yihanwang617's picture
Model save
23e8249 verified
raw
history blame
32.9 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9990265609790016,
"eval_steps": 200,
"global_step": 898,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.001112501738283966,
"grad_norm": 0.6684414708991117,
"learning_rate": 2.2222222222222225e-06,
"loss": 1.1751,
"step": 1
},
{
"epoch": 0.00556250869141983,
"grad_norm": 0.7639609960034875,
"learning_rate": 1.1111111111111112e-05,
"loss": 1.282,
"step": 5
},
{
"epoch": 0.01112501738283966,
"grad_norm": 0.8970088039778137,
"learning_rate": 2.2222222222222223e-05,
"loss": 1.276,
"step": 10
},
{
"epoch": 0.016687526074259492,
"grad_norm": 0.5983155959810567,
"learning_rate": 3.3333333333333335e-05,
"loss": 1.2166,
"step": 15
},
{
"epoch": 0.02225003476567932,
"grad_norm": 0.9387996319771594,
"learning_rate": 4.4444444444444447e-05,
"loss": 1.0526,
"step": 20
},
{
"epoch": 0.027812543457099152,
"grad_norm": 0.3387344315023908,
"learning_rate": 5.555555555555556e-05,
"loss": 0.9608,
"step": 25
},
{
"epoch": 0.033375052148518984,
"grad_norm": 0.23890923775202408,
"learning_rate": 6.666666666666667e-05,
"loss": 0.911,
"step": 30
},
{
"epoch": 0.03893756083993881,
"grad_norm": 0.22798914076978022,
"learning_rate": 7.777777777777778e-05,
"loss": 0.8445,
"step": 35
},
{
"epoch": 0.04450006953135864,
"grad_norm": 0.31587333267646156,
"learning_rate": 8.888888888888889e-05,
"loss": 0.8105,
"step": 40
},
{
"epoch": 0.05006257822277847,
"grad_norm": 0.19505366669382362,
"learning_rate": 0.0001,
"loss": 0.8265,
"step": 45
},
{
"epoch": 0.055625086914198305,
"grad_norm": 0.20253125415926881,
"learning_rate": 0.00011111111111111112,
"loss": 0.7973,
"step": 50
},
{
"epoch": 0.061187595605618136,
"grad_norm": 0.19874800607368992,
"learning_rate": 0.00012222222222222224,
"loss": 0.7554,
"step": 55
},
{
"epoch": 0.06675010429703797,
"grad_norm": 0.1603617233155014,
"learning_rate": 0.00013333333333333334,
"loss": 0.7737,
"step": 60
},
{
"epoch": 0.0723126129884578,
"grad_norm": 0.182313382422268,
"learning_rate": 0.00014444444444444444,
"loss": 0.7744,
"step": 65
},
{
"epoch": 0.07787512167987762,
"grad_norm": 0.1404552378108098,
"learning_rate": 0.00015555555555555556,
"loss": 0.7995,
"step": 70
},
{
"epoch": 0.08343763037129745,
"grad_norm": 0.15330163912523168,
"learning_rate": 0.0001666666666666667,
"loss": 0.7854,
"step": 75
},
{
"epoch": 0.08900013906271728,
"grad_norm": 0.15880144560375917,
"learning_rate": 0.00017777777777777779,
"loss": 0.7572,
"step": 80
},
{
"epoch": 0.09456264775413711,
"grad_norm": 0.17526658243677723,
"learning_rate": 0.00018888888888888888,
"loss": 0.7421,
"step": 85
},
{
"epoch": 0.10012515644555695,
"grad_norm": 0.1756854246552801,
"learning_rate": 0.0002,
"loss": 0.7569,
"step": 90
},
{
"epoch": 0.10568766513697678,
"grad_norm": 0.13988463897936598,
"learning_rate": 0.00019998110384864614,
"loss": 0.7271,
"step": 95
},
{
"epoch": 0.11125017382839661,
"grad_norm": 0.14983210481706255,
"learning_rate": 0.0001999244225358753,
"loss": 0.7444,
"step": 100
},
{
"epoch": 0.11681268251981644,
"grad_norm": 0.14861727909266345,
"learning_rate": 0.00019982997748286082,
"loss": 0.7783,
"step": 105
},
{
"epoch": 0.12237519121123627,
"grad_norm": 0.17979168387360328,
"learning_rate": 0.00019969780438256293,
"loss": 0.7651,
"step": 110
},
{
"epoch": 0.1279376999026561,
"grad_norm": 0.16202080618298298,
"learning_rate": 0.00019952795318623986,
"loss": 0.719,
"step": 115
},
{
"epoch": 0.13350020859407594,
"grad_norm": 0.13085394814942794,
"learning_rate": 0.0001993204880845699,
"loss": 0.7044,
"step": 120
},
{
"epoch": 0.13906271728549577,
"grad_norm": 0.1369700622873931,
"learning_rate": 0.00019907548748339222,
"loss": 0.7166,
"step": 125
},
{
"epoch": 0.1446252259769156,
"grad_norm": 0.12798471944493164,
"learning_rate": 0.0001987930439740757,
"loss": 0.7302,
"step": 130
},
{
"epoch": 0.15018773466833543,
"grad_norm": 0.16062970178106098,
"learning_rate": 0.0001984732642985263,
"loss": 0.7477,
"step": 135
},
{
"epoch": 0.15575024335975524,
"grad_norm": 0.12776472354551027,
"learning_rate": 0.0001981162693088471,
"loss": 0.7445,
"step": 140
},
{
"epoch": 0.16131275205117507,
"grad_norm": 0.13554426042598045,
"learning_rate": 0.00019772219392166519,
"loss": 0.7252,
"step": 145
},
{
"epoch": 0.1668752607425949,
"grad_norm": 0.13347482010697612,
"learning_rate": 0.00019729118706714375,
"loss": 0.7254,
"step": 150
},
{
"epoch": 0.17243776943401473,
"grad_norm": 0.134631392605464,
"learning_rate": 0.000196823411632698,
"loss": 0.7184,
"step": 155
},
{
"epoch": 0.17800027812543456,
"grad_norm": 0.1504769169328635,
"learning_rate": 0.00019631904440143612,
"loss": 0.7259,
"step": 160
},
{
"epoch": 0.1835627868168544,
"grad_norm": 0.13782774235003004,
"learning_rate": 0.00019577827598534885,
"loss": 0.7367,
"step": 165
},
{
"epoch": 0.18912529550827423,
"grad_norm": 0.1363344338061366,
"learning_rate": 0.00019520131075327298,
"loss": 0.722,
"step": 170
},
{
"epoch": 0.19468780419969406,
"grad_norm": 0.14621350415121373,
"learning_rate": 0.00019458836675365556,
"loss": 0.73,
"step": 175
},
{
"epoch": 0.2002503128911139,
"grad_norm": 0.15278741364826062,
"learning_rate": 0.00019393967563214833,
"loss": 0.6995,
"step": 180
},
{
"epoch": 0.20581282158253372,
"grad_norm": 0.13331150727297583,
"learning_rate": 0.00019325548254406352,
"loss": 0.7487,
"step": 185
},
{
"epoch": 0.21137533027395355,
"grad_norm": 0.1412877661682615,
"learning_rate": 0.00019253604606172417,
"loss": 0.7359,
"step": 190
},
{
"epoch": 0.2169378389653734,
"grad_norm": 0.15015640302548008,
"learning_rate": 0.0001917816380767434,
"loss": 0.7014,
"step": 195
},
{
"epoch": 0.22250034765679322,
"grad_norm": 0.13131810843452316,
"learning_rate": 0.0001909925436972706,
"loss": 0.7113,
"step": 200
},
{
"epoch": 0.22250034765679322,
"eval_loss": 0.7031363844871521,
"eval_runtime": 38.8569,
"eval_samples_per_second": 10.474,
"eval_steps_per_second": 0.669,
"step": 200
},
{
"epoch": 0.22806285634821305,
"grad_norm": 0.14737211911767406,
"learning_rate": 0.0001901690611402423,
"loss": 0.7264,
"step": 205
},
{
"epoch": 0.23362536503963288,
"grad_norm": 0.1357323208908195,
"learning_rate": 0.00018931150161867916,
"loss": 0.7366,
"step": 210
},
{
"epoch": 0.2391878737310527,
"grad_norm": 0.13993345760253112,
"learning_rate": 0.0001884201892240715,
"loss": 0.7117,
"step": 215
},
{
"epoch": 0.24475038242247255,
"grad_norm": 0.1369219623281007,
"learning_rate": 0.00018749546080389757,
"loss": 0.7197,
"step": 220
},
{
"epoch": 0.2503128911138924,
"grad_norm": 0.15142549765891067,
"learning_rate": 0.00018653766583432113,
"loss": 0.7402,
"step": 225
},
{
"epoch": 0.2558753998053122,
"grad_norm": 0.1268630478080886,
"learning_rate": 0.0001855471662881164,
"loss": 0.7061,
"step": 230
},
{
"epoch": 0.26143790849673204,
"grad_norm": 0.15250219181797606,
"learning_rate": 0.0001845243364978702,
"loss": 0.6856,
"step": 235
},
{
"epoch": 0.2670004171881519,
"grad_norm": 0.12578511590375505,
"learning_rate": 0.00018346956301451304,
"loss": 0.6933,
"step": 240
},
{
"epoch": 0.2725629258795717,
"grad_norm": 0.12595811486114722,
"learning_rate": 0.00018238324446123266,
"loss": 0.7216,
"step": 245
},
{
"epoch": 0.27812543457099154,
"grad_norm": 0.1372248567648991,
"learning_rate": 0.00018126579138282503,
"loss": 0.7092,
"step": 250
},
{
"epoch": 0.28368794326241137,
"grad_norm": 0.1268675904703008,
"learning_rate": 0.0001801176260905402,
"loss": 0.6885,
"step": 255
},
{
"epoch": 0.2892504519538312,
"grad_norm": 0.15050289967108632,
"learning_rate": 0.00017893918250248104,
"loss": 0.7095,
"step": 260
},
{
"epoch": 0.29481296064525103,
"grad_norm": 0.13336034584149736,
"learning_rate": 0.00017773090597961554,
"loss": 0.6966,
"step": 265
},
{
"epoch": 0.30037546933667086,
"grad_norm": 0.158284438999716,
"learning_rate": 0.00017649325315746478,
"loss": 0.7275,
"step": 270
},
{
"epoch": 0.3059379780280907,
"grad_norm": 0.14579684984990443,
"learning_rate": 0.00017522669177352977,
"loss": 0.7206,
"step": 275
},
{
"epoch": 0.31150048671951047,
"grad_norm": 0.12269074197800435,
"learning_rate": 0.0001739317004905227,
"loss": 0.697,
"step": 280
},
{
"epoch": 0.3170629954109303,
"grad_norm": 0.13835017103100894,
"learning_rate": 0.00017260876871546936,
"loss": 0.7174,
"step": 285
},
{
"epoch": 0.32262550410235014,
"grad_norm": 0.14300200863185467,
"learning_rate": 0.00017125839641475072,
"loss": 0.726,
"step": 290
},
{
"epoch": 0.32818801279376997,
"grad_norm": 0.12914092926857734,
"learning_rate": 0.0001698810939251543,
"loss": 0.6991,
"step": 295
},
{
"epoch": 0.3337505214851898,
"grad_norm": 0.1293105805311886,
"learning_rate": 0.00016847738176100632,
"loss": 0.7264,
"step": 300
},
{
"epoch": 0.33931303017660963,
"grad_norm": 0.1263743857475521,
"learning_rate": 0.00016704779041745686,
"loss": 0.6938,
"step": 305
},
{
"epoch": 0.34487553886802946,
"grad_norm": 0.16556508393117347,
"learning_rate": 0.000165592860169994,
"loss": 0.7251,
"step": 310
},
{
"epoch": 0.3504380475594493,
"grad_norm": 0.13892185415083144,
"learning_rate": 0.00016411314087026106,
"loss": 0.7439,
"step": 315
},
{
"epoch": 0.3560005562508691,
"grad_norm": 0.1427616131100252,
"learning_rate": 0.00016260919173825508,
"loss": 0.6905,
"step": 320
},
{
"epoch": 0.36156306494228896,
"grad_norm": 0.1187614399483276,
"learning_rate": 0.00016108158115098444,
"loss": 0.681,
"step": 325
},
{
"epoch": 0.3671255736337088,
"grad_norm": 0.1428657673910464,
"learning_rate": 0.0001595308864276666,
"loss": 0.705,
"step": 330
},
{
"epoch": 0.3726880823251286,
"grad_norm": 0.14224088322242012,
"learning_rate": 0.00015795769361154547,
"loss": 0.7139,
"step": 335
},
{
"epoch": 0.37825059101654845,
"grad_norm": 0.1456871215491553,
"learning_rate": 0.00015636259724841222,
"loss": 0.7292,
"step": 340
},
{
"epoch": 0.3838130997079683,
"grad_norm": 0.13983468342540284,
"learning_rate": 0.00015474620016191294,
"loss": 0.7222,
"step": 345
},
{
"epoch": 0.3893756083993881,
"grad_norm": 0.14315721096194434,
"learning_rate": 0.00015310911322572753,
"loss": 0.7112,
"step": 350
},
{
"epoch": 0.39493811709080795,
"grad_norm": 0.12717703522604246,
"learning_rate": 0.00015145195513270644,
"loss": 0.7018,
"step": 355
},
{
"epoch": 0.4005006257822278,
"grad_norm": 0.1405738893906113,
"learning_rate": 0.0001497753521610526,
"loss": 0.708,
"step": 360
},
{
"epoch": 0.4060631344736476,
"grad_norm": 0.12649904699680478,
"learning_rate": 0.00014807993793763619,
"loss": 0.6728,
"step": 365
},
{
"epoch": 0.41162564316506745,
"grad_norm": 0.1366397886559032,
"learning_rate": 0.00014636635319853275,
"loss": 0.6855,
"step": 370
},
{
"epoch": 0.4171881518564873,
"grad_norm": 0.1348575434185503,
"learning_rate": 0.00014463524554687399,
"loss": 0.6969,
"step": 375
},
{
"epoch": 0.4227506605479071,
"grad_norm": 0.13467406560845768,
"learning_rate": 0.0001428872692081038,
"loss": 0.7179,
"step": 380
},
{
"epoch": 0.42831316923932694,
"grad_norm": 0.13474764353512392,
"learning_rate": 0.00014112308478273145,
"loss": 0.6853,
"step": 385
},
{
"epoch": 0.4338756779307468,
"grad_norm": 0.12908327147450624,
"learning_rate": 0.00013934335899667527,
"loss": 0.7131,
"step": 390
},
{
"epoch": 0.4394381866221666,
"grad_norm": 0.12215743992138174,
"learning_rate": 0.00013754876444929166,
"loss": 0.6748,
"step": 395
},
{
"epoch": 0.44500069531358644,
"grad_norm": 0.12646747341185763,
"learning_rate": 0.0001357399793591844,
"loss": 0.7181,
"step": 400
},
{
"epoch": 0.44500069531358644,
"eval_loss": 0.6872018575668335,
"eval_runtime": 38.1286,
"eval_samples_per_second": 10.674,
"eval_steps_per_second": 0.682,
"step": 400
},
{
"epoch": 0.45056320400500627,
"grad_norm": 0.13122165542663003,
"learning_rate": 0.00013391768730789002,
"loss": 0.7121,
"step": 405
},
{
"epoch": 0.4561257126964261,
"grad_norm": 0.12302514176584399,
"learning_rate": 0.00013208257698153677,
"loss": 0.7122,
"step": 410
},
{
"epoch": 0.46168822138784593,
"grad_norm": 0.14440204630512954,
"learning_rate": 0.00013023534191057426,
"loss": 0.7239,
"step": 415
},
{
"epoch": 0.46725073007926576,
"grad_norm": 0.13547988803582495,
"learning_rate": 0.0001283766802076722,
"loss": 0.7206,
"step": 420
},
{
"epoch": 0.4728132387706856,
"grad_norm": 0.13999737919216407,
"learning_rate": 0.00012650729430388764,
"loss": 0.6826,
"step": 425
},
{
"epoch": 0.4783757474621054,
"grad_norm": 0.13048583251685064,
"learning_rate": 0.00012462789068320017,
"loss": 0.6926,
"step": 430
},
{
"epoch": 0.48393825615352526,
"grad_norm": 0.1496684308590174,
"learning_rate": 0.00012273917961551513,
"loss": 0.7342,
"step": 435
},
{
"epoch": 0.4895007648449451,
"grad_norm": 0.14327740803897993,
"learning_rate": 0.00012084187488823657,
"loss": 0.6966,
"step": 440
},
{
"epoch": 0.4950632735363649,
"grad_norm": 0.14722640475177795,
"learning_rate": 0.00011893669353651031,
"loss": 0.7045,
"step": 445
},
{
"epoch": 0.5006257822277848,
"grad_norm": 0.13182498566362877,
"learning_rate": 0.00011702435557223987,
"loss": 0.7205,
"step": 450
},
{
"epoch": 0.5061882909192046,
"grad_norm": 0.14649854622775277,
"learning_rate": 0.00011510558371197753,
"loss": 0.6972,
"step": 455
},
{
"epoch": 0.5117507996106244,
"grad_norm": 0.14603995632853023,
"learning_rate": 0.00011318110310379301,
"loss": 0.7455,
"step": 460
},
{
"epoch": 0.5173133083020443,
"grad_norm": 0.1413104318676706,
"learning_rate": 0.0001112516410532233,
"loss": 0.6917,
"step": 465
},
{
"epoch": 0.5228758169934641,
"grad_norm": 0.1319527122221744,
"learning_rate": 0.00010931792674840718,
"loss": 0.6855,
"step": 470
},
{
"epoch": 0.5284383256848839,
"grad_norm": 0.1245675456499167,
"learning_rate": 0.0001073806909845082,
"loss": 0.7028,
"step": 475
},
{
"epoch": 0.5340008343763037,
"grad_norm": 0.11698601668007995,
"learning_rate": 0.00010544066588753044,
"loss": 0.6933,
"step": 480
},
{
"epoch": 0.5395633430677236,
"grad_norm": 0.12677292396669446,
"learning_rate": 0.00010349858463763113,
"loss": 0.683,
"step": 485
},
{
"epoch": 0.5451258517591434,
"grad_norm": 0.13456238683420088,
"learning_rate": 0.0001015551811920351,
"loss": 0.7006,
"step": 490
},
{
"epoch": 0.5506883604505632,
"grad_norm": 0.15227986749806005,
"learning_rate": 9.961119000765531e-05,
"loss": 0.6894,
"step": 495
},
{
"epoch": 0.5562508691419831,
"grad_norm": 0.12847199230412348,
"learning_rate": 9.766734576352478e-05,
"loss": 0.6964,
"step": 500
},
{
"epoch": 0.5618133778334029,
"grad_norm": 0.13064877508080042,
"learning_rate": 9.572438308314446e-05,
"loss": 0.6885,
"step": 505
},
{
"epoch": 0.5673758865248227,
"grad_norm": 0.1372527364951275,
"learning_rate": 9.378303625685195e-05,
"loss": 0.7056,
"step": 510
},
{
"epoch": 0.5729383952162426,
"grad_norm": 0.13318320240144188,
"learning_rate": 9.18440389643165e-05,
"loss": 0.6792,
"step": 515
},
{
"epoch": 0.5785009039076624,
"grad_norm": 0.1389287540275015,
"learning_rate": 8.990812399726435e-05,
"loss": 0.6955,
"step": 520
},
{
"epoch": 0.5840634125990822,
"grad_norm": 0.15589038738911026,
"learning_rate": 8.797602298254004e-05,
"loss": 0.6995,
"step": 525
},
{
"epoch": 0.5896259212905021,
"grad_norm": 0.13296983699316012,
"learning_rate": 8.604846610560771e-05,
"loss": 0.6707,
"step": 530
},
{
"epoch": 0.5951884299819219,
"grad_norm": 0.12114356841208347,
"learning_rate": 8.412618183459708e-05,
"loss": 0.7015,
"step": 535
},
{
"epoch": 0.6007509386733417,
"grad_norm": 0.12351709388346521,
"learning_rate": 8.220989664499878e-05,
"loss": 0.7006,
"step": 540
},
{
"epoch": 0.6063134473647616,
"grad_norm": 0.1233117181776912,
"learning_rate": 8.030033474511249e-05,
"loss": 0.7061,
"step": 545
},
{
"epoch": 0.6118759560561814,
"grad_norm": 0.14966743246815337,
"learning_rate": 7.839821780235168e-05,
"loss": 0.7383,
"step": 550
},
{
"epoch": 0.6174384647476012,
"grad_norm": 0.13689976425741557,
"learning_rate": 7.650426467050926e-05,
"loss": 0.7083,
"step": 555
},
{
"epoch": 0.6230009734390209,
"grad_norm": 0.1540376322278511,
"learning_rate": 7.461919111808595e-05,
"loss": 0.6991,
"step": 560
},
{
"epoch": 0.6285634821304408,
"grad_norm": 0.15818277915085305,
"learning_rate": 7.274370955778498e-05,
"loss": 0.7147,
"step": 565
},
{
"epoch": 0.6341259908218606,
"grad_norm": 0.13733946492240742,
"learning_rate": 7.087852877727481e-05,
"loss": 0.7043,
"step": 570
},
{
"epoch": 0.6396884995132804,
"grad_norm": 0.12905771101166952,
"learning_rate": 6.902435367132208e-05,
"loss": 0.6952,
"step": 575
},
{
"epoch": 0.6452510082047003,
"grad_norm": 0.14917864740623707,
"learning_rate": 6.718188497539554e-05,
"loss": 0.6894,
"step": 580
},
{
"epoch": 0.6508135168961201,
"grad_norm": 0.17034610159423622,
"learning_rate": 6.535181900084206e-05,
"loss": 0.6906,
"step": 585
},
{
"epoch": 0.6563760255875399,
"grad_norm": 0.14577692793179953,
"learning_rate": 6.35348473717345e-05,
"loss": 0.6822,
"step": 590
},
{
"epoch": 0.6619385342789598,
"grad_norm": 0.1400663188779511,
"learning_rate": 6.173165676349103e-05,
"loss": 0.702,
"step": 595
},
{
"epoch": 0.6675010429703796,
"grad_norm": 0.13477519960817833,
"learning_rate": 5.9942928643364724e-05,
"loss": 0.7171,
"step": 600
},
{
"epoch": 0.6675010429703796,
"eval_loss": 0.6785927414894104,
"eval_runtime": 38.1279,
"eval_samples_per_second": 10.675,
"eval_steps_per_second": 0.682,
"step": 600
},
{
"epoch": 0.6730635516617994,
"grad_norm": 0.13058646321017944,
"learning_rate": 5.816933901290136e-05,
"loss": 0.6959,
"step": 605
},
{
"epoch": 0.6786260603532193,
"grad_norm": 0.14039443258781562,
"learning_rate": 5.6411558152462894e-05,
"loss": 0.6991,
"step": 610
},
{
"epoch": 0.6841885690446391,
"grad_norm": 0.1435099198472092,
"learning_rate": 5.4670250367913023e-05,
"loss": 0.7008,
"step": 615
},
{
"epoch": 0.6897510777360589,
"grad_norm": 0.1251912516827599,
"learning_rate": 5.2946073739560706e-05,
"loss": 0.6828,
"step": 620
},
{
"epoch": 0.6953135864274788,
"grad_norm": 0.12907022711465035,
"learning_rate": 5.1239679873456634e-05,
"loss": 0.6811,
"step": 625
},
{
"epoch": 0.7008760951188986,
"grad_norm": 0.13237387347735238,
"learning_rate": 4.955171365513603e-05,
"loss": 0.7235,
"step": 630
},
{
"epoch": 0.7064386038103184,
"grad_norm": 0.12376618467656812,
"learning_rate": 4.7882813005901696e-05,
"loss": 0.687,
"step": 635
},
{
"epoch": 0.7120011125017383,
"grad_norm": 0.1456941018180003,
"learning_rate": 4.623360864173893e-05,
"loss": 0.6813,
"step": 640
},
{
"epoch": 0.7175636211931581,
"grad_norm": 0.17032008806579718,
"learning_rate": 4.460472383495331e-05,
"loss": 0.6795,
"step": 645
},
{
"epoch": 0.7231261298845779,
"grad_norm": 0.12579166198761085,
"learning_rate": 4.2996774178621736e-05,
"loss": 0.6835,
"step": 650
},
{
"epoch": 0.7286886385759977,
"grad_norm": 0.14450683290791613,
"learning_rate": 4.141036735394574e-05,
"loss": 0.6862,
"step": 655
},
{
"epoch": 0.7342511472674176,
"grad_norm": 0.17119446089706108,
"learning_rate": 3.984610290059467e-05,
"loss": 0.7112,
"step": 660
},
{
"epoch": 0.7398136559588374,
"grad_norm": 0.14950992219628134,
"learning_rate": 3.830457199012585e-05,
"loss": 0.7259,
"step": 665
},
{
"epoch": 0.7453761646502572,
"grad_norm": 0.14485285236277412,
"learning_rate": 3.678635720256737e-05,
"loss": 0.6944,
"step": 670
},
{
"epoch": 0.7509386733416771,
"grad_norm": 0.13749155498971605,
"learning_rate": 3.529203230624747e-05,
"loss": 0.6758,
"step": 675
},
{
"epoch": 0.7565011820330969,
"grad_norm": 0.12869094791923574,
"learning_rate": 3.3822162040954354e-05,
"loss": 0.701,
"step": 680
},
{
"epoch": 0.7620636907245167,
"grad_norm": 0.12906896257797407,
"learning_rate": 3.237730190450816e-05,
"loss": 0.7059,
"step": 685
},
{
"epoch": 0.7676261994159366,
"grad_norm": 0.15164612968237504,
"learning_rate": 3.0957997942825336e-05,
"loss": 0.6925,
"step": 690
},
{
"epoch": 0.7731887081073564,
"grad_norm": 0.15092417362430902,
"learning_rate": 2.9564786543555388e-05,
"loss": 0.704,
"step": 695
},
{
"epoch": 0.7787512167987762,
"grad_norm": 0.12028807778273494,
"learning_rate": 2.819819423336775e-05,
"loss": 0.6615,
"step": 700
},
{
"epoch": 0.7843137254901961,
"grad_norm": 0.11668516511159772,
"learning_rate": 2.6858737478965035e-05,
"loss": 0.6761,
"step": 705
},
{
"epoch": 0.7898762341816159,
"grad_norm": 0.12013736025525155,
"learning_rate": 2.5546922491898495e-05,
"loss": 0.6812,
"step": 710
},
{
"epoch": 0.7954387428730357,
"grad_norm": 0.11406481330250699,
"learning_rate": 2.4263245037258995e-05,
"loss": 0.6653,
"step": 715
},
{
"epoch": 0.8010012515644556,
"grad_norm": 0.14003374126540685,
"learning_rate": 2.300819024631603e-05,
"loss": 0.7079,
"step": 720
},
{
"epoch": 0.8065637602558754,
"grad_norm": 0.14609209804734563,
"learning_rate": 2.178223243317532e-05,
"loss": 0.7126,
"step": 725
},
{
"epoch": 0.8121262689472952,
"grad_norm": 0.1412723606655406,
"learning_rate": 2.058583491552465e-05,
"loss": 0.6805,
"step": 730
},
{
"epoch": 0.8176887776387151,
"grad_norm": 0.13559476815952162,
"learning_rate": 1.941944983953552e-05,
"loss": 0.6687,
"step": 735
},
{
"epoch": 0.8232512863301349,
"grad_norm": 0.1212687591445283,
"learning_rate": 1.8283518008986567e-05,
"loss": 0.6949,
"step": 740
},
{
"epoch": 0.8288137950215547,
"grad_norm": 0.13418537722622348,
"learning_rate": 1.7178468718673714e-05,
"loss": 0.6947,
"step": 745
},
{
"epoch": 0.8343763037129746,
"grad_norm": 0.14626468131094617,
"learning_rate": 1.6104719592169902e-05,
"loss": 0.6974,
"step": 750
},
{
"epoch": 0.8399388124043944,
"grad_norm": 0.1505968834590085,
"learning_rate": 1.5062676423995247e-05,
"loss": 0.6699,
"step": 755
},
{
"epoch": 0.8455013210958142,
"grad_norm": 0.13840692663705956,
"learning_rate": 1.4052733026258281e-05,
"loss": 0.6775,
"step": 760
},
{
"epoch": 0.851063829787234,
"grad_norm": 0.15338747206444608,
"learning_rate": 1.3075271079825036e-05,
"loss": 0.6855,
"step": 765
},
{
"epoch": 0.8566263384786539,
"grad_norm": 0.15087298168068788,
"learning_rate": 1.2130659990073146e-05,
"loss": 0.7074,
"step": 770
},
{
"epoch": 0.8621888471700737,
"grad_norm": 0.15260125382082407,
"learning_rate": 1.1219256747285045e-05,
"loss": 0.6559,
"step": 775
},
{
"epoch": 0.8677513558614935,
"grad_norm": 0.13676984708600298,
"learning_rate": 1.0341405791733183e-05,
"loss": 0.7018,
"step": 780
},
{
"epoch": 0.8733138645529134,
"grad_norm": 0.12792881953620902,
"learning_rate": 9.49743888350798e-06,
"loss": 0.6758,
"step": 785
},
{
"epoch": 0.8788763732443332,
"grad_norm": 0.13024278184248392,
"learning_rate": 8.687674977138116e-06,
"loss": 0.703,
"step": 790
},
{
"epoch": 0.884438881935753,
"grad_norm": 0.15819074719833112,
"learning_rate": 7.912420101050367e-06,
"loss": 0.7026,
"step": 795
},
{
"epoch": 0.8900013906271729,
"grad_norm": 0.1363306447853743,
"learning_rate": 7.171967241914224e-06,
"loss": 0.7031,
"step": 800
},
{
"epoch": 0.8900013906271729,
"eval_loss": 0.6755391955375671,
"eval_runtime": 38.1439,
"eval_samples_per_second": 10.67,
"eval_steps_per_second": 0.682,
"step": 800
},
{
"epoch": 0.8955638993185927,
"grad_norm": 0.11599580684218645,
"learning_rate": 6.4665962339156005e-06,
"loss": 0.6559,
"step": 805
},
{
"epoch": 0.9011264080100125,
"grad_norm": 0.14301150354946243,
"learning_rate": 5.7965736530010916e-06,
"loss": 0.7436,
"step": 810
},
{
"epoch": 0.9066889167014324,
"grad_norm": 0.1454357862013425,
"learning_rate": 5.162152716132662e-06,
"loss": 0.6851,
"step": 815
},
{
"epoch": 0.9122514253928522,
"grad_norm": 0.1296885702439703,
"learning_rate": 4.563573185591219e-06,
"loss": 0.6799,
"step": 820
},
{
"epoch": 0.917813934084272,
"grad_norm": 0.13649350003605498,
"learning_rate": 4.0010612783648925e-06,
"loss": 0.6954,
"step": 825
},
{
"epoch": 0.9233764427756919,
"grad_norm": 0.11566680964002608,
"learning_rate": 3.4748295806564356e-06,
"loss": 0.6931,
"step": 830
},
{
"epoch": 0.9289389514671117,
"grad_norm": 0.1376951533333156,
"learning_rate": 2.9850769675419774e-06,
"loss": 0.7094,
"step": 835
},
{
"epoch": 0.9345014601585315,
"grad_norm": 0.13815883208860893,
"learning_rate": 2.5319885278115906e-06,
"loss": 0.6814,
"step": 840
},
{
"epoch": 0.9400639688499514,
"grad_norm": 0.12898446369151903,
"learning_rate": 2.115735494019966e-06,
"loss": 0.7252,
"step": 845
},
{
"epoch": 0.9456264775413712,
"grad_norm": 0.14801072441410335,
"learning_rate": 1.7364751777736332e-06,
"loss": 0.6701,
"step": 850
},
{
"epoch": 0.951188986232791,
"grad_norm": 0.1489167549426156,
"learning_rate": 1.394350910279385e-06,
"loss": 0.6852,
"step": 855
},
{
"epoch": 0.9567514949242109,
"grad_norm": 0.150107558255733,
"learning_rate": 1.089491988176017e-06,
"loss": 0.7177,
"step": 860
},
{
"epoch": 0.9623140036156307,
"grad_norm": 0.1462660934588179,
"learning_rate": 8.220136246701926e-07,
"loss": 0.6665,
"step": 865
},
{
"epoch": 0.9678765123070505,
"grad_norm": 0.13216262862731915,
"learning_rate": 5.920169059947411e-07,
"loss": 0.666,
"step": 870
},
{
"epoch": 0.9734390209984704,
"grad_norm": 0.1497443816075182,
"learning_rate": 3.9958875320580404e-07,
"loss": 0.6902,
"step": 875
},
{
"epoch": 0.9790015296898902,
"grad_norm": 0.1329859440152703,
"learning_rate": 2.448018893333681e-07,
"loss": 0.6959,
"step": 880
},
{
"epoch": 0.98456403838131,
"grad_norm": 0.14839381526763562,
"learning_rate": 1.277148118975835e-07,
"loss": 0.6884,
"step": 885
},
{
"epoch": 0.9901265470727298,
"grad_norm": 0.14357565386768453,
"learning_rate": 4.837177080119215e-08,
"loss": 0.6961,
"step": 890
},
{
"epoch": 0.9956890557641497,
"grad_norm": 0.12423764687193309,
"learning_rate": 6.8027516064606e-09,
"loss": 0.6802,
"step": 895
},
{
"epoch": 0.9990265609790016,
"step": 898,
"total_flos": 3899966689378304.0,
"train_loss": 0.7223505903458542,
"train_runtime": 18859.9671,
"train_samples_per_second": 3.05,
"train_steps_per_second": 0.048
}
],
"logging_steps": 5,
"max_steps": 898,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 3899966689378304.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}