{
"best_metric": 1.1517876386642456,
"best_model_checkpoint": "/media/user/Expansion/flan-t5-small-simplifier/checkpoint-80000",
"epoch": 2.0,
"eval_steps": 10000,
"global_step": 89920,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01112099644128114,
"grad_norm": 3.2175512313842773,
"learning_rate": 4.972197508896798e-05,
"loss": 1.6509,
"num_input_tokens_seen": 183488,
"step": 500
},
{
"epoch": 0.02224199288256228,
"grad_norm": 3.9029898643493652,
"learning_rate": 4.9443950177935946e-05,
"loss": 1.583,
"num_input_tokens_seen": 363744,
"step": 1000
},
{
"epoch": 0.03336298932384341,
"grad_norm": 3.4091885089874268,
"learning_rate": 4.9165925266903915e-05,
"loss": 1.5553,
"num_input_tokens_seen": 547584,
"step": 1500
},
{
"epoch": 0.04448398576512456,
"grad_norm": 2.5965495109558105,
"learning_rate": 4.888790035587189e-05,
"loss": 1.5444,
"num_input_tokens_seen": 732480,
"step": 2000
},
{
"epoch": 0.055604982206405695,
"grad_norm": 4.170108318328857,
"learning_rate": 4.860987544483986e-05,
"loss": 1.5137,
"num_input_tokens_seen": 916672,
"step": 2500
},
{
"epoch": 0.06672597864768683,
"grad_norm": 3.025068759918213,
"learning_rate": 4.8331850533807835e-05,
"loss": 1.5057,
"num_input_tokens_seen": 1095952,
"step": 3000
},
{
"epoch": 0.07784697508896797,
"grad_norm": 2.7047712802886963,
"learning_rate": 4.80538256227758e-05,
"loss": 1.5019,
"num_input_tokens_seen": 1276936,
"step": 3500
},
{
"epoch": 0.08896797153024912,
"grad_norm": 2.844285488128662,
"learning_rate": 4.777580071174377e-05,
"loss": 1.5074,
"num_input_tokens_seen": 1464080,
"step": 4000
},
{
"epoch": 0.10008896797153025,
"grad_norm": 3.055643081665039,
"learning_rate": 4.749777580071175e-05,
"loss": 1.4959,
"num_input_tokens_seen": 1647288,
"step": 4500
},
{
"epoch": 0.11120996441281139,
"grad_norm": 2.6245856285095215,
"learning_rate": 4.721975088967972e-05,
"loss": 1.4777,
"num_input_tokens_seen": 1836832,
"step": 5000
},
{
"epoch": 0.12233096085409252,
"grad_norm": 3.1758244037628174,
"learning_rate": 4.694172597864769e-05,
"loss": 1.4778,
"num_input_tokens_seen": 2020728,
"step": 5500
},
{
"epoch": 0.13345195729537365,
"grad_norm": 2.518728494644165,
"learning_rate": 4.666370106761566e-05,
"loss": 1.4758,
"num_input_tokens_seen": 2198968,
"step": 6000
},
{
"epoch": 0.1445729537366548,
"grad_norm": 3.8143937587738037,
"learning_rate": 4.638567615658363e-05,
"loss": 1.4846,
"num_input_tokens_seen": 2381968,
"step": 6500
},
{
"epoch": 0.15569395017793594,
"grad_norm": 2.766146421432495,
"learning_rate": 4.6107651245551604e-05,
"loss": 1.4468,
"num_input_tokens_seen": 2562176,
"step": 7000
},
{
"epoch": 0.16681494661921709,
"grad_norm": 3.891373634338379,
"learning_rate": 4.582962633451958e-05,
"loss": 1.4586,
"num_input_tokens_seen": 2744688,
"step": 7500
},
{
"epoch": 0.17793594306049823,
"grad_norm": 3.277316093444824,
"learning_rate": 4.555160142348754e-05,
"loss": 1.4606,
"num_input_tokens_seen": 2926000,
"step": 8000
},
{
"epoch": 0.18905693950177935,
"grad_norm": 3.242478132247925,
"learning_rate": 4.5273576512455517e-05,
"loss": 1.446,
"num_input_tokens_seen": 3108520,
"step": 8500
},
{
"epoch": 0.2001779359430605,
"grad_norm": 2.3061795234680176,
"learning_rate": 4.499555160142349e-05,
"loss": 1.4348,
"num_input_tokens_seen": 3289352,
"step": 9000
},
{
"epoch": 0.21129893238434164,
"grad_norm": 3.4106180667877197,
"learning_rate": 4.471752669039146e-05,
"loss": 1.4345,
"num_input_tokens_seen": 3473784,
"step": 9500
},
{
"epoch": 0.22241992882562278,
"grad_norm": 2.88779354095459,
"learning_rate": 4.443950177935943e-05,
"loss": 1.4423,
"num_input_tokens_seen": 3655312,
"step": 10000
},
{
"epoch": 0.22241992882562278,
"eval_loss": 1.2430843114852905,
"eval_runtime": 2.8218,
"eval_samples_per_second": 885.962,
"eval_steps_per_second": 110.922,
"num_input_tokens_seen": 3655312,
"step": 10000
},
{
"epoch": 0.23354092526690393,
"grad_norm": 2.829268217086792,
"learning_rate": 4.4161476868327405e-05,
"loss": 1.4321,
"num_input_tokens_seen": 3847328,
"step": 10500
},
{
"epoch": 0.24466192170818504,
"grad_norm": 3.807185173034668,
"learning_rate": 4.388345195729537e-05,
"loss": 1.4487,
"num_input_tokens_seen": 4025576,
"step": 11000
},
{
"epoch": 0.2557829181494662,
"grad_norm": 2.449007511138916,
"learning_rate": 4.360542704626335e-05,
"loss": 1.413,
"num_input_tokens_seen": 4213832,
"step": 11500
},
{
"epoch": 0.2669039145907473,
"grad_norm": 3.363555431365967,
"learning_rate": 4.3327402135231324e-05,
"loss": 1.4313,
"num_input_tokens_seen": 4395352,
"step": 12000
},
{
"epoch": 0.2780249110320285,
"grad_norm": 2.702747344970703,
"learning_rate": 4.3049377224199286e-05,
"loss": 1.4179,
"num_input_tokens_seen": 4582368,
"step": 12500
},
{
"epoch": 0.2891459074733096,
"grad_norm": 3.286044120788574,
"learning_rate": 4.277135231316726e-05,
"loss": 1.4137,
"num_input_tokens_seen": 4770640,
"step": 13000
},
{
"epoch": 0.30026690391459077,
"grad_norm": 2.6410391330718994,
"learning_rate": 4.249332740213524e-05,
"loss": 1.4303,
"num_input_tokens_seen": 4951280,
"step": 13500
},
{
"epoch": 0.3113879003558719,
"grad_norm": 3.239133358001709,
"learning_rate": 4.2215302491103205e-05,
"loss": 1.4185,
"num_input_tokens_seen": 5133552,
"step": 14000
},
{
"epoch": 0.322508896797153,
"grad_norm": 2.943094253540039,
"learning_rate": 4.1937277580071174e-05,
"loss": 1.4065,
"num_input_tokens_seen": 5319992,
"step": 14500
},
{
"epoch": 0.33362989323843417,
"grad_norm": 2.180136203765869,
"learning_rate": 4.165925266903915e-05,
"loss": 1.4096,
"num_input_tokens_seen": 5501592,
"step": 15000
},
{
"epoch": 0.3447508896797153,
"grad_norm": 2.4302403926849365,
"learning_rate": 4.138122775800712e-05,
"loss": 1.4203,
"num_input_tokens_seen": 5689312,
"step": 15500
},
{
"epoch": 0.35587188612099646,
"grad_norm": 2.850964069366455,
"learning_rate": 4.1103202846975093e-05,
"loss": 1.4291,
"num_input_tokens_seen": 5870384,
"step": 16000
},
{
"epoch": 0.3669928825622776,
"grad_norm": 1.9641114473342896,
"learning_rate": 4.082517793594306e-05,
"loss": 1.4054,
"num_input_tokens_seen": 6048624,
"step": 16500
},
{
"epoch": 0.3781138790035587,
"grad_norm": 2.287353754043579,
"learning_rate": 4.054715302491103e-05,
"loss": 1.4118,
"num_input_tokens_seen": 6229728,
"step": 17000
},
{
"epoch": 0.38923487544483987,
"grad_norm": 3.4425182342529297,
"learning_rate": 4.0269128113879006e-05,
"loss": 1.4061,
"num_input_tokens_seen": 6410968,
"step": 17500
},
{
"epoch": 0.400355871886121,
"grad_norm": 2.0604770183563232,
"learning_rate": 3.9991103202846975e-05,
"loss": 1.3871,
"num_input_tokens_seen": 6591312,
"step": 18000
},
{
"epoch": 0.41147686832740216,
"grad_norm": 2.671599864959717,
"learning_rate": 3.971307829181495e-05,
"loss": 1.42,
"num_input_tokens_seen": 6777912,
"step": 18500
},
{
"epoch": 0.4225978647686833,
"grad_norm": 2.176579475402832,
"learning_rate": 3.943505338078292e-05,
"loss": 1.417,
"num_input_tokens_seen": 6964240,
"step": 19000
},
{
"epoch": 0.4337188612099644,
"grad_norm": 3.2378785610198975,
"learning_rate": 3.915702846975089e-05,
"loss": 1.3992,
"num_input_tokens_seen": 7147288,
"step": 19500
},
{
"epoch": 0.44483985765124556,
"grad_norm": 3.0986084938049316,
"learning_rate": 3.887900355871886e-05,
"loss": 1.3884,
"num_input_tokens_seen": 7331520,
"step": 20000
},
{
"epoch": 0.44483985765124556,
"eval_loss": 1.2093411684036255,
"eval_runtime": 2.7984,
"eval_samples_per_second": 893.355,
"eval_steps_per_second": 111.848,
"num_input_tokens_seen": 7331520,
"step": 20000
},
{
"epoch": 0.4559608540925267,
"grad_norm": 3.4998202323913574,
"learning_rate": 3.860097864768684e-05,
"loss": 1.3915,
"num_input_tokens_seen": 7515512,
"step": 20500
},
{
"epoch": 0.46708185053380785,
"grad_norm": 3.0249533653259277,
"learning_rate": 3.832295373665481e-05,
"loss": 1.3967,
"num_input_tokens_seen": 7696752,
"step": 21000
},
{
"epoch": 0.47820284697508897,
"grad_norm": 6.8868513107299805,
"learning_rate": 3.8044928825622775e-05,
"loss": 1.4106,
"num_input_tokens_seen": 7878832,
"step": 21500
},
{
"epoch": 0.4893238434163701,
"grad_norm": 2.2134385108947754,
"learning_rate": 3.776690391459075e-05,
"loss": 1.3847,
"num_input_tokens_seen": 8059592,
"step": 22000
},
{
"epoch": 0.5004448398576512,
"grad_norm": 2.2698676586151123,
"learning_rate": 3.748887900355872e-05,
"loss": 1.3941,
"num_input_tokens_seen": 8245432,
"step": 22500
},
{
"epoch": 0.5115658362989324,
"grad_norm": 2.4593448638916016,
"learning_rate": 3.7210854092526695e-05,
"loss": 1.3716,
"num_input_tokens_seen": 8429400,
"step": 23000
},
{
"epoch": 0.5226868327402135,
"grad_norm": 2.5121207237243652,
"learning_rate": 3.6932829181494664e-05,
"loss": 1.3733,
"num_input_tokens_seen": 8618216,
"step": 23500
},
{
"epoch": 0.5338078291814946,
"grad_norm": 2.2493703365325928,
"learning_rate": 3.665480427046263e-05,
"loss": 1.3754,
"num_input_tokens_seen": 8801640,
"step": 24000
},
{
"epoch": 0.5449288256227758,
"grad_norm": 3.08921217918396,
"learning_rate": 3.637677935943061e-05,
"loss": 1.3694,
"num_input_tokens_seen": 8977800,
"step": 24500
},
{
"epoch": 0.556049822064057,
"grad_norm": 2.215222120285034,
"learning_rate": 3.609875444839858e-05,
"loss": 1.3674,
"num_input_tokens_seen": 9160904,
"step": 25000
},
{
"epoch": 0.5671708185053381,
"grad_norm": 2.553903818130493,
"learning_rate": 3.582072953736655e-05,
"loss": 1.3735,
"num_input_tokens_seen": 9348344,
"step": 25500
},
{
"epoch": 0.5782918149466192,
"grad_norm": 2.546022891998291,
"learning_rate": 3.554270462633452e-05,
"loss": 1.382,
"num_input_tokens_seen": 9532664,
"step": 26000
},
{
"epoch": 0.5894128113879004,
"grad_norm": 2.917534112930298,
"learning_rate": 3.5264679715302496e-05,
"loss": 1.3654,
"num_input_tokens_seen": 9717800,
"step": 26500
},
{
"epoch": 0.6005338078291815,
"grad_norm": 3.355299472808838,
"learning_rate": 3.4986654804270464e-05,
"loss": 1.3876,
"num_input_tokens_seen": 9902536,
"step": 27000
},
{
"epoch": 0.6116548042704626,
"grad_norm": 2.67924427986145,
"learning_rate": 3.470862989323844e-05,
"loss": 1.3575,
"num_input_tokens_seen": 10082216,
"step": 27500
},
{
"epoch": 0.6227758007117438,
"grad_norm": 3.040212392807007,
"learning_rate": 3.44306049822064e-05,
"loss": 1.3644,
"num_input_tokens_seen": 10263992,
"step": 28000
},
{
"epoch": 0.6338967971530249,
"grad_norm": 3.726254940032959,
"learning_rate": 3.415258007117438e-05,
"loss": 1.3616,
"num_input_tokens_seen": 10444128,
"step": 28500
},
{
"epoch": 0.645017793594306,
"grad_norm": 4.716592788696289,
"learning_rate": 3.387455516014235e-05,
"loss": 1.3662,
"num_input_tokens_seen": 10626928,
"step": 29000
},
{
"epoch": 0.6561387900355872,
"grad_norm": 2.9317166805267334,
"learning_rate": 3.359653024911032e-05,
"loss": 1.38,
"num_input_tokens_seen": 10808960,
"step": 29500
},
{
"epoch": 0.6672597864768683,
"grad_norm": 2.411684989929199,
"learning_rate": 3.331850533807829e-05,
"loss": 1.3782,
"num_input_tokens_seen": 10990432,
"step": 30000
},
{
"epoch": 0.6672597864768683,
"eval_loss": 1.185857892036438,
"eval_runtime": 2.9618,
"eval_samples_per_second": 844.079,
"eval_steps_per_second": 105.679,
"num_input_tokens_seen": 10990432,
"step": 30000
},
{
"epoch": 0.6783807829181495,
"grad_norm": 2.765089273452759,
"learning_rate": 3.3040480427046265e-05,
"loss": 1.3808,
"num_input_tokens_seen": 11172032,
"step": 30500
},
{
"epoch": 0.6895017793594306,
"grad_norm": 2.808806896209717,
"learning_rate": 3.2762455516014234e-05,
"loss": 1.3925,
"num_input_tokens_seen": 11356608,
"step": 31000
},
{
"epoch": 0.7006227758007118,
"grad_norm": 2.6583220958709717,
"learning_rate": 3.248443060498221e-05,
"loss": 1.3716,
"num_input_tokens_seen": 11538056,
"step": 31500
},
{
"epoch": 0.7117437722419929,
"grad_norm": 2.2725088596343994,
"learning_rate": 3.2206405693950184e-05,
"loss": 1.3689,
"num_input_tokens_seen": 11721264,
"step": 32000
},
{
"epoch": 0.722864768683274,
"grad_norm": 2.927656412124634,
"learning_rate": 3.1928380782918146e-05,
"loss": 1.3722,
"num_input_tokens_seen": 11896688,
"step": 32500
},
{
"epoch": 0.7339857651245552,
"grad_norm": 2.0601186752319336,
"learning_rate": 3.165035587188612e-05,
"loss": 1.3408,
"num_input_tokens_seen": 12084440,
"step": 33000
},
{
"epoch": 0.7451067615658363,
"grad_norm": 2.5769150257110596,
"learning_rate": 3.13723309608541e-05,
"loss": 1.3874,
"num_input_tokens_seen": 12264224,
"step": 33500
},
{
"epoch": 0.7562277580071174,
"grad_norm": 2.845653772354126,
"learning_rate": 3.1094306049822066e-05,
"loss": 1.3755,
"num_input_tokens_seen": 12446200,
"step": 34000
},
{
"epoch": 0.7673487544483986,
"grad_norm": 2.3848676681518555,
"learning_rate": 3.0816281138790034e-05,
"loss": 1.3463,
"num_input_tokens_seen": 12628992,
"step": 34500
},
{
"epoch": 0.7784697508896797,
"grad_norm": 3.2360849380493164,
"learning_rate": 3.053825622775801e-05,
"loss": 1.3678,
"num_input_tokens_seen": 12809808,
"step": 35000
},
{
"epoch": 0.7895907473309609,
"grad_norm": 2.3211023807525635,
"learning_rate": 3.026023131672598e-05,
"loss": 1.3732,
"num_input_tokens_seen": 12989912,
"step": 35500
},
{
"epoch": 0.800711743772242,
"grad_norm": 3.599958658218384,
"learning_rate": 2.9982206405693954e-05,
"loss": 1.3606,
"num_input_tokens_seen": 13170560,
"step": 36000
},
{
"epoch": 0.8118327402135231,
"grad_norm": 2.0861263275146484,
"learning_rate": 2.9704181494661926e-05,
"loss": 1.3475,
"num_input_tokens_seen": 13350424,
"step": 36500
},
{
"epoch": 0.8229537366548043,
"grad_norm": 2.043938159942627,
"learning_rate": 2.9426156583629895e-05,
"loss": 1.3753,
"num_input_tokens_seen": 13538640,
"step": 37000
},
{
"epoch": 0.8340747330960854,
"grad_norm": 2.4880995750427246,
"learning_rate": 2.9148131672597867e-05,
"loss": 1.3497,
"num_input_tokens_seen": 13730096,
"step": 37500
},
{
"epoch": 0.8451957295373665,
"grad_norm": 2.535860300064087,
"learning_rate": 2.8870106761565835e-05,
"loss": 1.3708,
"num_input_tokens_seen": 13909256,
"step": 38000
},
{
"epoch": 0.8563167259786477,
"grad_norm": 2.499455213546753,
"learning_rate": 2.8592081850533807e-05,
"loss": 1.3546,
"num_input_tokens_seen": 14101536,
"step": 38500
},
{
"epoch": 0.8674377224199288,
"grad_norm": 2.3696117401123047,
"learning_rate": 2.8314056939501783e-05,
"loss": 1.3593,
"num_input_tokens_seen": 14281392,
"step": 39000
},
{
"epoch": 0.87855871886121,
"grad_norm": 3.260430097579956,
"learning_rate": 2.803603202846975e-05,
"loss": 1.3506,
"num_input_tokens_seen": 14466320,
"step": 39500
},
{
"epoch": 0.8896797153024911,
"grad_norm": 4.11997652053833,
"learning_rate": 2.7758007117437723e-05,
"loss": 1.3595,
"num_input_tokens_seen": 14653328,
"step": 40000
},
{
"epoch": 0.8896797153024911,
"eval_loss": 1.1787019968032837,
"eval_runtime": 2.9698,
"eval_samples_per_second": 841.809,
"eval_steps_per_second": 105.394,
"num_input_tokens_seen": 14653328,
"step": 40000
},
{
"epoch": 0.9008007117437722,
"grad_norm": 1.932468056678772,
"learning_rate": 2.7479982206405695e-05,
"loss": 1.3559,
"num_input_tokens_seen": 14837624,
"step": 40500
},
{
"epoch": 0.9119217081850534,
"grad_norm": 2.6026477813720703,
"learning_rate": 2.7201957295373664e-05,
"loss": 1.3454,
"num_input_tokens_seen": 15020192,
"step": 41000
},
{
"epoch": 0.9230427046263345,
"grad_norm": 2.3455870151519775,
"learning_rate": 2.692393238434164e-05,
"loss": 1.3563,
"num_input_tokens_seen": 15204288,
"step": 41500
},
{
"epoch": 0.9341637010676157,
"grad_norm": 2.8757784366607666,
"learning_rate": 2.664590747330961e-05,
"loss": 1.3283,
"num_input_tokens_seen": 15393744,
"step": 42000
},
{
"epoch": 0.9452846975088968,
"grad_norm": 2.3972697257995605,
"learning_rate": 2.636788256227758e-05,
"loss": 1.3612,
"num_input_tokens_seen": 15569824,
"step": 42500
},
{
"epoch": 0.9564056939501779,
"grad_norm": 3.187290906906128,
"learning_rate": 2.6089857651245552e-05,
"loss": 1.3525,
"num_input_tokens_seen": 15753768,
"step": 43000
},
{
"epoch": 0.9675266903914591,
"grad_norm": 2.447659969329834,
"learning_rate": 2.5811832740213527e-05,
"loss": 1.3532,
"num_input_tokens_seen": 15934952,
"step": 43500
},
{
"epoch": 0.9786476868327402,
"grad_norm": 2.037935495376587,
"learning_rate": 2.5533807829181493e-05,
"loss": 1.3318,
"num_input_tokens_seen": 16117896,
"step": 44000
},
{
"epoch": 0.9897686832740213,
"grad_norm": 2.7559268474578857,
"learning_rate": 2.5255782918149468e-05,
"loss": 1.3325,
"num_input_tokens_seen": 16298928,
"step": 44500
},
{
"epoch": 1.0008896797153024,
"grad_norm": 2.2017595767974854,
"learning_rate": 2.4977758007117437e-05,
"loss": 1.3605,
"num_input_tokens_seen": 16481072,
"step": 45000
},
{
"epoch": 1.0120106761565837,
"grad_norm": 2.3097991943359375,
"learning_rate": 2.4699733096085412e-05,
"loss": 1.3404,
"num_input_tokens_seen": 16664336,
"step": 45500
},
{
"epoch": 1.0231316725978647,
"grad_norm": 2.6227993965148926,
"learning_rate": 2.4421708185053384e-05,
"loss": 1.3235,
"num_input_tokens_seen": 16845800,
"step": 46000
},
{
"epoch": 1.0342526690391458,
"grad_norm": 2.24474835395813,
"learning_rate": 2.4143683274021353e-05,
"loss": 1.3376,
"num_input_tokens_seen": 17029008,
"step": 46500
},
{
"epoch": 1.045373665480427,
"grad_norm": 2.7171192169189453,
"learning_rate": 2.3865658362989325e-05,
"loss": 1.3188,
"num_input_tokens_seen": 17215936,
"step": 47000
},
{
"epoch": 1.0564946619217082,
"grad_norm": 2.5323736667633057,
"learning_rate": 2.3587633451957297e-05,
"loss": 1.3369,
"num_input_tokens_seen": 17401232,
"step": 47500
},
{
"epoch": 1.0676156583629894,
"grad_norm": 2.267789363861084,
"learning_rate": 2.330960854092527e-05,
"loss": 1.3215,
"num_input_tokens_seen": 17583344,
"step": 48000
},
{
"epoch": 1.0787366548042705,
"grad_norm": 3.399862289428711,
"learning_rate": 2.3031583629893237e-05,
"loss": 1.3464,
"num_input_tokens_seen": 17771600,
"step": 48500
},
{
"epoch": 1.0898576512455516,
"grad_norm": 2.8749985694885254,
"learning_rate": 2.2753558718861213e-05,
"loss": 1.2928,
"num_input_tokens_seen": 17957200,
"step": 49000
},
{
"epoch": 1.1009786476868326,
"grad_norm": 2.6826517581939697,
"learning_rate": 2.247553380782918e-05,
"loss": 1.3191,
"num_input_tokens_seen": 18138560,
"step": 49500
},
{
"epoch": 1.112099644128114,
"grad_norm": 2.2963333129882812,
"learning_rate": 2.2197508896797153e-05,
"loss": 1.3059,
"num_input_tokens_seen": 18326104,
"step": 50000
},
{
"epoch": 1.112099644128114,
"eval_loss": 1.1665468215942383,
"eval_runtime": 3.0497,
"eval_samples_per_second": 819.748,
"eval_steps_per_second": 102.632,
"num_input_tokens_seen": 18326104,
"step": 50000
},
{
"epoch": 1.123220640569395,
"grad_norm": 2.2697386741638184,
"learning_rate": 2.1919483985765125e-05,
"loss": 1.321,
"num_input_tokens_seen": 18508560,
"step": 50500
},
{
"epoch": 1.1343416370106763,
"grad_norm": 2.436851739883423,
"learning_rate": 2.1641459074733097e-05,
"loss": 1.3528,
"num_input_tokens_seen": 18689264,
"step": 51000
},
{
"epoch": 1.1454626334519573,
"grad_norm": 2.297527313232422,
"learning_rate": 2.136343416370107e-05,
"loss": 1.2987,
"num_input_tokens_seen": 18874328,
"step": 51500
},
{
"epoch": 1.1565836298932384,
"grad_norm": 2.5088889598846436,
"learning_rate": 2.1085409252669038e-05,
"loss": 1.3139,
"num_input_tokens_seen": 19060472,
"step": 52000
},
{
"epoch": 1.1677046263345197,
"grad_norm": 2.067575454711914,
"learning_rate": 2.0807384341637014e-05,
"loss": 1.2961,
"num_input_tokens_seen": 19247416,
"step": 52500
},
{
"epoch": 1.1788256227758007,
"grad_norm": 2.467543363571167,
"learning_rate": 2.0529359430604982e-05,
"loss": 1.3279,
"num_input_tokens_seen": 19436888,
"step": 53000
},
{
"epoch": 1.1899466192170818,
"grad_norm": 3.4245800971984863,
"learning_rate": 2.0251334519572954e-05,
"loss": 1.3303,
"num_input_tokens_seen": 19616320,
"step": 53500
},
{
"epoch": 1.201067615658363,
"grad_norm": 2.759120464324951,
"learning_rate": 1.9973309608540926e-05,
"loss": 1.3201,
"num_input_tokens_seen": 19793576,
"step": 54000
},
{
"epoch": 1.2121886120996441,
"grad_norm": 2.7749531269073486,
"learning_rate": 1.9695284697508898e-05,
"loss": 1.3194,
"num_input_tokens_seen": 19980880,
"step": 54500
},
{
"epoch": 1.2233096085409252,
"grad_norm": 2.4467661380767822,
"learning_rate": 1.9417259786476867e-05,
"loss": 1.3293,
"num_input_tokens_seen": 20163048,
"step": 55000
},
{
"epoch": 1.2344306049822065,
"grad_norm": 3.4420840740203857,
"learning_rate": 1.9139234875444842e-05,
"loss": 1.3015,
"num_input_tokens_seen": 20340904,
"step": 55500
},
{
"epoch": 1.2455516014234875,
"grad_norm": 2.4761664867401123,
"learning_rate": 1.8861209964412814e-05,
"loss": 1.3292,
"num_input_tokens_seen": 20524328,
"step": 56000
},
{
"epoch": 1.2566725978647688,
"grad_norm": 3.0505285263061523,
"learning_rate": 1.8583185053380783e-05,
"loss": 1.317,
"num_input_tokens_seen": 20707808,
"step": 56500
},
{
"epoch": 1.2677935943060499,
"grad_norm": 2.361429214477539,
"learning_rate": 1.8305160142348755e-05,
"loss": 1.3176,
"num_input_tokens_seen": 20892744,
"step": 57000
},
{
"epoch": 1.278914590747331,
"grad_norm": 1.9151511192321777,
"learning_rate": 1.8027135231316727e-05,
"loss": 1.3267,
"num_input_tokens_seen": 21072216,
"step": 57500
},
{
"epoch": 1.290035587188612,
"grad_norm": 3.0513691902160645,
"learning_rate": 1.77491103202847e-05,
"loss": 1.2948,
"num_input_tokens_seen": 21256064,
"step": 58000
},
{
"epoch": 1.3011565836298933,
"grad_norm": 1.7151504755020142,
"learning_rate": 1.7471085409252668e-05,
"loss": 1.2967,
"num_input_tokens_seen": 21439648,
"step": 58500
},
{
"epoch": 1.3122775800711743,
"grad_norm": 2.3011133670806885,
"learning_rate": 1.7193060498220643e-05,
"loss": 1.3199,
"num_input_tokens_seen": 21621776,
"step": 59000
},
{
"epoch": 1.3233985765124556,
"grad_norm": 2.1768672466278076,
"learning_rate": 1.691503558718861e-05,
"loss": 1.3233,
"num_input_tokens_seen": 21810728,
"step": 59500
},
{
"epoch": 1.3345195729537367,
"grad_norm": 3.249089241027832,
"learning_rate": 1.6637010676156584e-05,
"loss": 1.3298,
"num_input_tokens_seen": 21991016,
"step": 60000
},
{
"epoch": 1.3345195729537367,
"eval_loss": 1.1589475870132446,
"eval_runtime": 3.0361,
"eval_samples_per_second": 823.416,
"eval_steps_per_second": 103.092,
"num_input_tokens_seen": 21991016,
"step": 60000
},
{
"epoch": 1.3456405693950177,
"grad_norm": 1.7282594442367554,
"learning_rate": 1.6358985765124556e-05,
"loss": 1.325,
"num_input_tokens_seen": 22178168,
"step": 60500
},
{
"epoch": 1.3567615658362988,
"grad_norm": 2.2659966945648193,
"learning_rate": 1.6080960854092528e-05,
"loss": 1.3321,
"num_input_tokens_seen": 22359360,
"step": 61000
},
{
"epoch": 1.36788256227758,
"grad_norm": 2.155791759490967,
"learning_rate": 1.58029359430605e-05,
"loss": 1.2995,
"num_input_tokens_seen": 22543952,
"step": 61500
},
{
"epoch": 1.3790035587188612,
"grad_norm": 2.902367353439331,
"learning_rate": 1.5524911032028472e-05,
"loss": 1.3061,
"num_input_tokens_seen": 22728064,
"step": 62000
},
{
"epoch": 1.3901245551601424,
"grad_norm": 2.168686866760254,
"learning_rate": 1.5246886120996442e-05,
"loss": 1.3231,
"num_input_tokens_seen": 22913608,
"step": 62500
},
{
"epoch": 1.4012455516014235,
"grad_norm": 2.682150363922119,
"learning_rate": 1.4968861209964412e-05,
"loss": 1.3091,
"num_input_tokens_seen": 23101416,
"step": 63000
},
{
"epoch": 1.4123665480427046,
"grad_norm": 2.82143497467041,
"learning_rate": 1.4690836298932384e-05,
"loss": 1.2922,
"num_input_tokens_seen": 23281200,
"step": 63500
},
{
"epoch": 1.4234875444839858,
"grad_norm": 2.1155049800872803,
"learning_rate": 1.4412811387900358e-05,
"loss": 1.3015,
"num_input_tokens_seen": 23462552,
"step": 64000
},
{
"epoch": 1.434608540925267,
"grad_norm": 2.6742069721221924,
"learning_rate": 1.4134786476868328e-05,
"loss": 1.3108,
"num_input_tokens_seen": 23640424,
"step": 64500
},
{
"epoch": 1.445729537366548,
"grad_norm": 2.575198173522949,
"learning_rate": 1.3856761565836299e-05,
"loss": 1.2987,
"num_input_tokens_seen": 23825392,
"step": 65000
},
{
"epoch": 1.4568505338078293,
"grad_norm": 3.2627415657043457,
"learning_rate": 1.3578736654804272e-05,
"loss": 1.3096,
"num_input_tokens_seen": 24004872,
"step": 65500
},
{
"epoch": 1.4679715302491103,
"grad_norm": 3.2727510929107666,
"learning_rate": 1.3300711743772243e-05,
"loss": 1.3108,
"num_input_tokens_seen": 24186064,
"step": 66000
},
{
"epoch": 1.4790925266903914,
"grad_norm": 2.187281608581543,
"learning_rate": 1.3022686832740213e-05,
"loss": 1.3239,
"num_input_tokens_seen": 24367968,
"step": 66500
},
{
"epoch": 1.4902135231316727,
"grad_norm": 3.5734856128692627,
"learning_rate": 1.2744661921708187e-05,
"loss": 1.3241,
"num_input_tokens_seen": 24551184,
"step": 67000
},
{
"epoch": 1.5013345195729537,
"grad_norm": 2.756578207015991,
"learning_rate": 1.2466637010676157e-05,
"loss": 1.307,
"num_input_tokens_seen": 24730288,
"step": 67500
},
{
"epoch": 1.512455516014235,
"grad_norm": 2.3752739429473877,
"learning_rate": 1.2188612099644127e-05,
"loss": 1.2963,
"num_input_tokens_seen": 24914816,
"step": 68000
},
{
"epoch": 1.523576512455516,
"grad_norm": 3.7108139991760254,
"learning_rate": 1.1910587188612101e-05,
"loss": 1.3126,
"num_input_tokens_seen": 25100648,
"step": 68500
},
{
"epoch": 1.5346975088967971,
"grad_norm": 2.5954089164733887,
"learning_rate": 1.1632562277580072e-05,
"loss": 1.3053,
"num_input_tokens_seen": 25282312,
"step": 69000
},
{
"epoch": 1.5458185053380782,
"grad_norm": 3.0645289421081543,
"learning_rate": 1.1354537366548044e-05,
"loss": 1.2976,
"num_input_tokens_seen": 25465504,
"step": 69500
},
{
"epoch": 1.5569395017793595,
"grad_norm": 2.3734166622161865,
"learning_rate": 1.1076512455516016e-05,
"loss": 1.2994,
"num_input_tokens_seen": 25656600,
"step": 70000
},
{
"epoch": 1.5569395017793595,
"eval_loss": 1.1562061309814453,
"eval_runtime": 2.9007,
"eval_samples_per_second": 861.851,
"eval_steps_per_second": 107.904,
"num_input_tokens_seen": 25656600,
"step": 70000
},
{
"epoch": 1.5680604982206405,
"grad_norm": 2.8284573554992676,
"learning_rate": 1.0798487544483986e-05,
"loss": 1.3125,
"num_input_tokens_seen": 25844216,
"step": 70500
},
{
"epoch": 1.5791814946619218,
"grad_norm": 1.8697574138641357,
"learning_rate": 1.0520462633451958e-05,
"loss": 1.3247,
"num_input_tokens_seen": 26026824,
"step": 71000
},
{
"epoch": 1.5903024911032029,
"grad_norm": 2.4746646881103516,
"learning_rate": 1.024243772241993e-05,
"loss": 1.3295,
"num_input_tokens_seen": 26211760,
"step": 71500
},
{
"epoch": 1.601423487544484,
"grad_norm": 2.415778398513794,
"learning_rate": 9.9644128113879e-06,
"loss": 1.3191,
"num_input_tokens_seen": 26393624,
"step": 72000
},
{
"epoch": 1.612544483985765,
"grad_norm": 2.263882875442505,
"learning_rate": 9.686387900355872e-06,
"loss": 1.3136,
"num_input_tokens_seen": 26576872,
"step": 72500
},
{
"epoch": 1.6236654804270463,
"grad_norm": 2.436645269393921,
"learning_rate": 9.408362989323843e-06,
"loss": 1.286,
"num_input_tokens_seen": 26763752,
"step": 73000
},
{
"epoch": 1.6347864768683276,
"grad_norm": 2.739278793334961,
"learning_rate": 9.130338078291816e-06,
"loss": 1.2712,
"num_input_tokens_seen": 26943856,
"step": 73500
},
{
"epoch": 1.6459074733096086,
"grad_norm": 2.406345844268799,
"learning_rate": 8.852313167259788e-06,
"loss": 1.305,
"num_input_tokens_seen": 27122192,
"step": 74000
},
{
"epoch": 1.6570284697508897,
"grad_norm": 2.1659858226776123,
"learning_rate": 8.574288256227759e-06,
"loss": 1.312,
"num_input_tokens_seen": 27305912,
"step": 74500
},
{
"epoch": 1.6681494661921707,
"grad_norm": 2.7831106185913086,
"learning_rate": 8.29626334519573e-06,
"loss": 1.2997,
"num_input_tokens_seen": 27485504,
"step": 75000
},
{
"epoch": 1.6792704626334518,
"grad_norm": 2.9885916709899902,
"learning_rate": 8.018238434163701e-06,
"loss": 1.3187,
"num_input_tokens_seen": 27665584,
"step": 75500
},
{
"epoch": 1.690391459074733,
"grad_norm": 2.5562667846679688,
"learning_rate": 7.740213523131673e-06,
"loss": 1.3142,
"num_input_tokens_seen": 27851448,
"step": 76000
},
{
"epoch": 1.7015124555160144,
"grad_norm": 2.9897525310516357,
"learning_rate": 7.462188612099645e-06,
"loss": 1.3071,
"num_input_tokens_seen": 28033824,
"step": 76500
},
{
"epoch": 1.7126334519572954,
"grad_norm": 2.6760592460632324,
"learning_rate": 7.184163701067615e-06,
"loss": 1.3049,
"num_input_tokens_seen": 28220000,
"step": 77000
},
{
"epoch": 1.7237544483985765,
"grad_norm": 2.314532995223999,
"learning_rate": 6.906138790035588e-06,
"loss": 1.3124,
"num_input_tokens_seen": 28403400,
"step": 77500
},
{
"epoch": 1.7348754448398576,
"grad_norm": 2.2853899002075195,
"learning_rate": 6.6281138790035586e-06,
"loss": 1.3188,
"num_input_tokens_seen": 28585168,
"step": 78000
},
{
"epoch": 1.7459964412811388,
"grad_norm": 2.462369918823242,
"learning_rate": 6.3500889679715306e-06,
"loss": 1.3146,
"num_input_tokens_seen": 28767960,
"step": 78500
},
{
"epoch": 1.75711743772242,
"grad_norm": 2.626847505569458,
"learning_rate": 6.072064056939502e-06,
"loss": 1.3082,
"num_input_tokens_seen": 28950344,
"step": 79000
},
{
"epoch": 1.7682384341637012,
"grad_norm": 3.058187484741211,
"learning_rate": 5.794039145907473e-06,
"loss": 1.301,
"num_input_tokens_seen": 29133248,
"step": 79500
},
{
"epoch": 1.7793594306049823,
"grad_norm": 1.932924747467041,
"learning_rate": 5.516014234875446e-06,
"loss": 1.2952,
"num_input_tokens_seen": 29314808,
"step": 80000
},
{
"epoch": 1.7793594306049823,
"eval_loss": 1.1517876386642456,
"eval_runtime": 3.0679,
"eval_samples_per_second": 814.88,
"eval_steps_per_second": 102.023,
"num_input_tokens_seen": 29314808,
"step": 80000
},
{
"epoch": 1.7904804270462633,
"grad_norm": 3.326586961746216,
"learning_rate": 5.237989323843417e-06,
"loss": 1.3161,
"num_input_tokens_seen": 29504040,
"step": 80500
},
{
"epoch": 1.8016014234875444,
"grad_norm": 3.139636754989624,
"learning_rate": 4.959964412811388e-06,
"loss": 1.3,
"num_input_tokens_seen": 29686712,
"step": 81000
},
{
"epoch": 1.8127224199288257,
"grad_norm": 2.5767743587493896,
"learning_rate": 4.681939501779359e-06,
"loss": 1.2934,
"num_input_tokens_seen": 29868136,
"step": 81500
},
{
"epoch": 1.8238434163701067,
"grad_norm": 2.1930339336395264,
"learning_rate": 4.4039145907473305e-06,
"loss": 1.3104,
"num_input_tokens_seen": 30050016,
"step": 82000
},
{
"epoch": 1.834964412811388,
"grad_norm": 2.9890389442443848,
"learning_rate": 4.125889679715303e-06,
"loss": 1.303,
"num_input_tokens_seen": 30234304,
"step": 82500
},
{
"epoch": 1.846085409252669,
"grad_norm": 2.9597034454345703,
"learning_rate": 3.8478647686832745e-06,
"loss": 1.3024,
"num_input_tokens_seen": 30419248,
"step": 83000
},
{
"epoch": 1.8572064056939501,
"grad_norm": 3.17301082611084,
"learning_rate": 3.5698398576512457e-06,
"loss": 1.2926,
"num_input_tokens_seen": 30601776,
"step": 83500
},
{
"epoch": 1.8683274021352312,
"grad_norm": 2.677340269088745,
"learning_rate": 3.291814946619217e-06,
"loss": 1.3184,
"num_input_tokens_seen": 30783800,
"step": 84000
},
{
"epoch": 1.8794483985765125,
"grad_norm": 3.5369062423706055,
"learning_rate": 3.013790035587189e-06,
"loss": 1.2858,
"num_input_tokens_seen": 30968472,
"step": 84500
},
{
"epoch": 1.8905693950177938,
"grad_norm": 2.870908737182617,
"learning_rate": 2.73576512455516e-06,
"loss": 1.3107,
"num_input_tokens_seen": 31147072,
"step": 85000
},
{
"epoch": 1.9016903914590748,
"grad_norm": 2.1105244159698486,
"learning_rate": 2.457740213523132e-06,
"loss": 1.2797,
"num_input_tokens_seen": 31330536,
"step": 85500
},
{
"epoch": 1.9128113879003559,
"grad_norm": 4.080565452575684,
"learning_rate": 2.1797153024911032e-06,
"loss": 1.2991,
"num_input_tokens_seen": 31507752,
"step": 86000
},
{
"epoch": 1.923932384341637,
"grad_norm": 3.036339044570923,
"learning_rate": 1.901690391459075e-06,
"loss": 1.2884,
"num_input_tokens_seen": 31685504,
"step": 86500
},
{
"epoch": 1.935053380782918,
"grad_norm": 2.3314597606658936,
"learning_rate": 1.6236654804270462e-06,
"loss": 1.2935,
"num_input_tokens_seen": 31865648,
"step": 87000
},
{
"epoch": 1.9461743772241993,
"grad_norm": 2.4721710681915283,
"learning_rate": 1.3456405693950178e-06,
"loss": 1.3124,
"num_input_tokens_seen": 32046704,
"step": 87500
},
{
"epoch": 1.9572953736654806,
"grad_norm": 2.4747238159179688,
"learning_rate": 1.0676156583629894e-06,
"loss": 1.2941,
"num_input_tokens_seen": 32235800,
"step": 88000
},
{
"epoch": 1.9684163701067616,
"grad_norm": 2.4421122074127197,
"learning_rate": 7.89590747330961e-07,
"loss": 1.3187,
"num_input_tokens_seen": 32415672,
"step": 88500
},
{
"epoch": 1.9795373665480427,
"grad_norm": 3.469534158706665,
"learning_rate": 5.115658362989324e-07,
"loss": 1.2897,
"num_input_tokens_seen": 32597136,
"step": 89000
},
{
"epoch": 1.9906583629893237,
"grad_norm": 3.1344103813171387,
"learning_rate": 2.335409252669039e-07,
"loss": 1.2902,
"num_input_tokens_seen": 32782688,
"step": 89500
},
{
"epoch": 2.0,
"num_input_tokens_seen": 32939232,
"step": 89920,
"total_flos": 1.1959161056722944e+16,
"train_loss": 1.358397777055082,
"train_runtime": 3698.2072,
"train_samples_per_second": 194.516,
"train_steps_per_second": 24.314,
"train_tokens_per_second": 8905.5
}
],
"logging_steps": 500,
"max_steps": 89920,
"num_input_tokens_seen": 32939232,
"num_train_epochs": 2,
"save_steps": 10000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.1959161056722944e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}