Download checkpoint-6076/trainer_state.json from yacineai7/sqlcoder: direct link, hf CLI and curl.
- Browser
- Download file 357 kB
-
https://huggingface.co/yacineai7/sqlcoder/resolve/main/checkpoint-6076/trainer_state.json
- Command line
-
hf download hf://yacineai7/sqlcoder/checkpoint-6076/trainer_state.json
-
curl -L -o trainer_state.json https://huggingface.co/yacineai7/sqlcoder/resolve/main/checkpoint-6076/trainer_state.json
357 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 608,
"global_step": 6076,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0,
"eval_entropy": 2.563803545593872,
"eval_loss": 1.3544182777404785,
"eval_mean_token_accuracy": 0.687368397152733,
"eval_num_tokens": 0.0,
"eval_runtime": 198.6529,
"eval_samples_per_second": 41.917,
"eval_steps_per_second": 6.987,
"step": 0
},
{
"entropy": 2.578546404838562,
"epoch": 0.0003291639236339697,
"grad_norm": 58.16326141357422,
"learning_rate": 0.0,
"loss": 1.4746,
"mean_token_accuracy": 0.6601307392120361,
"num_tokens": 2101.0,
"step": 1
},
{
"entropy": 2.547638326883316,
"epoch": 0.0016458196181698486,
"grad_norm": 50.99738311767578,
"learning_rate": 2.6315789473684213e-07,
"loss": 1.2383,
"mean_token_accuracy": 0.7003932222723961,
"num_tokens": 10649.0,
"step": 5
},
{
"entropy": 2.5566530227661133,
"epoch": 0.0032916392363396972,
"grad_norm": 38.35203552246094,
"learning_rate": 5.921052631578947e-07,
"loss": 1.0069,
"mean_token_accuracy": 0.7398434042930603,
"num_tokens": 21271.0,
"step": 10
},
{
"entropy": 2.452682948112488,
"epoch": 0.004937458854509546,
"grad_norm": 27.224767684936523,
"learning_rate": 9.210526315789474e-07,
"loss": 0.5295,
"mean_token_accuracy": 0.8550694227218628,
"num_tokens": 32112.0,
"step": 15
},
{
"entropy": 2.1593106389045715,
"epoch": 0.0065832784726793945,
"grad_norm": 23.221702575683594,
"learning_rate": 1.25e-06,
"loss": 0.2773,
"mean_token_accuracy": 0.9244555592536926,
"num_tokens": 42919.0,
"step": 20
},
{
"entropy": 1.6036222577095032,
"epoch": 0.008229098090849244,
"grad_norm": 15.987848281860352,
"learning_rate": 1.5789473684210526e-06,
"loss": 0.1681,
"mean_token_accuracy": 0.9467954277992249,
"num_tokens": 53618.0,
"step": 25
},
{
"entropy": 1.486020290851593,
"epoch": 0.009874917709019092,
"grad_norm": 11.537003517150879,
"learning_rate": 1.9078947368421057e-06,
"loss": 0.1446,
"mean_token_accuracy": 0.9633475482463837,
"num_tokens": 64755.0,
"step": 30
},
{
"entropy": 1.6325503945350648,
"epoch": 0.01152073732718894,
"grad_norm": 7.058774471282959,
"learning_rate": 2.236842105263158e-06,
"loss": 0.149,
"mean_token_accuracy": 0.9636693239212036,
"num_tokens": 75388.0,
"step": 35
},
{
"entropy": 1.7897157430648805,
"epoch": 0.013166556945358789,
"grad_norm": 6.356208324432373,
"learning_rate": 2.565789473684211e-06,
"loss": 0.0838,
"mean_token_accuracy": 0.9791653394699097,
"num_tokens": 86153.0,
"step": 40
},
{
"entropy": 1.7598371386528016,
"epoch": 0.014812376563528637,
"grad_norm": 11.368063926696777,
"learning_rate": 2.8947368421052634e-06,
"loss": 0.0981,
"mean_token_accuracy": 0.9723430037498474,
"num_tokens": 96649.0,
"step": 45
},
{
"entropy": 1.5412654876708984,
"epoch": 0.016458196181698487,
"grad_norm": 11.02826976776123,
"learning_rate": 3.223684210526316e-06,
"loss": 0.0802,
"mean_token_accuracy": 0.9778924286365509,
"num_tokens": 107209.0,
"step": 50
},
{
"entropy": 1.419243288040161,
"epoch": 0.018104015799868336,
"grad_norm": 14.368644714355469,
"learning_rate": 3.5526315789473687e-06,
"loss": 0.1422,
"mean_token_accuracy": 0.9643596112728119,
"num_tokens": 117665.0,
"step": 55
},
{
"entropy": 1.5144587397575378,
"epoch": 0.019749835418038184,
"grad_norm": 8.765934944152832,
"learning_rate": 3.8815789473684214e-06,
"loss": 0.1094,
"mean_token_accuracy": 0.9742422461509704,
"num_tokens": 128149.0,
"step": 60
},
{
"entropy": 1.6769550323486329,
"epoch": 0.021395655036208033,
"grad_norm": 9.26172161102295,
"learning_rate": 4.210526315789474e-06,
"loss": 0.0886,
"mean_token_accuracy": 0.9756901502609253,
"num_tokens": 138845.0,
"step": 65
},
{
"entropy": 1.804041564464569,
"epoch": 0.02304147465437788,
"grad_norm": 7.930450439453125,
"learning_rate": 4.539473684210527e-06,
"loss": 0.1117,
"mean_token_accuracy": 0.9712024033069611,
"num_tokens": 149695.0,
"step": 70
},
{
"entropy": 1.8376032829284668,
"epoch": 0.02468729427254773,
"grad_norm": 5.366224765777588,
"learning_rate": 4.8684210526315795e-06,
"loss": 0.0885,
"mean_token_accuracy": 0.9742565512657165,
"num_tokens": 160769.0,
"step": 75
},
{
"entropy": 1.8037242650985719,
"epoch": 0.026333113890717578,
"grad_norm": 50.815311431884766,
"learning_rate": 5.197368421052632e-06,
"loss": 0.0664,
"mean_token_accuracy": 0.9830366194248199,
"num_tokens": 171557.0,
"step": 80
},
{
"entropy": 1.7092326641082765,
"epoch": 0.027978933508887426,
"grad_norm": 22.222219467163086,
"learning_rate": 5.526315789473685e-06,
"loss": 0.0962,
"mean_token_accuracy": 0.9750658094882965,
"num_tokens": 182226.0,
"step": 85
},
{
"entropy": 1.6772519588470458,
"epoch": 0.029624753127057275,
"grad_norm": 7.502695560455322,
"learning_rate": 5.855263157894738e-06,
"loss": 0.1128,
"mean_token_accuracy": 0.9759301841259003,
"num_tokens": 193341.0,
"step": 90
},
{
"entropy": 1.8118021130561828,
"epoch": 0.031270572745227126,
"grad_norm": 4.66135311126709,
"learning_rate": 6.18421052631579e-06,
"loss": 0.0788,
"mean_token_accuracy": 0.9772343516349793,
"num_tokens": 204280.0,
"step": 95
},
{
"entropy": 1.7658962607383728,
"epoch": 0.032916392363396975,
"grad_norm": 9.973278045654297,
"learning_rate": 6.513157894736842e-06,
"loss": 0.0693,
"mean_token_accuracy": 0.9800639986991883,
"num_tokens": 214849.0,
"step": 100
},
{
"entropy": 1.6705022692680358,
"epoch": 0.03456221198156682,
"grad_norm": 12.13133430480957,
"learning_rate": 6.842105263157896e-06,
"loss": 0.1029,
"mean_token_accuracy": 0.9713725507259369,
"num_tokens": 225617.0,
"step": 105
},
{
"entropy": 1.6237557530403137,
"epoch": 0.03620803159973667,
"grad_norm": 12.934727668762207,
"learning_rate": 7.1710526315789475e-06,
"loss": 0.092,
"mean_token_accuracy": 0.9715362131595612,
"num_tokens": 236011.0,
"step": 110
},
{
"entropy": 1.6253692150115966,
"epoch": 0.03785385121790652,
"grad_norm": 5.850583076477051,
"learning_rate": 7.500000000000001e-06,
"loss": 0.0856,
"mean_token_accuracy": 0.9789658904075622,
"num_tokens": 246517.0,
"step": 115
},
{
"entropy": 1.699216902256012,
"epoch": 0.03949967083607637,
"grad_norm": 9.97748851776123,
"learning_rate": 7.828947368421054e-06,
"loss": 0.1076,
"mean_token_accuracy": 0.9751021385192871,
"num_tokens": 257039.0,
"step": 120
},
{
"entropy": 1.7488463997840882,
"epoch": 0.04114549045424622,
"grad_norm": 9.265748023986816,
"learning_rate": 8.157894736842106e-06,
"loss": 0.0841,
"mean_token_accuracy": 0.9726714611053466,
"num_tokens": 267666.0,
"step": 125
},
{
"entropy": 1.7969783902168275,
"epoch": 0.042791310072416065,
"grad_norm": 6.479616641998291,
"learning_rate": 8.486842105263159e-06,
"loss": 0.0759,
"mean_token_accuracy": 0.9793728470802308,
"num_tokens": 278047.0,
"step": 130
},
{
"entropy": 1.6162196516990661,
"epoch": 0.044437129690585914,
"grad_norm": 8.1745023727417,
"learning_rate": 8.81578947368421e-06,
"loss": 0.0684,
"mean_token_accuracy": 0.9798145830631256,
"num_tokens": 288638.0,
"step": 135
},
{
"entropy": 1.603439712524414,
"epoch": 0.04608294930875576,
"grad_norm": 11.199191093444824,
"learning_rate": 9.144736842105264e-06,
"loss": 0.1103,
"mean_token_accuracy": 0.973408317565918,
"num_tokens": 299280.0,
"step": 140
},
{
"entropy": 1.7563074707984925,
"epoch": 0.04772876892692561,
"grad_norm": 7.97125244140625,
"learning_rate": 9.473684210526315e-06,
"loss": 0.1014,
"mean_token_accuracy": 0.9699687659740448,
"num_tokens": 309815.0,
"step": 145
},
{
"entropy": 1.954578173160553,
"epoch": 0.04937458854509546,
"grad_norm": 5.959860324859619,
"learning_rate": 9.80263157894737e-06,
"loss": 0.0905,
"mean_token_accuracy": 0.9725431621074676,
"num_tokens": 320598.0,
"step": 150
},
{
"entropy": 1.997165060043335,
"epoch": 0.05102040816326531,
"grad_norm": 4.954457759857178,
"learning_rate": 1.0131578947368421e-05,
"loss": 0.0671,
"mean_token_accuracy": 0.983484423160553,
"num_tokens": 331592.0,
"step": 155
},
{
"entropy": 1.9534252166748047,
"epoch": 0.052666227781435156,
"grad_norm": 8.918072700500488,
"learning_rate": 1.0460526315789474e-05,
"loss": 0.0995,
"mean_token_accuracy": 0.9751366376876831,
"num_tokens": 342211.0,
"step": 160
},
{
"entropy": 1.8752238154411316,
"epoch": 0.054312047399605004,
"grad_norm": 8.222981452941895,
"learning_rate": 1.0789473684210528e-05,
"loss": 0.1081,
"mean_token_accuracy": 0.9731775224208832,
"num_tokens": 352894.0,
"step": 165
},
{
"entropy": 1.8922731161117554,
"epoch": 0.05595786701777485,
"grad_norm": 6.892168045043945,
"learning_rate": 1.111842105263158e-05,
"loss": 0.1093,
"mean_token_accuracy": 0.9666784107685089,
"num_tokens": 363571.0,
"step": 170
},
{
"entropy": 1.8998683929443358,
"epoch": 0.0576036866359447,
"grad_norm": 3.256152629852295,
"learning_rate": 1.1447368421052632e-05,
"loss": 0.0701,
"mean_token_accuracy": 0.9812213480472565,
"num_tokens": 374077.0,
"step": 175
},
{
"entropy": 1.8406203866004944,
"epoch": 0.05924950625411455,
"grad_norm": 5.9513983726501465,
"learning_rate": 1.1776315789473684e-05,
"loss": 0.0642,
"mean_token_accuracy": 0.9829223692417145,
"num_tokens": 384936.0,
"step": 180
},
{
"entropy": 1.8397523403167724,
"epoch": 0.0608953258722844,
"grad_norm": 5.223180770874023,
"learning_rate": 1.2105263157894737e-05,
"loss": 0.1066,
"mean_token_accuracy": 0.9780671417713165,
"num_tokens": 395309.0,
"step": 185
},
{
"entropy": 1.8732271909713745,
"epoch": 0.06254114549045425,
"grad_norm": 5.581716060638428,
"learning_rate": 1.2434210526315791e-05,
"loss": 0.1243,
"mean_token_accuracy": 0.9716349482536316,
"num_tokens": 406122.0,
"step": 190
},
{
"entropy": 1.8642462372779847,
"epoch": 0.0641869651086241,
"grad_norm": 4.00003719329834,
"learning_rate": 1.2763157894736844e-05,
"loss": 0.0881,
"mean_token_accuracy": 0.9758350074291229,
"num_tokens": 416622.0,
"step": 195
},
{
"entropy": 1.763167428970337,
"epoch": 0.06583278472679395,
"grad_norm": 4.925802707672119,
"learning_rate": 1.3092105263157895e-05,
"loss": 0.0865,
"mean_token_accuracy": 0.9732065200805664,
"num_tokens": 427374.0,
"step": 200
},
{
"entropy": 1.707397425174713,
"epoch": 0.06747860434496379,
"grad_norm": 3.0551066398620605,
"learning_rate": 1.3421052631578948e-05,
"loss": 0.081,
"mean_token_accuracy": 0.9832678198814392,
"num_tokens": 437906.0,
"step": 205
},
{
"entropy": 1.6210663080215455,
"epoch": 0.06912442396313365,
"grad_norm": 5.318910598754883,
"learning_rate": 1.375e-05,
"loss": 0.0959,
"mean_token_accuracy": 0.978535383939743,
"num_tokens": 448512.0,
"step": 210
},
{
"entropy": 1.7267163753509522,
"epoch": 0.07077024358130349,
"grad_norm": 18.240541458129883,
"learning_rate": 1.4078947368421055e-05,
"loss": 0.0878,
"mean_token_accuracy": 0.9779551386833191,
"num_tokens": 459296.0,
"step": 215
},
{
"entropy": 1.9074785947799682,
"epoch": 0.07241606319947334,
"grad_norm": 4.598775386810303,
"learning_rate": 1.4407894736842108e-05,
"loss": 0.1177,
"mean_token_accuracy": 0.9697455763816833,
"num_tokens": 469857.0,
"step": 220
},
{
"entropy": 2.076517927646637,
"epoch": 0.07406188281764318,
"grad_norm": 6.2975239753723145,
"learning_rate": 1.4736842105263159e-05,
"loss": 0.1191,
"mean_token_accuracy": 0.9704063773155213,
"num_tokens": 480679.0,
"step": 225
},
{
"entropy": 2.075191152095795,
"epoch": 0.07570770243581304,
"grad_norm": 5.611932754516602,
"learning_rate": 1.5065789473684211e-05,
"loss": 0.1126,
"mean_token_accuracy": 0.9733084321022034,
"num_tokens": 491266.0,
"step": 230
},
{
"entropy": 1.8972731590270997,
"epoch": 0.07735352205398288,
"grad_norm": 7.2364888191223145,
"learning_rate": 1.5394736842105264e-05,
"loss": 0.0767,
"mean_token_accuracy": 0.9798118114471436,
"num_tokens": 502150.0,
"step": 235
},
{
"entropy": 1.9421729803085328,
"epoch": 0.07899934167215274,
"grad_norm": 5.142597198486328,
"learning_rate": 1.572368421052632e-05,
"loss": 0.0984,
"mean_token_accuracy": 0.9767365634441376,
"num_tokens": 512761.0,
"step": 240
},
{
"entropy": 2.1024407386779784,
"epoch": 0.08064516129032258,
"grad_norm": 6.71290922164917,
"learning_rate": 1.605263157894737e-05,
"loss": 0.0941,
"mean_token_accuracy": 0.968624371290207,
"num_tokens": 523326.0,
"step": 245
},
{
"entropy": 2.108752429485321,
"epoch": 0.08229098090849243,
"grad_norm": 6.765599727630615,
"learning_rate": 1.638157894736842e-05,
"loss": 0.1249,
"mean_token_accuracy": 0.9727495968341827,
"num_tokens": 534182.0,
"step": 250
},
{
"entropy": 2.1626508831977844,
"epoch": 0.08393680052666228,
"grad_norm": 6.46060848236084,
"learning_rate": 1.6710526315789475e-05,
"loss": 0.0776,
"mean_token_accuracy": 0.9789380371570587,
"num_tokens": 544990.0,
"step": 255
},
{
"entropy": 2.010746192932129,
"epoch": 0.08558262014483213,
"grad_norm": 5.445334434509277,
"learning_rate": 1.703947368421053e-05,
"loss": 0.0855,
"mean_token_accuracy": 0.9773862421512604,
"num_tokens": 555445.0,
"step": 260
},
{
"entropy": 1.7812985062599183,
"epoch": 0.08722843976300197,
"grad_norm": 5.686192512512207,
"learning_rate": 1.736842105263158e-05,
"loss": 0.0891,
"mean_token_accuracy": 0.975337165594101,
"num_tokens": 566278.0,
"step": 265
},
{
"entropy": 1.6327629923820495,
"epoch": 0.08887425938117183,
"grad_norm": 6.93804931640625,
"learning_rate": 1.769736842105263e-05,
"loss": 0.1093,
"mean_token_accuracy": 0.9774836421012878,
"num_tokens": 577347.0,
"step": 270
},
{
"entropy": 1.6540838599205017,
"epoch": 0.09052007899934167,
"grad_norm": 4.582030773162842,
"learning_rate": 1.8026315789473685e-05,
"loss": 0.0712,
"mean_token_accuracy": 0.9831354200839997,
"num_tokens": 587873.0,
"step": 275
},
{
"entropy": 1.614749014377594,
"epoch": 0.09216589861751152,
"grad_norm": 4.63378381729126,
"learning_rate": 1.835526315789474e-05,
"loss": 0.0818,
"mean_token_accuracy": 0.9796861112117767,
"num_tokens": 598584.0,
"step": 280
},
{
"entropy": 1.7127745985984801,
"epoch": 0.09381171823568137,
"grad_norm": 5.947144508361816,
"learning_rate": 1.868421052631579e-05,
"loss": 0.0864,
"mean_token_accuracy": 0.9765213131904602,
"num_tokens": 608860.0,
"step": 285
},
{
"entropy": 1.871683084964752,
"epoch": 0.09545753785385122,
"grad_norm": 7.703008651733398,
"learning_rate": 1.9013157894736845e-05,
"loss": 0.0671,
"mean_token_accuracy": 0.9811056196689606,
"num_tokens": 619377.0,
"step": 290
},
{
"entropy": 1.9299902677536012,
"epoch": 0.09710335747202106,
"grad_norm": 6.416894912719727,
"learning_rate": 1.9342105263157896e-05,
"loss": 0.089,
"mean_token_accuracy": 0.9779207348823548,
"num_tokens": 630207.0,
"step": 295
},
{
"entropy": 2.0591522693634032,
"epoch": 0.09874917709019092,
"grad_norm": 2.4779272079467773,
"learning_rate": 1.9671052631578947e-05,
"loss": 0.0552,
"mean_token_accuracy": 0.9879588544368744,
"num_tokens": 640467.0,
"step": 300
},
{
"entropy": 2.069437396526337,
"epoch": 0.10039499670836076,
"grad_norm": 6.207917213439941,
"learning_rate": 2e-05,
"loss": 0.1163,
"mean_token_accuracy": 0.9745943665504455,
"num_tokens": 651341.0,
"step": 305
},
{
"entropy": 2.21907696723938,
"epoch": 0.10204081632653061,
"grad_norm": 4.81263542175293,
"learning_rate": 1.9999962969732823e-05,
"loss": 0.0853,
"mean_token_accuracy": 0.9776013195514679,
"num_tokens": 661968.0,
"step": 310
},
{
"entropy": 2.249755620956421,
"epoch": 0.10368663594470046,
"grad_norm": 5.646886348724365,
"learning_rate": 1.999985187920555e-05,
"loss": 0.114,
"mean_token_accuracy": 0.9764432370662689,
"num_tokens": 672739.0,
"step": 315
},
{
"entropy": 2.1538678646087646,
"epoch": 0.10533245556287031,
"grad_norm": 6.742599010467529,
"learning_rate": 1.9999666729240908e-05,
"loss": 0.0768,
"mean_token_accuracy": 0.9786048173904419,
"num_tokens": 683514.0,
"step": 320
},
{
"entropy": 1.9418102025985717,
"epoch": 0.10697827518104015,
"grad_norm": 5.256528854370117,
"learning_rate": 1.9999407521210143e-05,
"loss": 0.091,
"mean_token_accuracy": 0.9797745287418366,
"num_tokens": 694254.0,
"step": 325
},
{
"entropy": 1.8409188032150268,
"epoch": 0.10862409479921001,
"grad_norm": 6.152730464935303,
"learning_rate": 1.9999074257032953e-05,
"loss": 0.0924,
"mean_token_accuracy": 0.9798021256923676,
"num_tokens": 704933.0,
"step": 330
},
{
"entropy": 1.9683008909225463,
"epoch": 0.11026991441737985,
"grad_norm": 6.580195903778076,
"learning_rate": 1.9998666939177514e-05,
"loss": 0.1345,
"mean_token_accuracy": 0.9660695493221283,
"num_tokens": 715336.0,
"step": 335
},
{
"entropy": 2.197187530994415,
"epoch": 0.1119157340355497,
"grad_norm": 5.056039810180664,
"learning_rate": 1.9998185570660445e-05,
"loss": 0.1243,
"mean_token_accuracy": 0.9728079676628113,
"num_tokens": 726034.0,
"step": 340
},
{
"entropy": 2.3285026073455812,
"epoch": 0.11356155365371955,
"grad_norm": 6.029480934143066,
"learning_rate": 1.9997630155046784e-05,
"loss": 0.1302,
"mean_token_accuracy": 0.9665205538272857,
"num_tokens": 737083.0,
"step": 345
},
{
"entropy": 2.1871792316436767,
"epoch": 0.1152073732718894,
"grad_norm": 5.840641498565674,
"learning_rate": 1.9997000696449973e-05,
"loss": 0.087,
"mean_token_accuracy": 0.9797586441040039,
"num_tokens": 747987.0,
"step": 350
},
{
"entropy": 2.055765151977539,
"epoch": 0.11685319289005924,
"grad_norm": 10.675565719604492,
"learning_rate": 1.9996297199531813e-05,
"loss": 0.1199,
"mean_token_accuracy": 0.9669549405574799,
"num_tokens": 758532.0,
"step": 355
},
{
"entropy": 2.0415255546569826,
"epoch": 0.1184990125082291,
"grad_norm": 6.482625961303711,
"learning_rate": 1.9995519669502438e-05,
"loss": 0.1343,
"mean_token_accuracy": 0.9686403274536133,
"num_tokens": 769143.0,
"step": 360
},
{
"entropy": 2.146265912055969,
"epoch": 0.12014483212639894,
"grad_norm": 6.276276588439941,
"learning_rate": 1.9994668112120283e-05,
"loss": 0.0893,
"mean_token_accuracy": 0.9802504718303681,
"num_tokens": 779714.0,
"step": 365
},
{
"entropy": 2.267709231376648,
"epoch": 0.1217906517445688,
"grad_norm": 3.6624529361724854,
"learning_rate": 1.999374253369202e-05,
"loss": 0.1289,
"mean_token_accuracy": 0.967824923992157,
"num_tokens": 790409.0,
"step": 370
},
{
"entropy": 2.3208181142807005,
"epoch": 0.12343647136273865,
"grad_norm": 8.087434768676758,
"learning_rate": 1.999274294107254e-05,
"loss": 0.0958,
"mean_token_accuracy": 0.9716035008430481,
"num_tokens": 801020.0,
"step": 375
},
{
"entropy": 2.2529816150665285,
"epoch": 0.1250822909809085,
"grad_norm": 6.710372447967529,
"learning_rate": 1.9991669341664873e-05,
"loss": 0.1048,
"mean_token_accuracy": 0.971696001291275,
"num_tokens": 811861.0,
"step": 380
},
{
"entropy": 2.1597923040390015,
"epoch": 0.12672811059907835,
"grad_norm": 3.219034194946289,
"learning_rate": 1.9990521743420156e-05,
"loss": 0.1137,
"mean_token_accuracy": 0.9790872097015381,
"num_tokens": 822548.0,
"step": 385
},
{
"entropy": 2.06735520362854,
"epoch": 0.1283739302172482,
"grad_norm": 2.9641964435577393,
"learning_rate": 1.9989300154837564e-05,
"loss": 0.0935,
"mean_token_accuracy": 0.9800699293613434,
"num_tokens": 833368.0,
"step": 390
},
{
"entropy": 2.0970317363739013,
"epoch": 0.13001974983541803,
"grad_norm": 4.221366882324219,
"learning_rate": 1.9988004584964243e-05,
"loss": 0.0939,
"mean_token_accuracy": 0.9755605041980744,
"num_tokens": 844003.0,
"step": 395
},
{
"entropy": 2.073064410686493,
"epoch": 0.1316655694535879,
"grad_norm": 7.461073398590088,
"learning_rate": 1.9986635043395258e-05,
"loss": 0.1238,
"mean_token_accuracy": 0.9669863104820251,
"num_tokens": 854150.0,
"step": 400
},
{
"entropy": 1.975202488899231,
"epoch": 0.13331138907175774,
"grad_norm": 6.416271209716797,
"learning_rate": 1.9985191540273506e-05,
"loss": 0.0933,
"mean_token_accuracy": 0.9760487675666809,
"num_tokens": 864751.0,
"step": 405
},
{
"entropy": 1.9307495832443238,
"epoch": 0.13495720868992758,
"grad_norm": 6.192378044128418,
"learning_rate": 1.9983674086289647e-05,
"loss": 0.0756,
"mean_token_accuracy": 0.9799009144306183,
"num_tokens": 875364.0,
"step": 410
},
{
"entropy": 1.9320047855377198,
"epoch": 0.13660302830809742,
"grad_norm": 4.568058013916016,
"learning_rate": 1.9982082692682027e-05,
"loss": 0.0647,
"mean_token_accuracy": 0.9817326724529266,
"num_tokens": 885830.0,
"step": 415
},
{
"entropy": 1.8869742393493651,
"epoch": 0.1382488479262673,
"grad_norm": 6.1201019287109375,
"learning_rate": 1.998041737123659e-05,
"loss": 0.1027,
"mean_token_accuracy": 0.9717364609241486,
"num_tokens": 896622.0,
"step": 420
},
{
"entropy": 1.9822751045227052,
"epoch": 0.13989466754443713,
"grad_norm": 4.20133638381958,
"learning_rate": 1.9978678134286796e-05,
"loss": 0.1138,
"mean_token_accuracy": 0.9745773077011108,
"num_tokens": 907113.0,
"step": 425
},
{
"entropy": 2.109739363193512,
"epoch": 0.14154048716260698,
"grad_norm": 4.79110860824585,
"learning_rate": 1.997686499471353e-05,
"loss": 0.0759,
"mean_token_accuracy": 0.9797537565231323,
"num_tokens": 917765.0,
"step": 430
},
{
"entropy": 2.127437674999237,
"epoch": 0.14318630678077682,
"grad_norm": 7.135578632354736,
"learning_rate": 1.9974977965945e-05,
"loss": 0.0929,
"mean_token_accuracy": 0.9721114039421082,
"num_tokens": 928469.0,
"step": 435
},
{
"entropy": 1.9786985039710998,
"epoch": 0.1448321263989467,
"grad_norm": 5.621862888336182,
"learning_rate": 1.9973017061956638e-05,
"loss": 0.1085,
"mean_token_accuracy": 0.9724132001399994,
"num_tokens": 939212.0,
"step": 440
},
{
"entropy": 1.7866165399551392,
"epoch": 0.14647794601711653,
"grad_norm": 3.2050716876983643,
"learning_rate": 1.9970982297271007e-05,
"loss": 0.0794,
"mean_token_accuracy": 0.978398722410202,
"num_tokens": 950445.0,
"step": 445
},
{
"entropy": 1.7645732879638671,
"epoch": 0.14812376563528637,
"grad_norm": 6.321614742279053,
"learning_rate": 1.996887368695768e-05,
"loss": 0.0907,
"mean_token_accuracy": 0.9744451761245727,
"num_tokens": 961106.0,
"step": 450
},
{
"entropy": 1.7317794203758239,
"epoch": 0.1497695852534562,
"grad_norm": 4.161308765411377,
"learning_rate": 1.9966691246633143e-05,
"loss": 0.0625,
"mean_token_accuracy": 0.9846729040145874,
"num_tokens": 971898.0,
"step": 455
},
{
"entropy": 1.742343807220459,
"epoch": 0.15141540487162608,
"grad_norm": 4.4447503089904785,
"learning_rate": 1.9964434992460662e-05,
"loss": 0.0797,
"mean_token_accuracy": 0.9820732772350311,
"num_tokens": 982441.0,
"step": 460
},
{
"entropy": 1.8356537818908691,
"epoch": 0.15306122448979592,
"grad_norm": 10.55756664276123,
"learning_rate": 1.9962104941150177e-05,
"loss": 0.1023,
"mean_token_accuracy": 0.9745754778385163,
"num_tokens": 993128.0,
"step": 465
},
{
"entropy": 2.004080033302307,
"epoch": 0.15470704410796576,
"grad_norm": 3.8827741146087646,
"learning_rate": 1.995970110995817e-05,
"loss": 0.0735,
"mean_token_accuracy": 0.9818957507610321,
"num_tokens": 1003682.0,
"step": 470
},
{
"entropy": 2.0348508238792418,
"epoch": 0.1563528637261356,
"grad_norm": 5.195158958435059,
"learning_rate": 1.9957223516687545e-05,
"loss": 0.0934,
"mean_token_accuracy": 0.9747053384780884,
"num_tokens": 1014255.0,
"step": 475
},
{
"entropy": 2.0494486331939696,
"epoch": 0.15799868334430547,
"grad_norm": 6.356441497802734,
"learning_rate": 1.995467217968749e-05,
"loss": 0.1203,
"mean_token_accuracy": 0.9745293378829956,
"num_tokens": 1024794.0,
"step": 480
},
{
"entropy": 2.107176351547241,
"epoch": 0.15964450296247532,
"grad_norm": 4.578567028045654,
"learning_rate": 1.9952047117853345e-05,
"loss": 0.0774,
"mean_token_accuracy": 0.9803178489208222,
"num_tokens": 1035500.0,
"step": 485
},
{
"entropy": 2.1053681492805483,
"epoch": 0.16129032258064516,
"grad_norm": 3.887092113494873,
"learning_rate": 1.9949348350626456e-05,
"loss": 0.0903,
"mean_token_accuracy": 0.9732440710067749,
"num_tokens": 1046221.0,
"step": 490
},
{
"entropy": 2.17068989276886,
"epoch": 0.162936142198815,
"grad_norm": 8.976771354675293,
"learning_rate": 1.9946575897994042e-05,
"loss": 0.0549,
"mean_token_accuracy": 0.9809908986091613,
"num_tokens": 1056570.0,
"step": 495
},
{
"entropy": 2.05935001373291,
"epoch": 0.16458196181698487,
"grad_norm": 4.243111610412598,
"learning_rate": 1.994372978048903e-05,
"loss": 0.0691,
"mean_token_accuracy": 0.9819040298461914,
"num_tokens": 1067196.0,
"step": 500
},
{
"entropy": 2.047864282131195,
"epoch": 0.1662277814351547,
"grad_norm": 3.212517738342285,
"learning_rate": 1.9940810019189912e-05,
"loss": 0.0839,
"mean_token_accuracy": 0.9838090300559997,
"num_tokens": 1077680.0,
"step": 505
},
{
"entropy": 2.0122323632240295,
"epoch": 0.16787360105332455,
"grad_norm": 7.405489444732666,
"learning_rate": 1.9937816635720614e-05,
"loss": 0.1078,
"mean_token_accuracy": 0.9703534841537476,
"num_tokens": 1088097.0,
"step": 510
},
{
"entropy": 2.0119399070739745,
"epoch": 0.1695194206714944,
"grad_norm": 5.046123027801514,
"learning_rate": 1.9934749652250275e-05,
"loss": 0.085,
"mean_token_accuracy": 0.9790121138095855,
"num_tokens": 1098795.0,
"step": 515
},
{
"entropy": 2.0802261352539064,
"epoch": 0.17116524028966426,
"grad_norm": 4.109484672546387,
"learning_rate": 1.9931609091493154e-05,
"loss": 0.0873,
"mean_token_accuracy": 0.9734878361225128,
"num_tokens": 1109471.0,
"step": 520
},
{
"entropy": 2.095242714881897,
"epoch": 0.1728110599078341,
"grad_norm": 3.953889846801758,
"learning_rate": 1.9928394976708403e-05,
"loss": 0.089,
"mean_token_accuracy": 0.9813098549842835,
"num_tokens": 1119968.0,
"step": 525
},
{
"entropy": 2.2089867115020754,
"epoch": 0.17445687952600394,
"grad_norm": 5.200376510620117,
"learning_rate": 1.9925107331699928e-05,
"loss": 0.1003,
"mean_token_accuracy": 0.9750366508960724,
"num_tokens": 1130579.0,
"step": 530
},
{
"entropy": 2.2357950687408445,
"epoch": 0.17610269914417379,
"grad_norm": 5.884069919586182,
"learning_rate": 1.992174618081621e-05,
"loss": 0.0867,
"mean_token_accuracy": 0.979562520980835,
"num_tokens": 1140933.0,
"step": 535
},
{
"entropy": 2.1410197973251344,
"epoch": 0.17774851876234365,
"grad_norm": 6.01693058013916,
"learning_rate": 1.9918311548950102e-05,
"loss": 0.0957,
"mean_token_accuracy": 0.9744680047035217,
"num_tokens": 1151887.0,
"step": 540
},
{
"entropy": 2.228221225738525,
"epoch": 0.1793943383805135,
"grad_norm": 3.3614420890808105,
"learning_rate": 1.991480346153868e-05,
"loss": 0.0887,
"mean_token_accuracy": 0.9763470590114594,
"num_tokens": 1162403.0,
"step": 545
},
{
"entropy": 2.1968831777572633,
"epoch": 0.18104015799868334,
"grad_norm": 4.788167476654053,
"learning_rate": 1.9911221944563022e-05,
"loss": 0.0929,
"mean_token_accuracy": 0.9781689345836639,
"num_tokens": 1173014.0,
"step": 550
},
{
"entropy": 2.0954206466674803,
"epoch": 0.18268597761685318,
"grad_norm": 9.105097770690918,
"learning_rate": 1.9907567024548037e-05,
"loss": 0.0834,
"mean_token_accuracy": 0.9764321863651275,
"num_tokens": 1183721.0,
"step": 555
},
{
"entropy": 2.082006883621216,
"epoch": 0.18433179723502305,
"grad_norm": 5.6749267578125,
"learning_rate": 1.990383872856226e-05,
"loss": 0.0805,
"mean_token_accuracy": 0.9781780660152435,
"num_tokens": 1194206.0,
"step": 560
},
{
"entropy": 2.0844071984291075,
"epoch": 0.1859776168531929,
"grad_norm": 5.848039150238037,
"learning_rate": 1.9900037084217637e-05,
"loss": 0.1125,
"mean_token_accuracy": 0.9710779368877411,
"num_tokens": 1205005.0,
"step": 565
},
{
"entropy": 2.138554549217224,
"epoch": 0.18762343647136273,
"grad_norm": 4.573561668395996,
"learning_rate": 1.9896162119669367e-05,
"loss": 0.0784,
"mean_token_accuracy": 0.978986918926239,
"num_tokens": 1215653.0,
"step": 570
},
{
"entropy": 2.1205518007278443,
"epoch": 0.1892692560895326,
"grad_norm": 7.720390796661377,
"learning_rate": 1.9892213863615634e-05,
"loss": 0.1176,
"mean_token_accuracy": 0.9739040195941925,
"num_tokens": 1226075.0,
"step": 575
},
{
"entropy": 2.1874123096466063,
"epoch": 0.19091507570770244,
"grad_norm": 3.979853630065918,
"learning_rate": 1.9888192345297438e-05,
"loss": 0.092,
"mean_token_accuracy": 0.9774543702602386,
"num_tokens": 1236488.0,
"step": 580
},
{
"entropy": 2.240754795074463,
"epoch": 0.19256089532587228,
"grad_norm": 2.265650510787964,
"learning_rate": 1.9884097594498355e-05,
"loss": 0.0671,
"mean_token_accuracy": 0.9827798247337342,
"num_tokens": 1246837.0,
"step": 585
},
{
"entropy": 2.20197674036026,
"epoch": 0.19420671494404212,
"grad_norm": 8.468692779541016,
"learning_rate": 1.9879929641544328e-05,
"loss": 0.0818,
"mean_token_accuracy": 0.9744803309440613,
"num_tokens": 1257412.0,
"step": 590
},
{
"entropy": 2.1012311816215514,
"epoch": 0.195852534562212,
"grad_norm": 4.669355869293213,
"learning_rate": 1.9875688517303442e-05,
"loss": 0.0844,
"mean_token_accuracy": 0.9798676431179046,
"num_tokens": 1268044.0,
"step": 595
},
{
"entropy": 2.048796272277832,
"epoch": 0.19749835418038184,
"grad_norm": 3.2685132026672363,
"learning_rate": 1.987137425318569e-05,
"loss": 0.0601,
"mean_token_accuracy": 0.9844081580638886,
"num_tokens": 1278849.0,
"step": 600
},
{
"entropy": 2.0971350193023683,
"epoch": 0.19914417379855168,
"grad_norm": 5.5244317054748535,
"learning_rate": 1.9866986881142737e-05,
"loss": 0.1095,
"mean_token_accuracy": 0.9722402155399322,
"num_tokens": 1289463.0,
"step": 605
},
{
"epoch": 0.20013166556945358,
"eval_entropy": 2.089414228537584,
"eval_loss": 0.0923554077744484,
"eval_mean_token_accuracy": 0.9766228082200635,
"eval_num_tokens": 1295891.0,
"eval_runtime": 196.5996,
"eval_samples_per_second": 42.355,
"eval_steps_per_second": 7.06,
"step": 608
},
{
"entropy": 2.0954473495483397,
"epoch": 0.20078999341672152,
"grad_norm": 8.191391944885254,
"learning_rate": 1.9862526433667702e-05,
"loss": 0.1212,
"mean_token_accuracy": 0.97202570438385,
"num_tokens": 1300049.0,
"step": 610
},
{
"entropy": 2.0918750286102297,
"epoch": 0.2024358130348914,
"grad_norm": 5.4046478271484375,
"learning_rate": 1.9857992943794894e-05,
"loss": 0.0611,
"mean_token_accuracy": 0.984895896911621,
"num_tokens": 1310575.0,
"step": 615
},
{
"entropy": 2.0892592668533325,
"epoch": 0.20408163265306123,
"grad_norm": 4.6009840965271,
"learning_rate": 1.9853386445099585e-05,
"loss": 0.107,
"mean_token_accuracy": 0.9737550795078278,
"num_tokens": 1321474.0,
"step": 620
},
{
"entropy": 2.1745733857154845,
"epoch": 0.20572745227123107,
"grad_norm": 5.094756603240967,
"learning_rate": 1.9848706971697744e-05,
"loss": 0.0719,
"mean_token_accuracy": 0.9817912340164184,
"num_tokens": 1332086.0,
"step": 625
},
{
"entropy": 2.1878711700439455,
"epoch": 0.2073732718894009,
"grad_norm": 8.04099178314209,
"learning_rate": 1.98439545582458e-05,
"loss": 0.1345,
"mean_token_accuracy": 0.9673551738262176,
"num_tokens": 1342659.0,
"step": 630
},
{
"entropy": 2.1850104570388793,
"epoch": 0.20901909150757078,
"grad_norm": 5.618947982788086,
"learning_rate": 1.9839129239940392e-05,
"loss": 0.1271,
"mean_token_accuracy": 0.9722469210624695,
"num_tokens": 1353255.0,
"step": 635
},
{
"entropy": 2.2415198564529417,
"epoch": 0.21066491112574062,
"grad_norm": 4.641305923461914,
"learning_rate": 1.9834231052518074e-05,
"loss": 0.0918,
"mean_token_accuracy": 0.9772710859775543,
"num_tokens": 1363984.0,
"step": 640
},
{
"entropy": 2.126233756542206,
"epoch": 0.21231073074391046,
"grad_norm": 6.022129535675049,
"learning_rate": 1.9829260032255093e-05,
"loss": 0.1311,
"mean_token_accuracy": 0.9707091450691223,
"num_tokens": 1374522.0,
"step": 645
},
{
"entropy": 2.136021304130554,
"epoch": 0.2139565503620803,
"grad_norm": 5.552321910858154,
"learning_rate": 1.9824216215967082e-05,
"loss": 0.12,
"mean_token_accuracy": 0.9679823756217957,
"num_tokens": 1385283.0,
"step": 650
},
{
"entropy": 2.1720472097396852,
"epoch": 0.21560236998025017,
"grad_norm": 2.7927908897399902,
"learning_rate": 1.9819099641008817e-05,
"loss": 0.1065,
"mean_token_accuracy": 0.9764181435108185,
"num_tokens": 1395464.0,
"step": 655
},
{
"entropy": 2.1303334712982176,
"epoch": 0.21724818959842002,
"grad_norm": 6.158787250518799,
"learning_rate": 1.9813910345273927e-05,
"loss": 0.0986,
"mean_token_accuracy": 0.9694771349430085,
"num_tokens": 1406190.0,
"step": 660
},
{
"entropy": 2.0735477566719056,
"epoch": 0.21889400921658986,
"grad_norm": 3.1743812561035156,
"learning_rate": 1.9808648367194614e-05,
"loss": 0.1286,
"mean_token_accuracy": 0.9723605871200561,
"num_tokens": 1417233.0,
"step": 665
},
{
"entropy": 1.9879053115844727,
"epoch": 0.2205398288347597,
"grad_norm": 4.825967311859131,
"learning_rate": 1.980331374574137e-05,
"loss": 0.0633,
"mean_token_accuracy": 0.9823581516742707,
"num_tokens": 1427885.0,
"step": 670
},
{
"entropy": 1.953224778175354,
"epoch": 0.22218564845292957,
"grad_norm": 5.118729591369629,
"learning_rate": 1.979790652042268e-05,
"loss": 0.0851,
"mean_token_accuracy": 0.9767945885658265,
"num_tokens": 1438245.0,
"step": 675
},
{
"entropy": 1.881700599193573,
"epoch": 0.2238314680710994,
"grad_norm": 4.021531105041504,
"learning_rate": 1.9792426731284745e-05,
"loss": 0.1035,
"mean_token_accuracy": 0.9744091331958771,
"num_tokens": 1449016.0,
"step": 680
},
{
"entropy": 1.907604944705963,
"epoch": 0.22547728768926925,
"grad_norm": 3.6351332664489746,
"learning_rate": 1.9786874418911187e-05,
"loss": 0.069,
"mean_token_accuracy": 0.9822580933570861,
"num_tokens": 1459773.0,
"step": 685
},
{
"entropy": 1.8353791952133178,
"epoch": 0.2271231073074391,
"grad_norm": 5.1667022705078125,
"learning_rate": 1.9781249624422714e-05,
"loss": 0.0554,
"mean_token_accuracy": 0.9839705526828766,
"num_tokens": 1470406.0,
"step": 690
},
{
"entropy": 1.7690637350082397,
"epoch": 0.22876892692560896,
"grad_norm": 1.4933720827102661,
"learning_rate": 1.9775552389476865e-05,
"loss": 0.0929,
"mean_token_accuracy": 0.9805386900901795,
"num_tokens": 1481079.0,
"step": 695
},
{
"entropy": 1.7553959727287292,
"epoch": 0.2304147465437788,
"grad_norm": 7.114426612854004,
"learning_rate": 1.976978275626766e-05,
"loss": 0.0793,
"mean_token_accuracy": 0.9830655574798584,
"num_tokens": 1491605.0,
"step": 700
},
{
"entropy": 1.6573581099510193,
"epoch": 0.23206056616194864,
"grad_norm": 8.001904487609863,
"learning_rate": 1.976394076752531e-05,
"loss": 0.0476,
"mean_token_accuracy": 0.986677348613739,
"num_tokens": 1502176.0,
"step": 705
},
{
"entropy": 1.6726499915122985,
"epoch": 0.2337063857801185,
"grad_norm": 2.7234785556793213,
"learning_rate": 1.9758026466515902e-05,
"loss": 0.0755,
"mean_token_accuracy": 0.9787233471870422,
"num_tokens": 1512737.0,
"step": 710
},
{
"entropy": 1.6567459583282471,
"epoch": 0.23535220539828836,
"grad_norm": 4.0856547355651855,
"learning_rate": 1.975203989704106e-05,
"loss": 0.0923,
"mean_token_accuracy": 0.980503261089325,
"num_tokens": 1523545.0,
"step": 715
},
{
"entropy": 1.7129284858703613,
"epoch": 0.2369980250164582,
"grad_norm": 5.151584625244141,
"learning_rate": 1.9745981103437643e-05,
"loss": 0.0799,
"mean_token_accuracy": 0.979759806394577,
"num_tokens": 1534017.0,
"step": 720
},
{
"entropy": 1.7500203251838684,
"epoch": 0.23864384463462804,
"grad_norm": 4.301064968109131,
"learning_rate": 1.9739850130577393e-05,
"loss": 0.0907,
"mean_token_accuracy": 0.9790736913681031,
"num_tokens": 1544763.0,
"step": 725
},
{
"entropy": 1.741828978061676,
"epoch": 0.24028966425279788,
"grad_norm": 8.185233116149902,
"learning_rate": 1.973364702386663e-05,
"loss": 0.1146,
"mean_token_accuracy": 0.9718895375728607,
"num_tokens": 1555495.0,
"step": 730
},
{
"entropy": 1.706503689289093,
"epoch": 0.24193548387096775,
"grad_norm": 2.9231204986572266,
"learning_rate": 1.972737182924589e-05,
"loss": 0.0904,
"mean_token_accuracy": 0.9795828342437745,
"num_tokens": 1566520.0,
"step": 735
},
{
"entropy": 1.6863952279090881,
"epoch": 0.2435813034891376,
"grad_norm": 5.155311584472656,
"learning_rate": 1.9721024593189596e-05,
"loss": 0.0856,
"mean_token_accuracy": 0.9796004176139832,
"num_tokens": 1577011.0,
"step": 740
},
{
"entropy": 1.6566576838493348,
"epoch": 0.24522712310730743,
"grad_norm": 5.252721309661865,
"learning_rate": 1.9714605362705725e-05,
"loss": 0.0973,
"mean_token_accuracy": 0.9750127613544464,
"num_tokens": 1587496.0,
"step": 745
},
{
"entropy": 1.646758258342743,
"epoch": 0.2468729427254773,
"grad_norm": 6.778387069702148,
"learning_rate": 1.9708114185335434e-05,
"loss": 0.0768,
"mean_token_accuracy": 0.9819158792495728,
"num_tokens": 1598276.0,
"step": 750
},
{
"entropy": 1.5819305777549744,
"epoch": 0.24851876234364714,
"grad_norm": 2.566246747970581,
"learning_rate": 1.9701551109152732e-05,
"loss": 0.0955,
"mean_token_accuracy": 0.9769714951515198,
"num_tokens": 1609214.0,
"step": 755
},
{
"entropy": 1.5540341138839722,
"epoch": 0.250164581961817,
"grad_norm": 7.977406978607178,
"learning_rate": 1.9694916182764113e-05,
"loss": 0.1004,
"mean_token_accuracy": 0.9762414395809174,
"num_tokens": 1620096.0,
"step": 760
},
{
"entropy": 1.6054931163787842,
"epoch": 0.25181040157998685,
"grad_norm": 3.9608373641967773,
"learning_rate": 1.96882094553082e-05,
"loss": 0.0981,
"mean_token_accuracy": 0.9779304265975952,
"num_tokens": 1630872.0,
"step": 765
},
{
"entropy": 1.6776898503303528,
"epoch": 0.2534562211981567,
"grad_norm": 5.155510425567627,
"learning_rate": 1.9681430976455363e-05,
"loss": 0.0898,
"mean_token_accuracy": 0.9730437695980072,
"num_tokens": 1641601.0,
"step": 770
},
{
"entropy": 1.8331576704978942,
"epoch": 0.25510204081632654,
"grad_norm": 3.792520523071289,
"learning_rate": 1.9674580796407392e-05,
"loss": 0.098,
"mean_token_accuracy": 0.9796571433544159,
"num_tokens": 1652224.0,
"step": 775
},
{
"entropy": 1.8529532432556153,
"epoch": 0.2567478604344964,
"grad_norm": 5.989309310913086,
"learning_rate": 1.966765896589708e-05,
"loss": 0.1201,
"mean_token_accuracy": 0.9693608701229095,
"num_tokens": 1662822.0,
"step": 780
},
{
"entropy": 1.8614219188690186,
"epoch": 0.2583936800526662,
"grad_norm": 4.173523426055908,
"learning_rate": 1.9660665536187875e-05,
"loss": 0.092,
"mean_token_accuracy": 0.9794945657253266,
"num_tokens": 1673483.0,
"step": 785
},
{
"entropy": 1.8263102531433106,
"epoch": 0.26003949967083606,
"grad_norm": 3.7985875606536865,
"learning_rate": 1.965360055907349e-05,
"loss": 0.0998,
"mean_token_accuracy": 0.9791225135326386,
"num_tokens": 1684086.0,
"step": 790
},
{
"entropy": 1.8916141629219054,
"epoch": 0.2616853192890059,
"grad_norm": 5.6334052085876465,
"learning_rate": 1.9646464086877524e-05,
"loss": 0.0919,
"mean_token_accuracy": 0.9774609744548798,
"num_tokens": 1694892.0,
"step": 795
},
{
"entropy": 1.8735905647277833,
"epoch": 0.2633311389071758,
"grad_norm": 4.0586466789245605,
"learning_rate": 1.963925617245307e-05,
"loss": 0.0778,
"mean_token_accuracy": 0.9765025436878204,
"num_tokens": 1705771.0,
"step": 800
},
{
"entropy": 1.8850264430046082,
"epoch": 0.26497695852534564,
"grad_norm": 3.566467523574829,
"learning_rate": 1.963197686918233e-05,
"loss": 0.0735,
"mean_token_accuracy": 0.9790683507919311,
"num_tokens": 1716121.0,
"step": 805
},
{
"entropy": 1.8341678500175476,
"epoch": 0.2666227781435155,
"grad_norm": 3.038248062133789,
"learning_rate": 1.9624626230976208e-05,
"loss": 0.0674,
"mean_token_accuracy": 0.9794325351715087,
"num_tokens": 1726679.0,
"step": 810
},
{
"entropy": 1.8264055967330932,
"epoch": 0.2682685977616853,
"grad_norm": 7.074533462524414,
"learning_rate": 1.961720431227393e-05,
"loss": 0.0827,
"mean_token_accuracy": 0.9796413004398346,
"num_tokens": 1737353.0,
"step": 815
},
{
"entropy": 1.8825155735015868,
"epoch": 0.26991441737985516,
"grad_norm": 6.188788890838623,
"learning_rate": 1.9609711168042612e-05,
"loss": 0.0939,
"mean_token_accuracy": 0.9734999477863312,
"num_tokens": 1748061.0,
"step": 820
},
{
"entropy": 1.9580488204956055,
"epoch": 0.271560236998025,
"grad_norm": 3.9740562438964844,
"learning_rate": 1.9602146853776894e-05,
"loss": 0.0535,
"mean_token_accuracy": 0.9856287181377411,
"num_tokens": 1758768.0,
"step": 825
},
{
"entropy": 1.9614617347717285,
"epoch": 0.27320605661619485,
"grad_norm": 4.990133285522461,
"learning_rate": 1.9594511425498487e-05,
"loss": 0.0551,
"mean_token_accuracy": 0.9855091035366058,
"num_tokens": 1769139.0,
"step": 830
},
{
"entropy": 1.7971890330314637,
"epoch": 0.2748518762343647,
"grad_norm": 5.05476713180542,
"learning_rate": 1.958680493975578e-05,
"loss": 0.0687,
"mean_token_accuracy": 0.9821464836597442,
"num_tokens": 1780069.0,
"step": 835
},
{
"entropy": 1.7543489336967468,
"epoch": 0.2764976958525346,
"grad_norm": 4.147070407867432,
"learning_rate": 1.957902745362341e-05,
"loss": 0.0743,
"mean_token_accuracy": 0.9777528941631317,
"num_tokens": 1790802.0,
"step": 840
},
{
"entropy": 1.6835835099220275,
"epoch": 0.2781435154707044,
"grad_norm": 7.310057640075684,
"learning_rate": 1.957117902470187e-05,
"loss": 0.0913,
"mean_token_accuracy": 0.9787399888038635,
"num_tokens": 1801255.0,
"step": 845
},
{
"entropy": 1.7269906282424927,
"epoch": 0.27978933508887427,
"grad_norm": 6.00834321975708,
"learning_rate": 1.956325971111703e-05,
"loss": 0.1044,
"mean_token_accuracy": 0.9713066816329956,
"num_tokens": 1811973.0,
"step": 850
},
{
"entropy": 1.7431592702865601,
"epoch": 0.2814351547070441,
"grad_norm": 4.275935173034668,
"learning_rate": 1.955526957151976e-05,
"loss": 0.0821,
"mean_token_accuracy": 0.9789995789527893,
"num_tokens": 1822550.0,
"step": 855
},
{
"entropy": 1.7215226411819458,
"epoch": 0.28308097432521395,
"grad_norm": 2.8283443450927734,
"learning_rate": 1.954720866508546e-05,
"loss": 0.0592,
"mean_token_accuracy": 0.9890708029270172,
"num_tokens": 1833405.0,
"step": 860
},
{
"entropy": 1.7675551056861878,
"epoch": 0.2847267939433838,
"grad_norm": 4.092453956604004,
"learning_rate": 1.9539077051513624e-05,
"loss": 0.0992,
"mean_token_accuracy": 0.973135906457901,
"num_tokens": 1843944.0,
"step": 865
},
{
"entropy": 1.7665972590446473,
"epoch": 0.28637261356155364,
"grad_norm": 3.9631991386413574,
"learning_rate": 1.9530874791027425e-05,
"loss": 0.1248,
"mean_token_accuracy": 0.9706246018409729,
"num_tokens": 1854547.0,
"step": 870
},
{
"entropy": 1.7626410603523255,
"epoch": 0.2880184331797235,
"grad_norm": 5.616950988769531,
"learning_rate": 1.952260194437324e-05,
"loss": 0.0911,
"mean_token_accuracy": 0.9797743022441864,
"num_tokens": 1864880.0,
"step": 875
},
{
"entropy": 1.7034204244613647,
"epoch": 0.2896642527978934,
"grad_norm": 3.930392265319824,
"learning_rate": 1.9514258572820216e-05,
"loss": 0.0789,
"mean_token_accuracy": 0.9830069661140441,
"num_tokens": 1875669.0,
"step": 880
},
{
"entropy": 1.72216739654541,
"epoch": 0.2913100724160632,
"grad_norm": 3.14703631401062,
"learning_rate": 1.9505844738159807e-05,
"loss": 0.0851,
"mean_token_accuracy": 0.9783756732940674,
"num_tokens": 1886494.0,
"step": 885
},
{
"entropy": 1.671413540840149,
"epoch": 0.29295589203423306,
"grad_norm": 4.539898872375488,
"learning_rate": 1.949736050270532e-05,
"loss": 0.0766,
"mean_token_accuracy": 0.9854376316070557,
"num_tokens": 1897127.0,
"step": 890
},
{
"entropy": 1.6798272371292113,
"epoch": 0.2946017116524029,
"grad_norm": 3.341653347015381,
"learning_rate": 1.948880592929146e-05,
"loss": 0.0631,
"mean_token_accuracy": 0.9869305789470673,
"num_tokens": 1907594.0,
"step": 895
},
{
"entropy": 1.6980504512786865,
"epoch": 0.29624753127057274,
"grad_norm": 3.1306231021881104,
"learning_rate": 1.9480181081273846e-05,
"loss": 0.0964,
"mean_token_accuracy": 0.9781007289886474,
"num_tokens": 1918632.0,
"step": 900
},
{
"entropy": 1.7191000819206237,
"epoch": 0.2978933508887426,
"grad_norm": 3.709050416946411,
"learning_rate": 1.947148602252858e-05,
"loss": 0.0767,
"mean_token_accuracy": 0.9797729074954986,
"num_tokens": 1929270.0,
"step": 905
},
{
"entropy": 1.7052234768867494,
"epoch": 0.2995391705069124,
"grad_norm": 2.6057310104370117,
"learning_rate": 1.946272081745171e-05,
"loss": 0.0611,
"mean_token_accuracy": 0.9820807099342346,
"num_tokens": 1940143.0,
"step": 910
},
{
"entropy": 1.7210368633270263,
"epoch": 0.30118499012508226,
"grad_norm": 5.367160320281982,
"learning_rate": 1.9453885530958835e-05,
"loss": 0.0817,
"mean_token_accuracy": 0.9767018914222717,
"num_tokens": 1951059.0,
"step": 915
},
{
"entropy": 1.6308549642562866,
"epoch": 0.30283080974325216,
"grad_norm": 5.0577592849731445,
"learning_rate": 1.9444980228484542e-05,
"loss": 0.0641,
"mean_token_accuracy": 0.9800835251808167,
"num_tokens": 1962214.0,
"step": 920
},
{
"entropy": 1.6711305975914001,
"epoch": 0.304476629361422,
"grad_norm": 6.653420925140381,
"learning_rate": 1.9436004975981986e-05,
"loss": 0.108,
"mean_token_accuracy": 0.9743327260017395,
"num_tokens": 1972789.0,
"step": 925
},
{
"entropy": 1.679636800289154,
"epoch": 0.30612244897959184,
"grad_norm": 4.994918346405029,
"learning_rate": 1.9426959839922367e-05,
"loss": 0.0796,
"mean_token_accuracy": 0.9823659062385559,
"num_tokens": 1983264.0,
"step": 930
},
{
"entropy": 1.7234082102775574,
"epoch": 0.3077682685977617,
"grad_norm": 5.689176082611084,
"learning_rate": 1.941784488729444e-05,
"loss": 0.0774,
"mean_token_accuracy": 0.9792181611061096,
"num_tokens": 1993783.0,
"step": 935
},
{
"entropy": 1.7042930126190186,
"epoch": 0.3094140882159315,
"grad_norm": 2.4102768898010254,
"learning_rate": 1.9408660185604035e-05,
"loss": 0.0876,
"mean_token_accuracy": 0.9759888887405396,
"num_tokens": 2004295.0,
"step": 940
},
{
"entropy": 1.6130717277526856,
"epoch": 0.31105990783410137,
"grad_norm": 2.775099277496338,
"learning_rate": 1.939940580287354e-05,
"loss": 0.0805,
"mean_token_accuracy": 0.9795459330081939,
"num_tokens": 2015224.0,
"step": 945
},
{
"entropy": 1.5929541826248168,
"epoch": 0.3127057274522712,
"grad_norm": 3.403689384460449,
"learning_rate": 1.9390081807641413e-05,
"loss": 0.0899,
"mean_token_accuracy": 0.9769740760326385,
"num_tokens": 2025787.0,
"step": 950
},
{
"entropy": 1.5571842432022094,
"epoch": 0.3143515470704411,
"grad_norm": 3.0090384483337402,
"learning_rate": 1.938068826896166e-05,
"loss": 0.06,
"mean_token_accuracy": 0.9822336971759796,
"num_tokens": 2036384.0,
"step": 955
},
{
"entropy": 1.5760810375213623,
"epoch": 0.31599736668861095,
"grad_norm": 2.722409725189209,
"learning_rate": 1.9371225256403328e-05,
"loss": 0.0888,
"mean_token_accuracy": 0.9776059925556183,
"num_tokens": 2047090.0,
"step": 960
},
{
"entropy": 1.6288114428520202,
"epoch": 0.3176431863067808,
"grad_norm": 2.649474620819092,
"learning_rate": 1.9361692840049997e-05,
"loss": 0.066,
"mean_token_accuracy": 0.979485136270523,
"num_tokens": 2057684.0,
"step": 965
},
{
"entropy": 1.710892951488495,
"epoch": 0.31928900592495063,
"grad_norm": 2.8936843872070312,
"learning_rate": 1.935209109049925e-05,
"loss": 0.0862,
"mean_token_accuracy": 0.9751926064491272,
"num_tokens": 2068108.0,
"step": 970
},
{
"entropy": 1.7829145908355712,
"epoch": 0.32093482554312047,
"grad_norm": 2.203496217727661,
"learning_rate": 1.9342420078862153e-05,
"loss": 0.0902,
"mean_token_accuracy": 0.9730205178260803,
"num_tokens": 2078727.0,
"step": 975
},
{
"entropy": 1.6852322578430177,
"epoch": 0.3225806451612903,
"grad_norm": 2.909604072570801,
"learning_rate": 1.933267987676274e-05,
"loss": 0.0754,
"mean_token_accuracy": 0.9842035174369812,
"num_tokens": 2089939.0,
"step": 980
},
{
"entropy": 1.73391033411026,
"epoch": 0.32422646477946016,
"grad_norm": 2.8957767486572266,
"learning_rate": 1.9322870556337466e-05,
"loss": 0.0894,
"mean_token_accuracy": 0.9820044219493866,
"num_tokens": 2100675.0,
"step": 985
},
{
"entropy": 1.7834203481674193,
"epoch": 0.32587228439763,
"grad_norm": 5.757988929748535,
"learning_rate": 1.931299219023469e-05,
"loss": 0.085,
"mean_token_accuracy": 0.977233600616455,
"num_tokens": 2111220.0,
"step": 990
},
{
"entropy": 1.7618568420410157,
"epoch": 0.3275181040157999,
"grad_norm": 5.181485176086426,
"learning_rate": 1.9303044851614106e-05,
"loss": 0.0899,
"mean_token_accuracy": 0.9746578216552735,
"num_tokens": 2121728.0,
"step": 995
},
{
"entropy": 1.7923735499382019,
"epoch": 0.32916392363396973,
"grad_norm": 9.63333797454834,
"learning_rate": 1.9293028614146246e-05,
"loss": 0.1188,
"mean_token_accuracy": 0.9688325583934784,
"num_tokens": 2132434.0,
"step": 1000
},
{
"entropy": 1.860191786289215,
"epoch": 0.3308097432521396,
"grad_norm": 4.145689487457275,
"learning_rate": 1.9282943552011892e-05,
"loss": 0.0798,
"mean_token_accuracy": 0.9769950270652771,
"num_tokens": 2142917.0,
"step": 1005
},
{
"entropy": 1.833854353427887,
"epoch": 0.3324555628703094,
"grad_norm": 3.677757501602173,
"learning_rate": 1.927278973990156e-05,
"loss": 0.0828,
"mean_token_accuracy": 0.9798845946788788,
"num_tokens": 2153639.0,
"step": 1010
},
{
"entropy": 1.8186659574508668,
"epoch": 0.33410138248847926,
"grad_norm": 4.619242191314697,
"learning_rate": 1.9262567253014922e-05,
"loss": 0.0607,
"mean_token_accuracy": 0.9790556609630585,
"num_tokens": 2164673.0,
"step": 1015
},
{
"entropy": 1.8815788626670837,
"epoch": 0.3357472021066491,
"grad_norm": 6.019654273986816,
"learning_rate": 1.925227616706026e-05,
"loss": 0.0804,
"mean_token_accuracy": 0.9770128130912781,
"num_tokens": 2175078.0,
"step": 1020
},
{
"entropy": 1.8578821659088134,
"epoch": 0.33739302172481894,
"grad_norm": 4.046188831329346,
"learning_rate": 1.924191655825391e-05,
"loss": 0.0688,
"mean_token_accuracy": 0.979873389005661,
"num_tokens": 2185679.0,
"step": 1025
},
{
"entropy": 1.9458543419837953,
"epoch": 0.3390388413429888,
"grad_norm": 8.663795471191406,
"learning_rate": 1.9231488503319687e-05,
"loss": 0.1165,
"mean_token_accuracy": 0.9754189252853394,
"num_tokens": 2196306.0,
"step": 1030
},
{
"entropy": 1.9784336686134338,
"epoch": 0.3406846609611587,
"grad_norm": 4.623157501220703,
"learning_rate": 1.9220992079488322e-05,
"loss": 0.0863,
"mean_token_accuracy": 0.9824258208274841,
"num_tokens": 2207144.0,
"step": 1035
},
{
"entropy": 2.0585790038108827,
"epoch": 0.3423304805793285,
"grad_norm": 5.089199066162109,
"learning_rate": 1.9210427364496894e-05,
"loss": 0.0813,
"mean_token_accuracy": 0.9741666853427887,
"num_tokens": 2217699.0,
"step": 1040
},
{
"entropy": 1.9863484263420106,
"epoch": 0.34397630019749836,
"grad_norm": 2.4876182079315186,
"learning_rate": 1.9199794436588244e-05,
"loss": 0.0936,
"mean_token_accuracy": 0.9748422205448151,
"num_tokens": 2228579.0,
"step": 1045
},
{
"entropy": 2.034103310108185,
"epoch": 0.3456221198156682,
"grad_norm": 3.3714184761047363,
"learning_rate": 1.9189093374510403e-05,
"loss": 0.1052,
"mean_token_accuracy": 0.9673223495483398,
"num_tokens": 2239194.0,
"step": 1050
},
{
"entropy": 1.954673457145691,
"epoch": 0.34726793943383805,
"grad_norm": 3.50931453704834,
"learning_rate": 1.917832425751601e-05,
"loss": 0.0578,
"mean_token_accuracy": 0.9865812182426452,
"num_tokens": 2249697.0,
"step": 1055
},
{
"entropy": 1.8305863380432128,
"epoch": 0.3489137590520079,
"grad_norm": 5.005392074584961,
"learning_rate": 1.9167487165361726e-05,
"loss": 0.0775,
"mean_token_accuracy": 0.9806217133998871,
"num_tokens": 2260788.0,
"step": 1060
},
{
"entropy": 1.7882127285003662,
"epoch": 0.35055957867017773,
"grad_norm": 1.8662757873535156,
"learning_rate": 1.9156582178307625e-05,
"loss": 0.1092,
"mean_token_accuracy": 0.9730004489421844,
"num_tokens": 2271354.0,
"step": 1065
},
{
"entropy": 1.7463540315628052,
"epoch": 0.35220539828834757,
"grad_norm": 7.321222305297852,
"learning_rate": 1.9145609377116635e-05,
"loss": 0.0659,
"mean_token_accuracy": 0.9823782205581665,
"num_tokens": 2282285.0,
"step": 1070
},
{
"entropy": 1.7453363537788391,
"epoch": 0.35385121790651747,
"grad_norm": 3.2284722328186035,
"learning_rate": 1.9134568843053895e-05,
"loss": 0.0813,
"mean_token_accuracy": 0.9803031504154205,
"num_tokens": 2293037.0,
"step": 1075
},
{
"entropy": 1.76043781042099,
"epoch": 0.3554970375246873,
"grad_norm": 2.238819122314453,
"learning_rate": 1.91234606578862e-05,
"loss": 0.0761,
"mean_token_accuracy": 0.9841830551624298,
"num_tokens": 2303850.0,
"step": 1080
},
{
"entropy": 1.7409037947654724,
"epoch": 0.35714285714285715,
"grad_norm": 7.227831840515137,
"learning_rate": 1.911228490388136e-05,
"loss": 0.0743,
"mean_token_accuracy": 0.9812696516513825,
"num_tokens": 2314805.0,
"step": 1085
},
{
"entropy": 1.8239703059196473,
"epoch": 0.358788676761027,
"grad_norm": 5.955824851989746,
"learning_rate": 1.9101041663807606e-05,
"loss": 0.0915,
"mean_token_accuracy": 0.976921284198761,
"num_tokens": 2325296.0,
"step": 1090
},
{
"entropy": 1.8835476994514466,
"epoch": 0.36043449637919683,
"grad_norm": 2.8221843242645264,
"learning_rate": 1.9089731020932972e-05,
"loss": 0.0631,
"mean_token_accuracy": 0.9831080317497254,
"num_tokens": 2336116.0,
"step": 1095
},
{
"entropy": 1.9333994030952453,
"epoch": 0.3620803159973667,
"grad_norm": 6.433844089508057,
"learning_rate": 1.907835305902469e-05,
"loss": 0.0998,
"mean_token_accuracy": 0.9790532469749451,
"num_tokens": 2346756.0,
"step": 1100
},
{
"entropy": 2.0366763949394224,
"epoch": 0.3637261356155365,
"grad_norm": 13.404642105102539,
"learning_rate": 1.906690786234855e-05,
"loss": 0.0569,
"mean_token_accuracy": 0.9880981385707855,
"num_tokens": 2357587.0,
"step": 1105
},
{
"entropy": 2.0615396976470945,
"epoch": 0.36537195523370636,
"grad_norm": 4.546095371246338,
"learning_rate": 1.9055395515668288e-05,
"loss": 0.0903,
"mean_token_accuracy": 0.9765397369861603,
"num_tokens": 2368379.0,
"step": 1110
},
{
"entropy": 2.036512088775635,
"epoch": 0.36701777485187626,
"grad_norm": 4.192592144012451,
"learning_rate": 1.9043816104244964e-05,
"loss": 0.0779,
"mean_token_accuracy": 0.9820389151573181,
"num_tokens": 2379490.0,
"step": 1115
},
{
"entropy": 2.113160729408264,
"epoch": 0.3686635944700461,
"grad_norm": 4.126070976257324,
"learning_rate": 1.9032169713836314e-05,
"loss": 0.0884,
"mean_token_accuracy": 0.9770908057689667,
"num_tokens": 2390060.0,
"step": 1120
},
{
"entropy": 2.1656644344329834,
"epoch": 0.37030941408821594,
"grad_norm": 2.7143678665161133,
"learning_rate": 1.9020456430696126e-05,
"loss": 0.0493,
"mean_token_accuracy": 0.9862718224525452,
"num_tokens": 2400221.0,
"step": 1125
},
{
"entropy": 2.1137581944465635,
"epoch": 0.3719552337063858,
"grad_norm": 4.085227012634277,
"learning_rate": 1.9008676341573606e-05,
"loss": 0.0521,
"mean_token_accuracy": 0.988205760717392,
"num_tokens": 2410885.0,
"step": 1130
},
{
"entropy": 2.044228506088257,
"epoch": 0.3736010533245556,
"grad_norm": 5.65952205657959,
"learning_rate": 1.8996829533712723e-05,
"loss": 0.0629,
"mean_token_accuracy": 0.984690648317337,
"num_tokens": 2421494.0,
"step": 1135
},
{
"entropy": 2.0729115128517153,
"epoch": 0.37524687294272546,
"grad_norm": 3.8629510402679443,
"learning_rate": 1.898491609485156e-05,
"loss": 0.0862,
"mean_token_accuracy": 0.9760863423347473,
"num_tokens": 2432022.0,
"step": 1140
},
{
"entropy": 2.1294867634773254,
"epoch": 0.3768926925608953,
"grad_norm": 4.639437198638916,
"learning_rate": 1.8972936113221696e-05,
"loss": 0.0759,
"mean_token_accuracy": 0.9800122320652008,
"num_tokens": 2442748.0,
"step": 1145
},
{
"entropy": 2.141499698162079,
"epoch": 0.3785385121790652,
"grad_norm": 3.745588541030884,
"learning_rate": 1.8960889677547506e-05,
"loss": 0.0843,
"mean_token_accuracy": 0.9796210825443268,
"num_tokens": 2453552.0,
"step": 1150
},
{
"entropy": 2.2506140947341917,
"epoch": 0.38018433179723504,
"grad_norm": 1.9200206995010376,
"learning_rate": 1.8948776877045535e-05,
"loss": 0.0818,
"mean_token_accuracy": 0.9776529312133789,
"num_tokens": 2464201.0,
"step": 1155
},
{
"entropy": 2.2783297538757323,
"epoch": 0.3818301514154049,
"grad_norm": 3.2144906520843506,
"learning_rate": 1.893659780142384e-05,
"loss": 0.0748,
"mean_token_accuracy": 0.9799034297466278,
"num_tokens": 2475109.0,
"step": 1160
},
{
"entropy": 2.235105013847351,
"epoch": 0.3834759710335747,
"grad_norm": 11.074976921081543,
"learning_rate": 1.8924352540881298e-05,
"loss": 0.0759,
"mean_token_accuracy": 0.9808390080928803,
"num_tokens": 2485624.0,
"step": 1165
},
{
"entropy": 2.216382932662964,
"epoch": 0.38512179065174457,
"grad_norm": 5.54982328414917,
"learning_rate": 1.891204118610696e-05,
"loss": 0.1232,
"mean_token_accuracy": 0.9734435498714447,
"num_tokens": 2496301.0,
"step": 1170
},
{
"entropy": 2.362920618057251,
"epoch": 0.3867676102699144,
"grad_norm": 4.762061595916748,
"learning_rate": 1.8899663828279387e-05,
"loss": 0.0906,
"mean_token_accuracy": 0.9737257599830628,
"num_tokens": 2506985.0,
"step": 1175
},
{
"entropy": 2.3827850580215455,
"epoch": 0.38841342988808425,
"grad_norm": 4.1375732421875,
"learning_rate": 1.8887220559065946e-05,
"loss": 0.0868,
"mean_token_accuracy": 0.9808934390544891,
"num_tokens": 2517845.0,
"step": 1180
},
{
"entropy": 2.4273977756500242,
"epoch": 0.3900592495062541,
"grad_norm": 3.6323893070220947,
"learning_rate": 1.8874711470622152e-05,
"loss": 0.0928,
"mean_token_accuracy": 0.9732855081558227,
"num_tokens": 2528576.0,
"step": 1185
},
{
"entropy": 2.3609946250915526,
"epoch": 0.391705069124424,
"grad_norm": 2.174260377883911,
"learning_rate": 1.886213665559099e-05,
"loss": 0.0884,
"mean_token_accuracy": 0.9792202174663543,
"num_tokens": 2539169.0,
"step": 1190
},
{
"entropy": 2.293043828010559,
"epoch": 0.39335088874259383,
"grad_norm": 4.491889953613281,
"learning_rate": 1.8849496207102204e-05,
"loss": 0.0859,
"mean_token_accuracy": 0.9761190593242646,
"num_tokens": 2549674.0,
"step": 1195
},
{
"entropy": 2.2694995403289795,
"epoch": 0.39499670836076367,
"grad_norm": 5.4928812980651855,
"learning_rate": 1.8836790218771637e-05,
"loss": 0.0828,
"mean_token_accuracy": 0.9821516633033752,
"num_tokens": 2560364.0,
"step": 1200
},
{
"entropy": 2.3214454650878906,
"epoch": 0.3966425279789335,
"grad_norm": 3.3526692390441895,
"learning_rate": 1.882401878470052e-05,
"loss": 0.0621,
"mean_token_accuracy": 0.9838208556175232,
"num_tokens": 2570839.0,
"step": 1205
},
{
"entropy": 2.4305625438690184,
"epoch": 0.39828834759710335,
"grad_norm": 2.5236058235168457,
"learning_rate": 1.8811181999474763e-05,
"loss": 0.0772,
"mean_token_accuracy": 0.9798738360404968,
"num_tokens": 2581459.0,
"step": 1210
},
{
"entropy": 2.4986673831939696,
"epoch": 0.3999341672152732,
"grad_norm": 5.663600921630859,
"learning_rate": 1.8798279958164295e-05,
"loss": 0.0663,
"mean_token_accuracy": 0.9851954817771912,
"num_tokens": 2591722.0,
"step": 1215
},
{
"epoch": 0.40026333113890716,
"eval_entropy": 2.455367434179405,
"eval_loss": 0.07768969982862473,
"eval_mean_token_accuracy": 0.9805304530889912,
"eval_num_tokens": 2593845.0,
"eval_runtime": 196.539,
"eval_samples_per_second": 42.368,
"eval_steps_per_second": 7.062,
"step": 1216
},
{
"entropy": 2.4370908737182617,
"epoch": 0.40157998683344304,
"grad_norm": 3.0529227256774902,
"learning_rate": 1.878531275632232e-05,
"loss": 0.1109,
"mean_token_accuracy": 0.9730750441551208,
"num_tokens": 2602413.0,
"step": 1220
},
{
"entropy": 2.4589828968048097,
"epoch": 0.4032258064516129,
"grad_norm": 7.966182231903076,
"learning_rate": 1.8772280489984628e-05,
"loss": 0.0739,
"mean_token_accuracy": 0.9801295638084412,
"num_tokens": 2613250.0,
"step": 1225
},
{
"entropy": 2.4703043937683105,
"epoch": 0.4048716260697828,
"grad_norm": 4.381008148193359,
"learning_rate": 1.875918325566888e-05,
"loss": 0.0812,
"mean_token_accuracy": 0.9783048331737518,
"num_tokens": 2624053.0,
"step": 1230
},
{
"entropy": 2.4899261713027956,
"epoch": 0.4065174456879526,
"grad_norm": 4.341674327850342,
"learning_rate": 1.8746021150373892e-05,
"loss": 0.0768,
"mean_token_accuracy": 0.9822542846202851,
"num_tokens": 2634584.0,
"step": 1235
},
{
"entropy": 2.5025168657302856,
"epoch": 0.40816326530612246,
"grad_norm": 4.821120262145996,
"learning_rate": 1.873279427157892e-05,
"loss": 0.0649,
"mean_token_accuracy": 0.9847510397434235,
"num_tokens": 2645173.0,
"step": 1240
},
{
"entropy": 2.5438794136047362,
"epoch": 0.4098090849242923,
"grad_norm": 4.269495010375977,
"learning_rate": 1.8719502717242937e-05,
"loss": 0.0801,
"mean_token_accuracy": 0.9804035782814026,
"num_tokens": 2655404.0,
"step": 1245
},
{
"entropy": 2.467151665687561,
"epoch": 0.41145490454246214,
"grad_norm": 2.994798421859741,
"learning_rate": 1.8706146585803903e-05,
"loss": 0.0676,
"mean_token_accuracy": 0.9815965473651886,
"num_tokens": 2666063.0,
"step": 1250
},
{
"entropy": 2.3562156200408935,
"epoch": 0.413100724160632,
"grad_norm": 4.536716461181641,
"learning_rate": 1.8692725976178038e-05,
"loss": 0.0718,
"mean_token_accuracy": 0.9820040583610534,
"num_tokens": 2676929.0,
"step": 1255
},
{
"entropy": 2.3580038785934447,
"epoch": 0.4147465437788018,
"grad_norm": 3.2308404445648193,
"learning_rate": 1.8679240987759098e-05,
"loss": 0.0654,
"mean_token_accuracy": 0.9846524596214294,
"num_tokens": 2687554.0,
"step": 1260
},
{
"entropy": 2.24311740398407,
"epoch": 0.41639236339697167,
"grad_norm": 3.5214755535125732,
"learning_rate": 1.8665691720417624e-05,
"loss": 0.0644,
"mean_token_accuracy": 0.9793017745018006,
"num_tokens": 2698138.0,
"step": 1265
},
{
"entropy": 2.195026898384094,
"epoch": 0.41803818301514156,
"grad_norm": 2.515350818634033,
"learning_rate": 1.865207827450022e-05,
"loss": 0.0624,
"mean_token_accuracy": 0.9875331342220306,
"num_tokens": 2708644.0,
"step": 1270
},
{
"entropy": 2.1811726808547975,
"epoch": 0.4196840026333114,
"grad_norm": 3.122403144836426,
"learning_rate": 1.8638400750828793e-05,
"loss": 0.0525,
"mean_token_accuracy": 0.9861293792724609,
"num_tokens": 2719229.0,
"step": 1275
},
{
"entropy": 2.1384164810180666,
"epoch": 0.42132982225148125,
"grad_norm": 3.760925531387329,
"learning_rate": 1.8624659250699807e-05,
"loss": 0.1054,
"mean_token_accuracy": 0.973668885231018,
"num_tokens": 2729859.0,
"step": 1280
},
{
"entropy": 2.136362624168396,
"epoch": 0.4229756418696511,
"grad_norm": 3.5055840015411377,
"learning_rate": 1.8610853875883553e-05,
"loss": 0.0707,
"mean_token_accuracy": 0.980605673789978,
"num_tokens": 2740393.0,
"step": 1285
},
{
"entropy": 2.1360629320144655,
"epoch": 0.42462146148782093,
"grad_norm": 5.643499851226807,
"learning_rate": 1.8596984728623374e-05,
"loss": 0.0891,
"mean_token_accuracy": 0.9784270226955414,
"num_tokens": 2750842.0,
"step": 1290
},
{
"entropy": 2.1816317081451415,
"epoch": 0.42626728110599077,
"grad_norm": 2.8327138423919678,
"learning_rate": 1.858305191163491e-05,
"loss": 0.0915,
"mean_token_accuracy": 0.9775237262248992,
"num_tokens": 2761249.0,
"step": 1295
},
{
"entropy": 2.2412596225738524,
"epoch": 0.4279131007241606,
"grad_norm": 1.0889869928359985,
"learning_rate": 1.8569055528105356e-05,
"loss": 0.0952,
"mean_token_accuracy": 0.9791838228702545,
"num_tokens": 2771715.0,
"step": 1300
},
{
"entropy": 2.200755214691162,
"epoch": 0.4295589203423305,
"grad_norm": 2.3900723457336426,
"learning_rate": 1.855499568169267e-05,
"loss": 0.0761,
"mean_token_accuracy": 0.9767322957515716,
"num_tokens": 2782449.0,
"step": 1305
},
{
"entropy": 2.1307253360748293,
"epoch": 0.43120473996050035,
"grad_norm": 3.4063174724578857,
"learning_rate": 1.854087247652483e-05,
"loss": 0.0675,
"mean_token_accuracy": 0.9809297919273376,
"num_tokens": 2792843.0,
"step": 1310
},
{
"entropy": 2.0838756799697875,
"epoch": 0.4328505595786702,
"grad_norm": 2.3350863456726074,
"learning_rate": 1.8526686017199046e-05,
"loss": 0.0859,
"mean_token_accuracy": 0.9777861475944519,
"num_tokens": 2803434.0,
"step": 1315
},
{
"entropy": 2.061713254451752,
"epoch": 0.43449637919684003,
"grad_norm": 3.2840187549591064,
"learning_rate": 1.8512436408780995e-05,
"loss": 0.0994,
"mean_token_accuracy": 0.979932302236557,
"num_tokens": 2814246.0,
"step": 1320
},
{
"entropy": 2.1339207768440245,
"epoch": 0.4361421988150099,
"grad_norm": 3.665444850921631,
"learning_rate": 1.8498123756804038e-05,
"loss": 0.0704,
"mean_token_accuracy": 0.9794367909431457,
"num_tokens": 2825017.0,
"step": 1325
},
{
"entropy": 2.1423274278640747,
"epoch": 0.4377880184331797,
"grad_norm": 4.194466590881348,
"learning_rate": 1.848374816726844e-05,
"loss": 0.0636,
"mean_token_accuracy": 0.9790065705776214,
"num_tokens": 2835765.0,
"step": 1330
},
{
"entropy": 2.220040726661682,
"epoch": 0.43943383805134956,
"grad_norm": 2.7448790073394775,
"learning_rate": 1.8469309746640587e-05,
"loss": 0.0651,
"mean_token_accuracy": 0.9824341356754303,
"num_tokens": 2846294.0,
"step": 1335
},
{
"entropy": 2.1996994495391844,
"epoch": 0.4410796576695194,
"grad_norm": 3.1929495334625244,
"learning_rate": 1.845480860185219e-05,
"loss": 0.0431,
"mean_token_accuracy": 0.9881280601024628,
"num_tokens": 2856892.0,
"step": 1340
},
{
"entropy": 2.0707924485206606,
"epoch": 0.4427254772876893,
"grad_norm": 7.250471115112305,
"learning_rate": 1.8440244840299507e-05,
"loss": 0.0697,
"mean_token_accuracy": 0.983168751001358,
"num_tokens": 2867704.0,
"step": 1345
},
{
"entropy": 1.9964185953140259,
"epoch": 0.44437129690585914,
"grad_norm": 5.546008586883545,
"learning_rate": 1.8425618569842528e-05,
"loss": 0.0749,
"mean_token_accuracy": 0.9826294600963592,
"num_tokens": 2878605.0,
"step": 1350
},
{
"entropy": 2.0520959854125977,
"epoch": 0.446017116524029,
"grad_norm": 2.6965854167938232,
"learning_rate": 1.8410929898804197e-05,
"loss": 0.0748,
"mean_token_accuracy": 0.983626252412796,
"num_tokens": 2889180.0,
"step": 1355
},
{
"entropy": 2.033618211746216,
"epoch": 0.4476629361421988,
"grad_norm": 6.2912116050720215,
"learning_rate": 1.83961789359696e-05,
"loss": 0.0658,
"mean_token_accuracy": 0.9775195240974426,
"num_tokens": 2899792.0,
"step": 1360
},
{
"entropy": 2.0571701526641846,
"epoch": 0.44930875576036866,
"grad_norm": 5.728684425354004,
"learning_rate": 1.838136579058515e-05,
"loss": 0.1018,
"mean_token_accuracy": 0.9756032049655914,
"num_tokens": 2910450.0,
"step": 1365
},
{
"entropy": 2.135818696022034,
"epoch": 0.4509545753785385,
"grad_norm": 2.3949079513549805,
"learning_rate": 1.8366490572357798e-05,
"loss": 0.0688,
"mean_token_accuracy": 0.9789416313171386,
"num_tokens": 2921065.0,
"step": 1370
},
{
"entropy": 2.213190221786499,
"epoch": 0.45260039499670834,
"grad_norm": 3.1379809379577637,
"learning_rate": 1.8351553391454203e-05,
"loss": 0.0926,
"mean_token_accuracy": 0.9748626470565795,
"num_tokens": 2931783.0,
"step": 1375
},
{
"entropy": 2.259020519256592,
"epoch": 0.4542462146148782,
"grad_norm": 6.139921188354492,
"learning_rate": 1.8336554358499923e-05,
"loss": 0.0883,
"mean_token_accuracy": 0.9798992872238159,
"num_tokens": 2942729.0,
"step": 1380
},
{
"entropy": 2.2399872303009034,
"epoch": 0.4558920342330481,
"grad_norm": 4.462457180023193,
"learning_rate": 1.83214935845786e-05,
"loss": 0.0942,
"mean_token_accuracy": 0.9766364514827728,
"num_tokens": 2953444.0,
"step": 1385
},
{
"entropy": 2.3818634748458862,
"epoch": 0.4575378538512179,
"grad_norm": 2.3201687335968018,
"learning_rate": 1.830637118123113e-05,
"loss": 0.06,
"mean_token_accuracy": 0.987183290719986,
"num_tokens": 2963968.0,
"step": 1390
},
{
"entropy": 2.319228458404541,
"epoch": 0.45918367346938777,
"grad_norm": 1.5512436628341675,
"learning_rate": 1.8291187260454842e-05,
"loss": 0.0429,
"mean_token_accuracy": 0.9886379659175872,
"num_tokens": 2974544.0,
"step": 1395
},
{
"entropy": 2.212476873397827,
"epoch": 0.4608294930875576,
"grad_norm": 3.864231586456299,
"learning_rate": 1.827594193470266e-05,
"loss": 0.0719,
"mean_token_accuracy": 0.9817937731742858,
"num_tokens": 2985035.0,
"step": 1400
},
{
"entropy": 2.160155749320984,
"epoch": 0.46247531270572745,
"grad_norm": 4.27418851852417,
"learning_rate": 1.8260635316882288e-05,
"loss": 0.0752,
"mean_token_accuracy": 0.9822825253009796,
"num_tokens": 2995814.0,
"step": 1405
},
{
"entropy": 2.1970561265945436,
"epoch": 0.4641211323238973,
"grad_norm": 7.332748889923096,
"learning_rate": 1.8245267520355348e-05,
"loss": 0.0787,
"mean_token_accuracy": 0.9827774345874787,
"num_tokens": 3006260.0,
"step": 1410
},
{
"entropy": 2.339077877998352,
"epoch": 0.46576695194206713,
"grad_norm": 1.8340346813201904,
"learning_rate": 1.8229838658936566e-05,
"loss": 0.0763,
"mean_token_accuracy": 0.980904471874237,
"num_tokens": 3016932.0,
"step": 1415
},
{
"entropy": 2.3882060766220095,
"epoch": 0.467412771560237,
"grad_norm": 2.010164737701416,
"learning_rate": 1.8214348846892913e-05,
"loss": 0.0488,
"mean_token_accuracy": 0.9874413549900055,
"num_tokens": 3027677.0,
"step": 1420
},
{
"entropy": 2.291885328292847,
"epoch": 0.46905859117840687,
"grad_norm": 2.1564724445343018,
"learning_rate": 1.8198798198942768e-05,
"loss": 0.0702,
"mean_token_accuracy": 0.9820050477981568,
"num_tokens": 3038413.0,
"step": 1425
},
{
"entropy": 2.285910177230835,
"epoch": 0.4707044107965767,
"grad_norm": 3.993086576461792,
"learning_rate": 1.8183186830255058e-05,
"loss": 0.0707,
"mean_token_accuracy": 0.9839109480381012,
"num_tokens": 3049009.0,
"step": 1430
},
{
"entropy": 2.243725609779358,
"epoch": 0.47235023041474655,
"grad_norm": 3.17380952835083,
"learning_rate": 1.8167514856448413e-05,
"loss": 0.0509,
"mean_token_accuracy": 0.9843453109264374,
"num_tokens": 3059423.0,
"step": 1435
},
{
"entropy": 2.16868679523468,
"epoch": 0.4739960500329164,
"grad_norm": 2.891127824783325,
"learning_rate": 1.815178239359031e-05,
"loss": 0.0586,
"mean_token_accuracy": 0.9830320298671722,
"num_tokens": 3070009.0,
"step": 1440
},
{
"entropy": 2.055533969402313,
"epoch": 0.47564186965108624,
"grad_norm": 4.964085578918457,
"learning_rate": 1.8135989558196207e-05,
"loss": 0.039,
"mean_token_accuracy": 0.9861751973628998,
"num_tokens": 3080665.0,
"step": 1445
},
{
"entropy": 2.027339553833008,
"epoch": 0.4772876892692561,
"grad_norm": 7.009729862213135,
"learning_rate": 1.812013646722869e-05,
"loss": 0.096,
"mean_token_accuracy": 0.9795560419559479,
"num_tokens": 3090979.0,
"step": 1450
},
{
"entropy": 2.1618866443634035,
"epoch": 0.4789335088874259,
"grad_norm": 3.8036680221557617,
"learning_rate": 1.8104223238096596e-05,
"loss": 0.0902,
"mean_token_accuracy": 0.9777205526828766,
"num_tokens": 3101297.0,
"step": 1455
},
{
"entropy": 2.24229896068573,
"epoch": 0.48057932850559576,
"grad_norm": 2.5701446533203125,
"learning_rate": 1.808824998865415e-05,
"loss": 0.0397,
"mean_token_accuracy": 0.9876787722110748,
"num_tokens": 3111894.0,
"step": 1460
},
{
"entropy": 2.251754140853882,
"epoch": 0.48222514812376566,
"grad_norm": 5.5385847091674805,
"learning_rate": 1.8072216837200094e-05,
"loss": 0.0769,
"mean_token_accuracy": 0.9802028417587281,
"num_tokens": 3122674.0,
"step": 1465
},
{
"entropy": 2.185230827331543,
"epoch": 0.4838709677419355,
"grad_norm": 4.962352752685547,
"learning_rate": 1.80561239024768e-05,
"loss": 0.0529,
"mean_token_accuracy": 0.9858423829078674,
"num_tokens": 3133175.0,
"step": 1470
},
{
"entropy": 2.1357677221298217,
"epoch": 0.48551678736010534,
"grad_norm": 6.6793389320373535,
"learning_rate": 1.8039971303669407e-05,
"loss": 0.0698,
"mean_token_accuracy": 0.983429902791977,
"num_tokens": 3143713.0,
"step": 1475
},
{
"entropy": 2.1006932973861696,
"epoch": 0.4871626069782752,
"grad_norm": 1.9220439195632935,
"learning_rate": 1.8023759160404923e-05,
"loss": 0.0704,
"mean_token_accuracy": 0.9849179744720459,
"num_tokens": 3154604.0,
"step": 1480
},
{
"entropy": 2.1457018852233887,
"epoch": 0.488808426596445,
"grad_norm": 2.7263882160186768,
"learning_rate": 1.8007487592751343e-05,
"loss": 0.0578,
"mean_token_accuracy": 0.9825065851211547,
"num_tokens": 3165594.0,
"step": 1485
},
{
"entropy": 2.2146376371383667,
"epoch": 0.49045424621461486,
"grad_norm": 4.0069804191589355,
"learning_rate": 1.799115672121677e-05,
"loss": 0.0591,
"mean_token_accuracy": 0.9832662463188171,
"num_tokens": 3176301.0,
"step": 1490
},
{
"entropy": 2.2213491678237913,
"epoch": 0.4921000658327847,
"grad_norm": 2.8298332691192627,
"learning_rate": 1.7974766666748516e-05,
"loss": 0.0803,
"mean_token_accuracy": 0.9778595089912414,
"num_tokens": 3186898.0,
"step": 1495
},
{
"entropy": 2.203892719745636,
"epoch": 0.4937458854509546,
"grad_norm": 2.422091484069824,
"learning_rate": 1.7958317550732193e-05,
"loss": 0.0591,
"mean_token_accuracy": 0.9813643217086792,
"num_tokens": 3197716.0,
"step": 1500
},
{
"entropy": 2.1539814710617065,
"epoch": 0.49539170506912444,
"grad_norm": 1.6230992078781128,
"learning_rate": 1.7941809494990838e-05,
"loss": 0.0702,
"mean_token_accuracy": 0.9812954783439636,
"num_tokens": 3208681.0,
"step": 1505
},
{
"entropy": 2.0968565344810486,
"epoch": 0.4970375246872943,
"grad_norm": 4.276401042938232,
"learning_rate": 1.792524262178399e-05,
"loss": 0.0957,
"mean_token_accuracy": 0.9712316334247589,
"num_tokens": 3219313.0,
"step": 1510
},
{
"entropy": 1.9671455860137939,
"epoch": 0.4986833443054641,
"grad_norm": 2.3620660305023193,
"learning_rate": 1.7908617053806802e-05,
"loss": 0.0548,
"mean_token_accuracy": 0.9848778009414673,
"num_tokens": 3229908.0,
"step": 1515
},
{
"entropy": 1.9334992289543151,
"epoch": 0.500329163923634,
"grad_norm": 2.5839757919311523,
"learning_rate": 1.7891932914189112e-05,
"loss": 0.0542,
"mean_token_accuracy": 0.9889938116073609,
"num_tokens": 3240444.0,
"step": 1520
},
{
"entropy": 1.9328988552093507,
"epoch": 0.5019749835418038,
"grad_norm": 2.2053468227386475,
"learning_rate": 1.7875190326494552e-05,
"loss": 0.0706,
"mean_token_accuracy": 0.9844483017921448,
"num_tokens": 3251129.0,
"step": 1525
},
{
"entropy": 1.900172483921051,
"epoch": 0.5036208031599737,
"grad_norm": 3.773327112197876,
"learning_rate": 1.7858389414719628e-05,
"loss": 0.0567,
"mean_token_accuracy": 0.9843137383460998,
"num_tokens": 3261867.0,
"step": 1530
},
{
"entropy": 1.9451073169708253,
"epoch": 0.5052666227781435,
"grad_norm": 3.007178544998169,
"learning_rate": 1.7841530303292782e-05,
"loss": 0.0478,
"mean_token_accuracy": 0.9842455625534058,
"num_tokens": 3272277.0,
"step": 1535
},
{
"entropy": 1.9109816670417785,
"epoch": 0.5069124423963134,
"grad_norm": 6.694243431091309,
"learning_rate": 1.78246131170735e-05,
"loss": 0.0934,
"mean_token_accuracy": 0.9757687628269196,
"num_tokens": 3282740.0,
"step": 1540
},
{
"entropy": 1.945954716205597,
"epoch": 0.5085582620144832,
"grad_norm": 4.0865936279296875,
"learning_rate": 1.780763798135136e-05,
"loss": 0.055,
"mean_token_accuracy": 0.9839641332626343,
"num_tokens": 3293583.0,
"step": 1545
},
{
"entropy": 1.8939694523811341,
"epoch": 0.5102040816326531,
"grad_norm": 5.349853038787842,
"learning_rate": 1.779060502184513e-05,
"loss": 0.093,
"mean_token_accuracy": 0.9804658055305481,
"num_tokens": 3304579.0,
"step": 1550
},
{
"entropy": 1.9622906923294068,
"epoch": 0.5118499012508229,
"grad_norm": 4.101441860198975,
"learning_rate": 1.777351436470182e-05,
"loss": 0.0978,
"mean_token_accuracy": 0.9735360503196716,
"num_tokens": 3315316.0,
"step": 1555
},
{
"entropy": 2.1062827825546266,
"epoch": 0.5134957208689928,
"grad_norm": 2.2984564304351807,
"learning_rate": 1.775636613649574e-05,
"loss": 0.0659,
"mean_token_accuracy": 0.9824669599533081,
"num_tokens": 3325816.0,
"step": 1560
},
{
"entropy": 2.0575469732284546,
"epoch": 0.5151415404871627,
"grad_norm": 3.7546675205230713,
"learning_rate": 1.7739160464227593e-05,
"loss": 0.0902,
"mean_token_accuracy": 0.9760908842086792,
"num_tokens": 3336506.0,
"step": 1565
},
{
"entropy": 1.938406229019165,
"epoch": 0.5167873601053324,
"grad_norm": 2.3365416526794434,
"learning_rate": 1.7721897475323508e-05,
"loss": 0.084,
"mean_token_accuracy": 0.9817908525466919,
"num_tokens": 3346922.0,
"step": 1570
},
{
"entropy": 1.9214160919189454,
"epoch": 0.5184331797235023,
"grad_norm": 3.7224996089935303,
"learning_rate": 1.7704577297634096e-05,
"loss": 0.0595,
"mean_token_accuracy": 0.9839179992675782,
"num_tokens": 3357419.0,
"step": 1575
},
{
"entropy": 1.8380493998527527,
"epoch": 0.5200789993416721,
"grad_norm": 4.267509937286377,
"learning_rate": 1.768720005943353e-05,
"loss": 0.087,
"mean_token_accuracy": 0.9799594342708587,
"num_tokens": 3368205.0,
"step": 1580
},
{
"entropy": 1.8690474152565002,
"epoch": 0.521724818959842,
"grad_norm": 2.147115468978882,
"learning_rate": 1.7669765889418553e-05,
"loss": 0.1006,
"mean_token_accuracy": 0.9794728994369507,
"num_tokens": 3378592.0,
"step": 1585
},
{
"entropy": 1.880600130558014,
"epoch": 0.5233706385780118,
"grad_norm": 3.2792811393737793,
"learning_rate": 1.7652274916707566e-05,
"loss": 0.0709,
"mean_token_accuracy": 0.983178836107254,
"num_tokens": 3389050.0,
"step": 1590
},
{
"entropy": 1.9782520651817321,
"epoch": 0.5250164581961817,
"grad_norm": 2.2735812664031982,
"learning_rate": 1.7634727270839645e-05,
"loss": 0.0628,
"mean_token_accuracy": 0.9815650939941406,
"num_tokens": 3399987.0,
"step": 1595
},
{
"entropy": 2.048723006248474,
"epoch": 0.5266622778143516,
"grad_norm": 6.50067138671875,
"learning_rate": 1.761712308177359e-05,
"loss": 0.1481,
"mean_token_accuracy": 0.959316509962082,
"num_tokens": 3410677.0,
"step": 1600
},
{
"entropy": 2.094959032535553,
"epoch": 0.5283080974325214,
"grad_norm": 7.985395908355713,
"learning_rate": 1.7599462479886976e-05,
"loss": 0.0756,
"mean_token_accuracy": 0.9803730607032776,
"num_tokens": 3421071.0,
"step": 1605
},
{
"entropy": 2.046134078502655,
"epoch": 0.5299539170506913,
"grad_norm": 5.17471981048584,
"learning_rate": 1.7581745595975158e-05,
"loss": 0.0956,
"mean_token_accuracy": 0.9822999238967896,
"num_tokens": 3432117.0,
"step": 1610
},
{
"entropy": 2.171485185623169,
"epoch": 0.5315997366688611,
"grad_norm": 1.3401827812194824,
"learning_rate": 1.7563972561250323e-05,
"loss": 0.0513,
"mean_token_accuracy": 0.9851057410240174,
"num_tokens": 3442596.0,
"step": 1615
},
{
"entropy": 2.094390308856964,
"epoch": 0.533245556287031,
"grad_norm": 5.245898246765137,
"learning_rate": 1.7546143507340517e-05,
"loss": 0.068,
"mean_token_accuracy": 0.9802603662014008,
"num_tokens": 3453151.0,
"step": 1620
},
{
"entropy": 2.0529758810997008,
"epoch": 0.5348913759052007,
"grad_norm": 2.95365309715271,
"learning_rate": 1.7528258566288666e-05,
"loss": 0.0872,
"mean_token_accuracy": 0.9790965497493744,
"num_tokens": 3463685.0,
"step": 1625
},
{
"entropy": 2.0168527364730835,
"epoch": 0.5365371955233706,
"grad_norm": 3.4701573848724365,
"learning_rate": 1.75103178705516e-05,
"loss": 0.0809,
"mean_token_accuracy": 0.979897940158844,
"num_tokens": 3474671.0,
"step": 1630
},
{
"entropy": 2.0435917496681215,
"epoch": 0.5381830151415404,
"grad_norm": 2.95245361328125,
"learning_rate": 1.7492321552999076e-05,
"loss": 0.0423,
"mean_token_accuracy": 0.9878913462162018,
"num_tokens": 3485155.0,
"step": 1635
},
{
"entropy": 1.9816929221153259,
"epoch": 0.5398288347597103,
"grad_norm": 3.2681093215942383,
"learning_rate": 1.747426974691277e-05,
"loss": 0.053,
"mean_token_accuracy": 0.9853954315185547,
"num_tokens": 3495688.0,
"step": 1640
},
{
"entropy": 1.8952534437179565,
"epoch": 0.5414746543778802,
"grad_norm": 3.5420217514038086,
"learning_rate": 1.7456162585985335e-05,
"loss": 0.055,
"mean_token_accuracy": 0.9834610342979431,
"num_tokens": 3506316.0,
"step": 1645
},
{
"entropy": 1.9181370496749879,
"epoch": 0.54312047399605,
"grad_norm": 2.8061180114746094,
"learning_rate": 1.7438000204319365e-05,
"loss": 0.0615,
"mean_token_accuracy": 0.9796540081501007,
"num_tokens": 3517096.0,
"step": 1650
},
{
"entropy": 1.900493335723877,
"epoch": 0.5447662936142199,
"grad_norm": 6.022247791290283,
"learning_rate": 1.7419782736426433e-05,
"loss": 0.0686,
"mean_token_accuracy": 0.9824603438377381,
"num_tokens": 3527608.0,
"step": 1655
},
{
"entropy": 1.9555213689804076,
"epoch": 0.5464121132323897,
"grad_norm": 2.4826130867004395,
"learning_rate": 1.7401510317226077e-05,
"loss": 0.0602,
"mean_token_accuracy": 0.9860040962696075,
"num_tokens": 3538016.0,
"step": 1660
},
{
"entropy": 1.9975322604179382,
"epoch": 0.5480579328505596,
"grad_norm": 1.6209336519241333,
"learning_rate": 1.7383183082044814e-05,
"loss": 0.0632,
"mean_token_accuracy": 0.9838661909103393,
"num_tokens": 3548608.0,
"step": 1665
},
{
"entropy": 2.0439995765686034,
"epoch": 0.5497037524687294,
"grad_norm": 2.632179021835327,
"learning_rate": 1.7364801166615124e-05,
"loss": 0.0615,
"mean_token_accuracy": 0.9858087658882141,
"num_tokens": 3558846.0,
"step": 1670
},
{
"entropy": 2.084509313106537,
"epoch": 0.5513495720868993,
"grad_norm": 2.9909346103668213,
"learning_rate": 1.7346364707074453e-05,
"loss": 0.045,
"mean_token_accuracy": 0.9885495722293853,
"num_tokens": 3569553.0,
"step": 1675
},
{
"entropy": 2.1361532688140867,
"epoch": 0.5529953917050692,
"grad_norm": 2.4530577659606934,
"learning_rate": 1.732787383996421e-05,
"loss": 0.0783,
"mean_token_accuracy": 0.9758898675441742,
"num_tokens": 3580282.0,
"step": 1680
},
{
"entropy": 2.134485626220703,
"epoch": 0.554641211323239,
"grad_norm": 2.1198272705078125,
"learning_rate": 1.7309328702228742e-05,
"loss": 0.0757,
"mean_token_accuracy": 0.9795774221420288,
"num_tokens": 3591130.0,
"step": 1685
},
{
"entropy": 2.2107550859451295,
"epoch": 0.5562870309414089,
"grad_norm": 3.947378158569336,
"learning_rate": 1.729072943121433e-05,
"loss": 0.1107,
"mean_token_accuracy": 0.9773122131824493,
"num_tokens": 3601615.0,
"step": 1690
},
{
"entropy": 2.156628942489624,
"epoch": 0.5579328505595786,
"grad_norm": 5.117584228515625,
"learning_rate": 1.727207616466817e-05,
"loss": 0.0573,
"mean_token_accuracy": 0.9852468788623809,
"num_tokens": 3612527.0,
"step": 1695
},
{
"entropy": 2.1495004177093504,
"epoch": 0.5595786701777485,
"grad_norm": 3.0924577713012695,
"learning_rate": 1.725336904073735e-05,
"loss": 0.0499,
"mean_token_accuracy": 0.9855070650577545,
"num_tokens": 3623196.0,
"step": 1700
},
{
"entropy": 2.139863908290863,
"epoch": 0.5612244897959183,
"grad_norm": 2.3820176124572754,
"learning_rate": 1.723460819796783e-05,
"loss": 0.0439,
"mean_token_accuracy": 0.9874242305755615,
"num_tokens": 3634081.0,
"step": 1705
},
{
"entropy": 2.0737234473228456,
"epoch": 0.5628703094140882,
"grad_norm": 4.342321395874023,
"learning_rate": 1.7215793775303415e-05,
"loss": 0.0857,
"mean_token_accuracy": 0.9764466941356659,
"num_tokens": 3644499.0,
"step": 1710
},
{
"entropy": 2.0394096612930297,
"epoch": 0.5645161290322581,
"grad_norm": 2.6576485633850098,
"learning_rate": 1.719692591208472e-05,
"loss": 0.0679,
"mean_token_accuracy": 0.9818757474422455,
"num_tokens": 3655219.0,
"step": 1715
},
{
"entropy": 2.023374152183533,
"epoch": 0.5661619486504279,
"grad_norm": 1.6631029844284058,
"learning_rate": 1.7178004748048157e-05,
"loss": 0.0681,
"mean_token_accuracy": 0.9818334817886353,
"num_tokens": 3665785.0,
"step": 1720
},
{
"entropy": 2.0491710901260376,
"epoch": 0.5678077682685978,
"grad_norm": 1.3181400299072266,
"learning_rate": 1.7159030423324873e-05,
"loss": 0.0395,
"mean_token_accuracy": 0.9904541313648224,
"num_tokens": 3676536.0,
"step": 1725
},
{
"entropy": 1.9250925421714782,
"epoch": 0.5694535878867676,
"grad_norm": 3.0907676219940186,
"learning_rate": 1.7140003078439727e-05,
"loss": 0.0726,
"mean_token_accuracy": 0.9785450220108032,
"num_tokens": 3687519.0,
"step": 1730
},
{
"entropy": 1.9846270561218262,
"epoch": 0.5710994075049375,
"grad_norm": 2.5993564128875732,
"learning_rate": 1.712092285431026e-05,
"loss": 0.0729,
"mean_token_accuracy": 0.9780683517456055,
"num_tokens": 3698184.0,
"step": 1735
},
{
"entropy": 1.9407028794288634,
"epoch": 0.5727452271231073,
"grad_norm": 2.01981520652771,
"learning_rate": 1.710178989224562e-05,
"loss": 0.0693,
"mean_token_accuracy": 0.9816851794719696,
"num_tokens": 3709017.0,
"step": 1740
},
{
"entropy": 1.9194607377052306,
"epoch": 0.5743910467412772,
"grad_norm": 2.252833843231201,
"learning_rate": 1.7082604333945557e-05,
"loss": 0.0855,
"mean_token_accuracy": 0.9747888624668122,
"num_tokens": 3719609.0,
"step": 1745
},
{
"entropy": 1.9611817717552185,
"epoch": 0.576036866359447,
"grad_norm": 6.348813056945801,
"learning_rate": 1.7063366321499338e-05,
"loss": 0.0764,
"mean_token_accuracy": 0.9813452184200286,
"num_tokens": 3730007.0,
"step": 1750
},
{
"entropy": 1.9477728486061097,
"epoch": 0.5776826859776168,
"grad_norm": 2.917712926864624,
"learning_rate": 1.7044075997384707e-05,
"loss": 0.0559,
"mean_token_accuracy": 0.9842540562152863,
"num_tokens": 3740830.0,
"step": 1755
},
{
"entropy": 2.0349454164505003,
"epoch": 0.5793285055957867,
"grad_norm": 2.951350212097168,
"learning_rate": 1.7024733504466843e-05,
"loss": 0.0465,
"mean_token_accuracy": 0.9853815793991089,
"num_tokens": 3751273.0,
"step": 1760
},
{
"entropy": 2.0125353813171385,
"epoch": 0.5809743252139565,
"grad_norm": 2.382908582687378,
"learning_rate": 1.7005338985997273e-05,
"loss": 0.0593,
"mean_token_accuracy": 0.9872986257076264,
"num_tokens": 3761728.0,
"step": 1765
},
{
"entropy": 2.04194176197052,
"epoch": 0.5826201448321264,
"grad_norm": 5.703618049621582,
"learning_rate": 1.6985892585612848e-05,
"loss": 0.1025,
"mean_token_accuracy": 0.9684024691581726,
"num_tokens": 3772794.0,
"step": 1770
},
{
"entropy": 2.026109528541565,
"epoch": 0.5842659644502962,
"grad_norm": 2.672579288482666,
"learning_rate": 1.6966394447334638e-05,
"loss": 0.0634,
"mean_token_accuracy": 0.9840988755226135,
"num_tokens": 3783388.0,
"step": 1775
},
{
"entropy": 2.0058682441711424,
"epoch": 0.5859117840684661,
"grad_norm": 2.395892858505249,
"learning_rate": 1.69468447155669e-05,
"loss": 0.0432,
"mean_token_accuracy": 0.9877268433570862,
"num_tokens": 3793954.0,
"step": 1780
},
{
"entropy": 1.9445565462112426,
"epoch": 0.5875576036866359,
"grad_norm": 2.7134897708892822,
"learning_rate": 1.6927243535095995e-05,
"loss": 0.077,
"mean_token_accuracy": 0.9795953154563903,
"num_tokens": 3804724.0,
"step": 1785
},
{
"entropy": 1.936465561389923,
"epoch": 0.5892034233048058,
"grad_norm": 1.9062567949295044,
"learning_rate": 1.6907591051089313e-05,
"loss": 0.0685,
"mean_token_accuracy": 0.9803402423858643,
"num_tokens": 3815254.0,
"step": 1790
},
{
"entropy": 1.9493210196495057,
"epoch": 0.5908492429229757,
"grad_norm": 3.1633780002593994,
"learning_rate": 1.6887887409094195e-05,
"loss": 0.0475,
"mean_token_accuracy": 0.9850197255611419,
"num_tokens": 3825885.0,
"step": 1795
},
{
"entropy": 1.8948396682739257,
"epoch": 0.5924950625411455,
"grad_norm": 3.2793633937835693,
"learning_rate": 1.6868132755036875e-05,
"loss": 0.0539,
"mean_token_accuracy": 0.9861530363559723,
"num_tokens": 3836784.0,
"step": 1800
},
{
"entropy": 1.935602605342865,
"epoch": 0.5941408821593154,
"grad_norm": 3.2881906032562256,
"learning_rate": 1.6848327235221368e-05,
"loss": 0.0827,
"mean_token_accuracy": 0.9799518406391143,
"num_tokens": 3847307.0,
"step": 1805
},
{
"entropy": 1.9104020118713378,
"epoch": 0.5957867017774852,
"grad_norm": 1.5671672821044922,
"learning_rate": 1.6828470996328418e-05,
"loss": 0.0566,
"mean_token_accuracy": 0.9839206755161285,
"num_tokens": 3858140.0,
"step": 1810
},
{
"entropy": 1.9541502475738526,
"epoch": 0.5974325213956551,
"grad_norm": 3.0334932804107666,
"learning_rate": 1.680856418541439e-05,
"loss": 0.0717,
"mean_token_accuracy": 0.9826797246932983,
"num_tokens": 3868625.0,
"step": 1815
},
{
"entropy": 1.9071632623672485,
"epoch": 0.5990783410138248,
"grad_norm": 3.8940181732177734,
"learning_rate": 1.6788606949910188e-05,
"loss": 0.0591,
"mean_token_accuracy": 0.9818158507347107,
"num_tokens": 3879340.0,
"step": 1820
},
{
"epoch": 0.6003949967083607,
"eval_entropy": 1.9597396565445562,
"eval_loss": 0.06656259298324585,
"eval_mean_token_accuracy": 0.9824842680084602,
"eval_num_tokens": 3887690.0,
"eval_runtime": 195.6298,
"eval_samples_per_second": 42.565,
"eval_steps_per_second": 7.095,
"step": 1824
},
{
"entropy": 1.948818302154541,
"epoch": 0.6007241606319947,
"grad_norm": 2.146757125854492,
"learning_rate": 1.6768599437620166e-05,
"loss": 0.0598,
"mean_token_accuracy": 0.9824907779693604,
"num_tokens": 3889882.0,
"step": 1825
},
{
"entropy": 1.953722858428955,
"epoch": 0.6023699802501645,
"grad_norm": 4.7035112380981445,
"learning_rate": 1.6748541796721026e-05,
"loss": 0.0759,
"mean_token_accuracy": 0.9829168796539307,
"num_tokens": 3900763.0,
"step": 1830
},
{
"entropy": 1.9083998680114747,
"epoch": 0.6040157998683344,
"grad_norm": 3.567713737487793,
"learning_rate": 1.6728434175760733e-05,
"loss": 0.0602,
"mean_token_accuracy": 0.9855620503425598,
"num_tokens": 3911307.0,
"step": 1835
},
{
"entropy": 1.9374099254608155,
"epoch": 0.6056616194865043,
"grad_norm": 5.015064716339111,
"learning_rate": 1.6708276723657396e-05,
"loss": 0.0556,
"mean_token_accuracy": 0.986617261171341,
"num_tokens": 3921526.0,
"step": 1840
},
{
"entropy": 1.9269861102104187,
"epoch": 0.6073074391046741,
"grad_norm": 3.2826483249664307,
"learning_rate": 1.6688069589698194e-05,
"loss": 0.0727,
"mean_token_accuracy": 0.9803520023822785,
"num_tokens": 3932144.0,
"step": 1845
},
{
"entropy": 1.8954154014587403,
"epoch": 0.608953258722844,
"grad_norm": 6.194475173950195,
"learning_rate": 1.6667812923538226e-05,
"loss": 0.0841,
"mean_token_accuracy": 0.9779143512248993,
"num_tokens": 3943304.0,
"step": 1850
},
{
"entropy": 1.989688503742218,
"epoch": 0.6105990783410138,
"grad_norm": 4.787328243255615,
"learning_rate": 1.664750687519945e-05,
"loss": 0.0542,
"mean_token_accuracy": 0.9866823434829712,
"num_tokens": 3953946.0,
"step": 1855
},
{
"entropy": 2.018328094482422,
"epoch": 0.6122448979591837,
"grad_norm": 6.004918098449707,
"learning_rate": 1.662715159506955e-05,
"loss": 0.0683,
"mean_token_accuracy": 0.9783054530620575,
"num_tokens": 3964563.0,
"step": 1860
},
{
"entropy": 2.0147560954093935,
"epoch": 0.6138907175773535,
"grad_norm": 3.540149450302124,
"learning_rate": 1.6606747233900816e-05,
"loss": 0.0612,
"mean_token_accuracy": 0.983444994688034,
"num_tokens": 3974938.0,
"step": 1865
},
{
"entropy": 2.0170241355895997,
"epoch": 0.6155365371955234,
"grad_norm": 2.3520843982696533,
"learning_rate": 1.6586293942809034e-05,
"loss": 0.0675,
"mean_token_accuracy": 0.980956619977951,
"num_tokens": 3985380.0,
"step": 1870
},
{
"entropy": 2.022734725475311,
"epoch": 0.6171823568136933,
"grad_norm": 2.997664451599121,
"learning_rate": 1.6565791873272373e-05,
"loss": 0.0435,
"mean_token_accuracy": 0.9847745478153229,
"num_tokens": 3995690.0,
"step": 1875
},
{
"entropy": 2.0023676991462707,
"epoch": 0.618828176431863,
"grad_norm": 2.1239171028137207,
"learning_rate": 1.6545241177130254e-05,
"loss": 0.0486,
"mean_token_accuracy": 0.9879401743412017,
"num_tokens": 4006475.0,
"step": 1880
},
{
"entropy": 1.9986587166786194,
"epoch": 0.620473996050033,
"grad_norm": 3.1917483806610107,
"learning_rate": 1.652464200658223e-05,
"loss": 0.0688,
"mean_token_accuracy": 0.9839587926864624,
"num_tokens": 4017079.0,
"step": 1885
},
{
"entropy": 2.038959813117981,
"epoch": 0.6221198156682027,
"grad_norm": 2.038222551345825,
"learning_rate": 1.650399451418686e-05,
"loss": 0.0706,
"mean_token_accuracy": 0.98506378531456,
"num_tokens": 4027591.0,
"step": 1890
},
{
"entropy": 2.021359896659851,
"epoch": 0.6237656352863726,
"grad_norm": 3.2785909175872803,
"learning_rate": 1.6483298852860584e-05,
"loss": 0.063,
"mean_token_accuracy": 0.9845045149326325,
"num_tokens": 4038090.0,
"step": 1895
},
{
"entropy": 2.045257091522217,
"epoch": 0.6254114549045424,
"grad_norm": 1.1910496950149536,
"learning_rate": 1.646255517587656e-05,
"loss": 0.0417,
"mean_token_accuracy": 0.9895485579967499,
"num_tokens": 4048536.0,
"step": 1900
},
{
"entropy": 2.0391509175300597,
"epoch": 0.6270572745227123,
"grad_norm": 6.128629207611084,
"learning_rate": 1.644176363686358e-05,
"loss": 0.0824,
"mean_token_accuracy": 0.9755302667617798,
"num_tokens": 4059195.0,
"step": 1905
},
{
"entropy": 2.0614427804946898,
"epoch": 0.6287030941408822,
"grad_norm": 2.5974950790405273,
"learning_rate": 1.6420924389804887e-05,
"loss": 0.0469,
"mean_token_accuracy": 0.989526915550232,
"num_tokens": 4069610.0,
"step": 1910
},
{
"entropy": 2.0109457969665527,
"epoch": 0.630348913759052,
"grad_norm": 4.8466362953186035,
"learning_rate": 1.6400037589037062e-05,
"loss": 0.0814,
"mean_token_accuracy": 0.9819941699504853,
"num_tokens": 4080106.0,
"step": 1915
},
{
"entropy": 2.0402795314788817,
"epoch": 0.6319947333772219,
"grad_norm": 3.8476603031158447,
"learning_rate": 1.6379103389248867e-05,
"loss": 0.0909,
"mean_token_accuracy": 0.9787125289440155,
"num_tokens": 4090739.0,
"step": 1920
},
{
"entropy": 2.1076374292373656,
"epoch": 0.6336405529953917,
"grad_norm": 3.3376083374023438,
"learning_rate": 1.63581219454801e-05,
"loss": 0.0598,
"mean_token_accuracy": 0.98504838347435,
"num_tokens": 4101214.0,
"step": 1925
},
{
"entropy": 2.0517534494400023,
"epoch": 0.6352863726135616,
"grad_norm": 6.078008651733398,
"learning_rate": 1.6337093413120463e-05,
"loss": 0.0745,
"mean_token_accuracy": 0.9847012758255005,
"num_tokens": 4111702.0,
"step": 1930
},
{
"entropy": 2.0724000096321107,
"epoch": 0.6369321922317314,
"grad_norm": 1.9413197040557861,
"learning_rate": 1.631601794790838e-05,
"loss": 0.0533,
"mean_token_accuracy": 0.9855819940567017,
"num_tokens": 4122503.0,
"step": 1935
},
{
"entropy": 2.0283157587051392,
"epoch": 0.6385780118499013,
"grad_norm": 3.654207468032837,
"learning_rate": 1.629489570592988e-05,
"loss": 0.0704,
"mean_token_accuracy": 0.9817177176475524,
"num_tokens": 4132981.0,
"step": 1940
},
{
"entropy": 1.9633773922920228,
"epoch": 0.640223831468071,
"grad_norm": 5.509681701660156,
"learning_rate": 1.6273726843617413e-05,
"loss": 0.0735,
"mean_token_accuracy": 0.9831919133663177,
"num_tokens": 4143616.0,
"step": 1945
},
{
"entropy": 1.8812169671058654,
"epoch": 0.6418696510862409,
"grad_norm": 5.453320503234863,
"learning_rate": 1.625251151774871e-05,
"loss": 0.0743,
"mean_token_accuracy": 0.9811854064464569,
"num_tokens": 4154118.0,
"step": 1950
},
{
"entropy": 1.8234861969947815,
"epoch": 0.6435154707044108,
"grad_norm": 1.3750858306884766,
"learning_rate": 1.62312498854456e-05,
"loss": 0.0351,
"mean_token_accuracy": 0.987557852268219,
"num_tokens": 4165209.0,
"step": 1955
},
{
"entropy": 1.8196370124816894,
"epoch": 0.6451612903225806,
"grad_norm": 3.4758753776550293,
"learning_rate": 1.620994210417287e-05,
"loss": 0.0615,
"mean_token_accuracy": 0.9843147337436676,
"num_tokens": 4175834.0,
"step": 1960
},
{
"entropy": 1.8451374650001526,
"epoch": 0.6468071099407505,
"grad_norm": 0.8775233626365662,
"learning_rate": 1.6188588331737086e-05,
"loss": 0.1016,
"mean_token_accuracy": 0.9828826725482941,
"num_tokens": 4186493.0,
"step": 1965
},
{
"entropy": 1.8466584086418152,
"epoch": 0.6484529295589203,
"grad_norm": 2.0667057037353516,
"learning_rate": 1.6167188726285433e-05,
"loss": 0.041,
"mean_token_accuracy": 0.985245656967163,
"num_tokens": 4196992.0,
"step": 1970
},
{
"entropy": 1.820866048336029,
"epoch": 0.6500987491770902,
"grad_norm": 1.9866774082183838,
"learning_rate": 1.6145743446304524e-05,
"loss": 0.0521,
"mean_token_accuracy": 0.9827068746089935,
"num_tokens": 4207748.0,
"step": 1975
},
{
"entropy": 1.7237545490264892,
"epoch": 0.65174456879526,
"grad_norm": 4.054563045501709,
"learning_rate": 1.6124252650619257e-05,
"loss": 0.0721,
"mean_token_accuracy": 0.9837478876113892,
"num_tokens": 4218653.0,
"step": 1980
},
{
"entropy": 1.7705234289169312,
"epoch": 0.6533903884134299,
"grad_norm": 2.3912136554718018,
"learning_rate": 1.610271649839161e-05,
"loss": 0.0548,
"mean_token_accuracy": 0.9835083067417145,
"num_tokens": 4229354.0,
"step": 1985
},
{
"entropy": 1.733774983882904,
"epoch": 0.6550362080315998,
"grad_norm": 7.83612060546875,
"learning_rate": 1.608113514911948e-05,
"loss": 0.0886,
"mean_token_accuracy": 0.9790404379367829,
"num_tokens": 4240302.0,
"step": 1990
},
{
"entropy": 1.6919760227203369,
"epoch": 0.6566820276497696,
"grad_norm": 4.122557163238525,
"learning_rate": 1.6059508762635482e-05,
"loss": 0.0658,
"mean_token_accuracy": 0.9874409794807434,
"num_tokens": 4250965.0,
"step": 1995
},
{
"entropy": 1.727248990535736,
"epoch": 0.6583278472679395,
"grad_norm": 2.4015748500823975,
"learning_rate": 1.6037837499105804e-05,
"loss": 0.0619,
"mean_token_accuracy": 0.9821118116378784,
"num_tokens": 4261750.0,
"step": 2000
},
{
"entropy": 1.7900878310203552,
"epoch": 0.6599736668861093,
"grad_norm": 3.262315034866333,
"learning_rate": 1.601612151902897e-05,
"loss": 0.053,
"mean_token_accuracy": 0.9853106379508972,
"num_tokens": 4272360.0,
"step": 2005
},
{
"entropy": 1.8539373755455018,
"epoch": 0.6616194865042792,
"grad_norm": 3.7115604877471924,
"learning_rate": 1.5994360983234698e-05,
"loss": 0.0681,
"mean_token_accuracy": 0.9834797859191895,
"num_tokens": 4283012.0,
"step": 2010
},
{
"entropy": 1.888733983039856,
"epoch": 0.6632653061224489,
"grad_norm": 1.2049510478973389,
"learning_rate": 1.5972556052882672e-05,
"loss": 0.0399,
"mean_token_accuracy": 0.9876511216163635,
"num_tokens": 4293353.0,
"step": 2015
},
{
"entropy": 1.8981561660766602,
"epoch": 0.6649111257406188,
"grad_norm": 4.529847621917725,
"learning_rate": 1.595070688946138e-05,
"loss": 0.0757,
"mean_token_accuracy": 0.9844479858875275,
"num_tokens": 4303907.0,
"step": 2020
},
{
"entropy": 1.9379532098770142,
"epoch": 0.6665569453587886,
"grad_norm": 1.7552553415298462,
"learning_rate": 1.592881365478688e-05,
"loss": 0.0662,
"mean_token_accuracy": 0.983349347114563,
"num_tokens": 4314505.0,
"step": 2025
},
{
"entropy": 1.9738388657569885,
"epoch": 0.6682027649769585,
"grad_norm": 2.010423183441162,
"learning_rate": 1.590687651100165e-05,
"loss": 0.0707,
"mean_token_accuracy": 0.9822420716285706,
"num_tokens": 4325178.0,
"step": 2030
},
{
"entropy": 1.9870564699172975,
"epoch": 0.6698485845951284,
"grad_norm": 3.9561121463775635,
"learning_rate": 1.5884895620573346e-05,
"loss": 0.0568,
"mean_token_accuracy": 0.9854467332363128,
"num_tokens": 4335932.0,
"step": 2035
},
{
"entropy": 1.989486038684845,
"epoch": 0.6714944042132982,
"grad_norm": 2.525803565979004,
"learning_rate": 1.5862871146293616e-05,
"loss": 0.0593,
"mean_token_accuracy": 0.9792122423648835,
"num_tokens": 4346565.0,
"step": 2040
},
{
"entropy": 1.9590755701065063,
"epoch": 0.6731402238314681,
"grad_norm": 3.8919122219085693,
"learning_rate": 1.5840803251276892e-05,
"loss": 0.0805,
"mean_token_accuracy": 0.9800050914287567,
"num_tokens": 4357420.0,
"step": 2045
},
{
"entropy": 2.030180037021637,
"epoch": 0.6747860434496379,
"grad_norm": 3.5018019676208496,
"learning_rate": 1.5818692098959187e-05,
"loss": 0.099,
"mean_token_accuracy": 0.9742182731628418,
"num_tokens": 4368006.0,
"step": 2050
},
{
"entropy": 2.0822962045669557,
"epoch": 0.6764318630678078,
"grad_norm": 4.686304092407227,
"learning_rate": 1.5796537853096875e-05,
"loss": 0.0766,
"mean_token_accuracy": 0.9759474039077759,
"num_tokens": 4378686.0,
"step": 2055
},
{
"entropy": 2.134053909778595,
"epoch": 0.6780776826859776,
"grad_norm": 4.483613014221191,
"learning_rate": 1.5774340677765483e-05,
"loss": 0.0743,
"mean_token_accuracy": 0.9787837326526642,
"num_tokens": 4389381.0,
"step": 2060
},
{
"entropy": 2.101311683654785,
"epoch": 0.6797235023041475,
"grad_norm": 4.380194187164307,
"learning_rate": 1.575210073735848e-05,
"loss": 0.0875,
"mean_token_accuracy": 0.9795165956020355,
"num_tokens": 4400273.0,
"step": 2065
},
{
"entropy": 2.1265645384788514,
"epoch": 0.6813693219223174,
"grad_norm": 5.121950626373291,
"learning_rate": 1.572981819658605e-05,
"loss": 0.0539,
"mean_token_accuracy": 0.9842699348926545,
"num_tokens": 4411134.0,
"step": 2070
},
{
"entropy": 2.063078737258911,
"epoch": 0.6830151415404871,
"grad_norm": 1.915604829788208,
"learning_rate": 1.5707493220473886e-05,
"loss": 0.039,
"mean_token_accuracy": 0.9892246127128601,
"num_tokens": 4421711.0,
"step": 2075
},
{
"entropy": 2.0187936067581176,
"epoch": 0.684660961158657,
"grad_norm": 6.068004131317139,
"learning_rate": 1.568512597436195e-05,
"loss": 0.0724,
"mean_token_accuracy": 0.9800882399082184,
"num_tokens": 4432281.0,
"step": 2080
},
{
"entropy": 2.019049596786499,
"epoch": 0.6863067807768268,
"grad_norm": 3.6800379753112793,
"learning_rate": 1.566271662390326e-05,
"loss": 0.0666,
"mean_token_accuracy": 0.9861168086528778,
"num_tokens": 4442927.0,
"step": 2085
},
{
"entropy": 2.0945070028305053,
"epoch": 0.6879526003949967,
"grad_norm": 5.768988132476807,
"learning_rate": 1.564026533506267e-05,
"loss": 0.0637,
"mean_token_accuracy": 0.9825628995895386,
"num_tokens": 4453272.0,
"step": 2090
},
{
"entropy": 2.1388468503952027,
"epoch": 0.6895984200131665,
"grad_norm": 3.217071294784546,
"learning_rate": 1.5617772274115618e-05,
"loss": 0.0478,
"mean_token_accuracy": 0.9849779725074768,
"num_tokens": 4464006.0,
"step": 2095
},
{
"entropy": 2.1074278712272645,
"epoch": 0.6912442396313364,
"grad_norm": 2.122255563735962,
"learning_rate": 1.559523760764692e-05,
"loss": 0.1015,
"mean_token_accuracy": 0.9739717364311218,
"num_tokens": 4475018.0,
"step": 2100
},
{
"entropy": 2.1118789672851563,
"epoch": 0.6928900592495063,
"grad_norm": 2.033972978591919,
"learning_rate": 1.5572661502549514e-05,
"loss": 0.0485,
"mean_token_accuracy": 0.9862671077251435,
"num_tokens": 4485676.0,
"step": 2105
},
{
"entropy": 2.1021223068237305,
"epoch": 0.6945358788676761,
"grad_norm": 0.6050166487693787,
"learning_rate": 1.5550044126023245e-05,
"loss": 0.0395,
"mean_token_accuracy": 0.9876242458820343,
"num_tokens": 4496186.0,
"step": 2110
},
{
"entropy": 1.970346164703369,
"epoch": 0.696181698485846,
"grad_norm": 3.1835389137268066,
"learning_rate": 1.5527385645573613e-05,
"loss": 0.0538,
"mean_token_accuracy": 0.9837459921836853,
"num_tokens": 4507128.0,
"step": 2115
},
{
"entropy": 1.9633802771568298,
"epoch": 0.6978275181040158,
"grad_norm": 0.6275081634521484,
"learning_rate": 1.5504686229010535e-05,
"loss": 0.0388,
"mean_token_accuracy": 0.987060683965683,
"num_tokens": 4517561.0,
"step": 2120
},
{
"entropy": 1.889497458934784,
"epoch": 0.6994733377221857,
"grad_norm": 8.642451286315918,
"learning_rate": 1.54819460444471e-05,
"loss": 0.0617,
"mean_token_accuracy": 0.9837220013141632,
"num_tokens": 4528490.0,
"step": 2125
},
{
"entropy": 1.8756322503089904,
"epoch": 0.7011191573403555,
"grad_norm": 1.7640129327774048,
"learning_rate": 1.545916526029833e-05,
"loss": 0.0606,
"mean_token_accuracy": 0.9831870555877685,
"num_tokens": 4539249.0,
"step": 2130
},
{
"entropy": 1.84752117395401,
"epoch": 0.7027649769585254,
"grad_norm": 1.39552903175354,
"learning_rate": 1.5436344045279935e-05,
"loss": 0.079,
"mean_token_accuracy": 0.9791609466075897,
"num_tokens": 4550292.0,
"step": 2135
},
{
"entropy": 1.8886283040046692,
"epoch": 0.7044107965766951,
"grad_norm": 1.880411148071289,
"learning_rate": 1.5413482568407044e-05,
"loss": 0.0807,
"mean_token_accuracy": 0.9817405939102173,
"num_tokens": 4560802.0,
"step": 2140
},
{
"entropy": 1.8611098527908325,
"epoch": 0.706056616194865,
"grad_norm": 2.7298200130462646,
"learning_rate": 1.539058099899299e-05,
"loss": 0.0653,
"mean_token_accuracy": 0.9814600467681884,
"num_tokens": 4571690.0,
"step": 2145
},
{
"entropy": 1.791181170940399,
"epoch": 0.7077024358130349,
"grad_norm": 5.456151485443115,
"learning_rate": 1.5367639506648006e-05,
"loss": 0.0708,
"mean_token_accuracy": 0.9785441160202026,
"num_tokens": 4582204.0,
"step": 2150
},
{
"entropy": 1.752633535861969,
"epoch": 0.7093482554312047,
"grad_norm": 4.673461437225342,
"learning_rate": 1.5344658261278013e-05,
"loss": 0.0629,
"mean_token_accuracy": 0.982696121931076,
"num_tokens": 4592993.0,
"step": 2155
},
{
"entropy": 1.810698115825653,
"epoch": 0.7109940750493746,
"grad_norm": 4.658589839935303,
"learning_rate": 1.532163743308335e-05,
"loss": 0.0453,
"mean_token_accuracy": 0.990430223941803,
"num_tokens": 4603443.0,
"step": 2160
},
{
"entropy": 1.7778592824935913,
"epoch": 0.7126398946675444,
"grad_norm": 3.202155828475952,
"learning_rate": 1.5298577192557487e-05,
"loss": 0.069,
"mean_token_accuracy": 0.98820481300354,
"num_tokens": 4614058.0,
"step": 2165
},
{
"entropy": 1.7779869794845582,
"epoch": 0.7142857142857143,
"grad_norm": 1.4483567476272583,
"learning_rate": 1.5275477710485812e-05,
"loss": 0.0808,
"mean_token_accuracy": 0.9856044292449951,
"num_tokens": 4625236.0,
"step": 2170
},
{
"entropy": 1.900163722038269,
"epoch": 0.7159315339038841,
"grad_norm": 1.0799473524093628,
"learning_rate": 1.525233915794432e-05,
"loss": 0.0379,
"mean_token_accuracy": 0.9869786322116851,
"num_tokens": 4635887.0,
"step": 2175
},
{
"entropy": 1.8798472046852113,
"epoch": 0.717577353522054,
"grad_norm": 4.686890602111816,
"learning_rate": 1.522916170629836e-05,
"loss": 0.0359,
"mean_token_accuracy": 0.9911372125148773,
"num_tokens": 4646429.0,
"step": 2180
},
{
"entropy": 1.9202223777770997,
"epoch": 0.7192231731402239,
"grad_norm": 3.380596399307251,
"learning_rate": 1.5205945527201386e-05,
"loss": 0.0877,
"mean_token_accuracy": 0.9812184333801269,
"num_tokens": 4656951.0,
"step": 2185
},
{
"entropy": 1.916778802871704,
"epoch": 0.7208689927583937,
"grad_norm": 2.9656357765197754,
"learning_rate": 1.5182690792593659e-05,
"loss": 0.0866,
"mean_token_accuracy": 0.9801094174385071,
"num_tokens": 4667789.0,
"step": 2190
},
{
"entropy": 1.9705326914787293,
"epoch": 0.7225148123765636,
"grad_norm": 1.6277377605438232,
"learning_rate": 1.515939767470098e-05,
"loss": 0.0684,
"mean_token_accuracy": 0.9821272671222687,
"num_tokens": 4678462.0,
"step": 2195
},
{
"entropy": 2.0496770977973937,
"epoch": 0.7241606319947334,
"grad_norm": 2.6740543842315674,
"learning_rate": 1.5136066346033431e-05,
"loss": 0.0827,
"mean_token_accuracy": 0.9822946667671204,
"num_tokens": 4689142.0,
"step": 2200
},
{
"entropy": 1.99622882604599,
"epoch": 0.7258064516129032,
"grad_norm": 2.084319829940796,
"learning_rate": 1.5112696979384076e-05,
"loss": 0.0622,
"mean_token_accuracy": 0.9843969225883484,
"num_tokens": 4699742.0,
"step": 2205
},
{
"entropy": 1.9976887702941895,
"epoch": 0.727452271231073,
"grad_norm": 4.397096633911133,
"learning_rate": 1.5089289747827698e-05,
"loss": 0.0672,
"mean_token_accuracy": 0.9822742283344269,
"num_tokens": 4710529.0,
"step": 2210
},
{
"entropy": 2.00170783996582,
"epoch": 0.7290980908492429,
"grad_norm": 1.066505789756775,
"learning_rate": 1.5065844824719498e-05,
"loss": 0.0403,
"mean_token_accuracy": 0.9920450508594513,
"num_tokens": 4721625.0,
"step": 2215
},
{
"entropy": 2.0141910314559937,
"epoch": 0.7307439104674127,
"grad_norm": 3.045076370239258,
"learning_rate": 1.504236238369383e-05,
"loss": 0.0529,
"mean_token_accuracy": 0.9883207738399505,
"num_tokens": 4732195.0,
"step": 2220
},
{
"entropy": 2.031786823272705,
"epoch": 0.7323897300855826,
"grad_norm": 5.866887092590332,
"learning_rate": 1.5018842598662913e-05,
"loss": 0.08,
"mean_token_accuracy": 0.9810294449329376,
"num_tokens": 4742750.0,
"step": 2225
},
{
"entropy": 2.0390859127044676,
"epoch": 0.7340355497037525,
"grad_norm": 3.7776222229003906,
"learning_rate": 1.499528564381553e-05,
"loss": 0.0844,
"mean_token_accuracy": 0.9762147605419159,
"num_tokens": 4753095.0,
"step": 2230
},
{
"entropy": 2.058468997478485,
"epoch": 0.7356813693219223,
"grad_norm": 2.748897075653076,
"learning_rate": 1.497169169361574e-05,
"loss": 0.0628,
"mean_token_accuracy": 0.9859977424144745,
"num_tokens": 4763609.0,
"step": 2235
},
{
"entropy": 1.9909031867980957,
"epoch": 0.7373271889400922,
"grad_norm": 1.4906212091445923,
"learning_rate": 1.4948060922801609e-05,
"loss": 0.0506,
"mean_token_accuracy": 0.9845727860927582,
"num_tokens": 4774709.0,
"step": 2240
},
{
"entropy": 1.9794199824333192,
"epoch": 0.738973008558262,
"grad_norm": 5.407120227813721,
"learning_rate": 1.492439350638388e-05,
"loss": 0.0638,
"mean_token_accuracy": 0.9835636436939239,
"num_tokens": 4785745.0,
"step": 2245
},
{
"entropy": 2.052793502807617,
"epoch": 0.7406188281764319,
"grad_norm": 0.8577380776405334,
"learning_rate": 1.4900689619644709e-05,
"loss": 0.0635,
"mean_token_accuracy": 0.9855947732925415,
"num_tokens": 4796471.0,
"step": 2250
},
{
"entropy": 2.1061079859733582,
"epoch": 0.7422646477946017,
"grad_norm": 3.701122283935547,
"learning_rate": 1.4876949438136348e-05,
"loss": 0.0884,
"mean_token_accuracy": 0.9817906200885773,
"num_tokens": 4807169.0,
"step": 2255
},
{
"entropy": 2.124563980102539,
"epoch": 0.7439104674127716,
"grad_norm": 0.8693974018096924,
"learning_rate": 1.4853173137679845e-05,
"loss": 0.0565,
"mean_token_accuracy": 0.9871229946613311,
"num_tokens": 4818068.0,
"step": 2260
},
{
"entropy": 2.1583971261978148,
"epoch": 0.7455562870309415,
"grad_norm": 3.7275748252868652,
"learning_rate": 1.4829360894363755e-05,
"loss": 0.0685,
"mean_token_accuracy": 0.981871497631073,
"num_tokens": 4829134.0,
"step": 2265
},
{
"entropy": 2.1540443658828736,
"epoch": 0.7472021066491112,
"grad_norm": 5.0215373039245605,
"learning_rate": 1.4805512884542828e-05,
"loss": 0.1157,
"mean_token_accuracy": 0.9756322801113129,
"num_tokens": 4839612.0,
"step": 2270
},
{
"entropy": 2.0997665166854858,
"epoch": 0.7488479262672811,
"grad_norm": 1.8622218370437622,
"learning_rate": 1.4781629284836689e-05,
"loss": 0.076,
"mean_token_accuracy": 0.9799229741096497,
"num_tokens": 4850560.0,
"step": 2275
},
{
"entropy": 2.082018828392029,
"epoch": 0.7504937458854509,
"grad_norm": 3.323389768600464,
"learning_rate": 1.4757710272128562e-05,
"loss": 0.0679,
"mean_token_accuracy": 0.9854858696460724,
"num_tokens": 4861281.0,
"step": 2280
},
{
"entropy": 2.145115876197815,
"epoch": 0.7521395655036208,
"grad_norm": 3.6958255767822266,
"learning_rate": 1.4733756023563932e-05,
"loss": 0.0801,
"mean_token_accuracy": 0.9791579902172088,
"num_tokens": 4872072.0,
"step": 2285
},
{
"entropy": 2.2006231546401978,
"epoch": 0.7537853851217906,
"grad_norm": 2.5469164848327637,
"learning_rate": 1.4709766716549246e-05,
"loss": 0.0564,
"mean_token_accuracy": 0.9853972256183624,
"num_tokens": 4882541.0,
"step": 2290
},
{
"entropy": 2.155518102645874,
"epoch": 0.7554312047399605,
"grad_norm": 3.5733323097229004,
"learning_rate": 1.4685742528750584e-05,
"loss": 0.0609,
"mean_token_accuracy": 0.9829520225524903,
"num_tokens": 4893417.0,
"step": 2295
},
{
"entropy": 2.142457294464111,
"epoch": 0.7570770243581304,
"grad_norm": 3.7663276195526123,
"learning_rate": 1.4661683638092375e-05,
"loss": 0.0666,
"mean_token_accuracy": 0.9813594579696655,
"num_tokens": 4904063.0,
"step": 2300
},
{
"entropy": 2.1797530174255373,
"epoch": 0.7587228439763002,
"grad_norm": 3.873552083969116,
"learning_rate": 1.4637590222756041e-05,
"loss": 0.0482,
"mean_token_accuracy": 0.9882281720638275,
"num_tokens": 4914583.0,
"step": 2305
},
{
"entropy": 2.1009862065315246,
"epoch": 0.7603686635944701,
"grad_norm": 2.9890084266662598,
"learning_rate": 1.4613462461178714e-05,
"loss": 0.0542,
"mean_token_accuracy": 0.9857582569122314,
"num_tokens": 4925125.0,
"step": 2310
},
{
"entropy": 2.1062487721443177,
"epoch": 0.7620144832126399,
"grad_norm": 2.7217583656311035,
"learning_rate": 1.4589300532051875e-05,
"loss": 0.0649,
"mean_token_accuracy": 0.987279736995697,
"num_tokens": 4935529.0,
"step": 2315
},
{
"entropy": 2.0136227369308473,
"epoch": 0.7636603028308098,
"grad_norm": 6.62637996673584,
"learning_rate": 1.4565104614320065e-05,
"loss": 0.0809,
"mean_token_accuracy": 0.9829951286315918,
"num_tokens": 4946511.0,
"step": 2320
},
{
"entropy": 2.0690441846847536,
"epoch": 0.7653061224489796,
"grad_norm": 4.281233787536621,
"learning_rate": 1.4540874887179546e-05,
"loss": 0.0676,
"mean_token_accuracy": 0.9836612105369568,
"num_tokens": 4957109.0,
"step": 2325
},
{
"entropy": 2.0863526821136475,
"epoch": 0.7669519420671495,
"grad_norm": 2.6421709060668945,
"learning_rate": 1.451661153007697e-05,
"loss": 0.0828,
"mean_token_accuracy": 0.9836499035358429,
"num_tokens": 4967988.0,
"step": 2330
},
{
"entropy": 2.0738728761672975,
"epoch": 0.7685977616853192,
"grad_norm": 1.4690742492675781,
"learning_rate": 1.4492314722708057e-05,
"loss": 0.0488,
"mean_token_accuracy": 0.9871618390083313,
"num_tokens": 4978584.0,
"step": 2335
},
{
"entropy": 2.100434124469757,
"epoch": 0.7702435813034891,
"grad_norm": 3.107667922973633,
"learning_rate": 1.4467984645016259e-05,
"loss": 0.0678,
"mean_token_accuracy": 0.9847745180130005,
"num_tokens": 4989176.0,
"step": 2340
},
{
"entropy": 2.0772449016571044,
"epoch": 0.771889400921659,
"grad_norm": 2.9565625190734863,
"learning_rate": 1.4443621477191434e-05,
"loss": 0.0564,
"mean_token_accuracy": 0.9867813110351562,
"num_tokens": 4999990.0,
"step": 2345
},
{
"entropy": 2.0808992266654966,
"epoch": 0.7735352205398288,
"grad_norm": 4.1320390701293945,
"learning_rate": 1.4419225399668504e-05,
"loss": 0.0641,
"mean_token_accuracy": 0.982687771320343,
"num_tokens": 5010626.0,
"step": 2350
},
{
"entropy": 2.0873427987098694,
"epoch": 0.7751810401579987,
"grad_norm": 7.7029595375061035,
"learning_rate": 1.4394796593126121e-05,
"loss": 0.0628,
"mean_token_accuracy": 0.9848801851272583,
"num_tokens": 5021336.0,
"step": 2355
},
{
"entropy": 2.007732594013214,
"epoch": 0.7768268597761685,
"grad_norm": 1.4012236595153809,
"learning_rate": 1.4370335238485336e-05,
"loss": 0.042,
"mean_token_accuracy": 0.9886705160140992,
"num_tokens": 5032297.0,
"step": 2360
},
{
"entropy": 2.067774224281311,
"epoch": 0.7784726793943384,
"grad_norm": 3.532789945602417,
"learning_rate": 1.4345841516908244e-05,
"loss": 0.093,
"mean_token_accuracy": 0.9788289248943329,
"num_tokens": 5042834.0,
"step": 2365
},
{
"entropy": 2.1315778255462647,
"epoch": 0.7801184990125082,
"grad_norm": 3.001561403274536,
"learning_rate": 1.4321315609796656e-05,
"loss": 0.0596,
"mean_token_accuracy": 0.985314530134201,
"num_tokens": 5053499.0,
"step": 2370
},
{
"entropy": 2.088835906982422,
"epoch": 0.7817643186306781,
"grad_norm": 1.8496758937835693,
"learning_rate": 1.4296757698790752e-05,
"loss": 0.0544,
"mean_token_accuracy": 0.9835005939006806,
"num_tokens": 5064232.0,
"step": 2375
},
{
"entropy": 2.1136799812316895,
"epoch": 0.783410138248848,
"grad_norm": 3.9325993061065674,
"learning_rate": 1.4272167965767735e-05,
"loss": 0.0599,
"mean_token_accuracy": 0.9857981920242309,
"num_tokens": 5074681.0,
"step": 2380
},
{
"entropy": 2.082188093662262,
"epoch": 0.7850559578670178,
"grad_norm": 2.6712751388549805,
"learning_rate": 1.424754659284048e-05,
"loss": 0.0425,
"mean_token_accuracy": 0.9875874161720276,
"num_tokens": 5085364.0,
"step": 2385
},
{
"entropy": 2.0782926797866823,
"epoch": 0.7867017774851877,
"grad_norm": 2.5082037448883057,
"learning_rate": 1.422289376235619e-05,
"loss": 0.0924,
"mean_token_accuracy": 0.9773596286773681,
"num_tokens": 5095986.0,
"step": 2390
},
{
"entropy": 2.1379014015197755,
"epoch": 0.7883475971033574,
"grad_norm": 3.3415675163269043,
"learning_rate": 1.4198209656895045e-05,
"loss": 0.0538,
"mean_token_accuracy": 0.9802229166030884,
"num_tokens": 5106549.0,
"step": 2395
},
{
"entropy": 2.1798630714416505,
"epoch": 0.7899934167215273,
"grad_norm": 2.5034339427948,
"learning_rate": 1.4173494459268848e-05,
"loss": 0.0529,
"mean_token_accuracy": 0.9868557870388031,
"num_tokens": 5117206.0,
"step": 2400
},
{
"entropy": 2.1523491382598876,
"epoch": 0.7916392363396971,
"grad_norm": 2.866143226623535,
"learning_rate": 1.4148748352519677e-05,
"loss": 0.0434,
"mean_token_accuracy": 0.9868989825248718,
"num_tokens": 5127670.0,
"step": 2405
},
{
"entropy": 2.1185644268989563,
"epoch": 0.793285055957867,
"grad_norm": 3.293668270111084,
"learning_rate": 1.4123971519918516e-05,
"loss": 0.0704,
"mean_token_accuracy": 0.9803002297878265,
"num_tokens": 5138300.0,
"step": 2410
},
{
"entropy": 2.1225445747375487,
"epoch": 0.7949308755760369,
"grad_norm": 3.2386057376861572,
"learning_rate": 1.4099164144963914e-05,
"loss": 0.0745,
"mean_token_accuracy": 0.976332038640976,
"num_tokens": 5148898.0,
"step": 2415
},
{
"entropy": 2.0981351375579833,
"epoch": 0.7965766951942067,
"grad_norm": 2.420224666595459,
"learning_rate": 1.4074326411380617e-05,
"loss": 0.0554,
"mean_token_accuracy": 0.9834634006023407,
"num_tokens": 5159779.0,
"step": 2420
},
{
"entropy": 2.102435576915741,
"epoch": 0.7982225148123766,
"grad_norm": 4.499454975128174,
"learning_rate": 1.4049458503118206e-05,
"loss": 0.0728,
"mean_token_accuracy": 0.9828149318695069,
"num_tokens": 5170439.0,
"step": 2425
},
{
"entropy": 2.02862309217453,
"epoch": 0.7998683344305464,
"grad_norm": 1.3120136260986328,
"learning_rate": 1.4024560604349738e-05,
"loss": 0.07,
"mean_token_accuracy": 0.9814626455307007,
"num_tokens": 5181355.0,
"step": 2430
},
{
"epoch": 0.8005266622778143,
"eval_entropy": 2.0658205878150566,
"eval_loss": 0.05905340239405632,
"eval_mean_token_accuracy": 0.9841244661327054,
"eval_num_tokens": 5185715.0,
"eval_runtime": 197.2045,
"eval_samples_per_second": 42.225,
"eval_steps_per_second": 7.038,
"step": 2432
},
{
"entropy": 1.9992692589759826,
"epoch": 0.8015141540487163,
"grad_norm": 2.716399669647217,
"learning_rate": 1.3999632899470377e-05,
"loss": 0.0415,
"mean_token_accuracy": 0.9883556425571441,
"num_tokens": 5192402.0,
"step": 2435
},
{
"entropy": 2.073114001750946,
"epoch": 0.8031599736668861,
"grad_norm": 4.84559440612793,
"learning_rate": 1.3974675573096046e-05,
"loss": 0.0516,
"mean_token_accuracy": 0.9831765830516815,
"num_tokens": 5203184.0,
"step": 2440
},
{
"entropy": 2.0994094610214233,
"epoch": 0.804805793285056,
"grad_norm": 3.8086822032928467,
"learning_rate": 1.3949688810062033e-05,
"loss": 0.0714,
"mean_token_accuracy": 0.9829389989376068,
"num_tokens": 5213537.0,
"step": 2445
},
{
"entropy": 2.1006295680999756,
"epoch": 0.8064516129032258,
"grad_norm": 4.448069095611572,
"learning_rate": 1.3924672795421638e-05,
"loss": 0.0567,
"mean_token_accuracy": 0.9861303806304932,
"num_tokens": 5224028.0,
"step": 2450
},
{
"entropy": 2.0546552419662474,
"epoch": 0.8080974325213957,
"grad_norm": 2.4029576778411865,
"learning_rate": 1.389962771444481e-05,
"loss": 0.0561,
"mean_token_accuracy": 0.986747932434082,
"num_tokens": 5234600.0,
"step": 2455
},
{
"entropy": 2.0826905846595762,
"epoch": 0.8097432521395656,
"grad_norm": 3.8037831783294678,
"learning_rate": 1.3874553752616747e-05,
"loss": 0.0647,
"mean_token_accuracy": 0.9837893486022949,
"num_tokens": 5245430.0,
"step": 2460
},
{
"entropy": 2.101279282569885,
"epoch": 0.8113890717577353,
"grad_norm": 4.736576080322266,
"learning_rate": 1.3849451095636555e-05,
"loss": 0.0393,
"mean_token_accuracy": 0.9862057268619537,
"num_tokens": 5256031.0,
"step": 2465
},
{
"entropy": 2.0845310568809508,
"epoch": 0.8130348913759052,
"grad_norm": 3.912712812423706,
"learning_rate": 1.3824319929415856e-05,
"loss": 0.098,
"mean_token_accuracy": 0.9780865371227264,
"num_tokens": 5266625.0,
"step": 2470
},
{
"entropy": 1.9965445041656493,
"epoch": 0.814680710994075,
"grad_norm": 2.1379354000091553,
"learning_rate": 1.3799160440077407e-05,
"loss": 0.0305,
"mean_token_accuracy": 0.9910949647426606,
"num_tokens": 5277179.0,
"step": 2475
},
{
"entropy": 1.9245966672897339,
"epoch": 0.8163265306122449,
"grad_norm": 3.4994077682495117,
"learning_rate": 1.3773972813953726e-05,
"loss": 0.0606,
"mean_token_accuracy": 0.9876502931118012,
"num_tokens": 5287715.0,
"step": 2480
},
{
"entropy": 1.872160291671753,
"epoch": 0.8179723502304147,
"grad_norm": 4.733576774597168,
"learning_rate": 1.3748757237585729e-05,
"loss": 0.0691,
"mean_token_accuracy": 0.9822778880596161,
"num_tokens": 5298582.0,
"step": 2485
},
{
"entropy": 1.8784523367881776,
"epoch": 0.8196181698485846,
"grad_norm": 3.7372028827667236,
"learning_rate": 1.372351389772131e-05,
"loss": 0.0713,
"mean_token_accuracy": 0.9856967926025391,
"num_tokens": 5309140.0,
"step": 2490
},
{
"entropy": 1.9651299357414245,
"epoch": 0.8212639894667545,
"grad_norm": 3.022817611694336,
"learning_rate": 1.3698242981314e-05,
"loss": 0.0547,
"mean_token_accuracy": 0.9867731034755707,
"num_tokens": 5319522.0,
"step": 2495
},
{
"entropy": 1.943724513053894,
"epoch": 0.8229098090849243,
"grad_norm": 3.299440622329712,
"learning_rate": 1.3672944675521555e-05,
"loss": 0.0434,
"mean_token_accuracy": 0.985489410161972,
"num_tokens": 5330359.0,
"step": 2500
},
{
"entropy": 1.9148884296417237,
"epoch": 0.8245556287030942,
"grad_norm": 2.022271156311035,
"learning_rate": 1.3647619167704578e-05,
"loss": 0.0457,
"mean_token_accuracy": 0.9900146067142487,
"num_tokens": 5341225.0,
"step": 2505
},
{
"entropy": 1.8972573161125184,
"epoch": 0.826201448321264,
"grad_norm": 1.6173745393753052,
"learning_rate": 1.3622266645425135e-05,
"loss": 0.0371,
"mean_token_accuracy": 0.9884720623493195,
"num_tokens": 5351753.0,
"step": 2510
},
{
"entropy": 1.845130443572998,
"epoch": 0.8278472679394339,
"grad_norm": 3.775874376296997,
"learning_rate": 1.359688729644536e-05,
"loss": 0.0575,
"mean_token_accuracy": 0.985155212879181,
"num_tokens": 5362524.0,
"step": 2515
},
{
"entropy": 1.8224045038223267,
"epoch": 0.8294930875576036,
"grad_norm": 2.454981803894043,
"learning_rate": 1.3571481308726064e-05,
"loss": 0.0544,
"mean_token_accuracy": 0.9880367636680603,
"num_tokens": 5373427.0,
"step": 2520
},
{
"entropy": 1.854708468914032,
"epoch": 0.8311389071757735,
"grad_norm": 4.371317386627197,
"learning_rate": 1.3546048870425356e-05,
"loss": 0.0705,
"mean_token_accuracy": 0.9838429808616638,
"num_tokens": 5383996.0,
"step": 2525
},
{
"entropy": 1.8091017842292785,
"epoch": 0.8327847267939433,
"grad_norm": 2.3962645530700684,
"learning_rate": 1.3520590169897232e-05,
"loss": 0.0525,
"mean_token_accuracy": 0.9856068968772889,
"num_tokens": 5394554.0,
"step": 2530
},
{
"entropy": 1.827188992500305,
"epoch": 0.8344305464121132,
"grad_norm": 3.678561210632324,
"learning_rate": 1.3495105395690185e-05,
"loss": 0.0538,
"mean_token_accuracy": 0.9860463380813599,
"num_tokens": 5405311.0,
"step": 2535
},
{
"entropy": 1.8938659191131593,
"epoch": 0.8360763660302831,
"grad_norm": 4.154452323913574,
"learning_rate": 1.3469594736545816e-05,
"loss": 0.0566,
"mean_token_accuracy": 0.982678347826004,
"num_tokens": 5415855.0,
"step": 2540
},
{
"entropy": 1.8611347198486328,
"epoch": 0.8377221856484529,
"grad_norm": 5.7794365882873535,
"learning_rate": 1.344405838139743e-05,
"loss": 0.06,
"mean_token_accuracy": 0.9849008440971374,
"num_tokens": 5426534.0,
"step": 2545
},
{
"entropy": 1.8572628855705262,
"epoch": 0.8393680052666228,
"grad_norm": 3.2509214878082275,
"learning_rate": 1.341849651936864e-05,
"loss": 0.0449,
"mean_token_accuracy": 0.9872718393802643,
"num_tokens": 5437422.0,
"step": 2550
},
{
"entropy": 1.9231997847557067,
"epoch": 0.8410138248847926,
"grad_norm": 1.5836360454559326,
"learning_rate": 1.3392909339771957e-05,
"loss": 0.0687,
"mean_token_accuracy": 0.9836991250514984,
"num_tokens": 5447835.0,
"step": 2555
},
{
"entropy": 2.006228494644165,
"epoch": 0.8426596445029625,
"grad_norm": 2.372229814529419,
"learning_rate": 1.3367297032107404e-05,
"loss": 0.0748,
"mean_token_accuracy": 0.9789046466350555,
"num_tokens": 5458430.0,
"step": 2560
},
{
"entropy": 2.013643407821655,
"epoch": 0.8443054641211323,
"grad_norm": 2.6997487545013428,
"learning_rate": 1.33416597860611e-05,
"loss": 0.0693,
"mean_token_accuracy": 0.9819782853126526,
"num_tokens": 5469093.0,
"step": 2565
},
{
"entropy": 2.086654710769653,
"epoch": 0.8459512837393022,
"grad_norm": 5.783359050750732,
"learning_rate": 1.331599779150386e-05,
"loss": 0.0602,
"mean_token_accuracy": 0.9840903103351593,
"num_tokens": 5479883.0,
"step": 2570
},
{
"entropy": 1.9687761425971986,
"epoch": 0.8475971033574721,
"grad_norm": 1.4621000289916992,
"learning_rate": 1.3290311238489784e-05,
"loss": 0.037,
"mean_token_accuracy": 0.9909431576728821,
"num_tokens": 5490442.0,
"step": 2575
},
{
"entropy": 1.9323675155639648,
"epoch": 0.8492429229756419,
"grad_norm": 3.7601616382598877,
"learning_rate": 1.3264600317254854e-05,
"loss": 0.0535,
"mean_token_accuracy": 0.9863576173782349,
"num_tokens": 5501192.0,
"step": 2580
},
{
"entropy": 1.993829894065857,
"epoch": 0.8508887425938118,
"grad_norm": 2.846726655960083,
"learning_rate": 1.3238865218215535e-05,
"loss": 0.0735,
"mean_token_accuracy": 0.9818755805492401,
"num_tokens": 5511779.0,
"step": 2585
},
{
"entropy": 1.9815747618675232,
"epoch": 0.8525345622119815,
"grad_norm": 2.043933868408203,
"learning_rate": 1.3213106131967339e-05,
"loss": 0.0394,
"mean_token_accuracy": 0.9880852580070496,
"num_tokens": 5522208.0,
"step": 2590
},
{
"entropy": 1.9507970690727234,
"epoch": 0.8541803818301514,
"grad_norm": 3.3365821838378906,
"learning_rate": 1.3187323249283439e-05,
"loss": 0.0449,
"mean_token_accuracy": 0.987623393535614,
"num_tokens": 5532774.0,
"step": 2595
},
{
"entropy": 1.8794512629508973,
"epoch": 0.8558262014483212,
"grad_norm": 4.568225860595703,
"learning_rate": 1.316151676111324e-05,
"loss": 0.047,
"mean_token_accuracy": 0.9890295684337616,
"num_tokens": 5543614.0,
"step": 2600
},
{
"entropy": 1.8917754292488098,
"epoch": 0.8574720210664911,
"grad_norm": 1.8697177171707153,
"learning_rate": 1.313568685858097e-05,
"loss": 0.0604,
"mean_token_accuracy": 0.9812738180160523,
"num_tokens": 5554191.0,
"step": 2605
},
{
"entropy": 1.8788981676101684,
"epoch": 0.859117840684661,
"grad_norm": 2.4049696922302246,
"learning_rate": 1.3109833732984272e-05,
"loss": 0.0368,
"mean_token_accuracy": 0.9879702389240265,
"num_tokens": 5564836.0,
"step": 2610
},
{
"entropy": 1.8883351802825927,
"epoch": 0.8607636603028308,
"grad_norm": 3.053053379058838,
"learning_rate": 1.3083957575792772e-05,
"loss": 0.0608,
"mean_token_accuracy": 0.9815441846847535,
"num_tokens": 5575491.0,
"step": 2615
},
{
"entropy": 1.8524894714355469,
"epoch": 0.8624094799210007,
"grad_norm": 2.5347635746002197,
"learning_rate": 1.3058058578646673e-05,
"loss": 0.0676,
"mean_token_accuracy": 0.9812525391578675,
"num_tokens": 5586127.0,
"step": 2620
},
{
"entropy": 1.9090933799743652,
"epoch": 0.8640552995391705,
"grad_norm": 4.13297176361084,
"learning_rate": 1.3032136933355336e-05,
"loss": 0.0675,
"mean_token_accuracy": 0.9843396723270417,
"num_tokens": 5596745.0,
"step": 2625
},
{
"entropy": 1.8861599802970885,
"epoch": 0.8657011191573404,
"grad_norm": 3.5086593627929688,
"learning_rate": 1.3006192831895846e-05,
"loss": 0.0521,
"mean_token_accuracy": 0.9839343369007111,
"num_tokens": 5607882.0,
"step": 2630
},
{
"entropy": 1.9625224947929383,
"epoch": 0.8673469387755102,
"grad_norm": 2.0559775829315186,
"learning_rate": 1.2980226466411605e-05,
"loss": 0.06,
"mean_token_accuracy": 0.9822891354560852,
"num_tokens": 5618184.0,
"step": 2635
},
{
"entropy": 1.8167145729064942,
"epoch": 0.8689927583936801,
"grad_norm": 11.442062377929688,
"learning_rate": 1.2954238029210906e-05,
"loss": 0.0468,
"mean_token_accuracy": 0.9872013688087463,
"num_tokens": 5628846.0,
"step": 2640
},
{
"entropy": 1.7704446077346803,
"epoch": 0.8706385780118499,
"grad_norm": 3.554111957550049,
"learning_rate": 1.2928227712765504e-05,
"loss": 0.0784,
"mean_token_accuracy": 0.9803166925907135,
"num_tokens": 5639683.0,
"step": 2645
},
{
"entropy": 1.746169638633728,
"epoch": 0.8722843976300197,
"grad_norm": 4.623286247253418,
"learning_rate": 1.290219570970919e-05,
"loss": 0.0622,
"mean_token_accuracy": 0.9823844611644745,
"num_tokens": 5650210.0,
"step": 2650
},
{
"entropy": 1.671170949935913,
"epoch": 0.8739302172481896,
"grad_norm": 3.731656551361084,
"learning_rate": 1.2876142212836373e-05,
"loss": 0.0572,
"mean_token_accuracy": 0.9831017911434173,
"num_tokens": 5660768.0,
"step": 2655
},
{
"entropy": 1.7002404808998108,
"epoch": 0.8755760368663594,
"grad_norm": 3.3224618434906006,
"learning_rate": 1.2850067415100643e-05,
"loss": 0.0796,
"mean_token_accuracy": 0.9770964682102203,
"num_tokens": 5671563.0,
"step": 2660
},
{
"entropy": 1.7098248720169067,
"epoch": 0.8772218564845293,
"grad_norm": 4.906748294830322,
"learning_rate": 1.2823971509613338e-05,
"loss": 0.0776,
"mean_token_accuracy": 0.9807847082614899,
"num_tokens": 5682040.0,
"step": 2665
},
{
"entropy": 1.8245792269706727,
"epoch": 0.8788676761026991,
"grad_norm": 3.9868807792663574,
"learning_rate": 1.2797854689642136e-05,
"loss": 0.074,
"mean_token_accuracy": 0.9827134788036347,
"num_tokens": 5692607.0,
"step": 2670
},
{
"entropy": 1.8712535619735717,
"epoch": 0.880513495720869,
"grad_norm": 2.729797601699829,
"learning_rate": 1.2771717148609598e-05,
"loss": 0.0657,
"mean_token_accuracy": 0.9844698071479797,
"num_tokens": 5703254.0,
"step": 2675
},
{
"entropy": 1.8171938061714172,
"epoch": 0.8821593153390388,
"grad_norm": 2.3974361419677734,
"learning_rate": 1.2745559080091749e-05,
"loss": 0.0492,
"mean_token_accuracy": 0.98358274102211,
"num_tokens": 5713800.0,
"step": 2680
},
{
"entropy": 1.7879636526107787,
"epoch": 0.8838051349572087,
"grad_norm": 0.965469479560852,
"learning_rate": 1.2719380677816648e-05,
"loss": 0.0679,
"mean_token_accuracy": 0.9823775053024292,
"num_tokens": 5724150.0,
"step": 2685
},
{
"entropy": 1.752970778942108,
"epoch": 0.8854509545753786,
"grad_norm": 3.1124043464660645,
"learning_rate": 1.2693182135662933e-05,
"loss": 0.0431,
"mean_token_accuracy": 0.9858216881752014,
"num_tokens": 5734916.0,
"step": 2690
},
{
"entropy": 1.8105898737907409,
"epoch": 0.8870967741935484,
"grad_norm": 1.5042496919631958,
"learning_rate": 1.2666963647658413e-05,
"loss": 0.0514,
"mean_token_accuracy": 0.9884253561496734,
"num_tokens": 5745344.0,
"step": 2695
},
{
"entropy": 1.7806416988372802,
"epoch": 0.8887425938117183,
"grad_norm": 2.657801389694214,
"learning_rate": 1.2640725407978607e-05,
"loss": 0.0344,
"mean_token_accuracy": 0.9910404562950135,
"num_tokens": 5755871.0,
"step": 2700
},
{
"entropy": 1.7876773238182069,
"epoch": 0.8903884134298881,
"grad_norm": 2.7087864875793457,
"learning_rate": 1.2614467610945323e-05,
"loss": 0.0704,
"mean_token_accuracy": 0.9819835543632507,
"num_tokens": 5766351.0,
"step": 2705
},
{
"entropy": 1.7520806312561035,
"epoch": 0.892034233048058,
"grad_norm": 2.7425425052642822,
"learning_rate": 1.2588190451025209e-05,
"loss": 0.0391,
"mean_token_accuracy": 0.9887621998786926,
"num_tokens": 5776778.0,
"step": 2710
},
{
"entropy": 1.7028562068939208,
"epoch": 0.8936800526662277,
"grad_norm": 1.6712589263916016,
"learning_rate": 1.2561894122828315e-05,
"loss": 0.0588,
"mean_token_accuracy": 0.9863285422325134,
"num_tokens": 5787425.0,
"step": 2715
},
{
"entropy": 1.6849963307380675,
"epoch": 0.8953258722843976,
"grad_norm": 4.329936504364014,
"learning_rate": 1.2535578821106648e-05,
"loss": 0.0553,
"mean_token_accuracy": 0.984381890296936,
"num_tokens": 5798063.0,
"step": 2720
},
{
"entropy": 1.7037844061851501,
"epoch": 0.8969716919025674,
"grad_norm": 2.813596248626709,
"learning_rate": 1.2509244740752748e-05,
"loss": 0.0612,
"mean_token_accuracy": 0.9847327530384063,
"num_tokens": 5808513.0,
"step": 2725
},
{
"entropy": 1.763570249080658,
"epoch": 0.8986175115207373,
"grad_norm": 5.475368976593018,
"learning_rate": 1.2482892076798216e-05,
"loss": 0.0659,
"mean_token_accuracy": 0.9854476451873779,
"num_tokens": 5819032.0,
"step": 2730
},
{
"entropy": 1.8239656686782837,
"epoch": 0.9002633311389072,
"grad_norm": 2.5298304557800293,
"learning_rate": 1.2456521024412287e-05,
"loss": 0.0817,
"mean_token_accuracy": 0.983061420917511,
"num_tokens": 5829661.0,
"step": 2735
},
{
"entropy": 1.9393372416496277,
"epoch": 0.901909150757077,
"grad_norm": 2.5725841522216797,
"learning_rate": 1.243013177890039e-05,
"loss": 0.0466,
"mean_token_accuracy": 0.987521517276764,
"num_tokens": 5840288.0,
"step": 2740
},
{
"entropy": 2.0387110352516173,
"epoch": 0.9035549703752469,
"grad_norm": 2.8396382331848145,
"learning_rate": 1.2403724535702679e-05,
"loss": 0.1089,
"mean_token_accuracy": 0.9775011360645294,
"num_tokens": 5850912.0,
"step": 2745
},
{
"entropy": 2.0455712914466857,
"epoch": 0.9052007899934167,
"grad_norm": 3.0769193172454834,
"learning_rate": 1.2377299490392618e-05,
"loss": 0.0575,
"mean_token_accuracy": 0.9838365972042084,
"num_tokens": 5861530.0,
"step": 2750
},
{
"entropy": 2.0284469962120055,
"epoch": 0.9068466096115866,
"grad_norm": 3.171985626220703,
"learning_rate": 1.23508568386755e-05,
"loss": 0.0433,
"mean_token_accuracy": 0.986922162771225,
"num_tokens": 5872294.0,
"step": 2755
},
{
"entropy": 2.027714431285858,
"epoch": 0.9084924292297564,
"grad_norm": 1.4176881313323975,
"learning_rate": 1.2324396776387014e-05,
"loss": 0.0582,
"mean_token_accuracy": 0.9820313692092896,
"num_tokens": 5882926.0,
"step": 2760
},
{
"entropy": 1.9616046786308288,
"epoch": 0.9101382488479263,
"grad_norm": 1.9177738428115845,
"learning_rate": 1.2297919499491797e-05,
"loss": 0.0621,
"mean_token_accuracy": 0.9839075744152069,
"num_tokens": 5893515.0,
"step": 2765
},
{
"entropy": 1.8943446040153504,
"epoch": 0.9117840684660962,
"grad_norm": 3.43756103515625,
"learning_rate": 1.2271425204081981e-05,
"loss": 0.0688,
"mean_token_accuracy": 0.9814131796360016,
"num_tokens": 5904107.0,
"step": 2770
},
{
"entropy": 1.8612539172172546,
"epoch": 0.913429888084266,
"grad_norm": 2.007723808288574,
"learning_rate": 1.2244914086375726e-05,
"loss": 0.0566,
"mean_token_accuracy": 0.9836967706680297,
"num_tokens": 5914642.0,
"step": 2775
},
{
"entropy": 1.8655507802963256,
"epoch": 0.9150757077024358,
"grad_norm": 2.231388568878174,
"learning_rate": 1.2218386342715793e-05,
"loss": 0.0776,
"mean_token_accuracy": 0.9767457246780396,
"num_tokens": 5925381.0,
"step": 2780
},
{
"entropy": 1.9478851079940795,
"epoch": 0.9167215273206056,
"grad_norm": 3.511413335800171,
"learning_rate": 1.2191842169568067e-05,
"loss": 0.0627,
"mean_token_accuracy": 0.9826519787311554,
"num_tokens": 5936411.0,
"step": 2785
},
{
"entropy": 1.9535033702850342,
"epoch": 0.9183673469387755,
"grad_norm": 1.3818976879119873,
"learning_rate": 1.2165281763520106e-05,
"loss": 0.0284,
"mean_token_accuracy": 0.9931237936019898,
"num_tokens": 5947401.0,
"step": 2790
},
{
"entropy": 2.046694076061249,
"epoch": 0.9200131665569453,
"grad_norm": 2.708840847015381,
"learning_rate": 1.2138705321279709e-05,
"loss": 0.0357,
"mean_token_accuracy": 0.9874314427375793,
"num_tokens": 5957977.0,
"step": 2795
},
{
"entropy": 1.9919468641281128,
"epoch": 0.9216589861751152,
"grad_norm": 3.4317140579223633,
"learning_rate": 1.2112113039673418e-05,
"loss": 0.0491,
"mean_token_accuracy": 0.9871409773826599,
"num_tokens": 5968581.0,
"step": 2800
},
{
"entropy": 2.0010817289352416,
"epoch": 0.9233048057932851,
"grad_norm": 3.1390058994293213,
"learning_rate": 1.2085505115645095e-05,
"loss": 0.0463,
"mean_token_accuracy": 0.9875739812850952,
"num_tokens": 5979403.0,
"step": 2805
},
{
"entropy": 1.9707128882408143,
"epoch": 0.9249506254114549,
"grad_norm": 2.6084043979644775,
"learning_rate": 1.2058881746254447e-05,
"loss": 0.0593,
"mean_token_accuracy": 0.9853511095046997,
"num_tokens": 5990059.0,
"step": 2810
},
{
"entropy": 2.035856246948242,
"epoch": 0.9265964450296248,
"grad_norm": 1.959696888923645,
"learning_rate": 1.203224312867557e-05,
"loss": 0.0434,
"mean_token_accuracy": 0.9854652583599091,
"num_tokens": 6000522.0,
"step": 2815
},
{
"entropy": 2.0164458394050597,
"epoch": 0.9282422646477946,
"grad_norm": 5.868241786956787,
"learning_rate": 1.2005589460195486e-05,
"loss": 0.0773,
"mean_token_accuracy": 0.9825805127620697,
"num_tokens": 6011268.0,
"step": 2820
},
{
"entropy": 2.0600091218948364,
"epoch": 0.9298880842659645,
"grad_norm": 4.795652389526367,
"learning_rate": 1.1978920938212691e-05,
"loss": 0.0591,
"mean_token_accuracy": 0.984807425737381,
"num_tokens": 6022053.0,
"step": 2825
},
{
"entropy": 2.1145251750946046,
"epoch": 0.9315339038841343,
"grad_norm": 1.8055132627487183,
"learning_rate": 1.1952237760235686e-05,
"loss": 0.0611,
"mean_token_accuracy": 0.982151460647583,
"num_tokens": 6032622.0,
"step": 2830
},
{
"entropy": 2.1461195945739746,
"epoch": 0.9331797235023042,
"grad_norm": 3.280885934829712,
"learning_rate": 1.192554012388151e-05,
"loss": 0.0711,
"mean_token_accuracy": 0.9854592263698578,
"num_tokens": 6043224.0,
"step": 2835
},
{
"entropy": 2.1928713798522947,
"epoch": 0.934825543120474,
"grad_norm": 2.116999864578247,
"learning_rate": 1.1898828226874284e-05,
"loss": 0.0536,
"mean_token_accuracy": 0.9854759395122528,
"num_tokens": 6053910.0,
"step": 2840
},
{
"entropy": 2.153935670852661,
"epoch": 0.9364713627386438,
"grad_norm": 3.3620643615722656,
"learning_rate": 1.1872102267043748e-05,
"loss": 0.0525,
"mean_token_accuracy": 0.9844566106796264,
"num_tokens": 6064731.0,
"step": 2845
},
{
"entropy": 2.120525050163269,
"epoch": 0.9381171823568137,
"grad_norm": 2.1633312702178955,
"learning_rate": 1.1845362442323784e-05,
"loss": 0.0356,
"mean_token_accuracy": 0.9903881013393402,
"num_tokens": 6075595.0,
"step": 2850
},
{
"entropy": 2.168933653831482,
"epoch": 0.9397630019749835,
"grad_norm": 2.078237771987915,
"learning_rate": 1.1818608950750967e-05,
"loss": 0.045,
"mean_token_accuracy": 0.9854483067989349,
"num_tokens": 6085983.0,
"step": 2855
},
{
"entropy": 2.104006254673004,
"epoch": 0.9414088215931534,
"grad_norm": 1.4718915224075317,
"learning_rate": 1.1791841990463083e-05,
"loss": 0.0375,
"mean_token_accuracy": 0.9886513173580169,
"num_tokens": 6096996.0,
"step": 2860
},
{
"entropy": 2.048595929145813,
"epoch": 0.9430546412113232,
"grad_norm": 1.9670352935791016,
"learning_rate": 1.1765061759697669e-05,
"loss": 0.0316,
"mean_token_accuracy": 0.9901619255542755,
"num_tokens": 6107708.0,
"step": 2865
},
{
"entropy": 1.970208466053009,
"epoch": 0.9447004608294931,
"grad_norm": 5.284751892089844,
"learning_rate": 1.1738268456790548e-05,
"loss": 0.0672,
"mean_token_accuracy": 0.981162142753601,
"num_tokens": 6118639.0,
"step": 2870
},
{
"entropy": 1.8950253486633302,
"epoch": 0.9463462804476629,
"grad_norm": 4.18200159072876,
"learning_rate": 1.171146228017435e-05,
"loss": 0.0414,
"mean_token_accuracy": 0.9900772452354432,
"num_tokens": 6129781.0,
"step": 2875
},
{
"entropy": 1.8563172101974488,
"epoch": 0.9479921000658328,
"grad_norm": 2.1970016956329346,
"learning_rate": 1.1684643428377056e-05,
"loss": 0.058,
"mean_token_accuracy": 0.9863704919815064,
"num_tokens": 6140682.0,
"step": 2880
},
{
"entropy": 1.8867647051811218,
"epoch": 0.9496379196840027,
"grad_norm": 1.9754729270935059,
"learning_rate": 1.1657812100020507e-05,
"loss": 0.0588,
"mean_token_accuracy": 0.9856977164745331,
"num_tokens": 6151093.0,
"step": 2885
},
{
"entropy": 1.875185477733612,
"epoch": 0.9512837393021725,
"grad_norm": 4.4464592933654785,
"learning_rate": 1.1630968493818961e-05,
"loss": 0.062,
"mean_token_accuracy": 0.9841465830802918,
"num_tokens": 6161567.0,
"step": 2890
},
{
"entropy": 1.8699720621109008,
"epoch": 0.9529295589203424,
"grad_norm": 2.7016501426696777,
"learning_rate": 1.1604112808577603e-05,
"loss": 0.0727,
"mean_token_accuracy": 0.9823096752166748,
"num_tokens": 6172041.0,
"step": 2895
},
{
"entropy": 1.9755351066589355,
"epoch": 0.9545753785385122,
"grad_norm": 14.661170959472656,
"learning_rate": 1.1577245243191068e-05,
"loss": 0.0641,
"mean_token_accuracy": 0.979709017276764,
"num_tokens": 6182650.0,
"step": 2900
},
{
"entropy": 2.0633350372314454,
"epoch": 0.956221198156682,
"grad_norm": 3.1461470127105713,
"learning_rate": 1.155036599664198e-05,
"loss": 0.0693,
"mean_token_accuracy": 0.9809099555015564,
"num_tokens": 6193204.0,
"step": 2905
},
{
"entropy": 2.077620244026184,
"epoch": 0.9578670177748518,
"grad_norm": 1.7489817142486572,
"learning_rate": 1.1523475267999477e-05,
"loss": 0.0777,
"mean_token_accuracy": 0.9819590091705322,
"num_tokens": 6203825.0,
"step": 2910
},
{
"entropy": 2.087475097179413,
"epoch": 0.9595128373930217,
"grad_norm": 3.104898691177368,
"learning_rate": 1.1496573256417733e-05,
"loss": 0.0437,
"mean_token_accuracy": 0.9875694811344147,
"num_tokens": 6214779.0,
"step": 2915
},
{
"entropy": 2.038085675239563,
"epoch": 0.9611586570111915,
"grad_norm": 3.818824052810669,
"learning_rate": 1.1469660161134481e-05,
"loss": 0.0563,
"mean_token_accuracy": 0.9853502690792084,
"num_tokens": 6225540.0,
"step": 2920
},
{
"entropy": 1.9863147258758544,
"epoch": 0.9628044766293614,
"grad_norm": 3.5189995765686035,
"learning_rate": 1.1442736181469546e-05,
"loss": 0.0534,
"mean_token_accuracy": 0.9900292754173279,
"num_tokens": 6236269.0,
"step": 2925
},
{
"entropy": 1.8946941256523133,
"epoch": 0.9644502962475313,
"grad_norm": 1.297648549079895,
"learning_rate": 1.1415801516823358e-05,
"loss": 0.0523,
"mean_token_accuracy": 0.9893131017684936,
"num_tokens": 6247052.0,
"step": 2930
},
{
"entropy": 1.9294702768325807,
"epoch": 0.9660961158657011,
"grad_norm": 2.645850419998169,
"learning_rate": 1.1388856366675482e-05,
"loss": 0.0705,
"mean_token_accuracy": 0.9818682849407196,
"num_tokens": 6257581.0,
"step": 2935
},
{
"entropy": 1.9064095735549926,
"epoch": 0.967741935483871,
"grad_norm": 2.203904151916504,
"learning_rate": 1.1361900930583143e-05,
"loss": 0.0589,
"mean_token_accuracy": 0.9860333859920501,
"num_tokens": 6268381.0,
"step": 2940
},
{
"entropy": 1.8867696881294251,
"epoch": 0.9693877551020408,
"grad_norm": 4.154993534088135,
"learning_rate": 1.1334935408179736e-05,
"loss": 0.0495,
"mean_token_accuracy": 0.9842896819114685,
"num_tokens": 6279423.0,
"step": 2945
},
{
"entropy": 1.8813761711120605,
"epoch": 0.9710335747202107,
"grad_norm": 1.7893234491348267,
"learning_rate": 1.1307959999173362e-05,
"loss": 0.0524,
"mean_token_accuracy": 0.9866505920886993,
"num_tokens": 6290068.0,
"step": 2950
},
{
"entropy": 1.915200400352478,
"epoch": 0.9726793943383805,
"grad_norm": 2.213712692260742,
"learning_rate": 1.1280974903345347e-05,
"loss": 0.064,
"mean_token_accuracy": 0.9851447463035583,
"num_tokens": 6300817.0,
"step": 2955
},
{
"entropy": 1.9231054186820984,
"epoch": 0.9743252139565504,
"grad_norm": 1.9981787204742432,
"learning_rate": 1.1253980320548746e-05,
"loss": 0.0434,
"mean_token_accuracy": 0.9863059282302856,
"num_tokens": 6311420.0,
"step": 2960
},
{
"entropy": 1.9738001823425293,
"epoch": 0.9759710335747203,
"grad_norm": 0.5440634489059448,
"learning_rate": 1.1226976450706887e-05,
"loss": 0.041,
"mean_token_accuracy": 0.9861558496952056,
"num_tokens": 6321771.0,
"step": 2965
},
{
"entropy": 1.9686074256896973,
"epoch": 0.97761685319289,
"grad_norm": 5.0454936027526855,
"learning_rate": 1.119996349381187e-05,
"loss": 0.0772,
"mean_token_accuracy": 0.9804552078247071,
"num_tokens": 6332139.0,
"step": 2970
},
{
"entropy": 1.9591160893440247,
"epoch": 0.9792626728110599,
"grad_norm": 1.6458666324615479,
"learning_rate": 1.1172941649923096e-05,
"loss": 0.0463,
"mean_token_accuracy": 0.9881701529026031,
"num_tokens": 6342546.0,
"step": 2975
},
{
"entropy": 2.0319183111190795,
"epoch": 0.9809084924292297,
"grad_norm": 2.887859582901001,
"learning_rate": 1.1145911119165789e-05,
"loss": 0.0553,
"mean_token_accuracy": 0.9828070223331451,
"num_tokens": 6353442.0,
"step": 2980
},
{
"entropy": 2.058086669445038,
"epoch": 0.9825543120473996,
"grad_norm": 0.9121004343032837,
"learning_rate": 1.11188721017295e-05,
"loss": 0.0366,
"mean_token_accuracy": 0.9888731300830841,
"num_tokens": 6363972.0,
"step": 2985
},
{
"entropy": 1.9924060463905335,
"epoch": 0.9842001316655694,
"grad_norm": 1.7889047861099243,
"learning_rate": 1.1091824797866639e-05,
"loss": 0.0469,
"mean_token_accuracy": 0.9876609027385712,
"num_tokens": 6374648.0,
"step": 2990
},
{
"entropy": 1.9006078600883485,
"epoch": 0.9858459512837393,
"grad_norm": 3.3813469409942627,
"learning_rate": 1.1064769407890986e-05,
"loss": 0.0228,
"mean_token_accuracy": 0.9935931503772736,
"num_tokens": 6385573.0,
"step": 2995
},
{
"entropy": 1.8375752925872804,
"epoch": 0.9874917709019092,
"grad_norm": 4.215015888214111,
"learning_rate": 1.1037706132176197e-05,
"loss": 0.0313,
"mean_token_accuracy": 0.9893752455711364,
"num_tokens": 6396486.0,
"step": 3000
},
{
"entropy": 1.7838311195373535,
"epoch": 0.989137590520079,
"grad_norm": 1.322409749031067,
"learning_rate": 1.1010635171154342e-05,
"loss": 0.071,
"mean_token_accuracy": 0.9886986196041108,
"num_tokens": 6407411.0,
"step": 3005
},
{
"entropy": 1.7372970938682557,
"epoch": 0.9907834101382489,
"grad_norm": 2.0234224796295166,
"learning_rate": 1.098355672531441e-05,
"loss": 0.0549,
"mean_token_accuracy": 0.9890697836875916,
"num_tokens": 6418480.0,
"step": 3010
},
{
"entropy": 1.8203161001205443,
"epoch": 0.9924292297564187,
"grad_norm": 4.083996295928955,
"learning_rate": 1.0956470995200816e-05,
"loss": 0.0559,
"mean_token_accuracy": 0.9861119568347931,
"num_tokens": 6429376.0,
"step": 3015
},
{
"entropy": 1.8980080366134644,
"epoch": 0.9940750493745886,
"grad_norm": 2.3635177612304688,
"learning_rate": 1.0929378181411918e-05,
"loss": 0.0407,
"mean_token_accuracy": 0.9869163811206818,
"num_tokens": 6440084.0,
"step": 3020
},
{
"entropy": 1.9592399954795838,
"epoch": 0.9957208689927584,
"grad_norm": 1.469984769821167,
"learning_rate": 1.0902278484598549e-05,
"loss": 0.046,
"mean_token_accuracy": 0.9860125243663788,
"num_tokens": 6450632.0,
"step": 3025
},
{
"entropy": 1.943799901008606,
"epoch": 0.9973666886109283,
"grad_norm": 2.726168394088745,
"learning_rate": 1.0875172105462513e-05,
"loss": 0.0617,
"mean_token_accuracy": 0.9823821485042572,
"num_tokens": 6461324.0,
"step": 3030
},
{
"entropy": 1.979371726512909,
"epoch": 0.999012508229098,
"grad_norm": 3.588101625442505,
"learning_rate": 1.0848059244755093e-05,
"loss": 0.04,
"mean_token_accuracy": 0.9917591452598572,
"num_tokens": 6472058.0,
"step": 3035
},
{
"entropy": 2.0093634843826296,
"epoch": 1.000658327847268,
"grad_norm": 1.96505868434906,
"learning_rate": 1.0820940103275594e-05,
"loss": 0.0413,
"mean_token_accuracy": 0.9876884579658508,
"num_tokens": 6482261.0,
"step": 3040
},
{
"epoch": 1.000658327847268,
"eval_entropy": 2.00399693047966,
"eval_loss": 0.05508637800812721,
"eval_mean_token_accuracy": 0.9851879490383764,
"eval_num_tokens": 6482261.0,
"eval_runtime": 196.5531,
"eval_samples_per_second": 42.365,
"eval_steps_per_second": 7.062,
"step": 3040
},
{
"entropy": 2.0180766105651857,
"epoch": 1.0023041474654377,
"grad_norm": 1.1969290971755981,
"learning_rate": 1.079381488186982e-05,
"loss": 0.0215,
"mean_token_accuracy": 0.9949158251285553,
"num_tokens": 6492816.0,
"step": 3045
},
{
"entropy": 1.9490231990814209,
"epoch": 1.0039499670836076,
"grad_norm": 0.8785367012023926,
"learning_rate": 1.0766683781428617e-05,
"loss": 0.0451,
"mean_token_accuracy": 0.9905852675437927,
"num_tokens": 6503215.0,
"step": 3050
},
{
"entropy": 1.9013155460357667,
"epoch": 1.0055957867017775,
"grad_norm": 2.4878368377685547,
"learning_rate": 1.0739547002886361e-05,
"loss": 0.0177,
"mean_token_accuracy": 0.9950673639774322,
"num_tokens": 6514095.0,
"step": 3055
},
{
"entropy": 1.8226952791213988,
"epoch": 1.0072416063199474,
"grad_norm": 1.9651074409484863,
"learning_rate": 1.0712404747219481e-05,
"loss": 0.0253,
"mean_token_accuracy": 0.9917102456092834,
"num_tokens": 6525039.0,
"step": 3060
},
{
"entropy": 1.7761475205421449,
"epoch": 1.008887425938117,
"grad_norm": 0.24293629825115204,
"learning_rate": 1.0685257215444975e-05,
"loss": 0.0185,
"mean_token_accuracy": 0.9937527477741241,
"num_tokens": 6535691.0,
"step": 3065
},
{
"entropy": 1.7371309757232667,
"epoch": 1.010533245556287,
"grad_norm": 1.073920726776123,
"learning_rate": 1.0658104608618917e-05,
"loss": 0.0273,
"mean_token_accuracy": 0.9902878165245056,
"num_tokens": 6546449.0,
"step": 3070
},
{
"entropy": 1.7515403628349304,
"epoch": 1.0121790651744569,
"grad_norm": 1.7505379915237427,
"learning_rate": 1.0630947127834962e-05,
"loss": 0.0213,
"mean_token_accuracy": 0.9931905210018158,
"num_tokens": 6557347.0,
"step": 3075
},
{
"entropy": 1.723526620864868,
"epoch": 1.0138248847926268,
"grad_norm": 3.507692575454712,
"learning_rate": 1.0603784974222862e-05,
"loss": 0.0429,
"mean_token_accuracy": 0.9902272820472717,
"num_tokens": 6568263.0,
"step": 3080
},
{
"entropy": 1.7295622825622559,
"epoch": 1.0154707044107967,
"grad_norm": 1.3187499046325684,
"learning_rate": 1.0576618348946982e-05,
"loss": 0.0139,
"mean_token_accuracy": 0.9940691888332367,
"num_tokens": 6579159.0,
"step": 3085
},
{
"entropy": 1.7557840943336487,
"epoch": 1.0171165240289664,
"grad_norm": 0.35923171043395996,
"learning_rate": 1.0549447453204793e-05,
"loss": 0.0163,
"mean_token_accuracy": 0.9933087825775146,
"num_tokens": 6589930.0,
"step": 3090
},
{
"entropy": 1.7555920958518982,
"epoch": 1.0187623436471362,
"grad_norm": 2.311398983001709,
"learning_rate": 1.0522272488225411e-05,
"loss": 0.0226,
"mean_token_accuracy": 0.9911885142326355,
"num_tokens": 6600428.0,
"step": 3095
},
{
"entropy": 1.6874248266220093,
"epoch": 1.0204081632653061,
"grad_norm": 2.7232701778411865,
"learning_rate": 1.049509365526807e-05,
"loss": 0.0322,
"mean_token_accuracy": 0.9916168689727783,
"num_tokens": 6611008.0,
"step": 3100
},
{
"entropy": 1.6934003949165344,
"epoch": 1.022053982883476,
"grad_norm": 3.6890053749084473,
"learning_rate": 1.0467911155620664e-05,
"loss": 0.0252,
"mean_token_accuracy": 0.9931854546070099,
"num_tokens": 6621437.0,
"step": 3105
},
{
"entropy": 1.620060884952545,
"epoch": 1.0236998025016457,
"grad_norm": 1.3696199655532837,
"learning_rate": 1.044072519059824e-05,
"loss": 0.055,
"mean_token_accuracy": 0.9850696086883545,
"num_tokens": 6632746.0,
"step": 3110
},
{
"entropy": 1.725310730934143,
"epoch": 1.0253456221198156,
"grad_norm": 0.7018516659736633,
"learning_rate": 1.0413535961541499e-05,
"loss": 0.0239,
"mean_token_accuracy": 0.9919134438037872,
"num_tokens": 6643199.0,
"step": 3115
},
{
"entropy": 1.6991363763809204,
"epoch": 1.0269914417379855,
"grad_norm": 0.6958704590797424,
"learning_rate": 1.0386343669815333e-05,
"loss": 0.0304,
"mean_token_accuracy": 0.991345739364624,
"num_tokens": 6653990.0,
"step": 3120
},
{
"entropy": 1.7753076672554016,
"epoch": 1.0286372613561554,
"grad_norm": 3.0339064598083496,
"learning_rate": 1.0359148516807302e-05,
"loss": 0.0442,
"mean_token_accuracy": 0.9869567334651947,
"num_tokens": 6664947.0,
"step": 3125
},
{
"entropy": 1.7934301972389222,
"epoch": 1.0302830809743253,
"grad_norm": 3.500246047973633,
"learning_rate": 1.0331950703926165e-05,
"loss": 0.0589,
"mean_token_accuracy": 0.9841761589050293,
"num_tokens": 6675621.0,
"step": 3130
},
{
"entropy": 1.8426469326019288,
"epoch": 1.031928900592495,
"grad_norm": 1.5313292741775513,
"learning_rate": 1.0304750432600377e-05,
"loss": 0.0261,
"mean_token_accuracy": 0.9919860601425171,
"num_tokens": 6686232.0,
"step": 3135
},
{
"entropy": 1.8473934888839723,
"epoch": 1.0335747202106649,
"grad_norm": 2.6027140617370605,
"learning_rate": 1.02775479042766e-05,
"loss": 0.0334,
"mean_token_accuracy": 0.9884464621543885,
"num_tokens": 6696624.0,
"step": 3140
},
{
"entropy": 1.8324166655540466,
"epoch": 1.0352205398288348,
"grad_norm": 0.43423017859458923,
"learning_rate": 1.0250343320418215e-05,
"loss": 0.0232,
"mean_token_accuracy": 0.9935197830200195,
"num_tokens": 6707078.0,
"step": 3145
},
{
"entropy": 1.8373860239982605,
"epoch": 1.0368663594470047,
"grad_norm": 3.9265055656433105,
"learning_rate": 1.0223136882503821e-05,
"loss": 0.0292,
"mean_token_accuracy": 0.9914808392524719,
"num_tokens": 6717515.0,
"step": 3150
},
{
"entropy": 1.850171196460724,
"epoch": 1.0385121790651746,
"grad_norm": 1.000176191329956,
"learning_rate": 1.0195928792025754e-05,
"loss": 0.0236,
"mean_token_accuracy": 0.9916164934635162,
"num_tokens": 6727947.0,
"step": 3155
},
{
"entropy": 1.8611099600791932,
"epoch": 1.0401579986833442,
"grad_norm": 0.9246273636817932,
"learning_rate": 1.016871925048858e-05,
"loss": 0.0208,
"mean_token_accuracy": 0.9946603596210479,
"num_tokens": 6738673.0,
"step": 3160
},
{
"entropy": 1.8230043292045592,
"epoch": 1.0418038183015141,
"grad_norm": 2.162306785583496,
"learning_rate": 1.0141508459407622e-05,
"loss": 0.0322,
"mean_token_accuracy": 0.9894011855125427,
"num_tokens": 6749351.0,
"step": 3165
},
{
"entropy": 1.810911202430725,
"epoch": 1.043449637919684,
"grad_norm": 0.9168899655342102,
"learning_rate": 1.0114296620307455e-05,
"loss": 0.0227,
"mean_token_accuracy": 0.992879319190979,
"num_tokens": 6760013.0,
"step": 3170
},
{
"entropy": 1.7807167649269104,
"epoch": 1.045095457537854,
"grad_norm": 3.3147082328796387,
"learning_rate": 1.0087083934720407e-05,
"loss": 0.0245,
"mean_token_accuracy": 0.9914108633995056,
"num_tokens": 6770834.0,
"step": 3175
},
{
"entropy": 1.7569178104400636,
"epoch": 1.0467412771560236,
"grad_norm": 0.19522562623023987,
"learning_rate": 1.0059870604185087e-05,
"loss": 0.0244,
"mean_token_accuracy": 0.9945703566074371,
"num_tokens": 6781679.0,
"step": 3180
},
{
"entropy": 1.800702166557312,
"epoch": 1.0483870967741935,
"grad_norm": 0.2634085714817047,
"learning_rate": 1.003265683024487e-05,
"loss": 0.022,
"mean_token_accuracy": 0.9923857629299164,
"num_tokens": 6792206.0,
"step": 3185
},
{
"entropy": 1.78583163022995,
"epoch": 1.0500329163923634,
"grad_norm": 2.332631826400757,
"learning_rate": 1.0005442814446427e-05,
"loss": 0.0354,
"mean_token_accuracy": 0.9924177765846253,
"num_tokens": 6802785.0,
"step": 3190
},
{
"entropy": 1.7854429602622985,
"epoch": 1.0516787360105333,
"grad_norm": 3.857931613922119,
"learning_rate": 9.97822875833821e-06,
"loss": 0.0436,
"mean_token_accuracy": 0.9893366694450378,
"num_tokens": 6813246.0,
"step": 3195
},
{
"entropy": 1.7886322617530823,
"epoch": 1.0533245556287032,
"grad_norm": 0.9116196632385254,
"learning_rate": 9.951014863468971e-06,
"loss": 0.0308,
"mean_token_accuracy": 0.9903899192810058,
"num_tokens": 6824016.0,
"step": 3200
},
{
"entropy": 1.831316077709198,
"epoch": 1.0549703752468729,
"grad_norm": 4.522629737854004,
"learning_rate": 9.92380133138627e-06,
"loss": 0.0151,
"mean_token_accuracy": 0.9942168056964874,
"num_tokens": 6834519.0,
"step": 3205
},
{
"entropy": 1.8389117002487183,
"epoch": 1.0566161948650428,
"grad_norm": 3.8139638900756836,
"learning_rate": 9.896588363634983e-06,
"loss": 0.0348,
"mean_token_accuracy": 0.9891720294952393,
"num_tokens": 6845234.0,
"step": 3210
},
{
"entropy": 1.858761990070343,
"epoch": 1.0582620144832127,
"grad_norm": 1.8006324768066406,
"learning_rate": 9.869376161755797e-06,
"loss": 0.0308,
"mean_token_accuracy": 0.9904006004333497,
"num_tokens": 6855897.0,
"step": 3215
},
{
"entropy": 1.9025790572166443,
"epoch": 1.0599078341013826,
"grad_norm": 1.6231938600540161,
"learning_rate": 9.842164927283734e-06,
"loss": 0.0227,
"mean_token_accuracy": 0.9929196774959564,
"num_tokens": 6866586.0,
"step": 3220
},
{
"entropy": 1.8303877234458923,
"epoch": 1.0615536537195522,
"grad_norm": 2.040353298187256,
"learning_rate": 9.814954861746661e-06,
"loss": 0.0155,
"mean_token_accuracy": 0.9953079938888549,
"num_tokens": 6877250.0,
"step": 3225
},
{
"entropy": 1.799347722530365,
"epoch": 1.0631994733377221,
"grad_norm": 3.7308638095855713,
"learning_rate": 9.787746166663765e-06,
"loss": 0.0368,
"mean_token_accuracy": 0.9898776650428772,
"num_tokens": 6887880.0,
"step": 3230
},
{
"entropy": 1.7600205421447754,
"epoch": 1.064845292955892,
"grad_norm": 2.044337749481201,
"learning_rate": 9.760539043544105e-06,
"loss": 0.0464,
"mean_token_accuracy": 0.9879779398441315,
"num_tokens": 6898466.0,
"step": 3235
},
{
"entropy": 1.7838748335838317,
"epoch": 1.066491112574062,
"grad_norm": 0.9999270439147949,
"learning_rate": 9.733333693885078e-06,
"loss": 0.0282,
"mean_token_accuracy": 0.9914038121700287,
"num_tokens": 6909129.0,
"step": 3240
},
{
"entropy": 1.747081458568573,
"epoch": 1.0681369321922318,
"grad_norm": 0.998917281627655,
"learning_rate": 9.706130319170968e-06,
"loss": 0.0376,
"mean_token_accuracy": 0.9899788439273834,
"num_tokens": 6919953.0,
"step": 3245
},
{
"entropy": 1.8464920043945312,
"epoch": 1.0697827518104015,
"grad_norm": 0.4222843050956726,
"learning_rate": 9.678929120871414e-06,
"loss": 0.0173,
"mean_token_accuracy": 0.994493979215622,
"num_tokens": 6930503.0,
"step": 3250
},
{
"entropy": 1.8613513350486754,
"epoch": 1.0714285714285714,
"grad_norm": 5.423548698425293,
"learning_rate": 9.651730300439954e-06,
"loss": 0.0657,
"mean_token_accuracy": 0.9828275620937348,
"num_tokens": 6941390.0,
"step": 3255
},
{
"entropy": 1.809139084815979,
"epoch": 1.0730743910467413,
"grad_norm": 1.3602713346481323,
"learning_rate": 9.62453405931249e-06,
"loss": 0.0237,
"mean_token_accuracy": 0.9926668286323548,
"num_tokens": 6952211.0,
"step": 3260
},
{
"entropy": 1.8179643034934998,
"epoch": 1.0747202106649112,
"grad_norm": 1.4231082201004028,
"learning_rate": 9.597340598905851e-06,
"loss": 0.0248,
"mean_token_accuracy": 0.9913124799728393,
"num_tokens": 6962822.0,
"step": 3265
},
{
"entropy": 1.7864566802978517,
"epoch": 1.076366030283081,
"grad_norm": 1.2587391138076782,
"learning_rate": 9.57015012061625e-06,
"loss": 0.0269,
"mean_token_accuracy": 0.9924269318580627,
"num_tokens": 6973461.0,
"step": 3270
},
{
"entropy": 1.7307344913482665,
"epoch": 1.0780118499012508,
"grad_norm": 0.7557618021965027,
"learning_rate": 9.542962825817827e-06,
"loss": 0.0224,
"mean_token_accuracy": 0.9931135952472687,
"num_tokens": 6984212.0,
"step": 3275
},
{
"entropy": 1.7689846754074097,
"epoch": 1.0796576695194207,
"grad_norm": 3.2683939933776855,
"learning_rate": 9.515778915861136e-06,
"loss": 0.0494,
"mean_token_accuracy": 0.9892696917057038,
"num_tokens": 6994814.0,
"step": 3280
},
{
"entropy": 1.862963318824768,
"epoch": 1.0813034891375906,
"grad_norm": 2.394611120223999,
"learning_rate": 9.488598592071668e-06,
"loss": 0.0244,
"mean_token_accuracy": 0.9937925040721893,
"num_tokens": 7005314.0,
"step": 3285
},
{
"entropy": 1.8323248624801636,
"epoch": 1.0829493087557605,
"grad_norm": 0.8254493474960327,
"learning_rate": 9.461422055748357e-06,
"loss": 0.0258,
"mean_token_accuracy": 0.9926515042781829,
"num_tokens": 7015752.0,
"step": 3290
},
{
"entropy": 1.8604917049407959,
"epoch": 1.0845951283739301,
"grad_norm": 0.3408713936805725,
"learning_rate": 9.434249508162076e-06,
"loss": 0.0322,
"mean_token_accuracy": 0.991688358783722,
"num_tokens": 7026426.0,
"step": 3295
},
{
"entropy": 1.8841025471687316,
"epoch": 1.0862409479921,
"grad_norm": 2.080605983734131,
"learning_rate": 9.407081150554172e-06,
"loss": 0.0284,
"mean_token_accuracy": 0.9886221408843994,
"num_tokens": 7036922.0,
"step": 3300
},
{
"entropy": 1.9134224891662597,
"epoch": 1.08788676761027,
"grad_norm": 1.2799135446548462,
"learning_rate": 9.379917184134949e-06,
"loss": 0.0271,
"mean_token_accuracy": 0.9915643930435181,
"num_tokens": 7047343.0,
"step": 3305
},
{
"entropy": 1.8705042123794555,
"epoch": 1.0895325872284398,
"grad_norm": 0.8514411449432373,
"learning_rate": 9.352757810082196e-06,
"loss": 0.0239,
"mean_token_accuracy": 0.9933820009231568,
"num_tokens": 7057946.0,
"step": 3310
},
{
"entropy": 1.8473629593849181,
"epoch": 1.0911784068466095,
"grad_norm": 0.8632181286811829,
"learning_rate": 9.325603229539684e-06,
"loss": 0.0216,
"mean_token_accuracy": 0.9923954606056213,
"num_tokens": 7068260.0,
"step": 3315
},
{
"entropy": 1.8017066240310669,
"epoch": 1.0928242264647794,
"grad_norm": 1.6297268867492676,
"learning_rate": 9.298453643615692e-06,
"loss": 0.0211,
"mean_token_accuracy": 0.9935080289840699,
"num_tokens": 7078998.0,
"step": 3320
},
{
"entropy": 1.7489684343338012,
"epoch": 1.0944700460829493,
"grad_norm": 0.9504618048667908,
"learning_rate": 9.2713092533815e-06,
"loss": 0.0241,
"mean_token_accuracy": 0.9920541942119598,
"num_tokens": 7090092.0,
"step": 3325
},
{
"entropy": 1.753550660610199,
"epoch": 1.0961158657011192,
"grad_norm": 2.1300246715545654,
"learning_rate": 9.244170259869918e-06,
"loss": 0.0414,
"mean_token_accuracy": 0.9851863026618958,
"num_tokens": 7100582.0,
"step": 3330
},
{
"entropy": 1.7830381631851195,
"epoch": 1.097761685319289,
"grad_norm": 1.9206792116165161,
"learning_rate": 9.217036864073776e-06,
"loss": 0.0305,
"mean_token_accuracy": 0.989460003376007,
"num_tokens": 7110954.0,
"step": 3335
},
{
"entropy": 1.7082879304885865,
"epoch": 1.0994075049374588,
"grad_norm": 1.520218849182129,
"learning_rate": 9.189909266944459e-06,
"loss": 0.0367,
"mean_token_accuracy": 0.9915473163127899,
"num_tokens": 7121660.0,
"step": 3340
},
{
"entropy": 1.7186639666557313,
"epoch": 1.1010533245556287,
"grad_norm": 0.7555062770843506,
"learning_rate": 9.162787669390398e-06,
"loss": 0.0332,
"mean_token_accuracy": 0.9881877601146698,
"num_tokens": 7132239.0,
"step": 3345
},
{
"entropy": 1.7227881669998169,
"epoch": 1.1026991441737986,
"grad_norm": 1.7483959197998047,
"learning_rate": 9.135672272275593e-06,
"loss": 0.0258,
"mean_token_accuracy": 0.9922352850437164,
"num_tokens": 7143078.0,
"step": 3350
},
{
"entropy": 1.7311092615127563,
"epoch": 1.1043449637919684,
"grad_norm": 0.43958139419555664,
"learning_rate": 9.10856327641813e-06,
"loss": 0.0279,
"mean_token_accuracy": 0.9921256840229035,
"num_tokens": 7153701.0,
"step": 3355
},
{
"entropy": 1.6882481575012207,
"epoch": 1.1059907834101383,
"grad_norm": 2.0378565788269043,
"learning_rate": 9.081460882588668e-06,
"loss": 0.0168,
"mean_token_accuracy": 0.995065587759018,
"num_tokens": 7164253.0,
"step": 3360
},
{
"entropy": 1.6925435185432434,
"epoch": 1.107636603028308,
"grad_norm": 0.6830073595046997,
"learning_rate": 9.054365291508998e-06,
"loss": 0.0258,
"mean_token_accuracy": 0.9935117542743683,
"num_tokens": 7175017.0,
"step": 3365
},
{
"entropy": 1.7125192880630493,
"epoch": 1.109282422646478,
"grad_norm": 0.8072521090507507,
"learning_rate": 9.027276703850505e-06,
"loss": 0.0279,
"mean_token_accuracy": 0.9914268732070923,
"num_tokens": 7185582.0,
"step": 3370
},
{
"entropy": 1.6770400524139404,
"epoch": 1.1109282422646478,
"grad_norm": 3.0566606521606445,
"learning_rate": 9.000195320232724e-06,
"loss": 0.0457,
"mean_token_accuracy": 0.987758994102478,
"num_tokens": 7196254.0,
"step": 3375
},
{
"entropy": 1.7625356912612915,
"epoch": 1.1125740618828177,
"grad_norm": 2.011699914932251,
"learning_rate": 8.973121341221823e-06,
"loss": 0.0248,
"mean_token_accuracy": 0.9929955065250397,
"num_tokens": 7206923.0,
"step": 3380
},
{
"entropy": 1.7391006708145142,
"epoch": 1.1142198815009876,
"grad_norm": 2.011082887649536,
"learning_rate": 8.946054967329142e-06,
"loss": 0.0243,
"mean_token_accuracy": 0.9917931973934173,
"num_tokens": 7217251.0,
"step": 3385
},
{
"entropy": 1.732879626750946,
"epoch": 1.1158657011191573,
"grad_norm": 1.593976378440857,
"learning_rate": 8.918996399009689e-06,
"loss": 0.0246,
"mean_token_accuracy": 0.9882166802883148,
"num_tokens": 7227764.0,
"step": 3390
},
{
"entropy": 1.6700308084487916,
"epoch": 1.1175115207373272,
"grad_norm": 1.129323124885559,
"learning_rate": 8.891945836660673e-06,
"loss": 0.0283,
"mean_token_accuracy": 0.9915276169776917,
"num_tokens": 7238505.0,
"step": 3395
},
{
"entropy": 1.688683032989502,
"epoch": 1.119157340355497,
"grad_norm": 2.0232725143432617,
"learning_rate": 8.864903480619996e-06,
"loss": 0.0405,
"mean_token_accuracy": 0.9898577332496643,
"num_tokens": 7249030.0,
"step": 3400
},
{
"entropy": 1.6767473936080932,
"epoch": 1.120803159973667,
"grad_norm": 1.5392571687698364,
"learning_rate": 8.837869531164792e-06,
"loss": 0.0253,
"mean_token_accuracy": 0.9922107934951783,
"num_tokens": 7259760.0,
"step": 3405
},
{
"entropy": 1.6974989771842957,
"epoch": 1.1224489795918366,
"grad_norm": 1.918980360031128,
"learning_rate": 8.810844188509946e-06,
"loss": 0.0405,
"mean_token_accuracy": 0.9870936632156372,
"num_tokens": 7270389.0,
"step": 3410
},
{
"entropy": 1.6960701942443848,
"epoch": 1.1240947992100065,
"grad_norm": 1.4752129316329956,
"learning_rate": 8.783827652806577e-06,
"loss": 0.0225,
"mean_token_accuracy": 0.9912503600120545,
"num_tokens": 7281116.0,
"step": 3415
},
{
"entropy": 1.6978899121284485,
"epoch": 1.1257406188281764,
"grad_norm": 2.9839565753936768,
"learning_rate": 8.756820124140602e-06,
"loss": 0.0327,
"mean_token_accuracy": 0.9897141098976135,
"num_tokens": 7291516.0,
"step": 3420
},
{
"entropy": 1.6348261475563048,
"epoch": 1.1273864384463463,
"grad_norm": 2.958974599838257,
"learning_rate": 8.729821802531213e-06,
"loss": 0.0287,
"mean_token_accuracy": 0.9899660766124725,
"num_tokens": 7302127.0,
"step": 3425
},
{
"entropy": 1.671372151374817,
"epoch": 1.129032258064516,
"grad_norm": 1.5821384191513062,
"learning_rate": 8.70283288792943e-06,
"loss": 0.0213,
"mean_token_accuracy": 0.9928022623062134,
"num_tokens": 7312938.0,
"step": 3430
},
{
"entropy": 1.7028130054473878,
"epoch": 1.130678077682686,
"grad_norm": 1.062685489654541,
"learning_rate": 8.67585358021659e-06,
"loss": 0.037,
"mean_token_accuracy": 0.9904503941535949,
"num_tokens": 7323516.0,
"step": 3435
},
{
"entropy": 1.698065197467804,
"epoch": 1.1323238973008558,
"grad_norm": 3.0057718753814697,
"learning_rate": 8.648884079202896e-06,
"loss": 0.0186,
"mean_token_accuracy": 0.9931861937046051,
"num_tokens": 7333911.0,
"step": 3440
},
{
"entropy": 1.6505720615386963,
"epoch": 1.1339697169190257,
"grad_norm": 1.6598554849624634,
"learning_rate": 8.6219245846259e-06,
"loss": 0.0174,
"mean_token_accuracy": 0.9931141972541809,
"num_tokens": 7344558.0,
"step": 3445
},
{
"entropy": 1.6310924172401429,
"epoch": 1.1356155365371956,
"grad_norm": 2.144071578979492,
"learning_rate": 8.594975296149076e-06,
"loss": 0.0195,
"mean_token_accuracy": 0.9929544270038605,
"num_tokens": 7355291.0,
"step": 3450
},
{
"entropy": 1.600458061695099,
"epoch": 1.1372613561553653,
"grad_norm": 0.6878706216812134,
"learning_rate": 8.568036413360283e-06,
"loss": 0.0291,
"mean_token_accuracy": 0.9917424976825714,
"num_tokens": 7365889.0,
"step": 3455
},
{
"entropy": 1.6151419758796692,
"epoch": 1.1389071757735352,
"grad_norm": 1.7664144039154053,
"learning_rate": 8.541108135770327e-06,
"loss": 0.0225,
"mean_token_accuracy": 0.9897194743156433,
"num_tokens": 7376467.0,
"step": 3460
},
{
"entropy": 1.6531212329864502,
"epoch": 1.140552995391705,
"grad_norm": 3.0760936737060547,
"learning_rate": 8.514190662811477e-06,
"loss": 0.0251,
"mean_token_accuracy": 0.9931772351264954,
"num_tokens": 7387048.0,
"step": 3465
},
{
"entropy": 1.6679503798484803,
"epoch": 1.142198815009875,
"grad_norm": 2.6920888423919678,
"learning_rate": 8.48728419383597e-06,
"loss": 0.0298,
"mean_token_accuracy": 0.9894031941890716,
"num_tokens": 7397403.0,
"step": 3470
},
{
"entropy": 1.5688734650611877,
"epoch": 1.1438446346280449,
"grad_norm": 2.5779929161071777,
"learning_rate": 8.46038892811456e-06,
"loss": 0.0249,
"mean_token_accuracy": 0.9909287571907044,
"num_tokens": 7408276.0,
"step": 3475
},
{
"entropy": 1.6224292755126952,
"epoch": 1.1454904542462145,
"grad_norm": 0.8870710134506226,
"learning_rate": 8.433505064835012e-06,
"loss": 0.0293,
"mean_token_accuracy": 0.9906193792819977,
"num_tokens": 7418781.0,
"step": 3480
},
{
"entropy": 1.644513690471649,
"epoch": 1.1471362738643844,
"grad_norm": 1.0466939210891724,
"learning_rate": 8.406632803100665e-06,
"loss": 0.034,
"mean_token_accuracy": 0.9898303151130676,
"num_tokens": 7429487.0,
"step": 3485
},
{
"entropy": 1.6431127905845642,
"epoch": 1.1487820934825543,
"grad_norm": 2.5962440967559814,
"learning_rate": 8.379772341928916e-06,
"loss": 0.0348,
"mean_token_accuracy": 0.9878771305084229,
"num_tokens": 7439874.0,
"step": 3490
},
{
"entropy": 1.6485112071037293,
"epoch": 1.1504279131007242,
"grad_norm": 5.498775005340576,
"learning_rate": 8.352923880249784e-06,
"loss": 0.0329,
"mean_token_accuracy": 0.9888358414173126,
"num_tokens": 7450555.0,
"step": 3495
},
{
"entropy": 1.648984158039093,
"epoch": 1.1520737327188941,
"grad_norm": 1.443813681602478,
"learning_rate": 8.326087616904401e-06,
"loss": 0.038,
"mean_token_accuracy": 0.9887622356414795,
"num_tokens": 7461255.0,
"step": 3500
},
{
"entropy": 1.6465593934059144,
"epoch": 1.1537195523370638,
"grad_norm": 1.363283395767212,
"learning_rate": 8.299263750643577e-06,
"loss": 0.0244,
"mean_token_accuracy": 0.9922839045524597,
"num_tokens": 7472013.0,
"step": 3505
},
{
"entropy": 1.6541425347328187,
"epoch": 1.1553653719552337,
"grad_norm": 2.518134355545044,
"learning_rate": 8.272452480126292e-06,
"loss": 0.0341,
"mean_token_accuracy": 0.9866783797740937,
"num_tokens": 7483066.0,
"step": 3510
},
{
"entropy": 1.6475726962089539,
"epoch": 1.1570111915734036,
"grad_norm": 2.8579108715057373,
"learning_rate": 8.245654003918253e-06,
"loss": 0.0305,
"mean_token_accuracy": 0.9904314935207367,
"num_tokens": 7493486.0,
"step": 3515
},
{
"entropy": 1.6358031511306763,
"epoch": 1.1586570111915735,
"grad_norm": 2.28322172164917,
"learning_rate": 8.218868520490404e-06,
"loss": 0.0154,
"mean_token_accuracy": 0.9955400943756103,
"num_tokens": 7504061.0,
"step": 3520
},
{
"entropy": 1.6113605737686156,
"epoch": 1.1603028308097432,
"grad_norm": 1.8561283349990845,
"learning_rate": 8.192096228217464e-06,
"loss": 0.0138,
"mean_token_accuracy": 0.9960749268531799,
"num_tokens": 7514617.0,
"step": 3525
},
{
"entropy": 1.5644055128097534,
"epoch": 1.161948650427913,
"grad_norm": 2.0931262969970703,
"learning_rate": 8.165337325376467e-06,
"loss": 0.0282,
"mean_token_accuracy": 0.9914573907852173,
"num_tokens": 7525370.0,
"step": 3530
},
{
"entropy": 1.502221119403839,
"epoch": 1.163594470046083,
"grad_norm": 2.6928858757019043,
"learning_rate": 8.138592010145273e-06,
"loss": 0.0152,
"mean_token_accuracy": 0.9963742852210998,
"num_tokens": 7536226.0,
"step": 3535
},
{
"entropy": 1.5423286318778993,
"epoch": 1.1652402896642529,
"grad_norm": 2.7901740074157715,
"learning_rate": 8.111860480601117e-06,
"loss": 0.0464,
"mean_token_accuracy": 0.9899639308452606,
"num_tokens": 7546765.0,
"step": 3540
},
{
"entropy": 1.6044429183006286,
"epoch": 1.1668861092824225,
"grad_norm": 0.5412013530731201,
"learning_rate": 8.085142934719131e-06,
"loss": 0.015,
"mean_token_accuracy": 0.9947813093662262,
"num_tokens": 7557291.0,
"step": 3545
},
{
"entropy": 1.5845131874084473,
"epoch": 1.1685319289005924,
"grad_norm": 1.3245034217834473,
"learning_rate": 8.058439570370896e-06,
"loss": 0.0269,
"mean_token_accuracy": 0.9917524337768555,
"num_tokens": 7568110.0,
"step": 3550
},
{
"entropy": 1.576395034790039,
"epoch": 1.1701777485187623,
"grad_norm": 1.6202564239501953,
"learning_rate": 8.031750585322948e-06,
"loss": 0.0353,
"mean_token_accuracy": 0.9901875615119934,
"num_tokens": 7578544.0,
"step": 3555
},
{
"entropy": 1.6079319357872008,
"epoch": 1.1718235681369322,
"grad_norm": 1.0341594219207764,
"learning_rate": 8.005076177235337e-06,
"loss": 0.0217,
"mean_token_accuracy": 0.9924035966396332,
"num_tokens": 7589338.0,
"step": 3560
},
{
"entropy": 1.5887961030006408,
"epoch": 1.1734693877551021,
"grad_norm": 2.502521276473999,
"learning_rate": 7.978416543660157e-06,
"loss": 0.042,
"mean_token_accuracy": 0.9888204395771026,
"num_tokens": 7599872.0,
"step": 3565
},
{
"entropy": 1.5681854605674743,
"epoch": 1.1751152073732718,
"grad_norm": 0.6041249632835388,
"learning_rate": 7.951771882040082e-06,
"loss": 0.0141,
"mean_token_accuracy": 0.9941197037696838,
"num_tokens": 7610467.0,
"step": 3570
},
{
"entropy": 1.5940211772918702,
"epoch": 1.1767610269914417,
"grad_norm": 2.8404202461242676,
"learning_rate": 7.92514238970689e-06,
"loss": 0.0284,
"mean_token_accuracy": 0.9908346593379974,
"num_tokens": 7621740.0,
"step": 3575
},
{
"entropy": 1.6273322224617004,
"epoch": 1.1784068466096116,
"grad_norm": 2.6306891441345215,
"learning_rate": 7.898528263880032e-06,
"loss": 0.0421,
"mean_token_accuracy": 0.9906798958778381,
"num_tokens": 7632253.0,
"step": 3580
},
{
"entropy": 1.6058209180831908,
"epoch": 1.1800526662277815,
"grad_norm": 2.210401773452759,
"learning_rate": 7.871929701665147e-06,
"loss": 0.02,
"mean_token_accuracy": 0.9937019526958466,
"num_tokens": 7642945.0,
"step": 3585
},
{
"entropy": 1.6328086972236633,
"epoch": 1.1816984858459514,
"grad_norm": 1.6380540132522583,
"learning_rate": 7.8453469000526e-06,
"loss": 0.0288,
"mean_token_accuracy": 0.9929502904415131,
"num_tokens": 7653553.0,
"step": 3590
},
{
"entropy": 1.6912413835525513,
"epoch": 1.183344305464121,
"grad_norm": 4.160943508148193,
"learning_rate": 7.818780055916052e-06,
"loss": 0.0195,
"mean_token_accuracy": 0.9919560194015503,
"num_tokens": 7663891.0,
"step": 3595
},
{
"entropy": 1.6919382691383362,
"epoch": 1.184990125082291,
"grad_norm": 0.695886492729187,
"learning_rate": 7.792229366010959e-06,
"loss": 0.0248,
"mean_token_accuracy": 0.9903152227401734,
"num_tokens": 7674488.0,
"step": 3600
},
{
"entropy": 1.6337382435798644,
"epoch": 1.1866359447004609,
"grad_norm": 1.5294524431228638,
"learning_rate": 7.765695026973155e-06,
"loss": 0.0196,
"mean_token_accuracy": 0.9943537652492523,
"num_tokens": 7685143.0,
"step": 3605
},
{
"entropy": 1.6397574901580811,
"epoch": 1.1882817643186308,
"grad_norm": 2.282261610031128,
"learning_rate": 7.73917723531737e-06,
"loss": 0.0403,
"mean_token_accuracy": 0.9924940526485443,
"num_tokens": 7695569.0,
"step": 3610
},
{
"entropy": 1.5973503470420838,
"epoch": 1.1899275839368006,
"grad_norm": 5.261276721954346,
"learning_rate": 7.71267618743579e-06,
"loss": 0.0227,
"mean_token_accuracy": 0.9901642620563507,
"num_tokens": 7706303.0,
"step": 3615
},
{
"entropy": 1.6234167337417602,
"epoch": 1.1915734035549703,
"grad_norm": 2.048280954360962,
"learning_rate": 7.686192079596586e-06,
"loss": 0.0197,
"mean_token_accuracy": 0.9935329794883728,
"num_tokens": 7716665.0,
"step": 3620
},
{
"entropy": 1.6198643565177917,
"epoch": 1.1932192231731402,
"grad_norm": 1.0738399028778076,
"learning_rate": 7.659725107942484e-06,
"loss": 0.0267,
"mean_token_accuracy": 0.9920990586280822,
"num_tokens": 7727458.0,
"step": 3625
},
{
"entropy": 1.658357810974121,
"epoch": 1.1948650427913101,
"grad_norm": 1.3478312492370605,
"learning_rate": 7.633275468489278e-06,
"loss": 0.0254,
"mean_token_accuracy": 0.9936649382114411,
"num_tokens": 7738325.0,
"step": 3630
},
{
"entropy": 1.7477381587028504,
"epoch": 1.19651086240948,
"grad_norm": 2.4919769763946533,
"learning_rate": 7.606843357124426e-06,
"loss": 0.0399,
"mean_token_accuracy": 0.9883269786834716,
"num_tokens": 7748772.0,
"step": 3635
},
{
"entropy": 1.7540024757385253,
"epoch": 1.1981566820276497,
"grad_norm": 1.3142441511154175,
"learning_rate": 7.5804289696055445e-06,
"loss": 0.0355,
"mean_token_accuracy": 0.9913724660873413,
"num_tokens": 7759350.0,
"step": 3640
},
{
"entropy": 1.7711892247200012,
"epoch": 1.1998025016458196,
"grad_norm": 2.53527569770813,
"learning_rate": 7.554032501559002e-06,
"loss": 0.0313,
"mean_token_accuracy": 0.9935957908630371,
"num_tokens": 7770189.0,
"step": 3645
},
{
"epoch": 1.2007899934167214,
"eval_entropy": 1.8328242014223972,
"eval_loss": 0.055253952741622925,
"eval_mean_token_accuracy": 0.9852129074527483,
"eval_num_tokens": 7776505.0,
"eval_runtime": 197.2997,
"eval_samples_per_second": 42.205,
"eval_steps_per_second": 7.035,
"step": 3648
},
{
"entropy": 1.820277488231659,
"epoch": 1.2014483212639895,
"grad_norm": 1.2625054121017456,
"learning_rate": 7.52765414847846e-06,
"loss": 0.0367,
"mean_token_accuracy": 0.9870195090770721,
"num_tokens": 7780633.0,
"step": 3650
},
{
"entropy": 1.850312602519989,
"epoch": 1.2030941408821594,
"grad_norm": 2.3248748779296875,
"learning_rate": 7.5012941057234015e-06,
"loss": 0.0348,
"mean_token_accuracy": 0.988330626487732,
"num_tokens": 7791063.0,
"step": 3655
},
{
"entropy": 1.7486730933189392,
"epoch": 1.204739960500329,
"grad_norm": 2.7954654693603516,
"learning_rate": 7.4749525685177145e-06,
"loss": 0.0345,
"mean_token_accuracy": 0.9921814143657685,
"num_tokens": 7801794.0,
"step": 3660
},
{
"entropy": 1.7875786900520325,
"epoch": 1.206385780118499,
"grad_norm": 1.7330853939056396,
"learning_rate": 7.44862973194823e-06,
"loss": 0.0377,
"mean_token_accuracy": 0.9861465573310852,
"num_tokens": 7812680.0,
"step": 3665
},
{
"entropy": 1.7702386498451232,
"epoch": 1.2080315997366688,
"grad_norm": 1.6117557287216187,
"learning_rate": 7.422325790963286e-06,
"loss": 0.0228,
"mean_token_accuracy": 0.9921964347362519,
"num_tokens": 7823265.0,
"step": 3670
},
{
"entropy": 1.750121569633484,
"epoch": 1.2096774193548387,
"grad_norm": 1.6110137701034546,
"learning_rate": 7.39604094037127e-06,
"loss": 0.0226,
"mean_token_accuracy": 0.9932965815067292,
"num_tokens": 7833616.0,
"step": 3675
},
{
"entropy": 1.7086394667625426,
"epoch": 1.2113232389730086,
"grad_norm": 5.368286609649658,
"learning_rate": 7.369775374839196e-06,
"loss": 0.0257,
"mean_token_accuracy": 0.9906518340110779,
"num_tokens": 7844464.0,
"step": 3680
},
{
"entropy": 1.7639584064483642,
"epoch": 1.2129690585911783,
"grad_norm": 0.626293420791626,
"learning_rate": 7.343529288891239e-06,
"loss": 0.0251,
"mean_token_accuracy": 0.9907599687576294,
"num_tokens": 7855198.0,
"step": 3685
},
{
"entropy": 1.7602010250091553,
"epoch": 1.2146148782093482,
"grad_norm": 0.723297119140625,
"learning_rate": 7.317302876907318e-06,
"loss": 0.0312,
"mean_token_accuracy": 0.9907152950763702,
"num_tokens": 7865513.0,
"step": 3690
},
{
"entropy": 1.780686092376709,
"epoch": 1.2162606978275181,
"grad_norm": 0.540903627872467,
"learning_rate": 7.29109633312164e-06,
"loss": 0.0224,
"mean_token_accuracy": 0.9916972815990448,
"num_tokens": 7875954.0,
"step": 3695
},
{
"entropy": 1.7540069580078126,
"epoch": 1.217906517445688,
"grad_norm": 1.341984510421753,
"learning_rate": 7.2649098516212715e-06,
"loss": 0.0382,
"mean_token_accuracy": 0.9883949160575867,
"num_tokens": 7886845.0,
"step": 3700
},
{
"entropy": 1.759465503692627,
"epoch": 1.219552337063858,
"grad_norm": 1.8915423154830933,
"learning_rate": 7.2387436263446885e-06,
"loss": 0.0222,
"mean_token_accuracy": 0.9931784868240356,
"num_tokens": 7897311.0,
"step": 3705
},
{
"entropy": 1.768437671661377,
"epoch": 1.2211981566820276,
"grad_norm": 1.4099783897399902,
"learning_rate": 7.2125978510803565e-06,
"loss": 0.0268,
"mean_token_accuracy": 0.9915161907672883,
"num_tokens": 7907999.0,
"step": 3710
},
{
"entropy": 1.7534563660621643,
"epoch": 1.2228439763001975,
"grad_norm": 1.8560601472854614,
"learning_rate": 7.1864727194652874e-06,
"loss": 0.0295,
"mean_token_accuracy": 0.9926966905593873,
"num_tokens": 7918701.0,
"step": 3715
},
{
"entropy": 1.7628652215003968,
"epoch": 1.2244897959183674,
"grad_norm": 1.2373379468917847,
"learning_rate": 7.1603684249835966e-06,
"loss": 0.0226,
"mean_token_accuracy": 0.9931581020355225,
"num_tokens": 7929045.0,
"step": 3720
},
{
"entropy": 1.7709421277046205,
"epoch": 1.2261356155365373,
"grad_norm": 1.8868582248687744,
"learning_rate": 7.134285160965091e-06,
"loss": 0.0213,
"mean_token_accuracy": 0.9951976418495179,
"num_tokens": 7939549.0,
"step": 3725
},
{
"entropy": 1.7510177731513976,
"epoch": 1.2277814351547072,
"grad_norm": 1.6280816793441772,
"learning_rate": 7.108223120583806e-06,
"loss": 0.0431,
"mean_token_accuracy": 0.9894875645637512,
"num_tokens": 7950189.0,
"step": 3730
},
{
"entropy": 1.7418721914291382,
"epoch": 1.2294272547728768,
"grad_norm": 2.297192335128784,
"learning_rate": 7.0821824968566186e-06,
"loss": 0.0192,
"mean_token_accuracy": 0.9930408895015717,
"num_tokens": 7960790.0,
"step": 3735
},
{
"entropy": 1.7448366165161133,
"epoch": 1.2310730743910467,
"grad_norm": 2.935821294784546,
"learning_rate": 7.056163482641769e-06,
"loss": 0.027,
"mean_token_accuracy": 0.9896710455417633,
"num_tokens": 7971434.0,
"step": 3740
},
{
"entropy": 1.6850064516067504,
"epoch": 1.2327188940092166,
"grad_norm": 1.6496083736419678,
"learning_rate": 7.030166270637475e-06,
"loss": 0.0238,
"mean_token_accuracy": 0.9923489093780518,
"num_tokens": 7981902.0,
"step": 3745
},
{
"entropy": 1.6959288597106934,
"epoch": 1.2343647136273865,
"grad_norm": 4.362974643707275,
"learning_rate": 7.004191053380469e-06,
"loss": 0.028,
"mean_token_accuracy": 0.9899141073226929,
"num_tokens": 7992608.0,
"step": 3750
},
{
"entropy": 1.7090837240219117,
"epoch": 1.2360105332455562,
"grad_norm": 1.9156920909881592,
"learning_rate": 6.978238023244608e-06,
"loss": 0.0274,
"mean_token_accuracy": 0.9905971348285675,
"num_tokens": 8003032.0,
"step": 3755
},
{
"entropy": 1.6994237184524537,
"epoch": 1.237656352863726,
"grad_norm": 1.834991455078125,
"learning_rate": 6.952307372439411e-06,
"loss": 0.0264,
"mean_token_accuracy": 0.9915358006954194,
"num_tokens": 8013591.0,
"step": 3760
},
{
"entropy": 1.6717028617858887,
"epoch": 1.239302172481896,
"grad_norm": 1.8597207069396973,
"learning_rate": 6.926399293008668e-06,
"loss": 0.0233,
"mean_token_accuracy": 0.9935126841068268,
"num_tokens": 8024009.0,
"step": 3765
},
{
"entropy": 1.6924420833587646,
"epoch": 1.240947992100066,
"grad_norm": 2.2939343452453613,
"learning_rate": 6.900513976829e-06,
"loss": 0.0323,
"mean_token_accuracy": 0.9890012562274932,
"num_tokens": 8034690.0,
"step": 3770
},
{
"entropy": 1.746755886077881,
"epoch": 1.2425938117182356,
"grad_norm": 2.126143455505371,
"learning_rate": 6.874651615608441e-06,
"loss": 0.0174,
"mean_token_accuracy": 0.9934778928756713,
"num_tokens": 8045114.0,
"step": 3775
},
{
"entropy": 1.6977684855461121,
"epoch": 1.2442396313364055,
"grad_norm": 0.6126706600189209,
"learning_rate": 6.848812400885022e-06,
"loss": 0.0192,
"mean_token_accuracy": 0.9939151763916015,
"num_tokens": 8055532.0,
"step": 3780
},
{
"entropy": 1.6669098496437074,
"epoch": 1.2458854509545754,
"grad_norm": 0.8724939227104187,
"learning_rate": 6.822996524025343e-06,
"loss": 0.0197,
"mean_token_accuracy": 0.9912198483943939,
"num_tokens": 8066233.0,
"step": 3785
},
{
"entropy": 1.6577038049697876,
"epoch": 1.2475312705727453,
"grad_norm": 0.9917463064193726,
"learning_rate": 6.7972041762231735e-06,
"loss": 0.025,
"mean_token_accuracy": 0.990100759267807,
"num_tokens": 8076870.0,
"step": 3790
},
{
"entropy": 1.6371492862701416,
"epoch": 1.2491770901909152,
"grad_norm": 3.3986692428588867,
"learning_rate": 6.771435548498013e-06,
"loss": 0.035,
"mean_token_accuracy": 0.9852859020233155,
"num_tokens": 8087788.0,
"step": 3795
},
{
"entropy": 1.6349127650260926,
"epoch": 1.2508229098090848,
"grad_norm": 0.3032724857330322,
"learning_rate": 6.745690831693701e-06,
"loss": 0.0251,
"mean_token_accuracy": 0.993361896276474,
"num_tokens": 8098543.0,
"step": 3800
},
{
"entropy": 1.6615733742713927,
"epoch": 1.2524687294272547,
"grad_norm": 1.261225700378418,
"learning_rate": 6.719970216476982e-06,
"loss": 0.0253,
"mean_token_accuracy": 0.9910578489303589,
"num_tokens": 8109340.0,
"step": 3805
},
{
"entropy": 1.680140745639801,
"epoch": 1.2541145490454246,
"grad_norm": 1.3544028997421265,
"learning_rate": 6.694273893336112e-06,
"loss": 0.0229,
"mean_token_accuracy": 0.9918768882751465,
"num_tokens": 8119963.0,
"step": 3810
},
{
"entropy": 1.6792070388793945,
"epoch": 1.2557603686635945,
"grad_norm": 1.9016791582107544,
"learning_rate": 6.668602052579425e-06,
"loss": 0.0263,
"mean_token_accuracy": 0.9923485815525055,
"num_tokens": 8130505.0,
"step": 3815
},
{
"entropy": 1.7175181269645692,
"epoch": 1.2574061882817644,
"grad_norm": 0.3682861924171448,
"learning_rate": 6.6429548843339554e-06,
"loss": 0.0225,
"mean_token_accuracy": 0.991203784942627,
"num_tokens": 8141063.0,
"step": 3820
},
{
"entropy": 1.6892056941986084,
"epoch": 1.259052007899934,
"grad_norm": 5.687158107757568,
"learning_rate": 6.617332578543991e-06,
"loss": 0.0289,
"mean_token_accuracy": 0.9894017159938813,
"num_tokens": 8152016.0,
"step": 3825
},
{
"entropy": 1.7227007865905761,
"epoch": 1.260697827518104,
"grad_norm": 2.130305290222168,
"learning_rate": 6.591735324969703e-06,
"loss": 0.0423,
"mean_token_accuracy": 0.9899180591106415,
"num_tokens": 8162786.0,
"step": 3830
},
{
"entropy": 1.766884195804596,
"epoch": 1.262343647136274,
"grad_norm": 2.274033546447754,
"learning_rate": 6.566163313185725e-06,
"loss": 0.0261,
"mean_token_accuracy": 0.9887346982955932,
"num_tokens": 8173362.0,
"step": 3835
},
{
"entropy": 1.8262192487716675,
"epoch": 1.2639894667544438,
"grad_norm": 0.5273262858390808,
"learning_rate": 6.540616732579732e-06,
"loss": 0.0387,
"mean_token_accuracy": 0.9902793288230896,
"num_tokens": 8183767.0,
"step": 3840
},
{
"entropy": 1.8414955258369445,
"epoch": 1.2656352863726137,
"grad_norm": 0.5974478721618652,
"learning_rate": 6.515095772351072e-06,
"loss": 0.0124,
"mean_token_accuracy": 0.99626624584198,
"num_tokens": 8194098.0,
"step": 3845
},
{
"entropy": 1.8668495297431946,
"epoch": 1.2672811059907834,
"grad_norm": 1.7619863748550415,
"learning_rate": 6.489600621509338e-06,
"loss": 0.0206,
"mean_token_accuracy": 0.9931797504425048,
"num_tokens": 8204769.0,
"step": 3850
},
{
"entropy": 1.8715619444847107,
"epoch": 1.2689269256089533,
"grad_norm": 2.3929049968719482,
"learning_rate": 6.464131468872978e-06,
"loss": 0.0168,
"mean_token_accuracy": 0.9927472591400146,
"num_tokens": 8215368.0,
"step": 3855
},
{
"entropy": 1.8072044491767882,
"epoch": 1.2705727452271232,
"grad_norm": 2.2075207233428955,
"learning_rate": 6.4386885030679e-06,
"loss": 0.0332,
"mean_token_accuracy": 0.992115718126297,
"num_tokens": 8226508.0,
"step": 3860
},
{
"entropy": 1.905867862701416,
"epoch": 1.2722185648452928,
"grad_norm": 1.7269535064697266,
"learning_rate": 6.413271912526071e-06,
"loss": 0.0218,
"mean_token_accuracy": 0.993303781747818,
"num_tokens": 8237461.0,
"step": 3865
},
{
"entropy": 1.8554383277893067,
"epoch": 1.2738643844634627,
"grad_norm": 1.8244965076446533,
"learning_rate": 6.3878818854841095e-06,
"loss": 0.0202,
"mean_token_accuracy": 0.9931906521320343,
"num_tokens": 8248162.0,
"step": 3870
},
{
"entropy": 1.903454637527466,
"epoch": 1.2755102040816326,
"grad_norm": 1.2271250486373901,
"learning_rate": 6.36251860998192e-06,
"loss": 0.0204,
"mean_token_accuracy": 0.9948872745037078,
"num_tokens": 8258852.0,
"step": 3875
},
{
"entropy": 1.8506133675575256,
"epoch": 1.2771560236998025,
"grad_norm": 1.973997712135315,
"learning_rate": 6.337182273861273e-06,
"loss": 0.03,
"mean_token_accuracy": 0.9918236672878266,
"num_tokens": 8269756.0,
"step": 3880
},
{
"entropy": 1.8246717214584351,
"epoch": 1.2788018433179724,
"grad_norm": 1.5060253143310547,
"learning_rate": 6.311873064764429e-06,
"loss": 0.0203,
"mean_token_accuracy": 0.992903220653534,
"num_tokens": 8280446.0,
"step": 3885
},
{
"entropy": 1.7956886053085328,
"epoch": 1.280447662936142,
"grad_norm": 2.8490633964538574,
"learning_rate": 6.2865911701327445e-06,
"loss": 0.0434,
"mean_token_accuracy": 0.9881740510463715,
"num_tokens": 8291519.0,
"step": 3890
},
{
"entropy": 1.7901652932167054,
"epoch": 1.282093482554312,
"grad_norm": 2.23622727394104,
"learning_rate": 6.261336777205278e-06,
"loss": 0.0304,
"mean_token_accuracy": 0.9898859024047851,
"num_tokens": 8302444.0,
"step": 3895
},
{
"entropy": 1.795331597328186,
"epoch": 1.2837393021724819,
"grad_norm": 2.6051158905029297,
"learning_rate": 6.236110073017417e-06,
"loss": 0.0215,
"mean_token_accuracy": 0.9921227872371674,
"num_tokens": 8313239.0,
"step": 3900
},
{
"entropy": 1.8181136965751648,
"epoch": 1.2853851217906518,
"grad_norm": 1.3954198360443115,
"learning_rate": 6.210911244399477e-06,
"loss": 0.0241,
"mean_token_accuracy": 0.9936698138713836,
"num_tokens": 8323898.0,
"step": 3905
},
{
"entropy": 1.801988172531128,
"epoch": 1.2870309414088217,
"grad_norm": 1.1723895072937012,
"learning_rate": 6.185740477975335e-06,
"loss": 0.0183,
"mean_token_accuracy": 0.9923419177532196,
"num_tokens": 8334473.0,
"step": 3910
},
{
"entropy": 1.783213496208191,
"epoch": 1.2886767610269914,
"grad_norm": 4.490993499755859,
"learning_rate": 6.160597960161024e-06,
"loss": 0.0452,
"mean_token_accuracy": 0.9866098284721374,
"num_tokens": 8345014.0,
"step": 3915
},
{
"entropy": 1.73745299577713,
"epoch": 1.2903225806451613,
"grad_norm": 2.8388960361480713,
"learning_rate": 6.135483877163383e-06,
"loss": 0.0243,
"mean_token_accuracy": 0.9913448512554168,
"num_tokens": 8355724.0,
"step": 3920
},
{
"entropy": 1.7496454119682312,
"epoch": 1.2919684002633312,
"grad_norm": 1.7147318124771118,
"learning_rate": 6.1103984149786444e-06,
"loss": 0.0262,
"mean_token_accuracy": 0.9924720883369446,
"num_tokens": 8366189.0,
"step": 3925
},
{
"entropy": 1.7078065276145935,
"epoch": 1.293614219881501,
"grad_norm": 2.615743398666382,
"learning_rate": 6.085341759391089e-06,
"loss": 0.0138,
"mean_token_accuracy": 0.9948653101921081,
"num_tokens": 8376886.0,
"step": 3930
},
{
"entropy": 1.7135319232940673,
"epoch": 1.295260039499671,
"grad_norm": 0.48522359132766724,
"learning_rate": 6.060314095971641e-06,
"loss": 0.0258,
"mean_token_accuracy": 0.9912464320659637,
"num_tokens": 8387635.0,
"step": 3935
},
{
"entropy": 1.7192439675331115,
"epoch": 1.2969058591178406,
"grad_norm": 0.810149073600769,
"learning_rate": 6.035315610076518e-06,
"loss": 0.0246,
"mean_token_accuracy": 0.9908073008060455,
"num_tokens": 8398117.0,
"step": 3940
},
{
"entropy": 1.6878206253051757,
"epoch": 1.2985516787360105,
"grad_norm": 1.6084260940551758,
"learning_rate": 6.010346486845837e-06,
"loss": 0.0238,
"mean_token_accuracy": 0.9926587820053101,
"num_tokens": 8408773.0,
"step": 3945
},
{
"entropy": 1.682581651210785,
"epoch": 1.3001974983541804,
"grad_norm": 2.0567915439605713,
"learning_rate": 5.985406911202263e-06,
"loss": 0.0169,
"mean_token_accuracy": 0.9948483347892761,
"num_tokens": 8419245.0,
"step": 3950
},
{
"entropy": 1.6870299458503724,
"epoch": 1.3018433179723503,
"grad_norm": 2.500347375869751,
"learning_rate": 5.960497067849627e-06,
"loss": 0.0298,
"mean_token_accuracy": 0.9893729627132416,
"num_tokens": 8430019.0,
"step": 3955
},
{
"entropy": 1.7270505785942079,
"epoch": 1.3034891375905202,
"grad_norm": 2.2009787559509277,
"learning_rate": 5.935617141271554e-06,
"loss": 0.0255,
"mean_token_accuracy": 0.9894290864467621,
"num_tokens": 8440514.0,
"step": 3960
},
{
"entropy": 1.7091917991638184,
"epoch": 1.3051349572086899,
"grad_norm": 3.052794933319092,
"learning_rate": 5.910767315730119e-06,
"loss": 0.0366,
"mean_token_accuracy": 0.9881552219390869,
"num_tokens": 8451099.0,
"step": 3965
},
{
"entropy": 1.7440539956092835,
"epoch": 1.3067807768268598,
"grad_norm": 1.05941641330719,
"learning_rate": 5.885947775264447e-06,
"loss": 0.0241,
"mean_token_accuracy": 0.9918931663036347,
"num_tokens": 8461462.0,
"step": 3970
},
{
"entropy": 1.751047670841217,
"epoch": 1.3084265964450297,
"grad_norm": 1.5784897804260254,
"learning_rate": 5.861158703689389e-06,
"loss": 0.0216,
"mean_token_accuracy": 0.9915867269039154,
"num_tokens": 8472292.0,
"step": 3975
},
{
"entropy": 1.7028122067451477,
"epoch": 1.3100724160631994,
"grad_norm": 1.145462155342102,
"learning_rate": 5.836400284594126e-06,
"loss": 0.0247,
"mean_token_accuracy": 0.9926209390163422,
"num_tokens": 8482780.0,
"step": 3980
},
{
"entropy": 1.7101820945739745,
"epoch": 1.3117182356813692,
"grad_norm": 1.7238191366195679,
"learning_rate": 5.811672701340847e-06,
"loss": 0.0314,
"mean_token_accuracy": 0.9896239399909973,
"num_tokens": 8493301.0,
"step": 3985
},
{
"entropy": 1.7027180314064025,
"epoch": 1.3133640552995391,
"grad_norm": 1.4847729206085205,
"learning_rate": 5.786976137063336e-06,
"loss": 0.0209,
"mean_token_accuracy": 0.9931684076786041,
"num_tokens": 8504099.0,
"step": 3990
},
{
"entropy": 1.7060617804527283,
"epoch": 1.315009874917709,
"grad_norm": 0.6948561072349548,
"learning_rate": 5.762310774665682e-06,
"loss": 0.0404,
"mean_token_accuracy": 0.9896627187728881,
"num_tokens": 8514523.0,
"step": 3995
},
{
"entropy": 1.7185169339179993,
"epoch": 1.316655694535879,
"grad_norm": 1.8867969512939453,
"learning_rate": 5.737676796820871e-06,
"loss": 0.018,
"mean_token_accuracy": 0.9954242944717407,
"num_tokens": 8525127.0,
"step": 4000
},
{
"entropy": 1.6883883237838746,
"epoch": 1.3183015141540486,
"grad_norm": 2.294797420501709,
"learning_rate": 5.713074385969457e-06,
"loss": 0.018,
"mean_token_accuracy": 0.9943160593509675,
"num_tokens": 8535541.0,
"step": 4005
},
{
"entropy": 1.6336073637008668,
"epoch": 1.3199473337722185,
"grad_norm": 0.9374617338180542,
"learning_rate": 5.688503724318217e-06,
"loss": 0.0102,
"mean_token_accuracy": 0.9958161950111389,
"num_tokens": 8546220.0,
"step": 4010
},
{
"entropy": 1.6534652829170227,
"epoch": 1.3215931533903884,
"grad_norm": 1.0187243223190308,
"learning_rate": 5.663964993838779e-06,
"loss": 0.0204,
"mean_token_accuracy": 0.9947121858596801,
"num_tokens": 8556714.0,
"step": 4015
},
{
"entropy": 1.6586306810379028,
"epoch": 1.3232389730085583,
"grad_norm": 2.5042691230773926,
"learning_rate": 5.639458376266295e-06,
"loss": 0.0344,
"mean_token_accuracy": 0.9897667288780212,
"num_tokens": 8567099.0,
"step": 4020
},
{
"entropy": 1.6488097548484801,
"epoch": 1.3248847926267282,
"grad_norm": 0.3930192291736603,
"learning_rate": 5.614984053098076e-06,
"loss": 0.029,
"mean_token_accuracy": 0.9916586577892303,
"num_tokens": 8577866.0,
"step": 4025
},
{
"entropy": 1.6415641903877258,
"epoch": 1.3265306122448979,
"grad_norm": 1.2496612071990967,
"learning_rate": 5.590542205592283e-06,
"loss": 0.0288,
"mean_token_accuracy": 0.9903595089912415,
"num_tokens": 8588800.0,
"step": 4030
},
{
"entropy": 1.6364824771881104,
"epoch": 1.3281764318630678,
"grad_norm": 2.232058048248291,
"learning_rate": 5.566133014766526e-06,
"loss": 0.0333,
"mean_token_accuracy": 0.9903694272041321,
"num_tokens": 8599559.0,
"step": 4035
},
{
"entropy": 1.699565863609314,
"epoch": 1.3298222514812377,
"grad_norm": 2.588693618774414,
"learning_rate": 5.541756661396591e-06,
"loss": 0.0229,
"mean_token_accuracy": 0.9930707275867462,
"num_tokens": 8610049.0,
"step": 4040
},
{
"entropy": 1.6720836758613586,
"epoch": 1.3314680710994076,
"grad_norm": 1.1052114963531494,
"learning_rate": 5.517413326015046e-06,
"loss": 0.0325,
"mean_token_accuracy": 0.9897372424602509,
"num_tokens": 8620905.0,
"step": 4045
},
{
"entropy": 1.7122917652130127,
"epoch": 1.3331138907175775,
"grad_norm": 1.7371225357055664,
"learning_rate": 5.493103188909939e-06,
"loss": 0.0256,
"mean_token_accuracy": 0.9902097463607789,
"num_tokens": 8631605.0,
"step": 4050
},
{
"entropy": 1.7047467947006225,
"epoch": 1.3347597103357471,
"grad_norm": 3.5729825496673584,
"learning_rate": 5.468826430123435e-06,
"loss": 0.021,
"mean_token_accuracy": 0.992638248205185,
"num_tokens": 8642346.0,
"step": 4055
},
{
"entropy": 1.7031975507736206,
"epoch": 1.336405529953917,
"grad_norm": 2.91983962059021,
"learning_rate": 5.444583229450518e-06,
"loss": 0.0317,
"mean_token_accuracy": 0.9902465045452118,
"num_tokens": 8653100.0,
"step": 4060
},
{
"entropy": 1.697403860092163,
"epoch": 1.338051349572087,
"grad_norm": 1.3388501405715942,
"learning_rate": 5.4203737664376235e-06,
"loss": 0.0412,
"mean_token_accuracy": 0.9881835162639618,
"num_tokens": 8663816.0,
"step": 4065
},
{
"entropy": 1.7462014317512513,
"epoch": 1.3396971691902566,
"grad_norm": 5.225465774536133,
"learning_rate": 5.3961982203813215e-06,
"loss": 0.0383,
"mean_token_accuracy": 0.9885516107082367,
"num_tokens": 8674657.0,
"step": 4070
},
{
"entropy": 1.7663294553756714,
"epoch": 1.3413429888084267,
"grad_norm": 1.6548188924789429,
"learning_rate": 5.3720567703270135e-06,
"loss": 0.0239,
"mean_token_accuracy": 0.99189293384552,
"num_tokens": 8685457.0,
"step": 4075
},
{
"entropy": 1.7495159268379212,
"epoch": 1.3429888084265964,
"grad_norm": 1.063901662826538,
"learning_rate": 5.34794959506755e-06,
"loss": 0.0237,
"mean_token_accuracy": 0.9917633235454559,
"num_tokens": 8695979.0,
"step": 4080
},
{
"entropy": 1.7705445170402527,
"epoch": 1.3446346280447663,
"grad_norm": 0.8427658677101135,
"learning_rate": 5.323876873141973e-06,
"loss": 0.0224,
"mean_token_accuracy": 0.9938309490680695,
"num_tokens": 8706574.0,
"step": 4085
},
{
"entropy": 1.7015425443649292,
"epoch": 1.3462804476629362,
"grad_norm": 4.552727222442627,
"learning_rate": 5.299838782834141e-06,
"loss": 0.0273,
"mean_token_accuracy": 0.9894029378890992,
"num_tokens": 8717255.0,
"step": 4090
},
{
"entropy": 1.7069133877754212,
"epoch": 1.3479262672811059,
"grad_norm": 4.266534805297852,
"learning_rate": 5.275835502171439e-06,
"loss": 0.024,
"mean_token_accuracy": 0.992630559206009,
"num_tokens": 8728164.0,
"step": 4095
},
{
"entropy": 1.730119228363037,
"epoch": 1.3495720868992758,
"grad_norm": 2.2071967124938965,
"learning_rate": 5.251867208923439e-06,
"loss": 0.0151,
"mean_token_accuracy": 0.9952519774436951,
"num_tokens": 8738646.0,
"step": 4100
},
{
"entropy": 1.7257731676101684,
"epoch": 1.3512179065174457,
"grad_norm": 3.7456114292144775,
"learning_rate": 5.227934080600611e-06,
"loss": 0.032,
"mean_token_accuracy": 0.9912447273731232,
"num_tokens": 8749142.0,
"step": 4105
},
{
"entropy": 1.6916332483291625,
"epoch": 1.3528637261356156,
"grad_norm": 2.742368459701538,
"learning_rate": 5.2040362944529765e-06,
"loss": 0.0182,
"mean_token_accuracy": 0.9944073140621186,
"num_tokens": 8759739.0,
"step": 4110
},
{
"entropy": 1.6806737899780273,
"epoch": 1.3545095457537855,
"grad_norm": 1.4229061603546143,
"learning_rate": 5.180174027468818e-06,
"loss": 0.0184,
"mean_token_accuracy": 0.9935629963874817,
"num_tokens": 8770377.0,
"step": 4115
},
{
"entropy": 1.6647310853004456,
"epoch": 1.3561553653719551,
"grad_norm": 2.2806296348571777,
"learning_rate": 5.156347456373359e-06,
"loss": 0.0251,
"mean_token_accuracy": 0.9908474087715149,
"num_tokens": 8780910.0,
"step": 4120
},
{
"entropy": 1.6626045823097229,
"epoch": 1.357801184990125,
"grad_norm": 1.6852504014968872,
"learning_rate": 5.1325567576274595e-06,
"loss": 0.0328,
"mean_token_accuracy": 0.9896870791912079,
"num_tokens": 8791765.0,
"step": 4125
},
{
"entropy": 1.7072691798210144,
"epoch": 1.359447004608295,
"grad_norm": 0.524549126625061,
"learning_rate": 5.108802107426307e-06,
"loss": 0.0152,
"mean_token_accuracy": 0.9940847456455231,
"num_tokens": 8802312.0,
"step": 4130
},
{
"entropy": 1.7237987995147706,
"epoch": 1.3610928242264648,
"grad_norm": 2.021660566329956,
"learning_rate": 5.085083681698108e-06,
"loss": 0.0437,
"mean_token_accuracy": 0.9889907777309418,
"num_tokens": 8813114.0,
"step": 4135
},
{
"entropy": 1.7635319709777832,
"epoch": 1.3627386438446347,
"grad_norm": 3.0634493827819824,
"learning_rate": 5.061401656102791e-06,
"loss": 0.04,
"mean_token_accuracy": 0.9863172173500061,
"num_tokens": 8823776.0,
"step": 4140
},
{
"entropy": 1.7534740567207336,
"epoch": 1.3643844634628044,
"grad_norm": 1.1463490724563599,
"learning_rate": 5.0377562060307e-06,
"loss": 0.032,
"mean_token_accuracy": 0.9871594190597535,
"num_tokens": 8834704.0,
"step": 4145
},
{
"entropy": 1.7889039039611816,
"epoch": 1.3660302830809743,
"grad_norm": 0.34571316838264465,
"learning_rate": 5.014147506601308e-06,
"loss": 0.0239,
"mean_token_accuracy": 0.9936646819114685,
"num_tokens": 8845254.0,
"step": 4150
},
{
"entropy": 1.8070958495140075,
"epoch": 1.3676761026991442,
"grad_norm": 1.5474847555160522,
"learning_rate": 4.990575732661902e-06,
"loss": 0.0287,
"mean_token_accuracy": 0.9930408537387848,
"num_tokens": 8855930.0,
"step": 4155
},
{
"entropy": 1.7869633316993714,
"epoch": 1.369321922317314,
"grad_norm": 1.3527427911758423,
"learning_rate": 4.9670410587862995e-06,
"loss": 0.0341,
"mean_token_accuracy": 0.9886087715625763,
"num_tokens": 8866777.0,
"step": 4160
},
{
"entropy": 1.8497827887535094,
"epoch": 1.370967741935484,
"grad_norm": 1.8562260866165161,
"learning_rate": 4.943543659273548e-06,
"loss": 0.0476,
"mean_token_accuracy": 0.9888940870761871,
"num_tokens": 8877374.0,
"step": 4165
},
{
"entropy": 1.8215659737586976,
"epoch": 1.3726135615536537,
"grad_norm": 1.3257911205291748,
"learning_rate": 4.920083708146655e-06,
"loss": 0.0185,
"mean_token_accuracy": 0.9913751244544983,
"num_tokens": 8888152.0,
"step": 4170
},
{
"entropy": 1.8450712442398072,
"epoch": 1.3742593811718236,
"grad_norm": 0.5730196833610535,
"learning_rate": 4.896661379151259e-06,
"loss": 0.0186,
"mean_token_accuracy": 0.9925144612789154,
"num_tokens": 8898690.0,
"step": 4175
},
{
"entropy": 1.8470216035842895,
"epoch": 1.3759052007899935,
"grad_norm": 1.4637004137039185,
"learning_rate": 4.873276845754388e-06,
"loss": 0.0141,
"mean_token_accuracy": 0.9950567662715912,
"num_tokens": 8909192.0,
"step": 4180
},
{
"entropy": 1.7927292108535766,
"epoch": 1.3775510204081631,
"grad_norm": 2.084494113922119,
"learning_rate": 4.849930281143146e-06,
"loss": 0.0424,
"mean_token_accuracy": 0.9881355404853821,
"num_tokens": 8920016.0,
"step": 4185
},
{
"entropy": 1.815053677558899,
"epoch": 1.3791968400263332,
"grad_norm": 1.7285300493240356,
"learning_rate": 4.826621858223431e-06,
"loss": 0.0247,
"mean_token_accuracy": 0.992021232843399,
"num_tokens": 8930392.0,
"step": 4190
},
{
"entropy": 1.80452299118042,
"epoch": 1.380842659644503,
"grad_norm": 4.97409725189209,
"learning_rate": 4.803351749618679e-06,
"loss": 0.0234,
"mean_token_accuracy": 0.9906729996204376,
"num_tokens": 8940931.0,
"step": 4195
},
{
"entropy": 1.7994644165039062,
"epoch": 1.3824884792626728,
"grad_norm": 1.7542165517807007,
"learning_rate": 4.780120127668553e-06,
"loss": 0.019,
"mean_token_accuracy": 0.9934443414211274,
"num_tokens": 8951603.0,
"step": 4200
},
{
"entropy": 1.800536823272705,
"epoch": 1.3841342988808427,
"grad_norm": 1.1492308378219604,
"learning_rate": 4.756927164427685e-06,
"loss": 0.0318,
"mean_token_accuracy": 0.9890105724334717,
"num_tokens": 8962110.0,
"step": 4205
},
{
"entropy": 1.7841524839401246,
"epoch": 1.3857801184990124,
"grad_norm": 2.564023971557617,
"learning_rate": 4.733773031664398e-06,
"loss": 0.0292,
"mean_token_accuracy": 0.9909674048423767,
"num_tokens": 8972933.0,
"step": 4210
},
{
"entropy": 1.8291552066802979,
"epoch": 1.3874259381171823,
"grad_norm": 0.70340496301651,
"learning_rate": 4.710657900859447e-06,
"loss": 0.0205,
"mean_token_accuracy": 0.9959937572479248,
"num_tokens": 8983508.0,
"step": 4215
},
{
"entropy": 1.8385101079940795,
"epoch": 1.3890717577353522,
"grad_norm": 1.934988021850586,
"learning_rate": 4.687581943204711e-06,
"loss": 0.031,
"mean_token_accuracy": 0.9913569748401642,
"num_tokens": 8994226.0,
"step": 4220
},
{
"entropy": 1.8304347276687623,
"epoch": 1.390717577353522,
"grad_norm": 1.6579409837722778,
"learning_rate": 4.664545329601977e-06,
"loss": 0.0269,
"mean_token_accuracy": 0.9913767516613007,
"num_tokens": 9004843.0,
"step": 4225
},
{
"entropy": 1.842625379562378,
"epoch": 1.392363396971692,
"grad_norm": 1.2633566856384277,
"learning_rate": 4.641548230661633e-06,
"loss": 0.0327,
"mean_token_accuracy": 0.990666514635086,
"num_tokens": 9015138.0,
"step": 4230
},
{
"entropy": 1.822058653831482,
"epoch": 1.3940092165898617,
"grad_norm": 2.8293750286102295,
"learning_rate": 4.618590816701422e-06,
"loss": 0.0265,
"mean_token_accuracy": 0.9897272527217865,
"num_tokens": 9025856.0,
"step": 4235
},
{
"entropy": 1.8393550515174866,
"epoch": 1.3956550362080316,
"grad_norm": 0.7082259654998779,
"learning_rate": 4.5956732577451745e-06,
"loss": 0.0155,
"mean_token_accuracy": 0.9960557222366333,
"num_tokens": 9036445.0,
"step": 4240
},
{
"entropy": 1.8048294067382813,
"epoch": 1.3973008558262014,
"grad_norm": 0.9008603096008301,
"learning_rate": 4.572795723521565e-06,
"loss": 0.0173,
"mean_token_accuracy": 0.9946598470211029,
"num_tokens": 9047123.0,
"step": 4245
},
{
"entropy": 1.8322034358978272,
"epoch": 1.3989466754443713,
"grad_norm": 0.7377194166183472,
"learning_rate": 4.549958383462829e-06,
"loss": 0.0232,
"mean_token_accuracy": 0.9925904154777527,
"num_tokens": 9057591.0,
"step": 4250
},
{
"entropy": 1.8085259079933167,
"epoch": 1.4005924950625412,
"grad_norm": 0.7783912420272827,
"learning_rate": 4.527161406703525e-06,
"loss": 0.0381,
"mean_token_accuracy": 0.9926593840122223,
"num_tokens": 9068234.0,
"step": 4255
},
{
"epoch": 1.400921658986175,
"eval_entropy": 1.8214755838130325,
"eval_loss": 0.05032181739807129,
"eval_mean_token_accuracy": 0.9866692206959216,
"eval_num_tokens": 9070425.0,
"eval_runtime": 195.1937,
"eval_samples_per_second": 42.66,
"eval_steps_per_second": 7.111,
"step": 4256
},
{
"entropy": 1.8127322912216186,
"epoch": 1.402238314680711,
"grad_norm": 2.2625997066497803,
"learning_rate": 4.504404962079293e-06,
"loss": 0.0238,
"mean_token_accuracy": 0.9912942230701447,
"num_tokens": 9079026.0,
"step": 4260
},
{
"entropy": 1.825773298740387,
"epoch": 1.4038841342988808,
"grad_norm": 1.3523027896881104,
"learning_rate": 4.481689218125561e-06,
"loss": 0.0213,
"mean_token_accuracy": 0.9924202978610992,
"num_tokens": 9089666.0,
"step": 4265
},
{
"entropy": 1.790573537349701,
"epoch": 1.4055299539170507,
"grad_norm": 2.0441343784332275,
"learning_rate": 4.459014343076356e-06,
"loss": 0.0412,
"mean_token_accuracy": 0.9893443644046783,
"num_tokens": 9100438.0,
"step": 4270
},
{
"entropy": 1.8189255356788636,
"epoch": 1.4071757735352206,
"grad_norm": 1.1708464622497559,
"learning_rate": 4.436380504863008e-06,
"loss": 0.0169,
"mean_token_accuracy": 0.9938211143016815,
"num_tokens": 9111194.0,
"step": 4275
},
{
"entropy": 1.8196102857589722,
"epoch": 1.4088215931533905,
"grad_norm": 2.159729242324829,
"learning_rate": 4.413787871112934e-06,
"loss": 0.0197,
"mean_token_accuracy": 0.9932897388935089,
"num_tokens": 9122132.0,
"step": 4280
},
{
"entropy": 1.805807101726532,
"epoch": 1.4104674127715602,
"grad_norm": 2.360366106033325,
"learning_rate": 4.391236609148381e-06,
"loss": 0.0336,
"mean_token_accuracy": 0.9923444032669068,
"num_tokens": 9132718.0,
"step": 4285
},
{
"entropy": 1.813056230545044,
"epoch": 1.41211323238973,
"grad_norm": 0.6563708782196045,
"learning_rate": 4.3687268859852105e-06,
"loss": 0.0137,
"mean_token_accuracy": 0.9964501917362213,
"num_tokens": 9143366.0,
"step": 4290
},
{
"entropy": 1.8228686213493348,
"epoch": 1.4137590520079,
"grad_norm": 1.703528642654419,
"learning_rate": 4.34625886833163e-06,
"loss": 0.019,
"mean_token_accuracy": 0.9924368858337402,
"num_tokens": 9154131.0,
"step": 4295
},
{
"entropy": 1.763486886024475,
"epoch": 1.4154048716260696,
"grad_norm": 2.912816286087036,
"learning_rate": 4.323832722586979e-06,
"loss": 0.0141,
"mean_token_accuracy": 0.9942897677421569,
"num_tokens": 9165049.0,
"step": 4300
},
{
"entropy": 1.8135482668876648,
"epoch": 1.4170506912442398,
"grad_norm": 1.8574903011322021,
"learning_rate": 4.301448614840487e-06,
"loss": 0.0423,
"mean_token_accuracy": 0.9888120889663696,
"num_tokens": 9175582.0,
"step": 4305
},
{
"entropy": 1.8078377962112426,
"epoch": 1.4186965108624094,
"grad_norm": 1.1886160373687744,
"learning_rate": 4.279106710870059e-06,
"loss": 0.0164,
"mean_token_accuracy": 0.9940821826457977,
"num_tokens": 9186081.0,
"step": 4310
},
{
"entropy": 1.7859516978263854,
"epoch": 1.4203423304805793,
"grad_norm": 1.7242977619171143,
"learning_rate": 4.256807176141028e-06,
"loss": 0.0131,
"mean_token_accuracy": 0.99482541680336,
"num_tokens": 9196757.0,
"step": 4315
},
{
"entropy": 1.7669077515602112,
"epoch": 1.4219881500987492,
"grad_norm": 2.3228442668914795,
"learning_rate": 4.2345501758049365e-06,
"loss": 0.0278,
"mean_token_accuracy": 0.9920706391334534,
"num_tokens": 9207097.0,
"step": 4320
},
{
"entropy": 1.8107630848884582,
"epoch": 1.423633969716919,
"grad_norm": 2.6625027656555176,
"learning_rate": 4.2123358746983225e-06,
"loss": 0.041,
"mean_token_accuracy": 0.9908157408237457,
"num_tokens": 9217832.0,
"step": 4325
},
{
"entropy": 1.8029837489128113,
"epoch": 1.4252797893350888,
"grad_norm": 1.867008090019226,
"learning_rate": 4.190164437341479e-06,
"loss": 0.0204,
"mean_token_accuracy": 0.9934187710285187,
"num_tokens": 9228324.0,
"step": 4330
},
{
"entropy": 1.7727699518203734,
"epoch": 1.4269256089532587,
"grad_norm": 5.103766918182373,
"learning_rate": 4.168036027937267e-06,
"loss": 0.0227,
"mean_token_accuracy": 0.9946336686611176,
"num_tokens": 9239199.0,
"step": 4335
},
{
"entropy": 1.8098699450492859,
"epoch": 1.4285714285714286,
"grad_norm": 0.7097901105880737,
"learning_rate": 4.145950810369863e-06,
"loss": 0.0226,
"mean_token_accuracy": 0.9928686380386352,
"num_tokens": 9249956.0,
"step": 4340
},
{
"entropy": 1.7923429489135743,
"epoch": 1.4302172481895985,
"grad_norm": 1.6268037557601929,
"learning_rate": 4.1239089482035686e-06,
"loss": 0.0244,
"mean_token_accuracy": 0.9906105041503906,
"num_tokens": 9260596.0,
"step": 4345
},
{
"entropy": 1.7969419360160828,
"epoch": 1.4318630678077682,
"grad_norm": 1.7350503206253052,
"learning_rate": 4.10191060468159e-06,
"loss": 0.0182,
"mean_token_accuracy": 0.9944913208484649,
"num_tokens": 9271060.0,
"step": 4350
},
{
"entropy": 1.7492168545722961,
"epoch": 1.433508887425938,
"grad_norm": 3.6994593143463135,
"learning_rate": 4.079955942724845e-06,
"loss": 0.0326,
"mean_token_accuracy": 0.989514821767807,
"num_tokens": 9281476.0,
"step": 4355
},
{
"entropy": 1.720914614200592,
"epoch": 1.435154707044108,
"grad_norm": 0.9848082661628723,
"learning_rate": 4.0580451249307195e-06,
"loss": 0.0215,
"mean_token_accuracy": 0.9936264753341675,
"num_tokens": 9292136.0,
"step": 4360
},
{
"entropy": 1.7023021578788757,
"epoch": 1.4368005266622779,
"grad_norm": 0.8126395344734192,
"learning_rate": 4.03617831357191e-06,
"loss": 0.0172,
"mean_token_accuracy": 0.994192260503769,
"num_tokens": 9302979.0,
"step": 4365
},
{
"entropy": 1.6948873043060302,
"epoch": 1.4384463462804478,
"grad_norm": 4.326878547668457,
"learning_rate": 4.014355670595189e-06,
"loss": 0.0211,
"mean_token_accuracy": 0.9916200697422027,
"num_tokens": 9313652.0,
"step": 4370
},
{
"entropy": 1.7300003051757813,
"epoch": 1.4400921658986174,
"grad_norm": 0.8770890235900879,
"learning_rate": 3.99257735762021e-06,
"loss": 0.0311,
"mean_token_accuracy": 0.9879982471466064,
"num_tokens": 9324219.0,
"step": 4375
},
{
"entropy": 1.728563916683197,
"epoch": 1.4417379855167873,
"grad_norm": 1.0835927724838257,
"learning_rate": 3.970843535938332e-06,
"loss": 0.0218,
"mean_token_accuracy": 0.9914672791957855,
"num_tokens": 9334800.0,
"step": 4380
},
{
"entropy": 1.7277835011482239,
"epoch": 1.4433838051349572,
"grad_norm": 1.7556816339492798,
"learning_rate": 3.949154366511395e-06,
"loss": 0.024,
"mean_token_accuracy": 0.9926247537136078,
"num_tokens": 9345539.0,
"step": 4385
},
{
"entropy": 1.729397702217102,
"epoch": 1.4450296247531271,
"grad_norm": 0.777664840221405,
"learning_rate": 3.927510009970548e-06,
"loss": 0.0231,
"mean_token_accuracy": 0.9917639851570129,
"num_tokens": 9356267.0,
"step": 4390
},
{
"entropy": 1.7339849829673768,
"epoch": 1.446675444371297,
"grad_norm": 2.712010383605957,
"learning_rate": 3.905910626615046e-06,
"loss": 0.0407,
"mean_token_accuracy": 0.9871343851089478,
"num_tokens": 9367167.0,
"step": 4395
},
{
"entropy": 1.7719183206558227,
"epoch": 1.4483212639894667,
"grad_norm": 1.5582504272460938,
"learning_rate": 3.884356376411089e-06,
"loss": 0.0214,
"mean_token_accuracy": 0.9926240026950837,
"num_tokens": 9377539.0,
"step": 4400
},
{
"entropy": 1.7510367512702942,
"epoch": 1.4499670836076366,
"grad_norm": 1.32984459400177,
"learning_rate": 3.862847418990592e-06,
"loss": 0.0321,
"mean_token_accuracy": 0.9916741907596588,
"num_tokens": 9388175.0,
"step": 4405
},
{
"entropy": 1.7010186672210694,
"epoch": 1.4516129032258065,
"grad_norm": 2.487252950668335,
"learning_rate": 3.841383913650052e-06,
"loss": 0.0429,
"mean_token_accuracy": 0.9880555152893067,
"num_tokens": 9398803.0,
"step": 4410
},
{
"entropy": 1.7146742224693299,
"epoch": 1.4532587228439762,
"grad_norm": 1.8718352317810059,
"learning_rate": 3.819966019349334e-06,
"loss": 0.0141,
"mean_token_accuracy": 0.9953644871711731,
"num_tokens": 9409524.0,
"step": 4415
},
{
"entropy": 1.6977456569671632,
"epoch": 1.454904542462146,
"grad_norm": 0.736453115940094,
"learning_rate": 3.7985938947105073e-06,
"loss": 0.0157,
"mean_token_accuracy": 0.9962522923946381,
"num_tokens": 9420376.0,
"step": 4420
},
{
"entropy": 1.7264664053916932,
"epoch": 1.456550362080316,
"grad_norm": 1.4576939344406128,
"learning_rate": 3.7772676980166655e-06,
"loss": 0.0282,
"mean_token_accuracy": 0.9920338273048401,
"num_tokens": 9430868.0,
"step": 4425
},
{
"entropy": 1.6992250204086303,
"epoch": 1.4581961816984859,
"grad_norm": 1.5100425481796265,
"learning_rate": 3.7559875872107677e-06,
"loss": 0.0359,
"mean_token_accuracy": 0.9908050179481507,
"num_tokens": 9441532.0,
"step": 4430
},
{
"entropy": 1.7052590131759644,
"epoch": 1.4598420013166558,
"grad_norm": 2.6018645763397217,
"learning_rate": 3.7347537198944483e-06,
"loss": 0.0213,
"mean_token_accuracy": 0.992498379945755,
"num_tokens": 9452300.0,
"step": 4435
},
{
"entropy": 1.6945829153060914,
"epoch": 1.4614878209348254,
"grad_norm": 1.3602592945098877,
"learning_rate": 3.71356625332686e-06,
"loss": 0.0188,
"mean_token_accuracy": 0.9949553668498993,
"num_tokens": 9462853.0,
"step": 4440
},
{
"entropy": 1.6705161213874817,
"epoch": 1.4631336405529953,
"grad_norm": 0.6543161273002625,
"learning_rate": 3.6924253444235224e-06,
"loss": 0.0178,
"mean_token_accuracy": 0.9919328927993775,
"num_tokens": 9473560.0,
"step": 4445
},
{
"entropy": 1.6843745946884154,
"epoch": 1.4647794601711652,
"grad_norm": 1.519394874572754,
"learning_rate": 3.671331149755123e-06,
"loss": 0.0143,
"mean_token_accuracy": 0.9956124365329743,
"num_tokens": 9484022.0,
"step": 4450
},
{
"entropy": 1.6935065150260926,
"epoch": 1.4664252797893351,
"grad_norm": 5.51362943649292,
"learning_rate": 3.6502838255464045e-06,
"loss": 0.0319,
"mean_token_accuracy": 0.9897434711456299,
"num_tokens": 9494354.0,
"step": 4455
},
{
"entropy": 1.632759189605713,
"epoch": 1.468071099407505,
"grad_norm": 1.1584962606430054,
"learning_rate": 3.6292835276749715e-06,
"loss": 0.0238,
"mean_token_accuracy": 0.992765587568283,
"num_tokens": 9505092.0,
"step": 4460
},
{
"entropy": 1.6529253125190735,
"epoch": 1.4697169190256747,
"grad_norm": 0.6863316893577576,
"learning_rate": 3.6083304116701535e-06,
"loss": 0.018,
"mean_token_accuracy": 0.9955762267112732,
"num_tokens": 9515748.0,
"step": 4465
},
{
"entropy": 1.6669641017913819,
"epoch": 1.4713627386438446,
"grad_norm": 0.19445084035396576,
"learning_rate": 3.587424632711841e-06,
"loss": 0.0305,
"mean_token_accuracy": 0.9905413866043091,
"num_tokens": 9526278.0,
"step": 4470
},
{
"entropy": 1.6730605006217956,
"epoch": 1.4730085582620145,
"grad_norm": 2.6678473949432373,
"learning_rate": 3.56656634562936e-06,
"loss": 0.024,
"mean_token_accuracy": 0.9918078184127808,
"num_tokens": 9537087.0,
"step": 4475
},
{
"entropy": 1.6874117732048035,
"epoch": 1.4746543778801844,
"grad_norm": 0.8383910655975342,
"learning_rate": 3.5457557049002934e-06,
"loss": 0.0173,
"mean_token_accuracy": 0.9937337219715119,
"num_tokens": 9547735.0,
"step": 4480
},
{
"entropy": 1.68024080991745,
"epoch": 1.4763001974983543,
"grad_norm": 2.533155918121338,
"learning_rate": 3.52499286464936e-06,
"loss": 0.0287,
"mean_token_accuracy": 0.9925665855407715,
"num_tokens": 9558405.0,
"step": 4485
},
{
"entropy": 1.6580557942390441,
"epoch": 1.477946017116524,
"grad_norm": 4.7806830406188965,
"learning_rate": 3.5042779786472602e-06,
"loss": 0.0245,
"mean_token_accuracy": 0.9935833632946014,
"num_tokens": 9569152.0,
"step": 4490
},
{
"entropy": 1.6794435620307921,
"epoch": 1.4795918367346939,
"grad_norm": 2.641050100326538,
"learning_rate": 3.4836112003095524e-06,
"loss": 0.0237,
"mean_token_accuracy": 0.9917592287063599,
"num_tokens": 9579673.0,
"step": 4495
},
{
"entropy": 1.6873149633407594,
"epoch": 1.4812376563528638,
"grad_norm": 1.5926048755645752,
"learning_rate": 3.4629926826954997e-06,
"loss": 0.0347,
"mean_token_accuracy": 0.9919157564640045,
"num_tokens": 9590146.0,
"step": 4500
},
{
"entropy": 1.6695775389671326,
"epoch": 1.4828834759710336,
"grad_norm": 3.867030382156372,
"learning_rate": 3.4424225785069444e-06,
"loss": 0.0248,
"mean_token_accuracy": 0.993567830324173,
"num_tokens": 9600795.0,
"step": 4505
},
{
"entropy": 1.6816540002822875,
"epoch": 1.4845292955892035,
"grad_norm": 2.597640037536621,
"learning_rate": 3.421901040087177e-06,
"loss": 0.0253,
"mean_token_accuracy": 0.9921609818935394,
"num_tokens": 9611539.0,
"step": 4510
},
{
"entropy": 1.6816054821014403,
"epoch": 1.4861751152073732,
"grad_norm": 0.9247928857803345,
"learning_rate": 3.4014282194198045e-06,
"loss": 0.0328,
"mean_token_accuracy": 0.9923509240150452,
"num_tokens": 9621948.0,
"step": 4515
},
{
"entropy": 1.7215697646141053,
"epoch": 1.4878209348255431,
"grad_norm": 1.9585981369018555,
"learning_rate": 3.381004268127638e-06,
"loss": 0.0158,
"mean_token_accuracy": 0.995547354221344,
"num_tokens": 9632351.0,
"step": 4520
},
{
"entropy": 1.7168682098388672,
"epoch": 1.489466754443713,
"grad_norm": 1.5342450141906738,
"learning_rate": 3.360629337471548e-06,
"loss": 0.0276,
"mean_token_accuracy": 0.9930002510547637,
"num_tokens": 9643093.0,
"step": 4525
},
{
"entropy": 1.6930954933166504,
"epoch": 1.4911125740618827,
"grad_norm": 1.625235915184021,
"learning_rate": 3.340303578349361e-06,
"loss": 0.0175,
"mean_token_accuracy": 0.9939290165901185,
"num_tokens": 9653833.0,
"step": 4530
},
{
"entropy": 1.7007075071334838,
"epoch": 1.4927583936800526,
"grad_norm": 0.7999654412269592,
"learning_rate": 3.3200271412947294e-06,
"loss": 0.0154,
"mean_token_accuracy": 0.9946927070617676,
"num_tokens": 9664334.0,
"step": 4535
},
{
"entropy": 1.670901095867157,
"epoch": 1.4944042132982225,
"grad_norm": 0.8723247647285461,
"learning_rate": 3.2998001764760414e-06,
"loss": 0.0173,
"mean_token_accuracy": 0.9957589745521546,
"num_tokens": 9675067.0,
"step": 4540
},
{
"entropy": 1.7091269135475158,
"epoch": 1.4960500329163924,
"grad_norm": 3.8337857723236084,
"learning_rate": 3.2796228336952663e-06,
"loss": 0.0306,
"mean_token_accuracy": 0.9932047963142395,
"num_tokens": 9685487.0,
"step": 4545
},
{
"entropy": 1.7051352858543396,
"epoch": 1.4976958525345623,
"grad_norm": 2.0214622020721436,
"learning_rate": 3.259495262386888e-06,
"loss": 0.0257,
"mean_token_accuracy": 0.9929771959781647,
"num_tokens": 9696067.0,
"step": 4550
},
{
"entropy": 1.716159415245056,
"epoch": 1.499341672152732,
"grad_norm": 1.7375526428222656,
"learning_rate": 3.2394176116167818e-06,
"loss": 0.0188,
"mean_token_accuracy": 0.9942406177520752,
"num_tokens": 9706729.0,
"step": 4555
},
{
"entropy": 1.712713897228241,
"epoch": 1.500987491770902,
"grad_norm": 3.6430742740631104,
"learning_rate": 3.2193900300810908e-06,
"loss": 0.0252,
"mean_token_accuracy": 0.9919379532337189,
"num_tokens": 9717306.0,
"step": 4560
},
{
"entropy": 1.6966816425323485,
"epoch": 1.5026333113890717,
"grad_norm": 0.11951875686645508,
"learning_rate": 3.1994126661051628e-06,
"loss": 0.0111,
"mean_token_accuracy": 0.9968602657318115,
"num_tokens": 9728013.0,
"step": 4565
},
{
"entropy": 1.7037412881851197,
"epoch": 1.5042791310072416,
"grad_norm": 1.588755488395691,
"learning_rate": 3.179485667642419e-06,
"loss": 0.0284,
"mean_token_accuracy": 0.989536440372467,
"num_tokens": 9738874.0,
"step": 4570
},
{
"entropy": 1.6956771850585937,
"epoch": 1.5059249506254115,
"grad_norm": 0.9138298034667969,
"learning_rate": 3.1596091822732755e-06,
"loss": 0.013,
"mean_token_accuracy": 0.994870787858963,
"num_tokens": 9749398.0,
"step": 4575
},
{
"entropy": 1.7180339097976685,
"epoch": 1.5075707702435812,
"grad_norm": 3.6119987964630127,
"learning_rate": 3.1397833572040414e-06,
"loss": 0.0471,
"mean_token_accuracy": 0.9881613373756408,
"num_tokens": 9760121.0,
"step": 4580
},
{
"entropy": 1.7082261800765992,
"epoch": 1.5092165898617511,
"grad_norm": 3.047520875930786,
"learning_rate": 3.1200083392658464e-06,
"loss": 0.0208,
"mean_token_accuracy": 0.9932926952838897,
"num_tokens": 9770729.0,
"step": 4585
},
{
"entropy": 1.696124267578125,
"epoch": 1.510862409479921,
"grad_norm": 4.276211738586426,
"learning_rate": 3.1002842749135175e-06,
"loss": 0.025,
"mean_token_accuracy": 0.992525440454483,
"num_tokens": 9781733.0,
"step": 4590
},
{
"entropy": 1.689008605480194,
"epoch": 1.5125082290980907,
"grad_norm": 0.7173812389373779,
"learning_rate": 3.0806113102245395e-06,
"loss": 0.027,
"mean_token_accuracy": 0.9932247400283813,
"num_tokens": 9792414.0,
"step": 4595
},
{
"entropy": 1.6812505960464477,
"epoch": 1.5141540487162608,
"grad_norm": 1.7421464920043945,
"learning_rate": 3.0609895908979347e-06,
"loss": 0.0297,
"mean_token_accuracy": 0.9890223205089569,
"num_tokens": 9802799.0,
"step": 4600
},
{
"entropy": 1.7080901622772218,
"epoch": 1.5157998683344305,
"grad_norm": 2.5039777755737305,
"learning_rate": 3.041419262253208e-06,
"loss": 0.0217,
"mean_token_accuracy": 0.993123060464859,
"num_tokens": 9813536.0,
"step": 4605
},
{
"entropy": 1.703446888923645,
"epoch": 1.5174456879526004,
"grad_norm": 1.0789647102355957,
"learning_rate": 3.021900469229253e-06,
"loss": 0.0183,
"mean_token_accuracy": 0.9934683322906495,
"num_tokens": 9824215.0,
"step": 4610
},
{
"entropy": 1.7133625149726868,
"epoch": 1.5190915075707703,
"grad_norm": 1.688781499862671,
"learning_rate": 3.0024333563833007e-06,
"loss": 0.0276,
"mean_token_accuracy": 0.9893021941184997,
"num_tokens": 9834752.0,
"step": 4615
},
{
"entropy": 1.7214069485664367,
"epoch": 1.52073732718894,
"grad_norm": 3.265244722366333,
"learning_rate": 2.983018067889828e-06,
"loss": 0.0352,
"mean_token_accuracy": 0.9897504031658173,
"num_tokens": 9845611.0,
"step": 4620
},
{
"entropy": 1.711961841583252,
"epoch": 1.52238314680711,
"grad_norm": 6.8311333656311035,
"learning_rate": 2.963654747539494e-06,
"loss": 0.0225,
"mean_token_accuracy": 0.9920111656188965,
"num_tokens": 9856980.0,
"step": 4625
},
{
"entropy": 1.7189836978912354,
"epoch": 1.5240289664252797,
"grad_norm": 1.7447693347930908,
"learning_rate": 2.9443435387380936e-06,
"loss": 0.012,
"mean_token_accuracy": 0.9950068473815918,
"num_tokens": 9867771.0,
"step": 4630
},
{
"entropy": 1.7515931606292725,
"epoch": 1.5256747860434496,
"grad_norm": 1.2482762336730957,
"learning_rate": 2.9250845845054586e-06,
"loss": 0.0279,
"mean_token_accuracy": 0.9916983366012573,
"num_tokens": 9878291.0,
"step": 4635
},
{
"entropy": 1.746778416633606,
"epoch": 1.5273206056616195,
"grad_norm": 0.6369831562042236,
"learning_rate": 2.9058780274744426e-06,
"loss": 0.0213,
"mean_token_accuracy": 0.9942344844341278,
"num_tokens": 9888741.0,
"step": 4640
},
{
"entropy": 1.741689419746399,
"epoch": 1.5289664252797892,
"grad_norm": 1.0881776809692383,
"learning_rate": 2.8867240098898297e-06,
"loss": 0.0212,
"mean_token_accuracy": 0.9928946614265441,
"num_tokens": 9899481.0,
"step": 4645
},
{
"entropy": 1.715234887599945,
"epoch": 1.5306122448979593,
"grad_norm": 2.7303154468536377,
"learning_rate": 2.8676226736072975e-06,
"loss": 0.0286,
"mean_token_accuracy": 0.9908973276615143,
"num_tokens": 9910093.0,
"step": 4650
},
{
"entropy": 1.7300845026969909,
"epoch": 1.532258064516129,
"grad_norm": 0.8988632559776306,
"learning_rate": 2.848574160092362e-06,
"loss": 0.0265,
"mean_token_accuracy": 0.9913157403469086,
"num_tokens": 9920988.0,
"step": 4655
},
{
"entropy": 1.7755305886268615,
"epoch": 1.533903884134299,
"grad_norm": 1.2511268854141235,
"learning_rate": 2.829578610419337e-06,
"loss": 0.0183,
"mean_token_accuracy": 0.9922440350055695,
"num_tokens": 9931448.0,
"step": 4660
},
{
"entropy": 1.7372457265853882,
"epoch": 1.5355497037524688,
"grad_norm": 1.8746650218963623,
"learning_rate": 2.810636165270274e-06,
"loss": 0.0448,
"mean_token_accuracy": 0.9905406355857849,
"num_tokens": 9942334.0,
"step": 4665
},
{
"entropy": 1.7496211767196654,
"epoch": 1.5371955233706385,
"grad_norm": 3.1501920223236084,
"learning_rate": 2.7917469649339356e-06,
"loss": 0.0177,
"mean_token_accuracy": 0.9929421722888947,
"num_tokens": 9952925.0,
"step": 4670
},
{
"entropy": 1.769037902355194,
"epoch": 1.5388413429888086,
"grad_norm": 1.5442581176757812,
"learning_rate": 2.7729111493047458e-06,
"loss": 0.0212,
"mean_token_accuracy": 0.9939168155193329,
"num_tokens": 9963276.0,
"step": 4675
},
{
"entropy": 1.7426852703094482,
"epoch": 1.5404871626069783,
"grad_norm": 1.5778064727783203,
"learning_rate": 2.754128857881768e-06,
"loss": 0.022,
"mean_token_accuracy": 0.9926289975643158,
"num_tokens": 9974069.0,
"step": 4680
},
{
"entropy": 1.7343324899673462,
"epoch": 1.5421329822251482,
"grad_norm": 0.8580919504165649,
"learning_rate": 2.735400229767652e-06,
"loss": 0.0208,
"mean_token_accuracy": 0.9934548079967499,
"num_tokens": 9984603.0,
"step": 4685
},
{
"entropy": 1.7297178983688355,
"epoch": 1.543778801843318,
"grad_norm": 1.2886171340942383,
"learning_rate": 2.7167254036676183e-06,
"loss": 0.0261,
"mean_token_accuracy": 0.9902363300323487,
"num_tokens": 9995352.0,
"step": 4690
},
{
"entropy": 1.6793412566184998,
"epoch": 1.5454246214614877,
"grad_norm": 0.6308950185775757,
"learning_rate": 2.698104517888427e-06,
"loss": 0.0088,
"mean_token_accuracy": 0.9957121074199676,
"num_tokens": 10006198.0,
"step": 4695
},
{
"entropy": 1.712625253200531,
"epoch": 1.5470704410796576,
"grad_norm": 4.383965969085693,
"learning_rate": 2.679537710337352e-06,
"loss": 0.0301,
"mean_token_accuracy": 0.990826541185379,
"num_tokens": 10017039.0,
"step": 4700
},
{
"entropy": 1.7353214979171754,
"epoch": 1.5487162606978275,
"grad_norm": 0.49334779381752014,
"learning_rate": 2.6610251185211657e-06,
"loss": 0.029,
"mean_token_accuracy": 0.9952426433563233,
"num_tokens": 10027402.0,
"step": 4705
},
{
"entropy": 1.6911566019058228,
"epoch": 1.5503620803159972,
"grad_norm": 2.034087896347046,
"learning_rate": 2.6425668795451107e-06,
"loss": 0.0281,
"mean_token_accuracy": 0.9905034959316253,
"num_tokens": 10038046.0,
"step": 4710
},
{
"entropy": 1.69894437789917,
"epoch": 1.5520078999341673,
"grad_norm": 2.4216017723083496,
"learning_rate": 2.624163130111891e-06,
"loss": 0.0165,
"mean_token_accuracy": 0.9943257391452789,
"num_tokens": 10048564.0,
"step": 4715
},
{
"entropy": 1.7041826963424682,
"epoch": 1.553653719552337,
"grad_norm": 1.8711634874343872,
"learning_rate": 2.605814006520655e-06,
"loss": 0.0154,
"mean_token_accuracy": 0.9945900738239288,
"num_tokens": 10059017.0,
"step": 4720
},
{
"entropy": 1.6660914540290832,
"epoch": 1.555299539170507,
"grad_norm": 0.6987809538841248,
"learning_rate": 2.587519644666001e-06,
"loss": 0.03,
"mean_token_accuracy": 0.991539990901947,
"num_tokens": 10069792.0,
"step": 4725
},
{
"entropy": 1.6404382228851317,
"epoch": 1.5569453587886768,
"grad_norm": 0.9349547028541565,
"learning_rate": 2.5692801800369406e-06,
"loss": 0.0161,
"mean_token_accuracy": 0.9950057923793793,
"num_tokens": 10080522.0,
"step": 4730
},
{
"entropy": 1.6546894431114196,
"epoch": 1.5585911784068465,
"grad_norm": 2.1536412239074707,
"learning_rate": 2.5510957477159257e-06,
"loss": 0.0295,
"mean_token_accuracy": 0.9922216713428498,
"num_tokens": 10091760.0,
"step": 4735
},
{
"entropy": 1.6477965712547302,
"epoch": 1.5602369980250166,
"grad_norm": 0.56485915184021,
"learning_rate": 2.5329664823778444e-06,
"loss": 0.0114,
"mean_token_accuracy": 0.995609724521637,
"num_tokens": 10102446.0,
"step": 4740
},
{
"entropy": 1.699891173839569,
"epoch": 1.5618828176431863,
"grad_norm": 2.4956679344177246,
"learning_rate": 2.514892518288988e-06,
"loss": 0.0398,
"mean_token_accuracy": 0.9890912711620331,
"num_tokens": 10113234.0,
"step": 4745
},
{
"entropy": 1.6791865229606628,
"epoch": 1.5635286372613562,
"grad_norm": 2.926832914352417,
"learning_rate": 2.4968739893061132e-06,
"loss": 0.0185,
"mean_token_accuracy": 0.9936101496219635,
"num_tokens": 10124034.0,
"step": 4750
},
{
"entropy": 1.6816865921020507,
"epoch": 1.565174456879526,
"grad_norm": 1.028403401374817,
"learning_rate": 2.4789110288754038e-06,
"loss": 0.0326,
"mean_token_accuracy": 0.9936225533485412,
"num_tokens": 10134671.0,
"step": 4755
},
{
"entropy": 1.7139351129531861,
"epoch": 1.5668202764976957,
"grad_norm": 4.163893222808838,
"learning_rate": 2.461003770031504e-06,
"loss": 0.0191,
"mean_token_accuracy": 0.9935286998748779,
"num_tokens": 10145418.0,
"step": 4760
},
{
"entropy": 1.7017668962478638,
"epoch": 1.5684660961158658,
"grad_norm": 2.05292010307312,
"learning_rate": 2.4431523453965266e-06,
"loss": 0.0295,
"mean_token_accuracy": 0.9896788716316223,
"num_tokens": 10156080.0,
"step": 4765
},
{
"entropy": 1.7077699899673462,
"epoch": 1.5701119157340355,
"grad_norm": 1.2193189859390259,
"learning_rate": 2.4253568871790857e-06,
"loss": 0.0426,
"mean_token_accuracy": 0.9874286651611328,
"num_tokens": 10166779.0,
"step": 4770
},
{
"entropy": 1.7373961925506591,
"epoch": 1.5717577353522054,
"grad_norm": 0.09195420891046524,
"learning_rate": 2.407617527173285e-06,
"loss": 0.0234,
"mean_token_accuracy": 0.9929518520832061,
"num_tokens": 10177151.0,
"step": 4775
},
{
"entropy": 1.753466045856476,
"epoch": 1.5734035549703753,
"grad_norm": 1.3904253244400024,
"learning_rate": 2.3899343967577803e-06,
"loss": 0.0269,
"mean_token_accuracy": 0.990022599697113,
"num_tokens": 10187902.0,
"step": 4780
},
{
"entropy": 1.7257111310958861,
"epoch": 1.575049374588545,
"grad_norm": 1.9471051692962646,
"learning_rate": 2.3723076268947777e-06,
"loss": 0.0231,
"mean_token_accuracy": 0.9913094937801361,
"num_tokens": 10198342.0,
"step": 4785
},
{
"entropy": 1.6994914054870605,
"epoch": 1.576695194206715,
"grad_norm": 1.1079803705215454,
"learning_rate": 2.354737348129077e-06,
"loss": 0.0127,
"mean_token_accuracy": 0.994962739944458,
"num_tokens": 10209017.0,
"step": 4790
},
{
"entropy": 1.7078461527824402,
"epoch": 1.5783410138248848,
"grad_norm": 0.8264980316162109,
"learning_rate": 2.337223690587098e-06,
"loss": 0.019,
"mean_token_accuracy": 0.9939144611358642,
"num_tokens": 10219775.0,
"step": 4795
},
{
"entropy": 1.690078580379486,
"epoch": 1.5799868334430547,
"grad_norm": 1.1476809978485107,
"learning_rate": 2.3197667839759307e-06,
"loss": 0.0244,
"mean_token_accuracy": 0.9934214890003205,
"num_tokens": 10230654.0,
"step": 4800
},
{
"entropy": 1.7367305874824523,
"epoch": 1.5816326530612246,
"grad_norm": 0.9815186262130737,
"learning_rate": 2.302366757582355e-06,
"loss": 0.0142,
"mean_token_accuracy": 0.9940285742282867,
"num_tokens": 10241163.0,
"step": 4805
},
{
"entropy": 1.7170762538909912,
"epoch": 1.5832784726793943,
"grad_norm": 2.2614142894744873,
"learning_rate": 2.285023740271893e-06,
"loss": 0.0193,
"mean_token_accuracy": 0.994717925786972,
"num_tokens": 10252132.0,
"step": 4810
},
{
"entropy": 1.6848518610000611,
"epoch": 1.5849242922975642,
"grad_norm": 0.6454903483390808,
"learning_rate": 2.267737860487865e-06,
"loss": 0.0182,
"mean_token_accuracy": 0.9939591407775878,
"num_tokens": 10263052.0,
"step": 4815
},
{
"entropy": 1.7132264494895935,
"epoch": 1.586570111915734,
"grad_norm": 1.0455392599105835,
"learning_rate": 2.2505092462504153e-06,
"loss": 0.0134,
"mean_token_accuracy": 0.9961919903755188,
"num_tokens": 10273616.0,
"step": 4820
},
{
"entropy": 1.7440693497657775,
"epoch": 1.5882159315339037,
"grad_norm": 1.3470888137817383,
"learning_rate": 2.2333380251555826e-06,
"loss": 0.0279,
"mean_token_accuracy": 0.9919908702373504,
"num_tokens": 10284093.0,
"step": 4825
},
{
"entropy": 1.7089533686637879,
"epoch": 1.5898617511520738,
"grad_norm": 1.1258236169815063,
"learning_rate": 2.2162243243743485e-06,
"loss": 0.0233,
"mean_token_accuracy": 0.9929437875747681,
"num_tokens": 10294993.0,
"step": 4830
},
{
"entropy": 1.677238130569458,
"epoch": 1.5915075707702435,
"grad_norm": 3.9911301136016846,
"learning_rate": 2.1991682706516935e-06,
"loss": 0.0294,
"mean_token_accuracy": 0.9925793290138245,
"num_tokens": 10305763.0,
"step": 4835
},
{
"entropy": 1.679075539112091,
"epoch": 1.5931533903884134,
"grad_norm": 2.0645158290863037,
"learning_rate": 2.1821699903056627e-06,
"loss": 0.028,
"mean_token_accuracy": 0.9916234374046325,
"num_tokens": 10316288.0,
"step": 4840
},
{
"entropy": 1.6844815373420716,
"epoch": 1.5947992100065833,
"grad_norm": 0.7942723631858826,
"learning_rate": 2.1652296092264324e-06,
"loss": 0.0288,
"mean_token_accuracy": 0.9914738476276398,
"num_tokens": 10327136.0,
"step": 4845
},
{
"entropy": 1.681308126449585,
"epoch": 1.596445029624753,
"grad_norm": 2.3584067821502686,
"learning_rate": 2.148347252875368e-06,
"loss": 0.0106,
"mean_token_accuracy": 0.9962167620658875,
"num_tokens": 10337795.0,
"step": 4850
},
{
"entropy": 1.6970802187919616,
"epoch": 1.598090849242923,
"grad_norm": 0.7867595553398132,
"learning_rate": 2.1315230462840985e-06,
"loss": 0.0253,
"mean_token_accuracy": 0.9912108719348908,
"num_tokens": 10348719.0,
"step": 4855
},
{
"entropy": 1.706315529346466,
"epoch": 1.5997366688610928,
"grad_norm": 1.8649603128433228,
"learning_rate": 2.114757114053605e-06,
"loss": 0.0211,
"mean_token_accuracy": 0.992872542142868,
"num_tokens": 10359442.0,
"step": 4860
},
{
"epoch": 1.6010533245556287,
"eval_entropy": 1.6943830250834868,
"eval_loss": 0.048175811767578125,
"eval_mean_token_accuracy": 0.9870549200333505,
"eval_num_tokens": 10368037.0,
"eval_runtime": 196.4348,
"eval_samples_per_second": 42.391,
"eval_steps_per_second": 7.066,
"step": 4864
},
{
"entropy": 1.7008237361907959,
"epoch": 1.6013824884792627,
"grad_norm": 0.8595287799835205,
"learning_rate": 2.098049580353273e-06,
"loss": 0.0191,
"mean_token_accuracy": 0.9942748367786407,
"num_tokens": 10370122.0,
"step": 4865
},
{
"entropy": 1.697910237312317,
"epoch": 1.6030283080974326,
"grad_norm": 1.504879355430603,
"learning_rate": 2.08140056891999e-06,
"loss": 0.0133,
"mean_token_accuracy": 0.9958468794822692,
"num_tokens": 10380595.0,
"step": 4870
},
{
"entropy": 1.6834197640419006,
"epoch": 1.6046741277156022,
"grad_norm": 6.986445903778076,
"learning_rate": 2.0648102030572225e-06,
"loss": 0.03,
"mean_token_accuracy": 0.9888522446155548,
"num_tokens": 10391431.0,
"step": 4875
},
{
"entropy": 1.6846235394477844,
"epoch": 1.6063199473337724,
"grad_norm": 4.83704137802124,
"learning_rate": 2.048278605634113e-06,
"loss": 0.0198,
"mean_token_accuracy": 0.9933947384357452,
"num_tokens": 10402189.0,
"step": 4880
},
{
"entropy": 1.6738305926322936,
"epoch": 1.607965766951942,
"grad_norm": 1.9007184505462646,
"learning_rate": 2.0318058990845467e-06,
"loss": 0.0189,
"mean_token_accuracy": 0.9959712147712707,
"num_tokens": 10412659.0,
"step": 4885
},
{
"entropy": 1.668698501586914,
"epoch": 1.609611586570112,
"grad_norm": 5.049774169921875,
"learning_rate": 2.0153922054062758e-06,
"loss": 0.0212,
"mean_token_accuracy": 0.9936519682407379,
"num_tokens": 10423408.0,
"step": 4890
},
{
"entropy": 1.6653715133666993,
"epoch": 1.6112574061882818,
"grad_norm": 2.830352306365967,
"learning_rate": 1.999037646159989e-06,
"loss": 0.0142,
"mean_token_accuracy": 0.9954032123088836,
"num_tokens": 10434264.0,
"step": 4895
},
{
"entropy": 1.6867196083068847,
"epoch": 1.6129032258064515,
"grad_norm": 2.5599348545074463,
"learning_rate": 1.9827423424684267e-06,
"loss": 0.0332,
"mean_token_accuracy": 0.9906599521636963,
"num_tokens": 10445083.0,
"step": 4900
},
{
"entropy": 1.6548929691314698,
"epoch": 1.6145490454246214,
"grad_norm": 1.4400912523269653,
"learning_rate": 1.966506415015477e-06,
"loss": 0.0193,
"mean_token_accuracy": 0.994045513868332,
"num_tokens": 10455748.0,
"step": 4905
},
{
"entropy": 1.704959547519684,
"epoch": 1.6161948650427913,
"grad_norm": 1.0404367446899414,
"learning_rate": 1.9503299840452927e-06,
"loss": 0.0202,
"mean_token_accuracy": 0.9917968034744262,
"num_tokens": 10466258.0,
"step": 4910
},
{
"entropy": 1.7084056258201599,
"epoch": 1.6178406846609612,
"grad_norm": 1.2847148180007935,
"learning_rate": 1.9342131693613763e-06,
"loss": 0.0245,
"mean_token_accuracy": 0.9919559478759765,
"num_tokens": 10476806.0,
"step": 4915
},
{
"entropy": 1.6556385517120362,
"epoch": 1.619486504279131,
"grad_norm": 2.1901090145111084,
"learning_rate": 1.9181560903257234e-06,
"loss": 0.0242,
"mean_token_accuracy": 0.992319130897522,
"num_tokens": 10487881.0,
"step": 4920
},
{
"entropy": 1.6394746541976928,
"epoch": 1.6211323238973008,
"grad_norm": 1.6991859674453735,
"learning_rate": 1.9021588658579249e-06,
"loss": 0.0259,
"mean_token_accuracy": 0.9934229254722595,
"num_tokens": 10498864.0,
"step": 4925
},
{
"entropy": 1.7088184237480164,
"epoch": 1.6227781435154707,
"grad_norm": 1.4522404670715332,
"learning_rate": 1.8862216144342692e-06,
"loss": 0.0176,
"mean_token_accuracy": 0.9922803401947021,
"num_tokens": 10509379.0,
"step": 4930
},
{
"entropy": 1.6745522737503051,
"epoch": 1.6244239631336406,
"grad_norm": 0.5447269678115845,
"learning_rate": 1.870344454086901e-06,
"loss": 0.0315,
"mean_token_accuracy": 0.9911853075027466,
"num_tokens": 10519980.0,
"step": 4935
},
{
"entropy": 1.6869741559028626,
"epoch": 1.6260697827518102,
"grad_norm": 1.314532995223999,
"learning_rate": 1.8545275024029141e-06,
"loss": 0.0194,
"mean_token_accuracy": 0.9942934930324554,
"num_tokens": 10530726.0,
"step": 4940
},
{
"entropy": 1.6622373104095458,
"epoch": 1.6277156023699804,
"grad_norm": 1.3033603429794312,
"learning_rate": 1.838770876523498e-06,
"loss": 0.0211,
"mean_token_accuracy": 0.9951237618923188,
"num_tokens": 10541525.0,
"step": 4945
},
{
"entropy": 1.6855845212936402,
"epoch": 1.62936142198815,
"grad_norm": 2.0838241577148438,
"learning_rate": 1.8230746931430642e-06,
"loss": 0.0201,
"mean_token_accuracy": 0.9907946407794952,
"num_tokens": 10552191.0,
"step": 4950
},
{
"entropy": 1.6786072969436645,
"epoch": 1.63100724160632,
"grad_norm": 0.6673282384872437,
"learning_rate": 1.807439068508392e-06,
"loss": 0.0107,
"mean_token_accuracy": 0.9967107653617859,
"num_tokens": 10562922.0,
"step": 4955
},
{
"entropy": 1.6981304287910461,
"epoch": 1.6326530612244898,
"grad_norm": 0.6075028777122498,
"learning_rate": 1.7918641184177444e-06,
"loss": 0.0183,
"mean_token_accuracy": 0.992760181427002,
"num_tokens": 10573352.0,
"step": 4960
},
{
"entropy": 1.6855650663375854,
"epoch": 1.6342988808426595,
"grad_norm": 1.076545000076294,
"learning_rate": 1.7763499582200405e-06,
"loss": 0.023,
"mean_token_accuracy": 0.9909692645072937,
"num_tokens": 10584337.0,
"step": 4965
},
{
"entropy": 1.6684273719787597,
"epoch": 1.6359447004608296,
"grad_norm": 0.9483628869056702,
"learning_rate": 1.7608967028139767e-06,
"loss": 0.025,
"mean_token_accuracy": 0.9940790653228759,
"num_tokens": 10595236.0,
"step": 4970
},
{
"entropy": 1.6795395970344544,
"epoch": 1.6375905200789993,
"grad_norm": 2.4686009883880615,
"learning_rate": 1.7455044666471875e-06,
"loss": 0.0407,
"mean_token_accuracy": 0.9889774978160858,
"num_tokens": 10605904.0,
"step": 4975
},
{
"entropy": 1.7077094316482544,
"epoch": 1.6392363396971692,
"grad_norm": 1.020444393157959,
"learning_rate": 1.7301733637153994e-06,
"loss": 0.0318,
"mean_token_accuracy": 0.9912243783473969,
"num_tokens": 10616311.0,
"step": 4980
},
{
"entropy": 1.6756536841392518,
"epoch": 1.640882159315339,
"grad_norm": 4.183596134185791,
"learning_rate": 1.7149035075615795e-06,
"loss": 0.0325,
"mean_token_accuracy": 0.9907270252704621,
"num_tokens": 10627079.0,
"step": 4985
},
{
"entropy": 1.6901473999023438,
"epoch": 1.6425279789335088,
"grad_norm": 1.742541790008545,
"learning_rate": 1.6996950112750964e-06,
"loss": 0.0144,
"mean_token_accuracy": 0.9967920780181885,
"num_tokens": 10637698.0,
"step": 4990
},
{
"entropy": 1.665801465511322,
"epoch": 1.6441737985516789,
"grad_norm": 2.1923062801361084,
"learning_rate": 1.6845479874908865e-06,
"loss": 0.0236,
"mean_token_accuracy": 0.9916431307792664,
"num_tokens": 10648267.0,
"step": 4995
},
{
"entropy": 1.6393486261367798,
"epoch": 1.6458196181698486,
"grad_norm": 0.6840113401412964,
"learning_rate": 1.6694625483886195e-06,
"loss": 0.0164,
"mean_token_accuracy": 0.9945251405239105,
"num_tokens": 10659109.0,
"step": 5000
},
{
"entropy": 1.6628060221672059,
"epoch": 1.6474654377880185,
"grad_norm": 2.219745397567749,
"learning_rate": 1.6544388056918614e-06,
"loss": 0.0254,
"mean_token_accuracy": 0.9924133539199829,
"num_tokens": 10669828.0,
"step": 5005
},
{
"entropy": 1.6695847511291504,
"epoch": 1.6491112574061884,
"grad_norm": 1.5060310363769531,
"learning_rate": 1.6394768706672547e-06,
"loss": 0.0244,
"mean_token_accuracy": 0.9916559994220734,
"num_tokens": 10680094.0,
"step": 5010
},
{
"entropy": 1.659652292728424,
"epoch": 1.650757077024358,
"grad_norm": 3.3251304626464844,
"learning_rate": 1.6245768541236894e-06,
"loss": 0.0202,
"mean_token_accuracy": 0.9898216426372528,
"num_tokens": 10690770.0,
"step": 5015
},
{
"entropy": 1.6149708151817321,
"epoch": 1.652402896642528,
"grad_norm": 1.9098771810531616,
"learning_rate": 1.6097388664114833e-06,
"loss": 0.0147,
"mean_token_accuracy": 0.9929858863353729,
"num_tokens": 10702024.0,
"step": 5020
},
{
"entropy": 1.6477410674095154,
"epoch": 1.6540487162606978,
"grad_norm": 1.96040940284729,
"learning_rate": 1.5949630174215647e-06,
"loss": 0.0144,
"mean_token_accuracy": 0.9963162183761597,
"num_tokens": 10712833.0,
"step": 5025
},
{
"entropy": 1.6466076374053955,
"epoch": 1.6556945358788677,
"grad_norm": 0.9714366793632507,
"learning_rate": 1.580249416584666e-06,
"loss": 0.0365,
"mean_token_accuracy": 0.9881292939186096,
"num_tokens": 10723440.0,
"step": 5030
},
{
"entropy": 1.669557237625122,
"epoch": 1.6573403554970376,
"grad_norm": 1.7277604341506958,
"learning_rate": 1.5655981728704973e-06,
"loss": 0.0239,
"mean_token_accuracy": 0.991591501235962,
"num_tokens": 10733955.0,
"step": 5035
},
{
"entropy": 1.6399479150772094,
"epoch": 1.6589861751152073,
"grad_norm": 5.28700590133667,
"learning_rate": 1.5510093947869508e-06,
"loss": 0.0201,
"mean_token_accuracy": 0.9937320530414582,
"num_tokens": 10744942.0,
"step": 5040
},
{
"entropy": 1.688657033443451,
"epoch": 1.6606319947333772,
"grad_norm": 6.3327813148498535,
"learning_rate": 1.536483190379302e-06,
"loss": 0.0227,
"mean_token_accuracy": 0.9922419965267182,
"num_tokens": 10755594.0,
"step": 5045
},
{
"entropy": 1.6918665885925293,
"epoch": 1.662277814351547,
"grad_norm": 0.1568760871887207,
"learning_rate": 1.522019667229393e-06,
"loss": 0.009,
"mean_token_accuracy": 0.9979844152927398,
"num_tokens": 10766398.0,
"step": 5050
},
{
"entropy": 1.6314533591270446,
"epoch": 1.6639236339697168,
"grad_norm": 1.880570411682129,
"learning_rate": 1.5076189324548506e-06,
"loss": 0.0203,
"mean_token_accuracy": 0.9929390490055084,
"num_tokens": 10777255.0,
"step": 5055
},
{
"entropy": 1.6015793800354003,
"epoch": 1.6655694535878869,
"grad_norm": 1.6532872915267944,
"learning_rate": 1.493281092708283e-06,
"loss": 0.0153,
"mean_token_accuracy": 0.9948068380355835,
"num_tokens": 10788343.0,
"step": 5060
},
{
"entropy": 1.6586069226264955,
"epoch": 1.6672152732060566,
"grad_norm": 1.2856965065002441,
"learning_rate": 1.479006254176505e-06,
"loss": 0.0203,
"mean_token_accuracy": 0.992956918478012,
"num_tokens": 10799130.0,
"step": 5065
},
{
"entropy": 1.6501345872879027,
"epoch": 1.6688610928242265,
"grad_norm": 2.5434141159057617,
"learning_rate": 1.4647945225797244e-06,
"loss": 0.0261,
"mean_token_accuracy": 0.9912129878997803,
"num_tokens": 10809881.0,
"step": 5070
},
{
"entropy": 1.614508295059204,
"epoch": 1.6705069124423964,
"grad_norm": 0.4630468189716339,
"learning_rate": 1.4506460031707903e-06,
"loss": 0.0148,
"mean_token_accuracy": 0.9954249680042266,
"num_tokens": 10820794.0,
"step": 5075
},
{
"entropy": 1.645851767063141,
"epoch": 1.672152732060566,
"grad_norm": 2.122525930404663,
"learning_rate": 1.4365608007343922e-06,
"loss": 0.026,
"mean_token_accuracy": 0.9927963316440582,
"num_tokens": 10831234.0,
"step": 5080
},
{
"entropy": 1.6439926147460937,
"epoch": 1.6737985516787361,
"grad_norm": 0.7459424138069153,
"learning_rate": 1.4225390195862932e-06,
"loss": 0.0241,
"mean_token_accuracy": 0.9936447679996491,
"num_tokens": 10842096.0,
"step": 5085
},
{
"entropy": 1.6597126364707946,
"epoch": 1.6754443712969058,
"grad_norm": 1.7774500846862793,
"learning_rate": 1.4085807635725491e-06,
"loss": 0.0235,
"mean_token_accuracy": 0.9914094746112824,
"num_tokens": 10852586.0,
"step": 5090
},
{
"entropy": 1.7044395089149476,
"epoch": 1.6770901909150757,
"grad_norm": 1.1517224311828613,
"learning_rate": 1.3946861360687548e-06,
"loss": 0.0174,
"mean_token_accuracy": 0.9941137075424195,
"num_tokens": 10862934.0,
"step": 5095
},
{
"entropy": 1.6369341015815735,
"epoch": 1.6787360105332456,
"grad_norm": 0.2626267671585083,
"learning_rate": 1.380855239979264e-06,
"loss": 0.017,
"mean_token_accuracy": 0.99591423869133,
"num_tokens": 10873852.0,
"step": 5100
},
{
"entropy": 1.6440617442131042,
"epoch": 1.6803818301514153,
"grad_norm": 2.385711193084717,
"learning_rate": 1.3670881777364276e-06,
"loss": 0.0389,
"mean_token_accuracy": 0.9895196318626404,
"num_tokens": 10884490.0,
"step": 5105
},
{
"entropy": 1.6631293177604676,
"epoch": 1.6820276497695854,
"grad_norm": 2.767446994781494,
"learning_rate": 1.3533850512998515e-06,
"loss": 0.043,
"mean_token_accuracy": 0.9870490610599518,
"num_tokens": 10894958.0,
"step": 5110
},
{
"entropy": 1.671304202079773,
"epoch": 1.683673469387755,
"grad_norm": 1.706010103225708,
"learning_rate": 1.339745962155613e-06,
"loss": 0.0214,
"mean_token_accuracy": 0.9934303402900696,
"num_tokens": 10905373.0,
"step": 5115
},
{
"entropy": 1.652925717830658,
"epoch": 1.685319289005925,
"grad_norm": 1.8445219993591309,
"learning_rate": 1.3261710113155436e-06,
"loss": 0.0286,
"mean_token_accuracy": 0.9917166590690613,
"num_tokens": 10916144.0,
"step": 5120
},
{
"entropy": 1.6853961229324341,
"epoch": 1.6869651086240949,
"grad_norm": 0.4994262754917145,
"learning_rate": 1.3126602993164505e-06,
"loss": 0.0155,
"mean_token_accuracy": 0.9949258744716645,
"num_tokens": 10926960.0,
"step": 5125
},
{
"entropy": 1.6451457619667054,
"epoch": 1.6886109282422646,
"grad_norm": 0.9345157742500305,
"learning_rate": 1.2992139262193893e-06,
"loss": 0.0185,
"mean_token_accuracy": 0.9947284281253814,
"num_tokens": 10937405.0,
"step": 5130
},
{
"entropy": 1.624656081199646,
"epoch": 1.6902567478604344,
"grad_norm": 1.3270047903060913,
"learning_rate": 1.2858319916089156e-06,
"loss": 0.0168,
"mean_token_accuracy": 0.99341681599617,
"num_tokens": 10948172.0,
"step": 5135
},
{
"entropy": 1.659364914894104,
"epoch": 1.6919025674786043,
"grad_norm": 0.4350810647010803,
"learning_rate": 1.2725145945923588e-06,
"loss": 0.0183,
"mean_token_accuracy": 0.99389089345932,
"num_tokens": 10958763.0,
"step": 5140
},
{
"entropy": 1.6685426354408264,
"epoch": 1.6935483870967742,
"grad_norm": 1.704720377922058,
"learning_rate": 1.2592618337990647e-06,
"loss": 0.0285,
"mean_token_accuracy": 0.9913337886333465,
"num_tokens": 10969305.0,
"step": 5145
},
{
"entropy": 1.6044569730758667,
"epoch": 1.6951942067149441,
"grad_norm": 0.8705942034721375,
"learning_rate": 1.2460738073796929e-06,
"loss": 0.0238,
"mean_token_accuracy": 0.9927083373069763,
"num_tokens": 10980236.0,
"step": 5150
},
{
"entropy": 1.685863721370697,
"epoch": 1.6968400263331138,
"grad_norm": 0.5871350765228271,
"learning_rate": 1.2329506130054703e-06,
"loss": 0.0187,
"mean_token_accuracy": 0.9933398306369782,
"num_tokens": 10990582.0,
"step": 5155
},
{
"entropy": 1.6742502808570863,
"epoch": 1.6984858459512837,
"grad_norm": 2.4911105632781982,
"learning_rate": 1.219892347867474e-06,
"loss": 0.0298,
"mean_token_accuracy": 0.9904028534889221,
"num_tokens": 11000966.0,
"step": 5160
},
{
"entropy": 1.6669535398483277,
"epoch": 1.7001316655694536,
"grad_norm": 1.9511528015136719,
"learning_rate": 1.2068991086759175e-06,
"loss": 0.0377,
"mean_token_accuracy": 0.9925346195697784,
"num_tokens": 11011630.0,
"step": 5165
},
{
"entropy": 1.6653909087181091,
"epoch": 1.7017774851876233,
"grad_norm": 0.7113050222396851,
"learning_rate": 1.1939709916594222e-06,
"loss": 0.0297,
"mean_token_accuracy": 0.9918692409992218,
"num_tokens": 11022348.0,
"step": 5170
},
{
"entropy": 1.6599775552749634,
"epoch": 1.7034233048057934,
"grad_norm": 2.2692055702209473,
"learning_rate": 1.1811080925643125e-06,
"loss": 0.0344,
"mean_token_accuracy": 0.9922800958156586,
"num_tokens": 11032982.0,
"step": 5175
},
{
"entropy": 1.6723967909812927,
"epoch": 1.705069124423963,
"grad_norm": 1.5152289867401123,
"learning_rate": 1.1683105066539068e-06,
"loss": 0.0172,
"mean_token_accuracy": 0.9953593909740448,
"num_tokens": 11043325.0,
"step": 5180
},
{
"entropy": 1.6335657358169555,
"epoch": 1.706714944042133,
"grad_norm": 1.356180191040039,
"learning_rate": 1.1555783287078116e-06,
"loss": 0.0215,
"mean_token_accuracy": 0.9901393771171569,
"num_tokens": 11054285.0,
"step": 5185
},
{
"entropy": 1.6311080694198608,
"epoch": 1.7083607636603029,
"grad_norm": 1.751895785331726,
"learning_rate": 1.142911653021217e-06,
"loss": 0.0163,
"mean_token_accuracy": 0.994689530134201,
"num_tokens": 11064881.0,
"step": 5190
},
{
"entropy": 1.6046368837356568,
"epoch": 1.7100065832784725,
"grad_norm": 0.38154032826423645,
"learning_rate": 1.1303105734041996e-06,
"loss": 0.0181,
"mean_token_accuracy": 0.9927674353122711,
"num_tokens": 11076136.0,
"step": 5195
},
{
"entropy": 1.6612093567848205,
"epoch": 1.7116524028966427,
"grad_norm": 0.979333221912384,
"learning_rate": 1.1177751831810279e-06,
"loss": 0.0335,
"mean_token_accuracy": 0.9923636972904205,
"num_tokens": 11086778.0,
"step": 5200
},
{
"entropy": 1.6807520270347596,
"epoch": 1.7132982225148123,
"grad_norm": 1.7828707695007324,
"learning_rate": 1.1053055751894726e-06,
"loss": 0.0277,
"mean_token_accuracy": 0.993100905418396,
"num_tokens": 11097370.0,
"step": 5205
},
{
"entropy": 1.6403346300125121,
"epoch": 1.7149440421329822,
"grad_norm": 2.5002477169036865,
"learning_rate": 1.0929018417801129e-06,
"loss": 0.017,
"mean_token_accuracy": 0.9920848309993744,
"num_tokens": 11108166.0,
"step": 5210
},
{
"entropy": 1.6517832636833192,
"epoch": 1.7165898617511521,
"grad_norm": 1.3457746505737305,
"learning_rate": 1.0805640748156675e-06,
"loss": 0.0213,
"mean_token_accuracy": 0.9931346833705902,
"num_tokens": 11118902.0,
"step": 5215
},
{
"entropy": 1.6854795217514038,
"epoch": 1.7182356813693218,
"grad_norm": 0.07055312395095825,
"learning_rate": 1.0682923656702948e-06,
"loss": 0.0211,
"mean_token_accuracy": 0.9933703184127808,
"num_tokens": 11129409.0,
"step": 5220
},
{
"entropy": 1.7007460117340087,
"epoch": 1.719881500987492,
"grad_norm": 1.2991859912872314,
"learning_rate": 1.0560868052289253e-06,
"loss": 0.0244,
"mean_token_accuracy": 0.9902635872364044,
"num_tokens": 11140079.0,
"step": 5225
},
{
"entropy": 1.6376017451286315,
"epoch": 1.7215273206056616,
"grad_norm": 0.83013516664505,
"learning_rate": 1.0439474838865981e-06,
"loss": 0.013,
"mean_token_accuracy": 0.9975570976734162,
"num_tokens": 11150683.0,
"step": 5230
},
{
"entropy": 1.6345697283744811,
"epoch": 1.7231731402238315,
"grad_norm": 3.94348406791687,
"learning_rate": 1.031874491547773e-06,
"loss": 0.0189,
"mean_token_accuracy": 0.9938337445259094,
"num_tokens": 11161334.0,
"step": 5235
},
{
"entropy": 1.648432207107544,
"epoch": 1.7248189598420014,
"grad_norm": 1.0848654508590698,
"learning_rate": 1.0198679176256742e-06,
"loss": 0.0287,
"mean_token_accuracy": 0.9918507993221283,
"num_tokens": 11171950.0,
"step": 5240
},
{
"entropy": 1.660215425491333,
"epoch": 1.726464779460171,
"grad_norm": 1.0720781087875366,
"learning_rate": 1.0079278510416313e-06,
"loss": 0.0297,
"mean_token_accuracy": 0.9905134677886963,
"num_tokens": 11182536.0,
"step": 5245
},
{
"entropy": 1.6251054883003235,
"epoch": 1.728110599078341,
"grad_norm": 2.8627305030822754,
"learning_rate": 9.960543802244195e-07,
"loss": 0.0214,
"mean_token_accuracy": 0.991985023021698,
"num_tokens": 11193429.0,
"step": 5250
},
{
"entropy": 1.668004846572876,
"epoch": 1.7297564186965109,
"grad_norm": 0.5310274958610535,
"learning_rate": 9.842475931095895e-07,
"loss": 0.0398,
"mean_token_accuracy": 0.9901407361030579,
"num_tokens": 11204189.0,
"step": 5255
},
{
"entropy": 1.6230967044830322,
"epoch": 1.7314022383146808,
"grad_norm": 1.8132059574127197,
"learning_rate": 9.725075771388449e-07,
"loss": 0.0181,
"mean_token_accuracy": 0.9925452589988708,
"num_tokens": 11214880.0,
"step": 5260
},
{
"entropy": 1.629646909236908,
"epoch": 1.7330480579328507,
"grad_norm": 0.9202048182487488,
"learning_rate": 9.60834419259369e-07,
"loss": 0.0242,
"mean_token_accuracy": 0.9919378876686096,
"num_tokens": 11225689.0,
"step": 5265
},
{
"entropy": 1.6248352885246278,
"epoch": 1.7346938775510203,
"grad_norm": 0.7040166258811951,
"learning_rate": 9.492282059231917e-07,
"loss": 0.0173,
"mean_token_accuracy": 0.9930191397666931,
"num_tokens": 11236662.0,
"step": 5270
},
{
"entropy": 1.6555840373039246,
"epoch": 1.7363396971691902,
"grad_norm": 0.31907936930656433,
"learning_rate": 9.376890230865487e-07,
"loss": 0.0105,
"mean_token_accuracy": 0.9968835175037384,
"num_tokens": 11247227.0,
"step": 5275
},
{
"entropy": 1.6610607862472535,
"epoch": 1.7379855167873601,
"grad_norm": 1.0085439682006836,
"learning_rate": 9.262169562092483e-07,
"loss": 0.0167,
"mean_token_accuracy": 0.9929597020149231,
"num_tokens": 11258079.0,
"step": 5280
},
{
"entropy": 1.6653624534606934,
"epoch": 1.7396313364055298,
"grad_norm": 1.6425352096557617,
"learning_rate": 9.148120902540281e-07,
"loss": 0.0244,
"mean_token_accuracy": 0.9898296535015106,
"num_tokens": 11268656.0,
"step": 5285
},
{
"entropy": 1.6512943148612975,
"epoch": 1.7412771560237,
"grad_norm": 1.9848991632461548,
"learning_rate": 9.034745096859332e-07,
"loss": 0.0165,
"mean_token_accuracy": 0.9939022302627564,
"num_tokens": 11279189.0,
"step": 5290
},
{
"entropy": 1.633478820323944,
"epoch": 1.7429229756418696,
"grad_norm": 1.188881278038025,
"learning_rate": 8.922042984716972e-07,
"loss": 0.0211,
"mean_token_accuracy": 0.9940697014331817,
"num_tokens": 11289941.0,
"step": 5295
},
{
"entropy": 1.6428375720977784,
"epoch": 1.7445687952600395,
"grad_norm": 1.3060122728347778,
"learning_rate": 8.810015400790994e-07,
"loss": 0.0146,
"mean_token_accuracy": 0.9937053561210633,
"num_tokens": 11300919.0,
"step": 5300
},
{
"entropy": 1.644476592540741,
"epoch": 1.7462146148782094,
"grad_norm": 4.126039981842041,
"learning_rate": 8.69866317476371e-07,
"loss": 0.0436,
"mean_token_accuracy": 0.9879814326763153,
"num_tokens": 11311626.0,
"step": 5305
},
{
"entropy": 1.6524176239967345,
"epoch": 1.747860434496379,
"grad_norm": 0.9086925387382507,
"learning_rate": 8.587987131315656e-07,
"loss": 0.0384,
"mean_token_accuracy": 0.9912387907505036,
"num_tokens": 11322432.0,
"step": 5310
},
{
"entropy": 1.6595850586891174,
"epoch": 1.7495062541145492,
"grad_norm": 1.3107917308807373,
"learning_rate": 8.477988090119515e-07,
"loss": 0.0209,
"mean_token_accuracy": 0.9909784734249115,
"num_tokens": 11333086.0,
"step": 5315
},
{
"entropy": 1.6489954590797424,
"epoch": 1.7511520737327189,
"grad_norm": 0.6780989766120911,
"learning_rate": 8.36866686583404e-07,
"loss": 0.013,
"mean_token_accuracy": 0.9960228025913238,
"num_tokens": 11343698.0,
"step": 5320
},
{
"entropy": 1.6577156782150269,
"epoch": 1.7527978933508888,
"grad_norm": 2.5030269622802734,
"learning_rate": 8.260024268098121e-07,
"loss": 0.0407,
"mean_token_accuracy": 0.9897329092025757,
"num_tokens": 11354384.0,
"step": 5325
},
{
"entropy": 1.6797374844551087,
"epoch": 1.7544437129690587,
"grad_norm": 1.1785377264022827,
"learning_rate": 8.152061101524578e-07,
"loss": 0.0204,
"mean_token_accuracy": 0.992442113161087,
"num_tokens": 11364888.0,
"step": 5330
},
{
"entropy": 1.6517166137695312,
"epoch": 1.7560895325872283,
"grad_norm": 2.3063104152679443,
"learning_rate": 8.044778165694434e-07,
"loss": 0.0249,
"mean_token_accuracy": 0.9921338021755218,
"num_tokens": 11375641.0,
"step": 5335
},
{
"entropy": 1.644195032119751,
"epoch": 1.7577353522053984,
"grad_norm": 2.3999085426330566,
"learning_rate": 7.93817625515082e-07,
"loss": 0.0175,
"mean_token_accuracy": 0.9942924916744232,
"num_tokens": 11386312.0,
"step": 5340
},
{
"entropy": 1.6763506770133971,
"epoch": 1.7593811718235681,
"grad_norm": 2.3939199447631836,
"learning_rate": 7.832256159393181e-07,
"loss": 0.0242,
"mean_token_accuracy": 0.990136843919754,
"num_tokens": 11396648.0,
"step": 5345
},
{
"entropy": 1.7076772809028626,
"epoch": 1.761026991441738,
"grad_norm": 4.3696184158325195,
"learning_rate": 7.727018662871432e-07,
"loss": 0.0236,
"mean_token_accuracy": 0.9929444313049316,
"num_tokens": 11407191.0,
"step": 5350
},
{
"entropy": 1.6799395561218262,
"epoch": 1.762672811059908,
"grad_norm": 1.7837049961090088,
"learning_rate": 7.62246454498009e-07,
"loss": 0.0212,
"mean_token_accuracy": 0.993122935295105,
"num_tokens": 11417870.0,
"step": 5355
},
{
"entropy": 1.6892127633094787,
"epoch": 1.7643186306780776,
"grad_norm": 1.9029659032821655,
"learning_rate": 7.518594580052519e-07,
"loss": 0.0253,
"mean_token_accuracy": 0.9921809673309326,
"num_tokens": 11428181.0,
"step": 5360
},
{
"entropy": 1.6493794798851014,
"epoch": 1.7659644502962475,
"grad_norm": 0.7186495065689087,
"learning_rate": 7.415409537355222e-07,
"loss": 0.016,
"mean_token_accuracy": 0.9955087423324585,
"num_tokens": 11439027.0,
"step": 5365
},
{
"entropy": 1.668405246734619,
"epoch": 1.7676102699144174,
"grad_norm": 0.64583420753479,
"learning_rate": 7.312910181082178e-07,
"loss": 0.0149,
"mean_token_accuracy": 0.9964608550071716,
"num_tokens": 11449456.0,
"step": 5370
},
{
"entropy": 1.6470329999923705,
"epoch": 1.7692560895325873,
"grad_norm": 0.6577222347259521,
"learning_rate": 7.211097270349065e-07,
"loss": 0.0128,
"mean_token_accuracy": 0.9961907029151916,
"num_tokens": 11459922.0,
"step": 5375
},
{
"entropy": 1.6427839875221253,
"epoch": 1.7709019091507572,
"grad_norm": 1.2375580072402954,
"learning_rate": 7.109971559187767e-07,
"loss": 0.0251,
"mean_token_accuracy": 0.9925231218338013,
"num_tokens": 11470923.0,
"step": 5380
},
{
"entropy": 1.6566937565803528,
"epoch": 1.7725477287689269,
"grad_norm": 2.0263426303863525,
"learning_rate": 7.00953379654068e-07,
"loss": 0.012,
"mean_token_accuracy": 0.9966720283031464,
"num_tokens": 11481319.0,
"step": 5385
},
{
"entropy": 1.647239327430725,
"epoch": 1.7741935483870968,
"grad_norm": 0.439744770526886,
"learning_rate": 6.909784726255242e-07,
"loss": 0.0136,
"mean_token_accuracy": 0.9953515648841857,
"num_tokens": 11491892.0,
"step": 5390
},
{
"entropy": 1.653624951839447,
"epoch": 1.7758393680052666,
"grad_norm": 2.360499620437622,
"learning_rate": 6.810725087078395e-07,
"loss": 0.0177,
"mean_token_accuracy": 0.9931002259254456,
"num_tokens": 11502606.0,
"step": 5395
},
{
"entropy": 1.6672588109970092,
"epoch": 1.7774851876234363,
"grad_norm": 1.195021390914917,
"learning_rate": 6.712355612651145e-07,
"loss": 0.0175,
"mean_token_accuracy": 0.9928203105926514,
"num_tokens": 11513121.0,
"step": 5400
},
{
"entropy": 1.6415475726127624,
"epoch": 1.7791310072416064,
"grad_norm": 3.7338359355926514,
"learning_rate": 6.614677031503059e-07,
"loss": 0.0386,
"mean_token_accuracy": 0.9931011736392975,
"num_tokens": 11523757.0,
"step": 5405
},
{
"entropy": 1.630179488658905,
"epoch": 1.7807768268597761,
"grad_norm": 2.090383768081665,
"learning_rate": 6.517690067046922e-07,
"loss": 0.0193,
"mean_token_accuracy": 0.9923229575157165,
"num_tokens": 11534852.0,
"step": 5410
},
{
"entropy": 1.6325619697570801,
"epoch": 1.782422646477946,
"grad_norm": 0.8092725872993469,
"learning_rate": 6.421395437573386e-07,
"loss": 0.0219,
"mean_token_accuracy": 0.9927444994449616,
"num_tokens": 11545588.0,
"step": 5415
},
{
"entropy": 1.6684589862823487,
"epoch": 1.784068466096116,
"grad_norm": 1.0808240175247192,
"learning_rate": 6.325793856245632e-07,
"loss": 0.0084,
"mean_token_accuracy": 0.9980384469032287,
"num_tokens": 11555985.0,
"step": 5420
},
{
"entropy": 1.676847243309021,
"epoch": 1.7857142857142856,
"grad_norm": 1.1335194110870361,
"learning_rate": 6.230886031094063e-07,
"loss": 0.0168,
"mean_token_accuracy": 0.9945612967014312,
"num_tokens": 11566484.0,
"step": 5425
},
{
"entropy": 1.6188385486602783,
"epoch": 1.7873601053324557,
"grad_norm": 1.116150140762329,
"learning_rate": 6.136672665011089e-07,
"loss": 0.0198,
"mean_token_accuracy": 0.9933637619018555,
"num_tokens": 11577115.0,
"step": 5430
},
{
"entropy": 1.6665136098861695,
"epoch": 1.7890059249506254,
"grad_norm": 1.42592191696167,
"learning_rate": 6.043154455745981e-07,
"loss": 0.0205,
"mean_token_accuracy": 0.9911065042018891,
"num_tokens": 11587678.0,
"step": 5435
},
{
"entropy": 1.642556118965149,
"epoch": 1.7906517445687953,
"grad_norm": 1.1183552742004395,
"learning_rate": 5.950332095899547e-07,
"loss": 0.0191,
"mean_token_accuracy": 0.9941525161266327,
"num_tokens": 11598261.0,
"step": 5440
},
{
"entropy": 1.6583900451660156,
"epoch": 1.7922975641869652,
"grad_norm": 1.2433513402938843,
"learning_rate": 5.858206272919165e-07,
"loss": 0.0263,
"mean_token_accuracy": 0.9929942131042481,
"num_tokens": 11609100.0,
"step": 5445
},
{
"entropy": 1.630812132358551,
"epoch": 1.7939433838051349,
"grad_norm": 3.484126567840576,
"learning_rate": 5.766777669093604e-07,
"loss": 0.032,
"mean_token_accuracy": 0.9925698578357697,
"num_tokens": 11619934.0,
"step": 5450
},
{
"entropy": 1.595421350002289,
"epoch": 1.795589203423305,
"grad_norm": 0.8385356664657593,
"learning_rate": 5.676046961547987e-07,
"loss": 0.0335,
"mean_token_accuracy": 0.9886408507823944,
"num_tokens": 11630717.0,
"step": 5455
},
{
"entropy": 1.6572367787361144,
"epoch": 1.7972350230414746,
"grad_norm": 0.9135512113571167,
"learning_rate": 5.586014822238772e-07,
"loss": 0.0208,
"mean_token_accuracy": 0.9933992445468902,
"num_tokens": 11641288.0,
"step": 5460
},
{
"entropy": 1.5970224499702455,
"epoch": 1.7988808426596445,
"grad_norm": 1.7872576713562012,
"learning_rate": 5.496681917948809e-07,
"loss": 0.0154,
"mean_token_accuracy": 0.9937366545200348,
"num_tokens": 11652159.0,
"step": 5465
},
{
"entropy": 1.632554793357849,
"epoch": 1.8005266622778144,
"grad_norm": 0.10970994830131531,
"learning_rate": 5.408048910282348e-07,
"loss": 0.032,
"mean_token_accuracy": 0.9962066173553467,
"num_tokens": 11662941.0,
"step": 5470
},
{
"epoch": 1.8011849901250823,
"eval_entropy": 1.6544960326694962,
"eval_loss": 0.04707782715559006,
"eval_mean_token_accuracy": 0.9874065164100884,
"eval_num_tokens": 11667189.0,
"eval_runtime": 196.3646,
"eval_samples_per_second": 42.406,
"eval_steps_per_second": 7.068,
"step": 5472
},
{
"entropy": 1.6609861731529236,
"epoch": 1.8021724818959841,
"grad_norm": 1.7392816543579102,
"learning_rate": 5.320116455660185e-07,
"loss": 0.0206,
"mean_token_accuracy": 0.9918794929981232,
"num_tokens": 11673648.0,
"step": 5475
},
{
"entropy": 1.6410995721817017,
"epoch": 1.803818301514154,
"grad_norm": 1.1604136228561401,
"learning_rate": 5.232885205314797e-07,
"loss": 0.032,
"mean_token_accuracy": 0.9902990460395813,
"num_tokens": 11684381.0,
"step": 5480
},
{
"entropy": 1.6263534784317017,
"epoch": 1.805464121132324,
"grad_norm": 1.2235734462738037,
"learning_rate": 5.146355805285452e-07,
"loss": 0.0198,
"mean_token_accuracy": 0.994633013010025,
"num_tokens": 11694890.0,
"step": 5485
},
{
"entropy": 1.642207968235016,
"epoch": 1.8071099407504936,
"grad_norm": 3.1930456161499023,
"learning_rate": 5.06052889641353e-07,
"loss": 0.0133,
"mean_token_accuracy": 0.994309377670288,
"num_tokens": 11705533.0,
"step": 5490
},
{
"entropy": 1.6500659108161926,
"epoch": 1.8087557603686637,
"grad_norm": 3.067206621170044,
"learning_rate": 4.975405114337695e-07,
"loss": 0.0303,
"mean_token_accuracy": 0.9904295146465302,
"num_tokens": 11716171.0,
"step": 5495
},
{
"entropy": 1.636317217350006,
"epoch": 1.8104015799868334,
"grad_norm": 0.6358233094215393,
"learning_rate": 4.890985089489231e-07,
"loss": 0.0135,
"mean_token_accuracy": 0.9961594045162201,
"num_tokens": 11726724.0,
"step": 5500
},
{
"entropy": 1.6213058233261108,
"epoch": 1.8120473996050033,
"grad_norm": 1.792466402053833,
"learning_rate": 4.807269447087348e-07,
"loss": 0.019,
"mean_token_accuracy": 0.9931978344917297,
"num_tokens": 11737311.0,
"step": 5505
},
{
"entropy": 1.6493823170661925,
"epoch": 1.8136932192231732,
"grad_norm": 1.9440345764160156,
"learning_rate": 4.7242588071345965e-07,
"loss": 0.0278,
"mean_token_accuracy": 0.9932690799236298,
"num_tokens": 11747961.0,
"step": 5510
},
{
"entropy": 1.6495756149291991,
"epoch": 1.8153390388413428,
"grad_norm": 1.2750672101974487,
"learning_rate": 4.6419537844121565e-07,
"loss": 0.0189,
"mean_token_accuracy": 0.9953626215457916,
"num_tokens": 11758583.0,
"step": 5515
},
{
"entropy": 1.6222103834152222,
"epoch": 1.816984858459513,
"grad_norm": 2.3976049423217773,
"learning_rate": 4.5603549884754463e-07,
"loss": 0.0208,
"mean_token_accuracy": 0.9919763743877411,
"num_tokens": 11769636.0,
"step": 5520
},
{
"entropy": 1.6041205644607544,
"epoch": 1.8186306780776826,
"grad_norm": 1.3786771297454834,
"learning_rate": 4.479463023649555e-07,
"loss": 0.0127,
"mean_token_accuracy": 0.9947966277599335,
"num_tokens": 11780222.0,
"step": 5525
},
{
"entropy": 1.657669985294342,
"epoch": 1.8202764976958525,
"grad_norm": 3.0026445388793945,
"learning_rate": 4.3992784890246276e-07,
"loss": 0.0108,
"mean_token_accuracy": 0.9966068983078002,
"num_tokens": 11790566.0,
"step": 5530
},
{
"entropy": 1.6583294034004212,
"epoch": 1.8219223173140224,
"grad_norm": 2.0243186950683594,
"learning_rate": 4.319801978451654e-07,
"loss": 0.0269,
"mean_token_accuracy": 0.9926669538021088,
"num_tokens": 11801065.0,
"step": 5535
},
{
"entropy": 1.6717671513557435,
"epoch": 1.823568136932192,
"grad_norm": 0.7521846294403076,
"learning_rate": 4.241034080537909e-07,
"loss": 0.0156,
"mean_token_accuracy": 0.9947370052337646,
"num_tokens": 11811699.0,
"step": 5540
},
{
"entropy": 1.657283067703247,
"epoch": 1.8252139565503622,
"grad_norm": 2.917626142501831,
"learning_rate": 4.162975378642653e-07,
"loss": 0.0252,
"mean_token_accuracy": 0.9930805206298828,
"num_tokens": 11822213.0,
"step": 5545
},
{
"entropy": 1.6172433972358704,
"epoch": 1.826859776168532,
"grad_norm": 0.5599117875099182,
"learning_rate": 4.085626450872782e-07,
"loss": 0.0151,
"mean_token_accuracy": 0.9949965596199035,
"num_tokens": 11833106.0,
"step": 5550
},
{
"entropy": 1.6172014594078064,
"epoch": 1.8285055957867018,
"grad_norm": 1.6733261346817017,
"learning_rate": 4.008987870078629e-07,
"loss": 0.0168,
"mean_token_accuracy": 0.9940386712551117,
"num_tokens": 11843679.0,
"step": 5555
},
{
"entropy": 1.6307536125183106,
"epoch": 1.8301514154048717,
"grad_norm": 2.103327512741089,
"learning_rate": 3.9330602038495925e-07,
"loss": 0.0234,
"mean_token_accuracy": 0.9932344555854797,
"num_tokens": 11854343.0,
"step": 5560
},
{
"entropy": 1.6388854265213013,
"epoch": 1.8317972350230414,
"grad_norm": 0.9723978638648987,
"learning_rate": 3.8578440145100373e-07,
"loss": 0.0197,
"mean_token_accuracy": 0.9940811336040497,
"num_tokens": 11864727.0,
"step": 5565
},
{
"entropy": 1.6083611011505128,
"epoch": 1.8334430546412115,
"grad_norm": 1.644818902015686,
"learning_rate": 3.783339859115065e-07,
"loss": 0.0251,
"mean_token_accuracy": 0.990217787027359,
"num_tokens": 11875655.0,
"step": 5570
},
{
"entropy": 1.650509774684906,
"epoch": 1.8350888742593812,
"grad_norm": 4.321114540100098,
"learning_rate": 3.709548289446452e-07,
"loss": 0.0359,
"mean_token_accuracy": 0.9882233738899231,
"num_tokens": 11886276.0,
"step": 5575
},
{
"entropy": 1.6342867493629456,
"epoch": 1.836734693877551,
"grad_norm": 0.9216455221176147,
"learning_rate": 3.636469852008473e-07,
"loss": 0.0163,
"mean_token_accuracy": 0.9964061677455902,
"num_tokens": 11896970.0,
"step": 5580
},
{
"entropy": 1.6529955506324767,
"epoch": 1.838380513495721,
"grad_norm": 1.3248862028121948,
"learning_rate": 3.564105088023984e-07,
"loss": 0.0174,
"mean_token_accuracy": 0.9947122275829315,
"num_tokens": 11907415.0,
"step": 5585
},
{
"entropy": 1.6415375351905823,
"epoch": 1.8400263331138906,
"grad_norm": 1.5896852016448975,
"learning_rate": 3.4924545334302675e-07,
"loss": 0.0124,
"mean_token_accuracy": 0.9948625266551971,
"num_tokens": 11918125.0,
"step": 5590
},
{
"entropy": 1.6392775058746338,
"epoch": 1.8416721527320605,
"grad_norm": 0.6162107586860657,
"learning_rate": 3.421518718875161e-07,
"loss": 0.0204,
"mean_token_accuracy": 0.993281239271164,
"num_tokens": 11928566.0,
"step": 5595
},
{
"entropy": 1.6502568006515503,
"epoch": 1.8433179723502304,
"grad_norm": 0.8682472109794617,
"learning_rate": 3.351298169713102e-07,
"loss": 0.0241,
"mean_token_accuracy": 0.992335319519043,
"num_tokens": 11939271.0,
"step": 5600
},
{
"entropy": 1.6440783381462096,
"epoch": 1.8449637919684,
"grad_norm": 1.3868647813796997,
"learning_rate": 3.281793406001232e-07,
"loss": 0.0284,
"mean_token_accuracy": 0.9914430260658265,
"num_tokens": 11949781.0,
"step": 5605
},
{
"entropy": 1.6473740935325623,
"epoch": 1.8466096115865702,
"grad_norm": 1.2047520875930786,
"learning_rate": 3.213004942495546e-07,
"loss": 0.022,
"mean_token_accuracy": 0.9923123598098755,
"num_tokens": 11960650.0,
"step": 5610
},
{
"entropy": 1.6165038704872132,
"epoch": 1.84825543120474,
"grad_norm": 1.3083264827728271,
"learning_rate": 3.144933288647067e-07,
"loss": 0.01,
"mean_token_accuracy": 0.9968856811523438,
"num_tokens": 11971426.0,
"step": 5615
},
{
"entropy": 1.6375835180282592,
"epoch": 1.8499012508229098,
"grad_norm": 2.6903843879699707,
"learning_rate": 3.0775789485981254e-07,
"loss": 0.0134,
"mean_token_accuracy": 0.9949893295764923,
"num_tokens": 11982373.0,
"step": 5620
},
{
"entropy": 1.6399108648300171,
"epoch": 1.8515470704410797,
"grad_norm": 0.4139382839202881,
"learning_rate": 3.010942421178531e-07,
"loss": 0.0415,
"mean_token_accuracy": 0.9894947111606598,
"num_tokens": 11993082.0,
"step": 5625
},
{
"entropy": 1.6450653195381164,
"epoch": 1.8531928900592494,
"grad_norm": 1.1827541589736938,
"learning_rate": 2.9450241999020024e-07,
"loss": 0.0166,
"mean_token_accuracy": 0.9943609952926635,
"num_tokens": 12003839.0,
"step": 5630
},
{
"entropy": 1.6487621068954468,
"epoch": 1.8548387096774195,
"grad_norm": 0.9702763557434082,
"learning_rate": 2.879824772962381e-07,
"loss": 0.0148,
"mean_token_accuracy": 0.9935874819755555,
"num_tokens": 12014594.0,
"step": 5635
},
{
"entropy": 1.6262930750846862,
"epoch": 1.8564845292955892,
"grad_norm": 0.6445523500442505,
"learning_rate": 2.81534462323011e-07,
"loss": 0.0182,
"mean_token_accuracy": 0.9919591426849366,
"num_tokens": 12025486.0,
"step": 5640
},
{
"entropy": 1.6225888967514037,
"epoch": 1.858130348913759,
"grad_norm": 1.6470600366592407,
"learning_rate": 2.7515842282486274e-07,
"loss": 0.0263,
"mean_token_accuracy": 0.9922280848026276,
"num_tokens": 12035884.0,
"step": 5645
},
{
"entropy": 1.637891697883606,
"epoch": 1.859776168531929,
"grad_norm": 0.8527348637580872,
"learning_rate": 2.688544060230835e-07,
"loss": 0.0201,
"mean_token_accuracy": 0.9935072481632232,
"num_tokens": 12046847.0,
"step": 5650
},
{
"entropy": 1.6528636813163757,
"epoch": 1.8614219881500986,
"grad_norm": 0.8655990362167358,
"learning_rate": 2.626224586055581e-07,
"loss": 0.0191,
"mean_token_accuracy": 0.9932506561279297,
"num_tokens": 12057498.0,
"step": 5655
},
{
"entropy": 1.6207883477210998,
"epoch": 1.8630678077682687,
"grad_norm": 0.7990636229515076,
"learning_rate": 2.5646262672642033e-07,
"loss": 0.0169,
"mean_token_accuracy": 0.9947772800922394,
"num_tokens": 12068243.0,
"step": 5660
},
{
"entropy": 1.6112945556640625,
"epoch": 1.8647136273864384,
"grad_norm": 2.073958396911621,
"learning_rate": 2.503749560057178e-07,
"loss": 0.0185,
"mean_token_accuracy": 0.9919489622116089,
"num_tokens": 12079308.0,
"step": 5665
},
{
"entropy": 1.6480265259742737,
"epoch": 1.8663594470046083,
"grad_norm": 2.8769478797912598,
"learning_rate": 2.4435949152906144e-07,
"loss": 0.0235,
"mean_token_accuracy": 0.9915726184844971,
"num_tokens": 12090018.0,
"step": 5670
},
{
"entropy": 1.616892433166504,
"epoch": 1.8680052666227782,
"grad_norm": 0.648573100566864,
"learning_rate": 2.3841627784730536e-07,
"loss": 0.0174,
"mean_token_accuracy": 0.993311470746994,
"num_tokens": 12100850.0,
"step": 5675
},
{
"entropy": 1.6181252598762512,
"epoch": 1.869651086240948,
"grad_norm": 2.1162548065185547,
"learning_rate": 2.325453589762061e-07,
"loss": 0.0188,
"mean_token_accuracy": 0.9941606819629669,
"num_tokens": 12111601.0,
"step": 5680
},
{
"entropy": 1.6273619651794433,
"epoch": 1.8712969058591178,
"grad_norm": 3.485987901687622,
"learning_rate": 2.2674677839610305e-07,
"loss": 0.0351,
"mean_token_accuracy": 0.9911395490169526,
"num_tokens": 12122330.0,
"step": 5685
},
{
"entropy": 1.6523537158966064,
"epoch": 1.8729427254772877,
"grad_norm": 0.9486113786697388,
"learning_rate": 2.2102057905159292e-07,
"loss": 0.0188,
"mean_token_accuracy": 0.9935419499874115,
"num_tokens": 12132919.0,
"step": 5690
},
{
"entropy": 1.6415381669998168,
"epoch": 1.8745885450954576,
"grad_norm": 1.972302794456482,
"learning_rate": 2.1536680335121684e-07,
"loss": 0.0198,
"mean_token_accuracy": 0.9924857258796692,
"num_tokens": 12143394.0,
"step": 5695
},
{
"entropy": 1.6063251495361328,
"epoch": 1.8762343647136275,
"grad_norm": 0.8765217661857605,
"learning_rate": 2.0978549316713615e-07,
"loss": 0.0138,
"mean_token_accuracy": 0.9943941414356232,
"num_tokens": 12154201.0,
"step": 5700
},
{
"entropy": 1.6426220178604125,
"epoch": 1.8778801843317972,
"grad_norm": 1.477146029472351,
"learning_rate": 2.0427668983483361e-07,
"loss": 0.0221,
"mean_token_accuracy": 0.9906249940395355,
"num_tokens": 12164635.0,
"step": 5705
},
{
"entropy": 1.6509037852287292,
"epoch": 1.879526003949967,
"grad_norm": 1.7725666761398315,
"learning_rate": 1.9884043415280274e-07,
"loss": 0.0229,
"mean_token_accuracy": 0.9909674286842346,
"num_tokens": 12175320.0,
"step": 5710
},
{
"entropy": 1.649256443977356,
"epoch": 1.881171823568137,
"grad_norm": 1.5584267377853394,
"learning_rate": 1.9347676638224122e-07,
"loss": 0.0182,
"mean_token_accuracy": 0.9936787724494934,
"num_tokens": 12185966.0,
"step": 5715
},
{
"entropy": 1.6484339714050293,
"epoch": 1.8828176431863066,
"grad_norm": 0.69387286901474,
"learning_rate": 1.8818572624676124e-07,
"loss": 0.0179,
"mean_token_accuracy": 0.9950119018554687,
"num_tokens": 12196531.0,
"step": 5720
},
{
"entropy": 1.618002474308014,
"epoch": 1.8844634628044767,
"grad_norm": 3.0518853664398193,
"learning_rate": 1.8296735293208745e-07,
"loss": 0.0324,
"mean_token_accuracy": 0.991709417104721,
"num_tokens": 12207170.0,
"step": 5725
},
{
"entropy": 1.6460561990737914,
"epoch": 1.8861092824226464,
"grad_norm": 0.9351851940155029,
"learning_rate": 1.7782168508577168e-07,
"loss": 0.0178,
"mean_token_accuracy": 0.9925060391426086,
"num_tokens": 12217651.0,
"step": 5730
},
{
"entropy": 1.6722620725631714,
"epoch": 1.8877551020408163,
"grad_norm": 2.6632797718048096,
"learning_rate": 1.7274876081690429e-07,
"loss": 0.0423,
"mean_token_accuracy": 0.9879862844944001,
"num_tokens": 12227987.0,
"step": 5735
},
{
"entropy": 1.6359737396240235,
"epoch": 1.8894009216589862,
"grad_norm": 0.5745594501495361,
"learning_rate": 1.6774861769583538e-07,
"loss": 0.0291,
"mean_token_accuracy": 0.9925860285758972,
"num_tokens": 12238790.0,
"step": 5740
},
{
"entropy": 1.6011884927749633,
"epoch": 1.8910467412771559,
"grad_norm": 1.4916797876358032,
"learning_rate": 1.628212927538908e-07,
"loss": 0.0285,
"mean_token_accuracy": 0.9890454232692718,
"num_tokens": 12249683.0,
"step": 5745
},
{
"entropy": 1.6060463190078735,
"epoch": 1.892692560895326,
"grad_norm": 0.7883341312408447,
"learning_rate": 1.5796682248310214e-07,
"loss": 0.0156,
"mean_token_accuracy": 0.9944408297538757,
"num_tokens": 12260660.0,
"step": 5750
},
{
"entropy": 1.6521755695343017,
"epoch": 1.8943383805134957,
"grad_norm": 0.684786856174469,
"learning_rate": 1.5318524283593706e-07,
"loss": 0.0072,
"mean_token_accuracy": 0.9973987996578216,
"num_tokens": 12271559.0,
"step": 5755
},
{
"entropy": 1.6584128499031068,
"epoch": 1.8959842001316656,
"grad_norm": 3.097158432006836,
"learning_rate": 1.484765892250284e-07,
"loss": 0.0216,
"mean_token_accuracy": 0.9943334817886352,
"num_tokens": 12281996.0,
"step": 5760
},
{
"entropy": 1.6525746941566468,
"epoch": 1.8976300197498355,
"grad_norm": 4.315025329589844,
"learning_rate": 1.4384089652291544e-07,
"loss": 0.0293,
"mean_token_accuracy": 0.9913449168205262,
"num_tokens": 12292609.0,
"step": 5765
},
{
"entropy": 1.608083975315094,
"epoch": 1.8992758393680051,
"grad_norm": 0.8511555790901184,
"learning_rate": 1.3927819906178864e-07,
"loss": 0.0162,
"mean_token_accuracy": 0.9940433919429779,
"num_tokens": 12303345.0,
"step": 5770
},
{
"entropy": 1.654729962348938,
"epoch": 1.9009216589861753,
"grad_norm": 5.2224860191345215,
"learning_rate": 1.3478853063322862e-07,
"loss": 0.0203,
"mean_token_accuracy": 0.9930156350135804,
"num_tokens": 12313946.0,
"step": 5775
},
{
"entropy": 1.6443889737129211,
"epoch": 1.902567478604345,
"grad_norm": 0.4680110812187195,
"learning_rate": 1.3037192448795754e-07,
"loss": 0.0185,
"mean_token_accuracy": 0.9930983424186707,
"num_tokens": 12324639.0,
"step": 5780
},
{
"entropy": 1.634881365299225,
"epoch": 1.9042132982225148,
"grad_norm": 1.4533123970031738,
"learning_rate": 1.2602841333559934e-07,
"loss": 0.0181,
"mean_token_accuracy": 0.995710015296936,
"num_tokens": 12335427.0,
"step": 5785
},
{
"entropy": 1.6775338172912597,
"epoch": 1.9058591178406847,
"grad_norm": 1.9515726566314697,
"learning_rate": 1.217580293444276e-07,
"loss": 0.0208,
"mean_token_accuracy": 0.9887213408946991,
"num_tokens": 12345700.0,
"step": 5790
},
{
"entropy": 1.6608016729354858,
"epoch": 1.9075049374588544,
"grad_norm": 2.2838118076324463,
"learning_rate": 1.1756080414113691e-07,
"loss": 0.0221,
"mean_token_accuracy": 0.9925664782524108,
"num_tokens": 12356355.0,
"step": 5795
},
{
"entropy": 1.6454607725143433,
"epoch": 1.9091507570770243,
"grad_norm": 1.0771596431732178,
"learning_rate": 1.1343676881059751e-07,
"loss": 0.0106,
"mean_token_accuracy": 0.9954189419746399,
"num_tokens": 12366898.0,
"step": 5800
},
{
"entropy": 1.6636238932609557,
"epoch": 1.9107965766951942,
"grad_norm": 3.005011796951294,
"learning_rate": 1.0938595389563988e-07,
"loss": 0.0143,
"mean_token_accuracy": 0.994761872291565,
"num_tokens": 12377443.0,
"step": 5805
},
{
"entropy": 1.6605603098869324,
"epoch": 1.912442396313364,
"grad_norm": 2.230198621749878,
"learning_rate": 1.0540838939681164e-07,
"loss": 0.0242,
"mean_token_accuracy": 0.9926084816455841,
"num_tokens": 12388275.0,
"step": 5810
},
{
"entropy": 1.606773316860199,
"epoch": 1.914088215931534,
"grad_norm": 0.7968350052833557,
"learning_rate": 1.0150410477216987e-07,
"loss": 0.0105,
"mean_token_accuracy": 0.9964967429637909,
"num_tokens": 12399282.0,
"step": 5815
},
{
"entropy": 1.6078164100646972,
"epoch": 1.9157340355497037,
"grad_norm": 0.7457708716392517,
"learning_rate": 9.767312893705583e-08,
"loss": 0.0203,
"mean_token_accuracy": 0.9911931037902832,
"num_tokens": 12410043.0,
"step": 5820
},
{
"entropy": 1.6395422101020813,
"epoch": 1.9173798551678736,
"grad_norm": 0.5907423496246338,
"learning_rate": 9.391549026387948e-08,
"loss": 0.0238,
"mean_token_accuracy": 0.9929281890392303,
"num_tokens": 12420722.0,
"step": 5825
},
{
"entropy": 1.6461820960044862,
"epoch": 1.9190256747860435,
"grad_norm": 2.0906271934509277,
"learning_rate": 9.023121658191636e-08,
"loss": 0.0187,
"mean_token_accuracy": 0.993992280960083,
"num_tokens": 12431395.0,
"step": 5830
},
{
"entropy": 1.6568657994270324,
"epoch": 1.9206714944042131,
"grad_norm": 0.6122414469718933,
"learning_rate": 8.662033517709113e-08,
"loss": 0.0306,
"mean_token_accuracy": 0.9899275124073028,
"num_tokens": 12442077.0,
"step": 5835
},
{
"entropy": 1.6694289207458497,
"epoch": 1.9223173140223833,
"grad_norm": 0.9769027829170227,
"learning_rate": 8.308287279178651e-08,
"loss": 0.0214,
"mean_token_accuracy": 0.9897110819816589,
"num_tokens": 12452490.0,
"step": 5840
},
{
"entropy": 1.6428742289543152,
"epoch": 1.923963133640553,
"grad_norm": 1.7451066970825195,
"learning_rate": 7.961885562463689e-08,
"loss": 0.0205,
"mean_token_accuracy": 0.993678605556488,
"num_tokens": 12463113.0,
"step": 5845
},
{
"entropy": 1.6416840195655822,
"epoch": 1.9256089532587228,
"grad_norm": 0.20614351332187653,
"learning_rate": 7.622830933033954e-08,
"loss": 0.0159,
"mean_token_accuracy": 0.9947439730167389,
"num_tokens": 12473584.0,
"step": 5850
},
{
"entropy": 1.668267822265625,
"epoch": 1.9272547728768927,
"grad_norm": 1.4911457300186157,
"learning_rate": 7.291125901946027e-08,
"loss": 0.0293,
"mean_token_accuracy": 0.9900563299655915,
"num_tokens": 12483990.0,
"step": 5855
},
{
"entropy": 1.6057178854942322,
"epoch": 1.9289005924950624,
"grad_norm": 1.5931812524795532,
"learning_rate": 6.966772925825149e-08,
"loss": 0.0199,
"mean_token_accuracy": 0.9930034399032592,
"num_tokens": 12494711.0,
"step": 5860
},
{
"entropy": 1.6411123514175414,
"epoch": 1.9305464121132325,
"grad_norm": 2.541999578475952,
"learning_rate": 6.649774406846777e-08,
"loss": 0.0191,
"mean_token_accuracy": 0.9947657942771911,
"num_tokens": 12505189.0,
"step": 5865
},
{
"entropy": 1.6578918814659118,
"epoch": 1.9321922317314022,
"grad_norm": 1.0876408815383911,
"learning_rate": 6.340132692718936e-08,
"loss": 0.0143,
"mean_token_accuracy": 0.994833791255951,
"num_tokens": 12515680.0,
"step": 5870
},
{
"entropy": 1.620399296283722,
"epoch": 1.933838051349572,
"grad_norm": 1.0122355222702026,
"learning_rate": 6.037850076664686e-08,
"loss": 0.02,
"mean_token_accuracy": 0.9926087379455566,
"num_tokens": 12526219.0,
"step": 5875
},
{
"entropy": 1.619743776321411,
"epoch": 1.935483870967742,
"grad_norm": 2.4715607166290283,
"learning_rate": 5.742928797405234e-08,
"loss": 0.027,
"mean_token_accuracy": 0.9934347212314606,
"num_tokens": 12536968.0,
"step": 5880
},
{
"entropy": 1.6266910910606385,
"epoch": 1.9371296905859117,
"grad_norm": 1.2927361726760864,
"learning_rate": 5.455371039143176e-08,
"loss": 0.0296,
"mean_token_accuracy": 0.9922801196575165,
"num_tokens": 12547695.0,
"step": 5885
},
{
"entropy": 1.6311318516731261,
"epoch": 1.9387755102040818,
"grad_norm": 1.5303946733474731,
"learning_rate": 5.175178931546842e-08,
"loss": 0.0256,
"mean_token_accuracy": 0.9924924194812774,
"num_tokens": 12558345.0,
"step": 5890
},
{
"entropy": 1.6298449873924254,
"epoch": 1.9404213298222515,
"grad_norm": 0.6128053069114685,
"learning_rate": 4.902354549733979e-08,
"loss": 0.0268,
"mean_token_accuracy": 0.9905460000038147,
"num_tokens": 12568986.0,
"step": 5895
},
{
"entropy": 1.633393156528473,
"epoch": 1.9420671494404214,
"grad_norm": 1.1418019533157349,
"learning_rate": 4.636899914256421e-08,
"loss": 0.0184,
"mean_token_accuracy": 0.9936664879322052,
"num_tokens": 12579759.0,
"step": 5900
},
{
"entropy": 1.63808730840683,
"epoch": 1.9437129690585913,
"grad_norm": 2.710805654525757,
"learning_rate": 4.378816991085333e-08,
"loss": 0.0198,
"mean_token_accuracy": 0.9934788465499877,
"num_tokens": 12590306.0,
"step": 5905
},
{
"entropy": 1.6379319310188294,
"epoch": 1.945358788676761,
"grad_norm": 1.7254809141159058,
"learning_rate": 4.128107691596772e-08,
"loss": 0.0228,
"mean_token_accuracy": 0.9918417274951935,
"num_tokens": 12600767.0,
"step": 5910
},
{
"entropy": 1.6477123498916626,
"epoch": 1.9470046082949308,
"grad_norm": 0.8686738014221191,
"learning_rate": 3.884773872557035e-08,
"loss": 0.0285,
"mean_token_accuracy": 0.989762258529663,
"num_tokens": 12611296.0,
"step": 5915
},
{
"entropy": 1.6127866506576538,
"epoch": 1.9486504279131007,
"grad_norm": 1.5231475830078125,
"learning_rate": 3.648817336109556e-08,
"loss": 0.0185,
"mean_token_accuracy": 0.9937713503837585,
"num_tokens": 12622049.0,
"step": 5920
},
{
"entropy": 1.6393255352973939,
"epoch": 1.9502962475312706,
"grad_norm": 1.2301123142242432,
"learning_rate": 3.420239829761029e-08,
"loss": 0.0169,
"mean_token_accuracy": 0.9921809434890747,
"num_tokens": 12632714.0,
"step": 5925
},
{
"entropy": 1.661654818058014,
"epoch": 1.9519420671494405,
"grad_norm": 0.7805505990982056,
"learning_rate": 3.199043046368644e-08,
"loss": 0.0181,
"mean_token_accuracy": 0.9945761322975158,
"num_tokens": 12643222.0,
"step": 5930
},
{
"entropy": 1.6112356901168823,
"epoch": 1.9535878867676102,
"grad_norm": 1.137802004814148,
"learning_rate": 2.985228624127534e-08,
"loss": 0.0189,
"mean_token_accuracy": 0.9933175027370453,
"num_tokens": 12654137.0,
"step": 5935
},
{
"entropy": 1.6136484742164612,
"epoch": 1.95523370638578,
"grad_norm": 1.2796485424041748,
"learning_rate": 2.778798146558903e-08,
"loss": 0.0161,
"mean_token_accuracy": 0.9944588780403137,
"num_tokens": 12664918.0,
"step": 5940
},
{
"entropy": 1.615454363822937,
"epoch": 1.95687952600395,
"grad_norm": 1.3428839445114136,
"learning_rate": 2.5797531424976983e-08,
"loss": 0.0275,
"mean_token_accuracy": 0.993310171365738,
"num_tokens": 12675751.0,
"step": 5945
},
{
"entropy": 1.63991322517395,
"epoch": 1.9585253456221197,
"grad_norm": 1.0178812742233276,
"learning_rate": 2.388095086081954e-08,
"loss": 0.0189,
"mean_token_accuracy": 0.9943699061870575,
"num_tokens": 12686462.0,
"step": 5950
},
{
"entropy": 1.6252501964569093,
"epoch": 1.9601711652402898,
"grad_norm": 0.5898308157920837,
"learning_rate": 2.2038253967415768e-08,
"loss": 0.0262,
"mean_token_accuracy": 0.9889247179031372,
"num_tokens": 12697111.0,
"step": 5955
},
{
"entropy": 1.6657869219779968,
"epoch": 1.9618169848584595,
"grad_norm": 1.152711033821106,
"learning_rate": 2.0269454391874665e-08,
"loss": 0.0195,
"mean_token_accuracy": 0.9946586787700653,
"num_tokens": 12707827.0,
"step": 5960
},
{
"entropy": 1.6191688776016235,
"epoch": 1.9634628044766294,
"grad_norm": 1.1193846464157104,
"learning_rate": 1.8574565234023014e-08,
"loss": 0.0101,
"mean_token_accuracy": 0.9955754101276397,
"num_tokens": 12718563.0,
"step": 5965
},
{
"entropy": 1.603171467781067,
"epoch": 1.9651086240947993,
"grad_norm": 1.5283174514770508,
"learning_rate": 1.6953599046299895e-08,
"loss": 0.0175,
"mean_token_accuracy": 0.9934708416461945,
"num_tokens": 12729286.0,
"step": 5970
},
{
"entropy": 1.6272645711898803,
"epoch": 1.966754443712969,
"grad_norm": 0.8465014100074768,
"learning_rate": 1.5406567833666785e-08,
"loss": 0.0151,
"mean_token_accuracy": 0.9951033234596253,
"num_tokens": 12739732.0,
"step": 5975
},
{
"entropy": 1.644085705280304,
"epoch": 1.968400263331139,
"grad_norm": 0.6121142506599426,
"learning_rate": 1.3933483053519825e-08,
"loss": 0.0142,
"mean_token_accuracy": 0.9949011683464051,
"num_tokens": 12750492.0,
"step": 5980
},
{
"entropy": 1.6154101610183715,
"epoch": 1.9700460829493087,
"grad_norm": 2.3735482692718506,
"learning_rate": 1.2534355615603233e-08,
"loss": 0.033,
"mean_token_accuracy": 0.9925001919269562,
"num_tokens": 12761291.0,
"step": 5985
},
{
"entropy": 1.6425586223602295,
"epoch": 1.9716919025674786,
"grad_norm": 2.7295331954956055,
"learning_rate": 1.1209195881930479e-08,
"loss": 0.0247,
"mean_token_accuracy": 0.9938769578933716,
"num_tokens": 12771987.0,
"step": 5990
},
{
"entropy": 1.5966406106948852,
"epoch": 1.9733377221856485,
"grad_norm": 2.0931482315063477,
"learning_rate": 9.958013666704347e-09,
"loss": 0.0198,
"mean_token_accuracy": 0.9945861279964447,
"num_tokens": 12782717.0,
"step": 5995
},
{
"entropy": 1.6259140491485595,
"epoch": 1.9749835418038182,
"grad_norm": 0.6469723582267761,
"learning_rate": 8.780818236248101e-09,
"loss": 0.0242,
"mean_token_accuracy": 0.9942210376262665,
"num_tokens": 12793255.0,
"step": 6000
},
{
"entropy": 1.6549716949462892,
"epoch": 1.9766293614219883,
"grad_norm": 0.901390790939331,
"learning_rate": 7.67761830893443e-09,
"loss": 0.0137,
"mean_token_accuracy": 0.9956705689430236,
"num_tokens": 12804142.0,
"step": 6005
},
{
"entropy": 1.6143463373184204,
"epoch": 1.978275181040158,
"grad_norm": 1.2343865633010864,
"learning_rate": 6.648422055118842e-09,
"loss": 0.0127,
"mean_token_accuracy": 0.9951093852519989,
"num_tokens": 12815008.0,
"step": 6010
},
{
"entropy": 1.6510612845420838,
"epoch": 1.9799210006583279,
"grad_norm": 2.802551031112671,
"learning_rate": 5.693237097085247e-09,
"loss": 0.0162,
"mean_token_accuracy": 0.9935081660747528,
"num_tokens": 12825498.0,
"step": 6015
},
{
"entropy": 1.6498525738716125,
"epoch": 1.9815668202764978,
"grad_norm": 0.6338088512420654,
"learning_rate": 4.8120705089849116e-09,
"loss": 0.0147,
"mean_token_accuracy": 0.9955414116382599,
"num_tokens": 12835988.0,
"step": 6020
},
{
"entropy": 1.6122831344604491,
"epoch": 1.9832126398946675,
"grad_norm": 2.181898355484009,
"learning_rate": 4.00492881678427e-09,
"loss": 0.0294,
"mean_token_accuracy": 0.9928855717182159,
"num_tokens": 12846863.0,
"step": 6025
},
{
"entropy": 1.638089156150818,
"epoch": 1.9848584595128373,
"grad_norm": 2.5133092403411865,
"learning_rate": 3.271817998216076e-09,
"loss": 0.025,
"mean_token_accuracy": 0.9939551532268525,
"num_tokens": 12857465.0,
"step": 6030
},
{
"entropy": 1.5842849373817445,
"epoch": 1.9865042791310072,
"grad_norm": 3.628753662109375,
"learning_rate": 2.612743482741653e-09,
"loss": 0.0359,
"mean_token_accuracy": 0.9892144203186035,
"num_tokens": 12868438.0,
"step": 6035
},
{
"entropy": 1.676398503780365,
"epoch": 1.9881500987491771,
"grad_norm": 2.431900978088379,
"learning_rate": 2.0277101514987184e-09,
"loss": 0.0276,
"mean_token_accuracy": 0.991285365819931,
"num_tokens": 12878820.0,
"step": 6040
},
{
"entropy": 1.6270844221115113,
"epoch": 1.989795918367347,
"grad_norm": 3.490521192550659,
"learning_rate": 1.5167223372780648e-09,
"loss": 0.0291,
"mean_token_accuracy": 0.9905443787574768,
"num_tokens": 12889541.0,
"step": 6045
},
{
"entropy": 1.595809280872345,
"epoch": 1.9914417379855167,
"grad_norm": 2.535038709640503,
"learning_rate": 1.0797838244802627e-09,
"loss": 0.022,
"mean_token_accuracy": 0.9905419945716858,
"num_tokens": 12900791.0,
"step": 6050
},
{
"entropy": 1.594471561908722,
"epoch": 1.9930875576036866,
"grad_norm": 1.6428083181381226,
"learning_rate": 7.168978490978973e-10,
"loss": 0.0099,
"mean_token_accuracy": 0.9953214168548584,
"num_tokens": 12911687.0,
"step": 6055
},
{
"entropy": 1.6622145771980286,
"epoch": 1.9947333772218565,
"grad_norm": 1.0861161947250366,
"learning_rate": 4.2806709868115084e-10,
"loss": 0.0256,
"mean_token_accuracy": 0.9900858104228973,
"num_tokens": 12922176.0,
"step": 6060
},
{
"entropy": 1.616791796684265,
"epoch": 1.9963791968400262,
"grad_norm": 2.0424978733062744,
"learning_rate": 2.1329371232892138e-10,
"loss": 0.0261,
"mean_token_accuracy": 0.9941427707672119,
"num_tokens": 12932662.0,
"step": 6065
},
{
"entropy": 1.6235838294029237,
"epoch": 1.9980250164581963,
"grad_norm": 0.6039044260978699,
"learning_rate": 7.257928066217723e-11,
"loss": 0.0321,
"mean_token_accuracy": 0.9933849751949311,
"num_tokens": 12943359.0,
"step": 6070
},
{
"entropy": 1.6359187722206117,
"epoch": 1.999670836076366,
"grad_norm": 3.8299636840820312,
"learning_rate": 5.924845819516023e-12,
"loss": 0.0217,
"mean_token_accuracy": 0.9923797845840454,
"num_tokens": 12954081.0,
"step": 6075
}
],
"logging_steps": 5,
"max_steps": 6076,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 608,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 3.285029236308019e+16,
"train_batch_size": 6,
"trial_name": null,
"trial_params": null
}
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 608,
"global_step": 6076,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0,
"eval_entropy": 2.563803545593872,
"eval_loss": 1.3544182777404785,
"eval_mean_token_accuracy": 0.687368397152733,
"eval_num_tokens": 0.0,
"eval_runtime": 198.6529,
"eval_samples_per_second": 41.917,
"eval_steps_per_second": 6.987,
"step": 0
},
{
"entropy": 2.578546404838562,
"epoch": 0.0003291639236339697,
"grad_norm": 58.16326141357422,
"learning_rate": 0.0,
"loss": 1.4746,
"mean_token_accuracy": 0.6601307392120361,
"num_tokens": 2101.0,
"step": 1
},
{
"entropy": 2.547638326883316,
"epoch": 0.0016458196181698486,
"grad_norm": 50.99738311767578,
"learning_rate": 2.6315789473684213e-07,
"loss": 1.2383,
"mean_token_accuracy": 0.7003932222723961,
"num_tokens": 10649.0,
"step": 5
},
{
"entropy": 2.5566530227661133,
"epoch": 0.0032916392363396972,
"grad_norm": 38.35203552246094,
"learning_rate": 5.921052631578947e-07,
"loss": 1.0069,
"mean_token_accuracy": 0.7398434042930603,
"num_tokens": 21271.0,
"step": 10
},
{
"entropy": 2.452682948112488,
"epoch": 0.004937458854509546,
"grad_norm": 27.224767684936523,
"learning_rate": 9.210526315789474e-07,
"loss": 0.5295,
"mean_token_accuracy": 0.8550694227218628,
"num_tokens": 32112.0,
"step": 15
},
{
"entropy": 2.1593106389045715,
"epoch": 0.0065832784726793945,
"grad_norm": 23.221702575683594,
"learning_rate": 1.25e-06,
"loss": 0.2773,
"mean_token_accuracy": 0.9244555592536926,
"num_tokens": 42919.0,
"step": 20
},
{
"entropy": 1.6036222577095032,
"epoch": 0.008229098090849244,
"grad_norm": 15.987848281860352,
"learning_rate": 1.5789473684210526e-06,
"loss": 0.1681,
"mean_token_accuracy": 0.9467954277992249,
"num_tokens": 53618.0,
"step": 25
},
{
"entropy": 1.486020290851593,
"epoch": 0.009874917709019092,
"grad_norm": 11.537003517150879,
"learning_rate": 1.9078947368421057e-06,
"loss": 0.1446,
"mean_token_accuracy": 0.9633475482463837,
"num_tokens": 64755.0,
"step": 30
},
{
"entropy": 1.6325503945350648,
"epoch": 0.01152073732718894,
"grad_norm": 7.058774471282959,
"learning_rate": 2.236842105263158e-06,
"loss": 0.149,
"mean_token_accuracy": 0.9636693239212036,
"num_tokens": 75388.0,