical_loss": 3.545543288412607,
"tokens_seen": 1365704704
},
{
"epoch": 0.41,
"learning_rate": 0.00029607823470411234,
"loss": 2.3554,
"theoretical_loss": 3.5455280165286145,
"tokens_seen": 1365770240
},
{
"epoch": 0.41,
"learning_rate": 0.0002960682046138415,
"loss": 2.7736,
"theoretical_loss": 3.545512745582597,
"tokens_seen": 1365835776
},
{
"epoch": 0.41,
"learning_rate": 0.0002960581745235707,
"loss": 2.8273,
"theoretical_loss": 3.545497475574453,
"tokens_seen": 1365901312
},
{
"epoch": 0.41,
"learning_rate": 0.00029604814443329994,
"loss": 2.6343,
"theoretical_loss": 3.5454822065040794,
"tokens_seen": 1365966848
},
{
"epoch": 0.41,
"learning_rate": 0.00029603811434302906,
"loss": 2.7606,
"theoretical_loss": 3.5454669383713737,
"tokens_seen": 1366032384
},
{
"epoch": 0.41,
"learning_rate": 0.0002960280842527583,
"loss": 2.7351,
"theoretical_loss": 3.5454516711762327,
"tokens_seen": 1366097920
},
{
"epoch": 0.41,
"learning_rate": 0.0002960180541624874,
"loss": 2.7743,
"theoretical_loss": 3.5454364049185547,
"tokens_seen": 1366163456
},
{
"epoch": 0.41,
"learning_rate": 0.00029600802407221666,
"loss": 2.8133,
"theoretical_loss": 3.545421139598237,
"tokens_seen": 1366228992
},
{
"epoch": 0.41,
"learning_rate": 0.00029599799398194584,
"loss": 2.8305,
"theoretical_loss": 3.5454058752151765,
"tokens_seen": 1366294528
},
{
"epoch": 0.41,
"learning_rate": 0.000295987963891675,
"loss": 2.6717,
"theoretical_loss": 3.5453906117692715,
"tokens_seen": 1366360064
},
{
"epoch": 0.41,
"objective/train/docs_used": 773461,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.76285719871521,
"objective/train/theoretical_loss": 3.545375349260419,
"objective/train/tokens_used": 1386885600,
"theoretical_loss": 3.545375349260419,
"tokens_seen": 1366425600
},
{
"epoch": 0.41,
"learning_rate": 0.0002959779338014042,
"loss": 2.657,
"theoretical_loss": 3.545375349260419,
"tokens_seen": 1366425600
},
{
"epoch": 0.41,
"learning_rate": 0.0002959679037111334,
"loss": 2.666,
"theoretical_loss": 3.545360087688517,
"tokens_seen": 1366491136
},
{
"epoch": 0.41,
"learning_rate": 0.00029595787362086257,
"loss": 2.668,
"theoretical_loss": 3.545344827053462,
"tokens_seen": 1366556672
},
{
"epoch": 0.41,
"learning_rate": 0.0002959478435305918,
"loss": 2.7017,
"theoretical_loss": 3.5453295673551537,
"tokens_seen": 1366622208
},
{
"epoch": 0.41,
"learning_rate": 0.00029593781344032093,
"loss": 2.8002,
"theoretical_loss": 3.545314308593487,
"tokens_seen": 1366687744
},
{
"epoch": 0.41,
"learning_rate": 0.00029592778335005016,
"loss": 2.697,
"theoretical_loss": 3.5452990507683615,
"tokens_seen": 1366753280
},
{
"epoch": 0.41,
"learning_rate": 0.00029591775325977935,
"loss": 2.7421,
"theoretical_loss": 3.545283793879674,
"tokens_seen": 1366818816
},
{
"epoch": 0.41,
"learning_rate": 0.00029590772316950853,
"loss": 2.7339,
"theoretical_loss": 3.545268537927322,
"tokens_seen": 1366884352
},
{
"epoch": 0.41,
"learning_rate": 0.0002958976930792377,
"loss": 2.7107,
"theoretical_loss": 3.5452532829112036,
"tokens_seen": 1366949888
},
{
"epoch": 0.41,
"learning_rate": 0.0002958876629889669,
"loss": 2.755,
"theoretical_loss": 3.545238028831216,
"tokens_seen": 1367015424
},
{
"epoch": 0.41,
"learning_rate": 0.00029587763289869607,
"loss": 2.8069,
"theoretical_loss": 3.5452227756872574,
"tokens_seen": 1367080960
},
{
"epoch": 0.41,
"learning_rate": 0.0002958676028084253,
"loss": 2.6949,
"theoretical_loss": 3.545207523479225,
"tokens_seen": 1367146496
},
{
"epoch": 0.41,
"learning_rate": 0.00029585757271815443,
"loss": 2.8857,
"theoretical_loss": 3.545192272207017,
"tokens_seen": 1367212032
},
{
"epoch": 0.41,
"learning_rate": 0.00029584754262788367,
"loss": 2.6388,
"theoretical_loss": 3.545177021870531,
"tokens_seen": 1367277568
},
{
"epoch": 0.41,
"learning_rate": 0.0002958375125376128,
"loss": 2.5637,
"theoretical_loss": 3.5451617724696645,
"tokens_seen": 1367343104
},
{
"epoch": 0.41,
"learning_rate": 0.00029582748244734203,
"loss": 2.6331,
"theoretical_loss": 3.5451465240043145,
"tokens_seen": 1367408640
},
{
"epoch": 0.41,
"learning_rate": 0.0002958174523570712,
"loss": 2.567,
"theoretical_loss": 3.5451312764743808,
"tokens_seen": 1367474176
},
{
"epoch": 0.41,
"learning_rate": 0.0002958074222668004,
"loss": 2.4132,
"theoretical_loss": 3.5451160298797593,
"tokens_seen": 1367539712
},
{
"epoch": 0.41,
"learning_rate": 0.0002957973921765296,
"loss": 2.6808,
"theoretical_loss": 3.545100784220349,
"tokens_seen": 1367605248
},
{
"epoch": 0.41,
"learning_rate": 0.0002957873620862588,
"loss": 2.4931,
"theoretical_loss": 3.545085539496047,
"tokens_seen": 1367670784
},
{
"epoch": 0.41,
"learning_rate": 0.000295777331995988,
"loss": 2.6734,
"theoretical_loss": 3.5450702957067515,
"tokens_seen": 1367736320
},
{
"epoch": 0.41,
"learning_rate": 0.00029576730190571717,
"loss": 2.7366,
"theoretical_loss": 3.5450550528523603,
"tokens_seen": 1367801856
},
{
"epoch": 0.41,
"learning_rate": 0.00029575727181544635,
"loss": 2.8125,
"theoretical_loss": 3.545039810932771,
"tokens_seen": 1367867392
},
{
"epoch": 0.41,
"learning_rate": 0.00029574724172517553,
"loss": 2.8055,
"theoretical_loss": 3.5450245699478824,
"tokens_seen": 1367932928
},
{
"epoch": 0.41,
"learning_rate": 0.00029573721163490477,
"loss": 3.0026,
"theoretical_loss": 3.5450093298975913,
"tokens_seen": 1367998464
},
{
"epoch": 0.41,
"objective/train/docs_used": 774789,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.805393934249878,
"objective/train/theoretical_loss": 3.544994090781796,
"objective/train/tokens_used": 1388524000,
"theoretical_loss": 3.544994090781796,
"tokens_seen": 1368064000
},
{
"epoch": 0.41,
"learning_rate": 0.0002957271815446339,
"loss": 2.6313,
"theoretical_loss": 3.544994090781796,
"tokens_seen": 1368064000
},
{
"epoch": 0.41,
"learning_rate": 0.00029571715145436313,
"loss": 2.7141,
"theoretical_loss": 3.5449788526003942,
"tokens_seen": 1368129536
},
{
"epoch": 0.41,
"learning_rate": 0.00029570712136409226,
"loss": 2.7291,
"theoretical_loss": 3.5449636153532844,
"tokens_seen": 1368195072
},
{
"epoch": 0.41,
"learning_rate": 0.0002956970912738215,
"loss": 2.8375,
"theoretical_loss": 3.544948379040364,
"tokens_seen": 1368260608
},
{
"epoch": 0.41,
"learning_rate": 0.0002956870611835507,
"loss": 2.625,
"theoretical_loss": 3.544933143661532,
"tokens_seen": 1368326144
},
{
"epoch": 0.41,
"learning_rate": 0.00029567703109327986,
"loss": 2.7753,
"theoretical_loss": 3.5449179092166854,
"tokens_seen": 1368391680
},
{
"epoch": 0.41,
"learning_rate": 0.00029566700100300904,
"loss": 2.7018,
"theoretical_loss": 3.5449026757057225,
"tokens_seen": 1368457216
},
{
"epoch": 0.41,
"learning_rate": 0.0002956569709127382,
"loss": 2.5247,
"theoretical_loss": 3.544887443128541,
"tokens_seen": 1368522752
},
{
"epoch": 0.41,
"learning_rate": 0.0002956469408224674,
"loss": 2.563,
"theoretical_loss": 3.5448722114850395,
"tokens_seen": 1368588288
},
{
"epoch": 0.41,
"learning_rate": 0.00029563691073219663,
"loss": 2.8486,
"theoretical_loss": 3.544856980775116,
"tokens_seen": 1368653824
},
{
"epoch": 0.41,
"learning_rate": 0.00029562688064192576,
"loss": 2.7444,
"theoretical_loss": 3.5448417509986685,
"tokens_seen": 1368719360
},
{
"epoch": 0.41,
"learning_rate": 0.000295616850551655,
"loss": 2.6184,
"theoretical_loss": 3.544826522155595,
"tokens_seen": 1368784896
},
{
"epoch": 0.41,
"learning_rate": 0.0002956068204613842,
"loss": 2.6598,
"theoretical_loss": 3.544811294245793,
"tokens_seen": 1368850432
},
{
"epoch": 0.41,
"learning_rate": 0.00029559679037111336,
"loss": 2.5873,
"theoretical_loss": 3.5447960672691616,
"tokens_seen": 1368915968
},
{
"epoch": 0.41,
"learning_rate": 0.00029558676028084254,
"loss": 2.5046,
"theoretical_loss": 3.5447808412255988,
"tokens_seen": 1368981504
},
{
"epoch": 0.41,
"learning_rate": 0.0002955767301905717,
"loss": 2.6376,
"theoretical_loss": 3.5447656161150025,
"tokens_seen": 1369047040
},
{
"epoch": 0.41,
"learning_rate": 0.0002955667001003009,
"loss": 2.6882,
"theoretical_loss": 3.544750391937271,
"tokens_seen": 1369112576
},
{
"epoch": 0.41,
"learning_rate": 0.00029555667001003014,
"loss": 2.6794,
"theoretical_loss": 3.5447351686923025,
"tokens_seen": 1369178112
},
{
"epoch": 0.41,
"learning_rate": 0.00029554663991975926,
"loss": 2.6166,
"theoretical_loss": 3.544719946379995,
"tokens_seen": 1369243648
},
{
"epoch": 0.41,
"learning_rate": 0.0002955366098294885,
"loss": 2.6289,
"theoretical_loss": 3.544704725000247,
"tokens_seen": 1369309184
},
{
"epoch": 0.41,
"learning_rate": 0.0002955265797392176,
"loss": 2.5595,
"theoretical_loss": 3.5446895045529567,
"tokens_seen": 1369374720
},
{
"epoch": 0.41,
"learning_rate": 0.00029551654964894686,
"loss": 2.5521,
"theoretical_loss": 3.5446742850380217,
"tokens_seen": 1369440256
},
{
"epoch": 0.42,
"learning_rate": 0.00029550651955867604,
"loss": 2.7432,
"theoretical_loss": 3.544659066455341,
"tokens_seen": 1369505792
},
{
"epoch": 0.42,
"learning_rate": 0.0002954964894684052,
"loss": 2.6847,
"theoretical_loss": 3.5446438488048133,
"tokens_seen": 1369571328
},
{
"epoch": 0.42,
"learning_rate": 0.0002954864593781344,
"loss": 2.6526,
"theoretical_loss": 3.5446286320863356,
"tokens_seen": 1369636864
},
{
"epoch": 0.42,
"objective/train/docs_used": 775516,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.590693950653076,
"objective/train/theoretical_loss": 3.544613416299808,
"objective/train/tokens_used": 1390162400,
"theoretical_loss": 3.544613416299808,
"tokens_seen": 1369702400
},
{
"epoch": 0.42,
"learning_rate": 0.0002954764292878636,
"loss": 2.7433,
"theoretical_loss": 3.544613416299808,
"tokens_seen": 1369702400
},
{
"epoch": 0.42,
"learning_rate": 0.00029546639919759277,
"loss": 2.6204,
"theoretical_loss": 3.544598201445127,
"tokens_seen": 1369767936
},
{
"epoch": 0.42,
"learning_rate": 0.000295456369107322,
"loss": 2.8203,
"theoretical_loss": 3.5445829875221917,
"tokens_seen": 1369833472
},
{
"epoch": 0.42,
"learning_rate": 0.00029544633901705113,
"loss": 2.6493,
"theoretical_loss": 3.5445677745309006,
"tokens_seen": 1369899008
},
{
"epoch": 0.42,
"learning_rate": 0.00029543630892678037,
"loss": 2.5849,
"theoretical_loss": 3.544552562471152,
"tokens_seen": 1369964544
},
{
"epoch": 0.42,
"learning_rate": 0.00029542627883650955,
"loss": 2.4501,
"theoretical_loss": 3.544537351342844,
"tokens_seen": 1370030080
},
{
"epoch": 0.42,
"learning_rate": 0.00029541624874623873,
"loss": 2.4979,
"theoretical_loss": 3.544522141145876,
"tokens_seen": 1370095616
},
{
"epoch": 0.42,
"learning_rate": 0.0002954062186559679,
"loss": 2.9187,
"theoretical_loss": 3.5445069318801457,
"tokens_seen": 1370161152
},
{
"epoch": 0.42,
"learning_rate": 0.0002953961885656971,
"loss": 2.6342,
"theoretical_loss": 3.544491723545551,
"tokens_seen": 1370226688
},
{
"epoch": 0.42,
"learning_rate": 0.00029538615847542627,
"loss": 2.5932,
"theoretical_loss": 3.5444765161419913,
"tokens_seen": 1370292224
},
{
"epoch": 0.42,
"learning_rate": 0.0002953761283851555,
"loss": 2.5035,
"theoretical_loss": 3.5444613096693645,
"tokens_seen": 1370357760
},
{
"epoch": 0.42,
"learning_rate": 0.00029536609829488463,
"loss": 2.682,
"theoretical_loss": 3.5444461041275694,
"tokens_seen": 1370423296
},
{
"epoch": 0.42,
"learning_rate": 0.00029535606820461387,
"loss": 2.8673,
"theoretical_loss": 3.5444308995165046,
"tokens_seen": 1370488832
},
{
"epoch": 0.42,
"learning_rate": 0.000295346038114343,
"loss": 2.5488,
"theoretical_loss": 3.544415695836068,
"tokens_seen": 1370554368
},
{
"epoch": 0.42,
"learning_rate": 0.00029533600802407223,
"loss": 2.9489,
"theoretical_loss": 3.544400493086159,
"tokens_seen": 1370619904
},
{
"epoch": 0.42,
"learning_rate": 0.0002953259779338014,
"loss": 2.6802,
"theoretical_loss": 3.544385291266676,
"tokens_seen": 1370685440
},
{
"epoch": 0.42,
"learning_rate": 0.0002953159478435306,
"loss": 2.5938,
"theoretical_loss": 3.544370090377517,
"tokens_seen": 1370750976
},
{
"epoch": 0.42,
"learning_rate": 0.0002953059177532598,
"loss": 2.8505,
"theoretical_loss": 3.544354890418581,
"tokens_seen": 1370816512
},
{
"epoch": 0.42,
"learning_rate": 0.000295295887662989,
"loss": 2.5472,
"theoretical_loss": 3.5443396913897667,
"tokens_seen": 1370882048
},
{
"epoch": 0.42,
"learning_rate": 0.00029528585757271814,
"loss": 2.7028,
"theoretical_loss": 3.544324493290972,
"tokens_seen": 1370947584
},
{
"epoch": 0.42,
"learning_rate": 0.00029527582748244737,
"loss": 2.8363,
"theoretical_loss": 3.5443092961220968,
"tokens_seen": 1371013120
},
{
"epoch": 0.42,
"learning_rate": 0.0002952657973921765,
"loss": 2.7637,
"theoretical_loss": 3.5442940998830386,
"tokens_seen": 1371078656
},
{
"epoch": 0.42,
"learning_rate": 0.00029525576730190573,
"loss": 2.8298,
"theoretical_loss": 3.544278904573697,
"tokens_seen": 1371144192
},
{
"epoch": 0.42,
"learning_rate": 0.0002952457372116349,
"loss": 2.7616,
"theoretical_loss": 3.54426371019397,
"tokens_seen": 1371209728
},
{
"epoch": 0.42,
"learning_rate": 0.0002952357071213641,
"loss": 2.6977,
"theoretical_loss": 3.544248516743756,
"tokens_seen": 1371275264
},
{
"epoch": 0.42,
"objective/train/docs_used": 776338,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2038328647613525,
"objective/train/theoretical_loss": 3.5442333242229545,
"objective/train/tokens_used": 1391800800,
"theoretical_loss": 3.5442333242229545,
"tokens_seen": 1371340800
},
{
"epoch": 0.42,
"learning_rate": 0.0002952256770310933,
"loss": 2.5068,
"theoretical_loss": 3.5442333242229545,
"tokens_seen": 1371340800
},
{
"epoch": 0.42,
"learning_rate": 0.00029521564694082246,
"loss": 2.6847,
"theoretical_loss": 3.544218132631464,
"tokens_seen": 1371406336
},
{
"epoch": 0.42,
"learning_rate": 0.00029520561685055164,
"loss": 2.7255,
"theoretical_loss": 3.5442029419691834,
"tokens_seen": 1371471872
},
{
"epoch": 0.42,
"learning_rate": 0.0002951955867602809,
"loss": 2.6787,
"theoretical_loss": 3.5441877522360112,
"tokens_seen": 1371537408
},
{
"epoch": 0.42,
"learning_rate": 0.00029518555667001,
"loss": 2.51,
"theoretical_loss": 3.544172563431846,
"tokens_seen": 1371602944
},
{
"epoch": 0.42,
"learning_rate": 0.00029517552657973924,
"loss": 2.5442,
"theoretical_loss": 3.544157375556587,
"tokens_seen": 1371668480
},
{
"epoch": 0.42,
"learning_rate": 0.00029516549648946836,
"loss": 2.5799,
"theoretical_loss": 3.5441421886101336,
"tokens_seen": 1371734016
},
{
"epoch": 0.42,
"learning_rate": 0.0002951554663991976,
"loss": 2.7774,
"theoretical_loss": 3.544127002592383,
"tokens_seen": 1371799552
},
{
"epoch": 0.42,
"learning_rate": 0.0002951454363089268,
"loss": 2.6955,
"theoretical_loss": 3.544111817503235,
"tokens_seen": 1371865088
},
{
"epoch": 0.42,
"learning_rate": 0.00029513540621865596,
"loss": 2.6577,
"theoretical_loss": 3.5440966333425887,
"tokens_seen": 1371930624
},
{
"epoch": 0.42,
"learning_rate": 0.00029512537612838514,
"loss": 2.6909,
"theoretical_loss": 3.544081450110342,
"tokens_seen": 1371996160
},
{
"epoch": 0.42,
"learning_rate": 0.0002951153460381144,
"loss": 2.6876,
"theoretical_loss": 3.5440662678063957,
"tokens_seen": 1372061696
},
{
"epoch": 0.42,
"learning_rate": 0.0002951053159478435,
"loss": 2.7052,
"theoretical_loss": 3.5440510864306467,
"tokens_seen": 1372127232
},
{
"epoch": 0.42,
"learning_rate": 0.00029509528585757274,
"loss": 2.5855,
"theoretical_loss": 3.544035905982995,
"tokens_seen": 1372192768
},
{
"epoch": 0.42,
"learning_rate": 0.00029508525576730187,
"loss": 2.806,
"theoretical_loss": 3.544020726463339,
"tokens_seen": 1372258304
},
{
"epoch": 0.42,
"learning_rate": 0.0002950752256770311,
"loss": 2.6842,
"theoretical_loss": 3.5440055478715777,
"tokens_seen": 1372323840
},
{
"epoch": 0.42,
"learning_rate": 0.0002950651955867603,
"loss": 2.6767,
"theoretical_loss": 3.5439903702076108,
"tokens_seen": 1372389376
},
{
"epoch": 0.42,
"learning_rate": 0.00029505516549648946,
"loss": 2.4704,
"theoretical_loss": 3.5439751934713364,
"tokens_seen": 1372454912
},
{
"epoch": 0.42,
"learning_rate": 0.00029504513540621865,
"loss": 2.4159,
"theoretical_loss": 3.543960017662654,
"tokens_seen": 1372520448
},
{
"epoch": 0.42,
"learning_rate": 0.0002950351053159478,
"loss": 2.5305,
"theoretical_loss": 3.543944842781462,
"tokens_seen": 1372585984
},
{
"epoch": 0.42,
"learning_rate": 0.00029502507522567706,
"loss": 2.6152,
"theoretical_loss": 3.5439296688276607,
"tokens_seen": 1372651520
},
{
"epoch": 0.42,
"learning_rate": 0.00029501504513540624,
"loss": 2.6741,
"theoretical_loss": 3.5439144958011477,
"tokens_seen": 1372717056
},
{
"epoch": 0.42,
"learning_rate": 0.0002950050150451354,
"loss": 2.6453,
"theoretical_loss": 3.5438993237018233,
"tokens_seen": 1372782592
},
{
"epoch": 0.42,
"learning_rate": 0.0002949949849548646,
"loss": 2.7986,
"theoretical_loss": 3.5438841525295857,
"tokens_seen": 1372848128
},
{
"epoch": 0.42,
"learning_rate": 0.0002949849548645938,
"loss": 2.3249,
"theoretical_loss": 3.5438689822843346,
"tokens_seen": 1372913664
},
{
"epoch": 0.42,
"objective/train/docs_used": 777606,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0391976833343506,
"objective/train/theoretical_loss": 3.5438538129659687,
"objective/train/tokens_used": 1393439200,
"theoretical_loss": 3.5438538129659687,
"tokens_seen": 1372979200
},
{
"epoch": 0.42,
"learning_rate": 0.00029497492477432297,
"loss": 2.7255,
"theoretical_loss": 3.5438538129659687,
"tokens_seen": 1372979200
},
{
"epoch": 0.42,
"learning_rate": 0.0002949648946840522,
"loss": 2.7608,
"theoretical_loss": 3.543838644574387,
"tokens_seen": 1373044736
},
{
"epoch": 0.42,
"learning_rate": 0.00029495486459378133,
"loss": 2.7679,
"theoretical_loss": 3.543823477109489,
"tokens_seen": 1373110272
},
{
"epoch": 0.42,
"learning_rate": 0.00029494483450351057,
"loss": 2.6684,
"theoretical_loss": 3.543808310571174,
"tokens_seen": 1373175808
},
{
"epoch": 0.42,
"learning_rate": 0.00029493480441323975,
"loss": 2.7149,
"theoretical_loss": 3.5437931449593405,
"tokens_seen": 1373241344
},
{
"epoch": 0.42,
"learning_rate": 0.00029492477432296893,
"loss": 2.6783,
"theoretical_loss": 3.5437779802738887,
"tokens_seen": 1373306880
},
{
"epoch": 0.42,
"learning_rate": 0.0002949147442326981,
"loss": 2.671,
"theoretical_loss": 3.543762816514717,
"tokens_seen": 1373372416
},
{
"epoch": 0.42,
"learning_rate": 0.0002949047141424273,
"loss": 2.6672,
"theoretical_loss": 3.5437476536817245,
"tokens_seen": 1373437952
},
{
"epoch": 0.42,
"learning_rate": 0.00029489468405215647,
"loss": 2.7237,
"theoretical_loss": 3.5437324917748114,
"tokens_seen": 1373503488
},
{
"epoch": 0.42,
"learning_rate": 0.0002948846539618857,
"loss": 2.7714,
"theoretical_loss": 3.543717330793876,
"tokens_seen": 1373569024
},
{
"epoch": 0.42,
"learning_rate": 0.00029487462387161483,
"loss": 2.8441,
"theoretical_loss": 3.543702170738818,
"tokens_seen": 1373634560
},
{
"epoch": 0.42,
"learning_rate": 0.00029486459378134407,
"loss": 2.7025,
"theoretical_loss": 3.5436870116095367,
"tokens_seen": 1373700096
},
{
"epoch": 0.42,
"learning_rate": 0.0002948545636910732,
"loss": 2.6888,
"theoretical_loss": 3.543671853405931,
"tokens_seen": 1373765632
},
{
"epoch": 0.42,
"learning_rate": 0.00029484453360080243,
"loss": 2.7312,
"theoretical_loss": 3.543656696127901,
"tokens_seen": 1373831168
},
{
"epoch": 0.42,
"learning_rate": 0.0002948345035105316,
"loss": 2.839,
"theoretical_loss": 3.5436415397753454,
"tokens_seen": 1373896704
},
{
"epoch": 0.42,
"learning_rate": 0.0002948244734202608,
"loss": 2.755,
"theoretical_loss": 3.5436263843481637,
"tokens_seen": 1373962240
},
{
"epoch": 0.42,
"learning_rate": 0.00029481444332999,
"loss": 2.7661,
"theoretical_loss": 3.543611229846255,
"tokens_seen": 1374027776
},
{
"epoch": 0.42,
"learning_rate": 0.0002948044132397192,
"loss": 2.7571,
"theoretical_loss": 3.5435960762695196,
"tokens_seen": 1374093312
},
{
"epoch": 0.42,
"learning_rate": 0.00029479438314944834,
"loss": 2.8226,
"theoretical_loss": 3.543580923617856,
"tokens_seen": 1374158848
},
{
"epoch": 0.42,
"learning_rate": 0.00029478435305917757,
"loss": 2.788,
"theoretical_loss": 3.5435657718911635,
"tokens_seen": 1374224384
},
{
"epoch": 0.42,
"learning_rate": 0.0002947743229689067,
"loss": 2.8307,
"theoretical_loss": 3.5435506210893424,
"tokens_seen": 1374289920
},
{
"epoch": 0.42,
"learning_rate": 0.00029476429287863593,
"loss": 2.8162,
"theoretical_loss": 3.5435354712122917,
"tokens_seen": 1374355456
},
{
"epoch": 0.42,
"learning_rate": 0.0002947542627883651,
"loss": 2.734,
"theoretical_loss": 3.5435203222599103,
"tokens_seen": 1374420992
},
{
"epoch": 0.42,
"learning_rate": 0.0002947442326980943,
"loss": 2.6978,
"theoretical_loss": 3.5435051742320987,
"tokens_seen": 1374486528
},
{
"epoch": 0.42,
"learning_rate": 0.0002947342026078235,
"loss": 2.503,
"theoretical_loss": 3.5434900271287555,
"tokens_seen": 1374552064
},
{
"epoch": 0.42,
"objective/train/docs_used": 778164,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6816134452819824,
"objective/train/theoretical_loss": 3.5434748809497814,
"objective/train/tokens_used": 1395077600,
"theoretical_loss": 3.5434748809497814,
"tokens_seen": 1374617600
},
{
"epoch": 0.42,
"learning_rate": 0.00029472417251755266,
"loss": 2.6954,
"theoretical_loss": 3.5434748809497814,
"tokens_seen": 1374617600
},
{
"epoch": 0.42,
"learning_rate": 0.00029471414242728184,
"loss": 2.8582,
"theoretical_loss": 3.543459735695074,
"tokens_seen": 1374683136
},
{
"epoch": 0.42,
"learning_rate": 0.0002947041123370111,
"loss": 2.8721,
"theoretical_loss": 3.543444591364535,
"tokens_seen": 1374748672
},
{
"epoch": 0.42,
"learning_rate": 0.0002946940822467402,
"loss": 2.8491,
"theoretical_loss": 3.5434294479580624,
"tokens_seen": 1374814208
},
{
"epoch": 0.42,
"learning_rate": 0.00029468405215646944,
"loss": 2.7871,
"theoretical_loss": 3.5434143054755562,
"tokens_seen": 1374879744
},
{
"epoch": 0.42,
"learning_rate": 0.00029467402206619856,
"loss": 2.5729,
"theoretical_loss": 3.5433991639169165,
"tokens_seen": 1374945280
},
{
"epoch": 0.42,
"learning_rate": 0.0002946639919759278,
"loss": 2.9769,
"theoretical_loss": 3.5433840232820417,
"tokens_seen": 1375010816
},
{
"epoch": 0.42,
"learning_rate": 0.000294653961885657,
"loss": 2.62,
"theoretical_loss": 3.5433688835708326,
"tokens_seen": 1375076352
},
{
"epoch": 0.42,
"learning_rate": 0.00029464393179538616,
"loss": 2.8481,
"theoretical_loss": 3.5433537447831887,
"tokens_seen": 1375141888
},
{
"epoch": 0.42,
"learning_rate": 0.00029463390170511534,
"loss": 2.4481,
"theoretical_loss": 3.543338606919009,
"tokens_seen": 1375207424
},
{
"epoch": 0.42,
"learning_rate": 0.0002946238716148446,
"loss": 2.6294,
"theoretical_loss": 3.543323469978194,
"tokens_seen": 1375272960
},
{
"epoch": 0.42,
"learning_rate": 0.0002946138415245737,
"loss": 2.5583,
"theoretical_loss": 3.5433083339606424,
"tokens_seen": 1375338496
},
{
"epoch": 0.42,
"learning_rate": 0.00029460381143430294,
"loss": 2.8408,
"theoretical_loss": 3.543293198866255,
"tokens_seen": 1375404032
},
{
"epoch": 0.42,
"learning_rate": 0.00029459378134403207,
"loss": 2.4552,
"theoretical_loss": 3.5432780646949302,
"tokens_seen": 1375469568
},
{
"epoch": 0.42,
"learning_rate": 0.0002945837512537613,
"loss": 2.62,
"theoretical_loss": 3.543262931446569,
"tokens_seen": 1375535104
},
{
"epoch": 0.42,
"learning_rate": 0.0002945737211634905,
"loss": 2.7593,
"theoretical_loss": 3.5432477991210702,
"tokens_seen": 1375600640
},
{
"epoch": 0.42,
"learning_rate": 0.00029456369107321966,
"loss": 2.7814,
"theoretical_loss": 3.5432326677183346,
"tokens_seen": 1375666176
},
{
"epoch": 0.42,
"learning_rate": 0.00029455366098294885,
"loss": 2.6812,
"theoretical_loss": 3.5432175372382604,
"tokens_seen": 1375731712
},
{
"epoch": 0.42,
"learning_rate": 0.000294543630892678,
"loss": 2.7234,
"theoretical_loss": 3.543202407680749,
"tokens_seen": 1375797248
},
{
"epoch": 0.42,
"learning_rate": 0.0002945336008024072,
"loss": 2.6065,
"theoretical_loss": 3.543187279045699,
"tokens_seen": 1375862784
},
{
"epoch": 0.42,
"learning_rate": 0.00029452357071213644,
"loss": 2.544,
"theoretical_loss": 3.543172151333011,
"tokens_seen": 1375928320
},
{
"epoch": 0.42,
"learning_rate": 0.00029451354062186557,
"loss": 2.7469,
"theoretical_loss": 3.543157024542585,
"tokens_seen": 1375993856
},
{
"epoch": 0.42,
"learning_rate": 0.0002945035105315948,
"loss": 2.5139,
"theoretical_loss": 3.54314189867432,
"tokens_seen": 1376059392
},
{
"epoch": 0.42,
"learning_rate": 0.00029449348044132393,
"loss": 2.6732,
"theoretical_loss": 3.5431267737281167,
"tokens_seen": 1376124928
},
{
"epoch": 0.42,
"learning_rate": 0.00029448345035105317,
"loss": 2.5688,
"theoretical_loss": 3.5431116497038744,
"tokens_seen": 1376190464
},
{
"epoch": 0.42,
"objective/train/docs_used": 779255,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8736460208892822,
"objective/train/theoretical_loss": 3.5430965266014933,
"objective/train/tokens_used": 1396716000,
"theoretical_loss": 3.5430965266014933,
"tokens_seen": 1376256000
},
{
"epoch": 0.42,
"learning_rate": 0.00029447342026078235,
"loss": 2.6803,
"theoretical_loss": 3.5430965266014933,
"tokens_seen": 1376256000
},
{
"epoch": 0.42,
"learning_rate": 0.00029446339017051153,
"loss": 2.7525,
"theoretical_loss": 3.5430814044208727,
"tokens_seen": 1376321536
},
{
"epoch": 0.42,
"learning_rate": 0.0002944533600802407,
"loss": 2.8,
"theoretical_loss": 3.5430662831619135,
"tokens_seen": 1376387072
},
{
"epoch": 0.42,
"learning_rate": 0.00029444332998996995,
"loss": 2.7121,
"theoretical_loss": 3.543051162824515,
"tokens_seen": 1376452608
},
{
"epoch": 0.42,
"learning_rate": 0.0002944332998996991,
"loss": 2.5388,
"theoretical_loss": 3.5430360434085775,
"tokens_seen": 1376518144
},
{
"epoch": 0.42,
"learning_rate": 0.0002944232698094283,
"loss": 2.6063,
"theoretical_loss": 3.543020924914001,
"tokens_seen": 1376583680
},
{
"epoch": 0.42,
"learning_rate": 0.00029441323971915744,
"loss": 2.6542,
"theoretical_loss": 3.543005807340685,
"tokens_seen": 1376649216
},
{
"epoch": 0.42,
"learning_rate": 0.00029440320962888667,
"loss": 2.7366,
"theoretical_loss": 3.542990690688529,
"tokens_seen": 1376714752
},
{
"epoch": 0.42,
"learning_rate": 0.00029439317953861585,
"loss": 2.8056,
"theoretical_loss": 3.542975574957435,
"tokens_seen": 1376780288
},
{
"epoch": 0.42,
"learning_rate": 0.00029438314944834503,
"loss": 2.6106,
"theoretical_loss": 3.5429604601473015,
"tokens_seen": 1376845824
},
{
"epoch": 0.42,
"learning_rate": 0.0002943731193580742,
"loss": 2.4316,
"theoretical_loss": 3.5429453462580294,
"tokens_seen": 1376911360
},
{
"epoch": 0.42,
"learning_rate": 0.0002943630892678034,
"loss": 2.6154,
"theoretical_loss": 3.542930233289518,
"tokens_seen": 1376976896
},
{
"epoch": 0.42,
"learning_rate": 0.0002943530591775326,
"loss": 2.8072,
"theoretical_loss": 3.5429151212416676,
"tokens_seen": 1377042432
},
{
"epoch": 0.42,
"learning_rate": 0.0002943430290872618,
"loss": 2.646,
"theoretical_loss": 3.542900010114378,
"tokens_seen": 1377107968
},
{
"epoch": 0.42,
"learning_rate": 0.00029433299899699094,
"loss": 2.7846,
"theoretical_loss": 3.5428848999075506,
"tokens_seen": 1377173504
},
{
"epoch": 0.42,
"learning_rate": 0.0002943229689067202,
"loss": 2.6837,
"theoretical_loss": 3.542869790621084,
"tokens_seen": 1377239040
},
{
"epoch": 0.42,
"learning_rate": 0.0002943129388164493,
"loss": 2.5057,
"theoretical_loss": 3.5428546822548794,
"tokens_seen": 1377304576
},
{
"epoch": 0.42,
"learning_rate": 0.00029430290872617854,
"loss": 2.6859,
"theoretical_loss": 3.542839574808837,
"tokens_seen": 1377370112
},
{
"epoch": 0.42,
"learning_rate": 0.00029429287863590777,
"loss": 2.7809,
"theoretical_loss": 3.5428244682828556,
"tokens_seen": 1377435648
},
{
"epoch": 0.42,
"learning_rate": 0.0002942828485456369,
"loss": 2.7848,
"theoretical_loss": 3.5428093626768367,
"tokens_seen": 1377501184
},
{
"epoch": 0.42,
"learning_rate": 0.00029427281845536613,
"loss": 2.6144,
"theoretical_loss": 3.54279425799068,
"tokens_seen": 1377566720
},
{
"epoch": 0.42,
"learning_rate": 0.0002942627883650953,
"loss": 2.6084,
"theoretical_loss": 3.542779154224286,
"tokens_seen": 1377632256
},
{
"epoch": 0.42,
"learning_rate": 0.0002942527582748245,
"loss": 2.6928,
"theoretical_loss": 3.5427640513775556,
"tokens_seen": 1377697792
},
{
"epoch": 0.42,
"learning_rate": 0.0002942427281845537,
"loss": 2.5952,
"theoretical_loss": 3.542748949450387,
"tokens_seen": 1377763328
},
{
"epoch": 0.42,
"learning_rate": 0.00029423269809428286,
"loss": 2.7365,
"theoretical_loss": 3.5427338484426825,
"tokens_seen": 1377828864
},
{
"epoch": 0.42,
"objective/train/docs_used": 780014,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4970860481262207,
"objective/train/theoretical_loss": 3.542718748354342,
"objective/train/tokens_used": 1398354400,
"theoretical_loss": 3.542718748354342,
"tokens_seen": 1377894400
},
{
"epoch": 0.42,
"learning_rate": 0.00029422266800401204,
"loss": 2.6445,
"theoretical_loss": 3.542718748354342,
"tokens_seen": 1377894400
},
{
"epoch": 0.42,
"learning_rate": 0.0002942126379137413,
"loss": 2.3343,
"theoretical_loss": 3.542703649185265,
"tokens_seen": 1377959936
},
{
"epoch": 0.42,
"learning_rate": 0.0002942026078234704,
"loss": 2.8109,
"theoretical_loss": 3.542688550935352,
"tokens_seen": 1378025472
},
{
"epoch": 0.42,
"learning_rate": 0.00029419257773319964,
"loss": 2.5726,
"theoretical_loss": 3.542673453604504,
"tokens_seen": 1378091008
},
{
"epoch": 0.42,
"learning_rate": 0.00029418254764292876,
"loss": 2.6478,
"theoretical_loss": 3.5426583571926207,
"tokens_seen": 1378156544
},
{
"epoch": 0.42,
"learning_rate": 0.000294172517552658,
"loss": 2.6946,
"theoretical_loss": 3.5426432616996033,
"tokens_seen": 1378222080
},
{
"epoch": 0.42,
"learning_rate": 0.0002941624874623872,
"loss": 2.5964,
"theoretical_loss": 3.5426281671253506,
"tokens_seen": 1378287616
},
{
"epoch": 0.42,
"learning_rate": 0.00029415245737211636,
"loss": 2.7322,
"theoretical_loss": 3.542613073469765,
"tokens_seen": 1378353152
},
{
"epoch": 0.42,
"learning_rate": 0.00029414242728184554,
"loss": 2.7464,
"theoretical_loss": 3.5425979807327455,
"tokens_seen": 1378418688
},
{
"epoch": 0.42,
"learning_rate": 0.0002941323971915748,
"loss": 2.9246,
"theoretical_loss": 3.5425828889141933,
"tokens_seen": 1378484224
},
{
"epoch": 0.42,
"learning_rate": 0.0002941223671013039,
"loss": 2.8853,
"theoretical_loss": 3.542567798014008,
"tokens_seen": 1378549760
},
{
"epoch": 0.42,
"learning_rate": 0.00029411233701103314,
"loss": 2.6624,
"theoretical_loss": 3.542552708032091,
"tokens_seen": 1378615296
},
{
"epoch": 0.42,
"learning_rate": 0.00029410230692076227,
"loss": 2.6581,
"theoretical_loss": 3.5425376189683417,
"tokens_seen": 1378680832
},
{
"epoch": 0.42,
"learning_rate": 0.0002940922768304915,
"loss": 2.8978,
"theoretical_loss": 3.542522530822662,
"tokens_seen": 1378746368
},
{
"epoch": 0.42,
"learning_rate": 0.0002940822467402207,
"loss": 2.964,
"theoretical_loss": 3.5425074435949515,
"tokens_seen": 1378811904
},
{
"epoch": 0.42,
"learning_rate": 0.00029407221664994986,
"loss": 2.6298,
"theoretical_loss": 3.54249235728511,
"tokens_seen": 1378877440
},
{
"epoch": 0.42,
"learning_rate": 0.00029406218655967905,
"loss": 2.772,
"theoretical_loss": 3.54247727189304,
"tokens_seen": 1378942976
},
{
"epoch": 0.42,
"learning_rate": 0.0002940521564694082,
"loss": 2.6111,
"theoretical_loss": 3.5424621874186406,
"tokens_seen": 1379008512
},
{
"epoch": 0.42,
"learning_rate": 0.0002940421263791374,
"loss": 2.8884,
"theoretical_loss": 3.5424471038618126,
"tokens_seen": 1379074048
},
{
"epoch": 0.42,
"learning_rate": 0.00029403209628886664,
"loss": 2.5595,
"theoretical_loss": 3.5424320212224565,
"tokens_seen": 1379139584
},
{
"epoch": 0.42,
"learning_rate": 0.00029402206619859577,
"loss": 2.5775,
"theoretical_loss": 3.5424169395004736,
"tokens_seen": 1379205120
},
{
"epoch": 0.42,
"learning_rate": 0.000294012036108325,
"loss": 2.8067,
"theoretical_loss": 3.542401858695764,
"tokens_seen": 1379270656
},
{
"epoch": 0.42,
"learning_rate": 0.00029400200601805413,
"loss": 2.8204,
"theoretical_loss": 3.542386778808228,
"tokens_seen": 1379336192
},
{
"epoch": 0.42,
"learning_rate": 0.00029399197592778337,
"loss": 2.6327,
"theoretical_loss": 3.5423716998377666,
"tokens_seen": 1379401728
},
{
"epoch": 0.42,
"learning_rate": 0.00029398194583751255,
"loss": 2.6358,
"theoretical_loss": 3.542356621784281,
"tokens_seen": 1379467264
},
{
"epoch": 0.42,
"objective/train/docs_used": 781246,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8434011936187744,
"objective/train/theoretical_loss": 3.5423415446476705,
"objective/train/tokens_used": 1399992800,
"theoretical_loss": 3.5423415446476705,
"tokens_seen": 1379532800
},
{
"epoch": 0.42,
"learning_rate": 0.00029397191574724173,
"loss": 2.6442,
"theoretical_loss": 3.5423415446476705,
"tokens_seen": 1379532800
},
{
"epoch": 0.42,
"learning_rate": 0.0002939618856569709,
"loss": 2.5632,
"theoretical_loss": 3.542326468427837,
"tokens_seen": 1379598336
},
{
"epoch": 0.42,
"learning_rate": 0.00029395185556670015,
"loss": 2.7567,
"theoretical_loss": 3.542311393124681,
"tokens_seen": 1379663872
},
{
"epoch": 0.42,
"learning_rate": 0.0002939418254764293,
"loss": 2.6573,
"theoretical_loss": 3.5422963187381034,
"tokens_seen": 1379729408
},
{
"epoch": 0.42,
"learning_rate": 0.0002939317953861585,
"loss": 2.5441,
"theoretical_loss": 3.542281245268004,
"tokens_seen": 1379794944
},
{
"epoch": 0.42,
"learning_rate": 0.00029392176529588764,
"loss": 2.585,
"theoretical_loss": 3.5422661727142843,
"tokens_seen": 1379860480
},
{
"epoch": 0.42,
"learning_rate": 0.00029391173520561687,
"loss": 2.6571,
"theoretical_loss": 3.542251101076845,
"tokens_seen": 1379926016
},
{
"epoch": 0.42,
"learning_rate": 0.00029390170511534605,
"loss": 2.8359,
"theoretical_loss": 3.542236030355587,
"tokens_seen": 1379991552
},
{
"epoch": 0.42,
"learning_rate": 0.00029389167502507523,
"loss": 2.7616,
"theoretical_loss": 3.5422209605504107,
"tokens_seen": 1380057088
},
{
"epoch": 0.42,
"learning_rate": 0.0002938816449348044,
"loss": 2.843,
"theoretical_loss": 3.542205891661217,
"tokens_seen": 1380122624
},
{
"epoch": 0.42,
"learning_rate": 0.0002938716148445336,
"loss": 2.6628,
"theoretical_loss": 3.542190823687907,
"tokens_seen": 1380188160
},
{
"epoch": 0.42,
"learning_rate": 0.0002938615847542628,
"loss": 2.7361,
"theoretical_loss": 3.5421757566303818,
"tokens_seen": 1380253696
},
{
"epoch": 0.42,
"learning_rate": 0.000293851554663992,
"loss": 2.6609,
"theoretical_loss": 3.542160690488541,
"tokens_seen": 1380319232
},
{
"epoch": 0.42,
"learning_rate": 0.00029384152457372114,
"loss": 2.9,
"theoretical_loss": 3.542145625262287,
"tokens_seen": 1380384768
},
{
"epoch": 0.42,
"learning_rate": 0.0002938314944834504,
"loss": 2.7084,
"theoretical_loss": 3.54213056095152,
"tokens_seen": 1380450304
},
{
"epoch": 0.42,
"learning_rate": 0.0002938214643931795,
"loss": 2.8767,
"theoretical_loss": 3.542115497556141,
"tokens_seen": 1380515840
},
{
"epoch": 0.42,
"learning_rate": 0.00029381143430290874,
"loss": 2.4051,
"theoretical_loss": 3.5421004350760508,
"tokens_seen": 1380581376
},
{
"epoch": 0.42,
"learning_rate": 0.0002938014042126379,
"loss": 2.7797,
"theoretical_loss": 3.54208537351115,
"tokens_seen": 1380646912
},
{
"epoch": 0.42,
"learning_rate": 0.0002937913741223671,
"loss": 2.7532,
"theoretical_loss": 3.54207031286134,
"tokens_seen": 1380712448
},
{
"epoch": 0.42,
"learning_rate": 0.0002937813440320963,
"loss": 2.6683,
"theoretical_loss": 3.542055253126523,
"tokens_seen": 1380777984
},
{
"epoch": 0.42,
"learning_rate": 0.0002937713139418255,
"loss": 2.7589,
"theoretical_loss": 3.5420401943065976,
"tokens_seen": 1380843520
},
{
"epoch": 0.42,
"learning_rate": 0.00029376128385155464,
"loss": 2.7516,
"theoretical_loss": 3.542025136401466,
"tokens_seen": 1380909056
},
{
"epoch": 0.42,
"learning_rate": 0.0002937512537612839,
"loss": 2.5232,
"theoretical_loss": 3.542010079411029,
"tokens_seen": 1380974592
},
{
"epoch": 0.42,
"learning_rate": 0.000293741223671013,
"loss": 2.8305,
"theoretical_loss": 3.5419950233351885,
"tokens_seen": 1381040128
},
{
"epoch": 0.42,
"learning_rate": 0.00029373119358074224,
"loss": 2.6765,
"theoretical_loss": 3.5419799681738446,
"tokens_seen": 1381105664
},
{
"epoch": 0.42,
"objective/train/docs_used": 781914,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.805891752243042,
"objective/train/theoretical_loss": 3.5419649139268987,
"objective/train/tokens_used": 1401631200,
"theoretical_loss": 3.5419649139268987,
"tokens_seen": 1381171200
},
{
"epoch": 0.42,
"learning_rate": 0.0002937211634904714,
"loss": 2.5976,
"theoretical_loss": 3.5419649139268987,
"tokens_seen": 1381171200
},
{
"epoch": 0.42,
"learning_rate": 0.0002937111334002006,
"loss": 2.6595,
"theoretical_loss": 3.541949860594251,
"tokens_seen": 1381236736
},
{
"epoch": 0.42,
"learning_rate": 0.0002937011033099298,
"loss": 2.653,
"theoretical_loss": 3.541934808175804,
"tokens_seen": 1381302272
},
{
"epoch": 0.42,
"learning_rate": 0.00029369107321965896,
"loss": 2.5928,
"theoretical_loss": 3.541919756671458,
"tokens_seen": 1381367808
},
{
"epoch": 0.42,
"learning_rate": 0.00029368104312938815,
"loss": 2.5234,
"theoretical_loss": 3.541904706081114,
"tokens_seen": 1381433344
},
{
"epoch": 0.42,
"learning_rate": 0.0002936710130391174,
"loss": 2.6846,
"theoretical_loss": 3.541889656404674,
"tokens_seen": 1381498880
},
{
"epoch": 0.42,
"learning_rate": 0.0002936609829488465,
"loss": 2.7394,
"theoretical_loss": 3.541874607642039,
"tokens_seen": 1381564416
},
{
"epoch": 0.42,
"learning_rate": 0.00029365095285857574,
"loss": 2.8839,
"theoretical_loss": 3.541859559793109,
"tokens_seen": 1381629952
},
{
"epoch": 0.42,
"learning_rate": 0.00029364092276830487,
"loss": 2.8632,
"theoretical_loss": 3.5418445128577862,
"tokens_seen": 1381695488
},
{
"epoch": 0.42,
"learning_rate": 0.0002936308926780341,
"loss": 2.6887,
"theoretical_loss": 3.5418294668359716,
"tokens_seen": 1381761024
},
{
"epoch": 0.42,
"learning_rate": 0.0002936208625877633,
"loss": 2.8732,
"theoretical_loss": 3.541814421727566,
"tokens_seen": 1381826560
},
{
"epoch": 0.42,
"learning_rate": 0.00029361083249749247,
"loss": 2.7271,
"theoretical_loss": 3.5417993775324716,
"tokens_seen": 1381892096
},
{
"epoch": 0.42,
"learning_rate": 0.00029360080240722165,
"loss": 3.0045,
"theoretical_loss": 3.541784334250589,
"tokens_seen": 1381957632
},
{
"epoch": 0.42,
"learning_rate": 0.0002935907723169509,
"loss": 2.7398,
"theoretical_loss": 3.5417692918818187,
"tokens_seen": 1382023168
},
{
"epoch": 0.42,
"learning_rate": 0.00029358074222668,
"loss": 2.5375,
"theoretical_loss": 3.541754250426064,
"tokens_seen": 1382088704
},
{
"epoch": 0.42,
"learning_rate": 0.00029357071213640925,
"loss": 2.7706,
"theoretical_loss": 3.5417392098832243,
"tokens_seen": 1382154240
},
{
"epoch": 0.42,
"learning_rate": 0.0002935606820461384,
"loss": 2.7137,
"theoretical_loss": 3.541724170253201,
"tokens_seen": 1382219776
},
{
"epoch": 0.42,
"learning_rate": 0.0002935506519558676,
"loss": 2.8943,
"theoretical_loss": 3.5417091315358973,
"tokens_seen": 1382285312
},
{
"epoch": 0.42,
"learning_rate": 0.00029354062186559684,
"loss": 2.4373,
"theoretical_loss": 3.5416940937312127,
"tokens_seen": 1382350848
},
{
"epoch": 0.42,
"learning_rate": 0.00029353059177532597,
"loss": 2.5485,
"theoretical_loss": 3.5416790568390493,
"tokens_seen": 1382416384
},
{
"epoch": 0.42,
"learning_rate": 0.0002935205616850552,
"loss": 2.5533,
"theoretical_loss": 3.541664020859308,
"tokens_seen": 1382481920
},
{
"epoch": 0.42,
"learning_rate": 0.00029351053159478433,
"loss": 2.5909,
"theoretical_loss": 3.5416489857918902,
"tokens_seen": 1382547456
},
{
"epoch": 0.42,
"learning_rate": 0.00029350050150451357,
"loss": 2.6454,
"theoretical_loss": 3.5416339516366984,
"tokens_seen": 1382612992
},
{
"epoch": 0.42,
"learning_rate": 0.00029349047141424275,
"loss": 2.7256,
"theoretical_loss": 3.541618918393633,
"tokens_seen": 1382678528
},
{
"epoch": 0.42,
"learning_rate": 0.00029348044132397193,
"loss": 2.497,
"theoretical_loss": 3.541603886062595,
"tokens_seen": 1382744064
},
{
"epoch": 0.42,
"objective/train/docs_used": 783265,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7731690406799316,
"objective/train/theoretical_loss": 3.541588854643487,
"objective/train/tokens_used": 1403269600,
"theoretical_loss": 3.541588854643487,
"tokens_seen": 1382809600
},
{
"epoch": 0.42,
"learning_rate": 0.0002934704112337011,
"loss": 2.7865,
"theoretical_loss": 3.541588854643487,
"tokens_seen": 1382809600
},
{
"epoch": 0.42,
"learning_rate": 0.00029346038114343035,
"loss": 2.8549,
"theoretical_loss": 3.54157382413621,
"tokens_seen": 1382875136
},
{
"epoch": 0.42,
"learning_rate": 0.0002934503510531595,
"loss": 2.8729,
"theoretical_loss": 3.541558794540665,
"tokens_seen": 1382940672
},
{
"epoch": 0.42,
"learning_rate": 0.0002934403209628887,
"loss": 2.5464,
"theoretical_loss": 3.5415437658567543,
"tokens_seen": 1383006208
},
{
"epoch": 0.42,
"learning_rate": 0.00029343029087261784,
"loss": 2.8245,
"theoretical_loss": 3.541528738084379,
"tokens_seen": 1383071744
},
{
"epoch": 0.42,
"learning_rate": 0.00029342026078234707,
"loss": 2.6804,
"theoretical_loss": 3.5415137112234403,
"tokens_seen": 1383137280
},
{
"epoch": 0.42,
"learning_rate": 0.00029341023069207625,
"loss": 2.787,
"theoretical_loss": 3.54149868527384,
"tokens_seen": 1383202816
},
{
"epoch": 0.42,
"learning_rate": 0.00029340020060180543,
"loss": 2.7117,
"theoretical_loss": 3.541483660235481,
"tokens_seen": 1383268352
},
{
"epoch": 0.42,
"learning_rate": 0.0002933901705115346,
"loss": 2.5986,
"theoretical_loss": 3.541468636108262,
"tokens_seen": 1383333888
},
{
"epoch": 0.42,
"learning_rate": 0.0002933801404212638,
"loss": 2.8469,
"theoretical_loss": 3.541453612892087,
"tokens_seen": 1383399424
},
{
"epoch": 0.42,
"learning_rate": 0.000293370110330993,
"loss": 2.8191,
"theoretical_loss": 3.541438590586857,
"tokens_seen": 1383464960
},
{
"epoch": 0.42,
"learning_rate": 0.0002933600802407222,
"loss": 2.6033,
"theoretical_loss": 3.5414235691924727,
"tokens_seen": 1383530496
},
{
"epoch": 0.42,
"learning_rate": 0.00029335005015045134,
"loss": 2.669,
"theoretical_loss": 3.541408548708837,
"tokens_seen": 1383596032
},
{
"epoch": 0.42,
"learning_rate": 0.0002933400200601806,
"loss": 2.5886,
"theoretical_loss": 3.541393529135851,
"tokens_seen": 1383661568
},
{
"epoch": 0.42,
"learning_rate": 0.0002933299899699097,
"loss": 2.841,
"theoretical_loss": 3.5413785104734163,
"tokens_seen": 1383727104
},
{
"epoch": 0.42,
"learning_rate": 0.00029331995987963894,
"loss": 2.7297,
"theoretical_loss": 3.5413634927214344,
"tokens_seen": 1383792640
},
{
"epoch": 0.42,
"learning_rate": 0.0002933099297893681,
"loss": 2.6088,
"theoretical_loss": 3.5413484758798073,
"tokens_seen": 1383858176
},
{
"epoch": 0.42,
"learning_rate": 0.0002932998996990973,
"loss": 2.6383,
"theoretical_loss": 3.5413334599484365,
"tokens_seen": 1383923712
},
{
"epoch": 0.42,
"learning_rate": 0.0002932898696088265,
"loss": 2.7198,
"theoretical_loss": 3.5413184449272244,
"tokens_seen": 1383989248
},
{
"epoch": 0.42,
"learning_rate": 0.0002932798395185557,
"loss": 2.6039,
"theoretical_loss": 3.541303430816072,
"tokens_seen": 1384054784
},
{
"epoch": 0.42,
"learning_rate": 0.00029326980942828484,
"loss": 2.7195,
"theoretical_loss": 3.5412884176148807,
"tokens_seen": 1384120320
},
{
"epoch": 0.42,
"learning_rate": 0.0002932597793380141,
"loss": 2.6283,
"theoretical_loss": 3.541273405323553,
"tokens_seen": 1384185856
},
{
"epoch": 0.42,
"learning_rate": 0.0002932497492477432,
"loss": 2.7962,
"theoretical_loss": 3.5412583939419906,
"tokens_seen": 1384251392
},
{
"epoch": 0.42,
"learning_rate": 0.00029323971915747244,
"loss": 2.8197,
"theoretical_loss": 3.5412433834700954,
"tokens_seen": 1384316928
},
{
"epoch": 0.42,
"learning_rate": 0.0002932296890672016,
"loss": 2.828,
"theoretical_loss": 3.541228373907769,
"tokens_seen": 1384382464
},
{
"epoch": 0.42,
"objective/train/docs_used": 783958,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3266313076019287,
"objective/train/theoretical_loss": 3.541213365254913,
"objective/train/tokens_used": 1404908000,
"theoretical_loss": 3.541213365254913,
"tokens_seen": 1384448000
},
{
"epoch": 0.42,
"learning_rate": 0.0002932196589769308,
"loss": 2.6729,
"theoretical_loss": 3.541213365254913,
"tokens_seen": 1384448000
},
{
"epoch": 0.42,
"learning_rate": 0.00029320962888666,
"loss": 2.7296,
"theoretical_loss": 3.5411983575114294,
"tokens_seen": 1384513536
},
{
"epoch": 0.42,
"learning_rate": 0.00029319959879638916,
"loss": 2.4912,
"theoretical_loss": 3.5411833506772203,
"tokens_seen": 1384579072
},
{
"epoch": 0.42,
"learning_rate": 0.00029318956870611835,
"loss": 2.5321,
"theoretical_loss": 3.5411683447521876,
"tokens_seen": 1384644608
},
{
"epoch": 0.42,
"learning_rate": 0.0002931795386158476,
"loss": 2.6283,
"theoretical_loss": 3.5411533397362325,
"tokens_seen": 1384710144
},
{
"epoch": 0.42,
"learning_rate": 0.0002931695085255767,
"loss": 2.811,
"theoretical_loss": 3.541138335629258,
"tokens_seen": 1384775680
},
{
"epoch": 0.42,
"learning_rate": 0.00029315947843530594,
"loss": 2.7543,
"theoretical_loss": 3.541123332431165,
"tokens_seen": 1384841216
},
{
"epoch": 0.42,
"learning_rate": 0.00029314944834503507,
"loss": 2.4994,
"theoretical_loss": 3.541108330141856,
"tokens_seen": 1384906752
},
{
"epoch": 0.42,
"learning_rate": 0.0002931394182547643,
"loss": 2.6133,
"theoretical_loss": 3.5410933287612325,
"tokens_seen": 1384972288
},
{
"epoch": 0.42,
"learning_rate": 0.0002931293881644935,
"loss": 2.8683,
"theoretical_loss": 3.5410783282891973,
"tokens_seen": 1385037824
},
{
"epoch": 0.42,
"learning_rate": 0.00029311935807422267,
"loss": 2.8316,
"theoretical_loss": 3.541063328725652,
"tokens_seen": 1385103360
},
{
"epoch": 0.42,
"learning_rate": 0.00029310932798395185,
"loss": 2.7482,
"theoretical_loss": 3.541048330070498,
"tokens_seen": 1385168896
},
{
"epoch": 0.42,
"learning_rate": 0.0002930992978936811,
"loss": 2.7096,
"theoretical_loss": 3.541033332323638,
"tokens_seen": 1385234432
},
{
"epoch": 0.42,
"learning_rate": 0.0002930892678034102,
"loss": 2.548,
"theoretical_loss": 3.541018335484974,
"tokens_seen": 1385299968
},
{
"epoch": 0.42,
"learning_rate": 0.00029307923771313945,
"loss": 2.6794,
"theoretical_loss": 3.5410033395544076,
"tokens_seen": 1385365504
},
{
"epoch": 0.42,
"learning_rate": 0.0002930692076228686,
"loss": 2.6128,
"theoretical_loss": 3.540988344531841,
"tokens_seen": 1385431040
},
{
"epoch": 0.42,
"learning_rate": 0.0002930591775325978,
"loss": 2.6832,
"theoretical_loss": 3.5409733504171763,
"tokens_seen": 1385496576
},
{
"epoch": 0.42,
"learning_rate": 0.000293049147442327,
"loss": 2.7075,
"theoretical_loss": 3.540958357210316,
"tokens_seen": 1385562112
},
{
"epoch": 0.42,
"learning_rate": 0.00029303911735205617,
"loss": 2.8775,
"theoretical_loss": 3.5409433649111617,
"tokens_seen": 1385627648
},
{
"epoch": 0.42,
"learning_rate": 0.00029302908726178535,
"loss": 2.5821,
"theoretical_loss": 3.5409283735196153,
"tokens_seen": 1385693184
},
{
"epoch": 0.42,
"learning_rate": 0.00029301905717151453,
"loss": 2.8909,
"theoretical_loss": 3.54091338303558,
"tokens_seen": 1385758720
},
{
"epoch": 0.42,
"learning_rate": 0.0002930090270812437,
"loss": 2.6007,
"theoretical_loss": 3.5408983934589573,
"tokens_seen": 1385824256
},
{
"epoch": 0.42,
"learning_rate": 0.00029299899699097295,
"loss": 2.7121,
"theoretical_loss": 3.540883404789649,
"tokens_seen": 1385889792
},
{
"epoch": 0.42,
"learning_rate": 0.0002929889669007021,
"loss": 2.7258,
"theoretical_loss": 3.5408684170275575,
"tokens_seen": 1385955328
},
{
"epoch": 0.42,
"learning_rate": 0.0002929789368104313,
"loss": 2.7313,
"theoretical_loss": 3.5408534301725854,
"tokens_seen": 1386020864
},
{
"epoch": 0.42,
"objective/train/docs_used": 785103,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8746604919433594,
"objective/train/theoretical_loss": 3.5408384442246343,
"objective/train/tokens_used": 1406546400,
"theoretical_loss": 3.5408384442246343,
"tokens_seen": 1386086400
},
{
"epoch": 0.42,
"learning_rate": 0.0002929689067201605,
"loss": 2.7128,
"theoretical_loss": 3.5408384442246343,
"tokens_seen": 1386086400
},
{
"epoch": 0.42,
"learning_rate": 0.0002929588766298897,
"loss": 2.7031,
"theoretical_loss": 3.540823459183607,
"tokens_seen": 1386151936
},
{
"epoch": 0.42,
"learning_rate": 0.00029294884653961885,
"loss": 2.7255,
"theoretical_loss": 3.5408084750494053,
"tokens_seen": 1386217472
},
{
"epoch": 0.42,
"learning_rate": 0.00029293881644934804,
"loss": 2.5921,
"theoretical_loss": 3.540793491821932,
"tokens_seen": 1386283008
},
{
"epoch": 0.42,
"learning_rate": 0.0002929287863590772,
"loss": 2.541,
"theoretical_loss": 3.540778509501089,
"tokens_seen": 1386348544
},
{
"epoch": 0.42,
"learning_rate": 0.00029291875626880645,
"loss": 2.6487,
"theoretical_loss": 3.540763528086778,
"tokens_seen": 1386414080
},
{
"epoch": 0.42,
"learning_rate": 0.0002929087261785356,
"loss": 2.7102,
"theoretical_loss": 3.540748547578902,
"tokens_seen": 1386479616
},
{
"epoch": 0.42,
"learning_rate": 0.0002928986960882648,
"loss": 2.5887,
"theoretical_loss": 3.540733567977364,
"tokens_seen": 1386545152
},
{
"epoch": 0.42,
"learning_rate": 0.00029288866599799394,
"loss": 2.6243,
"theoretical_loss": 3.540718589282065,
"tokens_seen": 1386610688
},
{
"epoch": 0.42,
"learning_rate": 0.0002928786359077232,
"loss": 2.5606,
"theoretical_loss": 3.5407036114929076,
"tokens_seen": 1386676224
},
{
"epoch": 0.42,
"learning_rate": 0.00029286860581745236,
"loss": 2.7082,
"theoretical_loss": 3.540688634609795,
"tokens_seen": 1386741760
},
{
"epoch": 0.42,
"learning_rate": 0.00029285857572718154,
"loss": 2.6291,
"theoretical_loss": 3.5406736586326284,
"tokens_seen": 1386807296
},
{
"epoch": 0.42,
"learning_rate": 0.0002928485456369107,
"loss": 2.6003,
"theoretical_loss": 3.540658683561311,
"tokens_seen": 1386872832
},
{
"epoch": 0.42,
"learning_rate": 0.0002928385155466399,
"loss": 2.5735,
"theoretical_loss": 3.5406437093957455,
"tokens_seen": 1386938368
},
{
"epoch": 0.42,
"learning_rate": 0.0002928284854563691,
"loss": 2.7838,
"theoretical_loss": 3.540628736135834,
"tokens_seen": 1387003904
},
{
"epoch": 0.42,
"learning_rate": 0.0002928184553660983,
"loss": 2.6801,
"theoretical_loss": 3.540613763781478,
"tokens_seen": 1387069440
},
{
"epoch": 0.42,
"learning_rate": 0.00029280842527582744,
"loss": 2.6179,
"theoretical_loss": 3.540598792332581,
"tokens_seen": 1387134976
},
{
"epoch": 0.42,
"learning_rate": 0.0002927983951855567,
"loss": 2.5718,
"theoretical_loss": 3.5405838217890455,
"tokens_seen": 1387200512
},
{
"epoch": 0.42,
"learning_rate": 0.0002927883650952859,
"loss": 2.7263,
"theoretical_loss": 3.5405688521507734,
"tokens_seen": 1387266048
},
{
"epoch": 0.42,
"learning_rate": 0.00029277833500501504,
"loss": 2.4144,
"theoretical_loss": 3.5405538834176675,
"tokens_seen": 1387331584
},
{
"epoch": 0.42,
"learning_rate": 0.0002927683049147443,
"loss": 2.6998,
"theoretical_loss": 3.5405389155896305,
"tokens_seen": 1387397120
},
{
"epoch": 0.42,
"learning_rate": 0.0002927582748244734,
"loss": 2.7211,
"theoretical_loss": 3.5405239486665643,
"tokens_seen": 1387462656
},
{
"epoch": 0.42,
"learning_rate": 0.00029274824473420264,
"loss": 2.632,
"theoretical_loss": 3.540508982648372,
"tokens_seen": 1387528192
},
{
"epoch": 0.42,
"learning_rate": 0.0002927382146439318,
"loss": 2.5567,
"theoretical_loss": 3.5404940175349564,
"tokens_seen": 1387593728
},
{
"epoch": 0.42,
"learning_rate": 0.000292728184553661,
"loss": 2.8375,
"theoretical_loss": 3.5404790533262194,
"tokens_seen": 1387659264
},
{
"epoch": 0.42,
"objective/train/docs_used": 785794,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8001210689544678,
"objective/train/theoretical_loss": 3.5404640900220636,
"objective/train/tokens_used": 1408184800,
"theoretical_loss": 3.5404640900220636,
"tokens_seen": 1387724800
},
{
"epoch": 0.42,
"learning_rate": 0.0002927181544633902,
"loss": 2.5916,
"theoretical_loss": 3.5404640900220636,
"tokens_seen": 1387724800
},
{
"epoch": 0.42,
"learning_rate": 0.00029270812437311936,
"loss": 2.7585,
"theoretical_loss": 3.5404491276223924,
"tokens_seen": 1387790336
},
{
"epoch": 0.42,
"learning_rate": 0.00029269809428284855,
"loss": 2.4688,
"theoretical_loss": 3.540434166127108,
"tokens_seen": 1387855872
},
{
"epoch": 0.42,
"learning_rate": 0.0002926880641925778,
"loss": 2.7603,
"theoretical_loss": 3.5404192055361126,
"tokens_seen": 1387921408
},
{
"epoch": 0.42,
"learning_rate": 0.0002926780341023069,
"loss": 3.0173,
"theoretical_loss": 3.5404042458493086,
"tokens_seen": 1387986944
},
{
"epoch": 0.42,
"learning_rate": 0.00029266800401203614,
"loss": 2.5282,
"theoretical_loss": 3.5403892870666,
"tokens_seen": 1388052480
},
{
"epoch": 0.42,
"learning_rate": 0.00029265797392176527,
"loss": 2.8172,
"theoretical_loss": 3.5403743291878884,
"tokens_seen": 1388118016
},
{
"epoch": 0.42,
"learning_rate": 0.0002926479438314945,
"loss": 2.733,
"theoretical_loss": 3.5403593722130773,
"tokens_seen": 1388183552
},
{
"epoch": 0.42,
"learning_rate": 0.0002926379137412237,
"loss": 2.5028,
"theoretical_loss": 3.5403444161420685,
"tokens_seen": 1388249088
},
{
"epoch": 0.42,
"learning_rate": 0.00029262788365095287,
"loss": 2.8673,
"theoretical_loss": 3.5403294609747653,
"tokens_seen": 1388314624
},
{
"epoch": 0.42,
"learning_rate": 0.00029261785356068205,
"loss": 2.8456,
"theoretical_loss": 3.54031450671107,
"tokens_seen": 1388380160
},
{
"epoch": 0.42,
"learning_rate": 0.0002926078234704113,
"loss": 2.705,
"theoretical_loss": 3.540299553350886,
"tokens_seen": 1388445696
},
{
"epoch": 0.42,
"learning_rate": 0.0002925977933801404,
"loss": 2.67,
"theoretical_loss": 3.5402846008941156,
"tokens_seen": 1388511232
},
{
"epoch": 0.42,
"learning_rate": 0.00029258776328986965,
"loss": 2.7818,
"theoretical_loss": 3.5402696493406616,
"tokens_seen": 1388576768
},
{
"epoch": 0.42,
"learning_rate": 0.0002925777331995988,
"loss": 2.4424,
"theoretical_loss": 3.5402546986904273,
"tokens_seen": 1388642304
},
{
"epoch": 0.42,
"learning_rate": 0.000292567703109328,
"loss": 2.7327,
"theoretical_loss": 3.5402397489433146,
"tokens_seen": 1388707840
},
{
"epoch": 0.42,
"learning_rate": 0.0002925576730190572,
"loss": 3.0758,
"theoretical_loss": 3.540224800099227,
"tokens_seen": 1388773376
},
{
"epoch": 0.42,
"learning_rate": 0.00029254764292878637,
"loss": 2.8243,
"theoretical_loss": 3.5402098521580676,
"tokens_seen": 1388838912
},
{
"epoch": 0.42,
"learning_rate": 0.00029253761283851555,
"loss": 2.663,
"theoretical_loss": 3.540194905119738,
"tokens_seen": 1388904448
},
{
"epoch": 0.42,
"learning_rate": 0.00029252758274824473,
"loss": 2.5451,
"theoretical_loss": 3.5401799589841425,
"tokens_seen": 1388969984
},
{
"epoch": 0.42,
"learning_rate": 0.0002925175526579739,
"loss": 2.9159,
"theoretical_loss": 3.5401650137511833,
"tokens_seen": 1389035520
},
{
"epoch": 0.42,
"learning_rate": 0.00029250752256770315,
"loss": 2.8413,
"theoretical_loss": 3.5401500694207635,
"tokens_seen": 1389101056
},
{
"epoch": 0.42,
"learning_rate": 0.0002924974924774323,
"loss": 2.5923,
"theoretical_loss": 3.5401351259927853,
"tokens_seen": 1389166592
},
{
"epoch": 0.42,
"learning_rate": 0.0002924874623871615,
"loss": 2.501,
"theoretical_loss": 3.540120183467153,
"tokens_seen": 1389232128
},
{
"epoch": 0.42,
"learning_rate": 0.0002924774322968907,
"loss": 2.6568,
"theoretical_loss": 3.5401052418437686,
"tokens_seen": 1389297664
},
{
"epoch": 0.42,
"objective/train/docs_used": 786491,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4350569248199463,
"objective/train/theoretical_loss": 3.540090301122535,
"objective/train/tokens_used": 1409823200,
"theoretical_loss": 3.540090301122535,
"tokens_seen": 1389363200
},
{
"epoch": 0.42,
"learning_rate": 0.0002924674022066199,
"loss": 2.6373,
"theoretical_loss": 3.540090301122535,
"tokens_seen": 1389363200
},
{
"epoch": 0.42,
"learning_rate": 0.00029245737211634906,
"loss": 2.8875,
"theoretical_loss": 3.540075361303356,
"tokens_seen": 1389428736
},
{
"epoch": 0.42,
"learning_rate": 0.00029244734202607824,
"loss": 2.5514,
"theoretical_loss": 3.5400604223861336,
"tokens_seen": 1389494272
},
{
"epoch": 0.42,
"learning_rate": 0.0002924373119358074,
"loss": 2.6498,
"theoretical_loss": 3.540045484370771,
"tokens_seen": 1389559808
},
{
"epoch": 0.42,
"learning_rate": 0.00029242728184553665,
"loss": 2.9868,
"theoretical_loss": 3.5400305472571723,
"tokens_seen": 1389625344
},
{
"epoch": 0.42,
"learning_rate": 0.0002924172517552658,
"loss": 2.6669,
"theoretical_loss": 3.540015611045239,
"tokens_seen": 1389690880
},
{
"epoch": 0.42,
"learning_rate": 0.000292407221664995,
"loss": 2.5821,
"theoretical_loss": 3.540000675734875,
"tokens_seen": 1389756416
},
{
"epoch": 0.42,
"learning_rate": 0.00029239719157472414,
"loss": 2.6612,
"theoretical_loss": 3.5399857413259834,
"tokens_seen": 1389821952
},
{
"epoch": 0.42,
"learning_rate": 0.0002923871614844534,
"loss": 2.6244,
"theoretical_loss": 3.5399708078184675,
"tokens_seen": 1389887488
},
{
"epoch": 0.42,
"learning_rate": 0.00029237713139418256,
"loss": 2.6683,
"theoretical_loss": 3.5399558752122293,
"tokens_seen": 1389953024
},
{
"epoch": 0.42,
"learning_rate": 0.00029236710130391174,
"loss": 2.3443,
"theoretical_loss": 3.5399409435071734,
"tokens_seen": 1390018560
},
{
"epoch": 0.42,
"learning_rate": 0.0002923570712136409,
"loss": 2.5046,
"theoretical_loss": 3.5399260127032015,
"tokens_seen": 1390084096
},
{
"epoch": 0.42,
"learning_rate": 0.0002923470411233701,
"loss": 2.8247,
"theoretical_loss": 3.5399110828002174,
"tokens_seen": 1390149632
},
{
"epoch": 0.42,
"learning_rate": 0.0002923370110330993,
"loss": 2.6215,
"theoretical_loss": 3.5398961537981246,
"tokens_seen": 1390215168
},
{
"epoch": 0.42,
"learning_rate": 0.0002923269809428285,
"loss": 2.684,
"theoretical_loss": 3.539881225696826,
"tokens_seen": 1390280704
},
{
"epoch": 0.42,
"learning_rate": 0.00029231695085255765,
"loss": 2.733,
"theoretical_loss": 3.539866298496224,
"tokens_seen": 1390346240
},
{
"epoch": 0.42,
"learning_rate": 0.0002923069207622869,
"loss": 2.5882,
"theoretical_loss": 3.5398513721962237,
"tokens_seen": 1390411776
},
{
"epoch": 0.42,
"learning_rate": 0.00029229689067201606,
"loss": 2.6572,
"theoretical_loss": 3.5398364467967265,
"tokens_seen": 1390477312
},
{
"epoch": 0.42,
"learning_rate": 0.00029228686058174524,
"loss": 2.7224,
"theoretical_loss": 3.539821522297636,
"tokens_seen": 1390542848
},
{
"epoch": 0.42,
"learning_rate": 0.0002922768304914744,
"loss": 2.6217,
"theoretical_loss": 3.5398065986988563,
"tokens_seen": 1390608384
},
{
"epoch": 0.42,
"learning_rate": 0.0002922668004012036,
"loss": 2.8851,
"theoretical_loss": 3.53979167600029,
"tokens_seen": 1390673920
},
{
"epoch": 0.42,
"learning_rate": 0.0002922567703109328,
"loss": 2.8297,
"theoretical_loss": 3.5397767542018403,
"tokens_seen": 1390739456
},
{
"epoch": 0.42,
"learning_rate": 0.000292246740220662,
"loss": 2.5178,
"theoretical_loss": 3.5397618333034107,
"tokens_seen": 1390804992
},
{
"epoch": 0.42,
"learning_rate": 0.00029223671013039115,
"loss": 2.6979,
"theoretical_loss": 3.5397469133049047,
"tokens_seen": 1390870528
},
{
"epoch": 0.42,
"learning_rate": 0.0002922266800401204,
"loss": 2.6731,
"theoretical_loss": 3.5397319942062255,
"tokens_seen": 1390936064
},
{
"epoch": 0.42,
"objective/train/docs_used": 787676,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.47843337059021,
"objective/train/theoretical_loss": 3.539717076007276,
"objective/train/tokens_used": 1411461600,
"theoretical_loss": 3.539717076007276,
"tokens_seen": 1391001600
},
{
"epoch": 0.42,
"learning_rate": 0.0002922166499498495,
"loss": 2.5407,
"theoretical_loss": 3.539717076007276,
"tokens_seen": 1391001600
},
{
"epoch": 0.42,
"learning_rate": 0.00029220661985957875,
"loss": 2.513,
"theoretical_loss": 3.53970215870796,
"tokens_seen": 1391067136
},
{
"epoch": 0.42,
"learning_rate": 0.0002921965897693079,
"loss": 2.564,
"theoretical_loss": 3.539687242308181,
"tokens_seen": 1391132672
},
{
"epoch": 0.42,
"learning_rate": 0.0002921865596790371,
"loss": 2.738,
"theoretical_loss": 3.5396723268078416,
"tokens_seen": 1391198208
},
{
"epoch": 0.42,
"learning_rate": 0.0002921765295887663,
"loss": 2.6498,
"theoretical_loss": 3.539657412206846,
"tokens_seen": 1391263744
},
{
"epoch": 0.42,
"learning_rate": 0.00029216649949849547,
"loss": 2.6889,
"theoretical_loss": 3.5396424985050974,
"tokens_seen": 1391329280
},
{
"epoch": 0.42,
"learning_rate": 0.00029215646940822465,
"loss": 2.4986,
"theoretical_loss": 3.5396275857024992,
"tokens_seen": 1391394816
},
{
"epoch": 0.42,
"learning_rate": 0.0002921464393179539,
"loss": 2.5543,
"theoretical_loss": 3.5396126737989553,
"tokens_seen": 1391460352
},
{
"epoch": 0.42,
"learning_rate": 0.000292136409227683,
"loss": 2.5421,
"theoretical_loss": 3.5395977627943678,
"tokens_seen": 1391525888
},
{
"epoch": 0.42,
"learning_rate": 0.00029212637913741225,
"loss": 2.9765,
"theoretical_loss": 3.5395828526886417,
"tokens_seen": 1391591424
},
{
"epoch": 0.42,
"learning_rate": 0.00029211634904714143,
"loss": 2.6622,
"theoretical_loss": 3.5395679434816794,
"tokens_seen": 1391656960
},
{
"epoch": 0.42,
"learning_rate": 0.0002921063189568706,
"loss": 2.5637,
"theoretical_loss": 3.5395530351733853,
"tokens_seen": 1391722496
},
{
"epoch": 0.42,
"learning_rate": 0.0002920962888665998,
"loss": 2.9396,
"theoretical_loss": 3.539538127763662,
"tokens_seen": 1391788032
},
{
"epoch": 0.42,
"learning_rate": 0.000292086258776329,
"loss": 2.7754,
"theoretical_loss": 3.539523221252414,
"tokens_seen": 1391853568
},
{
"epoch": 0.42,
"learning_rate": 0.00029207622868605815,
"loss": 2.7768,
"theoretical_loss": 3.5395083156395444,
"tokens_seen": 1391919104
},
{
"epoch": 0.42,
"learning_rate": 0.0002920661985957874,
"loss": 2.8789,
"theoretical_loss": 3.5394934109249565,
"tokens_seen": 1391984640
},
{
"epoch": 0.42,
"learning_rate": 0.0002920561685055165,
"loss": 2.7123,
"theoretical_loss": 3.5394785071085537,
"tokens_seen": 1392050176
},
{
"epoch": 0.42,
"learning_rate": 0.00029204613841524575,
"loss": 2.8609,
"theoretical_loss": 3.5394636041902405,
"tokens_seen": 1392115712
},
{
"epoch": 0.42,
"learning_rate": 0.00029203610832497493,
"loss": 2.712,
"theoretical_loss": 3.5394487021699197,
"tokens_seen": 1392181248
},
{
"epoch": 0.42,
"learning_rate": 0.0002920260782347041,
"loss": 2.6514,
"theoretical_loss": 3.5394338010474957,
"tokens_seen": 1392246784
},
{
"epoch": 0.42,
"learning_rate": 0.00029201604814443335,
"loss": 2.8803,
"theoretical_loss": 3.539418900822871,
"tokens_seen": 1392312320
},
{
"epoch": 0.42,
"learning_rate": 0.0002920060180541625,
"loss": 2.8045,
"theoretical_loss": 3.5394040014959502,
"tokens_seen": 1392377856
},
{
"epoch": 0.42,
"learning_rate": 0.0002919959879638917,
"loss": 2.7945,
"theoretical_loss": 3.539389103066637,
"tokens_seen": 1392443392
},
{
"epoch": 0.42,
"learning_rate": 0.0002919859578736209,
"loss": 2.6441,
"theoretical_loss": 3.539374205534834,
"tokens_seen": 1392508928
},
{
"epoch": 0.42,
"learning_rate": 0.0002919759277833501,
"loss": 2.5839,
"theoretical_loss": 3.539359308900446,
"tokens_seen": 1392574464
},
{
"epoch": 0.42,
"objective/train/docs_used": 788318,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.430708169937134,
"objective/train/theoretical_loss": 3.5393444131633762,
"objective/train/tokens_used": 1413100000,
"theoretical_loss": 3.5393444131633762,
"tokens_seen": 1392640000
},
{
"epoch": 0.42,
"learning_rate": 0.00029196589769307926,
"loss": 2.6532,
"theoretical_loss": 3.5393444131633762,
"tokens_seen": 1392640000
},
{
"epoch": 0.42,
"learning_rate": 0.00029195586760280844,
"loss": 2.7376,
"theoretical_loss": 3.5393295183235294,
"tokens_seen": 1392705536
},
{
"epoch": 0.42,
"learning_rate": 0.0002919458375125376,
"loss": 2.7998,
"theoretical_loss": 3.5393146243808076,
"tokens_seen": 1392771072
},
{
"epoch": 0.42,
"learning_rate": 0.00029193580742226685,
"loss": 2.7025,
"theoretical_loss": 3.539299731335115,
"tokens_seen": 1392836608
},
{
"epoch": 0.42,
"learning_rate": 0.000291925777331996,
"loss": 2.5487,
"theoretical_loss": 3.5392848391863563,
"tokens_seen": 1392902144
},
{
"epoch": 0.42,
"learning_rate": 0.0002919157472417252,
"loss": 2.4653,
"theoretical_loss": 3.539269947934435,
"tokens_seen": 1392967680
},
{
"epoch": 0.42,
"learning_rate": 0.00029190571715145434,
"loss": 2.712,
"theoretical_loss": 3.539255057579254,
"tokens_seen": 1393033216
},
{
"epoch": 0.42,
"learning_rate": 0.0002918956870611836,
"loss": 2.5707,
"theoretical_loss": 3.539240168120718,
"tokens_seen": 1393098752
},
{
"epoch": 0.42,
"learning_rate": 0.00029188565697091276,
"loss": 2.8042,
"theoretical_loss": 3.539225279558731,
"tokens_seen": 1393164288
},
{
"epoch": 0.42,
"learning_rate": 0.00029187562688064194,
"loss": 2.6817,
"theoretical_loss": 3.539210391893196,
"tokens_seen": 1393229824
},
{
"epoch": 0.42,
"learning_rate": 0.0002918655967903711,
"loss": 2.7786,
"theoretical_loss": 3.539195505124017,
"tokens_seen": 1393295360
},
{
"epoch": 0.42,
"learning_rate": 0.0002918555667001003,
"loss": 2.6601,
"theoretical_loss": 3.5391806192510984,
"tokens_seen": 1393360896
},
{
"epoch": 0.42,
"learning_rate": 0.0002918455366098295,
"loss": 2.507,
"theoretical_loss": 3.539165734274344,
"tokens_seen": 1393426432
},
{
"epoch": 0.42,
"learning_rate": 0.0002918355065195587,
"loss": 2.4369,
"theoretical_loss": 3.539150850193658,
"tokens_seen": 1393491968
},
{
"epoch": 0.42,
"learning_rate": 0.00029182547642928785,
"loss": 2.6462,
"theoretical_loss": 3.539135967008943,
"tokens_seen": 1393557504
},
{
"epoch": 0.42,
"learning_rate": 0.0002918154463390171,
"loss": 2.7556,
"theoretical_loss": 3.5391210847201044,
"tokens_seen": 1393623040
},
{
"epoch": 0.42,
"learning_rate": 0.00029180541624874626,
"loss": 2.5547,
"theoretical_loss": 3.5391062033270453,
"tokens_seen": 1393688576
},
{
"epoch": 0.42,
"learning_rate": 0.00029179538615847544,
"loss": 2.7012,
"theoretical_loss": 3.5390913228296697,
"tokens_seen": 1393754112
},
{
"epoch": 0.42,
"learning_rate": 0.0002917853560682046,
"loss": 2.7014,
"theoretical_loss": 3.539076443227882,
"tokens_seen": 1393819648
},
{
"epoch": 0.42,
"learning_rate": 0.0002917753259779338,
"loss": 2.526,
"theoretical_loss": 3.539061564521586,
"tokens_seen": 1393885184
},
{
"epoch": 0.42,
"learning_rate": 0.000291765295887663,
"loss": 2.765,
"theoretical_loss": 3.5390466867106856,
"tokens_seen": 1393950720
},
{
"epoch": 0.42,
"learning_rate": 0.0002917552657973922,
"loss": 2.5763,
"theoretical_loss": 3.539031809795085,
"tokens_seen": 1394016256
},
{
"epoch": 0.42,
"learning_rate": 0.00029174523570712135,
"loss": 2.6533,
"theoretical_loss": 3.5390169337746884,
"tokens_seen": 1394081792
},
{
"epoch": 0.42,
"learning_rate": 0.0002917352056168506,
"loss": 2.4763,
"theoretical_loss": 3.539002058649399,
"tokens_seen": 1394147328
},
{
"epoch": 0.42,
"learning_rate": 0.0002917251755265797,
"loss": 2.8207,
"theoretical_loss": 3.5389871844191223,
"tokens_seen": 1394212864
},
{
"epoch": 0.42,
"objective/train/docs_used": 789787,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.706613540649414,
"objective/train/theoretical_loss": 3.5389723110837608,
"objective/train/tokens_used": 1414738400,
"theoretical_loss": 3.5389723110837608,
"tokens_seen": 1394278400
},
{
"epoch": 0.42,
"learning_rate": 0.00029171514543630895,
"loss": 2.5154,
"theoretical_loss": 3.5389723110837608,
"tokens_seen": 1394278400
},
{
"epoch": 0.42,
"learning_rate": 0.0002917051153460381,
"loss": 2.5897,
"theoretical_loss": 3.53895743864322,
"tokens_seen": 1394343936
},
{
"epoch": 0.42,
"learning_rate": 0.0002916950852557673,
"loss": 2.7571,
"theoretical_loss": 3.538942567097403,
"tokens_seen": 1394409472
},
{
"epoch": 0.42,
"learning_rate": 0.0002916850551654965,
"loss": 2.8474,
"theoretical_loss": 3.538927696446214,
"tokens_seen": 1394475008
},
{
"epoch": 0.42,
"learning_rate": 0.00029167502507522567,
"loss": 2.6402,
"theoretical_loss": 3.538912826689558,
"tokens_seen": 1394540544
},
{
"epoch": 0.42,
"learning_rate": 0.00029166499498495485,
"loss": 2.8137,
"theoretical_loss": 3.5388979578273383,
"tokens_seen": 1394606080
},
{
"epoch": 0.42,
"learning_rate": 0.0002916549648946841,
"loss": 2.7669,
"theoretical_loss": 3.538883089859459,
"tokens_seen": 1394671616
},
{
"epoch": 0.42,
"learning_rate": 0.0002916449348044132,
"loss": 2.8316,
"theoretical_loss": 3.5388682227858252,
"tokens_seen": 1394737152
},
{
"epoch": 0.42,
"learning_rate": 0.00029163490471414245,
"loss": 2.833,
"theoretical_loss": 3.5388533566063405,
"tokens_seen": 1394802688
},
{
"epoch": 0.42,
"learning_rate": 0.00029162487462387163,
"loss": 2.52,
"theoretical_loss": 3.538838491320909,
"tokens_seen": 1394868224
},
{
"epoch": 0.42,
"learning_rate": 0.0002916148445336008,
"loss": 2.6987,
"theoretical_loss": 3.5388236269294353,
"tokens_seen": 1394933760
},
{
"epoch": 0.42,
"learning_rate": 0.00029160481444333,
"loss": 2.7019,
"theoretical_loss": 3.538808763431823,
"tokens_seen": 1394999296
},
{
"epoch": 0.42,
"learning_rate": 0.0002915947843530592,
"loss": 2.4291,
"theoretical_loss": 3.5387939008279767,
"tokens_seen": 1395064832
},
{
"epoch": 0.42,
"learning_rate": 0.00029158475426278835,
"loss": 2.5269,
"theoretical_loss": 3.5387790391178013,
"tokens_seen": 1395130368
},
{
"epoch": 0.42,
"learning_rate": 0.0002915747241725176,
"loss": 2.6973,
"theoretical_loss": 3.5387641783012,
"tokens_seen": 1395195904
},
{
"epoch": 0.42,
"learning_rate": 0.0002915646940822467,
"loss": 2.6137,
"theoretical_loss": 3.5387493183780783,
"tokens_seen": 1395261440
},
{
"epoch": 0.42,
"learning_rate": 0.00029155466399197595,
"loss": 2.5393,
"theoretical_loss": 3.5387344593483396,
"tokens_seen": 1395326976
},
{
"epoch": 0.42,
"learning_rate": 0.0002915446339017051,
"loss": 2.5944,
"theoretical_loss": 3.5387196012118887,
"tokens_seen": 1395392512
},
{
"epoch": 0.42,
"learning_rate": 0.0002915346038114343,
"loss": 2.9075,
"theoretical_loss": 3.538704743968629,
"tokens_seen": 1395458048
},
{
"epoch": 0.42,
"learning_rate": 0.0002915245737211635,
"loss": 2.8089,
"theoretical_loss": 3.5386898876184665,
"tokens_seen": 1395523584
},
{
"epoch": 0.42,
"learning_rate": 0.0002915145436308927,
"loss": 2.6764,
"theoretical_loss": 3.5386750321613043,
"tokens_seen": 1395589120
},
{
"epoch": 0.42,
"learning_rate": 0.00029150451354062186,
"loss": 2.6925,
"theoretical_loss": 3.538660177597047,
"tokens_seen": 1395654656
},
{
"epoch": 0.42,
"learning_rate": 0.0002914944834503511,
"loss": 2.8556,
"theoretical_loss": 3.5386453239255995,
"tokens_seen": 1395720192
},
{
"epoch": 0.42,
"learning_rate": 0.0002914844533600802,
"loss": 2.8938,
"theoretical_loss": 3.538630471146866,
"tokens_seen": 1395785728
},
{
"epoch": 0.42,
"learning_rate": 0.00029147442326980946,
"loss": 2.7372,
"theoretical_loss": 3.5386156192607503,
"tokens_seen": 1395851264
},
{
"epoch": 0.42,
"objective/train/docs_used": 790377,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.550626277923584,
"objective/train/theoretical_loss": 3.5386007682671576,
"objective/train/tokens_used": 1416376800,
"theoretical_loss": 3.5386007682671576,
"tokens_seen": 1395916800
},
{
"epoch": 0.42,
"learning_rate": 0.0002914643931795386,
"loss": 2.6576,
"theoretical_loss": 3.5386007682671576,
"tokens_seen": 1395916800
},
{
"epoch": 0.42,
"learning_rate": 0.0002914543630892678,
"loss": 2.5163,
"theoretical_loss": 3.5385859181659924,
"tokens_seen": 1395982336
},
{
"epoch": 0.42,
"learning_rate": 0.000291444332998997,
"loss": 2.6514,
"theoretical_loss": 3.538571068957159,
"tokens_seen": 1396047872
},
{
"epoch": 0.42,
"learning_rate": 0.0002914343029087262,
"loss": 2.8007,
"theoretical_loss": 3.5385562206405616,
"tokens_seen": 1396113408
},
{
"epoch": 0.42,
"learning_rate": 0.00029142427281845536,
"loss": 2.8032,
"theoretical_loss": 3.5385413732161046,
"tokens_seen": 1396178944
},
{
"epoch": 0.42,
"learning_rate": 0.00029141424272818454,
"loss": 2.7857,
"theoretical_loss": 3.5385265266836936,
"tokens_seen": 1396244480
},
{
"epoch": 0.42,
"learning_rate": 0.0002914042126379137,
"loss": 2.8013,
"theoretical_loss": 3.538511681043232,
"tokens_seen": 1396310016
},
{
"epoch": 0.42,
"learning_rate": 0.00029139418254764296,
"loss": 2.7798,
"theoretical_loss": 3.538496836294625,
"tokens_seen": 1396375552
},
{
"epoch": 0.42,
"learning_rate": 0.0002913841524573721,
"loss": 2.9128,
"theoretical_loss": 3.5384819924377764,
"tokens_seen": 1396441088
},
{
"epoch": 0.42,
"learning_rate": 0.0002913741223671013,
"loss": 2.4505,
"theoretical_loss": 3.538467149472592,
"tokens_seen": 1396506624
},
{
"epoch": 0.42,
"learning_rate": 0.00029136409227683045,
"loss": 2.4388,
"theoretical_loss": 3.538452307398975,
"tokens_seen": 1396572160
},
{
"epoch": 0.42,
"learning_rate": 0.0002913540621865597,
"loss": 2.7848,
"theoretical_loss": 3.5384374662168314,
"tokens_seen": 1396637696
},
{
"epoch": 0.42,
"learning_rate": 0.00029134403209628886,
"loss": 2.5261,
"theoretical_loss": 3.5384226259260645,
"tokens_seen": 1396703232
},
{
"epoch": 0.42,
"learning_rate": 0.00029133400200601805,
"loss": 2.7067,
"theoretical_loss": 3.53840778652658,
"tokens_seen": 1396768768
},
{
"epoch": 0.42,
"learning_rate": 0.0002913239719157472,
"loss": 2.6646,
"theoretical_loss": 3.5383929480182825,
"tokens_seen": 1396834304
},
{
"epoch": 0.42,
"learning_rate": 0.00029131394182547646,
"loss": 2.9648,
"theoretical_loss": 3.5383781104010756,
"tokens_seen": 1396899840
},
{
"epoch": 0.42,
"learning_rate": 0.0002913039117352056,
"loss": 2.6559,
"theoretical_loss": 3.5383632736748654,
"tokens_seen": 1396965376
},
{
"epoch": 0.42,
"learning_rate": 0.0002912938816449348,
"loss": 2.7136,
"theoretical_loss": 3.5383484378395553,
"tokens_seen": 1397030912
},
{
"epoch": 0.42,
"learning_rate": 0.000291283851554664,
"loss": 2.3649,
"theoretical_loss": 3.538333602895051,
"tokens_seen": 1397096448
},
{
"epoch": 0.42,
"learning_rate": 0.0002912738214643932,
"loss": 2.4893,
"theoretical_loss": 3.5383187688412567,
"tokens_seen": 1397161984
},
{
"epoch": 0.42,
"learning_rate": 0.0002912637913741224,
"loss": 2.8944,
"theoretical_loss": 3.538303935678077,
"tokens_seen": 1397227520
},
{
"epoch": 0.42,
"learning_rate": 0.00029125376128385155,
"loss": 2.6969,
"theoretical_loss": 3.5382891034054174,
"tokens_seen": 1397293056
},
{
"epoch": 0.42,
"learning_rate": 0.0002912437311935808,
"loss": 2.6746,
"theoretical_loss": 3.538274272023182,
"tokens_seen": 1397358592
},
{
"epoch": 0.42,
"learning_rate": 0.0002912337011033099,
"loss": 2.7479,
"theoretical_loss": 3.538259441531276,
"tokens_seen": 1397424128
},
{
"epoch": 0.42,
"learning_rate": 0.00029122367101303915,
"loss": 2.6096,
"theoretical_loss": 3.5382446119296036,
"tokens_seen": 1397489664
},
{
"epoch": 0.42,
"objective/train/docs_used": 791674,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.729259729385376,
"objective/train/theoretical_loss": 3.5382297832180702,
"objective/train/tokens_used": 1418015200,
"theoretical_loss": 3.5382297832180702,
"tokens_seen": 1397555200
},
{
"epoch": 0.42,
"learning_rate": 0.00029121364092276833,
"loss": 2.8746,
"theoretical_loss": 3.5382297832180702,
"tokens_seen": 1397555200
},
{
"epoch": 0.42,
"learning_rate": 0.0002912036108324975,
"loss": 2.6599,
"theoretical_loss": 3.5382149553965805,
"tokens_seen": 1397620736
},
{
"epoch": 0.42,
"learning_rate": 0.0002911935807422267,
"loss": 2.5901,
"theoretical_loss": 3.5382001284650397,
"tokens_seen": 1397686272
},
{
"epoch": 0.42,
"learning_rate": 0.00029118355065195587,
"loss": 2.6947,
"theoretical_loss": 3.538185302423352,
"tokens_seen": 1397751808
},
{
"epoch": 0.42,
"learning_rate": 0.00029117352056168505,
"loss": 2.6294,
"theoretical_loss": 3.5381704772714224,
"tokens_seen": 1397817344
},
{
"epoch": 0.42,
"learning_rate": 0.0002911634904714143,
"loss": 2.6951,
"theoretical_loss": 3.5381556530091562,
"tokens_seen": 1397882880
},
{
"epoch": 0.42,
"learning_rate": 0.0002911534603811434,
"loss": 2.7635,
"theoretical_loss": 3.5381408296364576,
"tokens_seen": 1397948416
},
{
"epoch": 0.42,
"learning_rate": 0.00029114343029087265,
"loss": 2.8187,
"theoretical_loss": 3.5381260071532328,
"tokens_seen": 1398013952
},
{
"epoch": 0.42,
"learning_rate": 0.00029113340020060183,
"loss": 2.766,
"theoretical_loss": 3.538111185559385,
"tokens_seen": 1398079488
},
{
"epoch": 0.42,
"learning_rate": 0.000291123370110331,
"loss": 2.6815,
"theoretical_loss": 3.5380963648548205,
"tokens_seen": 1398145024
},
{
"epoch": 0.42,
"learning_rate": 0.0002911133400200602,
"loss": 2.6932,
"theoretical_loss": 3.538081545039444,
"tokens_seen": 1398210560
},
{
"epoch": 0.42,
"learning_rate": 0.0002911033099297894,
"loss": 2.7291,
"theoretical_loss": 3.5380667261131595,
"tokens_seen": 1398276096
},
{
"epoch": 0.42,
"learning_rate": 0.00029109327983951855,
"loss": 2.7262,
"theoretical_loss": 3.5380519080758734,
"tokens_seen": 1398341632
},
{
"epoch": 0.42,
"learning_rate": 0.0002910832497492478,
"loss": 2.5355,
"theoretical_loss": 3.5380370909274896,
"tokens_seen": 1398407168
},
{
"epoch": 0.42,
"learning_rate": 0.0002910732196589769,
"loss": 2.6787,
"theoretical_loss": 3.538022274667914,
"tokens_seen": 1398472704
},
{
"epoch": 0.42,
"learning_rate": 0.00029106318956870615,
"loss": 2.7814,
"theoretical_loss": 3.5380074592970514,
"tokens_seen": 1398538240
},
{
"epoch": 0.42,
"learning_rate": 0.0002910531594784353,
"loss": 2.6458,
"theoretical_loss": 3.537992644814807,
"tokens_seen": 1398603776
},
{
"epoch": 0.42,
"learning_rate": 0.0002910431293881645,
"loss": 2.7197,
"theoretical_loss": 3.5379778312210846,
"tokens_seen": 1398669312
},
{
"epoch": 0.42,
"learning_rate": 0.0002910330992978937,
"loss": 2.8392,
"theoretical_loss": 3.537963018515791,
"tokens_seen": 1398734848
},
{
"epoch": 0.42,
"learning_rate": 0.0002910230692076229,
"loss": 2.2893,
"theoretical_loss": 3.53794820669883,
"tokens_seen": 1398800384
},
{
"epoch": 0.42,
"learning_rate": 0.00029101303911735206,
"loss": 2.6234,
"theoretical_loss": 3.537933395770108,
"tokens_seen": 1398865920
},
{
"epoch": 0.42,
"learning_rate": 0.0002910030090270813,
"loss": 2.6486,
"theoretical_loss": 3.5379185857295283,
"tokens_seen": 1398931456
},
{
"epoch": 0.42,
"learning_rate": 0.0002909929789368104,
"loss": 2.6357,
"theoretical_loss": 3.5379037765769983,
"tokens_seen": 1398996992
},
{
"epoch": 0.42,
"learning_rate": 0.00029098294884653966,
"loss": 2.7425,
"theoretical_loss": 3.537888968312421,
"tokens_seen": 1399062528
},
{
"epoch": 0.42,
"learning_rate": 0.0002909729187562688,
"loss": 2.8131,
"theoretical_loss": 3.5378741609357034,
"tokens_seen": 1399128064
},
{
"epoch": 0.42,
"objective/train/docs_used": 792315,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5115575790405273,
"objective/train/theoretical_loss": 3.5378593544467494,
"objective/train/tokens_used": 1419653600,
"theoretical_loss": 3.5378593544467494,
"tokens_seen": 1399193600
},
{
"epoch": 0.42,
"learning_rate": 0.000290962888665998,
"loss": 2.4668,
"theoretical_loss": 3.5378593544467494,
"tokens_seen": 1399193600
},
{
"epoch": 0.42,
"learning_rate": 0.0002909528585757272,
"loss": 2.9751,
"theoretical_loss": 3.537844548845464,
"tokens_seen": 1399259136
},
{
"epoch": 0.42,
"learning_rate": 0.0002909428284854564,
"loss": 2.6839,
"theoretical_loss": 3.537829744131754,
"tokens_seen": 1399324672
},
{
"epoch": 0.42,
"learning_rate": 0.00029093279839518556,
"loss": 2.656,
"theoretical_loss": 3.5378149403055232,
"tokens_seen": 1399390208
},
{
"epoch": 0.42,
"learning_rate": 0.00029092276830491474,
"loss": 2.7067,
"theoretical_loss": 3.537800137366677,
"tokens_seen": 1399455744
},
{
"epoch": 0.42,
"learning_rate": 0.0002909127382146439,
"loss": 2.7079,
"theoretical_loss": 3.5377853353151214,
"tokens_seen": 1399521280
},
{
"epoch": 0.42,
"learning_rate": 0.00029090270812437316,
"loss": 2.5671,
"theoretical_loss": 3.537770534150761,
"tokens_seen": 1399586816
},
{
"epoch": 0.42,
"learning_rate": 0.0002908926780341023,
"loss": 2.6287,
"theoretical_loss": 3.5377557338735017,
"tokens_seen": 1399652352
},
{
"epoch": 0.42,
"learning_rate": 0.0002908826479438315,
"loss": 2.747,
"theoretical_loss": 3.537740934483248,
"tokens_seen": 1399717888
},
{
"epoch": 0.42,
"learning_rate": 0.00029087261785356065,
"loss": 2.9112,
"theoretical_loss": 3.5377261359799057,
"tokens_seen": 1399783424
},
{
"epoch": 0.42,
"learning_rate": 0.0002908625877632899,
"loss": 2.7011,
"theoretical_loss": 3.5377113383633803,
"tokens_seen": 1399848960
},
{
"epoch": 0.42,
"learning_rate": 0.00029085255767301906,
"loss": 2.7791,
"theoretical_loss": 3.5376965416335766,
"tokens_seen": 1399914496
},
{
"epoch": 0.42,
"learning_rate": 0.00029084252758274825,
"loss": 2.8184,
"theoretical_loss": 3.5376817457904,
"tokens_seen": 1399980032
},
{
"epoch": 0.42,
"learning_rate": 0.0002908324974924774,
"loss": 2.8436,
"theoretical_loss": 3.5376669508337564,
"tokens_seen": 1400045568
},
{
"epoch": 0.42,
"learning_rate": 0.00029082246740220666,
"loss": 2.8884,
"theoretical_loss": 3.537652156763551,
"tokens_seen": 1400111104
},
{
"epoch": 0.42,
"learning_rate": 0.0002908124373119358,
"loss": 2.5531,
"theoretical_loss": 3.537637363579689,
"tokens_seen": 1400176640
},
{
"epoch": 0.42,
"learning_rate": 0.000290802407221665,
"loss": 2.7356,
"theoretical_loss": 3.537622571282076,
"tokens_seen": 1400242176
},
{
"epoch": 0.42,
"learning_rate": 0.00029079237713139415,
"loss": 2.448,
"theoretical_loss": 3.5376077798706174,
"tokens_seen": 1400307712
},
{
"epoch": 0.42,
"learning_rate": 0.0002907823470411234,
"loss": 2.7759,
"theoretical_loss": 3.5375929893452187,
"tokens_seen": 1400373248
},
{
"epoch": 0.42,
"learning_rate": 0.00029077231695085257,
"loss": 2.8054,
"theoretical_loss": 3.537578199705785,
"tokens_seen": 1400438784
},
{
"epoch": 0.42,
"learning_rate": 0.00029076228686058175,
"loss": 2.7785,
"theoretical_loss": 3.537563410952222,
"tokens_seen": 1400504320
},
{
"epoch": 0.42,
"learning_rate": 0.00029075225677031093,
"loss": 2.7206,
"theoretical_loss": 3.5375486230844353,
"tokens_seen": 1400569856
},
{
"epoch": 0.42,
"learning_rate": 0.0002907422266800401,
"loss": 2.7487,
"theoretical_loss": 3.5375338361023303,
"tokens_seen": 1400635392
},
{
"epoch": 0.42,
"learning_rate": 0.0002907321965897693,
"loss": 2.6892,
"theoretical_loss": 3.5375190500058125,
"tokens_seen": 1400700928
},
{
"epoch": 0.42,
"learning_rate": 0.00029072216649949853,
"loss": 2.8509,
"theoretical_loss": 3.5375042647947876,
"tokens_seen": 1400766464
},
{
"epoch": 0.42,
"objective/train/docs_used": 793814,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7395641803741455,
"objective/train/theoretical_loss": 3.537489480469161,
"objective/train/tokens_used": 1421292000,
"theoretical_loss": 3.537489480469161,
"tokens_seen": 1400832000
},
{
"epoch": 0.42,
"learning_rate": 0.00029071213640922765,
"loss": 2.6685,
"theoretical_loss": 3.537489480469161,
"tokens_seen": 1400832000
},
{
"epoch": 0.42,
"learning_rate": 0.0002907021063189569,
"loss": 2.8707,
"theoretical_loss": 3.5374746970288378,
"tokens_seen": 1400897536
},
{
"epoch": 0.42,
"learning_rate": 0.000290692076228686,
"loss": 2.822,
"theoretical_loss": 3.5374599144737244,
"tokens_seen": 1400963072
},
{
"epoch": 0.42,
"learning_rate": 0.00029068204613841525,
"loss": 2.7127,
"theoretical_loss": 3.537445132803726,
"tokens_seen": 1401028608
},
{
"epoch": 0.42,
"learning_rate": 0.00029067201604814443,
"loss": 2.671,
"theoretical_loss": 3.537430352018749,
"tokens_seen": 1401094144
},
{
"epoch": 0.42,
"learning_rate": 0.0002906619859578736,
"loss": 2.6991,
"theoretical_loss": 3.537415572118697,
"tokens_seen": 1401159680
},
{
"epoch": 0.42,
"learning_rate": 0.0002906519558676028,
"loss": 2.8776,
"theoretical_loss": 3.5374007931034774,
"tokens_seen": 1401225216
},
{
"epoch": 0.42,
"learning_rate": 0.00029064192577733203,
"loss": 2.4408,
"theoretical_loss": 3.5373860149729954,
"tokens_seen": 1401290752
},
{
"epoch": 0.42,
"learning_rate": 0.00029063189568706116,
"loss": 2.5436,
"theoretical_loss": 3.5373712377271564,
"tokens_seen": 1401356288
},
{
"epoch": 0.42,
"learning_rate": 0.0002906218655967904,
"loss": 2.8714,
"theoretical_loss": 3.537356461365867,
"tokens_seen": 1401421824
},
{
"epoch": 0.42,
"learning_rate": 0.0002906118355065195,
"loss": 2.8638,
"theoretical_loss": 3.5373416858890314,
"tokens_seen": 1401487360
},
{
"epoch": 0.42,
"learning_rate": 0.00029060180541624876,
"loss": 2.5987,
"theoretical_loss": 3.5373269112965566,
"tokens_seen": 1401552896
},
{
"epoch": 0.42,
"learning_rate": 0.00029059177532597794,
"loss": 2.6079,
"theoretical_loss": 3.537312137588347,
"tokens_seen": 1401618432
},
{
"epoch": 0.42,
"learning_rate": 0.0002905817452357071,
"loss": 2.7251,
"theoretical_loss": 3.53729736476431,
"tokens_seen": 1401683968
},
{
"epoch": 0.42,
"learning_rate": 0.0002905717151454363,
"loss": 2.5087,
"theoretical_loss": 3.53728259282435,
"tokens_seen": 1401749504
},
{
"epoch": 0.42,
"learning_rate": 0.0002905616850551655,
"loss": 2.5654,
"theoretical_loss": 3.5372678217683733,
"tokens_seen": 1401815040
},
{
"epoch": 0.42,
"learning_rate": 0.00029055165496489466,
"loss": 2.5668,
"theoretical_loss": 3.5372530515962852,
"tokens_seen": 1401880576
},
{
"epoch": 0.42,
"learning_rate": 0.0002905416248746239,
"loss": 2.5963,
"theoretical_loss": 3.5372382823079924,
"tokens_seen": 1401946112
},
{
"epoch": 0.42,
"learning_rate": 0.0002905315947843531,
"loss": 2.6889,
"theoretical_loss": 3.5372235139034,
"tokens_seen": 1402011648
},
{
"epoch": 0.42,
"learning_rate": 0.00029052156469408226,
"loss": 2.68,
"theoretical_loss": 3.5372087463824142,
"tokens_seen": 1402077184
},
{
"epoch": 0.42,
"learning_rate": 0.0002905115346038115,
"loss": 2.729,
"theoretical_loss": 3.5371939797449405,
"tokens_seen": 1402142720
},
{
"epoch": 0.42,
"learning_rate": 0.0002905015045135406,
"loss": 2.7742,
"theoretical_loss": 3.5371792139908855,
"tokens_seen": 1402208256
},
{
"epoch": 0.42,
"learning_rate": 0.00029049147442326986,
"loss": 2.6424,
"theoretical_loss": 3.5371644491201537,
"tokens_seen": 1402273792
},
{
"epoch": 0.42,
"learning_rate": 0.000290481444332999,
"loss": 2.5355,
"theoretical_loss": 3.5371496851326523,
"tokens_seen": 1402339328
},
{
"epoch": 0.42,
"learning_rate": 0.0002904714142427282,
"loss": 2.6877,
"theoretical_loss": 3.5371349220282866,
"tokens_seen": 1402404864
},
{
"epoch": 0.42,
"objective/train/docs_used": 794381,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.337651014328003,
"objective/train/theoretical_loss": 3.537120159806962,
"objective/train/tokens_used": 1422930400,
"theoretical_loss": 3.537120159806962,
"tokens_seen": 1402470400
},
{
"epoch": 0.42,
"learning_rate": 0.0002904613841524574,
"loss": 2.6955,
"theoretical_loss": 3.537120159806962,
"tokens_seen": 1402470400
},
{
"epoch": 0.43,
"learning_rate": 0.0002904513540621866,
"loss": 2.6679,
"theoretical_loss": 3.5371053984685856,
"tokens_seen": 1402535936
},
{
"epoch": 0.43,
"learning_rate": 0.00029044132397191576,
"loss": 2.8402,
"theoretical_loss": 3.5370906380130624,
"tokens_seen": 1402601472
},
{
"epoch": 0.43,
"learning_rate": 0.00029043129388164494,
"loss": 2.6996,
"theoretical_loss": 3.537075878440299,
"tokens_seen": 1402667008
},
{
"epoch": 0.43,
"learning_rate": 0.0002904212637913741,
"loss": 2.6742,
"theoretical_loss": 3.5370611197502004,
"tokens_seen": 1402732544
},
{
"epoch": 0.43,
"learning_rate": 0.00029041123370110336,
"loss": 2.6735,
"theoretical_loss": 3.5370463619426733,
"tokens_seen": 1402798080
},
{
"epoch": 0.43,
"learning_rate": 0.0002904012036108325,
"loss": 2.6678,
"theoretical_loss": 3.5370316050176243,
"tokens_seen": 1402863616
},
{
"epoch": 0.43,
"learning_rate": 0.0002903911735205617,
"loss": 2.6585,
"theoretical_loss": 3.5370168489749583,
"tokens_seen": 1402929152
},
{
"epoch": 0.43,
"learning_rate": 0.00029038114343029085,
"loss": 2.7358,
"theoretical_loss": 3.5370020938145816,
"tokens_seen": 1402994688
},
{
"epoch": 0.43,
"learning_rate": 0.0002903711133400201,
"loss": 2.8123,
"theoretical_loss": 3.5369873395364007,
"tokens_seen": 1403060224
},
{
"epoch": 0.43,
"learning_rate": 0.00029036108324974926,
"loss": 2.6943,
"theoretical_loss": 3.5369725861403207,
"tokens_seen": 1403125760
},
{
"epoch": 0.43,
"learning_rate": 0.00029035105315947845,
"loss": 2.8623,
"theoretical_loss": 3.536957833626249,
"tokens_seen": 1403191296
},
{
"epoch": 0.43,
"learning_rate": 0.0002903410230692076,
"loss": 2.6871,
"theoretical_loss": 3.5369430819940906,
"tokens_seen": 1403256832
},
{
"epoch": 0.43,
"learning_rate": 0.00029033099297893686,
"loss": 2.7627,
"theoretical_loss": 3.536928331243752,
"tokens_seen": 1403322368
},
{
"epoch": 0.43,
"learning_rate": 0.000290320962888666,
"loss": 2.6316,
"theoretical_loss": 3.5369135813751393,
"tokens_seen": 1403387904
},
{
"epoch": 0.43,
"learning_rate": 0.0002903109327983952,
"loss": 2.719,
"theoretical_loss": 3.536898832388159,
"tokens_seen": 1403453440
},
{
"epoch": 0.43,
"learning_rate": 0.00029030090270812435,
"loss": 2.8812,
"theoretical_loss": 3.5368840842827165,
"tokens_seen": 1403518976
},
{
"epoch": 0.43,
"learning_rate": 0.0002902908726178536,
"loss": 2.625,
"theoretical_loss": 3.536869337058718,
"tokens_seen": 1403584512
},
{
"epoch": 0.43,
"learning_rate": 0.00029028084252758277,
"loss": 2.5014,
"theoretical_loss": 3.5368545907160707,
"tokens_seen": 1403650048
},
{
"epoch": 0.43,
"learning_rate": 0.00029027081243731195,
"loss": 2.6842,
"theoretical_loss": 3.536839845254679,
"tokens_seen": 1403715584
},
{
"epoch": 0.43,
"learning_rate": 0.00029026078234704113,
"loss": 2.8229,
"theoretical_loss": 3.536825100674451,
"tokens_seen": 1403781120
},
{
"epoch": 0.43,
"learning_rate": 0.0002902507522567703,
"loss": 2.754,
"theoretical_loss": 3.5368103569752916,
"tokens_seen": 1403846656
},
{
"epoch": 0.43,
"learning_rate": 0.0002902407221664995,
"loss": 2.5139,
"theoretical_loss": 3.5367956141571075,
"tokens_seen": 1403912192
},
{
"epoch": 0.43,
"learning_rate": 0.00029023069207622873,
"loss": 2.5574,
"theoretical_loss": 3.536780872219805,
"tokens_seen": 1403977728
},
{
"epoch": 0.43,
"learning_rate": 0.00029022066198595785,
"loss": 2.5354,
"theoretical_loss": 3.53676613116329,
"tokens_seen": 1404043264
},
{
"epoch": 0.43,
"objective/train/docs_used": 795506,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7257678508758545,
"objective/train/theoretical_loss": 3.5367513909874697,
"objective/train/tokens_used": 1424568800,
"theoretical_loss": 3.5367513909874697,
"tokens_seen": 1404108800
},
{
"epoch": 0.43,
"learning_rate": 0.0002902106318956871,
"loss": 2.7488,
"theoretical_loss": 3.5367513909874697,
"tokens_seen": 1404108800
},
{
"epoch": 0.43,
"learning_rate": 0.0002902006018054162,
"loss": 2.6347,
"theoretical_loss": 3.536736651692249,
"tokens_seen": 1404174336
},
{
"epoch": 0.43,
"learning_rate": 0.00029019057171514545,
"loss": 2.7227,
"theoretical_loss": 3.5367219132775354,
"tokens_seen": 1404239872
},
{
"epoch": 0.43,
"learning_rate": 0.00029018054162487463,
"loss": 2.7727,
"theoretical_loss": 3.5367071757432345,
"tokens_seen": 1404305408
},
{
"epoch": 0.43,
"learning_rate": 0.0002901705115346038,
"loss": 2.7482,
"theoretical_loss": 3.5366924390892525,
"tokens_seen": 1404370944
},
{
"epoch": 0.43,
"learning_rate": 0.000290160481444333,
"loss": 2.8169,
"theoretical_loss": 3.5366777033154966,
"tokens_seen": 1404436480
},
{
"epoch": 0.43,
"learning_rate": 0.00029015045135406223,
"loss": 2.8477,
"theoretical_loss": 3.536662968421872,
"tokens_seen": 1404502016
},
{
"epoch": 0.43,
"learning_rate": 0.00029014042126379136,
"loss": 2.5957,
"theoretical_loss": 3.536648234408286,
"tokens_seen": 1404567552
},
{
"epoch": 0.43,
"learning_rate": 0.0002901303911735206,
"loss": 2.7242,
"theoretical_loss": 3.536633501274645,
"tokens_seen": 1404633088
},
{
"epoch": 0.43,
"learning_rate": 0.0002901203610832497,
"loss": 2.351,
"theoretical_loss": 3.536618769020855,
"tokens_seen": 1404698624
},
{
"epoch": 0.43,
"learning_rate": 0.00029011033099297896,
"loss": 2.6724,
"theoretical_loss": 3.536604037646822,
"tokens_seen": 1404764160
},
{
"epoch": 0.43,
"learning_rate": 0.00029010030090270814,
"loss": 2.8175,
"theoretical_loss": 3.5365893071524535,
"tokens_seen": 1404829696
},
{
"epoch": 0.43,
"learning_rate": 0.0002900902708124373,
"loss": 2.531,
"theoretical_loss": 3.536574577537655,
"tokens_seen": 1404895232
},
{
"epoch": 0.43,
"learning_rate": 0.0002900802407221665,
"loss": 2.8148,
"theoretical_loss": 3.5365598488023333,
"tokens_seen": 1404960768
},
{
"epoch": 0.43,
"learning_rate": 0.0002900702106318957,
"loss": 2.6564,
"theoretical_loss": 3.536545120946395,
"tokens_seen": 1405026304
},
{
"epoch": 0.43,
"learning_rate": 0.00029006018054162486,
"loss": 2.726,
"theoretical_loss": 3.5365303939697466,
"tokens_seen": 1405091840
},
{
"epoch": 0.43,
"learning_rate": 0.0002900501504513541,
"loss": 2.8814,
"theoretical_loss": 3.536515667872294,
"tokens_seen": 1405157376
},
{
"epoch": 0.43,
"learning_rate": 0.0002900401203610832,
"loss": 2.6983,
"theoretical_loss": 3.5365009426539444,
"tokens_seen": 1405222912
},
{
"epoch": 0.43,
"learning_rate": 0.00029003009027081246,
"loss": 2.8594,
"theoretical_loss": 3.5364862183146037,
"tokens_seen": 1405288448
},
{
"epoch": 0.43,
"learning_rate": 0.0002900200601805416,
"loss": 2.7383,
"theoretical_loss": 3.5364714948541796,
"tokens_seen": 1405353984
},
{
"epoch": 0.43,
"learning_rate": 0.0002900100300902708,
"loss": 2.506,
"theoretical_loss": 3.536456772272577,
"tokens_seen": 1405419520
},
{
"epoch": 0.43,
"learning_rate": 0.00029,
"loss": 2.6917,
"theoretical_loss": 3.536442050569704,
"tokens_seen": 1405485056
},
{
"epoch": 0.43,
"learning_rate": 0.0002899899699097292,
"loss": 2.7939,
"theoretical_loss": 3.5364273297454663,
"tokens_seen": 1405550592
},
{
"epoch": 0.43,
"learning_rate": 0.00028997993981945836,
"loss": 2.7086,
"theoretical_loss": 3.5364126097997706,
"tokens_seen": 1405616128
},
{
"epoch": 0.43,
"learning_rate": 0.0002899699097291876,
"loss": 2.4815,
"theoretical_loss": 3.5363978907325233,
"tokens_seen": 1405681664
},
{
"epoch": 0.43,
"objective/train/docs_used": 796159,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7325944900512695,
"objective/train/theoretical_loss": 3.5363831725436317,
"objective/train/tokens_used": 1426207200,
"theoretical_loss": 3.5363831725436317,
"tokens_seen": 1405747200
},
{
"epoch": 0.43,
"learning_rate": 0.0002899598796389167,
"loss": 2.6028,
"theoretical_loss": 3.5363831725436317,
"tokens_seen": 1405747200
},
{
"epoch": 0.43,
"learning_rate": 0.00028994984954864596,
"loss": 2.695,
"theoretical_loss": 3.536368455233003,
"tokens_seen": 1405812736
},
{
"epoch": 0.43,
"learning_rate": 0.0002899398194583751,
"loss": 2.7604,
"theoretical_loss": 3.5363537388005417,
"tokens_seen": 1405878272
},
{
"epoch": 0.43,
"learning_rate": 0.0002899297893681043,
"loss": 2.6554,
"theoretical_loss": 3.536339023246156,
"tokens_seen": 1405943808
},
{
"epoch": 0.43,
"learning_rate": 0.0002899197592778335,
"loss": 2.6191,
"theoretical_loss": 3.5363243085697524,
"tokens_seen": 1406009344
},
{
"epoch": 0.43,
"learning_rate": 0.0002899097291875627,
"loss": 2.6882,
"theoretical_loss": 3.5363095947712377,
"tokens_seen": 1406074880
},
{
"epoch": 0.43,
"learning_rate": 0.00028989969909729187,
"loss": 2.7037,
"theoretical_loss": 3.5362948818505178,
"tokens_seen": 1406140416
},
{
"epoch": 0.43,
"learning_rate": 0.00028988966900702105,
"loss": 2.7306,
"theoretical_loss": 3.5362801698075006,
"tokens_seen": 1406205952
},
{
"epoch": 0.43,
"learning_rate": 0.00028987963891675023,
"loss": 2.6566,
"theoretical_loss": 3.536265458642092,
"tokens_seen": 1406271488
},
{
"epoch": 0.43,
"learning_rate": 0.00028986960882647946,
"loss": 2.7993,
"theoretical_loss": 3.536250748354199,
"tokens_seen": 1406337024
},
{
"epoch": 0.43,
"learning_rate": 0.0002898595787362086,
"loss": 2.6442,
"theoretical_loss": 3.536236038943729,
"tokens_seen": 1406402560
},
{
"epoch": 0.43,
"learning_rate": 0.0002898495486459378,
"loss": 2.5234,
"theoretical_loss": 3.536221330410587,
"tokens_seen": 1406468096
},
{
"epoch": 0.43,
"learning_rate": 0.000289839518555667,
"loss": 2.673,
"theoretical_loss": 3.5362066227546816,
"tokens_seen": 1406533632
},
{
"epoch": 0.43,
"learning_rate": 0.0002898294884653962,
"loss": 2.7651,
"theoretical_loss": 3.5361919159759196,
"tokens_seen": 1406599168
},
{
"epoch": 0.43,
"learning_rate": 0.00028981945837512537,
"loss": 2.7438,
"theoretical_loss": 3.5361772100742064,
"tokens_seen": 1406664704
},
{
"epoch": 0.43,
"learning_rate": 0.00028980942828485455,
"loss": 2.7419,
"theoretical_loss": 3.5361625050494494,
"tokens_seen": 1406730240
},
{
"epoch": 0.43,
"learning_rate": 0.00028979939819458373,
"loss": 2.8146,
"theoretical_loss": 3.5361478009015563,
"tokens_seen": 1406795776
},
{
"epoch": 0.43,
"learning_rate": 0.00028978936810431297,
"loss": 2.3188,
"theoretical_loss": 3.536133097630433,
"tokens_seen": 1406861312
},
{
"epoch": 0.43,
"learning_rate": 0.00028977933801404215,
"loss": 2.5511,
"theoretical_loss": 3.536118395235987,
"tokens_seen": 1406926848
},
{
"epoch": 0.43,
"learning_rate": 0.00028976930792377133,
"loss": 2.5375,
"theoretical_loss": 3.5361036937181254,
"tokens_seen": 1406992384
},
{
"epoch": 0.43,
"learning_rate": 0.0002897592778335005,
"loss": 2.8828,
"theoretical_loss": 3.5360889930767536,
"tokens_seen": 1407057920
},
{
"epoch": 0.43,
"learning_rate": 0.0002897492477432297,
"loss": 2.8633,
"theoretical_loss": 3.5360742933117804,
"tokens_seen": 1407123456
},
{
"epoch": 0.43,
"learning_rate": 0.00028973921765295893,
"loss": 3.1492,
"theoretical_loss": 3.5360595944231115,
"tokens_seen": 1407188992
},
{
"epoch": 0.43,
"learning_rate": 0.00028972918756268805,
"loss": 2.6349,
"theoretical_loss": 3.536044896410654,
"tokens_seen": 1407254528
},
{
"epoch": 0.43,
"learning_rate": 0.0002897191574724173,
"loss": 2.5684,
"theoretical_loss": 3.5360301992743155,
"tokens_seen": 1407320064
},
{
"epoch": 0.43,
"objective/train/docs_used": 797613,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.717883586883545,
"objective/train/theoretical_loss": 3.536015503014003,
"objective/train/tokens_used": 1427845600,
"theoretical_loss": 3.536015503014003,
"tokens_seen": 1407385600
},
{
"epoch": 0.43,
"learning_rate": 0.0002897091273821464,
"loss": 2.7889,
"theoretical_loss": 3.536015503014003,
"tokens_seen": 1407385600
},
{
"epoch": 0.43,
"learning_rate": 0.00028969909729187565,
"loss": 2.575,
"theoretical_loss": 3.5360008076296223,
"tokens_seen": 1407451136
},
{
"epoch": 0.43,
"learning_rate": 0.00028968906720160483,
"loss": 2.8083,
"theoretical_loss": 3.5359861131210817,
"tokens_seen": 1407516672
},
{
"epoch": 0.43,
"learning_rate": 0.000289679037111334,
"loss": 2.654,
"theoretical_loss": 3.5359714194882876,
"tokens_seen": 1407582208
},
{
"epoch": 0.43,
"learning_rate": 0.0002896690070210632,
"loss": 2.7783,
"theoretical_loss": 3.535956726731147,
"tokens_seen": 1407647744
},
{
"epoch": 0.43,
"learning_rate": 0.00028965897693079243,
"loss": 2.7623,
"theoretical_loss": 3.535942034849567,
"tokens_seen": 1407713280
},
{
"epoch": 0.43,
"learning_rate": 0.00028964894684052156,
"loss": 2.7288,
"theoretical_loss": 3.535927343843455,
"tokens_seen": 1407778816
},
{
"epoch": 0.43,
"learning_rate": 0.0002896389167502508,
"loss": 2.6913,
"theoretical_loss": 3.5359126537127183,
"tokens_seen": 1407844352
},
{
"epoch": 0.43,
"learning_rate": 0.0002896288866599799,
"loss": 2.5449,
"theoretical_loss": 3.535897964457263,
"tokens_seen": 1407909888
},
{
"epoch": 0.43,
"learning_rate": 0.00028961885656970916,
"loss": 2.7385,
"theoretical_loss": 3.535883276076997,
"tokens_seen": 1407975424
},
{
"epoch": 0.43,
"learning_rate": 0.00028960882647943834,
"loss": 2.5962,
"theoretical_loss": 3.5358685885718275,
"tokens_seen": 1408040960
},
{
"epoch": 0.43,
"learning_rate": 0.0002895987963891675,
"loss": 2.8114,
"theoretical_loss": 3.535853901941661,
"tokens_seen": 1408106496
},
{
"epoch": 0.43,
"learning_rate": 0.0002895887662988967,
"loss": 3.0106,
"theoretical_loss": 3.535839216186405,
"tokens_seen": 1408172032
},
{
"epoch": 0.43,
"learning_rate": 0.0002895787362086259,
"loss": 2.6134,
"theoretical_loss": 3.535824531305966,
"tokens_seen": 1408237568
},
{
"epoch": 0.43,
"learning_rate": 0.00028956870611835506,
"loss": 2.7067,
"theoretical_loss": 3.5358098473002526,
"tokens_seen": 1408303104
},
{
"epoch": 0.43,
"learning_rate": 0.0002895586760280843,
"loss": 2.7549,
"theoretical_loss": 3.5357951641691714,
"tokens_seen": 1408368640
},
{
"epoch": 0.43,
"learning_rate": 0.0002895486459378134,
"loss": 2.5324,
"theoretical_loss": 3.5357804819126293,
"tokens_seen": 1408434176
},
{
"epoch": 0.43,
"learning_rate": 0.00028953861584754266,
"loss": 2.5862,
"theoretical_loss": 3.5357658005305335,
"tokens_seen": 1408499712
},
{
"epoch": 0.43,
"learning_rate": 0.0002895285857572718,
"loss": 2.5186,
"theoretical_loss": 3.535751120022791,
"tokens_seen": 1408565248
},
{
"epoch": 0.43,
"learning_rate": 0.000289518555667001,
"loss": 2.5942,
"theoretical_loss": 3.53573644038931,
"tokens_seen": 1408630784
},
{
"epoch": 0.43,
"learning_rate": 0.0002895085255767302,
"loss": 2.6458,
"theoretical_loss": 3.535721761629997,
"tokens_seen": 1408696320
},
{
"epoch": 0.43,
"learning_rate": 0.0002894984954864594,
"loss": 2.7837,
"theoretical_loss": 3.53570708374476,
"tokens_seen": 1408761856
},
{
"epoch": 0.43,
"learning_rate": 0.00028948846539618856,
"loss": 2.6533,
"theoretical_loss": 3.5356924067335047,
"tokens_seen": 1408827392
},
{
"epoch": 0.43,
"learning_rate": 0.0002894784353059178,
"loss": 2.7312,
"theoretical_loss": 3.53567773059614,
"tokens_seen": 1408892928
},
{
"epoch": 0.43,
"learning_rate": 0.0002894684052156469,
"loss": 2.7949,
"theoretical_loss": 3.5356630553325727,
"tokens_seen": 1408958464
},
{
"epoch": 0.43,
"objective/train/docs_used": 798401,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6719274520874023,
"objective/train/theoretical_loss": 3.5356483809427104,
"objective/train/tokens_used": 1429484000,
"theoretical_loss": 3.5356483809427104,
"tokens_seen": 1409024000
},
{
"epoch": 0.43,
"learning_rate": 0.00028945837512537616,
"loss": 2.6603,
"theoretical_loss": 3.5356483809427104,
"tokens_seen": 1409024000
},
{
"epoch": 0.43,
"learning_rate": 0.0002894483450351053,
"loss": 2.6021,
"theoretical_loss": 3.5356337074264603,
"tokens_seen": 1409089536
},
{
"epoch": 0.43,
"learning_rate": 0.0002894383149448345,
"loss": 2.4586,
"theoretical_loss": 3.5356190347837293,
"tokens_seen": 1409155072
},
{
"epoch": 0.43,
"learning_rate": 0.0002894282848545637,
"loss": 3.0238,
"theoretical_loss": 3.535604363014425,
"tokens_seen": 1409220608
},
{
"epoch": 0.43,
"learning_rate": 0.0002894182547642929,
"loss": 2.8594,
"theoretical_loss": 3.535589692118455,
"tokens_seen": 1409286144
},
{
"epoch": 0.43,
"learning_rate": 0.00028940822467402207,
"loss": 2.8622,
"theoretical_loss": 3.5355750220957267,
"tokens_seen": 1409351680
},
{
"epoch": 0.43,
"learning_rate": 0.00028939819458375125,
"loss": 2.7049,
"theoretical_loss": 3.5355603529461477,
"tokens_seen": 1409417216
},
{
"epoch": 0.43,
"learning_rate": 0.00028938816449348043,
"loss": 2.6939,
"theoretical_loss": 3.535545684669625,
"tokens_seen": 1409482752
},
{
"epoch": 0.43,
"learning_rate": 0.00028937813440320966,
"loss": 2.753,
"theoretical_loss": 3.535531017266066,
"tokens_seen": 1409548288
},
{
"epoch": 0.43,
"learning_rate": 0.0002893681043129388,
"loss": 2.863,
"theoretical_loss": 3.535516350735379,
"tokens_seen": 1409613824
},
{
"epoch": 0.43,
"learning_rate": 0.000289358074222668,
"loss": 2.5096,
"theoretical_loss": 3.5355016850774703,
"tokens_seen": 1409679360
},
{
"epoch": 0.43,
"learning_rate": 0.0002893480441323972,
"loss": 2.9029,
"theoretical_loss": 3.535487020292248,
"tokens_seen": 1409744896
},
{
"epoch": 0.43,
"learning_rate": 0.0002893380140421264,
"loss": 2.7086,
"theoretical_loss": 3.5354723563796195,
"tokens_seen": 1409810432
},
{
"epoch": 0.43,
"learning_rate": 0.00028932798395185557,
"loss": 2.708,
"theoretical_loss": 3.5354576933394926,
"tokens_seen": 1409875968
},
{
"epoch": 0.43,
"learning_rate": 0.00028931795386158475,
"loss": 2.564,
"theoretical_loss": 3.5354430311717744,
"tokens_seen": 1409941504
},
{
"epoch": 0.43,
"learning_rate": 0.00028930792377131393,
"loss": 2.701,
"theoretical_loss": 3.535428369876373,
"tokens_seen": 1410007040
},
{
"epoch": 0.43,
"learning_rate": 0.00028929789368104317,
"loss": 2.9082,
"theoretical_loss": 3.535413709453196,
"tokens_seen": 1410072576
},
{
"epoch": 0.43,
"learning_rate": 0.0002892878635907723,
"loss": 2.8177,
"theoretical_loss": 3.53539904990215,
"tokens_seen": 1410138112
},
{
"epoch": 0.43,
"learning_rate": 0.00028927783350050153,
"loss": 2.8736,
"theoretical_loss": 3.5353843912231433,
"tokens_seen": 1410203648
},
{
"epoch": 0.43,
"learning_rate": 0.00028926780341023066,
"loss": 2.6187,
"theoretical_loss": 3.535369733416083,
"tokens_seen": 1410269184
},
{
"epoch": 0.43,
"learning_rate": 0.0002892577733199599,
"loss": 2.9008,
"theoretical_loss": 3.5353550764808777,
"tokens_seen": 1410334720
},
{
"epoch": 0.43,
"learning_rate": 0.0002892477432296891,
"loss": 2.6961,
"theoretical_loss": 3.535340420417434,
"tokens_seen": 1410400256
},
{
"epoch": 0.43,
"learning_rate": 0.00028923771313941825,
"loss": 2.7556,
"theoretical_loss": 3.53532576522566,
"tokens_seen": 1410465792
},
{
"epoch": 0.43,
"learning_rate": 0.00028922768304914744,
"loss": 2.8244,
"theoretical_loss": 3.5353111109054636,
"tokens_seen": 1410531328
},
{
"epoch": 0.43,
"learning_rate": 0.0002892176529588766,
"loss": 2.7357,
"theoretical_loss": 3.5352964574567523,
"tokens_seen": 1410596864
},
{
"epoch": 0.43,
"objective/train/docs_used": 799700,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5968167781829834,
"objective/train/theoretical_loss": 3.5352818048794337,
"objective/train/tokens_used": 1431122400,
"theoretical_loss": 3.5352818048794337,
"tokens_seen": 1410662400
},
{
"epoch": 0.43,
"learning_rate": 0.0002892076228686058,
"loss": 2.7217,
"theoretical_loss": 3.5352818048794337,
"tokens_seen": 1410662400
},
{
"epoch": 0.43,
"learning_rate": 0.00028919759277833503,
"loss": 2.7766,
"theoretical_loss": 3.5352671531734154,
"tokens_seen": 1410727936
},
{
"epoch": 0.43,
"learning_rate": 0.00028918756268806416,
"loss": 2.5968,
"theoretical_loss": 3.535252502338605,
"tokens_seen": 1410793472
},
{
"epoch": 0.43,
"learning_rate": 0.0002891775325977934,
"loss": 2.6668,
"theoretical_loss": 3.5352378523749106,
"tokens_seen": 1410859008
},
{
"epoch": 0.43,
"learning_rate": 0.0002891675025075226,
"loss": 2.7724,
"theoretical_loss": 3.53522320328224,
"tokens_seen": 1410924544
},
{
"epoch": 0.43,
"learning_rate": 0.00028915747241725176,
"loss": 2.8098,
"theoretical_loss": 3.5352085550605006,
"tokens_seen": 1410990080
},
{
"epoch": 0.43,
"learning_rate": 0.00028914744232698094,
"loss": 2.7642,
"theoretical_loss": 3.5351939077096004,
"tokens_seen": 1411055616
},
{
"epoch": 0.43,
"learning_rate": 0.0002891374122367101,
"loss": 2.7586,
"theoretical_loss": 3.5351792612294473,
"tokens_seen": 1411121152
},
{
"epoch": 0.43,
"learning_rate": 0.0002891273821464393,
"loss": 2.8735,
"theoretical_loss": 3.5351646156199488,
"tokens_seen": 1411186688
},
{
"epoch": 0.43,
"learning_rate": 0.00028911735205616854,
"loss": 2.7115,
"theoretical_loss": 3.5351499708810126,
"tokens_seen": 1411252224
},
{
"epoch": 0.43,
"learning_rate": 0.00028910732196589766,
"loss": 2.6691,
"theoretical_loss": 3.5351353270125463,
"tokens_seen": 1411317760
},
{
"epoch": 0.43,
"learning_rate": 0.0002890972918756269,
"loss": 2.5576,
"theoretical_loss": 3.5351206840144593,
"tokens_seen": 1411383296
},
{
"epoch": 0.43,
"learning_rate": 0.000289087261785356,
"loss": 2.5441,
"theoretical_loss": 3.5351060418866576,
"tokens_seen": 1411448832
},
{
"epoch": 0.43,
"learning_rate": 0.00028907723169508526,
"loss": 2.8106,
"theoretical_loss": 3.5350914006290504,
"tokens_seen": 1411514368
},
{
"epoch": 0.43,
"learning_rate": 0.00028906720160481444,
"loss": 2.9099,
"theoretical_loss": 3.5350767602415445,
"tokens_seen": 1411579904
},
{
"epoch": 0.43,
"learning_rate": 0.0002890571715145436,
"loss": 2.6194,
"theoretical_loss": 3.5350621207240485,
"tokens_seen": 1411645440
},
{
"epoch": 0.43,
"learning_rate": 0.0002890471414242728,
"loss": 2.7966,
"theoretical_loss": 3.53504748207647,
"tokens_seen": 1411710976
},
{
"epoch": 0.43,
"learning_rate": 0.000289037111334002,
"loss": 2.5429,
"theoretical_loss": 3.5350328442987173,
"tokens_seen": 1411776512
},
{
"epoch": 0.43,
"learning_rate": 0.0002890270812437312,
"loss": 2.7423,
"theoretical_loss": 3.535018207390698,
"tokens_seen": 1411842048
},
{
"epoch": 0.43,
"learning_rate": 0.0002890170511534604,
"loss": 2.6497,
"theoretical_loss": 3.53500357135232,
"tokens_seen": 1411907584
},
{
"epoch": 0.43,
"learning_rate": 0.0002890070210631896,
"loss": 2.7539,
"theoretical_loss": 3.534988936183492,
"tokens_seen": 1411973120
},
{
"epoch": 0.43,
"learning_rate": 0.00028899699097291876,
"loss": 2.6811,
"theoretical_loss": 3.5349743018841204,
"tokens_seen": 1412038656
},
{
"epoch": 0.43,
"learning_rate": 0.000288986960882648,
"loss": 2.602,
"theoretical_loss": 3.534959668454115,
"tokens_seen": 1412104192
},
{
"epoch": 0.43,
"learning_rate": 0.0002889769307923771,
"loss": 2.6275,
"theoretical_loss": 3.5349450358933825,
"tokens_seen": 1412169728
},
{
"epoch": 0.43,
"learning_rate": 0.00028896690070210636,
"loss": 2.5051,
"theoretical_loss": 3.5349304042018317,
"tokens_seen": 1412235264
},
{
"epoch": 0.43,
"objective/train/docs_used": 800754,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.226128101348877,
"objective/train/theoretical_loss": 3.53491577337937,
"objective/train/tokens_used": 1432760800,
"theoretical_loss": 3.53491577337937,
"tokens_seen": 1412300800
},
{
"epoch": 0.43,
"learning_rate": 0.0002889568706118355,
"loss": 2.8258,
"theoretical_loss": 3.53491577337937,
"tokens_seen": 1412300800
},
{
"epoch": 0.43,
"learning_rate": 0.0002889468405215647,
"loss": 2.6971,
"theoretical_loss": 3.5349011434259063,
"tokens_seen": 1412366336
},
{
"epoch": 0.43,
"learning_rate": 0.0002889368104312939,
"loss": 2.905,
"theoretical_loss": 3.5348865143413484,
"tokens_seen": 1412431872
},
{
"epoch": 0.43,
"learning_rate": 0.0002889267803410231,
"loss": 2.7844,
"theoretical_loss": 3.534871886125604,
"tokens_seen": 1412497408
},
{
"epoch": 0.43,
"learning_rate": 0.00028891675025075227,
"loss": 2.6861,
"theoretical_loss": 3.534857258778581,
"tokens_seen": 1412562944
},
{
"epoch": 0.43,
"learning_rate": 0.00028890672016048145,
"loss": 2.689,
"theoretical_loss": 3.5348426323001876,
"tokens_seen": 1412628480
},
{
"epoch": 0.43,
"learning_rate": 0.00028889669007021063,
"loss": 2.6111,
"theoretical_loss": 3.534828006690333,
"tokens_seen": 1412694016
},
{
"epoch": 0.43,
"learning_rate": 0.00028888665997993986,
"loss": 2.7045,
"theoretical_loss": 3.534813381948924,
"tokens_seen": 1412759552
},
{
"epoch": 0.43,
"learning_rate": 0.000288876629889669,
"loss": 2.8224,
"theoretical_loss": 3.53479875807587,
"tokens_seen": 1412825088
},
{
"epoch": 0.43,
"learning_rate": 0.00028886659979939823,
"loss": 2.6062,
"theoretical_loss": 3.534784135071077,
"tokens_seen": 1412890624
},
{
"epoch": 0.43,
"learning_rate": 0.0002888565697091274,
"loss": 2.7244,
"theoretical_loss": 3.5347695129344556,
"tokens_seen": 1412956160
},
{
"epoch": 0.43,
"learning_rate": 0.0002888465396188566,
"loss": 2.8829,
"theoretical_loss": 3.534754891665913,
"tokens_seen": 1413021696
},
{
"epoch": 0.43,
"learning_rate": 0.00028883650952858577,
"loss": 2.6506,
"theoretical_loss": 3.534740271265358,
"tokens_seen": 1413087232
},
{
"epoch": 0.43,
"learning_rate": 0.00028882647943831495,
"loss": 2.8766,
"theoretical_loss": 3.5347256517326975,
"tokens_seen": 1413152768
},
{
"epoch": 0.43,
"learning_rate": 0.00028881644934804413,
"loss": 2.7006,
"theoretical_loss": 3.5347110330678406,
"tokens_seen": 1413218304
},
{
"epoch": 0.43,
"learning_rate": 0.00028880641925777337,
"loss": 2.7408,
"theoretical_loss": 3.5346964152706954,
"tokens_seen": 1413283840
},
{
"epoch": 0.43,
"learning_rate": 0.0002887963891675025,
"loss": 2.5484,
"theoretical_loss": 3.53468179834117,
"tokens_seen": 1413349376
},
{
"epoch": 0.43,
"learning_rate": 0.00028878635907723173,
"loss": 2.5665,
"theoretical_loss": 3.534667182279173,
"tokens_seen": 1413414912
},
{
"epoch": 0.43,
"learning_rate": 0.00028877632898696086,
"loss": 2.8107,
"theoretical_loss": 3.5346525670846125,
"tokens_seen": 1413480448
},
{
"epoch": 0.43,
"learning_rate": 0.0002887662988966901,
"loss": 2.5601,
"theoretical_loss": 3.5346379527573974,
"tokens_seen": 1413545984
},
{
"epoch": 0.43,
"learning_rate": 0.0002887562688064193,
"loss": 2.6927,
"theoretical_loss": 3.534623339297435,
"tokens_seen": 1413611520
},
{
"epoch": 0.43,
"learning_rate": 0.00028874623871614845,
"loss": 3.0346,
"theoretical_loss": 3.534608726704634,
"tokens_seen": 1413677056
},
{
"epoch": 0.43,
"learning_rate": 0.00028873620862587764,
"loss": 2.6908,
"theoretical_loss": 3.534594114978903,
"tokens_seen": 1413742592
},
{
"epoch": 0.43,
"learning_rate": 0.0002887261785356068,
"loss": 2.8633,
"theoretical_loss": 3.5345795041201504,
"tokens_seen": 1413808128
},
{
"epoch": 0.43,
"learning_rate": 0.000288716148445336,
"loss": 2.9672,
"theoretical_loss": 3.5345648941282843,
"tokens_seen": 1413873664
},
{
"epoch": 0.43,
"objective/train/docs_used": 801653,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.687365770339966,
"objective/train/theoretical_loss": 3.5345502850032133,
"objective/train/tokens_used": 1434399200,
"theoretical_loss": 3.5345502850032133,
"tokens_seen": 1413939200
},
{
"epoch": 0.43,
"learning_rate": 0.00028870611835506523,
"loss": 2.6204,
"theoretical_loss": 3.5345502850032133,
"tokens_seen": 1413939200
},
{
"epoch": 0.43,
"learning_rate": 0.00028869608826479436,
"loss": 2.7615,
"theoretical_loss": 3.534535676744845,
"tokens_seen": 1414004736
},
{
"epoch": 0.43,
"learning_rate": 0.0002886860581745236,
"loss": 2.7819,
"theoretical_loss": 3.5345210693530893,
"tokens_seen": 1414070272
},
{
"epoch": 0.43,
"learning_rate": 0.0002886760280842528,
"loss": 2.6922,
"theoretical_loss": 3.5345064628278537,
"tokens_seen": 1414135808
},
{
"epoch": 0.43,
"learning_rate": 0.00028866599799398196,
"loss": 2.9312,
"theoretical_loss": 3.5344918571690465,
"tokens_seen": 1414201344
},
{
"epoch": 0.43,
"learning_rate": 0.00028865596790371114,
"loss": 2.6196,
"theoretical_loss": 3.5344772523765764,
"tokens_seen": 1414266880
},
{
"epoch": 0.43,
"learning_rate": 0.0002886459378134403,
"loss": 2.5937,
"theoretical_loss": 3.534462648450352,
"tokens_seen": 1414332416
},
{
"epoch": 0.43,
"learning_rate": 0.0002886359077231695,
"loss": 2.6143,
"theoretical_loss": 3.534448045390282,
"tokens_seen": 1414397952
},
{
"epoch": 0.43,
"learning_rate": 0.00028862587763289874,
"loss": 2.8008,
"theoretical_loss": 3.5344334431962743,
"tokens_seen": 1414463488
},
{
"epoch": 0.43,
"learning_rate": 0.00028861584754262786,
"loss": 2.7593,
"theoretical_loss": 3.5344188418682374,
"tokens_seen": 1414529024
},
{
"epoch": 0.43,
"learning_rate": 0.0002886058174523571,
"loss": 2.609,
"theoretical_loss": 3.5344042414060803,
"tokens_seen": 1414594560
},
{
"epoch": 0.43,
"learning_rate": 0.0002885957873620862,
"loss": 2.8667,
"theoretical_loss": 3.5343896418097116,
"tokens_seen": 1414660096
},
{
"epoch": 0.43,
"learning_rate": 0.00028858575727181546,
"loss": 2.9061,
"theoretical_loss": 3.5343750430790397,
"tokens_seen": 1414725632
},
{
"epoch": 0.43,
"learning_rate": 0.00028857572718154464,
"loss": 2.7897,
"theoretical_loss": 3.5343604452139727,
"tokens_seen": 1414791168
},
{
"epoch": 0.43,
"learning_rate": 0.0002885656970912738,
"loss": 2.7737,
"theoretical_loss": 3.5343458482144197,
"tokens_seen": 1414856704
},
{
"epoch": 0.43,
"learning_rate": 0.000288555667001003,
"loss": 2.7199,
"theoretical_loss": 3.534331252080289,
"tokens_seen": 1414922240
},
{
"epoch": 0.43,
"learning_rate": 0.0002885456369107322,
"loss": 2.5737,
"theoretical_loss": 3.5343166568114897,
"tokens_seen": 1414987776
},
{
"epoch": 0.43,
"learning_rate": 0.00028853560682046137,
"loss": 2.5739,
"theoretical_loss": 3.5343020624079298,
"tokens_seen": 1415053312
},
{
"epoch": 0.43,
"learning_rate": 0.0002885255767301906,
"loss": 2.5214,
"theoretical_loss": 3.5342874688695183,
"tokens_seen": 1415118848
},
{
"epoch": 0.43,
"learning_rate": 0.00028851554663991973,
"loss": 2.5039,
"theoretical_loss": 3.534272876196164,
"tokens_seen": 1415184384
},
{
"epoch": 0.43,
"learning_rate": 0.00028850551654964896,
"loss": 2.8454,
"theoretical_loss": 3.534258284387775,
"tokens_seen": 1415249920
},
{
"epoch": 0.43,
"learning_rate": 0.00028849548645937815,
"loss": 2.6405,
"theoretical_loss": 3.5342436934442607,
"tokens_seen": 1415315456
},
{
"epoch": 0.43,
"learning_rate": 0.0002884854563691073,
"loss": 2.6539,
"theoretical_loss": 3.5342291033655293,
"tokens_seen": 1415380992
},
{
"epoch": 0.43,
"learning_rate": 0.0002884754262788365,
"loss": 2.7267,
"theoretical_loss": 3.53421451415149,
"tokens_seen": 1415446528
},
{
"epoch": 0.43,
"learning_rate": 0.0002884653961885657,
"loss": 2.5181,
"theoretical_loss": 3.5341999258020502,
"tokens_seen": 1415512064
},
{
"epoch": 0.43,
"objective/train/docs_used": 802505,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5791609287261963,
"objective/train/theoretical_loss": 3.5341853383171205,
"objective/train/tokens_used": 1436037600,
"theoretical_loss": 3.5341853383171205,
"tokens_seen": 1415577600
},
{
"epoch": 0.43,
"learning_rate": 0.00028845536609829487,
"loss": 2.5191,
"theoretical_loss": 3.5341853383171205,
"tokens_seen": 1415577600
},
{
"epoch": 0.43,
"learning_rate": 0.0002884453360080241,
"loss": 2.7001,
"theoretical_loss": 3.5341707516966085,
"tokens_seen": 1415643136
},
{
"epoch": 0.43,
"learning_rate": 0.00028843530591775323,
"loss": 2.6914,
"theoretical_loss": 3.534156165940423,
"tokens_seen": 1415708672
},
{
"epoch": 0.43,
"learning_rate": 0.00028842527582748247,
"loss": 2.9653,
"theoretical_loss": 3.5341415810484733,
"tokens_seen": 1415774208
},
{
"epoch": 0.43,
"learning_rate": 0.0002884152457372116,
"loss": 2.5809,
"theoretical_loss": 3.5341269970206675,
"tokens_seen": 1415839744
},
{
"epoch": 0.43,
"learning_rate": 0.00028840521564694083,
"loss": 2.813,
"theoretical_loss": 3.534112413856915,
"tokens_seen": 1415905280
},
{
"epoch": 0.43,
"learning_rate": 0.00028839518555667,
"loss": 2.8915,
"theoretical_loss": 3.5340978315571245,
"tokens_seen": 1415970816
},
{
"epoch": 0.43,
"learning_rate": 0.0002883851554663992,
"loss": 2.7928,
"theoretical_loss": 3.534083250121204,
"tokens_seen": 1416036352
},
{
"epoch": 0.43,
"learning_rate": 0.0002883751253761284,
"loss": 2.7148,
"theoretical_loss": 3.5340686695490637,
"tokens_seen": 1416101888
},
{
"epoch": 0.43,
"learning_rate": 0.0002883650952858576,
"loss": 2.4765,
"theoretical_loss": 3.5340540898406116,
"tokens_seen": 1416167424
},
{
"epoch": 0.43,
"learning_rate": 0.00028835506519558674,
"loss": 2.7355,
"theoretical_loss": 3.534039510995757,
"tokens_seen": 1416232960
},
{
"epoch": 0.43,
"learning_rate": 0.00028834503510531597,
"loss": 2.7382,
"theoretical_loss": 3.5340249330144085,
"tokens_seen": 1416298496
},
{
"epoch": 0.43,
"learning_rate": 0.0002883350050150451,
"loss": 2.6379,
"theoretical_loss": 3.5340103558964753,
"tokens_seen": 1416364032
},
{
"epoch": 0.43,
"learning_rate": 0.00028832497492477433,
"loss": 2.6289,
"theoretical_loss": 3.5339957796418657,
"tokens_seen": 1416429568
},
{
"epoch": 0.43,
"learning_rate": 0.0002883149448345035,
"loss": 2.6745,
"theoretical_loss": 3.5339812042504897,
"tokens_seen": 1416495104
},
{
"epoch": 0.43,
"learning_rate": 0.0002883049147442327,
"loss": 2.5008,
"theoretical_loss": 3.533966629722255,
"tokens_seen": 1416560640
},
{
"epoch": 0.43,
"learning_rate": 0.0002882948846539619,
"loss": 2.8214,
"theoretical_loss": 3.533952056057071,
"tokens_seen": 1416626176
},
{
"epoch": 0.43,
"learning_rate": 0.00028828485456369106,
"loss": 2.5528,
"theoretical_loss": 3.5339374832548476,
"tokens_seen": 1416691712
},
{
"epoch": 0.43,
"learning_rate": 0.0002882748244734203,
"loss": 2.5616,
"theoretical_loss": 3.5339229113154924,
"tokens_seen": 1416757248
},
{
"epoch": 0.43,
"learning_rate": 0.0002882647943831495,
"loss": 2.5487,
"theoretical_loss": 3.5339083402389146,
"tokens_seen": 1416822784
},
{
"epoch": 0.43,
"learning_rate": 0.00028825476429287866,
"loss": 2.5233,
"theoretical_loss": 3.5338937700250246,
"tokens_seen": 1416888320
},
{
"epoch": 0.43,
"learning_rate": 0.00028824473420260784,
"loss": 2.7855,
"theoretical_loss": 3.53387920067373,
"tokens_seen": 1416953856
},
{
"epoch": 0.43,
"learning_rate": 0.000288234704112337,
"loss": 2.7091,
"theoretical_loss": 3.53386463218494,
"tokens_seen": 1417019392
},
{
"epoch": 0.43,
"learning_rate": 0.0002882246740220662,
"loss": 2.9369,
"theoretical_loss": 3.533850064558564,
"tokens_seen": 1417084928
},
{
"epoch": 0.43,
"learning_rate": 0.00028821464393179543,
"loss": 2.7856,
"theoretical_loss": 3.5338354977945112,
"tokens_seen": 1417150464
},
{
"epoch": 0.43,
"objective/train/docs_used": 803171,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0270919799804688,
"objective/train/theoretical_loss": 3.5338209318926905,
"objective/train/tokens_used": 1437676000,
"theoretical_loss": 3.5338209318926905,
"tokens_seen": 1417216000
},
{
"epoch": 0.43,
"learning_rate": 0.00028820461384152456,
"loss": 2.546,
"theoretical_loss": 3.5338209318926905,
"tokens_seen": 1417216000
},
{
"epoch": 0.43,
"learning_rate": 0.0002881945837512538,
"loss": 2.8082,
"theoretical_loss": 3.5338063668530104,
"tokens_seen": 1417281536
},
{
"epoch": 0.43,
"learning_rate": 0.000288184553660983,
"loss": 2.6209,
"theoretical_loss": 3.533791802675381,
"tokens_seen": 1417347072
},
{
"epoch": 0.43,
"learning_rate": 0.00028817452357071216,
"loss": 2.7957,
"theoretical_loss": 3.533777239359711,
"tokens_seen": 1417412608
},
{
"epoch": 0.43,
"learning_rate": 0.00028816449348044134,
"loss": 2.5751,
"theoretical_loss": 3.5337626769059094,
"tokens_seen": 1417478144
},
{
"epoch": 0.43,
"learning_rate": 0.0002881544633901705,
"loss": 2.6956,
"theoretical_loss": 3.5337481153138857,
"tokens_seen": 1417543680
},
{
"epoch": 0.43,
"learning_rate": 0.0002881444332998997,
"loss": 2.6488,
"theoretical_loss": 3.5337335545835487,
"tokens_seen": 1417609216
},
{
"epoch": 0.43,
"learning_rate": 0.00028813440320962894,
"loss": 2.5564,
"theoretical_loss": 3.5337189947148078,
"tokens_seen": 1417674752
},
{
"epoch": 0.43,
"learning_rate": 0.00028812437311935806,
"loss": 2.6537,
"theoretical_loss": 3.5337044357075715,
"tokens_seen": 1417740288
},
{
"epoch": 0.43,
"learning_rate": 0.0002881143430290873,
"loss": 2.7427,
"theoretical_loss": 3.5336898775617502,
"tokens_seen": 1417805824
},
{
"epoch": 0.43,
"learning_rate": 0.0002881043129388164,
"loss": 2.8817,
"theoretical_loss": 3.5336753202772524,
"tokens_seen": 1417871360
},
{
"epoch": 0.43,
"learning_rate": 0.00028809428284854566,
"loss": 2.717,
"theoretical_loss": 3.5336607638539874,
"tokens_seen": 1417936896
},
{
"epoch": 0.43,
"learning_rate": 0.00028808425275827484,
"loss": 2.8618,
"theoretical_loss": 3.5336462082918647,
"tokens_seen": 1418002432
},
{
"epoch": 0.43,
"learning_rate": 0.000288074222668004,
"loss": 2.5805,
"theoretical_loss": 3.5336316535907937,
"tokens_seen": 1418067968
},
{
"epoch": 0.43,
"learning_rate": 0.0002880641925777332,
"loss": 2.7424,
"theoretical_loss": 3.533617099750683,
"tokens_seen": 1418133504
},
{
"epoch": 0.43,
"learning_rate": 0.0002880541624874624,
"loss": 2.9195,
"theoretical_loss": 3.5336025467714416,
"tokens_seen": 1418199040
},
{
"epoch": 0.43,
"learning_rate": 0.00028804413239719157,
"loss": 2.494,
"theoretical_loss": 3.5335879946529802,
"tokens_seen": 1418264576
},
{
"epoch": 0.43,
"learning_rate": 0.0002880341023069208,
"loss": 2.7024,
"theoretical_loss": 3.5335734433952073,
"tokens_seen": 1418330112
},
{
"epoch": 0.43,
"learning_rate": 0.00028802407221664993,
"loss": 2.578,
"theoretical_loss": 3.5335588929980317,
"tokens_seen": 1418395648
},
{
"epoch": 0.43,
"learning_rate": 0.00028801404212637916,
"loss": 2.6966,
"theoretical_loss": 3.5335443434613643,
"tokens_seen": 1418461184
},
{
"epoch": 0.43,
"learning_rate": 0.00028800401203610835,
"loss": 2.8355,
"theoretical_loss": 3.5335297947851125,
"tokens_seen": 1418526720
},
{
"epoch": 0.43,
"learning_rate": 0.0002879939819458375,
"loss": 2.5621,
"theoretical_loss": 3.533515246969187,
"tokens_seen": 1418592256
},
{
"epoch": 0.43,
"learning_rate": 0.0002879839518555667,
"loss": 2.7903,
"theoretical_loss": 3.533500700013497,
"tokens_seen": 1418657792
},
{
"epoch": 0.43,
"learning_rate": 0.0002879739217652959,
"loss": 2.7391,
"theoretical_loss": 3.533486153917951,
"tokens_seen": 1418723328
},
{
"epoch": 0.43,
"learning_rate": 0.00028796389167502507,
"loss": 2.5435,
"theoretical_loss": 3.5334716086824605,
"tokens_seen": 1418788864
},
{
"epoch": 0.43,
"objective/train/docs_used": 804514,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.0574936866760254,
"objective/train/theoretical_loss": 3.5334570643069325,
"objective/train/tokens_used": 1439314400,
"theoretical_loss": 3.5334570643069325,
"tokens_seen": 1418854400
},
{
"epoch": 0.43,
"learning_rate": 0.0002879538615847543,
"loss": 2.2948,
"theoretical_loss": 3.5334570643069325,
"tokens_seen": 1418854400
},
{
"epoch": 0.43,
"learning_rate": 0.00028794383149448343,
"loss": 2.9707,
"theoretical_loss": 3.5334425207912776,
"tokens_seen": 1418919936
},
{
"epoch": 0.43,
"learning_rate": 0.00028793380140421267,
"loss": 2.9228,
"theoretical_loss": 3.533427978135405,
"tokens_seen": 1418985472
},
{
"epoch": 0.43,
"learning_rate": 0.0002879237713139418,
"loss": 2.6659,
"theoretical_loss": 3.5334134363392247,
"tokens_seen": 1419051008
},
{
"epoch": 0.43,
"learning_rate": 0.00028791374122367103,
"loss": 2.8051,
"theoretical_loss": 3.533398895402646,
"tokens_seen": 1419116544
},
{
"epoch": 0.43,
"learning_rate": 0.0002879037111334002,
"loss": 2.6917,
"theoretical_loss": 3.5333843553255777,
"tokens_seen": 1419182080
},
{
"epoch": 0.43,
"learning_rate": 0.0002878936810431294,
"loss": 2.7732,
"theoretical_loss": 3.53336981610793,
"tokens_seen": 1419247616
},
{
"epoch": 0.43,
"learning_rate": 0.0002878836509528586,
"loss": 2.6162,
"theoretical_loss": 3.5333552777496124,
"tokens_seen": 1419313152
},
{
"epoch": 0.43,
"learning_rate": 0.0002878736208625878,
"loss": 2.7316,
"theoretical_loss": 3.5333407402505337,
"tokens_seen": 1419378688
},
{
"epoch": 0.43,
"learning_rate": 0.00028786359077231694,
"loss": 2.4832,
"theoretical_loss": 3.5333262036106046,
"tokens_seen": 1419444224
},
{
"epoch": 0.43,
"learning_rate": 0.00028785356068204617,
"loss": 2.7362,
"theoretical_loss": 3.533311667829734,
"tokens_seen": 1419509760
},
{
"epoch": 0.43,
"learning_rate": 0.0002878435305917753,
"loss": 2.8183,
"theoretical_loss": 3.5332971329078315,
"tokens_seen": 1419575296
},
{
"epoch": 0.43,
"learning_rate": 0.00028783350050150453,
"loss": 2.4873,
"theoretical_loss": 3.5332825988448064,
"tokens_seen": 1419640832
},
{
"epoch": 0.43,
"learning_rate": 0.0002878234704112337,
"loss": 2.3214,
"theoretical_loss": 3.533268065640569,
"tokens_seen": 1419706368
},
{
"epoch": 0.43,
"learning_rate": 0.0002878134403209629,
"loss": 2.543,
"theoretical_loss": 3.533253533295028,
"tokens_seen": 1419771904
},
{
"epoch": 0.43,
"learning_rate": 0.0002878034102306921,
"loss": 2.7039,
"theoretical_loss": 3.533239001808094,
"tokens_seen": 1419837440
},
{
"epoch": 0.43,
"learning_rate": 0.00028779338014042126,
"loss": 2.699,
"theoretical_loss": 3.5332244711796763,
"tokens_seen": 1419902976
},
{
"epoch": 0.43,
"learning_rate": 0.00028778335005015044,
"loss": 2.7085,
"theoretical_loss": 3.533209941409684,
"tokens_seen": 1419968512
},
{
"epoch": 0.43,
"learning_rate": 0.0002877733199598797,
"loss": 2.87,
"theoretical_loss": 3.533195412498028,
"tokens_seen": 1420034048
},
{
"epoch": 0.43,
"learning_rate": 0.0002877632898696088,
"loss": 2.8065,
"theoretical_loss": 3.5331808844446164,
"tokens_seen": 1420099584
},
{
"epoch": 0.43,
"learning_rate": 0.00028775325977933804,
"loss": 2.9168,
"theoretical_loss": 3.5331663572493603,
"tokens_seen": 1420165120
},
{
"epoch": 0.43,
"learning_rate": 0.00028774322968906716,
"loss": 2.6171,
"theoretical_loss": 3.5331518309121686,
"tokens_seen": 1420230656
},
{
"epoch": 0.43,
"learning_rate": 0.0002877331995987964,
"loss": 2.5741,
"theoretical_loss": 3.533137305432951,
"tokens_seen": 1420296192
},
{
"epoch": 0.43,
"learning_rate": 0.0002877231695085256,
"loss": 2.6993,
"theoretical_loss": 3.5331227808116177,
"tokens_seen": 1420361728
},
{
"epoch": 0.43,
"learning_rate": 0.00028771313941825476,
"loss": 2.4322,
"theoretical_loss": 3.5331082570480787,
"tokens_seen": 1420427264
},
{
"epoch": 0.43,
"objective/train/docs_used": 804990,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1796112060546875,
"objective/train/theoretical_loss": 3.533093734142243,
"objective/train/tokens_used": 1440952800,
"theoretical_loss": 3.533093734142243,
"tokens_seen": 1420492800
},
{
"epoch": 0.43,
"learning_rate": 0.00028770310932798394,
"loss": 2.6025,
"theoretical_loss": 3.533093734142243,
"tokens_seen": 1420492800
},
{
"epoch": 0.43,
"learning_rate": 0.0002876930792377132,
"loss": 2.7426,
"theoretical_loss": 3.5330792120940204,
"tokens_seen": 1420558336
},
{
"epoch": 0.43,
"learning_rate": 0.0002876830491474423,
"loss": 2.7082,
"theoretical_loss": 3.5330646909033216,
"tokens_seen": 1420623872
},
{
"epoch": 0.43,
"learning_rate": 0.00028767301905717154,
"loss": 2.7508,
"theoretical_loss": 3.533050170570055,
"tokens_seen": 1420689408
},
{
"epoch": 0.43,
"learning_rate": 0.00028766298896690067,
"loss": 2.6596,
"theoretical_loss": 3.5330356510941323,
"tokens_seen": 1420754944
},
{
"epoch": 0.43,
"learning_rate": 0.0002876529588766299,
"loss": 2.6399,
"theoretical_loss": 3.5330211324754615,
"tokens_seen": 1420820480
},
{
"epoch": 0.43,
"learning_rate": 0.0002876429287863591,
"loss": 2.8822,
"theoretical_loss": 3.533006614713954,
"tokens_seen": 1420886016
},
{
"epoch": 0.43,
"learning_rate": 0.00028763289869608826,
"loss": 2.9332,
"theoretical_loss": 3.532992097809518,
"tokens_seen": 1420951552
},
{
"epoch": 0.43,
"learning_rate": 0.00028762286860581745,
"loss": 2.6263,
"theoretical_loss": 3.5329775817620646,
"tokens_seen": 1421017088
},
{
"epoch": 0.43,
"learning_rate": 0.0002876128385155466,
"loss": 2.8868,
"theoretical_loss": 3.532963066571503,
"tokens_seen": 1421082624
},
{
"epoch": 0.43,
"learning_rate": 0.0002876028084252758,
"loss": 2.6735,
"theoretical_loss": 3.5329485522377437,
"tokens_seen": 1421148160
},
{
"epoch": 0.43,
"learning_rate": 0.00028759277833500504,
"loss": 2.8405,
"theoretical_loss": 3.532934038760697,
"tokens_seen": 1421213696
},
{
"epoch": 0.43,
"learning_rate": 0.00028758274824473417,
"loss": 2.5779,
"theoretical_loss": 3.5329195261402715,
"tokens_seen": 1421279232
},
{
"epoch": 0.43,
"learning_rate": 0.0002875727181544634,
"loss": 2.7155,
"theoretical_loss": 3.532905014376378,
"tokens_seen": 1421344768
},
{
"epoch": 0.43,
"learning_rate": 0.00028756268806419253,
"loss": 2.542,
"theoretical_loss": 3.532890503468926,
"tokens_seen": 1421410304
},
{
"epoch": 0.43,
"learning_rate": 0.00028755265797392177,
"loss": 2.7286,
"theoretical_loss": 3.532875993417826,
"tokens_seen": 1421475840
},
{
"epoch": 0.43,
"learning_rate": 0.00028754262788365095,
"loss": 2.7564,
"theoretical_loss": 3.5328614842229875,
"tokens_seen": 1421541376
},
{
"epoch": 0.43,
"learning_rate": 0.00028753259779338013,
"loss": 2.9088,
"theoretical_loss": 3.532846975884321,
"tokens_seen": 1421606912
},
{
"epoch": 0.43,
"learning_rate": 0.00028752256770310936,
"loss": 2.9002,
"theoretical_loss": 3.5328324684017365,
"tokens_seen": 1421672448
},
{
"epoch": 0.43,
"learning_rate": 0.00028751253761283855,
"loss": 2.7563,
"theoretical_loss": 3.5328179617751436,
"tokens_seen": 1421737984
},
{
"epoch": 0.43,
"learning_rate": 0.0002875025075225677,
"loss": 2.5986,
"theoretical_loss": 3.5328034560044523,
"tokens_seen": 1421803520
},
{
"epoch": 0.43,
"learning_rate": 0.0002874924774322969,
"loss": 2.6294,
"theoretical_loss": 3.5327889510895734,
"tokens_seen": 1421869056
},
{
"epoch": 0.43,
"learning_rate": 0.0002874824473420261,
"loss": 2.7192,
"theoretical_loss": 3.532774447030416,
"tokens_seen": 1421934592
},
{
"epoch": 0.43,
"learning_rate": 0.00028747241725175527,
"loss": 2.6684,
"theoretical_loss": 3.532759943826891,
"tokens_seen": 1422000128
},
{
"epoch": 0.43,
"learning_rate": 0.0002874623871614845,
"loss": 2.8557,
"theoretical_loss": 3.532745441478908,
"tokens_seen": 1422065664
},
{
"epoch": 0.43,
"objective/train/docs_used": 806295,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1450142860412598,
"objective/train/theoretical_loss": 3.5327309399863767,
"objective/train/tokens_used": 1442591200,
"theoretical_loss": 3.5327309399863767,
"tokens_seen": 1422131200
},
{
"epoch": 0.43,
"learning_rate": 0.00028745235707121363,
"loss": 2.7963,
"theoretical_loss": 3.5327309399863767,
"tokens_seen": 1422131200
},
{
"epoch": 0.43,
"learning_rate": 0.00028744232698094287,
"loss": 2.6517,
"theoretical_loss": 3.5327164393492083,
"tokens_seen": 1422196736
},
{
"epoch": 0.43,
"learning_rate": 0.000287432296890672,
"loss": 2.8016,
"theoretical_loss": 3.5327019395673123,
"tokens_seen": 1422262272
},
{
"epoch": 0.43,
"learning_rate": 0.00028742226680040123,
"loss": 2.649,
"theoretical_loss": 3.532687440640599,
"tokens_seen": 1422327808
},
{
"epoch": 0.43,
"learning_rate": 0.0002874122367101304,
"loss": 2.694,
"theoretical_loss": 3.5326729425689782,
"tokens_seen": 1422393344
},
{
"epoch": 0.43,
"learning_rate": 0.0002874022066198596,
"loss": 2.7915,
"theoretical_loss": 3.5326584453523604,
"tokens_seen": 1422458880
},
{
"epoch": 0.43,
"learning_rate": 0.0002873921765295888,
"loss": 2.7365,
"theoretical_loss": 3.532643948990656,
"tokens_seen": 1422524416
},
{
"epoch": 0.43,
"learning_rate": 0.000287382146439318,
"loss": 2.5364,
"theoretical_loss": 3.532629453483775,
"tokens_seen": 1422589952
},
{
"epoch": 0.43,
"learning_rate": 0.00028737211634904714,
"loss": 2.5705,
"theoretical_loss": 3.532614958831627,
"tokens_seen": 1422655488
},
{
"epoch": 0.43,
"learning_rate": 0.00028736208625877637,
"loss": 2.5646,
"theoretical_loss": 3.5326004650341236,
"tokens_seen": 1422721024
},
{
"epoch": 0.43,
"learning_rate": 0.0002873520561685055,
"loss": 2.6553,
"theoretical_loss": 3.532585972091174,
"tokens_seen": 1422786560
},
{
"epoch": 0.43,
"learning_rate": 0.00028734202607823473,
"loss": 2.4984,
"theoretical_loss": 3.532571480002688,
"tokens_seen": 1422852096
},
{
"epoch": 0.43,
"learning_rate": 0.0002873319959879639,
"loss": 2.8716,
"theoretical_loss": 3.5325569887685777,
"tokens_seen": 1422917632
},
{
"epoch": 0.43,
"learning_rate": 0.0002873219658976931,
"loss": 2.8802,
"theoretical_loss": 3.532542498388752,
"tokens_seen": 1422983168
},
{
"epoch": 0.43,
"learning_rate": 0.0002873119358074223,
"loss": 2.7131,
"theoretical_loss": 3.532528008863121,
"tokens_seen": 1423048704
},
{
"epoch": 0.43,
"learning_rate": 0.00028730190571715146,
"loss": 2.6603,
"theoretical_loss": 3.5325135201915954,
"tokens_seen": 1423114240
},
{
"epoch": 0.43,
"learning_rate": 0.00028729187562688064,
"loss": 2.7697,
"theoretical_loss": 3.532499032374086,
"tokens_seen": 1423179776
},
{
"epoch": 0.43,
"learning_rate": 0.0002872818455366099,
"loss": 2.7163,
"theoretical_loss": 3.5324845454105027,
"tokens_seen": 1423245312
},
{
"epoch": 0.43,
"learning_rate": 0.000287271815446339,
"loss": 2.8083,
"theoretical_loss": 3.5324700593007554,
"tokens_seen": 1423310848
},
{
"epoch": 0.43,
"learning_rate": 0.00028726178535606824,
"loss": 2.6522,
"theoretical_loss": 3.5324555740447554,
"tokens_seen": 1423376384
},
{
"epoch": 0.43,
"learning_rate": 0.00028725175526579736,
"loss": 2.9201,
"theoretical_loss": 3.532441089642413,
"tokens_seen": 1423441920
},
{
"epoch": 0.43,
"learning_rate": 0.0002872417251755266,
"loss": 2.6896,
"theoretical_loss": 3.5324266060936376,
"tokens_seen": 1423507456
},
{
"epoch": 0.43,
"learning_rate": 0.0002872316950852558,
"loss": 2.7109,
"theoretical_loss": 3.5324121233983403,
"tokens_seen": 1423572992
},
{
"epoch": 0.43,
"learning_rate": 0.00028722166499498496,
"loss": 2.8008,
"theoretical_loss": 3.5323976415564315,
"tokens_seen": 1423638528
},
{
"epoch": 0.43,
"learning_rate": 0.00028721163490471414,
"loss": 2.7481,
"theoretical_loss": 3.5323831605678215,
"tokens_seen": 1423704064
},
{
"epoch": 0.43,
"objective/train/docs_used": 806909,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.23956036567688,
"objective/train/theoretical_loss": 3.5323686804324215,
"objective/train/tokens_used": 1444229600,
"theoretical_loss": 3.5323686804324215,
"tokens_seen": 1423769600
},
{
"epoch": 0.43,
"learning_rate": 0.0002872016048144434,
"loss": 2.5995,
"theoretical_loss": 3.5323686804324215,
"tokens_seen": 1423769600
},
{
"epoch": 0.43,
"learning_rate": 0.0002871915747241725,
"loss": 2.5933,
"theoretical_loss": 3.5323542011501408,
"tokens_seen": 1423835136
},
{
"epoch": 0.43,
"learning_rate": 0.00028718154463390174,
"loss": 2.7081,
"theoretical_loss": 3.53233972272089,
"tokens_seen": 1423900672
},
{
"epoch": 0.43,
"learning_rate": 0.00028717151454363087,
"loss": 2.7765,
"theoretical_loss": 3.5323252451445804,
"tokens_seen": 1423966208
},
{
"epoch": 0.43,
"learning_rate": 0.0002871614844533601,
"loss": 2.7327,
"theoretical_loss": 3.5323107684211217,
"tokens_seen": 1424031744
},
{
"epoch": 0.43,
"learning_rate": 0.0002871514543630893,
"loss": 2.5779,
"theoretical_loss": 3.5322962925504253,
"tokens_seen": 1424097280
},
{
"epoch": 0.43,
"learning_rate": 0.00028714142427281846,
"loss": 2.5486,
"theoretical_loss": 3.5322818175324007,
"tokens_seen": 1424162816
},
{
"epoch": 0.43,
"learning_rate": 0.00028713139418254765,
"loss": 2.7859,
"theoretical_loss": 3.532267343366959,
"tokens_seen": 1424228352
},
{
"epoch": 0.43,
"learning_rate": 0.0002871213640922768,
"loss": 2.7005,
"theoretical_loss": 3.532252870054011,
"tokens_seen": 1424293888
},
{
"epoch": 0.43,
"learning_rate": 0.000287111334002006,
"loss": 2.6481,
"theoretical_loss": 3.532238397593466,
"tokens_seen": 1424359424
},
{
"epoch": 0.43,
"learning_rate": 0.00028710130391173524,
"loss": 2.737,
"theoretical_loss": 3.532223925985237,
"tokens_seen": 1424424960
},
{
"epoch": 0.43,
"learning_rate": 0.00028709127382146437,
"loss": 2.7372,
"theoretical_loss": 3.532209455229232,
"tokens_seen": 1424490496
},
{
"epoch": 0.43,
"learning_rate": 0.0002870812437311936,
"loss": 2.9314,
"theoretical_loss": 3.532194985325363,
"tokens_seen": 1424556032
},
{
"epoch": 0.43,
"learning_rate": 0.00028707121364092273,
"loss": 2.5108,
"theoretical_loss": 3.5321805162735407,
"tokens_seen": 1424621568
},
{
"epoch": 0.43,
"learning_rate": 0.00028706118355065197,
"loss": 2.4583,
"theoretical_loss": 3.532166048073675,
"tokens_seen": 1424687104
},
{
"epoch": 0.43,
"learning_rate": 0.00028705115346038115,
"loss": 2.6545,
"theoretical_loss": 3.532151580725677,
"tokens_seen": 1424752640
},
{
"epoch": 0.43,
"learning_rate": 0.00028704112337011033,
"loss": 2.554,
"theoretical_loss": 3.5321371142294575,
"tokens_seen": 1424818176
},
{
"epoch": 0.43,
"learning_rate": 0.0002870310932798395,
"loss": 2.5647,
"theoretical_loss": 3.532122648584927,
"tokens_seen": 1424883712
},
{
"epoch": 0.43,
"learning_rate": 0.00028702106318956875,
"loss": 2.7175,
"theoretical_loss": 3.532108183791996,
"tokens_seen": 1424949248
},
{
"epoch": 0.43,
"learning_rate": 0.0002870110330992979,
"loss": 2.8437,
"theoretical_loss": 3.5320937198505753,
"tokens_seen": 1425014784
},
{
"epoch": 0.43,
"learning_rate": 0.0002870010030090271,
"loss": 2.8146,
"theoretical_loss": 3.532079256760576,
"tokens_seen": 1425080320
},
{
"epoch": 0.43,
"learning_rate": 0.00028699097291875624,
"loss": 2.6622,
"theoretical_loss": 3.532064794521908,
"tokens_seen": 1425145856
},
{
"epoch": 0.43,
"learning_rate": 0.00028698094282848547,
"loss": 2.6857,
"theoretical_loss": 3.5320503331344826,
"tokens_seen": 1425211392
},
{
"epoch": 0.43,
"learning_rate": 0.00028697091273821465,
"loss": 2.8247,
"theoretical_loss": 3.532035872598211,
"tokens_seen": 1425276928
},
{
"epoch": 0.43,
"learning_rate": 0.00028696088264794383,
"loss": 2.7583,
"theoretical_loss": 3.5320214129130036,
"tokens_seen": 1425342464
},
{
"epoch": 0.43,
"objective/train/docs_used": 808391,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5881307125091553,
"objective/train/theoretical_loss": 3.5320069540787706,
"objective/train/tokens_used": 1445868000,
"theoretical_loss": 3.5320069540787706,
"tokens_seen": 1425408000
},
{
"epoch": 0.43,
"learning_rate": 0.000286950852557673,
"loss": 2.4615,
"theoretical_loss": 3.5320069540787706,
"tokens_seen": 1425408000
},
{
"epoch": 0.43,
"learning_rate": 0.0002869408224674022,
"loss": 2.7377,
"theoretical_loss": 3.531992496095423,
"tokens_seen": 1425473536
},
{
"epoch": 0.43,
"learning_rate": 0.0002869307923771314,
"loss": 2.8657,
"theoretical_loss": 3.5319780389628725,
"tokens_seen": 1425539072
},
{
"epoch": 0.43,
"learning_rate": 0.0002869207622868606,
"loss": 2.7507,
"theoretical_loss": 3.531963582681029,
"tokens_seen": 1425604608
},
{
"epoch": 0.43,
"learning_rate": 0.00028691073219658974,
"loss": 2.8325,
"theoretical_loss": 3.531949127249804,
"tokens_seen": 1425670144
},
{
"epoch": 0.43,
"learning_rate": 0.000286900702106319,
"loss": 2.6006,
"theoretical_loss": 3.5319346726691077,
"tokens_seen": 1425735680
},
{
"epoch": 0.43,
"learning_rate": 0.0002868906720160481,
"loss": 2.6871,
"theoretical_loss": 3.5319202189388514,
"tokens_seen": 1425801216
},
{
"epoch": 0.43,
"learning_rate": 0.00028688064192577734,
"loss": 2.6515,
"theoretical_loss": 3.5319057660589452,
"tokens_seen": 1425866752
},
{
"epoch": 0.43,
"learning_rate": 0.0002868706118355065,
"loss": 2.771,
"theoretical_loss": 3.531891314029301,
"tokens_seen": 1425932288
},
{
"epoch": 0.43,
"learning_rate": 0.0002868605817452357,
"loss": 2.7467,
"theoretical_loss": 3.53187686284983,
"tokens_seen": 1425997824
},
{
"epoch": 0.43,
"learning_rate": 0.0002868505516549649,
"loss": 2.5652,
"theoretical_loss": 3.531862412520442,
"tokens_seen": 1426063360
},
{
"epoch": 0.43,
"learning_rate": 0.0002868405215646941,
"loss": 2.5719,
"theoretical_loss": 3.531847963041048,
"tokens_seen": 1426128896
},
{
"epoch": 0.43,
"learning_rate": 0.00028683049147442324,
"loss": 2.5668,
"theoretical_loss": 3.53183351441156,
"tokens_seen": 1426194432
},
{
"epoch": 0.43,
"learning_rate": 0.0002868204613841525,
"loss": 2.6897,
"theoretical_loss": 3.531819066631887,
"tokens_seen": 1426259968
},
{
"epoch": 0.43,
"learning_rate": 0.0002868104312938816,
"loss": 2.7118,
"theoretical_loss": 3.5318046197019424,
"tokens_seen": 1426325504
},
{
"epoch": 0.43,
"learning_rate": 0.00028680040120361084,
"loss": 2.5966,
"theoretical_loss": 3.531790173621636,
"tokens_seen": 1426391040
},
{
"epoch": 0.43,
"learning_rate": 0.00028679037111334,
"loss": 2.71,
"theoretical_loss": 3.531775728390879,
"tokens_seen": 1426456576
},
{
"epoch": 0.43,
"learning_rate": 0.0002867803410230692,
"loss": 2.5951,
"theoretical_loss": 3.5317612840095816,
"tokens_seen": 1426522112
},
{
"epoch": 0.43,
"learning_rate": 0.00028677031093279844,
"loss": 2.5853,
"theoretical_loss": 3.5317468404776555,
"tokens_seen": 1426587648
},
{
"epoch": 0.43,
"learning_rate": 0.00028676028084252756,
"loss": 2.7814,
"theoretical_loss": 3.531732397795012,
"tokens_seen": 1426653184
},
{
"epoch": 0.43,
"learning_rate": 0.0002867502507522568,
"loss": 2.6459,
"theoretical_loss": 3.531717955961562,
"tokens_seen": 1426718720
},
{
"epoch": 0.43,
"learning_rate": 0.000286740220661986,
"loss": 2.8278,
"theoretical_loss": 3.531703514977216,
"tokens_seen": 1426784256
},
{
"epoch": 0.43,
"learning_rate": 0.00028673019057171516,
"loss": 2.6427,
"theoretical_loss": 3.5316890748418857,
"tokens_seen": 1426849792
},
{
"epoch": 0.43,
"learning_rate": 0.00028672016048144434,
"loss": 2.6531,
"theoretical_loss": 3.531674635555482,
"tokens_seen": 1426915328
},
{
"epoch": 0.43,
"learning_rate": 0.0002867101303911736,
"loss": 2.7834,
"theoretical_loss": 3.5316601971179162,
"tokens_seen": 1426980864
},
{
"epoch": 0.43,
"objective/train/docs_used": 809012,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8582561016082764,
"objective/train/theoretical_loss": 3.531645759529099,
"objective/train/tokens_used": 1447506400,
"theoretical_loss": 3.531645759529099,
"tokens_seen": 1427046400
},
{
"epoch": 0.43,
"learning_rate": 0.0002867001003009027,
"loss": 2.7791,
"theoretical_loss": 3.531645759529099,
"tokens_seen": 1427046400
},
{
"epoch": 0.43,
"learning_rate": 0.00028669007021063194,
"loss": 2.7345,
"theoretical_loss": 3.5316313227889413,
"tokens_seen": 1427111936
},
{
"epoch": 0.43,
"learning_rate": 0.00028668004012036107,
"loss": 2.5567,
"theoretical_loss": 3.5316168868973556,
"tokens_seen": 1427177472
},
{
"epoch": 0.43,
"learning_rate": 0.0002866700100300903,
"loss": 2.7308,
"theoretical_loss": 3.5316024518542517,
"tokens_seen": 1427243008
},
{
"epoch": 0.43,
"learning_rate": 0.0002866599799398195,
"loss": 2.6548,
"theoretical_loss": 3.5315880176595416,
"tokens_seen": 1427308544
},
{
"epoch": 0.43,
"learning_rate": 0.00028664994984954866,
"loss": 2.5018,
"theoretical_loss": 3.5315735843131355,
"tokens_seen": 1427374080
},
{
"epoch": 0.43,
"learning_rate": 0.00028663991975927785,
"loss": 2.8204,
"theoretical_loss": 3.5315591518149456,
"tokens_seen": 1427439616
},
{
"epoch": 0.43,
"learning_rate": 0.000286629889669007,
"loss": 2.7367,
"theoretical_loss": 3.5315447201648826,
"tokens_seen": 1427505152
},
{
"epoch": 0.43,
"learning_rate": 0.0002866198595787362,
"loss": 2.8061,
"theoretical_loss": 3.531530289362858,
"tokens_seen": 1427570688
},
{
"epoch": 0.43,
"learning_rate": 0.00028660982948846544,
"loss": 2.4898,
"theoretical_loss": 3.531515859408783,
"tokens_seen": 1427636224
},
{
"epoch": 0.43,
"learning_rate": 0.00028659979939819457,
"loss": 2.5542,
"theoretical_loss": 3.5315014303025682,
"tokens_seen": 1427701760
},
{
"epoch": 0.43,
"learning_rate": 0.0002865897693079238,
"loss": 2.8644,
"theoretical_loss": 3.531487002044126,
"tokens_seen": 1427767296
},
{
"epoch": 0.43,
"learning_rate": 0.00028657973921765293,
"loss": 2.7728,
"theoretical_loss": 3.5314725746333666,
"tokens_seen": 1427832832
},
{
"epoch": 0.43,
"learning_rate": 0.00028656970912738217,
"loss": 2.5952,
"theoretical_loss": 3.531458148070202,
"tokens_seen": 1427898368
},
{
"epoch": 0.43,
"learning_rate": 0.00028655967903711135,
"loss": 2.6619,
"theoretical_loss": 3.5314437223545436,
"tokens_seen": 1427963904
},
{
"epoch": 0.43,
"learning_rate": 0.00028654964894684053,
"loss": 2.6874,
"theoretical_loss": 3.5314292974863024,
"tokens_seen": 1428029440
},
{
"epoch": 0.43,
"learning_rate": 0.0002865396188565697,
"loss": 2.8106,
"theoretical_loss": 3.5314148734653896,
"tokens_seen": 1428094976
},
{
"epoch": 0.43,
"learning_rate": 0.00028652958876629895,
"loss": 2.5036,
"theoretical_loss": 3.5314004502917165,
"tokens_seen": 1428160512
},
{
"epoch": 0.43,
"learning_rate": 0.0002865195586760281,
"loss": 2.5524,
"theoretical_loss": 3.531386027965195,
"tokens_seen": 1428226048
},
{
"epoch": 0.43,
"learning_rate": 0.0002865095285857573,
"loss": 2.4994,
"theoretical_loss": 3.5313716064857354,
"tokens_seen": 1428291584
},
{
"epoch": 0.43,
"learning_rate": 0.00028649949849548644,
"loss": 2.71,
"theoretical_loss": 3.531357185853251,
"tokens_seen": 1428357120
},
{
"epoch": 0.43,
"learning_rate": 0.00028648946840521567,
"loss": 2.6581,
"theoretical_loss": 3.5313427660676506,
"tokens_seen": 1428422656
},
{
"epoch": 0.43,
"learning_rate": 0.00028647943831494485,
"loss": 2.6594,
"theoretical_loss": 3.5313283471288477,
"tokens_seen": 1428488192
},
{
"epoch": 0.43,
"learning_rate": 0.00028646940822467403,
"loss": 2.5867,
"theoretical_loss": 3.5313139290367532,
"tokens_seen": 1428553728
},
{
"epoch": 0.43,
"learning_rate": 0.0002864593781344032,
"loss": 2.672,
"theoretical_loss": 3.5312995117912784,
"tokens_seen": 1428619264
},
{
"epoch": 0.43,
"objective/train/docs_used": 809683,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8703832626342773,
"objective/train/theoretical_loss": 3.531285095392335,
"objective/train/tokens_used": 1449144800,
"theoretical_loss": 3.531285095392335,
"tokens_seen": 1428684800
},
{
"epoch": 0.43,
"learning_rate": 0.0002864493480441324,
"loss": 2.621,
"theoretical_loss": 3.531285095392335,
"tokens_seen": 1428684800
},
{
"epoch": 0.43,
"learning_rate": 0.0002864393179538616,
"loss": 2.9404,
"theoretical_loss": 3.531270679839834,
"tokens_seen": 1428750336
},
{
"epoch": 0.43,
"learning_rate": 0.0002864292878635908,
"loss": 2.4935,
"theoretical_loss": 3.5312562651336865,
"tokens_seen": 1428815872
},
{
"epoch": 0.43,
"learning_rate": 0.00028641925777331994,
"loss": 2.6288,
"theoretical_loss": 3.5312418512738053,
"tokens_seen": 1428881408
},
{
"epoch": 0.43,
"learning_rate": 0.0002864092276830492,
"loss": 2.8679,
"theoretical_loss": 3.5312274382601005,
"tokens_seen": 1428946944
},
{
"epoch": 0.43,
"learning_rate": 0.0002863991975927783,
"loss": 2.6234,
"theoretical_loss": 3.5312130260924848,
"tokens_seen": 1429012480
},
{
"epoch": 0.43,
"learning_rate": 0.00028638916750250754,
"loss": 2.6032,
"theoretical_loss": 3.5311986147708696,
"tokens_seen": 1429078016
},
{
"epoch": 0.43,
"learning_rate": 0.0002863791374122367,
"loss": 2.8766,
"theoretical_loss": 3.5311842042951653,
"tokens_seen": 1429143552
},
{
"epoch": 0.43,
"learning_rate": 0.0002863691073219659,
"loss": 2.6628,
"theoretical_loss": 3.531169794665285,
"tokens_seen": 1429209088
},
{
"epoch": 0.43,
"learning_rate": 0.0002863590772316951,
"loss": 2.8693,
"theoretical_loss": 3.531155385881139,
"tokens_seen": 1429274624
},
{
"epoch": 0.43,
"learning_rate": 0.0002863490471414243,
"loss": 2.584,
"theoretical_loss": 3.5311409779426395,
"tokens_seen": 1429340160
},
{
"epoch": 0.43,
"learning_rate": 0.00028633901705115344,
"loss": 2.5916,
"theoretical_loss": 3.531126570849698,
"tokens_seen": 1429405696
},
{
"epoch": 0.43,
"learning_rate": 0.0002863289869608827,
"loss": 2.5924,
"theoretical_loss": 3.5311121646022263,
"tokens_seen": 1429471232
},
{
"epoch": 0.43,
"learning_rate": 0.0002863189568706118,
"loss": 2.7785,
"theoretical_loss": 3.5310977592001356,
"tokens_seen": 1429536768
},
{
"epoch": 0.43,
"learning_rate": 0.00028630892678034104,
"loss": 2.5969,
"theoretical_loss": 3.531083354643338,
"tokens_seen": 1429602304
},
{
"epoch": 0.43,
"learning_rate": 0.0002862988966900702,
"loss": 2.5377,
"theoretical_loss": 3.5310689509317448,
"tokens_seen": 1429667840
},
{
"epoch": 0.43,
"learning_rate": 0.0002862888665997994,
"loss": 2.8017,
"theoretical_loss": 3.531054548065268,
"tokens_seen": 1429733376
},
{
"epoch": 0.43,
"learning_rate": 0.0002862788365095286,
"loss": 2.5184,
"theoretical_loss": 3.5310401460438188,
"tokens_seen": 1429798912
},
{
"epoch": 0.43,
"learning_rate": 0.00028626880641925776,
"loss": 2.8011,
"theoretical_loss": 3.5310257448673092,
"tokens_seen": 1429864448
},
{
"epoch": 0.43,
"learning_rate": 0.00028625877632898694,
"loss": 2.7425,
"theoretical_loss": 3.531011344535651,
"tokens_seen": 1429929984
},
{
"epoch": 0.43,
"learning_rate": 0.0002862487462387162,
"loss": 2.7907,
"theoretical_loss": 3.5309969450487557,
"tokens_seen": 1429995520
},
{
"epoch": 0.43,
"learning_rate": 0.0002862387161484453,
"loss": 2.6573,
"theoretical_loss": 3.5309825464065354,
"tokens_seen": 1430061056
},
{
"epoch": 0.43,
"learning_rate": 0.00028622868605817454,
"loss": 2.8119,
"theoretical_loss": 3.5309681486089013,
"tokens_seen": 1430126592
},
{
"epoch": 0.43,
"learning_rate": 0.00028621865596790367,
"loss": 2.8224,
"theoretical_loss": 3.5309537516557654,
"tokens_seen": 1430192128
},
{
"epoch": 0.43,
"learning_rate": 0.0002862086258776329,
"loss": 2.6365,
"theoretical_loss": 3.53093935554704,
"tokens_seen": 1430257664
},
{
"epoch": 0.43,
"objective/train/docs_used": 810937,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9391539096832275,
"objective/train/theoretical_loss": 3.530924960282636,
"objective/train/tokens_used": 1450783200,
"theoretical_loss": 3.530924960282636,
"tokens_seen": 1430323200
},
{
"epoch": 0.43,
"learning_rate": 0.0002861985957873621,
"loss": 2.8021,
"theoretical_loss": 3.530924960282636,
"tokens_seen": 1430323200
},
{
"epoch": 0.43,
"learning_rate": 0.00028618856569709127,
"loss": 2.5136,
"theoretical_loss": 3.5309105658624658,
"tokens_seen": 1430388736
},
{
"epoch": 0.43,
"learning_rate": 0.00028617853560682045,
"loss": 2.8727,
"theoretical_loss": 3.5308961722864405,
"tokens_seen": 1430454272
},
{
"epoch": 0.43,
"learning_rate": 0.0002861685055165497,
"loss": 2.6961,
"theoretical_loss": 3.530881779554473,
"tokens_seen": 1430519808
},
{
"epoch": 0.43,
"learning_rate": 0.0002861584754262788,
"loss": 2.7515,
"theoretical_loss": 3.5308673876664742,
"tokens_seen": 1430585344
},
{
"epoch": 0.43,
"learning_rate": 0.00028614844533600805,
"loss": 2.6365,
"theoretical_loss": 3.5308529966223565,
"tokens_seen": 1430650880
},
{
"epoch": 0.43,
"learning_rate": 0.00028613841524573717,
"loss": 2.8144,
"theoretical_loss": 3.5308386064220314,
"tokens_seen": 1430716416
},
{
"epoch": 0.43,
"learning_rate": 0.0002861283851554664,
"loss": 2.974,
"theoretical_loss": 3.530824217065411,
"tokens_seen": 1430781952
},
{
"epoch": 0.43,
"learning_rate": 0.0002861183550651956,
"loss": 2.7546,
"theoretical_loss": 3.530809828552407,
"tokens_seen": 1430847488
},
{
"epoch": 0.43,
"learning_rate": 0.00028610832497492477,
"loss": 2.8187,
"theoretical_loss": 3.530795440882932,
"tokens_seen": 1430913024
},
{
"epoch": 0.43,
"learning_rate": 0.00028609829488465395,
"loss": 2.8346,
"theoretical_loss": 3.530781054056897,
"tokens_seen": 1430978560
},
{
"epoch": 0.43,
"learning_rate": 0.00028608826479438313,
"loss": 2.8139,
"theoretical_loss": 3.5307666680742145,
"tokens_seen": 1431044096
},
{
"epoch": 0.43,
"learning_rate": 0.0002860782347041123,
"loss": 2.4533,
"theoretical_loss": 3.530752282934796,
"tokens_seen": 1431109632
},
{
"epoch": 0.43,
"learning_rate": 0.00028606820461384155,
"loss": 2.5665,
"theoretical_loss": 3.530737898638554,
"tokens_seen": 1431175168
},
{
"epoch": 0.43,
"learning_rate": 0.0002860581745235707,
"loss": 2.9199,
"theoretical_loss": 3.5307235151854,
"tokens_seen": 1431240704
},
{
"epoch": 0.43,
"learning_rate": 0.0002860481444332999,
"loss": 2.9835,
"theoretical_loss": 3.5307091325752458,
"tokens_seen": 1431306240
},
{
"epoch": 0.43,
"learning_rate": 0.0002860381143430291,
"loss": 2.6532,
"theoretical_loss": 3.5306947508080047,
"tokens_seen": 1431371776
},
{
"epoch": 0.43,
"learning_rate": 0.0002860280842527583,
"loss": 2.5603,
"theoretical_loss": 3.530680369883587,
"tokens_seen": 1431437312
},
{
"epoch": 0.43,
"learning_rate": 0.0002860180541624875,
"loss": 2.8273,
"theoretical_loss": 3.5306659898019053,
"tokens_seen": 1431502848
},
{
"epoch": 0.43,
"learning_rate": 0.00028600802407221664,
"loss": 2.9164,
"theoretical_loss": 3.530651610562872,
"tokens_seen": 1431568384
},
{
"epoch": 0.43,
"learning_rate": 0.00028599799398194587,
"loss": 2.6954,
"theoretical_loss": 3.5306372321663995,
"tokens_seen": 1431633920
},
{
"epoch": 0.43,
"learning_rate": 0.00028598796389167505,
"loss": 2.5353,
"theoretical_loss": 3.530622854612399,
"tokens_seen": 1431699456
},
{
"epoch": 0.43,
"learning_rate": 0.00028597793380140423,
"loss": 2.6037,
"theoretical_loss": 3.530608477900783,
"tokens_seen": 1431764992
},
{
"epoch": 0.43,
"learning_rate": 0.0002859679037111334,
"loss": 2.5032,
"theoretical_loss": 3.5305941020314635,
"tokens_seen": 1431830528
},
{
"epoch": 0.43,
"learning_rate": 0.0002859578736208626,
"loss": 2.5836,
"theoretical_loss": 3.5305797270043526,
"tokens_seen": 1431896064
},
{
"epoch": 0.43,
"objective/train/docs_used": 811574,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6807103157043457,
"objective/train/theoretical_loss": 3.5305653528193623,
"objective/train/tokens_used": 1452421600,
"theoretical_loss": 3.5305653528193623,
"tokens_seen": 1431961600
},
{
"epoch": 0.43,
"learning_rate": 0.0002859478435305918,
"loss": 2.6973,
"theoretical_loss": 3.5305653528193623,
"tokens_seen": 1431961600
},
{
"epoch": 0.43,
"learning_rate": 0.000285937813440321,
"loss": 2.5562,
"theoretical_loss": 3.530550979476405,
"tokens_seen": 1432027136
},
{
"epoch": 0.43,
"learning_rate": 0.00028592778335005014,
"loss": 2.6549,
"theoretical_loss": 3.5305366069753927,
"tokens_seen": 1432092672
},
{
"epoch": 0.43,
"learning_rate": 0.0002859177532597794,
"loss": 2.7586,
"theoretical_loss": 3.530522235316237,
"tokens_seen": 1432158208
},
{
"epoch": 0.43,
"learning_rate": 0.0002859077231695085,
"loss": 2.5824,
"theoretical_loss": 3.5305078644988512,
"tokens_seen": 1432223744
},
{
"epoch": 0.43,
"learning_rate": 0.00028589769307923774,
"loss": 2.7833,
"theoretical_loss": 3.530493494523147,
"tokens_seen": 1432289280
},
{
"epoch": 0.43,
"learning_rate": 0.0002858876629889669,
"loss": 2.8422,
"theoretical_loss": 3.5304791253890366,
"tokens_seen": 1432354816
},
{
"epoch": 0.43,
"learning_rate": 0.0002858776328986961,
"loss": 2.7295,
"theoretical_loss": 3.530464757096432,
"tokens_seen": 1432420352
},
{
"epoch": 0.43,
"learning_rate": 0.0002858676028084253,
"loss": 2.6684,
"theoretical_loss": 3.530450389645245,
"tokens_seen": 1432485888
},
{
"epoch": 0.43,
"learning_rate": 0.0002858575727181545,
"loss": 2.7324,
"theoretical_loss": 3.5304360230353886,
"tokens_seen": 1432551424
},
{
"epoch": 0.43,
"learning_rate": 0.00028584754262788364,
"loss": 2.6677,
"theoretical_loss": 3.530421657266775,
"tokens_seen": 1432616960
},
{
"epoch": 0.43,
"learning_rate": 0.0002858375125376129,
"loss": 2.382,
"theoretical_loss": 3.5304072923393166,
"tokens_seen": 1432682496
},
{
"epoch": 0.43,
"learning_rate": 0.000285827482447342,
"loss": 2.7269,
"theoretical_loss": 3.530392928252925,
"tokens_seen": 1432748032
},
{
"epoch": 0.43,
"learning_rate": 0.00028581745235707124,
"loss": 2.8551,
"theoretical_loss": 3.5303785650075126,
"tokens_seen": 1432813568
},
{
"epoch": 0.43,
"learning_rate": 0.0002858074222668004,
"loss": 2.5926,
"theoretical_loss": 3.5303642026029918,
"tokens_seen": 1432879104
},
{
"epoch": 0.43,
"learning_rate": 0.0002857973921765296,
"loss": 2.5241,
"theoretical_loss": 3.5303498410392757,
"tokens_seen": 1432944640
},
{
"epoch": 0.43,
"learning_rate": 0.0002857873620862588,
"loss": 2.6993,
"theoretical_loss": 3.5303354803162756,
"tokens_seen": 1433010176
},
{
"epoch": 0.43,
"learning_rate": 0.00028577733199598796,
"loss": 2.8344,
"theoretical_loss": 3.5303211204339044,
"tokens_seen": 1433075712
},
{
"epoch": 0.43,
"learning_rate": 0.00028576730190571714,
"loss": 2.4379,
"theoretical_loss": 3.530306761392074,
"tokens_seen": 1433141248
},
{
"epoch": 0.43,
"learning_rate": 0.0002857572718154464,
"loss": 2.4978,
"theoretical_loss": 3.530292403190697,
"tokens_seen": 1433206784
},
{
"epoch": 0.43,
"learning_rate": 0.0002857472417251755,
"loss": 2.5881,
"theoretical_loss": 3.530278045829686,
"tokens_seen": 1433272320
},
{
"epoch": 0.43,
"learning_rate": 0.00028573721163490474,
"loss": 2.6333,
"theoretical_loss": 3.530263689308953,
"tokens_seen": 1433337856
},
{
"epoch": 0.43,
"learning_rate": 0.00028572718154463387,
"loss": 2.5297,
"theoretical_loss": 3.5302493336284106,
"tokens_seen": 1433403392
},
{
"epoch": 0.43,
"learning_rate": 0.0002857171514543631,
"loss": 2.608,
"theoretical_loss": 3.5302349787879717,
"tokens_seen": 1433468928
},
{
"epoch": 0.43,
"learning_rate": 0.0002857071213640923,
"loss": 2.6404,
"theoretical_loss": 3.530220624787548,
"tokens_seen": 1433534464
},
{
"epoch": 0.43,
"objective/train/docs_used": 812859,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9161672592163086,
"objective/train/theoretical_loss": 3.530206271627052,
"objective/train/tokens_used": 1454060000,
"theoretical_loss": 3.530206271627052,
"tokens_seen": 1433600000
},
{
"epoch": 0.43,
"learning_rate": 0.00028569709127382147,
"loss": 2.823,
"theoretical_loss": 3.530206271627052,
"tokens_seen": 1433600000
},
{
"epoch": 0.43,
"learning_rate": 0.00028568706118355065,
"loss": 2.7349,
"theoretical_loss": 3.5301919193063966,
"tokens_seen": 1433665536
},
{
"epoch": 0.43,
"learning_rate": 0.0002856770310932799,
"loss": 2.5872,
"theoretical_loss": 3.5301775678254943,
"tokens_seen": 1433731072
},
{
"epoch": 0.43,
"learning_rate": 0.000285667001003009,
"loss": 2.7491,
"theoretical_loss": 3.5301632171842567,
"tokens_seen": 1433796608
},
{
"epoch": 0.43,
"learning_rate": 0.00028565697091273825,
"loss": 2.7111,
"theoretical_loss": 3.5301488673825974,
"tokens_seen": 1433862144
},
{
"epoch": 0.43,
"learning_rate": 0.00028564694082246737,
"loss": 2.6845,
"theoretical_loss": 3.5301345184204282,
"tokens_seen": 1433927680
},
{
"epoch": 0.43,
"learning_rate": 0.0002856369107321966,
"loss": 2.5685,
"theoretical_loss": 3.5301201702976615,
"tokens_seen": 1433993216
},
{
"epoch": 0.43,
"learning_rate": 0.0002856268806419258,
"loss": 2.5816,
"theoretical_loss": 3.530105823014211,
"tokens_seen": 1434058752
},
{
"epoch": 0.43,
"learning_rate": 0.00028561685055165497,
"loss": 2.7417,
"theoretical_loss": 3.5300914765699876,
"tokens_seen": 1434124288
},
{
"epoch": 0.43,
"learning_rate": 0.00028560682046138415,
"loss": 2.8346,
"theoretical_loss": 3.5300771309649055,
"tokens_seen": 1434189824
},
{
"epoch": 0.43,
"learning_rate": 0.00028559679037111333,
"loss": 2.8161,
"theoretical_loss": 3.530062786198876,
"tokens_seen": 1434255360
},
{
"epoch": 0.43,
"learning_rate": 0.0002855867602808425,
"loss": 2.6281,
"theoretical_loss": 3.530048442271812,
"tokens_seen": 1434320896
},
{
"epoch": 0.43,
"learning_rate": 0.00028557673019057175,
"loss": 2.7378,
"theoretical_loss": 3.5300340991836268,
"tokens_seen": 1434386432
},
{
"epoch": 0.43,
"learning_rate": 0.0002855667001003009,
"loss": 2.6082,
"theoretical_loss": 3.530019756934232,
"tokens_seen": 1434451968
},
{
"epoch": 0.43,
"learning_rate": 0.0002855566700100301,
"loss": 2.623,
"theoretical_loss": 3.530005415523541,
"tokens_seen": 1434517504
},
{
"epoch": 0.43,
"learning_rate": 0.0002855466399197593,
"loss": 2.6378,
"theoretical_loss": 3.5299910749514662,
"tokens_seen": 1434583040
},
{
"epoch": 0.43,
"learning_rate": 0.0002855366098294885,
"loss": 2.8175,
"theoretical_loss": 3.52997673521792,
"tokens_seen": 1434648576
},
{
"epoch": 0.43,
"learning_rate": 0.00028552657973921765,
"loss": 2.4976,
"theoretical_loss": 3.529962396322815,
"tokens_seen": 1434714112
},
{
"epoch": 0.43,
"learning_rate": 0.00028551654964894684,
"loss": 2.5078,
"theoretical_loss": 3.529948058266065,
"tokens_seen": 1434779648
},
{
"epoch": 0.43,
"learning_rate": 0.000285506519558676,
"loss": 2.8611,
"theoretical_loss": 3.529933721047581,
"tokens_seen": 1434845184
},
{
"epoch": 0.43,
"learning_rate": 0.00028549648946840525,
"loss": 2.8136,
"theoretical_loss": 3.529919384667277,
"tokens_seen": 1434910720
},
{
"epoch": 0.43,
"learning_rate": 0.0002854864593781344,
"loss": 2.5679,
"theoretical_loss": 3.529905049125065,
"tokens_seen": 1434976256
},
{
"epoch": 0.43,
"learning_rate": 0.0002854764292878636,
"loss": 2.7412,
"theoretical_loss": 3.5298907144208584,
"tokens_seen": 1435041792
},
{
"epoch": 0.43,
"learning_rate": 0.00028546639919759274,
"loss": 2.3775,
"theoretical_loss": 3.529876380554569,
"tokens_seen": 1435107328
},
{
"epoch": 0.43,
"learning_rate": 0.000285456369107322,
"loss": 2.6197,
"theoretical_loss": 3.5298620475261107,
"tokens_seen": 1435172864
},
{
"epoch": 0.43,
"objective/train/docs_used": 813625,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7324283123016357,
"objective/train/theoretical_loss": 3.5298477153353955,
"objective/train/tokens_used": 1455698400,
"theoretical_loss": 3.5298477153353955,
"tokens_seen": 1435238400
},
{
"epoch": 0.43,
"learning_rate": 0.00028544633901705116,
"loss": 2.7406,
"theoretical_loss": 3.5298477153353955,
"tokens_seen": 1435238400
},
{
"epoch": 0.43,
"learning_rate": 0.00028543630892678034,
"loss": 2.5409,
"theoretical_loss": 3.5298333839823366,
"tokens_seen": 1435303936
},
{
"epoch": 0.43,
"learning_rate": 0.0002854262788365095,
"loss": 2.7674,
"theoretical_loss": 3.529819053466846,
"tokens_seen": 1435369472
},
{
"epoch": 0.43,
"learning_rate": 0.0002854162487462387,
"loss": 2.841,
"theoretical_loss": 3.5298047237888373,
"tokens_seen": 1435435008
},
{
"epoch": 0.44,
"learning_rate": 0.0002854062186559679,
"loss": 2.6596,
"theoretical_loss": 3.529790394948223,
"tokens_seen": 1435500544
},
{
"epoch": 0.44,
"learning_rate": 0.0002853961885656971,
"loss": 2.6146,
"theoretical_loss": 3.5297760669449163,
"tokens_seen": 1435566080
},
{
"epoch": 0.44,
"learning_rate": 0.00028538615847542624,
"loss": 2.7398,
"theoretical_loss": 3.52976173977883,
"tokens_seen": 1435631616
},
{
"epoch": 0.44,
"learning_rate": 0.0002853761283851555,
"loss": 2.535,
"theoretical_loss": 3.5297474134498765,
"tokens_seen": 1435697152
},
{
"epoch": 0.44,
"learning_rate": 0.00028536609829488466,
"loss": 2.4843,
"theoretical_loss": 3.529733087957969,
"tokens_seen": 1435762688
},
{
"epoch": 0.44,
"learning_rate": 0.00028535606820461384,
"loss": 2.5025,
"theoretical_loss": 3.5297187633030203,
"tokens_seen": 1435828224
},
{
"epoch": 0.44,
"learning_rate": 0.000285346038114343,
"loss": 2.8935,
"theoretical_loss": 3.5297044394849433,
"tokens_seen": 1435893760
},
{
"epoch": 0.44,
"learning_rate": 0.0002853360080240722,
"loss": 2.9017,
"theoretical_loss": 3.5296901165036507,
"tokens_seen": 1435959296
},
{
"epoch": 0.44,
"learning_rate": 0.0002853259779338014,
"loss": 2.5574,
"theoretical_loss": 3.529675794359056,
"tokens_seen": 1436024832
},
{
"epoch": 0.44,
"learning_rate": 0.0002853159478435306,
"loss": 2.6979,
"theoretical_loss": 3.529661473051072,
"tokens_seen": 1436090368
},
{
"epoch": 0.44,
"learning_rate": 0.00028530591775325975,
"loss": 2.7863,
"theoretical_loss": 3.529647152579611,
"tokens_seen": 1436155904
},
{
"epoch": 0.44,
"learning_rate": 0.000285295887662989,
"loss": 2.3401,
"theoretical_loss": 3.5296328329445865,
"tokens_seen": 1436221440
},
{
"epoch": 0.44,
"learning_rate": 0.0002852858575727181,
"loss": 2.662,
"theoretical_loss": 3.5296185141459118,
"tokens_seen": 1436286976
},
{
"epoch": 0.44,
"learning_rate": 0.00028527582748244735,
"loss": 2.7664,
"theoretical_loss": 3.529604196183499,
"tokens_seen": 1436352512
},
{
"epoch": 0.44,
"learning_rate": 0.0002852657973921766,
"loss": 2.6214,
"theoretical_loss": 3.529589879057262,
"tokens_seen": 1436418048
},
{
"epoch": 0.44,
"learning_rate": 0.0002852557673019057,
"loss": 2.6936,
"theoretical_loss": 3.5295755627671133,
"tokens_seen": 1436483584
},
{
"epoch": 0.44,
"learning_rate": 0.00028524573721163494,
"loss": 2.5677,
"theoretical_loss": 3.5295612473129663,
"tokens_seen": 1436549120
},
{
"epoch": 0.44,
"learning_rate": 0.00028523570712136407,
"loss": 2.5519,
"theoretical_loss": 3.5295469326947337,
"tokens_seen": 1436614656
},
{
"epoch": 0.44,
"learning_rate": 0.0002852256770310933,
"loss": 2.5751,
"theoretical_loss": 3.529532618912328,
"tokens_seen": 1436680192
},
{
"epoch": 0.44,
"learning_rate": 0.0002852156469408225,
"loss": 2.6265,
"theoretical_loss": 3.529518305965664,
"tokens_seen": 1436745728
},
{
"epoch": 0.44,
"learning_rate": 0.00028520561685055167,
"loss": 2.711,
"theoretical_loss": 3.5295039938546537,
"tokens_seen": 1436811264
},
{
"epoch": 0.44,
"objective/train/docs_used": 814744,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9006946086883545,
"objective/train/theoretical_loss": 3.52948968257921,
"objective/train/tokens_used": 1457336800,
"theoretical_loss": 3.52948968257921,
"tokens_seen": 1436876800
},
{
"epoch": 0.44,
"learning_rate": 0.00028519558676028085,
"loss": 2.7718,
"theoretical_loss": 3.52948968257921,
"tokens_seen": 1436876800
},
{
"epoch": 0.44,
"learning_rate": 0.0002851855566700101,
"loss": 2.6577,
"theoretical_loss": 3.5294753721392453,
"tokens_seen": 1436942336
},
{
"epoch": 0.44,
"learning_rate": 0.0002851755265797392,
"loss": 2.678,
"theoretical_loss": 3.5294610625346747,
"tokens_seen": 1437007872
},
{
"epoch": 0.44,
"learning_rate": 0.00028516549648946845,
"loss": 2.8287,
"theoretical_loss": 3.52944675376541,
"tokens_seen": 1437073408
},
{
"epoch": 0.44,
"learning_rate": 0.00028515546639919757,
"loss": 2.7042,
"theoretical_loss": 3.5294324458313646,
"tokens_seen": 1437138944
},
{
"epoch": 0.44,
"learning_rate": 0.0002851454363089268,
"loss": 2.5135,
"theoretical_loss": 3.529418138732452,
"tokens_seen": 1437204480
},
{
"epoch": 0.44,
"learning_rate": 0.000285135406218656,
"loss": 2.7003,
"theoretical_loss": 3.529403832468585,
"tokens_seen": 1437270016
},
{
"epoch": 0.44,
"learning_rate": 0.00028512537612838517,
"loss": 2.7305,
"theoretical_loss": 3.529389527039677,
"tokens_seen": 1437335552
},
{
"epoch": 0.44,
"learning_rate": 0.00028511534603811435,
"loss": 2.5053,
"theoretical_loss": 3.529375222445641,
"tokens_seen": 1437401088
},
{
"epoch": 0.44,
"learning_rate": 0.00028510531594784353,
"loss": 2.6113,
"theoretical_loss": 3.5293609186863906,
"tokens_seen": 1437466624
},
{
"epoch": 0.44,
"learning_rate": 0.0002850952858575727,
"loss": 2.7061,
"theoretical_loss": 3.5293466157618383,
"tokens_seen": 1437532160
},
{
"epoch": 0.44,
"learning_rate": 0.00028508525576730195,
"loss": 2.7495,
"theoretical_loss": 3.5293323136718975,
"tokens_seen": 1437597696
},
{
"epoch": 0.44,
"learning_rate": 0.0002850752256770311,
"loss": 2.5914,
"theoretical_loss": 3.529318012416482,
"tokens_seen": 1437663232
},
{
"epoch": 0.44,
"learning_rate": 0.0002850651955867603,
"loss": 2.7714,
"theoretical_loss": 3.529303711995505,
"tokens_seen": 1437728768
},
{
"epoch": 0.44,
"learning_rate": 0.0002850551654964895,
"loss": 2.9821,
"theoretical_loss": 3.52928941240888,
"tokens_seen": 1437794304
},
{
"epoch": 0.44,
"learning_rate": 0.0002850451354062187,
"loss": 2.7642,
"theoretical_loss": 3.529275113656519,
"tokens_seen": 1437859840
},
{
"epoch": 0.44,
"learning_rate": 0.00028503510531594785,
"loss": 2.6276,
"theoretical_loss": 3.5292608157383363,
"tokens_seen": 1437925376
},
{
"epoch": 0.44,
"learning_rate": 0.00028502507522567704,
"loss": 2.5966,
"theoretical_loss": 3.5292465186542454,
"tokens_seen": 1437990912
},
{
"epoch": 0.44,
"learning_rate": 0.0002850150451354062,
"loss": 2.5843,
"theoretical_loss": 3.529232222404159,
"tokens_seen": 1438056448
},
{
"epoch": 0.44,
"learning_rate": 0.00028500501504513545,
"loss": 2.5464,
"theoretical_loss": 3.5292179269879913,
"tokens_seen": 1438121984
},
{
"epoch": 0.44,
"learning_rate": 0.0002849949849548646,
"loss": 2.7872,
"theoretical_loss": 3.529203632405655,
"tokens_seen": 1438187520
},
{
"epoch": 0.44,
"learning_rate": 0.0002849849548645938,
"loss": 2.8913,
"theoretical_loss": 3.529189338657063,
"tokens_seen": 1438253056
},
{
"epoch": 0.44,
"learning_rate": 0.00028497492477432294,
"loss": 2.7219,
"theoretical_loss": 3.5291750457421296,
"tokens_seen": 1438318592
},
{
"epoch": 0.44,
"learning_rate": 0.0002849648946840522,
"loss": 2.5679,
"theoretical_loss": 3.5291607536607685,
"tokens_seen": 1438384128
},
{
"epoch": 0.44,
"learning_rate": 0.00028495486459378136,
"loss": 2.5677,
"theoretical_loss": 3.529146462412892,
"tokens_seen": 1438449664
},
{
"epoch": 0.44,
"objective/train/docs_used": 815309,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.642245054244995,
"objective/train/theoretical_loss": 3.5291321719984134,
"objective/train/tokens_used": 1458975200,
"theoretical_loss": 3.5291321719984134,
"tokens_seen": 1438515200
},
{
"epoch": 0.44,
"learning_rate": 0.00028494483450351054,
"loss": 2.6636,
"theoretical_loss": 3.5291321719984134,
"tokens_seen": 1438515200
},
{
"epoch": 0.44,
"learning_rate": 0.0002849348044132397,
"loss": 2.7145,
"theoretical_loss": 3.5291178824172476,
"tokens_seen": 1438580736
},
{
"epoch": 0.44,
"learning_rate": 0.0002849247743229689,
"loss": 2.7064,
"theoretical_loss": 3.5291035936693067,
"tokens_seen": 1438646272
},
{
"epoch": 0.44,
"learning_rate": 0.0002849147442326981,
"loss": 2.9105,
"theoretical_loss": 3.529089305754505,
"tokens_seen": 1438711808
},
{
"epoch": 0.44,
"learning_rate": 0.0002849047141424273,
"loss": 2.8239,
"theoretical_loss": 3.5290750186727555,
"tokens_seen": 1438777344
},
{
"epoch": 0.44,
"learning_rate": 0.00028489468405215644,
"loss": 2.5981,
"theoretical_loss": 3.529060732423972,
"tokens_seen": 1438842880
},
{
"epoch": 0.44,
"learning_rate": 0.0002848846539618857,
"loss": 2.6253,
"theoretical_loss": 3.5290464470080676,
"tokens_seen": 1438908416
},
{
"epoch": 0.44,
"learning_rate": 0.00028487462387161486,
"loss": 2.6984,
"theoretical_loss": 3.529032162424956,
"tokens_seen": 1438973952
},
{
"epoch": 0.44,
"learning_rate": 0.00028486459378134404,
"loss": 2.6967,
"theoretical_loss": 3.529017878674551,
"tokens_seen": 1439039488
},
{
"epoch": 0.44,
"learning_rate": 0.0002848545636910732,
"loss": 2.7384,
"theoretical_loss": 3.5290035957567656,
"tokens_seen": 1439105024
},
{
"epoch": 0.44,
"learning_rate": 0.0002848445336008024,
"loss": 2.5431,
"theoretical_loss": 3.528989313671514,
"tokens_seen": 1439170560
},
{
"epoch": 0.44,
"learning_rate": 0.0002848345035105316,
"loss": 2.4017,
"theoretical_loss": 3.5289750324187086,
"tokens_seen": 1439236096
},
{
"epoch": 0.44,
"learning_rate": 0.0002848244734202608,
"loss": 2.4889,
"theoretical_loss": 3.5289607519982646,
"tokens_seen": 1439301632
},
{
"epoch": 0.44,
"learning_rate": 0.00028481444332998995,
"loss": 2.7757,
"theoretical_loss": 3.5289464724100945,
"tokens_seen": 1439367168
},
{
"epoch": 0.44,
"learning_rate": 0.0002848044132397192,
"loss": 2.6878,
"theoretical_loss": 3.5289321936541125,
"tokens_seen": 1439432704
},
{
"epoch": 0.44,
"learning_rate": 0.0002847943831494483,
"loss": 2.7125,
"theoretical_loss": 3.5289179157302315,
"tokens_seen": 1439498240
},
{
"epoch": 0.44,
"learning_rate": 0.00028478435305917755,
"loss": 2.6191,
"theoretical_loss": 3.528903638638366,
"tokens_seen": 1439563776
},
{
"epoch": 0.44,
"learning_rate": 0.0002847743229689067,
"loss": 2.7126,
"theoretical_loss": 3.5288893623784285,
"tokens_seen": 1439629312
},
{
"epoch": 0.44,
"learning_rate": 0.0002847642928786359,
"loss": 2.4799,
"theoretical_loss": 3.5288750869503334,
"tokens_seen": 1439694848
},
{
"epoch": 0.44,
"learning_rate": 0.0002847542627883651,
"loss": 2.4438,
"theoretical_loss": 3.5288608123539946,
"tokens_seen": 1439760384
},
{
"epoch": 0.44,
"learning_rate": 0.00028474423269809427,
"loss": 2.622,
"theoretical_loss": 3.5288465385893257,
"tokens_seen": 1439825920
},
{
"epoch": 0.44,
"learning_rate": 0.00028473420260782345,
"loss": 2.6346,
"theoretical_loss": 3.5288322656562396,
"tokens_seen": 1439891456
},
{
"epoch": 0.44,
"learning_rate": 0.0002847241725175527,
"loss": 2.5995,
"theoretical_loss": 3.528817993554651,
"tokens_seen": 1439956992
},
{
"epoch": 0.44,
"learning_rate": 0.0002847141424272818,
"loss": 2.8446,
"theoretical_loss": 3.528803722284473,
"tokens_seen": 1440022528
},
{
"epoch": 0.44,
"learning_rate": 0.00028470411233701105,
"loss": 2.7215,
"theoretical_loss": 3.5287894518456193,
"tokens_seen": 1440088064
},
{
"epoch": 0.44,
"objective/train/docs_used": 816515,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7364134788513184,
"objective/train/theoretical_loss": 3.5287751822380047,
"objective/train/tokens_used": 1460613600,
"theoretical_loss": 3.5287751822380047,
"tokens_seen": 1440153600
},
{
"epoch": 0.44,
"learning_rate": 0.00028469408224674023,
"loss": 2.7771,
"theoretical_loss": 3.5287751822380047,
"tokens_seen": 1440153600
},
{
"epoch": 0.44,
"learning_rate": 0.0002846840521564694,
"loss": 2.5996,
"theoretical_loss": 3.528760913461541,
"tokens_seen": 1440219136
},
{
"epoch": 0.44,
"learning_rate": 0.0002846740220661986,
"loss": 2.8071,
"theoretical_loss": 3.528746645516144,
"tokens_seen": 1440284672
},
{
"epoch": 0.44,
"learning_rate": 0.0002846639919759278,
"loss": 2.7998,
"theoretical_loss": 3.5287323784017266,
"tokens_seen": 1440350208
},
{
"epoch": 0.44,
"learning_rate": 0.00028465396188565695,
"loss": 2.6073,
"theoretical_loss": 3.528718112118202,
"tokens_seen": 1440415744
},
{
"epoch": 0.44,
"learning_rate": 0.0002846439317953862,
"loss": 2.475,
"theoretical_loss": 3.5287038466654845,
"tokens_seen": 1440481280
},
{
"epoch": 0.44,
"learning_rate": 0.0002846339017051153,
"loss": 2.5914,
"theoretical_loss": 3.5286895820434885,
"tokens_seen": 1440546816
},
{
"epoch": 0.44,
"learning_rate": 0.00028462387161484455,
"loss": 2.4916,
"theoretical_loss": 3.5286753182521275,
"tokens_seen": 1440612352
},
{
"epoch": 0.44,
"learning_rate": 0.0002846138415245737,
"loss": 2.6168,
"theoretical_loss": 3.5286610552913147,
"tokens_seen": 1440677888
},
{
"epoch": 0.44,
"learning_rate": 0.0002846038114343029,
"loss": 2.4972,
"theoretical_loss": 3.5286467931609646,
"tokens_seen": 1440743424
},
{
"epoch": 0.44,
"learning_rate": 0.0002845937813440321,
"loss": 2.4997,
"theoretical_loss": 3.528632531860991,
"tokens_seen": 1440808960
},
{
"epoch": 0.44,
"learning_rate": 0.0002845837512537613,
"loss": 2.7004,
"theoretical_loss": 3.5286182713913075,
"tokens_seen": 1440874496
},
{
"epoch": 0.44,
"learning_rate": 0.00028457372116349046,
"loss": 2.5901,
"theoretical_loss": 3.5286040117518285,
"tokens_seen": 1440940032
},
{
"epoch": 0.44,
"learning_rate": 0.0002845636910732197,
"loss": 2.6852,
"theoretical_loss": 3.5285897529424677,
"tokens_seen": 1441005568
},
{
"epoch": 0.44,
"learning_rate": 0.0002845536609829488,
"loss": 2.9051,
"theoretical_loss": 3.5285754949631385,
"tokens_seen": 1441071104
},
{
"epoch": 0.44,
"learning_rate": 0.00028454363089267805,
"loss": 2.5472,
"theoretical_loss": 3.5285612378137556,
"tokens_seen": 1441136640
},
{
"epoch": 0.44,
"learning_rate": 0.00028453360080240724,
"loss": 2.4939,
"theoretical_loss": 3.528546981494233,
"tokens_seen": 1441202176
},
{
"epoch": 0.44,
"learning_rate": 0.0002845235707121364,
"loss": 2.8118,
"theoretical_loss": 3.5285327260044834,
"tokens_seen": 1441267712
},
{
"epoch": 0.44,
"learning_rate": 0.00028451354062186565,
"loss": 2.6424,
"theoretical_loss": 3.528518471344422,
"tokens_seen": 1441333248
},
{
"epoch": 0.44,
"learning_rate": 0.0002845035105315948,
"loss": 2.6612,
"theoretical_loss": 3.528504217513963,
"tokens_seen": 1441398784
},
{
"epoch": 0.44,
"learning_rate": 0.000284493480441324,
"loss": 2.7271,
"theoretical_loss": 3.528489964513019,
"tokens_seen": 1441464320
},
{
"epoch": 0.44,
"learning_rate": 0.00028448345035105314,
"loss": 2.7766,
"theoretical_loss": 3.528475712341506,
"tokens_seen": 1441529856
},
{
"epoch": 0.44,
"learning_rate": 0.0002844734202607824,
"loss": 2.7787,
"theoretical_loss": 3.528461460999336,
"tokens_seen": 1441595392
},
{
"epoch": 0.44,
"learning_rate": 0.00028446339017051156,
"loss": 2.6505,
"theoretical_loss": 3.5284472104864246,
"tokens_seen": 1441660928
},
{
"epoch": 0.44,
"learning_rate": 0.00028445336008024074,
"loss": 2.7905,
"theoretical_loss": 3.528432960802685,
"tokens_seen": 1441726464
},
{
"epoch": 0.44,
"objective/train/docs_used": 817059,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7759475708007812,
"objective/train/theoretical_loss": 3.5284187119480315,
"objective/train/tokens_used": 1462252000,
"theoretical_loss": 3.5284187119480315,
"tokens_seen": 1441792000
},
{
"epoch": 0.44,
"learning_rate": 0.0002844433299899699,
"loss": 2.7168,
"theoretical_loss": 3.5284187119480315,
"tokens_seen": 1441792000
},
{
"epoch": 0.44,
"learning_rate": 0.0002844332998996991,
"loss": 2.6099,
"theoretical_loss": 3.528404463922378,
"tokens_seen": 1441857536
},
{
"epoch": 0.44,
"learning_rate": 0.0002844232698094283,
"loss": 2.8601,
"theoretical_loss": 3.5283902167256382,
"tokens_seen": 1441923072
},
{
"epoch": 0.44,
"learning_rate": 0.0002844132397191575,
"loss": 2.4695,
"theoretical_loss": 3.528375970357728,
"tokens_seen": 1441988608
},
{
"epoch": 0.44,
"learning_rate": 0.00028440320962888664,
"loss": 2.5564,
"theoretical_loss": 3.5283617248185593,
"tokens_seen": 1442054144
},
{
"epoch": 0.44,
"learning_rate": 0.0002843931795386159,
"loss": 2.5468,
"theoretical_loss": 3.5283474801080477,
"tokens_seen": 1442119680
},
{
"epoch": 0.44,
"learning_rate": 0.00028438314944834506,
"loss": 2.8496,
"theoretical_loss": 3.528333236226106,
"tokens_seen": 1442185216
},
{
"epoch": 0.44,
"learning_rate": 0.00028437311935807424,
"loss": 2.5617,
"theoretical_loss": 3.5283189931726504,
"tokens_seen": 1442250752
},
{
"epoch": 0.44,
"learning_rate": 0.0002843630892678034,
"loss": 2.8286,
"theoretical_loss": 3.5283047509475933,
"tokens_seen": 1442316288
},
{
"epoch": 0.44,
"learning_rate": 0.0002843530591775326,
"loss": 2.9751,
"theoretical_loss": 3.528290509550849,
"tokens_seen": 1442381824
},
{
"epoch": 0.44,
"learning_rate": 0.0002843430290872618,
"loss": 2.681,
"theoretical_loss": 3.528276268982333,
"tokens_seen": 1442447360
},
{
"epoch": 0.44,
"learning_rate": 0.000284332998996991,
"loss": 2.5116,
"theoretical_loss": 3.528262029241958,
"tokens_seen": 1442512896
},
{
"epoch": 0.44,
"learning_rate": 0.00028432296890672015,
"loss": 2.6709,
"theoretical_loss": 3.5282477903296394,
"tokens_seen": 1442578432
},
{
"epoch": 0.44,
"learning_rate": 0.0002843129388164494,
"loss": 2.8372,
"theoretical_loss": 3.5282335522452906,
"tokens_seen": 1442643968
},
{
"epoch": 0.44,
"learning_rate": 0.0002843029087261785,
"loss": 2.6361,
"theoretical_loss": 3.528219314988826,
"tokens_seen": 1442709504
},
{
"epoch": 0.44,
"learning_rate": 0.00028429287863590775,
"loss": 2.8336,
"theoretical_loss": 3.52820507856016,
"tokens_seen": 1442775040
},
{
"epoch": 0.44,
"learning_rate": 0.0002842828485456369,
"loss": 3.0158,
"theoretical_loss": 3.528190842959207,
"tokens_seen": 1442840576
},
{
"epoch": 0.44,
"learning_rate": 0.0002842728184553661,
"loss": 2.6864,
"theoretical_loss": 3.5281766081858814,
"tokens_seen": 1442906112
},
{
"epoch": 0.44,
"learning_rate": 0.0002842627883650953,
"loss": 2.4544,
"theoretical_loss": 3.5281623742400967,
"tokens_seen": 1442971648
},
{
"epoch": 0.44,
"learning_rate": 0.00028425275827482447,
"loss": 2.7644,
"theoretical_loss": 3.5281481411217683,
"tokens_seen": 1443037184
},
{
"epoch": 0.44,
"learning_rate": 0.00028424272818455365,
"loss": 2.4086,
"theoretical_loss": 3.52813390883081,
"tokens_seen": 1443102720
},
{
"epoch": 0.44,
"learning_rate": 0.0002842326980942829,
"loss": 2.7891,
"theoretical_loss": 3.528119677367136,
"tokens_seen": 1443168256
},
{
"epoch": 0.44,
"learning_rate": 0.000284222668004012,
"loss": 2.6942,
"theoretical_loss": 3.5281054467306605,
"tokens_seen": 1443233792
},
{
"epoch": 0.44,
"learning_rate": 0.00028421263791374125,
"loss": 2.7092,
"theoretical_loss": 3.5280912169212986,
"tokens_seen": 1443299328
},
{
"epoch": 0.44,
"learning_rate": 0.00028420260782347043,
"loss": 2.8029,
"theoretical_loss": 3.528076987938964,
"tokens_seen": 1443364864
},
{
"epoch": 0.44,
"objective/train/docs_used": 818508,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.720475196838379,
"objective/train/theoretical_loss": 3.5280627597835705,
"objective/train/tokens_used": 1463890400,
"theoretical_loss": 3.5280627597835705,
"tokens_seen": 1443430400
},
{
"epoch": 0.44,
"learning_rate": 0.0002841925777331996,
"loss": 2.988,
"theoretical_loss": 3.5280627597835705,
"tokens_seen": 1443430400
},
{
"epoch": 0.44,
"learning_rate": 0.0002841825476429288,
"loss": 2.763,
"theoretical_loss": 3.528048532455034,
"tokens_seen": 1443495936
},
{
"epoch": 0.44,
"learning_rate": 0.000284172517552658,
"loss": 2.8729,
"theoretical_loss": 3.5280343059532684,
"tokens_seen": 1443561472
},
{
"epoch": 0.44,
"learning_rate": 0.00028416248746238715,
"loss": 2.6339,
"theoretical_loss": 3.528020080278188,
"tokens_seen": 1443627008
},
{
"epoch": 0.44,
"learning_rate": 0.0002841524573721164,
"loss": 2.7556,
"theoretical_loss": 3.528005855429707,
"tokens_seen": 1443692544
},
{
"epoch": 0.44,
"learning_rate": 0.0002841424272818455,
"loss": 2.8019,
"theoretical_loss": 3.5279916314077395,
"tokens_seen": 1443758080
},
{
"epoch": 0.44,
"learning_rate": 0.00028413239719157475,
"loss": 2.671,
"theoretical_loss": 3.527977408212201,
"tokens_seen": 1443823616
},
{
"epoch": 0.44,
"learning_rate": 0.0002841223671013039,
"loss": 2.8158,
"theoretical_loss": 3.527963185843005,
"tokens_seen": 1443889152
},
{
"epoch": 0.44,
"learning_rate": 0.0002841123370110331,
"loss": 2.7414,
"theoretical_loss": 3.5279489643000668,
"tokens_seen": 1443954688
},
{
"epoch": 0.44,
"learning_rate": 0.0002841023069207623,
"loss": 2.9441,
"theoretical_loss": 3.527934743583301,
"tokens_seen": 1444020224
},
{
"epoch": 0.44,
"learning_rate": 0.0002840922768304915,
"loss": 2.5164,
"theoretical_loss": 3.5279205236926208,
"tokens_seen": 1444085760
},
{
"epoch": 0.44,
"learning_rate": 0.00028408224674022066,
"loss": 2.7535,
"theoretical_loss": 3.5279063046279417,
"tokens_seen": 1444151296
},
{
"epoch": 0.44,
"learning_rate": 0.0002840722166499499,
"loss": 2.8436,
"theoretical_loss": 3.5278920863891785,
"tokens_seen": 1444216832
},
{
"epoch": 0.44,
"learning_rate": 0.000284062186559679,
"loss": 2.691,
"theoretical_loss": 3.527877868976245,
"tokens_seen": 1444282368
},
{
"epoch": 0.44,
"learning_rate": 0.00028405215646940825,
"loss": 2.7671,
"theoretical_loss": 3.5278636523890565,
"tokens_seen": 1444347904
},
{
"epoch": 0.44,
"learning_rate": 0.0002840421263791374,
"loss": 2.6465,
"theoretical_loss": 3.527849436627527,
"tokens_seen": 1444413440
},
{
"epoch": 0.44,
"learning_rate": 0.0002840320962888666,
"loss": 2.627,
"theoretical_loss": 3.5278352216915714,
"tokens_seen": 1444478976
},
{
"epoch": 0.44,
"learning_rate": 0.0002840220661985958,
"loss": 2.843,
"theoretical_loss": 3.527821007581104,
"tokens_seen": 1444544512
},
{
"epoch": 0.44,
"learning_rate": 0.000284012036108325,
"loss": 2.7805,
"theoretical_loss": 3.5278067942960396,
"tokens_seen": 1444610048
},
{
"epoch": 0.44,
"learning_rate": 0.00028400200601805416,
"loss": 2.5168,
"theoretical_loss": 3.5277925818362927,
"tokens_seen": 1444675584
},
{
"epoch": 0.44,
"learning_rate": 0.00028399197592778334,
"loss": 2.8453,
"theoretical_loss": 3.527778370201778,
"tokens_seen": 1444741120
},
{
"epoch": 0.44,
"learning_rate": 0.0002839819458375125,
"loss": 2.5906,
"theoretical_loss": 3.527764159392411,
"tokens_seen": 1444806656
},
{
"epoch": 0.44,
"learning_rate": 0.00028397191574724176,
"loss": 2.6565,
"theoretical_loss": 3.527749949408105,
"tokens_seen": 1444872192
},
{
"epoch": 0.44,
"learning_rate": 0.0002839618856569709,
"loss": 2.876,
"theoretical_loss": 3.5277357402487755,
"tokens_seen": 1444937728
},
{
"epoch": 0.44,
"learning_rate": 0.0002839518555667001,
"loss": 2.8525,
"theoretical_loss": 3.527721531914337,
"tokens_seen": 1445003264
},
{
"epoch": 0.44,
"objective/train/docs_used": 819330,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.570746421813965,
"objective/train/theoretical_loss": 3.527707324404704,
"objective/train/tokens_used": 1465528800,
"theoretical_loss": 3.527707324404704,
"tokens_seen": 1445068800
},
{
"epoch": 0.44,
"learning_rate": 0.00028394182547642925,
"loss": 2.6435,
"theoretical_loss": 3.527707324404704,
"tokens_seen": 1445068800
},
{
"epoch": 0.44,
"learning_rate": 0.0002839317953861585,
"loss": 2.7574,
"theoretical_loss": 3.5276931177197914,
"tokens_seen": 1445134336
},
{
"epoch": 0.44,
"learning_rate": 0.00028392176529588766,
"loss": 2.5464,
"theoretical_loss": 3.527678911859514,
"tokens_seen": 1445199872
},
{
"epoch": 0.44,
"learning_rate": 0.00028391173520561684,
"loss": 2.8376,
"theoretical_loss": 3.5276647068237867,
"tokens_seen": 1445265408
},
{
"epoch": 0.44,
"learning_rate": 0.000283901705115346,
"loss": 2.7723,
"theoretical_loss": 3.5276505026125236,
"tokens_seen": 1445330944
},
{
"epoch": 0.44,
"learning_rate": 0.00028389167502507526,
"loss": 2.5751,
"theoretical_loss": 3.52763629922564,
"tokens_seen": 1445396480
},
{
"epoch": 0.44,
"learning_rate": 0.0002838816449348044,
"loss": 2.6691,
"theoretical_loss": 3.5276220966630505,
"tokens_seen": 1445462016
},
{
"epoch": 0.44,
"learning_rate": 0.0002838716148445336,
"loss": 2.7676,
"theoretical_loss": 3.52760789492467,
"tokens_seen": 1445527552
},
{
"epoch": 0.44,
"learning_rate": 0.00028386158475426275,
"loss": 2.9706,
"theoretical_loss": 3.527593694010413,
"tokens_seen": 1445593088
},
{
"epoch": 0.44,
"learning_rate": 0.000283851554663992,
"loss": 2.6484,
"theoretical_loss": 3.5275794939201948,
"tokens_seen": 1445658624
},
{
"epoch": 0.44,
"learning_rate": 0.00028384152457372117,
"loss": 2.7474,
"theoretical_loss": 3.52756529465393,
"tokens_seen": 1445724160
},
{
"epoch": 0.44,
"learning_rate": 0.00028383149448345035,
"loss": 2.7487,
"theoretical_loss": 3.5275510962115337,
"tokens_seen": 1445789696
},
{
"epoch": 0.44,
"learning_rate": 0.00028382146439317953,
"loss": 2.708,
"theoretical_loss": 3.5275368985929205,
"tokens_seen": 1445855232
},
{
"epoch": 0.44,
"learning_rate": 0.0002838114343029087,
"loss": 2.7522,
"theoretical_loss": 3.527522701798005,
"tokens_seen": 1445920768
},
{
"epoch": 0.44,
"learning_rate": 0.0002838014042126379,
"loss": 2.7081,
"theoretical_loss": 3.5275085058267024,
"tokens_seen": 1445986304
},
{
"epoch": 0.44,
"learning_rate": 0.0002837913741223671,
"loss": 2.7469,
"theoretical_loss": 3.5274943106789274,
"tokens_seen": 1446051840
},
{
"epoch": 0.44,
"learning_rate": 0.0002837813440320963,
"loss": 2.535,
"theoretical_loss": 3.527480116354595,
"tokens_seen": 1446117376
},
{
"epoch": 0.44,
"learning_rate": 0.0002837713139418255,
"loss": 2.7216,
"theoretical_loss": 3.5274659228536205,
"tokens_seen": 1446182912
},
{
"epoch": 0.44,
"learning_rate": 0.00028376128385155467,
"loss": 2.6793,
"theoretical_loss": 3.5274517301759185,
"tokens_seen": 1446248448
},
{
"epoch": 0.44,
"learning_rate": 0.00028375125376128385,
"loss": 2.8913,
"theoretical_loss": 3.527437538321404,
"tokens_seen": 1446313984
},
{
"epoch": 0.44,
"learning_rate": 0.0002837412236710131,
"loss": 2.7687,
"theoretical_loss": 3.5274233472899916,
"tokens_seen": 1446379520
},
{
"epoch": 0.44,
"learning_rate": 0.0002837311935807422,
"loss": 2.4814,
"theoretical_loss": 3.527409157081597,
"tokens_seen": 1446445056
},
{
"epoch": 0.44,
"learning_rate": 0.00028372116349047145,
"loss": 2.6849,
"theoretical_loss": 3.527394967696134,
"tokens_seen": 1446510592
},
{
"epoch": 0.44,
"learning_rate": 0.00028371113340020063,
"loss": 2.8723,
"theoretical_loss": 3.5273807791335194,
"tokens_seen": 1446576128
},
{
"epoch": 0.44,
"learning_rate": 0.0002837011033099298,
"loss": 2.6644,
"theoretical_loss": 3.5273665913936663,
"tokens_seen": 1446641664
},
{
"epoch": 0.44,
"objective/train/docs_used": 820636,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.566915512084961,
"objective/train/theoretical_loss": 3.527352404476491,
"objective/train/tokens_used": 1467167200,
"theoretical_loss": 3.527352404476491,
"tokens_seen": 1446707200
},
{
"epoch": 0.44,
"learning_rate": 0.000283691073219659,
"loss": 2.6641,
"theoretical_loss": 3.527352404476491,
"tokens_seen": 1446707200
},
{
"epoch": 0.44,
"learning_rate": 0.0002836810431293882,
"loss": 2.7324,
"theoretical_loss": 3.527338218381908,
"tokens_seen": 1446772736
},
{
"epoch": 0.44,
"learning_rate": 0.00028367101303911735,
"loss": 2.7076,
"theoretical_loss": 3.5273240331098323,
"tokens_seen": 1446838272
},
{
"epoch": 0.44,
"learning_rate": 0.0002836609829488466,
"loss": 2.6947,
"theoretical_loss": 3.5273098486601793,
"tokens_seen": 1446903808
},
{
"epoch": 0.44,
"learning_rate": 0.0002836509528585757,
"loss": 2.5762,
"theoretical_loss": 3.527295665032864,
"tokens_seen": 1446969344
},
{
"epoch": 0.44,
"learning_rate": 0.00028364092276830495,
"loss": 2.8098,
"theoretical_loss": 3.527281482227801,
"tokens_seen": 1447034880
},
{
"epoch": 0.44,
"learning_rate": 0.0002836308926780341,
"loss": 2.7545,
"theoretical_loss": 3.527267300244906,
"tokens_seen": 1447100416
},
{
"epoch": 0.44,
"learning_rate": 0.0002836208625877633,
"loss": 2.6955,
"theoretical_loss": 3.5272531190840937,
"tokens_seen": 1447165952
},
{
"epoch": 0.44,
"learning_rate": 0.0002836108324974925,
"loss": 2.7348,
"theoretical_loss": 3.52723893874528,
"tokens_seen": 1447231488
},
{
"epoch": 0.44,
"learning_rate": 0.0002836008024072217,
"loss": 2.6358,
"theoretical_loss": 3.5272247592283787,
"tokens_seen": 1447297024
},
{
"epoch": 0.44,
"learning_rate": 0.00028359077231695086,
"loss": 2.829,
"theoretical_loss": 3.527210580533306,
"tokens_seen": 1447362560
},
{
"epoch": 0.44,
"learning_rate": 0.0002835807422266801,
"loss": 2.6084,
"theoretical_loss": 3.527196402659977,
"tokens_seen": 1447428096
},
{
"epoch": 0.44,
"learning_rate": 0.0002835707121364092,
"loss": 2.7588,
"theoretical_loss": 3.527182225608306,
"tokens_seen": 1447493632
},
{
"epoch": 0.44,
"learning_rate": 0.00028356068204613846,
"loss": 2.893,
"theoretical_loss": 3.527168049378209,
"tokens_seen": 1447559168
},
{
"epoch": 0.44,
"learning_rate": 0.0002835506519558676,
"loss": 2.7718,
"theoretical_loss": 3.5271538739696013,
"tokens_seen": 1447624704
},
{
"epoch": 0.44,
"learning_rate": 0.0002835406218655968,
"loss": 2.8303,
"theoretical_loss": 3.5271396993823974,
"tokens_seen": 1447690240
},
{
"epoch": 0.44,
"learning_rate": 0.000283530591775326,
"loss": 2.6782,
"theoretical_loss": 3.5271255256165133,
"tokens_seen": 1447755776
},
{
"epoch": 0.44,
"learning_rate": 0.0002835205616850552,
"loss": 2.8153,
"theoretical_loss": 3.527111352671864,
"tokens_seen": 1447821312
},
{
"epoch": 0.44,
"learning_rate": 0.00028351053159478436,
"loss": 2.7913,
"theoretical_loss": 3.527097180548364,
"tokens_seen": 1447886848
},
{
"epoch": 0.44,
"learning_rate": 0.00028350050150451354,
"loss": 2.5786,
"theoretical_loss": 3.5270830092459295,
"tokens_seen": 1447952384
},
{
"epoch": 0.44,
"learning_rate": 0.0002834904714142427,
"loss": 2.842,
"theoretical_loss": 3.527068838764475,
"tokens_seen": 1448017920
},
{
"epoch": 0.44,
"learning_rate": 0.00028348044132397196,
"loss": 2.7038,
"theoretical_loss": 3.527054669103917,
"tokens_seen": 1448083456
},
{
"epoch": 0.44,
"learning_rate": 0.0002834704112337011,
"loss": 2.8297,
"theoretical_loss": 3.527040500264169,
"tokens_seen": 1448148992
},
{
"epoch": 0.44,
"learning_rate": 0.0002834603811434303,
"loss": 2.5416,
"theoretical_loss": 3.527026332245148,
"tokens_seen": 1448214528
},
{
"epoch": 0.44,
"learning_rate": 0.00028345035105315945,
"loss": 2.7533,
"theoretical_loss": 3.527012165046769,
"tokens_seen": 1448280064
},
{
"epoch": 0.44,
"objective/train/docs_used": 821297,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.046020746231079,
"objective/train/theoretical_loss": 3.5269979986689464,
"objective/train/tokens_used": 1468805600,
"theoretical_loss": 3.5269979986689464,
"tokens_seen": 1448345600
},
{
"epoch": 0.44,
"learning_rate": 0.0002834403209628887,
"loss": 2.7483,
"theoretical_loss": 3.5269979986689464,
"tokens_seen": 1448345600
},
{
"epoch": 0.44,
"learning_rate": 0.00028343029087261786,
"loss": 2.6401,
"theoretical_loss": 3.526983833111596,
"tokens_seen": 1448411136
},
{
"epoch": 0.44,
"learning_rate": 0.00028342026078234705,
"loss": 2.853,
"theoretical_loss": 3.526969668374634,
"tokens_seen": 1448476672
},
{
"epoch": 0.44,
"learning_rate": 0.0002834102306920762,
"loss": 2.8231,
"theoretical_loss": 3.5269555044579746,
"tokens_seen": 1448542208
},
{
"epoch": 0.44,
"learning_rate": 0.00028340020060180546,
"loss": 2.8153,
"theoretical_loss": 3.5269413413615336,
"tokens_seen": 1448607744
},
{
"epoch": 0.44,
"learning_rate": 0.0002833901705115346,
"loss": 2.5768,
"theoretical_loss": 3.5269271790852263,
"tokens_seen": 1448673280
},
{
"epoch": 0.44,
"learning_rate": 0.0002833801404212638,
"loss": 2.4716,
"theoretical_loss": 3.526913017628969,
"tokens_seen": 1448738816
},
{
"epoch": 0.44,
"learning_rate": 0.00028337011033099295,
"loss": 2.7365,
"theoretical_loss": 3.5268988569926756,
"tokens_seen": 1448804352
},
{
"epoch": 0.44,
"learning_rate": 0.0002833600802407222,
"loss": 2.6879,
"theoretical_loss": 3.5268846971762624,
"tokens_seen": 1448869888
},
{
"epoch": 0.44,
"learning_rate": 0.00028335005015045137,
"loss": 2.693,
"theoretical_loss": 3.5268705381796455,
"tokens_seen": 1448935424
},
{
"epoch": 0.44,
"learning_rate": 0.00028334002006018055,
"loss": 2.872,
"theoretical_loss": 3.526856380002739,
"tokens_seen": 1449000960
},
{
"epoch": 0.44,
"learning_rate": 0.00028332998996990973,
"loss": 2.6497,
"theoretical_loss": 3.5268422226454588,
"tokens_seen": 1449066496
},
{
"epoch": 0.44,
"learning_rate": 0.0002833199598796389,
"loss": 2.6198,
"theoretical_loss": 3.526828066107721,
"tokens_seen": 1449132032
},
{
"epoch": 0.44,
"learning_rate": 0.0002833099297893681,
"loss": 2.6778,
"theoretical_loss": 3.5268139103894405,
"tokens_seen": 1449197568
},
{
"epoch": 0.44,
"learning_rate": 0.0002832998996990973,
"loss": 2.6399,
"theoretical_loss": 3.526799755490533,
"tokens_seen": 1449263104
},
{
"epoch": 0.44,
"learning_rate": 0.00028328986960882645,
"loss": 2.5742,
"theoretical_loss": 3.5267856014109142,
"tokens_seen": 1449328640
},
{
"epoch": 0.44,
"learning_rate": 0.0002832798395185557,
"loss": 2.6563,
"theoretical_loss": 3.526771448150499,
"tokens_seen": 1449394176
},
{
"epoch": 0.44,
"learning_rate": 0.0002832698094282848,
"loss": 2.6241,
"theoretical_loss": 3.5267572957092037,
"tokens_seen": 1449459712
},
{
"epoch": 0.44,
"learning_rate": 0.00028325977933801405,
"loss": 2.7155,
"theoretical_loss": 3.526743144086944,
"tokens_seen": 1449525248
},
{
"epoch": 0.44,
"learning_rate": 0.00028324974924774323,
"loss": 2.7836,
"theoretical_loss": 3.5267289932836343,
"tokens_seen": 1449590784
},
{
"epoch": 0.44,
"learning_rate": 0.0002832397191574724,
"loss": 2.676,
"theoretical_loss": 3.5267148432991915,
"tokens_seen": 1449656320
},
{
"epoch": 0.44,
"learning_rate": 0.0002832296890672016,
"loss": 2.6254,
"theoretical_loss": 3.52670069413353,
"tokens_seen": 1449721856
},
{
"epoch": 0.44,
"learning_rate": 0.00028321965897693083,
"loss": 2.6492,
"theoretical_loss": 3.526686545786567,
"tokens_seen": 1449787392
},
{
"epoch": 0.44,
"learning_rate": 0.00028320962888665996,
"loss": 2.6278,
"theoretical_loss": 3.526672398258216,
"tokens_seen": 1449852928
},
{
"epoch": 0.44,
"learning_rate": 0.0002831995987963892,
"loss": 2.5266,
"theoretical_loss": 3.5266582515483944,
"tokens_seen": 1449918464
},
{
"epoch": 0.44,
"objective/train/docs_used": 822517,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4528770446777344,
"objective/train/theoretical_loss": 3.5266441056570175,
"objective/train/tokens_used": 1470444000,
"theoretical_loss": 3.5266441056570175,
"tokens_seen": 1449984000
},
{
"epoch": 0.44,
"learning_rate": 0.0002831895687061183,
"loss": 2.6705,
"theoretical_loss": 3.5266441056570175,
"tokens_seen": 1449984000
},
{
"epoch": 0.44,
"learning_rate": 0.00028317953861584755,
"loss": 2.5696,
"theoretical_loss": 3.526629960584,
"tokens_seen": 1450049536
},
{
"epoch": 0.44,
"learning_rate": 0.00028316950852557674,
"loss": 2.5081,
"theoretical_loss": 3.526615816329259,
"tokens_seen": 1450115072
},
{
"epoch": 0.44,
"learning_rate": 0.0002831594784353059,
"loss": 2.7341,
"theoretical_loss": 3.526601672892709,
"tokens_seen": 1450180608
},
{
"epoch": 0.44,
"learning_rate": 0.0002831494483450351,
"loss": 2.5136,
"theoretical_loss": 3.526587530274266,
"tokens_seen": 1450246144
},
{
"epoch": 0.44,
"learning_rate": 0.0002831394182547643,
"loss": 2.7095,
"theoretical_loss": 3.5265733884738464,
"tokens_seen": 1450311680
},
{
"epoch": 0.44,
"learning_rate": 0.00028312938816449346,
"loss": 2.791,
"theoretical_loss": 3.526559247491365,
"tokens_seen": 1450377216
},
{
"epoch": 0.44,
"learning_rate": 0.0002831193580742227,
"loss": 2.4877,
"theoretical_loss": 3.526545107326738,
"tokens_seen": 1450442752
},
{
"epoch": 0.44,
"learning_rate": 0.0002831093279839518,
"loss": 2.6263,
"theoretical_loss": 3.5265309679798813,
"tokens_seen": 1450508288
},
{
"epoch": 0.44,
"learning_rate": 0.00028309929789368106,
"loss": 2.637,
"theoretical_loss": 3.52651682945071,
"tokens_seen": 1450573824
},
{
"epoch": 0.44,
"learning_rate": 0.0002830892678034102,
"loss": 2.554,
"theoretical_loss": 3.526502691739141,
"tokens_seen": 1450639360
},
{
"epoch": 0.44,
"learning_rate": 0.0002830792377131394,
"loss": 2.768,
"theoretical_loss": 3.5264885548450886,
"tokens_seen": 1450704896
},
{
"epoch": 0.44,
"learning_rate": 0.0002830692076228686,
"loss": 2.8381,
"theoretical_loss": 3.52647441876847,
"tokens_seen": 1450770432
},
{
"epoch": 0.44,
"learning_rate": 0.0002830591775325978,
"loss": 2.8026,
"theoretical_loss": 3.5264602835092003,
"tokens_seen": 1450835968
},
{
"epoch": 0.44,
"learning_rate": 0.00028304914744232696,
"loss": 2.8345,
"theoretical_loss": 3.526446149067195,
"tokens_seen": 1450901504
},
{
"epoch": 0.44,
"learning_rate": 0.0002830391173520562,
"loss": 2.5985,
"theoretical_loss": 3.5264320154423707,
"tokens_seen": 1450967040
},
{
"epoch": 0.44,
"learning_rate": 0.0002830290872617854,
"loss": 2.6703,
"theoretical_loss": 3.526417882634643,
"tokens_seen": 1451032576
},
{
"epoch": 0.44,
"learning_rate": 0.00028301905717151456,
"loss": 2.7993,
"theoretical_loss": 3.5264037506439276,
"tokens_seen": 1451098112
},
{
"epoch": 0.44,
"learning_rate": 0.00028300902708124374,
"loss": 2.6318,
"theoretical_loss": 3.52638961947014,
"tokens_seen": 1451163648
},
{
"epoch": 0.44,
"learning_rate": 0.0002829989969909729,
"loss": 2.5529,
"theoretical_loss": 3.5263754891131973,
"tokens_seen": 1451229184
},
{
"epoch": 0.44,
"learning_rate": 0.00028298896690070216,
"loss": 2.6613,
"theoretical_loss": 3.5263613595730146,
"tokens_seen": 1451294720
},
{
"epoch": 0.44,
"learning_rate": 0.0002829789368104313,
"loss": 2.8535,
"theoretical_loss": 3.526347230849507,
"tokens_seen": 1451360256
},
{
"epoch": 0.44,
"learning_rate": 0.0002829689067201605,
"loss": 2.7803,
"theoretical_loss": 3.526333102942592,
"tokens_seen": 1451425792
},
{
"epoch": 0.44,
"learning_rate": 0.00028295887662988965,
"loss": 2.9124,
"theoretical_loss": 3.5263189758521847,
"tokens_seen": 1451491328
},
{
"epoch": 0.44,
"learning_rate": 0.0002829488465396189,
"loss": 3.0537,
"theoretical_loss": 3.5263048495782012,
"tokens_seen": 1451556864
},
{
"epoch": 0.44,
"objective/train/docs_used": 823297,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.607311248779297,
"objective/train/theoretical_loss": 3.5262907241205568,
"objective/train/tokens_used": 1472082400,
"theoretical_loss": 3.5262907241205568,
"tokens_seen": 1451622400
},
{
"epoch": 0.44,
"learning_rate": 0.00028293881644934806,
"loss": 2.663,
"theoretical_loss": 3.5262907241205568,
"tokens_seen": 1451622400
},
{
"epoch": 0.44,
"learning_rate": 0.00028292878635907725,
"loss": 2.8232,
"theoretical_loss": 3.5262765994791687,
"tokens_seen": 1451687936
},
{
"epoch": 0.44,
"learning_rate": 0.0002829187562688064,
"loss": 2.7553,
"theoretical_loss": 3.5262624756539522,
"tokens_seen": 1451753472
},
{
"epoch": 0.44,
"learning_rate": 0.00028290872617853566,
"loss": 3.0206,
"theoretical_loss": 3.5262483526448234,
"tokens_seen": 1451819008
},
{
"epoch": 0.44,
"learning_rate": 0.0002828986960882648,
"loss": 2.8969,
"theoretical_loss": 3.526234230451698,
"tokens_seen": 1451884544
},
{
"epoch": 0.44,
"learning_rate": 0.000282888665997994,
"loss": 2.6809,
"theoretical_loss": 3.5262201090744925,
"tokens_seen": 1451950080
},
{
"epoch": 0.44,
"learning_rate": 0.00028287863590772315,
"loss": 2.7074,
"theoretical_loss": 3.526205988513123,
"tokens_seen": 1452015616
},
{
"epoch": 0.44,
"learning_rate": 0.0002828686058174524,
"loss": 2.6552,
"theoretical_loss": 3.5261918687675045,
"tokens_seen": 1452081152
},
{
"epoch": 0.44,
"learning_rate": 0.00028285857572718157,
"loss": 2.8008,
"theoretical_loss": 3.5261777498375544,
"tokens_seen": 1452146688
},
{
"epoch": 0.44,
"learning_rate": 0.00028284854563691075,
"loss": 2.5409,
"theoretical_loss": 3.526163631723188,
"tokens_seen": 1452212224
},
{
"epoch": 0.44,
"learning_rate": 0.00028283851554663993,
"loss": 2.6101,
"theoretical_loss": 3.526149514424322,
"tokens_seen": 1452277760
},
{
"epoch": 0.44,
"learning_rate": 0.0002828284854563691,
"loss": 2.6375,
"theoretical_loss": 3.5261353979408714,
"tokens_seen": 1452343296
},
{
"epoch": 0.44,
"learning_rate": 0.0002828184553660983,
"loss": 2.6801,
"theoretical_loss": 3.5261212822727535,
"tokens_seen": 1452408832
},
{
"epoch": 0.44,
"learning_rate": 0.0002828084252758275,
"loss": 2.5104,
"theoretical_loss": 3.526107167419884,
"tokens_seen": 1452474368
},
{
"epoch": 0.44,
"learning_rate": 0.00028279839518555665,
"loss": 2.5598,
"theoretical_loss": 3.526093053382179,
"tokens_seen": 1452539904
},
{
"epoch": 0.44,
"learning_rate": 0.0002827883650952859,
"loss": 2.7135,
"theoretical_loss": 3.5260789401595543,
"tokens_seen": 1452605440
},
{
"epoch": 0.44,
"learning_rate": 0.000282778335005015,
"loss": 2.6482,
"theoretical_loss": 3.5260648277519264,
"tokens_seen": 1452670976
},
{
"epoch": 0.44,
"learning_rate": 0.00028276830491474425,
"loss": 2.6403,
"theoretical_loss": 3.5260507161592116,
"tokens_seen": 1452736512
},
{
"epoch": 0.44,
"learning_rate": 0.00028275827482447343,
"loss": 2.8846,
"theoretical_loss": 3.526036605381326,
"tokens_seen": 1452802048
},
{
"epoch": 0.44,
"learning_rate": 0.0002827482447342026,
"loss": 2.6496,
"theoretical_loss": 3.5260224954181854,
"tokens_seen": 1452867584
},
{
"epoch": 0.44,
"learning_rate": 0.0002827382146439318,
"loss": 2.9562,
"theoretical_loss": 3.5260083862697065,
"tokens_seen": 1452933120
},
{
"epoch": 0.44,
"learning_rate": 0.00028272818455366103,
"loss": 2.6837,
"theoretical_loss": 3.5259942779358053,
"tokens_seen": 1452998656
},
{
"epoch": 0.44,
"learning_rate": 0.00028271815446339016,
"loss": 2.8021,
"theoretical_loss": 3.525980170416398,
"tokens_seen": 1453064192
},
{
"epoch": 0.44,
"learning_rate": 0.0002827081243731194,
"loss": 2.8636,
"theoretical_loss": 3.525966063711401,
"tokens_seen": 1453129728
},
{
"epoch": 0.44,
"learning_rate": 0.0002826980942828485,
"loss": 2.6776,
"theoretical_loss": 3.525951957820731,
"tokens_seen": 1453195264
},
{
"epoch": 0.44,
"objective/train/docs_used": 824661,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9723093509674072,
"objective/train/theoretical_loss": 3.5259378527443035,
"objective/train/tokens_used": 1473720800,
"theoretical_loss": 3.5259378527443035,
"tokens_seen": 1453260800
},
{
"epoch": 0.44,
"learning_rate": 0.00028268806419257775,
"loss": 2.5908,
"theoretical_loss": 3.5259378527443035,
"tokens_seen": 1453260800
},
{
"epoch": 0.44,
"learning_rate": 0.00028267803410230694,
"loss": 2.7709,
"theoretical_loss": 3.5259237484820347,
"tokens_seen": 1453326336
},
{
"epoch": 0.44,
"learning_rate": 0.0002826680040120361,
"loss": 2.5741,
"theoretical_loss": 3.525909645033841,
"tokens_seen": 1453391872
},
{
"epoch": 0.44,
"learning_rate": 0.0002826579739217653,
"loss": 2.877,
"theoretical_loss": 3.5258955423996396,
"tokens_seen": 1453457408
},
{
"epoch": 0.44,
"learning_rate": 0.0002826479438314945,
"loss": 2.8461,
"theoretical_loss": 3.525881440579346,
"tokens_seen": 1453522944
},
{
"epoch": 0.44,
"learning_rate": 0.00028263791374122366,
"loss": 2.6336,
"theoretical_loss": 3.5258673395728763,
"tokens_seen": 1453588480
},
{
"epoch": 0.44,
"learning_rate": 0.0002826278836509529,
"loss": 2.6231,
"theoretical_loss": 3.5258532393801474,
"tokens_seen": 1453654016
},
{
"epoch": 0.44,
"learning_rate": 0.000282617853560682,
"loss": 2.5197,
"theoretical_loss": 3.525839140001076,
"tokens_seen": 1453719552
},
{
"epoch": 0.44,
"learning_rate": 0.00028260782347041126,
"loss": 2.8398,
"theoretical_loss": 3.525825041435578,
"tokens_seen": 1453785088
},
{
"epoch": 0.44,
"learning_rate": 0.0002825977933801404,
"loss": 2.7179,
"theoretical_loss": 3.525810943683569,
"tokens_seen": 1453850624
},
{
"epoch": 0.44,
"learning_rate": 0.0002825877632898696,
"loss": 2.794,
"theoretical_loss": 3.5257968467449663,
"tokens_seen": 1453916160
},
{
"epoch": 0.44,
"learning_rate": 0.0002825777331995988,
"loss": 2.7625,
"theoretical_loss": 3.525782750619687,
"tokens_seen": 1453981696
},
{
"epoch": 0.44,
"learning_rate": 0.000282567703109328,
"loss": 2.6477,
"theoretical_loss": 3.5257686553076457,
"tokens_seen": 1454047232
},
{
"epoch": 0.44,
"learning_rate": 0.00028255767301905716,
"loss": 2.7184,
"theoretical_loss": 3.5257545608087604,
"tokens_seen": 1454112768
},
{
"epoch": 0.44,
"learning_rate": 0.0002825476429287864,
"loss": 2.7595,
"theoretical_loss": 3.5257404671229464,
"tokens_seen": 1454178304
},
{
"epoch": 0.44,
"learning_rate": 0.0002825376128385155,
"loss": 2.7913,
"theoretical_loss": 3.5257263742501213,
"tokens_seen": 1454243840
},
{
"epoch": 0.44,
"learning_rate": 0.00028252758274824476,
"loss": 2.6181,
"theoretical_loss": 3.5257122821902005,
"tokens_seen": 1454309376
},
{
"epoch": 0.44,
"learning_rate": 0.0002825175526579739,
"loss": 2.6331,
"theoretical_loss": 3.525698190943101,
"tokens_seen": 1454374912
},
{
"epoch": 0.44,
"learning_rate": 0.0002825075225677031,
"loss": 2.6784,
"theoretical_loss": 3.5256841005087396,
"tokens_seen": 1454440448
},
{
"epoch": 0.44,
"learning_rate": 0.0002824974924774323,
"loss": 2.5979,
"theoretical_loss": 3.5256700108870325,
"tokens_seen": 1454505984
},
{
"epoch": 0.44,
"learning_rate": 0.0002824874623871615,
"loss": 2.604,
"theoretical_loss": 3.5256559220778954,
"tokens_seen": 1454571520
},
{
"epoch": 0.44,
"learning_rate": 0.00028247743229689067,
"loss": 2.68,
"theoretical_loss": 3.525641834081246,
"tokens_seen": 1454637056
},
{
"epoch": 0.44,
"learning_rate": 0.00028246740220661985,
"loss": 2.7075,
"theoretical_loss": 3.525627746897001,
"tokens_seen": 1454702592
},
{
"epoch": 0.44,
"learning_rate": 0.00028245737211634903,
"loss": 2.7003,
"theoretical_loss": 3.5256136605250763,
"tokens_seen": 1454768128
},
{
"epoch": 0.44,
"learning_rate": 0.00028244734202607826,
"loss": 2.8688,
"theoretical_loss": 3.525599574965388,
"tokens_seen": 1454833664
},
{
"epoch": 0.44,
"objective/train/docs_used": 824875,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.707169771194458,
"objective/train/theoretical_loss": 3.5255854902178534,
"objective/train/tokens_used": 1475359200,
"theoretical_loss": 3.5255854902178534,
"tokens_seen": 1454899200
},
{
"epoch": 0.44,
"learning_rate": 0.0002824373119358074,
"loss": 2.6066,
"theoretical_loss": 3.5255854902178534,
"tokens_seen": 1454899200
},
{
"epoch": 0.44,
"learning_rate": 0.0002824272818455366,
"loss": 2.3995,
"theoretical_loss": 3.5255714062823893,
"tokens_seen": 1454964736
},
{
"epoch": 0.44,
"learning_rate": 0.0002824172517552658,
"loss": 2.724,
"theoretical_loss": 3.5255573231589117,
"tokens_seen": 1455030272
},
{
"epoch": 0.44,
"learning_rate": 0.000282407221664995,
"loss": 2.743,
"theoretical_loss": 3.5255432408473375,
"tokens_seen": 1455095808
},
{
"epoch": 0.44,
"learning_rate": 0.00028239719157472417,
"loss": 2.99,
"theoretical_loss": 3.525529159347583,
"tokens_seen": 1455161344
},
{
"epoch": 0.44,
"learning_rate": 0.00028238716148445335,
"loss": 2.9094,
"theoretical_loss": 3.5255150786595655,
"tokens_seen": 1455226880
},
{
"epoch": 0.44,
"learning_rate": 0.00028237713139418253,
"loss": 2.8286,
"theoretical_loss": 3.525500998783201,
"tokens_seen": 1455292416
},
{
"epoch": 0.44,
"learning_rate": 0.00028236710130391177,
"loss": 2.6355,
"theoretical_loss": 3.5254869197184067,
"tokens_seen": 1455357952
},
{
"epoch": 0.44,
"learning_rate": 0.0002823570712136409,
"loss": 2.5975,
"theoretical_loss": 3.525472841465099,
"tokens_seen": 1455423488
},
{
"epoch": 0.44,
"learning_rate": 0.00028234704112337013,
"loss": 2.6111,
"theoretical_loss": 3.5254587640231945,
"tokens_seen": 1455489024
},
{
"epoch": 0.44,
"learning_rate": 0.00028233701103309926,
"loss": 2.7363,
"theoretical_loss": 3.52544468739261,
"tokens_seen": 1455554560
},
{
"epoch": 0.44,
"learning_rate": 0.0002823269809428285,
"loss": 2.9252,
"theoretical_loss": 3.5254306115732623,
"tokens_seen": 1455620096
},
{
"epoch": 0.44,
"learning_rate": 0.0002823169508525577,
"loss": 2.7943,
"theoretical_loss": 3.525416536565068,
"tokens_seen": 1455685632
},
{
"epoch": 0.44,
"learning_rate": 0.00028230692076228685,
"loss": 2.7769,
"theoretical_loss": 3.525402462367944,
"tokens_seen": 1455751168
},
{
"epoch": 0.44,
"learning_rate": 0.00028229689067201604,
"loss": 2.7684,
"theoretical_loss": 3.525388388981807,
"tokens_seen": 1455816704
},
{
"epoch": 0.44,
"learning_rate": 0.0002822868605817452,
"loss": 2.8464,
"theoretical_loss": 3.525374316406574,
"tokens_seen": 1455882240
},
{
"epoch": 0.44,
"learning_rate": 0.00028227683049147445,
"loss": 2.8362,
"theoretical_loss": 3.525360244642161,
"tokens_seen": 1455947776
},
{
"epoch": 0.44,
"learning_rate": 0.00028226680040120363,
"loss": 2.7849,
"theoretical_loss": 3.5253461736884852,
"tokens_seen": 1456013312
},
{
"epoch": 0.44,
"learning_rate": 0.0002822567703109328,
"loss": 2.7832,
"theoretical_loss": 3.525332103545464,
"tokens_seen": 1456078848
},
{
"epoch": 0.44,
"learning_rate": 0.000282246740220662,
"loss": 2.7097,
"theoretical_loss": 3.5253180342130133,
"tokens_seen": 1456144384
},
{
"epoch": 0.44,
"learning_rate": 0.00028223671013039123,
"loss": 2.6544,
"theoretical_loss": 3.5253039656910508,
"tokens_seen": 1456209920
},
{
"epoch": 0.44,
"learning_rate": 0.00028222668004012036,
"loss": 2.7252,
"theoretical_loss": 3.5252898979794924,
"tokens_seen": 1456275456
},
{
"epoch": 0.44,
"learning_rate": 0.0002822166499498496,
"loss": 2.7505,
"theoretical_loss": 3.5252758310782557,
"tokens_seen": 1456340992
},
{
"epoch": 0.44,
"learning_rate": 0.0002822066198595787,
"loss": 2.7382,
"theoretical_loss": 3.525261764987257,
"tokens_seen": 1456406528
},
{
"epoch": 0.44,
"learning_rate": 0.00028219658976930795,
"loss": 2.7343,
"theoretical_loss": 3.5252476997064135,
"tokens_seen": 1456472064
},
{
"epoch": 0.44,
"objective/train/docs_used": 826174,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8232882022857666,
"objective/train/theoretical_loss": 3.525233635235642,
"objective/train/tokens_used": 1476997600,
"theoretical_loss": 3.525233635235642,
"tokens_seen": 1456537600
},
{
"epoch": 0.44,
"learning_rate": 0.00028218655967903714,
"loss": 2.7617,
"theoretical_loss": 3.525233635235642,
"tokens_seen": 1456537600
},
{
"epoch": 0.44,
"learning_rate": 0.0002821765295887663,
"loss": 2.6243,
"theoretical_loss": 3.5252195715748598,
"tokens_seen": 1456603136
},
{
"epoch": 0.44,
"learning_rate": 0.0002821664994984955,
"loss": 2.8443,
"theoretical_loss": 3.5252055087239835,
"tokens_seen": 1456668672
},
{
"epoch": 0.44,
"learning_rate": 0.0002821564694082247,
"loss": 2.7184,
"theoretical_loss": 3.5251914466829293,
"tokens_seen": 1456734208
},
{
"epoch": 0.44,
"learning_rate": 0.00028214643931795386,
"loss": 2.7687,
"theoretical_loss": 3.5251773854516153,
"tokens_seen": 1456799744
},
{
"epoch": 0.44,
"learning_rate": 0.0002821364092276831,
"loss": 2.8741,
"theoretical_loss": 3.5251633250299577,
"tokens_seen": 1456865280
},
{
"epoch": 0.44,
"learning_rate": 0.0002821263791374122,
"loss": 2.9155,
"theoretical_loss": 3.525149265417874,
"tokens_seen": 1456930816
},
{
"epoch": 0.44,
"learning_rate": 0.00028211634904714146,
"loss": 2.7256,
"theoretical_loss": 3.525135206615281,
"tokens_seen": 1456996352
},
{
"epoch": 0.44,
"learning_rate": 0.0002821063189568706,
"loss": 2.608,
"theoretical_loss": 3.5251211486220955,
"tokens_seen": 1457061888
},
{
"epoch": 0.44,
"learning_rate": 0.0002820962888665998,
"loss": 2.5863,
"theoretical_loss": 3.525107091438234,
"tokens_seen": 1457127424
},
{
"epoch": 0.44,
"learning_rate": 0.000282086258776329,
"loss": 2.6022,
"theoretical_loss": 3.5250930350636147,
"tokens_seen": 1457192960
},
{
"epoch": 0.44,
"learning_rate": 0.0002820762286860582,
"loss": 2.8464,
"theoretical_loss": 3.525078979498154,
"tokens_seen": 1457258496
},
{
"epoch": 0.44,
"learning_rate": 0.00028206619859578736,
"loss": 2.6429,
"theoretical_loss": 3.525064924741769,
"tokens_seen": 1457324032
},
{
"epoch": 0.44,
"learning_rate": 0.0002820561685055166,
"loss": 2.6513,
"theoretical_loss": 3.5250508707943764,
"tokens_seen": 1457389568
},
{
"epoch": 0.44,
"learning_rate": 0.0002820461384152457,
"loss": 2.6903,
"theoretical_loss": 3.5250368176558933,
"tokens_seen": 1457455104
},
{
"epoch": 0.44,
"learning_rate": 0.00028203610832497496,
"loss": 3.0793,
"theoretical_loss": 3.5250227653262374,
"tokens_seen": 1457520640
},
{
"epoch": 0.44,
"learning_rate": 0.0002820260782347041,
"loss": 2.7605,
"theoretical_loss": 3.5250087138053257,
"tokens_seen": 1457586176
},
{
"epoch": 0.44,
"learning_rate": 0.0002820160481444333,
"loss": 2.7006,
"theoretical_loss": 3.524994663093074,
"tokens_seen": 1457651712
},
{
"epoch": 0.44,
"learning_rate": 0.0002820060180541625,
"loss": 2.8052,
"theoretical_loss": 3.5249806131894013,
"tokens_seen": 1457717248
},
{
"epoch": 0.44,
"learning_rate": 0.0002819959879638917,
"loss": 2.7218,
"theoretical_loss": 3.524966564094224,
"tokens_seen": 1457782784
},
{
"epoch": 0.44,
"learning_rate": 0.00028198595787362087,
"loss": 2.6908,
"theoretical_loss": 3.524952515807458,
"tokens_seen": 1457848320
},
{
"epoch": 0.44,
"learning_rate": 0.00028197592778335005,
"loss": 2.9232,
"theoretical_loss": 3.5249384683290224,
"tokens_seen": 1457913856
},
{
"epoch": 0.44,
"learning_rate": 0.00028196589769307923,
"loss": 2.6262,
"theoretical_loss": 3.524924421658833,
"tokens_seen": 1457979392
},
{
"epoch": 0.44,
"learning_rate": 0.00028195586760280846,
"loss": 2.4915,
"theoretical_loss": 3.5249103757968077,
"tokens_seen": 1458044928
},
{
"epoch": 0.44,
"learning_rate": 0.0002819458375125376,
"loss": 2.8524,
"theoretical_loss": 3.5248963307428633,
"tokens_seen": 1458110464
},
{
"epoch": 0.44,
"objective/train/docs_used": 826810,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9254395961761475,
"objective/train/theoretical_loss": 3.524882286496917,
"objective/train/tokens_used": 1478636000,
"theoretical_loss": 3.524882286496917,
"tokens_seen": 1458176000
},
{
"epoch": 0.44,
"learning_rate": 0.0002819358074222668,
"loss": 2.7291,
"theoretical_loss": 3.524882286496917,
"tokens_seen": 1458176000
},
{
"epoch": 0.44,
"learning_rate": 0.000281925777331996,
"loss": 2.7154,
"theoretical_loss": 3.5248682430588865,
"tokens_seen": 1458241536
},
{
"epoch": 0.44,
"learning_rate": 0.0002819157472417252,
"loss": 2.718,
"theoretical_loss": 3.5248542004286882,
"tokens_seen": 1458307072
},
{
"epoch": 0.44,
"learning_rate": 0.00028190571715145437,
"loss": 2.4525,
"theoretical_loss": 3.52484015860624,
"tokens_seen": 1458372608
},
{
"epoch": 0.44,
"learning_rate": 0.00028189568706118355,
"loss": 2.4635,
"theoretical_loss": 3.5248261175914584,
"tokens_seen": 1458438144
},
{
"epoch": 0.44,
"learning_rate": 0.00028188565697091273,
"loss": 2.8467,
"theoretical_loss": 3.5248120773842615,
"tokens_seen": 1458503680
},
{
"epoch": 0.44,
"learning_rate": 0.00028187562688064197,
"loss": 2.5988,
"theoretical_loss": 3.5247980379845663,
"tokens_seen": 1458569216
},
{
"epoch": 0.44,
"learning_rate": 0.0002818655967903711,
"loss": 2.9422,
"theoretical_loss": 3.52478399939229,
"tokens_seen": 1458634752
},
{
"epoch": 0.44,
"learning_rate": 0.00028185556670010033,
"loss": 2.6671,
"theoretical_loss": 3.5247699616073502,
"tokens_seen": 1458700288
},
{
"epoch": 0.44,
"learning_rate": 0.00028184553660982946,
"loss": 2.6264,
"theoretical_loss": 3.5247559246296634,
"tokens_seen": 1458765824
},
{
"epoch": 0.44,
"learning_rate": 0.0002818355065195587,
"loss": 2.8233,
"theoretical_loss": 3.5247418884591473,
"tokens_seen": 1458831360
},
{
"epoch": 0.44,
"learning_rate": 0.0002818254764292879,
"loss": 2.8762,
"theoretical_loss": 3.52472785309572,
"tokens_seen": 1458896896
},
{
"epoch": 0.44,
"learning_rate": 0.00028181544633901705,
"loss": 2.8283,
"theoretical_loss": 3.524713818539297,
"tokens_seen": 1458962432
},
{
"epoch": 0.44,
"learning_rate": 0.00028180541624874624,
"loss": 2.8004,
"theoretical_loss": 3.5246997847897976,
"tokens_seen": 1459027968
},
{
"epoch": 0.44,
"learning_rate": 0.0002817953861584754,
"loss": 2.6478,
"theoretical_loss": 3.5246857518471386,
"tokens_seen": 1459093504
},
{
"epoch": 0.44,
"learning_rate": 0.0002817853560682046,
"loss": 2.7785,
"theoretical_loss": 3.524671719711237,
"tokens_seen": 1459159040
},
{
"epoch": 0.44,
"learning_rate": 0.00028177532597793383,
"loss": 2.8484,
"theoretical_loss": 3.52465768838201,
"tokens_seen": 1459224576
},
{
"epoch": 0.44,
"learning_rate": 0.00028176529588766296,
"loss": 2.7363,
"theoretical_loss": 3.5246436578593756,
"tokens_seen": 1459290112
},
{
"epoch": 0.44,
"learning_rate": 0.0002817552657973922,
"loss": 2.8804,
"theoretical_loss": 3.5246296281432508,
"tokens_seen": 1459355648
},
{
"epoch": 0.44,
"learning_rate": 0.0002817452357071214,
"loss": 2.7868,
"theoretical_loss": 3.5246155992335533,
"tokens_seen": 1459421184
},
{
"epoch": 0.44,
"learning_rate": 0.00028173520561685056,
"loss": 2.7835,
"theoretical_loss": 3.5246015711302006,
"tokens_seen": 1459486720
},
{
"epoch": 0.44,
"learning_rate": 0.00028172517552657974,
"loss": 2.6781,
"theoretical_loss": 3.5245875438331096,
"tokens_seen": 1459552256
},
{
"epoch": 0.44,
"learning_rate": 0.0002817151454363089,
"loss": 2.7683,
"theoretical_loss": 3.5245735173421986,
"tokens_seen": 1459617792
},
{
"epoch": 0.44,
"learning_rate": 0.0002817051153460381,
"loss": 2.818,
"theoretical_loss": 3.524559491657384,
"tokens_seen": 1459683328
},
{
"epoch": 0.44,
"learning_rate": 0.00028169508525576734,
"loss": 3.0611,
"theoretical_loss": 3.5245454667785845,
"tokens_seen": 1459748864
},
{
"epoch": 0.44,
"objective/train/docs_used": 827996,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.522461175918579,
"objective/train/theoretical_loss": 3.524531442705716,
"objective/train/tokens_used": 1480274400,
"theoretical_loss": 3.524531442705716,
"tokens_seen": 1459814400
},
{
"epoch": 0.44,
"learning_rate": 0.00028168505516549646,
"loss": 2.5689,
"theoretical_loss": 3.524531442705716,
"tokens_seen": 1459814400
},
{
"epoch": 0.44,
"learning_rate": 0.0002816750250752257,
"loss": 2.6952,
"theoretical_loss": 3.5245174194386983,
"tokens_seen": 1459879936
},
{
"epoch": 0.44,
"learning_rate": 0.0002816649949849548,
"loss": 2.8444,
"theoretical_loss": 3.5245033969774466,
"tokens_seen": 1459945472
},
{
"epoch": 0.44,
"learning_rate": 0.00028165496489468406,
"loss": 2.6002,
"theoretical_loss": 3.5244893753218802,
"tokens_seen": 1460011008
},
{
"epoch": 0.44,
"learning_rate": 0.00028164493480441324,
"loss": 2.7835,
"theoretical_loss": 3.5244753544719156,
"tokens_seen": 1460076544
},
{
"epoch": 0.44,
"learning_rate": 0.0002816349047141424,
"loss": 2.9149,
"theoretical_loss": 3.5244613344274707,
"tokens_seen": 1460142080
},
{
"epoch": 0.44,
"learning_rate": 0.0002816248746238716,
"loss": 2.8851,
"theoretical_loss": 3.5244473151884628,
"tokens_seen": 1460207616
},
{
"epoch": 0.44,
"learning_rate": 0.0002816148445336008,
"loss": 2.8797,
"theoretical_loss": 3.52443329675481,
"tokens_seen": 1460273152
},
{
"epoch": 0.44,
"learning_rate": 0.00028160481444332997,
"loss": 2.9093,
"theoretical_loss": 3.5244192791264295,
"tokens_seen": 1460338688
},
{
"epoch": 0.44,
"learning_rate": 0.0002815947843530592,
"loss": 2.7705,
"theoretical_loss": 3.524405262303239,
"tokens_seen": 1460404224
},
{
"epoch": 0.44,
"learning_rate": 0.00028158475426278833,
"loss": 2.8815,
"theoretical_loss": 3.524391246285156,
"tokens_seen": 1460469760
},
{
"epoch": 0.44,
"learning_rate": 0.00028157472417251756,
"loss": 2.7266,
"theoretical_loss": 3.524377231072098,
"tokens_seen": 1460535296
},
{
"epoch": 0.44,
"learning_rate": 0.00028156469408224674,
"loss": 2.6198,
"theoretical_loss": 3.5243632166639838,
"tokens_seen": 1460600832
},
{
"epoch": 0.44,
"learning_rate": 0.0002815546639919759,
"loss": 2.8849,
"theoretical_loss": 3.5243492030607295,
"tokens_seen": 1460666368
},
{
"epoch": 0.44,
"learning_rate": 0.0002815446339017051,
"loss": 2.448,
"theoretical_loss": 3.5243351902622537,
"tokens_seen": 1460731904
},
{
"epoch": 0.44,
"learning_rate": 0.0002815346038114343,
"loss": 2.6762,
"theoretical_loss": 3.5243211782684734,
"tokens_seen": 1460797440
},
{
"epoch": 0.44,
"learning_rate": 0.0002815245737211635,
"loss": 2.8437,
"theoretical_loss": 3.524307167079307,
"tokens_seen": 1460862976
},
{
"epoch": 0.44,
"learning_rate": 0.0002815145436308927,
"loss": 2.7668,
"theoretical_loss": 3.5242931566946716,
"tokens_seen": 1460928512
},
{
"epoch": 0.44,
"learning_rate": 0.0002815045135406219,
"loss": 2.6634,
"theoretical_loss": 3.524279147114486,
"tokens_seen": 1460994048
},
{
"epoch": 0.44,
"learning_rate": 0.00028149448345035107,
"loss": 2.6484,
"theoretical_loss": 3.5242651383386665,
"tokens_seen": 1461059584
},
{
"epoch": 0.44,
"learning_rate": 0.00028148445336008025,
"loss": 2.754,
"theoretical_loss": 3.5242511303671313,
"tokens_seen": 1461125120
},
{
"epoch": 0.44,
"learning_rate": 0.00028147442326980943,
"loss": 2.5412,
"theoretical_loss": 3.5242371231997986,
"tokens_seen": 1461190656
},
{
"epoch": 0.44,
"learning_rate": 0.00028146439317953866,
"loss": 2.7414,
"theoretical_loss": 3.524223116836586,
"tokens_seen": 1461256192
},
{
"epoch": 0.44,
"learning_rate": 0.0002814543630892678,
"loss": 2.4961,
"theoretical_loss": 3.524209111277411,
"tokens_seen": 1461321728
},
{
"epoch": 0.44,
"learning_rate": 0.000281444332998997,
"loss": 2.806,
"theoretical_loss": 3.5241951065221917,
"tokens_seen": 1461387264
},
{
"epoch": 0.44,
"objective/train/docs_used": 828576,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4366984367370605,
"objective/train/theoretical_loss": 3.524181102570846,
"objective/train/tokens_used": 1481912800,
"theoretical_loss": 3.524181102570846,
"tokens_seen": 1461452800
},
{
"epoch": 0.44,
"learning_rate": 0.0002814343029087262,
"loss": 2.7374,
"theoretical_loss": 3.524181102570846,
"tokens_seen": 1461452800
},
{
"epoch": 0.44,
"learning_rate": 0.0002814242728184554,
"loss": 2.7738,
"theoretical_loss": 3.524167099423291,
"tokens_seen": 1461518336
},
{
"epoch": 0.44,
"learning_rate": 0.00028141424272818457,
"loss": 2.9765,
"theoretical_loss": 3.5241530970794455,
"tokens_seen": 1461583872
},
{
"epoch": 0.44,
"learning_rate": 0.00028140421263791375,
"loss": 2.7582,
"theoretical_loss": 3.524139095539226,
"tokens_seen": 1461649408
},
{
"epoch": 0.44,
"learning_rate": 0.00028139418254764293,
"loss": 2.8063,
"theoretical_loss": 3.524125094802552,
"tokens_seen": 1461714944
},
{
"epoch": 0.44,
"learning_rate": 0.00028138415245737217,
"loss": 2.866,
"theoretical_loss": 3.5241110948693404,
"tokens_seen": 1461780480
},
{
"epoch": 0.44,
"learning_rate": 0.0002813741223671013,
"loss": 2.6453,
"theoretical_loss": 3.524097095739509,
"tokens_seen": 1461846016
},
{
"epoch": 0.44,
"learning_rate": 0.00028136409227683053,
"loss": 2.6011,
"theoretical_loss": 3.5240830974129764,
"tokens_seen": 1461911552
},
{
"epoch": 0.44,
"learning_rate": 0.00028135406218655966,
"loss": 2.733,
"theoretical_loss": 3.5240690998896596,
"tokens_seen": 1461977088
},
{
"epoch": 0.44,
"learning_rate": 0.0002813440320962889,
"loss": 2.6705,
"theoretical_loss": 3.524055103169477,
"tokens_seen": 1462042624
},
{
"epoch": 0.44,
"learning_rate": 0.0002813340020060181,
"loss": 2.8672,
"theoretical_loss": 3.524041107252346,
"tokens_seen": 1462108160
},
{
"epoch": 0.44,
"learning_rate": 0.00028132397191574725,
"loss": 2.8425,
"theoretical_loss": 3.524027112138186,
"tokens_seen": 1462173696
},
{
"epoch": 0.44,
"learning_rate": 0.00028131394182547644,
"loss": 2.5332,
"theoretical_loss": 3.524013117826913,
"tokens_seen": 1462239232
},
{
"epoch": 0.44,
"learning_rate": 0.0002813039117352056,
"loss": 2.6835,
"theoretical_loss": 3.523999124318446,
"tokens_seen": 1462304768
},
{
"epoch": 0.44,
"learning_rate": 0.0002812938816449348,
"loss": 2.7495,
"theoretical_loss": 3.5239851316127035,
"tokens_seen": 1462370304
},
{
"epoch": 0.44,
"learning_rate": 0.00028128385155466403,
"loss": 2.9577,
"theoretical_loss": 3.5239711397096025,
"tokens_seen": 1462435840
},
{
"epoch": 0.44,
"learning_rate": 0.00028127382146439316,
"loss": 2.8321,
"theoretical_loss": 3.523957148609061,
"tokens_seen": 1462501376
},
{
"epoch": 0.44,
"learning_rate": 0.0002812637913741224,
"loss": 2.6259,
"theoretical_loss": 3.5239431583109972,
"tokens_seen": 1462566912
},
{
"epoch": 0.44,
"learning_rate": 0.0002812537612838516,
"loss": 2.6485,
"theoretical_loss": 3.5239291688153296,
"tokens_seen": 1462632448
},
{
"epoch": 0.44,
"learning_rate": 0.00028124373119358076,
"loss": 2.8305,
"theoretical_loss": 3.5239151801219757,
"tokens_seen": 1462697984
},
{
"epoch": 0.44,
"learning_rate": 0.00028123370110330994,
"loss": 2.9485,
"theoretical_loss": 3.523901192230854,
"tokens_seen": 1462763520
},
{
"epoch": 0.44,
"learning_rate": 0.0002812236710130391,
"loss": 2.8801,
"theoretical_loss": 3.523887205141882,
"tokens_seen": 1462829056
},
{
"epoch": 0.44,
"learning_rate": 0.0002812136409227683,
"loss": 2.616,
"theoretical_loss": 3.5238732188549777,
"tokens_seen": 1462894592
},
{
"epoch": 0.44,
"learning_rate": 0.00028120361083249754,
"loss": 2.6114,
"theoretical_loss": 3.52385923337006,
"tokens_seen": 1462960128
},
{
"epoch": 0.44,
"learning_rate": 0.00028119358074222666,
"loss": 2.8037,
"theoretical_loss": 3.5238452486870466,
"tokens_seen": 1463025664
},
{
"epoch": 0.44,
"objective/train/docs_used": 830024,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2160961627960205,
"objective/train/theoretical_loss": 3.5238312648058554,
"objective/train/tokens_used": 1483551200,
"theoretical_loss": 3.5238312648058554,
"tokens_seen": 1463091200
},
{
"epoch": 0.44,
"learning_rate": 0.0002811835506519559,
"loss": 2.7384,
"theoretical_loss": 3.5238312648058554,
"tokens_seen": 1463091200
},
{
"epoch": 0.44,
"learning_rate": 0.000281173520561685,
"loss": 2.7401,
"theoretical_loss": 3.5238172817264046,
"tokens_seen": 1463156736
},
{
"epoch": 0.44,
"learning_rate": 0.00028116349047141426,
"loss": 2.5443,
"theoretical_loss": 3.523803299448612,
"tokens_seen": 1463222272
},
{
"epoch": 0.44,
"learning_rate": 0.00028115346038114344,
"loss": 2.6975,
"theoretical_loss": 3.5237893179723967,
"tokens_seen": 1463287808
},
{
"epoch": 0.44,
"learning_rate": 0.0002811434302908726,
"loss": 2.8031,
"theoretical_loss": 3.5237753372976757,
"tokens_seen": 1463353344
},
{
"epoch": 0.44,
"learning_rate": 0.0002811334002006018,
"loss": 2.6923,
"theoretical_loss": 3.523761357424368,
"tokens_seen": 1463418880
},
{
"epoch": 0.44,
"learning_rate": 0.000281123370110331,
"loss": 2.9628,
"theoretical_loss": 3.5237473783523914,
"tokens_seen": 1463484416
},
{
"epoch": 0.44,
"learning_rate": 0.00028111334002006017,
"loss": 2.8539,
"theoretical_loss": 3.523733400081664,
"tokens_seen": 1463549952
},
{
"epoch": 0.44,
"learning_rate": 0.0002811033099297894,
"loss": 2.4465,
"theoretical_loss": 3.5237194226121042,
"tokens_seen": 1463615488
},
{
"epoch": 0.44,
"learning_rate": 0.00028109327983951853,
"loss": 2.747,
"theoretical_loss": 3.52370544594363,
"tokens_seen": 1463681024
},
{
"epoch": 0.44,
"learning_rate": 0.00028108324974924776,
"loss": 2.7528,
"theoretical_loss": 3.5236914700761606,
"tokens_seen": 1463746560
},
{
"epoch": 0.44,
"learning_rate": 0.00028107321965897695,
"loss": 2.6155,
"theoretical_loss": 3.5236774950096126,
"tokens_seen": 1463812096
},
{
"epoch": 0.44,
"learning_rate": 0.0002810631895687061,
"loss": 2.8005,
"theoretical_loss": 3.523663520743905,
"tokens_seen": 1463877632
},
{
"epoch": 0.44,
"learning_rate": 0.0002810531594784353,
"loss": 2.7678,
"theoretical_loss": 3.5236495472789566,
"tokens_seen": 1463943168
},
{
"epoch": 0.44,
"learning_rate": 0.0002810431293881645,
"loss": 2.4125,
"theoretical_loss": 3.5236355746146852,
"tokens_seen": 1464008704
},
{
"epoch": 0.44,
"learning_rate": 0.00028103309929789367,
"loss": 2.7944,
"theoretical_loss": 3.5236216027510086,
"tokens_seen": 1464074240
},
{
"epoch": 0.44,
"learning_rate": 0.0002810230692076229,
"loss": 2.7227,
"theoretical_loss": 3.523607631687846,
"tokens_seen": 1464139776
},
{
"epoch": 0.44,
"learning_rate": 0.00028101303911735203,
"loss": 2.8579,
"theoretical_loss": 3.523593661425115,
"tokens_seen": 1464205312
},
{
"epoch": 0.44,
"learning_rate": 0.00028100300902708127,
"loss": 2.8443,
"theoretical_loss": 3.5235796919627345,
"tokens_seen": 1464270848
},
{
"epoch": 0.44,
"learning_rate": 0.0002809929789368104,
"loss": 2.9783,
"theoretical_loss": 3.5235657233006226,
"tokens_seen": 1464336384
},
{
"epoch": 0.44,
"learning_rate": 0.00028098294884653963,
"loss": 2.7819,
"theoretical_loss": 3.523551755438697,
"tokens_seen": 1464401920
},
{
"epoch": 0.44,
"learning_rate": 0.0002809729187562688,
"loss": 2.7878,
"theoretical_loss": 3.523537788376877,
"tokens_seen": 1464467456
},
{
"epoch": 0.44,
"learning_rate": 0.000280962888665998,
"loss": 2.7053,
"theoretical_loss": 3.523523822115081,
"tokens_seen": 1464532992
},
{
"epoch": 0.44,
"learning_rate": 0.00028095285857572717,
"loss": 2.5979,
"theoretical_loss": 3.5235098566532264,
"tokens_seen": 1464598528
},
{
"epoch": 0.44,
"learning_rate": 0.00028094282848545635,
"loss": 2.8607,
"theoretical_loss": 3.5234958919912316,
"tokens_seen": 1464664064
},
{
"epoch": 0.44,
"objective/train/docs_used": 830651,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4945149421691895,
"objective/train/theoretical_loss": 3.5234819281290166,
"objective/train/tokens_used": 1485189600,
"theoretical_loss": 3.5234819281290166,
"tokens_seen": 1464729600
},
{
"epoch": 0.44,
"learning_rate": 0.00028093279839518553,
"loss": 2.7503,
"theoretical_loss": 3.5234819281290166,
"tokens_seen": 1464729600
},
{
"epoch": 0.44,
"learning_rate": 0.00028092276830491477,
"loss": 2.7331,
"theoretical_loss": 3.5234679650664984,
"tokens_seen": 1464795136
},
{
"epoch": 0.44,
"learning_rate": 0.0002809127382146439,
"loss": 2.7503,
"theoretical_loss": 3.5234540028035957,
"tokens_seen": 1464860672
},
{
"epoch": 0.44,
"learning_rate": 0.00028090270812437313,
"loss": 2.7878,
"theoretical_loss": 3.523440041340227,
"tokens_seen": 1464926208
},
{
"epoch": 0.44,
"learning_rate": 0.0002808926780341023,
"loss": 2.7764,
"theoretical_loss": 3.523426080676311,
"tokens_seen": 1464991744
},
{
"epoch": 0.44,
"learning_rate": 0.0002808826479438315,
"loss": 2.5755,
"theoretical_loss": 3.523412120811766,
"tokens_seen": 1465057280
},
{
"epoch": 0.44,
"learning_rate": 0.0002808726178535607,
"loss": 2.5854,
"theoretical_loss": 3.5233981617465107,
"tokens_seen": 1465122816
},
{
"epoch": 0.44,
"learning_rate": 0.00028086258776328986,
"loss": 2.7784,
"theoretical_loss": 3.5233842034804628,
"tokens_seen": 1465188352
},
{
"epoch": 0.44,
"learning_rate": 0.00028085255767301904,
"loss": 2.6537,
"theoretical_loss": 3.523370246013541,
"tokens_seen": 1465253888
},
{
"epoch": 0.44,
"learning_rate": 0.0002808425275827483,
"loss": 2.7016,
"theoretical_loss": 3.5233562893456645,
"tokens_seen": 1465319424
},
{
"epoch": 0.44,
"learning_rate": 0.0002808324974924774,
"loss": 2.8815,
"theoretical_loss": 3.5233423334767515,
"tokens_seen": 1465384960
},
{
"epoch": 0.44,
"learning_rate": 0.00028082246740220664,
"loss": 2.8547,
"theoretical_loss": 3.5233283784067204,
"tokens_seen": 1465450496
},
{
"epoch": 0.44,
"learning_rate": 0.00028081243731193576,
"loss": 2.7318,
"theoretical_loss": 3.5233144241354903,
"tokens_seen": 1465516032
},
{
"epoch": 0.44,
"learning_rate": 0.000280802407221665,
"loss": 2.7548,
"theoretical_loss": 3.5233004706629787,
"tokens_seen": 1465581568
},
{
"epoch": 0.44,
"learning_rate": 0.0002807923771313942,
"loss": 2.6253,
"theoretical_loss": 3.5232865179891046,
"tokens_seen": 1465647104
},
{
"epoch": 0.44,
"learning_rate": 0.00028078234704112336,
"loss": 2.7259,
"theoretical_loss": 3.5232725661137874,
"tokens_seen": 1465712640
},
{
"epoch": 0.44,
"learning_rate": 0.0002807723169508526,
"loss": 2.6085,
"theoretical_loss": 3.5232586150369443,
"tokens_seen": 1465778176
},
{
"epoch": 0.44,
"learning_rate": 0.0002807622868605818,
"loss": 2.7145,
"theoretical_loss": 3.5232446647584954,
"tokens_seen": 1465843712
},
{
"epoch": 0.44,
"learning_rate": 0.00028075225677031096,
"loss": 2.8368,
"theoretical_loss": 3.5232307152783577,
"tokens_seen": 1465909248
},
{
"epoch": 0.44,
"learning_rate": 0.00028074222668004014,
"loss": 2.7335,
"theoretical_loss": 3.523216766596451,
"tokens_seen": 1465974784
},
{
"epoch": 0.44,
"learning_rate": 0.0002807321965897693,
"loss": 2.8439,
"theoretical_loss": 3.523202818712694,
"tokens_seen": 1466040320
},
{
"epoch": 0.44,
"learning_rate": 0.0002807221664994985,
"loss": 2.6083,
"theoretical_loss": 3.5231888716270046,
"tokens_seen": 1466105856
},
{
"epoch": 0.44,
"learning_rate": 0.00028071213640922774,
"loss": 2.7146,
"theoretical_loss": 3.523174925339302,
"tokens_seen": 1466171392
},
{
"epoch": 0.44,
"learning_rate": 0.00028070210631895686,
"loss": 2.9445,
"theoretical_loss": 3.5231609798495045,
"tokens_seen": 1466236928
},
{
"epoch": 0.44,
"learning_rate": 0.0002806920762286861,
"loss": 2.4936,
"theoretical_loss": 3.523147035157531,
"tokens_seen": 1466302464
},
{
"epoch": 0.44,
"objective/train/docs_used": 831840,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.9730173349380493,
"objective/train/theoretical_loss": 3.5231330912633005,
"objective/train/tokens_used": 1486828000,
"theoretical_loss": 3.5231330912633005,
"tokens_seen": 1466368000
},
{
"epoch": 0.44,
"learning_rate": 0.0002806820461384152,
"loss": 2.5983,
"theoretical_loss": 3.5231330912633005,
"tokens_seen": 1466368000
},
{
"epoch": 0.44,
"learning_rate": 0.00028067201604814446,
"loss": 2.7713,
"theoretical_loss": 3.523119148166731,
"tokens_seen": 1466433536
},
{
"epoch": 0.44,
"learning_rate": 0.00028066198595787364,
"loss": 2.8464,
"theoretical_loss": 3.5231052058677417,
"tokens_seen": 1466499072
},
{
"epoch": 0.44,
"learning_rate": 0.0002806519558676028,
"loss": 2.5492,
"theoretical_loss": 3.523091264366252,
"tokens_seen": 1466564608
},
{
"epoch": 0.44,
"learning_rate": 0.000280641925777332,
"loss": 2.6364,
"theoretical_loss": 3.523077323662179,
"tokens_seen": 1466630144
},
{
"epoch": 0.44,
"learning_rate": 0.0002806318956870612,
"loss": 2.5813,
"theoretical_loss": 3.523063383755443,
"tokens_seen": 1466695680
},
{
"epoch": 0.44,
"learning_rate": 0.00028062186559679037,
"loss": 2.6518,
"theoretical_loss": 3.523049444645962,
"tokens_seen": 1466761216
},
{
"epoch": 0.44,
"learning_rate": 0.0002806118355065196,
"loss": 2.7011,
"theoretical_loss": 3.5230355063336547,
"tokens_seen": 1466826752
},
{
"epoch": 0.44,
"learning_rate": 0.00028060180541624873,
"loss": 2.9344,
"theoretical_loss": 3.5230215688184403,
"tokens_seen": 1466892288
},
{
"epoch": 0.44,
"learning_rate": 0.00028059177532597796,
"loss": 2.8316,
"theoretical_loss": 3.5230076321002373,
"tokens_seen": 1466957824
},
{
"epoch": 0.44,
"learning_rate": 0.00028058174523570715,
"loss": 2.7117,
"theoretical_loss": 3.5229936961789647,
"tokens_seen": 1467023360
},
{
"epoch": 0.44,
"learning_rate": 0.0002805717151454363,
"loss": 2.807,
"theoretical_loss": 3.5229797610545415,
"tokens_seen": 1467088896
},
{
"epoch": 0.44,
"learning_rate": 0.0002805616850551655,
"loss": 2.7981,
"theoretical_loss": 3.522965826726886,
"tokens_seen": 1467154432
},
{
"epoch": 0.44,
"learning_rate": 0.0002805516549648947,
"loss": 2.6974,
"theoretical_loss": 3.522951893195918,
"tokens_seen": 1467219968
},
{
"epoch": 0.44,
"learning_rate": 0.00028054162487462387,
"loss": 2.5796,
"theoretical_loss": 3.5229379604615554,
"tokens_seen": 1467285504
},
{
"epoch": 0.44,
"learning_rate": 0.0002805315947843531,
"loss": 2.5888,
"theoretical_loss": 3.5229240285237173,
"tokens_seen": 1467351040
},
{
"epoch": 0.44,
"learning_rate": 0.00028052156469408223,
"loss": 2.5145,
"theoretical_loss": 3.522910097382323,
"tokens_seen": 1467416576
},
{
"epoch": 0.44,
"learning_rate": 0.00028051153460381147,
"loss": 2.611,
"theoretical_loss": 3.522896167037291,
"tokens_seen": 1467482112
},
{
"epoch": 0.44,
"learning_rate": 0.0002805015045135406,
"loss": 2.6284,
"theoretical_loss": 3.52288223748854,
"tokens_seen": 1467547648
},
{
"epoch": 0.44,
"learning_rate": 0.00028049147442326983,
"loss": 2.7133,
"theoretical_loss": 3.5228683087359896,
"tokens_seen": 1467613184
},
{
"epoch": 0.44,
"learning_rate": 0.000280481444332999,
"loss": 2.7561,
"theoretical_loss": 3.5228543807795587,
"tokens_seen": 1467678720
},
{
"epoch": 0.44,
"learning_rate": 0.0002804714142427282,
"loss": 2.4511,
"theoretical_loss": 3.5228404536191658,
"tokens_seen": 1467744256
},
{
"epoch": 0.44,
"learning_rate": 0.00028046138415245737,
"loss": 2.7242,
"theoretical_loss": 3.52282652725473,
"tokens_seen": 1467809792
},
{
"epoch": 0.44,
"learning_rate": 0.00028045135406218655,
"loss": 2.8203,
"theoretical_loss": 3.5228126016861703,
"tokens_seen": 1467875328
},
{
"epoch": 0.44,
"learning_rate": 0.00028044132397191574,
"loss": 2.6265,
"theoretical_loss": 3.522798676913405,
"tokens_seen": 1467940864
},
{
"epoch": 0.44,
"objective/train/docs_used": 832610,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.209239959716797,
"objective/train/theoretical_loss": 3.5227847529363547,
"objective/train/tokens_used": 1488466400,
"theoretical_loss": 3.5227847529363547,
"tokens_seen": 1468006400
},
{
"epoch": 0.44,
"learning_rate": 0.00028043129388164497,
"loss": 3.0612,
"theoretical_loss": 3.5227847529363547,
"tokens_seen": 1468006400
},
{
"epoch": 0.44,
"learning_rate": 0.0002804212637913741,
"loss": 2.613,
"theoretical_loss": 3.522770829754937,
"tokens_seen": 1468071936
},
{
"epoch": 0.44,
"learning_rate": 0.00028041123370110333,
"loss": 2.8475,
"theoretical_loss": 3.5227569073690717,
"tokens_seen": 1468137472
},
{
"epoch": 0.44,
"learning_rate": 0.0002804012036108325,
"loss": 2.7415,
"theoretical_loss": 3.522742985778678,
"tokens_seen": 1468203008
},
{
"epoch": 0.44,
"learning_rate": 0.0002803911735205617,
"loss": 2.4321,
"theoretical_loss": 3.5227290649836736,
"tokens_seen": 1468268544
},
{
"epoch": 0.44,
"learning_rate": 0.0002803811434302909,
"loss": 2.6987,
"theoretical_loss": 3.5227151449839793,
"tokens_seen": 1468334080
},
{
"epoch": 0.44,
"learning_rate": 0.00028037111334002006,
"loss": 2.6701,
"theoretical_loss": 3.5227012257795125,
"tokens_seen": 1468399616
},
{
"epoch": 0.44,
"learning_rate": 0.00028036108324974924,
"loss": 2.7002,
"theoretical_loss": 3.5226873073701936,
"tokens_seen": 1468465152
},
{
"epoch": 0.45,
"learning_rate": 0.0002803510531594785,
"loss": 2.7651,
"theoretical_loss": 3.522673389755941,
"tokens_seen": 1468530688
},
{
"epoch": 0.45,
"learning_rate": 0.0002803410230692076,
"loss": 2.8341,
"theoretical_loss": 3.522659472936674,
"tokens_seen": 1468596224
},
{
"epoch": 0.45,
"learning_rate": 0.00028033099297893684,
"loss": 2.7354,
"theoretical_loss": 3.522645556912312,
"tokens_seen": 1468661760
},
{
"epoch": 0.45,
"learning_rate": 0.00028032096288866596,
"loss": 2.6313,
"theoretical_loss": 3.5226316416827737,
"tokens_seen": 1468727296
},
{
"epoch": 0.45,
"learning_rate": 0.0002803109327983952,
"loss": 2.5563,
"theoretical_loss": 3.5226177272479786,
"tokens_seen": 1468792832
},
{
"epoch": 0.45,
"learning_rate": 0.0002803009027081244,
"loss": 2.9525,
"theoretical_loss": 3.5226038136078452,
"tokens_seen": 1468858368
},
{
"epoch": 0.45,
"learning_rate": 0.00028029087261785356,
"loss": 2.9583,
"theoretical_loss": 3.522589900762293,
"tokens_seen": 1468923904
},
{
"epoch": 0.45,
"learning_rate": 0.00028028084252758274,
"loss": 2.5417,
"theoretical_loss": 3.522575988711242,
"tokens_seen": 1468989440
},
{
"epoch": 0.45,
"learning_rate": 0.000280270812437312,
"loss": 2.6826,
"theoretical_loss": 3.5225620774546105,
"tokens_seen": 1469054976
},
{
"epoch": 0.45,
"learning_rate": 0.0002802607823470411,
"loss": 2.6585,
"theoretical_loss": 3.5225481669923173,
"tokens_seen": 1469120512
},
{
"epoch": 0.45,
"learning_rate": 0.00028025075225677034,
"loss": 2.7897,
"theoretical_loss": 3.522534257324282,
"tokens_seen": 1469186048
},
{
"epoch": 0.45,
"learning_rate": 0.00028024072216649947,
"loss": 2.6492,
"theoretical_loss": 3.522520348450425,
"tokens_seen": 1469251584
},
{
"epoch": 0.45,
"learning_rate": 0.0002802306920762287,
"loss": 2.8252,
"theoretical_loss": 3.5225064403706643,
"tokens_seen": 1469317120
},
{
"epoch": 0.45,
"learning_rate": 0.0002802206619859579,
"loss": 2.4647,
"theoretical_loss": 3.522492533084919,
"tokens_seen": 1469382656
},
{
"epoch": 0.45,
"learning_rate": 0.00028021063189568706,
"loss": 2.8274,
"theoretical_loss": 3.5224786265931085,
"tokens_seen": 1469448192
},
{
"epoch": 0.45,
"learning_rate": 0.00028020060180541624,
"loss": 2.7033,
"theoretical_loss": 3.5224647208951527,
"tokens_seen": 1469513728
},
{
"epoch": 0.45,
"learning_rate": 0.0002801905717151454,
"loss": 2.4854,
"theoretical_loss": 3.522450815990971,
"tokens_seen": 1469579264
},
{
"epoch": 0.45,
"objective/train/docs_used": 834187,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8797664642333984,
"objective/train/theoretical_loss": 3.5224369118804817,
"objective/train/tokens_used": 1490104800,
"theoretical_loss": 3.5224369118804817,
"tokens_seen": 1469644800
},
{
"epoch": 0.45,
"learning_rate": 0.0002801805416248746,
"loss": 2.8583,
"theoretical_loss": 3.5224369118804817,
"tokens_seen": 1469644800
},
{
"epoch": 0.45,
"learning_rate": 0.00028017051153460384,
"loss": 2.8169,
"theoretical_loss": 3.5224230085636044,
"tokens_seen": 1469710336
},
{
"epoch": 0.45,
"learning_rate": 0.00028016048144433297,
"loss": 2.8094,
"theoretical_loss": 3.522409106040259,
"tokens_seen": 1469775872
},
{
"epoch": 0.45,
"learning_rate": 0.0002801504513540622,
"loss": 2.6112,
"theoretical_loss": 3.522395204310364,
"tokens_seen": 1469841408
},
{
"epoch": 0.45,
"learning_rate": 0.00028014042126379133,
"loss": 2.7086,
"theoretical_loss": 3.5223813033738396,
"tokens_seen": 1469906944
},
{
"epoch": 0.45,
"learning_rate": 0.00028013039117352057,
"loss": 2.5905,
"theoretical_loss": 3.5223674032306045,
"tokens_seen": 1469972480
},
{
"epoch": 0.45,
"learning_rate": 0.00028012036108324975,
"loss": 2.7997,
"theoretical_loss": 3.5223535038805784,
"tokens_seen": 1470038016
},
{
"epoch": 0.45,
"learning_rate": 0.00028011033099297893,
"loss": 2.7152,
"theoretical_loss": 3.5223396053236806,
"tokens_seen": 1470103552
},
{
"epoch": 0.45,
"learning_rate": 0.0002801003009027081,
"loss": 2.8625,
"theoretical_loss": 3.5223257075598307,
"tokens_seen": 1470169088
},
{
"epoch": 0.45,
"learning_rate": 0.00028009027081243735,
"loss": 2.8143,
"theoretical_loss": 3.5223118105889473,
"tokens_seen": 1470234624
},
{
"epoch": 0.45,
"learning_rate": 0.00028008024072216647,
"loss": 2.4835,
"theoretical_loss": 3.5222979144109505,
"tokens_seen": 1470300160
},
{
"epoch": 0.45,
"learning_rate": 0.0002800702106318957,
"loss": 2.4791,
"theoretical_loss": 3.52228401902576,
"tokens_seen": 1470365696
},
{
"epoch": 0.45,
"learning_rate": 0.00028006018054162483,
"loss": 2.8509,
"theoretical_loss": 3.5222701244332946,
"tokens_seen": 1470431232
},
{
"epoch": 0.45,
"learning_rate": 0.00028005015045135407,
"loss": 2.7759,
"theoretical_loss": 3.5222562306334737,
"tokens_seen": 1470496768
},
{
"epoch": 0.45,
"learning_rate": 0.00028004012036108325,
"loss": 2.7796,
"theoretical_loss": 3.522242337626218,
"tokens_seen": 1470562304
},
{
"epoch": 0.45,
"learning_rate": 0.00028003009027081243,
"loss": 2.6936,
"theoretical_loss": 3.5222284454114448,
"tokens_seen": 1470627840
},
{
"epoch": 0.45,
"learning_rate": 0.00028002006018054167,
"loss": 2.7366,
"theoretical_loss": 3.5222145539890755,
"tokens_seen": 1470693376
},
{
"epoch": 0.45,
"learning_rate": 0.0002800100300902708,
"loss": 2.579,
"theoretical_loss": 3.522200663359029,
"tokens_seen": 1470758912
},
{
"epoch": 0.45,
"learning_rate": 0.00028000000000000003,
"loss": 2.8572,
"theoretical_loss": 3.5221867735212244,
"tokens_seen": 1470824448
},
{
"epoch": 0.45,
"learning_rate": 0.0002799899699097292,
"loss": 2.6206,
"theoretical_loss": 3.5221728844755815,
"tokens_seen": 1470889984
},
{
"epoch": 0.45,
"learning_rate": 0.0002799799398194584,
"loss": 2.6957,
"theoretical_loss": 3.5221589962220197,
"tokens_seen": 1470955520
},
{
"epoch": 0.45,
"learning_rate": 0.0002799699097291876,
"loss": 2.7695,
"theoretical_loss": 3.5221451087604594,
"tokens_seen": 1471021056
},
{
"epoch": 0.45,
"learning_rate": 0.00027995987963891675,
"loss": 2.9051,
"theoretical_loss": 3.5221312220908185,
"tokens_seen": 1471086592
},
{
"epoch": 0.45,
"learning_rate": 0.00027994984954864594,
"loss": 2.8924,
"theoretical_loss": 3.522117336213018,
"tokens_seen": 1471152128
},
{
"epoch": 0.45,
"learning_rate": 0.00027993981945837517,
"loss": 2.8103,
"theoretical_loss": 3.522103451126977,
"tokens_seen": 1471217664
},
{
"epoch": 0.45,
"objective/train/docs_used": 834912,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4597442150115967,
"objective/train/theoretical_loss": 3.522089566832615,
"objective/train/tokens_used": 1491743200,
"theoretical_loss": 3.522089566832615,
"tokens_seen": 1471283200
},
{
"epoch": 0.45,
"learning_rate": 0.0002799297893681043,
"loss": 2.5837,
"theoretical_loss": 3.522089566832615,
"tokens_seen": 1471283200
},
{
"epoch": 0.45,
"learning_rate": 0.00027991975927783353,
"loss": 2.7347,
"theoretical_loss": 3.5220756833298514,
"tokens_seen": 1471348736
},
{
"epoch": 0.45,
"learning_rate": 0.0002799097291875627,
"loss": 2.9641,
"theoretical_loss": 3.522061800618607,
"tokens_seen": 1471414272
},
{
"epoch": 0.45,
"learning_rate": 0.0002798996990972919,
"loss": 2.7043,
"theoretical_loss": 3.5220479186987994,
"tokens_seen": 1471479808
},
{
"epoch": 0.45,
"learning_rate": 0.0002798896690070211,
"loss": 2.5211,
"theoretical_loss": 3.52203403757035,
"tokens_seen": 1471545344
},
{
"epoch": 0.45,
"learning_rate": 0.00027987963891675026,
"loss": 2.6273,
"theoretical_loss": 3.5220201572331775,
"tokens_seen": 1471610880
},
{
"epoch": 0.45,
"learning_rate": 0.00027986960882647944,
"loss": 2.8245,
"theoretical_loss": 3.522006277687202,
"tokens_seen": 1471676416
},
{
"epoch": 0.45,
"learning_rate": 0.0002798595787362087,
"loss": 2.6443,
"theoretical_loss": 3.521992398932343,
"tokens_seen": 1471741952
},
{
"epoch": 0.45,
"learning_rate": 0.0002798495486459378,
"loss": 2.599,
"theoretical_loss": 3.52197852096852,
"tokens_seen": 1471807488
},
{
"epoch": 0.45,
"learning_rate": 0.00027983951855566704,
"loss": 2.7511,
"theoretical_loss": 3.521964643795653,
"tokens_seen": 1471873024
},
{
"epoch": 0.45,
"learning_rate": 0.00027982948846539616,
"loss": 2.6775,
"theoretical_loss": 3.521950767413662,
"tokens_seen": 1471938560
},
{
"epoch": 0.45,
"learning_rate": 0.0002798194583751254,
"loss": 2.6244,
"theoretical_loss": 3.521936891822466,
"tokens_seen": 1472004096
},
{
"epoch": 0.45,
"learning_rate": 0.0002798094282848546,
"loss": 2.6042,
"theoretical_loss": 3.521923017021985,
"tokens_seen": 1472069632
},
{
"epoch": 0.45,
"learning_rate": 0.00027979939819458376,
"loss": 2.5131,
"theoretical_loss": 3.521909143012139,
"tokens_seen": 1472135168
},
{
"epoch": 0.45,
"learning_rate": 0.00027978936810431294,
"loss": 2.701,
"theoretical_loss": 3.5218952697928474,
"tokens_seen": 1472200704
},
{
"epoch": 0.45,
"learning_rate": 0.0002797793380140422,
"loss": 2.6959,
"theoretical_loss": 3.5218813973640297,
"tokens_seen": 1472266240
},
{
"epoch": 0.45,
"learning_rate": 0.0002797693079237713,
"loss": 2.6492,
"theoretical_loss": 3.5218675257256065,
"tokens_seen": 1472331776
},
{
"epoch": 0.45,
"learning_rate": 0.00027975927783350054,
"loss": 2.5964,
"theoretical_loss": 3.5218536548774972,
"tokens_seen": 1472397312
},
{
"epoch": 0.45,
"learning_rate": 0.00027974924774322967,
"loss": 2.5911,
"theoretical_loss": 3.5218397848196217,
"tokens_seen": 1472462848
},
{
"epoch": 0.45,
"learning_rate": 0.0002797392176529589,
"loss": 2.7882,
"theoretical_loss": 3.521825915551899,
"tokens_seen": 1472528384
},
{
"epoch": 0.45,
"learning_rate": 0.0002797291875626881,
"loss": 2.4636,
"theoretical_loss": 3.52181204707425,
"tokens_seen": 1472593920
},
{
"epoch": 0.45,
"learning_rate": 0.00027971915747241726,
"loss": 2.7851,
"theoretical_loss": 3.5217981793865945,
"tokens_seen": 1472659456
},
{
"epoch": 0.45,
"learning_rate": 0.00027970912738214644,
"loss": 2.6309,
"theoretical_loss": 3.5217843124888515,
"tokens_seen": 1472724992
},
{
"epoch": 0.45,
"learning_rate": 0.0002796990972918756,
"loss": 2.5955,
"theoretical_loss": 3.5217704463809416,
"tokens_seen": 1472790528
},
{
"epoch": 0.45,
"learning_rate": 0.0002796890672016048,
"loss": 2.5336,
"theoretical_loss": 3.5217565810627844,
"tokens_seen": 1472856064
},
{
"epoch": 0.45,
"objective/train/docs_used": 835528,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.814811944961548,
"objective/train/theoretical_loss": 3.5217427165343,
"objective/train/tokens_used": 1493381600,
"theoretical_loss": 3.5217427165343,
"tokens_seen": 1472921600
},
{
"epoch": 0.45,
"learning_rate": 0.00027967903711133404,
"loss": 2.695,
"theoretical_loss": 3.5217427165343,
"tokens_seen": 1472921600
},
{
"epoch": 0.45,
"learning_rate": 0.00027966900702106317,
"loss": 2.8248,
"theoretical_loss": 3.521728852795407,
"tokens_seen": 1472987136
},
{
"epoch": 0.45,
"learning_rate": 0.0002796589769307924,
"loss": 2.5646,
"theoretical_loss": 3.5217149898460276,
"tokens_seen": 1473052672
},
{
"epoch": 0.45,
"learning_rate": 0.00027964894684052153,
"loss": 2.5825,
"theoretical_loss": 3.52170112768608,
"tokens_seen": 1473118208
},
{
"epoch": 0.45,
"learning_rate": 0.00027963891675025077,
"loss": 2.824,
"theoretical_loss": 3.5216872663154852,
"tokens_seen": 1473183744
},
{
"epoch": 0.45,
"learning_rate": 0.00027962888665997995,
"loss": 2.6189,
"theoretical_loss": 3.5216734057341617,
"tokens_seen": 1473249280
},
{
"epoch": 0.45,
"learning_rate": 0.00027961885656970913,
"loss": 2.547,
"theoretical_loss": 3.521659545942031,
"tokens_seen": 1473314816
},
{
"epoch": 0.45,
"learning_rate": 0.0002796088264794383,
"loss": 2.7638,
"theoretical_loss": 3.5216456869390123,
"tokens_seen": 1473380352
},
{
"epoch": 0.45,
"learning_rate": 0.00027959879638916755,
"loss": 2.8194,
"theoretical_loss": 3.5216318287250257,
"tokens_seen": 1473445888
},
{
"epoch": 0.45,
"learning_rate": 0.00027958876629889667,
"loss": 2.6555,
"theoretical_loss": 3.5216179712999907,
"tokens_seen": 1473511424
},
{
"epoch": 0.45,
"learning_rate": 0.0002795787362086259,
"loss": 2.6076,
"theoretical_loss": 3.5216041146638277,
"tokens_seen": 1473576960
},
{
"epoch": 0.45,
"learning_rate": 0.00027956870611835503,
"loss": 2.6189,
"theoretical_loss": 3.521590258816457,
"tokens_seen": 1473642496
},
{
"epoch": 0.45,
"learning_rate": 0.00027955867602808427,
"loss": 2.7623,
"theoretical_loss": 3.5215764037577983,
"tokens_seen": 1473708032
},
{
"epoch": 0.45,
"learning_rate": 0.00027954864593781345,
"loss": 2.6727,
"theoretical_loss": 3.521562549487772,
"tokens_seen": 1473773568
},
{
"epoch": 0.45,
"learning_rate": 0.00027953861584754263,
"loss": 2.4447,
"theoretical_loss": 3.521548696006298,
"tokens_seen": 1473839104
},
{
"epoch": 0.45,
"learning_rate": 0.0002795285857572718,
"loss": 2.7307,
"theoretical_loss": 3.521534843313296,
"tokens_seen": 1473904640
},
{
"epoch": 0.45,
"learning_rate": 0.000279518555667001,
"loss": 2.6957,
"theoretical_loss": 3.5215209914086865,
"tokens_seen": 1473970176
},
{
"epoch": 0.45,
"learning_rate": 0.0002795085255767302,
"loss": 2.8801,
"theoretical_loss": 3.521507140292389,
"tokens_seen": 1474035712
},
{
"epoch": 0.45,
"learning_rate": 0.0002794984954864594,
"loss": 2.4889,
"theoretical_loss": 3.521493289964324,
"tokens_seen": 1474101248
},
{
"epoch": 0.45,
"learning_rate": 0.00027948846539618854,
"loss": 2.5143,
"theoretical_loss": 3.521479440424412,
"tokens_seen": 1474166784
},
{
"epoch": 0.45,
"learning_rate": 0.0002794784353059178,
"loss": 2.7039,
"theoretical_loss": 3.5214655916725714,
"tokens_seen": 1474232320
},
{
"epoch": 0.45,
"learning_rate": 0.0002794684052156469,
"loss": 2.5857,
"theoretical_loss": 3.5214517437087247,
"tokens_seen": 1474297856
},
{
"epoch": 0.45,
"learning_rate": 0.00027945837512537614,
"loss": 2.6703,
"theoretical_loss": 3.521437896532791,
"tokens_seen": 1474363392
},
{
"epoch": 0.45,
"learning_rate": 0.0002794483450351053,
"loss": 2.7406,
"theoretical_loss": 3.52142405014469,
"tokens_seen": 1474428928
},
{
"epoch": 0.45,
"learning_rate": 0.0002794383149448345,
"loss": 2.5804,
"theoretical_loss": 3.5214102045443423,
"tokens_seen": 1474494464
},
{
"debugging/Self-BLEU-5": 0.3534733586091342,
"debugging/distinct-1-grams": 0.736222308537691,
"debugging/distinct-2-grams": 0.9376650413188702,
"debugging/entropy-1-grams": 5.598149618237377,
"debugging/entropy-2-grams": 6.390827090690173,
"debugging/length": 519.2222222222222,
"debugging/num_segments": 9,
"epoch": 0.45,
"objective/train/docs_used": 836610,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8507511615753174,
"objective/train/theoretical_loss": 3.521396359731668,
"objective/train/tokens_used": 1495020000,
"theoretical_loss": 3.521396359731668,
"tokens_seen": 1474560000
},
{
"epoch": 0.45,
"learning_rate": 0.0002794282848545637,
"loss": 2.9449,
"theoretical_loss": 3.521396359731668,
"tokens_seen": 1474560000
},
{
"epoch": 0.45,
"learning_rate": 0.0002794182547642929,
"loss": 2.6578,
"theoretical_loss": 3.5213825157065877,
"tokens_seen": 1474625536
},
{
"epoch": 0.45,
"learning_rate": 0.00027940822467402204,
"loss": 2.612,
"theoretical_loss": 3.5213686724690207,
"tokens_seen": 1474691072
},
{
"epoch": 0.45,
"learning_rate": 0.0002793981945837513,
"loss": 2.4529,
"theoretical_loss": 3.521354830018888,
"tokens_seen": 1474756608
},
{
"epoch": 0.45,
"learning_rate": 0.0002793881644934804,
"loss": 2.4685,
"theoretical_loss": 3.5213409883561093,
"tokens_seen": 1474822144
},
{
"epoch": 0.45,
"learning_rate": 0.00027937813440320964,
"loss": 2.5032,
"theoretical_loss": 3.521327147480605,
"tokens_seen": 1474887680
},
{
"epoch": 0.45,
"learning_rate": 0.0002793681043129388,
"loss": 2.695,
"theoretical_loss": 3.5213133073922958,
"tokens_seen": 1474953216
},
{
"epoch": 0.45,
"learning_rate": 0.000279358074222668,
"loss": 2.7743,
"theoretical_loss": 3.5212994680911014,
"tokens_seen": 1475018752
},
{
"epoch": 0.45,
"learning_rate": 0.0002793480441323972,
"loss": 2.7788,
"theoretical_loss": 3.5212856295769424,
"tokens_seen": 1475084288
},
{
"epoch": 0.45,
"learning_rate": 0.00027933801404212636,
"loss": 2.5684,
"theoretical_loss": 3.5212717918497383,
"tokens_seen": 1475149824
},
{
"epoch": 0.45,
"learning_rate": 0.00027932798395185554,
"loss": 2.527,
"theoretical_loss": 3.5212579549094105,
"tokens_seen": 1475215360
},
{
"epoch": 0.45,
"learning_rate": 0.0002793179538615848,
"loss": 2.6165,
"theoretical_loss": 3.5212441187558787,
"tokens_seen": 1475280896
},
{
"epoch": 0.45,
"learning_rate": 0.0002793079237713139,
"loss": 2.7696,
"theoretical_loss": 3.5212302833890634,
"tokens_seen": 1475346432
},
{
"epoch": 0.45,
"learning_rate": 0.00027929789368104314,
"loss": 2.5757,
"theoretical_loss": 3.5212164488088846,
"tokens_seen": 1475411968
},
{
"epoch": 0.45,
"learning_rate": 0.00027928786359077227,
"loss": 2.5978,
"theoretical_loss": 3.5212026150152633,
"tokens_seen": 1475477504
},
{
"epoch": 0.45,
"learning_rate": 0.0002792778335005015,
"loss": 3.0135,
"theoretical_loss": 3.521188782008119,
"tokens_seen": 1475543040
},
{
"epoch": 0.45,
"learning_rate": 0.00027926780341023074,
"loss": 2.2324,
"theoretical_loss": 3.5211749497873726,
"tokens_seen": 1475608576
},
{
"epoch": 0.45,
"learning_rate": 0.00027925777331995987,
"loss": 2.4582,
"theoretical_loss": 3.521161118352944,
"tokens_seen": 1475674112
},
{
"epoch": 0.45,
"learning_rate": 0.0002792477432296891,
"loss": 2.5089,
"theoretical_loss": 3.5211472877047543,
"tokens_seen": 1475739648
},
{
"epoch": 0.45,
"learning_rate": 0.0002792377131394183,
"loss": 2.73,
"theoretical_loss": 3.5211334578427236,
"tokens_seen": 1475805184
},
{
"epoch": 0.45,
"learning_rate": 0.00027922768304914746,
"loss": 2.4905,
"theoretical_loss": 3.5211196287667716,
"tokens_seen": 1475870720
},
{
"epoch": 0.45,
"learning_rate": 0.00027921765295887664,
"loss": 2.7907,
"theoretical_loss": 3.5211058004768203,
"tokens_seen": 1475936256
},
{
"epoch": 0.45,
"learning_rate": 0.0002792076228686058,
"loss": 2.558,
"theoretical_loss": 3.5210919729727888,
"tokens_seen": 1476001792
},
{
"epoch": 0.45,
"learning_rate": 0.000279197592778335,
"loss": 2.665,
"theoretical_loss": 3.521078146254598,
"tokens_seen": 1476067328
},
{
"epoch": 0.45,
"learning_rate": 0.00027918756268806424,
"loss": 2.5812,
"theoretical_loss": 3.521064320322168,
"tokens_seen": 1476132864
},
{
"epoch": 0.45,
"objective/train/docs_used": 837052,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5954623222351074,
"objective/train/theoretical_loss": 3.5210504951754196,
"objective/train/tokens_used": 1496658400,
"theoretical_loss": 3.5210504951754196,
"tokens_seen": 1476198400
},
{
"epoch": 0.45,
"learning_rate": 0.00027917753259779337,
"loss": 2.4089,
"theoretical_loss": 3.5210504951754196,
"tokens_seen": 1476198400
},
{
"epoch": 0.45,
"learning_rate": 0.0002791675025075226,
"loss": 2.6258,
"theoretical_loss": 3.521036670814273,
"tokens_seen": 1476263936
},
{
"epoch": 0.45,
"learning_rate": 0.00027915747241725173,
"loss": 2.7538,
"theoretical_loss": 3.5210228472386493,
"tokens_seen": 1476329472
},
{
"epoch": 0.45,
"learning_rate": 0.00027914744232698097,
"loss": 2.6118,
"theoretical_loss": 3.521009024448468,
"tokens_seen": 1476395008
},
{
"epoch": 0.45,
"learning_rate": 0.00027913741223671015,
"loss": 2.816,
"theoretical_loss": 3.5209952024436504,
"tokens_seen": 1476460544
},
{
"epoch": 0.45,
"learning_rate": 0.00027912738214643933,
"loss": 2.7838,
"theoretical_loss": 3.520981381224117,
"tokens_seen": 1476526080
},
{
"epoch": 0.45,
"learning_rate": 0.0002791173520561685,
"loss": 2.853,
"theoretical_loss": 3.520967560789788,
"tokens_seen": 1476591616
},
{
"epoch": 0.45,
"learning_rate": 0.00027910732196589775,
"loss": 2.5687,
"theoretical_loss": 3.5209537411405845,
"tokens_seen": 1476657152
},
{
"epoch": 0.45,
"learning_rate": 0.00027909729187562687,
"loss": 2.5436,
"theoretical_loss": 3.520939922276426,
"tokens_seen": 1476722688
},
{
"epoch": 0.45,
"learning_rate": 0.0002790872617853561,
"loss": 2.4236,
"theoretical_loss": 3.520926104197234,
"tokens_seen": 1476788224
},
{
"epoch": 0.45,
"learning_rate": 0.00027907723169508523,
"loss": 2.5552,
"theoretical_loss": 3.520912286902929,
"tokens_seen": 1476853760
},
{
"epoch": 0.45,
"learning_rate": 0.00027906720160481447,
"loss": 2.7566,
"theoretical_loss": 3.5208984703934316,
"tokens_seen": 1476919296
},
{
"epoch": 0.45,
"learning_rate": 0.00027905717151454365,
"loss": 2.8335,
"theoretical_loss": 3.520884654668662,
"tokens_seen": 1476984832
},
{
"epoch": 0.45,
"learning_rate": 0.00027904714142427283,
"loss": 2.7123,
"theoretical_loss": 3.5208708397285404,
"tokens_seen": 1477050368
},
{
"epoch": 0.45,
"learning_rate": 0.000279037111334002,
"loss": 2.6898,
"theoretical_loss": 3.5208570255729885,
"tokens_seen": 1477115904
},
{
"epoch": 0.45,
"learning_rate": 0.0002790270812437312,
"loss": 2.7458,
"theoretical_loss": 3.5208432122019264,
"tokens_seen": 1477181440
},
{
"epoch": 0.45,
"learning_rate": 0.0002790170511534604,
"loss": 2.687,
"theoretical_loss": 3.520829399615275,
"tokens_seen": 1477246976
},
{
"epoch": 0.45,
"learning_rate": 0.0002790070210631896,
"loss": 2.7015,
"theoretical_loss": 3.5208155878129546,
"tokens_seen": 1477312512
},
{
"epoch": 0.45,
"learning_rate": 0.00027899699097291874,
"loss": 2.8903,
"theoretical_loss": 3.5208017767948863,
"tokens_seen": 1477378048
},
{
"epoch": 0.45,
"learning_rate": 0.000278986960882648,
"loss": 2.9149,
"theoretical_loss": 3.5207879665609902,
"tokens_seen": 1477443584
},
{
"epoch": 0.45,
"learning_rate": 0.0002789769307923771,
"loss": 2.6869,
"theoretical_loss": 3.5207741571111875,
"tokens_seen": 1477509120
},
{
"epoch": 0.45,
"learning_rate": 0.00027896690070210634,
"loss": 2.6114,
"theoretical_loss": 3.520760348445399,
"tokens_seen": 1477574656
},
{
"epoch": 0.45,
"learning_rate": 0.0002789568706118355,
"loss": 2.7603,
"theoretical_loss": 3.520746540563544,
"tokens_seen": 1477640192
},
{
"epoch": 0.45,
"learning_rate": 0.0002789468405215647,
"loss": 2.869,
"theoretical_loss": 3.520732733465546,
"tokens_seen": 1477705728
},
{
"epoch": 0.45,
"learning_rate": 0.0002789368104312939,
"loss": 2.6386,
"theoretical_loss": 3.520718927151323,
"tokens_seen": 1477771264
},
{
"epoch": 0.45,
"objective/train/docs_used": 837925,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.80531644821167,
"objective/train/theoretical_loss": 3.520705121620797,
"objective/train/tokens_used": 1498296800,
"theoretical_loss": 3.520705121620797,
"tokens_seen": 1477836800
},
{
"epoch": 0.45,
"learning_rate": 0.0002789267803410231,
"loss": 2.7933,
"theoretical_loss": 3.520705121620797,
"tokens_seen": 1477836800
},
{
"epoch": 0.45,
"learning_rate": 0.00027891675025075224,
"loss": 2.6812,
"theoretical_loss": 3.5206913168738887,
"tokens_seen": 1477902336
},
{
"epoch": 0.45,
"learning_rate": 0.0002789067201604815,
"loss": 2.8027,
"theoretical_loss": 3.520677512910519,
"tokens_seen": 1477967872
},
{
"epoch": 0.45,
"learning_rate": 0.0002788966900702106,
"loss": 2.9317,
"theoretical_loss": 3.5206637097306084,
"tokens_seen": 1478033408
},
{
"epoch": 0.45,
"learning_rate": 0.00027888665997993984,
"loss": 2.7208,
"theoretical_loss": 3.520649907334078,
"tokens_seen": 1478098944
},
{
"epoch": 0.45,
"learning_rate": 0.000278876629889669,
"loss": 2.7695,
"theoretical_loss": 3.5206361057208477,
"tokens_seen": 1478164480
},
{
"epoch": 0.45,
"learning_rate": 0.0002788665997993982,
"loss": 2.513,
"theoretical_loss": 3.5206223048908396,
"tokens_seen": 1478230016
},
{
"epoch": 0.45,
"learning_rate": 0.0002788565697091274,
"loss": 2.9499,
"theoretical_loss": 3.520608504843974,
"tokens_seen": 1478295552
},
{
"epoch": 0.45,
"learning_rate": 0.00027884653961885656,
"loss": 2.887,
"theoretical_loss": 3.520594705580171,
"tokens_seen": 1478361088
},
{
"epoch": 0.45,
"learning_rate": 0.00027883650952858574,
"loss": 2.685,
"theoretical_loss": 3.520580907099353,
"tokens_seen": 1478426624
},
{
"epoch": 0.45,
"learning_rate": 0.000278826479438315,
"loss": 2.8641,
"theoretical_loss": 3.5205671094014397,
"tokens_seen": 1478492160
},
{
"epoch": 0.45,
"learning_rate": 0.0002788164493480441,
"loss": 2.7358,
"theoretical_loss": 3.5205533124863524,
"tokens_seen": 1478557696
},
{
"epoch": 0.45,
"learning_rate": 0.00027880641925777334,
"loss": 2.6395,
"theoretical_loss": 3.520539516354012,
"tokens_seen": 1478623232
},
{
"epoch": 0.45,
"learning_rate": 0.00027879638916750247,
"loss": 2.9032,
"theoretical_loss": 3.520525721004339,
"tokens_seen": 1478688768
},
{
"epoch": 0.45,
"learning_rate": 0.0002787863590772317,
"loss": 2.7999,
"theoretical_loss": 3.520511926437255,
"tokens_seen": 1478754304
},
{
"epoch": 0.45,
"learning_rate": 0.0002787763289869609,
"loss": 2.7922,
"theoretical_loss": 3.52049813265268,
"tokens_seen": 1478819840
},
{
"epoch": 0.45,
"learning_rate": 0.00027876629889669007,
"loss": 2.8563,
"theoretical_loss": 3.520484339650536,
"tokens_seen": 1478885376
},
{
"epoch": 0.45,
"learning_rate": 0.00027875626880641925,
"loss": 2.846,
"theoretical_loss": 3.520470547430743,
"tokens_seen": 1478950912
},
{
"epoch": 0.45,
"learning_rate": 0.0002787462387161485,
"loss": 2.7873,
"theoretical_loss": 3.520456755993222,
"tokens_seen": 1479016448
},
{
"epoch": 0.45,
"learning_rate": 0.0002787362086258776,
"loss": 2.712,
"theoretical_loss": 3.520442965337895,
"tokens_seen": 1479081984
},
{
"epoch": 0.45,
"learning_rate": 0.00027872617853560685,
"loss": 2.7765,
"theoretical_loss": 3.520429175464682,
"tokens_seen": 1479147520
},
{
"epoch": 0.45,
"learning_rate": 0.00027871614844533597,
"loss": 2.6589,
"theoretical_loss": 3.5204153863735046,
"tokens_seen": 1479213056
},
{
"epoch": 0.45,
"learning_rate": 0.0002787061183550652,
"loss": 2.9411,
"theoretical_loss": 3.5204015980642835,
"tokens_seen": 1479278592
},
{
"epoch": 0.45,
"learning_rate": 0.0002786960882647944,
"loss": 2.565,
"theoretical_loss": 3.5203878105369397,
"tokens_seen": 1479344128
},
{
"epoch": 0.45,
"learning_rate": 0.00027868605817452357,
"loss": 2.5308,
"theoretical_loss": 3.520374023791394,
"tokens_seen": 1479409664
},
{
"epoch": 0.45,
"objective/train/docs_used": 838738,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9416449069976807,
"objective/train/theoretical_loss": 3.520360237827568,
"objective/train/tokens_used": 1499935200,
"theoretical_loss": 3.520360237827568,
"tokens_seen": 1479475200
},
{
"epoch": 0.45,
"learning_rate": 0.00027867602808425275,
"loss": 2.848,
"theoretical_loss": 3.520360237827568,
"tokens_seen": 1479475200
},
{
"epoch": 0.45,
"learning_rate": 0.00027866599799398193,
"loss": 2.5653,
"theoretical_loss": 3.5203464526453825,
"tokens_seen": 1479540736
},
{
"epoch": 0.45,
"learning_rate": 0.0002786559679037111,
"loss": 2.5232,
"theoretical_loss": 3.5203326682447584,
"tokens_seen": 1479606272
},
{
"epoch": 0.45,
"learning_rate": 0.00027864593781344035,
"loss": 2.5874,
"theoretical_loss": 3.520318884625617,
"tokens_seen": 1479671808
},
{
"epoch": 0.45,
"learning_rate": 0.0002786359077231695,
"loss": 2.7358,
"theoretical_loss": 3.520305101787879,
"tokens_seen": 1479737344
},
{
"epoch": 0.45,
"learning_rate": 0.0002786258776328987,
"loss": 2.7574,
"theoretical_loss": 3.5202913197314665,
"tokens_seen": 1479802880
},
{
"epoch": 0.45,
"learning_rate": 0.0002786158475426279,
"loss": 2.802,
"theoretical_loss": 3.520277538456299,
"tokens_seen": 1479868416
},
{
"epoch": 0.45,
"learning_rate": 0.00027860581745235707,
"loss": 2.571,
"theoretical_loss": 3.520263757962299,
"tokens_seen": 1479933952
},
{
"epoch": 0.45,
"learning_rate": 0.00027859578736208625,
"loss": 2.8135,
"theoretical_loss": 3.520249978249387,
"tokens_seen": 1479999488
},
{
"epoch": 0.45,
"learning_rate": 0.00027858575727181544,
"loss": 2.6668,
"theoretical_loss": 3.5202361993174844,
"tokens_seen": 1480065024
},
{
"epoch": 0.45,
"learning_rate": 0.0002785757271815446,
"loss": 2.8963,
"theoretical_loss": 3.5202224211665123,
"tokens_seen": 1480130560
},
{
"epoch": 0.45,
"learning_rate": 0.00027856569709127385,
"loss": 2.6815,
"theoretical_loss": 3.5202086437963915,
"tokens_seen": 1480196096
},
{
"epoch": 0.45,
"learning_rate": 0.000278555667001003,
"loss": 2.4435,
"theoretical_loss": 3.520194867207044,
"tokens_seen": 1480261632
},
{
"epoch": 0.45,
"learning_rate": 0.0002785456369107322,
"loss": 2.4502,
"theoretical_loss": 3.52018109139839,
"tokens_seen": 1480327168
},
{
"epoch": 0.45,
"learning_rate": 0.00027853560682046134,
"loss": 2.8,
"theoretical_loss": 3.520167316370351,
"tokens_seen": 1480392704
},
{
"epoch": 0.45,
"learning_rate": 0.0002785255767301906,
"loss": 2.8177,
"theoretical_loss": 3.520153542122849,
"tokens_seen": 1480458240
},
{
"epoch": 0.45,
"learning_rate": 0.0002785155466399198,
"loss": 2.7375,
"theoretical_loss": 3.5201397686558042,
"tokens_seen": 1480523776
},
{
"epoch": 0.45,
"learning_rate": 0.00027850551654964894,
"loss": 2.5757,
"theoretical_loss": 3.5201259959691384,
"tokens_seen": 1480589312
},
{
"epoch": 0.45,
"learning_rate": 0.0002784954864593782,
"loss": 2.7746,
"theoretical_loss": 3.5201122240627725,
"tokens_seen": 1480654848
},
{
"epoch": 0.45,
"learning_rate": 0.0002784854563691073,
"loss": 2.6844,
"theoretical_loss": 3.520098452936628,
"tokens_seen": 1480720384
},
{
"epoch": 0.45,
"learning_rate": 0.00027847542627883654,
"loss": 2.7225,
"theoretical_loss": 3.5200846825906265,
"tokens_seen": 1480785920
},
{
"epoch": 0.45,
"learning_rate": 0.0002784653961885657,
"loss": 2.9973,
"theoretical_loss": 3.5200709130246883,
"tokens_seen": 1480851456
},
{
"epoch": 0.45,
"learning_rate": 0.0002784553660982949,
"loss": 2.6804,
"theoretical_loss": 3.5200571442387356,
"tokens_seen": 1480916992
},
{
"epoch": 0.45,
"learning_rate": 0.0002784453360080241,
"loss": 2.6745,
"theoretical_loss": 3.5200433762326897,
"tokens_seen": 1480982528
},
{
"epoch": 0.45,
"learning_rate": 0.0002784353059177533,
"loss": 2.7833,
"theoretical_loss": 3.520029609006471,
"tokens_seen": 1481048064
},
{
"epoch": 0.45,
"objective/train/docs_used": 839976,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5980851650238037,
"objective/train/theoretical_loss": 3.520015842560001,
"objective/train/tokens_used": 1501573600,
"theoretical_loss": 3.520015842560001,
"tokens_seen": 1481113600
},
{
"epoch": 0.45,
"learning_rate": 0.00027842527582748244,
"loss": 2.562,
"theoretical_loss": 3.520015842560001,
"tokens_seen": 1481113600
},
{
"epoch": 0.45,
"learning_rate": 0.0002784152457372117,
"loss": 2.8797,
"theoretical_loss": 3.5200020768932028,
"tokens_seen": 1481179136
},
{
"epoch": 0.45,
"learning_rate": 0.0002784052156469408,
"loss": 2.6289,
"theoretical_loss": 3.519988312005996,
"tokens_seen": 1481244672
},
{
"epoch": 0.45,
"learning_rate": 0.00027839518555667004,
"loss": 2.6994,
"theoretical_loss": 3.5199745478983013,
"tokens_seen": 1481310208
},
{
"epoch": 0.45,
"learning_rate": 0.0002783851554663992,
"loss": 2.7774,
"theoretical_loss": 3.519960784570042,
"tokens_seen": 1481375744
},
{
"epoch": 0.45,
"learning_rate": 0.0002783751253761284,
"loss": 2.7823,
"theoretical_loss": 3.519947022021139,
"tokens_seen": 1481441280
},
{
"epoch": 0.45,
"learning_rate": 0.0002783650952858576,
"loss": 2.8496,
"theoretical_loss": 3.5199332602515128,
"tokens_seen": 1481506816
},
{
"epoch": 0.45,
"learning_rate": 0.00027835506519558676,
"loss": 2.8219,
"theoretical_loss": 3.5199194992610856,
"tokens_seen": 1481572352
},
{
"epoch": 0.45,
"learning_rate": 0.00027834503510531594,
"loss": 2.7973,
"theoretical_loss": 3.5199057390497783,
"tokens_seen": 1481637888
},
{
"epoch": 0.45,
"learning_rate": 0.0002783350050150452,
"loss": 2.9173,
"theoretical_loss": 3.5198919796175128,
"tokens_seen": 1481703424
},
{
"epoch": 0.45,
"learning_rate": 0.0002783249749247743,
"loss": 2.5027,
"theoretical_loss": 3.51987822096421,
"tokens_seen": 1481768960
},
{
"epoch": 0.45,
"learning_rate": 0.00027831494483450354,
"loss": 2.652,
"theoretical_loss": 3.5198644630897915,
"tokens_seen": 1481834496
},
{
"epoch": 0.45,
"learning_rate": 0.00027830491474423267,
"loss": 2.8198,
"theoretical_loss": 3.5198507059941795,
"tokens_seen": 1481900032
},
{
"epoch": 0.45,
"learning_rate": 0.0002782948846539619,
"loss": 2.7191,
"theoretical_loss": 3.5198369496772948,
"tokens_seen": 1481965568
},
{
"epoch": 0.45,
"learning_rate": 0.0002782848545636911,
"loss": 2.8558,
"theoretical_loss": 3.5198231941390588,
"tokens_seen": 1482031104
},
{
"epoch": 0.45,
"learning_rate": 0.00027827482447342027,
"loss": 2.6271,
"theoretical_loss": 3.5198094393793933,
"tokens_seen": 1482096640
},
{
"epoch": 0.45,
"learning_rate": 0.00027826479438314945,
"loss": 2.6617,
"theoretical_loss": 3.5197956853982193,
"tokens_seen": 1482162176
},
{
"epoch": 0.45,
"learning_rate": 0.0002782547642928787,
"loss": 2.4507,
"theoretical_loss": 3.5197819321954595,
"tokens_seen": 1482227712
},
{
"epoch": 0.45,
"learning_rate": 0.0002782447342026078,
"loss": 2.5912,
"theoretical_loss": 3.5197681797710336,
"tokens_seen": 1482293248
},
{
"epoch": 0.45,
"learning_rate": 0.00027823470411233705,
"loss": 2.7931,
"theoretical_loss": 3.519754428124865,
"tokens_seen": 1482358784
},
{
"epoch": 0.45,
"learning_rate": 0.00027822467402206617,
"loss": 2.6252,
"theoretical_loss": 3.5197406772568742,
"tokens_seen": 1482424320
},
{
"epoch": 0.45,
"learning_rate": 0.0002782146439317954,
"loss": 2.6011,
"theoretical_loss": 3.519726927166983,
"tokens_seen": 1482489856
},
{
"epoch": 0.45,
"learning_rate": 0.0002782046138415246,
"loss": 2.7957,
"theoretical_loss": 3.519713177855113,
"tokens_seen": 1482555392
},
{
"epoch": 0.45,
"learning_rate": 0.00027819458375125377,
"loss": 2.6418,
"theoretical_loss": 3.519699429321186,
"tokens_seen": 1482620928
},
{
"epoch": 0.45,
"learning_rate": 0.00027818455366098295,
"loss": 2.6379,
"theoretical_loss": 3.519685681565123,
"tokens_seen": 1482686464
},
{
"epoch": 0.45,
"objective/train/docs_used": 840589,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5079002380371094,
"objective/train/theoretical_loss": 3.519671934586846,
"objective/train/tokens_used": 1503212000,
"theoretical_loss": 3.519671934586846,
"tokens_seen": 1482752000
},
{
"epoch": 0.45,
"learning_rate": 0.00027817452357071213,
"loss": 2.727,
"theoretical_loss": 3.519671934586846,
"tokens_seen": 1482752000
},
{
"epoch": 0.45,
"learning_rate": 0.0002781644934804413,
"loss": 2.46,
"theoretical_loss": 3.519658188386277,
"tokens_seen": 1482817536
},
{
"epoch": 0.45,
"learning_rate": 0.00027815446339017055,
"loss": 2.6114,
"theoretical_loss": 3.5196444429633367,
"tokens_seen": 1482883072
},
{
"epoch": 0.45,
"learning_rate": 0.0002781444332998997,
"loss": 2.5298,
"theoretical_loss": 3.5196306983179477,
"tokens_seen": 1482948608
},
{
"epoch": 0.45,
"learning_rate": 0.0002781344032096289,
"loss": 2.6922,
"theoretical_loss": 3.519616954450031,
"tokens_seen": 1483014144
},
{
"epoch": 0.45,
"learning_rate": 0.0002781243731193581,
"loss": 2.6056,
"theoretical_loss": 3.5196032113595086,
"tokens_seen": 1483079680
},
{
"epoch": 0.45,
"learning_rate": 0.0002781143430290873,
"loss": 2.6494,
"theoretical_loss": 3.519589469046302,
"tokens_seen": 1483145216
},
{
"epoch": 0.45,
"learning_rate": 0.00027810431293881645,
"loss": 2.7405,
"theoretical_loss": 3.519575727510333,
"tokens_seen": 1483210752
},
{
"epoch": 0.45,
"learning_rate": 0.00027809428284854564,
"loss": 2.634,
"theoretical_loss": 3.5195619867515235,
"tokens_seen": 1483276288
},
{
"epoch": 0.45,
"learning_rate": 0.0002780842527582748,
"loss": 2.6519,
"theoretical_loss": 3.5195482467697943,
"tokens_seen": 1483341824
},
{
"epoch": 0.45,
"learning_rate": 0.00027807422266800405,
"loss": 2.6196,
"theoretical_loss": 3.5195345075650684,
"tokens_seen": 1483407360
},
{
"epoch": 0.45,
"learning_rate": 0.0002780641925777332,
"loss": 2.9117,
"theoretical_loss": 3.5195207691372667,
"tokens_seen": 1483472896
},
{
"epoch": 0.45,
"learning_rate": 0.0002780541624874624,
"loss": 2.847,
"theoretical_loss": 3.5195070314863113,
"tokens_seen": 1483538432
},
{
"epoch": 0.45,
"learning_rate": 0.00027804413239719154,
"loss": 2.6208,
"theoretical_loss": 3.519493294612124,
"tokens_seen": 1483603968
},
{
"epoch": 0.45,
"learning_rate": 0.0002780341023069208,
"loss": 2.9106,
"theoretical_loss": 3.519479558514626,
"tokens_seen": 1483669504
},
{
"epoch": 0.45,
"learning_rate": 0.00027802407221664996,
"loss": 2.9355,
"theoretical_loss": 3.51946582319374,
"tokens_seen": 1483735040
},
{
"epoch": 0.45,
"learning_rate": 0.00027801404212637914,
"loss": 2.8249,
"theoretical_loss": 3.519452088649387,
"tokens_seen": 1483800576
},
{
"epoch": 0.45,
"learning_rate": 0.0002780040120361083,
"loss": 2.8558,
"theoretical_loss": 3.5194383548814896,
"tokens_seen": 1483866112
},
{
"epoch": 0.45,
"learning_rate": 0.0002779939819458375,
"loss": 2.6152,
"theoretical_loss": 3.5194246218899683,
"tokens_seen": 1483931648
},
{
"epoch": 0.45,
"learning_rate": 0.0002779839518555667,
"loss": 2.596,
"theoretical_loss": 3.519410889674746,
"tokens_seen": 1483997184
},
{
"epoch": 0.45,
"learning_rate": 0.0002779739217652959,
"loss": 2.6429,
"theoretical_loss": 3.519397158235745,
"tokens_seen": 1484062720
},
{
"epoch": 0.45,
"learning_rate": 0.00027796389167502504,
"loss": 2.7408,
"theoretical_loss": 3.5193834275728855,
"tokens_seen": 1484128256
},
{
"epoch": 0.45,
"learning_rate": 0.0002779538615847543,
"loss": 2.8187,
"theoretical_loss": 3.5193696976860904,
"tokens_seen": 1484193792
},
{
"epoch": 0.45,
"learning_rate": 0.00027794383149448346,
"loss": 2.7782,
"theoretical_loss": 3.519355968575282,
"tokens_seen": 1484259328
},
{
"epoch": 0.45,
"learning_rate": 0.00027793380140421264,
"loss": 2.6014,
"theoretical_loss": 3.5193422402403813,
"tokens_seen": 1484324864
},
{
"epoch": 0.45,
"objective/train/docs_used": 841910,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.920531749725342,
"objective/train/theoretical_loss": 3.5193285126813105,
"objective/train/tokens_used": 1504850400,
"theoretical_loss": 3.5193285126813105,
"tokens_seen": 1484390400
},
{
"epoch": 0.45,
"learning_rate": 0.0002779237713139418,
"loss": 2.8336,
"theoretical_loss": 3.5193285126813105,
"tokens_seen": 1484390400
},
{
"epoch": 0.45,
"learning_rate": 0.000277913741223671,
"loss": 2.4737,
"theoretical_loss": 3.5193147858979916,
"tokens_seen": 1484455936
},
{
"epoch": 0.45,
"learning_rate": 0.0002779037111334002,
"loss": 2.665,
"theoretical_loss": 3.5193010598903465,
"tokens_seen": 1484521472
},
{
"epoch": 0.45,
"learning_rate": 0.0002778936810431294,
"loss": 2.838,
"theoretical_loss": 3.519287334658297,
"tokens_seen": 1484587008
},
{
"epoch": 0.45,
"learning_rate": 0.00027788365095285855,
"loss": 2.6619,
"theoretical_loss": 3.5192736102017648,
"tokens_seen": 1484652544
},
{
"epoch": 0.45,
"learning_rate": 0.0002778736208625878,
"loss": 2.5163,
"theoretical_loss": 3.519259886520673,
"tokens_seen": 1484718080
},
{
"epoch": 0.45,
"learning_rate": 0.0002778635907723169,
"loss": 2.783,
"theoretical_loss": 3.519246163614942,
"tokens_seen": 1484783616
},
{
"epoch": 0.45,
"learning_rate": 0.00027785356068204614,
"loss": 2.8531,
"theoretical_loss": 3.519232441484495,
"tokens_seen": 1484849152
},
{
"epoch": 0.45,
"learning_rate": 0.0002778435305917753,
"loss": 2.8629,
"theoretical_loss": 3.519218720129253,
"tokens_seen": 1484914688
},
{
"epoch": 0.45,
"learning_rate": 0.0002778335005015045,
"loss": 2.6835,
"theoretical_loss": 3.519204999549139,
"tokens_seen": 1484980224
},
{
"epoch": 0.45,
"learning_rate": 0.0002778234704112337,
"loss": 2.2584,
"theoretical_loss": 3.519191279744074,
"tokens_seen": 1485045760
},
{
"epoch": 0.45,
"learning_rate": 0.00027781344032096287,
"loss": 2.4788,
"theoretical_loss": 3.519177560713981,
"tokens_seen": 1485111296
},
{
"epoch": 0.45,
"learning_rate": 0.00027780341023069205,
"loss": 2.6748,
"theoretical_loss": 3.5191638424587808,
"tokens_seen": 1485176832
},
{
"epoch": 0.45,
"learning_rate": 0.0002777933801404213,
"loss": 2.646,
"theoretical_loss": 3.519150124978397,
"tokens_seen": 1485242368
},
{
"epoch": 0.45,
"learning_rate": 0.0002777833500501504,
"loss": 2.3812,
"theoretical_loss": 3.5191364082727503,
"tokens_seen": 1485307904
},
{
"epoch": 0.45,
"learning_rate": 0.00027777331995987965,
"loss": 2.4955,
"theoretical_loss": 3.519122692341764,
"tokens_seen": 1485373440
},
{
"epoch": 0.45,
"learning_rate": 0.0002777632898696089,
"loss": 2.6461,
"theoretical_loss": 3.5191089771853585,
"tokens_seen": 1485438976
},
{
"epoch": 0.45,
"learning_rate": 0.000277753259779338,
"loss": 2.6055,
"theoretical_loss": 3.5190952628034573,
"tokens_seen": 1485504512
},
{
"epoch": 0.45,
"learning_rate": 0.00027774322968906725,
"loss": 2.6302,
"theoretical_loss": 3.519081549195982,
"tokens_seen": 1485570048
},
{
"epoch": 0.45,
"learning_rate": 0.00027773319959879637,
"loss": 2.602,
"theoretical_loss": 3.519067836362855,
"tokens_seen": 1485635584
},
{
"epoch": 0.45,
"learning_rate": 0.0002777231695085256,
"loss": 2.71,
"theoretical_loss": 3.519054124303998,
"tokens_seen": 1485701120
},
{
"epoch": 0.45,
"learning_rate": 0.0002777131394182548,
"loss": 2.7289,
"theoretical_loss": 3.519040413019334,
"tokens_seen": 1485766656
},
{
"epoch": 0.45,
"learning_rate": 0.00027770310932798397,
"loss": 2.7563,
"theoretical_loss": 3.519026702508783,
"tokens_seen": 1485832192
},
{
"epoch": 0.45,
"learning_rate": 0.00027769307923771315,
"loss": 2.5461,
"theoretical_loss": 3.5190129927722698,
"tokens_seen": 1485897728
},
{
"epoch": 0.45,
"learning_rate": 0.00027768304914744233,
"loss": 2.8839,
"theoretical_loss": 3.518999283809715,
"tokens_seen": 1485963264
},
{
"epoch": 0.45,
"objective/train/docs_used": 842517,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5579888820648193,
"objective/train/theoretical_loss": 3.5189855756210413,
"objective/train/tokens_used": 1506488800,
"theoretical_loss": 3.5189855756210413,
"tokens_seen": 1486028800
},
{
"epoch": 0.45,
"learning_rate": 0.0002776730190571715,
"loss": 2.5876,
"theoretical_loss": 3.5189855756210413,
"tokens_seen": 1486028800
},
{
"epoch": 0.45,
"learning_rate": 0.00027766298896690075,
"loss": 2.7306,
"theoretical_loss": 3.5189718682061706,
"tokens_seen": 1486094336
},
{
"epoch": 0.45,
"learning_rate": 0.0002776529588766299,
"loss": 2.7075,
"theoretical_loss": 3.518958161565025,
"tokens_seen": 1486159872
},
{
"epoch": 0.45,
"learning_rate": 0.0002776429287863591,
"loss": 2.5385,
"theoretical_loss": 3.518944455697527,
"tokens_seen": 1486225408
},
{
"epoch": 0.45,
"learning_rate": 0.0002776328986960883,
"loss": 2.596,
"theoretical_loss": 3.5189307506035994,
"tokens_seen": 1486290944
},
{
"epoch": 0.45,
"learning_rate": 0.0002776228686058175,
"loss": 2.5657,
"theoretical_loss": 3.518917046283163,
"tokens_seen": 1486356480
},
{
"epoch": 0.45,
"learning_rate": 0.00027761283851554665,
"loss": 2.6798,
"theoretical_loss": 3.5189033427361416,
"tokens_seen": 1486422016
},
{
"epoch": 0.45,
"learning_rate": 0.00027760280842527584,
"loss": 2.5493,
"theoretical_loss": 3.518889639962456,
"tokens_seen": 1486487552
},
{
"epoch": 0.45,
"learning_rate": 0.000277592778335005,
"loss": 2.6714,
"theoretical_loss": 3.51887593796203,
"tokens_seen": 1486553088
},
{
"epoch": 0.45,
"learning_rate": 0.00027758274824473425,
"loss": 2.7572,
"theoretical_loss": 3.5188622367347846,
"tokens_seen": 1486618624
},
{
"epoch": 0.45,
"learning_rate": 0.0002775727181544634,
"loss": 2.7044,
"theoretical_loss": 3.5188485362806423,
"tokens_seen": 1486684160
},
{
"epoch": 0.45,
"learning_rate": 0.0002775626880641926,
"loss": 2.6539,
"theoretical_loss": 3.5188348365995257,
"tokens_seen": 1486749696
},
{
"epoch": 0.45,
"learning_rate": 0.00027755265797392174,
"loss": 2.7874,
"theoretical_loss": 3.5188211376913574,
"tokens_seen": 1486815232
},
{
"epoch": 0.45,
"learning_rate": 0.000277542627883651,
"loss": 2.7924,
"theoretical_loss": 3.518807439556059,
"tokens_seen": 1486880768
},
{
"epoch": 0.45,
"learning_rate": 0.00027753259779338016,
"loss": 2.4949,
"theoretical_loss": 3.5187937421935533,
"tokens_seen": 1486946304
},
{
"epoch": 0.45,
"learning_rate": 0.00027752256770310934,
"loss": 2.5117,
"theoretical_loss": 3.518780045603763,
"tokens_seen": 1487011840
},
{
"epoch": 0.45,
"learning_rate": 0.0002775125376128385,
"loss": 2.7231,
"theoretical_loss": 3.5187663497866097,
"tokens_seen": 1487077376
},
{
"epoch": 0.45,
"learning_rate": 0.0002775025075225677,
"loss": 2.9194,
"theoretical_loss": 3.518752654742016,
"tokens_seen": 1487142912
},
{
"epoch": 0.45,
"learning_rate": 0.0002774924774322969,
"loss": 2.4597,
"theoretical_loss": 3.5187389604699044,
"tokens_seen": 1487208448
},
{
"epoch": 0.45,
"learning_rate": 0.0002774824473420261,
"loss": 2.6747,
"theoretical_loss": 3.518725266970197,
"tokens_seen": 1487273984
},
{
"epoch": 0.45,
"learning_rate": 0.00027747241725175524,
"loss": 2.7906,
"theoretical_loss": 3.5187115742428166,
"tokens_seen": 1487339520
},
{
"epoch": 0.45,
"learning_rate": 0.0002774623871614845,
"loss": 2.698,
"theoretical_loss": 3.5186978822876855,
"tokens_seen": 1487405056
},
{
"epoch": 0.45,
"learning_rate": 0.00027745235707121366,
"loss": 2.5835,
"theoretical_loss": 3.518684191104726,
"tokens_seen": 1487470592
},
{
"epoch": 0.45,
"learning_rate": 0.00027744232698094284,
"loss": 2.4448,
"theoretical_loss": 3.518670500693861,
"tokens_seen": 1487536128
},
{
"epoch": 0.45,
"learning_rate": 0.000277432296890672,
"loss": 2.5487,
"theoretical_loss": 3.518656811055012,
"tokens_seen": 1487601664
},
{
"epoch": 0.45,
"objective/train/docs_used": 843832,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7092649936676025,
"objective/train/theoretical_loss": 3.5186431221881023,
"objective/train/tokens_used": 1508127200,
"theoretical_loss": 3.5186431221881023,
"tokens_seen": 1487667200
},
{
"epoch": 0.45,
"learning_rate": 0.0002774222668004012,
"loss": 2.5233,
"theoretical_loss": 3.5186431221881023,
"tokens_seen": 1487667200
},
{
"epoch": 0.45,
"learning_rate": 0.0002774122367101304,
"loss": 2.6707,
"theoretical_loss": 3.5186294340930537,
"tokens_seen": 1487732736
},
{
"epoch": 0.45,
"learning_rate": 0.0002774022066198596,
"loss": 2.8334,
"theoretical_loss": 3.5186157467697896,
"tokens_seen": 1487798272
},
{
"epoch": 0.45,
"learning_rate": 0.00027739217652958875,
"loss": 2.5909,
"theoretical_loss": 3.5186020602182317,
"tokens_seen": 1487863808
},
{
"epoch": 0.45,
"learning_rate": 0.000277382146439318,
"loss": 2.526,
"theoretical_loss": 3.518588374438303,
"tokens_seen": 1487929344
},
{
"epoch": 0.45,
"learning_rate": 0.0002773721163490471,
"loss": 2.6173,
"theoretical_loss": 3.5185746894299257,
"tokens_seen": 1487994880
},
{
"epoch": 0.45,
"learning_rate": 0.00027736208625877634,
"loss": 2.6602,
"theoretical_loss": 3.518561005193022,
"tokens_seen": 1488060416
},
{
"epoch": 0.45,
"learning_rate": 0.0002773520561685055,
"loss": 2.6722,
"theoretical_loss": 3.5185473217275147,
"tokens_seen": 1488125952
},
{
"epoch": 0.45,
"learning_rate": 0.0002773420260782347,
"loss": 2.5965,
"theoretical_loss": 3.5185336390333273,
"tokens_seen": 1488191488
},
{
"epoch": 0.45,
"learning_rate": 0.0002773319959879639,
"loss": 2.6935,
"theoretical_loss": 3.5185199571103807,
"tokens_seen": 1488257024
},
{
"epoch": 0.45,
"learning_rate": 0.00027732196589769307,
"loss": 2.7184,
"theoretical_loss": 3.518506275958599,
"tokens_seen": 1488322560
},
{
"epoch": 0.45,
"learning_rate": 0.00027731193580742225,
"loss": 2.702,
"theoretical_loss": 3.5184925955779036,
"tokens_seen": 1488388096
},
{
"epoch": 0.45,
"learning_rate": 0.0002773019057171515,
"loss": 2.7448,
"theoretical_loss": 3.5184789159682177,
"tokens_seen": 1488453632
},
{
"epoch": 0.45,
"learning_rate": 0.0002772918756268806,
"loss": 2.8425,
"theoretical_loss": 3.518465237129464,
"tokens_seen": 1488519168
},
{
"epoch": 0.45,
"learning_rate": 0.00027728184553660985,
"loss": 2.6467,
"theoretical_loss": 3.518451559061565,
"tokens_seen": 1488584704
},
{
"epoch": 0.45,
"learning_rate": 0.00027727181544633903,
"loss": 2.621,
"theoretical_loss": 3.5184378817644424,
"tokens_seen": 1488650240
},
{
"epoch": 0.45,
"learning_rate": 0.0002772617853560682,
"loss": 2.6739,
"theoretical_loss": 3.5184242052380204,
"tokens_seen": 1488715776
},
{
"epoch": 0.45,
"learning_rate": 0.0002772517552657974,
"loss": 2.9045,
"theoretical_loss": 3.5184105294822206,
"tokens_seen": 1488781312
},
{
"epoch": 0.45,
"learning_rate": 0.00027724172517552657,
"loss": 2.6065,
"theoretical_loss": 3.5183968544969657,
"tokens_seen": 1488846848
},
{
"epoch": 0.45,
"learning_rate": 0.00027723169508525575,
"loss": 2.6008,
"theoretical_loss": 3.5183831802821794,
"tokens_seen": 1488912384
},
{
"epoch": 0.45,
"learning_rate": 0.000277221664994985,
"loss": 2.5943,
"theoretical_loss": 3.518369506837783,
"tokens_seen": 1488977920
},
{
"epoch": 0.45,
"learning_rate": 0.0002772116349047141,
"loss": 2.7132,
"theoretical_loss": 3.5183558341637,
"tokens_seen": 1489043456
},
{
"epoch": 0.45,
"learning_rate": 0.00027720160481444335,
"loss": 2.9433,
"theoretical_loss": 3.5183421622598527,
"tokens_seen": 1489108992
},
{
"epoch": 0.45,
"learning_rate": 0.0002771915747241725,
"loss": 2.6519,
"theoretical_loss": 3.518328491126164,
"tokens_seen": 1489174528
},
{
"epoch": 0.45,
"learning_rate": 0.0002771815446339017,
"loss": 2.6017,
"theoretical_loss": 3.5183148207625567,
"tokens_seen": 1489240064
},
{
"epoch": 0.45,
"objective/train/docs_used": 844519,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.601651668548584,
"objective/train/theoretical_loss": 3.518301151168954,
"objective/train/tokens_used": 1509765600,
"theoretical_loss": 3.518301151168954,
"tokens_seen": 1489305600
},
{
"epoch": 0.45,
"learning_rate": 0.0002771715145436309,
"loss": 2.6583,
"theoretical_loss": 3.518301151168954,
"tokens_seen": 1489305600
},
{
"epoch": 0.45,
"learning_rate": 0.0002771614844533601,
"loss": 2.7031,
"theoretical_loss": 3.5182874823452774,
"tokens_seen": 1489371136
},
{
"epoch": 0.45,
"learning_rate": 0.00027715145436308926,
"loss": 2.6006,
"theoretical_loss": 3.518273814291451,
"tokens_seen": 1489436672
},
{
"epoch": 0.45,
"learning_rate": 0.0002771414242728185,
"loss": 2.583,
"theoretical_loss": 3.518260147007396,
"tokens_seen": 1489502208
},
{
"epoch": 0.45,
"learning_rate": 0.0002771313941825476,
"loss": 2.7515,
"theoretical_loss": 3.5182464804930373,
"tokens_seen": 1489567744
},
{
"epoch": 0.45,
"learning_rate": 0.00027712136409227685,
"loss": 2.7054,
"theoretical_loss": 3.5182328147482957,
"tokens_seen": 1489633280
},
{
"epoch": 0.45,
"learning_rate": 0.000277111334002006,
"loss": 2.5475,
"theoretical_loss": 3.5182191497730955,
"tokens_seen": 1489698816
},
{
"epoch": 0.45,
"learning_rate": 0.0002771013039117352,
"loss": 2.7408,
"theoretical_loss": 3.5182054855673583,
"tokens_seen": 1489764352
},
{
"epoch": 0.45,
"learning_rate": 0.0002770912738214644,
"loss": 2.7602,
"theoretical_loss": 3.5181918221310076,
"tokens_seen": 1489829888
},
{
"epoch": 0.45,
"learning_rate": 0.0002770812437311936,
"loss": 2.8666,
"theoretical_loss": 3.5181781594639663,
"tokens_seen": 1489895424
},
{
"epoch": 0.45,
"learning_rate": 0.00027707121364092276,
"loss": 2.7151,
"theoretical_loss": 3.518164497566157,
"tokens_seen": 1489960960
},
{
"epoch": 0.45,
"learning_rate": 0.00027706118355065194,
"loss": 2.5637,
"theoretical_loss": 3.5181508364375023,
"tokens_seen": 1490026496
},
{
"epoch": 0.45,
"learning_rate": 0.0002770511534603811,
"loss": 2.8446,
"theoretical_loss": 3.518137176077926,
"tokens_seen": 1490092032
},
{
"epoch": 0.45,
"learning_rate": 0.00027704112337011036,
"loss": 2.7972,
"theoretical_loss": 3.51812351648735,
"tokens_seen": 1490157568
},
{
"epoch": 0.45,
"learning_rate": 0.0002770310932798395,
"loss": 2.5987,
"theoretical_loss": 3.518109857665697,
"tokens_seen": 1490223104
},
{
"epoch": 0.45,
"learning_rate": 0.0002770210631895687,
"loss": 2.7814,
"theoretical_loss": 3.518096199612891,
"tokens_seen": 1490288640
},
{
"epoch": 0.45,
"learning_rate": 0.0002770110330992979,
"loss": 2.6677,
"theoretical_loss": 3.5180825423288544,
"tokens_seen": 1490354176
},
{
"epoch": 0.45,
"learning_rate": 0.0002770010030090271,
"loss": 2.6737,
"theoretical_loss": 3.51806888581351,
"tokens_seen": 1490419712
},
{
"epoch": 0.45,
"learning_rate": 0.0002769909729187563,
"loss": 2.6941,
"theoretical_loss": 3.518055230066781,
"tokens_seen": 1490485248
},
{
"epoch": 0.45,
"learning_rate": 0.00027698094282848544,
"loss": 2.5956,
"theoretical_loss": 3.51804157508859,
"tokens_seen": 1490550784
},
{
"epoch": 0.45,
"learning_rate": 0.0002769709127382147,
"loss": 2.5648,
"theoretical_loss": 3.5180279208788603,
"tokens_seen": 1490616320
},
{
"epoch": 0.45,
"learning_rate": 0.00027696088264794386,
"loss": 2.6777,
"theoretical_loss": 3.5180142674375148,
"tokens_seen": 1490681856
},
{
"epoch": 0.45,
"learning_rate": 0.00027695085255767304,
"loss": 2.7308,
"theoretical_loss": 3.5180006147644756,
"tokens_seen": 1490747392
},
{
"epoch": 0.45,
"learning_rate": 0.0002769408224674022,
"loss": 2.4895,
"theoretical_loss": 3.517986962859667,
"tokens_seen": 1490812928
},
{
"epoch": 0.45,
"learning_rate": 0.0002769307923771314,
"loss": 2.6355,
"theoretical_loss": 3.5179733117230123,
"tokens_seen": 1490878464
},
{
"epoch": 0.45,
"objective/train/docs_used": 845874,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.340980291366577,
"objective/train/theoretical_loss": 3.517959661354433,
"objective/train/tokens_used": 1511404000,
"theoretical_loss": 3.517959661354433,
"tokens_seen": 1490944000
},
{
"epoch": 0.45,
"learning_rate": 0.0002769207622868606,
"loss": 2.6659,
"theoretical_loss": 3.517959661354433,
"tokens_seen": 1490944000
},
{
"epoch": 0.45,
"learning_rate": 0.0002769107321965898,
"loss": 2.5856,
"theoretical_loss": 3.5179460117538524,
"tokens_seen": 1491009536
},
{
"epoch": 0.45,
"learning_rate": 0.00027690070210631895,
"loss": 2.6164,
"theoretical_loss": 3.5179323629211945,
"tokens_seen": 1491075072
},
{
"epoch": 0.45,
"learning_rate": 0.0002768906720160482,
"loss": 2.5057,
"theoretical_loss": 3.517918714856382,
"tokens_seen": 1491140608
},
{
"epoch": 0.45,
"learning_rate": 0.0002768806419257773,
"loss": 2.8337,
"theoretical_loss": 3.5179050675593375,
"tokens_seen": 1491206144
},
{
"epoch": 0.45,
"learning_rate": 0.00027687061183550655,
"loss": 2.716,
"theoretical_loss": 3.517891421029984,
"tokens_seen": 1491271680
},
{
"epoch": 0.45,
"learning_rate": 0.0002768605817452357,
"loss": 2.365,
"theoretical_loss": 3.5178777752682455,
"tokens_seen": 1491337216
},
{
"epoch": 0.45,
"learning_rate": 0.0002768505516549649,
"loss": 2.6203,
"theoretical_loss": 3.5178641302740443,
"tokens_seen": 1491402752
},
{
"epoch": 0.45,
"learning_rate": 0.0002768405215646941,
"loss": 2.7501,
"theoretical_loss": 3.517850486047304,
"tokens_seen": 1491468288
},
{
"epoch": 0.45,
"learning_rate": 0.00027683049147442327,
"loss": 2.6383,
"theoretical_loss": 3.517836842587947,
"tokens_seen": 1491533824
},
{
"epoch": 0.45,
"learning_rate": 0.00027682046138415245,
"loss": 2.7941,
"theoretical_loss": 3.517823199895897,
"tokens_seen": 1491599360
},
{
"epoch": 0.45,
"learning_rate": 0.0002768104312938817,
"loss": 2.5831,
"theoretical_loss": 3.517809557971077,
"tokens_seen": 1491664896
},
{
"epoch": 0.45,
"learning_rate": 0.0002768004012036108,
"loss": 2.4848,
"theoretical_loss": 3.51779591681341,
"tokens_seen": 1491730432
},
{
"epoch": 0.45,
"learning_rate": 0.00027679037111334005,
"loss": 2.9691,
"theoretical_loss": 3.5177822764228193,
"tokens_seen": 1491795968
},
{
"epoch": 0.45,
"learning_rate": 0.00027678034102306923,
"loss": 2.6805,
"theoretical_loss": 3.5177686367992287,
"tokens_seen": 1491861504
},
{
"epoch": 0.45,
"learning_rate": 0.0002767703109327984,
"loss": 2.6693,
"theoretical_loss": 3.51775499794256,
"tokens_seen": 1491927040
},
{
"epoch": 0.45,
"learning_rate": 0.0002767602808425276,
"loss": 2.5418,
"theoretical_loss": 3.517741359852738,
"tokens_seen": 1491992576
},
{
"epoch": 0.45,
"learning_rate": 0.00027675025075225677,
"loss": 2.6598,
"theoretical_loss": 3.517727722529684,
"tokens_seen": 1492058112
},
{
"epoch": 0.45,
"learning_rate": 0.00027674022066198595,
"loss": 2.8101,
"theoretical_loss": 3.517714085973323,
"tokens_seen": 1492123648
},
{
"epoch": 0.45,
"learning_rate": 0.0002767301905717152,
"loss": 2.599,
"theoretical_loss": 3.517700450183577,
"tokens_seen": 1492189184
},
{
"epoch": 0.45,
"learning_rate": 0.0002767201604814443,
"loss": 2.6755,
"theoretical_loss": 3.5176868151603697,
"tokens_seen": 1492254720
},
{
"epoch": 0.45,
"learning_rate": 0.00027671013039117355,
"loss": 2.8115,
"theoretical_loss": 3.5176731809036252,
"tokens_seen": 1492320256
},
{
"epoch": 0.45,
"learning_rate": 0.0002767001003009027,
"loss": 2.7003,
"theoretical_loss": 3.517659547413265,
"tokens_seen": 1492385792
},
{
"epoch": 0.45,
"learning_rate": 0.0002766900702106319,
"loss": 2.4576,
"theoretical_loss": 3.517645914689214,
"tokens_seen": 1492451328
},
{
"epoch": 0.45,
"learning_rate": 0.0002766800401203611,
"loss": 2.4923,
"theoretical_loss": 3.5176322827313946,
"tokens_seen": 1492516864
},
{
"epoch": 0.45,
"objective/train/docs_used": 846561,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2721633911132812,
"objective/train/theoretical_loss": 3.5176186515397303,
"objective/train/tokens_used": 1513042400,
"theoretical_loss": 3.5176186515397303,
"tokens_seen": 1492582400
},
{
"epoch": 0.45,
"learning_rate": 0.0002766700100300903,
"loss": 2.6417,
"theoretical_loss": 3.5176186515397303,
"tokens_seen": 1492582400
},
{
"epoch": 0.45,
"learning_rate": 0.00027665997993981946,
"loss": 2.5733,
"theoretical_loss": 3.517605021114144,
"tokens_seen": 1492647936
},
{
"epoch": 0.45,
"learning_rate": 0.0002766499498495487,
"loss": 2.7199,
"theoretical_loss": 3.51759139145456,
"tokens_seen": 1492713472
},
{
"epoch": 0.45,
"learning_rate": 0.0002766399197592778,
"loss": 2.7722,
"theoretical_loss": 3.5175777625609004,
"tokens_seen": 1492779008
},
{
"epoch": 0.45,
"learning_rate": 0.00027662988966900705,
"loss": 2.7518,
"theoretical_loss": 3.5175641344330892,
"tokens_seen": 1492844544
},
{
"epoch": 0.45,
"learning_rate": 0.0002766198595787362,
"loss": 2.8116,
"theoretical_loss": 3.51755050707105,
"tokens_seen": 1492910080
},
{
"epoch": 0.45,
"learning_rate": 0.0002766098294884654,
"loss": 2.5341,
"theoretical_loss": 3.517536880474706,
"tokens_seen": 1492975616
},
{
"epoch": 0.45,
"learning_rate": 0.0002765997993981946,
"loss": 2.6646,
"theoretical_loss": 3.51752325464398,
"tokens_seen": 1493041152
},
{
"epoch": 0.45,
"learning_rate": 0.0002765897693079238,
"loss": 2.7839,
"theoretical_loss": 3.5175096295787958,
"tokens_seen": 1493106688
},
{
"epoch": 0.45,
"learning_rate": 0.00027657973921765296,
"loss": 2.6616,
"theoretical_loss": 3.517496005279077,
"tokens_seen": 1493172224
},
{
"epoch": 0.45,
"learning_rate": 0.00027656970912738214,
"loss": 2.6826,
"theoretical_loss": 3.5174823817447467,
"tokens_seen": 1493237760
},
{
"epoch": 0.45,
"learning_rate": 0.0002765596790371113,
"loss": 2.5071,
"theoretical_loss": 3.5174687589757285,
"tokens_seen": 1493303296
},
{
"epoch": 0.45,
"learning_rate": 0.00027654964894684056,
"loss": 2.5251,
"theoretical_loss": 3.517455136971946,
"tokens_seen": 1493368832
},
{
"epoch": 0.45,
"learning_rate": 0.0002765396188565697,
"loss": 2.5445,
"theoretical_loss": 3.5174415157333216,
"tokens_seen": 1493434368
},
{
"epoch": 0.45,
"learning_rate": 0.0002765295887662989,
"loss": 2.8631,
"theoretical_loss": 3.5174278952597806,
"tokens_seen": 1493499904
},
{
"epoch": 0.45,
"learning_rate": 0.00027651955867602805,
"loss": 2.6999,
"theoretical_loss": 3.5174142755512445,
"tokens_seen": 1493565440
},
{
"epoch": 0.45,
"learning_rate": 0.0002765095285857573,
"loss": 2.5781,
"theoretical_loss": 3.517400656607638,
"tokens_seen": 1493630976
},
{
"epoch": 0.45,
"learning_rate": 0.00027649949849548646,
"loss": 2.7064,
"theoretical_loss": 3.5173870384288843,
"tokens_seen": 1493696512
},
{
"epoch": 0.45,
"learning_rate": 0.00027648946840521564,
"loss": 2.7246,
"theoretical_loss": 3.5173734210149066,
"tokens_seen": 1493762048
},
{
"epoch": 0.45,
"learning_rate": 0.0002764794383149448,
"loss": 2.5557,
"theoretical_loss": 3.5173598043656282,
"tokens_seen": 1493827584
},
{
"epoch": 0.45,
"learning_rate": 0.00027646940822467406,
"loss": 2.816,
"theoretical_loss": 3.517346188480974,
"tokens_seen": 1493893120
},
{
"epoch": 0.45,
"learning_rate": 0.0002764593781344032,
"loss": 2.7012,
"theoretical_loss": 3.517332573360866,
"tokens_seen": 1493958656
},
{
"epoch": 0.45,
"learning_rate": 0.0002764493480441324,
"loss": 2.4528,
"theoretical_loss": 3.517318959005228,
"tokens_seen": 1494024192
},
{
"epoch": 0.45,
"learning_rate": 0.00027643931795386155,
"loss": 2.564,
"theoretical_loss": 3.517305345413984,
"tokens_seen": 1494089728
},
{
"epoch": 0.45,
"learning_rate": 0.0002764292878635908,
"loss": 2.7429,
"theoretical_loss": 3.517291732587058,
"tokens_seen": 1494155264
},
{
"epoch": 0.45,
"objective/train/docs_used": 848095,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6782610416412354,
"objective/train/theoretical_loss": 3.517278120524372,
"objective/train/tokens_used": 1514680800,
"theoretical_loss": 3.517278120524372,
"tokens_seen": 1494220800
},
{
"epoch": 0.45,
"learning_rate": 0.00027641925777331997,
"loss": 2.653,
"theoretical_loss": 3.517278120524372,
"tokens_seen": 1494220800
},
{
"epoch": 0.45,
"learning_rate": 0.00027640922768304915,
"loss": 2.6632,
"theoretical_loss": 3.517264509225851,
"tokens_seen": 1494286336
},
{
"epoch": 0.45,
"learning_rate": 0.00027639919759277833,
"loss": 2.8066,
"theoretical_loss": 3.517250898691418,
"tokens_seen": 1494351872
},
{
"epoch": 0.45,
"learning_rate": 0.0002763891675025075,
"loss": 2.4897,
"theoretical_loss": 3.517237288920997,
"tokens_seen": 1494417408
},
{
"epoch": 0.45,
"learning_rate": 0.0002763791374122367,
"loss": 2.635,
"theoretical_loss": 3.517223679914511,
"tokens_seen": 1494482944
},
{
"epoch": 0.45,
"learning_rate": 0.0002763691073219659,
"loss": 2.8805,
"theoretical_loss": 3.5172100716718844,
"tokens_seen": 1494548480
},
{
"epoch": 0.45,
"learning_rate": 0.00027635907723169505,
"loss": 2.6343,
"theoretical_loss": 3.5171964641930398,
"tokens_seen": 1494614016
},
{
"epoch": 0.45,
"learning_rate": 0.0002763490471414243,
"loss": 2.5003,
"theoretical_loss": 3.517182857477902,
"tokens_seen": 1494679552
},
{
"epoch": 0.45,
"learning_rate": 0.0002763390170511534,
"loss": 2.9079,
"theoretical_loss": 3.5171692515263935,
"tokens_seen": 1494745088
},
{
"epoch": 0.45,
"learning_rate": 0.00027632898696088265,
"loss": 2.7104,
"theoretical_loss": 3.5171556463384386,
"tokens_seen": 1494810624
},
{
"epoch": 0.45,
"learning_rate": 0.00027631895687061183,
"loss": 2.7981,
"theoretical_loss": 3.5171420419139614,
"tokens_seen": 1494876160
},
{
"epoch": 0.45,
"learning_rate": 0.000276308926780341,
"loss": 2.5827,
"theoretical_loss": 3.5171284382528847,
"tokens_seen": 1494941696
},
{
"epoch": 0.45,
"learning_rate": 0.0002762988966900702,
"loss": 2.5719,
"theoretical_loss": 3.517114835355133,
"tokens_seen": 1495007232
},
{
"epoch": 0.45,
"learning_rate": 0.00027628886659979943,
"loss": 2.7215,
"theoretical_loss": 3.5171012332206293,
"tokens_seen": 1495072768
},
{
"epoch": 0.45,
"learning_rate": 0.00027627883650952856,
"loss": 2.6334,
"theoretical_loss": 3.517087631849298,
"tokens_seen": 1495138304
},
{
"epoch": 0.45,
"learning_rate": 0.0002762688064192578,
"loss": 2.7117,
"theoretical_loss": 3.517074031241062,
"tokens_seen": 1495203840
},
{
"epoch": 0.45,
"learning_rate": 0.00027625877632898697,
"loss": 2.5609,
"theoretical_loss": 3.5170604313958456,
"tokens_seen": 1495269376
},
{
"epoch": 0.45,
"learning_rate": 0.00027624874623871615,
"loss": 2.5578,
"theoretical_loss": 3.5170468323135724,
"tokens_seen": 1495334912
},
{
"epoch": 0.45,
"learning_rate": 0.0002762387161484454,
"loss": 2.6222,
"theoretical_loss": 3.517033233994167,
"tokens_seen": 1495400448
},
{
"epoch": 0.45,
"learning_rate": 0.0002762286860581745,
"loss": 2.7651,
"theoretical_loss": 3.5170196364375514,
"tokens_seen": 1495465984
},
{
"epoch": 0.45,
"learning_rate": 0.00027621865596790375,
"loss": 2.8586,
"theoretical_loss": 3.5170060396436504,
"tokens_seen": 1495531520
},
{
"epoch": 0.45,
"learning_rate": 0.0002762086258776329,
"loss": 2.6816,
"theoretical_loss": 3.5169924436123887,
"tokens_seen": 1495597056
},
{
"epoch": 0.45,
"learning_rate": 0.0002761985957873621,
"loss": 2.711,
"theoretical_loss": 3.5169788483436886,
"tokens_seen": 1495662592
},
{
"epoch": 0.45,
"learning_rate": 0.0002761885656970913,
"loss": 2.6329,
"theoretical_loss": 3.5169652538374745,
"tokens_seen": 1495728128
},
{
"epoch": 0.45,
"learning_rate": 0.0002761785356068205,
"loss": 2.7441,
"theoretical_loss": 3.51695166009367,
"tokens_seen": 1495793664
},
{
"epoch": 0.45,
"objective/train/docs_used": 848766,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.476358413696289,
"objective/train/theoretical_loss": 3.5169380671122,
"objective/train/tokens_used": 1516319200,
"theoretical_loss": 3.5169380671122,
"tokens_seen": 1495859200
},
{
"epoch": 0.45,
"learning_rate": 0.00027616850551654966,
"loss": 2.6673,
"theoretical_loss": 3.5169380671122,
"tokens_seen": 1495859200
},
{
"epoch": 0.45,
"learning_rate": 0.0002761584754262789,
"loss": 2.6851,
"theoretical_loss": 3.516924474892987,
"tokens_seen": 1495924736
},
{
"epoch": 0.45,
"learning_rate": 0.000276148445336008,
"loss": 2.7359,
"theoretical_loss": 3.5169108834359557,
"tokens_seen": 1495990272
},
{
"epoch": 0.45,
"learning_rate": 0.00027613841524573725,
"loss": 2.7118,
"theoretical_loss": 3.5168972927410294,
"tokens_seen": 1496055808
},
{
"epoch": 0.45,
"learning_rate": 0.0002761283851554664,
"loss": 2.6572,
"theoretical_loss": 3.5168837028081326,
"tokens_seen": 1496121344
},
{
"epoch": 0.45,
"learning_rate": 0.0002761183550651956,
"loss": 2.5281,
"theoretical_loss": 3.5168701136371885,
"tokens_seen": 1496186880
},
{
"epoch": 0.45,
"learning_rate": 0.0002761083249749248,
"loss": 2.5437,
"theoretical_loss": 3.516856525228121,
"tokens_seen": 1496252416
},
{
"epoch": 0.45,
"learning_rate": 0.000276098294884654,
"loss": 2.7465,
"theoretical_loss": 3.516842937580855,
"tokens_seen": 1496317952
},
{
"epoch": 0.45,
"learning_rate": 0.00027608826479438316,
"loss": 2.3249,
"theoretical_loss": 3.516829350695314,
"tokens_seen": 1496383488
},
{
"epoch": 0.45,
"learning_rate": 0.00027607823470411234,
"loss": 2.5481,
"theoretical_loss": 3.5168157645714215,
"tokens_seen": 1496449024
},
{
"epoch": 0.45,
"learning_rate": 0.0002760682046138415,
"loss": 2.518,
"theoretical_loss": 3.516802179209101,
"tokens_seen": 1496514560
},
{
"epoch": 0.45,
"learning_rate": 0.00027605817452357076,
"loss": 2.4998,
"theoretical_loss": 3.516788594608278,
"tokens_seen": 1496580096
},
{
"epoch": 0.45,
"learning_rate": 0.0002760481444332999,
"loss": 2.8204,
"theoretical_loss": 3.5167750107688756,
"tokens_seen": 1496645632
},
{
"epoch": 0.45,
"learning_rate": 0.0002760381143430291,
"loss": 2.5531,
"theoretical_loss": 3.5167614276908177,
"tokens_seen": 1496711168
},
{
"epoch": 0.45,
"learning_rate": 0.00027602808425275825,
"loss": 2.6845,
"theoretical_loss": 3.5167478453740286,
"tokens_seen": 1496776704
},
{
"epoch": 0.45,
"learning_rate": 0.0002760180541624875,
"loss": 2.5706,
"theoretical_loss": 3.5167342638184316,
"tokens_seen": 1496842240
},
{
"epoch": 0.45,
"learning_rate": 0.00027600802407221666,
"loss": 2.7972,
"theoretical_loss": 3.5167206830239515,
"tokens_seen": 1496907776
},
{
"epoch": 0.45,
"learning_rate": 0.00027599799398194584,
"loss": 2.5509,
"theoretical_loss": 3.5167071029905124,
"tokens_seen": 1496973312
},
{
"epoch": 0.45,
"learning_rate": 0.000275987963891675,
"loss": 2.7023,
"theoretical_loss": 3.516693523718038,
"tokens_seen": 1497038848
},
{
"epoch": 0.45,
"learning_rate": 0.00027597793380140426,
"loss": 2.3785,
"theoretical_loss": 3.5166799452064517,
"tokens_seen": 1497104384
},
{
"epoch": 0.45,
"learning_rate": 0.0002759679037111334,
"loss": 2.6549,
"theoretical_loss": 3.5166663674556786,
"tokens_seen": 1497169920
},
{
"epoch": 0.45,
"learning_rate": 0.0002759578736208626,
"loss": 2.7072,
"theoretical_loss": 3.5166527904656424,
"tokens_seen": 1497235456
},
{
"epoch": 0.45,
"learning_rate": 0.00027594784353059175,
"loss": 2.5941,
"theoretical_loss": 3.5166392142362675,
"tokens_seen": 1497300992
},
{
"epoch": 0.45,
"learning_rate": 0.000275937813440321,
"loss": 2.7586,
"theoretical_loss": 3.516625638767477,
"tokens_seen": 1497366528
},
{
"epoch": 0.45,
"learning_rate": 0.00027592778335005017,
"loss": 2.7524,
"theoretical_loss": 3.516612064059196,
"tokens_seen": 1497432064
},
{
"epoch": 0.45,
"objective/train/docs_used": 849888,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6487090587615967,
"objective/train/theoretical_loss": 3.516598490111348,
"objective/train/tokens_used": 1517957600,
"theoretical_loss": 3.516598490111348,
"tokens_seen": 1497497600
},
{
"epoch": 0.45,
"learning_rate": 0.00027591775325977935,
"loss": 2.6396,
"theoretical_loss": 3.516598490111348,
"tokens_seen": 1497497600
},
{
"epoch": 0.45,
"learning_rate": 0.00027590772316950853,
"loss": 2.3488,
"theoretical_loss": 3.516584916923858,
"tokens_seen": 1497563136
},
{
"epoch": 0.45,
"learning_rate": 0.0002758976930792377,
"loss": 2.8883,
"theoretical_loss": 3.516571344496649,
"tokens_seen": 1497628672
},
{
"epoch": 0.45,
"learning_rate": 0.0002758876629889669,
"loss": 2.5636,
"theoretical_loss": 3.516557772829646,
"tokens_seen": 1497694208
},
{
"epoch": 0.45,
"learning_rate": 0.0002758776328986961,
"loss": 2.8297,
"theoretical_loss": 3.516544201922773,
"tokens_seen": 1497759744
},
{
"epoch": 0.45,
"learning_rate": 0.00027586760280842525,
"loss": 2.5523,
"theoretical_loss": 3.5165306317759537,
"tokens_seen": 1497825280
},
{
"epoch": 0.45,
"learning_rate": 0.0002758575727181545,
"loss": 2.6092,
"theoretical_loss": 3.516517062389112,
"tokens_seen": 1497890816
},
{
"epoch": 0.45,
"learning_rate": 0.0002758475426278836,
"loss": 2.5739,
"theoretical_loss": 3.5165034937621735,
"tokens_seen": 1497956352
},
{
"epoch": 0.45,
"learning_rate": 0.00027583751253761285,
"loss": 2.8135,
"theoretical_loss": 3.5164899258950615,
"tokens_seen": 1498021888
},
{
"epoch": 0.45,
"learning_rate": 0.00027582748244734203,
"loss": 2.5584,
"theoretical_loss": 3.5164763587877,
"tokens_seen": 1498087424
},
{
"epoch": 0.45,
"learning_rate": 0.0002758174523570712,
"loss": 2.698,
"theoretical_loss": 3.516462792440014,
"tokens_seen": 1498152960
},
{
"epoch": 0.45,
"learning_rate": 0.0002758074222668004,
"loss": 2.673,
"theoretical_loss": 3.5164492268519267,
"tokens_seen": 1498218496
},
{
"epoch": 0.45,
"learning_rate": 0.00027579739217652963,
"loss": 2.5653,
"theoretical_loss": 3.516435662023363,
"tokens_seen": 1498284032
},
{
"epoch": 0.45,
"learning_rate": 0.00027578736208625876,
"loss": 2.7921,
"theoretical_loss": 3.5164220979542478,
"tokens_seen": 1498349568
},
{
"epoch": 0.45,
"learning_rate": 0.000275777331995988,
"loss": 2.4283,
"theoretical_loss": 3.5164085346445035,
"tokens_seen": 1498415104
},
{
"epoch": 0.45,
"learning_rate": 0.0002757673019057171,
"loss": 2.4956,
"theoretical_loss": 3.516394972094056,
"tokens_seen": 1498480640
},
{
"epoch": 0.45,
"learning_rate": 0.00027575727181544635,
"loss": 2.6457,
"theoretical_loss": 3.516381410302829,
"tokens_seen": 1498546176
},
{
"epoch": 0.45,
"learning_rate": 0.00027574724172517554,
"loss": 2.4338,
"theoretical_loss": 3.5163678492707473,
"tokens_seen": 1498611712
},
{
"epoch": 0.45,
"learning_rate": 0.0002757372116349047,
"loss": 2.6423,
"theoretical_loss": 3.5163542889977344,
"tokens_seen": 1498677248
},
{
"epoch": 0.45,
"learning_rate": 0.0002757271815446339,
"loss": 2.7102,
"theoretical_loss": 3.516340729483715,
"tokens_seen": 1498742784
},
{
"epoch": 0.45,
"learning_rate": 0.0002757171514543631,
"loss": 2.6216,
"theoretical_loss": 3.516327170728614,
"tokens_seen": 1498808320
},
{
"epoch": 0.45,
"learning_rate": 0.00027570712136409226,
"loss": 2.6347,
"theoretical_loss": 3.5163136127323544,
"tokens_seen": 1498873856
},
{
"epoch": 0.45,
"learning_rate": 0.0002756970912738215,
"loss": 2.603,
"theoretical_loss": 3.516300055494862,
"tokens_seen": 1498939392
},
{
"epoch": 0.45,
"learning_rate": 0.0002756870611835506,
"loss": 2.6805,
"theoretical_loss": 3.5162864990160596,
"tokens_seen": 1499004928
},
{
"epoch": 0.45,
"learning_rate": 0.00027567703109327986,
"loss": 2.5052,
"theoretical_loss": 3.5162729432958733,
"tokens_seen": 1499070464
},
{
"epoch": 0.45,
"objective/train/docs_used": 850610,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2861216068267822,
"objective/train/theoretical_loss": 3.516259388334227,
"objective/train/tokens_used": 1519596000,
"theoretical_loss": 3.516259388334227,
"tokens_seen": 1499136000
},
{
"epoch": 0.45,
"learning_rate": 0.000275667001003009,
"loss": 2.6936,
"theoretical_loss": 3.516259388334227,
"tokens_seen": 1499136000
},
{
"epoch": 0.45,
"learning_rate": 0.0002756569709127382,
"loss": 2.5356,
"theoretical_loss": 3.516245834131044,
"tokens_seen": 1499201536
},
{
"epoch": 0.45,
"learning_rate": 0.0002756469408224674,
"loss": 2.7342,
"theoretical_loss": 3.5162322806862494,
"tokens_seen": 1499267072
},
{
"epoch": 0.45,
"learning_rate": 0.0002756369107321966,
"loss": 2.7266,
"theoretical_loss": 3.516218727999768,
"tokens_seen": 1499332608
},
{
"epoch": 0.45,
"learning_rate": 0.00027562688064192576,
"loss": 2.7281,
"theoretical_loss": 3.516205176071524,
"tokens_seen": 1499398144
},
{
"epoch": 0.45,
"learning_rate": 0.000275616850551655,
"loss": 2.5561,
"theoretical_loss": 3.5161916249014413,
"tokens_seen": 1499463680
},
{
"epoch": 0.45,
"learning_rate": 0.0002756068204613841,
"loss": 2.3772,
"theoretical_loss": 3.5161780744894453,
"tokens_seen": 1499529216
},
{
"epoch": 0.45,
"learning_rate": 0.00027559679037111336,
"loss": 2.5579,
"theoretical_loss": 3.51616452483546,
"tokens_seen": 1499594752
},
{
"epoch": 0.45,
"learning_rate": 0.0002755867602808425,
"loss": 2.6899,
"theoretical_loss": 3.5161509759394094,
"tokens_seen": 1499660288
},
{
"epoch": 0.45,
"learning_rate": 0.0002755767301905717,
"loss": 2.877,
"theoretical_loss": 3.516137427801219,
"tokens_seen": 1499725824
},
{
"epoch": 0.45,
"learning_rate": 0.0002755667001003009,
"loss": 2.3575,
"theoretical_loss": 3.516123880420812,
"tokens_seen": 1499791360
},
{
"epoch": 0.45,
"learning_rate": 0.0002755566700100301,
"loss": 2.4701,
"theoretical_loss": 3.5161103337981143,
"tokens_seen": 1499856896
},
{
"epoch": 0.45,
"learning_rate": 0.00027554663991975927,
"loss": 2.9197,
"theoretical_loss": 3.5160967879330496,
"tokens_seen": 1499922432
},
{
"epoch": 0.45,
"learning_rate": 0.00027553660982948845,
"loss": 2.6596,
"theoretical_loss": 3.516083242825542,
"tokens_seen": 1499987968
},
{
"epoch": 0.45,
"learning_rate": 0.00027552657973921763,
"loss": 2.7017,
"theoretical_loss": 3.5160696984755173,
"tokens_seen": 1500053504
},
{
"epoch": 0.45,
"learning_rate": 0.00027551654964894686,
"loss": 2.6792,
"theoretical_loss": 3.5160561548828992,
"tokens_seen": 1500119040
},
{
"epoch": 0.45,
"learning_rate": 0.00027550651955867604,
"loss": 2.5864,
"theoretical_loss": 3.5160426120476123,
"tokens_seen": 1500184576
},
{
"epoch": 0.45,
"learning_rate": 0.0002754964894684052,
"loss": 2.8472,
"theoretical_loss": 3.5160290699695813,
"tokens_seen": 1500250112
},
{
"epoch": 0.45,
"learning_rate": 0.00027548645937813446,
"loss": 2.7539,
"theoretical_loss": 3.5160155286487313,
"tokens_seen": 1500315648
},
{
"epoch": 0.45,
"learning_rate": 0.0002754764292878636,
"loss": 2.6775,
"theoretical_loss": 3.5160019880849855,
"tokens_seen": 1500381184
},
{
"epoch": 0.45,
"learning_rate": 0.0002754663991975928,
"loss": 2.7107,
"theoretical_loss": 3.5159884482782697,
"tokens_seen": 1500446720
},
{
"epoch": 0.45,
"learning_rate": 0.00027545636910732195,
"loss": 2.8488,
"theoretical_loss": 3.5159749092285084,
"tokens_seen": 1500512256
},
{
"epoch": 0.45,
"learning_rate": 0.0002754463390170512,
"loss": 2.6165,
"theoretical_loss": 3.5159613709356257,
"tokens_seen": 1500577792
},
{
"epoch": 0.45,
"learning_rate": 0.00027543630892678037,
"loss": 2.4482,
"theoretical_loss": 3.5159478333995464,
"tokens_seen": 1500643328
},
{
"epoch": 0.45,
"learning_rate": 0.00027542627883650955,
"loss": 2.5014,
"theoretical_loss": 3.5159342966201956,
"tokens_seen": 1500708864
},
{
"epoch": 0.45,
"objective/train/docs_used": 851359,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5217947959899902,
"objective/train/theoretical_loss": 3.515920760597498,
"objective/train/tokens_used": 1521234400,
"theoretical_loss": 3.515920760597498,
"tokens_seen": 1500774400
},
{
"epoch": 0.45,
"learning_rate": 0.00027541624874623873,
"loss": 2.5828,
"theoretical_loss": 3.515920760597498,
"tokens_seen": 1500774400
},
{
"epoch": 0.45,
"learning_rate": 0.0002754062186559679,
"loss": 2.5872,
"theoretical_loss": 3.515907225331377,
"tokens_seen": 1500839936
},
{
"epoch": 0.45,
"learning_rate": 0.0002753961885656971,
"loss": 2.6722,
"theoretical_loss": 3.515893690821759,
"tokens_seen": 1500905472
},
{
"epoch": 0.45,
"learning_rate": 0.0002753861584754263,
"loss": 2.6236,
"theoretical_loss": 3.515880157068567,
"tokens_seen": 1500971008
},
{
"epoch": 0.45,
"learning_rate": 0.00027537612838515545,
"loss": 2.6064,
"theoretical_loss": 3.5158666240717276,
"tokens_seen": 1501036544
},
{
"epoch": 0.45,
"learning_rate": 0.0002753660982948847,
"loss": 2.766,
"theoretical_loss": 3.515853091831164,
"tokens_seen": 1501102080
},
{
"epoch": 0.45,
"learning_rate": 0.0002753560682046138,
"loss": 2.6,
"theoretical_loss": 3.5158395603468016,
"tokens_seen": 1501167616
},
{
"epoch": 0.45,
"learning_rate": 0.00027534603811434305,
"loss": 2.5742,
"theoretical_loss": 3.515826029618564,
"tokens_seen": 1501233152
},
{
"epoch": 0.45,
"learning_rate": 0.00027533600802407223,
"loss": 2.6055,
"theoretical_loss": 3.515812499646378,
"tokens_seen": 1501298688
},
{
"epoch": 0.45,
"learning_rate": 0.0002753259779338014,
"loss": 2.7568,
"theoretical_loss": 3.5157989704301666,
"tokens_seen": 1501364224
},
{
"epoch": 0.45,
"learning_rate": 0.0002753159478435306,
"loss": 2.6491,
"theoretical_loss": 3.5157854419698555,
"tokens_seen": 1501429760
},
{
"epoch": 0.45,
"learning_rate": 0.00027530591775325983,
"loss": 2.7314,
"theoretical_loss": 3.515771914265369,
"tokens_seen": 1501495296
},
{
"epoch": 0.46,
"learning_rate": 0.00027529588766298896,
"loss": 2.598,
"theoretical_loss": 3.515758387316632,
"tokens_seen": 1501560832
},
{
"epoch": 0.46,
"learning_rate": 0.0002752858575727182,
"loss": 2.7708,
"theoretical_loss": 3.51574486112357,
"tokens_seen": 1501626368
},
{
"epoch": 0.46,
"learning_rate": 0.0002752758274824473,
"loss": 2.69,
"theoretical_loss": 3.5157313356861066,
"tokens_seen": 1501691904
},
{
"epoch": 0.46,
"learning_rate": 0.00027526579739217655,
"loss": 2.7741,
"theoretical_loss": 3.5157178110041674,
"tokens_seen": 1501757440
},
{
"epoch": 0.46,
"learning_rate": 0.00027525576730190574,
"loss": 2.6573,
"theoretical_loss": 3.5157042870776767,
"tokens_seen": 1501822976
},
{
"epoch": 0.46,
"learning_rate": 0.0002752457372116349,
"loss": 2.5598,
"theoretical_loss": 3.5156907639065595,
"tokens_seen": 1501888512
},
{
"epoch": 0.46,
"learning_rate": 0.0002752357071213641,
"loss": 2.8174,
"theoretical_loss": 3.5156772414907413,
"tokens_seen": 1501954048
},
{
"epoch": 0.46,
"learning_rate": 0.0002752256770310933,
"loss": 2.6147,
"theoretical_loss": 3.515663719830146,
"tokens_seen": 1502019584
},
{
"epoch": 0.46,
"learning_rate": 0.00027521564694082246,
"loss": 2.5259,
"theoretical_loss": 3.5156501989246993,
"tokens_seen": 1502085120
},
{
"epoch": 0.46,
"learning_rate": 0.0002752056168505517,
"loss": 2.9048,
"theoretical_loss": 3.5156366787743254,
"tokens_seen": 1502150656
},
{
"epoch": 0.46,
"learning_rate": 0.0002751955867602808,
"loss": 2.9369,
"theoretical_loss": 3.5156231593789498,
"tokens_seen": 1502216192
},
{
"epoch": 0.46,
"learning_rate": 0.00027518555667001006,
"loss": 2.8631,
"theoretical_loss": 3.5156096407384974,
"tokens_seen": 1502281728
},
{
"epoch": 0.46,
"learning_rate": 0.0002751755265797392,
"loss": 2.88,
"theoretical_loss": 3.515596122852892,
"tokens_seen": 1502347264
},
{
"epoch": 0.46,
"objective/train/docs_used": 852603,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6841578483581543,
"objective/train/theoretical_loss": 3.51558260572206,
"objective/train/tokens_used": 1522872800,
"theoretical_loss": 3.51558260572206,
"tokens_seen": 1502412800
},
{
"epoch": 0.46,
"learning_rate": 0.0002751654964894684,
"loss": 2.708,
"theoretical_loss": 3.51558260572206,
"tokens_seen": 1502412800
},
{
"epoch": 0.46,
"learning_rate": 0.0002751554663991976,
"loss": 2.6512,
"theoretical_loss": 3.5155690893459255,
"tokens_seen": 1502478336
},
{
"epoch": 0.46,
"learning_rate": 0.0002751454363089268,
"loss": 2.5085,
"theoretical_loss": 3.5155555737244133,
"tokens_seen": 1502543872
},
{
"epoch": 0.46,
"learning_rate": 0.00027513540621865596,
"loss": 2.7953,
"theoretical_loss": 3.515542058857449,
"tokens_seen": 1502609408
},
{
"epoch": 0.46,
"learning_rate": 0.0002751253761283852,
"loss": 2.8557,
"theoretical_loss": 3.5155285447449574,
"tokens_seen": 1502674944
},
{
"epoch": 0.46,
"learning_rate": 0.0002751153460381143,
"loss": 2.6766,
"theoretical_loss": 3.515515031386863,
"tokens_seen": 1502740480
},
{
"epoch": 0.46,
"learning_rate": 0.00027510531594784356,
"loss": 2.6592,
"theoretical_loss": 3.5155015187830916,
"tokens_seen": 1502806016
},
{
"epoch": 0.46,
"learning_rate": 0.0002750952858575727,
"loss": 2.8439,
"theoretical_loss": 3.515488006933567,
"tokens_seen": 1502871552
},
{
"epoch": 0.46,
"learning_rate": 0.0002750852557673019,
"loss": 2.8174,
"theoretical_loss": 3.515474495838216,
"tokens_seen": 1502937088
},
{
"epoch": 0.46,
"learning_rate": 0.0002750752256770311,
"loss": 2.8733,
"theoretical_loss": 3.5154609854969614,
"tokens_seen": 1503002624
},
{
"epoch": 0.46,
"learning_rate": 0.0002750651955867603,
"loss": 2.5403,
"theoretical_loss": 3.5154474759097303,
"tokens_seen": 1503068160
},
{
"epoch": 0.46,
"learning_rate": 0.00027505516549648947,
"loss": 2.5577,
"theoretical_loss": 3.5154339670764463,
"tokens_seen": 1503133696
},
{
"epoch": 0.46,
"learning_rate": 0.00027504513540621865,
"loss": 2.3886,
"theoretical_loss": 3.5154204589970353,
"tokens_seen": 1503199232
},
{
"epoch": 0.46,
"learning_rate": 0.00027503510531594783,
"loss": 2.7458,
"theoretical_loss": 3.515406951671422,
"tokens_seen": 1503264768
},
{
"epoch": 0.46,
"learning_rate": 0.00027502507522567706,
"loss": 2.6228,
"theoretical_loss": 3.5153934450995314,
"tokens_seen": 1503330304
},
{
"epoch": 0.46,
"learning_rate": 0.0002750150451354062,
"loss": 2.5597,
"theoretical_loss": 3.5153799392812886,
"tokens_seen": 1503395840
},
{
"epoch": 0.46,
"learning_rate": 0.0002750050150451354,
"loss": 2.683,
"theoretical_loss": 3.5153664342166193,
"tokens_seen": 1503461376
},
{
"epoch": 0.46,
"learning_rate": 0.00027499498495486455,
"loss": 2.5473,
"theoretical_loss": 3.515352929905448,
"tokens_seen": 1503526912
},
{
"epoch": 0.46,
"learning_rate": 0.0002749849548645938,
"loss": 2.7853,
"theoretical_loss": 3.5153394263476994,
"tokens_seen": 1503592448
},
{
"epoch": 0.46,
"learning_rate": 0.00027497492477432297,
"loss": 2.6969,
"theoretical_loss": 3.5153259235432994,
"tokens_seen": 1503657984
},
{
"epoch": 0.46,
"learning_rate": 0.00027496489468405215,
"loss": 2.7103,
"theoretical_loss": 3.5153124214921734,
"tokens_seen": 1503723520
},
{
"epoch": 0.46,
"learning_rate": 0.00027495486459378133,
"loss": 2.6891,
"theoretical_loss": 3.5152989201942457,
"tokens_seen": 1503789056
},
{
"epoch": 0.46,
"learning_rate": 0.00027494483450351057,
"loss": 2.7623,
"theoretical_loss": 3.515285419649442,
"tokens_seen": 1503854592
},
{
"epoch": 0.46,
"learning_rate": 0.0002749348044132397,
"loss": 2.5915,
"theoretical_loss": 3.515271919857687,
"tokens_seen": 1503920128
},
{
"epoch": 0.46,
"learning_rate": 0.00027492477432296893,
"loss": 2.6001,
"theoretical_loss": 3.515258420818906,
"tokens_seen": 1503985664
},
{
"epoch": 0.46,
"objective/train/docs_used": 853123,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4897100925445557,
"objective/train/theoretical_loss": 3.515244922533025,
"objective/train/tokens_used": 1524511200,
"theoretical_loss": 3.515244922533025,
"tokens_seen": 1504051200
},
{
"epoch": 0.46,
"learning_rate": 0.00027491474423269806,
"loss": 2.6291,
"theoretical_loss": 3.515244922533025,
"tokens_seen": 1504051200
},
{
"epoch": 0.46,
"learning_rate": 0.0002749047141424273,
"loss": 2.662,
"theoretical_loss": 3.5152314249999685,
"tokens_seen": 1504116736
},
{
"epoch": 0.46,
"learning_rate": 0.00027489468405215647,
"loss": 2.6613,
"theoretical_loss": 3.5152179282196614,
"tokens_seen": 1504182272
},
{
"epoch": 0.46,
"learning_rate": 0.00027488465396188565,
"loss": 2.6598,
"theoretical_loss": 3.5152044321920295,
"tokens_seen": 1504247808
},
{
"epoch": 0.46,
"learning_rate": 0.00027487462387161483,
"loss": 2.8206,
"theoretical_loss": 3.5151909369169982,
"tokens_seen": 1504313344
},
{
"epoch": 0.46,
"learning_rate": 0.000274864593781344,
"loss": 2.774,
"theoretical_loss": 3.515177442394492,
"tokens_seen": 1504378880
},
{
"epoch": 0.46,
"learning_rate": 0.0002748545636910732,
"loss": 2.8791,
"theoretical_loss": 3.5151639486244366,
"tokens_seen": 1504444416
},
{
"epoch": 0.46,
"learning_rate": 0.00027484453360080243,
"loss": 2.6973,
"theoretical_loss": 3.515150455606757,
"tokens_seen": 1504509952
},
{
"epoch": 0.46,
"learning_rate": 0.00027483450351053156,
"loss": 2.6744,
"theoretical_loss": 3.515136963341379,
"tokens_seen": 1504575488
},
{
"epoch": 0.46,
"learning_rate": 0.0002748244734202608,
"loss": 2.6286,
"theoretical_loss": 3.515123471828228,
"tokens_seen": 1504641024
},
{
"epoch": 0.46,
"learning_rate": 0.00027481444332999,
"loss": 2.5343,
"theoretical_loss": 3.515109981067228,
"tokens_seen": 1504706560
},
{
"epoch": 0.46,
"learning_rate": 0.00027480441323971916,
"loss": 2.6437,
"theoretical_loss": 3.5150964910583062,
"tokens_seen": 1504772096
},
{
"epoch": 0.46,
"learning_rate": 0.00027479438314944834,
"loss": 2.506,
"theoretical_loss": 3.5150830018013863,
"tokens_seen": 1504837632
},
{
"epoch": 0.46,
"learning_rate": 0.0002747843530591775,
"loss": 2.5433,
"theoretical_loss": 3.5150695132963943,
"tokens_seen": 1504903168
},
{
"epoch": 0.46,
"learning_rate": 0.00027477432296890675,
"loss": 2.5128,
"theoretical_loss": 3.515056025543256,
"tokens_seen": 1504968704
},
{
"epoch": 0.46,
"learning_rate": 0.00027476429287863594,
"loss": 2.6609,
"theoretical_loss": 3.5150425385418957,
"tokens_seen": 1505034240
},
{
"epoch": 0.46,
"learning_rate": 0.0002747542627883651,
"loss": 2.6762,
"theoretical_loss": 3.515029052292239,
"tokens_seen": 1505099776
},
{
"epoch": 0.46,
"learning_rate": 0.0002747442326980943,
"loss": 2.6938,
"theoretical_loss": 3.5150155667942125,
"tokens_seen": 1505165312
},
{
"epoch": 0.46,
"learning_rate": 0.0002747342026078235,
"loss": 2.7439,
"theoretical_loss": 3.51500208204774,
"tokens_seen": 1505230848
},
{
"epoch": 0.46,
"learning_rate": 0.00027472417251755266,
"loss": 2.8202,
"theoretical_loss": 3.514988598052748,
"tokens_seen": 1505296384
},
{
"epoch": 0.46,
"learning_rate": 0.0002747141424272819,
"loss": 2.5343,
"theoretical_loss": 3.5149751148091615,
"tokens_seen": 1505361920
},
{
"epoch": 0.46,
"learning_rate": 0.000274704112337011,
"loss": 2.7935,
"theoretical_loss": 3.514961632316906,
"tokens_seen": 1505427456
},
{
"epoch": 0.46,
"learning_rate": 0.00027469408224674026,
"loss": 2.7413,
"theoretical_loss": 3.5149481505759064,
"tokens_seen": 1505492992
},
{
"epoch": 0.46,
"learning_rate": 0.0002746840521564694,
"loss": 2.5229,
"theoretical_loss": 3.514934669586089,
"tokens_seen": 1505558528
},
{
"epoch": 0.46,
"learning_rate": 0.0002746740220661986,
"loss": 2.7799,
"theoretical_loss": 3.5149211893473784,
"tokens_seen": 1505624064
},
{
"epoch": 0.46,
"objective/train/docs_used": 854391,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4277405738830566,
"objective/train/theoretical_loss": 3.514907709859701,
"objective/train/tokens_used": 1526149600,
"theoretical_loss": 3.514907709859701,
"tokens_seen": 1505689600
},
{
"epoch": 0.46,
"learning_rate": 0.0002746639919759278,
"loss": 2.6277,
"theoretical_loss": 3.514907709859701,
"tokens_seen": 1505689600
},
{
"epoch": 0.46,
"learning_rate": 0.000274653961885657,
"loss": 3.0142,
"theoretical_loss": 3.514894231122981,
"tokens_seen": 1505755136
},
{
"epoch": 0.46,
"learning_rate": 0.00027464393179538616,
"loss": 2.7348,
"theoretical_loss": 3.5148807531371453,
"tokens_seen": 1505820672
},
{
"epoch": 0.46,
"learning_rate": 0.0002746339017051154,
"loss": 2.9283,
"theoretical_loss": 3.5148672759021182,
"tokens_seen": 1505886208
},
{
"epoch": 0.46,
"learning_rate": 0.0002746238716148445,
"loss": 2.6175,
"theoretical_loss": 3.514853799417826,
"tokens_seen": 1505951744
},
{
"epoch": 0.46,
"learning_rate": 0.00027461384152457376,
"loss": 2.6011,
"theoretical_loss": 3.514840323684194,
"tokens_seen": 1506017280
},
{
"epoch": 0.46,
"learning_rate": 0.0002746038114343029,
"loss": 2.6417,
"theoretical_loss": 3.514826848701148,
"tokens_seen": 1506082816
},
{
"epoch": 0.46,
"learning_rate": 0.0002745937813440321,
"loss": 2.8686,
"theoretical_loss": 3.5148133744686127,
"tokens_seen": 1506148352
},
{
"epoch": 0.46,
"learning_rate": 0.0002745837512537613,
"loss": 2.5809,
"theoretical_loss": 3.5147999009865143,
"tokens_seen": 1506213888
},
{
"epoch": 0.46,
"learning_rate": 0.0002745737211634905,
"loss": 2.7395,
"theoretical_loss": 3.5147864282547783,
"tokens_seen": 1506279424
},
{
"epoch": 0.46,
"learning_rate": 0.00027456369107321967,
"loss": 2.8658,
"theoretical_loss": 3.51477295627333,
"tokens_seen": 1506344960
},
{
"epoch": 0.46,
"learning_rate": 0.00027455366098294885,
"loss": 2.4288,
"theoretical_loss": 3.514759485042095,
"tokens_seen": 1506410496
},
{
"epoch": 0.46,
"learning_rate": 0.00027454363089267803,
"loss": 2.482,
"theoretical_loss": 3.514746014560999,
"tokens_seen": 1506476032
},
{
"epoch": 0.46,
"learning_rate": 0.00027453360080240726,
"loss": 2.5397,
"theoretical_loss": 3.514732544829968,
"tokens_seen": 1506541568
},
{
"epoch": 0.46,
"learning_rate": 0.0002745235707121364,
"loss": 2.5256,
"theoretical_loss": 3.5147190758489275,
"tokens_seen": 1506607104
},
{
"epoch": 0.46,
"learning_rate": 0.0002745135406218656,
"loss": 2.762,
"theoretical_loss": 3.5147056076178025,
"tokens_seen": 1506672640
},
{
"epoch": 0.46,
"learning_rate": 0.00027450351053159475,
"loss": 2.6661,
"theoretical_loss": 3.514692140136519,
"tokens_seen": 1506738176
},
{
"epoch": 0.46,
"learning_rate": 0.000274493480441324,
"loss": 2.8853,
"theoretical_loss": 3.514678673405003,
"tokens_seen": 1506803712
},
{
"epoch": 0.46,
"learning_rate": 0.00027448345035105317,
"loss": 2.6241,
"theoretical_loss": 3.514665207423179,
"tokens_seen": 1506869248
},
{
"epoch": 0.46,
"learning_rate": 0.00027447342026078235,
"loss": 2.6854,
"theoretical_loss": 3.514651742190974,
"tokens_seen": 1506934784
},
{
"epoch": 0.46,
"learning_rate": 0.00027446339017051153,
"loss": 2.6821,
"theoretical_loss": 3.514638277708313,
"tokens_seen": 1507000320
},
{
"epoch": 0.46,
"learning_rate": 0.00027445336008024077,
"loss": 2.5069,
"theoretical_loss": 3.5146248139751215,
"tokens_seen": 1507065856
},
{
"epoch": 0.46,
"learning_rate": 0.0002744433299899699,
"loss": 2.6914,
"theoretical_loss": 3.514611350991326,
"tokens_seen": 1507131392
},
{
"epoch": 0.46,
"learning_rate": 0.00027443329989969913,
"loss": 2.5428,
"theoretical_loss": 3.5145978887568514,
"tokens_seen": 1507196928
},
{
"epoch": 0.46,
"learning_rate": 0.00027442326980942826,
"loss": 2.7257,
"theoretical_loss": 3.5145844272716236,
"tokens_seen": 1507262464
},
{
"epoch": 0.46,
"objective/train/docs_used": 855086,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9375758171081543,
"objective/train/theoretical_loss": 3.514570966535569,
"objective/train/tokens_used": 1527788000,
"theoretical_loss": 3.514570966535569,
"tokens_seen": 1507328000
},
{
"epoch": 0.46,
"learning_rate": 0.0002744132397191575,
"loss": 2.7403,
"theoretical_loss": 3.514570966535569,
"tokens_seen": 1507328000
},
{
"epoch": 0.46,
"learning_rate": 0.00027440320962888667,
"loss": 2.8036,
"theoretical_loss": 3.5145575065486123,
"tokens_seen": 1507393536
},
{
"epoch": 0.46,
"learning_rate": 0.00027439317953861585,
"loss": 2.4739,
"theoretical_loss": 3.51454404731068,
"tokens_seen": 1507459072
},
{
"epoch": 0.46,
"learning_rate": 0.00027438314944834503,
"loss": 2.5874,
"theoretical_loss": 3.5145305888216973,
"tokens_seen": 1507524608
},
{
"epoch": 0.46,
"learning_rate": 0.0002743731193580742,
"loss": 2.3555,
"theoretical_loss": 3.5145171310815906,
"tokens_seen": 1507590144
},
{
"epoch": 0.46,
"learning_rate": 0.0002743630892678034,
"loss": 2.7126,
"theoretical_loss": 3.5145036740902853,
"tokens_seen": 1507655680
},
{
"epoch": 0.46,
"learning_rate": 0.00027435305917753263,
"loss": 2.7457,
"theoretical_loss": 3.514490217847707,
"tokens_seen": 1507721216
},
{
"epoch": 0.46,
"learning_rate": 0.00027434302908726176,
"loss": 2.5299,
"theoretical_loss": 3.5144767623537816,
"tokens_seen": 1507786752
},
{
"epoch": 0.46,
"learning_rate": 0.000274332998996991,
"loss": 2.6484,
"theoretical_loss": 3.514463307608435,
"tokens_seen": 1507852288
},
{
"epoch": 0.46,
"learning_rate": 0.0002743229689067202,
"loss": 2.6398,
"theoretical_loss": 3.514449853611594,
"tokens_seen": 1507917824
},
{
"epoch": 0.46,
"learning_rate": 0.00027431293881644936,
"loss": 2.6306,
"theoretical_loss": 3.5144364003631825,
"tokens_seen": 1507983360
},
{
"epoch": 0.46,
"learning_rate": 0.00027430290872617854,
"loss": 2.8692,
"theoretical_loss": 3.5144229478631273,
"tokens_seen": 1508048896
},
{
"epoch": 0.46,
"learning_rate": 0.0002742928786359077,
"loss": 2.73,
"theoretical_loss": 3.5144094961113543,
"tokens_seen": 1508114432
},
{
"epoch": 0.46,
"learning_rate": 0.0002742828485456369,
"loss": 2.767,
"theoretical_loss": 3.51439604510779,
"tokens_seen": 1508179968
},
{
"epoch": 0.46,
"learning_rate": 0.00027427281845536614,
"loss": 2.7589,
"theoretical_loss": 3.514382594852359,
"tokens_seen": 1508245504
},
{
"epoch": 0.46,
"learning_rate": 0.00027426278836509526,
"loss": 2.6004,
"theoretical_loss": 3.514369145344988,
"tokens_seen": 1508311040
},
{
"epoch": 0.46,
"learning_rate": 0.0002742527582748245,
"loss": 2.5687,
"theoretical_loss": 3.5143556965856027,
"tokens_seen": 1508376576
},
{
"epoch": 0.46,
"learning_rate": 0.0002742427281845536,
"loss": 2.7907,
"theoretical_loss": 3.514342248574129,
"tokens_seen": 1508442112
},
{
"epoch": 0.46,
"learning_rate": 0.00027423269809428286,
"loss": 2.7999,
"theoretical_loss": 3.5143288013104925,
"tokens_seen": 1508507648
},
{
"epoch": 0.46,
"learning_rate": 0.00027422266800401204,
"loss": 2.5563,
"theoretical_loss": 3.51431535479462,
"tokens_seen": 1508573184
},
{
"epoch": 0.46,
"learning_rate": 0.0002742126379137412,
"loss": 2.6116,
"theoretical_loss": 3.514301909026437,
"tokens_seen": 1508638720
},
{
"epoch": 0.46,
"learning_rate": 0.0002742026078234704,
"loss": 2.7187,
"theoretical_loss": 3.5142884640058685,
"tokens_seen": 1508704256
},
{
"epoch": 0.46,
"learning_rate": 0.0002741925777331996,
"loss": 2.6015,
"theoretical_loss": 3.5142750197328416,
"tokens_seen": 1508769792
},
{
"epoch": 0.46,
"learning_rate": 0.00027418254764292877,
"loss": 2.7058,
"theoretical_loss": 3.514261576207282,
"tokens_seen": 1508835328
},
{
"epoch": 0.46,
"learning_rate": 0.000274172517552658,
"loss": 2.7045,
"theoretical_loss": 3.514248133429116,
"tokens_seen": 1508900864
},
{
"epoch": 0.46,
"objective/train/docs_used": 856541,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3387999534606934,
"objective/train/theoretical_loss": 3.514234691398269,
"objective/train/tokens_used": 1529426400,
"theoretical_loss": 3.514234691398269,
"tokens_seen": 1508966400
},
{
"epoch": 0.46,
"learning_rate": 0.00027416248746238713,
"loss": 2.4903,
"theoretical_loss": 3.514234691398269,
"tokens_seen": 1508966400
},
{
"epoch": 0.46,
"learning_rate": 0.00027415245737211636,
"loss": 2.7534,
"theoretical_loss": 3.514221250114667,
"tokens_seen": 1509031936
},
{
"epoch": 0.46,
"learning_rate": 0.00027414242728184554,
"loss": 2.5341,
"theoretical_loss": 3.5142078095782363,
"tokens_seen": 1509097472
},
{
"epoch": 0.46,
"learning_rate": 0.0002741323971915747,
"loss": 2.653,
"theoretical_loss": 3.514194369788903,
"tokens_seen": 1509163008
},
{
"epoch": 0.46,
"learning_rate": 0.0002741223671013039,
"loss": 2.5298,
"theoretical_loss": 3.514180930746593,
"tokens_seen": 1509228544
},
{
"epoch": 0.46,
"learning_rate": 0.0002741123370110331,
"loss": 2.9339,
"theoretical_loss": 3.5141674924512323,
"tokens_seen": 1509294080
},
{
"epoch": 0.46,
"learning_rate": 0.00027410230692076227,
"loss": 2.7479,
"theoretical_loss": 3.514154054902747,
"tokens_seen": 1509359616
},
{
"epoch": 0.46,
"learning_rate": 0.0002740922768304915,
"loss": 2.6972,
"theoretical_loss": 3.514140618101063,
"tokens_seen": 1509425152
},
{
"epoch": 0.46,
"learning_rate": 0.00027408224674022063,
"loss": 2.7004,
"theoretical_loss": 3.5141271820461064,
"tokens_seen": 1509490688
},
{
"epoch": 0.46,
"learning_rate": 0.00027407221664994987,
"loss": 2.7246,
"theoretical_loss": 3.514113746737804,
"tokens_seen": 1509556224
},
{
"epoch": 0.46,
"learning_rate": 0.000274062186559679,
"loss": 2.8796,
"theoretical_loss": 3.5141003121760805,
"tokens_seen": 1509621760
},
{
"epoch": 0.46,
"learning_rate": 0.00027405215646940823,
"loss": 2.48,
"theoretical_loss": 3.514086878360863,
"tokens_seen": 1509687296
},
{
"epoch": 0.46,
"learning_rate": 0.0002740421263791374,
"loss": 2.5326,
"theoretical_loss": 3.5140734452920777,
"tokens_seen": 1509752832
},
{
"epoch": 0.46,
"learning_rate": 0.0002740320962888666,
"loss": 2.7493,
"theoretical_loss": 3.514060012969651,
"tokens_seen": 1509818368
},
{
"epoch": 0.46,
"learning_rate": 0.0002740220661985958,
"loss": 2.7927,
"theoretical_loss": 3.5140465813935076,
"tokens_seen": 1509883904
},
{
"epoch": 0.46,
"learning_rate": 0.00027401203610832495,
"loss": 2.7351,
"theoretical_loss": 3.5140331505635745,
"tokens_seen": 1509949440
},
{
"epoch": 0.46,
"learning_rate": 0.0002740020060180542,
"loss": 2.6574,
"theoretical_loss": 3.5140197204797783,
"tokens_seen": 1510014976
},
{
"epoch": 0.46,
"learning_rate": 0.00027399197592778337,
"loss": 2.5571,
"theoretical_loss": 3.514006291142045,
"tokens_seen": 1510080512
},
{
"epoch": 0.46,
"learning_rate": 0.00027398194583751255,
"loss": 2.6473,
"theoretical_loss": 3.5139928625503,
"tokens_seen": 1510146048
},
{
"epoch": 0.46,
"learning_rate": 0.00027397191574724173,
"loss": 2.6255,
"theoretical_loss": 3.5139794347044706,
"tokens_seen": 1510211584
},
{
"epoch": 0.46,
"learning_rate": 0.00027396188565697097,
"loss": 2.8232,
"theoretical_loss": 3.513966007604482,
"tokens_seen": 1510277120
},
{
"epoch": 0.46,
"learning_rate": 0.0002739518555667001,
"loss": 2.6141,
"theoretical_loss": 3.5139525812502606,
"tokens_seen": 1510342656
},
{
"epoch": 0.46,
"learning_rate": 0.00027394182547642933,
"loss": 2.8361,
"theoretical_loss": 3.513939155641733,
"tokens_seen": 1510408192
},
{
"epoch": 0.46,
"learning_rate": 0.00027393179538615846,
"loss": 2.6005,
"theoretical_loss": 3.5139257307788254,
"tokens_seen": 1510473728
},
{
"epoch": 0.46,
"learning_rate": 0.0002739217652958877,
"loss": 2.5553,
"theoretical_loss": 3.5139123066614646,
"tokens_seen": 1510539264
},
{
"epoch": 0.46,
"objective/train/docs_used": 857299,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5837247371673584,
"objective/train/theoretical_loss": 3.513898883289575,
"objective/train/tokens_used": 1531064800,
"theoretical_loss": 3.513898883289575,
"tokens_seen": 1510604800
},
{
"epoch": 0.46,
"learning_rate": 0.00027391173520561687,
"loss": 2.7965,
"theoretical_loss": 3.513898883289575,
"tokens_seen": 1510604800
},
{
"epoch": 0.46,
"learning_rate": 0.00027390170511534605,
"loss": 2.8215,
"theoretical_loss": 3.5138854606630847,
"tokens_seen": 1510670336
},
{
"epoch": 0.46,
"learning_rate": 0.00027389167502507524,
"loss": 2.6969,
"theoretical_loss": 3.5138720387819196,
"tokens_seen": 1510735872
},
{
"epoch": 0.46,
"learning_rate": 0.0002738816449348044,
"loss": 2.7607,
"theoretical_loss": 3.5138586176460054,
"tokens_seen": 1510801408
},
{
"epoch": 0.46,
"learning_rate": 0.0002738716148445336,
"loss": 2.4389,
"theoretical_loss": 3.513845197255269,
"tokens_seen": 1510866944
},
{
"epoch": 0.46,
"learning_rate": 0.00027386158475426283,
"loss": 2.4803,
"theoretical_loss": 3.513831777609636,
"tokens_seen": 1510932480
},
{
"epoch": 0.46,
"learning_rate": 0.00027385155466399196,
"loss": 2.7642,
"theoretical_loss": 3.513818358709033,
"tokens_seen": 1510998016
},
{
"epoch": 0.46,
"learning_rate": 0.0002738415245737212,
"loss": 2.6038,
"theoretical_loss": 3.5138049405533867,
"tokens_seen": 1511063552
},
{
"epoch": 0.46,
"learning_rate": 0.0002738314944834504,
"loss": 2.5003,
"theoretical_loss": 3.5137915231426233,
"tokens_seen": 1511129088
},
{
"epoch": 0.46,
"learning_rate": 0.00027382146439317956,
"loss": 2.5692,
"theoretical_loss": 3.5137781064766687,
"tokens_seen": 1511194624
},
{
"epoch": 0.46,
"learning_rate": 0.00027381143430290874,
"loss": 2.828,
"theoretical_loss": 3.51376469055545,
"tokens_seen": 1511260160
},
{
"epoch": 0.46,
"learning_rate": 0.0002738014042126379,
"loss": 2.7565,
"theoretical_loss": 3.513751275378893,
"tokens_seen": 1511325696
},
{
"epoch": 0.46,
"learning_rate": 0.0002737913741223671,
"loss": 2.597,
"theoretical_loss": 3.5137378609469243,
"tokens_seen": 1511391232
},
{
"epoch": 0.46,
"learning_rate": 0.00027378134403209634,
"loss": 2.4652,
"theoretical_loss": 3.51372444725947,
"tokens_seen": 1511456768
},
{
"epoch": 0.46,
"learning_rate": 0.00027377131394182546,
"loss": 2.419,
"theoretical_loss": 3.513711034316457,
"tokens_seen": 1511522304
},
{
"epoch": 0.46,
"learning_rate": 0.0002737612838515547,
"loss": 2.7526,
"theoretical_loss": 3.513697622117811,
"tokens_seen": 1511587840
},
{
"epoch": 0.46,
"learning_rate": 0.0002737512537612838,
"loss": 2.718,
"theoretical_loss": 3.5136842106634596,
"tokens_seen": 1511653376
},
{
"epoch": 0.46,
"learning_rate": 0.00027374122367101306,
"loss": 2.6191,
"theoretical_loss": 3.513670799953328,
"tokens_seen": 1511718912
},
{
"epoch": 0.46,
"learning_rate": 0.00027373119358074224,
"loss": 2.6107,
"theoretical_loss": 3.513657389987343,
"tokens_seen": 1511784448
},
{
"epoch": 0.46,
"learning_rate": 0.0002737211634904714,
"loss": 2.3703,
"theoretical_loss": 3.5136439807654316,
"tokens_seen": 1511849984
},
{
"epoch": 0.46,
"learning_rate": 0.0002737111334002006,
"loss": 2.8001,
"theoretical_loss": 3.5136305722875196,
"tokens_seen": 1511915520
},
{
"epoch": 0.46,
"learning_rate": 0.0002737011033099298,
"loss": 2.4589,
"theoretical_loss": 3.5136171645535335,
"tokens_seen": 1511981056
},
{
"epoch": 0.46,
"learning_rate": 0.00027369107321965897,
"loss": 2.5535,
"theoretical_loss": 3.5136037575634003,
"tokens_seen": 1512046592
},
{
"epoch": 0.46,
"learning_rate": 0.0002736810431293882,
"loss": 2.8103,
"theoretical_loss": 3.5135903513170463,
"tokens_seen": 1512112128
},
{
"epoch": 0.46,
"learning_rate": 0.00027367101303911733,
"loss": 2.5683,
"theoretical_loss": 3.513576945814398,
"tokens_seen": 1512177664
},
{
"epoch": 0.46,
"objective/train/docs_used": 858636,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4716200828552246,
"objective/train/theoretical_loss": 3.5135635410553814,
"objective/train/tokens_used": 1532703200,
"theoretical_loss": 3.5135635410553814,
"tokens_seen": 1512243200
},
{
"epoch": 0.46,
"learning_rate": 0.00027366098294884656,
"loss": 2.7043,
"theoretical_loss": 3.5135635410553814,
"tokens_seen": 1512243200
},
{
"epoch": 0.46,
"learning_rate": 0.00027365095285857574,
"loss": 2.6529,
"theoretical_loss": 3.513550137039924,
"tokens_seen": 1512308736
},
{
"epoch": 0.46,
"learning_rate": 0.0002736409227683049,
"loss": 2.4266,
"theoretical_loss": 3.513536733767951,
"tokens_seen": 1512374272
},
{
"epoch": 0.46,
"learning_rate": 0.0002736308926780341,
"loss": 2.6543,
"theoretical_loss": 3.5135233312393903,
"tokens_seen": 1512439808
},
{
"epoch": 0.46,
"learning_rate": 0.0002736208625877633,
"loss": 2.4313,
"theoretical_loss": 3.513509929454168,
"tokens_seen": 1512505344
},
{
"epoch": 0.46,
"learning_rate": 0.00027361083249749247,
"loss": 2.5758,
"theoretical_loss": 3.51349652841221,
"tokens_seen": 1512570880
},
{
"epoch": 0.46,
"learning_rate": 0.0002736008024072217,
"loss": 2.5588,
"theoretical_loss": 3.5134831281134433,
"tokens_seen": 1512636416
},
{
"epoch": 0.46,
"learning_rate": 0.00027359077231695083,
"loss": 2.4961,
"theoretical_loss": 3.513469728557795,
"tokens_seen": 1512701952
},
{
"epoch": 0.46,
"learning_rate": 0.00027358074222668007,
"loss": 2.6638,
"theoretical_loss": 3.5134563297451917,
"tokens_seen": 1512767488
},
{
"epoch": 0.46,
"learning_rate": 0.0002735707121364092,
"loss": 2.7532,
"theoretical_loss": 3.5134429316755593,
"tokens_seen": 1512833024
},
{
"epoch": 0.46,
"learning_rate": 0.00027356068204613843,
"loss": 2.8208,
"theoretical_loss": 3.513429534348824,
"tokens_seen": 1512898560
},
{
"epoch": 0.46,
"learning_rate": 0.0002735506519558676,
"loss": 2.7483,
"theoretical_loss": 3.5134161377649145,
"tokens_seen": 1512964096
},
{
"epoch": 0.46,
"learning_rate": 0.0002735406218655968,
"loss": 2.558,
"theoretical_loss": 3.5134027419237555,
"tokens_seen": 1513029632
},
{
"epoch": 0.46,
"learning_rate": 0.00027353059177532597,
"loss": 2.657,
"theoretical_loss": 3.513389346825274,
"tokens_seen": 1513095168
},
{
"epoch": 0.46,
"learning_rate": 0.00027352056168505515,
"loss": 2.8269,
"theoretical_loss": 3.513375952469397,
"tokens_seen": 1513160704
},
{
"epoch": 0.46,
"learning_rate": 0.00027351053159478433,
"loss": 2.7839,
"theoretical_loss": 3.513362558856051,
"tokens_seen": 1513226240
},
{
"epoch": 0.46,
"learning_rate": 0.00027350050150451357,
"loss": 2.5994,
"theoretical_loss": 3.5133491659851632,
"tokens_seen": 1513291776
},
{
"epoch": 0.46,
"learning_rate": 0.0002734904714142427,
"loss": 2.3467,
"theoretical_loss": 3.5133357738566597,
"tokens_seen": 1513357312
},
{
"epoch": 0.46,
"learning_rate": 0.00027348044132397193,
"loss": 2.731,
"theoretical_loss": 3.5133223824704674,
"tokens_seen": 1513422848
},
{
"epoch": 0.46,
"learning_rate": 0.0002734704112337011,
"loss": 2.7984,
"theoretical_loss": 3.513308991826513,
"tokens_seen": 1513488384
},
{
"epoch": 0.46,
"learning_rate": 0.0002734603811434303,
"loss": 2.5926,
"theoretical_loss": 3.513295601924723,
"tokens_seen": 1513553920
},
{
"epoch": 0.46,
"learning_rate": 0.0002734503510531595,
"loss": 2.8396,
"theoretical_loss": 3.513282212765024,
"tokens_seen": 1513619456
},
{
"epoch": 0.46,
"learning_rate": 0.00027344032096288866,
"loss": 2.5969,
"theoretical_loss": 3.5132688243473438,
"tokens_seen": 1513684992
},
{
"epoch": 0.46,
"learning_rate": 0.00027343029087261784,
"loss": 2.6251,
"theoretical_loss": 3.5132554366716082,
"tokens_seen": 1513750528
},
{
"epoch": 0.46,
"learning_rate": 0.0002734202607823471,
"loss": 2.881,
"theoretical_loss": 3.5132420497377437,
"tokens_seen": 1513816064
},
{
"epoch": 0.46,
"objective/train/docs_used": 859135,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.85482120513916,
"objective/train/theoretical_loss": 3.513228663545678,
"objective/train/tokens_used": 1534341600,
"theoretical_loss": 3.513228663545678,
"tokens_seen": 1513881600
},
{
"epoch": 0.46,
"learning_rate": 0.0002734102306920762,
"loss": 2.7353,
"theoretical_loss": 3.513228663545678,
"tokens_seen": 1513881600
},
{
"epoch": 0.46,
"learning_rate": 0.00027340020060180544,
"loss": 2.6836,
"theoretical_loss": 3.5132152780953376,
"tokens_seen": 1513947136
},
{
"epoch": 0.46,
"learning_rate": 0.00027339017051153456,
"loss": 2.7888,
"theoretical_loss": 3.513201893386649,
"tokens_seen": 1514012672
},
{
"epoch": 0.46,
"learning_rate": 0.0002733801404212638,
"loss": 2.7657,
"theoretical_loss": 3.513188509419539,
"tokens_seen": 1514078208
},
{
"epoch": 0.46,
"learning_rate": 0.000273370110330993,
"loss": 2.6271,
"theoretical_loss": 3.513175126193935,
"tokens_seen": 1514143744
},
{
"epoch": 0.46,
"learning_rate": 0.00027336008024072216,
"loss": 2.8075,
"theoretical_loss": 3.5131617437097633,
"tokens_seen": 1514209280
},
{
"epoch": 0.46,
"learning_rate": 0.00027335005015045134,
"loss": 2.7076,
"theoretical_loss": 3.5131483619669504,
"tokens_seen": 1514274816
},
{
"epoch": 0.46,
"learning_rate": 0.0002733400200601806,
"loss": 2.4047,
"theoretical_loss": 3.513134980965424,
"tokens_seen": 1514340352
},
{
"epoch": 0.46,
"learning_rate": 0.0002733299899699097,
"loss": 2.768,
"theoretical_loss": 3.5131216007051105,
"tokens_seen": 1514405888
},
{
"epoch": 0.46,
"learning_rate": 0.00027331995987963894,
"loss": 2.5087,
"theoretical_loss": 3.513108221185937,
"tokens_seen": 1514471424
},
{
"epoch": 0.46,
"learning_rate": 0.00027330992978936807,
"loss": 2.7685,
"theoretical_loss": 3.5130948424078294,
"tokens_seen": 1514536960
},
{
"epoch": 0.46,
"learning_rate": 0.0002732998996990973,
"loss": 2.6438,
"theoretical_loss": 3.5130814643707162,
"tokens_seen": 1514602496
},
{
"epoch": 0.46,
"learning_rate": 0.0002732898696088265,
"loss": 2.3986,
"theoretical_loss": 3.5130680870745232,
"tokens_seen": 1514668032
},
{
"epoch": 0.46,
"learning_rate": 0.00027327983951855566,
"loss": 2.5234,
"theoretical_loss": 3.5130547105191776,
"tokens_seen": 1514733568
},
{
"epoch": 0.46,
"learning_rate": 0.0002732698094282849,
"loss": 2.4483,
"theoretical_loss": 3.5130413347046066,
"tokens_seen": 1514799104
},
{
"epoch": 0.46,
"learning_rate": 0.000273259779338014,
"loss": 2.7852,
"theoretical_loss": 3.5130279596307363,
"tokens_seen": 1514864640
},
{
"epoch": 0.46,
"learning_rate": 0.00027324974924774326,
"loss": 2.67,
"theoretical_loss": 3.513014585297494,
"tokens_seen": 1514930176
},
{
"epoch": 0.46,
"learning_rate": 0.00027323971915747244,
"loss": 2.6812,
"theoretical_loss": 3.5130012117048075,
"tokens_seen": 1514995712
},
{
"epoch": 0.46,
"learning_rate": 0.0002732296890672016,
"loss": 2.4001,
"theoretical_loss": 3.5129878388526032,
"tokens_seen": 1515061248
},
{
"epoch": 0.46,
"learning_rate": 0.0002732196589769308,
"loss": 2.7063,
"theoretical_loss": 3.5129744667408076,
"tokens_seen": 1515126784
},
{
"epoch": 0.46,
"learning_rate": 0.00027320962888666,
"loss": 2.5908,
"theoretical_loss": 3.5129610953693478,
"tokens_seen": 1515192320
},
{
"epoch": 0.46,
"learning_rate": 0.00027319959879638917,
"loss": 2.6996,
"theoretical_loss": 3.5129477247381518,
"tokens_seen": 1515257856
},
{
"epoch": 0.46,
"learning_rate": 0.0002731895687061184,
"loss": 2.5592,
"theoretical_loss": 3.512934354847145,
"tokens_seen": 1515323392
},
{
"epoch": 0.46,
"learning_rate": 0.00027317953861584753,
"loss": 2.7911,
"theoretical_loss": 3.5129209856962555,
"tokens_seen": 1515388928
},
{
"epoch": 0.46,
"learning_rate": 0.00027316950852557676,
"loss": 2.6629,
"theoretical_loss": 3.5129076172854106,
"tokens_seen": 1515454464
},
{
"epoch": 0.46,
"objective/train/docs_used": 860397,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3560235500335693,
"objective/train/theoretical_loss": 3.5128942496145363,
"objective/train/tokens_used": 1535980000,
"theoretical_loss": 3.5128942496145363,
"tokens_seen": 1515520000
},
{
"epoch": 0.46,
"learning_rate": 0.00027315947843530594,
"loss": 2.651,
"theoretical_loss": 3.5128942496145363,
"tokens_seen": 1515520000
},
{
"epoch": 0.46,
"learning_rate": 0.0002731494483450351,
"loss": 3.004,
"theoretical_loss": 3.5128808826835605,
"tokens_seen": 1515585536
},
{
"epoch": 0.46,
"learning_rate": 0.0002731394182547643,
"loss": 2.6027,
"theoretical_loss": 3.5128675164924097,
"tokens_seen": 1515651072
},
{
"epoch": 0.46,
"learning_rate": 0.0002731293881644935,
"loss": 2.6915,
"theoretical_loss": 3.5128541510410116,
"tokens_seen": 1515716608
},
{
"epoch": 0.46,
"learning_rate": 0.00027311935807422267,
"loss": 2.6213,
"theoretical_loss": 3.5128407863292925,
"tokens_seen": 1515782144
},
{
"epoch": 0.46,
"learning_rate": 0.0002731093279839519,
"loss": 2.6825,
"theoretical_loss": 3.51282742235718,
"tokens_seen": 1515847680
},
{
"epoch": 0.46,
"learning_rate": 0.00027309929789368103,
"loss": 2.6402,
"theoretical_loss": 3.512814059124601,
"tokens_seen": 1515913216
},
{
"epoch": 0.46,
"learning_rate": 0.00027308926780341027,
"loss": 2.7184,
"theoretical_loss": 3.5128006966314826,
"tokens_seen": 1515978752
},
{
"epoch": 0.46,
"learning_rate": 0.0002730792377131394,
"loss": 2.8482,
"theoretical_loss": 3.5127873348777516,
"tokens_seen": 1516044288
},
{
"epoch": 0.46,
"learning_rate": 0.00027306920762286863,
"loss": 2.6866,
"theoretical_loss": 3.512773973863336,
"tokens_seen": 1516109824
},
{
"epoch": 0.46,
"learning_rate": 0.0002730591775325978,
"loss": 2.5805,
"theoretical_loss": 3.5127606135881626,
"tokens_seen": 1516175360
},
{
"epoch": 0.46,
"learning_rate": 0.000273049147442327,
"loss": 2.7237,
"theoretical_loss": 3.512747254052158,
"tokens_seen": 1516240896
},
{
"epoch": 0.46,
"learning_rate": 0.00027303911735205617,
"loss": 2.7322,
"theoretical_loss": 3.51273389525525,
"tokens_seen": 1516306432
},
{
"epoch": 0.46,
"learning_rate": 0.00027302908726178535,
"loss": 2.7344,
"theoretical_loss": 3.512720537197365,
"tokens_seen": 1516371968
},
{
"epoch": 0.46,
"learning_rate": 0.00027301905717151453,
"loss": 2.7319,
"theoretical_loss": 3.512707179878431,
"tokens_seen": 1516437504
},
{
"epoch": 0.46,
"learning_rate": 0.00027300902708124377,
"loss": 2.6716,
"theoretical_loss": 3.512693823298375,
"tokens_seen": 1516503040
},
{
"epoch": 0.46,
"learning_rate": 0.0002729989969909729,
"loss": 2.6373,
"theoretical_loss": 3.512680467457124,
"tokens_seen": 1516568576
},
{
"epoch": 0.46,
"learning_rate": 0.00027298896690070213,
"loss": 2.7732,
"theoretical_loss": 3.5126671123546056,
"tokens_seen": 1516634112
},
{
"epoch": 0.46,
"learning_rate": 0.0002729789368104313,
"loss": 2.4123,
"theoretical_loss": 3.5126537579907464,
"tokens_seen": 1516699648
},
{
"epoch": 0.46,
"learning_rate": 0.0002729689067201605,
"loss": 2.5421,
"theoretical_loss": 3.5126404043654738,
"tokens_seen": 1516765184
},
{
"epoch": 0.46,
"learning_rate": 0.0002729588766298897,
"loss": 2.8211,
"theoretical_loss": 3.5126270514787157,
"tokens_seen": 1516830720
},
{
"epoch": 0.46,
"learning_rate": 0.00027294884653961886,
"loss": 2.5806,
"theoretical_loss": 3.5126136993303985,
"tokens_seen": 1516896256
},
{
"epoch": 0.46,
"learning_rate": 0.00027293881644934804,
"loss": 2.5996,
"theoretical_loss": 3.5126003479204497,
"tokens_seen": 1516961792
},
{
"epoch": 0.46,
"learning_rate": 0.0002729287863590773,
"loss": 2.6083,
"theoretical_loss": 3.512586997248797,
"tokens_seen": 1517027328
},
{
"epoch": 0.46,
"learning_rate": 0.0002729187562688064,
"loss": 2.5585,
"theoretical_loss": 3.5125736473153673,
"tokens_seen": 1517092864
},
{
"epoch": 0.46,
"objective/train/docs_used": 860980,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.952728748321533,
"objective/train/theoretical_loss": 3.5125602981200874,
"objective/train/tokens_used": 1537618400,
"theoretical_loss": 3.5125602981200874,
"tokens_seen": 1517158400
},
{
"epoch": 0.46,
"learning_rate": 0.00027290872617853564,
"loss": 2.6249,
"theoretical_loss": 3.5125602981200874,
"tokens_seen": 1517158400
},
{
"epoch": 0.46,
"learning_rate": 0.00027289869608826476,
"loss": 2.6746,
"theoretical_loss": 3.5125469496628856,
"tokens_seen": 1517223936
},
{
"epoch": 0.46,
"learning_rate": 0.000272888665997994,
"loss": 2.8637,
"theoretical_loss": 3.512533601943689,
"tokens_seen": 1517289472
},
{
"epoch": 0.46,
"learning_rate": 0.0002728786359077232,
"loss": 2.6188,
"theoretical_loss": 3.512520254962424,
"tokens_seen": 1517355008
},
{
"epoch": 0.46,
"learning_rate": 0.00027286860581745236,
"loss": 2.7476,
"theoretical_loss": 3.5125069087190193,
"tokens_seen": 1517420544
},
{
"epoch": 0.46,
"learning_rate": 0.00027285857572718154,
"loss": 2.8184,
"theoretical_loss": 3.5124935632134013,
"tokens_seen": 1517486080
},
{
"epoch": 0.46,
"learning_rate": 0.0002728485456369108,
"loss": 2.7359,
"theoretical_loss": 3.5124802184454973,
"tokens_seen": 1517551616
},
{
"epoch": 0.46,
"learning_rate": 0.0002728385155466399,
"loss": 2.5872,
"theoretical_loss": 3.512466874415235,
"tokens_seen": 1517617152
},
{
"epoch": 0.46,
"learning_rate": 0.00027282848545636914,
"loss": 2.5372,
"theoretical_loss": 3.512453531122542,
"tokens_seen": 1517682688
},
{
"epoch": 0.46,
"learning_rate": 0.00027281845536609827,
"loss": 2.6663,
"theoretical_loss": 3.512440188567346,
"tokens_seen": 1517748224
},
{
"epoch": 0.46,
"learning_rate": 0.0002728084252758275,
"loss": 2.6228,
"theoretical_loss": 3.512426846749573,
"tokens_seen": 1517813760
},
{
"epoch": 0.46,
"learning_rate": 0.0002727983951855567,
"loss": 2.8295,
"theoretical_loss": 3.512413505669152,
"tokens_seen": 1517879296
},
{
"epoch": 0.46,
"learning_rate": 0.00027278836509528586,
"loss": 2.7251,
"theoretical_loss": 3.5124001653260093,
"tokens_seen": 1517944832
},
{
"epoch": 0.46,
"learning_rate": 0.00027277833500501504,
"loss": 2.7176,
"theoretical_loss": 3.512386825720072,
"tokens_seen": 1518010368
},
{
"epoch": 0.46,
"learning_rate": 0.0002727683049147442,
"loss": 2.4118,
"theoretical_loss": 3.5123734868512693,
"tokens_seen": 1518075904
},
{
"epoch": 0.46,
"learning_rate": 0.0002727582748244734,
"loss": 2.7872,
"theoretical_loss": 3.512360148719527,
"tokens_seen": 1518141440
},
{
"epoch": 0.46,
"learning_rate": 0.00027274824473420264,
"loss": 2.7048,
"theoretical_loss": 3.5123468113247736,
"tokens_seen": 1518206976
},
{
"epoch": 0.46,
"learning_rate": 0.00027273821464393177,
"loss": 2.4985,
"theoretical_loss": 3.5123334746669355,
"tokens_seen": 1518272512
},
{
"epoch": 0.46,
"learning_rate": 0.000272728184553661,
"loss": 2.5145,
"theoretical_loss": 3.512320138745941,
"tokens_seen": 1518338048
},
{
"epoch": 0.46,
"learning_rate": 0.00027271815446339013,
"loss": 2.9478,
"theoretical_loss": 3.5123068035617173,
"tokens_seen": 1518403584
},
{
"epoch": 0.46,
"learning_rate": 0.00027270812437311937,
"loss": 2.7021,
"theoretical_loss": 3.512293469114192,
"tokens_seen": 1518469120
},
{
"epoch": 0.46,
"learning_rate": 0.00027269809428284855,
"loss": 2.49,
"theoretical_loss": 3.5122801354032926,
"tokens_seen": 1518534656
},
{
"epoch": 0.46,
"learning_rate": 0.00027268806419257773,
"loss": 2.5551,
"theoretical_loss": 3.5122668024289467,
"tokens_seen": 1518600192
},
{
"epoch": 0.46,
"learning_rate": 0.0002726780341023069,
"loss": 2.7098,
"theoretical_loss": 3.5122534701910815,
"tokens_seen": 1518665728
},
{
"epoch": 0.46,
"learning_rate": 0.00027266800401203614,
"loss": 2.7791,
"theoretical_loss": 3.512240138689625,
"tokens_seen": 1518731264
},
{
"epoch": 0.46,
"objective/train/docs_used": 862373,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1491739749908447,
"objective/train/theoretical_loss": 3.5122268079245043,
"objective/train/tokens_used": 1539256800,
"theoretical_loss": 3.5122268079245043,
"tokens_seen": 1518796800
},
{
"epoch": 0.46,
"learning_rate": 0.00027265797392176527,
"loss": 2.4443,
"theoretical_loss": 3.5122268079245043,
"tokens_seen": 1518796800
},
{
"epoch": 0.46,
"learning_rate": 0.0002726479438314945,
"loss": 2.4448,
"theoretical_loss": 3.5122134778956475,
"tokens_seen": 1518862336
},
{
"epoch": 0.46,
"learning_rate": 0.00027263791374122363,
"loss": 2.7196,
"theoretical_loss": 3.5122001486029815,
"tokens_seen": 1518927872
},
{
"epoch": 0.46,
"learning_rate": 0.00027262788365095287,
"loss": 2.4489,
"theoretical_loss": 3.512186820046434,
"tokens_seen": 1518993408
},
{
"epoch": 0.46,
"learning_rate": 0.00027261785356068205,
"loss": 2.7707,
"theoretical_loss": 3.512173492225933,
"tokens_seen": 1519058944
},
{
"epoch": 0.46,
"learning_rate": 0.00027260782347041123,
"loss": 2.6532,
"theoretical_loss": 3.512160165141406,
"tokens_seen": 1519124480
},
{
"epoch": 0.46,
"learning_rate": 0.0002725977933801404,
"loss": 2.5755,
"theoretical_loss": 3.5121468387927806,
"tokens_seen": 1519190016
},
{
"epoch": 0.46,
"learning_rate": 0.0002725877632898696,
"loss": 2.6937,
"theoretical_loss": 3.5121335131799847,
"tokens_seen": 1519255552
},
{
"epoch": 0.46,
"learning_rate": 0.0002725777331995988,
"loss": 2.4515,
"theoretical_loss": 3.512120188302945,
"tokens_seen": 1519321088
},
{
"epoch": 0.46,
"learning_rate": 0.000272567703109328,
"loss": 2.7293,
"theoretical_loss": 3.5121068641615896,
"tokens_seen": 1519386624
},
{
"epoch": 0.46,
"learning_rate": 0.00027255767301905714,
"loss": 2.6014,
"theoretical_loss": 3.512093540755847,
"tokens_seen": 1519452160
},
{
"epoch": 0.46,
"learning_rate": 0.00027254764292878637,
"loss": 2.5583,
"theoretical_loss": 3.5120802180856434,
"tokens_seen": 1519517696
},
{
"epoch": 0.46,
"learning_rate": 0.0002725376128385155,
"loss": 2.4821,
"theoretical_loss": 3.512066896150907,
"tokens_seen": 1519583232
},
{
"epoch": 0.46,
"learning_rate": 0.00027252758274824473,
"loss": 2.6818,
"theoretical_loss": 3.5120535749515667,
"tokens_seen": 1519648768
},
{
"epoch": 0.46,
"learning_rate": 0.00027251755265797397,
"loss": 2.834,
"theoretical_loss": 3.5120402544875486,
"tokens_seen": 1519714304
},
{
"epoch": 0.46,
"learning_rate": 0.0002725075225677031,
"loss": 2.8,
"theoretical_loss": 3.512026934758781,
"tokens_seen": 1519779840
},
{
"epoch": 0.46,
"learning_rate": 0.00027249749247743233,
"loss": 2.792,
"theoretical_loss": 3.5120136157651913,
"tokens_seen": 1519845376
},
{
"epoch": 0.46,
"learning_rate": 0.0002724874623871615,
"loss": 2.4946,
"theoretical_loss": 3.5120002975067077,
"tokens_seen": 1519910912
},
{
"epoch": 0.46,
"learning_rate": 0.0002724774322968907,
"loss": 2.6387,
"theoretical_loss": 3.5119869799832584,
"tokens_seen": 1519976448
},
{
"epoch": 0.46,
"learning_rate": 0.0002724674022066199,
"loss": 2.5767,
"theoretical_loss": 3.5119736631947696,
"tokens_seen": 1520041984
},
{
"epoch": 0.46,
"learning_rate": 0.00027245737211634906,
"loss": 2.5785,
"theoretical_loss": 3.5119603471411702,
"tokens_seen": 1520107520
},
{
"epoch": 0.46,
"learning_rate": 0.00027244734202607824,
"loss": 2.599,
"theoretical_loss": 3.511947031822388,
"tokens_seen": 1520173056
},
{
"epoch": 0.46,
"learning_rate": 0.0002724373119358075,
"loss": 2.5924,
"theoretical_loss": 3.5119337172383505,
"tokens_seen": 1520238592
},
{
"epoch": 0.46,
"learning_rate": 0.0002724272818455366,
"loss": 2.6004,
"theoretical_loss": 3.5119204033889853,
"tokens_seen": 1520304128
},
{
"epoch": 0.46,
"learning_rate": 0.00027241725175526584,
"loss": 2.7071,
"theoretical_loss": 3.51190709027422,
"tokens_seen": 1520369664
},
{
"epoch": 0.46,
"objective/train/docs_used": 862945,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8024189472198486,
"objective/train/theoretical_loss": 3.5118937778939836,
"objective/train/tokens_used": 1540895200,
"theoretical_loss": 3.5118937778939836,
"tokens_seen": 1520435200
},
{
"epoch": 0.46,
"learning_rate": 0.00027240722166499496,
"loss": 2.6411,
"theoretical_loss": 3.5118937778939836,
"tokens_seen": 1520435200
},
{
"epoch": 0.46,
"learning_rate": 0.0002723971915747242,
"loss": 2.5435,
"theoretical_loss": 3.5118804662482024,
"tokens_seen": 1520500736
},
{
"epoch": 0.46,
"learning_rate": 0.0002723871614844534,
"loss": 2.5784,
"theoretical_loss": 3.5118671553368053,
"tokens_seen": 1520566272
},
{
"epoch": 0.46,
"learning_rate": 0.00027237713139418256,
"loss": 2.4846,
"theoretical_loss": 3.5118538451597194,
"tokens_seen": 1520631808
},
{
"epoch": 0.46,
"learning_rate": 0.00027236710130391174,
"loss": 2.5313,
"theoretical_loss": 3.511840535716873,
"tokens_seen": 1520697344
},
{
"epoch": 0.46,
"learning_rate": 0.000272357071213641,
"loss": 2.4843,
"theoretical_loss": 3.511827227008194,
"tokens_seen": 1520762880
},
{
"epoch": 0.46,
"learning_rate": 0.0002723470411233701,
"loss": 2.4172,
"theoretical_loss": 3.51181391903361,
"tokens_seen": 1520828416
},
{
"epoch": 0.46,
"learning_rate": 0.00027233701103309934,
"loss": 2.3251,
"theoretical_loss": 3.511800611793049,
"tokens_seen": 1520893952
},
{
"epoch": 0.46,
"learning_rate": 0.00027232698094282847,
"loss": 2.4814,
"theoretical_loss": 3.5117873052864392,
"tokens_seen": 1520959488
},
{
"epoch": 0.46,
"learning_rate": 0.0002723169508525577,
"loss": 2.6748,
"theoretical_loss": 3.511773999513708,
"tokens_seen": 1521025024
},
{
"epoch": 0.46,
"learning_rate": 0.0002723069207622869,
"loss": 2.5866,
"theoretical_loss": 3.5117606944747832,
"tokens_seen": 1521090560
},
{
"epoch": 0.46,
"learning_rate": 0.00027229689067201606,
"loss": 2.6666,
"theoretical_loss": 3.5117473901695933,
"tokens_seen": 1521156096
},
{
"epoch": 0.46,
"learning_rate": 0.00027228686058174524,
"loss": 2.7455,
"theoretical_loss": 3.511734086598066,
"tokens_seen": 1521221632
},
{
"epoch": 0.46,
"learning_rate": 0.0002722768304914744,
"loss": 2.6409,
"theoretical_loss": 3.511720783760129,
"tokens_seen": 1521287168
},
{
"epoch": 0.46,
"learning_rate": 0.0002722668004012036,
"loss": 2.6222,
"theoretical_loss": 3.5117074816557103,
"tokens_seen": 1521352704
},
{
"epoch": 0.46,
"learning_rate": 0.00027225677031093284,
"loss": 2.8297,
"theoretical_loss": 3.511694180284738,
"tokens_seen": 1521418240
},
{
"epoch": 0.46,
"learning_rate": 0.00027224674022066197,
"loss": 2.4674,
"theoretical_loss": 3.51168087964714,
"tokens_seen": 1521483776
},
{
"epoch": 0.46,
"learning_rate": 0.0002722367101303912,
"loss": 2.7031,
"theoretical_loss": 3.511667579742845,
"tokens_seen": 1521549312
},
{
"epoch": 0.46,
"learning_rate": 0.00027222668004012033,
"loss": 2.6709,
"theoretical_loss": 3.5116542805717796,
"tokens_seen": 1521614848
},
{
"epoch": 0.46,
"learning_rate": 0.00027221664994984957,
"loss": 3.0388,
"theoretical_loss": 3.5116409821338728,
"tokens_seen": 1521680384
},
{
"epoch": 0.46,
"learning_rate": 0.00027220661985957875,
"loss": 2.7709,
"theoretical_loss": 3.5116276844290524,
"tokens_seen": 1521745920
},
{
"epoch": 0.46,
"learning_rate": 0.00027219658976930793,
"loss": 2.3933,
"theoretical_loss": 3.511614387457246,
"tokens_seen": 1521811456
},
{
"epoch": 0.46,
"learning_rate": 0.0002721865596790371,
"loss": 2.6083,
"theoretical_loss": 3.511601091218382,
"tokens_seen": 1521876992
},
{
"epoch": 0.46,
"learning_rate": 0.00027217652958876635,
"loss": 2.7458,
"theoretical_loss": 3.511587795712389,
"tokens_seen": 1521942528
},
{
"epoch": 0.46,
"learning_rate": 0.00027216649949849547,
"loss": 2.6489,
"theoretical_loss": 3.5115745009391937,
"tokens_seen": 1522008064
},
{
"epoch": 0.46,
"objective/train/docs_used": 863367,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5443220138549805,
"objective/train/theoretical_loss": 3.5115612068987256,
"objective/train/tokens_used": 1542533600,
"theoretical_loss": 3.5115612068987256,
"tokens_seen": 1522073600
},
{
"epoch": 0.46,
"learning_rate": 0.0002721564694082247,
"loss": 2.6273,
"theoretical_loss": 3.5115612068987256,
"tokens_seen": 1522073600
},
{
"epoch": 0.46,
"learning_rate": 0.00027214643931795383,
"loss": 2.6306,
"theoretical_loss": 3.5115479135909116,
"tokens_seen": 1522139136
},
{
"epoch": 0.46,
"learning_rate": 0.00027213640922768307,
"loss": 2.6636,
"theoretical_loss": 3.5115346210156804,
"tokens_seen": 1522204672
},
{
"epoch": 0.46,
"learning_rate": 0.00027212637913741225,
"loss": 2.7407,
"theoretical_loss": 3.51152132917296,
"tokens_seen": 1522270208
},
{
"epoch": 0.46,
"learning_rate": 0.00027211634904714143,
"loss": 2.7171,
"theoretical_loss": 3.5115080380626784,
"tokens_seen": 1522335744
},
{
"epoch": 0.46,
"learning_rate": 0.0002721063189568706,
"loss": 2.6572,
"theoretical_loss": 3.5114947476847638,
"tokens_seen": 1522401280
},
{
"epoch": 0.46,
"learning_rate": 0.0002720962888665998,
"loss": 2.7933,
"theoretical_loss": 3.5114814580391442,
"tokens_seen": 1522466816
},
{
"epoch": 0.46,
"learning_rate": 0.000272086258776329,
"loss": 2.6271,
"theoretical_loss": 3.5114681691257483,
"tokens_seen": 1522532352
},
{
"epoch": 0.46,
"learning_rate": 0.0002720762286860582,
"loss": 2.6584,
"theoretical_loss": 3.5114548809445036,
"tokens_seen": 1522597888
},
{
"epoch": 0.46,
"learning_rate": 0.00027206619859578734,
"loss": 2.5067,
"theoretical_loss": 3.511441593495338,
"tokens_seen": 1522663424
},
{
"epoch": 0.46,
"learning_rate": 0.00027205616850551657,
"loss": 2.5795,
"theoretical_loss": 3.51142830677818,
"tokens_seen": 1522728960
},
{
"epoch": 0.46,
"learning_rate": 0.0002720461384152457,
"loss": 2.7684,
"theoretical_loss": 3.511415020792959,
"tokens_seen": 1522794496
},
{
"epoch": 0.46,
"learning_rate": 0.00027203610832497493,
"loss": 2.6176,
"theoretical_loss": 3.511401735539601,
"tokens_seen": 1522860032
},
{
"epoch": 0.46,
"learning_rate": 0.0002720260782347041,
"loss": 2.63,
"theoretical_loss": 3.5113884510180355,
"tokens_seen": 1522925568
},
{
"epoch": 0.46,
"learning_rate": 0.0002720160481444333,
"loss": 2.4433,
"theoretical_loss": 3.51137516722819,
"tokens_seen": 1522991104
},
{
"epoch": 0.46,
"learning_rate": 0.0002720060180541625,
"loss": 2.6799,
"theoretical_loss": 3.5113618841699936,
"tokens_seen": 1523056640
},
{
"epoch": 0.46,
"learning_rate": 0.0002719959879638917,
"loss": 2.7472,
"theoretical_loss": 3.511348601843374,
"tokens_seen": 1523122176
},
{
"epoch": 0.46,
"learning_rate": 0.00027198595787362084,
"loss": 2.4982,
"theoretical_loss": 3.51133532024826,
"tokens_seen": 1523187712
},
{
"epoch": 0.46,
"learning_rate": 0.0002719759277833501,
"loss": 2.8467,
"theoretical_loss": 3.511322039384578,
"tokens_seen": 1523253248
},
{
"epoch": 0.46,
"learning_rate": 0.0002719658976930792,
"loss": 2.7105,
"theoretical_loss": 3.511308759252259,
"tokens_seen": 1523318784
},
{
"epoch": 0.46,
"learning_rate": 0.00027195586760280844,
"loss": 2.5429,
"theoretical_loss": 3.5112954798512286,
"tokens_seen": 1523384320
},
{
"epoch": 0.46,
"learning_rate": 0.0002719458375125376,
"loss": 2.6304,
"theoretical_loss": 3.5112822011814173,
"tokens_seen": 1523449856
},
{
"epoch": 0.46,
"learning_rate": 0.0002719358074222668,
"loss": 2.5878,
"theoretical_loss": 3.511268923242752,
"tokens_seen": 1523515392
},
{
"epoch": 0.46,
"learning_rate": 0.000271925777331996,
"loss": 2.8193,
"theoretical_loss": 3.5112556460351616,
"tokens_seen": 1523580928
},
{
"epoch": 0.46,
"learning_rate": 0.00027191574724172516,
"loss": 2.7581,
"theoretical_loss": 3.511242369558574,
"tokens_seen": 1523646464
},
{
"epoch": 0.46,
"objective/train/docs_used": 864510,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5366392135620117,
"objective/train/theoretical_loss": 3.5112290938129176,
"objective/train/tokens_used": 1544172000,
"theoretical_loss": 3.5112290938129176,
"tokens_seen": 1523712000
},
{
"epoch": 0.46,
"learning_rate": 0.00027190571715145434,
"loss": 2.5353,
"theoretical_loss": 3.5112290938129176,
"tokens_seen": 1523712000
},
{
"epoch": 0.46,
"learning_rate": 0.0002718956870611836,
"loss": 2.8594,
"theoretical_loss": 3.5112158187981213,
"tokens_seen": 1523777536
},
{
"epoch": 0.46,
"learning_rate": 0.0002718856569709127,
"loss": 2.9529,
"theoretical_loss": 3.511202544514113,
"tokens_seen": 1523843072
},
{
"epoch": 0.46,
"learning_rate": 0.00027187562688064194,
"loss": 2.7592,
"theoretical_loss": 3.5111892709608203,
"tokens_seen": 1523908608
},
{
"epoch": 0.46,
"learning_rate": 0.00027186559679037107,
"loss": 2.7157,
"theoretical_loss": 3.5111759981381727,
"tokens_seen": 1523974144
},
{
"epoch": 0.46,
"learning_rate": 0.0002718555667001003,
"loss": 2.8916,
"theoretical_loss": 3.5111627260460985,
"tokens_seen": 1524039680
},
{
"epoch": 0.46,
"learning_rate": 0.0002718455366098295,
"loss": 2.5734,
"theoretical_loss": 3.511149454684525,
"tokens_seen": 1524105216
},
{
"epoch": 0.46,
"learning_rate": 0.00027183550651955867,
"loss": 2.5402,
"theoretical_loss": 3.511136184053382,
"tokens_seen": 1524170752
},
{
"epoch": 0.46,
"learning_rate": 0.00027182547642928785,
"loss": 2.6027,
"theoretical_loss": 3.5111229141525966,
"tokens_seen": 1524236288
},
{
"epoch": 0.46,
"learning_rate": 0.0002718154463390171,
"loss": 2.5986,
"theoretical_loss": 3.511109644982098,
"tokens_seen": 1524301824
},
{
"epoch": 0.46,
"learning_rate": 0.0002718054162487462,
"loss": 2.583,
"theoretical_loss": 3.511096376541815,
"tokens_seen": 1524367360
},
{
"epoch": 0.46,
"learning_rate": 0.00027179538615847544,
"loss": 2.7663,
"theoretical_loss": 3.5110831088316745,
"tokens_seen": 1524432896
},
{
"epoch": 0.46,
"learning_rate": 0.00027178535606820457,
"loss": 2.5402,
"theoretical_loss": 3.511069841851606,
"tokens_seen": 1524498432
},
{
"epoch": 0.46,
"learning_rate": 0.0002717753259779338,
"loss": 2.5404,
"theoretical_loss": 3.511056575601539,
"tokens_seen": 1524563968
},
{
"epoch": 0.46,
"learning_rate": 0.00027176529588766304,
"loss": 2.3173,
"theoretical_loss": 3.5110433100813996,
"tokens_seen": 1524629504
},
{
"epoch": 0.46,
"learning_rate": 0.00027175526579739217,
"loss": 2.7485,
"theoretical_loss": 3.5110300452911174,
"tokens_seen": 1524695040
},
{
"epoch": 0.46,
"learning_rate": 0.0002717452357071214,
"loss": 2.611,
"theoretical_loss": 3.511016781230621,
"tokens_seen": 1524760576
},
{
"epoch": 0.46,
"learning_rate": 0.00027173520561685053,
"loss": 2.654,
"theoretical_loss": 3.511003517899839,
"tokens_seen": 1524826112
},
{
"epoch": 0.46,
"learning_rate": 0.00027172517552657977,
"loss": 2.8729,
"theoretical_loss": 3.5109902552987,
"tokens_seen": 1524891648
},
{
"epoch": 0.46,
"learning_rate": 0.00027171514543630895,
"loss": 2.6274,
"theoretical_loss": 3.510976993427132,
"tokens_seen": 1524957184
},
{
"epoch": 0.46,
"learning_rate": 0.00027170511534603813,
"loss": 2.5094,
"theoretical_loss": 3.510963732285063,
"tokens_seen": 1525022720
},
{
"epoch": 0.46,
"learning_rate": 0.0002716950852557673,
"loss": 2.5265,
"theoretical_loss": 3.510950471872423,
"tokens_seen": 1525088256
},
{
"epoch": 0.46,
"learning_rate": 0.00027168505516549655,
"loss": 2.8013,
"theoretical_loss": 3.51093721218914,
"tokens_seen": 1525153792
},
{
"epoch": 0.46,
"learning_rate": 0.00027167502507522567,
"loss": 2.3634,
"theoretical_loss": 3.5109239532351415,
"tokens_seen": 1525219328
},
{
"epoch": 0.46,
"learning_rate": 0.0002716649949849549,
"loss": 2.5993,
"theoretical_loss": 3.5109106950103577,
"tokens_seen": 1525284864
},
{
"epoch": 0.46,
"objective/train/docs_used": 865194,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6970245838165283,
"objective/train/theoretical_loss": 3.5108974375147155,
"objective/train/tokens_used": 1545810400,
"theoretical_loss": 3.5108974375147155,
"tokens_seen": 1525350400
},
{
"epoch": 0.46,
"learning_rate": 0.00027165496489468403,
"loss": 2.6741,
"theoretical_loss": 3.5108974375147155,
"tokens_seen": 1525350400
},
{
"epoch": 0.46,
"learning_rate": 0.00027164493480441327,
"loss": 2.5339,
"theoretical_loss": 3.510884180748145,
"tokens_seen": 1525415936
},
{
"epoch": 0.46,
"learning_rate": 0.00027163490471414245,
"loss": 2.8664,
"theoretical_loss": 3.5108709247105736,
"tokens_seen": 1525481472
},
{
"epoch": 0.46,
"learning_rate": 0.00027162487462387163,
"loss": 2.7493,
"theoretical_loss": 3.510857669401931,
"tokens_seen": 1525547008
},
{
"epoch": 0.46,
"learning_rate": 0.0002716148445336008,
"loss": 2.6081,
"theoretical_loss": 3.5108444148221443,
"tokens_seen": 1525612544
},
{
"epoch": 0.46,
"learning_rate": 0.00027160481444333,
"loss": 2.6082,
"theoretical_loss": 3.5108311609711436,
"tokens_seen": 1525678080
},
{
"epoch": 0.46,
"learning_rate": 0.0002715947843530592,
"loss": 2.4469,
"theoretical_loss": 3.510817907848857,
"tokens_seen": 1525743616
},
{
"epoch": 0.46,
"learning_rate": 0.0002715847542627884,
"loss": 2.3276,
"theoretical_loss": 3.510804655455213,
"tokens_seen": 1525809152
},
{
"epoch": 0.46,
"learning_rate": 0.00027157472417251754,
"loss": 2.4452,
"theoretical_loss": 3.51079140379014,
"tokens_seen": 1525874688
},
{
"epoch": 0.46,
"learning_rate": 0.0002715646940822468,
"loss": 2.5347,
"theoretical_loss": 3.5107781528535673,
"tokens_seen": 1525940224
},
{
"epoch": 0.46,
"learning_rate": 0.0002715546639919759,
"loss": 2.7432,
"theoretical_loss": 3.5107649026454233,
"tokens_seen": 1526005760
},
{
"epoch": 0.46,
"learning_rate": 0.00027154463390170514,
"loss": 2.889,
"theoretical_loss": 3.510751653165636,
"tokens_seen": 1526071296
},
{
"epoch": 0.46,
"learning_rate": 0.0002715346038114343,
"loss": 2.5774,
"theoretical_loss": 3.510738404414136,
"tokens_seen": 1526136832
},
{
"epoch": 0.46,
"learning_rate": 0.0002715245737211635,
"loss": 2.6517,
"theoretical_loss": 3.5107251563908495,
"tokens_seen": 1526202368
},
{
"epoch": 0.46,
"learning_rate": 0.0002715145436308927,
"loss": 2.6862,
"theoretical_loss": 3.510711909095707,
"tokens_seen": 1526267904
},
{
"epoch": 0.46,
"learning_rate": 0.0002715045135406219,
"loss": 2.6491,
"theoretical_loss": 3.5106986625286365,
"tokens_seen": 1526333440
},
{
"epoch": 0.46,
"learning_rate": 0.00027149448345035104,
"loss": 2.5469,
"theoretical_loss": 3.510685416689567,
"tokens_seen": 1526398976
},
{
"epoch": 0.46,
"learning_rate": 0.0002714844533600803,
"loss": 2.3808,
"theoretical_loss": 3.5106721715784266,
"tokens_seen": 1526464512
},
{
"epoch": 0.46,
"learning_rate": 0.0002714744232698094,
"loss": 2.413,
"theoretical_loss": 3.510658927195145,
"tokens_seen": 1526530048
},
{
"epoch": 0.46,
"learning_rate": 0.00027146439317953864,
"loss": 2.6189,
"theoretical_loss": 3.5106456835396505,
"tokens_seen": 1526595584
},
{
"epoch": 0.46,
"learning_rate": 0.0002714543630892678,
"loss": 2.7087,
"theoretical_loss": 3.5106324406118716,
"tokens_seen": 1526661120
},
{
"epoch": 0.46,
"learning_rate": 0.000271444332998997,
"loss": 2.6135,
"theoretical_loss": 3.510619198411738,
"tokens_seen": 1526726656
},
{
"epoch": 0.46,
"learning_rate": 0.0002714343029087262,
"loss": 2.6273,
"theoretical_loss": 3.5106059569391768,
"tokens_seen": 1526792192
},
{
"epoch": 0.46,
"learning_rate": 0.00027142427281845536,
"loss": 2.5899,
"theoretical_loss": 3.510592716194118,
"tokens_seen": 1526857728
},
{
"epoch": 0.46,
"learning_rate": 0.00027141424272818454,
"loss": 2.3808,
"theoretical_loss": 3.510579476176491,
"tokens_seen": 1526923264
},
{
"epoch": 0.46,
"objective/train/docs_used": 866480,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3610594272613525,
"objective/train/theoretical_loss": 3.5105662368862234,
"objective/train/tokens_used": 1547448800,
"theoretical_loss": 3.5105662368862234,
"tokens_seen": 1526988800
},
{
"epoch": 0.46,
"learning_rate": 0.0002714042126379138,
"loss": 2.3806,
"theoretical_loss": 3.5105662368862234,
"tokens_seen": 1526988800
},
{
"epoch": 0.46,
"learning_rate": 0.0002713941825476429,
"loss": 2.5563,
"theoretical_loss": 3.510552998323244,
"tokens_seen": 1527054336
},
{
"epoch": 0.46,
"learning_rate": 0.00027138415245737214,
"loss": 2.6393,
"theoretical_loss": 3.5105397604874833,
"tokens_seen": 1527119872
},
{
"epoch": 0.46,
"learning_rate": 0.00027137412236710127,
"loss": 2.9569,
"theoretical_loss": 3.510526523378868,
"tokens_seen": 1527185408
},
{
"epoch": 0.46,
"learning_rate": 0.0002713640922768305,
"loss": 2.5562,
"theoretical_loss": 3.510513286997328,
"tokens_seen": 1527250944
},
{
"epoch": 0.46,
"learning_rate": 0.0002713540621865597,
"loss": 2.6357,
"theoretical_loss": 3.5105000513427926,
"tokens_seen": 1527316480
},
{
"epoch": 0.46,
"learning_rate": 0.00027134403209628887,
"loss": 2.4901,
"theoretical_loss": 3.5104868164151894,
"tokens_seen": 1527382016
},
{
"epoch": 0.46,
"learning_rate": 0.00027133400200601805,
"loss": 2.7927,
"theoretical_loss": 3.5104735822144484,
"tokens_seen": 1527447552
},
{
"epoch": 0.46,
"learning_rate": 0.0002713239719157473,
"loss": 2.5573,
"theoretical_loss": 3.510460348740498,
"tokens_seen": 1527513088
},
{
"epoch": 0.46,
"learning_rate": 0.0002713139418254764,
"loss": 2.5227,
"theoretical_loss": 3.5104471159932675,
"tokens_seen": 1527578624
},
{
"epoch": 0.46,
"learning_rate": 0.00027130391173520564,
"loss": 2.7479,
"theoretical_loss": 3.510433883972686,
"tokens_seen": 1527644160
},
{
"epoch": 0.46,
"learning_rate": 0.00027129388164493477,
"loss": 2.5529,
"theoretical_loss": 3.510420652678681,
"tokens_seen": 1527709696
},
{
"epoch": 0.46,
"learning_rate": 0.000271283851554664,
"loss": 2.4887,
"theoretical_loss": 3.510407422111183,
"tokens_seen": 1527775232
},
{
"epoch": 0.46,
"learning_rate": 0.0002712738214643932,
"loss": 2.7534,
"theoretical_loss": 3.5103941922701205,
"tokens_seen": 1527840768
},
{
"epoch": 0.46,
"learning_rate": 0.00027126379137412237,
"loss": 2.689,
"theoretical_loss": 3.5103809631554217,
"tokens_seen": 1527906304
},
{
"epoch": 0.46,
"learning_rate": 0.00027125376128385155,
"loss": 2.7496,
"theoretical_loss": 3.5103677347670166,
"tokens_seen": 1527971840
},
{
"epoch": 0.46,
"learning_rate": 0.00027124373119358073,
"loss": 2.7038,
"theoretical_loss": 3.510354507104834,
"tokens_seen": 1528037376
},
{
"epoch": 0.46,
"learning_rate": 0.0002712337011033099,
"loss": 2.6936,
"theoretical_loss": 3.5103412801688023,
"tokens_seen": 1528102912
},
{
"epoch": 0.46,
"learning_rate": 0.00027122367101303915,
"loss": 2.5142,
"theoretical_loss": 3.510328053958851,
"tokens_seen": 1528168448
},
{
"epoch": 0.46,
"learning_rate": 0.0002712136409227683,
"loss": 2.4874,
"theoretical_loss": 3.5103148284749093,
"tokens_seen": 1528233984
},
{
"epoch": 0.46,
"learning_rate": 0.0002712036108324975,
"loss": 2.5841,
"theoretical_loss": 3.5103016037169055,
"tokens_seen": 1528299520
},
{
"epoch": 0.46,
"learning_rate": 0.0002711935807422267,
"loss": 2.5183,
"theoretical_loss": 3.510288379684769,
"tokens_seen": 1528365056
},
{
"epoch": 0.46,
"learning_rate": 0.00027118355065195587,
"loss": 2.6532,
"theoretical_loss": 3.510275156378429,
"tokens_seen": 1528430592
},
{
"epoch": 0.46,
"learning_rate": 0.00027117352056168505,
"loss": 2.5997,
"theoretical_loss": 3.510261933797814,
"tokens_seen": 1528496128
},
{
"epoch": 0.46,
"learning_rate": 0.00027116349047141423,
"loss": 2.5936,
"theoretical_loss": 3.510248711942854,
"tokens_seen": 1528561664
},
{
"epoch": 0.46,
"objective/train/docs_used": 867112,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7275195121765137,
"objective/train/theoretical_loss": 3.5102354908134776,
"objective/train/tokens_used": 1549087200,
"theoretical_loss": 3.5102354908134776,
"tokens_seen": 1528627200
},
{
"epoch": 0.46,
"learning_rate": 0.0002711534603811434,
"loss": 2.6174,
"theoretical_loss": 3.5102354908134776,
"tokens_seen": 1528627200
},
{
"epoch": 0.46,
"learning_rate": 0.00027114343029087265,
"loss": 2.7103,
"theoretical_loss": 3.510222270409613,
"tokens_seen": 1528692736
},
{
"epoch": 0.46,
"learning_rate": 0.0002711334002006018,
"loss": 2.5268,
"theoretical_loss": 3.5102090507311905,
"tokens_seen": 1528758272
},
{
"epoch": 0.46,
"learning_rate": 0.000271123370110331,
"loss": 2.3953,
"theoretical_loss": 3.510195831778139,
"tokens_seen": 1528823808
},
{
"epoch": 0.46,
"learning_rate": 0.00027111334002006014,
"loss": 2.6818,
"theoretical_loss": 3.5101826135503873,
"tokens_seen": 1528889344
},
{
"epoch": 0.46,
"learning_rate": 0.0002711033099297894,
"loss": 2.5999,
"theoretical_loss": 3.5101693960478646,
"tokens_seen": 1528954880
},
{
"epoch": 0.46,
"learning_rate": 0.00027109327983951856,
"loss": 2.5154,
"theoretical_loss": 3.5101561792705,
"tokens_seen": 1529020416
},
{
"epoch": 0.46,
"learning_rate": 0.00027108324974924774,
"loss": 2.5211,
"theoretical_loss": 3.5101429632182226,
"tokens_seen": 1529085952
},
{
"epoch": 0.46,
"learning_rate": 0.0002710732196589769,
"loss": 2.5338,
"theoretical_loss": 3.5101297478909617,
"tokens_seen": 1529151488
},
{
"epoch": 0.46,
"learning_rate": 0.0002710631895687061,
"loss": 2.556,
"theoretical_loss": 3.510116533288646,
"tokens_seen": 1529217024
},
{
"epoch": 0.46,
"learning_rate": 0.0002710531594784353,
"loss": 2.6294,
"theoretical_loss": 3.5101033194112055,
"tokens_seen": 1529282560
},
{
"epoch": 0.46,
"learning_rate": 0.0002710431293881645,
"loss": 2.5404,
"theoretical_loss": 3.510090106258569,
"tokens_seen": 1529348096
},
{
"epoch": 0.46,
"learning_rate": 0.00027103309929789364,
"loss": 2.572,
"theoretical_loss": 3.5100768938306652,
"tokens_seen": 1529413632
},
{
"epoch": 0.46,
"learning_rate": 0.0002710230692076229,
"loss": 2.648,
"theoretical_loss": 3.5100636821274236,
"tokens_seen": 1529479168
},
{
"epoch": 0.46,
"learning_rate": 0.0002710130391173521,
"loss": 2.6791,
"theoretical_loss": 3.510050471148774,
"tokens_seen": 1529544704
},
{
"epoch": 0.46,
"learning_rate": 0.00027100300902708124,
"loss": 2.6543,
"theoretical_loss": 3.510037260894645,
"tokens_seen": 1529610240
},
{
"epoch": 0.46,
"learning_rate": 0.0002709929789368105,
"loss": 2.6352,
"theoretical_loss": 3.5100240513649656,
"tokens_seen": 1529675776
},
{
"epoch": 0.46,
"learning_rate": 0.0002709829488465396,
"loss": 2.539,
"theoretical_loss": 3.510010842559666,
"tokens_seen": 1529741312
},
{
"epoch": 0.46,
"learning_rate": 0.00027097291875626884,
"loss": 2.5768,
"theoretical_loss": 3.5099976344786743,
"tokens_seen": 1529806848
},
{
"epoch": 0.46,
"learning_rate": 0.000270962888665998,
"loss": 2.6157,
"theoretical_loss": 3.5099844271219203,
"tokens_seen": 1529872384
},
{
"epoch": 0.46,
"learning_rate": 0.0002709528585757272,
"loss": 2.4034,
"theoretical_loss": 3.5099712204893327,
"tokens_seen": 1529937920
},
{
"epoch": 0.46,
"learning_rate": 0.0002709428284854564,
"loss": 2.6553,
"theoretical_loss": 3.5099580145808424,
"tokens_seen": 1530003456
},
{
"epoch": 0.46,
"learning_rate": 0.00027093279839518556,
"loss": 2.7611,
"theoretical_loss": 3.509944809396377,
"tokens_seen": 1530068992
},
{
"epoch": 0.46,
"learning_rate": 0.00027092276830491474,
"loss": 2.668,
"theoretical_loss": 3.5099316049358666,
"tokens_seen": 1530134528
},
{
"epoch": 0.46,
"learning_rate": 0.000270912738214644,
"loss": 2.3986,
"theoretical_loss": 3.50991840119924,
"tokens_seen": 1530200064
},
{
"epoch": 0.46,
"objective/train/docs_used": 868323,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4359331130981445,
"objective/train/theoretical_loss": 3.5099051981864275,
"objective/train/tokens_used": 1550725600,
"theoretical_loss": 3.5099051981864275,
"tokens_seen": 1530265600
},
{
"epoch": 0.46,
"learning_rate": 0.0002709027081243731,
"loss": 2.699,
"theoretical_loss": 3.5099051981864275,
"tokens_seen": 1530265600
},
{
"epoch": 0.46,
"learning_rate": 0.00027089267803410234,
"loss": 2.4283,
"theoretical_loss": 3.509891995897357,
"tokens_seen": 1530331136
},
{
"epoch": 0.46,
"learning_rate": 0.00027088264794383147,
"loss": 2.5386,
"theoretical_loss": 3.509878794331959,
"tokens_seen": 1530396672
},
{
"epoch": 0.46,
"learning_rate": 0.0002708726178535607,
"loss": 2.6551,
"theoretical_loss": 3.5098655934901624,
"tokens_seen": 1530462208
},
{
"epoch": 0.46,
"learning_rate": 0.0002708625877632899,
"loss": 2.4708,
"theoretical_loss": 3.5098523933718964,
"tokens_seen": 1530527744
},
{
"epoch": 0.46,
"learning_rate": 0.00027085255767301907,
"loss": 2.6283,
"theoretical_loss": 3.5098391939770908,
"tokens_seen": 1530593280
},
{
"epoch": 0.46,
"learning_rate": 0.00027084252758274825,
"loss": 2.4623,
"theoretical_loss": 3.5098259953056745,
"tokens_seen": 1530658816
},
{
"epoch": 0.46,
"learning_rate": 0.0002708324974924775,
"loss": 2.5724,
"theoretical_loss": 3.5098127973575775,
"tokens_seen": 1530724352
},
{
"epoch": 0.46,
"learning_rate": 0.0002708224674022066,
"loss": 2.6483,
"theoretical_loss": 3.5097996001327285,
"tokens_seen": 1530789888
},
{
"epoch": 0.46,
"learning_rate": 0.00027081243731193584,
"loss": 2.7135,
"theoretical_loss": 3.509786403631057,
"tokens_seen": 1530855424
},
{
"epoch": 0.46,
"learning_rate": 0.00027080240722166497,
"loss": 2.1798,
"theoretical_loss": 3.509773207852493,
"tokens_seen": 1530920960
},
{
"epoch": 0.46,
"learning_rate": 0.0002707923771313942,
"loss": 2.7882,
"theoretical_loss": 3.5097600127969657,
"tokens_seen": 1530986496
},
{
"epoch": 0.46,
"learning_rate": 0.0002707823470411234,
"loss": 2.7092,
"theoretical_loss": 3.509746818464404,
"tokens_seen": 1531052032
},
{
"epoch": 0.46,
"learning_rate": 0.00027077231695085257,
"loss": 2.7324,
"theoretical_loss": 3.5097336248547375,
"tokens_seen": 1531117568
},
{
"epoch": 0.46,
"learning_rate": 0.00027076228686058175,
"loss": 2.5266,
"theoretical_loss": 3.5097204319678967,
"tokens_seen": 1531183104
},
{
"epoch": 0.46,
"learning_rate": 0.00027075225677031093,
"loss": 2.4453,
"theoretical_loss": 3.5097072398038094,
"tokens_seen": 1531248640
},
{
"epoch": 0.46,
"learning_rate": 0.0002707422266800401,
"loss": 2.6763,
"theoretical_loss": 3.509694048362406,
"tokens_seen": 1531314176
},
{
"epoch": 0.46,
"learning_rate": 0.00027073219658976935,
"loss": 2.7968,
"theoretical_loss": 3.509680857643617,
"tokens_seen": 1531379712
},
{
"epoch": 0.46,
"learning_rate": 0.0002707221664994985,
"loss": 2.4976,
"theoretical_loss": 3.5096676676473697,
"tokens_seen": 1531445248
},
{
"epoch": 0.46,
"learning_rate": 0.0002707121364092277,
"loss": 2.6461,
"theoretical_loss": 3.5096544783735952,
"tokens_seen": 1531510784
},
{
"epoch": 0.46,
"learning_rate": 0.0002707021063189569,
"loss": 2.6416,
"theoretical_loss": 3.509641289822222,
"tokens_seen": 1531576320
},
{
"epoch": 0.46,
"learning_rate": 0.00027069207622868607,
"loss": 2.6869,
"theoretical_loss": 3.5096281019931803,
"tokens_seen": 1531641856
},
{
"epoch": 0.46,
"learning_rate": 0.00027068204613841525,
"loss": 2.6674,
"theoretical_loss": 3.5096149148864,
"tokens_seen": 1531707392
},
{
"epoch": 0.46,
"learning_rate": 0.00027067201604814443,
"loss": 2.6467,
"theoretical_loss": 3.5096017285018095,
"tokens_seen": 1531772928
},
{
"epoch": 0.46,
"learning_rate": 0.0002706619859578736,
"loss": 2.6925,
"theoretical_loss": 3.5095885428393396,
"tokens_seen": 1531838464
},
{
"epoch": 0.46,
"objective/train/docs_used": 869121,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7407829761505127,
"objective/train/theoretical_loss": 3.5095753578989184,
"objective/train/tokens_used": 1552364000,
"theoretical_loss": 3.5095753578989184,
"tokens_seen": 1531904000
},
{
"epoch": 0.46,
"learning_rate": 0.00027065195586760285,
"loss": 2.5892,
"theoretical_loss": 3.5095753578989184,
"tokens_seen": 1531904000
},
{
"epoch": 0.46,
"learning_rate": 0.000270641925777332,
"loss": 2.7016,
"theoretical_loss": 3.509562173680477,
"tokens_seen": 1531969536
},
{
"epoch": 0.46,
"learning_rate": 0.0002706318956870612,
"loss": 2.5642,
"theoretical_loss": 3.5095489901839443,
"tokens_seen": 1532035072
},
{
"epoch": 0.46,
"learning_rate": 0.00027062186559679034,
"loss": 2.7733,
"theoretical_loss": 3.5095358074092493,
"tokens_seen": 1532100608
},
{
"epoch": 0.46,
"learning_rate": 0.0002706118355065196,
"loss": 2.9262,
"theoretical_loss": 3.5095226253563228,
"tokens_seen": 1532166144
},
{
"epoch": 0.46,
"learning_rate": 0.00027060180541624876,
"loss": 2.9327,
"theoretical_loss": 3.5095094440250936,
"tokens_seen": 1532231680
},
{
"epoch": 0.46,
"learning_rate": 0.00027059177532597794,
"loss": 2.6298,
"theoretical_loss": 3.5094962634154916,
"tokens_seen": 1532297216
},
{
"epoch": 0.46,
"learning_rate": 0.0002705817452357071,
"loss": 2.7335,
"theoretical_loss": 3.5094830835274458,
"tokens_seen": 1532362752
},
{
"epoch": 0.46,
"learning_rate": 0.0002705717151454363,
"loss": 2.4069,
"theoretical_loss": 3.509469904360887,
"tokens_seen": 1532428288
},
{
"epoch": 0.46,
"learning_rate": 0.0002705616850551655,
"loss": 2.8027,
"theoretical_loss": 3.509456725915744,
"tokens_seen": 1532493824
},
{
"epoch": 0.46,
"learning_rate": 0.0002705516549648947,
"loss": 2.781,
"theoretical_loss": 3.509443548191947,
"tokens_seen": 1532559360
},
{
"epoch": 0.46,
"learning_rate": 0.00027054162487462384,
"loss": 2.6362,
"theoretical_loss": 3.509430371189425,
"tokens_seen": 1532624896
},
{
"epoch": 0.46,
"learning_rate": 0.0002705315947843531,
"loss": 2.6142,
"theoretical_loss": 3.5094171949081083,
"tokens_seen": 1532690432
},
{
"epoch": 0.46,
"learning_rate": 0.00027052156469408226,
"loss": 2.4525,
"theoretical_loss": 3.509404019347926,
"tokens_seen": 1532755968
},
{
"epoch": 0.46,
"learning_rate": 0.00027051153460381144,
"loss": 2.5456,
"theoretical_loss": 3.5093908445088084,
"tokens_seen": 1532821504
},
{
"epoch": 0.46,
"learning_rate": 0.0002705015045135406,
"loss": 2.6618,
"theoretical_loss": 3.509377670390685,
"tokens_seen": 1532887040
},
{
"epoch": 0.46,
"learning_rate": 0.0002704914744232698,
"loss": 2.6254,
"theoretical_loss": 3.5093644969934856,
"tokens_seen": 1532952576
},
{
"epoch": 0.46,
"learning_rate": 0.000270481444332999,
"loss": 2.7338,
"theoretical_loss": 3.50935132431714,
"tokens_seen": 1533018112
},
{
"epoch": 0.46,
"learning_rate": 0.0002704714142427282,
"loss": 2.6663,
"theoretical_loss": 3.509338152361577,
"tokens_seen": 1533083648
},
{
"epoch": 0.46,
"learning_rate": 0.00027046138415245735,
"loss": 2.6552,
"theoretical_loss": 3.5093249811267277,
"tokens_seen": 1533149184
},
{
"epoch": 0.46,
"learning_rate": 0.0002704513540621866,
"loss": 2.6412,
"theoretical_loss": 3.5093118106125214,
"tokens_seen": 1533214720
},
{
"epoch": 0.46,
"learning_rate": 0.0002704413239719157,
"loss": 2.6595,
"theoretical_loss": 3.509298640818888,
"tokens_seen": 1533280256
},
{
"epoch": 0.46,
"learning_rate": 0.00027043129388164494,
"loss": 2.8005,
"theoretical_loss": 3.5092854717457564,
"tokens_seen": 1533345792
},
{
"epoch": 0.46,
"learning_rate": 0.0002704212637913741,
"loss": 2.7248,
"theoretical_loss": 3.5092723033930566,
"tokens_seen": 1533411328
},
{
"epoch": 0.46,
"learning_rate": 0.0002704112337011033,
"loss": 2.8162,
"theoretical_loss": 3.5092591357607197,
"tokens_seen": 1533476864
},
{
"epoch": 0.46,
"objective/train/docs_used": 869825,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.569054365158081,
"objective/train/theoretical_loss": 3.509245968848674,
"objective/train/tokens_used": 1554002400,
"theoretical_loss": 3.509245968848674,
"tokens_seen": 1533542400
},
{
"epoch": 0.46,
"learning_rate": 0.0002704012036108325,
"loss": 2.5022,
"theoretical_loss": 3.509245968848674,
"tokens_seen": 1533542400
},
{
"epoch": 0.46,
"learning_rate": 0.00027039117352056167,
"loss": 2.5432,
"theoretical_loss": 3.5092328026568507,
"tokens_seen": 1533607936
},
{
"epoch": 0.46,
"learning_rate": 0.00027038114343029085,
"loss": 2.5188,
"theoretical_loss": 3.5092196371851783,
"tokens_seen": 1533673472
},
{
"epoch": 0.46,
"learning_rate": 0.0002703711133400201,
"loss": 2.787,
"theoretical_loss": 3.509206472433587,
"tokens_seen": 1533739008
},
{
"epoch": 0.46,
"learning_rate": 0.0002703610832497492,
"loss": 2.5188,
"theoretical_loss": 3.5091933084020077,
"tokens_seen": 1533804544
},
{
"epoch": 0.46,
"learning_rate": 0.00027035105315947845,
"loss": 2.6884,
"theoretical_loss": 3.509180145090369,
"tokens_seen": 1533870080
},
{
"epoch": 0.46,
"learning_rate": 0.00027034102306920763,
"loss": 2.7183,
"theoretical_loss": 3.509166982498601,
"tokens_seen": 1533935616
},
{
"epoch": 0.46,
"learning_rate": 0.0002703309929789368,
"loss": 2.5028,
"theoretical_loss": 3.5091538206266337,
"tokens_seen": 1534001152
},
{
"epoch": 0.46,
"learning_rate": 0.000270320962888666,
"loss": 2.5323,
"theoretical_loss": 3.5091406594743972,
"tokens_seen": 1534066688
},
{
"epoch": 0.46,
"learning_rate": 0.00027031093279839517,
"loss": 2.6,
"theoretical_loss": 3.5091274990418215,
"tokens_seen": 1534132224
},
{
"epoch": 0.46,
"learning_rate": 0.00027030090270812435,
"loss": 2.6081,
"theoretical_loss": 3.5091143393288355,
"tokens_seen": 1534197760
},
{
"epoch": 0.46,
"learning_rate": 0.0002702908726178536,
"loss": 2.4306,
"theoretical_loss": 3.509101180335371,
"tokens_seen": 1534263296
},
{
"epoch": 0.46,
"learning_rate": 0.0002702808425275827,
"loss": 2.4238,
"theoretical_loss": 3.509088022061356,
"tokens_seen": 1534328832
},
{
"epoch": 0.46,
"learning_rate": 0.00027027081243731195,
"loss": 2.7012,
"theoretical_loss": 3.5090748645067213,
"tokens_seen": 1534394368
},
{
"epoch": 0.46,
"learning_rate": 0.00027026078234704113,
"loss": 2.5195,
"theoretical_loss": 3.509061707671397,
"tokens_seen": 1534459904
},
{
"epoch": 0.47,
"learning_rate": 0.0002702507522567703,
"loss": 2.7771,
"theoretical_loss": 3.5090485515553134,
"tokens_seen": 1534525440
},
{
"epoch": 0.47,
"learning_rate": 0.00027024072216649955,
"loss": 2.6224,
"theoretical_loss": 3.5090353961583993,
"tokens_seen": 1534590976
},
{
"epoch": 0.47,
"learning_rate": 0.0002702306920762287,
"loss": 2.6009,
"theoretical_loss": 3.509022241480585,
"tokens_seen": 1534656512
},
{
"epoch": 0.47,
"learning_rate": 0.0002702206619859579,
"loss": 2.7376,
"theoretical_loss": 3.509009087521801,
"tokens_seen": 1534722048
},
{
"epoch": 0.47,
"learning_rate": 0.0002702106318956871,
"loss": 2.4025,
"theoretical_loss": 3.5089959342819776,
"tokens_seen": 1534787584
},
{
"epoch": 0.47,
"learning_rate": 0.00027020060180541627,
"loss": 2.522,
"theoretical_loss": 3.508982781761044,
"tokens_seen": 1534853120
},
{
"epoch": 0.47,
"learning_rate": 0.00027019057171514545,
"loss": 2.4734,
"theoretical_loss": 3.50896962995893,
"tokens_seen": 1534918656
},
{
"epoch": 0.47,
"learning_rate": 0.00027018054162487463,
"loss": 2.5286,
"theoretical_loss": 3.5089564788755667,
"tokens_seen": 1534984192
},
{
"epoch": 0.47,
"learning_rate": 0.0002701705115346038,
"loss": 2.8928,
"theoretical_loss": 3.5089433285108838,
"tokens_seen": 1535049728
},
{
"epoch": 0.47,
"learning_rate": 0.00027016048144433305,
"loss": 2.6501,
"theoretical_loss": 3.5089301788648104,
"tokens_seen": 1535115264
},
{
"epoch": 0.47,
"objective/train/docs_used": 871111,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.875903844833374,
"objective/train/theoretical_loss": 3.5089170299372774,
"objective/train/tokens_used": 1555640800,
"theoretical_loss": 3.5089170299372774,
"tokens_seen": 1535180800
},
{
"epoch": 0.47,
"learning_rate": 0.0002701504513540622,
"loss": 2.8377,
"theoretical_loss": 3.5089170299372774,
"tokens_seen": 1535180800
},
{
"epoch": 0.47,
"learning_rate": 0.0002701404212637914,
"loss": 2.5943,
"theoretical_loss": 3.508903881728215,
"tokens_seen": 1535246336
},
{
"epoch": 0.47,
"learning_rate": 0.00027013039117352054,
"loss": 2.4603,
"theoretical_loss": 3.5088907342375526,
"tokens_seen": 1535311872
},
{
"epoch": 0.47,
"learning_rate": 0.0002701203610832498,
"loss": 2.8115,
"theoretical_loss": 3.5088775874652214,
"tokens_seen": 1535377408
},
{
"epoch": 0.47,
"learning_rate": 0.00027011033099297896,
"loss": 2.7334,
"theoretical_loss": 3.5088644414111503,
"tokens_seen": 1535442944
},
{
"epoch": 0.47,
"learning_rate": 0.00027010030090270814,
"loss": 2.6305,
"theoretical_loss": 3.50885129607527,
"tokens_seen": 1535508480
},
{
"epoch": 0.47,
"learning_rate": 0.0002700902708124373,
"loss": 2.546,
"theoretical_loss": 3.5088381514575104,
"tokens_seen": 1535574016
},
{
"epoch": 0.47,
"learning_rate": 0.0002700802407221665,
"loss": 2.5551,
"theoretical_loss": 3.5088250075578014,
"tokens_seen": 1535639552
},
{
"epoch": 0.47,
"learning_rate": 0.0002700702106318957,
"loss": 2.707,
"theoretical_loss": 3.508811864376074,
"tokens_seen": 1535705088
},
{
"epoch": 0.47,
"learning_rate": 0.0002700601805416249,
"loss": 2.7785,
"theoretical_loss": 3.5087987219122576,
"tokens_seen": 1535770624
},
{
"epoch": 0.47,
"learning_rate": 0.00027005015045135404,
"loss": 2.6691,
"theoretical_loss": 3.5087855801662826,
"tokens_seen": 1535836160
},
{
"epoch": 0.47,
"learning_rate": 0.0002700401203610833,
"loss": 2.5352,
"theoretical_loss": 3.5087724391380792,
"tokens_seen": 1535901696
},
{
"epoch": 0.47,
"learning_rate": 0.00027003009027081246,
"loss": 2.4964,
"theoretical_loss": 3.5087592988275773,
"tokens_seen": 1535967232
},
{
"epoch": 0.47,
"learning_rate": 0.00027002006018054164,
"loss": 2.4185,
"theoretical_loss": 3.5087461592347076,
"tokens_seen": 1536032768
},
{
"epoch": 0.47,
"learning_rate": 0.0002700100300902708,
"loss": 2.5986,
"theoretical_loss": 3.5087330203594,
"tokens_seen": 1536098304
},
{
"epoch": 0.47,
"learning_rate": 0.00027,
"loss": 2.5277,
"theoretical_loss": 3.508719882201584,
"tokens_seen": 1536163840
},
{
"epoch": 0.47,
"learning_rate": 0.0002699899699097292,
"loss": 2.6716,
"theoretical_loss": 3.5087067447611906,
"tokens_seen": 1536229376
},
{
"epoch": 0.47,
"learning_rate": 0.0002699799398194584,
"loss": 2.5744,
"theoretical_loss": 3.5086936080381506,
"tokens_seen": 1536294912
},
{
"epoch": 0.47,
"learning_rate": 0.00026996990972918755,
"loss": 2.5784,
"theoretical_loss": 3.5086804720323927,
"tokens_seen": 1536360448
},
{
"epoch": 0.47,
"learning_rate": 0.0002699598796389168,
"loss": 2.7211,
"theoretical_loss": 3.5086673367438483,
"tokens_seen": 1536425984
},
{
"epoch": 0.47,
"learning_rate": 0.0002699498495486459,
"loss": 2.6435,
"theoretical_loss": 3.5086542021724476,
"tokens_seen": 1536491520
},
{
"epoch": 0.47,
"learning_rate": 0.00026993981945837514,
"loss": 2.7082,
"theoretical_loss": 3.50864106831812,
"tokens_seen": 1536557056
},
{
"epoch": 0.47,
"learning_rate": 0.0002699297893681043,
"loss": 2.6394,
"theoretical_loss": 3.508627935180797,
"tokens_seen": 1536622592
},
{
"epoch": 0.47,
"learning_rate": 0.0002699197592778335,
"loss": 2.6257,
"theoretical_loss": 3.508614802760408,
"tokens_seen": 1536688128
},
{
"epoch": 0.47,
"learning_rate": 0.0002699097291875627,
"loss": 2.4059,
"theoretical_loss": 3.5086016710568835,
"tokens_seen": 1536753664
},
{
"epoch": 0.47,
"objective/train/docs_used": 871860,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7336935997009277,
"objective/train/theoretical_loss": 3.5085885400701535,
"objective/train/tokens_used": 1557279200,
"theoretical_loss": 3.5085885400701535,
"tokens_seen": 1536819200
},
{
"epoch": 0.47,
"learning_rate": 0.00026989969909729187,
"loss": 2.6045,
"theoretical_loss": 3.5085885400701535,
"tokens_seen": 1536819200
},
{
"epoch": 0.47,
"learning_rate": 0.00026988966900702105,
"loss": 2.3653,
"theoretical_loss": 3.508575409800149,
"tokens_seen": 1536884736
},
{
"epoch": 0.47,
"learning_rate": 0.0002698796389167503,
"loss": 2.667,
"theoretical_loss": 3.5085622802467995,
"tokens_seen": 1536950272
},
{
"epoch": 0.47,
"learning_rate": 0.0002698696088264794,
"loss": 2.8603,
"theoretical_loss": 3.508549151410037,
"tokens_seen": 1537015808
},
{
"epoch": 0.47,
"learning_rate": 0.00026985957873620865,
"loss": 2.6293,
"theoretical_loss": 3.508536023289789,
"tokens_seen": 1537081344
},
{
"epoch": 0.47,
"learning_rate": 0.00026984954864593783,
"loss": 2.5396,
"theoretical_loss": 3.5085228958859886,
"tokens_seen": 1537146880
},
{
"epoch": 0.47,
"learning_rate": 0.000269839518555667,
"loss": 2.7751,
"theoretical_loss": 3.5085097691985645,
"tokens_seen": 1537212416
},
{
"epoch": 0.47,
"learning_rate": 0.0002698294884653962,
"loss": 2.423,
"theoretical_loss": 3.508496643227448,
"tokens_seen": 1537277952
},
{
"epoch": 0.47,
"learning_rate": 0.00026981945837512537,
"loss": 2.6394,
"theoretical_loss": 3.508483517972569,
"tokens_seen": 1537343488
},
{
"epoch": 0.47,
"learning_rate": 0.00026980942828485455,
"loss": 2.6438,
"theoretical_loss": 3.5084703934338575,
"tokens_seen": 1537409024
},
{
"epoch": 0.47,
"learning_rate": 0.0002697993981945838,
"loss": 2.598,
"theoretical_loss": 3.5084572696112444,
"tokens_seen": 1537474560
},
{
"epoch": 0.47,
"learning_rate": 0.0002697893681043129,
"loss": 2.7805,
"theoretical_loss": 3.5084441465046607,
"tokens_seen": 1537540096
},
{
"epoch": 0.47,
"learning_rate": 0.00026977933801404215,
"loss": 2.7448,
"theoretical_loss": 3.508431024114035,
"tokens_seen": 1537605632
},
{
"epoch": 0.47,
"learning_rate": 0.0002697693079237713,
"loss": 2.6308,
"theoretical_loss": 3.5084179024393,
"tokens_seen": 1537671168
},
{
"epoch": 0.47,
"learning_rate": 0.0002697592778335005,
"loss": 2.6265,
"theoretical_loss": 3.508404781480385,
"tokens_seen": 1537736704
},
{
"epoch": 0.47,
"learning_rate": 0.0002697492477432297,
"loss": 2.7091,
"theoretical_loss": 3.50839166123722,
"tokens_seen": 1537802240
},
{
"epoch": 0.47,
"learning_rate": 0.0002697392176529589,
"loss": 2.6063,
"theoretical_loss": 3.508378541709736,
"tokens_seen": 1537867776
},
{
"epoch": 0.47,
"learning_rate": 0.00026972918756268806,
"loss": 2.6338,
"theoretical_loss": 3.508365422897864,
"tokens_seen": 1537933312
},
{
"epoch": 0.47,
"learning_rate": 0.0002697191574724173,
"loss": 2.7551,
"theoretical_loss": 3.508352304801533,
"tokens_seen": 1537998848
},
{
"epoch": 0.47,
"learning_rate": 0.0002697091273821464,
"loss": 2.6277,
"theoretical_loss": 3.508339187420675,
"tokens_seen": 1538064384
},
{
"epoch": 0.47,
"learning_rate": 0.00026969909729187565,
"loss": 2.5121,
"theoretical_loss": 3.5083260707552193,
"tokens_seen": 1538129920
},
{
"epoch": 0.47,
"learning_rate": 0.0002696890672016048,
"loss": 2.6352,
"theoretical_loss": 3.5083129548050973,
"tokens_seen": 1538195456
},
{
"epoch": 0.47,
"learning_rate": 0.000269679037111334,
"loss": 2.5408,
"theoretical_loss": 3.508299839570239,
"tokens_seen": 1538260992
},
{
"epoch": 0.47,
"learning_rate": 0.0002696690070210632,
"loss": 2.7734,
"theoretical_loss": 3.508286725050575,
"tokens_seen": 1538326528
},
{
"epoch": 0.47,
"learning_rate": 0.0002696589769307924,
"loss": 2.6441,
"theoretical_loss": 3.5082736112460364,
"tokens_seen": 1538392064
},
{
"epoch": 0.47,
"objective/train/docs_used": 873345,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.823885202407837,
"objective/train/theoretical_loss": 3.5082604981565533,
"objective/train/tokens_used": 1558917600,
"theoretical_loss": 3.5082604981565533,
"tokens_seen": 1538457600
},
{
"epoch": 0.47,
"learning_rate": 0.00026964894684052156,
"loss": 2.6733,
"theoretical_loss": 3.5082604981565533,
"tokens_seen": 1538457600
},
{
"epoch": 0.47,
"learning_rate": 0.00026963891675025074,
"loss": 2.8829,
"theoretical_loss": 3.508247385782056,
"tokens_seen": 1538523136
},
{
"epoch": 0.47,
"learning_rate": 0.0002696288866599799,
"loss": 2.8009,
"theoretical_loss": 3.508234274122475,
"tokens_seen": 1538588672
},
{
"epoch": 0.47,
"learning_rate": 0.00026961885656970916,
"loss": 2.558,
"theoretical_loss": 3.5082211631777414,
"tokens_seen": 1538654208
},
{
"epoch": 0.47,
"learning_rate": 0.0002696088264794383,
"loss": 2.5372,
"theoretical_loss": 3.5082080529477855,
"tokens_seen": 1538719744
},
{
"epoch": 0.47,
"learning_rate": 0.0002695987963891675,
"loss": 2.8936,
"theoretical_loss": 3.5081949434325383,
"tokens_seen": 1538785280
},
{
"epoch": 0.47,
"learning_rate": 0.00026958876629889665,
"loss": 2.6393,
"theoretical_loss": 3.5081818346319293,
"tokens_seen": 1538850816
},
{
"epoch": 0.47,
"learning_rate": 0.0002695787362086259,
"loss": 2.6773,
"theoretical_loss": 3.5081687265458905,
"tokens_seen": 1538916352
},
{
"epoch": 0.47,
"learning_rate": 0.00026956870611835506,
"loss": 2.8205,
"theoretical_loss": 3.5081556191743517,
"tokens_seen": 1538981888
},
{
"epoch": 0.47,
"learning_rate": 0.00026955867602808424,
"loss": 2.6504,
"theoretical_loss": 3.5081425125172436,
"tokens_seen": 1539047424
},
{
"epoch": 0.47,
"learning_rate": 0.0002695486459378134,
"loss": 2.5939,
"theoretical_loss": 3.508129406574497,
"tokens_seen": 1539112960
},
{
"epoch": 0.47,
"learning_rate": 0.00026953861584754266,
"loss": 2.7623,
"theoretical_loss": 3.5081163013460426,
"tokens_seen": 1539178496
},
{
"epoch": 0.47,
"learning_rate": 0.0002695285857572718,
"loss": 2.7835,
"theoretical_loss": 3.5081031968318106,
"tokens_seen": 1539244032
},
{
"epoch": 0.47,
"learning_rate": 0.000269518555667001,
"loss": 2.8182,
"theoretical_loss": 3.5080900930317327,
"tokens_seen": 1539309568
},
{
"epoch": 0.47,
"learning_rate": 0.0002695085255767302,
"loss": 2.4027,
"theoretical_loss": 3.508076989945738,
"tokens_seen": 1539375104
},
{
"epoch": 0.47,
"learning_rate": 0.0002694984954864594,
"loss": 2.525,
"theoretical_loss": 3.5080638875737584,
"tokens_seen": 1539440640
},
{
"epoch": 0.47,
"learning_rate": 0.0002694884653961886,
"loss": 2.4235,
"theoretical_loss": 3.5080507859157244,
"tokens_seen": 1539506176
},
{
"epoch": 0.47,
"learning_rate": 0.00026947843530591775,
"loss": 2.7236,
"theoretical_loss": 3.5080376849715664,
"tokens_seen": 1539571712
},
{
"epoch": 0.47,
"learning_rate": 0.000269468405215647,
"loss": 2.4357,
"theoretical_loss": 3.5080245847412153,
"tokens_seen": 1539637248
},
{
"epoch": 0.47,
"learning_rate": 0.0002694583751253761,
"loss": 2.537,
"theoretical_loss": 3.5080114852246025,
"tokens_seen": 1539702784
},
{
"epoch": 0.47,
"learning_rate": 0.00026944834503510534,
"loss": 2.5395,
"theoretical_loss": 3.507998386421657,
"tokens_seen": 1539768320
},
{
"epoch": 0.47,
"learning_rate": 0.0002694383149448345,
"loss": 2.6954,
"theoretical_loss": 3.5079852883323115,
"tokens_seen": 1539833856
},
{
"epoch": 0.47,
"learning_rate": 0.0002694282848545637,
"loss": 2.5131,
"theoretical_loss": 3.5079721909564956,
"tokens_seen": 1539899392
},
{
"epoch": 0.47,
"learning_rate": 0.0002694182547642929,
"loss": 2.6487,
"theoretical_loss": 3.50795909429414,
"tokens_seen": 1539964928
},
{
"epoch": 0.47,
"learning_rate": 0.00026940822467402207,
"loss": 2.5405,
"theoretical_loss": 3.5079459983451757,
"tokens_seen": 1540030464
},
{
"epoch": 0.47,
"objective/train/docs_used": 874032,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.321516752243042,
"objective/train/theoretical_loss": 3.507932903109534,
"objective/train/tokens_used": 1560556000,
"theoretical_loss": 3.507932903109534,
"tokens_seen": 1540096000
},
{
"epoch": 0.47,
"learning_rate": 0.00026939819458375125,
"loss": 2.7713,
"theoretical_loss": 3.507932903109534,
"tokens_seen": 1540096000
},
{
"epoch": 0.47,
"learning_rate": 0.0002693881644934805,
"loss": 2.6263,
"theoretical_loss": 3.5079198085871455,
"tokens_seen": 1540161536
},
{
"epoch": 0.47,
"learning_rate": 0.0002693781344032096,
"loss": 2.5708,
"theoretical_loss": 3.50790671477794,
"tokens_seen": 1540227072
},
{
"epoch": 0.47,
"learning_rate": 0.00026936810431293885,
"loss": 2.7338,
"theoretical_loss": 3.50789362168185,
"tokens_seen": 1540292608
},
{
"epoch": 0.47,
"learning_rate": 0.00026935807422266803,
"loss": 2.6086,
"theoretical_loss": 3.507880529298805,
"tokens_seen": 1540358144
},
{
"epoch": 0.47,
"learning_rate": 0.0002693480441323972,
"loss": 2.5859,
"theoretical_loss": 3.507867437628736,
"tokens_seen": 1540423680
},
{
"epoch": 0.47,
"learning_rate": 0.0002693380140421264,
"loss": 2.547,
"theoretical_loss": 3.507854346671575,
"tokens_seen": 1540489216
},
{
"epoch": 0.47,
"learning_rate": 0.00026932798395185557,
"loss": 2.3689,
"theoretical_loss": 3.507841256427251,
"tokens_seen": 1540554752
},
{
"epoch": 0.47,
"learning_rate": 0.00026931795386158475,
"loss": 2.5998,
"theoretical_loss": 3.5078281668956963,
"tokens_seen": 1540620288
},
{
"epoch": 0.47,
"learning_rate": 0.000269307923771314,
"loss": 2.6101,
"theoretical_loss": 3.507815078076841,
"tokens_seen": 1540685824
},
{
"epoch": 0.47,
"learning_rate": 0.0002692978936810431,
"loss": 2.6327,
"theoretical_loss": 3.5078019899706163,
"tokens_seen": 1540751360
},
{
"epoch": 0.47,
"learning_rate": 0.00026928786359077235,
"loss": 2.4734,
"theoretical_loss": 3.5077889025769533,
"tokens_seen": 1540816896
},
{
"epoch": 0.47,
"learning_rate": 0.0002692778335005015,
"loss": 2.6004,
"theoretical_loss": 3.5077758158957826,
"tokens_seen": 1540882432
},
{
"epoch": 0.47,
"learning_rate": 0.0002692678034102307,
"loss": 2.4377,
"theoretical_loss": 3.507762729927035,
"tokens_seen": 1540947968
},
{
"epoch": 0.47,
"learning_rate": 0.0002692577733199599,
"loss": 2.4125,
"theoretical_loss": 3.5077496446706418,
"tokens_seen": 1541013504
},
{
"epoch": 0.47,
"learning_rate": 0.0002692477432296891,
"loss": 2.6035,
"theoretical_loss": 3.5077365601265336,
"tokens_seen": 1541079040
},
{
"epoch": 0.47,
"learning_rate": 0.00026923771313941826,
"loss": 2.6364,
"theoretical_loss": 3.5077234762946414,
"tokens_seen": 1541144576
},
{
"epoch": 0.47,
"learning_rate": 0.0002692276830491475,
"loss": 2.5367,
"theoretical_loss": 3.5077103931748965,
"tokens_seen": 1541210112
},
{
"epoch": 0.47,
"learning_rate": 0.0002692176529588766,
"loss": 2.6759,
"theoretical_loss": 3.507697310767229,
"tokens_seen": 1541275648
},
{
"epoch": 0.47,
"learning_rate": 0.00026920762286860585,
"loss": 2.5766,
"theoretical_loss": 3.507684229071571,
"tokens_seen": 1541341184
},
{
"epoch": 0.47,
"learning_rate": 0.000269197592778335,
"loss": 2.549,
"theoretical_loss": 3.5076711480878524,
"tokens_seen": 1541406720
},
{
"epoch": 0.47,
"learning_rate": 0.0002691875626880642,
"loss": 2.8196,
"theoretical_loss": 3.5076580678160054,
"tokens_seen": 1541472256
},
{
"epoch": 0.47,
"learning_rate": 0.0002691775325977934,
"loss": 2.6958,
"theoretical_loss": 3.5076449882559597,
"tokens_seen": 1541537792
},
{
"epoch": 0.47,
"learning_rate": 0.0002691675025075226,
"loss": 2.752,
"theoretical_loss": 3.507631909407647,
"tokens_seen": 1541603328
},
{
"epoch": 0.47,
"learning_rate": 0.00026915747241725176,
"loss": 2.7693,
"theoretical_loss": 3.5076188312709986,
"tokens_seen": 1541668864
},
{
"epoch": 0.47,
"objective/train/docs_used": 874753,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.710665702819824,
"objective/train/theoretical_loss": 3.5076057538459446,
"objective/train/tokens_used": 1562194400,
"theoretical_loss": 3.5076057538459446,
"tokens_seen": 1541734400
},
{
"epoch": 0.47,
"learning_rate": 0.00026914744232698094,
"loss": 2.6451,
"theoretical_loss": 3.5076057538459446,
"tokens_seen": 1541734400
},
{
"epoch": 0.47,
"learning_rate": 0.0002691374122367101,
"loss": 2.7095,
"theoretical_loss": 3.5075926771324166,
"tokens_seen": 1541799936
},
{
"epoch": 0.47,
"learning_rate": 0.00026912738214643936,
"loss": 2.3524,
"theoretical_loss": 3.507579601130346,
"tokens_seen": 1541865472
},
{
"epoch": 0.47,
"learning_rate": 0.0002691173520561685,
"loss": 2.5905,
"theoretical_loss": 3.5075665258396636,
"tokens_seen": 1541931008
},
{
"epoch": 0.47,
"learning_rate": 0.0002691073219658977,
"loss": 2.5009,
"theoretical_loss": 3.5075534512603,
"tokens_seen": 1541996544
},
{
"epoch": 0.47,
"learning_rate": 0.00026909729187562685,
"loss": 2.65,
"theoretical_loss": 3.5075403773921865,
"tokens_seen": 1542062080
},
{
"epoch": 0.47,
"learning_rate": 0.0002690872617853561,
"loss": 2.5237,
"theoretical_loss": 3.5075273042352544,
"tokens_seen": 1542127616
},
{
"epoch": 0.47,
"learning_rate": 0.00026907723169508526,
"loss": 2.8182,
"theoretical_loss": 3.507514231789435,
"tokens_seen": 1542193152
},
{
"epoch": 0.47,
"learning_rate": 0.00026906720160481444,
"loss": 2.4154,
"theoretical_loss": 3.507501160054659,
"tokens_seen": 1542258688
},
{
"epoch": 0.47,
"learning_rate": 0.0002690571715145436,
"loss": 2.655,
"theoretical_loss": 3.5074880890308577,
"tokens_seen": 1542324224
},
{
"epoch": 0.47,
"learning_rate": 0.00026904714142427286,
"loss": 2.7248,
"theoretical_loss": 3.5074750187179617,
"tokens_seen": 1542389760
},
{
"epoch": 0.47,
"learning_rate": 0.000269037111334002,
"loss": 2.6448,
"theoretical_loss": 3.507461949115903,
"tokens_seen": 1542455296
},
{
"epoch": 0.47,
"learning_rate": 0.0002690270812437312,
"loss": 2.7606,
"theoretical_loss": 3.5074488802246115,
"tokens_seen": 1542520832
},
{
"epoch": 0.47,
"learning_rate": 0.00026901705115346035,
"loss": 2.5171,
"theoretical_loss": 3.5074358120440206,
"tokens_seen": 1542586368
},
{
"epoch": 0.47,
"learning_rate": 0.0002690070210631896,
"loss": 2.5509,
"theoretical_loss": 3.5074227445740584,
"tokens_seen": 1542651904
},
{
"epoch": 0.47,
"learning_rate": 0.00026899699097291877,
"loss": 2.6815,
"theoretical_loss": 3.5074096778146586,
"tokens_seen": 1542717440
},
{
"epoch": 0.47,
"learning_rate": 0.00026898696088264795,
"loss": 2.7656,
"theoretical_loss": 3.507396611765752,
"tokens_seen": 1542782976
},
{
"epoch": 0.47,
"learning_rate": 0.00026897693079237713,
"loss": 2.5843,
"theoretical_loss": 3.5073835464272682,
"tokens_seen": 1542848512
},
{
"epoch": 0.47,
"learning_rate": 0.0002689669007021063,
"loss": 2.657,
"theoretical_loss": 3.50737048179914,
"tokens_seen": 1542914048
},
{
"epoch": 0.47,
"learning_rate": 0.0002689568706118355,
"loss": 2.6702,
"theoretical_loss": 3.5073574178812974,
"tokens_seen": 1542979584
},
{
"epoch": 0.47,
"learning_rate": 0.0002689468405215647,
"loss": 2.707,
"theoretical_loss": 3.5073443546736733,
"tokens_seen": 1543045120
},
{
"epoch": 0.47,
"learning_rate": 0.00026893681043129385,
"loss": 2.5154,
"theoretical_loss": 3.507331292176197,
"tokens_seen": 1543110656
},
{
"epoch": 0.47,
"learning_rate": 0.0002689267803410231,
"loss": 2.4741,
"theoretical_loss": 3.507318230388801,
"tokens_seen": 1543176192
},
{
"epoch": 0.47,
"learning_rate": 0.0002689267803410231,
"loss": 2.3875,
"theoretical_loss": 3.507305169311416,
"tokens_seen": 1543241728
},
{
"epoch": 0.47,
"learning_rate": 0.0002689167502507522,
"loss": 2.8725,
"theoretical_loss": 3.507292108943974,
"tokens_seen": 1543307264
},
{
"epoch": 0.47,
"objective/train/docs_used": 875990,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9834370613098145,
"objective/train/theoretical_loss": 3.5072790492864057,
"objective/train/tokens_used": 1563832800,
"theoretical_loss": 3.5072790492864057,
"tokens_seen": 1543372800
},
{
"epoch": 0.47,
"learning_rate": 0.00026890672016048145,
"loss": 2.8922,
"theoretical_loss": 3.5072790492864057,
"tokens_seen": 1543372800
},
{
"epoch": 0.47,
"learning_rate": 0.00026889669007021063,
"loss": 2.8852,
"theoretical_loss": 3.5072659903386416,
"tokens_seen": 1543438336
},
{
"epoch": 0.47,
"learning_rate": 0.0002688866599799398,
"loss": 2.8282,
"theoretical_loss": 3.5072529321006147,
"tokens_seen": 1543503872
},
{
"epoch": 0.47,
"learning_rate": 0.000268876629889669,
"loss": 2.7014,
"theoretical_loss": 3.507239874572255,
"tokens_seen": 1543569408
},
{
"epoch": 0.47,
"learning_rate": 0.00026886659979939823,
"loss": 2.7442,
"theoretical_loss": 3.5072268177534935,
"tokens_seen": 1543634944
},
{
"epoch": 0.47,
"learning_rate": 0.00026885656970912736,
"loss": 2.6029,
"theoretical_loss": 3.5072137616442625,
"tokens_seen": 1543700480
},
{
"epoch": 0.47,
"learning_rate": 0.0002688465396188566,
"loss": 2.6481,
"theoretical_loss": 3.5072007062444936,
"tokens_seen": 1543766016
},
{
"epoch": 0.47,
"learning_rate": 0.0002688365095285857,
"loss": 2.8556,
"theoretical_loss": 3.5071876515541174,
"tokens_seen": 1543831552
},
{
"epoch": 0.47,
"learning_rate": 0.00026882647943831495,
"loss": 2.6467,
"theoretical_loss": 3.507174597573065,
"tokens_seen": 1543897088
},
{
"epoch": 0.47,
"learning_rate": 0.00026881644934804413,
"loss": 2.7457,
"theoretical_loss": 3.507161544301268,
"tokens_seen": 1543962624
},
{
"epoch": 0.47,
"learning_rate": 0.0002688064192577733,
"loss": 2.7254,
"theoretical_loss": 3.5071484917386586,
"tokens_seen": 1544028160
},
{
"epoch": 0.47,
"learning_rate": 0.0002687963891675025,
"loss": 2.6144,
"theoretical_loss": 3.5071354398851673,
"tokens_seen": 1544093696
},
{
"epoch": 0.47,
"learning_rate": 0.0002687863590772317,
"loss": 2.4903,
"theoretical_loss": 3.507122388740725,
"tokens_seen": 1544159232
},
{
"epoch": 0.47,
"learning_rate": 0.00026877632898696086,
"loss": 2.8003,
"theoretical_loss": 3.5071093383052645,
"tokens_seen": 1544224768
},
{
"epoch": 0.47,
"learning_rate": 0.0002687662988966901,
"loss": 2.5659,
"theoretical_loss": 3.5070962885787154,
"tokens_seen": 1544290304
},
{
"epoch": 0.47,
"learning_rate": 0.0002687562688064193,
"loss": 2.6771,
"theoretical_loss": 3.507083239561011,
"tokens_seen": 1544355840
},
{
"epoch": 0.47,
"learning_rate": 0.00026874623871614846,
"loss": 2.5019,
"theoretical_loss": 3.507070191252082,
"tokens_seen": 1544421376
},
{
"epoch": 0.47,
"learning_rate": 0.0002687362086258777,
"loss": 2.6153,
"theoretical_loss": 3.507057143651859,
"tokens_seen": 1544486912
},
{
"epoch": 0.47,
"learning_rate": 0.0002687261785356068,
"loss": 2.6927,
"theoretical_loss": 3.507044096760275,
"tokens_seen": 1544552448
},
{
"epoch": 0.47,
"learning_rate": 0.00026871614844533605,
"loss": 2.6341,
"theoretical_loss": 3.5070310505772597,
"tokens_seen": 1544617984
},
{
"epoch": 0.47,
"learning_rate": 0.0002687061183550652,
"loss": 2.785,
"theoretical_loss": 3.5070180051027458,
"tokens_seen": 1544683520
},
{
"epoch": 0.47,
"learning_rate": 0.0002686960882647944,
"loss": 2.4525,
"theoretical_loss": 3.5070049603366646,
"tokens_seen": 1544749056
},
{
"epoch": 0.47,
"learning_rate": 0.0002686860581745236,
"loss": 2.4881,
"theoretical_loss": 3.506991916278947,
"tokens_seen": 1544814592
},
{
"epoch": 0.47,
"learning_rate": 0.0002686760280842528,
"loss": 2.6636,
"theoretical_loss": 3.5069788729295253,
"tokens_seen": 1544880128
},
{
"epoch": 0.47,
"learning_rate": 0.00026866599799398196,
"loss": 2.6711,
"theoretical_loss": 3.5069658302883298,
"tokens_seen": 1544945664
},
{
"epoch": 0.47,
"objective/train/docs_used": 876526,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4204676151275635,
"objective/train/theoretical_loss": 3.5069527883552936,
"objective/train/tokens_used": 1565471200,
"theoretical_loss": 3.5069527883552936,
"tokens_seen": 1545011200
},
{
"epoch": 0.47,
"learning_rate": 0.00026865596790371114,
"loss": 2.6305,
"theoretical_loss": 3.5069527883552936,
"tokens_seen": 1545011200
},
{
"epoch": 0.47,
"learning_rate": 0.0002686459378134403,
"loss": 2.3358,
"theoretical_loss": 3.506939747130347,
"tokens_seen": 1545076736
},
{
"epoch": 0.47,
"learning_rate": 0.00026863590772316956,
"loss": 2.6685,
"theoretical_loss": 3.506926706613422,
"tokens_seen": 1545142272
},
{
"epoch": 0.47,
"learning_rate": 0.0002686258776328987,
"loss": 2.7144,
"theoretical_loss": 3.50691366680445,
"tokens_seen": 1545207808
},
{
"epoch": 0.47,
"learning_rate": 0.0002686158475426279,
"loss": 2.6337,
"theoretical_loss": 3.506900627703362,
"tokens_seen": 1545273344
},
{
"epoch": 0.47,
"learning_rate": 0.00026860581745235705,
"loss": 2.6793,
"theoretical_loss": 3.5068875893100913,
"tokens_seen": 1545338880
},
{
"epoch": 0.47,
"learning_rate": 0.0002685957873620863,
"loss": 2.6368,
"theoretical_loss": 3.506874551624567,
"tokens_seen": 1545404416
},
{
"epoch": 0.47,
"learning_rate": 0.00026858575727181546,
"loss": 2.593,
"theoretical_loss": 3.506861514646723,
"tokens_seen": 1545469952
},
{
"epoch": 0.47,
"learning_rate": 0.00026857572718154464,
"loss": 2.685,
"theoretical_loss": 3.5068484783764893,
"tokens_seen": 1545535488
},
{
"epoch": 0.47,
"learning_rate": 0.0002685656970912738,
"loss": 2.8052,
"theoretical_loss": 3.5068354428137978,
"tokens_seen": 1545601024
},
{
"epoch": 0.47,
"learning_rate": 0.00026855566700100306,
"loss": 2.5996,
"theoretical_loss": 3.5068224079585812,
"tokens_seen": 1545666560
},
{
"epoch": 0.47,
"learning_rate": 0.0002685456369107322,
"loss": 2.4635,
"theoretical_loss": 3.5068093738107695,
"tokens_seen": 1545732096
},
{
"epoch": 0.47,
"learning_rate": 0.0002685456369107322,
"loss": 2.6079,
"theoretical_loss": 3.506796340370295,
"tokens_seen": 1545797632
},
{
"epoch": 0.47,
"learning_rate": 0.0002685356068204614,
"loss": 2.6964,
"theoretical_loss": 3.5067833076370896,
"tokens_seen": 1545863168
},
{
"epoch": 0.47,
"learning_rate": 0.00026852557673019055,
"loss": 2.6646,
"theoretical_loss": 3.506770275611085,
"tokens_seen": 1545928704
},
{
"epoch": 0.47,
"learning_rate": 0.0002685155466399198,
"loss": 2.574,
"theoretical_loss": 3.5067572442922126,
"tokens_seen": 1545994240
},
{
"epoch": 0.47,
"learning_rate": 0.00026850551654964897,
"loss": 2.672,
"theoretical_loss": 3.5067442136804035,
"tokens_seen": 1546059776
},
{
"epoch": 0.47,
"learning_rate": 0.00026849548645937815,
"loss": 2.7595,
"theoretical_loss": 3.50673118377559,
"tokens_seen": 1546125312
},
{
"epoch": 0.47,
"learning_rate": 0.00026848545636910733,
"loss": 2.3981,
"theoretical_loss": 3.5067181545777037,
"tokens_seen": 1546190848
},
{
"epoch": 0.47,
"learning_rate": 0.0002684754262788365,
"loss": 2.597,
"theoretical_loss": 3.5067051260866764,
"tokens_seen": 1546256384
},
{
"epoch": 0.47,
"learning_rate": 0.0002684653961885657,
"loss": 2.7537,
"theoretical_loss": 3.50669209830244,
"tokens_seen": 1546321920
},
{
"epoch": 0.47,
"learning_rate": 0.0002684553660982949,
"loss": 2.7924,
"theoretical_loss": 3.5066790712249256,
"tokens_seen": 1546387456
},
{
"epoch": 0.47,
"learning_rate": 0.00026844533600802405,
"loss": 2.6317,
"theoretical_loss": 3.5066660448540645,
"tokens_seen": 1546452992
},
{
"epoch": 0.47,
"learning_rate": 0.0002684353059177533,
"loss": 2.6128,
"theoretical_loss": 3.50665301918979,
"tokens_seen": 1546518528
},
{
"epoch": 0.47,
"learning_rate": 0.0002684252758274824,
"loss": 2.4097,
"theoretical_loss": 3.5066399942320325,
"tokens_seen": 1546584064
},
{
"epoch": 0.47,
"objective/train/docs_used": 877732,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6479623317718506,
"objective/train/theoretical_loss": 3.506626969980724,
"objective/train/tokens_used": 1567109600,
"theoretical_loss": 3.506626969980724,
"tokens_seen": 1546649600
},
{
"epoch": 0.47,
"learning_rate": 0.00026841524573721165,
"loss": 2.6701,
"theoretical_loss": 3.506626969980724,
"tokens_seen": 1546649600
},
{
"epoch": 0.47,
"learning_rate": 0.00026840521564694083,
"loss": 2.576,
"theoretical_loss": 3.506613946435797,
"tokens_seen": 1546715136
},
{
"epoch": 0.47,
"learning_rate": 0.00026839518555667,
"loss": 2.5103,
"theoretical_loss": 3.5066009235971825,
"tokens_seen": 1546780672
},
{
"epoch": 0.47,
"learning_rate": 0.0002683851554663992,
"loss": 2.6414,
"theoretical_loss": 3.506587901464812,
"tokens_seen": 1546846208
},
{
"epoch": 0.47,
"learning_rate": 0.00026837512537612843,
"loss": 2.4312,
"theoretical_loss": 3.5065748800386176,
"tokens_seen": 1546911744
},
{
"epoch": 0.47,
"learning_rate": 0.00026836509528585756,
"loss": 2.5681,
"theoretical_loss": 3.506561859318532,
"tokens_seen": 1546977280
},
{
"epoch": 0.47,
"learning_rate": 0.0002683550651955868,
"loss": 2.6809,
"theoretical_loss": 3.506548839304486,
"tokens_seen": 1547042816
},
{
"epoch": 0.47,
"learning_rate": 0.0002683450351053159,
"loss": 2.5533,
"theoretical_loss": 3.506535819996411,
"tokens_seen": 1547108352
},
{
"epoch": 0.47,
"learning_rate": 0.00026833500501504515,
"loss": 2.5929,
"theoretical_loss": 3.50652280139424,
"tokens_seen": 1547173888
},
{
"epoch": 0.47,
"learning_rate": 0.00026832497492477433,
"loss": 2.8098,
"theoretical_loss": 3.506509783497904,
"tokens_seen": 1547239424
},
{
"epoch": 0.47,
"learning_rate": 0.0002683149448345035,
"loss": 2.6754,
"theoretical_loss": 3.506496766307335,
"tokens_seen": 1547304960
},
{
"epoch": 0.47,
"learning_rate": 0.0002683049147442327,
"loss": 2.6497,
"theoretical_loss": 3.5064837498224657,
"tokens_seen": 1547370496
},
{
"epoch": 0.47,
"learning_rate": 0.0002682948846539619,
"loss": 2.5738,
"theoretical_loss": 3.5064707340432264,
"tokens_seen": 1547436032
},
{
"epoch": 0.47,
"learning_rate": 0.00026828485456369106,
"loss": 2.6337,
"theoretical_loss": 3.5064577189695503,
"tokens_seen": 1547501568
},
{
"epoch": 0.47,
"learning_rate": 0.0002682748244734203,
"loss": 2.8737,
"theoretical_loss": 3.5064447046013685,
"tokens_seen": 1547567104
},
{
"epoch": 0.47,
"learning_rate": 0.0002682647943831494,
"loss": 2.8289,
"theoretical_loss": 3.506431690938613,
"tokens_seen": 1547632640
},
{
"epoch": 0.47,
"learning_rate": 0.00026825476429287866,
"loss": 2.9707,
"theoretical_loss": 3.506418677981216,
"tokens_seen": 1547698176
},
{
"epoch": 0.47,
"learning_rate": 0.0002682447342026078,
"loss": 2.6165,
"theoretical_loss": 3.5064056657291096,
"tokens_seen": 1547763712
},
{
"epoch": 0.47,
"learning_rate": 0.000268234704112337,
"loss": 2.5366,
"theoretical_loss": 3.506392654182225,
"tokens_seen": 1547829248
},
{
"epoch": 0.47,
"learning_rate": 0.0002682246740220662,
"loss": 2.7704,
"theoretical_loss": 3.5063796433404946,
"tokens_seen": 1547894784
},
{
"epoch": 0.47,
"learning_rate": 0.0002682146439317954,
"loss": 2.7348,
"theoretical_loss": 3.50636663320385,
"tokens_seen": 1547960320
},
{
"epoch": 0.47,
"learning_rate": 0.00026820461384152456,
"loss": 2.629,
"theoretical_loss": 3.506353623772223,
"tokens_seen": 1548025856
},
{
"epoch": 0.47,
"learning_rate": 0.0002681945837512538,
"loss": 2.7726,
"theoretical_loss": 3.5063406150455467,
"tokens_seen": 1548091392
},
{
"epoch": 0.47,
"learning_rate": 0.0002681845536609829,
"loss": 2.6825,
"theoretical_loss": 3.506327607023752,
"tokens_seen": 1548156928
},
{
"epoch": 0.47,
"learning_rate": 0.00026817452357071216,
"loss": 2.6144,
"theoretical_loss": 3.50631459970677,
"tokens_seen": 1548222464
},
{
"epoch": 0.47,
"objective/train/docs_used": 878530,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6239328384399414,
"objective/train/theoretical_loss": 3.5063015930945354,
"objective/train/tokens_used": 1568748000,
"theoretical_loss": 3.5063015930945354,
"tokens_seen": 1548288000
},
{
"epoch": 0.47,
"learning_rate": 0.0002681644934804413,
"loss": 2.499,
"theoretical_loss": 3.5063015930945354,
"tokens_seen": 1548288000
},
{
"epoch": 0.47,
"learning_rate": 0.0002681544633901705,
"loss": 2.8261,
"theoretical_loss": 3.506288587186978,
"tokens_seen": 1548353536
},
{
"epoch": 0.47,
"learning_rate": 0.0002681444332998997,
"loss": 2.6338,
"theoretical_loss": 3.50627558198403,
"tokens_seen": 1548419072
},
{
"epoch": 0.47,
"learning_rate": 0.0002681344032096289,
"loss": 2.7533,
"theoretical_loss": 3.506262577485624,
"tokens_seen": 1548484608
},
{
"epoch": 0.47,
"learning_rate": 0.00026812437311935807,
"loss": 2.9436,
"theoretical_loss": 3.506249573691692,
"tokens_seen": 1548550144
},
{
"epoch": 0.47,
"learning_rate": 0.00026811434302908725,
"loss": 2.7,
"theoretical_loss": 3.506236570602166,
"tokens_seen": 1548615680
},
{
"epoch": 0.47,
"learning_rate": 0.00026810431293881643,
"loss": 2.5738,
"theoretical_loss": 3.5062235682169773,
"tokens_seen": 1548681216
},
{
"epoch": 0.47,
"learning_rate": 0.00026809428284854566,
"loss": 2.7697,
"theoretical_loss": 3.5062105665360592,
"tokens_seen": 1548746752
},
{
"epoch": 0.47,
"learning_rate": 0.0002680842527582748,
"loss": 2.5787,
"theoretical_loss": 3.5061975655593423,
"tokens_seen": 1548812288
},
{
"epoch": 0.47,
"learning_rate": 0.000268074222668004,
"loss": 2.4572,
"theoretical_loss": 3.50618456528676,
"tokens_seen": 1548877824
},
{
"epoch": 0.47,
"learning_rate": 0.00026806419257773315,
"loss": 2.5176,
"theoretical_loss": 3.5061715657182435,
"tokens_seen": 1548943360
},
{
"epoch": 0.47,
"learning_rate": 0.0002680541624874624,
"loss": 2.6288,
"theoretical_loss": 3.506158566853725,
"tokens_seen": 1549008896
},
{
"epoch": 0.47,
"learning_rate": 0.00026804413239719157,
"loss": 2.7368,
"theoretical_loss": 3.506145568693137,
"tokens_seen": 1549074432
},
{
"epoch": 0.47,
"learning_rate": 0.00026803410230692075,
"loss": 2.8123,
"theoretical_loss": 3.506132571236412,
"tokens_seen": 1549139968
},
{
"epoch": 0.47,
"learning_rate": 0.00026802407221664993,
"loss": 2.8158,
"theoretical_loss": 3.5061195744834808,
"tokens_seen": 1549205504
},
{
"epoch": 0.47,
"learning_rate": 0.00026801404212637917,
"loss": 2.5059,
"theoretical_loss": 3.5061065784342764,
"tokens_seen": 1549271040
},
{
"epoch": 0.47,
"learning_rate": 0.00026800401203610835,
"loss": 2.5957,
"theoretical_loss": 3.506093583088731,
"tokens_seen": 1549336576
},
{
"epoch": 0.47,
"learning_rate": 0.00026799398194583753,
"loss": 2.6548,
"theoretical_loss": 3.5060805884467756,
"tokens_seen": 1549402112
},
{
"epoch": 0.47,
"learning_rate": 0.0002679839518555667,
"loss": 2.5968,
"theoretical_loss": 3.5060675945083437,
"tokens_seen": 1549467648
},
{
"epoch": 0.47,
"learning_rate": 0.0002679739217652959,
"loss": 2.4685,
"theoretical_loss": 3.5060546012733673,
"tokens_seen": 1549533184
},
{
"epoch": 0.47,
"learning_rate": 0.0002679638916750251,
"loss": 2.6454,
"theoretical_loss": 3.506041608741778,
"tokens_seen": 1549598720
},
{
"epoch": 0.47,
"learning_rate": 0.00026795386158475425,
"loss": 2.88,
"theoretical_loss": 3.5060286169135084,
"tokens_seen": 1549664256
},
{
"epoch": 0.47,
"learning_rate": 0.0002679438314944835,
"loss": 2.6665,
"theoretical_loss": 3.5060156257884905,
"tokens_seen": 1549729792
},
{
"epoch": 0.47,
"learning_rate": 0.0002679338014042126,
"loss": 2.5303,
"theoretical_loss": 3.5060026353666562,
"tokens_seen": 1549795328
},
{
"epoch": 0.47,
"learning_rate": 0.00026792377131394185,
"loss": 2.6048,
"theoretical_loss": 3.5059896456479382,
"tokens_seen": 1549860864
},
{
"epoch": 0.47,
"objective/train/docs_used": 879841,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.635636329650879,
"objective/train/theoretical_loss": 3.505976656632269,
"objective/train/tokens_used": 1570386400,
"theoretical_loss": 3.505976656632269,
"tokens_seen": 1549926400
},
{
"epoch": 0.47,
"learning_rate": 0.00026791374122367103,
"loss": 2.5817,
"theoretical_loss": 3.505976656632269,
"tokens_seen": 1549926400
},
{
"epoch": 0.47,
"learning_rate": 0.0002679037111334002,
"loss": 2.6367,
"theoretical_loss": 3.5059636683195796,
"tokens_seen": 1549991936
},
{
"epoch": 0.47,
"learning_rate": 0.0002678936810431294,
"loss": 2.6765,
"theoretical_loss": 3.5059506807098035,
"tokens_seen": 1550057472
},
{
"epoch": 0.47,
"learning_rate": 0.00026788365095285863,
"loss": 2.8322,
"theoretical_loss": 3.505937693802872,
"tokens_seen": 1550123008
},
{
"epoch": 0.47,
"learning_rate": 0.00026787362086258776,
"loss": 2.5005,
"theoretical_loss": 3.505924707598718,
"tokens_seen": 1550188544
},
{
"epoch": 0.47,
"learning_rate": 0.000267863590772317,
"loss": 2.7236,
"theoretical_loss": 3.5059117220972738,
"tokens_seen": 1550254080
},
{
"epoch": 0.47,
"learning_rate": 0.0002678535606820461,
"loss": 2.7141,
"theoretical_loss": 3.5058987372984713,
"tokens_seen": 1550319616
},
{
"epoch": 0.47,
"learning_rate": 0.00026784353059177535,
"loss": 2.5009,
"theoretical_loss": 3.5058857532022425,
"tokens_seen": 1550385152
},
{
"epoch": 0.47,
"learning_rate": 0.00026783350050150453,
"loss": 2.7587,
"theoretical_loss": 3.50587276980852,
"tokens_seen": 1550450688
},
{
"epoch": 0.47,
"learning_rate": 0.0002678234704112337,
"loss": 2.6192,
"theoretical_loss": 3.5058597871172372,
"tokens_seen": 1550516224
},
{
"epoch": 0.47,
"learning_rate": 0.0002678134403209629,
"loss": 2.6174,
"theoretical_loss": 3.5058468051283245,
"tokens_seen": 1550581760
},
{
"epoch": 0.47,
"learning_rate": 0.0002678034102306921,
"loss": 2.7398,
"theoretical_loss": 3.5058338238417157,
"tokens_seen": 1550647296
},
{
"epoch": 0.47,
"learning_rate": 0.00026779338014042126,
"loss": 2.6339,
"theoretical_loss": 3.505820843257342,
"tokens_seen": 1550712832
},
{
"epoch": 0.47,
"learning_rate": 0.0002677833500501505,
"loss": 2.8044,
"theoretical_loss": 3.5058078633751366,
"tokens_seen": 1550778368
},
{
"epoch": 0.47,
"learning_rate": 0.0002677733199598796,
"loss": 2.6142,
"theoretical_loss": 3.505794884195031,
"tokens_seen": 1550843904
},
{
"epoch": 0.47,
"learning_rate": 0.00026776328986960886,
"loss": 2.8606,
"theoretical_loss": 3.5057819057169586,
"tokens_seen": 1550909440
},
{
"epoch": 0.47,
"learning_rate": 0.000267753259779338,
"loss": 2.636,
"theoretical_loss": 3.505768927940851,
"tokens_seen": 1550974976
},
{
"epoch": 0.47,
"learning_rate": 0.0002677432296890672,
"loss": 2.5509,
"theoretical_loss": 3.505755950866641,
"tokens_seen": 1551040512
},
{
"epoch": 0.47,
"learning_rate": 0.0002677331995987964,
"loss": 2.9173,
"theoretical_loss": 3.505742974494261,
"tokens_seen": 1551106048
},
{
"epoch": 0.47,
"learning_rate": 0.0002677231695085256,
"loss": 2.8309,
"theoretical_loss": 3.5057299988236426,
"tokens_seen": 1551171584
},
{
"epoch": 0.47,
"learning_rate": 0.00026771313941825476,
"loss": 2.5229,
"theoretical_loss": 3.505717023854719,
"tokens_seen": 1551237120
},
{
"epoch": 0.47,
"learning_rate": 0.000267703109327984,
"loss": 2.7206,
"theoretical_loss": 3.5057040495874228,
"tokens_seen": 1551302656
},
{
"epoch": 0.47,
"learning_rate": 0.0002676930792377131,
"loss": 2.6336,
"theoretical_loss": 3.5056910760216855,
"tokens_seen": 1551368192
},
{
"epoch": 0.47,
"learning_rate": 0.00026768304914744236,
"loss": 2.4705,
"theoretical_loss": 3.5056781031574404,
"tokens_seen": 1551433728
},
{
"epoch": 0.47,
"learning_rate": 0.0002676730190571715,
"loss": 2.4987,
"theoretical_loss": 3.505665130994619,
"tokens_seen": 1551499264
},
{
"epoch": 0.47,
"objective/train/docs_used": 880523,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0017781257629395,
"objective/train/theoretical_loss": 3.505652159533155,
"objective/train/tokens_used": 1572024800,
"theoretical_loss": 3.505652159533155,
"tokens_seen": 1551564800
},
{
"epoch": 0.47,
"learning_rate": 0.0002676629889669007,
"loss": 2.7547,
"theoretical_loss": 3.505652159533155,
"tokens_seen": 1551564800
},
{
"epoch": 0.47,
"learning_rate": 0.0002676529588766299,
"loss": 2.519,
"theoretical_loss": 3.5056391887729803,
"tokens_seen": 1551630336
},
{
"epoch": 0.47,
"learning_rate": 0.0002676429287863591,
"loss": 2.6519,
"theoretical_loss": 3.505626218714027,
"tokens_seen": 1551695872
},
{
"epoch": 0.47,
"learning_rate": 0.00026763289869608827,
"loss": 2.7106,
"theoretical_loss": 3.5056132493562275,
"tokens_seen": 1551761408
},
{
"epoch": 0.47,
"learning_rate": 0.00026762286860581745,
"loss": 2.8936,
"theoretical_loss": 3.5056002806995155,
"tokens_seen": 1551826944
},
{
"epoch": 0.47,
"learning_rate": 0.00026761283851554663,
"loss": 2.9993,
"theoretical_loss": 3.5055873127438217,
"tokens_seen": 1551892480
},
{
"epoch": 0.47,
"learning_rate": 0.00026760280842527586,
"loss": 2.5602,
"theoretical_loss": 3.50557434548908,
"tokens_seen": 1551958016
},
{
"epoch": 0.47,
"learning_rate": 0.000267592778335005,
"loss": 2.6356,
"theoretical_loss": 3.505561378935223,
"tokens_seen": 1552023552
},
{
"epoch": 0.47,
"learning_rate": 0.0002675827482447342,
"loss": 2.5299,
"theoretical_loss": 3.5055484130821815,
"tokens_seen": 1552089088
},
{
"epoch": 0.47,
"learning_rate": 0.00026757271815446335,
"loss": 2.6635,
"theoretical_loss": 3.5055354479298897,
"tokens_seen": 1552154624
},
{
"epoch": 0.47,
"learning_rate": 0.0002675626880641926,
"loss": 2.7642,
"theoretical_loss": 3.5055224834782797,
"tokens_seen": 1552220160
},
{
"epoch": 0.47,
"learning_rate": 0.00026755265797392177,
"loss": 2.7377,
"theoretical_loss": 3.5055095197272843,
"tokens_seen": 1552285696
},
{
"epoch": 0.47,
"learning_rate": 0.00026754262788365095,
"loss": 2.6878,
"theoretical_loss": 3.5054965566768352,
"tokens_seen": 1552351232
},
{
"epoch": 0.47,
"learning_rate": 0.00026753259779338013,
"loss": 2.8357,
"theoretical_loss": 3.505483594326866,
"tokens_seen": 1552416768
},
{
"epoch": 0.47,
"learning_rate": 0.00026752256770310937,
"loss": 2.4021,
"theoretical_loss": 3.5054706326773086,
"tokens_seen": 1552482304
},
{
"epoch": 0.47,
"learning_rate": 0.0002675125376128385,
"loss": 2.5542,
"theoretical_loss": 3.5054576717280956,
"tokens_seen": 1552547840
},
{
"epoch": 0.47,
"learning_rate": 0.00026750250752256773,
"loss": 2.7443,
"theoretical_loss": 3.50544471147916,
"tokens_seen": 1552613376
},
{
"epoch": 0.47,
"learning_rate": 0.00026749247743229686,
"loss": 2.5975,
"theoretical_loss": 3.505431751930434,
"tokens_seen": 1552678912
},
{
"epoch": 0.47,
"learning_rate": 0.0002674824473420261,
"loss": 2.6074,
"theoretical_loss": 3.505418793081851,
"tokens_seen": 1552744448
},
{
"epoch": 0.47,
"learning_rate": 0.00026747241725175527,
"loss": 2.7654,
"theoretical_loss": 3.5054058349333426,
"tokens_seen": 1552809984
},
{
"epoch": 0.47,
"learning_rate": 0.00026746238716148445,
"loss": 2.5694,
"theoretical_loss": 3.5053928774848417,
"tokens_seen": 1552875520
},
{
"epoch": 0.47,
"learning_rate": 0.00026745235707121363,
"loss": 2.6294,
"theoretical_loss": 3.5053799207362815,
"tokens_seen": 1552941056
},
{
"epoch": 0.47,
"learning_rate": 0.0002674423269809428,
"loss": 2.723,
"theoretical_loss": 3.5053669646875942,
"tokens_seen": 1553006592
},
{
"epoch": 0.47,
"learning_rate": 0.000267432296890672,
"loss": 2.5161,
"theoretical_loss": 3.5053540093387126,
"tokens_seen": 1553072128
},
{
"epoch": 0.47,
"learning_rate": 0.00026742226680040123,
"loss": 2.7467,
"theoretical_loss": 3.50534105468957,
"tokens_seen": 1553137664
},
{
"epoch": 0.47,
"objective/train/docs_used": 881131,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1730709075927734,
"objective/train/theoretical_loss": 3.505328100740097,
"objective/train/tokens_used": 1573663200,
"theoretical_loss": 3.505328100740097,
"tokens_seen": 1553203200
},
{
"epoch": 0.47,
"learning_rate": 0.00026741223671013036,
"loss": 2.6465,
"theoretical_loss": 3.505328100740097,
"tokens_seen": 1553203200
},
{
"epoch": 0.47,
"learning_rate": 0.0002674022066198596,
"loss": 2.367,
"theoretical_loss": 3.5053151474902284,
"tokens_seen": 1553268736
},
{
"epoch": 0.47,
"learning_rate": 0.0002673921765295888,
"loss": 2.6143,
"theoretical_loss": 3.505302194939896,
"tokens_seen": 1553334272
},
{
"epoch": 0.47,
"learning_rate": 0.00026738214643931796,
"loss": 2.5055,
"theoretical_loss": 3.505289243089033,
"tokens_seen": 1553399808
},
{
"epoch": 0.47,
"learning_rate": 0.00026737211634904714,
"loss": 2.3749,
"theoretical_loss": 3.5052762919375717,
"tokens_seen": 1553465344
},
{
"epoch": 0.47,
"learning_rate": 0.0002673620862587763,
"loss": 2.7546,
"theoretical_loss": 3.5052633414854446,
"tokens_seen": 1553530880
},
{
"epoch": 0.47,
"learning_rate": 0.0002673520561685055,
"loss": 2.7103,
"theoretical_loss": 3.5052503917325852,
"tokens_seen": 1553596416
},
{
"epoch": 0.47,
"learning_rate": 0.00026734202607823474,
"loss": 2.813,
"theoretical_loss": 3.505237442678926,
"tokens_seen": 1553661952
},
{
"epoch": 0.47,
"learning_rate": 0.00026733199598796386,
"loss": 2.7728,
"theoretical_loss": 3.5052244943243993,
"tokens_seen": 1553727488
},
{
"epoch": 0.47,
"learning_rate": 0.0002673219658976931,
"loss": 2.5827,
"theoretical_loss": 3.505211546668938,
"tokens_seen": 1553793024
},
{
"epoch": 0.47,
"learning_rate": 0.0002673119358074222,
"loss": 2.373,
"theoretical_loss": 3.5051985997124757,
"tokens_seen": 1553858560
},
{
"epoch": 0.47,
"learning_rate": 0.00026730190571715146,
"loss": 2.7512,
"theoretical_loss": 3.5051856534549444,
"tokens_seen": 1553924096
},
{
"epoch": 0.47,
"learning_rate": 0.00026729187562688064,
"loss": 2.7125,
"theoretical_loss": 3.505172707896276,
"tokens_seen": 1553989632
},
{
"epoch": 0.47,
"learning_rate": 0.0002672818455366098,
"loss": 2.7901,
"theoretical_loss": 3.5051597630364055,
"tokens_seen": 1554055168
},
{
"epoch": 0.47,
"learning_rate": 0.000267271815446339,
"loss": 2.8147,
"theoretical_loss": 3.505146818875264,
"tokens_seen": 1554120704
},
{
"epoch": 0.47,
"learning_rate": 0.0002672617853560682,
"loss": 2.6613,
"theoretical_loss": 3.5051338754127848,
"tokens_seen": 1554186240
},
{
"epoch": 0.47,
"learning_rate": 0.0002672517552657974,
"loss": 2.7096,
"theoretical_loss": 3.505120932648901,
"tokens_seen": 1554251776
},
{
"epoch": 0.47,
"learning_rate": 0.0002672417251755266,
"loss": 2.4554,
"theoretical_loss": 3.505107990583545,
"tokens_seen": 1554317312
},
{
"epoch": 0.47,
"learning_rate": 0.0002672316950852558,
"loss": 2.5855,
"theoretical_loss": 3.50509504921665,
"tokens_seen": 1554382848
},
{
"epoch": 0.47,
"learning_rate": 0.00026722166499498496,
"loss": 2.6649,
"theoretical_loss": 3.505082108548149,
"tokens_seen": 1554448384
},
{
"epoch": 0.47,
"learning_rate": 0.0002672116349047142,
"loss": 2.5386,
"theoretical_loss": 3.5050691685779745,
"tokens_seen": 1554513920
},
{
"epoch": 0.47,
"learning_rate": 0.0002672016048144433,
"loss": 2.7686,
"theoretical_loss": 3.505056229306059,
"tokens_seen": 1554579456
},
{
"epoch": 0.47,
"learning_rate": 0.00026719157472417256,
"loss": 2.5,
"theoretical_loss": 3.5050432907323366,
"tokens_seen": 1554644992
},
{
"epoch": 0.47,
"learning_rate": 0.0002671815446339017,
"loss": 2.7718,
"theoretical_loss": 3.505030352856739,
"tokens_seen": 1554710528
},
{
"epoch": 0.47,
"learning_rate": 0.0002671715145436309,
"loss": 2.6597,
"theoretical_loss": 3.5050174156791996,
"tokens_seen": 1554776064
},
{
"epoch": 0.47,
"objective/train/docs_used": 882511,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9305777549743652,
"objective/train/theoretical_loss": 3.505004479199651,
"objective/train/tokens_used": 1575301600,
"theoretical_loss": 3.505004479199651,
"tokens_seen": 1554841600
},
{
"epoch": 0.47,
"learning_rate": 0.0002671614844533601,
"loss": 2.6047,
"theoretical_loss": 3.505004479199651,
"tokens_seen": 1554841600
},
{
"epoch": 0.47,
"learning_rate": 0.0002671514543630893,
"loss": 2.5781,
"theoretical_loss": 3.504991543418027,
"tokens_seen": 1554907136
},
{
"epoch": 0.47,
"learning_rate": 0.00026714142427281847,
"loss": 2.6586,
"theoretical_loss": 3.50497860833426,
"tokens_seen": 1554972672
},
{
"epoch": 0.47,
"learning_rate": 0.00026713139418254765,
"loss": 2.6383,
"theoretical_loss": 3.5049656739482824,
"tokens_seen": 1555038208
},
{
"epoch": 0.47,
"learning_rate": 0.00026712136409227683,
"loss": 2.5692,
"theoretical_loss": 3.504952740260028,
"tokens_seen": 1555103744
},
{
"epoch": 0.47,
"learning_rate": 0.00026711133400200606,
"loss": 2.7342,
"theoretical_loss": 3.504939807269429,
"tokens_seen": 1555169280
},
{
"epoch": 0.47,
"learning_rate": 0.0002671013039117352,
"loss": 2.2573,
"theoretical_loss": 3.5049268749764186,
"tokens_seen": 1555234816
},
{
"epoch": 0.47,
"learning_rate": 0.0002670912738214644,
"loss": 2.5,
"theoretical_loss": 3.504913943380931,
"tokens_seen": 1555300352
},
{
"epoch": 0.47,
"learning_rate": 0.00026708124373119355,
"loss": 2.7168,
"theoretical_loss": 3.5049010124828968,
"tokens_seen": 1555365888
},
{
"epoch": 0.47,
"learning_rate": 0.0002670712136409228,
"loss": 2.6414,
"theoretical_loss": 3.5048880822822515,
"tokens_seen": 1555431424
},
{
"epoch": 0.47,
"learning_rate": 0.00026706118355065197,
"loss": 2.7499,
"theoretical_loss": 3.5048751527789266,
"tokens_seen": 1555496960
},
{
"epoch": 0.47,
"learning_rate": 0.00026705115346038115,
"loss": 2.5104,
"theoretical_loss": 3.504862223972855,
"tokens_seen": 1555562496
},
{
"epoch": 0.47,
"learning_rate": 0.00026704112337011033,
"loss": 2.7877,
"theoretical_loss": 3.5048492958639708,
"tokens_seen": 1555628032
},
{
"epoch": 0.47,
"learning_rate": 0.00026703109327983957,
"loss": 2.7009,
"theoretical_loss": 3.5048363684522057,
"tokens_seen": 1555693568
},
{
"epoch": 0.47,
"learning_rate": 0.0002670210631895687,
"loss": 2.6163,
"theoretical_loss": 3.504823441737494,
"tokens_seen": 1555759104
},
{
"epoch": 0.47,
"learning_rate": 0.00026701103309929793,
"loss": 2.665,
"theoretical_loss": 3.504810515719768,
"tokens_seen": 1555824640
},
{
"epoch": 0.47,
"learning_rate": 0.00026700100300902706,
"loss": 2.9984,
"theoretical_loss": 3.5047975903989617,
"tokens_seen": 1555890176
},
{
"epoch": 0.47,
"learning_rate": 0.0002669909729187563,
"loss": 2.6251,
"theoretical_loss": 3.5047846657750066,
"tokens_seen": 1555955712
},
{
"epoch": 0.47,
"learning_rate": 0.00026698094282848547,
"loss": 2.6235,
"theoretical_loss": 3.504771741847837,
"tokens_seen": 1556021248
},
{
"epoch": 0.47,
"learning_rate": 0.00026697091273821465,
"loss": 2.6511,
"theoretical_loss": 3.5047588186173853,
"tokens_seen": 1556086784
},
{
"epoch": 0.47,
"learning_rate": 0.00026696088264794383,
"loss": 2.7389,
"theoretical_loss": 3.5047458960835853,
"tokens_seen": 1556152320
},
{
"epoch": 0.47,
"learning_rate": 0.000266950852557673,
"loss": 2.4737,
"theoretical_loss": 3.5047329742463695,
"tokens_seen": 1556217856
},
{
"epoch": 0.47,
"learning_rate": 0.0002669408224674022,
"loss": 2.6356,
"theoretical_loss": 3.5047200531056713,
"tokens_seen": 1556283392
},
{
"epoch": 0.47,
"learning_rate": 0.00026693079237713143,
"loss": 2.5974,
"theoretical_loss": 3.504707132661424,
"tokens_seen": 1556348928
},
{
"epoch": 0.47,
"learning_rate": 0.00026692076228686056,
"loss": 2.5318,
"theoretical_loss": 3.5046942129135603,
"tokens_seen": 1556414464
},
{
"epoch": 0.47,
"objective/train/docs_used": 882978,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.819887161254883,
"objective/train/theoretical_loss": 3.5046812938620135,
"objective/train/tokens_used": 1576940000,
"theoretical_loss": 3.5046812938620135,
"tokens_seen": 1556480000
},
{
"epoch": 0.47,
"learning_rate": 0.0002669107321965898,
"loss": 2.6064,
"theoretical_loss": 3.5046812938620135,
"tokens_seen": 1556480000
},
{
"epoch": 0.47,
"learning_rate": 0.000266900702106319,
"loss": 2.6159,
"theoretical_loss": 3.504668375506717,
"tokens_seen": 1556545536
},
{
"epoch": 0.47,
"learning_rate": 0.00026689067201604816,
"loss": 2.5725,
"theoretical_loss": 3.504655457847604,
"tokens_seen": 1556611072
},
{
"epoch": 0.47,
"learning_rate": 0.00026688064192577734,
"loss": 2.5182,
"theoretical_loss": 3.504642540884607,
"tokens_seen": 1556676608
},
{
"epoch": 0.47,
"learning_rate": 0.0002668706118355065,
"loss": 2.5463,
"theoretical_loss": 3.5046296246176594,
"tokens_seen": 1556742144
},
{
"epoch": 0.47,
"learning_rate": 0.0002668605817452357,
"loss": 2.6427,
"theoretical_loss": 3.5046167090466955,
"tokens_seen": 1556807680
},
{
"epoch": 0.47,
"learning_rate": 0.00026685055165496494,
"loss": 2.8632,
"theoretical_loss": 3.504603794171647,
"tokens_seen": 1556873216
},
{
"epoch": 0.47,
"learning_rate": 0.00026684052156469406,
"loss": 2.541,
"theoretical_loss": 3.504590879992448,
"tokens_seen": 1556938752
},
{
"epoch": 0.47,
"learning_rate": 0.0002668304914744233,
"loss": 2.4878,
"theoretical_loss": 3.5045779665090313,
"tokens_seen": 1557004288
},
{
"epoch": 0.47,
"learning_rate": 0.0002668204613841524,
"loss": 2.7127,
"theoretical_loss": 3.5045650537213304,
"tokens_seen": 1557069824
},
{
"epoch": 0.47,
"learning_rate": 0.00026681043129388166,
"loss": 2.5596,
"theoretical_loss": 3.5045521416292784,
"tokens_seen": 1557135360
},
{
"epoch": 0.47,
"learning_rate": 0.00026680040120361084,
"loss": 2.5201,
"theoretical_loss": 3.5045392302328087,
"tokens_seen": 1557200896
},
{
"epoch": 0.47,
"learning_rate": 0.00026679037111334,
"loss": 2.6212,
"theoretical_loss": 3.5045263195318546,
"tokens_seen": 1557266432
},
{
"epoch": 0.47,
"learning_rate": 0.0002667803410230692,
"loss": 2.605,
"theoretical_loss": 3.5045134095263486,
"tokens_seen": 1557331968
},
{
"epoch": 0.47,
"learning_rate": 0.0002667703109327984,
"loss": 2.6027,
"theoretical_loss": 3.504500500216225,
"tokens_seen": 1557397504
},
{
"epoch": 0.47,
"learning_rate": 0.00026676028084252757,
"loss": 2.5275,
"theoretical_loss": 3.504487591601417,
"tokens_seen": 1557463040
},
{
"epoch": 0.47,
"learning_rate": 0.0002667502507522568,
"loss": 2.6949,
"theoretical_loss": 3.504474683681857,
"tokens_seen": 1557528576
},
{
"epoch": 0.47,
"learning_rate": 0.00026674022066198593,
"loss": 2.7122,
"theoretical_loss": 3.504461776457479,
"tokens_seen": 1557594112
},
{
"epoch": 0.47,
"learning_rate": 0.00026673019057171516,
"loss": 2.7402,
"theoretical_loss": 3.504448869928216,
"tokens_seen": 1557659648
},
{
"epoch": 0.47,
"learning_rate": 0.00026672016048144434,
"loss": 2.7264,
"theoretical_loss": 3.504435964094002,
"tokens_seen": 1557725184
},
{
"epoch": 0.47,
"learning_rate": 0.0002667101303911735,
"loss": 2.5973,
"theoretical_loss": 3.504423058954769,
"tokens_seen": 1557790720
},
{
"epoch": 0.47,
"learning_rate": 0.0002667001003009027,
"loss": 2.6828,
"theoretical_loss": 3.504410154510452,
"tokens_seen": 1557856256
},
{
"epoch": 0.47,
"learning_rate": 0.0002666900702106319,
"loss": 2.4192,
"theoretical_loss": 3.5043972507609835,
"tokens_seen": 1557921792
},
{
"epoch": 0.47,
"learning_rate": 0.00026668004012036107,
"loss": 2.7397,
"theoretical_loss": 3.5043843477062966,
"tokens_seen": 1557987328
},
{
"epoch": 0.47,
"learning_rate": 0.0002666700100300903,
"loss": 2.6286,
"theoretical_loss": 3.504371445346325,
"tokens_seen": 1558052864
},
{
"epoch": 0.47,
"objective/train/docs_used": 884109,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5533640384674072,
"objective/train/theoretical_loss": 3.5043585436810014,
"objective/train/tokens_used": 1578578400,
"theoretical_loss": 3.5043585436810014,
"tokens_seen": 1558118400
},
{
"epoch": 0.47,
"learning_rate": 0.00026665997993981943,
"loss": 2.6537,
"theoretical_loss": 3.5043585436810014,
"tokens_seen": 1558118400
},
{
"epoch": 0.47,
"learning_rate": 0.00026664994984954867,
"loss": 2.5631,
"theoretical_loss": 3.504345642710261,
"tokens_seen": 1558183936
},
{
"epoch": 0.47,
"learning_rate": 0.0002666399197592778,
"loss": 2.6689,
"theoretical_loss": 3.504332742434035,
"tokens_seen": 1558249472
},
{
"epoch": 0.47,
"learning_rate": 0.00026662988966900703,
"loss": 2.6026,
"theoretical_loss": 3.504319842852258,
"tokens_seen": 1558315008
},
{
"epoch": 0.47,
"learning_rate": 0.0002666198595787362,
"loss": 2.5855,
"theoretical_loss": 3.5043069439648633,
"tokens_seen": 1558380544
},
{
"epoch": 0.47,
"learning_rate": 0.0002666098294884654,
"loss": 2.6296,
"theoretical_loss": 3.5042940457717844,
"tokens_seen": 1558446080
},
{
"epoch": 0.47,
"learning_rate": 0.00026659979939819457,
"loss": 2.5558,
"theoretical_loss": 3.5042811482729546,
"tokens_seen": 1558511616
},
{
"epoch": 0.47,
"learning_rate": 0.00026658976930792375,
"loss": 2.5571,
"theoretical_loss": 3.5042682514683072,
"tokens_seen": 1558577152
},
{
"epoch": 0.47,
"learning_rate": 0.00026657973921765293,
"loss": 2.852,
"theoretical_loss": 3.5042553553577758,
"tokens_seen": 1558642688
},
{
"epoch": 0.47,
"learning_rate": 0.00026656970912738217,
"loss": 2.6452,
"theoretical_loss": 3.5042424599412936,
"tokens_seen": 1558708224
},
{
"epoch": 0.47,
"learning_rate": 0.0002665596790371113,
"loss": 2.8871,
"theoretical_loss": 3.504229565218794,
"tokens_seen": 1558773760
},
{
"epoch": 0.47,
"learning_rate": 0.00026654964894684053,
"loss": 2.7596,
"theoretical_loss": 3.5042166711902114,
"tokens_seen": 1558839296
},
{
"epoch": 0.47,
"learning_rate": 0.0002665396188565697,
"loss": 2.948,
"theoretical_loss": 3.504203777855478,
"tokens_seen": 1558904832
},
{
"epoch": 0.47,
"learning_rate": 0.0002665295887662989,
"loss": 2.7319,
"theoretical_loss": 3.5041908852145287,
"tokens_seen": 1558970368
},
{
"epoch": 0.47,
"learning_rate": 0.0002665195586760281,
"loss": 2.4232,
"theoretical_loss": 3.5041779932672963,
"tokens_seen": 1559035904
},
{
"epoch": 0.47,
"learning_rate": 0.00026650952858575726,
"loss": 2.6914,
"theoretical_loss": 3.5041651020137135,
"tokens_seen": 1559101440
},
{
"epoch": 0.47,
"learning_rate": 0.0002664994984954865,
"loss": 2.7626,
"theoretical_loss": 3.5041522114537154,
"tokens_seen": 1559166976
},
{
"epoch": 0.47,
"learning_rate": 0.00026648946840521567,
"loss": 2.628,
"theoretical_loss": 3.504139321587234,
"tokens_seen": 1559232512
},
{
"epoch": 0.47,
"learning_rate": 0.00026647943831494485,
"loss": 2.7937,
"theoretical_loss": 3.504126432414204,
"tokens_seen": 1559298048
},
{
"epoch": 0.47,
"learning_rate": 0.00026646940822467403,
"loss": 2.8061,
"theoretical_loss": 3.504113543934558,
"tokens_seen": 1559363584
},
{
"epoch": 0.47,
"learning_rate": 0.0002664593781344032,
"loss": 2.7076,
"theoretical_loss": 3.5041006561482306,
"tokens_seen": 1559429120
},
{
"epoch": 0.47,
"learning_rate": 0.0002664493480441324,
"loss": 2.532,
"theoretical_loss": 3.504087769055155,
"tokens_seen": 1559494656
},
{
"epoch": 0.47,
"learning_rate": 0.00026643931795386163,
"loss": 2.7585,
"theoretical_loss": 3.504074882655264,
"tokens_seen": 1559560192
},
{
"epoch": 0.47,
"learning_rate": 0.00026642928786359076,
"loss": 2.6604,
"theoretical_loss": 3.5040619969484923,
"tokens_seen": 1559625728
},
{
"epoch": 0.47,
"learning_rate": 0.00026641925777332,
"loss": 2.58,
"theoretical_loss": 3.5040491119347728,
"tokens_seen": 1559691264
},
{
"epoch": 0.47,
"objective/train/docs_used": 884823,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6479320526123047,
"objective/train/theoretical_loss": 3.5040362276140393,
"objective/train/tokens_used": 1580216800,
"theoretical_loss": 3.5040362276140393,
"tokens_seen": 1559756800
},
{
"epoch": 0.47,
"learning_rate": 0.0002664092276830492,
"loss": 2.5364,
"theoretical_loss": 3.5040362276140393,
"tokens_seen": 1559756800
},
{
"epoch": 0.47,
"learning_rate": 0.00026639919759277836,
"loss": 2.8928,
"theoretical_loss": 3.5040233439862254,
"tokens_seen": 1559822336
},
{
"epoch": 0.47,
"learning_rate": 0.00026638916750250754,
"loss": 2.7021,
"theoretical_loss": 3.504010461051265,
"tokens_seen": 1559887872
},
{
"epoch": 0.47,
"learning_rate": 0.0002663791374122367,
"loss": 2.5435,
"theoretical_loss": 3.5039975788090914,
"tokens_seen": 1559953408
},
{
"epoch": 0.47,
"learning_rate": 0.0002663691073219659,
"loss": 2.7279,
"theoretical_loss": 3.503984697259638,
"tokens_seen": 1560018944
},
{
"epoch": 0.47,
"learning_rate": 0.00026635907723169514,
"loss": 2.7584,
"theoretical_loss": 3.503971816402839,
"tokens_seen": 1560084480
},
{
"epoch": 0.47,
"learning_rate": 0.00026634904714142426,
"loss": 2.5893,
"theoretical_loss": 3.503958936238628,
"tokens_seen": 1560150016
},
{
"epoch": 0.47,
"learning_rate": 0.0002663390170511535,
"loss": 2.7246,
"theoretical_loss": 3.503946056766938,
"tokens_seen": 1560215552
},
{
"epoch": 0.47,
"learning_rate": 0.0002663289869608826,
"loss": 2.5915,
"theoretical_loss": 3.5039331779877037,
"tokens_seen": 1560281088
},
{
"epoch": 0.47,
"learning_rate": 0.00026631895687061186,
"loss": 2.7421,
"theoretical_loss": 3.503920299900858,
"tokens_seen": 1560346624
},
{
"epoch": 0.47,
"learning_rate": 0.00026630892678034104,
"loss": 2.5641,
"theoretical_loss": 3.5039074225063347,
"tokens_seen": 1560412160
},
{
"epoch": 0.47,
"learning_rate": 0.0002662988966900702,
"loss": 2.9161,
"theoretical_loss": 3.5038945458040684,
"tokens_seen": 1560477696
},
{
"epoch": 0.47,
"learning_rate": 0.0002662888665997994,
"loss": 2.6934,
"theoretical_loss": 3.5038816697939916,
"tokens_seen": 1560543232
},
{
"epoch": 0.47,
"learning_rate": 0.0002662788365095286,
"loss": 2.772,
"theoretical_loss": 3.5038687944760385,
"tokens_seen": 1560608768
},
{
"epoch": 0.47,
"learning_rate": 0.00026626880641925777,
"loss": 2.7673,
"theoretical_loss": 3.503855919850143,
"tokens_seen": 1560674304
},
{
"epoch": 0.47,
"learning_rate": 0.000266258776328987,
"loss": 2.5933,
"theoretical_loss": 3.5038430459162386,
"tokens_seen": 1560739840
},
{
"epoch": 0.47,
"learning_rate": 0.00026624874623871613,
"loss": 2.8264,
"theoretical_loss": 3.503830172674259,
"tokens_seen": 1560805376
},
{
"epoch": 0.47,
"learning_rate": 0.00026623871614844536,
"loss": 2.4203,
"theoretical_loss": 3.503817300124138,
"tokens_seen": 1560870912
},
{
"epoch": 0.47,
"learning_rate": 0.00026622868605817454,
"loss": 2.5878,
"theoretical_loss": 3.50380442826581,
"tokens_seen": 1560936448
},
{
"epoch": 0.47,
"learning_rate": 0.0002662186559679037,
"loss": 2.6932,
"theoretical_loss": 3.503791557099208,
"tokens_seen": 1561001984
},
{
"epoch": 0.47,
"learning_rate": 0.0002662086258776329,
"loss": 2.409,
"theoretical_loss": 3.503778686624266,
"tokens_seen": 1561067520
},
{
"epoch": 0.47,
"learning_rate": 0.0002661985957873621,
"loss": 2.4701,
"theoretical_loss": 3.503765816840918,
"tokens_seen": 1561133056
},
{
"epoch": 0.47,
"learning_rate": 0.00026618856569709127,
"loss": 2.8541,
"theoretical_loss": 3.5037529477490974,
"tokens_seen": 1561198592
},
{
"epoch": 0.47,
"learning_rate": 0.0002661785356068205,
"loss": 2.7212,
"theoretical_loss": 3.5037400793487383,
"tokens_seen": 1561264128
},
{
"epoch": 0.47,
"learning_rate": 0.00026616850551654963,
"loss": 2.6886,
"theoretical_loss": 3.5037272116397746,
"tokens_seen": 1561329664
},
{
"epoch": 0.47,
"objective/train/docs_used": 885915,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7010347843170166,
"objective/train/theoretical_loss": 3.50371434462214,
"objective/train/tokens_used": 1581855200,
"theoretical_loss": 3.50371434462214,
"tokens_seen": 1561395200
},
{
"epoch": 0.47,
"learning_rate": 0.00026615847542627887,
"loss": 2.7246,
"theoretical_loss": 3.50371434462214,
"tokens_seen": 1561395200
},
{
"epoch": 0.47,
"learning_rate": 0.000266148445336008,
"loss": 2.6163,
"theoretical_loss": 3.5037014782957683,
"tokens_seen": 1561460736
},
{
"epoch": 0.47,
"learning_rate": 0.00026613841524573723,
"loss": 2.5536,
"theoretical_loss": 3.5036886126605937,
"tokens_seen": 1561526272
},
{
"epoch": 0.47,
"learning_rate": 0.0002661283851554664,
"loss": 2.5492,
"theoretical_loss": 3.5036757477165494,
"tokens_seen": 1561591808
},
{
"epoch": 0.47,
"learning_rate": 0.0002661183550651956,
"loss": 2.6099,
"theoretical_loss": 3.5036628834635697,
"tokens_seen": 1561657344
},
{
"epoch": 0.47,
"learning_rate": 0.00026610832497492477,
"loss": 2.8971,
"theoretical_loss": 3.5036500199015883,
"tokens_seen": 1561722880
},
{
"epoch": 0.47,
"learning_rate": 0.00026609829488465395,
"loss": 2.7427,
"theoretical_loss": 3.50363715703054,
"tokens_seen": 1561788416
},
{
"epoch": 0.47,
"learning_rate": 0.00026608826479438313,
"loss": 2.7519,
"theoretical_loss": 3.503624294850357,
"tokens_seen": 1561853952
},
{
"epoch": 0.47,
"learning_rate": 0.00026607823470411237,
"loss": 2.5574,
"theoretical_loss": 3.5036114333609745,
"tokens_seen": 1561919488
},
{
"epoch": 0.47,
"learning_rate": 0.0002660682046138415,
"loss": 2.6118,
"theoretical_loss": 3.5035985725623258,
"tokens_seen": 1561985024
},
{
"epoch": 0.47,
"learning_rate": 0.00026605817452357073,
"loss": 2.7323,
"theoretical_loss": 3.5035857124543455,
"tokens_seen": 1562050560
},
{
"epoch": 0.47,
"learning_rate": 0.0002660481444332999,
"loss": 2.5577,
"theoretical_loss": 3.503572853036967,
"tokens_seen": 1562116096
},
{
"epoch": 0.47,
"learning_rate": 0.0002660381143430291,
"loss": 2.6324,
"theoretical_loss": 3.503559994310124,
"tokens_seen": 1562181632
},
{
"epoch": 0.47,
"learning_rate": 0.0002660280842527583,
"loss": 2.7975,
"theoretical_loss": 3.503547136273751,
"tokens_seen": 1562247168
},
{
"epoch": 0.47,
"learning_rate": 0.00026601805416248746,
"loss": 2.4274,
"theoretical_loss": 3.5035342789277824,
"tokens_seen": 1562312704
},
{
"epoch": 0.47,
"learning_rate": 0.00026600802407221664,
"loss": 2.9621,
"theoretical_loss": 3.5035214222721507,
"tokens_seen": 1562378240
},
{
"epoch": 0.47,
"learning_rate": 0.00026599799398194587,
"loss": 2.5275,
"theoretical_loss": 3.503508566306791,
"tokens_seen": 1562443776
},
{
"epoch": 0.47,
"learning_rate": 0.000265987963891675,
"loss": 2.802,
"theoretical_loss": 3.5034957110316367,
"tokens_seen": 1562509312
},
{
"epoch": 0.47,
"learning_rate": 0.00026597793380140423,
"loss": 2.7849,
"theoretical_loss": 3.503482856446622,
"tokens_seen": 1562574848
},
{
"epoch": 0.47,
"learning_rate": 0.00026596790371113336,
"loss": 2.8213,
"theoretical_loss": 3.5034700025516816,
"tokens_seen": 1562640384
},
{
"epoch": 0.47,
"learning_rate": 0.0002659578736208626,
"loss": 2.7502,
"theoretical_loss": 3.5034571493467483,
"tokens_seen": 1562705920
},
{
"epoch": 0.47,
"learning_rate": 0.0002659478435305918,
"loss": 2.7614,
"theoretical_loss": 3.5034442968317574,
"tokens_seen": 1562771456
},
{
"epoch": 0.47,
"learning_rate": 0.00026593781344032096,
"loss": 2.7169,
"theoretical_loss": 3.5034314450066417,
"tokens_seen": 1562836992
},
{
"epoch": 0.47,
"learning_rate": 0.00026592778335005014,
"loss": 2.6779,
"theoretical_loss": 3.503418593871336,
"tokens_seen": 1562902528
},
{
"epoch": 0.47,
"learning_rate": 0.0002659177532597794,
"loss": 2.4373,
"theoretical_loss": 3.503405743425774,
"tokens_seen": 1562968064
},
{
"epoch": 0.47,
"objective/train/docs_used": 886454,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7232253551483154,
"objective/train/theoretical_loss": 3.5033928936698904,
"objective/train/tokens_used": 1583493600,
"theoretical_loss": 3.5033928936698904,
"tokens_seen": 1563033600
},
{
"epoch": 0.47,
"learning_rate": 0.0002659077231695085,
"loss": 2.7394,
"theoretical_loss": 3.5033928936698904,
"tokens_seen": 1563033600
},
{
"epoch": 0.47,
"learning_rate": 0.00026589769307923774,
"loss": 2.7732,
"theoretical_loss": 3.5033800446036185,
"tokens_seen": 1563099136
},
{
"epoch": 0.47,
"learning_rate": 0.00026588766298896686,
"loss": 2.5597,
"theoretical_loss": 3.5033671962268924,
"tokens_seen": 1563164672
},
{
"epoch": 0.47,
"learning_rate": 0.0002658776328986961,
"loss": 2.8412,
"theoretical_loss": 3.5033543485396472,
"tokens_seen": 1563230208
},
{
"epoch": 0.47,
"learning_rate": 0.0002658676028084253,
"loss": 2.6082,
"theoretical_loss": 3.503341501541816,
"tokens_seen": 1563295744
},
{
"epoch": 0.47,
"learning_rate": 0.00026585757271815446,
"loss": 2.6756,
"theoretical_loss": 3.5033286552333327,
"tokens_seen": 1563361280
},
{
"epoch": 0.47,
"learning_rate": 0.00026584754262788364,
"loss": 2.4135,
"theoretical_loss": 3.503315809614132,
"tokens_seen": 1563426816
},
{
"epoch": 0.47,
"learning_rate": 0.0002658375125376128,
"loss": 2.7178,
"theoretical_loss": 3.5033029646841483,
"tokens_seen": 1563492352
},
{
"epoch": 0.47,
"learning_rate": 0.000265827482447342,
"loss": 2.5815,
"theoretical_loss": 3.5032901204433156,
"tokens_seen": 1563557888
},
{
"epoch": 0.47,
"learning_rate": 0.00026581745235707124,
"loss": 2.5894,
"theoretical_loss": 3.503277276891567,
"tokens_seen": 1563623424
},
{
"epoch": 0.47,
"learning_rate": 0.00026580742226680037,
"loss": 2.4998,
"theoretical_loss": 3.5032644340288375,
"tokens_seen": 1563688960
},
{
"epoch": 0.47,
"learning_rate": 0.0002657973921765296,
"loss": 2.7567,
"theoretical_loss": 3.5032515918550615,
"tokens_seen": 1563754496
},
{
"epoch": 0.47,
"learning_rate": 0.00026578736208625873,
"loss": 2.9132,
"theoretical_loss": 3.503238750370173,
"tokens_seen": 1563820032
},
{
"epoch": 0.47,
"learning_rate": 0.00026577733199598797,
"loss": 2.7283,
"theoretical_loss": 3.503225909574106,
"tokens_seen": 1563885568
},
{
"epoch": 0.47,
"learning_rate": 0.00026576730190571715,
"loss": 2.7477,
"theoretical_loss": 3.5032130694667947,
"tokens_seen": 1563951104
},
{
"epoch": 0.47,
"learning_rate": 0.00026575727181544633,
"loss": 2.6688,
"theoretical_loss": 3.503200230048173,
"tokens_seen": 1564016640
},
{
"epoch": 0.47,
"learning_rate": 0.00026574724172517556,
"loss": 2.7339,
"theoretical_loss": 3.5031873913181757,
"tokens_seen": 1564082176
},
{
"epoch": 0.47,
"learning_rate": 0.00026573721163490474,
"loss": 2.6203,
"theoretical_loss": 3.503174553276737,
"tokens_seen": 1564147712
},
{
"epoch": 0.47,
"learning_rate": 0.0002657271815446339,
"loss": 2.6299,
"theoretical_loss": 3.5031617159237904,
"tokens_seen": 1564213248
},
{
"epoch": 0.47,
"learning_rate": 0.0002657171514543631,
"loss": 2.7354,
"theoretical_loss": 3.503148879259271,
"tokens_seen": 1564278784
},
{
"epoch": 0.47,
"learning_rate": 0.0002657071213640923,
"loss": 2.8946,
"theoretical_loss": 3.5031360432831127,
"tokens_seen": 1564344320
},
{
"epoch": 0.47,
"learning_rate": 0.00026569709127382147,
"loss": 2.6629,
"theoretical_loss": 3.5031232079952495,
"tokens_seen": 1564409856
},
{
"epoch": 0.47,
"learning_rate": 0.0002656870611835507,
"loss": 2.5688,
"theoretical_loss": 3.5031103733956157,
"tokens_seen": 1564475392
},
{
"epoch": 0.47,
"learning_rate": 0.00026567703109327983,
"loss": 2.7349,
"theoretical_loss": 3.5030975394841457,
"tokens_seen": 1564540928
},
{
"epoch": 0.47,
"learning_rate": 0.00026566700100300907,
"loss": 2.6643,
"theoretical_loss": 3.5030847062607746,
"tokens_seen": 1564606464
},
{
"epoch": 0.47,
"objective/train/docs_used": 887471,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2938284873962402,
"objective/train/theoretical_loss": 3.5030718737254354,
"objective/train/tokens_used": 1585132000,
"theoretical_loss": 3.5030718737254354,
"tokens_seen": 1564672000
},
{
"epoch": 0.47,
"learning_rate": 0.0002656569709127382,
"loss": 2.5791,
"theoretical_loss": 3.5030718737254354,
"tokens_seen": 1564672000
},
{
"epoch": 0.47,
"learning_rate": 0.00026564694082246743,
"loss": 2.6372,
"theoretical_loss": 3.5030590418780623,
"tokens_seen": 1564737536
},
{
"epoch": 0.47,
"learning_rate": 0.0002656369107321966,
"loss": 2.5939,
"theoretical_loss": 3.503046210718591,
"tokens_seen": 1564803072
},
{
"epoch": 0.47,
"learning_rate": 0.0002656268806419258,
"loss": 2.8375,
"theoretical_loss": 3.503033380246955,
"tokens_seen": 1564868608
},
{
"epoch": 0.47,
"learning_rate": 0.00026561685055165497,
"loss": 3.0359,
"theoretical_loss": 3.5030205504630882,
"tokens_seen": 1564934144
},
{
"epoch": 0.47,
"learning_rate": 0.00026560682046138415,
"loss": 2.4589,
"theoretical_loss": 3.5030077213669255,
"tokens_seen": 1564999680
},
{
"epoch": 0.47,
"learning_rate": 0.00026559679037111333,
"loss": 2.5053,
"theoretical_loss": 3.5029948929584016,
"tokens_seen": 1565065216
},
{
"epoch": 0.47,
"learning_rate": 0.00026558676028084257,
"loss": 2.6661,
"theoretical_loss": 3.5029820652374495,
"tokens_seen": 1565130752
},
{
"epoch": 0.47,
"learning_rate": 0.0002655767301905717,
"loss": 2.7472,
"theoretical_loss": 3.502969238204005,
"tokens_seen": 1565196288
},
{
"epoch": 0.47,
"learning_rate": 0.00026556670010030093,
"loss": 2.7089,
"theoretical_loss": 3.502956411858002,
"tokens_seen": 1565261824
},
{
"epoch": 0.47,
"learning_rate": 0.0002655566700100301,
"loss": 2.6483,
"theoretical_loss": 3.5029435861993745,
"tokens_seen": 1565327360
},
{
"epoch": 0.47,
"learning_rate": 0.0002655466399197593,
"loss": 2.7748,
"theoretical_loss": 3.502930761228057,
"tokens_seen": 1565392896
},
{
"epoch": 0.47,
"learning_rate": 0.0002655366098294885,
"loss": 2.9215,
"theoretical_loss": 3.5029179369439847,
"tokens_seen": 1565458432
},
{
"epoch": 0.47,
"learning_rate": 0.00026552657973921766,
"loss": 2.757,
"theoretical_loss": 3.5029051133470905,
"tokens_seen": 1565523968
},
{
"epoch": 0.47,
"learning_rate": 0.00026551654964894684,
"loss": 2.7415,
"theoretical_loss": 3.5028922904373103,
"tokens_seen": 1565589504
},
{
"epoch": 0.47,
"learning_rate": 0.00026550651955867607,
"loss": 2.6654,
"theoretical_loss": 3.5028794682145774,
"tokens_seen": 1565655040
},
{
"epoch": 0.47,
"learning_rate": 0.0002654964894684052,
"loss": 2.6742,
"theoretical_loss": 3.5028666466788274,
"tokens_seen": 1565720576
},
{
"epoch": 0.47,
"learning_rate": 0.00026548645937813443,
"loss": 2.6454,
"theoretical_loss": 3.5028538258299937,
"tokens_seen": 1565786112
},
{
"epoch": 0.47,
"learning_rate": 0.00026547642928786356,
"loss": 2.5541,
"theoretical_loss": 3.502841005668011,
"tokens_seen": 1565851648
},
{
"epoch": 0.47,
"learning_rate": 0.0002654663991975928,
"loss": 2.6732,
"theoretical_loss": 3.502828186192814,
"tokens_seen": 1565917184
},
{
"epoch": 0.47,
"learning_rate": 0.000265456369107322,
"loss": 2.7742,
"theoretical_loss": 3.502815367404337,
"tokens_seen": 1565982720
},
{
"epoch": 0.47,
"learning_rate": 0.00026544633901705116,
"loss": 2.8134,
"theoretical_loss": 3.5028025493025146,
"tokens_seen": 1566048256
},
{
"epoch": 0.47,
"learning_rate": 0.00026543630892678034,
"loss": 2.6239,
"theoretical_loss": 3.5027897318872814,
"tokens_seen": 1566113792
},
{
"epoch": 0.47,
"learning_rate": 0.0002654262788365096,
"loss": 2.6747,
"theoretical_loss": 3.502776915158571,
"tokens_seen": 1566179328
},
{
"epoch": 0.47,
"learning_rate": 0.0002654162487462387,
"loss": 2.5136,
"theoretical_loss": 3.5027640991163196,
"tokens_seen": 1566244864
},
{
"epoch": 0.47,
"objective/train/docs_used": 888126,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2683451175689697,
"objective/train/theoretical_loss": 3.50275128376046,
"objective/train/tokens_used": 1586770400,
"theoretical_loss": 3.50275128376046,
"tokens_seen": 1566310400
},
{
"epoch": 0.47,
"learning_rate": 0.00026540621865596794,
"loss": 2.614,
"theoretical_loss": 3.50275128376046,
"tokens_seen": 1566310400
},
{
"epoch": 0.47,
"learning_rate": 0.00026539618856569707,
"loss": 2.7061,
"theoretical_loss": 3.502738469090928,
"tokens_seen": 1566375936
},
{
"epoch": 0.47,
"learning_rate": 0.0002653861584754263,
"loss": 2.7256,
"theoretical_loss": 3.502725655107657,
"tokens_seen": 1566441472
},
{
"epoch": 0.47,
"learning_rate": 0.0002653761283851555,
"loss": 2.7901,
"theoretical_loss": 3.502712841810582,
"tokens_seen": 1566507008
},
{
"epoch": 0.47,
"learning_rate": 0.00026536609829488466,
"loss": 2.7311,
"theoretical_loss": 3.502700029199638,
"tokens_seen": 1566572544
},
{
"epoch": 0.47,
"learning_rate": 0.00026535606820461384,
"loss": 2.8912,
"theoretical_loss": 3.5026872172747594,
"tokens_seen": 1566638080
},
{
"epoch": 0.47,
"learning_rate": 0.000265346038114343,
"loss": 2.6776,
"theoretical_loss": 3.5026744060358803,
"tokens_seen": 1566703616
},
{
"epoch": 0.47,
"learning_rate": 0.0002653360080240722,
"loss": 2.8627,
"theoretical_loss": 3.5026615954829357,
"tokens_seen": 1566769152
},
{
"epoch": 0.47,
"learning_rate": 0.00026532597793380144,
"loss": 2.8377,
"theoretical_loss": 3.50264878561586,
"tokens_seen": 1566834688
},
{
"epoch": 0.47,
"learning_rate": 0.00026531594784353057,
"loss": 2.7911,
"theoretical_loss": 3.502635976434588,
"tokens_seen": 1566900224
},
{
"epoch": 0.47,
"learning_rate": 0.0002653059177532598,
"loss": 2.4296,
"theoretical_loss": 3.5026231679390536,
"tokens_seen": 1566965760
},
{
"epoch": 0.47,
"learning_rate": 0.00026529588766298893,
"loss": 2.99,
"theoretical_loss": 3.5026103601291925,
"tokens_seen": 1567031296
},
{
"epoch": 0.47,
"learning_rate": 0.00026528585757271817,
"loss": 2.7091,
"theoretical_loss": 3.5025975530049385,
"tokens_seen": 1567096832
},
{
"epoch": 0.47,
"learning_rate": 0.00026527582748244735,
"loss": 2.9422,
"theoretical_loss": 3.5025847465662268,
"tokens_seen": 1567162368
},
{
"epoch": 0.47,
"learning_rate": 0.00026526579739217653,
"loss": 2.6707,
"theoretical_loss": 3.5025719408129916,
"tokens_seen": 1567227904
},
{
"epoch": 0.47,
"learning_rate": 0.0002652557673019057,
"loss": 2.6783,
"theoretical_loss": 3.5025591357451673,
"tokens_seen": 1567293440
},
{
"epoch": 0.47,
"learning_rate": 0.00026524573721163494,
"loss": 2.8317,
"theoretical_loss": 3.5025463313626894,
"tokens_seen": 1567358976
},
{
"epoch": 0.47,
"learning_rate": 0.00026523570712136407,
"loss": 2.6124,
"theoretical_loss": 3.502533527665492,
"tokens_seen": 1567424512
},
{
"epoch": 0.47,
"learning_rate": 0.0002652256770310933,
"loss": 2.6238,
"theoretical_loss": 3.5025207246535097,
"tokens_seen": 1567490048
},
{
"epoch": 0.48,
"learning_rate": 0.00026521564694082243,
"loss": 2.8489,
"theoretical_loss": 3.5025079223266777,
"tokens_seen": 1567555584
},
{
"epoch": 0.48,
"learning_rate": 0.00026520561685055167,
"loss": 2.7145,
"theoretical_loss": 3.50249512068493,
"tokens_seen": 1567621120
},
{
"epoch": 0.48,
"learning_rate": 0.00026519558676028085,
"loss": 2.3955,
"theoretical_loss": 3.502482319728202,
"tokens_seen": 1567686656
},
{
"epoch": 0.48,
"learning_rate": 0.00026518555667001003,
"loss": 2.7588,
"theoretical_loss": 3.5024695194564277,
"tokens_seen": 1567752192
},
{
"epoch": 0.48,
"learning_rate": 0.0002651755265797392,
"loss": 2.6835,
"theoretical_loss": 3.502456719869542,
"tokens_seen": 1567817728
},
{
"epoch": 0.48,
"learning_rate": 0.0002651654964894684,
"loss": 2.7944,
"theoretical_loss": 3.5024439209674805,
"tokens_seen": 1567883264
},
{
"epoch": 0.48,
"objective/train/docs_used": 889384,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0547804832458496,
"objective/train/theoretical_loss": 3.5024311227501768,
"objective/train/tokens_used": 1588408800,
"theoretical_loss": 3.5024311227501768,
"tokens_seen": 1567948800
},
{
"epoch": 0.48,
"learning_rate": 0.0002651554663991976,
"loss": 2.894,
"theoretical_loss": 3.5024311227501768,
"tokens_seen": 1567948800
},
{
"epoch": 0.48,
"learning_rate": 0.0002651454363089268,
"loss": 2.6705,
"theoretical_loss": 3.5024183252175662,
"tokens_seen": 1568014336
},
{
"epoch": 0.48,
"learning_rate": 0.00026513540621865594,
"loss": 2.5851,
"theoretical_loss": 3.502405528369583,
"tokens_seen": 1568079872
},
{
"epoch": 0.48,
"learning_rate": 0.00026512537612838517,
"loss": 2.7046,
"theoretical_loss": 3.5023927322061637,
"tokens_seen": 1568145408
},
{
"epoch": 0.48,
"learning_rate": 0.0002651153460381143,
"loss": 2.6028,
"theoretical_loss": 3.50237993672724,
"tokens_seen": 1568210944
},
{
"epoch": 0.48,
"learning_rate": 0.00026510531594784353,
"loss": 2.6756,
"theoretical_loss": 3.5023671419327496,
"tokens_seen": 1568276480
},
{
"epoch": 0.48,
"learning_rate": 0.0002650952858575727,
"loss": 2.7097,
"theoretical_loss": 3.5023543478226253,
"tokens_seen": 1568342016
},
{
"epoch": 0.48,
"learning_rate": 0.0002650852557673019,
"loss": 2.5519,
"theoretical_loss": 3.5023415543968026,
"tokens_seen": 1568407552
},
{
"epoch": 0.48,
"learning_rate": 0.0002650752256770311,
"loss": 2.6901,
"theoretical_loss": 3.5023287616552166,
"tokens_seen": 1568473088
},
{
"epoch": 0.48,
"learning_rate": 0.0002650651955867603,
"loss": 2.7571,
"theoretical_loss": 3.5023159695978023,
"tokens_seen": 1568538624
},
{
"epoch": 0.48,
"learning_rate": 0.00026505516549648944,
"loss": 2.7922,
"theoretical_loss": 3.5023031782244933,
"tokens_seen": 1568604160
},
{
"epoch": 0.48,
"learning_rate": 0.0002650451354062187,
"loss": 2.6757,
"theoretical_loss": 3.502290387535226,
"tokens_seen": 1568669696
},
{
"epoch": 0.48,
"learning_rate": 0.0002650351053159478,
"loss": 2.5746,
"theoretical_loss": 3.5022775975299343,
"tokens_seen": 1568735232
},
{
"epoch": 0.48,
"learning_rate": 0.00026502507522567704,
"loss": 2.7272,
"theoretical_loss": 3.502264808208553,
"tokens_seen": 1568800768
},
{
"epoch": 0.48,
"learning_rate": 0.0002650150451354062,
"loss": 2.6675,
"theoretical_loss": 3.502252019571017,
"tokens_seen": 1568866304
},
{
"epoch": 0.48,
"learning_rate": 0.0002650050150451354,
"loss": 2.6134,
"theoretical_loss": 3.5022392316172617,
"tokens_seen": 1568931840
},
{
"epoch": 0.48,
"learning_rate": 0.00026499498495486464,
"loss": 2.6062,
"theoretical_loss": 3.502226444347222,
"tokens_seen": 1568997376
},
{
"epoch": 0.48,
"learning_rate": 0.00026498495486459376,
"loss": 2.4137,
"theoretical_loss": 3.502213657760832,
"tokens_seen": 1569062912
},
{
"epoch": 0.48,
"learning_rate": 0.000264974924774323,
"loss": 2.5753,
"theoretical_loss": 3.502200871858027,
"tokens_seen": 1569128448
},
{
"epoch": 0.48,
"learning_rate": 0.0002649648946840522,
"loss": 2.7103,
"theoretical_loss": 3.5021880866387423,
"tokens_seen": 1569193984
},
{
"epoch": 0.48,
"learning_rate": 0.00026495486459378136,
"loss": 2.8113,
"theoretical_loss": 3.5021753021029123,
"tokens_seen": 1569259520
},
{
"epoch": 0.48,
"learning_rate": 0.00026494483450351054,
"loss": 2.7555,
"theoretical_loss": 3.502162518250472,
"tokens_seen": 1569325056
},
{
"epoch": 0.48,
"learning_rate": 0.0002649348044132398,
"loss": 2.6035,
"theoretical_loss": 3.5021497350813564,
"tokens_seen": 1569390592
},
{
"epoch": 0.48,
"learning_rate": 0.0002649247743229689,
"loss": 2.7298,
"theoretical_loss": 3.5021369525955004,
"tokens_seen": 1569456128
},
{
"epoch": 0.48,
"learning_rate": 0.00026491474423269814,
"loss": 2.6887,
"theoretical_loss": 3.5021241707928397,
"tokens_seen": 1569521664
},
{
"epoch": 0.48,
"objective/train/docs_used": 890057,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.971144676208496,
"objective/train/theoretical_loss": 3.5021113896733076,
"objective/train/tokens_used": 1590047200,
"theoretical_loss": 3.5021113896733076,
"tokens_seen": 1569587200
},
{
"epoch": 0.48,
"learning_rate": 0.00026490471414242727,
"loss": 2.8338,
"theoretical_loss": 3.5021113896733076,
"tokens_seen": 1569587200
},
{
"epoch": 0.48,
"learning_rate": 0.0002648946840521565,
"loss": 2.7303,
"theoretical_loss": 3.5020986092368407,
"tokens_seen": 1569652736
},
{
"epoch": 0.48,
"learning_rate": 0.0002648846539618857,
"loss": 2.712,
"theoretical_loss": 3.5020858294833728,
"tokens_seen": 1569718272
},
{
"epoch": 0.48,
"learning_rate": 0.00026487462387161486,
"loss": 2.5344,
"theoretical_loss": 3.5020730504128395,
"tokens_seen": 1569783808
},
{
"epoch": 0.48,
"learning_rate": 0.00026486459378134404,
"loss": 2.7608,
"theoretical_loss": 3.5020602720251763,
"tokens_seen": 1569849344
},
{
"epoch": 0.48,
"learning_rate": 0.0002648545636910732,
"loss": 2.4189,
"theoretical_loss": 3.5020474943203173,
"tokens_seen": 1569914880
},
{
"epoch": 0.48,
"learning_rate": 0.0002648445336008024,
"loss": 2.7974,
"theoretical_loss": 3.502034717298198,
"tokens_seen": 1569980416
},
{
"epoch": 0.48,
"learning_rate": 0.00026483450351053164,
"loss": 2.9399,
"theoretical_loss": 3.5020219409587527,
"tokens_seen": 1570045952
},
{
"epoch": 0.48,
"learning_rate": 0.00026482447342026077,
"loss": 2.6916,
"theoretical_loss": 3.502009165301917,
"tokens_seen": 1570111488
},
{
"epoch": 0.48,
"learning_rate": 0.00026481444332999,
"loss": 2.7551,
"theoretical_loss": 3.5019963903276263,
"tokens_seen": 1570177024
},
{
"epoch": 0.48,
"learning_rate": 0.00026480441323971913,
"loss": 2.8331,
"theoretical_loss": 3.5019836160358153,
"tokens_seen": 1570242560
},
{
"epoch": 0.48,
"learning_rate": 0.00026479438314944837,
"loss": 2.7283,
"theoretical_loss": 3.5019708424264184,
"tokens_seen": 1570308096
},
{
"epoch": 0.48,
"learning_rate": 0.00026478435305917755,
"loss": 2.4873,
"theoretical_loss": 3.501958069499372,
"tokens_seen": 1570373632
},
{
"epoch": 0.48,
"learning_rate": 0.00026477432296890673,
"loss": 2.7615,
"theoretical_loss": 3.5019452972546103,
"tokens_seen": 1570439168
},
{
"epoch": 0.48,
"learning_rate": 0.0002647642928786359,
"loss": 2.8048,
"theoretical_loss": 3.5019325256920686,
"tokens_seen": 1570504704
},
{
"epoch": 0.48,
"learning_rate": 0.00026475426278836514,
"loss": 2.6309,
"theoretical_loss": 3.5019197548116816,
"tokens_seen": 1570570240
},
{
"epoch": 0.48,
"learning_rate": 0.00026474423269809427,
"loss": 2.8025,
"theoretical_loss": 3.501906984613385,
"tokens_seen": 1570635776
},
{
"epoch": 0.48,
"learning_rate": 0.0002647342026078235,
"loss": 2.5749,
"theoretical_loss": 3.5018942150971135,
"tokens_seen": 1570701312
},
{
"epoch": 0.48,
"learning_rate": 0.00026472417251755263,
"loss": 2.7553,
"theoretical_loss": 3.5018814462628027,
"tokens_seen": 1570766848
},
{
"epoch": 0.48,
"learning_rate": 0.00026471414242728187,
"loss": 2.3976,
"theoretical_loss": 3.5018686781103874,
"tokens_seen": 1570832384
},
{
"epoch": 0.48,
"learning_rate": 0.00026470411233701105,
"loss": 2.68,
"theoretical_loss": 3.5018559106398026,
"tokens_seen": 1570897920
},
{
"epoch": 0.48,
"learning_rate": 0.00026469408224674023,
"loss": 2.7,
"theoretical_loss": 3.501843143850983,
"tokens_seen": 1570963456
},
{
"epoch": 0.48,
"learning_rate": 0.0002646840521564694,
"loss": 2.7711,
"theoretical_loss": 3.5018303777438646,
"tokens_seen": 1571028992
},
{
"epoch": 0.48,
"learning_rate": 0.0002646740220661986,
"loss": 2.6569,
"theoretical_loss": 3.501817612318383,
"tokens_seen": 1571094528
},
{
"epoch": 0.48,
"learning_rate": 0.0002646639919759278,
"loss": 2.592,
"theoretical_loss": 3.5018048475744723,
"tokens_seen": 1571160064
},
{
"epoch": 0.48,
"objective/train/docs_used": 891281,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8096773624420166,
"objective/train/theoretical_loss": 3.501792083512068,
"objective/train/tokens_used": 1591685600,
"theoretical_loss": 3.501792083512068,
"tokens_seen": 1571225600
},
{
"epoch": 0.48,
"learning_rate": 0.000264653961885657,
"loss": 2.8892,
"theoretical_loss": 3.501792083512068,
"tokens_seen": 1571225600
},
{
"epoch": 0.48,
"learning_rate": 0.00026464393179538614,
"loss": 2.9393,
"theoretical_loss": 3.5017793201311056,
"tokens_seen": 1571291136
},
{
"epoch": 0.48,
"learning_rate": 0.00026463390170511537,
"loss": 2.6385,
"theoretical_loss": 3.5017665574315195,
"tokens_seen": 1571356672
},
{
"epoch": 0.48,
"learning_rate": 0.0002646238716148445,
"loss": 2.4161,
"theoretical_loss": 3.5017537954132463,
"tokens_seen": 1571422208
},
{
"epoch": 0.48,
"learning_rate": 0.00026461384152457373,
"loss": 2.6058,
"theoretical_loss": 3.5017410340762196,
"tokens_seen": 1571487744
},
{
"epoch": 0.48,
"learning_rate": 0.0002646038114343029,
"loss": 2.7054,
"theoretical_loss": 3.501728273420376,
"tokens_seen": 1571553280
},
{
"epoch": 0.48,
"learning_rate": 0.0002645937813440321,
"loss": 2.6353,
"theoretical_loss": 3.5017155134456495,
"tokens_seen": 1571618816
},
{
"epoch": 0.48,
"learning_rate": 0.0002645837512537613,
"loss": 2.7497,
"theoretical_loss": 3.5017027541519763,
"tokens_seen": 1571684352
},
{
"epoch": 0.48,
"learning_rate": 0.0002645737211634905,
"loss": 2.7459,
"theoretical_loss": 3.5016899955392917,
"tokens_seen": 1571749888
},
{
"epoch": 0.48,
"learning_rate": 0.00026456369107321964,
"loss": 2.858,
"theoretical_loss": 3.50167723760753,
"tokens_seen": 1571815424
},
{
"epoch": 0.48,
"learning_rate": 0.0002645536609829489,
"loss": 2.8354,
"theoretical_loss": 3.5016644803566277,
"tokens_seen": 1571880960
},
{
"epoch": 0.48,
"learning_rate": 0.000264543630892678,
"loss": 2.7065,
"theoretical_loss": 3.501651723786519,
"tokens_seen": 1571946496
},
{
"epoch": 0.48,
"learning_rate": 0.00026453360080240724,
"loss": 2.8862,
"theoretical_loss": 3.50163896789714,
"tokens_seen": 1572012032
},
{
"epoch": 0.48,
"learning_rate": 0.0002645235707121364,
"loss": 2.6915,
"theoretical_loss": 3.5016262126884254,
"tokens_seen": 1572077568
},
{
"epoch": 0.48,
"learning_rate": 0.0002645135406218656,
"loss": 2.7223,
"theoretical_loss": 3.5016134581603104,
"tokens_seen": 1572143104
},
{
"epoch": 0.48,
"learning_rate": 0.0002645035105315948,
"loss": 2.7089,
"theoretical_loss": 3.5016007043127315,
"tokens_seen": 1572208640
},
{
"epoch": 0.48,
"learning_rate": 0.00026449348044132396,
"loss": 2.7313,
"theoretical_loss": 3.5015879511456225,
"tokens_seen": 1572274176
},
{
"epoch": 0.48,
"learning_rate": 0.00026448345035105314,
"loss": 2.6133,
"theoretical_loss": 3.50157519865892,
"tokens_seen": 1572339712
},
{
"epoch": 0.48,
"learning_rate": 0.0002644734202607824,
"loss": 2.5419,
"theoretical_loss": 3.5015624468525584,
"tokens_seen": 1572405248
},
{
"epoch": 0.48,
"learning_rate": 0.0002644633901705115,
"loss": 2.8509,
"theoretical_loss": 3.5015496957264736,
"tokens_seen": 1572470784
},
{
"epoch": 0.48,
"learning_rate": 0.00026445336008024074,
"loss": 2.6063,
"theoretical_loss": 3.5015369452806,
"tokens_seen": 1572536320
},
{
"epoch": 0.48,
"learning_rate": 0.00026444332998996987,
"loss": 2.7732,
"theoretical_loss": 3.5015241955148744,
"tokens_seen": 1572601856
},
{
"epoch": 0.48,
"learning_rate": 0.0002644332998996991,
"loss": 2.5804,
"theoretical_loss": 3.501511446429232,
"tokens_seen": 1572667392
},
{
"epoch": 0.48,
"learning_rate": 0.0002644232698094283,
"loss": 2.6067,
"theoretical_loss": 3.5014986980236067,
"tokens_seen": 1572732928
},
{
"epoch": 0.48,
"learning_rate": 0.00026441323971915747,
"loss": 2.5086,
"theoretical_loss": 3.5014859502979356,
"tokens_seen": 1572798464
},
{
"epoch": 0.48,
"objective/train/docs_used": 891900,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.760402202606201,
"objective/train/theoretical_loss": 3.501473203252153,
"objective/train/tokens_used": 1593324000,
"theoretical_loss": 3.501473203252153,
"tokens_seen": 1572864000
},
{
"epoch": 0.48,
"learning_rate": 0.00026440320962888665,
"loss": 2.901,
"theoretical_loss": 3.501473203252153,
"tokens_seen": 1572864000
},
{
"epoch": 0.48,
"learning_rate": 0.0002643931795386159,
"loss": 2.7692,
"theoretical_loss": 3.501460456886195,
"tokens_seen": 1572929536
},
{
"epoch": 0.48,
"learning_rate": 0.000264383149448345,
"loss": 2.5585,
"theoretical_loss": 3.5014477111999964,
"tokens_seen": 1572995072
},
{
"epoch": 0.48,
"learning_rate": 0.00026437311935807424,
"loss": 2.8031,
"theoretical_loss": 3.5014349661934934,
"tokens_seen": 1573060608
},
{
"epoch": 0.48,
"learning_rate": 0.00026436308926780337,
"loss": 2.7396,
"theoretical_loss": 3.501422221866621,
"tokens_seen": 1573126144
},
{
"epoch": 0.48,
"learning_rate": 0.0002643530591775326,
"loss": 2.85,
"theoretical_loss": 3.5014094782193146,
"tokens_seen": 1573191680
},
{
"epoch": 0.48,
"learning_rate": 0.0002643430290872618,
"loss": 2.5394,
"theoretical_loss": 3.501396735251509,
"tokens_seen": 1573257216
},
{
"epoch": 0.48,
"learning_rate": 0.00026433299899699097,
"loss": 2.4725,
"theoretical_loss": 3.501383992963141,
"tokens_seen": 1573322752
},
{
"epoch": 0.48,
"learning_rate": 0.00026432296890672015,
"loss": 3.0966,
"theoretical_loss": 3.5013712513541453,
"tokens_seen": 1573388288
},
{
"epoch": 0.48,
"learning_rate": 0.00026431293881644933,
"loss": 2.7001,
"theoretical_loss": 3.501358510424458,
"tokens_seen": 1573453824
},
{
"epoch": 0.48,
"learning_rate": 0.0002643029087261785,
"loss": 2.8315,
"theoretical_loss": 3.501345770174013,
"tokens_seen": 1573519360
},
{
"epoch": 0.48,
"learning_rate": 0.00026429287863590775,
"loss": 2.7426,
"theoretical_loss": 3.501333030602748,
"tokens_seen": 1573584896
},
{
"epoch": 0.48,
"learning_rate": 0.0002642828485456369,
"loss": 2.739,
"theoretical_loss": 3.501320291710597,
"tokens_seen": 1573650432
},
{
"epoch": 0.48,
"learning_rate": 0.0002642728184553661,
"loss": 2.808,
"theoretical_loss": 3.501307553497496,
"tokens_seen": 1573715968
},
{
"epoch": 0.48,
"learning_rate": 0.00026426278836509534,
"loss": 2.8547,
"theoretical_loss": 3.501294815963381,
"tokens_seen": 1573781504
},
{
"epoch": 0.48,
"learning_rate": 0.00026425275827482447,
"loss": 2.6464,
"theoretical_loss": 3.5012820791081865,
"tokens_seen": 1573847040
},
{
"epoch": 0.48,
"learning_rate": 0.0002642427281845537,
"loss": 2.7497,
"theoretical_loss": 3.501269342931848,
"tokens_seen": 1573912576
},
{
"epoch": 0.48,
"learning_rate": 0.00026423269809428283,
"loss": 2.5446,
"theoretical_loss": 3.5012566074343026,
"tokens_seen": 1573978112
},
{
"epoch": 0.48,
"learning_rate": 0.00026422266800401207,
"loss": 2.6657,
"theoretical_loss": 3.5012438726154844,
"tokens_seen": 1574043648
},
{
"epoch": 0.48,
"learning_rate": 0.00026421263791374125,
"loss": 2.746,
"theoretical_loss": 3.501231138475329,
"tokens_seen": 1574109184
},
{
"epoch": 0.48,
"learning_rate": 0.00026420260782347043,
"loss": 2.7118,
"theoretical_loss": 3.5012184050137733,
"tokens_seen": 1574174720
},
{
"epoch": 0.48,
"learning_rate": 0.0002641925777331996,
"loss": 2.4394,
"theoretical_loss": 3.5012056722307516,
"tokens_seen": 1574240256
},
{
"epoch": 0.48,
"learning_rate": 0.0002641825476429288,
"loss": 2.7501,
"theoretical_loss": 3.5011929401261996,
"tokens_seen": 1574305792
},
{
"epoch": 0.48,
"learning_rate": 0.000264172517552658,
"loss": 2.7063,
"theoretical_loss": 3.5011802087000534,
"tokens_seen": 1574371328
},
{
"epoch": 0.48,
"learning_rate": 0.0002641624874623872,
"loss": 2.6933,
"theoretical_loss": 3.501167477952249,
"tokens_seen": 1574436864
},
{
"epoch": 0.48,
"objective/train/docs_used": 893256,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6786599159240723,
"objective/train/theoretical_loss": 3.5011547478827207,
"objective/train/tokens_used": 1594962400,
"theoretical_loss": 3.5011547478827207,
"tokens_seen": 1574502400
},
{
"epoch": 0.48,
"learning_rate": 0.00026415245737211634,
"loss": 2.6984,
"theoretical_loss": 3.5011547478827207,
"tokens_seen": 1574502400
},
{
"epoch": 0.48,
"learning_rate": 0.00026414242728184557,
"loss": 2.8852,
"theoretical_loss": 3.5011420184914055,
"tokens_seen": 1574567936
},
{
"epoch": 0.48,
"learning_rate": 0.0002641323971915747,
"loss": 2.485,
"theoretical_loss": 3.501129289778238,
"tokens_seen": 1574633472
},
{
"epoch": 0.48,
"learning_rate": 0.00026412236710130393,
"loss": 2.9047,
"theoretical_loss": 3.5011165617431543,
"tokens_seen": 1574699008
},
{
"epoch": 0.48,
"learning_rate": 0.0002641123370110331,
"loss": 2.769,
"theoretical_loss": 3.50110383438609,
"tokens_seen": 1574764544
},
{
"epoch": 0.48,
"learning_rate": 0.0002641023069207623,
"loss": 2.5368,
"theoretical_loss": 3.501091107706981,
"tokens_seen": 1574830080
},
{
"epoch": 0.48,
"learning_rate": 0.0002640922768304915,
"loss": 2.9257,
"theoretical_loss": 3.5010783817057627,
"tokens_seen": 1574895616
},
{
"epoch": 0.48,
"learning_rate": 0.0002640822467402207,
"loss": 2.5107,
"theoretical_loss": 3.501065656382371,
"tokens_seen": 1574961152
},
{
"epoch": 0.48,
"learning_rate": 0.00026407221664994984,
"loss": 2.5318,
"theoretical_loss": 3.5010529317367416,
"tokens_seen": 1575026688
},
{
"epoch": 0.48,
"learning_rate": 0.0002640621865596791,
"loss": 2.8075,
"theoretical_loss": 3.5010402077688094,
"tokens_seen": 1575092224
},
{
"epoch": 0.48,
"learning_rate": 0.0002640521564694082,
"loss": 2.824,
"theoretical_loss": 3.501027484478511,
"tokens_seen": 1575157760
},
{
"epoch": 0.48,
"learning_rate": 0.00026404212637913744,
"loss": 2.8091,
"theoretical_loss": 3.5010147618657825,
"tokens_seen": 1575223296
},
{
"epoch": 0.48,
"learning_rate": 0.0002640320962888666,
"loss": 2.6394,
"theoretical_loss": 3.501002039930558,
"tokens_seen": 1575288832
},
{
"epoch": 0.48,
"learning_rate": 0.0002640220661985958,
"loss": 2.7055,
"theoretical_loss": 3.500989318672775,
"tokens_seen": 1575354368
},
{
"epoch": 0.48,
"learning_rate": 0.000264012036108325,
"loss": 2.8721,
"theoretical_loss": 3.5009765980923686,
"tokens_seen": 1575419904
},
{
"epoch": 0.48,
"learning_rate": 0.00026400200601805416,
"loss": 2.7784,
"theoretical_loss": 3.5009638781892742,
"tokens_seen": 1575485440
},
{
"epoch": 0.48,
"learning_rate": 0.00026399197592778334,
"loss": 2.7691,
"theoretical_loss": 3.500951158963428,
"tokens_seen": 1575550976
},
{
"epoch": 0.48,
"learning_rate": 0.0002639819458375126,
"loss": 2.7565,
"theoretical_loss": 3.500938440414765,
"tokens_seen": 1575616512
},
{
"epoch": 0.48,
"learning_rate": 0.0002639719157472417,
"loss": 2.6728,
"theoretical_loss": 3.500925722543222,
"tokens_seen": 1575682048
},
{
"epoch": 0.48,
"learning_rate": 0.00026396188565697094,
"loss": 2.7173,
"theoretical_loss": 3.5009130053487345,
"tokens_seen": 1575747584
},
{
"epoch": 0.48,
"learning_rate": 0.00026395185556670007,
"loss": 2.685,
"theoretical_loss": 3.500900288831238,
"tokens_seen": 1575813120
},
{
"epoch": 0.48,
"learning_rate": 0.0002639418254764293,
"loss": 2.6076,
"theoretical_loss": 3.5008875729906688,
"tokens_seen": 1575878656
},
{
"epoch": 0.48,
"learning_rate": 0.0002639317953861585,
"loss": 2.5983,
"theoretical_loss": 3.500874857826962,
"tokens_seen": 1575944192
},
{
"epoch": 0.48,
"learning_rate": 0.00026392176529588767,
"loss": 2.7206,
"theoretical_loss": 3.500862143340054,
"tokens_seen": 1576009728
},
{
"epoch": 0.48,
"learning_rate": 0.00026391173520561685,
"loss": 2.5689,
"theoretical_loss": 3.50084942952988,
"tokens_seen": 1576075264
},
{
"epoch": 0.48,
"objective/train/docs_used": 893728,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.615064859390259,
"objective/train/theoretical_loss": 3.500836716396377,
"objective/train/tokens_used": 1596600800,
"theoretical_loss": 3.500836716396377,
"tokens_seen": 1576140800
},
{
"epoch": 0.48,
"learning_rate": 0.0002639017051153461,
"loss": 2.7297,
"theoretical_loss": 3.500836716396377,
"tokens_seen": 1576140800
},
{
"epoch": 0.48,
"learning_rate": 0.0002638916750250752,
"loss": 2.6096,
"theoretical_loss": 3.5008240039394796,
"tokens_seen": 1576206336
},
{
"epoch": 0.48,
"learning_rate": 0.00026388164493480444,
"loss": 2.8551,
"theoretical_loss": 3.5008112921591246,
"tokens_seen": 1576271872
},
{
"epoch": 0.48,
"learning_rate": 0.00026387161484453357,
"loss": 2.5224,
"theoretical_loss": 3.5007985810552475,
"tokens_seen": 1576337408
},
{
"epoch": 0.48,
"learning_rate": 0.0002638615847542628,
"loss": 2.9652,
"theoretical_loss": 3.5007858706277837,
"tokens_seen": 1576402944
},
{
"epoch": 0.48,
"learning_rate": 0.000263851554663992,
"loss": 2.8408,
"theoretical_loss": 3.50077316087667,
"tokens_seen": 1576468480
},
{
"epoch": 0.48,
"learning_rate": 0.00026384152457372117,
"loss": 2.5138,
"theoretical_loss": 3.500760451801842,
"tokens_seen": 1576534016
},
{
"epoch": 0.48,
"learning_rate": 0.00026383149448345035,
"loss": 2.7838,
"theoretical_loss": 3.5007477434032346,
"tokens_seen": 1576599552
},
{
"epoch": 0.48,
"learning_rate": 0.00026382146439317953,
"loss": 2.7135,
"theoretical_loss": 3.500735035680785,
"tokens_seen": 1576665088
},
{
"epoch": 0.48,
"learning_rate": 0.0002638114343029087,
"loss": 2.8436,
"theoretical_loss": 3.5007223286344287,
"tokens_seen": 1576730624
},
{
"epoch": 0.48,
"learning_rate": 0.00026380140421263795,
"loss": 2.6764,
"theoretical_loss": 3.500709622264102,
"tokens_seen": 1576796160
},
{
"epoch": 0.48,
"learning_rate": 0.0002637913741223671,
"loss": 2.7213,
"theoretical_loss": 3.50069691656974,
"tokens_seen": 1576861696
},
{
"epoch": 0.48,
"learning_rate": 0.0002637813440320963,
"loss": 2.6682,
"theoretical_loss": 3.5006842115512793,
"tokens_seen": 1576927232
},
{
"epoch": 0.48,
"learning_rate": 0.0002637713139418255,
"loss": 2.8113,
"theoretical_loss": 3.500671507208655,
"tokens_seen": 1576992768
},
{
"epoch": 0.48,
"learning_rate": 0.00026376128385155467,
"loss": 2.8464,
"theoretical_loss": 3.5006588035418043,
"tokens_seen": 1577058304
},
{
"epoch": 0.48,
"learning_rate": 0.00026375125376128385,
"loss": 2.5902,
"theoretical_loss": 3.5006461005506626,
"tokens_seen": 1577123840
},
{
"epoch": 0.48,
"learning_rate": 0.00026374122367101303,
"loss": 2.869,
"theoretical_loss": 3.5006333982351663,
"tokens_seen": 1577189376
},
{
"epoch": 0.48,
"learning_rate": 0.0002637311935807422,
"loss": 2.6552,
"theoretical_loss": 3.5006206965952504,
"tokens_seen": 1577254912
},
{
"epoch": 0.48,
"learning_rate": 0.00026372116349047145,
"loss": 2.7506,
"theoretical_loss": 3.500607995630851,
"tokens_seen": 1577320448
},
{
"epoch": 0.48,
"learning_rate": 0.0002637111334002006,
"loss": 2.8517,
"theoretical_loss": 3.5005952953419053,
"tokens_seen": 1577385984
},
{
"epoch": 0.48,
"learning_rate": 0.0002637011033099298,
"loss": 2.6846,
"theoretical_loss": 3.5005825957283485,
"tokens_seen": 1577451520
},
{
"epoch": 0.48,
"learning_rate": 0.00026369107321965894,
"loss": 2.8689,
"theoretical_loss": 3.5005698967901164,
"tokens_seen": 1577517056
},
{
"epoch": 0.48,
"learning_rate": 0.0002636810431293882,
"loss": 2.6501,
"theoretical_loss": 3.5005571985271455,
"tokens_seen": 1577582592
},
{
"epoch": 0.48,
"learning_rate": 0.00026367101303911736,
"loss": 2.7346,
"theoretical_loss": 3.5005445009393723,
"tokens_seen": 1577648128
},
{
"epoch": 0.48,
"learning_rate": 0.00026366098294884654,
"loss": 2.7278,
"theoretical_loss": 3.5005318040267315,
"tokens_seen": 1577713664
},
{
"epoch": 0.48,
"objective/train/docs_used": 894313,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.656029462814331,
"objective/train/theoretical_loss": 3.50051910778916,
"objective/train/tokens_used": 1598239200,
"theoretical_loss": 3.50051910778916,
"tokens_seen": 1577779200
},
{
"epoch": 0.48,
"learning_rate": 0.0002636509528585757,
"loss": 2.6279,
"theoretical_loss": 3.50051910778916,
"tokens_seen": 1577779200
},
{
"epoch": 0.48,
"learning_rate": 0.0002636409227683049,
"loss": 2.5671,
"theoretical_loss": 3.500506412226594,
"tokens_seen": 1577844736
},
{
"epoch": 0.48,
"learning_rate": 0.0002636308926780341,
"loss": 2.711,
"theoretical_loss": 3.500493717338969,
"tokens_seen": 1577910272
},
{
"epoch": 0.48,
"learning_rate": 0.0002636208625877633,
"loss": 2.8256,
"theoretical_loss": 3.5004810231262216,
"tokens_seen": 1577975808
},
{
"epoch": 0.48,
"learning_rate": 0.00026361083249749244,
"loss": 2.7922,
"theoretical_loss": 3.500468329588288,
"tokens_seen": 1578041344
},
{
"epoch": 0.48,
"learning_rate": 0.0002636008024072217,
"loss": 2.5442,
"theoretical_loss": 3.5004556367251043,
"tokens_seen": 1578106880
},
{
"epoch": 0.48,
"learning_rate": 0.00026359077231695086,
"loss": 2.676,
"theoretical_loss": 3.500442944536606,
"tokens_seen": 1578172416
},
{
"epoch": 0.48,
"learning_rate": 0.00026358074222668004,
"loss": 2.6103,
"theoretical_loss": 3.500430253022729,
"tokens_seen": 1578237952
},
{
"epoch": 0.48,
"learning_rate": 0.0002635707121364092,
"loss": 2.6836,
"theoretical_loss": 3.5004175621834106,
"tokens_seen": 1578303488
},
{
"epoch": 0.48,
"learning_rate": 0.0002635606820461384,
"loss": 2.6981,
"theoretical_loss": 3.5004048720185863,
"tokens_seen": 1578369024
},
{
"epoch": 0.48,
"learning_rate": 0.0002635506519558676,
"loss": 2.6014,
"theoretical_loss": 3.5003921825281923,
"tokens_seen": 1578434560
},
{
"epoch": 0.48,
"learning_rate": 0.0002635406218655968,
"loss": 2.6325,
"theoretical_loss": 3.500379493712165,
"tokens_seen": 1578500096
},
{
"epoch": 0.48,
"learning_rate": 0.00026353059177532595,
"loss": 2.8611,
"theoretical_loss": 3.50036680557044,
"tokens_seen": 1578565632
},
{
"epoch": 0.48,
"learning_rate": 0.0002635205616850552,
"loss": 2.9251,
"theoretical_loss": 3.500354118102954,
"tokens_seen": 1578631168
},
{
"epoch": 0.48,
"learning_rate": 0.00026351053159478436,
"loss": 2.5612,
"theoretical_loss": 3.500341431309643,
"tokens_seen": 1578696704
},
{
"epoch": 0.48,
"learning_rate": 0.00026350050150451354,
"loss": 2.7483,
"theoretical_loss": 3.5003287451904423,
"tokens_seen": 1578762240
},
{
"epoch": 0.48,
"learning_rate": 0.0002634904714142428,
"loss": 2.8014,
"theoretical_loss": 3.5003160597452903,
"tokens_seen": 1578827776
},
{
"epoch": 0.48,
"learning_rate": 0.0002634804413239719,
"loss": 2.9386,
"theoretical_loss": 3.500303374974121,
"tokens_seen": 1578893312
},
{
"epoch": 0.48,
"learning_rate": 0.00026347041123370114,
"loss": 2.8689,
"theoretical_loss": 3.500290690876872,
"tokens_seen": 1578958848
},
{
"epoch": 0.48,
"learning_rate": 0.00026346038114343027,
"loss": 2.8102,
"theoretical_loss": 3.5002780074534785,
"tokens_seen": 1579024384
},
{
"epoch": 0.48,
"learning_rate": 0.0002634503510531595,
"loss": 2.8983,
"theoretical_loss": 3.5002653247038777,
"tokens_seen": 1579089920
},
{
"epoch": 0.48,
"learning_rate": 0.0002634403209628887,
"loss": 2.6739,
"theoretical_loss": 3.5002526426280047,
"tokens_seen": 1579155456
},
{
"epoch": 0.48,
"learning_rate": 0.00026343029087261787,
"loss": 2.8154,
"theoretical_loss": 3.5002399612257973,
"tokens_seen": 1579220992
},
{
"epoch": 0.48,
"learning_rate": 0.00026342026078234705,
"loss": 2.599,
"theoretical_loss": 3.5002272804971906,
"tokens_seen": 1579286528
},
{
"epoch": 0.48,
"learning_rate": 0.0002634102306920763,
"loss": 2.726,
"theoretical_loss": 3.5002146004421215,
"tokens_seen": 1579352064
},
{
"epoch": 0.48,
"objective/train/docs_used": 895615,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9208004474639893,
"objective/train/theoretical_loss": 3.5002019210605253,
"objective/train/tokens_used": 1599877600,
"theoretical_loss": 3.5002019210605253,
"tokens_seen": 1579417600
},
{
"epoch": 0.48,
"learning_rate": 0.0002634002006018054,
"loss": 2.9608,
"theoretical_loss": 3.5002019210605253,
"tokens_seen": 1579417600
},
{
"epoch": 0.48,
"learning_rate": 0.00026339017051153464,
"loss": 2.8141,
"theoretical_loss": 3.5001892423523397,
"tokens_seen": 1579483136
},
{
"epoch": 0.48,
"learning_rate": 0.00026338014042126377,
"loss": 2.7341,
"theoretical_loss": 3.5001765643174996,
"tokens_seen": 1579548672
},
{
"epoch": 0.48,
"learning_rate": 0.000263370110330993,
"loss": 2.4907,
"theoretical_loss": 3.500163886955942,
"tokens_seen": 1579614208
},
{
"epoch": 0.48,
"learning_rate": 0.0002633600802407222,
"loss": 2.6707,
"theoretical_loss": 3.500151210267604,
"tokens_seen": 1579679744
},
{
"epoch": 0.48,
"learning_rate": 0.00026335005015045137,
"loss": 2.6937,
"theoretical_loss": 3.5001385342524203,
"tokens_seen": 1579745280
},
{
"epoch": 0.48,
"learning_rate": 0.00026334002006018055,
"loss": 2.7885,
"theoretical_loss": 3.5001258589103283,
"tokens_seen": 1579810816
},
{
"epoch": 0.48,
"learning_rate": 0.00026332998996990973,
"loss": 2.7356,
"theoretical_loss": 3.500113184241264,
"tokens_seen": 1579876352
},
{
"epoch": 0.48,
"learning_rate": 0.0002633199598796389,
"loss": 2.7413,
"theoretical_loss": 3.5001005102451637,
"tokens_seen": 1579941888
},
{
"epoch": 0.48,
"learning_rate": 0.00026330992978936815,
"loss": 2.6718,
"theoretical_loss": 3.500087836921964,
"tokens_seen": 1580007424
},
{
"epoch": 0.48,
"learning_rate": 0.0002632998996990973,
"loss": 2.6485,
"theoretical_loss": 3.500075164271601,
"tokens_seen": 1580072960
},
{
"epoch": 0.48,
"learning_rate": 0.0002632898696088265,
"loss": 2.7305,
"theoretical_loss": 3.500062492294011,
"tokens_seen": 1580138496
},
{
"epoch": 0.48,
"learning_rate": 0.0002632798395185557,
"loss": 2.6516,
"theoretical_loss": 3.5000498209891306,
"tokens_seen": 1580204032
},
{
"epoch": 0.48,
"learning_rate": 0.00026326980942828487,
"loss": 2.6273,
"theoretical_loss": 3.5000371503568966,
"tokens_seen": 1580269568
},
{
"epoch": 0.48,
"learning_rate": 0.00026325977933801405,
"loss": 2.7687,
"theoretical_loss": 3.5000244803972453,
"tokens_seen": 1580335104
},
{
"epoch": 0.48,
"learning_rate": 0.00026324974924774323,
"loss": 2.4526,
"theoretical_loss": 3.500011811110112,
"tokens_seen": 1580400640
},
{
"epoch": 0.48,
"learning_rate": 0.0002632397191574724,
"loss": 2.6272,
"theoretical_loss": 3.499999142495434,
"tokens_seen": 1580466176
},
{
"epoch": 0.48,
"learning_rate": 0.00026322968906720165,
"loss": 2.5854,
"theoretical_loss": 3.499986474553148,
"tokens_seen": 1580531712
},
{
"epoch": 0.48,
"learning_rate": 0.0002632196589769308,
"loss": 2.649,
"theoretical_loss": 3.49997380728319,
"tokens_seen": 1580597248
},
{
"epoch": 0.48,
"learning_rate": 0.00026320962888666,
"loss": 2.5222,
"theoretical_loss": 3.499961140685496,
"tokens_seen": 1580662784
},
{
"epoch": 0.48,
"learning_rate": 0.00026319959879638914,
"loss": 2.6583,
"theoretical_loss": 3.4999484747600036,
"tokens_seen": 1580728320
},
{
"epoch": 0.48,
"learning_rate": 0.0002631895687061184,
"loss": 2.6752,
"theoretical_loss": 3.499935809506648,
"tokens_seen": 1580793856
},
{
"epoch": 0.48,
"learning_rate": 0.00026317953861584756,
"loss": 2.5529,
"theoretical_loss": 3.499923144925367,
"tokens_seen": 1580859392
},
{
"epoch": 0.48,
"learning_rate": 0.00026316950852557674,
"loss": 2.7901,
"theoretical_loss": 3.499910481016096,
"tokens_seen": 1580924928
},
{
"epoch": 0.48,
"learning_rate": 0.0002631594784353059,
"loss": 2.5506,
"theoretical_loss": 3.4998978177787716,
"tokens_seen": 1580990464
},
{
"epoch": 0.48,
"objective/train/docs_used": 896135,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8393044471740723,
"objective/train/theoretical_loss": 3.499885155213331,
"objective/train/tokens_used": 1601516000,
"theoretical_loss": 3.499885155213331,
"tokens_seen": 1581056000
},
{
"epoch": 0.48,
"learning_rate": 0.0002631494483450351,
"loss": 2.9291,
"theoretical_loss": 3.499885155213331,
"tokens_seen": 1581056000
},
{
"epoch": 0.48,
"learning_rate": 0.0002631394182547643,
"loss": 2.5956,
"theoretical_loss": 3.4998724933197103,
"tokens_seen": 1581121536
},
{
"epoch": 0.48,
"learning_rate": 0.0002631293881644935,
"loss": 2.6563,
"theoretical_loss": 3.4998598320978456,
"tokens_seen": 1581187072
},
{
"epoch": 0.48,
"learning_rate": 0.00026311935807422264,
"loss": 2.7683,
"theoretical_loss": 3.4998471715476738,
"tokens_seen": 1581252608
},
{
"epoch": 0.48,
"learning_rate": 0.0002631093279839519,
"loss": 2.5546,
"theoretical_loss": 3.4998345116691314,
"tokens_seen": 1581318144
},
{
"epoch": 0.48,
"learning_rate": 0.00026309929789368106,
"loss": 2.8311,
"theoretical_loss": 3.499821852462155,
"tokens_seen": 1581383680
},
{
"epoch": 0.48,
"learning_rate": 0.00026308926780341024,
"loss": 2.9554,
"theoretical_loss": 3.499809193926681,
"tokens_seen": 1581449216
},
{
"epoch": 0.48,
"learning_rate": 0.0002630792377131394,
"loss": 2.8682,
"theoretical_loss": 3.4997965360626466,
"tokens_seen": 1581514752
},
{
"epoch": 0.48,
"learning_rate": 0.0002630692076228686,
"loss": 2.6978,
"theoretical_loss": 3.4997838788699873,
"tokens_seen": 1581580288
},
{
"epoch": 0.48,
"learning_rate": 0.0002630591775325978,
"loss": 2.5106,
"theoretical_loss": 3.49977122234864,
"tokens_seen": 1581645824
},
{
"epoch": 0.48,
"learning_rate": 0.000263049147442327,
"loss": 2.7514,
"theoretical_loss": 3.499758566498542,
"tokens_seen": 1581711360
},
{
"epoch": 0.48,
"learning_rate": 0.00026303911735205615,
"loss": 2.6671,
"theoretical_loss": 3.499745911319629,
"tokens_seen": 1581776896
},
{
"epoch": 0.48,
"learning_rate": 0.0002630290872617854,
"loss": 2.7454,
"theoretical_loss": 3.499733256811838,
"tokens_seen": 1581842432
},
{
"epoch": 0.48,
"learning_rate": 0.0002630190571715145,
"loss": 2.602,
"theoretical_loss": 3.4997206029751053,
"tokens_seen": 1581907968
},
{
"epoch": 0.48,
"learning_rate": 0.00026300902708124374,
"loss": 2.8683,
"theoretical_loss": 3.4997079498093684,
"tokens_seen": 1581973504
},
{
"epoch": 0.48,
"learning_rate": 0.0002629989969909729,
"loss": 2.9166,
"theoretical_loss": 3.499695297314563,
"tokens_seen": 1582039040
},
{
"epoch": 0.48,
"learning_rate": 0.0002629889669007021,
"loss": 2.5803,
"theoretical_loss": 3.4996826454906262,
"tokens_seen": 1582104576
},
{
"epoch": 0.48,
"learning_rate": 0.0002629789368104313,
"loss": 2.8282,
"theoretical_loss": 3.499669994337494,
"tokens_seen": 1582170112
},
{
"epoch": 0.48,
"learning_rate": 0.00026296890672016047,
"loss": 2.9128,
"theoretical_loss": 3.499657343855104,
"tokens_seen": 1582235648
},
{
"epoch": 0.48,
"learning_rate": 0.00026295887662988965,
"loss": 2.8043,
"theoretical_loss": 3.499644694043392,
"tokens_seen": 1582301184
},
{
"epoch": 0.48,
"learning_rate": 0.0002629488465396189,
"loss": 2.6758,
"theoretical_loss": 3.4996320449022953,
"tokens_seen": 1582366720
},
{
"epoch": 0.48,
"learning_rate": 0.000262938816449348,
"loss": 2.8503,
"theoretical_loss": 3.4996193964317506,
"tokens_seen": 1582432256
},
{
"epoch": 0.48,
"learning_rate": 0.00026292878635907725,
"loss": 2.8108,
"theoretical_loss": 3.4996067486316935,
"tokens_seen": 1582497792
},
{
"epoch": 0.48,
"learning_rate": 0.00026291875626880643,
"loss": 2.5832,
"theoretical_loss": 3.499594101502062,
"tokens_seen": 1582563328
},
{
"epoch": 0.48,
"learning_rate": 0.0002629087261785356,
"loss": 2.7444,
"theoretical_loss": 3.499581455042793,
"tokens_seen": 1582628864
},
{
"epoch": 0.48,
"objective/train/docs_used": 897089,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.815772771835327,
"objective/train/theoretical_loss": 3.4995688092538213,
"objective/train/tokens_used": 1603154400,
"theoretical_loss": 3.4995688092538213,
"tokens_seen": 1582694400
},
{
"epoch": 0.48,
"learning_rate": 0.0002628986960882648,
"loss": 2.6794,
"theoretical_loss": 3.4995688092538213,
"tokens_seen": 1582694400
},
{
"epoch": 0.48,
"learning_rate": 0.00026288866599799397,
"loss": 2.7824,
"theoretical_loss": 3.499556164135085,
"tokens_seen": 1582759936
},
{
"epoch": 0.48,
"learning_rate": 0.00026287863590772315,
"loss": 2.7896,
"theoretical_loss": 3.4995435196865214,
"tokens_seen": 1582825472
},
{
"epoch": 0.48,
"learning_rate": 0.0002628686058174524,
"loss": 2.8201,
"theoretical_loss": 3.499530875908066,
"tokens_seen": 1582891008
},
{
"epoch": 0.48,
"learning_rate": 0.0002628585757271815,
"loss": 2.7496,
"theoretical_loss": 3.499518232799656,
"tokens_seen": 1582956544
},
{
"epoch": 0.48,
"learning_rate": 0.00026284854563691075,
"loss": 2.6331,
"theoretical_loss": 3.4995055903612284,
"tokens_seen": 1583022080
},
{
"epoch": 0.48,
"learning_rate": 0.0002628385155466399,
"loss": 2.5491,
"theoretical_loss": 3.49949294859272,
"tokens_seen": 1583087616
},
{
"epoch": 0.48,
"learning_rate": 0.0002628284854563691,
"loss": 2.6358,
"theoretical_loss": 3.499480307494067,
"tokens_seen": 1583153152
},
{
"epoch": 0.48,
"learning_rate": 0.0002628184553660983,
"loss": 2.8029,
"theoretical_loss": 3.4994676670652067,
"tokens_seen": 1583218688
},
{
"epoch": 0.48,
"learning_rate": 0.0002628084252758275,
"loss": 2.585,
"theoretical_loss": 3.499455027306076,
"tokens_seen": 1583284224
},
{
"epoch": 0.48,
"learning_rate": 0.00026279839518555666,
"loss": 2.7336,
"theoretical_loss": 3.49944238821661,
"tokens_seen": 1583349760
},
{
"epoch": 0.48,
"learning_rate": 0.0002627883650952859,
"loss": 2.8357,
"theoretical_loss": 3.4994297497967484,
"tokens_seen": 1583415296
},
{
"epoch": 0.48,
"learning_rate": 0.000262778335005015,
"loss": 2.9743,
"theoretical_loss": 3.4994171120464257,
"tokens_seen": 1583480832
},
{
"epoch": 0.48,
"learning_rate": 0.00026276830491474425,
"loss": 2.7777,
"theoretical_loss": 3.49940447496558,
"tokens_seen": 1583546368
},
{
"epoch": 0.48,
"learning_rate": 0.00026275827482447343,
"loss": 2.7716,
"theoretical_loss": 3.499391838554147,
"tokens_seen": 1583611904
},
{
"epoch": 0.48,
"learning_rate": 0.0002627482447342026,
"loss": 2.5445,
"theoretical_loss": 3.499379202812065,
"tokens_seen": 1583677440
},
{
"epoch": 0.48,
"learning_rate": 0.00026273821464393185,
"loss": 2.7211,
"theoretical_loss": 3.49936656773927,
"tokens_seen": 1583742976
},
{
"epoch": 0.48,
"learning_rate": 0.000262728184553661,
"loss": 2.6747,
"theoretical_loss": 3.499353933335698,
"tokens_seen": 1583808512
},
{
"epoch": 0.48,
"learning_rate": 0.0002627181544633902,
"loss": 2.7656,
"theoretical_loss": 3.4993412996012876,
"tokens_seen": 1583874048
},
{
"epoch": 0.48,
"learning_rate": 0.00026270812437311934,
"loss": 2.651,
"theoretical_loss": 3.4993286665359746,
"tokens_seen": 1583939584
},
{
"epoch": 0.48,
"learning_rate": 0.0002626980942828486,
"loss": 2.5429,
"theoretical_loss": 3.4993160341396963,
"tokens_seen": 1584005120
},
{
"epoch": 0.48,
"learning_rate": 0.00026268806419257776,
"loss": 2.6365,
"theoretical_loss": 3.499303402412389,
"tokens_seen": 1584070656
},
{
"epoch": 0.48,
"learning_rate": 0.00026267803410230694,
"loss": 2.713,
"theoretical_loss": 3.4992907713539907,
"tokens_seen": 1584136192
},
{
"epoch": 0.48,
"learning_rate": 0.0002626680040120361,
"loss": 2.7274,
"theoretical_loss": 3.499278140964437,
"tokens_seen": 1584201728
},
{
"epoch": 0.48,
"learning_rate": 0.0002626579739217653,
"loss": 2.8013,
"theoretical_loss": 3.4992655112436655,
"tokens_seen": 1584267264
},
{
"epoch": 0.48,
"objective/train/docs_used": 897735,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.414405107498169,
"objective/train/theoretical_loss": 3.4992528821916133,
"objective/train/tokens_used": 1604792800,
"theoretical_loss": 3.4992528821916133,
"tokens_seen": 1584332800
},
{
"epoch": 0.48,
"learning_rate": 0.0002626479438314945,
"loss": 2.7031,
"theoretical_loss": 3.4992528821916133,
"tokens_seen": 1584332800
},
{
"epoch": 0.48,
"learning_rate": 0.0002626379137412237,
"loss": 2.8627,
"theoretical_loss": 3.4992402538082166,
"tokens_seen": 1584398336
},
{
"epoch": 0.48,
"learning_rate": 0.00026262788365095284,
"loss": 2.7633,
"theoretical_loss": 3.4992276260934134,
"tokens_seen": 1584463872
},
{
"epoch": 0.48,
"learning_rate": 0.0002626178535606821,
"loss": 2.6994,
"theoretical_loss": 3.49921499904714,
"tokens_seen": 1584529408
},
{
"epoch": 0.48,
"learning_rate": 0.00026260782347041126,
"loss": 2.6809,
"theoretical_loss": 3.499202372669333,
"tokens_seen": 1584594944
},
{
"epoch": 0.48,
"learning_rate": 0.00026259779338014044,
"loss": 2.7756,
"theoretical_loss": 3.4991897469599307,
"tokens_seen": 1584660480
},
{
"epoch": 0.48,
"learning_rate": 0.0002625877632898696,
"loss": 2.441,
"theoretical_loss": 3.4991771219188683,
"tokens_seen": 1584726016
},
{
"epoch": 0.48,
"learning_rate": 0.0002625777331995988,
"loss": 2.6641,
"theoretical_loss": 3.4991644975460843,
"tokens_seen": 1584791552
},
{
"epoch": 0.48,
"learning_rate": 0.000262567703109328,
"loss": 2.7362,
"theoretical_loss": 3.4991518738415146,
"tokens_seen": 1584857088
},
{
"epoch": 0.48,
"learning_rate": 0.0002625576730190572,
"loss": 2.7861,
"theoretical_loss": 3.499139250805097,
"tokens_seen": 1584922624
},
{
"epoch": 0.48,
"learning_rate": 0.00026254764292878635,
"loss": 2.7525,
"theoretical_loss": 3.499126628436768,
"tokens_seen": 1584988160
},
{
"epoch": 0.48,
"learning_rate": 0.0002625376128385156,
"loss": 2.7699,
"theoretical_loss": 3.4991140067364648,
"tokens_seen": 1585053696
},
{
"epoch": 0.48,
"learning_rate": 0.0002625275827482447,
"loss": 2.6066,
"theoretical_loss": 3.499101385704124,
"tokens_seen": 1585119232
},
{
"epoch": 0.48,
"learning_rate": 0.00026251755265797394,
"loss": 2.6854,
"theoretical_loss": 3.499088765339684,
"tokens_seen": 1585184768
},
{
"epoch": 0.48,
"learning_rate": 0.0002625075225677031,
"loss": 2.6444,
"theoretical_loss": 3.4990761456430803,
"tokens_seen": 1585250304
},
{
"epoch": 0.48,
"learning_rate": 0.0002624974924774323,
"loss": 2.764,
"theoretical_loss": 3.4990635266142505,
"tokens_seen": 1585315840
},
{
"epoch": 0.48,
"learning_rate": 0.0002624874623871615,
"loss": 2.7804,
"theoretical_loss": 3.4990509082531323,
"tokens_seen": 1585381376
},
{
"epoch": 0.48,
"learning_rate": 0.00026247743229689067,
"loss": 2.6483,
"theoretical_loss": 3.4990382905596618,
"tokens_seen": 1585446912
},
{
"epoch": 0.48,
"learning_rate": 0.00026246740220661985,
"loss": 2.8013,
"theoretical_loss": 3.4990256735337764,
"tokens_seen": 1585512448
},
{
"epoch": 0.48,
"learning_rate": 0.0002624573721163491,
"loss": 2.807,
"theoretical_loss": 3.4990130571754134,
"tokens_seen": 1585577984
},
{
"epoch": 0.48,
"learning_rate": 0.0002624473420260782,
"loss": 2.7142,
"theoretical_loss": 3.49900044148451,
"tokens_seen": 1585643520
},
{
"epoch": 0.48,
"learning_rate": 0.00026243731193580745,
"loss": 2.8318,
"theoretical_loss": 3.4989878264610024,
"tokens_seen": 1585709056
},
{
"epoch": 0.48,
"learning_rate": 0.00026242728184553663,
"loss": 2.8363,
"theoretical_loss": 3.4989752121048285,
"tokens_seen": 1585774592
},
{
"epoch": 0.48,
"learning_rate": 0.0002624172517552658,
"loss": 2.7209,
"theoretical_loss": 3.4989625984159254,
"tokens_seen": 1585840128
},
{
"epoch": 0.48,
"learning_rate": 0.000262407221664995,
"loss": 2.6337,
"theoretical_loss": 3.4989499853942307,
"tokens_seen": 1585905664
},
{
"epoch": 0.48,
"objective/train/docs_used": 899130,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8073930740356445,
"objective/train/theoretical_loss": 3.49893737303968,
"objective/train/tokens_used": 1606431200,
"theoretical_loss": 3.49893737303968,
"tokens_seen": 1585971200
},
{
"epoch": 0.48,
"learning_rate": 0.00026239719157472417,
"loss": 2.6316,
"theoretical_loss": 3.49893737303968,
"tokens_seen": 1585971200
},
{
"epoch": 0.48,
"learning_rate": 0.00026238716148445335,
"loss": 2.8764,
"theoretical_loss": 3.498924761352212,
"tokens_seen": 1586036736
},
{
"epoch": 0.48,
"learning_rate": 0.0002623771313941826,
"loss": 2.8662,
"theoretical_loss": 3.498912150331763,
"tokens_seen": 1586102272
},
{
"epoch": 0.48,
"learning_rate": 0.0002623671013039117,
"loss": 2.6703,
"theoretical_loss": 3.4988995399782707,
"tokens_seen": 1586167808
},
{
"epoch": 0.48,
"learning_rate": 0.00026235707121364095,
"loss": 2.8195,
"theoretical_loss": 3.498886930291672,
"tokens_seen": 1586233344
},
{
"epoch": 0.48,
"learning_rate": 0.0002623470411233701,
"loss": 2.5614,
"theoretical_loss": 3.498874321271904,
"tokens_seen": 1586298880
},
{
"epoch": 0.48,
"learning_rate": 0.0002623370110330993,
"loss": 2.6434,
"theoretical_loss": 3.4988617129189037,
"tokens_seen": 1586364416
},
{
"epoch": 0.48,
"learning_rate": 0.0002623269809428285,
"loss": 2.7715,
"theoretical_loss": 3.4988491052326087,
"tokens_seen": 1586429952
},
{
"epoch": 0.48,
"learning_rate": 0.0002623169508525577,
"loss": 2.8797,
"theoretical_loss": 3.498836498212956,
"tokens_seen": 1586495488
},
{
"epoch": 0.48,
"learning_rate": 0.00026230692076228686,
"loss": 2.8252,
"theoretical_loss": 3.498823891859883,
"tokens_seen": 1586561024
},
{
"epoch": 0.48,
"learning_rate": 0.0002622968906720161,
"loss": 2.7644,
"theoretical_loss": 3.4988112861733267,
"tokens_seen": 1586626560
},
{
"epoch": 0.48,
"learning_rate": 0.0002622868605817452,
"loss": 2.773,
"theoretical_loss": 3.498798681153225,
"tokens_seen": 1586692096
},
{
"epoch": 0.48,
"learning_rate": 0.00026227683049147445,
"loss": 2.6486,
"theoretical_loss": 3.4987860767995143,
"tokens_seen": 1586757632
},
{
"epoch": 0.48,
"learning_rate": 0.0002622668004012036,
"loss": 2.8301,
"theoretical_loss": 3.4987734731121316,
"tokens_seen": 1586823168
},
{
"epoch": 0.48,
"learning_rate": 0.0002622567703109328,
"loss": 2.7105,
"theoretical_loss": 3.498760870091015,
"tokens_seen": 1586888704
},
{
"epoch": 0.48,
"learning_rate": 0.000262246740220662,
"loss": 2.6464,
"theoretical_loss": 3.4987482677361017,
"tokens_seen": 1586954240
},
{
"epoch": 0.48,
"learning_rate": 0.0002622367101303912,
"loss": 2.7968,
"theoretical_loss": 3.4987356660473283,
"tokens_seen": 1587019776
},
{
"epoch": 0.48,
"learning_rate": 0.00026222668004012036,
"loss": 2.6032,
"theoretical_loss": 3.498723065024633,
"tokens_seen": 1587085312
},
{
"epoch": 0.48,
"learning_rate": 0.00026221664994984954,
"loss": 2.6696,
"theoretical_loss": 3.498710464667952,
"tokens_seen": 1587150848
},
{
"epoch": 0.48,
"learning_rate": 0.0002622066198595787,
"loss": 2.7157,
"theoretical_loss": 3.4986978649772236,
"tokens_seen": 1587216384
},
{
"epoch": 0.48,
"learning_rate": 0.00026219658976930796,
"loss": 2.7186,
"theoretical_loss": 3.4986852659523846,
"tokens_seen": 1587281920
},
{
"epoch": 0.48,
"learning_rate": 0.0002621865596790371,
"loss": 2.909,
"theoretical_loss": 3.4986726675933726,
"tokens_seen": 1587347456
},
{
"epoch": 0.48,
"learning_rate": 0.0002621765295887663,
"loss": 2.7193,
"theoretical_loss": 3.498660069900124,
"tokens_seen": 1587412992
},
{
"epoch": 0.48,
"learning_rate": 0.00026216649949849545,
"loss": 2.6583,
"theoretical_loss": 3.4986474728725776,
"tokens_seen": 1587478528
},
{
"epoch": 0.48,
"learning_rate": 0.0002621564694082247,
"loss": 2.5737,
"theoretical_loss": 3.4986348765106703,
"tokens_seen": 1587544064
},
{
"epoch": 0.48,
"objective/train/docs_used": 899656,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4070703983306885,
"objective/train/theoretical_loss": 3.4986222808143386,
"objective/train/tokens_used": 1608069600,
"theoretical_loss": 3.4986222808143386,
"tokens_seen": 1587609600
},
{
"epoch": 0.48,
"learning_rate": 0.00026214643931795386,
"loss": 2.727,
"theoretical_loss": 3.4986222808143386,
"tokens_seen": 1587609600
},
{
"epoch": 0.48,
"learning_rate": 0.00026213640922768304,
"loss": 2.8577,
"theoretical_loss": 3.4986096857835207,
"tokens_seen": 1587675136
},
{
"epoch": 0.48,
"learning_rate": 0.0002621263791374122,
"loss": 2.8464,
"theoretical_loss": 3.4985970914181532,
"tokens_seen": 1587740672
},
{
"epoch": 0.48,
"learning_rate": 0.00026211634904714146,
"loss": 2.6859,
"theoretical_loss": 3.4985844977181744,
"tokens_seen": 1587806208
},
{
"epoch": 0.48,
"learning_rate": 0.0002621063189568706,
"loss": 2.8247,
"theoretical_loss": 3.4985719046835215,
"tokens_seen": 1587871744
},
{
"epoch": 0.48,
"learning_rate": 0.0002620962888665998,
"loss": 2.5684,
"theoretical_loss": 3.498559312314131,
"tokens_seen": 1587937280
},
{
"epoch": 0.48,
"learning_rate": 0.00026208625877632895,
"loss": 2.8614,
"theoretical_loss": 3.4985467206099417,
"tokens_seen": 1588002816
},
{
"epoch": 0.48,
"learning_rate": 0.0002620762286860582,
"loss": 2.6739,
"theoretical_loss": 3.4985341295708903,
"tokens_seen": 1588068352
},
{
"epoch": 0.48,
"learning_rate": 0.00026206619859578737,
"loss": 2.5587,
"theoretical_loss": 3.4985215391969136,
"tokens_seen": 1588133888
},
{
"epoch": 0.48,
"learning_rate": 0.00026205616850551655,
"loss": 2.3027,
"theoretical_loss": 3.4985089494879498,
"tokens_seen": 1588199424
},
{
"epoch": 0.48,
"learning_rate": 0.00026204613841524573,
"loss": 2.788,
"theoretical_loss": 3.4984963604439363,
"tokens_seen": 1588264960
},
{
"epoch": 0.48,
"learning_rate": 0.0002620361083249749,
"loss": 2.7498,
"theoretical_loss": 3.49848377206481,
"tokens_seen": 1588330496
},
{
"epoch": 0.48,
"learning_rate": 0.0002620260782347041,
"loss": 2.8032,
"theoretical_loss": 3.4984711843505094,
"tokens_seen": 1588396032
},
{
"epoch": 0.48,
"learning_rate": 0.0002620160481444333,
"loss": 2.6354,
"theoretical_loss": 3.4984585973009708,
"tokens_seen": 1588461568
},
{
"epoch": 0.48,
"learning_rate": 0.0002620060180541625,
"loss": 2.5564,
"theoretical_loss": 3.4984460109161324,
"tokens_seen": 1588527104
},
{
"epoch": 0.48,
"learning_rate": 0.0002619959879638917,
"loss": 2.729,
"theoretical_loss": 3.4984334251959313,
"tokens_seen": 1588592640
},
{
"epoch": 0.48,
"learning_rate": 0.00026198595787362087,
"loss": 2.5689,
"theoretical_loss": 3.498420840140305,
"tokens_seen": 1588658176
},
{
"epoch": 0.48,
"learning_rate": 0.00026197592778335005,
"loss": 2.6821,
"theoretical_loss": 3.498408255749191,
"tokens_seen": 1588723712
},
{
"epoch": 0.48,
"learning_rate": 0.0002619658976930793,
"loss": 2.6462,
"theoretical_loss": 3.4983956720225278,
"tokens_seen": 1588789248
},
{
"epoch": 0.48,
"learning_rate": 0.0002619558676028084,
"loss": 2.7675,
"theoretical_loss": 3.498383088960251,
"tokens_seen": 1588854784
},
{
"epoch": 0.48,
"learning_rate": 0.00026194583751253765,
"loss": 2.6184,
"theoretical_loss": 3.4983705065623,
"tokens_seen": 1588920320
},
{
"epoch": 0.48,
"learning_rate": 0.00026193580742226683,
"loss": 2.5463,
"theoretical_loss": 3.4983579248286114,
"tokens_seen": 1588985856
},
{
"epoch": 0.48,
"learning_rate": 0.000261925777331996,
"loss": 2.5634,
"theoretical_loss": 3.4983453437591225,
"tokens_seen": 1589051392
},
{
"epoch": 0.48,
"learning_rate": 0.0002619157472417252,
"loss": 2.6999,
"theoretical_loss": 3.498332763353771,
"tokens_seen": 1589116928
},
{
"epoch": 0.48,
"learning_rate": 0.00026190571715145437,
"loss": 2.6279,
"theoretical_loss": 3.498320183612495,
"tokens_seen": 1589182464
},
{
"epoch": 0.48,
"objective/train/docs_used": 900745,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.173302412033081,
"objective/train/theoretical_loss": 3.4983076045352317,
"objective/train/tokens_used": 1609708000,
"theoretical_loss": 3.4983076045352317,
"tokens_seen": 1589248000
},
{
"epoch": 0.48,
"learning_rate": 0.00026189568706118355,
"loss": 2.8915,
"theoretical_loss": 3.4983076045352317,
"tokens_seen": 1589248000
},
{
"epoch": 0.48,
"learning_rate": 0.0002618856569709128,
"loss": 2.6834,
"theoretical_loss": 3.4982950261219186,
"tokens_seen": 1589313536
},
{
"epoch": 0.48,
"learning_rate": 0.0002618756268806419,
"loss": 2.629,
"theoretical_loss": 3.4982824483724935,
"tokens_seen": 1589379072
},
{
"epoch": 0.48,
"learning_rate": 0.00026186559679037115,
"loss": 2.7453,
"theoretical_loss": 3.498269871286894,
"tokens_seen": 1589444608
},
{
"epoch": 0.48,
"learning_rate": 0.0002618555667001003,
"loss": 2.6144,
"theoretical_loss": 3.4982572948650565,
"tokens_seen": 1589510144
},
{
"epoch": 0.48,
"learning_rate": 0.0002618455366098295,
"loss": 2.8416,
"theoretical_loss": 3.4982447191069204,
"tokens_seen": 1589575680
},
{
"epoch": 0.48,
"learning_rate": 0.0002618355065195587,
"loss": 2.9646,
"theoretical_loss": 3.4982321440124227,
"tokens_seen": 1589641216
},
{
"epoch": 0.48,
"learning_rate": 0.0002618254764292879,
"loss": 2.7694,
"theoretical_loss": 3.498219569581501,
"tokens_seen": 1589706752
},
{
"epoch": 0.48,
"learning_rate": 0.00026181544633901706,
"loss": 2.6492,
"theoretical_loss": 3.4982069958140922,
"tokens_seen": 1589772288
},
{
"epoch": 0.48,
"learning_rate": 0.0002618054162487463,
"loss": 2.6605,
"theoretical_loss": 3.4981944227101347,
"tokens_seen": 1589837824
},
{
"epoch": 0.48,
"learning_rate": 0.0002617953861584754,
"loss": 2.6945,
"theoretical_loss": 3.498181850269566,
"tokens_seen": 1589903360
},
{
"epoch": 0.48,
"learning_rate": 0.00026178535606820465,
"loss": 2.565,
"theoretical_loss": 3.498169278492324,
"tokens_seen": 1589968896
},
{
"epoch": 0.48,
"learning_rate": 0.0002617753259779338,
"loss": 2.7248,
"theoretical_loss": 3.498156707378346,
"tokens_seen": 1590034432
},
{
"epoch": 0.48,
"learning_rate": 0.000261765295887663,
"loss": 2.6367,
"theoretical_loss": 3.4981441369275696,
"tokens_seen": 1590099968
},
{
"epoch": 0.48,
"learning_rate": 0.0002617552657973922,
"loss": 2.5465,
"theoretical_loss": 3.4981315671399327,
"tokens_seen": 1590165504
},
{
"epoch": 0.48,
"learning_rate": 0.0002617452357071214,
"loss": 2.5025,
"theoretical_loss": 3.498118998015373,
"tokens_seen": 1590231040
},
{
"epoch": 0.48,
"learning_rate": 0.00026173520561685056,
"loss": 2.778,
"theoretical_loss": 3.498106429553828,
"tokens_seen": 1590296576
},
{
"epoch": 0.48,
"learning_rate": 0.00026172517552657974,
"loss": 2.8214,
"theoretical_loss": 3.498093861755236,
"tokens_seen": 1590362112
},
{
"epoch": 0.48,
"learning_rate": 0.0002617151454363089,
"loss": 2.793,
"theoretical_loss": 3.4980812946195337,
"tokens_seen": 1590427648
},
{
"epoch": 0.48,
"learning_rate": 0.00026170511534603816,
"loss": 2.7358,
"theoretical_loss": 3.4980687281466594,
"tokens_seen": 1590493184
},
{
"epoch": 0.48,
"learning_rate": 0.0002616950852557673,
"loss": 2.7095,
"theoretical_loss": 3.4980561623365514,
"tokens_seen": 1590558720
},
{
"epoch": 0.48,
"learning_rate": 0.0002616850551654965,
"loss": 2.5843,
"theoretical_loss": 3.498043597189146,
"tokens_seen": 1590624256
},
{
"epoch": 0.48,
"learning_rate": 0.00026167502507522565,
"loss": 2.7493,
"theoretical_loss": 3.4980310327043824,
"tokens_seen": 1590689792
},
{
"epoch": 0.48,
"learning_rate": 0.0002616649949849549,
"loss": 2.7446,
"theoretical_loss": 3.498018468882198,
"tokens_seen": 1590755328
},
{
"epoch": 0.48,
"learning_rate": 0.00026165496489468406,
"loss": 2.9456,
"theoretical_loss": 3.49800590572253,
"tokens_seen": 1590820864
},
{
"epoch": 0.48,
"objective/train/docs_used": 902130,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9453482627868652,
"objective/train/theoretical_loss": 3.4979933432253163,
"objective/train/tokens_used": 1611346400,
"theoretical_loss": 3.4979933432253163,
"tokens_seen": 1590886400
},
{
"epoch": 0.48,
"learning_rate": 0.00026164493480441324,
"loss": 2.7902,
"theoretical_loss": 3.4979933432253163,
"tokens_seen": 1590886400
},
{
"epoch": 0.48,
"learning_rate": 0.0002616349047141424,
"loss": 2.8172,
"theoretical_loss": 3.497980781390495,
"tokens_seen": 1590951936
},
{
"epoch": 0.48,
"learning_rate": 0.00026162487462387166,
"loss": 2.9388,
"theoretical_loss": 3.497968220218004,
"tokens_seen": 1591017472
},
{
"epoch": 0.48,
"learning_rate": 0.0002616148445336008,
"loss": 2.6931,
"theoretical_loss": 3.49795565970778,
"tokens_seen": 1591083008
},
{
"epoch": 0.48,
"learning_rate": 0.00026160481444333,
"loss": 2.5615,
"theoretical_loss": 3.4979430998597625,
"tokens_seen": 1591148544
},
{
"epoch": 0.48,
"learning_rate": 0.00026159478435305915,
"loss": 2.6003,
"theoretical_loss": 3.4979305406738885,
"tokens_seen": 1591214080
},
{
"epoch": 0.48,
"learning_rate": 0.0002615847542627884,
"loss": 2.5313,
"theoretical_loss": 3.4979179821500956,
"tokens_seen": 1591279616
},
{
"epoch": 0.48,
"learning_rate": 0.00026157472417251757,
"loss": 2.63,
"theoretical_loss": 3.497905424288322,
"tokens_seen": 1591345152
},
{
"epoch": 0.48,
"learning_rate": 0.00026156469408224675,
"loss": 2.7247,
"theoretical_loss": 3.497892867088505,
"tokens_seen": 1591410688
},
{
"epoch": 0.48,
"learning_rate": 0.00026155466399197593,
"loss": 2.4647,
"theoretical_loss": 3.4978803105505833,
"tokens_seen": 1591476224
},
{
"epoch": 0.48,
"learning_rate": 0.0002615446339017051,
"loss": 2.7272,
"theoretical_loss": 3.497867754674494,
"tokens_seen": 1591541760
},
{
"epoch": 0.48,
"learning_rate": 0.0002615346038114343,
"loss": 2.7753,
"theoretical_loss": 3.4978551994601754,
"tokens_seen": 1591607296
},
{
"epoch": 0.48,
"learning_rate": 0.0002615245737211635,
"loss": 2.675,
"theoretical_loss": 3.497842644907565,
"tokens_seen": 1591672832
},
{
"epoch": 0.48,
"learning_rate": 0.00026151454363089265,
"loss": 2.7009,
"theoretical_loss": 3.4978300910166014,
"tokens_seen": 1591738368
},
{
"epoch": 0.48,
"learning_rate": 0.0002615045135406219,
"loss": 2.8336,
"theoretical_loss": 3.4978175377872214,
"tokens_seen": 1591803904
},
{
"epoch": 0.48,
"learning_rate": 0.000261494483450351,
"loss": 2.6091,
"theoretical_loss": 3.497804985219364,
"tokens_seen": 1591869440
},
{
"epoch": 0.48,
"learning_rate": 0.00026148445336008025,
"loss": 2.7427,
"theoretical_loss": 3.4977924333129664,
"tokens_seen": 1591934976
},
{
"epoch": 0.48,
"learning_rate": 0.00026147442326980943,
"loss": 2.4583,
"theoretical_loss": 3.4977798820679666,
"tokens_seen": 1592000512
},
{
"epoch": 0.48,
"learning_rate": 0.0002614643931795386,
"loss": 2.4623,
"theoretical_loss": 3.4977673314843027,
"tokens_seen": 1592066048
},
{
"epoch": 0.48,
"learning_rate": 0.0002614543630892678,
"loss": 2.5587,
"theoretical_loss": 3.4977547815619126,
"tokens_seen": 1592131584
},
{
"epoch": 0.48,
"learning_rate": 0.00026144433299899703,
"loss": 2.6087,
"theoretical_loss": 3.497742232300734,
"tokens_seen": 1592197120
},
{
"epoch": 0.48,
"learning_rate": 0.00026143430290872616,
"loss": 2.5397,
"theoretical_loss": 3.4977296837007055,
"tokens_seen": 1592262656
},
{
"epoch": 0.48,
"learning_rate": 0.0002614242728184554,
"loss": 2.5022,
"theoretical_loss": 3.4977171357617642,
"tokens_seen": 1592328192
},
{
"epoch": 0.48,
"learning_rate": 0.0002614142427281845,
"loss": 2.7143,
"theoretical_loss": 3.497704588483849,
"tokens_seen": 1592393728
},
{
"epoch": 0.48,
"learning_rate": 0.00026140421263791375,
"loss": 2.8221,
"theoretical_loss": 3.497692041866897,
"tokens_seen": 1592459264
},
{
"epoch": 0.48,
"objective/train/docs_used": 902721,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7125089168548584,
"objective/train/theoretical_loss": 3.4976794959108464,
"objective/train/tokens_used": 1612984800,
"theoretical_loss": 3.4976794959108464,
"tokens_seen": 1592524800
},
{
"epoch": 0.48,
"learning_rate": 0.00026139418254764293,
"loss": 2.6174,
"theoretical_loss": 3.4976794959108464,
"tokens_seen": 1592524800
},
{
"epoch": 0.48,
"learning_rate": 0.0002613841524573721,
"loss": 2.7289,
"theoretical_loss": 3.4976669506156357,
"tokens_seen": 1592590336
},
{
"epoch": 0.48,
"learning_rate": 0.0002613741223671013,
"loss": 2.7524,
"theoretical_loss": 3.4976544059812023,
"tokens_seen": 1592655872
},
{
"epoch": 0.48,
"learning_rate": 0.0002613640922768305,
"loss": 2.7066,
"theoretical_loss": 3.4976418620074847,
"tokens_seen": 1592721408
},
{
"epoch": 0.48,
"learning_rate": 0.00026135406218655966,
"loss": 2.8915,
"theoretical_loss": 3.49762931869442,
"tokens_seen": 1592786944
},
{
"epoch": 0.48,
"learning_rate": 0.0002613440320962889,
"loss": 2.7387,
"theoretical_loss": 3.497616776041948,
"tokens_seen": 1592852480
},
{
"epoch": 0.48,
"learning_rate": 0.000261334002006018,
"loss": 2.51,
"theoretical_loss": 3.497604234050005,
"tokens_seen": 1592918016
},
{
"epoch": 0.48,
"learning_rate": 0.00026132397191574726,
"loss": 2.685,
"theoretical_loss": 3.4975916927185295,
"tokens_seen": 1592983552
},
{
"epoch": 0.48,
"learning_rate": 0.0002613139418254764,
"loss": 2.7992,
"theoretical_loss": 3.49757915204746,
"tokens_seen": 1593049088
},
{
"epoch": 0.48,
"learning_rate": 0.0002613039117352056,
"loss": 2.6267,
"theoretical_loss": 3.497566612036734,
"tokens_seen": 1593114624
},
{
"epoch": 0.48,
"learning_rate": 0.0002612938816449348,
"loss": 2.7275,
"theoretical_loss": 3.49755407268629,
"tokens_seen": 1593180160
},
{
"epoch": 0.48,
"learning_rate": 0.000261283851554664,
"loss": 2.4109,
"theoretical_loss": 3.497541533996066,
"tokens_seen": 1593245696
},
{
"epoch": 0.48,
"learning_rate": 0.00026127382146439316,
"loss": 2.6383,
"theoretical_loss": 3.497528995966,
"tokens_seen": 1593311232
},
{
"epoch": 0.48,
"learning_rate": 0.0002612637913741224,
"loss": 2.8103,
"theoretical_loss": 3.4975164585960297,
"tokens_seen": 1593376768
},
{
"epoch": 0.48,
"learning_rate": 0.0002612537612838516,
"loss": 2.7368,
"theoretical_loss": 3.497503921886094,
"tokens_seen": 1593442304
},
{
"epoch": 0.48,
"learning_rate": 0.00026124373119358076,
"loss": 2.6882,
"theoretical_loss": 3.4974913858361303,
"tokens_seen": 1593507840
},
{
"epoch": 0.48,
"learning_rate": 0.00026123370110330994,
"loss": 2.769,
"theoretical_loss": 3.497478850446077,
"tokens_seen": 1593573376
},
{
"epoch": 0.48,
"learning_rate": 0.0002612236710130391,
"loss": 2.619,
"theoretical_loss": 3.4974663157158727,
"tokens_seen": 1593638912
},
{
"epoch": 0.48,
"learning_rate": 0.00026121364092276836,
"loss": 2.6492,
"theoretical_loss": 3.497453781645455,
"tokens_seen": 1593704448
},
{
"epoch": 0.48,
"learning_rate": 0.0002612036108324975,
"loss": 2.8228,
"theoretical_loss": 3.497441248234762,
"tokens_seen": 1593769984
},
{
"epoch": 0.48,
"learning_rate": 0.0002611935807422267,
"loss": 2.3707,
"theoretical_loss": 3.4974287154837316,
"tokens_seen": 1593835520
},
{
"epoch": 0.48,
"learning_rate": 0.00026118355065195585,
"loss": 2.8959,
"theoretical_loss": 3.4974161833923025,
"tokens_seen": 1593901056
},
{
"epoch": 0.48,
"learning_rate": 0.0002611735205616851,
"loss": 2.7106,
"theoretical_loss": 3.497403651960413,
"tokens_seen": 1593966592
},
{
"epoch": 0.48,
"learning_rate": 0.00026116349047141426,
"loss": 2.8727,
"theoretical_loss": 3.4973911211880004,
"tokens_seen": 1594032128
},
{
"epoch": 0.48,
"learning_rate": 0.00026115346038114344,
"loss": 2.6681,
"theoretical_loss": 3.4973785910750035,
"tokens_seen": 1594097664
},
{
"epoch": 0.48,
"objective/train/docs_used": 903500,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7019095420837402,
"objective/train/theoretical_loss": 3.497366061621361,
"objective/train/tokens_used": 1614623200,
"theoretical_loss": 3.497366061621361,
"tokens_seen": 1594163200
},
{
"epoch": 0.48,
"learning_rate": 0.0002611434302908726,
"loss": 2.678,
"theoretical_loss": 3.497366061621361,
"tokens_seen": 1594163200
},
{
"epoch": 0.48,
"learning_rate": 0.00026113340020060186,
"loss": 2.82,
"theoretical_loss": 3.4973535328270104,
"tokens_seen": 1594228736
},
{
"epoch": 0.48,
"learning_rate": 0.000261123370110331,
"loss": 2.5539,
"theoretical_loss": 3.4973410046918896,
"tokens_seen": 1594294272
},
{
"epoch": 0.48,
"learning_rate": 0.0002611133400200602,
"loss": 2.7586,
"theoretical_loss": 3.4973284772159374,
"tokens_seen": 1594359808
},
{
"epoch": 0.48,
"learning_rate": 0.00026110330992978935,
"loss": 2.5912,
"theoretical_loss": 3.4973159503990923,
"tokens_seen": 1594425344
},
{
"epoch": 0.48,
"learning_rate": 0.0002610932798395186,
"loss": 2.819,
"theoretical_loss": 3.4973034242412915,
"tokens_seen": 1594490880
},
{
"epoch": 0.48,
"learning_rate": 0.00026108324974924777,
"loss": 2.664,
"theoretical_loss": 3.4972908987424747,
"tokens_seen": 1594556416
},
{
"epoch": 0.48,
"learning_rate": 0.00026107321965897695,
"loss": 2.8922,
"theoretical_loss": 3.4972783739025783,
"tokens_seen": 1594621952
},
{
"epoch": 0.48,
"learning_rate": 0.00026106318956870613,
"loss": 2.7895,
"theoretical_loss": 3.497265849721542,
"tokens_seen": 1594687488
},
{
"epoch": 0.48,
"learning_rate": 0.0002610531594784353,
"loss": 2.6689,
"theoretical_loss": 3.4972533261993037,
"tokens_seen": 1594753024
},
{
"epoch": 0.48,
"learning_rate": 0.0002610431293881645,
"loss": 2.7292,
"theoretical_loss": 3.497240803335802,
"tokens_seen": 1594818560
},
{
"epoch": 0.48,
"learning_rate": 0.0002610330992978937,
"loss": 2.8997,
"theoretical_loss": 3.4972282811309743,
"tokens_seen": 1594884096
},
{
"epoch": 0.48,
"learning_rate": 0.00026102306920762285,
"loss": 2.8973,
"theoretical_loss": 3.4972157595847593,
"tokens_seen": 1594949632
},
{
"epoch": 0.48,
"learning_rate": 0.0002610130391173521,
"loss": 2.7609,
"theoretical_loss": 3.4972032386970957,
"tokens_seen": 1595015168
},
{
"epoch": 0.48,
"learning_rate": 0.0002610030090270812,
"loss": 2.7334,
"theoretical_loss": 3.4971907184679214,
"tokens_seen": 1595080704
},
{
"epoch": 0.48,
"learning_rate": 0.00026099297893681045,
"loss": 2.7588,
"theoretical_loss": 3.4971781988971746,
"tokens_seen": 1595146240
},
{
"epoch": 0.48,
"learning_rate": 0.00026098294884653963,
"loss": 2.9154,
"theoretical_loss": 3.497165679984794,
"tokens_seen": 1595211776
},
{
"epoch": 0.48,
"learning_rate": 0.0002609729187562688,
"loss": 2.9821,
"theoretical_loss": 3.4971531617307177,
"tokens_seen": 1595277312
},
{
"epoch": 0.48,
"learning_rate": 0.000260962888665998,
"loss": 2.5743,
"theoretical_loss": 3.497140644134884,
"tokens_seen": 1595342848
},
{
"epoch": 0.48,
"learning_rate": 0.00026095285857572723,
"loss": 2.6973,
"theoretical_loss": 3.4971281271972314,
"tokens_seen": 1595408384
},
{
"epoch": 0.48,
"learning_rate": 0.00026094282848545636,
"loss": 2.6964,
"theoretical_loss": 3.4971156109176986,
"tokens_seen": 1595473920
},
{
"epoch": 0.48,
"learning_rate": 0.0002609327983951856,
"loss": 2.642,
"theoretical_loss": 3.4971030952962234,
"tokens_seen": 1595539456
},
{
"epoch": 0.48,
"learning_rate": 0.0002609227683049147,
"loss": 2.8027,
"theoretical_loss": 3.497090580332744,
"tokens_seen": 1595604992
},
{
"epoch": 0.48,
"learning_rate": 0.00026091273821464395,
"loss": 2.8233,
"theoretical_loss": 3.4970780660271994,
"tokens_seen": 1595670528
},
{
"epoch": 0.48,
"learning_rate": 0.00026090270812437313,
"loss": 2.6818,
"theoretical_loss": 3.497065552379528,
"tokens_seen": 1595736064
},
{
"epoch": 0.48,
"objective/train/docs_used": 904004,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.70963716506958,
"objective/train/theoretical_loss": 3.4970530393896673,
"objective/train/tokens_used": 1616261600,
"theoretical_loss": 3.4970530393896673,
"tokens_seen": 1595801600
},
{
"epoch": 0.48,
"learning_rate": 0.0002608926780341023,
"loss": 2.7064,
"theoretical_loss": 3.4970530393896673,
"tokens_seen": 1595801600
},
{
"epoch": 0.48,
"learning_rate": 0.0002608826479438315,
"loss": 2.3913,
"theoretical_loss": 3.497040527057557,
"tokens_seen": 1595867136
},
{
"epoch": 0.48,
"learning_rate": 0.0002608726178535607,
"loss": 2.8495,
"theoretical_loss": 3.497028015383134,
"tokens_seen": 1595932672
},
{
"epoch": 0.48,
"learning_rate": 0.00026086258776328986,
"loss": 2.6942,
"theoretical_loss": 3.4970155043663382,
"tokens_seen": 1595998208
},
{
"epoch": 0.48,
"learning_rate": 0.0002608525576730191,
"loss": 2.5338,
"theoretical_loss": 3.4970029940071075,
"tokens_seen": 1596063744
},
{
"epoch": 0.48,
"learning_rate": 0.0002608425275827482,
"loss": 2.6617,
"theoretical_loss": 3.49699048430538,
"tokens_seen": 1596129280
},
{
"epoch": 0.48,
"learning_rate": 0.00026083249749247746,
"loss": 2.5864,
"theoretical_loss": 3.4969779752610943,
"tokens_seen": 1596194816
},
{
"epoch": 0.48,
"learning_rate": 0.0002608224674022066,
"loss": 2.5698,
"theoretical_loss": 3.4969654668741894,
"tokens_seen": 1596260352
},
{
"epoch": 0.48,
"learning_rate": 0.0002608124373119358,
"loss": 2.7352,
"theoretical_loss": 3.496952959144603,
"tokens_seen": 1596325888
},
{
"epoch": 0.48,
"learning_rate": 0.000260802407221665,
"loss": 2.6666,
"theoretical_loss": 3.496940452072274,
"tokens_seen": 1596391424
},
{
"epoch": 0.48,
"learning_rate": 0.0002607923771313942,
"loss": 2.7448,
"theoretical_loss": 3.4969279456571405,
"tokens_seen": 1596456960
},
{
"epoch": 0.48,
"learning_rate": 0.00026078234704112336,
"loss": 2.5432,
"theoretical_loss": 3.4969154398991416,
"tokens_seen": 1596522496
},
{
"epoch": 0.48,
"learning_rate": 0.0002607723169508526,
"loss": 2.5906,
"theoretical_loss": 3.4969029347982152,
"tokens_seen": 1596588032
},
{
"epoch": 0.48,
"learning_rate": 0.0002607622868605817,
"loss": 2.8086,
"theoretical_loss": 3.4968904303543002,
"tokens_seen": 1596653568
},
{
"epoch": 0.48,
"learning_rate": 0.00026075225677031096,
"loss": 2.7998,
"theoretical_loss": 3.4968779265673353,
"tokens_seen": 1596719104
},
{
"epoch": 0.48,
"learning_rate": 0.0002607422266800401,
"loss": 2.8203,
"theoretical_loss": 3.4968654234372583,
"tokens_seen": 1596784640
},
{
"epoch": 0.48,
"learning_rate": 0.0002607321965897693,
"loss": 2.6637,
"theoretical_loss": 3.4968529209640087,
"tokens_seen": 1596850176
},
{
"epoch": 0.48,
"learning_rate": 0.0002607221664994985,
"loss": 2.9051,
"theoretical_loss": 3.4968404191475235,
"tokens_seen": 1596915712
},
{
"epoch": 0.48,
"learning_rate": 0.0002607121364092277,
"loss": 2.8052,
"theoretical_loss": 3.4968279179877433,
"tokens_seen": 1596981248
},
{
"epoch": 0.48,
"learning_rate": 0.00026070210631895687,
"loss": 2.6506,
"theoretical_loss": 3.496815417484605,
"tokens_seen": 1597046784
},
{
"epoch": 0.48,
"learning_rate": 0.00026069207622868605,
"loss": 2.7347,
"theoretical_loss": 3.496802917638048,
"tokens_seen": 1597112320
},
{
"epoch": 0.48,
"learning_rate": 0.00026068204613841523,
"loss": 2.8028,
"theoretical_loss": 3.4967904184480103,
"tokens_seen": 1597177856
},
{
"epoch": 0.48,
"learning_rate": 0.00026067201604814446,
"loss": 2.7946,
"theoretical_loss": 3.4967779199144307,
"tokens_seen": 1597243392
},
{
"epoch": 0.48,
"learning_rate": 0.0002606619859578736,
"loss": 2.6505,
"theoretical_loss": 3.4967654220372486,
"tokens_seen": 1597308928
},
{
"epoch": 0.48,
"learning_rate": 0.0002606519558676028,
"loss": 2.7504,
"theoretical_loss": 3.496752924816401,
"tokens_seen": 1597374464
},
{
"epoch": 0.48,
"objective/train/docs_used": 904558,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3360252380371094,
"objective/train/theoretical_loss": 3.496740428251828,
"objective/train/tokens_used": 1617900000,
"theoretical_loss": 3.496740428251828,
"tokens_seen": 1597440000
},
{
"epoch": 0.48,
"learning_rate": 0.00026064192577733195,
"loss": 2.7599,
"theoretical_loss": 3.496740428251828,
"tokens_seen": 1597440000
},
{
"epoch": 0.48,
"learning_rate": 0.0002606318956870612,
"loss": 2.7066,
"theoretical_loss": 3.4967279323434677,
"tokens_seen": 1597505536
},
{
"epoch": 0.48,
"learning_rate": 0.00026062186559679037,
"loss": 2.7552,
"theoretical_loss": 3.4967154370912583,
"tokens_seen": 1597571072
},
{
"epoch": 0.48,
"learning_rate": 0.00026061183550651955,
"loss": 2.7956,
"theoretical_loss": 3.4967029424951392,
"tokens_seen": 1597636608
},
{
"epoch": 0.48,
"learning_rate": 0.00026060180541624873,
"loss": 2.7706,
"theoretical_loss": 3.4966904485550483,
"tokens_seen": 1597702144
},
{
"epoch": 0.48,
"learning_rate": 0.00026059177532597797,
"loss": 2.475,
"theoretical_loss": 3.4966779552709246,
"tokens_seen": 1597767680
},
{
"epoch": 0.48,
"learning_rate": 0.0002605817452357071,
"loss": 2.6255,
"theoretical_loss": 3.496665462642707,
"tokens_seen": 1597833216
},
{
"epoch": 0.48,
"learning_rate": 0.00026057171514543633,
"loss": 2.7453,
"theoretical_loss": 3.4966529706703335,
"tokens_seen": 1597898752
},
{
"epoch": 0.48,
"learning_rate": 0.00026056168505516546,
"loss": 2.5997,
"theoretical_loss": 3.4966404793537436,
"tokens_seen": 1597964288
},
{
"epoch": 0.48,
"learning_rate": 0.0002605516549648947,
"loss": 2.7804,
"theoretical_loss": 3.4966279886928753,
"tokens_seen": 1598029824
},
{
"epoch": 0.48,
"learning_rate": 0.00026054162487462387,
"loss": 2.81,
"theoretical_loss": 3.4966154986876674,
"tokens_seen": 1598095360
},
{
"epoch": 0.48,
"learning_rate": 0.00026053159478435305,
"loss": 2.6182,
"theoretical_loss": 3.4966030093380587,
"tokens_seen": 1598160896
},
{
"epoch": 0.48,
"learning_rate": 0.00026052156469408223,
"loss": 2.6605,
"theoretical_loss": 3.496590520643988,
"tokens_seen": 1598226432
},
{
"epoch": 0.48,
"learning_rate": 0.0002605115346038114,
"loss": 2.7948,
"theoretical_loss": 3.496578032605394,
"tokens_seen": 1598291968
},
{
"epoch": 0.48,
"learning_rate": 0.00026050150451354065,
"loss": 2.9201,
"theoretical_loss": 3.4965655452222153,
"tokens_seen": 1598357504
},
{
"epoch": 0.48,
"learning_rate": 0.00026049147442326983,
"loss": 2.6467,
"theoretical_loss": 3.4965530584943907,
"tokens_seen": 1598423040
},
{
"epoch": 0.48,
"learning_rate": 0.000260481444332999,
"loss": 2.6954,
"theoretical_loss": 3.496540572421859,
"tokens_seen": 1598488576
},
{
"epoch": 0.48,
"learning_rate": 0.0002604714142427282,
"loss": 2.5789,
"theoretical_loss": 3.4965280870045587,
"tokens_seen": 1598554112
},
{
"epoch": 0.48,
"learning_rate": 0.00026046138415245743,
"loss": 2.5792,
"theoretical_loss": 3.496515602242429,
"tokens_seen": 1598619648
},
{
"epoch": 0.48,
"learning_rate": 0.00026045135406218656,
"loss": 2.6141,
"theoretical_loss": 3.4965031181354083,
"tokens_seen": 1598685184
},
{
"epoch": 0.48,
"learning_rate": 0.0002604413239719158,
"loss": 2.7123,
"theoretical_loss": 3.4964906346834352,
"tokens_seen": 1598750720
},
{
"epoch": 0.48,
"learning_rate": 0.0002604312938816449,
"loss": 2.8559,
"theoretical_loss": 3.496478151886449,
"tokens_seen": 1598816256
},
{
"epoch": 0.48,
"learning_rate": 0.00026042126379137415,
"loss": 2.7597,
"theoretical_loss": 3.4964656697443885,
"tokens_seen": 1598881792
},
{
"epoch": 0.48,
"learning_rate": 0.00026041123370110333,
"loss": 2.8234,
"theoretical_loss": 3.496453188257192,
"tokens_seen": 1598947328
},
{
"epoch": 0.48,
"learning_rate": 0.0002604012036108325,
"loss": 2.718,
"theoretical_loss": 3.4964407074247985,
"tokens_seen": 1599012864
},
{
"epoch": 0.48,
"objective/train/docs_used": 905731,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6145551204681396,
"objective/train/theoretical_loss": 3.4964282272471467,
"objective/train/tokens_used": 1619538400,
"theoretical_loss": 3.4964282272471467,
"tokens_seen": 1599078400
},
{
"epoch": 0.48,
"learning_rate": 0.0002603911735205617,
"loss": 2.6112,
"theoretical_loss": 3.4964282272471467,
"tokens_seen": 1599078400
},
{
"epoch": 0.48,
"learning_rate": 0.0002603811434302909,
"loss": 2.6037,
"theoretical_loss": 3.4964157477241757,
"tokens_seen": 1599143936
},
{
"epoch": 0.48,
"learning_rate": 0.00026037111334002006,
"loss": 2.9707,
"theoretical_loss": 3.4964032688558246,
"tokens_seen": 1599209472
},
{
"epoch": 0.48,
"learning_rate": 0.0002603610832497493,
"loss": 2.9657,
"theoretical_loss": 3.4963907906420313,
"tokens_seen": 1599275008
},
{
"epoch": 0.48,
"learning_rate": 0.0002603510531594784,
"loss": 2.8833,
"theoretical_loss": 3.4963783130827353,
"tokens_seen": 1599340544
},
{
"epoch": 0.48,
"learning_rate": 0.00026034102306920766,
"loss": 2.8338,
"theoretical_loss": 3.4963658361778753,
"tokens_seen": 1599406080
},
{
"epoch": 0.48,
"learning_rate": 0.0002603309929789368,
"loss": 2.6129,
"theoretical_loss": 3.49635335992739,
"tokens_seen": 1599471616
},
{
"epoch": 0.48,
"learning_rate": 0.000260320962888666,
"loss": 2.5819,
"theoretical_loss": 3.496340884331219,
"tokens_seen": 1599537152
},
{
"epoch": 0.48,
"learning_rate": 0.0002603109327983952,
"loss": 2.81,
"theoretical_loss": 3.4963284093893003,
"tokens_seen": 1599602688
},
{
"epoch": 0.48,
"learning_rate": 0.0002603009027081244,
"loss": 2.7558,
"theoretical_loss": 3.4963159351015736,
"tokens_seen": 1599668224
},
{
"epoch": 0.48,
"learning_rate": 0.00026029087261785356,
"loss": 2.4971,
"theoretical_loss": 3.496303461467977,
"tokens_seen": 1599733760
},
{
"epoch": 0.48,
"learning_rate": 0.0002602808425275828,
"loss": 2.7862,
"theoretical_loss": 3.4962909884884494,
"tokens_seen": 1599799296
},
{
"epoch": 0.48,
"learning_rate": 0.0002602708124373119,
"loss": 2.5784,
"theoretical_loss": 3.4962785161629304,
"tokens_seen": 1599864832
},
{
"epoch": 0.48,
"learning_rate": 0.00026026078234704116,
"loss": 2.7301,
"theoretical_loss": 3.4962660444913585,
"tokens_seen": 1599930368
},
{
"epoch": 0.48,
"learning_rate": 0.0002602507522567703,
"loss": 2.8232,
"theoretical_loss": 3.4962535734736733,
"tokens_seen": 1599995904
},
{
"epoch": 0.48,
"learning_rate": 0.0002602407221664995,
"loss": 2.6771,
"theoretical_loss": 3.496241103109812,
"tokens_seen": 1600061440
},
{
"epoch": 0.48,
"learning_rate": 0.0002602306920762287,
"loss": 2.7798,
"theoretical_loss": 3.4962286333997152,
"tokens_seen": 1600126976
},
{
"epoch": 0.48,
"learning_rate": 0.0002602206619859579,
"loss": 2.4794,
"theoretical_loss": 3.4962161643433216,
"tokens_seen": 1600192512
},
{
"epoch": 0.48,
"learning_rate": 0.00026021063189568707,
"loss": 2.5431,
"theoretical_loss": 3.4962036959405696,
"tokens_seen": 1600258048
},
{
"epoch": 0.48,
"learning_rate": 0.00026020060180541625,
"loss": 2.9577,
"theoretical_loss": 3.4961912281913987,
"tokens_seen": 1600323584
},
{
"epoch": 0.48,
"learning_rate": 0.00026019057171514543,
"loss": 2.6223,
"theoretical_loss": 3.496178761095747,
"tokens_seen": 1600389120
},
{
"epoch": 0.48,
"learning_rate": 0.00026018054162487466,
"loss": 2.8336,
"theoretical_loss": 3.4961662946535546,
"tokens_seen": 1600454656
},
{
"epoch": 0.49,
"learning_rate": 0.0002601705115346038,
"loss": 2.6917,
"theoretical_loss": 3.49615382886476,
"tokens_seen": 1600520192
},
{
"epoch": 0.49,
"learning_rate": 0.000260160481444333,
"loss": 2.7111,
"theoretical_loss": 3.496141363729302,
"tokens_seen": 1600585728
},
{
"epoch": 0.49,
"learning_rate": 0.00026015045135406215,
"loss": 2.4826,
"theoretical_loss": 3.49612889924712,
"tokens_seen": 1600651264
},
{
"epoch": 0.49,
"objective/train/docs_used": 906398,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.681668758392334,
"objective/train/theoretical_loss": 3.496116435418153,
"objective/train/tokens_used": 1621176800,
"theoretical_loss": 3.496116435418153,
"tokens_seen": 1600716800
},
{
"epoch": 0.49,
"learning_rate": 0.0002601404212637914,
"loss": 2.68,
"theoretical_loss": 3.496116435418153,
"tokens_seen": 1600716800
},
{
"epoch": 0.49,
"learning_rate": 0.00026013039117352057,
"loss": 2.6763,
"theoretical_loss": 3.4961039722423397,
"tokens_seen": 1600782336
},
{
"epoch": 0.49,
"learning_rate": 0.00026012036108324975,
"loss": 2.7982,
"theoretical_loss": 3.4960915097196192,
"tokens_seen": 1600847872
},
{
"epoch": 0.49,
"learning_rate": 0.00026011033099297893,
"loss": 2.6727,
"theoretical_loss": 3.49607904784993,
"tokens_seen": 1600913408
},
{
"epoch": 0.49,
"learning_rate": 0.00026010030090270817,
"loss": 2.6424,
"theoretical_loss": 3.496066586633213,
"tokens_seen": 1600978944
},
{
"epoch": 0.49,
"learning_rate": 0.0002600902708124373,
"loss": 2.6613,
"theoretical_loss": 3.4960541260694056,
"tokens_seen": 1601044480
},
{
"epoch": 0.49,
"learning_rate": 0.00026008024072216653,
"loss": 2.8658,
"theoretical_loss": 3.496041666158447,
"tokens_seen": 1601110016
},
{
"epoch": 0.49,
"learning_rate": 0.00026007021063189566,
"loss": 2.7089,
"theoretical_loss": 3.4960292069002765,
"tokens_seen": 1601175552
},
{
"epoch": 0.49,
"learning_rate": 0.0002600601805416249,
"loss": 2.7971,
"theoretical_loss": 3.496016748294834,
"tokens_seen": 1601241088
},
{
"epoch": 0.49,
"learning_rate": 0.00026005015045135407,
"loss": 2.8451,
"theoretical_loss": 3.496004290342057,
"tokens_seen": 1601306624
},
{
"epoch": 0.49,
"learning_rate": 0.00026004012036108325,
"loss": 2.8354,
"theoretical_loss": 3.495991833041886,
"tokens_seen": 1601372160
},
{
"epoch": 0.49,
"learning_rate": 0.00026003009027081243,
"loss": 2.6351,
"theoretical_loss": 3.49597937639426,
"tokens_seen": 1601437696
},
{
"epoch": 0.49,
"learning_rate": 0.0002600200601805416,
"loss": 2.7204,
"theoretical_loss": 3.495966920399117,
"tokens_seen": 1601503232
},
{
"epoch": 0.49,
"learning_rate": 0.0002600100300902708,
"loss": 2.814,
"theoretical_loss": 3.495954465056397,
"tokens_seen": 1601568768
},
{
"epoch": 0.49,
"learning_rate": 0.00026000000000000003,
"loss": 2.7127,
"theoretical_loss": 3.495942010366039,
"tokens_seen": 1601634304
},
{
"epoch": 0.49,
"learning_rate": 0.00025998996990972916,
"loss": 2.5944,
"theoretical_loss": 3.4959295563279817,
"tokens_seen": 1601699840
},
{
"epoch": 0.49,
"learning_rate": 0.0002599799398194584,
"loss": 2.7922,
"theoretical_loss": 3.495917102942165,
"tokens_seen": 1601765376
},
{
"epoch": 0.49,
"learning_rate": 0.0002599699097291876,
"loss": 2.7448,
"theoretical_loss": 3.4959046502085274,
"tokens_seen": 1601830912
},
{
"epoch": 0.49,
"learning_rate": 0.00025995987963891676,
"loss": 2.6806,
"theoretical_loss": 3.4958921981270086,
"tokens_seen": 1601896448
},
{
"epoch": 0.49,
"learning_rate": 0.00025994984954864594,
"loss": 2.9497,
"theoretical_loss": 3.4958797466975478,
"tokens_seen": 1601961984
},
{
"epoch": 0.49,
"learning_rate": 0.0002599398194583751,
"loss": 2.9598,
"theoretical_loss": 3.4958672959200836,
"tokens_seen": 1602027520
},
{
"epoch": 0.49,
"learning_rate": 0.0002599297893681043,
"loss": 2.6682,
"theoretical_loss": 3.4958548457945553,
"tokens_seen": 1602093056
},
{
"epoch": 0.49,
"learning_rate": 0.00025991975927783353,
"loss": 2.7328,
"theoretical_loss": 3.495842396320903,
"tokens_seen": 1602158592
},
{
"epoch": 0.49,
"learning_rate": 0.00025990972918756266,
"loss": 2.558,
"theoretical_loss": 3.4958299474990646,
"tokens_seen": 1602224128
},
{
"epoch": 0.49,
"learning_rate": 0.0002598996990972919,
"loss": 2.7215,
"theoretical_loss": 3.4958174993289797,
"tokens_seen": 1602289664
},
{
"epoch": 0.49,
"objective/train/docs_used": 907690,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6551146507263184,
"objective/train/theoretical_loss": 3.495805051810588,
"objective/train/tokens_used": 1622815200,
"theoretical_loss": 3.495805051810588,
"tokens_seen": 1602355200
},
{
"epoch": 0.49,
"learning_rate": 0.000259889669007021,
"loss": 2.7185,
"theoretical_loss": 3.495805051810588,
"tokens_seen": 1602355200
},
{
"epoch": 0.49,
"learning_rate": 0.00025987963891675026,
"loss": 2.7435,
"theoretical_loss": 3.4957926049438286,
"tokens_seen": 1602420736
},
{
"epoch": 0.49,
"learning_rate": 0.00025986960882647944,
"loss": 2.8299,
"theoretical_loss": 3.495780158728641,
"tokens_seen": 1602486272
},
{
"epoch": 0.49,
"learning_rate": 0.0002598595787362086,
"loss": 2.7299,
"theoretical_loss": 3.4957677131649634,
"tokens_seen": 1602551808
},
{
"epoch": 0.49,
"learning_rate": 0.0002598495486459378,
"loss": 2.4112,
"theoretical_loss": 3.4957552682527355,
"tokens_seen": 1602617344
},
{
"epoch": 0.49,
"learning_rate": 0.000259839518555667,
"loss": 2.7915,
"theoretical_loss": 3.4957428239918973,
"tokens_seen": 1602682880
},
{
"epoch": 0.49,
"learning_rate": 0.00025982948846539616,
"loss": 2.541,
"theoretical_loss": 3.4957303803823874,
"tokens_seen": 1602748416
},
{
"epoch": 0.49,
"learning_rate": 0.0002598194583751254,
"loss": 2.6613,
"theoretical_loss": 3.4957179374241454,
"tokens_seen": 1602813952
},
{
"epoch": 0.49,
"learning_rate": 0.0002598094282848545,
"loss": 2.6942,
"theoretical_loss": 3.49570549511711,
"tokens_seen": 1602879488
},
{
"epoch": 0.49,
"learning_rate": 0.00025979939819458376,
"loss": 2.8295,
"theoretical_loss": 3.4956930534612214,
"tokens_seen": 1602945024
},
{
"epoch": 0.49,
"learning_rate": 0.00025978936810431294,
"loss": 2.5928,
"theoretical_loss": 3.495680612456418,
"tokens_seen": 1603010560
},
{
"epoch": 0.49,
"learning_rate": 0.0002597793380140421,
"loss": 2.6078,
"theoretical_loss": 3.49566817210264,
"tokens_seen": 1603076096
},
{
"epoch": 0.49,
"learning_rate": 0.0002597693079237713,
"loss": 2.7019,
"theoretical_loss": 3.4956557323998254,
"tokens_seen": 1603141632
},
{
"epoch": 0.49,
"learning_rate": 0.0002597592778335005,
"loss": 2.6193,
"theoretical_loss": 3.495643293347915,
"tokens_seen": 1603207168
},
{
"epoch": 0.49,
"learning_rate": 0.0002597492477432297,
"loss": 2.7063,
"theoretical_loss": 3.4956308549468478,
"tokens_seen": 1603272704
},
{
"epoch": 0.49,
"learning_rate": 0.0002597392176529589,
"loss": 2.9676,
"theoretical_loss": 3.495618417196562,
"tokens_seen": 1603338240
},
{
"epoch": 0.49,
"learning_rate": 0.0002597291875626881,
"loss": 2.7352,
"theoretical_loss": 3.4956059800969985,
"tokens_seen": 1603403776
},
{
"epoch": 0.49,
"learning_rate": 0.00025971915747241727,
"loss": 2.3975,
"theoretical_loss": 3.4955935436480954,
"tokens_seen": 1603469312
},
{
"epoch": 0.49,
"learning_rate": 0.00025970912738214645,
"loss": 2.6224,
"theoretical_loss": 3.4955811078497927,
"tokens_seen": 1603534848
},
{
"epoch": 0.49,
"learning_rate": 0.00025969909729187563,
"loss": 2.7455,
"theoretical_loss": 3.49556867270203,
"tokens_seen": 1603600384
},
{
"epoch": 0.49,
"learning_rate": 0.00025968906720160486,
"loss": 2.6403,
"theoretical_loss": 3.4955562382047463,
"tokens_seen": 1603665920
},
{
"epoch": 0.49,
"learning_rate": 0.000259679037111334,
"loss": 2.6406,
"theoretical_loss": 3.4955438043578813,
"tokens_seen": 1603731456
},
{
"epoch": 0.49,
"learning_rate": 0.0002596690070210632,
"loss": 2.8513,
"theoretical_loss": 3.495531371161374,
"tokens_seen": 1603796992
},
{
"epoch": 0.49,
"learning_rate": 0.00025965897693079235,
"loss": 2.7006,
"theoretical_loss": 3.495518938615164,
"tokens_seen": 1603862528
},
{
"epoch": 0.49,
"learning_rate": 0.0002596489468405216,
"loss": 2.6186,
"theoretical_loss": 3.4955065067191904,
"tokens_seen": 1603928064
},
{
"epoch": 0.49,
"objective/train/docs_used": 908320,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9626288414001465,
"objective/train/theoretical_loss": 3.4954940754733936,
"objective/train/tokens_used": 1624453600,
"theoretical_loss": 3.4954940754733936,
"tokens_seen": 1603993600
},
{
"epoch": 0.49,
"learning_rate": 0.00025963891675025077,
"loss": 2.6275,
"theoretical_loss": 3.4954940754733936,
"tokens_seen": 1603993600
},
{
"epoch": 0.49,
"learning_rate": 0.00025962888665997995,
"loss": 2.5935,
"theoretical_loss": 3.4954816448777115,
"tokens_seen": 1604059136
},
{
"epoch": 0.49,
"learning_rate": 0.00025961885656970913,
"loss": 2.7174,
"theoretical_loss": 3.495469214932085,
"tokens_seen": 1604124672
},
{
"epoch": 0.49,
"learning_rate": 0.00025960882647943837,
"loss": 2.7257,
"theoretical_loss": 3.4954567856364527,
"tokens_seen": 1604190208
},
{
"epoch": 0.49,
"learning_rate": 0.0002595987963891675,
"loss": 2.7825,
"theoretical_loss": 3.4954443569907543,
"tokens_seen": 1604255744
},
{
"epoch": 0.49,
"learning_rate": 0.00025958876629889673,
"loss": 2.7304,
"theoretical_loss": 3.49543192899493,
"tokens_seen": 1604321280
},
{
"epoch": 0.49,
"learning_rate": 0.00025957873620862586,
"loss": 2.6078,
"theoretical_loss": 3.495419501648918,
"tokens_seen": 1604386816
},
{
"epoch": 0.49,
"learning_rate": 0.0002595687061183551,
"loss": 2.6527,
"theoretical_loss": 3.495407074952658,
"tokens_seen": 1604452352
},
{
"epoch": 0.49,
"learning_rate": 0.00025955867602808427,
"loss": 2.597,
"theoretical_loss": 3.4953946489060903,
"tokens_seen": 1604517888
},
{
"epoch": 0.49,
"learning_rate": 0.00025954864593781345,
"loss": 2.6924,
"theoretical_loss": 3.4953822235091536,
"tokens_seen": 1604583424
},
{
"epoch": 0.49,
"learning_rate": 0.00025953861584754263,
"loss": 2.8706,
"theoretical_loss": 3.495369798761788,
"tokens_seen": 1604648960
},
{
"epoch": 0.49,
"learning_rate": 0.0002595285857572718,
"loss": 2.6548,
"theoretical_loss": 3.4953573746639326,
"tokens_seen": 1604714496
},
{
"epoch": 0.49,
"learning_rate": 0.000259518555667001,
"loss": 2.6415,
"theoretical_loss": 3.4953449512155275,
"tokens_seen": 1604780032
},
{
"epoch": 0.49,
"learning_rate": 0.00025950852557673023,
"loss": 2.8359,
"theoretical_loss": 3.4953325284165113,
"tokens_seen": 1604845568
},
{
"epoch": 0.49,
"learning_rate": 0.00025949849548645936,
"loss": 2.6124,
"theoretical_loss": 3.495320106266824,
"tokens_seen": 1604911104
},
{
"epoch": 0.49,
"learning_rate": 0.0002594884653961886,
"loss": 2.8204,
"theoretical_loss": 3.4953076847664057,
"tokens_seen": 1604976640
},
{
"epoch": 0.49,
"learning_rate": 0.0002594784353059178,
"loss": 2.5975,
"theoretical_loss": 3.495295263915195,
"tokens_seen": 1605042176
},
{
"epoch": 0.49,
"learning_rate": 0.00025946840521564696,
"loss": 2.6757,
"theoretical_loss": 3.4952828437131322,
"tokens_seen": 1605107712
},
{
"epoch": 0.49,
"learning_rate": 0.00025945837512537614,
"loss": 2.9625,
"theoretical_loss": 3.4952704241601564,
"tokens_seen": 1605173248
},
{
"epoch": 0.49,
"learning_rate": 0.0002594483450351053,
"loss": 2.6808,
"theoretical_loss": 3.4952580052562077,
"tokens_seen": 1605238784
},
{
"epoch": 0.49,
"learning_rate": 0.0002594383149448345,
"loss": 2.6004,
"theoretical_loss": 3.4952455870012247,
"tokens_seen": 1605304320
},
{
"epoch": 0.49,
"learning_rate": 0.00025942828485456373,
"loss": 2.6937,
"theoretical_loss": 3.495233169395148,
"tokens_seen": 1605369856
},
{
"epoch": 0.49,
"learning_rate": 0.00025941825476429286,
"loss": 2.3482,
"theoretical_loss": 3.4952207524379166,
"tokens_seen": 1605435392
},
{
"epoch": 0.49,
"learning_rate": 0.0002594082246740221,
"loss": 2.7424,
"theoretical_loss": 3.495208336129471,
"tokens_seen": 1605500928
},
{
"epoch": 0.49,
"learning_rate": 0.0002593981945837512,
"loss": 2.8327,
"theoretical_loss": 3.4951959204697496,
"tokens_seen": 1605566464
},
{
"epoch": 0.49,
"objective/train/docs_used": 909776,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.91549015045166,
"objective/train/theoretical_loss": 3.4951835054586926,
"objective/train/tokens_used": 1626092000,
"theoretical_loss": 3.4951835054586926,
"tokens_seen": 1605632000
},
{
"epoch": 0.49,
"learning_rate": 0.00025938816449348046,
"loss": 2.6848,
"theoretical_loss": 3.4951835054586926,
"tokens_seen": 1605632000
},
{
"epoch": 0.49,
"learning_rate": 0.00025937813440320964,
"loss": 2.8588,
"theoretical_loss": 3.49517109109624,
"tokens_seen": 1605697536
},
{
"epoch": 0.49,
"learning_rate": 0.0002593681043129388,
"loss": 2.5755,
"theoretical_loss": 3.4951586773823307,
"tokens_seen": 1605763072
},
{
"epoch": 0.49,
"learning_rate": 0.000259358074222668,
"loss": 2.6378,
"theoretical_loss": 3.4951462643169053,
"tokens_seen": 1605828608
},
{
"epoch": 0.49,
"learning_rate": 0.0002593480441323972,
"loss": 2.5413,
"theoretical_loss": 3.4951338518999027,
"tokens_seen": 1605894144
},
{
"epoch": 0.49,
"learning_rate": 0.00025933801404212636,
"loss": 2.5055,
"theoretical_loss": 3.4951214401312622,
"tokens_seen": 1605959680
},
{
"epoch": 0.49,
"learning_rate": 0.0002593279839518556,
"loss": 2.7116,
"theoretical_loss": 3.4951090290109246,
"tokens_seen": 1606025216
},
{
"epoch": 0.49,
"learning_rate": 0.0002593179538615847,
"loss": 2.6358,
"theoretical_loss": 3.495096618538829,
"tokens_seen": 1606090752
},
{
"epoch": 0.49,
"learning_rate": 0.00025930792377131396,
"loss": 2.5851,
"theoretical_loss": 3.495084208714915,
"tokens_seen": 1606156288
},
{
"epoch": 0.49,
"learning_rate": 0.00025929789368104314,
"loss": 2.6771,
"theoretical_loss": 3.4950717995391223,
"tokens_seen": 1606221824
},
{
"epoch": 0.49,
"learning_rate": 0.0002592878635907723,
"loss": 2.7386,
"theoretical_loss": 3.4950593910113907,
"tokens_seen": 1606287360
},
{
"epoch": 0.49,
"learning_rate": 0.0002592778335005015,
"loss": 2.8324,
"theoretical_loss": 3.49504698313166,
"tokens_seen": 1606352896
},
{
"epoch": 0.49,
"learning_rate": 0.0002592678034102307,
"loss": 2.533,
"theoretical_loss": 3.4950345758998704,
"tokens_seen": 1606418432
},
{
"epoch": 0.49,
"learning_rate": 0.00025925777331995987,
"loss": 2.7091,
"theoretical_loss": 3.4950221693159604,
"tokens_seen": 1606483968
},
{
"epoch": 0.49,
"learning_rate": 0.0002592477432296891,
"loss": 2.6001,
"theoretical_loss": 3.495009763379871,
"tokens_seen": 1606549504
},
{
"epoch": 0.49,
"learning_rate": 0.00025923771313941823,
"loss": 2.694,
"theoretical_loss": 3.494997358091541,
"tokens_seen": 1606615040
},
{
"epoch": 0.49,
"learning_rate": 0.00025922768304914747,
"loss": 2.6026,
"theoretical_loss": 3.4949849534509108,
"tokens_seen": 1606680576
},
{
"epoch": 0.49,
"learning_rate": 0.0002592176529588766,
"loss": 2.8924,
"theoretical_loss": 3.49497254945792,
"tokens_seen": 1606746112
},
{
"epoch": 0.49,
"learning_rate": 0.00025920762286860583,
"loss": 2.7433,
"theoretical_loss": 3.4949601461125077,
"tokens_seen": 1606811648
},
{
"epoch": 0.49,
"learning_rate": 0.000259197592778335,
"loss": 2.6917,
"theoretical_loss": 3.494947743414615,
"tokens_seen": 1606877184
},
{
"epoch": 0.49,
"learning_rate": 0.0002591875626880642,
"loss": 2.6314,
"theoretical_loss": 3.4949353413641804,
"tokens_seen": 1606942720
},
{
"epoch": 0.49,
"learning_rate": 0.00025917753259779337,
"loss": 2.4699,
"theoretical_loss": 3.4949229399611443,
"tokens_seen": 1607008256
},
{
"epoch": 0.49,
"learning_rate": 0.00025916750250752255,
"loss": 2.591,
"theoretical_loss": 3.494910539205447,
"tokens_seen": 1607073792
},
{
"epoch": 0.49,
"learning_rate": 0.00025915747241725173,
"loss": 2.6776,
"theoretical_loss": 3.4948981390970273,
"tokens_seen": 1607139328
},
{
"epoch": 0.49,
"learning_rate": 0.00025914744232698097,
"loss": 2.8879,
"theoretical_loss": 3.4948857396358255,
"tokens_seen": 1607204864
},
{
"epoch": 0.49,
"objective/train/docs_used": 910493,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4678938388824463,
"objective/train/theoretical_loss": 3.4948733408217816,
"objective/train/tokens_used": 1627730400,
"theoretical_loss": 3.4948733408217816,
"tokens_seen": 1607270400
},
{
"epoch": 0.49,
"learning_rate": 0.0002591374122367101,
"loss": 2.5657,
"theoretical_loss": 3.4948733408217816,
"tokens_seen": 1607270400
},
{
"epoch": 0.49,
"learning_rate": 0.00025912738214643933,
"loss": 2.4477,
"theoretical_loss": 3.494860942654835,
"tokens_seen": 1607335936
},
{
"epoch": 0.49,
"learning_rate": 0.0002591173520561685,
"loss": 2.7826,
"theoretical_loss": 3.4948485451349263,
"tokens_seen": 1607401472
},
{
"epoch": 0.49,
"learning_rate": 0.0002591073219658977,
"loss": 2.7875,
"theoretical_loss": 3.494836148261995,
"tokens_seen": 1607467008
},
{
"epoch": 0.49,
"learning_rate": 0.0002590972918756269,
"loss": 2.4798,
"theoretical_loss": 3.4948237520359804,
"tokens_seen": 1607532544
},
{
"epoch": 0.49,
"learning_rate": 0.00025908726178535606,
"loss": 2.695,
"theoretical_loss": 3.4948113564568226,
"tokens_seen": 1607598080
},
{
"epoch": 0.49,
"learning_rate": 0.00025907723169508524,
"loss": 2.5153,
"theoretical_loss": 3.494798961524462,
"tokens_seen": 1607663616
},
{
"epoch": 0.49,
"learning_rate": 0.00025906720160481447,
"loss": 2.6589,
"theoretical_loss": 3.4947865672388376,
"tokens_seen": 1607729152
},
{
"epoch": 0.49,
"learning_rate": 0.0002590571715145436,
"loss": 2.9844,
"theoretical_loss": 3.4947741735998905,
"tokens_seen": 1607794688
},
{
"epoch": 0.49,
"learning_rate": 0.00025904714142427283,
"loss": 2.8268,
"theoretical_loss": 3.49476178060756,
"tokens_seen": 1607860224
},
{
"epoch": 0.49,
"learning_rate": 0.00025903711133400196,
"loss": 2.7809,
"theoretical_loss": 3.494749388261786,
"tokens_seen": 1607925760
},
{
"epoch": 0.49,
"learning_rate": 0.0002590270812437312,
"loss": 2.6381,
"theoretical_loss": 3.4947369965625077,
"tokens_seen": 1607991296
},
{
"epoch": 0.49,
"learning_rate": 0.0002590170511534604,
"loss": 2.6719,
"theoretical_loss": 3.494724605509666,
"tokens_seen": 1608056832
},
{
"epoch": 0.49,
"learning_rate": 0.00025900702106318956,
"loss": 2.7301,
"theoretical_loss": 3.494712215103201,
"tokens_seen": 1608122368
},
{
"epoch": 0.49,
"learning_rate": 0.0002589969909729188,
"loss": 2.8123,
"theoretical_loss": 3.494699825343052,
"tokens_seen": 1608187904
},
{
"epoch": 0.49,
"learning_rate": 0.000258986960882648,
"loss": 2.6621,
"theoretical_loss": 3.494687436229159,
"tokens_seen": 1608253440
},
{
"epoch": 0.49,
"learning_rate": 0.00025897693079237716,
"loss": 2.7921,
"theoretical_loss": 3.4946750477614623,
"tokens_seen": 1608318976
},
{
"epoch": 0.49,
"learning_rate": 0.00025896690070210634,
"loss": 2.8875,
"theoretical_loss": 3.494662659939901,
"tokens_seen": 1608384512
},
{
"epoch": 0.49,
"learning_rate": 0.0002589568706118355,
"loss": 2.6858,
"theoretical_loss": 3.4946502727644164,
"tokens_seen": 1608450048
},
{
"epoch": 0.49,
"learning_rate": 0.0002589468405215647,
"loss": 2.7549,
"theoretical_loss": 3.4946378862349476,
"tokens_seen": 1608515584
},
{
"epoch": 0.49,
"learning_rate": 0.00025893681043129393,
"loss": 2.6908,
"theoretical_loss": 3.494625500351435,
"tokens_seen": 1608581120
},
{
"epoch": 0.49,
"learning_rate": 0.00025892678034102306,
"loss": 2.7308,
"theoretical_loss": 3.4946131151138182,
"tokens_seen": 1608646656
},
{
"epoch": 0.49,
"learning_rate": 0.0002589167502507523,
"loss": 2.5994,
"theoretical_loss": 3.4946007305220377,
"tokens_seen": 1608712192
},
{
"epoch": 0.49,
"learning_rate": 0.0002589067201604814,
"loss": 2.8496,
"theoretical_loss": 3.494588346576033,
"tokens_seen": 1608777728
},
{
"epoch": 0.49,
"learning_rate": 0.00025889669007021066,
"loss": 2.7648,
"theoretical_loss": 3.4945759632757443,
"tokens_seen": 1608843264
},
{
"epoch": 0.49,
"objective/train/docs_used": 911872,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3492157459259033,
"objective/train/theoretical_loss": 3.4945635806211115,
"objective/train/tokens_used": 1629368800,
"theoretical_loss": 3.4945635806211115,
"tokens_seen": 1608908800
},
{
"epoch": 0.49,
"learning_rate": 0.00025888665997993984,
"loss": 2.4903,
"theoretical_loss": 3.4945635806211115,
"tokens_seen": 1608908800
},
{
"epoch": 0.49,
"learning_rate": 0.000258876629889669,
"loss": 2.5681,
"theoretical_loss": 3.494551198612075,
"tokens_seen": 1608974336
},
{
"epoch": 0.49,
"learning_rate": 0.0002588665997993982,
"loss": 2.4078,
"theoretical_loss": 3.494538817248575,
"tokens_seen": 1609039872
},
{
"epoch": 0.49,
"learning_rate": 0.0002588565697091274,
"loss": 2.6241,
"theoretical_loss": 3.494526436530551,
"tokens_seen": 1609105408
},
{
"epoch": 0.49,
"learning_rate": 0.00025884653961885657,
"loss": 2.5489,
"theoretical_loss": 3.494514056457943,
"tokens_seen": 1609170944
},
{
"epoch": 0.49,
"learning_rate": 0.0002588365095285858,
"loss": 2.8086,
"theoretical_loss": 3.4945016770306916,
"tokens_seen": 1609236480
},
{
"epoch": 0.49,
"learning_rate": 0.00025882647943831493,
"loss": 2.727,
"theoretical_loss": 3.4944892982487366,
"tokens_seen": 1609302016
},
{
"epoch": 0.49,
"learning_rate": 0.00025881644934804416,
"loss": 2.5722,
"theoretical_loss": 3.4944769201120183,
"tokens_seen": 1609367552
},
{
"epoch": 0.49,
"learning_rate": 0.00025880641925777334,
"loss": 2.6994,
"theoretical_loss": 3.4944645426204763,
"tokens_seen": 1609433088
},
{
"epoch": 0.49,
"learning_rate": 0.0002587963891675025,
"loss": 2.5787,
"theoretical_loss": 3.4944521657740513,
"tokens_seen": 1609498624
},
{
"epoch": 0.49,
"learning_rate": 0.0002587863590772317,
"loss": 2.6738,
"theoretical_loss": 3.494439789572683,
"tokens_seen": 1609564160
},
{
"epoch": 0.49,
"learning_rate": 0.0002587763289869609,
"loss": 2.5739,
"theoretical_loss": 3.494427414016312,
"tokens_seen": 1609629696
},
{
"epoch": 0.49,
"learning_rate": 0.00025876629889669007,
"loss": 2.6158,
"theoretical_loss": 3.494415039104877,
"tokens_seen": 1609695232
},
{
"epoch": 0.49,
"learning_rate": 0.0002587562688064193,
"loss": 2.6905,
"theoretical_loss": 3.4944026648383204,
"tokens_seen": 1609760768
},
{
"epoch": 0.49,
"learning_rate": 0.00025874623871614843,
"loss": 2.7332,
"theoretical_loss": 3.4943902912165807,
"tokens_seen": 1609826304
},
{
"epoch": 0.49,
"learning_rate": 0.00025873620862587767,
"loss": 2.4391,
"theoretical_loss": 3.494377918239598,
"tokens_seen": 1609891840
},
{
"epoch": 0.49,
"learning_rate": 0.0002587261785356068,
"loss": 2.5857,
"theoretical_loss": 3.4943655459073133,
"tokens_seen": 1609957376
},
{
"epoch": 0.49,
"learning_rate": 0.00025871614844533603,
"loss": 2.7155,
"theoretical_loss": 3.4943531742196665,
"tokens_seen": 1610022912
},
{
"epoch": 0.49,
"learning_rate": 0.0002587061183550652,
"loss": 2.5551,
"theoretical_loss": 3.4943408031765983,
"tokens_seen": 1610088448
},
{
"epoch": 0.49,
"learning_rate": 0.0002586960882647944,
"loss": 2.8175,
"theoretical_loss": 3.4943284327780475,
"tokens_seen": 1610153984
},
{
"epoch": 0.49,
"learning_rate": 0.00025868605817452357,
"loss": 2.7675,
"theoretical_loss": 3.4943160630239554,
"tokens_seen": 1610219520
},
{
"epoch": 0.49,
"learning_rate": 0.00025867602808425275,
"loss": 2.5961,
"theoretical_loss": 3.4943036939142615,
"tokens_seen": 1610285056
},
{
"epoch": 0.49,
"learning_rate": 0.00025866599799398193,
"loss": 2.72,
"theoretical_loss": 3.4942913254489065,
"tokens_seen": 1610350592
},
{
"epoch": 0.49,
"learning_rate": 0.00025865596790371117,
"loss": 2.7213,
"theoretical_loss": 3.494278957627831,
"tokens_seen": 1610416128
},
{
"epoch": 0.49,
"learning_rate": 0.0002586459378134403,
"loss": 2.4998,
"theoretical_loss": 3.494266590450974,
"tokens_seen": 1610481664
},
{
"epoch": 0.49,
"objective/train/docs_used": 912654,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2332985401153564,
"objective/train/theoretical_loss": 3.494254223918277,
"objective/train/tokens_used": 1631007200,
"theoretical_loss": 3.494254223918277,
"tokens_seen": 1610547200
},
{
"epoch": 0.49,
"learning_rate": 0.00025863590772316953,
"loss": 2.4914,
"theoretical_loss": 3.494254223918277,
"tokens_seen": 1610547200
},
{
"epoch": 0.49,
"learning_rate": 0.0002586258776328987,
"loss": 2.8895,
"theoretical_loss": 3.4942418580296795,
"tokens_seen": 1610612736
},
{
"epoch": 0.49,
"learning_rate": 0.0002586158475426279,
"loss": 2.7,
"theoretical_loss": 3.494229492785122,
"tokens_seen": 1610678272
},
{
"epoch": 0.49,
"learning_rate": 0.0002586058174523571,
"loss": 2.7346,
"theoretical_loss": 3.4942171281845447,
"tokens_seen": 1610743808
},
{
"epoch": 0.49,
"learning_rate": 0.00025859578736208626,
"loss": 2.5939,
"theoretical_loss": 3.494204764227887,
"tokens_seen": 1610809344
},
{
"epoch": 0.49,
"learning_rate": 0.00025858575727181544,
"loss": 2.7047,
"theoretical_loss": 3.4941924009150913,
"tokens_seen": 1610874880
},
{
"epoch": 0.49,
"learning_rate": 0.00025857572718154467,
"loss": 2.9011,
"theoretical_loss": 3.494180038246096,
"tokens_seen": 1610940416
},
{
"epoch": 0.49,
"learning_rate": 0.0002585656970912738,
"loss": 2.7759,
"theoretical_loss": 3.494167676220842,
"tokens_seen": 1611005952
},
{
"epoch": 0.49,
"learning_rate": 0.00025855566700100303,
"loss": 2.7356,
"theoretical_loss": 3.4941553148392694,
"tokens_seen": 1611071488
},
{
"epoch": 0.49,
"learning_rate": 0.00025854563691073216,
"loss": 2.6938,
"theoretical_loss": 3.494142954101319,
"tokens_seen": 1611137024
},
{
"epoch": 0.49,
"learning_rate": 0.0002585356068204614,
"loss": 2.7442,
"theoretical_loss": 3.4941305940069305,
"tokens_seen": 1611202560
},
{
"epoch": 0.49,
"learning_rate": 0.0002585255767301906,
"loss": 2.5258,
"theoretical_loss": 3.494118234556044,
"tokens_seen": 1611268096
},
{
"epoch": 0.49,
"learning_rate": 0.00025851554663991976,
"loss": 2.6695,
"theoretical_loss": 3.4941058757486014,
"tokens_seen": 1611333632
},
{
"epoch": 0.49,
"learning_rate": 0.00025850551654964894,
"loss": 2.6039,
"theoretical_loss": 3.494093517584542,
"tokens_seen": 1611399168
},
{
"epoch": 0.49,
"learning_rate": 0.0002584954864593782,
"loss": 2.7668,
"theoretical_loss": 3.4940811600638053,
"tokens_seen": 1611464704
},
{
"epoch": 0.49,
"learning_rate": 0.0002584854563691073,
"loss": 2.7252,
"theoretical_loss": 3.4940688031863325,
"tokens_seen": 1611530240
},
{
"epoch": 0.49,
"learning_rate": 0.00025847542627883654,
"loss": 2.5721,
"theoretical_loss": 3.4940564469520643,
"tokens_seen": 1611595776
},
{
"epoch": 0.49,
"learning_rate": 0.00025846539618856566,
"loss": 2.8015,
"theoretical_loss": 3.4940440913609407,
"tokens_seen": 1611661312
},
{
"epoch": 0.49,
"learning_rate": 0.0002584553660982949,
"loss": 2.8375,
"theoretical_loss": 3.494031736412902,
"tokens_seen": 1611726848
},
{
"epoch": 0.49,
"learning_rate": 0.0002584453360080241,
"loss": 2.7506,
"theoretical_loss": 3.494019382107888,
"tokens_seen": 1611792384
},
{
"epoch": 0.49,
"learning_rate": 0.00025843530591775326,
"loss": 2.8662,
"theoretical_loss": 3.49400702844584,
"tokens_seen": 1611857920
},
{
"epoch": 0.49,
"learning_rate": 0.00025842527582748244,
"loss": 2.5579,
"theoretical_loss": 3.4939946754266984,
"tokens_seen": 1611923456
},
{
"epoch": 0.49,
"learning_rate": 0.0002584152457372116,
"loss": 2.7099,
"theoretical_loss": 3.4939823230504037,
"tokens_seen": 1611988992
},
{
"epoch": 0.49,
"learning_rate": 0.0002584052156469408,
"loss": 2.4985,
"theoretical_loss": 3.4939699713168952,
"tokens_seen": 1612054528
},
{
"epoch": 0.49,
"learning_rate": 0.00025839518555667004,
"loss": 2.7429,
"theoretical_loss": 3.493957620226115,
"tokens_seen": 1612120064
},
{
"epoch": 0.49,
"objective/train/docs_used": 913987,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.606280565261841,
"objective/train/theoretical_loss": 3.493945269778002,
"objective/train/tokens_used": 1632645600,
"theoretical_loss": 3.493945269778002,
"tokens_seen": 1612185600
},
{
"epoch": 0.49,
"learning_rate": 0.00025838515546639917,
"loss": 2.4393,
"theoretical_loss": 3.493945269778002,
"tokens_seen": 1612185600
},
{
"epoch": 0.49,
"learning_rate": 0.0002583751253761284,
"loss": 2.6648,
"theoretical_loss": 3.4939329199724964,
"tokens_seen": 1612251136
},
{
"epoch": 0.49,
"learning_rate": 0.00025836509528585753,
"loss": 2.9545,
"theoretical_loss": 3.4939205708095407,
"tokens_seen": 1612316672
},
{
"epoch": 0.49,
"learning_rate": 0.00025835506519558677,
"loss": 2.5119,
"theoretical_loss": 3.4939082222890736,
"tokens_seen": 1612382208
},
{
"epoch": 0.49,
"learning_rate": 0.00025834503510531595,
"loss": 2.6821,
"theoretical_loss": 3.493895874411036,
"tokens_seen": 1612447744
},
{
"epoch": 0.49,
"learning_rate": 0.00025833500501504513,
"loss": 2.8393,
"theoretical_loss": 3.4938835271753685,
"tokens_seen": 1612513280
},
{
"epoch": 0.49,
"learning_rate": 0.0002583249749247743,
"loss": 2.7455,
"theoretical_loss": 3.493871180582012,
"tokens_seen": 1612578816
},
{
"epoch": 0.49,
"learning_rate": 0.00025831494483450354,
"loss": 2.5686,
"theoretical_loss": 3.493858834630906,
"tokens_seen": 1612644352
},
{
"epoch": 0.49,
"learning_rate": 0.00025830491474423267,
"loss": 2.6518,
"theoretical_loss": 3.4938464893219914,
"tokens_seen": 1612709888
},
{
"epoch": 0.49,
"learning_rate": 0.0002582948846539619,
"loss": 2.6431,
"theoretical_loss": 3.4938341446552093,
"tokens_seen": 1612775424
},
{
"epoch": 0.49,
"learning_rate": 0.00025828485456369103,
"loss": 2.8072,
"theoretical_loss": 3.4938218006304993,
"tokens_seen": 1612840960
},
{
"epoch": 0.49,
"learning_rate": 0.00025827482447342027,
"loss": 2.5223,
"theoretical_loss": 3.493809457247803,
"tokens_seen": 1612906496
},
{
"epoch": 0.49,
"learning_rate": 0.00025826479438314945,
"loss": 2.7228,
"theoretical_loss": 3.4937971145070597,
"tokens_seen": 1612972032
},
{
"epoch": 0.49,
"learning_rate": 0.00025825476429287863,
"loss": 2.457,
"theoretical_loss": 3.4937847724082105,
"tokens_seen": 1613037568
},
{
"epoch": 0.49,
"learning_rate": 0.00025824473420260787,
"loss": 2.5532,
"theoretical_loss": 3.493772430951196,
"tokens_seen": 1613103104
},
{
"epoch": 0.49,
"learning_rate": 0.000258234704112337,
"loss": 2.7081,
"theoretical_loss": 3.4937600901359565,
"tokens_seen": 1613168640
},
{
"epoch": 0.49,
"learning_rate": 0.00025822467402206623,
"loss": 2.5494,
"theoretical_loss": 3.493747749962433,
"tokens_seen": 1613234176
},
{
"epoch": 0.49,
"learning_rate": 0.0002582146439317954,
"loss": 2.4604,
"theoretical_loss": 3.4937354104305656,
"tokens_seen": 1613299712
},
{
"epoch": 0.49,
"learning_rate": 0.0002582046138415246,
"loss": 2.7003,
"theoretical_loss": 3.4937230715402956,
"tokens_seen": 1613365248
},
{
"epoch": 0.49,
"learning_rate": 0.00025819458375125377,
"loss": 2.6771,
"theoretical_loss": 3.4937107332915622,
"tokens_seen": 1613430784
},
{
"epoch": 0.49,
"learning_rate": 0.00025818455366098295,
"loss": 2.8266,
"theoretical_loss": 3.4936983956843073,
"tokens_seen": 1613496320
},
{
"epoch": 0.49,
"learning_rate": 0.00025817452357071213,
"loss": 2.877,
"theoretical_loss": 3.4936860587184713,
"tokens_seen": 1613561856
},
{
"epoch": 0.49,
"learning_rate": 0.00025816449348044137,
"loss": 2.5594,
"theoretical_loss": 3.4936737223939938,
"tokens_seen": 1613627392
},
{
"epoch": 0.49,
"learning_rate": 0.0002581544633901705,
"loss": 2.726,
"theoretical_loss": 3.4936613867108166,
"tokens_seen": 1613692928
},
{
"epoch": 0.49,
"learning_rate": 0.00025814443329989973,
"loss": 2.5411,
"theoretical_loss": 3.4936490516688803,
"tokens_seen": 1613758464
},
{
"epoch": 0.49,
"objective/train/docs_used": 914763,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.517117977142334,
"objective/train/theoretical_loss": 3.4936367172681244,
"objective/train/tokens_used": 1634284000,
"theoretical_loss": 3.4936367172681244,
"tokens_seen": 1613824000
},
{
"epoch": 0.49,
"learning_rate": 0.0002581344032096289,
"loss": 2.5916,
"theoretical_loss": 3.4936367172681244,
"tokens_seen": 1613824000
},
{
"epoch": 0.49,
"learning_rate": 0.0002581243731193581,
"loss": 2.7426,
"theoretical_loss": 3.4936243835084904,
"tokens_seen": 1613889536
},
{
"epoch": 0.49,
"learning_rate": 0.0002581143430290873,
"loss": 2.5287,
"theoretical_loss": 3.493612050389919,
"tokens_seen": 1613955072
},
{
"epoch": 0.49,
"learning_rate": 0.00025810431293881646,
"loss": 2.5535,
"theoretical_loss": 3.4935997179123506,
"tokens_seen": 1614020608
},
{
"epoch": 0.49,
"learning_rate": 0.00025809428284854564,
"loss": 2.6983,
"theoretical_loss": 3.493587386075726,
"tokens_seen": 1614086144
},
{
"epoch": 0.49,
"learning_rate": 0.00025808425275827487,
"loss": 2.4517,
"theoretical_loss": 3.493575054879985,
"tokens_seen": 1614151680
},
{
"epoch": 0.49,
"learning_rate": 0.000258074222668004,
"loss": 2.5943,
"theoretical_loss": 3.49356272432507,
"tokens_seen": 1614217216
},
{
"epoch": 0.49,
"learning_rate": 0.00025806419257773323,
"loss": 2.7531,
"theoretical_loss": 3.49355039441092,
"tokens_seen": 1614282752
},
{
"epoch": 0.49,
"learning_rate": 0.00025805416248746236,
"loss": 2.6228,
"theoretical_loss": 3.493538065137477,
"tokens_seen": 1614348288
},
{
"epoch": 0.49,
"learning_rate": 0.0002580441323971916,
"loss": 2.7219,
"theoretical_loss": 3.4935257365046803,
"tokens_seen": 1614413824
},
{
"epoch": 0.49,
"learning_rate": 0.0002580341023069208,
"loss": 2.4505,
"theoretical_loss": 3.493513408512472,
"tokens_seen": 1614479360
},
{
"epoch": 0.49,
"learning_rate": 0.00025802407221664996,
"loss": 2.7216,
"theoretical_loss": 3.493501081160792,
"tokens_seen": 1614544896
},
{
"epoch": 0.49,
"learning_rate": 0.00025801404212637914,
"loss": 2.8677,
"theoretical_loss": 3.493488754449581,
"tokens_seen": 1614610432
},
{
"epoch": 0.49,
"learning_rate": 0.0002580040120361084,
"loss": 2.3761,
"theoretical_loss": 3.4934764283787807,
"tokens_seen": 1614675968
},
{
"epoch": 0.49,
"learning_rate": 0.0002579939819458375,
"loss": 2.5361,
"theoretical_loss": 3.493464102948331,
"tokens_seen": 1614741504
},
{
"epoch": 0.49,
"learning_rate": 0.00025798395185556674,
"loss": 2.7253,
"theoretical_loss": 3.493451778158172,
"tokens_seen": 1614807040
},
{
"epoch": 0.49,
"learning_rate": 0.00025797392176529586,
"loss": 2.5595,
"theoretical_loss": 3.493439454008246,
"tokens_seen": 1614872576
},
{
"epoch": 0.49,
"learning_rate": 0.0002579638916750251,
"loss": 2.481,
"theoretical_loss": 3.4934271304984925,
"tokens_seen": 1614938112
},
{
"epoch": 0.49,
"learning_rate": 0.0002579538615847543,
"loss": 2.9287,
"theoretical_loss": 3.493414807628853,
"tokens_seen": 1615003648
},
{
"epoch": 0.49,
"learning_rate": 0.00025794383149448346,
"loss": 2.6319,
"theoretical_loss": 3.493402485399268,
"tokens_seen": 1615069184
},
{
"epoch": 0.49,
"learning_rate": 0.00025793380140421264,
"loss": 2.7209,
"theoretical_loss": 3.493390163809678,
"tokens_seen": 1615134720
},
{
"epoch": 0.49,
"learning_rate": 0.0002579237713139418,
"loss": 2.732,
"theoretical_loss": 3.493377842860024,
"tokens_seen": 1615200256
},
{
"epoch": 0.49,
"learning_rate": 0.000257913741223671,
"loss": 2.5619,
"theoretical_loss": 3.4933655225502473,
"tokens_seen": 1615265792
},
{
"epoch": 0.49,
"learning_rate": 0.00025790371113340024,
"loss": 2.4665,
"theoretical_loss": 3.4933532028802885,
"tokens_seen": 1615331328
},
{
"epoch": 0.49,
"learning_rate": 0.00025789368104312937,
"loss": 2.765,
"theoretical_loss": 3.493340883850088,
"tokens_seen": 1615396864
},
{
"epoch": 0.49,
"objective/train/docs_used": 916021,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8048737049102783,
"objective/train/theoretical_loss": 3.4933285654595867,
"objective/train/tokens_used": 1635922400,
"theoretical_loss": 3.4933285654595867,
"tokens_seen": 1615462400
},
{
"epoch": 0.49,
"learning_rate": 0.0002578836509528586,
"loss": 2.7848,
"theoretical_loss": 3.4933285654595867,
"tokens_seen": 1615462400
},
{
"epoch": 0.49,
"learning_rate": 0.00025787362086258773,
"loss": 2.5147,
"theoretical_loss": 3.4933162477087256,
"tokens_seen": 1615527936
},
{
"epoch": 0.49,
"learning_rate": 0.00025786359077231697,
"loss": 2.6695,
"theoretical_loss": 3.493303930597446,
"tokens_seen": 1615593472
},
{
"epoch": 0.49,
"learning_rate": 0.00025785356068204615,
"loss": 2.6998,
"theoretical_loss": 3.493291614125688,
"tokens_seen": 1615659008
},
{
"epoch": 0.49,
"learning_rate": 0.00025784353059177533,
"loss": 2.6888,
"theoretical_loss": 3.493279298293392,
"tokens_seen": 1615724544
},
{
"epoch": 0.49,
"learning_rate": 0.0002578335005015045,
"loss": 2.6354,
"theoretical_loss": 3.4932669831005008,
"tokens_seen": 1615790080
},
{
"epoch": 0.49,
"learning_rate": 0.00025782347041123374,
"loss": 2.6624,
"theoretical_loss": 3.493254668546953,
"tokens_seen": 1615855616
},
{
"epoch": 0.49,
"learning_rate": 0.00025781344032096287,
"loss": 2.8219,
"theoretical_loss": 3.4932423546326907,
"tokens_seen": 1615921152
},
{
"epoch": 0.49,
"learning_rate": 0.0002578034102306921,
"loss": 2.4463,
"theoretical_loss": 3.4932300413576556,
"tokens_seen": 1615986688
},
{
"epoch": 0.49,
"learning_rate": 0.00025779338014042123,
"loss": 2.6844,
"theoretical_loss": 3.4932177287217865,
"tokens_seen": 1616052224
},
{
"epoch": 0.49,
"learning_rate": 0.00025778335005015047,
"loss": 2.6718,
"theoretical_loss": 3.4932054167250257,
"tokens_seen": 1616117760
},
{
"epoch": 0.49,
"learning_rate": 0.00025777331995987965,
"loss": 2.5324,
"theoretical_loss": 3.4931931053673146,
"tokens_seen": 1616183296
},
{
"epoch": 0.49,
"learning_rate": 0.00025776328986960883,
"loss": 2.3974,
"theoretical_loss": 3.4931807946485924,
"tokens_seen": 1616248832
},
{
"epoch": 0.49,
"learning_rate": 0.000257753259779338,
"loss": 2.5832,
"theoretical_loss": 3.4931684845688014,
"tokens_seen": 1616314368
},
{
"epoch": 0.49,
"learning_rate": 0.0002577432296890672,
"loss": 2.5867,
"theoretical_loss": 3.4931561751278823,
"tokens_seen": 1616379904
},
{
"epoch": 0.49,
"learning_rate": 0.0002577331995987964,
"loss": 2.6487,
"theoretical_loss": 3.4931438663257754,
"tokens_seen": 1616445440
},
{
"epoch": 0.49,
"learning_rate": 0.0002577231695085256,
"loss": 2.7597,
"theoretical_loss": 3.4931315581624225,
"tokens_seen": 1616510976
},
{
"epoch": 0.49,
"learning_rate": 0.00025771313941825474,
"loss": 2.8443,
"theoretical_loss": 3.493119250637764,
"tokens_seen": 1616576512
},
{
"epoch": 0.49,
"learning_rate": 0.00025770310932798397,
"loss": 2.7674,
"theoretical_loss": 3.493106943751741,
"tokens_seen": 1616642048
},
{
"epoch": 0.49,
"learning_rate": 0.0002576930792377131,
"loss": 2.6171,
"theoretical_loss": 3.493094637504295,
"tokens_seen": 1616707584
},
{
"epoch": 0.49,
"learning_rate": 0.00025768304914744233,
"loss": 2.6201,
"theoretical_loss": 3.4930823318953657,
"tokens_seen": 1616773120
},
{
"epoch": 0.49,
"learning_rate": 0.0002576730190571715,
"loss": 2.7037,
"theoretical_loss": 3.4930700269248955,
"tokens_seen": 1616838656
},
{
"epoch": 0.49,
"learning_rate": 0.0002576629889669007,
"loss": 2.8793,
"theoretical_loss": 3.493057722592824,
"tokens_seen": 1616904192
},
{
"epoch": 0.49,
"learning_rate": 0.0002576529588766299,
"loss": 2.3839,
"theoretical_loss": 3.493045418899094,
"tokens_seen": 1616969728
},
{
"epoch": 0.49,
"learning_rate": 0.0002576429287863591,
"loss": 2.6682,
"theoretical_loss": 3.493033115843645,
"tokens_seen": 1617035264
},
{
"epoch": 0.49,
"objective/train/docs_used": 916782,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2013497352600098,
"objective/train/theoretical_loss": 3.4930208134264182,
"objective/train/tokens_used": 1637560800,
"theoretical_loss": 3.4930208134264182,
"tokens_seen": 1617100800
},
{
"epoch": 0.49,
"learning_rate": 0.00025763289869608824,
"loss": 2.5827,
"theoretical_loss": 3.4930208134264182,
"tokens_seen": 1617100800
},
{
"epoch": 0.49,
"learning_rate": 0.0002576228686058175,
"loss": 2.5014,
"theoretical_loss": 3.4930085116473553,
"tokens_seen": 1617166336
},
{
"epoch": 0.49,
"learning_rate": 0.0002576128385155466,
"loss": 2.5427,
"theoretical_loss": 3.492996210506397,
"tokens_seen": 1617231872
},
{
"epoch": 0.49,
"learning_rate": 0.00025760280842527584,
"loss": 2.7117,
"theoretical_loss": 3.4929839100034847,
"tokens_seen": 1617297408
},
{
"epoch": 0.49,
"learning_rate": 0.000257592778335005,
"loss": 2.6581,
"theoretical_loss": 3.4929716101385586,
"tokens_seen": 1617362944
},
{
"epoch": 0.49,
"learning_rate": 0.0002575827482447342,
"loss": 2.4765,
"theoretical_loss": 3.4929593109115604,
"tokens_seen": 1617428480
},
{
"epoch": 0.49,
"learning_rate": 0.0002575727181544634,
"loss": 2.5455,
"theoretical_loss": 3.492947012322431,
"tokens_seen": 1617494016
},
{
"epoch": 0.49,
"learning_rate": 0.00025756268806419256,
"loss": 2.6912,
"theoretical_loss": 3.4929347143711116,
"tokens_seen": 1617559552
},
{
"epoch": 0.49,
"learning_rate": 0.00025755265797392174,
"loss": 2.9282,
"theoretical_loss": 3.492922417057543,
"tokens_seen": 1617625088
},
{
"epoch": 0.49,
"learning_rate": 0.000257542627883651,
"loss": 2.7873,
"theoretical_loss": 3.4929101203816666,
"tokens_seen": 1617690624
},
{
"epoch": 0.49,
"learning_rate": 0.0002575325977933801,
"loss": 2.8849,
"theoretical_loss": 3.492897824343423,
"tokens_seen": 1617756160
},
{
"epoch": 0.49,
"learning_rate": 0.00025752256770310934,
"loss": 2.5545,
"theoretical_loss": 3.492885528942754,
"tokens_seen": 1617821696
},
{
"epoch": 0.49,
"learning_rate": 0.00025751253761283847,
"loss": 2.8906,
"theoretical_loss": 3.4928732341796005,
"tokens_seen": 1617887232
},
{
"epoch": 0.49,
"learning_rate": 0.0002575025075225677,
"loss": 2.734,
"theoretical_loss": 3.4928609400539035,
"tokens_seen": 1617952768
},
{
"epoch": 0.49,
"learning_rate": 0.00025749247743229694,
"loss": 2.6884,
"theoretical_loss": 3.4928486465656037,
"tokens_seen": 1618018304
},
{
"epoch": 0.49,
"learning_rate": 0.00025748244734202606,
"loss": 2.6138,
"theoretical_loss": 3.492836353714643,
"tokens_seen": 1618083840
},
{
"epoch": 0.49,
"learning_rate": 0.0002574724172517553,
"loss": 2.6093,
"theoretical_loss": 3.4928240615009623,
"tokens_seen": 1618149376
},
{
"epoch": 0.49,
"learning_rate": 0.0002574623871614845,
"loss": 2.7469,
"theoretical_loss": 3.4928117699245025,
"tokens_seen": 1618214912
},
{
"epoch": 0.49,
"learning_rate": 0.00025745235707121366,
"loss": 2.6516,
"theoretical_loss": 3.492799478985205,
"tokens_seen": 1618280448
},
{
"epoch": 0.49,
"learning_rate": 0.00025744232698094284,
"loss": 2.7308,
"theoretical_loss": 3.492787188683011,
"tokens_seen": 1618345984
},
{
"epoch": 0.49,
"learning_rate": 0.000257432296890672,
"loss": 2.5253,
"theoretical_loss": 3.4927748990178613,
"tokens_seen": 1618411520
},
{
"epoch": 0.49,
"learning_rate": 0.0002574222668004012,
"loss": 2.5673,
"theoretical_loss": 3.492762609989698,
"tokens_seen": 1618477056
},
{
"epoch": 0.49,
"learning_rate": 0.00025741223671013044,
"loss": 2.6626,
"theoretical_loss": 3.4927503215984608,
"tokens_seen": 1618542592
},
{
"epoch": 0.49,
"learning_rate": 0.00025740220661985957,
"loss": 2.5267,
"theoretical_loss": 3.492738033844092,
"tokens_seen": 1618608128
},
{
"epoch": 0.49,
"learning_rate": 0.0002573921765295888,
"loss": 2.5108,
"theoretical_loss": 3.492725746726533,
"tokens_seen": 1618673664
},
{
"epoch": 0.49,
"objective/train/docs_used": 918149,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6362714767456055,
"objective/train/theoretical_loss": 3.4927134602457244,
"objective/train/tokens_used": 1639199200,
"theoretical_loss": 3.4927134602457244,
"tokens_seen": 1618739200
},
{
"epoch": 0.49,
"learning_rate": 0.00025738214643931793,
"loss": 2.8211,
"theoretical_loss": 3.4927134602457244,
"tokens_seen": 1618739200
},
{
"epoch": 0.49,
"learning_rate": 0.00025737211634904717,
"loss": 2.5663,
"theoretical_loss": 3.492701174401608,
"tokens_seen": 1618804736
},
{
"epoch": 0.49,
"learning_rate": 0.00025736208625877635,
"loss": 2.6966,
"theoretical_loss": 3.492688889194124,
"tokens_seen": 1618870272
},
{
"epoch": 0.49,
"learning_rate": 0.00025735205616850553,
"loss": 2.7324,
"theoretical_loss": 3.492676604623215,
"tokens_seen": 1618935808
},
{
"epoch": 0.49,
"learning_rate": 0.0002573420260782347,
"loss": 2.5634,
"theoretical_loss": 3.492664320688821,
"tokens_seen": 1619001344
},
{
"epoch": 0.49,
"learning_rate": 0.00025733199598796394,
"loss": 2.8771,
"theoretical_loss": 3.492652037390884,
"tokens_seen": 1619066880
},
{
"epoch": 0.49,
"learning_rate": 0.00025732196589769307,
"loss": 2.6473,
"theoretical_loss": 3.4926397547293444,
"tokens_seen": 1619132416
},
{
"epoch": 0.49,
"learning_rate": 0.0002573119358074223,
"loss": 2.6062,
"theoretical_loss": 3.492627472704145,
"tokens_seen": 1619197952
},
{
"epoch": 0.49,
"learning_rate": 0.00025730190571715143,
"loss": 2.7021,
"theoretical_loss": 3.4926151913152257,
"tokens_seen": 1619263488
},
{
"epoch": 0.49,
"learning_rate": 0.00025729187562688067,
"loss": 2.6342,
"theoretical_loss": 3.4926029105625283,
"tokens_seen": 1619329024
},
{
"epoch": 0.49,
"learning_rate": 0.00025728184553660985,
"loss": 2.6501,
"theoretical_loss": 3.492590630445995,
"tokens_seen": 1619394560
},
{
"epoch": 0.49,
"learning_rate": 0.00025727181544633903,
"loss": 2.5709,
"theoretical_loss": 3.492578350965565,
"tokens_seen": 1619460096
},
{
"epoch": 0.49,
"learning_rate": 0.0002572617853560682,
"loss": 2.8151,
"theoretical_loss": 3.492566072121181,
"tokens_seen": 1619525632
},
{
"epoch": 0.49,
"learning_rate": 0.0002572517552657974,
"loss": 2.6201,
"theoretical_loss": 3.492553793912785,
"tokens_seen": 1619591168
},
{
"epoch": 0.49,
"learning_rate": 0.0002572417251755266,
"loss": 2.6165,
"theoretical_loss": 3.4925415163403164,
"tokens_seen": 1619656704
},
{
"epoch": 0.49,
"learning_rate": 0.0002572316950852558,
"loss": 2.5727,
"theoretical_loss": 3.4925292394037184,
"tokens_seen": 1619722240
},
{
"epoch": 0.49,
"learning_rate": 0.00025722166499498494,
"loss": 2.4516,
"theoretical_loss": 3.492516963102931,
"tokens_seen": 1619787776
},
{
"epoch": 0.49,
"learning_rate": 0.00025721163490471417,
"loss": 2.5525,
"theoretical_loss": 3.492504687437896,
"tokens_seen": 1619853312
},
{
"epoch": 0.49,
"learning_rate": 0.0002572016048144433,
"loss": 2.5803,
"theoretical_loss": 3.4924924124085557,
"tokens_seen": 1619918848
},
{
"epoch": 0.49,
"learning_rate": 0.00025719157472417253,
"loss": 2.509,
"theoretical_loss": 3.49248013801485,
"tokens_seen": 1619984384
},
{
"epoch": 0.49,
"learning_rate": 0.0002571815446339017,
"loss": 2.8017,
"theoretical_loss": 3.4924678642567204,
"tokens_seen": 1620049920
},
{
"epoch": 0.49,
"learning_rate": 0.0002571715145436309,
"loss": 2.6578,
"theoretical_loss": 3.4924555911341093,
"tokens_seen": 1620115456
},
{
"epoch": 0.49,
"learning_rate": 0.0002571614844533601,
"loss": 2.5204,
"theoretical_loss": 3.4924433186469575,
"tokens_seen": 1620180992
},
{
"epoch": 0.49,
"learning_rate": 0.0002571514543630893,
"loss": 2.6925,
"theoretical_loss": 3.492431046795206,
"tokens_seen": 1620246528
},
{
"epoch": 0.49,
"learning_rate": 0.00025714142427281844,
"loss": 2.554,
"theoretical_loss": 3.4924187755787974,
"tokens_seen": 1620312064
},
{
"epoch": 0.49,
"objective/train/docs_used": 918561,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6934638023376465,
"objective/train/theoretical_loss": 3.492406504997672,
"objective/train/tokens_used": 1640837600,
"theoretical_loss": 3.492406504997672,
"tokens_seen": 1620377600
},
{
"epoch": 0.49,
"learning_rate": 0.0002571313941825477,
"loss": 2.6277,
"theoretical_loss": 3.492406504997672,
"tokens_seen": 1620377600
},
{
"epoch": 0.49,
"learning_rate": 0.0002571213640922768,
"loss": 2.7164,
"theoretical_loss": 3.4923942350517714,
"tokens_seen": 1620443136
},
{
"epoch": 0.49,
"learning_rate": 0.00025711133400200604,
"loss": 2.4571,
"theoretical_loss": 3.4923819657410373,
"tokens_seen": 1620508672
},
{
"epoch": 0.49,
"learning_rate": 0.0002571013039117352,
"loss": 2.3672,
"theoretical_loss": 3.492369697065411,
"tokens_seen": 1620574208
},
{
"epoch": 0.49,
"learning_rate": 0.0002570912738214644,
"loss": 2.5746,
"theoretical_loss": 3.4923574290248336,
"tokens_seen": 1620639744
},
{
"epoch": 0.49,
"learning_rate": 0.0002570812437311936,
"loss": 2.6838,
"theoretical_loss": 3.492345161619248,
"tokens_seen": 1620705280
},
{
"epoch": 0.49,
"learning_rate": 0.00025707121364092276,
"loss": 2.4423,
"theoretical_loss": 3.4923328948485937,
"tokens_seen": 1620770816
},
{
"epoch": 0.49,
"learning_rate": 0.00025706118355065194,
"loss": 2.894,
"theoretical_loss": 3.492320628712813,
"tokens_seen": 1620836352
},
{
"epoch": 0.49,
"learning_rate": 0.0002570511534603812,
"loss": 2.7288,
"theoretical_loss": 3.4923083632118477,
"tokens_seen": 1620901888
},
{
"epoch": 0.49,
"learning_rate": 0.0002570411233701103,
"loss": 2.6343,
"theoretical_loss": 3.492296098345639,
"tokens_seen": 1620967424
},
{
"epoch": 0.49,
"learning_rate": 0.00025703109327983954,
"loss": 2.4868,
"theoretical_loss": 3.4922838341141285,
"tokens_seen": 1621032960
},
{
"epoch": 0.49,
"learning_rate": 0.00025702106318956867,
"loss": 2.6969,
"theoretical_loss": 3.492271570517257,
"tokens_seen": 1621098496
},
{
"epoch": 0.49,
"learning_rate": 0.0002570110330992979,
"loss": 2.769,
"theoretical_loss": 3.492259307554967,
"tokens_seen": 1621164032
},
{
"epoch": 0.49,
"learning_rate": 0.0002570010030090271,
"loss": 2.648,
"theoretical_loss": 3.4922470452272,
"tokens_seen": 1621229568
},
{
"epoch": 0.49,
"learning_rate": 0.00025699097291875626,
"loss": 2.7715,
"theoretical_loss": 3.4922347835338967,
"tokens_seen": 1621295104
},
{
"epoch": 0.49,
"learning_rate": 0.00025698094282848545,
"loss": 2.6912,
"theoretical_loss": 3.492222522474999,
"tokens_seen": 1621360640
},
{
"epoch": 0.49,
"learning_rate": 0.0002569709127382147,
"loss": 2.5782,
"theoretical_loss": 3.492210262050449,
"tokens_seen": 1621426176
},
{
"epoch": 0.49,
"learning_rate": 0.0002569608826479438,
"loss": 2.8373,
"theoretical_loss": 3.492198002260187,
"tokens_seen": 1621491712
},
{
"epoch": 0.49,
"learning_rate": 0.00025695085255767304,
"loss": 2.9014,
"theoretical_loss": 3.4921857431041556,
"tokens_seen": 1621557248
},
{
"epoch": 0.49,
"learning_rate": 0.00025694082246740217,
"loss": 2.5739,
"theoretical_loss": 3.4921734845822963,
"tokens_seen": 1621622784
},
{
"epoch": 0.49,
"learning_rate": 0.0002569307923771314,
"loss": 2.5584,
"theoretical_loss": 3.4921612266945496,
"tokens_seen": 1621688320
},
{
"epoch": 0.49,
"learning_rate": 0.0002569207622868606,
"loss": 2.8544,
"theoretical_loss": 3.4921489694408585,
"tokens_seen": 1621753856
},
{
"epoch": 0.49,
"learning_rate": 0.00025691073219658977,
"loss": 2.757,
"theoretical_loss": 3.4921367128211642,
"tokens_seen": 1621819392
},
{
"epoch": 0.49,
"learning_rate": 0.00025690070210631895,
"loss": 2.7715,
"theoretical_loss": 3.4921244568354077,
"tokens_seen": 1621884928
},
{
"epoch": 0.49,
"learning_rate": 0.00025689067201604813,
"loss": 2.6807,
"theoretical_loss": 3.492112201483531,
"tokens_seen": 1621950464
},
{
"epoch": 0.49,
"objective/train/docs_used": 919775,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.627126693725586,
"objective/train/theoretical_loss": 3.4920999467654754,
"objective/train/tokens_used": 1642476000,
"theoretical_loss": 3.4920999467654754,
"tokens_seen": 1622016000
},
{
"epoch": 0.49,
"learning_rate": 0.0002568806419257773,
"loss": 2.5371,
"theoretical_loss": 3.4920999467654754,
"tokens_seen": 1622016000
},
{
"epoch": 0.49,
"learning_rate": 0.00025687061183550655,
"loss": 2.6162,
"theoretical_loss": 3.4920876926811832,
"tokens_seen": 1622081536
},
{
"epoch": 0.49,
"learning_rate": 0.0002568605817452357,
"loss": 2.5638,
"theoretical_loss": 3.4920754392305957,
"tokens_seen": 1622147072
},
{
"epoch": 0.49,
"learning_rate": 0.0002568505516549649,
"loss": 2.467,
"theoretical_loss": 3.492063186413654,
"tokens_seen": 1622212608
},
{
"epoch": 0.49,
"learning_rate": 0.00025684052156469404,
"loss": 2.7373,
"theoretical_loss": 3.4920509342303,
"tokens_seen": 1622278144
},
{
"epoch": 0.49,
"learning_rate": 0.00025683049147442327,
"loss": 2.7583,
"theoretical_loss": 3.492038682680476,
"tokens_seen": 1622343680
},
{
"epoch": 0.49,
"learning_rate": 0.00025682046138415245,
"loss": 2.6648,
"theoretical_loss": 3.492026431764123,
"tokens_seen": 1622409216
},
{
"epoch": 0.49,
"learning_rate": 0.00025681043129388163,
"loss": 2.6463,
"theoretical_loss": 3.4920141814811827,
"tokens_seen": 1622474752
},
{
"epoch": 0.49,
"learning_rate": 0.0002568004012036108,
"loss": 2.7998,
"theoretical_loss": 3.492001931831597,
"tokens_seen": 1622540288
},
{
"epoch": 0.49,
"learning_rate": 0.00025679037111334005,
"loss": 2.6417,
"theoretical_loss": 3.4919896828153076,
"tokens_seen": 1622605824
},
{
"epoch": 0.49,
"learning_rate": 0.0002567803410230692,
"loss": 2.663,
"theoretical_loss": 3.491977434432256,
"tokens_seen": 1622671360
},
{
"epoch": 0.49,
"learning_rate": 0.0002567703109327984,
"loss": 2.7696,
"theoretical_loss": 3.4919651866823833,
"tokens_seen": 1622736896
},
{
"epoch": 0.49,
"learning_rate": 0.00025676028084252754,
"loss": 2.5497,
"theoretical_loss": 3.4919529395656324,
"tokens_seen": 1622802432
},
{
"epoch": 0.49,
"learning_rate": 0.0002567502507522568,
"loss": 2.6346,
"theoretical_loss": 3.4919406930819443,
"tokens_seen": 1622867968
},
{
"epoch": 0.49,
"learning_rate": 0.000256740220661986,
"loss": 2.5029,
"theoretical_loss": 3.4919284472312606,
"tokens_seen": 1622933504
},
{
"epoch": 0.49,
"learning_rate": 0.00025673019057171514,
"loss": 2.6192,
"theoretical_loss": 3.491916202013524,
"tokens_seen": 1622999040
},
{
"epoch": 0.49,
"learning_rate": 0.00025672016048144437,
"loss": 2.8522,
"theoretical_loss": 3.4919039574286748,
"tokens_seen": 1623064576
},
{
"epoch": 0.49,
"learning_rate": 0.0002567101303911735,
"loss": 2.5339,
"theoretical_loss": 3.4918917134766554,
"tokens_seen": 1623130112
},
{
"epoch": 0.49,
"learning_rate": 0.00025670010030090273,
"loss": 2.8449,
"theoretical_loss": 3.491879470157408,
"tokens_seen": 1623195648
},
{
"epoch": 0.49,
"learning_rate": 0.0002566900702106319,
"loss": 2.5441,
"theoretical_loss": 3.4918672274708737,
"tokens_seen": 1623261184
},
{
"epoch": 0.49,
"learning_rate": 0.0002566800401203611,
"loss": 2.4791,
"theoretical_loss": 3.4918549854169947,
"tokens_seen": 1623326720
},
{
"epoch": 0.49,
"learning_rate": 0.0002566700100300903,
"loss": 2.3101,
"theoretical_loss": 3.4918427439957123,
"tokens_seen": 1623392256
},
{
"epoch": 0.49,
"learning_rate": 0.0002566599799398195,
"loss": 2.7064,
"theoretical_loss": 3.4918305032069687,
"tokens_seen": 1623457792
},
{
"epoch": 0.49,
"learning_rate": 0.00025664994984954864,
"loss": 2.6411,
"theoretical_loss": 3.4918182630507055,
"tokens_seen": 1623523328
},
{
"epoch": 0.49,
"learning_rate": 0.0002566399197592779,
"loss": 2.7055,
"theoretical_loss": 3.4918060235268644,
"tokens_seen": 1623588864
},
{
"epoch": 0.49,
"objective/train/docs_used": 920322,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5939831733703613,
"objective/train/theoretical_loss": 3.4917937846353873,
"objective/train/tokens_used": 1644114400,
"theoretical_loss": 3.4917937846353873,
"tokens_seen": 1623654400
},
{
"epoch": 0.49,
"learning_rate": 0.000256629889669007,
"loss": 2.3791,
"theoretical_loss": 3.4917937846353873,
"tokens_seen": 1623654400
},
{
"epoch": 0.49,
"learning_rate": 0.00025661985957873624,
"loss": 2.9424,
"theoretical_loss": 3.491781546376216,
"tokens_seen": 1623719936
},
{
"epoch": 0.49,
"learning_rate": 0.0002566098294884654,
"loss": 2.6388,
"theoretical_loss": 3.4917693087492925,
"tokens_seen": 1623785472
},
{
"epoch": 0.49,
"learning_rate": 0.0002565997993981946,
"loss": 2.5897,
"theoretical_loss": 3.491757071754558,
"tokens_seen": 1623851008
},
{
"epoch": 0.49,
"learning_rate": 0.0002565897693079238,
"loss": 2.7318,
"theoretical_loss": 3.4917448353919553,
"tokens_seen": 1623916544
},
{
"epoch": 0.49,
"learning_rate": 0.00025657973921765296,
"loss": 2.6442,
"theoretical_loss": 3.4917325996614257,
"tokens_seen": 1623982080
},
{
"epoch": 0.49,
"learning_rate": 0.00025656970912738214,
"loss": 2.6853,
"theoretical_loss": 3.4917203645629105,
"tokens_seen": 1624047616
},
{
"epoch": 0.49,
"learning_rate": 0.0002565596790371114,
"loss": 2.6031,
"theoretical_loss": 3.4917081300963524,
"tokens_seen": 1624113152
},
{
"epoch": 0.49,
"learning_rate": 0.0002565496489468405,
"loss": 2.6539,
"theoretical_loss": 3.4916958962616933,
"tokens_seen": 1624178688
},
{
"epoch": 0.49,
"learning_rate": 0.00025653961885656974,
"loss": 2.541,
"theoretical_loss": 3.4916836630588746,
"tokens_seen": 1624244224
},
{
"epoch": 0.49,
"learning_rate": 0.00025652958876629887,
"loss": 2.6091,
"theoretical_loss": 3.491671430487838,
"tokens_seen": 1624309760
},
{
"epoch": 0.49,
"learning_rate": 0.0002565195586760281,
"loss": 2.4319,
"theoretical_loss": 3.491659198548526,
"tokens_seen": 1624375296
},
{
"epoch": 0.49,
"learning_rate": 0.0002565095285857573,
"loss": 2.4901,
"theoretical_loss": 3.49164696724088,
"tokens_seen": 1624440832
},
{
"epoch": 0.49,
"learning_rate": 0.00025649949849548647,
"loss": 2.6986,
"theoretical_loss": 3.4916347365648424,
"tokens_seen": 1624506368
},
{
"epoch": 0.49,
"learning_rate": 0.00025648946840521565,
"loss": 2.4345,
"theoretical_loss": 3.491622506520354,
"tokens_seen": 1624571904
},
{
"epoch": 0.49,
"learning_rate": 0.0002564794383149449,
"loss": 2.72,
"theoretical_loss": 3.491610277107358,
"tokens_seen": 1624637440
},
{
"epoch": 0.49,
"learning_rate": 0.000256469408224674,
"loss": 2.6809,
"theoretical_loss": 3.4915980483257965,
"tokens_seen": 1624702976
},
{
"epoch": 0.49,
"learning_rate": 0.00025645937813440324,
"loss": 2.574,
"theoretical_loss": 3.49158582017561,
"tokens_seen": 1624768512
},
{
"epoch": 0.49,
"learning_rate": 0.00025644934804413237,
"loss": 2.4513,
"theoretical_loss": 3.4915735926567413,
"tokens_seen": 1624834048
},
{
"epoch": 0.49,
"learning_rate": 0.0002564393179538616,
"loss": 2.6245,
"theoretical_loss": 3.4915613657691322,
"tokens_seen": 1624899584
},
{
"epoch": 0.49,
"learning_rate": 0.0002564292878635908,
"loss": 2.4643,
"theoretical_loss": 3.4915491395127245,
"tokens_seen": 1624965120
},
{
"epoch": 0.49,
"learning_rate": 0.00025641925777331997,
"loss": 2.7374,
"theoretical_loss": 3.491536913887461,
"tokens_seen": 1625030656
},
{
"epoch": 0.49,
"learning_rate": 0.00025640922768304915,
"loss": 2.4826,
"theoretical_loss": 3.491524688893282,
"tokens_seen": 1625096192
},
{
"epoch": 0.49,
"learning_rate": 0.00025639919759277833,
"loss": 2.3774,
"theoretical_loss": 3.4915124645301314,
"tokens_seen": 1625161728
},
{
"epoch": 0.49,
"learning_rate": 0.0002563891675025075,
"loss": 2.5467,
"theoretical_loss": 3.49150024079795,
"tokens_seen": 1625227264
},
{
"epoch": 0.49,
"objective/train/docs_used": 921030,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3893353939056396,
"objective/train/theoretical_loss": 3.49148801769668,
"objective/train/tokens_used": 1645752800,
"theoretical_loss": 3.49148801769668,
"tokens_seen": 1625292800
},
{
"epoch": 0.49,
"learning_rate": 0.00025637913741223675,
"loss": 2.6123,
"theoretical_loss": 3.49148801769668,
"tokens_seen": 1625292800
},
{
"epoch": 0.49,
"learning_rate": 0.0002563691073219659,
"loss": 2.5236,
"theoretical_loss": 3.4914757952262634,
"tokens_seen": 1625358336
},
{
"epoch": 0.49,
"learning_rate": 0.0002563590772316951,
"loss": 2.9421,
"theoretical_loss": 3.4914635733866426,
"tokens_seen": 1625423872
},
{
"epoch": 0.49,
"learning_rate": 0.00025634904714142424,
"loss": 2.6407,
"theoretical_loss": 3.4914513521777586,
"tokens_seen": 1625489408
},
{
"epoch": 0.49,
"learning_rate": 0.00025633901705115347,
"loss": 2.7449,
"theoretical_loss": 3.491439131599555,
"tokens_seen": 1625554944
},
{
"epoch": 0.49,
"learning_rate": 0.00025632898696088265,
"loss": 2.6938,
"theoretical_loss": 3.4914269116519723,
"tokens_seen": 1625620480
},
{
"epoch": 0.49,
"learning_rate": 0.00025631895687061183,
"loss": 2.7743,
"theoretical_loss": 3.4914146923349536,
"tokens_seen": 1625686016
},
{
"epoch": 0.49,
"learning_rate": 0.000256308926780341,
"loss": 2.6755,
"theoretical_loss": 3.49140247364844,
"tokens_seen": 1625751552
},
{
"epoch": 0.49,
"learning_rate": 0.00025629889669007025,
"loss": 2.5298,
"theoretical_loss": 3.491390255592374,
"tokens_seen": 1625817088
},
{
"epoch": 0.49,
"learning_rate": 0.0002562888665997994,
"loss": 2.4984,
"theoretical_loss": 3.4913780381666983,
"tokens_seen": 1625882624
},
{
"epoch": 0.49,
"learning_rate": 0.0002562788365095286,
"loss": 2.7314,
"theoretical_loss": 3.491365821371354,
"tokens_seen": 1625948160
},
{
"epoch": 0.49,
"learning_rate": 0.00025626880641925774,
"loss": 2.589,
"theoretical_loss": 3.491353605206284,
"tokens_seen": 1626013696
},
{
"epoch": 0.49,
"learning_rate": 0.000256258776328987,
"loss": 2.7327,
"theoretical_loss": 3.4913413896714296,
"tokens_seen": 1626079232
},
{
"epoch": 0.49,
"learning_rate": 0.00025624874623871616,
"loss": 2.6711,
"theoretical_loss": 3.491329174766733,
"tokens_seen": 1626144768
},
{
"epoch": 0.49,
"learning_rate": 0.00025623871614844534,
"loss": 2.5677,
"theoretical_loss": 3.491316960492137,
"tokens_seen": 1626210304
},
{
"epoch": 0.49,
"learning_rate": 0.0002562286860581745,
"loss": 2.6106,
"theoretical_loss": 3.4913047468475833,
"tokens_seen": 1626275840
},
{
"epoch": 0.49,
"learning_rate": 0.0002562186559679037,
"loss": 2.9482,
"theoretical_loss": 3.491292533833014,
"tokens_seen": 1626341376
},
{
"epoch": 0.49,
"learning_rate": 0.0002562086258776329,
"loss": 2.7469,
"theoretical_loss": 3.4912803214483707,
"tokens_seen": 1626406912
},
{
"epoch": 0.49,
"learning_rate": 0.0002561985957873621,
"loss": 2.5166,
"theoretical_loss": 3.4912681096935962,
"tokens_seen": 1626472448
},
{
"epoch": 0.49,
"learning_rate": 0.00025618856569709124,
"loss": 2.6686,
"theoretical_loss": 3.491255898568632,
"tokens_seen": 1626537984
},
{
"epoch": 0.49,
"learning_rate": 0.0002561785356068205,
"loss": 2.369,
"theoretical_loss": 3.4912436880734212,
"tokens_seen": 1626603520
},
{
"epoch": 0.49,
"learning_rate": 0.00025616850551654966,
"loss": 2.4844,
"theoretical_loss": 3.4912314782079057,
"tokens_seen": 1626669056
},
{
"epoch": 0.49,
"learning_rate": 0.00025615847542627884,
"loss": 2.3983,
"theoretical_loss": 3.491219268972027,
"tokens_seen": 1626734592
},
{
"epoch": 0.49,
"learning_rate": 0.000256148445336008,
"loss": 2.8237,
"theoretical_loss": 3.4912070603657277,
"tokens_seen": 1626800128
},
{
"epoch": 0.49,
"learning_rate": 0.0002561384152457372,
"loss": 2.6264,
"theoretical_loss": 3.49119485238895,
"tokens_seen": 1626865664
},
{
"epoch": 0.49,
"objective/train/docs_used": 922470,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4795773029327393,
"objective/train/theoretical_loss": 3.4911826450416363,
"objective/train/tokens_used": 1647391200,
"theoretical_loss": 3.4911826450416363,
"tokens_seen": 1626931200
},
{
"epoch": 0.49,
"learning_rate": 0.0002561283851554664,
"loss": 2.6602,
"theoretical_loss": 3.4911826450416363,
"tokens_seen": 1626931200
},
{
"epoch": 0.49,
"learning_rate": 0.0002561183550651956,
"loss": 2.6154,
"theoretical_loss": 3.491170438323728,
"tokens_seen": 1626996736
},
{
"epoch": 0.49,
"learning_rate": 0.00025610832497492475,
"loss": 2.8549,
"theoretical_loss": 3.4911582322351684,
"tokens_seen": 1627062272
},
{
"epoch": 0.49,
"learning_rate": 0.000256098294884654,
"loss": 2.4712,
"theoretical_loss": 3.4911460267758985,
"tokens_seen": 1627127808
},
{
"epoch": 0.49,
"learning_rate": 0.0002560882647943831,
"loss": 2.711,
"theoretical_loss": 3.491133821945861,
"tokens_seen": 1627193344
},
{
"epoch": 0.49,
"learning_rate": 0.00025607823470411234,
"loss": 2.5265,
"theoretical_loss": 3.491121617744999,
"tokens_seen": 1627258880
},
{
"epoch": 0.49,
"learning_rate": 0.0002560682046138415,
"loss": 2.4002,
"theoretical_loss": 3.4911094141732537,
"tokens_seen": 1627324416
},
{
"epoch": 0.49,
"learning_rate": 0.0002560581745235707,
"loss": 2.5883,
"theoretical_loss": 3.4910972112305676,
"tokens_seen": 1627389952
},
{
"epoch": 0.49,
"learning_rate": 0.0002560481444332999,
"loss": 2.5785,
"theoretical_loss": 3.491085008916883,
"tokens_seen": 1627455488
},
{
"epoch": 0.49,
"learning_rate": 0.00025603811434302907,
"loss": 2.7423,
"theoretical_loss": 3.4910728072321415,
"tokens_seen": 1627521024
},
{
"epoch": 0.49,
"learning_rate": 0.00025602808425275825,
"loss": 2.609,
"theoretical_loss": 3.491060606176287,
"tokens_seen": 1627586560
},
{
"epoch": 0.49,
"learning_rate": 0.0002560180541624875,
"loss": 2.6283,
"theoretical_loss": 3.49104840574926,
"tokens_seen": 1627652096
},
{
"epoch": 0.49,
"learning_rate": 0.0002560080240722166,
"loss": 2.7176,
"theoretical_loss": 3.491036205951003,
"tokens_seen": 1627717632
},
{
"epoch": 0.49,
"learning_rate": 0.00025599799398194585,
"loss": 2.4467,
"theoretical_loss": 3.49102400678146,
"tokens_seen": 1627783168
},
{
"epoch": 0.49,
"learning_rate": 0.0002559879638916751,
"loss": 2.8315,
"theoretical_loss": 3.4910118082405717,
"tokens_seen": 1627848704
},
{
"epoch": 0.49,
"learning_rate": 0.0002559779338014042,
"loss": 2.6251,
"theoretical_loss": 3.4909996103282803,
"tokens_seen": 1627914240
},
{
"epoch": 0.49,
"learning_rate": 0.00025596790371113344,
"loss": 2.5897,
"theoretical_loss": 3.490987413044529,
"tokens_seen": 1627979776
},
{
"epoch": 0.49,
"learning_rate": 0.00025595787362086257,
"loss": 2.5256,
"theoretical_loss": 3.490975216389259,
"tokens_seen": 1628045312
},
{
"epoch": 0.49,
"learning_rate": 0.0002559478435305918,
"loss": 2.6875,
"theoretical_loss": 3.4909630203624142,
"tokens_seen": 1628110848
},
{
"epoch": 0.49,
"learning_rate": 0.000255937813440321,
"loss": 2.587,
"theoretical_loss": 3.490950824963936,
"tokens_seen": 1628176384
},
{
"epoch": 0.49,
"learning_rate": 0.00025592778335005017,
"loss": 2.6906,
"theoretical_loss": 3.4909386301937664,
"tokens_seen": 1628241920
},
{
"epoch": 0.49,
"learning_rate": 0.00025591775325977935,
"loss": 2.7164,
"theoretical_loss": 3.490926436051848,
"tokens_seen": 1628307456
},
{
"epoch": 0.49,
"learning_rate": 0.00025590772316950853,
"loss": 2.5322,
"theoretical_loss": 3.4909142425381234,
"tokens_seen": 1628372992
},
{
"epoch": 0.49,
"learning_rate": 0.0002558976930792377,
"loss": 2.6291,
"theoretical_loss": 3.4909020496525347,
"tokens_seen": 1628438528
},
{
"epoch": 0.49,
"learning_rate": 0.00025588766298896695,
"loss": 2.641,
"theoretical_loss": 3.4908898573950244,
"tokens_seen": 1628504064
},
{
"epoch": 0.49,
"objective/train/docs_used": 922615,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8449294567108154,
"objective/train/theoretical_loss": 3.4908776657655354,
"objective/train/tokens_used": 1649029600,
"theoretical_loss": 3.4908776657655354,
"tokens_seen": 1628569600
},
{
"epoch": 0.49,
"learning_rate": 0.0002558776328986961,
"loss": 2.6792,
"theoretical_loss": 3.4908776657655354,
"tokens_seen": 1628569600
},
{
"epoch": 0.49,
"learning_rate": 0.0002558676028084253,
"loss": 2.8268,
"theoretical_loss": 3.490865474764009,
"tokens_seen": 1628635136
},
{
"epoch": 0.49,
"learning_rate": 0.00025585757271815444,
"loss": 2.7006,
"theoretical_loss": 3.490853284390388,
"tokens_seen": 1628700672
},
{
"epoch": 0.49,
"learning_rate": 0.00025584754262788367,
"loss": 2.5012,
"theoretical_loss": 3.4908410946446153,
"tokens_seen": 1628766208
},
{
"epoch": 0.49,
"learning_rate": 0.00025583751253761285,
"loss": 2.5619,
"theoretical_loss": 3.490828905526633,
"tokens_seen": 1628831744
},
{
"epoch": 0.49,
"learning_rate": 0.00025582748244734203,
"loss": 2.5895,
"theoretical_loss": 3.490816717036383,
"tokens_seen": 1628897280
},
{
"epoch": 0.49,
"learning_rate": 0.0002558174523570712,
"loss": 2.5632,
"theoretical_loss": 3.490804529173808,
"tokens_seen": 1628962816
},
{
"epoch": 0.49,
"learning_rate": 0.00025580742226680045,
"loss": 2.6842,
"theoretical_loss": 3.4907923419388514,
"tokens_seen": 1629028352
},
{
"epoch": 0.49,
"learning_rate": 0.0002557973921765296,
"loss": 2.7174,
"theoretical_loss": 3.4907801553314544,
"tokens_seen": 1629093888
},
{
"epoch": 0.49,
"learning_rate": 0.0002557873620862588,
"loss": 2.5266,
"theoretical_loss": 3.4907679693515594,
"tokens_seen": 1629159424
},
{
"epoch": 0.49,
"learning_rate": 0.00025577733199598794,
"loss": 2.6618,
"theoretical_loss": 3.4907557839991097,
"tokens_seen": 1629224960
},
{
"epoch": 0.49,
"learning_rate": 0.0002557673019057172,
"loss": 2.6317,
"theoretical_loss": 3.4907435992740474,
"tokens_seen": 1629290496
},
{
"epoch": 0.49,
"learning_rate": 0.00025575727181544636,
"loss": 2.6711,
"theoretical_loss": 3.490731415176315,
"tokens_seen": 1629356032
},
{
"epoch": 0.49,
"learning_rate": 0.00025574724172517554,
"loss": 2.6134,
"theoretical_loss": 3.4907192317058544,
"tokens_seen": 1629421568
},
{
"epoch": 0.49,
"learning_rate": 0.0002557372116349047,
"loss": 2.8892,
"theoretical_loss": 3.4907070488626086,
"tokens_seen": 1629487104
},
{
"epoch": 0.49,
"learning_rate": 0.0002557271815446339,
"loss": 2.8334,
"theoretical_loss": 3.4906948666465203,
"tokens_seen": 1629552640
},
{
"epoch": 0.49,
"learning_rate": 0.0002557171514543631,
"loss": 2.6227,
"theoretical_loss": 3.490682685057532,
"tokens_seen": 1629618176
},
{
"epoch": 0.49,
"learning_rate": 0.0002557071213640923,
"loss": 2.8948,
"theoretical_loss": 3.490670504095586,
"tokens_seen": 1629683712
},
{
"epoch": 0.49,
"learning_rate": 0.00025569709127382144,
"loss": 2.6179,
"theoretical_loss": 3.4906583237606243,
"tokens_seen": 1629749248
},
{
"epoch": 0.49,
"learning_rate": 0.0002556870611835507,
"loss": 2.625,
"theoretical_loss": 3.49064614405259,
"tokens_seen": 1629814784
},
{
"epoch": 0.49,
"learning_rate": 0.00025567703109327986,
"loss": 2.724,
"theoretical_loss": 3.4906339649714258,
"tokens_seen": 1629880320
},
{
"epoch": 0.49,
"learning_rate": 0.00025566700100300904,
"loss": 2.6884,
"theoretical_loss": 3.4906217865170737,
"tokens_seen": 1629945856
},
{
"epoch": 0.49,
"learning_rate": 0.0002556569709127382,
"loss": 2.6123,
"theoretical_loss": 3.4906096086894767,
"tokens_seen": 1630011392
},
{
"epoch": 0.49,
"learning_rate": 0.0002556469408224674,
"loss": 2.5109,
"theoretical_loss": 3.4905974314885766,
"tokens_seen": 1630076928
},
{
"epoch": 0.49,
"learning_rate": 0.0002556369107321966,
"loss": 2.6402,
"theoretical_loss": 3.4905852549143166,
"tokens_seen": 1630142464
},
{
"epoch": 0.49,
"objective/train/docs_used": 923942,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8311870098114014,
"objective/train/theoretical_loss": 3.49057307896664,
"objective/train/tokens_used": 1650668000,
"theoretical_loss": 3.49057307896664,
"tokens_seen": 1630208000
},
{
"epoch": 0.49,
"learning_rate": 0.0002556268806419258,
"loss": 2.8257,
"theoretical_loss": 3.49057307896664,
"tokens_seen": 1630208000
},
{
"epoch": 0.49,
"learning_rate": 0.00025561685055165495,
"loss": 2.7765,
"theoretical_loss": 3.4905609036454877,
"tokens_seen": 1630273536
},
{
"epoch": 0.49,
"learning_rate": 0.0002556068204613842,
"loss": 2.6133,
"theoretical_loss": 3.490548728950803,
"tokens_seen": 1630339072
},
{
"epoch": 0.49,
"learning_rate": 0.0002555967903711133,
"loss": 2.722,
"theoretical_loss": 3.4905365548825285,
"tokens_seen": 1630404608
},
{
"epoch": 0.49,
"learning_rate": 0.00025558676028084254,
"loss": 2.8325,
"theoretical_loss": 3.4905243814406073,
"tokens_seen": 1630470144
},
{
"epoch": 0.49,
"learning_rate": 0.0002555767301905717,
"loss": 2.6204,
"theoretical_loss": 3.490512208624981,
"tokens_seen": 1630535680
},
{
"epoch": 0.49,
"learning_rate": 0.0002555667001003009,
"loss": 2.7807,
"theoretical_loss": 3.4905000364355936,
"tokens_seen": 1630601216
},
{
"epoch": 0.49,
"learning_rate": 0.0002555566700100301,
"loss": 2.6689,
"theoretical_loss": 3.490487864872386,
"tokens_seen": 1630666752
},
{
"epoch": 0.49,
"learning_rate": 0.00025554663991975927,
"loss": 2.6796,
"theoretical_loss": 3.4904756939353025,
"tokens_seen": 1630732288
},
{
"epoch": 0.49,
"learning_rate": 0.00025553660982948845,
"loss": 2.7219,
"theoretical_loss": 3.4904635236242845,
"tokens_seen": 1630797824
},
{
"epoch": 0.49,
"learning_rate": 0.0002555265797392177,
"loss": 2.8822,
"theoretical_loss": 3.4904513539392754,
"tokens_seen": 1630863360
},
{
"epoch": 0.49,
"learning_rate": 0.0002555165496489468,
"loss": 2.5047,
"theoretical_loss": 3.490439184880217,
"tokens_seen": 1630928896
},
{
"epoch": 0.49,
"learning_rate": 0.00025550651955867605,
"loss": 2.8161,
"theoretical_loss": 3.490427016447053,
"tokens_seen": 1630994432
},
{
"epoch": 0.49,
"learning_rate": 0.00025549648946840523,
"loss": 2.7747,
"theoretical_loss": 3.490414848639725,
"tokens_seen": 1631059968
},
{
"epoch": 0.49,
"learning_rate": 0.0002554864593781344,
"loss": 2.8397,
"theoretical_loss": 3.490402681458177,
"tokens_seen": 1631125504
},
{
"epoch": 0.49,
"learning_rate": 0.0002554764292878636,
"loss": 2.7438,
"theoretical_loss": 3.4903905149023506,
"tokens_seen": 1631191040
},
{
"epoch": 0.49,
"learning_rate": 0.00025546639919759277,
"loss": 2.4294,
"theoretical_loss": 3.4903783489721882,
"tokens_seen": 1631256576
},
{
"epoch": 0.49,
"learning_rate": 0.00025545636910732195,
"loss": 2.5069,
"theoretical_loss": 3.490366183667634,
"tokens_seen": 1631322112
},
{
"epoch": 0.49,
"learning_rate": 0.0002554463390170512,
"loss": 2.7165,
"theoretical_loss": 3.4903540189886284,
"tokens_seen": 1631387648
},
{
"epoch": 0.49,
"learning_rate": 0.0002554363089267803,
"loss": 2.7762,
"theoretical_loss": 3.4903418549351164,
"tokens_seen": 1631453184
},
{
"epoch": 0.49,
"learning_rate": 0.00025542627883650955,
"loss": 2.6062,
"theoretical_loss": 3.49032969150704,
"tokens_seen": 1631518720
},
{
"epoch": 0.49,
"learning_rate": 0.0002554162487462387,
"loss": 2.7774,
"theoretical_loss": 3.490317528704341,
"tokens_seen": 1631584256
},
{
"epoch": 0.49,
"learning_rate": 0.0002554062186559679,
"loss": 2.6288,
"theoretical_loss": 3.4903053665269637,
"tokens_seen": 1631649792
},
{
"epoch": 0.49,
"learning_rate": 0.0002553961885656971,
"loss": 2.4666,
"theoretical_loss": 3.490293204974849,
"tokens_seen": 1631715328
},
{
"epoch": 0.49,
"learning_rate": 0.0002553861584754263,
"loss": 2.4247,
"theoretical_loss": 3.490281044047941,
"tokens_seen": 1631780864
},
{
"epoch": 0.49,
"objective/train/docs_used": 924504,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6483798027038574,
"objective/train/theoretical_loss": 3.4902688837461824,
"objective/train/tokens_used": 1652306400,
"theoretical_loss": 3.4902688837461824,
"tokens_seen": 1631846400
},
{
"epoch": 0.49,
"learning_rate": 0.00025537612838515546,
"loss": 2.6421,
"theoretical_loss": 3.4902688837461824,
"tokens_seen": 1631846400
},
{
"epoch": 0.49,
"learning_rate": 0.00025536609829488464,
"loss": 2.7229,
"theoretical_loss": 3.490256724069515,
"tokens_seen": 1631911936
},
{
"epoch": 0.49,
"learning_rate": 0.0002553560682046138,
"loss": 2.7193,
"theoretical_loss": 3.490244565017883,
"tokens_seen": 1631977472
},
{
"epoch": 0.49,
"learning_rate": 0.00025534603811434305,
"loss": 2.7639,
"theoretical_loss": 3.490232406591227,
"tokens_seen": 1632043008
},
{
"epoch": 0.49,
"learning_rate": 0.0002553360080240722,
"loss": 2.6664,
"theoretical_loss": 3.4902202487894924,
"tokens_seen": 1632108544
},
{
"epoch": 0.49,
"learning_rate": 0.0002553259779338014,
"loss": 2.5856,
"theoretical_loss": 3.49020809161262,
"tokens_seen": 1632174080
},
{
"epoch": 0.49,
"learning_rate": 0.0002553159478435306,
"loss": 2.6826,
"theoretical_loss": 3.490195935060554,
"tokens_seen": 1632239616
},
{
"epoch": 0.49,
"learning_rate": 0.0002553059177532598,
"loss": 2.4922,
"theoretical_loss": 3.490183779133236,
"tokens_seen": 1632305152
},
{
"epoch": 0.49,
"learning_rate": 0.00025529588766298896,
"loss": 2.8162,
"theoretical_loss": 3.4901716238306095,
"tokens_seen": 1632370688
},
{
"epoch": 0.49,
"learning_rate": 0.00025528585757271814,
"loss": 2.4459,
"theoretical_loss": 3.490159469152617,
"tokens_seen": 1632436224
},
{
"epoch": 0.49,
"learning_rate": 0.0002552758274824473,
"loss": 2.7211,
"theoretical_loss": 3.4901473150992017,
"tokens_seen": 1632501760
},
{
"epoch": 0.49,
"learning_rate": 0.00025526579739217656,
"loss": 2.7031,
"theoretical_loss": 3.490135161670306,
"tokens_seen": 1632567296
},
{
"epoch": 0.49,
"learning_rate": 0.0002552557673019057,
"loss": 2.7192,
"theoretical_loss": 3.490123008865873,
"tokens_seen": 1632632832
},
{
"epoch": 0.49,
"learning_rate": 0.0002552457372116349,
"loss": 2.4757,
"theoretical_loss": 3.490110856685846,
"tokens_seen": 1632698368
},
{
"epoch": 0.49,
"learning_rate": 0.0002552357071213641,
"loss": 2.6512,
"theoretical_loss": 3.4900987051301664,
"tokens_seen": 1632763904
},
{
"epoch": 0.49,
"learning_rate": 0.0002552256770310933,
"loss": 2.5557,
"theoretical_loss": 3.490086554198779,
"tokens_seen": 1632829440
},
{
"epoch": 0.49,
"learning_rate": 0.0002552156469408225,
"loss": 2.8718,
"theoretical_loss": 3.490074403891625,
"tokens_seen": 1632894976
},
{
"epoch": 0.49,
"learning_rate": 0.00025520561685055164,
"loss": 2.6034,
"theoretical_loss": 3.490062254208649,
"tokens_seen": 1632960512
},
{
"epoch": 0.49,
"learning_rate": 0.0002551955867602809,
"loss": 2.6655,
"theoretical_loss": 3.490050105149792,
"tokens_seen": 1633026048
},
{
"epoch": 0.49,
"learning_rate": 0.00025518555667001006,
"loss": 2.4938,
"theoretical_loss": 3.490037956714998,
"tokens_seen": 1633091584
},
{
"epoch": 0.49,
"learning_rate": 0.00025517552657973924,
"loss": 2.6353,
"theoretical_loss": 3.4900258089042095,
"tokens_seen": 1633157120
},
{
"epoch": 0.49,
"learning_rate": 0.0002551654964894684,
"loss": 2.6889,
"theoretical_loss": 3.490013661717369,
"tokens_seen": 1633222656
},
{
"epoch": 0.49,
"learning_rate": 0.0002551554663991976,
"loss": 2.6112,
"theoretical_loss": 3.490001515154421,
"tokens_seen": 1633288192
},
{
"epoch": 0.49,
"learning_rate": 0.0002551454363089268,
"loss": 2.7146,
"theoretical_loss": 3.4899893692153072,
"tokens_seen": 1633353728
},
{
"epoch": 0.49,
"learning_rate": 0.000255135406218656,
"loss": 2.6601,
"theoretical_loss": 3.4899772238999707,
"tokens_seen": 1633419264
},
{
"epoch": 0.49,
"objective/train/docs_used": 925849,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.755000591278076,
"objective/train/theoretical_loss": 3.4899650792083543,
"objective/train/tokens_used": 1653944800,
"theoretical_loss": 3.4899650792083543,
"tokens_seen": 1633484800
},
{
"epoch": 0.49,
"learning_rate": 0.00025512537612838515,
"loss": 2.794,
"theoretical_loss": 3.4899650792083543,
"tokens_seen": 1633484800
},
{
"epoch": 0.5,
"learning_rate": 0.0002551153460381144,
"loss": 2.7564,
"theoretical_loss": 3.4899529351404013,
"tokens_seen": 1633550336
},
{
"epoch": 0.5,
"learning_rate": 0.0002551053159478435,
"loss": 2.8585,
"theoretical_loss": 3.4899407916960543,
"tokens_seen": 1633615872
},
{
"epoch": 0.5,
"learning_rate": 0.00025509528585757274,
"loss": 2.5995,
"theoretical_loss": 3.4899286488752566,
"tokens_seen": 1633681408
},
{
"epoch": 0.5,
"learning_rate": 0.0002550852557673019,
"loss": 2.5416,
"theoretical_loss": 3.489916506677951,
"tokens_seen": 1633746944
},
{
"epoch": 0.5,
"learning_rate": 0.0002550752256770311,
"loss": 2.6123,
"theoretical_loss": 3.4899043651040804,
"tokens_seen": 1633812480
},
{
"epoch": 0.5,
"learning_rate": 0.0002550651955867603,
"loss": 2.4828,
"theoretical_loss": 3.4898922241535884,
"tokens_seen": 1633878016
},
{
"epoch": 0.5,
"learning_rate": 0.00025505516549648947,
"loss": 2.6824,
"theoretical_loss": 3.489880083826417,
"tokens_seen": 1633943552
},
{
"epoch": 0.5,
"learning_rate": 0.00025504513540621865,
"loss": 2.6761,
"theoretical_loss": 3.4898679441225093,
"tokens_seen": 1634009088
},
{
"epoch": 0.5,
"learning_rate": 0.0002550351053159479,
"loss": 2.7994,
"theoretical_loss": 3.4898558050418096,
"tokens_seen": 1634074624
},
{
"epoch": 0.5,
"learning_rate": 0.000255025075225677,
"loss": 2.6699,
"theoretical_loss": 3.4898436665842594,
"tokens_seen": 1634140160
},
{
"epoch": 0.5,
"learning_rate": 0.00025501504513540625,
"loss": 2.5031,
"theoretical_loss": 3.4898315287498023,
"tokens_seen": 1634205696
},
{
"epoch": 0.5,
"learning_rate": 0.00025500501504513543,
"loss": 2.6066,
"theoretical_loss": 3.489819391538381,
"tokens_seen": 1634271232
},
{
"epoch": 0.5,
"learning_rate": 0.0002549949849548646,
"loss": 2.7401,
"theoretical_loss": 3.4898072549499397,
"tokens_seen": 1634336768
},
{
"epoch": 0.5,
"learning_rate": 0.0002549849548645938,
"loss": 2.7164,
"theoretical_loss": 3.4897951189844205,
"tokens_seen": 1634402304
},
{
"epoch": 0.5,
"learning_rate": 0.00025497492477432297,
"loss": 2.5396,
"theoretical_loss": 3.489782983641766,
"tokens_seen": 1634467840
},
{
"epoch": 0.5,
"learning_rate": 0.00025496489468405215,
"loss": 2.8066,
"theoretical_loss": 3.4897708489219204,
"tokens_seen": 1634533376
},
{
"epoch": 0.5,
"learning_rate": 0.0002549548645937814,
"loss": 2.6488,
"theoretical_loss": 3.489758714824826,
"tokens_seen": 1634598912
},
{
"epoch": 0.5,
"learning_rate": 0.0002549448345035105,
"loss": 2.5027,
"theoretical_loss": 3.4897465813504263,
"tokens_seen": 1634664448
},
{
"epoch": 0.5,
"learning_rate": 0.00025493480441323975,
"loss": 2.5497,
"theoretical_loss": 3.489734448498664,
"tokens_seen": 1634729984
},
{
"epoch": 0.5,
"learning_rate": 0.0002549247743229689,
"loss": 2.6162,
"theoretical_loss": 3.489722316269482,
"tokens_seen": 1634795520
},
{
"epoch": 0.5,
"learning_rate": 0.0002549147442326981,
"loss": 2.6505,
"theoretical_loss": 3.4897101846628242,
"tokens_seen": 1634861056
},
{
"epoch": 0.5,
"learning_rate": 0.0002549047141424273,
"loss": 2.6997,
"theoretical_loss": 3.4896980536786333,
"tokens_seen": 1634926592
},
{
"epoch": 0.5,
"learning_rate": 0.0002548946840521565,
"loss": 2.5431,
"theoretical_loss": 3.489685923316852,
"tokens_seen": 1634992128
},
{
"epoch": 0.5,
"learning_rate": 0.00025488465396188566,
"loss": 2.71,
"theoretical_loss": 3.489673793577424,
"tokens_seen": 1635057664
},
{
"epoch": 0.5,
"objective/train/docs_used": 926453,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0141756534576416,
"objective/train/theoretical_loss": 3.489661664460292,
"objective/train/tokens_used": 1655583200,
"theoretical_loss": 3.489661664460292,
"tokens_seen": 1635123200
},
{
"epoch": 0.5,
"learning_rate": 0.00025487462387161484,
"loss": 2.8742,
"theoretical_loss": 3.489661664460292,
"tokens_seen": 1635123200
},
{
"epoch": 0.5,
"learning_rate": 0.000254864593781344,
"loss": 2.7643,
"theoretical_loss": 3.4896495359653996,
"tokens_seen": 1635188736
},
{
"epoch": 0.5,
"learning_rate": 0.00025485456369107325,
"loss": 2.8678,
"theoretical_loss": 3.4896374080926895,
"tokens_seen": 1635254272
},
{
"epoch": 0.5,
"learning_rate": 0.0002548445336008024,
"loss": 2.7599,
"theoretical_loss": 3.4896252808421053,
"tokens_seen": 1635319808
},
{
"epoch": 0.5,
"learning_rate": 0.0002548345035105316,
"loss": 2.8649,
"theoretical_loss": 3.4896131542135898,
"tokens_seen": 1635385344
},
{
"epoch": 0.5,
"learning_rate": 0.0002548244734202608,
"loss": 2.5686,
"theoretical_loss": 3.4896010282070864,
"tokens_seen": 1635450880
},
{
"epoch": 0.5,
"learning_rate": 0.00025481444332999,
"loss": 2.7794,
"theoretical_loss": 3.489588902822538,
"tokens_seen": 1635516416
},
{
"epoch": 0.5,
"learning_rate": 0.00025480441323971916,
"loss": 2.7288,
"theoretical_loss": 3.4895767780598876,
"tokens_seen": 1635581952
},
{
"epoch": 0.5,
"learning_rate": 0.00025479438314944834,
"loss": 2.7153,
"theoretical_loss": 3.489564653919079,
"tokens_seen": 1635647488
},
{
"epoch": 0.5,
"learning_rate": 0.0002547843530591775,
"loss": 2.6127,
"theoretical_loss": 3.489552530400055,
"tokens_seen": 1635713024
},
{
"epoch": 0.5,
"learning_rate": 0.00025477432296890676,
"loss": 2.7404,
"theoretical_loss": 3.489540407502759,
"tokens_seen": 1635778560
},
{
"epoch": 0.5,
"learning_rate": 0.0002547642928786359,
"loss": 2.5703,
"theoretical_loss": 3.489528285227134,
"tokens_seen": 1635844096
},
{
"epoch": 0.5,
"learning_rate": 0.0002547542627883651,
"loss": 2.9668,
"theoretical_loss": 3.489516163573123,
"tokens_seen": 1635909632
},
{
"epoch": 0.5,
"learning_rate": 0.00025474423269809425,
"loss": 2.6211,
"theoretical_loss": 3.4895040425406703,
"tokens_seen": 1635975168
},
{
"epoch": 0.5,
"learning_rate": 0.0002547342026078235,
"loss": 2.6471,
"theoretical_loss": 3.4894919221297176,
"tokens_seen": 1636040704
},
{
"epoch": 0.5,
"learning_rate": 0.00025472417251755266,
"loss": 2.4377,
"theoretical_loss": 3.4894798023402096,
"tokens_seen": 1636106240
},
{
"epoch": 0.5,
"learning_rate": 0.00025471414242728184,
"loss": 2.7476,
"theoretical_loss": 3.4894676831720886,
"tokens_seen": 1636171776
},
{
"epoch": 0.5,
"learning_rate": 0.000254704112337011,
"loss": 2.7174,
"theoretical_loss": 3.489455564625298,
"tokens_seen": 1636237312
},
{
"epoch": 0.5,
"learning_rate": 0.00025469408224674026,
"loss": 2.6257,
"theoretical_loss": 3.489443446699781,
"tokens_seen": 1636302848
},
{
"epoch": 0.5,
"learning_rate": 0.0002546840521564694,
"loss": 2.8827,
"theoretical_loss": 3.4894313293954813,
"tokens_seen": 1636368384
},
{
"epoch": 0.5,
"learning_rate": 0.0002546740220661986,
"loss": 2.5414,
"theoretical_loss": 3.489419212712342,
"tokens_seen": 1636433920
},
{
"epoch": 0.5,
"learning_rate": 0.00025466399197592775,
"loss": 2.8345,
"theoretical_loss": 3.4894070966503064,
"tokens_seen": 1636499456
},
{
"epoch": 0.5,
"learning_rate": 0.000254653961885657,
"loss": 2.5665,
"theoretical_loss": 3.4893949812093172,
"tokens_seen": 1636564992
},
{
"epoch": 0.5,
"learning_rate": 0.00025464393179538616,
"loss": 2.8427,
"theoretical_loss": 3.4893828663893185,
"tokens_seen": 1636630528
},
{
"epoch": 0.5,
"learning_rate": 0.00025463390170511535,
"loss": 2.6903,
"theoretical_loss": 3.4893707521902533,
"tokens_seen": 1636696064
},
{
"epoch": 0.5,
"objective/train/docs_used": 927487,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7059483528137207,
"objective/train/theoretical_loss": 3.4893586386120647,
"objective/train/tokens_used": 1657221600,
"theoretical_loss": 3.4893586386120647,
"tokens_seen": 1636761600
},
{
"epoch": 0.5,
"learning_rate": 0.00025462387161484453,
"loss": 2.6477,
"theoretical_loss": 3.4893586386120647,
"tokens_seen": 1636761600
},
{
"epoch": 0.5,
"learning_rate": 0.0002546138415245737,
"loss": 2.8268,
"theoretical_loss": 3.489346525654696,
"tokens_seen": 1636827136
},
{
"epoch": 0.5,
"learning_rate": 0.0002546038114343029,
"loss": 2.2032,
"theoretical_loss": 3.4893344133180912,
"tokens_seen": 1636892672
},
{
"epoch": 0.5,
"learning_rate": 0.0002545937813440321,
"loss": 2.7402,
"theoretical_loss": 3.4893223016021935,
"tokens_seen": 1636958208
},
{
"epoch": 0.5,
"learning_rate": 0.00025458375125376125,
"loss": 2.6706,
"theoretical_loss": 3.4893101905069455,
"tokens_seen": 1637023744
},
{
"epoch": 0.5,
"learning_rate": 0.0002545737211634905,
"loss": 2.704,
"theoretical_loss": 3.489298080032291,
"tokens_seen": 1637089280
},
{
"epoch": 0.5,
"learning_rate": 0.0002545636910732196,
"loss": 2.598,
"theoretical_loss": 3.4892859701781735,
"tokens_seen": 1637154816
},
{
"epoch": 0.5,
"learning_rate": 0.00025455366098294885,
"loss": 2.8076,
"theoretical_loss": 3.489273860944536,
"tokens_seen": 1637220352
},
{
"epoch": 0.5,
"learning_rate": 0.00025454363089267803,
"loss": 2.6332,
"theoretical_loss": 3.4892617523313225,
"tokens_seen": 1637285888
},
{
"epoch": 0.5,
"learning_rate": 0.0002545336008024072,
"loss": 2.785,
"theoretical_loss": 3.4892496443384755,
"tokens_seen": 1637351424
},
{
"epoch": 0.5,
"learning_rate": 0.0002545235707121364,
"loss": 2.5703,
"theoretical_loss": 3.4892375369659394,
"tokens_seen": 1637416960
},
{
"epoch": 0.5,
"learning_rate": 0.00025451354062186563,
"loss": 2.4838,
"theoretical_loss": 3.489225430213656,
"tokens_seen": 1637482496
},
{
"epoch": 0.5,
"learning_rate": 0.0002545035105315948,
"loss": 2.5987,
"theoretical_loss": 3.489213324081571,
"tokens_seen": 1637548032
},
{
"epoch": 0.5,
"learning_rate": 0.000254493480441324,
"loss": 2.7247,
"theoretical_loss": 3.489201218569626,
"tokens_seen": 1637613568
},
{
"epoch": 0.5,
"learning_rate": 0.00025448345035105317,
"loss": 2.7387,
"theoretical_loss": 3.4891891136777646,
"tokens_seen": 1637679104
},
{
"epoch": 0.5,
"learning_rate": 0.00025447342026078235,
"loss": 2.7947,
"theoretical_loss": 3.489177009405931,
"tokens_seen": 1637744640
},
{
"epoch": 0.5,
"learning_rate": 0.0002544633901705116,
"loss": 2.7966,
"theoretical_loss": 3.4891649057540683,
"tokens_seen": 1637810176
},
{
"epoch": 0.5,
"learning_rate": 0.0002544533600802407,
"loss": 2.6968,
"theoretical_loss": 3.4891528027221197,
"tokens_seen": 1637875712
},
{
"epoch": 0.5,
"learning_rate": 0.00025444332998996995,
"loss": 2.6775,
"theoretical_loss": 3.489140700310029,
"tokens_seen": 1637941248
},
{
"epoch": 0.5,
"learning_rate": 0.0002544332998996991,
"loss": 2.6369,
"theoretical_loss": 3.4891285985177394,
"tokens_seen": 1638006784
},
{
"epoch": 0.5,
"learning_rate": 0.0002544232698094283,
"loss": 2.7326,
"theoretical_loss": 3.4891164973451945,
"tokens_seen": 1638072320
},
{
"epoch": 0.5,
"learning_rate": 0.0002544132397191575,
"loss": 2.7592,
"theoretical_loss": 3.4891043967923374,
"tokens_seen": 1638137856
},
{
"epoch": 0.5,
"learning_rate": 0.0002544032096288867,
"loss": 2.7887,
"theoretical_loss": 3.4890922968591123,
"tokens_seen": 1638203392
},
{
"epoch": 0.5,
"learning_rate": 0.00025439317953861586,
"loss": 2.4962,
"theoretical_loss": 3.489080197545462,
"tokens_seen": 1638268928
},
{
"epoch": 0.5,
"learning_rate": 0.00025438314944834504,
"loss": 2.8056,
"theoretical_loss": 3.4890680988513303,
"tokens_seen": 1638334464
},
{
"debugging/Self-BLEU-5": 0.299523828417223,
"debugging/distinct-1-grams": 0.7931298937164184,
"debugging/distinct-2-grams": 0.9722027972027973,
"debugging/entropy-1-grams": 5.372256762784476,
"debugging/entropy-2-grams": 5.925287792100262,
"debugging/length": 545.4,
"debugging/num_segments": 5,
"epoch": 0.5,
"objective/train/docs_used": 928998,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9119627475738525,
"objective/train/theoretical_loss": 3.4890560007766607,
"objective/train/tokens_used": 1658860000,
"theoretical_loss": 3.4890560007766607,
"tokens_seen": 1638400000
},
{
"epoch": 0.5,
"learning_rate": 0.0002543731193580742,
"loss": 2.822,
"theoretical_loss": 3.4890560007766607,
"tokens_seen": 1638400000
},
{
"epoch": 0.5,
"learning_rate": 0.00025436308926780345,
"loss": 2.7626,
"theoretical_loss": 3.489043903321397,
"tokens_seen": 1638465536
},
{
"epoch": 0.5,
"learning_rate": 0.0002543530591775326,
"loss": 2.8282,
"theoretical_loss": 3.4890318064854817,
"tokens_seen": 1638531072
},
{
"epoch": 0.5,
"learning_rate": 0.0002543430290872618,
"loss": 2.7132,
"theoretical_loss": 3.4890197102688596,
"tokens_seen": 1638596608
},
{
"epoch": 0.5,
"learning_rate": 0.000254332998996991,
"loss": 2.8337,
"theoretical_loss": 3.489007614671473,
"tokens_seen": 1638662144
},
{
"epoch": 0.5,
"learning_rate": 0.0002543229689067202,
"loss": 2.5409,
"theoretical_loss": 3.4889955196932663,
"tokens_seen": 1638727680
},
{
"epoch": 0.5,
"learning_rate": 0.00025431293881644936,
"loss": 2.5905,
"theoretical_loss": 3.488983425334183,
"tokens_seen": 1638793216
},
{
"epoch": 0.5,
"learning_rate": 0.00025430290872617854,
"loss": 2.7771,
"theoretical_loss": 3.4889713315941666,
"tokens_seen": 1638858752
},
{
"epoch": 0.5,
"learning_rate": 0.0002542928786359077,
"loss": 2.4658,
"theoretical_loss": 3.4889592384731603,
"tokens_seen": 1638924288
},
{
"epoch": 0.5,
"learning_rate": 0.00025428284854563696,
"loss": 2.8103,
"theoretical_loss": 3.4889471459711077,
"tokens_seen": 1638989824
},
{
"epoch": 0.5,
"learning_rate": 0.0002542728184553661,
"loss": 2.8503,
"theoretical_loss": 3.4889350540879533,
"tokens_seen": 1639055360
},
{
"epoch": 0.5,
"learning_rate": 0.0002542627883650953,
"loss": 2.8874,
"theoretical_loss": 3.488922962823639,
"tokens_seen": 1639120896
},
{
"epoch": 0.5,
"learning_rate": 0.00025425275827482445,
"loss": 2.8791,
"theoretical_loss": 3.48891087217811,
"tokens_seen": 1639186432
},
{
"epoch": 0.5,
"learning_rate": 0.0002542427281845537,
"loss": 2.793,
"theoretical_loss": 3.488898782151309,
"tokens_seen": 1639251968
},
{
"epoch": 0.5,
"learning_rate": 0.00025423269809428286,
"loss": 2.8444,
"theoretical_loss": 3.4888866927431796,
"tokens_seen": 1639317504
},
{
"epoch": 0.5,
"learning_rate": 0.00025422266800401204,
"loss": 2.5596,
"theoretical_loss": 3.4888746039536658,
"tokens_seen": 1639383040
},
{
"epoch": 0.5,
"learning_rate": 0.0002542126379137412,
"loss": 2.6182,
"theoretical_loss": 3.488862515782711,
"tokens_seen": 1639448576
},
{
"epoch": 0.5,
"learning_rate": 0.00025420260782347046,
"loss": 2.6637,
"theoretical_loss": 3.4888504282302595,
"tokens_seen": 1639514112
},
{
"epoch": 0.5,
"learning_rate": 0.0002541925777331996,
"loss": 2.586,
"theoretical_loss": 3.4888383412962534,
"tokens_seen": 1639579648
},
{
"epoch": 0.5,
"learning_rate": 0.0002541825476429288,
"loss": 2.6117,
"theoretical_loss": 3.488826254980638,
"tokens_seen": 1639645184
},
{
"epoch": 0.5,
"learning_rate": 0.00025417251755265795,
"loss": 2.7666,
"theoretical_loss": 3.4888141692833554,
"tokens_seen": 1639710720
},
{
"epoch": 0.5,
"learning_rate": 0.0002541624874623872,
"loss": 2.764,
"theoretical_loss": 3.488802084204351,
"tokens_seen": 1639776256
},
{
"epoch": 0.5,
"learning_rate": 0.00025415245737211637,
"loss": 2.7038,
"theoretical_loss": 3.4887899997435667,
"tokens_seen": 1639841792
},
{
"epoch": 0.5,
"learning_rate": 0.00025414242728184555,
"loss": 2.5489,
"theoretical_loss": 3.488777915900947,
"tokens_seen": 1639907328
},
{
"epoch": 0.5,
"learning_rate": 0.00025413239719157473,
"loss": 2.5971,
"theoretical_loss": 3.488765832676436,
"tokens_seen": 1639972864
},
{
"epoch": 0.5,
"objective/train/docs_used": 929588,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.727292776107788,
"objective/train/theoretical_loss": 3.488753750069977,
"objective/train/tokens_used": 1660498400,
"theoretical_loss": 3.488753750069977,
"tokens_seen": 1640038400
},
{
"epoch": 0.5,
"learning_rate": 0.0002541223671013039,
"loss": 2.718,
"theoretical_loss": 3.488753750069977,
"tokens_seen": 1640038400
},
{
"epoch": 0.5,
"learning_rate": 0.0002541123370110331,
"loss": 2.7448,
"theoretical_loss": 3.488741668081513,
"tokens_seen": 1640103936
},
{
"epoch": 0.5,
"learning_rate": 0.0002541023069207623,
"loss": 2.5182,
"theoretical_loss": 3.4887295867109884,
"tokens_seen": 1640169472
},
{
"epoch": 0.5,
"learning_rate": 0.00025409227683049145,
"loss": 2.5351,
"theoretical_loss": 3.4887175059583475,
"tokens_seen": 1640235008
},
{
"epoch": 0.5,
"learning_rate": 0.0002540822467402207,
"loss": 2.6681,
"theoretical_loss": 3.4887054258235333,
"tokens_seen": 1640300544
},
{
"epoch": 0.5,
"learning_rate": 0.0002540722166499498,
"loss": 2.6697,
"theoretical_loss": 3.488693346306489,
"tokens_seen": 1640366080
},
{
"epoch": 0.5,
"learning_rate": 0.00025406218655967905,
"loss": 2.7561,
"theoretical_loss": 3.4886812674071592,
"tokens_seen": 1640431616
},
{
"epoch": 0.5,
"learning_rate": 0.00025405215646940823,
"loss": 2.7751,
"theoretical_loss": 3.4886691891254875,
"tokens_seen": 1640497152
},
{
"epoch": 0.5,
"learning_rate": 0.0002540421263791374,
"loss": 2.8421,
"theoretical_loss": 3.488657111461417,
"tokens_seen": 1640562688
},
{
"epoch": 0.5,
"learning_rate": 0.0002540320962888666,
"loss": 2.5648,
"theoretical_loss": 3.4886450344148923,
"tokens_seen": 1640628224
},
{
"epoch": 0.5,
"learning_rate": 0.00025402206619859583,
"loss": 2.7977,
"theoretical_loss": 3.488632957985857,
"tokens_seen": 1640693760
},
{
"epoch": 0.5,
"learning_rate": 0.00025401203610832495,
"loss": 2.6222,
"theoretical_loss": 3.488620882174254,
"tokens_seen": 1640759296
},
{
"epoch": 0.5,
"learning_rate": 0.0002540020060180542,
"loss": 2.513,
"theoretical_loss": 3.4886088069800283,
"tokens_seen": 1640824832
},
{
"epoch": 0.5,
"learning_rate": 0.0002539919759277833,
"loss": 2.7632,
"theoretical_loss": 3.488596732403123,
"tokens_seen": 1640890368
},
{
"epoch": 0.5,
"learning_rate": 0.00025398194583751255,
"loss": 2.6838,
"theoretical_loss": 3.488584658443482,
"tokens_seen": 1640955904
},
{
"epoch": 0.5,
"learning_rate": 0.00025397191574724173,
"loss": 2.7174,
"theoretical_loss": 3.488572585101049,
"tokens_seen": 1641021440
},
{
"epoch": 0.5,
"learning_rate": 0.0002539618856569709,
"loss": 2.638,
"theoretical_loss": 3.4885605123757677,
"tokens_seen": 1641086976
},
{
"epoch": 0.5,
"learning_rate": 0.0002539518555667001,
"loss": 2.913,
"theoretical_loss": 3.488548440267582,
"tokens_seen": 1641152512
},
{
"epoch": 0.5,
"learning_rate": 0.0002539418254764293,
"loss": 2.4725,
"theoretical_loss": 3.4885363687764364,
"tokens_seen": 1641218048
},
{
"epoch": 0.5,
"learning_rate": 0.00025393179538615846,
"loss": 2.6439,
"theoretical_loss": 3.488524297902274,
"tokens_seen": 1641283584
},
{
"epoch": 0.5,
"learning_rate": 0.0002539217652958877,
"loss": 2.7108,
"theoretical_loss": 3.488512227645039,
"tokens_seen": 1641349120
},
{
"epoch": 0.5,
"learning_rate": 0.0002539117352056168,
"loss": 2.6577,
"theoretical_loss": 3.4885001580046744,
"tokens_seen": 1641414656
},
{
"epoch": 0.5,
"learning_rate": 0.00025390170511534606,
"loss": 2.4195,
"theoretical_loss": 3.488488088981125,
"tokens_seen": 1641480192
},
{
"epoch": 0.5,
"learning_rate": 0.0002538916750250752,
"loss": 2.6274,
"theoretical_loss": 3.4884760205743337,
"tokens_seen": 1641545728
},
{
"epoch": 0.5,
"learning_rate": 0.0002538816449348044,
"loss": 2.7809,
"theoretical_loss": 3.4884639527842456,
"tokens_seen": 1641611264
},
{
"epoch": 0.5,
"objective/train/docs_used": 930445,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7779228687286377,
"objective/train/theoretical_loss": 3.4884518856108038,
"objective/train/tokens_used": 1662136800,
"theoretical_loss": 3.4884518856108038,
"tokens_seen": 1641676800
},
{
"epoch": 0.5,
"learning_rate": 0.0002538716148445336,
"loss": 2.7894,
"theoretical_loss": 3.4884518856108038,
"tokens_seen": 1641676800
},
{
"epoch": 0.5,
"learning_rate": 0.0002538615847542628,
"loss": 2.463,
"theoretical_loss": 3.4884398190539523,
"tokens_seen": 1641742336
},
{
"epoch": 0.5,
"learning_rate": 0.00025385155466399196,
"loss": 2.6753,
"theoretical_loss": 3.488427753113635,
"tokens_seen": 1641807872
},
{
"epoch": 0.5,
"learning_rate": 0.0002538415245737212,
"loss": 2.6284,
"theoretical_loss": 3.488415687789796,
"tokens_seen": 1641873408
},
{
"epoch": 0.5,
"learning_rate": 0.0002538314944834503,
"loss": 2.9136,
"theoretical_loss": 3.4884036230823785,
"tokens_seen": 1641938944
},
{
"epoch": 0.5,
"learning_rate": 0.00025382146439317956,
"loss": 2.5951,
"theoretical_loss": 3.4883915589913275,
"tokens_seen": 1642004480
},
{
"epoch": 0.5,
"learning_rate": 0.0002538114343029087,
"loss": 2.6064,
"theoretical_loss": 3.488379495516586,
"tokens_seen": 1642070016
},
{
"epoch": 0.5,
"learning_rate": 0.0002538014042126379,
"loss": 2.5636,
"theoretical_loss": 3.488367432658098,
"tokens_seen": 1642135552
},
{
"epoch": 0.5,
"learning_rate": 0.0002537913741223671,
"loss": 2.6256,
"theoretical_loss": 3.488355370415808,
"tokens_seen": 1642201088
},
{
"epoch": 0.5,
"learning_rate": 0.0002537813440320963,
"loss": 2.7674,
"theoretical_loss": 3.488343308789659,
"tokens_seen": 1642266624
},
{
"epoch": 0.5,
"learning_rate": 0.00025377131394182546,
"loss": 2.5536,
"theoretical_loss": 3.4883312477795965,
"tokens_seen": 1642332160
},
{
"epoch": 0.5,
"learning_rate": 0.00025376128385155465,
"loss": 2.663,
"theoretical_loss": 3.488319187385563,
"tokens_seen": 1642397696
},
{
"epoch": 0.5,
"learning_rate": 0.0002537512537612839,
"loss": 2.6474,
"theoretical_loss": 3.488307127607503,
"tokens_seen": 1642463232
},
{
"epoch": 0.5,
"learning_rate": 0.00025374122367101306,
"loss": 2.7424,
"theoretical_loss": 3.4882950684453604,
"tokens_seen": 1642528768
},
{
"epoch": 0.5,
"learning_rate": 0.00025373119358074224,
"loss": 2.659,
"theoretical_loss": 3.488283009899079,
"tokens_seen": 1642594304
},
{
"epoch": 0.5,
"learning_rate": 0.0002537211634904714,
"loss": 2.6738,
"theoretical_loss": 3.4882709519686026,
"tokens_seen": 1642659840
},
{
"epoch": 0.5,
"learning_rate": 0.00025371113340020066,
"loss": 2.8408,
"theoretical_loss": 3.488258894653876,
"tokens_seen": 1642725376
},
{
"epoch": 0.5,
"learning_rate": 0.0002537011033099298,
"loss": 2.7415,
"theoretical_loss": 3.4882468379548426,
"tokens_seen": 1642790912
},
{
"epoch": 0.5,
"learning_rate": 0.000253691073219659,
"loss": 2.7632,
"theoretical_loss": 3.4882347818714465,
"tokens_seen": 1642856448
},
{
"epoch": 0.5,
"learning_rate": 0.00025368104312938815,
"loss": 3.0534,
"theoretical_loss": 3.488222726403632,
"tokens_seen": 1642921984
},
{
"epoch": 0.5,
"learning_rate": 0.0002536710130391174,
"loss": 2.5772,
"theoretical_loss": 3.488210671551342,
"tokens_seen": 1642987520
},
{
"epoch": 0.5,
"learning_rate": 0.00025366098294884657,
"loss": 2.7903,
"theoretical_loss": 3.4881986173145227,
"tokens_seen": 1643053056
},
{
"epoch": 0.5,
"learning_rate": 0.00025365095285857575,
"loss": 2.4641,
"theoretical_loss": 3.488186563693116,
"tokens_seen": 1643118592
},
{
"epoch": 0.5,
"learning_rate": 0.00025364092276830493,
"loss": 2.734,
"theoretical_loss": 3.4881745106870667,
"tokens_seen": 1643184128
},
{
"epoch": 0.5,
"learning_rate": 0.0002536308926780341,
"loss": 2.3967,
"theoretical_loss": 3.488162458296319,
"tokens_seen": 1643249664
},
{
"epoch": 0.5,
"objective/train/docs_used": 931444,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.780547857284546,
"objective/train/theoretical_loss": 3.488150406520817,
"objective/train/tokens_used": 1663775200,
"theoretical_loss": 3.488150406520817,
"tokens_seen": 1643315200
},
{
"epoch": 0.5,
"learning_rate": 0.0002536208625877633,
"loss": 2.5836,
"theoretical_loss": 3.488150406520817,
"tokens_seen": 1643315200
},
{
"epoch": 0.5,
"learning_rate": 0.0002536108324974925,
"loss": 2.6563,
"theoretical_loss": 3.4881383553605043,
"tokens_seen": 1643380736
},
{
"epoch": 0.5,
"learning_rate": 0.00025360080240722165,
"loss": 2.5788,
"theoretical_loss": 3.4881263048153253,
"tokens_seen": 1643446272
},
{
"epoch": 0.5,
"learning_rate": 0.0002535907723169509,
"loss": 2.7509,
"theoretical_loss": 3.4881142548852244,
"tokens_seen": 1643511808
},
{
"epoch": 0.5,
"learning_rate": 0.00025358074222668,
"loss": 2.6448,
"theoretical_loss": 3.488102205570145,
"tokens_seen": 1643577344
},
{
"epoch": 0.5,
"learning_rate": 0.00025357071213640925,
"loss": 2.6782,
"theoretical_loss": 3.4880901568700313,
"tokens_seen": 1643642880
},
{
"epoch": 0.5,
"learning_rate": 0.00025356068204613843,
"loss": 2.6383,
"theoretical_loss": 3.488078108784828,
"tokens_seen": 1643708416
},
{
"epoch": 0.5,
"learning_rate": 0.0002535506519558676,
"loss": 2.8104,
"theoretical_loss": 3.4880660613144787,
"tokens_seen": 1643773952
},
{
"epoch": 0.5,
"learning_rate": 0.0002535406218655968,
"loss": 2.8409,
"theoretical_loss": 3.4880540144589274,
"tokens_seen": 1643839488
},
{
"epoch": 0.5,
"learning_rate": 0.00025353059177532603,
"loss": 2.5285,
"theoretical_loss": 3.488041968218118,
"tokens_seen": 1643905024
},
{
"epoch": 0.5,
"learning_rate": 0.00025352056168505516,
"loss": 2.557,
"theoretical_loss": 3.4880299225919957,
"tokens_seen": 1643970560
},
{
"epoch": 0.5,
"learning_rate": 0.0002535105315947844,
"loss": 2.905,
"theoretical_loss": 3.488017877580504,
"tokens_seen": 1644036096
},
{
"epoch": 0.5,
"learning_rate": 0.0002535005015045135,
"loss": 2.5689,
"theoretical_loss": 3.4880058331835873,
"tokens_seen": 1644101632
},
{
"epoch": 0.5,
"learning_rate": 0.00025349047141424275,
"loss": 2.568,
"theoretical_loss": 3.4879937894011888,
"tokens_seen": 1644167168
},
{
"epoch": 0.5,
"learning_rate": 0.00025348044132397193,
"loss": 2.5604,
"theoretical_loss": 3.4879817462332534,
"tokens_seen": 1644232704
},
{
"epoch": 0.5,
"learning_rate": 0.0002534704112337011,
"loss": 2.7694,
"theoretical_loss": 3.4879697036797257,
"tokens_seen": 1644298240
},
{
"epoch": 0.5,
"learning_rate": 0.0002534603811434303,
"loss": 2.7228,
"theoretical_loss": 3.4879576617405488,
"tokens_seen": 1644363776
},
{
"epoch": 0.5,
"learning_rate": 0.0002534503510531595,
"loss": 2.6577,
"theoretical_loss": 3.4879456204156676,
"tokens_seen": 1644429312
},
{
"epoch": 0.5,
"learning_rate": 0.00025344032096288866,
"loss": 2.5647,
"theoretical_loss": 3.487933579705026,
"tokens_seen": 1644494848
},
{
"epoch": 0.5,
"learning_rate": 0.0002534302908726179,
"loss": 2.5287,
"theoretical_loss": 3.4879215396085685,
"tokens_seen": 1644560384
},
{
"epoch": 0.5,
"learning_rate": 0.000253420260782347,
"loss": 2.5406,
"theoretical_loss": 3.4879095001262392,
"tokens_seen": 1644625920
},
{
"epoch": 0.5,
"learning_rate": 0.00025341023069207626,
"loss": 2.6967,
"theoretical_loss": 3.487897461257982,
"tokens_seen": 1644691456
},
{
"epoch": 0.5,
"learning_rate": 0.0002534002006018054,
"loss": 2.5627,
"theoretical_loss": 3.487885423003741,
"tokens_seen": 1644756992
},
{
"epoch": 0.5,
"learning_rate": 0.0002533901705115346,
"loss": 2.7543,
"theoretical_loss": 3.487873385363461,
"tokens_seen": 1644822528
},
{
"epoch": 0.5,
"learning_rate": 0.0002533801404212638,
"loss": 2.677,
"theoretical_loss": 3.487861348337086,
"tokens_seen": 1644888064
},
{
"epoch": 0.5,
"objective/train/docs_used": 932028,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2925543785095215,
"objective/train/theoretical_loss": 3.4878493119245606,
"objective/train/tokens_used": 1665413600,
"theoretical_loss": 3.4878493119245606,
"tokens_seen": 1644953600
},
{
"epoch": 0.5,
"learning_rate": 0.000253370110330993,
"loss": 2.5414,
"theoretical_loss": 3.4878493119245606,
"tokens_seen": 1644953600
},
{
"epoch": 0.5,
"learning_rate": 0.00025336008024072216,
"loss": 2.6505,
"theoretical_loss": 3.487837276125828,
"tokens_seen": 1645019136
},
{
"epoch": 0.5,
"learning_rate": 0.0002533500501504514,
"loss": 2.474,
"theoretical_loss": 3.4878252409408335,
"tokens_seen": 1645084672
},
{
"epoch": 0.5,
"learning_rate": 0.0002533400200601805,
"loss": 2.6088,
"theoretical_loss": 3.4878132063695206,
"tokens_seen": 1645150208
},
{
"epoch": 0.5,
"learning_rate": 0.00025332998996990976,
"loss": 2.7213,
"theoretical_loss": 3.4878011724118343,
"tokens_seen": 1645215744
},
{
"epoch": 0.5,
"learning_rate": 0.0002533199598796389,
"loss": 2.6199,
"theoretical_loss": 3.487789139067718,
"tokens_seen": 1645281280
},
{
"epoch": 0.5,
"learning_rate": 0.0002533099297893681,
"loss": 2.6846,
"theoretical_loss": 3.487777106337117,
"tokens_seen": 1645346816
},
{
"epoch": 0.5,
"learning_rate": 0.0002532998996990973,
"loss": 2.7638,
"theoretical_loss": 3.4877650742199746,
"tokens_seen": 1645412352
},
{
"epoch": 0.5,
"learning_rate": 0.0002532898696088265,
"loss": 2.6303,
"theoretical_loss": 3.4877530427162355,
"tokens_seen": 1645477888
},
{
"epoch": 0.5,
"learning_rate": 0.00025327983951855566,
"loss": 2.545,
"theoretical_loss": 3.487741011825844,
"tokens_seen": 1645543424
},
{
"epoch": 0.5,
"learning_rate": 0.00025326980942828485,
"loss": 2.4778,
"theoretical_loss": 3.4877289815487447,
"tokens_seen": 1645608960
},
{
"epoch": 0.5,
"learning_rate": 0.000253259779338014,
"loss": 2.7328,
"theoretical_loss": 3.487716951884882,
"tokens_seen": 1645674496
},
{
"epoch": 0.5,
"learning_rate": 0.00025324974924774326,
"loss": 2.4797,
"theoretical_loss": 3.487704922834199,
"tokens_seen": 1645740032
},
{
"epoch": 0.5,
"learning_rate": 0.0002532397191574724,
"loss": 2.5593,
"theoretical_loss": 3.487692894396641,
"tokens_seen": 1645805568
},
{
"epoch": 0.5,
"learning_rate": 0.0002532296890672016,
"loss": 2.5641,
"theoretical_loss": 3.487680866572153,
"tokens_seen": 1645871104
},
{
"epoch": 0.5,
"learning_rate": 0.00025321965897693075,
"loss": 2.6266,
"theoretical_loss": 3.487668839360678,
"tokens_seen": 1645936640
},
{
"epoch": 0.5,
"learning_rate": 0.00025320962888666,
"loss": 2.8061,
"theoretical_loss": 3.487656812762161,
"tokens_seen": 1646002176
},
{
"epoch": 0.5,
"learning_rate": 0.00025319959879638917,
"loss": 2.7052,
"theoretical_loss": 3.4876447867765465,
"tokens_seen": 1646067712
},
{
"epoch": 0.5,
"learning_rate": 0.00025318956870611835,
"loss": 2.5577,
"theoretical_loss": 3.4876327614037788,
"tokens_seen": 1646133248
},
{
"epoch": 0.5,
"learning_rate": 0.00025317953861584753,
"loss": 2.3043,
"theoretical_loss": 3.4876207366438017,
"tokens_seen": 1646198784
},
{
"epoch": 0.5,
"learning_rate": 0.00025316950852557677,
"loss": 2.7388,
"theoretical_loss": 3.4876087124965602,
"tokens_seen": 1646264320
},
{
"epoch": 0.5,
"learning_rate": 0.0002531594784353059,
"loss": 2.6308,
"theoretical_loss": 3.4875966889619985,
"tokens_seen": 1646329856
},
{
"epoch": 0.5,
"learning_rate": 0.00025314944834503513,
"loss": 2.7784,
"theoretical_loss": 3.4875846660400605,
"tokens_seen": 1646395392
},
{
"epoch": 0.5,
"learning_rate": 0.00025313941825476425,
"loss": 2.6506,
"theoretical_loss": 3.487572643730692,
"tokens_seen": 1646460928
},
{
"epoch": 0.5,
"learning_rate": 0.0002531293881644935,
"loss": 2.6844,
"theoretical_loss": 3.4875606220338358,
"tokens_seen": 1646526464
},
{
"epoch": 0.5,
"objective/train/docs_used": 933210,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7672715187072754,
"objective/train/theoretical_loss": 3.487548600949437,
"objective/train/tokens_used": 1667052000,
"theoretical_loss": 3.487548600949437,
"tokens_seen": 1646592000
},
{
"epoch": 0.5,
"learning_rate": 0.00025311935807422267,
"loss": 2.6184,
"theoretical_loss": 3.487548600949437,
"tokens_seen": 1646592000
},
{
"epoch": 0.5,
"learning_rate": 0.00025310932798395185,
"loss": 2.5069,
"theoretical_loss": 3.4875365804774408,
"tokens_seen": 1646657536
},
{
"epoch": 0.5,
"learning_rate": 0.00025309929789368103,
"loss": 2.5339,
"theoretical_loss": 3.48752456061779,
"tokens_seen": 1646723072
},
{
"epoch": 0.5,
"learning_rate": 0.0002530892678034102,
"loss": 2.7307,
"theoretical_loss": 3.48751254137043,
"tokens_seen": 1646788608
},
{
"epoch": 0.5,
"learning_rate": 0.0002530792377131394,
"loss": 2.6621,
"theoretical_loss": 3.4875005227353055,
"tokens_seen": 1646854144
},
{
"epoch": 0.5,
"learning_rate": 0.00025306920762286863,
"loss": 2.7978,
"theoretical_loss": 3.4874885047123607,
"tokens_seen": 1646919680
},
{
"epoch": 0.5,
"learning_rate": 0.00025305917753259776,
"loss": 2.6859,
"theoretical_loss": 3.4874764873015396,
"tokens_seen": 1646985216
},
{
"epoch": 0.5,
"learning_rate": 0.000253049147442327,
"loss": 2.7317,
"theoretical_loss": 3.487464470502787,
"tokens_seen": 1647050752
},
{
"epoch": 0.5,
"learning_rate": 0.0002530391173520562,
"loss": 2.6779,
"theoretical_loss": 3.487452454316048,
"tokens_seen": 1647116288
},
{
"epoch": 0.5,
"learning_rate": 0.00025302908726178536,
"loss": 2.6419,
"theoretical_loss": 3.487440438741266,
"tokens_seen": 1647181824
},
{
"epoch": 0.5,
"learning_rate": 0.00025301905717151454,
"loss": 2.7944,
"theoretical_loss": 3.4874284237783857,
"tokens_seen": 1647247360
},
{
"epoch": 0.5,
"learning_rate": 0.0002530090270812437,
"loss": 2.6736,
"theoretical_loss": 3.487416409427352,
"tokens_seen": 1647312896
},
{
"epoch": 0.5,
"learning_rate": 0.00025299899699097295,
"loss": 2.7594,
"theoretical_loss": 3.4874043956881096,
"tokens_seen": 1647378432
},
{
"epoch": 0.5,
"learning_rate": 0.00025298896690070213,
"loss": 2.7304,
"theoretical_loss": 3.4873923825606026,
"tokens_seen": 1647443968
},
{
"epoch": 0.5,
"learning_rate": 0.0002529789368104313,
"loss": 2.6567,
"theoretical_loss": 3.4873803700447756,
"tokens_seen": 1647509504
},
{
"epoch": 0.5,
"learning_rate": 0.0002529689067201605,
"loss": 2.6306,
"theoretical_loss": 3.487368358140573,
"tokens_seen": 1647575040
},
{
"epoch": 0.5,
"learning_rate": 0.0002529588766298897,
"loss": 2.4931,
"theoretical_loss": 3.4873563468479394,
"tokens_seen": 1647640576
},
{
"epoch": 0.5,
"learning_rate": 0.00025294884653961886,
"loss": 2.9931,
"theoretical_loss": 3.4873443361668195,
"tokens_seen": 1647706112
},
{
"epoch": 0.5,
"learning_rate": 0.0002529388164493481,
"loss": 2.7188,
"theoretical_loss": 3.4873323260971576,
"tokens_seen": 1647771648
},
{
"epoch": 0.5,
"learning_rate": 0.0002529287863590772,
"loss": 2.7649,
"theoretical_loss": 3.4873203166388986,
"tokens_seen": 1647837184
},
{
"epoch": 0.5,
"learning_rate": 0.00025291875626880646,
"loss": 2.5912,
"theoretical_loss": 3.487308307791987,
"tokens_seen": 1647902720
},
{
"epoch": 0.5,
"learning_rate": 0.0002529087261785356,
"loss": 2.779,
"theoretical_loss": 3.487296299556367,
"tokens_seen": 1647968256
},
{
"epoch": 0.5,
"learning_rate": 0.0002528986960882648,
"loss": 2.4655,
"theoretical_loss": 3.4872842919319833,
"tokens_seen": 1648033792
},
{
"epoch": 0.5,
"learning_rate": 0.000252888665997994,
"loss": 2.3951,
"theoretical_loss": 3.4872722849187805,
"tokens_seen": 1648099328
},
{
"epoch": 0.5,
"learning_rate": 0.0002528786359077232,
"loss": 2.7371,
"theoretical_loss": 3.4872602785167035,
"tokens_seen": 1648164864
},
{
"epoch": 0.5,
"objective/train/docs_used": 933788,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2070658206939697,
"objective/train/theoretical_loss": 3.4872482727256964,
"objective/train/tokens_used": 1668690400,
"theoretical_loss": 3.4872482727256964,
"tokens_seen": 1648230400
},
{
"epoch": 0.5,
"learning_rate": 0.00025286860581745236,
"loss": 2.51,
"theoretical_loss": 3.4872482727256964,
"tokens_seen": 1648230400
},
{
"epoch": 0.5,
"learning_rate": 0.0002528585757271816,
"loss": 2.5285,
"theoretical_loss": 3.4872362675457045,
"tokens_seen": 1648295936
},
{
"epoch": 0.5,
"learning_rate": 0.0002528485456369107,
"loss": 2.7369,
"theoretical_loss": 3.487224262976672,
"tokens_seen": 1648361472
},
{
"epoch": 0.5,
"learning_rate": 0.00025283851554663996,
"loss": 2.5617,
"theoretical_loss": 3.487212259018543,
"tokens_seen": 1648427008
},
{
"epoch": 0.5,
"learning_rate": 0.0002528284854563691,
"loss": 2.5873,
"theoretical_loss": 3.487200255671263,
"tokens_seen": 1648492544
},
{
"epoch": 0.5,
"learning_rate": 0.0002528184553660983,
"loss": 2.4994,
"theoretical_loss": 3.487188252934776,
"tokens_seen": 1648558080
},
{
"epoch": 0.5,
"learning_rate": 0.0002528084252758275,
"loss": 2.7864,
"theoretical_loss": 3.4871762508090267,
"tokens_seen": 1648623616
},
{
"epoch": 0.5,
"learning_rate": 0.0002527983951855567,
"loss": 2.7661,
"theoretical_loss": 3.4871642492939605,
"tokens_seen": 1648689152
},
{
"epoch": 0.5,
"learning_rate": 0.00025278836509528586,
"loss": 2.4356,
"theoretical_loss": 3.4871522483895214,
"tokens_seen": 1648754688
},
{
"epoch": 0.5,
"learning_rate": 0.00025277833500501505,
"loss": 2.7549,
"theoretical_loss": 3.4871402480956535,
"tokens_seen": 1648820224
},
{
"epoch": 0.5,
"learning_rate": 0.0002527683049147442,
"loss": 2.4734,
"theoretical_loss": 3.487128248412303,
"tokens_seen": 1648885760
},
{
"epoch": 0.5,
"learning_rate": 0.00025275827482447346,
"loss": 2.7182,
"theoretical_loss": 3.4871162493394134,
"tokens_seen": 1648951296
},
{
"epoch": 0.5,
"learning_rate": 0.0002527482447342026,
"loss": 2.5664,
"theoretical_loss": 3.4871042508769294,
"tokens_seen": 1649016832
},
{
"epoch": 0.5,
"learning_rate": 0.0002527382146439318,
"loss": 2.6568,
"theoretical_loss": 3.4870922530247963,
"tokens_seen": 1649082368
},
{
"epoch": 0.5,
"learning_rate": 0.00025272818455366095,
"loss": 2.6276,
"theoretical_loss": 3.4870802557829585,
"tokens_seen": 1649147904
},
{
"epoch": 0.5,
"learning_rate": 0.0002527181544633902,
"loss": 2.7853,
"theoretical_loss": 3.4870682591513606,
"tokens_seen": 1649213440
},
{
"epoch": 0.5,
"learning_rate": 0.00025270812437311937,
"loss": 2.7104,
"theoretical_loss": 3.4870562631299475,
"tokens_seen": 1649278976
},
{
"epoch": 0.5,
"learning_rate": 0.00025269809428284855,
"loss": 2.6854,
"theoretical_loss": 3.4870442677186637,
"tokens_seen": 1649344512
},
{
"epoch": 0.5,
"learning_rate": 0.00025268806419257773,
"loss": 2.6956,
"theoretical_loss": 3.487032272917454,
"tokens_seen": 1649410048
},
{
"epoch": 0.5,
"learning_rate": 0.00025267803410230697,
"loss": 2.916,
"theoretical_loss": 3.4870202787262636,
"tokens_seen": 1649475584
},
{
"epoch": 0.5,
"learning_rate": 0.0002526680040120361,
"loss": 2.5863,
"theoretical_loss": 3.4870082851450364,
"tokens_seen": 1649541120
},
{
"epoch": 0.5,
"learning_rate": 0.00025265797392176533,
"loss": 2.4404,
"theoretical_loss": 3.4869962921737176,
"tokens_seen": 1649606656
},
{
"epoch": 0.5,
"learning_rate": 0.00025264794383149445,
"loss": 2.815,
"theoretical_loss": 3.486984299812252,
"tokens_seen": 1649672192
},
{
"epoch": 0.5,
"learning_rate": 0.0002526379137412237,
"loss": 2.4832,
"theoretical_loss": 3.486972308060584,
"tokens_seen": 1649737728
},
{
"epoch": 0.5,
"learning_rate": 0.00025262788365095287,
"loss": 2.8596,
"theoretical_loss": 3.486960316918659,
"tokens_seen": 1649803264
},
{
"epoch": 0.5,
"objective/train/docs_used": 934285,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4778454303741455,
"objective/train/theoretical_loss": 3.4869483263864214,
"objective/train/tokens_used": 1670328800,
"theoretical_loss": 3.4869483263864214,
"tokens_seen": 1649868800
},
{
"epoch": 0.5,
"learning_rate": 0.00025261785356068205,
"loss": 2.5521,
"theoretical_loss": 3.4869483263864214,
"tokens_seen": 1649868800
},
{
"epoch": 0.5,
"learning_rate": 0.00025260782347041123,
"loss": 2.7061,
"theoretical_loss": 3.486936336463816,
"tokens_seen": 1649934336
},
{
"epoch": 0.5,
"learning_rate": 0.0002525977933801404,
"loss": 2.4675,
"theoretical_loss": 3.4869243471507874,
"tokens_seen": 1649999872
},
{
"epoch": 0.5,
"learning_rate": 0.0002525877632898696,
"loss": 2.7251,
"theoretical_loss": 3.486912358447281,
"tokens_seen": 1650065408
},
{
"epoch": 0.5,
"learning_rate": 0.00025257773319959883,
"loss": 2.8025,
"theoretical_loss": 3.486900370353241,
"tokens_seen": 1650130944
},
{
"epoch": 0.5,
"learning_rate": 0.00025256770310932796,
"loss": 2.7248,
"theoretical_loss": 3.486888382868612,
"tokens_seen": 1650196480
},
{
"epoch": 0.5,
"learning_rate": 0.0002525576730190572,
"loss": 2.4737,
"theoretical_loss": 3.48687639599334,
"tokens_seen": 1650262016
},
{
"epoch": 0.5,
"learning_rate": 0.0002525476429287864,
"loss": 2.7756,
"theoretical_loss": 3.4868644097273687,
"tokens_seen": 1650327552
},
{
"epoch": 0.5,
"learning_rate": 0.00025253761283851556,
"loss": 2.6907,
"theoretical_loss": 3.4868524240706433,
"tokens_seen": 1650393088
},
{
"epoch": 0.5,
"learning_rate": 0.00025252758274824474,
"loss": 2.7137,
"theoretical_loss": 3.4868404390231085,
"tokens_seen": 1650458624
},
{
"epoch": 0.5,
"learning_rate": 0.0002525175526579739,
"loss": 2.6196,
"theoretical_loss": 3.4868284545847095,
"tokens_seen": 1650524160
},
{
"epoch": 0.5,
"learning_rate": 0.0002525075225677031,
"loss": 2.7656,
"theoretical_loss": 3.4868164707553913,
"tokens_seen": 1650589696
},
{
"epoch": 0.5,
"learning_rate": 0.00025249749247743233,
"loss": 2.6397,
"theoretical_loss": 3.486804487535098,
"tokens_seen": 1650655232
},
{
"epoch": 0.5,
"learning_rate": 0.00025248746238716146,
"loss": 2.6367,
"theoretical_loss": 3.486792504923775,
"tokens_seen": 1650720768
},
{
"epoch": 0.5,
"learning_rate": 0.0002524774322968907,
"loss": 2.9656,
"theoretical_loss": 3.486780522921367,
"tokens_seen": 1650786304
},
{
"epoch": 0.5,
"learning_rate": 0.0002524674022066198,
"loss": 2.7899,
"theoretical_loss": 3.486768541527819,
"tokens_seen": 1650851840
},
{
"epoch": 0.5,
"learning_rate": 0.00025245737211634906,
"loss": 2.7382,
"theoretical_loss": 3.4867565607430757,
"tokens_seen": 1650917376
},
{
"epoch": 0.5,
"learning_rate": 0.00025244734202607824,
"loss": 2.5138,
"theoretical_loss": 3.4867445805670827,
"tokens_seen": 1650982912
},
{
"epoch": 0.5,
"learning_rate": 0.0002524373119358074,
"loss": 2.7407,
"theoretical_loss": 3.486732600999784,
"tokens_seen": 1651048448
},
{
"epoch": 0.5,
"learning_rate": 0.0002524272818455366,
"loss": 2.7124,
"theoretical_loss": 3.4867206220411244,
"tokens_seen": 1651113984
},
{
"epoch": 0.5,
"learning_rate": 0.0002524172517552658,
"loss": 2.583,
"theoretical_loss": 3.48670864369105,
"tokens_seen": 1651179520
},
{
"epoch": 0.5,
"learning_rate": 0.00025240722166499496,
"loss": 2.5857,
"theoretical_loss": 3.4866966659495047,
"tokens_seen": 1651245056
},
{
"epoch": 0.5,
"learning_rate": 0.0002523971915747242,
"loss": 2.4247,
"theoretical_loss": 3.4866846888164336,
"tokens_seen": 1651310592
},
{
"epoch": 0.5,
"learning_rate": 0.0002523871614844533,
"loss": 2.5654,
"theoretical_loss": 3.486672712291782,
"tokens_seen": 1651376128
},
{
"epoch": 0.5,
"learning_rate": 0.00025237713139418256,
"loss": 2.5408,
"theoretical_loss": 3.4866607363754945,
"tokens_seen": 1651441664
},
{
"epoch": 0.5,
"objective/train/docs_used": 935410,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9581127166748047,
"objective/train/theoretical_loss": 3.4866487610675163,
"objective/train/tokens_used": 1671967200,
"theoretical_loss": 3.4866487610675163,
"tokens_seen": 1651507200
},
{
"epoch": 0.5,
"learning_rate": 0.00025236710130391174,
"loss": 2.8177,
"theoretical_loss": 3.4866487610675163,
"tokens_seen": 1651507200
},
{
"epoch": 0.5,
"learning_rate": 0.0002523570712136409,
"loss": 2.6093,
"theoretical_loss": 3.4866367863677925,
"tokens_seen": 1651572736
},
{
"epoch": 0.5,
"learning_rate": 0.0002523470411233701,
"loss": 2.7451,
"theoretical_loss": 3.4866248122762675,
"tokens_seen": 1651638272
},
{
"epoch": 0.5,
"learning_rate": 0.0002523370110330993,
"loss": 2.7039,
"theoretical_loss": 3.4866128387928867,
"tokens_seen": 1651703808
},
{
"epoch": 0.5,
"learning_rate": 0.00025232698094282847,
"loss": 2.478,
"theoretical_loss": 3.4866008659175955,
"tokens_seen": 1651769344
},
{
"epoch": 0.5,
"learning_rate": 0.0002523169508525577,
"loss": 2.7107,
"theoretical_loss": 3.486588893650338,
"tokens_seen": 1651834880
},
{
"epoch": 0.5,
"learning_rate": 0.00025230692076228683,
"loss": 2.927,
"theoretical_loss": 3.486576921991059,
"tokens_seen": 1651900416
},
{
"epoch": 0.5,
"learning_rate": 0.00025229689067201606,
"loss": 2.8432,
"theoretical_loss": 3.4865649509397048,
"tokens_seen": 1651965952
},
{
"epoch": 0.5,
"learning_rate": 0.0002522868605817452,
"loss": 2.822,
"theoretical_loss": 3.4865529804962194,
"tokens_seen": 1652031488
},
{
"epoch": 0.5,
"learning_rate": 0.00025227683049147443,
"loss": 2.9273,
"theoretical_loss": 3.4865410106605483,
"tokens_seen": 1652097024
},
{
"epoch": 0.5,
"learning_rate": 0.0002522668004012036,
"loss": 2.455,
"theoretical_loss": 3.486529041432636,
"tokens_seen": 1652162560
},
{
"epoch": 0.5,
"learning_rate": 0.0002522567703109328,
"loss": 2.4305,
"theoretical_loss": 3.4865170728124286,
"tokens_seen": 1652228096
},
{
"epoch": 0.5,
"learning_rate": 0.000252246740220662,
"loss": 2.6555,
"theoretical_loss": 3.4865051047998703,
"tokens_seen": 1652293632
},
{
"epoch": 0.5,
"learning_rate": 0.00025223671013039115,
"loss": 2.7703,
"theoretical_loss": 3.4864931373949055,
"tokens_seen": 1652359168
},
{
"epoch": 0.5,
"learning_rate": 0.0002522266800401204,
"loss": 2.6624,
"theoretical_loss": 3.486481170597481,
"tokens_seen": 1652424704
},
{
"epoch": 0.5,
"learning_rate": 0.00025221664994984957,
"loss": 2.7455,
"theoretical_loss": 3.4864692044075403,
"tokens_seen": 1652490240
},
{
"epoch": 0.5,
"learning_rate": 0.00025220661985957875,
"loss": 2.7959,
"theoretical_loss": 3.4864572388250292,
"tokens_seen": 1652555776
},
{
"epoch": 0.5,
"learning_rate": 0.00025219658976930793,
"loss": 2.4331,
"theoretical_loss": 3.486445273849893,
"tokens_seen": 1652621312
},
{
"epoch": 0.5,
"learning_rate": 0.00025218655967903717,
"loss": 2.7188,
"theoretical_loss": 3.4864333094820763,
"tokens_seen": 1652686848
},
{
"epoch": 0.5,
"learning_rate": 0.0002521765295887663,
"loss": 2.3959,
"theoretical_loss": 3.4864213457215243,
"tokens_seen": 1652752384
},
{
"epoch": 0.5,
"learning_rate": 0.00025216649949849553,
"loss": 2.4991,
"theoretical_loss": 3.4864093825681817,
"tokens_seen": 1652817920
},
{
"epoch": 0.5,
"learning_rate": 0.00025215646940822465,
"loss": 2.8917,
"theoretical_loss": 3.4863974200219943,
"tokens_seen": 1652883456
},
{
"epoch": 0.5,
"learning_rate": 0.0002521464393179539,
"loss": 2.7663,
"theoretical_loss": 3.486385458082907,
"tokens_seen": 1652948992
},
{
"epoch": 0.5,
"learning_rate": 0.00025213640922768307,
"loss": 2.7564,
"theoretical_loss": 3.486373496750865,
"tokens_seen": 1653014528
},
{
"epoch": 0.5,
"learning_rate": 0.00025212637913741225,
"loss": 2.8214,
"theoretical_loss": 3.4863615360258136,
"tokens_seen": 1653080064
},
{
"epoch": 0.5,
"objective/train/docs_used": 935942,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.318352222442627,
"objective/train/theoretical_loss": 3.4863495759076972,
"objective/train/tokens_used": 1673605600,
"theoretical_loss": 3.4863495759076972,
"tokens_seen": 1653145600
},
{
"epoch": 0.5,
"learning_rate": 0.00025211634904714143,
"loss": 2.7659,
"theoretical_loss": 3.4863495759076972,
"tokens_seen": 1653145600
},
{
"epoch": 0.5,
"learning_rate": 0.0002521063189568706,
"loss": 2.6407,
"theoretical_loss": 3.4863376163964617,
"tokens_seen": 1653211136
},
{
"epoch": 0.5,
"learning_rate": 0.0002520962888665998,
"loss": 2.8199,
"theoretical_loss": 3.4863256574920514,
"tokens_seen": 1653276672
},
{
"epoch": 0.5,
"learning_rate": 0.00025208625877632903,
"loss": 2.6398,
"theoretical_loss": 3.4863136991944126,
"tokens_seen": 1653342208
},
{
"epoch": 0.5,
"learning_rate": 0.00025207622868605816,
"loss": 2.8069,
"theoretical_loss": 3.486301741503489,
"tokens_seen": 1653407744
},
{
"epoch": 0.5,
"learning_rate": 0.0002520661985957874,
"loss": 2.7541,
"theoretical_loss": 3.4862897844192275,
"tokens_seen": 1653473280
},
{
"epoch": 0.5,
"learning_rate": 0.0002520561685055166,
"loss": 2.5113,
"theoretical_loss": 3.486277827941572,
"tokens_seen": 1653538816
},
{
"epoch": 0.5,
"learning_rate": 0.00025204613841524576,
"loss": 2.683,
"theoretical_loss": 3.4862658720704682,
"tokens_seen": 1653604352
},
{
"epoch": 0.5,
"learning_rate": 0.00025203610832497494,
"loss": 2.8376,
"theoretical_loss": 3.486253916805861,
"tokens_seen": 1653669888
},
{
"epoch": 0.5,
"learning_rate": 0.0002520260782347041,
"loss": 2.5698,
"theoretical_loss": 3.4862419621476963,
"tokens_seen": 1653735424
},
{
"epoch": 0.5,
"learning_rate": 0.0002520160481444333,
"loss": 2.8357,
"theoretical_loss": 3.4862300080959185,
"tokens_seen": 1653800960
},
{
"epoch": 0.5,
"learning_rate": 0.00025200601805416253,
"loss": 2.4212,
"theoretical_loss": 3.4862180546504726,
"tokens_seen": 1653866496
},
{
"epoch": 0.5,
"learning_rate": 0.00025199598796389166,
"loss": 2.4431,
"theoretical_loss": 3.4862061018113053,
"tokens_seen": 1653932032
},
{
"epoch": 0.5,
"learning_rate": 0.0002519859578736209,
"loss": 2.733,
"theoretical_loss": 3.48619414957836,
"tokens_seen": 1653997568
},
{
"epoch": 0.5,
"learning_rate": 0.00025197592778335,
"loss": 2.753,
"theoretical_loss": 3.4861821979515835,
"tokens_seen": 1654063104
},
{
"epoch": 0.5,
"learning_rate": 0.00025196589769307926,
"loss": 2.5493,
"theoretical_loss": 3.4861702469309197,
"tokens_seen": 1654128640
},
{
"epoch": 0.5,
"learning_rate": 0.00025195586760280844,
"loss": 2.5577,
"theoretical_loss": 3.4861582965163147,
"tokens_seen": 1654194176
},
{
"epoch": 0.5,
"learning_rate": 0.0002519458375125376,
"loss": 2.6333,
"theoretical_loss": 3.486146346707714,
"tokens_seen": 1654259712
},
{
"epoch": 0.5,
"learning_rate": 0.0002519358074222668,
"loss": 2.4194,
"theoretical_loss": 3.4861343975050616,
"tokens_seen": 1654325248
},
{
"epoch": 0.5,
"learning_rate": 0.000251925777331996,
"loss": 2.7311,
"theoretical_loss": 3.4861224489083042,
"tokens_seen": 1654390784
},
{
"epoch": 0.5,
"learning_rate": 0.00025191574724172516,
"loss": 2.5183,
"theoretical_loss": 3.486110500917386,
"tokens_seen": 1654456320
},
{
"epoch": 0.5,
"learning_rate": 0.0002519057171514544,
"loss": 2.4539,
"theoretical_loss": 3.4860985535322526,
"tokens_seen": 1654521856
},
{
"epoch": 0.5,
"learning_rate": 0.0002518956870611835,
"loss": 2.4483,
"theoretical_loss": 3.48608660675285,
"tokens_seen": 1654587392
},
{
"epoch": 0.5,
"learning_rate": 0.00025188565697091276,
"loss": 2.643,
"theoretical_loss": 3.4860746605791224,
"tokens_seen": 1654652928
},
{
"epoch": 0.5,
"learning_rate": 0.00025187562688064194,
"loss": 2.7149,
"theoretical_loss": 3.486062715011016,
"tokens_seen": 1654718464
},
{
"epoch": 0.5,
"objective/train/docs_used": 937360,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7586429119110107,
"objective/train/theoretical_loss": 3.4860507700484757,
"objective/train/tokens_used": 1675244000,
"theoretical_loss": 3.4860507700484757,
"tokens_seen": 1654784000
},
{
"epoch": 0.5,
"learning_rate": 0.0002518655967903711,
"loss": 2.5052,
"theoretical_loss": 3.4860507700484757,
"tokens_seen": 1654784000
},
{
"epoch": 0.5,
"learning_rate": 0.0002518555667001003,
"loss": 2.5171,
"theoretical_loss": 3.486038825691447,
"tokens_seen": 1654849536
},
{
"epoch": 0.5,
"learning_rate": 0.0002518455366098295,
"loss": 2.6351,
"theoretical_loss": 3.4860268819398748,
"tokens_seen": 1654915072
},
{
"epoch": 0.5,
"learning_rate": 0.00025183550651955867,
"loss": 2.5702,
"theoretical_loss": 3.486014938793705,
"tokens_seen": 1654980608
},
{
"epoch": 0.5,
"learning_rate": 0.0002518254764292879,
"loss": 2.5823,
"theoretical_loss": 3.4860029962528825,
"tokens_seen": 1655046144
},
{
"epoch": 0.5,
"learning_rate": 0.00025181544633901703,
"loss": 2.5688,
"theoretical_loss": 3.4859910543173527,
"tokens_seen": 1655111680
},
{
"epoch": 0.5,
"learning_rate": 0.00025180541624874627,
"loss": 2.6971,
"theoretical_loss": 3.485979112987061,
"tokens_seen": 1655177216
},
{
"epoch": 0.5,
"learning_rate": 0.0002517953861584754,
"loss": 2.718,
"theoretical_loss": 3.4859671722619536,
"tokens_seen": 1655242752
},
{
"epoch": 0.5,
"learning_rate": 0.00025178535606820463,
"loss": 2.7038,
"theoretical_loss": 3.4859552321419742,
"tokens_seen": 1655308288
},
{
"epoch": 0.5,
"learning_rate": 0.0002517753259779338,
"loss": 2.3637,
"theoretical_loss": 3.4859432926270695,
"tokens_seen": 1655373824
},
{
"epoch": 0.5,
"learning_rate": 0.000251765295887663,
"loss": 2.6107,
"theoretical_loss": 3.4859313537171843,
"tokens_seen": 1655439360
},
{
"epoch": 0.5,
"learning_rate": 0.00025175526579739217,
"loss": 2.4522,
"theoretical_loss": 3.4859194154122646,
"tokens_seen": 1655504896
},
{
"epoch": 0.5,
"learning_rate": 0.00025174523570712135,
"loss": 2.8168,
"theoretical_loss": 3.4859074777122547,
"tokens_seen": 1655570432
},
{
"epoch": 0.5,
"learning_rate": 0.00025173520561685053,
"loss": 2.399,
"theoretical_loss": 3.485895540617101,
"tokens_seen": 1655635968
},
{
"epoch": 0.5,
"learning_rate": 0.00025172517552657977,
"loss": 2.6582,
"theoretical_loss": 3.485883604126749,
"tokens_seen": 1655701504
},
{
"epoch": 0.5,
"learning_rate": 0.0002517151454363089,
"loss": 2.6639,
"theoretical_loss": 3.485871668241143,
"tokens_seen": 1655767040
},
{
"epoch": 0.5,
"learning_rate": 0.00025170511534603813,
"loss": 2.576,
"theoretical_loss": 3.4858597329602294,
"tokens_seen": 1655832576
},
{
"epoch": 0.5,
"learning_rate": 0.0002516950852557673,
"loss": 2.895,
"theoretical_loss": 3.4858477982839533,
"tokens_seen": 1655898112
},
{
"epoch": 0.5,
"learning_rate": 0.0002516850551654965,
"loss": 2.5179,
"theoretical_loss": 3.4858358642122607,
"tokens_seen": 1655963648
},
{
"epoch": 0.5,
"learning_rate": 0.0002516750250752257,
"loss": 2.5616,
"theoretical_loss": 3.485823930745096,
"tokens_seen": 1656029184
},
{
"epoch": 0.5,
"learning_rate": 0.00025166499498495486,
"loss": 2.7319,
"theoretical_loss": 3.485811997882405,
"tokens_seen": 1656094720
},
{
"epoch": 0.5,
"learning_rate": 0.00025165496489468404,
"loss": 2.5798,
"theoretical_loss": 3.485800065624134,
"tokens_seen": 1656160256
},
{
"epoch": 0.5,
"learning_rate": 0.00025164493480441327,
"loss": 2.7706,
"theoretical_loss": 3.4857881339702272,
"tokens_seen": 1656225792
},
{
"epoch": 0.5,
"learning_rate": 0.0002516349047141424,
"loss": 2.6082,
"theoretical_loss": 3.4857762029206314,
"tokens_seen": 1656291328
},
{
"epoch": 0.5,
"learning_rate": 0.00025162487462387163,
"loss": 2.8266,
"theoretical_loss": 3.485764272475291,
"tokens_seen": 1656356864
},
{
"epoch": 0.5,
"objective/train/docs_used": 938134,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.107076644897461,
"objective/train/theoretical_loss": 3.485752342634152,
"objective/train/tokens_used": 1676882400,
"theoretical_loss": 3.485752342634152,
"tokens_seen": 1656422400
},
{
"epoch": 0.5,
"learning_rate": 0.00025161484453360076,
"loss": 2.8194,
"theoretical_loss": 3.485752342634152,
"tokens_seen": 1656422400
},
{
"epoch": 0.5,
"learning_rate": 0.00025160481444333,
"loss": 2.4422,
"theoretical_loss": 3.48574041339716,
"tokens_seen": 1656487936
},
{
"epoch": 0.5,
"learning_rate": 0.0002515947843530592,
"loss": 2.5503,
"theoretical_loss": 3.48572848476426,
"tokens_seen": 1656553472
},
{
"epoch": 0.5,
"learning_rate": 0.00025158475426278836,
"loss": 2.8605,
"theoretical_loss": 3.4857165567353974,
"tokens_seen": 1656619008
},
{
"epoch": 0.5,
"learning_rate": 0.00025157472417251754,
"loss": 2.6617,
"theoretical_loss": 3.4857046293105185,
"tokens_seen": 1656684544
},
{
"epoch": 0.5,
"learning_rate": 0.0002515646940822468,
"loss": 2.751,
"theoretical_loss": 3.4856927024895685,
"tokens_seen": 1656750080
},
{
"epoch": 0.5,
"learning_rate": 0.0002515546639919759,
"loss": 2.2641,
"theoretical_loss": 3.485680776272493,
"tokens_seen": 1656815616
},
{
"epoch": 0.5,
"learning_rate": 0.00025154463390170514,
"loss": 2.585,
"theoretical_loss": 3.4856688506592373,
"tokens_seen": 1656881152
},
{
"epoch": 0.5,
"learning_rate": 0.00025153460381143426,
"loss": 2.768,
"theoretical_loss": 3.4856569256497467,
"tokens_seen": 1656946688
},
{
"epoch": 0.5,
"learning_rate": 0.0002515245737211635,
"loss": 2.7102,
"theoretical_loss": 3.4856450012439675,
"tokens_seen": 1657012224
},
{
"epoch": 0.5,
"learning_rate": 0.0002515145436308927,
"loss": 2.6474,
"theoretical_loss": 3.485633077441845,
"tokens_seen": 1657077760
},
{
"epoch": 0.5,
"learning_rate": 0.00025150451354062186,
"loss": 2.564,
"theoretical_loss": 3.485621154243325,
"tokens_seen": 1657143296
},
{
"epoch": 0.5,
"learning_rate": 0.0002514944834503511,
"loss": 2.7374,
"theoretical_loss": 3.485609231648352,
"tokens_seen": 1657208832
},
{
"epoch": 0.5,
"learning_rate": 0.0002514844533600802,
"loss": 2.8022,
"theoretical_loss": 3.4855973096568724,
"tokens_seen": 1657274368
},
{
"epoch": 0.5,
"learning_rate": 0.00025147442326980946,
"loss": 2.664,
"theoretical_loss": 3.4855853882688317,
"tokens_seen": 1657339904
},
{
"epoch": 0.5,
"learning_rate": 0.00025146439317953864,
"loss": 2.6154,
"theoretical_loss": 3.485573467484176,
"tokens_seen": 1657405440
},
{
"epoch": 0.5,
"learning_rate": 0.0002514543630892678,
"loss": 2.5988,
"theoretical_loss": 3.48556154730285,
"tokens_seen": 1657470976
},
{
"epoch": 0.5,
"learning_rate": 0.000251444332998997,
"loss": 2.7223,
"theoretical_loss": 3.4855496277248,
"tokens_seen": 1657536512
},
{
"epoch": 0.5,
"learning_rate": 0.0002514343029087262,
"loss": 2.8675,
"theoretical_loss": 3.4855377087499706,
"tokens_seen": 1657602048
},
{
"epoch": 0.5,
"learning_rate": 0.00025142427281845536,
"loss": 3.0463,
"theoretical_loss": 3.4855257903783086,
"tokens_seen": 1657667584
},
{
"epoch": 0.5,
"learning_rate": 0.0002514142427281846,
"loss": 2.9043,
"theoretical_loss": 3.4855138726097596,
"tokens_seen": 1657733120
},
{
"epoch": 0.5,
"learning_rate": 0.0002514042126379137,
"loss": 2.9839,
"theoretical_loss": 3.485501955444268,
"tokens_seen": 1657798656
},
{
"epoch": 0.5,
"learning_rate": 0.00025139418254764296,
"loss": 2.7193,
"theoretical_loss": 3.485490038881781,
"tokens_seen": 1657864192
},
{
"epoch": 0.5,
"learning_rate": 0.00025138415245737214,
"loss": 2.5492,
"theoretical_loss": 3.485478122922243,
"tokens_seen": 1657929728
},
{
"epoch": 0.5,
"learning_rate": 0.0002513741223671013,
"loss": 2.5346,
"theoretical_loss": 3.485466207565601,
"tokens_seen": 1657995264
},
{
"epoch": 0.5,
"objective/train/docs_used": 939068,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8123443126678467,
"objective/train/theoretical_loss": 3.4854542928117986,
"objective/train/tokens_used": 1678520800,
"theoretical_loss": 3.4854542928117986,
"tokens_seen": 1658060800
},
{
"epoch": 0.5,
"learning_rate": 0.0002513640922768305,
"loss": 2.7642,
"theoretical_loss": 3.4854542928117986,
"tokens_seen": 1658060800
},
{
"epoch": 0.5,
"learning_rate": 0.0002513540621865597,
"loss": 2.767,
"theoretical_loss": 3.4854423786607835,
"tokens_seen": 1658126336
},
{
"epoch": 0.5,
"learning_rate": 0.00025134403209628887,
"loss": 2.8462,
"theoretical_loss": 3.4854304651125005,
"tokens_seen": 1658191872
},
{
"epoch": 0.5,
"learning_rate": 0.0002513340020060181,
"loss": 2.6498,
"theoretical_loss": 3.4854185521668954,
"tokens_seen": 1658257408
},
{
"epoch": 0.5,
"learning_rate": 0.00025132397191574723,
"loss": 2.933,
"theoretical_loss": 3.485406639823914,
"tokens_seen": 1658322944
},
{
"epoch": 0.5,
"learning_rate": 0.00025131394182547647,
"loss": 2.5583,
"theoretical_loss": 3.485394728083502,
"tokens_seen": 1658388480
},
{
"epoch": 0.5,
"learning_rate": 0.0002513039117352056,
"loss": 2.6016,
"theoretical_loss": 3.485382816945605,
"tokens_seen": 1658454016
},
{
"epoch": 0.5,
"learning_rate": 0.00025129388164493483,
"loss": 2.726,
"theoretical_loss": 3.485370906410168,
"tokens_seen": 1658519552
},
{
"epoch": 0.5,
"learning_rate": 0.000251283851554664,
"loss": 2.6549,
"theoretical_loss": 3.485358996477138,
"tokens_seen": 1658585088
},
{
"epoch": 0.5,
"learning_rate": 0.0002512738214643932,
"loss": 2.6745,
"theoretical_loss": 3.4853470871464602,
"tokens_seen": 1658650624
},
{
"epoch": 0.5,
"learning_rate": 0.00025126379137412237,
"loss": 2.8756,
"theoretical_loss": 3.4853351784180804,
"tokens_seen": 1658716160
},
{
"epoch": 0.5,
"learning_rate": 0.00025125376128385155,
"loss": 2.6573,
"theoretical_loss": 3.4853232702919437,
"tokens_seen": 1658781696
},
{
"epoch": 0.5,
"learning_rate": 0.00025124373119358073,
"loss": 2.7919,
"theoretical_loss": 3.4853113627679972,
"tokens_seen": 1658847232
},
{
"epoch": 0.5,
"learning_rate": 0.00025123370110330997,
"loss": 2.7807,
"theoretical_loss": 3.485299455846185,
"tokens_seen": 1658912768
},
{
"epoch": 0.5,
"learning_rate": 0.0002512236710130391,
"loss": 2.5404,
"theoretical_loss": 3.485287549526454,
"tokens_seen": 1658978304
},
{
"epoch": 0.5,
"learning_rate": 0.00025121364092276833,
"loss": 2.6067,
"theoretical_loss": 3.4852756438087504,
"tokens_seen": 1659043840
},
{
"epoch": 0.5,
"learning_rate": 0.0002512036108324975,
"loss": 2.7008,
"theoretical_loss": 3.4852637386930185,
"tokens_seen": 1659109376
},
{
"epoch": 0.5,
"learning_rate": 0.0002511935807422267,
"loss": 2.5233,
"theoretical_loss": 3.4852518341792056,
"tokens_seen": 1659174912
},
{
"epoch": 0.5,
"learning_rate": 0.0002511835506519559,
"loss": 2.9169,
"theoretical_loss": 3.485239930267256,
"tokens_seen": 1659240448
},
{
"epoch": 0.5,
"learning_rate": 0.00025117352056168506,
"loss": 2.7245,
"theoretical_loss": 3.4852280269571168,
"tokens_seen": 1659305984
},
{
"epoch": 0.5,
"learning_rate": 0.00025116349047141424,
"loss": 2.6187,
"theoretical_loss": 3.4852161242487325,
"tokens_seen": 1659371520
},
{
"epoch": 0.5,
"learning_rate": 0.00025115346038114347,
"loss": 2.5737,
"theoretical_loss": 3.4852042221420505,
"tokens_seen": 1659437056
},
{
"epoch": 0.5,
"learning_rate": 0.0002511434302908726,
"loss": 2.5197,
"theoretical_loss": 3.4851923206370157,
"tokens_seen": 1659502592
},
{
"epoch": 0.5,
"learning_rate": 0.00025113340020060183,
"loss": 2.6014,
"theoretical_loss": 3.485180419733574,
"tokens_seen": 1659568128
},
{
"epoch": 0.5,
"learning_rate": 0.00025112337011033096,
"loss": 2.803,
"theoretical_loss": 3.485168519431671,
"tokens_seen": 1659633664
},
{
"epoch": 0.5,
"objective/train/docs_used": 939826,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8426969051361084,
"objective/train/theoretical_loss": 3.485156619731253,
"objective/train/tokens_used": 1680159200,
"theoretical_loss": 3.485156619731253,
"tokens_seen": 1659699200
},
{
"epoch": 0.5,
"learning_rate": 0.0002511133400200602,
"loss": 2.6879,
"theoretical_loss": 3.485156619731253,
"tokens_seen": 1659699200
},
{
"epoch": 0.5,
"learning_rate": 0.0002511033099297894,
"loss": 2.8888,
"theoretical_loss": 3.4851447206322654,
"tokens_seen": 1659764736
},
{
"epoch": 0.5,
"learning_rate": 0.00025109327983951856,
"loss": 2.4908,
"theoretical_loss": 3.4851328221346547,
"tokens_seen": 1659830272
},
{
"epoch": 0.5,
"learning_rate": 0.00025108324974924774,
"loss": 2.5111,
"theoretical_loss": 3.485120924238366,
"tokens_seen": 1659895808
},
{
"epoch": 0.5,
"learning_rate": 0.000251073219658977,
"loss": 2.7824,
"theoretical_loss": 3.485109026943346,
"tokens_seen": 1659961344
},
{
"epoch": 0.5,
"learning_rate": 0.0002510631895687061,
"loss": 2.7458,
"theoretical_loss": 3.4850971302495397,
"tokens_seen": 1660026880
},
{
"epoch": 0.5,
"learning_rate": 0.00025105315947843534,
"loss": 2.6799,
"theoretical_loss": 3.4850852341568936,
"tokens_seen": 1660092416
},
{
"epoch": 0.5,
"learning_rate": 0.00025104312938816446,
"loss": 2.3934,
"theoretical_loss": 3.485073338665354,
"tokens_seen": 1660157952
},
{
"epoch": 0.5,
"learning_rate": 0.0002510330992978937,
"loss": 2.7097,
"theoretical_loss": 3.4850614437748657,
"tokens_seen": 1660223488
},
{
"epoch": 0.5,
"learning_rate": 0.0002510230692076229,
"loss": 2.8584,
"theoretical_loss": 3.4850495494853746,
"tokens_seen": 1660289024
},
{
"epoch": 0.5,
"learning_rate": 0.00025101303911735206,
"loss": 2.7306,
"theoretical_loss": 3.4850376557968277,
"tokens_seen": 1660354560
},
{
"epoch": 0.5,
"learning_rate": 0.00025100300902708124,
"loss": 2.8289,
"theoretical_loss": 3.4850257627091707,
"tokens_seen": 1660420096
},
{
"epoch": 0.5,
"learning_rate": 0.0002509929789368104,
"loss": 2.5776,
"theoretical_loss": 3.4850138702223488,
"tokens_seen": 1660485632
},
{
"epoch": 0.5,
"learning_rate": 0.0002509829488465396,
"loss": 2.7092,
"theoretical_loss": 3.485001978336308,
"tokens_seen": 1660551168
},
{
"epoch": 0.5,
"learning_rate": 0.00025097291875626884,
"loss": 3.0718,
"theoretical_loss": 3.4849900870509947,
"tokens_seen": 1660616704
},
{
"epoch": 0.5,
"learning_rate": 0.00025096288866599797,
"loss": 2.5621,
"theoretical_loss": 3.484978196366355,
"tokens_seen": 1660682240
},
{
"epoch": 0.5,
"learning_rate": 0.0002509528585757272,
"loss": 2.5954,
"theoretical_loss": 3.4849663062823346,
"tokens_seen": 1660747776
},
{
"epoch": 0.5,
"learning_rate": 0.00025094282848545633,
"loss": 2.6468,
"theoretical_loss": 3.484954416798879,
"tokens_seen": 1660813312
},
{
"epoch": 0.5,
"learning_rate": 0.00025093279839518556,
"loss": 2.7334,
"theoretical_loss": 3.4849425279159347,
"tokens_seen": 1660878848
},
{
"epoch": 0.5,
"learning_rate": 0.00025092276830491475,
"loss": 2.7161,
"theoretical_loss": 3.484930639633448,
"tokens_seen": 1660944384
},
{
"epoch": 0.5,
"learning_rate": 0.0002509127382146439,
"loss": 2.8489,
"theoretical_loss": 3.484918751951364,
"tokens_seen": 1661009920
},
{
"epoch": 0.5,
"learning_rate": 0.0002509027081243731,
"loss": 2.6612,
"theoretical_loss": 3.484906864869629,
"tokens_seen": 1661075456
},
{
"epoch": 0.5,
"learning_rate": 0.00025089267803410234,
"loss": 2.7213,
"theoretical_loss": 3.4848949783881893,
"tokens_seen": 1661140992
},
{
"epoch": 0.5,
"learning_rate": 0.00025088264794383147,
"loss": 2.6237,
"theoretical_loss": 3.484883092506991,
"tokens_seen": 1661206528
},
{
"epoch": 0.5,
"learning_rate": 0.0002508726178535607,
"loss": 2.3719,
"theoretical_loss": 3.4848712072259795,
"tokens_seen": 1661272064
},
{
"epoch": 0.5,
"objective/train/docs_used": 941080,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.887375831604004,
"objective/train/theoretical_loss": 3.484859322545101,
"objective/train/tokens_used": 1681797600,
"theoretical_loss": 3.484859322545101,
"tokens_seen": 1661337600
},
{
"epoch": 0.5,
"learning_rate": 0.00025086258776328983,
"loss": 2.7431,
"theoretical_loss": 3.484859322545101,
"tokens_seen": 1661337600
},
{
"epoch": 0.5,
"learning_rate": 0.00025085255767301907,
"loss": 2.7113,
"theoretical_loss": 3.4848474384643024,
"tokens_seen": 1661403136
},
{
"epoch": 0.5,
"learning_rate": 0.00025084252758274825,
"loss": 2.611,
"theoretical_loss": 3.484835554983529,
"tokens_seen": 1661468672
},
{
"epoch": 0.5,
"learning_rate": 0.00025083249749247743,
"loss": 2.6565,
"theoretical_loss": 3.484823672102726,
"tokens_seen": 1661534208
},
{
"epoch": 0.5,
"learning_rate": 0.0002508224674022066,
"loss": 2.706,
"theoretical_loss": 3.484811789821841,
"tokens_seen": 1661599744
},
{
"epoch": 0.5,
"learning_rate": 0.0002508124373119358,
"loss": 2.5169,
"theoretical_loss": 3.484799908140819,
"tokens_seen": 1661665280
},
{
"epoch": 0.5,
"learning_rate": 0.000250802407221665,
"loss": 2.5659,
"theoretical_loss": 3.4847880270596066,
"tokens_seen": 1661730816
},
{
"epoch": 0.5,
"learning_rate": 0.0002507923771313942,
"loss": 2.2855,
"theoretical_loss": 3.4847761465781497,
"tokens_seen": 1661796352
},
{
"epoch": 0.5,
"learning_rate": 0.00025078234704112334,
"loss": 2.74,
"theoretical_loss": 3.484764266696394,
"tokens_seen": 1661861888
},
{
"epoch": 0.5,
"learning_rate": 0.00025077231695085257,
"loss": 3.0564,
"theoretical_loss": 3.4847523874142867,
"tokens_seen": 1661927424
},
{
"epoch": 0.5,
"learning_rate": 0.0002507622868605817,
"loss": 2.6321,
"theoretical_loss": 3.4847405087317727,
"tokens_seen": 1661992960
},
{
"epoch": 0.5,
"learning_rate": 0.00025075225677031093,
"loss": 2.617,
"theoretical_loss": 3.4847286306487986,
"tokens_seen": 1662058496
},
{
"epoch": 0.5,
"learning_rate": 0.00025074222668004017,
"loss": 2.7945,
"theoretical_loss": 3.48471675316531,
"tokens_seen": 1662124032
},
{
"epoch": 0.5,
"learning_rate": 0.0002507321965897693,
"loss": 2.7132,
"theoretical_loss": 3.4847048762812536,
"tokens_seen": 1662189568
},
{
"epoch": 0.5,
"learning_rate": 0.00025072216649949853,
"loss": 2.7107,
"theoretical_loss": 3.4846929999965757,
"tokens_seen": 1662255104
},
{
"epoch": 0.5,
"learning_rate": 0.0002507121364092277,
"loss": 2.6719,
"theoretical_loss": 3.484681124311222,
"tokens_seen": 1662320640
},
{
"epoch": 0.5,
"learning_rate": 0.0002507021063189569,
"loss": 2.6054,
"theoretical_loss": 3.4846692492251385,
"tokens_seen": 1662386176
},
{
"epoch": 0.5,
"learning_rate": 0.0002506920762286861,
"loss": 2.6526,
"theoretical_loss": 3.484657374738272,
"tokens_seen": 1662451712
},
{
"epoch": 0.5,
"learning_rate": 0.00025068204613841526,
"loss": 2.6845,
"theoretical_loss": 3.4846455008505677,
"tokens_seen": 1662517248
},
{
"epoch": 0.5,
"learning_rate": 0.00025067201604814444,
"loss": 2.7385,
"theoretical_loss": 3.484633627561972,
"tokens_seen": 1662582784
},
{
"epoch": 0.5,
"learning_rate": 0.00025066198595787367,
"loss": 2.7237,
"theoretical_loss": 3.484621754872432,
"tokens_seen": 1662648320
},
{
"epoch": 0.5,
"learning_rate": 0.0002506519558676028,
"loss": 2.7437,
"theoretical_loss": 3.4846098827818928,
"tokens_seen": 1662713856
},
{
"epoch": 0.5,
"learning_rate": 0.00025064192577733203,
"loss": 2.654,
"theoretical_loss": 3.4845980112903003,
"tokens_seen": 1662779392
},
{
"epoch": 0.5,
"learning_rate": 0.00025063189568706116,
"loss": 2.5718,
"theoretical_loss": 3.484586140397602,
"tokens_seen": 1662844928
},
{
"epoch": 0.5,
"learning_rate": 0.0002506218655967904,
"loss": 2.563,
"theoretical_loss": 3.484574270103743,
"tokens_seen": 1662910464
},
{
"epoch": 0.5,
"objective/train/docs_used": 941634,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8897016048431396,
"objective/train/theoretical_loss": 3.484562400408671,
"objective/train/tokens_used": 1683436000,
"theoretical_loss": 3.484562400408671,
"tokens_seen": 1662976000
},
{
"epoch": 0.5,
"learning_rate": 0.0002506118355065196,
"loss": 2.9015,
"theoretical_loss": 3.484562400408671,
"tokens_seen": 1662976000
},
{
"epoch": 0.5,
"learning_rate": 0.00025060180541624876,
"loss": 2.6929,
"theoretical_loss": 3.4845505313123297,
"tokens_seen": 1663041536
},
{
"epoch": 0.5,
"learning_rate": 0.00025059177532597794,
"loss": 2.829,
"theoretical_loss": 3.4845386628146677,
"tokens_seen": 1663107072
},
{
"epoch": 0.5,
"learning_rate": 0.0002505817452357072,
"loss": 2.9222,
"theoretical_loss": 3.4845267949156296,
"tokens_seen": 1663172608
},
{
"epoch": 0.5,
"learning_rate": 0.0002505717151454363,
"loss": 2.4964,
"theoretical_loss": 3.4845149276151623,
"tokens_seen": 1663238144
},
{
"epoch": 0.5,
"learning_rate": 0.00025056168505516554,
"loss": 2.827,
"theoretical_loss": 3.484503060913212,
"tokens_seen": 1663303680
},
{
"epoch": 0.5,
"learning_rate": 0.00025055165496489466,
"loss": 2.6768,
"theoretical_loss": 3.484491194809725,
"tokens_seen": 1663369216
},
{
"epoch": 0.5,
"learning_rate": 0.0002505416248746239,
"loss": 2.6277,
"theoretical_loss": 3.484479329304647,
"tokens_seen": 1663434752
},
{
"epoch": 0.5,
"learning_rate": 0.0002505315947843531,
"loss": 2.8537,
"theoretical_loss": 3.484467464397925,
"tokens_seen": 1663500288
},
{
"epoch": 0.5,
"learning_rate": 0.00025052156469408226,
"loss": 2.8906,
"theoretical_loss": 3.484455600089505,
"tokens_seen": 1663565824
},
{
"epoch": 0.5,
"learning_rate": 0.00025051153460381144,
"loss": 2.7502,
"theoretical_loss": 3.4844437363793332,
"tokens_seen": 1663631360
},
{
"epoch": 0.5,
"learning_rate": 0.0002505015045135406,
"loss": 2.7507,
"theoretical_loss": 3.4844318732673556,
"tokens_seen": 1663696896
},
{
"epoch": 0.5,
"learning_rate": 0.0002504914744232698,
"loss": 2.8504,
"theoretical_loss": 3.484420010753519,
"tokens_seen": 1663762432
},
{
"epoch": 0.5,
"learning_rate": 0.00025048144433299904,
"loss": 2.7839,
"theoretical_loss": 3.484408148837769,
"tokens_seen": 1663827968
},
{
"epoch": 0.5,
"learning_rate": 0.00025047141424272817,
"loss": 2.6066,
"theoretical_loss": 3.4843962875200525,
"tokens_seen": 1663893504
},
{
"epoch": 0.5,
"learning_rate": 0.0002504613841524574,
"loss": 2.7869,
"theoretical_loss": 3.484384426800316,
"tokens_seen": 1663959040
},
{
"epoch": 0.5,
"learning_rate": 0.00025045135406218653,
"loss": 2.7156,
"theoretical_loss": 3.4843725666785046,
"tokens_seen": 1664024576
},
{
"epoch": 0.5,
"learning_rate": 0.00025044132397191576,
"loss": 2.7444,
"theoretical_loss": 3.484360707154566,
"tokens_seen": 1664090112
},
{
"epoch": 0.5,
"learning_rate": 0.00025043129388164495,
"loss": 2.6121,
"theoretical_loss": 3.484348848228446,
"tokens_seen": 1664155648
},
{
"epoch": 0.5,
"learning_rate": 0.0002504212637913741,
"loss": 2.6596,
"theoretical_loss": 3.4843369899000907,
"tokens_seen": 1664221184
},
{
"epoch": 0.5,
"learning_rate": 0.0002504112337011033,
"loss": 2.4153,
"theoretical_loss": 3.484325132169446,
"tokens_seen": 1664286720
},
{
"epoch": 0.5,
"learning_rate": 0.00025040120361083254,
"loss": 2.7088,
"theoretical_loss": 3.4843132750364596,
"tokens_seen": 1664352256
},
{
"epoch": 0.5,
"learning_rate": 0.00025039117352056167,
"loss": 2.9839,
"theoretical_loss": 3.4843014185010768,
"tokens_seen": 1664417792
},
{
"epoch": 0.5,
"learning_rate": 0.0002503811434302909,
"loss": 2.7318,
"theoretical_loss": 3.4842895625632444,
"tokens_seen": 1664483328
},
{
"epoch": 0.5,
"learning_rate": 0.00025037111334002003,
"loss": 2.8972,
"theoretical_loss": 3.4842777072229083,
"tokens_seen": 1664548864
},
{
"epoch": 0.5,
"objective/train/docs_used": 943199,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.909115791320801,
"objective/train/theoretical_loss": 3.484265852480015,
"objective/train/tokens_used": 1685074400,
"theoretical_loss": 3.484265852480015,
"tokens_seen": 1664614400
},
{
"epoch": 0.5,
"learning_rate": 0.00025036108324974927,
"loss": 2.8786,
"theoretical_loss": 3.484265852480015,
"tokens_seen": 1664614400
},
{
"epoch": 0.5,
"learning_rate": 0.00025035105315947845,
"loss": 2.6694,
"theoretical_loss": 3.4842539983345118,
"tokens_seen": 1664679936
},
{
"epoch": 0.5,
"learning_rate": 0.00025034102306920763,
"loss": 2.5698,
"theoretical_loss": 3.484242144786344,
"tokens_seen": 1664745472
},
{
"epoch": 0.5,
"learning_rate": 0.0002503309929789368,
"loss": 2.8168,
"theoretical_loss": 3.4842302918354577,
"tokens_seen": 1664811008
},
{
"epoch": 0.5,
"learning_rate": 0.000250320962888666,
"loss": 2.7094,
"theoretical_loss": 3.4842184394818,
"tokens_seen": 1664876544
},
{
"epoch": 0.5,
"learning_rate": 0.0002503109327983952,
"loss": 2.7248,
"theoretical_loss": 3.4842065877253177,
"tokens_seen": 1664942080
},
{
"epoch": 0.5,
"learning_rate": 0.0002503009027081244,
"loss": 2.4298,
"theoretical_loss": 3.4841947365659562,
"tokens_seen": 1665007616
},
{
"epoch": 0.5,
"learning_rate": 0.00025029087261785354,
"loss": 2.6989,
"theoretical_loss": 3.4841828860036625,
"tokens_seen": 1665073152
},
{
"epoch": 0.5,
"learning_rate": 0.00025028084252758277,
"loss": 2.5871,
"theoretical_loss": 3.484171036038383,
"tokens_seen": 1665138688
},
{
"epoch": 0.5,
"learning_rate": 0.0002502708124373119,
"loss": 2.5921,
"theoretical_loss": 3.4841591866700643,
"tokens_seen": 1665204224
},
{
"epoch": 0.5,
"learning_rate": 0.00025026078234704113,
"loss": 2.4726,
"theoretical_loss": 3.4841473378986523,
"tokens_seen": 1665269760
},
{
"epoch": 0.5,
"learning_rate": 0.0002502507522567703,
"loss": 2.5665,
"theoretical_loss": 3.4841354897240935,
"tokens_seen": 1665335296
},
{
"epoch": 0.5,
"learning_rate": 0.0002502407221664995,
"loss": 2.4692,
"theoretical_loss": 3.4841236421463346,
"tokens_seen": 1665400832
},
{
"epoch": 0.5,
"learning_rate": 0.0002502306920762287,
"loss": 2.6641,
"theoretical_loss": 3.4841117951653224,
"tokens_seen": 1665466368
},
{
"epoch": 0.5,
"learning_rate": 0.0002502206619859579,
"loss": 2.5357,
"theoretical_loss": 3.4840999487810027,
"tokens_seen": 1665531904
},
{
"epoch": 0.5,
"learning_rate": 0.00025021063189568704,
"loss": 2.6539,
"theoretical_loss": 3.4840881029933226,
"tokens_seen": 1665597440
},
{
"epoch": 0.5,
"learning_rate": 0.0002502006018054163,
"loss": 2.6219,
"theoretical_loss": 3.484076257802228,
"tokens_seen": 1665662976
},
{
"epoch": 0.5,
"learning_rate": 0.0002501905717151454,
"loss": 2.6376,
"theoretical_loss": 3.4840644132076655,
"tokens_seen": 1665728512
},
{
"epoch": 0.5,
"learning_rate": 0.00025018054162487464,
"loss": 2.6675,
"theoretical_loss": 3.484052569209582,
"tokens_seen": 1665794048
},
{
"epoch": 0.5,
"learning_rate": 0.0002501705115346038,
"loss": 2.6231,
"theoretical_loss": 3.4840407258079233,
"tokens_seen": 1665859584
},
{
"epoch": 0.5,
"learning_rate": 0.000250160481444333,
"loss": 2.7422,
"theoretical_loss": 3.4840288830026367,
"tokens_seen": 1665925120
},
{
"epoch": 0.5,
"learning_rate": 0.0002501504513540622,
"loss": 2.7113,
"theoretical_loss": 3.484017040793668,
"tokens_seen": 1665990656
},
{
"epoch": 0.5,
"learning_rate": 0.00025014042126379136,
"loss": 2.5508,
"theoretical_loss": 3.4840051991809644,
"tokens_seen": 1666056192
},
{
"epoch": 0.5,
"learning_rate": 0.00025013039117352054,
"loss": 2.6076,
"theoretical_loss": 3.4839933581644718,
"tokens_seen": 1666121728
},
{
"epoch": 0.5,
"learning_rate": 0.0002501203610832498,
"loss": 2.5103,
"theoretical_loss": 3.4839815177441373,
"tokens_seen": 1666187264
},
{
"epoch": 0.5,
"objective/train/docs_used": 943856,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.683974027633667,
"objective/train/theoretical_loss": 3.4839696779199065,
"objective/train/tokens_used": 1686712800,
"theoretical_loss": 3.4839696779199065,
"tokens_seen": 1666252800
},
{
"epoch": 0.5,
"learning_rate": 0.0002501103309929789,
"loss": 2.7177,
"theoretical_loss": 3.4839696779199065,
"tokens_seen": 1666252800
},
{
"epoch": 0.5,
"learning_rate": 0.00025010030090270814,
"loss": 2.7185,
"theoretical_loss": 3.483957838691727,
"tokens_seen": 1666318336
},
{
"epoch": 0.5,
"learning_rate": 0.00025009027081243727,
"loss": 2.7835,
"theoretical_loss": 3.483946000059545,
"tokens_seen": 1666383872
},
{
"epoch": 0.5,
"learning_rate": 0.0002500802407221665,
"loss": 2.566,
"theoretical_loss": 3.483934162023307,
"tokens_seen": 1666449408
},
{
"epoch": 0.51,
"learning_rate": 0.0002500702106318957,
"loss": 2.6403,
"theoretical_loss": 3.4839223245829594,
"tokens_seen": 1666514944
},
{
"epoch": 0.51,
"learning_rate": 0.00025006018054162486,
"loss": 2.6204,
"theoretical_loss": 3.4839104877384486,
"tokens_seen": 1666580480
},
{
"epoch": 0.51,
"learning_rate": 0.00025005015045135405,
"loss": 2.5078,
"theoretical_loss": 3.483898651489722,
"tokens_seen": 1666646016
},
{
"epoch": 0.51,
"learning_rate": 0.0002500401203610833,
"loss": 2.3353,
"theoretical_loss": 3.4838868158367253,
"tokens_seen": 1666711552
},
{
"epoch": 0.51,
"learning_rate": 0.0002500300902708124,
"loss": 2.7002,
"theoretical_loss": 3.4838749807794063,
"tokens_seen": 1666777088
},
{
"epoch": 0.51,
"learning_rate": 0.00025002006018054164,
"loss": 2.7502,
"theoretical_loss": 3.4838631463177103,
"tokens_seen": 1666842624
},
{
"epoch": 0.51,
"learning_rate": 0.00025001003009027077,
"loss": 2.3402,
"theoretical_loss": 3.483851312451584,
"tokens_seen": 1666908160
},
{
"epoch": 0.51,
"learning_rate": 0.00025,
"loss": 2.7361,
"theoretical_loss": 3.4838394791809746,
"tokens_seen": 1666973696
},
{
"epoch": 0.51,
"learning_rate": 0.0002499899699097292,
"loss": 2.7883,
"theoretical_loss": 3.4838276465058287,
"tokens_seen": 1667039232
},
{
"epoch": 0.51,
"learning_rate": 0.00024997993981945837,
"loss": 2.4586,
"theoretical_loss": 3.4838158144260927,
"tokens_seen": 1667104768
},
{
"epoch": 0.51,
"learning_rate": 0.00024996990972918755,
"loss": 2.4692,
"theoretical_loss": 3.483803982941714,
"tokens_seen": 1667170304
},
{
"epoch": 0.51,
"learning_rate": 0.00024995987963891673,
"loss": 2.7052,
"theoretical_loss": 3.4837921520526374,
"tokens_seen": 1667235840
},
{
"epoch": 0.51,
"learning_rate": 0.00024994984954864597,
"loss": 2.7284,
"theoretical_loss": 3.483780321758811,
"tokens_seen": 1667301376
},
{
"epoch": 0.51,
"learning_rate": 0.00024993981945837515,
"loss": 2.5605,
"theoretical_loss": 3.4837684920601815,
"tokens_seen": 1667366912
},
{
"epoch": 0.51,
"learning_rate": 0.00024992978936810433,
"loss": 2.7009,
"theoretical_loss": 3.483756662956695,
"tokens_seen": 1667432448
},
{
"epoch": 0.51,
"learning_rate": 0.0002499197592778335,
"loss": 2.7693,
"theoretical_loss": 3.483744834448298,
"tokens_seen": 1667497984
},
{
"epoch": 0.51,
"learning_rate": 0.0002499097291875627,
"loss": 2.8237,
"theoretical_loss": 3.483733006534938,
"tokens_seen": 1667563520
},
{
"epoch": 0.51,
"learning_rate": 0.00024989969909729187,
"loss": 2.6209,
"theoretical_loss": 3.483721179216561,
"tokens_seen": 1667629056
},
{
"epoch": 0.51,
"learning_rate": 0.00024988966900702105,
"loss": 2.7973,
"theoretical_loss": 3.4837093524931144,
"tokens_seen": 1667694592
},
{
"epoch": 0.51,
"learning_rate": 0.00024987963891675023,
"loss": 2.6427,
"theoretical_loss": 3.4836975263645438,
"tokens_seen": 1667760128
},
{
"epoch": 0.51,
"learning_rate": 0.0002498696088264794,
"loss": 2.4011,
"theoretical_loss": 3.483685700830797,
"tokens_seen": 1667825664
},
{
"epoch": 0.51,
"objective/train/docs_used": 945126,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7451326847076416,
"objective/train/theoretical_loss": 3.48367387589182,
"objective/train/tokens_used": 1688351200,
"theoretical_loss": 3.48367387589182,
"tokens_seen": 1667891200
},
{
"epoch": 0.51,
"learning_rate": 0.00024985957873620865,
"loss": 2.7824,
"theoretical_loss": 3.48367387589182,
"tokens_seen": 1667891200
},
{
"epoch": 0.51,
"learning_rate": 0.00024984954864593783,
"loss": 2.6761,
"theoretical_loss": 3.48366205154756,
"tokens_seen": 1667956736
},
{
"epoch": 0.51,
"learning_rate": 0.000249839518555667,
"loss": 2.6208,
"theoretical_loss": 3.4836502277979626,
"tokens_seen": 1668022272
},
{
"epoch": 0.51,
"learning_rate": 0.0002498294884653962,
"loss": 2.5316,
"theoretical_loss": 3.4836384046429765,
"tokens_seen": 1668087808
},
{
"epoch": 0.51,
"learning_rate": 0.0002498194583751254,
"loss": 2.4722,
"theoretical_loss": 3.4836265820825467,
"tokens_seen": 1668153344
},
{
"epoch": 0.51,
"learning_rate": 0.00024980942828485455,
"loss": 2.6576,
"theoretical_loss": 3.483614760116621,
"tokens_seen": 1668218880
},
{
"epoch": 0.51,
"learning_rate": 0.00024979939819458374,
"loss": 2.6727,
"theoretical_loss": 3.4836029387451455,
"tokens_seen": 1668284416
},
{
"epoch": 0.51,
"learning_rate": 0.0002497893681043129,
"loss": 2.6886,
"theoretical_loss": 3.483591117968067,
"tokens_seen": 1668349952
},
{
"epoch": 0.51,
"learning_rate": 0.0002497793380140421,
"loss": 2.7816,
"theoretical_loss": 3.483579297785333,
"tokens_seen": 1668415488
},
{
"epoch": 0.51,
"learning_rate": 0.00024976930792377133,
"loss": 2.3929,
"theoretical_loss": 3.483567478196889,
"tokens_seen": 1668481024
},
{
"epoch": 0.51,
"learning_rate": 0.0002497592778335005,
"loss": 2.6738,
"theoretical_loss": 3.483555659202683,
"tokens_seen": 1668546560
},
{
"epoch": 0.51,
"learning_rate": 0.0002497492477432297,
"loss": 2.5238,
"theoretical_loss": 3.4835438408026613,
"tokens_seen": 1668612096
},
{
"epoch": 0.51,
"learning_rate": 0.0002497392176529589,
"loss": 2.8027,
"theoretical_loss": 3.483532022996771,
"tokens_seen": 1668677632
},
{
"epoch": 0.51,
"learning_rate": 0.00024972918756268806,
"loss": 2.6289,
"theoretical_loss": 3.483520205784958,
"tokens_seen": 1668743168
},
{
"epoch": 0.51,
"learning_rate": 0.00024971915747241724,
"loss": 2.5418,
"theoretical_loss": 3.48350838916717,
"tokens_seen": 1668808704
},
{
"epoch": 0.51,
"learning_rate": 0.0002497091273821464,
"loss": 2.7873,
"theoretical_loss": 3.4834965731433534,
"tokens_seen": 1668874240
},
{
"epoch": 0.51,
"learning_rate": 0.0002496990972918756,
"loss": 2.7071,
"theoretical_loss": 3.4834847577134553,
"tokens_seen": 1668939776
},
{
"epoch": 0.51,
"learning_rate": 0.0002496890672016048,
"loss": 2.3717,
"theoretical_loss": 3.4834729428774223,
"tokens_seen": 1669005312
},
{
"epoch": 0.51,
"learning_rate": 0.000249679037111334,
"loss": 2.5794,
"theoretical_loss": 3.4834611286352013,
"tokens_seen": 1669070848
},
{
"epoch": 0.51,
"learning_rate": 0.0002496690070210632,
"loss": 2.5351,
"theoretical_loss": 3.4834493149867387,
"tokens_seen": 1669136384
},
{
"epoch": 0.51,
"learning_rate": 0.0002496589769307924,
"loss": 2.8416,
"theoretical_loss": 3.483437501931982,
"tokens_seen": 1669201920
},
{
"epoch": 0.51,
"learning_rate": 0.00024964894684052156,
"loss": 2.6162,
"theoretical_loss": 3.483425689470878,
"tokens_seen": 1669267456
},
{
"epoch": 0.51,
"learning_rate": 0.00024963891675025074,
"loss": 2.4763,
"theoretical_loss": 3.4834138776033736,
"tokens_seen": 1669332992
},
{
"epoch": 0.51,
"learning_rate": 0.0002496288866599799,
"loss": 2.7136,
"theoretical_loss": 3.483402066329415,
"tokens_seen": 1669398528
},
{
"epoch": 0.51,
"learning_rate": 0.00024961885656970916,
"loss": 2.5604,
"theoretical_loss": 3.48339025564895,
"tokens_seen": 1669464064
},
{
"epoch": 0.51,
"objective/train/docs_used": 945699,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4707672595977783,
"objective/train/theoretical_loss": 3.4833784455619248,
"objective/train/tokens_used": 1689989600,
"theoretical_loss": 3.4833784455619248,
"tokens_seen": 1669529600
},
{
"epoch": 0.51,
"learning_rate": 0.00024960882647943834,
"loss": 2.7026,
"theoretical_loss": 3.4833784455619248,
"tokens_seen": 1669529600
},
{
"epoch": 0.51,
"learning_rate": 0.0002495987963891675,
"loss": 2.574,
"theoretical_loss": 3.4833666360682867,
"tokens_seen": 1669595136
},
{
"epoch": 0.51,
"learning_rate": 0.0002495887662988967,
"loss": 2.5242,
"theoretical_loss": 3.4833548271679824,
"tokens_seen": 1669660672
},
{
"epoch": 0.51,
"learning_rate": 0.0002495787362086259,
"loss": 2.5236,
"theoretical_loss": 3.4833430188609587,
"tokens_seen": 1669726208
},
{
"epoch": 0.51,
"learning_rate": 0.00024956870611835506,
"loss": 2.9492,
"theoretical_loss": 3.4833312111471626,
"tokens_seen": 1669791744
},
{
"epoch": 0.51,
"learning_rate": 0.00024955867602808425,
"loss": 2.7523,
"theoretical_loss": 3.483319404026541,
"tokens_seen": 1669857280
},
{
"epoch": 0.51,
"learning_rate": 0.0002495486459378135,
"loss": 2.7197,
"theoretical_loss": 3.4833075974990404,
"tokens_seen": 1669922816
},
{
"epoch": 0.51,
"learning_rate": 0.00024953861584754266,
"loss": 2.6447,
"theoretical_loss": 3.4832957915646086,
"tokens_seen": 1669988352
},
{
"epoch": 0.51,
"learning_rate": 0.00024952858575727184,
"loss": 2.7091,
"theoretical_loss": 3.4832839862231926,
"tokens_seen": 1670053888
},
{
"epoch": 0.51,
"learning_rate": 0.000249518555667001,
"loss": 2.5728,
"theoretical_loss": 3.4832721814747383,
"tokens_seen": 1670119424
},
{
"epoch": 0.51,
"learning_rate": 0.0002495085255767302,
"loss": 2.6807,
"theoretical_loss": 3.4832603773191932,
"tokens_seen": 1670184960
},
{
"epoch": 0.51,
"learning_rate": 0.0002494984954864594,
"loss": 2.7171,
"theoretical_loss": 3.4832485737565047,
"tokens_seen": 1670250496
},
{
"epoch": 0.51,
"learning_rate": 0.00024948846539618857,
"loss": 2.7989,
"theoretical_loss": 3.483236770786619,
"tokens_seen": 1670316032
},
{
"epoch": 0.51,
"learning_rate": 0.00024947843530591775,
"loss": 2.8343,
"theoretical_loss": 3.4832249684094836,
"tokens_seen": 1670381568
},
{
"epoch": 0.51,
"learning_rate": 0.00024946840521564693,
"loss": 2.6609,
"theoretical_loss": 3.4832131666250454,
"tokens_seen": 1670447104
},
{
"epoch": 0.51,
"learning_rate": 0.00024945837512537617,
"loss": 2.6178,
"theoretical_loss": 3.483201365433251,
"tokens_seen": 1670512640
},
{
"epoch": 0.51,
"learning_rate": 0.00024944834503510535,
"loss": 2.5902,
"theoretical_loss": 3.4831895648340474,
"tokens_seen": 1670578176
},
{
"epoch": 0.51,
"learning_rate": 0.00024943831494483453,
"loss": 2.5369,
"theoretical_loss": 3.4831777648273823,
"tokens_seen": 1670643712
},
{
"epoch": 0.51,
"learning_rate": 0.0002494282848545637,
"loss": 2.8344,
"theoretical_loss": 3.4831659654132023,
"tokens_seen": 1670709248
},
{
"epoch": 0.51,
"learning_rate": 0.0002494182547642929,
"loss": 2.5683,
"theoretical_loss": 3.4831541665914543,
"tokens_seen": 1670774784
},
{
"epoch": 0.51,
"learning_rate": 0.00024940822467402207,
"loss": 2.569,
"theoretical_loss": 3.4831423683620852,
"tokens_seen": 1670840320
},
{
"epoch": 0.51,
"learning_rate": 0.00024939819458375125,
"loss": 2.5592,
"theoretical_loss": 3.4831305707250424,
"tokens_seen": 1670905856
},
{
"epoch": 0.51,
"learning_rate": 0.00024938816449348043,
"loss": 2.6314,
"theoretical_loss": 3.483118773680273,
"tokens_seen": 1670971392
},
{
"epoch": 0.51,
"learning_rate": 0.0002493781344032096,
"loss": 2.9109,
"theoretical_loss": 3.4831069772277234,
"tokens_seen": 1671036928
},
{
"epoch": 0.51,
"learning_rate": 0.00024936810431293885,
"loss": 2.6118,
"theoretical_loss": 3.483095181367341,
"tokens_seen": 1671102464
},
{
"epoch": 0.51,
"objective/train/docs_used": 946326,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.82181715965271,
"objective/train/theoretical_loss": 3.4830833860990733,
"objective/train/tokens_used": 1691628000,
"theoretical_loss": 3.4830833860990733,
"tokens_seen": 1671168000
},
{
"epoch": 0.51,
"learning_rate": 0.00024935807422266803,
"loss": 2.7126,
"theoretical_loss": 3.4830833860990733,
"tokens_seen": 1671168000
},
{
"epoch": 0.51,
"learning_rate": 0.0002493480441323972,
"loss": 2.7434,
"theoretical_loss": 3.4830715914228665,
"tokens_seen": 1671233536
},
{
"epoch": 0.51,
"learning_rate": 0.0002493380140421264,
"loss": 2.9203,
"theoretical_loss": 3.4830597973386683,
"tokens_seen": 1671299072
},
{
"epoch": 0.51,
"learning_rate": 0.0002493279839518556,
"loss": 2.4418,
"theoretical_loss": 3.4830480038464255,
"tokens_seen": 1671364608
},
{
"epoch": 0.51,
"learning_rate": 0.00024931795386158476,
"loss": 2.8067,
"theoretical_loss": 3.4830362109460857,
"tokens_seen": 1671430144
},
{
"epoch": 0.51,
"learning_rate": 0.00024930792377131394,
"loss": 2.9885,
"theoretical_loss": 3.483024418637595,
"tokens_seen": 1671495680
},
{
"epoch": 0.51,
"learning_rate": 0.0002492978936810431,
"loss": 2.8661,
"theoretical_loss": 3.4830126269209014,
"tokens_seen": 1671561216
},
{
"epoch": 0.51,
"learning_rate": 0.0002492878635907723,
"loss": 2.8255,
"theoretical_loss": 3.4830008357959517,
"tokens_seen": 1671626752
},
{
"epoch": 0.51,
"learning_rate": 0.00024927783350050153,
"loss": 2.5592,
"theoretical_loss": 3.4829890452626926,
"tokens_seen": 1671692288
},
{
"epoch": 0.51,
"learning_rate": 0.0002492678034102307,
"loss": 2.6909,
"theoretical_loss": 3.482977255321072,
"tokens_seen": 1671757824
},
{
"epoch": 0.51,
"learning_rate": 0.0002492577733199599,
"loss": 2.5329,
"theoretical_loss": 3.482965465971036,
"tokens_seen": 1671823360
},
{
"epoch": 0.51,
"learning_rate": 0.0002492477432296891,
"loss": 2.5824,
"theoretical_loss": 3.4829536772125325,
"tokens_seen": 1671888896
},
{
"epoch": 0.51,
"learning_rate": 0.00024923771313941826,
"loss": 2.6348,
"theoretical_loss": 3.482941889045509,
"tokens_seen": 1671954432
},
{
"epoch": 0.51,
"learning_rate": 0.00024922768304914744,
"loss": 2.7622,
"theoretical_loss": 3.4829301014699117,
"tokens_seen": 1672019968
},
{
"epoch": 0.51,
"learning_rate": 0.0002492176529588766,
"loss": 2.6095,
"theoretical_loss": 3.482918314485688,
"tokens_seen": 1672085504
},
{
"epoch": 0.51,
"learning_rate": 0.0002492076228686058,
"loss": 2.6158,
"theoretical_loss": 3.4829065280927853,
"tokens_seen": 1672151040
},
{
"epoch": 0.51,
"learning_rate": 0.000249197592778335,
"loss": 2.8073,
"theoretical_loss": 3.482894742291151,
"tokens_seen": 1672216576
},
{
"epoch": 0.51,
"learning_rate": 0.0002491875626880642,
"loss": 2.7055,
"theoretical_loss": 3.4828829570807316,
"tokens_seen": 1672282112
},
{
"epoch": 0.51,
"learning_rate": 0.0002491775325977934,
"loss": 2.5951,
"theoretical_loss": 3.4828711724614747,
"tokens_seen": 1672347648
},
{
"epoch": 0.51,
"learning_rate": 0.0002491675025075226,
"loss": 2.6642,
"theoretical_loss": 3.4828593884333277,
"tokens_seen": 1672413184
},
{
"epoch": 0.51,
"learning_rate": 0.00024915747241725176,
"loss": 2.6764,
"theoretical_loss": 3.4828476049962367,
"tokens_seen": 1672478720
},
{
"epoch": 0.51,
"learning_rate": 0.00024914744232698094,
"loss": 2.6143,
"theoretical_loss": 3.4828358221501503,
"tokens_seen": 1672544256
},
{
"epoch": 0.51,
"learning_rate": 0.0002491374122367101,
"loss": 2.6202,
"theoretical_loss": 3.4828240398950143,
"tokens_seen": 1672609792
},
{
"epoch": 0.51,
"learning_rate": 0.0002491273821464393,
"loss": 2.7614,
"theoretical_loss": 3.4828122582307777,
"tokens_seen": 1672675328
},
{
"epoch": 0.51,
"learning_rate": 0.0002491173520561685,
"loss": 2.4366,
"theoretical_loss": 3.482800477157386,
"tokens_seen": 1672740864
},
{
"epoch": 0.51,
"objective/train/docs_used": 947668,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3934407234191895,
"objective/train/theoretical_loss": 3.4827886966747874,
"objective/train/tokens_used": 1693266400,
"theoretical_loss": 3.4827886966747874,
"tokens_seen": 1672806400
},
{
"epoch": 0.51,
"learning_rate": 0.00024910732196589767,
"loss": 2.6807,
"theoretical_loss": 3.4827886966747874,
"tokens_seen": 1672806400
},
{
"epoch": 0.51,
"learning_rate": 0.0002490972918756269,
"loss": 2.6159,
"theoretical_loss": 3.482776916782929,
"tokens_seen": 1672871936
},
{
"epoch": 0.51,
"learning_rate": 0.0002490872617853561,
"loss": 2.4668,
"theoretical_loss": 3.482765137481757,
"tokens_seen": 1672937472
},
{
"epoch": 0.51,
"learning_rate": 0.00024907723169508526,
"loss": 2.5128,
"theoretical_loss": 3.4827533587712205,
"tokens_seen": 1673003008
},
{
"epoch": 0.51,
"learning_rate": 0.00024906720160481445,
"loss": 2.7598,
"theoretical_loss": 3.4827415806512656,
"tokens_seen": 1673068544
},
{
"epoch": 0.51,
"learning_rate": 0.0002490571715145436,
"loss": 2.7929,
"theoretical_loss": 3.4827298031218397,
"tokens_seen": 1673134080
},
{
"epoch": 0.51,
"learning_rate": 0.0002490471414242728,
"loss": 2.692,
"theoretical_loss": 3.48271802618289,
"tokens_seen": 1673199616
},
{
"epoch": 0.51,
"learning_rate": 0.000249037111334002,
"loss": 2.5748,
"theoretical_loss": 3.4827062498343637,
"tokens_seen": 1673265152
},
{
"epoch": 0.51,
"learning_rate": 0.00024902708124373117,
"loss": 2.7621,
"theoretical_loss": 3.4826944740762085,
"tokens_seen": 1673330688
},
{
"epoch": 0.51,
"learning_rate": 0.00024901705115346035,
"loss": 2.8067,
"theoretical_loss": 3.4826826989083712,
"tokens_seen": 1673396224
},
{
"epoch": 0.51,
"learning_rate": 0.0002490070210631896,
"loss": 2.7199,
"theoretical_loss": 3.4826709243307996,
"tokens_seen": 1673461760
},
{
"epoch": 0.51,
"learning_rate": 0.00024899699097291877,
"loss": 2.669,
"theoretical_loss": 3.4826591503434408,
"tokens_seen": 1673527296
},
{
"epoch": 0.51,
"learning_rate": 0.00024898696088264795,
"loss": 2.4181,
"theoretical_loss": 3.4826473769462414,
"tokens_seen": 1673592832
},
{
"epoch": 0.51,
"learning_rate": 0.00024897693079237713,
"loss": 2.6847,
"theoretical_loss": 3.4826356041391495,
"tokens_seen": 1673658368
},
{
"epoch": 0.51,
"learning_rate": 0.0002489669007021063,
"loss": 2.6478,
"theoretical_loss": 3.482623831922113,
"tokens_seen": 1673723904
},
{
"epoch": 0.51,
"learning_rate": 0.0002489568706118355,
"loss": 2.5446,
"theoretical_loss": 3.4826120602950774,
"tokens_seen": 1673789440
},
{
"epoch": 0.51,
"learning_rate": 0.0002489468405215647,
"loss": 2.6706,
"theoretical_loss": 3.4826002892579915,
"tokens_seen": 1673854976
},
{
"epoch": 0.51,
"learning_rate": 0.00024893681043129385,
"loss": 2.6023,
"theoretical_loss": 3.482588518810802,
"tokens_seen": 1673920512
},
{
"epoch": 0.51,
"learning_rate": 0.00024892678034102304,
"loss": 2.5673,
"theoretical_loss": 3.4825767489534574,
"tokens_seen": 1673986048
},
{
"epoch": 0.51,
"learning_rate": 0.00024891675025075227,
"loss": 2.7729,
"theoretical_loss": 3.482564979685903,
"tokens_seen": 1674051584
},
{
"epoch": 0.51,
"learning_rate": 0.00024890672016048145,
"loss": 2.3643,
"theoretical_loss": 3.482553211008088,
"tokens_seen": 1674117120
},
{
"epoch": 0.51,
"learning_rate": 0.00024889669007021063,
"loss": 2.5641,
"theoretical_loss": 3.482541442919959,
"tokens_seen": 1674182656
},
{
"epoch": 0.51,
"learning_rate": 0.0002488866599799398,
"loss": 2.6088,
"theoretical_loss": 3.482529675421463,
"tokens_seen": 1674248192
},
{
"epoch": 0.51,
"learning_rate": 0.000248876629889669,
"loss": 2.7589,
"theoretical_loss": 3.4825179085125484,
"tokens_seen": 1674313728
},
{
"epoch": 0.51,
"learning_rate": 0.00024886659979939823,
"loss": 2.7816,
"theoretical_loss": 3.4825061421931616,
"tokens_seen": 1674379264
},
{
"epoch": 0.51,
"objective/train/docs_used": 948408,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7572250366210938,
"objective/train/theoretical_loss": 3.4824943764632503,
"objective/train/tokens_used": 1694904800,
"theoretical_loss": 3.4824943764632503,
"tokens_seen": 1674444800
},
{
"epoch": 0.51,
"learning_rate": 0.0002488565697091274,
"loss": 2.4921,
"theoretical_loss": 3.4824943764632503,
"tokens_seen": 1674444800
},
{
"epoch": 0.51,
"learning_rate": 0.0002488465396188566,
"loss": 2.5524,
"theoretical_loss": 3.4824826113227623,
"tokens_seen": 1674510336
},
{
"epoch": 0.51,
"learning_rate": 0.0002488365095285858,
"loss": 2.5335,
"theoretical_loss": 3.4824708467716445,
"tokens_seen": 1674575872
},
{
"epoch": 0.51,
"learning_rate": 0.00024882647943831496,
"loss": 2.3231,
"theoretical_loss": 3.4824590828098447,
"tokens_seen": 1674641408
},
{
"epoch": 0.51,
"learning_rate": 0.00024881644934804414,
"loss": 2.759,
"theoretical_loss": 3.48244731943731,
"tokens_seen": 1674706944
},
{
"epoch": 0.51,
"learning_rate": 0.0002488064192577733,
"loss": 2.5952,
"theoretical_loss": 3.4824355566539875,
"tokens_seen": 1674772480
},
{
"epoch": 0.51,
"learning_rate": 0.0002487963891675025,
"loss": 2.9085,
"theoretical_loss": 3.4824237944598257,
"tokens_seen": 1674838016
},
{
"epoch": 0.51,
"learning_rate": 0.00024878635907723173,
"loss": 2.413,
"theoretical_loss": 3.482412032854771,
"tokens_seen": 1674903552
},
{
"epoch": 0.51,
"learning_rate": 0.0002487763289869609,
"loss": 2.7455,
"theoretical_loss": 3.482400271838771,
"tokens_seen": 1674969088
},
{
"epoch": 0.51,
"learning_rate": 0.0002487662988966901,
"loss": 2.4626,
"theoretical_loss": 3.482388511411774,
"tokens_seen": 1675034624
},
{
"epoch": 0.51,
"learning_rate": 0.0002487562688064193,
"loss": 2.4601,
"theoretical_loss": 3.4823767515737267,
"tokens_seen": 1675100160
},
{
"epoch": 0.51,
"learning_rate": 0.00024874623871614846,
"loss": 2.503,
"theoretical_loss": 3.4823649923245767,
"tokens_seen": 1675165696
},
{
"epoch": 0.51,
"learning_rate": 0.00024873620862587764,
"loss": 2.7555,
"theoretical_loss": 3.4823532336642717,
"tokens_seen": 1675231232
},
{
"epoch": 0.51,
"learning_rate": 0.0002487261785356068,
"loss": 2.5664,
"theoretical_loss": 3.482341475592759,
"tokens_seen": 1675296768
},
{
"epoch": 0.51,
"learning_rate": 0.000248716148445336,
"loss": 2.8844,
"theoretical_loss": 3.4823297181099857,
"tokens_seen": 1675362304
},
{
"epoch": 0.51,
"learning_rate": 0.0002487061183550652,
"loss": 2.597,
"theoretical_loss": 3.4823179612159,
"tokens_seen": 1675427840
},
{
"epoch": 0.51,
"learning_rate": 0.0002486960882647944,
"loss": 2.8291,
"theoretical_loss": 3.4823062049104485,
"tokens_seen": 1675493376
},
{
"epoch": 0.51,
"learning_rate": 0.0002486860581745236,
"loss": 2.4586,
"theoretical_loss": 3.4822944491935797,
"tokens_seen": 1675558912
},
{
"epoch": 0.51,
"learning_rate": 0.0002486760280842528,
"loss": 2.627,
"theoretical_loss": 3.4822826940652405,
"tokens_seen": 1675624448
},
{
"epoch": 0.51,
"learning_rate": 0.00024866599799398196,
"loss": 2.572,
"theoretical_loss": 3.482270939525379,
"tokens_seen": 1675689984
},
{
"epoch": 0.51,
"learning_rate": 0.00024865596790371114,
"loss": 2.5162,
"theoretical_loss": 3.482259185573942,
"tokens_seen": 1675755520
},
{
"epoch": 0.51,
"learning_rate": 0.0002486459378134403,
"loss": 2.6537,
"theoretical_loss": 3.4822474322108774,
"tokens_seen": 1675821056
},
{
"epoch": 0.51,
"learning_rate": 0.0002486359077231695,
"loss": 2.8905,
"theoretical_loss": 3.4822356794361324,
"tokens_seen": 1675886592
},
{
"epoch": 0.51,
"learning_rate": 0.0002486258776328987,
"loss": 2.663,
"theoretical_loss": 3.482223927249655,
"tokens_seen": 1675952128
},
{
"epoch": 0.51,
"learning_rate": 0.00024861584754262787,
"loss": 2.8771,
"theoretical_loss": 3.482212175651393,
"tokens_seen": 1676017664
},
{
"epoch": 0.51,
"objective/train/docs_used": 949481,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5695033073425293,
"objective/train/theoretical_loss": 3.4822004246412934,
"objective/train/tokens_used": 1696543200,
"theoretical_loss": 3.4822004246412934,
"tokens_seen": 1676083200
},
{
"epoch": 0.51,
"learning_rate": 0.0002486058174523571,
"loss": 2.6051,
"theoretical_loss": 3.4822004246412934,
"tokens_seen": 1676083200
},
{
"epoch": 0.51,
"learning_rate": 0.0002485957873620863,
"loss": 2.6324,
"theoretical_loss": 3.4821886742193042,
"tokens_seen": 1676148736
},
{
"epoch": 0.51,
"learning_rate": 0.00024858575727181546,
"loss": 2.5716,
"theoretical_loss": 3.4821769243853717,
"tokens_seen": 1676214272
},
{
"epoch": 0.51,
"learning_rate": 0.00024857572718154465,
"loss": 2.5833,
"theoretical_loss": 3.4821651751394453,
"tokens_seen": 1676279808
},
{
"epoch": 0.51,
"learning_rate": 0.0002485656970912738,
"loss": 2.589,
"theoretical_loss": 3.4821534264814717,
"tokens_seen": 1676345344
},
{
"epoch": 0.51,
"learning_rate": 0.000248555667001003,
"loss": 2.5742,
"theoretical_loss": 3.4821416784113985,
"tokens_seen": 1676410880
},
{
"epoch": 0.51,
"learning_rate": 0.0002485456369107322,
"loss": 2.7164,
"theoretical_loss": 3.4821299309291733,
"tokens_seen": 1676476416
},
{
"epoch": 0.51,
"learning_rate": 0.00024853560682046137,
"loss": 2.6401,
"theoretical_loss": 3.4821181840347437,
"tokens_seen": 1676541952
},
{
"epoch": 0.51,
"learning_rate": 0.00024852557673019055,
"loss": 2.3964,
"theoretical_loss": 3.4821064377280577,
"tokens_seen": 1676607488
},
{
"epoch": 0.51,
"learning_rate": 0.0002485155466399198,
"loss": 2.8223,
"theoretical_loss": 3.4820946920090625,
"tokens_seen": 1676673024
},
{
"epoch": 0.51,
"learning_rate": 0.00024850551654964897,
"loss": 2.6644,
"theoretical_loss": 3.4820829468777057,
"tokens_seen": 1676738560
},
{
"epoch": 0.51,
"learning_rate": 0.00024849548645937815,
"loss": 2.5649,
"theoretical_loss": 3.482071202333935,
"tokens_seen": 1676804096
},
{
"epoch": 0.51,
"learning_rate": 0.00024848545636910733,
"loss": 2.6795,
"theoretical_loss": 3.482059458377698,
"tokens_seen": 1676869632
},
{
"epoch": 0.51,
"learning_rate": 0.0002484754262788365,
"loss": 2.778,
"theoretical_loss": 3.4820477150089433,
"tokens_seen": 1676935168
},
{
"epoch": 0.51,
"learning_rate": 0.0002484653961885657,
"loss": 2.5942,
"theoretical_loss": 3.4820359722276164,
"tokens_seen": 1677000704
},
{
"epoch": 0.51,
"learning_rate": 0.0002484553660982949,
"loss": 2.3994,
"theoretical_loss": 3.4820242300336677,
"tokens_seen": 1677066240
},
{
"epoch": 0.51,
"learning_rate": 0.00024844533600802405,
"loss": 2.8308,
"theoretical_loss": 3.4820124884270425,
"tokens_seen": 1677131776
},
{
"epoch": 0.51,
"learning_rate": 0.00024843530591775324,
"loss": 2.5242,
"theoretical_loss": 3.4820007474076893,
"tokens_seen": 1677197312
},
{
"epoch": 0.51,
"learning_rate": 0.00024842527582748247,
"loss": 2.5995,
"theoretical_loss": 3.481989006975556,
"tokens_seen": 1677262848
},
{
"epoch": 0.51,
"learning_rate": 0.00024841524573721165,
"loss": 2.6422,
"theoretical_loss": 3.4819772671305906,
"tokens_seen": 1677328384
},
{
"epoch": 0.51,
"learning_rate": 0.00024840521564694083,
"loss": 2.6618,
"theoretical_loss": 3.48196552787274,
"tokens_seen": 1677393920
},
{
"epoch": 0.51,
"learning_rate": 0.00024839518555667,
"loss": 2.7352,
"theoretical_loss": 3.4819537892019525,
"tokens_seen": 1677459456
},
{
"epoch": 0.51,
"learning_rate": 0.0002483851554663992,
"loss": 2.7769,
"theoretical_loss": 3.4819420511181756,
"tokens_seen": 1677524992
},
{
"epoch": 0.51,
"learning_rate": 0.0002483751253761284,
"loss": 2.5896,
"theoretical_loss": 3.481930313621357,
"tokens_seen": 1677590528
},
{
"epoch": 0.51,
"learning_rate": 0.00024836509528585756,
"loss": 2.7067,
"theoretical_loss": 3.4819185767114442,
"tokens_seen": 1677656064
},
{
"epoch": 0.51,
"objective/train/docs_used": 950694,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9692914485931396,
"objective/train/theoretical_loss": 3.4819068403883855,
"objective/train/tokens_used": 1698181600,
"theoretical_loss": 3.4819068403883855,
"tokens_seen": 1677721600
},
{
"epoch": 0.51,
"learning_rate": 0.00024835506519558674,
"loss": 2.7975,
"theoretical_loss": 3.4819068403883855,
"tokens_seen": 1677721600
},
{
"epoch": 0.51,
"learning_rate": 0.000248345035105316,
"loss": 2.5935,
"theoretical_loss": 3.4818951046521276,
"tokens_seen": 1677787136
},
{
"epoch": 0.51,
"learning_rate": 0.00024833500501504516,
"loss": 2.7251,
"theoretical_loss": 3.4818833695026195,
"tokens_seen": 1677852672
},
{
"epoch": 0.51,
"learning_rate": 0.00024832497492477434,
"loss": 2.5254,
"theoretical_loss": 3.4818716349398082,
"tokens_seen": 1677918208
},
{
"epoch": 0.51,
"learning_rate": 0.0002483149448345035,
"loss": 2.5597,
"theoretical_loss": 3.481859900963642,
"tokens_seen": 1677983744
},
{
"epoch": 0.51,
"learning_rate": 0.0002483049147442327,
"loss": 2.6391,
"theoretical_loss": 3.481848167574068,
"tokens_seen": 1678049280
},
{
"epoch": 0.51,
"learning_rate": 0.0002482948846539619,
"loss": 2.7561,
"theoretical_loss": 3.481836434771034,
"tokens_seen": 1678114816
},
{
"epoch": 0.51,
"learning_rate": 0.00024828485456369106,
"loss": 2.5742,
"theoretical_loss": 3.4818247025544884,
"tokens_seen": 1678180352
},
{
"epoch": 0.51,
"learning_rate": 0.00024827482447342024,
"loss": 2.6453,
"theoretical_loss": 3.481812970924379,
"tokens_seen": 1678245888
},
{
"epoch": 0.51,
"learning_rate": 0.0002482647943831494,
"loss": 2.6419,
"theoretical_loss": 3.481801239880652,
"tokens_seen": 1678311424
},
{
"epoch": 0.51,
"learning_rate": 0.00024825476429287866,
"loss": 2.536,
"theoretical_loss": 3.4817895094232574,
"tokens_seen": 1678376960
},
{
"epoch": 0.51,
"learning_rate": 0.00024824473420260784,
"loss": 2.6823,
"theoretical_loss": 3.481777779552141,
"tokens_seen": 1678442496
},
{
"epoch": 0.51,
"learning_rate": 0.000248234704112337,
"loss": 2.5245,
"theoretical_loss": 3.481766050267253,
"tokens_seen": 1678508032
},
{
"epoch": 0.51,
"learning_rate": 0.0002482246740220662,
"loss": 2.5122,
"theoretical_loss": 3.481754321568539,
"tokens_seen": 1678573568
},
{
"epoch": 0.51,
"learning_rate": 0.0002482146439317954,
"loss": 2.6883,
"theoretical_loss": 3.4817425934559476,
"tokens_seen": 1678639104
},
{
"epoch": 0.51,
"learning_rate": 0.00024820461384152456,
"loss": 2.4606,
"theoretical_loss": 3.4817308659294266,
"tokens_seen": 1678704640
},
{
"epoch": 0.51,
"learning_rate": 0.00024819458375125375,
"loss": 2.5522,
"theoretical_loss": 3.4817191389889244,
"tokens_seen": 1678770176
},
{
"epoch": 0.51,
"learning_rate": 0.0002481845536609829,
"loss": 2.6324,
"theoretical_loss": 3.481707412634388,
"tokens_seen": 1678835712
},
{
"epoch": 0.51,
"learning_rate": 0.0002481745235707121,
"loss": 2.6005,
"theoretical_loss": 3.481695686865766,
"tokens_seen": 1678901248
},
{
"epoch": 0.51,
"learning_rate": 0.00024816449348044134,
"loss": 2.6226,
"theoretical_loss": 3.481683961683005,
"tokens_seen": 1678966784
},
{
"epoch": 0.51,
"learning_rate": 0.0002481544633901705,
"loss": 2.6933,
"theoretical_loss": 3.4816722370860544,
"tokens_seen": 1679032320
},
{
"epoch": 0.51,
"learning_rate": 0.0002481444332998997,
"loss": 2.5163,
"theoretical_loss": 3.4816605130748615,
"tokens_seen": 1679097856
},
{
"epoch": 0.51,
"learning_rate": 0.0002481344032096289,
"loss": 2.7915,
"theoretical_loss": 3.4816487896493733,
"tokens_seen": 1679163392
},
{
"epoch": 0.51,
"learning_rate": 0.00024812437311935807,
"loss": 2.7545,
"theoretical_loss": 3.4816370668095393,
"tokens_seen": 1679228928
},
{
"epoch": 0.51,
"learning_rate": 0.0002481143430290873,
"loss": 2.5137,
"theoretical_loss": 3.4816253445553063,
"tokens_seen": 1679294464
},
{
"epoch": 0.51,
"objective/train/docs_used": 951339,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4856183528900146,
"objective/train/theoretical_loss": 3.4816136228866217,
"objective/train/tokens_used": 1699820000,
"theoretical_loss": 3.4816136228866217,
"tokens_seen": 1679360000
},
{
"epoch": 0.51,
"learning_rate": 0.0002481043129388165,
"loss": 2.7202,
"theoretical_loss": 3.4816136228866217,
"tokens_seen": 1679360000
},
{
"epoch": 0.51,
"learning_rate": 0.00024809428284854566,
"loss": 2.6006,
"theoretical_loss": 3.481601901803435,
"tokens_seen": 1679425536
},
{
"epoch": 0.51,
"learning_rate": 0.00024808425275827485,
"loss": 2.5976,
"theoretical_loss": 3.481590181305693,
"tokens_seen": 1679491072
},
{
"epoch": 0.51,
"learning_rate": 0.00024807422266800403,
"loss": 2.6056,
"theoretical_loss": 3.4815784613933434,
"tokens_seen": 1679556608
},
{
"epoch": 0.51,
"learning_rate": 0.0002480641925777332,
"loss": 2.7389,
"theoretical_loss": 3.481566742066335,
"tokens_seen": 1679622144
},
{
"epoch": 0.51,
"learning_rate": 0.0002480541624874624,
"loss": 2.6726,
"theoretical_loss": 3.4815550233246153,
"tokens_seen": 1679687680
},
{
"epoch": 0.51,
"learning_rate": 0.00024804413239719157,
"loss": 2.6441,
"theoretical_loss": 3.481543305168132,
"tokens_seen": 1679753216
},
{
"epoch": 0.51,
"learning_rate": 0.00024803410230692075,
"loss": 2.6944,
"theoretical_loss": 3.4815315875968333,
"tokens_seen": 1679818752
},
{
"epoch": 0.51,
"learning_rate": 0.00024802407221665,
"loss": 2.9233,
"theoretical_loss": 3.4815198706106676,
"tokens_seen": 1679884288
},
{
"epoch": 0.51,
"learning_rate": 0.00024801404212637917,
"loss": 2.7773,
"theoretical_loss": 3.481508154209582,
"tokens_seen": 1679949824
},
{
"epoch": 0.51,
"learning_rate": 0.00024800401203610835,
"loss": 2.6279,
"theoretical_loss": 3.4814964383935245,
"tokens_seen": 1680015360
},
{
"epoch": 0.51,
"learning_rate": 0.00024799398194583753,
"loss": 2.7728,
"theoretical_loss": 3.4814847231624437,
"tokens_seen": 1680080896
},
{
"epoch": 0.51,
"learning_rate": 0.0002479839518555667,
"loss": 2.5147,
"theoretical_loss": 3.4814730085162875,
"tokens_seen": 1680146432
},
{
"epoch": 0.51,
"learning_rate": 0.0002479739217652959,
"loss": 2.5198,
"theoretical_loss": 3.4814612944550034,
"tokens_seen": 1680211968
},
{
"epoch": 0.51,
"learning_rate": 0.0002479638916750251,
"loss": 2.5498,
"theoretical_loss": 3.4814495809785395,
"tokens_seen": 1680277504
},
{
"epoch": 0.51,
"learning_rate": 0.00024795386158475425,
"loss": 2.5965,
"theoretical_loss": 3.4814378680868447,
"tokens_seen": 1680343040
},
{
"epoch": 0.51,
"learning_rate": 0.00024794383149448344,
"loss": 2.6287,
"theoretical_loss": 3.4814261557798654,
"tokens_seen": 1680408576
},
{
"epoch": 0.51,
"learning_rate": 0.00024793380140421267,
"loss": 2.6538,
"theoretical_loss": 3.4814144440575507,
"tokens_seen": 1680474112
},
{
"epoch": 0.51,
"learning_rate": 0.00024792377131394185,
"loss": 2.6636,
"theoretical_loss": 3.4814027329198485,
"tokens_seen": 1680539648
},
{
"epoch": 0.51,
"learning_rate": 0.00024791374122367103,
"loss": 2.5748,
"theoretical_loss": 3.481391022366707,
"tokens_seen": 1680605184
},
{
"epoch": 0.51,
"learning_rate": 0.0002479037111334002,
"loss": 2.7609,
"theoretical_loss": 3.481379312398073,
"tokens_seen": 1680670720
},
{
"epoch": 0.51,
"learning_rate": 0.0002478936810431294,
"loss": 2.6647,
"theoretical_loss": 3.4813676030138963,
"tokens_seen": 1680736256
},
{
"epoch": 0.51,
"learning_rate": 0.0002478836509528586,
"loss": 2.6718,
"theoretical_loss": 3.481355894214124,
"tokens_seen": 1680801792
},
{
"epoch": 0.51,
"learning_rate": 0.00024787362086258776,
"loss": 2.5461,
"theoretical_loss": 3.481344185998704,
"tokens_seen": 1680867328
},
{
"epoch": 0.51,
"learning_rate": 0.00024786359077231694,
"loss": 2.6993,
"theoretical_loss": 3.4813324783675847,
"tokens_seen": 1680932864
},
{
"epoch": 0.51,
"objective/train/docs_used": 952727,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6038718223571777,
"objective/train/theoretical_loss": 3.481320771320714,
"objective/train/tokens_used": 1701458400,
"theoretical_loss": 3.481320771320714,
"tokens_seen": 1680998400
},
{
"epoch": 0.51,
"learning_rate": 0.0002478535606820462,
"loss": 2.6243,
"theoretical_loss": 3.481320771320714,
"tokens_seen": 1680998400
},
{
"epoch": 0.51,
"learning_rate": 0.00024784353059177536,
"loss": 2.5981,
"theoretical_loss": 3.4813090648580403,
"tokens_seen": 1681063936
},
{
"epoch": 0.51,
"learning_rate": 0.00024783350050150454,
"loss": 2.6173,
"theoretical_loss": 3.4812973589795106,
"tokens_seen": 1681129472
},
{
"epoch": 0.51,
"learning_rate": 0.0002478234704112337,
"loss": 2.8379,
"theoretical_loss": 3.4812856536850747,
"tokens_seen": 1681195008
},
{
"epoch": 0.51,
"learning_rate": 0.0002478134403209629,
"loss": 2.8352,
"theoretical_loss": 3.4812739489746796,
"tokens_seen": 1681260544
},
{
"epoch": 0.51,
"learning_rate": 0.0002478034102306921,
"loss": 2.6722,
"theoretical_loss": 3.4812622448482733,
"tokens_seen": 1681326080
},
{
"epoch": 0.51,
"learning_rate": 0.00024779338014042126,
"loss": 2.6676,
"theoretical_loss": 3.481250541305804,
"tokens_seen": 1681391616
},
{
"epoch": 0.51,
"learning_rate": 0.00024778335005015044,
"loss": 2.7582,
"theoretical_loss": 3.4812388383472204,
"tokens_seen": 1681457152
},
{
"epoch": 0.51,
"learning_rate": 0.0002477733199598796,
"loss": 2.908,
"theoretical_loss": 3.4812271359724702,
"tokens_seen": 1681522688
},
{
"epoch": 0.51,
"learning_rate": 0.00024776328986960886,
"loss": 2.6319,
"theoretical_loss": 3.481215434181501,
"tokens_seen": 1681588224
},
{
"epoch": 0.51,
"learning_rate": 0.00024775325977933804,
"loss": 2.6467,
"theoretical_loss": 3.481203732974262,
"tokens_seen": 1681653760
},
{
"epoch": 0.51,
"learning_rate": 0.0002477432296890672,
"loss": 2.7919,
"theoretical_loss": 3.481192032350701,
"tokens_seen": 1681719296
},
{
"epoch": 0.51,
"learning_rate": 0.0002477331995987964,
"loss": 2.5796,
"theoretical_loss": 3.481180332310765,
"tokens_seen": 1681784832
},
{
"epoch": 0.51,
"learning_rate": 0.0002477231695085256,
"loss": 2.7339,
"theoretical_loss": 3.481168632854404,
"tokens_seen": 1681850368
},
{
"epoch": 0.51,
"learning_rate": 0.00024771313941825476,
"loss": 2.4857,
"theoretical_loss": 3.4811569339815644,
"tokens_seen": 1681915904
},
{
"epoch": 0.51,
"learning_rate": 0.00024770310932798395,
"loss": 2.7703,
"theoretical_loss": 3.4811452356921953,
"tokens_seen": 1681981440
},
{
"epoch": 0.51,
"learning_rate": 0.0002476930792377131,
"loss": 2.6917,
"theoretical_loss": 3.4811335379862456,
"tokens_seen": 1682046976
},
{
"epoch": 0.51,
"learning_rate": 0.0002476830491474423,
"loss": 2.5297,
"theoretical_loss": 3.4811218408636617,
"tokens_seen": 1682112512
},
{
"epoch": 0.51,
"learning_rate": 0.00024767301905717154,
"loss": 2.4179,
"theoretical_loss": 3.481110144324393,
"tokens_seen": 1682178048
},
{
"epoch": 0.51,
"learning_rate": 0.0002476629889669007,
"loss": 2.6188,
"theoretical_loss": 3.481098448368387,
"tokens_seen": 1682243584
},
{
"epoch": 0.51,
"learning_rate": 0.0002476529588766299,
"loss": 2.48,
"theoretical_loss": 3.4810867529955933,
"tokens_seen": 1682309120
},
{
"epoch": 0.51,
"learning_rate": 0.0002476429287863591,
"loss": 2.557,
"theoretical_loss": 3.481075058205958,
"tokens_seen": 1682374656
},
{
"epoch": 0.51,
"learning_rate": 0.00024763289869608827,
"loss": 2.509,
"theoretical_loss": 3.4810633639994313,
"tokens_seen": 1682440192
},
{
"epoch": 0.51,
"learning_rate": 0.00024762286860581745,
"loss": 2.6768,
"theoretical_loss": 3.48105167037596,
"tokens_seen": 1682505728
},
{
"epoch": 0.51,
"learning_rate": 0.00024761283851554663,
"loss": 2.563,
"theoretical_loss": 3.4810399773354925,
"tokens_seen": 1682571264
},
{
"epoch": 0.51,
"objective/train/docs_used": 953335,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.893624782562256,
"objective/train/theoretical_loss": 3.4810282848779783,
"objective/train/tokens_used": 1703096800,
"theoretical_loss": 3.4810282848779783,
"tokens_seen": 1682636800
},
{
"epoch": 0.51,
"learning_rate": 0.0002476028084252758,
"loss": 2.5565,
"theoretical_loss": 3.4810282848779783,
"tokens_seen": 1682636800
},
{
"epoch": 0.51,
"learning_rate": 0.000247592778335005,
"loss": 2.6724,
"theoretical_loss": 3.481016593003364,
"tokens_seen": 1682702336
},
{
"epoch": 0.51,
"learning_rate": 0.00024758274824473423,
"loss": 2.7617,
"theoretical_loss": 3.4810049017115983,
"tokens_seen": 1682767872
},
{
"epoch": 0.51,
"learning_rate": 0.0002475727181544634,
"loss": 2.7224,
"theoretical_loss": 3.4809932110026303,
"tokens_seen": 1682833408
},
{
"epoch": 0.51,
"learning_rate": 0.0002475626880641926,
"loss": 2.6866,
"theoretical_loss": 3.4809815208764077,
"tokens_seen": 1682898944
},
{
"epoch": 0.51,
"learning_rate": 0.00024755265797392177,
"loss": 2.6385,
"theoretical_loss": 3.480969831332878,
"tokens_seen": 1682964480
},
{
"epoch": 0.51,
"learning_rate": 0.00024754262788365095,
"loss": 2.4573,
"theoretical_loss": 3.4809581423719904,
"tokens_seen": 1683030016
},
{
"epoch": 0.51,
"learning_rate": 0.00024753259779338013,
"loss": 2.6785,
"theoretical_loss": 3.480946453993693,
"tokens_seen": 1683095552
},
{
"epoch": 0.51,
"learning_rate": 0.0002475225677031093,
"loss": 2.5849,
"theoretical_loss": 3.480934766197934,
"tokens_seen": 1683161088
},
{
"epoch": 0.51,
"learning_rate": 0.0002475125376128385,
"loss": 2.7282,
"theoretical_loss": 3.480923078984662,
"tokens_seen": 1683226624
},
{
"epoch": 0.51,
"learning_rate": 0.0002475025075225677,
"loss": 2.5553,
"theoretical_loss": 3.480911392353825,
"tokens_seen": 1683292160
},
{
"epoch": 0.51,
"learning_rate": 0.0002474924774322969,
"loss": 2.5595,
"theoretical_loss": 3.4808997063053706,
"tokens_seen": 1683357696
},
{
"epoch": 0.51,
"learning_rate": 0.0002474824473420261,
"loss": 2.6916,
"theoretical_loss": 3.480888020839248,
"tokens_seen": 1683423232
},
{
"epoch": 0.51,
"learning_rate": 0.0002474724172517553,
"loss": 2.7777,
"theoretical_loss": 3.480876335955406,
"tokens_seen": 1683488768
},
{
"epoch": 0.51,
"learning_rate": 0.00024746238716148445,
"loss": 2.79,
"theoretical_loss": 3.4808646516537918,
"tokens_seen": 1683554304
},
{
"epoch": 0.51,
"learning_rate": 0.00024745235707121364,
"loss": 2.5959,
"theoretical_loss": 3.4808529679343536,
"tokens_seen": 1683619840
},
{
"epoch": 0.51,
"learning_rate": 0.0002474423269809428,
"loss": 2.4946,
"theoretical_loss": 3.480841284797041,
"tokens_seen": 1683685376
},
{
"epoch": 0.51,
"learning_rate": 0.000247432296890672,
"loss": 2.8243,
"theoretical_loss": 3.4808296022418013,
"tokens_seen": 1683750912
},
{
"epoch": 0.51,
"learning_rate": 0.0002474222668004012,
"loss": 2.6303,
"theoretical_loss": 3.480817920268583,
"tokens_seen": 1683816448
},
{
"epoch": 0.51,
"learning_rate": 0.00024741223671013036,
"loss": 2.6817,
"theoretical_loss": 3.480806238877335,
"tokens_seen": 1683881984
},
{
"epoch": 0.51,
"learning_rate": 0.0002474022066198596,
"loss": 2.4385,
"theoretical_loss": 3.4807945580680055,
"tokens_seen": 1683947520
},
{
"epoch": 0.51,
"learning_rate": 0.0002473921765295888,
"loss": 2.4011,
"theoretical_loss": 3.480782877840542,
"tokens_seen": 1684013056
},
{
"epoch": 0.51,
"learning_rate": 0.00024738214643931796,
"loss": 2.8692,
"theoretical_loss": 3.480771198194894,
"tokens_seen": 1684078592
},
{
"epoch": 0.51,
"learning_rate": 0.00024737211634904714,
"loss": 2.7393,
"theoretical_loss": 3.4807595191310092,
"tokens_seen": 1684144128
},
{
"epoch": 0.51,
"learning_rate": 0.0002473620862587764,
"loss": 2.5403,
"theoretical_loss": 3.4807478406488364,
"tokens_seen": 1684209664
},
{
"epoch": 0.51,
"objective/train/docs_used": 954676,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6231210231781006,
"objective/train/theoretical_loss": 3.4807361627483235,
"objective/train/tokens_used": 1704735200,
"theoretical_loss": 3.4807361627483235,
"tokens_seen": 1684275200
},
{
"epoch": 0.51,
"learning_rate": 0.00024735205616850556,
"loss": 2.4159,
"theoretical_loss": 3.4807361627483235,
"tokens_seen": 1684275200
},
{
"epoch": 0.51,
"learning_rate": 0.00024734202607823474,
"loss": 2.4699,
"theoretical_loss": 3.4807244854294197,
"tokens_seen": 1684340736
},
{
"epoch": 0.51,
"learning_rate": 0.0002473319959879639,
"loss": 2.6387,
"theoretical_loss": 3.480712808692072,
"tokens_seen": 1684406272
},
{
"epoch": 0.51,
"learning_rate": 0.0002473219658976931,
"loss": 2.5981,
"theoretical_loss": 3.4807011325362307,
"tokens_seen": 1684471808
},
{
"epoch": 0.51,
"learning_rate": 0.0002473119358074223,
"loss": 2.6625,
"theoretical_loss": 3.4806894569618425,
"tokens_seen": 1684537344
},
{
"epoch": 0.51,
"learning_rate": 0.00024730190571715146,
"loss": 2.517,
"theoretical_loss": 3.480677781968857,
"tokens_seen": 1684602880
},
{
"epoch": 0.51,
"learning_rate": 0.00024729187562688064,
"loss": 2.5409,
"theoretical_loss": 3.480666107557221,
"tokens_seen": 1684668416
},
{
"epoch": 0.51,
"learning_rate": 0.0002472818455366098,
"loss": 2.5752,
"theoretical_loss": 3.4806544337268854,
"tokens_seen": 1684733952
},
{
"epoch": 0.51,
"learning_rate": 0.00024727181544633906,
"loss": 2.9091,
"theoretical_loss": 3.4806427604777967,
"tokens_seen": 1684799488
},
{
"epoch": 0.51,
"learning_rate": 0.00024726178535606824,
"loss": 2.8457,
"theoretical_loss": 3.4806310878099045,
"tokens_seen": 1684865024
},
{
"epoch": 0.51,
"learning_rate": 0.0002472517552657974,
"loss": 2.504,
"theoretical_loss": 3.4806194157231567,
"tokens_seen": 1684930560
},
{
"epoch": 0.51,
"learning_rate": 0.0002472417251755266,
"loss": 2.591,
"theoretical_loss": 3.480607744217502,
"tokens_seen": 1684996096
},
{
"epoch": 0.51,
"learning_rate": 0.0002472316950852558,
"loss": 2.8523,
"theoretical_loss": 3.4805960732928876,
"tokens_seen": 1685061632
},
{
"epoch": 0.51,
"learning_rate": 0.00024722166499498496,
"loss": 2.5836,
"theoretical_loss": 3.480584402949264,
"tokens_seen": 1685127168
},
{
"epoch": 0.51,
"learning_rate": 0.00024721163490471415,
"loss": 2.8698,
"theoretical_loss": 3.4805727331865786,
"tokens_seen": 1685192704
},
{
"epoch": 0.51,
"learning_rate": 0.0002472016048144433,
"loss": 2.7815,
"theoretical_loss": 3.48056106400478,
"tokens_seen": 1685258240
},
{
"epoch": 0.51,
"learning_rate": 0.0002471915747241725,
"loss": 2.7265,
"theoretical_loss": 3.480549395403817,
"tokens_seen": 1685323776
},
{
"epoch": 0.51,
"learning_rate": 0.00024718154463390174,
"loss": 2.6362,
"theoretical_loss": 3.4805377273836378,
"tokens_seen": 1685389312
},
{
"epoch": 0.51,
"learning_rate": 0.0002471715145436309,
"loss": 2.7158,
"theoretical_loss": 3.4805260599441903,
"tokens_seen": 1685454848
},
{
"epoch": 0.51,
"learning_rate": 0.0002471614844533601,
"loss": 2.8307,
"theoretical_loss": 3.480514393085424,
"tokens_seen": 1685520384
},
{
"epoch": 0.51,
"learning_rate": 0.0002471514543630893,
"loss": 2.8201,
"theoretical_loss": 3.4805027268072877,
"tokens_seen": 1685585920
},
{
"epoch": 0.51,
"learning_rate": 0.00024714142427281847,
"loss": 2.6145,
"theoretical_loss": 3.4804910611097286,
"tokens_seen": 1685651456
},
{
"epoch": 0.51,
"learning_rate": 0.00024713139418254765,
"loss": 2.5886,
"theoretical_loss": 3.4804793959926963,
"tokens_seen": 1685716992
},
{
"epoch": 0.51,
"learning_rate": 0.00024712136409227683,
"loss": 2.6087,
"theoretical_loss": 3.480467731456139,
"tokens_seen": 1685782528
},
{
"epoch": 0.51,
"learning_rate": 0.000247111334002006,
"loss": 2.7728,
"theoretical_loss": 3.4804560675000054,
"tokens_seen": 1685848064
},
{
"epoch": 0.51,
"objective/train/docs_used": 955358,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.922814130783081,
"objective/train/theoretical_loss": 3.4804444041242437,
"objective/train/tokens_used": 1706373600,
"theoretical_loss": 3.4804444041242437,
"tokens_seen": 1685913600
},
{
"epoch": 0.51,
"learning_rate": 0.0002471013039117352,
"loss": 2.7638,
"theoretical_loss": 3.4804444041242437,
"tokens_seen": 1685913600
},
{
"epoch": 0.51,
"learning_rate": 0.00024709127382146443,
"loss": 2.4838,
"theoretical_loss": 3.480432741328803,
"tokens_seen": 1685979136
},
{
"epoch": 0.51,
"learning_rate": 0.0002470812437311936,
"loss": 2.6885,
"theoretical_loss": 3.4804210791136314,
"tokens_seen": 1686044672
},
{
"epoch": 0.51,
"learning_rate": 0.0002470712136409228,
"loss": 2.4876,
"theoretical_loss": 3.4804094174786773,
"tokens_seen": 1686110208
},
{
"epoch": 0.51,
"learning_rate": 0.00024706118355065197,
"loss": 2.7682,
"theoretical_loss": 3.4803977564238897,
"tokens_seen": 1686175744
},
{
"epoch": 0.51,
"learning_rate": 0.00024705115346038115,
"loss": 2.2088,
"theoretical_loss": 3.4803860959492177,
"tokens_seen": 1686241280
},
{
"epoch": 0.51,
"learning_rate": 0.00024704112337011033,
"loss": 2.7202,
"theoretical_loss": 3.4803744360546087,
"tokens_seen": 1686306816
},
{
"epoch": 0.51,
"learning_rate": 0.0002470310932798395,
"loss": 2.8202,
"theoretical_loss": 3.4803627767400123,
"tokens_seen": 1686372352
},
{
"epoch": 0.51,
"learning_rate": 0.0002470210631895687,
"loss": 2.8677,
"theoretical_loss": 3.480351118005376,
"tokens_seen": 1686437888
},
{
"epoch": 0.51,
"learning_rate": 0.0002470110330992979,
"loss": 2.5866,
"theoretical_loss": 3.48033945985065,
"tokens_seen": 1686503424
},
{
"epoch": 0.51,
"learning_rate": 0.0002470010030090271,
"loss": 2.7014,
"theoretical_loss": 3.4803278022757818,
"tokens_seen": 1686568960
},
{
"epoch": 0.51,
"learning_rate": 0.0002469909729187563,
"loss": 2.793,
"theoretical_loss": 3.4803161452807196,
"tokens_seen": 1686634496
},
{
"epoch": 0.51,
"learning_rate": 0.0002469809428284855,
"loss": 2.4824,
"theoretical_loss": 3.4803044888654133,
"tokens_seen": 1686700032
},
{
"epoch": 0.51,
"learning_rate": 0.00024697091273821466,
"loss": 2.815,
"theoretical_loss": 3.480292833029811,
"tokens_seen": 1686765568
},
{
"epoch": 0.51,
"learning_rate": 0.00024696088264794384,
"loss": 2.6011,
"theoretical_loss": 3.480281177773861,
"tokens_seen": 1686831104
},
{
"epoch": 0.51,
"learning_rate": 0.000246950852557673,
"loss": 2.4531,
"theoretical_loss": 3.4802695230975127,
"tokens_seen": 1686896640
},
{
"epoch": 0.51,
"learning_rate": 0.0002469408224674022,
"loss": 2.5044,
"theoretical_loss": 3.480257869000714,
"tokens_seen": 1686962176
},
{
"epoch": 0.51,
"learning_rate": 0.0002469307923771314,
"loss": 2.6549,
"theoretical_loss": 3.4802462154834135,
"tokens_seen": 1687027712
},
{
"epoch": 0.51,
"learning_rate": 0.00024692076228686056,
"loss": 2.8623,
"theoretical_loss": 3.4802345625455606,
"tokens_seen": 1687093248
},
{
"epoch": 0.51,
"learning_rate": 0.0002469107321965898,
"loss": 2.4523,
"theoretical_loss": 3.480222910187104,
"tokens_seen": 1687158784
},
{
"epoch": 0.51,
"learning_rate": 0.000246900702106319,
"loss": 2.6886,
"theoretical_loss": 3.4802112584079916,
"tokens_seen": 1687224320
},
{
"epoch": 0.51,
"learning_rate": 0.00024689067201604816,
"loss": 2.6271,
"theoretical_loss": 3.4801996072081725,
"tokens_seen": 1687289856
},
{
"epoch": 0.51,
"learning_rate": 0.00024688064192577734,
"loss": 2.5127,
"theoretical_loss": 3.4801879565875957,
"tokens_seen": 1687355392
},
{
"epoch": 0.51,
"learning_rate": 0.0002468706118355065,
"loss": 2.6098,
"theoretical_loss": 3.4801763065462095,
"tokens_seen": 1687420928
},
{
"epoch": 0.51,
"learning_rate": 0.0002468605817452357,
"loss": 2.7558,
"theoretical_loss": 3.4801646570839626,
"tokens_seen": 1687486464
},
{
"epoch": 0.51,
"objective/train/docs_used": 956671,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8917486667633057,
"objective/train/theoretical_loss": 3.480153008200804,
"objective/train/tokens_used": 1708012000,
"theoretical_loss": 3.480153008200804,
"tokens_seen": 1687552000
},
{
"epoch": 0.51,
"learning_rate": 0.0002468505516549649,
"loss": 2.7244,
"theoretical_loss": 3.480153008200804,
"tokens_seen": 1687552000
},
{
"epoch": 0.51,
"learning_rate": 0.00024684052156469406,
"loss": 2.6498,
"theoretical_loss": 3.480141359896683,
"tokens_seen": 1687617536
},
{
"epoch": 0.51,
"learning_rate": 0.00024683049147442325,
"loss": 2.7708,
"theoretical_loss": 3.4801297121715464,
"tokens_seen": 1687683072
},
{
"epoch": 0.51,
"learning_rate": 0.0002468204613841525,
"loss": 2.7246,
"theoretical_loss": 3.4801180650253447,
"tokens_seen": 1687748608
},
{
"epoch": 0.51,
"learning_rate": 0.00024681043129388166,
"loss": 2.5536,
"theoretical_loss": 3.480106418458026,
"tokens_seen": 1687814144
},
{
"epoch": 0.51,
"learning_rate": 0.00024680040120361084,
"loss": 2.7861,
"theoretical_loss": 3.4800947724695392,
"tokens_seen": 1687879680
},
{
"epoch": 0.51,
"learning_rate": 0.00024679037111334,
"loss": 2.6922,
"theoretical_loss": 3.4800831270598334,
"tokens_seen": 1687945216
},
{
"epoch": 0.51,
"learning_rate": 0.0002467803410230692,
"loss": 2.5222,
"theoretical_loss": 3.4800714822288565,
"tokens_seen": 1688010752
},
{
"epoch": 0.51,
"learning_rate": 0.0002467703109327984,
"loss": 2.522,
"theoretical_loss": 3.4800598379765577,
"tokens_seen": 1688076288
},
{
"epoch": 0.51,
"learning_rate": 0.00024676028084252757,
"loss": 2.7297,
"theoretical_loss": 3.4800481943028863,
"tokens_seen": 1688141824
},
{
"epoch": 0.51,
"learning_rate": 0.00024675025075225675,
"loss": 2.7582,
"theoretical_loss": 3.48003655120779,
"tokens_seen": 1688207360
},
{
"epoch": 0.51,
"learning_rate": 0.00024674022066198593,
"loss": 2.5938,
"theoretical_loss": 3.4800249086912185,
"tokens_seen": 1688272896
},
{
"epoch": 0.51,
"learning_rate": 0.00024673019057171516,
"loss": 2.4862,
"theoretical_loss": 3.4800132667531205,
"tokens_seen": 1688338432
},
{
"epoch": 0.51,
"learning_rate": 0.00024672016048144435,
"loss": 2.5047,
"theoretical_loss": 3.4800016253934447,
"tokens_seen": 1688403968
},
{
"epoch": 0.51,
"learning_rate": 0.0002467101303911735,
"loss": 2.505,
"theoretical_loss": 3.4799899846121396,
"tokens_seen": 1688469504
},
{
"epoch": 0.51,
"learning_rate": 0.0002467001003009027,
"loss": 2.7019,
"theoretical_loss": 3.4799783444091545,
"tokens_seen": 1688535040
},
{
"epoch": 0.51,
"learning_rate": 0.0002466900702106319,
"loss": 2.5954,
"theoretical_loss": 3.4799667047844376,
"tokens_seen": 1688600576
},
{
"epoch": 0.51,
"learning_rate": 0.00024668004012036107,
"loss": 2.5804,
"theoretical_loss": 3.479955065737938,
"tokens_seen": 1688666112
},
{
"epoch": 0.51,
"learning_rate": 0.00024667001003009025,
"loss": 2.6949,
"theoretical_loss": 3.479943427269605,
"tokens_seen": 1688731648
},
{
"epoch": 0.51,
"learning_rate": 0.00024665997993981943,
"loss": 2.6449,
"theoretical_loss": 3.4799317893793873,
"tokens_seen": 1688797184
},
{
"epoch": 0.51,
"learning_rate": 0.0002466499498495486,
"loss": 2.7089,
"theoretical_loss": 3.479920152067233,
"tokens_seen": 1688862720
},
{
"epoch": 0.51,
"learning_rate": 0.00024663991975927785,
"loss": 2.5648,
"theoretical_loss": 3.4799085153330918,
"tokens_seen": 1688928256
},
{
"epoch": 0.51,
"learning_rate": 0.00024662988966900703,
"loss": 2.6195,
"theoretical_loss": 3.479896879176912,
"tokens_seen": 1688993792
},
{
"epoch": 0.51,
"learning_rate": 0.0002466198595787362,
"loss": 2.8753,
"theoretical_loss": 3.479885243598643,
"tokens_seen": 1689059328
},
{
"epoch": 0.51,
"learning_rate": 0.0002466098294884654,
"loss": 2.6734,
"theoretical_loss": 3.4798736085982336,
"tokens_seen": 1689124864
},
{
"epoch": 0.51,
"objective/train/docs_used": 957370,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.203357458114624,
"objective/train/theoretical_loss": 3.479861974175632,
"objective/train/tokens_used": 1709650400,
"theoretical_loss": 3.479861974175632,
"tokens_seen": 1689190400
},
{
"epoch": 0.51,
"learning_rate": 0.00024659979939819463,
"loss": 2.3916,
"theoretical_loss": 3.479861974175632,
"tokens_seen": 1689190400
},
{
"epoch": 0.51,
"learning_rate": 0.0002465897693079238,
"loss": 2.5434,
"theoretical_loss": 3.4798503403307874,
"tokens_seen": 1689255936
},
{
"epoch": 0.51,
"learning_rate": 0.000246579739217653,
"loss": 2.5062,
"theoretical_loss": 3.4798387070636494,
"tokens_seen": 1689321472
},
{
"epoch": 0.51,
"learning_rate": 0.00024656970912738217,
"loss": 2.632,
"theoretical_loss": 3.479827074374166,
"tokens_seen": 1689387008
},
{
"epoch": 0.51,
"learning_rate": 0.00024655967903711135,
"loss": 2.6542,
"theoretical_loss": 3.4798154422622867,
"tokens_seen": 1689452544
},
{
"epoch": 0.51,
"learning_rate": 0.00024654964894684053,
"loss": 2.5291,
"theoretical_loss": 3.4798038107279603,
"tokens_seen": 1689518080
},
{
"epoch": 0.51,
"learning_rate": 0.0002465396188565697,
"loss": 2.3895,
"theoretical_loss": 3.4797921797711355,
"tokens_seen": 1689583616
},
{
"epoch": 0.51,
"learning_rate": 0.0002465295887662989,
"loss": 2.8177,
"theoretical_loss": 3.4797805493917613,
"tokens_seen": 1689649152
},
{
"epoch": 0.51,
"learning_rate": 0.0002465195586760281,
"loss": 2.6103,
"theoretical_loss": 3.4797689195897865,
"tokens_seen": 1689714688
},
{
"epoch": 0.51,
"learning_rate": 0.0002465095285857573,
"loss": 2.7034,
"theoretical_loss": 3.4797572903651606,
"tokens_seen": 1689780224
},
{
"epoch": 0.51,
"learning_rate": 0.0002464994984954865,
"loss": 2.5226,
"theoretical_loss": 3.479745661717832,
"tokens_seen": 1689845760
},
{
"epoch": 0.51,
"learning_rate": 0.0002464894684052157,
"loss": 2.4451,
"theoretical_loss": 3.47973403364775,
"tokens_seen": 1689911296
},
{
"epoch": 0.51,
"learning_rate": 0.00024647943831494486,
"loss": 2.814,
"theoretical_loss": 3.4797224061548633,
"tokens_seen": 1689976832
},
{
"epoch": 0.51,
"learning_rate": 0.00024646940822467404,
"loss": 2.6171,
"theoretical_loss": 3.4797107792391206,
"tokens_seen": 1690042368
},
{
"epoch": 0.51,
"learning_rate": 0.0002464593781344032,
"loss": 2.5255,
"theoretical_loss": 3.4796991529004715,
"tokens_seen": 1690107904
},
{
"epoch": 0.51,
"learning_rate": 0.0002464493480441324,
"loss": 2.7988,
"theoretical_loss": 3.4796875271388648,
"tokens_seen": 1690173440
},
{
"epoch": 0.51,
"learning_rate": 0.0002464393179538616,
"loss": 3.0013,
"theoretical_loss": 3.4796759019542494,
"tokens_seen": 1690238976
},
{
"epoch": 0.51,
"learning_rate": 0.00024642928786359076,
"loss": 2.6229,
"theoretical_loss": 3.479664277346574,
"tokens_seen": 1690304512
},
{
"epoch": 0.51,
"learning_rate": 0.00024641925777332,
"loss": 2.4018,
"theoretical_loss": 3.479652653315788,
"tokens_seen": 1690370048
},
{
"epoch": 0.51,
"learning_rate": 0.0002464092276830492,
"loss": 2.8738,
"theoretical_loss": 3.4796410298618405,
"tokens_seen": 1690435584
},
{
"epoch": 0.51,
"learning_rate": 0.00024639919759277836,
"loss": 2.4213,
"theoretical_loss": 3.4796294069846803,
"tokens_seen": 1690501120
},
{
"epoch": 0.51,
"learning_rate": 0.00024638916750250754,
"loss": 2.726,
"theoretical_loss": 3.4796177846842564,
"tokens_seen": 1690566656
},
{
"epoch": 0.51,
"learning_rate": 0.0002463791374122367,
"loss": 2.5926,
"theoretical_loss": 3.479606162960518,
"tokens_seen": 1690632192
},
{
"epoch": 0.51,
"learning_rate": 0.0002463691073219659,
"loss": 2.8561,
"theoretical_loss": 3.4795945418134133,
"tokens_seen": 1690697728
},
{
"epoch": 0.51,
"learning_rate": 0.0002463590772316951,
"loss": 2.6873,
"theoretical_loss": 3.479582921242893,
"tokens_seen": 1690763264
},
{
"epoch": 0.51,
"objective/train/docs_used": 957881,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.909823179244995,
"objective/train/theoretical_loss": 3.4795713012489045,
"objective/train/tokens_used": 1711288800,
"theoretical_loss": 3.4795713012489045,
"tokens_seen": 1690828800
},
{
"epoch": 0.51,
"learning_rate": 0.00024634904714142426,
"loss": 2.8692,
"theoretical_loss": 3.4795713012489045,
"tokens_seen": 1690828800
},
{
"epoch": 0.51,
"learning_rate": 0.00024633901705115345,
"loss": 2.4031,
"theoretical_loss": 3.4795596818313976,
"tokens_seen": 1690894336
},
{
"epoch": 0.51,
"learning_rate": 0.0002463289869608827,
"loss": 2.6242,
"theoretical_loss": 3.4795480629903217,
"tokens_seen": 1690959872
},
{
"epoch": 0.51,
"learning_rate": 0.00024631895687061186,
"loss": 2.6184,
"theoretical_loss": 3.479536444725625,
"tokens_seen": 1691025408
},
{
"epoch": 0.51,
"learning_rate": 0.00024630892678034104,
"loss": 2.6527,
"theoretical_loss": 3.4795248270372574,
"tokens_seen": 1691090944
},
{
"epoch": 0.51,
"learning_rate": 0.0002462988966900702,
"loss": 2.7032,
"theoretical_loss": 3.479513209925167,
"tokens_seen": 1691156480
},
{
"epoch": 0.51,
"learning_rate": 0.0002462888665997994,
"loss": 2.6226,
"theoretical_loss": 3.479501593389304,
"tokens_seen": 1691222016
},
{
"epoch": 0.51,
"learning_rate": 0.0002462788365095286,
"loss": 2.238,
"theoretical_loss": 3.479489977429617,
"tokens_seen": 1691287552
},
{
"epoch": 0.51,
"learning_rate": 0.00024626880641925777,
"loss": 2.3959,
"theoretical_loss": 3.4794783620460548,
"tokens_seen": 1691353088
},
{
"epoch": 0.51,
"learning_rate": 0.00024625877632898695,
"loss": 2.6894,
"theoretical_loss": 3.479466747238567,
"tokens_seen": 1691418624
},
{
"epoch": 0.51,
"learning_rate": 0.00024624874623871613,
"loss": 2.4358,
"theoretical_loss": 3.4794551330071024,
"tokens_seen": 1691484160
},
{
"epoch": 0.51,
"learning_rate": 0.00024623871614844536,
"loss": 2.5118,
"theoretical_loss": 3.47944351935161,
"tokens_seen": 1691549696
},
{
"epoch": 0.51,
"learning_rate": 0.00024622868605817455,
"loss": 2.6147,
"theoretical_loss": 3.4794319062720396,
"tokens_seen": 1691615232
},
{
"epoch": 0.51,
"learning_rate": 0.0002462186559679037,
"loss": 2.5837,
"theoretical_loss": 3.479420293768339,
"tokens_seen": 1691680768
},
{
"epoch": 0.51,
"learning_rate": 0.0002462086258776329,
"loss": 2.6221,
"theoretical_loss": 3.479408681840459,
"tokens_seen": 1691746304
},
{
"epoch": 0.51,
"learning_rate": 0.0002461985957873621,
"loss": 3.0556,
"theoretical_loss": 3.4793970704883477,
"tokens_seen": 1691811840
},
{
"epoch": 0.51,
"learning_rate": 0.00024618856569709127,
"loss": 2.7471,
"theoretical_loss": 3.4793854597119545,
"tokens_seen": 1691877376
},
{
"epoch": 0.51,
"learning_rate": 0.00024617853560682045,
"loss": 2.7393,
"theoretical_loss": 3.4793738495112283,
"tokens_seen": 1691942912
},
{
"epoch": 0.51,
"learning_rate": 0.00024616850551654963,
"loss": 2.4923,
"theoretical_loss": 3.4793622398861186,
"tokens_seen": 1692008448
},
{
"epoch": 0.51,
"learning_rate": 0.0002461584754262788,
"loss": 2.5775,
"theoretical_loss": 3.4793506308365747,
"tokens_seen": 1692073984
},
{
"epoch": 0.51,
"learning_rate": 0.00024614844533600805,
"loss": 2.8141,
"theoretical_loss": 3.479339022362545,
"tokens_seen": 1692139520
},
{
"epoch": 0.51,
"learning_rate": 0.00024613841524573723,
"loss": 2.5837,
"theoretical_loss": 3.47932741446398,
"tokens_seen": 1692205056
},
{
"epoch": 0.51,
"learning_rate": 0.0002461283851554664,
"loss": 2.4673,
"theoretical_loss": 3.4793158071408277,
"tokens_seen": 1692270592
},
{
"epoch": 0.51,
"learning_rate": 0.0002461183550651956,
"loss": 2.7079,
"theoretical_loss": 3.4793042003930377,
"tokens_seen": 1692336128
},
{
"epoch": 0.51,
"learning_rate": 0.0002461083249749248,
"loss": 2.7266,
"theoretical_loss": 3.479292594220559,
"tokens_seen": 1692401664
},
{
"epoch": 0.51,
"objective/train/docs_used": 959159,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7859768867492676,
"objective/train/theoretical_loss": 3.4792809886233416,
"objective/train/tokens_used": 1712927200,
"theoretical_loss": 3.4792809886233416,
"tokens_seen": 1692467200
},
{
"epoch": 0.51,
"learning_rate": 0.00024609829488465395,
"loss": 2.7528,
"theoretical_loss": 3.4792809886233416,
"tokens_seen": 1692467200
},
{
"epoch": 0.51,
"learning_rate": 0.00024608826479438314,
"loss": 2.7169,
"theoretical_loss": 3.4792693836013333,
"tokens_seen": 1692532736
},
{
"epoch": 0.51,
"learning_rate": 0.0002460782347041123,
"loss": 2.4495,
"theoretical_loss": 3.4792577791544845,
"tokens_seen": 1692598272
},
{
"epoch": 0.51,
"learning_rate": 0.0002460682046138415,
"loss": 2.5689,
"theoretical_loss": 3.4792461752827446,
"tokens_seen": 1692663808
},
{
"epoch": 0.51,
"learning_rate": 0.00024605817452357073,
"loss": 2.6463,
"theoretical_loss": 3.479234571986062,
"tokens_seen": 1692729344
},
{
"epoch": 0.51,
"learning_rate": 0.0002460481444332999,
"loss": 2.7966,
"theoretical_loss": 3.479222969264386,
"tokens_seen": 1692794880
},
{
"epoch": 0.51,
"learning_rate": 0.0002460381143430291,
"loss": 2.881,
"theoretical_loss": 3.4792113671176663,
"tokens_seen": 1692860416
},
{
"epoch": 0.51,
"learning_rate": 0.0002460280842527583,
"loss": 2.6983,
"theoretical_loss": 3.479199765545852,
"tokens_seen": 1692925952
},
{
"epoch": 0.51,
"learning_rate": 0.00024601805416248746,
"loss": 2.6442,
"theoretical_loss": 3.4791881645488916,
"tokens_seen": 1692991488
},
{
"epoch": 0.51,
"learning_rate": 0.00024600802407221664,
"loss": 2.969,
"theoretical_loss": 3.479176564126736,
"tokens_seen": 1693057024
},
{
"epoch": 0.51,
"learning_rate": 0.0002459979939819458,
"loss": 2.8368,
"theoretical_loss": 3.4791649642793328,
"tokens_seen": 1693122560
},
{
"epoch": 0.51,
"learning_rate": 0.000245987963891675,
"loss": 2.6991,
"theoretical_loss": 3.479153365006632,
"tokens_seen": 1693188096
},
{
"epoch": 0.51,
"learning_rate": 0.0002459779338014042,
"loss": 2.7325,
"theoretical_loss": 3.479141766308584,
"tokens_seen": 1693253632
},
{
"epoch": 0.51,
"learning_rate": 0.0002459679037111334,
"loss": 2.7312,
"theoretical_loss": 3.479130168185136,
"tokens_seen": 1693319168
},
{
"epoch": 0.51,
"learning_rate": 0.0002459578736208626,
"loss": 2.5583,
"theoretical_loss": 3.4791185706362384,
"tokens_seen": 1693384704
},
{
"epoch": 0.51,
"learning_rate": 0.0002459478435305918,
"loss": 2.565,
"theoretical_loss": 3.4791069736618407,
"tokens_seen": 1693450240
},
{
"epoch": 0.51,
"learning_rate": 0.00024593781344032096,
"loss": 2.4533,
"theoretical_loss": 3.479095377261892,
"tokens_seen": 1693515776
},
{
"epoch": 0.51,
"learning_rate": 0.00024592778335005014,
"loss": 2.5664,
"theoretical_loss": 3.4790837814363407,
"tokens_seen": 1693581312
},
{
"epoch": 0.51,
"learning_rate": 0.0002459177532597793,
"loss": 2.6944,
"theoretical_loss": 3.479072186185138,
"tokens_seen": 1693646848
},
{
"epoch": 0.51,
"learning_rate": 0.0002459077231695085,
"loss": 2.453,
"theoretical_loss": 3.4790605915082313,
"tokens_seen": 1693712384
},
{
"epoch": 0.51,
"learning_rate": 0.0002458976930792377,
"loss": 2.7115,
"theoretical_loss": 3.479048997405571,
"tokens_seen": 1693777920
},
{
"epoch": 0.51,
"learning_rate": 0.00024588766298896687,
"loss": 2.7462,
"theoretical_loss": 3.479037403877107,
"tokens_seen": 1693843456
},
{
"epoch": 0.51,
"learning_rate": 0.0002458776328986961,
"loss": 2.5827,
"theoretical_loss": 3.4790258109227867,
"tokens_seen": 1693908992
},
{
"epoch": 0.51,
"learning_rate": 0.0002458676028084253,
"loss": 2.764,
"theoretical_loss": 3.479014218542561,
"tokens_seen": 1693974528
},
{
"epoch": 0.51,
"learning_rate": 0.00024585757271815446,
"loss": 2.7715,
"theoretical_loss": 3.4790026267363796,
"tokens_seen": 1694040064
},
{
"epoch": 0.51,
"objective/train/docs_used": 959847,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.432230234146118,
"objective/train/theoretical_loss": 3.478991035504191,
"objective/train/tokens_used": 1714565600,
"theoretical_loss": 3.478991035504191,
"tokens_seen": 1694105600
},
{
"epoch": 0.51,
"learning_rate": 0.00024584754262788365,
"loss": 2.4065,
"theoretical_loss": 3.478991035504191,
"tokens_seen": 1694105600
},
{
"epoch": 0.51,
"learning_rate": 0.0002458375125376129,
"loss": 2.7557,
"theoretical_loss": 3.4789794448459443,
"tokens_seen": 1694171136
},
{
"epoch": 0.51,
"learning_rate": 0.00024582748244734206,
"loss": 2.6149,
"theoretical_loss": 3.4789678547615894,
"tokens_seen": 1694236672
},
{
"epoch": 0.51,
"learning_rate": 0.00024581745235707124,
"loss": 2.674,
"theoretical_loss": 3.478956265251076,
"tokens_seen": 1694302208
},
{
"epoch": 0.51,
"learning_rate": 0.0002458074222668004,
"loss": 2.6172,
"theoretical_loss": 3.478944676314353,
"tokens_seen": 1694367744
},
{
"epoch": 0.51,
"learning_rate": 0.0002457973921765296,
"loss": 2.6708,
"theoretical_loss": 3.4789330879513694,
"tokens_seen": 1694433280
},
{
"epoch": 0.51,
"learning_rate": 0.0002457873620862588,
"loss": 2.6165,
"theoretical_loss": 3.478921500162076,
"tokens_seen": 1694498816
},
{
"epoch": 0.51,
"learning_rate": 0.00024577733199598797,
"loss": 2.682,
"theoretical_loss": 3.4789099129464205,
"tokens_seen": 1694564352
},
{
"epoch": 0.51,
"learning_rate": 0.00024576730190571715,
"loss": 2.5138,
"theoretical_loss": 3.4788983263043534,
"tokens_seen": 1694629888
},
{
"epoch": 0.51,
"learning_rate": 0.00024575727181544633,
"loss": 2.7513,
"theoretical_loss": 3.478886740235824,
"tokens_seen": 1694695424
},
{
"epoch": 0.51,
"learning_rate": 0.00024574724172517556,
"loss": 2.6109,
"theoretical_loss": 3.4788751547407815,
"tokens_seen": 1694760960
},
{
"epoch": 0.51,
"learning_rate": 0.00024573721163490475,
"loss": 3.0236,
"theoretical_loss": 3.478863569819176,
"tokens_seen": 1694826496
},
{
"epoch": 0.51,
"learning_rate": 0.00024572718154463393,
"loss": 2.6482,
"theoretical_loss": 3.478851985470956,
"tokens_seen": 1694892032
},
{
"epoch": 0.51,
"learning_rate": 0.0002457171514543631,
"loss": 2.5872,
"theoretical_loss": 3.478840401696071,
"tokens_seen": 1694957568
},
{
"epoch": 0.51,
"learning_rate": 0.0002457071213640923,
"loss": 2.6151,
"theoretical_loss": 3.4788288184944713,
"tokens_seen": 1695023104
},
{
"epoch": 0.51,
"learning_rate": 0.00024569709127382147,
"loss": 2.7049,
"theoretical_loss": 3.4788172358661056,
"tokens_seen": 1695088640
},
{
"epoch": 0.51,
"learning_rate": 0.00024568706118355065,
"loss": 2.6693,
"theoretical_loss": 3.478805653810924,
"tokens_seen": 1695154176
},
{
"epoch": 0.51,
"learning_rate": 0.00024567703109327983,
"loss": 2.6889,
"theoretical_loss": 3.478794072328875,
"tokens_seen": 1695219712
},
{
"epoch": 0.51,
"learning_rate": 0.000245667001003009,
"loss": 2.6943,
"theoretical_loss": 3.478782491419909,
"tokens_seen": 1695285248
},
{
"epoch": 0.51,
"learning_rate": 0.00024565697091273825,
"loss": 2.6048,
"theoretical_loss": 3.4787709110839753,
"tokens_seen": 1695350784
},
{
"epoch": 0.51,
"learning_rate": 0.00024564694082246743,
"loss": 2.8623,
"theoretical_loss": 3.478759331321023,
"tokens_seen": 1695416320
},
{
"epoch": 0.51,
"learning_rate": 0.0002456369107321966,
"loss": 2.407,
"theoretical_loss": 3.478747752131002,
"tokens_seen": 1695481856
},
{
"epoch": 0.51,
"learning_rate": 0.0002456268806419258,
"loss": 2.835,
"theoretical_loss": 3.4787361735138624,
"tokens_seen": 1695547392
},
{
"epoch": 0.51,
"learning_rate": 0.000245616850551655,
"loss": 2.437,
"theoretical_loss": 3.478724595469552,
"tokens_seen": 1695612928
},
{
"epoch": 0.51,
"learning_rate": 0.00024560682046138415,
"loss": 2.6151,
"theoretical_loss": 3.4787130179980217,
"tokens_seen": 1695678464
},
{
"epoch": 0.51,
"objective/train/docs_used": 961177,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.755793809890747,
"objective/train/theoretical_loss": 3.478701441099221,
"objective/train/tokens_used": 1716204000,
"theoretical_loss": 3.478701441099221,
"tokens_seen": 1695744000
},
{
"epoch": 0.51,
"learning_rate": 0.00024559679037111334,
"loss": 2.6471,
"theoretical_loss": 3.478701441099221,
"tokens_seen": 1695744000
},
{
"epoch": 0.51,
"learning_rate": 0.0002455867602808425,
"loss": 2.3603,
"theoretical_loss": 3.4786898647730986,
"tokens_seen": 1695809536
},
{
"epoch": 0.51,
"learning_rate": 0.0002455767301905717,
"loss": 2.7419,
"theoretical_loss": 3.478678289019605,
"tokens_seen": 1695875072
},
{
"epoch": 0.51,
"learning_rate": 0.00024556670010030093,
"loss": 2.8361,
"theoretical_loss": 3.478666713838689,
"tokens_seen": 1695940608
},
{
"epoch": 0.51,
"learning_rate": 0.0002455566700100301,
"loss": 2.5771,
"theoretical_loss": 3.478655139230301,
"tokens_seen": 1696006144
},
{
"epoch": 0.51,
"learning_rate": 0.0002455466399197593,
"loss": 2.7917,
"theoretical_loss": 3.478643565194389,
"tokens_seen": 1696071680
},
{
"epoch": 0.51,
"learning_rate": 0.0002455366098294885,
"loss": 2.757,
"theoretical_loss": 3.478631991730904,
"tokens_seen": 1696137216
},
{
"epoch": 0.51,
"learning_rate": 0.00024552657973921766,
"loss": 2.6907,
"theoretical_loss": 3.478620418839795,
"tokens_seen": 1696202752
},
{
"epoch": 0.51,
"learning_rate": 0.00024551654964894684,
"loss": 2.7171,
"theoretical_loss": 3.4786088465210123,
"tokens_seen": 1696268288
},
{
"epoch": 0.51,
"learning_rate": 0.000245506519558676,
"loss": 2.8916,
"theoretical_loss": 3.478597274774504,
"tokens_seen": 1696333824
},
{
"epoch": 0.51,
"learning_rate": 0.0002454964894684052,
"loss": 2.4452,
"theoretical_loss": 3.4785857036002215,
"tokens_seen": 1696399360
},
{
"epoch": 0.51,
"learning_rate": 0.0002454864593781344,
"loss": 2.6818,
"theoretical_loss": 3.478574132998113,
"tokens_seen": 1696464896
},
{
"epoch": 0.51,
"learning_rate": 0.0002454764292878636,
"loss": 2.6244,
"theoretical_loss": 3.478562562968129,
"tokens_seen": 1696530432
},
{
"epoch": 0.51,
"learning_rate": 0.0002454663991975928,
"loss": 2.8235,
"theoretical_loss": 3.4785509935102183,
"tokens_seen": 1696595968
},
{
"epoch": 0.51,
"learning_rate": 0.000245456369107322,
"loss": 2.8818,
"theoretical_loss": 3.4785394246243317,
"tokens_seen": 1696661504
},
{
"epoch": 0.51,
"learning_rate": 0.00024544633901705116,
"loss": 2.8209,
"theoretical_loss": 3.478527856310417,
"tokens_seen": 1696727040
},
{
"epoch": 0.51,
"learning_rate": 0.00024543630892678034,
"loss": 2.9438,
"theoretical_loss": 3.4785162885684255,
"tokens_seen": 1696792576
},
{
"epoch": 0.51,
"learning_rate": 0.0002454262788365095,
"loss": 2.8235,
"theoretical_loss": 3.478504721398306,
"tokens_seen": 1696858112
},
{
"epoch": 0.51,
"learning_rate": 0.0002454162487462387,
"loss": 2.5845,
"theoretical_loss": 3.4784931548000086,
"tokens_seen": 1696923648
},
{
"epoch": 0.51,
"learning_rate": 0.0002454062186559679,
"loss": 2.7245,
"theoretical_loss": 3.4784815887734823,
"tokens_seen": 1696989184
},
{
"epoch": 0.51,
"learning_rate": 0.00024539618856569707,
"loss": 2.5586,
"theoretical_loss": 3.4784700233186774,
"tokens_seen": 1697054720
},
{
"epoch": 0.51,
"learning_rate": 0.0002453861584754263,
"loss": 2.4902,
"theoretical_loss": 3.4784584584355436,
"tokens_seen": 1697120256
},
{
"epoch": 0.51,
"learning_rate": 0.0002453761283851555,
"loss": 2.5492,
"theoretical_loss": 3.47844689412403,
"tokens_seen": 1697185792
},
{
"epoch": 0.51,
"learning_rate": 0.00024536609829488466,
"loss": 2.7469,
"theoretical_loss": 3.4784353303840865,
"tokens_seen": 1697251328
},
{
"epoch": 0.51,
"learning_rate": 0.00024535606820461385,
"loss": 2.5275,
"theoretical_loss": 3.478423767215663,
"tokens_seen": 1697316864
},
{
"epoch": 0.51,
"objective/train/docs_used": 961828,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.719982385635376,
"objective/train/theoretical_loss": 3.478412204618709,
"objective/train/tokens_used": 1717842400,
"theoretical_loss": 3.478412204618709,
"tokens_seen": 1697382400
},
{
"epoch": 0.51,
"learning_rate": 0.000245346038114343,
"loss": 2.7691,
"theoretical_loss": 3.478412204618709,
"tokens_seen": 1697382400
},
{
"epoch": 0.51,
"learning_rate": 0.0002453360080240722,
"loss": 2.5782,
"theoretical_loss": 3.4784006425931744,
"tokens_seen": 1697447936
},
{
"epoch": 0.51,
"learning_rate": 0.0002453259779338014,
"loss": 2.5316,
"theoretical_loss": 3.478389081139009,
"tokens_seen": 1697513472
},
{
"epoch": 0.51,
"learning_rate": 0.00024531594784353057,
"loss": 2.3866,
"theoretical_loss": 3.4783775202561618,
"tokens_seen": 1697579008
},
{
"epoch": 0.51,
"learning_rate": 0.00024530591775325975,
"loss": 2.5912,
"theoretical_loss": 3.4783659599445826,
"tokens_seen": 1697644544
},
{
"epoch": 0.51,
"learning_rate": 0.000245295887662989,
"loss": 2.5956,
"theoretical_loss": 3.478354400204222,
"tokens_seen": 1697710080
},
{
"epoch": 0.51,
"learning_rate": 0.00024528585757271817,
"loss": 2.8235,
"theoretical_loss": 3.4783428410350297,
"tokens_seen": 1697775616
},
{
"epoch": 0.51,
"learning_rate": 0.00024527582748244735,
"loss": 2.8169,
"theoretical_loss": 3.4783312824369537,
"tokens_seen": 1697841152
},
{
"epoch": 0.51,
"learning_rate": 0.00024526579739217653,
"loss": 2.7098,
"theoretical_loss": 3.478319724409946,
"tokens_seen": 1697906688
},
{
"epoch": 0.51,
"learning_rate": 0.0002452557673019057,
"loss": 2.7913,
"theoretical_loss": 3.478308166953955,
"tokens_seen": 1697972224
},
{
"epoch": 0.51,
"learning_rate": 0.0002452457372116349,
"loss": 2.7773,
"theoretical_loss": 3.4782966100689308,
"tokens_seen": 1698037760
},
{
"epoch": 0.51,
"learning_rate": 0.0002452357071213641,
"loss": 2.5952,
"theoretical_loss": 3.4782850537548233,
"tokens_seen": 1698103296
},
{
"epoch": 0.51,
"learning_rate": 0.00024522567703109325,
"loss": 2.8039,
"theoretical_loss": 3.478273498011582,
"tokens_seen": 1698168832
},
{
"epoch": 0.51,
"learning_rate": 0.00024521564694082244,
"loss": 2.6668,
"theoretical_loss": 3.4782619428391564,
"tokens_seen": 1698234368
},
{
"epoch": 0.51,
"learning_rate": 0.00024520561685055167,
"loss": 2.4685,
"theoretical_loss": 3.478250388237497,
"tokens_seen": 1698299904
},
{
"epoch": 0.51,
"learning_rate": 0.00024519558676028085,
"loss": 2.4679,
"theoretical_loss": 3.478238834206553,
"tokens_seen": 1698365440
},
{
"epoch": 0.51,
"learning_rate": 0.00024518555667001003,
"loss": 2.7861,
"theoretical_loss": 3.4782272807462746,
"tokens_seen": 1698430976
},
{
"epoch": 0.51,
"learning_rate": 0.0002451755265797392,
"loss": 2.5515,
"theoretical_loss": 3.4782157278566115,
"tokens_seen": 1698496512
},
{
"epoch": 0.51,
"learning_rate": 0.0002451654964894684,
"loss": 2.6841,
"theoretical_loss": 3.4782041755375133,
"tokens_seen": 1698562048
},
{
"epoch": 0.51,
"learning_rate": 0.0002451554663991976,
"loss": 2.556,
"theoretical_loss": 3.47819262378893,
"tokens_seen": 1698627584
},
{
"epoch": 0.51,
"learning_rate": 0.00024514543630892676,
"loss": 2.815,
"theoretical_loss": 3.478181072610811,
"tokens_seen": 1698693120
},
{
"epoch": 0.51,
"learning_rate": 0.00024513540621865594,
"loss": 2.5977,
"theoretical_loss": 3.478169522003107,
"tokens_seen": 1698758656
},
{
"epoch": 0.51,
"learning_rate": 0.0002451253761283852,
"loss": 2.7824,
"theoretical_loss": 3.478157971965767,
"tokens_seen": 1698824192
},
{
"epoch": 0.51,
"learning_rate": 0.00024511534603811435,
"loss": 2.6552,
"theoretical_loss": 3.4781464224987415,
"tokens_seen": 1698889728
},
{
"epoch": 0.51,
"learning_rate": 0.00024510531594784354,
"loss": 2.8171,
"theoretical_loss": 3.478134873601979,
"tokens_seen": 1698955264
},
{
"epoch": 0.51,
"objective/train/docs_used": 963095,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4242331981658936,
"objective/train/theoretical_loss": 3.4781233252754316,
"objective/train/tokens_used": 1719480800,
"theoretical_loss": 3.4781233252754316,
"tokens_seen": 1699020800
},
{
"epoch": 0.51,
"learning_rate": 0.0002450952858575727,
"loss": 2.6469,
"theoretical_loss": 3.4781233252754316,
"tokens_seen": 1699020800
},
{
"epoch": 0.51,
"learning_rate": 0.0002450852557673019,
"loss": 2.7607,
"theoretical_loss": 3.4781117775190467,
"tokens_seen": 1699086336
},
{
"epoch": 0.51,
"learning_rate": 0.00024507522567703113,
"loss": 2.6419,
"theoretical_loss": 3.4781002303327764,
"tokens_seen": 1699151872
},
{
"epoch": 0.51,
"learning_rate": 0.0002450651955867603,
"loss": 2.6279,
"theoretical_loss": 3.4780886837165688,
"tokens_seen": 1699217408
},
{
"epoch": 0.51,
"learning_rate": 0.0002450551654964895,
"loss": 2.7564,
"theoretical_loss": 3.4780771376703745,
"tokens_seen": 1699282944
},
{
"epoch": 0.51,
"learning_rate": 0.0002450451354062187,
"loss": 2.7657,
"theoretical_loss": 3.4780655921941435,
"tokens_seen": 1699348480
},
{
"epoch": 0.51,
"learning_rate": 0.00024503510531594786,
"loss": 2.574,
"theoretical_loss": 3.4780540472878254,
"tokens_seen": 1699414016
},
{
"epoch": 0.51,
"learning_rate": 0.00024502507522567704,
"loss": 2.6681,
"theoretical_loss": 3.47804250295137,
"tokens_seen": 1699479552
},
{
"epoch": 0.52,
"learning_rate": 0.0002450150451354062,
"loss": 2.867,
"theoretical_loss": 3.478030959184728,
"tokens_seen": 1699545088
},
{
"epoch": 0.52,
"learning_rate": 0.0002450050150451354,
"loss": 2.7419,
"theoretical_loss": 3.4780194159878484,
"tokens_seen": 1699610624
},
{
"epoch": 0.52,
"learning_rate": 0.0002449949849548646,
"loss": 2.9158,
"theoretical_loss": 3.4780078733606814,
"tokens_seen": 1699676160
},
{
"epoch": 0.52,
"learning_rate": 0.0002449849548645938,
"loss": 2.5863,
"theoretical_loss": 3.477996331303177,
"tokens_seen": 1699741696
},
{
"epoch": 0.52,
"learning_rate": 0.000244974924774323,
"loss": 2.7245,
"theoretical_loss": 3.4779847898152854,
"tokens_seen": 1699807232
},
{
"epoch": 0.52,
"learning_rate": 0.0002449648946840522,
"loss": 2.5485,
"theoretical_loss": 3.477973248896956,
"tokens_seen": 1699872768
},
{
"epoch": 0.52,
"learning_rate": 0.00024495486459378136,
"loss": 2.4166,
"theoretical_loss": 3.4779617085481385,
"tokens_seen": 1699938304
},
{
"epoch": 0.52,
"learning_rate": 0.00024494483450351054,
"loss": 2.5625,
"theoretical_loss": 3.4779501687687837,
"tokens_seen": 1700003840
},
{
"epoch": 0.52,
"learning_rate": 0.0002449348044132397,
"loss": 2.6814,
"theoretical_loss": 3.477938629558841,
"tokens_seen": 1700069376
},
{
"epoch": 0.52,
"learning_rate": 0.0002449247743229689,
"loss": 2.4446,
"theoretical_loss": 3.4779270909182607,
"tokens_seen": 1700134912
},
{
"epoch": 0.52,
"learning_rate": 0.0002449147442326981,
"loss": 2.8038,
"theoretical_loss": 3.4779155528469925,
"tokens_seen": 1700200448
},
{
"epoch": 0.52,
"learning_rate": 0.00024490471414242727,
"loss": 2.6133,
"theoretical_loss": 3.477904015344986,
"tokens_seen": 1700265984
},
{
"epoch": 0.52,
"learning_rate": 0.0002448946840521565,
"loss": 2.5984,
"theoretical_loss": 3.4778924784121914,
"tokens_seen": 1700331520
},
{
"epoch": 0.52,
"learning_rate": 0.0002448846539618857,
"loss": 2.7032,
"theoretical_loss": 3.47788094204856,
"tokens_seen": 1700397056
},
{
"epoch": 0.52,
"learning_rate": 0.00024487462387161486,
"loss": 2.6714,
"theoretical_loss": 3.4778694062540394,
"tokens_seen": 1700462592
},
{
"epoch": 0.52,
"learning_rate": 0.00024486459378134405,
"loss": 2.6245,
"theoretical_loss": 3.4778578710285815,
"tokens_seen": 1700528128
},
{
"epoch": 0.52,
"learning_rate": 0.0002448545636910732,
"loss": 2.8354,
"theoretical_loss": 3.477846336372136,
"tokens_seen": 1700593664
},
{
"epoch": 0.52,
"objective/train/docs_used": 963811,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.716780424118042,
"objective/train/theoretical_loss": 3.4778348022846517,
"objective/train/tokens_used": 1721119200,
"theoretical_loss": 3.4778348022846517,
"tokens_seen": 1700659200
},
{
"epoch": 0.52,
"learning_rate": 0.0002448445336008024,
"loss": 2.7086,
"theoretical_loss": 3.4778348022846517,
"tokens_seen": 1700659200
},
{
"epoch": 0.52,
"learning_rate": 0.0002448345035105316,
"loss": 2.7536,
"theoretical_loss": 3.47782326876608,
"tokens_seen": 1700724736
},
{
"epoch": 0.52,
"learning_rate": 0.00024482447342026077,
"loss": 2.8421,
"theoretical_loss": 3.4778117358163705,
"tokens_seen": 1700790272
},
{
"epoch": 0.52,
"learning_rate": 0.00024481444332998995,
"loss": 2.7532,
"theoretical_loss": 3.4778002034354722,
"tokens_seen": 1700855808
},
{
"epoch": 0.52,
"learning_rate": 0.0002448044132397192,
"loss": 2.9043,
"theoretical_loss": 3.477788671623337,
"tokens_seen": 1700921344
},
{
"epoch": 0.52,
"learning_rate": 0.00024479438314944837,
"loss": 2.611,
"theoretical_loss": 3.4777771403799136,
"tokens_seen": 1700986880
},
{
"epoch": 0.52,
"learning_rate": 0.00024478435305917755,
"loss": 2.7456,
"theoretical_loss": 3.4777656097051524,
"tokens_seen": 1701052416
},
{
"epoch": 0.52,
"learning_rate": 0.00024477432296890673,
"loss": 2.5847,
"theoretical_loss": 3.4777540795990034,
"tokens_seen": 1701117952
},
{
"epoch": 0.52,
"learning_rate": 0.0002447642928786359,
"loss": 2.544,
"theoretical_loss": 3.477742550061417,
"tokens_seen": 1701183488
},
{
"epoch": 0.52,
"learning_rate": 0.0002447542627883651,
"loss": 2.6458,
"theoretical_loss": 3.4777310210923424,
"tokens_seen": 1701249024
},
{
"epoch": 0.52,
"learning_rate": 0.0002447442326980943,
"loss": 2.4067,
"theoretical_loss": 3.4777194926917305,
"tokens_seen": 1701314560
},
{
"epoch": 0.52,
"learning_rate": 0.00024473420260782345,
"loss": 2.8441,
"theoretical_loss": 3.4777079648595315,
"tokens_seen": 1701380096
},
{
"epoch": 0.52,
"learning_rate": 0.00024472417251755264,
"loss": 2.7678,
"theoretical_loss": 3.4776964375956947,
"tokens_seen": 1701445632
},
{
"epoch": 0.52,
"learning_rate": 0.00024471414242728187,
"loss": 2.641,
"theoretical_loss": 3.477684910900171,
"tokens_seen": 1701511168
},
{
"epoch": 0.52,
"learning_rate": 0.00024470411233701105,
"loss": 2.6133,
"theoretical_loss": 3.4776733847729098,
"tokens_seen": 1701576704
},
{
"epoch": 0.52,
"learning_rate": 0.00024469408224674023,
"loss": 2.4422,
"theoretical_loss": 3.4776618592138613,
"tokens_seen": 1701642240
},
{
"epoch": 0.52,
"learning_rate": 0.0002446840521564694,
"loss": 2.6135,
"theoretical_loss": 3.477650334222976,
"tokens_seen": 1701707776
},
{
"epoch": 0.52,
"learning_rate": 0.0002446740220661986,
"loss": 2.8614,
"theoretical_loss": 3.477638809800204,
"tokens_seen": 1701773312
},
{
"epoch": 0.52,
"learning_rate": 0.0002446639919759278,
"loss": 2.6162,
"theoretical_loss": 3.477627285945495,
"tokens_seen": 1701838848
},
{
"epoch": 0.52,
"learning_rate": 0.00024465396188565696,
"loss": 2.8307,
"theoretical_loss": 3.4776157626587993,
"tokens_seen": 1701904384
},
{
"epoch": 0.52,
"learning_rate": 0.00024464393179538614,
"loss": 2.6459,
"theoretical_loss": 3.4776042399400673,
"tokens_seen": 1701969920
},
{
"epoch": 0.52,
"learning_rate": 0.0002446339017051154,
"loss": 2.6029,
"theoretical_loss": 3.477592717789248,
"tokens_seen": 1702035456
},
{
"epoch": 0.52,
"learning_rate": 0.00024462387161484456,
"loss": 2.7246,
"theoretical_loss": 3.4775811962062937,
"tokens_seen": 1702100992
},
{
"epoch": 0.52,
"learning_rate": 0.00024461384152457374,
"loss": 2.6934,
"theoretical_loss": 3.4775696751911527,
"tokens_seen": 1702166528
},
{
"epoch": 0.52,
"learning_rate": 0.0002446038114343029,
"loss": 2.8219,
"theoretical_loss": 3.4775581547437757,
"tokens_seen": 1702232064
},
{
"epoch": 0.52,
"objective/train/docs_used": 965136,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3761980533599854,
"objective/train/theoretical_loss": 3.477546634864113,
"objective/train/tokens_used": 1722757600,
"theoretical_loss": 3.477546634864113,
"tokens_seen": 1702297600
},
{
"epoch": 0.52,
"learning_rate": 0.0002445937813440321,
"loss": 2.477,
"theoretical_loss": 3.477546634864113,
"tokens_seen": 1702297600
},
{
"epoch": 0.52,
"learning_rate": 0.0002445837512537613,
"loss": 2.6795,
"theoretical_loss": 3.477535115552115,
"tokens_seen": 1702363136
},
{
"epoch": 0.52,
"learning_rate": 0.00024457372116349046,
"loss": 2.8251,
"theoretical_loss": 3.4775235968077314,
"tokens_seen": 1702428672
},
{
"epoch": 0.52,
"learning_rate": 0.00024456369107321964,
"loss": 2.5046,
"theoretical_loss": 3.4775120786309124,
"tokens_seen": 1702494208
},
{
"epoch": 0.52,
"learning_rate": 0.0002445536609829488,
"loss": 2.6341,
"theoretical_loss": 3.4775005610216088,
"tokens_seen": 1702559744
},
{
"epoch": 0.52,
"learning_rate": 0.00024454363089267806,
"loss": 2.5919,
"theoretical_loss": 3.4774890439797694,
"tokens_seen": 1702625280
},
{
"epoch": 0.52,
"learning_rate": 0.00024453360080240724,
"loss": 2.6681,
"theoretical_loss": 3.4774775275053464,
"tokens_seen": 1702690816
},
{
"epoch": 0.52,
"learning_rate": 0.0002445235707121364,
"loss": 2.6202,
"theoretical_loss": 3.4774660115982883,
"tokens_seen": 1702756352
},
{
"epoch": 0.52,
"learning_rate": 0.0002445135406218656,
"loss": 2.6551,
"theoretical_loss": 3.477454496258546,
"tokens_seen": 1702821888
},
{
"epoch": 0.52,
"learning_rate": 0.0002445035105315948,
"loss": 2.5657,
"theoretical_loss": 3.47744298148607,
"tokens_seen": 1702887424
},
{
"epoch": 0.52,
"learning_rate": 0.00024449348044132396,
"loss": 2.7497,
"theoretical_loss": 3.4774314672808098,
"tokens_seen": 1702952960
},
{
"epoch": 0.52,
"learning_rate": 0.00024448345035105315,
"loss": 2.5205,
"theoretical_loss": 3.477419953642716,
"tokens_seen": 1703018496
},
{
"epoch": 0.52,
"learning_rate": 0.0002444734202607823,
"loss": 2.6631,
"theoretical_loss": 3.477408440571739,
"tokens_seen": 1703084032
},
{
"epoch": 0.52,
"learning_rate": 0.0002444633901705115,
"loss": 2.5832,
"theoretical_loss": 3.4773969280678294,
"tokens_seen": 1703149568
},
{
"epoch": 0.52,
"learning_rate": 0.00024445336008024074,
"loss": 2.5683,
"theoretical_loss": 3.4773854161309363,
"tokens_seen": 1703215104
},
{
"epoch": 0.52,
"learning_rate": 0.0002444433299899699,
"loss": 2.5714,
"theoretical_loss": 3.477373904761011,
"tokens_seen": 1703280640
},
{
"epoch": 0.52,
"learning_rate": 0.0002444332998996991,
"loss": 2.7385,
"theoretical_loss": 3.4773623939580034,
"tokens_seen": 1703346176
},
{
"epoch": 0.52,
"learning_rate": 0.0002444232698094283,
"loss": 2.7973,
"theoretical_loss": 3.477350883721863,
"tokens_seen": 1703411712
},
{
"epoch": 0.52,
"learning_rate": 0.00024441323971915747,
"loss": 2.9448,
"theoretical_loss": 3.477339374052542,
"tokens_seen": 1703477248
},
{
"epoch": 0.52,
"learning_rate": 0.00024440320962888665,
"loss": 2.6859,
"theoretical_loss": 3.477327864949989,
"tokens_seen": 1703542784
},
{
"epoch": 0.52,
"learning_rate": 0.00024439317953861583,
"loss": 2.8656,
"theoretical_loss": 3.4773163564141547,
"tokens_seen": 1703608320
},
{
"epoch": 0.52,
"learning_rate": 0.000244383149448345,
"loss": 2.7107,
"theoretical_loss": 3.4773048484449895,
"tokens_seen": 1703673856
},
{
"epoch": 0.52,
"learning_rate": 0.0002443731193580742,
"loss": 2.7467,
"theoretical_loss": 3.4772933410424436,
"tokens_seen": 1703739392
},
{
"epoch": 0.52,
"learning_rate": 0.0002443630892678034,
"loss": 2.6143,
"theoretical_loss": 3.477281834206468,
"tokens_seen": 1703804928
},
{
"epoch": 0.52,
"learning_rate": 0.0002443530591775326,
"loss": 2.663,
"theoretical_loss": 3.4772703279370116,
"tokens_seen": 1703870464
},
{
"epoch": 0.52,
"objective/train/docs_used": 965825,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5462608337402344,
"objective/train/theoretical_loss": 3.477258822234026,
"objective/train/tokens_used": 1724396000,
"theoretical_loss": 3.477258822234026,
"tokens_seen": 1703936000
},
{
"epoch": 0.52,
"learning_rate": 0.0002443430290872618,
"loss": 2.4402,
"theoretical_loss": 3.477258822234026,
"tokens_seen": 1703936000
},
{
"epoch": 0.52,
"learning_rate": 0.00024433299899699097,
"loss": 2.714,
"theoretical_loss": 3.4772473170974614,
"tokens_seen": 1704001536
},
{
"epoch": 0.52,
"learning_rate": 0.00024432296890672015,
"loss": 2.5284,
"theoretical_loss": 3.477235812527267,
"tokens_seen": 1704067072
},
{
"epoch": 0.52,
"learning_rate": 0.0002443129388164494,
"loss": 2.6859,
"theoretical_loss": 3.4772243085233945,
"tokens_seen": 1704132608
},
{
"epoch": 0.52,
"learning_rate": 0.00024430290872617857,
"loss": 2.7001,
"theoretical_loss": 3.4772128050857933,
"tokens_seen": 1704198144
},
{
"epoch": 0.52,
"learning_rate": 0.00024429287863590775,
"loss": 2.7035,
"theoretical_loss": 3.4772013022144144,
"tokens_seen": 1704263680
},
{
"epoch": 0.52,
"learning_rate": 0.00024428284854563693,
"loss": 2.6098,
"theoretical_loss": 3.4771897999092083,
"tokens_seen": 1704329216
},
{
"epoch": 0.52,
"learning_rate": 0.0002442728184553661,
"loss": 2.5166,
"theoretical_loss": 3.4771782981701245,
"tokens_seen": 1704394752
},
{
"epoch": 0.52,
"learning_rate": 0.0002442627883650953,
"loss": 2.8258,
"theoretical_loss": 3.4771667969971136,
"tokens_seen": 1704460288
},
{
"epoch": 0.52,
"learning_rate": 0.0002442527582748245,
"loss": 2.7535,
"theoretical_loss": 3.4771552963901264,
"tokens_seen": 1704525824
},
{
"epoch": 0.52,
"learning_rate": 0.00024424272818455365,
"loss": 2.7633,
"theoretical_loss": 3.477143796349113,
"tokens_seen": 1704591360
},
{
"epoch": 0.52,
"learning_rate": 0.00024423269809428284,
"loss": 2.6663,
"theoretical_loss": 3.477132296874024,
"tokens_seen": 1704656896
},
{
"epoch": 0.52,
"learning_rate": 0.00024422266800401207,
"loss": 2.6649,
"theoretical_loss": 3.47712079796481,
"tokens_seen": 1704722432
},
{
"epoch": 0.52,
"learning_rate": 0.00024421263791374125,
"loss": 2.6405,
"theoretical_loss": 3.4771092996214206,
"tokens_seen": 1704787968
},
{
"epoch": 0.52,
"learning_rate": 0.00024420260782347043,
"loss": 2.7841,
"theoretical_loss": 3.4770978018438066,
"tokens_seen": 1704853504
},
{
"epoch": 0.52,
"learning_rate": 0.0002441925777331996,
"loss": 2.7275,
"theoretical_loss": 3.4770863046319187,
"tokens_seen": 1704919040
},
{
"epoch": 0.52,
"learning_rate": 0.0002441825476429288,
"loss": 2.7383,
"theoretical_loss": 3.4770748079857072,
"tokens_seen": 1704984576
},
{
"epoch": 0.52,
"learning_rate": 0.000244172517552658,
"loss": 2.6688,
"theoretical_loss": 3.4770633119051224,
"tokens_seen": 1705050112
},
{
"epoch": 0.52,
"learning_rate": 0.00024416248746238716,
"loss": 2.6543,
"theoretical_loss": 3.4770518163901145,
"tokens_seen": 1705115648
},
{
"epoch": 0.52,
"learning_rate": 0.00024415245737211634,
"loss": 2.5527,
"theoretical_loss": 3.4770403214406347,
"tokens_seen": 1705181184
},
{
"epoch": 0.52,
"learning_rate": 0.0002441424272818456,
"loss": 2.3991,
"theoretical_loss": 3.4770288270566327,
"tokens_seen": 1705246720
},
{
"epoch": 0.52,
"learning_rate": 0.00024413239719157473,
"loss": 2.5828,
"theoretical_loss": 3.477017333238059,
"tokens_seen": 1705312256
},
{
"epoch": 0.52,
"learning_rate": 0.0002441223671013039,
"loss": 2.9272,
"theoretical_loss": 3.4770058399848645,
"tokens_seen": 1705377792
},
{
"epoch": 0.52,
"learning_rate": 0.00024411233701103312,
"loss": 2.9833,
"theoretical_loss": 3.476994347296999,
"tokens_seen": 1705443328
},
{
"epoch": 0.52,
"learning_rate": 0.0002441023069207623,
"loss": 2.5042,
"theoretical_loss": 3.476982855174414,
"tokens_seen": 1705508864
},
{
"epoch": 0.52,
"objective/train/docs_used": 966739,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3506920337677,
"objective/train/theoretical_loss": 3.476971363617059,
"objective/train/tokens_used": 1726034400,
"theoretical_loss": 3.476971363617059,
"tokens_seen": 1705574400
},
{
"epoch": 0.52,
"learning_rate": 0.00024409227683049148,
"loss": 2.6765,
"theoretical_loss": 3.476971363617059,
"tokens_seen": 1705574400
},
{
"epoch": 0.52,
"learning_rate": 0.00024408224674022066,
"loss": 2.5925,
"theoretical_loss": 3.476959872624885,
"tokens_seen": 1705639936
},
{
"epoch": 0.52,
"learning_rate": 0.00024407221664994987,
"loss": 2.7941,
"theoretical_loss": 3.4769483821978424,
"tokens_seen": 1705705472
},
{
"epoch": 0.52,
"learning_rate": 0.00024406218655967905,
"loss": 2.7736,
"theoretical_loss": 3.4769368923358814,
"tokens_seen": 1705771008
},
{
"epoch": 0.52,
"learning_rate": 0.00024405215646940823,
"loss": 2.4367,
"theoretical_loss": 3.476925403038953,
"tokens_seen": 1705836544
},
{
"epoch": 0.52,
"learning_rate": 0.0002440421263791374,
"loss": 2.4842,
"theoretical_loss": 3.4769139143070076,
"tokens_seen": 1705902080
},
{
"epoch": 0.52,
"learning_rate": 0.0002440320962888666,
"loss": 2.6787,
"theoretical_loss": 3.476902426139995,
"tokens_seen": 1705967616
},
{
"epoch": 0.52,
"learning_rate": 0.0002440220661985958,
"loss": 2.7628,
"theoretical_loss": 3.4768909385378666,
"tokens_seen": 1706033152
},
{
"epoch": 0.52,
"learning_rate": 0.00024401203610832498,
"loss": 2.4361,
"theoretical_loss": 3.4768794515005723,
"tokens_seen": 1706098688
},
{
"epoch": 0.52,
"learning_rate": 0.00024400200601805416,
"loss": 2.4319,
"theoretical_loss": 3.476867965028063,
"tokens_seen": 1706164224
},
{
"epoch": 0.52,
"learning_rate": 0.00024399197592778335,
"loss": 2.6707,
"theoretical_loss": 3.47685647912029,
"tokens_seen": 1706229760
},
{
"epoch": 0.52,
"learning_rate": 0.00024398194583751255,
"loss": 2.635,
"theoretical_loss": 3.476844993777202,
"tokens_seen": 1706295296
},
{
"epoch": 0.52,
"learning_rate": 0.00024397191574724173,
"loss": 2.3553,
"theoretical_loss": 3.476833508998751,
"tokens_seen": 1706360832
},
{
"epoch": 0.52,
"learning_rate": 0.00024396188565697092,
"loss": 2.5933,
"theoretical_loss": 3.4768220247848873,
"tokens_seen": 1706426368
},
{
"epoch": 0.52,
"learning_rate": 0.0002439518555667001,
"loss": 2.6862,
"theoretical_loss": 3.4768105411355608,
"tokens_seen": 1706491904
},
{
"epoch": 0.52,
"learning_rate": 0.00024394182547642928,
"loss": 2.6717,
"theoretical_loss": 3.476799058050723,
"tokens_seen": 1706557440
},
{
"epoch": 0.52,
"learning_rate": 0.00024393179538615849,
"loss": 2.5632,
"theoretical_loss": 3.4767875755303237,
"tokens_seen": 1706622976
},
{
"epoch": 0.52,
"learning_rate": 0.00024392176529588767,
"loss": 2.7323,
"theoretical_loss": 3.476776093574314,
"tokens_seen": 1706688512
},
{
"epoch": 0.52,
"learning_rate": 0.00024391173520561685,
"loss": 2.7355,
"theoretical_loss": 3.4767646121826443,
"tokens_seen": 1706754048
},
{
"epoch": 0.52,
"learning_rate": 0.00024390170511534603,
"loss": 2.6798,
"theoretical_loss": 3.4767531313552658,
"tokens_seen": 1706819584
},
{
"epoch": 0.52,
"learning_rate": 0.00024389167502507524,
"loss": 2.76,
"theoretical_loss": 3.4767416510921274,
"tokens_seen": 1706885120
},
{
"epoch": 0.52,
"learning_rate": 0.00024388164493480442,
"loss": 2.6444,
"theoretical_loss": 3.4767301713931813,
"tokens_seen": 1706950656
},
{
"epoch": 0.52,
"learning_rate": 0.0002438716148445336,
"loss": 2.8393,
"theoretical_loss": 3.4767186922583777,
"tokens_seen": 1707016192
},
{
"epoch": 0.52,
"learning_rate": 0.00024386158475426278,
"loss": 2.599,
"theoretical_loss": 3.4767072136876673,
"tokens_seen": 1707081728
},
{
"epoch": 0.52,
"learning_rate": 0.00024385155466399196,
"loss": 2.5116,
"theoretical_loss": 3.4766957356810004,
"tokens_seen": 1707147264
},
{
"epoch": 0.52,
"objective/train/docs_used": 967356,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8089561462402344,
"objective/train/theoretical_loss": 3.4766842582383277,
"objective/train/tokens_used": 1727672800,
"theoretical_loss": 3.4766842582383277,
"tokens_seen": 1707212800
},
{
"epoch": 0.52,
"learning_rate": 0.00024384152457372117,
"loss": 2.5245,
"theoretical_loss": 3.4766842582383277,
"tokens_seen": 1707212800
},
{
"epoch": 0.52,
"learning_rate": 0.00024383149448345035,
"loss": 2.6062,
"theoretical_loss": 3.4766727813596,
"tokens_seen": 1707278336
},
{
"epoch": 0.52,
"learning_rate": 0.00024382146439317953,
"loss": 2.5092,
"theoretical_loss": 3.476661305044768,
"tokens_seen": 1707343872
},
{
"epoch": 0.52,
"learning_rate": 0.00024381143430290871,
"loss": 2.617,
"theoretical_loss": 3.476649829293782,
"tokens_seen": 1707409408
},
{
"epoch": 0.52,
"learning_rate": 0.00024380140421263792,
"loss": 2.4444,
"theoretical_loss": 3.4766383541065933,
"tokens_seen": 1707474944
},
{
"epoch": 0.52,
"learning_rate": 0.0002437913741223671,
"loss": 2.8173,
"theoretical_loss": 3.4766268794831516,
"tokens_seen": 1707540480
},
{
"epoch": 0.52,
"learning_rate": 0.00024378134403209628,
"loss": 2.6418,
"theoretical_loss": 3.4766154054234084,
"tokens_seen": 1707606016
},
{
"epoch": 0.52,
"learning_rate": 0.00024377131394182547,
"loss": 2.6007,
"theoretical_loss": 3.4766039319273143,
"tokens_seen": 1707671552
},
{
"epoch": 0.52,
"learning_rate": 0.00024376128385155467,
"loss": 2.6993,
"theoretical_loss": 3.4765924589948196,
"tokens_seen": 1707737088
},
{
"epoch": 0.52,
"learning_rate": 0.00024375125376128385,
"loss": 2.5475,
"theoretical_loss": 3.4765809866258754,
"tokens_seen": 1707802624
},
{
"epoch": 0.52,
"learning_rate": 0.00024374122367101304,
"loss": 2.5985,
"theoretical_loss": 3.476569514820432,
"tokens_seen": 1707868160
},
{
"epoch": 0.52,
"learning_rate": 0.00024373119358074222,
"loss": 2.7105,
"theoretical_loss": 3.4765580435784402,
"tokens_seen": 1707933696
},
{
"epoch": 0.52,
"learning_rate": 0.0002437211634904714,
"loss": 2.6176,
"theoretical_loss": 3.4765465728998506,
"tokens_seen": 1707999232
},
{
"epoch": 0.52,
"learning_rate": 0.0002437111334002006,
"loss": 2.6235,
"theoretical_loss": 3.476535102784614,
"tokens_seen": 1708064768
},
{
"epoch": 0.52,
"learning_rate": 0.0002437011033099298,
"loss": 2.7317,
"theoretical_loss": 3.4765236332326817,
"tokens_seen": 1708130304
},
{
"epoch": 0.52,
"learning_rate": 0.00024369107321965897,
"loss": 2.7417,
"theoretical_loss": 3.4765121642440038,
"tokens_seen": 1708195840
},
{
"epoch": 0.52,
"learning_rate": 0.00024368104312938815,
"loss": 2.698,
"theoretical_loss": 3.476500695818531,
"tokens_seen": 1708261376
},
{
"epoch": 0.52,
"learning_rate": 0.00024367101303911736,
"loss": 2.397,
"theoretical_loss": 3.476489227956214,
"tokens_seen": 1708326912
},
{
"epoch": 0.52,
"learning_rate": 0.00024366098294884654,
"loss": 2.8463,
"theoretical_loss": 3.476477760657004,
"tokens_seen": 1708392448
},
{
"epoch": 0.52,
"learning_rate": 0.00024365095285857572,
"loss": 2.6436,
"theoretical_loss": 3.4764662939208515,
"tokens_seen": 1708457984
},
{
"epoch": 0.52,
"learning_rate": 0.0002436409227683049,
"loss": 2.7107,
"theoretical_loss": 3.476454827747707,
"tokens_seen": 1708523520
},
{
"epoch": 0.52,
"learning_rate": 0.00024363089267803408,
"loss": 2.8514,
"theoretical_loss": 3.476443362137522,
"tokens_seen": 1708589056
},
{
"epoch": 0.52,
"learning_rate": 0.0002436208625877633,
"loss": 2.4671,
"theoretical_loss": 3.4764318970902464,
"tokens_seen": 1708654592
},
{
"epoch": 0.52,
"learning_rate": 0.0002436108324974925,
"loss": 2.6182,
"theoretical_loss": 3.4764204326058317,
"tokens_seen": 1708720128
},
{
"epoch": 0.52,
"learning_rate": 0.00024360080240722168,
"loss": 2.8502,
"theoretical_loss": 3.476408968684228,
"tokens_seen": 1708785664
},
{
"epoch": 0.52,
"objective/train/docs_used": 967795,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.30680251121521,
"objective/train/theoretical_loss": 3.4763975053253864,
"objective/train/tokens_used": 1729311200,
"theoretical_loss": 3.4763975053253864,
"tokens_seen": 1708851200
},
{
"epoch": 0.52,
"learning_rate": 0.00024359077231695086,
"loss": 2.5534,
"theoretical_loss": 3.4763975053253864,
"tokens_seen": 1708851200
},
{
"epoch": 0.52,
"learning_rate": 0.00024358074222668007,
"loss": 2.8091,
"theoretical_loss": 3.4763860425292576,
"tokens_seen": 1708916736
},
{
"epoch": 0.52,
"learning_rate": 0.00024357071213640925,
"loss": 2.6447,
"theoretical_loss": 3.4763745802957926,
"tokens_seen": 1708982272
},
{
"epoch": 0.52,
"learning_rate": 0.00024356068204613843,
"loss": 2.7347,
"theoretical_loss": 3.476363118624943,
"tokens_seen": 1709047808
},
{
"epoch": 0.52,
"learning_rate": 0.0002435506519558676,
"loss": 2.7968,
"theoretical_loss": 3.4763516575166573,
"tokens_seen": 1709113344
},
{
"epoch": 0.52,
"learning_rate": 0.0002435406218655968,
"loss": 2.7241,
"theoretical_loss": 3.476340196970889,
"tokens_seen": 1709178880
},
{
"epoch": 0.52,
"learning_rate": 0.000243530591775326,
"loss": 2.5166,
"theoretical_loss": 3.476328736987587,
"tokens_seen": 1709244416
},
{
"epoch": 0.52,
"learning_rate": 0.00024352056168505518,
"loss": 2.7361,
"theoretical_loss": 3.476317277566703,
"tokens_seen": 1709309952
},
{
"epoch": 0.52,
"learning_rate": 0.00024351053159478436,
"loss": 2.755,
"theoretical_loss": 3.4763058187081874,
"tokens_seen": 1709375488
},
{
"epoch": 0.52,
"learning_rate": 0.00024350050150451355,
"loss": 2.7992,
"theoretical_loss": 3.476294360411991,
"tokens_seen": 1709441024
},
{
"epoch": 0.52,
"learning_rate": 0.00024349047141424275,
"loss": 2.3473,
"theoretical_loss": 3.4762829026780655,
"tokens_seen": 1709506560
},
{
"epoch": 0.52,
"learning_rate": 0.00024348044132397193,
"loss": 2.7117,
"theoretical_loss": 3.4762714455063612,
"tokens_seen": 1709572096
},
{
"epoch": 0.52,
"learning_rate": 0.00024347041123370112,
"loss": 2.8753,
"theoretical_loss": 3.4762599888968286,
"tokens_seen": 1709637632
},
{
"epoch": 0.52,
"learning_rate": 0.0002434603811434303,
"loss": 2.7423,
"theoretical_loss": 3.476248532849419,
"tokens_seen": 1709703168
},
{
"epoch": 0.52,
"learning_rate": 0.00024345035105315948,
"loss": 2.8425,
"theoretical_loss": 3.4762370773640834,
"tokens_seen": 1709768704
},
{
"epoch": 0.52,
"learning_rate": 0.00024344032096288869,
"loss": 2.5498,
"theoretical_loss": 3.476225622440772,
"tokens_seen": 1709834240
},
{
"epoch": 0.52,
"learning_rate": 0.00024343029087261787,
"loss": 2.5937,
"theoretical_loss": 3.4762141680794367,
"tokens_seen": 1709899776
},
{
"epoch": 0.52,
"learning_rate": 0.00024342026078234705,
"loss": 2.5858,
"theoretical_loss": 3.476202714280028,
"tokens_seen": 1709965312
},
{
"epoch": 0.52,
"learning_rate": 0.00024341023069207623,
"loss": 2.702,
"theoretical_loss": 3.476191261042496,
"tokens_seen": 1710030848
},
{
"epoch": 0.52,
"learning_rate": 0.00024340020060180544,
"loss": 2.4435,
"theoretical_loss": 3.4761798083667923,
"tokens_seen": 1710096384
},
{
"epoch": 0.52,
"learning_rate": 0.00024339017051153462,
"loss": 2.5085,
"theoretical_loss": 3.476168356252868,
"tokens_seen": 1710161920
},
{
"epoch": 0.52,
"learning_rate": 0.0002433801404212638,
"loss": 2.5794,
"theoretical_loss": 3.476156904700674,
"tokens_seen": 1710227456
},
{
"epoch": 0.52,
"learning_rate": 0.00024337011033099298,
"loss": 2.6355,
"theoretical_loss": 3.476145453710161,
"tokens_seen": 1710292992
},
{
"epoch": 0.52,
"learning_rate": 0.00024336008024072216,
"loss": 2.7641,
"theoretical_loss": 3.4761340032812793,
"tokens_seen": 1710358528
},
{
"epoch": 0.52,
"learning_rate": 0.00024335005015045137,
"loss": 2.8767,
"theoretical_loss": 3.4761225534139806,
"tokens_seen": 1710424064
},
{
"epoch": 0.52,
"objective/train/docs_used": 968886,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3432180881500244,
"objective/train/theoretical_loss": 3.476111104108216,
"objective/train/tokens_used": 1730949600,
"theoretical_loss": 3.476111104108216,
"tokens_seen": 1710489600
},
{
"epoch": 0.52,
"learning_rate": 0.00024334002006018055,
"loss": 2.9162,
"theoretical_loss": 3.476111104108216,
"tokens_seen": 1710489600
},
{
"epoch": 0.52,
"learning_rate": 0.00024332998996990973,
"loss": 2.9115,
"theoretical_loss": 3.476099655363936,
"tokens_seen": 1710555136
},
{
"epoch": 0.52,
"learning_rate": 0.00024331995987963891,
"loss": 2.552,
"theoretical_loss": 3.476088207181092,
"tokens_seen": 1710620672
},
{
"epoch": 0.52,
"learning_rate": 0.00024330992978936812,
"loss": 2.6301,
"theoretical_loss": 3.4760767595596347,
"tokens_seen": 1710686208
},
{
"epoch": 0.52,
"learning_rate": 0.0002432998996990973,
"loss": 2.5096,
"theoretical_loss": 3.4760653124995144,
"tokens_seen": 1710751744
},
{
"epoch": 0.52,
"learning_rate": 0.00024328986960882648,
"loss": 2.8855,
"theoretical_loss": 3.4760538660006834,
"tokens_seen": 1710817280
},
{
"epoch": 0.52,
"learning_rate": 0.00024327983951855567,
"loss": 2.7087,
"theoretical_loss": 3.4760424200630915,
"tokens_seen": 1710882816
},
{
"epoch": 0.52,
"learning_rate": 0.00024326980942828487,
"loss": 2.4159,
"theoretical_loss": 3.4760309746866906,
"tokens_seen": 1710948352
},
{
"epoch": 0.52,
"learning_rate": 0.00024325977933801405,
"loss": 2.7234,
"theoretical_loss": 3.476019529871431,
"tokens_seen": 1711013888
},
{
"epoch": 0.52,
"learning_rate": 0.00024324974924774324,
"loss": 2.6536,
"theoretical_loss": 3.4760080856172637,
"tokens_seen": 1711079424
},
{
"epoch": 0.52,
"learning_rate": 0.00024323971915747242,
"loss": 2.5838,
"theoretical_loss": 3.475996641924141,
"tokens_seen": 1711144960
},
{
"epoch": 0.52,
"learning_rate": 0.0002432296890672016,
"loss": 2.334,
"theoretical_loss": 3.475985198792012,
"tokens_seen": 1711210496
},
{
"epoch": 0.52,
"learning_rate": 0.0002432196589769308,
"loss": 2.7035,
"theoretical_loss": 3.4759737562208284,
"tokens_seen": 1711276032
},
{
"epoch": 0.52,
"learning_rate": 0.00024320962888666,
"loss": 2.4036,
"theoretical_loss": 3.4759623142105416,
"tokens_seen": 1711341568
},
{
"epoch": 0.52,
"learning_rate": 0.00024319959879638917,
"loss": 2.8487,
"theoretical_loss": 3.4759508727611026,
"tokens_seen": 1711407104
},
{
"epoch": 0.52,
"learning_rate": 0.00024318956870611835,
"loss": 2.8786,
"theoretical_loss": 3.475939431872462,
"tokens_seen": 1711472640
},
{
"epoch": 0.52,
"learning_rate": 0.00024317953861584756,
"loss": 2.5657,
"theoretical_loss": 3.4759279915445713,
"tokens_seen": 1711538176
},
{
"epoch": 0.52,
"learning_rate": 0.00024316950852557674,
"loss": 2.6136,
"theoretical_loss": 3.4759165517773813,
"tokens_seen": 1711603712
},
{
"epoch": 0.52,
"learning_rate": 0.00024315947843530592,
"loss": 2.5566,
"theoretical_loss": 3.4759051125708433,
"tokens_seen": 1711669248
},
{
"epoch": 0.52,
"learning_rate": 0.0002431494483450351,
"loss": 2.6157,
"theoretical_loss": 3.475893673924908,
"tokens_seen": 1711734784
},
{
"epoch": 0.52,
"learning_rate": 0.00024313941825476428,
"loss": 2.6476,
"theoretical_loss": 3.475882235839527,
"tokens_seen": 1711800320
},
{
"epoch": 0.52,
"learning_rate": 0.0002431293881644935,
"loss": 2.683,
"theoretical_loss": 3.4758707983146504,
"tokens_seen": 1711865856
},
{
"epoch": 0.52,
"learning_rate": 0.00024311935807422267,
"loss": 2.5054,
"theoretical_loss": 3.47585936135023,
"tokens_seen": 1711931392
},
{
"epoch": 0.52,
"learning_rate": 0.00024310932798395185,
"loss": 2.6697,
"theoretical_loss": 3.475847924946217,
"tokens_seen": 1711996928
},
{
"epoch": 0.52,
"learning_rate": 0.00024309929789368103,
"loss": 2.5059,
"theoretical_loss": 3.475836489102562,
"tokens_seen": 1712062464
},
{
"epoch": 0.52,
"objective/train/docs_used": 969614,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.838115692138672,
"objective/train/theoretical_loss": 3.4758250538192166,
"objective/train/tokens_used": 1732588000,
"theoretical_loss": 3.4758250538192166,
"tokens_seen": 1712128000
},
{
"epoch": 0.52,
"learning_rate": 0.00024308926780341024,
"loss": 2.7687,
"theoretical_loss": 3.4758250538192166,
"tokens_seen": 1712128000
},
{
"epoch": 0.52,
"learning_rate": 0.00024307923771313942,
"loss": 2.5316,
"theoretical_loss": 3.4758136190961313,
"tokens_seen": 1712193536
},
{
"epoch": 0.52,
"learning_rate": 0.0002430692076228686,
"loss": 2.6436,
"theoretical_loss": 3.475802184933258,
"tokens_seen": 1712259072
},
{
"epoch": 0.52,
"learning_rate": 0.00024305917753259779,
"loss": 2.705,
"theoretical_loss": 3.475790751330547,
"tokens_seen": 1712324608
},
{
"epoch": 0.52,
"learning_rate": 0.00024304914744232697,
"loss": 2.7353,
"theoretical_loss": 3.47577931828795,
"tokens_seen": 1712390144
},
{
"epoch": 0.52,
"learning_rate": 0.00024303911735205618,
"loss": 2.5359,
"theoretical_loss": 3.4757678858054177,
"tokens_seen": 1712455680
},
{
"epoch": 0.52,
"learning_rate": 0.00024302908726178536,
"loss": 2.7767,
"theoretical_loss": 3.4757564538829016,
"tokens_seen": 1712521216
},
{
"epoch": 0.52,
"learning_rate": 0.00024301905717151454,
"loss": 2.4807,
"theoretical_loss": 3.4757450225203526,
"tokens_seen": 1712586752
},
{
"epoch": 0.52,
"learning_rate": 0.00024300902708124372,
"loss": 2.3731,
"theoretical_loss": 3.475733591717722,
"tokens_seen": 1712652288
},
{
"epoch": 0.52,
"learning_rate": 0.00024299899699097293,
"loss": 3.0654,
"theoretical_loss": 3.4757221614749607,
"tokens_seen": 1712717824
},
{
"epoch": 0.52,
"learning_rate": 0.0002429889669007021,
"loss": 2.6095,
"theoretical_loss": 3.47571073179202,
"tokens_seen": 1712783360
},
{
"epoch": 0.52,
"learning_rate": 0.0002429789368104313,
"loss": 2.7297,
"theoretical_loss": 3.4756993026688514,
"tokens_seen": 1712848896
},
{
"epoch": 0.52,
"learning_rate": 0.00024296890672016047,
"loss": 2.5431,
"theoretical_loss": 3.475687874105406,
"tokens_seen": 1712914432
},
{
"epoch": 0.52,
"learning_rate": 0.00024295887662988965,
"loss": 2.6068,
"theoretical_loss": 3.4756764461016347,
"tokens_seen": 1712979968
},
{
"epoch": 0.52,
"learning_rate": 0.00024294884653961886,
"loss": 2.819,
"theoretical_loss": 3.4756650186574882,
"tokens_seen": 1713045504
},
{
"epoch": 0.52,
"learning_rate": 0.00024293881644934804,
"loss": 2.6397,
"theoretical_loss": 3.475653591772918,
"tokens_seen": 1713111040
},
{
"epoch": 0.52,
"learning_rate": 0.00024292878635907722,
"loss": 2.5004,
"theoretical_loss": 3.4756421654478764,
"tokens_seen": 1713176576
},
{
"epoch": 0.52,
"learning_rate": 0.0002429187562688064,
"loss": 2.7639,
"theoretical_loss": 3.475630739682313,
"tokens_seen": 1713242112
},
{
"epoch": 0.52,
"learning_rate": 0.0002429087261785356,
"loss": 2.6314,
"theoretical_loss": 3.47561931447618,
"tokens_seen": 1713307648
},
{
"epoch": 0.52,
"learning_rate": 0.0002428986960882648,
"loss": 2.6906,
"theoretical_loss": 3.475607889829428,
"tokens_seen": 1713373184
},
{
"epoch": 0.52,
"learning_rate": 0.00024288866599799397,
"loss": 2.7289,
"theoretical_loss": 3.475596465742009,
"tokens_seen": 1713438720
},
{
"epoch": 0.52,
"learning_rate": 0.00024287863590772315,
"loss": 2.7702,
"theoretical_loss": 3.475585042213874,
"tokens_seen": 1713504256
},
{
"epoch": 0.52,
"learning_rate": 0.00024286860581745234,
"loss": 2.496,
"theoretical_loss": 3.4755736192449733,
"tokens_seen": 1713569792
},
{
"epoch": 0.52,
"learning_rate": 0.00024285857572718157,
"loss": 2.7228,
"theoretical_loss": 3.4755621968352592,
"tokens_seen": 1713635328
},
{
"epoch": 0.52,
"learning_rate": 0.00024284854563691075,
"loss": 2.6308,
"theoretical_loss": 3.475550774984682,
"tokens_seen": 1713700864
},
{
"epoch": 0.52,
"objective/train/docs_used": 971016,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.490187883377075,
"objective/train/theoretical_loss": 3.4755393536931942,
"objective/train/tokens_used": 1734226400,
"theoretical_loss": 3.4755393536931942,
"tokens_seen": 1713766400
},
{
"epoch": 0.52,
"learning_rate": 0.00024283851554663993,
"loss": 2.7973,
"theoretical_loss": 3.4755393536931942,
"tokens_seen": 1713766400
},
{
"epoch": 0.52,
"learning_rate": 0.00024282848545636911,
"loss": 2.7912,
"theoretical_loss": 3.4755279329607465,
"tokens_seen": 1713831936
},
{
"epoch": 0.52,
"learning_rate": 0.00024281845536609832,
"loss": 2.783,
"theoretical_loss": 3.4755165127872893,
"tokens_seen": 1713897472
},
{
"epoch": 0.52,
"learning_rate": 0.0002428084252758275,
"loss": 2.7046,
"theoretical_loss": 3.4755050931727753,
"tokens_seen": 1713963008
},
{
"epoch": 0.52,
"learning_rate": 0.00024279839518555668,
"loss": 2.462,
"theoretical_loss": 3.4754936741171543,
"tokens_seen": 1714028544
},
{
"epoch": 0.52,
"learning_rate": 0.00024278836509528587,
"loss": 2.6709,
"theoretical_loss": 3.475482255620379,
"tokens_seen": 1714094080
},
{
"epoch": 0.52,
"learning_rate": 0.00024277833500501507,
"loss": 2.8093,
"theoretical_loss": 3.4754708376823995,
"tokens_seen": 1714159616
},
{
"epoch": 0.52,
"learning_rate": 0.00024276830491474426,
"loss": 2.5175,
"theoretical_loss": 3.475459420303168,
"tokens_seen": 1714225152
},
{
"epoch": 0.52,
"learning_rate": 0.00024275827482447344,
"loss": 2.646,
"theoretical_loss": 3.4754480034826356,
"tokens_seen": 1714290688
},
{
"epoch": 0.52,
"learning_rate": 0.00024274824473420262,
"loss": 2.611,
"theoretical_loss": 3.475436587220753,
"tokens_seen": 1714356224
},
{
"epoch": 0.52,
"learning_rate": 0.0002427382146439318,
"loss": 2.6447,
"theoretical_loss": 3.4754251715174727,
"tokens_seen": 1714421760
},
{
"epoch": 0.52,
"learning_rate": 0.000242728184553661,
"loss": 2.7469,
"theoretical_loss": 3.4754137563727445,
"tokens_seen": 1714487296
},
{
"epoch": 0.52,
"learning_rate": 0.000242728184553661,
"loss": 2.5922,
"theoretical_loss": 3.4754023417865207,
"tokens_seen": 1714552832
},
{
"epoch": 0.52,
"learning_rate": 0.0002427181544633902,
"loss": 2.6147,
"theoretical_loss": 3.4753909277587525,
"tokens_seen": 1714618368
},
{
"epoch": 0.52,
"learning_rate": 0.00024270812437311937,
"loss": 2.5997,
"theoretical_loss": 3.475379514289391,
"tokens_seen": 1714683904
},
{
"epoch": 0.52,
"learning_rate": 0.00024269809428284855,
"loss": 2.7215,
"theoretical_loss": 3.475368101378388,
"tokens_seen": 1714749440
},
{
"epoch": 0.52,
"learning_rate": 0.00024268806419257776,
"loss": 2.5948,
"theoretical_loss": 3.475356689025694,
"tokens_seen": 1714814976
},
{
"epoch": 0.52,
"learning_rate": 0.00024267803410230694,
"loss": 2.37,
"theoretical_loss": 3.4753452772312614,
"tokens_seen": 1714880512
},
{
"epoch": 0.52,
"learning_rate": 0.00024266800401203612,
"loss": 2.7453,
"theoretical_loss": 3.475333865995041,
"tokens_seen": 1714946048
},
{
"epoch": 0.52,
"learning_rate": 0.0002426579739217653,
"loss": 2.7854,
"theoretical_loss": 3.4753224553169835,
"tokens_seen": 1715011584
},
{
"epoch": 0.52,
"learning_rate": 0.00024264794383149448,
"loss": 2.6236,
"theoretical_loss": 3.4753110451970417,
"tokens_seen": 1715077120
},
{
"epoch": 0.52,
"learning_rate": 0.0002426379137412237,
"loss": 2.542,
"theoretical_loss": 3.475299635635166,
"tokens_seen": 1715142656
},
{
"epoch": 0.52,
"learning_rate": 0.00024262788365095287,
"loss": 2.609,
"theoretical_loss": 3.475288226631308,
"tokens_seen": 1715208192
},
{
"epoch": 0.52,
"learning_rate": 0.00024261785356068205,
"loss": 2.629,
"theoretical_loss": 3.475276818185419,
"tokens_seen": 1715273728
},
{
"epoch": 0.52,
"learning_rate": 0.00024260782347041123,
"loss": 2.6195,
"theoretical_loss": 3.475265410297451,
"tokens_seen": 1715339264
},
{
"epoch": 0.52,
"objective/train/docs_used": 971702,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.9550857543945312,
"objective/train/theoretical_loss": 3.4752540029673544,
"objective/train/tokens_used": 1735864800,
"theoretical_loss": 3.4752540029673544,
"tokens_seen": 1715404800
},
{
"epoch": 0.52,
"learning_rate": 0.00024259779338014044,
"loss": 2.3281,
"theoretical_loss": 3.4752540029673544,
"tokens_seen": 1715404800
},
{
"epoch": 0.52,
"learning_rate": 0.00024258776328986962,
"loss": 2.8496,
"theoretical_loss": 3.4752425961950815,
"tokens_seen": 1715470336
},
{
"epoch": 0.52,
"learning_rate": 0.0002425777331995988,
"loss": 2.6251,
"theoretical_loss": 3.4752311899805832,
"tokens_seen": 1715535872
},
{
"epoch": 0.52,
"learning_rate": 0.00024256770310932799,
"loss": 2.7034,
"theoretical_loss": 3.475219784323811,
"tokens_seen": 1715601408
},
{
"epoch": 0.52,
"learning_rate": 0.00024255767301905717,
"loss": 2.5147,
"theoretical_loss": 3.475208379224716,
"tokens_seen": 1715666944
},
{
"epoch": 0.52,
"learning_rate": 0.00024254764292878638,
"loss": 2.6617,
"theoretical_loss": 3.475196974683251,
"tokens_seen": 1715732480
},
{
"epoch": 0.52,
"learning_rate": 0.00024253761283851556,
"loss": 2.5541,
"theoretical_loss": 3.4751855706993657,
"tokens_seen": 1715798016
},
{
"epoch": 0.52,
"learning_rate": 0.00024252758274824474,
"loss": 2.484,
"theoretical_loss": 3.4751741672730123,
"tokens_seen": 1715863552
},
{
"epoch": 0.52,
"learning_rate": 0.00024251755265797392,
"loss": 2.6867,
"theoretical_loss": 3.4751627644041427,
"tokens_seen": 1715929088
},
{
"epoch": 0.52,
"learning_rate": 0.00024250752256770313,
"loss": 2.6565,
"theoretical_loss": 3.4751513620927073,
"tokens_seen": 1715994624
},
{
"epoch": 0.52,
"learning_rate": 0.0002424974924774323,
"loss": 2.5487,
"theoretical_loss": 3.475139960338659,
"tokens_seen": 1716060160
},
{
"epoch": 0.52,
"learning_rate": 0.0002424874623871615,
"loss": 2.6411,
"theoretical_loss": 3.4751285591419476,
"tokens_seen": 1716125696
},
{
"epoch": 0.52,
"learning_rate": 0.00024247743229689067,
"loss": 2.6879,
"theoretical_loss": 3.475117158502526,
"tokens_seen": 1716191232
},
{
"epoch": 0.52,
"learning_rate": 0.00024246740220661985,
"loss": 2.7014,
"theoretical_loss": 3.4751057584203444,
"tokens_seen": 1716256768
},
{
"epoch": 0.52,
"learning_rate": 0.00024245737211634906,
"loss": 2.5877,
"theoretical_loss": 3.4750943588953556,
"tokens_seen": 1716322304
},
{
"epoch": 0.52,
"learning_rate": 0.00024244734202607824,
"loss": 2.571,
"theoretical_loss": 3.47508295992751,
"tokens_seen": 1716387840
},
{
"epoch": 0.52,
"learning_rate": 0.00024243731193580742,
"loss": 2.6797,
"theoretical_loss": 3.47507156151676,
"tokens_seen": 1716453376
},
{
"epoch": 0.52,
"learning_rate": 0.0002424272818455366,
"loss": 2.4999,
"theoretical_loss": 3.4750601636630565,
"tokens_seen": 1716518912
},
{
"epoch": 0.52,
"learning_rate": 0.0002424172517552658,
"loss": 2.5819,
"theoretical_loss": 3.475048766366351,
"tokens_seen": 1716584448
},
{
"epoch": 0.52,
"learning_rate": 0.000242407221664995,
"loss": 2.5335,
"theoretical_loss": 3.4750373696265955,
"tokens_seen": 1716649984
},
{
"epoch": 0.52,
"learning_rate": 0.00024239719157472417,
"loss": 2.7512,
"theoretical_loss": 3.4750259734437408,
"tokens_seen": 1716715520
},
{
"epoch": 0.52,
"learning_rate": 0.00024238716148445335,
"loss": 2.5219,
"theoretical_loss": 3.4750145778177393,
"tokens_seen": 1716781056
},
{
"epoch": 0.52,
"learning_rate": 0.00024237713139418254,
"loss": 2.7014,
"theoretical_loss": 3.4750031827485417,
"tokens_seen": 1716846592
},
{
"epoch": 0.52,
"learning_rate": 0.00024236710130391174,
"loss": 2.6966,
"theoretical_loss": 3.4749917882360997,
"tokens_seen": 1716912128
},
{
"epoch": 0.52,
"learning_rate": 0.00024235707121364092,
"loss": 2.695,
"theoretical_loss": 3.4749803942803656,
"tokens_seen": 1716977664
},
{
"epoch": 0.52,
"objective/train/docs_used": 972817,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3506035804748535,
"objective/train/theoretical_loss": 3.4749690008812903,
"objective/train/tokens_used": 1737503200,
"theoretical_loss": 3.4749690008812903,
"tokens_seen": 1717043200
},
{
"epoch": 0.52,
"learning_rate": 0.0002423470411233701,
"loss": 2.5105,
"theoretical_loss": 3.4749690008812903,
"tokens_seen": 1717043200
},
{
"epoch": 0.52,
"learning_rate": 0.0002423370110330993,
"loss": 2.7486,
"theoretical_loss": 3.4749576080388254,
"tokens_seen": 1717108736
},
{
"epoch": 0.52,
"learning_rate": 0.0002423269809428285,
"loss": 2.8441,
"theoretical_loss": 3.4749462157529227,
"tokens_seen": 1717174272
},
{
"epoch": 0.52,
"learning_rate": 0.00024231695085255768,
"loss": 2.7228,
"theoretical_loss": 3.4749348240235336,
"tokens_seen": 1717239808
},
{
"epoch": 0.52,
"learning_rate": 0.00024230692076228686,
"loss": 2.5763,
"theoretical_loss": 3.47492343285061,
"tokens_seen": 1717305344
},
{
"epoch": 0.52,
"learning_rate": 0.00024229689067201604,
"loss": 2.6473,
"theoretical_loss": 3.474912042234102,
"tokens_seen": 1717370880
},
{
"epoch": 0.52,
"learning_rate": 0.00024228686058174522,
"loss": 2.7029,
"theoretical_loss": 3.4749006521739636,
"tokens_seen": 1717436416
},
{
"epoch": 0.52,
"learning_rate": 0.00024227683049147443,
"loss": 2.8855,
"theoretical_loss": 3.4748892626701444,
"tokens_seen": 1717501952
},
{
"epoch": 0.52,
"learning_rate": 0.0002422668004012036,
"loss": 2.522,
"theoretical_loss": 3.4748778737225976,
"tokens_seen": 1717567488
},
{
"epoch": 0.52,
"learning_rate": 0.0002422567703109328,
"loss": 2.7884,
"theoretical_loss": 3.4748664853312734,
"tokens_seen": 1717633024
},
{
"epoch": 0.52,
"learning_rate": 0.00024224674022066197,
"loss": 2.464,
"theoretical_loss": 3.474855097496124,
"tokens_seen": 1717698560
},
{
"epoch": 0.52,
"learning_rate": 0.00024223671013039118,
"loss": 2.6961,
"theoretical_loss": 3.474843710217101,
"tokens_seen": 1717764096
},
{
"epoch": 0.52,
"learning_rate": 0.00024222668004012036,
"loss": 2.7975,
"theoretical_loss": 3.474832323494156,
"tokens_seen": 1717829632
},
{
"epoch": 0.52,
"learning_rate": 0.00024221664994984954,
"loss": 2.6121,
"theoretical_loss": 3.474820937327241,
"tokens_seen": 1717895168
},
{
"epoch": 0.52,
"learning_rate": 0.00024220661985957872,
"loss": 2.7361,
"theoretical_loss": 3.474809551716307,
"tokens_seen": 1717960704
},
{
"epoch": 0.52,
"learning_rate": 0.00024219658976930793,
"loss": 2.5772,
"theoretical_loss": 3.474798166661306,
"tokens_seen": 1718026240
},
{
"epoch": 0.52,
"learning_rate": 0.0002421865596790371,
"loss": 2.7568,
"theoretical_loss": 3.4747867821621896,
"tokens_seen": 1718091776
},
{
"epoch": 0.52,
"learning_rate": 0.0002421765295887663,
"loss": 2.4584,
"theoretical_loss": 3.4747753982189096,
"tokens_seen": 1718157312
},
{
"epoch": 0.52,
"learning_rate": 0.00024216649949849547,
"loss": 2.4562,
"theoretical_loss": 3.474764014831417,
"tokens_seen": 1718222848
},
{
"epoch": 0.52,
"learning_rate": 0.00024215646940822466,
"loss": 2.4211,
"theoretical_loss": 3.4747526319996647,
"tokens_seen": 1718288384
},
{
"epoch": 0.52,
"learning_rate": 0.00024214643931795386,
"loss": 2.5976,
"theoretical_loss": 3.474741249723603,
"tokens_seen": 1718353920
},
{
"epoch": 0.52,
"learning_rate": 0.00024213640922768305,
"loss": 2.9066,
"theoretical_loss": 3.4747298680031844,
"tokens_seen": 1718419456
},
{
"epoch": 0.52,
"learning_rate": 0.00024212637913741223,
"loss": 2.5209,
"theoretical_loss": 3.4747184868383605,
"tokens_seen": 1718484992
},
{
"epoch": 0.52,
"learning_rate": 0.00024211634904714143,
"loss": 2.725,
"theoretical_loss": 3.474707106229083,
"tokens_seen": 1718550528
},
{
"epoch": 0.52,
"learning_rate": 0.00024210631895687064,
"loss": 2.4133,
"theoretical_loss": 3.4746957261753035,
"tokens_seen": 1718616064
},
{
"epoch": 0.52,
"objective/train/docs_used": 973424,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8540120124816895,
"objective/train/theoretical_loss": 3.4746843466769737,
"objective/train/tokens_used": 1739141600,
"theoretical_loss": 3.4746843466769737,
"tokens_seen": 1718681600
},
{
"epoch": 0.52,
"learning_rate": 0.00024209628886659982,
"loss": 2.7062,
"theoretical_loss": 3.4746843466769737,
"tokens_seen": 1718681600
},
{
"epoch": 0.52,
"learning_rate": 0.000242086258776329,
"loss": 2.509,
"theoretical_loss": 3.4746729677340453,
"tokens_seen": 1718747136
},
{
"epoch": 0.52,
"learning_rate": 0.00024207622868605819,
"loss": 2.3216,
"theoretical_loss": 3.47466158934647,
"tokens_seen": 1718812672
},
{
"epoch": 0.52,
"learning_rate": 0.00024206619859578737,
"loss": 2.6266,
"theoretical_loss": 3.4746502115141995,
"tokens_seen": 1718878208
},
{
"epoch": 0.52,
"learning_rate": 0.00024205616850551658,
"loss": 2.6836,
"theoretical_loss": 3.4746388342371857,
"tokens_seen": 1718943744
},
{
"epoch": 0.52,
"learning_rate": 0.00024204613841524576,
"loss": 2.6773,
"theoretical_loss": 3.47462745751538,
"tokens_seen": 1719009280
},
{
"epoch": 0.52,
"learning_rate": 0.00024203610832497494,
"loss": 2.6544,
"theoretical_loss": 3.474616081348735,
"tokens_seen": 1719074816
},
{
"epoch": 0.52,
"learning_rate": 0.00024202607823470412,
"loss": 2.6935,
"theoretical_loss": 3.4746047057372014,
"tokens_seen": 1719140352
},
{
"epoch": 0.52,
"learning_rate": 0.00024201604814443333,
"loss": 2.6323,
"theoretical_loss": 3.4745933306807313,
"tokens_seen": 1719205888
},
{
"epoch": 0.52,
"learning_rate": 0.0002420060180541625,
"loss": 2.5048,
"theoretical_loss": 3.474581956179277,
"tokens_seen": 1719271424
},
{
"epoch": 0.52,
"learning_rate": 0.0002419959879638917,
"loss": 2.7468,
"theoretical_loss": 3.474570582232789,
"tokens_seen": 1719336960
},
{
"epoch": 0.52,
"learning_rate": 0.00024198595787362087,
"loss": 2.8566,
"theoretical_loss": 3.4745592088412205,
"tokens_seen": 1719402496
},
{
"epoch": 0.52,
"learning_rate": 0.00024197592778335005,
"loss": 2.7774,
"theoretical_loss": 3.4745478360045228,
"tokens_seen": 1719468032
},
{
"epoch": 0.52,
"learning_rate": 0.00024196589769307926,
"loss": 2.5933,
"theoretical_loss": 3.4745364637226466,
"tokens_seen": 1719533568
},
{
"epoch": 0.52,
"learning_rate": 0.00024195586760280844,
"loss": 2.506,
"theoretical_loss": 3.4745250919955453,
"tokens_seen": 1719599104
},
{
"epoch": 0.52,
"learning_rate": 0.00024194583751253762,
"loss": 2.6552,
"theoretical_loss": 3.47451372082317,
"tokens_seen": 1719664640
},
{
"epoch": 0.52,
"learning_rate": 0.0002419358074222668,
"loss": 2.6128,
"theoretical_loss": 3.474502350205472,
"tokens_seen": 1719730176
},
{
"epoch": 0.52,
"learning_rate": 0.000241925777331996,
"loss": 2.6489,
"theoretical_loss": 3.474490980142404,
"tokens_seen": 1719795712
},
{
"epoch": 0.52,
"learning_rate": 0.0002419157472417252,
"loss": 2.6337,
"theoretical_loss": 3.4744796106339173,
"tokens_seen": 1719861248
},
{
"epoch": 0.52,
"learning_rate": 0.00024190571715145437,
"loss": 2.3488,
"theoretical_loss": 3.474468241679964,
"tokens_seen": 1719926784
},
{
"epoch": 0.52,
"learning_rate": 0.00024189568706118355,
"loss": 2.5449,
"theoretical_loss": 3.474456873280496,
"tokens_seen": 1719992320
},
{
"epoch": 0.52,
"learning_rate": 0.00024188565697091274,
"loss": 2.6173,
"theoretical_loss": 3.4744455054354644,
"tokens_seen": 1720057856
},
{
"epoch": 0.52,
"learning_rate": 0.00024187562688064194,
"loss": 2.4099,
"theoretical_loss": 3.474434138144822,
"tokens_seen": 1720123392
},
{
"epoch": 0.52,
"learning_rate": 0.00024186559679037113,
"loss": 2.6196,
"theoretical_loss": 3.47442277140852,
"tokens_seen": 1720188928
},
{
"epoch": 0.52,
"learning_rate": 0.0002418555667001003,
"loss": 2.5071,
"theoretical_loss": 3.47441140522651,
"tokens_seen": 1720254464
},
{
"epoch": 0.52,
"objective/train/docs_used": 974650,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.691516876220703,
"objective/train/theoretical_loss": 3.474400039598745,
"objective/train/tokens_used": 1740780000,
"theoretical_loss": 3.474400039598745,
"tokens_seen": 1720320000
},
{
"epoch": 0.52,
"learning_rate": 0.0002418455366098295,
"loss": 2.6352,
"theoretical_loss": 3.474400039598745,
"tokens_seen": 1720320000
},
{
"epoch": 0.52,
"learning_rate": 0.0002418355065195587,
"loss": 2.4576,
"theoretical_loss": 3.4743886745251755,
"tokens_seen": 1720385536
},
{
"epoch": 0.52,
"learning_rate": 0.00024182547642928788,
"loss": 2.6549,
"theoretical_loss": 3.4743773100057544,
"tokens_seen": 1720451072
},
{
"epoch": 0.52,
"learning_rate": 0.00024181544633901706,
"loss": 2.5698,
"theoretical_loss": 3.4743659460404333,
"tokens_seen": 1720516608
},
{
"epoch": 0.52,
"learning_rate": 0.00024180541624874624,
"loss": 2.6562,
"theoretical_loss": 3.4743545826291635,
"tokens_seen": 1720582144
},
{
"epoch": 0.52,
"learning_rate": 0.00024179538615847542,
"loss": 2.4442,
"theoretical_loss": 3.4743432197718978,
"tokens_seen": 1720647680
},
{
"epoch": 0.52,
"learning_rate": 0.00024178535606820463,
"loss": 2.66,
"theoretical_loss": 3.4743318574685875,
"tokens_seen": 1720713216
},
{
"epoch": 0.52,
"learning_rate": 0.0002417753259779338,
"loss": 2.5346,
"theoretical_loss": 3.4743204957191844,
"tokens_seen": 1720778752
},
{
"epoch": 0.52,
"learning_rate": 0.000241765295887663,
"loss": 2.5323,
"theoretical_loss": 3.474309134523641,
"tokens_seen": 1720844288
},
{
"epoch": 0.52,
"learning_rate": 0.00024175526579739217,
"loss": 2.5706,
"theoretical_loss": 3.4742977738819087,
"tokens_seen": 1720909824
},
{
"epoch": 0.52,
"learning_rate": 0.00024174523570712138,
"loss": 2.5497,
"theoretical_loss": 3.4742864137939398,
"tokens_seen": 1720975360
},
{
"epoch": 0.52,
"learning_rate": 0.00024173520561685056,
"loss": 2.6159,
"theoretical_loss": 3.474275054259685,
"tokens_seen": 1721040896
},
{
"epoch": 0.52,
"learning_rate": 0.00024172517552657974,
"loss": 2.7169,
"theoretical_loss": 3.4742636952790984,
"tokens_seen": 1721106432
},
{
"epoch": 0.52,
"learning_rate": 0.00024171514543630892,
"loss": 2.7591,
"theoretical_loss": 3.47425233685213,
"tokens_seen": 1721171968
},
{
"epoch": 0.52,
"learning_rate": 0.00024170511534603813,
"loss": 2.6439,
"theoretical_loss": 3.474240978978733,
"tokens_seen": 1721237504
},
{
"epoch": 0.52,
"learning_rate": 0.0002416950852557673,
"loss": 2.6968,
"theoretical_loss": 3.474229621658859,
"tokens_seen": 1721303040
},
{
"epoch": 0.52,
"learning_rate": 0.0002416850551654965,
"loss": 2.5067,
"theoretical_loss": 3.474218264892459,
"tokens_seen": 1721368576
},
{
"epoch": 0.52,
"learning_rate": 0.00024167502507522567,
"loss": 2.6719,
"theoretical_loss": 3.4742069086794864,
"tokens_seen": 1721434112
},
{
"epoch": 0.52,
"learning_rate": 0.00024166499498495486,
"loss": 2.6736,
"theoretical_loss": 3.474195553019892,
"tokens_seen": 1721499648
},
{
"epoch": 0.52,
"learning_rate": 0.00024165496489468406,
"loss": 2.6565,
"theoretical_loss": 3.4741841979136288,
"tokens_seen": 1721565184
},
{
"epoch": 0.52,
"learning_rate": 0.00024164493480441325,
"loss": 2.7174,
"theoretical_loss": 3.4741728433606482,
"tokens_seen": 1721630720
},
{
"epoch": 0.52,
"learning_rate": 0.00024163490471414243,
"loss": 2.495,
"theoretical_loss": 3.4741614893609016,
"tokens_seen": 1721696256
},
{
"epoch": 0.52,
"learning_rate": 0.0002416248746238716,
"loss": 2.7939,
"theoretical_loss": 3.4741501359143423,
"tokens_seen": 1721761792
},
{
"epoch": 0.52,
"learning_rate": 0.00024161484453360082,
"loss": 2.4119,
"theoretical_loss": 3.4741387830209214,
"tokens_seen": 1721827328
},
{
"epoch": 0.52,
"learning_rate": 0.00024160481444333,
"loss": 2.7231,
"theoretical_loss": 3.474127430680591,
"tokens_seen": 1721892864
},
{
"epoch": 0.52,
"objective/train/docs_used": 975130,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3939731121063232,
"objective/train/theoretical_loss": 3.474116078893303,
"objective/train/tokens_used": 1742418400,
"theoretical_loss": 3.474116078893303,
"tokens_seen": 1721958400
},
{
"epoch": 0.52,
"learning_rate": 0.00024159478435305918,
"loss": 2.6016,
"theoretical_loss": 3.474116078893303,
"tokens_seen": 1721958400
},
{
"epoch": 0.52,
"learning_rate": 0.00024158475426278836,
"loss": 2.6084,
"theoretical_loss": 3.47410472765901,
"tokens_seen": 1722023936
},
{
"epoch": 0.52,
"learning_rate": 0.00024157472417251754,
"loss": 2.5789,
"theoretical_loss": 3.474093376977664,
"tokens_seen": 1722089472
},
{
"epoch": 0.52,
"learning_rate": 0.00024156469408224675,
"loss": 2.4633,
"theoretical_loss": 3.474082026849216,
"tokens_seen": 1722155008
},
{
"epoch": 0.52,
"learning_rate": 0.00024155466399197593,
"loss": 2.7123,
"theoretical_loss": 3.4740706772736187,
"tokens_seen": 1722220544
},
{
"epoch": 0.52,
"learning_rate": 0.0002415446339017051,
"loss": 2.7213,
"theoretical_loss": 3.4740593282508243,
"tokens_seen": 1722286080
},
{
"epoch": 0.52,
"learning_rate": 0.0002415346038114343,
"loss": 2.6917,
"theoretical_loss": 3.474047979780785,
"tokens_seen": 1722351616
},
{
"epoch": 0.52,
"learning_rate": 0.0002415245737211635,
"loss": 2.6712,
"theoretical_loss": 3.474036631863452,
"tokens_seen": 1722417152
},
{
"epoch": 0.52,
"learning_rate": 0.00024151454363089268,
"loss": 2.5038,
"theoretical_loss": 3.4740252844987785,
"tokens_seen": 1722482688
},
{
"epoch": 0.52,
"learning_rate": 0.00024150451354062186,
"loss": 2.433,
"theoretical_loss": 3.474013937686715,
"tokens_seen": 1722548224
},
{
"epoch": 0.52,
"learning_rate": 0.00024149448345035104,
"loss": 2.6634,
"theoretical_loss": 3.4740025914272152,
"tokens_seen": 1722613760
},
{
"epoch": 0.52,
"learning_rate": 0.00024148445336008022,
"loss": 2.632,
"theoretical_loss": 3.4739912457202307,
"tokens_seen": 1722679296
},
{
"epoch": 0.52,
"learning_rate": 0.00024147442326980943,
"loss": 2.5113,
"theoretical_loss": 3.4739799005657126,
"tokens_seen": 1722744832
},
{
"epoch": 0.52,
"learning_rate": 0.00024146439317953861,
"loss": 2.5307,
"theoretical_loss": 3.473968555963614,
"tokens_seen": 1722810368
},
{
"epoch": 0.52,
"learning_rate": 0.0002414543630892678,
"loss": 2.6881,
"theoretical_loss": 3.473957211913887,
"tokens_seen": 1722875904
},
{
"epoch": 0.52,
"learning_rate": 0.00024144433299899698,
"loss": 2.5381,
"theoretical_loss": 3.4739458684164832,
"tokens_seen": 1722941440
},
{
"epoch": 0.52,
"learning_rate": 0.00024143430290872618,
"loss": 2.7319,
"theoretical_loss": 3.473934525471355,
"tokens_seen": 1723006976
},
{
"epoch": 0.52,
"learning_rate": 0.00024142427281845537,
"loss": 2.5789,
"theoretical_loss": 3.4739231830784543,
"tokens_seen": 1723072512
},
{
"epoch": 0.52,
"learning_rate": 0.00024141424272818455,
"loss": 2.6924,
"theoretical_loss": 3.473911841237733,
"tokens_seen": 1723138048
},
{
"epoch": 0.52,
"learning_rate": 0.00024140421263791373,
"loss": 2.7414,
"theoretical_loss": 3.473900499949144,
"tokens_seen": 1723203584
},
{
"epoch": 0.52,
"learning_rate": 0.0002413941825476429,
"loss": 2.7356,
"theoretical_loss": 3.4738891592126393,
"tokens_seen": 1723269120
},
{
"epoch": 0.52,
"learning_rate": 0.00024138415245737212,
"loss": 2.4644,
"theoretical_loss": 3.4738778190281705,
"tokens_seen": 1723334656
},
{
"epoch": 0.52,
"learning_rate": 0.0002413741223671013,
"loss": 2.7008,
"theoretical_loss": 3.4738664793956895,
"tokens_seen": 1723400192
},
{
"epoch": 0.52,
"learning_rate": 0.0002413640922768305,
"loss": 2.59,
"theoretical_loss": 3.473855140315149,
"tokens_seen": 1723465728
},
{
"epoch": 0.52,
"learning_rate": 0.0002413540621865597,
"loss": 2.7035,
"theoretical_loss": 3.473843801786501,
"tokens_seen": 1723531264
},
{
"epoch": 0.52,
"objective/train/docs_used": 976352,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.564732551574707,
"objective/train/theoretical_loss": 3.473832463809698,
"objective/train/tokens_used": 1744056800,
"theoretical_loss": 3.473832463809698,
"tokens_seen": 1723596800
},
{
"epoch": 0.52,
"learning_rate": 0.0002413440320962889,
"loss": 2.62,
"theoretical_loss": 3.473832463809698,
"tokens_seen": 1723596800
},
{
"epoch": 0.52,
"learning_rate": 0.00024133400200601808,
"loss": 2.5539,
"theoretical_loss": 3.4738211263846916,
"tokens_seen": 1723662336
},
{
"epoch": 0.52,
"learning_rate": 0.00024132397191574726,
"loss": 2.775,
"theoretical_loss": 3.473809789511434,
"tokens_seen": 1723727872
},
{
"epoch": 0.52,
"learning_rate": 0.00024131394182547644,
"loss": 2.7425,
"theoretical_loss": 3.473798453189878,
"tokens_seen": 1723793408
},
{
"epoch": 0.52,
"learning_rate": 0.00024130391173520562,
"loss": 2.6352,
"theoretical_loss": 3.4737871174199757,
"tokens_seen": 1723858944
},
{
"epoch": 0.52,
"learning_rate": 0.00024129388164493483,
"loss": 2.6554,
"theoretical_loss": 3.4737757822016784,
"tokens_seen": 1723924480
},
{
"epoch": 0.52,
"learning_rate": 0.000241283851554664,
"loss": 2.4559,
"theoretical_loss": 3.473764447534939,
"tokens_seen": 1723990016
},
{
"epoch": 0.52,
"learning_rate": 0.0002412738214643932,
"loss": 2.6863,
"theoretical_loss": 3.4737531134197095,
"tokens_seen": 1724055552
},
{
"epoch": 0.52,
"learning_rate": 0.00024126379137412237,
"loss": 2.4754,
"theoretical_loss": 3.473741779855942,
"tokens_seen": 1724121088
},
{
"epoch": 0.52,
"learning_rate": 0.00024125376128385158,
"loss": 2.3802,
"theoretical_loss": 3.4737304468435894,
"tokens_seen": 1724186624
},
{
"epoch": 0.52,
"learning_rate": 0.00024124373119358076,
"loss": 2.3724,
"theoretical_loss": 3.473719114382603,
"tokens_seen": 1724252160
},
{
"epoch": 0.52,
"learning_rate": 0.00024123370110330994,
"loss": 2.6392,
"theoretical_loss": 3.4737077824729354,
"tokens_seen": 1724317696
},
{
"epoch": 0.52,
"learning_rate": 0.00024122367101303912,
"loss": 2.7686,
"theoretical_loss": 3.4736964511145385,
"tokens_seen": 1724383232
},
{
"epoch": 0.52,
"learning_rate": 0.00024121364092276833,
"loss": 2.8224,
"theoretical_loss": 3.473685120307365,
"tokens_seen": 1724448768
},
{
"epoch": 0.52,
"learning_rate": 0.0002412036108324975,
"loss": 2.5544,
"theoretical_loss": 3.4736737900513672,
"tokens_seen": 1724514304
},
{
"epoch": 0.52,
"learning_rate": 0.0002411935807422267,
"loss": 2.4052,
"theoretical_loss": 3.473662460346497,
"tokens_seen": 1724579840
},
{
"epoch": 0.52,
"learning_rate": 0.00024118355065195587,
"loss": 2.6818,
"theoretical_loss": 3.4736511311927067,
"tokens_seen": 1724645376
},
{
"epoch": 0.52,
"learning_rate": 0.00024117352056168506,
"loss": 2.5562,
"theoretical_loss": 3.4736398025899486,
"tokens_seen": 1724710912
},
{
"epoch": 0.52,
"learning_rate": 0.00024116349047141426,
"loss": 2.6717,
"theoretical_loss": 3.4736284745381756,
"tokens_seen": 1724776448
},
{
"epoch": 0.52,
"learning_rate": 0.00024115346038114345,
"loss": 2.5471,
"theoretical_loss": 3.4736171470373387,
"tokens_seen": 1724841984
},
{
"epoch": 0.52,
"learning_rate": 0.00024114343029087263,
"loss": 2.7706,
"theoretical_loss": 3.473605820087391,
"tokens_seen": 1724907520
},
{
"epoch": 0.52,
"learning_rate": 0.0002411334002006018,
"loss": 2.8283,
"theoretical_loss": 3.4735944936882848,
"tokens_seen": 1724973056
},
{
"epoch": 0.52,
"learning_rate": 0.00024112337011033102,
"loss": 2.6034,
"theoretical_loss": 3.473583167839972,
"tokens_seen": 1725038592
},
{
"epoch": 0.52,
"learning_rate": 0.0002411133400200602,
"loss": 2.5186,
"theoretical_loss": 3.473571842542405,
"tokens_seen": 1725104128
},
{
"epoch": 0.52,
"learning_rate": 0.00024110330992978938,
"loss": 2.6476,
"theoretical_loss": 3.473560517795536,
"tokens_seen": 1725169664
},
{
"epoch": 0.52,
"objective/train/docs_used": 977030,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9039928913116455,
"objective/train/theoretical_loss": 3.473549193599318,
"objective/train/tokens_used": 1745695200,
"theoretical_loss": 3.473549193599318,
"tokens_seen": 1725235200
},
{
"epoch": 0.52,
"learning_rate": 0.00024109327983951856,
"loss": 2.7117,
"theoretical_loss": 3.473549193599318,
"tokens_seen": 1725235200
},
{
"epoch": 0.52,
"learning_rate": 0.00024108324974924774,
"loss": 2.369,
"theoretical_loss": 3.4735378699537027,
"tokens_seen": 1725300736
},
{
"epoch": 0.52,
"learning_rate": 0.00024107321965897695,
"loss": 2.6757,
"theoretical_loss": 3.4735265468586425,
"tokens_seen": 1725366272
},
{
"epoch": 0.52,
"learning_rate": 0.00024106318956870613,
"loss": 2.5216,
"theoretical_loss": 3.4735152243140894,
"tokens_seen": 1725431808
},
{
"epoch": 0.52,
"learning_rate": 0.0002410531594784353,
"loss": 2.4234,
"theoretical_loss": 3.4735039023199965,
"tokens_seen": 1725497344
},
{
"epoch": 0.52,
"learning_rate": 0.0002410431293881645,
"loss": 2.6757,
"theoretical_loss": 3.4734925808763153,
"tokens_seen": 1725562880
},
{
"epoch": 0.52,
"learning_rate": 0.0002410330992978937,
"loss": 2.7024,
"theoretical_loss": 3.473481259982999,
"tokens_seen": 1725628416
},
{
"epoch": 0.52,
"learning_rate": 0.00024102306920762288,
"loss": 2.6121,
"theoretical_loss": 3.4734699396399993,
"tokens_seen": 1725693952
},
{
"epoch": 0.52,
"learning_rate": 0.00024101303911735206,
"loss": 2.5135,
"theoretical_loss": 3.4734586198472686,
"tokens_seen": 1725759488
},
{
"epoch": 0.52,
"learning_rate": 0.00024100300902708124,
"loss": 2.6606,
"theoretical_loss": 3.4734473006047595,
"tokens_seen": 1725825024
},
{
"epoch": 0.52,
"learning_rate": 0.00024099297893681042,
"loss": 2.514,
"theoretical_loss": 3.4734359819124245,
"tokens_seen": 1725890560
},
{
"epoch": 0.52,
"learning_rate": 0.00024098294884653963,
"loss": 2.7979,
"theoretical_loss": 3.4734246637702153,
"tokens_seen": 1725956096
},
{
"epoch": 0.52,
"learning_rate": 0.00024097291875626881,
"loss": 2.5737,
"theoretical_loss": 3.4734133461780847,
"tokens_seen": 1726021632
},
{
"epoch": 0.52,
"learning_rate": 0.000240962888665998,
"loss": 2.5443,
"theoretical_loss": 3.4734020291359853,
"tokens_seen": 1726087168
},
{
"epoch": 0.52,
"learning_rate": 0.00024095285857572718,
"loss": 2.7937,
"theoretical_loss": 3.473390712643869,
"tokens_seen": 1726152704
},
{
"epoch": 0.52,
"learning_rate": 0.00024094282848545638,
"loss": 2.6381,
"theoretical_loss": 3.473379396701689,
"tokens_seen": 1726218240
},
{
"epoch": 0.52,
"learning_rate": 0.00024093279839518557,
"loss": 2.6073,
"theoretical_loss": 3.473368081309397,
"tokens_seen": 1726283776
},
{
"epoch": 0.52,
"learning_rate": 0.00024092276830491475,
"loss": 2.5338,
"theoretical_loss": 3.473356766466945,
"tokens_seen": 1726349312
},
{
"epoch": 0.52,
"learning_rate": 0.00024091273821464393,
"loss": 2.6578,
"theoretical_loss": 3.4733454521742866,
"tokens_seen": 1726414848
},
{
"epoch": 0.52,
"learning_rate": 0.0002409027081243731,
"loss": 2.6962,
"theoretical_loss": 3.4733341384313734,
"tokens_seen": 1726480384
},
{
"epoch": 0.52,
"learning_rate": 0.00024089267803410232,
"loss": 2.6974,
"theoretical_loss": 3.4733228252381574,
"tokens_seen": 1726545920
},
{
"epoch": 0.52,
"learning_rate": 0.0002408826479438315,
"loss": 2.6637,
"theoretical_loss": 3.4733115125945924,
"tokens_seen": 1726611456
},
{
"epoch": 0.52,
"learning_rate": 0.00024087261785356068,
"loss": 2.6925,
"theoretical_loss": 3.47330020050063,
"tokens_seen": 1726676992
},
{
"epoch": 0.52,
"learning_rate": 0.00024086258776328986,
"loss": 2.518,
"theoretical_loss": 3.4732888889562226,
"tokens_seen": 1726742528
},
{
"epoch": 0.52,
"learning_rate": 0.00024085255767301907,
"loss": 2.6195,
"theoretical_loss": 3.4732775779613227,
"tokens_seen": 1726808064
},
{
"epoch": 0.52,
"objective/train/docs_used": 977402,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6840083599090576,
"objective/train/theoretical_loss": 3.473266267515883,
"objective/train/tokens_used": 1747333600,
"theoretical_loss": 3.473266267515883,
"tokens_seen": 1726873600
},
{
"epoch": 0.52,
"learning_rate": 0.00024084252758274825,
"loss": 2.6496,
"theoretical_loss": 3.473266267515883,
"tokens_seen": 1726873600
},
{
"epoch": 0.52,
"learning_rate": 0.00024083249749247743,
"loss": 2.671,
"theoretical_loss": 3.4732549576198553,
"tokens_seen": 1726939136
},
{
"epoch": 0.52,
"learning_rate": 0.0002408224674022066,
"loss": 2.5729,
"theoretical_loss": 3.4732436482731925,
"tokens_seen": 1727004672
},
{
"epoch": 0.52,
"learning_rate": 0.0002408124373119358,
"loss": 2.6019,
"theoretical_loss": 3.4732323394758478,
"tokens_seen": 1727070208
},
{
"epoch": 0.52,
"learning_rate": 0.000240802407221665,
"loss": 2.5761,
"theoretical_loss": 3.473221031227773,
"tokens_seen": 1727135744
},
{
"epoch": 0.52,
"learning_rate": 0.00024079237713139418,
"loss": 2.5142,
"theoretical_loss": 3.47320972352892,
"tokens_seen": 1727201280
},
{
"epoch": 0.52,
"learning_rate": 0.00024078234704112336,
"loss": 2.6193,
"theoretical_loss": 3.473198416379242,
"tokens_seen": 1727266816
},
{
"epoch": 0.52,
"learning_rate": 0.00024077231695085254,
"loss": 2.5487,
"theoretical_loss": 3.4731871097786913,
"tokens_seen": 1727332352
},
{
"epoch": 0.52,
"learning_rate": 0.00024076228686058175,
"loss": 2.5982,
"theoretical_loss": 3.4731758037272202,
"tokens_seen": 1727397888
},
{
"epoch": 0.52,
"learning_rate": 0.00024075225677031093,
"loss": 2.7299,
"theoretical_loss": 3.473164498224782,
"tokens_seen": 1727463424
},
{
"epoch": 0.52,
"learning_rate": 0.00024074222668004012,
"loss": 2.6137,
"theoretical_loss": 3.4731531932713287,
"tokens_seen": 1727528960
},
{
"epoch": 0.52,
"learning_rate": 0.0002407321965897693,
"loss": 2.3614,
"theoretical_loss": 3.473141888866812,
"tokens_seen": 1727594496
},
{
"epoch": 0.52,
"learning_rate": 0.00024072216649949848,
"loss": 2.8096,
"theoretical_loss": 3.4731305850111855,
"tokens_seen": 1727660032
},
{
"epoch": 0.52,
"learning_rate": 0.00024071213640922769,
"loss": 2.6977,
"theoretical_loss": 3.473119281704402,
"tokens_seen": 1727725568
},
{
"epoch": 0.52,
"learning_rate": 0.00024070210631895687,
"loss": 2.6308,
"theoretical_loss": 3.473107978946413,
"tokens_seen": 1727791104
},
{
"epoch": 0.52,
"learning_rate": 0.00024069207622868605,
"loss": 2.6076,
"theoretical_loss": 3.4730966767371716,
"tokens_seen": 1727856640
},
{
"epoch": 0.52,
"learning_rate": 0.00024068204613841523,
"loss": 2.5304,
"theoretical_loss": 3.4730853750766304,
"tokens_seen": 1727922176
},
{
"epoch": 0.52,
"learning_rate": 0.00024067201604814444,
"loss": 2.7431,
"theoretical_loss": 3.4730740739647414,
"tokens_seen": 1727987712
},
{
"epoch": 0.52,
"learning_rate": 0.00024066198595787362,
"loss": 2.5978,
"theoretical_loss": 3.4730627734014576,
"tokens_seen": 1728053248
},
{
"epoch": 0.52,
"learning_rate": 0.0002406519558676028,
"loss": 2.6449,
"theoretical_loss": 3.4730514733867315,
"tokens_seen": 1728118784
},
{
"epoch": 0.52,
"learning_rate": 0.00024064192577733198,
"loss": 2.7191,
"theoretical_loss": 3.4730401739205163,
"tokens_seen": 1728184320
},
{
"epoch": 0.52,
"learning_rate": 0.00024063189568706116,
"loss": 2.6494,
"theoretical_loss": 3.4730288750027634,
"tokens_seen": 1728249856
},
{
"epoch": 0.52,
"learning_rate": 0.00024062186559679037,
"loss": 2.6994,
"theoretical_loss": 3.473017576633426,
"tokens_seen": 1728315392
},
{
"epoch": 0.52,
"learning_rate": 0.00024061183550651958,
"loss": 2.6576,
"theoretical_loss": 3.4730062788124565,
"tokens_seen": 1728380928
},
{
"epoch": 0.52,
"learning_rate": 0.00024060180541624876,
"loss": 2.5097,
"theoretical_loss": 3.472994981539808,
"tokens_seen": 1728446464
},
{
"epoch": 0.52,
"objective/train/docs_used": 978835,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.617340564727783,
"objective/train/theoretical_loss": 3.4729836848154325,
"objective/train/tokens_used": 1748972000,
"theoretical_loss": 3.4729836848154325,
"tokens_seen": 1728512000
},
{
"epoch": 0.52,
"learning_rate": 0.00024059177532597794,
"loss": 2.5492,
"theoretical_loss": 3.4729836848154325,
"tokens_seen": 1728512000
},
{
"epoch": 0.52,
"learning_rate": 0.00024058174523570715,
"loss": 2.6494,
"theoretical_loss": 3.472972388639283,
"tokens_seen": 1728577536
},
{
"epoch": 0.52,
"learning_rate": 0.00024057171514543633,
"loss": 2.7081,
"theoretical_loss": 3.4729610930113117,
"tokens_seen": 1728643072
},
{
"epoch": 0.52,
"learning_rate": 0.0002405616850551655,
"loss": 2.5013,
"theoretical_loss": 3.472949797931472,
"tokens_seen": 1728708608
},
{
"epoch": 0.52,
"learning_rate": 0.0002405516549648947,
"loss": 2.5517,
"theoretical_loss": 3.472938503399715,
"tokens_seen": 1728774144
},
{
"epoch": 0.52,
"learning_rate": 0.0002405416248746239,
"loss": 2.7114,
"theoretical_loss": 3.472927209415995,
"tokens_seen": 1728839680
},
{
"epoch": 0.52,
"learning_rate": 0.00024053159478435308,
"loss": 2.6792,
"theoretical_loss": 3.4729159159802636,
"tokens_seen": 1728905216
},
{
"epoch": 0.52,
"learning_rate": 0.00024052156469408226,
"loss": 2.4676,
"theoretical_loss": 3.4729046230924743,
"tokens_seen": 1728970752
},
{
"epoch": 0.52,
"learning_rate": 0.00024051153460381144,
"loss": 2.4256,
"theoretical_loss": 3.472893330752579,
"tokens_seen": 1729036288
},
{
"epoch": 0.52,
"learning_rate": 0.00024050150451354062,
"loss": 2.6138,
"theoretical_loss": 3.4728820389605306,
"tokens_seen": 1729101824
},
{
"epoch": 0.52,
"learning_rate": 0.00024049147442326983,
"loss": 2.8679,
"theoretical_loss": 3.4728707477162817,
"tokens_seen": 1729167360
},
{
"epoch": 0.52,
"learning_rate": 0.00024048144433299901,
"loss": 2.6657,
"theoretical_loss": 3.472859457019785,
"tokens_seen": 1729232896
},
{
"epoch": 0.52,
"learning_rate": 0.0002404714142427282,
"loss": 2.6521,
"theoretical_loss": 3.4728481668709934,
"tokens_seen": 1729298432
},
{
"epoch": 0.52,
"learning_rate": 0.00024046138415245738,
"loss": 2.3611,
"theoretical_loss": 3.4728368772698586,
"tokens_seen": 1729363968
},
{
"epoch": 0.52,
"learning_rate": 0.00024045135406218658,
"loss": 2.6055,
"theoretical_loss": 3.4728255882163346,
"tokens_seen": 1729429504
},
{
"epoch": 0.52,
"learning_rate": 0.00024044132397191577,
"loss": 2.7052,
"theoretical_loss": 3.4728142997103735,
"tokens_seen": 1729495040
},
{
"epoch": 0.52,
"learning_rate": 0.00024043129388164495,
"loss": 2.6801,
"theoretical_loss": 3.472803011751928,
"tokens_seen": 1729560576
},
{
"epoch": 0.52,
"learning_rate": 0.00024042126379137413,
"loss": 2.579,
"theoretical_loss": 3.472791724340951,
"tokens_seen": 1729626112
},
{
"epoch": 0.52,
"learning_rate": 0.0002404112337011033,
"loss": 2.3884,
"theoretical_loss": 3.472780437477395,
"tokens_seen": 1729691648
},
{
"epoch": 0.52,
"learning_rate": 0.00024040120361083252,
"loss": 2.5473,
"theoretical_loss": 3.472769151161213,
"tokens_seen": 1729757184
},
{
"epoch": 0.52,
"learning_rate": 0.0002403911735205617,
"loss": 2.5958,
"theoretical_loss": 3.4727578653923565,
"tokens_seen": 1729822720
},
{
"epoch": 0.52,
"learning_rate": 0.00024038114343029088,
"loss": 2.3295,
"theoretical_loss": 3.4727465801707798,
"tokens_seen": 1729888256
},
{
"epoch": 0.52,
"learning_rate": 0.00024037111334002006,
"loss": 2.6406,
"theoretical_loss": 3.472735295496435,
"tokens_seen": 1729953792
},
{
"epoch": 0.52,
"learning_rate": 0.00024036108324974927,
"loss": 2.5794,
"theoretical_loss": 3.4727240113692748,
"tokens_seen": 1730019328
},
{
"epoch": 0.52,
"learning_rate": 0.00024035105315947845,
"loss": 2.5206,
"theoretical_loss": 3.472712727789252,
"tokens_seen": 1730084864
},
{
"epoch": 0.52,
"objective/train/docs_used": 979527,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.767808437347412,
"objective/train/theoretical_loss": 3.4727014447563196,
"objective/train/tokens_used": 1750610400,
"theoretical_loss": 3.4727014447563196,
"tokens_seen": 1730150400
},
{
"epoch": 0.52,
"learning_rate": 0.00024034102306920763,
"loss": 2.8662,
"theoretical_loss": 3.4727014447563196,
"tokens_seen": 1730150400
},
{
"epoch": 0.52,
"learning_rate": 0.0002403309929789368,
"loss": 2.8813,
"theoretical_loss": 3.472690162270429,
"tokens_seen": 1730215936
},
{
"epoch": 0.52,
"learning_rate": 0.000240320962888666,
"loss": 2.8476,
"theoretical_loss": 3.472678880331535,
"tokens_seen": 1730281472
},
{
"epoch": 0.52,
"learning_rate": 0.0002403109327983952,
"loss": 2.4743,
"theoretical_loss": 3.4726675989395894,
"tokens_seen": 1730347008
},
{
"epoch": 0.52,
"learning_rate": 0.00024030090270812438,
"loss": 2.6661,
"theoretical_loss": 3.472656318094545,
"tokens_seen": 1730412544
},
{
"epoch": 0.52,
"learning_rate": 0.00024029087261785356,
"loss": 2.7985,
"theoretical_loss": 3.4726450377963545,
"tokens_seen": 1730478080
},
{
"epoch": 0.52,
"learning_rate": 0.00024028084252758274,
"loss": 2.5817,
"theoretical_loss": 3.47263375804497,
"tokens_seen": 1730543616
},
{
"epoch": 0.52,
"learning_rate": 0.00024027081243731195,
"loss": 2.5363,
"theoretical_loss": 3.472622478840346,
"tokens_seen": 1730609152
},
{
"epoch": 0.52,
"learning_rate": 0.00024026078234704113,
"loss": 2.7176,
"theoretical_loss": 3.472611200182434,
"tokens_seen": 1730674688
},
{
"epoch": 0.52,
"learning_rate": 0.00024025075225677032,
"loss": 2.4767,
"theoretical_loss": 3.472599922071187,
"tokens_seen": 1730740224
},
{
"epoch": 0.52,
"learning_rate": 0.0002402407221664995,
"loss": 2.7052,
"theoretical_loss": 3.472588644506558,
"tokens_seen": 1730805760
},
{
"epoch": 0.52,
"learning_rate": 0.00024023069207622868,
"loss": 2.7205,
"theoretical_loss": 3.4725773674884994,
"tokens_seen": 1730871296
},
{
"epoch": 0.52,
"learning_rate": 0.00024022066198595789,
"loss": 2.5732,
"theoretical_loss": 3.4725660910169647,
"tokens_seen": 1730936832
},
{
"epoch": 0.52,
"learning_rate": 0.00024021063189568707,
"loss": 2.535,
"theoretical_loss": 3.472554815091906,
"tokens_seen": 1731002368
},
{
"epoch": 0.52,
"learning_rate": 0.00024020060180541625,
"loss": 2.4127,
"theoretical_loss": 3.472543539713277,
"tokens_seen": 1731067904
},
{
"epoch": 0.52,
"learning_rate": 0.00024019057171514543,
"loss": 2.4635,
"theoretical_loss": 3.4725322648810297,
"tokens_seen": 1731133440
},
{
"epoch": 0.52,
"learning_rate": 0.00024018054162487464,
"loss": 2.6366,
"theoretical_loss": 3.4725209905951173,
"tokens_seen": 1731198976
},
{
"epoch": 0.52,
"learning_rate": 0.00024017051153460382,
"loss": 2.5269,
"theoretical_loss": 3.472509716855493,
"tokens_seen": 1731264512
},
{
"epoch": 0.52,
"learning_rate": 0.000240160481444333,
"loss": 2.6166,
"theoretical_loss": 3.4724984436621087,
"tokens_seen": 1731330048
},
{
"epoch": 0.52,
"learning_rate": 0.00024015045135406218,
"loss": 2.5634,
"theoretical_loss": 3.4724871710149183,
"tokens_seen": 1731395584
},
{
"epoch": 0.52,
"learning_rate": 0.00024014042126379136,
"loss": 2.7434,
"theoretical_loss": 3.4724758989138738,
"tokens_seen": 1731461120
},
{
"epoch": 0.52,
"learning_rate": 0.00024013039117352057,
"loss": 2.6719,
"theoretical_loss": 3.472464627358929,
"tokens_seen": 1731526656
},
{
"epoch": 0.52,
"learning_rate": 0.00024012036108324975,
"loss": 2.5765,
"theoretical_loss": 3.4724533563500355,
"tokens_seen": 1731592192
},
{
"epoch": 0.52,
"learning_rate": 0.00024011033099297893,
"loss": 2.4608,
"theoretical_loss": 3.472442085887147,
"tokens_seen": 1731657728
},
{
"epoch": 0.52,
"learning_rate": 0.00024010030090270811,
"loss": 2.6515,
"theoretical_loss": 3.472430815970217,
"tokens_seen": 1731723264
},
{
"epoch": 0.52,
"objective/train/docs_used": 980288,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.865384101867676,
"objective/train/theoretical_loss": 3.4724195465991974,
"objective/train/tokens_used": 1752248800,
"theoretical_loss": 3.4724195465991974,
"tokens_seen": 1731788800
},
{
"epoch": 0.52,
"learning_rate": 0.00024009027081243732,
"loss": 2.5767,
"theoretical_loss": 3.4724195465991974,
"tokens_seen": 1731788800
},
{
"epoch": 0.52,
"learning_rate": 0.0002400802407221665,
"loss": 2.885,
"theoretical_loss": 3.4724082777740413,
"tokens_seen": 1731854336
},
{
"epoch": 0.52,
"learning_rate": 0.00024007021063189568,
"loss": 2.6993,
"theoretical_loss": 3.472397009494702,
"tokens_seen": 1731919872
},
{
"epoch": 0.52,
"learning_rate": 0.00024006018054162487,
"loss": 2.5726,
"theoretical_loss": 3.4723857417611317,
"tokens_seen": 1731985408
},
{
"epoch": 0.52,
"learning_rate": 0.00024005015045135407,
"loss": 2.6048,
"theoretical_loss": 3.4723744745732836,
"tokens_seen": 1732050944
},
{
"epoch": 0.52,
"learning_rate": 0.00024004012036108325,
"loss": 2.7858,
"theoretical_loss": 3.472363207931111,
"tokens_seen": 1732116480
},
{
"epoch": 0.52,
"learning_rate": 0.00024003009027081244,
"loss": 2.4669,
"theoretical_loss": 3.4723519418345665,
"tokens_seen": 1732182016
},
{
"epoch": 0.52,
"learning_rate": 0.00024002006018054162,
"loss": 2.738,
"theoretical_loss": 3.4723406762836033,
"tokens_seen": 1732247552
},
{
"epoch": 0.52,
"learning_rate": 0.0002400100300902708,
"loss": 2.6058,
"theoretical_loss": 3.472329411278174,
"tokens_seen": 1732313088
},
{
"epoch": 0.52,
"learning_rate": 0.00024,
"loss": 2.8235,
"theoretical_loss": 3.472318146818232,
"tokens_seen": 1732378624
},
{
"epoch": 0.52,
"learning_rate": 0.0002399899699097292,
"loss": 2.7444,
"theoretical_loss": 3.4723068829037294,
"tokens_seen": 1732444160
},
{
"epoch": 0.53,
"learning_rate": 0.00023997993981945837,
"loss": 2.7223,
"theoretical_loss": 3.47229561953462,
"tokens_seen": 1732509696
},
{
"epoch": 0.53,
"learning_rate": 0.00023996990972918755,
"loss": 2.803,
"theoretical_loss": 3.4722843567108566,
"tokens_seen": 1732575232
},
{
"epoch": 0.53,
"learning_rate": 0.00023995987963891676,
"loss": 2.7514,
"theoretical_loss": 3.4722730944323916,
"tokens_seen": 1732640768
},
{
"epoch": 0.53,
"learning_rate": 0.00023994984954864594,
"loss": 2.7387,
"theoretical_loss": 3.4722618326991785,
"tokens_seen": 1732706304
},
{
"epoch": 0.53,
"learning_rate": 0.00023993981945837512,
"loss": 2.6156,
"theoretical_loss": 3.4722505715111707,
"tokens_seen": 1732771840
},
{
"epoch": 0.53,
"learning_rate": 0.0002399297893681043,
"loss": 2.7157,
"theoretical_loss": 3.4722393108683205,
"tokens_seen": 1732837376
},
{
"epoch": 0.53,
"learning_rate": 0.00023991975927783348,
"loss": 2.6092,
"theoretical_loss": 3.472228050770581,
"tokens_seen": 1732902912
},
{
"epoch": 0.53,
"learning_rate": 0.0002399097291875627,
"loss": 2.8596,
"theoretical_loss": 3.472216791217905,
"tokens_seen": 1732968448
},
{
"epoch": 0.53,
"learning_rate": 0.00023989969909729187,
"loss": 2.6144,
"theoretical_loss": 3.4722055322102463,
"tokens_seen": 1733033984
},
{
"epoch": 0.53,
"learning_rate": 0.00023988966900702105,
"loss": 2.6067,
"theoretical_loss": 3.472194273747557,
"tokens_seen": 1733099520
},
{
"epoch": 0.53,
"learning_rate": 0.00023987963891675023,
"loss": 2.4881,
"theoretical_loss": 3.47218301582979,
"tokens_seen": 1733165056
},
{
"epoch": 0.53,
"learning_rate": 0.00023986960882647944,
"loss": 2.7094,
"theoretical_loss": 3.4721717584569003,
"tokens_seen": 1733230592
},
{
"epoch": 0.53,
"learning_rate": 0.00023985957873620865,
"loss": 2.4774,
"theoretical_loss": 3.472160501628838,
"tokens_seen": 1733296128
},
{
"epoch": 0.53,
"learning_rate": 0.00023984954864593783,
"loss": 2.8638,
"theoretical_loss": 3.4721492453455585,
"tokens_seen": 1733361664
},
{
"epoch": 0.53,
"objective/train/docs_used": 981145,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8345024585723877,
"objective/train/theoretical_loss": 3.4721379896070133,
"objective/train/tokens_used": 1753887200,
"theoretical_loss": 3.4721379896070133,
"tokens_seen": 1733427200
},
{
"epoch": 0.53,
"learning_rate": 0.000239839518555667,
"loss": 2.649,
"theoretical_loss": 3.4721379896070133,
"tokens_seen": 1733427200
},
{
"epoch": 0.53,
"learning_rate": 0.0002398294884653962,
"loss": 2.5095,
"theoretical_loss": 3.4721267344131563,
"tokens_seen": 1733492736
},
{
"epoch": 0.53,
"learning_rate": 0.0002398194583751254,
"loss": 2.5694,
"theoretical_loss": 3.4721154797639406,
"tokens_seen": 1733558272
},
{
"epoch": 0.53,
"learning_rate": 0.00023980942828485458,
"loss": 2.6264,
"theoretical_loss": 3.4721042256593186,
"tokens_seen": 1733623808
},
{
"epoch": 0.53,
"learning_rate": 0.00023979939819458376,
"loss": 2.8336,
"theoretical_loss": 3.472092972099244,
"tokens_seen": 1733689344
},
{
"epoch": 0.53,
"learning_rate": 0.00023978936810431295,
"loss": 2.747,
"theoretical_loss": 3.4720817190836692,
"tokens_seen": 1733754880
},
{
"epoch": 0.53,
"learning_rate": 0.00023977933801404215,
"loss": 2.8868,
"theoretical_loss": 3.472070466612548,
"tokens_seen": 1733820416
},
{
"epoch": 0.53,
"learning_rate": 0.00023976930792377133,
"loss": 2.7084,
"theoretical_loss": 3.4720592146858333,
"tokens_seen": 1733885952
},
{
"epoch": 0.53,
"learning_rate": 0.00023975927783350052,
"loss": 2.7457,
"theoretical_loss": 3.4720479633034778,
"tokens_seen": 1733951488
},
{
"epoch": 0.53,
"learning_rate": 0.0002397492477432297,
"loss": 2.5377,
"theoretical_loss": 3.472036712465435,
"tokens_seen": 1734017024
},
{
"epoch": 0.53,
"learning_rate": 0.00023973921765295888,
"loss": 2.7785,
"theoretical_loss": 3.4720254621716578,
"tokens_seen": 1734082560
},
{
"epoch": 0.53,
"learning_rate": 0.00023972918756268809,
"loss": 2.7292,
"theoretical_loss": 3.472014212422099,
"tokens_seen": 1734148096
},
{
"epoch": 0.53,
"learning_rate": 0.00023971915747241727,
"loss": 2.6925,
"theoretical_loss": 3.4720029632167124,
"tokens_seen": 1734213632
},
{
"epoch": 0.53,
"learning_rate": 0.00023970912738214645,
"loss": 2.4948,
"theoretical_loss": 3.471991714555451,
"tokens_seen": 1734279168
},
{
"epoch": 0.53,
"learning_rate": 0.00023969909729187563,
"loss": 2.9792,
"theoretical_loss": 3.471980466438267,
"tokens_seen": 1734344704
},
{
"epoch": 0.53,
"learning_rate": 0.00023968906720160484,
"loss": 2.6948,
"theoretical_loss": 3.471969218865115,
"tokens_seen": 1734410240
},
{
"epoch": 0.53,
"learning_rate": 0.00023967903711133402,
"loss": 2.7173,
"theoretical_loss": 3.4719579718359466,
"tokens_seen": 1734475776
},
{
"epoch": 0.53,
"learning_rate": 0.0002396690070210632,
"loss": 2.4533,
"theoretical_loss": 3.4719467253507164,
"tokens_seen": 1734541312
},
{
"epoch": 0.53,
"learning_rate": 0.00023965897693079238,
"loss": 2.7136,
"theoretical_loss": 3.471935479409376,
"tokens_seen": 1734606848
},
{
"epoch": 0.53,
"learning_rate": 0.00023964894684052156,
"loss": 2.4764,
"theoretical_loss": 3.4719242340118797,
"tokens_seen": 1734672384
},
{
"epoch": 0.53,
"learning_rate": 0.00023963891675025077,
"loss": 2.4895,
"theoretical_loss": 3.471912989158181,
"tokens_seen": 1734737920
},
{
"epoch": 0.53,
"learning_rate": 0.00023962888665997995,
"loss": 2.7,
"theoretical_loss": 3.471901744848231,
"tokens_seen": 1734803456
},
{
"epoch": 0.53,
"learning_rate": 0.00023961885656970913,
"loss": 2.8935,
"theoretical_loss": 3.471890501081985,
"tokens_seen": 1734868992
},
{
"epoch": 0.53,
"learning_rate": 0.00023960882647943831,
"loss": 2.721,
"theoretical_loss": 3.4718792578593956,
"tokens_seen": 1734934528
},
{
"epoch": 0.53,
"learning_rate": 0.00023959879638916752,
"loss": 2.7056,
"theoretical_loss": 3.4718680151804158,
"tokens_seen": 1735000064
},
{
"epoch": 0.53,
"objective/train/docs_used": 982466,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.901257038116455,
"objective/train/theoretical_loss": 3.4718567730449985,
"objective/train/tokens_used": 1755525600,
"theoretical_loss": 3.4718567730449985,
"tokens_seen": 1735065600
},
{
"epoch": 0.53,
"learning_rate": 0.0002395887662988967,
"loss": 2.8475,
"theoretical_loss": 3.4718567730449985,
"tokens_seen": 1735065600
},
{
"epoch": 0.53,
"learning_rate": 0.00023957873620862588,
"loss": 2.4407,
"theoretical_loss": 3.4718455314530976,
"tokens_seen": 1735131136
},
{
"epoch": 0.53,
"learning_rate": 0.00023956870611835507,
"loss": 2.5301,
"theoretical_loss": 3.471834290404665,
"tokens_seen": 1735196672
},
{
"epoch": 0.53,
"learning_rate": 0.00023955867602808427,
"loss": 2.3615,
"theoretical_loss": 3.4718230498996556,
"tokens_seen": 1735262208
},
{
"epoch": 0.53,
"learning_rate": 0.00023954864593781345,
"loss": 2.4546,
"theoretical_loss": 3.4718118099380213,
"tokens_seen": 1735327744
},
{
"epoch": 0.53,
"learning_rate": 0.00023953861584754264,
"loss": 2.709,
"theoretical_loss": 3.4718005705197164,
"tokens_seen": 1735393280
},
{
"epoch": 0.53,
"learning_rate": 0.00023952858575727182,
"loss": 2.6454,
"theoretical_loss": 3.4717893316446933,
"tokens_seen": 1735458816
},
{
"epoch": 0.53,
"learning_rate": 0.000239518555667001,
"loss": 2.7955,
"theoretical_loss": 3.471778093312905,
"tokens_seen": 1735524352
},
{
"epoch": 0.53,
"learning_rate": 0.0002395085255767302,
"loss": 2.591,
"theoretical_loss": 3.4717668555243053,
"tokens_seen": 1735589888
},
{
"epoch": 0.53,
"learning_rate": 0.0002394984954864594,
"loss": 2.556,
"theoretical_loss": 3.4717556182788476,
"tokens_seen": 1735655424
},
{
"epoch": 0.53,
"learning_rate": 0.00023948846539618857,
"loss": 2.558,
"theoretical_loss": 3.471744381576485,
"tokens_seen": 1735720960
},
{
"epoch": 0.53,
"learning_rate": 0.00023947843530591775,
"loss": 2.7923,
"theoretical_loss": 3.4717331454171703,
"tokens_seen": 1735786496
},
{
"epoch": 0.53,
"learning_rate": 0.00023946840521564696,
"loss": 2.6013,
"theoretical_loss": 3.471721909800857,
"tokens_seen": 1735852032
},
{
"epoch": 0.53,
"learning_rate": 0.00023945837512537614,
"loss": 2.5876,
"theoretical_loss": 3.4717106747274986,
"tokens_seen": 1735917568
},
{
"epoch": 0.53,
"learning_rate": 0.00023944834503510532,
"loss": 2.5319,
"theoretical_loss": 3.471699440197048,
"tokens_seen": 1735983104
},
{
"epoch": 0.53,
"learning_rate": 0.0002394383149448345,
"loss": 2.5029,
"theoretical_loss": 3.4716882062094587,
"tokens_seen": 1736048640
},
{
"epoch": 0.53,
"learning_rate": 0.00023942828485456368,
"loss": 2.4574,
"theoretical_loss": 3.4716769727646835,
"tokens_seen": 1736114176
},
{
"epoch": 0.53,
"learning_rate": 0.0002394182547642929,
"loss": 2.5879,
"theoretical_loss": 3.4716657398626767,
"tokens_seen": 1736179712
},
{
"epoch": 0.53,
"learning_rate": 0.00023940822467402207,
"loss": 2.6598,
"theoretical_loss": 3.471654507503391,
"tokens_seen": 1736245248
},
{
"epoch": 0.53,
"learning_rate": 0.00023939819458375125,
"loss": 2.7946,
"theoretical_loss": 3.4716432756867794,
"tokens_seen": 1736310784
},
{
"epoch": 0.53,
"learning_rate": 0.00023938816449348043,
"loss": 2.4548,
"theoretical_loss": 3.4716320444127957,
"tokens_seen": 1736376320
},
{
"epoch": 0.53,
"learning_rate": 0.00023937813440320964,
"loss": 2.7882,
"theoretical_loss": 3.471620813681393,
"tokens_seen": 1736441856
},
{
"epoch": 0.53,
"learning_rate": 0.00023936810431293882,
"loss": 2.825,
"theoretical_loss": 3.4716095834925245,
"tokens_seen": 1736507392
},
{
"epoch": 0.53,
"learning_rate": 0.000239358074222668,
"loss": 2.6654,
"theoretical_loss": 3.4715983538461432,
"tokens_seen": 1736572928
},
{
"epoch": 0.53,
"learning_rate": 0.00023934804413239719,
"loss": 2.667,
"theoretical_loss": 3.4715871247422037,
"tokens_seen": 1736638464
},
{
"epoch": 0.53,
"objective/train/docs_used": 983129,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.610753297805786,
"objective/train/theoretical_loss": 3.4715758961806573,
"objective/train/tokens_used": 1757164000,
"theoretical_loss": 3.4715758961806573,
"tokens_seen": 1736704000
},
{
"epoch": 0.53,
"learning_rate": 0.00023933801404212637,
"loss": 2.8267,
"theoretical_loss": 3.4715758961806573,
"tokens_seen": 1736704000
},
{
"epoch": 0.53,
"learning_rate": 0.00023932798395185557,
"loss": 2.6163,
"theoretical_loss": 3.4715646681614594,
"tokens_seen": 1736769536
},
{
"epoch": 0.53,
"learning_rate": 0.00023931795386158476,
"loss": 2.6598,
"theoretical_loss": 3.4715534406845627,
"tokens_seen": 1736835072
},
{
"epoch": 0.53,
"learning_rate": 0.00023930792377131394,
"loss": 2.5863,
"theoretical_loss": 3.4715422137499194,
"tokens_seen": 1736900608
},
{
"epoch": 0.53,
"learning_rate": 0.00023929789368104312,
"loss": 2.856,
"theoretical_loss": 3.4715309873574847,
"tokens_seen": 1736966144
},
{
"epoch": 0.53,
"learning_rate": 0.00023928786359077233,
"loss": 2.6061,
"theoretical_loss": 3.4715197615072104,
"tokens_seen": 1737031680
},
{
"epoch": 0.53,
"learning_rate": 0.0002392778335005015,
"loss": 2.7639,
"theoretical_loss": 3.4715085361990505,
"tokens_seen": 1737097216
},
{
"epoch": 0.53,
"learning_rate": 0.0002392678034102307,
"loss": 2.4124,
"theoretical_loss": 3.471497311432959,
"tokens_seen": 1737162752
},
{
"epoch": 0.53,
"learning_rate": 0.00023925777331995987,
"loss": 2.7836,
"theoretical_loss": 3.4714860872088877,
"tokens_seen": 1737228288
},
{
"epoch": 0.53,
"learning_rate": 0.00023924774322968905,
"loss": 2.6241,
"theoretical_loss": 3.471474863526791,
"tokens_seen": 1737293824
},
{
"epoch": 0.53,
"learning_rate": 0.00023923771313941826,
"loss": 2.6454,
"theoretical_loss": 3.471463640386623,
"tokens_seen": 1737359360
},
{
"epoch": 0.53,
"learning_rate": 0.00023922768304914744,
"loss": 2.6374,
"theoretical_loss": 3.4714524177883352,
"tokens_seen": 1737424896
},
{
"epoch": 0.53,
"learning_rate": 0.00023921765295887662,
"loss": 2.785,
"theoretical_loss": 3.4714411957318827,
"tokens_seen": 1737490432
},
{
"epoch": 0.53,
"learning_rate": 0.0002392076228686058,
"loss": 2.7196,
"theoretical_loss": 3.4714299742172186,
"tokens_seen": 1737555968
},
{
"epoch": 0.53,
"learning_rate": 0.000239197592778335,
"loss": 2.7872,
"theoretical_loss": 3.4714187532442953,
"tokens_seen": 1737621504
},
{
"epoch": 0.53,
"learning_rate": 0.0002391875626880642,
"loss": 2.8094,
"theoretical_loss": 3.471407532813067,
"tokens_seen": 1737687040
},
{
"epoch": 0.53,
"learning_rate": 0.00023917753259779337,
"loss": 2.6587,
"theoretical_loss": 3.4713963129234875,
"tokens_seen": 1737752576
},
{
"epoch": 0.53,
"learning_rate": 0.00023916750250752255,
"loss": 2.4922,
"theoretical_loss": 3.4713850935755093,
"tokens_seen": 1737818112
},
{
"epoch": 0.53,
"learning_rate": 0.00023915747241725174,
"loss": 2.7313,
"theoretical_loss": 3.4713738747690863,
"tokens_seen": 1737883648
},
{
"epoch": 0.53,
"learning_rate": 0.00023914744232698094,
"loss": 2.4993,
"theoretical_loss": 3.471362656504172,
"tokens_seen": 1737949184
},
{
"epoch": 0.53,
"learning_rate": 0.00023913741223671012,
"loss": 2.4836,
"theoretical_loss": 3.4713514387807196,
"tokens_seen": 1738014720
},
{
"epoch": 0.53,
"learning_rate": 0.0002391273821464393,
"loss": 2.6591,
"theoretical_loss": 3.471340221598683,
"tokens_seen": 1738080256
},
{
"epoch": 0.53,
"learning_rate": 0.0002391173520561685,
"loss": 2.4776,
"theoretical_loss": 3.4713290049580157,
"tokens_seen": 1738145792
},
{
"epoch": 0.53,
"learning_rate": 0.00023910732196589772,
"loss": 2.567,
"theoretical_loss": 3.4713177888586704,
"tokens_seen": 1738211328
},
{
"epoch": 0.53,
"learning_rate": 0.0002390972918756269,
"loss": 2.5476,
"theoretical_loss": 3.471306573300601,
"tokens_seen": 1738276864
},
{
"epoch": 0.53,
"objective/train/docs_used": 984348,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.986797571182251,
"objective/train/theoretical_loss": 3.4712953582837613,
"objective/train/tokens_used": 1758802400,
"theoretical_loss": 3.4712953582837613,
"tokens_seen": 1738342400
},
{
"epoch": 0.53,
"learning_rate": 0.00023908726178535608,
"loss": 2.4408,
"theoretical_loss": 3.4712953582837613,
"tokens_seen": 1738342400
},
{
"epoch": 0.53,
"learning_rate": 0.00023907723169508527,
"loss": 2.7978,
"theoretical_loss": 3.471284143808104,
"tokens_seen": 1738407936
},
{
"epoch": 0.53,
"learning_rate": 0.00023906720160481447,
"loss": 2.5738,
"theoretical_loss": 3.4712729298735834,
"tokens_seen": 1738473472
},
{
"epoch": 0.53,
"learning_rate": 0.00023905717151454365,
"loss": 2.6355,
"theoretical_loss": 3.471261716480152,
"tokens_seen": 1738539008
},
{
"epoch": 0.53,
"learning_rate": 0.00023904714142427284,
"loss": 2.7452,
"theoretical_loss": 3.4712505036277648,
"tokens_seen": 1738604544
},
{
"epoch": 0.53,
"learning_rate": 0.00023903711133400202,
"loss": 2.7345,
"theoretical_loss": 3.4712392913163734,
"tokens_seen": 1738670080
},
{
"epoch": 0.53,
"learning_rate": 0.0002390270812437312,
"loss": 2.6134,
"theoretical_loss": 3.4712280795459334,
"tokens_seen": 1738735616
},
{
"epoch": 0.53,
"learning_rate": 0.0002390170511534604,
"loss": 2.7171,
"theoretical_loss": 3.4712168683163966,
"tokens_seen": 1738801152
},
{
"epoch": 0.53,
"learning_rate": 0.0002390070210631896,
"loss": 2.639,
"theoretical_loss": 3.4712056576277175,
"tokens_seen": 1738866688
},
{
"epoch": 0.53,
"learning_rate": 0.00023899699097291877,
"loss": 2.5487,
"theoretical_loss": 3.4711944474798493,
"tokens_seen": 1738932224
},
{
"epoch": 0.53,
"learning_rate": 0.00023898696088264795,
"loss": 2.786,
"theoretical_loss": 3.4711832378727454,
"tokens_seen": 1738997760
},
{
"epoch": 0.53,
"learning_rate": 0.00023897693079237716,
"loss": 2.5559,
"theoretical_loss": 3.4711720288063592,
"tokens_seen": 1739063296
},
{
"epoch": 0.53,
"learning_rate": 0.00023896690070210634,
"loss": 2.8159,
"theoretical_loss": 3.4711608202806445,
"tokens_seen": 1739128832
},
{
"epoch": 0.53,
"learning_rate": 0.00023895687061183552,
"loss": 2.5849,
"theoretical_loss": 3.471149612295555,
"tokens_seen": 1739194368
},
{
"epoch": 0.53,
"learning_rate": 0.0002389468405215647,
"loss": 2.6903,
"theoretical_loss": 3.4711384048510445,
"tokens_seen": 1739259904
},
{
"epoch": 0.53,
"learning_rate": 0.00023893681043129388,
"loss": 2.6293,
"theoretical_loss": 3.471127197947066,
"tokens_seen": 1739325440
},
{
"epoch": 0.53,
"learning_rate": 0.0002389267803410231,
"loss": 2.7075,
"theoretical_loss": 3.4711159915835728,
"tokens_seen": 1739390976
},
{
"epoch": 0.53,
"learning_rate": 0.00023891675025075227,
"loss": 2.4489,
"theoretical_loss": 3.4711047857605193,
"tokens_seen": 1739456512
},
{
"epoch": 0.53,
"learning_rate": 0.00023890672016048145,
"loss": 2.6144,
"theoretical_loss": 3.4710935804778584,
"tokens_seen": 1739522048
},
{
"epoch": 0.53,
"learning_rate": 0.00023889669007021063,
"loss": 2.5538,
"theoretical_loss": 3.471082375735544,
"tokens_seen": 1739587584
},
{
"epoch": 0.53,
"learning_rate": 0.00023888665997993984,
"loss": 2.7575,
"theoretical_loss": 3.47107117153353,
"tokens_seen": 1739653120
},
{
"epoch": 0.53,
"learning_rate": 0.00023887662988966902,
"loss": 2.6229,
"theoretical_loss": 3.471059967871769,
"tokens_seen": 1739718656
},
{
"epoch": 0.53,
"learning_rate": 0.0002388665997993982,
"loss": 2.6294,
"theoretical_loss": 3.4710487647502157,
"tokens_seen": 1739784192
},
{
"epoch": 0.53,
"learning_rate": 0.00023885656970912739,
"loss": 2.6611,
"theoretical_loss": 3.4710375621688234,
"tokens_seen": 1739849728
},
{
"epoch": 0.53,
"learning_rate": 0.00023884653961885657,
"loss": 2.6304,
"theoretical_loss": 3.471026360127545,
"tokens_seen": 1739915264
},
{
"epoch": 0.53,
"objective/train/docs_used": 985100,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4600489139556885,
"objective/train/theoretical_loss": 3.471015158626335,
"objective/train/tokens_used": 1760440800,
"theoretical_loss": 3.471015158626335,
"tokens_seen": 1739980800
},
{
"epoch": 0.53,
"learning_rate": 0.00023883650952858577,
"loss": 2.6934,
"theoretical_loss": 3.471015158626335,
"tokens_seen": 1739980800
},
{
"epoch": 0.53,
"learning_rate": 0.00023882647943831496,
"loss": 2.6192,
"theoretical_loss": 3.471003957665147,
"tokens_seen": 1740046336
},
{
"epoch": 0.53,
"learning_rate": 0.00023881644934804414,
"loss": 2.554,
"theoretical_loss": 3.470992757243934,
"tokens_seen": 1740111872
},
{
"epoch": 0.53,
"learning_rate": 0.00023880641925777332,
"loss": 2.5372,
"theoretical_loss": 3.4709815573626495,
"tokens_seen": 1740177408
},
{
"epoch": 0.53,
"learning_rate": 0.00023879638916750253,
"loss": 2.4947,
"theoretical_loss": 3.4709703580212485,
"tokens_seen": 1740242944
},
{
"epoch": 0.53,
"learning_rate": 0.0002387863590772317,
"loss": 2.5695,
"theoretical_loss": 3.470959159219683,
"tokens_seen": 1740308480
},
{
"epoch": 0.53,
"learning_rate": 0.0002387763289869609,
"loss": 2.7221,
"theoretical_loss": 3.4709479609579077,
"tokens_seen": 1740374016
},
{
"epoch": 0.53,
"learning_rate": 0.00023876629889669007,
"loss": 2.6542,
"theoretical_loss": 3.470936763235876,
"tokens_seen": 1740439552
},
{
"epoch": 0.53,
"learning_rate": 0.00023875626880641925,
"loss": 2.6259,
"theoretical_loss": 3.470925566053541,
"tokens_seen": 1740505088
},
{
"epoch": 0.53,
"learning_rate": 0.00023874623871614846,
"loss": 2.7039,
"theoretical_loss": 3.4709143694108575,
"tokens_seen": 1740570624
},
{
"epoch": 0.53,
"learning_rate": 0.00023873620862587764,
"loss": 2.3996,
"theoretical_loss": 3.4709031733077786,
"tokens_seen": 1740636160
},
{
"epoch": 0.53,
"learning_rate": 0.00023872617853560682,
"loss": 2.5446,
"theoretical_loss": 3.4708919777442575,
"tokens_seen": 1740701696
},
{
"epoch": 0.53,
"learning_rate": 0.000238716148445336,
"loss": 2.5837,
"theoretical_loss": 3.470880782720249,
"tokens_seen": 1740767232
},
{
"epoch": 0.53,
"learning_rate": 0.0002387061183550652,
"loss": 2.5995,
"theoretical_loss": 3.470869588235705,
"tokens_seen": 1740832768
},
{
"epoch": 0.53,
"learning_rate": 0.0002386960882647944,
"loss": 2.7412,
"theoretical_loss": 3.4708583942905813,
"tokens_seen": 1740898304
},
{
"epoch": 0.53,
"learning_rate": 0.00023868605817452357,
"loss": 2.6707,
"theoretical_loss": 3.4708472008848297,
"tokens_seen": 1740963840
},
{
"epoch": 0.53,
"learning_rate": 0.00023867602808425275,
"loss": 2.7406,
"theoretical_loss": 3.4708360080184057,
"tokens_seen": 1741029376
},
{
"epoch": 0.53,
"learning_rate": 0.00023866599799398194,
"loss": 2.6958,
"theoretical_loss": 3.4708248156912616,
"tokens_seen": 1741094912
},
{
"epoch": 0.53,
"learning_rate": 0.00023865596790371114,
"loss": 2.5735,
"theoretical_loss": 3.470813623903352,
"tokens_seen": 1741160448
},
{
"epoch": 0.53,
"learning_rate": 0.00023864593781344032,
"loss": 2.7787,
"theoretical_loss": 3.47080243265463,
"tokens_seen": 1741225984
},
{
"epoch": 0.53,
"learning_rate": 0.0002386359077231695,
"loss": 2.389,
"theoretical_loss": 3.47079124194505,
"tokens_seen": 1741291520
},
{
"epoch": 0.53,
"learning_rate": 0.0002386258776328987,
"loss": 2.7003,
"theoretical_loss": 3.4707800517745646,
"tokens_seen": 1741357056
},
{
"epoch": 0.53,
"learning_rate": 0.0002386158475426279,
"loss": 2.5043,
"theoretical_loss": 3.470768862143129,
"tokens_seen": 1741422592
},
{
"epoch": 0.53,
"learning_rate": 0.00023860581745235708,
"loss": 2.4054,
"theoretical_loss": 3.470757673050696,
"tokens_seen": 1741488128
},
{
"epoch": 0.53,
"learning_rate": 0.00023859578736208626,
"loss": 2.6552,
"theoretical_loss": 3.4707464844972193,
"tokens_seen": 1741553664
},
{
"epoch": 0.53,
"objective/train/docs_used": 986266,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.474574327468872,
"objective/train/theoretical_loss": 3.470735296482653,
"objective/train/tokens_used": 1762079200,
"theoretical_loss": 3.470735296482653,
"tokens_seen": 1741619200
},
{
"epoch": 0.53,
"learning_rate": 0.00023858575727181544,
"loss": 2.5732,
"theoretical_loss": 3.470735296482653,
"tokens_seen": 1741619200
},
{
"epoch": 0.53,
"learning_rate": 0.00023857572718154462,
"loss": 2.5914,
"theoretical_loss": 3.470724109006951,
"tokens_seen": 1741684736
},
{
"epoch": 0.53,
"learning_rate": 0.00023856569709127383,
"loss": 2.6578,
"theoretical_loss": 3.470712922070067,
"tokens_seen": 1741750272
},
{
"epoch": 0.53,
"learning_rate": 0.000238555667001003,
"loss": 2.6299,
"theoretical_loss": 3.4707017356719545,
"tokens_seen": 1741815808
},
{
"epoch": 0.53,
"learning_rate": 0.0002385456369107322,
"loss": 2.6605,
"theoretical_loss": 3.4706905498125673,
"tokens_seen": 1741881344
},
{
"epoch": 0.53,
"learning_rate": 0.00023853560682046137,
"loss": 2.6859,
"theoretical_loss": 3.47067936449186,
"tokens_seen": 1741946880
},
{
"epoch": 0.53,
"learning_rate": 0.00023852557673019058,
"loss": 2.7075,
"theoretical_loss": 3.4706681797097847,
"tokens_seen": 1742012416
},
{
"epoch": 0.53,
"learning_rate": 0.00023851554663991976,
"loss": 2.8545,
"theoretical_loss": 3.470656995466297,
"tokens_seen": 1742077952
},
{
"epoch": 0.53,
"learning_rate": 0.00023850551654964894,
"loss": 2.7652,
"theoretical_loss": 3.47064581176135,
"tokens_seen": 1742143488
},
{
"epoch": 0.53,
"learning_rate": 0.00023849548645937812,
"loss": 2.3598,
"theoretical_loss": 3.470634628594897,
"tokens_seen": 1742209024
},
{
"epoch": 0.53,
"learning_rate": 0.00023848545636910733,
"loss": 2.8063,
"theoretical_loss": 3.4706234459668925,
"tokens_seen": 1742274560
},
{
"epoch": 0.53,
"learning_rate": 0.0002384754262788365,
"loss": 2.6488,
"theoretical_loss": 3.47061226387729,
"tokens_seen": 1742340096
},
{
"epoch": 0.53,
"learning_rate": 0.0002384653961885657,
"loss": 2.4778,
"theoretical_loss": 3.470601082326043,
"tokens_seen": 1742405632
},
{
"epoch": 0.53,
"learning_rate": 0.00023845536609829487,
"loss": 2.7038,
"theoretical_loss": 3.4705899013131063,
"tokens_seen": 1742471168
},
{
"epoch": 0.53,
"learning_rate": 0.00023844533600802406,
"loss": 2.5251,
"theoretical_loss": 3.470578720838433,
"tokens_seen": 1742536704
},
{
"epoch": 0.53,
"learning_rate": 0.00023843530591775326,
"loss": 2.4857,
"theoretical_loss": 3.4705675409019774,
"tokens_seen": 1742602240
},
{
"epoch": 0.53,
"learning_rate": 0.00023842527582748244,
"loss": 2.6103,
"theoretical_loss": 3.4705563615036925,
"tokens_seen": 1742667776
},
{
"epoch": 0.53,
"learning_rate": 0.00023841524573721163,
"loss": 2.5615,
"theoretical_loss": 3.4705451826435327,
"tokens_seen": 1742733312
},
{
"epoch": 0.53,
"learning_rate": 0.0002384052156469408,
"loss": 2.6219,
"theoretical_loss": 3.4705340043214523,
"tokens_seen": 1742798848
},
{
"epoch": 0.53,
"learning_rate": 0.00023839518555667002,
"loss": 2.6777,
"theoretical_loss": 3.4705228265374046,
"tokens_seen": 1742864384
},
{
"epoch": 0.53,
"learning_rate": 0.0002383851554663992,
"loss": 2.5764,
"theoretical_loss": 3.470511649291344,
"tokens_seen": 1742929920
},
{
"epoch": 0.53,
"learning_rate": 0.00023837512537612838,
"loss": 2.6464,
"theoretical_loss": 3.4705004725832236,
"tokens_seen": 1742995456
},
{
"epoch": 0.53,
"learning_rate": 0.00023836509528585756,
"loss": 2.6805,
"theoretical_loss": 3.470489296412998,
"tokens_seen": 1743060992
},
{
"epoch": 0.53,
"learning_rate": 0.00023835506519558677,
"loss": 2.5805,
"theoretical_loss": 3.470478120780621,
"tokens_seen": 1743126528
},
{
"epoch": 0.53,
"learning_rate": 0.00023834503510531598,
"loss": 2.3771,
"theoretical_loss": 3.4704669456860455,
"tokens_seen": 1743192064
},
{
"epoch": 0.53,
"objective/train/docs_used": 986906,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5081822872161865,
"objective/train/theoretical_loss": 3.4704557711292265,
"objective/train/tokens_used": 1763717600,
"theoretical_loss": 3.4704557711292265,
"tokens_seen": 1743257600
},
{
"epoch": 0.53,
"learning_rate": 0.00023833500501504516,
"loss": 2.5734,
"theoretical_loss": 3.4704557711292265,
"tokens_seen": 1743257600
},
{
"epoch": 0.53,
"learning_rate": 0.00023832497492477434,
"loss": 2.7148,
"theoretical_loss": 3.470444597110118,
"tokens_seen": 1743323136
},
{
"epoch": 0.53,
"learning_rate": 0.00023831494483450352,
"loss": 2.6146,
"theoretical_loss": 3.4704334236286734,
"tokens_seen": 1743388672
},
{
"epoch": 0.53,
"learning_rate": 0.00023830491474423273,
"loss": 2.738,
"theoretical_loss": 3.4704222506848463,
"tokens_seen": 1743454208
},
{
"epoch": 0.53,
"learning_rate": 0.0002382948846539619,
"loss": 2.7572,
"theoretical_loss": 3.470411078278591,
"tokens_seen": 1743519744
},
{
"epoch": 0.53,
"learning_rate": 0.0002382848545636911,
"loss": 2.7853,
"theoretical_loss": 3.4703999064098623,
"tokens_seen": 1743585280
},
{
"epoch": 0.53,
"learning_rate": 0.00023827482447342027,
"loss": 2.4453,
"theoretical_loss": 3.470388735078613,
"tokens_seen": 1743650816
},
{
"epoch": 0.53,
"learning_rate": 0.00023826479438314945,
"loss": 2.6632,
"theoretical_loss": 3.4703775642847967,
"tokens_seen": 1743716352
},
{
"epoch": 0.53,
"learning_rate": 0.00023825476429287866,
"loss": 2.6405,
"theoretical_loss": 3.4703663940283684,
"tokens_seen": 1743781888
},
{
"epoch": 0.53,
"learning_rate": 0.00023824473420260784,
"loss": 2.4327,
"theoretical_loss": 3.470355224309282,
"tokens_seen": 1743847424
},
{
"epoch": 0.53,
"learning_rate": 0.00023823470411233702,
"loss": 2.6872,
"theoretical_loss": 3.470344055127491,
"tokens_seen": 1743912960
},
{
"epoch": 0.53,
"learning_rate": 0.0002382246740220662,
"loss": 2.7044,
"theoretical_loss": 3.4703328864829492,
"tokens_seen": 1743978496
},
{
"epoch": 0.53,
"learning_rate": 0.0002382146439317954,
"loss": 2.6411,
"theoretical_loss": 3.4703217183756108,
"tokens_seen": 1744044032
},
{
"epoch": 0.53,
"learning_rate": 0.0002382046138415246,
"loss": 2.3921,
"theoretical_loss": 3.47031055080543,
"tokens_seen": 1744109568
},
{
"epoch": 0.53,
"learning_rate": 0.00023819458375125377,
"loss": 2.5986,
"theoretical_loss": 3.470299383772361,
"tokens_seen": 1744175104
},
{
"epoch": 0.53,
"learning_rate": 0.00023818455366098295,
"loss": 2.6162,
"theoretical_loss": 3.470288217276357,
"tokens_seen": 1744240640
},
{
"epoch": 0.53,
"learning_rate": 0.00023817452357071214,
"loss": 2.606,
"theoretical_loss": 3.4702770513173724,
"tokens_seen": 1744306176
},
{
"epoch": 0.53,
"learning_rate": 0.00023816449348044134,
"loss": 2.7394,
"theoretical_loss": 3.470265885895361,
"tokens_seen": 1744371712
},
{
"epoch": 0.53,
"learning_rate": 0.00023815446339017052,
"loss": 2.573,
"theoretical_loss": 3.4702547210102774,
"tokens_seen": 1744437248
},
{
"epoch": 0.53,
"learning_rate": 0.0002381444332998997,
"loss": 2.7477,
"theoretical_loss": 3.470243556662075,
"tokens_seen": 1744502784
},
{
"epoch": 0.53,
"learning_rate": 0.0002381344032096289,
"loss": 2.6379,
"theoretical_loss": 3.470232392850708,
"tokens_seen": 1744568320
},
{
"epoch": 0.53,
"learning_rate": 0.0002381243731193581,
"loss": 2.5622,
"theoretical_loss": 3.470221229576131,
"tokens_seen": 1744633856
},
{
"epoch": 0.53,
"learning_rate": 0.00023811434302908728,
"loss": 2.4686,
"theoretical_loss": 3.4702100668382965,
"tokens_seen": 1744699392
},
{
"epoch": 0.53,
"learning_rate": 0.00023810431293881646,
"loss": 2.6195,
"theoretical_loss": 3.47019890463716,
"tokens_seen": 1744764928
},
{
"epoch": 0.53,
"learning_rate": 0.00023809428284854564,
"loss": 2.6679,
"theoretical_loss": 3.470187742972675,
"tokens_seen": 1744830464
},
{
"epoch": 0.53,
"objective/train/docs_used": 987618,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6582839488983154,
"objective/train/theoretical_loss": 3.470176581844796,
"objective/train/tokens_used": 1765356000,
"theoretical_loss": 3.470176581844796,
"tokens_seen": 1744896000
},
{
"epoch": 0.53,
"learning_rate": 0.00023808425275827482,
"loss": 2.5607,
"theoretical_loss": 3.470176581844796,
"tokens_seen": 1744896000
},
{
"epoch": 0.53,
"learning_rate": 0.00023807422266800403,
"loss": 2.4466,
"theoretical_loss": 3.4701654212534763,
"tokens_seen": 1744961536
},
{
"epoch": 0.53,
"learning_rate": 0.0002380641925777332,
"loss": 2.4177,
"theoretical_loss": 3.4701542611986698,
"tokens_seen": 1745027072
},
{
"epoch": 0.53,
"learning_rate": 0.0002380541624874624,
"loss": 2.591,
"theoretical_loss": 3.470143101680332,
"tokens_seen": 1745092608
},
{
"epoch": 0.53,
"learning_rate": 0.00023804413239719157,
"loss": 2.5331,
"theoretical_loss": 3.4701319426984156,
"tokens_seen": 1745158144
},
{
"epoch": 0.53,
"learning_rate": 0.00023803410230692078,
"loss": 2.6922,
"theoretical_loss": 3.4701207842528747,
"tokens_seen": 1745223680
},
{
"epoch": 0.53,
"learning_rate": 0.00023802407221664996,
"loss": 2.6048,
"theoretical_loss": 3.470109626343664,
"tokens_seen": 1745289216
},
{
"epoch": 0.53,
"learning_rate": 0.00023801404212637914,
"loss": 2.6707,
"theoretical_loss": 3.470098468970738,
"tokens_seen": 1745354752
},
{
"epoch": 0.53,
"learning_rate": 0.00023800401203610832,
"loss": 2.6944,
"theoretical_loss": 3.470087312134049,
"tokens_seen": 1745420288
},
{
"epoch": 0.53,
"learning_rate": 0.00023799398194583753,
"loss": 2.4416,
"theoretical_loss": 3.470076155833553,
"tokens_seen": 1745485824
},
{
"epoch": 0.53,
"learning_rate": 0.0002379839518555667,
"loss": 2.5548,
"theoretical_loss": 3.4700650000692033,
"tokens_seen": 1745551360
},
{
"epoch": 0.53,
"learning_rate": 0.0002379739217652959,
"loss": 2.6325,
"theoretical_loss": 3.4700538448409537,
"tokens_seen": 1745616896
},
{
"epoch": 0.53,
"learning_rate": 0.00023796389167502507,
"loss": 2.6339,
"theoretical_loss": 3.470042690148759,
"tokens_seen": 1745682432
},
{
"epoch": 0.53,
"learning_rate": 0.00023795386158475426,
"loss": 2.6304,
"theoretical_loss": 3.4700315359925726,
"tokens_seen": 1745747968
},
{
"epoch": 0.53,
"learning_rate": 0.00023794383149448346,
"loss": 2.6981,
"theoretical_loss": 3.470020382372349,
"tokens_seen": 1745813504
},
{
"epoch": 0.53,
"learning_rate": 0.00023793380140421265,
"loss": 2.544,
"theoretical_loss": 3.470009229288043,
"tokens_seen": 1745879040
},
{
"epoch": 0.53,
"learning_rate": 0.00023792377131394183,
"loss": 2.7768,
"theoretical_loss": 3.469998076739607,
"tokens_seen": 1745944576
},
{
"epoch": 0.53,
"learning_rate": 0.000237913741223671,
"loss": 2.4734,
"theoretical_loss": 3.469986924726997,
"tokens_seen": 1746010112
},
{
"epoch": 0.53,
"learning_rate": 0.00023790371113340022,
"loss": 2.7293,
"theoretical_loss": 3.469975773250166,
"tokens_seen": 1746075648
},
{
"epoch": 0.53,
"learning_rate": 0.0002378936810431294,
"loss": 2.5361,
"theoretical_loss": 3.4699646223090688,
"tokens_seen": 1746141184
},
{
"epoch": 0.53,
"learning_rate": 0.00023788365095285858,
"loss": 2.5805,
"theoretical_loss": 3.469953471903659,
"tokens_seen": 1746206720
},
{
"epoch": 0.53,
"learning_rate": 0.00023787362086258776,
"loss": 2.6594,
"theoretical_loss": 3.469942322033891,
"tokens_seen": 1746272256
},
{
"epoch": 0.53,
"learning_rate": 0.00023786359077231694,
"loss": 2.6786,
"theoretical_loss": 3.4699311726997193,
"tokens_seen": 1746337792
},
{
"epoch": 0.53,
"learning_rate": 0.00023785356068204615,
"loss": 2.3529,
"theoretical_loss": 3.469920023901097,
"tokens_seen": 1746403328
},
{
"epoch": 0.53,
"learning_rate": 0.00023784353059177533,
"loss": 2.5798,
"theoretical_loss": 3.46990887563798,
"tokens_seen": 1746468864
},
{
"epoch": 0.53,
"objective/train/docs_used": 989102,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.751021385192871,
"objective/train/theoretical_loss": 3.469897727910321,
"objective/train/tokens_used": 1766994400,
"theoretical_loss": 3.469897727910321,
"tokens_seen": 1746534400
},
{
"epoch": 0.53,
"learning_rate": 0.0002378335005015045,
"loss": 2.8258,
"theoretical_loss": 3.469897727910321,
"tokens_seen": 1746534400
},
{
"epoch": 0.53,
"learning_rate": 0.0002378234704112337,
"loss": 2.633,
"theoretical_loss": 3.4698865807180748,
"tokens_seen": 1746599936
},
{
"epoch": 0.53,
"learning_rate": 0.0002378134403209629,
"loss": 2.6394,
"theoretical_loss": 3.4698754340611955,
"tokens_seen": 1746665472
},
{
"epoch": 0.53,
"learning_rate": 0.00023780341023069208,
"loss": 2.4925,
"theoretical_loss": 3.4698642879396373,
"tokens_seen": 1746731008
},
{
"epoch": 0.53,
"learning_rate": 0.00023779338014042126,
"loss": 2.5172,
"theoretical_loss": 3.4698531423533545,
"tokens_seen": 1746796544
},
{
"epoch": 0.53,
"learning_rate": 0.00023778335005015044,
"loss": 2.405,
"theoretical_loss": 3.469841997302301,
"tokens_seen": 1746862080
},
{
"epoch": 0.53,
"learning_rate": 0.00023777331995987962,
"loss": 2.6185,
"theoretical_loss": 3.4698308527864317,
"tokens_seen": 1746927616
},
{
"epoch": 0.53,
"learning_rate": 0.00023776328986960883,
"loss": 2.7341,
"theoretical_loss": 3.4698197088056997,
"tokens_seen": 1746993152
},
{
"epoch": 0.53,
"learning_rate": 0.00023775325977933801,
"loss": 2.4654,
"theoretical_loss": 3.4698085653600605,
"tokens_seen": 1747058688
},
{
"epoch": 0.53,
"learning_rate": 0.0002377432296890672,
"loss": 2.6347,
"theoretical_loss": 3.469797422449467,
"tokens_seen": 1747124224
},
{
"epoch": 0.53,
"learning_rate": 0.00023773319959879638,
"loss": 2.7809,
"theoretical_loss": 3.4697862800738752,
"tokens_seen": 1747189760
},
{
"epoch": 0.53,
"learning_rate": 0.00023772316950852558,
"loss": 2.6746,
"theoretical_loss": 3.4697751382332376,
"tokens_seen": 1747255296
},
{
"epoch": 0.53,
"learning_rate": 0.00023771313941825477,
"loss": 2.6547,
"theoretical_loss": 3.46976399692751,
"tokens_seen": 1747320832
},
{
"epoch": 0.53,
"learning_rate": 0.00023770310932798395,
"loss": 2.5144,
"theoretical_loss": 3.469752856156645,
"tokens_seen": 1747386368
},
{
"epoch": 0.53,
"learning_rate": 0.00023769307923771313,
"loss": 2.4836,
"theoretical_loss": 3.469741715920598,
"tokens_seen": 1747451904
},
{
"epoch": 0.53,
"learning_rate": 0.0002376830491474423,
"loss": 2.5527,
"theoretical_loss": 3.4697305762193227,
"tokens_seen": 1747517440
},
{
"epoch": 0.53,
"learning_rate": 0.00023767301905717152,
"loss": 2.5179,
"theoretical_loss": 3.469719437052774,
"tokens_seen": 1747582976
},
{
"epoch": 0.53,
"learning_rate": 0.0002376629889669007,
"loss": 2.6635,
"theoretical_loss": 3.4697082984209056,
"tokens_seen": 1747648512
},
{
"epoch": 0.53,
"learning_rate": 0.00023765295887662988,
"loss": 2.7247,
"theoretical_loss": 3.469697160323672,
"tokens_seen": 1747714048
},
{
"epoch": 0.53,
"learning_rate": 0.00023764292878635906,
"loss": 2.644,
"theoretical_loss": 3.469686022761028,
"tokens_seen": 1747779584
},
{
"epoch": 0.53,
"learning_rate": 0.00023763289869608827,
"loss": 2.5844,
"theoretical_loss": 3.469674885732927,
"tokens_seen": 1747845120
},
{
"epoch": 0.53,
"learning_rate": 0.00023762286860581745,
"loss": 2.5977,
"theoretical_loss": 3.4696637492393236,
"tokens_seen": 1747910656
},
{
"epoch": 0.53,
"learning_rate": 0.00023761283851554663,
"loss": 2.5192,
"theoretical_loss": 3.4696526132801724,
"tokens_seen": 1747976192
},
{
"epoch": 0.53,
"learning_rate": 0.00023760280842527584,
"loss": 2.5404,
"theoretical_loss": 3.4696414778554274,
"tokens_seen": 1748041728
},
{
"epoch": 0.53,
"learning_rate": 0.00023759277833500502,
"loss": 2.7305,
"theoretical_loss": 3.4696303429650435,
"tokens_seen": 1748107264
},
{
"epoch": 0.53,
"objective/train/docs_used": 989762,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6953930854797363,
"objective/train/theoretical_loss": 3.469619208608974,
"objective/train/tokens_used": 1768632800,
"theoretical_loss": 3.469619208608974,
"tokens_seen": 1748172800
},
{
"epoch": 0.53,
"learning_rate": 0.00023758274824473423,
"loss": 2.5915,
"theoretical_loss": 3.469619208608974,
"tokens_seen": 1748172800
},
{
"epoch": 0.53,
"learning_rate": 0.0002375727181544634,
"loss": 2.6638,
"theoretical_loss": 3.469608074787174,
"tokens_seen": 1748238336
},
{
"epoch": 0.53,
"learning_rate": 0.0002375626880641926,
"loss": 2.7096,
"theoretical_loss": 3.4695969414995975,
"tokens_seen": 1748303872
},
{
"epoch": 0.53,
"learning_rate": 0.00023755265797392177,
"loss": 2.6085,
"theoretical_loss": 3.4695858087461993,
"tokens_seen": 1748369408
},
{
"epoch": 0.53,
"learning_rate": 0.00023754262788365098,
"loss": 2.4729,
"theoretical_loss": 3.469574676526933,
"tokens_seen": 1748434944
},
{
"epoch": 0.53,
"learning_rate": 0.00023753259779338016,
"loss": 2.4534,
"theoretical_loss": 3.469563544841754,
"tokens_seen": 1748500480
},
{
"epoch": 0.53,
"learning_rate": 0.00023752256770310934,
"loss": 2.5975,
"theoretical_loss": 3.4695524136906153,
"tokens_seen": 1748566016
},
{
"epoch": 0.53,
"learning_rate": 0.00023751253761283852,
"loss": 2.5878,
"theoretical_loss": 3.4695412830734726,
"tokens_seen": 1748631552
},
{
"epoch": 0.53,
"learning_rate": 0.00023750250752256773,
"loss": 2.6271,
"theoretical_loss": 3.46953015299028,
"tokens_seen": 1748697088
},
{
"epoch": 0.53,
"learning_rate": 0.0002374924774322969,
"loss": 2.4962,
"theoretical_loss": 3.469519023440991,
"tokens_seen": 1748762624
},
{
"epoch": 0.53,
"learning_rate": 0.0002374824473420261,
"loss": 2.4911,
"theoretical_loss": 3.4695078944255604,
"tokens_seen": 1748828160
},
{
"epoch": 0.53,
"learning_rate": 0.00023747241725175527,
"loss": 2.5157,
"theoretical_loss": 3.4694967659439433,
"tokens_seen": 1748893696
},
{
"epoch": 0.53,
"learning_rate": 0.00023746238716148446,
"loss": 2.7311,
"theoretical_loss": 3.469485637996093,
"tokens_seen": 1748959232
},
{
"epoch": 0.53,
"learning_rate": 0.00023745235707121366,
"loss": 2.8108,
"theoretical_loss": 3.4694745105819647,
"tokens_seen": 1749024768
},
{
"epoch": 0.53,
"learning_rate": 0.00023744232698094285,
"loss": 2.6555,
"theoretical_loss": 3.4694633837015125,
"tokens_seen": 1749090304
},
{
"epoch": 0.53,
"learning_rate": 0.00023743229689067203,
"loss": 2.4898,
"theoretical_loss": 3.469452257354691,
"tokens_seen": 1749155840
},
{
"epoch": 0.53,
"learning_rate": 0.0002374222668004012,
"loss": 2.5988,
"theoretical_loss": 3.469441131541454,
"tokens_seen": 1749221376
},
{
"epoch": 0.53,
"learning_rate": 0.00023741223671013042,
"loss": 2.5295,
"theoretical_loss": 3.4694300062617565,
"tokens_seen": 1749286912
},
{
"epoch": 0.53,
"learning_rate": 0.0002374022066198596,
"loss": 2.5749,
"theoretical_loss": 3.469418881515553,
"tokens_seen": 1749352448
},
{
"epoch": 0.53,
"learning_rate": 0.00023739217652958878,
"loss": 2.9024,
"theoretical_loss": 3.4694077573027977,
"tokens_seen": 1749417984
},
{
"epoch": 0.53,
"learning_rate": 0.00023738214643931796,
"loss": 2.6724,
"theoretical_loss": 3.469396633623445,
"tokens_seen": 1749483520
},
{
"epoch": 0.53,
"learning_rate": 0.00023737211634904714,
"loss": 2.4886,
"theoretical_loss": 3.4693855104774496,
"tokens_seen": 1749549056
},
{
"epoch": 0.53,
"learning_rate": 0.00023736208625877635,
"loss": 2.5911,
"theoretical_loss": 3.4693743878647654,
"tokens_seen": 1749614592
},
{
"epoch": 0.53,
"learning_rate": 0.00023735205616850553,
"loss": 2.2488,
"theoretical_loss": 3.469363265785348,
"tokens_seen": 1749680128
},
{
"epoch": 0.53,
"learning_rate": 0.0002373420260782347,
"loss": 2.5953,
"theoretical_loss": 3.46935214423915,
"tokens_seen": 1749745664
},
{
"epoch": 0.53,
"objective/train/docs_used": 990903,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.657815933227539,
"objective/train/theoretical_loss": 3.4693410232261277,
"objective/train/tokens_used": 1770271200,
"theoretical_loss": 3.4693410232261277,
"tokens_seen": 1749811200
},
{
"epoch": 0.53,
"learning_rate": 0.0002373319959879639,
"loss": 2.4783,
"theoretical_loss": 3.4693410232261277,
"tokens_seen": 1749811200
},
{
"epoch": 0.53,
"learning_rate": 0.0002373219658976931,
"loss": 2.6892,
"theoretical_loss": 3.4693299027462343,
"tokens_seen": 1749876736
},
{
"epoch": 0.53,
"learning_rate": 0.00023731193580742228,
"loss": 2.7501,
"theoretical_loss": 3.469318782799425,
"tokens_seen": 1749942272
},
{
"epoch": 0.53,
"learning_rate": 0.00023730190571715146,
"loss": 2.6358,
"theoretical_loss": 3.4693076633856545,
"tokens_seen": 1750007808
},
{
"epoch": 0.53,
"learning_rate": 0.00023729187562688064,
"loss": 2.6305,
"theoretical_loss": 3.4692965445048767,
"tokens_seen": 1750073344
},
{
"epoch": 0.53,
"learning_rate": 0.00023728184553660982,
"loss": 2.761,
"theoretical_loss": 3.469285426157046,
"tokens_seen": 1750138880
},
{
"epoch": 0.53,
"learning_rate": 0.00023727181544633903,
"loss": 2.5775,
"theoretical_loss": 3.469274308342117,
"tokens_seen": 1750204416
},
{
"epoch": 0.53,
"learning_rate": 0.00023726178535606821,
"loss": 2.8759,
"theoretical_loss": 3.469263191060045,
"tokens_seen": 1750269952
},
{
"epoch": 0.53,
"learning_rate": 0.0002372517552657974,
"loss": 2.5375,
"theoretical_loss": 3.4692520743107833,
"tokens_seen": 1750335488
},
{
"epoch": 0.53,
"learning_rate": 0.00023724172517552658,
"loss": 2.6293,
"theoretical_loss": 3.469240958094287,
"tokens_seen": 1750401024
},
{
"epoch": 0.53,
"learning_rate": 0.00023723169508525578,
"loss": 2.6242,
"theoretical_loss": 3.469229842410511,
"tokens_seen": 1750466560
},
{
"epoch": 0.53,
"learning_rate": 0.00023722166499498497,
"loss": 2.7694,
"theoretical_loss": 3.4692187272594093,
"tokens_seen": 1750532096
},
{
"epoch": 0.53,
"learning_rate": 0.00023721163490471415,
"loss": 2.5185,
"theoretical_loss": 3.469207612640936,
"tokens_seen": 1750597632
},
{
"epoch": 0.53,
"learning_rate": 0.00023720160481444333,
"loss": 2.5357,
"theoretical_loss": 3.469196498555047,
"tokens_seen": 1750663168
},
{
"epoch": 0.53,
"learning_rate": 0.0002371915747241725,
"loss": 2.6252,
"theoretical_loss": 3.469185385001696,
"tokens_seen": 1750728704
},
{
"epoch": 0.53,
"learning_rate": 0.00023718154463390172,
"loss": 2.713,
"theoretical_loss": 3.4691742719808376,
"tokens_seen": 1750794240
},
{
"epoch": 0.53,
"learning_rate": 0.0002371715145436309,
"loss": 2.5305,
"theoretical_loss": 3.469163159492426,
"tokens_seen": 1750859776
},
{
"epoch": 0.53,
"learning_rate": 0.00023716148445336008,
"loss": 2.7505,
"theoretical_loss": 3.469152047536416,
"tokens_seen": 1750925312
},
{
"epoch": 0.53,
"learning_rate": 0.00023715145436308926,
"loss": 2.727,
"theoretical_loss": 3.4691409361127628,
"tokens_seen": 1750990848
},
{
"epoch": 0.53,
"learning_rate": 0.00023714142427281847,
"loss": 2.6005,
"theoretical_loss": 3.46912982522142,
"tokens_seen": 1751056384
},
{
"epoch": 0.53,
"learning_rate": 0.00023713139418254765,
"loss": 2.6032,
"theoretical_loss": 3.469118714862343,
"tokens_seen": 1751121920
},
{
"epoch": 0.53,
"learning_rate": 0.00023712136409227683,
"loss": 2.8289,
"theoretical_loss": 3.469107605035486,
"tokens_seen": 1751187456
},
{
"epoch": 0.53,
"learning_rate": 0.000237111334002006,
"loss": 2.5104,
"theoretical_loss": 3.4690964957408035,
"tokens_seen": 1751252992
},
{
"epoch": 0.53,
"learning_rate": 0.0002371013039117352,
"loss": 2.3858,
"theoretical_loss": 3.46908538697825,
"tokens_seen": 1751318528
},
{
"epoch": 0.53,
"learning_rate": 0.0002370912738214644,
"loss": 2.3919,
"theoretical_loss": 3.4690742787477804,
"tokens_seen": 1751384064
},
{
"epoch": 0.53,
"objective/train/docs_used": 991641,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.490144729614258,
"objective/train/theoretical_loss": 3.4690631710493496,
"objective/train/tokens_used": 1771909600,
"theoretical_loss": 3.4690631710493496,
"tokens_seen": 1751449600
},
{
"epoch": 0.53,
"learning_rate": 0.00023708124373119358,
"loss": 2.6499,
"theoretical_loss": 3.4690631710493496,
"tokens_seen": 1751449600
},
{
"epoch": 0.53,
"learning_rate": 0.00023707121364092276,
"loss": 2.7306,
"theoretical_loss": 3.469052063882912,
"tokens_seen": 1751515136
},
{
"epoch": 0.53,
"learning_rate": 0.00023706118355065194,
"loss": 2.7702,
"theoretical_loss": 3.4690409572484215,
"tokens_seen": 1751580672
},
{
"epoch": 0.53,
"learning_rate": 0.00023705115346038115,
"loss": 2.4842,
"theoretical_loss": 3.469029851145833,
"tokens_seen": 1751646208
},
{
"epoch": 0.53,
"learning_rate": 0.00023704112337011033,
"loss": 2.6492,
"theoretical_loss": 3.469018745575102,
"tokens_seen": 1751711744
},
{
"epoch": 0.53,
"learning_rate": 0.00023703109327983952,
"loss": 2.5671,
"theoretical_loss": 3.469007640536182,
"tokens_seen": 1751777280
},
{
"epoch": 0.53,
"learning_rate": 0.0002370210631895687,
"loss": 2.5597,
"theoretical_loss": 3.4689965360290285,
"tokens_seen": 1751842816
},
{
"epoch": 0.53,
"learning_rate": 0.00023701103309929788,
"loss": 2.5239,
"theoretical_loss": 3.4689854320535956,
"tokens_seen": 1751908352
},
{
"epoch": 0.53,
"learning_rate": 0.00023700100300902709,
"loss": 2.5087,
"theoretical_loss": 3.4689743286098382,
"tokens_seen": 1751973888
},
{
"epoch": 0.53,
"learning_rate": 0.00023699097291875627,
"loss": 2.5755,
"theoretical_loss": 3.468963225697711,
"tokens_seen": 1752039424
},
{
"epoch": 0.53,
"learning_rate": 0.00023698094282848545,
"loss": 2.455,
"theoretical_loss": 3.4689521233171687,
"tokens_seen": 1752104960
},
{
"epoch": 0.53,
"learning_rate": 0.00023697091273821463,
"loss": 2.6717,
"theoretical_loss": 3.468941021468166,
"tokens_seen": 1752170496
},
{
"epoch": 0.53,
"learning_rate": 0.00023696088264794384,
"loss": 2.5451,
"theoretical_loss": 3.468929920150657,
"tokens_seen": 1752236032
},
{
"epoch": 0.53,
"learning_rate": 0.00023695085255767302,
"loss": 2.6476,
"theoretical_loss": 3.468918819364597,
"tokens_seen": 1752301568
},
{
"epoch": 0.53,
"learning_rate": 0.0002369408224674022,
"loss": 2.6612,
"theoretical_loss": 3.4689077191099402,
"tokens_seen": 1752367104
},
{
"epoch": 0.53,
"learning_rate": 0.00023693079237713138,
"loss": 2.5387,
"theoretical_loss": 3.4688966193866415,
"tokens_seen": 1752432640
},
{
"epoch": 0.53,
"learning_rate": 0.00023692076228686056,
"loss": 2.5906,
"theoretical_loss": 3.468885520194656,
"tokens_seen": 1752498176
},
{
"epoch": 0.53,
"learning_rate": 0.00023691073219658977,
"loss": 2.7705,
"theoretical_loss": 3.468874421533938,
"tokens_seen": 1752563712
},
{
"epoch": 0.53,
"learning_rate": 0.00023690070210631895,
"loss": 2.6656,
"theoretical_loss": 3.4688633234044426,
"tokens_seen": 1752629248
},
{
"epoch": 0.53,
"learning_rate": 0.00023689067201604813,
"loss": 2.6011,
"theoretical_loss": 3.468852225806123,
"tokens_seen": 1752694784
},
{
"epoch": 0.53,
"learning_rate": 0.0002368806419257773,
"loss": 2.4938,
"theoretical_loss": 3.4688411287389362,
"tokens_seen": 1752760320
},
{
"epoch": 0.53,
"learning_rate": 0.00023687061183550652,
"loss": 2.4159,
"theoretical_loss": 3.4688300322028356,
"tokens_seen": 1752825856
},
{
"epoch": 0.53,
"learning_rate": 0.00023686058174523573,
"loss": 2.5605,
"theoretical_loss": 3.468818936197776,
"tokens_seen": 1752891392
},
{
"epoch": 0.53,
"learning_rate": 0.0002368505516549649,
"loss": 2.6808,
"theoretical_loss": 3.4688078407237124,
"tokens_seen": 1752956928
},
{
"epoch": 0.53,
"learning_rate": 0.0002368405215646941,
"loss": 2.789,
"theoretical_loss": 3.4687967457805993,
"tokens_seen": 1753022464
},
{
"epoch": 0.53,
"objective/train/docs_used": 992944,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.461217164993286,
"objective/train/theoretical_loss": 3.4687856513683917,
"objective/train/tokens_used": 1773548000,
"theoretical_loss": 3.4687856513683917,
"tokens_seen": 1753088000
},
{
"epoch": 0.53,
"learning_rate": 0.0002368304914744233,
"loss": 2.7029,
"theoretical_loss": 3.4687856513683917,
"tokens_seen": 1753088000
},
{
"epoch": 0.53,
"learning_rate": 0.00023682046138415248,
"loss": 2.6477,
"theoretical_loss": 3.468774557487044,
"tokens_seen": 1753153536
},
{
"epoch": 0.53,
"learning_rate": 0.00023681043129388166,
"loss": 2.561,
"theoretical_loss": 3.4687634641365115,
"tokens_seen": 1753219072
},
{
"epoch": 0.53,
"learning_rate": 0.00023680040120361084,
"loss": 2.6093,
"theoretical_loss": 3.4687523713167483,
"tokens_seen": 1753284608
},
{
"epoch": 0.53,
"learning_rate": 0.00023679037111334002,
"loss": 2.6747,
"theoretical_loss": 3.468741279027709,
"tokens_seen": 1753350144
},
{
"epoch": 0.53,
"learning_rate": 0.00023678034102306923,
"loss": 2.6176,
"theoretical_loss": 3.4687301872693497,
"tokens_seen": 1753415680
},
{
"epoch": 0.53,
"learning_rate": 0.00023677031093279841,
"loss": 2.5935,
"theoretical_loss": 3.4687190960416245,
"tokens_seen": 1753481216
},
{
"epoch": 0.53,
"learning_rate": 0.0002367602808425276,
"loss": 2.547,
"theoretical_loss": 3.468708005344487,
"tokens_seen": 1753546752
},
{
"epoch": 0.53,
"learning_rate": 0.00023675025075225678,
"loss": 2.7267,
"theoretical_loss": 3.468696915177894,
"tokens_seen": 1753612288
},
{
"epoch": 0.53,
"learning_rate": 0.00023674022066198598,
"loss": 2.7836,
"theoretical_loss": 3.4686858255417987,
"tokens_seen": 1753677824
},
{
"epoch": 0.53,
"learning_rate": 0.00023673019057171517,
"loss": 2.415,
"theoretical_loss": 3.4686747364361565,
"tokens_seen": 1753743360
},
{
"epoch": 0.53,
"learning_rate": 0.00023672016048144435,
"loss": 2.6255,
"theoretical_loss": 3.4686636478609225,
"tokens_seen": 1753808896
},
{
"epoch": 0.53,
"learning_rate": 0.00023671013039117353,
"loss": 2.7422,
"theoretical_loss": 3.4686525598160514,
"tokens_seen": 1753874432
},
{
"epoch": 0.53,
"learning_rate": 0.0002367001003009027,
"loss": 2.4982,
"theoretical_loss": 3.4686414723014973,
"tokens_seen": 1753939968
},
{
"epoch": 0.53,
"learning_rate": 0.00023669007021063192,
"loss": 2.8011,
"theoretical_loss": 3.4686303853172165,
"tokens_seen": 1754005504
},
{
"epoch": 0.53,
"learning_rate": 0.0002366800401203611,
"loss": 2.8274,
"theoretical_loss": 3.468619298863162,
"tokens_seen": 1754071040
},
{
"epoch": 0.53,
"learning_rate": 0.00023667001003009028,
"loss": 2.6764,
"theoretical_loss": 3.46860821293929,
"tokens_seen": 1754136576
},
{
"epoch": 0.53,
"learning_rate": 0.00023665997993981946,
"loss": 2.8201,
"theoretical_loss": 3.4685971275455545,
"tokens_seen": 1754202112
},
{
"epoch": 0.53,
"learning_rate": 0.00023664994984954867,
"loss": 2.7517,
"theoretical_loss": 3.4685860426819106,
"tokens_seen": 1754267648
},
{
"epoch": 0.53,
"learning_rate": 0.00023663991975927785,
"loss": 2.6853,
"theoretical_loss": 3.468574958348314,
"tokens_seen": 1754333184
},
{
"epoch": 0.53,
"learning_rate": 0.00023662988966900703,
"loss": 2.7877,
"theoretical_loss": 3.468563874544718,
"tokens_seen": 1754398720
},
{
"epoch": 0.53,
"learning_rate": 0.0002366198595787362,
"loss": 2.8101,
"theoretical_loss": 3.4685527912710787,
"tokens_seen": 1754464256
},
{
"epoch": 0.53,
"learning_rate": 0.0002366098294884654,
"loss": 2.7018,
"theoretical_loss": 3.4685417085273507,
"tokens_seen": 1754529792
},
{
"epoch": 0.53,
"learning_rate": 0.0002365997993981946,
"loss": 2.6098,
"theoretical_loss": 3.468530626313488,
"tokens_seen": 1754595328
},
{
"epoch": 0.53,
"learning_rate": 0.00023658976930792378,
"loss": 2.5262,
"theoretical_loss": 3.468519544629447,
"tokens_seen": 1754660864
},
{
"epoch": 0.53,
"objective/train/docs_used": 993452,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3736958503723145,
"objective/train/theoretical_loss": 3.468508463475181,
"objective/train/tokens_used": 1775186400,
"theoretical_loss": 3.468508463475181,
"tokens_seen": 1754726400
},
{
"epoch": 0.53,
"learning_rate": 0.00023657973921765296,
"loss": 2.5631,
"theoretical_loss": 3.468508463475181,
"tokens_seen": 1754726400
},
{
"epoch": 0.53,
"learning_rate": 0.00023656970912738214,
"loss": 2.6177,
"theoretical_loss": 3.4684973828506465,
"tokens_seen": 1754791936
},
{
"epoch": 0.53,
"learning_rate": 0.00023655967903711135,
"loss": 2.463,
"theoretical_loss": 3.468486302755797,
"tokens_seen": 1754857472
},
{
"epoch": 0.53,
"learning_rate": 0.00023654964894684053,
"loss": 2.7462,
"theoretical_loss": 3.4684752231905875,
"tokens_seen": 1754923008
},
{
"epoch": 0.53,
"learning_rate": 0.00023653961885656972,
"loss": 2.6201,
"theoretical_loss": 3.468464144154974,
"tokens_seen": 1754988544
},
{
"epoch": 0.53,
"learning_rate": 0.0002365295887662989,
"loss": 2.6476,
"theoretical_loss": 3.468453065648911,
"tokens_seen": 1755054080
},
{
"epoch": 0.53,
"learning_rate": 0.00023651955867602808,
"loss": 2.4836,
"theoretical_loss": 3.468441987672353,
"tokens_seen": 1755119616
},
{
"epoch": 0.53,
"learning_rate": 0.00023650952858575729,
"loss": 2.5616,
"theoretical_loss": 3.468430910225255,
"tokens_seen": 1755185152
},
{
"epoch": 0.53,
"learning_rate": 0.00023649949849548647,
"loss": 2.4912,
"theoretical_loss": 3.468419833307572,
"tokens_seen": 1755250688
},
{
"epoch": 0.53,
"learning_rate": 0.00023648946840521565,
"loss": 2.7481,
"theoretical_loss": 3.468408756919259,
"tokens_seen": 1755316224
},
{
"epoch": 0.53,
"learning_rate": 0.00023647943831494483,
"loss": 2.764,
"theoretical_loss": 3.468397681060271,
"tokens_seen": 1755381760
},
{
"epoch": 0.53,
"learning_rate": 0.00023646940822467404,
"loss": 2.5238,
"theoretical_loss": 3.4683866057305623,
"tokens_seen": 1755447296
},
{
"epoch": 0.53,
"learning_rate": 0.00023645937813440322,
"loss": 2.6988,
"theoretical_loss": 3.468375530930089,
"tokens_seen": 1755512832
},
{
"epoch": 0.53,
"learning_rate": 0.0002364493480441324,
"loss": 2.5927,
"theoretical_loss": 3.468364456658805,
"tokens_seen": 1755578368
},
{
"epoch": 0.53,
"learning_rate": 0.00023643931795386158,
"loss": 2.618,
"theoretical_loss": 3.4683533829166664,
"tokens_seen": 1755643904
},
{
"epoch": 0.53,
"learning_rate": 0.00023642928786359076,
"loss": 2.7264,
"theoretical_loss": 3.468342309703627,
"tokens_seen": 1755709440
},
{
"epoch": 0.53,
"learning_rate": 0.00023641925777331997,
"loss": 2.717,
"theoretical_loss": 3.4683312370196426,
"tokens_seen": 1755774976
},
{
"epoch": 0.53,
"learning_rate": 0.00023640922768304915,
"loss": 2.5463,
"theoretical_loss": 3.4683201648646675,
"tokens_seen": 1755840512
},
{
"epoch": 0.53,
"learning_rate": 0.00023639919759277833,
"loss": 2.8782,
"theoretical_loss": 3.4683090932386573,
"tokens_seen": 1755906048
},
{
"epoch": 0.53,
"learning_rate": 0.0002363891675025075,
"loss": 2.5318,
"theoretical_loss": 3.4682980221415662,
"tokens_seen": 1755971584
},
{
"epoch": 0.53,
"learning_rate": 0.00023637913741223672,
"loss": 2.6843,
"theoretical_loss": 3.46828695157335,
"tokens_seen": 1756037120
},
{
"epoch": 0.53,
"learning_rate": 0.0002363691073219659,
"loss": 2.7029,
"theoretical_loss": 3.4682758815339634,
"tokens_seen": 1756102656
},
{
"epoch": 0.53,
"learning_rate": 0.00023635907723169508,
"loss": 2.4355,
"theoretical_loss": 3.4682648120233615,
"tokens_seen": 1756168192
},
{
"epoch": 0.53,
"learning_rate": 0.00023634904714142426,
"loss": 2.7103,
"theoretical_loss": 3.4682537430414992,
"tokens_seen": 1756233728
},
{
"epoch": 0.53,
"learning_rate": 0.00023633901705115347,
"loss": 2.7413,
"theoretical_loss": 3.468242674588331,
"tokens_seen": 1756299264
},
{
"epoch": 0.53,
"objective/train/docs_used": 994709,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.55454421043396,
"objective/train/theoretical_loss": 3.468231606663813,
"objective/train/tokens_used": 1776824800,
"theoretical_loss": 3.468231606663813,
"tokens_seen": 1756364800
},
{
"epoch": 0.53,
"learning_rate": 0.00023632898696088265,
"loss": 2.3709,
"theoretical_loss": 3.468231606663813,
"tokens_seen": 1756364800
},
{
"epoch": 0.53,
"learning_rate": 0.00023631895687061184,
"loss": 2.6721,
"theoretical_loss": 3.4682205392678993,
"tokens_seen": 1756430336
},
{
"epoch": 0.53,
"learning_rate": 0.00023630892678034102,
"loss": 2.5432,
"theoretical_loss": 3.4682094724005457,
"tokens_seen": 1756495872
},
{
"epoch": 0.53,
"learning_rate": 0.0002362988966900702,
"loss": 2.5623,
"theoretical_loss": 3.4681984060617066,
"tokens_seen": 1756561408
},
{
"epoch": 0.53,
"learning_rate": 0.0002362888665997994,
"loss": 2.626,
"theoretical_loss": 3.4681873402513372,
"tokens_seen": 1756626944
},
{
"epoch": 0.53,
"learning_rate": 0.0002362788365095286,
"loss": 2.6998,
"theoretical_loss": 3.4681762749693927,
"tokens_seen": 1756692480
},
{
"epoch": 0.53,
"learning_rate": 0.00023626880641925777,
"loss": 2.4866,
"theoretical_loss": 3.4681652102158282,
"tokens_seen": 1756758016
},
{
"epoch": 0.53,
"learning_rate": 0.00023625877632898695,
"loss": 2.5762,
"theoretical_loss": 3.4681541459905985,
"tokens_seen": 1756823552
},
{
"epoch": 0.53,
"learning_rate": 0.00023624874623871616,
"loss": 2.5069,
"theoretical_loss": 3.468143082293659,
"tokens_seen": 1756889088
},
{
"epoch": 0.53,
"learning_rate": 0.00023623871614844534,
"loss": 2.7263,
"theoretical_loss": 3.4681320191249645,
"tokens_seen": 1756954624
},
{
"epoch": 0.53,
"learning_rate": 0.00023622868605817452,
"loss": 2.5606,
"theoretical_loss": 3.46812095648447,
"tokens_seen": 1757020160
},
{
"epoch": 0.53,
"learning_rate": 0.0002362186559679037,
"loss": 2.7556,
"theoretical_loss": 3.468109894372131,
"tokens_seen": 1757085696
},
{
"epoch": 0.53,
"learning_rate": 0.00023620862587763288,
"loss": 2.6596,
"theoretical_loss": 3.468098832787902,
"tokens_seen": 1757151232
},
{
"epoch": 0.53,
"learning_rate": 0.0002361985957873621,
"loss": 2.638,
"theoretical_loss": 3.4680877717317387,
"tokens_seen": 1757216768
},
{
"epoch": 0.53,
"learning_rate": 0.00023618856569709127,
"loss": 2.5802,
"theoretical_loss": 3.4680767112035955,
"tokens_seen": 1757282304
},
{
"epoch": 0.53,
"learning_rate": 0.00023617853560682045,
"loss": 2.7255,
"theoretical_loss": 3.4680656512034282,
"tokens_seen": 1757347840
},
{
"epoch": 0.53,
"learning_rate": 0.00023616850551654963,
"loss": 2.5288,
"theoretical_loss": 3.4680545917311916,
"tokens_seen": 1757413376
},
{
"epoch": 0.53,
"learning_rate": 0.00023615847542627884,
"loss": 2.6731,
"theoretical_loss": 3.468043532786841,
"tokens_seen": 1757478912
},
{
"epoch": 0.53,
"learning_rate": 0.00023614844533600802,
"loss": 2.7107,
"theoretical_loss": 3.4680324743703315,
"tokens_seen": 1757544448
},
{
"epoch": 0.53,
"learning_rate": 0.0002361384152457372,
"loss": 2.703,
"theoretical_loss": 3.4680214164816174,
"tokens_seen": 1757609984
},
{
"epoch": 0.53,
"learning_rate": 0.00023612838515546639,
"loss": 2.5013,
"theoretical_loss": 3.468010359120655,
"tokens_seen": 1757675520
},
{
"epoch": 0.53,
"learning_rate": 0.00023611835506519557,
"loss": 2.608,
"theoretical_loss": 3.467999302287399,
"tokens_seen": 1757741056
},
{
"epoch": 0.53,
"learning_rate": 0.0002361083249749248,
"loss": 2.7633,
"theoretical_loss": 3.4679882459818043,
"tokens_seen": 1757806592
},
{
"epoch": 0.53,
"learning_rate": 0.00023609829488465398,
"loss": 2.617,
"theoretical_loss": 3.4679771902038263,
"tokens_seen": 1757872128
},
{
"epoch": 0.53,
"learning_rate": 0.00023608826479438316,
"loss": 2.682,
"theoretical_loss": 3.4679661349534197,
"tokens_seen": 1757937664
},
{
"epoch": 0.53,
"objective/train/docs_used": 995465,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5516903400421143,
"objective/train/theoretical_loss": 3.4679550802305408,
"objective/train/tokens_used": 1778463200,
"theoretical_loss": 3.4679550802305408,
"tokens_seen": 1758003200
},
{
"epoch": 0.53,
"learning_rate": 0.00023607823470411234,
"loss": 2.6434,
"theoretical_loss": 3.4679550802305408,
"tokens_seen": 1758003200
},
{
"epoch": 0.53,
"learning_rate": 0.00023606820461384155,
"loss": 2.7617,
"theoretical_loss": 3.467944026035144,
"tokens_seen": 1758068736
},
{
"epoch": 0.53,
"learning_rate": 0.00023605817452357073,
"loss": 2.4707,
"theoretical_loss": 3.4679329723671835,
"tokens_seen": 1758134272
},
{
"epoch": 0.53,
"learning_rate": 0.00023604814443329992,
"loss": 2.5974,
"theoretical_loss": 3.467921919226616,
"tokens_seen": 1758199808
},
{
"epoch": 0.53,
"learning_rate": 0.0002360381143430291,
"loss": 2.6065,
"theoretical_loss": 3.4679108666133964,
"tokens_seen": 1758265344
},
{
"epoch": 0.53,
"learning_rate": 0.00023602808425275828,
"loss": 2.7244,
"theoretical_loss": 3.4678998145274793,
"tokens_seen": 1758330880
},
{
"epoch": 0.53,
"learning_rate": 0.00023601805416248749,
"loss": 2.6737,
"theoretical_loss": 3.467888762968821,
"tokens_seen": 1758396416
},
{
"epoch": 0.53,
"learning_rate": 0.00023600802407221667,
"loss": 2.5925,
"theoretical_loss": 3.467877711937375,
"tokens_seen": 1758461952
},
{
"epoch": 0.53,
"learning_rate": 0.00023599799398194585,
"loss": 2.3466,
"theoretical_loss": 3.4678666614330975,
"tokens_seen": 1758527488
},
{
"epoch": 0.53,
"learning_rate": 0.00023598796389167503,
"loss": 2.6154,
"theoretical_loss": 3.467855611455944,
"tokens_seen": 1758593024
},
{
"epoch": 0.53,
"learning_rate": 0.00023597793380140424,
"loss": 2.7322,
"theoretical_loss": 3.467844562005869,
"tokens_seen": 1758658560
},
{
"epoch": 0.53,
"learning_rate": 0.00023596790371113342,
"loss": 2.5774,
"theoretical_loss": 3.4678335130828284,
"tokens_seen": 1758724096
},
{
"epoch": 0.53,
"learning_rate": 0.0002359578736208626,
"loss": 2.5544,
"theoretical_loss": 3.4678224646867766,
"tokens_seen": 1758789632
},
{
"epoch": 0.53,
"learning_rate": 0.00023594784353059178,
"loss": 2.2364,
"theoretical_loss": 3.4678114168176695,
"tokens_seen": 1758855168
},
{
"epoch": 0.53,
"learning_rate": 0.00023593781344032096,
"loss": 2.5555,
"theoretical_loss": 3.467800369475462,
"tokens_seen": 1758920704
},
{
"epoch": 0.53,
"learning_rate": 0.00023592778335005017,
"loss": 2.5876,
"theoretical_loss": 3.46778932266011,
"tokens_seen": 1758986240
},
{
"epoch": 0.53,
"learning_rate": 0.00023591775325977935,
"loss": 2.706,
"theoretical_loss": 3.467778276371568,
"tokens_seen": 1759051776
},
{
"epoch": 0.53,
"learning_rate": 0.00023590772316950853,
"loss": 2.7726,
"theoretical_loss": 3.467767230609791,
"tokens_seen": 1759117312
},
{
"epoch": 0.53,
"learning_rate": 0.0002358976930792377,
"loss": 2.647,
"theoretical_loss": 3.4677561853747356,
"tokens_seen": 1759182848
},
{
"epoch": 0.53,
"learning_rate": 0.00023588766298896692,
"loss": 2.5943,
"theoretical_loss": 3.4677451406663558,
"tokens_seen": 1759248384
},
{
"epoch": 0.53,
"learning_rate": 0.0002358776328986961,
"loss": 2.854,
"theoretical_loss": 3.467734096484607,
"tokens_seen": 1759313920
},
{
"epoch": 0.53,
"learning_rate": 0.00023586760280842528,
"loss": 2.7359,
"theoretical_loss": 3.467723052829445,
"tokens_seen": 1759379456
},
{
"epoch": 0.53,
"learning_rate": 0.00023585757271815447,
"loss": 2.478,
"theoretical_loss": 3.4677120097008247,
"tokens_seen": 1759444992
},
{
"epoch": 0.53,
"learning_rate": 0.00023584754262788367,
"loss": 2.8556,
"theoretical_loss": 3.4677009670987013,
"tokens_seen": 1759510528
},
{
"epoch": 0.53,
"learning_rate": 0.00023583751253761285,
"loss": 2.8915,
"theoretical_loss": 3.4676899250230306,
"tokens_seen": 1759576064
},
{
"epoch": 0.53,
"objective/train/docs_used": 995841,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.080861806869507,
"objective/train/theoretical_loss": 3.4676788834737677,
"objective/train/tokens_used": 1780101600,
"theoretical_loss": 3.4676788834737677,
"tokens_seen": 1759641600
},
{
"epoch": 0.53,
"learning_rate": 0.00023582748244734204,
"loss": 2.7479,
"theoretical_loss": 3.4676788834737677,
"tokens_seen": 1759641600
},
{
"epoch": 0.53,
"learning_rate": 0.00023581745235707122,
"loss": 2.5509,
"theoretical_loss": 3.467667842450868,
"tokens_seen": 1759707136
},
{
"epoch": 0.53,
"learning_rate": 0.0002358074222668004,
"loss": 2.7939,
"theoretical_loss": 3.467656801954286,
"tokens_seen": 1759772672
},
{
"epoch": 0.53,
"learning_rate": 0.0002357973921765296,
"loss": 2.8251,
"theoretical_loss": 3.4676457619839773,
"tokens_seen": 1759838208
},
{
"epoch": 0.53,
"learning_rate": 0.0002357873620862588,
"loss": 2.7373,
"theoretical_loss": 3.4676347225398985,
"tokens_seen": 1759903744
},
{
"epoch": 0.53,
"learning_rate": 0.00023577733199598797,
"loss": 2.6142,
"theoretical_loss": 3.467623683622003,
"tokens_seen": 1759969280
},
{
"epoch": 0.53,
"learning_rate": 0.00023576730190571715,
"loss": 2.5984,
"theoretical_loss": 3.467612645230248,
"tokens_seen": 1760034816
},
{
"epoch": 0.53,
"learning_rate": 0.00023575727181544636,
"loss": 2.8911,
"theoretical_loss": 3.4676016073645872,
"tokens_seen": 1760100352
},
{
"epoch": 0.53,
"learning_rate": 0.00023574724172517554,
"loss": 2.6623,
"theoretical_loss": 3.467590570024977,
"tokens_seen": 1760165888
},
{
"epoch": 0.53,
"learning_rate": 0.00023573721163490472,
"loss": 2.5288,
"theoretical_loss": 3.4675795332113726,
"tokens_seen": 1760231424
},
{
"epoch": 0.53,
"learning_rate": 0.0002357271815446339,
"loss": 2.7517,
"theoretical_loss": 3.467568496923729,
"tokens_seen": 1760296960
},
{
"epoch": 0.53,
"learning_rate": 0.00023571715145436308,
"loss": 2.6646,
"theoretical_loss": 3.4675574611620013,
"tokens_seen": 1760362496
},
{
"epoch": 0.53,
"learning_rate": 0.0002357071213640923,
"loss": 2.6965,
"theoretical_loss": 3.4675464259261455,
"tokens_seen": 1760428032
},
{
"epoch": 0.53,
"learning_rate": 0.00023569709127382147,
"loss": 2.5994,
"theoretical_loss": 3.467535391216116,
"tokens_seen": 1760493568
},
{
"epoch": 0.53,
"learning_rate": 0.00023568706118355065,
"loss": 2.6269,
"theoretical_loss": 3.46752435703187,
"tokens_seen": 1760559104
},
{
"epoch": 0.53,
"learning_rate": 0.00023567703109327983,
"loss": 2.4866,
"theoretical_loss": 3.4675133233733613,
"tokens_seen": 1760624640
},
{
"epoch": 0.53,
"learning_rate": 0.00023566700100300904,
"loss": 2.6289,
"theoretical_loss": 3.467502290240546,
"tokens_seen": 1760690176
},
{
"epoch": 0.53,
"learning_rate": 0.00023565697091273822,
"loss": 2.7678,
"theoretical_loss": 3.4674912576333785,
"tokens_seen": 1760755712
},
{
"epoch": 0.53,
"learning_rate": 0.0002356469408224674,
"loss": 2.4477,
"theoretical_loss": 3.4674802255518156,
"tokens_seen": 1760821248
},
{
"epoch": 0.53,
"learning_rate": 0.00023563691073219659,
"loss": 2.852,
"theoretical_loss": 3.467469193995812,
"tokens_seen": 1760886784
},
{
"epoch": 0.53,
"learning_rate": 0.00023562688064192577,
"loss": 2.7246,
"theoretical_loss": 3.4674581629653227,
"tokens_seen": 1760952320
},
{
"epoch": 0.53,
"learning_rate": 0.00023561685055165497,
"loss": 2.7179,
"theoretical_loss": 3.467447132460304,
"tokens_seen": 1761017856
},
{
"epoch": 0.53,
"learning_rate": 0.00023560682046138416,
"loss": 2.7568,
"theoretical_loss": 3.467436102480711,
"tokens_seen": 1761083392
},
{
"epoch": 0.53,
"learning_rate": 0.00023559679037111334,
"loss": 2.6958,
"theoretical_loss": 3.4674250730264986,
"tokens_seen": 1761148928
},
{
"epoch": 0.53,
"learning_rate": 0.00023558676028084252,
"loss": 2.7107,
"theoretical_loss": 3.4674140440976227,
"tokens_seen": 1761214464
},
{
"epoch": 0.53,
"objective/train/docs_used": 997243,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3875646591186523,
"objective/train/theoretical_loss": 3.467403015694039,
"objective/train/tokens_used": 1781740000,
"theoretical_loss": 3.467403015694039,
"tokens_seen": 1761280000
},
{
"epoch": 0.53,
"learning_rate": 0.00023557673019057173,
"loss": 2.567,
"theoretical_loss": 3.467403015694039,
"tokens_seen": 1761280000
},
{
"epoch": 0.53,
"learning_rate": 0.0002355667001003009,
"loss": 2.5493,
"theoretical_loss": 3.467391987815702,
"tokens_seen": 1761345536
},
{
"epoch": 0.53,
"learning_rate": 0.0002355566700100301,
"loss": 2.682,
"theoretical_loss": 3.4673809604625676,
"tokens_seen": 1761411072
},
{
"epoch": 0.53,
"learning_rate": 0.00023554663991975927,
"loss": 2.4985,
"theoretical_loss": 3.4673699336345916,
"tokens_seen": 1761476608
},
{
"epoch": 0.53,
"learning_rate": 0.00023553660982948845,
"loss": 2.3499,
"theoretical_loss": 3.4673589073317297,
"tokens_seen": 1761542144
},
{
"epoch": 0.53,
"learning_rate": 0.00023552657973921766,
"loss": 2.4864,
"theoretical_loss": 3.4673478815539363,
"tokens_seen": 1761607680
},
{
"epoch": 0.53,
"learning_rate": 0.00023551654964894684,
"loss": 2.7572,
"theoretical_loss": 3.4673368563011677,
"tokens_seen": 1761673216
},
{
"epoch": 0.53,
"learning_rate": 0.00023550651955867602,
"loss": 2.5762,
"theoretical_loss": 3.467325831573379,
"tokens_seen": 1761738752
},
{
"epoch": 0.53,
"learning_rate": 0.0002354964894684052,
"loss": 2.6959,
"theoretical_loss": 3.467314807370526,
"tokens_seen": 1761804288
},
{
"epoch": 0.53,
"learning_rate": 0.0002354864593781344,
"loss": 2.8041,
"theoretical_loss": 3.4673037836925635,
"tokens_seen": 1761869824
},
{
"epoch": 0.53,
"learning_rate": 0.0002354764292878636,
"loss": 2.7522,
"theoretical_loss": 3.467292760539448,
"tokens_seen": 1761935360
},
{
"epoch": 0.53,
"learning_rate": 0.00023546639919759277,
"loss": 2.7042,
"theoretical_loss": 3.467281737911134,
"tokens_seen": 1762000896
},
{
"epoch": 0.53,
"learning_rate": 0.00023545636910732195,
"loss": 2.6585,
"theoretical_loss": 3.4672707158075777,
"tokens_seen": 1762066432
},
{
"epoch": 0.53,
"learning_rate": 0.00023544633901705113,
"loss": 2.617,
"theoretical_loss": 3.467259694228734,
"tokens_seen": 1762131968
},
{
"epoch": 0.53,
"learning_rate": 0.00023543630892678034,
"loss": 2.7959,
"theoretical_loss": 3.467248673174559,
"tokens_seen": 1762197504
},
{
"epoch": 0.53,
"learning_rate": 0.00023542627883650952,
"loss": 2.5006,
"theoretical_loss": 3.467237652645008,
"tokens_seen": 1762263040
},
{
"epoch": 0.53,
"learning_rate": 0.0002354162487462387,
"loss": 2.5302,
"theoretical_loss": 3.4672266326400365,
"tokens_seen": 1762328576
},
{
"epoch": 0.53,
"learning_rate": 0.0002354062186559679,
"loss": 2.7479,
"theoretical_loss": 3.4672156131595995,
"tokens_seen": 1762394112
},
{
"epoch": 0.53,
"learning_rate": 0.0002353961885656971,
"loss": 2.662,
"theoretical_loss": 3.4672045942036536,
"tokens_seen": 1762459648
},
{
"epoch": 0.53,
"learning_rate": 0.00023538615847542628,
"loss": 2.5934,
"theoretical_loss": 3.4671935757721535,
"tokens_seen": 1762525184
},
{
"epoch": 0.53,
"learning_rate": 0.00023537612838515546,
"loss": 2.5721,
"theoretical_loss": 3.467182557865055,
"tokens_seen": 1762590720
},
{
"epoch": 0.53,
"learning_rate": 0.00023536609829488464,
"loss": 2.5624,
"theoretical_loss": 3.4671715404823136,
"tokens_seen": 1762656256
},
{
"epoch": 0.53,
"learning_rate": 0.00023535606820461387,
"loss": 2.5581,
"theoretical_loss": 3.4671605236238854,
"tokens_seen": 1762721792
},
{
"epoch": 0.53,
"learning_rate": 0.00023534603811434305,
"loss": 2.6831,
"theoretical_loss": 3.467149507289725,
"tokens_seen": 1762787328
},
{
"epoch": 0.53,
"learning_rate": 0.00023533600802407224,
"loss": 2.6657,
"theoretical_loss": 3.4671384914797887,
"tokens_seen": 1762852864
},
{
"epoch": 0.53,
"objective/train/docs_used": 997805,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.447045087814331,
"objective/train/theoretical_loss": 3.4671274761940314,
"objective/train/tokens_used": 1783378400,
"theoretical_loss": 3.4671274761940314,
"tokens_seen": 1762918400
},
{
"epoch": 0.53,
"learning_rate": 0.00023532597793380142,
"loss": 2.1604,
"theoretical_loss": 3.4671274761940314,
"tokens_seen": 1762918400
},
{
"epoch": 0.53,
"learning_rate": 0.0002353159478435306,
"loss": 2.679,
"theoretical_loss": 3.4671164614324095,
"tokens_seen": 1762983936
},
{
"epoch": 0.53,
"learning_rate": 0.0002353059177532598,
"loss": 2.7128,
"theoretical_loss": 3.4671054471948777,
"tokens_seen": 1763049472
},
{
"epoch": 0.53,
"learning_rate": 0.000235295887662989,
"loss": 2.6833,
"theoretical_loss": 3.467094433481392,
"tokens_seen": 1763115008
},
{
"epoch": 0.53,
"learning_rate": 0.00023528585757271817,
"loss": 2.7671,
"theoretical_loss": 3.4670834202919085,
"tokens_seen": 1763180544
},
{
"epoch": 0.53,
"learning_rate": 0.00023527582748244735,
"loss": 2.5977,
"theoretical_loss": 3.4670724076263815,
"tokens_seen": 1763246080
},
{
"epoch": 0.53,
"learning_rate": 0.00023526579739217656,
"loss": 2.7937,
"theoretical_loss": 3.4670613954847678,
"tokens_seen": 1763311616
},
{
"epoch": 0.53,
"learning_rate": 0.00023525576730190574,
"loss": 2.69,
"theoretical_loss": 3.467050383867023,
"tokens_seen": 1763377152
},
{
"epoch": 0.53,
"learning_rate": 0.00023524573721163492,
"loss": 2.7522,
"theoretical_loss": 3.467039372773102,
"tokens_seen": 1763442688
},
{
"epoch": 0.53,
"learning_rate": 0.0002352357071213641,
"loss": 2.8569,
"theoretical_loss": 3.467028362202961,
"tokens_seen": 1763508224
},
{
"epoch": 0.53,
"learning_rate": 0.00023522567703109328,
"loss": 2.6669,
"theoretical_loss": 3.467017352156555,
"tokens_seen": 1763573760
},
{
"epoch": 0.53,
"learning_rate": 0.0002352156469408225,
"loss": 2.6188,
"theoretical_loss": 3.46700634263384,
"tokens_seen": 1763639296
},
{
"epoch": 0.53,
"learning_rate": 0.00023520561685055167,
"loss": 2.5904,
"theoretical_loss": 3.466995333634772,
"tokens_seen": 1763704832
},
{
"epoch": 0.53,
"learning_rate": 0.00023519558676028085,
"loss": 2.7782,
"theoretical_loss": 3.4669843251593058,
"tokens_seen": 1763770368
},
{
"epoch": 0.53,
"learning_rate": 0.00023518555667001003,
"loss": 2.6278,
"theoretical_loss": 3.4669733172073975,
"tokens_seen": 1763835904
},
{
"epoch": 0.53,
"learning_rate": 0.00023517552657973924,
"loss": 2.5131,
"theoretical_loss": 3.466962309779003,
"tokens_seen": 1763901440
},
{
"epoch": 0.53,
"learning_rate": 0.00023516549648946842,
"loss": 2.4629,
"theoretical_loss": 3.4669513028740777,
"tokens_seen": 1763966976
},
{
"epoch": 0.53,
"learning_rate": 0.0002351554663991976,
"loss": 2.3806,
"theoretical_loss": 3.466940296492577,
"tokens_seen": 1764032512
},
{
"epoch": 0.53,
"learning_rate": 0.00023514543630892679,
"loss": 2.6504,
"theoretical_loss": 3.466929290634457,
"tokens_seen": 1764098048
},
{
"epoch": 0.53,
"learning_rate": 0.00023513540621865597,
"loss": 2.6639,
"theoretical_loss": 3.466918285299673,
"tokens_seen": 1764163584
},
{
"epoch": 0.53,
"learning_rate": 0.00023512537612838517,
"loss": 2.6775,
"theoretical_loss": 3.466907280488181,
"tokens_seen": 1764229120
},
{
"epoch": 0.53,
"learning_rate": 0.00023511534603811436,
"loss": 2.7153,
"theoretical_loss": 3.4668962761999365,
"tokens_seen": 1764294656
},
{
"epoch": 0.53,
"learning_rate": 0.00023510531594784354,
"loss": 2.565,
"theoretical_loss": 3.4668852724348955,
"tokens_seen": 1764360192
},
{
"epoch": 0.53,
"learning_rate": 0.00023509528585757272,
"loss": 2.5518,
"theoretical_loss": 3.4668742691930126,
"tokens_seen": 1764425728
},
{
"epoch": 0.53,
"learning_rate": 0.00023508525576730193,
"loss": 2.5543,
"theoretical_loss": 3.4668632664742454,
"tokens_seen": 1764491264
},
{
"epoch": 0.53,
"objective/train/docs_used": 998902,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7779269218444824,
"objective/train/theoretical_loss": 3.4668522642785478,
"objective/train/tokens_used": 1785016800,
"theoretical_loss": 3.4668522642785478,
"tokens_seen": 1764556800
},
{
"epoch": 0.53,
"learning_rate": 0.0002350752256770311,
"loss": 2.7277,
"theoretical_loss": 3.4668522642785478,
"tokens_seen": 1764556800
},
{
"epoch": 0.53,
"learning_rate": 0.0002350651955867603,
"loss": 2.6759,
"theoretical_loss": 3.466841262605876,
"tokens_seen": 1764622336
},
{
"epoch": 0.53,
"learning_rate": 0.00023505516549648947,
"loss": 2.7474,
"theoretical_loss": 3.466830261456187,
"tokens_seen": 1764687872
},
{
"epoch": 0.53,
"learning_rate": 0.00023504513540621865,
"loss": 2.565,
"theoretical_loss": 3.4668192608294346,
"tokens_seen": 1764753408
},
{
"epoch": 0.53,
"learning_rate": 0.00023503510531594786,
"loss": 2.5858,
"theoretical_loss": 3.4668082607255757,
"tokens_seen": 1764818944
},
{
"epoch": 0.53,
"learning_rate": 0.00023502507522567704,
"loss": 3.006,
"theoretical_loss": 3.4667972611445657,
"tokens_seen": 1764884480
},
{
"epoch": 0.53,
"learning_rate": 0.00023501504513540622,
"loss": 2.6648,
"theoretical_loss": 3.46678626208636,
"tokens_seen": 1764950016
},
{
"epoch": 0.53,
"learning_rate": 0.0002350050150451354,
"loss": 2.396,
"theoretical_loss": 3.466775263550915,
"tokens_seen": 1765015552
},
{
"epoch": 0.53,
"learning_rate": 0.0002349949849548646,
"loss": 2.5149,
"theoretical_loss": 3.466764265538186,
"tokens_seen": 1765081088
},
{
"epoch": 0.53,
"learning_rate": 0.0002349849548645938,
"loss": 2.5735,
"theoretical_loss": 3.466753268048129,
"tokens_seen": 1765146624
},
{
"epoch": 0.53,
"learning_rate": 0.00023497492477432297,
"loss": 2.7731,
"theoretical_loss": 3.4667422710806997,
"tokens_seen": 1765212160
},
{
"epoch": 0.53,
"learning_rate": 0.00023496489468405215,
"loss": 2.6629,
"theoretical_loss": 3.466731274635854,
"tokens_seen": 1765277696
},
{
"epoch": 0.53,
"learning_rate": 0.00023495486459378134,
"loss": 2.5833,
"theoretical_loss": 3.466720278713547,
"tokens_seen": 1765343232
},
{
"epoch": 0.53,
"learning_rate": 0.00023494483450351054,
"loss": 2.6963,
"theoretical_loss": 3.466709283313735,
"tokens_seen": 1765408768
},
{
"epoch": 0.53,
"learning_rate": 0.00023493480441323972,
"loss": 2.5906,
"theoretical_loss": 3.4666982884363735,
"tokens_seen": 1765474304
},
{
"epoch": 0.54,
"learning_rate": 0.0002349247743229689,
"loss": 2.701,
"theoretical_loss": 3.466687294081419,
"tokens_seen": 1765539840
},
{
"epoch": 0.54,
"learning_rate": 0.0002349147442326981,
"loss": 2.6676,
"theoretical_loss": 3.466676300248827,
"tokens_seen": 1765605376
},
{
"epoch": 0.54,
"learning_rate": 0.0002349047141424273,
"loss": 2.6125,
"theoretical_loss": 3.4666653069385527,
"tokens_seen": 1765670912
},
{
"epoch": 0.54,
"learning_rate": 0.00023489468405215648,
"loss": 2.527,
"theoretical_loss": 3.466654314150552,
"tokens_seen": 1765736448
},
{
"epoch": 0.54,
"learning_rate": 0.00023488465396188566,
"loss": 2.6362,
"theoretical_loss": 3.4666433218847814,
"tokens_seen": 1765801984
},
{
"epoch": 0.54,
"learning_rate": 0.00023487462387161484,
"loss": 2.5893,
"theoretical_loss": 3.466632330141196,
"tokens_seen": 1765867520
},
{
"epoch": 0.54,
"learning_rate": 0.00023486459378134402,
"loss": 2.9267,
"theoretical_loss": 3.4666213389197518,
"tokens_seen": 1765933056
},
{
"epoch": 0.54,
"learning_rate": 0.00023485456369107323,
"loss": 2.5059,
"theoretical_loss": 3.4666103482204056,
"tokens_seen": 1765998592
},
{
"epoch": 0.54,
"learning_rate": 0.0002348445336008024,
"loss": 2.7107,
"theoretical_loss": 3.466599358043111,
"tokens_seen": 1766064128
},
{
"epoch": 0.54,
"learning_rate": 0.0002348345035105316,
"loss": 2.6897,
"theoretical_loss": 3.4665883683878262,
"tokens_seen": 1766129664
},
{
"epoch": 0.54,
"objective/train/docs_used": 999561,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.554349422454834,
"objective/train/theoretical_loss": 3.466577379254506,
"objective/train/tokens_used": 1786655200,
"theoretical_loss": 3.466577379254506,
"tokens_seen": 1766195200
},
{
"epoch": 0.54,
"learning_rate": 0.00023482447342026077,
"loss": 2.6444,
"theoretical_loss": 3.466577379254506,
"tokens_seen": 1766195200
},
{
"epoch": 0.54,
"learning_rate": 0.00023481444332998998,
"loss": 2.7232,
"theoretical_loss": 3.4665663906431057,
"tokens_seen": 1766260736
},
{
"epoch": 0.54,
"learning_rate": 0.00023480441323971916,
"loss": 2.5647,
"theoretical_loss": 3.466555402553582,
"tokens_seen": 1766326272
},
{
"epoch": 0.54,
"learning_rate": 0.00023479438314944834,
"loss": 2.6412,
"theoretical_loss": 3.466544414985891,
"tokens_seen": 1766391808
},
{
"epoch": 0.54,
"learning_rate": 0.00023478435305917752,
"loss": 2.6796,
"theoretical_loss": 3.4665334279399875,
"tokens_seen": 1766457344
},
{
"epoch": 0.54,
"learning_rate": 0.00023477432296890673,
"loss": 2.5561,
"theoretical_loss": 3.4665224414158278,
"tokens_seen": 1766522880
},
{
"epoch": 0.54,
"learning_rate": 0.0002347642928786359,
"loss": 2.5876,
"theoretical_loss": 3.466511455413368,
"tokens_seen": 1766588416
},
{
"epoch": 0.54,
"learning_rate": 0.0002347542627883651,
"loss": 2.6932,
"theoretical_loss": 3.466500469932564,
"tokens_seen": 1766653952
},
{
"epoch": 0.54,
"learning_rate": 0.00023474423269809427,
"loss": 2.5416,
"theoretical_loss": 3.466489484973371,
"tokens_seen": 1766719488
},
{
"epoch": 0.54,
"learning_rate": 0.00023473420260782346,
"loss": 2.8007,
"theoretical_loss": 3.4664785005357457,
"tokens_seen": 1766785024
},
{
"epoch": 0.54,
"learning_rate": 0.00023472417251755266,
"loss": 3.0123,
"theoretical_loss": 3.4664675166196437,
"tokens_seen": 1766850560
},
{
"epoch": 0.54,
"learning_rate": 0.00023471414242728184,
"loss": 2.6382,
"theoretical_loss": 3.4664565332250215,
"tokens_seen": 1766916096
},
{
"epoch": 0.54,
"learning_rate": 0.00023470411233701103,
"loss": 2.5358,
"theoretical_loss": 3.4664455503518337,
"tokens_seen": 1766981632
},
{
"epoch": 0.54,
"learning_rate": 0.0002346940822467402,
"loss": 2.841,
"theoretical_loss": 3.466434568000037,
"tokens_seen": 1767047168
},
{
"epoch": 0.54,
"learning_rate": 0.00023468405215646942,
"loss": 2.6386,
"theoretical_loss": 3.4664235861695873,
"tokens_seen": 1767112704
},
{
"epoch": 0.54,
"learning_rate": 0.0002346740220661986,
"loss": 2.6807,
"theoretical_loss": 3.46641260486044,
"tokens_seen": 1767178240
},
{
"epoch": 0.54,
"learning_rate": 0.00023466399197592778,
"loss": 2.6743,
"theoretical_loss": 3.466401624072552,
"tokens_seen": 1767243776
},
{
"epoch": 0.54,
"learning_rate": 0.00023465396188565696,
"loss": 2.597,
"theoretical_loss": 3.4663906438058785,
"tokens_seen": 1767309312
},
{
"epoch": 0.54,
"learning_rate": 0.00023464393179538614,
"loss": 2.7118,
"theoretical_loss": 3.4663796640603755,
"tokens_seen": 1767374848
},
{
"epoch": 0.54,
"learning_rate": 0.00023463390170511535,
"loss": 2.5892,
"theoretical_loss": 3.466368684835999,
"tokens_seen": 1767440384
},
{
"epoch": 0.54,
"learning_rate": 0.00023462387161484453,
"loss": 2.6393,
"theoretical_loss": 3.466357706132705,
"tokens_seen": 1767505920
},
{
"epoch": 0.54,
"learning_rate": 0.0002346138415245737,
"loss": 2.5748,
"theoretical_loss": 3.46634672795045,
"tokens_seen": 1767571456
},
{
"epoch": 0.54,
"learning_rate": 0.00023460381143430292,
"loss": 2.7307,
"theoretical_loss": 3.466335750289189,
"tokens_seen": 1767636992
},
{
"epoch": 0.54,
"learning_rate": 0.00023459378134403213,
"loss": 2.5416,
"theoretical_loss": 3.466324773148878,
"tokens_seen": 1767702528
},
{
"epoch": 0.54,
"learning_rate": 0.0002345837512537613,
"loss": 2.6354,
"theoretical_loss": 3.466313796529474,
"tokens_seen": 1767768064
},
{
"epoch": 0.54,
"objective/train/docs_used": 1000589,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3767361640930176,
"objective/train/theoretical_loss": 3.466302820430932,
"objective/train/tokens_used": 1788293600,
"theoretical_loss": 3.466302820430932,
"tokens_seen": 1767833600
},
{
"epoch": 0.54,
"learning_rate": 0.0002345737211634905,
"loss": 2.6444,
"theoretical_loss": 3.466302820430932,
"tokens_seen": 1767833600
},
{
"epoch": 0.54,
"learning_rate": 0.00023456369107321967,
"loss": 2.5608,
"theoretical_loss": 3.466291844853208,
"tokens_seen": 1767899136
},
{
"epoch": 0.54,
"learning_rate": 0.00023455366098294885,
"loss": 2.5672,
"theoretical_loss": 3.4662808697962584,
"tokens_seen": 1767964672
},
{
"epoch": 0.54,
"learning_rate": 0.00023454363089267806,
"loss": 2.6785,
"theoretical_loss": 3.466269895260039,
"tokens_seen": 1768030208
},
{
"epoch": 0.54,
"learning_rate": 0.00023453360080240724,
"loss": 2.6573,
"theoretical_loss": 3.466258921244506,
"tokens_seen": 1768095744
},
{
"epoch": 0.54,
"learning_rate": 0.00023452357071213642,
"loss": 2.8148,
"theoretical_loss": 3.466247947749615,
"tokens_seen": 1768161280
},
{
"epoch": 0.54,
"learning_rate": 0.0002345135406218656,
"loss": 2.6635,
"theoretical_loss": 3.466236974775322,
"tokens_seen": 1768226816
},
{
"epoch": 0.54,
"learning_rate": 0.0002345035105315948,
"loss": 2.5661,
"theoretical_loss": 3.4662260023215836,
"tokens_seen": 1768292352
},
{
"epoch": 0.54,
"learning_rate": 0.000234493480441324,
"loss": 2.6739,
"theoretical_loss": 3.466215030388356,
"tokens_seen": 1768357888
},
{
"epoch": 0.54,
"learning_rate": 0.00023448345035105317,
"loss": 2.5984,
"theoretical_loss": 3.4662040589755936,
"tokens_seen": 1768423424
},
{
"epoch": 0.54,
"learning_rate": 0.00023447342026078235,
"loss": 2.7192,
"theoretical_loss": 3.466193088083254,
"tokens_seen": 1768488960
},
{
"epoch": 0.54,
"learning_rate": 0.00023446339017051154,
"loss": 2.5247,
"theoretical_loss": 3.466182117711293,
"tokens_seen": 1768554496
},
{
"epoch": 0.54,
"learning_rate": 0.00023445336008024074,
"loss": 2.6327,
"theoretical_loss": 3.466171147859666,
"tokens_seen": 1768620032
},
{
"epoch": 0.54,
"learning_rate": 0.00023444332998996992,
"loss": 2.7174,
"theoretical_loss": 3.46616017852833,
"tokens_seen": 1768685568
},
{
"epoch": 0.54,
"learning_rate": 0.0002344332998996991,
"loss": 2.4457,
"theoretical_loss": 3.4661492097172397,
"tokens_seen": 1768751104
},
{
"epoch": 0.54,
"learning_rate": 0.0002344232698094283,
"loss": 2.7142,
"theoretical_loss": 3.466138241426352,
"tokens_seen": 1768816640
},
{
"epoch": 0.54,
"learning_rate": 0.0002344132397191575,
"loss": 2.5869,
"theoretical_loss": 3.466127273655623,
"tokens_seen": 1768882176
},
{
"epoch": 0.54,
"learning_rate": 0.00023440320962888668,
"loss": 2.7404,
"theoretical_loss": 3.466116306405009,
"tokens_seen": 1768947712
},
{
"epoch": 0.54,
"learning_rate": 0.00023439317953861586,
"loss": 2.7957,
"theoretical_loss": 3.466105339674465,
"tokens_seen": 1769013248
},
{
"epoch": 0.54,
"learning_rate": 0.00023438314944834504,
"loss": 2.7113,
"theoretical_loss": 3.466094373463948,
"tokens_seen": 1769078784
},
{
"epoch": 0.54,
"learning_rate": 0.00023437311935807422,
"loss": 2.7382,
"theoretical_loss": 3.4660834077734144,
"tokens_seen": 1769144320
},
{
"epoch": 0.54,
"learning_rate": 0.00023436308926780343,
"loss": 2.5814,
"theoretical_loss": 3.466072442602819,
"tokens_seen": 1769209856
},
{
"epoch": 0.54,
"learning_rate": 0.0002343530591775326,
"loss": 2.7234,
"theoretical_loss": 3.4660614779521186,
"tokens_seen": 1769275392
},
{
"epoch": 0.54,
"learning_rate": 0.0002343430290872618,
"loss": 2.5498,
"theoretical_loss": 3.4660505138212696,
"tokens_seen": 1769340928
},
{
"epoch": 0.54,
"learning_rate": 0.00023433299899699097,
"loss": 2.8321,
"theoretical_loss": 3.466039550210228,
"tokens_seen": 1769406464
},
{
"epoch": 0.54,
"objective/train/docs_used": 1001288,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.400597333908081,
"objective/train/theoretical_loss": 3.4660285871189496,
"objective/train/tokens_used": 1789932000,
"theoretical_loss": 3.4660285871189496,
"tokens_seen": 1769472000
},
{
"epoch": 0.54,
"learning_rate": 0.00023432296890672018,
"loss": 2.3124,
"theoretical_loss": 3.4660285871189496,
"tokens_seen": 1769472000
},
{
"epoch": 0.54,
"learning_rate": 0.00023431293881644936,
"loss": 2.6551,
"theoretical_loss": 3.4660176245473906,
"tokens_seen": 1769537536
},
{
"epoch": 0.54,
"learning_rate": 0.00023430290872617854,
"loss": 2.5546,
"theoretical_loss": 3.4660066624955066,
"tokens_seen": 1769603072
},
{
"epoch": 0.54,
"learning_rate": 0.00023429287863590772,
"loss": 2.6494,
"theoretical_loss": 3.465995700963255,
"tokens_seen": 1769668608
},
{
"epoch": 0.54,
"learning_rate": 0.0002342828485456369,
"loss": 2.7659,
"theoretical_loss": 3.4659847399505903,
"tokens_seen": 1769734144
},
{
"epoch": 0.54,
"learning_rate": 0.0002342728184553661,
"loss": 2.6359,
"theoretical_loss": 3.46597377945747,
"tokens_seen": 1769799680
},
{
"epoch": 0.54,
"learning_rate": 0.0002342627883650953,
"loss": 2.5614,
"theoretical_loss": 3.46596281948385,
"tokens_seen": 1769865216
},
{
"epoch": 0.54,
"learning_rate": 0.00023425275827482447,
"loss": 2.4197,
"theoretical_loss": 3.465951860029686,
"tokens_seen": 1769930752
},
{
"epoch": 0.54,
"learning_rate": 0.00023424272818455366,
"loss": 2.5812,
"theoretical_loss": 3.4659409010949345,
"tokens_seen": 1769996288
},
{
"epoch": 0.54,
"learning_rate": 0.00023423269809428286,
"loss": 2.5446,
"theoretical_loss": 3.4659299426795513,
"tokens_seen": 1770061824
},
{
"epoch": 0.54,
"learning_rate": 0.00023422266800401204,
"loss": 2.4955,
"theoretical_loss": 3.4659189847834924,
"tokens_seen": 1770127360
},
{
"epoch": 0.54,
"learning_rate": 0.00023421263791374123,
"loss": 2.5399,
"theoretical_loss": 3.465908027406715,
"tokens_seen": 1770192896
},
{
"epoch": 0.54,
"learning_rate": 0.0002342026078234704,
"loss": 2.7821,
"theoretical_loss": 3.465897070549174,
"tokens_seen": 1770258432
},
{
"epoch": 0.54,
"learning_rate": 0.00023419257773319962,
"loss": 2.6134,
"theoretical_loss": 3.4658861142108264,
"tokens_seen": 1770323968
},
{
"epoch": 0.54,
"learning_rate": 0.0002341825476429288,
"loss": 2.7715,
"theoretical_loss": 3.4658751583916283,
"tokens_seen": 1770389504
},
{
"epoch": 0.54,
"learning_rate": 0.00023417251755265798,
"loss": 2.6354,
"theoretical_loss": 3.4658642030915354,
"tokens_seen": 1770455040
},
{
"epoch": 0.54,
"learning_rate": 0.00023416248746238716,
"loss": 2.4875,
"theoretical_loss": 3.4658532483105042,
"tokens_seen": 1770520576
},
{
"epoch": 0.54,
"learning_rate": 0.00023415245737211634,
"loss": 2.8572,
"theoretical_loss": 3.4658422940484908,
"tokens_seen": 1770586112
},
{
"epoch": 0.54,
"learning_rate": 0.00023414242728184555,
"loss": 2.6807,
"theoretical_loss": 3.465831340305452,
"tokens_seen": 1770651648
},
{
"epoch": 0.54,
"learning_rate": 0.00023413239719157473,
"loss": 2.6683,
"theoretical_loss": 3.4658203870813433,
"tokens_seen": 1770717184
},
{
"epoch": 0.54,
"learning_rate": 0.0002341223671013039,
"loss": 2.8582,
"theoretical_loss": 3.465809434376121,
"tokens_seen": 1770782720
},
{
"epoch": 0.54,
"learning_rate": 0.0002341123370110331,
"loss": 2.488,
"theoretical_loss": 3.465798482189742,
"tokens_seen": 1770848256
},
{
"epoch": 0.54,
"learning_rate": 0.0002341023069207623,
"loss": 2.5692,
"theoretical_loss": 3.465787530522161,
"tokens_seen": 1770913792
},
{
"epoch": 0.54,
"learning_rate": 0.00023409227683049148,
"loss": 2.5875,
"theoretical_loss": 3.4657765793733355,
"tokens_seen": 1770979328
},
{
"epoch": 0.54,
"learning_rate": 0.00023408224674022066,
"loss": 2.5156,
"theoretical_loss": 3.4657656287432212,
"tokens_seen": 1771044864
},
{
"epoch": 0.54,
"objective/train/docs_used": 1002447,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.771400213241577,
"objective/train/theoretical_loss": 3.465754678631775,
"objective/train/tokens_used": 1791570400,
"theoretical_loss": 3.465754678631775,
"tokens_seen": 1771110400
},
{
"epoch": 0.54,
"learning_rate": 0.00023407221664994984,
"loss": 2.6241,
"theoretical_loss": 3.465754678631775,
"tokens_seen": 1771110400
},
{
"epoch": 0.54,
"learning_rate": 0.00023406218655967902,
"loss": 2.5033,
"theoretical_loss": 3.4657437290389526,
"tokens_seen": 1771175936
},
{
"epoch": 0.54,
"learning_rate": 0.00023405215646940823,
"loss": 2.5985,
"theoretical_loss": 3.4657327799647106,
"tokens_seen": 1771241472
},
{
"epoch": 0.54,
"learning_rate": 0.0002340421263791374,
"loss": 2.4129,
"theoretical_loss": 3.4657218314090046,
"tokens_seen": 1771307008
},
{
"epoch": 0.54,
"learning_rate": 0.0002340320962888666,
"loss": 2.6515,
"theoretical_loss": 3.4657108833717913,
"tokens_seen": 1771372544
},
{
"epoch": 0.54,
"learning_rate": 0.00023402206619859578,
"loss": 2.7129,
"theoretical_loss": 3.465699935853027,
"tokens_seen": 1771438080
},
{
"epoch": 0.54,
"learning_rate": 0.00023401203610832498,
"loss": 2.6426,
"theoretical_loss": 3.465688988852668,
"tokens_seen": 1771503616
},
{
"epoch": 0.54,
"learning_rate": 0.00023400200601805416,
"loss": 2.6671,
"theoretical_loss": 3.46567804237067,
"tokens_seen": 1771569152
},
{
"epoch": 0.54,
"learning_rate": 0.00023399197592778335,
"loss": 2.5938,
"theoretical_loss": 3.4656670964069907,
"tokens_seen": 1771634688
},
{
"epoch": 0.54,
"learning_rate": 0.00023398194583751253,
"loss": 2.5793,
"theoretical_loss": 3.4656561509615846,
"tokens_seen": 1771700224
},
{
"epoch": 0.54,
"learning_rate": 0.0002339719157472417,
"loss": 2.5024,
"theoretical_loss": 3.4656452060344085,
"tokens_seen": 1771765760
},
{
"epoch": 0.54,
"learning_rate": 0.00023396188565697092,
"loss": 2.9541,
"theoretical_loss": 3.4656342616254197,
"tokens_seen": 1771831296
},
{
"epoch": 0.54,
"learning_rate": 0.0002339518555667001,
"loss": 2.6391,
"theoretical_loss": 3.4656233177345737,
"tokens_seen": 1771896832
},
{
"epoch": 0.54,
"learning_rate": 0.00023394182547642928,
"loss": 2.5872,
"theoretical_loss": 3.465612374361827,
"tokens_seen": 1771962368
},
{
"epoch": 0.54,
"learning_rate": 0.00023393179538615846,
"loss": 2.7571,
"theoretical_loss": 3.4656014315071357,
"tokens_seen": 1772027904
},
{
"epoch": 0.54,
"learning_rate": 0.00023392176529588767,
"loss": 2.662,
"theoretical_loss": 3.4655904891704563,
"tokens_seen": 1772093440
},
{
"epoch": 0.54,
"learning_rate": 0.00023391173520561685,
"loss": 2.7795,
"theoretical_loss": 3.4655795473517452,
"tokens_seen": 1772158976
},
{
"epoch": 0.54,
"learning_rate": 0.00023390170511534603,
"loss": 2.8549,
"theoretical_loss": 3.465568606050958,
"tokens_seen": 1772224512
},
{
"epoch": 0.54,
"learning_rate": 0.0002338916750250752,
"loss": 2.5986,
"theoretical_loss": 3.465557665268052,
"tokens_seen": 1772290048
},
{
"epoch": 0.54,
"learning_rate": 0.0002338816449348044,
"loss": 2.6809,
"theoretical_loss": 3.4655467250029837,
"tokens_seen": 1772355584
},
{
"epoch": 0.54,
"learning_rate": 0.0002338716148445336,
"loss": 2.8394,
"theoretical_loss": 3.4655357852557085,
"tokens_seen": 1772421120
},
{
"epoch": 0.54,
"learning_rate": 0.00023386158475426278,
"loss": 2.7635,
"theoretical_loss": 3.4655248460261827,
"tokens_seen": 1772486656
},
{
"epoch": 0.54,
"learning_rate": 0.000233851554663992,
"loss": 2.6906,
"theoretical_loss": 3.4655139073143637,
"tokens_seen": 1772552192
},
{
"epoch": 0.54,
"learning_rate": 0.00023384152457372117,
"loss": 2.593,
"theoretical_loss": 3.465502969120207,
"tokens_seen": 1772617728
},
{
"epoch": 0.54,
"learning_rate": 0.00023383149448345038,
"loss": 2.7418,
"theoretical_loss": 3.4654920314436697,
"tokens_seen": 1772683264
},
{
"epoch": 0.54,
"objective/train/docs_used": 1002969,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6199700832366943,
"objective/train/theoretical_loss": 3.465481094284707,
"objective/train/tokens_used": 1793208800,
"theoretical_loss": 3.465481094284707,
"tokens_seen": 1772748800
},
{
"epoch": 0.54,
"learning_rate": 0.00023382146439317956,
"loss": 2.6695,
"theoretical_loss": 3.465481094284707,
"tokens_seen": 1772748800
},
{
"epoch": 0.54,
"learning_rate": 0.00023381143430290874,
"loss": 2.5106,
"theoretical_loss": 3.465470157643277,
"tokens_seen": 1772814336
},
{
"epoch": 0.54,
"learning_rate": 0.00023380140421263792,
"loss": 2.6241,
"theoretical_loss": 3.4654592215193345,
"tokens_seen": 1772879872
},
{
"epoch": 0.54,
"learning_rate": 0.0002337913741223671,
"loss": 2.6699,
"theoretical_loss": 3.465448285912836,
"tokens_seen": 1772945408
},
{
"epoch": 0.54,
"learning_rate": 0.0002337813440320963,
"loss": 2.6816,
"theoretical_loss": 3.465437350823739,
"tokens_seen": 1773010944
},
{
"epoch": 0.54,
"learning_rate": 0.0002337713139418255,
"loss": 2.6562,
"theoretical_loss": 3.465426416251999,
"tokens_seen": 1773076480
},
{
"epoch": 0.54,
"learning_rate": 0.00023376128385155467,
"loss": 2.7867,
"theoretical_loss": 3.4654154821975727,
"tokens_seen": 1773142016
},
{
"epoch": 0.54,
"learning_rate": 0.00023375125376128386,
"loss": 2.5537,
"theoretical_loss": 3.465404548660416,
"tokens_seen": 1773207552
},
{
"epoch": 0.54,
"learning_rate": 0.00023374122367101306,
"loss": 2.6113,
"theoretical_loss": 3.4653936156404868,
"tokens_seen": 1773273088
},
{
"epoch": 0.54,
"learning_rate": 0.00023373119358074224,
"loss": 2.557,
"theoretical_loss": 3.4653826831377397,
"tokens_seen": 1773338624
},
{
"epoch": 0.54,
"learning_rate": 0.00023372116349047143,
"loss": 2.8998,
"theoretical_loss": 3.465371751152132,
"tokens_seen": 1773404160
},
{
"epoch": 0.54,
"learning_rate": 0.0002337111334002006,
"loss": 2.7604,
"theoretical_loss": 3.46536081968362,
"tokens_seen": 1773469696
},
{
"epoch": 0.54,
"learning_rate": 0.00023370110330992982,
"loss": 2.5808,
"theoretical_loss": 3.4653498887321605,
"tokens_seen": 1773535232
},
{
"epoch": 0.54,
"learning_rate": 0.000233691073219659,
"loss": 2.6718,
"theoretical_loss": 3.465338958297709,
"tokens_seen": 1773600768
},
{
"epoch": 0.54,
"learning_rate": 0.00023368104312938818,
"loss": 2.7933,
"theoretical_loss": 3.465328028380223,
"tokens_seen": 1773666304
},
{
"epoch": 0.54,
"learning_rate": 0.00023367101303911736,
"loss": 2.5657,
"theoretical_loss": 3.465317098979659,
"tokens_seen": 1773731840
},
{
"epoch": 0.54,
"learning_rate": 0.00023366098294884654,
"loss": 2.4841,
"theoretical_loss": 3.465306170095972,
"tokens_seen": 1773797376
},
{
"epoch": 0.54,
"learning_rate": 0.00023365095285857575,
"loss": 2.7055,
"theoretical_loss": 3.46529524172912,
"tokens_seen": 1773862912
},
{
"epoch": 0.54,
"learning_rate": 0.00023364092276830493,
"loss": 2.6717,
"theoretical_loss": 3.4652843138790583,
"tokens_seen": 1773928448
},
{
"epoch": 0.54,
"learning_rate": 0.0002336308926780341,
"loss": 2.5529,
"theoretical_loss": 3.4652733865457446,
"tokens_seen": 1773993984
},
{
"epoch": 0.54,
"learning_rate": 0.0002336208625877633,
"loss": 2.3943,
"theoretical_loss": 3.465262459729135,
"tokens_seen": 1774059520
},
{
"epoch": 0.54,
"learning_rate": 0.0002336108324974925,
"loss": 2.6865,
"theoretical_loss": 3.4652515334291847,
"tokens_seen": 1774125056
},
{
"epoch": 0.54,
"learning_rate": 0.00023360080240722168,
"loss": 2.8102,
"theoretical_loss": 3.4652406076458515,
"tokens_seen": 1774190592
},
{
"epoch": 0.54,
"learning_rate": 0.00023359077231695086,
"loss": 2.7445,
"theoretical_loss": 3.465229682379092,
"tokens_seen": 1774256128
},
{
"epoch": 0.54,
"learning_rate": 0.00023358074222668004,
"loss": 2.6502,
"theoretical_loss": 3.465218757628862,
"tokens_seen": 1774321664
},
{
"epoch": 0.54,
"objective/train/docs_used": 1004180,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2397451400756836,
"objective/train/theoretical_loss": 3.4652078333951186,
"objective/train/tokens_used": 1794847200,
"theoretical_loss": 3.4652078333951186,
"tokens_seen": 1774387200
},
{
"epoch": 0.54,
"learning_rate": 0.00023357071213640922,
"loss": 2.642,
"theoretical_loss": 3.4652078333951186,
"tokens_seen": 1774387200
},
{
"epoch": 0.54,
"learning_rate": 0.00023356068204613843,
"loss": 2.9635,
"theoretical_loss": 3.4651969096778177,
"tokens_seen": 1774452736
},
{
"epoch": 0.54,
"learning_rate": 0.0002335506519558676,
"loss": 2.69,
"theoretical_loss": 3.465185986476916,
"tokens_seen": 1774518272
},
{
"epoch": 0.54,
"learning_rate": 0.0002335406218655968,
"loss": 2.5567,
"theoretical_loss": 3.465175063792371,
"tokens_seen": 1774583808
},
{
"epoch": 0.54,
"learning_rate": 0.00023353059177532598,
"loss": 2.6275,
"theoretical_loss": 3.465164141624137,
"tokens_seen": 1774649344
},
{
"epoch": 0.54,
"learning_rate": 0.00023352056168505518,
"loss": 2.7185,
"theoretical_loss": 3.465153219972173,
"tokens_seen": 1774714880
},
{
"epoch": 0.54,
"learning_rate": 0.00023351053159478437,
"loss": 2.4568,
"theoretical_loss": 3.4651422988364335,
"tokens_seen": 1774780416
},
{
"epoch": 0.54,
"learning_rate": 0.00023350050150451355,
"loss": 2.7453,
"theoretical_loss": 3.4651313782168764,
"tokens_seen": 1774845952
},
{
"epoch": 0.54,
"learning_rate": 0.00023349047141424273,
"loss": 2.7383,
"theoretical_loss": 3.465120458113458,
"tokens_seen": 1774911488
},
{
"epoch": 0.54,
"learning_rate": 0.0002334804413239719,
"loss": 2.907,
"theoretical_loss": 3.4651095385261343,
"tokens_seen": 1774977024
},
{
"epoch": 0.54,
"learning_rate": 0.00023347041123370112,
"loss": 2.8174,
"theoretical_loss": 3.4650986194548623,
"tokens_seen": 1775042560
},
{
"epoch": 0.54,
"learning_rate": 0.0002334603811434303,
"loss": 2.573,
"theoretical_loss": 3.4650877008995984,
"tokens_seen": 1775108096
},
{
"epoch": 0.54,
"learning_rate": 0.00023345035105315948,
"loss": 2.706,
"theoretical_loss": 3.4650767828602995,
"tokens_seen": 1775173632
},
{
"epoch": 0.54,
"learning_rate": 0.00023344032096288866,
"loss": 2.8394,
"theoretical_loss": 3.4650658653369213,
"tokens_seen": 1775239168
},
{
"epoch": 0.54,
"learning_rate": 0.00023343029087261787,
"loss": 2.5896,
"theoretical_loss": 3.4650549483294215,
"tokens_seen": 1775304704
},
{
"epoch": 0.54,
"learning_rate": 0.00023342026078234705,
"loss": 2.7414,
"theoretical_loss": 3.4650440318377562,
"tokens_seen": 1775370240
},
{
"epoch": 0.54,
"learning_rate": 0.00023341023069207623,
"loss": 2.7347,
"theoretical_loss": 3.465033115861882,
"tokens_seen": 1775435776
},
{
"epoch": 0.54,
"learning_rate": 0.0002334002006018054,
"loss": 2.5633,
"theoretical_loss": 3.465022200401755,
"tokens_seen": 1775501312
},
{
"epoch": 0.54,
"learning_rate": 0.0002333901705115346,
"loss": 2.8618,
"theoretical_loss": 3.465011285457332,
"tokens_seen": 1775566848
},
{
"epoch": 0.54,
"learning_rate": 0.0002333801404212638,
"loss": 2.5858,
"theoretical_loss": 3.4650003710285704,
"tokens_seen": 1775632384
},
{
"epoch": 0.54,
"learning_rate": 0.00023337011033099298,
"loss": 2.5069,
"theoretical_loss": 3.4649894571154265,
"tokens_seen": 1775697920
},
{
"epoch": 0.54,
"learning_rate": 0.00023336008024072216,
"loss": 2.7162,
"theoretical_loss": 3.464978543717856,
"tokens_seen": 1775763456
},
{
"epoch": 0.54,
"learning_rate": 0.00023335005015045134,
"loss": 2.7284,
"theoretical_loss": 3.464967630835816,
"tokens_seen": 1775828992
},
{
"epoch": 0.54,
"learning_rate": 0.00023334002006018055,
"loss": 2.8193,
"theoretical_loss": 3.464956718469264,
"tokens_seen": 1775894528
},
{
"epoch": 0.54,
"learning_rate": 0.00023332998996990973,
"loss": 2.6603,
"theoretical_loss": 3.4649458066181555,
"tokens_seen": 1775960064
},
{
"epoch": 0.54,
"objective/train/docs_used": 1004852,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7583279609680176,
"objective/train/theoretical_loss": 3.4649348952824477,
"objective/train/tokens_used": 1796485600,
"theoretical_loss": 3.4649348952824477,
"tokens_seen": 1776025600
},
{
"epoch": 0.54,
"learning_rate": 0.00023331995987963891,
"loss": 2.6195,
"theoretical_loss": 3.4649348952824477,
"tokens_seen": 1776025600
},
{
"epoch": 0.54,
"learning_rate": 0.0002333099297893681,
"loss": 2.6191,
"theoretical_loss": 3.4649239844620974,
"tokens_seen": 1776091136
},
{
"epoch": 0.54,
"learning_rate": 0.00023329989969909728,
"loss": 2.6325,
"theoretical_loss": 3.46491307415706,
"tokens_seen": 1776156672
},
{
"epoch": 0.54,
"learning_rate": 0.00023328986960882649,
"loss": 2.9249,
"theoretical_loss": 3.464902164367294,
"tokens_seen": 1776222208
},
{
"epoch": 0.54,
"learning_rate": 0.00023327983951855567,
"loss": 2.6591,
"theoretical_loss": 3.4648912550927546,
"tokens_seen": 1776287744
},
{
"epoch": 0.54,
"learning_rate": 0.00023326980942828485,
"loss": 2.8109,
"theoretical_loss": 3.4648803463333993,
"tokens_seen": 1776353280
},
{
"epoch": 0.54,
"learning_rate": 0.00023325977933801403,
"loss": 2.7493,
"theoretical_loss": 3.4648694380891847,
"tokens_seen": 1776418816
},
{
"epoch": 0.54,
"learning_rate": 0.00023324974924774324,
"loss": 2.7045,
"theoretical_loss": 3.4648585303600665,
"tokens_seen": 1776484352
},
{
"epoch": 0.54,
"learning_rate": 0.00023323971915747242,
"loss": 2.4768,
"theoretical_loss": 3.464847623146003,
"tokens_seen": 1776549888
},
{
"epoch": 0.54,
"learning_rate": 0.0002332296890672016,
"loss": 2.6605,
"theoretical_loss": 3.464836716446949,
"tokens_seen": 1776615424
},
{
"epoch": 0.54,
"learning_rate": 0.00023321965897693078,
"loss": 2.7575,
"theoretical_loss": 3.464825810262863,
"tokens_seen": 1776680960
},
{
"epoch": 0.54,
"learning_rate": 0.00023320962888665996,
"loss": 2.5837,
"theoretical_loss": 3.4648149045937005,
"tokens_seen": 1776746496
},
{
"epoch": 0.54,
"learning_rate": 0.00023319959879638917,
"loss": 2.5962,
"theoretical_loss": 3.464803999439418,
"tokens_seen": 1776812032
},
{
"epoch": 0.54,
"learning_rate": 0.00023318956870611835,
"loss": 2.6525,
"theoretical_loss": 3.4647930947999743,
"tokens_seen": 1776877568
},
{
"epoch": 0.54,
"learning_rate": 0.00023317953861584753,
"loss": 2.4877,
"theoretical_loss": 3.464782190675323,
"tokens_seen": 1776943104
},
{
"epoch": 0.54,
"learning_rate": 0.0002331695085255767,
"loss": 2.7223,
"theoretical_loss": 3.4647712870654237,
"tokens_seen": 1777008640
},
{
"epoch": 0.54,
"learning_rate": 0.00023315947843530592,
"loss": 2.8539,
"theoretical_loss": 3.464760383970231,
"tokens_seen": 1777074176
},
{
"epoch": 0.54,
"learning_rate": 0.0002331494483450351,
"loss": 2.7792,
"theoretical_loss": 3.4647494813897026,
"tokens_seen": 1777139712
},
{
"epoch": 0.54,
"learning_rate": 0.00023313941825476428,
"loss": 2.7009,
"theoretical_loss": 3.464738579323795,
"tokens_seen": 1777205248
},
{
"epoch": 0.54,
"learning_rate": 0.00023312938816449346,
"loss": 2.6165,
"theoretical_loss": 3.4647276777724647,
"tokens_seen": 1777270784
},
{
"epoch": 0.54,
"learning_rate": 0.00023311935807422267,
"loss": 2.5678,
"theoretical_loss": 3.4647167767356692,
"tokens_seen": 1777336320
},
{
"epoch": 0.54,
"learning_rate": 0.00023310932798395185,
"loss": 2.8019,
"theoretical_loss": 3.4647058762133645,
"tokens_seen": 1777401856
},
{
"epoch": 0.54,
"learning_rate": 0.00023309929789368106,
"loss": 2.5501,
"theoretical_loss": 3.4646949762055077,
"tokens_seen": 1777467392
},
{
"epoch": 0.54,
"learning_rate": 0.00023308926780341024,
"loss": 2.7191,
"theoretical_loss": 3.4646840767120555,
"tokens_seen": 1777532928
},
{
"epoch": 0.54,
"learning_rate": 0.00023307923771313942,
"loss": 2.7653,
"theoretical_loss": 3.4646731777329647,
"tokens_seen": 1777598464
},
{
"epoch": 0.54,
"objective/train/docs_used": 1006253,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.710048198699951,
"objective/train/theoretical_loss": 3.4646622792681923,
"objective/train/tokens_used": 1798124000,
"theoretical_loss": 3.4646622792681923,
"tokens_seen": 1777664000
},
{
"epoch": 0.54,
"learning_rate": 0.00023306920762286863,
"loss": 2.6802,
"theoretical_loss": 3.4646622792681923,
"tokens_seen": 1777664000
},
{
"epoch": 0.54,
"learning_rate": 0.00023305917753259781,
"loss": 2.8105,
"theoretical_loss": 3.4646513813176942,
"tokens_seen": 1777729536
},
{
"epoch": 0.54,
"learning_rate": 0.000233049147442327,
"loss": 2.8093,
"theoretical_loss": 3.464640483881428,
"tokens_seen": 1777795072
},
{
"epoch": 0.54,
"learning_rate": 0.00023303911735205618,
"loss": 2.6295,
"theoretical_loss": 3.46462958695935,
"tokens_seen": 1777860608
},
{
"epoch": 0.54,
"learning_rate": 0.00023302908726178538,
"loss": 2.4049,
"theoretical_loss": 3.4646186905514176,
"tokens_seen": 1777926144
},
{
"epoch": 0.54,
"learning_rate": 0.00023301905717151457,
"loss": 2.3726,
"theoretical_loss": 3.464607794657587,
"tokens_seen": 1777991680
},
{
"epoch": 0.54,
"learning_rate": 0.00023300902708124375,
"loss": 2.4853,
"theoretical_loss": 3.464596899277815,
"tokens_seen": 1778057216
},
{
"epoch": 0.54,
"learning_rate": 0.00023299899699097293,
"loss": 2.6193,
"theoretical_loss": 3.464586004412059,
"tokens_seen": 1778122752
},
{
"epoch": 0.54,
"learning_rate": 0.0002329889669007021,
"loss": 2.8306,
"theoretical_loss": 3.464575110060275,
"tokens_seen": 1778188288
},
{
"epoch": 0.54,
"learning_rate": 0.00023297893681043132,
"loss": 2.8287,
"theoretical_loss": 3.4645642162224206,
"tokens_seen": 1778253824
},
{
"epoch": 0.54,
"learning_rate": 0.0002329689067201605,
"loss": 2.5982,
"theoretical_loss": 3.4645533228984515,
"tokens_seen": 1778319360
},
{
"epoch": 0.54,
"learning_rate": 0.00023295887662988968,
"loss": 2.7413,
"theoretical_loss": 3.464542430088326,
"tokens_seen": 1778384896
},
{
"epoch": 0.54,
"learning_rate": 0.00023294884653961886,
"loss": 2.6624,
"theoretical_loss": 3.4645315377920003,
"tokens_seen": 1778450432
},
{
"epoch": 0.54,
"learning_rate": 0.00023293881644934807,
"loss": 2.72,
"theoretical_loss": 3.46452064600943,
"tokens_seen": 1778515968
},
{
"epoch": 0.54,
"learning_rate": 0.00023292878635907725,
"loss": 2.6649,
"theoretical_loss": 3.4645097547405745,
"tokens_seen": 1778581504
},
{
"epoch": 0.54,
"learning_rate": 0.00023291875626880643,
"loss": 2.6235,
"theoretical_loss": 3.4644988639853884,
"tokens_seen": 1778647040
},
{
"epoch": 0.54,
"learning_rate": 0.0002329087261785356,
"loss": 2.6134,
"theoretical_loss": 3.4644879737438297,
"tokens_seen": 1778712576
},
{
"epoch": 0.54,
"learning_rate": 0.0002328986960882648,
"loss": 2.4148,
"theoretical_loss": 3.4644770840158543,
"tokens_seen": 1778778112
},
{
"epoch": 0.54,
"learning_rate": 0.000232888665997994,
"loss": 2.7701,
"theoretical_loss": 3.46446619480142,
"tokens_seen": 1778843648
},
{
"epoch": 0.54,
"learning_rate": 0.00023287863590772318,
"loss": 2.6888,
"theoretical_loss": 3.4644553061004832,
"tokens_seen": 1778909184
},
{
"epoch": 0.54,
"learning_rate": 0.00023286860581745236,
"loss": 2.5306,
"theoretical_loss": 3.4644444179130014,
"tokens_seen": 1778974720
},
{
"epoch": 0.54,
"learning_rate": 0.00023285857572718154,
"loss": 2.4614,
"theoretical_loss": 3.46443353023893,
"tokens_seen": 1779040256
},
{
"epoch": 0.54,
"learning_rate": 0.00023284854563691075,
"loss": 2.6351,
"theoretical_loss": 3.464422643078228,
"tokens_seen": 1779105792
},
{
"epoch": 0.54,
"learning_rate": 0.00023283851554663993,
"loss": 2.4523,
"theoretical_loss": 3.4644117564308505,
"tokens_seen": 1779171328
},
{
"epoch": 0.54,
"learning_rate": 0.00023282848545636911,
"loss": 2.6096,
"theoretical_loss": 3.464400870296755,
"tokens_seen": 1779236864
},
{
"epoch": 0.54,
"objective/train/docs_used": 1006973,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4340240955352783,
"objective/train/theoretical_loss": 3.4643899846758983,
"objective/train/tokens_used": 1799762400,
"theoretical_loss": 3.4643899846758983,
"tokens_seen": 1779302400
},
{
"epoch": 0.54,
"learning_rate": 0.0002328184553660983,
"loss": 2.5905,
"theoretical_loss": 3.4643899846758983,
"tokens_seen": 1779302400
},
{
"epoch": 0.54,
"learning_rate": 0.00023280842527582748,
"loss": 2.5131,
"theoretical_loss": 3.4643790995682373,
"tokens_seen": 1779367936
},
{
"epoch": 0.54,
"learning_rate": 0.00023279839518555669,
"loss": 2.813,
"theoretical_loss": 3.46436821497373,
"tokens_seen": 1779433472
},
{
"epoch": 0.54,
"learning_rate": 0.00023278836509528587,
"loss": 2.6404,
"theoretical_loss": 3.464357330892331,
"tokens_seen": 1779499008
},
{
"epoch": 0.54,
"learning_rate": 0.00023277833500501505,
"loss": 2.7701,
"theoretical_loss": 3.464346447323999,
"tokens_seen": 1779564544
},
{
"epoch": 0.54,
"learning_rate": 0.00023276830491474423,
"loss": 2.396,
"theoretical_loss": 3.4643355642686906,
"tokens_seen": 1779630080
},
{
"epoch": 0.54,
"learning_rate": 0.00023275827482447344,
"loss": 2.853,
"theoretical_loss": 3.464324681726363,
"tokens_seen": 1779695616
},
{
"epoch": 0.54,
"learning_rate": 0.00023274824473420262,
"loss": 2.6742,
"theoretical_loss": 3.4643137996969724,
"tokens_seen": 1779761152
},
{
"epoch": 0.54,
"learning_rate": 0.0002327382146439318,
"loss": 2.4806,
"theoretical_loss": 3.4643029181804756,
"tokens_seen": 1779826688
},
{
"epoch": 0.54,
"learning_rate": 0.00023272818455366098,
"loss": 2.3556,
"theoretical_loss": 3.464292037176831,
"tokens_seen": 1779892224
},
{
"epoch": 0.54,
"learning_rate": 0.00023271815446339016,
"loss": 2.7318,
"theoretical_loss": 3.4642811566859937,
"tokens_seen": 1779957760
},
{
"epoch": 0.54,
"learning_rate": 0.00023270812437311937,
"loss": 2.5109,
"theoretical_loss": 3.4642702767079214,
"tokens_seen": 1780023296
},
{
"epoch": 0.54,
"learning_rate": 0.00023269809428284855,
"loss": 2.8726,
"theoretical_loss": 3.4642593972425715,
"tokens_seen": 1780088832
},
{
"epoch": 0.54,
"learning_rate": 0.00023268806419257773,
"loss": 2.4838,
"theoretical_loss": 3.4642485182899008,
"tokens_seen": 1780154368
},
{
"epoch": 0.54,
"learning_rate": 0.0002326780341023069,
"loss": 2.5083,
"theoretical_loss": 3.464237639849866,
"tokens_seen": 1780219904
},
{
"epoch": 0.54,
"learning_rate": 0.00023266800401203612,
"loss": 2.8025,
"theoretical_loss": 3.464226761922424,
"tokens_seen": 1780285440
},
{
"epoch": 0.54,
"learning_rate": 0.0002326579739217653,
"loss": 2.7234,
"theoretical_loss": 3.464215884507532,
"tokens_seen": 1780350976
},
{
"epoch": 0.54,
"learning_rate": 0.00023264794383149448,
"loss": 2.5438,
"theoretical_loss": 3.464205007605147,
"tokens_seen": 1780416512
},
{
"epoch": 0.54,
"learning_rate": 0.00023263791374122366,
"loss": 2.943,
"theoretical_loss": 3.464194131215226,
"tokens_seen": 1780482048
},
{
"epoch": 0.54,
"learning_rate": 0.00023262788365095287,
"loss": 2.4689,
"theoretical_loss": 3.4641832553377254,
"tokens_seen": 1780547584
},
{
"epoch": 0.54,
"learning_rate": 0.00023261785356068205,
"loss": 2.727,
"theoretical_loss": 3.464172379972603,
"tokens_seen": 1780613120
},
{
"epoch": 0.54,
"learning_rate": 0.00023260782347041124,
"loss": 2.6549,
"theoretical_loss": 3.4641615051198156,
"tokens_seen": 1780678656
},
{
"epoch": 0.54,
"learning_rate": 0.00023259779338014042,
"loss": 2.5165,
"theoretical_loss": 3.4641506307793204,
"tokens_seen": 1780744192
},
{
"epoch": 0.54,
"learning_rate": 0.0002325877632898696,
"loss": 2.4701,
"theoretical_loss": 3.4641397569510737,
"tokens_seen": 1780809728
},
{
"epoch": 0.54,
"learning_rate": 0.0002325777331995988,
"loss": 2.8215,
"theoretical_loss": 3.464128883635033,
"tokens_seen": 1780875264
},
{
"epoch": 0.54,
"objective/train/docs_used": 1008497,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4147377014160156,
"objective/train/theoretical_loss": 3.4641180108311556,
"objective/train/tokens_used": 1801400800,
"theoretical_loss": 3.4641180108311556,
"tokens_seen": 1780940800
},
{
"epoch": 0.54,
"learning_rate": 0.000232567703109328,
"loss": 2.6321,
"theoretical_loss": 3.4641180108311556,
"tokens_seen": 1780940800
},
{
"epoch": 0.54,
"learning_rate": 0.00023255767301905717,
"loss": 2.5545,
"theoretical_loss": 3.4641071385393976,
"tokens_seen": 1781006336
},
{
"epoch": 0.54,
"learning_rate": 0.00023254764292878635,
"loss": 2.8342,
"theoretical_loss": 3.464096266759717,
"tokens_seen": 1781071872
},
{
"epoch": 0.54,
"learning_rate": 0.00023253761283851556,
"loss": 2.5986,
"theoretical_loss": 3.4640853954920705,
"tokens_seen": 1781137408
},
{
"epoch": 0.54,
"learning_rate": 0.00023252758274824474,
"loss": 2.5403,
"theoretical_loss": 3.4640745247364153,
"tokens_seen": 1781202944
},
{
"epoch": 0.54,
"learning_rate": 0.00023251755265797392,
"loss": 2.6456,
"theoretical_loss": 3.4640636544927075,
"tokens_seen": 1781268480
},
{
"epoch": 0.54,
"learning_rate": 0.0002325075225677031,
"loss": 2.7961,
"theoretical_loss": 3.4640527847609057,
"tokens_seen": 1781334016
},
{
"epoch": 0.54,
"learning_rate": 0.00023249749247743228,
"loss": 2.4495,
"theoretical_loss": 3.4640419155409665,
"tokens_seen": 1781399552
},
{
"epoch": 0.54,
"learning_rate": 0.0002324874623871615,
"loss": 2.615,
"theoretical_loss": 3.4640310468328455,
"tokens_seen": 1781465088
},
{
"epoch": 0.54,
"learning_rate": 0.00023247743229689067,
"loss": 2.4202,
"theoretical_loss": 3.4640201786365017,
"tokens_seen": 1781530624
},
{
"epoch": 0.54,
"learning_rate": 0.00023246740220661985,
"loss": 2.726,
"theoretical_loss": 3.4640093109518917,
"tokens_seen": 1781596160
},
{
"epoch": 0.54,
"learning_rate": 0.00023245737211634903,
"loss": 2.4844,
"theoretical_loss": 3.463998443778972,
"tokens_seen": 1781661696
},
{
"epoch": 0.54,
"learning_rate": 0.00023244734202607824,
"loss": 2.6153,
"theoretical_loss": 3.4639875771177,
"tokens_seen": 1781727232
},
{
"epoch": 0.54,
"learning_rate": 0.00023243731193580742,
"loss": 2.5647,
"theoretical_loss": 3.4639767109680326,
"tokens_seen": 1781792768
},
{
"epoch": 0.54,
"learning_rate": 0.0002324272818455366,
"loss": 2.5251,
"theoretical_loss": 3.463965845329927,
"tokens_seen": 1781858304
},
{
"epoch": 0.54,
"learning_rate": 0.00023241725175526578,
"loss": 2.6206,
"theoretical_loss": 3.46395498020334,
"tokens_seen": 1781923840
},
{
"epoch": 0.54,
"learning_rate": 0.00023240722166499497,
"loss": 2.6651,
"theoretical_loss": 3.4639441155882302,
"tokens_seen": 1781989376
},
{
"epoch": 0.54,
"learning_rate": 0.00023239719157472417,
"loss": 2.7277,
"theoretical_loss": 3.463933251484553,
"tokens_seen": 1782054912
},
{
"epoch": 0.54,
"learning_rate": 0.00023238716148445336,
"loss": 2.6994,
"theoretical_loss": 3.463922387892266,
"tokens_seen": 1782120448
},
{
"epoch": 0.54,
"learning_rate": 0.00023237713139418254,
"loss": 2.378,
"theoretical_loss": 3.4639115248113264,
"tokens_seen": 1782185984
},
{
"epoch": 0.54,
"learning_rate": 0.00023236710130391172,
"loss": 2.5169,
"theoretical_loss": 3.4639006622416915,
"tokens_seen": 1782251520
},
{
"epoch": 0.54,
"learning_rate": 0.00023235707121364095,
"loss": 2.3532,
"theoretical_loss": 3.4638898001833183,
"tokens_seen": 1782317056
},
{
"epoch": 0.54,
"learning_rate": 0.00023234704112337013,
"loss": 2.6716,
"theoretical_loss": 3.463878938636164,
"tokens_seen": 1782382592
},
{
"epoch": 0.54,
"learning_rate": 0.00023233701103309932,
"loss": 2.7313,
"theoretical_loss": 3.4638680776001856,
"tokens_seen": 1782448128
},
{
"epoch": 0.54,
"learning_rate": 0.0002323269809428285,
"loss": 2.7991,
"theoretical_loss": 3.4638572170753408,
"tokens_seen": 1782513664
},
{
"epoch": 0.54,
"objective/train/docs_used": 1009251,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7662036418914795,
"objective/train/theoretical_loss": 3.4638463570615854,
"objective/train/tokens_used": 1803039200,
"theoretical_loss": 3.4638463570615854,
"tokens_seen": 1782579200
},
{
"epoch": 0.54,
"learning_rate": 0.00023231695085255768,
"loss": 2.833,
"theoretical_loss": 3.4638463570615854,
"tokens_seen": 1782579200
},
{
"epoch": 0.54,
"learning_rate": 0.00023230692076228689,
"loss": 2.621,
"theoretical_loss": 3.463835497558878,
"tokens_seen": 1782644736
},
{
"epoch": 0.54,
"learning_rate": 0.00023229689067201607,
"loss": 2.7068,
"theoretical_loss": 3.463824638567175,
"tokens_seen": 1782710272
},
{
"epoch": 0.54,
"learning_rate": 0.00023228686058174525,
"loss": 2.4293,
"theoretical_loss": 3.4638137800864337,
"tokens_seen": 1782775808
},
{
"epoch": 0.54,
"learning_rate": 0.00023227683049147443,
"loss": 2.7233,
"theoretical_loss": 3.463802922116612,
"tokens_seen": 1782841344
},
{
"epoch": 0.54,
"learning_rate": 0.00023226680040120364,
"loss": 2.5339,
"theoretical_loss": 3.4637920646576656,
"tokens_seen": 1782906880
},
{
"epoch": 0.54,
"learning_rate": 0.00023225677031093282,
"loss": 2.7305,
"theoretical_loss": 3.463781207709553,
"tokens_seen": 1782972416
},
{
"epoch": 0.54,
"learning_rate": 0.000232246740220662,
"loss": 2.4868,
"theoretical_loss": 3.463770351272231,
"tokens_seen": 1783037952
},
{
"epoch": 0.54,
"learning_rate": 0.00023223671013039118,
"loss": 2.6533,
"theoretical_loss": 3.4637594953456565,
"tokens_seen": 1783103488
},
{
"epoch": 0.54,
"learning_rate": 0.00023222668004012036,
"loss": 2.9597,
"theoretical_loss": 3.463748639929787,
"tokens_seen": 1783169024
},
{
"epoch": 0.54,
"learning_rate": 0.00023221664994984957,
"loss": 2.6902,
"theoretical_loss": 3.4637377850245796,
"tokens_seen": 1783234560
},
{
"epoch": 0.54,
"learning_rate": 0.00023220661985957875,
"loss": 2.8305,
"theoretical_loss": 3.4637269306299916,
"tokens_seen": 1783300096
},
{
"epoch": 0.54,
"learning_rate": 0.00023219658976930793,
"loss": 2.6873,
"theoretical_loss": 3.46371607674598,
"tokens_seen": 1783365632
},
{
"epoch": 0.54,
"learning_rate": 0.0002321865596790371,
"loss": 2.636,
"theoretical_loss": 3.463705223372502,
"tokens_seen": 1783431168
},
{
"epoch": 0.54,
"learning_rate": 0.00023217652958876632,
"loss": 2.8623,
"theoretical_loss": 3.463694370509515,
"tokens_seen": 1783496704
},
{
"epoch": 0.54,
"learning_rate": 0.0002321664994984955,
"loss": 2.7636,
"theoretical_loss": 3.4636835181569765,
"tokens_seen": 1783562240
},
{
"epoch": 0.54,
"learning_rate": 0.00023215646940822468,
"loss": 2.7148,
"theoretical_loss": 3.4636726663148436,
"tokens_seen": 1783627776
},
{
"epoch": 0.54,
"learning_rate": 0.00023214643931795386,
"loss": 2.6213,
"theoretical_loss": 3.463661814983073,
"tokens_seen": 1783693312
},
{
"epoch": 0.54,
"learning_rate": 0.00023213640922768307,
"loss": 2.4321,
"theoretical_loss": 3.463650964161623,
"tokens_seen": 1783758848
},
{
"epoch": 0.54,
"learning_rate": 0.00023212637913741225,
"loss": 2.677,
"theoretical_loss": 3.4636401138504493,
"tokens_seen": 1783824384
},
{
"epoch": 0.54,
"learning_rate": 0.00023211634904714144,
"loss": 2.6166,
"theoretical_loss": 3.4636292640495103,
"tokens_seen": 1783889920
},
{
"epoch": 0.54,
"learning_rate": 0.00023210631895687062,
"loss": 2.6973,
"theoretical_loss": 3.4636184147587636,
"tokens_seen": 1783955456
},
{
"epoch": 0.54,
"learning_rate": 0.0002320962888665998,
"loss": 2.7192,
"theoretical_loss": 3.4636075659781653,
"tokens_seen": 1784020992
},
{
"epoch": 0.54,
"learning_rate": 0.000232086258776329,
"loss": 2.7284,
"theoretical_loss": 3.4635967177076736,
"tokens_seen": 1784086528
},
{
"epoch": 0.54,
"learning_rate": 0.0002320762286860582,
"loss": 2.7766,
"theoretical_loss": 3.463585869947245,
"tokens_seen": 1784152064
},
{
"epoch": 0.54,
"objective/train/docs_used": 1009780,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3898539543151855,
"objective/train/theoretical_loss": 3.463575022696838,
"objective/train/tokens_used": 1804677600,
"theoretical_loss": 3.463575022696838,
"tokens_seen": 1784217600
},
{
"epoch": 0.54,
"learning_rate": 0.00023206619859578737,
"loss": 2.4492,
"theoretical_loss": 3.463575022696838,
"tokens_seen": 1784217600
},
{
"epoch": 0.54,
"learning_rate": 0.00023205616850551655,
"loss": 2.9576,
"theoretical_loss": 3.4635641759564084,
"tokens_seen": 1784283136
},
{
"epoch": 0.54,
"learning_rate": 0.00023204613841524576,
"loss": 2.541,
"theoretical_loss": 3.4635533297259142,
"tokens_seen": 1784348672
},
{
"epoch": 0.54,
"learning_rate": 0.00023203610832497494,
"loss": 2.4122,
"theoretical_loss": 3.463542484005313,
"tokens_seen": 1784414208
},
{
"epoch": 0.54,
"learning_rate": 0.00023202607823470412,
"loss": 2.5427,
"theoretical_loss": 3.463531638794562,
"tokens_seen": 1784479744
},
{
"epoch": 0.54,
"learning_rate": 0.0002320160481444333,
"loss": 2.8968,
"theoretical_loss": 3.463520794093618,
"tokens_seen": 1784545280
},
{
"epoch": 0.54,
"learning_rate": 0.00023200601805416248,
"loss": 2.5133,
"theoretical_loss": 3.4635099499024387,
"tokens_seen": 1784610816
},
{
"epoch": 0.54,
"learning_rate": 0.0002319959879638917,
"loss": 2.8018,
"theoretical_loss": 3.463499106220981,
"tokens_seen": 1784676352
},
{
"epoch": 0.54,
"learning_rate": 0.00023198595787362087,
"loss": 2.7721,
"theoretical_loss": 3.463488263049203,
"tokens_seen": 1784741888
},
{
"epoch": 0.54,
"learning_rate": 0.00023197592778335005,
"loss": 2.6238,
"theoretical_loss": 3.4634774203870617,
"tokens_seen": 1784807424
},
{
"epoch": 0.54,
"learning_rate": 0.00023196589769307923,
"loss": 2.5674,
"theoretical_loss": 3.463466578234514,
"tokens_seen": 1784872960
},
{
"epoch": 0.54,
"learning_rate": 0.00023195586760280844,
"loss": 2.8772,
"theoretical_loss": 3.463455736591518,
"tokens_seen": 1784938496
},
{
"epoch": 0.54,
"learning_rate": 0.00023194583751253762,
"loss": 2.9253,
"theoretical_loss": 3.4634448954580304,
"tokens_seen": 1785004032
},
{
"epoch": 0.54,
"learning_rate": 0.0002319358074222668,
"loss": 2.6748,
"theoretical_loss": 3.463434054834009,
"tokens_seen": 1785069568
},
{
"epoch": 0.54,
"learning_rate": 0.00023192577733199599,
"loss": 2.7279,
"theoretical_loss": 3.463423214719411,
"tokens_seen": 1785135104
},
{
"epoch": 0.54,
"learning_rate": 0.00023191574724172517,
"loss": 2.7482,
"theoretical_loss": 3.463412375114193,
"tokens_seen": 1785200640
},
{
"epoch": 0.54,
"learning_rate": 0.00023190571715145437,
"loss": 2.5802,
"theoretical_loss": 3.4634015360183135,
"tokens_seen": 1785266176
},
{
"epoch": 0.54,
"learning_rate": 0.00023189568706118356,
"loss": 2.6402,
"theoretical_loss": 3.4633906974317292,
"tokens_seen": 1785331712
},
{
"epoch": 0.54,
"learning_rate": 0.00023188565697091274,
"loss": 2.7365,
"theoretical_loss": 3.463379859354398,
"tokens_seen": 1785397248
},
{
"epoch": 0.54,
"learning_rate": 0.00023187562688064192,
"loss": 2.6029,
"theoretical_loss": 3.463369021786277,
"tokens_seen": 1785462784
},
{
"epoch": 0.54,
"learning_rate": 0.00023186559679037113,
"loss": 2.7824,
"theoretical_loss": 3.463358184727323,
"tokens_seen": 1785528320
},
{
"epoch": 0.54,
"learning_rate": 0.0002318555667001003,
"loss": 2.6819,
"theoretical_loss": 3.4633473481774946,
"tokens_seen": 1785593856
},
{
"epoch": 0.54,
"learning_rate": 0.0002318455366098295,
"loss": 2.6123,
"theoretical_loss": 3.4633365121367485,
"tokens_seen": 1785659392
},
{
"epoch": 0.54,
"learning_rate": 0.00023183550651955867,
"loss": 2.6732,
"theoretical_loss": 3.463325676605042,
"tokens_seen": 1785724928
},
{
"epoch": 0.54,
"learning_rate": 0.00023182547642928785,
"loss": 2.5696,
"theoretical_loss": 3.4633148415823323,
"tokens_seen": 1785790464
},
{
"epoch": 0.54,
"objective/train/docs_used": 1010805,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.0527453422546387,
"objective/train/theoretical_loss": 3.4633040070685777,
"objective/train/tokens_used": 1806316000,
"theoretical_loss": 3.4633040070685777,
"tokens_seen": 1785856000
},
{
"epoch": 0.54,
"learning_rate": 0.00023181544633901706,
"loss": 2.7544,
"theoretical_loss": 3.4633040070685777,
"tokens_seen": 1785856000
},
{
"epoch": 0.54,
"learning_rate": 0.00023180541624874624,
"loss": 2.8198,
"theoretical_loss": 3.463293173063735,
"tokens_seen": 1785921536
},
{
"epoch": 0.54,
"learning_rate": 0.00023179538615847542,
"loss": 2.7906,
"theoretical_loss": 3.463282339567762,
"tokens_seen": 1785987072
},
{
"epoch": 0.54,
"learning_rate": 0.0002317853560682046,
"loss": 2.8655,
"theoretical_loss": 3.4632715065806154,
"tokens_seen": 1786052608
},
{
"epoch": 0.54,
"learning_rate": 0.0002317753259779338,
"loss": 2.7725,
"theoretical_loss": 3.463260674102253,
"tokens_seen": 1786118144
},
{
"epoch": 0.54,
"learning_rate": 0.000231765295887663,
"loss": 2.7607,
"theoretical_loss": 3.4632498421326328,
"tokens_seen": 1786183680
},
{
"epoch": 0.54,
"learning_rate": 0.00023175526579739217,
"loss": 2.6616,
"theoretical_loss": 3.4632390106717112,
"tokens_seen": 1786249216
},
{
"epoch": 0.54,
"learning_rate": 0.00023174523570712135,
"loss": 2.7375,
"theoretical_loss": 3.4632281797194464,
"tokens_seen": 1786314752
},
{
"epoch": 0.54,
"learning_rate": 0.00023173520561685053,
"loss": 2.3533,
"theoretical_loss": 3.463217349275796,
"tokens_seen": 1786380288
},
{
"epoch": 0.54,
"learning_rate": 0.00023172517552657974,
"loss": 2.7731,
"theoretical_loss": 3.4632065193407167,
"tokens_seen": 1786445824
},
{
"epoch": 0.54,
"learning_rate": 0.00023171514543630892,
"loss": 2.4936,
"theoretical_loss": 3.4631956899141665,
"tokens_seen": 1786511360
},
{
"epoch": 0.54,
"learning_rate": 0.0002317051153460381,
"loss": 2.4113,
"theoretical_loss": 3.463184860996103,
"tokens_seen": 1786576896
},
{
"epoch": 0.54,
"learning_rate": 0.00023169508525576729,
"loss": 2.5473,
"theoretical_loss": 3.4631740325864833,
"tokens_seen": 1786642432
},
{
"epoch": 0.54,
"learning_rate": 0.0002316850551654965,
"loss": 2.5822,
"theoretical_loss": 3.463163204685265,
"tokens_seen": 1786707968
},
{
"epoch": 0.54,
"learning_rate": 0.00023167502507522568,
"loss": 2.6592,
"theoretical_loss": 3.4631523772924053,
"tokens_seen": 1786773504
},
{
"epoch": 0.54,
"learning_rate": 0.00023166499498495486,
"loss": 2.767,
"theoretical_loss": 3.463141550407862,
"tokens_seen": 1786839040
},
{
"epoch": 0.54,
"learning_rate": 0.00023165496489468404,
"loss": 2.8778,
"theoretical_loss": 3.4631307240315934,
"tokens_seen": 1786904576
},
{
"epoch": 0.54,
"learning_rate": 0.00023164493480441322,
"loss": 2.7776,
"theoretical_loss": 3.463119898163555,
"tokens_seen": 1786970112
},
{
"epoch": 0.54,
"learning_rate": 0.00023163490471414243,
"loss": 2.7883,
"theoretical_loss": 3.463109072803706,
"tokens_seen": 1787035648
},
{
"epoch": 0.54,
"learning_rate": 0.0002316248746238716,
"loss": 2.6323,
"theoretical_loss": 3.4630982479520034,
"tokens_seen": 1787101184
},
{
"epoch": 0.54,
"learning_rate": 0.0002316148445336008,
"loss": 2.6087,
"theoretical_loss": 3.4630874236084046,
"tokens_seen": 1787166720
},
{
"epoch": 0.54,
"learning_rate": 0.00023160481444333,
"loss": 2.7653,
"theoretical_loss": 3.4630765997728674,
"tokens_seen": 1787232256
},
{
"epoch": 0.54,
"learning_rate": 0.0002315947843530592,
"loss": 2.6548,
"theoretical_loss": 3.463065776445349,
"tokens_seen": 1787297792
},
{
"epoch": 0.54,
"learning_rate": 0.0002315847542627884,
"loss": 2.5718,
"theoretical_loss": 3.463054953625807,
"tokens_seen": 1787363328
},
{
"epoch": 0.54,
"learning_rate": 0.00023157472417251757,
"loss": 2.9228,
"theoretical_loss": 3.4630441313141986,
"tokens_seen": 1787428864
},
{
"epoch": 0.54,
"objective/train/docs_used": 1011138,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.259239435195923,
"objective/train/theoretical_loss": 3.4630333095104824,
"objective/train/tokens_used": 1807954400,
"theoretical_loss": 3.4630333095104824,
"tokens_seen": 1787494400
},
{
"epoch": 0.54,
"learning_rate": 0.00023156469408224675,
"loss": 2.8853,
"theoretical_loss": 3.4630333095104824,
"tokens_seen": 1787494400
},
{
"epoch": 0.54,
"learning_rate": 0.00023155466399197596,
"loss": 2.7082,
"theoretical_loss": 3.463022488214615,
"tokens_seen": 1787559936
},
{
"epoch": 0.54,
"learning_rate": 0.00023154463390170514,
"loss": 2.768,
"theoretical_loss": 3.4630116674265548,
"tokens_seen": 1787625472
},
{
"epoch": 0.54,
"learning_rate": 0.00023153460381143432,
"loss": 2.448,
"theoretical_loss": 3.463000847146258,
"tokens_seen": 1787691008
},
{
"epoch": 0.54,
"learning_rate": 0.0002315245737211635,
"loss": 2.6728,
"theoretical_loss": 3.4629900273736833,
"tokens_seen": 1787756544
},
{
"epoch": 0.54,
"learning_rate": 0.00023151454363089268,
"loss": 2.6791,
"theoretical_loss": 3.462979208108788,
"tokens_seen": 1787822080
},
{
"epoch": 0.54,
"learning_rate": 0.0002315045135406219,
"loss": 2.7053,
"theoretical_loss": 3.462968389351529,
"tokens_seen": 1787887616
},
{
"epoch": 0.54,
"learning_rate": 0.00023149448345035107,
"loss": 2.5126,
"theoretical_loss": 3.4629575711018648,
"tokens_seen": 1787953152
},
{
"epoch": 0.54,
"learning_rate": 0.00023148445336008025,
"loss": 2.6075,
"theoretical_loss": 3.462946753359753,
"tokens_seen": 1788018688
},
{
"epoch": 0.54,
"learning_rate": 0.00023147442326980943,
"loss": 2.5717,
"theoretical_loss": 3.4629359361251506,
"tokens_seen": 1788084224
},
{
"epoch": 0.54,
"learning_rate": 0.00023146439317953864,
"loss": 2.5692,
"theoretical_loss": 3.4629251193980153,
"tokens_seen": 1788149760
},
{
"epoch": 0.54,
"learning_rate": 0.00023145436308926782,
"loss": 2.6387,
"theoretical_loss": 3.462914303178305,
"tokens_seen": 1788215296
},
{
"epoch": 0.54,
"learning_rate": 0.000231444332998997,
"loss": 2.9322,
"theoretical_loss": 3.4629034874659768,
"tokens_seen": 1788280832
},
{
"epoch": 0.54,
"learning_rate": 0.00023143430290872619,
"loss": 2.6838,
"theoretical_loss": 3.4628926722609887,
"tokens_seen": 1788346368
},
{
"epoch": 0.54,
"learning_rate": 0.00023142427281845537,
"loss": 2.6718,
"theoretical_loss": 3.4628818575632985,
"tokens_seen": 1788411904
},
{
"epoch": 0.54,
"learning_rate": 0.00023141424272818457,
"loss": 2.7545,
"theoretical_loss": 3.4628710433728633,
"tokens_seen": 1788477440
},
{
"epoch": 0.54,
"learning_rate": 0.00023140421263791376,
"loss": 2.8303,
"theoretical_loss": 3.462860229689641,
"tokens_seen": 1788542976
},
{
"epoch": 0.54,
"learning_rate": 0.00023139418254764294,
"loss": 2.5016,
"theoretical_loss": 3.4628494165135892,
"tokens_seen": 1788608512
},
{
"epoch": 0.54,
"learning_rate": 0.00023138415245737212,
"loss": 2.518,
"theoretical_loss": 3.4628386038446655,
"tokens_seen": 1788674048
},
{
"epoch": 0.54,
"learning_rate": 0.00023137412236710133,
"loss": 2.7324,
"theoretical_loss": 3.4628277916828276,
"tokens_seen": 1788739584
},
{
"epoch": 0.54,
"learning_rate": 0.0002313640922768305,
"loss": 2.6597,
"theoretical_loss": 3.462816980028033,
"tokens_seen": 1788805120
},
{
"epoch": 0.54,
"learning_rate": 0.0002313540621865597,
"loss": 2.7513,
"theoretical_loss": 3.46280616888024,
"tokens_seen": 1788870656
},
{
"epoch": 0.54,
"learning_rate": 0.00023134403209628887,
"loss": 2.8277,
"theoretical_loss": 3.462795358239405,
"tokens_seen": 1788936192
},
{
"epoch": 0.54,
"learning_rate": 0.00023133400200601805,
"loss": 2.7675,
"theoretical_loss": 3.462784548105487,
"tokens_seen": 1789001728
},
{
"epoch": 0.54,
"learning_rate": 0.00023132397191574726,
"loss": 2.7186,
"theoretical_loss": 3.4627737384784423,
"tokens_seen": 1789067264
},
{
"epoch": 0.54,
"objective/train/docs_used": 1012559,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.914393901824951,
"objective/train/theoretical_loss": 3.46276292935823,
"objective/train/tokens_used": 1809592800,
"theoretical_loss": 3.46276292935823,
"tokens_seen": 1789132800
},
{
"epoch": 0.54,
"learning_rate": 0.00023131394182547644,
"loss": 2.8175,
"theoretical_loss": 3.46276292935823,
"tokens_seen": 1789132800
},
{
"epoch": 0.54,
"learning_rate": 0.00023130391173520562,
"loss": 2.5053,
"theoretical_loss": 3.4627521207448066,
"tokens_seen": 1789198336
},
{
"epoch": 0.54,
"learning_rate": 0.0002312938816449348,
"loss": 2.5868,
"theoretical_loss": 3.4627413126381303,
"tokens_seen": 1789263872
},
{
"epoch": 0.54,
"learning_rate": 0.000231283851554664,
"loss": 2.8069,
"theoretical_loss": 3.462730505038159,
"tokens_seen": 1789329408
},
{
"epoch": 0.54,
"learning_rate": 0.0002312738214643932,
"loss": 2.6258,
"theoretical_loss": 3.46271969794485,
"tokens_seen": 1789394944
},
{
"epoch": 0.54,
"learning_rate": 0.00023126379137412237,
"loss": 2.5565,
"theoretical_loss": 3.4627088913581607,
"tokens_seen": 1789460480
},
{
"epoch": 0.54,
"learning_rate": 0.00023125376128385155,
"loss": 2.4959,
"theoretical_loss": 3.4626980852780496,
"tokens_seen": 1789526016
},
{
"epoch": 0.54,
"learning_rate": 0.00023124373119358073,
"loss": 2.6798,
"theoretical_loss": 3.4626872797044737,
"tokens_seen": 1789591552
},
{
"epoch": 0.54,
"learning_rate": 0.00023123370110330994,
"loss": 2.8085,
"theoretical_loss": 3.4626764746373913,
"tokens_seen": 1789657088
},
{
"epoch": 0.54,
"learning_rate": 0.00023122367101303912,
"loss": 2.8319,
"theoretical_loss": 3.4626656700767597,
"tokens_seen": 1789722624
},
{
"epoch": 0.54,
"learning_rate": 0.0002312136409227683,
"loss": 2.7105,
"theoretical_loss": 3.462654866022537,
"tokens_seen": 1789788160
},
{
"epoch": 0.54,
"learning_rate": 0.0002312036108324975,
"loss": 2.5143,
"theoretical_loss": 3.4626440624746797,
"tokens_seen": 1789853696
},
{
"epoch": 0.54,
"learning_rate": 0.0002311935807422267,
"loss": 2.6048,
"theoretical_loss": 3.4626332594331473,
"tokens_seen": 1789919232
},
{
"epoch": 0.54,
"learning_rate": 0.00023118355065195588,
"loss": 2.6016,
"theoretical_loss": 3.462622456897897,
"tokens_seen": 1789984768
},
{
"epoch": 0.54,
"learning_rate": 0.00023117352056168506,
"loss": 2.5794,
"theoretical_loss": 3.4626116548688857,
"tokens_seen": 1790050304
},
{
"epoch": 0.54,
"learning_rate": 0.00023116349047141424,
"loss": 2.804,
"theoretical_loss": 3.462600853346072,
"tokens_seen": 1790115840
},
{
"epoch": 0.54,
"learning_rate": 0.00023115346038114342,
"loss": 2.7752,
"theoretical_loss": 3.4625900523294133,
"tokens_seen": 1790181376
},
{
"epoch": 0.54,
"learning_rate": 0.00023114343029087263,
"loss": 2.6074,
"theoretical_loss": 3.462579251818867,
"tokens_seen": 1790246912
},
{
"epoch": 0.54,
"learning_rate": 0.0002311334002006018,
"loss": 2.739,
"theoretical_loss": 3.4625684518143918,
"tokens_seen": 1790312448
},
{
"epoch": 0.54,
"learning_rate": 0.000231123370110331,
"loss": 2.5022,
"theoretical_loss": 3.4625576523159447,
"tokens_seen": 1790377984
},
{
"epoch": 0.54,
"learning_rate": 0.00023111334002006017,
"loss": 2.5835,
"theoretical_loss": 3.4625468533234836,
"tokens_seen": 1790443520
},
{
"epoch": 0.54,
"learning_rate": 0.00023110330992978938,
"loss": 2.9041,
"theoretical_loss": 3.4625360548369666,
"tokens_seen": 1790509056
},
{
"epoch": 0.54,
"learning_rate": 0.00023109327983951856,
"loss": 2.5172,
"theoretical_loss": 3.462525256856351,
"tokens_seen": 1790574592
},
{
"epoch": 0.54,
"learning_rate": 0.00023108324974924774,
"loss": 2.8316,
"theoretical_loss": 3.462514459381595,
"tokens_seen": 1790640128
},
{
"epoch": 0.54,
"learning_rate": 0.00023107321965897692,
"loss": 2.5579,
"theoretical_loss": 3.462503662412656,
"tokens_seen": 1790705664
},
{
"epoch": 0.54,
"objective/train/docs_used": 1013169,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5323455333709717,
"objective/train/theoretical_loss": 3.4624928659494927,
"objective/train/tokens_used": 1811231200,
"theoretical_loss": 3.4624928659494927,
"tokens_seen": 1790771200
},
{
"epoch": 0.54,
"learning_rate": 0.0002310631895687061,
"loss": 2.5241,
"theoretical_loss": 3.4624928659494927,
"tokens_seen": 1790771200
},
{
"epoch": 0.54,
"learning_rate": 0.0002310531594784353,
"loss": 2.7941,
"theoretical_loss": 3.4624820699920615,
"tokens_seen": 1790836736
},
{
"epoch": 0.54,
"learning_rate": 0.0002310431293881645,
"loss": 2.472,
"theoretical_loss": 3.462471274540321,
"tokens_seen": 1790902272
},
{
"epoch": 0.54,
"learning_rate": 0.00023103309929789367,
"loss": 2.6845,
"theoretical_loss": 3.4624604795942293,
"tokens_seen": 1790967808
},
{
"epoch": 0.54,
"learning_rate": 0.00023102306920762286,
"loss": 2.6005,
"theoretical_loss": 3.462449685153744,
"tokens_seen": 1791033344
},
{
"epoch": 0.54,
"learning_rate": 0.00023101303911735206,
"loss": 2.462,
"theoretical_loss": 3.462438891218822,
"tokens_seen": 1791098880
},
{
"epoch": 0.54,
"learning_rate": 0.00023100300902708124,
"loss": 2.7168,
"theoretical_loss": 3.4624280977894224,
"tokens_seen": 1791164416
},
{
"epoch": 0.54,
"learning_rate": 0.00023099297893681043,
"loss": 2.8138,
"theoretical_loss": 3.462417304865502,
"tokens_seen": 1791229952
},
{
"epoch": 0.54,
"learning_rate": 0.0002309829488465396,
"loss": 2.6321,
"theoretical_loss": 3.4624065124470196,
"tokens_seen": 1791295488
},
{
"epoch": 0.54,
"learning_rate": 0.00023097291875626881,
"loss": 2.4497,
"theoretical_loss": 3.462395720533933,
"tokens_seen": 1791361024
},
{
"epoch": 0.54,
"learning_rate": 0.000230962888665998,
"loss": 2.8562,
"theoretical_loss": 3.462384929126199,
"tokens_seen": 1791426560
},
{
"epoch": 0.54,
"learning_rate": 0.00023095285857572718,
"loss": 2.831,
"theoretical_loss": 3.462374138223776,
"tokens_seen": 1791492096
},
{
"epoch": 0.54,
"learning_rate": 0.00023094282848545636,
"loss": 2.5548,
"theoretical_loss": 3.4623633478266225,
"tokens_seen": 1791557632
},
{
"epoch": 0.54,
"learning_rate": 0.00023093279839518554,
"loss": 2.7597,
"theoretical_loss": 3.4623525579346954,
"tokens_seen": 1791623168
},
{
"epoch": 0.54,
"learning_rate": 0.00023092276830491475,
"loss": 2.5487,
"theoretical_loss": 3.462341768547953,
"tokens_seen": 1791688704
},
{
"epoch": 0.54,
"learning_rate": 0.00023091273821464393,
"loss": 2.6556,
"theoretical_loss": 3.462330979666353,
"tokens_seen": 1791754240
},
{
"epoch": 0.54,
"learning_rate": 0.0002309027081243731,
"loss": 2.3974,
"theoretical_loss": 3.462320191289854,
"tokens_seen": 1791819776
},
{
"epoch": 0.54,
"learning_rate": 0.0002308926780341023,
"loss": 2.4414,
"theoretical_loss": 3.4623094034184128,
"tokens_seen": 1791885312
},
{
"epoch": 0.54,
"learning_rate": 0.0002308826479438315,
"loss": 2.5987,
"theoretical_loss": 3.4622986160519877,
"tokens_seen": 1791950848
},
{
"epoch": 0.54,
"learning_rate": 0.00023087261785356068,
"loss": 2.7139,
"theoretical_loss": 3.4622878291905366,
"tokens_seen": 1792016384
},
{
"epoch": 0.54,
"learning_rate": 0.00023086258776328986,
"loss": 2.5635,
"theoretical_loss": 3.4622770428340184,
"tokens_seen": 1792081920
},
{
"epoch": 0.54,
"learning_rate": 0.00023085255767301907,
"loss": 2.6961,
"theoretical_loss": 3.4622662569823888,
"tokens_seen": 1792147456
},
{
"epoch": 0.54,
"learning_rate": 0.00023084252758274825,
"loss": 2.4523,
"theoretical_loss": 3.4622554716356078,
"tokens_seen": 1792212992
},
{
"epoch": 0.54,
"learning_rate": 0.00023083249749247746,
"loss": 2.9522,
"theoretical_loss": 3.462244686793632,
"tokens_seen": 1792278528
},
{
"epoch": 0.54,
"learning_rate": 0.00023082246740220664,
"loss": 2.4954,
"theoretical_loss": 3.46223390245642,
"tokens_seen": 1792344064
},
{
"epoch": 0.54,
"objective/train/docs_used": 1014342,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.336982250213623,
"objective/train/theoretical_loss": 3.4622231186239296,
"objective/train/tokens_used": 1812869600,
"theoretical_loss": 3.4622231186239296,
"tokens_seen": 1792409600
},
{
"epoch": 0.54,
"learning_rate": 0.00023081243731193582,
"loss": 2.5574,
"theoretical_loss": 3.4622231186239296,
"tokens_seen": 1792409600
},
{
"epoch": 0.54,
"learning_rate": 0.000230802407221665,
"loss": 2.7628,
"theoretical_loss": 3.4622123352961185,
"tokens_seen": 1792475136
},
{
"epoch": 0.54,
"learning_rate": 0.0002307923771313942,
"loss": 2.6785,
"theoretical_loss": 3.462201552472945,
"tokens_seen": 1792540672
},
{
"epoch": 0.54,
"learning_rate": 0.0002307823470411234,
"loss": 2.6168,
"theoretical_loss": 3.4621907701543666,
"tokens_seen": 1792606208
},
{
"epoch": 0.54,
"learning_rate": 0.00023077231695085257,
"loss": 2.5759,
"theoretical_loss": 3.462179988340341,
"tokens_seen": 1792671744
},
{
"epoch": 0.54,
"learning_rate": 0.00023076228686058175,
"loss": 2.6671,
"theoretical_loss": 3.4621692070308274,
"tokens_seen": 1792737280
},
{
"epoch": 0.54,
"learning_rate": 0.00023075225677031094,
"loss": 2.5498,
"theoretical_loss": 3.4621584262257823,
"tokens_seen": 1792802816
},
{
"epoch": 0.54,
"learning_rate": 0.00023074222668004014,
"loss": 2.6399,
"theoretical_loss": 3.4621476459251648,
"tokens_seen": 1792868352
},
{
"epoch": 0.54,
"learning_rate": 0.00023073219658976932,
"loss": 2.7315,
"theoretical_loss": 3.4621368661289322,
"tokens_seen": 1792933888
},
{
"epoch": 0.54,
"learning_rate": 0.0002307221664994985,
"loss": 2.6579,
"theoretical_loss": 3.462126086837042,
"tokens_seen": 1792999424
},
{
"epoch": 0.54,
"learning_rate": 0.0002307121364092277,
"loss": 2.6615,
"theoretical_loss": 3.4621153080494538,
"tokens_seen": 1793064960
},
{
"epoch": 0.54,
"learning_rate": 0.0002307021063189569,
"loss": 2.426,
"theoretical_loss": 3.462104529766124,
"tokens_seen": 1793130496
},
{
"epoch": 0.54,
"learning_rate": 0.00023069207622868608,
"loss": 2.567,
"theoretical_loss": 3.462093751987011,
"tokens_seen": 1793196032
},
{
"epoch": 0.54,
"learning_rate": 0.00023068204613841526,
"loss": 2.8557,
"theoretical_loss": 3.4620829747120734,
"tokens_seen": 1793261568
},
{
"epoch": 0.54,
"learning_rate": 0.00023067201604814444,
"loss": 2.6829,
"theoretical_loss": 3.4620721979412687,
"tokens_seen": 1793327104
},
{
"epoch": 0.54,
"learning_rate": 0.00023066198595787362,
"loss": 2.5436,
"theoretical_loss": 3.462061421674554,
"tokens_seen": 1793392640
},
{
"epoch": 0.54,
"learning_rate": 0.00023065195586760283,
"loss": 2.6172,
"theoretical_loss": 3.46205064591189,
"tokens_seen": 1793458176
},
{
"epoch": 0.54,
"learning_rate": 0.000230641925777332,
"loss": 2.6146,
"theoretical_loss": 3.462039870653231,
"tokens_seen": 1793523712
},
{
"epoch": 0.54,
"learning_rate": 0.0002306318956870612,
"loss": 2.5651,
"theoretical_loss": 3.4620290958985382,
"tokens_seen": 1793589248
},
{
"epoch": 0.54,
"learning_rate": 0.00023062186559679037,
"loss": 2.5104,
"theoretical_loss": 3.462018321647768,
"tokens_seen": 1793654784
},
{
"epoch": 0.54,
"learning_rate": 0.00023061183550651958,
"loss": 2.8846,
"theoretical_loss": 3.462007547900879,
"tokens_seen": 1793720320
},
{
"epoch": 0.54,
"learning_rate": 0.00023060180541624876,
"loss": 2.6872,
"theoretical_loss": 3.4619967746578286,
"tokens_seen": 1793785856
},
{
"epoch": 0.54,
"learning_rate": 0.00023059177532597794,
"loss": 2.5191,
"theoretical_loss": 3.4619860019185755,
"tokens_seen": 1793851392
},
{
"epoch": 0.54,
"learning_rate": 0.00023058174523570712,
"loss": 2.9584,
"theoretical_loss": 3.4619752296830772,
"tokens_seen": 1793916928
},
{
"epoch": 0.54,
"learning_rate": 0.0002305717151454363,
"loss": 2.6816,
"theoretical_loss": 3.461964457951292,
"tokens_seen": 1793982464
},
{
"epoch": 0.54,
"objective/train/docs_used": 1015045,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.727816343307495,
"objective/train/theoretical_loss": 3.461953686723178,
"objective/train/tokens_used": 1814508000,
"theoretical_loss": 3.461953686723178,
"tokens_seen": 1794048000
},
{
"epoch": 0.54,
"learning_rate": 0.0002305616850551655,
"loss": 2.6138,
"theoretical_loss": 3.461953686723178,
"tokens_seen": 1794048000
},
{
"epoch": 0.54,
"learning_rate": 0.0002305516549648947,
"loss": 2.9462,
"theoretical_loss": 3.461942915998694,
"tokens_seen": 1794113536
},
{
"epoch": 0.54,
"learning_rate": 0.00023054162487462387,
"loss": 2.5719,
"theoretical_loss": 3.4619321457777965,
"tokens_seen": 1794179072
},
{
"epoch": 0.54,
"learning_rate": 0.00023053159478435306,
"loss": 2.5914,
"theoretical_loss": 3.4619213760604444,
"tokens_seen": 1794244608
},
{
"epoch": 0.54,
"learning_rate": 0.00023052156469408226,
"loss": 2.6493,
"theoretical_loss": 3.4619106068465957,
"tokens_seen": 1794310144
},
{
"epoch": 0.54,
"learning_rate": 0.00023051153460381144,
"loss": 2.6281,
"theoretical_loss": 3.4618998381362087,
"tokens_seen": 1794375680
},
{
"epoch": 0.54,
"learning_rate": 0.00023050150451354063,
"loss": 2.6562,
"theoretical_loss": 3.4618890699292413,
"tokens_seen": 1794441216
},
{
"epoch": 0.54,
"learning_rate": 0.0002304914744232698,
"loss": 2.6668,
"theoretical_loss": 3.461878302225651,
"tokens_seen": 1794506752
},
{
"epoch": 0.54,
"learning_rate": 0.00023048144433299902,
"loss": 2.8513,
"theoretical_loss": 3.4618675350253967,
"tokens_seen": 1794572288
},
{
"epoch": 0.54,
"learning_rate": 0.0002304714142427282,
"loss": 2.6781,
"theoretical_loss": 3.4618567683284365,
"tokens_seen": 1794637824
},
{
"epoch": 0.54,
"learning_rate": 0.00023046138415245738,
"loss": 2.701,
"theoretical_loss": 3.4618460021347275,
"tokens_seen": 1794703360
},
{
"epoch": 0.54,
"learning_rate": 0.00023045135406218656,
"loss": 2.9263,
"theoretical_loss": 3.4618352364442293,
"tokens_seen": 1794768896
},
{
"epoch": 0.54,
"learning_rate": 0.00023044132397191574,
"loss": 2.7185,
"theoretical_loss": 3.4618244712568984,
"tokens_seen": 1794834432
},
{
"epoch": 0.54,
"learning_rate": 0.00023043129388164495,
"loss": 2.7028,
"theoretical_loss": 3.4618137065726944,
"tokens_seen": 1794899968
},
{
"epoch": 0.54,
"learning_rate": 0.00023042126379137413,
"loss": 2.6061,
"theoretical_loss": 3.4618029423915746,
"tokens_seen": 1794965504
},
{
"epoch": 0.54,
"learning_rate": 0.0002304112337011033,
"loss": 2.7274,
"theoretical_loss": 3.461792178713497,
"tokens_seen": 1795031040
},
{
"epoch": 0.54,
"learning_rate": 0.0002304012036108325,
"loss": 2.7342,
"theoretical_loss": 3.46178141553842,
"tokens_seen": 1795096576
},
{
"epoch": 0.54,
"learning_rate": 0.0002303911735205617,
"loss": 2.8338,
"theoretical_loss": 3.461770652866302,
"tokens_seen": 1795162112
},
{
"epoch": 0.54,
"learning_rate": 0.00023038114343029088,
"loss": 2.4867,
"theoretical_loss": 3.4617598906971008,
"tokens_seen": 1795227648
},
{
"epoch": 0.54,
"learning_rate": 0.00023037111334002006,
"loss": 2.5932,
"theoretical_loss": 3.4617491290307743,
"tokens_seen": 1795293184
},
{
"epoch": 0.54,
"learning_rate": 0.00023036108324974924,
"loss": 2.7987,
"theoretical_loss": 3.461738367867281,
"tokens_seen": 1795358720
},
{
"epoch": 0.54,
"learning_rate": 0.00023035105315947842,
"loss": 2.4878,
"theoretical_loss": 3.461727607206579,
"tokens_seen": 1795424256
},
{
"epoch": 0.54,
"learning_rate": 0.00023034102306920763,
"loss": 2.7851,
"theoretical_loss": 3.461716847048627,
"tokens_seen": 1795489792
},
{
"epoch": 0.54,
"learning_rate": 0.0002303309929789368,
"loss": 2.6589,
"theoretical_loss": 3.461706087393382,
"tokens_seen": 1795555328
},
{
"epoch": 0.54,
"learning_rate": 0.000230320962888666,
"loss": 2.5346,
"theoretical_loss": 3.461695328240803,
"tokens_seen": 1795620864
},
{
"epoch": 0.54,
"objective/train/docs_used": 1016473,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1854147911071777,
"objective/train/theoretical_loss": 3.461684569590848,
"objective/train/tokens_used": 1816146400,
"theoretical_loss": 3.461684569590848,
"tokens_seen": 1795686400
},
{
"epoch": 0.54,
"learning_rate": 0.00023031093279839518,
"loss": 2.7095,
"theoretical_loss": 3.461684569590848,
"tokens_seen": 1795686400
},
{
"epoch": 0.54,
"learning_rate": 0.00023030090270812438,
"loss": 2.6162,
"theoretical_loss": 3.461673811443475,
"tokens_seen": 1795751936
},
{
"epoch": 0.54,
"learning_rate": 0.00023029087261785356,
"loss": 2.5227,
"theoretical_loss": 3.4616630537986426,
"tokens_seen": 1795817472
},
{
"epoch": 0.54,
"learning_rate": 0.00023028084252758275,
"loss": 2.6538,
"theoretical_loss": 3.4616522966563084,
"tokens_seen": 1795883008
},
{
"epoch": 0.54,
"learning_rate": 0.00023027081243731193,
"loss": 2.851,
"theoretical_loss": 3.461641540016431,
"tokens_seen": 1795948544
},
{
"epoch": 0.54,
"learning_rate": 0.0002302607823470411,
"loss": 2.6974,
"theoretical_loss": 3.4616307838789684,
"tokens_seen": 1796014080
},
{
"epoch": 0.54,
"learning_rate": 0.00023025075225677032,
"loss": 2.578,
"theoretical_loss": 3.4616200282438783,
"tokens_seen": 1796079616
},
{
"epoch": 0.54,
"learning_rate": 0.0002302407221664995,
"loss": 2.6867,
"theoretical_loss": 3.4616092731111205,
"tokens_seen": 1796145152
},
{
"epoch": 0.54,
"learning_rate": 0.00023023069207622868,
"loss": 2.5434,
"theoretical_loss": 3.4615985184806517,
"tokens_seen": 1796210688
},
{
"epoch": 0.54,
"learning_rate": 0.00023022066198595786,
"loss": 2.6489,
"theoretical_loss": 3.461587764352431,
"tokens_seen": 1796276224
},
{
"epoch": 0.54,
"learning_rate": 0.00023021063189568707,
"loss": 2.7372,
"theoretical_loss": 3.4615770107264154,
"tokens_seen": 1796341760
},
{
"epoch": 0.54,
"learning_rate": 0.00023020060180541625,
"loss": 2.5057,
"theoretical_loss": 3.461566257602565,
"tokens_seen": 1796407296
},
{
"epoch": 0.54,
"learning_rate": 0.00023019057171514543,
"loss": 2.6958,
"theoretical_loss": 3.4615555049808364,
"tokens_seen": 1796472832
},
{
"epoch": 0.54,
"learning_rate": 0.0002301805416248746,
"loss": 2.748,
"theoretical_loss": 3.4615447528611885,
"tokens_seen": 1796538368
},
{
"epoch": 0.54,
"learning_rate": 0.0002301705115346038,
"loss": 2.67,
"theoretical_loss": 3.46153400124358,
"tokens_seen": 1796603904
},
{
"epoch": 0.54,
"learning_rate": 0.000230160481444333,
"loss": 2.772,
"theoretical_loss": 3.461523250127968,
"tokens_seen": 1796669440
},
{
"epoch": 0.54,
"learning_rate": 0.00023015045135406218,
"loss": 2.5489,
"theoretical_loss": 3.4615124995143116,
"tokens_seen": 1796734976
},
{
"epoch": 0.54,
"learning_rate": 0.00023014042126379136,
"loss": 2.6411,
"theoretical_loss": 3.4615017494025686,
"tokens_seen": 1796800512
},
{
"epoch": 0.54,
"learning_rate": 0.00023013039117352054,
"loss": 2.6635,
"theoretical_loss": 3.461490999792698,
"tokens_seen": 1796866048
},
{
"epoch": 0.54,
"learning_rate": 0.00023012036108324975,
"loss": 2.5644,
"theoretical_loss": 3.461480250684657,
"tokens_seen": 1796931584
},
{
"epoch": 0.54,
"learning_rate": 0.00023011033099297893,
"loss": 2.6843,
"theoretical_loss": 3.4614695020784048,
"tokens_seen": 1796997120
},
{
"epoch": 0.54,
"learning_rate": 0.00023010030090270814,
"loss": 2.7799,
"theoretical_loss": 3.4614587539738992,
"tokens_seen": 1797062656
},
{
"epoch": 0.54,
"learning_rate": 0.00023009027081243732,
"loss": 2.6092,
"theoretical_loss": 3.4614480063710986,
"tokens_seen": 1797128192
},
{
"epoch": 0.54,
"learning_rate": 0.0002300802407221665,
"loss": 2.5475,
"theoretical_loss": 3.461437259269961,
"tokens_seen": 1797193728
},
{
"epoch": 0.54,
"learning_rate": 0.0002300702106318957,
"loss": 2.7429,
"theoretical_loss": 3.4614265126704455,
"tokens_seen": 1797259264
},
{
"epoch": 0.54,
"objective/train/docs_used": 1017203,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.150568723678589,
"objective/train/theoretical_loss": 3.46141576657251,
"objective/train/tokens_used": 1817784800,
"theoretical_loss": 3.46141576657251,
"tokens_seen": 1797324800
},
{
"epoch": 0.54,
"learning_rate": 0.0002300601805416249,
"loss": 2.3272,
"theoretical_loss": 3.46141576657251,
"tokens_seen": 1797324800
},
{
"epoch": 0.54,
"learning_rate": 0.00023005015045135407,
"loss": 2.3979,
"theoretical_loss": 3.461405020976112,
"tokens_seen": 1797390336
},
{
"epoch": 0.54,
"learning_rate": 0.00023004012036108326,
"loss": 2.6691,
"theoretical_loss": 3.4613942758812106,
"tokens_seen": 1797455872
},
{
"epoch": 0.54,
"learning_rate": 0.00023003009027081246,
"loss": 2.4962,
"theoretical_loss": 3.4613835312877645,
"tokens_seen": 1797521408
},
{
"epoch": 0.54,
"learning_rate": 0.00023002006018054164,
"loss": 2.7062,
"theoretical_loss": 3.461372787195731,
"tokens_seen": 1797586944
},
{
"epoch": 0.54,
"learning_rate": 0.00023001003009027083,
"loss": 2.738,
"theoretical_loss": 3.461362043605069,
"tokens_seen": 1797652480
},
{
"epoch": 0.54,
"learning_rate": 0.00023,
"loss": 2.5749,
"theoretical_loss": 3.4613513005157373,
"tokens_seen": 1797718016
},
{
"epoch": 0.54,
"learning_rate": 0.00022998996990972922,
"loss": 2.6642,
"theoretical_loss": 3.4613405579276932,
"tokens_seen": 1797783552
},
{
"epoch": 0.54,
"learning_rate": 0.0002299799398194584,
"loss": 2.5781,
"theoretical_loss": 3.4613298158408954,
"tokens_seen": 1797849088
},
{
"epoch": 0.54,
"learning_rate": 0.00022996990972918758,
"loss": 2.654,
"theoretical_loss": 3.461319074255303,
"tokens_seen": 1797914624
},
{
"epoch": 0.54,
"learning_rate": 0.00022995987963891676,
"loss": 2.6008,
"theoretical_loss": 3.461308333170873,
"tokens_seen": 1797980160
},
{
"epoch": 0.54,
"learning_rate": 0.00022994984954864594,
"loss": 2.6447,
"theoretical_loss": 3.461297592587565,
"tokens_seen": 1798045696
},
{
"epoch": 0.54,
"learning_rate": 0.00022993981945837515,
"loss": 2.6847,
"theoretical_loss": 3.461286852505336,
"tokens_seen": 1798111232
},
{
"epoch": 0.54,
"learning_rate": 0.00022992978936810433,
"loss": 2.6473,
"theoretical_loss": 3.4612761129241463,
"tokens_seen": 1798176768
},
{
"epoch": 0.54,
"learning_rate": 0.0002299197592778335,
"loss": 2.8162,
"theoretical_loss": 3.4612653738439523,
"tokens_seen": 1798242304
},
{
"epoch": 0.54,
"learning_rate": 0.0002299097291875627,
"loss": 2.7938,
"theoretical_loss": 3.4612546352647136,
"tokens_seen": 1798307840
},
{
"epoch": 0.54,
"learning_rate": 0.0002298996990972919,
"loss": 2.8273,
"theoretical_loss": 3.461243897186388,
"tokens_seen": 1798373376
},
{
"epoch": 0.54,
"learning_rate": 0.00022988966900702108,
"loss": 2.6853,
"theoretical_loss": 3.461233159608934,
"tokens_seen": 1798438912
},
{
"epoch": 0.55,
"learning_rate": 0.00022987963891675026,
"loss": 2.8128,
"theoretical_loss": 3.4612224225323103,
"tokens_seen": 1798504448
},
{
"epoch": 0.55,
"learning_rate": 0.00022986960882647944,
"loss": 2.5913,
"theoretical_loss": 3.461211685956475,
"tokens_seen": 1798569984
},
{
"epoch": 0.55,
"learning_rate": 0.00022985957873620862,
"loss": 2.804,
"theoretical_loss": 3.4612009498813867,
"tokens_seen": 1798635520
},
{
"epoch": 0.55,
"learning_rate": 0.00022984954864593783,
"loss": 2.6228,
"theoretical_loss": 3.4611902143070035,
"tokens_seen": 1798701056
},
{
"epoch": 0.55,
"learning_rate": 0.000229839518555667,
"loss": 2.5699,
"theoretical_loss": 3.4611794792332837,
"tokens_seen": 1798766592
},
{
"epoch": 0.55,
"learning_rate": 0.0002298294884653962,
"loss": 2.7163,
"theoretical_loss": 3.4611687446601866,
"tokens_seen": 1798832128
},
{
"epoch": 0.55,
"learning_rate": 0.00022981945837512538,
"loss": 2.7904,
"theoretical_loss": 3.4611580105876696,
"tokens_seen": 1798897664
},
{
"epoch": 0.55,
"objective/train/docs_used": 1018428,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7630767822265625,
"objective/train/theoretical_loss": 3.4611472770156917,
"objective/train/tokens_used": 1819423200,
"theoretical_loss": 3.4611472770156917,
"tokens_seen": 1798963200
},
{
"epoch": 0.55,
"learning_rate": 0.00022980942828485458,
"loss": 2.4388,
"theoretical_loss": 3.4611472770156917,
"tokens_seen": 1798963200
},
{
"epoch": 0.55,
"learning_rate": 0.00022979939819458376,
"loss": 2.6509,
"theoretical_loss": 3.4611365439442103,
"tokens_seen": 1799028736
},
{
"epoch": 0.55,
"learning_rate": 0.00022978936810431295,
"loss": 2.561,
"theoretical_loss": 3.4611258113731855,
"tokens_seen": 1799094272
},
{
"epoch": 0.55,
"learning_rate": 0.00022977933801404213,
"loss": 2.6517,
"theoretical_loss": 3.4611150793025747,
"tokens_seen": 1799159808
},
{
"epoch": 0.55,
"learning_rate": 0.0002297693079237713,
"loss": 2.7837,
"theoretical_loss": 3.461104347732337,
"tokens_seen": 1799225344
},
{
"epoch": 0.55,
"learning_rate": 0.00022975927783350052,
"loss": 2.8172,
"theoretical_loss": 3.4610936166624295,
"tokens_seen": 1799290880
},
{
"epoch": 0.55,
"learning_rate": 0.0002297492477432297,
"loss": 2.5426,
"theoretical_loss": 3.4610828860928122,
"tokens_seen": 1799356416
},
{
"epoch": 0.55,
"learning_rate": 0.00022973921765295888,
"loss": 2.7765,
"theoretical_loss": 3.4610721560234428,
"tokens_seen": 1799421952
},
{
"epoch": 0.55,
"learning_rate": 0.00022972918756268806,
"loss": 2.5277,
"theoretical_loss": 3.46106142645428,
"tokens_seen": 1799487488
},
{
"epoch": 0.55,
"learning_rate": 0.00022971915747241727,
"loss": 2.7364,
"theoretical_loss": 3.4610506973852813,
"tokens_seen": 1799553024
},
{
"epoch": 0.55,
"learning_rate": 0.00022970912738214645,
"loss": 2.6783,
"theoretical_loss": 3.4610399688164066,
"tokens_seen": 1799618560
},
{
"epoch": 0.55,
"learning_rate": 0.00022969909729187563,
"loss": 2.8637,
"theoretical_loss": 3.4610292407476138,
"tokens_seen": 1799684096
},
{
"epoch": 0.55,
"learning_rate": 0.0002296890672016048,
"loss": 2.6224,
"theoretical_loss": 3.4610185131788613,
"tokens_seen": 1799749632
},
{
"epoch": 0.55,
"learning_rate": 0.000229679037111334,
"loss": 2.7068,
"theoretical_loss": 3.461007786110108,
"tokens_seen": 1799815168
},
{
"epoch": 0.55,
"learning_rate": 0.0002296690070210632,
"loss": 2.7285,
"theoretical_loss": 3.4609970595413113,
"tokens_seen": 1799880704
},
{
"epoch": 0.55,
"learning_rate": 0.00022965897693079238,
"loss": 2.6106,
"theoretical_loss": 3.4609863334724307,
"tokens_seen": 1799946240
},
{
"epoch": 0.55,
"learning_rate": 0.00022964894684052156,
"loss": 2.7427,
"theoretical_loss": 3.460975607903425,
"tokens_seen": 1800011776
},
{
"epoch": 0.55,
"learning_rate": 0.00022963891675025074,
"loss": 2.6619,
"theoretical_loss": 3.460964882834251,
"tokens_seen": 1800077312
},
{
"epoch": 0.55,
"learning_rate": 0.00022962888665997995,
"loss": 2.4342,
"theoretical_loss": 3.460954158264869,
"tokens_seen": 1800142848
},
{
"epoch": 0.55,
"learning_rate": 0.00022961885656970913,
"loss": 2.5931,
"theoretical_loss": 3.460943434195237,
"tokens_seen": 1800208384
},
{
"epoch": 0.55,
"learning_rate": 0.00022960882647943831,
"loss": 2.4217,
"theoretical_loss": 3.460932710625313,
"tokens_seen": 1800273920
},
{
"epoch": 0.55,
"learning_rate": 0.0002295987963891675,
"loss": 2.4785,
"theoretical_loss": 3.4609219875550563,
"tokens_seen": 1800339456
},
{
"epoch": 0.55,
"learning_rate": 0.00022958876629889668,
"loss": 2.7098,
"theoretical_loss": 3.460911264984425,
"tokens_seen": 1800404992
},
{
"epoch": 0.55,
"learning_rate": 0.00022957873620862589,
"loss": 2.5124,
"theoretical_loss": 3.460900542913377,
"tokens_seen": 1800470528
},
{
"epoch": 0.55,
"learning_rate": 0.00022956870611835507,
"loss": 2.5245,
"theoretical_loss": 3.460889821341872,
"tokens_seen": 1800536064
},
{
"epoch": 0.55,
"objective/train/docs_used": 1018922,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5269439220428467,
"objective/train/theoretical_loss": 3.4608791002698682,
"objective/train/tokens_used": 1821061600,
"theoretical_loss": 3.4608791002698682,
"tokens_seen": 1800601600
},
{
"epoch": 0.55,
"learning_rate": 0.00022955867602808425,
"loss": 2.4634,
"theoretical_loss": 3.4608791002698682,
"tokens_seen": 1800601600
},
{
"epoch": 0.55,
"learning_rate": 0.00022954864593781343,
"loss": 2.6918,
"theoretical_loss": 3.460868379697324,
"tokens_seen": 1800667136
},
{
"epoch": 0.55,
"learning_rate": 0.00022953861584754264,
"loss": 2.6745,
"theoretical_loss": 3.460857659624198,
"tokens_seen": 1800732672
},
{
"epoch": 0.55,
"learning_rate": 0.00022952858575727182,
"loss": 2.7264,
"theoretical_loss": 3.4608469400504482,
"tokens_seen": 1800798208
},
{
"epoch": 0.55,
"learning_rate": 0.000229518555667001,
"loss": 2.6285,
"theoretical_loss": 3.4608362209760335,
"tokens_seen": 1800863744
},
{
"epoch": 0.55,
"learning_rate": 0.00022950852557673018,
"loss": 2.478,
"theoretical_loss": 3.4608255024009136,
"tokens_seen": 1800929280
},
{
"epoch": 0.55,
"learning_rate": 0.00022949849548645936,
"loss": 2.589,
"theoretical_loss": 3.4608147843250454,
"tokens_seen": 1800994816
},
{
"epoch": 0.55,
"learning_rate": 0.00022948846539618857,
"loss": 2.371,
"theoretical_loss": 3.4608040667483886,
"tokens_seen": 1801060352
},
{
"epoch": 0.55,
"learning_rate": 0.00022947843530591775,
"loss": 2.7455,
"theoretical_loss": 3.460793349670901,
"tokens_seen": 1801125888
},
{
"epoch": 0.55,
"learning_rate": 0.00022946840521564693,
"loss": 2.702,
"theoretical_loss": 3.460782633092542,
"tokens_seen": 1801191424
},
{
"epoch": 0.55,
"learning_rate": 0.0002294583751253761,
"loss": 2.6722,
"theoretical_loss": 3.4607719170132696,
"tokens_seen": 1801256960
},
{
"epoch": 0.55,
"learning_rate": 0.00022944834503510532,
"loss": 2.8343,
"theoretical_loss": 3.4607612014330424,
"tokens_seen": 1801322496
},
{
"epoch": 0.55,
"learning_rate": 0.0002294383149448345,
"loss": 2.5527,
"theoretical_loss": 3.4607504863518193,
"tokens_seen": 1801388032
},
{
"epoch": 0.55,
"learning_rate": 0.00022942828485456368,
"loss": 2.6188,
"theoretical_loss": 3.460739771769559,
"tokens_seen": 1801453568
},
{
"epoch": 0.55,
"learning_rate": 0.00022941825476429286,
"loss": 2.6889,
"theoretical_loss": 3.46072905768622,
"tokens_seen": 1801519104
},
{
"epoch": 0.55,
"learning_rate": 0.00022940822467402207,
"loss": 2.6582,
"theoretical_loss": 3.4607183441017604,
"tokens_seen": 1801584640
},
{
"epoch": 0.55,
"learning_rate": 0.00022939819458375125,
"loss": 2.3978,
"theoretical_loss": 3.460707631016139,
"tokens_seen": 1801650176
},
{
"epoch": 0.55,
"learning_rate": 0.00022938816449348043,
"loss": 2.7923,
"theoretical_loss": 3.4606969184293153,
"tokens_seen": 1801715712
},
{
"epoch": 0.55,
"learning_rate": 0.00022937813440320962,
"loss": 2.7343,
"theoretical_loss": 3.460686206341247,
"tokens_seen": 1801781248
},
{
"epoch": 0.55,
"learning_rate": 0.0002293681043129388,
"loss": 2.6453,
"theoretical_loss": 3.4606754947518934,
"tokens_seen": 1801846784
},
{
"epoch": 0.55,
"learning_rate": 0.000229358074222668,
"loss": 2.7677,
"theoretical_loss": 3.4606647836612128,
"tokens_seen": 1801912320
},
{
"epoch": 0.55,
"learning_rate": 0.0002293480441323972,
"loss": 2.4088,
"theoretical_loss": 3.460654073069163,
"tokens_seen": 1801977856
},
{
"epoch": 0.55,
"learning_rate": 0.0002293380140421264,
"loss": 2.7141,
"theoretical_loss": 3.4606433629757043,
"tokens_seen": 1802043392
},
{
"epoch": 0.55,
"learning_rate": 0.00022932798395185558,
"loss": 2.6025,
"theoretical_loss": 3.4606326533807943,
"tokens_seen": 1802108928
},
{
"epoch": 0.55,
"learning_rate": 0.00022931795386158478,
"loss": 2.6782,
"theoretical_loss": 3.4606219442843917,
"tokens_seen": 1802174464
},
{
"debugging/Self-BLEU-5": 0.4499445909709529,
"debugging/distinct-1-grams": 0.7807765957502695,
"debugging/distinct-2-grams": 0.9631601433789202,
"debugging/entropy-1-grams": 5.814570071853726,
"debugging/entropy-2-grams": 6.5584239085048655,
"debugging/length": 542.0,
"debugging/num_segments": 9,
"epoch": 0.55,
"objective/train/docs_used": 1019481,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.785318374633789,
"objective/train/theoretical_loss": 3.460611235686456,
"objective/train/tokens_used": 1822700000,
"theoretical_loss": 3.460611235686456,
"tokens_seen": 1802240000
},
{
"epoch": 0.55,
"learning_rate": 0.00022930792377131397,
"loss": 2.7932,
"theoretical_loss": 3.460611235686456,
"tokens_seen": 1802240000
},
{
"epoch": 0.55,
"learning_rate": 0.00022929789368104315,
"loss": 2.6964,
"theoretical_loss": 3.4606005275869447,
"tokens_seen": 1802305536
},
{
"epoch": 0.55,
"learning_rate": 0.00022928786359077233,
"loss": 2.6545,
"theoretical_loss": 3.460589819985817,
"tokens_seen": 1802371072
},
{
"epoch": 0.55,
"learning_rate": 0.0002292778335005015,
"loss": 2.5248,
"theoretical_loss": 3.460579112883032,
"tokens_seen": 1802436608
},
{
"epoch": 0.55,
"learning_rate": 0.00022926780341023072,
"loss": 2.7761,
"theoretical_loss": 3.460568406278548,
"tokens_seen": 1802502144
},
{
"epoch": 0.55,
"learning_rate": 0.0002292577733199599,
"loss": 2.4968,
"theoretical_loss": 3.4605577001723233,
"tokens_seen": 1802567680
},
{
"epoch": 0.55,
"learning_rate": 0.00022924774322968908,
"loss": 2.5541,
"theoretical_loss": 3.460546994564317,
"tokens_seen": 1802633216
},
{
"epoch": 0.55,
"learning_rate": 0.00022923771313941826,
"loss": 2.6626,
"theoretical_loss": 3.460536289454488,
"tokens_seen": 1802698752
},
{
"epoch": 0.55,
"learning_rate": 0.00022922768304914747,
"loss": 2.6556,
"theoretical_loss": 3.460525584842795,
"tokens_seen": 1802764288
},
{
"epoch": 0.55,
"learning_rate": 0.00022921765295887665,
"loss": 2.6656,
"theoretical_loss": 3.4605148807291966,
"tokens_seen": 1802829824
},
{
"epoch": 0.55,
"learning_rate": 0.00022920762286860583,
"loss": 2.6358,
"theoretical_loss": 3.460504177113651,
"tokens_seen": 1802895360
},
{
"epoch": 0.55,
"learning_rate": 0.000229197592778335,
"loss": 2.5158,
"theoretical_loss": 3.4604934739961175,
"tokens_seen": 1802960896
},
{
"epoch": 0.55,
"learning_rate": 0.0002291875626880642,
"loss": 2.6966,
"theoretical_loss": 3.460482771376555,
"tokens_seen": 1803026432
},
{
"epoch": 0.55,
"learning_rate": 0.0002291775325977934,
"loss": 3.0897,
"theoretical_loss": 3.460472069254922,
"tokens_seen": 1803091968
},
{
"epoch": 0.55,
"learning_rate": 0.00022916750250752258,
"loss": 2.6613,
"theoretical_loss": 3.4604613676311766,
"tokens_seen": 1803157504
},
{
"epoch": 0.55,
"learning_rate": 0.00022915747241725176,
"loss": 2.5836,
"theoretical_loss": 3.4604506665052783,
"tokens_seen": 1803223040
},
{
"epoch": 0.55,
"learning_rate": 0.00022914744232698094,
"loss": 2.5103,
"theoretical_loss": 3.4604399658771863,
"tokens_seen": 1803288576
},
{
"epoch": 0.55,
"learning_rate": 0.00022913741223671015,
"loss": 2.481,
"theoretical_loss": 3.460429265746858,
"tokens_seen": 1803354112
},
{
"epoch": 0.55,
"learning_rate": 0.00022912738214643933,
"loss": 2.6638,
"theoretical_loss": 3.460418566114253,
"tokens_seen": 1803419648
},
{
"epoch": 0.55,
"learning_rate": 0.00022911735205616851,
"loss": 2.5036,
"theoretical_loss": 3.4604078669793297,
"tokens_seen": 1803485184
},
{
"epoch": 0.55,
"learning_rate": 0.0002291073219658977,
"loss": 2.5183,
"theoretical_loss": 3.4603971683420474,
"tokens_seen": 1803550720
},
{
"epoch": 0.55,
"learning_rate": 0.00022909729187562688,
"loss": 2.8326,
"theoretical_loss": 3.4603864702023643,
"tokens_seen": 1803616256
},
{
"epoch": 0.55,
"learning_rate": 0.00022908726178535609,
"loss": 3.0168,
"theoretical_loss": 3.4603757725602398,
"tokens_seen": 1803681792
},
{
"epoch": 0.55,
"learning_rate": 0.00022907723169508527,
"loss": 2.5847,
"theoretical_loss": 3.4603650754156323,
"tokens_seen": 1803747328
},
{
"epoch": 0.55,
"learning_rate": 0.00022906720160481445,
"loss": 2.6566,
"theoretical_loss": 3.4603543787685,
"tokens_seen": 1803812864
},
{
"epoch": 0.55,
"objective/train/docs_used": 1020148,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.270443916320801,
"objective/train/theoretical_loss": 3.460343682618803,
"objective/train/tokens_used": 1824338400,
"theoretical_loss": 3.460343682618803,
"tokens_seen": 1803878400
},
{
"epoch": 0.55,
"learning_rate": 0.00022905717151454363,
"loss": 2.627,
"theoretical_loss": 3.460343682618803,
"tokens_seen": 1803878400
},
{
"epoch": 0.55,
"learning_rate": 0.00022904714142427284,
"loss": 2.5597,
"theoretical_loss": 3.460332986966499,
"tokens_seen": 1803943936
},
{
"epoch": 0.55,
"learning_rate": 0.00022903711133400202,
"loss": 2.6644,
"theoretical_loss": 3.4603222918115475,
"tokens_seen": 1804009472
},
{
"epoch": 0.55,
"learning_rate": 0.0002290270812437312,
"loss": 2.7348,
"theoretical_loss": 3.4603115971539067,
"tokens_seen": 1804075008
},
{
"epoch": 0.55,
"learning_rate": 0.00022901705115346038,
"loss": 2.5711,
"theoretical_loss": 3.460300902993536,
"tokens_seen": 1804140544
},
{
"epoch": 0.55,
"learning_rate": 0.00022900702106318956,
"loss": 2.5405,
"theoretical_loss": 3.4602902093303936,
"tokens_seen": 1804206080
},
{
"epoch": 0.55,
"learning_rate": 0.00022899699097291877,
"loss": 2.6522,
"theoretical_loss": 3.460279516164439,
"tokens_seen": 1804271616
},
{
"epoch": 0.55,
"learning_rate": 0.00022898696088264795,
"loss": 2.7227,
"theoretical_loss": 3.460268823495631,
"tokens_seen": 1804337152
},
{
"epoch": 0.55,
"learning_rate": 0.00022897693079237713,
"loss": 2.7128,
"theoretical_loss": 3.4602581313239273,
"tokens_seen": 1804402688
},
{
"epoch": 0.55,
"learning_rate": 0.0002289669007021063,
"loss": 2.6228,
"theoretical_loss": 3.460247439649288,
"tokens_seen": 1804468224
},
{
"epoch": 0.55,
"learning_rate": 0.00022895687061183552,
"loss": 2.7765,
"theoretical_loss": 3.4602367484716714,
"tokens_seen": 1804533760
},
{
"epoch": 0.55,
"learning_rate": 0.0002289468405215647,
"loss": 2.5737,
"theoretical_loss": 3.4602260577910364,
"tokens_seen": 1804599296
},
{
"epoch": 0.55,
"learning_rate": 0.00022893681043129388,
"loss": 2.5836,
"theoretical_loss": 3.4602153676073417,
"tokens_seen": 1804664832
},
{
"epoch": 0.55,
"learning_rate": 0.00022892678034102306,
"loss": 2.7331,
"theoretical_loss": 3.4602046779205464,
"tokens_seen": 1804730368
},
{
"epoch": 0.55,
"learning_rate": 0.00022891675025075227,
"loss": 2.6258,
"theoretical_loss": 3.4601939887306097,
"tokens_seen": 1804795904
},
{
"epoch": 0.55,
"learning_rate": 0.00022890672016048145,
"loss": 2.5309,
"theoretical_loss": 3.4601833000374898,
"tokens_seen": 1804861440
},
{
"epoch": 0.55,
"learning_rate": 0.00022889669007021063,
"loss": 2.6306,
"theoretical_loss": 3.4601726118411458,
"tokens_seen": 1804926976
},
{
"epoch": 0.55,
"learning_rate": 0.00022888665997993982,
"loss": 2.675,
"theoretical_loss": 3.460161924141537,
"tokens_seen": 1804992512
},
{
"epoch": 0.55,
"learning_rate": 0.000228876629889669,
"loss": 2.7637,
"theoretical_loss": 3.4601512369386214,
"tokens_seen": 1805058048
},
{
"epoch": 0.55,
"learning_rate": 0.0002288665997993982,
"loss": 2.9243,
"theoretical_loss": 3.4601405502323583,
"tokens_seen": 1805123584
},
{
"epoch": 0.55,
"learning_rate": 0.0002288565697091274,
"loss": 2.8187,
"theoretical_loss": 3.460129864022707,
"tokens_seen": 1805189120
},
{
"epoch": 0.55,
"learning_rate": 0.00022884653961885657,
"loss": 2.5989,
"theoretical_loss": 3.460119178309626,
"tokens_seen": 1805254656
},
{
"epoch": 0.55,
"learning_rate": 0.00022883650952858575,
"loss": 2.5765,
"theoretical_loss": 3.4601084930930743,
"tokens_seen": 1805320192
},
{
"epoch": 0.55,
"learning_rate": 0.00022882647943831496,
"loss": 2.7417,
"theoretical_loss": 3.4600978083730105,
"tokens_seen": 1805385728
},
{
"epoch": 0.55,
"learning_rate": 0.00022881644934804414,
"loss": 2.634,
"theoretical_loss": 3.460087124149394,
"tokens_seen": 1805451264
},
{
"epoch": 0.55,
"objective/train/docs_used": 1020822,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.752251148223877,
"objective/train/theoretical_loss": 3.4600764404221835,
"objective/train/tokens_used": 1825976800,
"theoretical_loss": 3.4600764404221835,
"tokens_seen": 1805516800
},
{
"epoch": 0.55,
"learning_rate": 0.00022880641925777332,
"loss": 2.7534,
"theoretical_loss": 3.4600764404221835,
"tokens_seen": 1805516800
},
{
"epoch": 0.55,
"learning_rate": 0.0002287963891675025,
"loss": 2.8177,
"theoretical_loss": 3.460065757191338,
"tokens_seen": 1805582336
},
{
"epoch": 0.55,
"learning_rate": 0.00022878635907723168,
"loss": 2.5871,
"theoretical_loss": 3.460055074456816,
"tokens_seen": 1805647872
},
{
"epoch": 0.55,
"learning_rate": 0.0002287763289869609,
"loss": 2.6218,
"theoretical_loss": 3.4600443922185766,
"tokens_seen": 1805713408
},
{
"epoch": 0.55,
"learning_rate": 0.00022876629889669007,
"loss": 2.5204,
"theoretical_loss": 3.4600337104765795,
"tokens_seen": 1805778944
},
{
"epoch": 0.55,
"learning_rate": 0.00022875626880641925,
"loss": 2.6144,
"theoretical_loss": 3.4600230292307828,
"tokens_seen": 1805844480
},
{
"epoch": 0.55,
"learning_rate": 0.00022874623871614843,
"loss": 2.9008,
"theoretical_loss": 3.4600123484811456,
"tokens_seen": 1805910016
},
{
"epoch": 0.55,
"learning_rate": 0.00022873620862587764,
"loss": 2.2547,
"theoretical_loss": 3.460001668227627,
"tokens_seen": 1805975552
},
{
"epoch": 0.55,
"learning_rate": 0.00022872617853560682,
"loss": 2.8188,
"theoretical_loss": 3.459990988470186,
"tokens_seen": 1806041088
},
{
"epoch": 0.55,
"learning_rate": 0.000228716148445336,
"loss": 2.5824,
"theoretical_loss": 3.459980309208781,
"tokens_seen": 1806106624
},
{
"epoch": 0.55,
"learning_rate": 0.00022870611835506518,
"loss": 2.704,
"theoretical_loss": 3.459969630443372,
"tokens_seen": 1806172160
},
{
"epoch": 0.55,
"learning_rate": 0.00022869608826479437,
"loss": 2.7493,
"theoretical_loss": 3.459958952173917,
"tokens_seen": 1806237696
},
{
"epoch": 0.55,
"learning_rate": 0.00022868605817452357,
"loss": 2.4929,
"theoretical_loss": 3.459948274400375,
"tokens_seen": 1806303232
},
{
"epoch": 0.55,
"learning_rate": 0.00022867602808425276,
"loss": 2.5881,
"theoretical_loss": 3.459937597122706,
"tokens_seen": 1806368768
},
{
"epoch": 0.55,
"learning_rate": 0.00022866599799398194,
"loss": 2.5689,
"theoretical_loss": 3.459926920340868,
"tokens_seen": 1806434304
},
{
"epoch": 0.55,
"learning_rate": 0.00022865596790371112,
"loss": 2.4208,
"theoretical_loss": 3.4599162440548197,
"tokens_seen": 1806499840
},
{
"epoch": 0.55,
"learning_rate": 0.00022864593781344033,
"loss": 2.689,
"theoretical_loss": 3.4599055682645217,
"tokens_seen": 1806565376
},
{
"epoch": 0.55,
"learning_rate": 0.0002286359077231695,
"loss": 2.6097,
"theoretical_loss": 3.459894892969931,
"tokens_seen": 1806630912
},
{
"epoch": 0.55,
"learning_rate": 0.0002286258776328987,
"loss": 2.7033,
"theoretical_loss": 3.459884218171008,
"tokens_seen": 1806696448
},
{
"epoch": 0.55,
"learning_rate": 0.00022861584754262787,
"loss": 2.689,
"theoretical_loss": 3.4598735438677117,
"tokens_seen": 1806761984
},
{
"epoch": 0.55,
"learning_rate": 0.00022860581745235705,
"loss": 2.6121,
"theoretical_loss": 3.4598628700600003,
"tokens_seen": 1806827520
},
{
"epoch": 0.55,
"learning_rate": 0.00022859578736208629,
"loss": 2.7091,
"theoretical_loss": 3.4598521967478337,
"tokens_seen": 1806893056
},
{
"epoch": 0.55,
"learning_rate": 0.00022858575727181547,
"loss": 2.6472,
"theoretical_loss": 3.4598415239311695,
"tokens_seen": 1806958592
},
{
"epoch": 0.55,
"learning_rate": 0.00022857572718154465,
"loss": 2.7967,
"theoretical_loss": 3.4598308516099685,
"tokens_seen": 1807024128
},
{
"epoch": 0.55,
"learning_rate": 0.00022856569709127383,
"loss": 2.6711,
"theoretical_loss": 3.4598201797841885,
"tokens_seen": 1807089664
},
{
"epoch": 0.55,
"objective/train/docs_used": 1022105,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.294231653213501,
"objective/train/theoretical_loss": 3.459809508453789,
"objective/train/tokens_used": 1827615200,
"theoretical_loss": 3.459809508453789,
"tokens_seen": 1807155200
},
{
"epoch": 0.55,
"learning_rate": 0.00022855566700100304,
"loss": 2.5223,
"theoretical_loss": 3.459809508453789,
"tokens_seen": 1807155200
},
{
"epoch": 0.55,
"learning_rate": 0.00022854563691073222,
"loss": 2.7111,
"theoretical_loss": 3.459798837618729,
"tokens_seen": 1807220736
},
{
"epoch": 0.55,
"learning_rate": 0.0002285356068204614,
"loss": 2.6248,
"theoretical_loss": 3.4597881672789677,
"tokens_seen": 1807286272
},
{
"epoch": 0.55,
"learning_rate": 0.00022852557673019058,
"loss": 2.7781,
"theoretical_loss": 3.4597774974344637,
"tokens_seen": 1807351808
},
{
"epoch": 0.55,
"learning_rate": 0.00022851554663991976,
"loss": 2.5567,
"theoretical_loss": 3.4597668280851765,
"tokens_seen": 1807417344
},
{
"epoch": 0.55,
"learning_rate": 0.00022850551654964897,
"loss": 2.5506,
"theoretical_loss": 3.4597561592310644,
"tokens_seen": 1807482880
},
{
"epoch": 0.55,
"learning_rate": 0.00022849548645937815,
"loss": 2.7236,
"theoretical_loss": 3.4597454908720877,
"tokens_seen": 1807548416
},
{
"epoch": 0.55,
"learning_rate": 0.00022848545636910733,
"loss": 2.5832,
"theoretical_loss": 3.459734823008205,
"tokens_seen": 1807613952
},
{
"epoch": 0.55,
"learning_rate": 0.0002284754262788365,
"loss": 2.7102,
"theoretical_loss": 3.4597241556393747,
"tokens_seen": 1807679488
},
{
"epoch": 0.55,
"learning_rate": 0.00022846539618856572,
"loss": 2.6845,
"theoretical_loss": 3.4597134887655567,
"tokens_seen": 1807745024
},
{
"epoch": 0.55,
"learning_rate": 0.0002284553660982949,
"loss": 2.7831,
"theoretical_loss": 3.459702822386709,
"tokens_seen": 1807810560
},
{
"epoch": 0.55,
"learning_rate": 0.00022844533600802408,
"loss": 2.5295,
"theoretical_loss": 3.4596921565027925,
"tokens_seen": 1807876096
},
{
"epoch": 0.55,
"learning_rate": 0.00022843530591775326,
"loss": 2.4676,
"theoretical_loss": 3.4596814911137646,
"tokens_seen": 1807941632
},
{
"epoch": 0.55,
"learning_rate": 0.00022842527582748247,
"loss": 2.6568,
"theoretical_loss": 3.459670826219585,
"tokens_seen": 1808007168
},
{
"epoch": 0.55,
"learning_rate": 0.00022841524573721165,
"loss": 2.8829,
"theoretical_loss": 3.459660161820213,
"tokens_seen": 1808072704
},
{
"epoch": 0.55,
"learning_rate": 0.00022840521564694084,
"loss": 2.7387,
"theoretical_loss": 3.459649497915608,
"tokens_seen": 1808138240
},
{
"epoch": 0.55,
"learning_rate": 0.00022839518555667002,
"loss": 2.77,
"theoretical_loss": 3.459638834505728,
"tokens_seen": 1808203776
},
{
"epoch": 0.55,
"learning_rate": 0.0002283851554663992,
"loss": 2.6029,
"theoretical_loss": 3.4596281715905333,
"tokens_seen": 1808269312
},
{
"epoch": 0.55,
"learning_rate": 0.0002283751253761284,
"loss": 2.7583,
"theoretical_loss": 3.4596175091699823,
"tokens_seen": 1808334848
},
{
"epoch": 0.55,
"learning_rate": 0.0002283650952858576,
"loss": 2.7685,
"theoretical_loss": 3.459606847244034,
"tokens_seen": 1808400384
},
{
"epoch": 0.55,
"learning_rate": 0.00022835506519558677,
"loss": 2.6894,
"theoretical_loss": 3.4595961858126483,
"tokens_seen": 1808465920
},
{
"epoch": 0.55,
"learning_rate": 0.00022834503510531595,
"loss": 2.7306,
"theoretical_loss": 3.459585524875784,
"tokens_seen": 1808531456
},
{
"epoch": 0.55,
"learning_rate": 0.00022833500501504516,
"loss": 2.8394,
"theoretical_loss": 3.4595748644334,
"tokens_seen": 1808596992
},
{
"epoch": 0.55,
"learning_rate": 0.00022832497492477434,
"loss": 2.8299,
"theoretical_loss": 3.4595642044854555,
"tokens_seen": 1808662528
},
{
"epoch": 0.55,
"learning_rate": 0.00022831494483450352,
"loss": 2.7707,
"theoretical_loss": 3.45955354503191,
"tokens_seen": 1808728064
},
{
"epoch": 0.55,
"objective/train/docs_used": 1022607,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.102808952331543,
"objective/train/theoretical_loss": 3.4595428860727218,
"objective/train/tokens_used": 1829253600,
"theoretical_loss": 3.4595428860727218,
"tokens_seen": 1808793600
},
{
"epoch": 0.55,
"learning_rate": 0.0002283049147442327,
"loss": 2.5414,
"theoretical_loss": 3.4595428860727218,
"tokens_seen": 1808793600
},
{
"epoch": 0.55,
"learning_rate": 0.00022829488465396188,
"loss": 2.6881,
"theoretical_loss": 3.4595322276078515,
"tokens_seen": 1808859136
},
{
"epoch": 0.55,
"learning_rate": 0.0002282848545636911,
"loss": 2.8559,
"theoretical_loss": 3.459521569637257,
"tokens_seen": 1808924672
},
{
"epoch": 0.55,
"learning_rate": 0.00022827482447342027,
"loss": 2.4626,
"theoretical_loss": 3.459510912160898,
"tokens_seen": 1808990208
},
{
"epoch": 0.55,
"learning_rate": 0.00022826479438314945,
"loss": 2.5678,
"theoretical_loss": 3.4595002551787335,
"tokens_seen": 1809055744
},
{
"epoch": 0.55,
"learning_rate": 0.00022825476429287863,
"loss": 2.7529,
"theoretical_loss": 3.4594895986907233,
"tokens_seen": 1809121280
},
{
"epoch": 0.55,
"learning_rate": 0.00022824473420260784,
"loss": 2.595,
"theoretical_loss": 3.4594789426968253,
"tokens_seen": 1809186816
},
{
"epoch": 0.55,
"learning_rate": 0.00022823470411233702,
"loss": 2.6924,
"theoretical_loss": 3.4594682871969997,
"tokens_seen": 1809252352
},
{
"epoch": 0.55,
"learning_rate": 0.0002282246740220662,
"loss": 2.5746,
"theoretical_loss": 3.4594576321912056,
"tokens_seen": 1809317888
},
{
"epoch": 0.55,
"learning_rate": 0.00022821464393179538,
"loss": 2.616,
"theoretical_loss": 3.459446977679402,
"tokens_seen": 1809383424
},
{
"epoch": 0.55,
"learning_rate": 0.00022820461384152457,
"loss": 2.6787,
"theoretical_loss": 3.4594363236615484,
"tokens_seen": 1809448960
},
{
"epoch": 0.55,
"learning_rate": 0.00022819458375125377,
"loss": 2.2807,
"theoretical_loss": 3.4594256701376036,
"tokens_seen": 1809514496
},
{
"epoch": 0.55,
"learning_rate": 0.00022818455366098296,
"loss": 2.6373,
"theoretical_loss": 3.459415017107527,
"tokens_seen": 1809580032
},
{
"epoch": 0.55,
"learning_rate": 0.00022817452357071214,
"loss": 2.7128,
"theoretical_loss": 3.4594043645712773,
"tokens_seen": 1809645568
},
{
"epoch": 0.55,
"learning_rate": 0.00022816449348044132,
"loss": 2.6449,
"theoretical_loss": 3.459393712528815,
"tokens_seen": 1809711104
},
{
"epoch": 0.55,
"learning_rate": 0.00022815446339017053,
"loss": 2.5317,
"theoretical_loss": 3.459383060980098,
"tokens_seen": 1809776640
},
{
"epoch": 0.55,
"learning_rate": 0.0002281444332998997,
"loss": 2.6862,
"theoretical_loss": 3.459372409925087,
"tokens_seen": 1809842176
},
{
"epoch": 0.55,
"learning_rate": 0.0002281344032096289,
"loss": 2.4135,
"theoretical_loss": 3.4593617593637394,
"tokens_seen": 1809907712
},
{
"epoch": 0.55,
"learning_rate": 0.00022812437311935807,
"loss": 2.6838,
"theoretical_loss": 3.4593511092960156,
"tokens_seen": 1809973248
},
{
"epoch": 0.55,
"learning_rate": 0.00022811434302908725,
"loss": 2.5702,
"theoretical_loss": 3.459340459721875,
"tokens_seen": 1810038784
},
{
"epoch": 0.55,
"learning_rate": 0.00022810431293881646,
"loss": 2.3772,
"theoretical_loss": 3.459329810641276,
"tokens_seen": 1810104320
},
{
"epoch": 0.55,
"learning_rate": 0.00022809428284854564,
"loss": 2.9218,
"theoretical_loss": 3.4593191620541783,
"tokens_seen": 1810169856
},
{
"epoch": 0.55,
"learning_rate": 0.00022808425275827482,
"loss": 2.6264,
"theoretical_loss": 3.4593085139605417,
"tokens_seen": 1810235392
},
{
"epoch": 0.55,
"learning_rate": 0.000228074222668004,
"loss": 2.5581,
"theoretical_loss": 3.459297866360325,
"tokens_seen": 1810300928
},
{
"epoch": 0.55,
"learning_rate": 0.0002280641925777332,
"loss": 2.7793,
"theoretical_loss": 3.459287219253487,
"tokens_seen": 1810366464
},
{
"epoch": 0.55,
"objective/train/docs_used": 1024104,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2906386852264404,
"objective/train/theoretical_loss": 3.459276572639988,
"objective/train/tokens_used": 1830892000,
"theoretical_loss": 3.459276572639988,
"tokens_seen": 1810432000
},
{
"epoch": 0.55,
"learning_rate": 0.0002280541624874624,
"loss": 2.5711,
"theoretical_loss": 3.459276572639988,
"tokens_seen": 1810432000
},
{
"epoch": 0.55,
"learning_rate": 0.00022804413239719157,
"loss": 2.619,
"theoretical_loss": 3.459265926519786,
"tokens_seen": 1810497536
},
{
"epoch": 0.55,
"learning_rate": 0.00022803410230692075,
"loss": 2.7205,
"theoretical_loss": 3.459255280892841,
"tokens_seen": 1810563072
},
{
"epoch": 0.55,
"learning_rate": 0.00022802407221664993,
"loss": 2.7462,
"theoretical_loss": 3.459244635759113,
"tokens_seen": 1810628608
},
{
"epoch": 0.55,
"learning_rate": 0.00022801404212637914,
"loss": 2.7817,
"theoretical_loss": 3.459233991118561,
"tokens_seen": 1810694144
},
{
"epoch": 0.55,
"learning_rate": 0.00022800401203610832,
"loss": 2.5177,
"theoretical_loss": 3.4592233469711426,
"tokens_seen": 1810759680
},
{
"epoch": 0.55,
"learning_rate": 0.0002279939819458375,
"loss": 2.6514,
"theoretical_loss": 3.459212703316819,
"tokens_seen": 1810825216
},
{
"epoch": 0.55,
"learning_rate": 0.00022798395185556669,
"loss": 2.4907,
"theoretical_loss": 3.4592020601555493,
"tokens_seen": 1810890752
},
{
"epoch": 0.55,
"learning_rate": 0.0002279739217652959,
"loss": 2.6703,
"theoretical_loss": 3.4591914174872915,
"tokens_seen": 1810956288
},
{
"epoch": 0.55,
"learning_rate": 0.00022796389167502508,
"loss": 2.7923,
"theoretical_loss": 3.4591807753120065,
"tokens_seen": 1811021824
},
{
"epoch": 0.55,
"learning_rate": 0.00022795386158475426,
"loss": 2.5196,
"theoretical_loss": 3.4591701336296534,
"tokens_seen": 1811087360
},
{
"epoch": 0.55,
"learning_rate": 0.00022794383149448344,
"loss": 2.6244,
"theoretical_loss": 3.4591594924401905,
"tokens_seen": 1811152896
},
{
"epoch": 0.55,
"learning_rate": 0.00022793380140421262,
"loss": 2.7121,
"theoretical_loss": 3.459148851743578,
"tokens_seen": 1811218432
},
{
"epoch": 0.55,
"learning_rate": 0.00022792377131394183,
"loss": 2.6193,
"theoretical_loss": 3.459138211539775,
"tokens_seen": 1811283968
},
{
"epoch": 0.55,
"learning_rate": 0.000227913741223671,
"loss": 2.7271,
"theoretical_loss": 3.4591275718287404,
"tokens_seen": 1811349504
},
{
"epoch": 0.55,
"learning_rate": 0.0002279037111334002,
"loss": 2.7389,
"theoretical_loss": 3.459116932610435,
"tokens_seen": 1811415040
},
{
"epoch": 0.55,
"learning_rate": 0.00022789368104312937,
"loss": 2.7876,
"theoretical_loss": 3.4591062938848163,
"tokens_seen": 1811480576
},
{
"epoch": 0.55,
"learning_rate": 0.00022788365095285858,
"loss": 2.5543,
"theoretical_loss": 3.4590956556518444,
"tokens_seen": 1811546112
},
{
"epoch": 0.55,
"learning_rate": 0.00022787362086258776,
"loss": 2.7286,
"theoretical_loss": 3.4590850179114794,
"tokens_seen": 1811611648
},
{
"epoch": 0.55,
"learning_rate": 0.00022786359077231694,
"loss": 2.74,
"theoretical_loss": 3.45907438066368,
"tokens_seen": 1811677184
},
{
"epoch": 0.55,
"learning_rate": 0.00022785356068204612,
"loss": 2.7666,
"theoretical_loss": 3.4590637439084055,
"tokens_seen": 1811742720
},
{
"epoch": 0.55,
"learning_rate": 0.00022784353059177536,
"loss": 2.8834,
"theoretical_loss": 3.459053107645615,
"tokens_seen": 1811808256
},
{
"epoch": 0.55,
"learning_rate": 0.00022783350050150454,
"loss": 2.5375,
"theoretical_loss": 3.459042471875269,
"tokens_seen": 1811873792
},
{
"epoch": 0.55,
"learning_rate": 0.00022782347041123372,
"loss": 2.6762,
"theoretical_loss": 3.4590318365973256,
"tokens_seen": 1811939328
},
{
"epoch": 0.55,
"learning_rate": 0.0002278134403209629,
"loss": 2.651,
"theoretical_loss": 3.459021201811745,
"tokens_seen": 1812004864
},
{
"epoch": 0.55,
"objective/train/docs_used": 1024786,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3896310329437256,
"objective/train/theoretical_loss": 3.4590105675184866,
"objective/train/tokens_used": 1832530400,
"theoretical_loss": 3.4590105675184866,
"tokens_seen": 1812070400
},
{
"epoch": 0.55,
"learning_rate": 0.00022780341023069208,
"loss": 2.6837,
"theoretical_loss": 3.4590105675184866,
"tokens_seen": 1812070400
},
{
"epoch": 0.55,
"learning_rate": 0.0002277933801404213,
"loss": 2.563,
"theoretical_loss": 3.458999933717509,
"tokens_seen": 1812135936
},
{
"epoch": 0.55,
"learning_rate": 0.00022778335005015047,
"loss": 2.7174,
"theoretical_loss": 3.4589893004087733,
"tokens_seen": 1812201472
},
{
"epoch": 0.55,
"learning_rate": 0.00022777331995987965,
"loss": 2.8446,
"theoretical_loss": 3.4589786675922367,
"tokens_seen": 1812267008
},
{
"epoch": 0.55,
"learning_rate": 0.00022776328986960883,
"loss": 2.7084,
"theoretical_loss": 3.4589680352678602,
"tokens_seen": 1812332544
},
{
"epoch": 0.55,
"learning_rate": 0.00022775325977933804,
"loss": 2.7773,
"theoretical_loss": 3.4589574034356025,
"tokens_seen": 1812398080
},
{
"epoch": 0.55,
"learning_rate": 0.00022774322968906722,
"loss": 2.6026,
"theoretical_loss": 3.4589467720954232,
"tokens_seen": 1812463616
},
{
"epoch": 0.55,
"learning_rate": 0.0002277331995987964,
"loss": 2.7163,
"theoretical_loss": 3.4589361412472828,
"tokens_seen": 1812529152
},
{
"epoch": 0.55,
"learning_rate": 0.00022772316950852558,
"loss": 2.3979,
"theoretical_loss": 3.4589255108911385,
"tokens_seen": 1812594688
},
{
"epoch": 0.55,
"learning_rate": 0.00022771313941825477,
"loss": 2.7601,
"theoretical_loss": 3.458914881026952,
"tokens_seen": 1812660224
},
{
"epoch": 0.55,
"learning_rate": 0.00022770310932798397,
"loss": 2.4381,
"theoretical_loss": 3.458904251654681,
"tokens_seen": 1812725760
},
{
"epoch": 0.55,
"learning_rate": 0.00022769307923771316,
"loss": 2.3462,
"theoretical_loss": 3.458893622774286,
"tokens_seen": 1812791296
},
{
"epoch": 0.55,
"learning_rate": 0.00022768304914744234,
"loss": 2.6592,
"theoretical_loss": 3.4588829943857258,
"tokens_seen": 1812856832
},
{
"epoch": 0.55,
"learning_rate": 0.00022767301905717152,
"loss": 2.6505,
"theoretical_loss": 3.4588723664889605,
"tokens_seen": 1812922368
},
{
"epoch": 0.55,
"learning_rate": 0.00022766298896690073,
"loss": 2.7688,
"theoretical_loss": 3.4588617390839493,
"tokens_seen": 1812987904
},
{
"epoch": 0.55,
"learning_rate": 0.0002276529588766299,
"loss": 2.6027,
"theoretical_loss": 3.4588511121706516,
"tokens_seen": 1813053440
},
{
"epoch": 0.55,
"learning_rate": 0.0002276429287863591,
"loss": 2.7167,
"theoretical_loss": 3.458840485749027,
"tokens_seen": 1813118976
},
{
"epoch": 0.55,
"learning_rate": 0.00022763289869608827,
"loss": 2.7222,
"theoretical_loss": 3.4588298598190352,
"tokens_seen": 1813184512
},
{
"epoch": 0.55,
"learning_rate": 0.00022762286860581745,
"loss": 2.9139,
"theoretical_loss": 3.458819234380635,
"tokens_seen": 1813250048
},
{
"epoch": 0.55,
"learning_rate": 0.00022761283851554666,
"loss": 2.5561,
"theoretical_loss": 3.4588086094337864,
"tokens_seen": 1813315584
},
{
"epoch": 0.55,
"learning_rate": 0.00022760280842527584,
"loss": 2.4775,
"theoretical_loss": 3.4587979849784487,
"tokens_seen": 1813381120
},
{
"epoch": 0.55,
"learning_rate": 0.00022759277833500502,
"loss": 2.5749,
"theoretical_loss": 3.458787361014582,
"tokens_seen": 1813446656
},
{
"epoch": 0.55,
"learning_rate": 0.0002275827482447342,
"loss": 2.7358,
"theoretical_loss": 3.4587767375421445,
"tokens_seen": 1813512192
},
{
"epoch": 0.55,
"learning_rate": 0.0002275727181544634,
"loss": 2.7181,
"theoretical_loss": 3.458766114561097,
"tokens_seen": 1813577728
},
{
"epoch": 0.55,
"learning_rate": 0.0002275626880641926,
"loss": 2.7556,
"theoretical_loss": 3.458755492071398,
"tokens_seen": 1813643264
},
{
"epoch": 0.55,
"objective/train/docs_used": 1026014,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.634758710861206,
"objective/train/theoretical_loss": 3.458744870073008,
"objective/train/tokens_used": 1834168800,
"theoretical_loss": 3.458744870073008,
"tokens_seen": 1813708800
},
{
"epoch": 0.55,
"learning_rate": 0.00022755265797392177,
"loss": 2.6264,
"theoretical_loss": 3.458744870073008,
"tokens_seen": 1813708800
},
{
"epoch": 0.55,
"learning_rate": 0.00022754262788365095,
"loss": 2.7619,
"theoretical_loss": 3.458734248565886,
"tokens_seen": 1813774336
},
{
"epoch": 0.55,
"learning_rate": 0.00022753259779338013,
"loss": 2.6479,
"theoretical_loss": 3.4587236275499915,
"tokens_seen": 1813839872
},
{
"epoch": 0.55,
"learning_rate": 0.00022752256770310934,
"loss": 2.5357,
"theoretical_loss": 3.458713007025284,
"tokens_seen": 1813905408
},
{
"epoch": 0.55,
"learning_rate": 0.00022751253761283852,
"loss": 2.8061,
"theoretical_loss": 3.4587023869917237,
"tokens_seen": 1813970944
},
{
"epoch": 0.55,
"learning_rate": 0.0002275025075225677,
"loss": 2.7945,
"theoretical_loss": 3.458691767449269,
"tokens_seen": 1814036480
},
{
"epoch": 0.55,
"learning_rate": 0.00022749247743229689,
"loss": 2.7428,
"theoretical_loss": 3.45868114839788,
"tokens_seen": 1814102016
},
{
"epoch": 0.55,
"learning_rate": 0.0002274824473420261,
"loss": 2.6371,
"theoretical_loss": 3.4586705298375167,
"tokens_seen": 1814167552
},
{
"epoch": 0.55,
"learning_rate": 0.00022747241725175528,
"loss": 2.7612,
"theoretical_loss": 3.4586599117681383,
"tokens_seen": 1814233088
},
{
"epoch": 0.55,
"learning_rate": 0.00022746238716148446,
"loss": 2.4688,
"theoretical_loss": 3.458649294189704,
"tokens_seen": 1814298624
},
{
"epoch": 0.55,
"learning_rate": 0.00022745235707121364,
"loss": 2.695,
"theoretical_loss": 3.458638677102174,
"tokens_seen": 1814364160
},
{
"epoch": 0.55,
"learning_rate": 0.00022744232698094282,
"loss": 2.8253,
"theoretical_loss": 3.4586280605055073,
"tokens_seen": 1814429696
},
{
"epoch": 0.55,
"learning_rate": 0.00022743229689067203,
"loss": 2.6178,
"theoretical_loss": 3.458617444399664,
"tokens_seen": 1814495232
},
{
"epoch": 0.55,
"learning_rate": 0.0002274222668004012,
"loss": 2.5768,
"theoretical_loss": 3.4586068287846032,
"tokens_seen": 1814560768
},
{
"epoch": 0.55,
"learning_rate": 0.0002274122367101304,
"loss": 2.8676,
"theoretical_loss": 3.458596213660285,
"tokens_seen": 1814626304
},
{
"epoch": 0.55,
"learning_rate": 0.00022740220661985957,
"loss": 2.5713,
"theoretical_loss": 3.4585855990266685,
"tokens_seen": 1814691840
},
{
"epoch": 0.55,
"learning_rate": 0.00022739217652958878,
"loss": 2.8937,
"theoretical_loss": 3.4585749848837133,
"tokens_seen": 1814757376
},
{
"epoch": 0.55,
"learning_rate": 0.00022738214643931796,
"loss": 2.7745,
"theoretical_loss": 3.45856437123138,
"tokens_seen": 1814822912
},
{
"epoch": 0.55,
"learning_rate": 0.00022737211634904714,
"loss": 2.6173,
"theoretical_loss": 3.4585537580696264,
"tokens_seen": 1814888448
},
{
"epoch": 0.55,
"learning_rate": 0.00022736208625877632,
"loss": 2.5409,
"theoretical_loss": 3.458543145398414,
"tokens_seen": 1814953984
},
{
"epoch": 0.55,
"learning_rate": 0.0002273520561685055,
"loss": 2.7487,
"theoretical_loss": 3.458532533217701,
"tokens_seen": 1815019520
},
{
"epoch": 0.55,
"learning_rate": 0.0002273420260782347,
"loss": 2.602,
"theoretical_loss": 3.4585219215274474,
"tokens_seen": 1815085056
},
{
"epoch": 0.55,
"learning_rate": 0.0002273319959879639,
"loss": 2.7759,
"theoretical_loss": 3.458511310327613,
"tokens_seen": 1815150592
},
{
"epoch": 0.55,
"learning_rate": 0.00022732196589769307,
"loss": 2.6712,
"theoretical_loss": 3.458500699618158,
"tokens_seen": 1815216128
},
{
"epoch": 0.55,
"learning_rate": 0.00022731193580742225,
"loss": 2.6527,
"theoretical_loss": 3.458490089399041,
"tokens_seen": 1815281664
},
{
"epoch": 0.55,
"objective/train/docs_used": 1026828,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.0092971324920654,
"objective/train/theoretical_loss": 3.4584794796702223,
"objective/train/tokens_used": 1835807200,
"theoretical_loss": 3.4584794796702223,
"tokens_seen": 1815347200
},
{
"epoch": 0.55,
"learning_rate": 0.00022730190571715146,
"loss": 2.5959,
"theoretical_loss": 3.4584794796702223,
"tokens_seen": 1815347200
},
{
"epoch": 0.55,
"learning_rate": 0.00022729187562688064,
"loss": 2.6648,
"theoretical_loss": 3.458468870431661,
"tokens_seen": 1815412736
},
{
"epoch": 0.55,
"learning_rate": 0.00022728184553660983,
"loss": 2.8715,
"theoretical_loss": 3.4584582616833175,
"tokens_seen": 1815478272
},
{
"epoch": 0.55,
"learning_rate": 0.000227271815446339,
"loss": 2.5651,
"theoretical_loss": 3.458447653425151,
"tokens_seen": 1815543808
},
{
"epoch": 0.55,
"learning_rate": 0.00022726178535606821,
"loss": 2.7024,
"theoretical_loss": 3.4584370456571207,
"tokens_seen": 1815609344
},
{
"epoch": 0.55,
"learning_rate": 0.0002272517552657974,
"loss": 2.8792,
"theoretical_loss": 3.4584264383791874,
"tokens_seen": 1815674880
},
{
"epoch": 0.55,
"learning_rate": 0.00022724172517552658,
"loss": 2.9049,
"theoretical_loss": 3.4584158315913096,
"tokens_seen": 1815740416
},
{
"epoch": 0.55,
"learning_rate": 0.00022723169508525576,
"loss": 2.5287,
"theoretical_loss": 3.458405225293448,
"tokens_seen": 1815805952
},
{
"epoch": 0.55,
"learning_rate": 0.00022722166499498494,
"loss": 2.4598,
"theoretical_loss": 3.458394619485561,
"tokens_seen": 1815871488
},
{
"epoch": 0.55,
"learning_rate": 0.00022721163490471415,
"loss": 3.0517,
"theoretical_loss": 3.4583840141676094,
"tokens_seen": 1815937024
},
{
"epoch": 0.55,
"learning_rate": 0.00022720160481444333,
"loss": 2.7051,
"theoretical_loss": 3.458373409339553,
"tokens_seen": 1816002560
},
{
"epoch": 0.55,
"learning_rate": 0.0002271915747241725,
"loss": 2.8352,
"theoretical_loss": 3.4583628050013506,
"tokens_seen": 1816068096
},
{
"epoch": 0.55,
"learning_rate": 0.0002271815446339017,
"loss": 2.5926,
"theoretical_loss": 3.4583522011529624,
"tokens_seen": 1816133632
},
{
"epoch": 0.55,
"learning_rate": 0.0002271715145436309,
"loss": 2.8284,
"theoretical_loss": 3.458341597794348,
"tokens_seen": 1816199168
},
{
"epoch": 0.55,
"learning_rate": 0.00022716148445336008,
"loss": 2.4703,
"theoretical_loss": 3.4583309949254675,
"tokens_seen": 1816264704
},
{
"epoch": 0.55,
"learning_rate": 0.00022715145436308926,
"loss": 2.6256,
"theoretical_loss": 3.4583203925462795,
"tokens_seen": 1816330240
},
{
"epoch": 0.55,
"learning_rate": 0.00022714142427281844,
"loss": 2.79,
"theoretical_loss": 3.4583097906567453,
"tokens_seen": 1816395776
},
{
"epoch": 0.55,
"learning_rate": 0.00022713139418254762,
"loss": 2.701,
"theoretical_loss": 3.458299189256823,
"tokens_seen": 1816461312
},
{
"epoch": 0.55,
"learning_rate": 0.00022712136409227683,
"loss": 2.5733,
"theoretical_loss": 3.458288588346474,
"tokens_seen": 1816526848
},
{
"epoch": 0.55,
"learning_rate": 0.000227111334002006,
"loss": 2.7726,
"theoretical_loss": 3.4582779879256567,
"tokens_seen": 1816592384
},
{
"epoch": 0.55,
"learning_rate": 0.00022710130391173522,
"loss": 2.3367,
"theoretical_loss": 3.458267387994331,
"tokens_seen": 1816657920
},
{
"epoch": 0.55,
"learning_rate": 0.0002270912738214644,
"loss": 2.4844,
"theoretical_loss": 3.458256788552457,
"tokens_seen": 1816723456
},
{
"epoch": 0.55,
"learning_rate": 0.0002270812437311936,
"loss": 2.8181,
"theoretical_loss": 3.4582461895999947,
"tokens_seen": 1816788992
},
{
"epoch": 0.55,
"learning_rate": 0.0002270712136409228,
"loss": 2.9987,
"theoretical_loss": 3.458235591136903,
"tokens_seen": 1816854528
},
{
"epoch": 0.55,
"learning_rate": 0.00022706118355065197,
"loss": 2.6856,
"theoretical_loss": 3.458224993163143,
"tokens_seen": 1816920064
},
{
"epoch": 0.55,
"objective/train/docs_used": 1028029,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4242141246795654,
"objective/train/theoretical_loss": 3.458214395678673,
"objective/train/tokens_used": 1837445600,
"theoretical_loss": 3.458214395678673,
"tokens_seen": 1816985600
},
{
"epoch": 0.55,
"learning_rate": 0.00022705115346038115,
"loss": 2.7533,
"theoretical_loss": 3.458214395678673,
"tokens_seen": 1816985600
},
{
"epoch": 0.55,
"learning_rate": 0.00022704112337011033,
"loss": 2.86,
"theoretical_loss": 3.4582037986834537,
"tokens_seen": 1817051136
},
{
"epoch": 0.55,
"learning_rate": 0.00022703109327983954,
"loss": 2.8266,
"theoretical_loss": 3.458193202177444,
"tokens_seen": 1817116672
},
{
"epoch": 0.55,
"learning_rate": 0.00022702106318956872,
"loss": 2.5635,
"theoretical_loss": 3.4581826061606047,
"tokens_seen": 1817182208
},
{
"epoch": 0.55,
"learning_rate": 0.0002270110330992979,
"loss": 2.6236,
"theoretical_loss": 3.458172010632895,
"tokens_seen": 1817247744
},
{
"epoch": 0.55,
"learning_rate": 0.00022700100300902709,
"loss": 2.8548,
"theoretical_loss": 3.458161415594275,
"tokens_seen": 1817313280
},
{
"epoch": 0.55,
"learning_rate": 0.0002269909729187563,
"loss": 2.8547,
"theoretical_loss": 3.458150821044704,
"tokens_seen": 1817378816
},
{
"epoch": 0.55,
"learning_rate": 0.00022698094282848548,
"loss": 2.5282,
"theoretical_loss": 3.458140226984142,
"tokens_seen": 1817444352
},
{
"epoch": 0.55,
"learning_rate": 0.00022697091273821466,
"loss": 2.7745,
"theoretical_loss": 3.458129633412549,
"tokens_seen": 1817509888
},
{
"epoch": 0.55,
"learning_rate": 0.00022696088264794384,
"loss": 2.8257,
"theoretical_loss": 3.4581190403298847,
"tokens_seen": 1817575424
},
{
"epoch": 0.55,
"learning_rate": 0.00022695085255767302,
"loss": 2.787,
"theoretical_loss": 3.458108447736109,
"tokens_seen": 1817640960
},
{
"epoch": 0.55,
"learning_rate": 0.00022694082246740223,
"loss": 2.726,
"theoretical_loss": 3.4580978556311814,
"tokens_seen": 1817706496
},
{
"epoch": 0.55,
"learning_rate": 0.0002269307923771314,
"loss": 2.6904,
"theoretical_loss": 3.458087264015062,
"tokens_seen": 1817772032
},
{
"epoch": 0.55,
"learning_rate": 0.0002269207622868606,
"loss": 2.5765,
"theoretical_loss": 3.45807667288771,
"tokens_seen": 1817837568
},
{
"epoch": 0.55,
"learning_rate": 0.00022691073219658977,
"loss": 2.5597,
"theoretical_loss": 3.4580660822490863,
"tokens_seen": 1817903104
},
{
"epoch": 0.55,
"learning_rate": 0.00022690070210631898,
"loss": 2.5947,
"theoretical_loss": 3.45805549209915,
"tokens_seen": 1817968640
},
{
"epoch": 0.55,
"learning_rate": 0.00022689067201604816,
"loss": 2.6368,
"theoretical_loss": 3.4580449024378614,
"tokens_seen": 1818034176
},
{
"epoch": 0.55,
"learning_rate": 0.00022688064192577734,
"loss": 2.7125,
"theoretical_loss": 3.4580343132651796,
"tokens_seen": 1818099712
},
{
"epoch": 0.55,
"learning_rate": 0.00022687061183550652,
"loss": 2.4421,
"theoretical_loss": 3.458023724581065,
"tokens_seen": 1818165248
},
{
"epoch": 0.55,
"learning_rate": 0.0002268605817452357,
"loss": 2.7761,
"theoretical_loss": 3.4580131363854782,
"tokens_seen": 1818230784
},
{
"epoch": 0.55,
"learning_rate": 0.0002268505516549649,
"loss": 2.5529,
"theoretical_loss": 3.458002548678377,
"tokens_seen": 1818296320
},
{
"epoch": 0.55,
"learning_rate": 0.0002268405215646941,
"loss": 2.6957,
"theoretical_loss": 3.457991961459723,
"tokens_seen": 1818361856
},
{
"epoch": 0.55,
"learning_rate": 0.00022683049147442327,
"loss": 2.7052,
"theoretical_loss": 3.4579813747294756,
"tokens_seen": 1818427392
},
{
"epoch": 0.55,
"learning_rate": 0.00022682046138415245,
"loss": 2.7017,
"theoretical_loss": 3.4579707884875948,
"tokens_seen": 1818492928
},
{
"epoch": 0.55,
"learning_rate": 0.00022681043129388166,
"loss": 2.4547,
"theoretical_loss": 3.4579602027340393,
"tokens_seen": 1818558464
},
{
"epoch": 0.55,
"objective/train/docs_used": 1028792,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5927464962005615,
"objective/train/theoretical_loss": 3.457949617468771,
"objective/train/tokens_used": 1839084000,
"theoretical_loss": 3.457949617468771,
"tokens_seen": 1818624000
},
{
"epoch": 0.55,
"learning_rate": 0.00022680040120361084,
"loss": 2.7141,
"theoretical_loss": 3.457949617468771,
"tokens_seen": 1818624000
},
{
"epoch": 0.55,
"learning_rate": 0.00022679037111334003,
"loss": 2.6938,
"theoretical_loss": 3.457939032691748,
"tokens_seen": 1818689536
},
{
"epoch": 0.55,
"learning_rate": 0.0002267803410230692,
"loss": 2.8209,
"theoretical_loss": 3.4579284484029316,
"tokens_seen": 1818755072
},
{
"epoch": 0.55,
"learning_rate": 0.00022677031093279841,
"loss": 2.7335,
"theoretical_loss": 3.45791786460228,
"tokens_seen": 1818820608
},
{
"epoch": 0.55,
"learning_rate": 0.0002267602808425276,
"loss": 2.5717,
"theoretical_loss": 3.4579072812897547,
"tokens_seen": 1818886144
},
{
"epoch": 0.55,
"learning_rate": 0.00022675025075225678,
"loss": 2.7329,
"theoretical_loss": 3.457896698465315,
"tokens_seen": 1818951680
},
{
"epoch": 0.55,
"learning_rate": 0.00022674022066198596,
"loss": 2.6791,
"theoretical_loss": 3.457886116128921,
"tokens_seen": 1819017216
},
{
"epoch": 0.55,
"learning_rate": 0.00022673019057171514,
"loss": 2.5895,
"theoretical_loss": 3.4578755342805323,
"tokens_seen": 1819082752
},
{
"epoch": 0.55,
"learning_rate": 0.00022672016048144435,
"loss": 2.6917,
"theoretical_loss": 3.457864952920109,
"tokens_seen": 1819148288
},
{
"epoch": 0.55,
"learning_rate": 0.00022671013039117353,
"loss": 2.6403,
"theoretical_loss": 3.4578543720476103,
"tokens_seen": 1819213824
},
{
"epoch": 0.55,
"learning_rate": 0.0002267001003009027,
"loss": 2.7765,
"theoretical_loss": 3.4578437916629974,
"tokens_seen": 1819279360
},
{
"epoch": 0.55,
"learning_rate": 0.0002266900702106319,
"loss": 2.7724,
"theoretical_loss": 3.4578332117662294,
"tokens_seen": 1819344896
},
{
"epoch": 0.55,
"learning_rate": 0.0002266800401203611,
"loss": 2.7127,
"theoretical_loss": 3.4578226323572663,
"tokens_seen": 1819410432
},
{
"epoch": 0.55,
"learning_rate": 0.00022667001003009028,
"loss": 2.5411,
"theoretical_loss": 3.4578120534360686,
"tokens_seen": 1819475968
},
{
"epoch": 0.55,
"learning_rate": 0.00022665997993981946,
"loss": 2.4434,
"theoretical_loss": 3.4578014750025954,
"tokens_seen": 1819541504
},
{
"epoch": 0.55,
"learning_rate": 0.00022664994984954864,
"loss": 2.6911,
"theoretical_loss": 3.457790897056807,
"tokens_seen": 1819607040
},
{
"epoch": 0.55,
"learning_rate": 0.00022663991975927782,
"loss": 2.6099,
"theoretical_loss": 3.4577803195986636,
"tokens_seen": 1819672576
},
{
"epoch": 0.55,
"learning_rate": 0.00022662988966900703,
"loss": 2.8166,
"theoretical_loss": 3.457769742628125,
"tokens_seen": 1819738112
},
{
"epoch": 0.55,
"learning_rate": 0.0002266198595787362,
"loss": 2.7286,
"theoretical_loss": 3.4577591661451517,
"tokens_seen": 1819803648
},
{
"epoch": 0.55,
"learning_rate": 0.0002266098294884654,
"loss": 2.9856,
"theoretical_loss": 3.457748590149703,
"tokens_seen": 1819869184
},
{
"epoch": 0.55,
"learning_rate": 0.00022659979939819458,
"loss": 2.8793,
"theoretical_loss": 3.457738014641738,
"tokens_seen": 1819934720
},
{
"epoch": 0.55,
"learning_rate": 0.00022658976930792378,
"loss": 2.5655,
"theoretical_loss": 3.4577274396212183,
"tokens_seen": 1820000256
},
{
"epoch": 0.55,
"learning_rate": 0.00022657973921765296,
"loss": 2.8939,
"theoretical_loss": 3.457716865088103,
"tokens_seen": 1820065792
},
{
"epoch": 0.55,
"learning_rate": 0.00022656970912738215,
"loss": 2.8657,
"theoretical_loss": 3.457706291042353,
"tokens_seen": 1820131328
},
{
"epoch": 0.55,
"learning_rate": 0.00022655967903711133,
"loss": 2.6213,
"theoretical_loss": 3.4576957174839267,
"tokens_seen": 1820196864
},
{
"epoch": 0.55,
"objective/train/docs_used": 1029550,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.575042486190796,
"objective/train/theoretical_loss": 3.4576851444127854,
"objective/train/tokens_used": 1840722400,
"theoretical_loss": 3.4576851444127854,
"tokens_seen": 1820262400
},
{
"epoch": 0.55,
"learning_rate": 0.0002265496489468405,
"loss": 2.5446,
"theoretical_loss": 3.4576851444127854,
"tokens_seen": 1820262400
},
{
"epoch": 0.55,
"learning_rate": 0.00022653961885656972,
"loss": 2.6461,
"theoretical_loss": 3.457674571828889,
"tokens_seen": 1820327936
},
{
"epoch": 0.55,
"learning_rate": 0.0002265295887662989,
"loss": 2.6348,
"theoretical_loss": 3.457663999732197,
"tokens_seen": 1820393472
},
{
"epoch": 0.55,
"learning_rate": 0.00022651955867602808,
"loss": 2.4401,
"theoretical_loss": 3.4576534281226694,
"tokens_seen": 1820459008
},
{
"epoch": 0.55,
"learning_rate": 0.00022650952858575726,
"loss": 2.6116,
"theoretical_loss": 3.457642857000267,
"tokens_seen": 1820524544
},
{
"epoch": 0.55,
"learning_rate": 0.00022649949849548647,
"loss": 2.5868,
"theoretical_loss": 3.457632286364949,
"tokens_seen": 1820590080
},
{
"epoch": 0.55,
"learning_rate": 0.00022648946840521565,
"loss": 2.7118,
"theoretical_loss": 3.4576217162166754,
"tokens_seen": 1820655616
},
{
"epoch": 0.55,
"learning_rate": 0.00022647943831494483,
"loss": 2.7007,
"theoretical_loss": 3.457611146555407,
"tokens_seen": 1820721152
},
{
"epoch": 0.55,
"learning_rate": 0.000226469408224674,
"loss": 2.7227,
"theoretical_loss": 3.457600577381103,
"tokens_seen": 1820786688
},
{
"epoch": 0.55,
"learning_rate": 0.0002264593781344032,
"loss": 2.6448,
"theoretical_loss": 3.4575900086937237,
"tokens_seen": 1820852224
},
{
"epoch": 0.55,
"learning_rate": 0.0002264493480441324,
"loss": 2.6891,
"theoretical_loss": 3.4575794404932294,
"tokens_seen": 1820917760
},
{
"epoch": 0.55,
"learning_rate": 0.00022643931795386158,
"loss": 2.4734,
"theoretical_loss": 3.4575688727795804,
"tokens_seen": 1820983296
},
{
"epoch": 0.55,
"learning_rate": 0.00022642928786359076,
"loss": 2.6677,
"theoretical_loss": 3.4575583055527357,
"tokens_seen": 1821048832
},
{
"epoch": 0.55,
"learning_rate": 0.00022641925777331994,
"loss": 2.9709,
"theoretical_loss": 3.4575477388126563,
"tokens_seen": 1821114368
},
{
"epoch": 0.55,
"learning_rate": 0.00022640922768304915,
"loss": 2.9994,
"theoretical_loss": 3.457537172559302,
"tokens_seen": 1821179904
},
{
"epoch": 0.55,
"learning_rate": 0.00022639919759277833,
"loss": 2.9089,
"theoretical_loss": 3.4575266067926327,
"tokens_seen": 1821245440
},
{
"epoch": 0.55,
"learning_rate": 0.00022638916750250751,
"loss": 2.9562,
"theoretical_loss": 3.4575160415126085,
"tokens_seen": 1821310976
},
{
"epoch": 0.55,
"learning_rate": 0.0002263791374122367,
"loss": 2.6481,
"theoretical_loss": 3.4575054767191897,
"tokens_seen": 1821376512
},
{
"epoch": 0.55,
"learning_rate": 0.00022636910732196588,
"loss": 2.7051,
"theoretical_loss": 3.4574949124123364,
"tokens_seen": 1821442048
},
{
"epoch": 0.55,
"learning_rate": 0.00022635907723169508,
"loss": 2.6937,
"theoretical_loss": 3.457484348592008,
"tokens_seen": 1821507584
},
{
"epoch": 0.55,
"learning_rate": 0.0002263490471414243,
"loss": 2.6771,
"theoretical_loss": 3.4574737852581654,
"tokens_seen": 1821573120
},
{
"epoch": 0.55,
"learning_rate": 0.00022633901705115347,
"loss": 2.7455,
"theoretical_loss": 3.4574632224107686,
"tokens_seen": 1821638656
},
{
"epoch": 0.55,
"learning_rate": 0.00022632898696088266,
"loss": 2.5539,
"theoretical_loss": 3.4574526600497775,
"tokens_seen": 1821704192
},
{
"epoch": 0.55,
"learning_rate": 0.00022631895687061186,
"loss": 2.7796,
"theoretical_loss": 3.457442098175152,
"tokens_seen": 1821769728
},
{
"epoch": 0.55,
"learning_rate": 0.00022630892678034104,
"loss": 2.818,
"theoretical_loss": 3.457431536786853,
"tokens_seen": 1821835264
},
{
"epoch": 0.55,
"objective/train/docs_used": 1030259,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5886473655700684,
"objective/train/theoretical_loss": 3.4574209758848395,
"objective/train/tokens_used": 1842360800,
"theoretical_loss": 3.4574209758848395,
"tokens_seen": 1821900800
},
{
"epoch": 0.55,
"learning_rate": 0.00022629889669007023,
"loss": 2.6606,
"theoretical_loss": 3.4574209758848395,
"tokens_seen": 1821900800
},
{
"epoch": 0.55,
"learning_rate": 0.0002262888665997994,
"loss": 2.4995,
"theoretical_loss": 3.457410415469072,
"tokens_seen": 1821966336
},
{
"epoch": 0.55,
"learning_rate": 0.00022627883650952861,
"loss": 2.5592,
"theoretical_loss": 3.4573998555395113,
"tokens_seen": 1822031872
},
{
"epoch": 0.55,
"learning_rate": 0.0002262688064192578,
"loss": 2.777,
"theoretical_loss": 3.4573892960961166,
"tokens_seen": 1822097408
},
{
"epoch": 0.55,
"learning_rate": 0.00022625877632898698,
"loss": 2.6319,
"theoretical_loss": 3.457378737138849,
"tokens_seen": 1822162944
},
{
"epoch": 0.55,
"learning_rate": 0.00022624874623871616,
"loss": 2.4684,
"theoretical_loss": 3.4573681786676675,
"tokens_seen": 1822228480
},
{
"epoch": 0.55,
"learning_rate": 0.00022623871614844534,
"loss": 2.7476,
"theoretical_loss": 3.457357620682533,
"tokens_seen": 1822294016
},
{
"epoch": 0.55,
"learning_rate": 0.00022622868605817455,
"loss": 2.6558,
"theoretical_loss": 3.4573470631834056,
"tokens_seen": 1822359552
},
{
"epoch": 0.55,
"learning_rate": 0.00022621865596790373,
"loss": 2.5614,
"theoretical_loss": 3.457336506170245,
"tokens_seen": 1822425088
},
{
"epoch": 0.55,
"learning_rate": 0.0002262086258776329,
"loss": 2.842,
"theoretical_loss": 3.457325949643012,
"tokens_seen": 1822490624
},
{
"epoch": 0.55,
"learning_rate": 0.0002261985957873621,
"loss": 2.5359,
"theoretical_loss": 3.4573153936016663,
"tokens_seen": 1822556160
},
{
"epoch": 0.55,
"learning_rate": 0.0002261885656970913,
"loss": 2.6835,
"theoretical_loss": 3.4573048380461686,
"tokens_seen": 1822621696
},
{
"epoch": 0.55,
"learning_rate": 0.00022617853560682048,
"loss": 2.5674,
"theoretical_loss": 3.4572942829764783,
"tokens_seen": 1822687232
},
{
"epoch": 0.55,
"learning_rate": 0.00022616850551654966,
"loss": 2.6835,
"theoretical_loss": 3.4572837283925564,
"tokens_seen": 1822752768
},
{
"epoch": 0.55,
"learning_rate": 0.00022615847542627884,
"loss": 2.7221,
"theoretical_loss": 3.457273174294362,
"tokens_seen": 1822818304
},
{
"epoch": 0.55,
"learning_rate": 0.00022614844533600802,
"loss": 2.7172,
"theoretical_loss": 3.457262620681856,
"tokens_seen": 1822883840
},
{
"epoch": 0.55,
"learning_rate": 0.00022613841524573723,
"loss": 2.5392,
"theoretical_loss": 3.4572520675549985,
"tokens_seen": 1822949376
},
{
"epoch": 0.55,
"learning_rate": 0.0002261283851554664,
"loss": 2.4944,
"theoretical_loss": 3.45724151491375,
"tokens_seen": 1823014912
},
{
"epoch": 0.55,
"learning_rate": 0.0002261183550651956,
"loss": 2.8742,
"theoretical_loss": 3.4572309627580706,
"tokens_seen": 1823080448
},
{
"epoch": 0.55,
"learning_rate": 0.00022610832497492478,
"loss": 2.8316,
"theoretical_loss": 3.4572204110879197,
"tokens_seen": 1823145984
},
{
"epoch": 0.55,
"learning_rate": 0.00022609829488465398,
"loss": 2.6102,
"theoretical_loss": 3.4572098599032586,
"tokens_seen": 1823211520
},
{
"epoch": 0.55,
"learning_rate": 0.00022608826479438316,
"loss": 2.7842,
"theoretical_loss": 3.4571993092040465,
"tokens_seen": 1823277056
},
{
"epoch": 0.55,
"learning_rate": 0.00022607823470411235,
"loss": 2.4737,
"theoretical_loss": 3.4571887589902444,
"tokens_seen": 1823342592
},
{
"epoch": 0.55,
"learning_rate": 0.00022606820461384153,
"loss": 2.7915,
"theoretical_loss": 3.457178209261812,
"tokens_seen": 1823408128
},
{
"epoch": 0.55,
"learning_rate": 0.0002260581745235707,
"loss": 2.7791,
"theoretical_loss": 3.457167660018711,
"tokens_seen": 1823473664
},
{
"epoch": 0.55,
"objective/train/docs_used": 1030966,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.731187582015991,
"objective/train/theoretical_loss": 3.4571571112608988,
"objective/train/tokens_used": 1843999200,
"theoretical_loss": 3.4571571112608988,
"tokens_seen": 1823539200
},
{
"epoch": 0.55,
"learning_rate": 0.00022604814443329992,
"loss": 2.4659,
"theoretical_loss": 3.4571571112608988,
"tokens_seen": 1823539200
},
{
"epoch": 0.55,
"learning_rate": 0.0002260381143430291,
"loss": 2.6075,
"theoretical_loss": 3.4571465629883384,
"tokens_seen": 1823604736
},
{
"epoch": 0.55,
"learning_rate": 0.00022602808425275828,
"loss": 2.6019,
"theoretical_loss": 3.4571360152009882,
"tokens_seen": 1823670272
},
{
"epoch": 0.55,
"learning_rate": 0.00022601805416248746,
"loss": 2.9266,
"theoretical_loss": 3.4571254678988095,
"tokens_seen": 1823735808
},
{
"epoch": 0.55,
"learning_rate": 0.00022600802407221667,
"loss": 2.3942,
"theoretical_loss": 3.4571149210817618,
"tokens_seen": 1823801344
},
{
"epoch": 0.55,
"learning_rate": 0.00022599799398194585,
"loss": 2.7659,
"theoretical_loss": 3.4571043747498065,
"tokens_seen": 1823866880
},
{
"epoch": 0.55,
"learning_rate": 0.00022598796389167503,
"loss": 2.4701,
"theoretical_loss": 3.4570938289029023,
"tokens_seen": 1823932416
},
{
"epoch": 0.55,
"learning_rate": 0.0002259779338014042,
"loss": 2.5838,
"theoretical_loss": 3.457083283541011,
"tokens_seen": 1823997952
},
{
"epoch": 0.55,
"learning_rate": 0.0002259679037111334,
"loss": 2.4068,
"theoretical_loss": 3.4570727386640914,
"tokens_seen": 1824063488
},
{
"epoch": 0.55,
"learning_rate": 0.0002259578736208626,
"loss": 2.5393,
"theoretical_loss": 3.4570621942721047,
"tokens_seen": 1824129024
},
{
"epoch": 0.55,
"learning_rate": 0.00022594784353059178,
"loss": 2.7426,
"theoretical_loss": 3.4570516503650115,
"tokens_seen": 1824194560
},
{
"epoch": 0.55,
"learning_rate": 0.00022593781344032096,
"loss": 2.6723,
"theoretical_loss": 3.4570411069427713,
"tokens_seen": 1824260096
},
{
"epoch": 0.55,
"learning_rate": 0.00022592778335005014,
"loss": 2.6455,
"theoretical_loss": 3.4570305640053443,
"tokens_seen": 1824325632
},
{
"epoch": 0.55,
"learning_rate": 0.00022591775325977935,
"loss": 2.3888,
"theoretical_loss": 3.457020021552691,
"tokens_seen": 1824391168
},
{
"epoch": 0.55,
"learning_rate": 0.00022590772316950853,
"loss": 2.8392,
"theoretical_loss": 3.4570094795847726,
"tokens_seen": 1824456704
},
{
"epoch": 0.55,
"learning_rate": 0.00022589769307923771,
"loss": 2.7531,
"theoretical_loss": 3.456998938101548,
"tokens_seen": 1824522240
},
{
"epoch": 0.55,
"learning_rate": 0.0002258876629889669,
"loss": 2.6771,
"theoretical_loss": 3.456988397102979,
"tokens_seen": 1824587776
},
{
"epoch": 0.55,
"learning_rate": 0.00022587763289869608,
"loss": 2.7412,
"theoretical_loss": 3.4569778565890243,
"tokens_seen": 1824653312
},
{
"epoch": 0.55,
"learning_rate": 0.00022586760280842528,
"loss": 2.6498,
"theoretical_loss": 3.456967316559645,
"tokens_seen": 1824718848
},
{
"epoch": 0.55,
"learning_rate": 0.00022585757271815447,
"loss": 2.7807,
"theoretical_loss": 3.4569567770148018,
"tokens_seen": 1824784384
},
{
"epoch": 0.55,
"learning_rate": 0.00022584754262788365,
"loss": 2.6607,
"theoretical_loss": 3.4569462379544547,
"tokens_seen": 1824849920
},
{
"epoch": 0.55,
"learning_rate": 0.00022583751253761283,
"loss": 2.5675,
"theoretical_loss": 3.4569356993785636,
"tokens_seen": 1824915456
},
{
"epoch": 0.55,
"learning_rate": 0.00022582748244734204,
"loss": 2.4471,
"theoretical_loss": 3.456925161287089,
"tokens_seen": 1824980992
},
{
"epoch": 0.55,
"learning_rate": 0.00022581745235707122,
"loss": 2.4413,
"theoretical_loss": 3.456914623679992,
"tokens_seen": 1825046528
},
{
"epoch": 0.55,
"learning_rate": 0.0002258074222668004,
"loss": 2.7595,
"theoretical_loss": 3.4569040865572322,
"tokens_seen": 1825112064
},
{
"epoch": 0.55,
"objective/train/docs_used": 1032081,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4678499698638916,
"objective/train/theoretical_loss": 3.4568935499187696,
"objective/train/tokens_used": 1845637600,
"theoretical_loss": 3.4568935499187696,
"tokens_seen": 1825177600
},
{
"epoch": 0.55,
"learning_rate": 0.00022579739217652958,
"loss": 2.6109,
"theoretical_loss": 3.4568935499187696,
"tokens_seen": 1825177600
},
{
"epoch": 0.55,
"learning_rate": 0.00022578736208625876,
"loss": 2.6245,
"theoretical_loss": 3.456883013764566,
"tokens_seen": 1825243136
},
{
"epoch": 0.55,
"learning_rate": 0.00022577733199598797,
"loss": 2.6317,
"theoretical_loss": 3.4568724780945796,
"tokens_seen": 1825308672
},
{
"epoch": 0.55,
"learning_rate": 0.00022576730190571715,
"loss": 2.5272,
"theoretical_loss": 3.456861942908773,
"tokens_seen": 1825374208
},
{
"epoch": 0.55,
"learning_rate": 0.00022575727181544633,
"loss": 2.729,
"theoretical_loss": 3.456851408207105,
"tokens_seen": 1825439744
},
{
"epoch": 0.55,
"learning_rate": 0.0002257472417251755,
"loss": 2.7754,
"theoretical_loss": 3.456840873989537,
"tokens_seen": 1825505280
},
{
"epoch": 0.55,
"learning_rate": 0.00022573721163490472,
"loss": 2.8737,
"theoretical_loss": 3.4568303402560288,
"tokens_seen": 1825570816
},
{
"epoch": 0.55,
"learning_rate": 0.0002257271815446339,
"loss": 2.6935,
"theoretical_loss": 3.4568198070065406,
"tokens_seen": 1825636352
},
{
"epoch": 0.55,
"learning_rate": 0.00022571715145436308,
"loss": 2.8063,
"theoretical_loss": 3.4568092742410332,
"tokens_seen": 1825701888
},
{
"epoch": 0.55,
"learning_rate": 0.00022570712136409226,
"loss": 2.7725,
"theoretical_loss": 3.456798741959467,
"tokens_seen": 1825767424
},
{
"epoch": 0.55,
"learning_rate": 0.00022569709127382147,
"loss": 2.6589,
"theoretical_loss": 3.456788210161802,
"tokens_seen": 1825832960
},
{
"epoch": 0.55,
"learning_rate": 0.00022568706118355065,
"loss": 2.6791,
"theoretical_loss": 3.456777678847999,
"tokens_seen": 1825898496
},
{
"epoch": 0.55,
"learning_rate": 0.00022567703109327983,
"loss": 2.791,
"theoretical_loss": 3.4567671480180184,
"tokens_seen": 1825964032
},
{
"epoch": 0.55,
"learning_rate": 0.00022566700100300902,
"loss": 2.6776,
"theoretical_loss": 3.4567566176718203,
"tokens_seen": 1826029568
},
{
"epoch": 0.55,
"learning_rate": 0.0002256569709127382,
"loss": 2.5527,
"theoretical_loss": 3.456746087809365,
"tokens_seen": 1826095104
},
{
"epoch": 0.55,
"learning_rate": 0.0002256469408224674,
"loss": 2.869,
"theoretical_loss": 3.456735558430614,
"tokens_seen": 1826160640
},
{
"epoch": 0.55,
"learning_rate": 0.00022563691073219659,
"loss": 2.8585,
"theoretical_loss": 3.4567250295355265,
"tokens_seen": 1826226176
},
{
"epoch": 0.55,
"learning_rate": 0.00022562688064192577,
"loss": 2.6914,
"theoretical_loss": 3.456714501124063,
"tokens_seen": 1826291712
},
{
"epoch": 0.55,
"learning_rate": 0.00022561685055165495,
"loss": 2.4155,
"theoretical_loss": 3.4567039731961846,
"tokens_seen": 1826357248
},
{
"epoch": 0.55,
"learning_rate": 0.00022560682046138416,
"loss": 2.7334,
"theoretical_loss": 3.4566934457518514,
"tokens_seen": 1826422784
},
{
"epoch": 0.55,
"learning_rate": 0.00022559679037111336,
"loss": 2.5001,
"theoretical_loss": 3.456682918791024,
"tokens_seen": 1826488320
},
{
"epoch": 0.55,
"learning_rate": 0.00022558676028084255,
"loss": 2.7526,
"theoretical_loss": 3.4566723923136626,
"tokens_seen": 1826553856
},
{
"epoch": 0.55,
"learning_rate": 0.00022557673019057173,
"loss": 2.6493,
"theoretical_loss": 3.4566618663197275,
"tokens_seen": 1826619392
},
{
"epoch": 0.55,
"learning_rate": 0.0002255667001003009,
"loss": 2.77,
"theoretical_loss": 3.45665134080918,
"tokens_seen": 1826684928
},
{
"epoch": 0.55,
"learning_rate": 0.00022555667001003012,
"loss": 2.6038,
"theoretical_loss": 3.4566408157819795,
"tokens_seen": 1826750464
},
{
"epoch": 0.55,
"objective/train/docs_used": 1032630,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1164731979370117,
"objective/train/theoretical_loss": 3.456630291238087,
"objective/train/tokens_used": 1847276000,
"theoretical_loss": 3.456630291238087,
"tokens_seen": 1826816000
},
{
"epoch": 0.55,
"learning_rate": 0.0002255466399197593,
"loss": 2.8416,
"theoretical_loss": 3.456630291238087,
"tokens_seen": 1826816000
},
{
"epoch": 0.55,
"learning_rate": 0.00022553660982948848,
"loss": 2.528,
"theoretical_loss": 3.4566197671774628,
"tokens_seen": 1826881536
},
{
"epoch": 0.55,
"learning_rate": 0.00022552657973921766,
"loss": 2.5913,
"theoretical_loss": 3.456609243600068,
"tokens_seen": 1826947072
},
{
"epoch": 0.55,
"learning_rate": 0.00022551654964894687,
"loss": 2.7065,
"theoretical_loss": 3.456598720505862,
"tokens_seen": 1827012608
},
{
"epoch": 0.55,
"learning_rate": 0.00022550651955867605,
"loss": 2.7562,
"theoretical_loss": 3.4565881978948063,
"tokens_seen": 1827078144
},
{
"epoch": 0.55,
"learning_rate": 0.00022549648946840523,
"loss": 2.7122,
"theoretical_loss": 3.4565776757668605,
"tokens_seen": 1827143680
},
{
"epoch": 0.55,
"learning_rate": 0.0002254864593781344,
"loss": 2.6408,
"theoretical_loss": 3.456567154121986,
"tokens_seen": 1827209216
},
{
"epoch": 0.55,
"learning_rate": 0.0002254764292878636,
"loss": 2.8434,
"theoretical_loss": 3.4565566329601425,
"tokens_seen": 1827274752
},
{
"epoch": 0.55,
"learning_rate": 0.0002254663991975928,
"loss": 2.9547,
"theoretical_loss": 3.4565461122812904,
"tokens_seen": 1827340288
},
{
"epoch": 0.55,
"learning_rate": 0.00022545636910732198,
"loss": 2.8556,
"theoretical_loss": 3.4565355920853915,
"tokens_seen": 1827405824
},
{
"epoch": 0.55,
"learning_rate": 0.00022544633901705116,
"loss": 2.7293,
"theoretical_loss": 3.456525072372405,
"tokens_seen": 1827471360
},
{
"epoch": 0.55,
"learning_rate": 0.00022543630892678034,
"loss": 2.5229,
"theoretical_loss": 3.456514553142292,
"tokens_seen": 1827536896
},
{
"epoch": 0.55,
"learning_rate": 0.00022542627883650955,
"loss": 2.6942,
"theoretical_loss": 3.456504034395013,
"tokens_seen": 1827602432
},
{
"epoch": 0.55,
"learning_rate": 0.00022541624874623873,
"loss": 2.8115,
"theoretical_loss": 3.4564935161305277,
"tokens_seen": 1827667968
},
{
"epoch": 0.55,
"learning_rate": 0.00022540621865596791,
"loss": 2.8297,
"theoretical_loss": 3.4564829983487977,
"tokens_seen": 1827733504
},
{
"epoch": 0.55,
"learning_rate": 0.0002253961885656971,
"loss": 2.7058,
"theoretical_loss": 3.4564724810497838,
"tokens_seen": 1827799040
},
{
"epoch": 0.55,
"learning_rate": 0.00022538615847542628,
"loss": 2.528,
"theoretical_loss": 3.456461964233445,
"tokens_seen": 1827864576
},
{
"epoch": 0.55,
"learning_rate": 0.00022537612838515548,
"loss": 2.8008,
"theoretical_loss": 3.4564514478997435,
"tokens_seen": 1827930112
},
{
"epoch": 0.55,
"learning_rate": 0.00022536609829488467,
"loss": 2.5659,
"theoretical_loss": 3.456440932048639,
"tokens_seen": 1827995648
},
{
"epoch": 0.55,
"learning_rate": 0.00022535606820461385,
"loss": 2.6914,
"theoretical_loss": 3.4564304166800914,
"tokens_seen": 1828061184
},
{
"epoch": 0.55,
"learning_rate": 0.00022534603811434303,
"loss": 2.6366,
"theoretical_loss": 3.456419901794063,
"tokens_seen": 1828126720
},
{
"epoch": 0.55,
"learning_rate": 0.00022533600802407224,
"loss": 2.5514,
"theoretical_loss": 3.4564093873905124,
"tokens_seen": 1828192256
},
{
"epoch": 0.55,
"learning_rate": 0.00022532597793380142,
"loss": 2.6612,
"theoretical_loss": 3.4563988734694018,
"tokens_seen": 1828257792
},
{
"epoch": 0.55,
"learning_rate": 0.0002253159478435306,
"loss": 2.6148,
"theoretical_loss": 3.456388360030691,
"tokens_seen": 1828323328
},
{
"epoch": 0.55,
"learning_rate": 0.00022530591775325978,
"loss": 2.6724,
"theoretical_loss": 3.4563778470743403,
"tokens_seen": 1828388864
},
{
"epoch": 0.55,
"objective/train/docs_used": 1034021,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6591670513153076,
"objective/train/theoretical_loss": 3.456367334600311,
"objective/train/tokens_used": 1848914400,
"theoretical_loss": 3.456367334600311,
"tokens_seen": 1828454400
},
{
"epoch": 0.55,
"learning_rate": 0.00022529588766298896,
"loss": 2.6596,
"theoretical_loss": 3.456367334600311,
"tokens_seen": 1828454400
},
{
"epoch": 0.55,
"learning_rate": 0.00022528585757271817,
"loss": 2.6059,
"theoretical_loss": 3.4563568226085635,
"tokens_seen": 1828519936
},
{
"epoch": 0.55,
"learning_rate": 0.00022527582748244735,
"loss": 2.5577,
"theoretical_loss": 3.456346311099058,
"tokens_seen": 1828585472
},
{
"epoch": 0.55,
"learning_rate": 0.00022526579739217653,
"loss": 2.9204,
"theoretical_loss": 3.4563358000717552,
"tokens_seen": 1828651008
},
{
"epoch": 0.55,
"learning_rate": 0.0002252557673019057,
"loss": 2.7624,
"theoretical_loss": 3.456325289526616,
"tokens_seen": 1828716544
},
{
"epoch": 0.55,
"learning_rate": 0.00022524573721163492,
"loss": 2.6035,
"theoretical_loss": 3.456314779463601,
"tokens_seen": 1828782080
},
{
"epoch": 0.55,
"learning_rate": 0.0002252357071213641,
"loss": 2.7851,
"theoretical_loss": 3.4563042698826707,
"tokens_seen": 1828847616
},
{
"epoch": 0.55,
"learning_rate": 0.00022522567703109328,
"loss": 2.7526,
"theoretical_loss": 3.456293760783786,
"tokens_seen": 1828913152
},
{
"epoch": 0.55,
"learning_rate": 0.00022521564694082246,
"loss": 2.6752,
"theoretical_loss": 3.456283252166906,
"tokens_seen": 1828978688
},
{
"epoch": 0.55,
"learning_rate": 0.00022520561685055167,
"loss": 2.7563,
"theoretical_loss": 3.4562727440319936,
"tokens_seen": 1829044224
},
{
"epoch": 0.55,
"learning_rate": 0.00022519558676028085,
"loss": 2.6124,
"theoretical_loss": 3.4562622363790076,
"tokens_seen": 1829109760
},
{
"epoch": 0.55,
"learning_rate": 0.00022518555667001003,
"loss": 2.5243,
"theoretical_loss": 3.45625172920791,
"tokens_seen": 1829175296
},
{
"epoch": 0.55,
"learning_rate": 0.00022517552657973922,
"loss": 2.7274,
"theoretical_loss": 3.45624122251866,
"tokens_seen": 1829240832
},
{
"epoch": 0.55,
"learning_rate": 0.0002251654964894684,
"loss": 2.3635,
"theoretical_loss": 3.45623071631122,
"tokens_seen": 1829306368
},
{
"epoch": 0.55,
"learning_rate": 0.0002251554663991976,
"loss": 2.8364,
"theoretical_loss": 3.456220210585549,
"tokens_seen": 1829371904
},
{
"epoch": 0.55,
"learning_rate": 0.00022514543630892679,
"loss": 2.7032,
"theoretical_loss": 3.4562097053416085,
"tokens_seen": 1829437440
},
{
"epoch": 0.55,
"learning_rate": 0.00022513540621865597,
"loss": 2.7508,
"theoretical_loss": 3.456199200579359,
"tokens_seen": 1829502976
},
{
"epoch": 0.55,
"learning_rate": 0.00022512537612838515,
"loss": 2.4447,
"theoretical_loss": 3.456188696298761,
"tokens_seen": 1829568512
},
{
"epoch": 0.55,
"learning_rate": 0.00022511534603811436,
"loss": 2.8453,
"theoretical_loss": 3.4561781924997756,
"tokens_seen": 1829634048
},
{
"epoch": 0.55,
"learning_rate": 0.00022510531594784354,
"loss": 2.6116,
"theoretical_loss": 3.456167689182363,
"tokens_seen": 1829699584
},
{
"epoch": 0.55,
"learning_rate": 0.00022509528585757272,
"loss": 2.9146,
"theoretical_loss": 3.456157186346484,
"tokens_seen": 1829765120
},
{
"epoch": 0.55,
"learning_rate": 0.0002250852557673019,
"loss": 2.7263,
"theoretical_loss": 3.4561466839921,
"tokens_seen": 1829830656
},
{
"epoch": 0.55,
"learning_rate": 0.00022507522567703108,
"loss": 2.5308,
"theoretical_loss": 3.4561361821191703,
"tokens_seen": 1829896192
},
{
"epoch": 0.55,
"learning_rate": 0.0002250651955867603,
"loss": 2.7385,
"theoretical_loss": 3.456125680727656,
"tokens_seen": 1829961728
},
{
"epoch": 0.55,
"learning_rate": 0.00022505516549648947,
"loss": 2.7679,
"theoretical_loss": 3.4561151798175187,
"tokens_seen": 1830027264
},
{
"epoch": 0.55,
"objective/train/docs_used": 1034491,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3891546726226807,
"objective/train/theoretical_loss": 3.4561046793887185,
"objective/train/tokens_used": 1850552800,
"theoretical_loss": 3.4561046793887185,
"tokens_seen": 1830092800
},
{
"epoch": 0.55,
"learning_rate": 0.00022504513540621865,
"loss": 2.849,
"theoretical_loss": 3.4561046793887185,
"tokens_seen": 1830092800
},
{
"epoch": 0.55,
"learning_rate": 0.00022503510531594783,
"loss": 2.7077,
"theoretical_loss": 3.456094179441216,
"tokens_seen": 1830158336
},
{
"epoch": 0.55,
"learning_rate": 0.00022502507522567704,
"loss": 2.7279,
"theoretical_loss": 3.456083679974972,
"tokens_seen": 1830223872
},
{
"epoch": 0.55,
"learning_rate": 0.00022501504513540622,
"loss": 2.8012,
"theoretical_loss": 3.456073180989947,
"tokens_seen": 1830289408
},
{
"epoch": 0.55,
"learning_rate": 0.0002250050150451354,
"loss": 2.5902,
"theoretical_loss": 3.456062682486102,
"tokens_seen": 1830354944
},
{
"epoch": 0.55,
"learning_rate": 0.00022499498495486458,
"loss": 2.5902,
"theoretical_loss": 3.4560521844633976,
"tokens_seen": 1830420480
},
{
"epoch": 0.55,
"learning_rate": 0.00022498495486459377,
"loss": 2.7499,
"theoretical_loss": 3.456041686921795,
"tokens_seen": 1830486016
},
{
"epoch": 0.55,
"learning_rate": 0.00022497492477432297,
"loss": 2.6154,
"theoretical_loss": 3.456031189861254,
"tokens_seen": 1830551552
},
{
"epoch": 0.55,
"learning_rate": 0.00022496489468405215,
"loss": 2.7783,
"theoretical_loss": 3.4560206932817357,
"tokens_seen": 1830617088
},
{
"epoch": 0.55,
"learning_rate": 0.00022495486459378134,
"loss": 2.5841,
"theoretical_loss": 3.4560101971832013,
"tokens_seen": 1830682624
},
{
"epoch": 0.55,
"learning_rate": 0.00022494483450351052,
"loss": 2.7085,
"theoretical_loss": 3.455999701565611,
"tokens_seen": 1830748160
},
{
"epoch": 0.55,
"learning_rate": 0.00022493480441323973,
"loss": 2.9216,
"theoretical_loss": 3.455989206428926,
"tokens_seen": 1830813696
},
{
"epoch": 0.55,
"learning_rate": 0.0002249247743229689,
"loss": 2.6977,
"theoretical_loss": 3.455978711773107,
"tokens_seen": 1830879232
},
{
"epoch": 0.55,
"learning_rate": 0.0002249147442326981,
"loss": 2.4173,
"theoretical_loss": 3.455968217598114,
"tokens_seen": 1830944768
},
{
"epoch": 0.55,
"learning_rate": 0.00022490471414242727,
"loss": 2.5563,
"theoretical_loss": 3.455957723903909,
"tokens_seen": 1831010304
},
{
"epoch": 0.55,
"learning_rate": 0.00022489468405215645,
"loss": 2.5997,
"theoretical_loss": 3.4559472306904517,
"tokens_seen": 1831075840
},
{
"epoch": 0.55,
"learning_rate": 0.00022488465396188566,
"loss": 2.7765,
"theoretical_loss": 3.4559367379577033,
"tokens_seen": 1831141376
},
{
"epoch": 0.55,
"learning_rate": 0.00022487462387161484,
"loss": 2.7139,
"theoretical_loss": 3.4559262457056246,
"tokens_seen": 1831206912
},
{
"epoch": 0.55,
"learning_rate": 0.00022486459378134402,
"loss": 2.7153,
"theoretical_loss": 3.4559157539341765,
"tokens_seen": 1831272448
},
{
"epoch": 0.55,
"learning_rate": 0.0002248545636910732,
"loss": 2.5196,
"theoretical_loss": 3.455905262643319,
"tokens_seen": 1831337984
},
{
"epoch": 0.55,
"learning_rate": 0.00022484453360080244,
"loss": 2.572,
"theoretical_loss": 3.4558947718330137,
"tokens_seen": 1831403520
},
{
"epoch": 0.55,
"learning_rate": 0.00022483450351053162,
"loss": 2.6303,
"theoretical_loss": 3.4558842815032214,
"tokens_seen": 1831469056
},
{
"epoch": 0.56,
"learning_rate": 0.0002248244734202608,
"loss": 2.4854,
"theoretical_loss": 3.455873791653903,
"tokens_seen": 1831534592
},
{
"epoch": 0.56,
"learning_rate": 0.00022481444332998998,
"loss": 2.7866,
"theoretical_loss": 3.4558633022850183,
"tokens_seen": 1831600128
},
{
"epoch": 0.56,
"learning_rate": 0.00022480441323971916,
"loss": 2.6226,
"theoretical_loss": 3.455852813396529,
"tokens_seen": 1831665664
},
{
"epoch": 0.56,
"objective/train/docs_used": 1035908,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2606523036956787,
"objective/train/theoretical_loss": 3.455842324988396,
"objective/train/tokens_used": 1852191200,
"theoretical_loss": 3.455842324988396,
"tokens_seen": 1831731200
},
{
"epoch": 0.56,
"learning_rate": 0.00022479438314944837,
"loss": 2.5822,
"theoretical_loss": 3.455842324988396,
"tokens_seen": 1831731200
},
{
"epoch": 0.56,
"learning_rate": 0.00022478435305917755,
"loss": 2.9024,
"theoretical_loss": 3.455831837060579,
"tokens_seen": 1831796736
},
{
"epoch": 0.56,
"learning_rate": 0.00022477432296890673,
"loss": 2.5809,
"theoretical_loss": 3.45582134961304,
"tokens_seen": 1831862272
},
{
"epoch": 0.56,
"learning_rate": 0.0002247642928786359,
"loss": 2.7395,
"theoretical_loss": 3.4558108626457402,
"tokens_seen": 1831927808
},
{
"epoch": 0.56,
"learning_rate": 0.00022475426278836512,
"loss": 2.8107,
"theoretical_loss": 3.455800376158639,
"tokens_seen": 1831993344
},
{
"epoch": 0.56,
"learning_rate": 0.0002247442326980943,
"loss": 2.6409,
"theoretical_loss": 3.4557898901516984,
"tokens_seen": 1832058880
},
{
"epoch": 0.56,
"learning_rate": 0.00022473420260782348,
"loss": 2.6311,
"theoretical_loss": 3.455779404624878,
"tokens_seen": 1832124416
},
{
"epoch": 0.56,
"learning_rate": 0.00022472417251755266,
"loss": 2.7774,
"theoretical_loss": 3.45576891957814,
"tokens_seen": 1832189952
},
{
"epoch": 0.56,
"learning_rate": 0.00022471414242728187,
"loss": 2.9757,
"theoretical_loss": 3.4557584350114445,
"tokens_seen": 1832255488
},
{
"epoch": 0.56,
"learning_rate": 0.00022470411233701105,
"loss": 2.7146,
"theoretical_loss": 3.4557479509247524,
"tokens_seen": 1832321024
},
{
"epoch": 0.56,
"learning_rate": 0.00022469408224674023,
"loss": 2.7883,
"theoretical_loss": 3.455737467318025,
"tokens_seen": 1832386560
},
{
"epoch": 0.56,
"learning_rate": 0.00022468405215646942,
"loss": 2.7964,
"theoretical_loss": 3.4557269841912226,
"tokens_seen": 1832452096
},
{
"epoch": 0.56,
"learning_rate": 0.0002246740220661986,
"loss": 2.6574,
"theoretical_loss": 3.4557165015443063,
"tokens_seen": 1832517632
},
{
"epoch": 0.56,
"learning_rate": 0.0002246639919759278,
"loss": 2.6693,
"theoretical_loss": 3.455706019377237,
"tokens_seen": 1832583168
},
{
"epoch": 0.56,
"learning_rate": 0.00022465396188565699,
"loss": 2.5414,
"theoretical_loss": 3.4556955376899756,
"tokens_seen": 1832648704
},
{
"epoch": 0.56,
"learning_rate": 0.00022464393179538617,
"loss": 2.666,
"theoretical_loss": 3.4556850564824826,
"tokens_seen": 1832714240
},
{
"epoch": 0.56,
"learning_rate": 0.00022463390170511535,
"loss": 2.3765,
"theoretical_loss": 3.4556745757547196,
"tokens_seen": 1832779776
},
{
"epoch": 0.56,
"learning_rate": 0.00022462387161484456,
"loss": 2.8034,
"theoretical_loss": 3.455664095506647,
"tokens_seen": 1832845312
},
{
"epoch": 0.56,
"learning_rate": 0.00022461384152457374,
"loss": 2.6777,
"theoretical_loss": 3.4556536157382256,
"tokens_seen": 1832910848
},
{
"epoch": 0.56,
"learning_rate": 0.00022460381143430292,
"loss": 2.8371,
"theoretical_loss": 3.455643136449417,
"tokens_seen": 1832976384
},
{
"epoch": 0.56,
"learning_rate": 0.0002245937813440321,
"loss": 2.5181,
"theoretical_loss": 3.455632657640181,
"tokens_seen": 1833041920
},
{
"epoch": 0.56,
"learning_rate": 0.00022458375125376128,
"loss": 2.5611,
"theoretical_loss": 3.4556221793104793,
"tokens_seen": 1833107456
},
{
"epoch": 0.56,
"learning_rate": 0.0002245737211634905,
"loss": 2.7099,
"theoretical_loss": 3.4556117014602727,
"tokens_seen": 1833172992
},
{
"epoch": 0.56,
"learning_rate": 0.00022456369107321967,
"loss": 2.7971,
"theoretical_loss": 3.455601224089522,
"tokens_seen": 1833238528
},
{
"epoch": 0.56,
"learning_rate": 0.00022455366098294885,
"loss": 2.632,
"theoretical_loss": 3.4555907471981886,
"tokens_seen": 1833304064
},
{
"epoch": 0.56,
"objective/train/docs_used": 1036404,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5784828662872314,
"objective/train/theoretical_loss": 3.455580270786232,
"objective/train/tokens_used": 1853829600,
"theoretical_loss": 3.455580270786232,
"tokens_seen": 1833369600
},
{
"epoch": 0.56,
"learning_rate": 0.00022454363089267803,
"loss": 2.5076,
"theoretical_loss": 3.455580270786232,
"tokens_seen": 1833369600
},
{
"epoch": 0.56,
"learning_rate": 0.00022453360080240724,
"loss": 2.6544,
"theoretical_loss": 3.455569794853615,
"tokens_seen": 1833435136
},
{
"epoch": 0.56,
"learning_rate": 0.00022452357071213642,
"loss": 2.6801,
"theoretical_loss": 3.4555593194002974,
"tokens_seen": 1833500672
},
{
"epoch": 0.56,
"learning_rate": 0.0002245135406218656,
"loss": 2.3346,
"theoretical_loss": 3.4555488444262403,
"tokens_seen": 1833566208
},
{
"epoch": 0.56,
"learning_rate": 0.00022450351053159478,
"loss": 2.5302,
"theoretical_loss": 3.4555383699314044,
"tokens_seen": 1833631744
},
{
"epoch": 0.56,
"learning_rate": 0.00022449348044132397,
"loss": 2.8549,
"theoretical_loss": 3.4555278959157514,
"tokens_seen": 1833697280
},
{
"epoch": 0.56,
"learning_rate": 0.00022448345035105317,
"loss": 2.7642,
"theoretical_loss": 3.455517422379242,
"tokens_seen": 1833762816
},
{
"epoch": 0.56,
"learning_rate": 0.00022447342026078236,
"loss": 2.5668,
"theoretical_loss": 3.455506949321837,
"tokens_seen": 1833828352
},
{
"epoch": 0.56,
"learning_rate": 0.00022446339017051154,
"loss": 2.8718,
"theoretical_loss": 3.455496476743497,
"tokens_seen": 1833893888
},
{
"epoch": 0.56,
"learning_rate": 0.00022445336008024072,
"loss": 2.8968,
"theoretical_loss": 3.4554860046441833,
"tokens_seen": 1833959424
},
{
"epoch": 0.56,
"learning_rate": 0.00022444332998996993,
"loss": 2.7655,
"theoretical_loss": 3.455475533023857,
"tokens_seen": 1834024960
},
{
"epoch": 0.56,
"learning_rate": 0.0002244332998996991,
"loss": 2.8947,
"theoretical_loss": 3.455465061882479,
"tokens_seen": 1834090496
},
{
"epoch": 0.56,
"learning_rate": 0.0002244232698094283,
"loss": 2.54,
"theoretical_loss": 3.4554545912200103,
"tokens_seen": 1834156032
},
{
"epoch": 0.56,
"learning_rate": 0.00022441323971915747,
"loss": 2.8814,
"theoretical_loss": 3.455444121036412,
"tokens_seen": 1834221568
},
{
"epoch": 0.56,
"learning_rate": 0.00022440320962888665,
"loss": 2.73,
"theoretical_loss": 3.455433651331645,
"tokens_seen": 1834287104
},
{
"epoch": 0.56,
"learning_rate": 0.00022439317953861586,
"loss": 2.5932,
"theoretical_loss": 3.45542318210567,
"tokens_seen": 1834352640
},
{
"epoch": 0.56,
"learning_rate": 0.00022438314944834504,
"loss": 2.729,
"theoretical_loss": 3.455412713358448,
"tokens_seen": 1834418176
},
{
"epoch": 0.56,
"learning_rate": 0.00022437311935807422,
"loss": 2.6028,
"theoretical_loss": 3.45540224508994,
"tokens_seen": 1834483712
},
{
"epoch": 0.56,
"learning_rate": 0.0002243630892678034,
"loss": 2.6228,
"theoretical_loss": 3.455391777300108,
"tokens_seen": 1834549248
},
{
"epoch": 0.56,
"learning_rate": 0.0002243530591775326,
"loss": 2.9009,
"theoretical_loss": 3.455381309988912,
"tokens_seen": 1834614784
},
{
"epoch": 0.56,
"learning_rate": 0.0002243430290872618,
"loss": 2.7279,
"theoretical_loss": 3.455370843156313,
"tokens_seen": 1834680320
},
{
"epoch": 0.56,
"learning_rate": 0.00022433299899699097,
"loss": 2.6002,
"theoretical_loss": 3.4553603768022727,
"tokens_seen": 1834745856
},
{
"epoch": 0.56,
"learning_rate": 0.00022432296890672015,
"loss": 2.7441,
"theoretical_loss": 3.4553499109267514,
"tokens_seen": 1834811392
},
{
"epoch": 0.56,
"learning_rate": 0.00022431293881644933,
"loss": 2.8284,
"theoretical_loss": 3.4553394455297104,
"tokens_seen": 1834876928
},
{
"epoch": 0.56,
"learning_rate": 0.00022430290872617854,
"loss": 2.3706,
"theoretical_loss": 3.455328980611111,
"tokens_seen": 1834942464
},
{
"epoch": 0.56,
"objective/train/docs_used": 1037817,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7909493446350098,
"objective/train/theoretical_loss": 3.455318516170914,
"objective/train/tokens_used": 1855468000,
"theoretical_loss": 3.455318516170914,
"tokens_seen": 1835008000
},
{
"epoch": 0.56,
"learning_rate": 0.00022429287863590772,
"loss": 2.7204,
"theoretical_loss": 3.455318516170914,
"tokens_seen": 1835008000
},
{
"epoch": 0.56,
"learning_rate": 0.0002242828485456369,
"loss": 2.4041,
"theoretical_loss": 3.4553080522090807,
"tokens_seen": 1835073536
},
{
"epoch": 0.56,
"learning_rate": 0.00022427281845536609,
"loss": 2.7923,
"theoretical_loss": 3.4552975887255717,
"tokens_seen": 1835139072
},
{
"epoch": 0.56,
"learning_rate": 0.0002242627883650953,
"loss": 2.6856,
"theoretical_loss": 3.455287125720348,
"tokens_seen": 1835204608
},
{
"epoch": 0.56,
"learning_rate": 0.00022425275827482448,
"loss": 2.4638,
"theoretical_loss": 3.455276663193372,
"tokens_seen": 1835270144
},
{
"epoch": 0.56,
"learning_rate": 0.00022424272818455366,
"loss": 2.6906,
"theoretical_loss": 3.4552662011446023,
"tokens_seen": 1835335680
},
{
"epoch": 0.56,
"learning_rate": 0.00022423269809428284,
"loss": 2.551,
"theoretical_loss": 3.4552557395740022,
"tokens_seen": 1835401216
},
{
"epoch": 0.56,
"learning_rate": 0.00022422266800401202,
"loss": 2.5844,
"theoretical_loss": 3.4552452784815317,
"tokens_seen": 1835466752
},
{
"epoch": 0.56,
"learning_rate": 0.00022421263791374123,
"loss": 2.773,
"theoretical_loss": 3.455234817867152,
"tokens_seen": 1835532288
},
{
"epoch": 0.56,
"learning_rate": 0.0002242026078234704,
"loss": 2.6156,
"theoretical_loss": 3.455224357730825,
"tokens_seen": 1835597824
},
{
"epoch": 0.56,
"learning_rate": 0.0002241925777331996,
"loss": 2.7709,
"theoretical_loss": 3.4552138980725102,
"tokens_seen": 1835663360
},
{
"epoch": 0.56,
"learning_rate": 0.00022418254764292877,
"loss": 2.8822,
"theoretical_loss": 3.4552034388921697,
"tokens_seen": 1835728896
},
{
"epoch": 0.56,
"learning_rate": 0.00022417251755265798,
"loss": 2.8095,
"theoretical_loss": 3.4551929801897647,
"tokens_seen": 1835794432
},
{
"epoch": 0.56,
"learning_rate": 0.00022416248746238716,
"loss": 2.6756,
"theoretical_loss": 3.455182521965256,
"tokens_seen": 1835859968
},
{
"epoch": 0.56,
"learning_rate": 0.00022415245737211634,
"loss": 2.7853,
"theoretical_loss": 3.455172064218605,
"tokens_seen": 1835925504
},
{
"epoch": 0.56,
"learning_rate": 0.00022414242728184552,
"loss": 2.6897,
"theoretical_loss": 3.4551616069497726,
"tokens_seen": 1835991040
},
{
"epoch": 0.56,
"learning_rate": 0.0002241323971915747,
"loss": 2.5902,
"theoretical_loss": 3.4551511501587195,
"tokens_seen": 1836056576
},
{
"epoch": 0.56,
"learning_rate": 0.0002241223671013039,
"loss": 2.9129,
"theoretical_loss": 3.4551406938454075,
"tokens_seen": 1836122112
},
{
"epoch": 0.56,
"learning_rate": 0.0002241123370110331,
"loss": 2.7974,
"theoretical_loss": 3.4551302380097972,
"tokens_seen": 1836187648
},
{
"epoch": 0.56,
"learning_rate": 0.00022410230692076227,
"loss": 2.8983,
"theoretical_loss": 3.4551197826518507,
"tokens_seen": 1836253184
},
{
"epoch": 0.56,
"learning_rate": 0.00022409227683049148,
"loss": 2.7825,
"theoretical_loss": 3.4551093277715275,
"tokens_seen": 1836318720
},
{
"epoch": 0.56,
"learning_rate": 0.0002240822467402207,
"loss": 2.7056,
"theoretical_loss": 3.45509887336879,
"tokens_seen": 1836384256
},
{
"epoch": 0.56,
"learning_rate": 0.00022407221664994987,
"loss": 2.8742,
"theoretical_loss": 3.455088419443599,
"tokens_seen": 1836449792
},
{
"epoch": 0.56,
"learning_rate": 0.00022406218655967905,
"loss": 2.8192,
"theoretical_loss": 3.4550779659959154,
"tokens_seen": 1836515328
},
{
"epoch": 0.56,
"learning_rate": 0.00022405215646940823,
"loss": 2.6948,
"theoretical_loss": 3.455067513025701,
"tokens_seen": 1836580864
},
{
"epoch": 0.56,
"objective/train/docs_used": 1038484,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8945260047912598,
"objective/train/theoretical_loss": 3.4550570605329165,
"objective/train/tokens_used": 1857106400,
"theoretical_loss": 3.4550570605329165,
"tokens_seen": 1836646400
},
{
"epoch": 0.56,
"learning_rate": 0.00022404212637913744,
"loss": 2.7636,
"theoretical_loss": 3.4550570605329165,
"tokens_seen": 1836646400
},
{
"epoch": 0.56,
"learning_rate": 0.00022403209628886662,
"loss": 2.7902,
"theoretical_loss": 3.455046608517523,
"tokens_seen": 1836711936
},
{
"epoch": 0.56,
"learning_rate": 0.0002240220661985958,
"loss": 2.5589,
"theoretical_loss": 3.4550361569794816,
"tokens_seen": 1836777472
},
{
"epoch": 0.56,
"learning_rate": 0.00022401203610832498,
"loss": 2.9416,
"theoretical_loss": 3.4550257059187537,
"tokens_seen": 1836843008
},
{
"epoch": 0.56,
"learning_rate": 0.00022400200601805417,
"loss": 2.3183,
"theoretical_loss": 3.4550152553353,
"tokens_seen": 1836908544
},
{
"epoch": 0.56,
"learning_rate": 0.00022399197592778337,
"loss": 2.7695,
"theoretical_loss": 3.4550048052290823,
"tokens_seen": 1836974080
},
{
"epoch": 0.56,
"learning_rate": 0.00022398194583751256,
"loss": 2.4918,
"theoretical_loss": 3.454994355600062,
"tokens_seen": 1837039616
},
{
"epoch": 0.56,
"learning_rate": 0.00022397191574724174,
"loss": 2.4971,
"theoretical_loss": 3.4549839064481995,
"tokens_seen": 1837105152
},
{
"epoch": 0.56,
"learning_rate": 0.00022396188565697092,
"loss": 2.6941,
"theoretical_loss": 3.4549734577734563,
"tokens_seen": 1837170688
},
{
"epoch": 0.56,
"learning_rate": 0.00022395185556670013,
"loss": 2.6882,
"theoretical_loss": 3.4549630095757937,
"tokens_seen": 1837236224
},
{
"epoch": 0.56,
"learning_rate": 0.0002239418254764293,
"loss": 2.7519,
"theoretical_loss": 3.4549525618551726,
"tokens_seen": 1837301760
},
{
"epoch": 0.56,
"learning_rate": 0.0002239317953861585,
"loss": 2.6934,
"theoretical_loss": 3.454942114611555,
"tokens_seen": 1837367296
},
{
"epoch": 0.56,
"learning_rate": 0.00022392176529588767,
"loss": 2.8406,
"theoretical_loss": 3.454931667844901,
"tokens_seen": 1837432832
},
{
"epoch": 0.56,
"learning_rate": 0.00022391173520561685,
"loss": 2.8072,
"theoretical_loss": 3.4549212215551726,
"tokens_seen": 1837498368
},
{
"epoch": 0.56,
"learning_rate": 0.00022390170511534606,
"loss": 2.8801,
"theoretical_loss": 3.4549107757423307,
"tokens_seen": 1837563904
},
{
"epoch": 0.56,
"learning_rate": 0.00022389167502507524,
"loss": 2.5573,
"theoretical_loss": 3.4549003304063364,
"tokens_seen": 1837629440
},
{
"epoch": 0.56,
"learning_rate": 0.00022388164493480442,
"loss": 2.8535,
"theoretical_loss": 3.4548898855471517,
"tokens_seen": 1837694976
},
{
"epoch": 0.56,
"learning_rate": 0.0002238716148445336,
"loss": 2.5822,
"theoretical_loss": 3.4548794411647368,
"tokens_seen": 1837760512
},
{
"epoch": 0.56,
"learning_rate": 0.0002238615847542628,
"loss": 2.7988,
"theoretical_loss": 3.4548689972590534,
"tokens_seen": 1837826048
},
{
"epoch": 0.56,
"learning_rate": 0.000223851554663992,
"loss": 2.6345,
"theoretical_loss": 3.4548585538300625,
"tokens_seen": 1837891584
},
{
"epoch": 0.56,
"learning_rate": 0.00022384152457372117,
"loss": 2.7044,
"theoretical_loss": 3.454848110877726,
"tokens_seen": 1837957120
},
{
"epoch": 0.56,
"learning_rate": 0.00022383149448345035,
"loss": 2.6498,
"theoretical_loss": 3.4548376684020043,
"tokens_seen": 1838022656
},
{
"epoch": 0.56,
"learning_rate": 0.00022382146439317953,
"loss": 2.335,
"theoretical_loss": 3.45482722640286,
"tokens_seen": 1838088192
},
{
"epoch": 0.56,
"learning_rate": 0.00022381143430290874,
"loss": 2.8653,
"theoretical_loss": 3.4548167848802525,
"tokens_seen": 1838153728
},
{
"epoch": 0.56,
"learning_rate": 0.00022380140421263792,
"loss": 2.4866,
"theoretical_loss": 3.454806343834144,
"tokens_seen": 1838219264
},
{
"epoch": 0.56,
"objective/train/docs_used": 1039769,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0204288959503174,
"objective/train/theoretical_loss": 3.454795903264496,
"objective/train/tokens_used": 1858744800,
"theoretical_loss": 3.454795903264496,
"tokens_seen": 1838284800
},
{
"epoch": 0.56,
"learning_rate": 0.0002237913741223671,
"loss": 2.706,
"theoretical_loss": 3.454795903264496,
"tokens_seen": 1838284800
},
{
"epoch": 0.56,
"learning_rate": 0.00022378134403209629,
"loss": 2.7008,
"theoretical_loss": 3.4547854631712696,
"tokens_seen": 1838350336
},
{
"epoch": 0.56,
"learning_rate": 0.0002237713139418255,
"loss": 2.4991,
"theoretical_loss": 3.454775023554426,
"tokens_seen": 1838415872
},
{
"epoch": 0.56,
"learning_rate": 0.00022376128385155468,
"loss": 2.5658,
"theoretical_loss": 3.454764584413927,
"tokens_seen": 1838481408
},
{
"epoch": 0.56,
"learning_rate": 0.00022375125376128386,
"loss": 2.5685,
"theoretical_loss": 3.4547541457497326,
"tokens_seen": 1838546944
},
{
"epoch": 0.56,
"learning_rate": 0.00022374122367101304,
"loss": 2.6944,
"theoretical_loss": 3.4547437075618053,
"tokens_seen": 1838612480
},
{
"epoch": 0.56,
"learning_rate": 0.00022373119358074222,
"loss": 2.6532,
"theoretical_loss": 3.454733269850106,
"tokens_seen": 1838678016
},
{
"epoch": 0.56,
"learning_rate": 0.00022372116349047143,
"loss": 2.6395,
"theoretical_loss": 3.4547228326145953,
"tokens_seen": 1838743552
},
{
"epoch": 0.56,
"learning_rate": 0.0002237111334002006,
"loss": 2.6411,
"theoretical_loss": 3.454712395855236,
"tokens_seen": 1838809088
},
{
"epoch": 0.56,
"learning_rate": 0.0002237011033099298,
"loss": 2.7521,
"theoretical_loss": 3.454701959571988,
"tokens_seen": 1838874624
},
{
"epoch": 0.56,
"learning_rate": 0.00022369107321965897,
"loss": 2.9125,
"theoretical_loss": 3.4546915237648137,
"tokens_seen": 1838940160
},
{
"epoch": 0.56,
"learning_rate": 0.00022368104312938818,
"loss": 2.615,
"theoretical_loss": 3.4546810884336736,
"tokens_seen": 1839005696
},
{
"epoch": 0.56,
"learning_rate": 0.00022367101303911736,
"loss": 2.609,
"theoretical_loss": 3.454670653578529,
"tokens_seen": 1839071232
},
{
"epoch": 0.56,
"learning_rate": 0.00022366098294884654,
"loss": 2.69,
"theoretical_loss": 3.4546602191993427,
"tokens_seen": 1839136768
},
{
"epoch": 0.56,
"learning_rate": 0.00022365095285857572,
"loss": 2.4407,
"theoretical_loss": 3.4546497852960742,
"tokens_seen": 1839202304
},
{
"epoch": 0.56,
"learning_rate": 0.0002236409227683049,
"loss": 2.7314,
"theoretical_loss": 3.4546393518686855,
"tokens_seen": 1839267840
},
{
"epoch": 0.56,
"learning_rate": 0.0002236308926780341,
"loss": 2.594,
"theoretical_loss": 3.4546289189171375,
"tokens_seen": 1839333376
},
{
"epoch": 0.56,
"learning_rate": 0.0002236208625877633,
"loss": 2.8213,
"theoretical_loss": 3.454618486441393,
"tokens_seen": 1839398912
},
{
"epoch": 0.56,
"learning_rate": 0.00022361083249749247,
"loss": 2.812,
"theoretical_loss": 3.4546080544414117,
"tokens_seen": 1839464448
},
{
"epoch": 0.56,
"learning_rate": 0.00022360080240722165,
"loss": 2.768,
"theoretical_loss": 3.4545976229171558,
"tokens_seen": 1839529984
},
{
"epoch": 0.56,
"learning_rate": 0.00022359077231695086,
"loss": 2.5775,
"theoretical_loss": 3.454587191868587,
"tokens_seen": 1839595520
},
{
"epoch": 0.56,
"learning_rate": 0.00022358074222668004,
"loss": 2.8522,
"theoretical_loss": 3.454576761295665,
"tokens_seen": 1839661056
},
{
"epoch": 0.56,
"learning_rate": 0.00022357071213640923,
"loss": 2.7306,
"theoretical_loss": 3.454566331198353,
"tokens_seen": 1839726592
},
{
"epoch": 0.56,
"learning_rate": 0.0002235606820461384,
"loss": 2.5497,
"theoretical_loss": 3.4545559015766116,
"tokens_seen": 1839792128
},
{
"epoch": 0.56,
"learning_rate": 0.00022355065195586761,
"loss": 2.7484,
"theoretical_loss": 3.4545454724304028,
"tokens_seen": 1839857664
},
{
"epoch": 0.56,
"objective/train/docs_used": 1040587,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7994730472564697,
"objective/train/theoretical_loss": 3.4545350437596865,
"objective/train/tokens_used": 1860383200,
"theoretical_loss": 3.4545350437596865,
"tokens_seen": 1839923200
},
{
"epoch": 0.56,
"learning_rate": 0.0002235406218655968,
"loss": 2.612,
"theoretical_loss": 3.4545350437596865,
"tokens_seen": 1839923200
},
{
"epoch": 0.56,
"learning_rate": 0.00022353059177532598,
"loss": 2.7495,
"theoretical_loss": 3.4545246155644254,
"tokens_seen": 1839988736
},
{
"epoch": 0.56,
"learning_rate": 0.00022352056168505516,
"loss": 2.4619,
"theoretical_loss": 3.4545141878445804,
"tokens_seen": 1840054272
},
{
"epoch": 0.56,
"learning_rate": 0.00022351053159478434,
"loss": 2.6253,
"theoretical_loss": 3.4545037606001134,
"tokens_seen": 1840119808
},
{
"epoch": 0.56,
"learning_rate": 0.00022350050150451355,
"loss": 2.8355,
"theoretical_loss": 3.4544933338309853,
"tokens_seen": 1840185344
},
{
"epoch": 0.56,
"learning_rate": 0.00022349047141424273,
"loss": 2.6094,
"theoretical_loss": 3.4544829075371575,
"tokens_seen": 1840250880
},
{
"epoch": 0.56,
"learning_rate": 0.0002234804413239719,
"loss": 2.5742,
"theoretical_loss": 3.4544724817185912,
"tokens_seen": 1840316416
},
{
"epoch": 0.56,
"learning_rate": 0.0002234704112337011,
"loss": 2.7647,
"theoretical_loss": 3.454462056375249,
"tokens_seen": 1840381952
},
{
"epoch": 0.56,
"learning_rate": 0.0002234603811434303,
"loss": 2.6601,
"theoretical_loss": 3.454451631507091,
"tokens_seen": 1840447488
},
{
"epoch": 0.56,
"learning_rate": 0.00022345035105315948,
"loss": 2.6614,
"theoretical_loss": 3.454441207114079,
"tokens_seen": 1840513024
},
{
"epoch": 0.56,
"learning_rate": 0.00022344032096288866,
"loss": 2.6522,
"theoretical_loss": 3.454430783196174,
"tokens_seen": 1840578560
},
{
"epoch": 0.56,
"learning_rate": 0.00022343029087261784,
"loss": 2.7687,
"theoretical_loss": 3.4544203597533385,
"tokens_seen": 1840644096
},
{
"epoch": 0.56,
"learning_rate": 0.00022342026078234702,
"loss": 2.5701,
"theoretical_loss": 3.454409936785533,
"tokens_seen": 1840709632
},
{
"epoch": 0.56,
"learning_rate": 0.00022341023069207623,
"loss": 2.6765,
"theoretical_loss": 3.45439951429272,
"tokens_seen": 1840775168
},
{
"epoch": 0.56,
"learning_rate": 0.0002234002006018054,
"loss": 2.6345,
"theoretical_loss": 3.4543890922748597,
"tokens_seen": 1840840704
},
{
"epoch": 0.56,
"learning_rate": 0.0002233901705115346,
"loss": 2.5124,
"theoretical_loss": 3.454378670731914,
"tokens_seen": 1840906240
},
{
"epoch": 0.56,
"learning_rate": 0.00022338014042126377,
"loss": 2.4154,
"theoretical_loss": 3.454368249663845,
"tokens_seen": 1840971776
},
{
"epoch": 0.56,
"learning_rate": 0.00022337011033099298,
"loss": 2.8341,
"theoretical_loss": 3.454357829070613,
"tokens_seen": 1841037312
},
{
"epoch": 0.56,
"learning_rate": 0.00022336008024072216,
"loss": 2.6048,
"theoretical_loss": 3.4543474089521804,
"tokens_seen": 1841102848
},
{
"epoch": 0.56,
"learning_rate": 0.00022335005015045135,
"loss": 2.8153,
"theoretical_loss": 3.4543369893085085,
"tokens_seen": 1841168384
},
{
"epoch": 0.56,
"learning_rate": 0.00022334002006018055,
"loss": 2.5361,
"theoretical_loss": 3.4543265701395582,
"tokens_seen": 1841233920
},
{
"epoch": 0.56,
"learning_rate": 0.00022332998996990973,
"loss": 2.7725,
"theoretical_loss": 3.4543161514452914,
"tokens_seen": 1841299456
},
{
"epoch": 0.56,
"learning_rate": 0.00022331995987963894,
"loss": 2.6524,
"theoretical_loss": 3.45430573322567,
"tokens_seen": 1841364992
},
{
"epoch": 0.56,
"learning_rate": 0.00022330992978936812,
"loss": 2.4462,
"theoretical_loss": 3.4542953154806546,
"tokens_seen": 1841430528
},
{
"epoch": 0.56,
"learning_rate": 0.0002232998996990973,
"loss": 2.7212,
"theoretical_loss": 3.4542848982102075,
"tokens_seen": 1841496064
},
{
"epoch": 0.56,
"objective/train/docs_used": 1041999,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3773317337036133,
"objective/train/theoretical_loss": 3.4542744814142896,
"objective/train/tokens_used": 1862021600,
"theoretical_loss": 3.4542744814142896,
"tokens_seen": 1841561600
},
{
"epoch": 0.56,
"learning_rate": 0.00022328986960882649,
"loss": 2.5731,
"theoretical_loss": 3.4542744814142896,
"tokens_seen": 1841561600
},
{
"epoch": 0.56,
"learning_rate": 0.0002232798395185557,
"loss": 2.8762,
"theoretical_loss": 3.454264065092863,
"tokens_seen": 1841627136
},
{
"epoch": 0.56,
"learning_rate": 0.00022326980942828488,
"loss": 2.7963,
"theoretical_loss": 3.454253649245888,
"tokens_seen": 1841692672
},
{
"epoch": 0.56,
"learning_rate": 0.00022325977933801406,
"loss": 2.6016,
"theoretical_loss": 3.4542432338733278,
"tokens_seen": 1841758208
},
{
"epoch": 0.56,
"learning_rate": 0.00022324974924774324,
"loss": 2.9082,
"theoretical_loss": 3.4542328189751426,
"tokens_seen": 1841823744
},
{
"epoch": 0.56,
"learning_rate": 0.00022323971915747242,
"loss": 2.6955,
"theoretical_loss": 3.4542224045512944,
"tokens_seen": 1841889280
},
{
"epoch": 0.56,
"learning_rate": 0.00022322968906720163,
"loss": 2.7354,
"theoretical_loss": 3.4542119906017446,
"tokens_seen": 1841954816
},
{
"epoch": 0.56,
"learning_rate": 0.0002232196589769308,
"loss": 2.7934,
"theoretical_loss": 3.4542015771264554,
"tokens_seen": 1842020352
},
{
"epoch": 0.56,
"learning_rate": 0.00022320962888666,
"loss": 2.6589,
"theoretical_loss": 3.454191164125387,
"tokens_seen": 1842085888
},
{
"epoch": 0.56,
"learning_rate": 0.00022319959879638917,
"loss": 2.4161,
"theoretical_loss": 3.4541807515985017,
"tokens_seen": 1842151424
},
{
"epoch": 0.56,
"learning_rate": 0.00022318956870611838,
"loss": 2.7133,
"theoretical_loss": 3.4541703395457617,
"tokens_seen": 1842216960
},
{
"epoch": 0.56,
"learning_rate": 0.00022317953861584756,
"loss": 2.6969,
"theoretical_loss": 3.4541599279671273,
"tokens_seen": 1842282496
},
{
"epoch": 0.56,
"learning_rate": 0.00022316950852557674,
"loss": 2.6709,
"theoretical_loss": 3.454149516862561,
"tokens_seen": 1842348032
},
{
"epoch": 0.56,
"learning_rate": 0.00022315947843530592,
"loss": 2.6857,
"theoretical_loss": 3.4541391062320237,
"tokens_seen": 1842413568
},
{
"epoch": 0.56,
"learning_rate": 0.0002231494483450351,
"loss": 2.7832,
"theoretical_loss": 3.454128696075477,
"tokens_seen": 1842479104
},
{
"epoch": 0.56,
"learning_rate": 0.0002231394182547643,
"loss": 2.8219,
"theoretical_loss": 3.454118286392883,
"tokens_seen": 1842544640
},
{
"epoch": 0.56,
"learning_rate": 0.0002231293881644935,
"loss": 2.5613,
"theoretical_loss": 3.454107877184203,
"tokens_seen": 1842610176
},
{
"epoch": 0.56,
"learning_rate": 0.00022311935807422267,
"loss": 2.6879,
"theoretical_loss": 3.4540974684493984,
"tokens_seen": 1842675712
},
{
"epoch": 0.56,
"learning_rate": 0.00022310932798395185,
"loss": 2.6532,
"theoretical_loss": 3.4540870601884306,
"tokens_seen": 1842741248
},
{
"epoch": 0.56,
"learning_rate": 0.00022309929789368106,
"loss": 2.6729,
"theoretical_loss": 3.4540766524012616,
"tokens_seen": 1842806784
},
{
"epoch": 0.56,
"learning_rate": 0.00022308926780341024,
"loss": 2.6681,
"theoretical_loss": 3.454066245087853,
"tokens_seen": 1842872320
},
{
"epoch": 0.56,
"learning_rate": 0.00022307923771313943,
"loss": 2.5534,
"theoretical_loss": 3.4540558382481663,
"tokens_seen": 1842937856
},
{
"epoch": 0.56,
"learning_rate": 0.0002230692076228686,
"loss": 2.8371,
"theoretical_loss": 3.4540454318821627,
"tokens_seen": 1843003392
},
{
"epoch": 0.56,
"learning_rate": 0.00022305917753259781,
"loss": 2.7557,
"theoretical_loss": 3.4540350259898043,
"tokens_seen": 1843068928
},
{
"epoch": 0.56,
"learning_rate": 0.000223049147442327,
"loss": 2.6824,
"theoretical_loss": 3.4540246205710528,
"tokens_seen": 1843134464
},
{
"epoch": 0.56,
"objective/train/docs_used": 1042423,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6538867950439453,
"objective/train/theoretical_loss": 3.454014215625869,
"objective/train/tokens_used": 1863660000,
"theoretical_loss": 3.454014215625869,
"tokens_seen": 1843200000
},
{
"epoch": 0.56,
"learning_rate": 0.00022303911735205618,
"loss": 2.6575,
"theoretical_loss": 3.454014215625869,
"tokens_seen": 1843200000
},
{
"epoch": 0.56,
"learning_rate": 0.00022302908726178536,
"loss": 2.8512,
"theoretical_loss": 3.4540038111542155,
"tokens_seen": 1843265536
},
{
"epoch": 0.56,
"learning_rate": 0.00022301905717151454,
"loss": 2.4685,
"theoretical_loss": 3.4539934071560534,
"tokens_seen": 1843331072
},
{
"epoch": 0.56,
"learning_rate": 0.00022300902708124375,
"loss": 2.4748,
"theoretical_loss": 3.453983003631344,
"tokens_seen": 1843396608
},
{
"epoch": 0.56,
"learning_rate": 0.00022299899699097293,
"loss": 2.7047,
"theoretical_loss": 3.45397260058005,
"tokens_seen": 1843462144
},
{
"epoch": 0.56,
"learning_rate": 0.0002229889669007021,
"loss": 2.6817,
"theoretical_loss": 3.453962198002132,
"tokens_seen": 1843527680
},
{
"epoch": 0.56,
"learning_rate": 0.0002229789368104313,
"loss": 2.7557,
"theoretical_loss": 3.4539517958975514,
"tokens_seen": 1843593216
},
{
"epoch": 0.56,
"learning_rate": 0.0002229689067201605,
"loss": 2.6906,
"theoretical_loss": 3.4539413942662707,
"tokens_seen": 1843658752
},
{
"epoch": 0.56,
"learning_rate": 0.00022295887662988968,
"loss": 2.6047,
"theoretical_loss": 3.4539309931082514,
"tokens_seen": 1843724288
},
{
"epoch": 0.56,
"learning_rate": 0.00022294884653961886,
"loss": 2.7369,
"theoretical_loss": 3.4539205924234553,
"tokens_seen": 1843789824
},
{
"epoch": 0.56,
"learning_rate": 0.00022293881644934804,
"loss": 2.375,
"theoretical_loss": 3.4539101922118434,
"tokens_seen": 1843855360
},
{
"epoch": 0.56,
"learning_rate": 0.00022292878635907722,
"loss": 2.6811,
"theoretical_loss": 3.4538997924733774,
"tokens_seen": 1843920896
},
{
"epoch": 0.56,
"learning_rate": 0.00022291875626880643,
"loss": 2.678,
"theoretical_loss": 3.4538893932080197,
"tokens_seen": 1843986432
},
{
"epoch": 0.56,
"learning_rate": 0.0002229087261785356,
"loss": 2.7019,
"theoretical_loss": 3.4538789944157315,
"tokens_seen": 1844051968
},
{
"epoch": 0.56,
"learning_rate": 0.0002228986960882648,
"loss": 2.6854,
"theoretical_loss": 3.4538685960964743,
"tokens_seen": 1844117504
},
{
"epoch": 0.56,
"learning_rate": 0.00022288866599799397,
"loss": 2.7312,
"theoretical_loss": 3.45385819825021,
"tokens_seen": 1844183040
},
{
"epoch": 0.56,
"learning_rate": 0.00022287863590772318,
"loss": 2.5643,
"theoretical_loss": 3.4538478008769,
"tokens_seen": 1844248576
},
{
"epoch": 0.56,
"learning_rate": 0.00022286860581745236,
"loss": 2.7435,
"theoretical_loss": 3.4538374039765065,
"tokens_seen": 1844314112
},
{
"epoch": 0.56,
"learning_rate": 0.00022285857572718155,
"loss": 2.7006,
"theoretical_loss": 3.4538270075489907,
"tokens_seen": 1844379648
},
{
"epoch": 0.56,
"learning_rate": 0.00022284854563691073,
"loss": 2.8479,
"theoretical_loss": 3.4538166115943145,
"tokens_seen": 1844445184
},
{
"epoch": 0.56,
"learning_rate": 0.0002228385155466399,
"loss": 2.607,
"theoretical_loss": 3.45380621611244,
"tokens_seen": 1844510720
},
{
"epoch": 0.56,
"learning_rate": 0.00022282848545636912,
"loss": 2.8793,
"theoretical_loss": 3.453795821103328,
"tokens_seen": 1844576256
},
{
"epoch": 0.56,
"learning_rate": 0.0002228184553660983,
"loss": 2.7633,
"theoretical_loss": 3.4537854265669408,
"tokens_seen": 1844641792
},
{
"epoch": 0.56,
"learning_rate": 0.00022280842527582748,
"loss": 2.5657,
"theoretical_loss": 3.4537750325032404,
"tokens_seen": 1844707328
},
{
"epoch": 0.56,
"learning_rate": 0.00022279839518555666,
"loss": 2.5877,
"theoretical_loss": 3.4537646389121877,
"tokens_seen": 1844772864
},
{
"epoch": 0.56,
"objective/train/docs_used": 1043833,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5953471660614014,
"objective/train/theoretical_loss": 3.4537542457937453,
"objective/train/tokens_used": 1865298400,
"theoretical_loss": 3.4537542457937453,
"tokens_seen": 1844838400
},
{
"epoch": 0.56,
"learning_rate": 0.00022278836509528587,
"loss": 2.7057,
"theoretical_loss": 3.4537542457937453,
"tokens_seen": 1844838400
},
{
"epoch": 0.56,
"learning_rate": 0.00022277833500501505,
"loss": 2.7583,
"theoretical_loss": 3.4537438531478735,
"tokens_seen": 1844903936
},
{
"epoch": 0.56,
"learning_rate": 0.00022276830491474423,
"loss": 2.6726,
"theoretical_loss": 3.4537334609745356,
"tokens_seen": 1844969472
},
{
"epoch": 0.56,
"learning_rate": 0.0002227582748244734,
"loss": 2.4224,
"theoretical_loss": 3.4537230692736927,
"tokens_seen": 1845035008
},
{
"epoch": 0.56,
"learning_rate": 0.0002227482447342026,
"loss": 2.543,
"theoretical_loss": 3.4537126780453065,
"tokens_seen": 1845100544
},
{
"epoch": 0.56,
"learning_rate": 0.0002227382146439318,
"loss": 2.6432,
"theoretical_loss": 3.453702287289339,
"tokens_seen": 1845166080
},
{
"epoch": 0.56,
"learning_rate": 0.00022272818455366098,
"loss": 2.7549,
"theoretical_loss": 3.4536918970057515,
"tokens_seen": 1845231616
},
{
"epoch": 0.56,
"learning_rate": 0.00022271815446339016,
"loss": 2.7982,
"theoretical_loss": 3.453681507194506,
"tokens_seen": 1845297152
},
{
"epoch": 0.56,
"learning_rate": 0.00022270812437311934,
"loss": 2.8075,
"theoretical_loss": 3.4536711178555644,
"tokens_seen": 1845362688
},
{
"epoch": 0.56,
"learning_rate": 0.00022269809428284855,
"loss": 2.6466,
"theoretical_loss": 3.453660728988888,
"tokens_seen": 1845428224
},
{
"epoch": 0.56,
"learning_rate": 0.00022268806419257773,
"loss": 2.9621,
"theoretical_loss": 3.453650340594439,
"tokens_seen": 1845493760
},
{
"epoch": 0.56,
"learning_rate": 0.00022267803410230691,
"loss": 2.8302,
"theoretical_loss": 3.453639952672179,
"tokens_seen": 1845559296
},
{
"epoch": 0.56,
"learning_rate": 0.0002226680040120361,
"loss": 2.7041,
"theoretical_loss": 3.4536295652220694,
"tokens_seen": 1845624832
},
{
"epoch": 0.56,
"learning_rate": 0.00022265797392176528,
"loss": 2.6486,
"theoretical_loss": 3.453619178244073,
"tokens_seen": 1845690368
},
{
"epoch": 0.56,
"learning_rate": 0.00022264794383149448,
"loss": 2.7241,
"theoretical_loss": 3.4536087917381506,
"tokens_seen": 1845755904
},
{
"epoch": 0.56,
"learning_rate": 0.00022263791374122367,
"loss": 2.4774,
"theoretical_loss": 3.4535984057042644,
"tokens_seen": 1845821440
},
{
"epoch": 0.56,
"learning_rate": 0.00022262788365095285,
"loss": 2.8701,
"theoretical_loss": 3.4535880201423765,
"tokens_seen": 1845886976
},
{
"epoch": 0.56,
"learning_rate": 0.00022261785356068203,
"loss": 2.6162,
"theoretical_loss": 3.4535776350524476,
"tokens_seen": 1845952512
},
{
"epoch": 0.56,
"learning_rate": 0.00022260782347041124,
"loss": 2.6393,
"theoretical_loss": 3.453567250434441,
"tokens_seen": 1846018048
},
{
"epoch": 0.56,
"learning_rate": 0.00022259779338014042,
"loss": 2.6163,
"theoretical_loss": 3.453556866288317,
"tokens_seen": 1846083584
},
{
"epoch": 0.56,
"learning_rate": 0.00022258776328986963,
"loss": 2.5316,
"theoretical_loss": 3.453546482614038,
"tokens_seen": 1846149120
},
{
"epoch": 0.56,
"learning_rate": 0.0002225777331995988,
"loss": 2.81,
"theoretical_loss": 3.4535360994115667,
"tokens_seen": 1846214656
},
{
"epoch": 0.56,
"learning_rate": 0.00022256770310932801,
"loss": 2.6259,
"theoretical_loss": 3.453525716680863,
"tokens_seen": 1846280192
},
{
"epoch": 0.56,
"learning_rate": 0.0002225576730190572,
"loss": 2.8541,
"theoretical_loss": 3.453515334421891,
"tokens_seen": 1846345728
},
{
"epoch": 0.56,
"learning_rate": 0.00022254764292878638,
"loss": 2.6477,
"theoretical_loss": 3.453504952634611,
"tokens_seen": 1846411264
},
{
"epoch": 0.56,
"objective/train/docs_used": 1044447,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.764051914215088,
"objective/train/theoretical_loss": 3.453494571318985,
"objective/train/tokens_used": 1866936800,
"theoretical_loss": 3.453494571318985,
"tokens_seen": 1846476800
},
{
"epoch": 0.56,
"learning_rate": 0.00022253761283851556,
"loss": 2.5833,
"theoretical_loss": 3.453494571318985,
"tokens_seen": 1846476800
},
{
"epoch": 0.56,
"learning_rate": 0.00022252758274824474,
"loss": 2.7661,
"theoretical_loss": 3.453484190474976,
"tokens_seen": 1846542336
},
{
"epoch": 0.56,
"learning_rate": 0.00022251755265797395,
"loss": 2.6866,
"theoretical_loss": 3.4534738101025435,
"tokens_seen": 1846607872
},
{
"epoch": 0.56,
"learning_rate": 0.00022250752256770313,
"loss": 2.5857,
"theoretical_loss": 3.4534634302016514,
"tokens_seen": 1846673408
},
{
"epoch": 0.56,
"learning_rate": 0.0002224974924774323,
"loss": 2.6434,
"theoretical_loss": 3.4534530507722607,
"tokens_seen": 1846738944
},
{
"epoch": 0.56,
"learning_rate": 0.0002224874623871615,
"loss": 2.711,
"theoretical_loss": 3.453442671814334,
"tokens_seen": 1846804480
},
{
"epoch": 0.56,
"learning_rate": 0.0002224774322968907,
"loss": 2.6722,
"theoretical_loss": 3.4534322933278316,
"tokens_seen": 1846870016
},
{
"epoch": 0.56,
"learning_rate": 0.00022246740220661988,
"loss": 2.7464,
"theoretical_loss": 3.453421915312717,
"tokens_seen": 1846935552
},
{
"epoch": 0.56,
"learning_rate": 0.00022245737211634906,
"loss": 2.5924,
"theoretical_loss": 3.453411537768951,
"tokens_seen": 1847001088
},
{
"epoch": 0.56,
"learning_rate": 0.00022244734202607824,
"loss": 2.6984,
"theoretical_loss": 3.4534011606964965,
"tokens_seen": 1847066624
},
{
"epoch": 0.56,
"learning_rate": 0.00022243731193580742,
"loss": 2.9715,
"theoretical_loss": 3.4533907840953146,
"tokens_seen": 1847132160
},
{
"epoch": 0.56,
"learning_rate": 0.00022242728184553663,
"loss": 2.6904,
"theoretical_loss": 3.453380407965367,
"tokens_seen": 1847197696
},
{
"epoch": 0.56,
"learning_rate": 0.0002224172517552658,
"loss": 2.4461,
"theoretical_loss": 3.4533700323066157,
"tokens_seen": 1847263232
},
{
"epoch": 0.56,
"learning_rate": 0.000222407221664995,
"loss": 2.6905,
"theoretical_loss": 3.4533596571190235,
"tokens_seen": 1847328768
},
{
"epoch": 0.56,
"learning_rate": 0.00022239719157472418,
"loss": 2.6273,
"theoretical_loss": 3.4533492824025513,
"tokens_seen": 1847394304
},
{
"epoch": 0.56,
"learning_rate": 0.00022238716148445338,
"loss": 2.7427,
"theoretical_loss": 3.453338908157161,
"tokens_seen": 1847459840
},
{
"epoch": 0.56,
"learning_rate": 0.00022237713139418256,
"loss": 2.6089,
"theoretical_loss": 3.4533285343828153,
"tokens_seen": 1847525376
},
{
"epoch": 0.56,
"learning_rate": 0.00022236710130391175,
"loss": 2.6183,
"theoretical_loss": 3.453318161079475,
"tokens_seen": 1847590912
},
{
"epoch": 0.56,
"learning_rate": 0.00022235707121364093,
"loss": 2.6254,
"theoretical_loss": 3.453307788247103,
"tokens_seen": 1847656448
},
{
"epoch": 0.56,
"learning_rate": 0.0002223470411233701,
"loss": 2.6802,
"theoretical_loss": 3.45329741588566,
"tokens_seen": 1847721984
},
{
"epoch": 0.56,
"learning_rate": 0.00022233701103309932,
"loss": 2.5594,
"theoretical_loss": 3.4532870439951098,
"tokens_seen": 1847787520
},
{
"epoch": 0.56,
"learning_rate": 0.0002223269809428285,
"loss": 2.6062,
"theoretical_loss": 3.453276672575413,
"tokens_seen": 1847853056
},
{
"epoch": 0.56,
"learning_rate": 0.00022231695085255768,
"loss": 2.7545,
"theoretical_loss": 3.4532663016265315,
"tokens_seen": 1847918592
},
{
"epoch": 0.56,
"learning_rate": 0.00022230692076228686,
"loss": 2.7301,
"theoretical_loss": 3.4532559311484277,
"tokens_seen": 1847984128
},
{
"epoch": 0.56,
"learning_rate": 0.00022229689067201607,
"loss": 2.7947,
"theoretical_loss": 3.4532455611410633,
"tokens_seen": 1848049664
},
{
"epoch": 0.56,
"objective/train/docs_used": 1045763,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8487067222595215,
"objective/train/theoretical_loss": 3.4532351916044,
"objective/train/tokens_used": 1868575200,
"theoretical_loss": 3.4532351916044,
"tokens_seen": 1848115200
},
{
"epoch": 0.56,
"learning_rate": 0.00022228686058174525,
"loss": 2.7269,
"theoretical_loss": 3.4532351916044,
"tokens_seen": 1848115200
},
{
"epoch": 0.56,
"learning_rate": 0.00022227683049147443,
"loss": 2.7555,
"theoretical_loss": 3.4532248225384006,
"tokens_seen": 1848180736
},
{
"epoch": 0.56,
"learning_rate": 0.0002222668004012036,
"loss": 2.6903,
"theoretical_loss": 3.453214453943026,
"tokens_seen": 1848246272
},
{
"epoch": 0.56,
"learning_rate": 0.0002222567703109328,
"loss": 2.6536,
"theoretical_loss": 3.4532040858182387,
"tokens_seen": 1848311808
},
{
"epoch": 0.56,
"learning_rate": 0.000222246740220662,
"loss": 2.6374,
"theoretical_loss": 3.453193718164001,
"tokens_seen": 1848377344
},
{
"epoch": 0.56,
"learning_rate": 0.00022223671013039118,
"loss": 2.6258,
"theoretical_loss": 3.4531833509802734,
"tokens_seen": 1848442880
},
{
"epoch": 0.56,
"learning_rate": 0.00022222668004012036,
"loss": 2.8551,
"theoretical_loss": 3.45317298426702,
"tokens_seen": 1848508416
},
{
"epoch": 0.56,
"learning_rate": 0.00022221664994984954,
"loss": 2.7154,
"theoretical_loss": 3.453162618024201,
"tokens_seen": 1848573952
},
{
"epoch": 0.56,
"learning_rate": 0.00022220661985957875,
"loss": 2.5289,
"theoretical_loss": 3.4531522522517797,
"tokens_seen": 1848639488
},
{
"epoch": 0.56,
"learning_rate": 0.00022219658976930793,
"loss": 2.6974,
"theoretical_loss": 3.453141886949717,
"tokens_seen": 1848705024
},
{
"epoch": 0.56,
"learning_rate": 0.00022218655967903711,
"loss": 2.666,
"theoretical_loss": 3.453131522117975,
"tokens_seen": 1848770560
},
{
"epoch": 0.56,
"learning_rate": 0.0002221765295887663,
"loss": 2.8357,
"theoretical_loss": 3.4531211577565166,
"tokens_seen": 1848836096
},
{
"epoch": 0.56,
"learning_rate": 0.00022216649949849548,
"loss": 2.6225,
"theoretical_loss": 3.453110793865303,
"tokens_seen": 1848901632
},
{
"epoch": 0.56,
"learning_rate": 0.00022215646940822468,
"loss": 2.3894,
"theoretical_loss": 3.4531004304442963,
"tokens_seen": 1848967168
},
{
"epoch": 0.56,
"learning_rate": 0.00022214643931795387,
"loss": 2.7841,
"theoretical_loss": 3.4530900674934584,
"tokens_seen": 1849032704
},
{
"epoch": 0.56,
"learning_rate": 0.00022213640922768305,
"loss": 2.6044,
"theoretical_loss": 3.453079705012752,
"tokens_seen": 1849098240
},
{
"epoch": 0.56,
"learning_rate": 0.00022212637913741223,
"loss": 2.6419,
"theoretical_loss": 3.4530693430021384,
"tokens_seen": 1849163776
},
{
"epoch": 0.56,
"learning_rate": 0.00022211634904714144,
"loss": 2.6215,
"theoretical_loss": 3.45305898146158,
"tokens_seen": 1849229312
},
{
"epoch": 0.56,
"learning_rate": 0.00022210631895687062,
"loss": 2.8906,
"theoretical_loss": 3.453048620391038,
"tokens_seen": 1849294848
},
{
"epoch": 0.56,
"learning_rate": 0.0002220962888665998,
"loss": 2.6584,
"theoretical_loss": 3.4530382597904756,
"tokens_seen": 1849360384
},
{
"epoch": 0.56,
"learning_rate": 0.00022208625877632898,
"loss": 2.6608,
"theoretical_loss": 3.4530278996598547,
"tokens_seen": 1849425920
},
{
"epoch": 0.56,
"learning_rate": 0.00022207622868605816,
"loss": 2.7369,
"theoretical_loss": 3.453017539999136,
"tokens_seen": 1849491456
},
{
"epoch": 0.56,
"learning_rate": 0.00022206619859578737,
"loss": 2.7363,
"theoretical_loss": 3.4530071808082834,
"tokens_seen": 1849556992
},
{
"epoch": 0.56,
"learning_rate": 0.00022205616850551655,
"loss": 2.4442,
"theoretical_loss": 3.4529968220872576,
"tokens_seen": 1849622528
},
{
"epoch": 0.56,
"learning_rate": 0.00022204613841524573,
"loss": 2.5643,
"theoretical_loss": 3.4529864638360213,
"tokens_seen": 1849688064
},
{
"epoch": 0.56,
"objective/train/docs_used": 1046397,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9440348148345947,
"objective/train/theoretical_loss": 3.452976106054536,
"objective/train/tokens_used": 1870213600,
"theoretical_loss": 3.452976106054536,
"tokens_seen": 1849753600
},
{
"epoch": 0.56,
"learning_rate": 0.0002220361083249749,
"loss": 2.723,
"theoretical_loss": 3.452976106054536,
"tokens_seen": 1849753600
},
{
"epoch": 0.56,
"learning_rate": 0.00022202607823470412,
"loss": 2.5828,
"theoretical_loss": 3.4529657487427636,
"tokens_seen": 1849819136
},
{
"epoch": 0.56,
"learning_rate": 0.0002220160481444333,
"loss": 2.5225,
"theoretical_loss": 3.4529553919006672,
"tokens_seen": 1849884672
},
{
"epoch": 0.56,
"learning_rate": 0.00022200601805416248,
"loss": 2.5904,
"theoretical_loss": 3.4529450355282085,
"tokens_seen": 1849950208
},
{
"epoch": 0.56,
"learning_rate": 0.00022199598796389166,
"loss": 2.8245,
"theoretical_loss": 3.452934679625349,
"tokens_seen": 1850015744
},
{
"epoch": 0.56,
"learning_rate": 0.00022198595787362084,
"loss": 2.6086,
"theoretical_loss": 3.4529243241920513,
"tokens_seen": 1850081280
},
{
"epoch": 0.56,
"learning_rate": 0.00022197592778335005,
"loss": 2.7035,
"theoretical_loss": 3.452913969228277,
"tokens_seen": 1850146816
},
{
"epoch": 0.56,
"learning_rate": 0.00022196589769307923,
"loss": 2.6919,
"theoretical_loss": 3.4529036147339887,
"tokens_seen": 1850212352
},
{
"epoch": 0.56,
"learning_rate": 0.00022195586760280842,
"loss": 2.656,
"theoretical_loss": 3.452893260709148,
"tokens_seen": 1850277888
},
{
"epoch": 0.56,
"learning_rate": 0.0002219458375125376,
"loss": 2.6005,
"theoretical_loss": 3.4528829071537177,
"tokens_seen": 1850343424
},
{
"epoch": 0.56,
"learning_rate": 0.0002219358074222668,
"loss": 2.6159,
"theoretical_loss": 3.4528725540676595,
"tokens_seen": 1850408960
},
{
"epoch": 0.56,
"learning_rate": 0.00022192577733199599,
"loss": 2.716,
"theoretical_loss": 3.452862201450935,
"tokens_seen": 1850474496
},
{
"epoch": 0.56,
"learning_rate": 0.00022191574724172517,
"loss": 2.6993,
"theoretical_loss": 3.452851849303507,
"tokens_seen": 1850540032
},
{
"epoch": 0.56,
"learning_rate": 0.00022190571715145435,
"loss": 2.7779,
"theoretical_loss": 3.452841497625337,
"tokens_seen": 1850605568
},
{
"epoch": 0.56,
"learning_rate": 0.00022189568706118356,
"loss": 2.7629,
"theoretical_loss": 3.452831146416388,
"tokens_seen": 1850671104
},
{
"epoch": 0.56,
"learning_rate": 0.00022188565697091274,
"loss": 2.5457,
"theoretical_loss": 3.4528207956766215,
"tokens_seen": 1850736640
},
{
"epoch": 0.56,
"learning_rate": 0.00022187562688064192,
"loss": 2.6808,
"theoretical_loss": 3.4528104454059996,
"tokens_seen": 1850802176
},
{
"epoch": 0.56,
"learning_rate": 0.0002218655967903711,
"loss": 2.5825,
"theoretical_loss": 3.4528000956044846,
"tokens_seen": 1850867712
},
{
"epoch": 0.56,
"learning_rate": 0.00022185556670010028,
"loss": 2.6881,
"theoretical_loss": 3.4527897462720385,
"tokens_seen": 1850933248
},
{
"epoch": 0.56,
"learning_rate": 0.00022184553660982952,
"loss": 2.7054,
"theoretical_loss": 3.452779397408624,
"tokens_seen": 1850998784
},
{
"epoch": 0.56,
"learning_rate": 0.0002218355065195587,
"loss": 2.4226,
"theoretical_loss": 3.452769049014202,
"tokens_seen": 1851064320
},
{
"epoch": 0.56,
"learning_rate": 0.00022182547642928788,
"loss": 2.8454,
"theoretical_loss": 3.4527587010887357,
"tokens_seen": 1851129856
},
{
"epoch": 0.56,
"learning_rate": 0.00022181544633901706,
"loss": 2.6173,
"theoretical_loss": 3.4527483536321864,
"tokens_seen": 1851195392
},
{
"epoch": 0.56,
"learning_rate": 0.00022180541624874627,
"loss": 2.6084,
"theoretical_loss": 3.4527380066445175,
"tokens_seen": 1851260928
},
{
"epoch": 0.56,
"learning_rate": 0.00022179538615847545,
"loss": 2.4941,
"theoretical_loss": 3.4527276601256904,
"tokens_seen": 1851326464
},
{
"epoch": 0.56,
"objective/train/docs_used": 1047036,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0464601516723633,
"objective/train/theoretical_loss": 3.4527173140756666,
"objective/train/tokens_used": 1871852000,
"theoretical_loss": 3.4527173140756666,
"tokens_seen": 1851392000
},
{
"epoch": 0.56,
"learning_rate": 0.00022178535606820463,
"loss": 2.7621,
"theoretical_loss": 3.4527173140756666,
"tokens_seen": 1851392000
},
{
"epoch": 0.56,
"learning_rate": 0.0002217753259779338,
"loss": 2.5125,
"theoretical_loss": 3.4527069684944096,
"tokens_seen": 1851457536
},
{
"epoch": 0.56,
"learning_rate": 0.000221765295887663,
"loss": 2.4327,
"theoretical_loss": 3.4526966233818808,
"tokens_seen": 1851523072
},
{
"epoch": 0.56,
"learning_rate": 0.0002217552657973922,
"loss": 2.5794,
"theoretical_loss": 3.4526862787380423,
"tokens_seen": 1851588608
},
{
"epoch": 0.56,
"learning_rate": 0.00022174523570712138,
"loss": 2.6596,
"theoretical_loss": 3.4526759345628566,
"tokens_seen": 1851654144
},
{
"epoch": 0.56,
"learning_rate": 0.00022173520561685056,
"loss": 2.7413,
"theoretical_loss": 3.452665590856286,
"tokens_seen": 1851719680
},
{
"epoch": 0.56,
"learning_rate": 0.00022172517552657974,
"loss": 2.7145,
"theoretical_loss": 3.452655247618292,
"tokens_seen": 1851785216
},
{
"epoch": 0.56,
"learning_rate": 0.00022171514543630895,
"loss": 2.5441,
"theoretical_loss": 3.452644904848838,
"tokens_seen": 1851850752
},
{
"epoch": 0.56,
"learning_rate": 0.00022170511534603813,
"loss": 2.6862,
"theoretical_loss": 3.4526345625478845,
"tokens_seen": 1851916288
},
{
"epoch": 0.56,
"learning_rate": 0.00022169508525576731,
"loss": 2.8407,
"theoretical_loss": 3.4526242207153954,
"tokens_seen": 1851981824
},
{
"epoch": 0.56,
"learning_rate": 0.0002216850551654965,
"loss": 2.6672,
"theoretical_loss": 3.4526138793513317,
"tokens_seen": 1852047360
},
{
"epoch": 0.56,
"learning_rate": 0.00022167502507522568,
"loss": 2.6199,
"theoretical_loss": 3.452603538455656,
"tokens_seen": 1852112896
},
{
"epoch": 0.56,
"learning_rate": 0.00022166499498495488,
"loss": 2.8354,
"theoretical_loss": 3.452593198028331,
"tokens_seen": 1852178432
},
{
"epoch": 0.56,
"learning_rate": 0.00022165496489468407,
"loss": 2.579,
"theoretical_loss": 3.4525828580693183,
"tokens_seen": 1852243968
},
{
"epoch": 0.56,
"learning_rate": 0.00022164493480441325,
"loss": 2.7259,
"theoretical_loss": 3.45257251857858,
"tokens_seen": 1852309504
},
{
"epoch": 0.56,
"learning_rate": 0.00022163490471414243,
"loss": 2.4799,
"theoretical_loss": 3.452562179556079,
"tokens_seen": 1852375040
},
{
"epoch": 0.56,
"learning_rate": 0.00022162487462387164,
"loss": 2.3783,
"theoretical_loss": 3.452551841001777,
"tokens_seen": 1852440576
},
{
"epoch": 0.56,
"learning_rate": 0.00022161484453360082,
"loss": 2.6217,
"theoretical_loss": 3.4525415029156363,
"tokens_seen": 1852506112
},
{
"epoch": 0.56,
"learning_rate": 0.00022160481444333,
"loss": 2.7146,
"theoretical_loss": 3.452531165297619,
"tokens_seen": 1852571648
},
{
"epoch": 0.56,
"learning_rate": 0.00022159478435305918,
"loss": 2.6759,
"theoretical_loss": 3.452520828147688,
"tokens_seen": 1852637184
},
{
"epoch": 0.56,
"learning_rate": 0.00022158475426278836,
"loss": 2.704,
"theoretical_loss": 3.452510491465805,
"tokens_seen": 1852702720
},
{
"epoch": 0.56,
"learning_rate": 0.00022157472417251757,
"loss": 2.493,
"theoretical_loss": 3.4525001552519323,
"tokens_seen": 1852768256
},
{
"epoch": 0.56,
"learning_rate": 0.00022156469408224675,
"loss": 2.6969,
"theoretical_loss": 3.4524898195060323,
"tokens_seen": 1852833792
},
{
"epoch": 0.56,
"learning_rate": 0.00022155466399197593,
"loss": 2.704,
"theoretical_loss": 3.4524794842280673,
"tokens_seen": 1852899328
},
{
"epoch": 0.56,
"learning_rate": 0.0002215446339017051,
"loss": 2.8103,
"theoretical_loss": 3.4524691494179995,
"tokens_seen": 1852964864
},
{
"epoch": 0.56,
"objective/train/docs_used": 1048172,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.369645595550537,
"objective/train/theoretical_loss": 3.452458815075791,
"objective/train/tokens_used": 1873490400,
"theoretical_loss": 3.452458815075791,
"tokens_seen": 1853030400
},
{
"epoch": 0.56,
"learning_rate": 0.00022153460381143432,
"loss": 2.3258,
"theoretical_loss": 3.452458815075791,
"tokens_seen": 1853030400
},
{
"epoch": 0.56,
"learning_rate": 0.0002215245737211635,
"loss": 2.698,
"theoretical_loss": 3.452448481201404,
"tokens_seen": 1853095936
},
{
"epoch": 0.56,
"learning_rate": 0.00022151454363089268,
"loss": 2.6235,
"theoretical_loss": 3.452438147794801,
"tokens_seen": 1853161472
},
{
"epoch": 0.56,
"learning_rate": 0.00022150451354062186,
"loss": 2.6674,
"theoretical_loss": 3.4524278148559446,
"tokens_seen": 1853227008
},
{
"epoch": 0.56,
"learning_rate": 0.00022149448345035105,
"loss": 2.6917,
"theoretical_loss": 3.4524174823847966,
"tokens_seen": 1853292544
},
{
"epoch": 0.56,
"learning_rate": 0.00022148445336008025,
"loss": 2.7929,
"theoretical_loss": 3.452407150381319,
"tokens_seen": 1853358080
},
{
"epoch": 0.56,
"learning_rate": 0.00022147442326980943,
"loss": 2.7743,
"theoretical_loss": 3.4523968188454752,
"tokens_seen": 1853423616
},
{
"epoch": 0.56,
"learning_rate": 0.00022146439317953862,
"loss": 2.4145,
"theoretical_loss": 3.452386487777227,
"tokens_seen": 1853489152
},
{
"epoch": 0.56,
"learning_rate": 0.0002214543630892678,
"loss": 2.8145,
"theoretical_loss": 3.452376157176536,
"tokens_seen": 1853554688
},
{
"epoch": 0.56,
"learning_rate": 0.000221444332998997,
"loss": 2.8773,
"theoretical_loss": 3.452365827043365,
"tokens_seen": 1853620224
},
{
"epoch": 0.56,
"learning_rate": 0.00022143430290872619,
"loss": 2.6333,
"theoretical_loss": 3.4523554973776767,
"tokens_seen": 1853685760
},
{
"epoch": 0.56,
"learning_rate": 0.00022142427281845537,
"loss": 2.5156,
"theoretical_loss": 3.452345168179433,
"tokens_seen": 1853751296
},
{
"epoch": 0.56,
"learning_rate": 0.00022141424272818455,
"loss": 2.5909,
"theoretical_loss": 3.4523348394485964,
"tokens_seen": 1853816832
},
{
"epoch": 0.56,
"learning_rate": 0.00022140421263791376,
"loss": 2.6676,
"theoretical_loss": 3.452324511185129,
"tokens_seen": 1853882368
},
{
"epoch": 0.56,
"learning_rate": 0.00022139418254764294,
"loss": 2.8603,
"theoretical_loss": 3.452314183388993,
"tokens_seen": 1853947904
},
{
"epoch": 0.56,
"learning_rate": 0.00022138415245737212,
"loss": 2.8564,
"theoretical_loss": 3.4523038560601513,
"tokens_seen": 1854013440
},
{
"epoch": 0.56,
"learning_rate": 0.0002213741223671013,
"loss": 2.4657,
"theoretical_loss": 3.452293529198566,
"tokens_seen": 1854078976
},
{
"epoch": 0.56,
"learning_rate": 0.00022136409227683048,
"loss": 2.5963,
"theoretical_loss": 3.4522832028041988,
"tokens_seen": 1854144512
},
{
"epoch": 0.56,
"learning_rate": 0.0002213540621865597,
"loss": 2.7837,
"theoretical_loss": 3.4522728768770135,
"tokens_seen": 1854210048
},
{
"epoch": 0.56,
"learning_rate": 0.00022134403209628887,
"loss": 2.6562,
"theoretical_loss": 3.452262551416971,
"tokens_seen": 1854275584
},
{
"epoch": 0.56,
"learning_rate": 0.00022133400200601805,
"loss": 2.9229,
"theoretical_loss": 3.4522522264240347,
"tokens_seen": 1854341120
},
{
"epoch": 0.56,
"learning_rate": 0.00022132397191574723,
"loss": 2.6772,
"theoretical_loss": 3.452241901898166,
"tokens_seen": 1854406656
},
{
"epoch": 0.56,
"learning_rate": 0.00022131394182547644,
"loss": 2.5544,
"theoretical_loss": 3.452231577839328,
"tokens_seen": 1854472192
},
{
"epoch": 0.56,
"learning_rate": 0.00022130391173520562,
"loss": 2.6049,
"theoretical_loss": 3.452221254247483,
"tokens_seen": 1854537728
},
{
"epoch": 0.56,
"learning_rate": 0.0002212938816449348,
"loss": 2.6442,
"theoretical_loss": 3.4522109311225933,
"tokens_seen": 1854603264
},
{
"epoch": 0.56,
"objective/train/docs_used": 1049323,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9285247325897217,
"objective/train/theoretical_loss": 3.4522006084646204,
"objective/train/tokens_used": 1875128800,
"theoretical_loss": 3.4522006084646204,
"tokens_seen": 1854668800
},
{
"epoch": 0.56,
"learning_rate": 0.00022128385155466398,
"loss": 2.8839,
"theoretical_loss": 3.4522006084646204,
"tokens_seen": 1854668800
},
{
"epoch": 0.56,
"learning_rate": 0.00022127382146439317,
"loss": 2.7335,
"theoretical_loss": 3.4521902862735283,
"tokens_seen": 1854734336
},
{
"epoch": 0.56,
"learning_rate": 0.00022126379137412237,
"loss": 2.7577,
"theoretical_loss": 3.4521799645492783,
"tokens_seen": 1854799872
},
{
"epoch": 0.56,
"learning_rate": 0.00022125376128385155,
"loss": 2.492,
"theoretical_loss": 3.452169643291833,
"tokens_seen": 1854865408
},
{
"epoch": 0.56,
"learning_rate": 0.00022124373119358074,
"loss": 2.5355,
"theoretical_loss": 3.452159322501155,
"tokens_seen": 1854930944
},
{
"epoch": 0.56,
"learning_rate": 0.00022123370110330992,
"loss": 2.5496,
"theoretical_loss": 3.4521490021772063,
"tokens_seen": 1854996480
},
{
"epoch": 0.56,
"learning_rate": 0.00022122367101303913,
"loss": 2.7079,
"theoretical_loss": 3.45213868231995,
"tokens_seen": 1855062016
},
{
"epoch": 0.56,
"learning_rate": 0.0002212136409227683,
"loss": 2.4927,
"theoretical_loss": 3.452128362929348,
"tokens_seen": 1855127552
},
{
"epoch": 0.56,
"learning_rate": 0.0002212036108324975,
"loss": 2.7722,
"theoretical_loss": 3.4521180440053625,
"tokens_seen": 1855193088
},
{
"epoch": 0.56,
"learning_rate": 0.00022119358074222667,
"loss": 2.6934,
"theoretical_loss": 3.452107725547956,
"tokens_seen": 1855258624
},
{
"epoch": 0.56,
"learning_rate": 0.00022118355065195585,
"loss": 2.5857,
"theoretical_loss": 3.4520974075570914,
"tokens_seen": 1855324160
},
{
"epoch": 0.56,
"learning_rate": 0.00022117352056168506,
"loss": 2.8036,
"theoretical_loss": 3.452087090032731,
"tokens_seen": 1855389696
},
{
"epoch": 0.56,
"learning_rate": 0.00022116349047141424,
"loss": 2.4284,
"theoretical_loss": 3.452076772974837,
"tokens_seen": 1855455232
},
{
"epoch": 0.56,
"learning_rate": 0.00022115346038114342,
"loss": 2.5939,
"theoretical_loss": 3.4520664563833723,
"tokens_seen": 1855520768
},
{
"epoch": 0.56,
"learning_rate": 0.0002211434302908726,
"loss": 2.7789,
"theoretical_loss": 3.452056140258298,
"tokens_seen": 1855586304
},
{
"epoch": 0.56,
"learning_rate": 0.0002211334002006018,
"loss": 2.8401,
"theoretical_loss": 3.4520458245995784,
"tokens_seen": 1855651840
},
{
"epoch": 0.56,
"learning_rate": 0.000221123370110331,
"loss": 2.7281,
"theoretical_loss": 3.4520355094071746,
"tokens_seen": 1855717376
},
{
"epoch": 0.56,
"learning_rate": 0.00022111334002006017,
"loss": 2.6031,
"theoretical_loss": 3.45202519468105,
"tokens_seen": 1855782912
},
{
"epoch": 0.56,
"learning_rate": 0.00022110330992978935,
"loss": 2.7838,
"theoretical_loss": 3.452014880421166,
"tokens_seen": 1855848448
},
{
"epoch": 0.56,
"learning_rate": 0.00022109327983951856,
"loss": 2.7669,
"theoretical_loss": 3.452004566627486,
"tokens_seen": 1855913984
},
{
"epoch": 0.56,
"learning_rate": 0.00022108324974924777,
"loss": 2.4421,
"theoretical_loss": 3.4519942532999712,
"tokens_seen": 1855979520
},
{
"epoch": 0.56,
"learning_rate": 0.00022107321965897695,
"loss": 2.6129,
"theoretical_loss": 3.451983940438586,
"tokens_seen": 1856045056
},
{
"epoch": 0.56,
"learning_rate": 0.00022106318956870613,
"loss": 2.7237,
"theoretical_loss": 3.4519736280432918,
"tokens_seen": 1856110592
},
{
"epoch": 0.56,
"learning_rate": 0.0002210531594784353,
"loss": 2.5836,
"theoretical_loss": 3.4519633161140506,
"tokens_seen": 1856176128
},
{
"epoch": 0.56,
"learning_rate": 0.00022104312938816452,
"loss": 2.5856,
"theoretical_loss": 3.4519530046508256,
"tokens_seen": 1856241664
},
{
"epoch": 0.56,
"objective/train/docs_used": 1049929,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.860739231109619,
"objective/train/theoretical_loss": 3.451942693653579,
"objective/train/tokens_used": 1876767200,
"theoretical_loss": 3.451942693653579,
"tokens_seen": 1856307200
},
{
"epoch": 0.56,
"learning_rate": 0.0002210330992978937,
"loss": 2.7941,
"theoretical_loss": 3.451942693653579,
"tokens_seen": 1856307200
},
{
"epoch": 0.56,
"learning_rate": 0.00022102306920762288,
"loss": 2.7826,
"theoretical_loss": 3.4519323831222737,
"tokens_seen": 1856372736
},
{
"epoch": 0.56,
"learning_rate": 0.00022101303911735206,
"loss": 2.7453,
"theoretical_loss": 3.4519220730568714,
"tokens_seen": 1856438272
},
{
"epoch": 0.56,
"learning_rate": 0.00022100300902708125,
"loss": 2.4343,
"theoretical_loss": 3.4519117634573355,
"tokens_seen": 1856503808
},
{
"epoch": 0.56,
"learning_rate": 0.00022099297893681045,
"loss": 2.5955,
"theoretical_loss": 3.451901454323628,
"tokens_seen": 1856569344
},
{
"epoch": 0.56,
"learning_rate": 0.00022098294884653963,
"loss": 2.5383,
"theoretical_loss": 3.4518911456557113,
"tokens_seen": 1856634880
},
{
"epoch": 0.56,
"learning_rate": 0.00022097291875626882,
"loss": 2.6791,
"theoretical_loss": 3.4518808374535483,
"tokens_seen": 1856700416
},
{
"epoch": 0.56,
"learning_rate": 0.000220962888665998,
"loss": 2.7089,
"theoretical_loss": 3.4518705297171013,
"tokens_seen": 1856765952
},
{
"epoch": 0.56,
"learning_rate": 0.0002209528585757272,
"loss": 2.7356,
"theoretical_loss": 3.4518602224463324,
"tokens_seen": 1856831488
},
{
"epoch": 0.56,
"learning_rate": 0.00022094282848545639,
"loss": 2.7376,
"theoretical_loss": 3.4518499156412052,
"tokens_seen": 1856897024
},
{
"epoch": 0.56,
"learning_rate": 0.00022093279839518557,
"loss": 2.831,
"theoretical_loss": 3.4518396093016817,
"tokens_seen": 1856962560
},
{
"epoch": 0.56,
"learning_rate": 0.00022092276830491475,
"loss": 2.6965,
"theoretical_loss": 3.451829303427724,
"tokens_seen": 1857028096
},
{
"epoch": 0.56,
"learning_rate": 0.00022091273821464396,
"loss": 2.3874,
"theoretical_loss": 3.451818998019295,
"tokens_seen": 1857093632
},
{
"epoch": 0.56,
"learning_rate": 0.00022090270812437314,
"loss": 2.6094,
"theoretical_loss": 3.451808693076357,
"tokens_seen": 1857159168
},
{
"epoch": 0.56,
"learning_rate": 0.00022089267803410232,
"loss": 2.7174,
"theoretical_loss": 3.4517983885988732,
"tokens_seen": 1857224704
},
{
"epoch": 0.56,
"learning_rate": 0.0002208826479438315,
"loss": 2.5549,
"theoretical_loss": 3.4517880845868056,
"tokens_seen": 1857290240
},
{
"epoch": 0.56,
"learning_rate": 0.00022087261785356068,
"loss": 2.831,
"theoretical_loss": 3.4517777810401165,
"tokens_seen": 1857355776
},
{
"epoch": 0.56,
"learning_rate": 0.0002208625877632899,
"loss": 2.5705,
"theoretical_loss": 3.4517674779587693,
"tokens_seen": 1857421312
},
{
"epoch": 0.56,
"learning_rate": 0.00022085255767301907,
"loss": 2.6188,
"theoretical_loss": 3.451757175342726,
"tokens_seen": 1857486848
},
{
"epoch": 0.56,
"learning_rate": 0.00022084252758274825,
"loss": 2.6715,
"theoretical_loss": 3.451746873191949,
"tokens_seen": 1857552384
},
{
"epoch": 0.56,
"learning_rate": 0.00022083249749247743,
"loss": 2.6822,
"theoretical_loss": 3.4517365715064017,
"tokens_seen": 1857617920
},
{
"epoch": 0.56,
"learning_rate": 0.00022082246740220664,
"loss": 2.8808,
"theoretical_loss": 3.451726270286046,
"tokens_seen": 1857683456
},
{
"epoch": 0.56,
"learning_rate": 0.00022081243731193582,
"loss": 2.6762,
"theoretical_loss": 3.451715969530844,
"tokens_seen": 1857748992
},
{
"epoch": 0.56,
"learning_rate": 0.000220802407221665,
"loss": 2.5584,
"theoretical_loss": 3.4517056692407597,
"tokens_seen": 1857814528
},
{
"epoch": 0.56,
"learning_rate": 0.00022079237713139418,
"loss": 2.4908,
"theoretical_loss": 3.451695369415755,
"tokens_seen": 1857880064
},
{
"epoch": 0.56,
"objective/train/docs_used": 1050597,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.4806439876556396,
"objective/train/theoretical_loss": 3.4516850700557917,
"objective/train/tokens_used": 1878405600,
"theoretical_loss": 3.4516850700557917,
"tokens_seen": 1857945600
},
{
"epoch": 0.56,
"learning_rate": 0.00022078234704112337,
"loss": 2.7129,
"theoretical_loss": 3.4516850700557917,
"tokens_seen": 1857945600
},
{
"epoch": 0.56,
"learning_rate": 0.00022077231695085257,
"loss": 2.7036,
"theoretical_loss": 3.451674771160833,
"tokens_seen": 1858011136
},
{
"epoch": 0.56,
"learning_rate": 0.00022076228686058175,
"loss": 2.7719,
"theoretical_loss": 3.451664472730842,
"tokens_seen": 1858076672
},
{
"epoch": 0.56,
"learning_rate": 0.00022075225677031094,
"loss": 2.5636,
"theoretical_loss": 3.451654174765781,
"tokens_seen": 1858142208
},
{
"epoch": 0.56,
"learning_rate": 0.00022074222668004012,
"loss": 2.812,
"theoretical_loss": 3.451643877265613,
"tokens_seen": 1858207744
},
{
"epoch": 0.56,
"learning_rate": 0.00022073219658976933,
"loss": 2.7165,
"theoretical_loss": 3.451633580230299,
"tokens_seen": 1858273280
},
{
"epoch": 0.56,
"learning_rate": 0.0002207221664994985,
"loss": 2.6035,
"theoretical_loss": 3.4516232836598033,
"tokens_seen": 1858338816
},
{
"epoch": 0.56,
"learning_rate": 0.0002207121364092277,
"loss": 2.654,
"theoretical_loss": 3.451612987554088,
"tokens_seen": 1858404352
},
{
"epoch": 0.56,
"learning_rate": 0.00022070210631895687,
"loss": 2.5037,
"theoretical_loss": 3.451602691913116,
"tokens_seen": 1858469888
},
{
"epoch": 0.56,
"learning_rate": 0.00022069207622868605,
"loss": 2.5497,
"theoretical_loss": 3.4515923967368494,
"tokens_seen": 1858535424
},
{
"epoch": 0.56,
"learning_rate": 0.00022068204613841526,
"loss": 2.468,
"theoretical_loss": 3.4515821020252506,
"tokens_seen": 1858600960
},
{
"epoch": 0.56,
"learning_rate": 0.00022067201604814444,
"loss": 2.5491,
"theoretical_loss": 3.4515718077782838,
"tokens_seen": 1858666496
},
{
"epoch": 0.56,
"learning_rate": 0.00022066198595787362,
"loss": 2.7516,
"theoretical_loss": 3.4515615139959097,
"tokens_seen": 1858732032
},
{
"epoch": 0.56,
"learning_rate": 0.0002206519558676028,
"loss": 2.7957,
"theoretical_loss": 3.451551220678092,
"tokens_seen": 1858797568
},
{
"epoch": 0.56,
"learning_rate": 0.000220641925777332,
"loss": 2.8106,
"theoretical_loss": 3.4515409278247935,
"tokens_seen": 1858863104
},
{
"epoch": 0.56,
"learning_rate": 0.0002206318956870612,
"loss": 2.4425,
"theoretical_loss": 3.4515306354359767,
"tokens_seen": 1858928640
},
{
"epoch": 0.56,
"learning_rate": 0.00022062186559679037,
"loss": 2.8055,
"theoretical_loss": 3.4515203435116035,
"tokens_seen": 1858994176
},
{
"epoch": 0.56,
"learning_rate": 0.00022061183550651955,
"loss": 2.8571,
"theoretical_loss": 3.4515100520516375,
"tokens_seen": 1859059712
},
{
"epoch": 0.56,
"learning_rate": 0.00022060180541624873,
"loss": 2.6798,
"theoretical_loss": 3.4514997610560414,
"tokens_seen": 1859125248
},
{
"epoch": 0.56,
"learning_rate": 0.00022059177532597794,
"loss": 2.6715,
"theoretical_loss": 3.451489470524777,
"tokens_seen": 1859190784
},
{
"epoch": 0.56,
"learning_rate": 0.00022058174523570712,
"loss": 2.6987,
"theoretical_loss": 3.4514791804578078,
"tokens_seen": 1859256320
},
{
"epoch": 0.56,
"learning_rate": 0.0002205717151454363,
"loss": 2.88,
"theoretical_loss": 3.4514688908550966,
"tokens_seen": 1859321856
},
{
"epoch": 0.56,
"learning_rate": 0.00022056168505516549,
"loss": 2.53,
"theoretical_loss": 3.451458601716605,
"tokens_seen": 1859387392
},
{
"epoch": 0.56,
"learning_rate": 0.0002205516549648947,
"loss": 2.8234,
"theoretical_loss": 3.451448313042297,
"tokens_seen": 1859452928
},
{
"epoch": 0.56,
"learning_rate": 0.00022054162487462387,
"loss": 2.8959,
"theoretical_loss": 3.451438024832134,
"tokens_seen": 1859518464
},
{
"epoch": 0.56,
"objective/train/docs_used": 1051049,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.410726547241211,
"objective/train/theoretical_loss": 3.45142773708608,
"objective/train/tokens_used": 1880044000,
"theoretical_loss": 3.45142773708608,
"tokens_seen": 1859584000
},
{
"epoch": 0.56,
"learning_rate": 0.00022053159478435306,
"loss": 2.5076,
"theoretical_loss": 3.45142773708608,
"tokens_seen": 1859584000
},
{
"epoch": 0.56,
"learning_rate": 0.00022052156469408224,
"loss": 2.6181,
"theoretical_loss": 3.451417449804097,
"tokens_seen": 1859649536
},
{
"epoch": 0.56,
"learning_rate": 0.00022051153460381142,
"loss": 2.6751,
"theoretical_loss": 3.451407162986148,
"tokens_seen": 1859715072
},
{
"epoch": 0.56,
"learning_rate": 0.00022050150451354063,
"loss": 2.5309,
"theoretical_loss": 3.451396876632195,
"tokens_seen": 1859780608
},
{
"epoch": 0.56,
"learning_rate": 0.0002204914744232698,
"loss": 2.4748,
"theoretical_loss": 3.4513865907422017,
"tokens_seen": 1859846144
},
{
"epoch": 0.56,
"learning_rate": 0.000220481444332999,
"loss": 2.6436,
"theoretical_loss": 3.4513763053161304,
"tokens_seen": 1859911680
},
{
"epoch": 0.56,
"learning_rate": 0.00022047141424272817,
"loss": 2.523,
"theoretical_loss": 3.4513660203539436,
"tokens_seen": 1859977216
},
{
"epoch": 0.56,
"learning_rate": 0.00022046138415245738,
"loss": 2.596,
"theoretical_loss": 3.4513557358556044,
"tokens_seen": 1860042752
},
{
"epoch": 0.56,
"learning_rate": 0.00022045135406218656,
"loss": 2.647,
"theoretical_loss": 3.4513454518210755,
"tokens_seen": 1860108288
},
{
"epoch": 0.56,
"learning_rate": 0.00022044132397191574,
"loss": 2.5282,
"theoretical_loss": 3.4513351682503193,
"tokens_seen": 1860173824
},
{
"epoch": 0.56,
"learning_rate": 0.00022043129388164492,
"loss": 2.6699,
"theoretical_loss": 3.4513248851432987,
"tokens_seen": 1860239360
},
{
"epoch": 0.56,
"learning_rate": 0.0002204212637913741,
"loss": 2.637,
"theoretical_loss": 3.451314602499977,
"tokens_seen": 1860304896
},
{
"epoch": 0.56,
"learning_rate": 0.0002204112337011033,
"loss": 2.7464,
"theoretical_loss": 3.451304320320316,
"tokens_seen": 1860370432
},
{
"epoch": 0.56,
"learning_rate": 0.0002204012036108325,
"loss": 2.7966,
"theoretical_loss": 3.451294038604279,
"tokens_seen": 1860435968
},
{
"epoch": 0.56,
"learning_rate": 0.00022039117352056167,
"loss": 2.4869,
"theoretical_loss": 3.4512837573518294,
"tokens_seen": 1860501504
},
{
"epoch": 0.56,
"learning_rate": 0.00022038114343029085,
"loss": 2.8064,
"theoretical_loss": 3.4512734765629283,
"tokens_seen": 1860567040
},
{
"epoch": 0.56,
"learning_rate": 0.00022037111334002006,
"loss": 2.4702,
"theoretical_loss": 3.45126319623754,
"tokens_seen": 1860632576
},
{
"epoch": 0.56,
"learning_rate": 0.00022036108324974924,
"loss": 2.7121,
"theoretical_loss": 3.451252916375627,
"tokens_seen": 1860698112
},
{
"epoch": 0.56,
"learning_rate": 0.00022035105315947842,
"loss": 2.7318,
"theoretical_loss": 3.4512426369771516,
"tokens_seen": 1860763648
},
{
"epoch": 0.56,
"learning_rate": 0.00022034102306920763,
"loss": 2.5609,
"theoretical_loss": 3.4512323580420765,
"tokens_seen": 1860829184
},
{
"epoch": 0.56,
"learning_rate": 0.00022033099297893684,
"loss": 2.7088,
"theoretical_loss": 3.451222079570365,
"tokens_seen": 1860894720
},
{
"epoch": 0.56,
"learning_rate": 0.00022032096288866602,
"loss": 2.6383,
"theoretical_loss": 3.4512118015619797,
"tokens_seen": 1860960256
},
{
"epoch": 0.56,
"learning_rate": 0.0002203109327983952,
"loss": 2.5622,
"theoretical_loss": 3.4512015240168834,
"tokens_seen": 1861025792
},
{
"epoch": 0.56,
"learning_rate": 0.00022030090270812438,
"loss": 2.812,
"theoretical_loss": 3.451191246935039,
"tokens_seen": 1861091328
},
{
"epoch": 0.56,
"learning_rate": 0.00022029087261785357,
"loss": 2.7721,
"theoretical_loss": 3.4511809703164094,
"tokens_seen": 1861156864
},
{
"epoch": 0.56,
"objective/train/docs_used": 1052031,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3999788761138916,
"objective/train/theoretical_loss": 3.4511706941609566,
"objective/train/tokens_used": 1881682400,
"theoretical_loss": 3.4511706941609566,
"tokens_seen": 1861222400
},
{
"epoch": 0.56,
"learning_rate": 0.00022028084252758277,
"loss": 2.6017,
"theoretical_loss": 3.4511706941609566,
"tokens_seen": 1861222400
},
{
"epoch": 0.56,
"learning_rate": 0.00022027081243731195,
"loss": 2.6344,
"theoretical_loss": 3.4511604184686444,
"tokens_seen": 1861287936
},
{
"epoch": 0.56,
"learning_rate": 0.00022026078234704114,
"loss": 2.7169,
"theoretical_loss": 3.4511501432394356,
"tokens_seen": 1861353472
},
{
"epoch": 0.56,
"learning_rate": 0.00022025075225677032,
"loss": 2.6026,
"theoretical_loss": 3.451139868473292,
"tokens_seen": 1861419008
},
{
"epoch": 0.56,
"learning_rate": 0.00022024072216649953,
"loss": 2.7406,
"theoretical_loss": 3.4511295941701774,
"tokens_seen": 1861484544
},
{
"epoch": 0.56,
"learning_rate": 0.0002202306920762287,
"loss": 2.8595,
"theoretical_loss": 3.4511193203300543,
"tokens_seen": 1861550080
},
{
"epoch": 0.56,
"learning_rate": 0.0002202206619859579,
"loss": 2.6373,
"theoretical_loss": 3.451109046952886,
"tokens_seen": 1861615616
},
{
"epoch": 0.56,
"learning_rate": 0.00022021063189568707,
"loss": 2.6711,
"theoretical_loss": 3.451098774038634,
"tokens_seen": 1861681152
},
{
"epoch": 0.56,
"learning_rate": 0.00022020060180541625,
"loss": 2.6204,
"theoretical_loss": 3.4510885015872628,
"tokens_seen": 1861746688
},
{
"epoch": 0.56,
"learning_rate": 0.00022019057171514546,
"loss": 2.6198,
"theoretical_loss": 3.4510782295987346,
"tokens_seen": 1861812224
},
{
"epoch": 0.56,
"learning_rate": 0.00022018054162487464,
"loss": 2.6662,
"theoretical_loss": 3.4510679580730117,
"tokens_seen": 1861877760
},
{
"epoch": 0.56,
"learning_rate": 0.00022017051153460382,
"loss": 2.8191,
"theoretical_loss": 3.4510576870100573,
"tokens_seen": 1861943296
},
{
"epoch": 0.56,
"learning_rate": 0.000220160481444333,
"loss": 2.6669,
"theoretical_loss": 3.4510474164098346,
"tokens_seen": 1862008832
},
{
"epoch": 0.56,
"learning_rate": 0.0002201504513540622,
"loss": 2.6887,
"theoretical_loss": 3.4510371462723066,
"tokens_seen": 1862074368
},
{
"epoch": 0.56,
"learning_rate": 0.0002201404212637914,
"loss": 2.6583,
"theoretical_loss": 3.4510268765974357,
"tokens_seen": 1862139904
},
{
"epoch": 0.56,
"learning_rate": 0.00022013039117352057,
"loss": 2.5998,
"theoretical_loss": 3.451016607385185,
"tokens_seen": 1862205440
},
{
"epoch": 0.56,
"learning_rate": 0.00022012036108324975,
"loss": 2.6056,
"theoretical_loss": 3.451006338635517,
"tokens_seen": 1862270976
},
{
"epoch": 0.56,
"learning_rate": 0.00022011033099297893,
"loss": 2.7994,
"theoretical_loss": 3.4509960703483946,
"tokens_seen": 1862336512
},
{
"epoch": 0.56,
"learning_rate": 0.00022010030090270814,
"loss": 2.5983,
"theoretical_loss": 3.450985802523782,
"tokens_seen": 1862402048
},
{
"epoch": 0.56,
"learning_rate": 0.00022009027081243732,
"loss": 2.5198,
"theoretical_loss": 3.45097553516164,
"tokens_seen": 1862467584
},
{
"epoch": 0.56,
"learning_rate": 0.0002200802407221665,
"loss": 2.7694,
"theoretical_loss": 3.4509652682619327,
"tokens_seen": 1862533120
},
{
"epoch": 0.56,
"learning_rate": 0.00022007021063189569,
"loss": 2.5644,
"theoretical_loss": 3.4509550018246236,
"tokens_seen": 1862598656
},
{
"epoch": 0.56,
"learning_rate": 0.0002200601805416249,
"loss": 2.7648,
"theoretical_loss": 3.4509447358496743,
"tokens_seen": 1862664192
},
{
"epoch": 0.56,
"learning_rate": 0.00022005015045135408,
"loss": 2.7848,
"theoretical_loss": 3.450934470337048,
"tokens_seen": 1862729728
},
{
"epoch": 0.56,
"learning_rate": 0.00022004012036108326,
"loss": 2.8084,
"theoretical_loss": 3.450924205286708,
"tokens_seen": 1862795264
},
{
"epoch": 0.56,
"objective/train/docs_used": 1052757,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6799139976501465,
"objective/train/theoretical_loss": 3.4509139406986176,
"objective/train/tokens_used": 1883320800,
"theoretical_loss": 3.4509139406986176,
"tokens_seen": 1862860800
},
{
"epoch": 0.56,
"learning_rate": 0.00022003009027081244,
"loss": 2.6831,
"theoretical_loss": 3.4509139406986176,
"tokens_seen": 1862860800
},
{
"epoch": 0.56,
"learning_rate": 0.00022002006018054162,
"loss": 2.5823,
"theoretical_loss": 3.450903676572739,
"tokens_seen": 1862926336
},
{
"epoch": 0.56,
"learning_rate": 0.00022001003009027083,
"loss": 2.6963,
"theoretical_loss": 3.4508934129090356,
"tokens_seen": 1862991872
},
{
"epoch": 0.56,
"learning_rate": 0.00022,
"loss": 2.8066,
"theoretical_loss": 3.4508831497074697,
"tokens_seen": 1863057408
},
{
"epoch": 0.56,
"learning_rate": 0.0002199899699097292,
"loss": 2.5454,
"theoretical_loss": 3.4508728869680043,
"tokens_seen": 1863122944
},
{
"epoch": 0.56,
"learning_rate": 0.00021997993981945837,
"loss": 2.5984,
"theoretical_loss": 3.450862624690603,
"tokens_seen": 1863188480
},
{
"epoch": 0.56,
"learning_rate": 0.00021996990972918758,
"loss": 2.8249,
"theoretical_loss": 3.450852362875229,
"tokens_seen": 1863254016
},
{
"epoch": 0.56,
"learning_rate": 0.00021995987963891676,
"loss": 2.7096,
"theoretical_loss": 3.450842101521844,
"tokens_seen": 1863319552
},
{
"epoch": 0.56,
"learning_rate": 0.00021994984954864594,
"loss": 2.6583,
"theoretical_loss": 3.4508318406304115,
"tokens_seen": 1863385088
},
{
"epoch": 0.56,
"learning_rate": 0.00021993981945837512,
"loss": 2.5253,
"theoretical_loss": 3.4508215802008952,
"tokens_seen": 1863450624
},
{
"epoch": 0.56,
"learning_rate": 0.0002199297893681043,
"loss": 2.594,
"theoretical_loss": 3.450811320233257,
"tokens_seen": 1863516160
},
{
"epoch": 0.56,
"learning_rate": 0.0002199197592778335,
"loss": 2.8765,
"theoretical_loss": 3.4508010607274606,
"tokens_seen": 1863581696
},
{
"epoch": 0.56,
"learning_rate": 0.0002199097291875627,
"loss": 2.5718,
"theoretical_loss": 3.4507908016834685,
"tokens_seen": 1863647232
},
{
"epoch": 0.56,
"learning_rate": 0.00021989969909729187,
"loss": 2.6626,
"theoretical_loss": 3.4507805431012435,
"tokens_seen": 1863712768
},
{
"epoch": 0.56,
"learning_rate": 0.00021988966900702105,
"loss": 2.558,
"theoretical_loss": 3.4507702849807496,
"tokens_seen": 1863778304
},
{
"epoch": 0.56,
"learning_rate": 0.00021987963891675026,
"loss": 2.55,
"theoretical_loss": 3.450760027321948,
"tokens_seen": 1863843840
},
{
"epoch": 0.56,
"learning_rate": 0.00021986960882647944,
"loss": 2.5156,
"theoretical_loss": 3.450749770124804,
"tokens_seen": 1863909376
},
{
"epoch": 0.56,
"learning_rate": 0.00021985957873620862,
"loss": 2.7998,
"theoretical_loss": 3.4507395133892786,
"tokens_seen": 1863974912
},
{
"epoch": 0.56,
"learning_rate": 0.0002198495486459378,
"loss": 2.3966,
"theoretical_loss": 3.450729257115336,
"tokens_seen": 1864040448
},
{
"epoch": 0.56,
"learning_rate": 0.00021983951855566701,
"loss": 2.6276,
"theoretical_loss": 3.4507190013029385,
"tokens_seen": 1864105984
},
{
"epoch": 0.56,
"learning_rate": 0.0002198294884653962,
"loss": 2.7039,
"theoretical_loss": 3.45070874595205,
"tokens_seen": 1864171520
},
{
"epoch": 0.56,
"learning_rate": 0.00021981945837512538,
"loss": 2.7382,
"theoretical_loss": 3.4506984910626324,
"tokens_seen": 1864237056
},
{
"epoch": 0.56,
"learning_rate": 0.00021980942828485456,
"loss": 2.7326,
"theoretical_loss": 3.450688236634649,
"tokens_seen": 1864302592
},
{
"epoch": 0.56,
"learning_rate": 0.00021979939819458374,
"loss": 2.7969,
"theoretical_loss": 3.450677982668063,
"tokens_seen": 1864368128
},
{
"epoch": 0.56,
"learning_rate": 0.00021978936810431295,
"loss": 2.5126,
"theoretical_loss": 3.450667729162838,
"tokens_seen": 1864433664
},
{
"epoch": 0.56,
"objective/train/docs_used": 1054143,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6187782287597656,
"objective/train/theoretical_loss": 3.450657476118936,
"objective/train/tokens_used": 1884959200,
"theoretical_loss": 3.450657476118936,
"tokens_seen": 1864499200
},
{
"epoch": 0.56,
"learning_rate": 0.00021977933801404213,
"loss": 2.4782,
"theoretical_loss": 3.450657476118936,
"tokens_seen": 1864499200
},
{
"epoch": 0.57,
"learning_rate": 0.0002197693079237713,
"loss": 2.5578,
"theoretical_loss": 3.450647223536321,
"tokens_seen": 1864564736
},
{
"epoch": 0.57,
"learning_rate": 0.0002197592778335005,
"loss": 2.5921,
"theoretical_loss": 3.450636971414955,
"tokens_seen": 1864630272
},
{
"epoch": 0.57,
"learning_rate": 0.0002197492477432297,
"loss": 2.7234,
"theoretical_loss": 3.4506267197548017,
"tokens_seen": 1864695808
},
{
"epoch": 0.57,
"learning_rate": 0.00021973921765295888,
"loss": 2.5954,
"theoretical_loss": 3.4506164685558245,
"tokens_seen": 1864761344
},
{
"epoch": 0.57,
"learning_rate": 0.00021972918756268806,
"loss": 2.6596,
"theoretical_loss": 3.4506062178179855,
"tokens_seen": 1864826880
},
{
"epoch": 0.57,
"learning_rate": 0.00021971915747241724,
"loss": 2.6824,
"theoretical_loss": 3.450595967541248,
"tokens_seen": 1864892416
},
{
"epoch": 0.57,
"learning_rate": 0.00021970912738214642,
"loss": 2.7332,
"theoretical_loss": 3.450585717725576,
"tokens_seen": 1864957952
},
{
"epoch": 0.57,
"learning_rate": 0.00021969909729187563,
"loss": 2.9049,
"theoretical_loss": 3.450575468370931,
"tokens_seen": 1865023488
},
{
"epoch": 0.57,
"learning_rate": 0.0002196890672016048,
"loss": 2.5465,
"theoretical_loss": 3.4505652194772773,
"tokens_seen": 1865089024
},
{
"epoch": 0.57,
"learning_rate": 0.000219679037111334,
"loss": 2.432,
"theoretical_loss": 3.4505549710445775,
"tokens_seen": 1865154560
},
{
"epoch": 0.57,
"learning_rate": 0.00021966900702106317,
"loss": 2.6996,
"theoretical_loss": 3.450544723072795,
"tokens_seen": 1865220096
},
{
"epoch": 0.57,
"learning_rate": 0.00021965897693079238,
"loss": 2.6155,
"theoretical_loss": 3.4505344755618923,
"tokens_seen": 1865285632
},
{
"epoch": 0.57,
"learning_rate": 0.00021964894684052156,
"loss": 2.5024,
"theoretical_loss": 3.450524228511833,
"tokens_seen": 1865351168
},
{
"epoch": 0.57,
"learning_rate": 0.00021963891675025075,
"loss": 2.7171,
"theoretical_loss": 3.4505139819225796,
"tokens_seen": 1865416704
},
{
"epoch": 0.57,
"learning_rate": 0.00021962888665997993,
"loss": 2.7429,
"theoretical_loss": 3.450503735794096,
"tokens_seen": 1865482240
},
{
"epoch": 0.57,
"learning_rate": 0.0002196188565697091,
"loss": 2.7152,
"theoretical_loss": 3.4504934901263447,
"tokens_seen": 1865547776
},
{
"epoch": 0.57,
"learning_rate": 0.00021960882647943832,
"loss": 2.5016,
"theoretical_loss": 3.450483244919289,
"tokens_seen": 1865613312
},
{
"epoch": 0.57,
"learning_rate": 0.0002195987963891675,
"loss": 2.7512,
"theoretical_loss": 3.450473000172892,
"tokens_seen": 1865678848
},
{
"epoch": 0.57,
"learning_rate": 0.0002195887662988967,
"loss": 2.6878,
"theoretical_loss": 3.450462755887117,
"tokens_seen": 1865744384
},
{
"epoch": 0.57,
"learning_rate": 0.00021957873620862589,
"loss": 2.6727,
"theoretical_loss": 3.4504525120619265,
"tokens_seen": 1865809920
},
{
"epoch": 0.57,
"learning_rate": 0.0002195687061183551,
"loss": 2.4596,
"theoretical_loss": 3.4504422686972838,
"tokens_seen": 1865875456
},
{
"epoch": 0.57,
"learning_rate": 0.00021955867602808428,
"loss": 2.6825,
"theoretical_loss": 3.4504320257931527,
"tokens_seen": 1865940992
},
{
"epoch": 0.57,
"learning_rate": 0.00021954864593781346,
"loss": 2.7313,
"theoretical_loss": 3.450421783349496,
"tokens_seen": 1866006528
},
{
"epoch": 0.57,
"learning_rate": 0.00021953861584754264,
"loss": 2.629,
"theoretical_loss": 3.450411541366276,
"tokens_seen": 1866072064
},
{
"epoch": 0.57,
"objective/train/docs_used": 1054832,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6300954818725586,
"objective/train/theoretical_loss": 3.450401299843457,
"objective/train/tokens_used": 1886597600,
"theoretical_loss": 3.450401299843457,
"tokens_seen": 1866137600
},
{
"epoch": 0.57,
"learning_rate": 0.00021952858575727182,
"loss": 2.5766,
"theoretical_loss": 3.450401299843457,
"tokens_seen": 1866137600
},
{
"epoch": 0.57,
"learning_rate": 0.00021951855566700103,
"loss": 2.691,
"theoretical_loss": 3.450391058781001,
"tokens_seen": 1866203136
},
{
"epoch": 0.57,
"learning_rate": 0.0002195085255767302,
"loss": 2.6154,
"theoretical_loss": 3.450380818178872,
"tokens_seen": 1866268672
},
{
"epoch": 0.57,
"learning_rate": 0.0002194984954864594,
"loss": 2.5903,
"theoretical_loss": 3.4503705780370337,
"tokens_seen": 1866334208
},
{
"epoch": 0.57,
"learning_rate": 0.00021948846539618857,
"loss": 2.7549,
"theoretical_loss": 3.4503603383554475,
"tokens_seen": 1866399744
},
{
"epoch": 0.57,
"learning_rate": 0.00021947843530591778,
"loss": 2.5433,
"theoretical_loss": 3.450350099134078,
"tokens_seen": 1866465280
},
{
"epoch": 0.57,
"learning_rate": 0.00021946840521564696,
"loss": 2.4681,
"theoretical_loss": 3.450339860372888,
"tokens_seen": 1866530816
},
{
"epoch": 0.57,
"learning_rate": 0.00021945837512537614,
"loss": 2.5455,
"theoretical_loss": 3.4503296220718402,
"tokens_seen": 1866596352
},
{
"epoch": 0.57,
"learning_rate": 0.00021944834503510532,
"loss": 2.6757,
"theoretical_loss": 3.4503193842308986,
"tokens_seen": 1866661888
},
{
"epoch": 0.57,
"learning_rate": 0.0002194383149448345,
"loss": 2.5536,
"theoretical_loss": 3.4503091468500253,
"tokens_seen": 1866727424
},
{
"epoch": 0.57,
"learning_rate": 0.0002194282848545637,
"loss": 2.5184,
"theoretical_loss": 3.4502989099291845,
"tokens_seen": 1866792960
},
{
"epoch": 0.57,
"learning_rate": 0.0002194182547642929,
"loss": 2.62,
"theoretical_loss": 3.450288673468339,
"tokens_seen": 1866858496
},
{
"epoch": 0.57,
"learning_rate": 0.00021940822467402207,
"loss": 2.5501,
"theoretical_loss": 3.4502784374674516,
"tokens_seen": 1866924032
},
{
"epoch": 0.57,
"learning_rate": 0.00021939819458375125,
"loss": 2.6397,
"theoretical_loss": 3.4502682019264856,
"tokens_seen": 1866989568
},
{
"epoch": 0.57,
"learning_rate": 0.00021938816449348046,
"loss": 2.6849,
"theoretical_loss": 3.450257966845405,
"tokens_seen": 1867055104
},
{
"epoch": 0.57,
"learning_rate": 0.00021937813440320964,
"loss": 2.4161,
"theoretical_loss": 3.450247732224172,
"tokens_seen": 1867120640
},
{
"epoch": 0.57,
"learning_rate": 0.00021936810431293882,
"loss": 2.5379,
"theoretical_loss": 3.4502374980627506,
"tokens_seen": 1867186176
},
{
"epoch": 0.57,
"learning_rate": 0.000219358074222668,
"loss": 2.6112,
"theoretical_loss": 3.4502272643611027,
"tokens_seen": 1867251712
},
{
"epoch": 0.57,
"learning_rate": 0.00021934804413239721,
"loss": 2.7584,
"theoretical_loss": 3.4502170311191933,
"tokens_seen": 1867317248
},
{
"epoch": 0.57,
"learning_rate": 0.0002193380140421264,
"loss": 2.642,
"theoretical_loss": 3.4502067983369846,
"tokens_seen": 1867382784
},
{
"epoch": 0.57,
"learning_rate": 0.00021932798395185558,
"loss": 2.7463,
"theoretical_loss": 3.45019656601444,
"tokens_seen": 1867448320
},
{
"epoch": 0.57,
"learning_rate": 0.00021931795386158476,
"loss": 2.5429,
"theoretical_loss": 3.450186334151522,
"tokens_seen": 1867513856
},
{
"epoch": 0.57,
"learning_rate": 0.00021930792377131394,
"loss": 2.7662,
"theoretical_loss": 3.4501761027481948,
"tokens_seen": 1867579392
},
{
"epoch": 0.57,
"learning_rate": 0.00021929789368104315,
"loss": 2.7896,
"theoretical_loss": 3.4501658718044217,
"tokens_seen": 1867644928
},
{
"epoch": 0.57,
"learning_rate": 0.00021928786359077233,
"loss": 2.7091,
"theoretical_loss": 3.4501556413201646,
"tokens_seen": 1867710464
},
{
"epoch": 0.57,
"objective/train/docs_used": 1056292,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2054543495178223,
"objective/train/theoretical_loss": 3.450145411295389,
"objective/train/tokens_used": 1888236000,
"theoretical_loss": 3.450145411295389,
"tokens_seen": 1867776000
},
{
"epoch": 0.57,
"learning_rate": 0.0002192778335005015,
"loss": 2.5827,
"theoretical_loss": 3.450145411295389,
"tokens_seen": 1867776000
},
{
"epoch": 0.57,
"learning_rate": 0.0002192678034102307,
"loss": 2.7055,
"theoretical_loss": 3.450135181730056,
"tokens_seen": 1867841536
},
{
"epoch": 0.57,
"learning_rate": 0.0002192577733199599,
"loss": 2.8107,
"theoretical_loss": 3.4501249526241295,
"tokens_seen": 1867907072
},
{
"epoch": 0.57,
"learning_rate": 0.00021924774322968908,
"loss": 2.5631,
"theoretical_loss": 3.450114723977573,
"tokens_seen": 1867972608
},
{
"epoch": 0.57,
"learning_rate": 0.00021923771313941826,
"loss": 2.7341,
"theoretical_loss": 3.4501044957903506,
"tokens_seen": 1868038144
},
{
"epoch": 0.57,
"learning_rate": 0.00021922768304914744,
"loss": 2.6546,
"theoretical_loss": 3.450094268062424,
"tokens_seen": 1868103680
},
{
"epoch": 0.57,
"learning_rate": 0.00021921765295887662,
"loss": 2.5337,
"theoretical_loss": 3.4500840407937567,
"tokens_seen": 1868169216
},
{
"epoch": 0.57,
"learning_rate": 0.00021920762286860583,
"loss": 2.7914,
"theoretical_loss": 3.450073813984313,
"tokens_seen": 1868234752
},
{
"epoch": 0.57,
"learning_rate": 0.000219197592778335,
"loss": 2.6335,
"theoretical_loss": 3.4500635876340553,
"tokens_seen": 1868300288
},
{
"epoch": 0.57,
"learning_rate": 0.0002191875626880642,
"loss": 2.6575,
"theoretical_loss": 3.450053361742947,
"tokens_seen": 1868365824
},
{
"epoch": 0.57,
"learning_rate": 0.00021917753259779337,
"loss": 2.6851,
"theoretical_loss": 3.450043136310952,
"tokens_seen": 1868431360
},
{
"epoch": 0.57,
"learning_rate": 0.00021916750250752258,
"loss": 2.6483,
"theoretical_loss": 3.450032911338032,
"tokens_seen": 1868496896
},
{
"epoch": 0.57,
"learning_rate": 0.00021915747241725176,
"loss": 2.5857,
"theoretical_loss": 3.4500226868241524,
"tokens_seen": 1868562432
},
{
"epoch": 0.57,
"learning_rate": 0.00021914744232698095,
"loss": 2.5466,
"theoretical_loss": 3.450012462769275,
"tokens_seen": 1868627968
},
{
"epoch": 0.57,
"learning_rate": 0.00021913741223671013,
"loss": 2.5652,
"theoretical_loss": 3.450002239173364,
"tokens_seen": 1868693504
},
{
"epoch": 0.57,
"learning_rate": 0.0002191273821464393,
"loss": 2.5233,
"theoretical_loss": 3.449992016036382,
"tokens_seen": 1868759040
},
{
"epoch": 0.57,
"learning_rate": 0.00021911735205616852,
"loss": 2.8728,
"theoretical_loss": 3.449981793358292,
"tokens_seen": 1868824576
},
{
"epoch": 0.57,
"learning_rate": 0.0002191073219658977,
"loss": 2.9047,
"theoretical_loss": 3.4499715711390584,
"tokens_seen": 1868890112
},
{
"epoch": 0.57,
"learning_rate": 0.00021909729187562688,
"loss": 2.5404,
"theoretical_loss": 3.4499613493786443,
"tokens_seen": 1868955648
},
{
"epoch": 0.57,
"learning_rate": 0.00021908726178535606,
"loss": 2.5774,
"theoretical_loss": 3.4499511280770125,
"tokens_seen": 1869021184
},
{
"epoch": 0.57,
"learning_rate": 0.00021907723169508527,
"loss": 2.7854,
"theoretical_loss": 3.4499409072341263,
"tokens_seen": 1869086720
},
{
"epoch": 0.57,
"learning_rate": 0.00021906720160481445,
"loss": 2.698,
"theoretical_loss": 3.4499306868499495,
"tokens_seen": 1869152256
},
{
"epoch": 0.57,
"learning_rate": 0.00021905717151454363,
"loss": 2.7029,
"theoretical_loss": 3.449920466924445,
"tokens_seen": 1869217792
},
{
"epoch": 0.57,
"learning_rate": 0.0002190471414242728,
"loss": 2.5455,
"theoretical_loss": 3.4499102474575762,
"tokens_seen": 1869283328
},
{
"epoch": 0.57,
"learning_rate": 0.000219037111334002,
"loss": 2.6987,
"theoretical_loss": 3.4499000284493073,
"tokens_seen": 1869348864
},
{
"epoch": 0.57,
"objective/train/docs_used": 1056874,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6355602741241455,
"objective/train/theoretical_loss": 3.4498898098996,
"objective/train/tokens_used": 1889874400,
"theoretical_loss": 3.4498898098996,
"tokens_seen": 1869414400
},
{
"epoch": 0.57,
"learning_rate": 0.0002190270812437312,
"loss": 2.7244,
"theoretical_loss": 3.4498898098996,
"tokens_seen": 1869414400
},
{
"epoch": 0.57,
"learning_rate": 0.00021901705115346038,
"loss": 2.5975,
"theoretical_loss": 3.4498795918084193,
"tokens_seen": 1869479936
},
{
"epoch": 0.57,
"learning_rate": 0.00021900702106318956,
"loss": 2.6968,
"theoretical_loss": 3.4498693741757274,
"tokens_seen": 1869545472
},
{
"epoch": 0.57,
"learning_rate": 0.00021899699097291874,
"loss": 2.3733,
"theoretical_loss": 3.4498591570014883,
"tokens_seen": 1869611008
},
{
"epoch": 0.57,
"learning_rate": 0.00021898696088264795,
"loss": 2.6747,
"theoretical_loss": 3.4498489402856647,
"tokens_seen": 1869676544
},
{
"epoch": 0.57,
"learning_rate": 0.00021897693079237713,
"loss": 2.5203,
"theoretical_loss": 3.4498387240282207,
"tokens_seen": 1869742080
},
{
"epoch": 0.57,
"learning_rate": 0.00021896690070210631,
"loss": 2.352,
"theoretical_loss": 3.4498285082291194,
"tokens_seen": 1869807616
},
{
"epoch": 0.57,
"learning_rate": 0.0002189568706118355,
"loss": 2.4223,
"theoretical_loss": 3.449818292888324,
"tokens_seen": 1869873152
},
{
"epoch": 0.57,
"learning_rate": 0.00021894684052156468,
"loss": 2.5942,
"theoretical_loss": 3.4498080780057983,
"tokens_seen": 1869938688
},
{
"epoch": 0.57,
"learning_rate": 0.00021893681043129388,
"loss": 2.9054,
"theoretical_loss": 3.449797863581505,
"tokens_seen": 1870004224
},
{
"epoch": 0.57,
"learning_rate": 0.00021892678034102307,
"loss": 2.5398,
"theoretical_loss": 3.4497876496154083,
"tokens_seen": 1870069760
},
{
"epoch": 0.57,
"learning_rate": 0.00021891675025075225,
"loss": 2.4978,
"theoretical_loss": 3.4497774361074707,
"tokens_seen": 1870135296
},
{
"epoch": 0.57,
"learning_rate": 0.00021890672016048143,
"loss": 2.6243,
"theoretical_loss": 3.4497672230576564,
"tokens_seen": 1870200832
},
{
"epoch": 0.57,
"learning_rate": 0.00021889669007021064,
"loss": 2.6614,
"theoretical_loss": 3.4497570104659285,
"tokens_seen": 1870266368
},
{
"epoch": 0.57,
"learning_rate": 0.00021888665997993982,
"loss": 2.5776,
"theoretical_loss": 3.4497467983322503,
"tokens_seen": 1870331904
},
{
"epoch": 0.57,
"learning_rate": 0.000218876629889669,
"loss": 2.7323,
"theoretical_loss": 3.4497365866565852,
"tokens_seen": 1870397440
},
{
"epoch": 0.57,
"learning_rate": 0.00021886659979939818,
"loss": 2.7206,
"theoretical_loss": 3.4497263754388965,
"tokens_seen": 1870462976
},
{
"epoch": 0.57,
"learning_rate": 0.00021885656970912736,
"loss": 2.7129,
"theoretical_loss": 3.449716164679148,
"tokens_seen": 1870528512
},
{
"epoch": 0.57,
"learning_rate": 0.00021884653961885657,
"loss": 2.6854,
"theoretical_loss": 3.449705954377303,
"tokens_seen": 1870594048
},
{
"epoch": 0.57,
"learning_rate": 0.00021883650952858578,
"loss": 2.8059,
"theoretical_loss": 3.449695744533325,
"tokens_seen": 1870659584
},
{
"epoch": 0.57,
"learning_rate": 0.00021882647943831496,
"loss": 2.8278,
"theoretical_loss": 3.449685535147177,
"tokens_seen": 1870725120
},
{
"epoch": 0.57,
"learning_rate": 0.00021881644934804414,
"loss": 2.5443,
"theoretical_loss": 3.4496753262188227,
"tokens_seen": 1870790656
},
{
"epoch": 0.57,
"learning_rate": 0.00021880641925777335,
"loss": 2.5995,
"theoretical_loss": 3.4496651177482254,
"tokens_seen": 1870856192
},
{
"epoch": 0.57,
"learning_rate": 0.00021879638916750253,
"loss": 2.4988,
"theoretical_loss": 3.449654909735349,
"tokens_seen": 1870921728
},
{
"epoch": 0.57,
"learning_rate": 0.0002187863590772317,
"loss": 2.6662,
"theoretical_loss": 3.449644702180156,
"tokens_seen": 1870987264
},
{
"epoch": 0.57,
"objective/train/docs_used": 1058498,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9856131076812744,
"objective/train/theoretical_loss": 3.4496344950826114,
"objective/train/tokens_used": 1891512800,
"theoretical_loss": 3.4496344950826114,
"tokens_seen": 1871052800
},
{
"epoch": 0.57,
"learning_rate": 0.0002187763289869609,
"loss": 2.5089,
"theoretical_loss": 3.4496344950826114,
"tokens_seen": 1871052800
},
{
"epoch": 0.57,
"learning_rate": 0.0002187662988966901,
"loss": 2.7941,
"theoretical_loss": 3.4496242884426773,
"tokens_seen": 1871118336
},
{
"epoch": 0.57,
"learning_rate": 0.00021875626880641928,
"loss": 2.7105,
"theoretical_loss": 3.449614082260317,
"tokens_seen": 1871183872
},
{
"epoch": 0.57,
"learning_rate": 0.00021874623871614846,
"loss": 2.61,
"theoretical_loss": 3.4496038765354955,
"tokens_seen": 1871249408
},
{
"epoch": 0.57,
"learning_rate": 0.00021873620862587764,
"loss": 2.5478,
"theoretical_loss": 3.449593671268175,
"tokens_seen": 1871314944
},
{
"epoch": 0.57,
"learning_rate": 0.00021872617853560682,
"loss": 2.4176,
"theoretical_loss": 3.449583466458319,
"tokens_seen": 1871380480
},
{
"epoch": 0.57,
"learning_rate": 0.00021871614844533603,
"loss": 2.548,
"theoretical_loss": 3.4495732621058917,
"tokens_seen": 1871446016
},
{
"epoch": 0.57,
"learning_rate": 0.0002187061183550652,
"loss": 2.6792,
"theoretical_loss": 3.449563058210856,
"tokens_seen": 1871511552
},
{
"epoch": 0.57,
"learning_rate": 0.0002186960882647944,
"loss": 2.6964,
"theoretical_loss": 3.4495528547731755,
"tokens_seen": 1871577088
},
{
"epoch": 0.57,
"learning_rate": 0.00021868605817452357,
"loss": 2.6626,
"theoretical_loss": 3.4495426517928136,
"tokens_seen": 1871642624
},
{
"epoch": 0.57,
"learning_rate": 0.00021867602808425278,
"loss": 2.7231,
"theoretical_loss": 3.449532449269734,
"tokens_seen": 1871708160
},
{
"epoch": 0.57,
"learning_rate": 0.00021866599799398196,
"loss": 2.5226,
"theoretical_loss": 3.4495222472039,
"tokens_seen": 1871773696
},
{
"epoch": 0.57,
"learning_rate": 0.00021865596790371115,
"loss": 2.6661,
"theoretical_loss": 3.449512045595275,
"tokens_seen": 1871839232
},
{
"epoch": 0.57,
"learning_rate": 0.00021864593781344033,
"loss": 2.7372,
"theoretical_loss": 3.449501844443823,
"tokens_seen": 1871904768
},
{
"epoch": 0.57,
"learning_rate": 0.0002186359077231695,
"loss": 2.6515,
"theoretical_loss": 3.449491643749507,
"tokens_seen": 1871970304
},
{
"epoch": 0.57,
"learning_rate": 0.00021862587763289872,
"loss": 2.4126,
"theoretical_loss": 3.449481443512291,
"tokens_seen": 1872035840
},
{
"epoch": 0.57,
"learning_rate": 0.0002186158475426279,
"loss": 2.7273,
"theoretical_loss": 3.449471243732138,
"tokens_seen": 1872101376
},
{
"epoch": 0.57,
"learning_rate": 0.00021860581745235708,
"loss": 2.5712,
"theoretical_loss": 3.449461044409012,
"tokens_seen": 1872166912
},
{
"epoch": 0.57,
"learning_rate": 0.00021859578736208626,
"loss": 2.8934,
"theoretical_loss": 3.449450845542876,
"tokens_seen": 1872232448
},
{
"epoch": 0.57,
"learning_rate": 0.00021858575727181547,
"loss": 2.5635,
"theoretical_loss": 3.4494406471336934,
"tokens_seen": 1872297984
},
{
"epoch": 0.57,
"learning_rate": 0.00021857572718154465,
"loss": 2.5822,
"theoretical_loss": 3.4494304491814285,
"tokens_seen": 1872363520
},
{
"epoch": 0.57,
"learning_rate": 0.00021856569709127383,
"loss": 2.8743,
"theoretical_loss": 3.449420251686045,
"tokens_seen": 1872429056
},
{
"epoch": 0.57,
"learning_rate": 0.000218555667001003,
"loss": 2.4279,
"theoretical_loss": 3.4494100546475055,
"tokens_seen": 1872494592
},
{
"epoch": 0.57,
"learning_rate": 0.0002185456369107322,
"loss": 2.6762,
"theoretical_loss": 3.4493998580657736,
"tokens_seen": 1872560128
},
{
"epoch": 0.57,
"learning_rate": 0.0002185356068204614,
"loss": 2.6567,
"theoretical_loss": 3.4493896619408138,
"tokens_seen": 1872625664
},
{
"epoch": 0.57,
"objective/train/docs_used": 1059136,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4825518131256104,
"objective/train/theoretical_loss": 3.449379466272589,
"objective/train/tokens_used": 1893151200,
"theoretical_loss": 3.449379466272589,
"tokens_seen": 1872691200
},
{
"epoch": 0.57,
"learning_rate": 0.00021852557673019058,
"loss": 2.5793,
"theoretical_loss": 3.449379466272589,
"tokens_seen": 1872691200
},
{
"epoch": 0.57,
"learning_rate": 0.00021851554663991976,
"loss": 2.4976,
"theoretical_loss": 3.4493692710610624,
"tokens_seen": 1872756736
},
{
"epoch": 0.57,
"learning_rate": 0.00021850551654964894,
"loss": 2.7383,
"theoretical_loss": 3.449359076306198,
"tokens_seen": 1872822272
},
{
"epoch": 0.57,
"learning_rate": 0.00021849548645937815,
"loss": 2.8148,
"theoretical_loss": 3.4493488820079596,
"tokens_seen": 1872887808
},
{
"epoch": 0.57,
"learning_rate": 0.00021848545636910733,
"loss": 2.5924,
"theoretical_loss": 3.4493386881663106,
"tokens_seen": 1872953344
},
{
"epoch": 0.57,
"learning_rate": 0.00021847542627883651,
"loss": 2.7275,
"theoretical_loss": 3.4493284947812146,
"tokens_seen": 1873018880
},
{
"epoch": 0.57,
"learning_rate": 0.0002184653961885657,
"loss": 2.5906,
"theoretical_loss": 3.4493183018526348,
"tokens_seen": 1873084416
},
{
"epoch": 0.57,
"learning_rate": 0.00021845536609829488,
"loss": 2.9778,
"theoretical_loss": 3.449308109380535,
"tokens_seen": 1873149952
},
{
"epoch": 0.57,
"learning_rate": 0.00021844533600802408,
"loss": 2.7796,
"theoretical_loss": 3.449297917364879,
"tokens_seen": 1873215488
},
{
"epoch": 0.57,
"learning_rate": 0.00021843530591775327,
"loss": 2.5879,
"theoretical_loss": 3.44928772580563,
"tokens_seen": 1873281024
},
{
"epoch": 0.57,
"learning_rate": 0.00021842527582748245,
"loss": 2.6287,
"theoretical_loss": 3.449277534702752,
"tokens_seen": 1873346560
},
{
"epoch": 0.57,
"learning_rate": 0.00021841524573721163,
"loss": 2.6887,
"theoretical_loss": 3.4492673440562083,
"tokens_seen": 1873412096
},
{
"epoch": 0.57,
"learning_rate": 0.00021840521564694084,
"loss": 2.774,
"theoretical_loss": 3.449257153865963,
"tokens_seen": 1873477632
},
{
"epoch": 0.57,
"learning_rate": 0.00021839518555667002,
"loss": 2.7001,
"theoretical_loss": 3.449246964131979,
"tokens_seen": 1873543168
},
{
"epoch": 0.57,
"learning_rate": 0.0002183851554663992,
"loss": 2.8263,
"theoretical_loss": 3.4492367748542203,
"tokens_seen": 1873608704
},
{
"epoch": 0.57,
"learning_rate": 0.00021837512537612838,
"loss": 2.6648,
"theoretical_loss": 3.4492265860326508,
"tokens_seen": 1873674240
},
{
"epoch": 0.57,
"learning_rate": 0.00021836509528585756,
"loss": 2.6239,
"theoretical_loss": 3.449216397667233,
"tokens_seen": 1873739776
},
{
"epoch": 0.57,
"learning_rate": 0.00021835506519558677,
"loss": 2.6981,
"theoretical_loss": 3.449206209757932,
"tokens_seen": 1873805312
},
{
"epoch": 0.57,
"learning_rate": 0.00021834503510531595,
"loss": 2.5792,
"theoretical_loss": 3.4491960223047107,
"tokens_seen": 1873870848
},
{
"epoch": 0.57,
"learning_rate": 0.00021833500501504513,
"loss": 2.5665,
"theoretical_loss": 3.4491858353075324,
"tokens_seen": 1873936384
},
{
"epoch": 0.57,
"learning_rate": 0.0002183249749247743,
"loss": 2.6201,
"theoretical_loss": 3.449175648766362,
"tokens_seen": 1874001920
},
{
"epoch": 0.57,
"learning_rate": 0.00021831494483450352,
"loss": 2.5902,
"theoretical_loss": 3.449165462681161,
"tokens_seen": 1874067456
},
{
"epoch": 0.57,
"learning_rate": 0.0002183049147442327,
"loss": 2.7632,
"theoretical_loss": 3.4491552770518954,
"tokens_seen": 1874132992
},
{
"epoch": 0.57,
"learning_rate": 0.00021829488465396188,
"loss": 2.5817,
"theoretical_loss": 3.449145091878527,
"tokens_seen": 1874198528
},
{
"epoch": 0.57,
"learning_rate": 0.00021828485456369106,
"loss": 2.5589,
"theoretical_loss": 3.4491349071610204,
"tokens_seen": 1874264064
},
{
"epoch": 0.57,
"objective/train/docs_used": 1059915,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6237382888793945,
"objective/train/theoretical_loss": 3.449124722899339,
"objective/train/tokens_used": 1894789600,
"theoretical_loss": 3.449124722899339,
"tokens_seen": 1874329600
},
{
"epoch": 0.57,
"learning_rate": 0.00021827482447342024,
"loss": 2.6194,
"theoretical_loss": 3.449124722899339,
"tokens_seen": 1874329600
},
{
"epoch": 0.57,
"learning_rate": 0.00021826479438314945,
"loss": 2.5369,
"theoretical_loss": 3.449114539093447,
"tokens_seen": 1874395136
},
{
"epoch": 0.57,
"learning_rate": 0.00021825476429287863,
"loss": 2.4653,
"theoretical_loss": 3.449104355743307,
"tokens_seen": 1874460672
},
{
"epoch": 0.57,
"learning_rate": 0.00021824473420260782,
"loss": 2.4569,
"theoretical_loss": 3.4490941728488838,
"tokens_seen": 1874526208
},
{
"epoch": 0.57,
"learning_rate": 0.000218234704112337,
"loss": 2.6477,
"theoretical_loss": 3.44908399041014,
"tokens_seen": 1874591744
},
{
"epoch": 0.57,
"learning_rate": 0.0002182246740220662,
"loss": 2.8331,
"theoretical_loss": 3.4490738084270403,
"tokens_seen": 1874657280
},
{
"epoch": 0.57,
"learning_rate": 0.00021821464393179539,
"loss": 2.6785,
"theoretical_loss": 3.4490636268995476,
"tokens_seen": 1874722816
},
{
"epoch": 0.57,
"learning_rate": 0.00021820461384152457,
"loss": 2.4524,
"theoretical_loss": 3.4490534458276256,
"tokens_seen": 1874788352
},
{
"epoch": 0.57,
"learning_rate": 0.00021819458375125375,
"loss": 2.7431,
"theoretical_loss": 3.449043265211239,
"tokens_seen": 1874853888
},
{
"epoch": 0.57,
"learning_rate": 0.00021818455366098296,
"loss": 2.6855,
"theoretical_loss": 3.4490330850503503,
"tokens_seen": 1874919424
},
{
"epoch": 0.57,
"learning_rate": 0.00021817452357071214,
"loss": 2.7464,
"theoretical_loss": 3.4490229053449237,
"tokens_seen": 1874984960
},
{
"epoch": 0.57,
"learning_rate": 0.00021816449348044132,
"loss": 2.6687,
"theoretical_loss": 3.449012726094923,
"tokens_seen": 1875050496
},
{
"epoch": 0.57,
"learning_rate": 0.0002181544633901705,
"loss": 2.6016,
"theoretical_loss": 3.449002547300312,
"tokens_seen": 1875116032
},
{
"epoch": 0.57,
"learning_rate": 0.00021814443329989968,
"loss": 2.429,
"theoretical_loss": 3.4489923689610538,
"tokens_seen": 1875181568
},
{
"epoch": 0.57,
"learning_rate": 0.0002181344032096289,
"loss": 2.9365,
"theoretical_loss": 3.4489821910771123,
"tokens_seen": 1875247104
},
{
"epoch": 0.57,
"learning_rate": 0.00021812437311935807,
"loss": 2.7628,
"theoretical_loss": 3.448972013648452,
"tokens_seen": 1875312640
},
{
"epoch": 0.57,
"learning_rate": 0.00021811434302908725,
"loss": 2.6083,
"theoretical_loss": 3.4489618366750356,
"tokens_seen": 1875378176
},
{
"epoch": 0.57,
"learning_rate": 0.00021810431293881643,
"loss": 2.8164,
"theoretical_loss": 3.448951660156828,
"tokens_seen": 1875443712
},
{
"epoch": 0.57,
"learning_rate": 0.00021809428284854564,
"loss": 2.58,
"theoretical_loss": 3.4489414840937913,
"tokens_seen": 1875509248
},
{
"epoch": 0.57,
"learning_rate": 0.00021808425275827485,
"loss": 2.5659,
"theoretical_loss": 3.448931308485891,
"tokens_seen": 1875574784
},
{
"epoch": 0.57,
"learning_rate": 0.00021807422266800403,
"loss": 2.7759,
"theoretical_loss": 3.4489211333330894,
"tokens_seen": 1875640320
},
{
"epoch": 0.57,
"learning_rate": 0.0002180641925777332,
"loss": 2.5309,
"theoretical_loss": 3.4489109586353512,
"tokens_seen": 1875705856
},
{
"epoch": 0.57,
"learning_rate": 0.0002180541624874624,
"loss": 2.6571,
"theoretical_loss": 3.44890078439264,
"tokens_seen": 1875771392
},
{
"epoch": 0.57,
"learning_rate": 0.0002180441323971916,
"loss": 2.7709,
"theoretical_loss": 3.448890610604918,
"tokens_seen": 1875836928
},
{
"epoch": 0.57,
"learning_rate": 0.00021803410230692078,
"loss": 2.6503,
"theoretical_loss": 3.4488804372721518,
"tokens_seen": 1875902464
},
{
"epoch": 0.57,
"objective/train/docs_used": 1060624,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8335423469543457,
"objective/train/theoretical_loss": 3.448870264394303,
"objective/train/tokens_used": 1896428000,
"theoretical_loss": 3.448870264394303,
"tokens_seen": 1875968000
},
{
"epoch": 0.57,
"learning_rate": 0.00021802407221664996,
"loss": 2.6851,
"theoretical_loss": 3.448870264394303,
"tokens_seen": 1875968000
},
{
"epoch": 0.57,
"learning_rate": 0.00021801404212637914,
"loss": 2.5137,
"theoretical_loss": 3.448860091971336,
"tokens_seen": 1876033536
},
{
"epoch": 0.57,
"learning_rate": 0.00021800401203610835,
"loss": 2.6206,
"theoretical_loss": 3.448849920003215,
"tokens_seen": 1876099072
},
{
"epoch": 0.57,
"learning_rate": 0.00021799398194583753,
"loss": 2.6897,
"theoretical_loss": 3.4488397484899034,
"tokens_seen": 1876164608
},
{
"epoch": 0.57,
"learning_rate": 0.00021798395185556671,
"loss": 2.5449,
"theoretical_loss": 3.448829577431365,
"tokens_seen": 1876230144
},
{
"epoch": 0.57,
"learning_rate": 0.0002179739217652959,
"loss": 2.6745,
"theoretical_loss": 3.448819406827563,
"tokens_seen": 1876295680
},
{
"epoch": 0.57,
"learning_rate": 0.00021796389167502508,
"loss": 2.6209,
"theoretical_loss": 3.448809236678462,
"tokens_seen": 1876361216
},
{
"epoch": 0.57,
"learning_rate": 0.00021795386158475428,
"loss": 2.6175,
"theoretical_loss": 3.4487990669840256,
"tokens_seen": 1876426752
},
{
"epoch": 0.57,
"learning_rate": 0.00021794383149448347,
"loss": 2.5958,
"theoretical_loss": 3.448788897744217,
"tokens_seen": 1876492288
},
{
"epoch": 0.57,
"learning_rate": 0.00021793380140421265,
"loss": 2.56,
"theoretical_loss": 3.4487787289590015,
"tokens_seen": 1876557824
},
{
"epoch": 0.57,
"learning_rate": 0.00021792377131394183,
"loss": 2.3724,
"theoretical_loss": 3.4487685606283414,
"tokens_seen": 1876623360
},
{
"epoch": 0.57,
"learning_rate": 0.00021791374122367104,
"loss": 2.6453,
"theoretical_loss": 3.4487583927522008,
"tokens_seen": 1876688896
},
{
"epoch": 0.57,
"learning_rate": 0.00021790371113340022,
"loss": 2.3816,
"theoretical_loss": 3.4487482253305437,
"tokens_seen": 1876754432
},
{
"epoch": 0.57,
"learning_rate": 0.0002178936810431294,
"loss": 2.6897,
"theoretical_loss": 3.4487380583633342,
"tokens_seen": 1876819968
},
{
"epoch": 0.57,
"learning_rate": 0.00021788365095285858,
"loss": 2.8264,
"theoretical_loss": 3.4487278918505364,
"tokens_seen": 1876885504
},
{
"epoch": 0.57,
"learning_rate": 0.00021787362086258776,
"loss": 2.6574,
"theoretical_loss": 3.448717725792113,
"tokens_seen": 1876951040
},
{
"epoch": 0.57,
"learning_rate": 0.00021786359077231697,
"loss": 2.6981,
"theoretical_loss": 3.448707560188028,
"tokens_seen": 1877016576
},
{
"epoch": 0.57,
"learning_rate": 0.00021785356068204615,
"loss": 2.6018,
"theoretical_loss": 3.4486973950382467,
"tokens_seen": 1877082112
},
{
"epoch": 0.57,
"learning_rate": 0.00021784353059177533,
"loss": 2.5834,
"theoretical_loss": 3.448687230342731,
"tokens_seen": 1877147648
},
{
"epoch": 0.57,
"learning_rate": 0.0002178335005015045,
"loss": 2.9554,
"theoretical_loss": 3.448677066101446,
"tokens_seen": 1877213184
},
{
"epoch": 0.57,
"learning_rate": 0.00021782347041123372,
"loss": 2.5066,
"theoretical_loss": 3.448666902314355,
"tokens_seen": 1877278720
},
{
"epoch": 0.57,
"learning_rate": 0.0002178134403209629,
"loss": 2.6876,
"theoretical_loss": 3.4486567389814224,
"tokens_seen": 1877344256
},
{
"epoch": 0.57,
"learning_rate": 0.00021780341023069208,
"loss": 2.5085,
"theoretical_loss": 3.4486465761026115,
"tokens_seen": 1877409792
},
{
"epoch": 0.57,
"learning_rate": 0.00021779338014042126,
"loss": 2.654,
"theoretical_loss": 3.448636413677886,
"tokens_seen": 1877475328
},
{
"epoch": 0.57,
"learning_rate": 0.00021778335005015044,
"loss": 2.7877,
"theoretical_loss": 3.4486262517072106,
"tokens_seen": 1877540864
},
{
"epoch": 0.57,
"objective/train/docs_used": 1061320,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4820566177368164,
"objective/train/theoretical_loss": 3.4486160901905487,
"objective/train/tokens_used": 1898066400,
"theoretical_loss": 3.4486160901905487,
"tokens_seen": 1877606400
},
{
"epoch": 0.57,
"learning_rate": 0.00021777331995987965,
"loss": 2.6811,
"theoretical_loss": 3.4486160901905487,
"tokens_seen": 1877606400
},
{
"epoch": 0.57,
"learning_rate": 0.00021776328986960883,
"loss": 2.2798,
"theoretical_loss": 3.448605929127864,
"tokens_seen": 1877671936
},
{
"epoch": 0.57,
"learning_rate": 0.00021775325977933802,
"loss": 2.7321,
"theoretical_loss": 3.4485957685191204,
"tokens_seen": 1877737472
},
{
"epoch": 0.57,
"learning_rate": 0.0002177432296890672,
"loss": 2.8374,
"theoretical_loss": 3.448585608364282,
"tokens_seen": 1877803008
},
{
"epoch": 0.57,
"learning_rate": 0.0002177331995987964,
"loss": 2.6185,
"theoretical_loss": 3.4485754486633127,
"tokens_seen": 1877868544
},
{
"epoch": 0.57,
"learning_rate": 0.00021772316950852559,
"loss": 2.6525,
"theoretical_loss": 3.448565289416176,
"tokens_seen": 1877934080
},
{
"epoch": 0.57,
"learning_rate": 0.00021771313941825477,
"loss": 2.5962,
"theoretical_loss": 3.448555130622836,
"tokens_seen": 1877999616
},
{
"epoch": 0.57,
"learning_rate": 0.00021770310932798395,
"loss": 2.6581,
"theoretical_loss": 3.4485449722832566,
"tokens_seen": 1878065152
},
{
"epoch": 0.57,
"learning_rate": 0.00021769307923771316,
"loss": 2.7001,
"theoretical_loss": 3.4485348143974024,
"tokens_seen": 1878130688
},
{
"epoch": 0.57,
"learning_rate": 0.00021768304914744234,
"loss": 2.4452,
"theoretical_loss": 3.448524656965236,
"tokens_seen": 1878196224
},
{
"epoch": 0.57,
"learning_rate": 0.00021767301905717152,
"loss": 2.698,
"theoretical_loss": 3.448514499986722,
"tokens_seen": 1878261760
},
{
"epoch": 0.57,
"learning_rate": 0.0002176629889669007,
"loss": 2.7423,
"theoretical_loss": 3.448504343461824,
"tokens_seen": 1878327296
},
{
"epoch": 0.57,
"learning_rate": 0.00021765295887662988,
"loss": 2.762,
"theoretical_loss": 3.448494187390507,
"tokens_seen": 1878392832
},
{
"epoch": 0.57,
"learning_rate": 0.0002176429287863591,
"loss": 2.5384,
"theoretical_loss": 3.448484031772734,
"tokens_seen": 1878458368
},
{
"epoch": 0.57,
"learning_rate": 0.00021763289869608827,
"loss": 2.3584,
"theoretical_loss": 3.4484738766084684,
"tokens_seen": 1878523904
},
{
"epoch": 0.57,
"learning_rate": 0.00021762286860581745,
"loss": 2.6882,
"theoretical_loss": 3.4484637218976752,
"tokens_seen": 1878589440
},
{
"epoch": 0.57,
"learning_rate": 0.00021761283851554663,
"loss": 2.3661,
"theoretical_loss": 3.4484535676403176,
"tokens_seen": 1878654976
},
{
"epoch": 0.57,
"learning_rate": 0.00021760280842527584,
"loss": 2.531,
"theoretical_loss": 3.44844341383636,
"tokens_seen": 1878720512
},
{
"epoch": 0.57,
"learning_rate": 0.00021759277833500502,
"loss": 2.6975,
"theoretical_loss": 3.448433260485766,
"tokens_seen": 1878786048
},
{
"epoch": 0.57,
"learning_rate": 0.0002175827482447342,
"loss": 2.6091,
"theoretical_loss": 3.4484231075885,
"tokens_seen": 1878851584
},
{
"epoch": 0.57,
"learning_rate": 0.00021757271815446338,
"loss": 2.585,
"theoretical_loss": 3.4484129551445255,
"tokens_seen": 1878917120
},
{
"epoch": 0.57,
"learning_rate": 0.00021756268806419257,
"loss": 2.7176,
"theoretical_loss": 3.4484028031538063,
"tokens_seen": 1878982656
},
{
"epoch": 0.57,
"learning_rate": 0.00021755265797392177,
"loss": 2.6405,
"theoretical_loss": 3.4483926516163073,
"tokens_seen": 1879048192
},
{
"epoch": 0.57,
"learning_rate": 0.00021754262788365095,
"loss": 2.5056,
"theoretical_loss": 3.4483825005319915,
"tokens_seen": 1879113728
},
{
"epoch": 0.57,
"learning_rate": 0.00021753259779338014,
"loss": 2.4966,
"theoretical_loss": 3.448372349900823,
"tokens_seen": 1879179264
},
{
"epoch": 0.57,
"objective/train/docs_used": 1062685,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.720628499984741,
"objective/train/theoretical_loss": 3.448362199722766,
"objective/train/tokens_used": 1899704800,
"theoretical_loss": 3.448362199722766,
"tokens_seen": 1879244800
},
{
"epoch": 0.57,
"learning_rate": 0.00021752256770310932,
"loss": 2.7414,
"theoretical_loss": 3.448362199722766,
"tokens_seen": 1879244800
},
{
"epoch": 0.57,
"learning_rate": 0.00021751253761283852,
"loss": 2.7563,
"theoretical_loss": 3.4483520499977844,
"tokens_seen": 1879310336
},
{
"epoch": 0.57,
"learning_rate": 0.0002175025075225677,
"loss": 2.6274,
"theoretical_loss": 3.4483419007258425,
"tokens_seen": 1879375872
},
{
"epoch": 0.57,
"learning_rate": 0.0002174924774322969,
"loss": 2.5781,
"theoretical_loss": 3.4483317519069043,
"tokens_seen": 1879441408
},
{
"epoch": 0.57,
"learning_rate": 0.00021748244734202607,
"loss": 2.5972,
"theoretical_loss": 3.4483216035409328,
"tokens_seen": 1879506944
},
{
"epoch": 0.57,
"learning_rate": 0.00021747241725175525,
"loss": 2.7683,
"theoretical_loss": 3.4483114556278927,
"tokens_seen": 1879572480
},
{
"epoch": 0.57,
"learning_rate": 0.00021746238716148446,
"loss": 2.616,
"theoretical_loss": 3.448301308167748,
"tokens_seen": 1879638016
},
{
"epoch": 0.57,
"learning_rate": 0.00021745235707121364,
"loss": 2.3682,
"theoretical_loss": 3.448291161160463,
"tokens_seen": 1879703552
},
{
"epoch": 0.57,
"learning_rate": 0.00021744232698094282,
"loss": 2.7526,
"theoretical_loss": 3.4482810146060014,
"tokens_seen": 1879769088
},
{
"epoch": 0.57,
"learning_rate": 0.000217432296890672,
"loss": 2.5659,
"theoretical_loss": 3.4482708685043266,
"tokens_seen": 1879834624
},
{
"epoch": 0.57,
"learning_rate": 0.0002174222668004012,
"loss": 2.7551,
"theoretical_loss": 3.4482607228554034,
"tokens_seen": 1879900160
},
{
"epoch": 0.57,
"learning_rate": 0.0002174122367101304,
"loss": 2.6165,
"theoretical_loss": 3.448250577659196,
"tokens_seen": 1879965696
},
{
"epoch": 0.57,
"learning_rate": 0.00021740220661985957,
"loss": 2.5452,
"theoretical_loss": 3.448240432915667,
"tokens_seen": 1880031232
},
{
"epoch": 0.57,
"learning_rate": 0.00021739217652958875,
"loss": 2.4651,
"theoretical_loss": 3.4482302886247824,
"tokens_seen": 1880096768
},
{
"epoch": 0.57,
"learning_rate": 0.00021738214643931793,
"loss": 2.7005,
"theoretical_loss": 3.448220144786505,
"tokens_seen": 1880162304
},
{
"epoch": 0.57,
"learning_rate": 0.00021737211634904714,
"loss": 2.5214,
"theoretical_loss": 3.448210001400799,
"tokens_seen": 1880227840
},
{
"epoch": 0.57,
"learning_rate": 0.00021736208625877632,
"loss": 2.7531,
"theoretical_loss": 3.4481998584676288,
"tokens_seen": 1880293376
},
{
"epoch": 0.57,
"learning_rate": 0.0002173520561685055,
"loss": 2.6815,
"theoretical_loss": 3.448189715986958,
"tokens_seen": 1880358912
},
{
"epoch": 0.57,
"learning_rate": 0.0002173420260782347,
"loss": 2.3201,
"theoretical_loss": 3.4481795739587504,
"tokens_seen": 1880424448
},
{
"epoch": 0.57,
"learning_rate": 0.00021733199598796392,
"loss": 2.7044,
"theoretical_loss": 3.448169432382971,
"tokens_seen": 1880489984
},
{
"epoch": 0.57,
"learning_rate": 0.0002173219658976931,
"loss": 2.7794,
"theoretical_loss": 3.448159291259583,
"tokens_seen": 1880555520
},
{
"epoch": 0.57,
"learning_rate": 0.00021731193580742228,
"loss": 2.6676,
"theoretical_loss": 3.448149150588551,
"tokens_seen": 1880621056
},
{
"epoch": 0.57,
"learning_rate": 0.00021730190571715146,
"loss": 2.3755,
"theoretical_loss": 3.4481390103698386,
"tokens_seen": 1880686592
},
{
"epoch": 0.57,
"learning_rate": 0.00021729187562688065,
"loss": 2.4822,
"theoretical_loss": 3.44812887060341,
"tokens_seen": 1880752128
},
{
"epoch": 0.57,
"learning_rate": 0.00021728184553660985,
"loss": 2.383,
"theoretical_loss": 3.448118731289229,
"tokens_seen": 1880817664
},
{
"epoch": 0.57,
"objective/train/docs_used": 1063493,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8093149662017822,
"objective/train/theoretical_loss": 3.448108592427261,
"objective/train/tokens_used": 1901343200,
"theoretical_loss": 3.448108592427261,
"tokens_seen": 1880883200
},
{
"epoch": 0.57,
"learning_rate": 0.00021727181544633903,
"loss": 2.7095,
"theoretical_loss": 3.448108592427261,
"tokens_seen": 1880883200
},
{
"epoch": 0.57,
"learning_rate": 0.00021726178535606822,
"loss": 2.6368,
"theoretical_loss": 3.448098454017469,
"tokens_seen": 1880948736
},
{
"epoch": 0.57,
"learning_rate": 0.0002172517552657974,
"loss": 2.643,
"theoretical_loss": 3.4480883160598164,
"tokens_seen": 1881014272
},
{
"epoch": 0.57,
"learning_rate": 0.0002172417251755266,
"loss": 2.6968,
"theoretical_loss": 3.448078178554268,
"tokens_seen": 1881079808
},
{
"epoch": 0.57,
"learning_rate": 0.00021723169508525579,
"loss": 2.7765,
"theoretical_loss": 3.4480680415007887,
"tokens_seen": 1881145344
},
{
"epoch": 0.57,
"learning_rate": 0.00021722166499498497,
"loss": 2.8107,
"theoretical_loss": 3.4480579048993416,
"tokens_seen": 1881210880
},
{
"epoch": 0.57,
"learning_rate": 0.00021721163490471415,
"loss": 2.6418,
"theoretical_loss": 3.448047768749891,
"tokens_seen": 1881276416
},
{
"epoch": 0.57,
"learning_rate": 0.00021720160481444336,
"loss": 2.6596,
"theoretical_loss": 3.448037633052401,
"tokens_seen": 1881341952
},
{
"epoch": 0.57,
"learning_rate": 0.00021719157472417254,
"loss": 2.7855,
"theoretical_loss": 3.4480274978068355,
"tokens_seen": 1881407488
},
{
"epoch": 0.57,
"learning_rate": 0.00021718154463390172,
"loss": 2.453,
"theoretical_loss": 3.448017363013159,
"tokens_seen": 1881473024
},
{
"epoch": 0.57,
"learning_rate": 0.0002171715145436309,
"loss": 2.6132,
"theoretical_loss": 3.4480072286713357,
"tokens_seen": 1881538560
},
{
"epoch": 0.57,
"learning_rate": 0.00021716148445336008,
"loss": 2.3995,
"theoretical_loss": 3.4479970947813294,
"tokens_seen": 1881604096
},
{
"epoch": 0.57,
"learning_rate": 0.0002171514543630893,
"loss": 2.5719,
"theoretical_loss": 3.447986961343104,
"tokens_seen": 1881669632
},
{
"epoch": 0.57,
"learning_rate": 0.00021714142427281847,
"loss": 2.6297,
"theoretical_loss": 3.4479768283566243,
"tokens_seen": 1881735168
},
{
"epoch": 0.57,
"learning_rate": 0.00021713139418254765,
"loss": 2.817,
"theoretical_loss": 3.447966695821854,
"tokens_seen": 1881800704
},
{
"epoch": 0.57,
"learning_rate": 0.00021712136409227683,
"loss": 2.7329,
"theoretical_loss": 3.447956563738757,
"tokens_seen": 1881866240
},
{
"epoch": 0.57,
"learning_rate": 0.00021711133400200604,
"loss": 2.7317,
"theoretical_loss": 3.4479464321072983,
"tokens_seen": 1881931776
},
{
"epoch": 0.57,
"learning_rate": 0.00021710130391173522,
"loss": 2.6194,
"theoretical_loss": 3.4479363009274406,
"tokens_seen": 1881997312
},
{
"epoch": 0.57,
"learning_rate": 0.0002170912738214644,
"loss": 2.7887,
"theoretical_loss": 3.4479261701991497,
"tokens_seen": 1882062848
},
{
"epoch": 0.57,
"learning_rate": 0.00021708124373119358,
"loss": 2.6026,
"theoretical_loss": 3.4479160399223887,
"tokens_seen": 1882128384
},
{
"epoch": 0.57,
"learning_rate": 0.00021707121364092277,
"loss": 2.6513,
"theoretical_loss": 3.4479059100971217,
"tokens_seen": 1882193920
},
{
"epoch": 0.57,
"learning_rate": 0.00021706118355065197,
"loss": 2.6192,
"theoretical_loss": 3.4478957807233135,
"tokens_seen": 1882259456
},
{
"epoch": 0.57,
"learning_rate": 0.00021705115346038115,
"loss": 2.5819,
"theoretical_loss": 3.447885651800928,
"tokens_seen": 1882324992
},
{
"epoch": 0.57,
"learning_rate": 0.00021704112337011034,
"loss": 2.6251,
"theoretical_loss": 3.4478755233299294,
"tokens_seen": 1882390528
},
{
"epoch": 0.57,
"learning_rate": 0.00021703109327983952,
"loss": 2.3386,
"theoretical_loss": 3.4478653953102816,
"tokens_seen": 1882456064
},
{
"epoch": 0.57,
"objective/train/docs_used": 1064917,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.531264305114746,
"objective/train/theoretical_loss": 3.4478552677419487,
"objective/train/tokens_used": 1902981600,
"theoretical_loss": 3.4478552677419487,
"tokens_seen": 1882521600
},
{
"epoch": 0.57,
"learning_rate": 0.00021702106318956873,
"loss": 2.6108,
"theoretical_loss": 3.4478552677419487,
"tokens_seen": 1882521600
},
{
"epoch": 0.57,
"learning_rate": 0.0002170110330992979,
"loss": 2.4503,
"theoretical_loss": 3.4478451406248953,
"tokens_seen": 1882587136
},
{
"epoch": 0.57,
"learning_rate": 0.0002170010030090271,
"loss": 2.5997,
"theoretical_loss": 3.4478350139590854,
"tokens_seen": 1882652672
},
{
"epoch": 0.57,
"learning_rate": 0.00021699097291875627,
"loss": 2.5448,
"theoretical_loss": 3.447824887744483,
"tokens_seen": 1882718208
},
{
"epoch": 0.57,
"learning_rate": 0.00021698094282848545,
"loss": 2.7602,
"theoretical_loss": 3.447814761981053,
"tokens_seen": 1882783744
},
{
"epoch": 0.57,
"learning_rate": 0.00021697091273821466,
"loss": 2.8065,
"theoretical_loss": 3.447804636668759,
"tokens_seen": 1882849280
},
{
"epoch": 0.57,
"learning_rate": 0.00021696088264794384,
"loss": 2.5905,
"theoretical_loss": 3.4477945118075652,
"tokens_seen": 1882914816
},
{
"epoch": 0.57,
"learning_rate": 0.00021695085255767302,
"loss": 2.5945,
"theoretical_loss": 3.4477843873974354,
"tokens_seen": 1882980352
},
{
"epoch": 0.57,
"learning_rate": 0.0002169408224674022,
"loss": 2.6824,
"theoretical_loss": 3.447774263438335,
"tokens_seen": 1883045888
},
{
"epoch": 0.57,
"learning_rate": 0.0002169307923771314,
"loss": 2.5947,
"theoretical_loss": 3.447764139930227,
"tokens_seen": 1883111424
},
{
"epoch": 0.57,
"learning_rate": 0.0002169207622868606,
"loss": 2.4539,
"theoretical_loss": 3.4477540168730765,
"tokens_seen": 1883176960
},
{
"epoch": 0.57,
"learning_rate": 0.00021691073219658977,
"loss": 2.5726,
"theoretical_loss": 3.4477438942668472,
"tokens_seen": 1883242496
},
{
"epoch": 0.57,
"learning_rate": 0.00021690070210631895,
"loss": 2.813,
"theoretical_loss": 3.447733772111503,
"tokens_seen": 1883308032
},
{
"epoch": 0.57,
"learning_rate": 0.00021689067201604813,
"loss": 2.6855,
"theoretical_loss": 3.447723650407009,
"tokens_seen": 1883373568
},
{
"epoch": 0.57,
"learning_rate": 0.00021688064192577734,
"loss": 2.4717,
"theoretical_loss": 3.447713529153329,
"tokens_seen": 1883439104
},
{
"epoch": 0.57,
"learning_rate": 0.00021687061183550652,
"loss": 2.7349,
"theoretical_loss": 3.447703408350427,
"tokens_seen": 1883504640
},
{
"epoch": 0.57,
"learning_rate": 0.0002168605817452357,
"loss": 2.4598,
"theoretical_loss": 3.447693287998268,
"tokens_seen": 1883570176
},
{
"epoch": 0.57,
"learning_rate": 0.00021685055165496489,
"loss": 2.7098,
"theoretical_loss": 3.447683168096815,
"tokens_seen": 1883635712
},
{
"epoch": 0.57,
"learning_rate": 0.0002168405215646941,
"loss": 2.9311,
"theoretical_loss": 3.447673048646034,
"tokens_seen": 1883701248
},
{
"epoch": 0.57,
"learning_rate": 0.00021683049147442327,
"loss": 2.5946,
"theoretical_loss": 3.4476629296458876,
"tokens_seen": 1883766784
},
{
"epoch": 0.57,
"learning_rate": 0.00021682046138415246,
"loss": 2.622,
"theoretical_loss": 3.447652811096341,
"tokens_seen": 1883832320
},
{
"epoch": 0.57,
"learning_rate": 0.00021681043129388164,
"loss": 2.6559,
"theoretical_loss": 3.447642692997358,
"tokens_seen": 1883897856
},
{
"epoch": 0.57,
"learning_rate": 0.00021680040120361082,
"loss": 2.5528,
"theoretical_loss": 3.4476325753489028,
"tokens_seen": 1883963392
},
{
"epoch": 0.57,
"learning_rate": 0.00021679037111334003,
"loss": 2.6048,
"theoretical_loss": 3.4476224581509403,
"tokens_seen": 1884028928
},
{
"epoch": 0.57,
"learning_rate": 0.0002167803410230692,
"loss": 2.8006,
"theoretical_loss": 3.447612341403434,
"tokens_seen": 1884094464
},
{
"epoch": 0.57,
"objective/train/docs_used": 1065450,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.745108127593994,
"objective/train/theoretical_loss": 3.4476022251063485,
"objective/train/tokens_used": 1904620000,
"theoretical_loss": 3.4476022251063485,
"tokens_seen": 1884160000
},
{
"epoch": 0.57,
"learning_rate": 0.0002167703109327984,
"loss": 2.5509,
"theoretical_loss": 3.4476022251063485,
"tokens_seen": 1884160000
},
{
"epoch": 0.57,
"learning_rate": 0.00021676028084252757,
"loss": 2.6413,
"theoretical_loss": 3.447592109259648,
"tokens_seen": 1884225536
},
{
"epoch": 0.57,
"learning_rate": 0.00021675025075225678,
"loss": 2.5321,
"theoretical_loss": 3.447581993863297,
"tokens_seen": 1884291072
},
{
"epoch": 0.57,
"learning_rate": 0.00021674022066198596,
"loss": 2.443,
"theoretical_loss": 3.44757187891726,
"tokens_seen": 1884356608
},
{
"epoch": 0.57,
"learning_rate": 0.00021673019057171514,
"loss": 2.6347,
"theoretical_loss": 3.4475617644215006,
"tokens_seen": 1884422144
},
{
"epoch": 0.57,
"learning_rate": 0.00021672016048144432,
"loss": 2.6269,
"theoretical_loss": 3.447551650375984,
"tokens_seen": 1884487680
},
{
"epoch": 0.57,
"learning_rate": 0.0002167101303911735,
"loss": 2.649,
"theoretical_loss": 3.4475415367806734,
"tokens_seen": 1884553216
},
{
"epoch": 0.57,
"learning_rate": 0.0002167001003009027,
"loss": 2.7416,
"theoretical_loss": 3.4475314236355343,
"tokens_seen": 1884618752
},
{
"epoch": 0.57,
"learning_rate": 0.0002166900702106319,
"loss": 2.6171,
"theoretical_loss": 3.44752131094053,
"tokens_seen": 1884684288
},
{
"epoch": 0.57,
"learning_rate": 0.00021668004012036107,
"loss": 2.7631,
"theoretical_loss": 3.447511198695625,
"tokens_seen": 1884749824
},
{
"epoch": 0.57,
"learning_rate": 0.00021667001003009025,
"loss": 2.626,
"theoretical_loss": 3.4475010869007843,
"tokens_seen": 1884815360
},
{
"epoch": 0.57,
"learning_rate": 0.00021665997993981946,
"loss": 2.5877,
"theoretical_loss": 3.4474909755559713,
"tokens_seen": 1884880896
},
{
"epoch": 0.57,
"learning_rate": 0.00021664994984954864,
"loss": 2.617,
"theoretical_loss": 3.4474808646611512,
"tokens_seen": 1884946432
},
{
"epoch": 0.57,
"learning_rate": 0.00021663991975927782,
"loss": 2.7296,
"theoretical_loss": 3.4474707542162877,
"tokens_seen": 1885011968
},
{
"epoch": 0.57,
"learning_rate": 0.000216629889669007,
"loss": 2.5098,
"theoretical_loss": 3.447460644221345,
"tokens_seen": 1885077504
},
{
"epoch": 0.57,
"learning_rate": 0.00021661985957873621,
"loss": 2.4389,
"theoretical_loss": 3.447450534676288,
"tokens_seen": 1885143040
},
{
"epoch": 0.57,
"learning_rate": 0.0002166098294884654,
"loss": 2.4837,
"theoretical_loss": 3.447440425581081,
"tokens_seen": 1885208576
},
{
"epoch": 0.57,
"learning_rate": 0.00021659979939819458,
"loss": 2.7538,
"theoretical_loss": 3.447430316935688,
"tokens_seen": 1885274112
},
{
"epoch": 0.57,
"learning_rate": 0.00021658976930792378,
"loss": 2.6291,
"theoretical_loss": 3.4474202087400734,
"tokens_seen": 1885339648
},
{
"epoch": 0.57,
"learning_rate": 0.00021657973921765297,
"loss": 2.6019,
"theoretical_loss": 3.447410100994202,
"tokens_seen": 1885405184
},
{
"epoch": 0.57,
"learning_rate": 0.00021656970912738217,
"loss": 2.72,
"theoretical_loss": 3.4473999936980375,
"tokens_seen": 1885470720
},
{
"epoch": 0.57,
"learning_rate": 0.00021655967903711135,
"loss": 2.6107,
"theoretical_loss": 3.4473898868515445,
"tokens_seen": 1885536256
},
{
"epoch": 0.57,
"learning_rate": 0.00021654964894684054,
"loss": 2.7048,
"theoretical_loss": 3.447379780454688,
"tokens_seen": 1885601792
},
{
"epoch": 0.57,
"learning_rate": 0.00021653961885656972,
"loss": 2.602,
"theoretical_loss": 3.4473696745074314,
"tokens_seen": 1885667328
},
{
"epoch": 0.57,
"learning_rate": 0.00021652958876629893,
"loss": 2.5732,
"theoretical_loss": 3.4473595690097394,
"tokens_seen": 1885732864
},
{
"epoch": 0.57,
"objective/train/docs_used": 1066425,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5568153858184814,
"objective/train/theoretical_loss": 3.4473494639615767,
"objective/train/tokens_used": 1906258400,
"theoretical_loss": 3.4473494639615767,
"tokens_seen": 1885798400
},
{
"epoch": 0.57,
"learning_rate": 0.0002165195586760281,
"loss": 2.7062,
"theoretical_loss": 3.4473494639615767,
"tokens_seen": 1885798400
},
{
"epoch": 0.57,
"learning_rate": 0.0002165095285857573,
"loss": 2.6144,
"theoretical_loss": 3.447339359362907,
"tokens_seen": 1885863936
},
{
"epoch": 0.57,
"learning_rate": 0.00021649949849548647,
"loss": 2.93,
"theoretical_loss": 3.447329255213696,
"tokens_seen": 1885929472
},
{
"epoch": 0.57,
"learning_rate": 0.00021648946840521565,
"loss": 2.7454,
"theoretical_loss": 3.4473191515139066,
"tokens_seen": 1885995008
},
{
"epoch": 0.57,
"learning_rate": 0.00021647943831494486,
"loss": 2.5049,
"theoretical_loss": 3.447309048263504,
"tokens_seen": 1886060544
},
{
"epoch": 0.57,
"learning_rate": 0.00021646940822467404,
"loss": 2.5762,
"theoretical_loss": 3.4472989454624523,
"tokens_seen": 1886126080
},
{
"epoch": 0.57,
"learning_rate": 0.00021645937813440322,
"loss": 2.6243,
"theoretical_loss": 3.447288843110716,
"tokens_seen": 1886191616
},
{
"epoch": 0.57,
"learning_rate": 0.0002164493480441324,
"loss": 2.6785,
"theoretical_loss": 3.4472787412082595,
"tokens_seen": 1886257152
},
{
"epoch": 0.57,
"learning_rate": 0.0002164393179538616,
"loss": 2.708,
"theoretical_loss": 3.4472686397550474,
"tokens_seen": 1886322688
},
{
"epoch": 0.57,
"learning_rate": 0.0002164292878635908,
"loss": 2.4966,
"theoretical_loss": 3.4472585387510435,
"tokens_seen": 1886388224
},
{
"epoch": 0.57,
"learning_rate": 0.00021641925777331997,
"loss": 2.5584,
"theoretical_loss": 3.447248438196213,
"tokens_seen": 1886453760
},
{
"epoch": 0.57,
"learning_rate": 0.00021640922768304915,
"loss": 2.6684,
"theoretical_loss": 3.4472383380905205,
"tokens_seen": 1886519296
},
{
"epoch": 0.57,
"learning_rate": 0.00021639919759277833,
"loss": 2.7055,
"theoretical_loss": 3.4472282384339294,
"tokens_seen": 1886584832
},
{
"epoch": 0.57,
"learning_rate": 0.00021638916750250754,
"loss": 2.6325,
"theoretical_loss": 3.4472181392264045,
"tokens_seen": 1886650368
},
{
"epoch": 0.57,
"learning_rate": 0.00021637913741223672,
"loss": 2.6428,
"theoretical_loss": 3.4472080404679106,
"tokens_seen": 1886715904
},
{
"epoch": 0.57,
"learning_rate": 0.0002163691073219659,
"loss": 2.6892,
"theoretical_loss": 3.447197942158412,
"tokens_seen": 1886781440
},
{
"epoch": 0.57,
"learning_rate": 0.00021635907723169509,
"loss": 2.6753,
"theoretical_loss": 3.447187844297873,
"tokens_seen": 1886846976
},
{
"epoch": 0.57,
"learning_rate": 0.0002163490471414243,
"loss": 2.7087,
"theoretical_loss": 3.447177746886258,
"tokens_seen": 1886912512
},
{
"epoch": 0.57,
"learning_rate": 0.00021633901705115347,
"loss": 2.7144,
"theoretical_loss": 3.4471676499235318,
"tokens_seen": 1886978048
},
{
"epoch": 0.57,
"learning_rate": 0.00021632898696088266,
"loss": 2.6578,
"theoretical_loss": 3.4471575534096583,
"tokens_seen": 1887043584
},
{
"epoch": 0.57,
"learning_rate": 0.00021631895687061184,
"loss": 2.5652,
"theoretical_loss": 3.4471474573446024,
"tokens_seen": 1887109120
},
{
"epoch": 0.57,
"learning_rate": 0.00021630892678034102,
"loss": 2.7119,
"theoretical_loss": 3.4471373617283287,
"tokens_seen": 1887174656
},
{
"epoch": 0.57,
"learning_rate": 0.00021629889669007023,
"loss": 2.4014,
"theoretical_loss": 3.447127266560801,
"tokens_seen": 1887240192
},
{
"epoch": 0.57,
"learning_rate": 0.0002162888665997994,
"loss": 2.5625,
"theoretical_loss": 3.4471171718419846,
"tokens_seen": 1887305728
},
{
"epoch": 0.57,
"learning_rate": 0.0002162788365095286,
"loss": 2.3705,
"theoretical_loss": 3.447107077571843,
"tokens_seen": 1887371264
},
{
"epoch": 0.57,
"objective/train/docs_used": 1067085,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.912510871887207,
"objective/train/theoretical_loss": 3.447096983750342,
"objective/train/tokens_used": 1907896800,
"theoretical_loss": 3.447096983750342,
"tokens_seen": 1887436800
},
{
"epoch": 0.57,
"learning_rate": 0.00021626880641925777,
"loss": 2.7049,
"theoretical_loss": 3.447096983750342,
"tokens_seen": 1887436800
},
{
"epoch": 0.57,
"learning_rate": 0.00021625877632898698,
"loss": 2.7827,
"theoretical_loss": 3.447086890377445,
"tokens_seen": 1887502336
},
{
"epoch": 0.57,
"learning_rate": 0.00021624874623871616,
"loss": 2.7463,
"theoretical_loss": 3.447076797453117,
"tokens_seen": 1887567872
},
{
"epoch": 0.57,
"learning_rate": 0.00021623871614844534,
"loss": 2.8071,
"theoretical_loss": 3.4470667049773214,
"tokens_seen": 1887633408
},
{
"epoch": 0.57,
"learning_rate": 0.00021622868605817452,
"loss": 2.4508,
"theoretical_loss": 3.4470566129500244,
"tokens_seen": 1887698944
},
{
"epoch": 0.57,
"learning_rate": 0.0002162186559679037,
"loss": 2.3958,
"theoretical_loss": 3.4470465213711896,
"tokens_seen": 1887764480
},
{
"epoch": 0.57,
"learning_rate": 0.0002162086258776329,
"loss": 2.6698,
"theoretical_loss": 3.4470364302407814,
"tokens_seen": 1887830016
},
{
"epoch": 0.57,
"learning_rate": 0.0002161985957873621,
"loss": 2.6842,
"theoretical_loss": 3.447026339558765,
"tokens_seen": 1887895552
},
{
"epoch": 0.57,
"learning_rate": 0.00021618856569709127,
"loss": 2.6051,
"theoretical_loss": 3.4470162493251038,
"tokens_seen": 1887961088
},
{
"epoch": 0.57,
"learning_rate": 0.00021617853560682045,
"loss": 2.7245,
"theoretical_loss": 3.4470061595397627,
"tokens_seen": 1888026624
},
{
"epoch": 0.57,
"learning_rate": 0.00021616850551654966,
"loss": 2.5864,
"theoretical_loss": 3.446996070202707,
"tokens_seen": 1888092160
},
{
"epoch": 0.57,
"learning_rate": 0.00021615847542627884,
"loss": 2.7766,
"theoretical_loss": 3.446985981313901,
"tokens_seen": 1888157696
},
{
"epoch": 0.57,
"learning_rate": 0.00021614844533600802,
"loss": 2.691,
"theoretical_loss": 3.4469758928733083,
"tokens_seen": 1888223232
},
{
"epoch": 0.57,
"learning_rate": 0.0002161384152457372,
"loss": 2.3261,
"theoretical_loss": 3.446965804880894,
"tokens_seen": 1888288768
},
{
"epoch": 0.57,
"learning_rate": 0.00021612838515546641,
"loss": 2.5697,
"theoretical_loss": 3.446955717336623,
"tokens_seen": 1888354304
},
{
"epoch": 0.57,
"learning_rate": 0.0002161183550651956,
"loss": 2.7338,
"theoretical_loss": 3.4469456302404597,
"tokens_seen": 1888419840
},
{
"epoch": 0.57,
"learning_rate": 0.00021610832497492478,
"loss": 2.7841,
"theoretical_loss": 3.4469355435923683,
"tokens_seen": 1888485376
},
{
"epoch": 0.57,
"learning_rate": 0.00021609829488465396,
"loss": 2.5672,
"theoretical_loss": 3.4469254573923136,
"tokens_seen": 1888550912
},
{
"epoch": 0.57,
"learning_rate": 0.00021608826479438314,
"loss": 2.6312,
"theoretical_loss": 3.4469153716402596,
"tokens_seen": 1888616448
},
{
"epoch": 0.57,
"learning_rate": 0.00021607823470411235,
"loss": 2.3429,
"theoretical_loss": 3.4469052863361713,
"tokens_seen": 1888681984
},
{
"epoch": 0.57,
"learning_rate": 0.00021606820461384153,
"loss": 2.5676,
"theoretical_loss": 3.4468952014800136,
"tokens_seen": 1888747520
},
{
"epoch": 0.57,
"learning_rate": 0.0002160581745235707,
"loss": 2.7003,
"theoretical_loss": 3.4468851170717505,
"tokens_seen": 1888813056
},
{
"epoch": 0.57,
"learning_rate": 0.0002160481444332999,
"loss": 2.8304,
"theoretical_loss": 3.446875033111347,
"tokens_seen": 1888878592
},
{
"epoch": 0.57,
"learning_rate": 0.0002160381143430291,
"loss": 2.5788,
"theoretical_loss": 3.4468649495987678,
"tokens_seen": 1888944128
},
{
"epoch": 0.57,
"learning_rate": 0.00021602808425275828,
"loss": 2.6678,
"theoretical_loss": 3.4468548665339767,
"tokens_seen": 1889009664
},
{
"epoch": 0.57,
"objective/train/docs_used": 1068281,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5290398597717285,
"objective/train/theoretical_loss": 3.446844783916939,
"objective/train/tokens_used": 1909535200,
"theoretical_loss": 3.446844783916939,
"tokens_seen": 1889075200
},
{
"epoch": 0.57,
"learning_rate": 0.00021601805416248746,
"loss": 2.5256,
"theoretical_loss": 3.446844783916939,
"tokens_seen": 1889075200
},
{
"epoch": 0.57,
"learning_rate": 0.00021600802407221664,
"loss": 2.7287,
"theoretical_loss": 3.4468347017476186,
"tokens_seen": 1889140736
},
{
"epoch": 0.57,
"learning_rate": 0.00021599799398194582,
"loss": 2.7085,
"theoretical_loss": 3.446824620025981,
"tokens_seen": 1889206272
},
{
"epoch": 0.57,
"learning_rate": 0.00021598796389167503,
"loss": 2.3207,
"theoretical_loss": 3.44681453875199,
"tokens_seen": 1889271808
},
{
"epoch": 0.57,
"learning_rate": 0.0002159779338014042,
"loss": 2.5497,
"theoretical_loss": 3.4468044579256105,
"tokens_seen": 1889337344
},
{
"epoch": 0.57,
"learning_rate": 0.0002159679037111334,
"loss": 2.7185,
"theoretical_loss": 3.446794377546807,
"tokens_seen": 1889402880
},
{
"epoch": 0.57,
"learning_rate": 0.00021595787362086257,
"loss": 2.7444,
"theoretical_loss": 3.4467842976155447,
"tokens_seen": 1889468416
},
{
"epoch": 0.57,
"learning_rate": 0.00021594784353059178,
"loss": 2.6208,
"theoretical_loss": 3.4467742181317873,
"tokens_seen": 1889533952
},
{
"epoch": 0.57,
"learning_rate": 0.00021593781344032096,
"loss": 2.5835,
"theoretical_loss": 3.4467641390955,
"tokens_seen": 1889599488
},
{
"epoch": 0.57,
"learning_rate": 0.00021592778335005014,
"loss": 2.7902,
"theoretical_loss": 3.446754060506647,
"tokens_seen": 1889665024
},
{
"epoch": 0.57,
"learning_rate": 0.00021591775325977933,
"loss": 2.7069,
"theoretical_loss": 3.4467439823651933,
"tokens_seen": 1889730560
},
{
"epoch": 0.57,
"learning_rate": 0.0002159077231695085,
"loss": 2.6815,
"theoretical_loss": 3.4467339046711034,
"tokens_seen": 1889796096
},
{
"epoch": 0.57,
"learning_rate": 0.00021589769307923772,
"loss": 2.647,
"theoretical_loss": 3.446723827424342,
"tokens_seen": 1889861632
},
{
"epoch": 0.57,
"learning_rate": 0.0002158876629889669,
"loss": 2.712,
"theoretical_loss": 3.4467137506248733,
"tokens_seen": 1889927168
},
{
"epoch": 0.57,
"learning_rate": 0.00021587763289869608,
"loss": 2.6076,
"theoretical_loss": 3.446703674272663,
"tokens_seen": 1889992704
},
{
"epoch": 0.57,
"learning_rate": 0.00021586760280842526,
"loss": 2.8003,
"theoretical_loss": 3.4466935983676743,
"tokens_seen": 1890058240
},
{
"epoch": 0.57,
"learning_rate": 0.00021585757271815447,
"loss": 2.6029,
"theoretical_loss": 3.446683522909873,
"tokens_seen": 1890123776
},
{
"epoch": 0.57,
"learning_rate": 0.00021584754262788365,
"loss": 2.6568,
"theoretical_loss": 3.446673447899223,
"tokens_seen": 1890189312
},
{
"epoch": 0.57,
"learning_rate": 0.00021583751253761286,
"loss": 2.6344,
"theoretical_loss": 3.4466633733356895,
"tokens_seen": 1890254848
},
{
"epoch": 0.57,
"learning_rate": 0.00021582748244734204,
"loss": 2.5272,
"theoretical_loss": 3.4466532992192365,
"tokens_seen": 1890320384
},
{
"epoch": 0.57,
"learning_rate": 0.00021581745235707122,
"loss": 2.7733,
"theoretical_loss": 3.4466432255498294,
"tokens_seen": 1890385920
},
{
"epoch": 0.57,
"learning_rate": 0.00021580742226680043,
"loss": 2.5987,
"theoretical_loss": 3.4466331523274327,
"tokens_seen": 1890451456
},
{
"epoch": 0.57,
"learning_rate": 0.0002157973921765296,
"loss": 2.4512,
"theoretical_loss": 3.446623079552011,
"tokens_seen": 1890516992
},
{
"epoch": 0.57,
"learning_rate": 0.0002157873620862588,
"loss": 2.4659,
"theoretical_loss": 3.4466130072235286,
"tokens_seen": 1890582528
},
{
"epoch": 0.57,
"learning_rate": 0.00021577733199598797,
"loss": 2.7511,
"theoretical_loss": 3.4466029353419505,
"tokens_seen": 1890648064
},
{
"epoch": 0.57,
"objective/train/docs_used": 1069021,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4789316654205322,
"objective/train/theoretical_loss": 3.4465928639072416,
"objective/train/tokens_used": 1911173600,
"theoretical_loss": 3.4465928639072416,
"tokens_seen": 1890713600
},
{
"epoch": 0.57,
"learning_rate": 0.00021576730190571718,
"loss": 2.7608,
"theoretical_loss": 3.4465928639072416,
"tokens_seen": 1890713600
},
{
"epoch": 0.57,
"learning_rate": 0.00021575727181544636,
"loss": 2.5624,
"theoretical_loss": 3.446582792919366,
"tokens_seen": 1890779136
},
{
"epoch": 0.57,
"learning_rate": 0.00021574724172517554,
"loss": 2.7936,
"theoretical_loss": 3.446572722378289,
"tokens_seen": 1890844672
},
{
"epoch": 0.57,
"learning_rate": 0.00021573721163490472,
"loss": 2.6016,
"theoretical_loss": 3.4465626522839745,
"tokens_seen": 1890910208
},
{
"epoch": 0.57,
"learning_rate": 0.0002157271815446339,
"loss": 2.871,
"theoretical_loss": 3.4465525826363885,
"tokens_seen": 1890975744
},
{
"epoch": 0.57,
"learning_rate": 0.0002157171514543631,
"loss": 2.5364,
"theoretical_loss": 3.4465425134354946,
"tokens_seen": 1891041280
},
{
"epoch": 0.57,
"learning_rate": 0.0002157071213640923,
"loss": 2.7465,
"theoretical_loss": 3.4465324446812575,
"tokens_seen": 1891106816
},
{
"epoch": 0.57,
"learning_rate": 0.00021569709127382147,
"loss": 2.4934,
"theoretical_loss": 3.4465223763736423,
"tokens_seen": 1891172352
},
{
"epoch": 0.57,
"learning_rate": 0.00021568706118355065,
"loss": 2.7727,
"theoretical_loss": 3.446512308512614,
"tokens_seen": 1891237888
},
{
"epoch": 0.57,
"learning_rate": 0.00021567703109327986,
"loss": 2.5336,
"theoretical_loss": 3.446502241098137,
"tokens_seen": 1891303424
},
{
"epoch": 0.57,
"learning_rate": 0.00021566700100300904,
"loss": 2.7451,
"theoretical_loss": 3.4464921741301757,
"tokens_seen": 1891368960
},
{
"epoch": 0.57,
"learning_rate": 0.00021565697091273822,
"loss": 2.6754,
"theoretical_loss": 3.4464821076086953,
"tokens_seen": 1891434496
},
{
"epoch": 0.57,
"learning_rate": 0.0002156469408224674,
"loss": 2.6523,
"theoretical_loss": 3.44647204153366,
"tokens_seen": 1891500032
},
{
"epoch": 0.57,
"learning_rate": 0.00021563691073219661,
"loss": 2.3661,
"theoretical_loss": 3.4464619759050352,
"tokens_seen": 1891565568
},
{
"epoch": 0.57,
"learning_rate": 0.0002156268806419258,
"loss": 2.6188,
"theoretical_loss": 3.4464519107227853,
"tokens_seen": 1891631104
},
{
"epoch": 0.57,
"learning_rate": 0.00021561685055165498,
"loss": 2.6293,
"theoretical_loss": 3.4464418459868753,
"tokens_seen": 1891696640
},
{
"epoch": 0.57,
"learning_rate": 0.00021560682046138416,
"loss": 2.8587,
"theoretical_loss": 3.446431781697269,
"tokens_seen": 1891762176
},
{
"epoch": 0.57,
"learning_rate": 0.00021559679037111334,
"loss": 2.6351,
"theoretical_loss": 3.4464217178539327,
"tokens_seen": 1891827712
},
{
"epoch": 0.57,
"learning_rate": 0.00021558676028084255,
"loss": 2.5828,
"theoretical_loss": 3.4464116544568295,
"tokens_seen": 1891893248
},
{
"epoch": 0.57,
"learning_rate": 0.00021557673019057173,
"loss": 2.6861,
"theoretical_loss": 3.446401591505925,
"tokens_seen": 1891958784
},
{
"epoch": 0.57,
"learning_rate": 0.0002155667001003009,
"loss": 2.7369,
"theoretical_loss": 3.4463915290011844,
"tokens_seen": 1892024320
},
{
"epoch": 0.57,
"learning_rate": 0.0002155566700100301,
"loss": 2.6527,
"theoretical_loss": 3.446381466942572,
"tokens_seen": 1892089856
},
{
"epoch": 0.57,
"learning_rate": 0.0002155466399197593,
"loss": 2.6726,
"theoretical_loss": 3.4463714053300523,
"tokens_seen": 1892155392
},
{
"epoch": 0.57,
"learning_rate": 0.00021553660982948848,
"loss": 2.7658,
"theoretical_loss": 3.44636134416359,
"tokens_seen": 1892220928
},
{
"epoch": 0.57,
"learning_rate": 0.00021552657973921766,
"loss": 2.7552,
"theoretical_loss": 3.4463512834431507,
"tokens_seen": 1892286464
},
{
"epoch": 0.57,
"objective/train/docs_used": 1069644,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.889479637145996,
"objective/train/theoretical_loss": 3.446341223168698,
"objective/train/tokens_used": 1912812000,
"theoretical_loss": 3.446341223168698,
"tokens_seen": 1892352000
},
{
"epoch": 0.57,
"learning_rate": 0.00021551654964894684,
"loss": 2.8947,
"theoretical_loss": 3.446341223168698,
"tokens_seen": 1892352000
},
{
"epoch": 0.57,
"learning_rate": 0.00021550651955867602,
"loss": 2.6903,
"theoretical_loss": 3.446331163340198,
"tokens_seen": 1892417536
},
{
"epoch": 0.57,
"learning_rate": 0.00021549648946840523,
"loss": 2.4468,
"theoretical_loss": 3.446321103957615,
"tokens_seen": 1892483072
},
{
"epoch": 0.57,
"learning_rate": 0.0002154864593781344,
"loss": 2.6,
"theoretical_loss": 3.4463110450209133,
"tokens_seen": 1892548608
},
{
"epoch": 0.57,
"learning_rate": 0.0002154764292878636,
"loss": 2.8163,
"theoretical_loss": 3.4463009865300585,
"tokens_seen": 1892614144
},
{
"epoch": 0.57,
"learning_rate": 0.00021546639919759277,
"loss": 2.5806,
"theoretical_loss": 3.4462909284850145,
"tokens_seen": 1892679680
},
{
"epoch": 0.57,
"learning_rate": 0.00021545636910732198,
"loss": 2.4528,
"theoretical_loss": 3.4462808708857464,
"tokens_seen": 1892745216
},
{
"epoch": 0.57,
"learning_rate": 0.00021544633901705116,
"loss": 2.5868,
"theoretical_loss": 3.4462708137322196,
"tokens_seen": 1892810752
},
{
"epoch": 0.57,
"learning_rate": 0.00021543630892678034,
"loss": 2.6894,
"theoretical_loss": 3.446260757024398,
"tokens_seen": 1892876288
},
{
"epoch": 0.57,
"learning_rate": 0.00021542627883650953,
"loss": 2.6535,
"theoretical_loss": 3.4462507007622474,
"tokens_seen": 1892941824
},
{
"epoch": 0.57,
"learning_rate": 0.0002154162487462387,
"loss": 2.5524,
"theoretical_loss": 3.4462406449457315,
"tokens_seen": 1893007360
},
{
"epoch": 0.57,
"learning_rate": 0.00021540621865596792,
"loss": 2.6942,
"theoretical_loss": 3.4462305895748164,
"tokens_seen": 1893072896
},
{
"epoch": 0.57,
"learning_rate": 0.0002153961885656971,
"loss": 2.6711,
"theoretical_loss": 3.4462205346494654,
"tokens_seen": 1893138432
},
{
"epoch": 0.57,
"learning_rate": 0.00021538615847542628,
"loss": 2.7763,
"theoretical_loss": 3.446210480169645,
"tokens_seen": 1893203968
},
{
"epoch": 0.57,
"learning_rate": 0.00021537612838515546,
"loss": 2.4621,
"theoretical_loss": 3.446200426135319,
"tokens_seen": 1893269504
},
{
"epoch": 0.57,
"learning_rate": 0.00021536609829488467,
"loss": 2.4134,
"theoretical_loss": 3.4461903725464524,
"tokens_seen": 1893335040
},
{
"epoch": 0.57,
"learning_rate": 0.00021535606820461385,
"loss": 2.7736,
"theoretical_loss": 3.4461803194030103,
"tokens_seen": 1893400576
},
{
"epoch": 0.57,
"learning_rate": 0.00021534603811434303,
"loss": 2.4039,
"theoretical_loss": 3.446170266704957,
"tokens_seen": 1893466112
},
{
"epoch": 0.57,
"learning_rate": 0.0002153360080240722,
"loss": 2.7189,
"theoretical_loss": 3.446160214452258,
"tokens_seen": 1893531648
},
{
"epoch": 0.57,
"learning_rate": 0.0002153259779338014,
"loss": 2.682,
"theoretical_loss": 3.4461501626448783,
"tokens_seen": 1893597184
},
{
"epoch": 0.57,
"learning_rate": 0.0002153259779338014,
"loss": 2.5592,
"theoretical_loss": 3.446140111282782,
"tokens_seen": 1893662720
},
{
"epoch": 0.57,
"learning_rate": 0.0002153159478435306,
"loss": 2.6307,
"theoretical_loss": 3.4461300603659337,
"tokens_seen": 1893728256
},
{
"epoch": 0.57,
"learning_rate": 0.00021530591775325978,
"loss": 2.5822,
"theoretical_loss": 3.4461200098942997,
"tokens_seen": 1893793792
},
{
"epoch": 0.57,
"learning_rate": 0.00021529588766298896,
"loss": 2.646,
"theoretical_loss": 3.4461099598678437,
"tokens_seen": 1893859328
},
{
"epoch": 0.57,
"learning_rate": 0.00021528585757271814,
"loss": 2.6033,
"theoretical_loss": 3.446099910286531,
"tokens_seen": 1893924864
},
{
"epoch": 0.57,
"objective/train/docs_used": 1070823,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.963459014892578,
"objective/train/theoretical_loss": 3.446089861150326,
"objective/train/tokens_used": 1914450400,
"theoretical_loss": 3.446089861150326,
"tokens_seen": 1893990400
},
{
"epoch": 0.57,
"learning_rate": 0.00021527582748244735,
"loss": 2.7621,
"theoretical_loss": 3.446089861150326,
"tokens_seen": 1893990400
},
{
"epoch": 0.57,
"learning_rate": 0.00021526579739217653,
"loss": 2.6453,
"theoretical_loss": 3.4460798124591947,
"tokens_seen": 1894055936
},
{
"epoch": 0.57,
"learning_rate": 0.0002152557673019057,
"loss": 2.4124,
"theoretical_loss": 3.446069764213101,
"tokens_seen": 1894121472
},
{
"epoch": 0.57,
"learning_rate": 0.0002152457372116349,
"loss": 2.7328,
"theoretical_loss": 3.4460597164120097,
"tokens_seen": 1894187008
},
{
"epoch": 0.57,
"learning_rate": 0.00021523570712136408,
"loss": 2.7664,
"theoretical_loss": 3.4460496690558866,
"tokens_seen": 1894252544
},
{
"epoch": 0.57,
"learning_rate": 0.00021522567703109328,
"loss": 2.5827,
"theoretical_loss": 3.4460396221446956,
"tokens_seen": 1894318080
},
{
"epoch": 0.57,
"learning_rate": 0.00021521564694082247,
"loss": 2.8078,
"theoretical_loss": 3.4460295756784025,
"tokens_seen": 1894383616
},
{
"epoch": 0.57,
"learning_rate": 0.00021520561685055165,
"loss": 2.6463,
"theoretical_loss": 3.446019529656972,
"tokens_seen": 1894449152
},
{
"epoch": 0.57,
"learning_rate": 0.00021519558676028083,
"loss": 2.7165,
"theoretical_loss": 3.446009484080368,
"tokens_seen": 1894514688
},
{
"epoch": 0.57,
"learning_rate": 0.00021518555667001004,
"loss": 2.6043,
"theoretical_loss": 3.445999438948557,
"tokens_seen": 1894580224
},
{
"epoch": 0.57,
"learning_rate": 0.00021517552657973922,
"loss": 2.875,
"theoretical_loss": 3.4459893942615025,
"tokens_seen": 1894645760
},
{
"epoch": 0.57,
"learning_rate": 0.0002151654964894684,
"loss": 2.6594,
"theoretical_loss": 3.4459793500191704,
"tokens_seen": 1894711296
},
{
"epoch": 0.57,
"learning_rate": 0.00021515546639919758,
"loss": 2.6343,
"theoretical_loss": 3.4459693062215253,
"tokens_seen": 1894776832
},
{
"epoch": 0.57,
"learning_rate": 0.00021514543630892676,
"loss": 2.4448,
"theoretical_loss": 3.445959262868532,
"tokens_seen": 1894842368
},
{
"epoch": 0.57,
"learning_rate": 0.00021513540621865597,
"loss": 2.561,
"theoretical_loss": 3.4459492199601556,
"tokens_seen": 1894907904
},
{
"epoch": 0.57,
"learning_rate": 0.00021512537612838515,
"loss": 2.6671,
"theoretical_loss": 3.445939177496361,
"tokens_seen": 1894973440
},
{
"epoch": 0.57,
"learning_rate": 0.00021511534603811433,
"loss": 2.575,
"theoretical_loss": 3.445929135477113,
"tokens_seen": 1895038976
},
{
"epoch": 0.57,
"learning_rate": 0.0002151053159478435,
"loss": 2.539,
"theoretical_loss": 3.445919093902377,
"tokens_seen": 1895104512
},
{
"epoch": 0.57,
"learning_rate": 0.00021509528585757272,
"loss": 2.8692,
"theoretical_loss": 3.4459090527721177,
"tokens_seen": 1895170048
},
{
"epoch": 0.57,
"learning_rate": 0.00021508525576730193,
"loss": 2.5268,
"theoretical_loss": 3.4458990120863,
"tokens_seen": 1895235584
},
{
"epoch": 0.57,
"learning_rate": 0.0002150752256770311,
"loss": 2.7028,
"theoretical_loss": 3.445888971844888,
"tokens_seen": 1895301120
},
{
"epoch": 0.57,
"learning_rate": 0.0002150651955867603,
"loss": 2.6923,
"theoretical_loss": 3.4458789320478487,
"tokens_seen": 1895366656
},
{
"epoch": 0.57,
"learning_rate": 0.0002150551654964895,
"loss": 2.711,
"theoretical_loss": 3.4458688926951453,
"tokens_seen": 1895432192
},
{
"epoch": 0.57,
"learning_rate": 0.00021504513540621868,
"loss": 2.5318,
"theoretical_loss": 3.4458588537867434,
"tokens_seen": 1895497728
},
{
"epoch": 0.57,
"learning_rate": 0.00021503510531594786,
"loss": 2.711,
"theoretical_loss": 3.4458488153226083,
"tokens_seen": 1895563264
},
{
"epoch": 0.57,
"objective/train/docs_used": 1071554,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6675522327423096,
"objective/train/theoretical_loss": 3.445838777302704,
"objective/train/tokens_used": 1916088800,
"theoretical_loss": 3.445838777302704,
"tokens_seen": 1895628800
},
{
"epoch": 0.57,
"learning_rate": 0.00021502507522567704,
"loss": 2.6792,
"theoretical_loss": 3.445838777302704,
"tokens_seen": 1895628800
},
{
"epoch": 0.57,
"learning_rate": 0.00021501504513540622,
"loss": 2.8236,
"theoretical_loss": 3.4458287397269967,
"tokens_seen": 1895694336
},
{
"epoch": 0.57,
"learning_rate": 0.00021500501504513543,
"loss": 2.5094,
"theoretical_loss": 3.4458187025954503,
"tokens_seen": 1895759872
},
{
"epoch": 0.57,
"learning_rate": 0.0002149949849548646,
"loss": 2.7307,
"theoretical_loss": 3.4458086659080305,
"tokens_seen": 1895825408
},
{
"epoch": 0.57,
"learning_rate": 0.0002149849548645938,
"loss": 2.3415,
"theoretical_loss": 3.4457986296647025,
"tokens_seen": 1895890944
},
{
"epoch": 0.57,
"learning_rate": 0.00021497492477432297,
"loss": 2.4975,
"theoretical_loss": 3.4457885938654305,
"tokens_seen": 1895956480
},
{
"epoch": 0.57,
"learning_rate": 0.00021496489468405218,
"loss": 2.6484,
"theoretical_loss": 3.44577855851018,
"tokens_seen": 1896022016
},
{
"epoch": 0.57,
"learning_rate": 0.00021495486459378136,
"loss": 2.5239,
"theoretical_loss": 3.4457685235989155,
"tokens_seen": 1896087552
},
{
"epoch": 0.57,
"learning_rate": 0.00021494483450351055,
"loss": 2.6194,
"theoretical_loss": 3.445758489131603,
"tokens_seen": 1896153088
},
{
"epoch": 0.57,
"learning_rate": 0.00021493480441323973,
"loss": 2.579,
"theoretical_loss": 3.4457484551082063,
"tokens_seen": 1896218624
},
{
"epoch": 0.57,
"learning_rate": 0.0002149247743229689,
"loss": 2.5588,
"theoretical_loss": 3.4457384215286915,
"tokens_seen": 1896284160
},
{
"epoch": 0.57,
"learning_rate": 0.00021491474423269812,
"loss": 2.435,
"theoretical_loss": 3.4457283883930234,
"tokens_seen": 1896349696
},
{
"epoch": 0.57,
"learning_rate": 0.0002149047141424273,
"loss": 2.6189,
"theoretical_loss": 3.4457183557011666,
"tokens_seen": 1896415232
},
{
"epoch": 0.57,
"learning_rate": 0.00021489468405215648,
"loss": 2.6425,
"theoretical_loss": 3.4457083234530863,
"tokens_seen": 1896480768
},
{
"epoch": 0.57,
"learning_rate": 0.00021488465396188566,
"loss": 2.5472,
"theoretical_loss": 3.445698291648747,
"tokens_seen": 1896546304
},
{
"epoch": 0.57,
"learning_rate": 0.00021487462387161487,
"loss": 2.7181,
"theoretical_loss": 3.445688260288115,
"tokens_seen": 1896611840
},
{
"epoch": 0.57,
"learning_rate": 0.00021486459378134405,
"loss": 2.6659,
"theoretical_loss": 3.445678229371154,
"tokens_seen": 1896677376
},
{
"epoch": 0.57,
"learning_rate": 0.00021485456369107323,
"loss": 2.7429,
"theoretical_loss": 3.4456681988978306,
"tokens_seen": 1896742912
},
{
"epoch": 0.57,
"learning_rate": 0.0002148445336008024,
"loss": 2.7123,
"theoretical_loss": 3.4456581688681087,
"tokens_seen": 1896808448
},
{
"epoch": 0.57,
"learning_rate": 0.0002148345035105316,
"loss": 2.4629,
"theoretical_loss": 3.4456481392819533,
"tokens_seen": 1896873984
},
{
"epoch": 0.57,
"learning_rate": 0.0002148244734202608,
"loss": 2.8385,
"theoretical_loss": 3.44563811013933,
"tokens_seen": 1896939520
},
{
"epoch": 0.57,
"learning_rate": 0.00021481444332998998,
"loss": 2.6444,
"theoretical_loss": 3.445628081440203,
"tokens_seen": 1897005056
},
{
"epoch": 0.57,
"learning_rate": 0.00021480441323971916,
"loss": 2.2774,
"theoretical_loss": 3.445618053184539,
"tokens_seen": 1897070592
},
{
"epoch": 0.57,
"learning_rate": 0.00021479438314944834,
"loss": 2.5319,
"theoretical_loss": 3.4456080253723016,
"tokens_seen": 1897136128
},
{
"epoch": 0.57,
"learning_rate": 0.00021478435305917755,
"loss": 2.5702,
"theoretical_loss": 3.445597998003456,
"tokens_seen": 1897201664
},
{
"epoch": 0.57,
"objective/train/docs_used": 1072893,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7457027435302734,
"objective/train/theoretical_loss": 3.445587971077968,
"objective/train/tokens_used": 1917727200,
"theoretical_loss": 3.445587971077968,
"tokens_seen": 1897267200
},
{
"epoch": 0.57,
"learning_rate": 0.00021477432296890673,
"loss": 2.5353,
"theoretical_loss": 3.445587971077968,
"tokens_seen": 1897267200
},
{
"epoch": 0.57,
"learning_rate": 0.00021476429287863591,
"loss": 2.6932,
"theoretical_loss": 3.4455779445958026,
"tokens_seen": 1897332736
},
{
"epoch": 0.57,
"learning_rate": 0.0002147542627883651,
"loss": 2.5293,
"theoretical_loss": 3.4455679185569243,
"tokens_seen": 1897398272
},
{
"epoch": 0.57,
"learning_rate": 0.00021474423269809428,
"loss": 2.6731,
"theoretical_loss": 3.4455578929612987,
"tokens_seen": 1897463808
},
{
"epoch": 0.58,
"learning_rate": 0.00021473420260782348,
"loss": 2.5969,
"theoretical_loss": 3.44554786780889,
"tokens_seen": 1897529344
},
{
"epoch": 0.58,
"learning_rate": 0.00021472417251755267,
"loss": 2.5528,
"theoretical_loss": 3.4455378430996646,
"tokens_seen": 1897594880
},
{
"epoch": 0.58,
"learning_rate": 0.00021471414242728185,
"loss": 2.9111,
"theoretical_loss": 3.445527818833587,
"tokens_seen": 1897660416
},
{
"epoch": 0.58,
"learning_rate": 0.00021470411233701103,
"loss": 2.5313,
"theoretical_loss": 3.4455177950106215,
"tokens_seen": 1897725952
},
{
"epoch": 0.58,
"learning_rate": 0.00021469408224674024,
"loss": 2.6312,
"theoretical_loss": 3.445507771630735,
"tokens_seen": 1897791488
},
{
"epoch": 0.58,
"learning_rate": 0.00021468405215646942,
"loss": 2.5215,
"theoretical_loss": 3.4454977486938914,
"tokens_seen": 1897857024
},
{
"epoch": 0.58,
"learning_rate": 0.0002146740220661986,
"loss": 2.5944,
"theoretical_loss": 3.4454877262000556,
"tokens_seen": 1897922560
},
{
"epoch": 0.58,
"learning_rate": 0.00021466399197592778,
"loss": 2.5181,
"theoretical_loss": 3.4454777041491935,
"tokens_seen": 1897988096
},
{
"epoch": 0.58,
"learning_rate": 0.00021465396188565696,
"loss": 2.3947,
"theoretical_loss": 3.44546768254127,
"tokens_seen": 1898053632
},
{
"epoch": 0.58,
"learning_rate": 0.00021464393179538617,
"loss": 2.3678,
"theoretical_loss": 3.44545766137625,
"tokens_seen": 1898119168
},
{
"epoch": 0.58,
"learning_rate": 0.00021463390170511535,
"loss": 2.719,
"theoretical_loss": 3.4454476406540984,
"tokens_seen": 1898184704
},
{
"epoch": 0.58,
"learning_rate": 0.00021462387161484453,
"loss": 2.6305,
"theoretical_loss": 3.445437620374781,
"tokens_seen": 1898250240
},
{
"epoch": 0.58,
"learning_rate": 0.0002146138415245737,
"loss": 2.785,
"theoretical_loss": 3.445427600538263,
"tokens_seen": 1898315776
},
{
"epoch": 0.58,
"learning_rate": 0.00021460381143430292,
"loss": 2.6376,
"theoretical_loss": 3.4454175811445085,
"tokens_seen": 1898381312
},
{
"epoch": 0.58,
"learning_rate": 0.0002145937813440321,
"loss": 2.8473,
"theoretical_loss": 3.445407562193484,
"tokens_seen": 1898446848
},
{
"epoch": 0.58,
"learning_rate": 0.00021458375125376128,
"loss": 2.7108,
"theoretical_loss": 3.4453975436851536,
"tokens_seen": 1898512384
},
{
"epoch": 0.58,
"learning_rate": 0.00021457372116349046,
"loss": 2.7527,
"theoretical_loss": 3.445387525619483,
"tokens_seen": 1898577920
},
{
"epoch": 0.58,
"learning_rate": 0.00021456369107321964,
"loss": 2.46,
"theoretical_loss": 3.445377507996437,
"tokens_seen": 1898643456
},
{
"epoch": 0.58,
"learning_rate": 0.00021455366098294885,
"loss": 2.5179,
"theoretical_loss": 3.445367490815981,
"tokens_seen": 1898708992
},
{
"epoch": 0.58,
"learning_rate": 0.00021454363089267803,
"loss": 2.678,
"theoretical_loss": 3.4453574740780804,
"tokens_seen": 1898774528
},
{
"epoch": 0.58,
"learning_rate": 0.00021453360080240721,
"loss": 2.7227,
"theoretical_loss": 3.4453474577826997,
"tokens_seen": 1898840064
},
{
"epoch": 0.58,
"objective/train/docs_used": 1073632,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4570674896240234,
"objective/train/theoretical_loss": 3.445337441929805,
"objective/train/tokens_used": 1919365600,
"theoretical_loss": 3.445337441929805,
"tokens_seen": 1898905600
},
{
"epoch": 0.58,
"learning_rate": 0.0002145235707121364,
"loss": 2.3388,
"theoretical_loss": 3.445337441929805,
"tokens_seen": 1898905600
},
{
"epoch": 0.58,
"learning_rate": 0.0002145135406218656,
"loss": 2.5921,
"theoretical_loss": 3.445327426519361,
"tokens_seen": 1898971136
},
{
"epoch": 0.58,
"learning_rate": 0.00021450351053159479,
"loss": 2.3905,
"theoretical_loss": 3.4453174115513323,
"tokens_seen": 1899036672
},
{
"epoch": 0.58,
"learning_rate": 0.00021449348044132397,
"loss": 2.6531,
"theoretical_loss": 3.445307397025685,
"tokens_seen": 1899102208
},
{
"epoch": 0.58,
"learning_rate": 0.00021448345035105315,
"loss": 2.2795,
"theoretical_loss": 3.445297382942384,
"tokens_seen": 1899167744
},
{
"epoch": 0.58,
"learning_rate": 0.00021447342026078236,
"loss": 2.6415,
"theoretical_loss": 3.4452873693013943,
"tokens_seen": 1899233280
},
{
"epoch": 0.58,
"learning_rate": 0.00021446339017051154,
"loss": 2.4747,
"theoretical_loss": 3.4452773561026815,
"tokens_seen": 1899298816
},
{
"epoch": 0.58,
"learning_rate": 0.00021445336008024072,
"loss": 2.631,
"theoretical_loss": 3.4452673433462104,
"tokens_seen": 1899364352
},
{
"epoch": 0.58,
"learning_rate": 0.0002144433299899699,
"loss": 2.5389,
"theoretical_loss": 3.4452573310319465,
"tokens_seen": 1899429888
},
{
"epoch": 0.58,
"learning_rate": 0.00021443329989969908,
"loss": 2.6905,
"theoretical_loss": 3.4452473191598547,
"tokens_seen": 1899495424
},
{
"epoch": 0.58,
"learning_rate": 0.0002144232698094283,
"loss": 2.4616,
"theoretical_loss": 3.4452373077299003,
"tokens_seen": 1899560960
},
{
"epoch": 0.58,
"learning_rate": 0.00021441323971915747,
"loss": 2.573,
"theoretical_loss": 3.4452272967420488,
"tokens_seen": 1899626496
},
{
"epoch": 0.58,
"learning_rate": 0.00021440320962888665,
"loss": 2.6893,
"theoretical_loss": 3.4452172861962653,
"tokens_seen": 1899692032
},
{
"epoch": 0.58,
"learning_rate": 0.00021439317953861583,
"loss": 2.5357,
"theoretical_loss": 3.445207276092515,
"tokens_seen": 1899757568
},
{
"epoch": 0.58,
"learning_rate": 0.00021438314944834504,
"loss": 2.5614,
"theoretical_loss": 3.445197266430763,
"tokens_seen": 1899823104
},
{
"epoch": 0.58,
"learning_rate": 0.00021437311935807422,
"loss": 2.5317,
"theoretical_loss": 3.4451872572109745,
"tokens_seen": 1899888640
},
{
"epoch": 0.58,
"learning_rate": 0.0002143630892678034,
"loss": 2.6521,
"theoretical_loss": 3.445177248433115,
"tokens_seen": 1899954176
},
{
"epoch": 0.58,
"learning_rate": 0.00021435305917753258,
"loss": 2.7039,
"theoretical_loss": 3.44516724009715,
"tokens_seen": 1900019712
},
{
"epoch": 0.58,
"learning_rate": 0.00021434302908726176,
"loss": 2.6037,
"theoretical_loss": 3.445157232203044,
"tokens_seen": 1900085248
},
{
"epoch": 0.58,
"learning_rate": 0.000214332998996991,
"loss": 2.5037,
"theoretical_loss": 3.4451472247507624,
"tokens_seen": 1900150784
},
{
"epoch": 0.58,
"learning_rate": 0.00021432296890672018,
"loss": 2.5708,
"theoretical_loss": 3.445137217740271,
"tokens_seen": 1900216320
},
{
"epoch": 0.58,
"learning_rate": 0.00021431293881644936,
"loss": 2.6008,
"theoretical_loss": 3.4451272111715348,
"tokens_seen": 1900281856
},
{
"epoch": 0.58,
"learning_rate": 0.00021430290872617854,
"loss": 2.6741,
"theoretical_loss": 3.445117205044519,
"tokens_seen": 1900347392
},
{
"epoch": 0.58,
"learning_rate": 0.00021429287863590775,
"loss": 2.3551,
"theoretical_loss": 3.445107199359189,
"tokens_seen": 1900412928
},
{
"epoch": 0.58,
"learning_rate": 0.00021428284854563693,
"loss": 2.4442,
"theoretical_loss": 3.44509719411551,
"tokens_seen": 1900478464
},
{
"epoch": 0.58,
"objective/train/docs_used": 1074230,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.367048740386963,
"objective/train/theoretical_loss": 3.4450871893134467,
"objective/train/tokens_used": 1921004000,
"theoretical_loss": 3.4450871893134467,
"tokens_seen": 1900544000
},
{
"epoch": 0.58,
"learning_rate": 0.00021427281845536611,
"loss": 2.6514,
"theoretical_loss": 3.4450871893134467,
"tokens_seen": 1900544000
},
{
"epoch": 0.58,
"learning_rate": 0.0002142627883650953,
"loss": 2.4855,
"theoretical_loss": 3.4450771849529653,
"tokens_seen": 1900609536
},
{
"epoch": 0.58,
"learning_rate": 0.00021425275827482448,
"loss": 2.7554,
"theoretical_loss": 3.4450671810340308,
"tokens_seen": 1900675072
},
{
"epoch": 0.58,
"learning_rate": 0.00021424272818455368,
"loss": 2.4228,
"theoretical_loss": 3.4450571775566083,
"tokens_seen": 1900740608
},
{
"epoch": 0.58,
"learning_rate": 0.00021423269809428287,
"loss": 2.7856,
"theoretical_loss": 3.445047174520663,
"tokens_seen": 1900806144
},
{
"epoch": 0.58,
"learning_rate": 0.00021422266800401205,
"loss": 2.6768,
"theoretical_loss": 3.4450371719261605,
"tokens_seen": 1900871680
},
{
"epoch": 0.58,
"learning_rate": 0.00021421263791374123,
"loss": 2.7847,
"theoretical_loss": 3.4450271697730663,
"tokens_seen": 1900937216
},
{
"epoch": 0.58,
"learning_rate": 0.00021420260782347044,
"loss": 2.4339,
"theoretical_loss": 3.445017168061345,
"tokens_seen": 1901002752
},
{
"epoch": 0.58,
"learning_rate": 0.00021419257773319962,
"loss": 2.6061,
"theoretical_loss": 3.4450071667909623,
"tokens_seen": 1901068288
},
{
"epoch": 0.58,
"learning_rate": 0.0002141825476429288,
"loss": 2.8247,
"theoretical_loss": 3.4449971659618837,
"tokens_seen": 1901133824
},
{
"epoch": 0.58,
"learning_rate": 0.00021417251755265798,
"loss": 2.4088,
"theoretical_loss": 3.4449871655740743,
"tokens_seen": 1901199360
},
{
"epoch": 0.58,
"learning_rate": 0.00021416248746238716,
"loss": 2.5997,
"theoretical_loss": 3.444977165627499,
"tokens_seen": 1901264896
},
{
"epoch": 0.58,
"learning_rate": 0.00021415245737211637,
"loss": 2.6285,
"theoretical_loss": 3.444967166122124,
"tokens_seen": 1901330432
},
{
"epoch": 0.58,
"learning_rate": 0.00021414242728184555,
"loss": 2.6457,
"theoretical_loss": 3.444957167057914,
"tokens_seen": 1901395968
},
{
"epoch": 0.58,
"learning_rate": 0.00021413239719157473,
"loss": 2.65,
"theoretical_loss": 3.444947168434835,
"tokens_seen": 1901461504
},
{
"epoch": 0.58,
"learning_rate": 0.0002141223671013039,
"loss": 2.9453,
"theoretical_loss": 3.4449371702528513,
"tokens_seen": 1901527040
},
{
"epoch": 0.58,
"learning_rate": 0.00021411233701103312,
"loss": 2.6177,
"theoretical_loss": 3.444927172511929,
"tokens_seen": 1901592576
},
{
"epoch": 0.58,
"learning_rate": 0.0002141023069207623,
"loss": 2.867,
"theoretical_loss": 3.4449171752120336,
"tokens_seen": 1901658112
},
{
"epoch": 0.58,
"learning_rate": 0.00021409227683049148,
"loss": 2.6176,
"theoretical_loss": 3.4449071783531293,
"tokens_seen": 1901723648
},
{
"epoch": 0.58,
"learning_rate": 0.00021408224674022066,
"loss": 2.4655,
"theoretical_loss": 3.4448971819351826,
"tokens_seen": 1901789184
},
{
"epoch": 0.58,
"learning_rate": 0.00021407221664994984,
"loss": 2.6193,
"theoretical_loss": 3.4448871859581587,
"tokens_seen": 1901854720
},
{
"epoch": 0.58,
"learning_rate": 0.00021406218655967905,
"loss": 2.6457,
"theoretical_loss": 3.444877190422023,
"tokens_seen": 1901920256
},
{
"epoch": 0.58,
"learning_rate": 0.00021405215646940823,
"loss": 2.4966,
"theoretical_loss": 3.44486719532674,
"tokens_seen": 1901985792
},
{
"epoch": 0.58,
"learning_rate": 0.00021404212637913742,
"loss": 2.5276,
"theoretical_loss": 3.444857200672276,
"tokens_seen": 1902051328
},
{
"epoch": 0.58,
"learning_rate": 0.0002140320962888666,
"loss": 2.4945,
"theoretical_loss": 3.4448472064585958,
"tokens_seen": 1902116864
},
{
"epoch": 0.58,
"objective/train/docs_used": 1075669,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.364870548248291,
"objective/train/theoretical_loss": 3.4448372126856652,
"objective/train/tokens_used": 1922642400,
"theoretical_loss": 3.4448372126856652,
"tokens_seen": 1902182400
},
{
"epoch": 0.58,
"learning_rate": 0.0002140220661985958,
"loss": 2.7279,
"theoretical_loss": 3.4448372126856652,
"tokens_seen": 1902182400
},
{
"epoch": 0.58,
"learning_rate": 0.00021401203610832499,
"loss": 2.6218,
"theoretical_loss": 3.4448272193534493,
"tokens_seen": 1902247936
},
{
"epoch": 0.58,
"learning_rate": 0.00021400200601805417,
"loss": 2.4877,
"theoretical_loss": 3.4448172264619137,
"tokens_seen": 1902313472
},
{
"epoch": 0.58,
"learning_rate": 0.00021399197592778335,
"loss": 2.6891,
"theoretical_loss": 3.444807234011024,
"tokens_seen": 1902379008
},
{
"epoch": 0.58,
"learning_rate": 0.00021398194583751256,
"loss": 2.74,
"theoretical_loss": 3.4447972420007447,
"tokens_seen": 1902444544
},
{
"epoch": 0.58,
"learning_rate": 0.00021397191574724174,
"loss": 2.6143,
"theoretical_loss": 3.444787250431042,
"tokens_seen": 1902510080
},
{
"epoch": 0.58,
"learning_rate": 0.00021396188565697092,
"loss": 2.4681,
"theoretical_loss": 3.444777259301881,
"tokens_seen": 1902575616
},
{
"epoch": 0.58,
"learning_rate": 0.0002139518555667001,
"loss": 2.7252,
"theoretical_loss": 3.4447672686132274,
"tokens_seen": 1902641152
},
{
"epoch": 0.58,
"learning_rate": 0.00021394182547642928,
"loss": 2.5799,
"theoretical_loss": 3.4447572783650457,
"tokens_seen": 1902706688
},
{
"epoch": 0.58,
"learning_rate": 0.0002139317953861585,
"loss": 2.5661,
"theoretical_loss": 3.444747288557303,
"tokens_seen": 1902772224
},
{
"epoch": 0.58,
"learning_rate": 0.00021392176529588767,
"loss": 2.648,
"theoretical_loss": 3.444737299189963,
"tokens_seen": 1902837760
},
{
"epoch": 0.58,
"learning_rate": 0.00021391173520561685,
"loss": 2.4367,
"theoretical_loss": 3.444727310262992,
"tokens_seen": 1902903296
},
{
"epoch": 0.58,
"learning_rate": 0.00021390170511534603,
"loss": 2.4941,
"theoretical_loss": 3.444717321776355,
"tokens_seen": 1902968832
},
{
"epoch": 0.58,
"learning_rate": 0.00021389167502507524,
"loss": 2.7642,
"theoretical_loss": 3.444707333730018,
"tokens_seen": 1903034368
},
{
"epoch": 0.58,
"learning_rate": 0.00021388164493480442,
"loss": 2.4321,
"theoretical_loss": 3.444697346123946,
"tokens_seen": 1903099904
},
{
"epoch": 0.58,
"learning_rate": 0.0002138716148445336,
"loss": 2.4237,
"theoretical_loss": 3.444687358958104,
"tokens_seen": 1903165440
},
{
"epoch": 0.58,
"learning_rate": 0.00021386158475426278,
"loss": 2.8138,
"theoretical_loss": 3.4446773722324586,
"tokens_seen": 1903230976
},
{
"epoch": 0.58,
"learning_rate": 0.00021385155466399196,
"loss": 2.1796,
"theoretical_loss": 3.4446673859469747,
"tokens_seen": 1903296512
},
{
"epoch": 0.58,
"learning_rate": 0.00021384152457372117,
"loss": 2.7698,
"theoretical_loss": 3.444657400101617,
"tokens_seen": 1903362048
},
{
"epoch": 0.58,
"learning_rate": 0.00021383149448345035,
"loss": 2.5416,
"theoretical_loss": 3.444647414696352,
"tokens_seen": 1903427584
},
{
"epoch": 0.58,
"learning_rate": 0.00021382146439317954,
"loss": 2.5338,
"theoretical_loss": 3.444637429731145,
"tokens_seen": 1903493120
},
{
"epoch": 0.58,
"learning_rate": 0.00021381143430290872,
"loss": 2.6585,
"theoretical_loss": 3.4446274452059606,
"tokens_seen": 1903558656
},
{
"epoch": 0.58,
"learning_rate": 0.00021380140421263792,
"loss": 2.6385,
"theoretical_loss": 3.444617461120765,
"tokens_seen": 1903624192
},
{
"epoch": 0.58,
"learning_rate": 0.0002137913741223671,
"loss": 2.684,
"theoretical_loss": 3.444607477475524,
"tokens_seen": 1903689728
},
{
"epoch": 0.58,
"learning_rate": 0.0002137813440320963,
"loss": 2.6281,
"theoretical_loss": 3.4445974942702025,
"tokens_seen": 1903755264
},
{
"epoch": 0.58,
"objective/train/docs_used": 1076340,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8268487453460693,
"objective/train/theoretical_loss": 3.4445875115047655,
"objective/train/tokens_used": 1924280800,
"theoretical_loss": 3.4445875115047655,
"tokens_seen": 1903820800
},
{
"epoch": 0.58,
"learning_rate": 0.00021377131394182547,
"loss": 2.6086,
"theoretical_loss": 3.4445875115047655,
"tokens_seen": 1903820800
},
{
"epoch": 0.58,
"learning_rate": 0.00021376128385155465,
"loss": 2.4826,
"theoretical_loss": 3.4445775291791794,
"tokens_seen": 1903886336
},
{
"epoch": 0.58,
"learning_rate": 0.00021375125376128386,
"loss": 2.483,
"theoretical_loss": 3.444567547293409,
"tokens_seen": 1903951872
},
{
"epoch": 0.58,
"learning_rate": 0.00021374122367101304,
"loss": 2.6328,
"theoretical_loss": 3.444557565847421,
"tokens_seen": 1904017408
},
{
"epoch": 0.58,
"learning_rate": 0.00021373119358074222,
"loss": 2.7965,
"theoretical_loss": 3.444547584841179,
"tokens_seen": 1904082944
},
{
"epoch": 0.58,
"learning_rate": 0.0002137211634904714,
"loss": 2.5692,
"theoretical_loss": 3.4445376042746503,
"tokens_seen": 1904148480
},
{
"epoch": 0.58,
"learning_rate": 0.0002137111334002006,
"loss": 2.4523,
"theoretical_loss": 3.444527624147799,
"tokens_seen": 1904214016
},
{
"epoch": 0.58,
"learning_rate": 0.0002137011033099298,
"loss": 2.7821,
"theoretical_loss": 3.444517644460591,
"tokens_seen": 1904279552
},
{
"epoch": 0.58,
"learning_rate": 0.00021369107321965897,
"loss": 2.5096,
"theoretical_loss": 3.4445076652129925,
"tokens_seen": 1904345088
},
{
"epoch": 0.58,
"learning_rate": 0.00021368104312938815,
"loss": 2.5903,
"theoretical_loss": 3.4444976864049686,
"tokens_seen": 1904410624
},
{
"epoch": 0.58,
"learning_rate": 0.00021367101303911733,
"loss": 2.7164,
"theoretical_loss": 3.444487708036484,
"tokens_seen": 1904476160
},
{
"epoch": 0.58,
"learning_rate": 0.00021366098294884654,
"loss": 2.3924,
"theoretical_loss": 3.444477730107505,
"tokens_seen": 1904541696
},
{
"epoch": 0.58,
"learning_rate": 0.00021365095285857572,
"loss": 2.3113,
"theoretical_loss": 3.4444677526179976,
"tokens_seen": 1904607232
},
{
"epoch": 0.58,
"learning_rate": 0.0002136409227683049,
"loss": 2.8032,
"theoretical_loss": 3.444457775567927,
"tokens_seen": 1904672768
},
{
"epoch": 0.58,
"learning_rate": 0.00021363089267803409,
"loss": 2.4249,
"theoretical_loss": 3.444447798957258,
"tokens_seen": 1904738304
},
{
"epoch": 0.58,
"learning_rate": 0.0002136208625877633,
"loss": 2.6812,
"theoretical_loss": 3.4444378227859564,
"tokens_seen": 1904803840
},
{
"epoch": 0.58,
"learning_rate": 0.00021361083249749247,
"loss": 2.7563,
"theoretical_loss": 3.4444278470539884,
"tokens_seen": 1904869376
},
{
"epoch": 0.58,
"learning_rate": 0.00021360080240722166,
"loss": 2.5859,
"theoretical_loss": 3.4444178717613188,
"tokens_seen": 1904934912
},
{
"epoch": 0.58,
"learning_rate": 0.00021359077231695084,
"loss": 2.5424,
"theoretical_loss": 3.4444078969079133,
"tokens_seen": 1905000448
},
{
"epoch": 0.58,
"learning_rate": 0.00021358074222668004,
"loss": 2.3694,
"theoretical_loss": 3.4443979224937378,
"tokens_seen": 1905065984
},
{
"epoch": 0.58,
"learning_rate": 0.00021357071213640925,
"loss": 2.5943,
"theoretical_loss": 3.4443879485187576,
"tokens_seen": 1905131520
},
{
"epoch": 0.58,
"learning_rate": 0.00021356068204613843,
"loss": 2.6716,
"theoretical_loss": 3.4443779749829386,
"tokens_seen": 1905197056
},
{
"epoch": 0.58,
"learning_rate": 0.00021355065195586762,
"loss": 2.5875,
"theoretical_loss": 3.4443680018862457,
"tokens_seen": 1905262592
},
{
"epoch": 0.58,
"learning_rate": 0.0002135406218655968,
"loss": 2.6514,
"theoretical_loss": 3.4443580292286446,
"tokens_seen": 1905328128
},
{
"epoch": 0.58,
"learning_rate": 0.000213530591775326,
"loss": 2.5152,
"theoretical_loss": 3.4443480570101017,
"tokens_seen": 1905393664
},
{
"epoch": 0.58,
"objective/train/docs_used": 1077487,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5943763256073,
"objective/train/theoretical_loss": 3.444338085230582,
"objective/train/tokens_used": 1925919200,
"theoretical_loss": 3.444338085230582,
"tokens_seen": 1905459200
},
{
"epoch": 0.58,
"learning_rate": 0.00021352056168505519,
"loss": 2.7474,
"theoretical_loss": 3.444338085230582,
"tokens_seen": 1905459200
},
{
"epoch": 0.58,
"learning_rate": 0.00021351053159478437,
"loss": 2.5512,
"theoretical_loss": 3.44432811389005,
"tokens_seen": 1905524736
},
{
"epoch": 0.58,
"learning_rate": 0.00021350050150451355,
"loss": 2.5654,
"theoretical_loss": 3.444318142988473,
"tokens_seen": 1905590272
},
{
"epoch": 0.58,
"learning_rate": 0.00021349047141424276,
"loss": 2.6987,
"theoretical_loss": 3.444308172525816,
"tokens_seen": 1905655808
},
{
"epoch": 0.58,
"learning_rate": 0.00021348044132397194,
"loss": 2.6556,
"theoretical_loss": 3.444298202502044,
"tokens_seen": 1905721344
},
{
"epoch": 0.58,
"learning_rate": 0.00021347041123370112,
"loss": 2.5371,
"theoretical_loss": 3.444288232917123,
"tokens_seen": 1905786880
},
{
"epoch": 0.58,
"learning_rate": 0.0002134603811434303,
"loss": 2.5652,
"theoretical_loss": 3.444278263771019,
"tokens_seen": 1905852416
},
{
"epoch": 0.58,
"learning_rate": 0.00021345035105315948,
"loss": 2.4582,
"theoretical_loss": 3.444268295063697,
"tokens_seen": 1905917952
},
{
"epoch": 0.58,
"learning_rate": 0.0002134403209628887,
"loss": 2.5884,
"theoretical_loss": 3.444258326795123,
"tokens_seen": 1905983488
},
{
"epoch": 0.58,
"learning_rate": 0.00021343029087261787,
"loss": 2.6521,
"theoretical_loss": 3.444248358965263,
"tokens_seen": 1906049024
},
{
"epoch": 0.58,
"learning_rate": 0.00021342026078234705,
"loss": 2.6582,
"theoretical_loss": 3.444238391574081,
"tokens_seen": 1906114560
},
{
"epoch": 0.58,
"learning_rate": 0.00021341023069207623,
"loss": 2.6192,
"theoretical_loss": 3.4442284246215444,
"tokens_seen": 1906180096
},
{
"epoch": 0.58,
"learning_rate": 0.00021340020060180544,
"loss": 2.4541,
"theoretical_loss": 3.444218458107618,
"tokens_seen": 1906245632
},
{
"epoch": 0.58,
"learning_rate": 0.00021339017051153462,
"loss": 2.5991,
"theoretical_loss": 3.444208492032267,
"tokens_seen": 1906311168
},
{
"epoch": 0.58,
"learning_rate": 0.0002133801404212638,
"loss": 2.7728,
"theoretical_loss": 3.444198526395458,
"tokens_seen": 1906376704
},
{
"epoch": 0.58,
"learning_rate": 0.00021337011033099298,
"loss": 2.4254,
"theoretical_loss": 3.444188561197156,
"tokens_seen": 1906442240
},
{
"epoch": 0.58,
"learning_rate": 0.00021336008024072216,
"loss": 2.5069,
"theoretical_loss": 3.4441785964373266,
"tokens_seen": 1906507776
},
{
"epoch": 0.58,
"learning_rate": 0.00021335005015045137,
"loss": 2.4357,
"theoretical_loss": 3.444168632115936,
"tokens_seen": 1906573312
},
{
"epoch": 0.58,
"learning_rate": 0.00021334002006018055,
"loss": 2.6574,
"theoretical_loss": 3.444158668232949,
"tokens_seen": 1906638848
},
{
"epoch": 0.58,
"learning_rate": 0.00021332998996990974,
"loss": 2.6171,
"theoretical_loss": 3.444148704788332,
"tokens_seen": 1906704384
},
{
"epoch": 0.58,
"learning_rate": 0.00021331995987963892,
"loss": 2.7178,
"theoretical_loss": 3.44413874178205,
"tokens_seen": 1906769920
},
{
"epoch": 0.58,
"learning_rate": 0.00021330992978936812,
"loss": 2.644,
"theoretical_loss": 3.4441287792140693,
"tokens_seen": 1906835456
},
{
"epoch": 0.58,
"learning_rate": 0.0002132998996990973,
"loss": 2.6461,
"theoretical_loss": 3.4441188170843553,
"tokens_seen": 1906900992
},
{
"epoch": 0.58,
"learning_rate": 0.0002132898696088265,
"loss": 2.613,
"theoretical_loss": 3.444108855392874,
"tokens_seen": 1906966528
},
{
"epoch": 0.58,
"learning_rate": 0.00021327983951855567,
"loss": 2.5033,
"theoretical_loss": 3.44409889413959,
"tokens_seen": 1907032064
},
{
"epoch": 0.58,
"objective/train/docs_used": 1078017,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7640483379364014,
"objective/train/theoretical_loss": 3.4440889333244695,
"objective/train/tokens_used": 1927557600,
"theoretical_loss": 3.4440889333244695,
"tokens_seen": 1907097600
},
{
"epoch": 0.58,
"learning_rate": 0.00021326980942828485,
"loss": 2.8029,
"theoretical_loss": 3.4440889333244695,
"tokens_seen": 1907097600
},
{
"epoch": 0.58,
"learning_rate": 0.00021325977933801406,
"loss": 2.4926,
"theoretical_loss": 3.444078972947479,
"tokens_seen": 1907163136
},
{
"epoch": 0.58,
"learning_rate": 0.00021324974924774324,
"loss": 2.7415,
"theoretical_loss": 3.4440690130085825,
"tokens_seen": 1907228672
},
{
"epoch": 0.58,
"learning_rate": 0.00021323971915747242,
"loss": 2.6666,
"theoretical_loss": 3.4440590535077478,
"tokens_seen": 1907294208
},
{
"epoch": 0.58,
"learning_rate": 0.0002132296890672016,
"loss": 2.7434,
"theoretical_loss": 3.4440490944449387,
"tokens_seen": 1907359744
},
{
"epoch": 0.58,
"learning_rate": 0.0002132196589769308,
"loss": 2.6468,
"theoretical_loss": 3.444039135820122,
"tokens_seen": 1907425280
},
{
"epoch": 0.58,
"learning_rate": 0.00021320962888666,
"loss": 2.6527,
"theoretical_loss": 3.444029177633263,
"tokens_seen": 1907490816
},
{
"epoch": 0.58,
"learning_rate": 0.00021319959879638917,
"loss": 2.6047,
"theoretical_loss": 3.4440192198843276,
"tokens_seen": 1907556352
},
{
"epoch": 0.58,
"learning_rate": 0.00021318956870611835,
"loss": 2.7659,
"theoretical_loss": 3.4440092625732808,
"tokens_seen": 1907621888
},
{
"epoch": 0.58,
"learning_rate": 0.00021317953861584753,
"loss": 2.5031,
"theoretical_loss": 3.443999305700089,
"tokens_seen": 1907687424
},
{
"epoch": 0.58,
"learning_rate": 0.00021316950852557674,
"loss": 2.6909,
"theoretical_loss": 3.443989349264718,
"tokens_seen": 1907752960
},
{
"epoch": 0.58,
"learning_rate": 0.00021315947843530592,
"loss": 2.6333,
"theoretical_loss": 3.4439793932671323,
"tokens_seen": 1907818496
},
{
"epoch": 0.58,
"learning_rate": 0.0002131494483450351,
"loss": 2.6868,
"theoretical_loss": 3.4439694377073,
"tokens_seen": 1907884032
},
{
"epoch": 0.58,
"learning_rate": 0.00021313941825476429,
"loss": 2.8199,
"theoretical_loss": 3.4439594825851843,
"tokens_seen": 1907949568
},
{
"epoch": 0.58,
"learning_rate": 0.0002131293881644935,
"loss": 2.6437,
"theoretical_loss": 3.4439495279007524,
"tokens_seen": 1908015104
},
{
"epoch": 0.58,
"learning_rate": 0.00021311935807422267,
"loss": 2.6314,
"theoretical_loss": 3.443939573653969,
"tokens_seen": 1908080640
},
{
"epoch": 0.58,
"learning_rate": 0.00021310932798395186,
"loss": 2.5589,
"theoretical_loss": 3.443929619844801,
"tokens_seen": 1908146176
},
{
"epoch": 0.58,
"learning_rate": 0.00021309929789368104,
"loss": 2.8064,
"theoretical_loss": 3.4439196664732137,
"tokens_seen": 1908211712
},
{
"epoch": 0.58,
"learning_rate": 0.00021308926780341022,
"loss": 2.5146,
"theoretical_loss": 3.4439097135391723,
"tokens_seen": 1908277248
},
{
"epoch": 0.58,
"learning_rate": 0.00021307923771313943,
"loss": 2.5921,
"theoretical_loss": 3.443899761042643,
"tokens_seen": 1908342784
},
{
"epoch": 0.58,
"learning_rate": 0.0002130692076228686,
"loss": 2.6834,
"theoretical_loss": 3.4438898089835916,
"tokens_seen": 1908408320
},
{
"epoch": 0.58,
"learning_rate": 0.0002130591775325978,
"loss": 2.7295,
"theoretical_loss": 3.443879857361984,
"tokens_seen": 1908473856
},
{
"epoch": 0.58,
"learning_rate": 0.00021304914744232697,
"loss": 2.7089,
"theoretical_loss": 3.4438699061777855,
"tokens_seen": 1908539392
},
{
"epoch": 0.58,
"learning_rate": 0.00021303911735205618,
"loss": 2.5532,
"theoretical_loss": 3.4438599554309617,
"tokens_seen": 1908604928
},
{
"epoch": 0.58,
"learning_rate": 0.00021302908726178536,
"loss": 2.5672,
"theoretical_loss": 3.4438500051214787,
"tokens_seen": 1908670464
},
{
"epoch": 0.58,
"objective/train/docs_used": 1079511,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3556151390075684,
"objective/train/theoretical_loss": 3.4438400552493027,
"objective/train/tokens_used": 1929196000,
"theoretical_loss": 3.4438400552493027,
"tokens_seen": 1908736000
},
{
"epoch": 0.58,
"learning_rate": 0.00021301905717151454,
"loss": 2.4686,
"theoretical_loss": 3.4438400552493027,
"tokens_seen": 1908736000
},
{
"epoch": 0.58,
"learning_rate": 0.00021300902708124372,
"loss": 2.5573,
"theoretical_loss": 3.4438301058143987,
"tokens_seen": 1908801536
},
{
"epoch": 0.58,
"learning_rate": 0.0002129989969909729,
"loss": 2.5877,
"theoretical_loss": 3.443820156816733,
"tokens_seen": 1908867072
},
{
"epoch": 0.58,
"learning_rate": 0.0002129889669007021,
"loss": 2.9083,
"theoretical_loss": 3.443810208256271,
"tokens_seen": 1908932608
},
{
"epoch": 0.58,
"learning_rate": 0.0002129789368104313,
"loss": 2.5769,
"theoretical_loss": 3.4438002601329787,
"tokens_seen": 1908998144
},
{
"epoch": 0.58,
"learning_rate": 0.00021296890672016047,
"loss": 2.4724,
"theoretical_loss": 3.443790312446822,
"tokens_seen": 1909063680
},
{
"epoch": 0.58,
"learning_rate": 0.00021295887662988965,
"loss": 2.5524,
"theoretical_loss": 3.4437803651977665,
"tokens_seen": 1909129216
},
{
"epoch": 0.58,
"learning_rate": 0.00021294884653961886,
"loss": 2.733,
"theoretical_loss": 3.4437704183857782,
"tokens_seen": 1909194752
},
{
"epoch": 0.58,
"learning_rate": 0.00021293881644934804,
"loss": 2.8178,
"theoretical_loss": 3.443760472010822,
"tokens_seen": 1909260288
},
{
"epoch": 0.58,
"learning_rate": 0.00021292878635907722,
"loss": 2.9041,
"theoretical_loss": 3.443750526072865,
"tokens_seen": 1909325824
},
{
"epoch": 0.58,
"learning_rate": 0.0002129187562688064,
"loss": 2.7514,
"theoretical_loss": 3.4437405805718724,
"tokens_seen": 1909391360
},
{
"epoch": 0.58,
"learning_rate": 0.0002129087261785356,
"loss": 3.046,
"theoretical_loss": 3.4437306355078094,
"tokens_seen": 1909456896
},
{
"epoch": 0.58,
"learning_rate": 0.0002128986960882648,
"loss": 2.682,
"theoretical_loss": 3.4437206908806433,
"tokens_seen": 1909522432
},
{
"epoch": 0.58,
"learning_rate": 0.00021288866599799398,
"loss": 2.4857,
"theoretical_loss": 3.4437107466903383,
"tokens_seen": 1909587968
},
{
"epoch": 0.58,
"learning_rate": 0.00021287863590772316,
"loss": 2.4035,
"theoretical_loss": 3.443700802936861,
"tokens_seen": 1909653504
},
{
"epoch": 0.58,
"learning_rate": 0.00021286860581745234,
"loss": 2.7955,
"theoretical_loss": 3.443690859620178,
"tokens_seen": 1909719040
},
{
"epoch": 0.58,
"learning_rate": 0.00021285857572718155,
"loss": 2.5173,
"theoretical_loss": 3.443680916740253,
"tokens_seen": 1909784576
},
{
"epoch": 0.58,
"learning_rate": 0.00021284854563691073,
"loss": 2.5386,
"theoretical_loss": 3.443670974297054,
"tokens_seen": 1909850112
},
{
"epoch": 0.58,
"learning_rate": 0.0002128385155466399,
"loss": 2.4342,
"theoretical_loss": 3.443661032290546,
"tokens_seen": 1909915648
},
{
"epoch": 0.58,
"learning_rate": 0.00021282848545636912,
"loss": 2.7896,
"theoretical_loss": 3.4436510907206945,
"tokens_seen": 1909981184
},
{
"epoch": 0.58,
"learning_rate": 0.00021281845536609832,
"loss": 2.6076,
"theoretical_loss": 3.4436411495874655,
"tokens_seen": 1910046720
},
{
"epoch": 0.58,
"learning_rate": 0.0002128084252758275,
"loss": 2.484,
"theoretical_loss": 3.4436312088908254,
"tokens_seen": 1910112256
},
{
"epoch": 0.58,
"learning_rate": 0.0002127983951855567,
"loss": 2.4196,
"theoretical_loss": 3.4436212686307393,
"tokens_seen": 1910177792
},
{
"epoch": 0.58,
"learning_rate": 0.00021278836509528587,
"loss": 2.5409,
"theoretical_loss": 3.4436113288071732,
"tokens_seen": 1910243328
},
{
"epoch": 0.58,
"learning_rate": 0.00021277833500501505,
"loss": 2.6553,
"theoretical_loss": 3.4436013894200936,
"tokens_seen": 1910308864
},
{
"epoch": 0.58,
"objective/train/docs_used": 1080108,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.663429021835327,
"objective/train/theoretical_loss": 3.443591450469466,
"objective/train/tokens_used": 1930834400,
"theoretical_loss": 3.443591450469466,
"tokens_seen": 1910374400
},
{
"epoch": 0.58,
"learning_rate": 0.00021276830491474426,
"loss": 2.5688,
"theoretical_loss": 3.443591450469466,
"tokens_seen": 1910374400
},
{
"epoch": 0.58,
"learning_rate": 0.00021275827482447344,
"loss": 2.6583,
"theoretical_loss": 3.4435815119552555,
"tokens_seen": 1910439936
},
{
"epoch": 0.58,
"learning_rate": 0.00021274824473420262,
"loss": 2.6882,
"theoretical_loss": 3.443571573877429,
"tokens_seen": 1910505472
},
{
"epoch": 0.58,
"learning_rate": 0.0002127382146439318,
"loss": 2.8896,
"theoretical_loss": 3.443561636235952,
"tokens_seen": 1910571008
},
{
"epoch": 0.58,
"learning_rate": 0.000212728184553661,
"loss": 2.3332,
"theoretical_loss": 3.4435516990307904,
"tokens_seen": 1910636544
},
{
"epoch": 0.58,
"learning_rate": 0.0002127181544633902,
"loss": 2.6919,
"theoretical_loss": 3.44354176226191,
"tokens_seen": 1910702080
},
{
"epoch": 0.58,
"learning_rate": 0.00021270812437311937,
"loss": 2.523,
"theoretical_loss": 3.4435318259292766,
"tokens_seen": 1910767616
},
{
"epoch": 0.58,
"learning_rate": 0.00021269809428284855,
"loss": 2.4923,
"theoretical_loss": 3.443521890032856,
"tokens_seen": 1910833152
},
{
"epoch": 0.58,
"learning_rate": 0.00021268806419257773,
"loss": 2.6433,
"theoretical_loss": 3.4435119545726147,
"tokens_seen": 1910898688
},
{
"epoch": 0.58,
"learning_rate": 0.00021267803410230694,
"loss": 2.4966,
"theoretical_loss": 3.443502019548518,
"tokens_seen": 1910964224
},
{
"epoch": 0.58,
"learning_rate": 0.00021266800401203612,
"loss": 2.7161,
"theoretical_loss": 3.4434920849605324,
"tokens_seen": 1911029760
},
{
"epoch": 0.58,
"learning_rate": 0.0002126579739217653,
"loss": 2.7163,
"theoretical_loss": 3.443482150808623,
"tokens_seen": 1911095296
},
{
"epoch": 0.58,
"learning_rate": 0.00021264794383149449,
"loss": 2.5018,
"theoretical_loss": 3.443472217092756,
"tokens_seen": 1911160832
},
{
"epoch": 0.58,
"learning_rate": 0.0002126379137412237,
"loss": 2.4867,
"theoretical_loss": 3.443462283812898,
"tokens_seen": 1911226368
},
{
"epoch": 0.58,
"learning_rate": 0.00021262788365095287,
"loss": 2.5275,
"theoretical_loss": 3.4434523509690136,
"tokens_seen": 1911291904
},
{
"epoch": 0.58,
"learning_rate": 0.00021261785356068206,
"loss": 2.7492,
"theoretical_loss": 3.44344241856107,
"tokens_seen": 1911357440
},
{
"epoch": 0.58,
"learning_rate": 0.00021260782347041124,
"loss": 2.4163,
"theoretical_loss": 3.4434324865890322,
"tokens_seen": 1911422976
},
{
"epoch": 0.58,
"learning_rate": 0.00021259779338014042,
"loss": 2.6131,
"theoretical_loss": 3.4434225550528668,
"tokens_seen": 1911488512
},
{
"epoch": 0.58,
"learning_rate": 0.00021258776328986963,
"loss": 2.7027,
"theoretical_loss": 3.443412623952539,
"tokens_seen": 1911554048
},
{
"epoch": 0.58,
"learning_rate": 0.0002125777331995988,
"loss": 2.6788,
"theoretical_loss": 3.4434026932880153,
"tokens_seen": 1911619584
},
{
"epoch": 0.58,
"learning_rate": 0.000212567703109328,
"loss": 2.5378,
"theoretical_loss": 3.4433927630592613,
"tokens_seen": 1911685120
},
{
"epoch": 0.58,
"learning_rate": 0.00021255767301905717,
"loss": 2.5788,
"theoretical_loss": 3.4433828332662437,
"tokens_seen": 1911750656
},
{
"epoch": 0.58,
"learning_rate": 0.00021254764292878638,
"loss": 2.9122,
"theoretical_loss": 3.4433729039089274,
"tokens_seen": 1911816192
},
{
"epoch": 0.58,
"learning_rate": 0.00021253761283851556,
"loss": 2.6191,
"theoretical_loss": 3.443362974987279,
"tokens_seen": 1911881728
},
{
"epoch": 0.58,
"learning_rate": 0.00021252758274824474,
"loss": 2.5387,
"theoretical_loss": 3.4433530465012643,
"tokens_seen": 1911947264
},
{
"epoch": 0.58,
"objective/train/docs_used": 1081319,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.712489604949951,
"objective/train/theoretical_loss": 3.443343118450849,
"objective/train/tokens_used": 1932472800,
"theoretical_loss": 3.443343118450849,
"tokens_seen": 1912012800
},
{
"epoch": 0.58,
"learning_rate": 0.00021251755265797392,
"loss": 2.6657,
"theoretical_loss": 3.443343118450849,
"tokens_seen": 1912012800
},
{
"epoch": 0.58,
"learning_rate": 0.0002125075225677031,
"loss": 2.5009,
"theoretical_loss": 3.4433331908359994,
"tokens_seen": 1912078336
},
{
"epoch": 0.58,
"learning_rate": 0.0002124974924774323,
"loss": 2.515,
"theoretical_loss": 3.4433232636566817,
"tokens_seen": 1912143872
},
{
"epoch": 0.58,
"learning_rate": 0.0002124874623871615,
"loss": 2.6425,
"theoretical_loss": 3.443313336912861,
"tokens_seen": 1912209408
},
{
"epoch": 0.58,
"learning_rate": 0.00021247743229689067,
"loss": 2.4343,
"theoretical_loss": 3.443303410604504,
"tokens_seen": 1912274944
},
{
"epoch": 0.58,
"learning_rate": 0.00021246740220661985,
"loss": 2.7541,
"theoretical_loss": 3.4432934847315764,
"tokens_seen": 1912340480
},
{
"epoch": 0.58,
"learning_rate": 0.00021245737211634906,
"loss": 2.9595,
"theoretical_loss": 3.443283559294044,
"tokens_seen": 1912406016
},
{
"epoch": 0.58,
"learning_rate": 0.00021244734202607824,
"loss": 2.9222,
"theoretical_loss": 3.4432736342918737,
"tokens_seen": 1912471552
},
{
"epoch": 0.58,
"learning_rate": 0.00021243731193580742,
"loss": 2.613,
"theoretical_loss": 3.44326370972503,
"tokens_seen": 1912537088
},
{
"epoch": 0.58,
"learning_rate": 0.0002124272818455366,
"loss": 2.6733,
"theoretical_loss": 3.4432537855934804,
"tokens_seen": 1912602624
},
{
"epoch": 0.58,
"learning_rate": 0.0002124172517552658,
"loss": 2.5553,
"theoretical_loss": 3.44324386189719,
"tokens_seen": 1912668160
},
{
"epoch": 0.58,
"learning_rate": 0.000212407221664995,
"loss": 2.6132,
"theoretical_loss": 3.4432339386361246,
"tokens_seen": 1912733696
},
{
"epoch": 0.58,
"learning_rate": 0.00021239719157472418,
"loss": 2.8167,
"theoretical_loss": 3.4432240158102507,
"tokens_seen": 1912799232
},
{
"epoch": 0.58,
"learning_rate": 0.00021238716148445336,
"loss": 2.6137,
"theoretical_loss": 3.443214093419534,
"tokens_seen": 1912864768
},
{
"epoch": 0.58,
"learning_rate": 0.00021237713139418254,
"loss": 2.7531,
"theoretical_loss": 3.4432041714639414,
"tokens_seen": 1912930304
},
{
"epoch": 0.58,
"learning_rate": 0.00021236710130391175,
"loss": 2.588,
"theoretical_loss": 3.4431942499434376,
"tokens_seen": 1912995840
},
{
"epoch": 0.58,
"learning_rate": 0.00021235707121364093,
"loss": 2.6679,
"theoretical_loss": 3.4431843288579893,
"tokens_seen": 1913061376
},
{
"epoch": 0.58,
"learning_rate": 0.0002123470411233701,
"loss": 2.8744,
"theoretical_loss": 3.443174408207563,
"tokens_seen": 1913126912
},
{
"epoch": 0.58,
"learning_rate": 0.0002123370110330993,
"loss": 2.5778,
"theoretical_loss": 3.4431644879921235,
"tokens_seen": 1913192448
},
{
"epoch": 0.58,
"learning_rate": 0.0002123269809428285,
"loss": 2.4867,
"theoretical_loss": 3.4431545682116376,
"tokens_seen": 1913257984
},
{
"epoch": 0.58,
"learning_rate": 0.00021231695085255768,
"loss": 2.5355,
"theoretical_loss": 3.4431446488660713,
"tokens_seen": 1913323520
},
{
"epoch": 0.58,
"learning_rate": 0.00021230692076228686,
"loss": 2.8216,
"theoretical_loss": 3.44313472995539,
"tokens_seen": 1913389056
},
{
"epoch": 0.58,
"learning_rate": 0.00021229689067201604,
"loss": 2.6807,
"theoretical_loss": 3.443124811479561,
"tokens_seen": 1913454592
},
{
"epoch": 0.58,
"learning_rate": 0.00021228686058174522,
"loss": 2.6641,
"theoretical_loss": 3.44311489343855,
"tokens_seen": 1913520128
},
{
"epoch": 0.58,
"learning_rate": 0.00021227683049147443,
"loss": 2.3608,
"theoretical_loss": 3.443104975832322,
"tokens_seen": 1913585664
},
{
"epoch": 0.58,
"objective/train/docs_used": 1081978,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9853310585021973,
"objective/train/theoretical_loss": 3.4430950586608438,
"objective/train/tokens_used": 1934111200,
"theoretical_loss": 3.4430950586608438,
"tokens_seen": 1913651200
},
{
"epoch": 0.58,
"learning_rate": 0.0002122668004012036,
"loss": 2.7099,
"theoretical_loss": 3.4430950586608438,
"tokens_seen": 1913651200
},
{
"epoch": 0.58,
"learning_rate": 0.0002122567703109328,
"loss": 2.9018,
"theoretical_loss": 3.4430851419240818,
"tokens_seen": 1913716736
},
{
"epoch": 0.58,
"learning_rate": 0.00021224674022066197,
"loss": 2.6578,
"theoretical_loss": 3.4430752256220014,
"tokens_seen": 1913782272
},
{
"epoch": 0.58,
"learning_rate": 0.00021223671013039118,
"loss": 2.5362,
"theoretical_loss": 3.443065309754569,
"tokens_seen": 1913847808
},
{
"epoch": 0.58,
"learning_rate": 0.00021222668004012036,
"loss": 2.6149,
"theoretical_loss": 3.4430553943217506,
"tokens_seen": 1913913344
},
{
"epoch": 0.58,
"learning_rate": 0.00021221664994984954,
"loss": 2.7666,
"theoretical_loss": 3.443045479323512,
"tokens_seen": 1913978880
},
{
"epoch": 0.58,
"learning_rate": 0.00021220661985957873,
"loss": 2.4946,
"theoretical_loss": 3.4430355647598194,
"tokens_seen": 1914044416
},
{
"epoch": 0.58,
"learning_rate": 0.0002121965897693079,
"loss": 2.4031,
"theoretical_loss": 3.4430256506306396,
"tokens_seen": 1914109952
},
{
"epoch": 0.58,
"learning_rate": 0.00021218655967903712,
"loss": 2.8009,
"theoretical_loss": 3.443015736935938,
"tokens_seen": 1914175488
},
{
"epoch": 0.58,
"learning_rate": 0.0002121765295887663,
"loss": 2.829,
"theoretical_loss": 3.4430058236756804,
"tokens_seen": 1914241024
},
{
"epoch": 0.58,
"learning_rate": 0.00021216649949849548,
"loss": 2.6284,
"theoretical_loss": 3.4429959108498336,
"tokens_seen": 1914306560
},
{
"epoch": 0.58,
"learning_rate": 0.00021215646940822466,
"loss": 2.5295,
"theoretical_loss": 3.4429859984583633,
"tokens_seen": 1914372096
},
{
"epoch": 0.58,
"learning_rate": 0.00021214643931795387,
"loss": 2.6139,
"theoretical_loss": 3.4429760865012353,
"tokens_seen": 1914437632
},
{
"epoch": 0.58,
"learning_rate": 0.00021213640922768305,
"loss": 2.6313,
"theoretical_loss": 3.442966174978416,
"tokens_seen": 1914503168
},
{
"epoch": 0.58,
"learning_rate": 0.00021212637913741223,
"loss": 2.4945,
"theoretical_loss": 3.4429562638898723,
"tokens_seen": 1914568704
},
{
"epoch": 0.58,
"learning_rate": 0.0002121163490471414,
"loss": 2.495,
"theoretical_loss": 3.442946353235569,
"tokens_seen": 1914634240
},
{
"epoch": 0.58,
"learning_rate": 0.0002121063189568706,
"loss": 2.7265,
"theoretical_loss": 3.4429364430154727,
"tokens_seen": 1914699776
},
{
"epoch": 0.58,
"learning_rate": 0.0002120962888665998,
"loss": 2.4907,
"theoretical_loss": 3.44292653322955,
"tokens_seen": 1914765312
},
{
"epoch": 0.58,
"learning_rate": 0.000212086258776329,
"loss": 2.7387,
"theoretical_loss": 3.4429166238777658,
"tokens_seen": 1914830848
},
{
"epoch": 0.58,
"learning_rate": 0.0002120762286860582,
"loss": 2.7192,
"theoretical_loss": 3.442906714960088,
"tokens_seen": 1914896384
},
{
"epoch": 0.58,
"learning_rate": 0.00021206619859578737,
"loss": 2.762,
"theoretical_loss": 3.4428968064764813,
"tokens_seen": 1914961920
},
{
"epoch": 0.58,
"learning_rate": 0.00021205616850551658,
"loss": 2.6507,
"theoretical_loss": 3.442886898426912,
"tokens_seen": 1915027456
},
{
"epoch": 0.58,
"learning_rate": 0.00021204613841524576,
"loss": 2.494,
"theoretical_loss": 3.442876990811347,
"tokens_seen": 1915092992
},
{
"epoch": 0.58,
"learning_rate": 0.00021203610832497494,
"loss": 2.638,
"theoretical_loss": 3.4428670836297517,
"tokens_seen": 1915158528
},
{
"epoch": 0.58,
"learning_rate": 0.00021202607823470412,
"loss": 2.6548,
"theoretical_loss": 3.442857176882093,
"tokens_seen": 1915224064
},
{
"epoch": 0.58,
"objective/train/docs_used": 1082665,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5702579021453857,
"objective/train/theoretical_loss": 3.442847270568336,
"objective/train/tokens_used": 1935749600,
"theoretical_loss": 3.442847270568336,
"tokens_seen": 1915289600
},
{
"epoch": 0.58,
"learning_rate": 0.0002120160481444333,
"loss": 2.6248,
"theoretical_loss": 3.442847270568336,
"tokens_seen": 1915289600
},
{
"epoch": 0.58,
"learning_rate": 0.0002120060180541625,
"loss": 2.7853,
"theoretical_loss": 3.4428373646884474,
"tokens_seen": 1915355136
},
{
"epoch": 0.58,
"learning_rate": 0.0002119959879638917,
"loss": 2.5953,
"theoretical_loss": 3.4428274592423938,
"tokens_seen": 1915420672
},
{
"epoch": 0.58,
"learning_rate": 0.00021198595787362087,
"loss": 2.6399,
"theoretical_loss": 3.4428175542301407,
"tokens_seen": 1915486208
},
{
"epoch": 0.58,
"learning_rate": 0.00021197592778335005,
"loss": 2.7868,
"theoretical_loss": 3.442807649651654,
"tokens_seen": 1915551744
},
{
"epoch": 0.58,
"learning_rate": 0.00021196589769307926,
"loss": 2.576,
"theoretical_loss": 3.4427977455069008,
"tokens_seen": 1915617280
},
{
"epoch": 0.58,
"learning_rate": 0.00021195586760280844,
"loss": 2.6254,
"theoretical_loss": 3.4427878417958464,
"tokens_seen": 1915682816
},
{
"epoch": 0.58,
"learning_rate": 0.00021194583751253762,
"loss": 2.6032,
"theoretical_loss": 3.442777938518458,
"tokens_seen": 1915748352
},
{
"epoch": 0.58,
"learning_rate": 0.0002119358074222668,
"loss": 2.6235,
"theoretical_loss": 3.4427680356747006,
"tokens_seen": 1915813888
},
{
"epoch": 0.58,
"learning_rate": 0.000211925777331996,
"loss": 2.6499,
"theoretical_loss": 3.442758133264541,
"tokens_seen": 1915879424
},
{
"epoch": 0.58,
"learning_rate": 0.0002119157472417252,
"loss": 2.6543,
"theoretical_loss": 3.442748231287946,
"tokens_seen": 1915944960
},
{
"epoch": 0.58,
"learning_rate": 0.00021190571715145438,
"loss": 2.5973,
"theoretical_loss": 3.4427383297448806,
"tokens_seen": 1916010496
},
{
"epoch": 0.58,
"learning_rate": 0.00021189568706118356,
"loss": 2.5777,
"theoretical_loss": 3.4427284286353115,
"tokens_seen": 1916076032
},
{
"epoch": 0.58,
"learning_rate": 0.00021188565697091274,
"loss": 2.6049,
"theoretical_loss": 3.442718527959205,
"tokens_seen": 1916141568
},
{
"epoch": 0.58,
"learning_rate": 0.00021187562688064195,
"loss": 2.4813,
"theoretical_loss": 3.4427086277165273,
"tokens_seen": 1916207104
},
{
"epoch": 0.58,
"learning_rate": 0.00021186559679037113,
"loss": 2.5412,
"theoretical_loss": 3.442698727907244,
"tokens_seen": 1916272640
},
{
"epoch": 0.58,
"learning_rate": 0.0002118555667001003,
"loss": 2.808,
"theoretical_loss": 3.442688828531322,
"tokens_seen": 1916338176
},
{
"epoch": 0.58,
"learning_rate": 0.0002118455366098295,
"loss": 2.6268,
"theoretical_loss": 3.4426789295887277,
"tokens_seen": 1916403712
},
{
"epoch": 0.58,
"learning_rate": 0.0002118355065195587,
"loss": 2.6964,
"theoretical_loss": 3.4426690310794266,
"tokens_seen": 1916469248
},
{
"epoch": 0.58,
"learning_rate": 0.00021182547642928788,
"loss": 2.5691,
"theoretical_loss": 3.442659133003385,
"tokens_seen": 1916534784
},
{
"epoch": 0.58,
"learning_rate": 0.00021181544633901706,
"loss": 2.6807,
"theoretical_loss": 3.4426492353605695,
"tokens_seen": 1916600320
},
{
"epoch": 0.58,
"learning_rate": 0.00021180541624874624,
"loss": 2.5551,
"theoretical_loss": 3.442639338150946,
"tokens_seen": 1916665856
},
{
"epoch": 0.58,
"learning_rate": 0.00021179538615847542,
"loss": 2.2766,
"theoretical_loss": 3.4426294413744816,
"tokens_seen": 1916731392
},
{
"epoch": 0.58,
"learning_rate": 0.00021178535606820463,
"loss": 2.5019,
"theoretical_loss": 3.4426195450311416,
"tokens_seen": 1916796928
},
{
"epoch": 0.58,
"learning_rate": 0.0002117753259779338,
"loss": 2.5378,
"theoretical_loss": 3.442609649120892,
"tokens_seen": 1916862464
},
{
"epoch": 0.58,
"objective/train/docs_used": 1083753,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.763125419616699,
"objective/train/theoretical_loss": 3.4425997536437,
"objective/train/tokens_used": 1937388000,
"theoretical_loss": 3.4425997536437,
"tokens_seen": 1916928000
},
{
"epoch": 0.58,
"learning_rate": 0.000211765295887663,
"loss": 2.6418,
"theoretical_loss": 3.4425997536437,
"tokens_seen": 1916928000
},
{
"epoch": 0.58,
"learning_rate": 0.00021175526579739217,
"loss": 2.6895,
"theoretical_loss": 3.4425898585995314,
"tokens_seen": 1916993536
},
{
"epoch": 0.58,
"learning_rate": 0.00021174523570712138,
"loss": 2.6179,
"theoretical_loss": 3.4425799639883525,
"tokens_seen": 1917059072
},
{
"epoch": 0.58,
"learning_rate": 0.00021173520561685056,
"loss": 2.4765,
"theoretical_loss": 3.442570069810129,
"tokens_seen": 1917124608
},
{
"epoch": 0.58,
"learning_rate": 0.00021172517552657974,
"loss": 2.5905,
"theoretical_loss": 3.442560176064828,
"tokens_seen": 1917190144
},
{
"epoch": 0.58,
"learning_rate": 0.00021171514543630893,
"loss": 2.5709,
"theoretical_loss": 3.442550282752415,
"tokens_seen": 1917255680
},
{
"epoch": 0.58,
"learning_rate": 0.0002117051153460381,
"loss": 2.7355,
"theoretical_loss": 3.4425403898728573,
"tokens_seen": 1917321216
},
{
"epoch": 0.58,
"learning_rate": 0.00021169508525576732,
"loss": 2.5111,
"theoretical_loss": 3.44253049742612,
"tokens_seen": 1917386752
},
{
"epoch": 0.58,
"learning_rate": 0.0002116850551654965,
"loss": 2.8182,
"theoretical_loss": 3.44252060541217,
"tokens_seen": 1917452288
},
{
"epoch": 0.58,
"learning_rate": 0.00021167502507522568,
"loss": 2.6781,
"theoretical_loss": 3.4425107138309734,
"tokens_seen": 1917517824
},
{
"epoch": 0.58,
"learning_rate": 0.00021166499498495486,
"loss": 2.6615,
"theoretical_loss": 3.4425008226824967,
"tokens_seen": 1917583360
},
{
"epoch": 0.58,
"learning_rate": 0.00021165496489468407,
"loss": 2.9214,
"theoretical_loss": 3.442490931966706,
"tokens_seen": 1917648896
},
{
"epoch": 0.58,
"learning_rate": 0.00021164493480441325,
"loss": 2.7825,
"theoretical_loss": 3.4424810416835676,
"tokens_seen": 1917714432
},
{
"epoch": 0.58,
"learning_rate": 0.00021163490471414243,
"loss": 2.2958,
"theoretical_loss": 3.4424711518330477,
"tokens_seen": 1917779968
},
{
"epoch": 0.58,
"learning_rate": 0.0002116248746238716,
"loss": 2.5675,
"theoretical_loss": 3.442461262415113,
"tokens_seen": 1917845504
},
{
"epoch": 0.58,
"learning_rate": 0.0002116148445336008,
"loss": 2.5826,
"theoretical_loss": 3.442451373429729,
"tokens_seen": 1917911040
},
{
"epoch": 0.58,
"learning_rate": 0.00021160481444333,
"loss": 2.6765,
"theoretical_loss": 3.4424414848768627,
"tokens_seen": 1917976576
},
{
"epoch": 0.58,
"learning_rate": 0.00021159478435305918,
"loss": 2.6829,
"theoretical_loss": 3.4424315967564807,
"tokens_seen": 1918042112
},
{
"epoch": 0.58,
"learning_rate": 0.00021158475426278836,
"loss": 2.7057,
"theoretical_loss": 3.4424217090685483,
"tokens_seen": 1918107648
},
{
"epoch": 0.58,
"learning_rate": 0.00021157472417251754,
"loss": 2.5116,
"theoretical_loss": 3.4424118218130326,
"tokens_seen": 1918173184
},
{
"epoch": 0.58,
"learning_rate": 0.00021156469408224675,
"loss": 2.6086,
"theoretical_loss": 3.442401934989899,
"tokens_seen": 1918238720
},
{
"epoch": 0.58,
"learning_rate": 0.00021155466399197593,
"loss": 2.6523,
"theoretical_loss": 3.442392048599115,
"tokens_seen": 1918304256
},
{
"epoch": 0.58,
"learning_rate": 0.0002115446339017051,
"loss": 2.4633,
"theoretical_loss": 3.442382162640646,
"tokens_seen": 1918369792
},
{
"epoch": 0.58,
"learning_rate": 0.0002115346038114343,
"loss": 2.4709,
"theoretical_loss": 3.442372277114459,
"tokens_seen": 1918435328
},
{
"epoch": 0.58,
"learning_rate": 0.00021152457372116348,
"loss": 2.7975,
"theoretical_loss": 3.44236239202052,
"tokens_seen": 1918500864
},
{
"epoch": 0.58,
"objective/train/docs_used": 1084225,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.681762456893921,
"objective/train/theoretical_loss": 3.4423525073587955,
"objective/train/tokens_used": 1939026400,
"theoretical_loss": 3.4423525073587955,
"tokens_seen": 1918566400
},
{
"epoch": 0.58,
"learning_rate": 0.00021151454363089268,
"loss": 2.5631,
"theoretical_loss": 3.4423525073587955,
"tokens_seen": 1918566400
},
{
"epoch": 0.58,
"learning_rate": 0.00021150451354062186,
"loss": 2.5202,
"theoretical_loss": 3.4423426231292518,
"tokens_seen": 1918631936
},
{
"epoch": 0.58,
"learning_rate": 0.00021149448345035105,
"loss": 2.4558,
"theoretical_loss": 3.442332739331855,
"tokens_seen": 1918697472
},
{
"epoch": 0.58,
"learning_rate": 0.00021148445336008023,
"loss": 2.8116,
"theoretical_loss": 3.4423228559665713,
"tokens_seen": 1918763008
},
{
"epoch": 0.58,
"learning_rate": 0.00021147442326980944,
"loss": 2.5238,
"theoretical_loss": 3.442312973033368,
"tokens_seen": 1918828544
},
{
"epoch": 0.58,
"learning_rate": 0.00021146439317953862,
"loss": 2.8165,
"theoretical_loss": 3.4423030905322105,
"tokens_seen": 1918894080
},
{
"epoch": 0.58,
"learning_rate": 0.0002114543630892678,
"loss": 2.5762,
"theoretical_loss": 3.4422932084630653,
"tokens_seen": 1918959616
},
{
"epoch": 0.58,
"learning_rate": 0.00021144433299899698,
"loss": 2.5632,
"theoretical_loss": 3.442283326825899,
"tokens_seen": 1919025152
},
{
"epoch": 0.58,
"learning_rate": 0.00021143430290872616,
"loss": 2.5601,
"theoretical_loss": 3.4422734456206783,
"tokens_seen": 1919090688
},
{
"epoch": 0.58,
"learning_rate": 0.00021142427281845537,
"loss": 2.7628,
"theoretical_loss": 3.4422635648473685,
"tokens_seen": 1919156224
},
{
"epoch": 0.58,
"learning_rate": 0.00021141424272818455,
"loss": 2.5787,
"theoretical_loss": 3.442253684505937,
"tokens_seen": 1919221760
},
{
"epoch": 0.58,
"learning_rate": 0.00021140421263791373,
"loss": 2.6771,
"theoretical_loss": 3.4422438045963495,
"tokens_seen": 1919287296
},
{
"epoch": 0.58,
"learning_rate": 0.0002113941825476429,
"loss": 2.808,
"theoretical_loss": 3.4422339251185727,
"tokens_seen": 1919352832
},
{
"epoch": 0.58,
"learning_rate": 0.00021138415245737212,
"loss": 2.6882,
"theoretical_loss": 3.4422240460725737,
"tokens_seen": 1919418368
},
{
"epoch": 0.58,
"learning_rate": 0.0002113741223671013,
"loss": 2.4038,
"theoretical_loss": 3.4422141674583173,
"tokens_seen": 1919483904
},
{
"epoch": 0.58,
"learning_rate": 0.00021136409227683048,
"loss": 2.6752,
"theoretical_loss": 3.4422042892757716,
"tokens_seen": 1919549440
},
{
"epoch": 0.58,
"learning_rate": 0.00021135406218655966,
"loss": 2.5791,
"theoretical_loss": 3.4421944115249015,
"tokens_seen": 1919614976
},
{
"epoch": 0.58,
"learning_rate": 0.00021134403209628884,
"loss": 2.4523,
"theoretical_loss": 3.442184534205674,
"tokens_seen": 1919680512
},
{
"epoch": 0.58,
"learning_rate": 0.00021133400200601808,
"loss": 2.6606,
"theoretical_loss": 3.4421746573180556,
"tokens_seen": 1919746048
},
{
"epoch": 0.58,
"learning_rate": 0.00021132397191574726,
"loss": 2.4474,
"theoretical_loss": 3.4421647808620133,
"tokens_seen": 1919811584
},
{
"epoch": 0.58,
"learning_rate": 0.00021131394182547644,
"loss": 2.6822,
"theoretical_loss": 3.442154904837512,
"tokens_seen": 1919877120
},
{
"epoch": 0.58,
"learning_rate": 0.00021130391173520562,
"loss": 2.808,
"theoretical_loss": 3.4421450292445197,
"tokens_seen": 1919942656
},
{
"epoch": 0.58,
"learning_rate": 0.00021129388164493483,
"loss": 2.5251,
"theoretical_loss": 3.442135154083002,
"tokens_seen": 1920008192
},
{
"epoch": 0.58,
"learning_rate": 0.000211283851554664,
"loss": 2.5758,
"theoretical_loss": 3.4421252793529247,
"tokens_seen": 1920073728
},
{
"epoch": 0.58,
"learning_rate": 0.0002112738214643932,
"loss": 2.5987,
"theoretical_loss": 3.4421154050542553,
"tokens_seen": 1920139264
},
{
"epoch": 0.58,
"objective/train/docs_used": 1085471,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5947773456573486,
"objective/train/theoretical_loss": 3.4421055311869604,
"objective/train/tokens_used": 1940664800,
"theoretical_loss": 3.4421055311869604,
"tokens_seen": 1920204800
},
{
"epoch": 0.58,
"learning_rate": 0.00021126379137412237,
"loss": 2.5052,
"theoretical_loss": 3.4421055311869604,
"tokens_seen": 1920204800
},
{
"epoch": 0.58,
"learning_rate": 0.00021125376128385158,
"loss": 2.6497,
"theoretical_loss": 3.442095657751006,
"tokens_seen": 1920270336
},
{
"epoch": 0.58,
"learning_rate": 0.00021124373119358076,
"loss": 2.3678,
"theoretical_loss": 3.4420857847463573,
"tokens_seen": 1920335872
},
{
"epoch": 0.58,
"learning_rate": 0.00021123370110330994,
"loss": 2.7756,
"theoretical_loss": 3.442075912172983,
"tokens_seen": 1920401408
},
{
"epoch": 0.58,
"learning_rate": 0.00021122367101303913,
"loss": 2.624,
"theoretical_loss": 3.4420660400308476,
"tokens_seen": 1920466944
},
{
"epoch": 0.58,
"learning_rate": 0.0002112136409227683,
"loss": 2.5489,
"theoretical_loss": 3.442056168319919,
"tokens_seen": 1920532480
},
{
"epoch": 0.58,
"learning_rate": 0.00021120361083249752,
"loss": 2.7843,
"theoretical_loss": 3.442046297040163,
"tokens_seen": 1920598016
},
{
"epoch": 0.58,
"learning_rate": 0.0002111935807422267,
"loss": 2.617,
"theoretical_loss": 3.442036426191546,
"tokens_seen": 1920663552
},
{
"epoch": 0.58,
"learning_rate": 0.00021118355065195588,
"loss": 2.5718,
"theoretical_loss": 3.442026555774034,
"tokens_seen": 1920729088
},
{
"epoch": 0.58,
"learning_rate": 0.00021117352056168506,
"loss": 2.5841,
"theoretical_loss": 3.442016685787595,
"tokens_seen": 1920794624
},
{
"epoch": 0.58,
"learning_rate": 0.00021116349047141427,
"loss": 2.7461,
"theoretical_loss": 3.442006816232194,
"tokens_seen": 1920860160
},
{
"epoch": 0.58,
"learning_rate": 0.00021115346038114345,
"loss": 2.4778,
"theoretical_loss": 3.441996947107798,
"tokens_seen": 1920925696
},
{
"epoch": 0.58,
"learning_rate": 0.00021114343029087263,
"loss": 2.5122,
"theoretical_loss": 3.441987078414373,
"tokens_seen": 1920991232
},
{
"epoch": 0.58,
"learning_rate": 0.0002111334002006018,
"loss": 2.5352,
"theoretical_loss": 3.4419772101518866,
"tokens_seen": 1921056768
},
{
"epoch": 0.58,
"learning_rate": 0.000211123370110331,
"loss": 2.5826,
"theoretical_loss": 3.4419673423203037,
"tokens_seen": 1921122304
},
{
"epoch": 0.58,
"learning_rate": 0.0002111133400200602,
"loss": 2.4657,
"theoretical_loss": 3.4419574749195925,
"tokens_seen": 1921187840
},
{
"epoch": 0.58,
"learning_rate": 0.00021110330992978938,
"loss": 2.6722,
"theoretical_loss": 3.441947607949718,
"tokens_seen": 1921253376
},
{
"epoch": 0.58,
"learning_rate": 0.00021109327983951856,
"loss": 2.2907,
"theoretical_loss": 3.441937741410648,
"tokens_seen": 1921318912
},
{
"epoch": 0.58,
"learning_rate": 0.00021108324974924774,
"loss": 2.4596,
"theoretical_loss": 3.441927875302348,
"tokens_seen": 1921384448
},
{
"epoch": 0.58,
"learning_rate": 0.00021107321965897695,
"loss": 2.5746,
"theoretical_loss": 3.441918009624785,
"tokens_seen": 1921449984
},
{
"epoch": 0.58,
"learning_rate": 0.00021106318956870613,
"loss": 2.4333,
"theoretical_loss": 3.4419081443779254,
"tokens_seen": 1921515520
},
{
"epoch": 0.58,
"learning_rate": 0.0002110531594784353,
"loss": 2.4937,
"theoretical_loss": 3.4418982795617357,
"tokens_seen": 1921581056
},
{
"epoch": 0.58,
"learning_rate": 0.0002110431293881645,
"loss": 2.7212,
"theoretical_loss": 3.441888415176182,
"tokens_seen": 1921646592
},
{
"epoch": 0.58,
"learning_rate": 0.00021103309929789368,
"loss": 2.7564,
"theoretical_loss": 3.4418785512212313,
"tokens_seen": 1921712128
},
{
"epoch": 0.58,
"learning_rate": 0.00021102306920762288,
"loss": 2.4207,
"theoretical_loss": 3.44186868769685,
"tokens_seen": 1921777664
},
{
"epoch": 0.58,
"objective/train/docs_used": 1086172,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.539794445037842,
"objective/train/theoretical_loss": 3.441858824603005,
"objective/train/tokens_used": 1942303200,
"theoretical_loss": 3.441858824603005,
"tokens_seen": 1921843200
},
{
"epoch": 0.58,
"learning_rate": 0.00021101303911735207,
"loss": 2.7923,
"theoretical_loss": 3.441858824603005,
"tokens_seen": 1921843200
},
{
"epoch": 0.58,
"learning_rate": 0.00021100300902708125,
"loss": 2.5245,
"theoretical_loss": 3.4418489619396624,
"tokens_seen": 1921908736
},
{
"epoch": 0.58,
"learning_rate": 0.00021099297893681043,
"loss": 2.7623,
"theoretical_loss": 3.4418390997067885,
"tokens_seen": 1921974272
},
{
"epoch": 0.58,
"learning_rate": 0.00021098294884653964,
"loss": 2.6693,
"theoretical_loss": 3.44182923790435,
"tokens_seen": 1922039808
},
{
"epoch": 0.58,
"learning_rate": 0.00021097291875626882,
"loss": 2.5848,
"theoretical_loss": 3.4418193765323135,
"tokens_seen": 1922105344
},
{
"epoch": 0.58,
"learning_rate": 0.000210962888665998,
"loss": 2.405,
"theoretical_loss": 3.4418095155906463,
"tokens_seen": 1922170880
},
{
"epoch": 0.58,
"learning_rate": 0.00021095285857572718,
"loss": 2.7712,
"theoretical_loss": 3.441799655079314,
"tokens_seen": 1922236416
},
{
"epoch": 0.58,
"learning_rate": 0.00021094282848545636,
"loss": 2.3336,
"theoretical_loss": 3.441789794998283,
"tokens_seen": 1922301952
},
{
"epoch": 0.58,
"learning_rate": 0.00021093279839518557,
"loss": 2.7085,
"theoretical_loss": 3.4417799353475202,
"tokens_seen": 1922367488
},
{
"epoch": 0.58,
"learning_rate": 0.00021092276830491475,
"loss": 2.395,
"theoretical_loss": 3.4417700761269927,
"tokens_seen": 1922433024
},
{
"epoch": 0.58,
"learning_rate": 0.00021091273821464393,
"loss": 2.6589,
"theoretical_loss": 3.4417602173366664,
"tokens_seen": 1922498560
},
{
"epoch": 0.58,
"learning_rate": 0.0002109027081243731,
"loss": 2.3189,
"theoretical_loss": 3.441750358976508,
"tokens_seen": 1922564096
},
{
"epoch": 0.58,
"learning_rate": 0.00021089267803410232,
"loss": 2.7088,
"theoretical_loss": 3.4417405010464837,
"tokens_seen": 1922629632
},
{
"epoch": 0.58,
"learning_rate": 0.0002108826479438315,
"loss": 2.6804,
"theoretical_loss": 3.441730643546561,
"tokens_seen": 1922695168
},
{
"epoch": 0.58,
"learning_rate": 0.00021087261785356068,
"loss": 2.7284,
"theoretical_loss": 3.441720786476706,
"tokens_seen": 1922760704
},
{
"epoch": 0.58,
"learning_rate": 0.00021086258776328986,
"loss": 2.5054,
"theoretical_loss": 3.441710929836885,
"tokens_seen": 1922826240
},
{
"epoch": 0.58,
"learning_rate": 0.00021085255767301904,
"loss": 2.6282,
"theoretical_loss": 3.4417010736270646,
"tokens_seen": 1922891776
},
{
"epoch": 0.58,
"learning_rate": 0.00021084252758274825,
"loss": 2.713,
"theoretical_loss": 3.441691217847212,
"tokens_seen": 1922957312
},
{
"epoch": 0.58,
"learning_rate": 0.00021083249749247743,
"loss": 2.7367,
"theoretical_loss": 3.4416813624972935,
"tokens_seen": 1923022848
},
{
"epoch": 0.58,
"learning_rate": 0.00021082246740220661,
"loss": 2.4991,
"theoretical_loss": 3.441671507577275,
"tokens_seen": 1923088384
},
{
"epoch": 0.58,
"learning_rate": 0.0002108124373119358,
"loss": 2.5891,
"theoretical_loss": 3.441661653087124,
"tokens_seen": 1923153920
},
{
"epoch": 0.58,
"learning_rate": 0.000210802407221665,
"loss": 3.0125,
"theoretical_loss": 3.4416517990268067,
"tokens_seen": 1923219456
},
{
"epoch": 0.58,
"learning_rate": 0.00021079237713139419,
"loss": 2.642,
"theoretical_loss": 3.44164194539629,
"tokens_seen": 1923284992
},
{
"epoch": 0.58,
"learning_rate": 0.00021078234704112337,
"loss": 2.6619,
"theoretical_loss": 3.4416320921955403,
"tokens_seen": 1923350528
},
{
"epoch": 0.58,
"learning_rate": 0.00021077231695085255,
"loss": 2.7116,
"theoretical_loss": 3.4416222394245235,
"tokens_seen": 1923416064
},
{
"epoch": 0.58,
"objective/train/docs_used": 1087440,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.235539197921753,
"objective/train/theoretical_loss": 3.441612387083208,
"objective/train/tokens_used": 1943941600,
"theoretical_loss": 3.441612387083208,
"tokens_seen": 1923481600
},
{
"epoch": 0.58,
"learning_rate": 0.00021076228686058176,
"loss": 2.5422,
"theoretical_loss": 3.441612387083208,
"tokens_seen": 1923481600
},
{
"epoch": 0.58,
"learning_rate": 0.00021075225677031094,
"loss": 2.7321,
"theoretical_loss": 3.4416025351715587,
"tokens_seen": 1923547136
},
{
"epoch": 0.58,
"learning_rate": 0.00021074222668004012,
"loss": 2.6197,
"theoretical_loss": 3.441592683689543,
"tokens_seen": 1923612672
},
{
"epoch": 0.58,
"learning_rate": 0.0002107321965897693,
"loss": 2.6081,
"theoretical_loss": 3.4415828326371276,
"tokens_seen": 1923678208
},
{
"epoch": 0.58,
"learning_rate": 0.00021072216649949848,
"loss": 2.7167,
"theoretical_loss": 3.4415729820142786,
"tokens_seen": 1923743744
},
{
"epoch": 0.58,
"learning_rate": 0.0002107121364092277,
"loss": 2.5914,
"theoretical_loss": 3.441563131820963,
"tokens_seen": 1923809280
},
{
"epoch": 0.58,
"learning_rate": 0.00021070210631895687,
"loss": 2.6545,
"theoretical_loss": 3.4415532820571473,
"tokens_seen": 1923874816
},
{
"epoch": 0.58,
"learning_rate": 0.00021069207622868605,
"loss": 2.603,
"theoretical_loss": 3.4415434327227983,
"tokens_seen": 1923940352
},
{
"epoch": 0.58,
"learning_rate": 0.00021068204613841523,
"loss": 2.5697,
"theoretical_loss": 3.441533583817883,
"tokens_seen": 1924005888
},
{
"epoch": 0.58,
"learning_rate": 0.00021067201604814444,
"loss": 2.6807,
"theoretical_loss": 3.4415237353423676,
"tokens_seen": 1924071424
},
{
"epoch": 0.58,
"learning_rate": 0.00021066198595787362,
"loss": 2.916,
"theoretical_loss": 3.4415138872962183,
"tokens_seen": 1924136960
},
{
"epoch": 0.58,
"learning_rate": 0.0002106519558676028,
"loss": 2.65,
"theoretical_loss": 3.4415040396794025,
"tokens_seen": 1924202496
},
{
"epoch": 0.58,
"learning_rate": 0.00021064192577733198,
"loss": 2.5015,
"theoretical_loss": 3.441494192491887,
"tokens_seen": 1924268032
},
{
"epoch": 0.58,
"learning_rate": 0.00021063189568706116,
"loss": 2.6574,
"theoretical_loss": 3.4414843457336373,
"tokens_seen": 1924333568
},
{
"epoch": 0.58,
"learning_rate": 0.00021062186559679037,
"loss": 2.6774,
"theoretical_loss": 3.441474499404621,
"tokens_seen": 1924399104
},
{
"epoch": 0.58,
"learning_rate": 0.00021061183550651955,
"loss": 2.5641,
"theoretical_loss": 3.441464653504805,
"tokens_seen": 1924464640
},
{
"epoch": 0.58,
"learning_rate": 0.00021060180541624873,
"loss": 2.6647,
"theoretical_loss": 3.4414548080341554,
"tokens_seen": 1924530176
},
{
"epoch": 0.58,
"learning_rate": 0.00021059177532597792,
"loss": 2.6022,
"theoretical_loss": 3.441444962992639,
"tokens_seen": 1924595712
},
{
"epoch": 0.58,
"learning_rate": 0.00021058174523570715,
"loss": 2.6536,
"theoretical_loss": 3.441435118380223,
"tokens_seen": 1924661248
},
{
"epoch": 0.58,
"learning_rate": 0.00021057171514543633,
"loss": 2.9487,
"theoretical_loss": 3.4414252741968734,
"tokens_seen": 1924726784
},
{
"epoch": 0.58,
"learning_rate": 0.00021056168505516551,
"loss": 2.5579,
"theoretical_loss": 3.4414154304425573,
"tokens_seen": 1924792320
},
{
"epoch": 0.58,
"learning_rate": 0.0002105516549648947,
"loss": 2.818,
"theoretical_loss": 3.4414055871172406,
"tokens_seen": 1924857856
},
{
"epoch": 0.58,
"learning_rate": 0.00021054162487462388,
"loss": 2.5976,
"theoretical_loss": 3.441395744220891,
"tokens_seen": 1924923392
},
{
"epoch": 0.58,
"learning_rate": 0.00021053159478435308,
"loss": 2.3158,
"theoretical_loss": 3.441385901753475,
"tokens_seen": 1924988928
},
{
"epoch": 0.58,
"learning_rate": 0.00021052156469408227,
"loss": 2.6507,
"theoretical_loss": 3.441376059714959,
"tokens_seen": 1925054464
},
{
"epoch": 0.58,
"objective/train/docs_used": 1088918,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7074055671691895,
"objective/train/theoretical_loss": 3.44136621810531,
"objective/train/tokens_used": 1945580000,
"theoretical_loss": 3.44136621810531,
"tokens_seen": 1925120000
},
{
"epoch": 0.58,
"learning_rate": 0.00021051153460381145,
"loss": 2.6543,
"theoretical_loss": 3.44136621810531,
"tokens_seen": 1925120000
},
{
"epoch": 0.58,
"learning_rate": 0.00021050150451354063,
"loss": 2.7493,
"theoretical_loss": 3.4413563769244946,
"tokens_seen": 1925185536
},
{
"epoch": 0.58,
"learning_rate": 0.00021049147442326984,
"loss": 2.377,
"theoretical_loss": 3.441346536172479,
"tokens_seen": 1925251072
},
{
"epoch": 0.58,
"learning_rate": 0.00021048144433299902,
"loss": 2.8341,
"theoretical_loss": 3.441336695849231,
"tokens_seen": 1925316608
},
{
"epoch": 0.58,
"learning_rate": 0.0002104714142427282,
"loss": 2.6436,
"theoretical_loss": 3.441326855954716,
"tokens_seen": 1925382144
},
{
"epoch": 0.58,
"learning_rate": 0.00021046138415245738,
"loss": 2.2671,
"theoretical_loss": 3.4413170164889024,
"tokens_seen": 1925447680
},
{
"epoch": 0.58,
"learning_rate": 0.00021045135406218656,
"loss": 2.706,
"theoretical_loss": 3.441307177451755,
"tokens_seen": 1925513216
},
{
"epoch": 0.58,
"learning_rate": 0.00021044132397191577,
"loss": 2.6657,
"theoretical_loss": 3.4412973388432424,
"tokens_seen": 1925578752
},
{
"epoch": 0.58,
"learning_rate": 0.00021043129388164495,
"loss": 2.7051,
"theoretical_loss": 3.44128750066333,
"tokens_seen": 1925644288
},
{
"epoch": 0.58,
"learning_rate": 0.00021042126379137413,
"loss": 2.7586,
"theoretical_loss": 3.4412776629119852,
"tokens_seen": 1925709824
},
{
"epoch": 0.58,
"learning_rate": 0.0002104112337011033,
"loss": 2.5533,
"theoretical_loss": 3.441267825589174,
"tokens_seen": 1925775360
},
{
"epoch": 0.58,
"learning_rate": 0.00021040120361083252,
"loss": 2.8904,
"theoretical_loss": 3.4412579886948644,
"tokens_seen": 1925840896
},
{
"epoch": 0.58,
"learning_rate": 0.0002103911735205617,
"loss": 2.693,
"theoretical_loss": 3.441248152229022,
"tokens_seen": 1925906432
},
{
"epoch": 0.58,
"learning_rate": 0.00021038114343029088,
"loss": 2.5337,
"theoretical_loss": 3.4412383161916145,
"tokens_seen": 1925971968
},
{
"epoch": 0.58,
"learning_rate": 0.00021037111334002006,
"loss": 2.5728,
"theoretical_loss": 3.4412284805826077,
"tokens_seen": 1926037504
},
{
"epoch": 0.58,
"learning_rate": 0.00021036108324974924,
"loss": 2.5121,
"theoretical_loss": 3.441218645401969,
"tokens_seen": 1926103040
},
{
"epoch": 0.58,
"learning_rate": 0.00021035105315947845,
"loss": 2.5986,
"theoretical_loss": 3.441208810649665,
"tokens_seen": 1926168576
},
{
"epoch": 0.58,
"learning_rate": 0.00021034102306920763,
"loss": 2.5666,
"theoretical_loss": 3.4411989763256625,
"tokens_seen": 1926234112
},
{
"epoch": 0.58,
"learning_rate": 0.00021033099297893681,
"loss": 2.841,
"theoretical_loss": 3.441189142429928,
"tokens_seen": 1926299648
},
{
"epoch": 0.58,
"learning_rate": 0.000210320962888666,
"loss": 2.5854,
"theoretical_loss": 3.441179308962429,
"tokens_seen": 1926365184
},
{
"epoch": 0.58,
"learning_rate": 0.0002103109327983952,
"loss": 2.6717,
"theoretical_loss": 3.4411694759231315,
"tokens_seen": 1926430720
},
{
"epoch": 0.58,
"learning_rate": 0.00021030090270812439,
"loss": 2.6007,
"theoretical_loss": 3.4411596433120026,
"tokens_seen": 1926496256
},
{
"epoch": 0.58,
"learning_rate": 0.00021029087261785357,
"loss": 2.673,
"theoretical_loss": 3.4411498111290095,
"tokens_seen": 1926561792
},
{
"epoch": 0.58,
"learning_rate": 0.00021028084252758275,
"loss": 2.5773,
"theoretical_loss": 3.4411399793741184,
"tokens_seen": 1926627328
},
{
"epoch": 0.58,
"learning_rate": 0.00021027081243731196,
"loss": 2.5324,
"theoretical_loss": 3.441130148047296,
"tokens_seen": 1926692864
},
{
"epoch": 0.58,
"objective/train/docs_used": 1089550,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8851699829101562,
"objective/train/theoretical_loss": 3.44112031714851,
"objective/train/tokens_used": 1947218400,
"theoretical_loss": 3.44112031714851,
"tokens_seen": 1926758400
},
{
"epoch": 0.58,
"learning_rate": 0.00021026078234704114,
"loss": 2.6885,
"theoretical_loss": 3.44112031714851,
"tokens_seen": 1926758400
},
{
"epoch": 0.58,
"learning_rate": 0.00021025075225677032,
"loss": 2.8023,
"theoretical_loss": 3.441110486677726,
"tokens_seen": 1926823936
},
{
"epoch": 0.58,
"learning_rate": 0.0002102407221664995,
"loss": 2.5292,
"theoretical_loss": 3.4411006566349114,
"tokens_seen": 1926889472
},
{
"epoch": 0.58,
"learning_rate": 0.00021023069207622868,
"loss": 2.829,
"theoretical_loss": 3.441090827020033,
"tokens_seen": 1926955008
},
{
"epoch": 0.58,
"learning_rate": 0.0002102206619859579,
"loss": 2.6029,
"theoretical_loss": 3.4410809978330583,
"tokens_seen": 1927020544
},
{
"epoch": 0.58,
"learning_rate": 0.00021021063189568707,
"loss": 2.5052,
"theoretical_loss": 3.441071169073953,
"tokens_seen": 1927086080
},
{
"epoch": 0.58,
"learning_rate": 0.00021020060180541625,
"loss": 2.7855,
"theoretical_loss": 3.441061340742684,
"tokens_seen": 1927151616
},
{
"epoch": 0.58,
"learning_rate": 0.00021019057171514543,
"loss": 2.7167,
"theoretical_loss": 3.4410515128392194,
"tokens_seen": 1927217152
},
{
"epoch": 0.58,
"learning_rate": 0.00021018054162487464,
"loss": 2.6487,
"theoretical_loss": 3.4410416853635244,
"tokens_seen": 1927282688
},
{
"epoch": 0.58,
"learning_rate": 0.00021017051153460382,
"loss": 2.8124,
"theoretical_loss": 3.441031858315567,
"tokens_seen": 1927348224
},
{
"epoch": 0.58,
"learning_rate": 0.000210160481444333,
"loss": 2.5577,
"theoretical_loss": 3.4410220316953133,
"tokens_seen": 1927413760
},
{
"epoch": 0.58,
"learning_rate": 0.00021015045135406218,
"loss": 2.6192,
"theoretical_loss": 3.441012205502731,
"tokens_seen": 1927479296
},
{
"epoch": 0.58,
"learning_rate": 0.00021014042126379136,
"loss": 2.5149,
"theoretical_loss": 3.441002379737786,
"tokens_seen": 1927544832
},
{
"epoch": 0.58,
"learning_rate": 0.00021013039117352057,
"loss": 2.6687,
"theoretical_loss": 3.4409925544004456,
"tokens_seen": 1927610368
},
{
"epoch": 0.58,
"learning_rate": 0.00021012036108324975,
"loss": 2.4944,
"theoretical_loss": 3.4409827294906767,
"tokens_seen": 1927675904
},
{
"epoch": 0.58,
"learning_rate": 0.00021011033099297894,
"loss": 2.4759,
"theoretical_loss": 3.4409729050084463,
"tokens_seen": 1927741440
},
{
"epoch": 0.58,
"learning_rate": 0.00021010030090270812,
"loss": 2.5308,
"theoretical_loss": 3.440963080953721,
"tokens_seen": 1927806976
},
{
"epoch": 0.58,
"learning_rate": 0.00021009027081243732,
"loss": 2.5908,
"theoretical_loss": 3.440953257326467,
"tokens_seen": 1927872512
},
{
"epoch": 0.58,
"learning_rate": 0.0002100802407221665,
"loss": 2.6168,
"theoretical_loss": 3.4409434341266527,
"tokens_seen": 1927938048
},
{
"epoch": 0.58,
"learning_rate": 0.0002100702106318957,
"loss": 2.6392,
"theoretical_loss": 3.440933611354244,
"tokens_seen": 1928003584
},
{
"epoch": 0.58,
"learning_rate": 0.00021006018054162487,
"loss": 2.5447,
"theoretical_loss": 3.4409237890092075,
"tokens_seen": 1928069120
},
{
"epoch": 0.58,
"learning_rate": 0.00021005015045135405,
"loss": 2.7343,
"theoretical_loss": 3.440913967091511,
"tokens_seen": 1928134656
},
{
"epoch": 0.58,
"learning_rate": 0.00021004012036108326,
"loss": 2.7091,
"theoretical_loss": 3.4409041456011202,
"tokens_seen": 1928200192
},
{
"epoch": 0.58,
"learning_rate": 0.00021003009027081244,
"loss": 2.6396,
"theoretical_loss": 3.4408943245380033,
"tokens_seen": 1928265728
},
{
"epoch": 0.58,
"learning_rate": 0.00021002006018054162,
"loss": 2.5118,
"theoretical_loss": 3.4408845039021263,
"tokens_seen": 1928331264
},
{
"epoch": 0.58,
"objective/train/docs_used": 1091004,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5786335468292236,
"objective/train/theoretical_loss": 3.4408746836934565,
"objective/train/tokens_used": 1948856800,
"theoretical_loss": 3.4408746836934565,
"tokens_seen": 1928396800
},
{
"epoch": 0.58,
"learning_rate": 0.0002100100300902708,
"loss": 2.576,
"theoretical_loss": 3.4408746836934565,
"tokens_seen": 1928396800
},
{
"epoch": 0.58,
"learning_rate": 0.00021,
"loss": 2.8235,
"theoretical_loss": 3.4408648639119606,
"tokens_seen": 1928462336
},
{
"epoch": 0.58,
"learning_rate": 0.0002099899699097292,
"loss": 2.5193,
"theoretical_loss": 3.4408550445576056,
"tokens_seen": 1928527872
},
{
"epoch": 0.58,
"learning_rate": 0.00020997993981945837,
"loss": 2.6454,
"theoretical_loss": 3.4408452256303583,
"tokens_seen": 1928593408
},
{
"epoch": 0.58,
"learning_rate": 0.00020996990972918755,
"loss": 2.7168,
"theoretical_loss": 3.4408354071301854,
"tokens_seen": 1928658944
},
{
"epoch": 0.58,
"learning_rate": 0.00020995987963891673,
"loss": 2.598,
"theoretical_loss": 3.4408255890570545,
"tokens_seen": 1928724480
},
{
"epoch": 0.58,
"learning_rate": 0.00020994984954864594,
"loss": 2.7352,
"theoretical_loss": 3.440815771410932,
"tokens_seen": 1928790016
},
{
"epoch": 0.58,
"learning_rate": 0.00020993981945837512,
"loss": 2.6839,
"theoretical_loss": 3.440805954191785,
"tokens_seen": 1928855552
},
{
"epoch": 0.58,
"learning_rate": 0.0002099297893681043,
"loss": 2.7554,
"theoretical_loss": 3.4407961373995803,
"tokens_seen": 1928921088
},
{
"epoch": 0.58,
"learning_rate": 0.00020991975927783348,
"loss": 2.8387,
"theoretical_loss": 3.4407863210342846,
"tokens_seen": 1928986624
},
{
"epoch": 0.58,
"learning_rate": 0.0002099097291875627,
"loss": 2.5128,
"theoretical_loss": 3.440776505095865,
"tokens_seen": 1929052160
},
{
"epoch": 0.58,
"learning_rate": 0.00020989969909729187,
"loss": 2.5505,
"theoretical_loss": 3.440766689584289,
"tokens_seen": 1929117696
},
{
"epoch": 0.58,
"learning_rate": 0.00020988966900702106,
"loss": 2.7821,
"theoretical_loss": 3.4407568744995234,
"tokens_seen": 1929183232
},
{
"epoch": 0.58,
"learning_rate": 0.00020987963891675024,
"loss": 2.5338,
"theoretical_loss": 3.4407470598415344,
"tokens_seen": 1929248768
},
{
"epoch": 0.58,
"learning_rate": 0.00020986960882647942,
"loss": 2.5557,
"theoretical_loss": 3.440737245610289,
"tokens_seen": 1929314304
},
{
"epoch": 0.58,
"learning_rate": 0.00020985957873620863,
"loss": 2.75,
"theoretical_loss": 3.4407274318057555,
"tokens_seen": 1929379840
},
{
"epoch": 0.58,
"learning_rate": 0.0002098495486459378,
"loss": 2.5565,
"theoretical_loss": 3.440717618427899,
"tokens_seen": 1929445376
},
{
"epoch": 0.58,
"learning_rate": 0.000209839518555667,
"loss": 2.6087,
"theoretical_loss": 3.4407078054766878,
"tokens_seen": 1929510912
},
{
"epoch": 0.58,
"learning_rate": 0.0002098294884653962,
"loss": 2.4574,
"theoretical_loss": 3.440697992952088,
"tokens_seen": 1929576448
},
{
"epoch": 0.58,
"learning_rate": 0.0002098194583751254,
"loss": 2.529,
"theoretical_loss": 3.4406881808540675,
"tokens_seen": 1929641984
},
{
"epoch": 0.58,
"learning_rate": 0.00020980942828485459,
"loss": 2.5376,
"theoretical_loss": 3.440678369182592,
"tokens_seen": 1929707520
},
{
"epoch": 0.58,
"learning_rate": 0.00020979939819458377,
"loss": 2.4811,
"theoretical_loss": 3.44066855793763,
"tokens_seen": 1929773056
},
{
"epoch": 0.58,
"learning_rate": 0.00020978936810431295,
"loss": 2.6077,
"theoretical_loss": 3.440658747119147,
"tokens_seen": 1929838592
},
{
"epoch": 0.58,
"learning_rate": 0.00020977933801404216,
"loss": 2.7556,
"theoretical_loss": 3.440648936727111,
"tokens_seen": 1929904128
},
{
"epoch": 0.58,
"learning_rate": 0.00020976930792377134,
"loss": 2.673,
"theoretical_loss": 3.4406391267614884,
"tokens_seen": 1929969664
},
{
"epoch": 0.58,
"objective/train/docs_used": 1091522,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8513214588165283,
"objective/train/theoretical_loss": 3.4406293172222466,
"objective/train/tokens_used": 1950495200,
"theoretical_loss": 3.4406293172222466,
"tokens_seen": 1930035200
},
{
"epoch": 0.58,
"learning_rate": 0.00020975927783350052,
"loss": 2.6767,
"theoretical_loss": 3.4406293172222466,
"tokens_seen": 1930035200
},
{
"epoch": 0.58,
"learning_rate": 0.0002097492477432297,
"loss": 2.4559,
"theoretical_loss": 3.4406195081093527,
"tokens_seen": 1930100736
},
{
"epoch": 0.58,
"learning_rate": 0.00020973921765295888,
"loss": 2.3288,
"theoretical_loss": 3.4406096994227733,
"tokens_seen": 1930166272
},
{
"epoch": 0.58,
"learning_rate": 0.0002097291875626881,
"loss": 2.3444,
"theoretical_loss": 3.440599891162475,
"tokens_seen": 1930231808
},
{
"epoch": 0.58,
"learning_rate": 0.00020971915747241727,
"loss": 2.6807,
"theoretical_loss": 3.4405900833284258,
"tokens_seen": 1930297344
},
{
"epoch": 0.58,
"learning_rate": 0.00020970912738214645,
"loss": 2.7168,
"theoretical_loss": 3.440580275920592,
"tokens_seen": 1930362880
},
{
"epoch": 0.58,
"learning_rate": 0.00020969909729187563,
"loss": 2.5145,
"theoretical_loss": 3.440570468938941,
"tokens_seen": 1930428416
},
{
"epoch": 0.58,
"learning_rate": 0.00020968906720160484,
"loss": 2.4488,
"theoretical_loss": 3.4405606623834393,
"tokens_seen": 1930493952
},
{
"epoch": 0.59,
"learning_rate": 0.00020967903711133402,
"loss": 2.5455,
"theoretical_loss": 3.4405508562540543,
"tokens_seen": 1930559488
},
{
"epoch": 0.59,
"learning_rate": 0.0002096690070210632,
"loss": 2.7534,
"theoretical_loss": 3.4405410505507534,
"tokens_seen": 1930625024
},
{
"epoch": 0.59,
"learning_rate": 0.00020965897693079238,
"loss": 2.5894,
"theoretical_loss": 3.440531245273503,
"tokens_seen": 1930690560
},
{
"epoch": 0.59,
"learning_rate": 0.00020964894684052156,
"loss": 2.677,
"theoretical_loss": 3.4405214404222706,
"tokens_seen": 1930756096
},
{
"epoch": 0.59,
"learning_rate": 0.00020963891675025077,
"loss": 2.6337,
"theoretical_loss": 3.4405116359970225,
"tokens_seen": 1930821632
},
{
"epoch": 0.59,
"learning_rate": 0.00020962888665997995,
"loss": 2.5375,
"theoretical_loss": 3.4405018319977265,
"tokens_seen": 1930887168
},
{
"epoch": 0.59,
"learning_rate": 0.00020961885656970914,
"loss": 2.564,
"theoretical_loss": 3.440492028424349,
"tokens_seen": 1930952704
},
{
"epoch": 0.59,
"learning_rate": 0.00020960882647943832,
"loss": 2.6399,
"theoretical_loss": 3.440482225276858,
"tokens_seen": 1931018240
},
{
"epoch": 0.59,
"learning_rate": 0.00020959879638916752,
"loss": 2.7438,
"theoretical_loss": 3.440472422555219,
"tokens_seen": 1931083776
},
{
"epoch": 0.59,
"learning_rate": 0.0002095887662988967,
"loss": 2.6903,
"theoretical_loss": 3.4404626202594004,
"tokens_seen": 1931149312
},
{
"epoch": 0.59,
"learning_rate": 0.0002095787362086259,
"loss": 2.7266,
"theoretical_loss": 3.4404528183893692,
"tokens_seen": 1931214848
},
{
"epoch": 0.59,
"learning_rate": 0.00020956870611835507,
"loss": 2.7944,
"theoretical_loss": 3.4404430169450917,
"tokens_seen": 1931280384
},
{
"epoch": 0.59,
"learning_rate": 0.00020955867602808425,
"loss": 2.8867,
"theoretical_loss": 3.4404332159265354,
"tokens_seen": 1931345920
},
{
"epoch": 0.59,
"learning_rate": 0.00020954864593781346,
"loss": 2.7837,
"theoretical_loss": 3.440423415333667,
"tokens_seen": 1931411456
},
{
"epoch": 0.59,
"learning_rate": 0.00020953861584754264,
"loss": 2.4087,
"theoretical_loss": 3.440413615166454,
"tokens_seen": 1931476992
},
{
"epoch": 0.59,
"learning_rate": 0.00020952858575727182,
"loss": 2.7186,
"theoretical_loss": 3.440403815424864,
"tokens_seen": 1931542528
},
{
"epoch": 0.59,
"learning_rate": 0.000209518555667001,
"loss": 2.6529,
"theoretical_loss": 3.440394016108863,
"tokens_seen": 1931608064
},
{
"epoch": 0.59,
"objective/train/docs_used": 1091970,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.781391143798828,
"objective/train/theoretical_loss": 3.4403842172184183,
"objective/train/tokens_used": 1952133600,
"theoretical_loss": 3.4403842172184183,
"tokens_seen": 1931673600
},
{
"epoch": 0.59,
"learning_rate": 0.0002095085255767302,
"loss": 2.5966,
"theoretical_loss": 3.4403842172184183,
"tokens_seen": 1931673600
},
{
"epoch": 0.59,
"learning_rate": 0.0002094984954864594,
"loss": 2.5333,
"theoretical_loss": 3.440374418753497,
"tokens_seen": 1931739136
},
{
"epoch": 0.59,
"learning_rate": 0.00020948846539618857,
"loss": 2.6782,
"theoretical_loss": 3.4403646207140666,
"tokens_seen": 1931804672
},
{
"epoch": 0.59,
"learning_rate": 0.00020947843530591775,
"loss": 2.6767,
"theoretical_loss": 3.440354823100094,
"tokens_seen": 1931870208
},
{
"epoch": 0.59,
"learning_rate": 0.00020946840521564693,
"loss": 2.7973,
"theoretical_loss": 3.4403450259115465,
"tokens_seen": 1931935744
},
{
"epoch": 0.59,
"learning_rate": 0.00020945837512537614,
"loss": 2.703,
"theoretical_loss": 3.4403352291483906,
"tokens_seen": 1932001280
},
{
"epoch": 0.59,
"learning_rate": 0.00020944834503510532,
"loss": 2.7084,
"theoretical_loss": 3.440325432810594,
"tokens_seen": 1932066816
},
{
"epoch": 0.59,
"learning_rate": 0.0002094383149448345,
"loss": 2.6117,
"theoretical_loss": 3.4403156368981236,
"tokens_seen": 1932132352
},
{
"epoch": 0.59,
"learning_rate": 0.00020942828485456368,
"loss": 2.8571,
"theoretical_loss": 3.4403058414109458,
"tokens_seen": 1932197888
},
{
"epoch": 0.59,
"learning_rate": 0.0002094182547642929,
"loss": 2.6124,
"theoretical_loss": 3.4402960463490286,
"tokens_seen": 1932263424
},
{
"epoch": 0.59,
"learning_rate": 0.00020940822467402207,
"loss": 2.5946,
"theoretical_loss": 3.4402862517123394,
"tokens_seen": 1932328960
},
{
"epoch": 0.59,
"learning_rate": 0.00020939819458375126,
"loss": 2.6284,
"theoretical_loss": 3.440276457500844,
"tokens_seen": 1932394496
},
{
"epoch": 0.59,
"learning_rate": 0.00020938816449348044,
"loss": 2.6871,
"theoretical_loss": 3.440266663714511,
"tokens_seen": 1932460032
},
{
"epoch": 0.59,
"learning_rate": 0.00020937813440320962,
"loss": 2.6665,
"theoretical_loss": 3.440256870353307,
"tokens_seen": 1932525568
},
{
"epoch": 0.59,
"learning_rate": 0.00020936810431293883,
"loss": 2.6524,
"theoretical_loss": 3.4402470774171987,
"tokens_seen": 1932591104
},
{
"epoch": 0.59,
"learning_rate": 0.000209358074222668,
"loss": 2.7694,
"theoretical_loss": 3.4402372849061535,
"tokens_seen": 1932656640
},
{
"epoch": 0.59,
"learning_rate": 0.0002093480441323972,
"loss": 2.5623,
"theoretical_loss": 3.4402274928201386,
"tokens_seen": 1932722176
},
{
"epoch": 0.59,
"learning_rate": 0.00020933801404212637,
"loss": 2.6972,
"theoretical_loss": 3.4402177011591206,
"tokens_seen": 1932787712
},
{
"epoch": 0.59,
"learning_rate": 0.00020932798395185558,
"loss": 2.7275,
"theoretical_loss": 3.4402079099230676,
"tokens_seen": 1932853248
},
{
"epoch": 0.59,
"learning_rate": 0.00020931795386158476,
"loss": 2.5805,
"theoretical_loss": 3.4401981191119466,
"tokens_seen": 1932918784
},
{
"epoch": 0.59,
"learning_rate": 0.00020930792377131394,
"loss": 2.7999,
"theoretical_loss": 3.440188328725724,
"tokens_seen": 1932984320
},
{
"epoch": 0.59,
"learning_rate": 0.00020929789368104312,
"loss": 2.6255,
"theoretical_loss": 3.4401785387643677,
"tokens_seen": 1933049856
},
{
"epoch": 0.59,
"learning_rate": 0.0002092878635907723,
"loss": 2.7345,
"theoretical_loss": 3.440168749227844,
"tokens_seen": 1933115392
},
{
"epoch": 0.59,
"learning_rate": 0.0002092778335005015,
"loss": 2.6249,
"theoretical_loss": 3.440158960116121,
"tokens_seen": 1933180928
},
{
"epoch": 0.59,
"learning_rate": 0.0002092678034102307,
"loss": 2.6457,
"theoretical_loss": 3.4401491714291654,
"tokens_seen": 1933246464
},
{
"epoch": 0.59,
"objective/train/docs_used": 1092955,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4713950157165527,
"objective/train/theoretical_loss": 3.4401393831669447,
"objective/train/tokens_used": 1953772000,
"theoretical_loss": 3.4401393831669447,
"tokens_seen": 1933312000
},
{
"epoch": 0.59,
"learning_rate": 0.00020925777331995987,
"loss": 2.1984,
"theoretical_loss": 3.4401393831669447,
"tokens_seen": 1933312000
},
{
"epoch": 0.59,
"learning_rate": 0.00020924774322968905,
"loss": 2.7649,
"theoretical_loss": 3.440129595329425,
"tokens_seen": 1933377536
},
{
"epoch": 0.59,
"learning_rate": 0.00020923771313941826,
"loss": 2.6434,
"theoretical_loss": 3.440119807916575,
"tokens_seen": 1933443072
},
{
"epoch": 0.59,
"learning_rate": 0.00020922768304914744,
"loss": 2.6962,
"theoretical_loss": 3.440110020928361,
"tokens_seen": 1933508608
},
{
"epoch": 0.59,
"learning_rate": 0.00020921765295887662,
"loss": 2.6908,
"theoretical_loss": 3.440100234364751,
"tokens_seen": 1933574144
},
{
"epoch": 0.59,
"learning_rate": 0.0002092076228686058,
"loss": 2.4424,
"theoretical_loss": 3.4400904482257104,
"tokens_seen": 1933639680
},
{
"epoch": 0.59,
"learning_rate": 0.00020919759277833499,
"loss": 2.6432,
"theoretical_loss": 3.4400806625112086,
"tokens_seen": 1933705216
},
{
"epoch": 0.59,
"learning_rate": 0.0002091875626880642,
"loss": 2.4815,
"theoretical_loss": 3.440070877221211,
"tokens_seen": 1933770752
},
{
"epoch": 0.59,
"learning_rate": 0.00020917753259779338,
"loss": 2.6012,
"theoretical_loss": 3.440061092355686,
"tokens_seen": 1933836288
},
{
"epoch": 0.59,
"learning_rate": 0.00020916750250752256,
"loss": 2.7036,
"theoretical_loss": 3.4400513079146,
"tokens_seen": 1933901824
},
{
"epoch": 0.59,
"learning_rate": 0.00020915747241725174,
"loss": 2.9219,
"theoretical_loss": 3.440041523897921,
"tokens_seen": 1933967360
},
{
"epoch": 0.59,
"learning_rate": 0.00020914744232698095,
"loss": 2.6707,
"theoretical_loss": 3.4400317403056153,
"tokens_seen": 1934032896
},
{
"epoch": 0.59,
"learning_rate": 0.00020913741223671013,
"loss": 2.4735,
"theoretical_loss": 3.4400219571376507,
"tokens_seen": 1934098432
},
{
"epoch": 0.59,
"learning_rate": 0.0002091273821464393,
"loss": 2.7646,
"theoretical_loss": 3.4400121743939946,
"tokens_seen": 1934163968
},
{
"epoch": 0.59,
"learning_rate": 0.0002091173520561685,
"loss": 2.6424,
"theoretical_loss": 3.4400023920746134,
"tokens_seen": 1934229504
},
{
"epoch": 0.59,
"learning_rate": 0.0002091073219658977,
"loss": 2.8384,
"theoretical_loss": 3.439992610179475,
"tokens_seen": 1934295040
},
{
"epoch": 0.59,
"learning_rate": 0.00020909729187562688,
"loss": 2.5301,
"theoretical_loss": 3.4399828287085468,
"tokens_seen": 1934360576
},
{
"epoch": 0.59,
"learning_rate": 0.00020908726178535606,
"loss": 2.5774,
"theoretical_loss": 3.4399730476617956,
"tokens_seen": 1934426112
},
{
"epoch": 0.59,
"learning_rate": 0.00020907723169508527,
"loss": 2.5898,
"theoretical_loss": 3.4399632670391886,
"tokens_seen": 1934491648
},
{
"epoch": 0.59,
"learning_rate": 0.00020906720160481445,
"loss": 2.6936,
"theoretical_loss": 3.439953486840693,
"tokens_seen": 1934557184
},
{
"epoch": 0.59,
"learning_rate": 0.00020905717151454366,
"loss": 2.5316,
"theoretical_loss": 3.4399437070662766,
"tokens_seen": 1934622720
},
{
"epoch": 0.59,
"learning_rate": 0.00020904714142427284,
"loss": 2.6154,
"theoretical_loss": 3.439933927715906,
"tokens_seen": 1934688256
},
{
"epoch": 0.59,
"learning_rate": 0.00020903711133400202,
"loss": 2.5992,
"theoretical_loss": 3.439924148789549,
"tokens_seen": 1934753792
},
{
"epoch": 0.59,
"learning_rate": 0.0002090270812437312,
"loss": 2.5632,
"theoretical_loss": 3.4399143702871724,
"tokens_seen": 1934819328
},
{
"epoch": 0.59,
"learning_rate": 0.0002090170511534604,
"loss": 2.8285,
"theoretical_loss": 3.4399045922087437,
"tokens_seen": 1934884864
},
{
"epoch": 0.59,
"objective/train/docs_used": 1093336,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.047774314880371,
"objective/train/theoretical_loss": 3.4398948145542296,
"objective/train/tokens_used": 1955410400,
"theoretical_loss": 3.4398948145542296,
"tokens_seen": 1934950400
},
{
"epoch": 0.59,
"learning_rate": 0.0002090070210631896,
"loss": 2.889,
"theoretical_loss": 3.4398948145542296,
"tokens_seen": 1934950400
},
{
"epoch": 0.59,
"learning_rate": 0.00020899699097291877,
"loss": 2.6421,
"theoretical_loss": 3.4398850373235987,
"tokens_seen": 1935015936
},
{
"epoch": 0.59,
"learning_rate": 0.00020898696088264795,
"loss": 2.7375,
"theoretical_loss": 3.439875260516817,
"tokens_seen": 1935081472
},
{
"epoch": 0.59,
"learning_rate": 0.00020897693079237713,
"loss": 2.4946,
"theoretical_loss": 3.439865484133852,
"tokens_seen": 1935147008
},
{
"epoch": 0.59,
"learning_rate": 0.00020896690070210634,
"loss": 2.5641,
"theoretical_loss": 3.4398557081746715,
"tokens_seen": 1935212544
},
{
"epoch": 0.59,
"learning_rate": 0.00020895687061183552,
"loss": 2.5532,
"theoretical_loss": 3.439845932639242,
"tokens_seen": 1935278080
},
{
"epoch": 0.59,
"learning_rate": 0.0002089468405215647,
"loss": 2.9805,
"theoretical_loss": 3.4398361575275316,
"tokens_seen": 1935343616
},
{
"epoch": 0.59,
"learning_rate": 0.00020893681043129389,
"loss": 2.7661,
"theoretical_loss": 3.439826382839507,
"tokens_seen": 1935409152
},
{
"epoch": 0.59,
"learning_rate": 0.0002089267803410231,
"loss": 2.7767,
"theoretical_loss": 3.4398166085751356,
"tokens_seen": 1935474688
},
{
"epoch": 0.59,
"learning_rate": 0.00020891675025075227,
"loss": 2.576,
"theoretical_loss": 3.4398068347343855,
"tokens_seen": 1935540224
},
{
"epoch": 0.59,
"learning_rate": 0.00020890672016048146,
"loss": 2.4034,
"theoretical_loss": 3.439797061317223,
"tokens_seen": 1935605760
},
{
"epoch": 0.59,
"learning_rate": 0.00020889669007021064,
"loss": 2.5821,
"theoretical_loss": 3.439787288323615,
"tokens_seen": 1935671296
},
{
"epoch": 0.59,
"learning_rate": 0.00020888665997993982,
"loss": 2.6796,
"theoretical_loss": 3.43977751575353,
"tokens_seen": 1935736832
},
{
"epoch": 0.59,
"learning_rate": 0.00020887662988966903,
"loss": 2.6657,
"theoretical_loss": 3.439767743606935,
"tokens_seen": 1935802368
},
{
"epoch": 0.59,
"learning_rate": 0.0002088665997993982,
"loss": 2.499,
"theoretical_loss": 3.439757971883797,
"tokens_seen": 1935867904
},
{
"epoch": 0.59,
"learning_rate": 0.0002088565697091274,
"loss": 2.7077,
"theoretical_loss": 3.4397482005840834,
"tokens_seen": 1935933440
},
{
"epoch": 0.59,
"learning_rate": 0.00020884653961885657,
"loss": 2.6855,
"theoretical_loss": 3.439738429707762,
"tokens_seen": 1935998976
},
{
"epoch": 0.59,
"learning_rate": 0.00020883650952858578,
"loss": 2.7175,
"theoretical_loss": 3.439728659254799,
"tokens_seen": 1936064512
},
{
"epoch": 0.59,
"learning_rate": 0.00020882647943831496,
"loss": 2.7057,
"theoretical_loss": 3.439718889225163,
"tokens_seen": 1936130048
},
{
"epoch": 0.59,
"learning_rate": 0.00020881644934804414,
"loss": 2.7183,
"theoretical_loss": 3.4397091196188203,
"tokens_seen": 1936195584
},
{
"epoch": 0.59,
"learning_rate": 0.00020880641925777332,
"loss": 2.6489,
"theoretical_loss": 3.4396993504357387,
"tokens_seen": 1936261120
},
{
"epoch": 0.59,
"learning_rate": 0.0002087963891675025,
"loss": 2.5515,
"theoretical_loss": 3.4396895816758857,
"tokens_seen": 1936326656
},
{
"epoch": 0.59,
"learning_rate": 0.0002087863590772317,
"loss": 2.6985,
"theoretical_loss": 3.439679813339229,
"tokens_seen": 1936392192
},
{
"epoch": 0.59,
"learning_rate": 0.0002087763289869609,
"loss": 2.7148,
"theoretical_loss": 3.439670045425734,
"tokens_seen": 1936457728
},
{
"epoch": 0.59,
"learning_rate": 0.00020876629889669007,
"loss": 2.6132,
"theoretical_loss": 3.4396602779353707,
"tokens_seen": 1936523264
},
{
"epoch": 0.59,
"objective/train/docs_used": 1094637,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.530860185623169,
"objective/train/theoretical_loss": 3.4396505108681046,
"objective/train/tokens_used": 1957048800,
"theoretical_loss": 3.4396505108681046,
"tokens_seen": 1936588800
},
{
"epoch": 0.59,
"learning_rate": 0.00020875626880641925,
"loss": 2.6887,
"theoretical_loss": 3.4396505108681046,
"tokens_seen": 1936588800
},
{
"epoch": 0.59,
"learning_rate": 0.00020874623871614846,
"loss": 2.797,
"theoretical_loss": 3.4396407442239036,
"tokens_seen": 1936654336
},
{
"epoch": 0.59,
"learning_rate": 0.00020873620862587764,
"loss": 2.6123,
"theoretical_loss": 3.4396309780027354,
"tokens_seen": 1936719872
},
{
"epoch": 0.59,
"learning_rate": 0.00020872617853560682,
"loss": 2.6552,
"theoretical_loss": 3.4396212122045666,
"tokens_seen": 1936785408
},
{
"epoch": 0.59,
"learning_rate": 0.000208716148445336,
"loss": 2.8438,
"theoretical_loss": 3.4396114468293657,
"tokens_seen": 1936850944
},
{
"epoch": 0.59,
"learning_rate": 0.00020870611835506519,
"loss": 2.6227,
"theoretical_loss": 3.4396016818770994,
"tokens_seen": 1936916480
},
{
"epoch": 0.59,
"learning_rate": 0.0002086960882647944,
"loss": 2.8266,
"theoretical_loss": 3.4395919173477347,
"tokens_seen": 1936982016
},
{
"epoch": 0.59,
"learning_rate": 0.00020868605817452358,
"loss": 2.7828,
"theoretical_loss": 3.43958215324124,
"tokens_seen": 1937047552
},
{
"epoch": 0.59,
"learning_rate": 0.00020867602808425276,
"loss": 2.461,
"theoretical_loss": 3.439572389557581,
"tokens_seen": 1937113088
},
{
"epoch": 0.59,
"learning_rate": 0.00020866599799398194,
"loss": 2.6503,
"theoretical_loss": 3.439562626296727,
"tokens_seen": 1937178624
},
{
"epoch": 0.59,
"learning_rate": 0.00020865596790371115,
"loss": 2.7743,
"theoretical_loss": 3.439552863458644,
"tokens_seen": 1937244160
},
{
"epoch": 0.59,
"learning_rate": 0.00020864593781344033,
"loss": 2.7305,
"theoretical_loss": 3.4395431010433,
"tokens_seen": 1937309696
},
{
"epoch": 0.59,
"learning_rate": 0.0002086359077231695,
"loss": 2.7314,
"theoretical_loss": 3.439533339050662,
"tokens_seen": 1937375232
},
{
"epoch": 0.59,
"learning_rate": 0.0002086258776328987,
"loss": 2.6689,
"theoretical_loss": 3.4395235774806983,
"tokens_seen": 1937440768
},
{
"epoch": 0.59,
"learning_rate": 0.0002086158475426279,
"loss": 2.6919,
"theoretical_loss": 3.439513816333376,
"tokens_seen": 1937506304
},
{
"epoch": 0.59,
"learning_rate": 0.00020860581745235708,
"loss": 2.5629,
"theoretical_loss": 3.439504055608661,
"tokens_seen": 1937571840
},
{
"epoch": 0.59,
"learning_rate": 0.00020859578736208626,
"loss": 2.6923,
"theoretical_loss": 3.439494295306523,
"tokens_seen": 1937637376
},
{
"epoch": 0.59,
"learning_rate": 0.00020858575727181544,
"loss": 2.7194,
"theoretical_loss": 3.439484535426928,
"tokens_seen": 1937702912
},
{
"epoch": 0.59,
"learning_rate": 0.00020857572718154462,
"loss": 2.5248,
"theoretical_loss": 3.4394747759698436,
"tokens_seen": 1937768448
},
{
"epoch": 0.59,
"learning_rate": 0.00020856569709127383,
"loss": 2.7251,
"theoretical_loss": 3.439465016935238,
"tokens_seen": 1937833984
},
{
"epoch": 0.59,
"learning_rate": 0.000208555667001003,
"loss": 2.6805,
"theoretical_loss": 3.4394552583230773,
"tokens_seen": 1937899520
},
{
"epoch": 0.59,
"learning_rate": 0.0002085456369107322,
"loss": 2.5863,
"theoretical_loss": 3.4394455001333295,
"tokens_seen": 1937965056
},
{
"epoch": 0.59,
"learning_rate": 0.00020853560682046137,
"loss": 2.5565,
"theoretical_loss": 3.439435742365963,
"tokens_seen": 1938030592
},
{
"epoch": 0.59,
"learning_rate": 0.00020852557673019058,
"loss": 2.7427,
"theoretical_loss": 3.439425985020944,
"tokens_seen": 1938096128
},
{
"epoch": 0.59,
"learning_rate": 0.00020851554663991976,
"loss": 2.7243,
"theoretical_loss": 3.43941622809824,
"tokens_seen": 1938161664
},
{
"epoch": 0.59,
"objective/train/docs_used": 1095318,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.602961778640747,
"objective/train/theoretical_loss": 3.439406471597819,
"objective/train/tokens_used": 1958687200,
"theoretical_loss": 3.439406471597819,
"tokens_seen": 1938227200
},
{
"epoch": 0.59,
"learning_rate": 0.00020850551654964894,
"loss": 2.5702,
"theoretical_loss": 3.439406471597819,
"tokens_seen": 1938227200
},
{
"epoch": 0.59,
"learning_rate": 0.00020849548645937813,
"loss": 2.5483,
"theoretical_loss": 3.4393967155196483,
"tokens_seen": 1938292736
},
{
"epoch": 0.59,
"learning_rate": 0.0002084854563691073,
"loss": 2.6978,
"theoretical_loss": 3.439386959863695,
"tokens_seen": 1938358272
},
{
"epoch": 0.59,
"learning_rate": 0.00020847542627883651,
"loss": 2.9934,
"theoretical_loss": 3.439377204629927,
"tokens_seen": 1938423808
},
{
"epoch": 0.59,
"learning_rate": 0.0002084653961885657,
"loss": 2.8884,
"theoretical_loss": 3.4393674498183118,
"tokens_seen": 1938489344
},
{
"epoch": 0.59,
"learning_rate": 0.00020845536609829488,
"loss": 2.7864,
"theoretical_loss": 3.439357695428816,
"tokens_seen": 1938554880
},
{
"epoch": 0.59,
"learning_rate": 0.00020844533600802406,
"loss": 2.7712,
"theoretical_loss": 3.4393479414614085,
"tokens_seen": 1938620416
},
{
"epoch": 0.59,
"learning_rate": 0.00020843530591775327,
"loss": 2.7474,
"theoretical_loss": 3.4393381879160554,
"tokens_seen": 1938685952
},
{
"epoch": 0.59,
"learning_rate": 0.00020842527582748245,
"loss": 2.5829,
"theoretical_loss": 3.439328434792725,
"tokens_seen": 1938751488
},
{
"epoch": 0.59,
"learning_rate": 0.00020841524573721163,
"loss": 2.6284,
"theoretical_loss": 3.439318682091385,
"tokens_seen": 1938817024
},
{
"epoch": 0.59,
"learning_rate": 0.0002084052156469408,
"loss": 2.5982,
"theoretical_loss": 3.4393089298120016,
"tokens_seen": 1938882560
},
{
"epoch": 0.59,
"learning_rate": 0.00020839518555667,
"loss": 2.6907,
"theoretical_loss": 3.4392991779545437,
"tokens_seen": 1938948096
},
{
"epoch": 0.59,
"learning_rate": 0.0002083851554663992,
"loss": 2.6107,
"theoretical_loss": 3.439289426518978,
"tokens_seen": 1939013632
},
{
"epoch": 0.59,
"learning_rate": 0.00020837512537612838,
"loss": 2.7665,
"theoretical_loss": 3.4392796755052712,
"tokens_seen": 1939079168
},
{
"epoch": 0.59,
"learning_rate": 0.00020836509528585756,
"loss": 2.7955,
"theoretical_loss": 3.439269924913393,
"tokens_seen": 1939144704
},
{
"epoch": 0.59,
"learning_rate": 0.00020835506519558674,
"loss": 2.6383,
"theoretical_loss": 3.439260174743309,
"tokens_seen": 1939210240
},
{
"epoch": 0.59,
"learning_rate": 0.00020834503510531595,
"loss": 2.899,
"theoretical_loss": 3.439250424994988,
"tokens_seen": 1939275776
},
{
"epoch": 0.59,
"learning_rate": 0.00020833500501504513,
"loss": 2.5666,
"theoretical_loss": 3.4392406756683958,
"tokens_seen": 1939341312
},
{
"epoch": 0.59,
"learning_rate": 0.00020832497492477434,
"loss": 2.7067,
"theoretical_loss": 3.4392309267635017,
"tokens_seen": 1939406848
},
{
"epoch": 0.59,
"learning_rate": 0.00020831494483450352,
"loss": 2.6712,
"theoretical_loss": 3.439221178280272,
"tokens_seen": 1939472384
},
{
"epoch": 0.59,
"learning_rate": 0.0002083049147442327,
"loss": 2.7877,
"theoretical_loss": 3.439211430218675,
"tokens_seen": 1939537920
},
{
"epoch": 0.59,
"learning_rate": 0.0002082948846539619,
"loss": 2.7508,
"theoretical_loss": 3.439201682578678,
"tokens_seen": 1939603456
},
{
"epoch": 0.59,
"learning_rate": 0.0002082848545636911,
"loss": 2.5676,
"theoretical_loss": 3.439191935360249,
"tokens_seen": 1939668992
},
{
"epoch": 0.59,
"learning_rate": 0.00020827482447342027,
"loss": 2.7664,
"theoretical_loss": 3.439182188563354,
"tokens_seen": 1939734528
},
{
"epoch": 0.59,
"learning_rate": 0.00020826479438314945,
"loss": 2.754,
"theoretical_loss": 3.4391724421879615,
"tokens_seen": 1939800064
},
{
"epoch": 0.59,
"objective/train/docs_used": 1096386,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4852352142333984,
"objective/train/theoretical_loss": 3.4391626962340394,
"objective/train/tokens_used": 1960325600,
"theoretical_loss": 3.4391626962340394,
"tokens_seen": 1939865600
},
{
"epoch": 0.59,
"learning_rate": 0.00020825476429287866,
"loss": 2.9056,
"theoretical_loss": 3.4391626962340394,
"tokens_seen": 1939865600
},
{
"epoch": 0.59,
"learning_rate": 0.00020824473420260784,
"loss": 2.8046,
"theoretical_loss": 3.439152950701555,
"tokens_seen": 1939931136
},
{
"epoch": 0.59,
"learning_rate": 0.00020823470411233702,
"loss": 2.674,
"theoretical_loss": 3.439143205590475,
"tokens_seen": 1939996672
},
{
"epoch": 0.59,
"learning_rate": 0.0002082246740220662,
"loss": 2.473,
"theoretical_loss": 3.439133460900768,
"tokens_seen": 1940062208
},
{
"epoch": 0.59,
"learning_rate": 0.0002082146439317954,
"loss": 2.9396,
"theoretical_loss": 3.439123716632401,
"tokens_seen": 1940127744
},
{
"epoch": 0.59,
"learning_rate": 0.0002082046138415246,
"loss": 2.8954,
"theoretical_loss": 3.439113972785342,
"tokens_seen": 1940193280
},
{
"epoch": 0.59,
"learning_rate": 0.00020819458375125378,
"loss": 2.662,
"theoretical_loss": 3.4391042293595584,
"tokens_seen": 1940258816
},
{
"epoch": 0.59,
"learning_rate": 0.00020818455366098296,
"loss": 2.7799,
"theoretical_loss": 3.4390944863550175,
"tokens_seen": 1940324352
},
{
"epoch": 0.59,
"learning_rate": 0.00020817452357071214,
"loss": 2.6855,
"theoretical_loss": 3.4390847437716863,
"tokens_seen": 1940389888
},
{
"epoch": 0.59,
"learning_rate": 0.00020816449348044135,
"loss": 2.5519,
"theoretical_loss": 3.4390750016095337,
"tokens_seen": 1940455424
},
{
"epoch": 0.59,
"learning_rate": 0.00020815446339017053,
"loss": 2.6376,
"theoretical_loss": 3.439065259868527,
"tokens_seen": 1940520960
},
{
"epoch": 0.59,
"learning_rate": 0.0002081444332998997,
"loss": 2.7254,
"theoretical_loss": 3.4390555185486327,
"tokens_seen": 1940586496
},
{
"epoch": 0.59,
"learning_rate": 0.0002081344032096289,
"loss": 2.6256,
"theoretical_loss": 3.4390457776498193,
"tokens_seen": 1940652032
},
{
"epoch": 0.59,
"learning_rate": 0.0002081243731193581,
"loss": 2.7027,
"theoretical_loss": 3.4390360371720545,
"tokens_seen": 1940717568
},
{
"epoch": 0.59,
"learning_rate": 0.00020811434302908728,
"loss": 2.4751,
"theoretical_loss": 3.439026297115305,
"tokens_seen": 1940783104
},
{
"epoch": 0.59,
"learning_rate": 0.00020810431293881646,
"loss": 2.7145,
"theoretical_loss": 3.4390165574795395,
"tokens_seen": 1940848640
},
{
"epoch": 0.59,
"learning_rate": 0.00020809428284854564,
"loss": 2.7121,
"theoretical_loss": 3.4390068182647244,
"tokens_seen": 1940914176
},
{
"epoch": 0.59,
"learning_rate": 0.00020808425275827482,
"loss": 2.748,
"theoretical_loss": 3.4389970794708287,
"tokens_seen": 1940979712
},
{
"epoch": 0.59,
"learning_rate": 0.00020807422266800403,
"loss": 2.611,
"theoretical_loss": 3.4389873410978185,
"tokens_seen": 1941045248
},
{
"epoch": 0.59,
"learning_rate": 0.0002080641925777332,
"loss": 2.5098,
"theoretical_loss": 3.4389776031456623,
"tokens_seen": 1941110784
},
{
"epoch": 0.59,
"learning_rate": 0.0002080541624874624,
"loss": 2.7992,
"theoretical_loss": 3.4389678656143277,
"tokens_seen": 1941176320
},
{
"epoch": 0.59,
"learning_rate": 0.00020804413239719157,
"loss": 2.6197,
"theoretical_loss": 3.4389581285037822,
"tokens_seen": 1941241856
},
{
"epoch": 0.59,
"learning_rate": 0.00020803410230692078,
"loss": 2.715,
"theoretical_loss": 3.4389483918139927,
"tokens_seen": 1941307392
},
{
"epoch": 0.59,
"learning_rate": 0.00020802407221664996,
"loss": 2.7149,
"theoretical_loss": 3.438938655544928,
"tokens_seen": 1941372928
},
{
"epoch": 0.59,
"learning_rate": 0.00020801404212637914,
"loss": 2.7448,
"theoretical_loss": 3.438928919696555,
"tokens_seen": 1941438464
},
{
"epoch": 0.59,
"objective/train/docs_used": 1097058,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.609701633453369,
"objective/train/theoretical_loss": 3.4389191842688414,
"objective/train/tokens_used": 1961964000,
"theoretical_loss": 3.4389191842688414,
"tokens_seen": 1941504000
},
{
"epoch": 0.59,
"learning_rate": 0.00020800401203610833,
"loss": 2.6264,
"theoretical_loss": 3.4389191842688414,
"tokens_seen": 1941504000
},
{
"epoch": 0.59,
"learning_rate": 0.0002079939819458375,
"loss": 2.7938,
"theoretical_loss": 3.438909449261755,
"tokens_seen": 1941569536
},
{
"epoch": 0.59,
"learning_rate": 0.00020798395185556671,
"loss": 2.791,
"theoretical_loss": 3.4388997146752636,
"tokens_seen": 1941635072
},
{
"epoch": 0.59,
"learning_rate": 0.0002079739217652959,
"loss": 2.8284,
"theoretical_loss": 3.4388899805093343,
"tokens_seen": 1941700608
},
{
"epoch": 0.59,
"learning_rate": 0.00020796389167502508,
"loss": 2.7867,
"theoretical_loss": 3.438880246763935,
"tokens_seen": 1941766144
},
{
"epoch": 0.59,
"learning_rate": 0.00020795386158475426,
"loss": 2.5886,
"theoretical_loss": 3.438870513439033,
"tokens_seen": 1941831680
},
{
"epoch": 0.59,
"learning_rate": 0.00020794383149448347,
"loss": 2.7791,
"theoretical_loss": 3.438860780534597,
"tokens_seen": 1941897216
},
{
"epoch": 0.59,
"learning_rate": 0.00020793380140421265,
"loss": 2.7264,
"theoretical_loss": 3.438851048050594,
"tokens_seen": 1941962752
},
{
"epoch": 0.59,
"learning_rate": 0.00020792377131394183,
"loss": 2.933,
"theoretical_loss": 3.4388413159869913,
"tokens_seen": 1942028288
},
{
"epoch": 0.59,
"learning_rate": 0.000207913741223671,
"loss": 2.7651,
"theoretical_loss": 3.438831584343757,
"tokens_seen": 1942093824
},
{
"epoch": 0.59,
"learning_rate": 0.0002079037111334002,
"loss": 2.5866,
"theoretical_loss": 3.4388218531208583,
"tokens_seen": 1942159360
},
{
"epoch": 0.59,
"learning_rate": 0.0002078936810431294,
"loss": 2.6308,
"theoretical_loss": 3.438812122318263,
"tokens_seen": 1942224896
},
{
"epoch": 0.59,
"learning_rate": 0.00020788365095285858,
"loss": 2.803,
"theoretical_loss": 3.4388023919359396,
"tokens_seen": 1942290432
},
{
"epoch": 0.59,
"learning_rate": 0.00020787362086258776,
"loss": 2.5284,
"theoretical_loss": 3.4387926619738547,
"tokens_seen": 1942355968
},
{
"epoch": 0.59,
"learning_rate": 0.00020786359077231694,
"loss": 2.6515,
"theoretical_loss": 3.438782932431976,
"tokens_seen": 1942421504
},
{
"epoch": 0.59,
"learning_rate": 0.00020785356068204615,
"loss": 2.6901,
"theoretical_loss": 3.438773203310272,
"tokens_seen": 1942487040
},
{
"epoch": 0.59,
"learning_rate": 0.00020784353059177533,
"loss": 2.6684,
"theoretical_loss": 3.4387634746087103,
"tokens_seen": 1942552576
},
{
"epoch": 0.59,
"learning_rate": 0.0002078335005015045,
"loss": 2.5815,
"theoretical_loss": 3.4387537463272575,
"tokens_seen": 1942618112
},
{
"epoch": 0.59,
"learning_rate": 0.0002078234704112337,
"loss": 2.7062,
"theoretical_loss": 3.438744018465882,
"tokens_seen": 1942683648
},
{
"epoch": 0.59,
"learning_rate": 0.00020781344032096288,
"loss": 2.7118,
"theoretical_loss": 3.438734291024552,
"tokens_seen": 1942749184
},
{
"epoch": 0.59,
"learning_rate": 0.00020780341023069208,
"loss": 2.5931,
"theoretical_loss": 3.438724564003235,
"tokens_seen": 1942814720
},
{
"epoch": 0.59,
"learning_rate": 0.00020779338014042126,
"loss": 2.7642,
"theoretical_loss": 3.438714837401897,
"tokens_seen": 1942880256
},
{
"epoch": 0.59,
"learning_rate": 0.00020778335005015045,
"loss": 2.7694,
"theoretical_loss": 3.4387051112205085,
"tokens_seen": 1942945792
},
{
"epoch": 0.59,
"learning_rate": 0.00020777331995987963,
"loss": 2.7553,
"theoretical_loss": 3.4386953854590354,
"tokens_seen": 1943011328
},
{
"epoch": 0.59,
"learning_rate": 0.00020776328986960884,
"loss": 2.7233,
"theoretical_loss": 3.4386856601174456,
"tokens_seen": 1943076864
},
{
"epoch": 0.59,
"objective/train/docs_used": 1098461,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.406571865081787,
"objective/train/theoretical_loss": 3.4386759351957066,
"objective/train/tokens_used": 1963602400,
"theoretical_loss": 3.4386759351957066,
"tokens_seen": 1943142400
},
{
"epoch": 0.59,
"learning_rate": 0.00020775325977933802,
"loss": 2.612,
"theoretical_loss": 3.4386759351957066,
"tokens_seen": 1943142400
},
{
"epoch": 0.59,
"learning_rate": 0.0002077432296890672,
"loss": 2.6914,
"theoretical_loss": 3.4386662106937873,
"tokens_seen": 1943207936
},
{
"epoch": 0.59,
"learning_rate": 0.00020773319959879638,
"loss": 2.7269,
"theoretical_loss": 3.438656486611654,
"tokens_seen": 1943273472
},
{
"epoch": 0.59,
"learning_rate": 0.00020772316950852556,
"loss": 2.4642,
"theoretical_loss": 3.4386467629492756,
"tokens_seen": 1943339008
},
{
"epoch": 0.59,
"learning_rate": 0.00020771313941825477,
"loss": 2.7527,
"theoretical_loss": 3.4386370397066193,
"tokens_seen": 1943404544
},
{
"epoch": 0.59,
"learning_rate": 0.00020770310932798395,
"loss": 2.5472,
"theoretical_loss": 3.4386273168836525,
"tokens_seen": 1943470080
},
{
"epoch": 0.59,
"learning_rate": 0.00020769307923771313,
"loss": 2.4666,
"theoretical_loss": 3.4386175944803434,
"tokens_seen": 1943535616
},
{
"epoch": 0.59,
"learning_rate": 0.0002076830491474423,
"loss": 2.7473,
"theoretical_loss": 3.4386078724966596,
"tokens_seen": 1943601152
},
{
"epoch": 0.59,
"learning_rate": 0.00020767301905717152,
"loss": 2.6247,
"theoretical_loss": 3.438598150932569,
"tokens_seen": 1943666688
},
{
"epoch": 0.59,
"learning_rate": 0.0002076629889669007,
"loss": 2.6612,
"theoretical_loss": 3.4385884297880387,
"tokens_seen": 1943732224
},
{
"epoch": 0.59,
"learning_rate": 0.00020765295887662988,
"loss": 2.6956,
"theoretical_loss": 3.4385787090630373,
"tokens_seen": 1943797760
},
{
"epoch": 0.59,
"learning_rate": 0.00020764292878635906,
"loss": 2.774,
"theoretical_loss": 3.438568988757532,
"tokens_seen": 1943863296
},
{
"epoch": 0.59,
"learning_rate": 0.00020763289869608824,
"loss": 2.6333,
"theoretical_loss": 3.438559268871491,
"tokens_seen": 1943928832
},
{
"epoch": 0.59,
"learning_rate": 0.00020762286860581745,
"loss": 2.5558,
"theoretical_loss": 3.4385495494048817,
"tokens_seen": 1943994368
},
{
"epoch": 0.59,
"learning_rate": 0.00020761283851554663,
"loss": 2.5808,
"theoretical_loss": 3.4385398303576715,
"tokens_seen": 1944059904
},
{
"epoch": 0.59,
"learning_rate": 0.00020760280842527581,
"loss": 2.8948,
"theoretical_loss": 3.438530111729829,
"tokens_seen": 1944125440
},
{
"epoch": 0.59,
"learning_rate": 0.000207592778335005,
"loss": 2.7491,
"theoretical_loss": 3.4385203935213218,
"tokens_seen": 1944190976
},
{
"epoch": 0.59,
"learning_rate": 0.0002075827482447342,
"loss": 2.6171,
"theoretical_loss": 3.4385106757321173,
"tokens_seen": 1944256512
},
{
"epoch": 0.59,
"learning_rate": 0.0002075727181544634,
"loss": 2.8227,
"theoretical_loss": 3.438500958362183,
"tokens_seen": 1944322048
},
{
"epoch": 0.59,
"learning_rate": 0.0002075626880641926,
"loss": 2.8719,
"theoretical_loss": 3.438491241411487,
"tokens_seen": 1944387584
},
{
"epoch": 0.59,
"learning_rate": 0.00020755265797392177,
"loss": 2.8159,
"theoretical_loss": 3.438481524879998,
"tokens_seen": 1944453120
},
{
"epoch": 0.59,
"learning_rate": 0.00020754262788365098,
"loss": 2.5465,
"theoretical_loss": 3.438471808767683,
"tokens_seen": 1944518656
},
{
"epoch": 0.59,
"learning_rate": 0.00020753259779338016,
"loss": 2.6726,
"theoretical_loss": 3.4384620930745093,
"tokens_seen": 1944584192
},
{
"epoch": 0.59,
"learning_rate": 0.00020752256770310934,
"loss": 2.576,
"theoretical_loss": 3.438452377800445,
"tokens_seen": 1944649728
},
{
"epoch": 0.59,
"learning_rate": 0.00020751253761283853,
"loss": 2.449,
"theoretical_loss": 3.4384426629454587,
"tokens_seen": 1944715264
},
{
"epoch": 0.59,
"objective/train/docs_used": 1099095,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7049999237060547,
"objective/train/theoretical_loss": 3.438432948509517,
"objective/train/tokens_used": 1965240800,
"theoretical_loss": 3.438432948509517,
"tokens_seen": 1944780800
},
{
"epoch": 0.59,
"learning_rate": 0.0002075025075225677,
"loss": 2.7617,
"theoretical_loss": 3.438432948509517,
"tokens_seen": 1944780800
},
{
"epoch": 0.59,
"learning_rate": 0.00020749247743229692,
"loss": 2.7651,
"theoretical_loss": 3.4384232344925882,
"tokens_seen": 1944846336
},
{
"epoch": 0.59,
"learning_rate": 0.0002074824473420261,
"loss": 2.7301,
"theoretical_loss": 3.438413520894641,
"tokens_seen": 1944911872
},
{
"epoch": 0.59,
"learning_rate": 0.00020747241725175528,
"loss": 2.9485,
"theoretical_loss": 3.438403807715641,
"tokens_seen": 1944977408
},
{
"epoch": 0.59,
"learning_rate": 0.00020746238716148446,
"loss": 2.7494,
"theoretical_loss": 3.4383940949555587,
"tokens_seen": 1945042944
},
{
"epoch": 0.59,
"learning_rate": 0.00020745235707121367,
"loss": 2.885,
"theoretical_loss": 3.43838438261436,
"tokens_seen": 1945108480
},
{
"epoch": 0.59,
"learning_rate": 0.00020744232698094285,
"loss": 2.6543,
"theoretical_loss": 3.4383746706920135,
"tokens_seen": 1945174016
},
{
"epoch": 0.59,
"learning_rate": 0.00020743229689067203,
"loss": 2.8275,
"theoretical_loss": 3.438364959188487,
"tokens_seen": 1945239552
},
{
"epoch": 0.59,
"learning_rate": 0.0002074222668004012,
"loss": 2.6417,
"theoretical_loss": 3.438355248103748,
"tokens_seen": 1945305088
},
{
"epoch": 0.59,
"learning_rate": 0.0002074122367101304,
"loss": 2.7073,
"theoretical_loss": 3.4383455374377645,
"tokens_seen": 1945370624
},
{
"epoch": 0.59,
"learning_rate": 0.0002074022066198596,
"loss": 2.6083,
"theoretical_loss": 3.4383358271905045,
"tokens_seen": 1945436160
},
{
"epoch": 0.59,
"learning_rate": 0.00020739217652958878,
"loss": 2.952,
"theoretical_loss": 3.4383261173619353,
"tokens_seen": 1945501696
},
{
"epoch": 0.59,
"learning_rate": 0.00020738214643931796,
"loss": 2.6783,
"theoretical_loss": 3.438316407952026,
"tokens_seen": 1945567232
},
{
"epoch": 0.59,
"learning_rate": 0.00020737211634904714,
"loss": 2.8211,
"theoretical_loss": 3.438306698960743,
"tokens_seen": 1945632768
},
{
"epoch": 0.59,
"learning_rate": 0.00020736208625877635,
"loss": 2.759,
"theoretical_loss": 3.4382969903880545,
"tokens_seen": 1945698304
},
{
"epoch": 0.59,
"learning_rate": 0.00020735205616850553,
"loss": 2.7112,
"theoretical_loss": 3.4382872822339294,
"tokens_seen": 1945763840
},
{
"epoch": 0.59,
"learning_rate": 0.0002073420260782347,
"loss": 2.7098,
"theoretical_loss": 3.4382775744983345,
"tokens_seen": 1945829376
},
{
"epoch": 0.59,
"learning_rate": 0.0002073319959879639,
"loss": 2.7478,
"theoretical_loss": 3.4382678671812377,
"tokens_seen": 1945894912
},
{
"epoch": 0.59,
"learning_rate": 0.00020732196589769308,
"loss": 2.6171,
"theoretical_loss": 3.438258160282607,
"tokens_seen": 1945960448
},
{
"epoch": 0.59,
"learning_rate": 0.00020731193580742228,
"loss": 2.7084,
"theoretical_loss": 3.4382484538024105,
"tokens_seen": 1946025984
},
{
"epoch": 0.59,
"learning_rate": 0.00020730190571715146,
"loss": 2.7807,
"theoretical_loss": 3.438238747740616,
"tokens_seen": 1946091520
},
{
"epoch": 0.59,
"learning_rate": 0.00020729187562688065,
"loss": 2.692,
"theoretical_loss": 3.4382290420971913,
"tokens_seen": 1946157056
},
{
"epoch": 0.59,
"learning_rate": 0.00020728184553660983,
"loss": 2.7801,
"theoretical_loss": 3.438219336872104,
"tokens_seen": 1946222592
},
{
"epoch": 0.59,
"learning_rate": 0.00020727181544633904,
"loss": 2.6912,
"theoretical_loss": 3.4382096320653224,
"tokens_seen": 1946288128
},
{
"epoch": 0.59,
"learning_rate": 0.00020726178535606822,
"loss": 2.5327,
"theoretical_loss": 3.4381999276768145,
"tokens_seen": 1946353664
},
{
"epoch": 0.59,
"objective/train/docs_used": 1099984,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7089033126831055,
"objective/train/theoretical_loss": 3.438190223706548,
"objective/train/tokens_used": 1966879200,
"theoretical_loss": 3.438190223706548,
"tokens_seen": 1946419200
},
{
"epoch": 0.59,
"learning_rate": 0.0002072517552657974,
"loss": 2.671,
"theoretical_loss": 3.438190223706548,
"tokens_seen": 1946419200
},
{
"epoch": 0.59,
"learning_rate": 0.00020724172517552658,
"loss": 2.6285,
"theoretical_loss": 3.43818052015449,
"tokens_seen": 1946484736
},
{
"epoch": 0.59,
"learning_rate": 0.00020723169508525576,
"loss": 2.7106,
"theoretical_loss": 3.43817081702061,
"tokens_seen": 1946550272
},
{
"epoch": 0.59,
"learning_rate": 0.00020722166499498497,
"loss": 2.778,
"theoretical_loss": 3.4381611143048745,
"tokens_seen": 1946615808
},
{
"epoch": 0.59,
"learning_rate": 0.00020721163490471415,
"loss": 2.6522,
"theoretical_loss": 3.4381514120072523,
"tokens_seen": 1946681344
},
{
"epoch": 0.59,
"learning_rate": 0.00020720160481444333,
"loss": 2.7189,
"theoretical_loss": 3.4381417101277107,
"tokens_seen": 1946746880
},
{
"epoch": 0.59,
"learning_rate": 0.0002071915747241725,
"loss": 2.9005,
"theoretical_loss": 3.438132008666218,
"tokens_seen": 1946812416
},
{
"epoch": 0.59,
"learning_rate": 0.00020718154463390172,
"loss": 2.7335,
"theoretical_loss": 3.438122307622742,
"tokens_seen": 1946877952
},
{
"epoch": 0.59,
"learning_rate": 0.0002071715145436309,
"loss": 2.8407,
"theoretical_loss": 3.43811260699725,
"tokens_seen": 1946943488
},
{
"epoch": 0.59,
"learning_rate": 0.00020716148445336008,
"loss": 2.8036,
"theoretical_loss": 3.4381029067897115,
"tokens_seen": 1947009024
},
{
"epoch": 0.59,
"learning_rate": 0.00020715145436308926,
"loss": 2.795,
"theoretical_loss": 3.438093207000093,
"tokens_seen": 1947074560
},
{
"epoch": 0.59,
"learning_rate": 0.00020714142427281844,
"loss": 2.8256,
"theoretical_loss": 3.4380835076283622,
"tokens_seen": 1947140096
},
{
"epoch": 0.59,
"learning_rate": 0.00020713139418254765,
"loss": 2.5133,
"theoretical_loss": 3.4380738086744884,
"tokens_seen": 1947205632
},
{
"epoch": 0.59,
"learning_rate": 0.00020712136409227683,
"loss": 2.8447,
"theoretical_loss": 3.4380641101384386,
"tokens_seen": 1947271168
},
{
"epoch": 0.59,
"learning_rate": 0.00020711133400200601,
"loss": 2.6122,
"theoretical_loss": 3.4380544120201812,
"tokens_seen": 1947336704
},
{
"epoch": 0.59,
"learning_rate": 0.0002071013039117352,
"loss": 2.8855,
"theoretical_loss": 3.438044714319684,
"tokens_seen": 1947402240
},
{
"epoch": 0.59,
"learning_rate": 0.0002070912738214644,
"loss": 2.752,
"theoretical_loss": 3.4380350170369143,
"tokens_seen": 1947467776
},
{
"epoch": 0.59,
"learning_rate": 0.00020708124373119358,
"loss": 2.79,
"theoretical_loss": 3.438025320171841,
"tokens_seen": 1947533312
},
{
"epoch": 0.59,
"learning_rate": 0.00020707121364092277,
"loss": 2.8276,
"theoretical_loss": 3.438015623724432,
"tokens_seen": 1947598848
},
{
"epoch": 0.59,
"learning_rate": 0.00020706118355065195,
"loss": 2.8808,
"theoretical_loss": 3.4380059276946544,
"tokens_seen": 1947664384
},
{
"epoch": 0.59,
"learning_rate": 0.00020705115346038116,
"loss": 2.8175,
"theoretical_loss": 3.437996232082477,
"tokens_seen": 1947729920
},
{
"epoch": 0.59,
"learning_rate": 0.00020704112337011034,
"loss": 2.7806,
"theoretical_loss": 3.437986536887867,
"tokens_seen": 1947795456
},
{
"epoch": 0.59,
"learning_rate": 0.00020703109327983952,
"loss": 2.7321,
"theoretical_loss": 3.437976842110793,
"tokens_seen": 1947860992
},
{
"epoch": 0.59,
"learning_rate": 0.0002070210631895687,
"loss": 2.9148,
"theoretical_loss": 3.4379671477512233,
"tokens_seen": 1947926528
},
{
"epoch": 0.59,
"learning_rate": 0.00020701103309929788,
"loss": 2.6502,
"theoretical_loss": 3.4379574538091244,
"tokens_seen": 1947992064
},
{
"epoch": 0.59,
"objective/train/docs_used": 1101410,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.496291160583496,
"objective/train/theoretical_loss": 3.437947760284466,
"objective/train/tokens_used": 1968517600,
"theoretical_loss": 3.437947760284466,
"tokens_seen": 1948057600
},
{
"epoch": 0.59,
"learning_rate": 0.0002070010030090271,
"loss": 2.7358,
"theoretical_loss": 3.437947760284466,
"tokens_seen": 1948057600
},
{
"epoch": 0.59,
"learning_rate": 0.00020699097291875627,
"loss": 2.9111,
"theoretical_loss": 3.437938067177215,
"tokens_seen": 1948123136
},
{
"epoch": 0.59,
"learning_rate": 0.00020698094282848545,
"loss": 2.7954,
"theoretical_loss": 3.43792837448734,
"tokens_seen": 1948188672
},
{
"epoch": 0.59,
"learning_rate": 0.00020697091273821463,
"loss": 2.8223,
"theoretical_loss": 3.437918682214808,
"tokens_seen": 1948254208
},
{
"epoch": 0.59,
"learning_rate": 0.00020696088264794384,
"loss": 2.7844,
"theoretical_loss": 3.4379089903595883,
"tokens_seen": 1948319744
},
{
"epoch": 0.59,
"learning_rate": 0.00020695085255767302,
"loss": 2.5401,
"theoretical_loss": 3.437899298921648,
"tokens_seen": 1948385280
},
{
"epoch": 0.59,
"learning_rate": 0.0002069408224674022,
"loss": 2.6615,
"theoretical_loss": 3.4378896079009555,
"tokens_seen": 1948450816
},
{
"epoch": 0.59,
"learning_rate": 0.00020693079237713138,
"loss": 2.5651,
"theoretical_loss": 3.437879917297479,
"tokens_seen": 1948516352
},
{
"epoch": 0.59,
"learning_rate": 0.00020692076228686056,
"loss": 2.6765,
"theoretical_loss": 3.437870227111186,
"tokens_seen": 1948581888
},
{
"epoch": 0.59,
"learning_rate": 0.00020691073219658977,
"loss": 2.9292,
"theoretical_loss": 3.4378605373420443,
"tokens_seen": 1948647424
},
{
"epoch": 0.59,
"learning_rate": 0.00020690070210631895,
"loss": 2.9369,
"theoretical_loss": 3.4378508479900227,
"tokens_seen": 1948712960
},
{
"epoch": 0.59,
"learning_rate": 0.00020689067201604813,
"loss": 2.6514,
"theoretical_loss": 3.437841159055089,
"tokens_seen": 1948778496
},
{
"epoch": 0.59,
"learning_rate": 0.00020688064192577732,
"loss": 2.6649,
"theoretical_loss": 3.4378314705372106,
"tokens_seen": 1948844032
},
{
"epoch": 0.59,
"learning_rate": 0.00020687061183550652,
"loss": 2.8419,
"theoretical_loss": 3.437821782436356,
"tokens_seen": 1948909568
},
{
"epoch": 0.59,
"learning_rate": 0.0002068605817452357,
"loss": 2.6036,
"theoretical_loss": 3.4378120947524935,
"tokens_seen": 1948975104
},
{
"epoch": 0.59,
"learning_rate": 0.00020685055165496489,
"loss": 2.6873,
"theoretical_loss": 3.4378024074855906,
"tokens_seen": 1949040640
},
{
"epoch": 0.59,
"learning_rate": 0.00020684052156469407,
"loss": 2.6786,
"theoretical_loss": 3.437792720635616,
"tokens_seen": 1949106176
},
{
"epoch": 0.59,
"learning_rate": 0.00020683049147442328,
"loss": 2.529,
"theoretical_loss": 3.437783034202537,
"tokens_seen": 1949171712
},
{
"epoch": 0.59,
"learning_rate": 0.00020682046138415248,
"loss": 2.654,
"theoretical_loss": 3.4377733481863215,
"tokens_seen": 1949237248
},
{
"epoch": 0.59,
"learning_rate": 0.00020681043129388166,
"loss": 2.6748,
"theoretical_loss": 3.4377636625869386,
"tokens_seen": 1949302784
},
{
"epoch": 0.59,
"learning_rate": 0.00020680040120361085,
"loss": 2.668,
"theoretical_loss": 3.4377539774043555,
"tokens_seen": 1949368320
},
{
"epoch": 0.59,
"learning_rate": 0.00020679037111334003,
"loss": 2.729,
"theoretical_loss": 3.4377442926385404,
"tokens_seen": 1949433856
},
{
"epoch": 0.59,
"learning_rate": 0.00020678034102306924,
"loss": 2.8314,
"theoretical_loss": 3.4377346082894618,
"tokens_seen": 1949499392
},
{
"epoch": 0.59,
"learning_rate": 0.00020677031093279842,
"loss": 2.4557,
"theoretical_loss": 3.437724924357087,
"tokens_seen": 1949564928
},
{
"epoch": 0.59,
"learning_rate": 0.0002067602808425276,
"loss": 2.8314,
"theoretical_loss": 3.437715240841385,
"tokens_seen": 1949630464
},
{
"epoch": 0.59,
"objective/train/docs_used": 1102181,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7315518856048584,
"objective/train/theoretical_loss": 3.437705557742323,
"objective/train/tokens_used": 1970156000,
"theoretical_loss": 3.437705557742323,
"tokens_seen": 1949696000
},
{
"epoch": 0.59,
"learning_rate": 0.00020675025075225678,
"loss": 2.7198,
"theoretical_loss": 3.437705557742323,
"tokens_seen": 1949696000
},
{
"epoch": 0.59,
"learning_rate": 0.00020674022066198596,
"loss": 2.9216,
"theoretical_loss": 3.4376958750598696,
"tokens_seen": 1949761536
},
{
"epoch": 0.59,
"learning_rate": 0.00020673019057171517,
"loss": 2.6321,
"theoretical_loss": 3.437686192793992,
"tokens_seen": 1949827072
},
{
"epoch": 0.59,
"learning_rate": 0.00020672016048144435,
"loss": 2.6753,
"theoretical_loss": 3.4376765109446596,
"tokens_seen": 1949892608
},
{
"epoch": 0.59,
"learning_rate": 0.00020671013039117353,
"loss": 2.7916,
"theoretical_loss": 3.4376668295118398,
"tokens_seen": 1949958144
},
{
"epoch": 0.59,
"learning_rate": 0.0002067001003009027,
"loss": 2.675,
"theoretical_loss": 3.4376571484955005,
"tokens_seen": 1950023680
},
{
"epoch": 0.59,
"learning_rate": 0.00020669007021063192,
"loss": 2.699,
"theoretical_loss": 3.4376474678956104,
"tokens_seen": 1950089216
},
{
"epoch": 0.59,
"learning_rate": 0.0002066800401203611,
"loss": 2.6766,
"theoretical_loss": 3.437637787712137,
"tokens_seen": 1950154752
},
{
"epoch": 0.59,
"learning_rate": 0.00020667001003009028,
"loss": 2.437,
"theoretical_loss": 3.4376281079450486,
"tokens_seen": 1950220288
},
{
"epoch": 0.59,
"learning_rate": 0.00020665997993981946,
"loss": 2.9064,
"theoretical_loss": 3.437618428594313,
"tokens_seen": 1950285824
},
{
"epoch": 0.59,
"learning_rate": 0.00020664994984954864,
"loss": 2.9135,
"theoretical_loss": 3.4376087496598986,
"tokens_seen": 1950351360
},
{
"epoch": 0.59,
"learning_rate": 0.00020663991975927785,
"loss": 2.8468,
"theoretical_loss": 3.437599071141774,
"tokens_seen": 1950416896
},
{
"epoch": 0.59,
"learning_rate": 0.00020662988966900703,
"loss": 2.6627,
"theoretical_loss": 3.4375893930399064,
"tokens_seen": 1950482432
},
{
"epoch": 0.59,
"learning_rate": 0.00020661985957873621,
"loss": 2.4759,
"theoretical_loss": 3.4375797153542647,
"tokens_seen": 1950547968
},
{
"epoch": 0.59,
"learning_rate": 0.0002066098294884654,
"loss": 2.9769,
"theoretical_loss": 3.437570038084816,
"tokens_seen": 1950613504
},
{
"epoch": 0.59,
"learning_rate": 0.0002065997993981946,
"loss": 2.7715,
"theoretical_loss": 3.4375603612315295,
"tokens_seen": 1950679040
},
{
"epoch": 0.59,
"learning_rate": 0.00020658976930792379,
"loss": 2.6444,
"theoretical_loss": 3.437550684794373,
"tokens_seen": 1950744576
},
{
"epoch": 0.59,
"learning_rate": 0.00020657973921765297,
"loss": 2.8026,
"theoretical_loss": 3.4375410087733145,
"tokens_seen": 1950810112
},
{
"epoch": 0.59,
"learning_rate": 0.00020656970912738215,
"loss": 2.6327,
"theoretical_loss": 3.437531333168322,
"tokens_seen": 1950875648
},
{
"epoch": 0.59,
"learning_rate": 0.00020655967903711136,
"loss": 2.8725,
"theoretical_loss": 3.4375216579793637,
"tokens_seen": 1950941184
},
{
"epoch": 0.59,
"learning_rate": 0.00020654964894684054,
"loss": 2.6778,
"theoretical_loss": 3.4375119832064076,
"tokens_seen": 1951006720
},
{
"epoch": 0.59,
"learning_rate": 0.00020653961885656972,
"loss": 2.6703,
"theoretical_loss": 3.4375023088494228,
"tokens_seen": 1951072256
},
{
"epoch": 0.59,
"learning_rate": 0.0002065295887662989,
"loss": 2.6438,
"theoretical_loss": 3.4374926349083763,
"tokens_seen": 1951137792
},
{
"epoch": 0.59,
"learning_rate": 0.00020651955867602808,
"loss": 2.9545,
"theoretical_loss": 3.4374829613832363,
"tokens_seen": 1951203328
},
{
"epoch": 0.59,
"learning_rate": 0.0002065095285857573,
"loss": 2.6205,
"theoretical_loss": 3.4374732882739716,
"tokens_seen": 1951268864
},
{
"epoch": 0.59,
"objective/train/docs_used": 1102842,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.493302345275879,
"objective/train/theoretical_loss": 3.43746361558055,
"objective/train/tokens_used": 1971794400,
"theoretical_loss": 3.43746361558055,
"tokens_seen": 1951334400
},
{
"epoch": 0.59,
"learning_rate": 0.00020649949849548647,
"loss": 2.4714,
"theoretical_loss": 3.43746361558055,
"tokens_seen": 1951334400
},
{
"epoch": 0.59,
"learning_rate": 0.00020648946840521565,
"loss": 2.763,
"theoretical_loss": 3.4374539433029394,
"tokens_seen": 1951399936
},
{
"epoch": 0.59,
"learning_rate": 0.00020647943831494483,
"loss": 2.784,
"theoretical_loss": 3.437444271441109,
"tokens_seen": 1951465472
},
{
"epoch": 0.59,
"learning_rate": 0.00020646940822467404,
"loss": 2.7434,
"theoretical_loss": 3.4374345999950253,
"tokens_seen": 1951531008
},
{
"epoch": 0.59,
"learning_rate": 0.00020645937813440322,
"loss": 2.6784,
"theoretical_loss": 3.437424928964658,
"tokens_seen": 1951596544
},
{
"epoch": 0.59,
"learning_rate": 0.0002064493480441324,
"loss": 2.6525,
"theoretical_loss": 3.437415258349975,
"tokens_seen": 1951662080
},
{
"epoch": 0.59,
"learning_rate": 0.00020643931795386158,
"loss": 2.6696,
"theoretical_loss": 3.437405588150943,
"tokens_seen": 1951727616
},
{
"epoch": 0.59,
"learning_rate": 0.00020642928786359076,
"loss": 2.8976,
"theoretical_loss": 3.437395918367532,
"tokens_seen": 1951793152
},
{
"epoch": 0.59,
"learning_rate": 0.00020641925777331997,
"loss": 2.8165,
"theoretical_loss": 3.43738624899971,
"tokens_seen": 1951858688
},
{
"epoch": 0.59,
"learning_rate": 0.00020640922768304915,
"loss": 2.5472,
"theoretical_loss": 3.437376580047444,
"tokens_seen": 1951924224
},
{
"epoch": 0.59,
"learning_rate": 0.00020639919759277833,
"loss": 2.6825,
"theoretical_loss": 3.437366911510703,
"tokens_seen": 1951989760
},
{
"epoch": 0.59,
"learning_rate": 0.00020638916750250752,
"loss": 2.6586,
"theoretical_loss": 3.437357243389455,
"tokens_seen": 1952055296
},
{
"epoch": 0.59,
"learning_rate": 0.00020637913741223672,
"loss": 2.5645,
"theoretical_loss": 3.437347575683668,
"tokens_seen": 1952120832
},
{
"epoch": 0.59,
"learning_rate": 0.0002063691073219659,
"loss": 2.5208,
"theoretical_loss": 3.437337908393311,
"tokens_seen": 1952186368
},
{
"epoch": 0.59,
"learning_rate": 0.00020635907723169509,
"loss": 2.4667,
"theoretical_loss": 3.437328241518352,
"tokens_seen": 1952251904
},
{
"epoch": 0.59,
"learning_rate": 0.00020634904714142427,
"loss": 2.7227,
"theoretical_loss": 3.4373185750587583,
"tokens_seen": 1952317440
},
{
"epoch": 0.59,
"learning_rate": 0.00020633901705115345,
"loss": 2.5957,
"theoretical_loss": 3.4373089090144986,
"tokens_seen": 1952382976
},
{
"epoch": 0.59,
"learning_rate": 0.00020632898696088266,
"loss": 2.6768,
"theoretical_loss": 3.4372992433855414,
"tokens_seen": 1952448512
},
{
"epoch": 0.59,
"learning_rate": 0.00020631895687061184,
"loss": 2.7777,
"theoretical_loss": 3.4372895781718547,
"tokens_seen": 1952514048
},
{
"epoch": 0.59,
"learning_rate": 0.00020630892678034102,
"loss": 2.5946,
"theoretical_loss": 3.437279913373407,
"tokens_seen": 1952579584
},
{
"epoch": 0.59,
"learning_rate": 0.0002062988966900702,
"loss": 2.7337,
"theoretical_loss": 3.437270248990166,
"tokens_seen": 1952645120
},
{
"epoch": 0.59,
"learning_rate": 0.0002062888665997994,
"loss": 2.6984,
"theoretical_loss": 3.4372605850221003,
"tokens_seen": 1952710656
},
{
"epoch": 0.59,
"learning_rate": 0.0002062788365095286,
"loss": 2.602,
"theoretical_loss": 3.437250921469178,
"tokens_seen": 1952776192
},
{
"epoch": 0.59,
"learning_rate": 0.00020626880641925777,
"loss": 2.6509,
"theoretical_loss": 3.437241258331367,
"tokens_seen": 1952841728
},
{
"epoch": 0.59,
"learning_rate": 0.00020625877632898695,
"loss": 2.5923,
"theoretical_loss": 3.437231595608636,
"tokens_seen": 1952907264
},
{
"epoch": 0.59,
"objective/train/docs_used": 1104173,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.045996904373169,
"objective/train/theoretical_loss": 3.4372219333009535,
"objective/train/tokens_used": 1973432800,
"theoretical_loss": 3.4372219333009535,
"tokens_seen": 1952972800
},
{
"epoch": 0.59,
"learning_rate": 0.00020624874623871613,
"loss": 2.8485,
"theoretical_loss": 3.4372219333009535,
"tokens_seen": 1952972800
},
{
"epoch": 0.59,
"learning_rate": 0.00020623871614844534,
"loss": 2.7965,
"theoretical_loss": 3.437212271408287,
"tokens_seen": 1953038336
},
{
"epoch": 0.59,
"learning_rate": 0.00020622868605817452,
"loss": 2.6931,
"theoretical_loss": 3.4372026099306057,
"tokens_seen": 1953103872
},
{
"epoch": 0.59,
"learning_rate": 0.0002062186559679037,
"loss": 2.6492,
"theoretical_loss": 3.4371929488678767,
"tokens_seen": 1953169408
},
{
"epoch": 0.59,
"learning_rate": 0.00020620862587763288,
"loss": 2.6936,
"theoretical_loss": 3.4371832882200692,
"tokens_seen": 1953234944
},
{
"epoch": 0.59,
"learning_rate": 0.0002061985957873621,
"loss": 2.5059,
"theoretical_loss": 3.4371736279871508,
"tokens_seen": 1953300480
},
{
"epoch": 0.59,
"learning_rate": 0.00020618856569709127,
"loss": 2.7039,
"theoretical_loss": 3.43716396816909,
"tokens_seen": 1953366016
},
{
"epoch": 0.59,
"learning_rate": 0.00020617853560682046,
"loss": 2.5376,
"theoretical_loss": 3.4371543087658556,
"tokens_seen": 1953431552
},
{
"epoch": 0.59,
"learning_rate": 0.00020616850551654964,
"loss": 2.4308,
"theoretical_loss": 3.437144649777415,
"tokens_seen": 1953497088
},
{
"epoch": 0.59,
"learning_rate": 0.00020615847542627882,
"loss": 2.6088,
"theoretical_loss": 3.4371349912037372,
"tokens_seen": 1953562624
},
{
"epoch": 0.59,
"learning_rate": 0.00020614844533600803,
"loss": 2.7009,
"theoretical_loss": 3.43712533304479,
"tokens_seen": 1953628160
},
{
"epoch": 0.59,
"learning_rate": 0.0002061384152457372,
"loss": 2.7198,
"theoretical_loss": 3.437115675300542,
"tokens_seen": 1953693696
},
{
"epoch": 0.59,
"learning_rate": 0.0002061283851554664,
"loss": 2.5417,
"theoretical_loss": 3.4371060179709607,
"tokens_seen": 1953759232
},
{
"epoch": 0.59,
"learning_rate": 0.00020611835506519557,
"loss": 2.7824,
"theoretical_loss": 3.4370963610560157,
"tokens_seen": 1953824768
},
{
"epoch": 0.59,
"learning_rate": 0.00020610832497492478,
"loss": 2.7042,
"theoretical_loss": 3.437086704555674,
"tokens_seen": 1953890304
},
{
"epoch": 0.59,
"learning_rate": 0.00020609829488465396,
"loss": 2.5964,
"theoretical_loss": 3.437077048469905,
"tokens_seen": 1953955840
},
{
"epoch": 0.59,
"learning_rate": 0.00020608826479438314,
"loss": 2.6188,
"theoretical_loss": 3.437067392798676,
"tokens_seen": 1954021376
},
{
"epoch": 0.59,
"learning_rate": 0.00020607823470411235,
"loss": 2.6058,
"theoretical_loss": 3.4370577375419566,
"tokens_seen": 1954086912
},
{
"epoch": 0.59,
"learning_rate": 0.00020606820461384156,
"loss": 2.6024,
"theoretical_loss": 3.4370480826997136,
"tokens_seen": 1954152448
},
{
"epoch": 0.59,
"learning_rate": 0.00020605817452357074,
"loss": 2.6146,
"theoretical_loss": 3.437038428271916,
"tokens_seen": 1954217984
},
{
"epoch": 0.59,
"learning_rate": 0.00020604814443329992,
"loss": 2.6052,
"theoretical_loss": 3.4370287742585326,
"tokens_seen": 1954283520
},
{
"epoch": 0.59,
"learning_rate": 0.0002060381143430291,
"loss": 2.6757,
"theoretical_loss": 3.4370191206595306,
"tokens_seen": 1954349056
},
{
"epoch": 0.59,
"learning_rate": 0.00020602808425275828,
"loss": 2.6971,
"theoretical_loss": 3.437009467474879,
"tokens_seen": 1954414592
},
{
"epoch": 0.59,
"learning_rate": 0.0002060180541624875,
"loss": 2.7031,
"theoretical_loss": 3.4369998147045466,
"tokens_seen": 1954480128
},
{
"epoch": 0.59,
"learning_rate": 0.00020600802407221667,
"loss": 2.909,
"theoretical_loss": 3.4369901623485006,
"tokens_seen": 1954545664
},
{
"epoch": 0.59,
"objective/train/docs_used": 1104920,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.47347354888916,
"objective/train/theoretical_loss": 3.43698051040671,
"objective/train/tokens_used": 1975071200,
"theoretical_loss": 3.43698051040671,
"tokens_seen": 1954611200
},
{
"epoch": 0.59,
"learning_rate": 0.00020599799398194585,
"loss": 2.6039,
"theoretical_loss": 3.43698051040671,
"tokens_seen": 1954611200
},
{
"epoch": 0.59,
"learning_rate": 0.00020598796389167503,
"loss": 2.9698,
"theoretical_loss": 3.4369708588791434,
"tokens_seen": 1954676736
},
{
"epoch": 0.59,
"learning_rate": 0.00020597793380140424,
"loss": 2.4051,
"theoretical_loss": 3.4369612077657687,
"tokens_seen": 1954742272
},
{
"epoch": 0.59,
"learning_rate": 0.00020596790371113342,
"loss": 2.5429,
"theoretical_loss": 3.4369515570665543,
"tokens_seen": 1954807808
},
{
"epoch": 0.59,
"learning_rate": 0.0002059578736208626,
"loss": 2.6479,
"theoretical_loss": 3.4369419067814686,
"tokens_seen": 1954873344
},
{
"epoch": 0.59,
"learning_rate": 0.00020594784353059178,
"loss": 2.7987,
"theoretical_loss": 3.4369322569104797,
"tokens_seen": 1954938880
},
{
"epoch": 0.59,
"learning_rate": 0.00020593781344032096,
"loss": 2.8533,
"theoretical_loss": 3.4369226074535564,
"tokens_seen": 1955004416
},
{
"epoch": 0.59,
"learning_rate": 0.00020592778335005017,
"loss": 2.741,
"theoretical_loss": 3.436912958410667,
"tokens_seen": 1955069952
},
{
"epoch": 0.59,
"learning_rate": 0.00020591775325977935,
"loss": 2.8116,
"theoretical_loss": 3.436903309781779,
"tokens_seen": 1955135488
},
{
"epoch": 0.59,
"learning_rate": 0.00020590772316950853,
"loss": 2.8952,
"theoretical_loss": 3.436893661566862,
"tokens_seen": 1955201024
},
{
"epoch": 0.59,
"learning_rate": 0.00020589769307923772,
"loss": 2.663,
"theoretical_loss": 3.4368840137658836,
"tokens_seen": 1955266560
},
{
"epoch": 0.59,
"learning_rate": 0.00020588766298896692,
"loss": 2.7145,
"theoretical_loss": 3.436874366378812,
"tokens_seen": 1955332096
},
{
"epoch": 0.59,
"learning_rate": 0.0002058776328986961,
"loss": 2.5681,
"theoretical_loss": 3.4368647194056168,
"tokens_seen": 1955397632
},
{
"epoch": 0.59,
"learning_rate": 0.0002058676028084253,
"loss": 2.712,
"theoretical_loss": 3.436855072846265,
"tokens_seen": 1955463168
},
{
"epoch": 0.59,
"learning_rate": 0.00020585757271815447,
"loss": 2.7563,
"theoretical_loss": 3.4368454267007253,
"tokens_seen": 1955528704
},
{
"epoch": 0.59,
"learning_rate": 0.00020584754262788365,
"loss": 2.434,
"theoretical_loss": 3.4368357809689667,
"tokens_seen": 1955594240
},
{
"epoch": 0.59,
"learning_rate": 0.00020583751253761286,
"loss": 2.5781,
"theoretical_loss": 3.436826135650957,
"tokens_seen": 1955659776
},
{
"epoch": 0.59,
"learning_rate": 0.00020582748244734204,
"loss": 2.6066,
"theoretical_loss": 3.436816490746665,
"tokens_seen": 1955725312
},
{
"epoch": 0.59,
"learning_rate": 0.00020581745235707122,
"loss": 2.4755,
"theoretical_loss": 3.436806846256058,
"tokens_seen": 1955790848
},
{
"epoch": 0.59,
"learning_rate": 0.0002058074222668004,
"loss": 2.789,
"theoretical_loss": 3.436797202179106,
"tokens_seen": 1955856384
},
{
"epoch": 0.59,
"learning_rate": 0.0002057973921765296,
"loss": 2.6429,
"theoretical_loss": 3.4367875585157766,
"tokens_seen": 1955921920
},
{
"epoch": 0.59,
"learning_rate": 0.0002057873620862588,
"loss": 2.8194,
"theoretical_loss": 3.4367779152660383,
"tokens_seen": 1955987456
},
{
"epoch": 0.59,
"learning_rate": 0.00020577733199598797,
"loss": 2.6799,
"theoretical_loss": 3.436768272429859,
"tokens_seen": 1956052992
},
{
"epoch": 0.59,
"learning_rate": 0.00020576730190571715,
"loss": 2.9619,
"theoretical_loss": 3.4367586300072075,
"tokens_seen": 1956118528
},
{
"epoch": 0.59,
"learning_rate": 0.00020575727181544633,
"loss": 2.721,
"theoretical_loss": 3.4367489879980524,
"tokens_seen": 1956184064
},
{
"epoch": 0.59,
"objective/train/docs_used": 1106102,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2858328819274902,
"objective/train/theoretical_loss": 3.436739346402362,
"objective/train/tokens_used": 1976709600,
"theoretical_loss": 3.436739346402362,
"tokens_seen": 1956249600
},
{
"epoch": 0.59,
"learning_rate": 0.00020574724172517554,
"loss": 2.6107,
"theoretical_loss": 3.436739346402362,
"tokens_seen": 1956249600
},
{
"epoch": 0.59,
"learning_rate": 0.00020573721163490472,
"loss": 2.6667,
"theoretical_loss": 3.436729705220105,
"tokens_seen": 1956315136
},
{
"epoch": 0.59,
"learning_rate": 0.0002057271815446339,
"loss": 2.7243,
"theoretical_loss": 3.436720064451249,
"tokens_seen": 1956380672
},
{
"epoch": 0.59,
"learning_rate": 0.00020571715145436308,
"loss": 2.6666,
"theoretical_loss": 3.4367104240957635,
"tokens_seen": 1956446208
},
{
"epoch": 0.59,
"learning_rate": 0.0002057071213640923,
"loss": 2.6779,
"theoretical_loss": 3.436700784153616,
"tokens_seen": 1956511744
},
{
"epoch": 0.59,
"learning_rate": 0.00020569709127382147,
"loss": 2.4491,
"theoretical_loss": 3.4366911446247754,
"tokens_seen": 1956577280
},
{
"epoch": 0.59,
"learning_rate": 0.00020568706118355066,
"loss": 2.7944,
"theoretical_loss": 3.43668150550921,
"tokens_seen": 1956642816
},
{
"epoch": 0.59,
"learning_rate": 0.00020567703109327984,
"loss": 2.4778,
"theoretical_loss": 3.4366718668068885,
"tokens_seen": 1956708352
},
{
"epoch": 0.59,
"learning_rate": 0.00020566700100300902,
"loss": 2.5865,
"theoretical_loss": 3.436662228517779,
"tokens_seen": 1956773888
},
{
"epoch": 0.59,
"learning_rate": 0.00020565697091273823,
"loss": 2.9145,
"theoretical_loss": 3.4366525906418497,
"tokens_seen": 1956839424
},
{
"epoch": 0.59,
"learning_rate": 0.0002056469408224674,
"loss": 2.6913,
"theoretical_loss": 3.43664295317907,
"tokens_seen": 1956904960
},
{
"epoch": 0.59,
"learning_rate": 0.0002056369107321966,
"loss": 2.806,
"theoretical_loss": 3.4366333161294076,
"tokens_seen": 1956970496
},
{
"epoch": 0.59,
"learning_rate": 0.00020562688064192577,
"loss": 2.5481,
"theoretical_loss": 3.436623679492831,
"tokens_seen": 1957036032
},
{
"epoch": 0.59,
"learning_rate": 0.00020561685055165498,
"loss": 2.6016,
"theoretical_loss": 3.436614043269309,
"tokens_seen": 1957101568
},
{
"epoch": 0.59,
"learning_rate": 0.00020560682046138416,
"loss": 2.587,
"theoretical_loss": 3.43660440745881,
"tokens_seen": 1957167104
},
{
"epoch": 0.59,
"learning_rate": 0.00020559679037111334,
"loss": 2.7733,
"theoretical_loss": 3.4365947720613015,
"tokens_seen": 1957232640
},
{
"epoch": 0.59,
"learning_rate": 0.00020558676028084252,
"loss": 2.6291,
"theoretical_loss": 3.4365851370767535,
"tokens_seen": 1957298176
},
{
"epoch": 0.59,
"learning_rate": 0.0002055767301905717,
"loss": 2.3775,
"theoretical_loss": 3.436575502505134,
"tokens_seen": 1957363712
},
{
"epoch": 0.59,
"learning_rate": 0.0002055667001003009,
"loss": 2.6375,
"theoretical_loss": 3.4365658683464106,
"tokens_seen": 1957429248
},
{
"epoch": 0.59,
"learning_rate": 0.0002055566700100301,
"loss": 2.5569,
"theoretical_loss": 3.4365562346005527,
"tokens_seen": 1957494784
},
{
"epoch": 0.59,
"learning_rate": 0.00020554663991975927,
"loss": 2.7044,
"theoretical_loss": 3.436546601267529,
"tokens_seen": 1957560320
},
{
"epoch": 0.59,
"learning_rate": 0.00020553660982948845,
"loss": 2.7237,
"theoretical_loss": 3.4365369683473066,
"tokens_seen": 1957625856
},
{
"epoch": 0.59,
"learning_rate": 0.00020552657973921766,
"loss": 2.5623,
"theoretical_loss": 3.4365273358398554,
"tokens_seen": 1957691392
},
{
"epoch": 0.59,
"learning_rate": 0.00020551654964894684,
"loss": 2.7241,
"theoretical_loss": 3.4365177037451433,
"tokens_seen": 1957756928
},
{
"epoch": 0.59,
"learning_rate": 0.00020550651955867602,
"loss": 2.6327,
"theoretical_loss": 3.4365080720631394,
"tokens_seen": 1957822464
},
{
"epoch": 0.59,
"objective/train/docs_used": 1106840,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.73449444770813,
"objective/train/theoretical_loss": 3.4364984407938115,
"objective/train/tokens_used": 1978348000,
"theoretical_loss": 3.4364984407938115,
"tokens_seen": 1957888000
},
{
"epoch": 0.59,
"learning_rate": 0.0002054964894684052,
"loss": 2.6142,
"theoretical_loss": 3.4364984407938115,
"tokens_seen": 1957888000
},
{
"epoch": 0.59,
"learning_rate": 0.00020548645937813439,
"loss": 2.5591,
"theoretical_loss": 3.436488809937128,
"tokens_seen": 1957953536
},
{
"epoch": 0.59,
"learning_rate": 0.0002054764292878636,
"loss": 2.7333,
"theoretical_loss": 3.4364791794930576,
"tokens_seen": 1958019072
},
{
"epoch": 0.59,
"learning_rate": 0.00020546639919759278,
"loss": 2.593,
"theoretical_loss": 3.4364695494615694,
"tokens_seen": 1958084608
},
{
"epoch": 0.59,
"learning_rate": 0.00020545636910732196,
"loss": 2.7209,
"theoretical_loss": 3.4364599198426307,
"tokens_seen": 1958150144
},
{
"epoch": 0.59,
"learning_rate": 0.00020544633901705114,
"loss": 2.7059,
"theoretical_loss": 3.4364502906362118,
"tokens_seen": 1958215680
},
{
"epoch": 0.59,
"learning_rate": 0.00020543630892678035,
"loss": 2.4575,
"theoretical_loss": 3.4364406618422794,
"tokens_seen": 1958281216
},
{
"epoch": 0.59,
"learning_rate": 0.00020542627883650953,
"loss": 2.4045,
"theoretical_loss": 3.436431033460803,
"tokens_seen": 1958346752
},
{
"epoch": 0.59,
"learning_rate": 0.0002054162487462387,
"loss": 2.8627,
"theoretical_loss": 3.4364214054917515,
"tokens_seen": 1958412288
},
{
"epoch": 0.59,
"learning_rate": 0.0002054062186559679,
"loss": 2.7185,
"theoretical_loss": 3.436411777935092,
"tokens_seen": 1958477824
},
{
"epoch": 0.59,
"learning_rate": 0.0002053961885656971,
"loss": 2.6017,
"theoretical_loss": 3.4364021507907943,
"tokens_seen": 1958543360
},
{
"epoch": 0.59,
"learning_rate": 0.00020538615847542628,
"loss": 2.6752,
"theoretical_loss": 3.4363925240588267,
"tokens_seen": 1958608896
},
{
"epoch": 0.59,
"learning_rate": 0.00020537612838515546,
"loss": 2.6908,
"theoretical_loss": 3.4363828977391577,
"tokens_seen": 1958674432
},
{
"epoch": 0.59,
"learning_rate": 0.00020536609829488464,
"loss": 2.5823,
"theoretical_loss": 3.436373271831755,
"tokens_seen": 1958739968
},
{
"epoch": 0.59,
"learning_rate": 0.00020535606820461382,
"loss": 2.6772,
"theoretical_loss": 3.4363636463365888,
"tokens_seen": 1958805504
},
{
"epoch": 0.59,
"learning_rate": 0.00020534603811434303,
"loss": 2.7917,
"theoretical_loss": 3.436354021253626,
"tokens_seen": 1958871040
},
{
"epoch": 0.59,
"learning_rate": 0.0002053360080240722,
"loss": 2.6313,
"theoretical_loss": 3.436344396582837,
"tokens_seen": 1958936576
},
{
"epoch": 0.59,
"learning_rate": 0.00020532597793380142,
"loss": 2.5085,
"theoretical_loss": 3.436334772324188,
"tokens_seen": 1959002112
},
{
"epoch": 0.59,
"learning_rate": 0.0002053159478435306,
"loss": 2.6477,
"theoretical_loss": 3.43632514847765,
"tokens_seen": 1959067648
},
{
"epoch": 0.59,
"learning_rate": 0.0002053059177532598,
"loss": 2.936,
"theoretical_loss": 3.4363155250431894,
"tokens_seen": 1959133184
},
{
"epoch": 0.59,
"learning_rate": 0.000205295887662989,
"loss": 2.6093,
"theoretical_loss": 3.4363059020207762,
"tokens_seen": 1959198720
},
{
"epoch": 0.59,
"learning_rate": 0.00020528585757271817,
"loss": 2.5901,
"theoretical_loss": 3.436296279410379,
"tokens_seen": 1959264256
},
{
"epoch": 0.59,
"learning_rate": 0.00020527582748244735,
"loss": 2.7789,
"theoretical_loss": 3.436286657211965,
"tokens_seen": 1959329792
},
{
"epoch": 0.59,
"learning_rate": 0.00020526579739217653,
"loss": 2.4933,
"theoretical_loss": 3.4362770354255043,
"tokens_seen": 1959395328
},
{
"epoch": 0.59,
"learning_rate": 0.00020525576730190574,
"loss": 2.7726,
"theoretical_loss": 3.436267414050965,
"tokens_seen": 1959460864
},
{
"epoch": 0.59,
"objective/train/docs_used": 1108280,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5348260402679443,
"objective/train/theoretical_loss": 3.4362577930883154,
"objective/train/tokens_used": 1979986400,
"theoretical_loss": 3.4362577930883154,
"tokens_seen": 1959526400
},
{
"epoch": 0.59,
"learning_rate": 0.00020524573721163492,
"loss": 2.6773,
"theoretical_loss": 3.4362577930883154,
"tokens_seen": 1959526400
},
{
"epoch": 0.59,
"learning_rate": 0.0002052357071213641,
"loss": 2.6622,
"theoretical_loss": 3.436248172537524,
"tokens_seen": 1959591936
},
{
"epoch": 0.59,
"learning_rate": 0.00020522567703109328,
"loss": 2.6266,
"theoretical_loss": 3.43623855239856,
"tokens_seen": 1959657472
},
{
"epoch": 0.59,
"learning_rate": 0.0002052156469408225,
"loss": 2.5899,
"theoretical_loss": 3.4362289326713915,
"tokens_seen": 1959723008
},
{
"epoch": 0.59,
"learning_rate": 0.00020520561685055167,
"loss": 2.8697,
"theoretical_loss": 3.4362193133559877,
"tokens_seen": 1959788544
},
{
"epoch": 0.59,
"learning_rate": 0.00020519558676028086,
"loss": 2.4893,
"theoretical_loss": 3.4362096944523164,
"tokens_seen": 1959854080
},
{
"epoch": 0.59,
"learning_rate": 0.00020518555667001004,
"loss": 2.6487,
"theoretical_loss": 3.436200075960347,
"tokens_seen": 1959919616
},
{
"epoch": 0.59,
"learning_rate": 0.00020517552657973922,
"loss": 2.6539,
"theoretical_loss": 3.4361904578800466,
"tokens_seen": 1959985152
},
{
"epoch": 0.59,
"learning_rate": 0.00020516549648946843,
"loss": 2.4064,
"theoretical_loss": 3.436180840211386,
"tokens_seen": 1960050688
},
{
"epoch": 0.59,
"learning_rate": 0.0002051554663991976,
"loss": 2.3973,
"theoretical_loss": 3.436171222954332,
"tokens_seen": 1960116224
},
{
"epoch": 0.59,
"learning_rate": 0.0002051454363089268,
"loss": 2.5482,
"theoretical_loss": 3.436161606108855,
"tokens_seen": 1960181760
},
{
"epoch": 0.59,
"learning_rate": 0.00020513540621865597,
"loss": 2.6437,
"theoretical_loss": 3.436151989674922,
"tokens_seen": 1960247296
},
{
"epoch": 0.59,
"learning_rate": 0.00020512537612838518,
"loss": 2.6752,
"theoretical_loss": 3.436142373652502,
"tokens_seen": 1960312832
},
{
"epoch": 0.59,
"learning_rate": 0.00020511534603811436,
"loss": 2.7144,
"theoretical_loss": 3.4361327580415644,
"tokens_seen": 1960378368
},
{
"epoch": 0.59,
"learning_rate": 0.00020510531594784354,
"loss": 2.4843,
"theoretical_loss": 3.436123142842077,
"tokens_seen": 1960443904
},
{
"epoch": 0.59,
"learning_rate": 0.00020509528585757272,
"loss": 2.4775,
"theoretical_loss": 3.4361135280540087,
"tokens_seen": 1960509440
},
{
"epoch": 0.59,
"learning_rate": 0.0002050852557673019,
"loss": 2.7041,
"theoretical_loss": 3.436103913677328,
"tokens_seen": 1960574976
},
{
"epoch": 0.59,
"learning_rate": 0.0002050752256770311,
"loss": 2.8246,
"theoretical_loss": 3.4360942997120043,
"tokens_seen": 1960640512
},
{
"epoch": 0.59,
"learning_rate": 0.0002050651955867603,
"loss": 2.3877,
"theoretical_loss": 3.436084686158005,
"tokens_seen": 1960706048
},
{
"epoch": 0.59,
"learning_rate": 0.00020505516549648947,
"loss": 2.5964,
"theoretical_loss": 3.4360750730153002,
"tokens_seen": 1960771584
},
{
"epoch": 0.59,
"learning_rate": 0.00020504513540621865,
"loss": 2.5999,
"theoretical_loss": 3.4360654602838574,
"tokens_seen": 1960837120
},
{
"epoch": 0.59,
"learning_rate": 0.00020503510531594786,
"loss": 2.7872,
"theoretical_loss": 3.4360558479636456,
"tokens_seen": 1960902656
},
{
"epoch": 0.59,
"learning_rate": 0.00020502507522567704,
"loss": 2.3559,
"theoretical_loss": 3.4360462360546333,
"tokens_seen": 1960968192
},
{
"epoch": 0.59,
"learning_rate": 0.00020501504513540622,
"loss": 2.4358,
"theoretical_loss": 3.43603662455679,
"tokens_seen": 1961033728
},
{
"epoch": 0.59,
"learning_rate": 0.0002050050150451354,
"loss": 2.4993,
"theoretical_loss": 3.436027013470083,
"tokens_seen": 1961099264
},
{
"epoch": 0.59,
"objective/train/docs_used": 1108943,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5256869792938232,
"objective/train/theoretical_loss": 3.436017402794482,
"objective/train/tokens_used": 1981624800,
"theoretical_loss": 3.436017402794482,
"tokens_seen": 1961164800
},
{
"epoch": 0.59,
"learning_rate": 0.00020499498495486459,
"loss": 2.5021,
"theoretical_loss": 3.436017402794482,
"tokens_seen": 1961164800
},
{
"epoch": 0.59,
"learning_rate": 0.0002049849548645938,
"loss": 2.8827,
"theoretical_loss": 3.4360077925299555,
"tokens_seen": 1961230336
},
{
"epoch": 0.59,
"learning_rate": 0.00020497492477432298,
"loss": 2.8604,
"theoretical_loss": 3.435998182676472,
"tokens_seen": 1961295872
},
{
"epoch": 0.59,
"learning_rate": 0.00020496489468405216,
"loss": 2.4665,
"theoretical_loss": 3.4359885732340008,
"tokens_seen": 1961361408
},
{
"epoch": 0.59,
"learning_rate": 0.00020495486459378134,
"loss": 2.7677,
"theoretical_loss": 3.4359789642025094,
"tokens_seen": 1961426944
},
{
"epoch": 0.59,
"learning_rate": 0.00020494483450351055,
"loss": 2.8036,
"theoretical_loss": 3.4359693555819675,
"tokens_seen": 1961492480
},
{
"epoch": 0.59,
"learning_rate": 0.00020493480441323973,
"loss": 2.563,
"theoretical_loss": 3.435959747372343,
"tokens_seen": 1961558016
},
{
"epoch": 0.59,
"learning_rate": 0.0002049247743229689,
"loss": 2.4747,
"theoretical_loss": 3.4359501395736056,
"tokens_seen": 1961623552
},
{
"epoch": 0.59,
"learning_rate": 0.0002049147442326981,
"loss": 2.7409,
"theoretical_loss": 3.4359405321857235,
"tokens_seen": 1961689088
},
{
"epoch": 0.59,
"learning_rate": 0.0002049047141424273,
"loss": 2.5966,
"theoretical_loss": 3.4359309252086647,
"tokens_seen": 1961754624
},
{
"epoch": 0.59,
"learning_rate": 0.00020489468405215648,
"loss": 2.4452,
"theoretical_loss": 3.435921318642399,
"tokens_seen": 1961820160
},
{
"epoch": 0.59,
"learning_rate": 0.00020488465396188566,
"loss": 2.4683,
"theoretical_loss": 3.4359117124868948,
"tokens_seen": 1961885696
},
{
"epoch": 0.59,
"learning_rate": 0.00020487462387161484,
"loss": 2.722,
"theoretical_loss": 3.4359021067421205,
"tokens_seen": 1961951232
},
{
"epoch": 0.59,
"learning_rate": 0.00020486459378134402,
"loss": 2.6119,
"theoretical_loss": 3.435892501408045,
"tokens_seen": 1962016768
},
{
"epoch": 0.59,
"learning_rate": 0.00020485456369107323,
"loss": 2.5362,
"theoretical_loss": 3.435882896484637,
"tokens_seen": 1962082304
},
{
"epoch": 0.59,
"learning_rate": 0.0002048445336008024,
"loss": 2.5996,
"theoretical_loss": 3.4358732919718653,
"tokens_seen": 1962147840
},
{
"epoch": 0.59,
"learning_rate": 0.0002048345035105316,
"loss": 2.7347,
"theoretical_loss": 3.435863687869699,
"tokens_seen": 1962213376
},
{
"epoch": 0.59,
"learning_rate": 0.00020482447342026077,
"loss": 2.7822,
"theoretical_loss": 3.4358540841781062,
"tokens_seen": 1962278912
},
{
"epoch": 0.59,
"learning_rate": 0.00020481444332998998,
"loss": 2.6281,
"theoretical_loss": 3.435844480897056,
"tokens_seen": 1962344448
},
{
"epoch": 0.59,
"learning_rate": 0.00020480441323971916,
"loss": 2.6539,
"theoretical_loss": 3.435834878026516,
"tokens_seen": 1962409984
},
{
"epoch": 0.59,
"learning_rate": 0.00020479438314944834,
"loss": 2.5971,
"theoretical_loss": 3.435825275566457,
"tokens_seen": 1962475520
},
{
"epoch": 0.59,
"learning_rate": 0.00020478435305917753,
"loss": 2.7252,
"theoretical_loss": 3.4358156735168466,
"tokens_seen": 1962541056
},
{
"epoch": 0.59,
"learning_rate": 0.0002047743229689067,
"loss": 2.6852,
"theoretical_loss": 3.4358060718776535,
"tokens_seen": 1962606592
},
{
"epoch": 0.59,
"learning_rate": 0.00020476429287863591,
"loss": 2.7721,
"theoretical_loss": 3.435796470648846,
"tokens_seen": 1962672128
},
{
"epoch": 0.59,
"learning_rate": 0.0002047542627883651,
"loss": 2.58,
"theoretical_loss": 3.4357868698303946,
"tokens_seen": 1962737664
},
{
"epoch": 0.59,
"objective/train/docs_used": 1109653,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7564241886138916,
"objective/train/theoretical_loss": 3.435777269422266,
"objective/train/tokens_used": 1983263200,
"theoretical_loss": 3.435777269422266,
"tokens_seen": 1962803200
},
{
"epoch": 0.59,
"learning_rate": 0.00020474423269809428,
"loss": 2.765,
"theoretical_loss": 3.435777269422266,
"tokens_seen": 1962803200
},
{
"epoch": 0.59,
"learning_rate": 0.00020473420260782346,
"loss": 2.6581,
"theoretical_loss": 3.4357676694244303,
"tokens_seen": 1962868736
},
{
"epoch": 0.59,
"learning_rate": 0.00020472417251755267,
"loss": 2.7998,
"theoretical_loss": 3.4357580698368557,
"tokens_seen": 1962934272
},
{
"epoch": 0.59,
"learning_rate": 0.00020471414242728185,
"loss": 2.3562,
"theoretical_loss": 3.435748470659511,
"tokens_seen": 1962999808
},
{
"epoch": 0.59,
"learning_rate": 0.00020470411233701103,
"loss": 2.7371,
"theoretical_loss": 3.4357388718923656,
"tokens_seen": 1963065344
},
{
"epoch": 0.59,
"learning_rate": 0.0002046940822467402,
"loss": 2.4786,
"theoretical_loss": 3.4357292735353875,
"tokens_seen": 1963130880
},
{
"epoch": 0.59,
"learning_rate": 0.0002046840521564694,
"loss": 2.6022,
"theoretical_loss": 3.4357196755885457,
"tokens_seen": 1963196416
},
{
"epoch": 0.59,
"learning_rate": 0.0002046740220661986,
"loss": 2.574,
"theoretical_loss": 3.435710078051809,
"tokens_seen": 1963261952
},
{
"epoch": 0.59,
"learning_rate": 0.00020466399197592778,
"loss": 2.67,
"theoretical_loss": 3.435700480925146,
"tokens_seen": 1963327488
},
{
"epoch": 0.59,
"learning_rate": 0.00020465396188565696,
"loss": 2.6329,
"theoretical_loss": 3.435690884208526,
"tokens_seen": 1963393024
},
{
"epoch": 0.59,
"learning_rate": 0.00020464393179538614,
"loss": 2.5739,
"theoretical_loss": 3.4356812879019176,
"tokens_seen": 1963458560
},
{
"epoch": 0.6,
"learning_rate": 0.00020463390170511535,
"loss": 2.6734,
"theoretical_loss": 3.4356716920052897,
"tokens_seen": 1963524096
},
{
"epoch": 0.6,
"learning_rate": 0.00020462387161484453,
"loss": 2.6824,
"theoretical_loss": 3.4356620965186107,
"tokens_seen": 1963589632
},
{
"epoch": 0.6,
"learning_rate": 0.0002046138415245737,
"loss": 2.4219,
"theoretical_loss": 3.435652501441849,
"tokens_seen": 1963655168
},
{
"epoch": 0.6,
"learning_rate": 0.0002046038114343029,
"loss": 2.3703,
"theoretical_loss": 3.4356429067749747,
"tokens_seen": 1963720704
},
{
"epoch": 0.6,
"learning_rate": 0.00020459378134403207,
"loss": 2.5283,
"theoretical_loss": 3.4356333125179557,
"tokens_seen": 1963786240
},
{
"epoch": 0.6,
"learning_rate": 0.00020458375125376128,
"loss": 2.614,
"theoretical_loss": 3.435623718670761,
"tokens_seen": 1963851776
},
{
"epoch": 0.6,
"learning_rate": 0.0002045737211634905,
"loss": 2.5237,
"theoretical_loss": 3.43561412523336,
"tokens_seen": 1963917312
},
{
"epoch": 0.6,
"learning_rate": 0.00020456369107321967,
"loss": 2.5806,
"theoretical_loss": 3.4356045322057205,
"tokens_seen": 1963982848
},
{
"epoch": 0.6,
"learning_rate": 0.00020455366098294885,
"loss": 2.4985,
"theoretical_loss": 3.435594939587812,
"tokens_seen": 1964048384
},
{
"epoch": 0.6,
"learning_rate": 0.00020454363089267806,
"loss": 2.624,
"theoretical_loss": 3.435585347379603,
"tokens_seen": 1964113920
},
{
"epoch": 0.6,
"learning_rate": 0.00020453360080240724,
"loss": 2.5391,
"theoretical_loss": 3.4355757555810627,
"tokens_seen": 1964179456
},
{
"epoch": 0.6,
"learning_rate": 0.00020452357071213642,
"loss": 2.5512,
"theoretical_loss": 3.435566164192159,
"tokens_seen": 1964244992
},
{
"epoch": 0.6,
"learning_rate": 0.0002045135406218656,
"loss": 2.75,
"theoretical_loss": 3.435556573212862,
"tokens_seen": 1964310528
},
{
"epoch": 0.6,
"learning_rate": 0.00020450351053159479,
"loss": 2.68,
"theoretical_loss": 3.4355469826431406,
"tokens_seen": 1964376064
},
{
"epoch": 0.6,
"objective/train/docs_used": 1110737,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.666476011276245,
"objective/train/theoretical_loss": 3.435537392482962,
"objective/train/tokens_used": 1984901600,
"theoretical_loss": 3.435537392482962,
"tokens_seen": 1964441600
},
{
"epoch": 0.6,
"learning_rate": 0.000204493480441324,
"loss": 2.5485,
"theoretical_loss": 3.435537392482962,
"tokens_seen": 1964441600
},
{
"epoch": 0.6,
"learning_rate": 0.00020448345035105318,
"loss": 2.6441,
"theoretical_loss": 3.4355278027322966,
"tokens_seen": 1964507136
},
{
"epoch": 0.6,
"learning_rate": 0.00020447342026078236,
"loss": 2.4684,
"theoretical_loss": 3.435518213391113,
"tokens_seen": 1964572672
},
{
"epoch": 0.6,
"learning_rate": 0.00020446339017051154,
"loss": 2.3968,
"theoretical_loss": 3.4355086244593793,
"tokens_seen": 1964638208
},
{
"epoch": 0.6,
"learning_rate": 0.00020445336008024075,
"loss": 2.6658,
"theoretical_loss": 3.435499035937065,
"tokens_seen": 1964703744
},
{
"epoch": 0.6,
"learning_rate": 0.00020444332998996993,
"loss": 2.7028,
"theoretical_loss": 3.435489447824139,
"tokens_seen": 1964769280
},
{
"epoch": 0.6,
"learning_rate": 0.0002044332998996991,
"loss": 2.4018,
"theoretical_loss": 3.43547986012057,
"tokens_seen": 1964834816
},
{
"epoch": 0.6,
"learning_rate": 0.0002044232698094283,
"loss": 2.3866,
"theoretical_loss": 3.4354702728263264,
"tokens_seen": 1964900352
},
{
"epoch": 0.6,
"learning_rate": 0.0002044132397191575,
"loss": 2.4924,
"theoretical_loss": 3.435460685941378,
"tokens_seen": 1964965888
},
{
"epoch": 0.6,
"learning_rate": 0.00020440320962888668,
"loss": 2.651,
"theoretical_loss": 3.435451099465693,
"tokens_seen": 1965031424
},
{
"epoch": 0.6,
"learning_rate": 0.00020439317953861586,
"loss": 2.5813,
"theoretical_loss": 3.4354415133992404,
"tokens_seen": 1965096960
},
{
"epoch": 0.6,
"learning_rate": 0.00020438314944834504,
"loss": 2.5772,
"theoretical_loss": 3.4354319277419894,
"tokens_seen": 1965162496
},
{
"epoch": 0.6,
"learning_rate": 0.00020437311935807422,
"loss": 2.4417,
"theoretical_loss": 3.4354223424939088,
"tokens_seen": 1965228032
},
{
"epoch": 0.6,
"learning_rate": 0.00020436308926780343,
"loss": 2.586,
"theoretical_loss": 3.435412757654967,
"tokens_seen": 1965293568
},
{
"epoch": 0.6,
"learning_rate": 0.0002043530591775326,
"loss": 2.3772,
"theoretical_loss": 3.4354031732251333,
"tokens_seen": 1965359104
},
{
"epoch": 0.6,
"learning_rate": 0.0002043430290872618,
"loss": 2.6859,
"theoretical_loss": 3.4353935892043768,
"tokens_seen": 1965424640
},
{
"epoch": 0.6,
"learning_rate": 0.00020433299899699097,
"loss": 2.5116,
"theoretical_loss": 3.435384005592666,
"tokens_seen": 1965490176
},
{
"epoch": 0.6,
"learning_rate": 0.00020432296890672018,
"loss": 2.5642,
"theoretical_loss": 3.4353744223899696,
"tokens_seen": 1965555712
},
{
"epoch": 0.6,
"learning_rate": 0.00020431293881644936,
"loss": 2.6969,
"theoretical_loss": 3.4353648395962573,
"tokens_seen": 1965621248
},
{
"epoch": 0.6,
"learning_rate": 0.00020430290872617854,
"loss": 2.886,
"theoretical_loss": 3.435355257211498,
"tokens_seen": 1965686784
},
{
"epoch": 0.6,
"learning_rate": 0.00020429287863590773,
"loss": 2.4566,
"theoretical_loss": 3.4353456752356593,
"tokens_seen": 1965752320
},
{
"epoch": 0.6,
"learning_rate": 0.0002042828485456369,
"loss": 2.703,
"theoretical_loss": 3.4353360936687114,
"tokens_seen": 1965817856
},
{
"epoch": 0.6,
"learning_rate": 0.00020427281845536611,
"loss": 2.7479,
"theoretical_loss": 3.4353265125106223,
"tokens_seen": 1965883392
},
{
"epoch": 0.6,
"learning_rate": 0.0002042627883650953,
"loss": 2.6055,
"theoretical_loss": 3.4353169317613617,
"tokens_seen": 1965948928
},
{
"epoch": 0.6,
"learning_rate": 0.00020425275827482448,
"loss": 2.4851,
"theoretical_loss": 3.4353073514208985,
"tokens_seen": 1966014464
},
{
"debugging/Self-BLEU-5": 0.33715398191719237,
"debugging/distinct-1-grams": 0.7635596180800125,
"debugging/distinct-2-grams": 0.9480781313546194,
"debugging/entropy-1-grams": 5.556007883515177,
"debugging/entropy-2-grams": 6.288683224115498,
"debugging/length": 474.0,
"debugging/num_segments": 8,
"epoch": 0.6,
"objective/train/docs_used": 1111454,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4314239025115967,
"objective/train/theoretical_loss": 3.4352977714892017,
"objective/train/tokens_used": 1986540000,
"theoretical_loss": 3.4352977714892017,
"tokens_seen": 1966080000
},
{
"epoch": 0.6,
"learning_rate": 0.00020424272818455366,
"loss": 2.4894,
"theoretical_loss": 3.4352977714892017,
"tokens_seen": 1966080000
},
{
"epoch": 0.6,
"learning_rate": 0.00020423269809428287,
"loss": 2.5708,
"theoretical_loss": 3.4352881919662392,
"tokens_seen": 1966145536
},
{
"epoch": 0.6,
"learning_rate": 0.00020422266800401205,
"loss": 2.5354,
"theoretical_loss": 3.435278612851981,
"tokens_seen": 1966211072
},
{
"epoch": 0.6,
"learning_rate": 0.00020421263791374123,
"loss": 2.8161,
"theoretical_loss": 3.435269034146396,
"tokens_seen": 1966276608
},
{
"epoch": 0.6,
"learning_rate": 0.0002042026078234704,
"loss": 2.6711,
"theoretical_loss": 3.4352594558494527,
"tokens_seen": 1966342144
},
{
"epoch": 0.6,
"learning_rate": 0.0002041925777331996,
"loss": 2.5911,
"theoretical_loss": 3.4352498779611196,
"tokens_seen": 1966407680
},
{
"epoch": 0.6,
"learning_rate": 0.0002041825476429288,
"loss": 2.5451,
"theoretical_loss": 3.435240300481367,
"tokens_seen": 1966473216
},
{
"epoch": 0.6,
"learning_rate": 0.00020417251755265798,
"loss": 2.4736,
"theoretical_loss": 3.4352307234101627,
"tokens_seen": 1966538752
},
{
"epoch": 0.6,
"learning_rate": 0.00020416248746238716,
"loss": 2.4779,
"theoretical_loss": 3.4352211467474767,
"tokens_seen": 1966604288
},
{
"epoch": 0.6,
"learning_rate": 0.00020415245737211634,
"loss": 2.4823,
"theoretical_loss": 3.4352115704932764,
"tokens_seen": 1966669824
},
{
"epoch": 0.6,
"learning_rate": 0.00020414242728184555,
"loss": 2.739,
"theoretical_loss": 3.4352019946475325,
"tokens_seen": 1966735360
},
{
"epoch": 0.6,
"learning_rate": 0.00020413239719157473,
"loss": 2.6224,
"theoretical_loss": 3.435192419210213,
"tokens_seen": 1966800896
},
{
"epoch": 0.6,
"learning_rate": 0.0002041223671013039,
"loss": 2.6394,
"theoretical_loss": 3.435182844181287,
"tokens_seen": 1966866432
},
{
"epoch": 0.6,
"learning_rate": 0.0002041123370110331,
"loss": 2.5879,
"theoretical_loss": 3.435173269560723,
"tokens_seen": 1966931968
},
{
"epoch": 0.6,
"learning_rate": 0.00020410230692076228,
"loss": 2.4744,
"theoretical_loss": 3.435163695348491,
"tokens_seen": 1966997504
},
{
"epoch": 0.6,
"learning_rate": 0.00020409227683049148,
"loss": 2.6258,
"theoretical_loss": 3.435154121544559,
"tokens_seen": 1967063040
},
{
"epoch": 0.6,
"learning_rate": 0.00020408224674022066,
"loss": 2.9074,
"theoretical_loss": 3.4351445481488967,
"tokens_seen": 1967128576
},
{
"epoch": 0.6,
"learning_rate": 0.00020407221664994985,
"loss": 2.6599,
"theoretical_loss": 3.435134975161473,
"tokens_seen": 1967194112
},
{
"epoch": 0.6,
"learning_rate": 0.00020406218655967903,
"loss": 2.3642,
"theoretical_loss": 3.4351254025822566,
"tokens_seen": 1967259648
},
{
"epoch": 0.6,
"learning_rate": 0.00020405215646940823,
"loss": 2.8896,
"theoretical_loss": 3.435115830411217,
"tokens_seen": 1967325184
},
{
"epoch": 0.6,
"learning_rate": 0.00020404212637913742,
"loss": 2.4752,
"theoretical_loss": 3.4351062586483225,
"tokens_seen": 1967390720
},
{
"epoch": 0.6,
"learning_rate": 0.0002040320962888666,
"loss": 2.5864,
"theoretical_loss": 3.4350966872935422,
"tokens_seen": 1967456256
},
{
"epoch": 0.6,
"learning_rate": 0.00020402206619859578,
"loss": 2.5142,
"theoretical_loss": 3.4350871163468457,
"tokens_seen": 1967521792
},
{
"epoch": 0.6,
"learning_rate": 0.00020401203610832496,
"loss": 2.5689,
"theoretical_loss": 3.435077545808202,
"tokens_seen": 1967587328
},
{
"epoch": 0.6,
"learning_rate": 0.00020400200601805417,
"loss": 2.3263,
"theoretical_loss": 3.435067975677579,
"tokens_seen": 1967652864
},
{
"epoch": 0.6,
"objective/train/docs_used": 1112624,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4626073837280273,
"objective/train/theoretical_loss": 3.4350584059549467,
"objective/train/tokens_used": 1988178400,
"theoretical_loss": 3.4350584059549467,
"tokens_seen": 1967718400
},
{
"epoch": 0.6,
"learning_rate": 0.00020399197592778335,
"loss": 2.5083,
"theoretical_loss": 3.4350584059549467,
"tokens_seen": 1967718400
},
{
"epoch": 0.6,
"learning_rate": 0.00020398194583751253,
"loss": 2.6046,
"theoretical_loss": 3.435048836640274,
"tokens_seen": 1967783936
},
{
"epoch": 0.6,
"learning_rate": 0.0002039719157472417,
"loss": 2.5089,
"theoretical_loss": 3.4350392677335297,
"tokens_seen": 1967849472
},
{
"epoch": 0.6,
"learning_rate": 0.00020396188565697092,
"loss": 2.5279,
"theoretical_loss": 3.435029699234683,
"tokens_seen": 1967915008
},
{
"epoch": 0.6,
"learning_rate": 0.0002039518555667001,
"loss": 2.5485,
"theoretical_loss": 3.435020131143703,
"tokens_seen": 1967980544
},
{
"epoch": 0.6,
"learning_rate": 0.00020394182547642928,
"loss": 2.7529,
"theoretical_loss": 3.4350105634605583,
"tokens_seen": 1968046080
},
{
"epoch": 0.6,
"learning_rate": 0.00020393179538615846,
"loss": 2.6326,
"theoretical_loss": 3.435000996185219,
"tokens_seen": 1968111616
},
{
"epoch": 0.6,
"learning_rate": 0.00020392176529588764,
"loss": 2.6291,
"theoretical_loss": 3.4349914293176527,
"tokens_seen": 1968177152
},
{
"epoch": 0.6,
"learning_rate": 0.00020391173520561685,
"loss": 2.6797,
"theoretical_loss": 3.4349818628578292,
"tokens_seen": 1968242688
},
{
"epoch": 0.6,
"learning_rate": 0.00020390170511534603,
"loss": 2.5083,
"theoretical_loss": 3.4349722968057175,
"tokens_seen": 1968308224
},
{
"epoch": 0.6,
"learning_rate": 0.00020389167502507521,
"loss": 2.5067,
"theoretical_loss": 3.4349627311612867,
"tokens_seen": 1968373760
},
{
"epoch": 0.6,
"learning_rate": 0.0002038816449348044,
"loss": 2.5141,
"theoretical_loss": 3.434953165924506,
"tokens_seen": 1968439296
},
{
"epoch": 0.6,
"learning_rate": 0.0002038716148445336,
"loss": 2.412,
"theoretical_loss": 3.434943601095344,
"tokens_seen": 1968504832
},
{
"epoch": 0.6,
"learning_rate": 0.00020386158475426278,
"loss": 2.6821,
"theoretical_loss": 3.43493403667377,
"tokens_seen": 1968570368
},
{
"epoch": 0.6,
"learning_rate": 0.00020385155466399197,
"loss": 2.4578,
"theoretical_loss": 3.4349244726597528,
"tokens_seen": 1968635904
},
{
"epoch": 0.6,
"learning_rate": 0.00020384152457372115,
"loss": 2.609,
"theoretical_loss": 3.434914909053262,
"tokens_seen": 1968701440
},
{
"epoch": 0.6,
"learning_rate": 0.00020383149448345036,
"loss": 2.8061,
"theoretical_loss": 3.434905345854266,
"tokens_seen": 1968766976
},
{
"epoch": 0.6,
"learning_rate": 0.00020382146439317956,
"loss": 2.6807,
"theoretical_loss": 3.4348957830627347,
"tokens_seen": 1968832512
},
{
"epoch": 0.6,
"learning_rate": 0.00020381143430290874,
"loss": 2.6506,
"theoretical_loss": 3.434886220678637,
"tokens_seen": 1968898048
},
{
"epoch": 0.6,
"learning_rate": 0.00020380140421263793,
"loss": 2.8273,
"theoretical_loss": 3.434876658701941,
"tokens_seen": 1968963584
},
{
"epoch": 0.6,
"learning_rate": 0.0002037913741223671,
"loss": 2.4247,
"theoretical_loss": 3.434867097132617,
"tokens_seen": 1969029120
},
{
"epoch": 0.6,
"learning_rate": 0.00020378134403209631,
"loss": 2.3773,
"theoretical_loss": 3.4348575359706333,
"tokens_seen": 1969094656
},
{
"epoch": 0.6,
"learning_rate": 0.0002037713139418255,
"loss": 2.4406,
"theoretical_loss": 3.43484797521596,
"tokens_seen": 1969160192
},
{
"epoch": 0.6,
"learning_rate": 0.00020376128385155468,
"loss": 2.6734,
"theoretical_loss": 3.4348384148685644,
"tokens_seen": 1969225728
},
{
"epoch": 0.6,
"learning_rate": 0.00020375125376128386,
"loss": 2.4204,
"theoretical_loss": 3.4348288549284174,
"tokens_seen": 1969291264
},
{
"epoch": 0.6,
"objective/train/docs_used": 1113292,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.859400987625122,
"objective/train/theoretical_loss": 3.434819295395487,
"objective/train/tokens_used": 1989816800,
"theoretical_loss": 3.434819295395487,
"tokens_seen": 1969356800
},
{
"epoch": 0.6,
"learning_rate": 0.00020374122367101307,
"loss": 2.7987,
"theoretical_loss": 3.434819295395487,
"tokens_seen": 1969356800
},
{
"epoch": 0.6,
"learning_rate": 0.00020373119358074225,
"loss": 2.7811,
"theoretical_loss": 3.434809736269743,
"tokens_seen": 1969422336
},
{
"epoch": 0.6,
"learning_rate": 0.00020372116349047143,
"loss": 2.4863,
"theoretical_loss": 3.4348001775511543,
"tokens_seen": 1969487872
},
{
"epoch": 0.6,
"learning_rate": 0.0002037111334002006,
"loss": 2.6329,
"theoretical_loss": 3.4347906192396893,
"tokens_seen": 1969553408
},
{
"epoch": 0.6,
"learning_rate": 0.0002037011033099298,
"loss": 2.5761,
"theoretical_loss": 3.4347810613353182,
"tokens_seen": 1969618944
},
{
"epoch": 0.6,
"learning_rate": 0.000203691073219659,
"loss": 2.5867,
"theoretical_loss": 3.434771503838009,
"tokens_seen": 1969684480
},
{
"epoch": 0.6,
"learning_rate": 0.00020368104312938818,
"loss": 2.5481,
"theoretical_loss": 3.434761946747732,
"tokens_seen": 1969750016
},
{
"epoch": 0.6,
"learning_rate": 0.00020367101303911736,
"loss": 2.7526,
"theoretical_loss": 3.434752390064456,
"tokens_seen": 1969815552
},
{
"epoch": 0.6,
"learning_rate": 0.00020366098294884654,
"loss": 2.475,
"theoretical_loss": 3.4347428337881496,
"tokens_seen": 1969881088
},
{
"epoch": 0.6,
"learning_rate": 0.00020365095285857575,
"loss": 2.6838,
"theoretical_loss": 3.434733277918782,
"tokens_seen": 1969946624
},
{
"epoch": 0.6,
"learning_rate": 0.00020364092276830493,
"loss": 2.6645,
"theoretical_loss": 3.4347237224563223,
"tokens_seen": 1970012160
},
{
"epoch": 0.6,
"learning_rate": 0.0002036308926780341,
"loss": 2.5252,
"theoretical_loss": 3.4347141674007404,
"tokens_seen": 1970077696
},
{
"epoch": 0.6,
"learning_rate": 0.0002036208625877633,
"loss": 2.6233,
"theoretical_loss": 3.4347046127520047,
"tokens_seen": 1970143232
},
{
"epoch": 0.6,
"learning_rate": 0.00020361083249749248,
"loss": 2.7747,
"theoretical_loss": 3.4346950585100853,
"tokens_seen": 1970208768
},
{
"epoch": 0.6,
"learning_rate": 0.00020360080240722168,
"loss": 2.7474,
"theoretical_loss": 3.43468550467495,
"tokens_seen": 1970274304
},
{
"epoch": 0.6,
"learning_rate": 0.00020359077231695086,
"loss": 2.4054,
"theoretical_loss": 3.434675951246568,
"tokens_seen": 1970339840
},
{
"epoch": 0.6,
"learning_rate": 0.00020358074222668005,
"loss": 2.6872,
"theoretical_loss": 3.43466639822491,
"tokens_seen": 1970405376
},
{
"epoch": 0.6,
"learning_rate": 0.00020357071213640923,
"loss": 2.4969,
"theoretical_loss": 3.434656845609944,
"tokens_seen": 1970470912
},
{
"epoch": 0.6,
"learning_rate": 0.00020356068204613844,
"loss": 2.8147,
"theoretical_loss": 3.434647293401639,
"tokens_seen": 1970536448
},
{
"epoch": 0.6,
"learning_rate": 0.00020355065195586762,
"loss": 2.4679,
"theoretical_loss": 3.434637741599965,
"tokens_seen": 1970601984
},
{
"epoch": 0.6,
"learning_rate": 0.0002035406218655968,
"loss": 2.5761,
"theoretical_loss": 3.4346281902048905,
"tokens_seen": 1970667520
},
{
"epoch": 0.6,
"learning_rate": 0.00020353059177532598,
"loss": 2.5349,
"theoretical_loss": 3.4346186392163847,
"tokens_seen": 1970733056
},
{
"epoch": 0.6,
"learning_rate": 0.00020352056168505516,
"loss": 2.7797,
"theoretical_loss": 3.434609088634417,
"tokens_seen": 1970798592
},
{
"epoch": 0.6,
"learning_rate": 0.00020351053159478437,
"loss": 2.4395,
"theoretical_loss": 3.4345995384589565,
"tokens_seen": 1970864128
},
{
"epoch": 0.6,
"learning_rate": 0.00020350050150451355,
"loss": 2.6219,
"theoretical_loss": 3.4345899886899725,
"tokens_seen": 1970929664
},
{
"epoch": 0.6,
"objective/train/docs_used": 1114737,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6968495845794678,
"objective/train/theoretical_loss": 3.434580439327434,
"objective/train/tokens_used": 1991455200,
"theoretical_loss": 3.434580439327434,
"tokens_seen": 1970995200
},
{
"epoch": 0.6,
"learning_rate": 0.00020349047141424273,
"loss": 2.4789,
"theoretical_loss": 3.434580439327434,
"tokens_seen": 1970995200
},
{
"epoch": 0.6,
"learning_rate": 0.0002034804413239719,
"loss": 2.74,
"theoretical_loss": 3.4345708903713104,
"tokens_seen": 1971060736
},
{
"epoch": 0.6,
"learning_rate": 0.00020347041123370112,
"loss": 2.7724,
"theoretical_loss": 3.4345613418215706,
"tokens_seen": 1971126272
},
{
"epoch": 0.6,
"learning_rate": 0.0002034603811434303,
"loss": 2.6291,
"theoretical_loss": 3.4345517936781835,
"tokens_seen": 1971191808
},
{
"epoch": 0.6,
"learning_rate": 0.00020345035105315948,
"loss": 2.5698,
"theoretical_loss": 3.4345422459411195,
"tokens_seen": 1971257344
},
{
"epoch": 0.6,
"learning_rate": 0.00020344032096288866,
"loss": 2.5111,
"theoretical_loss": 3.434532698610347,
"tokens_seen": 1971322880
},
{
"epoch": 0.6,
"learning_rate": 0.00020343029087261784,
"loss": 2.7635,
"theoretical_loss": 3.434523151685835,
"tokens_seen": 1971388416
},
{
"epoch": 0.6,
"learning_rate": 0.00020342026078234705,
"loss": 2.4649,
"theoretical_loss": 3.4345136051675524,
"tokens_seen": 1971453952
},
{
"epoch": 0.6,
"learning_rate": 0.00020341023069207623,
"loss": 2.2675,
"theoretical_loss": 3.43450405905547,
"tokens_seen": 1971519488
},
{
"epoch": 0.6,
"learning_rate": 0.00020340020060180541,
"loss": 2.7184,
"theoretical_loss": 3.4344945133495552,
"tokens_seen": 1971585024
},
{
"epoch": 0.6,
"learning_rate": 0.0002033901705115346,
"loss": 2.527,
"theoretical_loss": 3.4344849680497784,
"tokens_seen": 1971650560
},
{
"epoch": 0.6,
"learning_rate": 0.0002033801404212638,
"loss": 2.4048,
"theoretical_loss": 3.4344754231561083,
"tokens_seen": 1971716096
},
{
"epoch": 0.6,
"learning_rate": 0.00020337011033099298,
"loss": 2.5795,
"theoretical_loss": 3.4344658786685143,
"tokens_seen": 1971781632
},
{
"epoch": 0.6,
"learning_rate": 0.00020336008024072217,
"loss": 2.5067,
"theoretical_loss": 3.4344563345869656,
"tokens_seen": 1971847168
},
{
"epoch": 0.6,
"learning_rate": 0.00020335005015045135,
"loss": 2.4224,
"theoretical_loss": 3.4344467909114313,
"tokens_seen": 1971912704
},
{
"epoch": 0.6,
"learning_rate": 0.00020334002006018056,
"loss": 2.7343,
"theoretical_loss": 3.4344372476418807,
"tokens_seen": 1971978240
},
{
"epoch": 0.6,
"learning_rate": 0.00020332998996990974,
"loss": 2.6542,
"theoretical_loss": 3.434427704778283,
"tokens_seen": 1972043776
},
{
"epoch": 0.6,
"learning_rate": 0.00020331995987963892,
"loss": 2.5726,
"theoretical_loss": 3.434418162320608,
"tokens_seen": 1972109312
},
{
"epoch": 0.6,
"learning_rate": 0.0002033099297893681,
"loss": 2.5213,
"theoretical_loss": 3.434408620268824,
"tokens_seen": 1972174848
},
{
"epoch": 0.6,
"learning_rate": 0.00020329989969909728,
"loss": 2.5768,
"theoretical_loss": 3.434399078622901,
"tokens_seen": 1972240384
},
{
"epoch": 0.6,
"learning_rate": 0.0002032898696088265,
"loss": 2.7277,
"theoretical_loss": 3.434389537382808,
"tokens_seen": 1972305920
},
{
"epoch": 0.6,
"learning_rate": 0.00020327983951855567,
"loss": 2.4981,
"theoretical_loss": 3.434379996548514,
"tokens_seen": 1972371456
},
{
"epoch": 0.6,
"learning_rate": 0.00020326980942828485,
"loss": 2.5514,
"theoretical_loss": 3.4343704561199884,
"tokens_seen": 1972436992
},
{
"epoch": 0.6,
"learning_rate": 0.00020325977933801403,
"loss": 2.5959,
"theoretical_loss": 3.4343609160972006,
"tokens_seen": 1972502528
},
{
"epoch": 0.6,
"learning_rate": 0.00020324974924774324,
"loss": 2.554,
"theoretical_loss": 3.4343513764801203,
"tokens_seen": 1972568064
},
{
"epoch": 0.6,
"objective/train/docs_used": 1115366,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4654228687286377,
"objective/train/theoretical_loss": 3.4343418372687156,
"objective/train/tokens_used": 1993093600,
"theoretical_loss": 3.4343418372687156,
"tokens_seen": 1972633600
},
{
"epoch": 0.6,
"learning_rate": 0.00020323971915747242,
"loss": 2.4279,
"theoretical_loss": 3.4343418372687156,
"tokens_seen": 1972633600
},
{
"epoch": 0.6,
"learning_rate": 0.0002032296890672016,
"loss": 2.5783,
"theoretical_loss": 3.434332298462957,
"tokens_seen": 1972699136
},
{
"epoch": 0.6,
"learning_rate": 0.00020321965897693078,
"loss": 2.5862,
"theoretical_loss": 3.434322760062813,
"tokens_seen": 1972764672
},
{
"epoch": 0.6,
"learning_rate": 0.00020320962888665996,
"loss": 2.6565,
"theoretical_loss": 3.434313222068253,
"tokens_seen": 1972830208
},
{
"epoch": 0.6,
"learning_rate": 0.00020319959879638917,
"loss": 2.5998,
"theoretical_loss": 3.4343036844792465,
"tokens_seen": 1972895744
},
{
"epoch": 0.6,
"learning_rate": 0.00020318956870611835,
"loss": 2.6094,
"theoretical_loss": 3.4342941472957627,
"tokens_seen": 1972961280
},
{
"epoch": 0.6,
"learning_rate": 0.00020317953861584753,
"loss": 2.557,
"theoretical_loss": 3.4342846105177705,
"tokens_seen": 1973026816
},
{
"epoch": 0.6,
"learning_rate": 0.00020316950852557672,
"loss": 2.4615,
"theoretical_loss": 3.4342750741452397,
"tokens_seen": 1973092352
},
{
"epoch": 0.6,
"learning_rate": 0.00020315947843530592,
"loss": 2.3018,
"theoretical_loss": 3.43426553817814,
"tokens_seen": 1973157888
},
{
"epoch": 0.6,
"learning_rate": 0.0002031494483450351,
"loss": 2.3999,
"theoretical_loss": 3.4342560026164395,
"tokens_seen": 1973223424
},
{
"epoch": 0.6,
"learning_rate": 0.00020313941825476429,
"loss": 2.6573,
"theoretical_loss": 3.4342464674601083,
"tokens_seen": 1973288960
},
{
"epoch": 0.6,
"learning_rate": 0.00020312938816449347,
"loss": 2.3483,
"theoretical_loss": 3.4342369327091156,
"tokens_seen": 1973354496
},
{
"epoch": 0.6,
"learning_rate": 0.00020311935807422265,
"loss": 2.7563,
"theoretical_loss": 3.434227398363431,
"tokens_seen": 1973420032
},
{
"epoch": 0.6,
"learning_rate": 0.00020310932798395186,
"loss": 2.6032,
"theoretical_loss": 3.4342178644230232,
"tokens_seen": 1973485568
},
{
"epoch": 0.6,
"learning_rate": 0.00020309929789368104,
"loss": 2.599,
"theoretical_loss": 3.4342083308878615,
"tokens_seen": 1973551104
},
{
"epoch": 0.6,
"learning_rate": 0.00020308926780341022,
"loss": 2.5302,
"theoretical_loss": 3.4341987977579156,
"tokens_seen": 1973616640
},
{
"epoch": 0.6,
"learning_rate": 0.0002030792377131394,
"loss": 2.5513,
"theoretical_loss": 3.434189265033155,
"tokens_seen": 1973682176
},
{
"epoch": 0.6,
"learning_rate": 0.00020306920762286864,
"loss": 2.5615,
"theoretical_loss": 3.4341797327135484,
"tokens_seen": 1973747712
},
{
"epoch": 0.6,
"learning_rate": 0.00020305917753259782,
"loss": 2.4657,
"theoretical_loss": 3.4341702007990658,
"tokens_seen": 1973813248
},
{
"epoch": 0.6,
"learning_rate": 0.000203049147442327,
"loss": 2.565,
"theoretical_loss": 3.434160669289676,
"tokens_seen": 1973878784
},
{
"epoch": 0.6,
"learning_rate": 0.00020303911735205618,
"loss": 2.6913,
"theoretical_loss": 3.4341511381853485,
"tokens_seen": 1973944320
},
{
"epoch": 0.6,
"learning_rate": 0.00020302908726178536,
"loss": 2.5926,
"theoretical_loss": 3.434141607486053,
"tokens_seen": 1974009856
},
{
"epoch": 0.6,
"learning_rate": 0.00020301905717151457,
"loss": 2.739,
"theoretical_loss": 3.4341320771917583,
"tokens_seen": 1974075392
},
{
"epoch": 0.6,
"learning_rate": 0.00020300902708124375,
"loss": 2.5738,
"theoretical_loss": 3.434122547302434,
"tokens_seen": 1974140928
},
{
"epoch": 0.6,
"learning_rate": 0.00020299899699097293,
"loss": 2.55,
"theoretical_loss": 3.4341130178180492,
"tokens_seen": 1974206464
},
{
"epoch": 0.6,
"objective/train/docs_used": 1116806,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2048540115356445,
"objective/train/theoretical_loss": 3.4341034887385735,
"objective/train/tokens_used": 1994732000,
"theoretical_loss": 3.4341034887385735,
"tokens_seen": 1974272000
},
{
"epoch": 0.6,
"learning_rate": 0.0002029889669007021,
"loss": 2.4266,
"theoretical_loss": 3.4341034887385735,
"tokens_seen": 1974272000
},
{
"epoch": 0.6,
"learning_rate": 0.00020297893681043132,
"loss": 2.5755,
"theoretical_loss": 3.4340939600639766,
"tokens_seen": 1974337536
},
{
"epoch": 0.6,
"learning_rate": 0.0002029689067201605,
"loss": 2.4185,
"theoretical_loss": 3.4340844317942274,
"tokens_seen": 1974403072
},
{
"epoch": 0.6,
"learning_rate": 0.00020295887662988968,
"loss": 2.5427,
"theoretical_loss": 3.4340749039292957,
"tokens_seen": 1974468608
},
{
"epoch": 0.6,
"learning_rate": 0.00020294884653961886,
"loss": 2.366,
"theoretical_loss": 3.4340653764691496,
"tokens_seen": 1974534144
},
{
"epoch": 0.6,
"learning_rate": 0.00020293881644934804,
"loss": 2.5359,
"theoretical_loss": 3.43405584941376,
"tokens_seen": 1974599680
},
{
"epoch": 0.6,
"learning_rate": 0.00020292878635907725,
"loss": 2.5047,
"theoretical_loss": 3.4340463227630957,
"tokens_seen": 1974665216
},
{
"epoch": 0.6,
"learning_rate": 0.00020291875626880643,
"loss": 2.5769,
"theoretical_loss": 3.434036796517126,
"tokens_seen": 1974730752
},
{
"epoch": 0.6,
"learning_rate": 0.00020290872617853561,
"loss": 2.4115,
"theoretical_loss": 3.43402727067582,
"tokens_seen": 1974796288
},
{
"epoch": 0.6,
"learning_rate": 0.0002028986960882648,
"loss": 2.4932,
"theoretical_loss": 3.4340177452391476,
"tokens_seen": 1974861824
},
{
"epoch": 0.6,
"learning_rate": 0.000202888665997994,
"loss": 2.5937,
"theoretical_loss": 3.4340082202070774,
"tokens_seen": 1974927360
},
{
"epoch": 0.6,
"learning_rate": 0.00020287863590772318,
"loss": 2.7067,
"theoretical_loss": 3.43399869557958,
"tokens_seen": 1974992896
},
{
"epoch": 0.6,
"learning_rate": 0.00020286860581745237,
"loss": 2.3403,
"theoretical_loss": 3.4339891713566244,
"tokens_seen": 1975058432
},
{
"epoch": 0.6,
"learning_rate": 0.00020285857572718155,
"loss": 2.5488,
"theoretical_loss": 3.433979647538179,
"tokens_seen": 1975123968
},
{
"epoch": 0.6,
"learning_rate": 0.00020284854563691073,
"loss": 2.5848,
"theoretical_loss": 3.4339701241242144,
"tokens_seen": 1975189504
},
{
"epoch": 0.6,
"learning_rate": 0.00020283851554663994,
"loss": 2.4181,
"theoretical_loss": 3.4339606011146993,
"tokens_seen": 1975255040
},
{
"epoch": 0.6,
"learning_rate": 0.00020282848545636912,
"loss": 2.5332,
"theoretical_loss": 3.4339510785096037,
"tokens_seen": 1975320576
},
{
"epoch": 0.6,
"learning_rate": 0.0002028184553660983,
"loss": 2.5331,
"theoretical_loss": 3.4339415563088966,
"tokens_seen": 1975386112
},
{
"epoch": 0.6,
"learning_rate": 0.00020280842527582748,
"loss": 2.6641,
"theoretical_loss": 3.433932034512547,
"tokens_seen": 1975451648
},
{
"epoch": 0.6,
"learning_rate": 0.0002027983951855567,
"loss": 2.6287,
"theoretical_loss": 3.4339225131205255,
"tokens_seen": 1975517184
},
{
"epoch": 0.6,
"learning_rate": 0.00020278836509528587,
"loss": 2.7651,
"theoretical_loss": 3.4339129921328,
"tokens_seen": 1975582720
},
{
"epoch": 0.6,
"learning_rate": 0.00020277833500501505,
"loss": 2.5387,
"theoretical_loss": 3.4339034715493417,
"tokens_seen": 1975648256
},
{
"epoch": 0.6,
"learning_rate": 0.00020276830491474423,
"loss": 2.5429,
"theoretical_loss": 3.433893951370118,
"tokens_seen": 1975713792
},
{
"epoch": 0.6,
"learning_rate": 0.00020275827482447344,
"loss": 2.5823,
"theoretical_loss": 3.4338844315951,
"tokens_seen": 1975779328
},
{
"epoch": 0.6,
"learning_rate": 0.00020274824473420262,
"loss": 2.5618,
"theoretical_loss": 3.4338749122242564,
"tokens_seen": 1975844864
},
{
"epoch": 0.6,
"objective/train/docs_used": 1117380,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.480229377746582,
"objective/train/theoretical_loss": 3.433865393257557,
"objective/train/tokens_used": 1996370400,
"theoretical_loss": 3.433865393257557,
"tokens_seen": 1975910400
},
{
"epoch": 0.6,
"learning_rate": 0.0002027382146439318,
"loss": 2.598,
"theoretical_loss": 3.433865393257557,
"tokens_seen": 1975910400
},
{
"epoch": 0.6,
"learning_rate": 0.00020272818455366098,
"loss": 2.7631,
"theoretical_loss": 3.433855874694971,
"tokens_seen": 1975975936
},
{
"epoch": 0.6,
"learning_rate": 0.00020271815446339016,
"loss": 2.503,
"theoretical_loss": 3.433846356536468,
"tokens_seen": 1976041472
},
{
"epoch": 0.6,
"learning_rate": 0.00020270812437311937,
"loss": 2.6559,
"theoretical_loss": 3.4338368387820166,
"tokens_seen": 1976107008
},
{
"epoch": 0.6,
"learning_rate": 0.00020269809428284855,
"loss": 2.5059,
"theoretical_loss": 3.433827321431587,
"tokens_seen": 1976172544
},
{
"epoch": 0.6,
"learning_rate": 0.00020268806419257773,
"loss": 2.8965,
"theoretical_loss": 3.433817804485149,
"tokens_seen": 1976238080
},
{
"epoch": 0.6,
"learning_rate": 0.00020267803410230692,
"loss": 2.6844,
"theoretical_loss": 3.4338082879426715,
"tokens_seen": 1976303616
},
{
"epoch": 0.6,
"learning_rate": 0.00020266800401203612,
"loss": 2.5098,
"theoretical_loss": 3.433798771804124,
"tokens_seen": 1976369152
},
{
"epoch": 0.6,
"learning_rate": 0.0002026579739217653,
"loss": 2.3285,
"theoretical_loss": 3.433789256069476,
"tokens_seen": 1976434688
},
{
"epoch": 0.6,
"learning_rate": 0.00020264794383149449,
"loss": 2.4844,
"theoretical_loss": 3.433779740738697,
"tokens_seen": 1976500224
},
{
"epoch": 0.6,
"learning_rate": 0.00020263791374122367,
"loss": 2.5684,
"theoretical_loss": 3.433770225811757,
"tokens_seen": 1976565760
},
{
"epoch": 0.6,
"learning_rate": 0.00020262788365095285,
"loss": 2.4572,
"theoretical_loss": 3.433760711288624,
"tokens_seen": 1976631296
},
{
"epoch": 0.6,
"learning_rate": 0.00020261785356068206,
"loss": 2.7182,
"theoretical_loss": 3.4337511971692694,
"tokens_seen": 1976696832
},
{
"epoch": 0.6,
"learning_rate": 0.00020260782347041124,
"loss": 2.6503,
"theoretical_loss": 3.433741683453661,
"tokens_seen": 1976762368
},
{
"epoch": 0.6,
"learning_rate": 0.00020259779338014042,
"loss": 2.3068,
"theoretical_loss": 3.4337321701417696,
"tokens_seen": 1976827904
},
{
"epoch": 0.6,
"learning_rate": 0.0002025877632898696,
"loss": 2.497,
"theoretical_loss": 3.433722657233564,
"tokens_seen": 1976893440
},
{
"epoch": 0.6,
"learning_rate": 0.0002025777331995988,
"loss": 2.4798,
"theoretical_loss": 3.4337131447290132,
"tokens_seen": 1976958976
},
{
"epoch": 0.6,
"learning_rate": 0.000202567703109328,
"loss": 2.625,
"theoretical_loss": 3.4337036326280876,
"tokens_seen": 1977024512
},
{
"epoch": 0.6,
"learning_rate": 0.00020255767301905717,
"loss": 2.3689,
"theoretical_loss": 3.433694120930756,
"tokens_seen": 1977090048
},
{
"epoch": 0.6,
"learning_rate": 0.00020254764292878635,
"loss": 2.6761,
"theoretical_loss": 3.4336846096369893,
"tokens_seen": 1977155584
},
{
"epoch": 0.6,
"learning_rate": 0.00020253761283851553,
"loss": 2.5578,
"theoretical_loss": 3.433675098746755,
"tokens_seen": 1977221120
},
{
"epoch": 0.6,
"learning_rate": 0.00020252758274824474,
"loss": 2.6576,
"theoretical_loss": 3.4336655882600238,
"tokens_seen": 1977286656
},
{
"epoch": 0.6,
"learning_rate": 0.00020251755265797392,
"loss": 2.6556,
"theoretical_loss": 3.433656078176765,
"tokens_seen": 1977352192
},
{
"epoch": 0.6,
"learning_rate": 0.0002025075225677031,
"loss": 2.3872,
"theoretical_loss": 3.4336465684969477,
"tokens_seen": 1977417728
},
{
"epoch": 0.6,
"learning_rate": 0.00020249749247743228,
"loss": 2.2835,
"theoretical_loss": 3.4336370592205423,
"tokens_seen": 1977483264
},
{
"epoch": 0.6,
"objective/train/docs_used": 1118619,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5790343284606934,
"objective/train/theoretical_loss": 3.4336275503475178,
"objective/train/tokens_used": 1998008800,
"theoretical_loss": 3.4336275503475178,
"tokens_seen": 1977548800
},
{
"epoch": 0.6,
"learning_rate": 0.0002024874623871615,
"loss": 2.5669,
"theoretical_loss": 3.4336275503475178,
"tokens_seen": 1977548800
},
{
"epoch": 0.6,
"learning_rate": 0.00020247743229689067,
"loss": 2.6471,
"theoretical_loss": 3.4336180418778435,
"tokens_seen": 1977614336
},
{
"epoch": 0.6,
"learning_rate": 0.00020246740220661985,
"loss": 2.6056,
"theoretical_loss": 3.4336085338114897,
"tokens_seen": 1977679872
},
{
"epoch": 0.6,
"learning_rate": 0.00020245737211634904,
"loss": 2.6313,
"theoretical_loss": 3.433599026148425,
"tokens_seen": 1977745408
},
{
"epoch": 0.6,
"learning_rate": 0.00020244734202607822,
"loss": 2.6348,
"theoretical_loss": 3.4335895188886187,
"tokens_seen": 1977810944
},
{
"epoch": 0.6,
"learning_rate": 0.00020243731193580743,
"loss": 2.6918,
"theoretical_loss": 3.433580012032042,
"tokens_seen": 1977876480
},
{
"epoch": 0.6,
"learning_rate": 0.0002024272818455366,
"loss": 2.4868,
"theoretical_loss": 3.433570505578663,
"tokens_seen": 1977942016
},
{
"epoch": 0.6,
"learning_rate": 0.0002024172517552658,
"loss": 2.6077,
"theoretical_loss": 3.433560999528451,
"tokens_seen": 1978007552
},
{
"epoch": 0.6,
"learning_rate": 0.00020240722166499497,
"loss": 2.508,
"theoretical_loss": 3.433551493881377,
"tokens_seen": 1978073088
},
{
"epoch": 0.6,
"learning_rate": 0.00020239719157472418,
"loss": 2.6041,
"theoretical_loss": 3.4335419886374092,
"tokens_seen": 1978138624
},
{
"epoch": 0.6,
"learning_rate": 0.00020238716148445336,
"loss": 2.7332,
"theoretical_loss": 3.433532483796518,
"tokens_seen": 1978204160
},
{
"epoch": 0.6,
"learning_rate": 0.00020237713139418254,
"loss": 2.5664,
"theoretical_loss": 3.433522979358673,
"tokens_seen": 1978269696
},
{
"epoch": 0.6,
"learning_rate": 0.00020236710130391172,
"loss": 2.7153,
"theoretical_loss": 3.433513475323843,
"tokens_seen": 1978335232
},
{
"epoch": 0.6,
"learning_rate": 0.0002023570712136409,
"loss": 2.3886,
"theoretical_loss": 3.433503971691998,
"tokens_seen": 1978400768
},
{
"epoch": 0.6,
"learning_rate": 0.0002023470411233701,
"loss": 2.5659,
"theoretical_loss": 3.433494468463107,
"tokens_seen": 1978466304
},
{
"epoch": 0.6,
"learning_rate": 0.0002023370110330993,
"loss": 2.5975,
"theoretical_loss": 3.4334849656371405,
"tokens_seen": 1978531840
},
{
"epoch": 0.6,
"learning_rate": 0.0002023269809428285,
"loss": 2.447,
"theoretical_loss": 3.4334754632140676,
"tokens_seen": 1978597376
},
{
"epoch": 0.6,
"learning_rate": 0.00020231695085255768,
"loss": 2.5464,
"theoretical_loss": 3.433465961193858,
"tokens_seen": 1978662912
},
{
"epoch": 0.6,
"learning_rate": 0.0002023069207622869,
"loss": 2.7993,
"theoretical_loss": 3.4334564595764814,
"tokens_seen": 1978728448
},
{
"epoch": 0.6,
"learning_rate": 0.00020229689067201607,
"loss": 2.512,
"theoretical_loss": 3.4334469583619067,
"tokens_seen": 1978793984
},
{
"epoch": 0.6,
"learning_rate": 0.00020228686058174525,
"loss": 2.4429,
"theoretical_loss": 3.433437457550104,
"tokens_seen": 1978859520
},
{
"epoch": 0.6,
"learning_rate": 0.00020227683049147443,
"loss": 2.6064,
"theoretical_loss": 3.4334279571410438,
"tokens_seen": 1978925056
},
{
"epoch": 0.6,
"learning_rate": 0.00020226680040120364,
"loss": 2.7416,
"theoretical_loss": 3.4334184571346933,
"tokens_seen": 1978990592
},
{
"epoch": 0.6,
"learning_rate": 0.00020225677031093282,
"loss": 2.6407,
"theoretical_loss": 3.4334089575310247,
"tokens_seen": 1979056128
},
{
"epoch": 0.6,
"learning_rate": 0.000202246740220662,
"loss": 2.3884,
"theoretical_loss": 3.433399458330006,
"tokens_seen": 1979121664
},
{
"epoch": 0.6,
"objective/train/docs_used": 1119354,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.7723928689956665,
"objective/train/theoretical_loss": 3.433389959531607,
"objective/train/tokens_used": 1999647200,
"theoretical_loss": 3.433389959531607,
"tokens_seen": 1979187200
},
{
"epoch": 0.6,
"learning_rate": 0.00020223671013039118,
"loss": 2.1013,
"theoretical_loss": 3.433389959531607,
"tokens_seen": 1979187200
},
{
"epoch": 0.6,
"learning_rate": 0.00020222668004012036,
"loss": 2.5931,
"theoretical_loss": 3.433380461135798,
"tokens_seen": 1979252736
},
{
"epoch": 0.6,
"learning_rate": 0.00020221664994984957,
"loss": 2.6986,
"theoretical_loss": 3.433370963142548,
"tokens_seen": 1979318272
},
{
"epoch": 0.6,
"learning_rate": 0.00020220661985957875,
"loss": 2.7779,
"theoretical_loss": 3.433361465551827,
"tokens_seen": 1979383808
},
{
"epoch": 0.6,
"learning_rate": 0.00020219658976930793,
"loss": 2.7416,
"theoretical_loss": 3.4333519683636036,
"tokens_seen": 1979449344
},
{
"epoch": 0.6,
"learning_rate": 0.00020218655967903712,
"loss": 2.5905,
"theoretical_loss": 3.433342471577849,
"tokens_seen": 1979514880
},
{
"epoch": 0.6,
"learning_rate": 0.00020217652958876632,
"loss": 2.7701,
"theoretical_loss": 3.4333329751945314,
"tokens_seen": 1979580416
},
{
"epoch": 0.6,
"learning_rate": 0.0002021664994984955,
"loss": 2.5795,
"theoretical_loss": 3.433323479213622,
"tokens_seen": 1979645952
},
{
"epoch": 0.6,
"learning_rate": 0.00020215646940822469,
"loss": 2.6844,
"theoretical_loss": 3.4333139836350886,
"tokens_seen": 1979711488
},
{
"epoch": 0.6,
"learning_rate": 0.00020214643931795387,
"loss": 2.5715,
"theoretical_loss": 3.433304488458902,
"tokens_seen": 1979777024
},
{
"epoch": 0.6,
"learning_rate": 0.00020213640922768305,
"loss": 2.5584,
"theoretical_loss": 3.433294993685031,
"tokens_seen": 1979842560
},
{
"epoch": 0.6,
"learning_rate": 0.00020212637913741226,
"loss": 2.3478,
"theoretical_loss": 3.4332854993134463,
"tokens_seen": 1979908096
},
{
"epoch": 0.6,
"learning_rate": 0.00020211634904714144,
"loss": 2.7309,
"theoretical_loss": 3.4332760053441174,
"tokens_seen": 1979973632
},
{
"epoch": 0.6,
"learning_rate": 0.00020210631895687062,
"loss": 2.6796,
"theoretical_loss": 3.4332665117770125,
"tokens_seen": 1980039168
},
{
"epoch": 0.6,
"learning_rate": 0.0002020962888665998,
"loss": 2.4474,
"theoretical_loss": 3.433257018612103,
"tokens_seen": 1980104704
},
{
"epoch": 0.6,
"learning_rate": 0.000202086258776329,
"loss": 2.4142,
"theoretical_loss": 3.433247525849358,
"tokens_seen": 1980170240
},
{
"epoch": 0.6,
"learning_rate": 0.0002020762286860582,
"loss": 2.5832,
"theoretical_loss": 3.433238033488747,
"tokens_seen": 1980235776
},
{
"epoch": 0.6,
"learning_rate": 0.00020206619859578737,
"loss": 2.6216,
"theoretical_loss": 3.433228541530239,
"tokens_seen": 1980301312
},
{
"epoch": 0.6,
"learning_rate": 0.00020205616850551655,
"loss": 2.5499,
"theoretical_loss": 3.433219049973805,
"tokens_seen": 1980366848
},
{
"epoch": 0.6,
"learning_rate": 0.00020204613841524573,
"loss": 2.6023,
"theoretical_loss": 3.4332095588194136,
"tokens_seen": 1980432384
},
{
"epoch": 0.6,
"learning_rate": 0.00020203610832497494,
"loss": 2.6211,
"theoretical_loss": 3.433200068067035,
"tokens_seen": 1980497920
},
{
"epoch": 0.6,
"learning_rate": 0.00020202607823470412,
"loss": 2.7138,
"theoretical_loss": 3.4331905777166387,
"tokens_seen": 1980563456
},
{
"epoch": 0.6,
"learning_rate": 0.0002020160481444333,
"loss": 2.5634,
"theoretical_loss": 3.4331810877681948,
"tokens_seen": 1980628992
},
{
"epoch": 0.6,
"learning_rate": 0.00020200601805416248,
"loss": 2.1954,
"theoretical_loss": 3.433171598221672,
"tokens_seen": 1980694528
},
{
"epoch": 0.6,
"learning_rate": 0.0002019959879638917,
"loss": 2.6191,
"theoretical_loss": 3.43316210907704,
"tokens_seen": 1980760064
},
{
"epoch": 0.6,
"objective/train/docs_used": 1120511,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5323708057403564,
"objective/train/theoretical_loss": 3.43315262033427,
"objective/train/tokens_used": 2001285600,
"theoretical_loss": 3.43315262033427,
"tokens_seen": 1980825600
},
{
"epoch": 0.6,
"learning_rate": 0.00020198595787362087,
"loss": 2.6631,
"theoretical_loss": 3.43315262033427,
"tokens_seen": 1980825600
},
{
"epoch": 0.6,
"learning_rate": 0.00020197592778335005,
"loss": 2.5179,
"theoretical_loss": 3.4331431319933303,
"tokens_seen": 1980891136
},
{
"epoch": 0.6,
"learning_rate": 0.00020196589769307924,
"loss": 2.6225,
"theoretical_loss": 3.4331336440541915,
"tokens_seen": 1980956672
},
{
"epoch": 0.6,
"learning_rate": 0.00020195586760280842,
"loss": 2.5537,
"theoretical_loss": 3.433124156516823,
"tokens_seen": 1981022208
},
{
"epoch": 0.6,
"learning_rate": 0.00020194583751253763,
"loss": 2.7932,
"theoretical_loss": 3.4331146693811934,
"tokens_seen": 1981087744
},
{
"epoch": 0.6,
"learning_rate": 0.0002019358074222668,
"loss": 2.9028,
"theoretical_loss": 3.4331051826472736,
"tokens_seen": 1981153280
},
{
"epoch": 0.6,
"learning_rate": 0.000201925777331996,
"loss": 2.4817,
"theoretical_loss": 3.433095696315033,
"tokens_seen": 1981218816
},
{
"epoch": 0.6,
"learning_rate": 0.00020191574724172517,
"loss": 2.6063,
"theoretical_loss": 3.4330862103844417,
"tokens_seen": 1981284352
},
{
"epoch": 0.6,
"learning_rate": 0.00020190571715145438,
"loss": 2.6879,
"theoretical_loss": 3.4330767248554688,
"tokens_seen": 1981349888
},
{
"epoch": 0.6,
"learning_rate": 0.00020189568706118356,
"loss": 2.6835,
"theoretical_loss": 3.433067239728084,
"tokens_seen": 1981415424
},
{
"epoch": 0.6,
"learning_rate": 0.00020188565697091274,
"loss": 2.4371,
"theoretical_loss": 3.4330577550022574,
"tokens_seen": 1981480960
},
{
"epoch": 0.6,
"learning_rate": 0.00020187562688064192,
"loss": 2.5715,
"theoretical_loss": 3.4330482706779586,
"tokens_seen": 1981546496
},
{
"epoch": 0.6,
"learning_rate": 0.0002018655967903711,
"loss": 2.7381,
"theoretical_loss": 3.4330387867551573,
"tokens_seen": 1981612032
},
{
"epoch": 0.6,
"learning_rate": 0.0002018555667001003,
"loss": 2.5401,
"theoretical_loss": 3.4330293032338233,
"tokens_seen": 1981677568
},
{
"epoch": 0.6,
"learning_rate": 0.0002018455366098295,
"loss": 2.8218,
"theoretical_loss": 3.433019820113926,
"tokens_seen": 1981743104
},
{
"epoch": 0.6,
"learning_rate": 0.00020183550651955867,
"loss": 2.8752,
"theoretical_loss": 3.4330103373954355,
"tokens_seen": 1981808640
},
{
"epoch": 0.6,
"learning_rate": 0.00020182547642928785,
"loss": 2.7674,
"theoretical_loss": 3.4330008550783213,
"tokens_seen": 1981874176
},
{
"epoch": 0.6,
"learning_rate": 0.00020181544633901706,
"loss": 2.84,
"theoretical_loss": 3.4329913731625536,
"tokens_seen": 1981939712
},
{
"epoch": 0.6,
"learning_rate": 0.00020180541624874624,
"loss": 2.6552,
"theoretical_loss": 3.4329818916481014,
"tokens_seen": 1982005248
},
{
"epoch": 0.6,
"learning_rate": 0.00020179538615847542,
"loss": 2.7176,
"theoretical_loss": 3.4329724105349353,
"tokens_seen": 1982070784
},
{
"epoch": 0.6,
"learning_rate": 0.0002017853560682046,
"loss": 2.8087,
"theoretical_loss": 3.432962929823024,
"tokens_seen": 1982136320
},
{
"epoch": 0.6,
"learning_rate": 0.00020177532597793379,
"loss": 2.7465,
"theoretical_loss": 3.432953449512338,
"tokens_seen": 1982201856
},
{
"epoch": 0.6,
"learning_rate": 0.000201765295887663,
"loss": 2.6213,
"theoretical_loss": 3.432943969602847,
"tokens_seen": 1982267392
},
{
"epoch": 0.6,
"learning_rate": 0.00020175526579739218,
"loss": 2.7821,
"theoretical_loss": 3.432934490094521,
"tokens_seen": 1982332928
},
{
"epoch": 0.6,
"learning_rate": 0.00020174523570712136,
"loss": 2.6115,
"theoretical_loss": 3.4329250109873293,
"tokens_seen": 1982398464
},
{
"epoch": 0.6,
"objective/train/docs_used": 1121183,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.696559429168701,
"objective/train/theoretical_loss": 3.4329155322812417,
"objective/train/tokens_used": 2002924000,
"theoretical_loss": 3.4329155322812417,
"tokens_seen": 1982464000
},
{
"epoch": 0.6,
"learning_rate": 0.00020173520561685054,
"loss": 2.5687,
"theoretical_loss": 3.4329155322812417,
"tokens_seen": 1982464000
},
{
"epoch": 0.6,
"learning_rate": 0.00020172517552657975,
"loss": 2.5187,
"theoretical_loss": 3.432906053976228,
"tokens_seen": 1982529536
},
{
"epoch": 0.6,
"learning_rate": 0.00020171514543630893,
"loss": 2.613,
"theoretical_loss": 3.432896576072258,
"tokens_seen": 1982595072
},
{
"epoch": 0.6,
"learning_rate": 0.0002017051153460381,
"loss": 2.5527,
"theoretical_loss": 3.432887098569301,
"tokens_seen": 1982660608
},
{
"epoch": 0.6,
"learning_rate": 0.0002016950852557673,
"loss": 2.6708,
"theoretical_loss": 3.4328776214673278,
"tokens_seen": 1982726144
},
{
"epoch": 0.6,
"learning_rate": 0.0002016850551654965,
"loss": 2.7673,
"theoretical_loss": 3.432868144766308,
"tokens_seen": 1982791680
},
{
"epoch": 0.6,
"learning_rate": 0.00020167502507522568,
"loss": 2.7018,
"theoretical_loss": 3.432858668466211,
"tokens_seen": 1982857216
},
{
"epoch": 0.6,
"learning_rate": 0.00020166499498495486,
"loss": 2.6459,
"theoretical_loss": 3.432849192567006,
"tokens_seen": 1982922752
},
{
"epoch": 0.6,
"learning_rate": 0.00020165496489468404,
"loss": 2.502,
"theoretical_loss": 3.4328397170686635,
"tokens_seen": 1982988288
},
{
"epoch": 0.6,
"learning_rate": 0.00020164493480441322,
"loss": 2.4582,
"theoretical_loss": 3.4328302419711534,
"tokens_seen": 1983053824
},
{
"epoch": 0.6,
"learning_rate": 0.00020163490471414243,
"loss": 2.7742,
"theoretical_loss": 3.4328207672744453,
"tokens_seen": 1983119360
},
{
"epoch": 0.6,
"learning_rate": 0.0002016248746238716,
"loss": 2.3785,
"theoretical_loss": 3.432811292978509,
"tokens_seen": 1983184896
},
{
"epoch": 0.6,
"learning_rate": 0.0002016148445336008,
"loss": 2.777,
"theoretical_loss": 3.432801819083315,
"tokens_seen": 1983250432
},
{
"epoch": 0.6,
"learning_rate": 0.00020160481444332997,
"loss": 2.6914,
"theoretical_loss": 3.4327923455888314,
"tokens_seen": 1983315968
},
{
"epoch": 0.6,
"learning_rate": 0.00020159478435305918,
"loss": 2.4797,
"theoretical_loss": 3.4327828724950296,
"tokens_seen": 1983381504
},
{
"epoch": 0.6,
"learning_rate": 0.00020158475426278836,
"loss": 2.6505,
"theoretical_loss": 3.432773399801879,
"tokens_seen": 1983447040
},
{
"epoch": 0.6,
"learning_rate": 0.00020157472417251757,
"loss": 2.9179,
"theoretical_loss": 3.432763927509349,
"tokens_seen": 1983512576
},
{
"epoch": 0.6,
"learning_rate": 0.00020156469408224675,
"loss": 2.6086,
"theoretical_loss": 3.4327544556174097,
"tokens_seen": 1983578112
},
{
"epoch": 0.6,
"learning_rate": 0.00020155466399197593,
"loss": 2.658,
"theoretical_loss": 3.432744984126031,
"tokens_seen": 1983643648
},
{
"epoch": 0.6,
"learning_rate": 0.00020154463390170514,
"loss": 2.8361,
"theoretical_loss": 3.4327355130351824,
"tokens_seen": 1983709184
},
{
"epoch": 0.6,
"learning_rate": 0.00020153460381143432,
"loss": 2.63,
"theoretical_loss": 3.4327260423448345,
"tokens_seen": 1983774720
},
{
"epoch": 0.6,
"learning_rate": 0.0002015245737211635,
"loss": 2.5986,
"theoretical_loss": 3.4327165720549564,
"tokens_seen": 1983840256
},
{
"epoch": 0.6,
"learning_rate": 0.00020151454363089268,
"loss": 2.4981,
"theoretical_loss": 3.4327071021655176,
"tokens_seen": 1983905792
},
{
"epoch": 0.6,
"learning_rate": 0.0002015045135406219,
"loss": 2.6175,
"theoretical_loss": 3.432697632676489,
"tokens_seen": 1983971328
},
{
"epoch": 0.6,
"learning_rate": 0.00020149448345035107,
"loss": 2.6047,
"theoretical_loss": 3.43268816358784,
"tokens_seen": 1984036864
},
{
"epoch": 0.6,
"objective/train/docs_used": 1122413,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4781854152679443,
"objective/train/theoretical_loss": 3.4326786948995407,
"objective/train/tokens_used": 2004562400,
"theoretical_loss": 3.4326786948995407,
"tokens_seen": 1984102400
},
{
"epoch": 0.6,
"learning_rate": 0.00020148445336008026,
"loss": 2.5026,
"theoretical_loss": 3.4326786948995407,
"tokens_seen": 1984102400
},
{
"epoch": 0.6,
"learning_rate": 0.00020147442326980944,
"loss": 2.518,
"theoretical_loss": 3.43266922661156,
"tokens_seen": 1984167936
},
{
"epoch": 0.6,
"learning_rate": 0.00020146439317953862,
"loss": 2.7189,
"theoretical_loss": 3.4326597587238687,
"tokens_seen": 1984233472
},
{
"epoch": 0.6,
"learning_rate": 0.00020145436308926783,
"loss": 2.5907,
"theoretical_loss": 3.432650291236436,
"tokens_seen": 1984299008
},
{
"epoch": 0.6,
"learning_rate": 0.000201444332998997,
"loss": 2.7007,
"theoretical_loss": 3.4326408241492326,
"tokens_seen": 1984364544
},
{
"epoch": 0.6,
"learning_rate": 0.0002014343029087262,
"loss": 2.5739,
"theoretical_loss": 3.4326313574622276,
"tokens_seen": 1984430080
},
{
"epoch": 0.6,
"learning_rate": 0.00020142427281845537,
"loss": 2.6113,
"theoretical_loss": 3.432621891175391,
"tokens_seen": 1984495616
},
{
"epoch": 0.6,
"learning_rate": 0.00020141424272818458,
"loss": 2.387,
"theoretical_loss": 3.432612425288693,
"tokens_seen": 1984561152
},
{
"epoch": 0.6,
"learning_rate": 0.00020140421263791376,
"loss": 2.6862,
"theoretical_loss": 3.4326029598021033,
"tokens_seen": 1984626688
},
{
"epoch": 0.6,
"learning_rate": 0.00020139418254764294,
"loss": 2.663,
"theoretical_loss": 3.432593494715592,
"tokens_seen": 1984692224
},
{
"epoch": 0.6,
"learning_rate": 0.00020138415245737212,
"loss": 2.8333,
"theoretical_loss": 3.432584030029128,
"tokens_seen": 1984757760
},
{
"epoch": 0.6,
"learning_rate": 0.0002013741223671013,
"loss": 2.59,
"theoretical_loss": 3.4325745657426827,
"tokens_seen": 1984823296
},
{
"epoch": 0.6,
"learning_rate": 0.0002013640922768305,
"loss": 2.6789,
"theoretical_loss": 3.432565101856224,
"tokens_seen": 1984888832
},
{
"epoch": 0.6,
"learning_rate": 0.0002013540621865597,
"loss": 2.6136,
"theoretical_loss": 3.4325556383697244,
"tokens_seen": 1984954368
},
{
"epoch": 0.6,
"learning_rate": 0.00020134403209628887,
"loss": 2.6919,
"theoretical_loss": 3.432546175283152,
"tokens_seen": 1985019904
},
{
"epoch": 0.6,
"learning_rate": 0.00020133400200601805,
"loss": 2.7094,
"theoretical_loss": 3.4325367125964767,
"tokens_seen": 1985085440
},
{
"epoch": 0.6,
"learning_rate": 0.00020132397191574726,
"loss": 2.6325,
"theoretical_loss": 3.4325272503096693,
"tokens_seen": 1985150976
},
{
"epoch": 0.6,
"learning_rate": 0.00020131394182547644,
"loss": 2.5005,
"theoretical_loss": 3.4325177884226985,
"tokens_seen": 1985216512
},
{
"epoch": 0.6,
"learning_rate": 0.00020130391173520562,
"loss": 2.6425,
"theoretical_loss": 3.4325083269355354,
"tokens_seen": 1985282048
},
{
"epoch": 0.6,
"learning_rate": 0.0002012938816449348,
"loss": 2.409,
"theoretical_loss": 3.4324988658481494,
"tokens_seen": 1985347584
},
{
"epoch": 0.6,
"learning_rate": 0.00020128385155466399,
"loss": 2.6137,
"theoretical_loss": 3.43248940516051,
"tokens_seen": 1985413120
},
{
"epoch": 0.6,
"learning_rate": 0.0002012738214643932,
"loss": 2.5291,
"theoretical_loss": 3.432479944872588,
"tokens_seen": 1985478656
},
{
"epoch": 0.6,
"learning_rate": 0.00020126379137412238,
"loss": 2.5948,
"theoretical_loss": 3.4324704849843526,
"tokens_seen": 1985544192
},
{
"epoch": 0.6,
"learning_rate": 0.00020125376128385156,
"loss": 2.6363,
"theoretical_loss": 3.4324610254957744,
"tokens_seen": 1985609728
},
{
"epoch": 0.6,
"learning_rate": 0.00020124373119358074,
"loss": 2.7139,
"theoretical_loss": 3.4324515664068223,
"tokens_seen": 1985675264
},
{
"epoch": 0.6,
"objective/train/docs_used": 1123215,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.331287145614624,
"objective/train/theoretical_loss": 3.432442107717467,
"objective/train/tokens_used": 2006200800,
"theoretical_loss": 3.432442107717467,
"tokens_seen": 1985740800
},
{
"epoch": 0.6,
"learning_rate": 0.00020123370110330995,
"loss": 2.6268,
"theoretical_loss": 3.432442107717467,
"tokens_seen": 1985740800
},
{
"epoch": 0.6,
"learning_rate": 0.00020122367101303913,
"loss": 2.615,
"theoretical_loss": 3.4324326494276782,
"tokens_seen": 1985806336
},
{
"epoch": 0.6,
"learning_rate": 0.0002012136409227683,
"loss": 2.823,
"theoretical_loss": 3.432423191537426,
"tokens_seen": 1985871872
},
{
"epoch": 0.6,
"learning_rate": 0.0002012036108324975,
"loss": 2.6214,
"theoretical_loss": 3.4324137340466807,
"tokens_seen": 1985937408
},
{
"epoch": 0.6,
"learning_rate": 0.0002011935807422267,
"loss": 2.773,
"theoretical_loss": 3.432404276955411,
"tokens_seen": 1986002944
},
{
"epoch": 0.6,
"learning_rate": 0.00020118355065195588,
"loss": 2.4202,
"theoretical_loss": 3.432394820263588,
"tokens_seen": 1986068480
},
{
"epoch": 0.6,
"learning_rate": 0.00020117352056168506,
"loss": 2.6942,
"theoretical_loss": 3.432385363971181,
"tokens_seen": 1986134016
},
{
"epoch": 0.6,
"learning_rate": 0.00020116349047141424,
"loss": 2.7872,
"theoretical_loss": 3.4323759080781606,
"tokens_seen": 1986199552
},
{
"epoch": 0.6,
"learning_rate": 0.00020115346038114342,
"loss": 2.7499,
"theoretical_loss": 3.432366452584496,
"tokens_seen": 1986265088
},
{
"epoch": 0.6,
"learning_rate": 0.00020114343029087263,
"loss": 2.5759,
"theoretical_loss": 3.4323569974901575,
"tokens_seen": 1986330624
},
{
"epoch": 0.6,
"learning_rate": 0.0002011334002006018,
"loss": 2.7159,
"theoretical_loss": 3.4323475427951156,
"tokens_seen": 1986396160
},
{
"epoch": 0.6,
"learning_rate": 0.000201123370110331,
"loss": 2.458,
"theoretical_loss": 3.4323380884993395,
"tokens_seen": 1986461696
},
{
"epoch": 0.6,
"learning_rate": 0.00020111334002006017,
"loss": 2.8084,
"theoretical_loss": 3.4323286346027997,
"tokens_seen": 1986527232
},
{
"epoch": 0.6,
"learning_rate": 0.00020110330992978938,
"loss": 2.6634,
"theoretical_loss": 3.4323191811054654,
"tokens_seen": 1986592768
},
{
"epoch": 0.6,
"learning_rate": 0.00020109327983951856,
"loss": 2.7405,
"theoretical_loss": 3.432309728007307,
"tokens_seen": 1986658304
},
{
"epoch": 0.6,
"learning_rate": 0.00020108324974924774,
"loss": 2.6382,
"theoretical_loss": 3.432300275308295,
"tokens_seen": 1986723840
},
{
"epoch": 0.6,
"learning_rate": 0.00020107321965897692,
"loss": 2.7143,
"theoretical_loss": 3.4322908230083984,
"tokens_seen": 1986789376
},
{
"epoch": 0.6,
"learning_rate": 0.0002010631895687061,
"loss": 2.6339,
"theoretical_loss": 3.432281371107588,
"tokens_seen": 1986854912
},
{
"epoch": 0.6,
"learning_rate": 0.00020105315947843531,
"loss": 2.5259,
"theoretical_loss": 3.4322719196058338,
"tokens_seen": 1986920448
},
{
"epoch": 0.6,
"learning_rate": 0.0002010431293881645,
"loss": 2.6712,
"theoretical_loss": 3.4322624685031053,
"tokens_seen": 1986985984
},
{
"epoch": 0.6,
"learning_rate": 0.00020103309929789368,
"loss": 2.552,
"theoretical_loss": 3.4322530177993724,
"tokens_seen": 1987051520
},
{
"epoch": 0.6,
"learning_rate": 0.00020102306920762286,
"loss": 2.5998,
"theoretical_loss": 3.4322435674946057,
"tokens_seen": 1987117056
},
{
"epoch": 0.6,
"learning_rate": 0.00020101303911735207,
"loss": 2.4349,
"theoretical_loss": 3.4322341175887745,
"tokens_seen": 1987182592
},
{
"epoch": 0.6,
"learning_rate": 0.00020100300902708125,
"loss": 2.6042,
"theoretical_loss": 3.4322246680818496,
"tokens_seen": 1987248128
},
{
"epoch": 0.6,
"learning_rate": 0.00020099297893681043,
"loss": 2.6272,
"theoretical_loss": 3.4322152189738,
"tokens_seen": 1987313664
},
{
"epoch": 0.6,
"objective/train/docs_used": 1124511,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6344871520996094,
"objective/train/theoretical_loss": 3.4322057702645967,
"objective/train/tokens_used": 2007839200,
"theoretical_loss": 3.4322057702645967,
"tokens_seen": 1987379200
},
{
"epoch": 0.6,
"learning_rate": 0.0002009829488465396,
"loss": 2.544,
"theoretical_loss": 3.4322057702645967,
"tokens_seen": 1987379200
},
{
"epoch": 0.6,
"learning_rate": 0.0002009729187562688,
"loss": 2.4529,
"theoretical_loss": 3.4321963219542098,
"tokens_seen": 1987444736
},
{
"epoch": 0.6,
"learning_rate": 0.000200962888665998,
"loss": 2.5859,
"theoretical_loss": 3.4321868740426082,
"tokens_seen": 1987510272
},
{
"epoch": 0.6,
"learning_rate": 0.00020095285857572718,
"loss": 2.712,
"theoretical_loss": 3.432177426529763,
"tokens_seen": 1987575808
},
{
"epoch": 0.6,
"learning_rate": 0.00020094282848545636,
"loss": 2.4933,
"theoretical_loss": 3.4321679794156434,
"tokens_seen": 1987641344
},
{
"epoch": 0.6,
"learning_rate": 0.00020093279839518554,
"loss": 2.6133,
"theoretical_loss": 3.4321585327002198,
"tokens_seen": 1987706880
},
{
"epoch": 0.6,
"learning_rate": 0.00020092276830491475,
"loss": 2.5922,
"theoretical_loss": 3.432149086383462,
"tokens_seen": 1987772416
},
{
"epoch": 0.6,
"learning_rate": 0.00020091273821464393,
"loss": 2.6491,
"theoretical_loss": 3.4321396404653406,
"tokens_seen": 1987837952
},
{
"epoch": 0.6,
"learning_rate": 0.0002009027081243731,
"loss": 2.6913,
"theoretical_loss": 3.4321301949458256,
"tokens_seen": 1987903488
},
{
"epoch": 0.6,
"learning_rate": 0.0002008926780341023,
"loss": 2.5613,
"theoretical_loss": 3.432120749824886,
"tokens_seen": 1987969024
},
{
"epoch": 0.6,
"learning_rate": 0.00020088264794383147,
"loss": 2.4847,
"theoretical_loss": 3.4321113051024934,
"tokens_seen": 1988034560
},
{
"epoch": 0.6,
"learning_rate": 0.00020087261785356068,
"loss": 2.6195,
"theoretical_loss": 3.4321018607786167,
"tokens_seen": 1988100096
},
{
"epoch": 0.6,
"learning_rate": 0.00020086258776328986,
"loss": 2.8115,
"theoretical_loss": 3.4320924168532256,
"tokens_seen": 1988165632
},
{
"epoch": 0.6,
"learning_rate": 0.00020085255767301905,
"loss": 2.7888,
"theoretical_loss": 3.432082973326292,
"tokens_seen": 1988231168
},
{
"epoch": 0.6,
"learning_rate": 0.00020084252758274823,
"loss": 2.5526,
"theoretical_loss": 3.432073530197784,
"tokens_seen": 1988296704
},
{
"epoch": 0.6,
"learning_rate": 0.00020083249749247743,
"loss": 2.6244,
"theoretical_loss": 3.4320640874676727,
"tokens_seen": 1988362240
},
{
"epoch": 0.6,
"learning_rate": 0.00020082246740220664,
"loss": 2.6596,
"theoretical_loss": 3.4320546451359277,
"tokens_seen": 1988427776
},
{
"epoch": 0.6,
"learning_rate": 0.00020081243731193582,
"loss": 2.7242,
"theoretical_loss": 3.4320452032025193,
"tokens_seen": 1988493312
},
{
"epoch": 0.6,
"learning_rate": 0.000200802407221665,
"loss": 2.788,
"theoretical_loss": 3.4320357616674175,
"tokens_seen": 1988558848
},
{
"epoch": 0.6,
"learning_rate": 0.00020079237713139419,
"loss": 2.65,
"theoretical_loss": 3.4320263205305928,
"tokens_seen": 1988624384
},
{
"epoch": 0.6,
"learning_rate": 0.0002007823470411234,
"loss": 2.5966,
"theoretical_loss": 3.4320168797920143,
"tokens_seen": 1988689920
},
{
"epoch": 0.6,
"learning_rate": 0.00020077231695085258,
"loss": 2.5685,
"theoretical_loss": 3.432007439451653,
"tokens_seen": 1988755456
},
{
"epoch": 0.6,
"learning_rate": 0.00020076228686058176,
"loss": 2.6215,
"theoretical_loss": 3.4319979995094787,
"tokens_seen": 1988820992
},
{
"epoch": 0.6,
"learning_rate": 0.00020075225677031094,
"loss": 2.5364,
"theoretical_loss": 3.431988559965461,
"tokens_seen": 1988886528
},
{
"epoch": 0.6,
"learning_rate": 0.00020074222668004015,
"loss": 2.4655,
"theoretical_loss": 3.431979120819571,
"tokens_seen": 1988952064
},
{
"epoch": 0.6,
"objective/train/docs_used": 1125297,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5770888328552246,
"objective/train/theoretical_loss": 3.4319696820717778,
"objective/train/tokens_used": 2009477600,
"theoretical_loss": 3.4319696820717778,
"tokens_seen": 1989017600
},
{
"epoch": 0.6,
"learning_rate": 0.00020073219658976933,
"loss": 2.5541,
"theoretical_loss": 3.4319696820717778,
"tokens_seen": 1989017600
},
{
"epoch": 0.6,
"learning_rate": 0.0002007221664994985,
"loss": 2.5472,
"theoretical_loss": 3.431960243722052,
"tokens_seen": 1989083136
},
{
"epoch": 0.6,
"learning_rate": 0.0002007121364092277,
"loss": 2.6244,
"theoretical_loss": 3.4319508057703634,
"tokens_seen": 1989148672
},
{
"epoch": 0.6,
"learning_rate": 0.0002007021063189569,
"loss": 2.5015,
"theoretical_loss": 3.4319413682166826,
"tokens_seen": 1989214208
},
{
"epoch": 0.6,
"learning_rate": 0.00020069207622868608,
"loss": 2.794,
"theoretical_loss": 3.431931931060979,
"tokens_seen": 1989279744
},
{
"epoch": 0.6,
"learning_rate": 0.00020068204613841526,
"loss": 2.6056,
"theoretical_loss": 3.4319224943032234,
"tokens_seen": 1989345280
},
{
"epoch": 0.6,
"learning_rate": 0.00020067201604814444,
"loss": 2.5884,
"theoretical_loss": 3.431913057943386,
"tokens_seen": 1989410816
},
{
"epoch": 0.6,
"learning_rate": 0.00020066198595787362,
"loss": 2.7906,
"theoretical_loss": 3.4319036219814363,
"tokens_seen": 1989476352
},
{
"epoch": 0.6,
"learning_rate": 0.00020065195586760283,
"loss": 2.6446,
"theoretical_loss": 3.4318941864173445,
"tokens_seen": 1989541888
},
{
"epoch": 0.6,
"learning_rate": 0.000200641925777332,
"loss": 2.4929,
"theoretical_loss": 3.4318847512510806,
"tokens_seen": 1989607424
},
{
"epoch": 0.6,
"learning_rate": 0.0002006318956870612,
"loss": 2.6389,
"theoretical_loss": 3.431875316482615,
"tokens_seen": 1989672960
},
{
"epoch": 0.6,
"learning_rate": 0.00020062186559679037,
"loss": 2.6823,
"theoretical_loss": 3.4318658821119183,
"tokens_seen": 1989738496
},
{
"epoch": 0.6,
"learning_rate": 0.00020061183550651958,
"loss": 2.4796,
"theoretical_loss": 3.43185644813896,
"tokens_seen": 1989804032
},
{
"epoch": 0.6,
"learning_rate": 0.00020060180541624876,
"loss": 2.4299,
"theoretical_loss": 3.43184701456371,
"tokens_seen": 1989869568
},
{
"epoch": 0.6,
"learning_rate": 0.00020059177532597794,
"loss": 2.6316,
"theoretical_loss": 3.431837581386139,
"tokens_seen": 1989935104
},
{
"epoch": 0.6,
"learning_rate": 0.00020058174523570713,
"loss": 2.4635,
"theoretical_loss": 3.4318281486062174,
"tokens_seen": 1990000640
},
{
"epoch": 0.6,
"learning_rate": 0.0002005717151454363,
"loss": 2.5731,
"theoretical_loss": 3.4318187162239147,
"tokens_seen": 1990066176
},
{
"epoch": 0.6,
"learning_rate": 0.00020056168505516551,
"loss": 2.609,
"theoretical_loss": 3.4318092842392005,
"tokens_seen": 1990131712
},
{
"epoch": 0.6,
"learning_rate": 0.0002005516549648947,
"loss": 2.5496,
"theoretical_loss": 3.4317998526520466,
"tokens_seen": 1990197248
},
{
"epoch": 0.6,
"learning_rate": 0.00020054162487462388,
"loss": 2.6141,
"theoretical_loss": 3.4317904214624217,
"tokens_seen": 1990262784
},
{
"epoch": 0.6,
"learning_rate": 0.00020053159478435306,
"loss": 2.5639,
"theoretical_loss": 3.4317809906702967,
"tokens_seen": 1990328320
},
{
"epoch": 0.6,
"learning_rate": 0.00020052156469408227,
"loss": 2.8157,
"theoretical_loss": 3.4317715602756413,
"tokens_seen": 1990393856
},
{
"epoch": 0.6,
"learning_rate": 0.00020051153460381145,
"loss": 2.8178,
"theoretical_loss": 3.4317621302784262,
"tokens_seen": 1990459392
},
{
"epoch": 0.6,
"learning_rate": 0.00020050150451354063,
"loss": 2.5964,
"theoretical_loss": 3.4317527006786217,
"tokens_seen": 1990524928
},
{
"epoch": 0.6,
"learning_rate": 0.0002004914744232698,
"loss": 2.6623,
"theoretical_loss": 3.431743271476197,
"tokens_seen": 1990590464
},
{
"epoch": 0.6,
"objective/train/docs_used": 1126404,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3342862129211426,
"objective/train/theoretical_loss": 3.4317338426711226,
"objective/train/tokens_used": 2011116000,
"theoretical_loss": 3.4317338426711226,
"tokens_seen": 1990656000
},
{
"epoch": 0.6,
"learning_rate": 0.000200481444332999,
"loss": 2.4303,
"theoretical_loss": 3.4317338426711226,
"tokens_seen": 1990656000
},
{
"epoch": 0.6,
"learning_rate": 0.0002004714142427282,
"loss": 2.4879,
"theoretical_loss": 3.4317244142633694,
"tokens_seen": 1990721536
},
{
"epoch": 0.6,
"learning_rate": 0.00020046138415245738,
"loss": 2.481,
"theoretical_loss": 3.4317149862529073,
"tokens_seen": 1990787072
},
{
"epoch": 0.6,
"learning_rate": 0.00020045135406218656,
"loss": 2.693,
"theoretical_loss": 3.4317055586397056,
"tokens_seen": 1990852608
},
{
"epoch": 0.6,
"learning_rate": 0.00020044132397191574,
"loss": 2.7728,
"theoretical_loss": 3.4316961314237355,
"tokens_seen": 1990918144
},
{
"epoch": 0.6,
"learning_rate": 0.00020043129388164495,
"loss": 2.7932,
"theoretical_loss": 3.431686704604967,
"tokens_seen": 1990983680
},
{
"epoch": 0.6,
"learning_rate": 0.00020042126379137413,
"loss": 2.5672,
"theoretical_loss": 3.4316772781833693,
"tokens_seen": 1991049216
},
{
"epoch": 0.6,
"learning_rate": 0.0002004112337011033,
"loss": 2.4807,
"theoretical_loss": 3.431667852158914,
"tokens_seen": 1991114752
},
{
"epoch": 0.6,
"learning_rate": 0.0002004012036108325,
"loss": 2.5105,
"theoretical_loss": 3.4316584265315706,
"tokens_seen": 1991180288
},
{
"epoch": 0.6,
"learning_rate": 0.00020039117352056167,
"loss": 2.48,
"theoretical_loss": 3.43164900130131,
"tokens_seen": 1991245824
},
{
"epoch": 0.6,
"learning_rate": 0.00020038114343029088,
"loss": 2.4618,
"theoretical_loss": 3.4316395764681014,
"tokens_seen": 1991311360
},
{
"epoch": 0.6,
"learning_rate": 0.00020037111334002006,
"loss": 2.549,
"theoretical_loss": 3.4316301520319152,
"tokens_seen": 1991376896
},
{
"epoch": 0.6,
"learning_rate": 0.00020036108324974925,
"loss": 2.6786,
"theoretical_loss": 3.4316207279927218,
"tokens_seen": 1991442432
},
{
"epoch": 0.6,
"learning_rate": 0.00020035105315947843,
"loss": 2.7542,
"theoretical_loss": 3.4316113043504917,
"tokens_seen": 1991507968
},
{
"epoch": 0.6,
"learning_rate": 0.00020034102306920763,
"loss": 2.1359,
"theoretical_loss": 3.4316018811051947,
"tokens_seen": 1991573504
},
{
"epoch": 0.6,
"learning_rate": 0.00020033099297893682,
"loss": 2.7766,
"theoretical_loss": 3.4315924582568016,
"tokens_seen": 1991639040
},
{
"epoch": 0.6,
"learning_rate": 0.000200320962888666,
"loss": 2.6955,
"theoretical_loss": 3.4315830358052817,
"tokens_seen": 1991704576
},
{
"epoch": 0.6,
"learning_rate": 0.00020031093279839518,
"loss": 2.5396,
"theoretical_loss": 3.431573613750606,
"tokens_seen": 1991770112
},
{
"epoch": 0.6,
"learning_rate": 0.00020030090270812436,
"loss": 2.3725,
"theoretical_loss": 3.431564192092744,
"tokens_seen": 1991835648
},
{
"epoch": 0.6,
"learning_rate": 0.00020029087261785357,
"loss": 2.7523,
"theoretical_loss": 3.431554770831667,
"tokens_seen": 1991901184
},
{
"epoch": 0.6,
"learning_rate": 0.00020028084252758275,
"loss": 2.5134,
"theoretical_loss": 3.4315453499673443,
"tokens_seen": 1991966720
},
{
"epoch": 0.6,
"learning_rate": 0.00020027081243731193,
"loss": 2.6905,
"theoretical_loss": 3.4315359294997467,
"tokens_seen": 1992032256
},
{
"epoch": 0.6,
"learning_rate": 0.0002002607823470411,
"loss": 2.6512,
"theoretical_loss": 3.431526509428844,
"tokens_seen": 1992097792
},
{
"epoch": 0.6,
"learning_rate": 0.00020025075225677032,
"loss": 2.5763,
"theoretical_loss": 3.431517089754607,
"tokens_seen": 1992163328
},
{
"epoch": 0.6,
"learning_rate": 0.0002002407221664995,
"loss": 2.6469,
"theoretical_loss": 3.4315076704770053,
"tokens_seen": 1992228864
},
{
"epoch": 0.6,
"objective/train/docs_used": 1126980,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8636534214019775,
"objective/train/theoretical_loss": 3.4314982515960093,
"objective/train/tokens_used": 2012754400,
"theoretical_loss": 3.4314982515960093,
"tokens_seen": 1992294400
},
{
"epoch": 0.6,
"learning_rate": 0.00020023069207622868,
"loss": 2.656,
"theoretical_loss": 3.4314982515960093,
"tokens_seen": 1992294400
},
{
"epoch": 0.6,
"learning_rate": 0.00020022066198595786,
"loss": 2.6137,
"theoretical_loss": 3.4314888331115894,
"tokens_seen": 1992359936
},
{
"epoch": 0.6,
"learning_rate": 0.00020021063189568704,
"loss": 2.5492,
"theoretical_loss": 3.4314794150237162,
"tokens_seen": 1992425472
},
{
"epoch": 0.6,
"learning_rate": 0.00020020060180541625,
"loss": 2.4938,
"theoretical_loss": 3.4314699973323597,
"tokens_seen": 1992491008
},
{
"epoch": 0.6,
"learning_rate": 0.00020019057171514543,
"loss": 2.5735,
"theoretical_loss": 3.4314605800374895,
"tokens_seen": 1992556544
},
{
"epoch": 0.6,
"learning_rate": 0.00020018054162487461,
"loss": 2.3878,
"theoretical_loss": 3.4314511631390774,
"tokens_seen": 1992622080
},
{
"epoch": 0.6,
"learning_rate": 0.0002001705115346038,
"loss": 2.6584,
"theoretical_loss": 3.431441746637092,
"tokens_seen": 1992687616
},
{
"epoch": 0.6,
"learning_rate": 0.000200160481444333,
"loss": 2.462,
"theoretical_loss": 3.4314323305315044,
"tokens_seen": 1992753152
},
{
"epoch": 0.6,
"learning_rate": 0.00020015045135406218,
"loss": 2.5287,
"theoretical_loss": 3.431422914822285,
"tokens_seen": 1992818688
},
{
"epoch": 0.6,
"learning_rate": 0.00020014042126379137,
"loss": 2.7109,
"theoretical_loss": 3.431413499509404,
"tokens_seen": 1992884224
},
{
"epoch": 0.6,
"learning_rate": 0.00020013039117352055,
"loss": 2.7126,
"theoretical_loss": 3.431404084592831,
"tokens_seen": 1992949760
},
{
"epoch": 0.6,
"learning_rate": 0.00020012036108324973,
"loss": 2.4587,
"theoretical_loss": 3.4313946700725375,
"tokens_seen": 1993015296
},
{
"epoch": 0.6,
"learning_rate": 0.00020011033099297894,
"loss": 2.52,
"theoretical_loss": 3.4313852559484928,
"tokens_seen": 1993080832
},
{
"epoch": 0.6,
"learning_rate": 0.00020010030090270812,
"loss": 2.4799,
"theoretical_loss": 3.4313758422206675,
"tokens_seen": 1993146368
},
{
"epoch": 0.6,
"learning_rate": 0.0002000902708124373,
"loss": 2.7896,
"theoretical_loss": 3.4313664288890324,
"tokens_seen": 1993211904
},
{
"epoch": 0.6,
"learning_rate": 0.00020008024072216648,
"loss": 2.7177,
"theoretical_loss": 3.431357015953557,
"tokens_seen": 1993277440
},
{
"epoch": 0.6,
"learning_rate": 0.00020007021063189571,
"loss": 2.5497,
"theoretical_loss": 3.4313476034142116,
"tokens_seen": 1993342976
},
{
"epoch": 0.6,
"learning_rate": 0.0002000601805416249,
"loss": 2.6153,
"theoretical_loss": 3.431338191270967,
"tokens_seen": 1993408512
},
{
"epoch": 0.6,
"learning_rate": 0.00020005015045135408,
"loss": 2.5958,
"theoretical_loss": 3.4313287795237932,
"tokens_seen": 1993474048
},
{
"epoch": 0.6,
"learning_rate": 0.00020004012036108326,
"loss": 2.623,
"theoretical_loss": 3.4313193681726606,
"tokens_seen": 1993539584
},
{
"epoch": 0.6,
"learning_rate": 0.00020003009027081247,
"loss": 2.6444,
"theoretical_loss": 3.4313099572175405,
"tokens_seen": 1993605120
},
{
"epoch": 0.6,
"learning_rate": 0.00020002006018054165,
"loss": 2.8482,
"theoretical_loss": 3.4313005466584015,
"tokens_seen": 1993670656
},
{
"epoch": 0.6,
"learning_rate": 0.00020001003009027083,
"loss": 2.3883,
"theoretical_loss": 3.431291136495215,
"tokens_seen": 1993736192
},
{
"epoch": 0.6,
"learning_rate": 0.0002,
"loss": 2.8274,
"theoretical_loss": 3.4312817267279505,
"tokens_seen": 1993801728
},
{
"epoch": 0.6,
"learning_rate": 0.0001999899699097292,
"loss": 2.6056,
"theoretical_loss": 3.431272317356579,
"tokens_seen": 1993867264
},
{
"epoch": 0.6,
"objective/train/docs_used": 1127596,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3768882751464844,
"objective/train/theoretical_loss": 3.431262908381071,
"objective/train/tokens_used": 2014392800,
"theoretical_loss": 3.431262908381071,
"tokens_seen": 1993932800
},
{
"epoch": 0.6,
"learning_rate": 0.0001999799398194584,
"loss": 2.845,
"theoretical_loss": 3.431262908381071,
"tokens_seen": 1993932800
},
{
"epoch": 0.6,
"learning_rate": 0.00019996990972918758,
"loss": 2.5958,
"theoretical_loss": 3.4312534998013966,
"tokens_seen": 1993998336
},
{
"epoch": 0.6,
"learning_rate": 0.00019995987963891676,
"loss": 2.614,
"theoretical_loss": 3.431244091617526,
"tokens_seen": 1994063872
},
{
"epoch": 0.6,
"learning_rate": 0.00019994984954864594,
"loss": 2.5922,
"theoretical_loss": 3.4312346838294294,
"tokens_seen": 1994129408
},
{
"epoch": 0.6,
"learning_rate": 0.00019993981945837515,
"loss": 2.7681,
"theoretical_loss": 3.431225276437078,
"tokens_seen": 1994194944
},
{
"epoch": 0.6,
"learning_rate": 0.00019992978936810433,
"loss": 2.6554,
"theoretical_loss": 3.431215869440441,
"tokens_seen": 1994260480
},
{
"epoch": 0.6,
"learning_rate": 0.0001999197592778335,
"loss": 2.6081,
"theoretical_loss": 3.431206462839489,
"tokens_seen": 1994326016
},
{
"epoch": 0.6,
"learning_rate": 0.0001999097291875627,
"loss": 2.6327,
"theoretical_loss": 3.4311970566341934,
"tokens_seen": 1994391552
},
{
"epoch": 0.6,
"learning_rate": 0.00019989969909729187,
"loss": 2.5546,
"theoretical_loss": 3.4311876508245227,
"tokens_seen": 1994457088
},
{
"epoch": 0.6,
"learning_rate": 0.00019988966900702108,
"loss": 2.6168,
"theoretical_loss": 3.4311782454104494,
"tokens_seen": 1994522624
},
{
"epoch": 0.6,
"learning_rate": 0.00019987963891675026,
"loss": 2.4941,
"theoretical_loss": 3.4311688403919423,
"tokens_seen": 1994588160
},
{
"epoch": 0.6,
"learning_rate": 0.00019986960882647945,
"loss": 2.4786,
"theoretical_loss": 3.431159435768972,
"tokens_seen": 1994653696
},
{
"epoch": 0.6,
"learning_rate": 0.00019985957873620863,
"loss": 2.5746,
"theoretical_loss": 3.4311500315415095,
"tokens_seen": 1994719232
},
{
"epoch": 0.6,
"learning_rate": 0.00019984954864593783,
"loss": 2.7045,
"theoretical_loss": 3.431140627709525,
"tokens_seen": 1994784768
},
{
"epoch": 0.6,
"learning_rate": 0.00019983951855566702,
"loss": 2.5571,
"theoretical_loss": 3.4311312242729883,
"tokens_seen": 1994850304
},
{
"epoch": 0.6,
"learning_rate": 0.0001998294884653962,
"loss": 2.6139,
"theoretical_loss": 3.43112182123187,
"tokens_seen": 1994915840
},
{
"epoch": 0.6,
"learning_rate": 0.00019981945837512538,
"loss": 2.393,
"theoretical_loss": 3.431112418586141,
"tokens_seen": 1994981376
},
{
"epoch": 0.6,
"learning_rate": 0.00019980942828485456,
"loss": 2.7276,
"theoretical_loss": 3.431103016335771,
"tokens_seen": 1995046912
},
{
"epoch": 0.6,
"learning_rate": 0.00019979939819458377,
"loss": 2.5107,
"theoretical_loss": 3.4310936144807314,
"tokens_seen": 1995112448
},
{
"epoch": 0.6,
"learning_rate": 0.00019978936810431295,
"loss": 2.6525,
"theoretical_loss": 3.4310842130209913,
"tokens_seen": 1995177984
},
{
"epoch": 0.6,
"learning_rate": 0.00019977933801404213,
"loss": 2.4777,
"theoretical_loss": 3.431074811956522,
"tokens_seen": 1995243520
},
{
"epoch": 0.6,
"learning_rate": 0.0001997693079237713,
"loss": 2.8106,
"theoretical_loss": 3.4310654112872934,
"tokens_seen": 1995309056
},
{
"epoch": 0.6,
"learning_rate": 0.00019975927783350052,
"loss": 2.5843,
"theoretical_loss": 3.4310560110132764,
"tokens_seen": 1995374592
},
{
"epoch": 0.6,
"learning_rate": 0.0001997492477432297,
"loss": 2.6041,
"theoretical_loss": 3.431046611134441,
"tokens_seen": 1995440128
},
{
"epoch": 0.6,
"learning_rate": 0.00019973921765295888,
"loss": 2.6167,
"theoretical_loss": 3.4310372116507573,
"tokens_seen": 1995505664
},
{
"epoch": 0.6,
"objective/train/docs_used": 1128921,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.722097635269165,
"objective/train/theoretical_loss": 3.4310278125621965,
"objective/train/tokens_used": 2016031200,
"theoretical_loss": 3.4310278125621965,
"tokens_seen": 1995571200
},
{
"epoch": 0.6,
"learning_rate": 0.00019972918756268806,
"loss": 2.4771,
"theoretical_loss": 3.4310278125621965,
"tokens_seen": 1995571200
},
{
"epoch": 0.6,
"learning_rate": 0.00019971915747241724,
"loss": 2.6832,
"theoretical_loss": 3.431018413868729,
"tokens_seen": 1995636736
},
{
"epoch": 0.6,
"learning_rate": 0.00019970912738214645,
"loss": 2.2566,
"theoretical_loss": 3.4310090155703246,
"tokens_seen": 1995702272
},
{
"epoch": 0.6,
"learning_rate": 0.00019969909729187563,
"loss": 2.4638,
"theoretical_loss": 3.430999617666954,
"tokens_seen": 1995767808
},
{
"epoch": 0.6,
"learning_rate": 0.00019968906720160481,
"loss": 2.6222,
"theoretical_loss": 3.430990220158588,
"tokens_seen": 1995833344
},
{
"epoch": 0.6,
"learning_rate": 0.000199679037111334,
"loss": 2.5972,
"theoretical_loss": 3.4309808230451955,
"tokens_seen": 1995898880
},
{
"epoch": 0.6,
"learning_rate": 0.0001996690070210632,
"loss": 2.3397,
"theoretical_loss": 3.430971426326749,
"tokens_seen": 1995964416
},
{
"epoch": 0.6,
"learning_rate": 0.00019965897693079238,
"loss": 2.4669,
"theoretical_loss": 3.4309620300032178,
"tokens_seen": 1996029952
},
{
"epoch": 0.6,
"learning_rate": 0.00019964894684052157,
"loss": 2.3157,
"theoretical_loss": 3.430952634074573,
"tokens_seen": 1996095488
},
{
"epoch": 0.6,
"learning_rate": 0.00019963891675025075,
"loss": 2.4344,
"theoretical_loss": 3.4309432385407836,
"tokens_seen": 1996161024
},
{
"epoch": 0.6,
"learning_rate": 0.00019962888665997993,
"loss": 2.5019,
"theoretical_loss": 3.4309338434018217,
"tokens_seen": 1996226560
},
{
"epoch": 0.6,
"learning_rate": 0.00019961885656970914,
"loss": 2.6694,
"theoretical_loss": 3.430924448657657,
"tokens_seen": 1996292096
},
{
"epoch": 0.6,
"learning_rate": 0.00019960882647943832,
"loss": 2.6165,
"theoretical_loss": 3.43091505430826,
"tokens_seen": 1996357632
},
{
"epoch": 0.6,
"learning_rate": 0.0001995987963891675,
"loss": 2.7511,
"theoretical_loss": 3.4309056603536012,
"tokens_seen": 1996423168
},
{
"epoch": 0.6,
"learning_rate": 0.00019958876629889668,
"loss": 2.3224,
"theoretical_loss": 3.430896266793651,
"tokens_seen": 1996488704
},
{
"epoch": 0.61,
"learning_rate": 0.0001995787362086259,
"loss": 2.5314,
"theoretical_loss": 3.43088687362838,
"tokens_seen": 1996554240
},
{
"epoch": 0.61,
"learning_rate": 0.00019956870611835507,
"loss": 2.7181,
"theoretical_loss": 3.4308774808577587,
"tokens_seen": 1996619776
},
{
"epoch": 0.61,
"learning_rate": 0.00019955867602808425,
"loss": 2.5873,
"theoretical_loss": 3.4308680884817573,
"tokens_seen": 1996685312
},
{
"epoch": 0.61,
"learning_rate": 0.00019954864593781343,
"loss": 2.5753,
"theoretical_loss": 3.430858696500346,
"tokens_seen": 1996750848
},
{
"epoch": 0.61,
"learning_rate": 0.00019953861584754264,
"loss": 2.5597,
"theoretical_loss": 3.430849304913496,
"tokens_seen": 1996816384
},
{
"epoch": 0.61,
"learning_rate": 0.00019952858575727182,
"loss": 2.6156,
"theoretical_loss": 3.4308399137211776,
"tokens_seen": 1996881920
},
{
"epoch": 0.61,
"learning_rate": 0.000199518555667001,
"loss": 2.676,
"theoretical_loss": 3.430830522923361,
"tokens_seen": 1996947456
},
{
"epoch": 0.61,
"learning_rate": 0.00019950852557673018,
"loss": 2.4787,
"theoretical_loss": 3.4308211325200166,
"tokens_seen": 1997012992
},
{
"epoch": 0.61,
"learning_rate": 0.00019949849548645936,
"loss": 2.6819,
"theoretical_loss": 3.430811742511115,
"tokens_seen": 1997078528
},
{
"epoch": 0.61,
"learning_rate": 0.00019948846539618857,
"loss": 2.4907,
"theoretical_loss": 3.4308023528966274,
"tokens_seen": 1997144064
},
{
"epoch": 0.61,
"objective/train/docs_used": 1129541,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3653786182403564,
"objective/train/theoretical_loss": 3.4307929636765233,
"objective/train/tokens_used": 2017669600,
"theoretical_loss": 3.4307929636765233,
"tokens_seen": 1997209600
},
{
"epoch": 0.61,
"learning_rate": 0.00019947843530591775,
"loss": 2.4624,
"theoretical_loss": 3.4307929636765233,
"tokens_seen": 1997209600
},
{
"epoch": 0.61,
"learning_rate": 0.00019946840521564693,
"loss": 2.5293,
"theoretical_loss": 3.4307835748507736,
"tokens_seen": 1997275136
},
{
"epoch": 0.61,
"learning_rate": 0.00019945837512537612,
"loss": 2.662,
"theoretical_loss": 3.430774186419349,
"tokens_seen": 1997340672
},
{
"epoch": 0.61,
"learning_rate": 0.00019944834503510532,
"loss": 2.742,
"theoretical_loss": 3.4307647983822194,
"tokens_seen": 1997406208
},
{
"epoch": 0.61,
"learning_rate": 0.0001994383149448345,
"loss": 2.3246,
"theoretical_loss": 3.430755410739356,
"tokens_seen": 1997471744
},
{
"epoch": 0.61,
"learning_rate": 0.00019942828485456369,
"loss": 2.5897,
"theoretical_loss": 3.4307460234907285,
"tokens_seen": 1997537280
},
{
"epoch": 0.61,
"learning_rate": 0.00019941825476429287,
"loss": 2.6472,
"theoretical_loss": 3.430736636636308,
"tokens_seen": 1997602816
},
{
"epoch": 0.61,
"learning_rate": 0.00019940822467402205,
"loss": 2.3168,
"theoretical_loss": 3.430727250176065,
"tokens_seen": 1997668352
},
{
"epoch": 0.61,
"learning_rate": 0.00019939819458375126,
"loss": 2.5027,
"theoretical_loss": 3.4307178641099703,
"tokens_seen": 1997733888
},
{
"epoch": 0.61,
"learning_rate": 0.00019938816449348044,
"loss": 2.5896,
"theoretical_loss": 3.4307084784379933,
"tokens_seen": 1997799424
},
{
"epoch": 0.61,
"learning_rate": 0.00019937813440320962,
"loss": 2.5902,
"theoretical_loss": 3.430699093160106,
"tokens_seen": 1997864960
},
{
"epoch": 0.61,
"learning_rate": 0.0001993681043129388,
"loss": 2.6683,
"theoretical_loss": 3.4306897082762777,
"tokens_seen": 1997930496
},
{
"epoch": 0.61,
"learning_rate": 0.000199358074222668,
"loss": 2.3998,
"theoretical_loss": 3.4306803237864796,
"tokens_seen": 1997996032
},
{
"epoch": 0.61,
"learning_rate": 0.0001993480441323972,
"loss": 2.6569,
"theoretical_loss": 3.4306709396906823,
"tokens_seen": 1998061568
},
{
"epoch": 0.61,
"learning_rate": 0.00019933801404212637,
"loss": 2.6114,
"theoretical_loss": 3.4306615559888556,
"tokens_seen": 1998127104
},
{
"epoch": 0.61,
"learning_rate": 0.00019932798395185555,
"loss": 2.7003,
"theoretical_loss": 3.4306521726809707,
"tokens_seen": 1998192640
},
{
"epoch": 0.61,
"learning_rate": 0.00019931795386158476,
"loss": 2.4545,
"theoretical_loss": 3.430642789766998,
"tokens_seen": 1998258176
},
{
"epoch": 0.61,
"learning_rate": 0.00019930792377131397,
"loss": 2.5004,
"theoretical_loss": 3.4306334072469085,
"tokens_seen": 1998323712
},
{
"epoch": 0.61,
"learning_rate": 0.00019929789368104315,
"loss": 2.804,
"theoretical_loss": 3.4306240251206717,
"tokens_seen": 1998389248
},
{
"epoch": 0.61,
"learning_rate": 0.00019928786359077233,
"loss": 2.3503,
"theoretical_loss": 3.4306146433882585,
"tokens_seen": 1998454784
},
{
"epoch": 0.61,
"learning_rate": 0.0001992778335005015,
"loss": 2.5792,
"theoretical_loss": 3.43060526204964,
"tokens_seen": 1998520320
},
{
"epoch": 0.61,
"learning_rate": 0.00019926780341023072,
"loss": 2.6819,
"theoretical_loss": 3.4305958811047863,
"tokens_seen": 1998585856
},
{
"epoch": 0.61,
"learning_rate": 0.0001992577733199599,
"loss": 2.4819,
"theoretical_loss": 3.4305865005536686,
"tokens_seen": 1998651392
},
{
"epoch": 0.61,
"learning_rate": 0.00019924774322968908,
"loss": 2.6408,
"theoretical_loss": 3.430577120396256,
"tokens_seen": 1998716928
},
{
"epoch": 0.61,
"learning_rate": 0.00019923771313941826,
"loss": 2.6853,
"theoretical_loss": 3.4305677406325206,
"tokens_seen": 1998782464
},
{
"epoch": 0.61,
"objective/train/docs_used": 1130465,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.561919689178467,
"objective/train/theoretical_loss": 3.430558361262433,
"objective/train/tokens_used": 2019308000,
"theoretical_loss": 3.430558361262433,
"tokens_seen": 1998848000
},
{
"epoch": 0.61,
"learning_rate": 0.00019922768304914744,
"loss": 2.3352,
"theoretical_loss": 3.430558361262433,
"tokens_seen": 1998848000
},
{
"epoch": 0.61,
"learning_rate": 0.00019921765295887665,
"loss": 2.5142,
"theoretical_loss": 3.430548982285962,
"tokens_seen": 1998913536
},
{
"epoch": 0.61,
"learning_rate": 0.00019920762286860583,
"loss": 2.8509,
"theoretical_loss": 3.43053960370308,
"tokens_seen": 1998979072
},
{
"epoch": 0.61,
"learning_rate": 0.00019919759277833501,
"loss": 2.656,
"theoretical_loss": 3.4305302255137566,
"tokens_seen": 1999044608
},
{
"epoch": 0.61,
"learning_rate": 0.0001991875626880642,
"loss": 2.4121,
"theoretical_loss": 3.430520847717963,
"tokens_seen": 1999110144
},
{
"epoch": 0.61,
"learning_rate": 0.0001991775325977934,
"loss": 2.51,
"theoretical_loss": 3.430511470315669,
"tokens_seen": 1999175680
},
{
"epoch": 0.61,
"learning_rate": 0.00019916750250752258,
"loss": 2.4133,
"theoretical_loss": 3.4305020933068464,
"tokens_seen": 1999241216
},
{
"epoch": 0.61,
"learning_rate": 0.00019915747241725177,
"loss": 2.4906,
"theoretical_loss": 3.4304927166914645,
"tokens_seen": 1999306752
},
{
"epoch": 0.61,
"learning_rate": 0.00019914744232698095,
"loss": 2.4412,
"theoretical_loss": 3.430483340469495,
"tokens_seen": 1999372288
},
{
"epoch": 0.61,
"learning_rate": 0.00019913741223671013,
"loss": 2.5389,
"theoretical_loss": 3.4304739646409077,
"tokens_seen": 1999437824
},
{
"epoch": 0.61,
"learning_rate": 0.00019912738214643934,
"loss": 2.4754,
"theoretical_loss": 3.4304645892056733,
"tokens_seen": 1999503360
},
{
"epoch": 0.61,
"learning_rate": 0.00019911735205616852,
"loss": 2.6572,
"theoretical_loss": 3.4304552141637625,
"tokens_seen": 1999568896
},
{
"epoch": 0.61,
"learning_rate": 0.0001991073219658977,
"loss": 2.4458,
"theoretical_loss": 3.4304458395151465,
"tokens_seen": 1999634432
},
{
"epoch": 0.61,
"learning_rate": 0.00019909729187562688,
"loss": 2.554,
"theoretical_loss": 3.430436465259795,
"tokens_seen": 1999699968
},
{
"epoch": 0.61,
"learning_rate": 0.0001990872617853561,
"loss": 2.5535,
"theoretical_loss": 3.4304270913976795,
"tokens_seen": 1999765504
},
{
"epoch": 0.61,
"learning_rate": 0.00019907723169508527,
"loss": 2.7016,
"theoretical_loss": 3.430417717928769,
"tokens_seen": 1999831040
},
{
"epoch": 0.61,
"learning_rate": 0.00019906720160481445,
"loss": 2.7387,
"theoretical_loss": 3.4304083448530367,
"tokens_seen": 1999896576
},
{
"epoch": 0.61,
"learning_rate": 0.00019905717151454363,
"loss": 2.3132,
"theoretical_loss": 3.4303989721704506,
"tokens_seen": 1999962112
},
{
"epoch": 0.61,
"learning_rate": 0.00019904714142427284,
"loss": 2.4685,
"theoretical_loss": 3.430389599880983,
"tokens_seen": 2000027648
},
{
"epoch": 0.61,
"learning_rate": 0.00019903711133400202,
"loss": 2.5416,
"theoretical_loss": 3.430380227984604,
"tokens_seen": 2000093184
},
{
"epoch": 0.61,
"learning_rate": 0.0001990270812437312,
"loss": 2.4704,
"theoretical_loss": 3.4303708564812845,
"tokens_seen": 2000158720
},
{
"epoch": 0.61,
"learning_rate": 0.00019901705115346038,
"loss": 2.6585,
"theoretical_loss": 3.4303614853709945,
"tokens_seen": 2000224256
},
{
"epoch": 0.61,
"learning_rate": 0.00019900702106318956,
"loss": 2.6106,
"theoretical_loss": 3.4303521146537053,
"tokens_seen": 2000289792
},
{
"epoch": 0.61,
"learning_rate": 0.00019899699097291877,
"loss": 2.4571,
"theoretical_loss": 3.430342744329387,
"tokens_seen": 2000355328
},
{
"epoch": 0.61,
"learning_rate": 0.00019898696088264795,
"loss": 2.6423,
"theoretical_loss": 3.430333374398011,
"tokens_seen": 2000420864
},
{
"epoch": 0.61,
"objective/train/docs_used": 1131148,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.267143726348877,
"objective/train/theoretical_loss": 3.430324004859547,
"objective/train/tokens_used": 2020946400,
"theoretical_loss": 3.430324004859547,
"tokens_seen": 2000486400
},
{
"epoch": 0.61,
"learning_rate": 0.00019897693079237713,
"loss": 2.4961,
"theoretical_loss": 3.430324004859547,
"tokens_seen": 2000486400
},
{
"epoch": 0.61,
"learning_rate": 0.00019896690070210632,
"loss": 2.4303,
"theoretical_loss": 3.430314635713967,
"tokens_seen": 2000551936
},
{
"epoch": 0.61,
"learning_rate": 0.00019895687061183552,
"loss": 2.7236,
"theoretical_loss": 3.4303052669612395,
"tokens_seen": 2000617472
},
{
"epoch": 0.61,
"learning_rate": 0.0001989468405215647,
"loss": 2.5235,
"theoretical_loss": 3.4302958986013374,
"tokens_seen": 2000683008
},
{
"epoch": 0.61,
"learning_rate": 0.00019893681043129389,
"loss": 2.4802,
"theoretical_loss": 3.43028653063423,
"tokens_seen": 2000748544
},
{
"epoch": 0.61,
"learning_rate": 0.00019892678034102307,
"loss": 2.6187,
"theoretical_loss": 3.4302771630598885,
"tokens_seen": 2000814080
},
{
"epoch": 0.61,
"learning_rate": 0.00019891675025075225,
"loss": 2.5419,
"theoretical_loss": 3.4302677958782835,
"tokens_seen": 2000879616
},
{
"epoch": 0.61,
"learning_rate": 0.00019890672016048146,
"loss": 2.6603,
"theoretical_loss": 3.4302584290893856,
"tokens_seen": 2000945152
},
{
"epoch": 0.61,
"learning_rate": 0.00019889669007021064,
"loss": 2.4083,
"theoretical_loss": 3.430249062693165,
"tokens_seen": 2001010688
},
{
"epoch": 0.61,
"learning_rate": 0.00019888665997993982,
"loss": 2.6815,
"theoretical_loss": 3.430239696689594,
"tokens_seen": 2001076224
},
{
"epoch": 0.61,
"learning_rate": 0.000198876629889669,
"loss": 2.4604,
"theoretical_loss": 3.4302303310786417,
"tokens_seen": 2001141760
},
{
"epoch": 0.61,
"learning_rate": 0.0001988665997993982,
"loss": 2.6137,
"theoretical_loss": 3.4302209658602787,
"tokens_seen": 2001207296
},
{
"epoch": 0.61,
"learning_rate": 0.0001988565697091274,
"loss": 2.4468,
"theoretical_loss": 3.4302116010344763,
"tokens_seen": 2001272832
},
{
"epoch": 0.61,
"learning_rate": 0.00019884653961885657,
"loss": 2.5314,
"theoretical_loss": 3.4302022366012057,
"tokens_seen": 2001338368
},
{
"epoch": 0.61,
"learning_rate": 0.00019883650952858575,
"loss": 2.7766,
"theoretical_loss": 3.4301928725604367,
"tokens_seen": 2001403904
},
{
"epoch": 0.61,
"learning_rate": 0.00019882647943831493,
"loss": 2.6956,
"theoretical_loss": 3.4301835089121404,
"tokens_seen": 2001469440
},
{
"epoch": 0.61,
"learning_rate": 0.00019881644934804414,
"loss": 2.4811,
"theoretical_loss": 3.430174145656287,
"tokens_seen": 2001534976
},
{
"epoch": 0.61,
"learning_rate": 0.00019880641925777332,
"loss": 2.6568,
"theoretical_loss": 3.4301647827928483,
"tokens_seen": 2001600512
},
{
"epoch": 0.61,
"learning_rate": 0.0001987963891675025,
"loss": 2.6198,
"theoretical_loss": 3.430155420321794,
"tokens_seen": 2001666048
},
{
"epoch": 0.61,
"learning_rate": 0.00019878635907723168,
"loss": 2.4891,
"theoretical_loss": 3.430146058243095,
"tokens_seen": 2001731584
},
{
"epoch": 0.61,
"learning_rate": 0.0001987763289869609,
"loss": 2.8104,
"theoretical_loss": 3.430136696556722,
"tokens_seen": 2001797120
},
{
"epoch": 0.61,
"learning_rate": 0.00019876629889669007,
"loss": 2.5711,
"theoretical_loss": 3.4301273352626462,
"tokens_seen": 2001862656
},
{
"epoch": 0.61,
"learning_rate": 0.00019875626880641925,
"loss": 2.5512,
"theoretical_loss": 3.430117974360838,
"tokens_seen": 2001928192
},
{
"epoch": 0.61,
"learning_rate": 0.00019874623871614844,
"loss": 2.3893,
"theoretical_loss": 3.4301086138512678,
"tokens_seen": 2001993728
},
{
"epoch": 0.61,
"learning_rate": 0.00019873620862587762,
"loss": 2.5464,
"theoretical_loss": 3.430099253733907,
"tokens_seen": 2002059264
},
{
"epoch": 0.61,
"objective/train/docs_used": 1132271,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5540459156036377,
"objective/train/theoretical_loss": 3.4300898940087254,
"objective/train/tokens_used": 2022584800,
"theoretical_loss": 3.4300898940087254,
"tokens_seen": 2002124800
},
{
"epoch": 0.61,
"learning_rate": 0.00019872617853560683,
"loss": 2.4961,
"theoretical_loss": 3.4300898940087254,
"tokens_seen": 2002124800
},
{
"epoch": 0.61,
"learning_rate": 0.000198716148445336,
"loss": 2.63,
"theoretical_loss": 3.430080534675695,
"tokens_seen": 2002190336
},
{
"epoch": 0.61,
"learning_rate": 0.0001987061183550652,
"loss": 2.7026,
"theoretical_loss": 3.4300711757347853,
"tokens_seen": 2002255872
},
{
"epoch": 0.61,
"learning_rate": 0.00019869608826479437,
"loss": 2.4633,
"theoretical_loss": 3.4300618171859676,
"tokens_seen": 2002321408
},
{
"epoch": 0.61,
"learning_rate": 0.00019868605817452358,
"loss": 2.5541,
"theoretical_loss": 3.4300524590292127,
"tokens_seen": 2002386944
},
{
"epoch": 0.61,
"learning_rate": 0.00019867602808425276,
"loss": 2.5957,
"theoretical_loss": 3.4300431012644914,
"tokens_seen": 2002452480
},
{
"epoch": 0.61,
"learning_rate": 0.00019866599799398194,
"loss": 2.8056,
"theoretical_loss": 3.430033743891774,
"tokens_seen": 2002518016
},
{
"epoch": 0.61,
"learning_rate": 0.00019865596790371112,
"loss": 2.564,
"theoretical_loss": 3.4300243869110316,
"tokens_seen": 2002583552
},
{
"epoch": 0.61,
"learning_rate": 0.0001986459378134403,
"loss": 2.5724,
"theoretical_loss": 3.430015030322235,
"tokens_seen": 2002649088
},
{
"epoch": 0.61,
"learning_rate": 0.0001986359077231695,
"loss": 2.5818,
"theoretical_loss": 3.4300056741253546,
"tokens_seen": 2002714624
},
{
"epoch": 0.61,
"learning_rate": 0.0001986258776328987,
"loss": 2.6691,
"theoretical_loss": 3.4299963183203617,
"tokens_seen": 2002780160
},
{
"epoch": 0.61,
"learning_rate": 0.00019861584754262787,
"loss": 2.7556,
"theoretical_loss": 3.4299869629072264,
"tokens_seen": 2002845696
},
{
"epoch": 0.61,
"learning_rate": 0.00019860581745235705,
"loss": 2.6147,
"theoretical_loss": 3.4299776078859203,
"tokens_seen": 2002911232
},
{
"epoch": 0.61,
"learning_rate": 0.00019859578736208626,
"loss": 2.6692,
"theoretical_loss": 3.4299682532564137,
"tokens_seen": 2002976768
},
{
"epoch": 0.61,
"learning_rate": 0.00019858575727181544,
"loss": 2.6154,
"theoretical_loss": 3.4299588990186773,
"tokens_seen": 2003042304
},
{
"epoch": 0.61,
"learning_rate": 0.00019857572718154462,
"loss": 2.8048,
"theoretical_loss": 3.429949545172682,
"tokens_seen": 2003107840
},
{
"epoch": 0.61,
"learning_rate": 0.00019856569709127383,
"loss": 2.6741,
"theoretical_loss": 3.429940191718398,
"tokens_seen": 2003173376
},
{
"epoch": 0.61,
"learning_rate": 0.00019855566700100304,
"loss": 2.669,
"theoretical_loss": 3.429930838655797,
"tokens_seen": 2003238912
},
{
"epoch": 0.61,
"learning_rate": 0.00019854563691073222,
"loss": 2.717,
"theoretical_loss": 3.4299214859848495,
"tokens_seen": 2003304448
},
{
"epoch": 0.61,
"learning_rate": 0.0001985356068204614,
"loss": 2.5268,
"theoretical_loss": 3.429912133705526,
"tokens_seen": 2003369984
},
{
"epoch": 0.61,
"learning_rate": 0.00019852557673019058,
"loss": 2.8259,
"theoretical_loss": 3.4299027818177974,
"tokens_seen": 2003435520
},
{
"epoch": 0.61,
"learning_rate": 0.00019851554663991976,
"loss": 2.7352,
"theoretical_loss": 3.429893430321635,
"tokens_seen": 2003501056
},
{
"epoch": 0.61,
"learning_rate": 0.00019850551654964897,
"loss": 2.6325,
"theoretical_loss": 3.4298840792170084,
"tokens_seen": 2003566592
},
{
"epoch": 0.61,
"learning_rate": 0.00019849548645937815,
"loss": 2.599,
"theoretical_loss": 3.42987472850389,
"tokens_seen": 2003632128
},
{
"epoch": 0.61,
"learning_rate": 0.00019848545636910733,
"loss": 2.4763,
"theoretical_loss": 3.4298653781822495,
"tokens_seen": 2003697664
},
{
"epoch": 0.61,
"objective/train/docs_used": 1132794,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.33974027633667,
"objective/train/theoretical_loss": 3.429856028252058,
"objective/train/tokens_used": 2024223200,
"theoretical_loss": 3.429856028252058,
"tokens_seen": 2003763200
},
{
"epoch": 0.61,
"learning_rate": 0.00019847542627883652,
"loss": 2.5125,
"theoretical_loss": 3.429856028252058,
"tokens_seen": 2003763200
},
{
"epoch": 0.61,
"learning_rate": 0.00019846539618856572,
"loss": 2.6191,
"theoretical_loss": 3.429846678713286,
"tokens_seen": 2003828736
},
{
"epoch": 0.61,
"learning_rate": 0.0001984553660982949,
"loss": 2.4281,
"theoretical_loss": 3.4298373295659053,
"tokens_seen": 2003894272
},
{
"epoch": 0.61,
"learning_rate": 0.00019844533600802409,
"loss": 2.5432,
"theoretical_loss": 3.4298279808098853,
"tokens_seen": 2003959808
},
{
"epoch": 0.61,
"learning_rate": 0.00019843530591775327,
"loss": 2.3809,
"theoretical_loss": 3.4298186324451976,
"tokens_seen": 2004025344
},
{
"epoch": 0.61,
"learning_rate": 0.00019842527582748245,
"loss": 2.7134,
"theoretical_loss": 3.4298092844718133,
"tokens_seen": 2004090880
},
{
"epoch": 0.61,
"learning_rate": 0.00019841524573721166,
"loss": 2.6994,
"theoretical_loss": 3.4297999368897027,
"tokens_seen": 2004156416
},
{
"epoch": 0.61,
"learning_rate": 0.00019840521564694084,
"loss": 2.6735,
"theoretical_loss": 3.429790589698837,
"tokens_seen": 2004221952
},
{
"epoch": 0.61,
"learning_rate": 0.00019839518555667002,
"loss": 2.5129,
"theoretical_loss": 3.4297812428991867,
"tokens_seen": 2004287488
},
{
"epoch": 0.61,
"learning_rate": 0.0001983851554663992,
"loss": 2.6858,
"theoretical_loss": 3.429771896490723,
"tokens_seen": 2004353024
},
{
"epoch": 0.61,
"learning_rate": 0.0001983751253761284,
"loss": 2.5897,
"theoretical_loss": 3.4297625504734164,
"tokens_seen": 2004418560
},
{
"epoch": 0.61,
"learning_rate": 0.0001983650952858576,
"loss": 2.693,
"theoretical_loss": 3.4297532048472377,
"tokens_seen": 2004484096
},
{
"epoch": 0.61,
"learning_rate": 0.00019835506519558677,
"loss": 2.375,
"theoretical_loss": 3.4297438596121577,
"tokens_seen": 2004549632
},
{
"epoch": 0.61,
"learning_rate": 0.00019834503510531595,
"loss": 2.3871,
"theoretical_loss": 3.429734514768148,
"tokens_seen": 2004615168
},
{
"epoch": 0.61,
"learning_rate": 0.00019833500501504513,
"loss": 2.3369,
"theoretical_loss": 3.4297251703151783,
"tokens_seen": 2004680704
},
{
"epoch": 0.61,
"learning_rate": 0.00019832497492477434,
"loss": 2.5353,
"theoretical_loss": 3.4297158262532204,
"tokens_seen": 2004746240
},
{
"epoch": 0.61,
"learning_rate": 0.00019831494483450352,
"loss": 2.6324,
"theoretical_loss": 3.4297064825822448,
"tokens_seen": 2004811776
},
{
"epoch": 0.61,
"learning_rate": 0.0001983049147442327,
"loss": 2.3622,
"theoretical_loss": 3.4296971393022226,
"tokens_seen": 2004877312
},
{
"epoch": 0.61,
"learning_rate": 0.00019829488465396188,
"loss": 2.5397,
"theoretical_loss": 3.4296877964131243,
"tokens_seen": 2004942848
},
{
"epoch": 0.61,
"learning_rate": 0.0001982848545636911,
"loss": 2.5601,
"theoretical_loss": 3.4296784539149208,
"tokens_seen": 2005008384
},
{
"epoch": 0.61,
"learning_rate": 0.00019827482447342027,
"loss": 2.8443,
"theoretical_loss": 3.4296691118075833,
"tokens_seen": 2005073920
},
{
"epoch": 0.61,
"learning_rate": 0.00019826479438314945,
"loss": 2.3661,
"theoretical_loss": 3.4296597700910825,
"tokens_seen": 2005139456
},
{
"epoch": 0.61,
"learning_rate": 0.00019825476429287864,
"loss": 2.8714,
"theoretical_loss": 3.429650428765389,
"tokens_seen": 2005204992
},
{
"epoch": 0.61,
"learning_rate": 0.00019824473420260782,
"loss": 2.705,
"theoretical_loss": 3.4296410878304737,
"tokens_seen": 2005270528
},
{
"epoch": 0.61,
"learning_rate": 0.00019823470411233703,
"loss": 2.725,
"theoretical_loss": 3.429631747286308,
"tokens_seen": 2005336064
},
{
"epoch": 0.61,
"objective/train/docs_used": 1134045,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3661653995513916,
"objective/train/theoretical_loss": 3.4296224071328623,
"objective/train/tokens_used": 2025861600,
"theoretical_loss": 3.4296224071328623,
"tokens_seen": 2005401600
},
{
"epoch": 0.61,
"learning_rate": 0.0001982246740220662,
"loss": 2.5734,
"theoretical_loss": 3.4296224071328623,
"tokens_seen": 2005401600
},
{
"epoch": 0.61,
"learning_rate": 0.0001982146439317954,
"loss": 2.5709,
"theoretical_loss": 3.4296130673701075,
"tokens_seen": 2005467136
},
{
"epoch": 0.61,
"learning_rate": 0.00019820461384152457,
"loss": 2.7964,
"theoretical_loss": 3.429603727998015,
"tokens_seen": 2005532672
},
{
"epoch": 0.61,
"learning_rate": 0.00019819458375125378,
"loss": 2.6307,
"theoretical_loss": 3.429594389016555,
"tokens_seen": 2005598208
},
{
"epoch": 0.61,
"learning_rate": 0.00019818455366098296,
"loss": 2.5209,
"theoretical_loss": 3.429585050425699,
"tokens_seen": 2005663744
},
{
"epoch": 0.61,
"learning_rate": 0.00019817452357071214,
"loss": 2.4889,
"theoretical_loss": 3.429575712225418,
"tokens_seen": 2005729280
},
{
"epoch": 0.61,
"learning_rate": 0.00019816449348044132,
"loss": 2.5012,
"theoretical_loss": 3.429566374415682,
"tokens_seen": 2005794816
},
{
"epoch": 0.61,
"learning_rate": 0.0001981544633901705,
"loss": 2.6098,
"theoretical_loss": 3.4295570369964623,
"tokens_seen": 2005860352
},
{
"epoch": 0.61,
"learning_rate": 0.0001981444332998997,
"loss": 2.5598,
"theoretical_loss": 3.4295476999677303,
"tokens_seen": 2005925888
},
{
"epoch": 0.61,
"learning_rate": 0.0001981344032096289,
"loss": 2.5943,
"theoretical_loss": 3.4295383633294563,
"tokens_seen": 2005991424
},
{
"epoch": 0.61,
"learning_rate": 0.00019812437311935807,
"loss": 2.8159,
"theoretical_loss": 3.429529027081612,
"tokens_seen": 2006056960
},
{
"epoch": 0.61,
"learning_rate": 0.00019811434302908725,
"loss": 2.6984,
"theoretical_loss": 3.4295196912241677,
"tokens_seen": 2006122496
},
{
"epoch": 0.61,
"learning_rate": 0.00019810431293881646,
"loss": 2.6025,
"theoretical_loss": 3.429510355757094,
"tokens_seen": 2006188032
},
{
"epoch": 0.61,
"learning_rate": 0.00019809428284854564,
"loss": 2.6786,
"theoretical_loss": 3.4295010206803624,
"tokens_seen": 2006253568
},
{
"epoch": 0.61,
"learning_rate": 0.00019808425275827482,
"loss": 2.4728,
"theoretical_loss": 3.429491685993944,
"tokens_seen": 2006319104
},
{
"epoch": 0.61,
"learning_rate": 0.000198074222668004,
"loss": 2.4968,
"theoretical_loss": 3.4294823516978092,
"tokens_seen": 2006384640
},
{
"epoch": 0.61,
"learning_rate": 0.00019806419257773319,
"loss": 2.5768,
"theoretical_loss": 3.4294730177919295,
"tokens_seen": 2006450176
},
{
"epoch": 0.61,
"learning_rate": 0.0001980541624874624,
"loss": 2.6674,
"theoretical_loss": 3.429463684276275,
"tokens_seen": 2006515712
},
{
"epoch": 0.61,
"learning_rate": 0.00019804413239719157,
"loss": 2.5493,
"theoretical_loss": 3.429454351150817,
"tokens_seen": 2006581248
},
{
"epoch": 0.61,
"learning_rate": 0.00019803410230692076,
"loss": 2.4649,
"theoretical_loss": 3.429445018415527,
"tokens_seen": 2006646784
},
{
"epoch": 0.61,
"learning_rate": 0.00019802407221664994,
"loss": 2.6005,
"theoretical_loss": 3.4294356860703754,
"tokens_seen": 2006712320
},
{
"epoch": 0.61,
"learning_rate": 0.00019801404212637915,
"loss": 2.4563,
"theoretical_loss": 3.429426354115334,
"tokens_seen": 2006777856
},
{
"epoch": 0.61,
"learning_rate": 0.00019800401203610833,
"loss": 2.568,
"theoretical_loss": 3.4294170225503717,
"tokens_seen": 2006843392
},
{
"epoch": 0.61,
"learning_rate": 0.0001979939819458375,
"loss": 2.9098,
"theoretical_loss": 3.429407691375461,
"tokens_seen": 2006908928
},
{
"epoch": 0.61,
"learning_rate": 0.0001979839518555667,
"loss": 2.4482,
"theoretical_loss": 3.4293983605905733,
"tokens_seen": 2006974464
},
{
"epoch": 0.61,
"objective/train/docs_used": 1134700,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8493223190307617,
"objective/train/theoretical_loss": 3.429389030195679,
"objective/train/tokens_used": 2027500000,
"theoretical_loss": 3.429389030195679,
"tokens_seen": 2007040000
},
{
"epoch": 0.61,
"learning_rate": 0.0001979739217652959,
"loss": 2.6135,
"theoretical_loss": 3.429389030195679,
"tokens_seen": 2007040000
},
{
"epoch": 0.61,
"learning_rate": 0.00019796389167502508,
"loss": 2.4343,
"theoretical_loss": 3.4293797001907484,
"tokens_seen": 2007105536
},
{
"epoch": 0.61,
"learning_rate": 0.00019795386158475426,
"loss": 2.2752,
"theoretical_loss": 3.4293703705757537,
"tokens_seen": 2007171072
},
{
"epoch": 0.61,
"learning_rate": 0.00019794383149448344,
"loss": 2.3667,
"theoretical_loss": 3.4293610413506643,
"tokens_seen": 2007236608
},
{
"epoch": 0.61,
"learning_rate": 0.00019793380140421262,
"loss": 2.7261,
"theoretical_loss": 3.4293517125154525,
"tokens_seen": 2007302144
},
{
"epoch": 0.61,
"learning_rate": 0.00019792377131394183,
"loss": 2.5883,
"theoretical_loss": 3.429342384070089,
"tokens_seen": 2007367680
},
{
"epoch": 0.61,
"learning_rate": 0.000197913741223671,
"loss": 2.5429,
"theoretical_loss": 3.4293330560145447,
"tokens_seen": 2007433216
},
{
"epoch": 0.61,
"learning_rate": 0.0001979037111334002,
"loss": 2.4867,
"theoretical_loss": 3.4293237283487903,
"tokens_seen": 2007498752
},
{
"epoch": 0.61,
"learning_rate": 0.00019789368104312937,
"loss": 2.7255,
"theoretical_loss": 3.4293144010727974,
"tokens_seen": 2007564288
},
{
"epoch": 0.61,
"learning_rate": 0.00019788365095285858,
"loss": 2.5245,
"theoretical_loss": 3.4293050741865363,
"tokens_seen": 2007629824
},
{
"epoch": 0.61,
"learning_rate": 0.00019787362086258776,
"loss": 2.3925,
"theoretical_loss": 3.429295747689978,
"tokens_seen": 2007695360
},
{
"epoch": 0.61,
"learning_rate": 0.00019786359077231694,
"loss": 2.7668,
"theoretical_loss": 3.4292864215830945,
"tokens_seen": 2007760896
},
{
"epoch": 0.61,
"learning_rate": 0.00019785356068204612,
"loss": 2.5238,
"theoretical_loss": 3.4292770958658556,
"tokens_seen": 2007826432
},
{
"epoch": 0.61,
"learning_rate": 0.0001978435305917753,
"loss": 2.4242,
"theoretical_loss": 3.4292677705382335,
"tokens_seen": 2007891968
},
{
"epoch": 0.61,
"learning_rate": 0.00019783350050150451,
"loss": 2.6756,
"theoretical_loss": 3.4292584456001975,
"tokens_seen": 2007957504
},
{
"epoch": 0.61,
"learning_rate": 0.0001978234704112337,
"loss": 2.5051,
"theoretical_loss": 3.42924912105172,
"tokens_seen": 2008023040
},
{
"epoch": 0.61,
"learning_rate": 0.0001978134403209629,
"loss": 2.9158,
"theoretical_loss": 3.429239796892772,
"tokens_seen": 2008088576
},
{
"epoch": 0.61,
"learning_rate": 0.00019780341023069208,
"loss": 2.5502,
"theoretical_loss": 3.429230473123324,
"tokens_seen": 2008154112
},
{
"epoch": 0.61,
"learning_rate": 0.0001977933801404213,
"loss": 2.321,
"theoretical_loss": 3.429221149743347,
"tokens_seen": 2008219648
},
{
"epoch": 0.61,
"learning_rate": 0.00019778335005015047,
"loss": 2.5222,
"theoretical_loss": 3.4292118267528124,
"tokens_seen": 2008285184
},
{
"epoch": 0.61,
"learning_rate": 0.00019777331995987965,
"loss": 2.827,
"theoretical_loss": 3.429202504151691,
"tokens_seen": 2008350720
},
{
"epoch": 0.61,
"learning_rate": 0.00019776328986960884,
"loss": 2.7021,
"theoretical_loss": 3.4291931819399535,
"tokens_seen": 2008416256
},
{
"epoch": 0.61,
"learning_rate": 0.00019775325977933802,
"loss": 2.626,
"theoretical_loss": 3.429183860117572,
"tokens_seen": 2008481792
},
{
"epoch": 0.61,
"learning_rate": 0.00019774322968906723,
"loss": 2.5718,
"theoretical_loss": 3.4291745386845163,
"tokens_seen": 2008547328
},
{
"epoch": 0.61,
"learning_rate": 0.0001977331995987964,
"loss": 2.5465,
"theoretical_loss": 3.429165217640758,
"tokens_seen": 2008612864
},
{
"epoch": 0.61,
"objective/train/docs_used": 1135755,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.757493257522583,
"objective/train/theoretical_loss": 3.429155896986268,
"objective/train/tokens_used": 2029138400,
"theoretical_loss": 3.429155896986268,
"tokens_seen": 2008678400
},
{
"epoch": 0.61,
"learning_rate": 0.0001977231695085256,
"loss": 2.5722,
"theoretical_loss": 3.429155896986268,
"tokens_seen": 2008678400
},
{
"epoch": 0.61,
"learning_rate": 0.00019771313941825477,
"loss": 2.7177,
"theoretical_loss": 3.4291465767210174,
"tokens_seen": 2008743936
},
{
"epoch": 0.61,
"learning_rate": 0.00019770310932798398,
"loss": 2.5953,
"theoretical_loss": 3.429137256844977,
"tokens_seen": 2008809472
},
{
"epoch": 0.61,
"learning_rate": 0.00019769307923771316,
"loss": 2.7352,
"theoretical_loss": 3.429127937358119,
"tokens_seen": 2008875008
},
{
"epoch": 0.61,
"learning_rate": 0.00019768304914744234,
"loss": 2.4409,
"theoretical_loss": 3.429118618260413,
"tokens_seen": 2008940544
},
{
"epoch": 0.61,
"learning_rate": 0.00019767301905717152,
"loss": 2.4634,
"theoretical_loss": 3.429109299551831,
"tokens_seen": 2009006080
},
{
"epoch": 0.61,
"learning_rate": 0.0001976629889669007,
"loss": 2.7417,
"theoretical_loss": 3.4290999812323433,
"tokens_seen": 2009071616
},
{
"epoch": 0.61,
"learning_rate": 0.0001976529588766299,
"loss": 2.4664,
"theoretical_loss": 3.4290906633019214,
"tokens_seen": 2009137152
},
{
"epoch": 0.61,
"learning_rate": 0.0001976429287863591,
"loss": 2.696,
"theoretical_loss": 3.4290813457605362,
"tokens_seen": 2009202688
},
{
"epoch": 0.61,
"learning_rate": 0.00019763289869608827,
"loss": 2.3693,
"theoretical_loss": 3.429072028608159,
"tokens_seen": 2009268224
},
{
"epoch": 0.61,
"learning_rate": 0.00019762286860581745,
"loss": 2.8273,
"theoretical_loss": 3.429062711844761,
"tokens_seen": 2009333760
},
{
"epoch": 0.61,
"learning_rate": 0.00019761283851554666,
"loss": 2.7025,
"theoretical_loss": 3.429053395470313,
"tokens_seen": 2009399296
},
{
"epoch": 0.61,
"learning_rate": 0.00019760280842527584,
"loss": 2.3127,
"theoretical_loss": 3.429044079484786,
"tokens_seen": 2009464832
},
{
"epoch": 0.61,
"learning_rate": 0.00019759277833500502,
"loss": 2.6539,
"theoretical_loss": 3.4290347638881515,
"tokens_seen": 2009530368
},
{
"epoch": 0.61,
"learning_rate": 0.0001975827482447342,
"loss": 2.5861,
"theoretical_loss": 3.42902544868038,
"tokens_seen": 2009595904
},
{
"epoch": 0.61,
"learning_rate": 0.00019757271815446339,
"loss": 2.6059,
"theoretical_loss": 3.4290161338614427,
"tokens_seen": 2009661440
},
{
"epoch": 0.61,
"learning_rate": 0.0001975626880641926,
"loss": 2.7206,
"theoretical_loss": 3.4290068194313115,
"tokens_seen": 2009726976
},
{
"epoch": 0.61,
"learning_rate": 0.00019755265797392178,
"loss": 2.5691,
"theoretical_loss": 3.428997505389956,
"tokens_seen": 2009792512
},
{
"epoch": 0.61,
"learning_rate": 0.00019754262788365096,
"loss": 2.6318,
"theoretical_loss": 3.4289881917373486,
"tokens_seen": 2009858048
},
{
"epoch": 0.61,
"learning_rate": 0.00019753259779338014,
"loss": 2.6218,
"theoretical_loss": 3.42897887847346,
"tokens_seen": 2009923584
},
{
"epoch": 0.61,
"learning_rate": 0.00019752256770310935,
"loss": 2.9213,
"theoretical_loss": 3.428969565598261,
"tokens_seen": 2009989120
},
{
"epoch": 0.61,
"learning_rate": 0.00019751253761283853,
"loss": 2.6323,
"theoretical_loss": 3.428960253111723,
"tokens_seen": 2010054656
},
{
"epoch": 0.61,
"learning_rate": 0.0001975025075225677,
"loss": 2.6263,
"theoretical_loss": 3.4289509410138175,
"tokens_seen": 2010120192
},
{
"epoch": 0.61,
"learning_rate": 0.0001974924774322969,
"loss": 2.7049,
"theoretical_loss": 3.4289416293045147,
"tokens_seen": 2010185728
},
{
"epoch": 0.61,
"learning_rate": 0.0001974824473420261,
"loss": 2.5061,
"theoretical_loss": 3.4289323179837865,
"tokens_seen": 2010251264
},
{
"epoch": 0.61,
"objective/train/docs_used": 1136321,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.628479480743408,
"objective/train/theoretical_loss": 3.4289230070516035,
"objective/train/tokens_used": 2030776800,
"theoretical_loss": 3.4289230070516035,
"tokens_seen": 2010316800
},
{
"epoch": 0.61,
"learning_rate": 0.00019747241725175528,
"loss": 2.7715,
"theoretical_loss": 3.4289230070516035,
"tokens_seen": 2010316800
},
{
"epoch": 0.61,
"learning_rate": 0.00019746238716148446,
"loss": 2.4945,
"theoretical_loss": 3.428913696507937,
"tokens_seen": 2010382336
},
{
"epoch": 0.61,
"learning_rate": 0.00019745235707121364,
"loss": 2.6173,
"theoretical_loss": 3.428904386352758,
"tokens_seen": 2010447872
},
{
"epoch": 0.61,
"learning_rate": 0.00019744232698094282,
"loss": 2.7806,
"theoretical_loss": 3.428895076586038,
"tokens_seen": 2010513408
},
{
"epoch": 0.61,
"learning_rate": 0.00019743229689067203,
"loss": 2.866,
"theoretical_loss": 3.428885767207748,
"tokens_seen": 2010578944
},
{
"epoch": 0.61,
"learning_rate": 0.0001974222668004012,
"loss": 2.6775,
"theoretical_loss": 3.428876458217859,
"tokens_seen": 2010644480
},
{
"epoch": 0.61,
"learning_rate": 0.0001974122367101304,
"loss": 2.7411,
"theoretical_loss": 3.4288671496163423,
"tokens_seen": 2010710016
},
{
"epoch": 0.61,
"learning_rate": 0.00019740220661985957,
"loss": 2.4813,
"theoretical_loss": 3.4288578414031687,
"tokens_seen": 2010775552
},
{
"epoch": 0.61,
"learning_rate": 0.00019739217652958878,
"loss": 2.6416,
"theoretical_loss": 3.42884853357831,
"tokens_seen": 2010841088
},
{
"epoch": 0.61,
"learning_rate": 0.00019738214643931796,
"loss": 2.5594,
"theoretical_loss": 3.4288392261417364,
"tokens_seen": 2010906624
},
{
"epoch": 0.61,
"learning_rate": 0.00019737211634904714,
"loss": 2.5906,
"theoretical_loss": 3.4288299190934195,
"tokens_seen": 2010972160
},
{
"epoch": 0.61,
"learning_rate": 0.00019736208625877632,
"loss": 2.6935,
"theoretical_loss": 3.4288206124333307,
"tokens_seen": 2011037696
},
{
"epoch": 0.61,
"learning_rate": 0.0001973520561685055,
"loss": 2.5625,
"theoretical_loss": 3.4288113061614407,
"tokens_seen": 2011103232
},
{
"epoch": 0.61,
"learning_rate": 0.00019734202607823471,
"loss": 2.7808,
"theoretical_loss": 3.4288020002777215,
"tokens_seen": 2011168768
},
{
"epoch": 0.61,
"learning_rate": 0.0001973319959879639,
"loss": 2.581,
"theoretical_loss": 3.4287926947821434,
"tokens_seen": 2011234304
},
{
"epoch": 0.61,
"learning_rate": 0.00019732196589769308,
"loss": 2.8481,
"theoretical_loss": 3.4287833896746776,
"tokens_seen": 2011299840
},
{
"epoch": 0.61,
"learning_rate": 0.00019731193580742226,
"loss": 2.7461,
"theoretical_loss": 3.428774084955296,
"tokens_seen": 2011365376
},
{
"epoch": 0.61,
"learning_rate": 0.00019730190571715147,
"loss": 2.8201,
"theoretical_loss": 3.428764780623969,
"tokens_seen": 2011430912
},
{
"epoch": 0.61,
"learning_rate": 0.00019729187562688065,
"loss": 2.6489,
"theoretical_loss": 3.428755476680668,
"tokens_seen": 2011496448
},
{
"epoch": 0.61,
"learning_rate": 0.00019728184553660983,
"loss": 2.7491,
"theoretical_loss": 3.428746173125364,
"tokens_seen": 2011561984
},
{
"epoch": 0.61,
"learning_rate": 0.000197271815446339,
"loss": 2.5158,
"theoretical_loss": 3.428736869958029,
"tokens_seen": 2011627520
},
{
"epoch": 0.61,
"learning_rate": 0.0001972617853560682,
"loss": 2.4838,
"theoretical_loss": 3.4287275671786333,
"tokens_seen": 2011693056
},
{
"epoch": 0.61,
"learning_rate": 0.0001972517552657974,
"loss": 2.8106,
"theoretical_loss": 3.4287182647871486,
"tokens_seen": 2011758592
},
{
"epoch": 0.61,
"learning_rate": 0.00019724172517552658,
"loss": 2.7368,
"theoretical_loss": 3.4287089627835456,
"tokens_seen": 2011824128
},
{
"epoch": 0.61,
"learning_rate": 0.00019723169508525576,
"loss": 2.7102,
"theoretical_loss": 3.428699661167796,
"tokens_seen": 2011889664
},
{
"epoch": 0.61,
"objective/train/docs_used": 1137326,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.778999090194702,
"objective/train/theoretical_loss": 3.4286903599398704,
"objective/train/tokens_used": 2032415200,
"theoretical_loss": 3.4286903599398704,
"tokens_seen": 2011955200
},
{
"epoch": 0.61,
"learning_rate": 0.00019722166499498494,
"loss": 2.5695,
"theoretical_loss": 3.4286903599398704,
"tokens_seen": 2011955200
},
{
"epoch": 0.61,
"learning_rate": 0.00019721163490471415,
"loss": 2.7374,
"theoretical_loss": 3.4286810590997403,
"tokens_seen": 2012020736
},
{
"epoch": 0.61,
"learning_rate": 0.00019720160481444333,
"loss": 2.7258,
"theoretical_loss": 3.4286717586473774,
"tokens_seen": 2012086272
},
{
"epoch": 0.61,
"learning_rate": 0.0001971915747241725,
"loss": 2.6143,
"theoretical_loss": 3.4286624585827523,
"tokens_seen": 2012151808
},
{
"epoch": 0.61,
"learning_rate": 0.0001971815446339017,
"loss": 2.7693,
"theoretical_loss": 3.428653158905836,
"tokens_seen": 2012217344
},
{
"epoch": 0.61,
"learning_rate": 0.00019717151454363087,
"loss": 2.7554,
"theoretical_loss": 3.4286438596166007,
"tokens_seen": 2012282880
},
{
"epoch": 0.61,
"learning_rate": 0.00019716148445336008,
"loss": 2.9142,
"theoretical_loss": 3.428634560715017,
"tokens_seen": 2012348416
},
{
"epoch": 0.61,
"learning_rate": 0.00019715145436308926,
"loss": 2.527,
"theoretical_loss": 3.428625262201056,
"tokens_seen": 2012413952
},
{
"epoch": 0.61,
"learning_rate": 0.00019714142427281844,
"loss": 2.7906,
"theoretical_loss": 3.4286159640746883,
"tokens_seen": 2012479488
},
{
"epoch": 0.61,
"learning_rate": 0.00019713139418254763,
"loss": 2.701,
"theoretical_loss": 3.4286066663358863,
"tokens_seen": 2012545024
},
{
"epoch": 0.61,
"learning_rate": 0.00019712136409227683,
"loss": 2.9224,
"theoretical_loss": 3.428597368984621,
"tokens_seen": 2012610560
},
{
"epoch": 0.61,
"learning_rate": 0.00019711133400200602,
"loss": 2.6177,
"theoretical_loss": 3.4285880720208635,
"tokens_seen": 2012676096
},
{
"epoch": 0.61,
"learning_rate": 0.0001971013039117352,
"loss": 2.5838,
"theoretical_loss": 3.428578775444585,
"tokens_seen": 2012741632
},
{
"epoch": 0.61,
"learning_rate": 0.00019709127382146438,
"loss": 2.5428,
"theoretical_loss": 3.428569479255756,
"tokens_seen": 2012807168
},
{
"epoch": 0.61,
"learning_rate": 0.00019708124373119356,
"loss": 2.943,
"theoretical_loss": 3.4285601834543487,
"tokens_seen": 2012872704
},
{
"epoch": 0.61,
"learning_rate": 0.0001970712136409228,
"loss": 2.7838,
"theoretical_loss": 3.428550888040334,
"tokens_seen": 2012938240
},
{
"epoch": 0.61,
"learning_rate": 0.00019706118355065198,
"loss": 2.637,
"theoretical_loss": 3.4285415930136836,
"tokens_seen": 2013003776
},
{
"epoch": 0.61,
"learning_rate": 0.00019705115346038116,
"loss": 2.6378,
"theoretical_loss": 3.4285322983743676,
"tokens_seen": 2013069312
},
{
"epoch": 0.61,
"learning_rate": 0.00019704112337011034,
"loss": 2.6403,
"theoretical_loss": 3.4285230041223587,
"tokens_seen": 2013134848
},
{
"epoch": 0.61,
"learning_rate": 0.00019703109327983955,
"loss": 2.6885,
"theoretical_loss": 3.4285137102576275,
"tokens_seen": 2013200384
},
{
"epoch": 0.61,
"learning_rate": 0.00019702106318956873,
"loss": 2.5325,
"theoretical_loss": 3.428504416780145,
"tokens_seen": 2013265920
},
{
"epoch": 0.61,
"learning_rate": 0.0001970110330992979,
"loss": 2.4706,
"theoretical_loss": 3.428495123689882,
"tokens_seen": 2013331456
},
{
"epoch": 0.61,
"learning_rate": 0.0001970010030090271,
"loss": 2.8818,
"theoretical_loss": 3.428485830986811,
"tokens_seen": 2013396992
},
{
"epoch": 0.61,
"learning_rate": 0.0001969909729187563,
"loss": 2.758,
"theoretical_loss": 3.4284765386709033,
"tokens_seen": 2013462528
},
{
"epoch": 0.61,
"learning_rate": 0.00019698094282848548,
"loss": 2.7328,
"theoretical_loss": 3.4284672467421284,
"tokens_seen": 2013528064
},
{
"epoch": 0.61,
"objective/train/docs_used": 1138031,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.748216152191162,
"objective/train/theoretical_loss": 3.4284579552004595,
"objective/train/tokens_used": 2034053600,
"theoretical_loss": 3.4284579552004595,
"tokens_seen": 2013593600
},
{
"epoch": 0.61,
"learning_rate": 0.00019697091273821466,
"loss": 2.7598,
"theoretical_loss": 3.4284579552004595,
"tokens_seen": 2013593600
},
{
"epoch": 0.61,
"learning_rate": 0.00019696088264794384,
"loss": 2.5884,
"theoretical_loss": 3.4284486640458667,
"tokens_seen": 2013659136
},
{
"epoch": 0.61,
"learning_rate": 0.00019695085255767302,
"loss": 2.7709,
"theoretical_loss": 3.428439373278322,
"tokens_seen": 2013724672
},
{
"epoch": 0.61,
"learning_rate": 0.00019694082246740223,
"loss": 2.6651,
"theoretical_loss": 3.4284300828977967,
"tokens_seen": 2013790208
},
{
"epoch": 0.61,
"learning_rate": 0.0001969307923771314,
"loss": 2.713,
"theoretical_loss": 3.428420792904261,
"tokens_seen": 2013855744
},
{
"epoch": 0.61,
"learning_rate": 0.0001969207622868606,
"loss": 2.6154,
"theoretical_loss": 3.428411503297687,
"tokens_seen": 2013921280
},
{
"epoch": 0.61,
"learning_rate": 0.00019691073219658977,
"loss": 2.7477,
"theoretical_loss": 3.428402214078046,
"tokens_seen": 2013986816
},
{
"epoch": 0.61,
"learning_rate": 0.00019690070210631898,
"loss": 2.6804,
"theoretical_loss": 3.4283929252453094,
"tokens_seen": 2014052352
},
{
"epoch": 0.61,
"learning_rate": 0.00019689067201604816,
"loss": 2.7215,
"theoretical_loss": 3.4283836367994485,
"tokens_seen": 2014117888
},
{
"epoch": 0.61,
"learning_rate": 0.00019688064192577734,
"loss": 2.5362,
"theoretical_loss": 3.428374348740434,
"tokens_seen": 2014183424
},
{
"epoch": 0.61,
"learning_rate": 0.00019687061183550652,
"loss": 2.9546,
"theoretical_loss": 3.428365061068238,
"tokens_seen": 2014248960
},
{
"epoch": 0.61,
"learning_rate": 0.0001968605817452357,
"loss": 2.447,
"theoretical_loss": 3.4283557737828314,
"tokens_seen": 2014314496
},
{
"epoch": 0.61,
"learning_rate": 0.00019685055165496491,
"loss": 2.5441,
"theoretical_loss": 3.4283464868841853,
"tokens_seen": 2014380032
},
{
"epoch": 0.61,
"learning_rate": 0.0001968405215646941,
"loss": 2.6947,
"theoretical_loss": 3.428337200372271,
"tokens_seen": 2014445568
},
{
"epoch": 0.61,
"learning_rate": 0.00019683049147442328,
"loss": 2.6512,
"theoretical_loss": 3.4283279142470606,
"tokens_seen": 2014511104
},
{
"epoch": 0.61,
"learning_rate": 0.00019682046138415246,
"loss": 2.7395,
"theoretical_loss": 3.4283186285085243,
"tokens_seen": 2014576640
},
{
"epoch": 0.61,
"learning_rate": 0.00019681043129388167,
"loss": 2.5896,
"theoretical_loss": 3.4283093431566343,
"tokens_seen": 2014642176
},
{
"epoch": 0.61,
"learning_rate": 0.00019680040120361085,
"loss": 2.8238,
"theoretical_loss": 3.428300058191362,
"tokens_seen": 2014707712
},
{
"epoch": 0.61,
"learning_rate": 0.00019679037111334003,
"loss": 2.5994,
"theoretical_loss": 3.428290773612678,
"tokens_seen": 2014773248
},
{
"epoch": 0.61,
"learning_rate": 0.0001967803410230692,
"loss": 2.6641,
"theoretical_loss": 3.428281489420554,
"tokens_seen": 2014838784
},
{
"epoch": 0.61,
"learning_rate": 0.0001967703109327984,
"loss": 2.5539,
"theoretical_loss": 3.4282722056149613,
"tokens_seen": 2014904320
},
{
"epoch": 0.61,
"learning_rate": 0.0001967602808425276,
"loss": 2.7556,
"theoretical_loss": 3.428262922195871,
"tokens_seen": 2014969856
},
{
"epoch": 0.61,
"learning_rate": 0.00019675025075225678,
"loss": 2.9302,
"theoretical_loss": 3.428253639163255,
"tokens_seen": 2015035392
},
{
"epoch": 0.61,
"learning_rate": 0.00019674022066198596,
"loss": 2.8124,
"theoretical_loss": 3.428244356517084,
"tokens_seen": 2015100928
},
{
"epoch": 0.61,
"learning_rate": 0.00019673019057171514,
"loss": 2.7121,
"theoretical_loss": 3.42823507425733,
"tokens_seen": 2015166464
},
{
"epoch": 0.61,
"objective/train/docs_used": 1139429,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9175148010253906,
"objective/train/theoretical_loss": 3.428225792383964,
"objective/train/tokens_used": 2035692000,
"theoretical_loss": 3.428225792383964,
"tokens_seen": 2015232000
},
{
"epoch": 0.61,
"learning_rate": 0.00019672016048144435,
"loss": 2.9722,
"theoretical_loss": 3.428225792383964,
"tokens_seen": 2015232000
},
{
"epoch": 0.61,
"learning_rate": 0.00019671013039117353,
"loss": 2.642,
"theoretical_loss": 3.4282165108969576,
"tokens_seen": 2015297536
},
{
"epoch": 0.61,
"learning_rate": 0.0001967001003009027,
"loss": 2.7426,
"theoretical_loss": 3.428207229796281,
"tokens_seen": 2015363072
},
{
"epoch": 0.61,
"learning_rate": 0.0001966900702106319,
"loss": 2.722,
"theoretical_loss": 3.4281979490819072,
"tokens_seen": 2015428608
},
{
"epoch": 0.61,
"learning_rate": 0.00019668004012036107,
"loss": 2.5946,
"theoretical_loss": 3.4281886687538066,
"tokens_seen": 2015494144
},
{
"epoch": 0.61,
"learning_rate": 0.00019667001003009028,
"loss": 2.4678,
"theoretical_loss": 3.428179388811951,
"tokens_seen": 2015559680
},
{
"epoch": 0.61,
"learning_rate": 0.00019665997993981946,
"loss": 2.6947,
"theoretical_loss": 3.4281701092563117,
"tokens_seen": 2015625216
},
{
"epoch": 0.61,
"learning_rate": 0.00019664994984954865,
"loss": 2.5225,
"theoretical_loss": 3.42816083008686,
"tokens_seen": 2015690752
},
{
"epoch": 0.61,
"learning_rate": 0.00019663991975927783,
"loss": 2.8564,
"theoretical_loss": 3.428151551303567,
"tokens_seen": 2015756288
},
{
"epoch": 0.61,
"learning_rate": 0.00019662988966900703,
"loss": 2.751,
"theoretical_loss": 3.428142272906404,
"tokens_seen": 2015821824
},
{
"epoch": 0.61,
"learning_rate": 0.00019661985957873622,
"loss": 2.7467,
"theoretical_loss": 3.4281329948953427,
"tokens_seen": 2015887360
},
{
"epoch": 0.61,
"learning_rate": 0.0001966098294884654,
"loss": 2.7189,
"theoretical_loss": 3.428123717270355,
"tokens_seen": 2015952896
},
{
"epoch": 0.61,
"learning_rate": 0.00019659979939819458,
"loss": 2.5674,
"theoretical_loss": 3.428114440031411,
"tokens_seen": 2016018432
},
{
"epoch": 0.61,
"learning_rate": 0.00019658976930792376,
"loss": 2.6599,
"theoretical_loss": 3.4281051631784836,
"tokens_seen": 2016083968
},
{
"epoch": 0.61,
"learning_rate": 0.00019657973921765297,
"loss": 2.6479,
"theoretical_loss": 3.428095886711543,
"tokens_seen": 2016149504
},
{
"epoch": 0.61,
"learning_rate": 0.00019656970912738215,
"loss": 2.7257,
"theoretical_loss": 3.4280866106305607,
"tokens_seen": 2016215040
},
{
"epoch": 0.61,
"learning_rate": 0.00019655967903711133,
"loss": 2.6233,
"theoretical_loss": 3.4280773349355087,
"tokens_seen": 2016280576
},
{
"epoch": 0.61,
"learning_rate": 0.0001965496489468405,
"loss": 2.4712,
"theoretical_loss": 3.4280680596263586,
"tokens_seen": 2016346112
},
{
"epoch": 0.61,
"learning_rate": 0.00019653961885656972,
"loss": 2.7224,
"theoretical_loss": 3.4280587847030812,
"tokens_seen": 2016411648
},
{
"epoch": 0.61,
"learning_rate": 0.0001965295887662989,
"loss": 2.6064,
"theoretical_loss": 3.428049510165647,
"tokens_seen": 2016477184
},
{
"epoch": 0.61,
"learning_rate": 0.00019651955867602808,
"loss": 2.5133,
"theoretical_loss": 3.4280402360140294,
"tokens_seen": 2016542720
},
{
"epoch": 0.61,
"learning_rate": 0.00019650952858575726,
"loss": 2.5432,
"theoretical_loss": 3.4280309622481986,
"tokens_seen": 2016608256
},
{
"epoch": 0.61,
"learning_rate": 0.00019649949849548644,
"loss": 2.5954,
"theoretical_loss": 3.4280216888681263,
"tokens_seen": 2016673792
},
{
"epoch": 0.61,
"learning_rate": 0.00019648946840521565,
"loss": 2.8129,
"theoretical_loss": 3.4280124158737837,
"tokens_seen": 2016739328
},
{
"epoch": 0.61,
"learning_rate": 0.00019647943831494483,
"loss": 2.8902,
"theoretical_loss": 3.4280031432651423,
"tokens_seen": 2016804864
},
{
"epoch": 0.61,
"objective/train/docs_used": 1140637,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4223930835723877,
"objective/train/theoretical_loss": 3.4279938710421742,
"objective/train/tokens_used": 2037330400,
"theoretical_loss": 3.4279938710421742,
"tokens_seen": 2016870400
},
{
"epoch": 0.61,
"learning_rate": 0.00019646940822467401,
"loss": 2.5043,
"theoretical_loss": 3.4279938710421742,
"tokens_seen": 2016870400
},
{
"epoch": 0.61,
"learning_rate": 0.0001964593781344032,
"loss": 2.5791,
"theoretical_loss": 3.4279845992048497,
"tokens_seen": 2016935936
},
{
"epoch": 0.61,
"learning_rate": 0.0001964493480441324,
"loss": 2.708,
"theoretical_loss": 3.427975327753141,
"tokens_seen": 2017001472
},
{
"epoch": 0.61,
"learning_rate": 0.00019643931795386158,
"loss": 2.8594,
"theoretical_loss": 3.427966056687019,
"tokens_seen": 2017067008
},
{
"epoch": 0.61,
"learning_rate": 0.00019642928786359077,
"loss": 2.8172,
"theoretical_loss": 3.427956786006456,
"tokens_seen": 2017132544
},
{
"epoch": 0.61,
"learning_rate": 0.00019641925777331995,
"loss": 2.6388,
"theoretical_loss": 3.4279475157114225,
"tokens_seen": 2017198080
},
{
"epoch": 0.61,
"learning_rate": 0.00019640922768304913,
"loss": 2.676,
"theoretical_loss": 3.42793824580189,
"tokens_seen": 2017263616
},
{
"epoch": 0.61,
"learning_rate": 0.00019639919759277834,
"loss": 2.837,
"theoretical_loss": 3.427928976277831,
"tokens_seen": 2017329152
},
{
"epoch": 0.61,
"learning_rate": 0.00019638916750250752,
"loss": 2.6084,
"theoretical_loss": 3.427919707139216,
"tokens_seen": 2017394688
},
{
"epoch": 0.61,
"learning_rate": 0.0001963791374122367,
"loss": 2.91,
"theoretical_loss": 3.4279104383860166,
"tokens_seen": 2017460224
},
{
"epoch": 0.61,
"learning_rate": 0.00019636910732196588,
"loss": 2.5701,
"theoretical_loss": 3.4279011700182043,
"tokens_seen": 2017525760
},
{
"epoch": 0.61,
"learning_rate": 0.0001963590772316951,
"loss": 2.8335,
"theoretical_loss": 3.427891902035751,
"tokens_seen": 2017591296
},
{
"epoch": 0.61,
"learning_rate": 0.00019634904714142427,
"loss": 2.8224,
"theoretical_loss": 3.427882634438627,
"tokens_seen": 2017656832
},
{
"epoch": 0.61,
"learning_rate": 0.00019633901705115345,
"loss": 2.522,
"theoretical_loss": 3.4278733672268045,
"tokens_seen": 2017722368
},
{
"epoch": 0.61,
"learning_rate": 0.00019632898696088263,
"loss": 2.6309,
"theoretical_loss": 3.4278641004002557,
"tokens_seen": 2017787904
},
{
"epoch": 0.61,
"learning_rate": 0.00019631895687061187,
"loss": 2.7805,
"theoretical_loss": 3.4278548339589507,
"tokens_seen": 2017853440
},
{
"epoch": 0.61,
"learning_rate": 0.00019630892678034105,
"loss": 2.7375,
"theoretical_loss": 3.427845567902862,
"tokens_seen": 2017918976
},
{
"epoch": 0.61,
"learning_rate": 0.00019629889669007023,
"loss": 2.6818,
"theoretical_loss": 3.427836302231961,
"tokens_seen": 2017984512
},
{
"epoch": 0.61,
"learning_rate": 0.0001962888665997994,
"loss": 2.5532,
"theoretical_loss": 3.4278270369462183,
"tokens_seen": 2018050048
},
{
"epoch": 0.61,
"learning_rate": 0.0001962788365095286,
"loss": 2.5836,
"theoretical_loss": 3.4278177720456062,
"tokens_seen": 2018115584
},
{
"epoch": 0.61,
"learning_rate": 0.0001962688064192578,
"loss": 2.725,
"theoretical_loss": 3.427808507530096,
"tokens_seen": 2018181120
},
{
"epoch": 0.61,
"learning_rate": 0.00019625877632898698,
"loss": 2.8654,
"theoretical_loss": 3.427799243399659,
"tokens_seen": 2018246656
},
{
"epoch": 0.61,
"learning_rate": 0.00019624874623871616,
"loss": 2.7568,
"theoretical_loss": 3.427789979654267,
"tokens_seen": 2018312192
},
{
"epoch": 0.61,
"learning_rate": 0.00019623871614844534,
"loss": 2.6427,
"theoretical_loss": 3.427780716293891,
"tokens_seen": 2018377728
},
{
"epoch": 0.61,
"learning_rate": 0.00019622868605817455,
"loss": 2.5828,
"theoretical_loss": 3.427771453318503,
"tokens_seen": 2018443264
},
{
"epoch": 0.61,
"objective/train/docs_used": 1141300,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8086745738983154,
"objective/train/theoretical_loss": 3.4277621907280746,
"objective/train/tokens_used": 2038968800,
"theoretical_loss": 3.4277621907280746,
"tokens_seen": 2018508800
},
{
"epoch": 0.61,
"learning_rate": 0.00019621865596790373,
"loss": 2.7695,
"theoretical_loss": 3.4277621907280746,
"tokens_seen": 2018508800
},
{
"epoch": 0.61,
"learning_rate": 0.0001962086258776329,
"loss": 2.7233,
"theoretical_loss": 3.4277529285225765,
"tokens_seen": 2018574336
},
{
"epoch": 0.61,
"learning_rate": 0.0001961985957873621,
"loss": 2.4795,
"theoretical_loss": 3.427743666701981,
"tokens_seen": 2018639872
},
{
"epoch": 0.61,
"learning_rate": 0.00019618856569709127,
"loss": 2.6475,
"theoretical_loss": 3.4277344052662597,
"tokens_seen": 2018705408
},
{
"epoch": 0.61,
"learning_rate": 0.00019617853560682048,
"loss": 2.7576,
"theoretical_loss": 3.4277251442153833,
"tokens_seen": 2018770944
},
{
"epoch": 0.61,
"learning_rate": 0.00019616850551654966,
"loss": 2.8064,
"theoretical_loss": 3.4277158835493236,
"tokens_seen": 2018836480
},
{
"epoch": 0.61,
"learning_rate": 0.00019615847542627885,
"loss": 2.6874,
"theoretical_loss": 3.427706623268053,
"tokens_seen": 2018902016
},
{
"epoch": 0.61,
"learning_rate": 0.00019614844533600803,
"loss": 2.8197,
"theoretical_loss": 3.4276973633715415,
"tokens_seen": 2018967552
},
{
"epoch": 0.61,
"learning_rate": 0.00019613841524573723,
"loss": 2.8167,
"theoretical_loss": 3.4276881038597615,
"tokens_seen": 2019033088
},
{
"epoch": 0.61,
"learning_rate": 0.00019612838515546642,
"loss": 2.8099,
"theoretical_loss": 3.427678844732685,
"tokens_seen": 2019098624
},
{
"epoch": 0.61,
"learning_rate": 0.0001961183550651956,
"loss": 2.4404,
"theoretical_loss": 3.427669585990283,
"tokens_seen": 2019164160
},
{
"epoch": 0.61,
"learning_rate": 0.00019610832497492478,
"loss": 2.8634,
"theoretical_loss": 3.427660327632527,
"tokens_seen": 2019229696
},
{
"epoch": 0.61,
"learning_rate": 0.00019609829488465396,
"loss": 2.7114,
"theoretical_loss": 3.4276510696593876,
"tokens_seen": 2019295232
},
{
"epoch": 0.61,
"learning_rate": 0.00019608826479438317,
"loss": 2.6398,
"theoretical_loss": 3.4276418120708385,
"tokens_seen": 2019360768
},
{
"epoch": 0.61,
"learning_rate": 0.00019607823470411235,
"loss": 2.5292,
"theoretical_loss": 3.4276325548668494,
"tokens_seen": 2019426304
},
{
"epoch": 0.61,
"learning_rate": 0.00019606820461384153,
"loss": 2.5366,
"theoretical_loss": 3.4276232980473926,
"tokens_seen": 2019491840
},
{
"epoch": 0.61,
"learning_rate": 0.0001960581745235707,
"loss": 2.6519,
"theoretical_loss": 3.42761404161244,
"tokens_seen": 2019557376
},
{
"epoch": 0.61,
"learning_rate": 0.00019604814443329992,
"loss": 2.8661,
"theoretical_loss": 3.4276047855619622,
"tokens_seen": 2019622912
},
{
"epoch": 0.61,
"learning_rate": 0.0001960381143430291,
"loss": 2.6737,
"theoretical_loss": 3.4275955298959313,
"tokens_seen": 2019688448
},
{
"epoch": 0.61,
"learning_rate": 0.00019602808425275828,
"loss": 2.8997,
"theoretical_loss": 3.4275862746143186,
"tokens_seen": 2019753984
},
{
"epoch": 0.61,
"learning_rate": 0.00019601805416248746,
"loss": 2.5031,
"theoretical_loss": 3.4275770197170963,
"tokens_seen": 2019819520
},
{
"epoch": 0.61,
"learning_rate": 0.00019600802407221664,
"loss": 2.7566,
"theoretical_loss": 3.427567765204236,
"tokens_seen": 2019885056
},
{
"epoch": 0.61,
"learning_rate": 0.00019599799398194585,
"loss": 2.8912,
"theoretical_loss": 3.4275585110757074,
"tokens_seen": 2019950592
},
{
"epoch": 0.61,
"learning_rate": 0.00019598796389167503,
"loss": 2.6259,
"theoretical_loss": 3.4275492573314845,
"tokens_seen": 2020016128
},
{
"epoch": 0.61,
"learning_rate": 0.00019597793380140421,
"loss": 2.805,
"theoretical_loss": 3.4275400039715374,
"tokens_seen": 2020081664
},
{
"epoch": 0.61,
"objective/train/docs_used": 1142665,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7612686157226562,
"objective/train/theoretical_loss": 3.427530750995839,
"objective/train/tokens_used": 2040607200,
"theoretical_loss": 3.427530750995839,
"tokens_seen": 2020147200
},
{
"epoch": 0.61,
"learning_rate": 0.0001959679037111334,
"loss": 2.7855,
"theoretical_loss": 3.427530750995839,
"tokens_seen": 2020147200
},
{
"epoch": 0.61,
"learning_rate": 0.0001959578736208626,
"loss": 2.8139,
"theoretical_loss": 3.427521498404359,
"tokens_seen": 2020212736
},
{
"epoch": 0.61,
"learning_rate": 0.00019594784353059178,
"loss": 2.6536,
"theoretical_loss": 3.42751224619707,
"tokens_seen": 2020278272
},
{
"epoch": 0.61,
"learning_rate": 0.00019593781344032097,
"loss": 2.7708,
"theoretical_loss": 3.427502994373944,
"tokens_seen": 2020343808
},
{
"epoch": 0.61,
"learning_rate": 0.00019592778335005015,
"loss": 2.5769,
"theoretical_loss": 3.427493742934952,
"tokens_seen": 2020409344
},
{
"epoch": 0.61,
"learning_rate": 0.00019591775325977933,
"loss": 2.5843,
"theoretical_loss": 3.4274844918800658,
"tokens_seen": 2020474880
},
{
"epoch": 0.61,
"learning_rate": 0.00019590772316950854,
"loss": 2.5662,
"theoretical_loss": 3.4274752412092573,
"tokens_seen": 2020540416
},
{
"epoch": 0.61,
"learning_rate": 0.00019589769307923772,
"loss": 2.5897,
"theoretical_loss": 3.427465990922497,
"tokens_seen": 2020605952
},
{
"epoch": 0.61,
"learning_rate": 0.0001958876629889669,
"loss": 2.6641,
"theoretical_loss": 3.427456741019758,
"tokens_seen": 2020671488
},
{
"epoch": 0.61,
"learning_rate": 0.00019587763289869608,
"loss": 2.6835,
"theoretical_loss": 3.4274474915010105,
"tokens_seen": 2020737024
},
{
"epoch": 0.61,
"learning_rate": 0.0001958676028084253,
"loss": 2.6267,
"theoretical_loss": 3.427438242366227,
"tokens_seen": 2020802560
},
{
"epoch": 0.61,
"learning_rate": 0.00019585757271815447,
"loss": 2.6507,
"theoretical_loss": 3.427428993615379,
"tokens_seen": 2020868096
},
{
"epoch": 0.61,
"learning_rate": 0.00019584754262788365,
"loss": 2.6877,
"theoretical_loss": 3.4274197452484376,
"tokens_seen": 2020933632
},
{
"epoch": 0.61,
"learning_rate": 0.00019583751253761283,
"loss": 2.7496,
"theoretical_loss": 3.4274104972653747,
"tokens_seen": 2020999168
},
{
"epoch": 0.61,
"learning_rate": 0.00019582748244734204,
"loss": 2.6803,
"theoretical_loss": 3.4274012496661626,
"tokens_seen": 2021064704
},
{
"epoch": 0.61,
"learning_rate": 0.00019581745235707122,
"loss": 2.5626,
"theoretical_loss": 3.4273920024507722,
"tokens_seen": 2021130240
},
{
"epoch": 0.61,
"learning_rate": 0.0001958074222668004,
"loss": 2.6728,
"theoretical_loss": 3.427382755619175,
"tokens_seen": 2021195776
},
{
"epoch": 0.61,
"learning_rate": 0.00019579739217652958,
"loss": 2.6438,
"theoretical_loss": 3.427373509171343,
"tokens_seen": 2021261312
},
{
"epoch": 0.61,
"learning_rate": 0.00019578736208625876,
"loss": 2.649,
"theoretical_loss": 3.427364263107248,
"tokens_seen": 2021326848
},
{
"epoch": 0.61,
"learning_rate": 0.00019577733199598797,
"loss": 2.7454,
"theoretical_loss": 3.4273550174268603,
"tokens_seen": 2021392384
},
{
"epoch": 0.61,
"learning_rate": 0.00019576730190571715,
"loss": 2.5579,
"theoretical_loss": 3.4273457721301535,
"tokens_seen": 2021457920
},
{
"epoch": 0.61,
"learning_rate": 0.00019575727181544633,
"loss": 2.7141,
"theoretical_loss": 3.4273365272170984,
"tokens_seen": 2021523456
},
{
"epoch": 0.61,
"learning_rate": 0.00019574724172517552,
"loss": 2.6174,
"theoretical_loss": 3.427327282687666,
"tokens_seen": 2021588992
},
{
"epoch": 0.61,
"learning_rate": 0.00019573721163490472,
"loss": 2.678,
"theoretical_loss": 3.427318038541829,
"tokens_seen": 2021654528
},
{
"epoch": 0.61,
"learning_rate": 0.0001957271815446339,
"loss": 2.5276,
"theoretical_loss": 3.4273087947795586,
"tokens_seen": 2021720064
},
{
"epoch": 0.61,
"objective/train/docs_used": 1143239,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.834577798843384,
"objective/train/theoretical_loss": 3.427299551400826,
"objective/train/tokens_used": 2042245600,
"theoretical_loss": 3.427299551400826,
"tokens_seen": 2021785600
},
{
"epoch": 0.61,
"learning_rate": 0.00019571715145436309,
"loss": 2.8237,
"theoretical_loss": 3.427299551400826,
"tokens_seen": 2021785600
},
{
"epoch": 0.61,
"learning_rate": 0.00019570712136409227,
"loss": 2.5652,
"theoretical_loss": 3.4272903084056034,
"tokens_seen": 2021851136
},
{
"epoch": 0.61,
"learning_rate": 0.00019569709127382145,
"loss": 2.6525,
"theoretical_loss": 3.4272810657938626,
"tokens_seen": 2021916672
},
{
"epoch": 0.61,
"learning_rate": 0.00019568706118355066,
"loss": 2.6948,
"theoretical_loss": 3.427271823565574,
"tokens_seen": 2021982208
},
{
"epoch": 0.61,
"learning_rate": 0.00019567703109327984,
"loss": 2.5167,
"theoretical_loss": 3.4272625817207114,
"tokens_seen": 2022047744
},
{
"epoch": 0.61,
"learning_rate": 0.00019566700100300902,
"loss": 2.4845,
"theoretical_loss": 3.427253340259245,
"tokens_seen": 2022113280
},
{
"epoch": 0.61,
"learning_rate": 0.0001956569709127382,
"loss": 2.5498,
"theoretical_loss": 3.427244099181147,
"tokens_seen": 2022178816
},
{
"epoch": 0.61,
"learning_rate": 0.0001956469408224674,
"loss": 2.6874,
"theoretical_loss": 3.427234858486388,
"tokens_seen": 2022244352
},
{
"epoch": 0.61,
"learning_rate": 0.0001956369107321966,
"loss": 2.6872,
"theoretical_loss": 3.4272256181749414,
"tokens_seen": 2022309888
},
{
"epoch": 0.61,
"learning_rate": 0.00019562688064192577,
"loss": 2.8147,
"theoretical_loss": 3.4272163782467775,
"tokens_seen": 2022375424
},
{
"epoch": 0.61,
"learning_rate": 0.00019561685055165495,
"loss": 2.8251,
"theoretical_loss": 3.4272071387018688,
"tokens_seen": 2022440960
},
{
"epoch": 0.61,
"learning_rate": 0.00019560682046138413,
"loss": 2.7662,
"theoretical_loss": 3.427197899540187,
"tokens_seen": 2022506496
},
{
"epoch": 0.61,
"learning_rate": 0.00019559679037111334,
"loss": 2.4545,
"theoretical_loss": 3.4271886607617024,
"tokens_seen": 2022572032
},
{
"epoch": 0.61,
"learning_rate": 0.00019558676028084252,
"loss": 2.5156,
"theoretical_loss": 3.4271794223663887,
"tokens_seen": 2022637568
},
{
"epoch": 0.61,
"learning_rate": 0.0001955767301905717,
"loss": 2.8846,
"theoretical_loss": 3.427170184354216,
"tokens_seen": 2022703104
},
{
"epoch": 0.61,
"learning_rate": 0.0001955667001003009,
"loss": 2.7781,
"theoretical_loss": 3.427160946725157,
"tokens_seen": 2022768640
},
{
"epoch": 0.61,
"learning_rate": 0.00019555667001003012,
"loss": 2.5998,
"theoretical_loss": 3.4271517094791832,
"tokens_seen": 2022834176
},
{
"epoch": 0.61,
"learning_rate": 0.0001955466399197593,
"loss": 2.797,
"theoretical_loss": 3.427142472616266,
"tokens_seen": 2022899712
},
{
"epoch": 0.61,
"learning_rate": 0.00019553660982948848,
"loss": 2.8696,
"theoretical_loss": 3.4271332361363775,
"tokens_seen": 2022965248
},
{
"epoch": 0.61,
"learning_rate": 0.00019552657973921766,
"loss": 2.43,
"theoretical_loss": 3.4271240000394885,
"tokens_seen": 2023030784
},
{
"epoch": 0.61,
"learning_rate": 0.00019551654964894684,
"loss": 2.771,
"theoretical_loss": 3.427114764325572,
"tokens_seen": 2023096320
},
{
"epoch": 0.61,
"learning_rate": 0.00019550651955867605,
"loss": 2.4059,
"theoretical_loss": 3.427105528994599,
"tokens_seen": 2023161856
},
{
"epoch": 0.61,
"learning_rate": 0.00019549648946840523,
"loss": 2.7317,
"theoretical_loss": 3.427096294046541,
"tokens_seen": 2023227392
},
{
"epoch": 0.61,
"learning_rate": 0.00019548645937813441,
"loss": 2.7124,
"theoretical_loss": 3.4270870594813703,
"tokens_seen": 2023292928
},
{
"epoch": 0.61,
"learning_rate": 0.0001954764292878636,
"loss": 2.6951,
"theoretical_loss": 3.427077825299058,
"tokens_seen": 2023358464
},
{
"epoch": 0.61,
"objective/train/docs_used": 1144661,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4763267040252686,
"objective/train/theoretical_loss": 3.4270685914995767,
"objective/train/tokens_used": 2043884000,
"theoretical_loss": 3.4270685914995767,
"tokens_seen": 2023424000
},
{
"epoch": 0.61,
"learning_rate": 0.0001954663991975928,
"loss": 2.4885,
"theoretical_loss": 3.4270685914995767,
"tokens_seen": 2023424000
},
{
"epoch": 0.61,
"learning_rate": 0.00019545636910732198,
"loss": 2.5793,
"theoretical_loss": 3.4270593580828974,
"tokens_seen": 2023489536
},
{
"epoch": 0.61,
"learning_rate": 0.00019544633901705117,
"loss": 2.765,
"theoretical_loss": 3.427050125048992,
"tokens_seen": 2023555072
},
{
"epoch": 0.61,
"learning_rate": 0.00019543630892678035,
"loss": 2.6756,
"theoretical_loss": 3.4270408923978324,
"tokens_seen": 2023620608
},
{
"epoch": 0.61,
"learning_rate": 0.00019542627883650953,
"loss": 2.7258,
"theoretical_loss": 3.42703166012939,
"tokens_seen": 2023686144
},
{
"epoch": 0.61,
"learning_rate": 0.00019541624874623874,
"loss": 2.9164,
"theoretical_loss": 3.427022428243637,
"tokens_seen": 2023751680
},
{
"epoch": 0.61,
"learning_rate": 0.00019540621865596792,
"loss": 2.8545,
"theoretical_loss": 3.427013196740545,
"tokens_seen": 2023817216
},
{
"epoch": 0.61,
"learning_rate": 0.0001953961885656971,
"loss": 2.6285,
"theoretical_loss": 3.4270039656200852,
"tokens_seen": 2023882752
},
{
"epoch": 0.61,
"learning_rate": 0.00019538615847542628,
"loss": 2.6245,
"theoretical_loss": 3.4269947348822303,
"tokens_seen": 2023948288
},
{
"epoch": 0.61,
"learning_rate": 0.0001953761283851555,
"loss": 2.7652,
"theoretical_loss": 3.426985504526951,
"tokens_seen": 2024013824
},
{
"epoch": 0.61,
"learning_rate": 0.00019536609829488467,
"loss": 2.4865,
"theoretical_loss": 3.42697627455422,
"tokens_seen": 2024079360
},
{
"epoch": 0.61,
"learning_rate": 0.00019535606820461385,
"loss": 2.6816,
"theoretical_loss": 3.426967044964009,
"tokens_seen": 2024144896
},
{
"epoch": 0.61,
"learning_rate": 0.00019534603811434303,
"loss": 2.7344,
"theoretical_loss": 3.426957815756289,
"tokens_seen": 2024210432
},
{
"epoch": 0.61,
"learning_rate": 0.00019533600802407224,
"loss": 2.7292,
"theoretical_loss": 3.426948586931032,
"tokens_seen": 2024275968
},
{
"epoch": 0.61,
"learning_rate": 0.00019532597793380142,
"loss": 2.6195,
"theoretical_loss": 3.4269393584882106,
"tokens_seen": 2024341504
},
{
"epoch": 0.61,
"learning_rate": 0.0001953159478435306,
"loss": 2.8543,
"theoretical_loss": 3.4269301304277953,
"tokens_seen": 2024407040
},
{
"epoch": 0.61,
"learning_rate": 0.00019530591775325978,
"loss": 2.6515,
"theoretical_loss": 3.4269209027497594,
"tokens_seen": 2024472576
},
{
"epoch": 0.61,
"learning_rate": 0.00019529588766298896,
"loss": 2.5843,
"theoretical_loss": 3.4269116754540727,
"tokens_seen": 2024538112
},
{
"epoch": 0.61,
"learning_rate": 0.00019528585757271817,
"loss": 2.495,
"theoretical_loss": 3.4269024485407087,
"tokens_seen": 2024603648
},
{
"epoch": 0.61,
"learning_rate": 0.00019527582748244735,
"loss": 2.6557,
"theoretical_loss": 3.426893222009639,
"tokens_seen": 2024669184
},
{
"epoch": 0.61,
"learning_rate": 0.00019526579739217653,
"loss": 2.6446,
"theoretical_loss": 3.426883995860834,
"tokens_seen": 2024734720
},
{
"epoch": 0.61,
"learning_rate": 0.00019525576730190572,
"loss": 2.6622,
"theoretical_loss": 3.4268747700942668,
"tokens_seen": 2024800256
},
{
"epoch": 0.61,
"learning_rate": 0.00019524573721163492,
"loss": 2.6818,
"theoretical_loss": 3.426865544709909,
"tokens_seen": 2024865792
},
{
"epoch": 0.61,
"learning_rate": 0.0001952357071213641,
"loss": 2.7589,
"theoretical_loss": 3.426856319707732,
"tokens_seen": 2024931328
},
{
"epoch": 0.61,
"learning_rate": 0.00019522567703109329,
"loss": 2.722,
"theoretical_loss": 3.426847095087708,
"tokens_seen": 2024996864
},
{
"epoch": 0.61,
"objective/train/docs_used": 1145363,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.959200620651245,
"objective/train/theoretical_loss": 3.4268378708498086,
"objective/train/tokens_used": 2045522400,
"theoretical_loss": 3.4268378708498086,
"tokens_seen": 2025062400
},
{
"epoch": 0.61,
"learning_rate": 0.00019521564694082247,
"loss": 2.8166,
"theoretical_loss": 3.4268378708498086,
"tokens_seen": 2025062400
},
{
"epoch": 0.61,
"learning_rate": 0.00019520561685055165,
"loss": 2.4636,
"theoretical_loss": 3.426828646994006,
"tokens_seen": 2025127936
},
{
"epoch": 0.61,
"learning_rate": 0.00019519558676028086,
"loss": 2.755,
"theoretical_loss": 3.4268194235202714,
"tokens_seen": 2025193472
},
{
"epoch": 0.61,
"learning_rate": 0.00019518555667001004,
"loss": 2.6255,
"theoretical_loss": 3.4268102004285765,
"tokens_seen": 2025259008
},
{
"epoch": 0.61,
"learning_rate": 0.00019517552657973922,
"loss": 2.6164,
"theoretical_loss": 3.4268009777188935,
"tokens_seen": 2025324544
},
{
"epoch": 0.61,
"learning_rate": 0.0001951654964894684,
"loss": 2.7295,
"theoretical_loss": 3.4267917553911946,
"tokens_seen": 2025390080
},
{
"epoch": 0.61,
"learning_rate": 0.0001951554663991976,
"loss": 2.5185,
"theoretical_loss": 3.426782533445451,
"tokens_seen": 2025455616
},
{
"epoch": 0.61,
"learning_rate": 0.0001951454363089268,
"loss": 2.531,
"theoretical_loss": 3.426773311881635,
"tokens_seen": 2025521152
},
{
"epoch": 0.61,
"learning_rate": 0.00019513540621865597,
"loss": 2.6816,
"theoretical_loss": 3.426764090699718,
"tokens_seen": 2025586688
},
{
"epoch": 0.61,
"learning_rate": 0.00019512537612838515,
"loss": 2.7462,
"theoretical_loss": 3.4267548698996717,
"tokens_seen": 2025652224
},
{
"epoch": 0.61,
"learning_rate": 0.00019511534603811433,
"loss": 2.7617,
"theoretical_loss": 3.4267456494814685,
"tokens_seen": 2025717760
},
{
"epoch": 0.61,
"learning_rate": 0.00019510531594784354,
"loss": 2.4845,
"theoretical_loss": 3.4267364294450795,
"tokens_seen": 2025783296
},
{
"epoch": 0.61,
"learning_rate": 0.00019509528585757272,
"loss": 2.696,
"theoretical_loss": 3.426727209790478,
"tokens_seen": 2025848832
},
{
"epoch": 0.61,
"learning_rate": 0.0001950852557673019,
"loss": 2.7224,
"theoretical_loss": 3.426717990517634,
"tokens_seen": 2025914368
},
{
"epoch": 0.61,
"learning_rate": 0.00019507522567703108,
"loss": 2.4884,
"theoretical_loss": 3.4267087716265205,
"tokens_seen": 2025979904
},
{
"epoch": 0.61,
"learning_rate": 0.0001950651955867603,
"loss": 2.7297,
"theoretical_loss": 3.4266995531171087,
"tokens_seen": 2026045440
},
{
"epoch": 0.61,
"learning_rate": 0.00019505516549648947,
"loss": 2.6236,
"theoretical_loss": 3.426690334989371,
"tokens_seen": 2026110976
},
{
"epoch": 0.61,
"learning_rate": 0.00019504513540621865,
"loss": 2.6268,
"theoretical_loss": 3.426681117243279,
"tokens_seen": 2026176512
},
{
"epoch": 0.61,
"learning_rate": 0.00019503510531594784,
"loss": 2.6978,
"theoretical_loss": 3.426671899878804,
"tokens_seen": 2026242048
},
{
"epoch": 0.61,
"learning_rate": 0.00019502507522567702,
"loss": 2.8411,
"theoretical_loss": 3.4266626828959197,
"tokens_seen": 2026307584
},
{
"epoch": 0.61,
"learning_rate": 0.00019501504513540622,
"loss": 2.706,
"theoretical_loss": 3.4266534662945958,
"tokens_seen": 2026373120
},
{
"epoch": 0.61,
"learning_rate": 0.0001950050150451354,
"loss": 2.9289,
"theoretical_loss": 3.4266442500748053,
"tokens_seen": 2026438656
},
{
"epoch": 0.61,
"learning_rate": 0.0001949949849548646,
"loss": 2.8102,
"theoretical_loss": 3.42663503423652,
"tokens_seen": 2026504192
},
{
"epoch": 0.61,
"learning_rate": 0.00019498495486459377,
"loss": 2.544,
"theoretical_loss": 3.426625818779711,
"tokens_seen": 2026569728
},
{
"epoch": 0.61,
"learning_rate": 0.00019497492477432298,
"loss": 2.5753,
"theoretical_loss": 3.4266166037043515,
"tokens_seen": 2026635264
},
{
"epoch": 0.61,
"objective/train/docs_used": 1146045,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.682671546936035,
"objective/train/theoretical_loss": 3.4266073890104125,
"objective/train/tokens_used": 2047160800,
"theoretical_loss": 3.4266073890104125,
"tokens_seen": 2026700800
},
{
"epoch": 0.61,
"learning_rate": 0.00019496489468405216,
"loss": 2.6835,
"theoretical_loss": 3.4266073890104125,
"tokens_seen": 2026700800
},
{
"epoch": 0.61,
"learning_rate": 0.00019495486459378134,
"loss": 2.6482,
"theoretical_loss": 3.426598174697866,
"tokens_seen": 2026766336
},
{
"epoch": 0.61,
"learning_rate": 0.00019494483450351052,
"loss": 2.7601,
"theoretical_loss": 3.4265889607666837,
"tokens_seen": 2026831872
},
{
"epoch": 0.61,
"learning_rate": 0.0001949348044132397,
"loss": 2.5781,
"theoretical_loss": 3.426579747216838,
"tokens_seen": 2026897408
},
{
"epoch": 0.61,
"learning_rate": 0.0001949247743229689,
"loss": 2.5865,
"theoretical_loss": 3.4265705340483,
"tokens_seen": 2026962944
},
{
"epoch": 0.61,
"learning_rate": 0.0001949147442326981,
"loss": 2.6575,
"theoretical_loss": 3.426561321261042,
"tokens_seen": 2027028480
},
{
"epoch": 0.61,
"learning_rate": 0.00019490471414242727,
"loss": 2.6291,
"theoretical_loss": 3.426552108855037,
"tokens_seen": 2027094016
},
{
"epoch": 0.61,
"learning_rate": 0.00019489468405215645,
"loss": 2.8315,
"theoretical_loss": 3.426542896830255,
"tokens_seen": 2027159552
},
{
"epoch": 0.61,
"learning_rate": 0.00019488465396188566,
"loss": 2.4605,
"theoretical_loss": 3.4265336851866692,
"tokens_seen": 2027225088
},
{
"epoch": 0.61,
"learning_rate": 0.00019487462387161484,
"loss": 2.5117,
"theoretical_loss": 3.4265244739242506,
"tokens_seen": 2027290624
},
{
"epoch": 0.61,
"learning_rate": 0.00019486459378134402,
"loss": 2.6658,
"theoretical_loss": 3.4265152630429725,
"tokens_seen": 2027356160
},
{
"epoch": 0.61,
"learning_rate": 0.0001948545636910732,
"loss": 2.7904,
"theoretical_loss": 3.426506052542805,
"tokens_seen": 2027421696
},
{
"epoch": 0.61,
"learning_rate": 0.00019484453360080239,
"loss": 2.7405,
"theoretical_loss": 3.426496842423721,
"tokens_seen": 2027487232
},
{
"epoch": 0.61,
"learning_rate": 0.0001948345035105316,
"loss": 2.949,
"theoretical_loss": 3.4264876326856926,
"tokens_seen": 2027552768
},
{
"epoch": 0.61,
"learning_rate": 0.00019482447342026077,
"loss": 2.5988,
"theoretical_loss": 3.4264784233286916,
"tokens_seen": 2027618304
},
{
"epoch": 0.61,
"learning_rate": 0.00019481444332998998,
"loss": 2.6435,
"theoretical_loss": 3.4264692143526894,
"tokens_seen": 2027683840
},
{
"epoch": 0.61,
"learning_rate": 0.00019480441323971916,
"loss": 2.792,
"theoretical_loss": 3.4264600057576584,
"tokens_seen": 2027749376
},
{
"epoch": 0.61,
"learning_rate": 0.00019479438314944837,
"loss": 2.6279,
"theoretical_loss": 3.4264507975435707,
"tokens_seen": 2027814912
},
{
"epoch": 0.61,
"learning_rate": 0.00019478435305917755,
"loss": 2.7576,
"theoretical_loss": 3.4264415897103975,
"tokens_seen": 2027880448
},
{
"epoch": 0.61,
"learning_rate": 0.00019477432296890673,
"loss": 2.4928,
"theoretical_loss": 3.4264323822581115,
"tokens_seen": 2027945984
},
{
"epoch": 0.61,
"learning_rate": 0.00019476429287863592,
"loss": 3.0044,
"theoretical_loss": 3.426423175186684,
"tokens_seen": 2028011520
},
{
"epoch": 0.61,
"learning_rate": 0.00019475426278836512,
"loss": 2.5536,
"theoretical_loss": 3.4264139684960875,
"tokens_seen": 2028077056
},
{
"epoch": 0.61,
"learning_rate": 0.0001947442326980943,
"loss": 2.7566,
"theoretical_loss": 3.4264047621862934,
"tokens_seen": 2028142592
},
{
"epoch": 0.61,
"learning_rate": 0.00019473420260782349,
"loss": 2.5446,
"theoretical_loss": 3.4263955562572743,
"tokens_seen": 2028208128
},
{
"epoch": 0.61,
"learning_rate": 0.00019472417251755267,
"loss": 2.6536,
"theoretical_loss": 3.4263863507090018,
"tokens_seen": 2028273664
},
{
"epoch": 0.61,
"objective/train/docs_used": 1147311,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7167694568634033,
"objective/train/theoretical_loss": 3.426377145541448,
"objective/train/tokens_used": 2048799200,
"theoretical_loss": 3.426377145541448,
"tokens_seen": 2028339200
},
{
"epoch": 0.61,
"learning_rate": 0.00019471414242728185,
"loss": 2.7033,
"theoretical_loss": 3.426377145541448,
"tokens_seen": 2028339200
},
{
"epoch": 0.61,
"learning_rate": 0.00019470411233701106,
"loss": 2.5118,
"theoretical_loss": 3.426367940754584,
"tokens_seen": 2028404736
},
{
"epoch": 0.61,
"learning_rate": 0.00019469408224674024,
"loss": 2.6686,
"theoretical_loss": 3.4263587363483827,
"tokens_seen": 2028470272
},
{
"epoch": 0.61,
"learning_rate": 0.00019468405215646942,
"loss": 2.848,
"theoretical_loss": 3.426349532322816,
"tokens_seen": 2028535808
},
{
"epoch": 0.61,
"learning_rate": 0.0001946740220661986,
"loss": 2.6319,
"theoretical_loss": 3.4263403286778558,
"tokens_seen": 2028601344
},
{
"epoch": 0.61,
"learning_rate": 0.0001946639919759278,
"loss": 2.5663,
"theoretical_loss": 3.4263311254134736,
"tokens_seen": 2028666880
},
{
"epoch": 0.61,
"learning_rate": 0.000194653961885657,
"loss": 2.6691,
"theoretical_loss": 3.4263219225296417,
"tokens_seen": 2028732416
},
{
"epoch": 0.61,
"learning_rate": 0.00019464393179538617,
"loss": 2.7504,
"theoretical_loss": 3.4263127200263326,
"tokens_seen": 2028797952
},
{
"epoch": 0.61,
"learning_rate": 0.00019463390170511535,
"loss": 2.5805,
"theoretical_loss": 3.4263035179035173,
"tokens_seen": 2028863488
},
{
"epoch": 0.61,
"learning_rate": 0.00019462387161484453,
"loss": 2.6083,
"theoretical_loss": 3.4262943161611688,
"tokens_seen": 2028929024
},
{
"epoch": 0.61,
"learning_rate": 0.00019461384152457374,
"loss": 2.9695,
"theoretical_loss": 3.4262851147992577,
"tokens_seen": 2028994560
},
{
"epoch": 0.61,
"learning_rate": 0.00019460381143430292,
"loss": 2.6299,
"theoretical_loss": 3.426275913817757,
"tokens_seen": 2029060096
},
{
"epoch": 0.61,
"learning_rate": 0.0001945937813440321,
"loss": 2.8273,
"theoretical_loss": 3.4262667132166387,
"tokens_seen": 2029125632
},
{
"epoch": 0.61,
"learning_rate": 0.00019458375125376128,
"loss": 2.6038,
"theoretical_loss": 3.4262575129958748,
"tokens_seen": 2029191168
},
{
"epoch": 0.61,
"learning_rate": 0.0001945737211634905,
"loss": 2.731,
"theoretical_loss": 3.426248313155437,
"tokens_seen": 2029256704
},
{
"epoch": 0.61,
"learning_rate": 0.00019456369107321967,
"loss": 2.4657,
"theoretical_loss": 3.426239113695297,
"tokens_seen": 2029322240
},
{
"epoch": 0.61,
"learning_rate": 0.00019455366098294885,
"loss": 2.6251,
"theoretical_loss": 3.426229914615427,
"tokens_seen": 2029387776
},
{
"epoch": 0.61,
"learning_rate": 0.00019454363089267804,
"loss": 2.8589,
"theoretical_loss": 3.4262207159158,
"tokens_seen": 2029453312
},
{
"epoch": 0.62,
"learning_rate": 0.00019453360080240722,
"loss": 2.7922,
"theoretical_loss": 3.4262115175963865,
"tokens_seen": 2029518848
},
{
"epoch": 0.62,
"learning_rate": 0.00019452357071213642,
"loss": 2.6651,
"theoretical_loss": 3.4262023196571594,
"tokens_seen": 2029584384
},
{
"epoch": 0.62,
"learning_rate": 0.0001945135406218656,
"loss": 2.4895,
"theoretical_loss": 3.4261931220980903,
"tokens_seen": 2029649920
},
{
"epoch": 0.62,
"learning_rate": 0.0001945035105315948,
"loss": 2.5262,
"theoretical_loss": 3.426183924919152,
"tokens_seen": 2029715456
},
{
"epoch": 0.62,
"learning_rate": 0.00019449348044132397,
"loss": 2.9889,
"theoretical_loss": 3.4261747281203148,
"tokens_seen": 2029780992
},
{
"epoch": 0.62,
"learning_rate": 0.00019448345035105318,
"loss": 2.7515,
"theoretical_loss": 3.4261655317015527,
"tokens_seen": 2029846528
},
{
"epoch": 0.62,
"learning_rate": 0.00019447342026078236,
"loss": 2.7914,
"theoretical_loss": 3.426156335662837,
"tokens_seen": 2029912064
},
{
"epoch": 0.62,
"objective/train/docs_used": 1147924,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2753171920776367,
"objective/train/theoretical_loss": 3.426147140004139,
"objective/train/tokens_used": 2050437600,
"theoretical_loss": 3.426147140004139,
"tokens_seen": 2029977600
},
{
"epoch": 0.62,
"learning_rate": 0.00019446339017051154,
"loss": 2.5965,
"theoretical_loss": 3.426147140004139,
"tokens_seen": 2029977600
},
{
"epoch": 0.62,
"learning_rate": 0.00019445336008024072,
"loss": 2.7128,
"theoretical_loss": 3.4261379447254314,
"tokens_seen": 2030043136
},
{
"epoch": 0.62,
"learning_rate": 0.0001944433299899699,
"loss": 3.0194,
"theoretical_loss": 3.4261287498266864,
"tokens_seen": 2030108672
},
{
"epoch": 0.62,
"learning_rate": 0.0001944332998996991,
"loss": 2.5592,
"theoretical_loss": 3.4261195553078756,
"tokens_seen": 2030174208
},
{
"epoch": 0.62,
"learning_rate": 0.0001944232698094283,
"loss": 2.3204,
"theoretical_loss": 3.4261103611689716,
"tokens_seen": 2030239744
},
{
"epoch": 0.62,
"learning_rate": 0.00019441323971915747,
"loss": 2.6324,
"theoretical_loss": 3.4261011674099455,
"tokens_seen": 2030305280
},
{
"epoch": 0.62,
"learning_rate": 0.00019440320962888665,
"loss": 2.7268,
"theoretical_loss": 3.4260919740307703,
"tokens_seen": 2030370816
},
{
"epoch": 0.62,
"learning_rate": 0.00019439317953861586,
"loss": 2.5545,
"theoretical_loss": 3.4260827810314174,
"tokens_seen": 2030436352
},
{
"epoch": 0.62,
"learning_rate": 0.00019438314944834504,
"loss": 2.616,
"theoretical_loss": 3.426073588411859,
"tokens_seen": 2030501888
},
{
"epoch": 0.62,
"learning_rate": 0.00019437311935807422,
"loss": 2.6957,
"theoretical_loss": 3.426064396172068,
"tokens_seen": 2030567424
},
{
"epoch": 0.62,
"learning_rate": 0.0001943630892678034,
"loss": 2.6125,
"theoretical_loss": 3.4260552043120147,
"tokens_seen": 2030632960
},
{
"epoch": 0.62,
"learning_rate": 0.00019435305917753259,
"loss": 2.6386,
"theoretical_loss": 3.4260460128316725,
"tokens_seen": 2030698496
},
{
"epoch": 0.62,
"learning_rate": 0.0001943430290872618,
"loss": 2.7823,
"theoretical_loss": 3.4260368217310133,
"tokens_seen": 2030764032
},
{
"epoch": 0.62,
"learning_rate": 0.00019433299899699097,
"loss": 2.5205,
"theoretical_loss": 3.426027631010009,
"tokens_seen": 2030829568
},
{
"epoch": 0.62,
"learning_rate": 0.00019432296890672016,
"loss": 2.6728,
"theoretical_loss": 3.4260184406686314,
"tokens_seen": 2030895104
},
{
"epoch": 0.62,
"learning_rate": 0.00019431293881644934,
"loss": 2.6202,
"theoretical_loss": 3.426009250706853,
"tokens_seen": 2030960640
},
{
"epoch": 0.62,
"learning_rate": 0.00019430290872617855,
"loss": 2.3049,
"theoretical_loss": 3.4260000611246455,
"tokens_seen": 2031026176
},
{
"epoch": 0.62,
"learning_rate": 0.00019429287863590773,
"loss": 2.7034,
"theoretical_loss": 3.4259908719219814,
"tokens_seen": 2031091712
},
{
"epoch": 0.62,
"learning_rate": 0.0001942828485456369,
"loss": 2.5536,
"theoretical_loss": 3.4259816830988328,
"tokens_seen": 2031157248
},
{
"epoch": 0.62,
"learning_rate": 0.0001942728184553661,
"loss": 2.5187,
"theoretical_loss": 3.425972494655171,
"tokens_seen": 2031222784
},
{
"epoch": 0.62,
"learning_rate": 0.0001942627883650953,
"loss": 2.5648,
"theoretical_loss": 3.425963306590969,
"tokens_seen": 2031288320
},
{
"epoch": 0.62,
"learning_rate": 0.00019425275827482448,
"loss": 2.6512,
"theoretical_loss": 3.4259541189061977,
"tokens_seen": 2031353856
},
{
"epoch": 0.62,
"learning_rate": 0.00019424272818455366,
"loss": 2.3923,
"theoretical_loss": 3.4259449316008306,
"tokens_seen": 2031419392
},
{
"epoch": 0.62,
"learning_rate": 0.00019423269809428284,
"loss": 2.7136,
"theoretical_loss": 3.4259357446748395,
"tokens_seen": 2031484928
},
{
"epoch": 0.62,
"learning_rate": 0.00019422266800401202,
"loss": 2.6926,
"theoretical_loss": 3.4259265581281957,
"tokens_seen": 2031550464
},
{
"epoch": 0.62,
"objective/train/docs_used": 1149081,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0337157249450684,
"objective/train/theoretical_loss": 3.425917371960872,
"objective/train/tokens_used": 2052076000,
"theoretical_loss": 3.425917371960872,
"tokens_seen": 2031616000
},
{
"epoch": 0.62,
"learning_rate": 0.00019421263791374123,
"loss": 2.5494,
"theoretical_loss": 3.425917371960872,
"tokens_seen": 2031616000
},
{
"epoch": 0.62,
"learning_rate": 0.0001942026078234704,
"loss": 2.8723,
"theoretical_loss": 3.4259081861728404,
"tokens_seen": 2031681536
},
{
"epoch": 0.62,
"learning_rate": 0.0001941925777331996,
"loss": 2.6242,
"theoretical_loss": 3.4258990007640726,
"tokens_seen": 2031747072
},
{
"epoch": 0.62,
"learning_rate": 0.00019418254764292877,
"loss": 2.3752,
"theoretical_loss": 3.425889815734541,
"tokens_seen": 2031812608
},
{
"epoch": 0.62,
"learning_rate": 0.00019417251755265798,
"loss": 2.6828,
"theoretical_loss": 3.425880631084218,
"tokens_seen": 2031878144
},
{
"epoch": 0.62,
"learning_rate": 0.00019416248746238716,
"loss": 2.6017,
"theoretical_loss": 3.4258714468130753,
"tokens_seen": 2031943680
},
{
"epoch": 0.62,
"learning_rate": 0.00019415245737211634,
"loss": 2.6625,
"theoretical_loss": 3.425862262921085,
"tokens_seen": 2032009216
},
{
"epoch": 0.62,
"learning_rate": 0.00019414242728184552,
"loss": 2.8432,
"theoretical_loss": 3.425853079408219,
"tokens_seen": 2032074752
},
{
"epoch": 0.62,
"learning_rate": 0.0001941323971915747,
"loss": 2.7025,
"theoretical_loss": 3.4258438962744506,
"tokens_seen": 2032140288
},
{
"epoch": 0.62,
"learning_rate": 0.00019412236710130391,
"loss": 2.7289,
"theoretical_loss": 3.4258347135197504,
"tokens_seen": 2032205824
},
{
"epoch": 0.62,
"learning_rate": 0.0001941123370110331,
"loss": 2.7999,
"theoretical_loss": 3.425825531144092,
"tokens_seen": 2032271360
},
{
"epoch": 0.62,
"learning_rate": 0.00019410230692076228,
"loss": 2.5397,
"theoretical_loss": 3.4258163491474463,
"tokens_seen": 2032336896
},
{
"epoch": 0.62,
"learning_rate": 0.00019409227683049146,
"loss": 2.3858,
"theoretical_loss": 3.4258071675297854,
"tokens_seen": 2032402432
},
{
"epoch": 0.62,
"learning_rate": 0.00019408224674022067,
"loss": 2.5696,
"theoretical_loss": 3.425797986291083,
"tokens_seen": 2032467968
},
{
"epoch": 0.62,
"learning_rate": 0.00019407221664994985,
"loss": 2.6689,
"theoretical_loss": 3.425788805431309,
"tokens_seen": 2032533504
},
{
"epoch": 0.62,
"learning_rate": 0.00019406218655967905,
"loss": 2.5418,
"theoretical_loss": 3.4257796249504375,
"tokens_seen": 2032599040
},
{
"epoch": 0.62,
"learning_rate": 0.00019405215646940824,
"loss": 2.5888,
"theoretical_loss": 3.42577044484844,
"tokens_seen": 2032664576
},
{
"epoch": 0.62,
"learning_rate": 0.00019404212637913742,
"loss": 2.4848,
"theoretical_loss": 3.425761265125288,
"tokens_seen": 2032730112
},
{
"epoch": 0.62,
"learning_rate": 0.00019403209628886663,
"loss": 2.5593,
"theoretical_loss": 3.4257520857809545,
"tokens_seen": 2032795648
},
{
"epoch": 0.62,
"learning_rate": 0.0001940220661985958,
"loss": 2.5071,
"theoretical_loss": 3.4257429068154113,
"tokens_seen": 2032861184
},
{
"epoch": 0.62,
"learning_rate": 0.000194012036108325,
"loss": 2.6193,
"theoretical_loss": 3.4257337282286304,
"tokens_seen": 2032926720
},
{
"epoch": 0.62,
"learning_rate": 0.00019400200601805417,
"loss": 2.489,
"theoretical_loss": 3.4257245500205844,
"tokens_seen": 2032992256
},
{
"epoch": 0.62,
"learning_rate": 0.00019399197592778338,
"loss": 2.6039,
"theoretical_loss": 3.4257153721912452,
"tokens_seen": 2033057792
},
{
"epoch": 0.62,
"learning_rate": 0.00019398194583751256,
"loss": 2.556,
"theoretical_loss": 3.4257061947405845,
"tokens_seen": 2033123328
},
{
"epoch": 0.62,
"learning_rate": 0.00019397191574724174,
"loss": 2.7106,
"theoretical_loss": 3.4256970176685755,
"tokens_seen": 2033188864
},
{
"epoch": 0.62,
"objective/train/docs_used": 1149676,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.0733673572540283,
"objective/train/theoretical_loss": 3.42568784097519,
"objective/train/tokens_used": 2053714400,
"theoretical_loss": 3.42568784097519,
"tokens_seen": 2033254400
},
{
"epoch": 0.62,
"learning_rate": 0.00019396188565697092,
"loss": 2.6203,
"theoretical_loss": 3.42568784097519,
"tokens_seen": 2033254400
},
{
"epoch": 0.62,
"learning_rate": 0.0001939518555667001,
"loss": 2.5479,
"theoretical_loss": 3.425678664660399,
"tokens_seen": 2033319936
},
{
"epoch": 0.62,
"learning_rate": 0.0001939418254764293,
"loss": 2.7581,
"theoretical_loss": 3.4256694887241768,
"tokens_seen": 2033385472
},
{
"epoch": 0.62,
"learning_rate": 0.0001939317953861585,
"loss": 2.3908,
"theoretical_loss": 3.4256603131664933,
"tokens_seen": 2033451008
},
{
"epoch": 0.62,
"learning_rate": 0.00019392176529588767,
"loss": 2.7448,
"theoretical_loss": 3.4256511379873227,
"tokens_seen": 2033516544
},
{
"epoch": 0.62,
"learning_rate": 0.00019391173520561685,
"loss": 2.6026,
"theoretical_loss": 3.425641963186636,
"tokens_seen": 2033582080
},
{
"epoch": 0.62,
"learning_rate": 0.00019390170511534606,
"loss": 2.8747,
"theoretical_loss": 3.425632788764406,
"tokens_seen": 2033647616
},
{
"epoch": 0.62,
"learning_rate": 0.00019389167502507524,
"loss": 2.5858,
"theoretical_loss": 3.425623614720604,
"tokens_seen": 2033713152
},
{
"epoch": 0.62,
"learning_rate": 0.00019388164493480442,
"loss": 2.584,
"theoretical_loss": 3.4256144410552034,
"tokens_seen": 2033778688
},
{
"epoch": 0.62,
"learning_rate": 0.0001938716148445336,
"loss": 2.6144,
"theoretical_loss": 3.4256052677681756,
"tokens_seen": 2033844224
},
{
"epoch": 0.62,
"learning_rate": 0.00019386158475426279,
"loss": 2.621,
"theoretical_loss": 3.4255960948594932,
"tokens_seen": 2033909760
},
{
"epoch": 0.62,
"learning_rate": 0.000193851554663992,
"loss": 2.6339,
"theoretical_loss": 3.425586922329128,
"tokens_seen": 2033975296
},
{
"epoch": 0.62,
"learning_rate": 0.00019384152457372117,
"loss": 2.7636,
"theoretical_loss": 3.4255777501770526,
"tokens_seen": 2034040832
},
{
"epoch": 0.62,
"learning_rate": 0.00019383149448345036,
"loss": 2.6431,
"theoretical_loss": 3.4255685784032393,
"tokens_seen": 2034106368
},
{
"epoch": 0.62,
"learning_rate": 0.00019382146439317954,
"loss": 2.6226,
"theoretical_loss": 3.4255594070076594,
"tokens_seen": 2034171904
},
{
"epoch": 0.62,
"learning_rate": 0.00019381143430290875,
"loss": 2.7341,
"theoretical_loss": 3.4255502359902863,
"tokens_seen": 2034237440
},
{
"epoch": 0.62,
"learning_rate": 0.00019380140421263793,
"loss": 2.5978,
"theoretical_loss": 3.4255410653510916,
"tokens_seen": 2034302976
},
{
"epoch": 0.62,
"learning_rate": 0.0001937913741223671,
"loss": 2.9551,
"theoretical_loss": 3.425531895090047,
"tokens_seen": 2034368512
},
{
"epoch": 0.62,
"learning_rate": 0.0001937813440320963,
"loss": 2.5231,
"theoretical_loss": 3.425522725207126,
"tokens_seen": 2034434048
},
{
"epoch": 0.62,
"learning_rate": 0.0001937713139418255,
"loss": 2.5399,
"theoretical_loss": 3.4255135557022998,
"tokens_seen": 2034499584
},
{
"epoch": 0.62,
"learning_rate": 0.00019376128385155468,
"loss": 2.3599,
"theoretical_loss": 3.4255043865755415,
"tokens_seen": 2034565120
},
{
"epoch": 0.62,
"learning_rate": 0.00019375125376128386,
"loss": 2.5661,
"theoretical_loss": 3.425495217826822,
"tokens_seen": 2034630656
},
{
"epoch": 0.62,
"learning_rate": 0.00019374122367101304,
"loss": 2.6419,
"theoretical_loss": 3.425486049456115,
"tokens_seen": 2034696192
},
{
"epoch": 0.62,
"learning_rate": 0.00019373119358074222,
"loss": 2.7082,
"theoretical_loss": 3.425476881463392,
"tokens_seen": 2034761728
},
{
"epoch": 0.62,
"learning_rate": 0.00019372116349047143,
"loss": 2.498,
"theoretical_loss": 3.425467713848625,
"tokens_seen": 2034827264
},
{
"epoch": 0.62,
"objective/train/docs_used": 1150523,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9078264236450195,
"objective/train/theoretical_loss": 3.425458546611787,
"objective/train/tokens_used": 2055352800,
"theoretical_loss": 3.425458546611787,
"tokens_seen": 2034892800
},
{
"epoch": 0.62,
"learning_rate": 0.0001937111334002006,
"loss": 2.771,
"theoretical_loss": 3.425458546611787,
"tokens_seen": 2034892800
},
{
"epoch": 0.62,
"learning_rate": 0.0001937011033099298,
"loss": 2.5281,
"theoretical_loss": 3.42544937975285,
"tokens_seen": 2034958336
},
{
"epoch": 0.62,
"learning_rate": 0.00019369107321965897,
"loss": 2.7322,
"theoretical_loss": 3.4254402132717856,
"tokens_seen": 2035023872
},
{
"epoch": 0.62,
"learning_rate": 0.00019368104312938818,
"loss": 2.6025,
"theoretical_loss": 3.4254310471685665,
"tokens_seen": 2035089408
},
{
"epoch": 0.62,
"learning_rate": 0.00019367101303911736,
"loss": 2.5978,
"theoretical_loss": 3.4254218814431656,
"tokens_seen": 2035154944
},
{
"epoch": 0.62,
"learning_rate": 0.00019366098294884654,
"loss": 2.6449,
"theoretical_loss": 3.425412716095554,
"tokens_seen": 2035220480
},
{
"epoch": 0.62,
"learning_rate": 0.00019365095285857572,
"loss": 2.5325,
"theoretical_loss": 3.4254035511257053,
"tokens_seen": 2035286016
},
{
"epoch": 0.62,
"learning_rate": 0.0001936409227683049,
"loss": 2.5583,
"theoretical_loss": 3.4253943865335903,
"tokens_seen": 2035351552
},
{
"epoch": 0.62,
"learning_rate": 0.00019363089267803411,
"loss": 2.6154,
"theoretical_loss": 3.425385222319182,
"tokens_seen": 2035417088
},
{
"epoch": 0.62,
"learning_rate": 0.0001936208625877633,
"loss": 2.727,
"theoretical_loss": 3.425376058482453,
"tokens_seen": 2035482624
},
{
"epoch": 0.62,
"learning_rate": 0.00019361083249749248,
"loss": 2.5939,
"theoretical_loss": 3.425366895023375,
"tokens_seen": 2035548160
},
{
"epoch": 0.62,
"learning_rate": 0.00019360080240722166,
"loss": 2.7045,
"theoretical_loss": 3.425357731941921,
"tokens_seen": 2035613696
},
{
"epoch": 0.62,
"learning_rate": 0.00019359077231695087,
"loss": 2.6486,
"theoretical_loss": 3.425348569238062,
"tokens_seen": 2035679232
},
{
"epoch": 0.62,
"learning_rate": 0.00019358074222668005,
"loss": 2.7118,
"theoretical_loss": 3.4253394069117715,
"tokens_seen": 2035744768
},
{
"epoch": 0.62,
"learning_rate": 0.00019357071213640923,
"loss": 2.73,
"theoretical_loss": 3.425330244963021,
"tokens_seen": 2035810304
},
{
"epoch": 0.62,
"learning_rate": 0.0001935606820461384,
"loss": 2.7438,
"theoretical_loss": 3.4253210833917835,
"tokens_seen": 2035875840
},
{
"epoch": 0.62,
"learning_rate": 0.0001935506519558676,
"loss": 2.7352,
"theoretical_loss": 3.425311922198031,
"tokens_seen": 2035941376
},
{
"epoch": 0.62,
"learning_rate": 0.0001935406218655968,
"loss": 2.5672,
"theoretical_loss": 3.4253027613817357,
"tokens_seen": 2036006912
},
{
"epoch": 0.62,
"learning_rate": 0.00019353059177532598,
"loss": 2.8482,
"theoretical_loss": 3.4252936009428696,
"tokens_seen": 2036072448
},
{
"epoch": 0.62,
"learning_rate": 0.00019352056168505516,
"loss": 2.7137,
"theoretical_loss": 3.4252844408814056,
"tokens_seen": 2036137984
},
{
"epoch": 0.62,
"learning_rate": 0.00019351053159478434,
"loss": 2.5126,
"theoretical_loss": 3.4252752811973157,
"tokens_seen": 2036203520
},
{
"epoch": 0.62,
"learning_rate": 0.00019350050150451355,
"loss": 2.6855,
"theoretical_loss": 3.4252661218905724,
"tokens_seen": 2036269056
},
{
"epoch": 0.62,
"learning_rate": 0.00019349047141424273,
"loss": 2.8819,
"theoretical_loss": 3.4252569629611473,
"tokens_seen": 2036334592
},
{
"epoch": 0.62,
"learning_rate": 0.0001934804413239719,
"loss": 2.6631,
"theoretical_loss": 3.4252478044090138,
"tokens_seen": 2036400128
},
{
"epoch": 0.62,
"learning_rate": 0.0001934704112337011,
"loss": 2.5673,
"theoretical_loss": 3.4252386462341438,
"tokens_seen": 2036465664
},
{
"epoch": 0.62,
"objective/train/docs_used": 1151127,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5571155548095703,
"objective/train/theoretical_loss": 3.4252294884365093,
"objective/train/tokens_used": 2056991200,
"theoretical_loss": 3.4252294884365093,
"tokens_seen": 2036531200
},
{
"epoch": 0.62,
"learning_rate": 0.00019346038114343027,
"loss": 2.5723,
"theoretical_loss": 3.4252294884365093,
"tokens_seen": 2036531200
},
{
"epoch": 0.62,
"learning_rate": 0.00019345035105315948,
"loss": 2.6007,
"theoretical_loss": 3.425220331016083,
"tokens_seen": 2036596736
},
{
"epoch": 0.62,
"learning_rate": 0.00019344032096288866,
"loss": 2.3993,
"theoretical_loss": 3.4252111739728366,
"tokens_seen": 2036662272
},
{
"epoch": 0.62,
"learning_rate": 0.00019343029087261784,
"loss": 2.6803,
"theoretical_loss": 3.4252020173067432,
"tokens_seen": 2036727808
},
{
"epoch": 0.62,
"learning_rate": 0.00019342026078234703,
"loss": 2.6811,
"theoretical_loss": 3.4251928610177744,
"tokens_seen": 2036793344
},
{
"epoch": 0.62,
"learning_rate": 0.00019341023069207623,
"loss": 2.6056,
"theoretical_loss": 3.4251837051059035,
"tokens_seen": 2036858880
},
{
"epoch": 0.62,
"learning_rate": 0.00019340020060180542,
"loss": 2.6942,
"theoretical_loss": 3.425174549571102,
"tokens_seen": 2036924416
},
{
"epoch": 0.62,
"learning_rate": 0.0001933901705115346,
"loss": 2.84,
"theoretical_loss": 3.4251653944133427,
"tokens_seen": 2036989952
},
{
"epoch": 0.62,
"learning_rate": 0.00019338014042126378,
"loss": 2.7274,
"theoretical_loss": 3.4251562396325976,
"tokens_seen": 2037055488
},
{
"epoch": 0.62,
"learning_rate": 0.00019337011033099296,
"loss": 2.6297,
"theoretical_loss": 3.4251470852288395,
"tokens_seen": 2037121024
},
{
"epoch": 0.62,
"learning_rate": 0.00019336008024072217,
"loss": 2.7411,
"theoretical_loss": 3.4251379312020402,
"tokens_seen": 2037186560
},
{
"epoch": 0.62,
"learning_rate": 0.00019335005015045135,
"loss": 2.5864,
"theoretical_loss": 3.4251287775521724,
"tokens_seen": 2037252096
},
{
"epoch": 0.62,
"learning_rate": 0.00019334002006018053,
"loss": 2.6711,
"theoretical_loss": 3.4251196242792084,
"tokens_seen": 2037317632
},
{
"epoch": 0.62,
"learning_rate": 0.0001933299899699097,
"loss": 2.629,
"theoretical_loss": 3.42511047138312,
"tokens_seen": 2037383168
},
{
"epoch": 0.62,
"learning_rate": 0.00019331995987963892,
"loss": 2.5975,
"theoretical_loss": 3.425101318863881,
"tokens_seen": 2037448704
},
{
"epoch": 0.62,
"learning_rate": 0.00019330992978936813,
"loss": 2.4196,
"theoretical_loss": 3.4250921667214627,
"tokens_seen": 2037514240
},
{
"epoch": 0.62,
"learning_rate": 0.0001932998996990973,
"loss": 2.6175,
"theoretical_loss": 3.4250830149558373,
"tokens_seen": 2037579776
},
{
"epoch": 0.62,
"learning_rate": 0.0001932898696088265,
"loss": 2.5804,
"theoretical_loss": 3.4250738635669773,
"tokens_seen": 2037645312
},
{
"epoch": 0.62,
"learning_rate": 0.0001932798395185557,
"loss": 2.7219,
"theoretical_loss": 3.425064712554856,
"tokens_seen": 2037710848
},
{
"epoch": 0.62,
"learning_rate": 0.00019326980942828488,
"loss": 2.8197,
"theoretical_loss": 3.425055561919444,
"tokens_seen": 2037776384
},
{
"epoch": 0.62,
"learning_rate": 0.00019325977933801406,
"loss": 2.3939,
"theoretical_loss": 3.4250464116607153,
"tokens_seen": 2037841920
},
{
"epoch": 0.62,
"learning_rate": 0.00019324974924774324,
"loss": 2.4768,
"theoretical_loss": 3.4250372617786415,
"tokens_seen": 2037907456
},
{
"epoch": 0.62,
"learning_rate": 0.00019323971915747242,
"loss": 2.8809,
"theoretical_loss": 3.4250281122731954,
"tokens_seen": 2037972992
},
{
"epoch": 0.62,
"learning_rate": 0.00019322968906720163,
"loss": 2.5676,
"theoretical_loss": 3.4250189631443493,
"tokens_seen": 2038038528
},
{
"epoch": 0.62,
"learning_rate": 0.0001932196589769308,
"loss": 2.6237,
"theoretical_loss": 3.425009814392075,
"tokens_seen": 2038104064
},
{
"epoch": 0.62,
"objective/train/docs_used": 1152471,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.45070481300354,
"objective/train/theoretical_loss": 3.425000666016346,
"objective/train/tokens_used": 2058629600,
"theoretical_loss": 3.425000666016346,
"tokens_seen": 2038169600
},
{
"epoch": 0.62,
"learning_rate": 0.00019320962888666,
"loss": 2.4284,
"theoretical_loss": 3.425000666016346,
"tokens_seen": 2038169600
},
{
"epoch": 0.62,
"learning_rate": 0.00019319959879638917,
"loss": 2.5328,
"theoretical_loss": 3.4249915180171335,
"tokens_seen": 2038235136
},
{
"epoch": 0.62,
"learning_rate": 0.00019318956870611838,
"loss": 2.559,
"theoretical_loss": 3.4249823703944102,
"tokens_seen": 2038300672
},
{
"epoch": 0.62,
"learning_rate": 0.00019317953861584756,
"loss": 2.7165,
"theoretical_loss": 3.4249732231481493,
"tokens_seen": 2038366208
},
{
"epoch": 0.62,
"learning_rate": 0.00019316950852557674,
"loss": 2.6975,
"theoretical_loss": 3.4249640762783224,
"tokens_seen": 2038431744
},
{
"epoch": 0.62,
"learning_rate": 0.00019315947843530592,
"loss": 2.5604,
"theoretical_loss": 3.424954929784902,
"tokens_seen": 2038497280
},
{
"epoch": 0.62,
"learning_rate": 0.0001931494483450351,
"loss": 2.4539,
"theoretical_loss": 3.424945783667861,
"tokens_seen": 2038562816
},
{
"epoch": 0.62,
"learning_rate": 0.00019313941825476431,
"loss": 2.8036,
"theoretical_loss": 3.4249366379271713,
"tokens_seen": 2038628352
},
{
"epoch": 0.62,
"learning_rate": 0.0001931293881644935,
"loss": 2.5126,
"theoretical_loss": 3.4249274925628055,
"tokens_seen": 2038693888
},
{
"epoch": 0.62,
"learning_rate": 0.00019311935807422268,
"loss": 2.4754,
"theoretical_loss": 3.424918347574736,
"tokens_seen": 2038759424
},
{
"epoch": 0.62,
"learning_rate": 0.00019310932798395186,
"loss": 2.3994,
"theoretical_loss": 3.4249092029629353,
"tokens_seen": 2038824960
},
{
"epoch": 0.62,
"learning_rate": 0.00019309929789368107,
"loss": 2.6974,
"theoretical_loss": 3.424900058727376,
"tokens_seen": 2038890496
},
{
"epoch": 0.62,
"learning_rate": 0.00019308926780341025,
"loss": 2.5124,
"theoretical_loss": 3.4248909148680298,
"tokens_seen": 2038956032
},
{
"epoch": 0.62,
"learning_rate": 0.00019307923771313943,
"loss": 2.6429,
"theoretical_loss": 3.42488177138487,
"tokens_seen": 2039021568
},
{
"epoch": 0.62,
"learning_rate": 0.0001930692076228686,
"loss": 2.6916,
"theoretical_loss": 3.424872628277868,
"tokens_seen": 2039087104
},
{
"epoch": 0.62,
"learning_rate": 0.0001930591775325978,
"loss": 2.675,
"theoretical_loss": 3.4248634855469975,
"tokens_seen": 2039152640
},
{
"epoch": 0.62,
"learning_rate": 0.000193049147442327,
"loss": 2.5876,
"theoretical_loss": 3.42485434319223,
"tokens_seen": 2039218176
},
{
"epoch": 0.62,
"learning_rate": 0.00019303911735205618,
"loss": 2.7067,
"theoretical_loss": 3.4248452012135386,
"tokens_seen": 2039283712
},
{
"epoch": 0.62,
"learning_rate": 0.00019302908726178536,
"loss": 2.5438,
"theoretical_loss": 3.424836059610895,
"tokens_seen": 2039349248
},
{
"epoch": 0.62,
"learning_rate": 0.00019301905717151454,
"loss": 2.4181,
"theoretical_loss": 3.424826918384272,
"tokens_seen": 2039414784
},
{
"epoch": 0.62,
"learning_rate": 0.00019300902708124375,
"loss": 2.4998,
"theoretical_loss": 3.424817777533643,
"tokens_seen": 2039480320
},
{
"epoch": 0.62,
"learning_rate": 0.00019299899699097293,
"loss": 2.7014,
"theoretical_loss": 3.4248086370589785,
"tokens_seen": 2039545856
},
{
"epoch": 0.62,
"learning_rate": 0.0001929889669007021,
"loss": 2.457,
"theoretical_loss": 3.4247994969602527,
"tokens_seen": 2039611392
},
{
"epoch": 0.62,
"learning_rate": 0.0001929789368104313,
"loss": 2.5896,
"theoretical_loss": 3.4247903572374367,
"tokens_seen": 2039676928
},
{
"epoch": 0.62,
"learning_rate": 0.00019296890672016047,
"loss": 2.424,
"theoretical_loss": 3.424781217890504,
"tokens_seen": 2039742464
},
{
"epoch": 0.62,
"objective/train/docs_used": 1153219,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6190526485443115,
"objective/train/theoretical_loss": 3.4247720789194274,
"objective/train/tokens_used": 2060268000,
"theoretical_loss": 3.4247720789194274,
"tokens_seen": 2039808000
},
{
"epoch": 0.62,
"learning_rate": 0.00019295887662988968,
"loss": 2.5121,
"theoretical_loss": 3.4247720789194274,
"tokens_seen": 2039808000
},
{
"epoch": 0.62,
"learning_rate": 0.00019294884653961886,
"loss": 2.4417,
"theoretical_loss": 3.424762940324178,
"tokens_seen": 2039873536
},
{
"epoch": 0.62,
"learning_rate": 0.00019293881644934804,
"loss": 2.8336,
"theoretical_loss": 3.424753802104729,
"tokens_seen": 2039939072
},
{
"epoch": 0.62,
"learning_rate": 0.00019292878635907723,
"loss": 2.479,
"theoretical_loss": 3.4247446642610524,
"tokens_seen": 2040004608
},
{
"epoch": 0.62,
"learning_rate": 0.00019291875626880643,
"loss": 2.4775,
"theoretical_loss": 3.4247355267931217,
"tokens_seen": 2040070144
},
{
"epoch": 0.62,
"learning_rate": 0.00019290872617853562,
"loss": 2.6304,
"theoretical_loss": 3.4247263897009086,
"tokens_seen": 2040135680
},
{
"epoch": 0.62,
"learning_rate": 0.0001928986960882648,
"loss": 2.7086,
"theoretical_loss": 3.4247172529843857,
"tokens_seen": 2040201216
},
{
"epoch": 0.62,
"learning_rate": 0.00019288866599799398,
"loss": 2.445,
"theoretical_loss": 3.4247081166435254,
"tokens_seen": 2040266752
},
{
"epoch": 0.62,
"learning_rate": 0.00019287863590772316,
"loss": 2.6082,
"theoretical_loss": 3.4246989806783006,
"tokens_seen": 2040332288
},
{
"epoch": 0.62,
"learning_rate": 0.00019286860581745237,
"loss": 2.4628,
"theoretical_loss": 3.4246898450886833,
"tokens_seen": 2040397824
},
{
"epoch": 0.62,
"learning_rate": 0.00019285857572718155,
"loss": 2.5345,
"theoretical_loss": 3.4246807098746466,
"tokens_seen": 2040463360
},
{
"epoch": 0.62,
"learning_rate": 0.00019284854563691073,
"loss": 2.6229,
"theoretical_loss": 3.424671575036162,
"tokens_seen": 2040528896
},
{
"epoch": 0.62,
"learning_rate": 0.0001928385155466399,
"loss": 2.7211,
"theoretical_loss": 3.424662440573203,
"tokens_seen": 2040594432
},
{
"epoch": 0.62,
"learning_rate": 0.00019282848545636912,
"loss": 2.824,
"theoretical_loss": 3.4246533064857414,
"tokens_seen": 2040659968
},
{
"epoch": 0.62,
"learning_rate": 0.0001928184553660983,
"loss": 2.6643,
"theoretical_loss": 3.4246441727737507,
"tokens_seen": 2040725504
},
{
"epoch": 0.62,
"learning_rate": 0.00019280842527582748,
"loss": 2.7087,
"theoretical_loss": 3.424635039437202,
"tokens_seen": 2040791040
},
{
"epoch": 0.62,
"learning_rate": 0.00019279839518555666,
"loss": 2.7354,
"theoretical_loss": 3.4246259064760687,
"tokens_seen": 2040856576
},
{
"epoch": 0.62,
"learning_rate": 0.00019278836509528584,
"loss": 2.5844,
"theoretical_loss": 3.4246167738903233,
"tokens_seen": 2040922112
},
{
"epoch": 0.62,
"learning_rate": 0.00019277833500501505,
"loss": 2.7324,
"theoretical_loss": 3.424607641679938,
"tokens_seen": 2040987648
},
{
"epoch": 0.62,
"learning_rate": 0.00019276830491474423,
"loss": 2.6033,
"theoretical_loss": 3.424598509844886,
"tokens_seen": 2041053184
},
{
"epoch": 0.62,
"learning_rate": 0.0001927582748244734,
"loss": 2.6159,
"theoretical_loss": 3.4245893783851384,
"tokens_seen": 2041118720
},
{
"epoch": 0.62,
"learning_rate": 0.0001927482447342026,
"loss": 2.4945,
"theoretical_loss": 3.4245802473006695,
"tokens_seen": 2041184256
},
{
"epoch": 0.62,
"learning_rate": 0.0001927382146439318,
"loss": 2.586,
"theoretical_loss": 3.4245711165914505,
"tokens_seen": 2041249792
},
{
"epoch": 0.62,
"learning_rate": 0.00019272818455366098,
"loss": 2.7799,
"theoretical_loss": 3.4245619862574546,
"tokens_seen": 2041315328
},
{
"epoch": 0.62,
"learning_rate": 0.00019271815446339017,
"loss": 2.5622,
"theoretical_loss": 3.424552856298654,
"tokens_seen": 2041380864
},
{
"epoch": 0.62,
"objective/train/docs_used": 1154501,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.464268684387207,
"objective/train/theoretical_loss": 3.4245437267150214,
"objective/train/tokens_used": 2061906400,
"theoretical_loss": 3.4245437267150214,
"tokens_seen": 2041446400
},
{
"epoch": 0.62,
"learning_rate": 0.00019270812437311935,
"loss": 2.6052,
"theoretical_loss": 3.4245437267150214,
"tokens_seen": 2041446400
},
{
"epoch": 0.62,
"learning_rate": 0.00019269809428284853,
"loss": 2.6345,
"theoretical_loss": 3.4245345975065296,
"tokens_seen": 2041511936
},
{
"epoch": 0.62,
"learning_rate": 0.00019268806419257774,
"loss": 2.5677,
"theoretical_loss": 3.4245254686731506,
"tokens_seen": 2041577472
},
{
"epoch": 0.62,
"learning_rate": 0.00019267803410230692,
"loss": 2.6508,
"theoretical_loss": 3.424516340214857,
"tokens_seen": 2041643008
},
{
"epoch": 0.62,
"learning_rate": 0.0001926680040120361,
"loss": 2.7815,
"theoretical_loss": 3.424507212131622,
"tokens_seen": 2041708544
},
{
"epoch": 0.62,
"learning_rate": 0.00019265797392176528,
"loss": 2.8135,
"theoretical_loss": 3.4244980844234174,
"tokens_seen": 2041774080
},
{
"epoch": 0.62,
"learning_rate": 0.0001926479438314945,
"loss": 2.4182,
"theoretical_loss": 3.424488957090216,
"tokens_seen": 2041839616
},
{
"epoch": 0.62,
"learning_rate": 0.00019263791374122367,
"loss": 2.7736,
"theoretical_loss": 3.4244798301319905,
"tokens_seen": 2041905152
},
{
"epoch": 0.62,
"learning_rate": 0.00019262788365095285,
"loss": 2.5531,
"theoretical_loss": 3.424470703548714,
"tokens_seen": 2041970688
},
{
"epoch": 0.62,
"learning_rate": 0.00019261785356068203,
"loss": 2.6185,
"theoretical_loss": 3.4244615773403577,
"tokens_seen": 2042036224
},
{
"epoch": 0.62,
"learning_rate": 0.00019260782347041124,
"loss": 2.7454,
"theoretical_loss": 3.4244524515068946,
"tokens_seen": 2042101760
},
{
"epoch": 0.62,
"learning_rate": 0.00019259779338014042,
"loss": 2.5345,
"theoretical_loss": 3.4244433260482983,
"tokens_seen": 2042167296
},
{
"epoch": 0.62,
"learning_rate": 0.0001925877632898696,
"loss": 2.9425,
"theoretical_loss": 3.4244342009645403,
"tokens_seen": 2042232832
},
{
"epoch": 0.62,
"learning_rate": 0.00019257773319959878,
"loss": 2.7188,
"theoretical_loss": 3.4244250762555937,
"tokens_seen": 2042298368
},
{
"epoch": 0.62,
"learning_rate": 0.00019256770310932796,
"loss": 2.67,
"theoretical_loss": 3.424415951921431,
"tokens_seen": 2042363904
},
{
"epoch": 0.62,
"learning_rate": 0.0001925576730190572,
"loss": 2.5817,
"theoretical_loss": 3.4244068279620246,
"tokens_seen": 2042429440
},
{
"epoch": 0.62,
"learning_rate": 0.00019254764292878638,
"loss": 2.6052,
"theoretical_loss": 3.4243977043773466,
"tokens_seen": 2042494976
},
{
"epoch": 0.62,
"learning_rate": 0.00019253761283851556,
"loss": 2.6215,
"theoretical_loss": 3.4243885811673707,
"tokens_seen": 2042560512
},
{
"epoch": 0.62,
"learning_rate": 0.00019252758274824474,
"loss": 2.5968,
"theoretical_loss": 3.424379458332069,
"tokens_seen": 2042626048
},
{
"epoch": 0.62,
"learning_rate": 0.00019251755265797395,
"loss": 2.7998,
"theoretical_loss": 3.424370335871414,
"tokens_seen": 2042691584
},
{
"epoch": 0.62,
"learning_rate": 0.00019250752256770313,
"loss": 2.5462,
"theoretical_loss": 3.424361213785378,
"tokens_seen": 2042757120
},
{
"epoch": 0.62,
"learning_rate": 0.0001924974924774323,
"loss": 2.5581,
"theoretical_loss": 3.424352092073934,
"tokens_seen": 2042822656
},
{
"epoch": 0.62,
"learning_rate": 0.0001924874623871615,
"loss": 2.7203,
"theoretical_loss": 3.424342970737055,
"tokens_seen": 2042888192
},
{
"epoch": 0.62,
"learning_rate": 0.00019247743229689067,
"loss": 2.6954,
"theoretical_loss": 3.4243338497747127,
"tokens_seen": 2042953728
},
{
"epoch": 0.62,
"learning_rate": 0.00019246740220661988,
"loss": 2.5815,
"theoretical_loss": 3.42432472918688,
"tokens_seen": 2043019264
},
{
"epoch": 0.62,
"objective/train/docs_used": 1155127,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1598868370056152,
"objective/train/theoretical_loss": 3.4243156089735303,
"objective/train/tokens_used": 2063544800,
"theoretical_loss": 3.4243156089735303,
"tokens_seen": 2043084800
},
{
"epoch": 0.62,
"learning_rate": 0.00019245737211634906,
"loss": 2.7316,
"theoretical_loss": 3.4243156089735303,
"tokens_seen": 2043084800
},
{
"epoch": 0.62,
"learning_rate": 0.00019244734202607825,
"loss": 2.6577,
"theoretical_loss": 3.424306489134635,
"tokens_seen": 2043150336
},
{
"epoch": 0.62,
"learning_rate": 0.00019243731193580743,
"loss": 2.6119,
"theoretical_loss": 3.4242973696701675,
"tokens_seen": 2043215872
},
{
"epoch": 0.62,
"learning_rate": 0.00019242728184553663,
"loss": 2.805,
"theoretical_loss": 3.4242882505801,
"tokens_seen": 2043281408
},
{
"epoch": 0.62,
"learning_rate": 0.00019241725175526582,
"loss": 2.9693,
"theoretical_loss": 3.424279131864406,
"tokens_seen": 2043346944
},
{
"epoch": 0.62,
"learning_rate": 0.000192407221664995,
"loss": 2.623,
"theoretical_loss": 3.4242700135230564,
"tokens_seen": 2043412480
},
{
"epoch": 0.62,
"learning_rate": 0.00019239719157472418,
"loss": 2.6305,
"theoretical_loss": 3.424260895556025,
"tokens_seen": 2043478016
},
{
"epoch": 0.62,
"learning_rate": 0.00019238716148445336,
"loss": 2.724,
"theoretical_loss": 3.424251777963285,
"tokens_seen": 2043543552
},
{
"epoch": 0.62,
"learning_rate": 0.00019237713139418257,
"loss": 2.6783,
"theoretical_loss": 3.4242426607448078,
"tokens_seen": 2043609088
},
{
"epoch": 0.62,
"learning_rate": 0.00019236710130391175,
"loss": 2.3636,
"theoretical_loss": 3.424233543900567,
"tokens_seen": 2043674624
},
{
"epoch": 0.62,
"learning_rate": 0.00019235707121364093,
"loss": 2.5605,
"theoretical_loss": 3.4242244274305347,
"tokens_seen": 2043740160
},
{
"epoch": 0.62,
"learning_rate": 0.0001923470411233701,
"loss": 2.5675,
"theoretical_loss": 3.4242153113346836,
"tokens_seen": 2043805696
},
{
"epoch": 0.62,
"learning_rate": 0.00019233701103309932,
"loss": 2.6107,
"theoretical_loss": 3.424206195612986,
"tokens_seen": 2043871232
},
{
"epoch": 0.62,
"learning_rate": 0.0001923269809428285,
"loss": 2.6693,
"theoretical_loss": 3.424197080265415,
"tokens_seen": 2043936768
},
{
"epoch": 0.62,
"learning_rate": 0.00019231695085255768,
"loss": 2.7251,
"theoretical_loss": 3.424187965291944,
"tokens_seen": 2044002304
},
{
"epoch": 0.62,
"learning_rate": 0.00019230692076228686,
"loss": 2.736,
"theoretical_loss": 3.4241788506925444,
"tokens_seen": 2044067840
},
{
"epoch": 0.62,
"learning_rate": 0.00019229689067201604,
"loss": 2.4979,
"theoretical_loss": 3.4241697364671886,
"tokens_seen": 2044133376
},
{
"epoch": 0.62,
"learning_rate": 0.00019228686058174525,
"loss": 2.6138,
"theoretical_loss": 3.424160622615851,
"tokens_seen": 2044198912
},
{
"epoch": 0.62,
"learning_rate": 0.00019227683049147443,
"loss": 2.8631,
"theoretical_loss": 3.424151509138502,
"tokens_seen": 2044264448
},
{
"epoch": 0.62,
"learning_rate": 0.00019226680040120361,
"loss": 2.8781,
"theoretical_loss": 3.4241423960351165,
"tokens_seen": 2044329984
},
{
"epoch": 0.62,
"learning_rate": 0.0001922567703109328,
"loss": 2.7764,
"theoretical_loss": 3.4241332833056664,
"tokens_seen": 2044395520
},
{
"epoch": 0.62,
"learning_rate": 0.000192246740220662,
"loss": 2.6588,
"theoretical_loss": 3.4241241709501233,
"tokens_seen": 2044461056
},
{
"epoch": 0.62,
"learning_rate": 0.00019223671013039118,
"loss": 2.3583,
"theoretical_loss": 3.424115058968461,
"tokens_seen": 2044526592
},
{
"epoch": 0.62,
"learning_rate": 0.00019222668004012037,
"loss": 2.7006,
"theoretical_loss": 3.424105947360652,
"tokens_seen": 2044592128
},
{
"epoch": 0.62,
"learning_rate": 0.00019221664994984955,
"loss": 2.7727,
"theoretical_loss": 3.424096836126669,
"tokens_seen": 2044657664
},
{
"epoch": 0.62,
"objective/train/docs_used": 1156711,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.398145914077759,
"objective/train/theoretical_loss": 3.4240877252664834,
"objective/train/tokens_used": 2065183200,
"theoretical_loss": 3.4240877252664834,
"tokens_seen": 2044723200
},
{
"epoch": 0.62,
"learning_rate": 0.00019220661985957873,
"loss": 2.6944,
"theoretical_loss": 3.4240877252664834,
"tokens_seen": 2044723200
},
{
"epoch": 0.62,
"learning_rate": 0.00019219658976930794,
"loss": 2.7828,
"theoretical_loss": 3.42407861478007,
"tokens_seen": 2044788736
},
{
"epoch": 0.62,
"learning_rate": 0.00019218655967903712,
"loss": 2.6729,
"theoretical_loss": 3.4240695046674006,
"tokens_seen": 2044854272
},
{
"epoch": 0.62,
"learning_rate": 0.0001921765295887663,
"loss": 2.7758,
"theoretical_loss": 3.424060394928447,
"tokens_seen": 2044919808
},
{
"epoch": 0.62,
"learning_rate": 0.00019216649949849548,
"loss": 2.8586,
"theoretical_loss": 3.4240512855631833,
"tokens_seen": 2044985344
},
{
"epoch": 0.62,
"learning_rate": 0.0001921564694082247,
"loss": 2.6391,
"theoretical_loss": 3.424042176571582,
"tokens_seen": 2045050880
},
{
"epoch": 0.62,
"learning_rate": 0.00019214643931795387,
"loss": 2.7294,
"theoretical_loss": 3.424033067953615,
"tokens_seen": 2045116416
},
{
"epoch": 0.62,
"learning_rate": 0.00019213640922768305,
"loss": 2.6463,
"theoretical_loss": 3.4240239597092548,
"tokens_seen": 2045181952
},
{
"epoch": 0.62,
"learning_rate": 0.00019212637913741223,
"loss": 2.8163,
"theoretical_loss": 3.424014851838475,
"tokens_seen": 2045247488
},
{
"epoch": 0.62,
"learning_rate": 0.00019211634904714144,
"loss": 2.8199,
"theoretical_loss": 3.4240057443412484,
"tokens_seen": 2045313024
},
{
"epoch": 0.62,
"learning_rate": 0.00019210631895687062,
"loss": 2.5874,
"theoretical_loss": 3.423996637217547,
"tokens_seen": 2045378560
},
{
"epoch": 0.62,
"learning_rate": 0.0001920962888665998,
"loss": 2.7891,
"theoretical_loss": 3.4239875304673437,
"tokens_seen": 2045444096
},
{
"epoch": 0.62,
"learning_rate": 0.00019208625877632898,
"loss": 2.5731,
"theoretical_loss": 3.4239784240906115,
"tokens_seen": 2045509632
},
{
"epoch": 0.62,
"learning_rate": 0.00019207622868605816,
"loss": 2.7294,
"theoretical_loss": 3.423969318087323,
"tokens_seen": 2045575168
},
{
"epoch": 0.62,
"learning_rate": 0.00019206619859578737,
"loss": 2.5899,
"theoretical_loss": 3.423960212457451,
"tokens_seen": 2045640704
},
{
"epoch": 0.62,
"learning_rate": 0.00019205616850551655,
"loss": 2.9364,
"theoretical_loss": 3.423951107200968,
"tokens_seen": 2045706240
},
{
"epoch": 0.62,
"learning_rate": 0.00019204613841524573,
"loss": 2.6556,
"theoretical_loss": 3.423942002317846,
"tokens_seen": 2045771776
},
{
"epoch": 0.62,
"learning_rate": 0.00019203610832497491,
"loss": 2.5629,
"theoretical_loss": 3.4239328978080596,
"tokens_seen": 2045837312
},
{
"epoch": 0.62,
"learning_rate": 0.00019202607823470412,
"loss": 2.6445,
"theoretical_loss": 3.4239237936715803,
"tokens_seen": 2045902848
},
{
"epoch": 0.62,
"learning_rate": 0.0001920160481444333,
"loss": 2.7134,
"theoretical_loss": 3.4239146899083805,
"tokens_seen": 2045968384
},
{
"epoch": 0.62,
"learning_rate": 0.00019200601805416249,
"loss": 2.8017,
"theoretical_loss": 3.4239055865184342,
"tokens_seen": 2046033920
},
{
"epoch": 0.62,
"learning_rate": 0.00019199598796389167,
"loss": 2.5811,
"theoretical_loss": 3.423896483501713,
"tokens_seen": 2046099456
},
{
"epoch": 0.62,
"learning_rate": 0.00019198595787362085,
"loss": 2.3936,
"theoretical_loss": 3.42388738085819,
"tokens_seen": 2046164992
},
{
"epoch": 0.62,
"learning_rate": 0.00019197592778335006,
"loss": 2.6732,
"theoretical_loss": 3.423878278587838,
"tokens_seen": 2046230528
},
{
"epoch": 0.62,
"learning_rate": 0.00019196589769307924,
"loss": 2.8101,
"theoretical_loss": 3.42386917669063,
"tokens_seen": 2046296064
},
{
"epoch": 0.62,
"objective/train/docs_used": 1157531,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.84330153465271,
"objective/train/theoretical_loss": 3.4238600751665382,
"objective/train/tokens_used": 2066821600,
"theoretical_loss": 3.4238600751665382,
"tokens_seen": 2046361600
},
{
"epoch": 0.62,
"learning_rate": 0.00019195586760280842,
"loss": 2.6263,
"theoretical_loss": 3.4238600751665382,
"tokens_seen": 2046361600
},
{
"epoch": 0.62,
"learning_rate": 0.0001919458375125376,
"loss": 2.7822,
"theoretical_loss": 3.4238509740155356,
"tokens_seen": 2046427136
},
{
"epoch": 0.62,
"learning_rate": 0.0001919358074222668,
"loss": 2.5943,
"theoretical_loss": 3.4238418732375955,
"tokens_seen": 2046492672
},
{
"epoch": 0.62,
"learning_rate": 0.000191925777331996,
"loss": 2.8974,
"theoretical_loss": 3.42383277283269,
"tokens_seen": 2046558208
},
{
"epoch": 0.62,
"learning_rate": 0.00019191574724172517,
"loss": 2.5987,
"theoretical_loss": 3.4238236728007916,
"tokens_seen": 2046623744
},
{
"epoch": 0.62,
"learning_rate": 0.00019190571715145435,
"loss": 2.7413,
"theoretical_loss": 3.4238145731418737,
"tokens_seen": 2046689280
},
{
"epoch": 0.62,
"learning_rate": 0.00019189568706118353,
"loss": 2.4227,
"theoretical_loss": 3.4238054738559094,
"tokens_seen": 2046754816
},
{
"epoch": 0.62,
"learning_rate": 0.00019188565697091274,
"loss": 2.8275,
"theoretical_loss": 3.4237963749428704,
"tokens_seen": 2046820352
},
{
"epoch": 0.62,
"learning_rate": 0.00019187562688064192,
"loss": 2.8371,
"theoretical_loss": 3.4237872764027304,
"tokens_seen": 2046885888
},
{
"epoch": 0.62,
"learning_rate": 0.0001918655967903711,
"loss": 2.6842,
"theoretical_loss": 3.4237781782354615,
"tokens_seen": 2046951424
},
{
"epoch": 0.62,
"learning_rate": 0.00019185556670010028,
"loss": 2.6572,
"theoretical_loss": 3.423769080441037,
"tokens_seen": 2047016960
},
{
"epoch": 0.62,
"learning_rate": 0.0001918455366098295,
"loss": 2.7031,
"theoretical_loss": 3.423759983019429,
"tokens_seen": 2047082496
},
{
"epoch": 0.62,
"learning_rate": 0.00019183550651955867,
"loss": 2.8445,
"theoretical_loss": 3.4237508859706116,
"tokens_seen": 2047148032
},
{
"epoch": 0.62,
"learning_rate": 0.00019182547642928785,
"loss": 2.6003,
"theoretical_loss": 3.423741789294556,
"tokens_seen": 2047213568
},
{
"epoch": 0.62,
"learning_rate": 0.00019181544633901706,
"loss": 2.5834,
"theoretical_loss": 3.423732692991236,
"tokens_seen": 2047279104
},
{
"epoch": 0.62,
"learning_rate": 0.00019180541624874624,
"loss": 2.7113,
"theoretical_loss": 3.4237235970606243,
"tokens_seen": 2047344640
},
{
"epoch": 0.62,
"learning_rate": 0.00019179538615847545,
"loss": 2.5475,
"theoretical_loss": 3.4237145015026935,
"tokens_seen": 2047410176
},
{
"epoch": 0.62,
"learning_rate": 0.00019178535606820463,
"loss": 2.7632,
"theoretical_loss": 3.4237054063174166,
"tokens_seen": 2047475712
},
{
"epoch": 0.62,
"learning_rate": 0.00019177532597793381,
"loss": 2.628,
"theoretical_loss": 3.4236963115047656,
"tokens_seen": 2047541248
},
{
"epoch": 0.62,
"learning_rate": 0.000191765295887663,
"loss": 2.6519,
"theoretical_loss": 3.423687217064715,
"tokens_seen": 2047606784
},
{
"epoch": 0.62,
"learning_rate": 0.0001917552657973922,
"loss": 2.6431,
"theoretical_loss": 3.4236781229972353,
"tokens_seen": 2047672320
},
{
"epoch": 0.62,
"learning_rate": 0.00019174523570712138,
"loss": 2.6769,
"theoretical_loss": 3.4236690293023018,
"tokens_seen": 2047737856
},
{
"epoch": 0.62,
"learning_rate": 0.00019173520561685057,
"loss": 2.6589,
"theoretical_loss": 3.4236599359798854,
"tokens_seen": 2047803392
},
{
"epoch": 0.62,
"learning_rate": 0.00019172517552657975,
"loss": 2.8388,
"theoretical_loss": 3.42365084302996,
"tokens_seen": 2047868928
},
{
"epoch": 0.62,
"learning_rate": 0.00019171514543630893,
"loss": 2.6591,
"theoretical_loss": 3.4236417504524974,
"tokens_seen": 2047934464
},
{
"epoch": 0.62,
"objective/train/docs_used": 1159006,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.789802074432373,
"objective/train/theoretical_loss": 3.4236326582474716,
"objective/train/tokens_used": 2068460000,
"theoretical_loss": 3.4236326582474716,
"tokens_seen": 2048000000
},
{
"epoch": 0.62,
"learning_rate": 0.00019170511534603814,
"loss": 2.784,
"theoretical_loss": 3.4236326582474716,
"tokens_seen": 2048000000
},
{
"epoch": 0.62,
"learning_rate": 0.00019169508525576732,
"loss": 2.5171,
"theoretical_loss": 3.4236235664148547,
"tokens_seen": 2048065536
},
{
"epoch": 0.62,
"learning_rate": 0.0001916850551654965,
"loss": 2.4609,
"theoretical_loss": 3.42361447495462,
"tokens_seen": 2048131072
},
{
"epoch": 0.62,
"learning_rate": 0.00019167502507522568,
"loss": 2.5107,
"theoretical_loss": 3.4236053838667395,
"tokens_seen": 2048196608
},
{
"epoch": 0.62,
"learning_rate": 0.0001916649949849549,
"loss": 2.6838,
"theoretical_loss": 3.4235962931511867,
"tokens_seen": 2048262144
},
{
"epoch": 0.62,
"learning_rate": 0.00019165496489468407,
"loss": 2.2762,
"theoretical_loss": 3.4235872028079353,
"tokens_seen": 2048327680
},
{
"epoch": 0.62,
"learning_rate": 0.00019164493480441325,
"loss": 2.6733,
"theoretical_loss": 3.423578112836956,
"tokens_seen": 2048393216
},
{
"epoch": 0.62,
"learning_rate": 0.00019163490471414243,
"loss": 2.4983,
"theoretical_loss": 3.4235690232382234,
"tokens_seen": 2048458752
},
{
"epoch": 0.62,
"learning_rate": 0.00019162487462387164,
"loss": 2.6129,
"theoretical_loss": 3.4235599340117098,
"tokens_seen": 2048524288
},
{
"epoch": 0.62,
"learning_rate": 0.00019161484453360082,
"loss": 2.5187,
"theoretical_loss": 3.4235508451573873,
"tokens_seen": 2048589824
},
{
"epoch": 0.62,
"learning_rate": 0.00019160481444333,
"loss": 2.6603,
"theoretical_loss": 3.42354175667523,
"tokens_seen": 2048655360
},
{
"epoch": 0.62,
"learning_rate": 0.00019159478435305918,
"loss": 2.6708,
"theoretical_loss": 3.42353266856521,
"tokens_seen": 2048720896
},
{
"epoch": 0.62,
"learning_rate": 0.00019158475426278836,
"loss": 2.3095,
"theoretical_loss": 3.4235235808273012,
"tokens_seen": 2048786432
},
{
"epoch": 0.62,
"learning_rate": 0.00019157472417251757,
"loss": 2.7056,
"theoretical_loss": 3.4235144934614747,
"tokens_seen": 2048851968
},
{
"epoch": 0.62,
"learning_rate": 0.00019156469408224675,
"loss": 2.3348,
"theoretical_loss": 3.423505406467705,
"tokens_seen": 2048917504
},
{
"epoch": 0.62,
"learning_rate": 0.00019155466399197593,
"loss": 2.7324,
"theoretical_loss": 3.4234963198459636,
"tokens_seen": 2048983040
},
{
"epoch": 0.62,
"learning_rate": 0.00019154463390170512,
"loss": 2.7702,
"theoretical_loss": 3.4234872335962248,
"tokens_seen": 2049048576
},
{
"epoch": 0.62,
"learning_rate": 0.00019153460381143432,
"loss": 2.573,
"theoretical_loss": 3.42347814771846,
"tokens_seen": 2049114112
},
{
"epoch": 0.62,
"learning_rate": 0.0001915245737211635,
"loss": 2.6557,
"theoretical_loss": 3.423469062212643,
"tokens_seen": 2049179648
},
{
"epoch": 0.62,
"learning_rate": 0.00019151454363089269,
"loss": 2.4851,
"theoretical_loss": 3.4234599770787466,
"tokens_seen": 2049245184
},
{
"epoch": 0.62,
"learning_rate": 0.00019150451354062187,
"loss": 2.8064,
"theoretical_loss": 3.4234508923167435,
"tokens_seen": 2049310720
},
{
"epoch": 0.62,
"learning_rate": 0.00019149448345035105,
"loss": 2.7087,
"theoretical_loss": 3.423441807926607,
"tokens_seen": 2049376256
},
{
"epoch": 0.62,
"learning_rate": 0.00019148445336008026,
"loss": 2.5187,
"theoretical_loss": 3.423432723908309,
"tokens_seen": 2049441792
},
{
"epoch": 0.62,
"learning_rate": 0.00019147442326980944,
"loss": 2.5419,
"theoretical_loss": 3.423423640261823,
"tokens_seen": 2049507328
},
{
"epoch": 0.62,
"learning_rate": 0.00019146439317953862,
"loss": 2.6645,
"theoretical_loss": 3.4234145569871224,
"tokens_seen": 2049572864
},
{
"epoch": 0.62,
"objective/train/docs_used": 1159734,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4614672660827637,
"objective/train/theoretical_loss": 3.4234054740841797,
"objective/train/tokens_used": 2070098400,
"theoretical_loss": 3.4234054740841797,
"tokens_seen": 2049638400
},
{
"epoch": 0.62,
"learning_rate": 0.0001914543630892678,
"loss": 2.5905,
"theoretical_loss": 3.4234054740841797,
"tokens_seen": 2049638400
},
{
"epoch": 0.62,
"learning_rate": 0.000191444332998997,
"loss": 2.6779,
"theoretical_loss": 3.4233963915529673,
"tokens_seen": 2049703936
},
{
"epoch": 0.62,
"learning_rate": 0.0001914343029087262,
"loss": 2.5203,
"theoretical_loss": 3.4233873093934584,
"tokens_seen": 2049769472
},
{
"epoch": 0.62,
"learning_rate": 0.00019142427281845537,
"loss": 2.609,
"theoretical_loss": 3.4233782276056264,
"tokens_seen": 2049835008
},
{
"epoch": 0.62,
"learning_rate": 0.00019141424272818455,
"loss": 2.4339,
"theoretical_loss": 3.4233691461894438,
"tokens_seen": 2049900544
},
{
"epoch": 0.62,
"learning_rate": 0.00019140421263791373,
"loss": 2.6849,
"theoretical_loss": 3.4233600651448834,
"tokens_seen": 2049966080
},
{
"epoch": 0.62,
"learning_rate": 0.00019139418254764294,
"loss": 2.7758,
"theoretical_loss": 3.4233509844719183,
"tokens_seen": 2050031616
},
{
"epoch": 0.62,
"learning_rate": 0.00019138415245737212,
"loss": 2.6908,
"theoretical_loss": 3.4233419041705213,
"tokens_seen": 2050097152
},
{
"epoch": 0.62,
"learning_rate": 0.0001913741223671013,
"loss": 2.6223,
"theoretical_loss": 3.4233328242406653,
"tokens_seen": 2050162688
},
{
"epoch": 0.62,
"learning_rate": 0.00019136409227683048,
"loss": 2.7096,
"theoretical_loss": 3.4233237446823237,
"tokens_seen": 2050228224
},
{
"epoch": 0.62,
"learning_rate": 0.0001913540621865597,
"loss": 2.7091,
"theoretical_loss": 3.4233146654954685,
"tokens_seen": 2050293760
},
{
"epoch": 0.62,
"learning_rate": 0.00019134403209628887,
"loss": 2.8327,
"theoretical_loss": 3.4233055866800735,
"tokens_seen": 2050359296
},
{
"epoch": 0.62,
"learning_rate": 0.00019133400200601805,
"loss": 2.6935,
"theoretical_loss": 3.4232965082361106,
"tokens_seen": 2050424832
},
{
"epoch": 0.62,
"learning_rate": 0.00019132397191574724,
"loss": 2.6861,
"theoretical_loss": 3.4232874301635543,
"tokens_seen": 2050490368
},
{
"epoch": 0.62,
"learning_rate": 0.00019131394182547642,
"loss": 2.745,
"theoretical_loss": 3.4232783524623764,
"tokens_seen": 2050555904
},
{
"epoch": 0.62,
"learning_rate": 0.00019130391173520562,
"loss": 2.5679,
"theoretical_loss": 3.42326927513255,
"tokens_seen": 2050621440
},
{
"epoch": 0.62,
"learning_rate": 0.0001912938816449348,
"loss": 2.5704,
"theoretical_loss": 3.423260198174048,
"tokens_seen": 2050686976
},
{
"epoch": 0.62,
"learning_rate": 0.000191283851554664,
"loss": 2.7086,
"theoretical_loss": 3.4232511215868433,
"tokens_seen": 2050752512
},
{
"epoch": 0.62,
"learning_rate": 0.00019127382146439317,
"loss": 2.6658,
"theoretical_loss": 3.42324204537091,
"tokens_seen": 2050818048
},
{
"epoch": 0.62,
"learning_rate": 0.00019126379137412238,
"loss": 2.644,
"theoretical_loss": 3.423232969526219,
"tokens_seen": 2050883584
},
{
"epoch": 0.62,
"learning_rate": 0.00019125376128385156,
"loss": 2.7304,
"theoretical_loss": 3.423223894052745,
"tokens_seen": 2050949120
},
{
"epoch": 0.62,
"learning_rate": 0.00019124373119358074,
"loss": 2.5961,
"theoretical_loss": 3.4232148189504596,
"tokens_seen": 2051014656
},
{
"epoch": 0.62,
"learning_rate": 0.00019123370110330992,
"loss": 2.705,
"theoretical_loss": 3.423205744219337,
"tokens_seen": 2051080192
},
{
"epoch": 0.62,
"learning_rate": 0.0001912236710130391,
"loss": 2.4883,
"theoretical_loss": 3.4231966698593492,
"tokens_seen": 2051145728
},
{
"epoch": 0.62,
"learning_rate": 0.0001912136409227683,
"loss": 2.6158,
"theoretical_loss": 3.42318759587047,
"tokens_seen": 2051211264
},
{
"epoch": 0.62,
"objective/train/docs_used": 1160185,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.753434419631958,
"objective/train/theoretical_loss": 3.423178522252672,
"objective/train/tokens_used": 2071736800,
"theoretical_loss": 3.423178522252672,
"tokens_seen": 2051276800
},
{
"epoch": 0.62,
"learning_rate": 0.0001912036108324975,
"loss": 2.4551,
"theoretical_loss": 3.423178522252672,
"tokens_seen": 2051276800
},
{
"epoch": 0.62,
"learning_rate": 0.00019119358074222667,
"loss": 2.6507,
"theoretical_loss": 3.4231694490059272,
"tokens_seen": 2051342336
},
{
"epoch": 0.62,
"learning_rate": 0.00019118355065195585,
"loss": 2.5683,
"theoretical_loss": 3.4231603761302107,
"tokens_seen": 2051407872
},
{
"epoch": 0.62,
"learning_rate": 0.00019117352056168506,
"loss": 2.6095,
"theoretical_loss": 3.4231513036254935,
"tokens_seen": 2051473408
},
{
"epoch": 0.62,
"learning_rate": 0.00019116349047141424,
"loss": 2.5899,
"theoretical_loss": 3.4231422314917497,
"tokens_seen": 2051538944
},
{
"epoch": 0.62,
"learning_rate": 0.00019115346038114342,
"loss": 2.6045,
"theoretical_loss": 3.423133159728951,
"tokens_seen": 2051604480
},
{
"epoch": 0.62,
"learning_rate": 0.0001911434302908726,
"loss": 2.5484,
"theoretical_loss": 3.4231240883370724,
"tokens_seen": 2051670016
},
{
"epoch": 0.62,
"learning_rate": 0.00019113340020060178,
"loss": 2.4996,
"theoretical_loss": 3.423115017316085,
"tokens_seen": 2051735552
},
{
"epoch": 0.62,
"learning_rate": 0.000191123370110331,
"loss": 2.669,
"theoretical_loss": 3.423105946665963,
"tokens_seen": 2051801088
},
{
"epoch": 0.62,
"learning_rate": 0.00019111334002006017,
"loss": 2.5772,
"theoretical_loss": 3.423096876386679,
"tokens_seen": 2051866624
},
{
"epoch": 0.62,
"learning_rate": 0.00019110330992978936,
"loss": 2.6502,
"theoretical_loss": 3.4230878064782058,
"tokens_seen": 2051932160
},
{
"epoch": 0.62,
"learning_rate": 0.00019109327983951854,
"loss": 2.5187,
"theoretical_loss": 3.4230787369405165,
"tokens_seen": 2051997696
},
{
"epoch": 0.62,
"learning_rate": 0.00019108324974924774,
"loss": 2.4233,
"theoretical_loss": 3.4230696677735843,
"tokens_seen": 2052063232
},
{
"epoch": 0.62,
"learning_rate": 0.00019107321965897693,
"loss": 2.3449,
"theoretical_loss": 3.423060598977382,
"tokens_seen": 2052128768
},
{
"epoch": 0.62,
"learning_rate": 0.00019106318956870613,
"loss": 2.7475,
"theoretical_loss": 3.4230515305518825,
"tokens_seen": 2052194304
},
{
"epoch": 0.62,
"learning_rate": 0.00019105315947843532,
"loss": 2.8176,
"theoretical_loss": 3.423042462497059,
"tokens_seen": 2052259840
},
{
"epoch": 0.62,
"learning_rate": 0.00019104312938816452,
"loss": 2.6732,
"theoretical_loss": 3.423033394812885,
"tokens_seen": 2052325376
},
{
"epoch": 0.62,
"learning_rate": 0.0001910330992978937,
"loss": 2.9159,
"theoretical_loss": 3.4230243274993324,
"tokens_seen": 2052390912
},
{
"epoch": 0.62,
"learning_rate": 0.00019102306920762289,
"loss": 2.5307,
"theoretical_loss": 3.423015260556375,
"tokens_seen": 2052456448
},
{
"epoch": 0.62,
"learning_rate": 0.00019101303911735207,
"loss": 2.6533,
"theoretical_loss": 3.4230061939839858,
"tokens_seen": 2052521984
},
{
"epoch": 0.62,
"learning_rate": 0.00019100300902708125,
"loss": 2.5328,
"theoretical_loss": 3.4229971277821374,
"tokens_seen": 2052587520
},
{
"epoch": 0.62,
"learning_rate": 0.00019099297893681046,
"loss": 2.8518,
"theoretical_loss": 3.4229880619508033,
"tokens_seen": 2052653056
},
{
"epoch": 0.62,
"learning_rate": 0.00019098294884653964,
"loss": 2.7951,
"theoretical_loss": 3.422978996489956,
"tokens_seen": 2052718592
},
{
"epoch": 0.62,
"learning_rate": 0.00019097291875626882,
"loss": 2.3702,
"theoretical_loss": 3.422969931399569,
"tokens_seen": 2052784128
},
{
"epoch": 0.62,
"learning_rate": 0.000190962888665998,
"loss": 2.693,
"theoretical_loss": 3.422960866679616,
"tokens_seen": 2052849664
},
{
"epoch": 0.62,
"objective/train/docs_used": 1161710,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.017590284347534,
"objective/train/theoretical_loss": 3.422951802330068,
"objective/train/tokens_used": 2073375200,
"theoretical_loss": 3.422951802330068,
"tokens_seen": 2052915200
},
{
"epoch": 0.62,
"learning_rate": 0.0001909528585757272,
"loss": 2.2831,
"theoretical_loss": 3.422951802330068,
"tokens_seen": 2052915200
},
{
"epoch": 0.62,
"learning_rate": 0.0001909428284854564,
"loss": 2.7414,
"theoretical_loss": 3.4229427383508995,
"tokens_seen": 2052980736
},
{
"epoch": 0.62,
"learning_rate": 0.00019093279839518557,
"loss": 2.7106,
"theoretical_loss": 3.422933674742084,
"tokens_seen": 2053046272
},
{
"epoch": 0.62,
"learning_rate": 0.00019092276830491475,
"loss": 2.5065,
"theoretical_loss": 3.422924611503593,
"tokens_seen": 2053111808
},
{
"epoch": 0.62,
"learning_rate": 0.00019091273821464393,
"loss": 2.8004,
"theoretical_loss": 3.422915548635401,
"tokens_seen": 2053177344
},
{
"epoch": 0.62,
"learning_rate": 0.00019090270812437314,
"loss": 2.4745,
"theoretical_loss": 3.4229064861374803,
"tokens_seen": 2053242880
},
{
"epoch": 0.62,
"learning_rate": 0.00019089267803410232,
"loss": 2.4718,
"theoretical_loss": 3.422897424009804,
"tokens_seen": 2053308416
},
{
"epoch": 0.62,
"learning_rate": 0.0001908826479438315,
"loss": 2.7257,
"theoretical_loss": 3.422888362252345,
"tokens_seen": 2053373952
},
{
"epoch": 0.62,
"learning_rate": 0.00019087261785356068,
"loss": 2.8535,
"theoretical_loss": 3.4228793008650773,
"tokens_seen": 2053439488
},
{
"epoch": 0.62,
"learning_rate": 0.0001908625877632899,
"loss": 2.4114,
"theoretical_loss": 3.422870239847973,
"tokens_seen": 2053505024
},
{
"epoch": 0.62,
"learning_rate": 0.00019085255767301907,
"loss": 2.449,
"theoretical_loss": 3.4228611792010053,
"tokens_seen": 2053570560
},
{
"epoch": 0.62,
"learning_rate": 0.00019084252758274825,
"loss": 2.694,
"theoretical_loss": 3.4228521189241476,
"tokens_seen": 2053636096
},
{
"epoch": 0.62,
"learning_rate": 0.00019083249749247744,
"loss": 2.4213,
"theoretical_loss": 3.4228430590173726,
"tokens_seen": 2053701632
},
{
"epoch": 0.62,
"learning_rate": 0.00019082246740220662,
"loss": 2.724,
"theoretical_loss": 3.4228339994806536,
"tokens_seen": 2053767168
},
{
"epoch": 0.62,
"learning_rate": 0.00019081243731193582,
"loss": 2.7756,
"theoretical_loss": 3.422824940313964,
"tokens_seen": 2053832704
},
{
"epoch": 0.62,
"learning_rate": 0.000190802407221665,
"loss": 2.8274,
"theoretical_loss": 3.422815881517276,
"tokens_seen": 2053898240
},
{
"epoch": 0.62,
"learning_rate": 0.0001907923771313942,
"loss": 2.3091,
"theoretical_loss": 3.4228068230905633,
"tokens_seen": 2053963776
},
{
"epoch": 0.62,
"learning_rate": 0.00019078234704112337,
"loss": 2.5254,
"theoretical_loss": 3.422797765033799,
"tokens_seen": 2054029312
},
{
"epoch": 0.62,
"learning_rate": 0.00019077231695085258,
"loss": 2.6842,
"theoretical_loss": 3.4227887073469563,
"tokens_seen": 2054094848
},
{
"epoch": 0.62,
"learning_rate": 0.00019076228686058176,
"loss": 2.5255,
"theoretical_loss": 3.422779650030008,
"tokens_seen": 2054160384
},
{
"epoch": 0.62,
"learning_rate": 0.00019075225677031094,
"loss": 2.7112,
"theoretical_loss": 3.422770593082927,
"tokens_seen": 2054225920
},
{
"epoch": 0.62,
"learning_rate": 0.00019074222668004012,
"loss": 2.7683,
"theoretical_loss": 3.422761536505687,
"tokens_seen": 2054291456
},
{
"epoch": 0.62,
"learning_rate": 0.0001907321965897693,
"loss": 2.4191,
"theoretical_loss": 3.4227524802982607,
"tokens_seen": 2054356992
},
{
"epoch": 0.62,
"learning_rate": 0.0001907221664994985,
"loss": 2.6046,
"theoretical_loss": 3.422743424460621,
"tokens_seen": 2054422528
},
{
"epoch": 0.62,
"learning_rate": 0.0001907121364092277,
"loss": 2.6386,
"theoretical_loss": 3.4227343689927414,
"tokens_seen": 2054488064
},
{
"epoch": 0.62,
"objective/train/docs_used": 1162341,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1584722995758057,
"objective/train/theoretical_loss": 3.4227253138945946,
"objective/train/tokens_used": 2075013600,
"theoretical_loss": 3.4227253138945946,
"tokens_seen": 2054553600
},
{
"epoch": 0.62,
"learning_rate": 0.00019070210631895687,
"loss": 2.417,
"theoretical_loss": 3.4227253138945946,
"tokens_seen": 2054553600
},
{
"epoch": 0.62,
"learning_rate": 0.00019069207622868605,
"loss": 2.5468,
"theoretical_loss": 3.4227162591661546,
"tokens_seen": 2054619136
},
{
"epoch": 0.62,
"learning_rate": 0.00019068204613841526,
"loss": 2.6082,
"theoretical_loss": 3.422707204807393,
"tokens_seen": 2054684672
},
{
"epoch": 0.62,
"learning_rate": 0.00019067201604814444,
"loss": 2.431,
"theoretical_loss": 3.4226981508182845,
"tokens_seen": 2054750208
},
{
"epoch": 0.62,
"learning_rate": 0.00019066198595787362,
"loss": 2.9812,
"theoretical_loss": 3.422689097198802,
"tokens_seen": 2054815744
},
{
"epoch": 0.62,
"learning_rate": 0.0001906519558676028,
"loss": 2.5801,
"theoretical_loss": 3.422680043948917,
"tokens_seen": 2054881280
},
{
"epoch": 0.62,
"learning_rate": 0.00019064192577733199,
"loss": 2.5917,
"theoretical_loss": 3.422670991068605,
"tokens_seen": 2054946816
},
{
"epoch": 0.62,
"learning_rate": 0.0001906318956870612,
"loss": 2.6198,
"theoretical_loss": 3.422661938557837,
"tokens_seen": 2055012352
},
{
"epoch": 0.62,
"learning_rate": 0.00019062186559679037,
"loss": 2.6007,
"theoretical_loss": 3.4226528864165875,
"tokens_seen": 2055077888
},
{
"epoch": 0.62,
"learning_rate": 0.00019061183550651956,
"loss": 2.5528,
"theoretical_loss": 3.4226438346448287,
"tokens_seen": 2055143424
},
{
"epoch": 0.62,
"learning_rate": 0.00019060180541624874,
"loss": 2.4108,
"theoretical_loss": 3.422634783242535,
"tokens_seen": 2055208960
},
{
"epoch": 0.62,
"learning_rate": 0.00019059177532597794,
"loss": 2.5183,
"theoretical_loss": 3.422625732209678,
"tokens_seen": 2055274496
},
{
"epoch": 0.62,
"learning_rate": 0.00019058174523570713,
"loss": 2.8544,
"theoretical_loss": 3.4226166815462316,
"tokens_seen": 2055340032
},
{
"epoch": 0.62,
"learning_rate": 0.0001905717151454363,
"loss": 2.5815,
"theoretical_loss": 3.42260763125217,
"tokens_seen": 2055405568
},
{
"epoch": 0.62,
"learning_rate": 0.0001905616850551655,
"loss": 2.6848,
"theoretical_loss": 3.422598581327464,
"tokens_seen": 2055471104
},
{
"epoch": 0.62,
"learning_rate": 0.00019055165496489467,
"loss": 2.6285,
"theoretical_loss": 3.4225895317720885,
"tokens_seen": 2055536640
},
{
"epoch": 0.62,
"learning_rate": 0.00019054162487462388,
"loss": 2.6907,
"theoretical_loss": 3.4225804825860164,
"tokens_seen": 2055602176
},
{
"epoch": 0.62,
"learning_rate": 0.00019053159478435306,
"loss": 2.5329,
"theoretical_loss": 3.42257143376922,
"tokens_seen": 2055667712
},
{
"epoch": 0.62,
"learning_rate": 0.00019052156469408224,
"loss": 2.4378,
"theoretical_loss": 3.422562385321674,
"tokens_seen": 2055733248
},
{
"epoch": 0.62,
"learning_rate": 0.00019051153460381142,
"loss": 2.4771,
"theoretical_loss": 3.4225533372433503,
"tokens_seen": 2055798784
},
{
"epoch": 0.62,
"learning_rate": 0.00019050150451354063,
"loss": 2.5765,
"theoretical_loss": 3.422544289534222,
"tokens_seen": 2055864320
},
{
"epoch": 0.62,
"learning_rate": 0.0001904914744232698,
"loss": 2.3323,
"theoretical_loss": 3.422535242194263,
"tokens_seen": 2055929856
},
{
"epoch": 0.62,
"learning_rate": 0.000190481444332999,
"loss": 2.695,
"theoretical_loss": 3.4225261952234467,
"tokens_seen": 2055995392
},
{
"epoch": 0.62,
"learning_rate": 0.00019047141424272817,
"loss": 2.5557,
"theoretical_loss": 3.422517148621745,
"tokens_seen": 2056060928
},
{
"epoch": 0.62,
"learning_rate": 0.00019046138415245738,
"loss": 2.7329,
"theoretical_loss": 3.4225081023891324,
"tokens_seen": 2056126464
},
{
"epoch": 0.62,
"objective/train/docs_used": 1163822,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9361259937286377,
"objective/train/theoretical_loss": 3.422499056525581,
"objective/train/tokens_used": 2076652000,
"theoretical_loss": 3.422499056525581,
"tokens_seen": 2056192000
},
{
"epoch": 0.62,
"learning_rate": 0.00019045135406218656,
"loss": 2.8898,
"theoretical_loss": 3.422499056525581,
"tokens_seen": 2056192000
},
{
"epoch": 0.62,
"learning_rate": 0.00019044132397191574,
"loss": 2.6842,
"theoretical_loss": 3.4224900110310648,
"tokens_seen": 2056257536
},
{
"epoch": 0.62,
"learning_rate": 0.00019043129388164492,
"loss": 2.69,
"theoretical_loss": 3.422480965905556,
"tokens_seen": 2056323072
},
{
"epoch": 0.62,
"learning_rate": 0.0001904212637913741,
"loss": 2.503,
"theoretical_loss": 3.4224719211490293,
"tokens_seen": 2056388608
},
{
"epoch": 0.62,
"learning_rate": 0.0001904112337011033,
"loss": 2.5064,
"theoretical_loss": 3.4224628767614567,
"tokens_seen": 2056454144
},
{
"epoch": 0.62,
"learning_rate": 0.0001904012036108325,
"loss": 2.6291,
"theoretical_loss": 3.4224538327428116,
"tokens_seen": 2056519680
},
{
"epoch": 0.62,
"learning_rate": 0.00019039117352056168,
"loss": 2.7908,
"theoretical_loss": 3.4224447890930674,
"tokens_seen": 2056585216
},
{
"epoch": 0.62,
"learning_rate": 0.00019038114343029086,
"loss": 2.582,
"theoretical_loss": 3.422435745812197,
"tokens_seen": 2056650752
},
{
"epoch": 0.62,
"learning_rate": 0.00019037111334002007,
"loss": 2.6876,
"theoretical_loss": 3.4224267029001743,
"tokens_seen": 2056716288
},
{
"epoch": 0.62,
"learning_rate": 0.00019036108324974925,
"loss": 2.6431,
"theoretical_loss": 3.422417660356972,
"tokens_seen": 2056781824
},
{
"epoch": 0.62,
"learning_rate": 0.00019035105315947843,
"loss": 2.6515,
"theoretical_loss": 3.422408618182563,
"tokens_seen": 2056847360
},
{
"epoch": 0.62,
"learning_rate": 0.0001903410230692076,
"loss": 2.6275,
"theoretical_loss": 3.422399576376921,
"tokens_seen": 2056912896
},
{
"epoch": 0.62,
"learning_rate": 0.0001903309929789368,
"loss": 2.587,
"theoretical_loss": 3.422390534940019,
"tokens_seen": 2056978432
},
{
"epoch": 0.62,
"learning_rate": 0.000190320962888666,
"loss": 2.5196,
"theoretical_loss": 3.42238149387183,
"tokens_seen": 2057043968
},
{
"epoch": 0.62,
"learning_rate": 0.0001903109327983952,
"loss": 2.4764,
"theoretical_loss": 3.4223724531723283,
"tokens_seen": 2057109504
},
{
"epoch": 0.62,
"learning_rate": 0.0001903009027081244,
"loss": 2.6486,
"theoretical_loss": 3.4223634128414853,
"tokens_seen": 2057175040
},
{
"epoch": 0.62,
"learning_rate": 0.00019029087261785357,
"loss": 2.6248,
"theoretical_loss": 3.422354372879276,
"tokens_seen": 2057240576
},
{
"epoch": 0.62,
"learning_rate": 0.00019028084252758278,
"loss": 2.7329,
"theoretical_loss": 3.4223453332856724,
"tokens_seen": 2057306112
},
{
"epoch": 0.62,
"learning_rate": 0.00019027081243731196,
"loss": 2.6376,
"theoretical_loss": 3.4223362940606483,
"tokens_seen": 2057371648
},
{
"epoch": 0.62,
"learning_rate": 0.00019026078234704114,
"loss": 2.4595,
"theoretical_loss": 3.422327255204177,
"tokens_seen": 2057437184
},
{
"epoch": 0.62,
"learning_rate": 0.00019025075225677032,
"loss": 2.8023,
"theoretical_loss": 3.4223182167162314,
"tokens_seen": 2057502720
},
{
"epoch": 0.62,
"learning_rate": 0.0001902407221664995,
"loss": 2.6158,
"theoretical_loss": 3.4223091785967847,
"tokens_seen": 2057568256
},
{
"epoch": 0.62,
"learning_rate": 0.0001902306920762287,
"loss": 2.5756,
"theoretical_loss": 3.4223001408458105,
"tokens_seen": 2057633792
},
{
"epoch": 0.62,
"learning_rate": 0.0001902206619859579,
"loss": 2.6575,
"theoretical_loss": 3.422291103463282,
"tokens_seen": 2057699328
},
{
"epoch": 0.62,
"learning_rate": 0.00019021063189568707,
"loss": 2.4445,
"theoretical_loss": 3.422282066449172,
"tokens_seen": 2057764864
},
{
"epoch": 0.62,
"objective/train/docs_used": 1164494,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1543006896972656,
"objective/train/theoretical_loss": 3.4222730298034545,
"objective/train/tokens_used": 2078290400,
"theoretical_loss": 3.4222730298034545,
"tokens_seen": 2057830400
},
{
"epoch": 0.62,
"learning_rate": 0.00019020060180541625,
"loss": 2.6993,
"theoretical_loss": 3.4222730298034545,
"tokens_seen": 2057830400
},
{
"epoch": 0.62,
"learning_rate": 0.00019019057171514546,
"loss": 2.5665,
"theoretical_loss": 3.4222639935261014,
"tokens_seen": 2057895936
},
{
"epoch": 0.62,
"learning_rate": 0.00019018054162487464,
"loss": 2.7336,
"theoretical_loss": 3.4222549576170875,
"tokens_seen": 2057961472
},
{
"epoch": 0.62,
"learning_rate": 0.00019017051153460382,
"loss": 2.7858,
"theoretical_loss": 3.422245922076386,
"tokens_seen": 2058027008
},
{
"epoch": 0.62,
"learning_rate": 0.000190160481444333,
"loss": 2.525,
"theoretical_loss": 3.4222368869039688,
"tokens_seen": 2058092544
},
{
"epoch": 0.62,
"learning_rate": 0.00019015045135406219,
"loss": 2.7094,
"theoretical_loss": 3.42222785209981,
"tokens_seen": 2058158080
},
{
"epoch": 0.62,
"learning_rate": 0.0001901404212637914,
"loss": 2.6946,
"theoretical_loss": 3.4222188176638833,
"tokens_seen": 2058223616
},
{
"epoch": 0.62,
"learning_rate": 0.00019013039117352057,
"loss": 2.6014,
"theoretical_loss": 3.422209783596161,
"tokens_seen": 2058289152
},
{
"epoch": 0.62,
"learning_rate": 0.00019012036108324976,
"loss": 2.5161,
"theoretical_loss": 3.422200749896617,
"tokens_seen": 2058354688
},
{
"epoch": 0.62,
"learning_rate": 0.00019011033099297894,
"loss": 2.6768,
"theoretical_loss": 3.4221917165652243,
"tokens_seen": 2058420224
},
{
"epoch": 0.62,
"learning_rate": 0.00019010030090270815,
"loss": 2.7023,
"theoretical_loss": 3.4221826836019567,
"tokens_seen": 2058485760
},
{
"epoch": 0.62,
"learning_rate": 0.00019009027081243733,
"loss": 2.8728,
"theoretical_loss": 3.4221736510067866,
"tokens_seen": 2058551296
},
{
"epoch": 0.62,
"learning_rate": 0.0001900802407221665,
"loss": 2.586,
"theoretical_loss": 3.422164618779688,
"tokens_seen": 2058616832
},
{
"epoch": 0.62,
"learning_rate": 0.0001900702106318957,
"loss": 2.6272,
"theoretical_loss": 3.422155586920634,
"tokens_seen": 2058682368
},
{
"epoch": 0.62,
"learning_rate": 0.00019006018054162487,
"loss": 2.7645,
"theoretical_loss": 3.4221465554295976,
"tokens_seen": 2058747904
},
{
"epoch": 0.62,
"learning_rate": 0.00019005015045135408,
"loss": 2.5346,
"theoretical_loss": 3.422137524306552,
"tokens_seen": 2058813440
},
{
"epoch": 0.62,
"learning_rate": 0.00019004012036108326,
"loss": 2.9382,
"theoretical_loss": 3.4221284935514715,
"tokens_seen": 2058878976
},
{
"epoch": 0.62,
"learning_rate": 0.00019003009027081244,
"loss": 2.4485,
"theoretical_loss": 3.4221194631643286,
"tokens_seen": 2058944512
},
{
"epoch": 0.62,
"learning_rate": 0.00019002006018054162,
"loss": 2.7878,
"theoretical_loss": 3.4221104331450967,
"tokens_seen": 2059010048
},
{
"epoch": 0.62,
"learning_rate": 0.00019001003009027083,
"loss": 2.7596,
"theoretical_loss": 3.4221014034937487,
"tokens_seen": 2059075584
},
{
"epoch": 0.62,
"learning_rate": 0.00019,
"loss": 2.8256,
"theoretical_loss": 3.4220923742102585,
"tokens_seen": 2059141120
},
{
"epoch": 0.62,
"learning_rate": 0.0001899899699097292,
"loss": 2.8228,
"theoretical_loss": 3.4220833452945993,
"tokens_seen": 2059206656
},
{
"epoch": 0.62,
"learning_rate": 0.00018997993981945837,
"loss": 2.3647,
"theoretical_loss": 3.4220743167467442,
"tokens_seen": 2059272192
},
{
"epoch": 0.62,
"learning_rate": 0.00018996990972918758,
"loss": 2.655,
"theoretical_loss": 3.422065288566667,
"tokens_seen": 2059337728
},
{
"epoch": 0.62,
"learning_rate": 0.00018995987963891676,
"loss": 2.5226,
"theoretical_loss": 3.4220562607543403,
"tokens_seen": 2059403264
},
{
"epoch": 0.62,
"objective/train/docs_used": 1165816,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9599318504333496,
"objective/train/theoretical_loss": 3.422047233309738,
"objective/train/tokens_used": 2079928800,
"theoretical_loss": 3.422047233309738,
"tokens_seen": 2059468800
},
{
"epoch": 0.62,
"learning_rate": 0.00018994984954864594,
"loss": 2.9149,
"theoretical_loss": 3.422047233309738,
"tokens_seen": 2059468800
},
{
"epoch": 0.62,
"learning_rate": 0.00018993981945837512,
"loss": 2.5724,
"theoretical_loss": 3.422038206232833,
"tokens_seen": 2059534336
},
{
"epoch": 0.62,
"learning_rate": 0.0001899297893681043,
"loss": 2.5456,
"theoretical_loss": 3.422029179523599,
"tokens_seen": 2059599872
},
{
"epoch": 0.62,
"learning_rate": 0.00018991975927783351,
"loss": 2.2937,
"theoretical_loss": 3.422020153182009,
"tokens_seen": 2059665408
},
{
"epoch": 0.62,
"learning_rate": 0.0001899097291875627,
"loss": 2.7068,
"theoretical_loss": 3.4220111272080365,
"tokens_seen": 2059730944
},
{
"epoch": 0.62,
"learning_rate": 0.00018989969909729188,
"loss": 2.5053,
"theoretical_loss": 3.4220021016016546,
"tokens_seen": 2059796480
},
{
"epoch": 0.62,
"learning_rate": 0.00018988966900702106,
"loss": 2.4197,
"theoretical_loss": 3.4219930763628374,
"tokens_seen": 2059862016
},
{
"epoch": 0.62,
"learning_rate": 0.00018987963891675027,
"loss": 2.6841,
"theoretical_loss": 3.421984051491557,
"tokens_seen": 2059927552
},
{
"epoch": 0.62,
"learning_rate": 0.00018986960882647945,
"loss": 2.489,
"theoretical_loss": 3.4219750269877878,
"tokens_seen": 2059993088
},
{
"epoch": 0.62,
"learning_rate": 0.00018985957873620863,
"loss": 2.6856,
"theoretical_loss": 3.4219660028515024,
"tokens_seen": 2060058624
},
{
"epoch": 0.62,
"learning_rate": 0.0001898495486459378,
"loss": 2.6531,
"theoretical_loss": 3.421956979082675,
"tokens_seen": 2060124160
},
{
"epoch": 0.62,
"learning_rate": 0.000189839518555667,
"loss": 2.5392,
"theoretical_loss": 3.4219479556812784,
"tokens_seen": 2060189696
},
{
"epoch": 0.62,
"learning_rate": 0.0001898294884653962,
"loss": 2.6976,
"theoretical_loss": 3.421938932647286,
"tokens_seen": 2060255232
},
{
"epoch": 0.62,
"learning_rate": 0.00018981945837512538,
"loss": 2.4544,
"theoretical_loss": 3.421929909980671,
"tokens_seen": 2060320768
},
{
"epoch": 0.62,
"learning_rate": 0.00018980942828485456,
"loss": 2.4327,
"theoretical_loss": 3.4219208876814067,
"tokens_seen": 2060386304
},
{
"epoch": 0.62,
"learning_rate": 0.00018979939819458374,
"loss": 2.4313,
"theoretical_loss": 3.4219118657494674,
"tokens_seen": 2060451840
},
{
"epoch": 0.62,
"learning_rate": 0.00018978936810431295,
"loss": 2.7074,
"theoretical_loss": 3.421902844184825,
"tokens_seen": 2060517376
},
{
"epoch": 0.62,
"learning_rate": 0.00018977933801404213,
"loss": 2.4237,
"theoretical_loss": 3.421893822987454,
"tokens_seen": 2060582912
},
{
"epoch": 0.62,
"learning_rate": 0.0001897693079237713,
"loss": 2.8009,
"theoretical_loss": 3.421884802157327,
"tokens_seen": 2060648448
},
{
"epoch": 0.62,
"learning_rate": 0.0001897592778335005,
"loss": 2.5843,
"theoretical_loss": 3.4218757816944185,
"tokens_seen": 2060713984
},
{
"epoch": 0.62,
"learning_rate": 0.00018974924774322967,
"loss": 2.617,
"theoretical_loss": 3.4218667615987,
"tokens_seen": 2060779520
},
{
"epoch": 0.62,
"learning_rate": 0.00018973921765295888,
"loss": 2.6568,
"theoretical_loss": 3.421857741870147,
"tokens_seen": 2060845056
},
{
"epoch": 0.62,
"learning_rate": 0.00018972918756268806,
"loss": 2.5479,
"theoretical_loss": 3.4218487225087317,
"tokens_seen": 2060910592
},
{
"epoch": 0.62,
"learning_rate": 0.00018971915747241724,
"loss": 2.3539,
"theoretical_loss": 3.4218397035144275,
"tokens_seen": 2060976128
},
{
"epoch": 0.62,
"learning_rate": 0.00018970912738214643,
"loss": 2.4044,
"theoretical_loss": 3.4218306848872073,
"tokens_seen": 2061041664
},
{
"epoch": 0.62,
"objective/train/docs_used": 1166525,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.595794439315796,
"objective/train/theoretical_loss": 3.421821666627046,
"objective/train/tokens_used": 2081567200,
"theoretical_loss": 3.421821666627046,
"tokens_seen": 2061107200
},
{
"epoch": 0.62,
"learning_rate": 0.00018969909729187563,
"loss": 2.5213,
"theoretical_loss": 3.421821666627046,
"tokens_seen": 2061107200
},
{
"epoch": 0.62,
"learning_rate": 0.00018968906720160481,
"loss": 2.5867,
"theoretical_loss": 3.4218126487339156,
"tokens_seen": 2061172736
},
{
"epoch": 0.62,
"learning_rate": 0.000189679037111334,
"loss": 2.5141,
"theoretical_loss": 3.4218036312077906,
"tokens_seen": 2061238272
},
{
"epoch": 0.62,
"learning_rate": 0.00018966900702106318,
"loss": 2.528,
"theoretical_loss": 3.4217946140486433,
"tokens_seen": 2061303808
},
{
"epoch": 0.62,
"learning_rate": 0.00018965897693079236,
"loss": 2.8796,
"theoretical_loss": 3.4217855972564477,
"tokens_seen": 2061369344
},
{
"epoch": 0.62,
"learning_rate": 0.00018964894684052157,
"loss": 2.8376,
"theoretical_loss": 3.4217765808311773,
"tokens_seen": 2061434880
},
{
"epoch": 0.62,
"learning_rate": 0.00018963891675025075,
"loss": 2.5917,
"theoretical_loss": 3.4217675647728054,
"tokens_seen": 2061500416
},
{
"epoch": 0.62,
"learning_rate": 0.00018962888665997993,
"loss": 2.7944,
"theoretical_loss": 3.4217585490813045,
"tokens_seen": 2061565952
},
{
"epoch": 0.62,
"learning_rate": 0.0001896188565697091,
"loss": 2.5754,
"theoretical_loss": 3.4217495337566497,
"tokens_seen": 2061631488
},
{
"epoch": 0.62,
"learning_rate": 0.00018960882647943832,
"loss": 2.5688,
"theoretical_loss": 3.4217405187988135,
"tokens_seen": 2061697024
},
{
"epoch": 0.62,
"learning_rate": 0.0001895987963891675,
"loss": 2.6523,
"theoretical_loss": 3.421731504207769,
"tokens_seen": 2061762560
},
{
"epoch": 0.62,
"learning_rate": 0.00018958876629889668,
"loss": 2.6599,
"theoretical_loss": 3.42172248998349,
"tokens_seen": 2061828096
},
{
"epoch": 0.62,
"learning_rate": 0.00018957873620862586,
"loss": 2.6335,
"theoretical_loss": 3.42171347612595,
"tokens_seen": 2061893632
},
{
"epoch": 0.62,
"learning_rate": 0.00018956870611835504,
"loss": 2.5781,
"theoretical_loss": 3.4217044626351223,
"tokens_seen": 2061959168
},
{
"epoch": 0.62,
"learning_rate": 0.00018955867602808428,
"loss": 2.8634,
"theoretical_loss": 3.42169544951098,
"tokens_seen": 2062024704
},
{
"epoch": 0.62,
"learning_rate": 0.00018954864593781346,
"loss": 2.4852,
"theoretical_loss": 3.4216864367534976,
"tokens_seen": 2062090240
},
{
"epoch": 0.62,
"learning_rate": 0.00018953861584754264,
"loss": 2.5268,
"theoretical_loss": 3.4216774243626475,
"tokens_seen": 2062155776
},
{
"epoch": 0.62,
"learning_rate": 0.00018952858575727182,
"loss": 2.8157,
"theoretical_loss": 3.4216684123384034,
"tokens_seen": 2062221312
},
{
"epoch": 0.62,
"learning_rate": 0.00018951855566700103,
"loss": 2.6418,
"theoretical_loss": 3.421659400680739,
"tokens_seen": 2062286848
},
{
"epoch": 0.62,
"learning_rate": 0.0001895085255767302,
"loss": 2.5636,
"theoretical_loss": 3.421650389389627,
"tokens_seen": 2062352384
},
{
"epoch": 0.62,
"learning_rate": 0.0001894984954864594,
"loss": 2.6441,
"theoretical_loss": 3.4216413784650417,
"tokens_seen": 2062417920
},
{
"epoch": 0.62,
"learning_rate": 0.00018948846539618857,
"loss": 2.5113,
"theoretical_loss": 3.4216323679069562,
"tokens_seen": 2062483456
},
{
"epoch": 0.63,
"learning_rate": 0.00018947843530591778,
"loss": 2.7392,
"theoretical_loss": 3.421623357715344,
"tokens_seen": 2062548992
},
{
"epoch": 0.63,
"learning_rate": 0.00018946840521564696,
"loss": 2.5736,
"theoretical_loss": 3.421614347890178,
"tokens_seen": 2062614528
},
{
"epoch": 0.63,
"learning_rate": 0.00018945837512537614,
"loss": 2.6497,
"theoretical_loss": 3.4216053384314327,
"tokens_seen": 2062680064
},
{
"epoch": 0.63,
"objective/train/docs_used": 1167825,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7563376426696777,
"objective/train/theoretical_loss": 3.4215963293390805,
"objective/train/tokens_used": 2083205600,
"theoretical_loss": 3.4215963293390805,
"tokens_seen": 2062745600
},
{
"epoch": 0.63,
"learning_rate": 0.00018944834503510532,
"loss": 2.4735,
"theoretical_loss": 3.4215963293390805,
"tokens_seen": 2062745600
},
{
"epoch": 0.63,
"learning_rate": 0.0001894383149448345,
"loss": 2.4363,
"theoretical_loss": 3.4215873206130962,
"tokens_seen": 2062811136
},
{
"epoch": 0.63,
"learning_rate": 0.00018942828485456371,
"loss": 2.6852,
"theoretical_loss": 3.4215783122534518,
"tokens_seen": 2062876672
},
{
"epoch": 0.63,
"learning_rate": 0.0001894182547642929,
"loss": 2.6533,
"theoretical_loss": 3.421569304260122,
"tokens_seen": 2062942208
},
{
"epoch": 0.63,
"learning_rate": 0.00018940822467402208,
"loss": 2.7151,
"theoretical_loss": 3.421560296633079,
"tokens_seen": 2063007744
},
{
"epoch": 0.63,
"learning_rate": 0.00018939819458375126,
"loss": 2.6456,
"theoretical_loss": 3.4215512893722972,
"tokens_seen": 2063073280
},
{
"epoch": 0.63,
"learning_rate": 0.00018938816449348047,
"loss": 2.2553,
"theoretical_loss": 3.4215422824777497,
"tokens_seen": 2063138816
},
{
"epoch": 0.63,
"learning_rate": 0.00018937813440320965,
"loss": 2.7584,
"theoretical_loss": 3.42153327594941,
"tokens_seen": 2063204352
},
{
"epoch": 0.63,
"learning_rate": 0.00018936810431293883,
"loss": 2.6193,
"theoretical_loss": 3.421524269787252,
"tokens_seen": 2063269888
},
{
"epoch": 0.63,
"learning_rate": 0.000189358074222668,
"loss": 2.7495,
"theoretical_loss": 3.4215152639912487,
"tokens_seen": 2063335424
},
{
"epoch": 0.63,
"learning_rate": 0.0001893480441323972,
"loss": 2.6031,
"theoretical_loss": 3.421506258561374,
"tokens_seen": 2063400960
},
{
"epoch": 0.63,
"learning_rate": 0.0001893380140421264,
"loss": 2.529,
"theoretical_loss": 3.421497253497601,
"tokens_seen": 2063466496
},
{
"epoch": 0.63,
"learning_rate": 0.00018932798395185558,
"loss": 2.7815,
"theoretical_loss": 3.4214882487999034,
"tokens_seen": 2063532032
},
{
"epoch": 0.63,
"learning_rate": 0.00018931795386158476,
"loss": 2.5333,
"theoretical_loss": 3.4214792444682542,
"tokens_seen": 2063597568
},
{
"epoch": 0.63,
"learning_rate": 0.00018930792377131394,
"loss": 2.4915,
"theoretical_loss": 3.421470240502628,
"tokens_seen": 2063663104
},
{
"epoch": 0.63,
"learning_rate": 0.00018929789368104315,
"loss": 2.5403,
"theoretical_loss": 3.4214612369029966,
"tokens_seen": 2063728640
},
{
"epoch": 0.63,
"learning_rate": 0.00018928786359077233,
"loss": 2.6494,
"theoretical_loss": 3.421452233669335,
"tokens_seen": 2063794176
},
{
"epoch": 0.63,
"learning_rate": 0.0001892778335005015,
"loss": 2.5573,
"theoretical_loss": 3.421443230801616,
"tokens_seen": 2063859712
},
{
"epoch": 0.63,
"learning_rate": 0.0001892678034102307,
"loss": 2.4877,
"theoretical_loss": 3.421434228299814,
"tokens_seen": 2063925248
},
{
"epoch": 0.63,
"learning_rate": 0.00018925777331995987,
"loss": 2.5916,
"theoretical_loss": 3.421425226163901,
"tokens_seen": 2063990784
},
{
"epoch": 0.63,
"learning_rate": 0.00018924774322968908,
"loss": 2.6538,
"theoretical_loss": 3.4214162243938517,
"tokens_seen": 2064056320
},
{
"epoch": 0.63,
"learning_rate": 0.00018923771313941826,
"loss": 2.4528,
"theoretical_loss": 3.4214072229896395,
"tokens_seen": 2064121856
},
{
"epoch": 0.63,
"learning_rate": 0.00018922768304914744,
"loss": 2.8476,
"theoretical_loss": 3.4213982219512373,
"tokens_seen": 2064187392
},
{
"epoch": 0.63,
"learning_rate": 0.00018921765295887663,
"loss": 2.5467,
"theoretical_loss": 3.421389221278619,
"tokens_seen": 2064252928
},
{
"epoch": 0.63,
"learning_rate": 0.00018920762286860583,
"loss": 2.433,
"theoretical_loss": 3.4213802209717583,
"tokens_seen": 2064318464
},
{
"epoch": 0.63,
"objective/train/docs_used": 1168450,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8020882606506348,
"objective/train/theoretical_loss": 3.4213712210306286,
"objective/train/tokens_used": 2084844000,
"theoretical_loss": 3.4213712210306286,
"tokens_seen": 2064384000
},
{
"epoch": 0.63,
"learning_rate": 0.00018919759277833502,
"loss": 2.7339,
"theoretical_loss": 3.4213712210306286,
"tokens_seen": 2064384000
},
{
"epoch": 0.63,
"learning_rate": 0.0001891875626880642,
"loss": 2.4962,
"theoretical_loss": 3.421362221455203,
"tokens_seen": 2064449536
},
{
"epoch": 0.63,
"learning_rate": 0.00018917753259779338,
"loss": 2.7161,
"theoretical_loss": 3.4213532222454557,
"tokens_seen": 2064515072
},
{
"epoch": 0.63,
"learning_rate": 0.00018916750250752256,
"loss": 2.725,
"theoretical_loss": 3.42134422340136,
"tokens_seen": 2064580608
},
{
"epoch": 0.63,
"learning_rate": 0.00018915747241725177,
"loss": 2.5538,
"theoretical_loss": 3.421335224922889,
"tokens_seen": 2064646144
},
{
"epoch": 0.63,
"learning_rate": 0.00018914744232698095,
"loss": 2.6663,
"theoretical_loss": 3.4213262268100166,
"tokens_seen": 2064711680
},
{
"epoch": 0.63,
"learning_rate": 0.00018913741223671013,
"loss": 2.4297,
"theoretical_loss": 3.4213172290627165,
"tokens_seen": 2064777216
},
{
"epoch": 0.63,
"learning_rate": 0.0001891273821464393,
"loss": 2.7275,
"theoretical_loss": 3.421308231680962,
"tokens_seen": 2064842752
},
{
"epoch": 0.63,
"learning_rate": 0.00018911735205616852,
"loss": 2.6007,
"theoretical_loss": 3.421299234664727,
"tokens_seen": 2064908288
},
{
"epoch": 0.63,
"learning_rate": 0.0001891073219658977,
"loss": 2.6367,
"theoretical_loss": 3.421290238013985,
"tokens_seen": 2064973824
},
{
"epoch": 0.63,
"learning_rate": 0.00018909729187562688,
"loss": 2.6335,
"theoretical_loss": 3.4212812417287086,
"tokens_seen": 2065039360
},
{
"epoch": 0.63,
"learning_rate": 0.00018908726178535606,
"loss": 2.5314,
"theoretical_loss": 3.421272245808873,
"tokens_seen": 2065104896
},
{
"epoch": 0.63,
"learning_rate": 0.00018907723169508524,
"loss": 2.5862,
"theoretical_loss": 3.42126325025445,
"tokens_seen": 2065170432
},
{
"epoch": 0.63,
"learning_rate": 0.00018906720160481445,
"loss": 2.4878,
"theoretical_loss": 3.4212542550654144,
"tokens_seen": 2065235968
},
{
"epoch": 0.63,
"learning_rate": 0.00018905717151454363,
"loss": 2.5199,
"theoretical_loss": 3.4212452602417396,
"tokens_seen": 2065301504
},
{
"epoch": 0.63,
"learning_rate": 0.0001890471414242728,
"loss": 2.5031,
"theoretical_loss": 3.421236265783399,
"tokens_seen": 2065367040
},
{
"epoch": 0.63,
"learning_rate": 0.000189037111334002,
"loss": 2.4933,
"theoretical_loss": 3.4212272716903653,
"tokens_seen": 2065432576
},
{
"epoch": 0.63,
"learning_rate": 0.0001890270812437312,
"loss": 2.8464,
"theoretical_loss": 3.421218277962614,
"tokens_seen": 2065498112
},
{
"epoch": 0.63,
"learning_rate": 0.00018901705115346038,
"loss": 2.6188,
"theoretical_loss": 3.421209284600117,
"tokens_seen": 2065563648
},
{
"epoch": 0.63,
"learning_rate": 0.00018900702106318956,
"loss": 2.4066,
"theoretical_loss": 3.4212002916028483,
"tokens_seen": 2065629184
},
{
"epoch": 0.63,
"learning_rate": 0.00018899699097291875,
"loss": 2.5997,
"theoretical_loss": 3.421191298970782,
"tokens_seen": 2065694720
},
{
"epoch": 0.63,
"learning_rate": 0.00018898696088264793,
"loss": 2.6671,
"theoretical_loss": 3.4211823067038907,
"tokens_seen": 2065760256
},
{
"epoch": 0.63,
"learning_rate": 0.00018897693079237714,
"loss": 2.444,
"theoretical_loss": 3.4211733148021493,
"tokens_seen": 2065825792
},
{
"epoch": 0.63,
"learning_rate": 0.00018896690070210632,
"loss": 2.6483,
"theoretical_loss": 3.42116432326553,
"tokens_seen": 2065891328
},
{
"epoch": 0.63,
"learning_rate": 0.0001889568706118355,
"loss": 2.572,
"theoretical_loss": 3.421155332094008,
"tokens_seen": 2065956864
},
{
"epoch": 0.63,
"objective/train/docs_used": 1168888,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.04369854927063,
"objective/train/theoretical_loss": 3.4211463412875553,
"objective/train/tokens_used": 2086482400,
"theoretical_loss": 3.4211463412875553,
"tokens_seen": 2066022400
},
{
"epoch": 0.63,
"learning_rate": 0.00018894684052156468,
"loss": 2.6654,
"theoretical_loss": 3.4211463412875553,
"tokens_seen": 2066022400
},
{
"epoch": 0.63,
"learning_rate": 0.0001889368104312939,
"loss": 2.8321,
"theoretical_loss": 3.4211373508461462,
"tokens_seen": 2066087936
},
{
"epoch": 0.63,
"learning_rate": 0.00018892678034102307,
"loss": 2.654,
"theoretical_loss": 3.4211283607697545,
"tokens_seen": 2066153472
},
{
"epoch": 0.63,
"learning_rate": 0.00018891675025075225,
"loss": 2.5639,
"theoretical_loss": 3.421119371058354,
"tokens_seen": 2066219008
},
{
"epoch": 0.63,
"learning_rate": 0.00018890672016048143,
"loss": 2.6171,
"theoretical_loss": 3.421110381711917,
"tokens_seen": 2066284544
},
{
"epoch": 0.63,
"learning_rate": 0.00018889669007021064,
"loss": 2.6398,
"theoretical_loss": 3.421101392730418,
"tokens_seen": 2066350080
},
{
"epoch": 0.63,
"learning_rate": 0.00018888665997993982,
"loss": 2.3719,
"theoretical_loss": 3.421092404113831,
"tokens_seen": 2066415616
},
{
"epoch": 0.63,
"learning_rate": 0.000188876629889669,
"loss": 2.5716,
"theoretical_loss": 3.421083415862129,
"tokens_seen": 2066481152
},
{
"epoch": 0.63,
"learning_rate": 0.00018886659979939818,
"loss": 2.3512,
"theoretical_loss": 3.4210744279752863,
"tokens_seen": 2066546688
},
{
"epoch": 0.63,
"learning_rate": 0.00018885656970912736,
"loss": 2.4636,
"theoretical_loss": 3.421065440453276,
"tokens_seen": 2066612224
},
{
"epoch": 0.63,
"learning_rate": 0.00018884653961885657,
"loss": 2.6775,
"theoretical_loss": 3.421056453296071,
"tokens_seen": 2066677760
},
{
"epoch": 0.63,
"learning_rate": 0.00018883650952858575,
"loss": 2.6788,
"theoretical_loss": 3.4210474665036466,
"tokens_seen": 2066743296
},
{
"epoch": 0.63,
"learning_rate": 0.00018882647943831493,
"loss": 2.6004,
"theoretical_loss": 3.421038480075975,
"tokens_seen": 2066808832
},
{
"epoch": 0.63,
"learning_rate": 0.00018881644934804411,
"loss": 2.7649,
"theoretical_loss": 3.421029494013031,
"tokens_seen": 2066874368
},
{
"epoch": 0.63,
"learning_rate": 0.00018880641925777335,
"loss": 2.7073,
"theoretical_loss": 3.421020508314787,
"tokens_seen": 2066939904
},
{
"epoch": 0.63,
"learning_rate": 0.00018879638916750253,
"loss": 2.3229,
"theoretical_loss": 3.421011522981217,
"tokens_seen": 2067005440
},
{
"epoch": 0.63,
"learning_rate": 0.0001887863590772317,
"loss": 2.5042,
"theoretical_loss": 3.4210025380122953,
"tokens_seen": 2067070976
},
{
"epoch": 0.63,
"learning_rate": 0.0001887763289869609,
"loss": 2.8402,
"theoretical_loss": 3.420993553407995,
"tokens_seen": 2067136512
},
{
"epoch": 0.63,
"learning_rate": 0.00018876629889669007,
"loss": 2.6585,
"theoretical_loss": 3.42098456916829,
"tokens_seen": 2067202048
},
{
"epoch": 0.63,
"learning_rate": 0.00018875626880641928,
"loss": 2.7347,
"theoretical_loss": 3.4209755852931534,
"tokens_seen": 2067267584
},
{
"epoch": 0.63,
"learning_rate": 0.00018874623871614846,
"loss": 2.7054,
"theoretical_loss": 3.4209666017825597,
"tokens_seen": 2067333120
},
{
"epoch": 0.63,
"learning_rate": 0.00018873620862587764,
"loss": 2.6621,
"theoretical_loss": 3.4209576186364816,
"tokens_seen": 2067398656
},
{
"epoch": 0.63,
"learning_rate": 0.00018872617853560683,
"loss": 2.6857,
"theoretical_loss": 3.4209486358548933,
"tokens_seen": 2067464192
},
{
"epoch": 0.63,
"learning_rate": 0.00018871614844533603,
"loss": 2.6505,
"theoretical_loss": 3.420939653437769,
"tokens_seen": 2067529728
},
{
"epoch": 0.63,
"learning_rate": 0.00018870611835506522,
"loss": 2.4386,
"theoretical_loss": 3.420930671385081,
"tokens_seen": 2067595264
},
{
"epoch": 0.63,
"objective/train/docs_used": 1170265,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.410999059677124,
"objective/train/theoretical_loss": 3.4209216896968044,
"objective/train/tokens_used": 2088120800,
"theoretical_loss": 3.4209216896968044,
"tokens_seen": 2067660800
},
{
"epoch": 0.63,
"learning_rate": 0.0001886960882647944,
"loss": 2.48,
"theoretical_loss": 3.4209216896968044,
"tokens_seen": 2067660800
},
{
"epoch": 0.63,
"learning_rate": 0.00018868605817452358,
"loss": 2.5217,
"theoretical_loss": 3.420912708372912,
"tokens_seen": 2067726336
},
{
"epoch": 0.63,
"learning_rate": 0.00018867602808425276,
"loss": 2.6745,
"theoretical_loss": 3.4209037274133776,
"tokens_seen": 2067791872
},
{
"epoch": 0.63,
"learning_rate": 0.00018866599799398197,
"loss": 2.5173,
"theoretical_loss": 3.420894746818175,
"tokens_seen": 2067857408
},
{
"epoch": 0.63,
"learning_rate": 0.00018865596790371115,
"loss": 2.7154,
"theoretical_loss": 3.4208857665872774,
"tokens_seen": 2067922944
},
{
"epoch": 0.63,
"learning_rate": 0.00018864593781344033,
"loss": 2.4668,
"theoretical_loss": 3.4208767867206595,
"tokens_seen": 2067988480
},
{
"epoch": 0.63,
"learning_rate": 0.0001886359077231695,
"loss": 2.6379,
"theoretical_loss": 3.4208678072182943,
"tokens_seen": 2068054016
},
{
"epoch": 0.63,
"learning_rate": 0.00018862587763289872,
"loss": 2.4141,
"theoretical_loss": 3.4208588280801555,
"tokens_seen": 2068119552
},
{
"epoch": 0.63,
"learning_rate": 0.0001886158475426279,
"loss": 2.668,
"theoretical_loss": 3.420849849306216,
"tokens_seen": 2068185088
},
{
"epoch": 0.63,
"learning_rate": 0.00018860581745235708,
"loss": 2.776,
"theoretical_loss": 3.4208408708964515,
"tokens_seen": 2068250624
},
{
"epoch": 0.63,
"learning_rate": 0.00018859578736208626,
"loss": 2.5637,
"theoretical_loss": 3.420831892850834,
"tokens_seen": 2068316160
},
{
"epoch": 0.63,
"learning_rate": 0.00018858575727181544,
"loss": 2.5516,
"theoretical_loss": 3.420822915169338,
"tokens_seen": 2068381696
},
{
"epoch": 0.63,
"learning_rate": 0.00018857572718154465,
"loss": 2.5221,
"theoretical_loss": 3.4208139378519364,
"tokens_seen": 2068447232
},
{
"epoch": 0.63,
"learning_rate": 0.00018856569709127383,
"loss": 2.3892,
"theoretical_loss": 3.4208049608986038,
"tokens_seen": 2068512768
},
{
"epoch": 0.63,
"learning_rate": 0.000188555667001003,
"loss": 2.6151,
"theoretical_loss": 3.420795984309313,
"tokens_seen": 2068578304
},
{
"epoch": 0.63,
"learning_rate": 0.0001885456369107322,
"loss": 2.6905,
"theoretical_loss": 3.420787008084039,
"tokens_seen": 2068643840
},
{
"epoch": 0.63,
"learning_rate": 0.0001885356068204614,
"loss": 2.6267,
"theoretical_loss": 3.4207780322227537,
"tokens_seen": 2068709376
},
{
"epoch": 0.63,
"learning_rate": 0.00018852557673019058,
"loss": 2.4724,
"theoretical_loss": 3.4207690567254327,
"tokens_seen": 2068774912
},
{
"epoch": 0.63,
"learning_rate": 0.00018851554663991976,
"loss": 2.6168,
"theoretical_loss": 3.4207600815920487,
"tokens_seen": 2068840448
},
{
"epoch": 0.63,
"learning_rate": 0.00018850551654964895,
"loss": 2.6772,
"theoretical_loss": 3.4207511068225753,
"tokens_seen": 2068905984
},
{
"epoch": 0.63,
"learning_rate": 0.00018849548645937813,
"loss": 2.551,
"theoretical_loss": 3.4207421324169864,
"tokens_seen": 2068971520
},
{
"epoch": 0.63,
"learning_rate": 0.00018848545636910734,
"loss": 2.5961,
"theoretical_loss": 3.4207331583752563,
"tokens_seen": 2069037056
},
{
"epoch": 0.63,
"learning_rate": 0.00018847542627883652,
"loss": 2.5807,
"theoretical_loss": 3.4207241846973577,
"tokens_seen": 2069102592
},
{
"epoch": 0.63,
"learning_rate": 0.0001884653961885657,
"loss": 2.6816,
"theoretical_loss": 3.420715211383265,
"tokens_seen": 2069168128
},
{
"epoch": 0.63,
"learning_rate": 0.00018845536609829488,
"loss": 2.8721,
"theoretical_loss": 3.4207062384329516,
"tokens_seen": 2069233664
},
{
"epoch": 0.63,
"objective/train/docs_used": 1170942,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4281811714172363,
"objective/train/theoretical_loss": 3.4206972658463917,
"objective/train/tokens_used": 2089759200,
"theoretical_loss": 3.4206972658463917,
"tokens_seen": 2069299200
},
{
"epoch": 0.63,
"learning_rate": 0.0001884453360080241,
"loss": 2.5292,
"theoretical_loss": 3.4206972658463917,
"tokens_seen": 2069299200
},
{
"epoch": 0.63,
"learning_rate": 0.00018843530591775327,
"loss": 2.7115,
"theoretical_loss": 3.420688293623559,
"tokens_seen": 2069364736
},
{
"epoch": 0.63,
"learning_rate": 0.00018842527582748245,
"loss": 2.8248,
"theoretical_loss": 3.4206793217644265,
"tokens_seen": 2069430272
},
{
"epoch": 0.63,
"learning_rate": 0.00018841524573721163,
"loss": 2.4566,
"theoretical_loss": 3.420670350268969,
"tokens_seen": 2069495808
},
{
"epoch": 0.63,
"learning_rate": 0.00018840521564694084,
"loss": 2.8497,
"theoretical_loss": 3.420661379137159,
"tokens_seen": 2069561344
},
{
"epoch": 0.63,
"learning_rate": 0.00018839518555667002,
"loss": 2.7813,
"theoretical_loss": 3.420652408368971,
"tokens_seen": 2069626880
},
{
"epoch": 0.63,
"learning_rate": 0.0001883851554663992,
"loss": 2.6486,
"theoretical_loss": 3.4206434379643786,
"tokens_seen": 2069692416
},
{
"epoch": 0.63,
"learning_rate": 0.00018837512537612838,
"loss": 2.7511,
"theoretical_loss": 3.420634467923356,
"tokens_seen": 2069757952
},
{
"epoch": 0.63,
"learning_rate": 0.00018836509528585756,
"loss": 2.6723,
"theoretical_loss": 3.420625498245876,
"tokens_seen": 2069823488
},
{
"epoch": 0.63,
"learning_rate": 0.00018835506519558677,
"loss": 2.6247,
"theoretical_loss": 3.4206165289319133,
"tokens_seen": 2069889024
},
{
"epoch": 0.63,
"learning_rate": 0.00018834503510531595,
"loss": 2.4753,
"theoretical_loss": 3.420607559981441,
"tokens_seen": 2069954560
},
{
"epoch": 0.63,
"learning_rate": 0.00018833500501504513,
"loss": 2.453,
"theoretical_loss": 3.420598591394433,
"tokens_seen": 2070020096
},
{
"epoch": 0.63,
"learning_rate": 0.00018832497492477431,
"loss": 2.4668,
"theoretical_loss": 3.420589623170864,
"tokens_seen": 2070085632
},
{
"epoch": 0.63,
"learning_rate": 0.00018831494483450352,
"loss": 2.4894,
"theoretical_loss": 3.4205806553107063,
"tokens_seen": 2070151168
},
{
"epoch": 0.63,
"learning_rate": 0.0001883049147442327,
"loss": 2.5664,
"theoretical_loss": 3.4205716878139345,
"tokens_seen": 2070216704
},
{
"epoch": 0.63,
"learning_rate": 0.00018829488465396189,
"loss": 2.4012,
"theoretical_loss": 3.420562720680522,
"tokens_seen": 2070282240
},
{
"epoch": 0.63,
"learning_rate": 0.00018828485456369107,
"loss": 2.5385,
"theoretical_loss": 3.420553753910443,
"tokens_seen": 2070347776
},
{
"epoch": 0.63,
"learning_rate": 0.00018827482447342025,
"loss": 2.6195,
"theoretical_loss": 3.4205447875036707,
"tokens_seen": 2070413312
},
{
"epoch": 0.63,
"learning_rate": 0.00018826479438314946,
"loss": 2.5573,
"theoretical_loss": 3.42053582146018,
"tokens_seen": 2070478848
},
{
"epoch": 0.63,
"learning_rate": 0.00018825476429287864,
"loss": 2.5369,
"theoretical_loss": 3.4205268557799435,
"tokens_seen": 2070544384
},
{
"epoch": 0.63,
"learning_rate": 0.00018824473420260782,
"loss": 2.5714,
"theoretical_loss": 3.420517890462935,
"tokens_seen": 2070609920
},
{
"epoch": 0.63,
"learning_rate": 0.000188234704112337,
"loss": 2.6145,
"theoretical_loss": 3.420508925509129,
"tokens_seen": 2070675456
},
{
"epoch": 0.63,
"learning_rate": 0.0001882246740220662,
"loss": 2.5222,
"theoretical_loss": 3.420499960918499,
"tokens_seen": 2070740992
},
{
"epoch": 0.63,
"learning_rate": 0.0001882146439317954,
"loss": 2.4642,
"theoretical_loss": 3.420490996691019,
"tokens_seen": 2070806528
},
{
"epoch": 0.63,
"learning_rate": 0.00018820461384152457,
"loss": 2.8352,
"theoretical_loss": 3.420482032826662,
"tokens_seen": 2070872064
},
{
"epoch": 0.63,
"objective/train/docs_used": 1172086,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.609699249267578,
"objective/train/theoretical_loss": 3.4204730693254035,
"objective/train/tokens_used": 2091397600,
"theoretical_loss": 3.4204730693254035,
"tokens_seen": 2070937600
},
{
"epoch": 0.63,
"learning_rate": 0.00018819458375125375,
"loss": 2.5049,
"theoretical_loss": 3.4204730693254035,
"tokens_seen": 2070937600
},
{
"epoch": 0.63,
"learning_rate": 0.00018818455366098293,
"loss": 2.5942,
"theoretical_loss": 3.4204641061872154,
"tokens_seen": 2071003136
},
{
"epoch": 0.63,
"learning_rate": 0.00018817452357071214,
"loss": 2.6538,
"theoretical_loss": 3.420455143412072,
"tokens_seen": 2071068672
},
{
"epoch": 0.63,
"learning_rate": 0.00018816449348044132,
"loss": 2.5563,
"theoretical_loss": 3.4204461809999476,
"tokens_seen": 2071134208
},
{
"epoch": 0.63,
"learning_rate": 0.0001881544633901705,
"loss": 2.5945,
"theoretical_loss": 3.4204372189508163,
"tokens_seen": 2071199744
},
{
"epoch": 0.63,
"learning_rate": 0.00018814443329989968,
"loss": 2.425,
"theoretical_loss": 3.4204282572646507,
"tokens_seen": 2071265280
},
{
"epoch": 0.63,
"learning_rate": 0.0001881344032096289,
"loss": 2.5492,
"theoretical_loss": 3.420419295941426,
"tokens_seen": 2071330816
},
{
"epoch": 0.63,
"learning_rate": 0.00018812437311935807,
"loss": 2.6566,
"theoretical_loss": 3.4204103349811152,
"tokens_seen": 2071396352
},
{
"epoch": 0.63,
"learning_rate": 0.00018811434302908725,
"loss": 2.8555,
"theoretical_loss": 3.420401374383692,
"tokens_seen": 2071461888
},
{
"epoch": 0.63,
"learning_rate": 0.00018810431293881643,
"loss": 2.6662,
"theoretical_loss": 3.4203924141491306,
"tokens_seen": 2071527424
},
{
"epoch": 0.63,
"learning_rate": 0.00018809428284854562,
"loss": 2.7245,
"theoretical_loss": 3.4203834542774043,
"tokens_seen": 2071592960
},
{
"epoch": 0.63,
"learning_rate": 0.00018808425275827482,
"loss": 2.5484,
"theoretical_loss": 3.420374494768488,
"tokens_seen": 2071658496
},
{
"epoch": 0.63,
"learning_rate": 0.000188074222668004,
"loss": 2.567,
"theoretical_loss": 3.420365535622355,
"tokens_seen": 2071724032
},
{
"epoch": 0.63,
"learning_rate": 0.00018806419257773319,
"loss": 2.5166,
"theoretical_loss": 3.4203565768389783,
"tokens_seen": 2071789568
},
{
"epoch": 0.63,
"learning_rate": 0.0001880541624874624,
"loss": 2.5189,
"theoretical_loss": 3.420347618418333,
"tokens_seen": 2071855104
},
{
"epoch": 0.63,
"learning_rate": 0.0001880441323971916,
"loss": 2.5822,
"theoretical_loss": 3.4203386603603922,
"tokens_seen": 2071920640
},
{
"epoch": 0.63,
"learning_rate": 0.00018803410230692078,
"loss": 2.6114,
"theoretical_loss": 3.4203297026651303,
"tokens_seen": 2071986176
},
{
"epoch": 0.63,
"learning_rate": 0.00018802407221664997,
"loss": 2.6808,
"theoretical_loss": 3.4203207453325204,
"tokens_seen": 2072051712
},
{
"epoch": 0.63,
"learning_rate": 0.00018801404212637915,
"loss": 2.6441,
"theoretical_loss": 3.4203117883625365,
"tokens_seen": 2072117248
},
{
"epoch": 0.63,
"learning_rate": 0.00018800401203610833,
"loss": 2.5798,
"theoretical_loss": 3.420302831755153,
"tokens_seen": 2072182784
},
{
"epoch": 0.63,
"learning_rate": 0.00018799398194583754,
"loss": 2.8367,
"theoretical_loss": 3.4202938755103434,
"tokens_seen": 2072248320
},
{
"epoch": 0.63,
"learning_rate": 0.00018798395185556672,
"loss": 2.3833,
"theoretical_loss": 3.420284919628082,
"tokens_seen": 2072313856
},
{
"epoch": 0.63,
"learning_rate": 0.0001879739217652959,
"loss": 2.5607,
"theoretical_loss": 3.4202759641083413,
"tokens_seen": 2072379392
},
{
"epoch": 0.63,
"learning_rate": 0.00018796389167502508,
"loss": 2.689,
"theoretical_loss": 3.4202670089510967,
"tokens_seen": 2072444928
},
{
"epoch": 0.63,
"learning_rate": 0.0001879538615847543,
"loss": 2.4349,
"theoretical_loss": 3.4202580541563217,
"tokens_seen": 2072510464
},
{
"epoch": 0.63,
"objective/train/docs_used": 1172787,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0199029445648193,
"objective/train/theoretical_loss": 3.4202490997239896,
"objective/train/tokens_used": 2093036000,
"theoretical_loss": 3.4202490997239896,
"tokens_seen": 2072576000
},
{
"epoch": 0.63,
"learning_rate": 0.00018794383149448347,
"loss": 2.7779,
"theoretical_loss": 3.4202490997239896,
"tokens_seen": 2072576000
},
{
"epoch": 0.63,
"learning_rate": 0.00018793380140421265,
"loss": 2.5619,
"theoretical_loss": 3.4202401456540747,
"tokens_seen": 2072641536
},
{
"epoch": 0.63,
"learning_rate": 0.00018792377131394183,
"loss": 2.6431,
"theoretical_loss": 3.420231191946551,
"tokens_seen": 2072707072
},
{
"epoch": 0.63,
"learning_rate": 0.00018791374122367104,
"loss": 2.7133,
"theoretical_loss": 3.420222238601392,
"tokens_seen": 2072772608
},
{
"epoch": 0.63,
"learning_rate": 0.00018790371113340022,
"loss": 2.6227,
"theoretical_loss": 3.420213285618572,
"tokens_seen": 2072838144
},
{
"epoch": 0.63,
"learning_rate": 0.0001878936810431294,
"loss": 2.626,
"theoretical_loss": 3.420204332998064,
"tokens_seen": 2072903680
},
{
"epoch": 0.63,
"learning_rate": 0.00018788365095285858,
"loss": 2.4832,
"theoretical_loss": 3.420195380739843,
"tokens_seen": 2072969216
},
{
"epoch": 0.63,
"learning_rate": 0.00018787362086258776,
"loss": 2.6245,
"theoretical_loss": 3.4201864288438824,
"tokens_seen": 2073034752
},
{
"epoch": 0.63,
"learning_rate": 0.00018786359077231697,
"loss": 2.563,
"theoretical_loss": 3.4201774773101556,
"tokens_seen": 2073100288
},
{
"epoch": 0.63,
"learning_rate": 0.00018785356068204615,
"loss": 2.6526,
"theoretical_loss": 3.4201685261386374,
"tokens_seen": 2073165824
},
{
"epoch": 0.63,
"learning_rate": 0.00018784353059177533,
"loss": 2.7508,
"theoretical_loss": 3.4201595753293015,
"tokens_seen": 2073231360
},
{
"epoch": 0.63,
"learning_rate": 0.00018783350050150451,
"loss": 2.7142,
"theoretical_loss": 3.420150624882121,
"tokens_seen": 2073296896
},
{
"epoch": 0.63,
"learning_rate": 0.00018782347041123372,
"loss": 2.7651,
"theoretical_loss": 3.4201416747970708,
"tokens_seen": 2073362432
},
{
"epoch": 0.63,
"learning_rate": 0.0001878134403209629,
"loss": 2.6925,
"theoretical_loss": 3.4201327250741245,
"tokens_seen": 2073427968
},
{
"epoch": 0.63,
"learning_rate": 0.00018780341023069209,
"loss": 2.5677,
"theoretical_loss": 3.4201237757132557,
"tokens_seen": 2073493504
},
{
"epoch": 0.63,
"learning_rate": 0.00018779338014042127,
"loss": 2.4889,
"theoretical_loss": 3.420114826714438,
"tokens_seen": 2073559040
},
{
"epoch": 0.63,
"learning_rate": 0.00018778335005015045,
"loss": 2.4428,
"theoretical_loss": 3.4201058780776465,
"tokens_seen": 2073624576
},
{
"epoch": 0.63,
"learning_rate": 0.00018777331995987966,
"loss": 2.5756,
"theoretical_loss": 3.4200969298028543,
"tokens_seen": 2073690112
},
{
"epoch": 0.63,
"learning_rate": 0.00018776328986960884,
"loss": 2.7139,
"theoretical_loss": 3.4200879818900347,
"tokens_seen": 2073755648
},
{
"epoch": 0.63,
"learning_rate": 0.00018775325977933802,
"loss": 2.4333,
"theoretical_loss": 3.420079034339163,
"tokens_seen": 2073821184
},
{
"epoch": 0.63,
"learning_rate": 0.0001877432296890672,
"loss": 2.5958,
"theoretical_loss": 3.4200700871502123,
"tokens_seen": 2073886720
},
{
"epoch": 0.63,
"learning_rate": 0.0001877331995987964,
"loss": 2.7431,
"theoretical_loss": 3.420061140323157,
"tokens_seen": 2073952256
},
{
"epoch": 0.63,
"learning_rate": 0.0001877231695085256,
"loss": 2.6682,
"theoretical_loss": 3.4200521938579707,
"tokens_seen": 2074017792
},
{
"epoch": 0.63,
"learning_rate": 0.00018771313941825477,
"loss": 2.8442,
"theoretical_loss": 3.420043247754627,
"tokens_seen": 2074083328
},
{
"epoch": 0.63,
"learning_rate": 0.00018770310932798395,
"loss": 2.563,
"theoretical_loss": 3.4200343020131,
"tokens_seen": 2074148864
},
{
"epoch": 0.63,
"objective/train/docs_used": 1174027,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6113691329956055,
"objective/train/theoretical_loss": 3.4200253566333645,
"objective/train/tokens_used": 2094674400,
"theoretical_loss": 3.4200253566333645,
"tokens_seen": 2074214400
},
{
"epoch": 0.63,
"learning_rate": 0.00018769307923771313,
"loss": 2.6778,
"theoretical_loss": 3.4200253566333645,
"tokens_seen": 2074214400
},
{
"epoch": 0.63,
"learning_rate": 0.00018768304914744234,
"loss": 2.504,
"theoretical_loss": 3.4200164116153937,
"tokens_seen": 2074279936
},
{
"epoch": 0.63,
"learning_rate": 0.00018767301905717152,
"loss": 2.6182,
"theoretical_loss": 3.420007466959161,
"tokens_seen": 2074345472
},
{
"epoch": 0.63,
"learning_rate": 0.0001876629889669007,
"loss": 2.7745,
"theoretical_loss": 3.419998522664641,
"tokens_seen": 2074411008
},
{
"epoch": 0.63,
"learning_rate": 0.00018765295887662988,
"loss": 2.5735,
"theoretical_loss": 3.4199895787318084,
"tokens_seen": 2074476544
},
{
"epoch": 0.63,
"learning_rate": 0.0001876429287863591,
"loss": 2.6076,
"theoretical_loss": 3.4199806351606354,
"tokens_seen": 2074542080
},
{
"epoch": 0.63,
"learning_rate": 0.00018763289869608827,
"loss": 2.8037,
"theoretical_loss": 3.4199716919510976,
"tokens_seen": 2074607616
},
{
"epoch": 0.63,
"learning_rate": 0.00018762286860581745,
"loss": 2.8939,
"theoretical_loss": 3.4199627491031674,
"tokens_seen": 2074673152
},
{
"epoch": 0.63,
"learning_rate": 0.00018761283851554663,
"loss": 2.5408,
"theoretical_loss": 3.41995380661682,
"tokens_seen": 2074738688
},
{
"epoch": 0.63,
"learning_rate": 0.00018760280842527582,
"loss": 2.6645,
"theoretical_loss": 3.419944864492029,
"tokens_seen": 2074804224
},
{
"epoch": 0.63,
"learning_rate": 0.00018759277833500502,
"loss": 2.7742,
"theoretical_loss": 3.4199359227287687,
"tokens_seen": 2074869760
},
{
"epoch": 0.63,
"learning_rate": 0.0001875827482447342,
"loss": 2.6442,
"theoretical_loss": 3.4199269813270123,
"tokens_seen": 2074935296
},
{
"epoch": 0.63,
"learning_rate": 0.0001875727181544634,
"loss": 2.6782,
"theoretical_loss": 3.419918040286734,
"tokens_seen": 2075000832
},
{
"epoch": 0.63,
"learning_rate": 0.00018756268806419257,
"loss": 2.6808,
"theoretical_loss": 3.4199090996079082,
"tokens_seen": 2075066368
},
{
"epoch": 0.63,
"learning_rate": 0.00018755265797392178,
"loss": 2.8448,
"theoretical_loss": 3.419900159290509,
"tokens_seen": 2075131904
},
{
"epoch": 0.63,
"learning_rate": 0.00018754262788365096,
"loss": 2.6093,
"theoretical_loss": 3.419891219334509,
"tokens_seen": 2075197440
},
{
"epoch": 0.63,
"learning_rate": 0.00018753259779338014,
"loss": 2.8182,
"theoretical_loss": 3.419882279739884,
"tokens_seen": 2075262976
},
{
"epoch": 0.63,
"learning_rate": 0.00018752256770310932,
"loss": 2.5472,
"theoretical_loss": 3.4198733405066064,
"tokens_seen": 2075328512
},
{
"epoch": 0.63,
"learning_rate": 0.0001875125376128385,
"loss": 2.7209,
"theoretical_loss": 3.4198644016346513,
"tokens_seen": 2075394048
},
{
"epoch": 0.63,
"learning_rate": 0.0001875025075225677,
"loss": 2.7372,
"theoretical_loss": 3.4198554631239926,
"tokens_seen": 2075459584
},
{
"epoch": 0.63,
"learning_rate": 0.0001874924774322969,
"loss": 2.9249,
"theoretical_loss": 3.4198465249746035,
"tokens_seen": 2075525120
},
{
"epoch": 0.63,
"learning_rate": 0.00018748244734202607,
"loss": 2.5123,
"theoretical_loss": 3.419837587186459,
"tokens_seen": 2075590656
},
{
"epoch": 0.63,
"learning_rate": 0.00018747241725175525,
"loss": 2.4134,
"theoretical_loss": 3.419828649759532,
"tokens_seen": 2075656192
},
{
"epoch": 0.63,
"learning_rate": 0.00018746238716148446,
"loss": 2.6955,
"theoretical_loss": 3.419819712693797,
"tokens_seen": 2075721728
},
{
"epoch": 0.63,
"learning_rate": 0.00018745235707121364,
"loss": 2.6507,
"theoretical_loss": 3.4198107759892284,
"tokens_seen": 2075787264
},
{
"epoch": 0.63,
"objective/train/docs_used": 1174812,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.684016227722168,
"objective/train/theoretical_loss": 3.4198018396457996,
"objective/train/tokens_used": 2096312800,
"theoretical_loss": 3.4198018396457996,
"tokens_seen": 2075852800
},
{
"epoch": 0.63,
"learning_rate": 0.00018744232698094282,
"loss": 2.8204,
"theoretical_loss": 3.4198018396457996,
"tokens_seen": 2075852800
},
{
"epoch": 0.63,
"learning_rate": 0.000187432296890672,
"loss": 2.499,
"theoretical_loss": 3.4197929036634855,
"tokens_seen": 2075918336
},
{
"epoch": 0.63,
"learning_rate": 0.00018742226680040118,
"loss": 2.7253,
"theoretical_loss": 3.4197839680422586,
"tokens_seen": 2075983872
},
{
"epoch": 0.63,
"learning_rate": 0.0001874122367101304,
"loss": 2.7948,
"theoretical_loss": 3.419775032782095,
"tokens_seen": 2076049408
},
{
"epoch": 0.63,
"learning_rate": 0.00018740220661985957,
"loss": 2.5243,
"theoretical_loss": 3.4197660978829667,
"tokens_seen": 2076114944
},
{
"epoch": 0.63,
"learning_rate": 0.00018739217652958876,
"loss": 2.5922,
"theoretical_loss": 3.4197571633448485,
"tokens_seen": 2076180480
},
{
"epoch": 0.63,
"learning_rate": 0.00018738214643931794,
"loss": 2.5333,
"theoretical_loss": 3.4197482291677144,
"tokens_seen": 2076246016
},
{
"epoch": 0.63,
"learning_rate": 0.00018737211634904714,
"loss": 2.5595,
"theoretical_loss": 3.4197392953515386,
"tokens_seen": 2076311552
},
{
"epoch": 0.63,
"learning_rate": 0.00018736208625877633,
"loss": 2.5933,
"theoretical_loss": 3.419730361896295,
"tokens_seen": 2076377088
},
{
"epoch": 0.63,
"learning_rate": 0.0001873520561685055,
"loss": 2.4001,
"theoretical_loss": 3.4197214288019575,
"tokens_seen": 2076442624
},
{
"epoch": 0.63,
"learning_rate": 0.0001873420260782347,
"loss": 2.5846,
"theoretical_loss": 3.4197124960685006,
"tokens_seen": 2076508160
},
{
"epoch": 0.63,
"learning_rate": 0.00018733199598796387,
"loss": 2.6549,
"theoretical_loss": 3.419703563695898,
"tokens_seen": 2076573696
},
{
"epoch": 0.63,
"learning_rate": 0.00018732196589769308,
"loss": 2.6221,
"theoretical_loss": 3.4196946316841235,
"tokens_seen": 2076639232
},
{
"epoch": 0.63,
"learning_rate": 0.00018731193580742229,
"loss": 2.7714,
"theoretical_loss": 3.419685700033151,
"tokens_seen": 2076704768
},
{
"epoch": 0.63,
"learning_rate": 0.00018730190571715147,
"loss": 2.7147,
"theoretical_loss": 3.4196767687429555,
"tokens_seen": 2076770304
},
{
"epoch": 0.63,
"learning_rate": 0.00018729187562688065,
"loss": 2.4846,
"theoretical_loss": 3.41966783781351,
"tokens_seen": 2076835840
},
{
"epoch": 0.63,
"learning_rate": 0.00018728184553660986,
"loss": 2.4919,
"theoretical_loss": 3.4196589072447896,
"tokens_seen": 2076901376
},
{
"epoch": 0.63,
"learning_rate": 0.00018727181544633904,
"loss": 2.6371,
"theoretical_loss": 3.4196499770367668,
"tokens_seen": 2076966912
},
{
"epoch": 0.63,
"learning_rate": 0.00018726178535606822,
"loss": 2.7513,
"theoretical_loss": 3.419641047189417,
"tokens_seen": 2077032448
},
{
"epoch": 0.63,
"learning_rate": 0.0001872517552657974,
"loss": 2.7396,
"theoretical_loss": 3.4196321177027142,
"tokens_seen": 2077097984
},
{
"epoch": 0.63,
"learning_rate": 0.0001872417251755266,
"loss": 2.8971,
"theoretical_loss": 3.4196231885766317,
"tokens_seen": 2077163520
},
{
"epoch": 0.63,
"learning_rate": 0.0001872316950852558,
"loss": 2.6279,
"theoretical_loss": 3.4196142598111443,
"tokens_seen": 2077229056
},
{
"epoch": 0.63,
"learning_rate": 0.00018722166499498497,
"loss": 2.594,
"theoretical_loss": 3.4196053314062254,
"tokens_seen": 2077294592
},
{
"epoch": 0.63,
"learning_rate": 0.00018721163490471415,
"loss": 2.666,
"theoretical_loss": 3.419596403361849,
"tokens_seen": 2077360128
},
{
"epoch": 0.63,
"learning_rate": 0.00018720160481444333,
"loss": 2.5784,
"theoretical_loss": 3.4195874756779903,
"tokens_seen": 2077425664
},
{
"epoch": 0.63,
"objective/train/docs_used": 1175335,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.927333116531372,
"objective/train/theoretical_loss": 3.4195785483546226,
"objective/train/tokens_used": 2097951200,
"theoretical_loss": 3.4195785483546226,
"tokens_seen": 2077491200
},
{
"epoch": 0.63,
"learning_rate": 0.00018719157472417254,
"loss": 2.6646,
"theoretical_loss": 3.4195785483546226,
"tokens_seen": 2077491200
},
{
"epoch": 0.63,
"learning_rate": 0.00018718154463390172,
"loss": 2.7905,
"theoretical_loss": 3.4195696213917195,
"tokens_seen": 2077556736
},
{
"epoch": 0.63,
"learning_rate": 0.0001871715145436309,
"loss": 2.7097,
"theoretical_loss": 3.419560694789256,
"tokens_seen": 2077622272
},
{
"epoch": 0.63,
"learning_rate": 0.00018716148445336008,
"loss": 2.5246,
"theoretical_loss": 3.419551768547205,
"tokens_seen": 2077687808
},
{
"epoch": 0.63,
"learning_rate": 0.0001871514543630893,
"loss": 2.4627,
"theoretical_loss": 3.419542842665542,
"tokens_seen": 2077753344
},
{
"epoch": 0.63,
"learning_rate": 0.00018714142427281847,
"loss": 2.8617,
"theoretical_loss": 3.41953391714424,
"tokens_seen": 2077818880
},
{
"epoch": 0.63,
"learning_rate": 0.00018713139418254765,
"loss": 2.6541,
"theoretical_loss": 3.4195249919832738,
"tokens_seen": 2077884416
},
{
"epoch": 0.63,
"learning_rate": 0.00018712136409227684,
"loss": 2.5942,
"theoretical_loss": 3.419516067182617,
"tokens_seen": 2077949952
},
{
"epoch": 0.63,
"learning_rate": 0.00018711133400200602,
"loss": 2.6319,
"theoretical_loss": 3.419507142742244,
"tokens_seen": 2078015488
},
{
"epoch": 0.63,
"learning_rate": 0.00018710130391173522,
"loss": 2.7043,
"theoretical_loss": 3.4194982186621288,
"tokens_seen": 2078081024
},
{
"epoch": 0.63,
"learning_rate": 0.0001870912738214644,
"loss": 2.4497,
"theoretical_loss": 3.4194892949422453,
"tokens_seen": 2078146560
},
{
"epoch": 0.63,
"learning_rate": 0.0001870812437311936,
"loss": 2.4304,
"theoretical_loss": 3.419480371582568,
"tokens_seen": 2078212096
},
{
"epoch": 0.63,
"learning_rate": 0.00018707121364092277,
"loss": 2.4773,
"theoretical_loss": 3.419471448583071,
"tokens_seen": 2078277632
},
{
"epoch": 0.63,
"learning_rate": 0.00018706118355065198,
"loss": 2.5506,
"theoretical_loss": 3.419462525943727,
"tokens_seen": 2078343168
},
{
"epoch": 0.63,
"learning_rate": 0.00018705115346038116,
"loss": 2.5558,
"theoretical_loss": 3.419453603664512,
"tokens_seen": 2078408704
},
{
"epoch": 0.63,
"learning_rate": 0.00018704112337011034,
"loss": 2.6408,
"theoretical_loss": 3.4194446817453996,
"tokens_seen": 2078474240
},
{
"epoch": 0.63,
"learning_rate": 0.00018703109327983952,
"loss": 2.7308,
"theoretical_loss": 3.4194357601863636,
"tokens_seen": 2078539776
},
{
"epoch": 0.63,
"learning_rate": 0.0001870210631895687,
"loss": 2.7108,
"theoretical_loss": 3.419426838987378,
"tokens_seen": 2078605312
},
{
"epoch": 0.63,
"learning_rate": 0.0001870110330992979,
"loss": 2.6785,
"theoretical_loss": 3.419417918148417,
"tokens_seen": 2078670848
},
{
"epoch": 0.63,
"learning_rate": 0.0001870010030090271,
"loss": 2.6456,
"theoretical_loss": 3.4194089976694553,
"tokens_seen": 2078736384
},
{
"epoch": 0.63,
"learning_rate": 0.00018699097291875627,
"loss": 2.7362,
"theoretical_loss": 3.419400077550466,
"tokens_seen": 2078801920
},
{
"epoch": 0.63,
"learning_rate": 0.00018698094282848545,
"loss": 2.6247,
"theoretical_loss": 3.4193911577914244,
"tokens_seen": 2078867456
},
{
"epoch": 0.63,
"learning_rate": 0.00018697091273821466,
"loss": 2.7805,
"theoretical_loss": 3.419382238392304,
"tokens_seen": 2078932992
},
{
"epoch": 0.63,
"learning_rate": 0.00018696088264794384,
"loss": 2.6895,
"theoretical_loss": 3.4193733193530784,
"tokens_seen": 2078998528
},
{
"epoch": 0.63,
"learning_rate": 0.00018695085255767302,
"loss": 2.7621,
"theoretical_loss": 3.419364400673723,
"tokens_seen": 2079064064
},
{
"epoch": 0.63,
"objective/train/docs_used": 1176572,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7253220081329346,
"objective/train/theoretical_loss": 3.4193554823542107,
"objective/train/tokens_used": 2099589600,
"theoretical_loss": 3.4193554823542107,
"tokens_seen": 2079129600
},
{
"epoch": 0.63,
"learning_rate": 0.0001869408224674022,
"loss": 2.8258,
"theoretical_loss": 3.4193554823542107,
"tokens_seen": 2079129600
},
{
"epoch": 0.63,
"learning_rate": 0.00018693079237713138,
"loss": 2.5988,
"theoretical_loss": 3.419346564394516,
"tokens_seen": 2079195136
},
{
"epoch": 0.63,
"learning_rate": 0.0001869207622868606,
"loss": 2.7477,
"theoretical_loss": 3.419337646794614,
"tokens_seen": 2079260672
},
{
"epoch": 0.63,
"learning_rate": 0.00018691073219658977,
"loss": 2.758,
"theoretical_loss": 3.4193287295544774,
"tokens_seen": 2079326208
},
{
"epoch": 0.63,
"learning_rate": 0.00018690070210631896,
"loss": 2.6472,
"theoretical_loss": 3.419319812674081,
"tokens_seen": 2079391744
},
{
"epoch": 0.63,
"learning_rate": 0.00018689067201604814,
"loss": 2.6227,
"theoretical_loss": 3.4193108961533993,
"tokens_seen": 2079457280
},
{
"epoch": 0.63,
"learning_rate": 0.00018688064192577734,
"loss": 2.9767,
"theoretical_loss": 3.4193019799924063,
"tokens_seen": 2079522816
},
{
"epoch": 0.63,
"learning_rate": 0.00018687061183550653,
"loss": 2.5315,
"theoretical_loss": 3.4192930641910753,
"tokens_seen": 2079588352
},
{
"epoch": 0.63,
"learning_rate": 0.0001868605817452357,
"loss": 2.8023,
"theoretical_loss": 3.4192841487493815,
"tokens_seen": 2079653888
},
{
"epoch": 0.63,
"learning_rate": 0.0001868505516549649,
"loss": 2.6194,
"theoretical_loss": 3.4192752336672987,
"tokens_seen": 2079719424
},
{
"epoch": 0.63,
"learning_rate": 0.00018684052156469407,
"loss": 2.7734,
"theoretical_loss": 3.4192663189448007,
"tokens_seen": 2079784960
},
{
"epoch": 0.63,
"learning_rate": 0.00018683049147442328,
"loss": 2.5221,
"theoretical_loss": 3.4192574045818622,
"tokens_seen": 2079850496
},
{
"epoch": 0.63,
"learning_rate": 0.00018682046138415246,
"loss": 2.6171,
"theoretical_loss": 3.4192484905784575,
"tokens_seen": 2079916032
},
{
"epoch": 0.63,
"learning_rate": 0.00018681043129388164,
"loss": 2.6241,
"theoretical_loss": 3.41923957693456,
"tokens_seen": 2079981568
},
{
"epoch": 0.63,
"learning_rate": 0.00018680040120361082,
"loss": 2.6631,
"theoretical_loss": 3.4192306636501444,
"tokens_seen": 2080047104
},
{
"epoch": 0.63,
"learning_rate": 0.00018679037111334003,
"loss": 2.6611,
"theoretical_loss": 3.4192217507251854,
"tokens_seen": 2080112640
},
{
"epoch": 0.63,
"learning_rate": 0.0001867803410230692,
"loss": 2.6275,
"theoretical_loss": 3.419212838159656,
"tokens_seen": 2080178176
},
{
"epoch": 0.63,
"learning_rate": 0.0001867703109327984,
"loss": 2.7389,
"theoretical_loss": 3.4192039259535307,
"tokens_seen": 2080243712
},
{
"epoch": 0.63,
"learning_rate": 0.00018676028084252757,
"loss": 2.605,
"theoretical_loss": 3.4191950141067844,
"tokens_seen": 2080309248
},
{
"epoch": 0.63,
"learning_rate": 0.00018675025075225678,
"loss": 2.6758,
"theoretical_loss": 3.419186102619391,
"tokens_seen": 2080374784
},
{
"epoch": 0.63,
"learning_rate": 0.00018674022066198596,
"loss": 2.5464,
"theoretical_loss": 3.419177191491324,
"tokens_seen": 2080440320
},
{
"epoch": 0.63,
"learning_rate": 0.00018673019057171514,
"loss": 2.5688,
"theoretical_loss": 3.4191682807225585,
"tokens_seen": 2080505856
},
{
"epoch": 0.63,
"learning_rate": 0.00018672016048144432,
"loss": 2.474,
"theoretical_loss": 3.419159370313068,
"tokens_seen": 2080571392
},
{
"epoch": 0.63,
"learning_rate": 0.0001867101303911735,
"loss": 2.7104,
"theoretical_loss": 3.419150460262827,
"tokens_seen": 2080636928
},
{
"epoch": 0.63,
"learning_rate": 0.0001867001003009027,
"loss": 2.8442,
"theoretical_loss": 3.41914155057181,
"tokens_seen": 2080702464
},
{
"epoch": 0.63,
"objective/train/docs_used": 1176843,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.042849063873291,
"objective/train/theoretical_loss": 3.419132641239991,
"objective/train/tokens_used": 2101228000,
"theoretical_loss": 3.419132641239991,
"tokens_seen": 2080768000
},
{
"epoch": 0.63,
"learning_rate": 0.0001866900702106319,
"loss": 2.658,
"theoretical_loss": 3.419132641239991,
"tokens_seen": 2080768000
},
{
"epoch": 0.63,
"learning_rate": 0.00018668004012036108,
"loss": 2.894,
"theoretical_loss": 3.419123732267344,
"tokens_seen": 2080833536
},
{
"epoch": 0.63,
"learning_rate": 0.00018667001003009026,
"loss": 2.5585,
"theoretical_loss": 3.4191148236538433,
"tokens_seen": 2080899072
},
{
"epoch": 0.63,
"learning_rate": 0.00018665997993981946,
"loss": 2.7622,
"theoretical_loss": 3.419105915399463,
"tokens_seen": 2080964608
},
{
"epoch": 0.63,
"learning_rate": 0.00018664994984954865,
"loss": 2.7443,
"theoretical_loss": 3.4190970075041776,
"tokens_seen": 2081030144
},
{
"epoch": 0.63,
"learning_rate": 0.00018663991975927783,
"loss": 2.6347,
"theoretical_loss": 3.4190880999679614,
"tokens_seen": 2081095680
},
{
"epoch": 0.63,
"learning_rate": 0.000186629889669007,
"loss": 2.8024,
"theoretical_loss": 3.4190791927907878,
"tokens_seen": 2081161216
},
{
"epoch": 0.63,
"learning_rate": 0.0001866198595787362,
"loss": 2.4775,
"theoretical_loss": 3.419070285972632,
"tokens_seen": 2081226752
},
{
"epoch": 0.63,
"learning_rate": 0.0001866098294884654,
"loss": 2.7961,
"theoretical_loss": 3.419061379513468,
"tokens_seen": 2081292288
},
{
"epoch": 0.63,
"learning_rate": 0.00018659979939819458,
"loss": 2.5222,
"theoretical_loss": 3.4190524734132692,
"tokens_seen": 2081357824
},
{
"epoch": 0.63,
"learning_rate": 0.00018658976930792376,
"loss": 2.6849,
"theoretical_loss": 3.419043567672011,
"tokens_seen": 2081423360
},
{
"epoch": 0.63,
"learning_rate": 0.00018657973921765294,
"loss": 2.3631,
"theoretical_loss": 3.4190346622896675,
"tokens_seen": 2081488896
},
{
"epoch": 0.63,
"learning_rate": 0.00018656970912738215,
"loss": 2.7032,
"theoretical_loss": 3.419025757266212,
"tokens_seen": 2081554432
},
{
"epoch": 0.63,
"learning_rate": 0.00018655967903711136,
"loss": 2.7879,
"theoretical_loss": 3.4190168526016196,
"tokens_seen": 2081619968
},
{
"epoch": 0.63,
"learning_rate": 0.00018654964894684054,
"loss": 2.8393,
"theoretical_loss": 3.4190079482958637,
"tokens_seen": 2081685504
},
{
"epoch": 0.63,
"learning_rate": 0.00018653961885656972,
"loss": 2.5647,
"theoretical_loss": 3.4189990443489195,
"tokens_seen": 2081751040
},
{
"epoch": 0.63,
"learning_rate": 0.0001865295887662989,
"loss": 2.4578,
"theoretical_loss": 3.4189901407607612,
"tokens_seen": 2081816576
},
{
"epoch": 0.63,
"learning_rate": 0.0001865195586760281,
"loss": 2.6112,
"theoretical_loss": 3.418981237531362,
"tokens_seen": 2081882112
},
{
"epoch": 0.63,
"learning_rate": 0.0001865095285857573,
"loss": 2.8683,
"theoretical_loss": 3.4189723346606975,
"tokens_seen": 2081947648
},
{
"epoch": 0.63,
"learning_rate": 0.00018649949849548647,
"loss": 2.617,
"theoretical_loss": 3.4189634321487405,
"tokens_seen": 2082013184
},
{
"epoch": 0.63,
"learning_rate": 0.00018648946840521565,
"loss": 2.6749,
"theoretical_loss": 3.4189545299954665,
"tokens_seen": 2082078720
},
{
"epoch": 0.63,
"learning_rate": 0.00018647943831494486,
"loss": 2.7244,
"theoretical_loss": 3.4189456282008495,
"tokens_seen": 2082144256
},
{
"epoch": 0.63,
"learning_rate": 0.00018646940822467404,
"loss": 2.6183,
"theoretical_loss": 3.4189367267648634,
"tokens_seen": 2082209792
},
{
"epoch": 0.63,
"learning_rate": 0.00018645937813440322,
"loss": 2.7123,
"theoretical_loss": 3.4189278256874824,
"tokens_seen": 2082275328
},
{
"epoch": 0.63,
"learning_rate": 0.0001864493480441324,
"loss": 2.8333,
"theoretical_loss": 3.4189189249686813,
"tokens_seen": 2082340864
},
{
"epoch": 0.63,
"objective/train/docs_used": 1177725,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2277188301086426,
"objective/train/theoretical_loss": 3.418910024608434,
"objective/train/tokens_used": 2102866400,
"theoretical_loss": 3.418910024608434,
"tokens_seen": 2082406400
},
{
"epoch": 0.63,
"learning_rate": 0.00018643931795386159,
"loss": 2.8371,
"theoretical_loss": 3.418910024608434,
"tokens_seen": 2082406400
},
{
"epoch": 0.63,
"learning_rate": 0.0001864292878635908,
"loss": 2.8566,
"theoretical_loss": 3.4189011246067147,
"tokens_seen": 2082471936
},
{
"epoch": 0.63,
"learning_rate": 0.00018641925777331997,
"loss": 2.546,
"theoretical_loss": 3.418892224963498,
"tokens_seen": 2082537472
},
{
"epoch": 0.63,
"learning_rate": 0.00018640922768304916,
"loss": 2.8875,
"theoretical_loss": 3.418883325678758,
"tokens_seen": 2082603008
},
{
"epoch": 0.63,
"learning_rate": 0.00018639919759277834,
"loss": 2.5737,
"theoretical_loss": 3.4188744267524687,
"tokens_seen": 2082668544
},
{
"epoch": 0.63,
"learning_rate": 0.00018638916750250754,
"loss": 2.5642,
"theoretical_loss": 3.4188655281846048,
"tokens_seen": 2082734080
},
{
"epoch": 0.63,
"learning_rate": 0.00018637913741223673,
"loss": 2.572,
"theoretical_loss": 3.4188566299751404,
"tokens_seen": 2082799616
},
{
"epoch": 0.63,
"learning_rate": 0.0001863691073219659,
"loss": 2.5783,
"theoretical_loss": 3.41884773212405,
"tokens_seen": 2082865152
},
{
"epoch": 0.63,
"learning_rate": 0.0001863590772316951,
"loss": 2.5082,
"theoretical_loss": 3.418838834631307,
"tokens_seen": 2082930688
},
{
"epoch": 0.63,
"learning_rate": 0.00018634904714142427,
"loss": 2.7645,
"theoretical_loss": 3.4188299374968874,
"tokens_seen": 2082996224
},
{
"epoch": 0.63,
"learning_rate": 0.00018633901705115348,
"loss": 2.7466,
"theoretical_loss": 3.418821040720764,
"tokens_seen": 2083061760
},
{
"epoch": 0.63,
"learning_rate": 0.00018632898696088266,
"loss": 2.6959,
"theoretical_loss": 3.418812144302912,
"tokens_seen": 2083127296
},
{
"epoch": 0.63,
"learning_rate": 0.00018631895687061184,
"loss": 2.487,
"theoretical_loss": 3.418803248243305,
"tokens_seen": 2083192832
},
{
"epoch": 0.63,
"learning_rate": 0.00018630892678034102,
"loss": 2.5576,
"theoretical_loss": 3.4187943525419175,
"tokens_seen": 2083258368
},
{
"epoch": 0.63,
"learning_rate": 0.00018629889669007023,
"loss": 2.6181,
"theoretical_loss": 3.4187854571987244,
"tokens_seen": 2083323904
},
{
"epoch": 0.63,
"learning_rate": 0.0001862888665997994,
"loss": 2.6394,
"theoretical_loss": 3.418776562213699,
"tokens_seen": 2083389440
},
{
"epoch": 0.63,
"learning_rate": 0.0001862788365095286,
"loss": 2.7187,
"theoretical_loss": 3.418767667586817,
"tokens_seen": 2083454976
},
{
"epoch": 0.63,
"learning_rate": 0.00018626880641925777,
"loss": 2.8385,
"theoretical_loss": 3.418758773318051,
"tokens_seen": 2083520512
},
{
"epoch": 0.63,
"learning_rate": 0.00018625877632898698,
"loss": 2.6557,
"theoretical_loss": 3.4187498794073763,
"tokens_seen": 2083586048
},
{
"epoch": 0.63,
"learning_rate": 0.00018624874623871616,
"loss": 2.8125,
"theoretical_loss": 3.4187409858547673,
"tokens_seen": 2083651584
},
{
"epoch": 0.63,
"learning_rate": 0.00018623871614844534,
"loss": 2.6952,
"theoretical_loss": 3.418732092660198,
"tokens_seen": 2083717120
},
{
"epoch": 0.63,
"learning_rate": 0.00018622868605817452,
"loss": 2.6008,
"theoretical_loss": 3.418723199823643,
"tokens_seen": 2083782656
},
{
"epoch": 0.63,
"learning_rate": 0.0001862186559679037,
"loss": 2.5127,
"theoretical_loss": 3.418714307345076,
"tokens_seen": 2083848192
},
{
"epoch": 0.63,
"learning_rate": 0.0001862086258776329,
"loss": 2.748,
"theoretical_loss": 3.4187054152244727,
"tokens_seen": 2083913728
},
{
"epoch": 0.63,
"learning_rate": 0.0001861985957873621,
"loss": 2.797,
"theoretical_loss": 3.418696523461806,
"tokens_seen": 2083979264
},
{
"epoch": 0.63,
"objective/train/docs_used": 1178158,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.548515796661377,
"objective/train/theoretical_loss": 3.4186876320570505,
"objective/train/tokens_used": 2104504800,
"theoretical_loss": 3.4186876320570505,
"tokens_seen": 2084044800
},
{
"epoch": 0.63,
"learning_rate": 0.00018618856569709128,
"loss": 2.6672,
"theoretical_loss": 3.4186876320570505,
"tokens_seen": 2084044800
},
{
"epoch": 0.63,
"learning_rate": 0.00018617853560682046,
"loss": 2.6587,
"theoretical_loss": 3.4186787410101815,
"tokens_seen": 2084110336
},
{
"epoch": 0.63,
"learning_rate": 0.00018616850551654966,
"loss": 2.4583,
"theoretical_loss": 3.418669850321172,
"tokens_seen": 2084175872
},
{
"epoch": 0.63,
"learning_rate": 0.00018615847542627885,
"loss": 2.8443,
"theoretical_loss": 3.4186609599899977,
"tokens_seen": 2084241408
},
{
"epoch": 0.63,
"learning_rate": 0.00018614844533600803,
"loss": 2.8049,
"theoretical_loss": 3.418652070016632,
"tokens_seen": 2084306944
},
{
"epoch": 0.63,
"learning_rate": 0.0001861384152457372,
"loss": 2.6827,
"theoretical_loss": 3.418643180401049,
"tokens_seen": 2084372480
},
{
"epoch": 0.63,
"learning_rate": 0.0001861283851554664,
"loss": 2.4613,
"theoretical_loss": 3.418634291143224,
"tokens_seen": 2084438016
},
{
"epoch": 0.63,
"learning_rate": 0.0001861183550651956,
"loss": 2.5636,
"theoretical_loss": 3.418625402243131,
"tokens_seen": 2084503552
},
{
"epoch": 0.63,
"learning_rate": 0.00018610832497492478,
"loss": 2.7524,
"theoretical_loss": 3.4186165137007443,
"tokens_seen": 2084569088
},
{
"epoch": 0.63,
"learning_rate": 0.00018609829488465396,
"loss": 2.6204,
"theoretical_loss": 3.418607625516038,
"tokens_seen": 2084634624
},
{
"epoch": 0.63,
"learning_rate": 0.00018608826479438314,
"loss": 2.5718,
"theoretical_loss": 3.4185987376889866,
"tokens_seen": 2084700160
},
{
"epoch": 0.63,
"learning_rate": 0.00018607823470411235,
"loss": 2.8425,
"theoretical_loss": 3.418589850219565,
"tokens_seen": 2084765696
},
{
"epoch": 0.63,
"learning_rate": 0.00018606820461384153,
"loss": 2.6377,
"theoretical_loss": 3.418580963107746,
"tokens_seen": 2084831232
},
{
"epoch": 0.63,
"learning_rate": 0.0001860581745235707,
"loss": 2.6382,
"theoretical_loss": 3.418572076353506,
"tokens_seen": 2084896768
},
{
"epoch": 0.63,
"learning_rate": 0.0001860481444332999,
"loss": 2.6347,
"theoretical_loss": 3.4185631899568185,
"tokens_seen": 2084962304
},
{
"epoch": 0.63,
"learning_rate": 0.00018603811434302907,
"loss": 2.7144,
"theoretical_loss": 3.4185543039176576,
"tokens_seen": 2085027840
},
{
"epoch": 0.63,
"learning_rate": 0.00018602808425275828,
"loss": 2.831,
"theoretical_loss": 3.4185454182359982,
"tokens_seen": 2085093376
},
{
"epoch": 0.63,
"learning_rate": 0.00018601805416248746,
"loss": 2.6146,
"theoretical_loss": 3.418536532911814,
"tokens_seen": 2085158912
},
{
"epoch": 0.63,
"learning_rate": 0.00018600802407221664,
"loss": 2.4189,
"theoretical_loss": 3.41852764794508,
"tokens_seen": 2085224448
},
{
"epoch": 0.63,
"learning_rate": 0.00018599799398194583,
"loss": 2.4939,
"theoretical_loss": 3.4185187633357703,
"tokens_seen": 2085289984
},
{
"epoch": 0.63,
"learning_rate": 0.00018598796389167503,
"loss": 2.5085,
"theoretical_loss": 3.4185098790838593,
"tokens_seen": 2085355520
},
{
"epoch": 0.63,
"learning_rate": 0.00018597793380140421,
"loss": 2.6685,
"theoretical_loss": 3.418500995189322,
"tokens_seen": 2085421056
},
{
"epoch": 0.63,
"learning_rate": 0.0001859679037111334,
"loss": 2.6556,
"theoretical_loss": 3.4184921116521316,
"tokens_seen": 2085486592
},
{
"epoch": 0.63,
"learning_rate": 0.00018595787362086258,
"loss": 2.8437,
"theoretical_loss": 3.4184832284722635,
"tokens_seen": 2085552128
},
{
"epoch": 0.63,
"learning_rate": 0.00018594784353059176,
"loss": 2.9332,
"theoretical_loss": 3.4184743456496918,
"tokens_seen": 2085617664
},
{
"epoch": 0.63,
"objective/train/docs_used": 1179467,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.884235143661499,
"objective/train/theoretical_loss": 3.4184654631843907,
"objective/train/tokens_used": 2106143200,
"theoretical_loss": 3.4184654631843907,
"tokens_seen": 2085683200
},
{
"epoch": 0.63,
"learning_rate": 0.00018593781344032097,
"loss": 2.7006,
"theoretical_loss": 3.4184654631843907,
"tokens_seen": 2085683200
},
{
"epoch": 0.63,
"learning_rate": 0.00018592778335005015,
"loss": 2.6801,
"theoretical_loss": 3.4184565810763345,
"tokens_seen": 2085748736
},
{
"epoch": 0.63,
"learning_rate": 0.00018591775325977933,
"loss": 2.7564,
"theoretical_loss": 3.4184476993254984,
"tokens_seen": 2085814272
},
{
"epoch": 0.63,
"learning_rate": 0.0001859077231695085,
"loss": 2.7461,
"theoretical_loss": 3.4184388179318557,
"tokens_seen": 2085879808
},
{
"epoch": 0.63,
"learning_rate": 0.00018589769307923772,
"loss": 2.5649,
"theoretical_loss": 3.4184299368953814,
"tokens_seen": 2085945344
},
{
"epoch": 0.63,
"learning_rate": 0.0001858876629889669,
"loss": 2.5254,
"theoretical_loss": 3.4184210562160504,
"tokens_seen": 2086010880
},
{
"epoch": 0.63,
"learning_rate": 0.00018587763289869608,
"loss": 2.4777,
"theoretical_loss": 3.4184121758938364,
"tokens_seen": 2086076416
},
{
"epoch": 0.63,
"learning_rate": 0.00018586760280842526,
"loss": 2.7084,
"theoretical_loss": 3.418403295928714,
"tokens_seen": 2086141952
},
{
"epoch": 0.63,
"learning_rate": 0.00018585757271815444,
"loss": 2.357,
"theoretical_loss": 3.4183944163206577,
"tokens_seen": 2086207488
},
{
"epoch": 0.63,
"learning_rate": 0.00018584754262788365,
"loss": 2.7029,
"theoretical_loss": 3.418385537069642,
"tokens_seen": 2086273024
},
{
"epoch": 0.63,
"learning_rate": 0.00018583751253761283,
"loss": 2.6751,
"theoretical_loss": 3.4183766581756405,
"tokens_seen": 2086338560
},
{
"epoch": 0.63,
"learning_rate": 0.000185827482447342,
"loss": 2.4704,
"theoretical_loss": 3.418367779638629,
"tokens_seen": 2086404096
},
{
"epoch": 0.63,
"learning_rate": 0.0001858174523570712,
"loss": 2.8805,
"theoretical_loss": 3.418358901458581,
"tokens_seen": 2086469632
},
{
"epoch": 0.63,
"learning_rate": 0.00018580742226680043,
"loss": 2.6886,
"theoretical_loss": 3.4183500236354716,
"tokens_seen": 2086535168
},
{
"epoch": 0.63,
"learning_rate": 0.0001857973921765296,
"loss": 2.7252,
"theoretical_loss": 3.4183411461692748,
"tokens_seen": 2086600704
},
{
"epoch": 0.63,
"learning_rate": 0.0001857873620862588,
"loss": 2.6741,
"theoretical_loss": 3.418332269059965,
"tokens_seen": 2086666240
},
{
"epoch": 0.63,
"learning_rate": 0.00018577733199598797,
"loss": 2.7531,
"theoretical_loss": 3.4183233923075163,
"tokens_seen": 2086731776
},
{
"epoch": 0.63,
"learning_rate": 0.00018576730190571718,
"loss": 2.8118,
"theoretical_loss": 3.418314515911904,
"tokens_seen": 2086797312
},
{
"epoch": 0.63,
"learning_rate": 0.00018575727181544636,
"loss": 2.6855,
"theoretical_loss": 3.418305639873102,
"tokens_seen": 2086862848
},
{
"epoch": 0.63,
"learning_rate": 0.00018574724172517554,
"loss": 2.7203,
"theoretical_loss": 3.4182967641910853,
"tokens_seen": 2086928384
},
{
"epoch": 0.63,
"learning_rate": 0.00018573721163490472,
"loss": 2.7866,
"theoretical_loss": 3.4182878888658275,
"tokens_seen": 2086993920
},
{
"epoch": 0.63,
"learning_rate": 0.0001857271815446339,
"loss": 2.6729,
"theoretical_loss": 3.418279013897304,
"tokens_seen": 2087059456
},
{
"epoch": 0.63,
"learning_rate": 0.00018571715145436311,
"loss": 2.568,
"theoretical_loss": 3.418270139285488,
"tokens_seen": 2087124992
},
{
"epoch": 0.63,
"learning_rate": 0.0001857071213640923,
"loss": 2.6805,
"theoretical_loss": 3.418261265030355,
"tokens_seen": 2087190528
},
{
"epoch": 0.63,
"learning_rate": 0.00018569709127382148,
"loss": 2.3915,
"theoretical_loss": 3.4182523911318796,
"tokens_seen": 2087256064
},
{
"epoch": 0.63,
"objective/train/docs_used": 1180081,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.552173376083374,
"objective/train/theoretical_loss": 3.418243517590036,
"objective/train/tokens_used": 2107781600,
"theoretical_loss": 3.418243517590036,
"tokens_seen": 2087321600
},
{
"epoch": 0.63,
"learning_rate": 0.00018568706118355066,
"loss": 2.7766,
"theoretical_loss": 3.418243517590036,
"tokens_seen": 2087321600
},
{
"epoch": 0.63,
"learning_rate": 0.00018567703109327987,
"loss": 2.7712,
"theoretical_loss": 3.418234644404798,
"tokens_seen": 2087387136
},
{
"epoch": 0.63,
"learning_rate": 0.00018566700100300905,
"loss": 2.494,
"theoretical_loss": 3.4182257715761404,
"tokens_seen": 2087452672
},
{
"epoch": 0.63,
"learning_rate": 0.00018565697091273823,
"loss": 2.5718,
"theoretical_loss": 3.4182168991040385,
"tokens_seen": 2087518208
},
{
"epoch": 0.63,
"learning_rate": 0.0001856469408224674,
"loss": 2.4742,
"theoretical_loss": 3.4182080269884656,
"tokens_seen": 2087583744
},
{
"epoch": 0.63,
"learning_rate": 0.0001856369107321966,
"loss": 2.6744,
"theoretical_loss": 3.4181991552293973,
"tokens_seen": 2087649280
},
{
"epoch": 0.63,
"learning_rate": 0.0001856268806419258,
"loss": 2.7131,
"theoretical_loss": 3.4181902838268075,
"tokens_seen": 2087714816
},
{
"epoch": 0.63,
"learning_rate": 0.00018561685055165498,
"loss": 2.5361,
"theoretical_loss": 3.4181814127806702,
"tokens_seen": 2087780352
},
{
"epoch": 0.63,
"learning_rate": 0.00018560682046138416,
"loss": 2.6566,
"theoretical_loss": 3.4181725420909608,
"tokens_seen": 2087845888
},
{
"epoch": 0.63,
"learning_rate": 0.00018559679037111334,
"loss": 2.6744,
"theoretical_loss": 3.4181636717576533,
"tokens_seen": 2087911424
},
{
"epoch": 0.63,
"learning_rate": 0.00018558676028084255,
"loss": 2.6222,
"theoretical_loss": 3.418154801780722,
"tokens_seen": 2087976960
},
{
"epoch": 0.63,
"learning_rate": 0.00018557673019057173,
"loss": 2.7899,
"theoretical_loss": 3.418145932160142,
"tokens_seen": 2088042496
},
{
"epoch": 0.63,
"learning_rate": 0.0001855667001003009,
"loss": 2.7219,
"theoretical_loss": 3.4181370628958874,
"tokens_seen": 2088108032
},
{
"epoch": 0.63,
"learning_rate": 0.0001855566700100301,
"loss": 2.4674,
"theoretical_loss": 3.4181281939879327,
"tokens_seen": 2088173568
},
{
"epoch": 0.63,
"learning_rate": 0.00018554663991975927,
"loss": 2.5409,
"theoretical_loss": 3.418119325436253,
"tokens_seen": 2088239104
},
{
"epoch": 0.63,
"learning_rate": 0.00018553660982948848,
"loss": 2.598,
"theoretical_loss": 3.4181104572408216,
"tokens_seen": 2088304640
},
{
"epoch": 0.63,
"learning_rate": 0.00018552657973921766,
"loss": 2.7653,
"theoretical_loss": 3.4181015894016142,
"tokens_seen": 2088370176
},
{
"epoch": 0.63,
"learning_rate": 0.00018551654964894684,
"loss": 2.6262,
"theoretical_loss": 3.4180927219186046,
"tokens_seen": 2088435712
},
{
"epoch": 0.63,
"learning_rate": 0.00018550651955867603,
"loss": 2.5368,
"theoretical_loss": 3.418083854791768,
"tokens_seen": 2088501248
},
{
"epoch": 0.63,
"learning_rate": 0.00018549648946840523,
"loss": 2.8405,
"theoretical_loss": 3.4180749880210777,
"tokens_seen": 2088566784
},
{
"epoch": 0.63,
"learning_rate": 0.00018548645937813441,
"loss": 2.8817,
"theoretical_loss": 3.4180661216065094,
"tokens_seen": 2088632320
},
{
"epoch": 0.63,
"learning_rate": 0.0001854764292878636,
"loss": 2.4982,
"theoretical_loss": 3.418057255548037,
"tokens_seen": 2088697856
},
{
"epoch": 0.63,
"learning_rate": 0.00018546639919759278,
"loss": 2.6798,
"theoretical_loss": 3.4180483898456355,
"tokens_seen": 2088763392
},
{
"epoch": 0.63,
"learning_rate": 0.00018545636910732196,
"loss": 2.6476,
"theoretical_loss": 3.418039524499279,
"tokens_seen": 2088828928
},
{
"epoch": 0.63,
"learning_rate": 0.00018544633901705117,
"loss": 2.5695,
"theoretical_loss": 3.418030659508942,
"tokens_seen": 2088894464
},
{
"epoch": 0.63,
"objective/train/docs_used": 1181528,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4753005504608154,
"objective/train/theoretical_loss": 3.4180217948745995,
"objective/train/tokens_used": 2109420000,
"theoretical_loss": 3.4180217948745995,
"tokens_seen": 2088960000
},
{
"epoch": 0.63,
"learning_rate": 0.00018543630892678035,
"loss": 2.6595,
"theoretical_loss": 3.4180217948745995,
"tokens_seen": 2088960000
},
{
"epoch": 0.63,
"learning_rate": 0.00018542627883650953,
"loss": 2.7449,
"theoretical_loss": 3.418012930596226,
"tokens_seen": 2089025536
},
{
"epoch": 0.63,
"learning_rate": 0.0001854162487462387,
"loss": 2.3901,
"theoretical_loss": 3.418004066673795,
"tokens_seen": 2089091072
},
{
"epoch": 0.63,
"learning_rate": 0.00018540621865596792,
"loss": 2.5624,
"theoretical_loss": 3.4179952031072824,
"tokens_seen": 2089156608
},
{
"epoch": 0.63,
"learning_rate": 0.0001853961885656971,
"loss": 2.7629,
"theoretical_loss": 3.417986339896662,
"tokens_seen": 2089222144
},
{
"epoch": 0.63,
"learning_rate": 0.00018538615847542628,
"loss": 2.6689,
"theoretical_loss": 3.4179774770419087,
"tokens_seen": 2089287680
},
{
"epoch": 0.63,
"learning_rate": 0.00018537612838515546,
"loss": 2.4688,
"theoretical_loss": 3.417968614542997,
"tokens_seen": 2089353216
},
{
"epoch": 0.63,
"learning_rate": 0.00018536609829488464,
"loss": 2.7279,
"theoretical_loss": 3.4179597523999012,
"tokens_seen": 2089418752
},
{
"epoch": 0.63,
"learning_rate": 0.00018535606820461385,
"loss": 2.477,
"theoretical_loss": 3.4179508906125955,
"tokens_seen": 2089484288
},
{
"epoch": 0.63,
"learning_rate": 0.00018534603811434303,
"loss": 2.8172,
"theoretical_loss": 3.417942029181056,
"tokens_seen": 2089549824
},
{
"epoch": 0.63,
"learning_rate": 0.0001853360080240722,
"loss": 2.749,
"theoretical_loss": 3.417933168105255,
"tokens_seen": 2089615360
},
{
"epoch": 0.63,
"learning_rate": 0.0001853259779338014,
"loss": 2.7645,
"theoretical_loss": 3.417924307385169,
"tokens_seen": 2089680896
},
{
"epoch": 0.63,
"learning_rate": 0.0001853159478435306,
"loss": 2.7168,
"theoretical_loss": 3.4179154470207718,
"tokens_seen": 2089746432
},
{
"epoch": 0.63,
"learning_rate": 0.00018530591775325978,
"loss": 2.5691,
"theoretical_loss": 3.4179065870120375,
"tokens_seen": 2089811968
},
{
"epoch": 0.63,
"learning_rate": 0.00018529588766298896,
"loss": 2.7846,
"theoretical_loss": 3.417897727358942,
"tokens_seen": 2089877504
},
{
"epoch": 0.63,
"learning_rate": 0.00018528585757271815,
"loss": 2.7072,
"theoretical_loss": 3.4178888680614583,
"tokens_seen": 2089943040
},
{
"epoch": 0.63,
"learning_rate": 0.00018527582748244733,
"loss": 2.6556,
"theoretical_loss": 3.4178800091195622,
"tokens_seen": 2090008576
},
{
"epoch": 0.63,
"learning_rate": 0.00018526579739217654,
"loss": 3.0792,
"theoretical_loss": 3.4178711505332275,
"tokens_seen": 2090074112
},
{
"epoch": 0.63,
"learning_rate": 0.00018525576730190572,
"loss": 2.9169,
"theoretical_loss": 3.417862292302429,
"tokens_seen": 2090139648
},
{
"epoch": 0.63,
"learning_rate": 0.0001852457372116349,
"loss": 2.8247,
"theoretical_loss": 3.4178534344271414,
"tokens_seen": 2090205184
},
{
"epoch": 0.63,
"learning_rate": 0.00018523570712136408,
"loss": 2.6776,
"theoretical_loss": 3.4178445769073393,
"tokens_seen": 2090270720
},
{
"epoch": 0.63,
"learning_rate": 0.0001852256770310933,
"loss": 2.7467,
"theoretical_loss": 3.417835719742998,
"tokens_seen": 2090336256
},
{
"epoch": 0.63,
"learning_rate": 0.00018521564694082247,
"loss": 2.5505,
"theoretical_loss": 3.41782686293409,
"tokens_seen": 2090401792
},
{
"epoch": 0.63,
"learning_rate": 0.00018520561685055165,
"loss": 2.5549,
"theoretical_loss": 3.417818006480592,
"tokens_seen": 2090467328
},
{
"epoch": 0.63,
"learning_rate": 0.00018519558676028083,
"loss": 2.7486,
"theoretical_loss": 3.4178091503824777,
"tokens_seen": 2090532864
},
{
"epoch": 0.63,
"objective/train/docs_used": 1182167,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4885737895965576,
"objective/train/theoretical_loss": 3.4178002946397217,
"objective/train/tokens_used": 2111058400,
"theoretical_loss": 3.4178002946397217,
"tokens_seen": 2090598400
},
{
"epoch": 0.63,
"learning_rate": 0.00018518555667001004,
"loss": 2.7703,
"theoretical_loss": 3.4178002946397217,
"tokens_seen": 2090598400
},
{
"epoch": 0.63,
"learning_rate": 0.00018517552657973922,
"loss": 2.7661,
"theoretical_loss": 3.417791439252299,
"tokens_seen": 2090663936
},
{
"epoch": 0.63,
"learning_rate": 0.0001851654964894684,
"loss": 2.8989,
"theoretical_loss": 3.4177825842201837,
"tokens_seen": 2090729472
},
{
"epoch": 0.63,
"learning_rate": 0.00018515546639919758,
"loss": 2.7103,
"theoretical_loss": 3.4177737295433506,
"tokens_seen": 2090795008
},
{
"epoch": 0.63,
"learning_rate": 0.00018514543630892676,
"loss": 2.7185,
"theoretical_loss": 3.417764875221774,
"tokens_seen": 2090860544
},
{
"epoch": 0.63,
"learning_rate": 0.00018513540621865597,
"loss": 2.6995,
"theoretical_loss": 3.4177560212554297,
"tokens_seen": 2090926080
},
{
"epoch": 0.63,
"learning_rate": 0.00018512537612838515,
"loss": 2.6336,
"theoretical_loss": 3.4177471676442908,
"tokens_seen": 2090991616
},
{
"epoch": 0.63,
"learning_rate": 0.00018511534603811433,
"loss": 2.568,
"theoretical_loss": 3.417738314388333,
"tokens_seen": 2091057152
},
{
"epoch": 0.63,
"learning_rate": 0.00018510531594784351,
"loss": 2.7119,
"theoretical_loss": 3.4177294614875304,
"tokens_seen": 2091122688
},
{
"epoch": 0.63,
"learning_rate": 0.00018509528585757272,
"loss": 2.6706,
"theoretical_loss": 3.4177206089418575,
"tokens_seen": 2091188224
},
{
"epoch": 0.63,
"learning_rate": 0.0001850852557673019,
"loss": 2.7142,
"theoretical_loss": 3.4177117567512894,
"tokens_seen": 2091253760
},
{
"epoch": 0.63,
"learning_rate": 0.00018507522567703108,
"loss": 2.4727,
"theoretical_loss": 3.4177029049158003,
"tokens_seen": 2091319296
},
{
"epoch": 0.63,
"learning_rate": 0.00018506519558676027,
"loss": 2.632,
"theoretical_loss": 3.4176940534353655,
"tokens_seen": 2091384832
},
{
"epoch": 0.63,
"learning_rate": 0.00018505516549648947,
"loss": 2.7663,
"theoretical_loss": 3.417685202309958,
"tokens_seen": 2091450368
},
{
"epoch": 0.63,
"learning_rate": 0.00018504513540621868,
"loss": 2.7089,
"theoretical_loss": 3.417676351539555,
"tokens_seen": 2091515904
},
{
"epoch": 0.63,
"learning_rate": 0.00018503510531594786,
"loss": 2.8825,
"theoretical_loss": 3.4176675011241286,
"tokens_seen": 2091581440
},
{
"epoch": 0.63,
"learning_rate": 0.00018502507522567704,
"loss": 2.7217,
"theoretical_loss": 3.417658651063655,
"tokens_seen": 2091646976
},
{
"epoch": 0.63,
"learning_rate": 0.00018501504513540623,
"loss": 2.6336,
"theoretical_loss": 3.4176498013581087,
"tokens_seen": 2091712512
},
{
"epoch": 0.63,
"learning_rate": 0.00018500501504513543,
"loss": 2.5632,
"theoretical_loss": 3.4176409520074635,
"tokens_seen": 2091778048
},
{
"epoch": 0.63,
"learning_rate": 0.00018499498495486462,
"loss": 2.4692,
"theoretical_loss": 3.417632103011695,
"tokens_seen": 2091843584
},
{
"epoch": 0.63,
"learning_rate": 0.0001849849548645938,
"loss": 2.6568,
"theoretical_loss": 3.417623254370777,
"tokens_seen": 2091909120
},
{
"epoch": 0.63,
"learning_rate": 0.00018497492477432298,
"loss": 2.6637,
"theoretical_loss": 3.4176144060846854,
"tokens_seen": 2091974656
},
{
"epoch": 0.63,
"learning_rate": 0.00018496489468405216,
"loss": 2.8401,
"theoretical_loss": 3.4176055581533933,
"tokens_seen": 2092040192
},
{
"epoch": 0.63,
"learning_rate": 0.00018495486459378137,
"loss": 2.6756,
"theoretical_loss": 3.417596710576876,
"tokens_seen": 2092105728
},
{
"epoch": 0.63,
"learning_rate": 0.00018494483450351055,
"loss": 2.597,
"theoretical_loss": 3.417587863355109,
"tokens_seen": 2092171264
},
{
"epoch": 0.63,
"objective/train/docs_used": 1183291,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.898303270339966,
"objective/train/theoretical_loss": 3.417579016488066,
"objective/train/tokens_used": 2112696800,
"theoretical_loss": 3.417579016488066,
"tokens_seen": 2092236800
},
{
"epoch": 0.63,
"learning_rate": 0.00018493480441323973,
"loss": 2.6887,
"theoretical_loss": 3.417579016488066,
"tokens_seen": 2092236800
},
{
"epoch": 0.63,
"learning_rate": 0.0001849247743229689,
"loss": 2.5607,
"theoretical_loss": 3.417570169975722,
"tokens_seen": 2092302336
},
{
"epoch": 0.63,
"learning_rate": 0.00018491474423269812,
"loss": 2.588,
"theoretical_loss": 3.417561323818051,
"tokens_seen": 2092367872
},
{
"epoch": 0.63,
"learning_rate": 0.0001849047141424273,
"loss": 2.7534,
"theoretical_loss": 3.417552478015029,
"tokens_seen": 2092433408
},
{
"epoch": 0.63,
"learning_rate": 0.00018489468405215648,
"loss": 2.6116,
"theoretical_loss": 3.4175436325666295,
"tokens_seen": 2092498944
},
{
"epoch": 0.63,
"learning_rate": 0.00018488465396188566,
"loss": 2.8568,
"theoretical_loss": 3.4175347874728277,
"tokens_seen": 2092564480
},
{
"epoch": 0.63,
"learning_rate": 0.00018487462387161484,
"loss": 2.6143,
"theoretical_loss": 3.417525942733598,
"tokens_seen": 2092630016
},
{
"epoch": 0.63,
"learning_rate": 0.00018486459378134405,
"loss": 2.4986,
"theoretical_loss": 3.417517098348916,
"tokens_seen": 2092695552
},
{
"epoch": 0.63,
"learning_rate": 0.00018485456369107323,
"loss": 2.6912,
"theoretical_loss": 3.4175082543187547,
"tokens_seen": 2092761088
},
{
"epoch": 0.63,
"learning_rate": 0.0001848445336008024,
"loss": 2.6318,
"theoretical_loss": 3.4174994106430905,
"tokens_seen": 2092826624
},
{
"epoch": 0.63,
"learning_rate": 0.0001848345035105316,
"loss": 2.6607,
"theoretical_loss": 3.417490567321897,
"tokens_seen": 2092892160
},
{
"epoch": 0.63,
"learning_rate": 0.0001848244734202608,
"loss": 2.626,
"theoretical_loss": 3.4174817243551487,
"tokens_seen": 2092957696
},
{
"epoch": 0.63,
"learning_rate": 0.00018481444332998998,
"loss": 2.7155,
"theoretical_loss": 3.4174728817428215,
"tokens_seen": 2093023232
},
{
"epoch": 0.63,
"learning_rate": 0.00018480441323971916,
"loss": 2.5284,
"theoretical_loss": 3.417464039484889,
"tokens_seen": 2093088768
},
{
"epoch": 0.63,
"learning_rate": 0.00018479438314944835,
"loss": 2.7007,
"theoretical_loss": 3.4174551975813268,
"tokens_seen": 2093154304
},
{
"epoch": 0.63,
"learning_rate": 0.00018478435305917753,
"loss": 2.522,
"theoretical_loss": 3.417446356032109,
"tokens_seen": 2093219840
},
{
"epoch": 0.63,
"learning_rate": 0.00018477432296890674,
"loss": 2.8091,
"theoretical_loss": 3.4174375148372103,
"tokens_seen": 2093285376
},
{
"epoch": 0.63,
"learning_rate": 0.00018476429287863592,
"loss": 2.6974,
"theoretical_loss": 3.4174286739966053,
"tokens_seen": 2093350912
},
{
"epoch": 0.63,
"learning_rate": 0.0001847542627883651,
"loss": 2.4737,
"theoretical_loss": 3.4174198335102695,
"tokens_seen": 2093416448
},
{
"epoch": 0.63,
"learning_rate": 0.00018474423269809428,
"loss": 2.6042,
"theoretical_loss": 3.417410993378177,
"tokens_seen": 2093481984
},
{
"epoch": 0.63,
"learning_rate": 0.0001847342026078235,
"loss": 2.7392,
"theoretical_loss": 3.4174021536003023,
"tokens_seen": 2093547520
},
{
"epoch": 0.63,
"learning_rate": 0.00018472417251755267,
"loss": 2.404,
"theoretical_loss": 3.4173933141766204,
"tokens_seen": 2093613056
},
{
"epoch": 0.63,
"learning_rate": 0.00018471414242728185,
"loss": 2.6962,
"theoretical_loss": 3.417384475107106,
"tokens_seen": 2093678592
},
{
"epoch": 0.63,
"learning_rate": 0.00018470411233701103,
"loss": 2.5133,
"theoretical_loss": 3.417375636391734,
"tokens_seen": 2093744128
},
{
"epoch": 0.63,
"learning_rate": 0.00018469408224674024,
"loss": 2.6032,
"theoretical_loss": 3.417366798030479,
"tokens_seen": 2093809664
},
{
"epoch": 0.63,
"objective/train/docs_used": 1183994,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.442464828491211,
"objective/train/theoretical_loss": 3.4173579600233155,
"objective/train/tokens_used": 2114335200,
"theoretical_loss": 3.4173579600233155,
"tokens_seen": 2093875200
},
{
"epoch": 0.63,
"learning_rate": 0.00018468405215646942,
"loss": 2.6993,
"theoretical_loss": 3.4173579600233155,
"tokens_seen": 2093875200
},
{
"epoch": 0.63,
"learning_rate": 0.0001846740220661986,
"loss": 2.7025,
"theoretical_loss": 3.4173491223702186,
"tokens_seen": 2093940736
},
{
"epoch": 0.63,
"learning_rate": 0.00018466399197592778,
"loss": 2.6029,
"theoretical_loss": 3.417340285071163,
"tokens_seen": 2094006272
},
{
"epoch": 0.63,
"learning_rate": 0.00018465396188565696,
"loss": 2.5967,
"theoretical_loss": 3.4173314481261228,
"tokens_seen": 2094071808
},
{
"epoch": 0.63,
"learning_rate": 0.00018464393179538617,
"loss": 2.7252,
"theoretical_loss": 3.4173226115350737,
"tokens_seen": 2094137344
},
{
"epoch": 0.63,
"learning_rate": 0.00018463390170511535,
"loss": 2.8011,
"theoretical_loss": 3.41731377529799,
"tokens_seen": 2094202880
},
{
"epoch": 0.63,
"learning_rate": 0.00018462387161484453,
"loss": 2.6752,
"theoretical_loss": 3.417304939414846,
"tokens_seen": 2094268416
},
{
"epoch": 0.63,
"learning_rate": 0.00018461384152457371,
"loss": 2.6122,
"theoretical_loss": 3.4172961038856173,
"tokens_seen": 2094333952
},
{
"epoch": 0.63,
"learning_rate": 0.00018460381143430292,
"loss": 2.7733,
"theoretical_loss": 3.417287268710278,
"tokens_seen": 2094399488
},
{
"epoch": 0.63,
"learning_rate": 0.0001845937813440321,
"loss": 2.5769,
"theoretical_loss": 3.4172784338888036,
"tokens_seen": 2094465024
},
{
"epoch": 0.63,
"learning_rate": 0.00018458375125376128,
"loss": 2.4374,
"theoretical_loss": 3.417269599421168,
"tokens_seen": 2094530560
},
{
"epoch": 0.63,
"learning_rate": 0.00018457372116349047,
"loss": 2.7426,
"theoretical_loss": 3.417260765307346,
"tokens_seen": 2094596096
},
{
"epoch": 0.63,
"learning_rate": 0.00018456369107321965,
"loss": 2.4614,
"theoretical_loss": 3.417251931547313,
"tokens_seen": 2094661632
},
{
"epoch": 0.63,
"learning_rate": 0.00018455366098294886,
"loss": 2.5143,
"theoretical_loss": 3.417243098141043,
"tokens_seen": 2094727168
},
{
"epoch": 0.63,
"learning_rate": 0.00018454363089267804,
"loss": 2.8556,
"theoretical_loss": 3.4172342650885117,
"tokens_seen": 2094792704
},
{
"epoch": 0.63,
"learning_rate": 0.00018453360080240722,
"loss": 2.5761,
"theoretical_loss": 3.417225432389693,
"tokens_seen": 2094858240
},
{
"epoch": 0.63,
"learning_rate": 0.0001845235707121364,
"loss": 2.549,
"theoretical_loss": 3.4172166000445623,
"tokens_seen": 2094923776
},
{
"epoch": 0.63,
"learning_rate": 0.0001845135406218656,
"loss": 2.6833,
"theoretical_loss": 3.417207768053094,
"tokens_seen": 2094989312
},
{
"epoch": 0.63,
"learning_rate": 0.0001845035105315948,
"loss": 2.6896,
"theoretical_loss": 3.4171989364152626,
"tokens_seen": 2095054848
},
{
"epoch": 0.63,
"learning_rate": 0.00018449348044132397,
"loss": 2.8827,
"theoretical_loss": 3.417190105131044,
"tokens_seen": 2095120384
},
{
"epoch": 0.63,
"learning_rate": 0.00018448345035105315,
"loss": 2.5187,
"theoretical_loss": 3.4171812742004115,
"tokens_seen": 2095185920
},
{
"epoch": 0.63,
"learning_rate": 0.00018447342026078233,
"loss": 2.7906,
"theoretical_loss": 3.417172443623341,
"tokens_seen": 2095251456
},
{
"epoch": 0.63,
"learning_rate": 0.00018446339017051154,
"loss": 2.7618,
"theoretical_loss": 3.417163613399807,
"tokens_seen": 2095316992
},
{
"epoch": 0.63,
"learning_rate": 0.00018445336008024072,
"loss": 2.6557,
"theoretical_loss": 3.417154783529784,
"tokens_seen": 2095382528
},
{
"epoch": 0.63,
"learning_rate": 0.0001844433299899699,
"loss": 2.5704,
"theoretical_loss": 3.4171459540132467,
"tokens_seen": 2095448064
},
{
"epoch": 0.63,
"objective/train/docs_used": 1184873,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.674896001815796,
"objective/train/theoretical_loss": 3.4171371248501705,
"objective/train/tokens_used": 2115973600,
"theoretical_loss": 3.4171371248501705,
"tokens_seen": 2095513600
},
{
"epoch": 0.64,
"learning_rate": 0.00018443329989969908,
"loss": 2.7423,
"theoretical_loss": 3.4171371248501705,
"tokens_seen": 2095513600
},
{
"epoch": 0.64,
"learning_rate": 0.0001844232698094283,
"loss": 2.798,
"theoretical_loss": 3.41712829604053,
"tokens_seen": 2095579136
},
{
"epoch": 0.64,
"learning_rate": 0.00018441323971915747,
"loss": 2.7631,
"theoretical_loss": 3.4171194675842997,
"tokens_seen": 2095644672
},
{
"epoch": 0.64,
"learning_rate": 0.00018440320962888665,
"loss": 2.551,
"theoretical_loss": 3.4171106394814545,
"tokens_seen": 2095710208
},
{
"epoch": 0.64,
"learning_rate": 0.00018439317953861583,
"loss": 2.8782,
"theoretical_loss": 3.41710181173197,
"tokens_seen": 2095775744
},
{
"epoch": 0.64,
"learning_rate": 0.00018438314944834502,
"loss": 2.5507,
"theoretical_loss": 3.4170929843358193,
"tokens_seen": 2095841280
},
{
"epoch": 0.64,
"learning_rate": 0.00018437311935807422,
"loss": 2.7172,
"theoretical_loss": 3.4170841572929787,
"tokens_seen": 2095906816
},
{
"epoch": 0.64,
"learning_rate": 0.0001843630892678034,
"loss": 2.49,
"theoretical_loss": 3.4170753306034225,
"tokens_seen": 2095972352
},
{
"epoch": 0.64,
"learning_rate": 0.00018435305917753259,
"loss": 2.672,
"theoretical_loss": 3.417066504267125,
"tokens_seen": 2096037888
},
{
"epoch": 0.64,
"learning_rate": 0.00018434302908726177,
"loss": 2.3685,
"theoretical_loss": 3.4170576782840625,
"tokens_seen": 2096103424
},
{
"epoch": 0.64,
"learning_rate": 0.00018433299899699098,
"loss": 2.5901,
"theoretical_loss": 3.417048852654208,
"tokens_seen": 2096168960
},
{
"epoch": 0.64,
"learning_rate": 0.00018432296890672016,
"loss": 2.6382,
"theoretical_loss": 3.4170400273775376,
"tokens_seen": 2096234496
},
{
"epoch": 0.64,
"learning_rate": 0.00018431293881644934,
"loss": 2.567,
"theoretical_loss": 3.4170312024540257,
"tokens_seen": 2096300032
},
{
"epoch": 0.64,
"learning_rate": 0.00018430290872617855,
"loss": 2.5857,
"theoretical_loss": 3.4170223778836473,
"tokens_seen": 2096365568
},
{
"epoch": 0.64,
"learning_rate": 0.00018429287863590773,
"loss": 2.55,
"theoretical_loss": 3.4170135536663766,
"tokens_seen": 2096431104
},
{
"epoch": 0.64,
"learning_rate": 0.00018428284854563694,
"loss": 2.7658,
"theoretical_loss": 3.4170047298021893,
"tokens_seen": 2096496640
},
{
"epoch": 0.64,
"learning_rate": 0.00018427281845536612,
"loss": 2.7637,
"theoretical_loss": 3.41699590629106,
"tokens_seen": 2096562176
},
{
"epoch": 0.64,
"learning_rate": 0.0001842627883650953,
"loss": 2.5125,
"theoretical_loss": 3.4169870831329625,
"tokens_seen": 2096627712
},
{
"epoch": 0.64,
"learning_rate": 0.00018425275827482448,
"loss": 2.8328,
"theoretical_loss": 3.4169782603278733,
"tokens_seen": 2096693248
},
{
"epoch": 0.64,
"learning_rate": 0.0001842427281845537,
"loss": 2.2922,
"theoretical_loss": 3.4169694378757662,
"tokens_seen": 2096758784
},
{
"epoch": 0.64,
"learning_rate": 0.00018423269809428287,
"loss": 2.5465,
"theoretical_loss": 3.4169606157766164,
"tokens_seen": 2096824320
},
{
"epoch": 0.64,
"learning_rate": 0.00018422266800401205,
"loss": 2.6386,
"theoretical_loss": 3.416951794030399,
"tokens_seen": 2096889856
},
{
"epoch": 0.64,
"learning_rate": 0.00018421263791374123,
"loss": 2.7995,
"theoretical_loss": 3.416942972637088,
"tokens_seen": 2096955392
},
{
"epoch": 0.64,
"learning_rate": 0.00018420260782347044,
"loss": 2.5853,
"theoretical_loss": 3.4169341515966583,
"tokens_seen": 2097020928
},
{
"epoch": 0.64,
"learning_rate": 0.00018419257773319962,
"loss": 2.4994,
"theoretical_loss": 3.416925330909086,
"tokens_seen": 2097086464
},
{
"epoch": 0.64,
"objective/train/docs_used": 1185597,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.072784900665283,
"objective/train/theoretical_loss": 3.416916510574345,
"objective/train/tokens_used": 2117612000,
"theoretical_loss": 3.416916510574345,
"tokens_seen": 2097152000
},
{
"epoch": 0.64,
"learning_rate": 0.0001841825476429288,
"loss": 2.4472,
"theoretical_loss": 3.416916510574345,
"tokens_seen": 2097152000
},
{
"epoch": 0.64,
"learning_rate": 0.00018417251755265798,
"loss": 2.5637,
"theoretical_loss": 3.41690769059241,
"tokens_seen": 2097217536
},
{
"epoch": 0.64,
"learning_rate": 0.00018416248746238716,
"loss": 2.6737,
"theoretical_loss": 3.416898870963257,
"tokens_seen": 2097283072
},
{
"epoch": 0.64,
"learning_rate": 0.00018415245737211637,
"loss": 2.6031,
"theoretical_loss": 3.416890051686859,
"tokens_seen": 2097348608
},
{
"epoch": 0.64,
"learning_rate": 0.00018414242728184555,
"loss": 2.6606,
"theoretical_loss": 3.4168812327631928,
"tokens_seen": 2097414144
},
{
"epoch": 0.64,
"learning_rate": 0.00018413239719157473,
"loss": 2.5972,
"theoretical_loss": 3.416872414192232,
"tokens_seen": 2097479680
},
{
"epoch": 0.64,
"learning_rate": 0.00018412236710130391,
"loss": 2.7903,
"theoretical_loss": 3.416863595973952,
"tokens_seen": 2097545216
},
{
"epoch": 0.64,
"learning_rate": 0.00018411233701103312,
"loss": 2.6449,
"theoretical_loss": 3.4168547781083274,
"tokens_seen": 2097610752
},
{
"epoch": 0.64,
"learning_rate": 0.0001841023069207623,
"loss": 2.6749,
"theoretical_loss": 3.4168459605953334,
"tokens_seen": 2097676288
},
{
"epoch": 0.64,
"learning_rate": 0.00018409227683049149,
"loss": 2.6962,
"theoretical_loss": 3.416837143434945,
"tokens_seen": 2097741824
},
{
"epoch": 0.64,
"learning_rate": 0.00018408224674022067,
"loss": 2.4381,
"theoretical_loss": 3.416828326627136,
"tokens_seen": 2097807360
},
{
"epoch": 0.64,
"learning_rate": 0.00018407221664994985,
"loss": 2.6072,
"theoretical_loss": 3.416819510171883,
"tokens_seen": 2097872896
},
{
"epoch": 0.64,
"learning_rate": 0.00018406218655967906,
"loss": 2.8063,
"theoretical_loss": 3.4168106940691594,
"tokens_seen": 2097938432
},
{
"epoch": 0.64,
"learning_rate": 0.00018405215646940824,
"loss": 2.5789,
"theoretical_loss": 3.4168018783189407,
"tokens_seen": 2098003968
},
{
"epoch": 0.64,
"learning_rate": 0.00018404212637913742,
"loss": 2.7629,
"theoretical_loss": 3.4167930629212018,
"tokens_seen": 2098069504
},
{
"epoch": 0.64,
"learning_rate": 0.0001840320962888666,
"loss": 2.6968,
"theoretical_loss": 3.416784247875918,
"tokens_seen": 2098135040
},
{
"epoch": 0.64,
"learning_rate": 0.0001840220661985958,
"loss": 2.526,
"theoretical_loss": 3.416775433183063,
"tokens_seen": 2098200576
},
{
"epoch": 0.64,
"learning_rate": 0.000184012036108325,
"loss": 2.5715,
"theoretical_loss": 3.4167666188426127,
"tokens_seen": 2098266112
},
{
"epoch": 0.64,
"learning_rate": 0.00018400200601805417,
"loss": 2.4708,
"theoretical_loss": 3.4167578048545417,
"tokens_seen": 2098331648
},
{
"epoch": 0.64,
"learning_rate": 0.00018399197592778335,
"loss": 2.6585,
"theoretical_loss": 3.4167489912188254,
"tokens_seen": 2098397184
},
{
"epoch": 0.64,
"learning_rate": 0.00018398194583751253,
"loss": 2.4972,
"theoretical_loss": 3.4167401779354383,
"tokens_seen": 2098462720
},
{
"epoch": 0.64,
"learning_rate": 0.00018397191574724174,
"loss": 2.6505,
"theoretical_loss": 3.416731365004355,
"tokens_seen": 2098528256
},
{
"epoch": 0.64,
"learning_rate": 0.00018396188565697092,
"loss": 2.7716,
"theoretical_loss": 3.4167225524255507,
"tokens_seen": 2098593792
},
{
"epoch": 0.64,
"learning_rate": 0.0001839518555667001,
"loss": 2.7242,
"theoretical_loss": 3.4167137401990004,
"tokens_seen": 2098659328
},
{
"epoch": 0.64,
"learning_rate": 0.00018394182547642928,
"loss": 2.5983,
"theoretical_loss": 3.4167049283246786,
"tokens_seen": 2098724864
},
{
"epoch": 0.64,
"objective/train/docs_used": 1186813,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2012434005737305,
"objective/train/theoretical_loss": 3.4166961168025614,
"objective/train/tokens_used": 2119250400,
"theoretical_loss": 3.4166961168025614,
"tokens_seen": 2098790400
},
{
"epoch": 0.64,
"learning_rate": 0.0001839317953861585,
"loss": 2.6735,
"theoretical_loss": 3.4166961168025614,
"tokens_seen": 2098790400
},
{
"epoch": 0.64,
"learning_rate": 0.00018392176529588767,
"loss": 2.5679,
"theoretical_loss": 3.4166873056326224,
"tokens_seen": 2098855936
},
{
"epoch": 0.64,
"learning_rate": 0.00018391173520561685,
"loss": 2.8334,
"theoretical_loss": 3.416678494814837,
"tokens_seen": 2098921472
},
{
"epoch": 0.64,
"learning_rate": 0.00018390170511534603,
"loss": 2.4391,
"theoretical_loss": 3.4166696843491806,
"tokens_seen": 2098987008
},
{
"epoch": 0.64,
"learning_rate": 0.00018389167502507522,
"loss": 2.5914,
"theoretical_loss": 3.4166608742356273,
"tokens_seen": 2099052544
},
{
"epoch": 0.64,
"learning_rate": 0.00018388164493480442,
"loss": 2.7365,
"theoretical_loss": 3.4166520644741523,
"tokens_seen": 2099118080
},
{
"epoch": 0.64,
"learning_rate": 0.0001838716148445336,
"loss": 2.7082,
"theoretical_loss": 3.4166432550647308,
"tokens_seen": 2099183616
},
{
"epoch": 0.64,
"learning_rate": 0.00018386158475426279,
"loss": 2.4556,
"theoretical_loss": 3.4166344460073375,
"tokens_seen": 2099249152
},
{
"epoch": 0.64,
"learning_rate": 0.00018385155466399197,
"loss": 2.4727,
"theoretical_loss": 3.416625637301948,
"tokens_seen": 2099314688
},
{
"epoch": 0.64,
"learning_rate": 0.00018384152457372118,
"loss": 2.5716,
"theoretical_loss": 3.416616828948536,
"tokens_seen": 2099380224
},
{
"epoch": 0.64,
"learning_rate": 0.00018383149448345036,
"loss": 2.5461,
"theoretical_loss": 3.4166080209470775,
"tokens_seen": 2099445760
},
{
"epoch": 0.64,
"learning_rate": 0.00018382146439317954,
"loss": 2.5572,
"theoretical_loss": 3.416599213297547,
"tokens_seen": 2099511296
},
{
"epoch": 0.64,
"learning_rate": 0.00018381143430290872,
"loss": 2.8406,
"theoretical_loss": 3.4165904059999193,
"tokens_seen": 2099576832
},
{
"epoch": 0.64,
"learning_rate": 0.0001838014042126379,
"loss": 2.7057,
"theoretical_loss": 3.41658159905417,
"tokens_seen": 2099642368
},
{
"epoch": 0.64,
"learning_rate": 0.0001837913741223671,
"loss": 2.6278,
"theoretical_loss": 3.416572792460274,
"tokens_seen": 2099707904
},
{
"epoch": 0.64,
"learning_rate": 0.0001837813440320963,
"loss": 2.6218,
"theoretical_loss": 3.4165639862182053,
"tokens_seen": 2099773440
},
{
"epoch": 0.64,
"learning_rate": 0.00018377131394182547,
"loss": 2.5612,
"theoretical_loss": 3.4165551803279395,
"tokens_seen": 2099838976
},
{
"epoch": 0.64,
"learning_rate": 0.00018376128385155465,
"loss": 2.5859,
"theoretical_loss": 3.416546374789452,
"tokens_seen": 2099904512
},
{
"epoch": 0.64,
"learning_rate": 0.00018375125376128386,
"loss": 2.4653,
"theoretical_loss": 3.416537569602717,
"tokens_seen": 2099970048
},
{
"epoch": 0.64,
"learning_rate": 0.00018374122367101304,
"loss": 2.3809,
"theoretical_loss": 3.41652876476771,
"tokens_seen": 2100035584
},
{
"epoch": 0.64,
"learning_rate": 0.00018373119358074222,
"loss": 2.6275,
"theoretical_loss": 3.4165199602844054,
"tokens_seen": 2100101120
},
{
"epoch": 0.64,
"learning_rate": 0.0001837211634904714,
"loss": 2.7359,
"theoretical_loss": 3.416511156152779,
"tokens_seen": 2100166656
},
{
"epoch": 0.64,
"learning_rate": 0.00018371113340020058,
"loss": 2.6137,
"theoretical_loss": 3.416502352372805,
"tokens_seen": 2100232192
},
{
"epoch": 0.64,
"learning_rate": 0.0001837011033099298,
"loss": 2.6456,
"theoretical_loss": 3.416493548944459,
"tokens_seen": 2100297728
},
{
"epoch": 0.64,
"learning_rate": 0.00018369107321965897,
"loss": 2.4286,
"theoretical_loss": 3.4164847458677157,
"tokens_seen": 2100363264
},
{
"epoch": 0.64,
"objective/train/docs_used": 1187470,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7260966300964355,
"objective/train/theoretical_loss": 3.4164759431425495,
"objective/train/tokens_used": 2120888800,
"theoretical_loss": 3.4164759431425495,
"tokens_seen": 2100428800
},
{
"epoch": 0.64,
"learning_rate": 0.00018368104312938815,
"loss": 2.722,
"theoretical_loss": 3.4164759431425495,
"tokens_seen": 2100428800
},
{
"epoch": 0.64,
"learning_rate": 0.00018367101303911734,
"loss": 2.4549,
"theoretical_loss": 3.4164671407689364,
"tokens_seen": 2100494336
},
{
"epoch": 0.64,
"learning_rate": 0.00018366098294884654,
"loss": 2.6145,
"theoretical_loss": 3.4164583387468515,
"tokens_seen": 2100559872
},
{
"epoch": 0.64,
"learning_rate": 0.00018365095285857573,
"loss": 2.7341,
"theoretical_loss": 3.4164495370762684,
"tokens_seen": 2100625408
},
{
"epoch": 0.64,
"learning_rate": 0.0001836409227683049,
"loss": 2.6869,
"theoretical_loss": 3.4164407357571633,
"tokens_seen": 2100690944
},
{
"epoch": 0.64,
"learning_rate": 0.0001836308926780341,
"loss": 2.8076,
"theoretical_loss": 3.416431934789511,
"tokens_seen": 2100756480
},
{
"epoch": 0.64,
"learning_rate": 0.00018362086258776327,
"loss": 2.4362,
"theoretical_loss": 3.416423134173286,
"tokens_seen": 2100822016
},
{
"epoch": 0.64,
"learning_rate": 0.00018361083249749248,
"loss": 2.6207,
"theoretical_loss": 3.416414333908464,
"tokens_seen": 2100887552
},
{
"epoch": 0.64,
"learning_rate": 0.00018360080240722166,
"loss": 2.6947,
"theoretical_loss": 3.4164055339950195,
"tokens_seen": 2100953088
},
{
"epoch": 0.64,
"learning_rate": 0.00018359077231695084,
"loss": 2.675,
"theoretical_loss": 3.4163967344329276,
"tokens_seen": 2101018624
},
{
"epoch": 0.64,
"learning_rate": 0.00018358074222668002,
"loss": 2.633,
"theoretical_loss": 3.416387935222164,
"tokens_seen": 2101084160
},
{
"epoch": 0.64,
"learning_rate": 0.00018357071213640923,
"loss": 2.5979,
"theoretical_loss": 3.4163791363627025,
"tokens_seen": 2101149696
},
{
"epoch": 0.64,
"learning_rate": 0.0001835606820461384,
"loss": 2.5212,
"theoretical_loss": 3.4163703378545187,
"tokens_seen": 2101215232
},
{
"epoch": 0.64,
"learning_rate": 0.00018355065195586762,
"loss": 2.7443,
"theoretical_loss": 3.4163615396975877,
"tokens_seen": 2101280768
},
{
"epoch": 0.64,
"learning_rate": 0.0001835406218655968,
"loss": 2.5325,
"theoretical_loss": 3.4163527418918846,
"tokens_seen": 2101346304
},
{
"epoch": 0.64,
"learning_rate": 0.000183530591775326,
"loss": 2.846,
"theoretical_loss": 3.4163439444373846,
"tokens_seen": 2101411840
},
{
"epoch": 0.64,
"learning_rate": 0.0001835205616850552,
"loss": 2.679,
"theoretical_loss": 3.416335147334062,
"tokens_seen": 2101477376
},
{
"epoch": 0.64,
"learning_rate": 0.00018351053159478437,
"loss": 2.716,
"theoretical_loss": 3.416326350581892,
"tokens_seen": 2101542912
},
{
"epoch": 0.64,
"learning_rate": 0.00018350050150451355,
"loss": 2.749,
"theoretical_loss": 3.416317554180851,
"tokens_seen": 2101608448
},
{
"epoch": 0.64,
"learning_rate": 0.00018349047141424273,
"loss": 2.7541,
"theoretical_loss": 3.4163087581309117,
"tokens_seen": 2101673984
},
{
"epoch": 0.64,
"learning_rate": 0.00018348044132397194,
"loss": 2.8103,
"theoretical_loss": 3.416299962432051,
"tokens_seen": 2101739520
},
{
"epoch": 0.64,
"learning_rate": 0.00018347041123370112,
"loss": 2.8595,
"theoretical_loss": 3.4162911670842435,
"tokens_seen": 2101805056
},
{
"epoch": 0.64,
"learning_rate": 0.0001834603811434303,
"loss": 2.6761,
"theoretical_loss": 3.4162823720874633,
"tokens_seen": 2101870592
},
{
"epoch": 0.64,
"learning_rate": 0.00018345035105315948,
"loss": 2.5531,
"theoretical_loss": 3.416273577441687,
"tokens_seen": 2101936128
},
{
"epoch": 0.64,
"learning_rate": 0.0001834403209628887,
"loss": 2.8527,
"theoretical_loss": 3.416264783146888,
"tokens_seen": 2102001664
},
{
"epoch": 0.64,
"objective/train/docs_used": 1188148,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5995919704437256,
"objective/train/theoretical_loss": 3.4162559892030426,
"objective/train/tokens_used": 2122527200,
"theoretical_loss": 3.4162559892030426,
"tokens_seen": 2102067200
},
{
"epoch": 0.64,
"learning_rate": 0.00018343029087261787,
"loss": 2.5518,
"theoretical_loss": 3.4162559892030426,
"tokens_seen": 2102067200
},
{
"epoch": 0.64,
"learning_rate": 0.00018342026078234705,
"loss": 2.5674,
"theoretical_loss": 3.4162471956101257,
"tokens_seen": 2102132736
},
{
"epoch": 0.64,
"learning_rate": 0.00018341023069207623,
"loss": 2.3065,
"theoretical_loss": 3.416238402368112,
"tokens_seen": 2102198272
},
{
"epoch": 0.64,
"learning_rate": 0.00018340020060180542,
"loss": 2.5609,
"theoretical_loss": 3.4162296094769764,
"tokens_seen": 2102263808
},
{
"epoch": 0.64,
"learning_rate": 0.00018339017051153462,
"loss": 2.5144,
"theoretical_loss": 3.4162208169366943,
"tokens_seen": 2102329344
},
{
"epoch": 0.64,
"learning_rate": 0.0001833801404212638,
"loss": 2.6898,
"theoretical_loss": 3.416212024747241,
"tokens_seen": 2102394880
},
{
"epoch": 0.64,
"learning_rate": 0.000183370110330993,
"loss": 2.7572,
"theoretical_loss": 3.416203232908591,
"tokens_seen": 2102460416
},
{
"epoch": 0.64,
"learning_rate": 0.00018336008024072217,
"loss": 2.6736,
"theoretical_loss": 3.4161944414207195,
"tokens_seen": 2102525952
},
{
"epoch": 0.64,
"learning_rate": 0.00018335005015045138,
"loss": 2.8285,
"theoretical_loss": 3.4161856502836017,
"tokens_seen": 2102591488
},
{
"epoch": 0.64,
"learning_rate": 0.00018334002006018056,
"loss": 2.5447,
"theoretical_loss": 3.416176859497213,
"tokens_seen": 2102657024
},
{
"epoch": 0.64,
"learning_rate": 0.00018332998996990974,
"loss": 2.6286,
"theoretical_loss": 3.416168069061528,
"tokens_seen": 2102722560
},
{
"epoch": 0.64,
"learning_rate": 0.00018331995987963892,
"loss": 2.7834,
"theoretical_loss": 3.416159278976522,
"tokens_seen": 2102788096
},
{
"epoch": 0.64,
"learning_rate": 0.0001833099297893681,
"loss": 2.5596,
"theoretical_loss": 3.4161504892421695,
"tokens_seen": 2102853632
},
{
"epoch": 0.64,
"learning_rate": 0.0001832998996990973,
"loss": 2.7557,
"theoretical_loss": 3.4161416998584464,
"tokens_seen": 2102919168
},
{
"epoch": 0.64,
"learning_rate": 0.0001832898696088265,
"loss": 2.6955,
"theoretical_loss": 3.4161329108253278,
"tokens_seen": 2102984704
},
{
"epoch": 0.64,
"learning_rate": 0.00018327983951855567,
"loss": 2.7716,
"theoretical_loss": 3.4161241221427883,
"tokens_seen": 2103050240
},
{
"epoch": 0.64,
"learning_rate": 0.00018326980942828485,
"loss": 2.72,
"theoretical_loss": 3.416115333810803,
"tokens_seen": 2103115776
},
{
"epoch": 0.64,
"learning_rate": 0.00018325977933801406,
"loss": 2.5348,
"theoretical_loss": 3.416106545829347,
"tokens_seen": 2103181312
},
{
"epoch": 0.64,
"learning_rate": 0.00018324974924774324,
"loss": 2.8141,
"theoretical_loss": 3.4160977581983962,
"tokens_seen": 2103246848
},
{
"epoch": 0.64,
"learning_rate": 0.00018323971915747242,
"loss": 2.6187,
"theoretical_loss": 3.4160889709179245,
"tokens_seen": 2103312384
},
{
"epoch": 0.64,
"learning_rate": 0.0001832296890672016,
"loss": 2.7276,
"theoretical_loss": 3.4160801839879076,
"tokens_seen": 2103377920
},
{
"epoch": 0.64,
"learning_rate": 0.00018321965897693078,
"loss": 2.4063,
"theoretical_loss": 3.4160713974083206,
"tokens_seen": 2103443456
},
{
"epoch": 0.64,
"learning_rate": 0.00018320962888666,
"loss": 2.6632,
"theoretical_loss": 3.4160626111791386,
"tokens_seen": 2103508992
},
{
"epoch": 0.64,
"learning_rate": 0.00018319959879638917,
"loss": 2.8343,
"theoretical_loss": 3.4160538253003367,
"tokens_seen": 2103574528
},
{
"epoch": 0.64,
"learning_rate": 0.00018318956870611836,
"loss": 2.6471,
"theoretical_loss": 3.41604503977189,
"tokens_seen": 2103640064
},
{
"epoch": 0.64,
"objective/train/docs_used": 1189669,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9627673625946045,
"objective/train/theoretical_loss": 3.4160362545937732,
"objective/train/tokens_used": 2124165600,
"theoretical_loss": 3.4160362545937732,
"tokens_seen": 2103705600
},
{
"epoch": 0.64,
"learning_rate": 0.00018317953861584754,
"loss": 2.8063,
"theoretical_loss": 3.4160362545937732,
"tokens_seen": 2103705600
},
{
"epoch": 0.64,
"learning_rate": 0.00018316950852557674,
"loss": 2.6787,
"theoretical_loss": 3.4160274697659623,
"tokens_seen": 2103771136
},
{
"epoch": 0.64,
"learning_rate": 0.00018315947843530593,
"loss": 2.8301,
"theoretical_loss": 3.4160186852884316,
"tokens_seen": 2103836672
},
{
"epoch": 0.64,
"learning_rate": 0.0001831494483450351,
"loss": 2.7467,
"theoretical_loss": 3.416009901161157,
"tokens_seen": 2103902208
},
{
"epoch": 0.64,
"learning_rate": 0.0001831394182547643,
"loss": 2.9456,
"theoretical_loss": 3.4160011173841127,
"tokens_seen": 2103967744
},
{
"epoch": 0.64,
"learning_rate": 0.00018312938816449347,
"loss": 2.5074,
"theoretical_loss": 3.4159923339572744,
"tokens_seen": 2104033280
},
{
"epoch": 0.64,
"learning_rate": 0.00018311935807422268,
"loss": 2.7733,
"theoretical_loss": 3.4159835508806173,
"tokens_seen": 2104098816
},
{
"epoch": 0.64,
"learning_rate": 0.00018310932798395186,
"loss": 2.8233,
"theoretical_loss": 3.4159747681541166,
"tokens_seen": 2104164352
},
{
"epoch": 0.64,
"learning_rate": 0.00018309929789368104,
"loss": 2.6057,
"theoretical_loss": 3.4159659857777465,
"tokens_seen": 2104229888
},
{
"epoch": 0.64,
"learning_rate": 0.00018308926780341022,
"loss": 2.7356,
"theoretical_loss": 3.415957203751484,
"tokens_seen": 2104295424
},
{
"epoch": 0.64,
"learning_rate": 0.00018307923771313943,
"loss": 2.6276,
"theoretical_loss": 3.415948422075302,
"tokens_seen": 2104360960
},
{
"epoch": 0.64,
"learning_rate": 0.0001830692076228686,
"loss": 2.6637,
"theoretical_loss": 3.415939640749177,
"tokens_seen": 2104426496
},
{
"epoch": 0.64,
"learning_rate": 0.0001830591775325978,
"loss": 2.6416,
"theoretical_loss": 3.415930859773084,
"tokens_seen": 2104492032
},
{
"epoch": 0.64,
"learning_rate": 0.00018304914744232697,
"loss": 2.7155,
"theoretical_loss": 3.415922079146998,
"tokens_seen": 2104557568
},
{
"epoch": 0.64,
"learning_rate": 0.00018303911735205618,
"loss": 2.6334,
"theoretical_loss": 3.415913298870894,
"tokens_seen": 2104623104
},
{
"epoch": 0.64,
"learning_rate": 0.00018302908726178536,
"loss": 2.7081,
"theoretical_loss": 3.4159045189447474,
"tokens_seen": 2104688640
},
{
"epoch": 0.64,
"learning_rate": 0.00018301905717151454,
"loss": 2.7649,
"theoretical_loss": 3.4158957393685334,
"tokens_seen": 2104754176
},
{
"epoch": 0.64,
"learning_rate": 0.00018300902708124372,
"loss": 2.4384,
"theoretical_loss": 3.415886960142227,
"tokens_seen": 2104819712
},
{
"epoch": 0.64,
"learning_rate": 0.0001829989969909729,
"loss": 2.8887,
"theoretical_loss": 3.4158781812658034,
"tokens_seen": 2104885248
},
{
"epoch": 0.64,
"learning_rate": 0.0001829889669007021,
"loss": 2.3953,
"theoretical_loss": 3.415869402739238,
"tokens_seen": 2104950784
},
{
"epoch": 0.64,
"learning_rate": 0.0001829789368104313,
"loss": 2.6655,
"theoretical_loss": 3.415860624562505,
"tokens_seen": 2105016320
},
{
"epoch": 0.64,
"learning_rate": 0.00018296890672016048,
"loss": 2.4885,
"theoretical_loss": 3.415851846735581,
"tokens_seen": 2105081856
},
{
"epoch": 0.64,
"learning_rate": 0.00018295887662988966,
"loss": 2.7518,
"theoretical_loss": 3.4158430692584405,
"tokens_seen": 2105147392
},
{
"epoch": 0.64,
"learning_rate": 0.00018294884653961886,
"loss": 2.6468,
"theoretical_loss": 3.4158342921310583,
"tokens_seen": 2105212928
},
{
"epoch": 0.64,
"learning_rate": 0.00018293881644934805,
"loss": 2.8344,
"theoretical_loss": 3.4158255153534096,
"tokens_seen": 2105278464
},
{
"epoch": 0.64,
"objective/train/docs_used": 1190193,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.0119941234588623,
"objective/train/theoretical_loss": 3.4158167389254706,
"objective/train/tokens_used": 2125804000,
"theoretical_loss": 3.4158167389254706,
"tokens_seen": 2105344000
},
{
"epoch": 0.64,
"learning_rate": 0.00018292878635907723,
"loss": 2.5623,
"theoretical_loss": 3.4158167389254706,
"tokens_seen": 2105344000
},
{
"epoch": 0.64,
"learning_rate": 0.0001829187562688064,
"loss": 2.6892,
"theoretical_loss": 3.4158079628472153,
"tokens_seen": 2105409536
},
{
"epoch": 0.64,
"learning_rate": 0.0001829087261785356,
"loss": 2.6442,
"theoretical_loss": 3.415799187118619,
"tokens_seen": 2105475072
},
{
"epoch": 0.64,
"learning_rate": 0.0001828986960882648,
"loss": 2.5817,
"theoretical_loss": 3.415790411739658,
"tokens_seen": 2105540608
},
{
"epoch": 0.64,
"learning_rate": 0.00018288866599799398,
"loss": 2.6958,
"theoretical_loss": 3.4157816367103067,
"tokens_seen": 2105606144
},
{
"epoch": 0.64,
"learning_rate": 0.00018287863590772316,
"loss": 2.4971,
"theoretical_loss": 3.41577286203054,
"tokens_seen": 2105671680
},
{
"epoch": 0.64,
"learning_rate": 0.00018286860581745234,
"loss": 2.7573,
"theoretical_loss": 3.4157640877003335,
"tokens_seen": 2105737216
},
{
"epoch": 0.64,
"learning_rate": 0.00018285857572718155,
"loss": 2.5269,
"theoretical_loss": 3.415755313719662,
"tokens_seen": 2105802752
},
{
"epoch": 0.64,
"learning_rate": 0.00018284854563691073,
"loss": 2.7193,
"theoretical_loss": 3.4157465400885014,
"tokens_seen": 2105868288
},
{
"epoch": 0.64,
"learning_rate": 0.0001828385155466399,
"loss": 2.4987,
"theoretical_loss": 3.415737766806826,
"tokens_seen": 2105933824
},
{
"epoch": 0.64,
"learning_rate": 0.0001828284854563691,
"loss": 2.6408,
"theoretical_loss": 3.4157289938746125,
"tokens_seen": 2105999360
},
{
"epoch": 0.64,
"learning_rate": 0.00018281845536609827,
"loss": 2.7189,
"theoretical_loss": 3.415720221291834,
"tokens_seen": 2106064896
},
{
"epoch": 0.64,
"learning_rate": 0.00018280842527582748,
"loss": 2.6296,
"theoretical_loss": 3.4157114490584677,
"tokens_seen": 2106130432
},
{
"epoch": 0.64,
"learning_rate": 0.0001827983951855567,
"loss": 2.8604,
"theoretical_loss": 3.4157026771744876,
"tokens_seen": 2106195968
},
{
"epoch": 0.64,
"learning_rate": 0.00018278836509528587,
"loss": 2.6379,
"theoretical_loss": 3.415693905639869,
"tokens_seen": 2106261504
},
{
"epoch": 0.64,
"learning_rate": 0.00018277833500501505,
"loss": 2.5115,
"theoretical_loss": 3.4156851344545878,
"tokens_seen": 2106327040
},
{
"epoch": 0.64,
"learning_rate": 0.00018276830491474426,
"loss": 2.5204,
"theoretical_loss": 3.4156763636186183,
"tokens_seen": 2106392576
},
{
"epoch": 0.64,
"learning_rate": 0.00018275827482447344,
"loss": 2.7161,
"theoretical_loss": 3.4156675931319365,
"tokens_seen": 2106458112
},
{
"epoch": 0.64,
"learning_rate": 0.00018274824473420262,
"loss": 2.4431,
"theoretical_loss": 3.4156588229945175,
"tokens_seen": 2106523648
},
{
"epoch": 0.64,
"learning_rate": 0.0001827382146439318,
"loss": 2.625,
"theoretical_loss": 3.415650053206336,
"tokens_seen": 2106589184
},
{
"epoch": 0.64,
"learning_rate": 0.00018272818455366098,
"loss": 2.562,
"theoretical_loss": 3.4156412837673678,
"tokens_seen": 2106654720
},
{
"epoch": 0.64,
"learning_rate": 0.0001827181544633902,
"loss": 2.5238,
"theoretical_loss": 3.415632514677588,
"tokens_seen": 2106720256
},
{
"epoch": 0.64,
"learning_rate": 0.00018270812437311937,
"loss": 2.5962,
"theoretical_loss": 3.415623745936972,
"tokens_seen": 2106785792
},
{
"epoch": 0.64,
"learning_rate": 0.00018269809428284856,
"loss": 2.4742,
"theoretical_loss": 3.415614977545494,
"tokens_seen": 2106851328
},
{
"epoch": 0.64,
"learning_rate": 0.00018268806419257774,
"loss": 2.681,
"theoretical_loss": 3.4156062095031308,
"tokens_seen": 2106916864
},
{
"epoch": 0.64,
"objective/train/docs_used": 1191719,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7179348468780518,
"objective/train/theoretical_loss": 3.4155974418098562,
"objective/train/tokens_used": 2127442400,
"theoretical_loss": 3.4155974418098562,
"tokens_seen": 2106982400
},
{
"epoch": 0.64,
"learning_rate": 0.00018267803410230694,
"loss": 2.7089,
"theoretical_loss": 3.4155974418098562,
"tokens_seen": 2106982400
},
{
"epoch": 0.64,
"learning_rate": 0.00018266800401203613,
"loss": 2.5453,
"theoretical_loss": 3.4155886744656465,
"tokens_seen": 2107047936
},
{
"epoch": 0.64,
"learning_rate": 0.0001826579739217653,
"loss": 2.7227,
"theoretical_loss": 3.4155799074704762,
"tokens_seen": 2107113472
},
{
"epoch": 0.64,
"learning_rate": 0.0001826479438314945,
"loss": 2.8224,
"theoretical_loss": 3.415571140824321,
"tokens_seen": 2107179008
},
{
"epoch": 0.64,
"learning_rate": 0.00018263791374122367,
"loss": 2.6104,
"theoretical_loss": 3.4155623745271564,
"tokens_seen": 2107244544
},
{
"epoch": 0.64,
"learning_rate": 0.00018262788365095288,
"loss": 2.7527,
"theoretical_loss": 3.415553608578957,
"tokens_seen": 2107310080
},
{
"epoch": 0.64,
"learning_rate": 0.00018261785356068206,
"loss": 2.5263,
"theoretical_loss": 3.4155448429796986,
"tokens_seen": 2107375616
},
{
"epoch": 0.64,
"learning_rate": 0.00018260782347041124,
"loss": 2.6807,
"theoretical_loss": 3.415536077729356,
"tokens_seen": 2107441152
},
{
"epoch": 0.64,
"learning_rate": 0.00018259779338014042,
"loss": 2.727,
"theoretical_loss": 3.415527312827905,
"tokens_seen": 2107506688
},
{
"epoch": 0.64,
"learning_rate": 0.00018258776328986963,
"loss": 2.61,
"theoretical_loss": 3.4155185482753203,
"tokens_seen": 2107572224
},
{
"epoch": 0.64,
"learning_rate": 0.0001825777331995988,
"loss": 2.7127,
"theoretical_loss": 3.4155097840715776,
"tokens_seen": 2107637760
},
{
"epoch": 0.64,
"learning_rate": 0.000182567703109328,
"loss": 2.5179,
"theoretical_loss": 3.415501020216652,
"tokens_seen": 2107703296
},
{
"epoch": 0.64,
"learning_rate": 0.00018255767301905717,
"loss": 2.3244,
"theoretical_loss": 3.415492256710518,
"tokens_seen": 2107768832
},
{
"epoch": 0.64,
"learning_rate": 0.00018254764292878638,
"loss": 2.5427,
"theoretical_loss": 3.4154834935531526,
"tokens_seen": 2107834368
},
{
"epoch": 0.64,
"learning_rate": 0.00018253761283851556,
"loss": 2.7534,
"theoretical_loss": 3.41547473074453,
"tokens_seen": 2107899904
},
{
"epoch": 0.64,
"learning_rate": 0.00018252758274824474,
"loss": 2.7843,
"theoretical_loss": 3.415465968284625,
"tokens_seen": 2107965440
},
{
"epoch": 0.64,
"learning_rate": 0.00018251755265797392,
"loss": 2.7725,
"theoretical_loss": 3.4154572061734143,
"tokens_seen": 2108030976
},
{
"epoch": 0.64,
"learning_rate": 0.0001825075225677031,
"loss": 2.4929,
"theoretical_loss": 3.4154484444108717,
"tokens_seen": 2108096512
},
{
"epoch": 0.64,
"learning_rate": 0.0001824974924774323,
"loss": 2.5405,
"theoretical_loss": 3.4154396829969733,
"tokens_seen": 2108162048
},
{
"epoch": 0.64,
"learning_rate": 0.0001824874623871615,
"loss": 2.5934,
"theoretical_loss": 3.4154309219316943,
"tokens_seen": 2108227584
},
{
"epoch": 0.64,
"learning_rate": 0.00018247743229689068,
"loss": 2.6871,
"theoretical_loss": 3.4154221612150097,
"tokens_seen": 2108293120
},
{
"epoch": 0.64,
"learning_rate": 0.00018246740220661986,
"loss": 2.5932,
"theoretical_loss": 3.4154134008468953,
"tokens_seen": 2108358656
},
{
"epoch": 0.64,
"learning_rate": 0.00018245737211634906,
"loss": 2.7881,
"theoretical_loss": 3.415404640827326,
"tokens_seen": 2108424192
},
{
"epoch": 0.64,
"learning_rate": 0.00018244734202607825,
"loss": 2.5922,
"theoretical_loss": 3.4153958811562775,
"tokens_seen": 2108489728
},
{
"epoch": 0.64,
"learning_rate": 0.00018243731193580743,
"loss": 2.6742,
"theoretical_loss": 3.4153871218337244,
"tokens_seen": 2108555264
},
{
"epoch": 0.64,
"objective/train/docs_used": 1192553,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.772707223892212,
"objective/train/theoretical_loss": 3.4153783628596424,
"objective/train/tokens_used": 2129080800,
"theoretical_loss": 3.4153783628596424,
"tokens_seen": 2108620800
},
{
"epoch": 0.64,
"learning_rate": 0.0001824272818455366,
"loss": 2.4799,
"theoretical_loss": 3.4153783628596424,
"tokens_seen": 2108620800
},
{
"epoch": 0.64,
"learning_rate": 0.0001824172517552658,
"loss": 2.8177,
"theoretical_loss": 3.415369604234007,
"tokens_seen": 2108686336
},
{
"epoch": 0.64,
"learning_rate": 0.000182407221664995,
"loss": 2.4638,
"theoretical_loss": 3.4153608459567932,
"tokens_seen": 2108751872
},
{
"epoch": 0.64,
"learning_rate": 0.00018239719157472418,
"loss": 2.4879,
"theoretical_loss": 3.415352088027977,
"tokens_seen": 2108817408
},
{
"epoch": 0.64,
"learning_rate": 0.00018238716148445336,
"loss": 2.6279,
"theoretical_loss": 3.4153433304475325,
"tokens_seen": 2108882944
},
{
"epoch": 0.64,
"learning_rate": 0.00018237713139418254,
"loss": 2.7294,
"theoretical_loss": 3.415334573215436,
"tokens_seen": 2108948480
},
{
"epoch": 0.64,
"learning_rate": 0.00018236710130391175,
"loss": 2.5318,
"theoretical_loss": 3.4153258163316624,
"tokens_seen": 2109014016
},
{
"epoch": 0.64,
"learning_rate": 0.00018235707121364093,
"loss": 2.5772,
"theoretical_loss": 3.4153170597961875,
"tokens_seen": 2109079552
},
{
"epoch": 0.64,
"learning_rate": 0.0001823470411233701,
"loss": 2.5613,
"theoretical_loss": 3.4153083036089855,
"tokens_seen": 2109145088
},
{
"epoch": 0.64,
"learning_rate": 0.0001823370110330993,
"loss": 2.613,
"theoretical_loss": 3.415299547770033,
"tokens_seen": 2109210624
},
{
"epoch": 0.64,
"learning_rate": 0.00018232698094282847,
"loss": 2.8454,
"theoretical_loss": 3.415290792279305,
"tokens_seen": 2109276160
},
{
"epoch": 0.64,
"learning_rate": 0.00018231695085255768,
"loss": 2.5326,
"theoretical_loss": 3.415282037136776,
"tokens_seen": 2109341696
},
{
"epoch": 0.64,
"learning_rate": 0.00018230692076228686,
"loss": 2.5627,
"theoretical_loss": 3.4152732823424223,
"tokens_seen": 2109407232
},
{
"epoch": 0.64,
"learning_rate": 0.00018229689067201604,
"loss": 2.5462,
"theoretical_loss": 3.415264527896219,
"tokens_seen": 2109472768
},
{
"epoch": 0.64,
"learning_rate": 0.00018228686058174523,
"loss": 2.3246,
"theoretical_loss": 3.4152557737981413,
"tokens_seen": 2109538304
},
{
"epoch": 0.64,
"learning_rate": 0.00018227683049147443,
"loss": 2.6278,
"theoretical_loss": 3.4152470200481644,
"tokens_seen": 2109603840
},
{
"epoch": 0.64,
"learning_rate": 0.00018226680040120361,
"loss": 2.4245,
"theoretical_loss": 3.4152382666462637,
"tokens_seen": 2109669376
},
{
"epoch": 0.64,
"learning_rate": 0.0001822567703109328,
"loss": 2.5179,
"theoretical_loss": 3.4152295135924153,
"tokens_seen": 2109734912
},
{
"epoch": 0.64,
"learning_rate": 0.00018224674022066198,
"loss": 2.5173,
"theoretical_loss": 3.4152207608865934,
"tokens_seen": 2109800448
},
{
"epoch": 0.64,
"learning_rate": 0.00018223671013039116,
"loss": 2.4665,
"theoretical_loss": 3.4152120085287745,
"tokens_seen": 2109865984
},
{
"epoch": 0.64,
"learning_rate": 0.00018222668004012037,
"loss": 2.6608,
"theoretical_loss": 3.4152032565189323,
"tokens_seen": 2109931520
},
{
"epoch": 0.64,
"learning_rate": 0.00018221664994984955,
"loss": 2.5452,
"theoretical_loss": 3.415194504857044,
"tokens_seen": 2109997056
},
{
"epoch": 0.64,
"learning_rate": 0.00018220661985957873,
"loss": 2.6356,
"theoretical_loss": 3.415185753543084,
"tokens_seen": 2110062592
},
{
"epoch": 0.64,
"learning_rate": 0.0001821965897693079,
"loss": 2.707,
"theoretical_loss": 3.415177002577028,
"tokens_seen": 2110128128
},
{
"epoch": 0.64,
"learning_rate": 0.00018218655967903712,
"loss": 2.6658,
"theoretical_loss": 3.4151682519588507,
"tokens_seen": 2110193664
},
{
"epoch": 0.64,
"objective/train/docs_used": 1193312,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7820699214935303,
"objective/train/theoretical_loss": 3.415159501688528,
"objective/train/tokens_used": 2130719200,
"theoretical_loss": 3.415159501688528,
"tokens_seen": 2110259200
},
{
"epoch": 0.64,
"learning_rate": 0.0001821765295887663,
"loss": 2.4133,
"theoretical_loss": 3.415159501688528,
"tokens_seen": 2110259200
},
{
"epoch": 0.64,
"learning_rate": 0.00018216649949849548,
"loss": 2.7574,
"theoretical_loss": 3.4151507517660358,
"tokens_seen": 2110324736
},
{
"epoch": 0.64,
"learning_rate": 0.00018215646940822466,
"loss": 2.6511,
"theoretical_loss": 3.4151420021913483,
"tokens_seen": 2110390272
},
{
"epoch": 0.64,
"learning_rate": 0.00018214643931795384,
"loss": 2.8395,
"theoretical_loss": 3.415133252964442,
"tokens_seen": 2110455808
},
{
"epoch": 0.64,
"learning_rate": 0.00018213640922768305,
"loss": 2.5798,
"theoretical_loss": 3.4151245040852913,
"tokens_seen": 2110521344
},
{
"epoch": 0.64,
"learning_rate": 0.00018212637913741223,
"loss": 2.568,
"theoretical_loss": 3.415115755553872,
"tokens_seen": 2110586880
},
{
"epoch": 0.64,
"learning_rate": 0.0001821163490471414,
"loss": 2.354,
"theoretical_loss": 3.4151070073701595,
"tokens_seen": 2110652416
},
{
"epoch": 0.64,
"learning_rate": 0.0001821063189568706,
"loss": 2.619,
"theoretical_loss": 3.4150982595341297,
"tokens_seen": 2110717952
},
{
"epoch": 0.64,
"learning_rate": 0.0001820962888665998,
"loss": 2.623,
"theoretical_loss": 3.415089512045757,
"tokens_seen": 2110783488
},
{
"epoch": 0.64,
"learning_rate": 0.00018208625877632898,
"loss": 2.5698,
"theoretical_loss": 3.415080764905017,
"tokens_seen": 2110849024
},
{
"epoch": 0.64,
"learning_rate": 0.00018207622868605816,
"loss": 2.5046,
"theoretical_loss": 3.415072018111886,
"tokens_seen": 2110914560
},
{
"epoch": 0.64,
"learning_rate": 0.00018206619859578735,
"loss": 2.4274,
"theoretical_loss": 3.4150632716663387,
"tokens_seen": 2110980096
},
{
"epoch": 0.64,
"learning_rate": 0.00018205616850551658,
"loss": 2.3801,
"theoretical_loss": 3.41505452556835,
"tokens_seen": 2111045632
},
{
"epoch": 0.64,
"learning_rate": 0.00018204613841524576,
"loss": 2.6191,
"theoretical_loss": 3.4150457798178966,
"tokens_seen": 2111111168
},
{
"epoch": 0.64,
"learning_rate": 0.00018203610832497494,
"loss": 2.5005,
"theoretical_loss": 3.4150370344149525,
"tokens_seen": 2111176704
},
{
"epoch": 0.64,
"learning_rate": 0.00018202607823470412,
"loss": 2.5251,
"theoretical_loss": 3.415028289359494,
"tokens_seen": 2111242240
},
{
"epoch": 0.64,
"learning_rate": 0.0001820160481444333,
"loss": 2.5091,
"theoretical_loss": 3.4150195446514964,
"tokens_seen": 2111307776
},
{
"epoch": 0.64,
"learning_rate": 0.0001820060180541625,
"loss": 2.4751,
"theoretical_loss": 3.415010800290935,
"tokens_seen": 2111373312
},
{
"epoch": 0.64,
"learning_rate": 0.0001819959879638917,
"loss": 2.6479,
"theoretical_loss": 3.4150020562777854,
"tokens_seen": 2111438848
},
{
"epoch": 0.64,
"learning_rate": 0.00018198595787362088,
"loss": 2.6895,
"theoretical_loss": 3.4149933126120224,
"tokens_seen": 2111504384
},
{
"epoch": 0.64,
"learning_rate": 0.00018197592778335006,
"loss": 2.7311,
"theoretical_loss": 3.4149845692936216,
"tokens_seen": 2111569920
},
{
"epoch": 0.64,
"learning_rate": 0.00018196589769307926,
"loss": 2.6444,
"theoretical_loss": 3.414975826322559,
"tokens_seen": 2111635456
},
{
"epoch": 0.64,
"learning_rate": 0.00018195586760280845,
"loss": 2.3734,
"theoretical_loss": 3.4149670836988095,
"tokens_seen": 2111700992
},
{
"epoch": 0.64,
"learning_rate": 0.00018194583751253763,
"loss": 2.8941,
"theoretical_loss": 3.4149583414223486,
"tokens_seen": 2111766528
},
{
"epoch": 0.64,
"learning_rate": 0.0001819358074222668,
"loss": 2.6304,
"theoretical_loss": 3.414949599493152,
"tokens_seen": 2111832064
},
{
"epoch": 0.64,
"objective/train/docs_used": 1194755,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4481608867645264,
"objective/train/theoretical_loss": 3.414940857911195,
"objective/train/tokens_used": 2132357600,
"theoretical_loss": 3.414940857911195,
"tokens_seen": 2111897600
},
{
"epoch": 0.64,
"learning_rate": 0.000181925777331996,
"loss": 2.7124,
"theoretical_loss": 3.414940857911195,
"tokens_seen": 2111897600
},
{
"epoch": 0.64,
"learning_rate": 0.0001819157472417252,
"loss": 2.6461,
"theoretical_loss": 3.4149321166764532,
"tokens_seen": 2111963136
},
{
"epoch": 0.64,
"learning_rate": 0.00018190571715145438,
"loss": 2.4856,
"theoretical_loss": 3.4149233757889013,
"tokens_seen": 2112028672
},
{
"epoch": 0.64,
"learning_rate": 0.00018189568706118356,
"loss": 2.5368,
"theoretical_loss": 3.4149146352485156,
"tokens_seen": 2112094208
},
{
"epoch": 0.64,
"learning_rate": 0.00018188565697091274,
"loss": 2.5156,
"theoretical_loss": 3.414905895055271,
"tokens_seen": 2112159744
},
{
"epoch": 0.64,
"learning_rate": 0.00018187562688064195,
"loss": 2.4766,
"theoretical_loss": 3.414897155209143,
"tokens_seen": 2112225280
},
{
"epoch": 0.64,
"learning_rate": 0.00018186559679037113,
"loss": 2.4885,
"theoretical_loss": 3.4148884157101076,
"tokens_seen": 2112290816
},
{
"epoch": 0.64,
"learning_rate": 0.0001818555667001003,
"loss": 2.4454,
"theoretical_loss": 3.4148796765581397,
"tokens_seen": 2112356352
},
{
"epoch": 0.64,
"learning_rate": 0.0001818455366098295,
"loss": 2.5786,
"theoretical_loss": 3.414870937753215,
"tokens_seen": 2112421888
},
{
"epoch": 0.64,
"learning_rate": 0.00018183550651955867,
"loss": 2.5912,
"theoretical_loss": 3.4148621992953085,
"tokens_seen": 2112487424
},
{
"epoch": 0.64,
"learning_rate": 0.00018182547642928788,
"loss": 2.5874,
"theoretical_loss": 3.4148534611843964,
"tokens_seen": 2112552960
},
{
"epoch": 0.64,
"learning_rate": 0.00018181544633901706,
"loss": 2.6301,
"theoretical_loss": 3.414844723420453,
"tokens_seen": 2112618496
},
{
"epoch": 0.64,
"learning_rate": 0.00018180541624874624,
"loss": 2.6097,
"theoretical_loss": 3.414835986003455,
"tokens_seen": 2112684032
},
{
"epoch": 0.64,
"learning_rate": 0.00018179538615847543,
"loss": 2.5245,
"theoretical_loss": 3.4148272489333773,
"tokens_seen": 2112749568
},
{
"epoch": 0.64,
"learning_rate": 0.00018178535606820463,
"loss": 2.376,
"theoretical_loss": 3.4148185122101955,
"tokens_seen": 2112815104
},
{
"epoch": 0.64,
"learning_rate": 0.00018177532597793381,
"loss": 2.4556,
"theoretical_loss": 3.414809775833885,
"tokens_seen": 2112880640
},
{
"epoch": 0.64,
"learning_rate": 0.000181765295887663,
"loss": 2.3115,
"theoretical_loss": 3.414801039804421,
"tokens_seen": 2112946176
},
{
"epoch": 0.64,
"learning_rate": 0.00018175526579739218,
"loss": 2.6706,
"theoretical_loss": 3.4147923041217796,
"tokens_seen": 2113011712
},
{
"epoch": 0.64,
"learning_rate": 0.00018174523570712136,
"loss": 2.5575,
"theoretical_loss": 3.4147835687859356,
"tokens_seen": 2113077248
},
{
"epoch": 0.64,
"learning_rate": 0.00018173520561685057,
"loss": 2.5919,
"theoretical_loss": 3.414774833796865,
"tokens_seen": 2113142784
},
{
"epoch": 0.64,
"learning_rate": 0.00018172517552657975,
"loss": 2.6051,
"theoretical_loss": 3.414766099154543,
"tokens_seen": 2113208320
},
{
"epoch": 0.64,
"learning_rate": 0.00018171514543630893,
"loss": 2.5856,
"theoretical_loss": 3.414757364858945,
"tokens_seen": 2113273856
},
{
"epoch": 0.64,
"learning_rate": 0.0001817051153460381,
"loss": 2.7819,
"theoretical_loss": 3.414748630910047,
"tokens_seen": 2113339392
},
{
"epoch": 0.64,
"learning_rate": 0.00018169508525576732,
"loss": 2.5824,
"theoretical_loss": 3.414739897307824,
"tokens_seen": 2113404928
},
{
"epoch": 0.64,
"learning_rate": 0.0001816850551654965,
"loss": 2.7111,
"theoretical_loss": 3.4147311640522515,
"tokens_seen": 2113470464
},
{
"epoch": 0.64,
"objective/train/docs_used": 1195575,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.184577703475952,
"objective/train/theoretical_loss": 3.4147224311433053,
"objective/train/tokens_used": 2133996000,
"theoretical_loss": 3.4147224311433053,
"tokens_seen": 2113536000
},
{
"epoch": 0.64,
"learning_rate": 0.00018167502507522568,
"loss": 2.598,
"theoretical_loss": 3.4147224311433053,
"tokens_seen": 2113536000
},
{
"epoch": 0.64,
"learning_rate": 0.00018166499498495486,
"loss": 2.7262,
"theoretical_loss": 3.41471369858096,
"tokens_seen": 2113601536
},
{
"epoch": 0.64,
"learning_rate": 0.00018165496489468404,
"loss": 2.621,
"theoretical_loss": 3.4147049663651927,
"tokens_seen": 2113667072
},
{
"epoch": 0.64,
"learning_rate": 0.00018164493480441325,
"loss": 2.622,
"theoretical_loss": 3.4146962344959775,
"tokens_seen": 2113732608
},
{
"epoch": 0.64,
"learning_rate": 0.00018163490471414243,
"loss": 2.3986,
"theoretical_loss": 3.4146875029732904,
"tokens_seen": 2113798144
},
{
"epoch": 0.64,
"learning_rate": 0.0001816248746238716,
"loss": 2.7694,
"theoretical_loss": 3.4146787717971074,
"tokens_seen": 2113863680
},
{
"epoch": 0.64,
"learning_rate": 0.0001816148445336008,
"loss": 2.7552,
"theoretical_loss": 3.414670040967403,
"tokens_seen": 2113929216
},
{
"epoch": 0.64,
"learning_rate": 0.00018160481444333,
"loss": 2.457,
"theoretical_loss": 3.4146613104841532,
"tokens_seen": 2113994752
},
{
"epoch": 0.64,
"learning_rate": 0.00018159478435305918,
"loss": 2.4325,
"theoretical_loss": 3.414652580347334,
"tokens_seen": 2114060288
},
{
"epoch": 0.64,
"learning_rate": 0.00018158475426278836,
"loss": 2.7953,
"theoretical_loss": 3.41464385055692,
"tokens_seen": 2114125824
},
{
"epoch": 0.64,
"learning_rate": 0.00018157472417251755,
"loss": 2.6307,
"theoretical_loss": 3.4146351211128874,
"tokens_seen": 2114191360
},
{
"epoch": 0.64,
"learning_rate": 0.00018156469408224673,
"loss": 2.5925,
"theoretical_loss": 3.4146263920152116,
"tokens_seen": 2114256896
},
{
"epoch": 0.64,
"learning_rate": 0.00018155466399197593,
"loss": 2.599,
"theoretical_loss": 3.4146176632638676,
"tokens_seen": 2114322432
},
{
"epoch": 0.64,
"learning_rate": 0.00018154463390170512,
"loss": 2.5751,
"theoretical_loss": 3.4146089348588315,
"tokens_seen": 2114387968
},
{
"epoch": 0.64,
"learning_rate": 0.0001815346038114343,
"loss": 2.4655,
"theoretical_loss": 3.4146002068000785,
"tokens_seen": 2114453504
},
{
"epoch": 0.64,
"learning_rate": 0.00018152457372116348,
"loss": 2.5738,
"theoretical_loss": 3.4145914790875844,
"tokens_seen": 2114519040
},
{
"epoch": 0.64,
"learning_rate": 0.00018151454363089269,
"loss": 2.7619,
"theoretical_loss": 3.414582751721325,
"tokens_seen": 2114584576
},
{
"epoch": 0.64,
"learning_rate": 0.00018150451354062187,
"loss": 2.3665,
"theoretical_loss": 3.4145740247012752,
"tokens_seen": 2114650112
},
{
"epoch": 0.64,
"learning_rate": 0.00018149448345035105,
"loss": 2.5778,
"theoretical_loss": 3.414565298027411,
"tokens_seen": 2114715648
},
{
"epoch": 0.64,
"learning_rate": 0.00018148445336008023,
"loss": 2.4957,
"theoretical_loss": 3.4145565716997073,
"tokens_seen": 2114781184
},
{
"epoch": 0.64,
"learning_rate": 0.0001814744232698094,
"loss": 2.5489,
"theoretical_loss": 3.41454784571814,
"tokens_seen": 2114846720
},
{
"epoch": 0.64,
"learning_rate": 0.00018146439317953862,
"loss": 2.4506,
"theoretical_loss": 3.4145391200826856,
"tokens_seen": 2114912256
},
{
"epoch": 0.64,
"learning_rate": 0.0001814543630892678,
"loss": 2.5852,
"theoretical_loss": 3.414530394793318,
"tokens_seen": 2114977792
},
{
"epoch": 0.64,
"learning_rate": 0.00018144433299899698,
"loss": 2.5821,
"theoretical_loss": 3.4145216698500134,
"tokens_seen": 2115043328
},
{
"epoch": 0.64,
"learning_rate": 0.00018143430290872616,
"loss": 2.7482,
"theoretical_loss": 3.414512945252748,
"tokens_seen": 2115108864
},
{
"epoch": 0.64,
"objective/train/docs_used": 1196855,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4625720977783203,
"objective/train/theoretical_loss": 3.414504221001496,
"objective/train/tokens_used": 2135634400,
"theoretical_loss": 3.414504221001496,
"tokens_seen": 2115174400
},
{
"epoch": 0.64,
"learning_rate": 0.00018142427281845537,
"loss": 2.6316,
"theoretical_loss": 3.414504221001496,
"tokens_seen": 2115174400
},
{
"epoch": 0.64,
"learning_rate": 0.00018141424272818455,
"loss": 2.4523,
"theoretical_loss": 3.414495497096235,
"tokens_seen": 2115239936
},
{
"epoch": 0.64,
"learning_rate": 0.00018140421263791373,
"loss": 2.5295,
"theoretical_loss": 3.4144867735369386,
"tokens_seen": 2115305472
},
{
"epoch": 0.64,
"learning_rate": 0.00018139418254764291,
"loss": 2.4652,
"theoretical_loss": 3.414478050323583,
"tokens_seen": 2115371008
},
{
"epoch": 0.64,
"learning_rate": 0.00018138415245737212,
"loss": 2.7452,
"theoretical_loss": 3.4144693274561444,
"tokens_seen": 2115436544
},
{
"epoch": 0.64,
"learning_rate": 0.0001813741223671013,
"loss": 2.7273,
"theoretical_loss": 3.414460604934597,
"tokens_seen": 2115502080
},
{
"epoch": 0.64,
"learning_rate": 0.00018136409227683048,
"loss": 2.328,
"theoretical_loss": 3.414451882758918,
"tokens_seen": 2115567616
},
{
"epoch": 0.64,
"learning_rate": 0.00018135406218655967,
"loss": 2.6348,
"theoretical_loss": 3.414443160929082,
"tokens_seen": 2115633152
},
{
"epoch": 0.64,
"learning_rate": 0.00018134403209628885,
"loss": 2.6815,
"theoretical_loss": 3.4144344394450643,
"tokens_seen": 2115698688
},
{
"epoch": 0.64,
"learning_rate": 0.00018133400200601805,
"loss": 2.5552,
"theoretical_loss": 3.414425718306841,
"tokens_seen": 2115764224
},
{
"epoch": 0.64,
"learning_rate": 0.00018132397191574724,
"loss": 2.5105,
"theoretical_loss": 3.4144169975143885,
"tokens_seen": 2115829760
},
{
"epoch": 0.64,
"learning_rate": 0.00018131394182547642,
"loss": 2.8175,
"theoretical_loss": 3.4144082770676807,
"tokens_seen": 2115895296
},
{
"epoch": 0.64,
"learning_rate": 0.00018130391173520563,
"loss": 2.6059,
"theoretical_loss": 3.4143995569666936,
"tokens_seen": 2115960832
},
{
"epoch": 0.64,
"learning_rate": 0.00018129388164493483,
"loss": 2.3676,
"theoretical_loss": 3.4143908372114042,
"tokens_seen": 2116026368
},
{
"epoch": 0.64,
"learning_rate": 0.00018128385155466401,
"loss": 2.5323,
"theoretical_loss": 3.4143821178017864,
"tokens_seen": 2116091904
},
{
"epoch": 0.64,
"learning_rate": 0.0001812738214643932,
"loss": 2.571,
"theoretical_loss": 3.4143733987378164,
"tokens_seen": 2116157440
},
{
"epoch": 0.64,
"learning_rate": 0.00018126379137412238,
"loss": 2.6626,
"theoretical_loss": 3.41436468001947,
"tokens_seen": 2116222976
},
{
"epoch": 0.64,
"learning_rate": 0.00018125376128385156,
"loss": 2.4461,
"theoretical_loss": 3.4143559616467227,
"tokens_seen": 2116288512
},
{
"epoch": 0.64,
"learning_rate": 0.00018124373119358077,
"loss": 2.5682,
"theoretical_loss": 3.41434724361955,
"tokens_seen": 2116354048
},
{
"epoch": 0.64,
"learning_rate": 0.00018123370110330995,
"loss": 2.6133,
"theoretical_loss": 3.4143385259379277,
"tokens_seen": 2116419584
},
{
"epoch": 0.64,
"learning_rate": 0.00018122367101303913,
"loss": 2.439,
"theoretical_loss": 3.4143298086018308,
"tokens_seen": 2116485120
},
{
"epoch": 0.64,
"learning_rate": 0.0001812136409227683,
"loss": 2.4292,
"theoretical_loss": 3.4143210916112356,
"tokens_seen": 2116550656
},
{
"epoch": 0.64,
"learning_rate": 0.00018120361083249752,
"loss": 2.4931,
"theoretical_loss": 3.4143123749661175,
"tokens_seen": 2116616192
},
{
"epoch": 0.64,
"learning_rate": 0.0001811935807422267,
"loss": 2.757,
"theoretical_loss": 3.414303658666452,
"tokens_seen": 2116681728
},
{
"epoch": 0.64,
"learning_rate": 0.00018118355065195588,
"loss": 2.8528,
"theoretical_loss": 3.414294942712215,
"tokens_seen": 2116747264
},
{
"epoch": 0.64,
"objective/train/docs_used": 1197590,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7894322872161865,
"objective/train/theoretical_loss": 3.414286227103381,
"objective/train/tokens_used": 2137272800,
"theoretical_loss": 3.414286227103381,
"tokens_seen": 2116812800
},
{
"epoch": 0.64,
"learning_rate": 0.00018117352056168506,
"loss": 2.8394,
"theoretical_loss": 3.414286227103381,
"tokens_seen": 2116812800
},
{
"epoch": 0.64,
"learning_rate": 0.00018116349047141424,
"loss": 2.6647,
"theoretical_loss": 3.4142775118399276,
"tokens_seen": 2116878336
},
{
"epoch": 0.64,
"learning_rate": 0.00018115346038114345,
"loss": 2.7627,
"theoretical_loss": 3.414268796921829,
"tokens_seen": 2116943872
},
{
"epoch": 0.64,
"learning_rate": 0.00018114343029087263,
"loss": 2.7222,
"theoretical_loss": 3.414260082349061,
"tokens_seen": 2117009408
},
{
"epoch": 0.64,
"learning_rate": 0.0001811334002006018,
"loss": 2.797,
"theoretical_loss": 3.414251368121599,
"tokens_seen": 2117074944
},
{
"epoch": 0.64,
"learning_rate": 0.000181123370110331,
"loss": 2.5599,
"theoretical_loss": 3.4142426542394197,
"tokens_seen": 2117140480
},
{
"epoch": 0.64,
"learning_rate": 0.0001811133400200602,
"loss": 2.9002,
"theoretical_loss": 3.414233940702498,
"tokens_seen": 2117206016
},
{
"epoch": 0.64,
"learning_rate": 0.00018110330992978938,
"loss": 2.4219,
"theoretical_loss": 3.4142252275108094,
"tokens_seen": 2117271552
},
{
"epoch": 0.64,
"learning_rate": 0.00018109327983951856,
"loss": 2.5783,
"theoretical_loss": 3.4142165146643295,
"tokens_seen": 2117337088
},
{
"epoch": 0.64,
"learning_rate": 0.00018108324974924775,
"loss": 2.6472,
"theoretical_loss": 3.4142078021630344,
"tokens_seen": 2117402624
},
{
"epoch": 0.64,
"learning_rate": 0.00018107321965897693,
"loss": 2.3964,
"theoretical_loss": 3.4141990900068997,
"tokens_seen": 2117468160
},
{
"epoch": 0.64,
"learning_rate": 0.00018106318956870613,
"loss": 2.6591,
"theoretical_loss": 3.4141903781959004,
"tokens_seen": 2117533696
},
{
"epoch": 0.64,
"learning_rate": 0.00018105315947843532,
"loss": 2.7258,
"theoretical_loss": 3.4141816667300127,
"tokens_seen": 2117599232
},
{
"epoch": 0.64,
"learning_rate": 0.0001810431293881645,
"loss": 2.6961,
"theoretical_loss": 3.414172955609212,
"tokens_seen": 2117664768
},
{
"epoch": 0.64,
"learning_rate": 0.00018103309929789368,
"loss": 2.4577,
"theoretical_loss": 3.414164244833475,
"tokens_seen": 2117730304
},
{
"epoch": 0.64,
"learning_rate": 0.0001810230692076229,
"loss": 2.5196,
"theoretical_loss": 3.4141555344027754,
"tokens_seen": 2117795840
},
{
"epoch": 0.64,
"learning_rate": 0.00018101303911735207,
"loss": 2.5041,
"theoretical_loss": 3.41414682431709,
"tokens_seen": 2117861376
},
{
"epoch": 0.64,
"learning_rate": 0.00018100300902708125,
"loss": 2.3757,
"theoretical_loss": 3.414138114576395,
"tokens_seen": 2117926912
},
{
"epoch": 0.64,
"learning_rate": 0.00018099297893681043,
"loss": 2.4492,
"theoretical_loss": 3.414129405180665,
"tokens_seen": 2117992448
},
{
"epoch": 0.64,
"learning_rate": 0.0001809829488465396,
"loss": 2.6857,
"theoretical_loss": 3.414120696129876,
"tokens_seen": 2118057984
},
{
"epoch": 0.64,
"learning_rate": 0.00018097291875626882,
"loss": 2.4634,
"theoretical_loss": 3.4141119874240036,
"tokens_seen": 2118123520
},
{
"epoch": 0.64,
"learning_rate": 0.000180962888665998,
"loss": 2.7013,
"theoretical_loss": 3.414103279063024,
"tokens_seen": 2118189056
},
{
"epoch": 0.64,
"learning_rate": 0.00018095285857572718,
"loss": 2.4553,
"theoretical_loss": 3.4140945710469124,
"tokens_seen": 2118254592
},
{
"epoch": 0.64,
"learning_rate": 0.00018094282848545636,
"loss": 2.6421,
"theoretical_loss": 3.414085863375644,
"tokens_seen": 2118320128
},
{
"epoch": 0.64,
"learning_rate": 0.00018093279839518557,
"loss": 2.6383,
"theoretical_loss": 3.4140771560491956,
"tokens_seen": 2118385664
},
{
"epoch": 0.64,
"objective/train/docs_used": 1199039,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9594292640686035,
"objective/train/theoretical_loss": 3.4140684490675426,
"objective/train/tokens_used": 2138911200,
"theoretical_loss": 3.4140684490675426,
"tokens_seen": 2118451200
},
{
"epoch": 0.64,
"learning_rate": 0.00018092276830491475,
"loss": 2.7399,
"theoretical_loss": 3.4140684490675426,
"tokens_seen": 2118451200
},
{
"epoch": 0.64,
"learning_rate": 0.00018091273821464393,
"loss": 2.3562,
"theoretical_loss": 3.41405974243066,
"tokens_seen": 2118516736
},
{
"epoch": 0.64,
"learning_rate": 0.00018090270812437311,
"loss": 2.5007,
"theoretical_loss": 3.414051036138524,
"tokens_seen": 2118582272
},
{
"epoch": 0.64,
"learning_rate": 0.00018089267803410232,
"loss": 2.703,
"theoretical_loss": 3.41404233019111,
"tokens_seen": 2118647808
},
{
"epoch": 0.64,
"learning_rate": 0.0001808826479438315,
"loss": 2.4813,
"theoretical_loss": 3.4140336245883933,
"tokens_seen": 2118713344
},
{
"epoch": 0.64,
"learning_rate": 0.00018087261785356068,
"loss": 2.7247,
"theoretical_loss": 3.414024919330351,
"tokens_seen": 2118778880
},
{
"epoch": 0.64,
"learning_rate": 0.00018086258776328987,
"loss": 2.6722,
"theoretical_loss": 3.4140162144169577,
"tokens_seen": 2118844416
},
{
"epoch": 0.64,
"learning_rate": 0.00018085255767301905,
"loss": 2.6622,
"theoretical_loss": 3.4140075098481892,
"tokens_seen": 2118909952
},
{
"epoch": 0.64,
"learning_rate": 0.00018084252758274826,
"loss": 2.8223,
"theoretical_loss": 3.4139988056240216,
"tokens_seen": 2118975488
},
{
"epoch": 0.64,
"learning_rate": 0.00018083249749247744,
"loss": 2.5812,
"theoretical_loss": 3.41399010174443,
"tokens_seen": 2119041024
},
{
"epoch": 0.64,
"learning_rate": 0.00018082246740220662,
"loss": 2.5014,
"theoretical_loss": 3.4139813982093905,
"tokens_seen": 2119106560
},
{
"epoch": 0.64,
"learning_rate": 0.0001808124373119358,
"loss": 2.7256,
"theoretical_loss": 3.413972695018879,
"tokens_seen": 2119172096
},
{
"epoch": 0.64,
"learning_rate": 0.000180802407221665,
"loss": 2.6476,
"theoretical_loss": 3.413963992172871,
"tokens_seen": 2119237632
},
{
"epoch": 0.64,
"learning_rate": 0.0001807923771313942,
"loss": 2.3125,
"theoretical_loss": 3.4139552896713417,
"tokens_seen": 2119303168
},
{
"epoch": 0.64,
"learning_rate": 0.00018078234704112337,
"loss": 2.7353,
"theoretical_loss": 3.4139465875142676,
"tokens_seen": 2119368704
},
{
"epoch": 0.64,
"learning_rate": 0.00018077231695085255,
"loss": 2.5002,
"theoretical_loss": 3.413937885701624,
"tokens_seen": 2119434240
},
{
"epoch": 0.64,
"learning_rate": 0.00018076228686058173,
"loss": 2.5338,
"theoretical_loss": 3.4139291842333863,
"tokens_seen": 2119499776
},
{
"epoch": 0.64,
"learning_rate": 0.00018075225677031094,
"loss": 2.4649,
"theoretical_loss": 3.413920483109531,
"tokens_seen": 2119565312
},
{
"epoch": 0.64,
"learning_rate": 0.00018074222668004012,
"loss": 2.6776,
"theoretical_loss": 3.413911782330034,
"tokens_seen": 2119630848
},
{
"epoch": 0.64,
"learning_rate": 0.0001807321965897693,
"loss": 2.497,
"theoretical_loss": 3.41390308189487,
"tokens_seen": 2119696384
},
{
"epoch": 0.64,
"learning_rate": 0.00018072216649949848,
"loss": 2.6289,
"theoretical_loss": 3.413894381804015,
"tokens_seen": 2119761920
},
{
"epoch": 0.64,
"learning_rate": 0.0001807121364092277,
"loss": 2.4438,
"theoretical_loss": 3.413885682057445,
"tokens_seen": 2119827456
},
{
"epoch": 0.64,
"learning_rate": 0.00018070210631895687,
"loss": 2.5114,
"theoretical_loss": 3.413876982655135,
"tokens_seen": 2119892992
},
{
"epoch": 0.64,
"learning_rate": 0.00018069207622868605,
"loss": 2.8506,
"theoretical_loss": 3.4138682835970626,
"tokens_seen": 2119958528
},
{
"epoch": 0.64,
"learning_rate": 0.00018068204613841523,
"loss": 2.5756,
"theoretical_loss": 3.413859584883202,
"tokens_seen": 2120024064
},
{
"epoch": 0.64,
"objective/train/docs_used": 1199714,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3925373554229736,
"objective/train/theoretical_loss": 3.413850886513529,
"objective/train/tokens_used": 2140549600,
"theoretical_loss": 3.413850886513529,
"tokens_seen": 2120089600
},
{
"epoch": 0.64,
"learning_rate": 0.00018067201604814442,
"loss": 2.5539,
"theoretical_loss": 3.413850886513529,
"tokens_seen": 2120089600
},
{
"epoch": 0.64,
"learning_rate": 0.00018066198595787362,
"loss": 2.6856,
"theoretical_loss": 3.4138421884880197,
"tokens_seen": 2120155136
},
{
"epoch": 0.64,
"learning_rate": 0.0001806519558676028,
"loss": 2.5931,
"theoretical_loss": 3.41383349080665,
"tokens_seen": 2120220672
},
{
"epoch": 0.64,
"learning_rate": 0.00018064192577733199,
"loss": 2.5001,
"theoretical_loss": 3.413824793469395,
"tokens_seen": 2120286208
},
{
"epoch": 0.64,
"learning_rate": 0.00018063189568706117,
"loss": 2.7017,
"theoretical_loss": 3.4138160964762307,
"tokens_seen": 2120351744
},
{
"epoch": 0.64,
"learning_rate": 0.00018062186559679038,
"loss": 2.6889,
"theoretical_loss": 3.4138073998271334,
"tokens_seen": 2120417280
},
{
"epoch": 0.64,
"learning_rate": 0.00018061183550651956,
"loss": 2.5577,
"theoretical_loss": 3.4137987035220787,
"tokens_seen": 2120482816
},
{
"epoch": 0.64,
"learning_rate": 0.00018060180541624874,
"loss": 2.666,
"theoretical_loss": 3.4137900075610412,
"tokens_seen": 2120548352
},
{
"epoch": 0.64,
"learning_rate": 0.00018059177532597792,
"loss": 2.5801,
"theoretical_loss": 3.413781311943999,
"tokens_seen": 2120613888
},
{
"epoch": 0.64,
"learning_rate": 0.0001805817452357071,
"loss": 2.7018,
"theoretical_loss": 3.413772616670925,
"tokens_seen": 2120679424
},
{
"epoch": 0.64,
"learning_rate": 0.0001805717151454363,
"loss": 2.4921,
"theoretical_loss": 3.413763921741797,
"tokens_seen": 2120744960
},
{
"epoch": 0.64,
"learning_rate": 0.0001805616850551655,
"loss": 2.4735,
"theoretical_loss": 3.4137552271565905,
"tokens_seen": 2120810496
},
{
"epoch": 0.64,
"learning_rate": 0.0001805516549648947,
"loss": 2.6371,
"theoretical_loss": 3.4137465329152805,
"tokens_seen": 2120876032
},
{
"epoch": 0.64,
"learning_rate": 0.00018054162487462388,
"loss": 2.6825,
"theoretical_loss": 3.4137378390178434,
"tokens_seen": 2120941568
},
{
"epoch": 0.64,
"learning_rate": 0.0001805315947843531,
"loss": 2.5398,
"theoretical_loss": 3.413729145464255,
"tokens_seen": 2121007104
},
{
"epoch": 0.64,
"learning_rate": 0.00018052156469408227,
"loss": 2.5573,
"theoretical_loss": 3.4137204522544904,
"tokens_seen": 2121072640
},
{
"epoch": 0.64,
"learning_rate": 0.00018051153460381145,
"loss": 2.8421,
"theoretical_loss": 3.413711759388526,
"tokens_seen": 2121138176
},
{
"epoch": 0.64,
"learning_rate": 0.00018050150451354063,
"loss": 2.6357,
"theoretical_loss": 3.4137030668663373,
"tokens_seen": 2121203712
},
{
"epoch": 0.64,
"learning_rate": 0.0001804914744232698,
"loss": 2.6724,
"theoretical_loss": 3.4136943746879007,
"tokens_seen": 2121269248
},
{
"epoch": 0.64,
"learning_rate": 0.00018048144433299902,
"loss": 2.5775,
"theoretical_loss": 3.413685682853191,
"tokens_seen": 2121334784
},
{
"epoch": 0.64,
"learning_rate": 0.0001804714142427282,
"loss": 2.4543,
"theoretical_loss": 3.413676991362185,
"tokens_seen": 2121400320
},
{
"epoch": 0.64,
"learning_rate": 0.00018046138415245738,
"loss": 2.2496,
"theoretical_loss": 3.413668300214858,
"tokens_seen": 2121465856
},
{
"epoch": 0.64,
"learning_rate": 0.00018045135406218656,
"loss": 2.499,
"theoretical_loss": 3.413659609411185,
"tokens_seen": 2121531392
},
{
"epoch": 0.64,
"learning_rate": 0.00018044132397191577,
"loss": 2.609,
"theoretical_loss": 3.4136509189511433,
"tokens_seen": 2121596928
},
{
"epoch": 0.64,
"learning_rate": 0.00018043129388164495,
"loss": 2.6273,
"theoretical_loss": 3.4136422288347075,
"tokens_seen": 2121662464
},
{
"epoch": 0.64,
"objective/train/docs_used": 1200493,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.096813917160034,
"objective/train/theoretical_loss": 3.4136335390618537,
"objective/train/tokens_used": 2142188000,
"theoretical_loss": 3.4136335390618537,
"tokens_seen": 2121728000
},
{
"epoch": 0.64,
"learning_rate": 0.00018042126379137413,
"loss": 2.6141,
"theoretical_loss": 3.4136335390618537,
"tokens_seen": 2121728000
},
{
"epoch": 0.64,
"learning_rate": 0.00018041123370110331,
"loss": 2.6562,
"theoretical_loss": 3.4136248496325585,
"tokens_seen": 2121793536
},
{
"epoch": 0.64,
"learning_rate": 0.00018040120361083252,
"loss": 2.7098,
"theoretical_loss": 3.413616160546797,
"tokens_seen": 2121859072
},
{
"epoch": 0.64,
"learning_rate": 0.0001803911735205617,
"loss": 2.3597,
"theoretical_loss": 3.4136074718045446,
"tokens_seen": 2121924608
},
{
"epoch": 0.64,
"learning_rate": 0.00018038114343029088,
"loss": 2.7033,
"theoretical_loss": 3.413598783405778,
"tokens_seen": 2121990144
},
{
"epoch": 0.64,
"learning_rate": 0.00018037111334002007,
"loss": 2.6893,
"theoretical_loss": 3.4135900953504725,
"tokens_seen": 2122055680
},
{
"epoch": 0.64,
"learning_rate": 0.00018036108324974925,
"loss": 2.3445,
"theoretical_loss": 3.4135814076386044,
"tokens_seen": 2122121216
},
{
"epoch": 0.64,
"learning_rate": 0.00018035105315947846,
"loss": 2.6131,
"theoretical_loss": 3.4135727202701487,
"tokens_seen": 2122186752
},
{
"epoch": 0.64,
"learning_rate": 0.00018034102306920764,
"loss": 2.6132,
"theoretical_loss": 3.4135640332450814,
"tokens_seen": 2122252288
},
{
"epoch": 0.64,
"learning_rate": 0.00018033099297893682,
"loss": 2.6819,
"theoretical_loss": 3.4135553465633794,
"tokens_seen": 2122317824
},
{
"epoch": 0.64,
"learning_rate": 0.000180320962888666,
"loss": 2.516,
"theoretical_loss": 3.413546660225017,
"tokens_seen": 2122383360
},
{
"epoch": 0.64,
"learning_rate": 0.0001803109327983952,
"loss": 2.6707,
"theoretical_loss": 3.413537974229971,
"tokens_seen": 2122448896
},
{
"epoch": 0.64,
"learning_rate": 0.0001803009027081244,
"loss": 2.6088,
"theoretical_loss": 3.413529288578217,
"tokens_seen": 2122514432
},
{
"epoch": 0.64,
"learning_rate": 0.00018029087261785357,
"loss": 2.5117,
"theoretical_loss": 3.4135206032697307,
"tokens_seen": 2122579968
},
{
"epoch": 0.64,
"learning_rate": 0.00018028084252758275,
"loss": 2.4931,
"theoretical_loss": 3.413511918304488,
"tokens_seen": 2122645504
},
{
"epoch": 0.64,
"learning_rate": 0.00018027081243731193,
"loss": 2.6036,
"theoretical_loss": 3.413503233682465,
"tokens_seen": 2122711040
},
{
"epoch": 0.64,
"learning_rate": 0.00018026078234704114,
"loss": 2.5983,
"theoretical_loss": 3.4134945494036373,
"tokens_seen": 2122776576
},
{
"epoch": 0.64,
"learning_rate": 0.00018025075225677032,
"loss": 2.714,
"theoretical_loss": 3.4134858654679805,
"tokens_seen": 2122842112
},
{
"epoch": 0.64,
"learning_rate": 0.0001802407221664995,
"loss": 2.6452,
"theoretical_loss": 3.4134771818754714,
"tokens_seen": 2122907648
},
{
"epoch": 0.64,
"learning_rate": 0.00018023069207622868,
"loss": 2.4402,
"theoretical_loss": 3.4134684986260844,
"tokens_seen": 2122973184
},
{
"epoch": 0.64,
"learning_rate": 0.0001802206619859579,
"loss": 2.5299,
"theoretical_loss": 3.413459815719796,
"tokens_seen": 2123038720
},
{
"epoch": 0.64,
"learning_rate": 0.00018021063189568707,
"loss": 2.5459,
"theoretical_loss": 3.413451133156583,
"tokens_seen": 2123104256
},
{
"epoch": 0.64,
"learning_rate": 0.00018020060180541625,
"loss": 2.7145,
"theoretical_loss": 3.41344245093642,
"tokens_seen": 2123169792
},
{
"epoch": 0.64,
"learning_rate": 0.00018019057171514543,
"loss": 2.5119,
"theoretical_loss": 3.4134337690592833,
"tokens_seen": 2123235328
},
{
"epoch": 0.64,
"learning_rate": 0.00018018054162487462,
"loss": 2.562,
"theoretical_loss": 3.413425087525149,
"tokens_seen": 2123300864
},
{
"epoch": 0.64,
"objective/train/docs_used": 1201963,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5963897705078125,
"objective/train/theoretical_loss": 3.413416406333992,
"objective/train/tokens_used": 2143826400,
"theoretical_loss": 3.413416406333992,
"tokens_seen": 2123366400
},
{
"epoch": 0.64,
"learning_rate": 0.00018017051153460382,
"loss": 2.5936,
"theoretical_loss": 3.413416406333992,
"tokens_seen": 2123366400
},
{
"epoch": 0.64,
"learning_rate": 0.000180160481444333,
"loss": 2.5911,
"theoretical_loss": 3.413407725485789,
"tokens_seen": 2123431936
},
{
"epoch": 0.64,
"learning_rate": 0.00018015045135406219,
"loss": 2.5894,
"theoretical_loss": 3.4133990449805163,
"tokens_seen": 2123497472
},
{
"epoch": 0.64,
"learning_rate": 0.00018014042126379137,
"loss": 2.4545,
"theoretical_loss": 3.4133903648181487,
"tokens_seen": 2123563008
},
{
"epoch": 0.64,
"learning_rate": 0.00018013039117352058,
"loss": 2.6515,
"theoretical_loss": 3.413381684998663,
"tokens_seen": 2123628544
},
{
"epoch": 0.64,
"learning_rate": 0.00018012036108324976,
"loss": 2.6141,
"theoretical_loss": 3.4133730055220344,
"tokens_seen": 2123694080
},
{
"epoch": 0.64,
"learning_rate": 0.00018011033099297894,
"loss": 2.5107,
"theoretical_loss": 3.4133643263882387,
"tokens_seen": 2123759616
},
{
"epoch": 0.64,
"learning_rate": 0.00018010030090270812,
"loss": 2.4273,
"theoretical_loss": 3.413355647597253,
"tokens_seen": 2123825152
},
{
"epoch": 0.64,
"learning_rate": 0.0001800902708124373,
"loss": 2.5707,
"theoretical_loss": 3.4133469691490514,
"tokens_seen": 2123890688
},
{
"epoch": 0.64,
"learning_rate": 0.0001800802407221665,
"loss": 2.5632,
"theoretical_loss": 3.413338291043611,
"tokens_seen": 2123956224
},
{
"epoch": 0.64,
"learning_rate": 0.0001800702106318957,
"loss": 2.4204,
"theoretical_loss": 3.4133296132809074,
"tokens_seen": 2124021760
},
{
"epoch": 0.64,
"learning_rate": 0.00018006018054162487,
"loss": 2.6901,
"theoretical_loss": 3.4133209358609164,
"tokens_seen": 2124087296
},
{
"epoch": 0.64,
"learning_rate": 0.00018005015045135405,
"loss": 2.5714,
"theoretical_loss": 3.413312258783614,
"tokens_seen": 2124152832
},
{
"epoch": 0.64,
"learning_rate": 0.00018004012036108326,
"loss": 2.6081,
"theoretical_loss": 3.413303582048976,
"tokens_seen": 2124218368
},
{
"epoch": 0.64,
"learning_rate": 0.00018003009027081244,
"loss": 2.7136,
"theoretical_loss": 3.4132949056569784,
"tokens_seen": 2124283904
},
{
"epoch": 0.64,
"learning_rate": 0.00018002006018054162,
"loss": 2.5506,
"theoretical_loss": 3.4132862296075968,
"tokens_seen": 2124349440
},
{
"epoch": 0.64,
"learning_rate": 0.0001800100300902708,
"loss": 2.4326,
"theoretical_loss": 3.413277553900808,
"tokens_seen": 2124414976
},
{
"epoch": 0.64,
"learning_rate": 0.00017999999999999998,
"loss": 2.5923,
"theoretical_loss": 3.413268878536586,
"tokens_seen": 2124480512
},
{
"epoch": 0.64,
"learning_rate": 0.0001799899699097292,
"loss": 2.2943,
"theoretical_loss": 3.413260203514909,
"tokens_seen": 2124546048
},
{
"epoch": 0.64,
"learning_rate": 0.00017997993981945837,
"loss": 2.6065,
"theoretical_loss": 3.413251528835751,
"tokens_seen": 2124611584
},
{
"epoch": 0.64,
"learning_rate": 0.00017996990972918755,
"loss": 2.651,
"theoretical_loss": 3.4132428544990896,
"tokens_seen": 2124677120
},
{
"epoch": 0.64,
"learning_rate": 0.00017995987963891674,
"loss": 2.3564,
"theoretical_loss": 3.4132341805048996,
"tokens_seen": 2124742656
},
{
"epoch": 0.64,
"learning_rate": 0.00017994984954864594,
"loss": 2.4764,
"theoretical_loss": 3.4132255068531574,
"tokens_seen": 2124808192
},
{
"epoch": 0.64,
"learning_rate": 0.00017993981945837513,
"loss": 2.8094,
"theoretical_loss": 3.413216833543838,
"tokens_seen": 2124873728
},
{
"epoch": 0.64,
"learning_rate": 0.0001799297893681043,
"loss": 2.5877,
"theoretical_loss": 3.413208160576919,
"tokens_seen": 2124939264
},
{
"epoch": 0.64,
"objective/train/docs_used": 1202675,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8241257667541504,
"objective/train/theoretical_loss": 3.413199487952374,
"objective/train/tokens_used": 2145464800,
"theoretical_loss": 3.413199487952374,
"tokens_seen": 2125004800
},
{
"epoch": 0.64,
"learning_rate": 0.0001799197592778335,
"loss": 2.6061,
"theoretical_loss": 3.413199487952374,
"tokens_seen": 2125004800
},
{
"epoch": 0.64,
"learning_rate": 0.00017990972918756267,
"loss": 2.3405,
"theoretical_loss": 3.4131908156701813,
"tokens_seen": 2125070336
},
{
"epoch": 0.64,
"learning_rate": 0.00017989969909729188,
"loss": 2.5908,
"theoretical_loss": 3.4131821437303156,
"tokens_seen": 2125135872
},
{
"epoch": 0.64,
"learning_rate": 0.00017988966900702106,
"loss": 2.3997,
"theoretical_loss": 3.4131734721327533,
"tokens_seen": 2125201408
},
{
"epoch": 0.64,
"learning_rate": 0.00017987963891675024,
"loss": 2.2152,
"theoretical_loss": 3.41316480087747,
"tokens_seen": 2125266944
},
{
"epoch": 0.64,
"learning_rate": 0.00017986960882647942,
"loss": 2.6141,
"theoretical_loss": 3.4131561299644417,
"tokens_seen": 2125332480
},
{
"epoch": 0.64,
"learning_rate": 0.00017985957873620863,
"loss": 2.4961,
"theoretical_loss": 3.413147459393644,
"tokens_seen": 2125398016
},
{
"epoch": 0.64,
"learning_rate": 0.0001798495486459378,
"loss": 2.3617,
"theoretical_loss": 3.4131387891650533,
"tokens_seen": 2125463552
},
{
"epoch": 0.64,
"learning_rate": 0.000179839518555667,
"loss": 2.7575,
"theoretical_loss": 3.4131301192786454,
"tokens_seen": 2125529088
},
{
"epoch": 0.64,
"learning_rate": 0.00017982948846539617,
"loss": 2.5997,
"theoretical_loss": 3.4131214497343967,
"tokens_seen": 2125594624
},
{
"epoch": 0.64,
"learning_rate": 0.00017981945837512538,
"loss": 2.5348,
"theoretical_loss": 3.4131127805322823,
"tokens_seen": 2125660160
},
{
"epoch": 0.64,
"learning_rate": 0.00017980942828485456,
"loss": 2.3582,
"theoretical_loss": 3.4131041116722787,
"tokens_seen": 2125725696
},
{
"epoch": 0.64,
"learning_rate": 0.00017979939819458377,
"loss": 2.5176,
"theoretical_loss": 3.413095443154362,
"tokens_seen": 2125791232
},
{
"epoch": 0.64,
"learning_rate": 0.00017978936810431295,
"loss": 2.5627,
"theoretical_loss": 3.4130867749785074,
"tokens_seen": 2125856768
},
{
"epoch": 0.64,
"learning_rate": 0.00017977933801404213,
"loss": 2.4276,
"theoretical_loss": 3.4130781071446914,
"tokens_seen": 2125922304
},
{
"epoch": 0.64,
"learning_rate": 0.00017976930792377134,
"loss": 2.3166,
"theoretical_loss": 3.4130694396528902,
"tokens_seen": 2125987840
},
{
"epoch": 0.64,
"learning_rate": 0.00017975927783350052,
"loss": 2.5536,
"theoretical_loss": 3.413060772503079,
"tokens_seen": 2126053376
},
{
"epoch": 0.64,
"learning_rate": 0.0001797492477432297,
"loss": 2.7782,
"theoretical_loss": 3.4130521056952348,
"tokens_seen": 2126118912
},
{
"epoch": 0.64,
"learning_rate": 0.00017973921765295888,
"loss": 2.5253,
"theoretical_loss": 3.413043439229333,
"tokens_seen": 2126184448
},
{
"epoch": 0.64,
"learning_rate": 0.0001797291875626881,
"loss": 2.4431,
"theoretical_loss": 3.413034773105349,
"tokens_seen": 2126249984
},
{
"epoch": 0.64,
"learning_rate": 0.00017971915747241727,
"loss": 2.7045,
"theoretical_loss": 3.41302610732326,
"tokens_seen": 2126315520
},
{
"epoch": 0.64,
"learning_rate": 0.00017970912738214645,
"loss": 2.4159,
"theoretical_loss": 3.4130174418830403,
"tokens_seen": 2126381056
},
{
"epoch": 0.64,
"learning_rate": 0.00017969909729187563,
"loss": 2.7764,
"theoretical_loss": 3.4130087767846677,
"tokens_seen": 2126446592
},
{
"epoch": 0.64,
"learning_rate": 0.00017968906720160482,
"loss": 2.717,
"theoretical_loss": 3.4130001120281173,
"tokens_seen": 2126512128
},
{
"epoch": 0.64,
"learning_rate": 0.00017967903711133402,
"loss": 2.6629,
"theoretical_loss": 3.412991447613365,
"tokens_seen": 2126577664
},
{
"epoch": 0.64,
"objective/train/docs_used": 1203810,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.116706371307373,
"objective/train/theoretical_loss": 3.412982783540387,
"objective/train/tokens_used": 2147103200,
"theoretical_loss": 3.412982783540387,
"tokens_seen": 2126643200
},
{
"epoch": 0.64,
"learning_rate": 0.0001796690070210632,
"loss": 2.5746,
"theoretical_loss": 3.412982783540387,
"tokens_seen": 2126643200
},
{
"epoch": 0.64,
"learning_rate": 0.00017965897693079239,
"loss": 2.6852,
"theoretical_loss": 3.412974119809159,
"tokens_seen": 2126708736
},
{
"epoch": 0.64,
"learning_rate": 0.00017964894684052157,
"loss": 2.6675,
"theoretical_loss": 3.4129654564196574,
"tokens_seen": 2126774272
},
{
"epoch": 0.64,
"learning_rate": 0.00017963891675025078,
"loss": 2.53,
"theoretical_loss": 3.4129567933718583,
"tokens_seen": 2126839808
},
{
"epoch": 0.64,
"learning_rate": 0.00017962888665997996,
"loss": 2.4999,
"theoretical_loss": 3.412948130665737,
"tokens_seen": 2126905344
},
{
"epoch": 0.64,
"learning_rate": 0.00017961885656970914,
"loss": 2.4847,
"theoretical_loss": 3.4129394683012704,
"tokens_seen": 2126970880
},
{
"epoch": 0.64,
"learning_rate": 0.00017960882647943832,
"loss": 2.5991,
"theoretical_loss": 3.4129308062784336,
"tokens_seen": 2127036416
},
{
"epoch": 0.64,
"learning_rate": 0.0001795987963891675,
"loss": 2.5535,
"theoretical_loss": 3.4129221445972027,
"tokens_seen": 2127101952
},
{
"epoch": 0.64,
"learning_rate": 0.0001795887662988967,
"loss": 2.4038,
"theoretical_loss": 3.4129134832575545,
"tokens_seen": 2127167488
},
{
"epoch": 0.64,
"learning_rate": 0.0001795787362086259,
"loss": 2.4632,
"theoretical_loss": 3.4129048222594642,
"tokens_seen": 2127233024
},
{
"epoch": 0.64,
"learning_rate": 0.00017956870611835507,
"loss": 2.6595,
"theoretical_loss": 3.4128961616029083,
"tokens_seen": 2127298560
},
{
"epoch": 0.64,
"learning_rate": 0.00017955867602808425,
"loss": 2.245,
"theoretical_loss": 3.4128875012878623,
"tokens_seen": 2127364096
},
{
"epoch": 0.64,
"learning_rate": 0.00017954864593781346,
"loss": 2.6976,
"theoretical_loss": 3.4128788413143027,
"tokens_seen": 2127429632
},
{
"epoch": 0.64,
"learning_rate": 0.00017953861584754264,
"loss": 2.4293,
"theoretical_loss": 3.412870181682206,
"tokens_seen": 2127495168
},
{
"epoch": 0.64,
"learning_rate": 0.00017952858575727182,
"loss": 2.3657,
"theoretical_loss": 3.412861522391547,
"tokens_seen": 2127560704
},
{
"epoch": 0.64,
"learning_rate": 0.000179518555667001,
"loss": 2.4916,
"theoretical_loss": 3.412852863442302,
"tokens_seen": 2127626240
},
{
"epoch": 0.64,
"learning_rate": 0.00017950852557673018,
"loss": 2.5549,
"theoretical_loss": 3.412844204834448,
"tokens_seen": 2127691776
},
{
"epoch": 0.64,
"learning_rate": 0.0001794984954864594,
"loss": 2.5367,
"theoretical_loss": 3.4128355465679596,
"tokens_seen": 2127757312
},
{
"epoch": 0.64,
"learning_rate": 0.00017948846539618857,
"loss": 2.8451,
"theoretical_loss": 3.412826888642814,
"tokens_seen": 2127822848
},
{
"epoch": 0.64,
"learning_rate": 0.00017947843530591775,
"loss": 2.5745,
"theoretical_loss": 3.4128182310589867,
"tokens_seen": 2127888384
},
{
"epoch": 0.64,
"learning_rate": 0.00017946840521564694,
"loss": 2.4317,
"theoretical_loss": 3.4128095738164537,
"tokens_seen": 2127953920
},
{
"epoch": 0.64,
"learning_rate": 0.00017945837512537614,
"loss": 2.6437,
"theoretical_loss": 3.4128009169151916,
"tokens_seen": 2128019456
},
{
"epoch": 0.64,
"learning_rate": 0.00017944834503510533,
"loss": 2.5642,
"theoretical_loss": 3.4127922603551752,
"tokens_seen": 2128084992
},
{
"epoch": 0.64,
"learning_rate": 0.0001794383149448345,
"loss": 2.4739,
"theoretical_loss": 3.412783604136382,
"tokens_seen": 2128150528
},
{
"epoch": 0.64,
"learning_rate": 0.0001794282848545637,
"loss": 2.5277,
"theoretical_loss": 3.4127749482587872,
"tokens_seen": 2128216064
},
{
"epoch": 0.64,
"objective/train/docs_used": 1204521,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.630894660949707,
"objective/train/theoretical_loss": 3.412766292722367,
"objective/train/tokens_used": 2148741600,
"theoretical_loss": 3.412766292722367,
"tokens_seen": 2128281600
},
{
"epoch": 0.64,
"learning_rate": 0.00017941825476429287,
"loss": 2.6489,
"theoretical_loss": 3.412766292722367,
"tokens_seen": 2128281600
},
{
"epoch": 0.64,
"learning_rate": 0.00017940822467402208,
"loss": 2.8245,
"theoretical_loss": 3.412757637527098,
"tokens_seen": 2128347136
},
{
"epoch": 0.64,
"learning_rate": 0.00017939819458375126,
"loss": 2.6067,
"theoretical_loss": 3.412748982672955,
"tokens_seen": 2128412672
},
{
"epoch": 0.64,
"learning_rate": 0.00017938816449348044,
"loss": 2.6097,
"theoretical_loss": 3.412740328159915,
"tokens_seen": 2128478208
},
{
"epoch": 0.65,
"learning_rate": 0.00017937813440320962,
"loss": 2.4454,
"theoretical_loss": 3.4127316739879543,
"tokens_seen": 2128543744
},
{
"epoch": 0.65,
"learning_rate": 0.00017936810431293883,
"loss": 2.4479,
"theoretical_loss": 3.4127230201570478,
"tokens_seen": 2128609280
},
{
"epoch": 0.65,
"learning_rate": 0.000179358074222668,
"loss": 2.6835,
"theoretical_loss": 3.4127143666671724,
"tokens_seen": 2128674816
},
{
"epoch": 0.65,
"learning_rate": 0.0001793480441323972,
"loss": 2.6535,
"theoretical_loss": 3.4127057135183043,
"tokens_seen": 2128740352
},
{
"epoch": 0.65,
"learning_rate": 0.00017933801404212637,
"loss": 2.4877,
"theoretical_loss": 3.412697060710419,
"tokens_seen": 2128805888
},
{
"epoch": 0.65,
"learning_rate": 0.00017932798395185558,
"loss": 2.4205,
"theoretical_loss": 3.4126884082434934,
"tokens_seen": 2128871424
},
{
"epoch": 0.65,
"learning_rate": 0.00017931795386158476,
"loss": 2.5421,
"theoretical_loss": 3.412679756117502,
"tokens_seen": 2128936960
},
{
"epoch": 0.65,
"learning_rate": 0.00017930792377131394,
"loss": 2.5116,
"theoretical_loss": 3.4126711043324227,
"tokens_seen": 2129002496
},
{
"epoch": 0.65,
"learning_rate": 0.00017929789368104312,
"loss": 2.3174,
"theoretical_loss": 3.4126624528882306,
"tokens_seen": 2129068032
},
{
"epoch": 0.65,
"learning_rate": 0.0001792878635907723,
"loss": 2.4599,
"theoretical_loss": 3.4126538017849017,
"tokens_seen": 2129133568
},
{
"epoch": 0.65,
"learning_rate": 0.0001792778335005015,
"loss": 2.5874,
"theoretical_loss": 3.4126451510224127,
"tokens_seen": 2129199104
},
{
"epoch": 0.65,
"learning_rate": 0.0001792678034102307,
"loss": 2.3984,
"theoretical_loss": 3.4126365006007386,
"tokens_seen": 2129264640
},
{
"epoch": 0.65,
"learning_rate": 0.00017925777331995988,
"loss": 2.4866,
"theoretical_loss": 3.412627850519857,
"tokens_seen": 2129330176
},
{
"epoch": 0.65,
"learning_rate": 0.00017924774322968906,
"loss": 2.5432,
"theoretical_loss": 3.412619200779743,
"tokens_seen": 2129395712
},
{
"epoch": 0.65,
"learning_rate": 0.00017923771313941826,
"loss": 2.467,
"theoretical_loss": 3.4126105513803724,
"tokens_seen": 2129461248
},
{
"epoch": 0.65,
"learning_rate": 0.00017922768304914745,
"loss": 2.6885,
"theoretical_loss": 3.4126019023217222,
"tokens_seen": 2129526784
},
{
"epoch": 0.65,
"learning_rate": 0.00017921765295887663,
"loss": 2.5411,
"theoretical_loss": 3.4125932536037675,
"tokens_seen": 2129592320
},
{
"epoch": 0.65,
"learning_rate": 0.0001792076228686058,
"loss": 2.6385,
"theoretical_loss": 3.4125846052264857,
"tokens_seen": 2129657856
},
{
"epoch": 0.65,
"learning_rate": 0.000179197592778335,
"loss": 2.5643,
"theoretical_loss": 3.4125759571898513,
"tokens_seen": 2129723392
},
{
"epoch": 0.65,
"learning_rate": 0.0001791875626880642,
"loss": 2.654,
"theoretical_loss": 3.412567309493842,
"tokens_seen": 2129788928
},
{
"epoch": 0.65,
"learning_rate": 0.00017917753259779338,
"loss": 2.5953,
"theoretical_loss": 3.4125586621384327,
"tokens_seen": 2129854464
},
{
"debugging/Self-BLEU-5": 0.19806458433645993,
"debugging/distinct-1-grams": 0.7162394027462491,
"debugging/distinct-2-grams": 0.937001466202286,
"debugging/entropy-1-grams": 4.7826233728189305,
"debugging/entropy-2-grams": 5.409961017479109,
"debugging/length": 511.6666666666667,
"debugging/num_segments": 3,
"epoch": 0.65,
"objective/train/docs_used": 1206012,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6803126335144043,
"objective/train/theoretical_loss": 3.4125500151236,
"objective/train/tokens_used": 2150380000,
"theoretical_loss": 3.4125500151236,
"tokens_seen": 2129920000
},
{
"epoch": 0.65,
"learning_rate": 0.00017916750250752256,
"loss": 2.4829,
"theoretical_loss": 3.4125500151236,
"tokens_seen": 2129920000
},
{
"epoch": 0.65,
"learning_rate": 0.00017915747241725174,
"loss": 2.8185,
"theoretical_loss": 3.41254136844932,
"tokens_seen": 2129985536
},
{
"epoch": 0.65,
"learning_rate": 0.00017914744232698095,
"loss": 2.6452,
"theoretical_loss": 3.412532722115569,
"tokens_seen": 2130051072
},
{
"epoch": 0.65,
"learning_rate": 0.00017913741223671013,
"loss": 2.4374,
"theoretical_loss": 3.412524076122322,
"tokens_seen": 2130116608
},
{
"epoch": 0.65,
"learning_rate": 0.0001791273821464393,
"loss": 2.888,
"theoretical_loss": 3.412515430469557,
"tokens_seen": 2130182144
},
{
"epoch": 0.65,
"learning_rate": 0.0001791173520561685,
"loss": 2.636,
"theoretical_loss": 3.412506785157248,
"tokens_seen": 2130247680
},
{
"epoch": 0.65,
"learning_rate": 0.00017910732196589767,
"loss": 2.4828,
"theoretical_loss": 3.4124981401853733,
"tokens_seen": 2130313216
},
{
"epoch": 0.65,
"learning_rate": 0.00017909729187562688,
"loss": 2.4779,
"theoretical_loss": 3.412489495553908,
"tokens_seen": 2130378752
},
{
"epoch": 0.65,
"learning_rate": 0.00017908726178535606,
"loss": 2.4902,
"theoretical_loss": 3.4124808512628277,
"tokens_seen": 2130444288
},
{
"epoch": 0.65,
"learning_rate": 0.00017907723169508524,
"loss": 2.5402,
"theoretical_loss": 3.412472207312109,
"tokens_seen": 2130509824
},
{
"epoch": 0.65,
"learning_rate": 0.00017906720160481442,
"loss": 2.4411,
"theoretical_loss": 3.4124635637017278,
"tokens_seen": 2130575360
},
{
"epoch": 0.65,
"learning_rate": 0.00017905717151454363,
"loss": 2.3624,
"theoretical_loss": 3.4124549204316605,
"tokens_seen": 2130640896
},
{
"epoch": 0.65,
"learning_rate": 0.00017904714142427284,
"loss": 2.466,
"theoretical_loss": 3.412446277501884,
"tokens_seen": 2130706432
},
{
"epoch": 0.65,
"learning_rate": 0.00017903711133400202,
"loss": 2.5972,
"theoretical_loss": 3.4124376349123726,
"tokens_seen": 2130771968
},
{
"epoch": 0.65,
"learning_rate": 0.0001790270812437312,
"loss": 2.7302,
"theoretical_loss": 3.412428992663104,
"tokens_seen": 2130837504
},
{
"epoch": 0.65,
"learning_rate": 0.00017901705115346038,
"loss": 2.7343,
"theoretical_loss": 3.4124203507540534,
"tokens_seen": 2130903040
},
{
"epoch": 0.65,
"learning_rate": 0.0001790070210631896,
"loss": 2.5262,
"theoretical_loss": 3.4124117091851978,
"tokens_seen": 2130968576
},
{
"epoch": 0.65,
"learning_rate": 0.00017899699097291877,
"loss": 2.7524,
"theoretical_loss": 3.4124030679565127,
"tokens_seen": 2131034112
},
{
"epoch": 0.65,
"learning_rate": 0.00017898696088264796,
"loss": 2.6559,
"theoretical_loss": 3.4123944270679742,
"tokens_seen": 2131099648
},
{
"epoch": 0.65,
"learning_rate": 0.00017897693079237714,
"loss": 2.3846,
"theoretical_loss": 3.4123857865195593,
"tokens_seen": 2131165184
},
{
"epoch": 0.65,
"learning_rate": 0.00017896690070210634,
"loss": 2.5302,
"theoretical_loss": 3.4123771463112433,
"tokens_seen": 2131230720
},
{
"epoch": 0.65,
"learning_rate": 0.00017895687061183553,
"loss": 2.5293,
"theoretical_loss": 3.412368506443002,
"tokens_seen": 2131296256
},
{
"epoch": 0.65,
"learning_rate": 0.0001789468405215647,
"loss": 2.5804,
"theoretical_loss": 3.412359866914813,
"tokens_seen": 2131361792
},
{
"epoch": 0.65,
"learning_rate": 0.0001789368104312939,
"loss": 2.714,
"theoretical_loss": 3.4123512277266514,
"tokens_seen": 2131427328
},
{
"epoch": 0.65,
"learning_rate": 0.00017892678034102307,
"loss": 2.6942,
"theoretical_loss": 3.4123425888784933,
"tokens_seen": 2131492864
},
{
"epoch": 0.65,
"objective/train/docs_used": 1206751,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3085784912109375,
"objective/train/theoretical_loss": 3.4123339503703156,
"objective/train/tokens_used": 2152018400,
"theoretical_loss": 3.4123339503703156,
"tokens_seen": 2131558400
},
{
"epoch": 0.65,
"learning_rate": 0.00017891675025075228,
"loss": 2.6906,
"theoretical_loss": 3.4123339503703156,
"tokens_seen": 2131558400
},
{
"epoch": 0.65,
"learning_rate": 0.00017890672016048146,
"loss": 2.3556,
"theoretical_loss": 3.4123253122020936,
"tokens_seen": 2131623936
},
{
"epoch": 0.65,
"learning_rate": 0.00017889669007021064,
"loss": 2.4672,
"theoretical_loss": 3.4123166743738045,
"tokens_seen": 2131689472
},
{
"epoch": 0.65,
"learning_rate": 0.00017888665997993982,
"loss": 2.6088,
"theoretical_loss": 3.412308036885423,
"tokens_seen": 2131755008
},
{
"epoch": 0.65,
"learning_rate": 0.00017887662988966903,
"loss": 2.3365,
"theoretical_loss": 3.412299399736927,
"tokens_seen": 2131820544
},
{
"epoch": 0.65,
"learning_rate": 0.0001788665997993982,
"loss": 2.5822,
"theoretical_loss": 3.4122907629282913,
"tokens_seen": 2131886080
},
{
"epoch": 0.65,
"learning_rate": 0.0001788565697091274,
"loss": 2.6369,
"theoretical_loss": 3.412282126459493,
"tokens_seen": 2131951616
},
{
"epoch": 0.65,
"learning_rate": 0.00017884653961885657,
"loss": 2.8055,
"theoretical_loss": 3.412273490330507,
"tokens_seen": 2132017152
},
{
"epoch": 0.65,
"learning_rate": 0.00017883650952858578,
"loss": 2.7091,
"theoretical_loss": 3.412264854541311,
"tokens_seen": 2132082688
},
{
"epoch": 0.65,
"learning_rate": 0.00017882647943831496,
"loss": 2.5808,
"theoretical_loss": 3.412256219091881,
"tokens_seen": 2132148224
},
{
"epoch": 0.65,
"learning_rate": 0.00017881644934804414,
"loss": 2.6134,
"theoretical_loss": 3.412247583982192,
"tokens_seen": 2132213760
},
{
"epoch": 0.65,
"learning_rate": 0.00017880641925777332,
"loss": 2.5391,
"theoretical_loss": 3.412238949212221,
"tokens_seen": 2132279296
},
{
"epoch": 0.65,
"learning_rate": 0.0001787963891675025,
"loss": 2.5653,
"theoretical_loss": 3.4122303147819446,
"tokens_seen": 2132344832
},
{
"epoch": 0.65,
"learning_rate": 0.0001787863590772317,
"loss": 2.6997,
"theoretical_loss": 3.412221680691338,
"tokens_seen": 2132410368
},
{
"epoch": 0.65,
"learning_rate": 0.0001787763289869609,
"loss": 2.5477,
"theoretical_loss": 3.412213046940378,
"tokens_seen": 2132475904
},
{
"epoch": 0.65,
"learning_rate": 0.00017876629889669008,
"loss": 2.5772,
"theoretical_loss": 3.4122044135290412,
"tokens_seen": 2132541440
},
{
"epoch": 0.65,
"learning_rate": 0.00017875626880641926,
"loss": 2.6938,
"theoretical_loss": 3.412195780457303,
"tokens_seen": 2132606976
},
{
"epoch": 0.65,
"learning_rate": 0.00017874623871614846,
"loss": 2.7279,
"theoretical_loss": 3.41218714772514,
"tokens_seen": 2132672512
},
{
"epoch": 0.65,
"learning_rate": 0.00017873620862587765,
"loss": 2.6179,
"theoretical_loss": 3.412178515332528,
"tokens_seen": 2132738048
},
{
"epoch": 0.65,
"learning_rate": 0.00017872617853560683,
"loss": 2.4455,
"theoretical_loss": 3.412169883279444,
"tokens_seen": 2132803584
},
{
"epoch": 0.65,
"learning_rate": 0.000178716148445336,
"loss": 2.4701,
"theoretical_loss": 3.412161251565864,
"tokens_seen": 2132869120
},
{
"epoch": 0.65,
"learning_rate": 0.0001787061183550652,
"loss": 2.3517,
"theoretical_loss": 3.4121526201917636,
"tokens_seen": 2132934656
},
{
"epoch": 0.65,
"learning_rate": 0.0001786960882647944,
"loss": 2.5117,
"theoretical_loss": 3.4121439891571193,
"tokens_seen": 2133000192
},
{
"epoch": 0.65,
"learning_rate": 0.00017868605817452358,
"loss": 2.4956,
"theoretical_loss": 3.4121353584619074,
"tokens_seen": 2133065728
},
{
"epoch": 0.65,
"learning_rate": 0.00017867602808425276,
"loss": 2.6445,
"theoretical_loss": 3.412126728106104,
"tokens_seen": 2133131264
},
{
"epoch": 0.65,
"objective/train/docs_used": 1207326,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3070077896118164,
"objective/train/theoretical_loss": 3.412118098089686,
"objective/train/tokens_used": 2153656800,
"theoretical_loss": 3.412118098089686,
"tokens_seen": 2133196800
},
{
"epoch": 0.65,
"learning_rate": 0.00017866599799398194,
"loss": 2.4031,
"theoretical_loss": 3.412118098089686,
"tokens_seen": 2133196800
},
{
"epoch": 0.65,
"learning_rate": 0.00017865596790371115,
"loss": 2.7279,
"theoretical_loss": 3.4121094684126287,
"tokens_seen": 2133262336
},
{
"epoch": 0.65,
"learning_rate": 0.00017864593781344033,
"loss": 2.48,
"theoretical_loss": 3.412100839074909,
"tokens_seen": 2133327872
},
{
"epoch": 0.65,
"learning_rate": 0.0001786359077231695,
"loss": 2.6017,
"theoretical_loss": 3.412092210076503,
"tokens_seen": 2133393408
},
{
"epoch": 0.65,
"learning_rate": 0.0001786258776328987,
"loss": 2.545,
"theoretical_loss": 3.412083581417386,
"tokens_seen": 2133458944
},
{
"epoch": 0.65,
"learning_rate": 0.00017861584754262787,
"loss": 2.4055,
"theoretical_loss": 3.412074953097536,
"tokens_seen": 2133524480
},
{
"epoch": 0.65,
"learning_rate": 0.00017860581745235708,
"loss": 2.6714,
"theoretical_loss": 3.412066325116928,
"tokens_seen": 2133590016
},
{
"epoch": 0.65,
"learning_rate": 0.00017859578736208626,
"loss": 2.6237,
"theoretical_loss": 3.4120576974755386,
"tokens_seen": 2133655552
},
{
"epoch": 0.65,
"learning_rate": 0.00017858575727181544,
"loss": 2.6797,
"theoretical_loss": 3.4120490701733432,
"tokens_seen": 2133721088
},
{
"epoch": 0.65,
"learning_rate": 0.00017857572718154462,
"loss": 2.7279,
"theoretical_loss": 3.4120404432103193,
"tokens_seen": 2133786624
},
{
"epoch": 0.65,
"learning_rate": 0.00017856569709127383,
"loss": 2.7392,
"theoretical_loss": 3.4120318165864427,
"tokens_seen": 2133852160
},
{
"epoch": 0.65,
"learning_rate": 0.00017855566700100301,
"loss": 2.1904,
"theoretical_loss": 3.41202319030169,
"tokens_seen": 2133917696
},
{
"epoch": 0.65,
"learning_rate": 0.0001785456369107322,
"loss": 2.7289,
"theoretical_loss": 3.4120145643560367,
"tokens_seen": 2133983232
},
{
"epoch": 0.65,
"learning_rate": 0.00017853560682046138,
"loss": 2.4837,
"theoretical_loss": 3.4120059387494592,
"tokens_seen": 2134048768
},
{
"epoch": 0.65,
"learning_rate": 0.00017852557673019056,
"loss": 2.5895,
"theoretical_loss": 3.411997313481934,
"tokens_seen": 2134114304
},
{
"epoch": 0.65,
"learning_rate": 0.00017851554663991977,
"loss": 2.6991,
"theoretical_loss": 3.411988688553438,
"tokens_seen": 2134179840
},
{
"epoch": 0.65,
"learning_rate": 0.00017850551654964895,
"loss": 2.9117,
"theoretical_loss": 3.411980063963946,
"tokens_seen": 2134245376
},
{
"epoch": 0.65,
"learning_rate": 0.00017849548645937813,
"loss": 2.7422,
"theoretical_loss": 3.4119714397134357,
"tokens_seen": 2134310912
},
{
"epoch": 0.65,
"learning_rate": 0.0001784854563691073,
"loss": 2.8395,
"theoretical_loss": 3.4119628158018824,
"tokens_seen": 2134376448
},
{
"epoch": 0.65,
"learning_rate": 0.00017847542627883652,
"loss": 2.8697,
"theoretical_loss": 3.411954192229263,
"tokens_seen": 2134441984
},
{
"epoch": 0.65,
"learning_rate": 0.0001784653961885657,
"loss": 2.5913,
"theoretical_loss": 3.411945568995553,
"tokens_seen": 2134507520
},
{
"epoch": 0.65,
"learning_rate": 0.00017845536609829488,
"loss": 2.6018,
"theoretical_loss": 3.411936946100729,
"tokens_seen": 2134573056
},
{
"epoch": 0.65,
"learning_rate": 0.00017844533600802406,
"loss": 2.5401,
"theoretical_loss": 3.4119283235447684,
"tokens_seen": 2134638592
},
{
"epoch": 0.65,
"learning_rate": 0.00017843530591775324,
"loss": 2.5467,
"theoretical_loss": 3.411919701327646,
"tokens_seen": 2134704128
},
{
"epoch": 0.65,
"learning_rate": 0.00017842527582748245,
"loss": 2.9455,
"theoretical_loss": 3.4119110794493386,
"tokens_seen": 2134769664
},
{
"epoch": 0.65,
"objective/train/docs_used": 1208575,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.554659843444824,
"objective/train/theoretical_loss": 3.4119024579098225,
"objective/train/tokens_used": 2155295200,
"theoretical_loss": 3.4119024579098225,
"tokens_seen": 2134835200
},
{
"epoch": 0.65,
"learning_rate": 0.00017841524573721163,
"loss": 2.8367,
"theoretical_loss": 3.4119024579098225,
"tokens_seen": 2134835200
},
{
"epoch": 0.65,
"learning_rate": 0.0001784052156469408,
"loss": 2.564,
"theoretical_loss": 3.4118938367090736,
"tokens_seen": 2134900736
},
{
"epoch": 0.65,
"learning_rate": 0.00017839518555667,
"loss": 2.6896,
"theoretical_loss": 3.411885215847069,
"tokens_seen": 2134966272
},
{
"epoch": 0.65,
"learning_rate": 0.0001783851554663992,
"loss": 2.5051,
"theoretical_loss": 3.4118765953237844,
"tokens_seen": 2135031808
},
{
"epoch": 0.65,
"learning_rate": 0.00017837512537612838,
"loss": 2.4437,
"theoretical_loss": 3.4118679751391965,
"tokens_seen": 2135097344
},
{
"epoch": 0.65,
"learning_rate": 0.00017836509528585756,
"loss": 2.6674,
"theoretical_loss": 3.411859355293281,
"tokens_seen": 2135162880
},
{
"epoch": 0.65,
"learning_rate": 0.00017835506519558675,
"loss": 2.4315,
"theoretical_loss": 3.4118507357860146,
"tokens_seen": 2135228416
},
{
"epoch": 0.65,
"learning_rate": 0.00017834503510531593,
"loss": 2.6593,
"theoretical_loss": 3.4118421166173736,
"tokens_seen": 2135293952
},
{
"epoch": 0.65,
"learning_rate": 0.00017833500501504513,
"loss": 2.8114,
"theoretical_loss": 3.4118334977873346,
"tokens_seen": 2135359488
},
{
"epoch": 0.65,
"learning_rate": 0.00017832497492477432,
"loss": 2.8185,
"theoretical_loss": 3.4118248792958736,
"tokens_seen": 2135425024
},
{
"epoch": 0.65,
"learning_rate": 0.0001783149448345035,
"loss": 2.6455,
"theoretical_loss": 3.4118162611429663,
"tokens_seen": 2135490560
},
{
"epoch": 0.65,
"learning_rate": 0.00017830491474423268,
"loss": 2.7522,
"theoretical_loss": 3.41180764332859,
"tokens_seen": 2135556096
},
{
"epoch": 0.65,
"learning_rate": 0.0001782948846539619,
"loss": 2.7362,
"theoretical_loss": 3.41179902585272,
"tokens_seen": 2135621632
},
{
"epoch": 0.65,
"learning_rate": 0.0001782848545636911,
"loss": 2.3461,
"theoretical_loss": 3.411790408715334,
"tokens_seen": 2135687168
},
{
"epoch": 0.65,
"learning_rate": 0.00017827482447342028,
"loss": 2.5139,
"theoretical_loss": 3.411781791916407,
"tokens_seen": 2135752704
},
{
"epoch": 0.65,
"learning_rate": 0.00017826479438314946,
"loss": 2.6445,
"theoretical_loss": 3.411773175455916,
"tokens_seen": 2135818240
},
{
"epoch": 0.65,
"learning_rate": 0.00017825476429287866,
"loss": 2.5532,
"theoretical_loss": 3.4117645593338377,
"tokens_seen": 2135883776
},
{
"epoch": 0.65,
"learning_rate": 0.00017824473420260785,
"loss": 2.7327,
"theoretical_loss": 3.411755943550147,
"tokens_seen": 2135949312
},
{
"epoch": 0.65,
"learning_rate": 0.00017823470411233703,
"loss": 2.6386,
"theoretical_loss": 3.4117473281048216,
"tokens_seen": 2136014848
},
{
"epoch": 0.65,
"learning_rate": 0.0001782246740220662,
"loss": 2.639,
"theoretical_loss": 3.4117387129978374,
"tokens_seen": 2136080384
},
{
"epoch": 0.65,
"learning_rate": 0.0001782146439317954,
"loss": 2.4235,
"theoretical_loss": 3.4117300982291705,
"tokens_seen": 2136145920
},
{
"epoch": 0.65,
"learning_rate": 0.0001782046138415246,
"loss": 2.7047,
"theoretical_loss": 3.4117214837987975,
"tokens_seen": 2136211456
},
{
"epoch": 0.65,
"learning_rate": 0.00017819458375125378,
"loss": 2.6141,
"theoretical_loss": 3.4117128697066947,
"tokens_seen": 2136276992
},
{
"epoch": 0.65,
"learning_rate": 0.00017818455366098296,
"loss": 2.4044,
"theoretical_loss": 3.4117042559528388,
"tokens_seen": 2136342528
},
{
"epoch": 0.65,
"learning_rate": 0.00017817452357071214,
"loss": 2.6703,
"theoretical_loss": 3.411695642537205,
"tokens_seen": 2136408064
},
{
"epoch": 0.65,
"objective/train/docs_used": 1209281,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6299500465393066,
"objective/train/theoretical_loss": 3.411687029459771,
"objective/train/tokens_used": 2156933600,
"theoretical_loss": 3.411687029459771,
"tokens_seen": 2136473600
},
{
"epoch": 0.65,
"learning_rate": 0.00017816449348044135,
"loss": 2.536,
"theoretical_loss": 3.411687029459771,
"tokens_seen": 2136473600
},
{
"epoch": 0.65,
"learning_rate": 0.00017815446339017053,
"loss": 2.7019,
"theoretical_loss": 3.4116784167205125,
"tokens_seen": 2136539136
},
{
"epoch": 0.65,
"learning_rate": 0.0001781444332998997,
"loss": 2.6802,
"theoretical_loss": 3.4116698043194056,
"tokens_seen": 2136604672
},
{
"epoch": 0.65,
"learning_rate": 0.0001781344032096289,
"loss": 2.4199,
"theoretical_loss": 3.411661192256427,
"tokens_seen": 2136670208
},
{
"epoch": 0.65,
"learning_rate": 0.00017812437311935807,
"loss": 2.4708,
"theoretical_loss": 3.4116525805315527,
"tokens_seen": 2136735744
},
{
"epoch": 0.65,
"learning_rate": 0.00017811434302908728,
"loss": 2.5925,
"theoretical_loss": 3.4116439691447598,
"tokens_seen": 2136801280
},
{
"epoch": 0.65,
"learning_rate": 0.00017810431293881646,
"loss": 2.6081,
"theoretical_loss": 3.4116353580960244,
"tokens_seen": 2136866816
},
{
"epoch": 0.65,
"learning_rate": 0.00017809428284854564,
"loss": 2.4894,
"theoretical_loss": 3.411626747385322,
"tokens_seen": 2136932352
},
{
"epoch": 0.65,
"learning_rate": 0.00017808425275827483,
"loss": 2.4098,
"theoretical_loss": 3.4116181370126304,
"tokens_seen": 2136997888
},
{
"epoch": 0.65,
"learning_rate": 0.00017807422266800403,
"loss": 2.4084,
"theoretical_loss": 3.4116095269779247,
"tokens_seen": 2137063424
},
{
"epoch": 0.65,
"learning_rate": 0.00017806419257773321,
"loss": 2.4259,
"theoretical_loss": 3.4116009172811816,
"tokens_seen": 2137128960
},
{
"epoch": 0.65,
"learning_rate": 0.0001780541624874624,
"loss": 2.6384,
"theoretical_loss": 3.411592307922378,
"tokens_seen": 2137194496
},
{
"epoch": 0.65,
"learning_rate": 0.00017804413239719158,
"loss": 2.3443,
"theoretical_loss": 3.41158369890149,
"tokens_seen": 2137260032
},
{
"epoch": 0.65,
"learning_rate": 0.00017803410230692076,
"loss": 2.5242,
"theoretical_loss": 3.411575090218494,
"tokens_seen": 2137325568
},
{
"epoch": 0.65,
"learning_rate": 0.00017802407221664997,
"loss": 2.4248,
"theoretical_loss": 3.4115664818733658,
"tokens_seen": 2137391104
},
{
"epoch": 0.65,
"learning_rate": 0.00017801404212637915,
"loss": 2.7587,
"theoretical_loss": 3.411557873866083,
"tokens_seen": 2137456640
},
{
"epoch": 0.65,
"learning_rate": 0.00017800401203610833,
"loss": 2.4212,
"theoretical_loss": 3.4115492661966202,
"tokens_seen": 2137522176
},
{
"epoch": 0.65,
"learning_rate": 0.0001779939819458375,
"loss": 2.5798,
"theoretical_loss": 3.411540658864955,
"tokens_seen": 2137587712
},
{
"epoch": 0.65,
"learning_rate": 0.00017798395185556672,
"loss": 2.6245,
"theoretical_loss": 3.411532051871064,
"tokens_seen": 2137653248
},
{
"epoch": 0.65,
"learning_rate": 0.0001779739217652959,
"loss": 2.3373,
"theoretical_loss": 3.4115234452149235,
"tokens_seen": 2137718784
},
{
"epoch": 0.65,
"learning_rate": 0.00017796389167502508,
"loss": 2.4038,
"theoretical_loss": 3.411514838896509,
"tokens_seen": 2137784320
},
{
"epoch": 0.65,
"learning_rate": 0.00017795386158475426,
"loss": 2.6411,
"theoretical_loss": 3.411506232915798,
"tokens_seen": 2137849856
},
{
"epoch": 0.65,
"learning_rate": 0.00017794383149448344,
"loss": 2.4901,
"theoretical_loss": 3.411497627272766,
"tokens_seen": 2137915392
},
{
"epoch": 0.65,
"learning_rate": 0.00017793380140421265,
"loss": 2.654,
"theoretical_loss": 3.41148902196739,
"tokens_seen": 2137980928
},
{
"epoch": 0.65,
"learning_rate": 0.00017792377131394183,
"loss": 2.7421,
"theoretical_loss": 3.4114804169996464,
"tokens_seen": 2138046464
},
{
"epoch": 0.65,
"objective/train/docs_used": 1210838,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2557053565979004,
"objective/train/theoretical_loss": 3.411471812369511,
"objective/train/tokens_used": 2158572000,
"theoretical_loss": 3.411471812369511,
"tokens_seen": 2138112000
},
{
"epoch": 0.65,
"learning_rate": 0.000177913741223671,
"loss": 2.5102,
"theoretical_loss": 3.411471812369511,
"tokens_seen": 2138112000
},
{
"epoch": 0.65,
"learning_rate": 0.0001779037111334002,
"loss": 2.6754,
"theoretical_loss": 3.4114632080769605,
"tokens_seen": 2138177536
},
{
"epoch": 0.65,
"learning_rate": 0.0001778936810431294,
"loss": 2.3975,
"theoretical_loss": 3.4114546041219715,
"tokens_seen": 2138243072
},
{
"epoch": 0.65,
"learning_rate": 0.00017788365095285858,
"loss": 2.5553,
"theoretical_loss": 3.4114460005045206,
"tokens_seen": 2138308608
},
{
"epoch": 0.65,
"learning_rate": 0.00017787362086258776,
"loss": 2.7034,
"theoretical_loss": 3.4114373972245837,
"tokens_seen": 2138374144
},
{
"epoch": 0.65,
"learning_rate": 0.00017786359077231695,
"loss": 2.6248,
"theoretical_loss": 3.4114287942821377,
"tokens_seen": 2138439680
},
{
"epoch": 0.65,
"learning_rate": 0.00017785356068204613,
"loss": 2.4796,
"theoretical_loss": 3.4114201916771583,
"tokens_seen": 2138505216
},
{
"epoch": 0.65,
"learning_rate": 0.00017784353059177533,
"loss": 2.5992,
"theoretical_loss": 3.4114115894096226,
"tokens_seen": 2138570752
},
{
"epoch": 0.65,
"learning_rate": 0.00017783350050150452,
"loss": 2.8155,
"theoretical_loss": 3.411402987479507,
"tokens_seen": 2138636288
},
{
"epoch": 0.65,
"learning_rate": 0.0001778234704112337,
"loss": 2.6061,
"theoretical_loss": 3.4113943858867875,
"tokens_seen": 2138701824
},
{
"epoch": 0.65,
"learning_rate": 0.00017781344032096288,
"loss": 2.5752,
"theoretical_loss": 3.411385784631441,
"tokens_seen": 2138767360
},
{
"epoch": 0.65,
"learning_rate": 0.00017780341023069209,
"loss": 2.7388,
"theoretical_loss": 3.4113771837134435,
"tokens_seen": 2138832896
},
{
"epoch": 0.65,
"learning_rate": 0.00017779338014042127,
"loss": 2.5193,
"theoretical_loss": 3.411368583132772,
"tokens_seen": 2138898432
},
{
"epoch": 0.65,
"learning_rate": 0.00017778335005015045,
"loss": 2.5799,
"theoretical_loss": 3.411359982889402,
"tokens_seen": 2138963968
},
{
"epoch": 0.65,
"learning_rate": 0.00017777331995987963,
"loss": 2.5239,
"theoretical_loss": 3.4113513829833106,
"tokens_seen": 2139029504
},
{
"epoch": 0.65,
"learning_rate": 0.0001777632898696088,
"loss": 2.5709,
"theoretical_loss": 3.411342783414474,
"tokens_seen": 2139095040
},
{
"epoch": 0.65,
"learning_rate": 0.00017775325977933802,
"loss": 2.6271,
"theoretical_loss": 3.411334184182869,
"tokens_seen": 2139160576
},
{
"epoch": 0.65,
"learning_rate": 0.0001777432296890672,
"loss": 2.6543,
"theoretical_loss": 3.411325585288472,
"tokens_seen": 2139226112
},
{
"epoch": 0.65,
"learning_rate": 0.00017773319959879638,
"loss": 2.7498,
"theoretical_loss": 3.411316986731259,
"tokens_seen": 2139291648
},
{
"epoch": 0.65,
"learning_rate": 0.00017772316950852556,
"loss": 2.8135,
"theoretical_loss": 3.411308388511207,
"tokens_seen": 2139357184
},
{
"epoch": 0.65,
"learning_rate": 0.00017771313941825477,
"loss": 2.6262,
"theoretical_loss": 3.4112997906282923,
"tokens_seen": 2139422720
},
{
"epoch": 0.65,
"learning_rate": 0.00017770310932798395,
"loss": 2.5642,
"theoretical_loss": 3.4112911930824907,
"tokens_seen": 2139488256
},
{
"epoch": 0.65,
"learning_rate": 0.00017769307923771313,
"loss": 2.3489,
"theoretical_loss": 3.4112825958737796,
"tokens_seen": 2139553792
},
{
"epoch": 0.65,
"learning_rate": 0.00017768304914744231,
"loss": 2.684,
"theoretical_loss": 3.411273999002135,
"tokens_seen": 2139619328
},
{
"epoch": 0.65,
"learning_rate": 0.00017767301905717152,
"loss": 2.5586,
"theoretical_loss": 3.4112654024675333,
"tokens_seen": 2139684864
},
{
"epoch": 0.65,
"objective/train/docs_used": 1211569,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.387763261795044,
"objective/train/theoretical_loss": 3.411256806269951,
"objective/train/tokens_used": 2160210400,
"theoretical_loss": 3.411256806269951,
"tokens_seen": 2139750400
},
{
"epoch": 0.65,
"learning_rate": 0.0001776629889669007,
"loss": 2.53,
"theoretical_loss": 3.411256806269951,
"tokens_seen": 2139750400
},
{
"epoch": 0.65,
"learning_rate": 0.00017765295887662988,
"loss": 2.6496,
"theoretical_loss": 3.411248210409364,
"tokens_seen": 2139815936
},
{
"epoch": 0.65,
"learning_rate": 0.00017764292878635907,
"loss": 2.5823,
"theoretical_loss": 3.4112396148857504,
"tokens_seen": 2139881472
},
{
"epoch": 0.65,
"learning_rate": 0.00017763289869608825,
"loss": 2.5997,
"theoretical_loss": 3.411231019699086,
"tokens_seen": 2139947008
},
{
"epoch": 0.65,
"learning_rate": 0.00017762286860581745,
"loss": 2.4716,
"theoretical_loss": 3.411222424849346,
"tokens_seen": 2140012544
},
{
"epoch": 0.65,
"learning_rate": 0.00017761283851554664,
"loss": 2.5856,
"theoretical_loss": 3.411213830336508,
"tokens_seen": 2140078080
},
{
"epoch": 0.65,
"learning_rate": 0.00017760280842527582,
"loss": 2.3719,
"theoretical_loss": 3.4112052361605487,
"tokens_seen": 2140143616
},
{
"epoch": 0.65,
"learning_rate": 0.000177592778335005,
"loss": 2.4915,
"theoretical_loss": 3.4111966423214435,
"tokens_seen": 2140209152
},
{
"epoch": 0.65,
"learning_rate": 0.0001775827482447342,
"loss": 2.6491,
"theoretical_loss": 3.41118804881917,
"tokens_seen": 2140274688
},
{
"epoch": 0.65,
"learning_rate": 0.0001775727181544634,
"loss": 2.6376,
"theoretical_loss": 3.4111794556537047,
"tokens_seen": 2140340224
},
{
"epoch": 0.65,
"learning_rate": 0.00017756268806419257,
"loss": 2.6277,
"theoretical_loss": 3.411170862825023,
"tokens_seen": 2140405760
},
{
"epoch": 0.65,
"learning_rate": 0.00017755265797392175,
"loss": 2.6927,
"theoretical_loss": 3.4111622703331017,
"tokens_seen": 2140471296
},
{
"epoch": 0.65,
"learning_rate": 0.00017754262788365096,
"loss": 2.5417,
"theoretical_loss": 3.4111536781779184,
"tokens_seen": 2140536832
},
{
"epoch": 0.65,
"learning_rate": 0.00017753259779338017,
"loss": 2.6766,
"theoretical_loss": 3.4111450863594484,
"tokens_seen": 2140602368
},
{
"epoch": 0.65,
"learning_rate": 0.00017752256770310935,
"loss": 2.5712,
"theoretical_loss": 3.4111364948776686,
"tokens_seen": 2140667904
},
{
"epoch": 0.65,
"learning_rate": 0.00017751253761283853,
"loss": 2.5318,
"theoretical_loss": 3.411127903732556,
"tokens_seen": 2140733440
},
{
"epoch": 0.65,
"learning_rate": 0.0001775025075225677,
"loss": 2.4099,
"theoretical_loss": 3.411119312924086,
"tokens_seen": 2140798976
},
{
"epoch": 0.65,
"learning_rate": 0.00017749247743229692,
"loss": 2.5908,
"theoretical_loss": 3.4111107224522357,
"tokens_seen": 2140864512
},
{
"epoch": 0.65,
"learning_rate": 0.0001774824473420261,
"loss": 2.4425,
"theoretical_loss": 3.411102132316982,
"tokens_seen": 2140930048
},
{
"epoch": 0.65,
"learning_rate": 0.00017747241725175528,
"loss": 2.4382,
"theoretical_loss": 3.411093542518301,
"tokens_seen": 2140995584
},
{
"epoch": 0.65,
"learning_rate": 0.00017746238716148446,
"loss": 2.6759,
"theoretical_loss": 3.411084953056169,
"tokens_seen": 2141061120
},
{
"epoch": 0.65,
"learning_rate": 0.00017745235707121364,
"loss": 2.6807,
"theoretical_loss": 3.411076363930563,
"tokens_seen": 2141126656
},
{
"epoch": 0.65,
"learning_rate": 0.00017744232698094285,
"loss": 2.6496,
"theoretical_loss": 3.411067775141459,
"tokens_seen": 2141192192
},
{
"epoch": 0.65,
"learning_rate": 0.00017743229689067203,
"loss": 2.6305,
"theoretical_loss": 3.411059186688834,
"tokens_seen": 2141257728
},
{
"epoch": 0.65,
"learning_rate": 0.0001774222668004012,
"loss": 2.7431,
"theoretical_loss": 3.411050598572664,
"tokens_seen": 2141323264
},
{
"epoch": 0.65,
"objective/train/docs_used": 1213144,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.718681573867798,
"objective/train/theoretical_loss": 3.4110420107929262,
"objective/train/tokens_used": 2161848800,
"theoretical_loss": 3.4110420107929262,
"tokens_seen": 2141388800
},
{
"epoch": 0.65,
"learning_rate": 0.0001774122367101304,
"loss": 2.3593,
"theoretical_loss": 3.4110420107929262,
"tokens_seen": 2141388800
},
{
"epoch": 0.65,
"learning_rate": 0.0001774022066198596,
"loss": 2.4249,
"theoretical_loss": 3.4110334233495965,
"tokens_seen": 2141454336
},
{
"epoch": 0.65,
"learning_rate": 0.00017739217652958878,
"loss": 2.5001,
"theoretical_loss": 3.4110248362426514,
"tokens_seen": 2141519872
},
{
"epoch": 0.65,
"learning_rate": 0.00017738214643931796,
"loss": 2.5808,
"theoretical_loss": 3.4110162494720684,
"tokens_seen": 2141585408
},
{
"epoch": 0.65,
"learning_rate": 0.00017737211634904715,
"loss": 2.4715,
"theoretical_loss": 3.4110076630378225,
"tokens_seen": 2141650944
},
{
"epoch": 0.65,
"learning_rate": 0.00017736208625877633,
"loss": 2.7513,
"theoretical_loss": 3.410999076939892,
"tokens_seen": 2141716480
},
{
"epoch": 0.65,
"learning_rate": 0.00017735205616850553,
"loss": 2.7197,
"theoretical_loss": 3.410990491178252,
"tokens_seen": 2141782016
},
{
"epoch": 0.65,
"learning_rate": 0.00017734202607823472,
"loss": 2.5294,
"theoretical_loss": 3.4109819057528794,
"tokens_seen": 2141847552
},
{
"epoch": 0.65,
"learning_rate": 0.0001773319959879639,
"loss": 2.58,
"theoretical_loss": 3.4109733206637514,
"tokens_seen": 2141913088
},
{
"epoch": 0.65,
"learning_rate": 0.00017732196589769308,
"loss": 2.698,
"theoretical_loss": 3.4109647359108437,
"tokens_seen": 2141978624
},
{
"epoch": 0.65,
"learning_rate": 0.00017731193580742229,
"loss": 2.6247,
"theoretical_loss": 3.4109561514941333,
"tokens_seen": 2142044160
},
{
"epoch": 0.65,
"learning_rate": 0.00017730190571715147,
"loss": 2.5492,
"theoretical_loss": 3.4109475674135963,
"tokens_seen": 2142109696
},
{
"epoch": 0.65,
"learning_rate": 0.00017729187562688065,
"loss": 2.48,
"theoretical_loss": 3.41093898366921,
"tokens_seen": 2142175232
},
{
"epoch": 0.65,
"learning_rate": 0.00017728184553660983,
"loss": 2.707,
"theoretical_loss": 3.41093040026095,
"tokens_seen": 2142240768
},
{
"epoch": 0.65,
"learning_rate": 0.000177271815446339,
"loss": 2.7641,
"theoretical_loss": 3.410921817188794,
"tokens_seen": 2142306304
},
{
"epoch": 0.65,
"learning_rate": 0.00017726178535606822,
"loss": 2.6138,
"theoretical_loss": 3.4109132344527175,
"tokens_seen": 2142371840
},
{
"epoch": 0.65,
"learning_rate": 0.0001772517552657974,
"loss": 2.4884,
"theoretical_loss": 3.4109046520526975,
"tokens_seen": 2142437376
},
{
"epoch": 0.65,
"learning_rate": 0.00017724172517552658,
"loss": 2.6699,
"theoretical_loss": 3.410896069988711,
"tokens_seen": 2142502912
},
{
"epoch": 0.65,
"learning_rate": 0.00017723169508525576,
"loss": 2.3238,
"theoretical_loss": 3.410887488260734,
"tokens_seen": 2142568448
},
{
"epoch": 0.65,
"learning_rate": 0.00017722166499498497,
"loss": 2.6163,
"theoretical_loss": 3.410878906868743,
"tokens_seen": 2142633984
},
{
"epoch": 0.65,
"learning_rate": 0.00017721163490471415,
"loss": 2.7469,
"theoretical_loss": 3.4108703258127147,
"tokens_seen": 2142699520
},
{
"epoch": 0.65,
"learning_rate": 0.00017720160481444333,
"loss": 2.5502,
"theoretical_loss": 3.4108617450926255,
"tokens_seen": 2142765056
},
{
"epoch": 0.65,
"learning_rate": 0.00017719157472417251,
"loss": 2.8245,
"theoretical_loss": 3.4108531647084526,
"tokens_seen": 2142830592
},
{
"epoch": 0.65,
"learning_rate": 0.00017718154463390172,
"loss": 2.5295,
"theoretical_loss": 3.4108445846601727,
"tokens_seen": 2142896128
},
{
"epoch": 0.65,
"learning_rate": 0.0001771715145436309,
"loss": 2.507,
"theoretical_loss": 3.410836004947761,
"tokens_seen": 2142961664
},
{
"epoch": 0.65,
"objective/train/docs_used": 1213818,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.875296115875244,
"objective/train/theoretical_loss": 3.4108274255711954,
"objective/train/tokens_used": 2163487200,
"theoretical_loss": 3.4108274255711954,
"tokens_seen": 2143027200
},
{
"epoch": 0.65,
"learning_rate": 0.00017716148445336008,
"loss": 2.5119,
"theoretical_loss": 3.4108274255711954,
"tokens_seen": 2143027200
},
{
"epoch": 0.65,
"learning_rate": 0.00017715145436308927,
"loss": 2.6274,
"theoretical_loss": 3.410818846530452,
"tokens_seen": 2143092736
},
{
"epoch": 0.65,
"learning_rate": 0.00017714142427281845,
"loss": 2.5899,
"theoretical_loss": 3.4108102678255072,
"tokens_seen": 2143158272
},
{
"epoch": 0.65,
"learning_rate": 0.00017713139418254765,
"loss": 2.8387,
"theoretical_loss": 3.410801689456338,
"tokens_seen": 2143223808
},
{
"epoch": 0.65,
"learning_rate": 0.00017712136409227684,
"loss": 2.6252,
"theoretical_loss": 3.4107931114229206,
"tokens_seen": 2143289344
},
{
"epoch": 0.65,
"learning_rate": 0.00017711133400200602,
"loss": 2.6747,
"theoretical_loss": 3.410784533725232,
"tokens_seen": 2143354880
},
{
"epoch": 0.65,
"learning_rate": 0.0001771013039117352,
"loss": 2.3798,
"theoretical_loss": 3.4107759563632487,
"tokens_seen": 2143420416
},
{
"epoch": 0.65,
"learning_rate": 0.0001770912738214644,
"loss": 2.5385,
"theoretical_loss": 3.4107673793369466,
"tokens_seen": 2143485952
},
{
"epoch": 0.65,
"learning_rate": 0.0001770812437311936,
"loss": 2.8671,
"theoretical_loss": 3.4107588026463036,
"tokens_seen": 2143551488
},
{
"epoch": 0.65,
"learning_rate": 0.00017707121364092277,
"loss": 2.5468,
"theoretical_loss": 3.410750226291295,
"tokens_seen": 2143617024
},
{
"epoch": 0.65,
"learning_rate": 0.00017706118355065195,
"loss": 2.6335,
"theoretical_loss": 3.4107416502718984,
"tokens_seen": 2143682560
},
{
"epoch": 0.65,
"learning_rate": 0.00017705115346038113,
"loss": 2.4554,
"theoretical_loss": 3.41073307458809,
"tokens_seen": 2143748096
},
{
"epoch": 0.65,
"learning_rate": 0.00017704112337011034,
"loss": 2.5986,
"theoretical_loss": 3.4107244992398464,
"tokens_seen": 2143813632
},
{
"epoch": 0.65,
"learning_rate": 0.00017703109327983952,
"loss": 2.6381,
"theoretical_loss": 3.4107159242271443,
"tokens_seen": 2143879168
},
{
"epoch": 0.65,
"learning_rate": 0.0001770210631895687,
"loss": 2.6588,
"theoretical_loss": 3.41070734954996,
"tokens_seen": 2143944704
},
{
"epoch": 0.65,
"learning_rate": 0.00017701103309929788,
"loss": 2.4972,
"theoretical_loss": 3.41069877520827,
"tokens_seen": 2144010240
},
{
"epoch": 0.65,
"learning_rate": 0.0001770010030090271,
"loss": 2.8373,
"theoretical_loss": 3.410690201202052,
"tokens_seen": 2144075776
},
{
"epoch": 0.65,
"learning_rate": 0.00017699097291875627,
"loss": 2.3428,
"theoretical_loss": 3.4106816275312815,
"tokens_seen": 2144141312
},
{
"epoch": 0.65,
"learning_rate": 0.00017698094282848545,
"loss": 2.3266,
"theoretical_loss": 3.410673054195936,
"tokens_seen": 2144206848
},
{
"epoch": 0.65,
"learning_rate": 0.00017697091273821463,
"loss": 2.5674,
"theoretical_loss": 3.410664481195991,
"tokens_seen": 2144272384
},
{
"epoch": 0.65,
"learning_rate": 0.00017696088264794382,
"loss": 2.728,
"theoretical_loss": 3.4106559085314236,
"tokens_seen": 2144337920
},
{
"epoch": 0.65,
"learning_rate": 0.00017695085255767302,
"loss": 2.6459,
"theoretical_loss": 3.4106473362022114,
"tokens_seen": 2144403456
},
{
"epoch": 0.65,
"learning_rate": 0.0001769408224674022,
"loss": 2.5484,
"theoretical_loss": 3.41063876420833,
"tokens_seen": 2144468992
},
{
"epoch": 0.65,
"learning_rate": 0.00017693079237713139,
"loss": 2.4781,
"theoretical_loss": 3.410630192549756,
"tokens_seen": 2144534528
},
{
"epoch": 0.65,
"learning_rate": 0.00017692076228686057,
"loss": 2.324,
"theoretical_loss": 3.4106216212264666,
"tokens_seen": 2144600064
},
{
"epoch": 0.65,
"objective/train/docs_used": 1215022,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3268635272979736,
"objective/train/theoretical_loss": 3.4106130502384375,
"objective/train/tokens_used": 2165125600,
"theoretical_loss": 3.4106130502384375,
"tokens_seen": 2144665600
},
{
"epoch": 0.65,
"learning_rate": 0.00017691073219658978,
"loss": 2.4777,
"theoretical_loss": 3.4106130502384375,
"tokens_seen": 2144665600
},
{
"epoch": 0.65,
"learning_rate": 0.00017690070210631896,
"loss": 2.3819,
"theoretical_loss": 3.410604479585647,
"tokens_seen": 2144731136
},
{
"epoch": 0.65,
"learning_rate": 0.00017689067201604814,
"loss": 2.5412,
"theoretical_loss": 3.4105959092680695,
"tokens_seen": 2144796672
},
{
"epoch": 0.65,
"learning_rate": 0.00017688064192577732,
"loss": 2.4943,
"theoretical_loss": 3.410587339285684,
"tokens_seen": 2144862208
},
{
"epoch": 0.65,
"learning_rate": 0.0001768706118355065,
"loss": 2.5956,
"theoretical_loss": 3.4105787696384655,
"tokens_seen": 2144927744
},
{
"epoch": 0.65,
"learning_rate": 0.0001768605817452357,
"loss": 2.552,
"theoretical_loss": 3.4105702003263914,
"tokens_seen": 2144993280
},
{
"epoch": 0.65,
"learning_rate": 0.0001768505516549649,
"loss": 2.5973,
"theoretical_loss": 3.410561631349438,
"tokens_seen": 2145058816
},
{
"epoch": 0.65,
"learning_rate": 0.00017684052156469407,
"loss": 2.7724,
"theoretical_loss": 3.410553062707582,
"tokens_seen": 2145124352
},
{
"epoch": 0.65,
"learning_rate": 0.00017683049147442325,
"loss": 2.7437,
"theoretical_loss": 3.4105444944008,
"tokens_seen": 2145189888
},
{
"epoch": 0.65,
"learning_rate": 0.00017682046138415246,
"loss": 2.6389,
"theoretical_loss": 3.4105359264290693,
"tokens_seen": 2145255424
},
{
"epoch": 0.65,
"learning_rate": 0.00017681043129388164,
"loss": 2.4541,
"theoretical_loss": 3.4105273587923657,
"tokens_seen": 2145320960
},
{
"epoch": 0.65,
"learning_rate": 0.00017680040120361085,
"loss": 2.3729,
"theoretical_loss": 3.4105187914906665,
"tokens_seen": 2145386496
},
{
"epoch": 0.65,
"learning_rate": 0.00017679037111334003,
"loss": 2.6536,
"theoretical_loss": 3.4105102245239483,
"tokens_seen": 2145452032
},
{
"epoch": 0.65,
"learning_rate": 0.0001767803410230692,
"loss": 2.6434,
"theoretical_loss": 3.410501657892187,
"tokens_seen": 2145517568
},
{
"epoch": 0.65,
"learning_rate": 0.00017677031093279842,
"loss": 2.471,
"theoretical_loss": 3.4104930915953604,
"tokens_seen": 2145583104
},
{
"epoch": 0.65,
"learning_rate": 0.0001767602808425276,
"loss": 2.474,
"theoretical_loss": 3.410484525633444,
"tokens_seen": 2145648640
},
{
"epoch": 0.65,
"learning_rate": 0.00017675025075225678,
"loss": 2.414,
"theoretical_loss": 3.4104759600064156,
"tokens_seen": 2145714176
},
{
"epoch": 0.65,
"learning_rate": 0.00017674022066198596,
"loss": 2.4408,
"theoretical_loss": 3.410467394714251,
"tokens_seen": 2145779712
},
{
"epoch": 0.65,
"learning_rate": 0.00017673019057171517,
"loss": 2.6457,
"theoretical_loss": 3.4104588297569283,
"tokens_seen": 2145845248
},
{
"epoch": 0.65,
"learning_rate": 0.00017672016048144435,
"loss": 2.6484,
"theoretical_loss": 3.410450265134422,
"tokens_seen": 2145910784
},
{
"epoch": 0.65,
"learning_rate": 0.00017671013039117353,
"loss": 2.58,
"theoretical_loss": 3.41044170084671,
"tokens_seen": 2145976320
},
{
"epoch": 0.65,
"learning_rate": 0.00017670010030090271,
"loss": 2.4516,
"theoretical_loss": 3.4104331368937695,
"tokens_seen": 2146041856
},
{
"epoch": 0.65,
"learning_rate": 0.00017669007021063192,
"loss": 2.506,
"theoretical_loss": 3.4104245732755762,
"tokens_seen": 2146107392
},
{
"epoch": 0.65,
"learning_rate": 0.0001766800401203611,
"loss": 2.6321,
"theoretical_loss": 3.4104160099921077,
"tokens_seen": 2146172928
},
{
"epoch": 0.65,
"learning_rate": 0.00017667001003009028,
"loss": 2.5626,
"theoretical_loss": 3.41040744704334,
"tokens_seen": 2146238464
},
{
"epoch": 0.65,
"objective/train/docs_used": 1215651,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.312642812728882,
"objective/train/theoretical_loss": 3.4103988844292497,
"objective/train/tokens_used": 2166764000,
"theoretical_loss": 3.4103988844292497,
"tokens_seen": 2146304000
},
{
"epoch": 0.65,
"learning_rate": 0.00017665997993981947,
"loss": 2.5198,
"theoretical_loss": 3.4103988844292497,
"tokens_seen": 2146304000
},
{
"epoch": 0.65,
"learning_rate": 0.00017664994984954865,
"loss": 2.7462,
"theoretical_loss": 3.410390322149814,
"tokens_seen": 2146369536
},
{
"epoch": 0.65,
"learning_rate": 0.00017663991975927786,
"loss": 2.6555,
"theoretical_loss": 3.4103817602050093,
"tokens_seen": 2146435072
},
{
"epoch": 0.65,
"learning_rate": 0.00017662988966900704,
"loss": 2.7149,
"theoretical_loss": 3.4103731985948125,
"tokens_seen": 2146500608
},
{
"epoch": 0.65,
"learning_rate": 0.00017661985957873622,
"loss": 2.591,
"theoretical_loss": 3.4103646373192005,
"tokens_seen": 2146566144
},
{
"epoch": 0.65,
"learning_rate": 0.0001766098294884654,
"loss": 2.5796,
"theoretical_loss": 3.4103560763781493,
"tokens_seen": 2146631680
},
{
"epoch": 0.65,
"learning_rate": 0.0001765997993981946,
"loss": 2.6252,
"theoretical_loss": 3.4103475157716363,
"tokens_seen": 2146697216
},
{
"epoch": 0.65,
"learning_rate": 0.0001765897693079238,
"loss": 2.5943,
"theoretical_loss": 3.410338955499638,
"tokens_seen": 2146762752
},
{
"epoch": 0.65,
"learning_rate": 0.00017657973921765297,
"loss": 2.6472,
"theoretical_loss": 3.4103303955621307,
"tokens_seen": 2146828288
},
{
"epoch": 0.65,
"learning_rate": 0.00017656970912738215,
"loss": 2.5983,
"theoretical_loss": 3.410321835959092,
"tokens_seen": 2146893824
},
{
"epoch": 0.65,
"learning_rate": 0.00017655967903711133,
"loss": 2.7088,
"theoretical_loss": 3.4103132766904976,
"tokens_seen": 2146959360
},
{
"epoch": 0.65,
"learning_rate": 0.00017654964894684054,
"loss": 2.696,
"theoretical_loss": 3.4103047177563255,
"tokens_seen": 2147024896
},
{
"epoch": 0.65,
"learning_rate": 0.00017653961885656972,
"loss": 2.8773,
"theoretical_loss": 3.410296159156551,
"tokens_seen": 2147090432
},
{
"epoch": 0.65,
"learning_rate": 0.0001765295887662989,
"loss": 2.5828,
"theoretical_loss": 3.410287600891152,
"tokens_seen": 2147155968
},
{
"epoch": 0.65,
"learning_rate": 0.00017651955867602808,
"loss": 2.7094,
"theoretical_loss": 3.410279042960104,
"tokens_seen": 2147221504
},
{
"epoch": 0.65,
"learning_rate": 0.0001765095285857573,
"loss": 2.7085,
"theoretical_loss": 3.4102704853633847,
"tokens_seen": 2147287040
},
{
"epoch": 0.65,
"learning_rate": 0.00017649949849548647,
"loss": 2.5373,
"theoretical_loss": 3.4102619281009714,
"tokens_seen": 2147352576
},
{
"epoch": 0.65,
"learning_rate": 0.00017648946840521565,
"loss": 2.6434,
"theoretical_loss": 3.410253371172839,
"tokens_seen": 2147418112
},
{
"epoch": 0.65,
"learning_rate": 0.00017647943831494483,
"loss": 2.5995,
"theoretical_loss": 3.4102448145789657,
"tokens_seen": 2147483648
},
{
"epoch": 0.65,
"learning_rate": 0.00017646940822467402,
"loss": 2.5694,
"theoretical_loss": 3.410236258319328,
"tokens_seen": 2147549184
},
{
"epoch": 0.65,
"learning_rate": 0.00017645937813440322,
"loss": 2.6345,
"theoretical_loss": 3.4102277023939025,
"tokens_seen": 2147614720
},
{
"epoch": 0.65,
"learning_rate": 0.0001764493480441324,
"loss": 2.5879,
"theoretical_loss": 3.4102191468026657,
"tokens_seen": 2147680256
},
{
"epoch": 0.65,
"learning_rate": 0.00017643931795386159,
"loss": 2.661,
"theoretical_loss": 3.4102105915455945,
"tokens_seen": 2147745792
},
{
"epoch": 0.65,
"learning_rate": 0.00017642928786359077,
"loss": 2.7249,
"theoretical_loss": 3.4102020366226657,
"tokens_seen": 2147811328
},
{
"epoch": 0.65,
"learning_rate": 0.00017641925777331998,
"loss": 2.4326,
"theoretical_loss": 3.410193482033856,
"tokens_seen": 2147876864
},
{
"epoch": 0.65,
"objective/train/docs_used": 1216251,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5646564960479736,
"objective/train/theoretical_loss": 3.4101849277791425,
"objective/train/tokens_used": 2168402400,
"theoretical_loss": 3.4101849277791425,
"tokens_seen": 2147942400
},
{
"epoch": 0.65,
"learning_rate": 0.00017640922768304916,
"loss": 2.4756,
"theoretical_loss": 3.4101849277791425,
"tokens_seen": 2147942400
},
{
"epoch": 0.65,
"learning_rate": 0.00017639919759277834,
"loss": 2.5452,
"theoretical_loss": 3.4101763738585014,
"tokens_seen": 2148007936
},
{
"epoch": 0.65,
"learning_rate": 0.00017638916750250752,
"loss": 2.3855,
"theoretical_loss": 3.4101678202719103,
"tokens_seen": 2148073472
},
{
"epoch": 0.65,
"learning_rate": 0.0001763791374122367,
"loss": 2.5292,
"theoretical_loss": 3.4101592670193446,
"tokens_seen": 2148139008
},
{
"epoch": 0.65,
"learning_rate": 0.0001763691073219659,
"loss": 2.8049,
"theoretical_loss": 3.4101507141007827,
"tokens_seen": 2148204544
},
{
"epoch": 0.65,
"learning_rate": 0.0001763590772316951,
"loss": 2.5174,
"theoretical_loss": 3.4101421615162,
"tokens_seen": 2148270080
},
{
"epoch": 0.65,
"learning_rate": 0.00017634904714142427,
"loss": 2.5629,
"theoretical_loss": 3.410133609265574,
"tokens_seen": 2148335616
},
{
"epoch": 0.65,
"learning_rate": 0.00017633901705115345,
"loss": 2.5029,
"theoretical_loss": 3.410125057348881,
"tokens_seen": 2148401152
},
{
"epoch": 0.65,
"learning_rate": 0.00017632898696088266,
"loss": 2.6862,
"theoretical_loss": 3.4101165057660987,
"tokens_seen": 2148466688
},
{
"epoch": 0.65,
"learning_rate": 0.00017631895687061184,
"loss": 2.7941,
"theoretical_loss": 3.4101079545172026,
"tokens_seen": 2148532224
},
{
"epoch": 0.65,
"learning_rate": 0.00017630892678034102,
"loss": 2.7439,
"theoretical_loss": 3.41009940360217,
"tokens_seen": 2148597760
},
{
"epoch": 0.65,
"learning_rate": 0.0001762988966900702,
"loss": 2.4822,
"theoretical_loss": 3.4100908530209786,
"tokens_seen": 2148663296
},
{
"epoch": 0.65,
"learning_rate": 0.00017628886659979938,
"loss": 2.7426,
"theoretical_loss": 3.4100823027736036,
"tokens_seen": 2148728832
},
{
"epoch": 0.65,
"learning_rate": 0.0001762788365095286,
"loss": 2.4467,
"theoretical_loss": 3.410073752860023,
"tokens_seen": 2148794368
},
{
"epoch": 0.65,
"learning_rate": 0.00017626880641925777,
"loss": 2.6608,
"theoretical_loss": 3.4100652032802135,
"tokens_seen": 2148859904
},
{
"epoch": 0.65,
"learning_rate": 0.00017625877632898695,
"loss": 2.381,
"theoretical_loss": 3.410056654034151,
"tokens_seen": 2148925440
},
{
"epoch": 0.65,
"learning_rate": 0.00017624874623871614,
"loss": 2.5904,
"theoretical_loss": 3.4100481051218132,
"tokens_seen": 2148990976
},
{
"epoch": 0.65,
"learning_rate": 0.00017623871614844534,
"loss": 2.547,
"theoretical_loss": 3.4100395565431763,
"tokens_seen": 2149056512
},
{
"epoch": 0.65,
"learning_rate": 0.00017622868605817452,
"loss": 2.6132,
"theoretical_loss": 3.4100310082982173,
"tokens_seen": 2149122048
},
{
"epoch": 0.65,
"learning_rate": 0.0001762186559679037,
"loss": 2.5202,
"theoretical_loss": 3.410022460386913,
"tokens_seen": 2149187584
},
{
"epoch": 0.65,
"learning_rate": 0.0001762086258776329,
"loss": 2.7424,
"theoretical_loss": 3.4100139128092404,
"tokens_seen": 2149253120
},
{
"epoch": 0.65,
"learning_rate": 0.00017619859578736207,
"loss": 2.5365,
"theoretical_loss": 3.4100053655651763,
"tokens_seen": 2149318656
},
{
"epoch": 0.65,
"learning_rate": 0.00017618856569709128,
"loss": 2.411,
"theoretical_loss": 3.409996818654697,
"tokens_seen": 2149384192
},
{
"epoch": 0.65,
"learning_rate": 0.00017617853560682046,
"loss": 2.5075,
"theoretical_loss": 3.4099882720777797,
"tokens_seen": 2149449728
},
{
"epoch": 0.65,
"learning_rate": 0.00017616850551654964,
"loss": 2.4369,
"theoretical_loss": 3.4099797258344013,
"tokens_seen": 2149515264
},
{
"epoch": 0.65,
"objective/train/docs_used": 1217415,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.540207862854004,
"objective/train/theoretical_loss": 3.4099711799245385,
"objective/train/tokens_used": 2170040800,
"theoretical_loss": 3.4099711799245385,
"tokens_seen": 2149580800
},
{
"epoch": 0.65,
"learning_rate": 0.00017615847542627882,
"loss": 2.574,
"theoretical_loss": 3.4099711799245385,
"tokens_seen": 2149580800
},
{
"epoch": 0.65,
"learning_rate": 0.00017614844533600803,
"loss": 2.5918,
"theoretical_loss": 3.4099626343481684,
"tokens_seen": 2149646336
},
{
"epoch": 0.65,
"learning_rate": 0.0001761384152457372,
"loss": 2.7151,
"theoretical_loss": 3.409954089105267,
"tokens_seen": 2149711872
},
{
"epoch": 0.65,
"learning_rate": 0.0001761283851554664,
"loss": 2.6573,
"theoretical_loss": 3.4099455441958124,
"tokens_seen": 2149777408
},
{
"epoch": 0.65,
"learning_rate": 0.00017611835506519557,
"loss": 2.4264,
"theoretical_loss": 3.40993699961978,
"tokens_seen": 2149842944
},
{
"epoch": 0.65,
"learning_rate": 0.00017610832497492478,
"loss": 2.595,
"theoretical_loss": 3.4099284553771474,
"tokens_seen": 2149908480
},
{
"epoch": 0.65,
"learning_rate": 0.00017609829488465396,
"loss": 2.6571,
"theoretical_loss": 3.4099199114678918,
"tokens_seen": 2149974016
},
{
"epoch": 0.65,
"learning_rate": 0.00017608826479438314,
"loss": 2.4376,
"theoretical_loss": 3.409911367891989,
"tokens_seen": 2150039552
},
{
"epoch": 0.65,
"learning_rate": 0.00017607823470411232,
"loss": 2.5251,
"theoretical_loss": 3.4099028246494165,
"tokens_seen": 2150105088
},
{
"epoch": 0.65,
"learning_rate": 0.0001760682046138415,
"loss": 2.6515,
"theoretical_loss": 3.409894281740151,
"tokens_seen": 2150170624
},
{
"epoch": 0.65,
"learning_rate": 0.0001760581745235707,
"loss": 2.5343,
"theoretical_loss": 3.40988573916417,
"tokens_seen": 2150236160
},
{
"epoch": 0.65,
"learning_rate": 0.00017604814443329992,
"loss": 2.6314,
"theoretical_loss": 3.4098771969214487,
"tokens_seen": 2150301696
},
{
"epoch": 0.65,
"learning_rate": 0.0001760381143430291,
"loss": 2.4113,
"theoretical_loss": 3.409868655011966,
"tokens_seen": 2150367232
},
{
"epoch": 0.65,
"learning_rate": 0.00017602808425275828,
"loss": 2.6195,
"theoretical_loss": 3.4098601134356965,
"tokens_seen": 2150432768
},
{
"epoch": 0.65,
"learning_rate": 0.0001760180541624875,
"loss": 2.4525,
"theoretical_loss": 3.4098515721926193,
"tokens_seen": 2150498304
},
{
"epoch": 0.65,
"learning_rate": 0.00017600802407221667,
"loss": 2.7922,
"theoretical_loss": 3.4098430312827097,
"tokens_seen": 2150563840
},
{
"epoch": 0.65,
"learning_rate": 0.00017599799398194585,
"loss": 2.6155,
"theoretical_loss": 3.4098344907059452,
"tokens_seen": 2150629376
},
{
"epoch": 0.65,
"learning_rate": 0.00017598796389167503,
"loss": 2.5453,
"theoretical_loss": 3.4098259504623023,
"tokens_seen": 2150694912
},
{
"epoch": 0.65,
"learning_rate": 0.00017597793380140422,
"loss": 2.6499,
"theoretical_loss": 3.4098174105517582,
"tokens_seen": 2150760448
},
{
"epoch": 0.65,
"learning_rate": 0.00017596790371113342,
"loss": 2.411,
"theoretical_loss": 3.40980887097429,
"tokens_seen": 2150825984
},
{
"epoch": 0.65,
"learning_rate": 0.0001759578736208626,
"loss": 2.4044,
"theoretical_loss": 3.4098003317298735,
"tokens_seen": 2150891520
},
{
"epoch": 0.65,
"learning_rate": 0.00017594784353059179,
"loss": 2.4874,
"theoretical_loss": 3.4097917928184867,
"tokens_seen": 2150957056
},
{
"epoch": 0.65,
"learning_rate": 0.00017593781344032097,
"loss": 2.4475,
"theoretical_loss": 3.4097832542401063,
"tokens_seen": 2151022592
},
{
"epoch": 0.65,
"learning_rate": 0.00017592778335005018,
"loss": 2.3284,
"theoretical_loss": 3.4097747159947085,
"tokens_seen": 2151088128
},
{
"epoch": 0.65,
"learning_rate": 0.00017591775325977936,
"loss": 2.8699,
"theoretical_loss": 3.40976617808227,
"tokens_seen": 2151153664
},
{
"epoch": 0.65,
"objective/train/docs_used": 1217907,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5594141483306885,
"objective/train/theoretical_loss": 3.409757640502769,
"objective/train/tokens_used": 2171679200,
"theoretical_loss": 3.409757640502769,
"tokens_seen": 2151219200
},
{
"epoch": 0.65,
"learning_rate": 0.00017590772316950854,
"loss": 2.5608,
"theoretical_loss": 3.409757640502769,
"tokens_seen": 2151219200
},
{
"epoch": 0.65,
"learning_rate": 0.00017589769307923772,
"loss": 2.8561,
"theoretical_loss": 3.409749103256181,
"tokens_seen": 2151284736
},
{
"epoch": 0.65,
"learning_rate": 0.0001758876629889669,
"loss": 2.5473,
"theoretical_loss": 3.4097405663424842,
"tokens_seen": 2151350272
},
{
"epoch": 0.65,
"learning_rate": 0.0001758776328986961,
"loss": 2.8665,
"theoretical_loss": 3.4097320297616545,
"tokens_seen": 2151415808
},
{
"epoch": 0.65,
"learning_rate": 0.0001758676028084253,
"loss": 2.5065,
"theoretical_loss": 3.4097234935136687,
"tokens_seen": 2151481344
},
{
"epoch": 0.65,
"learning_rate": 0.00017585757271815447,
"loss": 2.4571,
"theoretical_loss": 3.4097149575985037,
"tokens_seen": 2151546880
},
{
"epoch": 0.65,
"learning_rate": 0.00017584754262788365,
"loss": 2.4712,
"theoretical_loss": 3.4097064220161375,
"tokens_seen": 2151612416
},
{
"epoch": 0.65,
"learning_rate": 0.00017583751253761286,
"loss": 2.6183,
"theoretical_loss": 3.409697886766546,
"tokens_seen": 2151677952
},
{
"epoch": 0.65,
"learning_rate": 0.00017582748244734204,
"loss": 2.5124,
"theoretical_loss": 3.409689351849706,
"tokens_seen": 2151743488
},
{
"epoch": 0.65,
"learning_rate": 0.00017581745235707122,
"loss": 2.6855,
"theoretical_loss": 3.409680817265595,
"tokens_seen": 2151809024
},
{
"epoch": 0.65,
"learning_rate": 0.0001758074222668004,
"loss": 2.3889,
"theoretical_loss": 3.4096722830141895,
"tokens_seen": 2151874560
},
{
"epoch": 0.65,
"learning_rate": 0.00017579739217652958,
"loss": 2.5835,
"theoretical_loss": 3.409663749095466,
"tokens_seen": 2151940096
},
{
"epoch": 0.65,
"learning_rate": 0.0001757873620862588,
"loss": 2.6903,
"theoretical_loss": 3.4096552155094026,
"tokens_seen": 2152005632
},
{
"epoch": 0.65,
"learning_rate": 0.00017577733199598797,
"loss": 2.4603,
"theoretical_loss": 3.409646682255975,
"tokens_seen": 2152071168
},
{
"epoch": 0.65,
"learning_rate": 0.00017576730190571715,
"loss": 2.3969,
"theoretical_loss": 3.409638149335161,
"tokens_seen": 2152136704
},
{
"epoch": 0.65,
"learning_rate": 0.00017575727181544634,
"loss": 2.185,
"theoretical_loss": 3.4096296167469364,
"tokens_seen": 2152202240
},
{
"epoch": 0.65,
"learning_rate": 0.00017574724172517554,
"loss": 2.5502,
"theoretical_loss": 3.4096210844912793,
"tokens_seen": 2152267776
},
{
"epoch": 0.65,
"learning_rate": 0.00017573721163490473,
"loss": 2.4666,
"theoretical_loss": 3.409612552568166,
"tokens_seen": 2152333312
},
{
"epoch": 0.65,
"learning_rate": 0.0001757271815446339,
"loss": 2.488,
"theoretical_loss": 3.4096040209775733,
"tokens_seen": 2152398848
},
{
"epoch": 0.65,
"learning_rate": 0.0001757171514543631,
"loss": 2.5101,
"theoretical_loss": 3.4095954897194787,
"tokens_seen": 2152464384
},
{
"epoch": 0.65,
"learning_rate": 0.00017570712136409227,
"loss": 2.6382,
"theoretical_loss": 3.4095869587938585,
"tokens_seen": 2152529920
},
{
"epoch": 0.65,
"learning_rate": 0.00017569709127382148,
"loss": 2.7211,
"theoretical_loss": 3.40957842820069,
"tokens_seen": 2152595456
},
{
"epoch": 0.65,
"learning_rate": 0.00017568706118355066,
"loss": 2.7488,
"theoretical_loss": 3.4095698979399502,
"tokens_seen": 2152660992
},
{
"epoch": 0.65,
"learning_rate": 0.00017567703109327984,
"loss": 2.8525,
"theoretical_loss": 3.4095613680116155,
"tokens_seen": 2152726528
},
{
"epoch": 0.65,
"learning_rate": 0.00017566700100300902,
"loss": 2.5481,
"theoretical_loss": 3.4095528384156633,
"tokens_seen": 2152792064
},
{
"epoch": 0.65,
"objective/train/docs_used": 1219413,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.686859369277954,
"objective/train/theoretical_loss": 3.4095443091520705,
"objective/train/tokens_used": 2173317600,
"theoretical_loss": 3.4095443091520705,
"tokens_seen": 2152857600
},
{
"epoch": 0.65,
"learning_rate": 0.00017565697091273823,
"loss": 2.6433,
"theoretical_loss": 3.4095443091520705,
"tokens_seen": 2152857600
},
{
"epoch": 0.65,
"learning_rate": 0.0001756469408224674,
"loss": 2.5783,
"theoretical_loss": 3.4095357802208133,
"tokens_seen": 2152923136
},
{
"epoch": 0.65,
"learning_rate": 0.0001756369107321966,
"loss": 2.6975,
"theoretical_loss": 3.4095272516218698,
"tokens_seen": 2152988672
},
{
"epoch": 0.65,
"learning_rate": 0.00017562688064192577,
"loss": 2.4429,
"theoretical_loss": 3.4095187233552164,
"tokens_seen": 2153054208
},
{
"epoch": 0.65,
"learning_rate": 0.00017561685055165498,
"loss": 2.4087,
"theoretical_loss": 3.40951019542083,
"tokens_seen": 2153119744
},
{
"epoch": 0.65,
"learning_rate": 0.00017560682046138416,
"loss": 2.5664,
"theoretical_loss": 3.409501667818687,
"tokens_seen": 2153185280
},
{
"epoch": 0.65,
"learning_rate": 0.00017559679037111334,
"loss": 2.7274,
"theoretical_loss": 3.409493140548766,
"tokens_seen": 2153250816
},
{
"epoch": 0.65,
"learning_rate": 0.00017558676028084252,
"loss": 2.8435,
"theoretical_loss": 3.409484613611042,
"tokens_seen": 2153316352
},
{
"epoch": 0.65,
"learning_rate": 0.0001755767301905717,
"loss": 2.7899,
"theoretical_loss": 3.409476087005493,
"tokens_seen": 2153381888
},
{
"epoch": 0.65,
"learning_rate": 0.0001755667001003009,
"loss": 2.588,
"theoretical_loss": 3.4094675607320957,
"tokens_seen": 2153447424
},
{
"epoch": 0.65,
"learning_rate": 0.0001755566700100301,
"loss": 2.6394,
"theoretical_loss": 3.4094590347908276,
"tokens_seen": 2153512960
},
{
"epoch": 0.65,
"learning_rate": 0.00017554663991975927,
"loss": 2.5378,
"theoretical_loss": 3.4094505091816645,
"tokens_seen": 2153578496
},
{
"epoch": 0.65,
"learning_rate": 0.00017553660982948846,
"loss": 2.5994,
"theoretical_loss": 3.4094419839045846,
"tokens_seen": 2153644032
},
{
"epoch": 0.65,
"learning_rate": 0.00017552657973921766,
"loss": 2.8213,
"theoretical_loss": 3.409433458959564,
"tokens_seen": 2153709568
},
{
"epoch": 0.65,
"learning_rate": 0.00017551654964894685,
"loss": 2.3038,
"theoretical_loss": 3.40942493434658,
"tokens_seen": 2153775104
},
{
"epoch": 0.65,
"learning_rate": 0.00017550651955867603,
"loss": 2.6465,
"theoretical_loss": 3.4094164100656092,
"tokens_seen": 2153840640
},
{
"epoch": 0.65,
"learning_rate": 0.0001754964894684052,
"loss": 2.7399,
"theoretical_loss": 3.4094078861166297,
"tokens_seen": 2153906176
},
{
"epoch": 0.65,
"learning_rate": 0.0001754864593781344,
"loss": 2.4095,
"theoretical_loss": 3.4093993624996166,
"tokens_seen": 2153971712
},
{
"epoch": 0.65,
"learning_rate": 0.0001754764292878636,
"loss": 2.4577,
"theoretical_loss": 3.4093908392145487,
"tokens_seen": 2154037248
},
{
"epoch": 0.65,
"learning_rate": 0.00017546639919759278,
"loss": 2.6869,
"theoretical_loss": 3.409382316261402,
"tokens_seen": 2154102784
},
{
"epoch": 0.65,
"learning_rate": 0.00017545636910732196,
"loss": 2.3743,
"theoretical_loss": 3.4093737936401536,
"tokens_seen": 2154168320
},
{
"epoch": 0.65,
"learning_rate": 0.00017544633901705114,
"loss": 2.639,
"theoretical_loss": 3.4093652713507803,
"tokens_seen": 2154233856
},
{
"epoch": 0.65,
"learning_rate": 0.00017543630892678035,
"loss": 2.5345,
"theoretical_loss": 3.4093567493932597,
"tokens_seen": 2154299392
},
{
"epoch": 0.65,
"learning_rate": 0.00017542627883650953,
"loss": 2.7281,
"theoretical_loss": 3.4093482277675684,
"tokens_seen": 2154364928
},
{
"epoch": 0.65,
"learning_rate": 0.0001754162487462387,
"loss": 2.7373,
"theoretical_loss": 3.4093397064736832,
"tokens_seen": 2154430464
},
{
"epoch": 0.65,
"objective/train/docs_used": 1220074,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7935123443603516,
"objective/train/theoretical_loss": 3.4093311855115815,
"objective/train/tokens_used": 2174956000,
"theoretical_loss": 3.4093311855115815,
"tokens_seen": 2154496000
},
{
"epoch": 0.65,
"learning_rate": 0.0001754062186559679,
"loss": 2.8227,
"theoretical_loss": 3.4093311855115815,
"tokens_seen": 2154496000
},
{
"epoch": 0.65,
"learning_rate": 0.00017539618856569707,
"loss": 2.8463,
"theoretical_loss": 3.4093226648812403,
"tokens_seen": 2154561536
},
{
"epoch": 0.65,
"learning_rate": 0.00017538615847542628,
"loss": 2.6946,
"theoretical_loss": 3.409314144582636,
"tokens_seen": 2154627072
},
{
"epoch": 0.65,
"learning_rate": 0.00017537612838515546,
"loss": 2.6742,
"theoretical_loss": 3.4093056246157456,
"tokens_seen": 2154692608
},
{
"epoch": 0.65,
"learning_rate": 0.00017536609829488464,
"loss": 2.6839,
"theoretical_loss": 3.4092971049805474,
"tokens_seen": 2154758144
},
{
"epoch": 0.65,
"learning_rate": 0.00017535606820461382,
"loss": 2.832,
"theoretical_loss": 3.4092885856770168,
"tokens_seen": 2154823680
},
{
"epoch": 0.65,
"learning_rate": 0.00017534603811434303,
"loss": 2.5846,
"theoretical_loss": 3.4092800667051315,
"tokens_seen": 2154889216
},
{
"epoch": 0.65,
"learning_rate": 0.00017533600802407221,
"loss": 2.4917,
"theoretical_loss": 3.4092715480648685,
"tokens_seen": 2154954752
},
{
"epoch": 0.65,
"learning_rate": 0.0001753259779338014,
"loss": 2.6033,
"theoretical_loss": 3.4092630297562048,
"tokens_seen": 2155020288
},
{
"epoch": 0.65,
"learning_rate": 0.00017531594784353058,
"loss": 2.7059,
"theoretical_loss": 3.4092545117791175,
"tokens_seen": 2155085824
},
{
"epoch": 0.65,
"learning_rate": 0.00017530591775325976,
"loss": 2.6237,
"theoretical_loss": 3.4092459941335838,
"tokens_seen": 2155151360
},
{
"epoch": 0.65,
"learning_rate": 0.000175295887662989,
"loss": 2.7094,
"theoretical_loss": 3.4092374768195794,
"tokens_seen": 2155216896
},
{
"epoch": 0.65,
"learning_rate": 0.00017528585757271817,
"loss": 2.3996,
"theoretical_loss": 3.4092289598370833,
"tokens_seen": 2155282432
},
{
"epoch": 0.65,
"learning_rate": 0.00017527582748244735,
"loss": 2.7398,
"theoretical_loss": 3.4092204431860713,
"tokens_seen": 2155347968
},
{
"epoch": 0.65,
"learning_rate": 0.00017526579739217654,
"loss": 2.5095,
"theoretical_loss": 3.4092119268665204,
"tokens_seen": 2155413504
},
{
"epoch": 0.65,
"learning_rate": 0.00017525576730190574,
"loss": 2.5579,
"theoretical_loss": 3.409203410878408,
"tokens_seen": 2155479040
},
{
"epoch": 0.65,
"learning_rate": 0.00017524573721163493,
"loss": 2.7182,
"theoretical_loss": 3.4091948952217113,
"tokens_seen": 2155544576
},
{
"epoch": 0.65,
"learning_rate": 0.0001752357071213641,
"loss": 2.5531,
"theoretical_loss": 3.4091863798964064,
"tokens_seen": 2155610112
},
{
"epoch": 0.65,
"learning_rate": 0.0001752256770310933,
"loss": 2.6976,
"theoretical_loss": 3.409177864902471,
"tokens_seen": 2155675648
},
{
"epoch": 0.65,
"learning_rate": 0.00017521564694082247,
"loss": 2.6007,
"theoretical_loss": 3.4091693502398828,
"tokens_seen": 2155741184
},
{
"epoch": 0.65,
"learning_rate": 0.00017520561685055168,
"loss": 2.8097,
"theoretical_loss": 3.409160835908618,
"tokens_seen": 2155806720
},
{
"epoch": 0.65,
"learning_rate": 0.00017519558676028086,
"loss": 2.5895,
"theoretical_loss": 3.4091523219086532,
"tokens_seen": 2155872256
},
{
"epoch": 0.65,
"learning_rate": 0.00017518555667001004,
"loss": 2.8011,
"theoretical_loss": 3.4091438082399668,
"tokens_seen": 2155937792
},
{
"epoch": 0.65,
"learning_rate": 0.00017517552657973922,
"loss": 2.6325,
"theoretical_loss": 3.4091352949025344,
"tokens_seen": 2156003328
},
{
"epoch": 0.65,
"learning_rate": 0.00017516549648946843,
"loss": 2.7068,
"theoretical_loss": 3.4091267818963336,
"tokens_seen": 2156068864
},
{
"epoch": 0.65,
"objective/train/docs_used": 1221515,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4459400177001953,
"objective/train/theoretical_loss": 3.4091182692213415,
"objective/train/tokens_used": 2176594400,
"theoretical_loss": 3.4091182692213415,
"tokens_seen": 2156134400
},
{
"epoch": 0.65,
"learning_rate": 0.0001751554663991976,
"loss": 2.7167,
"theoretical_loss": 3.4091182692213415,
"tokens_seen": 2156134400
},
{
"epoch": 0.65,
"learning_rate": 0.0001751454363089268,
"loss": 2.5528,
"theoretical_loss": 3.4091097568775357,
"tokens_seen": 2156199936
},
{
"epoch": 0.65,
"learning_rate": 0.00017513540621865597,
"loss": 2.7352,
"theoretical_loss": 3.4091012448648925,
"tokens_seen": 2156265472
},
{
"epoch": 0.65,
"learning_rate": 0.00017512537612838518,
"loss": 2.5609,
"theoretical_loss": 3.409092733183389,
"tokens_seen": 2156331008
},
{
"epoch": 0.65,
"learning_rate": 0.00017511534603811436,
"loss": 2.7199,
"theoretical_loss": 3.4090842218330026,
"tokens_seen": 2156396544
},
{
"epoch": 0.65,
"learning_rate": 0.00017510531594784354,
"loss": 2.4587,
"theoretical_loss": 3.40907571081371,
"tokens_seen": 2156462080
},
{
"epoch": 0.65,
"learning_rate": 0.00017509528585757272,
"loss": 2.609,
"theoretical_loss": 3.409067200125489,
"tokens_seen": 2156527616
},
{
"epoch": 0.65,
"learning_rate": 0.0001750852557673019,
"loss": 2.5744,
"theoretical_loss": 3.4090586897683153,
"tokens_seen": 2156593152
},
{
"epoch": 0.65,
"learning_rate": 0.0001750752256770311,
"loss": 2.3137,
"theoretical_loss": 3.4090501797421675,
"tokens_seen": 2156658688
},
{
"epoch": 0.65,
"learning_rate": 0.0001750651955867603,
"loss": 2.4648,
"theoretical_loss": 3.4090416700470216,
"tokens_seen": 2156724224
},
{
"epoch": 0.65,
"learning_rate": 0.00017505516549648947,
"loss": 2.6102,
"theoretical_loss": 3.409033160682855,
"tokens_seen": 2156789760
},
{
"epoch": 0.65,
"learning_rate": 0.00017504513540621866,
"loss": 2.5978,
"theoretical_loss": 3.409024651649645,
"tokens_seen": 2156855296
},
{
"epoch": 0.65,
"learning_rate": 0.00017503510531594786,
"loss": 2.5182,
"theoretical_loss": 3.409016142947368,
"tokens_seen": 2156920832
},
{
"epoch": 0.65,
"learning_rate": 0.00017502507522567705,
"loss": 2.6285,
"theoretical_loss": 3.409007634576002,
"tokens_seen": 2156986368
},
{
"epoch": 0.65,
"learning_rate": 0.00017501504513540623,
"loss": 2.6584,
"theoretical_loss": 3.408999126535523,
"tokens_seen": 2157051904
},
{
"epoch": 0.65,
"learning_rate": 0.0001750050150451354,
"loss": 2.5951,
"theoretical_loss": 3.4089906188259094,
"tokens_seen": 2157117440
},
{
"epoch": 0.65,
"learning_rate": 0.0001749949849548646,
"loss": 2.6629,
"theoretical_loss": 3.408982111447137,
"tokens_seen": 2157182976
},
{
"epoch": 0.65,
"learning_rate": 0.0001749849548645938,
"loss": 2.7417,
"theoretical_loss": 3.4089736043991836,
"tokens_seen": 2157248512
},
{
"epoch": 0.65,
"learning_rate": 0.00017497492477432298,
"loss": 2.5277,
"theoretical_loss": 3.4089650976820263,
"tokens_seen": 2157314048
},
{
"epoch": 0.65,
"learning_rate": 0.00017496489468405216,
"loss": 2.4775,
"theoretical_loss": 3.4089565912956417,
"tokens_seen": 2157379584
},
{
"epoch": 0.65,
"learning_rate": 0.00017495486459378134,
"loss": 2.4737,
"theoretical_loss": 3.4089480852400076,
"tokens_seen": 2157445120
},
{
"epoch": 0.65,
"learning_rate": 0.00017494483450351055,
"loss": 2.6011,
"theoretical_loss": 3.408939579515101,
"tokens_seen": 2157510656
},
{
"epoch": 0.65,
"learning_rate": 0.00017493480441323973,
"loss": 2.5441,
"theoretical_loss": 3.408931074120898,
"tokens_seen": 2157576192
},
{
"epoch": 0.65,
"learning_rate": 0.0001749247743229689,
"loss": 2.9322,
"theoretical_loss": 3.4089225690573763,
"tokens_seen": 2157641728
},
{
"epoch": 0.65,
"learning_rate": 0.0001749147442326981,
"loss": 2.5175,
"theoretical_loss": 3.4089140643245135,
"tokens_seen": 2157707264
},
{
"epoch": 0.65,
"objective/train/docs_used": 1222265,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7969579696655273,
"objective/train/theoretical_loss": 3.408905559922286,
"objective/train/tokens_used": 2178232800,
"theoretical_loss": 3.408905559922286,
"tokens_seen": 2157772800
},
{
"epoch": 0.65,
"learning_rate": 0.00017490471414242727,
"loss": 2.6381,
"theoretical_loss": 3.408905559922286,
"tokens_seen": 2157772800
},
{
"epoch": 0.65,
"learning_rate": 0.00017489468405215648,
"loss": 2.7074,
"theoretical_loss": 3.4088970558506717,
"tokens_seen": 2157838336
},
{
"epoch": 0.65,
"learning_rate": 0.00017488465396188566,
"loss": 2.7728,
"theoretical_loss": 3.408888552109647,
"tokens_seen": 2157903872
},
{
"epoch": 0.65,
"learning_rate": 0.00017487462387161484,
"loss": 2.649,
"theoretical_loss": 3.408880048699189,
"tokens_seen": 2157969408
},
{
"epoch": 0.65,
"learning_rate": 0.00017486459378134402,
"loss": 2.4719,
"theoretical_loss": 3.4088715456192755,
"tokens_seen": 2158034944
},
{
"epoch": 0.65,
"learning_rate": 0.00017485456369107323,
"loss": 2.4781,
"theoretical_loss": 3.4088630428698825,
"tokens_seen": 2158100480
},
{
"epoch": 0.65,
"learning_rate": 0.00017484453360080241,
"loss": 2.6025,
"theoretical_loss": 3.4088545404509882,
"tokens_seen": 2158166016
},
{
"epoch": 0.65,
"learning_rate": 0.0001748345035105316,
"loss": 2.3528,
"theoretical_loss": 3.408846038362569,
"tokens_seen": 2158231552
},
{
"epoch": 0.65,
"learning_rate": 0.00017482447342026078,
"loss": 2.7018,
"theoretical_loss": 3.4088375366046026,
"tokens_seen": 2158297088
},
{
"epoch": 0.65,
"learning_rate": 0.00017481444332998996,
"loss": 2.6829,
"theoretical_loss": 3.4088290351770656,
"tokens_seen": 2158362624
},
{
"epoch": 0.65,
"learning_rate": 0.00017480441323971917,
"loss": 2.807,
"theoretical_loss": 3.4088205340799353,
"tokens_seen": 2158428160
},
{
"epoch": 0.65,
"learning_rate": 0.00017479438314944835,
"loss": 2.679,
"theoretical_loss": 3.4088120333131893,
"tokens_seen": 2158493696
},
{
"epoch": 0.65,
"learning_rate": 0.00017478435305917753,
"loss": 2.4306,
"theoretical_loss": 3.408803532876804,
"tokens_seen": 2158559232
},
{
"epoch": 0.65,
"learning_rate": 0.0001747743229689067,
"loss": 2.3975,
"theoretical_loss": 3.408795032770757,
"tokens_seen": 2158624768
},
{
"epoch": 0.65,
"learning_rate": 0.00017476429287863592,
"loss": 2.592,
"theoretical_loss": 3.408786532995025,
"tokens_seen": 2158690304
},
{
"epoch": 0.65,
"learning_rate": 0.0001747542627883651,
"loss": 2.5764,
"theoretical_loss": 3.408778033549585,
"tokens_seen": 2158755840
},
{
"epoch": 0.65,
"learning_rate": 0.00017474423269809428,
"loss": 2.6931,
"theoretical_loss": 3.4087695344344153,
"tokens_seen": 2158821376
},
{
"epoch": 0.65,
"learning_rate": 0.00017473420260782346,
"loss": 2.5188,
"theoretical_loss": 3.4087610356494924,
"tokens_seen": 2158886912
},
{
"epoch": 0.65,
"learning_rate": 0.00017472417251755264,
"loss": 2.7114,
"theoretical_loss": 3.408752537194793,
"tokens_seen": 2158952448
},
{
"epoch": 0.65,
"learning_rate": 0.00017471414242728185,
"loss": 2.6557,
"theoretical_loss": 3.408744039070294,
"tokens_seen": 2159017984
},
{
"epoch": 0.65,
"learning_rate": 0.00017470411233701103,
"loss": 2.7809,
"theoretical_loss": 3.408735541275974,
"tokens_seen": 2159083520
},
{
"epoch": 0.65,
"learning_rate": 0.0001746940822467402,
"loss": 2.5513,
"theoretical_loss": 3.408727043811809,
"tokens_seen": 2159149056
},
{
"epoch": 0.65,
"learning_rate": 0.0001746840521564694,
"loss": 2.7633,
"theoretical_loss": 3.4087185466777763,
"tokens_seen": 2159214592
},
{
"epoch": 0.65,
"learning_rate": 0.0001746740220661986,
"loss": 2.4475,
"theoretical_loss": 3.4087100498738536,
"tokens_seen": 2159280128
},
{
"epoch": 0.65,
"learning_rate": 0.00017466399197592778,
"loss": 2.4614,
"theoretical_loss": 3.408701553400017,
"tokens_seen": 2159345664
},
{
"epoch": 0.65,
"objective/train/docs_used": 1223379,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.288496971130371,
"objective/train/theoretical_loss": 3.4086930572562446,
"objective/train/tokens_used": 2179871200,
"theoretical_loss": 3.4086930572562446,
"tokens_seen": 2159411200
},
{
"epoch": 0.65,
"learning_rate": 0.00017465396188565696,
"loss": 2.309,
"theoretical_loss": 3.4086930572562446,
"tokens_seen": 2159411200
},
{
"epoch": 0.65,
"learning_rate": 0.00017464393179538614,
"loss": 2.7282,
"theoretical_loss": 3.4086845614425134,
"tokens_seen": 2159476736
},
{
"epoch": 0.65,
"learning_rate": 0.00017463390170511533,
"loss": 2.6857,
"theoretical_loss": 3.4086760659588,
"tokens_seen": 2159542272
},
{
"epoch": 0.65,
"learning_rate": 0.00017462387161484453,
"loss": 2.7225,
"theoretical_loss": 3.408667570805082,
"tokens_seen": 2159607808
},
{
"epoch": 0.65,
"learning_rate": 0.00017461384152457372,
"loss": 2.6502,
"theoretical_loss": 3.408659075981337,
"tokens_seen": 2159673344
},
{
"epoch": 0.65,
"learning_rate": 0.0001746038114343029,
"loss": 2.5178,
"theoretical_loss": 3.4086505814875414,
"tokens_seen": 2159738880
},
{
"epoch": 0.65,
"learning_rate": 0.00017459378134403208,
"loss": 2.5917,
"theoretical_loss": 3.408642087323673,
"tokens_seen": 2159804416
},
{
"epoch": 0.65,
"learning_rate": 0.00017458375125376129,
"loss": 2.6775,
"theoretical_loss": 3.4086335934897085,
"tokens_seen": 2159869952
},
{
"epoch": 0.65,
"learning_rate": 0.00017457372116349047,
"loss": 2.7077,
"theoretical_loss": 3.4086250999856254,
"tokens_seen": 2159935488
},
{
"epoch": 0.65,
"learning_rate": 0.00017456369107321965,
"loss": 2.5795,
"theoretical_loss": 3.4086166068114006,
"tokens_seen": 2160001024
},
{
"epoch": 0.65,
"learning_rate": 0.00017455366098294883,
"loss": 2.2853,
"theoretical_loss": 3.408608113967011,
"tokens_seen": 2160066560
},
{
"epoch": 0.65,
"learning_rate": 0.00017454363089267806,
"loss": 2.6878,
"theoretical_loss": 3.408599621452435,
"tokens_seen": 2160132096
},
{
"epoch": 0.65,
"learning_rate": 0.00017453360080240725,
"loss": 2.5471,
"theoretical_loss": 3.4085911292676485,
"tokens_seen": 2160197632
},
{
"epoch": 0.65,
"learning_rate": 0.00017452357071213643,
"loss": 2.6298,
"theoretical_loss": 3.4085826374126293,
"tokens_seen": 2160263168
},
{
"epoch": 0.65,
"learning_rate": 0.0001745135406218656,
"loss": 2.602,
"theoretical_loss": 3.4085741458873544,
"tokens_seen": 2160328704
},
{
"epoch": 0.65,
"learning_rate": 0.0001745035105315948,
"loss": 2.5208,
"theoretical_loss": 3.408565654691801,
"tokens_seen": 2160394240
},
{
"epoch": 0.65,
"learning_rate": 0.000174493480441324,
"loss": 2.8559,
"theoretical_loss": 3.408557163825946,
"tokens_seen": 2160459776
},
{
"epoch": 0.65,
"learning_rate": 0.00017448345035105318,
"loss": 2.6837,
"theoretical_loss": 3.4085486732897676,
"tokens_seen": 2160525312
},
{
"epoch": 0.65,
"learning_rate": 0.00017447342026078236,
"loss": 2.49,
"theoretical_loss": 3.4085401830832422,
"tokens_seen": 2160590848
},
{
"epoch": 0.65,
"learning_rate": 0.00017446339017051154,
"loss": 2.2599,
"theoretical_loss": 3.408531693206347,
"tokens_seen": 2160656384
},
{
"epoch": 0.65,
"learning_rate": 0.00017445336008024075,
"loss": 2.4981,
"theoretical_loss": 3.408523203659059,
"tokens_seen": 2160721920
},
{
"epoch": 0.65,
"learning_rate": 0.00017444332998996993,
"loss": 2.5268,
"theoretical_loss": 3.4085147144413566,
"tokens_seen": 2160787456
},
{
"epoch": 0.65,
"learning_rate": 0.0001744332998996991,
"loss": 2.6788,
"theoretical_loss": 3.4085062255532157,
"tokens_seen": 2160852992
},
{
"epoch": 0.65,
"learning_rate": 0.0001744232698094283,
"loss": 2.67,
"theoretical_loss": 3.408497736994614,
"tokens_seen": 2160918528
},
{
"epoch": 0.65,
"learning_rate": 0.00017441323971915747,
"loss": 2.451,
"theoretical_loss": 3.4084892487655285,
"tokens_seen": 2160984064
},
{
"epoch": 0.65,
"objective/train/docs_used": 1223894,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.681525707244873,
"objective/train/theoretical_loss": 3.408480760865937,
"objective/train/tokens_used": 2181509600,
"theoretical_loss": 3.408480760865937,
"tokens_seen": 2161049600
},
{
"epoch": 0.65,
"learning_rate": 0.00017440320962888668,
"loss": 2.7461,
"theoretical_loss": 3.408480760865937,
"tokens_seen": 2161049600
},
{
"epoch": 0.65,
"learning_rate": 0.00017439317953861586,
"loss": 2.6539,
"theoretical_loss": 3.408472273295816,
"tokens_seen": 2161115136
},
{
"epoch": 0.65,
"learning_rate": 0.00017438314944834504,
"loss": 2.5566,
"theoretical_loss": 3.408463786055143,
"tokens_seen": 2161180672
},
{
"epoch": 0.65,
"learning_rate": 0.00017437311935807422,
"loss": 2.5518,
"theoretical_loss": 3.408455299143896,
"tokens_seen": 2161246208
},
{
"epoch": 0.65,
"learning_rate": 0.00017436308926780343,
"loss": 2.6799,
"theoretical_loss": 3.408446812562051,
"tokens_seen": 2161311744
},
{
"epoch": 0.65,
"learning_rate": 0.00017435305917753261,
"loss": 2.6267,
"theoretical_loss": 3.4084383263095854,
"tokens_seen": 2161377280
},
{
"epoch": 0.65,
"learning_rate": 0.0001743430290872618,
"loss": 2.6139,
"theoretical_loss": 3.4084298403864772,
"tokens_seen": 2161442816
},
{
"epoch": 0.66,
"learning_rate": 0.00017433299899699098,
"loss": 2.686,
"theoretical_loss": 3.408421354792703,
"tokens_seen": 2161508352
},
{
"epoch": 0.66,
"learning_rate": 0.00017432296890672016,
"loss": 2.6141,
"theoretical_loss": 3.4084128695282403,
"tokens_seen": 2161573888
},
{
"epoch": 0.66,
"learning_rate": 0.00017431293881644937,
"loss": 2.466,
"theoretical_loss": 3.408404384593066,
"tokens_seen": 2161639424
},
{
"epoch": 0.66,
"learning_rate": 0.00017430290872617855,
"loss": 2.7223,
"theoretical_loss": 3.4083958999871578,
"tokens_seen": 2161704960
},
{
"epoch": 0.66,
"learning_rate": 0.00017429287863590773,
"loss": 2.3784,
"theoretical_loss": 3.408387415710493,
"tokens_seen": 2161770496
},
{
"epoch": 0.66,
"learning_rate": 0.0001742828485456369,
"loss": 2.5114,
"theoretical_loss": 3.408378931763048,
"tokens_seen": 2161836032
},
{
"epoch": 0.66,
"learning_rate": 0.00017427281845536612,
"loss": 2.439,
"theoretical_loss": 3.4083704481448014,
"tokens_seen": 2161901568
},
{
"epoch": 0.66,
"learning_rate": 0.0001742627883650953,
"loss": 2.6521,
"theoretical_loss": 3.408361964855729,
"tokens_seen": 2161967104
},
{
"epoch": 0.66,
"learning_rate": 0.00017425275827482448,
"loss": 2.6587,
"theoretical_loss": 3.408353481895809,
"tokens_seen": 2162032640
},
{
"epoch": 0.66,
"learning_rate": 0.00017424272818455366,
"loss": 2.8271,
"theoretical_loss": 3.4083449992650183,
"tokens_seen": 2162098176
},
{
"epoch": 0.66,
"learning_rate": 0.00017423269809428284,
"loss": 2.6869,
"theoretical_loss": 3.408336516963334,
"tokens_seen": 2162163712
},
{
"epoch": 0.66,
"learning_rate": 0.00017422266800401205,
"loss": 2.3946,
"theoretical_loss": 3.408328034990734,
"tokens_seen": 2162229248
},
{
"epoch": 0.66,
"learning_rate": 0.00017421263791374123,
"loss": 2.5485,
"theoretical_loss": 3.4083195533471953,
"tokens_seen": 2162294784
},
{
"epoch": 0.66,
"learning_rate": 0.0001742026078234704,
"loss": 2.5581,
"theoretical_loss": 3.4083110720326943,
"tokens_seen": 2162360320
},
{
"epoch": 0.66,
"learning_rate": 0.0001741925777331996,
"loss": 2.8213,
"theoretical_loss": 3.4083025910472093,
"tokens_seen": 2162425856
},
{
"epoch": 0.66,
"learning_rate": 0.0001741825476429288,
"loss": 2.8712,
"theoretical_loss": 3.4082941103907176,
"tokens_seen": 2162491392
},
{
"epoch": 0.66,
"learning_rate": 0.00017417251755265798,
"loss": 2.4918,
"theoretical_loss": 3.4082856300631956,
"tokens_seen": 2162556928
},
{
"epoch": 0.66,
"learning_rate": 0.00017416248746238716,
"loss": 2.7453,
"theoretical_loss": 3.408277150064621,
"tokens_seen": 2162622464
},
{
"epoch": 0.66,
"objective/train/docs_used": 1225280,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.330841064453125,
"objective/train/theoretical_loss": 3.4082686703949716,
"objective/train/tokens_used": 2183148000,
"theoretical_loss": 3.4082686703949716,
"tokens_seen": 2162688000
},
{
"epoch": 0.66,
"learning_rate": 0.00017415245737211634,
"loss": 2.7278,
"theoretical_loss": 3.4082686703949716,
"tokens_seen": 2162688000
},
{
"epoch": 0.66,
"learning_rate": 0.00017414242728184553,
"loss": 2.6722,
"theoretical_loss": 3.4082601910542243,
"tokens_seen": 2162753536
},
{
"epoch": 0.66,
"learning_rate": 0.00017413239719157473,
"loss": 2.5308,
"theoretical_loss": 3.4082517120423557,
"tokens_seen": 2162819072
},
{
"epoch": 0.66,
"learning_rate": 0.00017412236710130392,
"loss": 2.5612,
"theoretical_loss": 3.408243233359344,
"tokens_seen": 2162884608
},
{
"epoch": 0.66,
"learning_rate": 0.0001741123370110331,
"loss": 2.4832,
"theoretical_loss": 3.408234755005166,
"tokens_seen": 2162950144
},
{
"epoch": 0.66,
"learning_rate": 0.00017410230692076228,
"loss": 2.6858,
"theoretical_loss": 3.4082262769797995,
"tokens_seen": 2163015680
},
{
"epoch": 0.66,
"learning_rate": 0.00017409227683049149,
"loss": 2.7547,
"theoretical_loss": 3.4082177992832214,
"tokens_seen": 2163081216
},
{
"epoch": 0.66,
"learning_rate": 0.00017408224674022067,
"loss": 2.5458,
"theoretical_loss": 3.408209321915409,
"tokens_seen": 2163146752
},
{
"epoch": 0.66,
"learning_rate": 0.00017407221664994985,
"loss": 2.647,
"theoretical_loss": 3.4082008448763395,
"tokens_seen": 2163212288
},
{
"epoch": 0.66,
"learning_rate": 0.00017406218655967903,
"loss": 2.48,
"theoretical_loss": 3.40819236816599,
"tokens_seen": 2163277824
},
{
"epoch": 0.66,
"learning_rate": 0.0001740521564694082,
"loss": 2.5512,
"theoretical_loss": 3.408183891784338,
"tokens_seen": 2163343360
},
{
"epoch": 0.66,
"learning_rate": 0.00017404212637913742,
"loss": 2.296,
"theoretical_loss": 3.4081754157313617,
"tokens_seen": 2163408896
},
{
"epoch": 0.66,
"learning_rate": 0.0001740320962888666,
"loss": 2.5254,
"theoretical_loss": 3.408166940007037,
"tokens_seen": 2163474432
},
{
"epoch": 0.66,
"learning_rate": 0.00017402206619859578,
"loss": 2.4113,
"theoretical_loss": 3.408158464611342,
"tokens_seen": 2163539968
},
{
"epoch": 0.66,
"learning_rate": 0.00017401203610832496,
"loss": 2.6528,
"theoretical_loss": 3.4081499895442535,
"tokens_seen": 2163605504
},
{
"epoch": 0.66,
"learning_rate": 0.00017400200601805417,
"loss": 2.5469,
"theoretical_loss": 3.4081415148057497,
"tokens_seen": 2163671040
},
{
"epoch": 0.66,
"learning_rate": 0.00017399197592778335,
"loss": 2.6639,
"theoretical_loss": 3.4081330403958066,
"tokens_seen": 2163736576
},
{
"epoch": 0.66,
"learning_rate": 0.00017398194583751253,
"loss": 2.5738,
"theoretical_loss": 3.4081245663144024,
"tokens_seen": 2163802112
},
{
"epoch": 0.66,
"learning_rate": 0.00017397191574724171,
"loss": 2.6919,
"theoretical_loss": 3.4081160925615146,
"tokens_seen": 2163867648
},
{
"epoch": 0.66,
"learning_rate": 0.00017396188565697092,
"loss": 2.5313,
"theoretical_loss": 3.40810761913712,
"tokens_seen": 2163933184
},
{
"epoch": 0.66,
"learning_rate": 0.0001739518555667001,
"loss": 2.6881,
"theoretical_loss": 3.408099146041196,
"tokens_seen": 2163998720
},
{
"epoch": 0.66,
"learning_rate": 0.00017394182547642928,
"loss": 2.4776,
"theoretical_loss": 3.40809067327372,
"tokens_seen": 2164064256
},
{
"epoch": 0.66,
"learning_rate": 0.00017393179538615847,
"loss": 2.6251,
"theoretical_loss": 3.408082200834669,
"tokens_seen": 2164129792
},
{
"epoch": 0.66,
"learning_rate": 0.00017392176529588765,
"loss": 2.6254,
"theoretical_loss": 3.408073728724021,
"tokens_seen": 2164195328
},
{
"epoch": 0.66,
"learning_rate": 0.00017391173520561685,
"loss": 2.7914,
"theoretical_loss": 3.408065256941753,
"tokens_seen": 2164260864
},
{
"epoch": 0.66,
"objective/train/docs_used": 1225734,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3772361278533936,
"objective/train/theoretical_loss": 3.408056785487842,
"objective/train/tokens_used": 2184786400,
"theoretical_loss": 3.408056785487842,
"tokens_seen": 2164326400
},
{
"epoch": 0.66,
"learning_rate": 0.00017390170511534604,
"loss": 2.4467,
"theoretical_loss": 3.408056785487842,
"tokens_seen": 2164326400
},
{
"epoch": 0.66,
"learning_rate": 0.00017389167502507522,
"loss": 2.9082,
"theoretical_loss": 3.408048314362266,
"tokens_seen": 2164391936
},
{
"epoch": 0.66,
"learning_rate": 0.0001738816449348044,
"loss": 2.6417,
"theoretical_loss": 3.408039843565002,
"tokens_seen": 2164457472
},
{
"epoch": 0.66,
"learning_rate": 0.0001738716148445336,
"loss": 2.5599,
"theoretical_loss": 3.4080313730960268,
"tokens_seen": 2164523008
},
{
"epoch": 0.66,
"learning_rate": 0.0001738615847542628,
"loss": 2.6552,
"theoretical_loss": 3.4080229029553184,
"tokens_seen": 2164588544
},
{
"epoch": 0.66,
"learning_rate": 0.00017385155466399197,
"loss": 2.466,
"theoretical_loss": 3.408014433142854,
"tokens_seen": 2164654080
},
{
"epoch": 0.66,
"learning_rate": 0.00017384152457372115,
"loss": 2.5175,
"theoretical_loss": 3.408005963658611,
"tokens_seen": 2164719616
},
{
"epoch": 0.66,
"learning_rate": 0.00017383149448345033,
"loss": 2.6701,
"theoretical_loss": 3.407997494502567,
"tokens_seen": 2164785152
},
{
"epoch": 0.66,
"learning_rate": 0.00017382146439317954,
"loss": 2.6045,
"theoretical_loss": 3.4079890256746985,
"tokens_seen": 2164850688
},
{
"epoch": 0.66,
"learning_rate": 0.00017381143430290872,
"loss": 2.6151,
"theoretical_loss": 3.4079805571749833,
"tokens_seen": 2164916224
},
{
"epoch": 0.66,
"learning_rate": 0.0001738014042126379,
"loss": 2.742,
"theoretical_loss": 3.4079720890033993,
"tokens_seen": 2164981760
},
{
"epoch": 0.66,
"learning_rate": 0.0001737913741223671,
"loss": 2.7502,
"theoretical_loss": 3.4079636211599227,
"tokens_seen": 2165047296
},
{
"epoch": 0.66,
"learning_rate": 0.00017378134403209632,
"loss": 2.4655,
"theoretical_loss": 3.407955153644532,
"tokens_seen": 2165112832
},
{
"epoch": 0.66,
"learning_rate": 0.0001737713139418255,
"loss": 2.3148,
"theoretical_loss": 3.407946686457204,
"tokens_seen": 2165178368
},
{
"epoch": 0.66,
"learning_rate": 0.00017376128385155468,
"loss": 2.7476,
"theoretical_loss": 3.407938219597916,
"tokens_seen": 2165243904
},
{
"epoch": 0.66,
"learning_rate": 0.00017375125376128386,
"loss": 2.5391,
"theoretical_loss": 3.4079297530666457,
"tokens_seen": 2165309440
},
{
"epoch": 0.66,
"learning_rate": 0.00017374122367101304,
"loss": 2.7471,
"theoretical_loss": 3.40792128686337,
"tokens_seen": 2165374976
},
{
"epoch": 0.66,
"learning_rate": 0.00017373119358074225,
"loss": 2.493,
"theoretical_loss": 3.4079128209880665,
"tokens_seen": 2165440512
},
{
"epoch": 0.66,
"learning_rate": 0.00017372116349047143,
"loss": 2.8309,
"theoretical_loss": 3.4079043554407127,
"tokens_seen": 2165506048
},
{
"epoch": 0.66,
"learning_rate": 0.0001737111334002006,
"loss": 2.7088,
"theoretical_loss": 3.407895890221286,
"tokens_seen": 2165571584
},
{
"epoch": 0.66,
"learning_rate": 0.0001737011033099298,
"loss": 2.7865,
"theoretical_loss": 3.4078874253297635,
"tokens_seen": 2165637120
},
{
"epoch": 0.66,
"learning_rate": 0.000173691073219659,
"loss": 2.3726,
"theoretical_loss": 3.407878960766123,
"tokens_seen": 2165702656
},
{
"epoch": 0.66,
"learning_rate": 0.00017368104312938818,
"loss": 2.7729,
"theoretical_loss": 3.407870496530341,
"tokens_seen": 2165768192
},
{
"epoch": 0.66,
"learning_rate": 0.00017367101303911736,
"loss": 2.7049,
"theoretical_loss": 3.4078620326223956,
"tokens_seen": 2165833728
},
{
"epoch": 0.66,
"learning_rate": 0.00017366098294884655,
"loss": 2.4814,
"theoretical_loss": 3.407853569042264,
"tokens_seen": 2165899264
},
{
"epoch": 0.66,
"objective/train/docs_used": 1226526,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.572643280029297,
"objective/train/theoretical_loss": 3.4078451057899244,
"objective/train/tokens_used": 2186424800,
"theoretical_loss": 3.4078451057899244,
"tokens_seen": 2165964800
},
{
"epoch": 0.66,
"learning_rate": 0.00017365095285857573,
"loss": 2.6467,
"theoretical_loss": 3.4078451057899244,
"tokens_seen": 2165964800
},
{
"epoch": 0.66,
"learning_rate": 0.00017364092276830493,
"loss": 2.7739,
"theoretical_loss": 3.4078366428653526,
"tokens_seen": 2166030336
},
{
"epoch": 0.66,
"learning_rate": 0.00017363089267803412,
"loss": 2.6836,
"theoretical_loss": 3.4078281802685275,
"tokens_seen": 2166095872
},
{
"epoch": 0.66,
"learning_rate": 0.0001736208625877633,
"loss": 2.7237,
"theoretical_loss": 3.407819717999425,
"tokens_seen": 2166161408
},
{
"epoch": 0.66,
"learning_rate": 0.00017361083249749248,
"loss": 2.7349,
"theoretical_loss": 3.4078112560580243,
"tokens_seen": 2166226944
},
{
"epoch": 0.66,
"learning_rate": 0.00017360080240722169,
"loss": 2.353,
"theoretical_loss": 3.4078027944443012,
"tokens_seen": 2166292480
},
{
"epoch": 0.66,
"learning_rate": 0.00017359077231695087,
"loss": 2.562,
"theoretical_loss": 3.407794333158234,
"tokens_seen": 2166358016
},
{
"epoch": 0.66,
"learning_rate": 0.00017358074222668005,
"loss": 2.5926,
"theoretical_loss": 3.407785872199799,
"tokens_seen": 2166423552
},
{
"epoch": 0.66,
"learning_rate": 0.00017357071213640923,
"loss": 2.6463,
"theoretical_loss": 3.407777411568975,
"tokens_seen": 2166489088
},
{
"epoch": 0.66,
"learning_rate": 0.0001735606820461384,
"loss": 2.8183,
"theoretical_loss": 3.407768951265739,
"tokens_seen": 2166554624
},
{
"epoch": 0.66,
"learning_rate": 0.00017355065195586762,
"loss": 2.6016,
"theoretical_loss": 3.407760491290068,
"tokens_seen": 2166620160
},
{
"epoch": 0.66,
"learning_rate": 0.0001735406218655968,
"loss": 2.7151,
"theoretical_loss": 3.4077520316419396,
"tokens_seen": 2166685696
},
{
"epoch": 0.66,
"learning_rate": 0.00017353059177532598,
"loss": 2.8535,
"theoretical_loss": 3.4077435723213316,
"tokens_seen": 2166751232
},
{
"epoch": 0.66,
"learning_rate": 0.00017352056168505516,
"loss": 2.4965,
"theoretical_loss": 3.4077351133282203,
"tokens_seen": 2166816768
},
{
"epoch": 0.66,
"learning_rate": 0.00017351053159478437,
"loss": 2.6886,
"theoretical_loss": 3.407726654662585,
"tokens_seen": 2166882304
},
{
"epoch": 0.66,
"learning_rate": 0.00017350050150451355,
"loss": 2.9618,
"theoretical_loss": 3.4077181963244008,
"tokens_seen": 2166947840
},
{
"epoch": 0.66,
"learning_rate": 0.00017349047141424273,
"loss": 2.4524,
"theoretical_loss": 3.4077097383136468,
"tokens_seen": 2167013376
},
{
"epoch": 0.66,
"learning_rate": 0.00017348044132397191,
"loss": 2.5297,
"theoretical_loss": 3.4077012806302998,
"tokens_seen": 2167078912
},
{
"epoch": 0.66,
"learning_rate": 0.00017347041123370112,
"loss": 2.8488,
"theoretical_loss": 3.4076928232743375,
"tokens_seen": 2167144448
},
{
"epoch": 0.66,
"learning_rate": 0.0001734603811434303,
"loss": 2.5812,
"theoretical_loss": 3.4076843662457375,
"tokens_seen": 2167209984
},
{
"epoch": 0.66,
"learning_rate": 0.00017345035105315948,
"loss": 2.6685,
"theoretical_loss": 3.4076759095444764,
"tokens_seen": 2167275520
},
{
"epoch": 0.66,
"learning_rate": 0.00017344032096288867,
"loss": 2.6731,
"theoretical_loss": 3.407667453170532,
"tokens_seen": 2167341056
},
{
"epoch": 0.66,
"learning_rate": 0.00017343029087261785,
"loss": 2.6166,
"theoretical_loss": 3.4076589971238826,
"tokens_seen": 2167406592
},
{
"epoch": 0.66,
"learning_rate": 0.00017342026078234705,
"loss": 2.6356,
"theoretical_loss": 3.4076505414045046,
"tokens_seen": 2167472128
},
{
"epoch": 0.66,
"learning_rate": 0.00017341023069207624,
"loss": 2.6163,
"theoretical_loss": 3.4076420860123755,
"tokens_seen": 2167537664
},
{
"epoch": 0.66,
"objective/train/docs_used": 1227923,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4693679809570312,
"objective/train/theoretical_loss": 3.4076336309474735,
"objective/train/tokens_used": 2188063200,
"theoretical_loss": 3.4076336309474735,
"tokens_seen": 2167603200
},
{
"epoch": 0.66,
"learning_rate": 0.00017340020060180542,
"loss": 2.4881,
"theoretical_loss": 3.4076336309474735,
"tokens_seen": 2167603200
},
{
"epoch": 0.66,
"learning_rate": 0.0001733901705115346,
"loss": 2.7753,
"theoretical_loss": 3.407625176209775,
"tokens_seen": 2167668736
},
{
"epoch": 0.66,
"learning_rate": 0.0001733801404212638,
"loss": 2.6787,
"theoretical_loss": 3.407616721799258,
"tokens_seen": 2167734272
},
{
"epoch": 0.66,
"learning_rate": 0.000173370110330993,
"loss": 2.3901,
"theoretical_loss": 3.4076082677159003,
"tokens_seen": 2167799808
},
{
"epoch": 0.66,
"learning_rate": 0.00017336008024072217,
"loss": 2.7623,
"theoretical_loss": 3.4075998139596786,
"tokens_seen": 2167865344
},
{
"epoch": 0.66,
"learning_rate": 0.00017335005015045135,
"loss": 2.7018,
"theoretical_loss": 3.407591360530571,
"tokens_seen": 2167930880
},
{
"epoch": 0.66,
"learning_rate": 0.00017334002006018053,
"loss": 2.7033,
"theoretical_loss": 3.4075829074285546,
"tokens_seen": 2167996416
},
{
"epoch": 0.66,
"learning_rate": 0.00017332998996990974,
"loss": 2.6407,
"theoretical_loss": 3.4075744546536075,
"tokens_seen": 2168061952
},
{
"epoch": 0.66,
"learning_rate": 0.00017331995987963892,
"loss": 2.6276,
"theoretical_loss": 3.407566002205706,
"tokens_seen": 2168127488
},
{
"epoch": 0.66,
"learning_rate": 0.0001733099297893681,
"loss": 2.6601,
"theoretical_loss": 3.407557550084828,
"tokens_seen": 2168193024
},
{
"epoch": 0.66,
"learning_rate": 0.00017329989969909728,
"loss": 2.6138,
"theoretical_loss": 3.4075490982909518,
"tokens_seen": 2168258560
},
{
"epoch": 0.66,
"learning_rate": 0.0001732898696088265,
"loss": 2.7069,
"theoretical_loss": 3.4075406468240534,
"tokens_seen": 2168324096
},
{
"epoch": 0.66,
"learning_rate": 0.00017327983951855567,
"loss": 2.757,
"theoretical_loss": 3.4075321956841114,
"tokens_seen": 2168389632
},
{
"epoch": 0.66,
"learning_rate": 0.00017326980942828485,
"loss": 2.5348,
"theoretical_loss": 3.4075237448711033,
"tokens_seen": 2168455168
},
{
"epoch": 0.66,
"learning_rate": 0.00017325977933801403,
"loss": 2.6751,
"theoretical_loss": 3.4075152943850058,
"tokens_seen": 2168520704
},
{
"epoch": 0.66,
"learning_rate": 0.00017324974924774322,
"loss": 2.7038,
"theoretical_loss": 3.407506844225797,
"tokens_seen": 2168586240
},
{
"epoch": 0.66,
"learning_rate": 0.00017323971915747242,
"loss": 2.7503,
"theoretical_loss": 3.4074983943934543,
"tokens_seen": 2168651776
},
{
"epoch": 0.66,
"learning_rate": 0.0001732296890672016,
"loss": 2.8206,
"theoretical_loss": 3.4074899448879545,
"tokens_seen": 2168717312
},
{
"epoch": 0.66,
"learning_rate": 0.00017321965897693079,
"loss": 2.5905,
"theoretical_loss": 3.407481495709276,
"tokens_seen": 2168782848
},
{
"epoch": 0.66,
"learning_rate": 0.00017320962888665997,
"loss": 2.5826,
"theoretical_loss": 3.4074730468573957,
"tokens_seen": 2168848384
},
{
"epoch": 0.66,
"learning_rate": 0.00017319959879638917,
"loss": 2.6345,
"theoretical_loss": 3.4074645983322913,
"tokens_seen": 2168913920
},
{
"epoch": 0.66,
"learning_rate": 0.00017318956870611836,
"loss": 2.6321,
"theoretical_loss": 3.4074561501339407,
"tokens_seen": 2168979456
},
{
"epoch": 0.66,
"learning_rate": 0.00017317953861584754,
"loss": 2.6559,
"theoretical_loss": 3.4074477022623206,
"tokens_seen": 2169044992
},
{
"epoch": 0.66,
"learning_rate": 0.00017316950852557672,
"loss": 2.602,
"theoretical_loss": 3.407439254717409,
"tokens_seen": 2169110528
},
{
"epoch": 0.66,
"learning_rate": 0.0001731594784353059,
"loss": 2.3586,
"theoretical_loss": 3.407430807499183,
"tokens_seen": 2169176064
},
{
"epoch": 0.66,
"objective/train/docs_used": 1228616,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.481062889099121,
"objective/train/theoretical_loss": 3.4074223606076206,
"objective/train/tokens_used": 2189701600,
"theoretical_loss": 3.4074223606076206,
"tokens_seen": 2169241600
},
{
"epoch": 0.66,
"learning_rate": 0.0001731494483450351,
"loss": 2.5341,
"theoretical_loss": 3.4074223606076206,
"tokens_seen": 2169241600
},
{
"epoch": 0.66,
"learning_rate": 0.0001731394182547643,
"loss": 2.5066,
"theoretical_loss": 3.407413914042699,
"tokens_seen": 2169307136
},
{
"epoch": 0.66,
"learning_rate": 0.00017312938816449347,
"loss": 2.7129,
"theoretical_loss": 3.407405467804396,
"tokens_seen": 2169372672
},
{
"epoch": 0.66,
"learning_rate": 0.00017311935807422265,
"loss": 2.5353,
"theoretical_loss": 3.407397021892688,
"tokens_seen": 2169438208
},
{
"epoch": 0.66,
"learning_rate": 0.00017310932798395186,
"loss": 2.6978,
"theoretical_loss": 3.407388576307554,
"tokens_seen": 2169503744
},
{
"epoch": 0.66,
"learning_rate": 0.00017309929789368104,
"loss": 2.571,
"theoretical_loss": 3.407380131048971,
"tokens_seen": 2169569280
},
{
"epoch": 0.66,
"learning_rate": 0.00017308926780341022,
"loss": 2.6138,
"theoretical_loss": 3.407371686116916,
"tokens_seen": 2169634816
},
{
"epoch": 0.66,
"learning_rate": 0.0001730792377131394,
"loss": 2.4737,
"theoretical_loss": 3.4073632415113675,
"tokens_seen": 2169700352
},
{
"epoch": 0.66,
"learning_rate": 0.00017306920762286858,
"loss": 2.6237,
"theoretical_loss": 3.4073547972323017,
"tokens_seen": 2169765888
},
{
"epoch": 0.66,
"learning_rate": 0.0001730591775325978,
"loss": 2.5801,
"theoretical_loss": 3.407346353279697,
"tokens_seen": 2169831424
},
{
"epoch": 0.66,
"learning_rate": 0.00017304914744232697,
"loss": 2.6146,
"theoretical_loss": 3.4073379096535312,
"tokens_seen": 2169896960
},
{
"epoch": 0.66,
"learning_rate": 0.00017303911735205618,
"loss": 2.6322,
"theoretical_loss": 3.4073294663537808,
"tokens_seen": 2169962496
},
{
"epoch": 0.66,
"learning_rate": 0.00017302908726178536,
"loss": 2.9724,
"theoretical_loss": 3.4073210233804243,
"tokens_seen": 2170028032
},
{
"epoch": 0.66,
"learning_rate": 0.00017301905717151457,
"loss": 2.69,
"theoretical_loss": 3.4073125807334383,
"tokens_seen": 2170093568
},
{
"epoch": 0.66,
"learning_rate": 0.00017300902708124375,
"loss": 2.682,
"theoretical_loss": 3.407304138412801,
"tokens_seen": 2170159104
},
{
"epoch": 0.66,
"learning_rate": 0.00017299899699097293,
"loss": 2.7437,
"theoretical_loss": 3.4072956964184904,
"tokens_seen": 2170224640
},
{
"epoch": 0.66,
"learning_rate": 0.00017298896690070211,
"loss": 2.6752,
"theoretical_loss": 3.4072872547504827,
"tokens_seen": 2170290176
},
{
"epoch": 0.66,
"learning_rate": 0.00017297893681043132,
"loss": 2.7654,
"theoretical_loss": 3.4072788134087566,
"tokens_seen": 2170355712
},
{
"epoch": 0.66,
"learning_rate": 0.0001729689067201605,
"loss": 2.6417,
"theoretical_loss": 3.407270372393289,
"tokens_seen": 2170421248
},
{
"epoch": 0.66,
"learning_rate": 0.00017295887662988968,
"loss": 2.6036,
"theoretical_loss": 3.4072619317040576,
"tokens_seen": 2170486784
},
{
"epoch": 0.66,
"learning_rate": 0.00017294884653961887,
"loss": 2.6387,
"theoretical_loss": 3.4072534913410397,
"tokens_seen": 2170552320
},
{
"epoch": 0.66,
"learning_rate": 0.00017293881644934805,
"loss": 2.7027,
"theoretical_loss": 3.4072450513042134,
"tokens_seen": 2170617856
},
{
"epoch": 0.66,
"learning_rate": 0.00017292878635907725,
"loss": 2.4306,
"theoretical_loss": 3.407236611593556,
"tokens_seen": 2170683392
},
{
"epoch": 0.66,
"learning_rate": 0.00017291875626880644,
"loss": 2.6412,
"theoretical_loss": 3.4072281722090447,
"tokens_seen": 2170748928
},
{
"epoch": 0.66,
"learning_rate": 0.00017290872617853562,
"loss": 2.5513,
"theoretical_loss": 3.4072197331506575,
"tokens_seen": 2170814464
},
{
"epoch": 0.66,
"objective/train/docs_used": 1230033,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.610170364379883,
"objective/train/theoretical_loss": 3.407211294418372,
"objective/train/tokens_used": 2191340000,
"theoretical_loss": 3.407211294418372,
"tokens_seen": 2170880000
},
{
"epoch": 0.66,
"learning_rate": 0.0001728986960882648,
"loss": 2.3675,
"theoretical_loss": 3.407211294418372,
"tokens_seen": 2170880000
},
{
"epoch": 0.66,
"learning_rate": 0.000172888665997994,
"loss": 2.5536,
"theoretical_loss": 3.4072028560121654,
"tokens_seen": 2170945536
},
{
"epoch": 0.66,
"learning_rate": 0.0001728786359077232,
"loss": 2.4932,
"theoretical_loss": 3.4071944179320153,
"tokens_seen": 2171011072
},
{
"epoch": 0.66,
"learning_rate": 0.00017286860581745237,
"loss": 2.5775,
"theoretical_loss": 3.4071859801778994,
"tokens_seen": 2171076608
},
{
"epoch": 0.66,
"learning_rate": 0.00017285857572718155,
"loss": 2.6368,
"theoretical_loss": 3.4071775427497952,
"tokens_seen": 2171142144
},
{
"epoch": 0.66,
"learning_rate": 0.00017284854563691073,
"loss": 2.6206,
"theoretical_loss": 3.4071691056476805,
"tokens_seen": 2171207680
},
{
"epoch": 0.66,
"learning_rate": 0.00017283851554663994,
"loss": 2.4787,
"theoretical_loss": 3.4071606688715326,
"tokens_seen": 2171273216
},
{
"epoch": 0.66,
"learning_rate": 0.00017282848545636912,
"loss": 2.5772,
"theoretical_loss": 3.4071522324213293,
"tokens_seen": 2171338752
},
{
"epoch": 0.66,
"learning_rate": 0.0001728184553660983,
"loss": 2.4906,
"theoretical_loss": 3.4071437962970474,
"tokens_seen": 2171404288
},
{
"epoch": 0.66,
"learning_rate": 0.00017280842527582748,
"loss": 2.5513,
"theoretical_loss": 3.407135360498666,
"tokens_seen": 2171469824
},
{
"epoch": 0.66,
"learning_rate": 0.0001727983951855567,
"loss": 2.6378,
"theoretical_loss": 3.4071269250261604,
"tokens_seen": 2171535360
},
{
"epoch": 0.66,
"learning_rate": 0.00017278836509528587,
"loss": 2.5347,
"theoretical_loss": 3.407118489879511,
"tokens_seen": 2171600896
},
{
"epoch": 0.66,
"learning_rate": 0.00017277833500501505,
"loss": 2.4012,
"theoretical_loss": 3.407110055058693,
"tokens_seen": 2171666432
},
{
"epoch": 0.66,
"learning_rate": 0.00017276830491474423,
"loss": 2.6662,
"theoretical_loss": 3.407101620563685,
"tokens_seen": 2171731968
},
{
"epoch": 0.66,
"learning_rate": 0.00017275827482447342,
"loss": 2.4809,
"theoretical_loss": 3.4070931863944645,
"tokens_seen": 2171797504
},
{
"epoch": 0.66,
"learning_rate": 0.00017274824473420262,
"loss": 2.3943,
"theoretical_loss": 3.4070847525510093,
"tokens_seen": 2171863040
},
{
"epoch": 0.66,
"learning_rate": 0.0001727382146439318,
"loss": 2.5548,
"theoretical_loss": 3.4070763190332967,
"tokens_seen": 2171928576
},
{
"epoch": 0.66,
"learning_rate": 0.00017272818455366099,
"loss": 2.7812,
"theoretical_loss": 3.407067885841304,
"tokens_seen": 2171994112
},
{
"epoch": 0.66,
"learning_rate": 0.00017271815446339017,
"loss": 2.6545,
"theoretical_loss": 3.4070594529750093,
"tokens_seen": 2172059648
},
{
"epoch": 0.66,
"learning_rate": 0.00017270812437311937,
"loss": 2.5106,
"theoretical_loss": 3.4070510204343902,
"tokens_seen": 2172125184
},
{
"epoch": 0.66,
"learning_rate": 0.00017269809428284856,
"loss": 2.5139,
"theoretical_loss": 3.407042588219424,
"tokens_seen": 2172190720
},
{
"epoch": 0.66,
"learning_rate": 0.00017268806419257774,
"loss": 2.4208,
"theoretical_loss": 3.407034156330088,
"tokens_seen": 2172256256
},
{
"epoch": 0.66,
"learning_rate": 0.00017267803410230692,
"loss": 2.6633,
"theoretical_loss": 3.4070257247663607,
"tokens_seen": 2172321792
},
{
"epoch": 0.66,
"learning_rate": 0.0001726680040120361,
"loss": 2.5905,
"theoretical_loss": 3.4070172935282192,
"tokens_seen": 2172387328
},
{
"epoch": 0.66,
"learning_rate": 0.0001726579739217653,
"loss": 2.5218,
"theoretical_loss": 3.407008862615641,
"tokens_seen": 2172452864
},
{
"epoch": 0.66,
"objective/train/docs_used": 1230759,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.712217092514038,
"objective/train/theoretical_loss": 3.4070004320286036,
"objective/train/tokens_used": 2192978400,
"theoretical_loss": 3.4070004320286036,
"tokens_seen": 2172518400
},
{
"epoch": 0.66,
"learning_rate": 0.0001726479438314945,
"loss": 2.7697,
"theoretical_loss": 3.4070004320286036,
"tokens_seen": 2172518400
},
{
"epoch": 0.66,
"learning_rate": 0.00017263791374122367,
"loss": 2.4487,
"theoretical_loss": 3.4069920017670854,
"tokens_seen": 2172583936
},
{
"epoch": 0.66,
"learning_rate": 0.00017262788365095285,
"loss": 2.5715,
"theoretical_loss": 3.4069835718310633,
"tokens_seen": 2172649472
},
{
"epoch": 0.66,
"learning_rate": 0.00017261785356068206,
"loss": 2.5398,
"theoretical_loss": 3.4069751422205146,
"tokens_seen": 2172715008
},
{
"epoch": 0.66,
"learning_rate": 0.00017260782347041124,
"loss": 2.5489,
"theoretical_loss": 3.406966712935418,
"tokens_seen": 2172780544
},
{
"epoch": 0.66,
"learning_rate": 0.00017259779338014042,
"loss": 2.5845,
"theoretical_loss": 3.40695828397575,
"tokens_seen": 2172846080
},
{
"epoch": 0.66,
"learning_rate": 0.0001725877632898696,
"loss": 2.7403,
"theoretical_loss": 3.406949855341489,
"tokens_seen": 2172911616
},
{
"epoch": 0.66,
"learning_rate": 0.00017257773319959878,
"loss": 2.5426,
"theoretical_loss": 3.4069414270326126,
"tokens_seen": 2172977152
},
{
"epoch": 0.66,
"learning_rate": 0.000172567703109328,
"loss": 2.6167,
"theoretical_loss": 3.4069329990490975,
"tokens_seen": 2173042688
},
{
"epoch": 0.66,
"learning_rate": 0.00017255767301905717,
"loss": 2.7394,
"theoretical_loss": 3.4069245713909226,
"tokens_seen": 2173108224
},
{
"epoch": 0.66,
"learning_rate": 0.00017254764292878635,
"loss": 2.6712,
"theoretical_loss": 3.406916144058065,
"tokens_seen": 2173173760
},
{
"epoch": 0.66,
"learning_rate": 0.00017253761283851554,
"loss": 2.5855,
"theoretical_loss": 3.4069077170505015,
"tokens_seen": 2173239296
},
{
"epoch": 0.66,
"learning_rate": 0.00017252758274824474,
"loss": 2.3628,
"theoretical_loss": 3.4068992903682114,
"tokens_seen": 2173304832
},
{
"epoch": 0.66,
"learning_rate": 0.00017251755265797392,
"loss": 2.7758,
"theoretical_loss": 3.406890864011171,
"tokens_seen": 2173370368
},
{
"epoch": 0.66,
"learning_rate": 0.0001725075225677031,
"loss": 2.6498,
"theoretical_loss": 3.406882437979358,
"tokens_seen": 2173435904
},
{
"epoch": 0.66,
"learning_rate": 0.0001724974924774323,
"loss": 2.4082,
"theoretical_loss": 3.406874012272751,
"tokens_seen": 2173501440
},
{
"epoch": 0.66,
"learning_rate": 0.00017248746238716147,
"loss": 2.6441,
"theoretical_loss": 3.4068655868913273,
"tokens_seen": 2173566976
},
{
"epoch": 0.66,
"learning_rate": 0.00017247743229689068,
"loss": 2.6247,
"theoretical_loss": 3.406857161835064,
"tokens_seen": 2173632512
},
{
"epoch": 0.66,
"learning_rate": 0.00017246740220661986,
"loss": 2.5961,
"theoretical_loss": 3.406848737103939,
"tokens_seen": 2173698048
},
{
"epoch": 0.66,
"learning_rate": 0.00017245737211634904,
"loss": 2.5346,
"theoretical_loss": 3.4068403126979296,
"tokens_seen": 2173763584
},
{
"epoch": 0.66,
"learning_rate": 0.00017244734202607822,
"loss": 2.7081,
"theoretical_loss": 3.406831888617014,
"tokens_seen": 2173829120
},
{
"epoch": 0.66,
"learning_rate": 0.00017243731193580743,
"loss": 2.5832,
"theoretical_loss": 3.40682346486117,
"tokens_seen": 2173894656
},
{
"epoch": 0.66,
"learning_rate": 0.0001724272818455366,
"loss": 2.6044,
"theoretical_loss": 3.4068150414303755,
"tokens_seen": 2173960192
},
{
"epoch": 0.66,
"learning_rate": 0.0001724172517552658,
"loss": 2.5476,
"theoretical_loss": 3.406806618324607,
"tokens_seen": 2174025728
},
{
"epoch": 0.66,
"learning_rate": 0.00017240722166499497,
"loss": 2.6038,
"theoretical_loss": 3.4067981955438427,
"tokens_seen": 2174091264
},
{
"epoch": 0.66,
"objective/train/docs_used": 1232037,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.322192907333374,
"objective/train/theoretical_loss": 3.4067897730880605,
"objective/train/tokens_used": 2194616800,
"theoretical_loss": 3.4067897730880605,
"tokens_seen": 2174156800
},
{
"epoch": 0.66,
"learning_rate": 0.00017239719157472418,
"loss": 2.5408,
"theoretical_loss": 3.4067897730880605,
"tokens_seen": 2174156800
},
{
"epoch": 0.66,
"learning_rate": 0.00017238716148445336,
"loss": 2.8156,
"theoretical_loss": 3.406781350957238,
"tokens_seen": 2174222336
},
{
"epoch": 0.66,
"learning_rate": 0.00017237713139418254,
"loss": 2.7868,
"theoretical_loss": 3.406772929151353,
"tokens_seen": 2174287872
},
{
"epoch": 0.66,
"learning_rate": 0.00017236710130391172,
"loss": 2.769,
"theoretical_loss": 3.4067645076703825,
"tokens_seen": 2174353408
},
{
"epoch": 0.66,
"learning_rate": 0.0001723570712136409,
"loss": 2.5981,
"theoretical_loss": 3.4067560865143047,
"tokens_seen": 2174418944
},
{
"epoch": 0.66,
"learning_rate": 0.0001723470411233701,
"loss": 2.5962,
"theoretical_loss": 3.4067476656830973,
"tokens_seen": 2174484480
},
{
"epoch": 0.66,
"learning_rate": 0.0001723370110330993,
"loss": 2.6615,
"theoretical_loss": 3.406739245176738,
"tokens_seen": 2174550016
},
{
"epoch": 0.66,
"learning_rate": 0.00017232698094282847,
"loss": 2.667,
"theoretical_loss": 3.406730824995204,
"tokens_seen": 2174615552
},
{
"epoch": 0.66,
"learning_rate": 0.00017231695085255766,
"loss": 2.6348,
"theoretical_loss": 3.406722405138474,
"tokens_seen": 2174681088
},
{
"epoch": 0.66,
"learning_rate": 0.00017230692076228686,
"loss": 2.9274,
"theoretical_loss": 3.4067139856065247,
"tokens_seen": 2174746624
},
{
"epoch": 0.66,
"learning_rate": 0.00017229689067201607,
"loss": 2.5207,
"theoretical_loss": 3.406705566399334,
"tokens_seen": 2174812160
},
{
"epoch": 0.66,
"learning_rate": 0.00017228686058174525,
"loss": 2.6282,
"theoretical_loss": 3.40669714751688,
"tokens_seen": 2174877696
},
{
"epoch": 0.66,
"learning_rate": 0.00017227683049147443,
"loss": 2.5861,
"theoretical_loss": 3.4066887289591397,
"tokens_seen": 2174943232
},
{
"epoch": 0.66,
"learning_rate": 0.00017226680040120362,
"loss": 2.6303,
"theoretical_loss": 3.4066803107260912,
"tokens_seen": 2175008768
},
{
"epoch": 0.66,
"learning_rate": 0.00017225677031093282,
"loss": 2.6519,
"theoretical_loss": 3.4066718928177124,
"tokens_seen": 2175074304
},
{
"epoch": 0.66,
"learning_rate": 0.000172246740220662,
"loss": 2.6372,
"theoretical_loss": 3.4066634752339806,
"tokens_seen": 2175139840
},
{
"epoch": 0.66,
"learning_rate": 0.00017223671013039119,
"loss": 2.5806,
"theoretical_loss": 3.4066550579748736,
"tokens_seen": 2175205376
},
{
"epoch": 0.66,
"learning_rate": 0.00017222668004012037,
"loss": 2.4595,
"theoretical_loss": 3.4066466410403695,
"tokens_seen": 2175270912
},
{
"epoch": 0.66,
"learning_rate": 0.00017221664994984958,
"loss": 2.5859,
"theoretical_loss": 3.4066382244304454,
"tokens_seen": 2175336448
},
{
"epoch": 0.66,
"learning_rate": 0.00017220661985957876,
"loss": 2.5398,
"theoretical_loss": 3.4066298081450794,
"tokens_seen": 2175401984
},
{
"epoch": 0.66,
"learning_rate": 0.00017219658976930794,
"loss": 2.496,
"theoretical_loss": 3.406621392184249,
"tokens_seen": 2175467520
},
{
"epoch": 0.66,
"learning_rate": 0.00017218655967903712,
"loss": 2.6845,
"theoretical_loss": 3.4066129765479323,
"tokens_seen": 2175533056
},
{
"epoch": 0.66,
"learning_rate": 0.0001721765295887663,
"loss": 2.8011,
"theoretical_loss": 3.406604561236106,
"tokens_seen": 2175598592
},
{
"epoch": 0.66,
"learning_rate": 0.0001721664994984955,
"loss": 2.8159,
"theoretical_loss": 3.406596146248749,
"tokens_seen": 2175664128
},
{
"epoch": 0.66,
"learning_rate": 0.0001721564694082247,
"loss": 2.4398,
"theoretical_loss": 3.4065877315858386,
"tokens_seen": 2175729664
},
{
"epoch": 0.66,
"objective/train/docs_used": 1232648,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5028202533721924,
"objective/train/theoretical_loss": 3.4065793172473526,
"objective/train/tokens_used": 2196255200,
"theoretical_loss": 3.4065793172473526,
"tokens_seen": 2175795200
},
{
"epoch": 0.66,
"learning_rate": 0.00017214643931795387,
"loss": 2.592,
"theoretical_loss": 3.4065793172473526,
"tokens_seen": 2175795200
},
{
"epoch": 0.66,
"learning_rate": 0.00017213640922768305,
"loss": 2.5333,
"theoretical_loss": 3.406570903233268,
"tokens_seen": 2175860736
},
{
"epoch": 0.66,
"learning_rate": 0.00017212637913741226,
"loss": 2.5496,
"theoretical_loss": 3.4065624895435636,
"tokens_seen": 2175926272
},
{
"epoch": 0.66,
"learning_rate": 0.00017211634904714144,
"loss": 2.5301,
"theoretical_loss": 3.4065540761782165,
"tokens_seen": 2175991808
},
{
"epoch": 0.66,
"learning_rate": 0.00017210631895687062,
"loss": 2.6833,
"theoretical_loss": 3.406545663137204,
"tokens_seen": 2176057344
},
{
"epoch": 0.66,
"learning_rate": 0.0001720962888665998,
"loss": 2.4902,
"theoretical_loss": 3.4065372504205054,
"tokens_seen": 2176122880
},
{
"epoch": 0.66,
"learning_rate": 0.00017208625877632898,
"loss": 2.5423,
"theoretical_loss": 3.406528838028097,
"tokens_seen": 2176188416
},
{
"epoch": 0.66,
"learning_rate": 0.0001720762286860582,
"loss": 2.6745,
"theoretical_loss": 3.4065204259599566,
"tokens_seen": 2176253952
},
{
"epoch": 0.66,
"learning_rate": 0.00017206619859578737,
"loss": 2.7293,
"theoretical_loss": 3.4065120142160628,
"tokens_seen": 2176319488
},
{
"epoch": 0.66,
"learning_rate": 0.00017205616850551655,
"loss": 2.652,
"theoretical_loss": 3.4065036027963926,
"tokens_seen": 2176385024
},
{
"epoch": 0.66,
"learning_rate": 0.00017204613841524574,
"loss": 2.3774,
"theoretical_loss": 3.406495191700924,
"tokens_seen": 2176450560
},
{
"epoch": 0.66,
"learning_rate": 0.00017203610832497494,
"loss": 2.6725,
"theoretical_loss": 3.4064867809296344,
"tokens_seen": 2176516096
},
{
"epoch": 0.66,
"learning_rate": 0.00017202607823470412,
"loss": 2.4723,
"theoretical_loss": 3.4064783704825023,
"tokens_seen": 2176581632
},
{
"epoch": 0.66,
"learning_rate": 0.0001720160481444333,
"loss": 2.6738,
"theoretical_loss": 3.406469960359505,
"tokens_seen": 2176647168
},
{
"epoch": 0.66,
"learning_rate": 0.0001720060180541625,
"loss": 2.5779,
"theoretical_loss": 3.4064615505606195,
"tokens_seen": 2176712704
},
{
"epoch": 0.66,
"learning_rate": 0.00017199598796389167,
"loss": 2.6852,
"theoretical_loss": 3.4064531410858248,
"tokens_seen": 2176778240
},
{
"epoch": 0.66,
"learning_rate": 0.00017198595787362088,
"loss": 2.4115,
"theoretical_loss": 3.4064447319350983,
"tokens_seen": 2176843776
},
{
"epoch": 0.66,
"learning_rate": 0.00017197592778335006,
"loss": 2.4876,
"theoretical_loss": 3.4064363231084176,
"tokens_seen": 2176909312
},
{
"epoch": 0.66,
"learning_rate": 0.00017196589769307924,
"loss": 2.6704,
"theoretical_loss": 3.4064279146057603,
"tokens_seen": 2176974848
},
{
"epoch": 0.66,
"learning_rate": 0.00017195586760280842,
"loss": 2.7927,
"theoretical_loss": 3.4064195064271043,
"tokens_seen": 2177040384
},
{
"epoch": 0.66,
"learning_rate": 0.00017194583751253763,
"loss": 2.7191,
"theoretical_loss": 3.406411098572428,
"tokens_seen": 2177105920
},
{
"epoch": 0.66,
"learning_rate": 0.0001719358074222668,
"loss": 2.6484,
"theoretical_loss": 3.4064026910417073,
"tokens_seen": 2177171456
},
{
"epoch": 0.66,
"learning_rate": 0.000171925777331996,
"loss": 2.6256,
"theoretical_loss": 3.4063942838349224,
"tokens_seen": 2177236992
},
{
"epoch": 0.66,
"learning_rate": 0.00017191574724172517,
"loss": 2.5196,
"theoretical_loss": 3.4063858769520494,
"tokens_seen": 2177302528
},
{
"epoch": 0.66,
"learning_rate": 0.00017190571715145438,
"loss": 2.6796,
"theoretical_loss": 3.4063774703930667,
"tokens_seen": 2177368064
},
{
"epoch": 0.66,
"objective/train/docs_used": 1233682,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4729695320129395,
"objective/train/theoretical_loss": 3.4063690641579516,
"objective/train/tokens_used": 2197893600,
"theoretical_loss": 3.4063690641579516,
"tokens_seen": 2177433600
},
{
"epoch": 0.66,
"learning_rate": 0.00017189568706118356,
"loss": 2.5376,
"theoretical_loss": 3.4063690641579516,
"tokens_seen": 2177433600
},
{
"epoch": 0.66,
"learning_rate": 0.00017188565697091274,
"loss": 2.735,
"theoretical_loss": 3.4063606582466823,
"tokens_seen": 2177499136
},
{
"epoch": 0.66,
"learning_rate": 0.00017187562688064192,
"loss": 2.4594,
"theoretical_loss": 3.4063522526592367,
"tokens_seen": 2177564672
},
{
"epoch": 0.66,
"learning_rate": 0.0001718655967903711,
"loss": 2.6457,
"theoretical_loss": 3.4063438473955925,
"tokens_seen": 2177630208
},
{
"epoch": 0.66,
"learning_rate": 0.0001718555667001003,
"loss": 2.5858,
"theoretical_loss": 3.406335442455727,
"tokens_seen": 2177695744
},
{
"epoch": 0.66,
"learning_rate": 0.0001718455366098295,
"loss": 2.4823,
"theoretical_loss": 3.4063270378396187,
"tokens_seen": 2177761280
},
{
"epoch": 0.66,
"learning_rate": 0.00017183550651955867,
"loss": 2.5959,
"theoretical_loss": 3.4063186335472446,
"tokens_seen": 2177826816
},
{
"epoch": 0.66,
"learning_rate": 0.00017182547642928786,
"loss": 2.5855,
"theoretical_loss": 3.406310229578583,
"tokens_seen": 2177892352
},
{
"epoch": 0.66,
"learning_rate": 0.00017181544633901706,
"loss": 2.3502,
"theoretical_loss": 3.4063018259336117,
"tokens_seen": 2177957888
},
{
"epoch": 0.66,
"learning_rate": 0.00017180541624874625,
"loss": 2.6755,
"theoretical_loss": 3.4062934226123085,
"tokens_seen": 2178023424
},
{
"epoch": 0.66,
"learning_rate": 0.00017179538615847543,
"loss": 2.4802,
"theoretical_loss": 3.406285019614651,
"tokens_seen": 2178088960
},
{
"epoch": 0.66,
"learning_rate": 0.0001717853560682046,
"loss": 2.6722,
"theoretical_loss": 3.406276616940617,
"tokens_seen": 2178154496
},
{
"epoch": 0.66,
"learning_rate": 0.0001717753259779338,
"loss": 2.5478,
"theoretical_loss": 3.4062682145901846,
"tokens_seen": 2178220032
},
{
"epoch": 0.66,
"learning_rate": 0.000171765295887663,
"loss": 2.4767,
"theoretical_loss": 3.406259812563331,
"tokens_seen": 2178285568
},
{
"epoch": 0.66,
"learning_rate": 0.00017175526579739218,
"loss": 2.6965,
"theoretical_loss": 3.4062514108600346,
"tokens_seen": 2178351104
},
{
"epoch": 0.66,
"learning_rate": 0.00017174523570712136,
"loss": 2.8953,
"theoretical_loss": 3.406243009480273,
"tokens_seen": 2178416640
},
{
"epoch": 0.66,
"learning_rate": 0.00017173520561685054,
"loss": 2.6896,
"theoretical_loss": 3.406234608424024,
"tokens_seen": 2178482176
},
{
"epoch": 0.66,
"learning_rate": 0.00017172517552657975,
"loss": 2.5756,
"theoretical_loss": 3.4062262076912653,
"tokens_seen": 2178547712
},
{
"epoch": 0.66,
"learning_rate": 0.00017171514543630893,
"loss": 2.6493,
"theoretical_loss": 3.4062178072819753,
"tokens_seen": 2178613248
},
{
"epoch": 0.66,
"learning_rate": 0.0001717051153460381,
"loss": 2.6304,
"theoretical_loss": 3.406209407196131,
"tokens_seen": 2178678784
},
{
"epoch": 0.66,
"learning_rate": 0.0001716950852557673,
"loss": 2.493,
"theoretical_loss": 3.406201007433711,
"tokens_seen": 2178744320
},
{
"epoch": 0.66,
"learning_rate": 0.00017168505516549647,
"loss": 2.502,
"theoretical_loss": 3.406192607994692,
"tokens_seen": 2178809856
},
{
"epoch": 0.66,
"learning_rate": 0.00017167502507522568,
"loss": 2.6025,
"theoretical_loss": 3.4061842088790524,
"tokens_seen": 2178875392
},
{
"epoch": 0.66,
"learning_rate": 0.00017166499498495486,
"loss": 2.442,
"theoretical_loss": 3.4061758100867707,
"tokens_seen": 2178940928
},
{
"epoch": 0.66,
"learning_rate": 0.00017165496489468404,
"loss": 2.8157,
"theoretical_loss": 3.4061674116178238,
"tokens_seen": 2179006464
},
{
"epoch": 0.66,
"objective/train/docs_used": 1234346,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6162099838256836,
"objective/train/theoretical_loss": 3.40615901347219,
"objective/train/tokens_used": 2199532000,
"theoretical_loss": 3.40615901347219,
"tokens_seen": 2179072000
},
{
"epoch": 0.66,
"learning_rate": 0.00017164493480441322,
"loss": 2.6073,
"theoretical_loss": 3.40615901347219,
"tokens_seen": 2179072000
},
{
"epoch": 0.66,
"learning_rate": 0.00017163490471414243,
"loss": 2.5826,
"theoretical_loss": 3.4061506156498473,
"tokens_seen": 2179137536
},
{
"epoch": 0.66,
"learning_rate": 0.00017162487462387161,
"loss": 2.7512,
"theoretical_loss": 3.406142218150773,
"tokens_seen": 2179203072
},
{
"epoch": 0.66,
"learning_rate": 0.0001716148445336008,
"loss": 2.6875,
"theoretical_loss": 3.406133820974945,
"tokens_seen": 2179268608
},
{
"epoch": 0.66,
"learning_rate": 0.00017160481444332998,
"loss": 2.5507,
"theoretical_loss": 3.4061254241223415,
"tokens_seen": 2179334144
},
{
"epoch": 0.66,
"learning_rate": 0.00017159478435305916,
"loss": 2.541,
"theoretical_loss": 3.4061170275929404,
"tokens_seen": 2179399680
},
{
"epoch": 0.66,
"learning_rate": 0.00017158475426278837,
"loss": 2.5089,
"theoretical_loss": 3.406108631386719,
"tokens_seen": 2179465216
},
{
"epoch": 0.66,
"learning_rate": 0.00017157472417251755,
"loss": 2.6032,
"theoretical_loss": 3.4061002355036556,
"tokens_seen": 2179530752
},
{
"epoch": 0.66,
"learning_rate": 0.00017156469408224673,
"loss": 2.6094,
"theoretical_loss": 3.406091839943728,
"tokens_seen": 2179596288
},
{
"epoch": 0.66,
"learning_rate": 0.0001715546639919759,
"loss": 2.3776,
"theoretical_loss": 3.4060834447069133,
"tokens_seen": 2179661824
},
{
"epoch": 0.66,
"learning_rate": 0.00017154463390170514,
"loss": 2.7762,
"theoretical_loss": 3.4060750497931904,
"tokens_seen": 2179727360
},
{
"epoch": 0.66,
"learning_rate": 0.00017153460381143433,
"loss": 2.7607,
"theoretical_loss": 3.4060666552025367,
"tokens_seen": 2179792896
},
{
"epoch": 0.66,
"learning_rate": 0.0001715245737211635,
"loss": 2.5352,
"theoretical_loss": 3.4060582609349304,
"tokens_seen": 2179858432
},
{
"epoch": 0.66,
"learning_rate": 0.0001715145436308927,
"loss": 2.5605,
"theoretical_loss": 3.4060498669903483,
"tokens_seen": 2179923968
},
{
"epoch": 0.66,
"learning_rate": 0.00017150451354062187,
"loss": 2.4628,
"theoretical_loss": 3.4060414733687696,
"tokens_seen": 2179989504
},
{
"epoch": 0.66,
"learning_rate": 0.00017149448345035108,
"loss": 2.8187,
"theoretical_loss": 3.4060330800701717,
"tokens_seen": 2180055040
},
{
"epoch": 0.66,
"learning_rate": 0.00017148445336008026,
"loss": 2.4601,
"theoretical_loss": 3.406024687094532,
"tokens_seen": 2180120576
},
{
"epoch": 0.66,
"learning_rate": 0.00017147442326980944,
"loss": 2.2992,
"theoretical_loss": 3.4060162944418284,
"tokens_seen": 2180186112
},
{
"epoch": 0.66,
"learning_rate": 0.00017146439317953862,
"loss": 2.6616,
"theoretical_loss": 3.40600790211204,
"tokens_seen": 2180251648
},
{
"epoch": 0.66,
"learning_rate": 0.00017145436308926783,
"loss": 2.7926,
"theoretical_loss": 3.4059995101051426,
"tokens_seen": 2180317184
},
{
"epoch": 0.66,
"learning_rate": 0.000171444332998997,
"loss": 2.6235,
"theoretical_loss": 3.405991118421116,
"tokens_seen": 2180382720
},
{
"epoch": 0.66,
"learning_rate": 0.0001714343029087262,
"loss": 2.787,
"theoretical_loss": 3.405982727059937,
"tokens_seen": 2180448256
},
{
"epoch": 0.66,
"learning_rate": 0.00017142427281845537,
"loss": 2.4921,
"theoretical_loss": 3.405974336021584,
"tokens_seen": 2180513792
},
{
"epoch": 0.66,
"learning_rate": 0.00017141424272818455,
"loss": 2.9329,
"theoretical_loss": 3.4059659453060345,
"tokens_seen": 2180579328
},
{
"epoch": 0.66,
"learning_rate": 0.00017140421263791376,
"loss": 2.6301,
"theoretical_loss": 3.4059575549132663,
"tokens_seen": 2180644864
},
{
"epoch": 0.66,
"objective/train/docs_used": 1235717,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8363263607025146,
"objective/train/theoretical_loss": 3.4059491648432574,
"objective/train/tokens_used": 2201170400,
"theoretical_loss": 3.4059491648432574,
"tokens_seen": 2180710400
},
{
"epoch": 0.66,
"learning_rate": 0.00017139418254764294,
"loss": 2.8309,
"theoretical_loss": 3.4059491648432574,
"tokens_seen": 2180710400
},
{
"epoch": 0.66,
"learning_rate": 0.00017138415245737212,
"loss": 2.7891,
"theoretical_loss": 3.4059407750959863,
"tokens_seen": 2180775936
},
{
"epoch": 0.66,
"learning_rate": 0.0001713741223671013,
"loss": 2.5105,
"theoretical_loss": 3.40593238567143,
"tokens_seen": 2180841472
},
{
"epoch": 0.66,
"learning_rate": 0.0001713640922768305,
"loss": 2.6646,
"theoretical_loss": 3.405923996569567,
"tokens_seen": 2180907008
},
{
"epoch": 0.66,
"learning_rate": 0.0001713540621865597,
"loss": 2.8409,
"theoretical_loss": 3.4059156077903747,
"tokens_seen": 2180972544
},
{
"epoch": 0.66,
"learning_rate": 0.00017134403209628887,
"loss": 2.4456,
"theoretical_loss": 3.405907219333831,
"tokens_seen": 2181038080
},
{
"epoch": 0.66,
"learning_rate": 0.00017133400200601806,
"loss": 2.7178,
"theoretical_loss": 3.4058988311999148,
"tokens_seen": 2181103616
},
{
"epoch": 0.66,
"learning_rate": 0.00017132397191574726,
"loss": 2.4982,
"theoretical_loss": 3.4058904433886026,
"tokens_seen": 2181169152
},
{
"epoch": 0.66,
"learning_rate": 0.00017131394182547645,
"loss": 2.8126,
"theoretical_loss": 3.4058820558998733,
"tokens_seen": 2181234688
},
{
"epoch": 0.66,
"learning_rate": 0.00017130391173520563,
"loss": 2.5861,
"theoretical_loss": 3.4058736687337046,
"tokens_seen": 2181300224
},
{
"epoch": 0.66,
"learning_rate": 0.0001712938816449348,
"loss": 2.5205,
"theoretical_loss": 3.4058652818900734,
"tokens_seen": 2181365760
},
{
"epoch": 0.66,
"learning_rate": 0.000171283851554664,
"loss": 2.7303,
"theoretical_loss": 3.4058568953689594,
"tokens_seen": 2181431296
},
{
"epoch": 0.66,
"learning_rate": 0.0001712738214643932,
"loss": 2.6367,
"theoretical_loss": 3.405848509170339,
"tokens_seen": 2181496832
},
{
"epoch": 0.66,
"learning_rate": 0.00017126379137412238,
"loss": 2.4997,
"theoretical_loss": 3.405840123294191,
"tokens_seen": 2181562368
},
{
"epoch": 0.66,
"learning_rate": 0.00017125376128385156,
"loss": 2.4952,
"theoretical_loss": 3.4058317377404927,
"tokens_seen": 2181627904
},
{
"epoch": 0.66,
"learning_rate": 0.00017124373119358074,
"loss": 2.8158,
"theoretical_loss": 3.4058233525092225,
"tokens_seen": 2181693440
},
{
"epoch": 0.66,
"learning_rate": 0.00017123370110330995,
"loss": 2.5035,
"theoretical_loss": 3.405814967600358,
"tokens_seen": 2181758976
},
{
"epoch": 0.66,
"learning_rate": 0.00017122367101303913,
"loss": 2.4613,
"theoretical_loss": 3.4058065830138773,
"tokens_seen": 2181824512
},
{
"epoch": 0.66,
"learning_rate": 0.0001712136409227683,
"loss": 2.6436,
"theoretical_loss": 3.405798198749758,
"tokens_seen": 2181890048
},
{
"epoch": 0.66,
"learning_rate": 0.0001712036108324975,
"loss": 2.8638,
"theoretical_loss": 3.405789814807979,
"tokens_seen": 2181955584
},
{
"epoch": 0.66,
"learning_rate": 0.00017119358074222667,
"loss": 2.7076,
"theoretical_loss": 3.4057814311885166,
"tokens_seen": 2182021120
},
{
"epoch": 0.66,
"learning_rate": 0.00017118355065195588,
"loss": 2.6262,
"theoretical_loss": 3.4057730478913504,
"tokens_seen": 2182086656
},
{
"epoch": 0.66,
"learning_rate": 0.00017117352056168506,
"loss": 2.5538,
"theoretical_loss": 3.405764664916457,
"tokens_seen": 2182152192
},
{
"epoch": 0.66,
"learning_rate": 0.00017116349047141424,
"loss": 2.7559,
"theoretical_loss": 3.405756282263815,
"tokens_seen": 2182217728
},
{
"epoch": 0.66,
"learning_rate": 0.00017115346038114342,
"loss": 2.5249,
"theoretical_loss": 3.4057478999334023,
"tokens_seen": 2182283264
},
{
"epoch": 0.66,
"objective/train/docs_used": 1236411,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.595909833908081,
"objective/train/theoretical_loss": 3.405739517925197,
"objective/train/tokens_used": 2202808800,
"theoretical_loss": 3.405739517925197,
"tokens_seen": 2182348800
},
{
"epoch": 0.66,
"learning_rate": 0.00017114343029087263,
"loss": 2.6883,
"theoretical_loss": 3.405739517925197,
"tokens_seen": 2182348800
},
{
"epoch": 0.66,
"learning_rate": 0.00017113340020060181,
"loss": 2.7543,
"theoretical_loss": 3.4057311362391767,
"tokens_seen": 2182414336
},
{
"epoch": 0.66,
"learning_rate": 0.000171123370110331,
"loss": 2.4915,
"theoretical_loss": 3.4057227548753195,
"tokens_seen": 2182479872
},
{
"epoch": 0.66,
"learning_rate": 0.00017111334002006018,
"loss": 2.576,
"theoretical_loss": 3.405714373833603,
"tokens_seen": 2182545408
},
{
"epoch": 0.66,
"learning_rate": 0.00017110330992978936,
"loss": 2.7381,
"theoretical_loss": 3.405705993114006,
"tokens_seen": 2182610944
},
{
"epoch": 0.66,
"learning_rate": 0.00017109327983951857,
"loss": 2.687,
"theoretical_loss": 3.4056976127165055,
"tokens_seen": 2182676480
},
{
"epoch": 0.66,
"learning_rate": 0.00017108324974924775,
"loss": 2.7033,
"theoretical_loss": 3.40568923264108,
"tokens_seen": 2182742016
},
{
"epoch": 0.66,
"learning_rate": 0.00017107321965897693,
"loss": 2.5937,
"theoretical_loss": 3.405680852887707,
"tokens_seen": 2182807552
},
{
"epoch": 0.66,
"learning_rate": 0.0001710631895687061,
"loss": 2.8664,
"theoretical_loss": 3.405672473456365,
"tokens_seen": 2182873088
},
{
"epoch": 0.66,
"learning_rate": 0.00017105315947843532,
"loss": 2.3803,
"theoretical_loss": 3.4056640943470318,
"tokens_seen": 2182938624
},
{
"epoch": 0.66,
"learning_rate": 0.0001710431293881645,
"loss": 2.5178,
"theoretical_loss": 3.405655715559685,
"tokens_seen": 2183004160
},
{
"epoch": 0.66,
"learning_rate": 0.00017103309929789368,
"loss": 2.604,
"theoretical_loss": 3.405647337094303,
"tokens_seen": 2183069696
},
{
"epoch": 0.66,
"learning_rate": 0.00017102306920762286,
"loss": 2.5179,
"theoretical_loss": 3.4056389589508633,
"tokens_seen": 2183135232
},
{
"epoch": 0.66,
"learning_rate": 0.00017101303911735204,
"loss": 2.6351,
"theoretical_loss": 3.4056305811293446,
"tokens_seen": 2183200768
},
{
"epoch": 0.66,
"learning_rate": 0.00017100300902708125,
"loss": 2.8377,
"theoretical_loss": 3.4056222036297243,
"tokens_seen": 2183266304
},
{
"epoch": 0.66,
"learning_rate": 0.00017099297893681043,
"loss": 2.6374,
"theoretical_loss": 3.4056138264519804,
"tokens_seen": 2183331840
},
{
"epoch": 0.66,
"learning_rate": 0.0001709829488465396,
"loss": 2.4149,
"theoretical_loss": 3.405605449596091,
"tokens_seen": 2183397376
},
{
"epoch": 0.66,
"learning_rate": 0.0001709729187562688,
"loss": 2.5342,
"theoretical_loss": 3.405597073062034,
"tokens_seen": 2183462912
},
{
"epoch": 0.66,
"learning_rate": 0.000170962888665998,
"loss": 2.8407,
"theoretical_loss": 3.4055886968497875,
"tokens_seen": 2183528448
},
{
"epoch": 0.66,
"learning_rate": 0.00017095285857572718,
"loss": 2.6932,
"theoretical_loss": 3.4055803209593294,
"tokens_seen": 2183593984
},
{
"epoch": 0.66,
"learning_rate": 0.00017094282848545636,
"loss": 2.7689,
"theoretical_loss": 3.4055719453906375,
"tokens_seen": 2183659520
},
{
"epoch": 0.66,
"learning_rate": 0.00017093279839518554,
"loss": 2.5313,
"theoretical_loss": 3.40556357014369,
"tokens_seen": 2183725056
},
{
"epoch": 0.66,
"learning_rate": 0.00017092276830491473,
"loss": 2.5715,
"theoretical_loss": 3.405555195218465,
"tokens_seen": 2183790592
},
{
"epoch": 0.66,
"learning_rate": 0.00017091273821464393,
"loss": 2.6077,
"theoretical_loss": 3.40554682061494,
"tokens_seen": 2183856128
},
{
"epoch": 0.66,
"learning_rate": 0.00017090270812437312,
"loss": 2.8187,
"theoretical_loss": 3.405538446333094,
"tokens_seen": 2183921664
},
{
"epoch": 0.66,
"objective/train/docs_used": 1237677,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.069289207458496,
"objective/train/theoretical_loss": 3.405530072372904,
"objective/train/tokens_used": 2204447200,
"theoretical_loss": 3.405530072372904,
"tokens_seen": 2183987200
},
{
"epoch": 0.66,
"learning_rate": 0.0001708926780341023,
"loss": 2.9309,
"theoretical_loss": 3.405530072372904,
"tokens_seen": 2183987200
},
{
"epoch": 0.66,
"learning_rate": 0.00017088264794383148,
"loss": 2.7069,
"theoretical_loss": 3.405521698734348,
"tokens_seen": 2184052736
},
{
"epoch": 0.66,
"learning_rate": 0.00017087261785356069,
"loss": 2.6762,
"theoretical_loss": 3.405513325417404,
"tokens_seen": 2184118272
},
{
"epoch": 0.66,
"learning_rate": 0.00017086258776328987,
"loss": 2.4913,
"theoretical_loss": 3.405504952422051,
"tokens_seen": 2184183808
},
{
"epoch": 0.66,
"learning_rate": 0.00017085255767301905,
"loss": 2.7618,
"theoretical_loss": 3.405496579748266,
"tokens_seen": 2184249344
},
{
"epoch": 0.66,
"learning_rate": 0.00017084252758274823,
"loss": 2.4848,
"theoretical_loss": 3.4054882073960275,
"tokens_seen": 2184314880
},
{
"epoch": 0.66,
"learning_rate": 0.0001708324974924774,
"loss": 2.8526,
"theoretical_loss": 3.405479835365313,
"tokens_seen": 2184380416
},
{
"epoch": 0.66,
"learning_rate": 0.00017082246740220662,
"loss": 2.7242,
"theoretical_loss": 3.4054714636561014,
"tokens_seen": 2184445952
},
{
"epoch": 0.66,
"learning_rate": 0.0001708124373119358,
"loss": 2.4631,
"theoretical_loss": 3.4054630922683695,
"tokens_seen": 2184511488
},
{
"epoch": 0.66,
"learning_rate": 0.00017080240722166498,
"loss": 2.5839,
"theoretical_loss": 3.405454721202096,
"tokens_seen": 2184577024
},
{
"epoch": 0.66,
"learning_rate": 0.0001707923771313942,
"loss": 2.4905,
"theoretical_loss": 3.405446350457259,
"tokens_seen": 2184642560
},
{
"epoch": 0.66,
"learning_rate": 0.0001707823470411234,
"loss": 2.5045,
"theoretical_loss": 3.4054379800338364,
"tokens_seen": 2184708096
},
{
"epoch": 0.66,
"learning_rate": 0.00017077231695085258,
"loss": 2.6327,
"theoretical_loss": 3.405429609931806,
"tokens_seen": 2184773632
},
{
"epoch": 0.66,
"learning_rate": 0.00017076228686058176,
"loss": 2.5839,
"theoretical_loss": 3.4054212401511466,
"tokens_seen": 2184839168
},
{
"epoch": 0.66,
"learning_rate": 0.00017075225677031094,
"loss": 2.4927,
"theoretical_loss": 3.4054128706918347,
"tokens_seen": 2184904704
},
{
"epoch": 0.66,
"learning_rate": 0.00017074222668004015,
"loss": 2.4752,
"theoretical_loss": 3.40540450155385,
"tokens_seen": 2184970240
},
{
"epoch": 0.66,
"learning_rate": 0.00017073219658976933,
"loss": 2.7477,
"theoretical_loss": 3.4053961327371693,
"tokens_seen": 2185035776
},
{
"epoch": 0.66,
"learning_rate": 0.0001707221664994985,
"loss": 2.6083,
"theoretical_loss": 3.4053877642417714,
"tokens_seen": 2185101312
},
{
"epoch": 0.66,
"learning_rate": 0.0001707121364092277,
"loss": 2.6839,
"theoretical_loss": 3.4053793960676337,
"tokens_seen": 2185166848
},
{
"epoch": 0.66,
"learning_rate": 0.00017070210631895687,
"loss": 2.6818,
"theoretical_loss": 3.405371028214735,
"tokens_seen": 2185232384
},
{
"epoch": 0.66,
"learning_rate": 0.00017069207622868608,
"loss": 2.5732,
"theoretical_loss": 3.4053626606830525,
"tokens_seen": 2185297920
},
{
"epoch": 0.66,
"learning_rate": 0.00017068204613841526,
"loss": 2.6739,
"theoretical_loss": 3.405354293472565,
"tokens_seen": 2185363456
},
{
"epoch": 0.66,
"learning_rate": 0.00017067201604814444,
"loss": 2.6075,
"theoretical_loss": 3.4053459265832497,
"tokens_seen": 2185428992
},
{
"epoch": 0.66,
"learning_rate": 0.00017066198595787362,
"loss": 2.671,
"theoretical_loss": 3.4053375600150853,
"tokens_seen": 2185494528
},
{
"epoch": 0.66,
"learning_rate": 0.00017065195586760283,
"loss": 2.6981,
"theoretical_loss": 3.4053291937680497,
"tokens_seen": 2185560064
},
{
"epoch": 0.66,
"objective/train/docs_used": 1238334,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.319587230682373,
"objective/train/theoretical_loss": 3.4053208278421208,
"objective/train/tokens_used": 2206085600,
"theoretical_loss": 3.4053208278421208,
"tokens_seen": 2185625600
},
{
"epoch": 0.66,
"learning_rate": 0.00017064192577733201,
"loss": 2.5252,
"theoretical_loss": 3.4053208278421208,
"tokens_seen": 2185625600
},
{
"epoch": 0.66,
"learning_rate": 0.0001706318956870612,
"loss": 2.5807,
"theoretical_loss": 3.405312462237277,
"tokens_seen": 2185691136
},
{
"epoch": 0.66,
"learning_rate": 0.00017062186559679038,
"loss": 2.6902,
"theoretical_loss": 3.405304096953496,
"tokens_seen": 2185756672
},
{
"epoch": 0.66,
"learning_rate": 0.00017061183550651956,
"loss": 2.7282,
"theoretical_loss": 3.4052957319907557,
"tokens_seen": 2185822208
},
{
"epoch": 0.66,
"learning_rate": 0.00017060180541624877,
"loss": 2.6745,
"theoretical_loss": 3.4052873673490347,
"tokens_seen": 2185887744
},
{
"epoch": 0.66,
"learning_rate": 0.00017059177532597795,
"loss": 2.5132,
"theoretical_loss": 3.4052790030283107,
"tokens_seen": 2185953280
},
{
"epoch": 0.66,
"learning_rate": 0.00017058174523570713,
"loss": 2.6547,
"theoretical_loss": 3.405270639028562,
"tokens_seen": 2186018816
},
{
"epoch": 0.66,
"learning_rate": 0.0001705717151454363,
"loss": 2.7322,
"theoretical_loss": 3.405262275349766,
"tokens_seen": 2186084352
},
{
"epoch": 0.66,
"learning_rate": 0.00017056168505516552,
"loss": 2.6761,
"theoretical_loss": 3.4052539119919016,
"tokens_seen": 2186149888
},
{
"epoch": 0.66,
"learning_rate": 0.0001705516549648947,
"loss": 2.6113,
"theoretical_loss": 3.4052455489549462,
"tokens_seen": 2186215424
},
{
"epoch": 0.66,
"learning_rate": 0.00017054162487462388,
"loss": 2.5952,
"theoretical_loss": 3.4052371862388786,
"tokens_seen": 2186280960
},
{
"epoch": 0.66,
"learning_rate": 0.00017053159478435306,
"loss": 2.6086,
"theoretical_loss": 3.405228823843676,
"tokens_seen": 2186346496
},
{
"epoch": 0.66,
"learning_rate": 0.00017052156469408224,
"loss": 2.5844,
"theoretical_loss": 3.4052204617693174,
"tokens_seen": 2186412032
},
{
"epoch": 0.66,
"learning_rate": 0.00017051153460381145,
"loss": 2.6803,
"theoretical_loss": 3.4052121000157802,
"tokens_seen": 2186477568
},
{
"epoch": 0.66,
"learning_rate": 0.00017050150451354063,
"loss": 2.7017,
"theoretical_loss": 3.4052037385830425,
"tokens_seen": 2186543104
},
{
"epoch": 0.66,
"learning_rate": 0.0001704914744232698,
"loss": 2.7461,
"theoretical_loss": 3.405195377471083,
"tokens_seen": 2186608640
},
{
"epoch": 0.66,
"learning_rate": 0.000170481444332999,
"loss": 2.6201,
"theoretical_loss": 3.4051870166798786,
"tokens_seen": 2186674176
},
{
"epoch": 0.66,
"learning_rate": 0.0001704714142427282,
"loss": 2.6293,
"theoretical_loss": 3.4051786562094084,
"tokens_seen": 2186739712
},
{
"epoch": 0.66,
"learning_rate": 0.00017046138415245738,
"loss": 2.7808,
"theoretical_loss": 3.4051702960596506,
"tokens_seen": 2186805248
},
{
"epoch": 0.66,
"learning_rate": 0.00017045135406218656,
"loss": 2.5688,
"theoretical_loss": 3.4051619362305825,
"tokens_seen": 2186870784
},
{
"epoch": 0.66,
"learning_rate": 0.00017044132397191574,
"loss": 2.6292,
"theoretical_loss": 3.4051535767221823,
"tokens_seen": 2186936320
},
{
"epoch": 0.66,
"learning_rate": 0.00017043129388164493,
"loss": 2.4659,
"theoretical_loss": 3.405145217534429,
"tokens_seen": 2187001856
},
{
"epoch": 0.66,
"learning_rate": 0.00017042126379137413,
"loss": 2.8742,
"theoretical_loss": 3.4051368586672996,
"tokens_seen": 2187067392
},
{
"epoch": 0.66,
"learning_rate": 0.00017041123370110332,
"loss": 2.6781,
"theoretical_loss": 3.4051285001207727,
"tokens_seen": 2187132928
},
{
"epoch": 0.66,
"learning_rate": 0.0001704012036108325,
"loss": 2.5837,
"theoretical_loss": 3.4051201418948263,
"tokens_seen": 2187198464
},
{
"epoch": 0.66,
"objective/train/docs_used": 1239462,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.359999656677246,
"objective/train/theoretical_loss": 3.4051117839894385,
"objective/train/tokens_used": 2207724000,
"theoretical_loss": 3.4051117839894385,
"tokens_seen": 2187264000
},
{
"epoch": 0.66,
"learning_rate": 0.00017039117352056168,
"loss": 2.5545,
"theoretical_loss": 3.4051117839894385,
"tokens_seen": 2187264000
},
{
"epoch": 0.66,
"learning_rate": 0.00017038114343029089,
"loss": 2.7622,
"theoretical_loss": 3.4051034264045876,
"tokens_seen": 2187329536
},
{
"epoch": 0.66,
"learning_rate": 0.00017037111334002007,
"loss": 2.3982,
"theoretical_loss": 3.4050950691402515,
"tokens_seen": 2187395072
},
{
"epoch": 0.66,
"learning_rate": 0.00017036108324974925,
"loss": 2.6087,
"theoretical_loss": 3.4050867121964083,
"tokens_seen": 2187460608
},
{
"epoch": 0.66,
"learning_rate": 0.00017035105315947843,
"loss": 2.6781,
"theoretical_loss": 3.4050783555730364,
"tokens_seen": 2187526144
},
{
"epoch": 0.66,
"learning_rate": 0.0001703410230692076,
"loss": 2.6026,
"theoretical_loss": 3.4050699992701134,
"tokens_seen": 2187591680
},
{
"epoch": 0.66,
"learning_rate": 0.00017033099297893682,
"loss": 2.5763,
"theoretical_loss": 3.4050616432876177,
"tokens_seen": 2187657216
},
{
"epoch": 0.66,
"learning_rate": 0.000170320962888666,
"loss": 2.7582,
"theoretical_loss": 3.4050532876255275,
"tokens_seen": 2187722752
},
{
"epoch": 0.66,
"learning_rate": 0.00017031093279839518,
"loss": 2.6501,
"theoretical_loss": 3.405044932283821,
"tokens_seen": 2187788288
},
{
"epoch": 0.66,
"learning_rate": 0.00017030090270812436,
"loss": 2.9003,
"theoretical_loss": 3.405036577262476,
"tokens_seen": 2187853824
},
{
"epoch": 0.66,
"learning_rate": 0.00017029087261785357,
"loss": 2.6015,
"theoretical_loss": 3.4050282225614703,
"tokens_seen": 2187919360
},
{
"epoch": 0.66,
"learning_rate": 0.00017028084252758275,
"loss": 2.7912,
"theoretical_loss": 3.405019868180783,
"tokens_seen": 2187984896
},
{
"epoch": 0.66,
"learning_rate": 0.00017027081243731193,
"loss": 2.451,
"theoretical_loss": 3.4050115141203916,
"tokens_seen": 2188050432
},
{
"epoch": 0.66,
"learning_rate": 0.0001702607823470411,
"loss": 2.3967,
"theoretical_loss": 3.405003160380274,
"tokens_seen": 2188115968
},
{
"epoch": 0.66,
"learning_rate": 0.00017025075225677032,
"loss": 2.7059,
"theoretical_loss": 3.4049948069604095,
"tokens_seen": 2188181504
},
{
"epoch": 0.66,
"learning_rate": 0.0001702407221664995,
"loss": 2.6276,
"theoretical_loss": 3.404986453860775,
"tokens_seen": 2188247040
},
{
"epoch": 0.66,
"learning_rate": 0.00017023069207622868,
"loss": 2.6246,
"theoretical_loss": 3.4049781010813485,
"tokens_seen": 2188312576
},
{
"epoch": 0.66,
"learning_rate": 0.00017022066198595786,
"loss": 2.9552,
"theoretical_loss": 3.404969748622109,
"tokens_seen": 2188378112
},
{
"epoch": 0.66,
"learning_rate": 0.00017021063189568705,
"loss": 2.5498,
"theoretical_loss": 3.404961396483034,
"tokens_seen": 2188443648
},
{
"epoch": 0.66,
"learning_rate": 0.00017020060180541625,
"loss": 2.5769,
"theoretical_loss": 3.4049530446641025,
"tokens_seen": 2188509184
},
{
"epoch": 0.66,
"learning_rate": 0.00017019057171514544,
"loss": 2.682,
"theoretical_loss": 3.404944693165292,
"tokens_seen": 2188574720
},
{
"epoch": 0.66,
"learning_rate": 0.00017018054162487462,
"loss": 2.5997,
"theoretical_loss": 3.4049363419865806,
"tokens_seen": 2188640256
},
{
"epoch": 0.66,
"learning_rate": 0.0001701705115346038,
"loss": 2.6474,
"theoretical_loss": 3.404927991127946,
"tokens_seen": 2188705792
},
{
"epoch": 0.66,
"learning_rate": 0.000170160481444333,
"loss": 2.627,
"theoretical_loss": 3.404919640589368,
"tokens_seen": 2188771328
},
{
"epoch": 0.66,
"learning_rate": 0.0001701504513540622,
"loss": 2.5888,
"theoretical_loss": 3.4049112903708227,
"tokens_seen": 2188836864
},
{
"epoch": 0.66,
"objective/train/docs_used": 1240051,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7275075912475586,
"objective/train/theoretical_loss": 3.4049029404722897,
"objective/train/tokens_used": 2209362400,
"theoretical_loss": 3.4049029404722897,
"tokens_seen": 2188902400
},
{
"epoch": 0.66,
"learning_rate": 0.00017014042126379137,
"loss": 2.6758,
"theoretical_loss": 3.4049029404722897,
"tokens_seen": 2188902400
},
{
"epoch": 0.66,
"learning_rate": 0.00017013039117352055,
"loss": 2.4749,
"theoretical_loss": 3.404894590893747,
"tokens_seen": 2188967936
},
{
"epoch": 0.66,
"learning_rate": 0.00017012036108324973,
"loss": 2.8934,
"theoretical_loss": 3.4048862416351717,
"tokens_seen": 2189033472
},
{
"epoch": 0.66,
"learning_rate": 0.00017011033099297894,
"loss": 2.7144,
"theoretical_loss": 3.4048778926965433,
"tokens_seen": 2189099008
},
{
"epoch": 0.66,
"learning_rate": 0.00017010030090270812,
"loss": 2.4533,
"theoretical_loss": 3.404869544077839,
"tokens_seen": 2189164544
},
{
"epoch": 0.66,
"learning_rate": 0.0001700902708124373,
"loss": 2.4072,
"theoretical_loss": 3.404861195779038,
"tokens_seen": 2189230080
},
{
"epoch": 0.66,
"learning_rate": 0.00017008024072216648,
"loss": 2.492,
"theoretical_loss": 3.4048528478001168,
"tokens_seen": 2189295616
},
{
"epoch": 0.66,
"learning_rate": 0.0001700702106318957,
"loss": 2.6187,
"theoretical_loss": 3.404844500141055,
"tokens_seen": 2189361152
},
{
"epoch": 0.66,
"learning_rate": 0.00017006018054162487,
"loss": 2.6326,
"theoretical_loss": 3.4048361528018303,
"tokens_seen": 2189426688
},
{
"epoch": 0.66,
"learning_rate": 0.00017005015045135405,
"loss": 2.5849,
"theoretical_loss": 3.404827805782421,
"tokens_seen": 2189492224
},
{
"epoch": 0.66,
"learning_rate": 0.00017004012036108326,
"loss": 2.6112,
"theoretical_loss": 3.404819459082805,
"tokens_seen": 2189557760
},
{
"epoch": 0.66,
"learning_rate": 0.00017003009027081244,
"loss": 2.5822,
"theoretical_loss": 3.4048111127029603,
"tokens_seen": 2189623296
},
{
"epoch": 0.66,
"learning_rate": 0.00017002006018054165,
"loss": 2.8656,
"theoretical_loss": 3.404802766642866,
"tokens_seen": 2189688832
},
{
"epoch": 0.66,
"learning_rate": 0.00017001003009027083,
"loss": 2.7741,
"theoretical_loss": 3.4047944209025,
"tokens_seen": 2189754368
},
{
"epoch": 0.66,
"learning_rate": 0.00017,
"loss": 2.5979,
"theoretical_loss": 3.4047860754818395,
"tokens_seen": 2189819904
},
{
"epoch": 0.66,
"learning_rate": 0.0001699899699097292,
"loss": 2.6725,
"theoretical_loss": 3.4047777303808635,
"tokens_seen": 2189885440
},
{
"epoch": 0.66,
"learning_rate": 0.0001699799398194584,
"loss": 2.5411,
"theoretical_loss": 3.4047693855995504,
"tokens_seen": 2189950976
},
{
"epoch": 0.66,
"learning_rate": 0.00016996990972918758,
"loss": 2.5308,
"theoretical_loss": 3.4047610411378777,
"tokens_seen": 2190016512
},
{
"epoch": 0.66,
"learning_rate": 0.00016995987963891676,
"loss": 2.749,
"theoretical_loss": 3.404752696995824,
"tokens_seen": 2190082048
},
{
"epoch": 0.66,
"learning_rate": 0.00016994984954864594,
"loss": 2.5983,
"theoretical_loss": 3.404744353173368,
"tokens_seen": 2190147584
},
{
"epoch": 0.66,
"learning_rate": 0.00016993981945837513,
"loss": 2.7556,
"theoretical_loss": 3.4047360096704864,
"tokens_seen": 2190213120
},
{
"epoch": 0.66,
"learning_rate": 0.00016992978936810433,
"loss": 2.6164,
"theoretical_loss": 3.404727666487159,
"tokens_seen": 2190278656
},
{
"epoch": 0.66,
"learning_rate": 0.00016991975927783352,
"loss": 2.7097,
"theoretical_loss": 3.4047193236233633,
"tokens_seen": 2190344192
},
{
"epoch": 0.66,
"learning_rate": 0.0001699097291875627,
"loss": 2.339,
"theoretical_loss": 3.404710981079077,
"tokens_seen": 2190409728
},
{
"epoch": 0.66,
"learning_rate": 0.00016989969909729188,
"loss": 2.8372,
"theoretical_loss": 3.4047026388542796,
"tokens_seen": 2190475264
},
{
"epoch": 0.66,
"objective/train/docs_used": 1241523,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.893355369567871,
"objective/train/theoretical_loss": 3.404694296948948,
"objective/train/tokens_used": 2211000800,
"theoretical_loss": 3.404694296948948,
"tokens_seen": 2190540800
},
{
"epoch": 0.66,
"learning_rate": 0.00016988966900702109,
"loss": 2.5169,
"theoretical_loss": 3.404694296948948,
"tokens_seen": 2190540800
},
{
"epoch": 0.66,
"learning_rate": 0.00016987963891675027,
"loss": 2.5234,
"theoretical_loss": 3.4046859553630613,
"tokens_seen": 2190606336
},
{
"epoch": 0.66,
"learning_rate": 0.00016986960882647945,
"loss": 2.577,
"theoretical_loss": 3.4046776140965975,
"tokens_seen": 2190671872
},
{
"epoch": 0.66,
"learning_rate": 0.00016985957873620863,
"loss": 2.7572,
"theoretical_loss": 3.4046692731495343,
"tokens_seen": 2190737408
},
{
"epoch": 0.66,
"learning_rate": 0.0001698495486459378,
"loss": 2.7481,
"theoretical_loss": 3.40466093252185,
"tokens_seen": 2190802944
},
{
"epoch": 0.66,
"learning_rate": 0.00016983951855566702,
"loss": 2.5976,
"theoretical_loss": 3.404652592213524,
"tokens_seen": 2190868480
},
{
"epoch": 0.66,
"learning_rate": 0.0001698294884653962,
"loss": 2.5028,
"theoretical_loss": 3.404644252224533,
"tokens_seen": 2190934016
},
{
"epoch": 0.66,
"learning_rate": 0.00016981945837512538,
"loss": 2.512,
"theoretical_loss": 3.404635912554856,
"tokens_seen": 2190999552
},
{
"epoch": 0.66,
"learning_rate": 0.00016980942828485456,
"loss": 2.577,
"theoretical_loss": 3.404627573204471,
"tokens_seen": 2191065088
},
{
"epoch": 0.66,
"learning_rate": 0.00016979939819458377,
"loss": 2.7565,
"theoretical_loss": 3.4046192341733565,
"tokens_seen": 2191130624
},
{
"epoch": 0.66,
"learning_rate": 0.00016978936810431295,
"loss": 2.6776,
"theoretical_loss": 3.4046108954614906,
"tokens_seen": 2191196160
},
{
"epoch": 0.66,
"learning_rate": 0.00016977933801404213,
"loss": 2.5877,
"theoretical_loss": 3.404602557068851,
"tokens_seen": 2191261696
},
{
"epoch": 0.66,
"learning_rate": 0.0001697693079237713,
"loss": 2.8046,
"theoretical_loss": 3.404594218995417,
"tokens_seen": 2191327232
},
{
"epoch": 0.66,
"learning_rate": 0.00016975927783350052,
"loss": 2.6643,
"theoretical_loss": 3.4045858812411662,
"tokens_seen": 2191392768
},
{
"epoch": 0.66,
"learning_rate": 0.0001697492477432297,
"loss": 2.5789,
"theoretical_loss": 3.4045775438060764,
"tokens_seen": 2191458304
},
{
"epoch": 0.66,
"learning_rate": 0.00016973921765295888,
"loss": 2.6565,
"theoretical_loss": 3.4045692066901263,
"tokens_seen": 2191523840
},
{
"epoch": 0.66,
"learning_rate": 0.00016972918756268807,
"loss": 2.7082,
"theoretical_loss": 3.4045608698932943,
"tokens_seen": 2191589376
},
{
"epoch": 0.66,
"learning_rate": 0.00016971915747241725,
"loss": 2.4756,
"theoretical_loss": 3.4045525334155586,
"tokens_seen": 2191654912
},
{
"epoch": 0.66,
"learning_rate": 0.00016970912738214645,
"loss": 2.5987,
"theoretical_loss": 3.4045441972568975,
"tokens_seen": 2191720448
},
{
"epoch": 0.66,
"learning_rate": 0.00016969909729187564,
"loss": 2.8301,
"theoretical_loss": 3.4045358614172887,
"tokens_seen": 2191785984
},
{
"epoch": 0.66,
"learning_rate": 0.00016968906720160482,
"loss": 2.7687,
"theoretical_loss": 3.404527525896711,
"tokens_seen": 2191851520
},
{
"epoch": 0.66,
"learning_rate": 0.000169679037111334,
"loss": 2.569,
"theoretical_loss": 3.404519190695143,
"tokens_seen": 2191917056
},
{
"epoch": 0.66,
"learning_rate": 0.0001696690070210632,
"loss": 2.5448,
"theoretical_loss": 3.4045108558125614,
"tokens_seen": 2191982592
},
{
"epoch": 0.66,
"learning_rate": 0.0001696589769307924,
"loss": 2.3578,
"theoretical_loss": 3.404502521248946,
"tokens_seen": 2192048128
},
{
"epoch": 0.66,
"learning_rate": 0.00016964894684052157,
"loss": 2.481,
"theoretical_loss": 3.404494187004275,
"tokens_seen": 2192113664
},
{
"epoch": 0.66,
"objective/train/docs_used": 1242176,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6534042358398438,
"objective/train/theoretical_loss": 3.4044858530785254,
"objective/train/tokens_used": 2212639200,
"theoretical_loss": 3.4044858530785254,
"tokens_seen": 2192179200
},
{
"epoch": 0.66,
"learning_rate": 0.00016963891675025075,
"loss": 2.5481,
"theoretical_loss": 3.4044858530785254,
"tokens_seen": 2192179200
},
{
"epoch": 0.66,
"learning_rate": 0.00016962888665997993,
"loss": 2.6063,
"theoretical_loss": 3.4044775194716768,
"tokens_seen": 2192244736
},
{
"epoch": 0.66,
"learning_rate": 0.00016961885656970914,
"loss": 2.5382,
"theoretical_loss": 3.404469186183707,
"tokens_seen": 2192310272
},
{
"epoch": 0.66,
"learning_rate": 0.00016960882647943832,
"loss": 2.4622,
"theoretical_loss": 3.404460853214594,
"tokens_seen": 2192375808
},
{
"epoch": 0.66,
"learning_rate": 0.0001695987963891675,
"loss": 2.8632,
"theoretical_loss": 3.404452520564316,
"tokens_seen": 2192441344
},
{
"epoch": 0.66,
"learning_rate": 0.00016958876629889668,
"loss": 2.5829,
"theoretical_loss": 3.4044441882328518,
"tokens_seen": 2192506880
},
{
"epoch": 0.66,
"learning_rate": 0.0001695787362086259,
"loss": 2.5959,
"theoretical_loss": 3.4044358562201795,
"tokens_seen": 2192572416
},
{
"epoch": 0.66,
"learning_rate": 0.00016956870611835507,
"loss": 2.4053,
"theoretical_loss": 3.404427524526277,
"tokens_seen": 2192637952
},
{
"epoch": 0.66,
"learning_rate": 0.00016955867602808425,
"loss": 2.5612,
"theoretical_loss": 3.4044191931511234,
"tokens_seen": 2192703488
},
{
"epoch": 0.66,
"learning_rate": 0.00016954864593781343,
"loss": 2.5678,
"theoretical_loss": 3.404410862094696,
"tokens_seen": 2192769024
},
{
"epoch": 0.66,
"learning_rate": 0.00016953861584754261,
"loss": 2.6569,
"theoretical_loss": 3.4044025313569737,
"tokens_seen": 2192834560
},
{
"epoch": 0.66,
"learning_rate": 0.00016952858575727182,
"loss": 2.6772,
"theoretical_loss": 3.4043942009379347,
"tokens_seen": 2192900096
},
{
"epoch": 0.66,
"learning_rate": 0.000169518555667001,
"loss": 2.7103,
"theoretical_loss": 3.404385870837557,
"tokens_seen": 2192965632
},
{
"epoch": 0.66,
"learning_rate": 0.00016950852557673019,
"loss": 2.5956,
"theoretical_loss": 3.404377541055819,
"tokens_seen": 2193031168
},
{
"epoch": 0.66,
"learning_rate": 0.00016949849548645937,
"loss": 2.6681,
"theoretical_loss": 3.4043692115926993,
"tokens_seen": 2193096704
},
{
"epoch": 0.66,
"learning_rate": 0.00016948846539618857,
"loss": 2.7757,
"theoretical_loss": 3.4043608824481764,
"tokens_seen": 2193162240
},
{
"epoch": 0.66,
"learning_rate": 0.00016947843530591776,
"loss": 2.5615,
"theoretical_loss": 3.4043525536222274,
"tokens_seen": 2193227776
},
{
"epoch": 0.66,
"learning_rate": 0.00016946840521564694,
"loss": 2.5615,
"theoretical_loss": 3.4043442251148317,
"tokens_seen": 2193293312
},
{
"epoch": 0.66,
"learning_rate": 0.00016945837512537612,
"loss": 2.5911,
"theoretical_loss": 3.404335896925967,
"tokens_seen": 2193358848
},
{
"epoch": 0.66,
"learning_rate": 0.0001694483450351053,
"loss": 2.4555,
"theoretical_loss": 3.4043275690556123,
"tokens_seen": 2193424384
},
{
"epoch": 0.66,
"learning_rate": 0.0001694383149448345,
"loss": 2.824,
"theoretical_loss": 3.404319241503745,
"tokens_seen": 2193489920
},
{
"epoch": 0.66,
"learning_rate": 0.0001694282848545637,
"loss": 2.5315,
"theoretical_loss": 3.404310914270344,
"tokens_seen": 2193555456
},
{
"epoch": 0.66,
"learning_rate": 0.00016941825476429287,
"loss": 2.6233,
"theoretical_loss": 3.404302587355388,
"tokens_seen": 2193620992
},
{
"epoch": 0.66,
"learning_rate": 0.00016940822467402205,
"loss": 2.5537,
"theoretical_loss": 3.404294260758854,
"tokens_seen": 2193686528
},
{
"epoch": 0.66,
"learning_rate": 0.00016939819458375126,
"loss": 2.6546,
"theoretical_loss": 3.4042859344807215,
"tokens_seen": 2193752064
},
{
"epoch": 0.66,
"objective/train/docs_used": 1243185,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2621278762817383,
"objective/train/theoretical_loss": 3.4042776085209683,
"objective/train/tokens_used": 2214277600,
"theoretical_loss": 3.4042776085209683,
"tokens_seen": 2193817600
},
{
"epoch": 0.66,
"learning_rate": 0.00016938816449348044,
"loss": 2.8916,
"theoretical_loss": 3.4042776085209683,
"tokens_seen": 2193817600
},
{
"epoch": 0.66,
"learning_rate": 0.00016937813440320962,
"loss": 2.6665,
"theoretical_loss": 3.404269282879573,
"tokens_seen": 2193883136
},
{
"epoch": 0.66,
"learning_rate": 0.0001693681043129388,
"loss": 2.6777,
"theoretical_loss": 3.4042609575565135,
"tokens_seen": 2193948672
},
{
"epoch": 0.66,
"learning_rate": 0.00016935807422266798,
"loss": 2.699,
"theoretical_loss": 3.404252632551769,
"tokens_seen": 2194014208
},
{
"epoch": 0.66,
"learning_rate": 0.0001693480441323972,
"loss": 2.6771,
"theoretical_loss": 3.404244307865316,
"tokens_seen": 2194079744
},
{
"epoch": 0.66,
"learning_rate": 0.00016933801404212637,
"loss": 2.5982,
"theoretical_loss": 3.4042359834971347,
"tokens_seen": 2194145280
},
{
"epoch": 0.66,
"learning_rate": 0.00016932798395185555,
"loss": 2.7997,
"theoretical_loss": 3.4042276594472027,
"tokens_seen": 2194210816
},
{
"epoch": 0.66,
"learning_rate": 0.00016931795386158473,
"loss": 2.4679,
"theoretical_loss": 3.4042193357154984,
"tokens_seen": 2194276352
},
{
"epoch": 0.66,
"learning_rate": 0.00016930792377131394,
"loss": 2.6776,
"theoretical_loss": 3.404211012302,
"tokens_seen": 2194341888
},
{
"epoch": 0.66,
"learning_rate": 0.00016929789368104312,
"loss": 2.7016,
"theoretical_loss": 3.404202689206686,
"tokens_seen": 2194407424
},
{
"epoch": 0.66,
"learning_rate": 0.00016928786359077233,
"loss": 2.7381,
"theoretical_loss": 3.4041943664295347,
"tokens_seen": 2194472960
},
{
"epoch": 0.67,
"learning_rate": 0.00016927783350050151,
"loss": 2.6596,
"theoretical_loss": 3.4041860439705243,
"tokens_seen": 2194538496
},
{
"epoch": 0.67,
"learning_rate": 0.00016926780341023072,
"loss": 2.515,
"theoretical_loss": 3.404177721829633,
"tokens_seen": 2194604032
},
{
"epoch": 0.67,
"learning_rate": 0.0001692577733199599,
"loss": 2.4929,
"theoretical_loss": 3.4041694000068397,
"tokens_seen": 2194669568
},
{
"epoch": 0.67,
"learning_rate": 0.00016924774322968908,
"loss": 2.6466,
"theoretical_loss": 3.4041610785021224,
"tokens_seen": 2194735104
},
{
"epoch": 0.67,
"learning_rate": 0.00016923771313941827,
"loss": 2.4787,
"theoretical_loss": 3.4041527573154595,
"tokens_seen": 2194800640
},
{
"epoch": 0.67,
"learning_rate": 0.00016922768304914745,
"loss": 2.5728,
"theoretical_loss": 3.404144436446829,
"tokens_seen": 2194866176
},
{
"epoch": 0.67,
"learning_rate": 0.00016921765295887665,
"loss": 2.7183,
"theoretical_loss": 3.40413611589621,
"tokens_seen": 2194931712
},
{
"epoch": 0.67,
"learning_rate": 0.00016920762286860584,
"loss": 2.7259,
"theoretical_loss": 3.40412779566358,
"tokens_seen": 2194997248
},
{
"epoch": 0.67,
"learning_rate": 0.00016919759277833502,
"loss": 2.5947,
"theoretical_loss": 3.4041194757489177,
"tokens_seen": 2195062784
},
{
"epoch": 0.67,
"learning_rate": 0.0001691875626880642,
"loss": 2.5897,
"theoretical_loss": 3.4041111561522017,
"tokens_seen": 2195128320
},
{
"epoch": 0.67,
"learning_rate": 0.0001691775325977934,
"loss": 2.5016,
"theoretical_loss": 3.4041028368734105,
"tokens_seen": 2195193856
},
{
"epoch": 0.67,
"learning_rate": 0.0001691675025075226,
"loss": 2.6553,
"theoretical_loss": 3.4040945179125215,
"tokens_seen": 2195259392
},
{
"epoch": 0.67,
"learning_rate": 0.00016915747241725177,
"loss": 2.6585,
"theoretical_loss": 3.404086199269514,
"tokens_seen": 2195324928
},
{
"epoch": 0.67,
"learning_rate": 0.00016914744232698095,
"loss": 2.5367,
"theoretical_loss": 3.4040778809443664,
"tokens_seen": 2195390464
},
{
"epoch": 0.67,
"objective/train/docs_used": 1243800,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.115630626678467,
"objective/train/theoretical_loss": 3.4040695629370563,
"objective/train/tokens_used": 2215916000,
"theoretical_loss": 3.4040695629370563,
"tokens_seen": 2195456000
},
{
"epoch": 0.67,
"learning_rate": 0.00016913741223671013,
"loss": 2.3451,
"theoretical_loss": 3.4040695629370563,
"tokens_seen": 2195456000
},
{
"epoch": 0.67,
"learning_rate": 0.00016912738214643934,
"loss": 2.3738,
"theoretical_loss": 3.4040612452475627,
"tokens_seen": 2195521536
},
{
"epoch": 0.67,
"learning_rate": 0.00016911735205616852,
"loss": 2.3365,
"theoretical_loss": 3.4040529278758638,
"tokens_seen": 2195587072
},
{
"epoch": 0.67,
"learning_rate": 0.0001691073219658977,
"loss": 2.6812,
"theoretical_loss": 3.4040446108219378,
"tokens_seen": 2195652608
},
{
"epoch": 0.67,
"learning_rate": 0.00016909729187562688,
"loss": 2.6853,
"theoretical_loss": 3.4040362940857634,
"tokens_seen": 2195718144
},
{
"epoch": 0.67,
"learning_rate": 0.0001690872617853561,
"loss": 2.4258,
"theoretical_loss": 3.4040279776673183,
"tokens_seen": 2195783680
},
{
"epoch": 0.67,
"learning_rate": 0.00016907723169508527,
"loss": 2.5751,
"theoretical_loss": 3.404019661566582,
"tokens_seen": 2195849216
},
{
"epoch": 0.67,
"learning_rate": 0.00016906720160481445,
"loss": 2.4419,
"theoretical_loss": 3.4040113457835317,
"tokens_seen": 2195914752
},
{
"epoch": 0.67,
"learning_rate": 0.00016905717151454363,
"loss": 2.633,
"theoretical_loss": 3.4040030303181465,
"tokens_seen": 2195980288
},
{
"epoch": 0.67,
"learning_rate": 0.00016904714142427281,
"loss": 2.634,
"theoretical_loss": 3.403994715170405,
"tokens_seen": 2196045824
},
{
"epoch": 0.67,
"learning_rate": 0.00016903711133400202,
"loss": 2.629,
"theoretical_loss": 3.403986400340285,
"tokens_seen": 2196111360
},
{
"epoch": 0.67,
"learning_rate": 0.0001690270812437312,
"loss": 2.6138,
"theoretical_loss": 3.403978085827765,
"tokens_seen": 2196176896
},
{
"epoch": 0.67,
"learning_rate": 0.00016901705115346039,
"loss": 2.4832,
"theoretical_loss": 3.4039697716328234,
"tokens_seen": 2196242432
},
{
"epoch": 0.67,
"learning_rate": 0.00016900702106318957,
"loss": 2.3563,
"theoretical_loss": 3.4039614577554387,
"tokens_seen": 2196307968
},
{
"epoch": 0.67,
"learning_rate": 0.00016899699097291877,
"loss": 2.6424,
"theoretical_loss": 3.4039531441955897,
"tokens_seen": 2196373504
},
{
"epoch": 0.67,
"learning_rate": 0.00016898696088264796,
"loss": 2.7651,
"theoretical_loss": 3.403944830953254,
"tokens_seen": 2196439040
},
{
"epoch": 0.67,
"learning_rate": 0.00016897693079237714,
"loss": 2.589,
"theoretical_loss": 3.4039365180284102,
"tokens_seen": 2196504576
},
{
"epoch": 0.67,
"learning_rate": 0.00016896690070210632,
"loss": 2.5374,
"theoretical_loss": 3.403928205421037,
"tokens_seen": 2196570112
},
{
"epoch": 0.67,
"learning_rate": 0.0001689568706118355,
"loss": 2.4881,
"theoretical_loss": 3.403919893131113,
"tokens_seen": 2196635648
},
{
"epoch": 0.67,
"learning_rate": 0.0001689468405215647,
"loss": 2.6964,
"theoretical_loss": 3.4039115811586163,
"tokens_seen": 2196701184
},
{
"epoch": 0.67,
"learning_rate": 0.0001689368104312939,
"loss": 2.6705,
"theoretical_loss": 3.403903269503525,
"tokens_seen": 2196766720
},
{
"epoch": 0.67,
"learning_rate": 0.00016892678034102307,
"loss": 2.4666,
"theoretical_loss": 3.403894958165818,
"tokens_seen": 2196832256
},
{
"epoch": 0.67,
"learning_rate": 0.00016891675025075225,
"loss": 2.6425,
"theoretical_loss": 3.4038866471454736,
"tokens_seen": 2196897792
},
{
"epoch": 0.67,
"learning_rate": 0.00016890672016048146,
"loss": 2.8078,
"theoretical_loss": 3.4038783364424696,
"tokens_seen": 2196963328
},
{
"epoch": 0.67,
"learning_rate": 0.00016889669007021064,
"loss": 2.4974,
"theoretical_loss": 3.4038700260567856,
"tokens_seen": 2197028864
},
{
"epoch": 0.67,
"objective/train/docs_used": 1244898,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.569892644882202,
"objective/train/theoretical_loss": 3.4038617159883993,
"objective/train/tokens_used": 2217554400,
"theoretical_loss": 3.4038617159883993,
"tokens_seen": 2197094400
},
{
"epoch": 0.67,
"learning_rate": 0.00016888665997993982,
"loss": 2.7584,
"theoretical_loss": 3.4038617159883993,
"tokens_seen": 2197094400
},
{
"epoch": 0.67,
"learning_rate": 0.000168876629889669,
"loss": 2.918,
"theoretical_loss": 3.403853406237289,
"tokens_seen": 2197159936
},
{
"epoch": 0.67,
"learning_rate": 0.00016886659979939818,
"loss": 2.7737,
"theoretical_loss": 3.4038450968034333,
"tokens_seen": 2197225472
},
{
"epoch": 0.67,
"learning_rate": 0.0001688565697091274,
"loss": 2.9325,
"theoretical_loss": 3.4038367876868105,
"tokens_seen": 2197291008
},
{
"epoch": 0.67,
"learning_rate": 0.00016884653961885657,
"loss": 2.5684,
"theoretical_loss": 3.4038284788873994,
"tokens_seen": 2197356544
},
{
"epoch": 0.67,
"learning_rate": 0.00016883650952858575,
"loss": 2.9598,
"theoretical_loss": 3.403820170405178,
"tokens_seen": 2197422080
},
{
"epoch": 0.67,
"learning_rate": 0.00016882647943831494,
"loss": 2.6373,
"theoretical_loss": 3.403811862240125,
"tokens_seen": 2197487616
},
{
"epoch": 0.67,
"learning_rate": 0.00016881644934804414,
"loss": 2.8173,
"theoretical_loss": 3.403803554392219,
"tokens_seen": 2197553152
},
{
"epoch": 0.67,
"learning_rate": 0.00016880641925777332,
"loss": 2.5885,
"theoretical_loss": 3.403795246861438,
"tokens_seen": 2197618688
},
{
"epoch": 0.67,
"learning_rate": 0.0001687963891675025,
"loss": 2.8074,
"theoretical_loss": 3.4037869396477607,
"tokens_seen": 2197684224
},
{
"epoch": 0.67,
"learning_rate": 0.0001687863590772317,
"loss": 2.6254,
"theoretical_loss": 3.4037786327511657,
"tokens_seen": 2197749760
},
{
"epoch": 0.67,
"learning_rate": 0.00016877632898696087,
"loss": 2.5983,
"theoretical_loss": 3.4037703261716308,
"tokens_seen": 2197815296
},
{
"epoch": 0.67,
"learning_rate": 0.00016876629889669008,
"loss": 2.7866,
"theoretical_loss": 3.403762019909135,
"tokens_seen": 2197880832
},
{
"epoch": 0.67,
"learning_rate": 0.00016875626880641926,
"loss": 2.6434,
"theoretical_loss": 3.4037537139636567,
"tokens_seen": 2197946368
},
{
"epoch": 0.67,
"learning_rate": 0.00016874623871614844,
"loss": 2.6339,
"theoretical_loss": 3.4037454083351744,
"tokens_seen": 2198011904
},
{
"epoch": 0.67,
"learning_rate": 0.00016873620862587762,
"loss": 2.7315,
"theoretical_loss": 3.403737103023666,
"tokens_seen": 2198077440
},
{
"epoch": 0.67,
"learning_rate": 0.00016872617853560683,
"loss": 2.5364,
"theoretical_loss": 3.403728798029111,
"tokens_seen": 2198142976
},
{
"epoch": 0.67,
"learning_rate": 0.000168716148445336,
"loss": 2.6764,
"theoretical_loss": 3.4037204933514866,
"tokens_seen": 2198208512
},
{
"epoch": 0.67,
"learning_rate": 0.0001687061183550652,
"loss": 2.6329,
"theoretical_loss": 3.403712188990772,
"tokens_seen": 2198274048
},
{
"epoch": 0.67,
"learning_rate": 0.00016869608826479437,
"loss": 2.7673,
"theoretical_loss": 3.403703884946946,
"tokens_seen": 2198339584
},
{
"epoch": 0.67,
"learning_rate": 0.00016868605817452355,
"loss": 2.5392,
"theoretical_loss": 3.4036955812199863,
"tokens_seen": 2198405120
},
{
"epoch": 0.67,
"learning_rate": 0.00016867602808425276,
"loss": 2.4961,
"theoretical_loss": 3.4036872778098717,
"tokens_seen": 2198470656
},
{
"epoch": 0.67,
"learning_rate": 0.00016866599799398194,
"loss": 2.6968,
"theoretical_loss": 3.4036789747165805,
"tokens_seen": 2198536192
},
{
"epoch": 0.67,
"learning_rate": 0.00016865596790371112,
"loss": 2.5497,
"theoretical_loss": 3.403670671940091,
"tokens_seen": 2198601728
},
{
"epoch": 0.67,
"learning_rate": 0.0001686459378134403,
"loss": 2.6262,
"theoretical_loss": 3.403662369480383,
"tokens_seen": 2198667264
},
{
"epoch": 0.67,
"objective/train/docs_used": 1245692,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8267266750335693,
"objective/train/theoretical_loss": 3.403654067337433,
"objective/train/tokens_used": 2219192800,
"theoretical_loss": 3.403654067337433,
"tokens_seen": 2198732800
},
{
"epoch": 0.67,
"learning_rate": 0.0001686359077231695,
"loss": 2.6625,
"theoretical_loss": 3.403654067337433,
"tokens_seen": 2198732800
},
{
"epoch": 0.67,
"learning_rate": 0.0001686258776328987,
"loss": 2.6199,
"theoretical_loss": 3.4036457655112207,
"tokens_seen": 2198798336
},
{
"epoch": 0.67,
"learning_rate": 0.00016861584754262787,
"loss": 2.5458,
"theoretical_loss": 3.403637464001724,
"tokens_seen": 2198863872
},
{
"epoch": 0.67,
"learning_rate": 0.00016860581745235706,
"loss": 2.6421,
"theoretical_loss": 3.4036291628089224,
"tokens_seen": 2198929408
},
{
"epoch": 0.67,
"learning_rate": 0.00016859578736208626,
"loss": 2.8342,
"theoretical_loss": 3.403620861932793,
"tokens_seen": 2198994944
},
{
"epoch": 0.67,
"learning_rate": 0.00016858575727181544,
"loss": 2.3107,
"theoretical_loss": 3.4036125613733152,
"tokens_seen": 2199060480
},
{
"epoch": 0.67,
"learning_rate": 0.00016857572718154463,
"loss": 2.6717,
"theoretical_loss": 3.403604261130467,
"tokens_seen": 2199126016
},
{
"epoch": 0.67,
"learning_rate": 0.0001685656970912738,
"loss": 2.7143,
"theoretical_loss": 3.4035959612042275,
"tokens_seen": 2199191552
},
{
"epoch": 0.67,
"learning_rate": 0.000168555667001003,
"loss": 2.6003,
"theoretical_loss": 3.4035876615945746,
"tokens_seen": 2199257088
},
{
"epoch": 0.67,
"learning_rate": 0.0001685456369107322,
"loss": 2.6276,
"theoretical_loss": 3.4035793623014867,
"tokens_seen": 2199322624
},
{
"epoch": 0.67,
"learning_rate": 0.0001685356068204614,
"loss": 2.7264,
"theoretical_loss": 3.403571063324943,
"tokens_seen": 2199388160
},
{
"epoch": 0.67,
"learning_rate": 0.00016852557673019059,
"loss": 2.7234,
"theoretical_loss": 3.403562764664921,
"tokens_seen": 2199453696
},
{
"epoch": 0.67,
"learning_rate": 0.00016851554663991977,
"loss": 2.6821,
"theoretical_loss": 3.4035544663214004,
"tokens_seen": 2199519232
},
{
"epoch": 0.67,
"learning_rate": 0.00016850551654964897,
"loss": 2.4969,
"theoretical_loss": 3.403546168294359,
"tokens_seen": 2199584768
},
{
"epoch": 0.67,
"learning_rate": 0.00016849548645937816,
"loss": 2.4979,
"theoretical_loss": 3.403537870583775,
"tokens_seen": 2199650304
},
{
"epoch": 0.67,
"learning_rate": 0.00016848545636910734,
"loss": 2.5943,
"theoretical_loss": 3.4035295731896276,
"tokens_seen": 2199715840
},
{
"epoch": 0.67,
"learning_rate": 0.00016847542627883652,
"loss": 2.8006,
"theoretical_loss": 3.4035212761118947,
"tokens_seen": 2199781376
},
{
"epoch": 0.67,
"learning_rate": 0.0001684653961885657,
"loss": 2.739,
"theoretical_loss": 3.4035129793505554,
"tokens_seen": 2199846912
},
{
"epoch": 0.67,
"learning_rate": 0.0001684553660982949,
"loss": 2.8127,
"theoretical_loss": 3.403504682905588,
"tokens_seen": 2199912448
},
{
"epoch": 0.67,
"learning_rate": 0.0001684453360080241,
"loss": 2.5552,
"theoretical_loss": 3.40349638677697,
"tokens_seen": 2199977984
},
{
"epoch": 0.67,
"learning_rate": 0.00016843530591775327,
"loss": 2.6895,
"theoretical_loss": 3.403488090964682,
"tokens_seen": 2200043520
},
{
"epoch": 0.67,
"learning_rate": 0.00016842527582748245,
"loss": 2.8276,
"theoretical_loss": 3.4034797954687006,
"tokens_seen": 2200109056
},
{
"epoch": 0.67,
"learning_rate": 0.00016841524573721166,
"loss": 2.4559,
"theoretical_loss": 3.403471500289005,
"tokens_seen": 2200174592
},
{
"epoch": 0.67,
"learning_rate": 0.00016840521564694084,
"loss": 2.7295,
"theoretical_loss": 3.4034632054255742,
"tokens_seen": 2200240128
},
{
"epoch": 0.67,
"learning_rate": 0.00016839518555667002,
"loss": 2.5812,
"theoretical_loss": 3.403454910878386,
"tokens_seen": 2200305664
},
{
"epoch": 0.67,
"objective/train/docs_used": 1246987,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.8895443677902222,
"objective/train/theoretical_loss": 3.4034466166474195,
"objective/train/tokens_used": 2220831200,
"theoretical_loss": 3.4034466166474195,
"tokens_seen": 2200371200
},
{
"epoch": 0.67,
"learning_rate": 0.0001683851554663992,
"loss": 2.5704,
"theoretical_loss": 3.4034466166474195,
"tokens_seen": 2200371200
},
{
"epoch": 0.67,
"learning_rate": 0.00016837512537612838,
"loss": 2.7951,
"theoretical_loss": 3.4034383227326526,
"tokens_seen": 2200436736
},
{
"epoch": 0.67,
"learning_rate": 0.0001683650952858576,
"loss": 2.6233,
"theoretical_loss": 3.4034300291340647,
"tokens_seen": 2200502272
},
{
"epoch": 0.67,
"learning_rate": 0.00016835506519558677,
"loss": 2.5604,
"theoretical_loss": 3.403421735851633,
"tokens_seen": 2200567808
},
{
"epoch": 0.67,
"learning_rate": 0.00016834503510531595,
"loss": 2.7977,
"theoretical_loss": 3.4034134428853378,
"tokens_seen": 2200633344
},
{
"epoch": 0.67,
"learning_rate": 0.00016833500501504514,
"loss": 2.4156,
"theoretical_loss": 3.403405150235156,
"tokens_seen": 2200698880
},
{
"epoch": 0.67,
"learning_rate": 0.00016832497492477434,
"loss": 2.4242,
"theoretical_loss": 3.403396857901067,
"tokens_seen": 2200764416
},
{
"epoch": 0.67,
"learning_rate": 0.00016831494483450352,
"loss": 2.4588,
"theoretical_loss": 3.4033885658830494,
"tokens_seen": 2200829952
},
{
"epoch": 0.67,
"learning_rate": 0.0001683049147442327,
"loss": 2.6253,
"theoretical_loss": 3.403380274181081,
"tokens_seen": 2200895488
},
{
"epoch": 0.67,
"learning_rate": 0.0001682948846539619,
"loss": 2.8183,
"theoretical_loss": 3.403371982795141,
"tokens_seen": 2200961024
},
{
"epoch": 0.67,
"learning_rate": 0.00016828485456369107,
"loss": 2.5162,
"theoretical_loss": 3.403363691725208,
"tokens_seen": 2201026560
},
{
"epoch": 0.67,
"learning_rate": 0.00016827482447342028,
"loss": 2.5569,
"theoretical_loss": 3.40335540097126,
"tokens_seen": 2201092096
},
{
"epoch": 0.67,
"learning_rate": 0.00016826479438314946,
"loss": 2.5456,
"theoretical_loss": 3.403347110533276,
"tokens_seen": 2201157632
},
{
"epoch": 0.67,
"learning_rate": 0.00016825476429287864,
"loss": 2.5551,
"theoretical_loss": 3.4033388204112347,
"tokens_seen": 2201223168
},
{
"epoch": 0.67,
"learning_rate": 0.00016824473420260782,
"loss": 2.7831,
"theoretical_loss": 3.403330530605114,
"tokens_seen": 2201288704
},
{
"epoch": 0.67,
"learning_rate": 0.00016823470411233703,
"loss": 2.7611,
"theoretical_loss": 3.403322241114893,
"tokens_seen": 2201354240
},
{
"epoch": 0.67,
"learning_rate": 0.0001682246740220662,
"loss": 2.8999,
"theoretical_loss": 3.40331395194055,
"tokens_seen": 2201419776
},
{
"epoch": 0.67,
"learning_rate": 0.0001682146439317954,
"loss": 2.6527,
"theoretical_loss": 3.403305663082064,
"tokens_seen": 2201485312
},
{
"epoch": 0.67,
"learning_rate": 0.00016820461384152457,
"loss": 2.764,
"theoretical_loss": 3.403297374539413,
"tokens_seen": 2201550848
},
{
"epoch": 0.67,
"learning_rate": 0.00016819458375125375,
"loss": 2.7001,
"theoretical_loss": 3.4032890863125758,
"tokens_seen": 2201616384
},
{
"epoch": 0.67,
"learning_rate": 0.00016818455366098296,
"loss": 2.624,
"theoretical_loss": 3.403280798401531,
"tokens_seen": 2201681920
},
{
"epoch": 0.67,
"learning_rate": 0.00016817452357071214,
"loss": 2.7296,
"theoretical_loss": 3.403272510806257,
"tokens_seen": 2201747456
},
{
"epoch": 0.67,
"learning_rate": 0.00016816449348044132,
"loss": 2.5265,
"theoretical_loss": 3.4032642235267327,
"tokens_seen": 2201812992
},
{
"epoch": 0.67,
"learning_rate": 0.0001681544633901705,
"loss": 2.5131,
"theoretical_loss": 3.4032559365629362,
"tokens_seen": 2201878528
},
{
"epoch": 0.67,
"learning_rate": 0.0001681444332998997,
"loss": 2.6166,
"theoretical_loss": 3.4032476499148467,
"tokens_seen": 2201944064
},
{
"epoch": 0.67,
"objective/train/docs_used": 1247624,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.593736171722412,
"objective/train/theoretical_loss": 3.4032393635824416,
"objective/train/tokens_used": 2222469600,
"theoretical_loss": 3.4032393635824416,
"tokens_seen": 2202009600
},
{
"epoch": 0.67,
"learning_rate": 0.0001681344032096289,
"loss": 2.7114,
"theoretical_loss": 3.4032393635824416,
"tokens_seen": 2202009600
},
{
"epoch": 0.67,
"learning_rate": 0.00016812437311935807,
"loss": 2.4779,
"theoretical_loss": 3.4032310775657013,
"tokens_seen": 2202075136
},
{
"epoch": 0.67,
"learning_rate": 0.00016811434302908726,
"loss": 2.7311,
"theoretical_loss": 3.4032227918646027,
"tokens_seen": 2202140672
},
{
"epoch": 0.67,
"learning_rate": 0.00016810431293881646,
"loss": 2.5685,
"theoretical_loss": 3.4032145064791255,
"tokens_seen": 2202206208
},
{
"epoch": 0.67,
"learning_rate": 0.00016809428284854564,
"loss": 2.7959,
"theoretical_loss": 3.4032062214092473,
"tokens_seen": 2202271744
},
{
"epoch": 0.67,
"learning_rate": 0.00016808425275827483,
"loss": 2.7044,
"theoretical_loss": 3.403197936654948,
"tokens_seen": 2202337280
},
{
"epoch": 0.67,
"learning_rate": 0.000168074222668004,
"loss": 2.4229,
"theoretical_loss": 3.403189652216205,
"tokens_seen": 2202402816
},
{
"epoch": 0.67,
"learning_rate": 0.0001680641925777332,
"loss": 2.3409,
"theoretical_loss": 3.403181368092997,
"tokens_seen": 2202468352
},
{
"epoch": 0.67,
"learning_rate": 0.0001680541624874624,
"loss": 2.6024,
"theoretical_loss": 3.4031730842853034,
"tokens_seen": 2202533888
},
{
"epoch": 0.67,
"learning_rate": 0.00016804413239719158,
"loss": 2.6587,
"theoretical_loss": 3.403164800793102,
"tokens_seen": 2202599424
},
{
"epoch": 0.67,
"learning_rate": 0.00016803410230692076,
"loss": 2.7411,
"theoretical_loss": 3.4031565176163716,
"tokens_seen": 2202664960
},
{
"epoch": 0.67,
"learning_rate": 0.00016802407221664994,
"loss": 2.6551,
"theoretical_loss": 3.403148234755091,
"tokens_seen": 2202730496
},
{
"epoch": 0.67,
"learning_rate": 0.00016801404212637915,
"loss": 2.716,
"theoretical_loss": 3.403139952209239,
"tokens_seen": 2202796032
},
{
"epoch": 0.67,
"learning_rate": 0.00016800401203610833,
"loss": 2.522,
"theoretical_loss": 3.4031316699787935,
"tokens_seen": 2202861568
},
{
"epoch": 0.67,
"learning_rate": 0.0001679939819458375,
"loss": 2.7084,
"theoretical_loss": 3.403123388063734,
"tokens_seen": 2202927104
},
{
"epoch": 0.67,
"learning_rate": 0.0001679839518555667,
"loss": 2.5519,
"theoretical_loss": 3.403115106464038,
"tokens_seen": 2202992640
},
{
"epoch": 0.67,
"learning_rate": 0.00016797392176529587,
"loss": 2.563,
"theoretical_loss": 3.403106825179685,
"tokens_seen": 2203058176
},
{
"epoch": 0.67,
"learning_rate": 0.00016796389167502508,
"loss": 2.561,
"theoretical_loss": 3.403098544210654,
"tokens_seen": 2203123712
},
{
"epoch": 0.67,
"learning_rate": 0.00016795386158475426,
"loss": 2.7129,
"theoretical_loss": 3.403090263556922,
"tokens_seen": 2203189248
},
{
"epoch": 0.67,
"learning_rate": 0.00016794383149448344,
"loss": 2.6082,
"theoretical_loss": 3.403081983218469,
"tokens_seen": 2203254784
},
{
"epoch": 0.67,
"learning_rate": 0.00016793380140421262,
"loss": 2.854,
"theoretical_loss": 3.403073703195273,
"tokens_seen": 2203320320
},
{
"epoch": 0.67,
"learning_rate": 0.00016792377131394183,
"loss": 2.4567,
"theoretical_loss": 3.403065423487313,
"tokens_seen": 2203385856
},
{
"epoch": 0.67,
"learning_rate": 0.000167913741223671,
"loss": 2.5555,
"theoretical_loss": 3.4030571440945674,
"tokens_seen": 2203451392
},
{
"epoch": 0.67,
"learning_rate": 0.0001679037111334002,
"loss": 2.7845,
"theoretical_loss": 3.4030488650170145,
"tokens_seen": 2203516928
},
{
"epoch": 0.67,
"learning_rate": 0.00016789368104312938,
"loss": 2.6286,
"theoretical_loss": 3.4030405862546336,
"tokens_seen": 2203582464
},
{
"epoch": 0.67,
"objective/train/docs_used": 1248832,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6334705352783203,
"objective/train/theoretical_loss": 3.403032307807403,
"objective/train/tokens_used": 2224108000,
"theoretical_loss": 3.403032307807403,
"tokens_seen": 2203648000
},
{
"epoch": 0.67,
"learning_rate": 0.00016788365095285856,
"loss": 2.6616,
"theoretical_loss": 3.403032307807403,
"tokens_seen": 2203648000
},
{
"epoch": 0.67,
"learning_rate": 0.00016787362086258776,
"loss": 2.404,
"theoretical_loss": 3.4030240296753016,
"tokens_seen": 2203713536
},
{
"epoch": 0.67,
"learning_rate": 0.00016786359077231695,
"loss": 2.6099,
"theoretical_loss": 3.4030157518583075,
"tokens_seen": 2203779072
},
{
"epoch": 0.67,
"learning_rate": 0.00016785356068204613,
"loss": 2.7397,
"theoretical_loss": 3.4030074743563996,
"tokens_seen": 2203844608
},
{
"epoch": 0.67,
"learning_rate": 0.0001678435305917753,
"loss": 2.8265,
"theoretical_loss": 3.4029991971695566,
"tokens_seen": 2203910144
},
{
"epoch": 0.67,
"learning_rate": 0.00016783350050150452,
"loss": 2.4189,
"theoretical_loss": 3.402990920297757,
"tokens_seen": 2203975680
},
{
"epoch": 0.67,
"learning_rate": 0.0001678234704112337,
"loss": 2.7722,
"theoretical_loss": 3.40298264374098,
"tokens_seen": 2204041216
},
{
"epoch": 0.67,
"learning_rate": 0.00016781344032096288,
"loss": 2.4771,
"theoretical_loss": 3.402974367499203,
"tokens_seen": 2204106752
},
{
"epoch": 0.67,
"learning_rate": 0.00016780341023069206,
"loss": 2.5161,
"theoretical_loss": 3.402966091572406,
"tokens_seen": 2204172288
},
{
"epoch": 0.67,
"learning_rate": 0.00016779338014042124,
"loss": 2.661,
"theoretical_loss": 3.4029578159605665,
"tokens_seen": 2204237824
},
{
"epoch": 0.67,
"learning_rate": 0.00016778335005015048,
"loss": 2.6802,
"theoretical_loss": 3.4029495406636645,
"tokens_seen": 2204303360
},
{
"epoch": 0.67,
"learning_rate": 0.00016777331995987966,
"loss": 2.6968,
"theoretical_loss": 3.4029412656816773,
"tokens_seen": 2204368896
},
{
"epoch": 0.67,
"learning_rate": 0.00016776328986960884,
"loss": 2.5617,
"theoretical_loss": 3.4029329910145845,
"tokens_seen": 2204434432
},
{
"epoch": 0.67,
"learning_rate": 0.00016775325977933802,
"loss": 2.646,
"theoretical_loss": 3.4029247166623637,
"tokens_seen": 2204499968
},
{
"epoch": 0.67,
"learning_rate": 0.00016774322968906723,
"loss": 2.6217,
"theoretical_loss": 3.402916442624995,
"tokens_seen": 2204565504
},
{
"epoch": 0.67,
"learning_rate": 0.0001677331995987964,
"loss": 2.597,
"theoretical_loss": 3.402908168902456,
"tokens_seen": 2204631040
},
{
"epoch": 0.67,
"learning_rate": 0.0001677231695085256,
"loss": 2.5402,
"theoretical_loss": 3.4028998954947256,
"tokens_seen": 2204696576
},
{
"epoch": 0.67,
"learning_rate": 0.00016771313941825477,
"loss": 2.4166,
"theoretical_loss": 3.4028916224017824,
"tokens_seen": 2204762112
},
{
"epoch": 0.67,
"learning_rate": 0.00016770310932798395,
"loss": 2.6201,
"theoretical_loss": 3.4028833496236057,
"tokens_seen": 2204827648
},
{
"epoch": 0.67,
"learning_rate": 0.00016769307923771316,
"loss": 2.5525,
"theoretical_loss": 3.4028750771601732,
"tokens_seen": 2204893184
},
{
"epoch": 0.67,
"learning_rate": 0.00016768304914744234,
"loss": 2.7269,
"theoretical_loss": 3.402866805011464,
"tokens_seen": 2204958720
},
{
"epoch": 0.67,
"learning_rate": 0.00016767301905717152,
"loss": 2.5531,
"theoretical_loss": 3.402858533177457,
"tokens_seen": 2205024256
},
{
"epoch": 0.67,
"learning_rate": 0.0001676629889669007,
"loss": 2.5713,
"theoretical_loss": 3.4028502616581306,
"tokens_seen": 2205089792
},
{
"epoch": 0.67,
"learning_rate": 0.0001676529588766299,
"loss": 2.8122,
"theoretical_loss": 3.4028419904534637,
"tokens_seen": 2205155328
},
{
"epoch": 0.67,
"learning_rate": 0.0001676429287863591,
"loss": 2.7566,
"theoretical_loss": 3.402833719563435,
"tokens_seen": 2205220864
},
{
"epoch": 0.67,
"objective/train/docs_used": 1249474,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7113823890686035,
"objective/train/theoretical_loss": 3.4028254489880227,
"objective/train/tokens_used": 2225746400,
"theoretical_loss": 3.4028254489880227,
"tokens_seen": 2205286400
},
{
"epoch": 0.67,
"learning_rate": 0.00016763289869608827,
"loss": 2.6895,
"theoretical_loss": 3.4028254489880227,
"tokens_seen": 2205286400
},
{
"epoch": 0.67,
"learning_rate": 0.00016762286860581746,
"loss": 2.7755,
"theoretical_loss": 3.4028171787272057,
"tokens_seen": 2205351936
},
{
"epoch": 0.67,
"learning_rate": 0.00016761283851554666,
"loss": 2.4647,
"theoretical_loss": 3.402808908780963,
"tokens_seen": 2205417472
},
{
"epoch": 0.67,
"learning_rate": 0.00016760280842527584,
"loss": 2.7157,
"theoretical_loss": 3.402800639149273,
"tokens_seen": 2205483008
},
{
"epoch": 0.67,
"learning_rate": 0.00016759277833500503,
"loss": 2.2956,
"theoretical_loss": 3.4027923698321145,
"tokens_seen": 2205548544
},
{
"epoch": 0.67,
"learning_rate": 0.0001675827482447342,
"loss": 2.6861,
"theoretical_loss": 3.402784100829466,
"tokens_seen": 2205614080
},
{
"epoch": 0.67,
"learning_rate": 0.0001675727181544634,
"loss": 2.7458,
"theoretical_loss": 3.402775832141307,
"tokens_seen": 2205679616
},
{
"epoch": 0.67,
"learning_rate": 0.0001675626880641926,
"loss": 2.7516,
"theoretical_loss": 3.4027675637676147,
"tokens_seen": 2205745152
},
{
"epoch": 0.67,
"learning_rate": 0.00016755265797392178,
"loss": 2.7115,
"theoretical_loss": 3.402759295708369,
"tokens_seen": 2205810688
},
{
"epoch": 0.67,
"learning_rate": 0.00016754262788365096,
"loss": 2.5623,
"theoretical_loss": 3.4027510279635482,
"tokens_seen": 2205876224
},
{
"epoch": 0.67,
"learning_rate": 0.00016753259779338014,
"loss": 2.4431,
"theoretical_loss": 3.402742760533131,
"tokens_seen": 2205941760
},
{
"epoch": 0.67,
"learning_rate": 0.00016752256770310935,
"loss": 2.5018,
"theoretical_loss": 3.402734493417096,
"tokens_seen": 2206007296
},
{
"epoch": 0.67,
"learning_rate": 0.00016751253761283853,
"loss": 2.7763,
"theoretical_loss": 3.402726226615423,
"tokens_seen": 2206072832
},
{
"epoch": 0.67,
"learning_rate": 0.0001675025075225677,
"loss": 2.4671,
"theoretical_loss": 3.4027179601280886,
"tokens_seen": 2206138368
},
{
"epoch": 0.67,
"learning_rate": 0.0001674924774322969,
"loss": 2.6445,
"theoretical_loss": 3.402709693955073,
"tokens_seen": 2206203904
},
{
"epoch": 0.67,
"learning_rate": 0.00016748244734202607,
"loss": 2.5929,
"theoretical_loss": 3.402701428096355,
"tokens_seen": 2206269440
},
{
"epoch": 0.67,
"learning_rate": 0.00016747241725175528,
"loss": 2.5056,
"theoretical_loss": 3.402693162551912,
"tokens_seen": 2206334976
},
{
"epoch": 0.67,
"learning_rate": 0.00016746238716148446,
"loss": 2.8821,
"theoretical_loss": 3.4026848973217243,
"tokens_seen": 2206400512
},
{
"epoch": 0.67,
"learning_rate": 0.00016745235707121364,
"loss": 2.752,
"theoretical_loss": 3.40267663240577,
"tokens_seen": 2206466048
},
{
"epoch": 0.67,
"learning_rate": 0.00016744232698094282,
"loss": 2.5964,
"theoretical_loss": 3.4026683678040275,
"tokens_seen": 2206531584
},
{
"epoch": 0.67,
"learning_rate": 0.00016743229689067203,
"loss": 2.7902,
"theoretical_loss": 3.402660103516476,
"tokens_seen": 2206597120
},
{
"epoch": 0.67,
"learning_rate": 0.0001674222668004012,
"loss": 2.6607,
"theoretical_loss": 3.402651839543094,
"tokens_seen": 2206662656
},
{
"epoch": 0.67,
"learning_rate": 0.0001674122367101304,
"loss": 2.5376,
"theoretical_loss": 3.4026435758838596,
"tokens_seen": 2206728192
},
{
"epoch": 0.67,
"learning_rate": 0.00016740220661985958,
"loss": 2.8959,
"theoretical_loss": 3.4026353125387527,
"tokens_seen": 2206793728
},
{
"epoch": 0.67,
"learning_rate": 0.00016739217652958876,
"loss": 2.6611,
"theoretical_loss": 3.4026270495077515,
"tokens_seen": 2206859264
},
{
"epoch": 0.67,
"objective/train/docs_used": 1250567,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9655425548553467,
"objective/train/theoretical_loss": 3.4026187867908346,
"objective/train/tokens_used": 2227384800,
"theoretical_loss": 3.4026187867908346,
"tokens_seen": 2206924800
},
{
"epoch": 0.67,
"learning_rate": 0.00016738214643931797,
"loss": 2.5615,
"theoretical_loss": 3.4026187867908346,
"tokens_seen": 2206924800
},
{
"epoch": 0.67,
"learning_rate": 0.00016737211634904715,
"loss": 2.7374,
"theoretical_loss": 3.4026105243879807,
"tokens_seen": 2206990336
},
{
"epoch": 0.67,
"learning_rate": 0.00016736208625877633,
"loss": 2.7852,
"theoretical_loss": 3.402602262299169,
"tokens_seen": 2207055872
},
{
"epoch": 0.67,
"learning_rate": 0.0001673520561685055,
"loss": 2.5782,
"theoretical_loss": 3.4025940005243775,
"tokens_seen": 2207121408
},
{
"epoch": 0.67,
"learning_rate": 0.00016734202607823472,
"loss": 2.6617,
"theoretical_loss": 3.4025857390635856,
"tokens_seen": 2207186944
},
{
"epoch": 0.67,
"learning_rate": 0.0001673319959879639,
"loss": 2.6942,
"theoretical_loss": 3.402577477916772,
"tokens_seen": 2207252480
},
{
"epoch": 0.67,
"learning_rate": 0.00016732196589769308,
"loss": 2.625,
"theoretical_loss": 3.402569217083915,
"tokens_seen": 2207318016
},
{
"epoch": 0.67,
"learning_rate": 0.00016731193580742226,
"loss": 2.7622,
"theoretical_loss": 3.4025609565649937,
"tokens_seen": 2207383552
},
{
"epoch": 0.67,
"learning_rate": 0.00016730190571715144,
"loss": 2.6159,
"theoretical_loss": 3.4025526963599866,
"tokens_seen": 2207449088
},
{
"epoch": 0.67,
"learning_rate": 0.00016729187562688065,
"loss": 2.4607,
"theoretical_loss": 3.402544436468873,
"tokens_seen": 2207514624
},
{
"epoch": 0.67,
"learning_rate": 0.00016728184553660983,
"loss": 2.7509,
"theoretical_loss": 3.402536176891631,
"tokens_seen": 2207580160
},
{
"epoch": 0.67,
"learning_rate": 0.000167271815446339,
"loss": 2.8594,
"theoretical_loss": 3.4025279176282397,
"tokens_seen": 2207645696
},
{
"epoch": 0.67,
"learning_rate": 0.0001672617853560682,
"loss": 2.7354,
"theoretical_loss": 3.4025196586786777,
"tokens_seen": 2207711232
},
{
"epoch": 0.67,
"learning_rate": 0.0001672517552657974,
"loss": 2.6177,
"theoretical_loss": 3.4025114000429237,
"tokens_seen": 2207776768
},
{
"epoch": 0.67,
"learning_rate": 0.00016724172517552658,
"loss": 2.564,
"theoretical_loss": 3.402503141720957,
"tokens_seen": 2207842304
},
{
"epoch": 0.67,
"learning_rate": 0.00016723169508525576,
"loss": 2.5203,
"theoretical_loss": 3.4024948837127558,
"tokens_seen": 2207907840
},
{
"epoch": 0.67,
"learning_rate": 0.00016722166499498494,
"loss": 2.7575,
"theoretical_loss": 3.4024866260182987,
"tokens_seen": 2207973376
},
{
"epoch": 0.67,
"learning_rate": 0.00016721163490471413,
"loss": 2.7646,
"theoretical_loss": 3.4024783686375653,
"tokens_seen": 2208038912
},
{
"epoch": 0.67,
"learning_rate": 0.00016720160481444333,
"loss": 2.6366,
"theoretical_loss": 3.4024701115705334,
"tokens_seen": 2208104448
},
{
"epoch": 0.67,
"learning_rate": 0.00016719157472417251,
"loss": 2.3241,
"theoretical_loss": 3.4024618548171826,
"tokens_seen": 2208169984
},
{
"epoch": 0.67,
"learning_rate": 0.0001671815446339017,
"loss": 2.6221,
"theoretical_loss": 3.402453598377491,
"tokens_seen": 2208235520
},
{
"epoch": 0.67,
"learning_rate": 0.00016717151454363088,
"loss": 2.7422,
"theoretical_loss": 3.402445342251438,
"tokens_seen": 2208301056
},
{
"epoch": 0.67,
"learning_rate": 0.00016716148445336009,
"loss": 2.7602,
"theoretical_loss": 3.4024370864390017,
"tokens_seen": 2208366592
},
{
"epoch": 0.67,
"learning_rate": 0.00016715145436308927,
"loss": 2.885,
"theoretical_loss": 3.4024288309401616,
"tokens_seen": 2208432128
},
{
"epoch": 0.67,
"learning_rate": 0.00016714142427281845,
"loss": 2.574,
"theoretical_loss": 3.402420575754896,
"tokens_seen": 2208497664
},
{
"epoch": 0.67,
"objective/train/docs_used": 1251065,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6675820350646973,
"objective/train/theoretical_loss": 3.4024123208831836,
"objective/train/tokens_used": 2229023200,
"theoretical_loss": 3.4024123208831836,
"tokens_seen": 2208563200
},
{
"epoch": 0.67,
"learning_rate": 0.00016713139418254763,
"loss": 2.4309,
"theoretical_loss": 3.4024123208831836,
"tokens_seen": 2208563200
},
{
"epoch": 0.67,
"learning_rate": 0.0001671213640922768,
"loss": 2.5745,
"theoretical_loss": 3.402404066325004,
"tokens_seen": 2208628736
},
{
"epoch": 0.67,
"learning_rate": 0.00016711133400200602,
"loss": 2.528,
"theoretical_loss": 3.402395812080335,
"tokens_seen": 2208694272
},
{
"epoch": 0.67,
"learning_rate": 0.0001671013039117352,
"loss": 2.5742,
"theoretical_loss": 3.4023875581491554,
"tokens_seen": 2208759808
},
{
"epoch": 0.67,
"learning_rate": 0.00016709127382146438,
"loss": 2.6714,
"theoretical_loss": 3.4023793045314448,
"tokens_seen": 2208825344
},
{
"epoch": 0.67,
"learning_rate": 0.00016708124373119356,
"loss": 2.4912,
"theoretical_loss": 3.402371051227181,
"tokens_seen": 2208890880
},
{
"epoch": 0.67,
"learning_rate": 0.00016707121364092277,
"loss": 2.7302,
"theoretical_loss": 3.4023627982363442,
"tokens_seen": 2208956416
},
{
"epoch": 0.67,
"learning_rate": 0.00016706118355065195,
"loss": 2.7011,
"theoretical_loss": 3.402354545558912,
"tokens_seen": 2209021952
},
{
"epoch": 0.67,
"learning_rate": 0.00016705115346038113,
"loss": 2.722,
"theoretical_loss": 3.4023462931948636,
"tokens_seen": 2209087488
},
{
"epoch": 0.67,
"learning_rate": 0.00016704112337011034,
"loss": 2.4734,
"theoretical_loss": 3.4023380411441777,
"tokens_seen": 2209153024
},
{
"epoch": 0.67,
"learning_rate": 0.00016703109327983955,
"loss": 2.603,
"theoretical_loss": 3.402329789406833,
"tokens_seen": 2209218560
},
{
"epoch": 0.67,
"learning_rate": 0.00016702106318956873,
"loss": 2.562,
"theoretical_loss": 3.402321537982809,
"tokens_seen": 2209284096
},
{
"epoch": 0.67,
"learning_rate": 0.0001670110330992979,
"loss": 2.4257,
"theoretical_loss": 3.402313286872084,
"tokens_seen": 2209349632
},
{
"epoch": 0.67,
"learning_rate": 0.0001670010030090271,
"loss": 2.509,
"theoretical_loss": 3.4023050360746363,
"tokens_seen": 2209415168
},
{
"epoch": 0.67,
"learning_rate": 0.00016699097291875627,
"loss": 2.7746,
"theoretical_loss": 3.4022967855904453,
"tokens_seen": 2209480704
},
{
"epoch": 0.67,
"learning_rate": 0.00016698094282848548,
"loss": 2.4786,
"theoretical_loss": 3.4022885354194896,
"tokens_seen": 2209546240
},
{
"epoch": 0.67,
"learning_rate": 0.00016697091273821466,
"loss": 2.606,
"theoretical_loss": 3.4022802855617487,
"tokens_seen": 2209611776
},
{
"epoch": 0.67,
"learning_rate": 0.00016696088264794384,
"loss": 2.5186,
"theoretical_loss": 3.4022720360172007,
"tokens_seen": 2209677312
},
{
"epoch": 0.67,
"learning_rate": 0.00016695085255767302,
"loss": 2.4595,
"theoretical_loss": 3.402263786785824,
"tokens_seen": 2209742848
},
{
"epoch": 0.67,
"learning_rate": 0.00016694082246740223,
"loss": 2.6374,
"theoretical_loss": 3.4022555378675987,
"tokens_seen": 2209808384
},
{
"epoch": 0.67,
"learning_rate": 0.00016693079237713141,
"loss": 2.8346,
"theoretical_loss": 3.402247289262503,
"tokens_seen": 2209873920
},
{
"epoch": 0.67,
"learning_rate": 0.0001669207622868606,
"loss": 2.6372,
"theoretical_loss": 3.402239040970515,
"tokens_seen": 2209939456
},
{
"epoch": 0.67,
"learning_rate": 0.00016691073219658978,
"loss": 2.6193,
"theoretical_loss": 3.4022307929916145,
"tokens_seen": 2210004992
},
{
"epoch": 0.67,
"learning_rate": 0.00016690070210631896,
"loss": 2.7749,
"theoretical_loss": 3.40222254532578,
"tokens_seen": 2210070528
},
{
"epoch": 0.67,
"learning_rate": 0.00016689067201604817,
"loss": 2.7723,
"theoretical_loss": 3.4022142979729906,
"tokens_seen": 2210136064
},
{
"epoch": 0.67,
"objective/train/docs_used": 1252332,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.716803550720215,
"objective/train/theoretical_loss": 3.402206050933225,
"objective/train/tokens_used": 2230661600,
"theoretical_loss": 3.402206050933225,
"tokens_seen": 2210201600
},
{
"epoch": 0.67,
"learning_rate": 0.00016688064192577735,
"loss": 2.5954,
"theoretical_loss": 3.402206050933225,
"tokens_seen": 2210201600
},
{
"epoch": 0.67,
"learning_rate": 0.00016687061183550653,
"loss": 2.5507,
"theoretical_loss": 3.4021978042064616,
"tokens_seen": 2210267136
},
{
"epoch": 0.67,
"learning_rate": 0.0001668605817452357,
"loss": 2.5781,
"theoretical_loss": 3.4021895577926795,
"tokens_seen": 2210332672
},
{
"epoch": 0.67,
"learning_rate": 0.00016685055165496492,
"loss": 2.6742,
"theoretical_loss": 3.402181311691858,
"tokens_seen": 2210398208
},
{
"epoch": 0.67,
"learning_rate": 0.0001668405215646941,
"loss": 2.4919,
"theoretical_loss": 3.402173065903975,
"tokens_seen": 2210463744
},
{
"epoch": 0.67,
"learning_rate": 0.00016683049147442328,
"loss": 2.6115,
"theoretical_loss": 3.4021648204290105,
"tokens_seen": 2210529280
},
{
"epoch": 0.67,
"learning_rate": 0.00016682046138415246,
"loss": 2.5251,
"theoretical_loss": 3.4021565752669423,
"tokens_seen": 2210594816
},
{
"epoch": 0.67,
"learning_rate": 0.00016681043129388164,
"loss": 2.5372,
"theoretical_loss": 3.40214833041775,
"tokens_seen": 2210660352
},
{
"epoch": 0.67,
"learning_rate": 0.00016680040120361085,
"loss": 2.7147,
"theoretical_loss": 3.4021400858814124,
"tokens_seen": 2210725888
},
{
"epoch": 0.67,
"learning_rate": 0.00016679037111334003,
"loss": 2.5543,
"theoretical_loss": 3.4021318416579076,
"tokens_seen": 2210791424
},
{
"epoch": 0.67,
"learning_rate": 0.0001667803410230692,
"loss": 2.5952,
"theoretical_loss": 3.4021235977472153,
"tokens_seen": 2210856960
},
{
"epoch": 0.67,
"learning_rate": 0.0001667703109327984,
"loss": 2.4037,
"theoretical_loss": 3.4021153541493137,
"tokens_seen": 2210922496
},
{
"epoch": 0.67,
"learning_rate": 0.0001667602808425276,
"loss": 2.5779,
"theoretical_loss": 3.4021071108641823,
"tokens_seen": 2210988032
},
{
"epoch": 0.67,
"learning_rate": 0.00016675025075225678,
"loss": 2.5323,
"theoretical_loss": 3.4020988678918,
"tokens_seen": 2211053568
},
{
"epoch": 0.67,
"learning_rate": 0.00016674022066198596,
"loss": 2.5013,
"theoretical_loss": 3.4020906252321446,
"tokens_seen": 2211119104
},
{
"epoch": 0.67,
"learning_rate": 0.00016673019057171514,
"loss": 2.5335,
"theoretical_loss": 3.4020823828851956,
"tokens_seen": 2211184640
},
{
"epoch": 0.67,
"learning_rate": 0.00016672016048144433,
"loss": 2.6592,
"theoretical_loss": 3.4020741408509325,
"tokens_seen": 2211250176
},
{
"epoch": 0.67,
"learning_rate": 0.00016671013039117353,
"loss": 2.6996,
"theoretical_loss": 3.4020658991293335,
"tokens_seen": 2211315712
},
{
"epoch": 0.67,
"learning_rate": 0.00016670010030090271,
"loss": 2.5574,
"theoretical_loss": 3.402057657720378,
"tokens_seen": 2211381248
},
{
"epoch": 0.67,
"learning_rate": 0.0001666900702106319,
"loss": 2.8207,
"theoretical_loss": 3.4020494166240436,
"tokens_seen": 2211446784
},
{
"epoch": 0.67,
"learning_rate": 0.00016668004012036108,
"loss": 2.5527,
"theoretical_loss": 3.402041175840311,
"tokens_seen": 2211512320
},
{
"epoch": 0.67,
"learning_rate": 0.00016667001003009029,
"loss": 2.5584,
"theoretical_loss": 3.4020329353691574,
"tokens_seen": 2211577856
},
{
"epoch": 0.67,
"learning_rate": 0.00016665997993981947,
"loss": 2.739,
"theoretical_loss": 3.402024695210563,
"tokens_seen": 2211643392
},
{
"epoch": 0.67,
"learning_rate": 0.00016664994984954865,
"loss": 2.6189,
"theoretical_loss": 3.4020164553645054,
"tokens_seen": 2211708928
},
{
"epoch": 0.67,
"learning_rate": 0.00016663991975927783,
"loss": 2.8751,
"theoretical_loss": 3.4020082158309646,
"tokens_seen": 2211774464
},
{
"epoch": 0.67,
"objective/train/docs_used": 1253043,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4169273376464844,
"objective/train/theoretical_loss": 3.4019999766099187,
"objective/train/tokens_used": 2232300000,
"theoretical_loss": 3.4019999766099187,
"tokens_seen": 2211840000
},
{
"epoch": 0.67,
"learning_rate": 0.000166629889669007,
"loss": 2.5851,
"theoretical_loss": 3.4019999766099187,
"tokens_seen": 2211840000
},
{
"epoch": 0.67,
"learning_rate": 0.00016661985957873622,
"loss": 2.69,
"theoretical_loss": 3.4019917377013473,
"tokens_seen": 2211905536
},
{
"epoch": 0.67,
"learning_rate": 0.0001666098294884654,
"loss": 2.7486,
"theoretical_loss": 3.401983499105229,
"tokens_seen": 2211971072
},
{
"epoch": 0.67,
"learning_rate": 0.00016659979939819458,
"loss": 2.5317,
"theoretical_loss": 3.4019752608215423,
"tokens_seen": 2212036608
},
{
"epoch": 0.67,
"learning_rate": 0.00016658976930792376,
"loss": 2.638,
"theoretical_loss": 3.401967022850267,
"tokens_seen": 2212102144
},
{
"epoch": 0.67,
"learning_rate": 0.00016657973921765297,
"loss": 2.646,
"theoretical_loss": 3.401958785191381,
"tokens_seen": 2212167680
},
{
"epoch": 0.67,
"learning_rate": 0.00016656970912738215,
"loss": 2.3757,
"theoretical_loss": 3.401950547844864,
"tokens_seen": 2212233216
},
{
"epoch": 0.67,
"learning_rate": 0.00016655967903711133,
"loss": 2.3933,
"theoretical_loss": 3.4019423108106945,
"tokens_seen": 2212298752
},
{
"epoch": 0.67,
"learning_rate": 0.0001665496489468405,
"loss": 2.6434,
"theoretical_loss": 3.4019340740888513,
"tokens_seen": 2212364288
},
{
"epoch": 0.67,
"learning_rate": 0.00016653961885656972,
"loss": 2.4092,
"theoretical_loss": 3.401925837679313,
"tokens_seen": 2212429824
},
{
"epoch": 0.67,
"learning_rate": 0.0001665295887662989,
"loss": 2.5277,
"theoretical_loss": 3.40191760158206,
"tokens_seen": 2212495360
},
{
"epoch": 0.67,
"learning_rate": 0.00016651955867602808,
"loss": 2.4854,
"theoretical_loss": 3.401909365797069,
"tokens_seen": 2212560896
},
{
"epoch": 0.67,
"learning_rate": 0.00016650952858575726,
"loss": 2.447,
"theoretical_loss": 3.4019011303243207,
"tokens_seen": 2212626432
},
{
"epoch": 0.67,
"learning_rate": 0.00016649949849548645,
"loss": 2.5849,
"theoretical_loss": 3.4018928951637935,
"tokens_seen": 2212691968
},
{
"epoch": 0.67,
"learning_rate": 0.00016648946840521565,
"loss": 2.7874,
"theoretical_loss": 3.401884660315466,
"tokens_seen": 2212757504
},
{
"epoch": 0.67,
"learning_rate": 0.00016647943831494484,
"loss": 2.6556,
"theoretical_loss": 3.4018764257793173,
"tokens_seen": 2212823040
},
{
"epoch": 0.67,
"learning_rate": 0.00016646940822467402,
"loss": 2.4826,
"theoretical_loss": 3.4018681915553266,
"tokens_seen": 2212888576
},
{
"epoch": 0.67,
"learning_rate": 0.0001664593781344032,
"loss": 2.7519,
"theoretical_loss": 3.4018599576434725,
"tokens_seen": 2212954112
},
{
"epoch": 0.67,
"learning_rate": 0.0001664493480441324,
"loss": 2.4165,
"theoretical_loss": 3.4018517240437336,
"tokens_seen": 2213019648
},
{
"epoch": 0.67,
"learning_rate": 0.0001664393179538616,
"loss": 2.2541,
"theoretical_loss": 3.4018434907560895,
"tokens_seen": 2213085184
},
{
"epoch": 0.67,
"learning_rate": 0.00016642928786359077,
"loss": 2.5185,
"theoretical_loss": 3.401835257780519,
"tokens_seen": 2213150720
},
{
"epoch": 0.67,
"learning_rate": 0.00016641925777331995,
"loss": 2.5441,
"theoretical_loss": 3.401827025117001,
"tokens_seen": 2213216256
},
{
"epoch": 0.67,
"learning_rate": 0.00016640922768304913,
"loss": 2.6057,
"theoretical_loss": 3.401818792765514,
"tokens_seen": 2213281792
},
{
"epoch": 0.67,
"learning_rate": 0.00016639919759277834,
"loss": 2.6328,
"theoretical_loss": 3.4018105607260374,
"tokens_seen": 2213347328
},
{
"epoch": 0.67,
"learning_rate": 0.00016638916750250752,
"loss": 2.5074,
"theoretical_loss": 3.40180232899855,
"tokens_seen": 2213412864
},
{
"epoch": 0.67,
"objective/train/docs_used": 1254214,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1435799598693848,
"objective/train/theoretical_loss": 3.4017940975830308,
"objective/train/tokens_used": 2233938400,
"theoretical_loss": 3.4017940975830308,
"tokens_seen": 2213478400
},
{
"epoch": 0.67,
"learning_rate": 0.0001663791374122367,
"loss": 2.2176,
"theoretical_loss": 3.4017940975830308,
"tokens_seen": 2213478400
},
{
"epoch": 0.67,
"learning_rate": 0.00016636910732196588,
"loss": 2.5571,
"theoretical_loss": 3.4017858664794582,
"tokens_seen": 2213543936
},
{
"epoch": 0.67,
"learning_rate": 0.0001663590772316951,
"loss": 2.7346,
"theoretical_loss": 3.401777635687812,
"tokens_seen": 2213609472
},
{
"epoch": 0.67,
"learning_rate": 0.00016634904714142427,
"loss": 2.847,
"theoretical_loss": 3.4017694052080705,
"tokens_seen": 2213675008
},
{
"epoch": 0.67,
"learning_rate": 0.00016633901705115345,
"loss": 2.5443,
"theoretical_loss": 3.4017611750402135,
"tokens_seen": 2213740544
},
{
"epoch": 0.67,
"learning_rate": 0.00016632898696088263,
"loss": 2.4444,
"theoretical_loss": 3.401752945184219,
"tokens_seen": 2213806080
},
{
"epoch": 0.67,
"learning_rate": 0.00016631895687061181,
"loss": 2.4024,
"theoretical_loss": 3.401744715640066,
"tokens_seen": 2213871616
},
{
"epoch": 0.67,
"learning_rate": 0.00016630892678034102,
"loss": 2.5942,
"theoretical_loss": 3.401736486407734,
"tokens_seen": 2213937152
},
{
"epoch": 0.67,
"learning_rate": 0.0001662988966900702,
"loss": 2.5781,
"theoretical_loss": 3.4017282574872016,
"tokens_seen": 2214002688
},
{
"epoch": 0.67,
"learning_rate": 0.0001662888665997994,
"loss": 2.7554,
"theoretical_loss": 3.401720028878448,
"tokens_seen": 2214068224
},
{
"epoch": 0.67,
"learning_rate": 0.0001662788365095286,
"loss": 2.5104,
"theoretical_loss": 3.4017118005814524,
"tokens_seen": 2214133760
},
{
"epoch": 0.67,
"learning_rate": 0.0001662688064192578,
"loss": 2.4563,
"theoretical_loss": 3.4017035725961926,
"tokens_seen": 2214199296
},
{
"epoch": 0.67,
"learning_rate": 0.00016625877632898698,
"loss": 2.6019,
"theoretical_loss": 3.401695344922649,
"tokens_seen": 2214264832
},
{
"epoch": 0.67,
"learning_rate": 0.00016624874623871616,
"loss": 2.5611,
"theoretical_loss": 3.4016871175607997,
"tokens_seen": 2214330368
},
{
"epoch": 0.67,
"learning_rate": 0.00016623871614844534,
"loss": 2.5671,
"theoretical_loss": 3.401678890510624,
"tokens_seen": 2214395904
},
{
"epoch": 0.67,
"learning_rate": 0.00016622868605817453,
"loss": 2.5406,
"theoretical_loss": 3.4016706637721006,
"tokens_seen": 2214461440
},
{
"epoch": 0.67,
"learning_rate": 0.00016621865596790373,
"loss": 2.5971,
"theoretical_loss": 3.4016624373452085,
"tokens_seen": 2214526976
},
{
"epoch": 0.67,
"learning_rate": 0.00016620862587763292,
"loss": 2.4816,
"theoretical_loss": 3.401654211229927,
"tokens_seen": 2214592512
},
{
"epoch": 0.67,
"learning_rate": 0.0001661985957873621,
"loss": 2.3571,
"theoretical_loss": 3.4016459854262346,
"tokens_seen": 2214658048
},
{
"epoch": 0.67,
"learning_rate": 0.00016618856569709128,
"loss": 2.6164,
"theoretical_loss": 3.4016377599341108,
"tokens_seen": 2214723584
},
{
"epoch": 0.67,
"learning_rate": 0.00016617853560682049,
"loss": 2.5963,
"theoretical_loss": 3.4016295347535346,
"tokens_seen": 2214789120
},
{
"epoch": 0.67,
"learning_rate": 0.00016616850551654967,
"loss": 2.6234,
"theoretical_loss": 3.4016213098844843,
"tokens_seen": 2214854656
},
{
"epoch": 0.67,
"learning_rate": 0.00016615847542627885,
"loss": 2.4717,
"theoretical_loss": 3.4016130853269395,
"tokens_seen": 2214920192
},
{
"epoch": 0.67,
"learning_rate": 0.00016614844533600803,
"loss": 2.6848,
"theoretical_loss": 3.4016048610808785,
"tokens_seen": 2214985728
},
{
"epoch": 0.67,
"learning_rate": 0.0001661384152457372,
"loss": 2.6747,
"theoretical_loss": 3.401596637146281,
"tokens_seen": 2215051264
},
{
"epoch": 0.67,
"objective/train/docs_used": 1254739,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.448244571685791,
"objective/train/theoretical_loss": 3.4015884135231262,
"objective/train/tokens_used": 2235576800,
"theoretical_loss": 3.4015884135231262,
"tokens_seen": 2215116800
},
{
"epoch": 0.67,
"learning_rate": 0.00016612838515546642,
"loss": 2.5416,
"theoretical_loss": 3.4015884135231262,
"tokens_seen": 2215116800
},
{
"epoch": 0.67,
"learning_rate": 0.0001661183550651956,
"loss": 2.5513,
"theoretical_loss": 3.4015801902113925,
"tokens_seen": 2215182336
},
{
"epoch": 0.67,
"learning_rate": 0.00016610832497492478,
"loss": 2.6168,
"theoretical_loss": 3.401571967211059,
"tokens_seen": 2215247872
},
{
"epoch": 0.67,
"learning_rate": 0.00016609829488465396,
"loss": 2.6919,
"theoretical_loss": 3.4015637445221043,
"tokens_seen": 2215313408
},
{
"epoch": 0.67,
"learning_rate": 0.00016608826479438317,
"loss": 2.3968,
"theoretical_loss": 3.4015555221445086,
"tokens_seen": 2215378944
},
{
"epoch": 0.67,
"learning_rate": 0.00016607823470411235,
"loss": 2.4934,
"theoretical_loss": 3.4015473000782497,
"tokens_seen": 2215444480
},
{
"epoch": 0.67,
"learning_rate": 0.00016606820461384153,
"loss": 2.7449,
"theoretical_loss": 3.401539078323307,
"tokens_seen": 2215510016
},
{
"epoch": 0.67,
"learning_rate": 0.0001660581745235707,
"loss": 2.4487,
"theoretical_loss": 3.40153085687966,
"tokens_seen": 2215575552
},
{
"epoch": 0.67,
"learning_rate": 0.00016604814443329992,
"loss": 2.7099,
"theoretical_loss": 3.4015226357472867,
"tokens_seen": 2215641088
},
{
"epoch": 0.67,
"learning_rate": 0.0001660381143430291,
"loss": 2.4476,
"theoretical_loss": 3.401514414926167,
"tokens_seen": 2215706624
},
{
"epoch": 0.67,
"learning_rate": 0.00016602808425275828,
"loss": 2.4858,
"theoretical_loss": 3.401506194416279,
"tokens_seen": 2215772160
},
{
"epoch": 0.67,
"learning_rate": 0.00016601805416248746,
"loss": 2.639,
"theoretical_loss": 3.4014979742176026,
"tokens_seen": 2215837696
},
{
"epoch": 0.67,
"learning_rate": 0.00016600802407221665,
"loss": 2.4846,
"theoretical_loss": 3.401489754330117,
"tokens_seen": 2215903232
},
{
"epoch": 0.67,
"learning_rate": 0.00016599799398194585,
"loss": 2.6558,
"theoretical_loss": 3.4014815347538,
"tokens_seen": 2215968768
},
{
"epoch": 0.67,
"learning_rate": 0.00016598796389167504,
"loss": 2.594,
"theoretical_loss": 3.401473315488632,
"tokens_seen": 2216034304
},
{
"epoch": 0.67,
"learning_rate": 0.00016597793380140422,
"loss": 2.6619,
"theoretical_loss": 3.4014650965345905,
"tokens_seen": 2216099840
},
{
"epoch": 0.67,
"learning_rate": 0.0001659679037111334,
"loss": 2.4621,
"theoretical_loss": 3.401456877891656,
"tokens_seen": 2216165376
},
{
"epoch": 0.67,
"learning_rate": 0.0001659578736208626,
"loss": 2.7338,
"theoretical_loss": 3.401448659559807,
"tokens_seen": 2216230912
},
{
"epoch": 0.67,
"learning_rate": 0.0001659478435305918,
"loss": 2.5813,
"theoretical_loss": 3.401440441539022,
"tokens_seen": 2216296448
},
{
"epoch": 0.67,
"learning_rate": 0.00016593781344032097,
"loss": 2.5816,
"theoretical_loss": 3.4014322238292807,
"tokens_seen": 2216361984
},
{
"epoch": 0.67,
"learning_rate": 0.00016592778335005015,
"loss": 2.4671,
"theoretical_loss": 3.401424006430562,
"tokens_seen": 2216427520
},
{
"epoch": 0.67,
"learning_rate": 0.00016591775325977933,
"loss": 2.5647,
"theoretical_loss": 3.4014157893428445,
"tokens_seen": 2216493056
},
{
"epoch": 0.67,
"learning_rate": 0.00016590772316950854,
"loss": 2.5486,
"theoretical_loss": 3.4014075725661077,
"tokens_seen": 2216558592
},
{
"epoch": 0.67,
"learning_rate": 0.00016589769307923772,
"loss": 2.7278,
"theoretical_loss": 3.40139935610033,
"tokens_seen": 2216624128
},
{
"epoch": 0.67,
"learning_rate": 0.0001658876629889669,
"loss": 2.4623,
"theoretical_loss": 3.401391139945492,
"tokens_seen": 2216689664
},
{
"epoch": 0.67,
"objective/train/docs_used": 1256282,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.406754732131958,
"objective/train/theoretical_loss": 3.401382924101571,
"objective/train/tokens_used": 2237215200,
"theoretical_loss": 3.401382924101571,
"tokens_seen": 2216755200
},
{
"epoch": 0.67,
"learning_rate": 0.00016587763289869608,
"loss": 2.5486,
"theoretical_loss": 3.401382924101571,
"tokens_seen": 2216755200
},
{
"epoch": 0.67,
"learning_rate": 0.0001658676028084253,
"loss": 2.7743,
"theoretical_loss": 3.4013747085685466,
"tokens_seen": 2216820736
},
{
"epoch": 0.67,
"learning_rate": 0.00016585757271815447,
"loss": 2.7765,
"theoretical_loss": 3.4013664933463983,
"tokens_seen": 2216886272
},
{
"epoch": 0.67,
"learning_rate": 0.00016584754262788365,
"loss": 2.7512,
"theoretical_loss": 3.4013582784351044,
"tokens_seen": 2216951808
},
{
"epoch": 0.67,
"learning_rate": 0.00016583751253761283,
"loss": 2.6279,
"theoretical_loss": 3.401350063834645,
"tokens_seen": 2217017344
},
{
"epoch": 0.67,
"learning_rate": 0.00016582748244734201,
"loss": 2.8325,
"theoretical_loss": 3.401341849544998,
"tokens_seen": 2217082880
},
{
"epoch": 0.67,
"learning_rate": 0.00016581745235707122,
"loss": 2.4849,
"theoretical_loss": 3.401333635566143,
"tokens_seen": 2217148416
},
{
"epoch": 0.67,
"learning_rate": 0.0001658074222668004,
"loss": 2.495,
"theoretical_loss": 3.401325421898059,
"tokens_seen": 2217213952
},
{
"epoch": 0.67,
"learning_rate": 0.00016579739217652959,
"loss": 2.6179,
"theoretical_loss": 3.4013172085407253,
"tokens_seen": 2217279488
},
{
"epoch": 0.67,
"learning_rate": 0.00016578736208625877,
"loss": 2.4784,
"theoretical_loss": 3.4013089954941202,
"tokens_seen": 2217345024
},
{
"epoch": 0.67,
"learning_rate": 0.00016577733199598797,
"loss": 2.6264,
"theoretical_loss": 3.401300782758224,
"tokens_seen": 2217410560
},
{
"epoch": 0.67,
"learning_rate": 0.00016576730190571716,
"loss": 2.6308,
"theoretical_loss": 3.4012925703330144,
"tokens_seen": 2217476096
},
{
"epoch": 0.67,
"learning_rate": 0.00016575727181544634,
"loss": 2.6397,
"theoretical_loss": 3.4012843582184717,
"tokens_seen": 2217541632
},
{
"epoch": 0.67,
"learning_rate": 0.00016574724172517552,
"loss": 2.6032,
"theoretical_loss": 3.401276146414574,
"tokens_seen": 2217607168
},
{
"epoch": 0.67,
"learning_rate": 0.0001657372116349047,
"loss": 2.7005,
"theoretical_loss": 3.401267934921301,
"tokens_seen": 2217672704
},
{
"epoch": 0.67,
"learning_rate": 0.0001657271815446339,
"loss": 2.608,
"theoretical_loss": 3.4012597237386313,
"tokens_seen": 2217738240
},
{
"epoch": 0.67,
"learning_rate": 0.0001657171514543631,
"loss": 2.5687,
"theoretical_loss": 3.401251512866544,
"tokens_seen": 2217803776
},
{
"epoch": 0.67,
"learning_rate": 0.00016570712136409227,
"loss": 2.3585,
"theoretical_loss": 3.4012433023050193,
"tokens_seen": 2217869312
},
{
"epoch": 0.67,
"learning_rate": 0.00016569709127382145,
"loss": 2.4916,
"theoretical_loss": 3.4012350920540344,
"tokens_seen": 2217934848
},
{
"epoch": 0.67,
"learning_rate": 0.00016568706118355066,
"loss": 2.6444,
"theoretical_loss": 3.4012268821135696,
"tokens_seen": 2218000384
},
{
"epoch": 0.67,
"learning_rate": 0.00016567703109327984,
"loss": 2.5694,
"theoretical_loss": 3.4012186724836035,
"tokens_seen": 2218065920
},
{
"epoch": 0.67,
"learning_rate": 0.00016566700100300902,
"loss": 2.5518,
"theoretical_loss": 3.401210463164116,
"tokens_seen": 2218131456
},
{
"epoch": 0.67,
"learning_rate": 0.0001656569709127382,
"loss": 2.5545,
"theoretical_loss": 3.401202254155085,
"tokens_seen": 2218196992
},
{
"epoch": 0.67,
"learning_rate": 0.00016564694082246738,
"loss": 2.6887,
"theoretical_loss": 3.4011940454564904,
"tokens_seen": 2218262528
},
{
"epoch": 0.67,
"learning_rate": 0.0001656369107321966,
"loss": 2.6081,
"theoretical_loss": 3.401185837068311,
"tokens_seen": 2218328064
},
{
"epoch": 0.67,
"objective/train/docs_used": 1256744,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3241775035858154,
"objective/train/theoretical_loss": 3.401177628990526,
"objective/train/tokens_used": 2238853600,
"theoretical_loss": 3.401177628990526,
"tokens_seen": 2218393600
},
{
"epoch": 0.67,
"learning_rate": 0.00016562688064192577,
"loss": 2.4363,
"theoretical_loss": 3.401177628990526,
"tokens_seen": 2218393600
},
{
"epoch": 0.67,
"learning_rate": 0.00016561685055165495,
"loss": 2.6181,
"theoretical_loss": 3.4011694212231145,
"tokens_seen": 2218459136
},
{
"epoch": 0.67,
"learning_rate": 0.00016560682046138413,
"loss": 2.6413,
"theoretical_loss": 3.401161213766055,
"tokens_seen": 2218524672
},
{
"epoch": 0.67,
"learning_rate": 0.00016559679037111334,
"loss": 2.5081,
"theoretical_loss": 3.401153006619328,
"tokens_seen": 2218590208
},
{
"epoch": 0.67,
"learning_rate": 0.00016558676028084252,
"loss": 2.6271,
"theoretical_loss": 3.4011447997829114,
"tokens_seen": 2218655744
},
{
"epoch": 0.67,
"learning_rate": 0.0001655767301905717,
"loss": 2.4697,
"theoretical_loss": 3.4011365932567843,
"tokens_seen": 2218721280
},
{
"epoch": 0.67,
"learning_rate": 0.00016556670010030089,
"loss": 2.7864,
"theoretical_loss": 3.4011283870409263,
"tokens_seen": 2218786816
},
{
"epoch": 0.67,
"learning_rate": 0.00016555667001003007,
"loss": 2.5393,
"theoretical_loss": 3.4011201811353167,
"tokens_seen": 2218852352
},
{
"epoch": 0.67,
"learning_rate": 0.00016554663991975928,
"loss": 2.5957,
"theoretical_loss": 3.4011119755399335,
"tokens_seen": 2218917888
},
{
"epoch": 0.67,
"learning_rate": 0.00016553660982948848,
"loss": 2.7207,
"theoretical_loss": 3.4011037702547573,
"tokens_seen": 2218983424
},
{
"epoch": 0.67,
"learning_rate": 0.00016552657973921767,
"loss": 2.5341,
"theoretical_loss": 3.4010955652797663,
"tokens_seen": 2219048960
},
{
"epoch": 0.67,
"learning_rate": 0.00016551654964894685,
"loss": 2.777,
"theoretical_loss": 3.4010873606149397,
"tokens_seen": 2219114496
},
{
"epoch": 0.67,
"learning_rate": 0.00016550651955867605,
"loss": 2.6366,
"theoretical_loss": 3.4010791562602565,
"tokens_seen": 2219180032
},
{
"epoch": 0.67,
"learning_rate": 0.00016549648946840524,
"loss": 2.4311,
"theoretical_loss": 3.4010709522156963,
"tokens_seen": 2219245568
},
{
"epoch": 0.67,
"learning_rate": 0.00016548645937813442,
"loss": 2.6274,
"theoretical_loss": 3.401062748481238,
"tokens_seen": 2219311104
},
{
"epoch": 0.67,
"learning_rate": 0.0001654764292878636,
"loss": 2.544,
"theoretical_loss": 3.4010545450568603,
"tokens_seen": 2219376640
},
{
"epoch": 0.67,
"learning_rate": 0.0001654663991975928,
"loss": 2.413,
"theoretical_loss": 3.4010463419425427,
"tokens_seen": 2219442176
},
{
"epoch": 0.67,
"learning_rate": 0.000165456369107322,
"loss": 2.5489,
"theoretical_loss": 3.4010381391382647,
"tokens_seen": 2219507712
},
{
"epoch": 0.67,
"learning_rate": 0.00016544633901705117,
"loss": 2.7102,
"theoretical_loss": 3.401029936644005,
"tokens_seen": 2219573248
},
{
"epoch": 0.67,
"learning_rate": 0.00016543630892678035,
"loss": 2.5301,
"theoretical_loss": 3.4010217344597424,
"tokens_seen": 2219638784
},
{
"epoch": 0.67,
"learning_rate": 0.00016542627883650953,
"loss": 2.5912,
"theoretical_loss": 3.4010135325854565,
"tokens_seen": 2219704320
},
{
"epoch": 0.67,
"learning_rate": 0.00016541624874623874,
"loss": 2.5712,
"theoretical_loss": 3.4010053310211266,
"tokens_seen": 2219769856
},
{
"epoch": 0.67,
"learning_rate": 0.00016540621865596792,
"loss": 2.7202,
"theoretical_loss": 3.4009971297667314,
"tokens_seen": 2219835392
},
{
"epoch": 0.67,
"learning_rate": 0.0001653961885656971,
"loss": 2.8324,
"theoretical_loss": 3.40098892882225,
"tokens_seen": 2219900928
},
{
"epoch": 0.67,
"learning_rate": 0.00016538615847542628,
"loss": 2.4281,
"theoretical_loss": 3.4009807281876623,
"tokens_seen": 2219966464
},
{
"epoch": 0.67,
"objective/train/docs_used": 1258010,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.10628080368042,
"objective/train/theoretical_loss": 3.4009725278629466,
"objective/train/tokens_used": 2240492000,
"theoretical_loss": 3.4009725278629466,
"tokens_seen": 2220032000
},
{
"epoch": 0.67,
"learning_rate": 0.0001653761283851555,
"loss": 2.4056,
"theoretical_loss": 3.4009725278629466,
"tokens_seen": 2220032000
},
{
"epoch": 0.67,
"learning_rate": 0.00016536609829488467,
"loss": 2.4907,
"theoretical_loss": 3.400964327848082,
"tokens_seen": 2220097536
},
{
"epoch": 0.67,
"learning_rate": 0.00016535606820461385,
"loss": 2.6789,
"theoretical_loss": 3.4009561281430485,
"tokens_seen": 2220163072
},
{
"epoch": 0.67,
"learning_rate": 0.00016534603811434303,
"loss": 2.7728,
"theoretical_loss": 3.4009479287478244,
"tokens_seen": 2220228608
},
{
"epoch": 0.67,
"learning_rate": 0.00016533600802407221,
"loss": 2.6864,
"theoretical_loss": 3.4009397296623893,
"tokens_seen": 2220294144
},
{
"epoch": 0.67,
"learning_rate": 0.00016532597793380142,
"loss": 2.5108,
"theoretical_loss": 3.400931530886722,
"tokens_seen": 2220359680
},
{
"epoch": 0.67,
"learning_rate": 0.0001653159478435306,
"loss": 2.4042,
"theoretical_loss": 3.4009233324208026,
"tokens_seen": 2220425216
},
{
"epoch": 0.67,
"learning_rate": 0.00016530591775325979,
"loss": 2.5014,
"theoretical_loss": 3.4009151342646087,
"tokens_seen": 2220490752
},
{
"epoch": 0.67,
"learning_rate": 0.00016529588766298897,
"loss": 2.5218,
"theoretical_loss": 3.400906936418121,
"tokens_seen": 2220556288
},
{
"epoch": 0.67,
"learning_rate": 0.00016528585757271817,
"loss": 2.6735,
"theoretical_loss": 3.400898738881317,
"tokens_seen": 2220621824
},
{
"epoch": 0.67,
"learning_rate": 0.00016527582748244736,
"loss": 2.755,
"theoretical_loss": 3.400890541654178,
"tokens_seen": 2220687360
},
{
"epoch": 0.67,
"learning_rate": 0.00016526579739217654,
"loss": 2.5553,
"theoretical_loss": 3.400882344736681,
"tokens_seen": 2220752896
},
{
"epoch": 0.67,
"learning_rate": 0.00016525576730190572,
"loss": 2.6277,
"theoretical_loss": 3.400874148128807,
"tokens_seen": 2220818432
},
{
"epoch": 0.67,
"learning_rate": 0.0001652457372116349,
"loss": 2.4783,
"theoretical_loss": 3.4008659518305335,
"tokens_seen": 2220883968
},
{
"epoch": 0.67,
"learning_rate": 0.0001652357071213641,
"loss": 2.8127,
"theoretical_loss": 3.400857755841841,
"tokens_seen": 2220949504
},
{
"epoch": 0.67,
"learning_rate": 0.0001652256770310933,
"loss": 2.4741,
"theoretical_loss": 3.4008495601627082,
"tokens_seen": 2221015040
},
{
"epoch": 0.67,
"learning_rate": 0.00016521564694082247,
"loss": 2.7036,
"theoretical_loss": 3.4008413647931137,
"tokens_seen": 2221080576
},
{
"epoch": 0.67,
"learning_rate": 0.00016520561685055165,
"loss": 2.7798,
"theoretical_loss": 3.400833169733038,
"tokens_seen": 2221146112
},
{
"epoch": 0.67,
"learning_rate": 0.00016519558676028086,
"loss": 2.3687,
"theoretical_loss": 3.400824974982459,
"tokens_seen": 2221211648
},
{
"epoch": 0.67,
"learning_rate": 0.00016518555667001004,
"loss": 2.5065,
"theoretical_loss": 3.4008167805413563,
"tokens_seen": 2221277184
},
{
"epoch": 0.67,
"learning_rate": 0.00016517552657973922,
"loss": 2.7384,
"theoretical_loss": 3.4008085864097093,
"tokens_seen": 2221342720
},
{
"epoch": 0.67,
"learning_rate": 0.0001651654964894684,
"loss": 2.5894,
"theoretical_loss": 3.400800392587497,
"tokens_seen": 2221408256
},
{
"epoch": 0.67,
"learning_rate": 0.00016515546639919758,
"loss": 2.7845,
"theoretical_loss": 3.4007921990746985,
"tokens_seen": 2221473792
},
{
"epoch": 0.67,
"learning_rate": 0.0001651454363089268,
"loss": 2.4396,
"theoretical_loss": 3.4007840058712935,
"tokens_seen": 2221539328
},
{
"epoch": 0.67,
"learning_rate": 0.00016513540621865597,
"loss": 2.3663,
"theoretical_loss": 3.4007758129772605,
"tokens_seen": 2221604864
},
{
"epoch": 0.67,
"objective/train/docs_used": 1258716,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.008598566055298,
"objective/train/theoretical_loss": 3.400767620392579,
"objective/train/tokens_used": 2242130400,
"theoretical_loss": 3.400767620392579,
"tokens_seen": 2221670400
},
{
"epoch": 0.67,
"learning_rate": 0.00016512537612838515,
"loss": 2.8315,
"theoretical_loss": 3.400767620392579,
"tokens_seen": 2221670400
},
{
"epoch": 0.67,
"learning_rate": 0.00016511534603811433,
"loss": 2.6254,
"theoretical_loss": 3.400759428117228,
"tokens_seen": 2221735936
},
{
"epoch": 0.67,
"learning_rate": 0.00016510531594784354,
"loss": 2.6692,
"theoretical_loss": 3.4007512361511876,
"tokens_seen": 2221801472
},
{
"epoch": 0.67,
"learning_rate": 0.00016509528585757272,
"loss": 2.6479,
"theoretical_loss": 3.400743044494436,
"tokens_seen": 2221867008
},
{
"epoch": 0.67,
"learning_rate": 0.0001650852557673019,
"loss": 2.5154,
"theoretical_loss": 3.400734853146952,
"tokens_seen": 2221932544
},
{
"epoch": 0.67,
"learning_rate": 0.0001650752256770311,
"loss": 2.5227,
"theoretical_loss": 3.400726662108716,
"tokens_seen": 2221998080
},
{
"epoch": 0.67,
"learning_rate": 0.00016506519558676027,
"loss": 2.6954,
"theoretical_loss": 3.4007184713797067,
"tokens_seen": 2222063616
},
{
"epoch": 0.67,
"learning_rate": 0.00016505516549648948,
"loss": 2.5278,
"theoretical_loss": 3.4007102809599035,
"tokens_seen": 2222129152
},
{
"epoch": 0.67,
"learning_rate": 0.00016504513540621866,
"loss": 2.4187,
"theoretical_loss": 3.4007020908492853,
"tokens_seen": 2222194688
},
{
"epoch": 0.67,
"learning_rate": 0.00016503510531594784,
"loss": 2.6409,
"theoretical_loss": 3.4006939010478314,
"tokens_seen": 2222260224
},
{
"epoch": 0.67,
"learning_rate": 0.00016502507522567702,
"loss": 2.5326,
"theoretical_loss": 3.400685711555521,
"tokens_seen": 2222325760
},
{
"epoch": 0.67,
"learning_rate": 0.00016501504513540623,
"loss": 2.7323,
"theoretical_loss": 3.4006775223723333,
"tokens_seen": 2222391296
},
{
"epoch": 0.67,
"learning_rate": 0.0001650050150451354,
"loss": 2.9618,
"theoretical_loss": 3.4006693334982474,
"tokens_seen": 2222456832
},
{
"epoch": 0.67,
"learning_rate": 0.0001649949849548646,
"loss": 2.7346,
"theoretical_loss": 3.4006611449332427,
"tokens_seen": 2222522368
},
{
"epoch": 0.67,
"learning_rate": 0.00016498495486459377,
"loss": 2.7362,
"theoretical_loss": 3.4006529566772987,
"tokens_seen": 2222587904
},
{
"epoch": 0.67,
"learning_rate": 0.00016497492477432295,
"loss": 2.6449,
"theoretical_loss": 3.4006447687303942,
"tokens_seen": 2222653440
},
{
"epoch": 0.67,
"learning_rate": 0.00016496489468405216,
"loss": 2.7538,
"theoretical_loss": 3.4006365810925088,
"tokens_seen": 2222718976
},
{
"epoch": 0.67,
"learning_rate": 0.00016495486459378134,
"loss": 2.7226,
"theoretical_loss": 3.400628393763621,
"tokens_seen": 2222784512
},
{
"epoch": 0.67,
"learning_rate": 0.00016494483450351052,
"loss": 2.5704,
"theoretical_loss": 3.4006202067437106,
"tokens_seen": 2222850048
},
{
"epoch": 0.67,
"learning_rate": 0.0001649348044132397,
"loss": 2.3627,
"theoretical_loss": 3.400612020032757,
"tokens_seen": 2222915584
},
{
"epoch": 0.67,
"learning_rate": 0.0001649247743229689,
"loss": 2.5991,
"theoretical_loss": 3.400603833630739,
"tokens_seen": 2222981120
},
{
"epoch": 0.67,
"learning_rate": 0.0001649147442326981,
"loss": 2.5462,
"theoretical_loss": 3.400595647537636,
"tokens_seen": 2223046656
},
{
"epoch": 0.67,
"learning_rate": 0.00016490471414242727,
"loss": 2.7166,
"theoretical_loss": 3.4005874617534273,
"tokens_seen": 2223112192
},
{
"epoch": 0.67,
"learning_rate": 0.00016489468405215646,
"loss": 2.7095,
"theoretical_loss": 3.400579276278092,
"tokens_seen": 2223177728
},
{
"epoch": 0.67,
"learning_rate": 0.00016488465396188566,
"loss": 2.6366,
"theoretical_loss": 3.400571091111609,
"tokens_seen": 2223243264
},
{
"epoch": 0.67,
"objective/train/docs_used": 1259970,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.1125638484954834,
"objective/train/theoretical_loss": 3.400562906253959,
"objective/train/tokens_used": 2243768800,
"theoretical_loss": 3.400562906253959,
"tokens_seen": 2223308800
},
{
"epoch": 0.67,
"learning_rate": 0.00016487462387161484,
"loss": 2.6511,
"theoretical_loss": 3.400562906253959,
"tokens_seen": 2223308800
},
{
"epoch": 0.67,
"learning_rate": 0.00016486459378134403,
"loss": 2.6008,
"theoretical_loss": 3.4005547217051193,
"tokens_seen": 2223374336
},
{
"epoch": 0.67,
"learning_rate": 0.0001648545636910732,
"loss": 2.5793,
"theoretical_loss": 3.4005465374650705,
"tokens_seen": 2223439872
},
{
"epoch": 0.67,
"learning_rate": 0.0001648445336008024,
"loss": 2.472,
"theoretical_loss": 3.400538353533791,
"tokens_seen": 2223505408
},
{
"epoch": 0.67,
"learning_rate": 0.0001648345035105316,
"loss": 2.5275,
"theoretical_loss": 3.400530169911261,
"tokens_seen": 2223570944
},
{
"epoch": 0.67,
"learning_rate": 0.00016482447342026078,
"loss": 2.5958,
"theoretical_loss": 3.4005219865974587,
"tokens_seen": 2223636480
},
{
"epoch": 0.67,
"learning_rate": 0.00016481444332998996,
"loss": 2.71,
"theoretical_loss": 3.4005138035923643,
"tokens_seen": 2223702016
},
{
"epoch": 0.67,
"learning_rate": 0.00016480441323971914,
"loss": 2.5719,
"theoretical_loss": 3.4005056208959563,
"tokens_seen": 2223767552
},
{
"epoch": 0.67,
"learning_rate": 0.00016479438314944835,
"loss": 2.6916,
"theoretical_loss": 3.4004974385082143,
"tokens_seen": 2223833088
},
{
"epoch": 0.67,
"learning_rate": 0.00016478435305917756,
"loss": 2.8231,
"theoretical_loss": 3.400489256429118,
"tokens_seen": 2223898624
},
{
"epoch": 0.67,
"learning_rate": 0.00016477432296890674,
"loss": 2.6189,
"theoretical_loss": 3.4004810746586456,
"tokens_seen": 2223964160
},
{
"epoch": 0.67,
"learning_rate": 0.00016476429287863592,
"loss": 2.586,
"theoretical_loss": 3.400472893196777,
"tokens_seen": 2224029696
},
{
"epoch": 0.67,
"learning_rate": 0.0001647542627883651,
"loss": 2.6541,
"theoretical_loss": 3.4004647120434917,
"tokens_seen": 2224095232
},
{
"epoch": 0.67,
"learning_rate": 0.0001647442326980943,
"loss": 2.6805,
"theoretical_loss": 3.4004565311987687,
"tokens_seen": 2224160768
},
{
"epoch": 0.67,
"learning_rate": 0.0001647342026078235,
"loss": 2.4867,
"theoretical_loss": 3.400448350662587,
"tokens_seen": 2224226304
},
{
"epoch": 0.67,
"learning_rate": 0.00016472417251755267,
"loss": 2.6332,
"theoretical_loss": 3.4004401704349263,
"tokens_seen": 2224291840
},
{
"epoch": 0.67,
"learning_rate": 0.00016471414242728185,
"loss": 2.6278,
"theoretical_loss": 3.400431990515766,
"tokens_seen": 2224357376
},
{
"epoch": 0.67,
"learning_rate": 0.00016470411233701106,
"loss": 2.7068,
"theoretical_loss": 3.400423810905085,
"tokens_seen": 2224422912
},
{
"epoch": 0.67,
"learning_rate": 0.00016469408224674024,
"loss": 2.7033,
"theoretical_loss": 3.400415631602862,
"tokens_seen": 2224488448
},
{
"epoch": 0.67,
"learning_rate": 0.00016468405215646942,
"loss": 2.6039,
"theoretical_loss": 3.4004074526090777,
"tokens_seen": 2224553984
},
{
"epoch": 0.67,
"learning_rate": 0.0001646740220661986,
"loss": 2.5558,
"theoretical_loss": 3.4003992739237106,
"tokens_seen": 2224619520
},
{
"epoch": 0.67,
"learning_rate": 0.00016466399197592778,
"loss": 2.6991,
"theoretical_loss": 3.4003910955467394,
"tokens_seen": 2224685056
},
{
"epoch": 0.67,
"learning_rate": 0.000164653961885657,
"loss": 2.7423,
"theoretical_loss": 3.4003829174781446,
"tokens_seen": 2224750592
},
{
"epoch": 0.67,
"learning_rate": 0.00016464393179538617,
"loss": 2.6369,
"theoretical_loss": 3.400374739717905,
"tokens_seen": 2224816128
},
{
"epoch": 0.67,
"learning_rate": 0.00016463390170511535,
"loss": 2.6448,
"theoretical_loss": 3.400366562265999,
"tokens_seen": 2224881664
},
{
"epoch": 0.67,
"objective/train/docs_used": 1260594,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.703911542892456,
"objective/train/theoretical_loss": 3.4003583851224075,
"objective/train/tokens_used": 2245407200,
"theoretical_loss": 3.4003583851224075,
"tokens_seen": 2224947200
},
{
"epoch": 0.67,
"learning_rate": 0.00016462387161484454,
"loss": 2.3974,
"theoretical_loss": 3.4003583851224075,
"tokens_seen": 2224947200
},
{
"epoch": 0.67,
"learning_rate": 0.00016461384152457374,
"loss": 2.6156,
"theoretical_loss": 3.4003502082871084,
"tokens_seen": 2225012736
},
{
"epoch": 0.67,
"learning_rate": 0.00016460381143430292,
"loss": 2.7348,
"theoretical_loss": 3.400342031760082,
"tokens_seen": 2225078272
},
{
"epoch": 0.67,
"learning_rate": 0.0001645937813440321,
"loss": 2.2872,
"theoretical_loss": 3.400333855541307,
"tokens_seen": 2225143808
},
{
"epoch": 0.67,
"learning_rate": 0.0001645837512537613,
"loss": 2.6321,
"theoretical_loss": 3.400325679630763,
"tokens_seen": 2225209344
},
{
"epoch": 0.67,
"learning_rate": 0.00016457372116349047,
"loss": 2.6484,
"theoretical_loss": 3.400317504028429,
"tokens_seen": 2225274880
},
{
"epoch": 0.67,
"learning_rate": 0.00016456369107321968,
"loss": 2.7213,
"theoretical_loss": 3.4003093287342843,
"tokens_seen": 2225340416
},
{
"epoch": 0.67,
"learning_rate": 0.00016455366098294886,
"loss": 2.5086,
"theoretical_loss": 3.4003011537483085,
"tokens_seen": 2225405952
},
{
"epoch": 0.67,
"learning_rate": 0.00016454363089267804,
"loss": 2.694,
"theoretical_loss": 3.400292979070481,
"tokens_seen": 2225471488
},
{
"epoch": 0.67,
"learning_rate": 0.00016453360080240722,
"loss": 2.6411,
"theoretical_loss": 3.4002848047007808,
"tokens_seen": 2225537024
},
{
"epoch": 0.67,
"learning_rate": 0.00016452357071213643,
"loss": 2.6322,
"theoretical_loss": 3.4002766306391874,
"tokens_seen": 2225602560
},
{
"epoch": 0.67,
"learning_rate": 0.0001645135406218656,
"loss": 2.7011,
"theoretical_loss": 3.40026845688568,
"tokens_seen": 2225668096
},
{
"epoch": 0.67,
"learning_rate": 0.0001645035105315948,
"loss": 2.592,
"theoretical_loss": 3.400260283440238,
"tokens_seen": 2225733632
},
{
"epoch": 0.67,
"learning_rate": 0.00016449348044132397,
"loss": 2.3368,
"theoretical_loss": 3.4002521103028407,
"tokens_seen": 2225799168
},
{
"epoch": 0.67,
"learning_rate": 0.00016448345035105315,
"loss": 2.4259,
"theoretical_loss": 3.400243937473467,
"tokens_seen": 2225864704
},
{
"epoch": 0.67,
"learning_rate": 0.00016447342026078236,
"loss": 2.4876,
"theoretical_loss": 3.4002357649520967,
"tokens_seen": 2225930240
},
{
"epoch": 0.67,
"learning_rate": 0.00016446339017051154,
"loss": 2.7743,
"theoretical_loss": 3.4002275927387093,
"tokens_seen": 2225995776
},
{
"epoch": 0.67,
"learning_rate": 0.00016445336008024072,
"loss": 2.8682,
"theoretical_loss": 3.4002194208332837,
"tokens_seen": 2226061312
},
{
"epoch": 0.67,
"learning_rate": 0.0001644433299899699,
"loss": 2.5714,
"theoretical_loss": 3.4002112492357996,
"tokens_seen": 2226126848
},
{
"epoch": 0.67,
"learning_rate": 0.0001644332998996991,
"loss": 2.6151,
"theoretical_loss": 3.400203077946236,
"tokens_seen": 2226192384
},
{
"epoch": 0.67,
"learning_rate": 0.0001644232698094283,
"loss": 2.7914,
"theoretical_loss": 3.4001949069645723,
"tokens_seen": 2226257920
},
{
"epoch": 0.67,
"learning_rate": 0.00016441323971915747,
"loss": 2.3955,
"theoretical_loss": 3.400186736290788,
"tokens_seen": 2226323456
},
{
"epoch": 0.67,
"learning_rate": 0.00016440320962888666,
"loss": 2.7466,
"theoretical_loss": 3.400178565924862,
"tokens_seen": 2226388992
},
{
"epoch": 0.67,
"learning_rate": 0.00016439317953861586,
"loss": 2.5334,
"theoretical_loss": 3.400170395866774,
"tokens_seen": 2226454528
},
{
"epoch": 0.67,
"learning_rate": 0.00016438314944834504,
"loss": 2.5332,
"theoretical_loss": 3.400162226116504,
"tokens_seen": 2226520064
},
{
"epoch": 0.67,
"objective/train/docs_used": 1261463,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.3414580821990967,
"objective/train/theoretical_loss": 3.40015405667403,
"objective/train/tokens_used": 2247045600,
"theoretical_loss": 3.40015405667403,
"tokens_seen": 2226585600
},
{
"epoch": 0.67,
"learning_rate": 0.00016437311935807423,
"loss": 2.783,
"theoretical_loss": 3.40015405667403,
"tokens_seen": 2226585600
},
{
"epoch": 0.67,
"learning_rate": 0.0001643630892678034,
"loss": 2.6451,
"theoretical_loss": 3.4001458875393324,
"tokens_seen": 2226651136
},
{
"epoch": 0.67,
"learning_rate": 0.0001643530591775326,
"loss": 2.6217,
"theoretical_loss": 3.40013771871239,
"tokens_seen": 2226716672
},
{
"epoch": 0.67,
"learning_rate": 0.0001643430290872618,
"loss": 2.5343,
"theoretical_loss": 3.4001295501931823,
"tokens_seen": 2226782208
},
{
"epoch": 0.67,
"learning_rate": 0.00016433299899699098,
"loss": 2.6281,
"theoretical_loss": 3.4001213819816885,
"tokens_seen": 2226847744
},
{
"epoch": 0.67,
"learning_rate": 0.00016432296890672016,
"loss": 2.4573,
"theoretical_loss": 3.400113214077888,
"tokens_seen": 2226913280
},
{
"epoch": 0.67,
"learning_rate": 0.00016431293881644934,
"loss": 2.5187,
"theoretical_loss": 3.40010504648176,
"tokens_seen": 2226978816
},
{
"epoch": 0.67,
"learning_rate": 0.00016430290872617855,
"loss": 2.6034,
"theoretical_loss": 3.400096879193285,
"tokens_seen": 2227044352
},
{
"epoch": 0.67,
"learning_rate": 0.00016429287863590773,
"loss": 2.5568,
"theoretical_loss": 3.400088712212441,
"tokens_seen": 2227109888
},
{
"epoch": 0.67,
"learning_rate": 0.0001642828485456369,
"loss": 2.6298,
"theoretical_loss": 3.4000805455392076,
"tokens_seen": 2227175424
},
{
"epoch": 0.67,
"learning_rate": 0.0001642728184553661,
"loss": 2.5374,
"theoretical_loss": 3.4000723791735643,
"tokens_seen": 2227240960
},
{
"epoch": 0.67,
"learning_rate": 0.00016426278836509527,
"loss": 2.4207,
"theoretical_loss": 3.400064213115491,
"tokens_seen": 2227306496
},
{
"epoch": 0.67,
"learning_rate": 0.00016425275827482448,
"loss": 2.8279,
"theoretical_loss": 3.4000560473649664,
"tokens_seen": 2227372032
},
{
"epoch": 0.67,
"learning_rate": 0.00016424272818455366,
"loss": 2.8134,
"theoretical_loss": 3.40004788192197,
"tokens_seen": 2227437568
},
{
"epoch": 0.68,
"learning_rate": 0.00016423269809428284,
"loss": 2.5106,
"theoretical_loss": 3.4000397167864813,
"tokens_seen": 2227503104
},
{
"epoch": 0.68,
"learning_rate": 0.00016422266800401202,
"loss": 2.4542,
"theoretical_loss": 3.40003155195848,
"tokens_seen": 2227568640
},
{
"epoch": 0.68,
"learning_rate": 0.00016421263791374123,
"loss": 2.4856,
"theoretical_loss": 3.400023387437945,
"tokens_seen": 2227634176
},
{
"epoch": 0.68,
"learning_rate": 0.0001642026078234704,
"loss": 2.6002,
"theoretical_loss": 3.4000152232248553,
"tokens_seen": 2227699712
},
{
"epoch": 0.68,
"learning_rate": 0.0001641925777331996,
"loss": 2.6462,
"theoretical_loss": 3.400007059319191,
"tokens_seen": 2227765248
},
{
"epoch": 0.68,
"learning_rate": 0.00016418254764292878,
"loss": 2.785,
"theoretical_loss": 3.3999988957209313,
"tokens_seen": 2227830784
},
{
"epoch": 0.68,
"learning_rate": 0.00016417251755265796,
"loss": 2.5923,
"theoretical_loss": 3.3999907324300556,
"tokens_seen": 2227896320
},
{
"epoch": 0.68,
"learning_rate": 0.00016416248746238716,
"loss": 2.5189,
"theoretical_loss": 3.399982569446543,
"tokens_seen": 2227961856
},
{
"epoch": 0.68,
"learning_rate": 0.00016415245737211635,
"loss": 2.43,
"theoretical_loss": 3.399974406770373,
"tokens_seen": 2228027392
},
{
"epoch": 0.68,
"learning_rate": 0.00016414242728184553,
"loss": 2.7393,
"theoretical_loss": 3.3999662444015257,
"tokens_seen": 2228092928
},
{
"epoch": 0.68,
"learning_rate": 0.0001641323971915747,
"loss": 2.6337,
"theoretical_loss": 3.399958082339979,
"tokens_seen": 2228158464
},
{
"epoch": 0.68,
"objective/train/docs_used": 1262743,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6381685733795166,
"objective/train/theoretical_loss": 3.3999499205857138,
"objective/train/tokens_used": 2248684000,
"theoretical_loss": 3.3999499205857138,
"tokens_seen": 2228224000
},
{
"epoch": 0.68,
"learning_rate": 0.00016412236710130392,
"loss": 2.5684,
"theoretical_loss": 3.3999499205857138,
"tokens_seen": 2228224000
},
{
"epoch": 0.68,
"learning_rate": 0.0001641123370110331,
"loss": 2.5899,
"theoretical_loss": 3.399941759138709,
"tokens_seen": 2228289536
},
{
"epoch": 0.68,
"learning_rate": 0.00016410230692076228,
"loss": 2.5885,
"theoretical_loss": 3.399933597998943,
"tokens_seen": 2228355072
},
{
"epoch": 0.68,
"learning_rate": 0.00016409227683049146,
"loss": 2.5943,
"theoretical_loss": 3.399925437166397,
"tokens_seen": 2228420608
},
{
"epoch": 0.68,
"learning_rate": 0.00016408224674022064,
"loss": 2.5846,
"theoretical_loss": 3.399917276641049,
"tokens_seen": 2228486144
},
{
"epoch": 0.68,
"learning_rate": 0.00016407221664994985,
"loss": 2.6707,
"theoretical_loss": 3.399909116422879,
"tokens_seen": 2228551680
},
{
"epoch": 0.68,
"learning_rate": 0.00016406218655967903,
"loss": 2.545,
"theoretical_loss": 3.399900956511866,
"tokens_seen": 2228617216
},
{
"epoch": 0.68,
"learning_rate": 0.0001640521564694082,
"loss": 2.5003,
"theoretical_loss": 3.3998927969079897,
"tokens_seen": 2228682752
},
{
"epoch": 0.68,
"learning_rate": 0.0001640421263791374,
"loss": 2.6562,
"theoretical_loss": 3.3998846376112297,
"tokens_seen": 2228748288
},
{
"epoch": 0.68,
"learning_rate": 0.00016403209628886663,
"loss": 2.8298,
"theoretical_loss": 3.3998764786215654,
"tokens_seen": 2228813824
},
{
"epoch": 0.68,
"learning_rate": 0.0001640220661985958,
"loss": 2.4408,
"theoretical_loss": 3.3998683199389754,
"tokens_seen": 2228879360
},
{
"epoch": 0.68,
"learning_rate": 0.000164012036108325,
"loss": 2.6097,
"theoretical_loss": 3.3998601615634403,
"tokens_seen": 2228944896
},
{
"epoch": 0.68,
"learning_rate": 0.00016400200601805417,
"loss": 2.6873,
"theoretical_loss": 3.3998520034949387,
"tokens_seen": 2229010432
},
{
"epoch": 0.68,
"learning_rate": 0.00016399197592778335,
"loss": 2.5394,
"theoretical_loss": 3.39984384573345,
"tokens_seen": 2229075968
},
{
"epoch": 0.68,
"learning_rate": 0.00016398194583751256,
"loss": 2.5735,
"theoretical_loss": 3.3998356882789538,
"tokens_seen": 2229141504
},
{
"epoch": 0.68,
"learning_rate": 0.00016397191574724174,
"loss": 2.7286,
"theoretical_loss": 3.39982753113143,
"tokens_seen": 2229207040
},
{
"epoch": 0.68,
"learning_rate": 0.00016396188565697092,
"loss": 2.5702,
"theoretical_loss": 3.3998193742908573,
"tokens_seen": 2229272576
},
{
"epoch": 0.68,
"learning_rate": 0.0001639518555667001,
"loss": 2.6108,
"theoretical_loss": 3.3998112177572155,
"tokens_seen": 2229338112
},
{
"epoch": 0.68,
"learning_rate": 0.0001639418254764293,
"loss": 2.5892,
"theoretical_loss": 3.399803061530484,
"tokens_seen": 2229403648
},
{
"epoch": 0.68,
"learning_rate": 0.0001639317953861585,
"loss": 2.3623,
"theoretical_loss": 3.3997949056106425,
"tokens_seen": 2229469184
},
{
"epoch": 0.68,
"learning_rate": 0.00016392176529588767,
"loss": 2.4772,
"theoretical_loss": 3.39978674999767,
"tokens_seen": 2229534720
},
{
"epoch": 0.68,
"learning_rate": 0.00016391173520561686,
"loss": 2.6769,
"theoretical_loss": 3.399778594691546,
"tokens_seen": 2229600256
},
{
"epoch": 0.68,
"learning_rate": 0.00016390170511534606,
"loss": 2.5272,
"theoretical_loss": 3.39977043969225,
"tokens_seen": 2229665792
},
{
"epoch": 0.68,
"learning_rate": 0.00016389167502507524,
"loss": 2.5808,
"theoretical_loss": 3.3997622849997615,
"tokens_seen": 2229731328
},
{
"epoch": 0.68,
"learning_rate": 0.00016388164493480443,
"loss": 2.712,
"theoretical_loss": 3.3997541306140597,
"tokens_seen": 2229796864
},
{
"epoch": 0.68,
"objective/train/docs_used": 1263359,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3071436882019043,
"objective/train/theoretical_loss": 3.399745976535124,
"objective/train/tokens_used": 2250322400,
"theoretical_loss": 3.399745976535124,
"tokens_seen": 2229862400
},
{
"epoch": 0.68,
"learning_rate": 0.0001638716148445336,
"loss": 2.5434,
"theoretical_loss": 3.399745976535124,
"tokens_seen": 2229862400
},
{
"epoch": 0.68,
"learning_rate": 0.0001638615847542628,
"loss": 2.478,
"theoretical_loss": 3.3997378227629347,
"tokens_seen": 2229927936
},
{
"epoch": 0.68,
"learning_rate": 0.000163851554663992,
"loss": 2.5653,
"theoretical_loss": 3.39972966929747,
"tokens_seen": 2229993472
},
{
"epoch": 0.68,
"learning_rate": 0.00016384152457372118,
"loss": 2.6119,
"theoretical_loss": 3.3997215161387104,
"tokens_seen": 2230059008
},
{
"epoch": 0.68,
"learning_rate": 0.00016383149448345036,
"loss": 2.675,
"theoretical_loss": 3.3997133632866348,
"tokens_seen": 2230124544
},
{
"epoch": 0.68,
"learning_rate": 0.00016382146439317954,
"loss": 2.6427,
"theoretical_loss": 3.3997052107412227,
"tokens_seen": 2230190080
},
{
"epoch": 0.68,
"learning_rate": 0.00016381143430290875,
"loss": 2.596,
"theoretical_loss": 3.3996970585024537,
"tokens_seen": 2230255616
},
{
"epoch": 0.68,
"learning_rate": 0.00016380140421263793,
"loss": 2.5091,
"theoretical_loss": 3.399688906570307,
"tokens_seen": 2230321152
},
{
"epoch": 0.68,
"learning_rate": 0.0001637913741223671,
"loss": 2.6163,
"theoretical_loss": 3.3996807549447627,
"tokens_seen": 2230386688
},
{
"epoch": 0.68,
"learning_rate": 0.0001637813440320963,
"loss": 2.5888,
"theoretical_loss": 3.3996726036257994,
"tokens_seen": 2230452224
},
{
"epoch": 0.68,
"learning_rate": 0.00016377131394182547,
"loss": 2.7551,
"theoretical_loss": 3.399664452613397,
"tokens_seen": 2230517760
},
{
"epoch": 0.68,
"learning_rate": 0.00016376128385155468,
"loss": 2.6664,
"theoretical_loss": 3.3996563019075348,
"tokens_seen": 2230583296
},
{
"epoch": 0.68,
"learning_rate": 0.00016375125376128386,
"loss": 2.5233,
"theoretical_loss": 3.3996481515081927,
"tokens_seen": 2230648832
},
{
"epoch": 0.68,
"learning_rate": 0.00016374122367101304,
"loss": 2.5515,
"theoretical_loss": 3.3996400014153494,
"tokens_seen": 2230714368
},
{
"epoch": 0.68,
"learning_rate": 0.00016373119358074222,
"loss": 2.8734,
"theoretical_loss": 3.399631851628985,
"tokens_seen": 2230779904
},
{
"epoch": 0.68,
"learning_rate": 0.00016372116349047143,
"loss": 2.7659,
"theoretical_loss": 3.399623702149079,
"tokens_seen": 2230845440
},
{
"epoch": 0.68,
"learning_rate": 0.0001637111334002006,
"loss": 2.607,
"theoretical_loss": 3.3996155529756105,
"tokens_seen": 2230910976
},
{
"epoch": 0.68,
"learning_rate": 0.0001637011033099298,
"loss": 2.462,
"theoretical_loss": 3.399607404108559,
"tokens_seen": 2230976512
},
{
"epoch": 0.68,
"learning_rate": 0.00016369107321965898,
"loss": 2.4282,
"theoretical_loss": 3.3995992555479044,
"tokens_seen": 2231042048
},
{
"epoch": 0.68,
"learning_rate": 0.00016368104312938816,
"loss": 2.5396,
"theoretical_loss": 3.3995911072936256,
"tokens_seen": 2231107584
},
{
"epoch": 0.68,
"learning_rate": 0.00016367101303911736,
"loss": 2.2135,
"theoretical_loss": 3.399582959345703,
"tokens_seen": 2231173120
},
{
"epoch": 0.68,
"learning_rate": 0.00016366098294884655,
"loss": 2.7006,
"theoretical_loss": 3.399574811704115,
"tokens_seen": 2231238656
},
{
"epoch": 0.68,
"learning_rate": 0.00016365095285857573,
"loss": 2.6357,
"theoretical_loss": 3.3995666643688414,
"tokens_seen": 2231304192
},
{
"epoch": 0.68,
"learning_rate": 0.0001636409227683049,
"loss": 2.4764,
"theoretical_loss": 3.399558517339862,
"tokens_seen": 2231369728
},
{
"epoch": 0.68,
"learning_rate": 0.00016363089267803412,
"loss": 2.43,
"theoretical_loss": 3.399550370617156,
"tokens_seen": 2231435264
},
{
"epoch": 0.68,
"objective/train/docs_used": 1264337,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8029980659484863,
"objective/train/theoretical_loss": 3.3995422242007036,
"objective/train/tokens_used": 2251960800,
"theoretical_loss": 3.3995422242007036,
"tokens_seen": 2231500800
},
{
"epoch": 0.68,
"learning_rate": 0.0001636208625877633,
"loss": 2.6398,
"theoretical_loss": 3.3995422242007036,
"tokens_seen": 2231500800
},
{
"epoch": 0.68,
"learning_rate": 0.00016361083249749248,
"loss": 2.4955,
"theoretical_loss": 3.399534078090483,
"tokens_seen": 2231566336
},
{
"epoch": 0.68,
"learning_rate": 0.00016360080240722166,
"loss": 2.6546,
"theoretical_loss": 3.399525932286475,
"tokens_seen": 2231631872
},
{
"epoch": 0.68,
"learning_rate": 0.00016359077231695084,
"loss": 2.3292,
"theoretical_loss": 3.3995177867886577,
"tokens_seen": 2231697408
},
{
"epoch": 0.68,
"learning_rate": 0.00016358074222668005,
"loss": 2.5475,
"theoretical_loss": 3.399509641597012,
"tokens_seen": 2231762944
},
{
"epoch": 0.68,
"learning_rate": 0.00016357071213640923,
"loss": 2.5473,
"theoretical_loss": 3.399501496711517,
"tokens_seen": 2231828480
},
{
"epoch": 0.68,
"learning_rate": 0.0001635606820461384,
"loss": 2.3882,
"theoretical_loss": 3.3994933521321515,
"tokens_seen": 2231894016
},
{
"epoch": 0.68,
"learning_rate": 0.0001635506519558676,
"loss": 2.6156,
"theoretical_loss": 3.3994852078588957,
"tokens_seen": 2231959552
},
{
"epoch": 0.68,
"learning_rate": 0.0001635406218655968,
"loss": 2.4964,
"theoretical_loss": 3.3994770638917293,
"tokens_seen": 2232025088
},
{
"epoch": 0.68,
"learning_rate": 0.00016353059177532598,
"loss": 2.4702,
"theoretical_loss": 3.399468920230631,
"tokens_seen": 2232090624
},
{
"epoch": 0.68,
"learning_rate": 0.00016352056168505516,
"loss": 2.4514,
"theoretical_loss": 3.3994607768755807,
"tokens_seen": 2232156160
},
{
"epoch": 0.68,
"learning_rate": 0.00016351053159478434,
"loss": 2.8075,
"theoretical_loss": 3.399452633826558,
"tokens_seen": 2232221696
},
{
"epoch": 0.68,
"learning_rate": 0.00016350050150451353,
"loss": 2.6988,
"theoretical_loss": 3.3994444910835426,
"tokens_seen": 2232287232
},
{
"epoch": 0.68,
"learning_rate": 0.00016349047141424273,
"loss": 2.428,
"theoretical_loss": 3.3994363486465136,
"tokens_seen": 2232352768
},
{
"epoch": 0.68,
"learning_rate": 0.00016348044132397191,
"loss": 2.605,
"theoretical_loss": 3.399428206515451,
"tokens_seen": 2232418304
},
{
"epoch": 0.68,
"learning_rate": 0.0001634704112337011,
"loss": 2.627,
"theoretical_loss": 3.3994200646903336,
"tokens_seen": 2232483840
},
{
"epoch": 0.68,
"learning_rate": 0.00016346038114343028,
"loss": 2.8121,
"theoretical_loss": 3.3994119231711415,
"tokens_seen": 2232549376
},
{
"epoch": 0.68,
"learning_rate": 0.00016345035105315949,
"loss": 2.7927,
"theoretical_loss": 3.3994037819578544,
"tokens_seen": 2232614912
},
{
"epoch": 0.68,
"learning_rate": 0.00016344032096288867,
"loss": 2.7152,
"theoretical_loss": 3.399395641050451,
"tokens_seen": 2232680448
},
{
"epoch": 0.68,
"learning_rate": 0.00016343029087261785,
"loss": 2.6266,
"theoretical_loss": 3.3993875004489116,
"tokens_seen": 2232745984
},
{
"epoch": 0.68,
"learning_rate": 0.00016342026078234703,
"loss": 2.6203,
"theoretical_loss": 3.3993793601532154,
"tokens_seen": 2232811520
},
{
"epoch": 0.68,
"learning_rate": 0.0001634102306920762,
"loss": 2.6545,
"theoretical_loss": 3.399371220163342,
"tokens_seen": 2232877056
},
{
"epoch": 0.68,
"learning_rate": 0.00016340020060180542,
"loss": 2.4695,
"theoretical_loss": 3.399363080479271,
"tokens_seen": 2232942592
},
{
"epoch": 0.68,
"learning_rate": 0.0001633901705115346,
"loss": 2.5598,
"theoretical_loss": 3.399354941100982,
"tokens_seen": 2233008128
},
{
"epoch": 0.68,
"learning_rate": 0.00016338014042126378,
"loss": 2.5467,
"theoretical_loss": 3.399346802028454,
"tokens_seen": 2233073664
},
{
"epoch": 0.68,
"objective/train/docs_used": 1265060,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7534539699554443,
"objective/train/theoretical_loss": 3.3993386632616676,
"objective/train/tokens_used": 2253599200,
"theoretical_loss": 3.3993386632616676,
"tokens_seen": 2233139200
},
{
"epoch": 0.68,
"learning_rate": 0.00016337011033099296,
"loss": 2.7515,
"theoretical_loss": 3.3993386632616676,
"tokens_seen": 2233139200
},
{
"epoch": 0.68,
"learning_rate": 0.00016336008024072217,
"loss": 2.7081,
"theoretical_loss": 3.399330524800601,
"tokens_seen": 2233204736
},
{
"epoch": 0.68,
"learning_rate": 0.00016335005015045135,
"loss": 2.6341,
"theoretical_loss": 3.3993223866452347,
"tokens_seen": 2233270272
},
{
"epoch": 0.68,
"learning_rate": 0.00016334002006018053,
"loss": 2.8194,
"theoretical_loss": 3.399314248795548,
"tokens_seen": 2233335808
},
{
"epoch": 0.68,
"learning_rate": 0.0001633299899699097,
"loss": 2.6966,
"theoretical_loss": 3.3993061112515206,
"tokens_seen": 2233401344
},
{
"epoch": 0.68,
"learning_rate": 0.00016331995987963892,
"loss": 2.5882,
"theoretical_loss": 3.3992979740131317,
"tokens_seen": 2233466880
},
{
"epoch": 0.68,
"learning_rate": 0.0001633099297893681,
"loss": 2.5989,
"theoretical_loss": 3.399289837080361,
"tokens_seen": 2233532416
},
{
"epoch": 0.68,
"learning_rate": 0.00016329989969909728,
"loss": 2.4088,
"theoretical_loss": 3.399281700453188,
"tokens_seen": 2233597952
},
{
"epoch": 0.68,
"learning_rate": 0.00016328986960882646,
"loss": 2.6481,
"theoretical_loss": 3.399273564131593,
"tokens_seen": 2233663488
},
{
"epoch": 0.68,
"learning_rate": 0.00016327983951855567,
"loss": 2.5637,
"theoretical_loss": 3.3992654281155543,
"tokens_seen": 2233729024
},
{
"epoch": 0.68,
"learning_rate": 0.00016326980942828488,
"loss": 2.3709,
"theoretical_loss": 3.399257292405052,
"tokens_seen": 2233794560
},
{
"epoch": 0.68,
"learning_rate": 0.00016325977933801406,
"loss": 2.6419,
"theoretical_loss": 3.399249157000066,
"tokens_seen": 2233860096
},
{
"epoch": 0.68,
"learning_rate": 0.00016324974924774324,
"loss": 2.7084,
"theoretical_loss": 3.3992410219005755,
"tokens_seen": 2233925632
},
{
"epoch": 0.68,
"learning_rate": 0.00016323971915747242,
"loss": 2.6563,
"theoretical_loss": 3.3992328871065602,
"tokens_seen": 2233991168
},
{
"epoch": 0.68,
"learning_rate": 0.00016322968906720163,
"loss": 2.6596,
"theoretical_loss": 3.3992247526179997,
"tokens_seen": 2234056704
},
{
"epoch": 0.68,
"learning_rate": 0.0001632196589769308,
"loss": 2.6817,
"theoretical_loss": 3.3992166184348735,
"tokens_seen": 2234122240
},
{
"epoch": 0.68,
"learning_rate": 0.00016320962888666,
"loss": 2.669,
"theoretical_loss": 3.3992084845571613,
"tokens_seen": 2234187776
},
{
"epoch": 0.68,
"learning_rate": 0.00016319959879638918,
"loss": 2.5205,
"theoretical_loss": 3.399200350984842,
"tokens_seen": 2234253312
},
{
"epoch": 0.68,
"learning_rate": 0.00016318956870611836,
"loss": 2.8766,
"theoretical_loss": 3.3991922177178964,
"tokens_seen": 2234318848
},
{
"epoch": 0.68,
"learning_rate": 0.00016317953861584757,
"loss": 2.689,
"theoretical_loss": 3.399184084756303,
"tokens_seen": 2234384384
},
{
"epoch": 0.68,
"learning_rate": 0.00016316950852557675,
"loss": 2.7227,
"theoretical_loss": 3.3991759521000415,
"tokens_seen": 2234449920
},
{
"epoch": 0.68,
"learning_rate": 0.00016315947843530593,
"loss": 2.5651,
"theoretical_loss": 3.399167819749092,
"tokens_seen": 2234515456
},
{
"epoch": 0.68,
"learning_rate": 0.0001631494483450351,
"loss": 2.5838,
"theoretical_loss": 3.3991596877034342,
"tokens_seen": 2234580992
},
{
"epoch": 0.68,
"learning_rate": 0.00016313941825476432,
"loss": 2.6081,
"theoretical_loss": 3.399151555963047,
"tokens_seen": 2234646528
},
{
"epoch": 0.68,
"learning_rate": 0.0001631293881644935,
"loss": 2.8524,
"theoretical_loss": 3.3991434245279106,
"tokens_seen": 2234712064
},
{
"epoch": 0.68,
"objective/train/docs_used": 1266280,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6935136318206787,
"objective/train/theoretical_loss": 3.399135293398004,
"objective/train/tokens_used": 2255237600,
"theoretical_loss": 3.399135293398004,
"tokens_seen": 2234777600
},
{
"epoch": 0.68,
"learning_rate": 0.00016311935807422268,
"loss": 2.7109,
"theoretical_loss": 3.399135293398004,
"tokens_seen": 2234777600
},
{
"epoch": 0.68,
"learning_rate": 0.00016310932798395186,
"loss": 2.5535,
"theoretical_loss": 3.3991271625733073,
"tokens_seen": 2234843136
},
{
"epoch": 0.68,
"learning_rate": 0.00016309929789368104,
"loss": 2.8206,
"theoretical_loss": 3.3991190320538,
"tokens_seen": 2234908672
},
{
"epoch": 0.68,
"learning_rate": 0.00016308926780341025,
"loss": 2.4898,
"theoretical_loss": 3.399110901839461,
"tokens_seen": 2234974208
},
{
"epoch": 0.68,
"learning_rate": 0.00016307923771313943,
"loss": 2.6022,
"theoretical_loss": 3.3991027719302713,
"tokens_seen": 2235039744
},
{
"epoch": 0.68,
"learning_rate": 0.0001630692076228686,
"loss": 2.512,
"theoretical_loss": 3.399094642326209,
"tokens_seen": 2235105280
},
{
"epoch": 0.68,
"learning_rate": 0.0001630591775325978,
"loss": 2.4174,
"theoretical_loss": 3.399086513027255,
"tokens_seen": 2235170816
},
{
"epoch": 0.68,
"learning_rate": 0.000163049147442327,
"loss": 2.7684,
"theoretical_loss": 3.399078384033388,
"tokens_seen": 2235236352
},
{
"epoch": 0.68,
"learning_rate": 0.00016303911735205618,
"loss": 2.6466,
"theoretical_loss": 3.3990702553445877,
"tokens_seen": 2235301888
},
{
"epoch": 0.68,
"learning_rate": 0.00016302908726178536,
"loss": 2.6513,
"theoretical_loss": 3.399062126960834,
"tokens_seen": 2235367424
},
{
"epoch": 0.68,
"learning_rate": 0.00016301905717151454,
"loss": 2.4917,
"theoretical_loss": 3.3990539988821067,
"tokens_seen": 2235432960
},
{
"epoch": 0.68,
"learning_rate": 0.00016300902708124373,
"loss": 2.8966,
"theoretical_loss": 3.399045871108385,
"tokens_seen": 2235498496
},
{
"epoch": 0.68,
"learning_rate": 0.00016299899699097293,
"loss": 2.7452,
"theoretical_loss": 3.399037743639649,
"tokens_seen": 2235564032
},
{
"epoch": 0.68,
"learning_rate": 0.00016298896690070211,
"loss": 2.4712,
"theoretical_loss": 3.399029616475877,
"tokens_seen": 2235629568
},
{
"epoch": 0.68,
"learning_rate": 0.0001629789368104313,
"loss": 2.527,
"theoretical_loss": 3.3990214896170503,
"tokens_seen": 2235695104
},
{
"epoch": 0.68,
"learning_rate": 0.00016296890672016048,
"loss": 2.7138,
"theoretical_loss": 3.3990133630631476,
"tokens_seen": 2235760640
},
{
"epoch": 0.68,
"learning_rate": 0.00016295887662988969,
"loss": 2.6915,
"theoretical_loss": 3.399005236814149,
"tokens_seen": 2235826176
},
{
"epoch": 0.68,
"learning_rate": 0.00016294884653961887,
"loss": 2.6478,
"theoretical_loss": 3.3989971108700336,
"tokens_seen": 2235891712
},
{
"epoch": 0.68,
"learning_rate": 0.00016293881644934805,
"loss": 2.4487,
"theoretical_loss": 3.3989889852307815,
"tokens_seen": 2235957248
},
{
"epoch": 0.68,
"learning_rate": 0.00016292878635907723,
"loss": 2.8535,
"theoretical_loss": 3.3989808598963718,
"tokens_seen": 2236022784
},
{
"epoch": 0.68,
"learning_rate": 0.0001629187562688064,
"loss": 2.816,
"theoretical_loss": 3.3989727348667844,
"tokens_seen": 2236088320
},
{
"epoch": 0.68,
"learning_rate": 0.00016290872617853562,
"loss": 2.5421,
"theoretical_loss": 3.398964610141999,
"tokens_seen": 2236153856
},
{
"epoch": 0.68,
"learning_rate": 0.0001628986960882648,
"loss": 2.6456,
"theoretical_loss": 3.3989564857219956,
"tokens_seen": 2236219392
},
{
"epoch": 0.68,
"learning_rate": 0.00016288866599799398,
"loss": 2.7569,
"theoretical_loss": 3.398948361606753,
"tokens_seen": 2236284928
},
{
"epoch": 0.68,
"learning_rate": 0.00016287863590772316,
"loss": 2.7037,
"theoretical_loss": 3.398940237796251,
"tokens_seen": 2236350464
},
{
"epoch": 0.68,
"objective/train/docs_used": 1266935,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.868516683578491,
"objective/train/theoretical_loss": 3.3989321142904703,
"objective/train/tokens_used": 2256876000,
"theoretical_loss": 3.3989321142904703,
"tokens_seen": 2236416000
},
{
"epoch": 0.68,
"learning_rate": 0.00016286860581745237,
"loss": 2.7521,
"theoretical_loss": 3.3989321142904703,
"tokens_seen": 2236416000
},
{
"epoch": 0.68,
"learning_rate": 0.00016285857572718155,
"loss": 2.4876,
"theoretical_loss": 3.398923991089389,
"tokens_seen": 2236481536
},
{
"epoch": 0.68,
"learning_rate": 0.00016284854563691073,
"loss": 2.5602,
"theoretical_loss": 3.3989158681929883,
"tokens_seen": 2236547072
},
{
"epoch": 0.68,
"learning_rate": 0.0001628385155466399,
"loss": 2.6485,
"theoretical_loss": 3.3989077456012464,
"tokens_seen": 2236612608
},
{
"epoch": 0.68,
"learning_rate": 0.00016282848545636912,
"loss": 2.6401,
"theoretical_loss": 3.3988996233141435,
"tokens_seen": 2236678144
},
{
"epoch": 0.68,
"learning_rate": 0.0001628184553660983,
"loss": 2.6891,
"theoretical_loss": 3.3988915013316596,
"tokens_seen": 2236743680
},
{
"epoch": 0.68,
"learning_rate": 0.00016280842527582748,
"loss": 2.571,
"theoretical_loss": 3.398883379653774,
"tokens_seen": 2236809216
},
{
"epoch": 0.68,
"learning_rate": 0.00016279839518555666,
"loss": 2.7941,
"theoretical_loss": 3.3988752582804667,
"tokens_seen": 2236874752
},
{
"epoch": 0.68,
"learning_rate": 0.00016278836509528585,
"loss": 2.7294,
"theoretical_loss": 3.3988671372117167,
"tokens_seen": 2236940288
},
{
"epoch": 0.68,
"learning_rate": 0.00016277833500501505,
"loss": 2.6138,
"theoretical_loss": 3.398859016447504,
"tokens_seen": 2237005824
},
{
"epoch": 0.68,
"learning_rate": 0.00016276830491474423,
"loss": 2.5786,
"theoretical_loss": 3.3988508959878088,
"tokens_seen": 2237071360
},
{
"epoch": 0.68,
"learning_rate": 0.00016275827482447342,
"loss": 2.585,
"theoretical_loss": 3.39884277583261,
"tokens_seen": 2237136896
},
{
"epoch": 0.68,
"learning_rate": 0.0001627482447342026,
"loss": 2.3273,
"theoretical_loss": 3.3988346559818874,
"tokens_seen": 2237202432
},
{
"epoch": 0.68,
"learning_rate": 0.0001627382146439318,
"loss": 2.7236,
"theoretical_loss": 3.3988265364356205,
"tokens_seen": 2237267968
},
{
"epoch": 0.68,
"learning_rate": 0.000162728184553661,
"loss": 2.7675,
"theoretical_loss": 3.3988184171937896,
"tokens_seen": 2237333504
},
{
"epoch": 0.68,
"learning_rate": 0.00016271815446339017,
"loss": 2.4842,
"theoretical_loss": 3.3988102982563744,
"tokens_seen": 2237399040
},
{
"epoch": 0.68,
"learning_rate": 0.00016270812437311935,
"loss": 2.4126,
"theoretical_loss": 3.3988021796233534,
"tokens_seen": 2237464576
},
{
"epoch": 0.68,
"learning_rate": 0.00016269809428284853,
"loss": 2.4372,
"theoretical_loss": 3.3987940612947076,
"tokens_seen": 2237530112
},
{
"epoch": 0.68,
"learning_rate": 0.00016268806419257774,
"loss": 2.6073,
"theoretical_loss": 3.398785943270416,
"tokens_seen": 2237595648
},
{
"epoch": 0.68,
"learning_rate": 0.00016267803410230692,
"loss": 2.5831,
"theoretical_loss": 3.3987778255504586,
"tokens_seen": 2237661184
},
{
"epoch": 0.68,
"learning_rate": 0.0001626680040120361,
"loss": 2.6906,
"theoretical_loss": 3.3987697081348145,
"tokens_seen": 2237726720
},
{
"epoch": 0.68,
"learning_rate": 0.00016265797392176528,
"loss": 2.5975,
"theoretical_loss": 3.3987615910234643,
"tokens_seen": 2237792256
},
{
"epoch": 0.68,
"learning_rate": 0.0001626479438314945,
"loss": 2.5625,
"theoretical_loss": 3.3987534742163867,
"tokens_seen": 2237857792
},
{
"epoch": 0.68,
"learning_rate": 0.00016263791374122367,
"loss": 2.5303,
"theoretical_loss": 3.398745357713562,
"tokens_seen": 2237923328
},
{
"epoch": 0.68,
"learning_rate": 0.00016262788365095285,
"loss": 2.5275,
"theoretical_loss": 3.39873724151497,
"tokens_seen": 2237988864
},
{
"epoch": 0.68,
"objective/train/docs_used": 1268187,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.826174259185791,
"objective/train/theoretical_loss": 3.39872912562059,
"objective/train/tokens_used": 2258514400,
"theoretical_loss": 3.39872912562059,
"tokens_seen": 2238054400
},
{
"epoch": 0.68,
"learning_rate": 0.00016261785356068203,
"loss": 2.57,
"theoretical_loss": 3.39872912562059,
"tokens_seen": 2238054400
},
{
"epoch": 0.68,
"learning_rate": 0.00016260782347041121,
"loss": 2.7293,
"theoretical_loss": 3.3987210100304015,
"tokens_seen": 2238119936
},
{
"epoch": 0.68,
"learning_rate": 0.00016259779338014042,
"loss": 2.8125,
"theoretical_loss": 3.3987128947443845,
"tokens_seen": 2238185472
},
{
"epoch": 0.68,
"learning_rate": 0.0001625877632898696,
"loss": 2.7876,
"theoretical_loss": 3.398704779762519,
"tokens_seen": 2238251008
},
{
"epoch": 0.68,
"learning_rate": 0.00016257773319959878,
"loss": 2.9749,
"theoretical_loss": 3.398696665084784,
"tokens_seen": 2238316544
},
{
"epoch": 0.68,
"learning_rate": 0.00016256770310932797,
"loss": 2.7142,
"theoretical_loss": 3.39868855071116,
"tokens_seen": 2238382080
},
{
"epoch": 0.68,
"learning_rate": 0.00016255767301905717,
"loss": 2.718,
"theoretical_loss": 3.398680436641626,
"tokens_seen": 2238447616
},
{
"epoch": 0.68,
"learning_rate": 0.00016254764292878636,
"loss": 2.7367,
"theoretical_loss": 3.398672322876162,
"tokens_seen": 2238513152
},
{
"epoch": 0.68,
"learning_rate": 0.00016253761283851554,
"loss": 2.545,
"theoretical_loss": 3.398664209414748,
"tokens_seen": 2238578688
},
{
"epoch": 0.68,
"learning_rate": 0.00016252758274824474,
"loss": 2.481,
"theoretical_loss": 3.398656096257363,
"tokens_seen": 2238644224
},
{
"epoch": 0.68,
"learning_rate": 0.00016251755265797393,
"loss": 2.6524,
"theoretical_loss": 3.398647983403987,
"tokens_seen": 2238709760
},
{
"epoch": 0.68,
"learning_rate": 0.00016250752256770313,
"loss": 2.971,
"theoretical_loss": 3.3986398708546,
"tokens_seen": 2238775296
},
{
"epoch": 0.68,
"learning_rate": 0.00016249749247743231,
"loss": 2.632,
"theoretical_loss": 3.3986317586091817,
"tokens_seen": 2238840832
},
{
"epoch": 0.68,
"learning_rate": 0.0001624874623871615,
"loss": 2.608,
"theoretical_loss": 3.398623646667712,
"tokens_seen": 2238906368
},
{
"epoch": 0.68,
"learning_rate": 0.00016247743229689068,
"loss": 2.5555,
"theoretical_loss": 3.3986155350301694,
"tokens_seen": 2238971904
},
{
"epoch": 0.68,
"learning_rate": 0.00016246740220661989,
"loss": 2.6393,
"theoretical_loss": 3.398607423696535,
"tokens_seen": 2239037440
},
{
"epoch": 0.68,
"learning_rate": 0.00016245737211634907,
"loss": 2.718,
"theoretical_loss": 3.3985993126667875,
"tokens_seen": 2239102976
},
{
"epoch": 0.68,
"learning_rate": 0.00016244734202607825,
"loss": 2.5336,
"theoretical_loss": 3.3985912019409077,
"tokens_seen": 2239168512
},
{
"epoch": 0.68,
"learning_rate": 0.00016243731193580743,
"loss": 2.7767,
"theoretical_loss": 3.3985830915188746,
"tokens_seen": 2239234048
},
{
"epoch": 0.68,
"learning_rate": 0.0001624272818455366,
"loss": 2.6098,
"theoretical_loss": 3.398574981400668,
"tokens_seen": 2239299584
},
{
"epoch": 0.68,
"learning_rate": 0.00016241725175526582,
"loss": 2.7449,
"theoretical_loss": 3.3985668715862674,
"tokens_seen": 2239365120
},
{
"epoch": 0.68,
"learning_rate": 0.000162407221664995,
"loss": 2.5242,
"theoretical_loss": 3.398558762075653,
"tokens_seen": 2239430656
},
{
"epoch": 0.68,
"learning_rate": 0.00016239719157472418,
"loss": 2.4218,
"theoretical_loss": 3.3985506528688045,
"tokens_seen": 2239496192
},
{
"epoch": 0.68,
"learning_rate": 0.00016238716148445336,
"loss": 2.4815,
"theoretical_loss": 3.3985425439657013,
"tokens_seen": 2239561728
},
{
"epoch": 0.68,
"learning_rate": 0.00016237713139418257,
"loss": 2.5248,
"theoretical_loss": 3.398534435366323,
"tokens_seen": 2239627264
},
{
"epoch": 0.68,
"objective/train/docs_used": 1268791,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9225525856018066,
"objective/train/theoretical_loss": 3.3985263270706505,
"objective/train/tokens_used": 2260152800,
"theoretical_loss": 3.3985263270706505,
"tokens_seen": 2239692800
},
{
"epoch": 0.68,
"learning_rate": 0.00016236710130391175,
"loss": 2.7999,
"theoretical_loss": 3.3985263270706505,
"tokens_seen": 2239692800
},
{
"epoch": 0.68,
"learning_rate": 0.00016235707121364093,
"loss": 2.7577,
"theoretical_loss": 3.398518219078662,
"tokens_seen": 2239758336
},
{
"epoch": 0.68,
"learning_rate": 0.0001623470411233701,
"loss": 2.7913,
"theoretical_loss": 3.398510111390338,
"tokens_seen": 2239823872
},
{
"epoch": 0.68,
"learning_rate": 0.00016233701103309932,
"loss": 2.5985,
"theoretical_loss": 3.3985020040056586,
"tokens_seen": 2239889408
},
{
"epoch": 0.68,
"learning_rate": 0.0001623269809428285,
"loss": 2.508,
"theoretical_loss": 3.3984938969246024,
"tokens_seen": 2239954944
},
{
"epoch": 0.68,
"learning_rate": 0.00016231695085255768,
"loss": 2.5874,
"theoretical_loss": 3.3984857901471504,
"tokens_seen": 2240020480
},
{
"epoch": 0.68,
"learning_rate": 0.00016230692076228686,
"loss": 2.7864,
"theoretical_loss": 3.3984776836732817,
"tokens_seen": 2240086016
},
{
"epoch": 0.68,
"learning_rate": 0.00016229689067201605,
"loss": 2.4765,
"theoretical_loss": 3.398469577502976,
"tokens_seen": 2240151552
},
{
"epoch": 0.68,
"learning_rate": 0.00016228686058174525,
"loss": 2.5675,
"theoretical_loss": 3.3984614716362134,
"tokens_seen": 2240217088
},
{
"epoch": 0.68,
"learning_rate": 0.00016227683049147444,
"loss": 2.8003,
"theoretical_loss": 3.398453366072973,
"tokens_seen": 2240282624
},
{
"epoch": 0.68,
"learning_rate": 0.00016226680040120362,
"loss": 2.4794,
"theoretical_loss": 3.3984452608132356,
"tokens_seen": 2240348160
},
{
"epoch": 0.68,
"learning_rate": 0.0001622567703109328,
"loss": 2.5088,
"theoretical_loss": 3.39843715585698,
"tokens_seen": 2240413696
},
{
"epoch": 0.68,
"learning_rate": 0.000162246740220662,
"loss": 2.5514,
"theoretical_loss": 3.398429051204187,
"tokens_seen": 2240479232
},
{
"epoch": 0.68,
"learning_rate": 0.0001622367101303912,
"loss": 2.5178,
"theoretical_loss": 3.398420946854835,
"tokens_seen": 2240544768
},
{
"epoch": 0.68,
"learning_rate": 0.00016222668004012037,
"loss": 2.5689,
"theoretical_loss": 3.398412842808905,
"tokens_seen": 2240610304
},
{
"epoch": 0.68,
"learning_rate": 0.00016221664994984955,
"loss": 2.4033,
"theoretical_loss": 3.3984047390663754,
"tokens_seen": 2240675840
},
{
"epoch": 0.68,
"learning_rate": 0.00016220661985957873,
"loss": 2.4692,
"theoretical_loss": 3.3983966356272277,
"tokens_seen": 2240741376
},
{
"epoch": 0.68,
"learning_rate": 0.00016219658976930794,
"loss": 2.5592,
"theoretical_loss": 3.39838853249144,
"tokens_seen": 2240806912
},
{
"epoch": 0.68,
"learning_rate": 0.00016218655967903712,
"loss": 2.4402,
"theoretical_loss": 3.3983804296589932,
"tokens_seen": 2240872448
},
{
"epoch": 0.68,
"learning_rate": 0.0001621765295887663,
"loss": 2.4726,
"theoretical_loss": 3.398372327129867,
"tokens_seen": 2240937984
},
{
"epoch": 0.68,
"learning_rate": 0.00016216649949849548,
"loss": 2.5682,
"theoretical_loss": 3.3983642249040407,
"tokens_seen": 2241003520
},
{
"epoch": 0.68,
"learning_rate": 0.0001621564694082247,
"loss": 2.6982,
"theoretical_loss": 3.398356122981494,
"tokens_seen": 2241069056
},
{
"epoch": 0.68,
"learning_rate": 0.00016214643931795387,
"loss": 2.5616,
"theoretical_loss": 3.398348021362207,
"tokens_seen": 2241134592
},
{
"epoch": 0.68,
"learning_rate": 0.00016213640922768305,
"loss": 2.7988,
"theoretical_loss": 3.3983399200461597,
"tokens_seen": 2241200128
},
{
"epoch": 0.68,
"learning_rate": 0.00016212637913741223,
"loss": 2.508,
"theoretical_loss": 3.3983318190333316,
"tokens_seen": 2241265664
},
{
"epoch": 0.68,
"objective/train/docs_used": 1270246,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1365742683410645,
"objective/train/theoretical_loss": 3.3983237183237023,
"objective/train/tokens_used": 2261791200,
"theoretical_loss": 3.3983237183237023,
"tokens_seen": 2241331200
},
{
"epoch": 0.68,
"learning_rate": 0.00016211634904714141,
"loss": 2.6681,
"theoretical_loss": 3.3983237183237023,
"tokens_seen": 2241331200
},
{
"epoch": 0.68,
"learning_rate": 0.00016210631895687062,
"loss": 2.7746,
"theoretical_loss": 3.398315617917252,
"tokens_seen": 2241396736
},
{
"epoch": 0.68,
"learning_rate": 0.0001620962888665998,
"loss": 2.4718,
"theoretical_loss": 3.3983075178139597,
"tokens_seen": 2241462272
},
{
"epoch": 0.68,
"learning_rate": 0.00016208625877632898,
"loss": 2.5242,
"theoretical_loss": 3.3982994180138064,
"tokens_seen": 2241527808
},
{
"epoch": 0.68,
"learning_rate": 0.00016207622868605817,
"loss": 2.5092,
"theoretical_loss": 3.3982913185167707,
"tokens_seen": 2241593344
},
{
"epoch": 0.68,
"learning_rate": 0.00016206619859578737,
"loss": 2.6148,
"theoretical_loss": 3.3982832193228334,
"tokens_seen": 2241658880
},
{
"epoch": 0.68,
"learning_rate": 0.00016205616850551656,
"loss": 2.7099,
"theoretical_loss": 3.3982751204319737,
"tokens_seen": 2241724416
},
{
"epoch": 0.68,
"learning_rate": 0.00016204613841524574,
"loss": 2.362,
"theoretical_loss": 3.3982670218441715,
"tokens_seen": 2241789952
},
{
"epoch": 0.68,
"learning_rate": 0.00016203610832497492,
"loss": 2.447,
"theoretical_loss": 3.398258923559407,
"tokens_seen": 2241855488
},
{
"epoch": 0.68,
"learning_rate": 0.0001620260782347041,
"loss": 2.4835,
"theoretical_loss": 3.398250825577659,
"tokens_seen": 2241921024
},
{
"epoch": 0.68,
"learning_rate": 0.0001620160481444333,
"loss": 2.4261,
"theoretical_loss": 3.398242727898908,
"tokens_seen": 2241986560
},
{
"epoch": 0.68,
"learning_rate": 0.0001620060180541625,
"loss": 2.7573,
"theoretical_loss": 3.3982346305231346,
"tokens_seen": 2242052096
},
{
"epoch": 0.68,
"learning_rate": 0.00016199598796389167,
"loss": 2.7369,
"theoretical_loss": 3.398226533450317,
"tokens_seen": 2242117632
},
{
"epoch": 0.68,
"learning_rate": 0.00016198595787362085,
"loss": 2.6649,
"theoretical_loss": 3.3982184366804358,
"tokens_seen": 2242183168
},
{
"epoch": 0.68,
"learning_rate": 0.00016197592778335006,
"loss": 2.4128,
"theoretical_loss": 3.3982103402134713,
"tokens_seen": 2242248704
},
{
"epoch": 0.68,
"learning_rate": 0.00016196589769307924,
"loss": 2.509,
"theoretical_loss": 3.3982022440494024,
"tokens_seen": 2242314240
},
{
"epoch": 0.68,
"learning_rate": 0.00016195586760280842,
"loss": 2.5262,
"theoretical_loss": 3.3981941481882094,
"tokens_seen": 2242379776
},
{
"epoch": 0.68,
"learning_rate": 0.0001619458375125376,
"loss": 2.5535,
"theoretical_loss": 3.398186052629872,
"tokens_seen": 2242445312
},
{
"epoch": 0.68,
"learning_rate": 0.00016193580742226678,
"loss": 2.5085,
"theoretical_loss": 3.39817795737437,
"tokens_seen": 2242510848
},
{
"epoch": 0.68,
"learning_rate": 0.000161925777331996,
"loss": 2.5365,
"theoretical_loss": 3.3981698624216836,
"tokens_seen": 2242576384
},
{
"epoch": 0.68,
"learning_rate": 0.00016191574724172517,
"loss": 2.3924,
"theoretical_loss": 3.398161767771792,
"tokens_seen": 2242641920
},
{
"epoch": 0.68,
"learning_rate": 0.00016190571715145435,
"loss": 2.5364,
"theoretical_loss": 3.3981536734246753,
"tokens_seen": 2242707456
},
{
"epoch": 0.68,
"learning_rate": 0.00016189568706118353,
"loss": 2.6112,
"theoretical_loss": 3.3981455793803135,
"tokens_seen": 2242772992
},
{
"epoch": 0.68,
"learning_rate": 0.00016188565697091274,
"loss": 2.6729,
"theoretical_loss": 3.3981374856386863,
"tokens_seen": 2242838528
},
{
"epoch": 0.68,
"learning_rate": 0.00016187562688064192,
"loss": 2.5329,
"theoretical_loss": 3.3981293921997735,
"tokens_seen": 2242904064
},
{
"epoch": 0.68,
"objective/train/docs_used": 1270875,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4042739868164062,
"objective/train/theoretical_loss": 3.398121299063555,
"objective/train/tokens_used": 2263429600,
"theoretical_loss": 3.398121299063555,
"tokens_seen": 2242969600
},
{
"epoch": 0.68,
"learning_rate": 0.0001618655967903711,
"loss": 2.4867,
"theoretical_loss": 3.398121299063555,
"tokens_seen": 2242969600
},
{
"epoch": 0.68,
"learning_rate": 0.00016185556670010029,
"loss": 2.6634,
"theoretical_loss": 3.3981132062300103,
"tokens_seen": 2243035136
},
{
"epoch": 0.68,
"learning_rate": 0.00016184553660982947,
"loss": 2.6742,
"theoretical_loss": 3.3981051136991196,
"tokens_seen": 2243100672
},
{
"epoch": 0.68,
"learning_rate": 0.00016183550651955868,
"loss": 2.5438,
"theoretical_loss": 3.398097021470863,
"tokens_seen": 2243166208
},
{
"epoch": 0.68,
"learning_rate": 0.00016182547642928786,
"loss": 2.6446,
"theoretical_loss": 3.39808892954522,
"tokens_seen": 2243231744
},
{
"epoch": 0.68,
"learning_rate": 0.00016181544633901704,
"loss": 2.7393,
"theoretical_loss": 3.39808083792217,
"tokens_seen": 2243297280
},
{
"epoch": 0.68,
"learning_rate": 0.00016180541624874622,
"loss": 2.7556,
"theoretical_loss": 3.3980727466016933,
"tokens_seen": 2243362816
},
{
"epoch": 0.68,
"learning_rate": 0.00016179538615847543,
"loss": 2.4903,
"theoretical_loss": 3.39806465558377,
"tokens_seen": 2243428352
},
{
"epoch": 0.68,
"learning_rate": 0.00016178535606820464,
"loss": 2.4546,
"theoretical_loss": 3.39805656486838,
"tokens_seen": 2243493888
},
{
"epoch": 0.68,
"learning_rate": 0.00016177532597793382,
"loss": 2.7222,
"theoretical_loss": 3.3980484744555026,
"tokens_seen": 2243559424
},
{
"epoch": 0.68,
"learning_rate": 0.000161765295887663,
"loss": 2.5511,
"theoretical_loss": 3.3980403843451175,
"tokens_seen": 2243624960
},
{
"epoch": 0.68,
"learning_rate": 0.0001617552657973922,
"loss": 2.3736,
"theoretical_loss": 3.3980322945372055,
"tokens_seen": 2243690496
},
{
"epoch": 0.68,
"learning_rate": 0.0001617452357071214,
"loss": 2.7351,
"theoretical_loss": 3.3980242050317457,
"tokens_seen": 2243756032
},
{
"epoch": 0.68,
"learning_rate": 0.00016173520561685057,
"loss": 2.3555,
"theoretical_loss": 3.3980161158287183,
"tokens_seen": 2243821568
},
{
"epoch": 0.68,
"learning_rate": 0.00016172517552657975,
"loss": 2.8177,
"theoretical_loss": 3.3980080269281028,
"tokens_seen": 2243887104
},
{
"epoch": 0.68,
"learning_rate": 0.00016171514543630893,
"loss": 2.847,
"theoretical_loss": 3.397999938329879,
"tokens_seen": 2243952640
},
{
"epoch": 0.68,
"learning_rate": 0.00016170511534603814,
"loss": 2.5126,
"theoretical_loss": 3.3979918500340274,
"tokens_seen": 2244018176
},
{
"epoch": 0.68,
"learning_rate": 0.00016169508525576732,
"loss": 2.7088,
"theoretical_loss": 3.3979837620405275,
"tokens_seen": 2244083712
},
{
"epoch": 0.68,
"learning_rate": 0.0001616850551654965,
"loss": 2.5685,
"theoretical_loss": 3.397975674349359,
"tokens_seen": 2244149248
},
{
"epoch": 0.68,
"learning_rate": 0.00016167502507522568,
"loss": 2.7847,
"theoretical_loss": 3.3979675869605024,
"tokens_seen": 2244214784
},
{
"epoch": 0.68,
"learning_rate": 0.0001616649949849549,
"loss": 2.5934,
"theoretical_loss": 3.397959499873937,
"tokens_seen": 2244280320
},
{
"epoch": 0.68,
"learning_rate": 0.00016165496489468407,
"loss": 2.4919,
"theoretical_loss": 3.3979514130896424,
"tokens_seen": 2244345856
},
{
"epoch": 0.68,
"learning_rate": 0.00016164493480441325,
"loss": 2.6915,
"theoretical_loss": 3.3979433266075993,
"tokens_seen": 2244411392
},
{
"epoch": 0.68,
"learning_rate": 0.00016163490471414243,
"loss": 2.7122,
"theoretical_loss": 3.3979352404277874,
"tokens_seen": 2244476928
},
{
"epoch": 0.68,
"learning_rate": 0.00016162487462387161,
"loss": 2.5681,
"theoretical_loss": 3.3979271545501857,
"tokens_seen": 2244542464
},
{
"epoch": 0.68,
"objective/train/docs_used": 1272103,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.3544654846191406,
"objective/train/theoretical_loss": 3.397919068974775,
"objective/train/tokens_used": 2265068000,
"theoretical_loss": 3.397919068974775,
"tokens_seen": 2244608000
},
{
"epoch": 0.68,
"learning_rate": 0.00016161484453360082,
"loss": 2.51,
"theoretical_loss": 3.397919068974775,
"tokens_seen": 2244608000
},
{
"epoch": 0.68,
"learning_rate": 0.00016160481444333,
"loss": 2.5392,
"theoretical_loss": 3.397910983701535,
"tokens_seen": 2244673536
},
{
"epoch": 0.68,
"learning_rate": 0.00016159478435305918,
"loss": 2.5777,
"theoretical_loss": 3.3979028987304454,
"tokens_seen": 2244739072
},
{
"epoch": 0.68,
"learning_rate": 0.00016158475426278837,
"loss": 2.6156,
"theoretical_loss": 3.397894814061486,
"tokens_seen": 2244804608
},
{
"epoch": 0.68,
"learning_rate": 0.00016157472417251757,
"loss": 2.6433,
"theoretical_loss": 3.397886729694637,
"tokens_seen": 2244870144
},
{
"epoch": 0.68,
"learning_rate": 0.00016156469408224676,
"loss": 2.629,
"theoretical_loss": 3.397878645629878,
"tokens_seen": 2244935680
},
{
"epoch": 0.68,
"learning_rate": 0.00016155466399197594,
"loss": 2.7537,
"theoretical_loss": 3.3978705618671894,
"tokens_seen": 2245001216
},
{
"epoch": 0.68,
"learning_rate": 0.00016154463390170512,
"loss": 2.4609,
"theoretical_loss": 3.3978624784065508,
"tokens_seen": 2245066752
},
{
"epoch": 0.68,
"learning_rate": 0.0001615346038114343,
"loss": 2.4564,
"theoretical_loss": 3.3978543952479416,
"tokens_seen": 2245132288
},
{
"epoch": 0.68,
"learning_rate": 0.0001615245737211635,
"loss": 2.5001,
"theoretical_loss": 3.3978463123913425,
"tokens_seen": 2245197824
},
{
"epoch": 0.68,
"learning_rate": 0.0001615145436308927,
"loss": 2.6899,
"theoretical_loss": 3.397838229836733,
"tokens_seen": 2245263360
},
{
"epoch": 0.68,
"learning_rate": 0.00016150451354062187,
"loss": 2.53,
"theoretical_loss": 3.397830147584093,
"tokens_seen": 2245328896
},
{
"epoch": 0.68,
"learning_rate": 0.00016149448345035105,
"loss": 2.3155,
"theoretical_loss": 3.3978220656334024,
"tokens_seen": 2245394432
},
{
"epoch": 0.68,
"learning_rate": 0.00016148445336008026,
"loss": 2.6424,
"theoretical_loss": 3.3978139839846406,
"tokens_seen": 2245459968
},
{
"epoch": 0.68,
"learning_rate": 0.00016147442326980944,
"loss": 2.4676,
"theoretical_loss": 3.397805902637789,
"tokens_seen": 2245525504
},
{
"epoch": 0.68,
"learning_rate": 0.00016146439317953862,
"loss": 2.5071,
"theoretical_loss": 3.397797821592826,
"tokens_seen": 2245591040
},
{
"epoch": 0.68,
"learning_rate": 0.0001614543630892678,
"loss": 2.7383,
"theoretical_loss": 3.397789740849732,
"tokens_seen": 2245656576
},
{
"epoch": 0.68,
"learning_rate": 0.00016144433299899698,
"loss": 2.4919,
"theoretical_loss": 3.3977816604084876,
"tokens_seen": 2245722112
},
{
"epoch": 0.68,
"learning_rate": 0.0001614343029087262,
"loss": 2.4749,
"theoretical_loss": 3.397773580269072,
"tokens_seen": 2245787648
},
{
"epoch": 0.68,
"learning_rate": 0.00016142427281845537,
"loss": 2.5456,
"theoretical_loss": 3.397765500431465,
"tokens_seen": 2245853184
},
{
"epoch": 0.68,
"learning_rate": 0.00016141424272818455,
"loss": 2.5451,
"theoretical_loss": 3.3977574208956467,
"tokens_seen": 2245918720
},
{
"epoch": 0.68,
"learning_rate": 0.00016140421263791373,
"loss": 2.4315,
"theoretical_loss": 3.3977493416615974,
"tokens_seen": 2245984256
},
{
"epoch": 0.68,
"learning_rate": 0.00016139418254764294,
"loss": 2.7294,
"theoretical_loss": 3.3977412627292964,
"tokens_seen": 2246049792
},
{
"epoch": 0.68,
"learning_rate": 0.00016138415245737212,
"loss": 2.4474,
"theoretical_loss": 3.397733184098724,
"tokens_seen": 2246115328
},
{
"epoch": 0.68,
"learning_rate": 0.0001613741223671013,
"loss": 2.5241,
"theoretical_loss": 3.39772510576986,
"tokens_seen": 2246180864
},
{
"epoch": 0.68,
"objective/train/docs_used": 1272482,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.981611728668213,
"objective/train/theoretical_loss": 3.3977170277426847,
"objective/train/tokens_used": 2266706400,
"theoretical_loss": 3.3977170277426847,
"tokens_seen": 2246246400
},
{
"epoch": 0.68,
"learning_rate": 0.00016136409227683049,
"loss": 2.9096,
"theoretical_loss": 3.3977170277426847,
"tokens_seen": 2246246400
},
{
"epoch": 0.68,
"learning_rate": 0.00016135406218655967,
"loss": 2.583,
"theoretical_loss": 3.3977089500171775,
"tokens_seen": 2246311936
},
{
"epoch": 0.68,
"learning_rate": 0.00016134403209628888,
"loss": 2.2987,
"theoretical_loss": 3.3977008725933184,
"tokens_seen": 2246377472
},
{
"epoch": 0.68,
"learning_rate": 0.00016133400200601806,
"loss": 2.5489,
"theoretical_loss": 3.3976927954710874,
"tokens_seen": 2246443008
},
{
"epoch": 0.68,
"learning_rate": 0.00016132397191574724,
"loss": 2.667,
"theoretical_loss": 3.397684718650465,
"tokens_seen": 2246508544
},
{
"epoch": 0.68,
"learning_rate": 0.00016131394182547642,
"loss": 2.5022,
"theoretical_loss": 3.3976766421314304,
"tokens_seen": 2246574080
},
{
"epoch": 0.68,
"learning_rate": 0.00016130391173520563,
"loss": 2.3212,
"theoretical_loss": 3.397668565913964,
"tokens_seen": 2246639616
},
{
"epoch": 0.68,
"learning_rate": 0.0001612938816449348,
"loss": 2.6529,
"theoretical_loss": 3.3976604899980454,
"tokens_seen": 2246705152
},
{
"epoch": 0.68,
"learning_rate": 0.000161283851554664,
"loss": 2.6762,
"theoretical_loss": 3.3976524143836544,
"tokens_seen": 2246770688
},
{
"epoch": 0.68,
"learning_rate": 0.00016127382146439317,
"loss": 2.6184,
"theoretical_loss": 3.3976443390707716,
"tokens_seen": 2246836224
},
{
"epoch": 0.68,
"learning_rate": 0.00016126379137412235,
"loss": 2.7587,
"theoretical_loss": 3.3976362640593765,
"tokens_seen": 2246901760
},
{
"epoch": 0.68,
"learning_rate": 0.00016125376128385156,
"loss": 2.5885,
"theoretical_loss": 3.397628189349449,
"tokens_seen": 2246967296
},
{
"epoch": 0.68,
"learning_rate": 0.00016124373119358074,
"loss": 2.6109,
"theoretical_loss": 3.3976201149409695,
"tokens_seen": 2247032832
},
{
"epoch": 0.68,
"learning_rate": 0.00016123370110330992,
"loss": 2.6002,
"theoretical_loss": 3.3976120408339177,
"tokens_seen": 2247098368
},
{
"epoch": 0.68,
"learning_rate": 0.0001612236710130391,
"loss": 2.6977,
"theoretical_loss": 3.3976039670282736,
"tokens_seen": 2247163904
},
{
"epoch": 0.68,
"learning_rate": 0.0001612136409227683,
"loss": 2.4379,
"theoretical_loss": 3.3975958935240165,
"tokens_seen": 2247229440
},
{
"epoch": 0.68,
"learning_rate": 0.0001612036108324975,
"loss": 2.7949,
"theoretical_loss": 3.397587820321127,
"tokens_seen": 2247294976
},
{
"epoch": 0.68,
"learning_rate": 0.00016119358074222667,
"loss": 2.7464,
"theoretical_loss": 3.3975797474195857,
"tokens_seen": 2247360512
},
{
"epoch": 0.68,
"learning_rate": 0.00016118355065195585,
"loss": 2.8757,
"theoretical_loss": 3.397571674819371,
"tokens_seen": 2247426048
},
{
"epoch": 0.68,
"learning_rate": 0.00016117352056168506,
"loss": 2.5159,
"theoretical_loss": 3.3975636025204645,
"tokens_seen": 2247491584
},
{
"epoch": 0.68,
"learning_rate": 0.00016116349047141424,
"loss": 2.6037,
"theoretical_loss": 3.397555530522845,
"tokens_seen": 2247557120
},
{
"epoch": 0.68,
"learning_rate": 0.00016115346038114343,
"loss": 2.5772,
"theoretical_loss": 3.3975474588264927,
"tokens_seen": 2247622656
},
{
"epoch": 0.68,
"learning_rate": 0.0001611434302908726,
"loss": 2.5924,
"theoretical_loss": 3.397539387431388,
"tokens_seen": 2247688192
},
{
"epoch": 0.68,
"learning_rate": 0.0001611334002006018,
"loss": 2.7038,
"theoretical_loss": 3.3975313163375107,
"tokens_seen": 2247753728
},
{
"epoch": 0.68,
"learning_rate": 0.000161123370110331,
"loss": 2.4761,
"theoretical_loss": 3.39752324554484,
"tokens_seen": 2247819264
},
{
"epoch": 0.68,
"objective/train/docs_used": 1273740,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7811596393585205,
"objective/train/theoretical_loss": 3.3975151750533574,
"objective/train/tokens_used": 2268344800,
"theoretical_loss": 3.3975151750533574,
"tokens_seen": 2247884800
},
{
"epoch": 0.68,
"learning_rate": 0.00016111334002006018,
"loss": 2.6169,
"theoretical_loss": 3.3975151750533574,
"tokens_seen": 2247884800
},
{
"epoch": 0.68,
"learning_rate": 0.00016110330992978936,
"loss": 2.5977,
"theoretical_loss": 3.397507104863042,
"tokens_seen": 2247950336
},
{
"epoch": 0.68,
"learning_rate": 0.00016109327983951854,
"loss": 2.7507,
"theoretical_loss": 3.397499034973873,
"tokens_seen": 2248015872
},
{
"epoch": 0.68,
"learning_rate": 0.00016108324974924775,
"loss": 2.538,
"theoretical_loss": 3.397490965385832,
"tokens_seen": 2248081408
},
{
"epoch": 0.68,
"learning_rate": 0.00016107321965897693,
"loss": 2.7831,
"theoretical_loss": 3.397482896098898,
"tokens_seen": 2248146944
},
{
"epoch": 0.68,
"learning_rate": 0.0001610631895687061,
"loss": 2.5654,
"theoretical_loss": 3.397474827113051,
"tokens_seen": 2248212480
},
{
"epoch": 0.68,
"learning_rate": 0.0001610531594784353,
"loss": 2.2318,
"theoretical_loss": 3.397466758428271,
"tokens_seen": 2248278016
},
{
"epoch": 0.68,
"learning_rate": 0.00016104312938816447,
"loss": 2.5827,
"theoretical_loss": 3.3974586900445383,
"tokens_seen": 2248343552
},
{
"epoch": 0.68,
"learning_rate": 0.0001610330992978937,
"loss": 2.4757,
"theoretical_loss": 3.3974506219618332,
"tokens_seen": 2248409088
},
{
"epoch": 0.68,
"learning_rate": 0.0001610230692076229,
"loss": 2.5008,
"theoretical_loss": 3.3974425541801345,
"tokens_seen": 2248474624
},
{
"epoch": 0.68,
"learning_rate": 0.00016101303911735207,
"loss": 2.4856,
"theoretical_loss": 3.3974344866994235,
"tokens_seen": 2248540160
},
{
"epoch": 0.68,
"learning_rate": 0.00016100300902708125,
"loss": 2.6101,
"theoretical_loss": 3.3974264195196797,
"tokens_seen": 2248605696
},
{
"epoch": 0.68,
"learning_rate": 0.00016099297893681046,
"loss": 2.8839,
"theoretical_loss": 3.3974183526408828,
"tokens_seen": 2248671232
},
{
"epoch": 0.68,
"learning_rate": 0.00016098294884653964,
"loss": 2.5854,
"theoretical_loss": 3.397410286063013,
"tokens_seen": 2248736768
},
{
"epoch": 0.68,
"learning_rate": 0.00016097291875626882,
"loss": 2.6272,
"theoretical_loss": 3.3974022197860503,
"tokens_seen": 2248802304
},
{
"epoch": 0.68,
"learning_rate": 0.000160962888665998,
"loss": 2.5251,
"theoretical_loss": 3.3973941538099752,
"tokens_seen": 2248867840
},
{
"epoch": 0.68,
"learning_rate": 0.00016095285857572718,
"loss": 2.4237,
"theoretical_loss": 3.397386088134767,
"tokens_seen": 2248933376
},
{
"epoch": 0.68,
"learning_rate": 0.0001609428284854564,
"loss": 2.8061,
"theoretical_loss": 3.397378022760406,
"tokens_seen": 2248998912
},
{
"epoch": 0.68,
"learning_rate": 0.00016093279839518557,
"loss": 2.5356,
"theoretical_loss": 3.397369957686872,
"tokens_seen": 2249064448
},
{
"epoch": 0.68,
"learning_rate": 0.00016092276830491475,
"loss": 2.8981,
"theoretical_loss": 3.3973618929141454,
"tokens_seen": 2249129984
},
{
"epoch": 0.68,
"learning_rate": 0.00016091273821464393,
"loss": 2.7667,
"theoretical_loss": 3.397353828442206,
"tokens_seen": 2249195520
},
{
"epoch": 0.68,
"learning_rate": 0.00016090270812437314,
"loss": 2.6615,
"theoretical_loss": 3.3973457642710336,
"tokens_seen": 2249261056
},
{
"epoch": 0.68,
"learning_rate": 0.00016089267803410232,
"loss": 2.6565,
"theoretical_loss": 3.397337700400609,
"tokens_seen": 2249326592
},
{
"epoch": 0.68,
"learning_rate": 0.0001608826479438315,
"loss": 2.5347,
"theoretical_loss": 3.397329636830911,
"tokens_seen": 2249392128
},
{
"epoch": 0.68,
"learning_rate": 0.00016087261785356069,
"loss": 2.6024,
"theoretical_loss": 3.397321573561921,
"tokens_seen": 2249457664
},
{
"epoch": 0.68,
"objective/train/docs_used": 1274460,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.955536365509033,
"objective/train/theoretical_loss": 3.3973135105936176,
"objective/train/tokens_used": 2269983200,
"theoretical_loss": 3.3973135105936176,
"tokens_seen": 2249523200
},
{
"epoch": 0.68,
"learning_rate": 0.00016086258776328987,
"loss": 2.6984,
"theoretical_loss": 3.3973135105936176,
"tokens_seen": 2249523200
},
{
"epoch": 0.68,
"learning_rate": 0.00016085255767301908,
"loss": 2.4274,
"theoretical_loss": 3.397305447925982,
"tokens_seen": 2249588736
},
{
"epoch": 0.68,
"learning_rate": 0.00016084252758274826,
"loss": 2.7124,
"theoretical_loss": 3.3972973855589936,
"tokens_seen": 2249654272
},
{
"epoch": 0.68,
"learning_rate": 0.00016083249749247744,
"loss": 2.4484,
"theoretical_loss": 3.397289323492633,
"tokens_seen": 2249719808
},
{
"epoch": 0.68,
"learning_rate": 0.00016082246740220662,
"loss": 2.785,
"theoretical_loss": 3.3972812617268797,
"tokens_seen": 2249785344
},
{
"epoch": 0.68,
"learning_rate": 0.00016081243731193583,
"loss": 2.8009,
"theoretical_loss": 3.397273200261713,
"tokens_seen": 2249850880
},
{
"epoch": 0.68,
"learning_rate": 0.000160802407221665,
"loss": 2.6423,
"theoretical_loss": 3.397265139097115,
"tokens_seen": 2249916416
},
{
"epoch": 0.68,
"learning_rate": 0.0001607923771313942,
"loss": 2.5031,
"theoretical_loss": 3.397257078233064,
"tokens_seen": 2249981952
},
{
"epoch": 0.68,
"learning_rate": 0.00016078234704112337,
"loss": 2.4829,
"theoretical_loss": 3.397249017669541,
"tokens_seen": 2250047488
},
{
"epoch": 0.68,
"learning_rate": 0.00016077231695085255,
"loss": 2.5639,
"theoretical_loss": 3.397240957406525,
"tokens_seen": 2250113024
},
{
"epoch": 0.68,
"learning_rate": 0.00016076228686058176,
"loss": 2.5817,
"theoretical_loss": 3.397232897443997,
"tokens_seen": 2250178560
},
{
"epoch": 0.68,
"learning_rate": 0.00016075225677031094,
"loss": 2.6785,
"theoretical_loss": 3.397224837781937,
"tokens_seen": 2250244096
},
{
"epoch": 0.68,
"learning_rate": 0.00016074222668004012,
"loss": 2.5273,
"theoretical_loss": 3.397216778420325,
"tokens_seen": 2250309632
},
{
"epoch": 0.68,
"learning_rate": 0.0001607321965897693,
"loss": 2.6218,
"theoretical_loss": 3.3972087193591403,
"tokens_seen": 2250375168
},
{
"epoch": 0.68,
"learning_rate": 0.0001607221664994985,
"loss": 2.7792,
"theoretical_loss": 3.3972006605983633,
"tokens_seen": 2250440704
},
{
"epoch": 0.68,
"learning_rate": 0.0001607121364092277,
"loss": 2.5769,
"theoretical_loss": 3.397192602137975,
"tokens_seen": 2250506240
},
{
"epoch": 0.68,
"learning_rate": 0.00016070210631895687,
"loss": 2.768,
"theoretical_loss": 3.397184543977954,
"tokens_seen": 2250571776
},
{
"epoch": 0.68,
"learning_rate": 0.00016069207622868605,
"loss": 2.6492,
"theoretical_loss": 3.3971764861182816,
"tokens_seen": 2250637312
},
{
"epoch": 0.68,
"learning_rate": 0.00016068204613841526,
"loss": 2.496,
"theoretical_loss": 3.397168428558937,
"tokens_seen": 2250702848
},
{
"epoch": 0.68,
"learning_rate": 0.00016067201604814444,
"loss": 2.5405,
"theoretical_loss": 3.3971603712999006,
"tokens_seen": 2250768384
},
{
"epoch": 0.68,
"learning_rate": 0.00016066198595787363,
"loss": 2.4043,
"theoretical_loss": 3.3971523143411524,
"tokens_seen": 2250833920
},
{
"epoch": 0.68,
"learning_rate": 0.0001606519558676028,
"loss": 2.5563,
"theoretical_loss": 3.3971442576826725,
"tokens_seen": 2250899456
},
{
"epoch": 0.68,
"learning_rate": 0.000160641925777332,
"loss": 2.5669,
"theoretical_loss": 3.3971362013244413,
"tokens_seen": 2250964992
},
{
"epoch": 0.68,
"learning_rate": 0.0001606318956870612,
"loss": 2.3639,
"theoretical_loss": 3.3971281452664384,
"tokens_seen": 2251030528
},
{
"epoch": 0.68,
"learning_rate": 0.00016062186559679038,
"loss": 2.648,
"theoretical_loss": 3.3971200895086437,
"tokens_seen": 2251096064
},
{
"epoch": 0.68,
"objective/train/docs_used": 1275535,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8239128589630127,
"objective/train/theoretical_loss": 3.3971120340510375,
"objective/train/tokens_used": 2271621600,
"theoretical_loss": 3.3971120340510375,
"tokens_seen": 2251161600
},
{
"epoch": 0.68,
"learning_rate": 0.00016061183550651956,
"loss": 2.6891,
"theoretical_loss": 3.3971120340510375,
"tokens_seen": 2251161600
},
{
"epoch": 0.68,
"learning_rate": 0.00016060180541624874,
"loss": 2.8034,
"theoretical_loss": 3.3971039788936004,
"tokens_seen": 2251227136
},
{
"epoch": 0.68,
"learning_rate": 0.00016059177532597795,
"loss": 2.5459,
"theoretical_loss": 3.3970959240363117,
"tokens_seen": 2251292672
},
{
"epoch": 0.68,
"learning_rate": 0.00016058174523570713,
"loss": 2.5617,
"theoretical_loss": 3.397087869479152,
"tokens_seen": 2251358208
},
{
"epoch": 0.68,
"learning_rate": 0.0001605717151454363,
"loss": 2.5624,
"theoretical_loss": 3.3970798152221016,
"tokens_seen": 2251423744
},
{
"epoch": 0.68,
"learning_rate": 0.0001605616850551655,
"loss": 2.4609,
"theoretical_loss": 3.3970717612651398,
"tokens_seen": 2251489280
},
{
"epoch": 0.68,
"learning_rate": 0.00016055165496489467,
"loss": 2.6009,
"theoretical_loss": 3.397063707608247,
"tokens_seen": 2251554816
},
{
"epoch": 0.68,
"learning_rate": 0.00016054162487462388,
"loss": 2.6516,
"theoretical_loss": 3.397055654251403,
"tokens_seen": 2251620352
},
{
"epoch": 0.68,
"learning_rate": 0.00016053159478435306,
"loss": 2.6531,
"theoretical_loss": 3.397047601194589,
"tokens_seen": 2251685888
},
{
"epoch": 0.68,
"learning_rate": 0.00016052156469408224,
"loss": 2.5198,
"theoretical_loss": 3.397039548437784,
"tokens_seen": 2251751424
},
{
"epoch": 0.68,
"learning_rate": 0.00016051153460381142,
"loss": 2.625,
"theoretical_loss": 3.3970314959809684,
"tokens_seen": 2251816960
},
{
"epoch": 0.68,
"learning_rate": 0.00016050150451354063,
"loss": 2.8663,
"theoretical_loss": 3.3970234438241222,
"tokens_seen": 2251882496
},
{
"epoch": 0.68,
"learning_rate": 0.0001604914744232698,
"loss": 2.5788,
"theoretical_loss": 3.397015391967226,
"tokens_seen": 2251948032
},
{
"epoch": 0.68,
"learning_rate": 0.000160481444332999,
"loss": 2.8986,
"theoretical_loss": 3.3970073404102594,
"tokens_seen": 2252013568
},
{
"epoch": 0.68,
"learning_rate": 0.00016047141424272818,
"loss": 2.4722,
"theoretical_loss": 3.3969992891532024,
"tokens_seen": 2252079104
},
{
"epoch": 0.68,
"learning_rate": 0.00016046138415245736,
"loss": 2.5344,
"theoretical_loss": 3.3969912381960357,
"tokens_seen": 2252144640
},
{
"epoch": 0.68,
"learning_rate": 0.00016045135406218656,
"loss": 2.8131,
"theoretical_loss": 3.396983187538739,
"tokens_seen": 2252210176
},
{
"epoch": 0.68,
"learning_rate": 0.00016044132397191575,
"loss": 2.7275,
"theoretical_loss": 3.3969751371812915,
"tokens_seen": 2252275712
},
{
"epoch": 0.68,
"learning_rate": 0.00016043129388164493,
"loss": 2.6091,
"theoretical_loss": 3.396967087123675,
"tokens_seen": 2252341248
},
{
"epoch": 0.68,
"learning_rate": 0.0001604212637913741,
"loss": 2.6176,
"theoretical_loss": 3.396959037365869,
"tokens_seen": 2252406784
},
{
"epoch": 0.68,
"learning_rate": 0.00016041123370110332,
"loss": 2.7437,
"theoretical_loss": 3.3969509879078528,
"tokens_seen": 2252472320
},
{
"epoch": 0.68,
"learning_rate": 0.0001604012036108325,
"loss": 2.724,
"theoretical_loss": 3.396942938749608,
"tokens_seen": 2252537856
},
{
"epoch": 0.68,
"learning_rate": 0.00016039117352056168,
"loss": 2.3677,
"theoretical_loss": 3.396934889891113,
"tokens_seen": 2252603392
},
{
"epoch": 0.68,
"learning_rate": 0.00016038114343029086,
"loss": 2.5036,
"theoretical_loss": 3.3969268413323497,
"tokens_seen": 2252668928
},
{
"epoch": 0.68,
"learning_rate": 0.00016037111334002004,
"loss": 2.5513,
"theoretical_loss": 3.3969187930732967,
"tokens_seen": 2252734464
},
{
"epoch": 0.68,
"objective/train/docs_used": 1276002,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.853318214416504,
"objective/train/theoretical_loss": 3.396910745113935,
"objective/train/tokens_used": 2273260000,
"theoretical_loss": 3.396910745113935,
"tokens_seen": 2252800000
},
{
"epoch": 0.68,
"learning_rate": 0.00016036108324974925,
"loss": 2.7658,
"theoretical_loss": 3.396910745113935,
"tokens_seen": 2252800000
},
{
"epoch": 0.68,
"learning_rate": 0.00016035105315947843,
"loss": 2.7115,
"theoretical_loss": 3.3969026974542444,
"tokens_seen": 2252865536
},
{
"epoch": 0.68,
"learning_rate": 0.0001603410230692076,
"loss": 2.6107,
"theoretical_loss": 3.3968946500942048,
"tokens_seen": 2252931072
},
{
"epoch": 0.68,
"learning_rate": 0.0001603309929789368,
"loss": 2.7316,
"theoretical_loss": 3.396886603033797,
"tokens_seen": 2252996608
},
{
"epoch": 0.68,
"learning_rate": 0.000160320962888666,
"loss": 2.7354,
"theoretical_loss": 3.3968785562730006,
"tokens_seen": 2253062144
},
{
"epoch": 0.68,
"learning_rate": 0.00016031093279839518,
"loss": 2.7214,
"theoretical_loss": 3.396870509811796,
"tokens_seen": 2253127680
},
{
"epoch": 0.68,
"learning_rate": 0.00016030090270812436,
"loss": 2.5751,
"theoretical_loss": 3.396862463650163,
"tokens_seen": 2253193216
},
{
"epoch": 0.68,
"learning_rate": 0.00016029087261785354,
"loss": 2.2692,
"theoretical_loss": 3.3968544177880817,
"tokens_seen": 2253258752
},
{
"epoch": 0.68,
"learning_rate": 0.00016028084252758275,
"loss": 2.5415,
"theoretical_loss": 3.396846372225533,
"tokens_seen": 2253324288
},
{
"epoch": 0.68,
"learning_rate": 0.00016027081243731196,
"loss": 2.471,
"theoretical_loss": 3.3968383269624964,
"tokens_seen": 2253389824
},
{
"epoch": 0.68,
"learning_rate": 0.00016026078234704114,
"loss": 2.6484,
"theoretical_loss": 3.396830281998952,
"tokens_seen": 2253455360
},
{
"epoch": 0.68,
"learning_rate": 0.00016025075225677032,
"loss": 2.5116,
"theoretical_loss": 3.3968222373348804,
"tokens_seen": 2253520896
},
{
"epoch": 0.68,
"learning_rate": 0.0001602407221664995,
"loss": 2.6536,
"theoretical_loss": 3.396814192970261,
"tokens_seen": 2253586432
},
{
"epoch": 0.68,
"learning_rate": 0.0001602306920762287,
"loss": 2.6462,
"theoretical_loss": 3.3968061489050747,
"tokens_seen": 2253651968
},
{
"epoch": 0.68,
"learning_rate": 0.0001602206619859579,
"loss": 2.464,
"theoretical_loss": 3.3967981051393012,
"tokens_seen": 2253717504
},
{
"epoch": 0.68,
"learning_rate": 0.00016021063189568707,
"loss": 2.7839,
"theoretical_loss": 3.3967900616729207,
"tokens_seen": 2253783040
},
{
"epoch": 0.68,
"learning_rate": 0.00016020060180541626,
"loss": 2.6529,
"theoretical_loss": 3.3967820185059137,
"tokens_seen": 2253848576
},
{
"epoch": 0.68,
"learning_rate": 0.00016019057171514546,
"loss": 2.7564,
"theoretical_loss": 3.3967739756382596,
"tokens_seen": 2253914112
},
{
"epoch": 0.68,
"learning_rate": 0.00016018054162487464,
"loss": 2.934,
"theoretical_loss": 3.3967659330699393,
"tokens_seen": 2253979648
},
{
"epoch": 0.68,
"learning_rate": 0.00016017051153460383,
"loss": 2.6744,
"theoretical_loss": 3.396757890800933,
"tokens_seen": 2254045184
},
{
"epoch": 0.68,
"learning_rate": 0.000160160481444333,
"loss": 2.3583,
"theoretical_loss": 3.39674984883122,
"tokens_seen": 2254110720
},
{
"epoch": 0.68,
"learning_rate": 0.0001601504513540622,
"loss": 2.9607,
"theoretical_loss": 3.3967418071607813,
"tokens_seen": 2254176256
},
{
"epoch": 0.68,
"learning_rate": 0.0001601404212637914,
"loss": 2.8352,
"theoretical_loss": 3.396733765789597,
"tokens_seen": 2254241792
},
{
"epoch": 0.68,
"learning_rate": 0.00016013039117352058,
"loss": 2.6286,
"theoretical_loss": 3.3967257247176468,
"tokens_seen": 2254307328
},
{
"epoch": 0.68,
"learning_rate": 0.00016012036108324976,
"loss": 2.5735,
"theoretical_loss": 3.396717683944911,
"tokens_seen": 2254372864
},
{
"epoch": 0.68,
"objective/train/docs_used": 1277060,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5808379650115967,
"objective/train/theoretical_loss": 3.39670964347137,
"objective/train/tokens_used": 2274898400,
"theoretical_loss": 3.39670964347137,
"tokens_seen": 2254438400
},
{
"epoch": 0.68,
"learning_rate": 0.00016011033099297894,
"loss": 2.5503,
"theoretical_loss": 3.39670964347137,
"tokens_seen": 2254438400
},
{
"epoch": 0.68,
"learning_rate": 0.00016010030090270815,
"loss": 2.6541,
"theoretical_loss": 3.3967016032970037,
"tokens_seen": 2254503936
},
{
"epoch": 0.68,
"learning_rate": 0.00016009027081243733,
"loss": 2.4582,
"theoretical_loss": 3.3966935634217927,
"tokens_seen": 2254569472
},
{
"epoch": 0.68,
"learning_rate": 0.0001600802407221665,
"loss": 2.5742,
"theoretical_loss": 3.3966855238457168,
"tokens_seen": 2254635008
},
{
"epoch": 0.68,
"learning_rate": 0.0001600702106318957,
"loss": 2.6238,
"theoretical_loss": 3.396677484568756,
"tokens_seen": 2254700544
},
{
"epoch": 0.68,
"learning_rate": 0.00016006018054162487,
"loss": 2.6898,
"theoretical_loss": 3.396669445590891,
"tokens_seen": 2254766080
},
{
"epoch": 0.68,
"learning_rate": 0.00016005015045135408,
"loss": 2.7523,
"theoretical_loss": 3.3966614069121017,
"tokens_seen": 2254831616
},
{
"epoch": 0.68,
"learning_rate": 0.00016004012036108326,
"loss": 2.5272,
"theoretical_loss": 3.3966533685323683,
"tokens_seen": 2254897152
},
{
"epoch": 0.68,
"learning_rate": 0.00016003009027081244,
"loss": 2.7975,
"theoretical_loss": 3.3966453304516713,
"tokens_seen": 2254962688
},
{
"epoch": 0.68,
"learning_rate": 0.00016002006018054162,
"loss": 2.9059,
"theoretical_loss": 3.39663729266999,
"tokens_seen": 2255028224
},
{
"epoch": 0.68,
"learning_rate": 0.00016001003009027083,
"loss": 2.6861,
"theoretical_loss": 3.3966292551873054,
"tokens_seen": 2255093760
},
{
"epoch": 0.68,
"learning_rate": 0.00016,
"loss": 2.7759,
"theoretical_loss": 3.3966212180035975,
"tokens_seen": 2255159296
},
{
"epoch": 0.68,
"learning_rate": 0.0001599899699097292,
"loss": 2.8271,
"theoretical_loss": 3.3966131811188465,
"tokens_seen": 2255224832
},
{
"epoch": 0.68,
"learning_rate": 0.00015997993981945838,
"loss": 2.809,
"theoretical_loss": 3.3966051445330323,
"tokens_seen": 2255290368
},
{
"epoch": 0.68,
"learning_rate": 0.00015996990972918756,
"loss": 2.5855,
"theoretical_loss": 3.3965971082461355,
"tokens_seen": 2255355904
},
{
"epoch": 0.68,
"learning_rate": 0.00015995987963891676,
"loss": 2.6092,
"theoretical_loss": 3.396589072258136,
"tokens_seen": 2255421440
},
{
"epoch": 0.68,
"learning_rate": 0.00015994984954864595,
"loss": 2.897,
"theoretical_loss": 3.3965810365690143,
"tokens_seen": 2255486976
},
{
"epoch": 0.68,
"learning_rate": 0.00015993981945837513,
"loss": 2.698,
"theoretical_loss": 3.3965730011787505,
"tokens_seen": 2255552512
},
{
"epoch": 0.68,
"learning_rate": 0.0001599297893681043,
"loss": 2.6336,
"theoretical_loss": 3.3965649660873245,
"tokens_seen": 2255618048
},
{
"epoch": 0.68,
"learning_rate": 0.00015991975927783352,
"loss": 2.7566,
"theoretical_loss": 3.3965569312947164,
"tokens_seen": 2255683584
},
{
"epoch": 0.68,
"learning_rate": 0.0001599097291875627,
"loss": 2.5431,
"theoretical_loss": 3.3965488968009074,
"tokens_seen": 2255749120
},
{
"epoch": 0.68,
"learning_rate": 0.00015989969909729188,
"loss": 2.6635,
"theoretical_loss": 3.3965408626058764,
"tokens_seen": 2255814656
},
{
"epoch": 0.68,
"learning_rate": 0.00015988966900702106,
"loss": 2.4745,
"theoretical_loss": 3.3965328287096046,
"tokens_seen": 2255880192
},
{
"epoch": 0.68,
"learning_rate": 0.00015987963891675024,
"loss": 2.5221,
"theoretical_loss": 3.396524795112072,
"tokens_seen": 2255945728
},
{
"epoch": 0.68,
"learning_rate": 0.00015986960882647945,
"loss": 2.8535,
"theoretical_loss": 3.396516761813259,
"tokens_seen": 2256011264
},
{
"epoch": 0.68,
"objective/train/docs_used": 1277665,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9245715141296387,
"objective/train/theoretical_loss": 3.396508728813145,
"objective/train/tokens_used": 2276536800,
"theoretical_loss": 3.396508728813145,
"tokens_seen": 2256076800
},
{
"epoch": 0.68,
"learning_rate": 0.00015985957873620863,
"loss": 2.5463,
"theoretical_loss": 3.396508728813145,
"tokens_seen": 2256076800
},
{
"epoch": 0.68,
"learning_rate": 0.0001598495486459378,
"loss": 2.6997,
"theoretical_loss": 3.3965006961117106,
"tokens_seen": 2256142336
},
{
"epoch": 0.68,
"learning_rate": 0.000159839518555667,
"loss": 2.8574,
"theoretical_loss": 3.396492663708936,
"tokens_seen": 2256207872
},
{
"epoch": 0.68,
"learning_rate": 0.0001598294884653962,
"loss": 2.482,
"theoretical_loss": 3.396484631604802,
"tokens_seen": 2256273408
},
{
"epoch": 0.68,
"learning_rate": 0.00015981945837512538,
"loss": 2.7682,
"theoretical_loss": 3.3964765997992883,
"tokens_seen": 2256338944
},
{
"epoch": 0.68,
"learning_rate": 0.00015980942828485456,
"loss": 2.6429,
"theoretical_loss": 3.396468568292375,
"tokens_seen": 2256404480
},
{
"epoch": 0.68,
"learning_rate": 0.00015979939819458374,
"loss": 2.573,
"theoretical_loss": 3.3964605370840424,
"tokens_seen": 2256470016
},
{
"epoch": 0.68,
"learning_rate": 0.00015978936810431293,
"loss": 2.6471,
"theoretical_loss": 3.3964525061742714,
"tokens_seen": 2256535552
},
{
"epoch": 0.68,
"learning_rate": 0.00015977933801404213,
"loss": 2.4935,
"theoretical_loss": 3.396444475563041,
"tokens_seen": 2256601088
},
{
"epoch": 0.68,
"learning_rate": 0.00015976930792377131,
"loss": 2.7068,
"theoretical_loss": 3.3964364452503326,
"tokens_seen": 2256666624
},
{
"epoch": 0.68,
"learning_rate": 0.0001597592778335005,
"loss": 2.5511,
"theoretical_loss": 3.3964284152361257,
"tokens_seen": 2256732160
},
{
"epoch": 0.68,
"learning_rate": 0.00015974924774322968,
"loss": 2.8261,
"theoretical_loss": 3.396420385520401,
"tokens_seen": 2256797696
},
{
"epoch": 0.68,
"learning_rate": 0.00015973921765295888,
"loss": 2.5883,
"theoretical_loss": 3.3964123561031383,
"tokens_seen": 2256863232
},
{
"epoch": 0.68,
"learning_rate": 0.00015972918756268807,
"loss": 2.6528,
"theoretical_loss": 3.3964043269843183,
"tokens_seen": 2256928768
},
{
"epoch": 0.68,
"learning_rate": 0.00015971915747241725,
"loss": 2.4173,
"theoretical_loss": 3.396396298163921,
"tokens_seen": 2256994304
},
{
"epoch": 0.68,
"learning_rate": 0.00015970912738214643,
"loss": 2.7895,
"theoretical_loss": 3.3963882696419265,
"tokens_seen": 2257059840
},
{
"epoch": 0.68,
"learning_rate": 0.0001596990972918756,
"loss": 2.5922,
"theoretical_loss": 3.3963802414183153,
"tokens_seen": 2257125376
},
{
"epoch": 0.68,
"learning_rate": 0.00015968906720160482,
"loss": 2.7153,
"theoretical_loss": 3.396372213493067,
"tokens_seen": 2257190912
},
{
"epoch": 0.68,
"learning_rate": 0.000159679037111334,
"loss": 2.6861,
"theoretical_loss": 3.396364185866163,
"tokens_seen": 2257256448
},
{
"epoch": 0.68,
"learning_rate": 0.00015966900702106318,
"loss": 2.551,
"theoretical_loss": 3.396356158537583,
"tokens_seen": 2257321984
},
{
"epoch": 0.68,
"learning_rate": 0.00015965897693079236,
"loss": 2.5071,
"theoretical_loss": 3.396348131507307,
"tokens_seen": 2257387520
},
{
"epoch": 0.68,
"learning_rate": 0.00015964894684052157,
"loss": 2.4667,
"theoretical_loss": 3.3963401047753154,
"tokens_seen": 2257453056
},
{
"epoch": 0.68,
"learning_rate": 0.00015963891675025075,
"loss": 2.646,
"theoretical_loss": 3.3963320783415885,
"tokens_seen": 2257518592
},
{
"epoch": 0.68,
"learning_rate": 0.00015962888665997993,
"loss": 2.5447,
"theoretical_loss": 3.3963240522061064,
"tokens_seen": 2257584128
},
{
"epoch": 0.68,
"learning_rate": 0.0001596188565697091,
"loss": 2.4835,
"theoretical_loss": 3.39631602636885,
"tokens_seen": 2257649664
},
{
"epoch": 0.68,
"objective/train/docs_used": 1279074,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5337157249450684,
"objective/train/theoretical_loss": 3.396308000829799,
"objective/train/tokens_used": 2278175200,
"theoretical_loss": 3.396308000829799,
"tokens_seen": 2257715200
},
{
"epoch": 0.68,
"learning_rate": 0.0001596088264794383,
"loss": 2.4464,
"theoretical_loss": 3.396308000829799,
"tokens_seen": 2257715200
},
{
"epoch": 0.68,
"learning_rate": 0.0001595987963891675,
"loss": 2.7097,
"theoretical_loss": 3.396299975588933,
"tokens_seen": 2257780736
},
{
"epoch": 0.68,
"learning_rate": 0.00015958876629889668,
"loss": 2.8055,
"theoretical_loss": 3.396291950646234,
"tokens_seen": 2257846272
},
{
"epoch": 0.68,
"learning_rate": 0.00015957873620862586,
"loss": 2.7803,
"theoretical_loss": 3.396283926001681,
"tokens_seen": 2257911808
},
{
"epoch": 0.68,
"learning_rate": 0.00015956870611835505,
"loss": 2.513,
"theoretical_loss": 3.396275901655254,
"tokens_seen": 2257977344
},
{
"epoch": 0.68,
"learning_rate": 0.00015955867602808425,
"loss": 2.6277,
"theoretical_loss": 3.3962678776069346,
"tokens_seen": 2258042880
},
{
"epoch": 0.68,
"learning_rate": 0.00015954864593781343,
"loss": 2.894,
"theoretical_loss": 3.396259853856702,
"tokens_seen": 2258108416
},
{
"epoch": 0.68,
"learning_rate": 0.00015953861584754262,
"loss": 2.6191,
"theoretical_loss": 3.3962518304045366,
"tokens_seen": 2258173952
},
{
"epoch": 0.68,
"learning_rate": 0.00015952858575727182,
"loss": 2.5684,
"theoretical_loss": 3.396243807250419,
"tokens_seen": 2258239488
},
{
"epoch": 0.68,
"learning_rate": 0.00015951855566700103,
"loss": 2.4342,
"theoretical_loss": 3.3962357843943294,
"tokens_seen": 2258305024
},
{
"epoch": 0.68,
"learning_rate": 0.0001595085255767302,
"loss": 2.2871,
"theoretical_loss": 3.396227761836248,
"tokens_seen": 2258370560
},
{
"epoch": 0.68,
"learning_rate": 0.0001594984954864594,
"loss": 2.3774,
"theoretical_loss": 3.396219739576155,
"tokens_seen": 2258436096
},
{
"epoch": 0.68,
"learning_rate": 0.00015948846539618858,
"loss": 2.512,
"theoretical_loss": 3.3962117176140305,
"tokens_seen": 2258501632
},
{
"epoch": 0.68,
"learning_rate": 0.00015947843530591776,
"loss": 2.4849,
"theoretical_loss": 3.396203695949856,
"tokens_seen": 2258567168
},
{
"epoch": 0.68,
"learning_rate": 0.00015946840521564696,
"loss": 2.5814,
"theoretical_loss": 3.39619567458361,
"tokens_seen": 2258632704
},
{
"epoch": 0.68,
"learning_rate": 0.00015945837512537615,
"loss": 2.5147,
"theoretical_loss": 3.396187653515274,
"tokens_seen": 2258698240
},
{
"epoch": 0.68,
"learning_rate": 0.00015944834503510533,
"loss": 2.3623,
"theoretical_loss": 3.3961796327448277,
"tokens_seen": 2258763776
},
{
"epoch": 0.68,
"learning_rate": 0.0001594383149448345,
"loss": 2.7303,
"theoretical_loss": 3.396171612272252,
"tokens_seen": 2258829312
},
{
"epoch": 0.68,
"learning_rate": 0.00015942828485456372,
"loss": 2.5004,
"theoretical_loss": 3.3961635920975266,
"tokens_seen": 2258894848
},
{
"epoch": 0.68,
"learning_rate": 0.0001594182547642929,
"loss": 2.674,
"theoretical_loss": 3.396155572220632,
"tokens_seen": 2258960384
},
{
"epoch": 0.68,
"learning_rate": 0.00015940822467402208,
"loss": 2.6034,
"theoretical_loss": 3.3961475526415486,
"tokens_seen": 2259025920
},
{
"epoch": 0.68,
"learning_rate": 0.00015939819458375126,
"loss": 2.4426,
"theoretical_loss": 3.396139533360256,
"tokens_seen": 2259091456
},
{
"epoch": 0.68,
"learning_rate": 0.00015938816449348044,
"loss": 2.7095,
"theoretical_loss": 3.3961315143767363,
"tokens_seen": 2259156992
},
{
"epoch": 0.68,
"learning_rate": 0.00015937813440320965,
"loss": 2.598,
"theoretical_loss": 3.396123495690968,
"tokens_seen": 2259222528
},
{
"epoch": 0.68,
"learning_rate": 0.00015936810431293883,
"loss": 2.6494,
"theoretical_loss": 3.3961154773029323,
"tokens_seen": 2259288064
},
{
"epoch": 0.68,
"objective/train/docs_used": 1279794,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7419610023498535,
"objective/train/theoretical_loss": 3.396107459212609,
"objective/train/tokens_used": 2279813600,
"theoretical_loss": 3.396107459212609,
"tokens_seen": 2259353600
},
{
"epoch": 0.68,
"learning_rate": 0.000159358074222668,
"loss": 2.7909,
"theoretical_loss": 3.396107459212609,
"tokens_seen": 2259353600
},
{
"epoch": 0.68,
"learning_rate": 0.0001593480441323972,
"loss": 2.4938,
"theoretical_loss": 3.3960994414199783,
"tokens_seen": 2259419136
},
{
"epoch": 0.68,
"learning_rate": 0.0001593380140421264,
"loss": 2.7027,
"theoretical_loss": 3.3960914239250215,
"tokens_seen": 2259484672
},
{
"epoch": 0.68,
"learning_rate": 0.00015932798395185558,
"loss": 2.3945,
"theoretical_loss": 3.396083406727718,
"tokens_seen": 2259550208
},
{
"epoch": 0.68,
"learning_rate": 0.00015931795386158476,
"loss": 2.7701,
"theoretical_loss": 3.3960753898280487,
"tokens_seen": 2259615744
},
{
"epoch": 0.68,
"learning_rate": 0.00015930792377131394,
"loss": 2.7996,
"theoretical_loss": 3.3960673732259936,
"tokens_seen": 2259681280
},
{
"epoch": 0.68,
"learning_rate": 0.00015929789368104313,
"loss": 2.4798,
"theoretical_loss": 3.396059356921533,
"tokens_seen": 2259746816
},
{
"epoch": 0.68,
"learning_rate": 0.00015928786359077233,
"loss": 2.5921,
"theoretical_loss": 3.396051340914647,
"tokens_seen": 2259812352
},
{
"epoch": 0.68,
"learning_rate": 0.00015927783350050151,
"loss": 2.7213,
"theoretical_loss": 3.3960433252053166,
"tokens_seen": 2259877888
},
{
"epoch": 0.68,
"learning_rate": 0.0001592678034102307,
"loss": 2.6322,
"theoretical_loss": 3.3960353097935214,
"tokens_seen": 2259943424
},
{
"epoch": 0.68,
"learning_rate": 0.00015925777331995988,
"loss": 2.7179,
"theoretical_loss": 3.396027294679242,
"tokens_seen": 2260008960
},
{
"epoch": 0.68,
"learning_rate": 0.00015924774322968909,
"loss": 2.771,
"theoretical_loss": 3.3960192798624593,
"tokens_seen": 2260074496
},
{
"epoch": 0.68,
"learning_rate": 0.00015923771313941827,
"loss": 2.5088,
"theoretical_loss": 3.3960112653431525,
"tokens_seen": 2260140032
},
{
"epoch": 0.68,
"learning_rate": 0.00015922768304914745,
"loss": 2.7564,
"theoretical_loss": 3.396003251121303,
"tokens_seen": 2260205568
},
{
"epoch": 0.68,
"learning_rate": 0.00015921765295887663,
"loss": 2.7081,
"theoretical_loss": 3.3959952371968907,
"tokens_seen": 2260271104
},
{
"epoch": 0.68,
"learning_rate": 0.0001592076228686058,
"loss": 2.7277,
"theoretical_loss": 3.3959872235698954,
"tokens_seen": 2260336640
},
{
"epoch": 0.68,
"learning_rate": 0.00015919759277833502,
"loss": 2.7085,
"theoretical_loss": 3.3959792102402986,
"tokens_seen": 2260402176
},
{
"epoch": 0.68,
"learning_rate": 0.0001591875626880642,
"loss": 2.535,
"theoretical_loss": 3.3959711972080795,
"tokens_seen": 2260467712
},
{
"epoch": 0.69,
"learning_rate": 0.00015917753259779338,
"loss": 2.559,
"theoretical_loss": 3.395963184473219,
"tokens_seen": 2260533248
},
{
"epoch": 0.69,
"learning_rate": 0.00015916750250752256,
"loss": 2.4842,
"theoretical_loss": 3.3959551720356975,
"tokens_seen": 2260598784
},
{
"epoch": 0.69,
"learning_rate": 0.00015915747241725177,
"loss": 2.7283,
"theoretical_loss": 3.3959471598954956,
"tokens_seen": 2260664320
},
{
"epoch": 0.69,
"learning_rate": 0.00015914744232698095,
"loss": 2.6982,
"theoretical_loss": 3.3959391480525927,
"tokens_seen": 2260729856
},
{
"epoch": 0.69,
"learning_rate": 0.00015913741223671013,
"loss": 2.6517,
"theoretical_loss": 3.39593113650697,
"tokens_seen": 2260795392
},
{
"epoch": 0.69,
"learning_rate": 0.0001591273821464393,
"loss": 2.819,
"theoretical_loss": 3.3959231252586077,
"tokens_seen": 2260860928
},
{
"epoch": 0.69,
"learning_rate": 0.0001591173520561685,
"loss": 2.3179,
"theoretical_loss": 3.3959151143074857,
"tokens_seen": 2260926464
},
{
"epoch": 0.69,
"objective/train/docs_used": 1281070,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6713414192199707,
"objective/train/theoretical_loss": 3.395907103653585,
"objective/train/tokens_used": 2281452000,
"theoretical_loss": 3.395907103653585,
"tokens_seen": 2260992000
},
{
"epoch": 0.69,
"learning_rate": 0.0001591073219658977,
"loss": 2.4445,
"theoretical_loss": 3.395907103653585,
"tokens_seen": 2260992000
},
{
"epoch": 0.69,
"learning_rate": 0.00015909729187562688,
"loss": 2.5021,
"theoretical_loss": 3.3958990932968858,
"tokens_seen": 2261057536
},
{
"epoch": 0.69,
"learning_rate": 0.00015908726178535606,
"loss": 2.6393,
"theoretical_loss": 3.395891083237368,
"tokens_seen": 2261123072
},
{
"epoch": 0.69,
"learning_rate": 0.00015907723169508525,
"loss": 2.6166,
"theoretical_loss": 3.3958830734750123,
"tokens_seen": 2261188608
},
{
"epoch": 0.69,
"learning_rate": 0.00015906720160481445,
"loss": 2.5725,
"theoretical_loss": 3.3958750640097994,
"tokens_seen": 2261254144
},
{
"epoch": 0.69,
"learning_rate": 0.00015905717151454363,
"loss": 2.3361,
"theoretical_loss": 3.395867054841709,
"tokens_seen": 2261319680
},
{
"epoch": 0.69,
"learning_rate": 0.00015904714142427282,
"loss": 2.3884,
"theoretical_loss": 3.3958590459707216,
"tokens_seen": 2261385216
},
{
"epoch": 0.69,
"learning_rate": 0.000159037111334002,
"loss": 2.3416,
"theoretical_loss": 3.3958510373968176,
"tokens_seen": 2261450752
},
{
"epoch": 0.69,
"learning_rate": 0.0001590270812437312,
"loss": 2.4737,
"theoretical_loss": 3.395843029119978,
"tokens_seen": 2261516288
},
{
"epoch": 0.69,
"learning_rate": 0.00015901705115346039,
"loss": 2.2378,
"theoretical_loss": 3.395835021140183,
"tokens_seen": 2261581824
},
{
"epoch": 0.69,
"learning_rate": 0.00015900702106318957,
"loss": 2.6032,
"theoretical_loss": 3.395827013457412,
"tokens_seen": 2261647360
},
{
"epoch": 0.69,
"learning_rate": 0.00015899699097291875,
"loss": 2.5837,
"theoretical_loss": 3.395819006071646,
"tokens_seen": 2261712896
},
{
"epoch": 0.69,
"learning_rate": 0.00015898696088264793,
"loss": 2.5229,
"theoretical_loss": 3.3958109989828658,
"tokens_seen": 2261778432
},
{
"epoch": 0.69,
"learning_rate": 0.00015897693079237714,
"loss": 2.5796,
"theoretical_loss": 3.3958029921910513,
"tokens_seen": 2261843968
},
{
"epoch": 0.69,
"learning_rate": 0.00015896690070210632,
"loss": 2.6741,
"theoretical_loss": 3.395794985696183,
"tokens_seen": 2261909504
},
{
"epoch": 0.69,
"learning_rate": 0.0001589568706118355,
"loss": 2.8534,
"theoretical_loss": 3.395786979498241,
"tokens_seen": 2261975040
},
{
"epoch": 0.69,
"learning_rate": 0.00015894684052156468,
"loss": 2.6307,
"theoretical_loss": 3.3957789735972064,
"tokens_seen": 2262040576
},
{
"epoch": 0.69,
"learning_rate": 0.0001589368104312939,
"loss": 2.4846,
"theoretical_loss": 3.3957709679930588,
"tokens_seen": 2262106112
},
{
"epoch": 0.69,
"learning_rate": 0.00015892678034102307,
"loss": 2.4787,
"theoretical_loss": 3.395762962685779,
"tokens_seen": 2262171648
},
{
"epoch": 0.69,
"learning_rate": 0.00015891675025075225,
"loss": 2.6135,
"theoretical_loss": 3.395754957675347,
"tokens_seen": 2262237184
},
{
"epoch": 0.69,
"learning_rate": 0.00015890672016048143,
"loss": 2.5439,
"theoretical_loss": 3.395746952961744,
"tokens_seen": 2262302720
},
{
"epoch": 0.69,
"learning_rate": 0.00015889669007021061,
"loss": 2.8846,
"theoretical_loss": 3.39573894854495,
"tokens_seen": 2262368256
},
{
"epoch": 0.69,
"learning_rate": 0.00015888665997993982,
"loss": 2.6097,
"theoretical_loss": 3.395730944424945,
"tokens_seen": 2262433792
},
{
"epoch": 0.69,
"learning_rate": 0.000158876629889669,
"loss": 2.5972,
"theoretical_loss": 3.3957229406017095,
"tokens_seen": 2262499328
},
{
"epoch": 0.69,
"learning_rate": 0.00015886659979939818,
"loss": 2.4145,
"theoretical_loss": 3.3957149370752244,
"tokens_seen": 2262564864
},
{
"epoch": 0.69,
"objective/train/docs_used": 1281765,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.591778516769409,
"objective/train/theoretical_loss": 3.39570693384547,
"objective/train/tokens_used": 2283090400,
"theoretical_loss": 3.39570693384547,
"tokens_seen": 2262630400
},
{
"epoch": 0.69,
"learning_rate": 0.00015885656970912737,
"loss": 2.6197,
"theoretical_loss": 3.39570693384547,
"tokens_seen": 2262630400
},
{
"epoch": 0.69,
"learning_rate": 0.00015884653961885657,
"loss": 2.75,
"theoretical_loss": 3.3956989309124257,
"tokens_seen": 2262695936
},
{
"epoch": 0.69,
"learning_rate": 0.00015883650952858575,
"loss": 2.4819,
"theoretical_loss": 3.3956909282760734,
"tokens_seen": 2262761472
},
{
"epoch": 0.69,
"learning_rate": 0.00015882647943831494,
"loss": 2.5319,
"theoretical_loss": 3.3956829259363928,
"tokens_seen": 2262827008
},
{
"epoch": 0.69,
"learning_rate": 0.00015881644934804412,
"loss": 2.4677,
"theoretical_loss": 3.395674923893364,
"tokens_seen": 2262892544
},
{
"epoch": 0.69,
"learning_rate": 0.0001588064192577733,
"loss": 2.3959,
"theoretical_loss": 3.395666922146968,
"tokens_seen": 2262958080
},
{
"epoch": 0.69,
"learning_rate": 0.0001587963891675025,
"loss": 2.5616,
"theoretical_loss": 3.3956589206971852,
"tokens_seen": 2263023616
},
{
"epoch": 0.69,
"learning_rate": 0.0001587863590772317,
"loss": 2.7634,
"theoretical_loss": 3.3956509195439954,
"tokens_seen": 2263089152
},
{
"epoch": 0.69,
"learning_rate": 0.0001587763289869609,
"loss": 2.537,
"theoretical_loss": 3.3956429186873796,
"tokens_seen": 2263154688
},
{
"epoch": 0.69,
"learning_rate": 0.00015876629889669008,
"loss": 2.6879,
"theoretical_loss": 3.395634918127318,
"tokens_seen": 2263220224
},
{
"epoch": 0.69,
"learning_rate": 0.00015875626880641929,
"loss": 2.6877,
"theoretical_loss": 3.3956269178637912,
"tokens_seen": 2263285760
},
{
"epoch": 0.69,
"learning_rate": 0.00015874623871614847,
"loss": 2.6602,
"theoretical_loss": 3.395618917896779,
"tokens_seen": 2263351296
},
{
"epoch": 0.69,
"learning_rate": 0.00015873620862587765,
"loss": 2.9071,
"theoretical_loss": 3.3956109182262626,
"tokens_seen": 2263416832
},
{
"epoch": 0.69,
"learning_rate": 0.00015872617853560683,
"loss": 2.6878,
"theoretical_loss": 3.395602918852222,
"tokens_seen": 2263482368
},
{
"epoch": 0.69,
"learning_rate": 0.000158716148445336,
"loss": 2.5971,
"theoretical_loss": 3.395594919774638,
"tokens_seen": 2263547904
},
{
"epoch": 0.69,
"learning_rate": 0.00015870611835506522,
"loss": 2.297,
"theoretical_loss": 3.39558692099349,
"tokens_seen": 2263613440
},
{
"epoch": 0.69,
"learning_rate": 0.0001586960882647944,
"loss": 2.2881,
"theoretical_loss": 3.39557892250876,
"tokens_seen": 2263678976
},
{
"epoch": 0.69,
"learning_rate": 0.00015868605817452358,
"loss": 2.5777,
"theoretical_loss": 3.3955709243204275,
"tokens_seen": 2263744512
},
{
"epoch": 0.69,
"learning_rate": 0.00015867602808425276,
"loss": 2.6093,
"theoretical_loss": 3.3955629264284726,
"tokens_seen": 2263810048
},
{
"epoch": 0.69,
"learning_rate": 0.00015866599799398197,
"loss": 2.6172,
"theoretical_loss": 3.3955549288328766,
"tokens_seen": 2263875584
},
{
"epoch": 0.69,
"learning_rate": 0.00015865596790371115,
"loss": 2.7729,
"theoretical_loss": 3.3955469315336195,
"tokens_seen": 2263941120
},
{
"epoch": 0.69,
"learning_rate": 0.00015864593781344033,
"loss": 2.627,
"theoretical_loss": 3.395538934530682,
"tokens_seen": 2264006656
},
{
"epoch": 0.69,
"learning_rate": 0.0001586359077231695,
"loss": 2.716,
"theoretical_loss": 3.395530937824044,
"tokens_seen": 2264072192
},
{
"epoch": 0.69,
"learning_rate": 0.0001586258776328987,
"loss": 2.5613,
"theoretical_loss": 3.3955229414136863,
"tokens_seen": 2264137728
},
{
"epoch": 0.69,
"learning_rate": 0.0001586158475426279,
"loss": 2.6261,
"theoretical_loss": 3.3955149452995896,
"tokens_seen": 2264203264
},
{
"epoch": 0.69,
"objective/train/docs_used": 1282394,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9338395595550537,
"objective/train/theoretical_loss": 3.395506949481734,
"objective/train/tokens_used": 2284728800,
"theoretical_loss": 3.395506949481734,
"tokens_seen": 2264268800
},
{
"epoch": 0.69,
"learning_rate": 0.00015860581745235708,
"loss": 2.8281,
"theoretical_loss": 3.395506949481734,
"tokens_seen": 2264268800
},
{
"epoch": 0.69,
"learning_rate": 0.00015859578736208626,
"loss": 2.6865,
"theoretical_loss": 3.3954989539601,
"tokens_seen": 2264334336
},
{
"epoch": 0.69,
"learning_rate": 0.00015858575727181545,
"loss": 2.5175,
"theoretical_loss": 3.395490958734668,
"tokens_seen": 2264399872
},
{
"epoch": 0.69,
"learning_rate": 0.00015857572718154465,
"loss": 2.4627,
"theoretical_loss": 3.3954829638054185,
"tokens_seen": 2264465408
},
{
"epoch": 0.69,
"learning_rate": 0.00015856569709127383,
"loss": 2.7441,
"theoretical_loss": 3.395474969172332,
"tokens_seen": 2264530944
},
{
"epoch": 0.69,
"learning_rate": 0.00015855566700100302,
"loss": 2.7389,
"theoretical_loss": 3.395466974835389,
"tokens_seen": 2264596480
},
{
"epoch": 0.69,
"learning_rate": 0.0001585456369107322,
"loss": 2.5086,
"theoretical_loss": 3.39545898079457,
"tokens_seen": 2264662016
},
{
"epoch": 0.69,
"learning_rate": 0.0001585356068204614,
"loss": 2.6234,
"theoretical_loss": 3.3954509870498555,
"tokens_seen": 2264727552
},
{
"epoch": 0.69,
"learning_rate": 0.0001585255767301906,
"loss": 2.7187,
"theoretical_loss": 3.3954429936012254,
"tokens_seen": 2264793088
},
{
"epoch": 0.69,
"learning_rate": 0.00015851554663991977,
"loss": 2.6342,
"theoretical_loss": 3.3954350004486606,
"tokens_seen": 2264858624
},
{
"epoch": 0.69,
"learning_rate": 0.00015850551654964895,
"loss": 2.5737,
"theoretical_loss": 3.3954270075921418,
"tokens_seen": 2264924160
},
{
"epoch": 0.69,
"learning_rate": 0.00015849548645937813,
"loss": 2.5077,
"theoretical_loss": 3.3954190150316492,
"tokens_seen": 2264989696
},
{
"epoch": 0.69,
"learning_rate": 0.00015848545636910734,
"loss": 2.6902,
"theoretical_loss": 3.3954110227671634,
"tokens_seen": 2265055232
},
{
"epoch": 0.69,
"learning_rate": 0.00015847542627883652,
"loss": 2.8635,
"theoretical_loss": 3.395403030798665,
"tokens_seen": 2265120768
},
{
"epoch": 0.69,
"learning_rate": 0.0001584653961885657,
"loss": 2.6332,
"theoretical_loss": 3.395395039126134,
"tokens_seen": 2265186304
},
{
"epoch": 0.69,
"learning_rate": 0.00015845536609829488,
"loss": 2.8078,
"theoretical_loss": 3.3953870477495514,
"tokens_seen": 2265251840
},
{
"epoch": 0.69,
"learning_rate": 0.0001584453360080241,
"loss": 2.7769,
"theoretical_loss": 3.395379056668897,
"tokens_seen": 2265317376
},
{
"epoch": 0.69,
"learning_rate": 0.00015843530591775327,
"loss": 2.7118,
"theoretical_loss": 3.395371065884152,
"tokens_seen": 2265382912
},
{
"epoch": 0.69,
"learning_rate": 0.00015842527582748245,
"loss": 2.721,
"theoretical_loss": 3.395363075395296,
"tokens_seen": 2265448448
},
{
"epoch": 0.69,
"learning_rate": 0.00015841524573721163,
"loss": 2.5105,
"theoretical_loss": 3.395355085202311,
"tokens_seen": 2265513984
},
{
"epoch": 0.69,
"learning_rate": 0.00015840521564694081,
"loss": 2.6121,
"theoretical_loss": 3.395347095305176,
"tokens_seen": 2265579520
},
{
"epoch": 0.69,
"learning_rate": 0.00015839518555667002,
"loss": 2.6289,
"theoretical_loss": 3.3953391057038718,
"tokens_seen": 2265645056
},
{
"epoch": 0.69,
"learning_rate": 0.0001583851554663992,
"loss": 2.4586,
"theoretical_loss": 3.3953311163983795,
"tokens_seen": 2265710592
},
{
"epoch": 0.69,
"learning_rate": 0.00015837512537612838,
"loss": 2.6472,
"theoretical_loss": 3.395323127388679,
"tokens_seen": 2265776128
},
{
"epoch": 0.69,
"learning_rate": 0.00015836509528585757,
"loss": 2.6881,
"theoretical_loss": 3.3953151386747513,
"tokens_seen": 2265841664
},
{
"epoch": 0.69,
"objective/train/docs_used": 1283537,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2421860694885254,
"objective/train/theoretical_loss": 3.3953071502565764,
"objective/train/tokens_used": 2286367200,
"theoretical_loss": 3.3953071502565764,
"tokens_seen": 2265907200
},
{
"epoch": 0.69,
"learning_rate": 0.00015835506519558677,
"loss": 2.5327,
"theoretical_loss": 3.3953071502565764,
"tokens_seen": 2265907200
},
{
"epoch": 0.69,
"learning_rate": 0.00015834503510531596,
"loss": 2.5141,
"theoretical_loss": 3.395299162134135,
"tokens_seen": 2265972736
},
{
"epoch": 0.69,
"learning_rate": 0.00015833500501504514,
"loss": 2.6883,
"theoretical_loss": 3.395291174307408,
"tokens_seen": 2266038272
},
{
"epoch": 0.69,
"learning_rate": 0.00015832497492477432,
"loss": 2.5713,
"theoretical_loss": 3.395283186776375,
"tokens_seen": 2266103808
},
{
"epoch": 0.69,
"learning_rate": 0.0001583149448345035,
"loss": 2.3721,
"theoretical_loss": 3.3952751995410173,
"tokens_seen": 2266169344
},
{
"epoch": 0.69,
"learning_rate": 0.0001583049147442327,
"loss": 2.4999,
"theoretical_loss": 3.395267212601315,
"tokens_seen": 2266234880
},
{
"epoch": 0.69,
"learning_rate": 0.0001582948846539619,
"loss": 2.6162,
"theoretical_loss": 3.3952592259572487,
"tokens_seen": 2266300416
},
{
"epoch": 0.69,
"learning_rate": 0.00015828485456369107,
"loss": 2.5439,
"theoretical_loss": 3.395251239608799,
"tokens_seen": 2266365952
},
{
"epoch": 0.69,
"learning_rate": 0.00015827482447342025,
"loss": 2.5138,
"theoretical_loss": 3.3952432535559463,
"tokens_seen": 2266431488
},
{
"epoch": 0.69,
"learning_rate": 0.00015826479438314946,
"loss": 2.5727,
"theoretical_loss": 3.3952352677986712,
"tokens_seen": 2266497024
},
{
"epoch": 0.69,
"learning_rate": 0.00015825476429287864,
"loss": 2.6602,
"theoretical_loss": 3.395227282336954,
"tokens_seen": 2266562560
},
{
"epoch": 0.69,
"learning_rate": 0.00015824473420260782,
"loss": 2.6114,
"theoretical_loss": 3.3952192971707755,
"tokens_seen": 2266628096
},
{
"epoch": 0.69,
"learning_rate": 0.000158234704112337,
"loss": 2.659,
"theoretical_loss": 3.395211312300116,
"tokens_seen": 2266693632
},
{
"epoch": 0.69,
"learning_rate": 0.00015822467402206618,
"loss": 2.462,
"theoretical_loss": 3.3952033277249565,
"tokens_seen": 2266759168
},
{
"epoch": 0.69,
"learning_rate": 0.0001582146439317954,
"loss": 2.7472,
"theoretical_loss": 3.395195343445277,
"tokens_seen": 2266824704
},
{
"epoch": 0.69,
"learning_rate": 0.00015820461384152457,
"loss": 2.5937,
"theoretical_loss": 3.3951873594610578,
"tokens_seen": 2266890240
},
{
"epoch": 0.69,
"learning_rate": 0.00015819458375125375,
"loss": 2.522,
"theoretical_loss": 3.3951793757722797,
"tokens_seen": 2266955776
},
{
"epoch": 0.69,
"learning_rate": 0.00015818455366098293,
"loss": 2.6971,
"theoretical_loss": 3.395171392378924,
"tokens_seen": 2267021312
},
{
"epoch": 0.69,
"learning_rate": 0.00015817452357071214,
"loss": 2.7352,
"theoretical_loss": 3.39516340928097,
"tokens_seen": 2267086848
},
{
"epoch": 0.69,
"learning_rate": 0.00015816449348044132,
"loss": 2.6578,
"theoretical_loss": 3.395155426478399,
"tokens_seen": 2267152384
},
{
"epoch": 0.69,
"learning_rate": 0.0001581544633901705,
"loss": 2.5685,
"theoretical_loss": 3.395147443971191,
"tokens_seen": 2267217920
},
{
"epoch": 0.69,
"learning_rate": 0.00015814443329989969,
"loss": 2.5352,
"theoretical_loss": 3.395139461759327,
"tokens_seen": 2267283456
},
{
"epoch": 0.69,
"learning_rate": 0.00015813440320962887,
"loss": 2.5913,
"theoretical_loss": 3.395131479842788,
"tokens_seen": 2267348992
},
{
"epoch": 0.69,
"learning_rate": 0.00015812437311935808,
"loss": 2.5027,
"theoretical_loss": 3.3951234982215532,
"tokens_seen": 2267414528
},
{
"epoch": 0.69,
"learning_rate": 0.00015811434302908726,
"loss": 2.4869,
"theoretical_loss": 3.395115516895604,
"tokens_seen": 2267480064
},
{
"epoch": 0.69,
"objective/train/docs_used": 1284224,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.692258834838867,
"objective/train/theoretical_loss": 3.3951075358649208,
"objective/train/tokens_used": 2288005600,
"theoretical_loss": 3.3951075358649208,
"tokens_seen": 2267545600
},
{
"epoch": 0.69,
"learning_rate": 0.00015810431293881644,
"loss": 2.6894,
"theoretical_loss": 3.3951075358649208,
"tokens_seen": 2267545600
},
{
"epoch": 0.69,
"learning_rate": 0.00015809428284854562,
"loss": 2.4754,
"theoretical_loss": 3.3950995551294842,
"tokens_seen": 2267611136
},
{
"epoch": 0.69,
"learning_rate": 0.00015808425275827483,
"loss": 2.389,
"theoretical_loss": 3.3950915746892747,
"tokens_seen": 2267676672
},
{
"epoch": 0.69,
"learning_rate": 0.000158074222668004,
"loss": 2.5474,
"theoretical_loss": 3.395083594544273,
"tokens_seen": 2267742208
},
{
"epoch": 0.69,
"learning_rate": 0.0001580641925777332,
"loss": 2.6996,
"theoretical_loss": 3.395075614694459,
"tokens_seen": 2267807744
},
{
"epoch": 0.69,
"learning_rate": 0.00015805416248746237,
"loss": 2.4913,
"theoretical_loss": 3.395067635139814,
"tokens_seen": 2267873280
},
{
"epoch": 0.69,
"learning_rate": 0.00015804413239719155,
"loss": 2.567,
"theoretical_loss": 3.3950596558803183,
"tokens_seen": 2267938816
},
{
"epoch": 0.69,
"learning_rate": 0.00015803410230692076,
"loss": 2.6243,
"theoretical_loss": 3.3950516769159527,
"tokens_seen": 2268004352
},
{
"epoch": 0.69,
"learning_rate": 0.00015802407221664997,
"loss": 2.4795,
"theoretical_loss": 3.395043698246697,
"tokens_seen": 2268069888
},
{
"epoch": 0.69,
"learning_rate": 0.00015801404212637915,
"loss": 2.5717,
"theoretical_loss": 3.3950357198725327,
"tokens_seen": 2268135424
},
{
"epoch": 0.69,
"learning_rate": 0.00015800401203610833,
"loss": 2.474,
"theoretical_loss": 3.39502774179344,
"tokens_seen": 2268200960
},
{
"epoch": 0.69,
"learning_rate": 0.00015799398194583754,
"loss": 2.6502,
"theoretical_loss": 3.395019764009399,
"tokens_seen": 2268266496
},
{
"epoch": 0.69,
"learning_rate": 0.00015798395185556672,
"loss": 2.4946,
"theoretical_loss": 3.3950117865203904,
"tokens_seen": 2268332032
},
{
"epoch": 0.69,
"learning_rate": 0.0001579739217652959,
"loss": 2.4954,
"theoretical_loss": 3.3950038093263952,
"tokens_seen": 2268397568
},
{
"epoch": 0.69,
"learning_rate": 0.00015796389167502508,
"loss": 2.6507,
"theoretical_loss": 3.3949958324273943,
"tokens_seen": 2268463104
},
{
"epoch": 0.69,
"learning_rate": 0.0001579538615847543,
"loss": 2.5397,
"theoretical_loss": 3.3949878558233673,
"tokens_seen": 2268528640
},
{
"epoch": 0.69,
"learning_rate": 0.00015794383149448347,
"loss": 2.5428,
"theoretical_loss": 3.3949798795142954,
"tokens_seen": 2268594176
},
{
"epoch": 0.69,
"learning_rate": 0.00015793380140421265,
"loss": 2.6445,
"theoretical_loss": 3.3949719035001586,
"tokens_seen": 2268659712
},
{
"epoch": 0.69,
"learning_rate": 0.00015792377131394183,
"loss": 2.573,
"theoretical_loss": 3.394963927780938,
"tokens_seen": 2268725248
},
{
"epoch": 0.69,
"learning_rate": 0.00015791374122367101,
"loss": 2.5274,
"theoretical_loss": 3.3949559523566144,
"tokens_seen": 2268790784
},
{
"epoch": 0.69,
"learning_rate": 0.00015790371113340022,
"loss": 2.4344,
"theoretical_loss": 3.394947977227168,
"tokens_seen": 2268856320
},
{
"epoch": 0.69,
"learning_rate": 0.0001578936810431294,
"loss": 2.5605,
"theoretical_loss": 3.3949400023925786,
"tokens_seen": 2268921856
},
{
"epoch": 0.69,
"learning_rate": 0.00015788365095285858,
"loss": 2.7779,
"theoretical_loss": 3.3949320278528283,
"tokens_seen": 2268987392
},
{
"epoch": 0.69,
"learning_rate": 0.00015787362086258777,
"loss": 2.5461,
"theoretical_loss": 3.394924053607897,
"tokens_seen": 2269052928
},
{
"epoch": 0.69,
"learning_rate": 0.00015786359077231697,
"loss": 2.6766,
"theoretical_loss": 3.394916079657765,
"tokens_seen": 2269118464
},
{
"epoch": 0.69,
"objective/train/docs_used": 1285503,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.632453203201294,
"objective/train/theoretical_loss": 3.394908106002413,
"objective/train/tokens_used": 2289644000,
"theoretical_loss": 3.394908106002413,
"tokens_seen": 2269184000
},
{
"epoch": 0.69,
"learning_rate": 0.00015785356068204616,
"loss": 2.6916,
"theoretical_loss": 3.394908106002413,
"tokens_seen": 2269184000
},
{
"epoch": 0.69,
"learning_rate": 0.00015784353059177534,
"loss": 2.5499,
"theoretical_loss": 3.394900132641822,
"tokens_seen": 2269249536
},
{
"epoch": 0.69,
"learning_rate": 0.00015783350050150452,
"loss": 2.6319,
"theoretical_loss": 3.3948921595759725,
"tokens_seen": 2269315072
},
{
"epoch": 0.69,
"learning_rate": 0.0001578234704112337,
"loss": 2.493,
"theoretical_loss": 3.3948841868048447,
"tokens_seen": 2269380608
},
{
"epoch": 0.69,
"learning_rate": 0.0001578134403209629,
"loss": 2.5298,
"theoretical_loss": 3.3948762143284195,
"tokens_seen": 2269446144
},
{
"epoch": 0.69,
"learning_rate": 0.0001578034102306921,
"loss": 2.775,
"theoretical_loss": 3.394868242146677,
"tokens_seen": 2269511680
},
{
"epoch": 0.69,
"learning_rate": 0.00015779338014042127,
"loss": 2.681,
"theoretical_loss": 3.3948602702595987,
"tokens_seen": 2269577216
},
{
"epoch": 0.69,
"learning_rate": 0.00015778335005015045,
"loss": 2.7046,
"theoretical_loss": 3.3948522986671645,
"tokens_seen": 2269642752
},
{
"epoch": 0.69,
"learning_rate": 0.00015777331995987966,
"loss": 2.4895,
"theoretical_loss": 3.394844327369355,
"tokens_seen": 2269708288
},
{
"epoch": 0.69,
"learning_rate": 0.00015776328986960884,
"loss": 2.545,
"theoretical_loss": 3.3948363563661514,
"tokens_seen": 2269773824
},
{
"epoch": 0.69,
"learning_rate": 0.00015775325977933802,
"loss": 2.5634,
"theoretical_loss": 3.394828385657534,
"tokens_seen": 2269839360
},
{
"epoch": 0.69,
"learning_rate": 0.0001577432296890672,
"loss": 2.4877,
"theoretical_loss": 3.394820415243483,
"tokens_seen": 2269904896
},
{
"epoch": 0.69,
"learning_rate": 0.00015773319959879638,
"loss": 2.5507,
"theoretical_loss": 3.394812445123979,
"tokens_seen": 2269970432
},
{
"epoch": 0.69,
"learning_rate": 0.0001577231695085256,
"loss": 2.6278,
"theoretical_loss": 3.3948044752990034,
"tokens_seen": 2270035968
},
{
"epoch": 0.69,
"learning_rate": 0.00015771313941825477,
"loss": 2.8882,
"theoretical_loss": 3.3947965057685363,
"tokens_seen": 2270101504
},
{
"epoch": 0.69,
"learning_rate": 0.00015770310932798395,
"loss": 2.441,
"theoretical_loss": 3.3947885365325585,
"tokens_seen": 2270167040
},
{
"epoch": 0.69,
"learning_rate": 0.00015769307923771313,
"loss": 2.7342,
"theoretical_loss": 3.3947805675910505,
"tokens_seen": 2270232576
},
{
"epoch": 0.69,
"learning_rate": 0.00015768304914744234,
"loss": 2.3795,
"theoretical_loss": 3.3947725989439927,
"tokens_seen": 2270298112
},
{
"epoch": 0.69,
"learning_rate": 0.00015767301905717152,
"loss": 2.5452,
"theoretical_loss": 3.3947646305913657,
"tokens_seen": 2270363648
},
{
"epoch": 0.69,
"learning_rate": 0.0001576629889669007,
"loss": 2.6472,
"theoretical_loss": 3.394756662533151,
"tokens_seen": 2270429184
},
{
"epoch": 0.69,
"learning_rate": 0.00015765295887662989,
"loss": 2.4812,
"theoretical_loss": 3.394748694769328,
"tokens_seen": 2270494720
},
{
"epoch": 0.69,
"learning_rate": 0.00015764292878635907,
"loss": 2.6827,
"theoretical_loss": 3.394740727299878,
"tokens_seen": 2270560256
},
{
"epoch": 0.69,
"learning_rate": 0.00015763289869608828,
"loss": 2.6315,
"theoretical_loss": 3.394732760124782,
"tokens_seen": 2270625792
},
{
"epoch": 0.69,
"learning_rate": 0.00015762286860581746,
"loss": 2.5902,
"theoretical_loss": 3.39472479324402,
"tokens_seen": 2270691328
},
{
"epoch": 0.69,
"learning_rate": 0.00015761283851554664,
"loss": 2.5197,
"theoretical_loss": 3.3947168266575725,
"tokens_seen": 2270756864
},
{
"epoch": 0.69,
"objective/train/docs_used": 1286076,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.946425437927246,
"objective/train/theoretical_loss": 3.3947088603654203,
"objective/train/tokens_used": 2291282400,
"theoretical_loss": 3.3947088603654203,
"tokens_seen": 2270822400
},
{
"epoch": 0.69,
"learning_rate": 0.00015760280842527582,
"loss": 2.7201,
"theoretical_loss": 3.3947088603654203,
"tokens_seen": 2270822400
},
{
"epoch": 0.69,
"learning_rate": 0.00015759277833500503,
"loss": 2.4436,
"theoretical_loss": 3.3947008943675447,
"tokens_seen": 2270887936
},
{
"epoch": 0.69,
"learning_rate": 0.0001575827482447342,
"loss": 2.7654,
"theoretical_loss": 3.3946929286639254,
"tokens_seen": 2270953472
},
{
"epoch": 0.69,
"learning_rate": 0.0001575727181544634,
"loss": 2.5878,
"theoretical_loss": 3.3946849632545435,
"tokens_seen": 2271019008
},
{
"epoch": 0.69,
"learning_rate": 0.00015756268806419257,
"loss": 2.6162,
"theoretical_loss": 3.3946769981393796,
"tokens_seen": 2271084544
},
{
"epoch": 0.69,
"learning_rate": 0.00015755265797392175,
"loss": 2.5105,
"theoretical_loss": 3.394669033318414,
"tokens_seen": 2271150080
},
{
"epoch": 0.69,
"learning_rate": 0.00015754262788365096,
"loss": 2.5622,
"theoretical_loss": 3.3946610687916285,
"tokens_seen": 2271215616
},
{
"epoch": 0.69,
"learning_rate": 0.00015753259779338014,
"loss": 2.5991,
"theoretical_loss": 3.3946531045590023,
"tokens_seen": 2271281152
},
{
"epoch": 0.69,
"learning_rate": 0.00015752256770310932,
"loss": 2.7496,
"theoretical_loss": 3.3946451406205167,
"tokens_seen": 2271346688
},
{
"epoch": 0.69,
"learning_rate": 0.0001575125376128385,
"loss": 2.6368,
"theoretical_loss": 3.3946371769761527,
"tokens_seen": 2271412224
},
{
"epoch": 0.69,
"learning_rate": 0.0001575025075225677,
"loss": 2.7649,
"theoretical_loss": 3.3946292136258904,
"tokens_seen": 2271477760
},
{
"epoch": 0.69,
"learning_rate": 0.0001574924774322969,
"loss": 2.732,
"theoretical_loss": 3.3946212505697106,
"tokens_seen": 2271543296
},
{
"epoch": 0.69,
"learning_rate": 0.00015748244734202607,
"loss": 2.4062,
"theoretical_loss": 3.3946132878075934,
"tokens_seen": 2271608832
},
{
"epoch": 0.69,
"learning_rate": 0.00015747241725175525,
"loss": 2.4719,
"theoretical_loss": 3.394605325339521,
"tokens_seen": 2271674368
},
{
"epoch": 0.69,
"learning_rate": 0.00015746238716148446,
"loss": 2.6699,
"theoretical_loss": 3.3945973631654724,
"tokens_seen": 2271739904
},
{
"epoch": 0.69,
"learning_rate": 0.00015745235707121364,
"loss": 2.6203,
"theoretical_loss": 3.394589401285429,
"tokens_seen": 2271805440
},
{
"epoch": 0.69,
"learning_rate": 0.00015744232698094283,
"loss": 2.4823,
"theoretical_loss": 3.3945814396993716,
"tokens_seen": 2271870976
},
{
"epoch": 0.69,
"learning_rate": 0.000157432296890672,
"loss": 2.5496,
"theoretical_loss": 3.394573478407281,
"tokens_seen": 2271936512
},
{
"epoch": 0.69,
"learning_rate": 0.0001574222668004012,
"loss": 2.4364,
"theoretical_loss": 3.394565517409137,
"tokens_seen": 2272002048
},
{
"epoch": 0.69,
"learning_rate": 0.0001574122367101304,
"loss": 2.7477,
"theoretical_loss": 3.394557556704921,
"tokens_seen": 2272067584
},
{
"epoch": 0.69,
"learning_rate": 0.00015740220661985958,
"loss": 2.5433,
"theoretical_loss": 3.3945495962946137,
"tokens_seen": 2272133120
},
{
"epoch": 0.69,
"learning_rate": 0.00015739217652958876,
"loss": 2.5796,
"theoretical_loss": 3.3945416361781953,
"tokens_seen": 2272198656
},
{
"epoch": 0.69,
"learning_rate": 0.00015738214643931794,
"loss": 2.5583,
"theoretical_loss": 3.394533676355647,
"tokens_seen": 2272264192
},
{
"epoch": 0.69,
"learning_rate": 0.00015737211634904715,
"loss": 2.6147,
"theoretical_loss": 3.3945257168269487,
"tokens_seen": 2272329728
},
{
"epoch": 0.69,
"learning_rate": 0.00015736208625877633,
"loss": 2.5406,
"theoretical_loss": 3.394517757592082,
"tokens_seen": 2272395264
},
{
"epoch": 0.69,
"objective/train/docs_used": 1287335,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6375226974487305,
"objective/train/theoretical_loss": 3.3945097986510273,
"objective/train/tokens_used": 2292920800,
"theoretical_loss": 3.3945097986510273,
"tokens_seen": 2272460800
},
{
"epoch": 0.69,
"learning_rate": 0.0001573520561685055,
"loss": 2.7132,
"theoretical_loss": 3.3945097986510273,
"tokens_seen": 2272460800
},
{
"epoch": 0.69,
"learning_rate": 0.0001573420260782347,
"loss": 2.5363,
"theoretical_loss": 3.3945018400037648,
"tokens_seen": 2272526336
},
{
"epoch": 0.69,
"learning_rate": 0.00015733199598796387,
"loss": 2.5252,
"theoretical_loss": 3.3944938816502757,
"tokens_seen": 2272591872
},
{
"epoch": 0.69,
"learning_rate": 0.00015732196589769308,
"loss": 2.7452,
"theoretical_loss": 3.3944859235905405,
"tokens_seen": 2272657408
},
{
"epoch": 0.69,
"learning_rate": 0.00015731193580742226,
"loss": 2.6197,
"theoretical_loss": 3.3944779658245396,
"tokens_seen": 2272722944
},
{
"epoch": 0.69,
"learning_rate": 0.00015730190571715144,
"loss": 2.8012,
"theoretical_loss": 3.394470008352254,
"tokens_seen": 2272788480
},
{
"epoch": 0.69,
"learning_rate": 0.00015729187562688062,
"loss": 2.5266,
"theoretical_loss": 3.394462051173665,
"tokens_seen": 2272854016
},
{
"epoch": 0.69,
"learning_rate": 0.00015728184553660986,
"loss": 2.7962,
"theoretical_loss": 3.3944540942887524,
"tokens_seen": 2272919552
},
{
"epoch": 0.69,
"learning_rate": 0.00015727181544633904,
"loss": 2.2605,
"theoretical_loss": 3.394446137697497,
"tokens_seen": 2272985088
},
{
"epoch": 0.69,
"learning_rate": 0.00015726178535606822,
"loss": 2.5739,
"theoretical_loss": 3.3944381813998796,
"tokens_seen": 2273050624
},
{
"epoch": 0.69,
"learning_rate": 0.0001572517552657974,
"loss": 2.4813,
"theoretical_loss": 3.3944302253958814,
"tokens_seen": 2273116160
},
{
"epoch": 0.69,
"learning_rate": 0.00015724172517552658,
"loss": 2.4272,
"theoretical_loss": 3.394422269685482,
"tokens_seen": 2273181696
},
{
"epoch": 0.69,
"learning_rate": 0.0001572316950852558,
"loss": 2.8419,
"theoretical_loss": 3.394414314268663,
"tokens_seen": 2273247232
},
{
"epoch": 0.69,
"learning_rate": 0.00015722166499498497,
"loss": 2.5075,
"theoretical_loss": 3.394406359145405,
"tokens_seen": 2273312768
},
{
"epoch": 0.69,
"learning_rate": 0.00015721163490471415,
"loss": 2.4722,
"theoretical_loss": 3.3943984043156887,
"tokens_seen": 2273378304
},
{
"epoch": 0.69,
"learning_rate": 0.00015720160481444333,
"loss": 2.6392,
"theoretical_loss": 3.3943904497794946,
"tokens_seen": 2273443840
},
{
"epoch": 0.69,
"learning_rate": 0.00015719157472417254,
"loss": 2.4461,
"theoretical_loss": 3.3943824955368034,
"tokens_seen": 2273509376
},
{
"epoch": 0.69,
"learning_rate": 0.00015718154463390172,
"loss": 2.6855,
"theoretical_loss": 3.3943745415875957,
"tokens_seen": 2273574912
},
{
"epoch": 0.69,
"learning_rate": 0.0001571715145436309,
"loss": 2.7162,
"theoretical_loss": 3.3943665879318528,
"tokens_seen": 2273640448
},
{
"epoch": 0.69,
"learning_rate": 0.00015716148445336009,
"loss": 2.6479,
"theoretical_loss": 3.3943586345695547,
"tokens_seen": 2273705984
},
{
"epoch": 0.69,
"learning_rate": 0.00015715145436308927,
"loss": 2.5508,
"theoretical_loss": 3.3943506815006823,
"tokens_seen": 2273771520
},
{
"epoch": 0.69,
"learning_rate": 0.00015714142427281848,
"loss": 2.5007,
"theoretical_loss": 3.394342728725217,
"tokens_seen": 2273837056
},
{
"epoch": 0.69,
"learning_rate": 0.00015713139418254766,
"loss": 2.5625,
"theoretical_loss": 3.3943347762431384,
"tokens_seen": 2273902592
},
{
"epoch": 0.69,
"learning_rate": 0.00015712136409227684,
"loss": 2.608,
"theoretical_loss": 3.394326824054428,
"tokens_seen": 2273968128
},
{
"epoch": 0.69,
"learning_rate": 0.00015711133400200602,
"loss": 2.8141,
"theoretical_loss": 3.3943188721590665,
"tokens_seen": 2274033664
},
{
"epoch": 0.69,
"objective/train/docs_used": 1288139,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.1960086822509766,
"objective/train/theoretical_loss": 3.394310920557034,
"objective/train/tokens_used": 2294559200,
"theoretical_loss": 3.394310920557034,
"tokens_seen": 2274099200
},
{
"epoch": 0.69,
"learning_rate": 0.00015710130391173523,
"loss": 2.7814,
"theoretical_loss": 3.394310920557034,
"tokens_seen": 2274099200
},
{
"epoch": 0.69,
"learning_rate": 0.0001570912738214644,
"loss": 2.7087,
"theoretical_loss": 3.3943029692483124,
"tokens_seen": 2274164736
},
{
"epoch": 0.69,
"learning_rate": 0.0001570812437311936,
"loss": 2.5996,
"theoretical_loss": 3.394295018232881,
"tokens_seen": 2274230272
},
{
"epoch": 0.69,
"learning_rate": 0.00015707121364092277,
"loss": 2.5461,
"theoretical_loss": 3.3942870675107217,
"tokens_seen": 2274295808
},
{
"epoch": 0.69,
"learning_rate": 0.00015706118355065195,
"loss": 2.5063,
"theoretical_loss": 3.3942791170818145,
"tokens_seen": 2274361344
},
{
"epoch": 0.69,
"learning_rate": 0.00015705115346038116,
"loss": 2.6275,
"theoretical_loss": 3.3942711669461407,
"tokens_seen": 2274426880
},
{
"epoch": 0.69,
"learning_rate": 0.00015704112337011034,
"loss": 2.8823,
"theoretical_loss": 3.3942632171036804,
"tokens_seen": 2274492416
},
{
"epoch": 0.69,
"learning_rate": 0.00015703109327983952,
"loss": 2.5241,
"theoretical_loss": 3.394255267554415,
"tokens_seen": 2274557952
},
{
"epoch": 0.69,
"learning_rate": 0.0001570210631895687,
"loss": 2.6238,
"theoretical_loss": 3.3942473182983246,
"tokens_seen": 2274623488
},
{
"epoch": 0.69,
"learning_rate": 0.0001570110330992979,
"loss": 2.8145,
"theoretical_loss": 3.3942393693353905,
"tokens_seen": 2274689024
},
{
"epoch": 0.69,
"learning_rate": 0.0001570010030090271,
"loss": 2.4712,
"theoretical_loss": 3.3942314206655926,
"tokens_seen": 2274754560
},
{
"epoch": 0.69,
"learning_rate": 0.00015699097291875627,
"loss": 2.5946,
"theoretical_loss": 3.3942234722889126,
"tokens_seen": 2274820096
},
{
"epoch": 0.69,
"learning_rate": 0.00015698094282848545,
"loss": 2.6076,
"theoretical_loss": 3.394215524205331,
"tokens_seen": 2274885632
},
{
"epoch": 0.69,
"learning_rate": 0.00015697091273821466,
"loss": 2.7,
"theoretical_loss": 3.3942075764148285,
"tokens_seen": 2274951168
},
{
"epoch": 0.69,
"learning_rate": 0.00015696088264794384,
"loss": 2.6002,
"theoretical_loss": 3.3941996289173857,
"tokens_seen": 2275016704
},
{
"epoch": 0.69,
"learning_rate": 0.00015695085255767303,
"loss": 2.5979,
"theoretical_loss": 3.3941916817129836,
"tokens_seen": 2275082240
},
{
"epoch": 0.69,
"learning_rate": 0.0001569408224674022,
"loss": 2.6127,
"theoretical_loss": 3.3941837348016026,
"tokens_seen": 2275147776
},
{
"epoch": 0.69,
"learning_rate": 0.0001569307923771314,
"loss": 2.7099,
"theoretical_loss": 3.3941757881832233,
"tokens_seen": 2275213312
},
{
"epoch": 0.69,
"learning_rate": 0.0001569207622868606,
"loss": 2.6766,
"theoretical_loss": 3.3941678418578274,
"tokens_seen": 2275278848
},
{
"epoch": 0.69,
"learning_rate": 0.00015691073219658978,
"loss": 2.2861,
"theoretical_loss": 3.394159895825395,
"tokens_seen": 2275344384
},
{
"epoch": 0.69,
"learning_rate": 0.00015690070210631896,
"loss": 2.8211,
"theoretical_loss": 3.3941519500859063,
"tokens_seen": 2275409920
},
{
"epoch": 0.69,
"learning_rate": 0.00015689067201604814,
"loss": 2.7956,
"theoretical_loss": 3.3941440046393434,
"tokens_seen": 2275475456
},
{
"epoch": 0.69,
"learning_rate": 0.00015688064192577735,
"loss": 2.5817,
"theoretical_loss": 3.394136059485686,
"tokens_seen": 2275540992
},
{
"epoch": 0.69,
"learning_rate": 0.00015687061183550653,
"loss": 2.511,
"theoretical_loss": 3.394128114624915,
"tokens_seen": 2275606528
},
{
"epoch": 0.69,
"learning_rate": 0.0001568605817452357,
"loss": 2.5843,
"theoretical_loss": 3.394120170057012,
"tokens_seen": 2275672064
},
{
"epoch": 0.69,
"objective/train/docs_used": 1288752,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.481455087661743,
"objective/train/theoretical_loss": 3.394112225781957,
"objective/train/tokens_used": 2296197600,
"theoretical_loss": 3.394112225781957,
"tokens_seen": 2275737600
},
{
"epoch": 0.69,
"learning_rate": 0.0001568505516549649,
"loss": 2.4554,
"theoretical_loss": 3.394112225781957,
"tokens_seen": 2275737600
},
{
"epoch": 0.69,
"learning_rate": 0.00015684052156469407,
"loss": 2.6084,
"theoretical_loss": 3.3941042817997307,
"tokens_seen": 2275803136
},
{
"epoch": 0.69,
"learning_rate": 0.00015683049147442328,
"loss": 2.6397,
"theoretical_loss": 3.394096338110314,
"tokens_seen": 2275868672
},
{
"epoch": 0.69,
"learning_rate": 0.00015682046138415246,
"loss": 2.6167,
"theoretical_loss": 3.394088394713688,
"tokens_seen": 2275934208
},
{
"epoch": 0.69,
"learning_rate": 0.00015681043129388164,
"loss": 2.4726,
"theoretical_loss": 3.3940804516098333,
"tokens_seen": 2275999744
},
{
"epoch": 0.69,
"learning_rate": 0.00015680040120361082,
"loss": 2.4559,
"theoretical_loss": 3.3940725087987307,
"tokens_seen": 2276065280
},
{
"epoch": 0.69,
"learning_rate": 0.00015679037111334003,
"loss": 2.9476,
"theoretical_loss": 3.3940645662803606,
"tokens_seen": 2276130816
},
{
"epoch": 0.69,
"learning_rate": 0.0001567803410230692,
"loss": 2.7321,
"theoretical_loss": 3.3940566240547043,
"tokens_seen": 2276196352
},
{
"epoch": 0.69,
"learning_rate": 0.0001567703109327984,
"loss": 2.7827,
"theoretical_loss": 3.3940486821217424,
"tokens_seen": 2276261888
},
{
"epoch": 0.69,
"learning_rate": 0.00015676028084252757,
"loss": 2.6977,
"theoretical_loss": 3.3940407404814557,
"tokens_seen": 2276327424
},
{
"epoch": 0.69,
"learning_rate": 0.00015675025075225676,
"loss": 2.6357,
"theoretical_loss": 3.394032799133825,
"tokens_seen": 2276392960
},
{
"epoch": 0.69,
"learning_rate": 0.00015674022066198596,
"loss": 2.6095,
"theoretical_loss": 3.394024858078831,
"tokens_seen": 2276458496
},
{
"epoch": 0.69,
"learning_rate": 0.00015673019057171515,
"loss": 2.7015,
"theoretical_loss": 3.3940169173164545,
"tokens_seen": 2276524032
},
{
"epoch": 0.69,
"learning_rate": 0.00015672016048144433,
"loss": 2.5926,
"theoretical_loss": 3.394008976846676,
"tokens_seen": 2276589568
},
{
"epoch": 0.69,
"learning_rate": 0.0001567101303911735,
"loss": 2.8372,
"theoretical_loss": 3.394001036669477,
"tokens_seen": 2276655104
},
{
"epoch": 0.69,
"learning_rate": 0.00015670010030090272,
"loss": 2.6841,
"theoretical_loss": 3.393993096784838,
"tokens_seen": 2276720640
},
{
"epoch": 0.69,
"learning_rate": 0.0001566900702106319,
"loss": 2.7488,
"theoretical_loss": 3.3939851571927395,
"tokens_seen": 2276786176
},
{
"epoch": 0.69,
"learning_rate": 0.00015668004012036108,
"loss": 2.728,
"theoretical_loss": 3.393977217893163,
"tokens_seen": 2276851712
},
{
"epoch": 0.69,
"learning_rate": 0.00015667001003009026,
"loss": 2.6139,
"theoretical_loss": 3.3939692788860887,
"tokens_seen": 2276917248
},
{
"epoch": 0.69,
"learning_rate": 0.00015665997993981944,
"loss": 2.3613,
"theoretical_loss": 3.393961340171497,
"tokens_seen": 2276982784
},
{
"epoch": 0.69,
"learning_rate": 0.00015664994984954865,
"loss": 2.4877,
"theoretical_loss": 3.39395340174937,
"tokens_seen": 2277048320
},
{
"epoch": 0.69,
"learning_rate": 0.00015663991975927783,
"loss": 2.4323,
"theoretical_loss": 3.393945463619687,
"tokens_seen": 2277113856
},
{
"epoch": 0.69,
"learning_rate": 0.000156629889669007,
"loss": 2.4568,
"theoretical_loss": 3.3939375257824302,
"tokens_seen": 2277179392
},
{
"epoch": 0.69,
"learning_rate": 0.0001566198595787362,
"loss": 2.5868,
"theoretical_loss": 3.3939295882375795,
"tokens_seen": 2277244928
},
{
"epoch": 0.69,
"learning_rate": 0.0001566098294884654,
"loss": 2.603,
"theoretical_loss": 3.3939216509851162,
"tokens_seen": 2277310464
},
{
"epoch": 0.69,
"objective/train/docs_used": 1289790,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6909701824188232,
"objective/train/theoretical_loss": 3.393913714025021,
"objective/train/tokens_used": 2297836000,
"theoretical_loss": 3.393913714025021,
"tokens_seen": 2277376000
},
{
"epoch": 0.69,
"learning_rate": 0.00015659979939819458,
"loss": 2.5397,
"theoretical_loss": 3.393913714025021,
"tokens_seen": 2277376000
},
{
"epoch": 0.69,
"learning_rate": 0.00015658976930792376,
"loss": 2.4798,
"theoretical_loss": 3.393905777357274,
"tokens_seen": 2277441536
},
{
"epoch": 0.69,
"learning_rate": 0.00015657973921765294,
"loss": 2.7517,
"theoretical_loss": 3.393897840981857,
"tokens_seen": 2277507072
},
{
"epoch": 0.69,
"learning_rate": 0.00015656970912738212,
"loss": 2.8014,
"theoretical_loss": 3.3938899048987508,
"tokens_seen": 2277572608
},
{
"epoch": 0.69,
"learning_rate": 0.00015655967903711133,
"loss": 2.658,
"theoretical_loss": 3.3938819691079356,
"tokens_seen": 2277638144
},
{
"epoch": 0.69,
"learning_rate": 0.00015654964894684051,
"loss": 2.475,
"theoretical_loss": 3.393874033609392,
"tokens_seen": 2277703680
},
{
"epoch": 0.69,
"learning_rate": 0.0001565396188565697,
"loss": 2.6062,
"theoretical_loss": 3.393866098403102,
"tokens_seen": 2277769216
},
{
"epoch": 0.69,
"learning_rate": 0.0001565295887662989,
"loss": 2.5402,
"theoretical_loss": 3.3938581634890457,
"tokens_seen": 2277834752
},
{
"epoch": 0.69,
"learning_rate": 0.0001565195586760281,
"loss": 2.6291,
"theoretical_loss": 3.393850228867204,
"tokens_seen": 2277900288
},
{
"epoch": 0.69,
"learning_rate": 0.0001565095285857573,
"loss": 2.8506,
"theoretical_loss": 3.3938422945375577,
"tokens_seen": 2277965824
},
{
"epoch": 0.69,
"learning_rate": 0.00015649949849548647,
"loss": 2.9952,
"theoretical_loss": 3.3938343605000876,
"tokens_seen": 2278031360
},
{
"epoch": 0.69,
"learning_rate": 0.00015648946840521565,
"loss": 2.7764,
"theoretical_loss": 3.3938264267547744,
"tokens_seen": 2278096896
},
{
"epoch": 0.69,
"learning_rate": 0.00015647943831494486,
"loss": 2.1268,
"theoretical_loss": 3.3938184933015996,
"tokens_seen": 2278162432
},
{
"epoch": 0.69,
"learning_rate": 0.00015646940822467404,
"loss": 2.7928,
"theoretical_loss": 3.3938105601405435,
"tokens_seen": 2278227968
},
{
"epoch": 0.69,
"learning_rate": 0.00015645937813440323,
"loss": 2.708,
"theoretical_loss": 3.3938026272715867,
"tokens_seen": 2278293504
},
{
"epoch": 0.69,
"learning_rate": 0.0001564493480441324,
"loss": 2.7036,
"theoretical_loss": 3.3937946946947104,
"tokens_seen": 2278359040
},
{
"epoch": 0.69,
"learning_rate": 0.0001564393179538616,
"loss": 2.6363,
"theoretical_loss": 3.393786762409896,
"tokens_seen": 2278424576
},
{
"epoch": 0.69,
"learning_rate": 0.0001564292878635908,
"loss": 2.7116,
"theoretical_loss": 3.3937788304171232,
"tokens_seen": 2278490112
},
{
"epoch": 0.69,
"learning_rate": 0.00015641925777331998,
"loss": 2.4648,
"theoretical_loss": 3.3937708987163733,
"tokens_seen": 2278555648
},
{
"epoch": 0.69,
"learning_rate": 0.00015640922768304916,
"loss": 2.5896,
"theoretical_loss": 3.393762967307628,
"tokens_seen": 2278621184
},
{
"epoch": 0.69,
"learning_rate": 0.00015639919759277834,
"loss": 2.7044,
"theoretical_loss": 3.393755036190867,
"tokens_seen": 2278686720
},
{
"epoch": 0.69,
"learning_rate": 0.00015638916750250755,
"loss": 2.6178,
"theoretical_loss": 3.393747105366071,
"tokens_seen": 2278752256
},
{
"epoch": 0.69,
"learning_rate": 0.00015637913741223673,
"loss": 2.4334,
"theoretical_loss": 3.393739174833222,
"tokens_seen": 2278817792
},
{
"epoch": 0.69,
"learning_rate": 0.0001563691073219659,
"loss": 2.6681,
"theoretical_loss": 3.3937312445923,
"tokens_seen": 2278883328
},
{
"epoch": 0.69,
"learning_rate": 0.0001563590772316951,
"loss": 2.4757,
"theoretical_loss": 3.393723314643286,
"tokens_seen": 2278948864
},
{
"epoch": 0.69,
"objective/train/docs_used": 1290345,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8778507709503174,
"objective/train/theoretical_loss": 3.393715384986162,
"objective/train/tokens_used": 2299474400,
"theoretical_loss": 3.393715384986162,
"tokens_seen": 2279014400
},
{
"epoch": 0.69,
"learning_rate": 0.00015634904714142427,
"loss": 2.7713,
"theoretical_loss": 3.393715384986162,
"tokens_seen": 2279014400
},
{
"epoch": 0.69,
"learning_rate": 0.00015633901705115348,
"loss": 2.4739,
"theoretical_loss": 3.393707455620907,
"tokens_seen": 2279079936
},
{
"epoch": 0.69,
"learning_rate": 0.00015632898696088266,
"loss": 2.8061,
"theoretical_loss": 3.3936995265475027,
"tokens_seen": 2279145472
},
{
"epoch": 0.69,
"learning_rate": 0.00015631895687061184,
"loss": 2.2674,
"theoretical_loss": 3.3936915977659305,
"tokens_seen": 2279211008
},
{
"epoch": 0.69,
"learning_rate": 0.00015630892678034102,
"loss": 2.507,
"theoretical_loss": 3.39368366927617,
"tokens_seen": 2279276544
},
{
"epoch": 0.69,
"learning_rate": 0.00015629889669007023,
"loss": 2.7721,
"theoretical_loss": 3.3936757410782032,
"tokens_seen": 2279342080
},
{
"epoch": 0.69,
"learning_rate": 0.0001562888665997994,
"loss": 2.6785,
"theoretical_loss": 3.393667813172011,
"tokens_seen": 2279407616
},
{
"epoch": 0.69,
"learning_rate": 0.0001562788365095286,
"loss": 2.6004,
"theoretical_loss": 3.3936598855575735,
"tokens_seen": 2279473152
},
{
"epoch": 0.69,
"learning_rate": 0.00015626880641925778,
"loss": 2.6415,
"theoretical_loss": 3.393651958234872,
"tokens_seen": 2279538688
},
{
"epoch": 0.69,
"learning_rate": 0.00015625877632898696,
"loss": 2.5045,
"theoretical_loss": 3.393644031203887,
"tokens_seen": 2279604224
},
{
"epoch": 0.69,
"learning_rate": 0.00015624874623871616,
"loss": 2.6668,
"theoretical_loss": 3.3936361044646,
"tokens_seen": 2279669760
},
{
"epoch": 0.69,
"learning_rate": 0.00015623871614844535,
"loss": 2.7463,
"theoretical_loss": 3.393628178016992,
"tokens_seen": 2279735296
},
{
"epoch": 0.69,
"learning_rate": 0.00015622868605817453,
"loss": 2.6636,
"theoretical_loss": 3.3936202518610425,
"tokens_seen": 2279800832
},
{
"epoch": 0.69,
"learning_rate": 0.0001562186559679037,
"loss": 2.6592,
"theoretical_loss": 3.393612325996734,
"tokens_seen": 2279866368
},
{
"epoch": 0.69,
"learning_rate": 0.00015620862587763292,
"loss": 2.9584,
"theoretical_loss": 3.3936044004240467,
"tokens_seen": 2279931904
},
{
"epoch": 0.69,
"learning_rate": 0.0001561985957873621,
"loss": 2.528,
"theoretical_loss": 3.3935964751429615,
"tokens_seen": 2279997440
},
{
"epoch": 0.69,
"learning_rate": 0.00015618856569709128,
"loss": 2.7534,
"theoretical_loss": 3.393588550153459,
"tokens_seen": 2280062976
},
{
"epoch": 0.69,
"learning_rate": 0.00015617853560682046,
"loss": 2.7644,
"theoretical_loss": 3.393580625455521,
"tokens_seen": 2280128512
},
{
"epoch": 0.69,
"learning_rate": 0.00015616850551654964,
"loss": 2.7496,
"theoretical_loss": 3.3935727010491274,
"tokens_seen": 2280194048
},
{
"epoch": 0.69,
"learning_rate": 0.00015615847542627885,
"loss": 2.5668,
"theoretical_loss": 3.3935647769342596,
"tokens_seen": 2280259584
},
{
"epoch": 0.69,
"learning_rate": 0.00015614844533600803,
"loss": 2.773,
"theoretical_loss": 3.3935568531108986,
"tokens_seen": 2280325120
},
{
"epoch": 0.69,
"learning_rate": 0.0001561384152457372,
"loss": 2.5729,
"theoretical_loss": 3.393548929579025,
"tokens_seen": 2280390656
},
{
"epoch": 0.69,
"learning_rate": 0.0001561283851554664,
"loss": 2.3199,
"theoretical_loss": 3.3935410063386193,
"tokens_seen": 2280456192
},
{
"epoch": 0.69,
"learning_rate": 0.0001561183550651956,
"loss": 2.3127,
"theoretical_loss": 3.3935330833896638,
"tokens_seen": 2280521728
},
{
"epoch": 0.69,
"learning_rate": 0.00015610832497492478,
"loss": 2.5751,
"theoretical_loss": 3.3935251607321377,
"tokens_seen": 2280587264
},
{
"epoch": 0.69,
"objective/train/docs_used": 1291619,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.737004518508911,
"objective/train/theoretical_loss": 3.393517238366023,
"objective/train/tokens_used": 2301112800,
"theoretical_loss": 3.393517238366023,
"tokens_seen": 2280652800
},
{
"epoch": 0.69,
"learning_rate": 0.00015609829488465396,
"loss": 2.636,
"theoretical_loss": 3.393517238366023,
"tokens_seen": 2280652800
},
{
"epoch": 0.69,
"learning_rate": 0.00015608826479438314,
"loss": 2.8125,
"theoretical_loss": 3.3935093162913,
"tokens_seen": 2280718336
},
{
"epoch": 0.69,
"learning_rate": 0.00015607823470411232,
"loss": 2.4539,
"theoretical_loss": 3.3935013945079504,
"tokens_seen": 2280783872
},
{
"epoch": 0.69,
"learning_rate": 0.00015606820461384153,
"loss": 2.7814,
"theoretical_loss": 3.3934934730159547,
"tokens_seen": 2280849408
},
{
"epoch": 0.69,
"learning_rate": 0.00015605817452357071,
"loss": 2.5231,
"theoretical_loss": 3.3934855518152935,
"tokens_seen": 2280914944
},
{
"epoch": 0.69,
"learning_rate": 0.0001560481444332999,
"loss": 2.8628,
"theoretical_loss": 3.393477630905948,
"tokens_seen": 2280980480
},
{
"epoch": 0.69,
"learning_rate": 0.00015603811434302908,
"loss": 2.6195,
"theoretical_loss": 3.393469710287899,
"tokens_seen": 2281046016
},
{
"epoch": 0.69,
"learning_rate": 0.00015602808425275828,
"loss": 2.5517,
"theoretical_loss": 3.3934617899611275,
"tokens_seen": 2281111552
},
{
"epoch": 0.69,
"learning_rate": 0.00015601805416248747,
"loss": 2.3465,
"theoretical_loss": 3.3934538699256143,
"tokens_seen": 2281177088
},
{
"epoch": 0.69,
"learning_rate": 0.00015600802407221665,
"loss": 2.5143,
"theoretical_loss": 3.393445950181341,
"tokens_seen": 2281242624
},
{
"epoch": 0.69,
"learning_rate": 0.00015599799398194583,
"loss": 2.6628,
"theoretical_loss": 3.393438030728287,
"tokens_seen": 2281308160
},
{
"epoch": 0.69,
"learning_rate": 0.000155987963891675,
"loss": 2.5985,
"theoretical_loss": 3.393430111566435,
"tokens_seen": 2281373696
},
{
"epoch": 0.69,
"learning_rate": 0.00015597793380140422,
"loss": 2.6833,
"theoretical_loss": 3.3934221926957653,
"tokens_seen": 2281439232
},
{
"epoch": 0.69,
"learning_rate": 0.0001559679037111334,
"loss": 2.7851,
"theoretical_loss": 3.393414274116258,
"tokens_seen": 2281504768
},
{
"epoch": 0.69,
"learning_rate": 0.00015595787362086258,
"loss": 2.628,
"theoretical_loss": 3.393406355827895,
"tokens_seen": 2281570304
},
{
"epoch": 0.69,
"learning_rate": 0.00015594784353059176,
"loss": 2.5959,
"theoretical_loss": 3.393398437830657,
"tokens_seen": 2281635840
},
{
"epoch": 0.69,
"learning_rate": 0.00015593781344032097,
"loss": 2.594,
"theoretical_loss": 3.3933905201245245,
"tokens_seen": 2281701376
},
{
"epoch": 0.69,
"learning_rate": 0.00015592778335005015,
"loss": 2.6387,
"theoretical_loss": 3.393382602709479,
"tokens_seen": 2281766912
},
{
"epoch": 0.69,
"learning_rate": 0.00015591775325977933,
"loss": 2.6528,
"theoretical_loss": 3.3933746855855014,
"tokens_seen": 2281832448
},
{
"epoch": 0.69,
"learning_rate": 0.0001559077231695085,
"loss": 2.5953,
"theoretical_loss": 3.393366768752572,
"tokens_seen": 2281897984
},
{
"epoch": 0.69,
"learning_rate": 0.0001558976930792377,
"loss": 2.5648,
"theoretical_loss": 3.393358852210673,
"tokens_seen": 2281963520
},
{
"epoch": 0.69,
"learning_rate": 0.0001558876629889669,
"loss": 2.6454,
"theoretical_loss": 3.3933509359597838,
"tokens_seen": 2282029056
},
{
"epoch": 0.69,
"learning_rate": 0.00015587763289869608,
"loss": 2.4084,
"theoretical_loss": 3.393343019999886,
"tokens_seen": 2282094592
},
{
"epoch": 0.69,
"learning_rate": 0.00015586760280842526,
"loss": 2.495,
"theoretical_loss": 3.393335104330961,
"tokens_seen": 2282160128
},
{
"epoch": 0.69,
"learning_rate": 0.00015585757271815444,
"loss": 2.5695,
"theoretical_loss": 3.3933271889529895,
"tokens_seen": 2282225664
},
{
"epoch": 0.69,
"objective/train/docs_used": 1292123,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.370457172393799,
"objective/train/theoretical_loss": 3.393319273865952,
"objective/train/tokens_used": 2302751200,
"theoretical_loss": 3.393319273865952,
"tokens_seen": 2282291200
},
{
"epoch": 0.69,
"learning_rate": 0.00015584754262788365,
"loss": 2.5735,
"theoretical_loss": 3.393319273865952,
"tokens_seen": 2282291200
},
{
"epoch": 0.69,
"learning_rate": 0.00015583751253761283,
"loss": 2.4734,
"theoretical_loss": 3.3933113590698305,
"tokens_seen": 2282356736
},
{
"epoch": 0.69,
"learning_rate": 0.00015582748244734202,
"loss": 2.6264,
"theoretical_loss": 3.3933034445646046,
"tokens_seen": 2282422272
},
{
"epoch": 0.69,
"learning_rate": 0.0001558174523570712,
"loss": 2.8239,
"theoretical_loss": 3.393295530350256,
"tokens_seen": 2282487808
},
{
"epoch": 0.69,
"learning_rate": 0.0001558074222668004,
"loss": 2.6471,
"theoretical_loss": 3.3932876164267656,
"tokens_seen": 2282553344
},
{
"epoch": 0.69,
"learning_rate": 0.00015579739217652959,
"loss": 2.5259,
"theoretical_loss": 3.3932797027941146,
"tokens_seen": 2282618880
},
{
"epoch": 0.69,
"learning_rate": 0.00015578736208625877,
"loss": 2.4319,
"theoretical_loss": 3.3932717894522835,
"tokens_seen": 2282684416
},
{
"epoch": 0.69,
"learning_rate": 0.00015577733199598798,
"loss": 2.6318,
"theoretical_loss": 3.393263876401253,
"tokens_seen": 2282749952
},
{
"epoch": 0.69,
"learning_rate": 0.00015576730190571716,
"loss": 2.6735,
"theoretical_loss": 3.393255963641005,
"tokens_seen": 2282815488
},
{
"epoch": 0.69,
"learning_rate": 0.00015575727181544636,
"loss": 2.7341,
"theoretical_loss": 3.39324805117152,
"tokens_seen": 2282881024
},
{
"epoch": 0.69,
"learning_rate": 0.00015574724172517555,
"loss": 2.775,
"theoretical_loss": 3.393240138992779,
"tokens_seen": 2282946560
},
{
"epoch": 0.69,
"learning_rate": 0.00015573721163490473,
"loss": 2.5637,
"theoretical_loss": 3.393232227104763,
"tokens_seen": 2283012096
},
{
"epoch": 0.69,
"learning_rate": 0.0001557271815446339,
"loss": 2.7303,
"theoretical_loss": 3.393224315507452,
"tokens_seen": 2283077632
},
{
"epoch": 0.69,
"learning_rate": 0.00015571715145436312,
"loss": 2.8171,
"theoretical_loss": 3.393216404200829,
"tokens_seen": 2283143168
},
{
"epoch": 0.69,
"learning_rate": 0.0001557071213640923,
"loss": 2.5338,
"theoretical_loss": 3.3932084931848734,
"tokens_seen": 2283208704
},
{
"epoch": 0.69,
"learning_rate": 0.00015569709127382148,
"loss": 2.3888,
"theoretical_loss": 3.3932005824595666,
"tokens_seen": 2283274240
},
{
"epoch": 0.69,
"learning_rate": 0.00015568706118355066,
"loss": 2.3746,
"theoretical_loss": 3.3931926720248895,
"tokens_seen": 2283339776
},
{
"epoch": 0.69,
"learning_rate": 0.00015567703109327984,
"loss": 2.4343,
"theoretical_loss": 3.3931847618808235,
"tokens_seen": 2283405312
},
{
"epoch": 0.69,
"learning_rate": 0.00015566700100300905,
"loss": 2.5923,
"theoretical_loss": 3.393176852027349,
"tokens_seen": 2283470848
},
{
"epoch": 0.69,
"learning_rate": 0.00015565697091273823,
"loss": 2.2655,
"theoretical_loss": 3.393168942464447,
"tokens_seen": 2283536384
},
{
"epoch": 0.69,
"learning_rate": 0.0001556469408224674,
"loss": 2.6038,
"theoretical_loss": 3.3931610331920994,
"tokens_seen": 2283601920
},
{
"epoch": 0.69,
"learning_rate": 0.0001556369107321966,
"loss": 2.5219,
"theoretical_loss": 3.3931531242102864,
"tokens_seen": 2283667456
},
{
"epoch": 0.69,
"learning_rate": 0.0001556268806419258,
"loss": 2.7135,
"theoretical_loss": 3.393145215518989,
"tokens_seen": 2283732992
},
{
"epoch": 0.69,
"learning_rate": 0.00015561685055165498,
"loss": 2.6057,
"theoretical_loss": 3.393137307118188,
"tokens_seen": 2283798528
},
{
"epoch": 0.69,
"learning_rate": 0.00015560682046138416,
"loss": 2.6579,
"theoretical_loss": 3.393129399007865,
"tokens_seen": 2283864064
},
{
"epoch": 0.69,
"objective/train/docs_used": 1293247,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4629571437835693,
"objective/train/theoretical_loss": 3.3931214911880008,
"objective/train/tokens_used": 2304389600,
"theoretical_loss": 3.3931214911880008,
"tokens_seen": 2283929600
},
{
"epoch": 0.69,
"learning_rate": 0.00015559679037111334,
"loss": 2.5301,
"theoretical_loss": 3.3931214911880008,
"tokens_seen": 2283929600
},
{
"epoch": 0.69,
"learning_rate": 0.00015558676028084252,
"loss": 2.7543,
"theoretical_loss": 3.393113583658576,
"tokens_seen": 2283995136
},
{
"epoch": 0.69,
"learning_rate": 0.00015557673019057173,
"loss": 2.5213,
"theoretical_loss": 3.393105676419572,
"tokens_seen": 2284060672
},
{
"epoch": 0.69,
"learning_rate": 0.00015556670010030091,
"loss": 2.5575,
"theoretical_loss": 3.3930977694709696,
"tokens_seen": 2284126208
},
{
"epoch": 0.69,
"learning_rate": 0.0001555566700100301,
"loss": 2.6428,
"theoretical_loss": 3.3930898628127504,
"tokens_seen": 2284191744
},
{
"epoch": 0.69,
"learning_rate": 0.00015554663991975928,
"loss": 2.5818,
"theoretical_loss": 3.3930819564448944,
"tokens_seen": 2284257280
},
{
"epoch": 0.69,
"learning_rate": 0.00015553660982948848,
"loss": 2.7222,
"theoretical_loss": 3.3930740503673835,
"tokens_seen": 2284322816
},
{
"epoch": 0.69,
"learning_rate": 0.00015552657973921767,
"loss": 2.6333,
"theoretical_loss": 3.3930661445801977,
"tokens_seen": 2284388352
},
{
"epoch": 0.69,
"learning_rate": 0.00015551654964894685,
"loss": 2.7854,
"theoretical_loss": 3.3930582390833193,
"tokens_seen": 2284453888
},
{
"epoch": 0.69,
"learning_rate": 0.00015550651955867603,
"loss": 2.7072,
"theoretical_loss": 3.3930503338767286,
"tokens_seen": 2284519424
},
{
"epoch": 0.69,
"learning_rate": 0.0001554964894684052,
"loss": 2.7495,
"theoretical_loss": 3.393042428960406,
"tokens_seen": 2284584960
},
{
"epoch": 0.69,
"learning_rate": 0.00015548645937813442,
"loss": 2.7621,
"theoretical_loss": 3.393034524334334,
"tokens_seen": 2284650496
},
{
"epoch": 0.69,
"learning_rate": 0.0001554764292878636,
"loss": 2.5738,
"theoretical_loss": 3.3930266199984924,
"tokens_seen": 2284716032
},
{
"epoch": 0.69,
"learning_rate": 0.00015546639919759278,
"loss": 2.8329,
"theoretical_loss": 3.3930187159528624,
"tokens_seen": 2284781568
},
{
"epoch": 0.69,
"learning_rate": 0.00015545636910732196,
"loss": 2.6537,
"theoretical_loss": 3.393010812197425,
"tokens_seen": 2284847104
},
{
"epoch": 0.69,
"learning_rate": 0.00015544633901705117,
"loss": 2.6113,
"theoretical_loss": 3.3930029087321625,
"tokens_seen": 2284912640
},
{
"epoch": 0.69,
"learning_rate": 0.00015543630892678035,
"loss": 2.8128,
"theoretical_loss": 3.3929950055570544,
"tokens_seen": 2284978176
},
{
"epoch": 0.69,
"learning_rate": 0.00015542627883650953,
"loss": 2.7073,
"theoretical_loss": 3.3929871026720817,
"tokens_seen": 2285043712
},
{
"epoch": 0.69,
"learning_rate": 0.0001554162487462387,
"loss": 2.5685,
"theoretical_loss": 3.3929792000772263,
"tokens_seen": 2285109248
},
{
"epoch": 0.69,
"learning_rate": 0.0001554062186559679,
"loss": 2.4657,
"theoretical_loss": 3.392971297772469,
"tokens_seen": 2285174784
},
{
"epoch": 0.69,
"learning_rate": 0.0001553961885656971,
"loss": 2.5955,
"theoretical_loss": 3.3929633957577905,
"tokens_seen": 2285240320
},
{
"epoch": 0.69,
"learning_rate": 0.00015538615847542628,
"loss": 2.7045,
"theoretical_loss": 3.3929554940331723,
"tokens_seen": 2285305856
},
{
"epoch": 0.69,
"learning_rate": 0.00015537612838515546,
"loss": 2.4741,
"theoretical_loss": 3.392947592598595,
"tokens_seen": 2285371392
},
{
"epoch": 0.69,
"learning_rate": 0.00015536609829488465,
"loss": 2.6968,
"theoretical_loss": 3.3929396914540395,
"tokens_seen": 2285436928
},
{
"epoch": 0.69,
"learning_rate": 0.00015535606820461385,
"loss": 2.7413,
"theoretical_loss": 3.3929317905994876,
"tokens_seen": 2285502464
},
{
"epoch": 0.69,
"objective/train/docs_used": 1293760,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8127799034118652,
"objective/train/theoretical_loss": 3.3929238900349197,
"objective/train/tokens_used": 2306028000,
"theoretical_loss": 3.3929238900349197,
"tokens_seen": 2285568000
},
{
"epoch": 0.69,
"learning_rate": 0.00015534603811434303,
"loss": 2.6929,
"theoretical_loss": 3.3929238900349197,
"tokens_seen": 2285568000
},
{
"epoch": 0.69,
"learning_rate": 0.00015533600802407222,
"loss": 2.7211,
"theoretical_loss": 3.3929159897603167,
"tokens_seen": 2285633536
},
{
"epoch": 0.69,
"learning_rate": 0.0001553259779338014,
"loss": 2.5427,
"theoretical_loss": 3.3929080897756605,
"tokens_seen": 2285699072
},
{
"epoch": 0.69,
"learning_rate": 0.0001553159478435306,
"loss": 2.473,
"theoretical_loss": 3.3929001900809315,
"tokens_seen": 2285764608
},
{
"epoch": 0.69,
"learning_rate": 0.00015530591775325979,
"loss": 2.6913,
"theoretical_loss": 3.3928922906761105,
"tokens_seen": 2285830144
},
{
"epoch": 0.69,
"learning_rate": 0.00015529588766298897,
"loss": 2.5545,
"theoretical_loss": 3.392884391561179,
"tokens_seen": 2285895680
},
{
"epoch": 0.69,
"learning_rate": 0.00015528585757271815,
"loss": 2.6398,
"theoretical_loss": 3.3928764927361184,
"tokens_seen": 2285961216
},
{
"epoch": 0.69,
"learning_rate": 0.00015527582748244733,
"loss": 2.5188,
"theoretical_loss": 3.392868594200909,
"tokens_seen": 2286026752
},
{
"epoch": 0.69,
"learning_rate": 0.00015526579739217654,
"loss": 2.4816,
"theoretical_loss": 3.392860695955532,
"tokens_seen": 2286092288
},
{
"epoch": 0.69,
"learning_rate": 0.00015525576730190572,
"loss": 2.7617,
"theoretical_loss": 3.3928527979999688,
"tokens_seen": 2286157824
},
{
"epoch": 0.69,
"learning_rate": 0.0001552457372116349,
"loss": 2.5809,
"theoretical_loss": 3.3928449003342003,
"tokens_seen": 2286223360
},
{
"epoch": 0.69,
"learning_rate": 0.00015523570712136408,
"loss": 2.45,
"theoretical_loss": 3.3928370029582076,
"tokens_seen": 2286288896
},
{
"epoch": 0.69,
"learning_rate": 0.0001552256770310933,
"loss": 2.607,
"theoretical_loss": 3.3928291058719715,
"tokens_seen": 2286354432
},
{
"epoch": 0.69,
"learning_rate": 0.00015521564694082247,
"loss": 2.5091,
"theoretical_loss": 3.3928212090754735,
"tokens_seen": 2286419968
},
{
"epoch": 0.69,
"learning_rate": 0.00015520561685055165,
"loss": 2.6901,
"theoretical_loss": 3.3928133125686943,
"tokens_seen": 2286485504
},
{
"epoch": 0.69,
"learning_rate": 0.00015519558676028083,
"loss": 2.8055,
"theoretical_loss": 3.3928054163516155,
"tokens_seen": 2286551040
},
{
"epoch": 0.69,
"learning_rate": 0.00015518555667001001,
"loss": 2.6808,
"theoretical_loss": 3.392797520424217,
"tokens_seen": 2286616576
},
{
"epoch": 0.69,
"learning_rate": 0.00015517552657973922,
"loss": 2.6304,
"theoretical_loss": 3.3927896247864813,
"tokens_seen": 2286682112
},
{
"epoch": 0.69,
"learning_rate": 0.0001551654964894684,
"loss": 2.6631,
"theoretical_loss": 3.392781729438388,
"tokens_seen": 2286747648
},
{
"epoch": 0.69,
"learning_rate": 0.00015515546639919758,
"loss": 2.5533,
"theoretical_loss": 3.39277383437992,
"tokens_seen": 2286813184
},
{
"epoch": 0.69,
"learning_rate": 0.00015514543630892677,
"loss": 2.4344,
"theoretical_loss": 3.3927659396110568,
"tokens_seen": 2286878720
},
{
"epoch": 0.69,
"learning_rate": 0.00015513540621865597,
"loss": 2.382,
"theoretical_loss": 3.39275804513178,
"tokens_seen": 2286944256
},
{
"epoch": 0.69,
"learning_rate": 0.00015512537612838515,
"loss": 2.5724,
"theoretical_loss": 3.3927501509420708,
"tokens_seen": 2287009792
},
{
"epoch": 0.69,
"learning_rate": 0.00015511534603811434,
"loss": 2.3724,
"theoretical_loss": 3.3927422570419106,
"tokens_seen": 2287075328
},
{
"epoch": 0.69,
"learning_rate": 0.00015510531594784352,
"loss": 2.7785,
"theoretical_loss": 3.3927343634312797,
"tokens_seen": 2287140864
},
{
"epoch": 0.69,
"objective/train/docs_used": 1294930,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.754033327102661,
"objective/train/theoretical_loss": 3.3927264701101594,
"objective/train/tokens_used": 2307666400,
"theoretical_loss": 3.3927264701101594,
"tokens_seen": 2287206400
},
{
"epoch": 0.69,
"learning_rate": 0.0001550952858575727,
"loss": 2.5893,
"theoretical_loss": 3.3927264701101594,
"tokens_seen": 2287206400
},
{
"epoch": 0.69,
"learning_rate": 0.0001550852557673019,
"loss": 2.7488,
"theoretical_loss": 3.392718577078531,
"tokens_seen": 2287271936
},
{
"epoch": 0.69,
"learning_rate": 0.0001550752256770311,
"loss": 2.7435,
"theoretical_loss": 3.392710684336376,
"tokens_seen": 2287337472
},
{
"epoch": 0.69,
"learning_rate": 0.00015506519558676027,
"loss": 2.6203,
"theoretical_loss": 3.3927027918836745,
"tokens_seen": 2287403008
},
{
"epoch": 0.69,
"learning_rate": 0.00015505516549648945,
"loss": 2.701,
"theoretical_loss": 3.3926948997204085,
"tokens_seen": 2287468544
},
{
"epoch": 0.69,
"learning_rate": 0.00015504513540621866,
"loss": 2.6112,
"theoretical_loss": 3.3926870078465585,
"tokens_seen": 2287534080
},
{
"epoch": 0.69,
"learning_rate": 0.00015503510531594784,
"loss": 2.6776,
"theoretical_loss": 3.392679116262106,
"tokens_seen": 2287599616
},
{
"epoch": 0.69,
"learning_rate": 0.00015502507522567705,
"loss": 2.7171,
"theoretical_loss": 3.3926712249670317,
"tokens_seen": 2287665152
},
{
"epoch": 0.69,
"learning_rate": 0.00015501504513540623,
"loss": 2.5129,
"theoretical_loss": 3.392663333961317,
"tokens_seen": 2287730688
},
{
"epoch": 0.69,
"learning_rate": 0.0001550050150451354,
"loss": 2.652,
"theoretical_loss": 3.392655443244943,
"tokens_seen": 2287796224
},
{
"epoch": 0.69,
"learning_rate": 0.00015499498495486462,
"loss": 2.3884,
"theoretical_loss": 3.3926475528178908,
"tokens_seen": 2287861760
},
{
"epoch": 0.69,
"learning_rate": 0.0001549849548645938,
"loss": 2.7117,
"theoretical_loss": 3.392639662680141,
"tokens_seen": 2287927296
},
{
"epoch": 0.69,
"learning_rate": 0.00015497492477432298,
"loss": 2.699,
"theoretical_loss": 3.3926317728316757,
"tokens_seen": 2287992832
},
{
"epoch": 0.69,
"learning_rate": 0.00015496489468405216,
"loss": 2.5964,
"theoretical_loss": 3.392623883272475,
"tokens_seen": 2288058368
},
{
"epoch": 0.69,
"learning_rate": 0.00015495486459378137,
"loss": 2.8309,
"theoretical_loss": 3.3926159940025205,
"tokens_seen": 2288123904
},
{
"epoch": 0.69,
"learning_rate": 0.00015494483450351055,
"loss": 2.5975,
"theoretical_loss": 3.3926081050217936,
"tokens_seen": 2288189440
},
{
"epoch": 0.69,
"learning_rate": 0.00015493480441323973,
"loss": 2.6604,
"theoretical_loss": 3.392600216330275,
"tokens_seen": 2288254976
},
{
"epoch": 0.69,
"learning_rate": 0.0001549247743229689,
"loss": 2.8049,
"theoretical_loss": 3.3925923279279453,
"tokens_seen": 2288320512
},
{
"epoch": 0.69,
"learning_rate": 0.0001549147442326981,
"loss": 2.7092,
"theoretical_loss": 3.3925844398147866,
"tokens_seen": 2288386048
},
{
"epoch": 0.69,
"learning_rate": 0.0001549047141424273,
"loss": 2.6125,
"theoretical_loss": 3.39257655199078,
"tokens_seen": 2288451584
},
{
"epoch": 0.69,
"learning_rate": 0.00015489468405215648,
"loss": 2.6853,
"theoretical_loss": 3.3925686644559057,
"tokens_seen": 2288517120
},
{
"epoch": 0.69,
"learning_rate": 0.00015488465396188566,
"loss": 2.7757,
"theoretical_loss": 3.3925607772101456,
"tokens_seen": 2288582656
},
{
"epoch": 0.69,
"learning_rate": 0.00015487462387161485,
"loss": 2.4254,
"theoretical_loss": 3.39255289025348,
"tokens_seen": 2288648192
},
{
"epoch": 0.69,
"learning_rate": 0.00015486459378134405,
"loss": 2.6783,
"theoretical_loss": 3.3925450035858913,
"tokens_seen": 2288713728
},
{
"epoch": 0.69,
"learning_rate": 0.00015485456369107323,
"loss": 2.5126,
"theoretical_loss": 3.3925371172073597,
"tokens_seen": 2288779264
},
{
"epoch": 0.69,
"objective/train/docs_used": 1295606,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5279576778411865,
"objective/train/theoretical_loss": 3.3925292311178668,
"objective/train/tokens_used": 2309304800,
"theoretical_loss": 3.3925292311178668,
"tokens_seen": 2288844800
},
{
"epoch": 0.69,
"learning_rate": 0.00015484453360080242,
"loss": 2.7726,
"theoretical_loss": 3.3925292311178668,
"tokens_seen": 2288844800
},
{
"epoch": 0.69,
"learning_rate": 0.0001548345035105316,
"loss": 2.4907,
"theoretical_loss": 3.3925213453173932,
"tokens_seen": 2288910336
},
{
"epoch": 0.69,
"learning_rate": 0.0001548244734202608,
"loss": 2.6412,
"theoretical_loss": 3.3925134598059206,
"tokens_seen": 2288975872
},
{
"epoch": 0.69,
"learning_rate": 0.00015481444332998999,
"loss": 2.6264,
"theoretical_loss": 3.3925055745834296,
"tokens_seen": 2289041408
},
{
"epoch": 0.69,
"learning_rate": 0.00015480441323971917,
"loss": 2.5841,
"theoretical_loss": 3.392497689649902,
"tokens_seen": 2289106944
},
{
"epoch": 0.69,
"learning_rate": 0.00015479438314944835,
"loss": 2.5632,
"theoretical_loss": 3.3924898050053183,
"tokens_seen": 2289172480
},
{
"epoch": 0.69,
"learning_rate": 0.00015478435305917753,
"loss": 2.6681,
"theoretical_loss": 3.3924819206496597,
"tokens_seen": 2289238016
},
{
"epoch": 0.69,
"learning_rate": 0.00015477432296890674,
"loss": 2.4573,
"theoretical_loss": 3.392474036582908,
"tokens_seen": 2289303552
},
{
"epoch": 0.69,
"learning_rate": 0.00015476429287863592,
"loss": 2.718,
"theoretical_loss": 3.3924661528050435,
"tokens_seen": 2289369088
},
{
"epoch": 0.69,
"learning_rate": 0.0001547542627883651,
"loss": 2.5882,
"theoretical_loss": 3.3924582693160477,
"tokens_seen": 2289434624
},
{
"epoch": 0.69,
"learning_rate": 0.00015474423269809428,
"loss": 2.6645,
"theoretical_loss": 3.3924503861159017,
"tokens_seen": 2289500160
},
{
"epoch": 0.69,
"learning_rate": 0.0001547342026078235,
"loss": 2.7494,
"theoretical_loss": 3.392442503204587,
"tokens_seen": 2289565696
},
{
"epoch": 0.69,
"learning_rate": 0.00015472417251755267,
"loss": 2.6978,
"theoretical_loss": 3.392434620582084,
"tokens_seen": 2289631232
},
{
"epoch": 0.69,
"learning_rate": 0.00015471414242728185,
"loss": 2.4154,
"theoretical_loss": 3.392426738248375,
"tokens_seen": 2289696768
},
{
"epoch": 0.69,
"learning_rate": 0.00015470411233701103,
"loss": 2.6906,
"theoretical_loss": 3.39241885620344,
"tokens_seen": 2289762304
},
{
"epoch": 0.69,
"learning_rate": 0.00015469408224674021,
"loss": 2.5141,
"theoretical_loss": 3.3924109744472606,
"tokens_seen": 2289827840
},
{
"epoch": 0.69,
"learning_rate": 0.00015468405215646942,
"loss": 2.3348,
"theoretical_loss": 3.3924030929798183,
"tokens_seen": 2289893376
},
{
"epoch": 0.69,
"learning_rate": 0.0001546740220661986,
"loss": 2.6074,
"theoretical_loss": 3.3923952118010936,
"tokens_seen": 2289958912
},
{
"epoch": 0.69,
"learning_rate": 0.00015466399197592778,
"loss": 2.7382,
"theoretical_loss": 3.392387330911068,
"tokens_seen": 2290024448
},
{
"epoch": 0.69,
"learning_rate": 0.00015465396188565697,
"loss": 2.8263,
"theoretical_loss": 3.392379450309723,
"tokens_seen": 2290089984
},
{
"epoch": 0.69,
"learning_rate": 0.00015464393179538617,
"loss": 2.4688,
"theoretical_loss": 3.392371569997039,
"tokens_seen": 2290155520
},
{
"epoch": 0.69,
"learning_rate": 0.00015463390170511535,
"loss": 2.5034,
"theoretical_loss": 3.392363689972998,
"tokens_seen": 2290221056
},
{
"epoch": 0.69,
"learning_rate": 0.00015462387161484454,
"loss": 3.0071,
"theoretical_loss": 3.3923558102375804,
"tokens_seen": 2290286592
},
{
"epoch": 0.69,
"learning_rate": 0.00015461384152457372,
"loss": 2.7413,
"theoretical_loss": 3.392347930790768,
"tokens_seen": 2290352128
},
{
"epoch": 0.69,
"learning_rate": 0.0001546038114343029,
"loss": 2.8473,
"theoretical_loss": 3.3923400516325417,
"tokens_seen": 2290417664
},
{
"epoch": 0.69,
"objective/train/docs_used": 1296233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.582214117050171,
"objective/train/theoretical_loss": 3.392332172762883,
"objective/train/tokens_used": 2310943200,
"theoretical_loss": 3.392332172762883,
"tokens_seen": 2290483200
},
{
"epoch": 0.69,
"learning_rate": 0.0001545937813440321,
"loss": 2.6855,
"theoretical_loss": 3.392332172762883,
"tokens_seen": 2290483200
},
{
"epoch": 0.69,
"learning_rate": 0.0001545837512537613,
"loss": 2.6838,
"theoretical_loss": 3.3923242941817717,
"tokens_seen": 2290548736
},
{
"epoch": 0.69,
"learning_rate": 0.00015457372116349047,
"loss": 2.5226,
"theoretical_loss": 3.3923164158891908,
"tokens_seen": 2290614272
},
{
"epoch": 0.69,
"learning_rate": 0.00015456369107321965,
"loss": 2.528,
"theoretical_loss": 3.3923085378851208,
"tokens_seen": 2290679808
},
{
"epoch": 0.69,
"learning_rate": 0.00015455366098294886,
"loss": 2.8495,
"theoretical_loss": 3.3923006601695427,
"tokens_seen": 2290745344
},
{
"epoch": 0.69,
"learning_rate": 0.00015454363089267804,
"loss": 2.6534,
"theoretical_loss": 3.3922927827424374,
"tokens_seen": 2290810880
},
{
"epoch": 0.69,
"learning_rate": 0.00015453360080240722,
"loss": 2.8208,
"theoretical_loss": 3.3922849056037867,
"tokens_seen": 2290876416
},
{
"epoch": 0.69,
"learning_rate": 0.0001545235707121364,
"loss": 2.6387,
"theoretical_loss": 3.392277028753572,
"tokens_seen": 2290941952
},
{
"epoch": 0.69,
"learning_rate": 0.00015451354062186558,
"loss": 2.687,
"theoretical_loss": 3.3922691521917736,
"tokens_seen": 2291007488
},
{
"epoch": 0.69,
"learning_rate": 0.0001545035105315948,
"loss": 2.7175,
"theoretical_loss": 3.392261275918373,
"tokens_seen": 2291073024
},
{
"epoch": 0.69,
"learning_rate": 0.00015449348044132397,
"loss": 2.4453,
"theoretical_loss": 3.3922533999333515,
"tokens_seen": 2291138560
},
{
"epoch": 0.69,
"learning_rate": 0.00015448345035105315,
"loss": 2.5462,
"theoretical_loss": 3.3922455242366905,
"tokens_seen": 2291204096
},
{
"epoch": 0.69,
"learning_rate": 0.00015447342026078233,
"loss": 2.6868,
"theoretical_loss": 3.392237648828371,
"tokens_seen": 2291269632
},
{
"epoch": 0.69,
"learning_rate": 0.00015446339017051154,
"loss": 2.6614,
"theoretical_loss": 3.392229773708374,
"tokens_seen": 2291335168
},
{
"epoch": 0.69,
"learning_rate": 0.00015445336008024072,
"loss": 2.7653,
"theoretical_loss": 3.3922218988766812,
"tokens_seen": 2291400704
},
{
"epoch": 0.69,
"learning_rate": 0.0001544433299899699,
"loss": 2.6727,
"theoretical_loss": 3.3922140243332732,
"tokens_seen": 2291466240
},
{
"epoch": 0.69,
"learning_rate": 0.00015443329989969909,
"loss": 2.6497,
"theoretical_loss": 3.3922061500781315,
"tokens_seen": 2291531776
},
{
"epoch": 0.69,
"learning_rate": 0.00015442326980942827,
"loss": 2.8892,
"theoretical_loss": 3.3921982761112375,
"tokens_seen": 2291597312
},
{
"epoch": 0.69,
"learning_rate": 0.00015441323971915747,
"loss": 2.6577,
"theoretical_loss": 3.392190402432572,
"tokens_seen": 2291662848
},
{
"epoch": 0.69,
"learning_rate": 0.00015440320962888666,
"loss": 2.3838,
"theoretical_loss": 3.392182529042117,
"tokens_seen": 2291728384
},
{
"epoch": 0.69,
"learning_rate": 0.00015439317953861584,
"loss": 2.5699,
"theoretical_loss": 3.3921746559398525,
"tokens_seen": 2291793920
},
{
"epoch": 0.69,
"learning_rate": 0.00015438314944834502,
"loss": 2.6296,
"theoretical_loss": 3.3921667831257607,
"tokens_seen": 2291859456
},
{
"epoch": 0.69,
"learning_rate": 0.00015437311935807423,
"loss": 2.8505,
"theoretical_loss": 3.392158910599822,
"tokens_seen": 2291924992
},
{
"epoch": 0.69,
"learning_rate": 0.0001543630892678034,
"loss": 2.9119,
"theoretical_loss": 3.3921510383620186,
"tokens_seen": 2291990528
},
{
"epoch": 0.69,
"learning_rate": 0.0001543530591775326,
"loss": 2.6388,
"theoretical_loss": 3.3921431664123305,
"tokens_seen": 2292056064
},
{
"epoch": 0.69,
"objective/train/docs_used": 1296233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.038449287414551,
"objective/train/theoretical_loss": 3.39213529475074,
"objective/train/tokens_used": 2312581600,
"theoretical_loss": 3.39213529475074,
"tokens_seen": 2292121600
},
{
"epoch": 0.69,
"learning_rate": 0.00015434302908726177,
"loss": 2.5577,
"theoretical_loss": 3.39213529475074,
"tokens_seen": 2292121600
},
{
"epoch": 0.69,
"learning_rate": 0.00015433299899699095,
"loss": 2.5476,
"theoretical_loss": 3.392127423377228,
"tokens_seen": 2292187136
},
{
"epoch": 0.69,
"learning_rate": 0.00015432296890672016,
"loss": 2.3747,
"theoretical_loss": 3.3921195522917755,
"tokens_seen": 2292252672
},
{
"epoch": 0.69,
"learning_rate": 0.00015431293881644934,
"loss": 2.83,
"theoretical_loss": 3.392111681494364,
"tokens_seen": 2292318208
},
{
"epoch": 0.69,
"learning_rate": 0.00015430290872617852,
"loss": 2.6868,
"theoretical_loss": 3.3921038109849744,
"tokens_seen": 2292383744
},
{
"epoch": 0.69,
"learning_rate": 0.0001542928786359077,
"loss": 2.7735,
"theoretical_loss": 3.3920959407635882,
"tokens_seen": 2292449280
},
{
"epoch": 0.69,
"learning_rate": 0.0001542828485456369,
"loss": 2.575,
"theoretical_loss": 3.3920880708301864,
"tokens_seen": 2292514816
},
{
"epoch": 0.69,
"learning_rate": 0.00015427281845536612,
"loss": 2.7506,
"theoretical_loss": 3.3920802011847506,
"tokens_seen": 2292580352
},
{
"epoch": 0.69,
"learning_rate": 0.0001542627883650953,
"loss": 2.6799,
"theoretical_loss": 3.3920723318272614,
"tokens_seen": 2292645888
},
{
"epoch": 0.69,
"learning_rate": 0.00015425275827482448,
"loss": 2.8005,
"theoretical_loss": 3.392064462757701,
"tokens_seen": 2292711424
},
{
"epoch": 0.69,
"learning_rate": 0.0001542427281845537,
"loss": 2.7933,
"theoretical_loss": 3.39205659397605,
"tokens_seen": 2292776960
},
{
"epoch": 0.69,
"learning_rate": 0.00015423269809428287,
"loss": 2.7906,
"theoretical_loss": 3.3920487254822893,
"tokens_seen": 2292842496
},
{
"epoch": 0.69,
"learning_rate": 0.00015422266800401205,
"loss": 2.7088,
"theoretical_loss": 3.3920408572764007,
"tokens_seen": 2292908032
},
{
"epoch": 0.69,
"learning_rate": 0.00015421263791374123,
"loss": 2.7092,
"theoretical_loss": 3.392032989358365,
"tokens_seen": 2292973568
},
{
"epoch": 0.69,
"learning_rate": 0.00015420260782347041,
"loss": 2.7581,
"theoretical_loss": 3.392025121728164,
"tokens_seen": 2293039104
},
{
"epoch": 0.69,
"learning_rate": 0.00015419257773319962,
"loss": 2.8399,
"theoretical_loss": 3.3920172543857787,
"tokens_seen": 2293104640
},
{
"epoch": 0.69,
"learning_rate": 0.0001541825476429288,
"loss": 2.6602,
"theoretical_loss": 3.39200938733119,
"tokens_seen": 2293170176
},
{
"epoch": 0.69,
"learning_rate": 0.00015417251755265798,
"loss": 2.9519,
"theoretical_loss": 3.39200152056438,
"tokens_seen": 2293235712
},
{
"epoch": 0.69,
"learning_rate": 0.00015416248746238717,
"loss": 2.8215,
"theoretical_loss": 3.3919936540853293,
"tokens_seen": 2293301248
},
{
"epoch": 0.69,
"learning_rate": 0.00015415245737211637,
"loss": 2.8487,
"theoretical_loss": 3.3919857878940194,
"tokens_seen": 2293366784
},
{
"epoch": 0.69,
"learning_rate": 0.00015414242728184555,
"loss": 2.6305,
"theoretical_loss": 3.391977921990431,
"tokens_seen": 2293432320
},
{
"epoch": 0.69,
"learning_rate": 0.00015413239719157474,
"loss": 2.9629,
"theoretical_loss": 3.3919700563745456,
"tokens_seen": 2293497856
},
{
"epoch": 0.7,
"learning_rate": 0.00015412236710130392,
"loss": 2.6741,
"theoretical_loss": 3.3919621910463453,
"tokens_seen": 2293563392
},
{
"epoch": 0.7,
"learning_rate": 0.0001541123370110331,
"loss": 3.0912,
"theoretical_loss": 3.39195432600581,
"tokens_seen": 2293628928
},
{
"epoch": 0.7,
"learning_rate": 0.0001541023069207623,
"loss": 2.6439,
"theoretical_loss": 3.391946461252922,
"tokens_seen": 2293694464
},
{
"debugging/Self-BLEU-5": 0.21086513998351097,
"debugging/distinct-1-grams": 0.8447142012939409,
"debugging/distinct-2-grams": 0.9847426470588235,
"debugging/entropy-1-grams": 5.159037198858333,
"debugging/entropy-2-grams": 5.60072408146037,
"debugging/length": 464.75,
"debugging/num_segments": 4,
"epoch": 0.7,
"objective/train/docs_used": 1296233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.2612814903259277,
"objective/train/theoretical_loss": 3.3919385967876625,
"objective/train/tokens_used": 2314220000,
"theoretical_loss": 3.3919385967876625,
"tokens_seen": 2293760000
},
{
"epoch": 0.7,
"learning_rate": 0.0001540922768304915,
"loss": 2.9603,
"theoretical_loss": 3.3919385967876625,
"tokens_seen": 2293760000
},
{
"epoch": 0.7,
"learning_rate": 0.00015408224674022067,
"loss": 2.8806,
"theoretical_loss": 3.391930732610012,
"tokens_seen": 2293825536
},
{
"epoch": 0.7,
"learning_rate": 0.00015407221664994985,
"loss": 2.6327,
"theoretical_loss": 3.3919228687199525,
"tokens_seen": 2293891072
},
{
"epoch": 0.7,
"learning_rate": 0.00015406218655967906,
"loss": 2.9537,
"theoretical_loss": 3.3919150051174647,
"tokens_seen": 2293956608
},
{
"epoch": 0.7,
"learning_rate": 0.00015405215646940824,
"loss": 2.7843,
"theoretical_loss": 3.3919071418025304,
"tokens_seen": 2294022144
},
{
"epoch": 0.7,
"learning_rate": 0.00015404212637913742,
"loss": 2.5092,
"theoretical_loss": 3.3918992787751305,
"tokens_seen": 2294087680
},
{
"epoch": 0.7,
"learning_rate": 0.0001540320962888666,
"loss": 3.0998,
"theoretical_loss": 3.391891416035247,
"tokens_seen": 2294153216
},
{
"epoch": 0.7,
"learning_rate": 0.00015402206619859578,
"loss": 2.8212,
"theoretical_loss": 3.39188355358286,
"tokens_seen": 2294218752
},
{
"epoch": 0.7,
"learning_rate": 0.000154012036108325,
"loss": 2.7717,
"theoretical_loss": 3.3918756914179515,
"tokens_seen": 2294284288
},
{
"epoch": 0.7,
"learning_rate": 0.00015400200601805417,
"loss": 2.9407,
"theoretical_loss": 3.3918678295405025,
"tokens_seen": 2294349824
},
{
"epoch": 0.7,
"learning_rate": 0.00015399197592778335,
"loss": 2.6485,
"theoretical_loss": 3.3918599679504946,
"tokens_seen": 2294415360
},
{
"epoch": 0.7,
"learning_rate": 0.00015398194583751253,
"loss": 2.7524,
"theoretical_loss": 3.391852106647909,
"tokens_seen": 2294480896
},
{
"epoch": 0.7,
"learning_rate": 0.00015397191574724174,
"loss": 2.5755,
"theoretical_loss": 3.3918442456327265,
"tokens_seen": 2294546432
},
{
"epoch": 0.7,
"learning_rate": 0.00015396188565697092,
"loss": 2.8874,
"theoretical_loss": 3.391836384904929,
"tokens_seen": 2294611968
},
{
"epoch": 0.7,
"learning_rate": 0.0001539518555667001,
"loss": 2.6937,
"theoretical_loss": 3.3918285244644975,
"tokens_seen": 2294677504
},
{
"epoch": 0.7,
"learning_rate": 0.00015394182547642929,
"loss": 2.5561,
"theoretical_loss": 3.3918206643114135,
"tokens_seen": 2294743040
},
{
"epoch": 0.7,
"learning_rate": 0.00015393179538615847,
"loss": 2.7027,
"theoretical_loss": 3.391812804445658,
"tokens_seen": 2294808576
},
{
"epoch": 0.7,
"learning_rate": 0.00015392176529588768,
"loss": 2.8232,
"theoretical_loss": 3.3918049448672125,
"tokens_seen": 2294874112
},
{
"epoch": 0.7,
"learning_rate": 0.00015391173520561686,
"loss": 2.6085,
"theoretical_loss": 3.391797085576058,
"tokens_seen": 2294939648
},
{
"epoch": 0.7,
"learning_rate": 0.00015390170511534604,
"loss": 2.8504,
"theoretical_loss": 3.3917892265721763,
"tokens_seen": 2295005184
},
{
"epoch": 0.7,
"learning_rate": 0.00015389167502507522,
"loss": 2.7212,
"theoretical_loss": 3.391781367855548,
"tokens_seen": 2295070720
},
{
"epoch": 0.7,
"learning_rate": 0.00015388164493480443,
"loss": 2.7541,
"theoretical_loss": 3.391773509426155,
"tokens_seen": 2295136256
},
{
"epoch": 0.7,
"learning_rate": 0.0001538716148445336,
"loss": 2.664,
"theoretical_loss": 3.3917656512839787,
"tokens_seen": 2295201792
},
{
"epoch": 0.7,
"learning_rate": 0.0001538615847542628,
"loss": 2.9392,
"theoretical_loss": 3.391757793429,
"tokens_seen": 2295267328
},
{
"epoch": 0.7,
"learning_rate": 0.00015385155466399197,
"loss": 2.7541,
"theoretical_loss": 3.3917499358612,
"tokens_seen": 2295332864
},
{
"epoch": 0.7,
"objective/train/docs_used": 1296233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.2856571674346924,
"objective/train/theoretical_loss": 3.391742078580561,
"objective/train/tokens_used": 2315858400,
"theoretical_loss": 3.391742078580561,
"tokens_seen": 2295398400
},
{
"epoch": 0.7,
"learning_rate": 0.00015384152457372115,
"loss": 2.6852,
"theoretical_loss": 3.391742078580561,
"tokens_seen": 2295398400
},
{
"epoch": 0.7,
"learning_rate": 0.00015383149448345036,
"loss": 2.7681,
"theoretical_loss": 3.391734221587063,
"tokens_seen": 2295463936
},
{
"epoch": 0.7,
"learning_rate": 0.00015382146439317954,
"loss": 2.7094,
"theoretical_loss": 3.391726364880688,
"tokens_seen": 2295529472
},
{
"epoch": 0.7,
"learning_rate": 0.00015381143430290872,
"loss": 2.7944,
"theoretical_loss": 3.3917185084614174,
"tokens_seen": 2295595008
},
{
"epoch": 0.7,
"learning_rate": 0.0001538014042126379,
"loss": 2.8674,
"theoretical_loss": 3.391710652329232,
"tokens_seen": 2295660544
},
{
"epoch": 0.7,
"learning_rate": 0.0001537913741223671,
"loss": 2.9441,
"theoretical_loss": 3.391702796484114,
"tokens_seen": 2295726080
},
{
"epoch": 0.7,
"learning_rate": 0.0001537813440320963,
"loss": 2.7029,
"theoretical_loss": 3.391694940926044,
"tokens_seen": 2295791616
},
{
"epoch": 0.7,
"learning_rate": 0.00015377131394182547,
"loss": 2.6489,
"theoretical_loss": 3.3916870856550037,
"tokens_seen": 2295857152
},
{
"epoch": 0.7,
"learning_rate": 0.00015376128385155465,
"loss": 2.914,
"theoretical_loss": 3.3916792306709738,
"tokens_seen": 2295922688
},
{
"epoch": 0.7,
"learning_rate": 0.00015375125376128386,
"loss": 2.7531,
"theoretical_loss": 3.391671375973936,
"tokens_seen": 2295988224
},
{
"epoch": 0.7,
"learning_rate": 0.00015374122367101304,
"loss": 2.7955,
"theoretical_loss": 3.3916635215638724,
"tokens_seen": 2296053760
},
{
"epoch": 0.7,
"learning_rate": 0.00015373119358074222,
"loss": 2.8183,
"theoretical_loss": 3.391655667440763,
"tokens_seen": 2296119296
},
{
"epoch": 0.7,
"learning_rate": 0.0001537211634904714,
"loss": 3.0187,
"theoretical_loss": 3.39164781360459,
"tokens_seen": 2296184832
},
{
"epoch": 0.7,
"learning_rate": 0.0001537111334002006,
"loss": 2.8705,
"theoretical_loss": 3.391639960055334,
"tokens_seen": 2296250368
},
{
"epoch": 0.7,
"learning_rate": 0.0001537011033099298,
"loss": 2.8395,
"theoretical_loss": 3.391632106792977,
"tokens_seen": 2296315904
},
{
"epoch": 0.7,
"learning_rate": 0.00015369107321965898,
"loss": 2.9822,
"theoretical_loss": 3.3916242538175005,
"tokens_seen": 2296381440
},
{
"epoch": 0.7,
"learning_rate": 0.00015368104312938816,
"loss": 2.9315,
"theoretical_loss": 3.3916164011288847,
"tokens_seen": 2296446976
},
{
"epoch": 0.7,
"learning_rate": 0.00015367101303911734,
"loss": 2.8208,
"theoretical_loss": 3.3916085487271124,
"tokens_seen": 2296512512
},
{
"epoch": 0.7,
"learning_rate": 0.00015366098294884655,
"loss": 2.6686,
"theoretical_loss": 3.391600696612164,
"tokens_seen": 2296578048
},
{
"epoch": 0.7,
"learning_rate": 0.00015365095285857573,
"loss": 2.7874,
"theoretical_loss": 3.3915928447840207,
"tokens_seen": 2296643584
},
{
"epoch": 0.7,
"learning_rate": 0.0001536409227683049,
"loss": 2.9008,
"theoretical_loss": 3.3915849932426645,
"tokens_seen": 2296709120
},
{
"epoch": 0.7,
"learning_rate": 0.0001536308926780341,
"loss": 2.9182,
"theoretical_loss": 3.3915771419880762,
"tokens_seen": 2296774656
},
{
"epoch": 0.7,
"learning_rate": 0.00015362086258776327,
"loss": 2.6995,
"theoretical_loss": 3.3915692910202377,
"tokens_seen": 2296840192
},
{
"epoch": 0.7,
"learning_rate": 0.00015361083249749248,
"loss": 2.739,
"theoretical_loss": 3.39156144033913,
"tokens_seen": 2296905728
},
{
"epoch": 0.7,
"learning_rate": 0.00015360080240722166,
"loss": 2.8458,
"theoretical_loss": 3.3915535899447344,
"tokens_seen": 2296971264
},
{
"epoch": 0.7,
"objective/train/docs_used": 1296233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.0080392360687256,
"objective/train/theoretical_loss": 3.3915457398370323,
"objective/train/tokens_used": 2317496800,
"theoretical_loss": 3.3915457398370323,
"tokens_seen": 2297036800
},
{
"epoch": 0.7,
"learning_rate": 0.00015359077231695084,
"loss": 2.8356,
"theoretical_loss": 3.3915457398370323,
"tokens_seen": 2297036800
},
{
"epoch": 0.7,
"learning_rate": 0.00015358074222668002,
"loss": 3.1263,
"theoretical_loss": 3.391537890016005,
"tokens_seen": 2297102336
},
{
"epoch": 0.7,
"learning_rate": 0.00015357071213640923,
"loss": 3.1339,
"theoretical_loss": 3.3915300404816344,
"tokens_seen": 2297167872
},
{
"epoch": 0.7,
"learning_rate": 0.0001535606820461384,
"loss": 3.0284,
"theoretical_loss": 3.391522191233901,
"tokens_seen": 2297233408
},
{
"epoch": 0.7,
"learning_rate": 0.0001535506519558676,
"loss": 2.717,
"theoretical_loss": 3.391514342272786,
"tokens_seen": 2297298944
},
{
"epoch": 0.7,
"learning_rate": 0.00015354062186559677,
"loss": 2.8315,
"theoretical_loss": 3.391506493598272,
"tokens_seen": 2297364480
},
{
"epoch": 0.7,
"learning_rate": 0.00015353059177532596,
"loss": 2.844,
"theoretical_loss": 3.3914986452103397,
"tokens_seen": 2297430016
},
{
"epoch": 0.7,
"learning_rate": 0.0001535205616850552,
"loss": 2.8497,
"theoretical_loss": 3.39149079710897,
"tokens_seen": 2297495552
},
{
"epoch": 0.7,
"learning_rate": 0.00015351053159478437,
"loss": 2.8926,
"theoretical_loss": 3.391482949294145,
"tokens_seen": 2297561088
},
{
"epoch": 0.7,
"learning_rate": 0.00015350050150451355,
"loss": 2.9836,
"theoretical_loss": 3.3914751017658458,
"tokens_seen": 2297626624
},
{
"epoch": 0.7,
"learning_rate": 0.00015349047141424273,
"loss": 2.7462,
"theoretical_loss": 3.3914672545240534,
"tokens_seen": 2297692160
},
{
"epoch": 0.7,
"learning_rate": 0.00015348044132397194,
"loss": 2.8487,
"theoretical_loss": 3.3914594075687496,
"tokens_seen": 2297757696
},
{
"epoch": 0.7,
"learning_rate": 0.00015347041123370112,
"loss": 2.8568,
"theoretical_loss": 3.391451560899916,
"tokens_seen": 2297823232
},
{
"epoch": 0.7,
"learning_rate": 0.0001534603811434303,
"loss": 2.6583,
"theoretical_loss": 3.391443714517533,
"tokens_seen": 2297888768
},
{
"epoch": 0.7,
"learning_rate": 0.00015345035105315949,
"loss": 2.9483,
"theoretical_loss": 3.391435868421583,
"tokens_seen": 2297954304
},
{
"epoch": 0.7,
"learning_rate": 0.00015344032096288867,
"loss": 2.7125,
"theoretical_loss": 3.391428022612047,
"tokens_seen": 2298019840
},
{
"epoch": 0.7,
"learning_rate": 0.00015343029087261788,
"loss": 2.6959,
"theoretical_loss": 3.391420177088906,
"tokens_seen": 2298085376
},
{
"epoch": 0.7,
"learning_rate": 0.00015342026078234706,
"loss": 2.9709,
"theoretical_loss": 3.391412331852142,
"tokens_seen": 2298150912
},
{
"epoch": 0.7,
"learning_rate": 0.00015341023069207624,
"loss": 3.082,
"theoretical_loss": 3.391404486901736,
"tokens_seen": 2298216448
},
{
"epoch": 0.7,
"learning_rate": 0.00015340020060180542,
"loss": 2.729,
"theoretical_loss": 3.3913966422376696,
"tokens_seen": 2298281984
},
{
"epoch": 0.7,
"learning_rate": 0.00015339017051153463,
"loss": 2.7919,
"theoretical_loss": 3.391388797859924,
"tokens_seen": 2298347520
},
{
"epoch": 0.7,
"learning_rate": 0.0001533801404212638,
"loss": 2.8373,
"theoretical_loss": 3.3913809537684805,
"tokens_seen": 2298413056
},
{
"epoch": 0.7,
"learning_rate": 0.000153370110330993,
"loss": 2.7673,
"theoretical_loss": 3.391373109963321,
"tokens_seen": 2298478592
},
{
"epoch": 0.7,
"learning_rate": 0.00015336008024072217,
"loss": 2.905,
"theoretical_loss": 3.391365266444426,
"tokens_seen": 2298544128
},
{
"epoch": 0.7,
"learning_rate": 0.00015335005015045135,
"loss": 2.6718,
"theoretical_loss": 3.3913574232117782,
"tokens_seen": 2298609664
},
{
"epoch": 0.7,
"objective/train/docs_used": 1296233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.186466693878174,
"objective/train/theoretical_loss": 3.3913495802653575,
"objective/train/tokens_used": 2319135200,
"theoretical_loss": 3.3913495802653575,
"tokens_seen": 2298675200
},
{
"epoch": 0.7,
"learning_rate": 0.00015334002006018056,
"loss": 2.7623,
"theoretical_loss": 3.3913495802653575,
"tokens_seen": 2298675200
},
{
"epoch": 0.7,
"learning_rate": 0.00015332998996990974,
"loss": 2.8059,
"theoretical_loss": 3.3913417376051465,
"tokens_seen": 2298740736
},
{
"epoch": 0.7,
"learning_rate": 0.00015331995987963892,
"loss": 2.5515,
"theoretical_loss": 3.391333895231126,
"tokens_seen": 2298806272
},
{
"epoch": 0.7,
"learning_rate": 0.0001533099297893681,
"loss": 2.8552,
"theoretical_loss": 3.3913260531432776,
"tokens_seen": 2298871808
},
{
"epoch": 0.7,
"learning_rate": 0.0001532998996990973,
"loss": 2.7633,
"theoretical_loss": 3.3913182113415825,
"tokens_seen": 2298937344
},
{
"epoch": 0.7,
"learning_rate": 0.0001532898696088265,
"loss": 2.7611,
"theoretical_loss": 3.391310369826022,
"tokens_seen": 2299002880
},
{
"epoch": 0.7,
"learning_rate": 0.00015327983951855567,
"loss": 2.7135,
"theoretical_loss": 3.3913025285965777,
"tokens_seen": 2299068416
},
{
"epoch": 0.7,
"learning_rate": 0.00015326980942828485,
"loss": 2.9125,
"theoretical_loss": 3.3912946876532315,
"tokens_seen": 2299133952
},
{
"epoch": 0.7,
"learning_rate": 0.00015325977933801406,
"loss": 2.8592,
"theoretical_loss": 3.391286846995964,
"tokens_seen": 2299199488
},
{
"epoch": 0.7,
"learning_rate": 0.00015324974924774324,
"loss": 2.8784,
"theoretical_loss": 3.3912790066247567,
"tokens_seen": 2299265024
},
{
"epoch": 0.7,
"learning_rate": 0.00015323971915747243,
"loss": 3.0436,
"theoretical_loss": 3.3912711665395916,
"tokens_seen": 2299330560
},
{
"epoch": 0.7,
"learning_rate": 0.0001532296890672016,
"loss": 2.8454,
"theoretical_loss": 3.3912633267404497,
"tokens_seen": 2299396096
},
{
"epoch": 0.7,
"learning_rate": 0.0001532196589769308,
"loss": 2.7817,
"theoretical_loss": 3.391255487227313,
"tokens_seen": 2299461632
},
{
"epoch": 0.7,
"learning_rate": 0.00015320962888666,
"loss": 2.8149,
"theoretical_loss": 3.3912476480001614,
"tokens_seen": 2299527168
},
{
"epoch": 0.7,
"learning_rate": 0.00015319959879638918,
"loss": 2.9108,
"theoretical_loss": 3.3912398090589777,
"tokens_seen": 2299592704
},
{
"epoch": 0.7,
"learning_rate": 0.00015318956870611836,
"loss": 2.9443,
"theoretical_loss": 3.391231970403743,
"tokens_seen": 2299658240
},
{
"epoch": 0.7,
"learning_rate": 0.00015317953861584754,
"loss": 2.6008,
"theoretical_loss": 3.3912241320344387,
"tokens_seen": 2299723776
},
{
"epoch": 0.7,
"learning_rate": 0.00015316950852557675,
"loss": 2.873,
"theoretical_loss": 3.3912162939510457,
"tokens_seen": 2299789312
},
{
"epoch": 0.7,
"learning_rate": 0.00015315947843530593,
"loss": 2.9337,
"theoretical_loss": 3.3912084561535467,
"tokens_seen": 2299854848
},
{
"epoch": 0.7,
"learning_rate": 0.0001531494483450351,
"loss": 2.766,
"theoretical_loss": 3.3912006186419217,
"tokens_seen": 2299920384
},
{
"epoch": 0.7,
"learning_rate": 0.0001531394182547643,
"loss": 2.8907,
"theoretical_loss": 3.391192781416153,
"tokens_seen": 2299985920
},
{
"epoch": 0.7,
"learning_rate": 0.00015312938816449347,
"loss": 2.6308,
"theoretical_loss": 3.3911849444762217,
"tokens_seen": 2300051456
},
{
"epoch": 0.7,
"learning_rate": 0.00015311935807422268,
"loss": 2.9451,
"theoretical_loss": 3.3911771078221093,
"tokens_seen": 2300116992
},
{
"epoch": 0.7,
"learning_rate": 0.00015310932798395186,
"loss": 2.8736,
"theoretical_loss": 3.391169271453797,
"tokens_seen": 2300182528
},
{
"epoch": 0.7,
"learning_rate": 0.00015309929789368104,
"loss": 2.9509,
"theoretical_loss": 3.3911614353712665,
"tokens_seen": 2300248064
},
{
"epoch": 0.7,
"objective/train/docs_used": 1296233,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 3.130023241043091,
"objective/train/theoretical_loss": 3.3911535995744995,
"objective/train/tokens_used": 2320773600,
"theoretical_loss": 3.3911535995744995,
"tokens_seen": 2300313600
},
{
"epoch": 0.7,
"learning_rate": 0.00015308926780341022,
"loss": 2.8927,
"theoretical_loss": 3.3911535995744995,
"tokens_seen": 2300313600
},
{
"epoch": 0.7,
"learning_rate": 0.00015307923771313943,
"loss": 2.8513,
"theoretical_loss": 3.3911457640634772,
"tokens_seen": 2300379136
},
{
"epoch": 0.7,
"learning_rate": 0.0001530692076228686,
"loss": 2.6882,
"theoretical_loss": 3.3911379288381807,
"tokens_seen": 2300444672
},
{
"epoch": 0.7,
"learning_rate": 0.0001530591775325978,
"loss": 2.8767,
"theoretical_loss": 3.391130093898592,
"tokens_seen": 2300510208
},
{
"epoch": 0.7,
"learning_rate": 0.00015304914744232697,
"loss": 2.9125,
"theoretical_loss": 3.391122259244692,
"tokens_seen": 2300575744
},
{
"epoch": 0.7,
"learning_rate": 0.00015303911735205616,
"loss": 3.0513,
"theoretical_loss": 3.3911144248764624,
"tokens_seen": 2300641280
},
{
"epoch": 0.7,
"learning_rate": 0.00015302908726178536,
"loss": 3.0137,
"theoretical_loss": 3.3911065907938847,
"tokens_seen": 2300706816
},
{
"epoch": 0.7,
"learning_rate": 0.00015301905717151455,
"loss": 2.6932,
"theoretical_loss": 3.3910987569969406,
"tokens_seen": 2300772352
},
{
"epoch": 0.7,
"learning_rate": 0.00015300902708124373,
"loss": 2.986,
"theoretical_loss": 3.391090923485611,
"tokens_seen": 2300837888
},
{
"epoch": 0.7,
"learning_rate": 0.0001529989969909729,
"loss": 2.7524,
"theoretical_loss": 3.391083090259878,
"tokens_seen": 2300903424
},
{
"epoch": 0.7,
"learning_rate": 0.00015298896690070212,
"loss": 2.72,
"theoretical_loss": 3.391075257319722,
"tokens_seen": 2300968960
},
{
"epoch": 0.7,
"learning_rate": 0.0001529789368104313,
"loss": 2.7409,
"theoretical_loss": 3.391067424665126,
"tokens_seen": 2301034496
},
{
"epoch": 0.7,
"learning_rate": 0.00015296890672016048,
"loss": 2.732,
"theoretical_loss": 3.3910595922960693,
"tokens_seen": 2301100032
},
{
"epoch": 0.7,
"learning_rate": 0.00015295887662988966,
"loss": 2.6171,
"theoretical_loss": 3.3910517602125356,
"tokens_seen": 2301165568
},
{
"epoch": 0.7,
"learning_rate": 0.00015294884653961884,
"loss": 2.7893,
"theoretical_loss": 3.391043928414505,
"tokens_seen": 2301231104
},
{
"epoch": 0.7,
"learning_rate": 0.00015293881644934805,
"loss": 2.9301,
"theoretical_loss": 3.3910360969019595,
"tokens_seen": 2301296640
},
{
"epoch": 0.7,
"learning_rate": 0.00015292878635907723,
"loss": 2.7953,
"theoretical_loss": 3.3910282656748807,
"tokens_seen": 2301362176
},
{
"epoch": 0.7,
"learning_rate": 0.0001529187562688064,
"loss": 3.017,
"theoretical_loss": 3.391020434733249,
"tokens_seen": 2301427712
},
{
"epoch": 0.7,
"learning_rate": 0.0001529087261785356,
"loss": 2.6943,
"theoretical_loss": 3.3910126040770474,
"tokens_seen": 2301493248
},
{
"epoch": 0.7,
"learning_rate": 0.0001528986960882648,
"loss": 2.8533,
"theoretical_loss": 3.3910047737062565,
"tokens_seen": 2301558784
},
{
"epoch": 0.7,
"learning_rate": 0.00015288866599799398,
"loss": 2.7834,
"theoretical_loss": 3.390996943620858,
"tokens_seen": 2301624320
},
{
"epoch": 0.7,
"learning_rate": 0.00015287863590772316,
"loss": 2.6424,
"theoretical_loss": 3.3909891138208326,
"tokens_seen": 2301689856
},
{
"epoch": 0.7,
"learning_rate": 0.00015286860581745234,
"loss": 2.9186,
"theoretical_loss": 3.390981284306163,
"tokens_seen": 2301755392
},
{
"epoch": 0.7,
"learning_rate": 0.00015285857572718152,
"loss": 3.0323,
"theoretical_loss": 3.3909734550768302,
"tokens_seen": 2301820928
},
{
"epoch": 0.7,
"learning_rate": 0.00015284854563691073,
"loss": 2.6057,
"theoretical_loss": 3.3909656261328154,
"tokens_seen": 2301886464
},
{
"epoch": 0.7,
"objective/train/docs_used": 1297386,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9890995025634766,
"objective/train/theoretical_loss": 3.3909577974741003,
"objective/train/tokens_used": 2322412000,
"theoretical_loss": 3.3909577974741003,
"tokens_seen": 2301952000
},
{
"epoch": 0.7,
"learning_rate": 0.00015283851554663991,
"loss": 2.8053,
"theoretical_loss": 3.3909577974741003,
"tokens_seen": 2301952000
},
{
"epoch": 0.7,
"learning_rate": 0.0001528284854563691,
"loss": 2.7626,
"theoretical_loss": 3.390949969100666,
"tokens_seen": 2302017536
},
{
"epoch": 0.7,
"learning_rate": 0.00015281845536609828,
"loss": 2.7239,
"theoretical_loss": 3.390942141012495,
"tokens_seen": 2302083072
},
{
"epoch": 0.7,
"learning_rate": 0.00015280842527582748,
"loss": 2.9962,
"theoretical_loss": 3.3909343132095673,
"tokens_seen": 2302148608
},
{
"epoch": 0.7,
"learning_rate": 0.00015279839518555667,
"loss": 2.8683,
"theoretical_loss": 3.3909264856918657,
"tokens_seen": 2302214144
},
{
"epoch": 0.7,
"learning_rate": 0.00015278836509528585,
"loss": 2.7754,
"theoretical_loss": 3.3909186584593716,
"tokens_seen": 2302279680
},
{
"epoch": 0.7,
"learning_rate": 0.00015277833500501503,
"loss": 2.8668,
"theoretical_loss": 3.3909108315120653,
"tokens_seen": 2302345216
},
{
"epoch": 0.7,
"learning_rate": 0.00015276830491474426,
"loss": 2.8809,
"theoretical_loss": 3.3909030048499296,
"tokens_seen": 2302410752
},
{
"epoch": 0.7,
"learning_rate": 0.00015275827482447344,
"loss": 2.8217,
"theoretical_loss": 3.3908951784729453,
"tokens_seen": 2302476288
},
{
"epoch": 0.7,
"learning_rate": 0.00015274824473420263,
"loss": 2.8207,
"theoretical_loss": 3.3908873523810943,
"tokens_seen": 2302541824
},
{
"epoch": 0.7,
"learning_rate": 0.0001527382146439318,
"loss": 2.7007,
"theoretical_loss": 3.3908795265743574,
"tokens_seen": 2302607360
},
{
"epoch": 0.7,
"learning_rate": 0.000152728184553661,
"loss": 2.8228,
"theoretical_loss": 3.390871701052717,
"tokens_seen": 2302672896
},
{
"epoch": 0.7,
"learning_rate": 0.0001527181544633902,
"loss": 2.8494,
"theoretical_loss": 3.3908638758161542,
"tokens_seen": 2302738432
},
{
"epoch": 0.7,
"learning_rate": 0.00015270812437311938,
"loss": 2.9229,
"theoretical_loss": 3.3908560508646497,
"tokens_seen": 2302803968
},
{
"epoch": 0.7,
"learning_rate": 0.00015269809428284856,
"loss": 2.8376,
"theoretical_loss": 3.3908482261981865,
"tokens_seen": 2302869504
},
{
"epoch": 0.7,
"learning_rate": 0.00015268806419257774,
"loss": 2.7469,
"theoretical_loss": 3.390840401816745,
"tokens_seen": 2302935040
},
{
"epoch": 0.7,
"learning_rate": 0.00015267803410230695,
"loss": 2.8172,
"theoretical_loss": 3.3908325777203077,
"tokens_seen": 2303000576
},
{
"epoch": 0.7,
"learning_rate": 0.00015266800401203613,
"loss": 3.0507,
"theoretical_loss": 3.3908247539088547,
"tokens_seen": 2303066112
},
{
"epoch": 0.7,
"learning_rate": 0.0001526579739217653,
"loss": 2.7507,
"theoretical_loss": 3.3908169303823685,
"tokens_seen": 2303131648
},
{
"epoch": 0.7,
"learning_rate": 0.0001526479438314945,
"loss": 2.9436,
"theoretical_loss": 3.390809107140831,
"tokens_seen": 2303197184
},
{
"epoch": 0.7,
"learning_rate": 0.00015263791374122367,
"loss": 2.8543,
"theoretical_loss": 3.3908012841842226,
"tokens_seen": 2303262720
},
{
"epoch": 0.7,
"learning_rate": 0.00015262788365095288,
"loss": 2.6726,
"theoretical_loss": 3.3907934615125255,
"tokens_seen": 2303328256
},
{
"epoch": 0.7,
"learning_rate": 0.00015261785356068206,
"loss": 2.7604,
"theoretical_loss": 3.390785639125721,
"tokens_seen": 2303393792
},
{
"epoch": 0.7,
"learning_rate": 0.00015260782347041124,
"loss": 2.7248,
"theoretical_loss": 3.3907778170237908,
"tokens_seen": 2303459328
},
{
"epoch": 0.7,
"learning_rate": 0.00015259779338014042,
"loss": 2.8559,
"theoretical_loss": 3.390769995206716,
"tokens_seen": 2303524864
},
{
"epoch": 0.7,
"objective/train/docs_used": 1298119,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6796865463256836,
"objective/train/theoretical_loss": 3.390762173674479,
"objective/train/tokens_used": 2324050400,
"theoretical_loss": 3.390762173674479,
"tokens_seen": 2303590400
},
{
"epoch": 0.7,
"learning_rate": 0.00015258776328986963,
"loss": 2.8942,
"theoretical_loss": 3.390762173674479,
"tokens_seen": 2303590400
},
{
"epoch": 0.7,
"learning_rate": 0.0001525777331995988,
"loss": 2.8376,
"theoretical_loss": 3.3907543524270602,
"tokens_seen": 2303655936
},
{
"epoch": 0.7,
"learning_rate": 0.000152567703109328,
"loss": 2.8303,
"theoretical_loss": 3.3907465314644414,
"tokens_seen": 2303721472
},
{
"epoch": 0.7,
"learning_rate": 0.00015255767301905717,
"loss": 2.8836,
"theoretical_loss": 3.3907387107866054,
"tokens_seen": 2303787008
},
{
"epoch": 0.7,
"learning_rate": 0.00015254764292878636,
"loss": 2.6534,
"theoretical_loss": 3.390730890393532,
"tokens_seen": 2303852544
},
{
"epoch": 0.7,
"learning_rate": 0.00015253761283851556,
"loss": 2.7487,
"theoretical_loss": 3.3907230702852034,
"tokens_seen": 2303918080
},
{
"epoch": 0.7,
"learning_rate": 0.00015252758274824475,
"loss": 2.8332,
"theoretical_loss": 3.3907152504616014,
"tokens_seen": 2303983616
},
{
"epoch": 0.7,
"learning_rate": 0.00015251755265797393,
"loss": 2.8272,
"theoretical_loss": 3.390707430922707,
"tokens_seen": 2304049152
},
{
"epoch": 0.7,
"learning_rate": 0.0001525075225677031,
"loss": 2.6812,
"theoretical_loss": 3.390699611668503,
"tokens_seen": 2304114688
},
{
"epoch": 0.7,
"learning_rate": 0.00015249749247743232,
"loss": 2.7374,
"theoretical_loss": 3.3906917926989695,
"tokens_seen": 2304180224
},
{
"epoch": 0.7,
"learning_rate": 0.0001524874623871615,
"loss": 2.9085,
"theoretical_loss": 3.390683974014088,
"tokens_seen": 2304245760
},
{
"epoch": 0.7,
"learning_rate": 0.00015247743229689068,
"loss": 2.7387,
"theoretical_loss": 3.3906761556138414,
"tokens_seen": 2304311296
},
{
"epoch": 0.7,
"learning_rate": 0.00015246740220661986,
"loss": 2.7303,
"theoretical_loss": 3.39066833749821,
"tokens_seen": 2304376832
},
{
"epoch": 0.7,
"learning_rate": 0.00015245737211634904,
"loss": 2.848,
"theoretical_loss": 3.390660519667176,
"tokens_seen": 2304442368
},
{
"epoch": 0.7,
"learning_rate": 0.00015244734202607825,
"loss": 2.789,
"theoretical_loss": 3.3906527021207205,
"tokens_seen": 2304507904
},
{
"epoch": 0.7,
"learning_rate": 0.00015243731193580743,
"loss": 2.8972,
"theoretical_loss": 3.3906448848588253,
"tokens_seen": 2304573440
},
{
"epoch": 0.7,
"learning_rate": 0.0001524272818455366,
"loss": 2.7773,
"theoretical_loss": 3.390637067881472,
"tokens_seen": 2304638976
},
{
"epoch": 0.7,
"learning_rate": 0.0001524172517552658,
"loss": 2.746,
"theoretical_loss": 3.390629251188642,
"tokens_seen": 2304704512
},
{
"epoch": 0.7,
"learning_rate": 0.000152407221664995,
"loss": 2.6556,
"theoretical_loss": 3.390621434780317,
"tokens_seen": 2304770048
},
{
"epoch": 0.7,
"learning_rate": 0.00015239719157472418,
"loss": 2.8135,
"theoretical_loss": 3.3906136186564786,
"tokens_seen": 2304835584
},
{
"epoch": 0.7,
"learning_rate": 0.00015238716148445336,
"loss": 2.9719,
"theoretical_loss": 3.3906058028171078,
"tokens_seen": 2304901120
},
{
"epoch": 0.7,
"learning_rate": 0.00015237713139418254,
"loss": 2.7483,
"theoretical_loss": 3.390597987262187,
"tokens_seen": 2304966656
},
{
"epoch": 0.7,
"learning_rate": 0.00015236710130391172,
"loss": 2.7779,
"theoretical_loss": 3.3905901719916973,
"tokens_seen": 2305032192
},
{
"epoch": 0.7,
"learning_rate": 0.00015235707121364093,
"loss": 2.9079,
"theoretical_loss": 3.3905823570056204,
"tokens_seen": 2305097728
},
{
"epoch": 0.7,
"learning_rate": 0.00015234704112337011,
"loss": 2.8897,
"theoretical_loss": 3.3905745423039377,
"tokens_seen": 2305163264
},
{
"epoch": 0.7,
"objective/train/docs_used": 1299212,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9192726612091064,
"objective/train/theoretical_loss": 3.3905667278866307,
"objective/train/tokens_used": 2325688800,
"theoretical_loss": 3.3905667278866307,
"tokens_seen": 2305228800
},
{
"epoch": 0.7,
"learning_rate": 0.0001523370110330993,
"loss": 2.7795,
"theoretical_loss": 3.3905667278866307,
"tokens_seen": 2305228800
},
{
"epoch": 0.7,
"learning_rate": 0.00015232698094282848,
"loss": 2.8723,
"theoretical_loss": 3.390558913753681,
"tokens_seen": 2305294336
},
{
"epoch": 0.7,
"learning_rate": 0.00015231695085255768,
"loss": 2.8289,
"theoretical_loss": 3.3905510999050708,
"tokens_seen": 2305359872
},
{
"epoch": 0.7,
"learning_rate": 0.00015230692076228687,
"loss": 2.7307,
"theoretical_loss": 3.3905432863407805,
"tokens_seen": 2305425408
},
{
"epoch": 0.7,
"learning_rate": 0.00015229689067201605,
"loss": 2.6849,
"theoretical_loss": 3.390535473060793,
"tokens_seen": 2305490944
},
{
"epoch": 0.7,
"learning_rate": 0.00015228686058174523,
"loss": 2.7097,
"theoretical_loss": 3.390527660065089,
"tokens_seen": 2305556480
},
{
"epoch": 0.7,
"learning_rate": 0.0001522768304914744,
"loss": 2.9281,
"theoretical_loss": 3.3905198473536498,
"tokens_seen": 2305622016
},
{
"epoch": 0.7,
"learning_rate": 0.00015226680040120362,
"loss": 2.7888,
"theoretical_loss": 3.3905120349264584,
"tokens_seen": 2305687552
},
{
"epoch": 0.7,
"learning_rate": 0.0001522567703109328,
"loss": 2.9242,
"theoretical_loss": 3.3905042227834947,
"tokens_seen": 2305753088
},
{
"epoch": 0.7,
"learning_rate": 0.00015224674022066198,
"loss": 2.8205,
"theoretical_loss": 3.390496410924741,
"tokens_seen": 2305818624
},
{
"epoch": 0.7,
"learning_rate": 0.00015223671013039116,
"loss": 2.7546,
"theoretical_loss": 3.390488599350179,
"tokens_seen": 2305884160
},
{
"epoch": 0.7,
"learning_rate": 0.00015222668004012037,
"loss": 2.5691,
"theoretical_loss": 3.3904807880597905,
"tokens_seen": 2305949696
},
{
"epoch": 0.7,
"learning_rate": 0.00015221664994984955,
"loss": 2.8155,
"theoretical_loss": 3.390472977053556,
"tokens_seen": 2306015232
},
{
"epoch": 0.7,
"learning_rate": 0.00015220661985957873,
"loss": 2.9052,
"theoretical_loss": 3.3904651663314587,
"tokens_seen": 2306080768
},
{
"epoch": 0.7,
"learning_rate": 0.0001521965897693079,
"loss": 3.0109,
"theoretical_loss": 3.390457355893479,
"tokens_seen": 2306146304
},
{
"epoch": 0.7,
"learning_rate": 0.0001521865596790371,
"loss": 2.7894,
"theoretical_loss": 3.390449545739599,
"tokens_seen": 2306211840
},
{
"epoch": 0.7,
"learning_rate": 0.0001521765295887663,
"loss": 2.6849,
"theoretical_loss": 3.3904417358698002,
"tokens_seen": 2306277376
},
{
"epoch": 0.7,
"learning_rate": 0.00015216649949849548,
"loss": 2.6875,
"theoretical_loss": 3.390433926284064,
"tokens_seen": 2306342912
},
{
"epoch": 0.7,
"learning_rate": 0.00015215646940822466,
"loss": 2.6722,
"theoretical_loss": 3.390426116982372,
"tokens_seen": 2306408448
},
{
"epoch": 0.7,
"learning_rate": 0.00015214643931795384,
"loss": 2.7556,
"theoretical_loss": 3.390418307964706,
"tokens_seen": 2306473984
},
{
"epoch": 0.7,
"learning_rate": 0.00015213640922768305,
"loss": 2.9347,
"theoretical_loss": 3.3904104992310478,
"tokens_seen": 2306539520
},
{
"epoch": 0.7,
"learning_rate": 0.00015212637913741223,
"loss": 2.7451,
"theoretical_loss": 3.3904026907813787,
"tokens_seen": 2306605056
},
{
"epoch": 0.7,
"learning_rate": 0.00015211634904714142,
"loss": 2.7769,
"theoretical_loss": 3.39039488261568,
"tokens_seen": 2306670592
},
{
"epoch": 0.7,
"learning_rate": 0.0001521063189568706,
"loss": 2.6271,
"theoretical_loss": 3.3903870747339333,
"tokens_seen": 2306736128
},
{
"epoch": 0.7,
"learning_rate": 0.0001520962888665998,
"loss": 2.6652,
"theoretical_loss": 3.3903792671361215,
"tokens_seen": 2306801664
},
{
"epoch": 0.7,
"objective/train/docs_used": 1299822,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.890716314315796,
"objective/train/theoretical_loss": 3.3903714598222243,
"objective/train/tokens_used": 2327327200,
"theoretical_loss": 3.3903714598222243,
"tokens_seen": 2306867200
},
{
"epoch": 0.7,
"learning_rate": 0.00015208625877632899,
"loss": 2.7883,
"theoretical_loss": 3.3903714598222243,
"tokens_seen": 2306867200
},
{
"epoch": 0.7,
"learning_rate": 0.00015207622868605817,
"loss": 2.9536,
"theoretical_loss": 3.390363652792225,
"tokens_seen": 2306932736
},
{
"epoch": 0.7,
"learning_rate": 0.00015206619859578735,
"loss": 2.9128,
"theoretical_loss": 3.390355846046104,
"tokens_seen": 2306998272
},
{
"epoch": 0.7,
"learning_rate": 0.00015205616850551653,
"loss": 2.8185,
"theoretical_loss": 3.3903480395838437,
"tokens_seen": 2307063808
},
{
"epoch": 0.7,
"learning_rate": 0.00015204613841524574,
"loss": 2.8292,
"theoretical_loss": 3.3903402334054253,
"tokens_seen": 2307129344
},
{
"epoch": 0.7,
"learning_rate": 0.00015203610832497492,
"loss": 2.8106,
"theoretical_loss": 3.39033242751083,
"tokens_seen": 2307194880
},
{
"epoch": 0.7,
"learning_rate": 0.00015202607823470413,
"loss": 2.7436,
"theoretical_loss": 3.390324621900041,
"tokens_seen": 2307260416
},
{
"epoch": 0.7,
"learning_rate": 0.0001520160481444333,
"loss": 2.7281,
"theoretical_loss": 3.390316816573038,
"tokens_seen": 2307325952
},
{
"epoch": 0.7,
"learning_rate": 0.00015200601805416252,
"loss": 2.7517,
"theoretical_loss": 3.390309011529804,
"tokens_seen": 2307391488
},
{
"epoch": 0.7,
"learning_rate": 0.0001519959879638917,
"loss": 2.9472,
"theoretical_loss": 3.39030120677032,
"tokens_seen": 2307457024
},
{
"epoch": 0.7,
"learning_rate": 0.00015198595787362088,
"loss": 2.6697,
"theoretical_loss": 3.390293402294567,
"tokens_seen": 2307522560
},
{
"epoch": 0.7,
"learning_rate": 0.00015197592778335006,
"loss": 2.8398,
"theoretical_loss": 3.3902855981025284,
"tokens_seen": 2307588096
},
{
"epoch": 0.7,
"learning_rate": 0.00015196589769307924,
"loss": 2.7429,
"theoretical_loss": 3.3902777941941844,
"tokens_seen": 2307653632
},
{
"epoch": 0.7,
"learning_rate": 0.00015195586760280845,
"loss": 2.6985,
"theoretical_loss": 3.390269990569517,
"tokens_seen": 2307719168
},
{
"epoch": 0.7,
"learning_rate": 0.00015194583751253763,
"loss": 2.7743,
"theoretical_loss": 3.3902621872285077,
"tokens_seen": 2307784704
},
{
"epoch": 0.7,
"learning_rate": 0.0001519358074222668,
"loss": 2.6672,
"theoretical_loss": 3.3902543841711386,
"tokens_seen": 2307850240
},
{
"epoch": 0.7,
"learning_rate": 0.000151925777331996,
"loss": 2.7313,
"theoretical_loss": 3.3902465813973905,
"tokens_seen": 2307915776
},
{
"epoch": 0.7,
"learning_rate": 0.0001519157472417252,
"loss": 2.785,
"theoretical_loss": 3.390238778907246,
"tokens_seen": 2307981312
},
{
"epoch": 0.7,
"learning_rate": 0.00015190571715145438,
"loss": 2.6458,
"theoretical_loss": 3.3902309767006864,
"tokens_seen": 2308046848
},
{
"epoch": 0.7,
"learning_rate": 0.00015189568706118356,
"loss": 2.9111,
"theoretical_loss": 3.390223174777693,
"tokens_seen": 2308112384
},
{
"epoch": 0.7,
"learning_rate": 0.00015188565697091274,
"loss": 2.6538,
"theoretical_loss": 3.390215373138248,
"tokens_seen": 2308177920
},
{
"epoch": 0.7,
"learning_rate": 0.00015187562688064192,
"loss": 2.851,
"theoretical_loss": 3.390207571782332,
"tokens_seen": 2308243456
},
{
"epoch": 0.7,
"learning_rate": 0.00015186559679037113,
"loss": 2.8492,
"theoretical_loss": 3.390199770709928,
"tokens_seen": 2308308992
},
{
"epoch": 0.7,
"learning_rate": 0.00015185556670010031,
"loss": 2.7934,
"theoretical_loss": 3.390191969921017,
"tokens_seen": 2308374528
},
{
"epoch": 0.7,
"learning_rate": 0.0001518455366098295,
"loss": 2.7932,
"theoretical_loss": 3.3901841694155803,
"tokens_seen": 2308440064
},
{
"epoch": 0.7,
"objective/train/docs_used": 1301383,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7817068099975586,
"objective/train/theoretical_loss": 3.3901763691936,
"objective/train/tokens_used": 2328965600,
"theoretical_loss": 3.3901763691936,
"tokens_seen": 2308505600
},
{
"epoch": 0.7,
"learning_rate": 0.00015183550651955868,
"loss": 2.8557,
"theoretical_loss": 3.3901763691936,
"tokens_seen": 2308505600
},
{
"epoch": 0.7,
"learning_rate": 0.00015182547642928788,
"loss": 2.9114,
"theoretical_loss": 3.390168569255058,
"tokens_seen": 2308571136
},
{
"epoch": 0.7,
"learning_rate": 0.00015181544633901707,
"loss": 2.4937,
"theoretical_loss": 3.3901607695999356,
"tokens_seen": 2308636672
},
{
"epoch": 0.7,
"learning_rate": 0.00015180541624874625,
"loss": 2.7778,
"theoretical_loss": 3.390152970228214,
"tokens_seen": 2308702208
},
{
"epoch": 0.7,
"learning_rate": 0.00015179538615847543,
"loss": 2.7489,
"theoretical_loss": 3.390145171139876,
"tokens_seen": 2308767744
},
{
"epoch": 0.7,
"learning_rate": 0.0001517853560682046,
"loss": 2.9741,
"theoretical_loss": 3.390137372334902,
"tokens_seen": 2308833280
},
{
"epoch": 0.7,
"learning_rate": 0.00015177532597793382,
"loss": 2.5403,
"theoretical_loss": 3.3901295738132746,
"tokens_seen": 2308898816
},
{
"epoch": 0.7,
"learning_rate": 0.000151765295887663,
"loss": 2.5911,
"theoretical_loss": 3.390121775574975,
"tokens_seen": 2308964352
},
{
"epoch": 0.7,
"learning_rate": 0.00015175526579739218,
"loss": 2.8601,
"theoretical_loss": 3.3901139776199853,
"tokens_seen": 2309029888
},
{
"epoch": 0.7,
"learning_rate": 0.00015174523570712136,
"loss": 2.6998,
"theoretical_loss": 3.390106179948287,
"tokens_seen": 2309095424
},
{
"epoch": 0.7,
"learning_rate": 0.00015173520561685057,
"loss": 2.703,
"theoretical_loss": 3.390098382559861,
"tokens_seen": 2309160960
},
{
"epoch": 0.7,
"learning_rate": 0.00015172517552657975,
"loss": 2.8365,
"theoretical_loss": 3.39009058545469,
"tokens_seen": 2309226496
},
{
"epoch": 0.7,
"learning_rate": 0.00015171514543630893,
"loss": 3.0203,
"theoretical_loss": 3.390082788632755,
"tokens_seen": 2309292032
},
{
"epoch": 0.7,
"learning_rate": 0.0001517051153460381,
"loss": 2.7461,
"theoretical_loss": 3.3900749920940383,
"tokens_seen": 2309357568
},
{
"epoch": 0.7,
"learning_rate": 0.0001516950852557673,
"loss": 3.0156,
"theoretical_loss": 3.3900671958385207,
"tokens_seen": 2309423104
},
{
"epoch": 0.7,
"learning_rate": 0.0001516850551654965,
"loss": 2.8159,
"theoretical_loss": 3.390059399866185,
"tokens_seen": 2309488640
},
{
"epoch": 0.7,
"learning_rate": 0.00015167502507522568,
"loss": 2.881,
"theoretical_loss": 3.390051604177012,
"tokens_seen": 2309554176
},
{
"epoch": 0.7,
"learning_rate": 0.00015166499498495486,
"loss": 2.8883,
"theoretical_loss": 3.3900438087709834,
"tokens_seen": 2309619712
},
{
"epoch": 0.7,
"learning_rate": 0.00015165496489468404,
"loss": 2.8767,
"theoretical_loss": 3.3900360136480816,
"tokens_seen": 2309685248
},
{
"epoch": 0.7,
"learning_rate": 0.00015164493480441325,
"loss": 2.8674,
"theoretical_loss": 3.3900282188082875,
"tokens_seen": 2309750784
},
{
"epoch": 0.7,
"learning_rate": 0.00015163490471414243,
"loss": 2.8861,
"theoretical_loss": 3.3900204242515835,
"tokens_seen": 2309816320
},
{
"epoch": 0.7,
"learning_rate": 0.00015162487462387162,
"loss": 2.5879,
"theoretical_loss": 3.3900126299779503,
"tokens_seen": 2309881856
},
{
"epoch": 0.7,
"learning_rate": 0.0001516148445336008,
"loss": 2.8589,
"theoretical_loss": 3.3900048359873707,
"tokens_seen": 2309947392
},
{
"epoch": 0.7,
"learning_rate": 0.00015160481444333,
"loss": 2.746,
"theoretical_loss": 3.3899970422798256,
"tokens_seen": 2310012928
},
{
"epoch": 0.7,
"learning_rate": 0.00015159478435305919,
"loss": 2.6061,
"theoretical_loss": 3.389989248855297,
"tokens_seen": 2310078464
},
{
"epoch": 0.7,
"objective/train/docs_used": 1301708,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.9675357341766357,
"objective/train/theoretical_loss": 3.389981455713767,
"objective/train/tokens_used": 2330604000,
"theoretical_loss": 3.389981455713767,
"tokens_seen": 2310144000
},
{
"epoch": 0.7,
"learning_rate": 0.00015158475426278837,
"loss": 2.8403,
"theoretical_loss": 3.389981455713767,
"tokens_seen": 2310144000
},
{
"epoch": 0.7,
"learning_rate": 0.00015157472417251755,
"loss": 2.8445,
"theoretical_loss": 3.3899736628552164,
"tokens_seen": 2310209536
},
{
"epoch": 0.7,
"learning_rate": 0.00015156469408224673,
"loss": 2.4856,
"theoretical_loss": 3.389965870279627,
"tokens_seen": 2310275072
},
{
"epoch": 0.7,
"learning_rate": 0.00015155466399197594,
"loss": 2.876,
"theoretical_loss": 3.389958077986982,
"tokens_seen": 2310340608
},
{
"epoch": 0.7,
"learning_rate": 0.00015154463390170512,
"loss": 2.7771,
"theoretical_loss": 3.3899502859772612,
"tokens_seen": 2310406144
},
{
"epoch": 0.7,
"learning_rate": 0.0001515346038114343,
"loss": 2.7798,
"theoretical_loss": 3.389942494250447,
"tokens_seen": 2310471680
},
{
"epoch": 0.7,
"learning_rate": 0.00015152457372116348,
"loss": 2.8172,
"theoretical_loss": 3.389934702806521,
"tokens_seen": 2310537216
},
{
"epoch": 0.7,
"learning_rate": 0.0001515145436308927,
"loss": 2.7067,
"theoretical_loss": 3.389926911645466,
"tokens_seen": 2310602752
},
{
"epoch": 0.7,
"learning_rate": 0.00015150451354062187,
"loss": 2.8379,
"theoretical_loss": 3.389919120767262,
"tokens_seen": 2310668288
},
{
"epoch": 0.7,
"learning_rate": 0.00015149448345035105,
"loss": 2.762,
"theoretical_loss": 3.3899113301718913,
"tokens_seen": 2310733824
},
{
"epoch": 0.7,
"learning_rate": 0.00015148445336008023,
"loss": 2.7784,
"theoretical_loss": 3.3899035398593362,
"tokens_seen": 2310799360
},
{
"epoch": 0.7,
"learning_rate": 0.0001514744232698094,
"loss": 2.8784,
"theoretical_loss": 3.389895749829578,
"tokens_seen": 2310864896
},
{
"epoch": 0.7,
"learning_rate": 0.00015146439317953862,
"loss": 2.7945,
"theoretical_loss": 3.389887960082598,
"tokens_seen": 2310930432
},
{
"epoch": 0.7,
"learning_rate": 0.0001514543630892678,
"loss": 2.8487,
"theoretical_loss": 3.3898801706183788,
"tokens_seen": 2310995968
},
{
"epoch": 0.7,
"learning_rate": 0.00015144433299899698,
"loss": 2.9228,
"theoretical_loss": 3.3898723814369016,
"tokens_seen": 2311061504
},
{
"epoch": 0.7,
"learning_rate": 0.00015143430290872617,
"loss": 2.7585,
"theoretical_loss": 3.389864592538148,
"tokens_seen": 2311127040
},
{
"epoch": 0.7,
"learning_rate": 0.00015142427281845537,
"loss": 2.7243,
"theoretical_loss": 3.3898568039221004,
"tokens_seen": 2311192576
},
{
"epoch": 0.7,
"learning_rate": 0.00015141424272818455,
"loss": 2.9123,
"theoretical_loss": 3.3898490155887395,
"tokens_seen": 2311258112
},
{
"epoch": 0.7,
"learning_rate": 0.00015140421263791374,
"loss": 2.8025,
"theoretical_loss": 3.3898412275380476,
"tokens_seen": 2311323648
},
{
"epoch": 0.7,
"learning_rate": 0.00015139418254764292,
"loss": 2.8009,
"theoretical_loss": 3.3898334397700065,
"tokens_seen": 2311389184
},
{
"epoch": 0.7,
"learning_rate": 0.0001513841524573721,
"loss": 2.8958,
"theoretical_loss": 3.389825652284598,
"tokens_seen": 2311454720
},
{
"epoch": 0.7,
"learning_rate": 0.0001513741223671013,
"loss": 2.8863,
"theoretical_loss": 3.3898178650818034,
"tokens_seen": 2311520256
},
{
"epoch": 0.7,
"learning_rate": 0.0001513640922768305,
"loss": 2.9089,
"theoretical_loss": 3.389810078161605,
"tokens_seen": 2311585792
},
{
"epoch": 0.7,
"learning_rate": 0.0001513640922768305,
"loss": 2.6727,
"theoretical_loss": 3.3898022915239836,
"tokens_seen": 2311651328
},
{
"epoch": 0.7,
"learning_rate": 0.00015135406218655967,
"loss": 2.8618,
"theoretical_loss": 3.389794505168922,
"tokens_seen": 2311716864
},
{
"epoch": 0.7,
"objective/train/docs_used": 1302992,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.8403515815734863,
"objective/train/theoretical_loss": 3.389786719096401,
"objective/train/tokens_used": 2332242400,
"theoretical_loss": 3.389786719096401,
"tokens_seen": 2311782400
},
{
"epoch": 0.7,
"learning_rate": 0.00015134403209628885,
"loss": 2.7295,
"theoretical_loss": 3.389786719096401,
"tokens_seen": 2311782400
},
{
"epoch": 0.7,
"learning_rate": 0.00015133400200601806,
"loss": 2.8005,
"theoretical_loss": 3.389778933306403,
"tokens_seen": 2311847936
},
{
"epoch": 0.7,
"learning_rate": 0.00015132397191574724,
"loss": 2.7881,
"theoretical_loss": 3.3897711477989096,
"tokens_seen": 2311913472
},
{
"epoch": 0.7,
"learning_rate": 0.00015131394182547642,
"loss": 2.906,
"theoretical_loss": 3.3897633625739028,
"tokens_seen": 2311979008
},
{
"epoch": 0.7,
"learning_rate": 0.0001513039117352056,
"loss": 2.771,
"theoretical_loss": 3.3897555776313637,
"tokens_seen": 2312044544
},
{
"epoch": 0.7,
"learning_rate": 0.00015129388164493478,
"loss": 2.847,
"theoretical_loss": 3.3897477929712743,
"tokens_seen": 2312110080
},
{
"epoch": 0.7,
"learning_rate": 0.000151283851554664,
"loss": 2.8661,
"theoretical_loss": 3.3897400085936167,
"tokens_seen": 2312175616
},
{
"epoch": 0.7,
"learning_rate": 0.0001512738214643932,
"loss": 2.6742,
"theoretical_loss": 3.3897322244983723,
"tokens_seen": 2312241152
},
{
"epoch": 0.7,
"learning_rate": 0.00015126379137412238,
"loss": 3.0103,
"theoretical_loss": 3.389724440685523,
"tokens_seen": 2312306688
},
{
"epoch": 0.7,
"learning_rate": 0.00015125376128385156,
"loss": 2.6936,
"theoretical_loss": 3.3897166571550503,
"tokens_seen": 2312372224
},
{
"epoch": 0.7,
"learning_rate": 0.00015124373119358077,
"loss": 2.6984,
"theoretical_loss": 3.389708873906936,
"tokens_seen": 2312437760
},
{
"epoch": 0.7,
"learning_rate": 0.00015123370110330995,
"loss": 2.7324,
"theoretical_loss": 3.3897010909411622,
"tokens_seen": 2312503296
},
{
"epoch": 0.7,
"learning_rate": 0.00015122367101303913,
"loss": 2.9167,
"theoretical_loss": 3.3896933082577103,
"tokens_seen": 2312568832
},
{
"epoch": 0.7,
"learning_rate": 0.0001512136409227683,
"loss": 2.7821,
"theoretical_loss": 3.3896855258565624,
"tokens_seen": 2312634368
},
{
"epoch": 0.7,
"learning_rate": 0.0001512036108324975,
"loss": 2.5027,
"theoretical_loss": 3.3896777437376997,
"tokens_seen": 2312699904
},
{
"epoch": 0.7,
"learning_rate": 0.0001511935807422267,
"loss": 2.7627,
"theoretical_loss": 3.3896699619011046,
"tokens_seen": 2312765440
},
{
"epoch": 0.7,
"learning_rate": 0.00015118355065195588,
"loss": 2.494,
"theoretical_loss": 3.389662180346759,
"tokens_seen": 2312830976
},
{
"epoch": 0.7,
"learning_rate": 0.00015117352056168506,
"loss": 2.6384,
"theoretical_loss": 3.3896543990746437,
"tokens_seen": 2312896512
},
{
"epoch": 0.7,
"learning_rate": 0.00015116349047141425,
"loss": 2.696,
"theoretical_loss": 3.3896466180847407,
"tokens_seen": 2312962048
},
{
"epoch": 0.7,
"learning_rate": 0.00015115346038114345,
"loss": 2.6502,
"theoretical_loss": 3.3896388373770328,
"tokens_seen": 2313027584
},
{
"epoch": 0.7,
"learning_rate": 0.00015114343029087263,
"loss": 2.9052,
"theoretical_loss": 3.389631056951501,
"tokens_seen": 2313093120
},
{
"epoch": 0.7,
"learning_rate": 0.00015113340020060182,
"loss": 2.8058,
"theoretical_loss": 3.389623276808127,
"tokens_seen": 2313158656
},
{
"epoch": 0.7,
"learning_rate": 0.000151123370110331,
"loss": 2.7504,
"theoretical_loss": 3.389615496946892,
"tokens_seen": 2313224192
},
{
"epoch": 0.7,
"learning_rate": 0.0001511133400200602,
"loss": 2.923,
"theoretical_loss": 3.389607717367779,
"tokens_seen": 2313289728
},
{
"epoch": 0.7,
"learning_rate": 0.00015110330992978939,
"loss": 2.8558,
"theoretical_loss": 3.3895999380707695,
"tokens_seen": 2313355264
},
{
"epoch": 0.7,
"objective/train/docs_used": 1303654,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4339420795440674,
"objective/train/theoretical_loss": 3.389592159055845,
"objective/train/tokens_used": 2333880800,
"theoretical_loss": 3.389592159055845,
"tokens_seen": 2313420800
},
{
"epoch": 0.7,
"learning_rate": 0.00015109327983951857,
"loss": 2.6258,
"theoretical_loss": 3.389592159055845,
"tokens_seen": 2313420800
},
{
"epoch": 0.7,
"learning_rate": 0.00015108324974924775,
"loss": 2.7654,
"theoretical_loss": 3.389584380322987,
"tokens_seen": 2313486336
},
{
"epoch": 0.7,
"learning_rate": 0.00015107321965897693,
"loss": 2.9493,
"theoretical_loss": 3.389576601872178,
"tokens_seen": 2313551872
},
{
"epoch": 0.7,
"learning_rate": 0.00015106318956870614,
"loss": 2.6432,
"theoretical_loss": 3.389568823703399,
"tokens_seen": 2313617408
},
{
"epoch": 0.7,
"learning_rate": 0.00015105315947843532,
"loss": 2.5808,
"theoretical_loss": 3.3895610458166323,
"tokens_seen": 2313682944
},
{
"epoch": 0.7,
"learning_rate": 0.0001510431293881645,
"loss": 2.7568,
"theoretical_loss": 3.3895532682118596,
"tokens_seen": 2313748480
},
{
"epoch": 0.7,
"learning_rate": 0.00015103309929789368,
"loss": 2.7581,
"theoretical_loss": 3.3895454908890628,
"tokens_seen": 2313814016
},
{
"epoch": 0.7,
"learning_rate": 0.0001510230692076229,
"loss": 2.7539,
"theoretical_loss": 3.389537713848223,
"tokens_seen": 2313879552
},
{
"epoch": 0.7,
"learning_rate": 0.00015101303911735207,
"loss": 2.6652,
"theoretical_loss": 3.389529937089323,
"tokens_seen": 2313945088
},
{
"epoch": 0.7,
"learning_rate": 0.00015100300902708125,
"loss": 2.9207,
"theoretical_loss": 3.3895221606123442,
"tokens_seen": 2314010624
},
{
"epoch": 0.7,
"learning_rate": 0.00015099297893681043,
"loss": 2.6359,
"theoretical_loss": 3.389514384417268,
"tokens_seen": 2314076160
},
{
"epoch": 0.7,
"learning_rate": 0.00015098294884653961,
"loss": 2.9392,
"theoretical_loss": 3.389506608504077,
"tokens_seen": 2314141696
},
{
"epoch": 0.7,
"learning_rate": 0.00015097291875626882,
"loss": 2.7996,
"theoretical_loss": 3.3894988328727518,
"tokens_seen": 2314207232
},
{
"epoch": 0.7,
"learning_rate": 0.000150962888665998,
"loss": 2.7658,
"theoretical_loss": 3.3894910575232755,
"tokens_seen": 2314272768
},
{
"epoch": 0.7,
"learning_rate": 0.00015095285857572718,
"loss": 2.8016,
"theoretical_loss": 3.389483282455629,
"tokens_seen": 2314338304
},
{
"epoch": 0.7,
"learning_rate": 0.00015094282848545637,
"loss": 2.831,
"theoretical_loss": 3.389475507669794,
"tokens_seen": 2314403840
},
{
"epoch": 0.7,
"learning_rate": 0.00015093279839518557,
"loss": 2.6282,
"theoretical_loss": 3.3894677331657537,
"tokens_seen": 2314469376
},
{
"epoch": 0.7,
"learning_rate": 0.00015092276830491475,
"loss": 2.7907,
"theoretical_loss": 3.3894599589434886,
"tokens_seen": 2314534912
},
{
"epoch": 0.7,
"learning_rate": 0.00015091273821464394,
"loss": 2.6954,
"theoretical_loss": 3.3894521850029804,
"tokens_seen": 2314600448
},
{
"epoch": 0.7,
"learning_rate": 0.00015090270812437312,
"loss": 2.8794,
"theoretical_loss": 3.3894444113442117,
"tokens_seen": 2314665984
},
{
"epoch": 0.7,
"learning_rate": 0.0001508926780341023,
"loss": 2.7874,
"theoretical_loss": 3.389436637967164,
"tokens_seen": 2314731520
},
{
"epoch": 0.7,
"learning_rate": 0.0001508826479438315,
"loss": 2.7137,
"theoretical_loss": 3.389428864871819,
"tokens_seen": 2314797056
},
{
"epoch": 0.7,
"learning_rate": 0.0001508726178535607,
"loss": 2.5175,
"theoretical_loss": 3.389421092058159,
"tokens_seen": 2314862592
},
{
"epoch": 0.7,
"learning_rate": 0.00015086258776328987,
"loss": 2.7887,
"theoretical_loss": 3.389413319526165,
"tokens_seen": 2314928128
},
{
"epoch": 0.7,
"learning_rate": 0.00015085255767301905,
"loss": 2.837,
"theoretical_loss": 3.389405547275819,
"tokens_seen": 2314993664
},
{
"epoch": 0.7,
"objective/train/docs_used": 1304892,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7171783447265625,
"objective/train/theoretical_loss": 3.3893977753071036,
"objective/train/tokens_used": 2335519200,
"theoretical_loss": 3.3893977753071036,
"tokens_seen": 2315059200
},
{
"epoch": 0.7,
"learning_rate": 0.00015084252758274826,
"loss": 2.8092,
"theoretical_loss": 3.3893977753071036,
"tokens_seen": 2315059200
},
{
"epoch": 0.7,
"learning_rate": 0.00015083249749247744,
"loss": 2.7669,
"theoretical_loss": 3.3893900036199995,
"tokens_seen": 2315124736
},
{
"epoch": 0.7,
"learning_rate": 0.00015082246740220662,
"loss": 2.7962,
"theoretical_loss": 3.3893822322144898,
"tokens_seen": 2315190272
},
{
"epoch": 0.7,
"learning_rate": 0.0001508124373119358,
"loss": 2.672,
"theoretical_loss": 3.389374461090555,
"tokens_seen": 2315255808
},
{
"epoch": 0.7,
"learning_rate": 0.00015080240722166498,
"loss": 2.7185,
"theoretical_loss": 3.3893666902481776,
"tokens_seen": 2315321344
},
{
"epoch": 0.7,
"learning_rate": 0.0001507923771313942,
"loss": 2.8075,
"theoretical_loss": 3.38935891968734,
"tokens_seen": 2315386880
},
{
"epoch": 0.7,
"learning_rate": 0.00015078234704112337,
"loss": 2.6874,
"theoretical_loss": 3.389351149408023,
"tokens_seen": 2315452416
},
{
"epoch": 0.7,
"learning_rate": 0.00015077231695085255,
"loss": 2.8724,
"theoretical_loss": 3.389343379410209,
"tokens_seen": 2315517952
},
{
"epoch": 0.7,
"learning_rate": 0.00015076228686058173,
"loss": 2.8493,
"theoretical_loss": 3.3893356096938794,
"tokens_seen": 2315583488
},
{
"epoch": 0.7,
"learning_rate": 0.00015075225677031094,
"loss": 2.7059,
"theoretical_loss": 3.389327840259017,
"tokens_seen": 2315649024
},
{
"epoch": 0.7,
"learning_rate": 0.00015074222668004012,
"loss": 2.6791,
"theoretical_loss": 3.389320071105602,
"tokens_seen": 2315714560
},
{
"epoch": 0.7,
"learning_rate": 0.0001507321965897693,
"loss": 2.6129,
"theoretical_loss": 3.3893123022336176,
"tokens_seen": 2315780096
},
{
"epoch": 0.7,
"learning_rate": 0.00015072216649949849,
"loss": 2.8105,
"theoretical_loss": 3.3893045336430454,
"tokens_seen": 2315845632
},
{
"epoch": 0.7,
"learning_rate": 0.00015071213640922767,
"loss": 2.8395,
"theoretical_loss": 3.389296765333867,
"tokens_seen": 2315911168
},
{
"epoch": 0.7,
"learning_rate": 0.00015070210631895687,
"loss": 2.775,
"theoretical_loss": 3.3892889973060645,
"tokens_seen": 2315976704
},
{
"epoch": 0.7,
"learning_rate": 0.00015069207622868606,
"loss": 2.7474,
"theoretical_loss": 3.3892812295596193,
"tokens_seen": 2316042240
},
{
"epoch": 0.7,
"learning_rate": 0.00015068204613841524,
"loss": 2.7186,
"theoretical_loss": 3.3892734620945135,
"tokens_seen": 2316107776
},
{
"epoch": 0.7,
"learning_rate": 0.00015067201604814442,
"loss": 2.6161,
"theoretical_loss": 3.3892656949107294,
"tokens_seen": 2316173312
},
{
"epoch": 0.7,
"learning_rate": 0.00015066198595787363,
"loss": 2.9049,
"theoretical_loss": 3.3892579280082478,
"tokens_seen": 2316238848
},
{
"epoch": 0.7,
"learning_rate": 0.0001506519558676028,
"loss": 2.9633,
"theoretical_loss": 3.389250161387052,
"tokens_seen": 2316304384
},
{
"epoch": 0.7,
"learning_rate": 0.000150641925777332,
"loss": 2.5955,
"theoretical_loss": 3.3892423950471224,
"tokens_seen": 2316369920
},
{
"epoch": 0.7,
"learning_rate": 0.00015063189568706117,
"loss": 2.7058,
"theoretical_loss": 3.389234628988442,
"tokens_seen": 2316435456
},
{
"epoch": 0.7,
"learning_rate": 0.00015062186559679035,
"loss": 2.7482,
"theoretical_loss": 3.3892268632109914,
"tokens_seen": 2316500992
},
{
"epoch": 0.7,
"learning_rate": 0.00015061183550651956,
"loss": 2.7869,
"theoretical_loss": 3.389219097714754,
"tokens_seen": 2316566528
},
{
"epoch": 0.7,
"learning_rate": 0.00015060180541624874,
"loss": 2.7607,
"theoretical_loss": 3.38921133249971,
"tokens_seen": 2316632064
},
{
"epoch": 0.7,
"objective/train/docs_used": 1305637,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4803452491760254,
"objective/train/theoretical_loss": 3.389203567565843,
"objective/train/tokens_used": 2337157600,
"theoretical_loss": 3.389203567565843,
"tokens_seen": 2316697600
},
{
"epoch": 0.7,
"learning_rate": 0.00015059177532597792,
"loss": 2.5511,
"theoretical_loss": 3.389203567565843,
"tokens_seen": 2316697600
},
{
"epoch": 0.7,
"learning_rate": 0.0001505817452357071,
"loss": 2.6416,
"theoretical_loss": 3.3891958029131333,
"tokens_seen": 2316763136
},
{
"epoch": 0.7,
"learning_rate": 0.0001505717151454363,
"loss": 2.6929,
"theoretical_loss": 3.389188038541564,
"tokens_seen": 2316828672
},
{
"epoch": 0.7,
"learning_rate": 0.0001505616850551655,
"loss": 2.5269,
"theoretical_loss": 3.3891802744511166,
"tokens_seen": 2316894208
},
{
"epoch": 0.7,
"learning_rate": 0.00015055165496489467,
"loss": 2.4093,
"theoretical_loss": 3.3891725106417723,
"tokens_seen": 2316959744
},
{
"epoch": 0.7,
"learning_rate": 0.00015054162487462385,
"loss": 2.6467,
"theoretical_loss": 3.3891647471135133,
"tokens_seen": 2317025280
},
{
"epoch": 0.7,
"learning_rate": 0.00015053159478435304,
"loss": 2.9476,
"theoretical_loss": 3.3891569838663225,
"tokens_seen": 2317090816
},
{
"epoch": 0.7,
"learning_rate": 0.00015052156469408227,
"loss": 2.5233,
"theoretical_loss": 3.38914922090018,
"tokens_seen": 2317156352
},
{
"epoch": 0.7,
"learning_rate": 0.00015051153460381145,
"loss": 2.6592,
"theoretical_loss": 3.389141458215069,
"tokens_seen": 2317221888
},
{
"epoch": 0.7,
"learning_rate": 0.00015050150451354063,
"loss": 2.5042,
"theoretical_loss": 3.3891336958109712,
"tokens_seen": 2317287424
},
{
"epoch": 0.7,
"learning_rate": 0.00015049147442326981,
"loss": 2.7221,
"theoretical_loss": 3.3891259336878683,
"tokens_seen": 2317352960
},
{
"epoch": 0.7,
"learning_rate": 0.00015048144433299902,
"loss": 2.4874,
"theoretical_loss": 3.389118171845742,
"tokens_seen": 2317418496
},
{
"epoch": 0.7,
"learning_rate": 0.0001504714142427282,
"loss": 2.4247,
"theoretical_loss": 3.389110410284574,
"tokens_seen": 2317484032
},
{
"epoch": 0.7,
"learning_rate": 0.00015046138415245738,
"loss": 2.5496,
"theoretical_loss": 3.3891026490043465,
"tokens_seen": 2317549568
},
{
"epoch": 0.7,
"learning_rate": 0.00015045135406218657,
"loss": 2.6206,
"theoretical_loss": 3.389094888005042,
"tokens_seen": 2317615104
},
{
"epoch": 0.7,
"learning_rate": 0.00015044132397191577,
"loss": 2.6191,
"theoretical_loss": 3.3890871272866416,
"tokens_seen": 2317680640
},
{
"epoch": 0.7,
"learning_rate": 0.00015043129388164495,
"loss": 2.5785,
"theoretical_loss": 3.389079366849127,
"tokens_seen": 2317746176
},
{
"epoch": 0.7,
"learning_rate": 0.00015042126379137414,
"loss": 2.5723,
"theoretical_loss": 3.3890716066924806,
"tokens_seen": 2317811712
},
{
"epoch": 0.7,
"learning_rate": 0.00015041123370110332,
"loss": 2.4477,
"theoretical_loss": 3.3890638468166845,
"tokens_seen": 2317877248
},
{
"epoch": 0.7,
"learning_rate": 0.0001504012036108325,
"loss": 2.5637,
"theoretical_loss": 3.38905608722172,
"tokens_seen": 2317942784
},
{
"epoch": 0.7,
"learning_rate": 0.0001503911735205617,
"loss": 2.8322,
"theoretical_loss": 3.3890483279075694,
"tokens_seen": 2318008320
},
{
"epoch": 0.7,
"learning_rate": 0.0001503811434302909,
"loss": 2.628,
"theoretical_loss": 3.3890405688742145,
"tokens_seen": 2318073856
},
{
"epoch": 0.7,
"learning_rate": 0.00015037111334002007,
"loss": 2.5576,
"theoretical_loss": 3.3890328101216367,
"tokens_seen": 2318139392
},
{
"epoch": 0.7,
"learning_rate": 0.00015036108324974925,
"loss": 2.5031,
"theoretical_loss": 3.389025051649819,
"tokens_seen": 2318204928
},
{
"epoch": 0.7,
"learning_rate": 0.00015035105315947846,
"loss": 2.6636,
"theoretical_loss": 3.389017293458742,
"tokens_seen": 2318270464
},
{
"epoch": 0.7,
"objective/train/docs_used": 1306410,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.4640891551971436,
"objective/train/theoretical_loss": 3.389009535548389,
"objective/train/tokens_used": 2338796000,
"theoretical_loss": 3.389009535548389,
"tokens_seen": 2318336000
},
{
"epoch": 0.7,
"learning_rate": 0.00015034102306920764,
"loss": 2.5322,
"theoretical_loss": 3.389009535548389,
"tokens_seen": 2318336000
},
{
"epoch": 0.7,
"learning_rate": 0.00015033099297893682,
"loss": 2.8184,
"theoretical_loss": 3.3890017779187405,
"tokens_seen": 2318401536
},
{
"epoch": 0.7,
"learning_rate": 0.000150320962888666,
"loss": 2.7075,
"theoretical_loss": 3.3889940205697795,
"tokens_seen": 2318467072
},
{
"epoch": 0.7,
"learning_rate": 0.00015031093279839518,
"loss": 2.6677,
"theoretical_loss": 3.3889862635014874,
"tokens_seen": 2318532608
},
{
"epoch": 0.7,
"learning_rate": 0.0001503009027081244,
"loss": 2.5963,
"theoretical_loss": 3.3889785067138463,
"tokens_seen": 2318598144
},
{
"epoch": 0.7,
"learning_rate": 0.00015029087261785357,
"loss": 2.6575,
"theoretical_loss": 3.3889707502068376,
"tokens_seen": 2318663680
},
{
"epoch": 0.7,
"learning_rate": 0.00015028084252758275,
"loss": 2.847,
"theoretical_loss": 3.388962993980444,
"tokens_seen": 2318729216
},
{
"epoch": 0.7,
"learning_rate": 0.00015027081243731193,
"loss": 2.6007,
"theoretical_loss": 3.3889552380346473,
"tokens_seen": 2318794752
},
{
"epoch": 0.7,
"learning_rate": 0.00015026078234704114,
"loss": 2.5342,
"theoretical_loss": 3.3889474823694288,
"tokens_seen": 2318860288
},
{
"epoch": 0.7,
"learning_rate": 0.00015025075225677032,
"loss": 2.5773,
"theoretical_loss": 3.3889397269847708,
"tokens_seen": 2318925824
},
{
"epoch": 0.7,
"learning_rate": 0.0001502407221664995,
"loss": 2.6941,
"theoretical_loss": 3.3889319718806554,
"tokens_seen": 2318991360
},
{
"epoch": 0.7,
"learning_rate": 0.00015023069207622869,
"loss": 2.4312,
"theoretical_loss": 3.388924217057064,
"tokens_seen": 2319056896
},
{
"epoch": 0.7,
"learning_rate": 0.00015022066198595787,
"loss": 2.7074,
"theoretical_loss": 3.388916462513979,
"tokens_seen": 2319122432
},
{
"epoch": 0.7,
"learning_rate": 0.00015021063189568707,
"loss": 2.4902,
"theoretical_loss": 3.3889087082513827,
"tokens_seen": 2319187968
},
{
"epoch": 0.7,
"learning_rate": 0.00015020060180541626,
"loss": 2.7004,
"theoretical_loss": 3.388900954269256,
"tokens_seen": 2319253504
},
{
"epoch": 0.7,
"learning_rate": 0.00015019057171514544,
"loss": 2.7672,
"theoretical_loss": 3.388893200567581,
"tokens_seen": 2319319040
},
{
"epoch": 0.7,
"learning_rate": 0.00015018054162487462,
"loss": 2.6768,
"theoretical_loss": 3.3888854471463405,
"tokens_seen": 2319384576
},
{
"epoch": 0.7,
"learning_rate": 0.00015017051153460383,
"loss": 2.7587,
"theoretical_loss": 3.388877694005516,
"tokens_seen": 2319450112
},
{
"epoch": 0.7,
"learning_rate": 0.000150160481444333,
"loss": 2.749,
"theoretical_loss": 3.388869941145089,
"tokens_seen": 2319515648
},
{
"epoch": 0.7,
"learning_rate": 0.0001501504513540622,
"loss": 2.6993,
"theoretical_loss": 3.3888621885650423,
"tokens_seen": 2319581184
},
{
"epoch": 0.7,
"learning_rate": 0.00015014042126379137,
"loss": 2.8766,
"theoretical_loss": 3.388854436265357,
"tokens_seen": 2319646720
},
{
"epoch": 0.7,
"learning_rate": 0.00015013039117352055,
"loss": 2.8515,
"theoretical_loss": 3.3888466842460154,
"tokens_seen": 2319712256
},
{
"epoch": 0.7,
"learning_rate": 0.00015012036108324976,
"loss": 2.7364,
"theoretical_loss": 3.3888389325069994,
"tokens_seen": 2319777792
},
{
"epoch": 0.7,
"learning_rate": 0.00015011033099297894,
"loss": 2.6966,
"theoretical_loss": 3.3888311810482907,
"tokens_seen": 2319843328
},
{
"epoch": 0.7,
"learning_rate": 0.00015010030090270812,
"loss": 2.7712,
"theoretical_loss": 3.388823429869872,
"tokens_seen": 2319908864
},
{
"epoch": 0.7,
"objective/train/docs_used": 1307847,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.7233824729919434,
"objective/train/theoretical_loss": 3.3888156789717243,
"objective/train/tokens_used": 2340434400,
"theoretical_loss": 3.3888156789717243,
"tokens_seen": 2319974400
},
{
"epoch": 0.7,
"learning_rate": 0.0001500902708124373,
"loss": 2.6795,
"theoretical_loss": 3.3888156789717243,
"tokens_seen": 2319974400
},
{
"epoch": 0.7,
"learning_rate": 0.0001500802407221665,
"loss": 2.4224,
"theoretical_loss": 3.38880792835383,
"tokens_seen": 2320039936
},
{
"epoch": 0.7,
"learning_rate": 0.0001500702106318957,
"loss": 2.4721,
"theoretical_loss": 3.3888001780161714,
"tokens_seen": 2320105472
},
{
"epoch": 0.7,
"learning_rate": 0.00015006018054162487,
"loss": 2.4984,
"theoretical_loss": 3.3887924279587303,
"tokens_seen": 2320171008
},
{
"epoch": 0.7,
"learning_rate": 0.00015005015045135405,
"loss": 2.5876,
"theoretical_loss": 3.388784678181488,
"tokens_seen": 2320236544
},
{
"epoch": 0.7,
"learning_rate": 0.00015004012036108324,
"loss": 2.7932,
"theoretical_loss": 3.388776928684427,
"tokens_seen": 2320302080
},
{
"epoch": 0.7,
"learning_rate": 0.00015003009027081244,
"loss": 2.6231,
"theoretical_loss": 3.388769179467529,
"tokens_seen": 2320367616
},
{
"epoch": 0.7,
"learning_rate": 0.00015002006018054162,
"loss": 2.501,
"theoretical_loss": 3.3887614305307765,
"tokens_seen": 2320433152
},
{
"epoch": 0.7,
"learning_rate": 0.0001500100300902708,
"loss": 2.6844,
"theoretical_loss": 3.388753681874151,
"tokens_seen": 2320498688
},
{
"epoch": 0.7,
"learning_rate": 0.00015,
"loss": 2.6135,
"theoretical_loss": 3.3887459334976344,
"tokens_seen": 2320564224
},
{
"epoch": 0.7,
"learning_rate": 0.0001499899699097292,
"loss": 2.7376,
"theoretical_loss": 3.3887381854012086,
"tokens_seen": 2320629760
},
{
"epoch": 0.7,
"learning_rate": 0.00014997993981945838,
"loss": 2.3689,
"theoretical_loss": 3.388730437584856,
"tokens_seen": 2320695296
},
{
"epoch": 0.7,
"learning_rate": 0.00014996990972918756,
"loss": 2.5498,
"theoretical_loss": 3.3887226900485583,
"tokens_seen": 2320760832
},
{
"epoch": 0.7,
"learning_rate": 0.00014995987963891674,
"loss": 2.8022,
"theoretical_loss": 3.388714942792298,
"tokens_seen": 2320826368
},
{
"epoch": 0.7,
"learning_rate": 0.00014994984954864595,
"loss": 2.5586,
"theoretical_loss": 3.388707195816056,
"tokens_seen": 2320891904
},
{
"epoch": 0.7,
"learning_rate": 0.00014993981945837513,
"loss": 2.7205,
"theoretical_loss": 3.388699449119815,
"tokens_seen": 2320957440
},
{
"epoch": 0.7,
"learning_rate": 0.0001499297893681043,
"loss": 2.5764,
"theoretical_loss": 3.3886917027035572,
"tokens_seen": 2321022976
},
{
"epoch": 0.7,
"learning_rate": 0.0001499197592778335,
"loss": 2.682,
"theoretical_loss": 3.3886839565672635,
"tokens_seen": 2321088512
},
{
"epoch": 0.7,
"learning_rate": 0.00014990972918756267,
"loss": 2.4343,
"theoretical_loss": 3.3886762107109174,
"tokens_seen": 2321154048
},
{
"epoch": 0.7,
"learning_rate": 0.00014989969909729188,
"loss": 2.8809,
"theoretical_loss": 3.3886684651344994,
"tokens_seen": 2321219584
},
{
"epoch": 0.7,
"learning_rate": 0.00014988966900702106,
"loss": 2.7351,
"theoretical_loss": 3.3886607198379926,
"tokens_seen": 2321285120
},
{
"epoch": 0.7,
"learning_rate": 0.00014987963891675024,
"loss": 2.7589,
"theoretical_loss": 3.388652974821378,
"tokens_seen": 2321350656
},
{
"epoch": 0.7,
"learning_rate": 0.00014986960882647942,
"loss": 2.5862,
"theoretical_loss": 3.3886452300846384,
"tokens_seen": 2321416192
},
{
"epoch": 0.7,
"learning_rate": 0.00014985957873620863,
"loss": 2.6519,
"theoretical_loss": 3.3886374856277555,
"tokens_seen": 2321481728
},
{
"epoch": 0.7,
"learning_rate": 0.0001498495486459378,
"loss": 2.4881,
"theoretical_loss": 3.3886297414507114,
"tokens_seen": 2321547264
},
{
"epoch": 0.7,
"objective/train/docs_used": 1308577,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.331028938293457,
"objective/train/theoretical_loss": 3.388621997553488,
"objective/train/tokens_used": 2342072800,
"theoretical_loss": 3.388621997553488,
"tokens_seen": 2321612800
},
{
"epoch": 0.7,
"learning_rate": 0.000149839518555667,
"loss": 2.6692,
"theoretical_loss": 3.388621997553488,
"tokens_seen": 2321612800
},
{
"epoch": 0.7,
"learning_rate": 0.00014982948846539617,
"loss": 2.7153,
"theoretical_loss": 3.388614253936067,
"tokens_seen": 2321678336
},
{
"epoch": 0.7,
"learning_rate": 0.00014981945837512536,
"loss": 2.5099,
"theoretical_loss": 3.3886065105984304,
"tokens_seen": 2321743872
},
{
"epoch": 0.7,
"learning_rate": 0.00014980942828485456,
"loss": 2.5467,
"theoretical_loss": 3.3885987675405613,
"tokens_seen": 2321809408
},
{
"epoch": 0.7,
"learning_rate": 0.00014979939819458374,
"loss": 2.703,
"theoretical_loss": 3.38859102476244,
"tokens_seen": 2321874944
},
{
"epoch": 0.7,
"learning_rate": 0.00014978936810431293,
"loss": 2.8124,
"theoretical_loss": 3.38858328226405,
"tokens_seen": 2321940480
},
{
"epoch": 0.7,
"learning_rate": 0.0001497793380140421,
"loss": 2.6055,
"theoretical_loss": 3.3885755400453723,
"tokens_seen": 2322006016
},
{
"epoch": 0.7,
"learning_rate": 0.00014976930792377134,
"loss": 2.6206,
"theoretical_loss": 3.388567798106389,
"tokens_seen": 2322071552
},
{
"epoch": 0.7,
"learning_rate": 0.00014975927783350052,
"loss": 2.9521,
"theoretical_loss": 3.388560056447083,
"tokens_seen": 2322137088
},
{
"epoch": 0.7,
"learning_rate": 0.0001497492477432297,
"loss": 2.7289,
"theoretical_loss": 3.3885523150674355,
"tokens_seen": 2322202624
},
{
"epoch": 0.7,
"learning_rate": 0.00014973921765295889,
"loss": 2.792,
"theoretical_loss": 3.3885445739674283,
"tokens_seen": 2322268160
},
{
"epoch": 0.7,
"learning_rate": 0.00014972918756268807,
"loss": 2.7656,
"theoretical_loss": 3.388536833147044,
"tokens_seen": 2322333696
},
{
"epoch": 0.7,
"learning_rate": 0.00014971915747241728,
"loss": 2.6936,
"theoretical_loss": 3.388529092606264,
"tokens_seen": 2322399232
},
{
"epoch": 0.7,
"learning_rate": 0.00014970912738214646,
"loss": 2.7308,
"theoretical_loss": 3.3885213523450712,
"tokens_seen": 2322464768
},
{
"epoch": 0.7,
"learning_rate": 0.00014969909729187564,
"loss": 2.6703,
"theoretical_loss": 3.3885136123634467,
"tokens_seen": 2322530304
},
{
"epoch": 0.7,
"learning_rate": 0.00014968906720160482,
"loss": 2.5558,
"theoretical_loss": 3.3885058726613733,
"tokens_seen": 2322595840
},
{
"epoch": 0.7,
"learning_rate": 0.00014967903711133403,
"loss": 2.5321,
"theoretical_loss": 3.388498133238832,
"tokens_seen": 2322661376
},
{
"epoch": 0.7,
"learning_rate": 0.0001496690070210632,
"loss": 2.7112,
"theoretical_loss": 3.388490394095806,
"tokens_seen": 2322726912
},
{
"epoch": 0.7,
"learning_rate": 0.0001496589769307924,
"loss": 2.6606,
"theoretical_loss": 3.3884826552322767,
"tokens_seen": 2322792448
},
{
"epoch": 0.7,
"learning_rate": 0.00014964894684052157,
"loss": 2.6783,
"theoretical_loss": 3.388474916648226,
"tokens_seen": 2322857984
},
{
"epoch": 0.7,
"learning_rate": 0.00014963891675025075,
"loss": 2.72,
"theoretical_loss": 3.388467178343636,
"tokens_seen": 2322923520
},
{
"epoch": 0.7,
"learning_rate": 0.00014962888665997996,
"loss": 2.6508,
"theoretical_loss": 3.3884594403184893,
"tokens_seen": 2322989056
},
{
"epoch": 0.7,
"learning_rate": 0.00014961885656970914,
"loss": 2.5672,
"theoretical_loss": 3.3884517025727674,
"tokens_seen": 2323054592
},
{
"epoch": 0.7,
"learning_rate": 0.00014960882647943832,
"loss": 2.63,
"theoretical_loss": 3.388443965106452,
"tokens_seen": 2323120128
},
{
"epoch": 0.7,
"learning_rate": 0.0001495987963891675,
"loss": 2.8846,
"theoretical_loss": 3.3884362279195255,
"tokens_seen": 2323185664
},
{
"epoch": 0.7,
"objective/train/docs_used": 1309762,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.6374409198760986,
"objective/train/theoretical_loss": 3.3884284910119704,
"objective/train/tokens_used": 2343711200,
"theoretical_loss": 3.3884284910119704,
"tokens_seen": 2323251200
},
{
"epoch": 0.7,
"learning_rate": 0.0001495887662988967,
"loss": 2.5795,
"theoretical_loss": 3.3884284910119704,
"tokens_seen": 2323251200
},
{
"epoch": 0.7,
"learning_rate": 0.0001495787362086259,
"loss": 2.6827,
"theoretical_loss": 3.388420754383768,
"tokens_seen": 2323316736
},
{
"epoch": 0.7,
"learning_rate": 0.00014956870611835507,
"loss": 2.7197,
"theoretical_loss": 3.3884130180349006,
"tokens_seen": 2323382272
},
{
"epoch": 0.7,
"learning_rate": 0.00014955867602808425,
"loss": 2.9289,
"theoretical_loss": 3.3884052819653503,
"tokens_seen": 2323447808
},
{
"epoch": 0.7,
"learning_rate": 0.00014954864593781344,
"loss": 2.6861,
"theoretical_loss": 3.388397546175099,
"tokens_seen": 2323513344
},
{
"epoch": 0.7,
"learning_rate": 0.00014953861584754264,
"loss": 2.6662,
"theoretical_loss": 3.388389810664129,
"tokens_seen": 2323578880
},
{
"epoch": 0.7,
"learning_rate": 0.00014952858575727182,
"loss": 2.7095,
"theoretical_loss": 3.3883820754324216,
"tokens_seen": 2323644416
},
{
"epoch": 0.7,
"learning_rate": 0.000149518555667001,
"loss": 2.7562,
"theoretical_loss": 3.3883743404799596,
"tokens_seen": 2323709952
},
{
"epoch": 0.7,
"learning_rate": 0.0001495085255767302,
"loss": 2.7546,
"theoretical_loss": 3.388366605806725,
"tokens_seen": 2323775488
},
{
"epoch": 0.7,
"learning_rate": 0.0001494984954864594,
"loss": 2.6396,
"theoretical_loss": 3.3883588714127,
"tokens_seen": 2323841024
},
{
"epoch": 0.7,
"learning_rate": 0.00014948846539618858,
"loss": 2.7041,
"theoretical_loss": 3.388351137297866,
"tokens_seen": 2323906560
},
{
"epoch": 0.7,
"learning_rate": 0.00014947843530591776,
"loss": 2.7174,
"theoretical_loss": 3.3883434034622057,
"tokens_seen": 2323972096
},
{
"epoch": 0.7,
"learning_rate": 0.00014946840521564694,
"loss": 2.4956,
"theoretical_loss": 3.3883356699057003,
"tokens_seen": 2324037632
},
{
"epoch": 0.7,
"learning_rate": 0.00014945837512537615,
"loss": 2.5527,
"theoretical_loss": 3.3883279366283325,
"tokens_seen": 2324103168
},
{
"epoch": 0.7,
"learning_rate": 0.00014944834503510533,
"loss": 2.4806,
"theoretical_loss": 3.3883202036300846,
"tokens_seen": 2324168704
},
{
"epoch": 0.7,
"learning_rate": 0.0001494383149448345,
"loss": 2.6707,
"theoretical_loss": 3.3883124709109382,
"tokens_seen": 2324234240
},
{
"epoch": 0.7,
"learning_rate": 0.0001494282848545637,
"loss": 2.5619,
"theoretical_loss": 3.3883047384708753,
"tokens_seen": 2324299776
},
{
"epoch": 0.7,
"learning_rate": 0.00014941825476429287,
"loss": 2.6821,
"theoretical_loss": 3.3882970063098785,
"tokens_seen": 2324365312
},
{
"epoch": 0.7,
"learning_rate": 0.00014940822467402208,
"loss": 2.6425,
"theoretical_loss": 3.3882892744279287,
"tokens_seen": 2324430848
},
{
"epoch": 0.7,
"learning_rate": 0.00014939819458375126,
"loss": 2.7358,
"theoretical_loss": 3.3882815428250095,
"tokens_seen": 2324496384
},
{
"epoch": 0.7,
"learning_rate": 0.00014938816449348044,
"loss": 2.8452,
"theoretical_loss": 3.3882738115011017,
"tokens_seen": 2324561920
},
{
"epoch": 0.7,
"learning_rate": 0.00014937813440320962,
"loss": 2.733,
"theoretical_loss": 3.388266080456188,
"tokens_seen": 2324627456
},
{
"epoch": 0.7,
"learning_rate": 0.00014936810431293883,
"loss": 2.5552,
"theoretical_loss": 3.388258349690251,
"tokens_seen": 2324692992
},
{
"epoch": 0.7,
"learning_rate": 0.000149358074222668,
"loss": 2.5997,
"theoretical_loss": 3.388250619203271,
"tokens_seen": 2324758528
},
{
"epoch": 0.7,
"learning_rate": 0.0001493480441323972,
"loss": 2.6671,
"theoretical_loss": 3.388242888995232,
"tokens_seen": 2324824064
},
{
"epoch": 0.7,
"objective/train/docs_used": 1310527,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 2.5369784832000732,
"objective/train/theoretical_loss": 3.388235159066115,
"objective/train/tokens_used": 2345349600,
"theoretical_loss": 3.388235159066115,
"tokens_seen": 2324889600
},
{
"epoch": 0.7,
"learning_rate": 0.00014933801404212637,
"loss": 2.7336,
"theoretical_loss": 3.388235159066115,
"tokens_seen": 2324889600
},
{
"epoch": 0.7,
"learning_rate": 0.00014932798395185556,
"loss": 2.5742,
"theoretical_loss": 3.388227429415902,
"tokens_seen": 2324955136
},
{
"epoch": 0.7,
"learning_rate": 0.00014931795386158476,
"loss": 2.6767,
"theoretical_loss": 3.3882197000445755,
"tokens_seen": 2325020672
},
{
"epoch": 0.7,
"learning_rate": 0.00014930792377131394,
"loss": 2.7307,
"theoretical_loss": 3.388211970952118,
"tokens_seen": 2325086208
},
{
"epoch": 0.7,
"learning_rate": 0.00014929789368104313,
"loss": 2.5592,
"theoretical_loss": 3.3882042421385106,
"tokens_seen": 2325151744
},
{
"epoch": 0.7,
"learning_rate": 0.0001492878635907723,
"loss": 2.7935,
"theoretical_loss": 3.388196513603736,
"tokens_seen": 2325217280
},
{
"epoch": 0.7,
"learning_rate": 0.00014927783350050152,
"loss": 2.5273,
"theoretical_loss": 3.388188785347776,
"tokens_seen": 2325282816
},
{
"epoch": 0.7,
"learning_rate": 0.0001492678034102307,
"loss": 2.8864,
"theoretical_loss": 3.3881810573706126,
"tokens_seen": 2325348352
},
{
"epoch": 0.7,
"learning_rate": 0.00014925777331995988,
"loss": 2.8276,
"theoretical_loss": 3.3881733296722283,
"tokens_seen": 2325413888
},
{
"epoch": 0.7,
"learning_rate": 0.00014924774322968906,
"loss": 2.5809,
"theoretical_loss": 3.388165602252605,
"tokens_seen": 2325479424
},
{
"epoch": 0.7,
"learning_rate": 0.00014923771313941824,
"loss": 2.6551,
"theoretical_loss": 3.388157875111725,
"tokens_seen": 2325544960
},
{
"epoch": 0.7,
"learning_rate": 0.00014922768304914745,
"loss": 2.836,
"theoretical_loss": 3.38815014824957,
"tokens_seen": 2325610496
},
{
"epoch": 0.7,
"learning_rate": 0.00014921765295887663,
"loss": 2.6413,
"theoretical_loss": 3.3881424216661222,
"tokens_seen": 2325676032
},
{
"epoch": 0.7,
"learning_rate": 0.0001492076228686058,
"loss": 2.6693,
"theoretical_loss": 3.3881346953613636,
"tokens_seen": 2325741568
},
{
"epoch": 0.7,
"learning_rate": 0.000149197592778335,
"loss": 2.6445,
"theoretical_loss": 3.3881269693352762,
"tokens_seen": 2325807104
},
{
"epoch": 0.7,
"learning_rate": 0.0001491875626880642,
"loss": 2.8334,
"theoretical_loss": 3.388119243587843,
"tokens_seen": 2325872640
}
],
"max_steps": 50354,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.18697653960704e+18,
"trial_name": null,
"trial_params": null
}