fpadovani/tur-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
tur-latn-100mb-after-ppt-sh…/checkpoint-9500/trainer_state.json
ModelHub XC 28185ab911 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/tur-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407
Source: Original Platform
2026-07-30 04:30:20 +08:00

19244 lines
526 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 9.443560417702635,
"eval_steps": 500,
"global_step": 9500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.695967006683349,
"epoch": 0.004972650422675286,
"grad_norm": 14.4375,
"learning_rate": 2e-06,
"loss": 14.4267,
"mean_token_accuracy": 0.00011848341673612595,
"num_tokens": 9089.0,
"step": 5
},
{
"entropy": 4.677842998504639,
"epoch": 0.009945300845350571,
"grad_norm": 15.3125,
"learning_rate": 4.5e-06,
"loss": 14.5334,
"mean_token_accuracy": 0.00015810276381671428,
"num_tokens": 21650.0,
"step": 10
},
{
"entropy": 4.76645073890686,
"epoch": 0.014917951268025857,
"grad_norm": 19.25,
"learning_rate": 7e-06,
"loss": 14.113,
"mean_token_accuracy": 0.00020675990963354708,
"num_tokens": 31067.0,
"step": 15
},
{
"entropy": 5.015227270126343,
"epoch": 0.019890601690701143,
"grad_norm": 34.25,
"learning_rate": 9.5e-06,
"loss": 13.4863,
"mean_token_accuracy": 0.0,
"num_tokens": 40816.0,
"step": 20
},
{
"entropy": 7.510676717758178,
"epoch": 0.02486325211337643,
"grad_norm": 13.875,
"learning_rate": 1.2e-05,
"loss": 11.7346,
"mean_token_accuracy": 9.78473573923111e-05,
"num_tokens": 51065.0,
"step": 25
},
{
"entropy": 10.519583129882813,
"epoch": 0.029835902536051714,
"grad_norm": 3.4375,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.7082,
"mean_token_accuracy": 0.005600619851611555,
"num_tokens": 60539.0,
"step": 30
},
{
"entropy": 10.700261497497559,
"epoch": 0.034808552958727,
"grad_norm": 2.640625,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.5537,
"mean_token_accuracy": 0.028791588079184294,
"num_tokens": 71010.0,
"step": 35
},
{
"entropy": 10.671732521057129,
"epoch": 0.039781203381402286,
"grad_norm": 2.265625,
"learning_rate": 1.95e-05,
"loss": 10.2564,
"mean_token_accuracy": 0.04515027813613415,
"num_tokens": 81394.0,
"step": 40
},
{
"entropy": 10.483489990234375,
"epoch": 0.04475385380407757,
"grad_norm": 1.8046875,
"learning_rate": 2.2e-05,
"loss": 10.1167,
"mean_token_accuracy": 0.04342293459922075,
"num_tokens": 92491.0,
"step": 45
},
{
"entropy": 10.399119567871093,
"epoch": 0.04972650422675286,
"grad_norm": 1.7109375,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.9358,
"mean_token_accuracy": 0.043518274836242196,
"num_tokens": 102113.0,
"step": 50
},
{
"entropy": 10.48739938735962,
"epoch": 0.05469915464942814,
"grad_norm": 2.1875,
"learning_rate": 2.7e-05,
"loss": 9.8944,
"mean_token_accuracy": 0.04352726638317108,
"num_tokens": 112211.0,
"step": 55
},
{
"entropy": 10.497431468963622,
"epoch": 0.05967180507210343,
"grad_norm": 1.5390625,
"learning_rate": 2.95e-05,
"loss": 9.7893,
"mean_token_accuracy": 0.040843716263771056,
"num_tokens": 122166.0,
"step": 60
},
{
"entropy": 10.438471221923828,
"epoch": 0.06464445549477872,
"grad_norm": 1.4921875,
"learning_rate": 3.2e-05,
"loss": 9.6965,
"mean_token_accuracy": 0.04762680903077125,
"num_tokens": 131016.0,
"step": 65
},
{
"entropy": 10.383831214904784,
"epoch": 0.069617105917454,
"grad_norm": 1.515625,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.6514,
"mean_token_accuracy": 0.04989047236740589,
"num_tokens": 141369.0,
"step": 70
},
{
"entropy": 10.395554351806641,
"epoch": 0.07458975634012929,
"grad_norm": 1.46875,
"learning_rate": 3.7e-05,
"loss": 9.6059,
"mean_token_accuracy": 0.045283081382513045,
"num_tokens": 150939.0,
"step": 75
},
{
"entropy": 10.431264877319336,
"epoch": 0.07956240676280457,
"grad_norm": 1.5078125,
"learning_rate": 3.95e-05,
"loss": 9.5567,
"mean_token_accuracy": 0.05196296907961369,
"num_tokens": 159981.0,
"step": 80
},
{
"entropy": 10.404547595977784,
"epoch": 0.08453505718547986,
"grad_norm": 1.4296875,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.4901,
"mean_token_accuracy": 0.05384046658873558,
"num_tokens": 170139.0,
"step": 85
},
{
"entropy": 10.323073673248292,
"epoch": 0.08950770760815514,
"grad_norm": 1.578125,
"learning_rate": 4.45e-05,
"loss": 9.4153,
"mean_token_accuracy": 0.05128877349197865,
"num_tokens": 180746.0,
"step": 90
},
{
"entropy": 10.234126758575439,
"epoch": 0.09448035803083044,
"grad_norm": 1.46875,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.2695,
"mean_token_accuracy": 0.057958731055259706,
"num_tokens": 190250.0,
"step": 95
},
{
"entropy": 10.200442790985107,
"epoch": 0.09945300845350571,
"grad_norm": 1.3125,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.3164,
"mean_token_accuracy": 0.061629551276564595,
"num_tokens": 201185.0,
"step": 100
},
{
"entropy": 10.167423057556153,
"epoch": 0.10442565887618101,
"grad_norm": 1.359375,
"learning_rate": 5.2e-05,
"loss": 9.2292,
"mean_token_accuracy": 0.057263224571943286,
"num_tokens": 211476.0,
"step": 105
},
{
"entropy": 10.147284317016602,
"epoch": 0.10939830929885629,
"grad_norm": 1.2109375,
"learning_rate": 5.45e-05,
"loss": 9.1279,
"mean_token_accuracy": 0.054998861998319624,
"num_tokens": 220965.0,
"step": 110
},
{
"entropy": 10.161615943908691,
"epoch": 0.11437095972153158,
"grad_norm": 1.4453125,
"learning_rate": 5.7e-05,
"loss": 9.0647,
"mean_token_accuracy": 0.04764294326305389,
"num_tokens": 231539.0,
"step": 115
},
{
"entropy": 10.009593105316162,
"epoch": 0.11934361014420686,
"grad_norm": 1.09375,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.9565,
"mean_token_accuracy": 0.0596030842512846,
"num_tokens": 241206.0,
"step": 120
},
{
"entropy": 9.899549961090088,
"epoch": 0.12431626056688215,
"grad_norm": 1.1796875,
"learning_rate": 6.2e-05,
"loss": 8.8673,
"mean_token_accuracy": 0.056177464127540586,
"num_tokens": 250629.0,
"step": 125
},
{
"entropy": 9.890707683563232,
"epoch": 0.12928891098955744,
"grad_norm": 1.0703125,
"learning_rate": 6.450000000000001e-05,
"loss": 8.8412,
"mean_token_accuracy": 0.05115443356335163,
"num_tokens": 260705.0,
"step": 130
},
{
"entropy": 9.684529113769532,
"epoch": 0.13426156141223272,
"grad_norm": 0.98046875,
"learning_rate": 6.7e-05,
"loss": 8.7676,
"mean_token_accuracy": 0.05398021526634693,
"num_tokens": 270499.0,
"step": 135
},
{
"entropy": 9.696455478668213,
"epoch": 0.139234211834908,
"grad_norm": 1.0546875,
"learning_rate": 6.950000000000001e-05,
"loss": 8.7417,
"mean_token_accuracy": 0.05484265871345997,
"num_tokens": 279998.0,
"step": 140
},
{
"entropy": 9.45594367980957,
"epoch": 0.14420686225758328,
"grad_norm": 1.0390625,
"learning_rate": 7.2e-05,
"loss": 8.6184,
"mean_token_accuracy": 0.058821605145931245,
"num_tokens": 290322.0,
"step": 145
},
{
"entropy": 9.307820510864257,
"epoch": 0.14917951268025859,
"grad_norm": 1.03125,
"learning_rate": 7.45e-05,
"loss": 8.5725,
"mean_token_accuracy": 0.06243101172149181,
"num_tokens": 299893.0,
"step": 150
},
{
"entropy": 9.206488990783692,
"epoch": 0.15415216310293386,
"grad_norm": 0.9140625,
"learning_rate": 7.7e-05,
"loss": 8.5498,
"mean_token_accuracy": 0.06206914372742176,
"num_tokens": 309191.0,
"step": 155
},
{
"entropy": 9.06636447906494,
"epoch": 0.15912481352560914,
"grad_norm": 0.88671875,
"learning_rate": 7.950000000000001e-05,
"loss": 8.4815,
"mean_token_accuracy": 0.06140335313975811,
"num_tokens": 318494.0,
"step": 160
},
{
"entropy": 9.044106483459473,
"epoch": 0.16409746394828442,
"grad_norm": 0.90234375,
"learning_rate": 8.2e-05,
"loss": 8.4775,
"mean_token_accuracy": 0.0596194114536047,
"num_tokens": 328299.0,
"step": 165
},
{
"entropy": 8.939890956878662,
"epoch": 0.16907011437095973,
"grad_norm": 0.81640625,
"learning_rate": 8.450000000000001e-05,
"loss": 8.5442,
"mean_token_accuracy": 0.057165585830807684,
"num_tokens": 338878.0,
"step": 170
},
{
"entropy": 8.874685859680175,
"epoch": 0.174042764793635,
"grad_norm": 1.3359375,
"learning_rate": 8.7e-05,
"loss": 8.468,
"mean_token_accuracy": 0.05600897297263145,
"num_tokens": 348508.0,
"step": 175
},
{
"entropy": 8.797854232788087,
"epoch": 0.1790154152163103,
"grad_norm": 0.921875,
"learning_rate": 8.95e-05,
"loss": 8.4283,
"mean_token_accuracy": 0.06139532402157784,
"num_tokens": 358233.0,
"step": 180
},
{
"entropy": 8.801108264923096,
"epoch": 0.1839880656389856,
"grad_norm": 0.953125,
"learning_rate": 9.2e-05,
"loss": 8.3868,
"mean_token_accuracy": 0.06326259039342404,
"num_tokens": 367297.0,
"step": 185
},
{
"entropy": 8.760680103302002,
"epoch": 0.18896071606166087,
"grad_norm": 0.87109375,
"learning_rate": 9.45e-05,
"loss": 8.3884,
"mean_token_accuracy": 0.058847269788384435,
"num_tokens": 376880.0,
"step": 190
},
{
"entropy": 8.77523078918457,
"epoch": 0.19393336648433615,
"grad_norm": 0.81640625,
"learning_rate": 9.7e-05,
"loss": 8.3879,
"mean_token_accuracy": 0.06135675199329853,
"num_tokens": 386806.0,
"step": 195
},
{
"entropy": 8.665618419647217,
"epoch": 0.19890601690701143,
"grad_norm": 1.203125,
"learning_rate": 9.95e-05,
"loss": 8.349,
"mean_token_accuracy": 0.06269419603049756,
"num_tokens": 396161.0,
"step": 200
},
{
"entropy": 8.676361560821533,
"epoch": 0.20387866732968674,
"grad_norm": 1.03125,
"learning_rate": 0.000102,
"loss": 8.3986,
"mean_token_accuracy": 0.06113377064466476,
"num_tokens": 405643.0,
"step": 205
},
{
"entropy": 8.612883186340332,
"epoch": 0.20885131775236201,
"grad_norm": 0.9375,
"learning_rate": 0.00010449999999999999,
"loss": 8.3532,
"mean_token_accuracy": 0.06992297060787678,
"num_tokens": 414367.0,
"step": 210
},
{
"entropy": 8.678384780883789,
"epoch": 0.2138239681750373,
"grad_norm": 0.90234375,
"learning_rate": 0.000107,
"loss": 8.4084,
"mean_token_accuracy": 0.06210144981741905,
"num_tokens": 424430.0,
"step": 215
},
{
"entropy": 8.648359966278075,
"epoch": 0.21879661859771257,
"grad_norm": 0.98828125,
"learning_rate": 0.0001095,
"loss": 8.3776,
"mean_token_accuracy": 0.06364397667348384,
"num_tokens": 435176.0,
"step": 220
},
{
"entropy": 8.576762390136718,
"epoch": 0.22376926902038788,
"grad_norm": 0.86328125,
"learning_rate": 0.000112,
"loss": 8.3963,
"mean_token_accuracy": 0.06432987973093987,
"num_tokens": 445611.0,
"step": 225
},
{
"entropy": 8.57848825454712,
"epoch": 0.22874191944306316,
"grad_norm": 0.984375,
"learning_rate": 0.0001145,
"loss": 8.3247,
"mean_token_accuracy": 0.0661028627306223,
"num_tokens": 455143.0,
"step": 230
},
{
"entropy": 8.5846848487854,
"epoch": 0.23371456986573844,
"grad_norm": 1.7421875,
"learning_rate": 0.00011700000000000001,
"loss": 8.4127,
"mean_token_accuracy": 0.06825519613921642,
"num_tokens": 465507.0,
"step": 235
},
{
"entropy": 8.642884063720704,
"epoch": 0.23868722028841372,
"grad_norm": 0.95703125,
"learning_rate": 0.00011949999999999999,
"loss": 8.413,
"mean_token_accuracy": 0.06387550197541714,
"num_tokens": 476152.0,
"step": 240
},
{
"entropy": 8.646145343780518,
"epoch": 0.24365987071108902,
"grad_norm": 0.92578125,
"learning_rate": 0.000122,
"loss": 8.2597,
"mean_token_accuracy": 0.06847746260464191,
"num_tokens": 486341.0,
"step": 245
},
{
"entropy": 8.544342041015625,
"epoch": 0.2486325211337643,
"grad_norm": 0.84375,
"learning_rate": 0.0001245,
"loss": 8.218,
"mean_token_accuracy": 0.06961140334606171,
"num_tokens": 495915.0,
"step": 250
},
{
"entropy": 8.424121761322022,
"epoch": 0.2536051715564396,
"grad_norm": 0.84375,
"learning_rate": 0.000127,
"loss": 8.2949,
"mean_token_accuracy": 0.07351046912372113,
"num_tokens": 505666.0,
"step": 255
},
{
"entropy": 8.567151832580567,
"epoch": 0.2585778219791149,
"grad_norm": 0.9609375,
"learning_rate": 0.0001295,
"loss": 8.2218,
"mean_token_accuracy": 0.06917047686874866,
"num_tokens": 515411.0,
"step": 260
},
{
"entropy": 8.503114604949952,
"epoch": 0.26355047240179014,
"grad_norm": 1.015625,
"learning_rate": 0.000132,
"loss": 8.2975,
"mean_token_accuracy": 0.06331334635615349,
"num_tokens": 525018.0,
"step": 265
},
{
"entropy": 8.563315486907959,
"epoch": 0.26852312282446544,
"grad_norm": 1.5390625,
"learning_rate": 0.00013450000000000002,
"loss": 8.255,
"mean_token_accuracy": 0.06826438158750534,
"num_tokens": 535056.0,
"step": 270
},
{
"entropy": 8.556770515441894,
"epoch": 0.27349577324714075,
"grad_norm": 0.97265625,
"learning_rate": 0.00013700000000000002,
"loss": 8.2081,
"mean_token_accuracy": 0.06950257569551468,
"num_tokens": 545115.0,
"step": 275
},
{
"entropy": 8.51298542022705,
"epoch": 0.278468423669816,
"grad_norm": 2.4375,
"learning_rate": 0.0001395,
"loss": 8.2583,
"mean_token_accuracy": 0.06896053329110145,
"num_tokens": 554429.0,
"step": 280
},
{
"entropy": 8.453236293792724,
"epoch": 0.2834410740924913,
"grad_norm": 1.1015625,
"learning_rate": 0.00014199999999999998,
"loss": 8.2007,
"mean_token_accuracy": 0.07584292404353618,
"num_tokens": 563357.0,
"step": 285
},
{
"entropy": 8.559890079498292,
"epoch": 0.28841372451516656,
"grad_norm": 1.09375,
"learning_rate": 0.0001445,
"loss": 8.3728,
"mean_token_accuracy": 0.06644579172134399,
"num_tokens": 574480.0,
"step": 290
},
{
"entropy": 8.503904342651367,
"epoch": 0.29338637493784187,
"grad_norm": 1.484375,
"learning_rate": 0.000147,
"loss": 8.2295,
"mean_token_accuracy": 0.06927415728569031,
"num_tokens": 583722.0,
"step": 295
},
{
"entropy": 8.454716300964355,
"epoch": 0.29835902536051717,
"grad_norm": 1.0703125,
"learning_rate": 0.0001495,
"loss": 8.1738,
"mean_token_accuracy": 0.07493777833878994,
"num_tokens": 593644.0,
"step": 300
},
{
"entropy": 8.49923391342163,
"epoch": 0.3033316757831924,
"grad_norm": 1.109375,
"learning_rate": 0.000152,
"loss": 8.2147,
"mean_token_accuracy": 0.07160871513187886,
"num_tokens": 603547.0,
"step": 305
},
{
"entropy": 8.44572868347168,
"epoch": 0.30830432620586773,
"grad_norm": 1.125,
"learning_rate": 0.00015450000000000001,
"loss": 8.2459,
"mean_token_accuracy": 0.06570485346019268,
"num_tokens": 613880.0,
"step": 310
},
{
"entropy": 8.436261844635009,
"epoch": 0.31327697662854304,
"grad_norm": 1.609375,
"learning_rate": 0.000157,
"loss": 8.2238,
"mean_token_accuracy": 0.07062076106667518,
"num_tokens": 624213.0,
"step": 315
},
{
"entropy": 8.490541934967041,
"epoch": 0.3182496270512183,
"grad_norm": 1.0546875,
"learning_rate": 0.0001595,
"loss": 8.2437,
"mean_token_accuracy": 0.07502996549010277,
"num_tokens": 634350.0,
"step": 320
},
{
"entropy": 8.546083354949952,
"epoch": 0.3232222774738936,
"grad_norm": 1.4296875,
"learning_rate": 0.000162,
"loss": 8.2438,
"mean_token_accuracy": 0.07064831554889679,
"num_tokens": 644040.0,
"step": 325
},
{
"entropy": 8.492809295654297,
"epoch": 0.32819492789656884,
"grad_norm": 1.0546875,
"learning_rate": 0.00016450000000000001,
"loss": 8.2033,
"mean_token_accuracy": 0.07537714764475822,
"num_tokens": 652915.0,
"step": 330
},
{
"entropy": 8.426844501495362,
"epoch": 0.33316757831924415,
"grad_norm": 0.859375,
"learning_rate": 0.00016700000000000002,
"loss": 8.1858,
"mean_token_accuracy": 0.06924654506146907,
"num_tokens": 662665.0,
"step": 335
},
{
"entropy": 8.427862071990967,
"epoch": 0.33814022874191946,
"grad_norm": 1.015625,
"learning_rate": 0.00016950000000000003,
"loss": 8.1955,
"mean_token_accuracy": 0.07384301200509072,
"num_tokens": 671667.0,
"step": 340
},
{
"entropy": 8.440595722198486,
"epoch": 0.3431128791645947,
"grad_norm": 1.125,
"learning_rate": 0.00017199999999999998,
"loss": 8.2177,
"mean_token_accuracy": 0.06726174391806125,
"num_tokens": 680432.0,
"step": 345
},
{
"entropy": 8.477052116394043,
"epoch": 0.34808552958727,
"grad_norm": 0.8828125,
"learning_rate": 0.00017449999999999999,
"loss": 8.2011,
"mean_token_accuracy": 0.0729417447000742,
"num_tokens": 690184.0,
"step": 350
},
{
"entropy": 8.372831344604492,
"epoch": 0.3530581800099453,
"grad_norm": 1.296875,
"learning_rate": 0.000177,
"loss": 8.092,
"mean_token_accuracy": 0.07851836830377579,
"num_tokens": 699147.0,
"step": 355
},
{
"entropy": 8.329792785644532,
"epoch": 0.3580308304326206,
"grad_norm": 1.0,
"learning_rate": 0.0001795,
"loss": 8.1201,
"mean_token_accuracy": 0.0765114389359951,
"num_tokens": 708696.0,
"step": 360
},
{
"entropy": 8.382805728912354,
"epoch": 0.3630034808552959,
"grad_norm": 1.078125,
"learning_rate": 0.000182,
"loss": 8.0897,
"mean_token_accuracy": 0.07439367957413197,
"num_tokens": 718438.0,
"step": 365
},
{
"entropy": 8.393447971343994,
"epoch": 0.3679761312779712,
"grad_norm": 1.0625,
"learning_rate": 0.0001845,
"loss": 8.2278,
"mean_token_accuracy": 0.0699504878371954,
"num_tokens": 728157.0,
"step": 370
},
{
"entropy": 8.350840854644776,
"epoch": 0.37294878170064644,
"grad_norm": 1.390625,
"learning_rate": 0.000187,
"loss": 8.1211,
"mean_token_accuracy": 0.07720480635762214,
"num_tokens": 737007.0,
"step": 375
},
{
"entropy": 8.357489013671875,
"epoch": 0.37792143212332174,
"grad_norm": 1.1875,
"learning_rate": 0.0001895,
"loss": 8.0726,
"mean_token_accuracy": 0.07758348956704139,
"num_tokens": 746873.0,
"step": 380
},
{
"entropy": 8.33244686126709,
"epoch": 0.382894082545997,
"grad_norm": 1.03125,
"learning_rate": 0.000192,
"loss": 7.9852,
"mean_token_accuracy": 0.08200759738683701,
"num_tokens": 755486.0,
"step": 385
},
{
"entropy": 8.285024738311767,
"epoch": 0.3878667329686723,
"grad_norm": 1.234375,
"learning_rate": 0.0001945,
"loss": 8.0944,
"mean_token_accuracy": 0.07533938214182853,
"num_tokens": 765481.0,
"step": 390
},
{
"entropy": 8.345268440246581,
"epoch": 0.3928393833913476,
"grad_norm": 1.1640625,
"learning_rate": 0.00019700000000000002,
"loss": 8.0464,
"mean_token_accuracy": 0.07712380215525627,
"num_tokens": 774335.0,
"step": 395
},
{
"entropy": 8.38193359375,
"epoch": 0.39781203381402286,
"grad_norm": 1.1875,
"learning_rate": 0.00019950000000000002,
"loss": 8.1296,
"mean_token_accuracy": 0.07453210800886154,
"num_tokens": 784307.0,
"step": 400
},
{
"entropy": 8.411716270446778,
"epoch": 0.40278468423669817,
"grad_norm": 0.96484375,
"learning_rate": 0.000202,
"loss": 8.152,
"mean_token_accuracy": 0.07242761701345443,
"num_tokens": 795058.0,
"step": 405
},
{
"entropy": 8.320662117004394,
"epoch": 0.40775733465937347,
"grad_norm": 1.6953125,
"learning_rate": 0.00020449999999999998,
"loss": 8.0567,
"mean_token_accuracy": 0.07517294511198998,
"num_tokens": 805687.0,
"step": 410
},
{
"entropy": 8.350330257415772,
"epoch": 0.4127299850820487,
"grad_norm": 0.99609375,
"learning_rate": 0.000207,
"loss": 8.0842,
"mean_token_accuracy": 0.07857826426625251,
"num_tokens": 815138.0,
"step": 415
},
{
"entropy": 8.33483533859253,
"epoch": 0.41770263550472403,
"grad_norm": 1.1796875,
"learning_rate": 0.0002095,
"loss": 8.0588,
"mean_token_accuracy": 0.07447804734110833,
"num_tokens": 825009.0,
"step": 420
},
{
"entropy": 8.324224948883057,
"epoch": 0.4226752859273993,
"grad_norm": 1.96875,
"learning_rate": 0.000212,
"loss": 8.0464,
"mean_token_accuracy": 0.07719542868435383,
"num_tokens": 834708.0,
"step": 425
},
{
"entropy": 8.2837308883667,
"epoch": 0.4276479363500746,
"grad_norm": 1.2421875,
"learning_rate": 0.0002145,
"loss": 8.0994,
"mean_token_accuracy": 0.07782356329262256,
"num_tokens": 843888.0,
"step": 430
},
{
"entropy": 8.280590438842774,
"epoch": 0.4326205867727499,
"grad_norm": 1.109375,
"learning_rate": 0.00021700000000000002,
"loss": 8.0652,
"mean_token_accuracy": 0.07787982597947121,
"num_tokens": 853057.0,
"step": 435
},
{
"entropy": 8.313239860534669,
"epoch": 0.43759323719542514,
"grad_norm": 1.546875,
"learning_rate": 0.0002195,
"loss": 8.1067,
"mean_token_accuracy": 0.0732115276157856,
"num_tokens": 862701.0,
"step": 440
},
{
"entropy": 8.202727890014648,
"epoch": 0.44256588761810045,
"grad_norm": 1.140625,
"learning_rate": 0.000222,
"loss": 8.0037,
"mean_token_accuracy": 0.07995785847306251,
"num_tokens": 872129.0,
"step": 445
},
{
"entropy": 8.249781703948974,
"epoch": 0.44753853804077576,
"grad_norm": 1.0078125,
"learning_rate": 0.0002245,
"loss": 8.0233,
"mean_token_accuracy": 0.0809432752430439,
"num_tokens": 883166.0,
"step": 450
},
{
"entropy": 8.261981916427612,
"epoch": 0.452511188463451,
"grad_norm": 1.3828125,
"learning_rate": 0.00022700000000000002,
"loss": 7.8287,
"mean_token_accuracy": 0.09492504075169564,
"num_tokens": 894159.0,
"step": 455
},
{
"entropy": 8.228504180908203,
"epoch": 0.4574838388861263,
"grad_norm": 0.95703125,
"learning_rate": 0.00022950000000000002,
"loss": 8.0265,
"mean_token_accuracy": 0.07806817516684532,
"num_tokens": 904588.0,
"step": 460
},
{
"entropy": 8.212932109832764,
"epoch": 0.46245648930880157,
"grad_norm": 1.6015625,
"learning_rate": 0.00023200000000000003,
"loss": 8.0094,
"mean_token_accuracy": 0.07935516387224198,
"num_tokens": 915554.0,
"step": 465
},
{
"entropy": 8.191864347457885,
"epoch": 0.4674291397314769,
"grad_norm": 1.1171875,
"learning_rate": 0.00023449999999999998,
"loss": 8.03,
"mean_token_accuracy": 0.07978576868772506,
"num_tokens": 925546.0,
"step": 470
},
{
"entropy": 8.266866683959961,
"epoch": 0.4724017901541522,
"grad_norm": 1.1640625,
"learning_rate": 0.000237,
"loss": 7.9894,
"mean_token_accuracy": 0.082433320581913,
"num_tokens": 936615.0,
"step": 475
},
{
"entropy": 8.207193756103516,
"epoch": 0.47737444057682743,
"grad_norm": 1.3671875,
"learning_rate": 0.0002395,
"loss": 8.023,
"mean_token_accuracy": 0.08058698140084744,
"num_tokens": 947374.0,
"step": 480
},
{
"entropy": 8.204709148406982,
"epoch": 0.48234709099950274,
"grad_norm": 1.0859375,
"learning_rate": 0.000242,
"loss": 7.955,
"mean_token_accuracy": 0.08018621355295182,
"num_tokens": 957772.0,
"step": 485
},
{
"entropy": 8.176499080657958,
"epoch": 0.48731974142217804,
"grad_norm": 1.0625,
"learning_rate": 0.0002445,
"loss": 8.0309,
"mean_token_accuracy": 0.08311204165220261,
"num_tokens": 967724.0,
"step": 490
},
{
"entropy": 8.139270210266114,
"epoch": 0.4922923918448533,
"grad_norm": 1.0859375,
"learning_rate": 0.000247,
"loss": 7.9264,
"mean_token_accuracy": 0.08439484089612961,
"num_tokens": 976497.0,
"step": 495
},
{
"entropy": 8.097719097137452,
"epoch": 0.4972650422675286,
"grad_norm": 1.1640625,
"learning_rate": 0.0002495,
"loss": 7.9124,
"mean_token_accuracy": 0.0810380220413208,
"num_tokens": 986207.0,
"step": 500
},
{
"epoch": 0.4972650422675286,
"eval_entropy": 8.243832484917707,
"eval_loss": 8.042750358581543,
"eval_mean_token_accuracy": 0.07886241382623117,
"eval_num_tokens": 986207.0,
"eval_runtime": 2.7268,
"eval_samples_per_second": 1270.725,
"eval_steps_per_second": 159.162,
"step": 500
},
{
"entropy": 8.125858736038207,
"epoch": 0.5022376926902039,
"grad_norm": 0.97265625,
"learning_rate": 0.000252,
"loss": 7.8922,
"mean_token_accuracy": 0.08268110826611519,
"num_tokens": 995568.0,
"step": 505
},
{
"entropy": 8.18411340713501,
"epoch": 0.5072103431128792,
"grad_norm": 1.15625,
"learning_rate": 0.0002545,
"loss": 7.9499,
"mean_token_accuracy": 0.08079156577587128,
"num_tokens": 1005829.0,
"step": 510
},
{
"entropy": 8.089786481857299,
"epoch": 0.5121829935355544,
"grad_norm": 1.40625,
"learning_rate": 0.000257,
"loss": 7.9077,
"mean_token_accuracy": 0.08108234629034997,
"num_tokens": 1015290.0,
"step": 515
},
{
"entropy": 8.183901023864745,
"epoch": 0.5171556439582298,
"grad_norm": 1.21875,
"learning_rate": 0.0002595,
"loss": 7.872,
"mean_token_accuracy": 0.08664105087518692,
"num_tokens": 1024479.0,
"step": 520
},
{
"entropy": 7.893880844116211,
"epoch": 0.522128294380905,
"grad_norm": 2.078125,
"learning_rate": 0.000262,
"loss": 7.7961,
"mean_token_accuracy": 0.08195992782711983,
"num_tokens": 1034092.0,
"step": 525
},
{
"entropy": 8.25371503829956,
"epoch": 0.5271009448035803,
"grad_norm": 1.140625,
"learning_rate": 0.00026450000000000003,
"loss": 8.0162,
"mean_token_accuracy": 0.08469353094696999,
"num_tokens": 1043844.0,
"step": 530
},
{
"entropy": 8.041243362426759,
"epoch": 0.5320735952262556,
"grad_norm": 1.3671875,
"learning_rate": 0.00026700000000000004,
"loss": 7.8648,
"mean_token_accuracy": 0.08864305019378663,
"num_tokens": 1052646.0,
"step": 535
},
{
"entropy": 8.153491306304932,
"epoch": 0.5370462456489309,
"grad_norm": 1.078125,
"learning_rate": 0.00026950000000000005,
"loss": 7.9181,
"mean_token_accuracy": 0.08187793269753456,
"num_tokens": 1062345.0,
"step": 540
},
{
"entropy": 8.129785537719727,
"epoch": 0.5420188960716061,
"grad_norm": 1.1328125,
"learning_rate": 0.00027200000000000005,
"loss": 7.9003,
"mean_token_accuracy": 0.079953121393919,
"num_tokens": 1071116.0,
"step": 545
},
{
"entropy": 8.065836000442506,
"epoch": 0.5469915464942815,
"grad_norm": 1.078125,
"learning_rate": 0.0002745,
"loss": 7.8489,
"mean_token_accuracy": 0.08677002638578415,
"num_tokens": 1080562.0,
"step": 550
},
{
"entropy": 8.08962984085083,
"epoch": 0.5519641969169568,
"grad_norm": 1.09375,
"learning_rate": 0.000277,
"loss": 7.9092,
"mean_token_accuracy": 0.08326611518859864,
"num_tokens": 1091944.0,
"step": 555
},
{
"entropy": 8.061810493469238,
"epoch": 0.556936847339632,
"grad_norm": 1.109375,
"learning_rate": 0.0002795,
"loss": 7.8361,
"mean_token_accuracy": 0.08510831445455551,
"num_tokens": 1101412.0,
"step": 560
},
{
"entropy": 8.109366941452027,
"epoch": 0.5619094977623074,
"grad_norm": 1.078125,
"learning_rate": 0.00028199999999999997,
"loss": 7.9156,
"mean_token_accuracy": 0.08460526764392853,
"num_tokens": 1110652.0,
"step": 565
},
{
"entropy": 8.040374612808227,
"epoch": 0.5668821481849826,
"grad_norm": 1.1875,
"learning_rate": 0.0002845,
"loss": 7.8352,
"mean_token_accuracy": 0.08915429562330246,
"num_tokens": 1120420.0,
"step": 570
},
{
"entropy": 8.038780927658081,
"epoch": 0.5718547986076579,
"grad_norm": 1.296875,
"learning_rate": 0.000287,
"loss": 7.8934,
"mean_token_accuracy": 0.0821109727025032,
"num_tokens": 1130046.0,
"step": 575
},
{
"entropy": 8.008350706100464,
"epoch": 0.5768274490303331,
"grad_norm": 1.2578125,
"learning_rate": 0.0002895,
"loss": 7.8662,
"mean_token_accuracy": 0.08604678623378277,
"num_tokens": 1138946.0,
"step": 580
},
{
"entropy": 8.048417139053345,
"epoch": 0.5818000994530085,
"grad_norm": 1.2421875,
"learning_rate": 0.000292,
"loss": 7.6038,
"mean_token_accuracy": 0.09397232159972191,
"num_tokens": 1149440.0,
"step": 585
},
{
"entropy": 8.026965951919555,
"epoch": 0.5867727498756837,
"grad_norm": 1.1015625,
"learning_rate": 0.0002945,
"loss": 7.8972,
"mean_token_accuracy": 0.08445911929011345,
"num_tokens": 1158636.0,
"step": 590
},
{
"entropy": 7.959886932373047,
"epoch": 0.591745400298359,
"grad_norm": 1.109375,
"learning_rate": 0.000297,
"loss": 7.7368,
"mean_token_accuracy": 0.09059922322630883,
"num_tokens": 1167776.0,
"step": 595
},
{
"entropy": 7.994223403930664,
"epoch": 0.5967180507210343,
"grad_norm": 1.03125,
"learning_rate": 0.0002995,
"loss": 7.719,
"mean_token_accuracy": 0.08690877109766007,
"num_tokens": 1177919.0,
"step": 600
},
{
"entropy": 7.917198991775512,
"epoch": 0.6016907011437096,
"grad_norm": 1.046875,
"learning_rate": 0.000302,
"loss": 7.7869,
"mean_token_accuracy": 0.0975386694073677,
"num_tokens": 1187363.0,
"step": 605
},
{
"entropy": 8.191771316528321,
"epoch": 0.6066633515663848,
"grad_norm": 1.1953125,
"learning_rate": 0.0003045,
"loss": 7.9566,
"mean_token_accuracy": 0.08074369877576829,
"num_tokens": 1197329.0,
"step": 610
},
{
"entropy": 7.946739959716797,
"epoch": 0.6116360019890602,
"grad_norm": 1.140625,
"learning_rate": 0.000307,
"loss": 7.8811,
"mean_token_accuracy": 0.08208825774490833,
"num_tokens": 1207408.0,
"step": 615
},
{
"entropy": 8.007125759124756,
"epoch": 0.6166086524117355,
"grad_norm": 1.2421875,
"learning_rate": 0.0003095,
"loss": 7.8551,
"mean_token_accuracy": 0.08459711819887161,
"num_tokens": 1219258.0,
"step": 620
},
{
"entropy": 8.056624031066894,
"epoch": 0.6215813028344107,
"grad_norm": 1.1875,
"learning_rate": 0.000312,
"loss": 7.8472,
"mean_token_accuracy": 0.08643570393323899,
"num_tokens": 1228561.0,
"step": 625
},
{
"entropy": 7.8858905792236325,
"epoch": 0.6265539532570861,
"grad_norm": 0.97265625,
"learning_rate": 0.0003145,
"loss": 7.7861,
"mean_token_accuracy": 0.0854301854968071,
"num_tokens": 1239665.0,
"step": 630
},
{
"entropy": 7.968693017959595,
"epoch": 0.6315266036797613,
"grad_norm": 1.046875,
"learning_rate": 0.000317,
"loss": 7.7938,
"mean_token_accuracy": 0.08755213171243667,
"num_tokens": 1249490.0,
"step": 635
},
{
"entropy": 7.931836175918579,
"epoch": 0.6364992541024366,
"grad_norm": 1.15625,
"learning_rate": 0.0003195,
"loss": 7.8103,
"mean_token_accuracy": 0.08826896399259568,
"num_tokens": 1259328.0,
"step": 640
},
{
"entropy": 8.089329528808594,
"epoch": 0.6414719045251119,
"grad_norm": 1.1328125,
"learning_rate": 0.000322,
"loss": 7.7524,
"mean_token_accuracy": 0.08410023339092731,
"num_tokens": 1269523.0,
"step": 645
},
{
"entropy": 7.8148744106292725,
"epoch": 0.6464445549477872,
"grad_norm": 1.1953125,
"learning_rate": 0.00032450000000000003,
"loss": 7.8255,
"mean_token_accuracy": 0.08130960762500763,
"num_tokens": 1277961.0,
"step": 650
},
{
"entropy": 8.05358533859253,
"epoch": 0.6514172053704624,
"grad_norm": 1.140625,
"learning_rate": 0.00032700000000000003,
"loss": 7.7896,
"mean_token_accuracy": 0.08684217259287834,
"num_tokens": 1288228.0,
"step": 655
},
{
"entropy": 7.969742012023926,
"epoch": 0.6563898557931377,
"grad_norm": 0.94140625,
"learning_rate": 0.00032950000000000004,
"loss": 7.7255,
"mean_token_accuracy": 0.09051954224705697,
"num_tokens": 1299636.0,
"step": 660
},
{
"entropy": 7.837801790237426,
"epoch": 0.661362506215813,
"grad_norm": 1.109375,
"learning_rate": 0.00033200000000000005,
"loss": 7.7596,
"mean_token_accuracy": 0.08984493017196656,
"num_tokens": 1310035.0,
"step": 665
},
{
"entropy": 7.899158906936646,
"epoch": 0.6663351566384883,
"grad_norm": 1.0078125,
"learning_rate": 0.00033450000000000005,
"loss": 7.7654,
"mean_token_accuracy": 0.09294063225388527,
"num_tokens": 1320751.0,
"step": 670
},
{
"entropy": 7.864286184310913,
"epoch": 0.6713078070611636,
"grad_norm": 1.546875,
"learning_rate": 0.000337,
"loss": 7.6445,
"mean_token_accuracy": 0.0892346404492855,
"num_tokens": 1331023.0,
"step": 675
},
{
"entropy": 7.845469665527344,
"epoch": 0.6762804574838389,
"grad_norm": 1.1875,
"learning_rate": 0.0003395,
"loss": 7.7513,
"mean_token_accuracy": 0.0856916643679142,
"num_tokens": 1341814.0,
"step": 680
},
{
"entropy": 7.911331701278686,
"epoch": 0.6812531079065142,
"grad_norm": 1.046875,
"learning_rate": 0.000342,
"loss": 7.6408,
"mean_token_accuracy": 0.0899275153875351,
"num_tokens": 1351026.0,
"step": 685
},
{
"entropy": 7.953000831604004,
"epoch": 0.6862257583291894,
"grad_norm": 1.2421875,
"learning_rate": 0.00034449999999999997,
"loss": 7.855,
"mean_token_accuracy": 0.07497294135391712,
"num_tokens": 1361004.0,
"step": 690
},
{
"entropy": 7.813990640640259,
"epoch": 0.6911984087518648,
"grad_norm": 1.3671875,
"learning_rate": 0.000347,
"loss": 7.7092,
"mean_token_accuracy": 0.08678068146109581,
"num_tokens": 1370729.0,
"step": 695
},
{
"entropy": 7.900962209701538,
"epoch": 0.69617105917454,
"grad_norm": 1.234375,
"learning_rate": 0.0003495,
"loss": 7.6649,
"mean_token_accuracy": 0.08870695382356644,
"num_tokens": 1379525.0,
"step": 700
},
{
"entropy": 7.831973600387573,
"epoch": 0.7011437095972153,
"grad_norm": 1.15625,
"learning_rate": 0.000352,
"loss": 7.7323,
"mean_token_accuracy": 0.09046533703804016,
"num_tokens": 1389863.0,
"step": 705
},
{
"entropy": 7.800638961791992,
"epoch": 0.7061163600198906,
"grad_norm": 1.421875,
"learning_rate": 0.0003545,
"loss": 7.8508,
"mean_token_accuracy": 0.08676985278725624,
"num_tokens": 1399468.0,
"step": 710
},
{
"entropy": 7.86581072807312,
"epoch": 0.7110890104425659,
"grad_norm": 0.953125,
"learning_rate": 0.000357,
"loss": 7.6529,
"mean_token_accuracy": 0.09256174638867379,
"num_tokens": 1408785.0,
"step": 715
},
{
"entropy": 7.747158002853394,
"epoch": 0.7160616608652411,
"grad_norm": 1.2890625,
"learning_rate": 0.0003595,
"loss": 7.6904,
"mean_token_accuracy": 0.09273576214909554,
"num_tokens": 1418666.0,
"step": 720
},
{
"entropy": 7.985917377471924,
"epoch": 0.7210343112879165,
"grad_norm": 1.1171875,
"learning_rate": 0.000362,
"loss": 7.8001,
"mean_token_accuracy": 0.08961994498968125,
"num_tokens": 1428699.0,
"step": 725
},
{
"entropy": 7.853034973144531,
"epoch": 0.7260069617105918,
"grad_norm": 1.1640625,
"learning_rate": 0.0003645,
"loss": 7.6893,
"mean_token_accuracy": 0.09192831739783287,
"num_tokens": 1437354.0,
"step": 730
},
{
"entropy": 7.813079929351806,
"epoch": 0.730979612133267,
"grad_norm": 1.203125,
"learning_rate": 0.000367,
"loss": 7.708,
"mean_token_accuracy": 0.08974602445960045,
"num_tokens": 1447327.0,
"step": 735
},
{
"entropy": 7.871495628356934,
"epoch": 0.7359522625559424,
"grad_norm": 1.015625,
"learning_rate": 0.0003695,
"loss": 7.7563,
"mean_token_accuracy": 0.09072970226407051,
"num_tokens": 1457793.0,
"step": 740
},
{
"entropy": 7.713533926010132,
"epoch": 0.7409249129786176,
"grad_norm": 1.203125,
"learning_rate": 0.000372,
"loss": 7.6719,
"mean_token_accuracy": 0.08988175988197326,
"num_tokens": 1467143.0,
"step": 745
},
{
"entropy": 7.857532787322998,
"epoch": 0.7458975634012929,
"grad_norm": 1.2890625,
"learning_rate": 0.0003745,
"loss": 7.7148,
"mean_token_accuracy": 0.08926208540797234,
"num_tokens": 1476422.0,
"step": 750
},
{
"entropy": 7.813498783111572,
"epoch": 0.7508702138239681,
"grad_norm": 1.09375,
"learning_rate": 0.000377,
"loss": 7.656,
"mean_token_accuracy": 0.09503143355250358,
"num_tokens": 1486253.0,
"step": 755
},
{
"entropy": 7.716996860504151,
"epoch": 0.7558428642466435,
"grad_norm": 1.21875,
"learning_rate": 0.0003795,
"loss": 7.6467,
"mean_token_accuracy": 0.09296249076724053,
"num_tokens": 1496080.0,
"step": 760
},
{
"entropy": 7.760905981063843,
"epoch": 0.7608155146693187,
"grad_norm": 1.03125,
"learning_rate": 0.000382,
"loss": 7.6075,
"mean_token_accuracy": 0.09641017168760299,
"num_tokens": 1505532.0,
"step": 765
},
{
"entropy": 7.783897256851196,
"epoch": 0.765788165091994,
"grad_norm": 1.046875,
"learning_rate": 0.0003845,
"loss": 7.5671,
"mean_token_accuracy": 0.0936719074845314,
"num_tokens": 1515705.0,
"step": 770
},
{
"entropy": 7.653465604782104,
"epoch": 0.7707608155146694,
"grad_norm": 1.03125,
"learning_rate": 0.00038700000000000003,
"loss": 7.6041,
"mean_token_accuracy": 0.09394907802343369,
"num_tokens": 1525651.0,
"step": 775
},
{
"entropy": 7.739487743377685,
"epoch": 0.7757334659373446,
"grad_norm": 1.125,
"learning_rate": 0.00038950000000000003,
"loss": 7.5965,
"mean_token_accuracy": 0.09230046570301056,
"num_tokens": 1534576.0,
"step": 780
},
{
"entropy": 7.608673191070556,
"epoch": 0.7807061163600199,
"grad_norm": 1.8046875,
"learning_rate": 0.00039200000000000004,
"loss": 7.4234,
"mean_token_accuracy": 0.1033897265791893,
"num_tokens": 1544765.0,
"step": 785
},
{
"entropy": 7.687298154830932,
"epoch": 0.7856787667826952,
"grad_norm": 1.109375,
"learning_rate": 0.00039450000000000005,
"loss": 7.5766,
"mean_token_accuracy": 0.09803976342082024,
"num_tokens": 1553636.0,
"step": 790
},
{
"entropy": 7.740061044692993,
"epoch": 0.7906514172053705,
"grad_norm": 1.03125,
"learning_rate": 0.00039700000000000005,
"loss": 7.5557,
"mean_token_accuracy": 0.09136517718434334,
"num_tokens": 1563356.0,
"step": 795
},
{
"entropy": 7.585089302062988,
"epoch": 0.7956240676280457,
"grad_norm": 1.15625,
"learning_rate": 0.0003995,
"loss": 7.5708,
"mean_token_accuracy": 0.09769159182906151,
"num_tokens": 1572508.0,
"step": 800
},
{
"entropy": 7.73101921081543,
"epoch": 0.8005967180507211,
"grad_norm": 1.0078125,
"learning_rate": 0.000402,
"loss": 7.6024,
"mean_token_accuracy": 0.094760861992836,
"num_tokens": 1581583.0,
"step": 805
},
{
"entropy": 7.694145679473877,
"epoch": 0.8055693684733963,
"grad_norm": 1.1015625,
"learning_rate": 0.0004045,
"loss": 7.5936,
"mean_token_accuracy": 0.09348807856440544,
"num_tokens": 1592114.0,
"step": 810
},
{
"entropy": 7.6239536762237545,
"epoch": 0.8105420188960716,
"grad_norm": 1.1953125,
"learning_rate": 0.00040699999999999997,
"loss": 7.4752,
"mean_token_accuracy": 0.10072982311248779,
"num_tokens": 1602417.0,
"step": 815
},
{
"entropy": 7.628807640075683,
"epoch": 0.8155146693187469,
"grad_norm": 1.34375,
"learning_rate": 0.0004095,
"loss": 7.5729,
"mean_token_accuracy": 0.09305822476744652,
"num_tokens": 1611753.0,
"step": 820
},
{
"entropy": 7.685155630111694,
"epoch": 0.8204873197414222,
"grad_norm": 1.171875,
"learning_rate": 0.000412,
"loss": 7.6614,
"mean_token_accuracy": 0.09417398199439049,
"num_tokens": 1621959.0,
"step": 825
},
{
"entropy": 7.682513904571533,
"epoch": 0.8254599701640974,
"grad_norm": 1.0390625,
"learning_rate": 0.0004145,
"loss": 7.5577,
"mean_token_accuracy": 0.09433014839887618,
"num_tokens": 1630909.0,
"step": 830
},
{
"entropy": 7.712544775009155,
"epoch": 0.8304326205867727,
"grad_norm": 0.95703125,
"learning_rate": 0.000417,
"loss": 7.5309,
"mean_token_accuracy": 0.09536156728863716,
"num_tokens": 1641316.0,
"step": 835
},
{
"entropy": 7.516957235336304,
"epoch": 0.8354052710094481,
"grad_norm": 1.1953125,
"learning_rate": 0.0004195,
"loss": 7.4478,
"mean_token_accuracy": 0.10105396658182145,
"num_tokens": 1651506.0,
"step": 840
},
{
"entropy": 7.540144920349121,
"epoch": 0.8403779214321233,
"grad_norm": 1.1171875,
"learning_rate": 0.000422,
"loss": 7.4276,
"mean_token_accuracy": 0.0987382858991623,
"num_tokens": 1659871.0,
"step": 845
},
{
"entropy": 7.714676856994629,
"epoch": 0.8453505718547986,
"grad_norm": 1.140625,
"learning_rate": 0.0004245,
"loss": 7.6448,
"mean_token_accuracy": 0.0904959000647068,
"num_tokens": 1670178.0,
"step": 850
},
{
"entropy": 7.6358119487762455,
"epoch": 0.8503232222774739,
"grad_norm": 1.1171875,
"learning_rate": 0.000427,
"loss": 7.5413,
"mean_token_accuracy": 0.09731325581669807,
"num_tokens": 1680131.0,
"step": 855
},
{
"entropy": 7.569293355941772,
"epoch": 0.8552958727001492,
"grad_norm": 1.1953125,
"learning_rate": 0.0004295,
"loss": 7.4575,
"mean_token_accuracy": 0.10155909061431885,
"num_tokens": 1689568.0,
"step": 860
},
{
"entropy": 7.681883001327515,
"epoch": 0.8602685231228244,
"grad_norm": 1.109375,
"learning_rate": 0.000432,
"loss": 7.5101,
"mean_token_accuracy": 0.1035876877605915,
"num_tokens": 1698039.0,
"step": 865
},
{
"entropy": 7.606202030181885,
"epoch": 0.8652411735454998,
"grad_norm": 1.078125,
"learning_rate": 0.0004345,
"loss": 7.5097,
"mean_token_accuracy": 0.10067008286714554,
"num_tokens": 1707659.0,
"step": 870
},
{
"entropy": 7.514298009872436,
"epoch": 0.870213823968175,
"grad_norm": 1.0859375,
"learning_rate": 0.000437,
"loss": 7.4873,
"mean_token_accuracy": 0.10052895620465278,
"num_tokens": 1716952.0,
"step": 875
},
{
"entropy": 7.705683279037475,
"epoch": 0.8751864743908503,
"grad_norm": 1.1796875,
"learning_rate": 0.0004395,
"loss": 7.5457,
"mean_token_accuracy": 0.0922851774841547,
"num_tokens": 1726323.0,
"step": 880
},
{
"entropy": 7.634090089797974,
"epoch": 0.8801591248135257,
"grad_norm": 1.09375,
"learning_rate": 0.000442,
"loss": 7.5293,
"mean_token_accuracy": 0.0906990833580494,
"num_tokens": 1736265.0,
"step": 885
},
{
"entropy": 7.553630065917969,
"epoch": 0.8851317752362009,
"grad_norm": 1.0625,
"learning_rate": 0.0004445,
"loss": 7.4983,
"mean_token_accuracy": 0.0946983590722084,
"num_tokens": 1745150.0,
"step": 890
},
{
"entropy": 7.669640970230103,
"epoch": 0.8901044256588762,
"grad_norm": 0.9609375,
"learning_rate": 0.000447,
"loss": 7.5938,
"mean_token_accuracy": 0.09150914251804351,
"num_tokens": 1754641.0,
"step": 895
},
{
"entropy": 7.481142282485962,
"epoch": 0.8950770760815515,
"grad_norm": 1.1171875,
"learning_rate": 0.00044950000000000003,
"loss": 7.4017,
"mean_token_accuracy": 0.10149217993021012,
"num_tokens": 1763468.0,
"step": 900
},
{
"entropy": 7.524608373641968,
"epoch": 0.9000497265042268,
"grad_norm": 1.015625,
"learning_rate": 0.00045200000000000004,
"loss": 7.5274,
"mean_token_accuracy": 0.09709789380431175,
"num_tokens": 1774845.0,
"step": 905
},
{
"entropy": 7.516222858428955,
"epoch": 0.905022376926902,
"grad_norm": 0.99609375,
"learning_rate": 0.00045450000000000004,
"loss": 7.4645,
"mean_token_accuracy": 0.10099468976259232,
"num_tokens": 1784606.0,
"step": 910
},
{
"entropy": 7.644159364700317,
"epoch": 0.9099950273495774,
"grad_norm": 1.09375,
"learning_rate": 0.00045700000000000005,
"loss": 7.4937,
"mean_token_accuracy": 0.09957759752869606,
"num_tokens": 1793893.0,
"step": 915
},
{
"entropy": 7.51349983215332,
"epoch": 0.9149676777722526,
"grad_norm": 1.03125,
"learning_rate": 0.00045950000000000006,
"loss": 7.4704,
"mean_token_accuracy": 0.10551093220710754,
"num_tokens": 1804095.0,
"step": 920
},
{
"entropy": 7.605443716049194,
"epoch": 0.9199403281949279,
"grad_norm": 1.1328125,
"learning_rate": 0.000462,
"loss": 7.4784,
"mean_token_accuracy": 0.10306655913591385,
"num_tokens": 1813475.0,
"step": 925
},
{
"entropy": 7.447127199172973,
"epoch": 0.9249129786176031,
"grad_norm": 1.1640625,
"learning_rate": 0.0004645,
"loss": 7.3992,
"mean_token_accuracy": 0.10492947027087211,
"num_tokens": 1822779.0,
"step": 930
},
{
"entropy": 7.615777015686035,
"epoch": 0.9298856290402785,
"grad_norm": 1.1796875,
"learning_rate": 0.000467,
"loss": 7.4335,
"mean_token_accuracy": 0.09923516064882279,
"num_tokens": 1832239.0,
"step": 935
},
{
"entropy": 7.387023115158081,
"epoch": 0.9348582794629537,
"grad_norm": 1.4375,
"learning_rate": 0.0004695,
"loss": 7.4093,
"mean_token_accuracy": 0.09377154260873795,
"num_tokens": 1843395.0,
"step": 940
},
{
"entropy": 7.5788719177246096,
"epoch": 0.939830929885629,
"grad_norm": 1.2265625,
"learning_rate": 0.000472,
"loss": 7.5225,
"mean_token_accuracy": 0.10159855633974076,
"num_tokens": 1853861.0,
"step": 945
},
{
"entropy": 7.520095157623291,
"epoch": 0.9448035803083044,
"grad_norm": 1.140625,
"learning_rate": 0.0004745,
"loss": 7.4465,
"mean_token_accuracy": 0.10155273377895355,
"num_tokens": 1862992.0,
"step": 950
},
{
"entropy": 7.476743268966675,
"epoch": 0.9497762307309796,
"grad_norm": 1.046875,
"learning_rate": 0.000477,
"loss": 7.462,
"mean_token_accuracy": 0.09680022820830345,
"num_tokens": 1871684.0,
"step": 955
},
{
"entropy": 7.5863439559936525,
"epoch": 0.9547488811536549,
"grad_norm": 1.1015625,
"learning_rate": 0.0004795,
"loss": 7.4698,
"mean_token_accuracy": 0.09647404700517655,
"num_tokens": 1881956.0,
"step": 960
},
{
"entropy": 7.474717378616333,
"epoch": 0.9597215315763302,
"grad_norm": 1.109375,
"learning_rate": 0.000482,
"loss": 7.5371,
"mean_token_accuracy": 0.09288978725671768,
"num_tokens": 1891860.0,
"step": 965
},
{
"entropy": 7.579680776596069,
"epoch": 0.9646941819990055,
"grad_norm": 1.09375,
"learning_rate": 0.0004845,
"loss": 7.4486,
"mean_token_accuracy": 0.09725009277462959,
"num_tokens": 1901521.0,
"step": 970
},
{
"entropy": 7.453162288665771,
"epoch": 0.9696668324216807,
"grad_norm": 1.28125,
"learning_rate": 0.000487,
"loss": 7.4141,
"mean_token_accuracy": 0.09932923316955566,
"num_tokens": 1910438.0,
"step": 975
},
{
"entropy": 7.387307262420654,
"epoch": 0.9746394828443561,
"grad_norm": 1.2421875,
"learning_rate": 0.0004895,
"loss": 7.4446,
"mean_token_accuracy": 0.1021581083536148,
"num_tokens": 1922045.0,
"step": 980
},
{
"entropy": 7.487140321731568,
"epoch": 0.9796121332670313,
"grad_norm": 1.125,
"learning_rate": 0.000492,
"loss": 7.2728,
"mean_token_accuracy": 0.111979690939188,
"num_tokens": 1930330.0,
"step": 985
},
{
"entropy": 7.266748380661011,
"epoch": 0.9845847836897066,
"grad_norm": 1.0,
"learning_rate": 0.0004945,
"loss": 7.2387,
"mean_token_accuracy": 0.10433561950922013,
"num_tokens": 1939485.0,
"step": 990
},
{
"entropy": 7.47536039352417,
"epoch": 0.989557434112382,
"grad_norm": 0.96875,
"learning_rate": 0.000497,
"loss": 7.3665,
"mean_token_accuracy": 0.09609432145953178,
"num_tokens": 1949448.0,
"step": 995
},
{
"entropy": 7.554550075531006,
"epoch": 0.9945300845350572,
"grad_norm": 0.84375,
"learning_rate": 0.0004995,
"loss": 7.467,
"mean_token_accuracy": 0.0964827798306942,
"num_tokens": 1959923.0,
"step": 1000
},
{
"epoch": 0.9945300845350572,
"eval_entropy": 7.274206740515573,
"eval_loss": 7.491804599761963,
"eval_mean_token_accuracy": 0.09507872324763081,
"eval_num_tokens": 1959923.0,
"eval_runtime": 2.531,
"eval_samples_per_second": 1369.031,
"eval_steps_per_second": 171.475,
"step": 1000
},
{
"entropy": 7.353235244750977,
"epoch": 0.9995027349577325,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999997830922735,
"loss": 7.3984,
"mean_token_accuracy": 0.10374441519379615,
"num_tokens": 1969105.0,
"step": 1005
},
{
"entropy": 7.457019170125325,
"epoch": 1.0039781203381402,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999989019053515,
"loss": 7.2227,
"mean_token_accuracy": 0.10618321845928828,
"num_tokens": 1979124.0,
"step": 1010
},
{
"entropy": 7.439343357086182,
"epoch": 1.0089507707608154,
"grad_norm": 1.0625,
"learning_rate": 0.0004999973428851536,
"loss": 7.2425,
"mean_token_accuracy": 0.10488953292369843,
"num_tokens": 1987813.0,
"step": 1015
},
{
"entropy": 7.403989601135254,
"epoch": 1.0139234211834909,
"grad_norm": 1.125,
"learning_rate": 0.0004999951060363766,
"loss": 7.2194,
"mean_token_accuracy": 0.10040127784013748,
"num_tokens": 1996743.0,
"step": 1020
},
{
"entropy": 7.332245922088623,
"epoch": 1.0188960716061661,
"grad_norm": 1.09375,
"learning_rate": 0.0004999921913657592,
"loss": 7.2289,
"mean_token_accuracy": 0.10728532820940018,
"num_tokens": 2006499.0,
"step": 1025
},
{
"entropy": 7.397444581985473,
"epoch": 1.0238687220288414,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999885988820821,
"loss": 7.2181,
"mean_token_accuracy": 0.10751654133200646,
"num_tokens": 2015103.0,
"step": 1030
},
{
"entropy": 7.333607339859009,
"epoch": 1.0288413724515166,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999843285961683,
"loss": 7.1521,
"mean_token_accuracy": 0.10657267794013023,
"num_tokens": 2024681.0,
"step": 1035
},
{
"entropy": 7.361648988723755,
"epoch": 1.0338140228741919,
"grad_norm": 1.0,
"learning_rate": 0.0004999793805208822,
"loss": 7.2333,
"mean_token_accuracy": 0.10235757604241372,
"num_tokens": 2034388.0,
"step": 1040
},
{
"entropy": 7.3061192512512205,
"epoch": 1.0387866732968671,
"grad_norm": 1.015625,
"learning_rate": 0.0004999737546711305,
"loss": 7.1629,
"mean_token_accuracy": 0.10662485361099243,
"num_tokens": 2042729.0,
"step": 1045
},
{
"entropy": 7.468704652786255,
"epoch": 1.0437593237195426,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999674510638618,
"loss": 7.2491,
"mean_token_accuracy": 0.10004433616995811,
"num_tokens": 2051669.0,
"step": 1050
},
{
"entropy": 7.315274286270141,
"epoch": 1.0487319741422179,
"grad_norm": 1.03125,
"learning_rate": 0.0004999604697180661,
"loss": 7.203,
"mean_token_accuracy": 0.10867782533168793,
"num_tokens": 2061002.0,
"step": 1055
},
{
"entropy": 7.257018661499023,
"epoch": 1.053704624564893,
"grad_norm": 1.125,
"learning_rate": 0.0004999528106547759,
"loss": 7.1119,
"mean_token_accuracy": 0.11166723147034645,
"num_tokens": 2070222.0,
"step": 1060
},
{
"entropy": 7.385891675949097,
"epoch": 1.0586772749875684,
"grad_norm": 1.03125,
"learning_rate": 0.0004999444738970644,
"loss": 7.2811,
"mean_token_accuracy": 0.10769186615943908,
"num_tokens": 2079683.0,
"step": 1065
},
{
"entropy": 7.287356996536255,
"epoch": 1.0636499254102436,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999354594700474,
"loss": 7.2799,
"mean_token_accuracy": 0.10403456911444664,
"num_tokens": 2089659.0,
"step": 1070
},
{
"entropy": 7.447647762298584,
"epoch": 1.0686225758329189,
"grad_norm": 1.078125,
"learning_rate": 0.0004999257674008817,
"loss": 7.1489,
"mean_token_accuracy": 0.11329402998089791,
"num_tokens": 2099766.0,
"step": 1075
},
{
"entropy": 7.274552202224731,
"epoch": 1.0735952262555943,
"grad_norm": 1.015625,
"learning_rate": 0.0004999153977187656,
"loss": 7.3182,
"mean_token_accuracy": 0.0980657048523426,
"num_tokens": 2110244.0,
"step": 1080
},
{
"entropy": 7.460256767272949,
"epoch": 1.0785678766782696,
"grad_norm": 1.3671875,
"learning_rate": 0.000499904350454939,
"loss": 7.2313,
"mean_token_accuracy": 0.10618335977196694,
"num_tokens": 2120068.0,
"step": 1085
},
{
"entropy": 7.2404412746429445,
"epoch": 1.0835405271009448,
"grad_norm": 1.09375,
"learning_rate": 0.0004998926256426829,
"loss": 7.2233,
"mean_token_accuracy": 0.10707401782274246,
"num_tokens": 2128862.0,
"step": 1090
},
{
"entropy": 7.458929872512817,
"epoch": 1.08851317752362,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998802233173196,
"loss": 7.2764,
"mean_token_accuracy": 0.09662094824016094,
"num_tokens": 2140618.0,
"step": 1095
},
{
"entropy": 7.288017702102661,
"epoch": 1.0934858279462953,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998671435162123,
"loss": 7.194,
"mean_token_accuracy": 0.10394087433815002,
"num_tokens": 2151890.0,
"step": 1100
},
{
"entropy": 7.270646715164185,
"epoch": 1.0984584783689706,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998533862787657,
"loss": 7.1826,
"mean_token_accuracy": 0.11115405187010766,
"num_tokens": 2160862.0,
"step": 1105
},
{
"entropy": 7.330387210845947,
"epoch": 1.1034311287916458,
"grad_norm": 1.015625,
"learning_rate": 0.0004998389516464244,
"loss": 7.2229,
"mean_token_accuracy": 0.11168113499879836,
"num_tokens": 2171631.0,
"step": 1110
},
{
"entropy": 7.2536815166473385,
"epoch": 1.1084037792143213,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998238396626746,
"loss": 7.1376,
"mean_token_accuracy": 0.10918443202972412,
"num_tokens": 2181564.0,
"step": 1115
},
{
"entropy": 7.330545425415039,
"epoch": 1.1133764296369966,
"grad_norm": 0.984375,
"learning_rate": 0.0004998080503730427,
"loss": 7.1867,
"mean_token_accuracy": 0.11050143316388131,
"num_tokens": 2191288.0,
"step": 1120
},
{
"entropy": 7.261415958404541,
"epoch": 1.1183490800596718,
"grad_norm": 1.03125,
"learning_rate": 0.0004997915838250956,
"loss": 7.1565,
"mean_token_accuracy": 0.10519161149859428,
"num_tokens": 2201254.0,
"step": 1125
},
{
"entropy": 7.230209684371948,
"epoch": 1.123321730482347,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997744400684406,
"loss": 7.1055,
"mean_token_accuracy": 0.11186393052339554,
"num_tokens": 2210325.0,
"step": 1130
},
{
"entropy": 7.281809711456299,
"epoch": 1.1282943809050223,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997566191547251,
"loss": 7.2071,
"mean_token_accuracy": 0.10663560330867768,
"num_tokens": 2220357.0,
"step": 1135
},
{
"entropy": 7.23015570640564,
"epoch": 1.1332670313276976,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997381211376364,
"loss": 7.1303,
"mean_token_accuracy": 0.11602199077606201,
"num_tokens": 2229208.0,
"step": 1140
},
{
"entropy": 7.2548665523529055,
"epoch": 1.138239681750373,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997189460729019,
"loss": 7.1328,
"mean_token_accuracy": 0.11204535439610482,
"num_tokens": 2238780.0,
"step": 1145
},
{
"entropy": 7.141768455505371,
"epoch": 1.1432123321730483,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996990940182884,
"loss": 7.167,
"mean_token_accuracy": 0.10382938161492347,
"num_tokens": 2249457.0,
"step": 1150
},
{
"entropy": 7.285257005691529,
"epoch": 1.1481849825957235,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996785650336024,
"loss": 7.1663,
"mean_token_accuracy": 0.1068244844675064,
"num_tokens": 2259565.0,
"step": 1155
},
{
"entropy": 7.224769735336304,
"epoch": 1.1531576330183988,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996573591806898,
"loss": 7.0625,
"mean_token_accuracy": 0.11556584015488625,
"num_tokens": 2268301.0,
"step": 1160
},
{
"entropy": 7.239184761047364,
"epoch": 1.158130283441074,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996354765234353,
"loss": 7.1984,
"mean_token_accuracy": 0.11132131665945053,
"num_tokens": 2278298.0,
"step": 1165
},
{
"entropy": 7.2060218334198,
"epoch": 1.1631029338637493,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996129171277628,
"loss": 7.1351,
"mean_token_accuracy": 0.10933703258633613,
"num_tokens": 2288442.0,
"step": 1170
},
{
"entropy": 7.241116619110107,
"epoch": 1.1680755842864246,
"grad_norm": 1.078125,
"learning_rate": 0.0004995896810616349,
"loss": 7.1301,
"mean_token_accuracy": 0.10417242571711541,
"num_tokens": 2298785.0,
"step": 1175
},
{
"entropy": 7.203887319564819,
"epoch": 1.1730482347091,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995657683950528,
"loss": 7.19,
"mean_token_accuracy": 0.10378496199846268,
"num_tokens": 2308881.0,
"step": 1180
},
{
"entropy": 7.240233612060547,
"epoch": 1.1780208851317753,
"grad_norm": 1.0625,
"learning_rate": 0.0004995411792000559,
"loss": 7.1108,
"mean_token_accuracy": 0.11426399871706963,
"num_tokens": 2318447.0,
"step": 1185
},
{
"entropy": 7.245753383636474,
"epoch": 1.1829935355544505,
"grad_norm": 1.125,
"learning_rate": 0.0004995159135507218,
"loss": 7.1087,
"mean_token_accuracy": 0.11610123813152314,
"num_tokens": 2327162.0,
"step": 1190
},
{
"entropy": 7.276870441436768,
"epoch": 1.1879661859771258,
"grad_norm": 1.0390625,
"learning_rate": 0.000499489971523166,
"loss": 7.1426,
"mean_token_accuracy": 0.10571493431925774,
"num_tokens": 2337207.0,
"step": 1195
},
{
"entropy": 7.15098819732666,
"epoch": 1.192938836399801,
"grad_norm": 0.92578125,
"learning_rate": 0.0004994633531955415,
"loss": 7.1241,
"mean_token_accuracy": 0.10898757874965667,
"num_tokens": 2347381.0,
"step": 1200
},
{
"entropy": 7.295102882385254,
"epoch": 1.1979114868224765,
"grad_norm": 0.9375,
"learning_rate": 0.0004994360586480391,
"loss": 7.1662,
"mean_token_accuracy": 0.10816890597343445,
"num_tokens": 2357942.0,
"step": 1205
},
{
"entropy": 7.127454805374145,
"epoch": 1.2028841372451518,
"grad_norm": 1.09375,
"learning_rate": 0.0004994080879628864,
"loss": 7.0819,
"mean_token_accuracy": 0.11559660658240319,
"num_tokens": 2367242.0,
"step": 1210
},
{
"entropy": 7.35596718788147,
"epoch": 1.207856787667827,
"grad_norm": 1.15625,
"learning_rate": 0.0004993794412243481,
"loss": 7.1466,
"mean_token_accuracy": 0.11849364861845971,
"num_tokens": 2376344.0,
"step": 1215
},
{
"entropy": 7.073469018936157,
"epoch": 1.2128294380905023,
"grad_norm": 0.92578125,
"learning_rate": 0.0004993501185187255,
"loss": 7.1563,
"mean_token_accuracy": 0.11420170515775681,
"num_tokens": 2387206.0,
"step": 1220
},
{
"entropy": 7.336416816711425,
"epoch": 1.2178020885131775,
"grad_norm": 1.015625,
"learning_rate": 0.0004993201199343565,
"loss": 7.1873,
"mean_token_accuracy": 0.10830738544464111,
"num_tokens": 2397532.0,
"step": 1225
},
{
"entropy": 7.1107689380645756,
"epoch": 1.2227747389358528,
"grad_norm": 0.921875,
"learning_rate": 0.0004992894455616149,
"loss": 7.0445,
"mean_token_accuracy": 0.10626394525170327,
"num_tokens": 2409165.0,
"step": 1230
},
{
"entropy": 7.190313863754272,
"epoch": 1.227747389358528,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992580954929108,
"loss": 7.149,
"mean_token_accuracy": 0.10830143168568611,
"num_tokens": 2420657.0,
"step": 1235
},
{
"entropy": 7.180588579177856,
"epoch": 1.2327200397812033,
"grad_norm": 0.98828125,
"learning_rate": 0.0004992260698226894,
"loss": 7.083,
"mean_token_accuracy": 0.109988534450531,
"num_tokens": 2429936.0,
"step": 1240
},
{
"entropy": 7.22004280090332,
"epoch": 1.2376926902038787,
"grad_norm": 1.078125,
"learning_rate": 0.0004991933686474315,
"loss": 7.0766,
"mean_token_accuracy": 0.1176595076918602,
"num_tokens": 2440386.0,
"step": 1245
},
{
"entropy": 7.133073568344116,
"epoch": 1.242665340626554,
"grad_norm": 0.984375,
"learning_rate": 0.0004991599920656529,
"loss": 7.2361,
"mean_token_accuracy": 0.10853235200047492,
"num_tokens": 2449086.0,
"step": 1250
},
{
"entropy": 7.390012121200561,
"epoch": 1.2476379910492292,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991259401779042,
"loss": 7.2336,
"mean_token_accuracy": 0.10847590416669846,
"num_tokens": 2459380.0,
"step": 1255
},
{
"entropy": 7.199709463119507,
"epoch": 1.2526106414719045,
"grad_norm": 1.03125,
"learning_rate": 0.0004990912130867704,
"loss": 7.0777,
"mean_token_accuracy": 0.11947175413370133,
"num_tokens": 2468623.0,
"step": 1260
},
{
"entropy": 7.199767065048218,
"epoch": 1.2575832918945797,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990558108968703,
"loss": 7.0294,
"mean_token_accuracy": 0.1127948857843876,
"num_tokens": 2479817.0,
"step": 1265
},
{
"entropy": 7.164726114273071,
"epoch": 1.2625559423172552,
"grad_norm": 2.578125,
"learning_rate": 0.0004990197337148571,
"loss": 7.0859,
"mean_token_accuracy": 0.11037703976035118,
"num_tokens": 2490544.0,
"step": 1270
},
{
"entropy": 7.167629146575928,
"epoch": 1.2675285927399305,
"grad_norm": 1.0234375,
"learning_rate": 0.000498982981649417,
"loss": 7.0454,
"mean_token_accuracy": 0.12425570264458656,
"num_tokens": 2500516.0,
"step": 1275
},
{
"entropy": 7.172258043289185,
"epoch": 1.2725012431626057,
"grad_norm": 0.96875,
"learning_rate": 0.0004989455548112695,
"loss": 7.1754,
"mean_token_accuracy": 0.10876649245619774,
"num_tokens": 2510508.0,
"step": 1280
},
{
"entropy": 7.203859663009643,
"epoch": 1.277473893585281,
"grad_norm": 1.03125,
"learning_rate": 0.0004989074533131671,
"loss": 7.1213,
"mean_token_accuracy": 0.10957314744591713,
"num_tokens": 2520532.0,
"step": 1285
},
{
"entropy": 7.155911636352539,
"epoch": 1.2824465440079562,
"grad_norm": 1.03125,
"learning_rate": 0.0004988686772698945,
"loss": 7.0755,
"mean_token_accuracy": 0.11014840230345727,
"num_tokens": 2529912.0,
"step": 1290
},
{
"entropy": 7.191475534439087,
"epoch": 1.2874191944306315,
"grad_norm": 0.9296875,
"learning_rate": 0.0004988292267982689,
"loss": 7.1632,
"mean_token_accuracy": 0.10724633038043976,
"num_tokens": 2540700.0,
"step": 1295
},
{
"entropy": 7.206296586990357,
"epoch": 1.2923918448533067,
"grad_norm": 1.125,
"learning_rate": 0.0004987891020171387,
"loss": 7.0413,
"mean_token_accuracy": 0.11765994355082512,
"num_tokens": 2549343.0,
"step": 1300
},
{
"entropy": 7.083246183395386,
"epoch": 1.297364495275982,
"grad_norm": 1.109375,
"learning_rate": 0.0004987483030473845,
"loss": 7.0668,
"mean_token_accuracy": 0.1165860801935196,
"num_tokens": 2558630.0,
"step": 1305
},
{
"entropy": 7.1589103698730465,
"epoch": 1.3023371456986574,
"grad_norm": 1.1015625,
"learning_rate": 0.0004987068300119173,
"loss": 7.0718,
"mean_token_accuracy": 0.11376485601067543,
"num_tokens": 2568559.0,
"step": 1310
},
{
"entropy": 7.175602960586548,
"epoch": 1.3073097961213327,
"grad_norm": 0.9609375,
"learning_rate": 0.0004986646830356792,
"loss": 7.1026,
"mean_token_accuracy": 0.11608376652002335,
"num_tokens": 2578088.0,
"step": 1315
},
{
"entropy": 7.084215831756592,
"epoch": 1.312282446544008,
"grad_norm": 0.859375,
"learning_rate": 0.0004986218622456423,
"loss": 7.0601,
"mean_token_accuracy": 0.12005995959043503,
"num_tokens": 2588518.0,
"step": 1320
},
{
"entropy": 7.294498062133789,
"epoch": 1.3172550969666832,
"grad_norm": 1.03125,
"learning_rate": 0.000498578367770809,
"loss": 7.2039,
"mean_token_accuracy": 0.10467370301485061,
"num_tokens": 2598850.0,
"step": 1325
},
{
"entropy": 7.135764741897583,
"epoch": 1.3222277473893587,
"grad_norm": 0.9453125,
"learning_rate": 0.0004985341997422111,
"loss": 7.1688,
"mean_token_accuracy": 0.10237733125686646,
"num_tokens": 2610667.0,
"step": 1330
},
{
"entropy": 7.185567140579224,
"epoch": 1.327200397812034,
"grad_norm": 1.0546875,
"learning_rate": 0.0004984893582929093,
"loss": 7.0257,
"mean_token_accuracy": 0.12162859663367272,
"num_tokens": 2620430.0,
"step": 1335
},
{
"entropy": 6.903904771804809,
"epoch": 1.3321730482347092,
"grad_norm": 1.0859375,
"learning_rate": 0.0004984438435579933,
"loss": 6.8987,
"mean_token_accuracy": 0.12102525010704994,
"num_tokens": 2629433.0,
"step": 1340
},
{
"entropy": 7.157380104064941,
"epoch": 1.3371456986573844,
"grad_norm": 1.1015625,
"learning_rate": 0.0004983976556745812,
"loss": 7.0789,
"mean_token_accuracy": 0.10664870142936707,
"num_tokens": 2639611.0,
"step": 1345
},
{
"entropy": 7.200573539733886,
"epoch": 1.3421183490800597,
"grad_norm": 0.90234375,
"learning_rate": 0.000498350794781819,
"loss": 7.0774,
"mean_token_accuracy": 0.11245412677526474,
"num_tokens": 2649494.0,
"step": 1350
},
{
"entropy": 7.080723524093628,
"epoch": 1.347090999502735,
"grad_norm": 0.9453125,
"learning_rate": 0.0004983032610208801,
"loss": 7.0022,
"mean_token_accuracy": 0.11420606151223182,
"num_tokens": 2658723.0,
"step": 1355
},
{
"entropy": 7.136773586273193,
"epoch": 1.3520636499254102,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982550545349649,
"loss": 7.0418,
"mean_token_accuracy": 0.11387985423207284,
"num_tokens": 2668530.0,
"step": 1360
},
{
"entropy": 7.171313905715943,
"epoch": 1.3570363003480854,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982061754693008,
"loss": 7.1111,
"mean_token_accuracy": 0.113609828799963,
"num_tokens": 2678788.0,
"step": 1365
},
{
"entropy": 7.13408350944519,
"epoch": 1.362008950770761,
"grad_norm": 0.984375,
"learning_rate": 0.0004981566239711412,
"loss": 7.0156,
"mean_token_accuracy": 0.11711029559373856,
"num_tokens": 2688308.0,
"step": 1370
},
{
"entropy": 6.921298694610596,
"epoch": 1.3669816011934361,
"grad_norm": 2.578125,
"learning_rate": 0.0004981064001897651,
"loss": 6.973,
"mean_token_accuracy": 0.11835426688194275,
"num_tokens": 2698146.0,
"step": 1375
},
{
"entropy": 7.197758626937866,
"epoch": 1.3719542516161114,
"grad_norm": 1.2421875,
"learning_rate": 0.0004980555042764772,
"loss": 7.0236,
"mean_token_accuracy": 0.11459743902087212,
"num_tokens": 2707336.0,
"step": 1380
},
{
"entropy": 7.096755838394165,
"epoch": 1.3769269020387866,
"grad_norm": 1.0234375,
"learning_rate": 0.0004980039363846069,
"loss": 7.0015,
"mean_token_accuracy": 0.12133366167545319,
"num_tokens": 2717573.0,
"step": 1385
},
{
"entropy": 7.0911478996276855,
"epoch": 1.381899552461462,
"grad_norm": 1.0234375,
"learning_rate": 0.000497951696669508,
"loss": 6.9963,
"mean_token_accuracy": 0.12023369744420051,
"num_tokens": 2727488.0,
"step": 1390
},
{
"entropy": 7.039147090911865,
"epoch": 1.3868722028841374,
"grad_norm": 0.97265625,
"learning_rate": 0.0004978987852885579,
"loss": 6.9939,
"mean_token_accuracy": 0.12316683754324913,
"num_tokens": 2738010.0,
"step": 1395
},
{
"entropy": 7.114057350158691,
"epoch": 1.3918448533068126,
"grad_norm": 1.0546875,
"learning_rate": 0.0004978452024011582,
"loss": 7.0293,
"mean_token_accuracy": 0.1164513386785984,
"num_tokens": 2746984.0,
"step": 1400
},
{
"entropy": 7.1252563953399655,
"epoch": 1.3968175037294879,
"grad_norm": 1.0703125,
"learning_rate": 0.000497790948168733,
"loss": 7.0438,
"mean_token_accuracy": 0.11343937739729881,
"num_tokens": 2756578.0,
"step": 1405
},
{
"entropy": 7.112103080749511,
"epoch": 1.4017901541521631,
"grad_norm": 0.94921875,
"learning_rate": 0.0004977360227547288,
"loss": 6.9922,
"mean_token_accuracy": 0.12000504955649376,
"num_tokens": 2765773.0,
"step": 1410
},
{
"entropy": 7.04692587852478,
"epoch": 1.4067628045748384,
"grad_norm": 1.0234375,
"learning_rate": 0.0004976804263246146,
"loss": 6.9818,
"mean_token_accuracy": 0.12030813694000245,
"num_tokens": 2775002.0,
"step": 1415
},
{
"entropy": 7.180310916900635,
"epoch": 1.4117354549975136,
"grad_norm": 0.953125,
"learning_rate": 0.0004976241590458803,
"loss": 7.0527,
"mean_token_accuracy": 0.1163438878953457,
"num_tokens": 2784682.0,
"step": 1420
},
{
"entropy": 7.036463880538941,
"epoch": 1.4167081054201889,
"grad_norm": 0.96875,
"learning_rate": 0.0004975672210880372,
"loss": 6.9593,
"mean_token_accuracy": 0.12377962619066238,
"num_tokens": 2794371.0,
"step": 1425
},
{
"entropy": 7.1429462909698485,
"epoch": 1.4216807558428641,
"grad_norm": 0.97265625,
"learning_rate": 0.0004975096126226171,
"loss": 7.0222,
"mean_token_accuracy": 0.11636790558695793,
"num_tokens": 2803840.0,
"step": 1430
},
{
"entropy": 6.9931422710418705,
"epoch": 1.4266534062655396,
"grad_norm": 1.0234375,
"learning_rate": 0.0004974513338231715,
"loss": 6.9398,
"mean_token_accuracy": 0.12235129252076149,
"num_tokens": 2813536.0,
"step": 1435
},
{
"entropy": 7.061179304122925,
"epoch": 1.4316260566882149,
"grad_norm": 1.015625,
"learning_rate": 0.0004973923848652715,
"loss": 7.0019,
"mean_token_accuracy": 0.11406521499156952,
"num_tokens": 2822741.0,
"step": 1440
},
{
"entropy": 7.118486928939819,
"epoch": 1.43659870711089,
"grad_norm": 1.140625,
"learning_rate": 0.0004973327659265073,
"loss": 6.9595,
"mean_token_accuracy": 0.1193142868578434,
"num_tokens": 2832208.0,
"step": 1445
},
{
"entropy": 6.882198286056519,
"epoch": 1.4415713575335654,
"grad_norm": 1.09375,
"learning_rate": 0.0004972724771864874,
"loss": 6.8508,
"mean_token_accuracy": 0.1252315267920494,
"num_tokens": 2841672.0,
"step": 1450
},
{
"entropy": 7.089263820648194,
"epoch": 1.4465440079562406,
"grad_norm": 1.0,
"learning_rate": 0.0004972115188268378,
"loss": 6.9647,
"mean_token_accuracy": 0.11878615766763687,
"num_tokens": 2852323.0,
"step": 1455
},
{
"entropy": 7.0160314559936525,
"epoch": 1.451516658378916,
"grad_norm": 1.0546875,
"learning_rate": 0.0004971498910312024,
"loss": 7.025,
"mean_token_accuracy": 0.11984933465719223,
"num_tokens": 2861863.0,
"step": 1460
},
{
"entropy": 7.071931934356689,
"epoch": 1.4564893088015913,
"grad_norm": 1.0546875,
"learning_rate": 0.0004970875939852412,
"loss": 6.9709,
"mean_token_accuracy": 0.12041502445936203,
"num_tokens": 2871608.0,
"step": 1465
},
{
"entropy": 6.973705291748047,
"epoch": 1.4614619592242666,
"grad_norm": 1.0625,
"learning_rate": 0.000497024627876631,
"loss": 6.9334,
"mean_token_accuracy": 0.11892005652189255,
"num_tokens": 2882091.0,
"step": 1470
},
{
"entropy": 7.082184648513794,
"epoch": 1.4664346096469418,
"grad_norm": 1.171875,
"learning_rate": 0.0004969609928950639,
"loss": 6.8939,
"mean_token_accuracy": 0.12786136120557784,
"num_tokens": 2892499.0,
"step": 1475
},
{
"entropy": 7.019110584259034,
"epoch": 1.471407260069617,
"grad_norm": 0.98046875,
"learning_rate": 0.0004968966892322468,
"loss": 6.9454,
"mean_token_accuracy": 0.11744871884584426,
"num_tokens": 2902208.0,
"step": 1480
},
{
"entropy": 7.049245834350586,
"epoch": 1.4763799104922923,
"grad_norm": 0.99609375,
"learning_rate": 0.0004968317170819018,
"loss": 6.9257,
"mean_token_accuracy": 0.11979928463697434,
"num_tokens": 2911196.0,
"step": 1485
},
{
"entropy": 6.953201770782471,
"epoch": 1.4813525609149676,
"grad_norm": 1.1171875,
"learning_rate": 0.0004967660766397642,
"loss": 6.9696,
"mean_token_accuracy": 0.11339190155267716,
"num_tokens": 2921339.0,
"step": 1490
},
{
"entropy": 7.122575187683106,
"epoch": 1.4863252113376428,
"grad_norm": 1.109375,
"learning_rate": 0.0004966997681035829,
"loss": 6.9165,
"mean_token_accuracy": 0.117047119140625,
"num_tokens": 2930795.0,
"step": 1495
},
{
"entropy": 6.9450513362884525,
"epoch": 1.4912978617603183,
"grad_norm": 0.984375,
"learning_rate": 0.0004966327916731195,
"loss": 6.8855,
"mean_token_accuracy": 0.12101696357131005,
"num_tokens": 2939647.0,
"step": 1500
},
{
"epoch": 1.4912978617603183,
"eval_entropy": 6.896390275471771,
"eval_loss": 7.153826713562012,
"eval_mean_token_accuracy": 0.111372493739639,
"eval_num_tokens": 2939647.0,
"eval_runtime": 2.5473,
"eval_samples_per_second": 1360.259,
"eval_steps_per_second": 170.376,
"step": 1500
},
{
"entropy": 7.024867534637451,
"epoch": 1.4962705121829936,
"grad_norm": 1.0703125,
"learning_rate": 0.0004965651475501476,
"loss": 6.9212,
"mean_token_accuracy": 0.12069288194179535,
"num_tokens": 2948365.0,
"step": 1505
},
{
"entropy": 6.94093337059021,
"epoch": 1.5012431626056688,
"grad_norm": 1.0,
"learning_rate": 0.0004964968359384524,
"loss": 6.9325,
"mean_token_accuracy": 0.11558735221624375,
"num_tokens": 2957447.0,
"step": 1510
},
{
"entropy": 7.112627077102661,
"epoch": 1.506215813028344,
"grad_norm": 1.1171875,
"learning_rate": 0.0004964278570438299,
"loss": 6.9271,
"mean_token_accuracy": 0.12053594887256622,
"num_tokens": 2965938.0,
"step": 1515
},
{
"entropy": 7.008978652954101,
"epoch": 1.5111884634510195,
"grad_norm": 1.0546875,
"learning_rate": 0.0004963582110740865,
"loss": 6.9347,
"mean_token_accuracy": 0.1205656386911869,
"num_tokens": 2975830.0,
"step": 1520
},
{
"entropy": 7.047356414794922,
"epoch": 1.5161611138736948,
"grad_norm": 1.015625,
"learning_rate": 0.0004962878982390381,
"loss": 6.9197,
"mean_token_accuracy": 0.1240025982260704,
"num_tokens": 2985391.0,
"step": 1525
},
{
"entropy": 7.048643589019775,
"epoch": 1.52113376429637,
"grad_norm": 1.1484375,
"learning_rate": 0.0004962169187505097,
"loss": 6.9336,
"mean_token_accuracy": 0.12923261746764184,
"num_tokens": 2994397.0,
"step": 1530
},
{
"entropy": 6.9428564548492435,
"epoch": 1.5261064147190453,
"grad_norm": 0.9609375,
"learning_rate": 0.0004961452728223343,
"loss": 6.953,
"mean_token_accuracy": 0.11560544520616531,
"num_tokens": 3005085.0,
"step": 1535
},
{
"entropy": 7.029836559295655,
"epoch": 1.5310790651417205,
"grad_norm": 0.96484375,
"learning_rate": 0.0004960729606703535,
"loss": 6.9578,
"mean_token_accuracy": 0.11529956981539727,
"num_tokens": 3015500.0,
"step": 1540
},
{
"entropy": 7.058086109161377,
"epoch": 1.5360517155643958,
"grad_norm": 0.99609375,
"learning_rate": 0.0004959999825124147,
"loss": 7.0565,
"mean_token_accuracy": 0.11921946704387665,
"num_tokens": 3024862.0,
"step": 1545
},
{
"entropy": 7.083318996429443,
"epoch": 1.541024365987071,
"grad_norm": 1.0703125,
"learning_rate": 0.000495926338568373,
"loss": 6.8992,
"mean_token_accuracy": 0.11814992874860764,
"num_tokens": 3034775.0,
"step": 1550
},
{
"entropy": 6.957399702072143,
"epoch": 1.5459970164097463,
"grad_norm": 1.0703125,
"learning_rate": 0.0004958520290600884,
"loss": 6.9628,
"mean_token_accuracy": 0.11769941002130509,
"num_tokens": 3044048.0,
"step": 1555
},
{
"entropy": 7.14118275642395,
"epoch": 1.5509696668324215,
"grad_norm": 1.0390625,
"learning_rate": 0.0004957770542114262,
"loss": 6.9626,
"mean_token_accuracy": 0.11674395352602004,
"num_tokens": 3054237.0,
"step": 1560
},
{
"entropy": 6.794984579086304,
"epoch": 1.555942317255097,
"grad_norm": 0.9140625,
"learning_rate": 0.0004957014142482563,
"loss": 6.7674,
"mean_token_accuracy": 0.1281619429588318,
"num_tokens": 3064787.0,
"step": 1565
},
{
"entropy": 6.9928360939025875,
"epoch": 1.5609149676777723,
"grad_norm": 1.0390625,
"learning_rate": 0.0004956251093984521,
"loss": 6.9458,
"mean_token_accuracy": 0.12229207009077073,
"num_tokens": 3074002.0,
"step": 1570
},
{
"entropy": 6.850202512741089,
"epoch": 1.5658876181004475,
"grad_norm": 1.0546875,
"learning_rate": 0.0004955481398918902,
"loss": 6.7999,
"mean_token_accuracy": 0.12233345210552216,
"num_tokens": 3084270.0,
"step": 1575
},
{
"entropy": 7.043655920028686,
"epoch": 1.5708602685231228,
"grad_norm": 0.91796875,
"learning_rate": 0.0004954705059604495,
"loss": 6.9182,
"mean_token_accuracy": 0.12310172468423844,
"num_tokens": 3095236.0,
"step": 1580
},
{
"entropy": 6.941450357437134,
"epoch": 1.5758329189457982,
"grad_norm": 1.0390625,
"learning_rate": 0.0004953922078380104,
"loss": 6.9168,
"mean_token_accuracy": 0.12208462804555893,
"num_tokens": 3104912.0,
"step": 1585
},
{
"entropy": 6.970426225662232,
"epoch": 1.5808055693684735,
"grad_norm": 1.0,
"learning_rate": 0.0004953132457604544,
"loss": 6.9082,
"mean_token_accuracy": 0.12240447178483009,
"num_tokens": 3114782.0,
"step": 1590
},
{
"entropy": 7.107024383544922,
"epoch": 1.5857782197911487,
"grad_norm": 1.046875,
"learning_rate": 0.0004952336199656632,
"loss": 6.9601,
"mean_token_accuracy": 0.12547790259122849,
"num_tokens": 3123606.0,
"step": 1595
},
{
"entropy": 6.788683700561523,
"epoch": 1.590750870213824,
"grad_norm": 1.046875,
"learning_rate": 0.0004951533306935181,
"loss": 6.7625,
"mean_token_accuracy": 0.13145018517971038,
"num_tokens": 3133343.0,
"step": 1600
},
{
"entropy": 7.108981418609619,
"epoch": 1.5957235206364992,
"grad_norm": 1.1015625,
"learning_rate": 0.0004950723781858991,
"loss": 7.0058,
"mean_token_accuracy": 0.11305271163582802,
"num_tokens": 3143137.0,
"step": 1605
},
{
"entropy": 6.947832918167114,
"epoch": 1.6006961710591745,
"grad_norm": 0.99609375,
"learning_rate": 0.0004949907626866844,
"loss": 6.8644,
"mean_token_accuracy": 0.1278637908399105,
"num_tokens": 3152595.0,
"step": 1610
},
{
"entropy": 6.939337062835693,
"epoch": 1.6056688214818498,
"grad_norm": 1.1015625,
"learning_rate": 0.0004949084844417493,
"loss": 6.9363,
"mean_token_accuracy": 0.11795696318149566,
"num_tokens": 3162333.0,
"step": 1615
},
{
"entropy": 6.9427478313446045,
"epoch": 1.610641471904525,
"grad_norm": 0.94921875,
"learning_rate": 0.0004948255436989657,
"loss": 6.8453,
"mean_token_accuracy": 0.12352253571152687,
"num_tokens": 3172765.0,
"step": 1620
},
{
"entropy": 7.045589303970337,
"epoch": 1.6156141223272003,
"grad_norm": 0.89453125,
"learning_rate": 0.0004947419407082017,
"loss": 6.9929,
"mean_token_accuracy": 0.11614625081419945,
"num_tokens": 3183613.0,
"step": 1625
},
{
"entropy": 7.03980655670166,
"epoch": 1.6205867727498757,
"grad_norm": 1.09375,
"learning_rate": 0.0004946576757213202,
"loss": 6.8766,
"mean_token_accuracy": 0.12009989842772484,
"num_tokens": 3193020.0,
"step": 1630
},
{
"entropy": 6.980470466613769,
"epoch": 1.625559423172551,
"grad_norm": 0.96484375,
"learning_rate": 0.0004945727489921784,
"loss": 6.8784,
"mean_token_accuracy": 0.11922077238559722,
"num_tokens": 3202845.0,
"step": 1635
},
{
"entropy": 6.8759448528289795,
"epoch": 1.6305320735952262,
"grad_norm": 1.1796875,
"learning_rate": 0.0004944871607766273,
"loss": 6.8369,
"mean_token_accuracy": 0.11566570103168487,
"num_tokens": 3211195.0,
"step": 1640
},
{
"entropy": 6.975606393814087,
"epoch": 1.6355047240179017,
"grad_norm": 0.94921875,
"learning_rate": 0.0004944009113325104,
"loss": 6.9086,
"mean_token_accuracy": 0.12444080635905266,
"num_tokens": 3221549.0,
"step": 1645
},
{
"entropy": 6.922389078140259,
"epoch": 1.640477374440577,
"grad_norm": 1.0234375,
"learning_rate": 0.0004943140009196635,
"loss": 6.8006,
"mean_token_accuracy": 0.12442531585693359,
"num_tokens": 3232206.0,
"step": 1650
},
{
"entropy": 6.991663599014283,
"epoch": 1.6454500248632522,
"grad_norm": 0.98046875,
"learning_rate": 0.0004942264297999135,
"loss": 6.878,
"mean_token_accuracy": 0.11768145188689232,
"num_tokens": 3241613.0,
"step": 1655
},
{
"entropy": 6.811205768585205,
"epoch": 1.6504226752859275,
"grad_norm": 0.9140625,
"learning_rate": 0.0004941381982370778,
"loss": 6.8443,
"mean_token_accuracy": 0.11838877126574517,
"num_tokens": 3251968.0,
"step": 1660
},
{
"entropy": 6.960154819488525,
"epoch": 1.6553953257086027,
"grad_norm": 1.0078125,
"learning_rate": 0.0004940493064969633,
"loss": 6.8302,
"mean_token_accuracy": 0.12369426786899566,
"num_tokens": 3261969.0,
"step": 1665
},
{
"entropy": 6.932631254196167,
"epoch": 1.660367976131278,
"grad_norm": 1.09375,
"learning_rate": 0.000493959754847366,
"loss": 6.8695,
"mean_token_accuracy": 0.12236511632800103,
"num_tokens": 3270894.0,
"step": 1670
},
{
"entropy": 6.932890605926514,
"epoch": 1.6653406265539532,
"grad_norm": 1.0546875,
"learning_rate": 0.0004938695435580698,
"loss": 6.9159,
"mean_token_accuracy": 0.11275593265891075,
"num_tokens": 3280949.0,
"step": 1675
},
{
"entropy": 7.057922506332398,
"epoch": 1.6703132769766285,
"grad_norm": 1.15625,
"learning_rate": 0.0004937786729008459,
"loss": 6.8769,
"mean_token_accuracy": 0.1186174876987934,
"num_tokens": 3291449.0,
"step": 1680
},
{
"entropy": 6.8796507835388185,
"epoch": 1.6752859273993037,
"grad_norm": 1.1015625,
"learning_rate": 0.0004936871431494517,
"loss": 6.8854,
"mean_token_accuracy": 0.12050152122974396,
"num_tokens": 3301632.0,
"step": 1685
},
{
"entropy": 6.933141851425171,
"epoch": 1.680258577821979,
"grad_norm": 1.234375,
"learning_rate": 0.0004935949545796307,
"loss": 6.7486,
"mean_token_accuracy": 0.12816066890954972,
"num_tokens": 3310802.0,
"step": 1690
},
{
"entropy": 6.8287769794464115,
"epoch": 1.6852312282446544,
"grad_norm": 0.94140625,
"learning_rate": 0.0004935021074691106,
"loss": 6.8149,
"mean_token_accuracy": 0.12629651874303818,
"num_tokens": 3320833.0,
"step": 1695
},
{
"entropy": 6.9207744121551515,
"epoch": 1.6902038786673297,
"grad_norm": 0.94921875,
"learning_rate": 0.0004934086020976032,
"loss": 6.7941,
"mean_token_accuracy": 0.129004967212677,
"num_tokens": 3330030.0,
"step": 1700
},
{
"entropy": 6.759958076477051,
"epoch": 1.695176529090005,
"grad_norm": 1.1171875,
"learning_rate": 0.0004933144387468037,
"loss": 6.5858,
"mean_token_accuracy": 0.14270602613687516,
"num_tokens": 3338745.0,
"step": 1705
},
{
"entropy": 6.899774885177612,
"epoch": 1.7001491795126804,
"grad_norm": 1.0390625,
"learning_rate": 0.0004932196177003891,
"loss": 6.8278,
"mean_token_accuracy": 0.12718854397535323,
"num_tokens": 3348067.0,
"step": 1710
},
{
"entropy": 6.859093141555786,
"epoch": 1.7051218299353557,
"grad_norm": 0.96875,
"learning_rate": 0.000493124139244018,
"loss": 6.7337,
"mean_token_accuracy": 0.13416660204529762,
"num_tokens": 3357061.0,
"step": 1715
},
{
"entropy": 6.898359632492065,
"epoch": 1.710094480358031,
"grad_norm": 1.0234375,
"learning_rate": 0.0004930280036653294,
"loss": 6.7571,
"mean_token_accuracy": 0.12454862222075462,
"num_tokens": 3366573.0,
"step": 1720
},
{
"entropy": 6.931832551956177,
"epoch": 1.7150671307807062,
"grad_norm": 0.97265625,
"learning_rate": 0.000492931211253942,
"loss": 6.8438,
"mean_token_accuracy": 0.11833840906620026,
"num_tokens": 3376996.0,
"step": 1725
},
{
"entropy": 6.953691625595093,
"epoch": 1.7200397812033814,
"grad_norm": 1.0,
"learning_rate": 0.0004928337623014535,
"loss": 6.8752,
"mean_token_accuracy": 0.1171141505241394,
"num_tokens": 3387615.0,
"step": 1730
},
{
"entropy": 6.756794214248657,
"epoch": 1.7250124316260567,
"grad_norm": 0.9609375,
"learning_rate": 0.000492735657101439,
"loss": 6.6557,
"mean_token_accuracy": 0.1345459997653961,
"num_tokens": 3397578.0,
"step": 1735
},
{
"entropy": 6.858054876327515,
"epoch": 1.729985082048732,
"grad_norm": 1.0625,
"learning_rate": 0.0004926368959494513,
"loss": 6.8261,
"mean_token_accuracy": 0.12104411125183105,
"num_tokens": 3406697.0,
"step": 1740
},
{
"entropy": 6.958321046829224,
"epoch": 1.7349577324714072,
"grad_norm": 0.90234375,
"learning_rate": 0.0004925374791430186,
"loss": 6.7952,
"mean_token_accuracy": 0.12863764241337777,
"num_tokens": 3416631.0,
"step": 1745
},
{
"entropy": 6.879059791564941,
"epoch": 1.7399303828940824,
"grad_norm": 1.0859375,
"learning_rate": 0.0004924374069816448,
"loss": 6.8912,
"mean_token_accuracy": 0.11450941115617752,
"num_tokens": 3426564.0,
"step": 1750
},
{
"entropy": 7.018344354629517,
"epoch": 1.7449030333167577,
"grad_norm": 1.1015625,
"learning_rate": 0.0004923366797668081,
"loss": 6.896,
"mean_token_accuracy": 0.12408864945173263,
"num_tokens": 3436200.0,
"step": 1755
},
{
"entropy": 6.8397636890411375,
"epoch": 1.7498756837394331,
"grad_norm": 0.97265625,
"learning_rate": 0.0004922352978019597,
"loss": 6.8571,
"mean_token_accuracy": 0.12426263466477394,
"num_tokens": 3446460.0,
"step": 1760
},
{
"entropy": 6.954912996292114,
"epoch": 1.7548483341621084,
"grad_norm": 1.953125,
"learning_rate": 0.0004921332613925239,
"loss": 6.742,
"mean_token_accuracy": 0.13110286071896554,
"num_tokens": 3457629.0,
"step": 1765
},
{
"entropy": 6.879143714904785,
"epoch": 1.7598209845847836,
"grad_norm": 1.3515625,
"learning_rate": 0.0004920305708458964,
"loss": 6.9071,
"mean_token_accuracy": 0.11985552161931992,
"num_tokens": 3467754.0,
"step": 1770
},
{
"entropy": 6.863123798370362,
"epoch": 1.7647936350074591,
"grad_norm": 1.0625,
"learning_rate": 0.0004919272264714431,
"loss": 6.8413,
"mean_token_accuracy": 0.12256991863250732,
"num_tokens": 3477094.0,
"step": 1775
},
{
"entropy": 6.883700656890869,
"epoch": 1.7697662854301344,
"grad_norm": 1.03125,
"learning_rate": 0.0004918232285805004,
"loss": 6.7897,
"mean_token_accuracy": 0.12876719385385513,
"num_tokens": 3487276.0,
"step": 1780
},
{
"entropy": 6.943530464172364,
"epoch": 1.7747389358528096,
"grad_norm": 0.984375,
"learning_rate": 0.0004917185774863727,
"loss": 6.7975,
"mean_token_accuracy": 0.12761382833123208,
"num_tokens": 3498118.0,
"step": 1785
},
{
"entropy": 6.754543590545654,
"epoch": 1.7797115862754849,
"grad_norm": 0.93359375,
"learning_rate": 0.0004916132735043329,
"loss": 6.6284,
"mean_token_accuracy": 0.12865251079201698,
"num_tokens": 3508286.0,
"step": 1790
},
{
"entropy": 6.811681509017944,
"epoch": 1.7846842366981601,
"grad_norm": 1.015625,
"learning_rate": 0.0004915073169516204,
"loss": 6.7476,
"mean_token_accuracy": 0.13177093118429184,
"num_tokens": 3517979.0,
"step": 1795
},
{
"entropy": 6.850424098968506,
"epoch": 1.7896568871208354,
"grad_norm": 1.0546875,
"learning_rate": 0.0004914007081474406,
"loss": 6.7761,
"mean_token_accuracy": 0.12440782338380814,
"num_tokens": 3528003.0,
"step": 1800
},
{
"entropy": 7.005227136611938,
"epoch": 1.7946295375435106,
"grad_norm": 1.1171875,
"learning_rate": 0.0004912934474129641,
"loss": 6.8417,
"mean_token_accuracy": 0.12152071818709373,
"num_tokens": 3537932.0,
"step": 1805
},
{
"entropy": 6.810896348953247,
"epoch": 1.7996021879661859,
"grad_norm": 0.984375,
"learning_rate": 0.0004911855350713251,
"loss": 6.9556,
"mean_token_accuracy": 0.1197297677397728,
"num_tokens": 3548584.0,
"step": 1810
},
{
"entropy": 7.012252855300903,
"epoch": 1.8045748383888611,
"grad_norm": 1.109375,
"learning_rate": 0.0004910769714476214,
"loss": 6.781,
"mean_token_accuracy": 0.1271994046866894,
"num_tokens": 3558117.0,
"step": 1815
},
{
"entropy": 6.8483014583587645,
"epoch": 1.8095474888115366,
"grad_norm": 1.0390625,
"learning_rate": 0.0004909677568689122,
"loss": 6.9036,
"mean_token_accuracy": 0.11816685497760773,
"num_tokens": 3569226.0,
"step": 1820
},
{
"entropy": 6.974257755279541,
"epoch": 1.8145201392342118,
"grad_norm": 1.046875,
"learning_rate": 0.0004908578916642183,
"loss": 6.7629,
"mean_token_accuracy": 0.12144647464156151,
"num_tokens": 3578375.0,
"step": 1825
},
{
"entropy": 6.866056156158447,
"epoch": 1.819492789656887,
"grad_norm": 1.078125,
"learning_rate": 0.0004907473761645205,
"loss": 6.7794,
"mean_token_accuracy": 0.13146526888012885,
"num_tokens": 3587356.0,
"step": 1830
},
{
"entropy": 6.8879882335662845,
"epoch": 1.8244654400795626,
"grad_norm": 1.09375,
"learning_rate": 0.0004906362107027583,
"loss": 6.8589,
"mean_token_accuracy": 0.12173376381397247,
"num_tokens": 3596785.0,
"step": 1835
},
{
"entropy": 6.89839768409729,
"epoch": 1.8294380905022378,
"grad_norm": 1.046875,
"learning_rate": 0.0004905243956138298,
"loss": 6.7551,
"mean_token_accuracy": 0.13092348501086234,
"num_tokens": 3605357.0,
"step": 1840
},
{
"entropy": 6.862800073623657,
"epoch": 1.834410740924913,
"grad_norm": 0.94921875,
"learning_rate": 0.0004904119312345896,
"loss": 6.7526,
"mean_token_accuracy": 0.12731626331806184,
"num_tokens": 3614979.0,
"step": 1845
},
{
"entropy": 6.827274322509766,
"epoch": 1.8393833913475883,
"grad_norm": 1.1171875,
"learning_rate": 0.0004902988179038492,
"loss": 6.757,
"mean_token_accuracy": 0.12851415947079659,
"num_tokens": 3624594.0,
"step": 1850
},
{
"entropy": 6.864847850799561,
"epoch": 1.8443560417702636,
"grad_norm": 1.078125,
"learning_rate": 0.000490185055962374,
"loss": 6.7487,
"mean_token_accuracy": 0.12592514380812644,
"num_tokens": 3633998.0,
"step": 1855
},
{
"entropy": 6.851823806762695,
"epoch": 1.8493286921929388,
"grad_norm": 1.171875,
"learning_rate": 0.0004900706457528845,
"loss": 6.7163,
"mean_token_accuracy": 0.12974942922592164,
"num_tokens": 3642406.0,
"step": 1860
},
{
"entropy": 6.770096349716186,
"epoch": 1.854301342615614,
"grad_norm": 0.94921875,
"learning_rate": 0.0004899555876200535,
"loss": 6.7885,
"mean_token_accuracy": 0.12484269142150879,
"num_tokens": 3652905.0,
"step": 1865
},
{
"entropy": 6.815859794616699,
"epoch": 1.8592739930382893,
"grad_norm": 0.9765625,
"learning_rate": 0.0004898398819105059,
"loss": 6.7069,
"mean_token_accuracy": 0.12600032165646552,
"num_tokens": 3662518.0,
"step": 1870
},
{
"entropy": 6.9039140224456785,
"epoch": 1.8642466434609646,
"grad_norm": 1.09375,
"learning_rate": 0.0004897235289728178,
"loss": 6.7629,
"mean_token_accuracy": 0.13054750561714173,
"num_tokens": 3672102.0,
"step": 1875
},
{
"entropy": 6.855216646194458,
"epoch": 1.8692192938836398,
"grad_norm": 1.046875,
"learning_rate": 0.0004896065291575145,
"loss": 6.7937,
"mean_token_accuracy": 0.12748010605573654,
"num_tokens": 3681821.0,
"step": 1880
},
{
"entropy": 7.0097535133361815,
"epoch": 1.8741919443063153,
"grad_norm": 1.1484375,
"learning_rate": 0.0004894888828170708,
"loss": 6.8683,
"mean_token_accuracy": 0.1209042452275753,
"num_tokens": 3690753.0,
"step": 1885
},
{
"entropy": 6.8094244480133055,
"epoch": 1.8791645947289906,
"grad_norm": 0.98828125,
"learning_rate": 0.0004893705903059087,
"loss": 6.8339,
"mean_token_accuracy": 0.12141335383057594,
"num_tokens": 3701694.0,
"step": 1890
},
{
"entropy": 6.893871736526489,
"epoch": 1.8841372451516658,
"grad_norm": 1.046875,
"learning_rate": 0.0004892516519803971,
"loss": 6.7541,
"mean_token_accuracy": 0.12454791739583015,
"num_tokens": 3710894.0,
"step": 1895
},
{
"entropy": 6.784770774841308,
"epoch": 1.8891098955743413,
"grad_norm": 1.046875,
"learning_rate": 0.0004891320681988504,
"loss": 6.691,
"mean_token_accuracy": 0.13386452198028564,
"num_tokens": 3720081.0,
"step": 1900
},
{
"entropy": 6.87687315940857,
"epoch": 1.8940825459970165,
"grad_norm": 1.0390625,
"learning_rate": 0.0004890118393215277,
"loss": 6.8052,
"mean_token_accuracy": 0.12933627292513847,
"num_tokens": 3731168.0,
"step": 1905
},
{
"entropy": 6.751729583740234,
"epoch": 1.8990551964196918,
"grad_norm": 1.015625,
"learning_rate": 0.0004888909657106311,
"loss": 6.6062,
"mean_token_accuracy": 0.13786736354231835,
"num_tokens": 3740533.0,
"step": 1910
},
{
"entropy": 6.782799530029297,
"epoch": 1.904027846842367,
"grad_norm": 0.9453125,
"learning_rate": 0.0004887694477303055,
"loss": 6.785,
"mean_token_accuracy": 0.12551278993487358,
"num_tokens": 3750523.0,
"step": 1915
},
{
"entropy": 6.876261758804321,
"epoch": 1.9090004972650423,
"grad_norm": 0.92578125,
"learning_rate": 0.0004886472857466368,
"loss": 6.8015,
"mean_token_accuracy": 0.12344402298331261,
"num_tokens": 3760323.0,
"step": 1920
},
{
"entropy": 6.664627456665039,
"epoch": 1.9139731476877175,
"grad_norm": 0.96875,
"learning_rate": 0.0004885244801276509,
"loss": 6.5688,
"mean_token_accuracy": 0.14272117540240287,
"num_tokens": 3770970.0,
"step": 1925
},
{
"entropy": 6.79260401725769,
"epoch": 1.9189457981103928,
"grad_norm": 1.046875,
"learning_rate": 0.0004884010312433131,
"loss": 6.8004,
"mean_token_accuracy": 0.1279281996190548,
"num_tokens": 3780810.0,
"step": 1930
},
{
"entropy": 6.853330516815186,
"epoch": 1.923918448533068,
"grad_norm": 1.078125,
"learning_rate": 0.00048827693946552615,
"loss": 6.7076,
"mean_token_accuracy": 0.13404083177447318,
"num_tokens": 3791461.0,
"step": 1935
},
{
"entropy": 6.7495325088500975,
"epoch": 1.9288910989557433,
"grad_norm": 0.98828125,
"learning_rate": 0.00048815220516813004,
"loss": 6.5249,
"mean_token_accuracy": 0.13892961144447327,
"num_tokens": 3800353.0,
"step": 1940
},
{
"entropy": 6.772555494308472,
"epoch": 1.9338637493784185,
"grad_norm": 0.9609375,
"learning_rate": 0.0004880268287269,
"loss": 6.7006,
"mean_token_accuracy": 0.13255587890744208,
"num_tokens": 3810633.0,
"step": 1945
},
{
"entropy": 6.88034839630127,
"epoch": 1.938836399801094,
"grad_norm": 1.0703125,
"learning_rate": 0.00048790081051954605,
"loss": 6.842,
"mean_token_accuracy": 0.1255967289209366,
"num_tokens": 3820254.0,
"step": 1950
},
{
"entropy": 6.9175169467926025,
"epoch": 1.9438090502237693,
"grad_norm": 1.125,
"learning_rate": 0.0004877741509257116,
"loss": 6.8171,
"mean_token_accuracy": 0.12266699597239494,
"num_tokens": 3829976.0,
"step": 1955
},
{
"entropy": 6.797274971008301,
"epoch": 1.9487817006464445,
"grad_norm": 0.98828125,
"learning_rate": 0.0004876468503269721,
"loss": 6.767,
"mean_token_accuracy": 0.1263387106359005,
"num_tokens": 3840593.0,
"step": 1960
},
{
"entropy": 6.817362308502197,
"epoch": 1.95375435106912,
"grad_norm": 1.0625,
"learning_rate": 0.00048751890910683434,
"loss": 6.7407,
"mean_token_accuracy": 0.13059170991182328,
"num_tokens": 3850063.0,
"step": 1965
},
{
"entropy": 6.871044397354126,
"epoch": 1.9587270014917952,
"grad_norm": 0.9453125,
"learning_rate": 0.00048739032765073487,
"loss": 6.7777,
"mean_token_accuracy": 0.126081869751215,
"num_tokens": 3860132.0,
"step": 1970
},
{
"entropy": 6.796940660476684,
"epoch": 1.9636996519144705,
"grad_norm": 0.95703125,
"learning_rate": 0.00048726110634603913,
"loss": 6.7346,
"mean_token_accuracy": 0.130959652364254,
"num_tokens": 3869914.0,
"step": 1975
},
{
"entropy": 6.699121189117432,
"epoch": 1.9686723023371457,
"grad_norm": 0.9609375,
"learning_rate": 0.00048713124558204013,
"loss": 6.674,
"mean_token_accuracy": 0.14097589403390884,
"num_tokens": 3879825.0,
"step": 1980
},
{
"entropy": 6.92541275024414,
"epoch": 1.973644952759821,
"grad_norm": 1.0,
"learning_rate": 0.0004870007457499573,
"loss": 6.7464,
"mean_token_accuracy": 0.13307701125741006,
"num_tokens": 3889318.0,
"step": 1985
},
{
"entropy": 6.78137845993042,
"epoch": 1.9786176031824962,
"grad_norm": 1.0625,
"learning_rate": 0.00048686960724293537,
"loss": 6.7402,
"mean_token_accuracy": 0.12863369882106782,
"num_tokens": 3899017.0,
"step": 1990
},
{
"entropy": 6.874488973617554,
"epoch": 1.9835902536051715,
"grad_norm": 1.03125,
"learning_rate": 0.0004867378304560432,
"loss": 6.6951,
"mean_token_accuracy": 0.13685127720236778,
"num_tokens": 3907766.0,
"step": 1995
},
{
"entropy": 6.797482347488403,
"epoch": 1.9885629040278467,
"grad_norm": 1.0859375,
"learning_rate": 0.0004866054157862723,
"loss": 6.6783,
"mean_token_accuracy": 0.1312019035220146,
"num_tokens": 3916323.0,
"step": 2000
},
{
"epoch": 1.9885629040278467,
"eval_entropy": 6.5960938523991315,
"eval_loss": 6.9205851554870605,
"eval_mean_token_accuracy": 0.11949630909972751,
"eval_num_tokens": 3916323.0,
"eval_runtime": 2.5415,
"eval_samples_per_second": 1363.394,
"eval_steps_per_second": 170.769,
"step": 2000
},
{
"entropy": 6.743474960327148,
"epoch": 1.993535554450522,
"grad_norm": 0.94140625,
"learning_rate": 0.0004864723636325361,
"loss": 6.7726,
"mean_token_accuracy": 0.12707573622465135,
"num_tokens": 3927639.0,
"step": 2005
},
{
"entropy": 6.870388746261597,
"epoch": 1.9985082048731975,
"grad_norm": 1.875,
"learning_rate": 0.00048633867439566855,
"loss": 6.6874,
"mean_token_accuracy": 0.12734511867165565,
"num_tokens": 3937452.0,
"step": 2010
},
{
"entropy": 6.868780401017931,
"epoch": 2.002983590253605,
"grad_norm": 0.921875,
"learning_rate": 0.0004862043484784227,
"loss": 6.6415,
"mean_token_accuracy": 0.13130029953188366,
"num_tokens": 3946737.0,
"step": 2015
},
{
"entropy": 6.816501617431641,
"epoch": 2.0079562406762803,
"grad_norm": 1.0625,
"learning_rate": 0.00048606938628546975,
"loss": 6.4258,
"mean_token_accuracy": 0.12974144965410234,
"num_tokens": 3957541.0,
"step": 2020
},
{
"entropy": 6.793483448028565,
"epoch": 2.0129288910989556,
"grad_norm": 1.015625,
"learning_rate": 0.0004859337882233978,
"loss": 6.5056,
"mean_token_accuracy": 0.1279767021536827,
"num_tokens": 3967621.0,
"step": 2025
},
{
"entropy": 6.799717760086059,
"epoch": 2.017901541521631,
"grad_norm": 1.109375,
"learning_rate": 0.0004857975547007106,
"loss": 6.5055,
"mean_token_accuracy": 0.13455797657370566,
"num_tokens": 3977920.0,
"step": 2030
},
{
"entropy": 6.578008365631104,
"epoch": 2.0228741919443065,
"grad_norm": 0.96484375,
"learning_rate": 0.00048566068612782595,
"loss": 6.1722,
"mean_token_accuracy": 0.15097661018371583,
"num_tokens": 3988128.0,
"step": 2035
},
{
"entropy": 6.703016328811645,
"epoch": 2.0278468423669818,
"grad_norm": 1.0234375,
"learning_rate": 0.00048552318291707533,
"loss": 6.3662,
"mean_token_accuracy": 0.14040629491209983,
"num_tokens": 3998467.0,
"step": 2040
},
{
"entropy": 6.733289527893066,
"epoch": 2.032819492789657,
"grad_norm": 1.1015625,
"learning_rate": 0.00048538504548270177,
"loss": 6.3358,
"mean_token_accuracy": 0.13190700858831406,
"num_tokens": 4007513.0,
"step": 2045
},
{
"entropy": 6.574963092803955,
"epoch": 2.0377921432123323,
"grad_norm": 1.0390625,
"learning_rate": 0.00048524627424085913,
"loss": 6.4053,
"mean_token_accuracy": 0.1385435938835144,
"num_tokens": 4016538.0,
"step": 2050
},
{
"entropy": 6.686594772338867,
"epoch": 2.0427647936350075,
"grad_norm": 1.0546875,
"learning_rate": 0.00048510686960961057,
"loss": 6.3102,
"mean_token_accuracy": 0.14076657891273497,
"num_tokens": 4026850.0,
"step": 2055
},
{
"entropy": 6.68578085899353,
"epoch": 2.0477374440576828,
"grad_norm": 0.98828125,
"learning_rate": 0.0004849668320089276,
"loss": 6.308,
"mean_token_accuracy": 0.1393464133143425,
"num_tokens": 4037105.0,
"step": 2060
},
{
"entropy": 6.621056842803955,
"epoch": 2.052710094480358,
"grad_norm": 1.046875,
"learning_rate": 0.00048482616186068826,
"loss": 6.3641,
"mean_token_accuracy": 0.14485840871930122,
"num_tokens": 4045694.0,
"step": 2065
},
{
"entropy": 6.611151647567749,
"epoch": 2.0576827449030333,
"grad_norm": 1.140625,
"learning_rate": 0.00048468485958867664,
"loss": 6.2588,
"mean_token_accuracy": 0.14561159759759904,
"num_tokens": 4056425.0,
"step": 2070
},
{
"entropy": 6.588506889343262,
"epoch": 2.0626553953257085,
"grad_norm": 1.109375,
"learning_rate": 0.00048454292561858083,
"loss": 6.292,
"mean_token_accuracy": 0.1491833969950676,
"num_tokens": 4066257.0,
"step": 2075
},
{
"entropy": 6.6618874073028564,
"epoch": 2.0676280457483838,
"grad_norm": 0.9765625,
"learning_rate": 0.0004844003603779921,
"loss": 6.3697,
"mean_token_accuracy": 0.13735560178756714,
"num_tokens": 4076835.0,
"step": 2080
},
{
"entropy": 6.6407470703125,
"epoch": 2.072600696171059,
"grad_norm": 0.9609375,
"learning_rate": 0.0004842571642964036,
"loss": 6.492,
"mean_token_accuracy": 0.13370660170912743,
"num_tokens": 4087644.0,
"step": 2085
},
{
"entropy": 6.70997428894043,
"epoch": 2.0775733465937343,
"grad_norm": 1.0390625,
"learning_rate": 0.00048411333780520876,
"loss": 6.4182,
"mean_token_accuracy": 0.13545572832226754,
"num_tokens": 4097040.0,
"step": 2090
},
{
"entropy": 6.660089588165283,
"epoch": 2.0825459970164095,
"grad_norm": 1.109375,
"learning_rate": 0.00048396888133770034,
"loss": 6.385,
"mean_token_accuracy": 0.1400207258760929,
"num_tokens": 4106698.0,
"step": 2095
},
{
"entropy": 6.67029595375061,
"epoch": 2.087518647439085,
"grad_norm": 0.9921875,
"learning_rate": 0.00048382379532906897,
"loss": 6.3724,
"mean_token_accuracy": 0.14506559818983078,
"num_tokens": 4116804.0,
"step": 2100
},
{
"entropy": 6.649099445343017,
"epoch": 2.0924912978617605,
"grad_norm": 1.0703125,
"learning_rate": 0.0004836780802164016,
"loss": 6.481,
"mean_token_accuracy": 0.13108737394213676,
"num_tokens": 4126658.0,
"step": 2105
},
{
"entropy": 6.663601350784302,
"epoch": 2.0974639482844357,
"grad_norm": 1.0234375,
"learning_rate": 0.00048353173643868064,
"loss": 6.2641,
"mean_token_accuracy": 0.14222084060311319,
"num_tokens": 4136323.0,
"step": 2110
},
{
"entropy": 6.552397966384888,
"epoch": 2.102436598707111,
"grad_norm": 0.94140625,
"learning_rate": 0.0004833847644367826,
"loss": 6.4093,
"mean_token_accuracy": 0.14488788247108458,
"num_tokens": 4145714.0,
"step": 2115
},
{
"entropy": 6.74545841217041,
"epoch": 2.107409249129786,
"grad_norm": 0.92578125,
"learning_rate": 0.00048323716465347615,
"loss": 6.3883,
"mean_token_accuracy": 0.13914027959108352,
"num_tokens": 4156059.0,
"step": 2120
},
{
"entropy": 6.599608993530273,
"epoch": 2.1123818995524615,
"grad_norm": 1.015625,
"learning_rate": 0.0004830889375334215,
"loss": 6.4329,
"mean_token_accuracy": 0.13543619066476822,
"num_tokens": 4166990.0,
"step": 2125
},
{
"entropy": 6.644566059112549,
"epoch": 2.1173545499751367,
"grad_norm": 1.1640625,
"learning_rate": 0.00048294008352316875,
"loss": 6.4255,
"mean_token_accuracy": 0.1432893082499504,
"num_tokens": 4175716.0,
"step": 2130
},
{
"entropy": 6.638896894454956,
"epoch": 2.122327200397812,
"grad_norm": 1.0625,
"learning_rate": 0.0004827906030711564,
"loss": 6.4268,
"mean_token_accuracy": 0.13586034253239632,
"num_tokens": 4184832.0,
"step": 2135
},
{
"entropy": 6.468638515472412,
"epoch": 2.1272998508204872,
"grad_norm": 1.1328125,
"learning_rate": 0.00048264049662771053,
"loss": 6.2182,
"mean_token_accuracy": 0.15223108381032943,
"num_tokens": 4195678.0,
"step": 2140
},
{
"entropy": 6.667506408691406,
"epoch": 2.1322725012431625,
"grad_norm": 1.0546875,
"learning_rate": 0.00048248976464504266,
"loss": 6.4117,
"mean_token_accuracy": 0.13893484696745872,
"num_tokens": 4205357.0,
"step": 2145
},
{
"entropy": 6.682303857803345,
"epoch": 2.1372451516658377,
"grad_norm": 1.015625,
"learning_rate": 0.00048233840757724926,
"loss": 6.4726,
"mean_token_accuracy": 0.13612781465053558,
"num_tokens": 4215416.0,
"step": 2150
},
{
"entropy": 6.6814305782318115,
"epoch": 2.142217802088513,
"grad_norm": 1.109375,
"learning_rate": 0.00048218642588030943,
"loss": 6.5098,
"mean_token_accuracy": 0.1335015468299389,
"num_tokens": 4225894.0,
"step": 2155
},
{
"entropy": 6.718389368057251,
"epoch": 2.1471904525111887,
"grad_norm": 1.046875,
"learning_rate": 0.00048203382001208453,
"loss": 6.4266,
"mean_token_accuracy": 0.14146106615662574,
"num_tokens": 4235973.0,
"step": 2160
},
{
"entropy": 6.5530835628509525,
"epoch": 2.152163102933864,
"grad_norm": 1.140625,
"learning_rate": 0.00048188059043231585,
"loss": 6.3416,
"mean_token_accuracy": 0.14535260200500488,
"num_tokens": 4246411.0,
"step": 2165
},
{
"entropy": 6.699462938308716,
"epoch": 2.157135753356539,
"grad_norm": 1.203125,
"learning_rate": 0.00048172673760262414,
"loss": 6.4057,
"mean_token_accuracy": 0.13765638917684556,
"num_tokens": 4255754.0,
"step": 2170
},
{
"entropy": 6.578954648971558,
"epoch": 2.1621084037792144,
"grad_norm": 1.0859375,
"learning_rate": 0.00048157226198650735,
"loss": 6.3426,
"mean_token_accuracy": 0.14024818763136865,
"num_tokens": 4265263.0,
"step": 2175
},
{
"entropy": 6.69226040840149,
"epoch": 2.1670810542018897,
"grad_norm": 0.9921875,
"learning_rate": 0.0004814171640493398,
"loss": 6.5192,
"mean_token_accuracy": 0.13299463614821433,
"num_tokens": 4275262.0,
"step": 2180
},
{
"entropy": 6.60386552810669,
"epoch": 2.172053704624565,
"grad_norm": 0.9765625,
"learning_rate": 0.0004812614442583708,
"loss": 6.4364,
"mean_token_accuracy": 0.1357870452105999,
"num_tokens": 4285467.0,
"step": 2185
},
{
"entropy": 6.650628185272216,
"epoch": 2.17702635504724,
"grad_norm": 0.984375,
"learning_rate": 0.0004811051030827227,
"loss": 6.3541,
"mean_token_accuracy": 0.13564409241080283,
"num_tokens": 4294680.0,
"step": 2190
},
{
"entropy": 6.605967903137207,
"epoch": 2.1819990054699154,
"grad_norm": 1.09375,
"learning_rate": 0.00048094814099339006,
"loss": 6.4335,
"mean_token_accuracy": 0.1353071413934231,
"num_tokens": 4304605.0,
"step": 2195
},
{
"entropy": 6.5511720180511475,
"epoch": 2.1869716558925907,
"grad_norm": 1.0546875,
"learning_rate": 0.0004807905584632381,
"loss": 6.2601,
"mean_token_accuracy": 0.1484084963798523,
"num_tokens": 4314112.0,
"step": 2200
},
{
"entropy": 6.510998249053955,
"epoch": 2.191944306315266,
"grad_norm": 1.0703125,
"learning_rate": 0.00048063235596700087,
"loss": 6.3169,
"mean_token_accuracy": 0.14116548001766205,
"num_tokens": 4323872.0,
"step": 2205
},
{
"entropy": 6.636221885681152,
"epoch": 2.196916956737941,
"grad_norm": 1.078125,
"learning_rate": 0.00048047353398128056,
"loss": 6.2781,
"mean_token_accuracy": 0.14961938709020614,
"num_tokens": 4333002.0,
"step": 2210
},
{
"entropy": 6.5612975597381595,
"epoch": 2.2018896071606164,
"grad_norm": 1.296875,
"learning_rate": 0.0004803140929845452,
"loss": 6.3989,
"mean_token_accuracy": 0.14100862443447112,
"num_tokens": 4342012.0,
"step": 2215
},
{
"entropy": 6.630680561065674,
"epoch": 2.2068622575832917,
"grad_norm": 1.171875,
"learning_rate": 0.00048015403345712796,
"loss": 6.335,
"mean_token_accuracy": 0.1386937290430069,
"num_tokens": 4351959.0,
"step": 2220
},
{
"entropy": 6.611843776702881,
"epoch": 2.2118349080059674,
"grad_norm": 0.984375,
"learning_rate": 0.0004799933558812252,
"loss": 6.3142,
"mean_token_accuracy": 0.1452218174934387,
"num_tokens": 4361267.0,
"step": 2225
},
{
"entropy": 6.592637825012207,
"epoch": 2.2168075584286426,
"grad_norm": 1.078125,
"learning_rate": 0.00047983206074089546,
"loss": 6.3592,
"mean_token_accuracy": 0.1453488864004612,
"num_tokens": 4370104.0,
"step": 2230
},
{
"entropy": 6.297759294509888,
"epoch": 2.221780208851318,
"grad_norm": 1.15625,
"learning_rate": 0.0004796701485220576,
"loss": 6.1391,
"mean_token_accuracy": 0.1663060151040554,
"num_tokens": 4380724.0,
"step": 2235
},
{
"entropy": 6.461687994003296,
"epoch": 2.226752859273993,
"grad_norm": 1.0546875,
"learning_rate": 0.0004795076197124893,
"loss": 6.2111,
"mean_token_accuracy": 0.15564971938729286,
"num_tokens": 4390521.0,
"step": 2240
},
{
"entropy": 6.5855005264282225,
"epoch": 2.2317255096966684,
"grad_norm": 1.0625,
"learning_rate": 0.00047934447480182626,
"loss": 6.39,
"mean_token_accuracy": 0.14133716821670533,
"num_tokens": 4400075.0,
"step": 2245
},
{
"entropy": 6.576947164535523,
"epoch": 2.2366981601193436,
"grad_norm": 1.0078125,
"learning_rate": 0.0004791807142815598,
"loss": 6.3309,
"mean_token_accuracy": 0.14028072953224183,
"num_tokens": 4409913.0,
"step": 2250
},
{
"entropy": 6.559875965118408,
"epoch": 2.241670810542019,
"grad_norm": 1.0234375,
"learning_rate": 0.00047901633864503614,
"loss": 6.3491,
"mean_token_accuracy": 0.14033740609884263,
"num_tokens": 4418551.0,
"step": 2255
},
{
"entropy": 6.563806486129761,
"epoch": 2.246643460964694,
"grad_norm": 1.21875,
"learning_rate": 0.00047885134838745433,
"loss": 6.5427,
"mean_token_accuracy": 0.1390278995037079,
"num_tokens": 4428129.0,
"step": 2260
},
{
"entropy": 6.722531986236572,
"epoch": 2.2516161113873694,
"grad_norm": 1.0234375,
"learning_rate": 0.0004786857440058654,
"loss": 6.4167,
"mean_token_accuracy": 0.14224035739898683,
"num_tokens": 4437311.0,
"step": 2265
},
{
"entropy": 6.425306606292724,
"epoch": 2.2565887618100446,
"grad_norm": 1.015625,
"learning_rate": 0.0004785195259991701,
"loss": 6.2592,
"mean_token_accuracy": 0.15068233460187913,
"num_tokens": 4447387.0,
"step": 2270
},
{
"entropy": 6.519154500961304,
"epoch": 2.26156141223272,
"grad_norm": 1.09375,
"learning_rate": 0.000478352694868118,
"loss": 6.4323,
"mean_token_accuracy": 0.1365874394774437,
"num_tokens": 4456435.0,
"step": 2275
},
{
"entropy": 6.466250038146972,
"epoch": 2.266534062655395,
"grad_norm": 1.1484375,
"learning_rate": 0.0004781852511153058,
"loss": 6.2613,
"mean_token_accuracy": 0.15041158795356752,
"num_tokens": 4466464.0,
"step": 2280
},
{
"entropy": 6.588946151733398,
"epoch": 2.271506713078071,
"grad_norm": 1.2109375,
"learning_rate": 0.00047801719524517573,
"loss": 6.3699,
"mean_token_accuracy": 0.14319706484675407,
"num_tokens": 4476793.0,
"step": 2285
},
{
"entropy": 6.551668357849121,
"epoch": 2.276479363500746,
"grad_norm": 1.09375,
"learning_rate": 0.00047784852776401403,
"loss": 6.3408,
"mean_token_accuracy": 0.14303407892584802,
"num_tokens": 4485811.0,
"step": 2290
},
{
"entropy": 6.616492700576782,
"epoch": 2.2814520139234213,
"grad_norm": 1.109375,
"learning_rate": 0.0004776792491799495,
"loss": 6.4544,
"mean_token_accuracy": 0.13649573624134065,
"num_tokens": 4495018.0,
"step": 2295
},
{
"entropy": 6.587541723251343,
"epoch": 2.2864246643460966,
"grad_norm": 0.99609375,
"learning_rate": 0.00047750936000295213,
"loss": 6.4237,
"mean_token_accuracy": 0.13602023273706437,
"num_tokens": 4505847.0,
"step": 2300
},
{
"entropy": 6.593376111984253,
"epoch": 2.291397314768772,
"grad_norm": 1.109375,
"learning_rate": 0.00047733886074483115,
"loss": 6.4319,
"mean_token_accuracy": 0.13742442131042482,
"num_tokens": 4516104.0,
"step": 2305
},
{
"entropy": 6.4319030284881595,
"epoch": 2.296369965191447,
"grad_norm": 1.1328125,
"learning_rate": 0.0004771677519192339,
"loss": 6.1951,
"mean_token_accuracy": 0.1542932316660881,
"num_tokens": 4524875.0,
"step": 2310
},
{
"entropy": 6.522582721710205,
"epoch": 2.3013426156141223,
"grad_norm": 1.1953125,
"learning_rate": 0.0004769960340416439,
"loss": 6.4475,
"mean_token_accuracy": 0.13546053245663642,
"num_tokens": 4533836.0,
"step": 2315
},
{
"entropy": 6.370718002319336,
"epoch": 2.3063152660367976,
"grad_norm": 1.078125,
"learning_rate": 0.00047682370762937976,
"loss": 5.9434,
"mean_token_accuracy": 0.1733056738972664,
"num_tokens": 4544323.0,
"step": 2320
},
{
"entropy": 6.515668678283691,
"epoch": 2.311287916459473,
"grad_norm": 1.09375,
"learning_rate": 0.00047665077320159315,
"loss": 6.3566,
"mean_token_accuracy": 0.14200845062732698,
"num_tokens": 4553811.0,
"step": 2325
},
{
"entropy": 6.400250387191773,
"epoch": 2.316260566882148,
"grad_norm": 1.2578125,
"learning_rate": 0.00047647723127926765,
"loss": 6.1887,
"mean_token_accuracy": 0.16027804613113403,
"num_tokens": 4563098.0,
"step": 2330
},
{
"entropy": 6.615733480453491,
"epoch": 2.3212332173048233,
"grad_norm": 1.171875,
"learning_rate": 0.0004763030823852168,
"loss": 6.4867,
"mean_token_accuracy": 0.13570496514439584,
"num_tokens": 4573589.0,
"step": 2335
},
{
"entropy": 6.541930723190307,
"epoch": 2.3262058677274986,
"grad_norm": 1.234375,
"learning_rate": 0.0004761283270440829,
"loss": 6.3805,
"mean_token_accuracy": 0.1394104018807411,
"num_tokens": 4583059.0,
"step": 2340
},
{
"entropy": 6.6634218215942385,
"epoch": 2.331178518150174,
"grad_norm": 1.1328125,
"learning_rate": 0.0004759529657823349,
"loss": 6.5747,
"mean_token_accuracy": 0.1389290988445282,
"num_tokens": 4593371.0,
"step": 2345
},
{
"entropy": 6.5851836681365965,
"epoch": 2.336151168572849,
"grad_norm": 0.93359375,
"learning_rate": 0.00047577699912826755,
"loss": 6.4681,
"mean_token_accuracy": 0.1362654224038124,
"num_tokens": 4602862.0,
"step": 2350
},
{
"entropy": 6.585508203506469,
"epoch": 2.341123818995525,
"grad_norm": 1.0,
"learning_rate": 0.00047560042761199915,
"loss": 6.3364,
"mean_token_accuracy": 0.1408190168440342,
"num_tokens": 4612827.0,
"step": 2355
},
{
"entropy": 6.567585754394531,
"epoch": 2.3460964694182,
"grad_norm": 1.203125,
"learning_rate": 0.0004754232517654704,
"loss": 6.3909,
"mean_token_accuracy": 0.13844477534294128,
"num_tokens": 4621706.0,
"step": 2360
},
{
"entropy": 6.587166833877563,
"epoch": 2.3510691198408753,
"grad_norm": 1.0625,
"learning_rate": 0.0004752454721224424,
"loss": 6.3494,
"mean_token_accuracy": 0.1418745554983616,
"num_tokens": 4630878.0,
"step": 2365
},
{
"entropy": 6.643035554885865,
"epoch": 2.3560417702635505,
"grad_norm": 1.1171875,
"learning_rate": 0.00047506708921849535,
"loss": 6.4744,
"mean_token_accuracy": 0.13255613371729852,
"num_tokens": 4640582.0,
"step": 2370
},
{
"entropy": 6.524079608917236,
"epoch": 2.361014420686226,
"grad_norm": 1.0390625,
"learning_rate": 0.0004748881035910269,
"loss": 6.394,
"mean_token_accuracy": 0.14071099683642388,
"num_tokens": 4650799.0,
"step": 2375
},
{
"entropy": 6.599751138687134,
"epoch": 2.365987071108901,
"grad_norm": 1.046875,
"learning_rate": 0.0004747085157792504,
"loss": 6.3499,
"mean_token_accuracy": 0.14517031535506247,
"num_tokens": 4660566.0,
"step": 2380
},
{
"entropy": 6.573225212097168,
"epoch": 2.3709597215315763,
"grad_norm": 1.109375,
"learning_rate": 0.0004745283263241933,
"loss": 6.4453,
"mean_token_accuracy": 0.13409614190459251,
"num_tokens": 4670234.0,
"step": 2385
},
{
"entropy": 6.540919303894043,
"epoch": 2.3759323719542516,
"grad_norm": 1.0546875,
"learning_rate": 0.0004743475357686957,
"loss": 6.3862,
"mean_token_accuracy": 0.14218721091747283,
"num_tokens": 4681114.0,
"step": 2390
},
{
"entropy": 6.505688333511353,
"epoch": 2.380905022376927,
"grad_norm": 1.03125,
"learning_rate": 0.00047416614465740843,
"loss": 6.1865,
"mean_token_accuracy": 0.16146760135889054,
"num_tokens": 4691970.0,
"step": 2395
},
{
"entropy": 6.524224042892456,
"epoch": 2.385877672799602,
"grad_norm": 1.03125,
"learning_rate": 0.00047398415353679167,
"loss": 6.4015,
"mean_token_accuracy": 0.13508354276418685,
"num_tokens": 4701723.0,
"step": 2400
},
{
"entropy": 6.523842716217041,
"epoch": 2.3908503232222773,
"grad_norm": 1.171875,
"learning_rate": 0.0004738015629551132,
"loss": 6.3611,
"mean_token_accuracy": 0.1428997755050659,
"num_tokens": 4711356.0,
"step": 2405
},
{
"entropy": 6.502122688293457,
"epoch": 2.395822973644953,
"grad_norm": 1.1015625,
"learning_rate": 0.00047361837346244655,
"loss": 6.4016,
"mean_token_accuracy": 0.14505933225154877,
"num_tokens": 4721765.0,
"step": 2410
},
{
"entropy": 6.5327929019927975,
"epoch": 2.4007956240676283,
"grad_norm": 1.0625,
"learning_rate": 0.0004734345856106698,
"loss": 6.2746,
"mean_token_accuracy": 0.14620665311813355,
"num_tokens": 4731678.0,
"step": 2415
},
{
"entropy": 6.482340097427368,
"epoch": 2.4057682744903035,
"grad_norm": 1.109375,
"learning_rate": 0.00047325019995346355,
"loss": 6.3168,
"mean_token_accuracy": 0.14606815874576567,
"num_tokens": 4740960.0,
"step": 2420
},
{
"entropy": 6.501438188552856,
"epoch": 2.4107409249129788,
"grad_norm": 1.0078125,
"learning_rate": 0.0004730652170463091,
"loss": 6.4316,
"mean_token_accuracy": 0.13504584059119223,
"num_tokens": 4751753.0,
"step": 2425
},
{
"entropy": 6.555155038833618,
"epoch": 2.415713575335654,
"grad_norm": 1.015625,
"learning_rate": 0.0004728796374464875,
"loss": 6.2847,
"mean_token_accuracy": 0.14447317868471146,
"num_tokens": 4761274.0,
"step": 2430
},
{
"entropy": 6.510864734649658,
"epoch": 2.4206862257583293,
"grad_norm": 1.171875,
"learning_rate": 0.00047269346171307716,
"loss": 6.4626,
"mean_token_accuracy": 0.14278751909732817,
"num_tokens": 4770852.0,
"step": 2435
},
{
"entropy": 6.688248014450073,
"epoch": 2.4256588761810045,
"grad_norm": 1.078125,
"learning_rate": 0.0004725066904069523,
"loss": 6.3209,
"mean_token_accuracy": 0.1462319627404213,
"num_tokens": 4780857.0,
"step": 2440
},
{
"entropy": 6.426857948303223,
"epoch": 2.4306315266036798,
"grad_norm": 1.140625,
"learning_rate": 0.0004723193240907814,
"loss": 6.2768,
"mean_token_accuracy": 0.1379391722381115,
"num_tokens": 4790018.0,
"step": 2445
},
{
"entropy": 6.520555019378662,
"epoch": 2.435604177026355,
"grad_norm": 1.09375,
"learning_rate": 0.0004721313633290257,
"loss": 6.3764,
"mean_token_accuracy": 0.1447875529527664,
"num_tokens": 4799747.0,
"step": 2450
},
{
"entropy": 6.63818507194519,
"epoch": 2.4405768274490303,
"grad_norm": 1.0703125,
"learning_rate": 0.000471942808687937,
"loss": 6.4086,
"mean_token_accuracy": 0.13961422741413115,
"num_tokens": 4809710.0,
"step": 2455
},
{
"entropy": 6.511287641525269,
"epoch": 2.4455494778717055,
"grad_norm": 1.15625,
"learning_rate": 0.00047175366073555654,
"loss": 6.3511,
"mean_token_accuracy": 0.14600509107112886,
"num_tokens": 4819387.0,
"step": 2460
},
{
"entropy": 6.471326446533203,
"epoch": 2.4505221282943808,
"grad_norm": 1.0546875,
"learning_rate": 0.0004715639200417125,
"loss": 6.29,
"mean_token_accuracy": 0.1439615435898304,
"num_tokens": 4829541.0,
"step": 2465
},
{
"entropy": 6.4300775051116945,
"epoch": 2.455494778717056,
"grad_norm": 1.078125,
"learning_rate": 0.0004713735871780192,
"loss": 6.2899,
"mean_token_accuracy": 0.14560722708702087,
"num_tokens": 4838699.0,
"step": 2470
},
{
"entropy": 6.544748401641845,
"epoch": 2.4604674291397313,
"grad_norm": 1.0546875,
"learning_rate": 0.0004711826627178747,
"loss": 6.4294,
"mean_token_accuracy": 0.14071948751807212,
"num_tokens": 4848295.0,
"step": 2475
},
{
"entropy": 6.534548664093018,
"epoch": 2.4654400795624065,
"grad_norm": 1.1875,
"learning_rate": 0.00047099114723645944,
"loss": 6.2891,
"mean_token_accuracy": 0.14699965268373488,
"num_tokens": 4857649.0,
"step": 2480
},
{
"entropy": 6.547018003463745,
"epoch": 2.470412729985082,
"grad_norm": 1.09375,
"learning_rate": 0.00047079904131073434,
"loss": 6.3242,
"mean_token_accuracy": 0.1475421205163002,
"num_tokens": 4867332.0,
"step": 2485
},
{
"entropy": 6.512684297561646,
"epoch": 2.4753853804077575,
"grad_norm": 1.1015625,
"learning_rate": 0.0004706063455194389,
"loss": 6.3853,
"mean_token_accuracy": 0.1304320976138115,
"num_tokens": 4878726.0,
"step": 2490
},
{
"entropy": 6.583101511001587,
"epoch": 2.4803580308304327,
"grad_norm": 1.1796875,
"learning_rate": 0.00047041306044308986,
"loss": 6.4034,
"mean_token_accuracy": 0.13978520780801773,
"num_tokens": 4888030.0,
"step": 2495
},
{
"entropy": 6.534812021255493,
"epoch": 2.485330681253108,
"grad_norm": 1.0625,
"learning_rate": 0.0004702191866639792,
"loss": 6.3693,
"mean_token_accuracy": 0.1370741046965122,
"num_tokens": 4897191.0,
"step": 2500
},
{
"epoch": 2.485330681253108,
"eval_entropy": 6.471034149969777,
"eval_loss": 6.821934700012207,
"eval_mean_token_accuracy": 0.1270321078581332,
"eval_num_tokens": 4897191.0,
"eval_runtime": 2.7348,
"eval_samples_per_second": 1267.006,
"eval_steps_per_second": 158.696,
"step": 2500
},
{
"entropy": 6.540355014801025,
"epoch": 2.490303331675783,
"grad_norm": 1.078125,
"learning_rate": 0.0004700247247661726,
"loss": 6.3653,
"mean_token_accuracy": 0.14543634057044982,
"num_tokens": 4907154.0,
"step": 2505
},
{
"entropy": 6.318873071670533,
"epoch": 2.4952759820984585,
"grad_norm": 1.421875,
"learning_rate": 0.0004698296753355072,
"loss": 6.0402,
"mean_token_accuracy": 0.1681458830833435,
"num_tokens": 4917582.0,
"step": 2510
},
{
"entropy": 6.464159154891968,
"epoch": 2.5002486325211337,
"grad_norm": 1.1484375,
"learning_rate": 0.0004696340389595903,
"loss": 6.2884,
"mean_token_accuracy": 0.1530615746974945,
"num_tokens": 4927473.0,
"step": 2515
},
{
"entropy": 6.559104919433594,
"epoch": 2.505221282943809,
"grad_norm": 1.0078125,
"learning_rate": 0.00046943781622779733,
"loss": 6.4407,
"mean_token_accuracy": 0.13506940230727196,
"num_tokens": 4938037.0,
"step": 2520
},
{
"entropy": 6.527531147003174,
"epoch": 2.510193933366484,
"grad_norm": 1.0,
"learning_rate": 0.00046924100773127055,
"loss": 6.398,
"mean_token_accuracy": 0.13462310209870337,
"num_tokens": 4948009.0,
"step": 2525
},
{
"entropy": 6.499567365646362,
"epoch": 2.5151665837891595,
"grad_norm": 1.015625,
"learning_rate": 0.00046904361406291633,
"loss": 6.371,
"mean_token_accuracy": 0.14283587485551835,
"num_tokens": 4957971.0,
"step": 2530
},
{
"entropy": 6.621564435958862,
"epoch": 2.520139234211835,
"grad_norm": 1.0859375,
"learning_rate": 0.0004688456358174044,
"loss": 6.3498,
"mean_token_accuracy": 0.1347363695502281,
"num_tokens": 4968040.0,
"step": 2535
},
{
"entropy": 6.478594255447388,
"epoch": 2.5251118846345104,
"grad_norm": 1.1171875,
"learning_rate": 0.0004686470735911653,
"loss": 6.3009,
"mean_token_accuracy": 0.14506892040371894,
"num_tokens": 4977142.0,
"step": 2540
},
{
"entropy": 6.497954607009888,
"epoch": 2.5300845350571857,
"grad_norm": 1.1875,
"learning_rate": 0.0004684479279823891,
"loss": 6.4449,
"mean_token_accuracy": 0.14204382449388503,
"num_tokens": 4986726.0,
"step": 2545
},
{
"entropy": 6.622348499298096,
"epoch": 2.535057185479861,
"grad_norm": 1.0390625,
"learning_rate": 0.0004682481995910232,
"loss": 6.3631,
"mean_token_accuracy": 0.13624119982123375,
"num_tokens": 4996690.0,
"step": 2550
},
{
"entropy": 6.484410619735717,
"epoch": 2.540029835902536,
"grad_norm": 1.1640625,
"learning_rate": 0.00046804788901877084,
"loss": 6.4084,
"mean_token_accuracy": 0.14261124581098555,
"num_tokens": 5005804.0,
"step": 2555
},
{
"entropy": 6.4933616638183596,
"epoch": 2.5450024863252114,
"grad_norm": 1.0625,
"learning_rate": 0.00046784699686908886,
"loss": 6.3458,
"mean_token_accuracy": 0.14723647236824036,
"num_tokens": 5015666.0,
"step": 2560
},
{
"entropy": 6.542861223220825,
"epoch": 2.5499751367478867,
"grad_norm": 1.03125,
"learning_rate": 0.00046764552374718656,
"loss": 6.2782,
"mean_token_accuracy": 0.15213844776153565,
"num_tokens": 5025897.0,
"step": 2565
},
{
"entropy": 6.4771161556243895,
"epoch": 2.554947787170562,
"grad_norm": 1.046875,
"learning_rate": 0.00046744347026002315,
"loss": 6.3501,
"mean_token_accuracy": 0.14428422451019288,
"num_tokens": 5035195.0,
"step": 2570
},
{
"entropy": 6.505103015899659,
"epoch": 2.559920437593237,
"grad_norm": 1.109375,
"learning_rate": 0.0004672408370163066,
"loss": 6.3556,
"mean_token_accuracy": 0.1479845479130745,
"num_tokens": 5044472.0,
"step": 2575
},
{
"entropy": 6.518973064422608,
"epoch": 2.5648930880159124,
"grad_norm": 1.1171875,
"learning_rate": 0.000467037624626491,
"loss": 6.4082,
"mean_token_accuracy": 0.13811224997043609,
"num_tokens": 5053593.0,
"step": 2580
},
{
"entropy": 6.514780187606812,
"epoch": 2.5698657384385877,
"grad_norm": 1.1015625,
"learning_rate": 0.00046683383370277554,
"loss": 6.3632,
"mean_token_accuracy": 0.1389802947640419,
"num_tokens": 5063604.0,
"step": 2585
},
{
"entropy": 6.471893930435181,
"epoch": 2.574838388861263,
"grad_norm": 1.0,
"learning_rate": 0.00046662946485910226,
"loss": 6.2686,
"mean_token_accuracy": 0.14278726428747177,
"num_tokens": 5073013.0,
"step": 2590
},
{
"entropy": 6.501459407806396,
"epoch": 2.579811039283938,
"grad_norm": 1.03125,
"learning_rate": 0.000466424518711154,
"loss": 6.353,
"mean_token_accuracy": 0.14196558594703673,
"num_tokens": 5082590.0,
"step": 2595
},
{
"entropy": 6.558322525024414,
"epoch": 2.5847836897066134,
"grad_norm": 1.0859375,
"learning_rate": 0.0004662189958763531,
"loss": 6.3299,
"mean_token_accuracy": 0.14453306496143342,
"num_tokens": 5091860.0,
"step": 2600
},
{
"entropy": 6.4649756908416744,
"epoch": 2.5897563401292887,
"grad_norm": 1.09375,
"learning_rate": 0.0004660128969738591,
"loss": 6.2484,
"mean_token_accuracy": 0.1543172523379326,
"num_tokens": 5101164.0,
"step": 2605
},
{
"entropy": 6.371197319030761,
"epoch": 2.594728990551964,
"grad_norm": 0.9453125,
"learning_rate": 0.0004658062226245668,
"loss": 6.3489,
"mean_token_accuracy": 0.14326864182949067,
"num_tokens": 5112885.0,
"step": 2610
},
{
"entropy": 6.522914695739746,
"epoch": 2.5997016409746396,
"grad_norm": 0.92578125,
"learning_rate": 0.0004655989734511052,
"loss": 6.3226,
"mean_token_accuracy": 0.14789563715457915,
"num_tokens": 5124063.0,
"step": 2615
},
{
"entropy": 6.366082334518433,
"epoch": 2.604674291397315,
"grad_norm": 1.0078125,
"learning_rate": 0.0004653911500778343,
"loss": 6.1761,
"mean_token_accuracy": 0.1581413634121418,
"num_tokens": 5134286.0,
"step": 2620
},
{
"entropy": 6.534527587890625,
"epoch": 2.60964694181999,
"grad_norm": 1.1328125,
"learning_rate": 0.0004651827531308443,
"loss": 6.4104,
"mean_token_accuracy": 0.14058353900909423,
"num_tokens": 5144421.0,
"step": 2625
},
{
"entropy": 6.610892343521118,
"epoch": 2.6146195922426654,
"grad_norm": 1.0625,
"learning_rate": 0.0004649737832379534,
"loss": 6.3732,
"mean_token_accuracy": 0.1419479578733444,
"num_tokens": 5153466.0,
"step": 2630
},
{
"entropy": 6.443132543563843,
"epoch": 2.6195922426653406,
"grad_norm": 0.99609375,
"learning_rate": 0.0004647642410287057,
"loss": 6.4243,
"mean_token_accuracy": 0.14085619822144507,
"num_tokens": 5164034.0,
"step": 2635
},
{
"entropy": 6.504536724090576,
"epoch": 2.624564893088016,
"grad_norm": 1.03125,
"learning_rate": 0.0004645541271343696,
"loss": 6.3538,
"mean_token_accuracy": 0.14277144595980645,
"num_tokens": 5173867.0,
"step": 2640
},
{
"entropy": 6.540040922164917,
"epoch": 2.629537543510691,
"grad_norm": 1.109375,
"learning_rate": 0.00046434344218793566,
"loss": 6.399,
"mean_token_accuracy": 0.14028803780674934,
"num_tokens": 5183182.0,
"step": 2645
},
{
"entropy": 6.518236303329468,
"epoch": 2.6345101939333664,
"grad_norm": 1.0234375,
"learning_rate": 0.00046413218682411485,
"loss": 6.3741,
"mean_token_accuracy": 0.14554732367396356,
"num_tokens": 5192537.0,
"step": 2650
},
{
"entropy": 6.556453895568848,
"epoch": 2.6394828443560416,
"grad_norm": 0.98828125,
"learning_rate": 0.0004639203616793365,
"loss": 6.3197,
"mean_token_accuracy": 0.1478041559457779,
"num_tokens": 5201879.0,
"step": 2655
},
{
"entropy": 6.537673616409302,
"epoch": 2.6444554947787173,
"grad_norm": 1.046875,
"learning_rate": 0.00046370796739174657,
"loss": 6.3579,
"mean_token_accuracy": 0.13904747143387794,
"num_tokens": 5211555.0,
"step": 2660
},
{
"entropy": 6.459664249420166,
"epoch": 2.6494281452013926,
"grad_norm": 1.140625,
"learning_rate": 0.00046349500460120556,
"loss": 6.2529,
"mean_token_accuracy": 0.15000981390476226,
"num_tokens": 5220895.0,
"step": 2665
},
{
"entropy": 6.539349603652954,
"epoch": 2.654400795624068,
"grad_norm": 1.0390625,
"learning_rate": 0.00046328147394928684,
"loss": 6.4012,
"mean_token_accuracy": 0.14086081832647324,
"num_tokens": 5230101.0,
"step": 2670
},
{
"entropy": 6.559596300125122,
"epoch": 2.659373446046743,
"grad_norm": 1.0078125,
"learning_rate": 0.0004630673760792741,
"loss": 6.3814,
"mean_token_accuracy": 0.13394856974482536,
"num_tokens": 5239472.0,
"step": 2675
},
{
"entropy": 6.457580184936523,
"epoch": 2.6643460964694183,
"grad_norm": 1.21875,
"learning_rate": 0.0004628527116361603,
"loss": 6.3162,
"mean_token_accuracy": 0.1445642292499542,
"num_tokens": 5248823.0,
"step": 2680
},
{
"entropy": 6.471943759918213,
"epoch": 2.6693187468920936,
"grad_norm": 1.1015625,
"learning_rate": 0.0004626374812666449,
"loss": 6.3291,
"mean_token_accuracy": 0.13978914096951484,
"num_tokens": 5258551.0,
"step": 2685
},
{
"entropy": 6.450301027297973,
"epoch": 2.674291397314769,
"grad_norm": 1.0390625,
"learning_rate": 0.00046242168561913256,
"loss": 6.1373,
"mean_token_accuracy": 0.1550034001469612,
"num_tokens": 5267933.0,
"step": 2690
},
{
"entropy": 6.545872449874878,
"epoch": 2.679264047737444,
"grad_norm": 1.1171875,
"learning_rate": 0.00046220532534373067,
"loss": 6.3955,
"mean_token_accuracy": 0.14434008225798606,
"num_tokens": 5277856.0,
"step": 2695
},
{
"entropy": 6.508489274978638,
"epoch": 2.6842366981601193,
"grad_norm": 1.046875,
"learning_rate": 0.0004619884010922477,
"loss": 6.3651,
"mean_token_accuracy": 0.13867145925760269,
"num_tokens": 5288127.0,
"step": 2700
},
{
"entropy": 6.476871538162231,
"epoch": 2.6892093485827946,
"grad_norm": 1.0703125,
"learning_rate": 0.0004617709135181913,
"loss": 6.3386,
"mean_token_accuracy": 0.14043029323220252,
"num_tokens": 5297143.0,
"step": 2705
},
{
"entropy": 6.433478450775146,
"epoch": 2.69418199900547,
"grad_norm": 1.2578125,
"learning_rate": 0.00046155286327676594,
"loss": 6.315,
"mean_token_accuracy": 0.14332007616758347,
"num_tokens": 5307067.0,
"step": 2710
},
{
"entropy": 6.450258541107178,
"epoch": 2.699154649428145,
"grad_norm": 1.046875,
"learning_rate": 0.00046133425102487134,
"loss": 6.2566,
"mean_token_accuracy": 0.14392766803503038,
"num_tokens": 5317471.0,
"step": 2715
},
{
"entropy": 6.559765481948853,
"epoch": 2.7041272998508203,
"grad_norm": 1.0546875,
"learning_rate": 0.00046111507742110035,
"loss": 6.3351,
"mean_token_accuracy": 0.15104418098926545,
"num_tokens": 5326803.0,
"step": 2720
},
{
"entropy": 6.460714960098267,
"epoch": 2.7090999502734956,
"grad_norm": 1.0703125,
"learning_rate": 0.000460895343125737,
"loss": 6.311,
"mean_token_accuracy": 0.14636903926730155,
"num_tokens": 5336342.0,
"step": 2725
},
{
"entropy": 6.327510404586792,
"epoch": 2.714072600696171,
"grad_norm": 1.125,
"learning_rate": 0.00046067504880075425,
"loss": 6.1531,
"mean_token_accuracy": 0.16241607964038848,
"num_tokens": 5345694.0,
"step": 2730
},
{
"entropy": 6.50901460647583,
"epoch": 2.719045251118846,
"grad_norm": 1.046875,
"learning_rate": 0.0004604541951098122,
"loss": 6.2289,
"mean_token_accuracy": 0.15025691464543342,
"num_tokens": 5355277.0,
"step": 2735
},
{
"entropy": 6.397147083282471,
"epoch": 2.724017901541522,
"grad_norm": 1.09375,
"learning_rate": 0.0004602327827182565,
"loss": 6.2651,
"mean_token_accuracy": 0.14940724670886993,
"num_tokens": 5364650.0,
"step": 2740
},
{
"entropy": 6.462321996688843,
"epoch": 2.728990551964197,
"grad_norm": 1.171875,
"learning_rate": 0.0004600108122931156,
"loss": 6.2447,
"mean_token_accuracy": 0.14868565499782563,
"num_tokens": 5373151.0,
"step": 2745
},
{
"entropy": 6.449673080444336,
"epoch": 2.7339632023868723,
"grad_norm": 0.921875,
"learning_rate": 0.00045978828450309907,
"loss": 6.3625,
"mean_token_accuracy": 0.14481861144304276,
"num_tokens": 5383093.0,
"step": 2750
},
{
"entropy": 6.545788192749024,
"epoch": 2.7389358528095475,
"grad_norm": 2.015625,
"learning_rate": 0.0004595652000185958,
"loss": 6.2861,
"mean_token_accuracy": 0.14802829027175904,
"num_tokens": 5392060.0,
"step": 2755
},
{
"entropy": 6.469894981384277,
"epoch": 2.743908503232223,
"grad_norm": 1.0,
"learning_rate": 0.00045934155951167163,
"loss": 6.3925,
"mean_token_accuracy": 0.1392727740108967,
"num_tokens": 5402119.0,
"step": 2760
},
{
"entropy": 6.519701528549194,
"epoch": 2.748881153654898,
"grad_norm": 1.1328125,
"learning_rate": 0.0004591173636560675,
"loss": 6.2894,
"mean_token_accuracy": 0.15115260183811188,
"num_tokens": 5411804.0,
"step": 2765
},
{
"entropy": 6.454624652862549,
"epoch": 2.7538538040775733,
"grad_norm": 1.0078125,
"learning_rate": 0.0004588926131271974,
"loss": 6.3411,
"mean_token_accuracy": 0.1430234581232071,
"num_tokens": 5423808.0,
"step": 2770
},
{
"entropy": 6.478226995468139,
"epoch": 2.7588264545002485,
"grad_norm": 1.03125,
"learning_rate": 0.00045866730860214655,
"loss": 6.3122,
"mean_token_accuracy": 0.14709860384464263,
"num_tokens": 5432154.0,
"step": 2775
},
{
"entropy": 6.300262498855591,
"epoch": 2.763799104922924,
"grad_norm": 1.03125,
"learning_rate": 0.0004584414507596687,
"loss": 6.0542,
"mean_token_accuracy": 0.17085577547550201,
"num_tokens": 5443006.0,
"step": 2780
},
{
"entropy": 6.417803001403809,
"epoch": 2.768771755345599,
"grad_norm": 1.03125,
"learning_rate": 0.00045821504028018494,
"loss": 6.2355,
"mean_token_accuracy": 0.14497338235378265,
"num_tokens": 5453596.0,
"step": 2785
},
{
"entropy": 6.4752153873443605,
"epoch": 2.7737444057682747,
"grad_norm": 1.0703125,
"learning_rate": 0.0004579880778457811,
"loss": 6.3689,
"mean_token_accuracy": 0.14066243916749954,
"num_tokens": 5463048.0,
"step": 2790
},
{
"entropy": 6.467712926864624,
"epoch": 2.77871705619095,
"grad_norm": 1.1328125,
"learning_rate": 0.0004577605641402056,
"loss": 6.2947,
"mean_token_accuracy": 0.14769297540187837,
"num_tokens": 5472409.0,
"step": 2795
},
{
"entropy": 6.441144371032715,
"epoch": 2.7836897066136252,
"grad_norm": 1.0859375,
"learning_rate": 0.0004575324998488681,
"loss": 6.2354,
"mean_token_accuracy": 0.15422018766403198,
"num_tokens": 5481380.0,
"step": 2800
},
{
"entropy": 6.496242666244507,
"epoch": 2.7886623570363005,
"grad_norm": 0.99609375,
"learning_rate": 0.0004573038856588364,
"loss": 6.4379,
"mean_token_accuracy": 0.1381148561835289,
"num_tokens": 5491157.0,
"step": 2805
},
{
"entropy": 6.40307035446167,
"epoch": 2.7936350074589757,
"grad_norm": 1.015625,
"learning_rate": 0.0004570747222588354,
"loss": 6.3196,
"mean_token_accuracy": 0.14346076548099518,
"num_tokens": 5500856.0,
"step": 2810
},
{
"entropy": 6.47411961555481,
"epoch": 2.798607657881651,
"grad_norm": 1.15625,
"learning_rate": 0.00045684501033924443,
"loss": 6.2046,
"mean_token_accuracy": 0.14963931888341903,
"num_tokens": 5509654.0,
"step": 2815
},
{
"entropy": 6.401515960693359,
"epoch": 2.8035803083043263,
"grad_norm": 1.140625,
"learning_rate": 0.00045661475059209503,
"loss": 6.2211,
"mean_token_accuracy": 0.15217607766389846,
"num_tokens": 5519009.0,
"step": 2820
},
{
"entropy": 6.397446584701538,
"epoch": 2.8085529587270015,
"grad_norm": 1.2265625,
"learning_rate": 0.0004563839437110695,
"loss": 6.2338,
"mean_token_accuracy": 0.15214656740427018,
"num_tokens": 5528316.0,
"step": 2825
},
{
"entropy": 6.471298694610596,
"epoch": 2.8135256091496768,
"grad_norm": 1.1015625,
"learning_rate": 0.00045615259039149826,
"loss": 6.3488,
"mean_token_accuracy": 0.1476473778486252,
"num_tokens": 5538416.0,
"step": 2830
},
{
"entropy": 6.4038327693939205,
"epoch": 2.818498259572352,
"grad_norm": 0.9921875,
"learning_rate": 0.00045592069133035783,
"loss": 6.2869,
"mean_token_accuracy": 0.1492191307246685,
"num_tokens": 5548600.0,
"step": 2835
},
{
"entropy": 6.528113079071045,
"epoch": 2.8234709099950273,
"grad_norm": 1.0546875,
"learning_rate": 0.0004556882472262691,
"loss": 6.3799,
"mean_token_accuracy": 0.1402927130460739,
"num_tokens": 5559454.0,
"step": 2840
},
{
"entropy": 6.46011061668396,
"epoch": 2.8284435604177025,
"grad_norm": 1.234375,
"learning_rate": 0.0004554552587794948,
"loss": 6.3378,
"mean_token_accuracy": 0.13970609605312348,
"num_tokens": 5567748.0,
"step": 2845
},
{
"entropy": 6.495553874969483,
"epoch": 2.8334162108403778,
"grad_norm": 1.0390625,
"learning_rate": 0.0004552217266919375,
"loss": 6.4256,
"mean_token_accuracy": 0.13631942868232727,
"num_tokens": 5577617.0,
"step": 2850
},
{
"entropy": 6.5852532386779785,
"epoch": 2.838388861263053,
"grad_norm": 1.1171875,
"learning_rate": 0.0004549876516671376,
"loss": 6.3907,
"mean_token_accuracy": 0.14207628220319748,
"num_tokens": 5586600.0,
"step": 2855
},
{
"entropy": 6.4586786270141605,
"epoch": 2.8433615116857283,
"grad_norm": 1.0703125,
"learning_rate": 0.00045475303441027114,
"loss": 6.2216,
"mean_token_accuracy": 0.1577283725142479,
"num_tokens": 5596526.0,
"step": 2860
},
{
"entropy": 6.426614618301391,
"epoch": 2.8483341621084035,
"grad_norm": 1.078125,
"learning_rate": 0.0004545178756281479,
"loss": 6.2625,
"mean_token_accuracy": 0.15038749873638152,
"num_tokens": 5605780.0,
"step": 2865
},
{
"entropy": 6.41419792175293,
"epoch": 2.853306812531079,
"grad_norm": 1.0,
"learning_rate": 0.00045428217602920876,
"loss": 6.3216,
"mean_token_accuracy": 0.14313239157199859,
"num_tokens": 5616109.0,
"step": 2870
},
{
"entropy": 6.429047727584839,
"epoch": 2.8582794629537545,
"grad_norm": 1.1484375,
"learning_rate": 0.00045404593632352385,
"loss": 6.275,
"mean_token_accuracy": 0.15239612609148026,
"num_tokens": 5625423.0,
"step": 2875
},
{
"entropy": 6.475142240524292,
"epoch": 2.8632521133764297,
"grad_norm": 1.28125,
"learning_rate": 0.0004538091572227909,
"loss": 6.3661,
"mean_token_accuracy": 0.14628186374902724,
"num_tokens": 5634538.0,
"step": 2880
},
{
"entropy": 6.36572036743164,
"epoch": 2.868224763799105,
"grad_norm": 1.0703125,
"learning_rate": 0.0004535718394403319,
"loss": 6.1448,
"mean_token_accuracy": 0.16420113444328308,
"num_tokens": 5644087.0,
"step": 2885
},
{
"entropy": 6.331579542160034,
"epoch": 2.87319741422178,
"grad_norm": 1.1484375,
"learning_rate": 0.0004533339836910924,
"loss": 6.1773,
"mean_token_accuracy": 0.15930747836828232,
"num_tokens": 5654241.0,
"step": 2890
},
{
"entropy": 6.4233019828796385,
"epoch": 2.8781700646444555,
"grad_norm": 1.046875,
"learning_rate": 0.0004530955906916382,
"loss": 6.2268,
"mean_token_accuracy": 0.1520026817917824,
"num_tokens": 5663827.0,
"step": 2895
},
{
"entropy": 6.43547625541687,
"epoch": 2.8831427150671307,
"grad_norm": 1.0390625,
"learning_rate": 0.0004528566611601536,
"loss": 6.3226,
"mean_token_accuracy": 0.14533980414271355,
"num_tokens": 5673427.0,
"step": 2900
},
{
"entropy": 6.36797866821289,
"epoch": 2.888115365489806,
"grad_norm": 0.98046875,
"learning_rate": 0.0004526171958164396,
"loss": 6.2047,
"mean_token_accuracy": 0.14768766909837722,
"num_tokens": 5684762.0,
"step": 2905
},
{
"entropy": 6.3772743225097654,
"epoch": 2.893088015912481,
"grad_norm": 1.125,
"learning_rate": 0.00045237719538191107,
"loss": 6.309,
"mean_token_accuracy": 0.14853577613830565,
"num_tokens": 5694938.0,
"step": 2910
},
{
"entropy": 6.417752361297607,
"epoch": 2.898060666335157,
"grad_norm": 1.078125,
"learning_rate": 0.000452136660579595,
"loss": 6.2306,
"mean_token_accuracy": 0.1528899848461151,
"num_tokens": 5704910.0,
"step": 2915
},
{
"entropy": 6.385972118377685,
"epoch": 2.903033316757832,
"grad_norm": 1.0859375,
"learning_rate": 0.00045189559213412825,
"loss": 6.3146,
"mean_token_accuracy": 0.14834412410855294,
"num_tokens": 5714325.0,
"step": 2920
},
{
"entropy": 6.4172662734985355,
"epoch": 2.9080059671805074,
"grad_norm": 1.109375,
"learning_rate": 0.0004516539907717553,
"loss": 6.2152,
"mean_token_accuracy": 0.157168248295784,
"num_tokens": 5724382.0,
"step": 2925
},
{
"entropy": 6.403017520904541,
"epoch": 2.9129786176031827,
"grad_norm": 1.0703125,
"learning_rate": 0.00045141185722032617,
"loss": 6.2618,
"mean_token_accuracy": 0.1492059126496315,
"num_tokens": 5734277.0,
"step": 2930
},
{
"entropy": 6.416959381103515,
"epoch": 2.917951268025858,
"grad_norm": 1.1015625,
"learning_rate": 0.00045116919220929407,
"loss": 6.254,
"mean_token_accuracy": 0.14605386927723885,
"num_tokens": 5742772.0,
"step": 2935
},
{
"entropy": 6.405718898773193,
"epoch": 2.922923918448533,
"grad_norm": 1.03125,
"learning_rate": 0.0004509259964697133,
"loss": 6.2743,
"mean_token_accuracy": 0.14779454469680786,
"num_tokens": 5754011.0,
"step": 2940
},
{
"entropy": 6.556613206863403,
"epoch": 2.9278965688712084,
"grad_norm": 0.9765625,
"learning_rate": 0.000450682270734237,
"loss": 6.3732,
"mean_token_accuracy": 0.13893503844738006,
"num_tokens": 5765445.0,
"step": 2945
},
{
"entropy": 6.4057858943939205,
"epoch": 2.9328692192938837,
"grad_norm": 1.0546875,
"learning_rate": 0.0004504380157371152,
"loss": 6.3165,
"mean_token_accuracy": 0.15030984133481978,
"num_tokens": 5775529.0,
"step": 2950
},
{
"entropy": 6.451057004928589,
"epoch": 2.937841869716559,
"grad_norm": 1.109375,
"learning_rate": 0.00045019323221419204,
"loss": 6.4231,
"mean_token_accuracy": 0.14229757115244865,
"num_tokens": 5785594.0,
"step": 2955
},
{
"entropy": 6.43241868019104,
"epoch": 2.942814520139234,
"grad_norm": 1.0703125,
"learning_rate": 0.00044994792090290425,
"loss": 6.1403,
"mean_token_accuracy": 0.16009896695613862,
"num_tokens": 5796071.0,
"step": 2960
},
{
"entropy": 6.375617933273316,
"epoch": 2.9477871705619094,
"grad_norm": 1.15625,
"learning_rate": 0.00044970208254227836,
"loss": 6.2772,
"mean_token_accuracy": 0.1521676555275917,
"num_tokens": 5805127.0,
"step": 2965
},
{
"entropy": 6.472930192947388,
"epoch": 2.9527598209845847,
"grad_norm": 1.0390625,
"learning_rate": 0.00044945571787292866,
"loss": 6.3395,
"mean_token_accuracy": 0.1493378311395645,
"num_tokens": 5815001.0,
"step": 2970
},
{
"entropy": 6.419169139862061,
"epoch": 2.95773247140726,
"grad_norm": 1.03125,
"learning_rate": 0.0004492088276370552,
"loss": 6.2974,
"mean_token_accuracy": 0.14073021113872528,
"num_tokens": 5826915.0,
"step": 2975
},
{
"entropy": 6.408809566497803,
"epoch": 2.962705121829935,
"grad_norm": 1.015625,
"learning_rate": 0.0004489614125784411,
"loss": 6.3013,
"mean_token_accuracy": 0.14516500532627105,
"num_tokens": 5837444.0,
"step": 2980
},
{
"entropy": 6.355221128463745,
"epoch": 2.9676777722526104,
"grad_norm": 1.125,
"learning_rate": 0.0004487134734424509,
"loss": 6.1628,
"mean_token_accuracy": 0.15315280109643936,
"num_tokens": 5846514.0,
"step": 2985
},
{
"entropy": 6.480633068084717,
"epoch": 2.9726504226752857,
"grad_norm": 1.046875,
"learning_rate": 0.0004484650109760276,
"loss": 6.4218,
"mean_token_accuracy": 0.14197667092084884,
"num_tokens": 5857227.0,
"step": 2990
},
{
"entropy": 6.504110956192017,
"epoch": 2.9776230730979614,
"grad_norm": 1.109375,
"learning_rate": 0.0004482160259276911,
"loss": 6.4022,
"mean_token_accuracy": 0.14497765749692917,
"num_tokens": 5866877.0,
"step": 2995
},
{
"entropy": 6.437817049026489,
"epoch": 2.9825957235206366,
"grad_norm": 1.171875,
"learning_rate": 0.0004479665190475355,
"loss": 6.2861,
"mean_token_accuracy": 0.14751956462860108,
"num_tokens": 5877127.0,
"step": 3000
},
{
"epoch": 2.9825957235206366,
"eval_entropy": 6.340936762945993,
"eval_loss": 6.701125144958496,
"eval_mean_token_accuracy": 0.13151445259524655,
"eval_num_tokens": 5877127.0,
"eval_runtime": 2.5452,
"eval_samples_per_second": 1361.38,
"eval_steps_per_second": 170.516,
"step": 3000
},
{
"entropy": 6.398074674606323,
"epoch": 2.987568373943312,
"grad_norm": 1.0703125,
"learning_rate": 0.00044771649108722695,
"loss": 6.3128,
"mean_token_accuracy": 0.14532861709594727,
"num_tokens": 5886434.0,
"step": 3005
},
{
"entropy": 6.389483833312989,
"epoch": 2.992541024365987,
"grad_norm": 1.140625,
"learning_rate": 0.0004474659428000014,
"loss": 6.2912,
"mean_token_accuracy": 0.15597790628671646,
"num_tokens": 5895462.0,
"step": 3010
},
{
"entropy": 6.451540756225586,
"epoch": 2.9975136747886624,
"grad_norm": 1.09375,
"learning_rate": 0.00044721487494066254,
"loss": 6.2394,
"mean_token_accuracy": 0.15187835693359375,
"num_tokens": 5905274.0,
"step": 3015
},
{
"entropy": 6.340188609229194,
"epoch": 3.00198906016907,
"grad_norm": 1.1796875,
"learning_rate": 0.000446963288265579,
"loss": 5.9949,
"mean_token_accuracy": 0.16156921121809217,
"num_tokens": 5914164.0,
"step": 3020
},
{
"entropy": 6.394064807891846,
"epoch": 3.0069617105917454,
"grad_norm": 0.984375,
"learning_rate": 0.00044671118353268257,
"loss": 5.8749,
"mean_token_accuracy": 0.15735693871974946,
"num_tokens": 5923159.0,
"step": 3025
},
{
"entropy": 6.308403873443604,
"epoch": 3.0119343610144207,
"grad_norm": 1.0078125,
"learning_rate": 0.00044645856150146593,
"loss": 5.8305,
"mean_token_accuracy": 0.16631314903497696,
"num_tokens": 5932708.0,
"step": 3030
},
{
"entropy": 6.30610146522522,
"epoch": 3.016907011437096,
"grad_norm": 1.1015625,
"learning_rate": 0.00044620542293297974,
"loss": 5.8641,
"mean_token_accuracy": 0.1618802011013031,
"num_tokens": 5941370.0,
"step": 3035
},
{
"entropy": 6.352138042449951,
"epoch": 3.021879661859771,
"grad_norm": 0.984375,
"learning_rate": 0.0004459517685898312,
"loss": 5.9036,
"mean_token_accuracy": 0.16566911041736604,
"num_tokens": 5950567.0,
"step": 3040
},
{
"entropy": 6.388893699645996,
"epoch": 3.0268523122824464,
"grad_norm": 1.1328125,
"learning_rate": 0.00044569759923618113,
"loss": 5.9756,
"mean_token_accuracy": 0.15771342366933822,
"num_tokens": 5959242.0,
"step": 3045
},
{
"entropy": 6.324466896057129,
"epoch": 3.0318249627051217,
"grad_norm": 1.0859375,
"learning_rate": 0.0004454429156377419,
"loss": 5.9412,
"mean_token_accuracy": 0.16309747397899627,
"num_tokens": 5968365.0,
"step": 3050
},
{
"entropy": 6.359807634353638,
"epoch": 3.036797613127797,
"grad_norm": 1.0703125,
"learning_rate": 0.0004451877185617751,
"loss": 5.9374,
"mean_token_accuracy": 0.15564917623996735,
"num_tokens": 5978407.0,
"step": 3055
},
{
"entropy": 6.239137601852417,
"epoch": 3.041770263550472,
"grad_norm": 1.1171875,
"learning_rate": 0.0004449320087770894,
"loss": 5.6659,
"mean_token_accuracy": 0.18490099608898164,
"num_tokens": 5988865.0,
"step": 3060
},
{
"entropy": 6.2805784225463865,
"epoch": 3.046742913973148,
"grad_norm": 1.03125,
"learning_rate": 0.0004446757870540376,
"loss": 5.851,
"mean_token_accuracy": 0.16899575367569925,
"num_tokens": 5998873.0,
"step": 3065
},
{
"entropy": 6.305179262161255,
"epoch": 3.051715564395823,
"grad_norm": 1.109375,
"learning_rate": 0.0004444190541645153,
"loss": 5.89,
"mean_token_accuracy": 0.16174780428409577,
"num_tokens": 6007021.0,
"step": 3070
},
{
"entropy": 6.248967933654785,
"epoch": 3.0566882148184984,
"grad_norm": 1.5625,
"learning_rate": 0.00044416181088195776,
"loss": 5.7122,
"mean_token_accuracy": 0.17303745299577714,
"num_tokens": 6015884.0,
"step": 3075
},
{
"entropy": 6.237709283828735,
"epoch": 3.0616608652411736,
"grad_norm": 1.0703125,
"learning_rate": 0.0004439040579813379,
"loss": 5.9689,
"mean_token_accuracy": 0.15646934062242507,
"num_tokens": 6026110.0,
"step": 3080
},
{
"entropy": 6.072254180908203,
"epoch": 3.066633515663849,
"grad_norm": 0.984375,
"learning_rate": 0.00044364579623916385,
"loss": 5.7213,
"mean_token_accuracy": 0.1819382682442665,
"num_tokens": 6037479.0,
"step": 3085
},
{
"entropy": 6.319383955001831,
"epoch": 3.071606166086524,
"grad_norm": 1.0859375,
"learning_rate": 0.0004433870264334767,
"loss": 5.8795,
"mean_token_accuracy": 0.15825727283954621,
"num_tokens": 6046743.0,
"step": 3090
},
{
"entropy": 6.270925521850586,
"epoch": 3.0765788165091994,
"grad_norm": 1.0078125,
"learning_rate": 0.0004431277493438484,
"loss": 5.9616,
"mean_token_accuracy": 0.1608537882566452,
"num_tokens": 6056982.0,
"step": 3095
},
{
"entropy": 6.2930597305297855,
"epoch": 3.0815514669318746,
"grad_norm": 1.1171875,
"learning_rate": 0.0004428679657513789,
"loss": 5.9251,
"mean_token_accuracy": 0.1641475811600685,
"num_tokens": 6066016.0,
"step": 3100
},
{
"entropy": 6.192068672180175,
"epoch": 3.08652411735455,
"grad_norm": 1.1640625,
"learning_rate": 0.0004426076764386939,
"loss": 5.8679,
"mean_token_accuracy": 0.15714501738548278,
"num_tokens": 6075149.0,
"step": 3105
},
{
"entropy": 6.311230945587158,
"epoch": 3.091496767777225,
"grad_norm": 1.078125,
"learning_rate": 0.00044234688218994293,
"loss": 5.8356,
"mean_token_accuracy": 0.1684956133365631,
"num_tokens": 6085736.0,
"step": 3110
},
{
"entropy": 6.12915735244751,
"epoch": 3.0964694181999004,
"grad_norm": 1.0546875,
"learning_rate": 0.0004420855837907967,
"loss": 5.7768,
"mean_token_accuracy": 0.16718606948852538,
"num_tokens": 6096557.0,
"step": 3115
},
{
"entropy": 6.4019359111785885,
"epoch": 3.1014420686225757,
"grad_norm": 1.2265625,
"learning_rate": 0.00044182378202844427,
"loss": 6.0902,
"mean_token_accuracy": 0.1507420502603054,
"num_tokens": 6105735.0,
"step": 3120
},
{
"entropy": 6.39102954864502,
"epoch": 3.106414719045251,
"grad_norm": 1.15625,
"learning_rate": 0.00044156147769159183,
"loss": 5.9196,
"mean_token_accuracy": 0.16611894369125366,
"num_tokens": 6114132.0,
"step": 3125
},
{
"entropy": 6.251520824432373,
"epoch": 3.1113873694679266,
"grad_norm": 1.21875,
"learning_rate": 0.0004412986715704591,
"loss": 5.9318,
"mean_token_accuracy": 0.15787497460842131,
"num_tokens": 6124341.0,
"step": 3130
},
{
"entropy": 6.195437812805176,
"epoch": 3.116360019890602,
"grad_norm": 0.96875,
"learning_rate": 0.0004410353644567777,
"loss": 5.8262,
"mean_token_accuracy": 0.16261177211999894,
"num_tokens": 6134418.0,
"step": 3135
},
{
"entropy": 6.287057399749756,
"epoch": 3.121332670313277,
"grad_norm": 1.0703125,
"learning_rate": 0.00044077155714378836,
"loss": 5.9001,
"mean_token_accuracy": 0.15995267778635025,
"num_tokens": 6143659.0,
"step": 3140
},
{
"entropy": 6.270061492919922,
"epoch": 3.1263053207359524,
"grad_norm": 1.2421875,
"learning_rate": 0.0004405072504262391,
"loss": 5.9224,
"mean_token_accuracy": 0.1607741340994835,
"num_tokens": 6152706.0,
"step": 3145
},
{
"entropy": 6.331454181671143,
"epoch": 3.1312779711586276,
"grad_norm": 1.1953125,
"learning_rate": 0.000440242445100382,
"loss": 5.9809,
"mean_token_accuracy": 0.15174245834350586,
"num_tokens": 6161390.0,
"step": 3150
},
{
"entropy": 6.332275772094727,
"epoch": 3.136250621581303,
"grad_norm": 1.125,
"learning_rate": 0.00043997714196397143,
"loss": 5.9242,
"mean_token_accuracy": 0.1620955288410187,
"num_tokens": 6170982.0,
"step": 3155
},
{
"entropy": 6.339143848419189,
"epoch": 3.141223272003978,
"grad_norm": 1.2578125,
"learning_rate": 0.00043971134181626173,
"loss": 5.9776,
"mean_token_accuracy": 0.15909559056162834,
"num_tokens": 6179356.0,
"step": 3160
},
{
"entropy": 6.3014768123626705,
"epoch": 3.1461959224266534,
"grad_norm": 1.0859375,
"learning_rate": 0.00043944504545800396,
"loss": 5.9115,
"mean_token_accuracy": 0.16268313229084014,
"num_tokens": 6188270.0,
"step": 3165
},
{
"entropy": 6.296312236785889,
"epoch": 3.1511685728493286,
"grad_norm": 1.09375,
"learning_rate": 0.0004391782536914447,
"loss": 5.8739,
"mean_token_accuracy": 0.16030744910240174,
"num_tokens": 6197624.0,
"step": 3170
},
{
"entropy": 6.347575855255127,
"epoch": 3.156141223272004,
"grad_norm": 1.2265625,
"learning_rate": 0.00043891096732032257,
"loss": 5.9272,
"mean_token_accuracy": 0.15927883833646775,
"num_tokens": 6207174.0,
"step": 3175
},
{
"entropy": 6.13545241355896,
"epoch": 3.161113873694679,
"grad_norm": 1.109375,
"learning_rate": 0.00043864318714986625,
"loss": 5.7614,
"mean_token_accuracy": 0.17330558001995086,
"num_tokens": 6217939.0,
"step": 3180
},
{
"entropy": 6.245046329498291,
"epoch": 3.1660865241173544,
"grad_norm": 1.046875,
"learning_rate": 0.00043837491398679234,
"loss": 5.9426,
"mean_token_accuracy": 0.15701133906841278,
"num_tokens": 6227857.0,
"step": 3185
},
{
"entropy": 6.2896781921386715,
"epoch": 3.17105917454003,
"grad_norm": 1.109375,
"learning_rate": 0.0004381061486393025,
"loss": 5.9872,
"mean_token_accuracy": 0.15053638741374015,
"num_tokens": 6236832.0,
"step": 3190
},
{
"entropy": 6.253927230834961,
"epoch": 3.1760318249627053,
"grad_norm": 1.28125,
"learning_rate": 0.0004378368919170809,
"loss": 5.9146,
"mean_token_accuracy": 0.15767164677381515,
"num_tokens": 6245879.0,
"step": 3195
},
{
"entropy": 6.241726875305176,
"epoch": 3.1810044753853806,
"grad_norm": 1.1171875,
"learning_rate": 0.00043756714463129244,
"loss": 5.9443,
"mean_token_accuracy": 0.15676286220550537,
"num_tokens": 6256350.0,
"step": 3200
},
{
"entropy": 6.198462724685669,
"epoch": 3.185977125808056,
"grad_norm": 1.1953125,
"learning_rate": 0.0004372969075945796,
"loss": 5.7284,
"mean_token_accuracy": 0.17455279529094697,
"num_tokens": 6266085.0,
"step": 3205
},
{
"entropy": 6.287790966033936,
"epoch": 3.190949776230731,
"grad_norm": 1.1875,
"learning_rate": 0.0004370261816210604,
"loss": 5.9766,
"mean_token_accuracy": 0.15546519458293914,
"num_tokens": 6276499.0,
"step": 3210
},
{
"entropy": 6.302801609039307,
"epoch": 3.1959224266534063,
"grad_norm": 1.0546875,
"learning_rate": 0.0004367549675263258,
"loss": 5.9758,
"mean_token_accuracy": 0.15671393871307374,
"num_tokens": 6286630.0,
"step": 3215
},
{
"entropy": 6.248741674423218,
"epoch": 3.2008950770760816,
"grad_norm": 1.1953125,
"learning_rate": 0.0004364832661274373,
"loss": 5.9279,
"mean_token_accuracy": 0.15730811208486556,
"num_tokens": 6295907.0,
"step": 3220
},
{
"entropy": 6.208601236343384,
"epoch": 3.205867727498757,
"grad_norm": 1.203125,
"learning_rate": 0.00043621107824292444,
"loss": 5.8547,
"mean_token_accuracy": 0.15954403430223466,
"num_tokens": 6304304.0,
"step": 3225
},
{
"entropy": 6.187569808959961,
"epoch": 3.210840377921432,
"grad_norm": 1.078125,
"learning_rate": 0.0004359384046927824,
"loss": 5.9667,
"mean_token_accuracy": 0.1540994629263878,
"num_tokens": 6315041.0,
"step": 3230
},
{
"entropy": 6.283197927474975,
"epoch": 3.2158130283441073,
"grad_norm": 1.140625,
"learning_rate": 0.0004356652462984693,
"loss": 5.8468,
"mean_token_accuracy": 0.16110004335641862,
"num_tokens": 6324551.0,
"step": 3235
},
{
"entropy": 6.227240324020386,
"epoch": 3.2207856787667826,
"grad_norm": 1.0703125,
"learning_rate": 0.00043539160388290416,
"loss": 5.9145,
"mean_token_accuracy": 0.1652422845363617,
"num_tokens": 6335512.0,
"step": 3240
},
{
"entropy": 6.237622880935669,
"epoch": 3.225758329189458,
"grad_norm": 1.09375,
"learning_rate": 0.0004351174782704641,
"loss": 5.9783,
"mean_token_accuracy": 0.15737072601914406,
"num_tokens": 6345987.0,
"step": 3245
},
{
"entropy": 6.113334465026855,
"epoch": 3.230730979612133,
"grad_norm": 1.1015625,
"learning_rate": 0.00043484287028698176,
"loss": 5.7628,
"mean_token_accuracy": 0.17646268904209136,
"num_tokens": 6357617.0,
"step": 3250
},
{
"entropy": 6.15558614730835,
"epoch": 3.2357036300348083,
"grad_norm": 1.2265625,
"learning_rate": 0.00043456778075974306,
"loss": 5.8489,
"mean_token_accuracy": 0.1625670999288559,
"num_tokens": 6366123.0,
"step": 3255
},
{
"entropy": 6.211847257614136,
"epoch": 3.240676280457484,
"grad_norm": 1.109375,
"learning_rate": 0.00043429221051748487,
"loss": 5.9587,
"mean_token_accuracy": 0.1544474706053734,
"num_tokens": 6375776.0,
"step": 3260
},
{
"entropy": 6.3595692157745365,
"epoch": 3.2456489308801593,
"grad_norm": 1.078125,
"learning_rate": 0.00043401616039039174,
"loss": 6.0042,
"mean_token_accuracy": 0.14871479719877242,
"num_tokens": 6386235.0,
"step": 3265
},
{
"entropy": 6.327882146835327,
"epoch": 3.2506215813028345,
"grad_norm": 1.0859375,
"learning_rate": 0.0004337396312100944,
"loss": 5.9654,
"mean_token_accuracy": 0.1534355416893959,
"num_tokens": 6396919.0,
"step": 3270
},
{
"entropy": 6.242443513870239,
"epoch": 3.2555942317255098,
"grad_norm": 1.171875,
"learning_rate": 0.0004334626238096665,
"loss": 5.92,
"mean_token_accuracy": 0.1631858080625534,
"num_tokens": 6406769.0,
"step": 3275
},
{
"entropy": 6.1702470779418945,
"epoch": 3.260566882148185,
"grad_norm": 1.1015625,
"learning_rate": 0.0004331851390236227,
"loss": 5.9474,
"mean_token_accuracy": 0.16255069971084596,
"num_tokens": 6416208.0,
"step": 3280
},
{
"entropy": 6.277315378189087,
"epoch": 3.2655395325708603,
"grad_norm": 1.1171875,
"learning_rate": 0.00043290717768791533,
"loss": 5.896,
"mean_token_accuracy": 0.1659083768725395,
"num_tokens": 6426125.0,
"step": 3285
},
{
"entropy": 6.2302977561950685,
"epoch": 3.2705121829935355,
"grad_norm": 1.1328125,
"learning_rate": 0.00043262874063993285,
"loss": 5.8787,
"mean_token_accuracy": 0.15293095484375954,
"num_tokens": 6436671.0,
"step": 3290
},
{
"entropy": 6.230917596817017,
"epoch": 3.2754848334162108,
"grad_norm": 1.1484375,
"learning_rate": 0.0004323498287184967,
"loss": 5.9823,
"mean_token_accuracy": 0.15813344120979309,
"num_tokens": 6447396.0,
"step": 3295
},
{
"entropy": 6.268582439422607,
"epoch": 3.280457483838886,
"grad_norm": 1.2109375,
"learning_rate": 0.00043207044276385884,
"loss": 5.9319,
"mean_token_accuracy": 0.15607068538665772,
"num_tokens": 6457045.0,
"step": 3300
},
{
"entropy": 6.140562963485718,
"epoch": 3.2854301342615613,
"grad_norm": 1.171875,
"learning_rate": 0.00043179058361769924,
"loss": 5.9539,
"mean_token_accuracy": 0.1630365252494812,
"num_tokens": 6466741.0,
"step": 3305
},
{
"entropy": 6.092462396621704,
"epoch": 3.2904027846842365,
"grad_norm": 1.5625,
"learning_rate": 0.0004315102521231237,
"loss": 5.6903,
"mean_token_accuracy": 0.17723899930715561,
"num_tokens": 6477769.0,
"step": 3310
},
{
"entropy": 6.383918190002442,
"epoch": 3.295375435106912,
"grad_norm": 1.140625,
"learning_rate": 0.0004312294491246609,
"loss": 6.0325,
"mean_token_accuracy": 0.15231009274721147,
"num_tokens": 6486991.0,
"step": 3315
},
{
"entropy": 6.201223230361938,
"epoch": 3.3003480855295875,
"grad_norm": 1.1328125,
"learning_rate": 0.0004309481754682597,
"loss": 5.9198,
"mean_token_accuracy": 0.16021429598331452,
"num_tokens": 6496772.0,
"step": 3320
},
{
"entropy": 6.219602870941162,
"epoch": 3.3053207359522627,
"grad_norm": 1.125,
"learning_rate": 0.00043066643200128724,
"loss": 5.9686,
"mean_token_accuracy": 0.15603076964616774,
"num_tokens": 6507577.0,
"step": 3325
},
{
"entropy": 6.3069655418396,
"epoch": 3.310293386374938,
"grad_norm": 1.03125,
"learning_rate": 0.0004303842195725259,
"loss": 5.9816,
"mean_token_accuracy": 0.15907459557056428,
"num_tokens": 6518040.0,
"step": 3330
},
{
"entropy": 6.228147172927857,
"epoch": 3.3152660367976132,
"grad_norm": 1.1484375,
"learning_rate": 0.00043010153903217074,
"loss": 5.9129,
"mean_token_accuracy": 0.15827998369932175,
"num_tokens": 6529261.0,
"step": 3335
},
{
"entropy": 6.192533159255982,
"epoch": 3.3202386872202885,
"grad_norm": 1.2421875,
"learning_rate": 0.0004298183912318273,
"loss": 5.836,
"mean_token_accuracy": 0.16489026099443435,
"num_tokens": 6537527.0,
"step": 3340
},
{
"entropy": 6.279222059249878,
"epoch": 3.3252113376429637,
"grad_norm": 1.1484375,
"learning_rate": 0.00042953477702450845,
"loss": 5.9592,
"mean_token_accuracy": 0.15893186032772064,
"num_tokens": 6547205.0,
"step": 3345
},
{
"entropy": 6.2540380477905275,
"epoch": 3.330183988065639,
"grad_norm": 1.265625,
"learning_rate": 0.0004292506972646324,
"loss": 6.0344,
"mean_token_accuracy": 0.1568805992603302,
"num_tokens": 6556328.0,
"step": 3350
},
{
"entropy": 6.263769674301147,
"epoch": 3.3351566384883142,
"grad_norm": 1.03125,
"learning_rate": 0.00042896615280802006,
"loss": 5.9791,
"mean_token_accuracy": 0.15798300355672837,
"num_tokens": 6566414.0,
"step": 3355
},
{
"entropy": 6.155578088760376,
"epoch": 3.3401292889109895,
"grad_norm": 1.0390625,
"learning_rate": 0.00042868114451189164,
"loss": 5.7989,
"mean_token_accuracy": 0.17771409898996354,
"num_tokens": 6577365.0,
"step": 3360
},
{
"entropy": 6.248938512802124,
"epoch": 3.3451019393336647,
"grad_norm": 1.1015625,
"learning_rate": 0.0004283956732348655,
"loss": 5.9143,
"mean_token_accuracy": 0.15745315849781036,
"num_tokens": 6586647.0,
"step": 3365
},
{
"entropy": 6.279117202758789,
"epoch": 3.35007458975634,
"grad_norm": 1.1796875,
"learning_rate": 0.00042810973983695426,
"loss": 6.0087,
"mean_token_accuracy": 0.1535283401608467,
"num_tokens": 6596955.0,
"step": 3370
},
{
"entropy": 6.301529741287231,
"epoch": 3.3550472401790152,
"grad_norm": 1.1953125,
"learning_rate": 0.0004278233451795629,
"loss": 5.8908,
"mean_token_accuracy": 0.16477383077144622,
"num_tokens": 6606161.0,
"step": 3375
},
{
"entropy": 6.083950757980347,
"epoch": 3.3600198906016905,
"grad_norm": 1.1484375,
"learning_rate": 0.000427536490125486,
"loss": 5.7778,
"mean_token_accuracy": 0.17313984781503677,
"num_tokens": 6616540.0,
"step": 3380
},
{
"entropy": 6.167257642745971,
"epoch": 3.364992541024366,
"grad_norm": 1.125,
"learning_rate": 0.00042724917553890507,
"loss": 5.9014,
"mean_token_accuracy": 0.16311111748218537,
"num_tokens": 6626109.0,
"step": 3385
},
{
"entropy": 6.247101974487305,
"epoch": 3.3699651914470414,
"grad_norm": 1.1875,
"learning_rate": 0.00042696140228538616,
"loss": 5.9134,
"mean_token_accuracy": 0.16419617831707,
"num_tokens": 6635576.0,
"step": 3390
},
{
"entropy": 6.22693657875061,
"epoch": 3.3749378418697167,
"grad_norm": 1.078125,
"learning_rate": 0.00042667317123187696,
"loss": 6.0019,
"mean_token_accuracy": 0.15808911621570587,
"num_tokens": 6645728.0,
"step": 3395
},
{
"entropy": 6.271794652938842,
"epoch": 3.379910492292392,
"grad_norm": 1.0859375,
"learning_rate": 0.00042638448324670424,
"loss": 5.8776,
"mean_token_accuracy": 0.16340596377849578,
"num_tokens": 6654538.0,
"step": 3400
},
{
"entropy": 6.158373880386352,
"epoch": 3.384883142715067,
"grad_norm": 1.1171875,
"learning_rate": 0.0004260953391995716,
"loss": 5.8444,
"mean_token_accuracy": 0.16797399520874023,
"num_tokens": 6664811.0,
"step": 3405
},
{
"entropy": 6.19985933303833,
"epoch": 3.3898557931377424,
"grad_norm": 1.046875,
"learning_rate": 0.00042580573996155646,
"loss": 5.9396,
"mean_token_accuracy": 0.16423666924238206,
"num_tokens": 6674185.0,
"step": 3410
},
{
"entropy": 6.239108467102051,
"epoch": 3.3948284435604177,
"grad_norm": 1.15625,
"learning_rate": 0.0004255156864051074,
"loss": 5.9207,
"mean_token_accuracy": 0.1551107332110405,
"num_tokens": 6684103.0,
"step": 3415
},
{
"entropy": 6.03340106010437,
"epoch": 3.399801093983093,
"grad_norm": 1.171875,
"learning_rate": 0.00042522517940404186,
"loss": 5.6243,
"mean_token_accuracy": 0.18921671211719512,
"num_tokens": 6693631.0,
"step": 3420
},
{
"entropy": 6.243697547912598,
"epoch": 3.404773744405768,
"grad_norm": 1.1015625,
"learning_rate": 0.0004249342198335433,
"loss": 5.962,
"mean_token_accuracy": 0.1601146638393402,
"num_tokens": 6704465.0,
"step": 3425
},
{
"entropy": 6.178900337219238,
"epoch": 3.4097463948284434,
"grad_norm": 1.21875,
"learning_rate": 0.00042464280857015856,
"loss": 5.9297,
"mean_token_accuracy": 0.16500024497509003,
"num_tokens": 6713592.0,
"step": 3430
},
{
"entropy": 6.232093381881714,
"epoch": 3.4147190452511187,
"grad_norm": 1.1015625,
"learning_rate": 0.00042435094649179525,
"loss": 5.9642,
"mean_token_accuracy": 0.15766821950674056,
"num_tokens": 6726252.0,
"step": 3435
},
{
"entropy": 6.284474229812622,
"epoch": 3.419691695673794,
"grad_norm": 1.0390625,
"learning_rate": 0.0004240586344777191,
"loss": 6.0436,
"mean_token_accuracy": 0.15264040231704712,
"num_tokens": 6736301.0,
"step": 3440
},
{
"entropy": 6.282935619354248,
"epoch": 3.4246643460964696,
"grad_norm": 1.1328125,
"learning_rate": 0.0004237658734085514,
"loss": 5.9903,
"mean_token_accuracy": 0.15537417978048323,
"num_tokens": 6745706.0,
"step": 3445
},
{
"entropy": 6.241247701644897,
"epoch": 3.429636996519145,
"grad_norm": 1.15625,
"learning_rate": 0.00042347266416626624,
"loss": 5.9844,
"mean_token_accuracy": 0.15316892564296722,
"num_tokens": 6755811.0,
"step": 3450
},
{
"entropy": 6.303303241729736,
"epoch": 3.43460964694182,
"grad_norm": 1.0859375,
"learning_rate": 0.0004231790076341877,
"loss": 5.9194,
"mean_token_accuracy": 0.16049495637416838,
"num_tokens": 6766365.0,
"step": 3455
},
{
"entropy": 6.1602260112762455,
"epoch": 3.4395822973644954,
"grad_norm": 1.140625,
"learning_rate": 0.0004228849046969876,
"loss": 5.8743,
"mean_token_accuracy": 0.16943498998880385,
"num_tokens": 6775806.0,
"step": 3460
},
{
"entropy": 6.2135313034057615,
"epoch": 3.4445549477871706,
"grad_norm": 1.203125,
"learning_rate": 0.0004225903562406826,
"loss": 5.9336,
"mean_token_accuracy": 0.15922852456569672,
"num_tokens": 6785996.0,
"step": 3465
},
{
"entropy": 6.222315454483033,
"epoch": 3.449527598209846,
"grad_norm": 0.9765625,
"learning_rate": 0.0004222953631526313,
"loss": 5.8644,
"mean_token_accuracy": 0.17208046913146974,
"num_tokens": 6795528.0,
"step": 3470
},
{
"entropy": 6.184564733505249,
"epoch": 3.454500248632521,
"grad_norm": 1.15625,
"learning_rate": 0.000421999926321532,
"loss": 5.8814,
"mean_token_accuracy": 0.16285580843687059,
"num_tokens": 6804592.0,
"step": 3475
},
{
"entropy": 6.1636364459991455,
"epoch": 3.4594728990551964,
"grad_norm": 1.09375,
"learning_rate": 0.0004217040466374199,
"loss": 5.9094,
"mean_token_accuracy": 0.16223583221435547,
"num_tokens": 6813242.0,
"step": 3480
},
{
"entropy": 6.266526794433593,
"epoch": 3.4644455494778716,
"grad_norm": 1.0390625,
"learning_rate": 0.00042140772499166403,
"loss": 5.9562,
"mean_token_accuracy": 0.15989834815263748,
"num_tokens": 6823223.0,
"step": 3485
},
{
"entropy": 6.076566362380982,
"epoch": 3.469418199900547,
"grad_norm": 1.1953125,
"learning_rate": 0.0004211109622769652,
"loss": 5.774,
"mean_token_accuracy": 0.16934829950332642,
"num_tokens": 6832425.0,
"step": 3490
},
{
"entropy": 6.163863039016723,
"epoch": 3.474390850323222,
"grad_norm": 1.140625,
"learning_rate": 0.00042081375938735283,
"loss": 5.9707,
"mean_token_accuracy": 0.15563918948173522,
"num_tokens": 6842549.0,
"step": 3495
},
{
"entropy": 6.238685083389282,
"epoch": 3.4793635007458974,
"grad_norm": 1.2734375,
"learning_rate": 0.00042051611721818233,
"loss": 6.0601,
"mean_token_accuracy": 0.15095847398042678,
"num_tokens": 6851652.0,
"step": 3500
},
{
"epoch": 3.4793635007458974,
"eval_entropy": 6.172711926121866,
"eval_loss": 6.67270040512085,
"eval_mean_token_accuracy": 0.13566318958620047,
"eval_num_tokens": 6851652.0,
"eval_runtime": 2.5275,
"eval_samples_per_second": 1370.934,
"eval_steps_per_second": 171.713,
"step": 3500
},
{
"entropy": 6.201852750778198,
"epoch": 3.4843361511685726,
"grad_norm": 1.15625,
"learning_rate": 0.00042021803666613273,
"loss": 5.8964,
"mean_token_accuracy": 0.16534167975187303,
"num_tokens": 6860716.0,
"step": 3505
},
{
"entropy": 6.16337308883667,
"epoch": 3.489308801591248,
"grad_norm": 1.0859375,
"learning_rate": 0.00041991951862920365,
"loss": 5.9095,
"mean_token_accuracy": 0.15994026511907578,
"num_tokens": 6870074.0,
"step": 3510
},
{
"entropy": 6.331715488433838,
"epoch": 3.4942814520139236,
"grad_norm": 1.109375,
"learning_rate": 0.0004196205640067124,
"loss": 6.1444,
"mean_token_accuracy": 0.14215195477008818,
"num_tokens": 6880569.0,
"step": 3515
},
{
"entropy": 6.314612197875976,
"epoch": 3.499254102436599,
"grad_norm": 1.1640625,
"learning_rate": 0.000419321173699292,
"loss": 5.9999,
"mean_token_accuracy": 0.1541007474064827,
"num_tokens": 6890522.0,
"step": 3520
},
{
"entropy": 6.205518817901611,
"epoch": 3.504226752859274,
"grad_norm": 1.015625,
"learning_rate": 0.0004190213486088877,
"loss": 5.937,
"mean_token_accuracy": 0.1579175189137459,
"num_tokens": 6900388.0,
"step": 3525
},
{
"entropy": 6.2201203346252445,
"epoch": 3.5091994032819493,
"grad_norm": 1.140625,
"learning_rate": 0.0004187210896387546,
"loss": 5.9266,
"mean_token_accuracy": 0.16327783614397048,
"num_tokens": 6910467.0,
"step": 3530
},
{
"entropy": 6.187130546569824,
"epoch": 3.5141720537046246,
"grad_norm": 1.1484375,
"learning_rate": 0.0004184203976934552,
"loss": 5.958,
"mean_token_accuracy": 0.16534241512417794,
"num_tokens": 6919370.0,
"step": 3535
},
{
"entropy": 6.0957060813903805,
"epoch": 3.5191447041273,
"grad_norm": 1.203125,
"learning_rate": 0.00041811927367885605,
"loss": 5.7468,
"mean_token_accuracy": 0.17413049638271333,
"num_tokens": 6929688.0,
"step": 3540
},
{
"entropy": 6.214923286437989,
"epoch": 3.524117354549975,
"grad_norm": 1.1875,
"learning_rate": 0.0004178177185021254,
"loss": 5.993,
"mean_token_accuracy": 0.1586630493402481,
"num_tokens": 6939861.0,
"step": 3545
},
{
"entropy": 6.277981901168824,
"epoch": 3.5290900049726504,
"grad_norm": 1.109375,
"learning_rate": 0.0004175157330717305,
"loss": 6.012,
"mean_token_accuracy": 0.152120178937912,
"num_tokens": 6949746.0,
"step": 3550
},
{
"entropy": 6.279894161224365,
"epoch": 3.5340626553953256,
"grad_norm": 1.0859375,
"learning_rate": 0.000417213318297435,
"loss": 5.9477,
"mean_token_accuracy": 0.15829975455999373,
"num_tokens": 6961408.0,
"step": 3555
},
{
"entropy": 6.155437850952149,
"epoch": 3.539035305818001,
"grad_norm": 1.140625,
"learning_rate": 0.00041691047509029555,
"loss": 5.9903,
"mean_token_accuracy": 0.16253448277711868,
"num_tokens": 6972003.0,
"step": 3560
},
{
"entropy": 6.238606929779053,
"epoch": 3.5440079562406765,
"grad_norm": 1.1484375,
"learning_rate": 0.00041660720436265987,
"loss": 5.8919,
"mean_token_accuracy": 0.16272868663072587,
"num_tokens": 6983528.0,
"step": 3565
},
{
"entropy": 6.24536771774292,
"epoch": 3.548980606663352,
"grad_norm": 1.140625,
"learning_rate": 0.0004163035070281634,
"loss": 5.9737,
"mean_token_accuracy": 0.16169923469424247,
"num_tokens": 6992766.0,
"step": 3570
},
{
"entropy": 6.18948974609375,
"epoch": 3.553953257086027,
"grad_norm": 1.2109375,
"learning_rate": 0.00041599938400172686,
"loss": 5.8557,
"mean_token_accuracy": 0.1610173135995865,
"num_tokens": 7001797.0,
"step": 3575
},
{
"entropy": 6.242516040802002,
"epoch": 3.5589259075087023,
"grad_norm": 1.140625,
"learning_rate": 0.00041569483619955354,
"loss": 5.8976,
"mean_token_accuracy": 0.16617249846458435,
"num_tokens": 7011215.0,
"step": 3580
},
{
"entropy": 6.202178335189819,
"epoch": 3.5638985579313776,
"grad_norm": 1.0390625,
"learning_rate": 0.00041538986453912616,
"loss": 5.9411,
"mean_token_accuracy": 0.15492074564099312,
"num_tokens": 7021513.0,
"step": 3585
},
{
"entropy": 6.168223142623901,
"epoch": 3.568871208354053,
"grad_norm": 1.0859375,
"learning_rate": 0.0004150844699392045,
"loss": 5.9705,
"mean_token_accuracy": 0.15568189769983293,
"num_tokens": 7031654.0,
"step": 3590
},
{
"entropy": 6.140969467163086,
"epoch": 3.573843858776728,
"grad_norm": 1.2265625,
"learning_rate": 0.00041477865331982264,
"loss": 5.9243,
"mean_token_accuracy": 0.1648896038532257,
"num_tokens": 7040614.0,
"step": 3595
},
{
"entropy": 6.2059790134429935,
"epoch": 3.5788165091994033,
"grad_norm": 1.09375,
"learning_rate": 0.0004144724156022857,
"loss": 5.9264,
"mean_token_accuracy": 0.16396153271198272,
"num_tokens": 7050877.0,
"step": 3600
},
{
"entropy": 6.350143241882324,
"epoch": 3.5837891596220786,
"grad_norm": 1.125,
"learning_rate": 0.00041416575770916763,
"loss": 6.1312,
"mean_token_accuracy": 0.14578675478696823,
"num_tokens": 7061330.0,
"step": 3605
},
{
"entropy": 6.270376348495484,
"epoch": 3.588761810044754,
"grad_norm": 1.0078125,
"learning_rate": 0.00041385868056430827,
"loss": 6.0061,
"mean_token_accuracy": 0.15612463057041168,
"num_tokens": 7071290.0,
"step": 3610
},
{
"entropy": 6.197152137756348,
"epoch": 3.593734460467429,
"grad_norm": 1.2421875,
"learning_rate": 0.0004135511850928102,
"loss": 5.9823,
"mean_token_accuracy": 0.15387657284736633,
"num_tokens": 7080112.0,
"step": 3615
},
{
"entropy": 6.247121429443359,
"epoch": 3.5987071108901043,
"grad_norm": 1.0859375,
"learning_rate": 0.00041324327222103655,
"loss": 6.1274,
"mean_token_accuracy": 0.14585825949907302,
"num_tokens": 7091467.0,
"step": 3620
},
{
"entropy": 6.221949338912964,
"epoch": 3.6036797613127796,
"grad_norm": 1.3359375,
"learning_rate": 0.00041293494287660785,
"loss": 5.7527,
"mean_token_accuracy": 0.17463432848453522,
"num_tokens": 7101744.0,
"step": 3625
},
{
"entropy": 6.246480560302734,
"epoch": 3.608652411735455,
"grad_norm": 1.0859375,
"learning_rate": 0.00041262619798839914,
"loss": 5.9563,
"mean_token_accuracy": 0.16098868250846862,
"num_tokens": 7112854.0,
"step": 3630
},
{
"entropy": 6.0680344581604,
"epoch": 3.61362506215813,
"grad_norm": 1.2109375,
"learning_rate": 0.00041231703848653746,
"loss": 5.9483,
"mean_token_accuracy": 0.16464419662952423,
"num_tokens": 7121613.0,
"step": 3635
},
{
"entropy": 6.24086217880249,
"epoch": 3.6185977125808053,
"grad_norm": 1.078125,
"learning_rate": 0.00041200746530239903,
"loss": 5.9938,
"mean_token_accuracy": 0.15724552422761917,
"num_tokens": 7131357.0,
"step": 3640
},
{
"entropy": 6.184099435806274,
"epoch": 3.623570363003481,
"grad_norm": 1.109375,
"learning_rate": 0.0004116974793686061,
"loss": 5.8765,
"mean_token_accuracy": 0.17357155233621596,
"num_tokens": 7140200.0,
"step": 3645
},
{
"entropy": 6.190071725845337,
"epoch": 3.6285430134261563,
"grad_norm": 1.1015625,
"learning_rate": 0.0004113870816190244,
"loss": 5.9801,
"mean_token_accuracy": 0.16244304478168486,
"num_tokens": 7149367.0,
"step": 3650
},
{
"entropy": 6.152639007568359,
"epoch": 3.6335156638488315,
"grad_norm": 1.0859375,
"learning_rate": 0.0004110762729887605,
"loss": 5.9362,
"mean_token_accuracy": 0.16452273428440095,
"num_tokens": 7158088.0,
"step": 3655
},
{
"entropy": 6.190124607086181,
"epoch": 3.6384883142715068,
"grad_norm": 1.03125,
"learning_rate": 0.00041076505441415873,
"loss": 5.9339,
"mean_token_accuracy": 0.16660187393426895,
"num_tokens": 7168141.0,
"step": 3660
},
{
"entropy": 6.254987573623657,
"epoch": 3.643460964694182,
"grad_norm": 1.1875,
"learning_rate": 0.0004104534268327982,
"loss": 5.8644,
"mean_token_accuracy": 0.1626141905784607,
"num_tokens": 7177646.0,
"step": 3665
},
{
"entropy": 6.238340234756469,
"epoch": 3.6484336151168573,
"grad_norm": 1.171875,
"learning_rate": 0.0004101413911834904,
"loss": 6.017,
"mean_token_accuracy": 0.1559474915266037,
"num_tokens": 7186445.0,
"step": 3670
},
{
"entropy": 6.2091888904571535,
"epoch": 3.6534062655395325,
"grad_norm": 1.1328125,
"learning_rate": 0.00040982894840627614,
"loss": 5.9566,
"mean_token_accuracy": 0.16231908202171325,
"num_tokens": 7195812.0,
"step": 3675
},
{
"entropy": 6.149083423614502,
"epoch": 3.6583789159622078,
"grad_norm": 1.0625,
"learning_rate": 0.0004095160994424228,
"loss": 5.8106,
"mean_token_accuracy": 0.1724393531680107,
"num_tokens": 7204756.0,
"step": 3680
},
{
"entropy": 6.216005420684814,
"epoch": 3.663351566384883,
"grad_norm": 1.140625,
"learning_rate": 0.00040920284523442125,
"loss": 5.9843,
"mean_token_accuracy": 0.16606335788965226,
"num_tokens": 7214708.0,
"step": 3685
},
{
"entropy": 6.237343263626099,
"epoch": 3.6683242168075583,
"grad_norm": 1.046875,
"learning_rate": 0.00040888918672598337,
"loss": 6.0235,
"mean_token_accuracy": 0.15607056766748428,
"num_tokens": 7224623.0,
"step": 3690
},
{
"entropy": 6.20723705291748,
"epoch": 3.673296867230234,
"grad_norm": 0.97265625,
"learning_rate": 0.000408575124862039,
"loss": 5.925,
"mean_token_accuracy": 0.16391729712486267,
"num_tokens": 7235228.0,
"step": 3695
},
{
"entropy": 6.186968851089477,
"epoch": 3.678269517652909,
"grad_norm": 1.1875,
"learning_rate": 0.0004082606605887331,
"loss": 6.0144,
"mean_token_accuracy": 0.16542051434516908,
"num_tokens": 7245318.0,
"step": 3700
},
{
"entropy": 5.970559120178223,
"epoch": 3.6832421680755845,
"grad_norm": 1.21875,
"learning_rate": 0.00040794579485342304,
"loss": 5.5855,
"mean_token_accuracy": 0.19518422335386276,
"num_tokens": 7255468.0,
"step": 3705
},
{
"entropy": 6.256650352478028,
"epoch": 3.6882148184982597,
"grad_norm": 1.09375,
"learning_rate": 0.00040763052860467547,
"loss": 5.9628,
"mean_token_accuracy": 0.15635033398866655,
"num_tokens": 7265006.0,
"step": 3710
},
{
"entropy": 6.160571527481079,
"epoch": 3.693187468920935,
"grad_norm": 1.2109375,
"learning_rate": 0.0004073148627922637,
"loss": 5.935,
"mean_token_accuracy": 0.1624176725745201,
"num_tokens": 7273835.0,
"step": 3715
},
{
"entropy": 6.266654777526855,
"epoch": 3.69816011934361,
"grad_norm": 1.2109375,
"learning_rate": 0.0004069987983671648,
"loss": 6.0389,
"mean_token_accuracy": 0.1518197163939476,
"num_tokens": 7284617.0,
"step": 3720
},
{
"entropy": 6.080435180664063,
"epoch": 3.7031327697662855,
"grad_norm": 1.1484375,
"learning_rate": 0.00040668233628155664,
"loss": 5.7677,
"mean_token_accuracy": 0.17647801116108894,
"num_tokens": 7293707.0,
"step": 3725
},
{
"entropy": 6.213000202178955,
"epoch": 3.7081054201889607,
"grad_norm": 1.171875,
"learning_rate": 0.0004063654774888153,
"loss": 5.9872,
"mean_token_accuracy": 0.14929335862398146,
"num_tokens": 7303227.0,
"step": 3730
},
{
"entropy": 6.149737310409546,
"epoch": 3.713078070611636,
"grad_norm": 1.0390625,
"learning_rate": 0.0004060482229435116,
"loss": 5.8579,
"mean_token_accuracy": 0.17009468525648117,
"num_tokens": 7312782.0,
"step": 3735
},
{
"entropy": 6.218086004257202,
"epoch": 3.7180507210343112,
"grad_norm": 1.1796875,
"learning_rate": 0.0004057305736014089,
"loss": 5.9983,
"mean_token_accuracy": 0.15850173234939574,
"num_tokens": 7322249.0,
"step": 3740
},
{
"entropy": 6.258641719818115,
"epoch": 3.7230233714569865,
"grad_norm": 1.15625,
"learning_rate": 0.0004054125304194598,
"loss": 5.9575,
"mean_token_accuracy": 0.16373585313558578,
"num_tokens": 7331076.0,
"step": 3745
},
{
"entropy": 6.139318656921387,
"epoch": 3.7279960218796617,
"grad_norm": 1.109375,
"learning_rate": 0.00040509409435580355,
"loss": 5.9144,
"mean_token_accuracy": 0.16399238407611846,
"num_tokens": 7340039.0,
"step": 3750
},
{
"entropy": 6.157795095443726,
"epoch": 3.732968672302337,
"grad_norm": 1.0703125,
"learning_rate": 0.0004047752663697627,
"loss": 5.9026,
"mean_token_accuracy": 0.16509526073932648,
"num_tokens": 7349300.0,
"step": 3755
},
{
"entropy": 6.329801225662232,
"epoch": 3.7379413227250122,
"grad_norm": 1.125,
"learning_rate": 0.0004044560474218409,
"loss": 6.0081,
"mean_token_accuracy": 0.16038140654563904,
"num_tokens": 7358363.0,
"step": 3760
},
{
"entropy": 6.180526304244995,
"epoch": 3.7429139731476875,
"grad_norm": 1.1328125,
"learning_rate": 0.00040413643847371923,
"loss": 5.9073,
"mean_token_accuracy": 0.1686038687825203,
"num_tokens": 7369391.0,
"step": 3765
},
{
"entropy": 6.182810735702515,
"epoch": 3.7478866235703627,
"grad_norm": 1.234375,
"learning_rate": 0.0004038164404882539,
"loss": 5.9715,
"mean_token_accuracy": 0.1539320930838585,
"num_tokens": 7378272.0,
"step": 3770
},
{
"entropy": 6.178169631958008,
"epoch": 3.7528592739930384,
"grad_norm": 1.1171875,
"learning_rate": 0.0004034960544294731,
"loss": 5.8548,
"mean_token_accuracy": 0.16293878555297853,
"num_tokens": 7387590.0,
"step": 3775
},
{
"entropy": 6.199071216583252,
"epoch": 3.7578319244157137,
"grad_norm": 1.15625,
"learning_rate": 0.00040317528126257413,
"loss": 5.9395,
"mean_token_accuracy": 0.15719935297966003,
"num_tokens": 7397733.0,
"step": 3780
},
{
"entropy": 6.15965461730957,
"epoch": 3.762804574838389,
"grad_norm": 1.1328125,
"learning_rate": 0.0004028541219539205,
"loss": 5.8109,
"mean_token_accuracy": 0.16741475462913513,
"num_tokens": 7406748.0,
"step": 3785
},
{
"entropy": 6.16231632232666,
"epoch": 3.767777225261064,
"grad_norm": 1.140625,
"learning_rate": 0.0004025325774710391,
"loss": 6.0358,
"mean_token_accuracy": 0.15851458609104158,
"num_tokens": 7416419.0,
"step": 3790
},
{
"entropy": 6.185123634338379,
"epoch": 3.7727498756837394,
"grad_norm": 1.265625,
"learning_rate": 0.00040221064878261684,
"loss": 5.9069,
"mean_token_accuracy": 0.16041389405727385,
"num_tokens": 7425257.0,
"step": 3795
},
{
"entropy": 6.274077224731445,
"epoch": 3.7777225261064147,
"grad_norm": 1.1328125,
"learning_rate": 0.0004018883368584985,
"loss": 6.0027,
"mean_token_accuracy": 0.15457284897565843,
"num_tokens": 7436914.0,
"step": 3800
},
{
"entropy": 6.19541301727295,
"epoch": 3.78269517652909,
"grad_norm": 1.1171875,
"learning_rate": 0.00040156564266968336,
"loss": 6.008,
"mean_token_accuracy": 0.16219560652971268,
"num_tokens": 7447568.0,
"step": 3805
},
{
"entropy": 6.157040357589722,
"epoch": 3.787667826951765,
"grad_norm": 1.109375,
"learning_rate": 0.00040124256718832214,
"loss": 5.9362,
"mean_token_accuracy": 0.16297836005687713,
"num_tokens": 7457717.0,
"step": 3810
},
{
"entropy": 6.241963338851929,
"epoch": 3.7926404773744404,
"grad_norm": 1.171875,
"learning_rate": 0.0004009191113877141,
"loss": 5.9447,
"mean_token_accuracy": 0.1600917473435402,
"num_tokens": 7467123.0,
"step": 3815
},
{
"entropy": 6.153872013092041,
"epoch": 3.797613127797116,
"grad_norm": 1.0859375,
"learning_rate": 0.0004005952762423048,
"loss": 5.9096,
"mean_token_accuracy": 0.165580677986145,
"num_tokens": 7476987.0,
"step": 3820
},
{
"entropy": 6.235847520828247,
"epoch": 3.8025857782197914,
"grad_norm": 1.15625,
"learning_rate": 0.0004002710627276821,
"loss": 6.0076,
"mean_token_accuracy": 0.16175391674041747,
"num_tokens": 7487024.0,
"step": 3825
},
{
"entropy": 6.230613899230957,
"epoch": 3.8075584286424666,
"grad_norm": 1.0390625,
"learning_rate": 0.0003999464718205739,
"loss": 6.0186,
"mean_token_accuracy": 0.16084508448839188,
"num_tokens": 7496826.0,
"step": 3830
},
{
"entropy": 6.1878491878509525,
"epoch": 3.812531079065142,
"grad_norm": 1.125,
"learning_rate": 0.0003996215044988451,
"loss": 5.9516,
"mean_token_accuracy": 0.15903948694467546,
"num_tokens": 7506981.0,
"step": 3835
},
{
"entropy": 6.100557088851929,
"epoch": 3.817503729487817,
"grad_norm": 1.59375,
"learning_rate": 0.00039929616174149454,
"loss": 5.7443,
"mean_token_accuracy": 0.17773568779230117,
"num_tokens": 7516925.0,
"step": 3840
},
{
"entropy": 6.201454401016235,
"epoch": 3.8224763799104924,
"grad_norm": 1.1796875,
"learning_rate": 0.00039897044452865207,
"loss": 5.9044,
"mean_token_accuracy": 0.16549646556377412,
"num_tokens": 7525789.0,
"step": 3845
},
{
"entropy": 6.124452829360962,
"epoch": 3.8274490303331676,
"grad_norm": 1.1015625,
"learning_rate": 0.0003986443538415756,
"loss": 5.8609,
"mean_token_accuracy": 0.16523757129907607,
"num_tokens": 7535907.0,
"step": 3850
},
{
"entropy": 6.203757095336914,
"epoch": 3.832421680755843,
"grad_norm": 1.171875,
"learning_rate": 0.0003983178906626484,
"loss": 5.8838,
"mean_token_accuracy": 0.16845598071813583,
"num_tokens": 7546985.0,
"step": 3855
},
{
"entropy": 6.063313245773315,
"epoch": 3.837394331178518,
"grad_norm": 1.09375,
"learning_rate": 0.00039799105597537564,
"loss": 5.7248,
"mean_token_accuracy": 0.18246889263391494,
"num_tokens": 7556912.0,
"step": 3860
},
{
"entropy": 6.163967084884644,
"epoch": 3.8423669816011934,
"grad_norm": 1.1328125,
"learning_rate": 0.0003976638507643818,
"loss": 5.9558,
"mean_token_accuracy": 0.1610849991440773,
"num_tokens": 7567156.0,
"step": 3865
},
{
"entropy": 6.187885379791259,
"epoch": 3.8473396320238686,
"grad_norm": 1.1328125,
"learning_rate": 0.0003973362760154076,
"loss": 5.9082,
"mean_token_accuracy": 0.17026485204696656,
"num_tokens": 7576423.0,
"step": 3870
},
{
"entropy": 6.229396629333496,
"epoch": 3.852312282446544,
"grad_norm": 0.984375,
"learning_rate": 0.0003970083327153071,
"loss": 5.9005,
"mean_token_accuracy": 0.16606468260288237,
"num_tokens": 7585612.0,
"step": 3875
},
{
"entropy": 6.051190519332886,
"epoch": 3.857284932869219,
"grad_norm": 1.3046875,
"learning_rate": 0.0003966800218520445,
"loss": 5.925,
"mean_token_accuracy": 0.16799391359090804,
"num_tokens": 7595581.0,
"step": 3880
},
{
"entropy": 6.238703393936158,
"epoch": 3.8622575832918944,
"grad_norm": 1.15625,
"learning_rate": 0.0003963513444146914,
"loss": 5.9248,
"mean_token_accuracy": 0.16054464280605316,
"num_tokens": 7605202.0,
"step": 3885
},
{
"entropy": 6.2128133296966555,
"epoch": 3.8672302337145696,
"grad_norm": 1.0390625,
"learning_rate": 0.0003960223013934239,
"loss": 5.8363,
"mean_token_accuracy": 0.1716415360569954,
"num_tokens": 7615783.0,
"step": 3890
},
{
"entropy": 6.20598292350769,
"epoch": 3.872202884137245,
"grad_norm": 1.1015625,
"learning_rate": 0.0003956928937795192,
"loss": 6.0374,
"mean_token_accuracy": 0.1488328218460083,
"num_tokens": 7626609.0,
"step": 3895
},
{
"entropy": 6.1895088195800785,
"epoch": 3.8771755345599206,
"grad_norm": 1.234375,
"learning_rate": 0.000395363122565353,
"loss": 5.9054,
"mean_token_accuracy": 0.16600026860833167,
"num_tokens": 7636676.0,
"step": 3900
},
{
"entropy": 6.227712249755859,
"epoch": 3.882148184982596,
"grad_norm": 1.125,
"learning_rate": 0.00039503298874439655,
"loss": 5.9127,
"mean_token_accuracy": 0.1655281513929367,
"num_tokens": 7646523.0,
"step": 3905
},
{
"entropy": 6.089673137664795,
"epoch": 3.887120835405271,
"grad_norm": 1.1484375,
"learning_rate": 0.0003947024933112131,
"loss": 5.8043,
"mean_token_accuracy": 0.1747158259153366,
"num_tokens": 7656267.0,
"step": 3910
},
{
"entropy": 6.189454221725464,
"epoch": 3.8920934858279463,
"grad_norm": 1.234375,
"learning_rate": 0.0003943716372614556,
"loss": 5.8879,
"mean_token_accuracy": 0.16677381247282028,
"num_tokens": 7665529.0,
"step": 3915
},
{
"entropy": 6.224945831298828,
"epoch": 3.8970661362506216,
"grad_norm": 1.1015625,
"learning_rate": 0.0003940404215918634,
"loss": 5.9675,
"mean_token_accuracy": 0.16120045632123947,
"num_tokens": 7675477.0,
"step": 3920
},
{
"entropy": 5.92595272064209,
"epoch": 3.902038786673297,
"grad_norm": 1.46875,
"learning_rate": 0.00039370884730025897,
"loss": 5.5843,
"mean_token_accuracy": 0.19233729541301728,
"num_tokens": 7685675.0,
"step": 3925
},
{
"entropy": 6.115845489501953,
"epoch": 3.907011437095972,
"grad_norm": 1.0625,
"learning_rate": 0.0003933769153855455,
"loss": 5.9479,
"mean_token_accuracy": 0.1654975861310959,
"num_tokens": 7695821.0,
"step": 3930
},
{
"entropy": 6.182166719436646,
"epoch": 3.9119840875186473,
"grad_norm": 1.203125,
"learning_rate": 0.0003930446268477034,
"loss": 5.9437,
"mean_token_accuracy": 0.16409701406955718,
"num_tokens": 7706163.0,
"step": 3935
},
{
"entropy": 6.1722980499267575,
"epoch": 3.9169567379413226,
"grad_norm": 1.1328125,
"learning_rate": 0.0003927119826877875,
"loss": 5.8664,
"mean_token_accuracy": 0.16642989367246627,
"num_tokens": 7715125.0,
"step": 3940
},
{
"entropy": 6.141762924194336,
"epoch": 3.921929388363998,
"grad_norm": 1.1328125,
"learning_rate": 0.0003923789839079238,
"loss": 5.8855,
"mean_token_accuracy": 0.16860341429710388,
"num_tokens": 7724546.0,
"step": 3945
},
{
"entropy": 5.987500953674316,
"epoch": 3.9269020387866735,
"grad_norm": 1.4375,
"learning_rate": 0.00039204563151130683,
"loss": 5.7059,
"mean_token_accuracy": 0.17774517983198165,
"num_tokens": 7734829.0,
"step": 3950
},
{
"entropy": 6.170799255371094,
"epoch": 3.931874689209349,
"grad_norm": 1.0390625,
"learning_rate": 0.0003917119265021965,
"loss": 5.8843,
"mean_token_accuracy": 0.1613735541701317,
"num_tokens": 7744612.0,
"step": 3955
},
{
"entropy": 6.208910751342773,
"epoch": 3.936847339632024,
"grad_norm": 1.1953125,
"learning_rate": 0.0003913778698859146,
"loss": 5.9705,
"mean_token_accuracy": 0.1621137499809265,
"num_tokens": 7754042.0,
"step": 3960
},
{
"entropy": 6.196197128295898,
"epoch": 3.9418199900546993,
"grad_norm": 1.2109375,
"learning_rate": 0.0003910434626688427,
"loss": 5.8676,
"mean_token_accuracy": 0.16624239832162857,
"num_tokens": 7762608.0,
"step": 3965
},
{
"entropy": 6.097027254104614,
"epoch": 3.9467926404773745,
"grad_norm": 1.2109375,
"learning_rate": 0.0003907087058584183,
"loss": 5.8146,
"mean_token_accuracy": 0.1681312456727028,
"num_tokens": 7771477.0,
"step": 3970
},
{
"entropy": 6.09872784614563,
"epoch": 3.95176529090005,
"grad_norm": 1.3203125,
"learning_rate": 0.00039037360046313186,
"loss": 5.8922,
"mean_token_accuracy": 0.17521725371479988,
"num_tokens": 7782785.0,
"step": 3975
},
{
"entropy": 6.3034168720245365,
"epoch": 3.956737941322725,
"grad_norm": 1.03125,
"learning_rate": 0.00039003814749252454,
"loss": 6.0725,
"mean_token_accuracy": 0.15178293958306313,
"num_tokens": 7793572.0,
"step": 3980
},
{
"entropy": 6.195157480239868,
"epoch": 3.9617105917454003,
"grad_norm": 1.1015625,
"learning_rate": 0.00038970234795718424,
"loss": 5.856,
"mean_token_accuracy": 0.16383777558803558,
"num_tokens": 7803170.0,
"step": 3985
},
{
"entropy": 6.1456903457641605,
"epoch": 3.9666832421680756,
"grad_norm": 0.99609375,
"learning_rate": 0.000389366202868743,
"loss": 6.0241,
"mean_token_accuracy": 0.1551950082182884,
"num_tokens": 7813445.0,
"step": 3990
},
{
"entropy": 6.195073509216309,
"epoch": 3.971655892590751,
"grad_norm": 1.1640625,
"learning_rate": 0.0003890297132398739,
"loss": 6.0116,
"mean_token_accuracy": 0.16142124235630034,
"num_tokens": 7823089.0,
"step": 3995
},
{
"entropy": 6.200736808776855,
"epoch": 3.976628543013426,
"grad_norm": 1.21875,
"learning_rate": 0.0003886928800842879,
"loss": 5.8244,
"mean_token_accuracy": 0.1657528504729271,
"num_tokens": 7832487.0,
"step": 4000
},
{
"epoch": 3.976628543013426,
"eval_entropy": 6.082162658190398,
"eval_loss": 6.595086574554443,
"eval_mean_token_accuracy": 0.13894746185619436,
"eval_num_tokens": 7832487.0,
"eval_runtime": 2.5484,
"eval_samples_per_second": 1359.685,
"eval_steps_per_second": 170.304,
"step": 4000
},
{
"entropy": 6.113853406906128,
"epoch": 3.9816011934361013,
"grad_norm": 1.1171875,
"learning_rate": 0.00038835570441673095,
"loss": 5.9105,
"mean_token_accuracy": 0.1635222017765045,
"num_tokens": 7842540.0,
"step": 4005
},
{
"entropy": 6.2019185543060305,
"epoch": 3.9865738438587766,
"grad_norm": 1.1328125,
"learning_rate": 0.0003880181872529809,
"loss": 5.9621,
"mean_token_accuracy": 0.1631697878241539,
"num_tokens": 7852384.0,
"step": 4010
},
{
"entropy": 6.171451997756958,
"epoch": 3.991546494281452,
"grad_norm": 1.046875,
"learning_rate": 0.0003876803296098443,
"loss": 5.9513,
"mean_token_accuracy": 0.16824467927217485,
"num_tokens": 7864082.0,
"step": 4015
},
{
"entropy": 6.178514528274536,
"epoch": 3.996519144704127,
"grad_norm": 1.0546875,
"learning_rate": 0.0003873421325051535,
"loss": 5.8956,
"mean_token_accuracy": 0.16187321320176123,
"num_tokens": 7873677.0,
"step": 4020
},
{
"entropy": 6.214059988657634,
"epoch": 4.000994530084535,
"grad_norm": 1.09375,
"learning_rate": 0.00038700359695776346,
"loss": 5.8598,
"mean_token_accuracy": 0.160642276207606,
"num_tokens": 7882368.0,
"step": 4025
},
{
"entropy": 6.179887580871582,
"epoch": 4.00596718050721,
"grad_norm": 1.09375,
"learning_rate": 0.0003866647239875487,
"loss": 5.532,
"mean_token_accuracy": 0.1760445863008499,
"num_tokens": 7892327.0,
"step": 4030
},
{
"entropy": 6.11782283782959,
"epoch": 4.010939830929885,
"grad_norm": 1.0625,
"learning_rate": 0.00038632551461540023,
"loss": 5.5491,
"mean_token_accuracy": 0.18085032552480698,
"num_tokens": 7902048.0,
"step": 4035
},
{
"entropy": 6.083871650695801,
"epoch": 4.015912481352561,
"grad_norm": 1.0625,
"learning_rate": 0.0003859859698632228,
"loss": 5.5917,
"mean_token_accuracy": 0.17614275962114334,
"num_tokens": 7911512.0,
"step": 4040
},
{
"entropy": 6.145664548873901,
"epoch": 4.020885131775236,
"grad_norm": 1.09375,
"learning_rate": 0.0003856460907539312,
"loss": 5.6743,
"mean_token_accuracy": 0.16649484783411025,
"num_tokens": 7921832.0,
"step": 4045
},
{
"entropy": 6.190884065628052,
"epoch": 4.025857782197911,
"grad_norm": 1.125,
"learning_rate": 0.0003853058783114478,
"loss": 5.5828,
"mean_token_accuracy": 0.17876727730035782,
"num_tokens": 7931143.0,
"step": 4050
},
{
"entropy": 6.009689617156982,
"epoch": 4.030830432620586,
"grad_norm": 1.046875,
"learning_rate": 0.0003849653335606987,
"loss": 5.4871,
"mean_token_accuracy": 0.18637479096651077,
"num_tokens": 7941108.0,
"step": 4055
},
{
"entropy": 6.043899965286255,
"epoch": 4.035803083043262,
"grad_norm": 1.1015625,
"learning_rate": 0.0003846244575276117,
"loss": 5.5587,
"mean_token_accuracy": 0.1737949013710022,
"num_tokens": 7951464.0,
"step": 4060
},
{
"entropy": 6.200726842880249,
"epoch": 4.040775733465938,
"grad_norm": 1.4140625,
"learning_rate": 0.00038428325123911205,
"loss": 5.6606,
"mean_token_accuracy": 0.1685145765542984,
"num_tokens": 7961626.0,
"step": 4065
},
{
"entropy": 6.159210205078125,
"epoch": 4.045748383888613,
"grad_norm": 1.1796875,
"learning_rate": 0.0003839417157231205,
"loss": 5.6431,
"mean_token_accuracy": 0.17365633100271224,
"num_tokens": 7970135.0,
"step": 4070
},
{
"entropy": 6.033663606643676,
"epoch": 4.050721034311288,
"grad_norm": 1.3828125,
"learning_rate": 0.0003835998520085492,
"loss": 5.5779,
"mean_token_accuracy": 0.17881902307271957,
"num_tokens": 7982177.0,
"step": 4075
},
{
"entropy": 6.10365571975708,
"epoch": 4.0556936847339635,
"grad_norm": 1.21875,
"learning_rate": 0.0003832576611252993,
"loss": 5.5528,
"mean_token_accuracy": 0.18295233845710754,
"num_tokens": 7991845.0,
"step": 4080
},
{
"entropy": 6.0836097240448,
"epoch": 4.060666335156639,
"grad_norm": 1.140625,
"learning_rate": 0.0003829151441042572,
"loss": 5.5083,
"mean_token_accuracy": 0.18952144533395768,
"num_tokens": 8000731.0,
"step": 4085
},
{
"entropy": 6.06826605796814,
"epoch": 4.065638985579314,
"grad_norm": 1.0625,
"learning_rate": 0.00038257230197729237,
"loss": 5.5476,
"mean_token_accuracy": 0.17833086103200912,
"num_tokens": 8011137.0,
"step": 4090
},
{
"entropy": 6.11496787071228,
"epoch": 4.070611636001989,
"grad_norm": 1.1796875,
"learning_rate": 0.00038222913577725325,
"loss": 5.5899,
"mean_token_accuracy": 0.17287869453430177,
"num_tokens": 8020413.0,
"step": 4095
},
{
"entropy": 6.044083786010742,
"epoch": 4.0755842864246645,
"grad_norm": 1.2265625,
"learning_rate": 0.00038188564653796487,
"loss": 5.5381,
"mean_token_accuracy": 0.18592326194047928,
"num_tokens": 8029727.0,
"step": 4100
},
{
"entropy": 6.095187664031982,
"epoch": 4.08055693684734,
"grad_norm": 1.09375,
"learning_rate": 0.0003815418352942252,
"loss": 5.6096,
"mean_token_accuracy": 0.18130715936422348,
"num_tokens": 8038707.0,
"step": 4105
},
{
"entropy": 5.9706159114837645,
"epoch": 4.085529587270015,
"grad_norm": 1.15625,
"learning_rate": 0.0003811977030818026,
"loss": 5.3653,
"mean_token_accuracy": 0.19662608206272125,
"num_tokens": 8048604.0,
"step": 4110
},
{
"entropy": 6.062938785552978,
"epoch": 4.09050223769269,
"grad_norm": 1.15625,
"learning_rate": 0.000380853250937432,
"loss": 5.5782,
"mean_token_accuracy": 0.183298459649086,
"num_tokens": 8058089.0,
"step": 4115
},
{
"entropy": 6.228862714767456,
"epoch": 4.0954748881153655,
"grad_norm": 1.265625,
"learning_rate": 0.00038050847989881236,
"loss": 5.5829,
"mean_token_accuracy": 0.177506659924984,
"num_tokens": 8068078.0,
"step": 4120
},
{
"entropy": 5.995114278793335,
"epoch": 4.100447538538041,
"grad_norm": 1.125,
"learning_rate": 0.0003801633910046035,
"loss": 5.4587,
"mean_token_accuracy": 0.19045730233192443,
"num_tokens": 8076918.0,
"step": 4125
},
{
"entropy": 5.867263746261597,
"epoch": 4.105420188960716,
"grad_norm": 1.109375,
"learning_rate": 0.0003798179852944225,
"loss": 5.3961,
"mean_token_accuracy": 0.19580039530992507,
"num_tokens": 8087104.0,
"step": 4130
},
{
"entropy": 6.055717945098877,
"epoch": 4.110392839383391,
"grad_norm": 1.1953125,
"learning_rate": 0.00037947226380884115,
"loss": 5.654,
"mean_token_accuracy": 0.16547372192144394,
"num_tokens": 8096976.0,
"step": 4135
},
{
"entropy": 6.102532052993775,
"epoch": 4.1153654898060665,
"grad_norm": 1.046875,
"learning_rate": 0.00037912622758938243,
"loss": 5.5414,
"mean_token_accuracy": 0.18029424250125886,
"num_tokens": 8106707.0,
"step": 4140
},
{
"entropy": 6.086570358276367,
"epoch": 4.120338140228742,
"grad_norm": 1.0390625,
"learning_rate": 0.0003787798776785174,
"loss": 5.5819,
"mean_token_accuracy": 0.17331225275993348,
"num_tokens": 8116822.0,
"step": 4145
},
{
"entropy": 5.974531888961792,
"epoch": 4.125310790651417,
"grad_norm": 1.28125,
"learning_rate": 0.0003784332151196622,
"loss": 5.5124,
"mean_token_accuracy": 0.1836848735809326,
"num_tokens": 8125677.0,
"step": 4150
},
{
"entropy": 5.90705246925354,
"epoch": 4.130283441074092,
"grad_norm": 1.1953125,
"learning_rate": 0.00037808624095717503,
"loss": 5.4274,
"mean_token_accuracy": 0.19424668103456497,
"num_tokens": 8134878.0,
"step": 4155
},
{
"entropy": 6.076256513595581,
"epoch": 4.1352560914967675,
"grad_norm": 1.2109375,
"learning_rate": 0.0003777389562363526,
"loss": 5.5763,
"mean_token_accuracy": 0.18200969099998474,
"num_tokens": 8144491.0,
"step": 4160
},
{
"entropy": 6.089936447143555,
"epoch": 4.140228741919443,
"grad_norm": 1.2265625,
"learning_rate": 0.00037739136200342743,
"loss": 5.6102,
"mean_token_accuracy": 0.17207376658916473,
"num_tokens": 8155163.0,
"step": 4165
},
{
"entropy": 6.004365491867065,
"epoch": 4.145201392342118,
"grad_norm": 1.1328125,
"learning_rate": 0.00037704345930556426,
"loss": 5.5741,
"mean_token_accuracy": 0.17346905469894408,
"num_tokens": 8164626.0,
"step": 4170
},
{
"entropy": 6.169218111038208,
"epoch": 4.150174042764793,
"grad_norm": 1.1171875,
"learning_rate": 0.00037669524919085716,
"loss": 5.7291,
"mean_token_accuracy": 0.1693943291902542,
"num_tokens": 8174681.0,
"step": 4175
},
{
"entropy": 6.204012489318847,
"epoch": 4.1551466931874685,
"grad_norm": 1.125,
"learning_rate": 0.0003763467327083264,
"loss": 5.7866,
"mean_token_accuracy": 0.16910551488399506,
"num_tokens": 8184627.0,
"step": 4180
},
{
"entropy": 5.9840844631195065,
"epoch": 4.160119343610144,
"grad_norm": 1.15625,
"learning_rate": 0.00037599791090791527,
"loss": 5.4112,
"mean_token_accuracy": 0.1853877931833267,
"num_tokens": 8195029.0,
"step": 4185
},
{
"entropy": 6.041340446472168,
"epoch": 4.165091994032819,
"grad_norm": 2.09375,
"learning_rate": 0.0003756487848404868,
"loss": 5.4611,
"mean_token_accuracy": 0.1942393809556961,
"num_tokens": 8204897.0,
"step": 4190
},
{
"entropy": 6.056771850585937,
"epoch": 4.170064644455495,
"grad_norm": 1.0,
"learning_rate": 0.0003752993555578205,
"loss": 5.6834,
"mean_token_accuracy": 0.1679566189646721,
"num_tokens": 8215859.0,
"step": 4195
},
{
"entropy": 6.0442667484283445,
"epoch": 4.17503729487817,
"grad_norm": 1.203125,
"learning_rate": 0.0003749496241126095,
"loss": 5.4976,
"mean_token_accuracy": 0.18313408195972442,
"num_tokens": 8225377.0,
"step": 4200
},
{
"entropy": 6.076041746139526,
"epoch": 4.180009945300846,
"grad_norm": 1.140625,
"learning_rate": 0.00037459959155845726,
"loss": 5.691,
"mean_token_accuracy": 0.16364823058247566,
"num_tokens": 8235436.0,
"step": 4205
},
{
"entropy": 6.083760166168213,
"epoch": 4.184982595723521,
"grad_norm": 1.1796875,
"learning_rate": 0.0003742492589498744,
"loss": 5.6245,
"mean_token_accuracy": 0.17680579125881196,
"num_tokens": 8245129.0,
"step": 4210
},
{
"entropy": 6.039215135574341,
"epoch": 4.189955246146196,
"grad_norm": 1.1953125,
"learning_rate": 0.00037389862734227525,
"loss": 5.5549,
"mean_token_accuracy": 0.17913154512643814,
"num_tokens": 8254870.0,
"step": 4215
},
{
"entropy": 6.116287899017334,
"epoch": 4.194927896568871,
"grad_norm": 1.2109375,
"learning_rate": 0.00037354769779197514,
"loss": 5.6223,
"mean_token_accuracy": 0.17344698458909988,
"num_tokens": 8266078.0,
"step": 4220
},
{
"entropy": 6.081928539276123,
"epoch": 4.199900546991547,
"grad_norm": 1.2109375,
"learning_rate": 0.00037319647135618695,
"loss": 5.676,
"mean_token_accuracy": 0.17546181678771972,
"num_tokens": 8275833.0,
"step": 4225
},
{
"entropy": 5.996638298034668,
"epoch": 4.204873197414222,
"grad_norm": 1.1484375,
"learning_rate": 0.00037284494909301793,
"loss": 5.581,
"mean_token_accuracy": 0.17470307350158693,
"num_tokens": 8286866.0,
"step": 4230
},
{
"entropy": 6.186329460144043,
"epoch": 4.209845847836897,
"grad_norm": 1.2421875,
"learning_rate": 0.0003724931320614663,
"loss": 5.5989,
"mean_token_accuracy": 0.17571214735507965,
"num_tokens": 8296316.0,
"step": 4235
},
{
"entropy": 6.1434248924255375,
"epoch": 4.214818498259572,
"grad_norm": 1.1328125,
"learning_rate": 0.00037214102132141874,
"loss": 5.741,
"mean_token_accuracy": 0.1624046102166176,
"num_tokens": 8306023.0,
"step": 4240
},
{
"entropy": 5.93798131942749,
"epoch": 4.219791148682248,
"grad_norm": 1.109375,
"learning_rate": 0.0003717886179336465,
"loss": 5.4997,
"mean_token_accuracy": 0.18289718627929688,
"num_tokens": 8315808.0,
"step": 4245
},
{
"entropy": 6.094416522979737,
"epoch": 4.224763799104923,
"grad_norm": 1.1484375,
"learning_rate": 0.00037143592295980243,
"loss": 5.6042,
"mean_token_accuracy": 0.17310193926095963,
"num_tokens": 8324951.0,
"step": 4250
},
{
"entropy": 6.009879922866821,
"epoch": 4.229736449527598,
"grad_norm": 1.140625,
"learning_rate": 0.0003710829374624181,
"loss": 5.5899,
"mean_token_accuracy": 0.1784605026245117,
"num_tokens": 8335019.0,
"step": 4255
},
{
"entropy": 6.119559049606323,
"epoch": 4.2347090999502734,
"grad_norm": 1.2109375,
"learning_rate": 0.0003707296625048999,
"loss": 5.561,
"mean_token_accuracy": 0.1809312954545021,
"num_tokens": 8344496.0,
"step": 4260
},
{
"entropy": 6.071708011627197,
"epoch": 4.239681750372949,
"grad_norm": 1.140625,
"learning_rate": 0.0003703760991515264,
"loss": 5.609,
"mean_token_accuracy": 0.16997691243886948,
"num_tokens": 8353766.0,
"step": 4265
},
{
"entropy": 6.073993825912476,
"epoch": 4.244654400795624,
"grad_norm": 1.21875,
"learning_rate": 0.00037002224846744526,
"loss": 5.5947,
"mean_token_accuracy": 0.1682877093553543,
"num_tokens": 8363576.0,
"step": 4270
},
{
"entropy": 6.196797513961792,
"epoch": 4.249627051218299,
"grad_norm": 1.1484375,
"learning_rate": 0.00036966811151866943,
"loss": 5.6808,
"mean_token_accuracy": 0.17006082832813263,
"num_tokens": 8372714.0,
"step": 4275
},
{
"entropy": 5.970323657989502,
"epoch": 4.2545997016409745,
"grad_norm": 1.2578125,
"learning_rate": 0.00036931368937207436,
"loss": 5.5833,
"mean_token_accuracy": 0.1856955111026764,
"num_tokens": 8383563.0,
"step": 4280
},
{
"entropy": 6.016225719451905,
"epoch": 4.25957235206365,
"grad_norm": 1.171875,
"learning_rate": 0.0003689589830953948,
"loss": 5.5085,
"mean_token_accuracy": 0.18958783894777298,
"num_tokens": 8393352.0,
"step": 4285
},
{
"entropy": 6.0590564727783205,
"epoch": 4.264545002486325,
"grad_norm": 1.25,
"learning_rate": 0.0003686039937572213,
"loss": 5.6474,
"mean_token_accuracy": 0.17412520200014114,
"num_tokens": 8403219.0,
"step": 4290
},
{
"entropy": 5.990196228027344,
"epoch": 4.269517652909,
"grad_norm": 1.234375,
"learning_rate": 0.0003682487224269972,
"loss": 5.5013,
"mean_token_accuracy": 0.1872052416205406,
"num_tokens": 8411596.0,
"step": 4295
},
{
"entropy": 6.035375308990479,
"epoch": 4.2744903033316755,
"grad_norm": 1.3125,
"learning_rate": 0.0003678931701750155,
"loss": 5.5982,
"mean_token_accuracy": 0.1793920785188675,
"num_tokens": 8421472.0,
"step": 4300
},
{
"entropy": 6.040344381332398,
"epoch": 4.279462953754351,
"grad_norm": 1.0859375,
"learning_rate": 0.00036753733807241536,
"loss": 5.5206,
"mean_token_accuracy": 0.1836888685822487,
"num_tokens": 8431489.0,
"step": 4305
},
{
"entropy": 5.880771589279175,
"epoch": 4.284435604177026,
"grad_norm": 1.390625,
"learning_rate": 0.00036718122719117907,
"loss": 5.3547,
"mean_token_accuracy": 0.20107638984918594,
"num_tokens": 8441108.0,
"step": 4310
},
{
"entropy": 5.893637561798096,
"epoch": 4.289408254599701,
"grad_norm": 1.203125,
"learning_rate": 0.0003668248386041288,
"loss": 5.4273,
"mean_token_accuracy": 0.19837938100099564,
"num_tokens": 8451664.0,
"step": 4315
},
{
"entropy": 6.074677515029907,
"epoch": 4.294380905022377,
"grad_norm": 1.234375,
"learning_rate": 0.00036646817338492334,
"loss": 5.6321,
"mean_token_accuracy": 0.1751484215259552,
"num_tokens": 8460531.0,
"step": 4320
},
{
"entropy": 6.093930625915528,
"epoch": 4.299353555445053,
"grad_norm": 1.234375,
"learning_rate": 0.0003661112326080548,
"loss": 5.5971,
"mean_token_accuracy": 0.18014349937438964,
"num_tokens": 8470018.0,
"step": 4325
},
{
"entropy": 5.9427752017974855,
"epoch": 4.304326205867728,
"grad_norm": 1.2421875,
"learning_rate": 0.0003657540173488455,
"loss": 5.5249,
"mean_token_accuracy": 0.18636006712913514,
"num_tokens": 8480095.0,
"step": 4330
},
{
"entropy": 6.090675592422485,
"epoch": 4.309298856290403,
"grad_norm": 1.296875,
"learning_rate": 0.0003653965286834446,
"loss": 5.6639,
"mean_token_accuracy": 0.16785015910863876,
"num_tokens": 8489817.0,
"step": 4335
},
{
"entropy": 6.118750476837159,
"epoch": 4.314271506713078,
"grad_norm": 1.234375,
"learning_rate": 0.0003650387676888251,
"loss": 5.6793,
"mean_token_accuracy": 0.17175161987543106,
"num_tokens": 8499333.0,
"step": 4340
},
{
"entropy": 6.061108827590942,
"epoch": 4.319244157135754,
"grad_norm": 1.21875,
"learning_rate": 0.0003646807354427802,
"loss": 5.6465,
"mean_token_accuracy": 0.1705202966928482,
"num_tokens": 8508732.0,
"step": 4345
},
{
"entropy": 6.051059722900391,
"epoch": 4.324216807558429,
"grad_norm": 1.078125,
"learning_rate": 0.00036432243302392025,
"loss": 5.6395,
"mean_token_accuracy": 0.17406514883041382,
"num_tokens": 8519525.0,
"step": 4350
},
{
"entropy": 6.003647899627685,
"epoch": 4.329189457981104,
"grad_norm": 1.171875,
"learning_rate": 0.00036396386151166975,
"loss": 5.4996,
"mean_token_accuracy": 0.191575026512146,
"num_tokens": 8529559.0,
"step": 4355
},
{
"entropy": 6.058806085586548,
"epoch": 4.334162108403779,
"grad_norm": 1.25,
"learning_rate": 0.0003636050219862638,
"loss": 5.553,
"mean_token_accuracy": 0.17373328357934953,
"num_tokens": 8538615.0,
"step": 4360
},
{
"entropy": 6.00287880897522,
"epoch": 4.339134758826455,
"grad_norm": 1.0859375,
"learning_rate": 0.00036324591552874493,
"loss": 5.5368,
"mean_token_accuracy": 0.18158403038978577,
"num_tokens": 8548432.0,
"step": 4365
},
{
"entropy": 5.98505048751831,
"epoch": 4.34410740924913,
"grad_norm": 1.2109375,
"learning_rate": 0.00036288654322095973,
"loss": 5.5162,
"mean_token_accuracy": 0.18717821091413497,
"num_tokens": 8558216.0,
"step": 4370
},
{
"entropy": 6.035119962692261,
"epoch": 4.349080059671805,
"grad_norm": 1.2578125,
"learning_rate": 0.00036252690614555557,
"loss": 5.438,
"mean_token_accuracy": 0.18823740482330323,
"num_tokens": 8568234.0,
"step": 4375
},
{
"entropy": 6.015963745117188,
"epoch": 4.35405271009448,
"grad_norm": 1.25,
"learning_rate": 0.0003621670053859778,
"loss": 5.6118,
"mean_token_accuracy": 0.1740383729338646,
"num_tokens": 8578140.0,
"step": 4380
},
{
"entropy": 6.069340801239013,
"epoch": 4.359025360517156,
"grad_norm": 1.2734375,
"learning_rate": 0.000361806842026466,
"loss": 5.6116,
"mean_token_accuracy": 0.17976488918066025,
"num_tokens": 8587976.0,
"step": 4385
},
{
"entropy": 6.0301127433776855,
"epoch": 4.363998010939831,
"grad_norm": 1.3203125,
"learning_rate": 0.00036144641715205077,
"loss": 5.6386,
"mean_token_accuracy": 0.17513138949871063,
"num_tokens": 8597739.0,
"step": 4390
},
{
"entropy": 6.033037996292114,
"epoch": 4.368970661362506,
"grad_norm": 1.15625,
"learning_rate": 0.0003610857318485507,
"loss": 5.6775,
"mean_token_accuracy": 0.1767078623175621,
"num_tokens": 8608609.0,
"step": 4395
},
{
"entropy": 6.064388275146484,
"epoch": 4.373943311785181,
"grad_norm": 1.1796875,
"learning_rate": 0.00036072478720256885,
"loss": 5.5924,
"mean_token_accuracy": 0.17492693215608596,
"num_tokens": 8617870.0,
"step": 4400
},
{
"entropy": 6.047692203521729,
"epoch": 4.378915962207857,
"grad_norm": 1.265625,
"learning_rate": 0.0003603635843014895,
"loss": 5.5824,
"mean_token_accuracy": 0.17925060093402861,
"num_tokens": 8627811.0,
"step": 4405
},
{
"entropy": 6.116177988052368,
"epoch": 4.383888612630532,
"grad_norm": 1.2578125,
"learning_rate": 0.00036000212423347523,
"loss": 5.6609,
"mean_token_accuracy": 0.175652177631855,
"num_tokens": 8636989.0,
"step": 4410
},
{
"entropy": 6.024384880065918,
"epoch": 4.388861263053207,
"grad_norm": 1.125,
"learning_rate": 0.000359640408087463,
"loss": 5.5036,
"mean_token_accuracy": 0.1824796810746193,
"num_tokens": 8646090.0,
"step": 4415
},
{
"entropy": 5.991547870635986,
"epoch": 4.393833913475882,
"grad_norm": 1.1796875,
"learning_rate": 0.0003592784369531615,
"loss": 5.5312,
"mean_token_accuracy": 0.17987066209316255,
"num_tokens": 8656244.0,
"step": 4420
},
{
"entropy": 5.945890426635742,
"epoch": 4.398806563898558,
"grad_norm": 1.125,
"learning_rate": 0.00035891621192104764,
"loss": 5.4804,
"mean_token_accuracy": 0.19014694839715957,
"num_tokens": 8665783.0,
"step": 4425
},
{
"entropy": 6.097377109527588,
"epoch": 4.403779214321233,
"grad_norm": 1.1953125,
"learning_rate": 0.00035855373408236294,
"loss": 5.6944,
"mean_token_accuracy": 0.17162372320890426,
"num_tokens": 8676736.0,
"step": 4430
},
{
"entropy": 6.032937335968017,
"epoch": 4.408751864743908,
"grad_norm": 1.1171875,
"learning_rate": 0.0003581910045291107,
"loss": 5.6249,
"mean_token_accuracy": 0.1742832511663437,
"num_tokens": 8687035.0,
"step": 4435
},
{
"entropy": 6.100921726226806,
"epoch": 4.413724515166583,
"grad_norm": 1.296875,
"learning_rate": 0.0003578280243540526,
"loss": 5.656,
"mean_token_accuracy": 0.17648518830537796,
"num_tokens": 8696939.0,
"step": 4440
},
{
"entropy": 6.021227693557739,
"epoch": 4.4186971655892595,
"grad_norm": 1.1171875,
"learning_rate": 0.00035746479465070515,
"loss": 5.6298,
"mean_token_accuracy": 0.1803789719939232,
"num_tokens": 8707213.0,
"step": 4445
},
{
"entropy": 5.972588682174683,
"epoch": 4.423669816011935,
"grad_norm": 1.1953125,
"learning_rate": 0.00035710131651333683,
"loss": 5.5832,
"mean_token_accuracy": 0.18056459873914718,
"num_tokens": 8717066.0,
"step": 4450
},
{
"entropy": 6.038032865524292,
"epoch": 4.42864246643461,
"grad_norm": 1.1953125,
"learning_rate": 0.00035673759103696447,
"loss": 5.6037,
"mean_token_accuracy": 0.17720283716917037,
"num_tokens": 8727421.0,
"step": 4455
},
{
"entropy": 6.098892688751221,
"epoch": 4.433615116857285,
"grad_norm": 1.1875,
"learning_rate": 0.0003563736193173499,
"loss": 5.6504,
"mean_token_accuracy": 0.17696666717529297,
"num_tokens": 8736086.0,
"step": 4460
},
{
"entropy": 6.114102602005005,
"epoch": 4.4385877672799605,
"grad_norm": 1.3203125,
"learning_rate": 0.0003560094024509969,
"loss": 5.692,
"mean_token_accuracy": 0.17179251909255983,
"num_tokens": 8746496.0,
"step": 4465
},
{
"entropy": 6.0173756122589115,
"epoch": 4.443560417702636,
"grad_norm": 1.2890625,
"learning_rate": 0.00035564494153514777,
"loss": 5.5374,
"mean_token_accuracy": 0.1894352987408638,
"num_tokens": 8755528.0,
"step": 4470
},
{
"entropy": 6.068454360961914,
"epoch": 4.448533068125311,
"grad_norm": 1.25,
"learning_rate": 0.00035528023766778,
"loss": 5.62,
"mean_token_accuracy": 0.1667730301618576,
"num_tokens": 8765486.0,
"step": 4475
},
{
"entropy": 6.055483961105347,
"epoch": 4.453505718547986,
"grad_norm": 1.046875,
"learning_rate": 0.0003549152919476031,
"loss": 5.6708,
"mean_token_accuracy": 0.17374666333198546,
"num_tokens": 8775315.0,
"step": 4480
},
{
"entropy": 6.0540163040161135,
"epoch": 4.4584783689706615,
"grad_norm": 1.3203125,
"learning_rate": 0.00035455010547405513,
"loss": 5.6622,
"mean_token_accuracy": 0.16853180080652236,
"num_tokens": 8784976.0,
"step": 4485
},
{
"entropy": 6.062339735031128,
"epoch": 4.463451019393337,
"grad_norm": 1.1796875,
"learning_rate": 0.00035418467934729934,
"loss": 5.6929,
"mean_token_accuracy": 0.16901296079158784,
"num_tokens": 8795412.0,
"step": 4490
},
{
"entropy": 6.008714962005615,
"epoch": 4.468423669816012,
"grad_norm": 1.3125,
"learning_rate": 0.00035381901466822113,
"loss": 5.5399,
"mean_token_accuracy": 0.18363385647535324,
"num_tokens": 8805277.0,
"step": 4495
},
{
"entropy": 6.026242113113403,
"epoch": 4.473396320238687,
"grad_norm": 1.2421875,
"learning_rate": 0.0003534531125384244,
"loss": 5.6114,
"mean_token_accuracy": 0.1756073758006096,
"num_tokens": 8814308.0,
"step": 4500
},
{
"epoch": 4.473396320238687,
"eval_entropy": 5.932372247019122,
"eval_loss": 6.589542865753174,
"eval_mean_token_accuracy": 0.14009550782049307,
"eval_num_tokens": 8814308.0,
"eval_runtime": 2.5353,
"eval_samples_per_second": 1366.692,
"eval_steps_per_second": 171.182,
"step": 4500
},
{
"entropy": 5.998975419998169,
"epoch": 4.4783689706613625,
"grad_norm": 1.0703125,
"learning_rate": 0.00035308697406022865,
"loss": 5.6552,
"mean_token_accuracy": 0.17434559911489486,
"num_tokens": 8824836.0,
"step": 4505
},
{
"entropy": 5.943307113647461,
"epoch": 4.483341621084038,
"grad_norm": 1.109375,
"learning_rate": 0.00035272060033666503,
"loss": 5.4808,
"mean_token_accuracy": 0.19201576411724092,
"num_tokens": 8834390.0,
"step": 4510
},
{
"entropy": 6.094746208190918,
"epoch": 4.488314271506713,
"grad_norm": 1.5234375,
"learning_rate": 0.0003523539924714738,
"loss": 5.7066,
"mean_token_accuracy": 0.17179358899593353,
"num_tokens": 8844266.0,
"step": 4515
},
{
"entropy": 6.07678632736206,
"epoch": 4.493286921929388,
"grad_norm": 1.328125,
"learning_rate": 0.00035198715156910016,
"loss": 5.5804,
"mean_token_accuracy": 0.1798890933394432,
"num_tokens": 8852876.0,
"step": 4520
},
{
"entropy": 5.8529962539672855,
"epoch": 4.4982595723520635,
"grad_norm": 1.1640625,
"learning_rate": 0.00035162007873469175,
"loss": 5.3023,
"mean_token_accuracy": 0.2094032034277916,
"num_tokens": 8862861.0,
"step": 4525
},
{
"entropy": 6.002284145355224,
"epoch": 4.503232222774739,
"grad_norm": 1.2109375,
"learning_rate": 0.00035125277507409483,
"loss": 5.5646,
"mean_token_accuracy": 0.18428260535001756,
"num_tokens": 8872177.0,
"step": 4530
},
{
"entropy": 5.93495922088623,
"epoch": 4.508204873197414,
"grad_norm": 1.2265625,
"learning_rate": 0.00035088524169385075,
"loss": 5.5945,
"mean_token_accuracy": 0.17870858162641526,
"num_tokens": 8882202.0,
"step": 4535
},
{
"entropy": 6.024176549911499,
"epoch": 4.513177523620089,
"grad_norm": 1.1640625,
"learning_rate": 0.00035051747970119354,
"loss": 5.6016,
"mean_token_accuracy": 0.1791429489850998,
"num_tokens": 8892361.0,
"step": 4540
},
{
"entropy": 6.099696445465088,
"epoch": 4.5181501740427645,
"grad_norm": 1.125,
"learning_rate": 0.00035014949020404523,
"loss": 5.6884,
"mean_token_accuracy": 0.17060996890068053,
"num_tokens": 8903014.0,
"step": 4545
},
{
"entropy": 6.021279954910279,
"epoch": 4.52312282446544,
"grad_norm": 1.1796875,
"learning_rate": 0.00034978127431101365,
"loss": 5.6539,
"mean_token_accuracy": 0.17107139080762862,
"num_tokens": 8912145.0,
"step": 4550
},
{
"entropy": 6.008196640014648,
"epoch": 4.528095474888115,
"grad_norm": 1.140625,
"learning_rate": 0.00034941283313138865,
"loss": 5.5662,
"mean_token_accuracy": 0.18208493739366532,
"num_tokens": 8921748.0,
"step": 4555
},
{
"entropy": 6.018685579299927,
"epoch": 4.53306812531079,
"grad_norm": 1.1171875,
"learning_rate": 0.0003490441677751387,
"loss": 5.6681,
"mean_token_accuracy": 0.1748025581240654,
"num_tokens": 8932327.0,
"step": 4560
},
{
"entropy": 6.054449415206909,
"epoch": 4.5380407757334655,
"grad_norm": 1.2734375,
"learning_rate": 0.00034867527935290773,
"loss": 5.6834,
"mean_token_accuracy": 0.17018242329359054,
"num_tokens": 8942781.0,
"step": 4565
},
{
"entropy": 6.147935771942139,
"epoch": 4.543013426156142,
"grad_norm": 1.28125,
"learning_rate": 0.00034830616897601146,
"loss": 5.6472,
"mean_token_accuracy": 0.17360907047986984,
"num_tokens": 8951887.0,
"step": 4570
},
{
"entropy": 6.03584303855896,
"epoch": 4.547986076578816,
"grad_norm": 1.171875,
"learning_rate": 0.0003479368377564344,
"loss": 5.6853,
"mean_token_accuracy": 0.172353096306324,
"num_tokens": 8961454.0,
"step": 4575
},
{
"entropy": 5.9941725730896,
"epoch": 4.552958727001492,
"grad_norm": 1.1875,
"learning_rate": 0.0003475672868068265,
"loss": 5.5551,
"mean_token_accuracy": 0.17939372211694718,
"num_tokens": 8970218.0,
"step": 4580
},
{
"entropy": 6.082084035873413,
"epoch": 4.557931377424167,
"grad_norm": 1.2890625,
"learning_rate": 0.00034719751724049955,
"loss": 5.5806,
"mean_token_accuracy": 0.17678709626197814,
"num_tokens": 8980141.0,
"step": 4585
},
{
"entropy": 5.962429332733154,
"epoch": 4.562904027846843,
"grad_norm": 1.328125,
"learning_rate": 0.00034682753017142387,
"loss": 5.5814,
"mean_token_accuracy": 0.1805876150727272,
"num_tokens": 8989514.0,
"step": 4590
},
{
"entropy": 6.002709293365479,
"epoch": 4.567876678269518,
"grad_norm": 1.203125,
"learning_rate": 0.00034645732671422517,
"loss": 5.7014,
"mean_token_accuracy": 0.17099539488554,
"num_tokens": 9000054.0,
"step": 4595
},
{
"entropy": 6.024324035644531,
"epoch": 4.572849328692193,
"grad_norm": 1.1171875,
"learning_rate": 0.0003460869079841809,
"loss": 5.7023,
"mean_token_accuracy": 0.1759747162461281,
"num_tokens": 9010070.0,
"step": 4600
},
{
"entropy": 6.116684293746948,
"epoch": 4.577821979114868,
"grad_norm": 1.1875,
"learning_rate": 0.0003457162750972172,
"loss": 5.6751,
"mean_token_accuracy": 0.16745835095643996,
"num_tokens": 9019785.0,
"step": 4605
},
{
"entropy": 6.051382446289063,
"epoch": 4.582794629537544,
"grad_norm": 1.1875,
"learning_rate": 0.00034534542916990534,
"loss": 5.6663,
"mean_token_accuracy": 0.1736306443810463,
"num_tokens": 9028281.0,
"step": 4610
},
{
"entropy": 5.8645693302154545,
"epoch": 4.587767279960219,
"grad_norm": 1.1640625,
"learning_rate": 0.00034497437131945837,
"loss": 5.433,
"mean_token_accuracy": 0.18875787109136583,
"num_tokens": 9037628.0,
"step": 4615
},
{
"entropy": 6.1155859470367435,
"epoch": 4.592739930382894,
"grad_norm": 1.15625,
"learning_rate": 0.0003446031026637277,
"loss": 5.681,
"mean_token_accuracy": 0.1732897162437439,
"num_tokens": 9047939.0,
"step": 4620
},
{
"entropy": 6.038040828704834,
"epoch": 4.597712580805569,
"grad_norm": 1.2421875,
"learning_rate": 0.00034423162432119995,
"loss": 5.6296,
"mean_token_accuracy": 0.17677718102931977,
"num_tokens": 9056583.0,
"step": 4625
},
{
"entropy": 5.994035863876343,
"epoch": 4.602685231228245,
"grad_norm": 1.25,
"learning_rate": 0.0003438599374109934,
"loss": 5.5988,
"mean_token_accuracy": 0.17510462403297425,
"num_tokens": 9067610.0,
"step": 4630
},
{
"entropy": 6.043559217453003,
"epoch": 4.60765788165092,
"grad_norm": 1.0703125,
"learning_rate": 0.00034348804305285466,
"loss": 5.5982,
"mean_token_accuracy": 0.17402266561985016,
"num_tokens": 9077958.0,
"step": 4635
},
{
"entropy": 6.03648157119751,
"epoch": 4.612630532073595,
"grad_norm": 1.3046875,
"learning_rate": 0.0003431159423671552,
"loss": 5.6928,
"mean_token_accuracy": 0.17147929072380066,
"num_tokens": 9088821.0,
"step": 4640
},
{
"entropy": 5.997366046905517,
"epoch": 4.61760318249627,
"grad_norm": 1.21875,
"learning_rate": 0.00034274363647488826,
"loss": 5.5408,
"mean_token_accuracy": 0.1787536010146141,
"num_tokens": 9098495.0,
"step": 4645
},
{
"entropy": 6.018900680541992,
"epoch": 4.622575832918946,
"grad_norm": 1.2109375,
"learning_rate": 0.0003423711264976652,
"loss": 5.594,
"mean_token_accuracy": 0.1797969177365303,
"num_tokens": 9107279.0,
"step": 4650
},
{
"entropy": 6.076930236816406,
"epoch": 4.627548483341621,
"grad_norm": 1.203125,
"learning_rate": 0.00034199841355771226,
"loss": 5.6326,
"mean_token_accuracy": 0.17793766409158707,
"num_tokens": 9118182.0,
"step": 4655
},
{
"entropy": 6.016601800918579,
"epoch": 4.632521133764296,
"grad_norm": 1.234375,
"learning_rate": 0.00034162549877786685,
"loss": 5.5795,
"mean_token_accuracy": 0.1833501473069191,
"num_tokens": 9128299.0,
"step": 4660
},
{
"entropy": 6.075742292404175,
"epoch": 4.6374937841869714,
"grad_norm": 1.125,
"learning_rate": 0.0003412523832815748,
"loss": 5.7282,
"mean_token_accuracy": 0.16387763023376464,
"num_tokens": 9138097.0,
"step": 4665
},
{
"entropy": 5.9717409133911135,
"epoch": 4.642466434609647,
"grad_norm": 1.0546875,
"learning_rate": 0.0003408790681928865,
"loss": 5.502,
"mean_token_accuracy": 0.18098540902137755,
"num_tokens": 9148951.0,
"step": 4670
},
{
"entropy": 6.051633644104004,
"epoch": 4.647439085032322,
"grad_norm": 1.25,
"learning_rate": 0.0003405055546364535,
"loss": 5.5969,
"mean_token_accuracy": 0.17990484535694123,
"num_tokens": 9158372.0,
"step": 4675
},
{
"entropy": 5.9092339992523195,
"epoch": 4.652411735454997,
"grad_norm": 1.359375,
"learning_rate": 0.00034013184373752557,
"loss": 5.48,
"mean_token_accuracy": 0.18555311113595963,
"num_tokens": 9168308.0,
"step": 4680
},
{
"entropy": 5.935237789154053,
"epoch": 4.6573843858776725,
"grad_norm": 1.28125,
"learning_rate": 0.0003397579366219466,
"loss": 5.5154,
"mean_token_accuracy": 0.19447309523820877,
"num_tokens": 9178200.0,
"step": 4685
},
{
"entropy": 6.030032205581665,
"epoch": 4.662357036300348,
"grad_norm": 1.234375,
"learning_rate": 0.0003393838344161519,
"loss": 5.6528,
"mean_token_accuracy": 0.17138334065675737,
"num_tokens": 9187759.0,
"step": 4690
},
{
"entropy": 5.801412057876587,
"epoch": 4.667329686723024,
"grad_norm": 1.4609375,
"learning_rate": 0.00033900953824716443,
"loss": 5.281,
"mean_token_accuracy": 0.20511690080165862,
"num_tokens": 9198530.0,
"step": 4695
},
{
"entropy": 5.9435622215271,
"epoch": 4.672302337145698,
"grad_norm": 1.234375,
"learning_rate": 0.0003386350492425913,
"loss": 5.6327,
"mean_token_accuracy": 0.1728418454527855,
"num_tokens": 9207668.0,
"step": 4700
},
{
"entropy": 6.06575117111206,
"epoch": 4.677274987568374,
"grad_norm": 1.296875,
"learning_rate": 0.00033826036853062095,
"loss": 5.7421,
"mean_token_accuracy": 0.1739937499165535,
"num_tokens": 9216856.0,
"step": 4705
},
{
"entropy": 6.0592506408691404,
"epoch": 4.68224763799105,
"grad_norm": 1.1796875,
"learning_rate": 0.00033788549724001897,
"loss": 5.6419,
"mean_token_accuracy": 0.17309482544660568,
"num_tokens": 9226307.0,
"step": 4710
},
{
"entropy": 5.942525053024292,
"epoch": 4.687220288413725,
"grad_norm": 1.2109375,
"learning_rate": 0.00033751043650012534,
"loss": 5.6072,
"mean_token_accuracy": 0.18039609491825104,
"num_tokens": 9235919.0,
"step": 4715
},
{
"entropy": 5.887523221969604,
"epoch": 4.6921929388364,
"grad_norm": 1.1953125,
"learning_rate": 0.0003371351874408506,
"loss": 5.3639,
"mean_token_accuracy": 0.19560288041830062,
"num_tokens": 9245542.0,
"step": 4720
},
{
"entropy": 6.125654506683349,
"epoch": 4.697165589259075,
"grad_norm": 1.1640625,
"learning_rate": 0.0003367597511926728,
"loss": 5.7179,
"mean_token_accuracy": 0.17309315353631974,
"num_tokens": 9256146.0,
"step": 4725
},
{
"entropy": 6.075051927566529,
"epoch": 4.702138239681751,
"grad_norm": 1.15625,
"learning_rate": 0.00033638412888663374,
"loss": 5.741,
"mean_token_accuracy": 0.17186035960912704,
"num_tokens": 9266694.0,
"step": 4730
},
{
"entropy": 6.003686714172363,
"epoch": 4.707110890104426,
"grad_norm": 1.2265625,
"learning_rate": 0.0003360083216543358,
"loss": 5.659,
"mean_token_accuracy": 0.17265633791685103,
"num_tokens": 9276155.0,
"step": 4735
},
{
"entropy": 6.024498319625854,
"epoch": 4.712083540527101,
"grad_norm": 1.2578125,
"learning_rate": 0.0003356323306279387,
"loss": 5.6337,
"mean_token_accuracy": 0.17213286757469176,
"num_tokens": 9286048.0,
"step": 4740
},
{
"entropy": 6.135942173004151,
"epoch": 4.717056190949776,
"grad_norm": 0.984375,
"learning_rate": 0.00033525615694015545,
"loss": 5.7317,
"mean_token_accuracy": 0.16332919001579285,
"num_tokens": 9296515.0,
"step": 4745
},
{
"entropy": 5.899558115005493,
"epoch": 4.722028841372452,
"grad_norm": 1.265625,
"learning_rate": 0.00033487980172424955,
"loss": 5.5268,
"mean_token_accuracy": 0.19124788492918016,
"num_tokens": 9306737.0,
"step": 4750
},
{
"entropy": 6.026609706878662,
"epoch": 4.727001491795127,
"grad_norm": 1.203125,
"learning_rate": 0.00033450326611403136,
"loss": 5.6808,
"mean_token_accuracy": 0.17124763876199722,
"num_tokens": 9316453.0,
"step": 4755
},
{
"entropy": 6.034600734710693,
"epoch": 4.731974142217802,
"grad_norm": 1.171875,
"learning_rate": 0.00033412655124385494,
"loss": 5.6069,
"mean_token_accuracy": 0.1798119679093361,
"num_tokens": 9327422.0,
"step": 4760
},
{
"entropy": 6.054593706130982,
"epoch": 4.736946792640477,
"grad_norm": 1.171875,
"learning_rate": 0.0003337496582486139,
"loss": 5.5777,
"mean_token_accuracy": 0.18113671690225602,
"num_tokens": 9336890.0,
"step": 4765
},
{
"entropy": 6.041345262527466,
"epoch": 4.741919443063153,
"grad_norm": 1.171875,
"learning_rate": 0.00033337258826373896,
"loss": 5.6546,
"mean_token_accuracy": 0.1782251015305519,
"num_tokens": 9346807.0,
"step": 4770
},
{
"entropy": 6.061462640762329,
"epoch": 4.746892093485828,
"grad_norm": 1.109375,
"learning_rate": 0.00033299534242519355,
"loss": 5.6259,
"mean_token_accuracy": 0.17500995099544525,
"num_tokens": 9356343.0,
"step": 4775
},
{
"entropy": 6.053284788131714,
"epoch": 4.751864743908503,
"grad_norm": 1.1171875,
"learning_rate": 0.00033261792186947133,
"loss": 5.636,
"mean_token_accuracy": 0.1725785732269287,
"num_tokens": 9366936.0,
"step": 4780
},
{
"entropy": 6.043677186965942,
"epoch": 4.756837394331178,
"grad_norm": 1.03125,
"learning_rate": 0.00033224032773359203,
"loss": 5.6052,
"mean_token_accuracy": 0.18076443374156953,
"num_tokens": 9377520.0,
"step": 4785
},
{
"entropy": 6.052340507507324,
"epoch": 4.761810044753854,
"grad_norm": 1.2109375,
"learning_rate": 0.00033186256115509845,
"loss": 5.721,
"mean_token_accuracy": 0.1731985554099083,
"num_tokens": 9386498.0,
"step": 4790
},
{
"entropy": 6.0284703254699705,
"epoch": 4.766782695176529,
"grad_norm": 1.2890625,
"learning_rate": 0.0003314846232720528,
"loss": 5.6535,
"mean_token_accuracy": 0.17431531399488448,
"num_tokens": 9396153.0,
"step": 4795
},
{
"entropy": 6.079997587203979,
"epoch": 4.771755345599204,
"grad_norm": 1.1875,
"learning_rate": 0.00033110651522303346,
"loss": 5.6564,
"mean_token_accuracy": 0.1774495542049408,
"num_tokens": 9406019.0,
"step": 4800
},
{
"entropy": 6.00224118232727,
"epoch": 4.776727996021879,
"grad_norm": 1.2265625,
"learning_rate": 0.0003307282381471314,
"loss": 5.6343,
"mean_token_accuracy": 0.17681074738502503,
"num_tokens": 9416019.0,
"step": 4805
},
{
"entropy": 5.992507362365723,
"epoch": 4.781700646444555,
"grad_norm": 1.28125,
"learning_rate": 0.00033034979318394676,
"loss": 5.5951,
"mean_token_accuracy": 0.17914985567331315,
"num_tokens": 9425874.0,
"step": 4810
},
{
"entropy": 6.050097990036011,
"epoch": 4.78667329686723,
"grad_norm": 1.296875,
"learning_rate": 0.00032997118147358554,
"loss": 5.6595,
"mean_token_accuracy": 0.17315989434719087,
"num_tokens": 9435252.0,
"step": 4815
},
{
"entropy": 6.051864671707153,
"epoch": 4.791645947289906,
"grad_norm": 1.1953125,
"learning_rate": 0.000329592404156656,
"loss": 5.6102,
"mean_token_accuracy": 0.17624711990356445,
"num_tokens": 9444696.0,
"step": 4820
},
{
"entropy": 6.07463059425354,
"epoch": 4.79661859771258,
"grad_norm": 1.2578125,
"learning_rate": 0.00032921346237426536,
"loss": 5.6361,
"mean_token_accuracy": 0.17828792482614517,
"num_tokens": 9454163.0,
"step": 4825
},
{
"entropy": 5.964988279342651,
"epoch": 4.8015912481352565,
"grad_norm": 1.125,
"learning_rate": 0.00032883435726801635,
"loss": 5.4416,
"mean_token_accuracy": 0.19737583845853807,
"num_tokens": 9463058.0,
"step": 4830
},
{
"entropy": 5.960331106185913,
"epoch": 4.806563898557932,
"grad_norm": 1.2578125,
"learning_rate": 0.0003284550899800036,
"loss": 5.5807,
"mean_token_accuracy": 0.1837708070874214,
"num_tokens": 9472302.0,
"step": 4835
},
{
"entropy": 6.077133893966675,
"epoch": 4.811536548980607,
"grad_norm": 1.3046875,
"learning_rate": 0.00032807566165281055,
"loss": 5.6519,
"mean_token_accuracy": 0.17796968221664428,
"num_tokens": 9482607.0,
"step": 4840
},
{
"entropy": 6.022154664993286,
"epoch": 4.816509199403282,
"grad_norm": 1.390625,
"learning_rate": 0.0003276960734295054,
"loss": 5.6163,
"mean_token_accuracy": 0.17916265577077867,
"num_tokens": 9492451.0,
"step": 4845
},
{
"entropy": 6.028244829177856,
"epoch": 4.8214818498259575,
"grad_norm": 1.125,
"learning_rate": 0.00032731632645363856,
"loss": 5.6566,
"mean_token_accuracy": 0.1692410185933113,
"num_tokens": 9502633.0,
"step": 4850
},
{
"entropy": 5.966323137283325,
"epoch": 4.826454500248633,
"grad_norm": 1.125,
"learning_rate": 0.0003269364218692384,
"loss": 5.5467,
"mean_token_accuracy": 0.18376618027687072,
"num_tokens": 9512887.0,
"step": 4855
},
{
"entropy": 5.971394872665405,
"epoch": 4.831427150671308,
"grad_norm": 1.28125,
"learning_rate": 0.0003265563608208081,
"loss": 5.5741,
"mean_token_accuracy": 0.18436105996370317,
"num_tokens": 9522210.0,
"step": 4860
},
{
"entropy": 5.887615728378296,
"epoch": 4.836399801093983,
"grad_norm": 1.234375,
"learning_rate": 0.0003261761444533221,
"loss": 5.4716,
"mean_token_accuracy": 0.18887942284345627,
"num_tokens": 9532790.0,
"step": 4865
},
{
"entropy": 6.042507553100586,
"epoch": 4.8413724515166585,
"grad_norm": 1.1015625,
"learning_rate": 0.0003257957739122231,
"loss": 5.6201,
"mean_token_accuracy": 0.17621637284755706,
"num_tokens": 9543429.0,
"step": 4870
},
{
"entropy": 6.070701694488525,
"epoch": 4.846345101939334,
"grad_norm": 1.1875,
"learning_rate": 0.00032541525034341805,
"loss": 5.6938,
"mean_token_accuracy": 0.17254300713539122,
"num_tokens": 9552722.0,
"step": 4875
},
{
"entropy": 6.004253339767456,
"epoch": 4.851317752362009,
"grad_norm": 1.3125,
"learning_rate": 0.0003250345748932749,
"loss": 5.576,
"mean_token_accuracy": 0.1779184639453888,
"num_tokens": 9560880.0,
"step": 4880
},
{
"entropy": 6.070407962799072,
"epoch": 4.856290402784684,
"grad_norm": 1.28125,
"learning_rate": 0.00032465374870861913,
"loss": 5.652,
"mean_token_accuracy": 0.1710787296295166,
"num_tokens": 9570553.0,
"step": 4885
},
{
"entropy": 5.991409873962402,
"epoch": 4.8612630532073595,
"grad_norm": 1.15625,
"learning_rate": 0.00032427277293673053,
"loss": 5.6775,
"mean_token_accuracy": 0.1701473355293274,
"num_tokens": 9581313.0,
"step": 4890
},
{
"entropy": 6.03027024269104,
"epoch": 4.866235703630035,
"grad_norm": 1.25,
"learning_rate": 0.0003238916487253393,
"loss": 5.623,
"mean_token_accuracy": 0.173647241294384,
"num_tokens": 9589545.0,
"step": 4895
},
{
"entropy": 6.067294263839722,
"epoch": 4.87120835405271,
"grad_norm": 1.15625,
"learning_rate": 0.0003235103772226229,
"loss": 5.662,
"mean_token_accuracy": 0.17482498437166213,
"num_tokens": 9599489.0,
"step": 4900
},
{
"entropy": 6.045927238464356,
"epoch": 4.876181004475385,
"grad_norm": 1.1796875,
"learning_rate": 0.0003231289595772026,
"loss": 5.7172,
"mean_token_accuracy": 0.17421929389238358,
"num_tokens": 9609346.0,
"step": 4905
},
{
"entropy": 5.8537744045257565,
"epoch": 4.8811536548980605,
"grad_norm": 1.171875,
"learning_rate": 0.0003227473969381398,
"loss": 5.368,
"mean_token_accuracy": 0.20267200022935866,
"num_tokens": 9618808.0,
"step": 4910
},
{
"entropy": 6.020894718170166,
"epoch": 4.886126305320736,
"grad_norm": 1.234375,
"learning_rate": 0.00032236569045493283,
"loss": 5.5618,
"mean_token_accuracy": 0.17821861207485198,
"num_tokens": 9629312.0,
"step": 4915
},
{
"entropy": 6.06228404045105,
"epoch": 4.891098955743411,
"grad_norm": 1.2421875,
"learning_rate": 0.00032198384127751344,
"loss": 5.7663,
"mean_token_accuracy": 0.16749885827302932,
"num_tokens": 9639439.0,
"step": 4920
},
{
"entropy": 6.1211097240448,
"epoch": 4.896071606166086,
"grad_norm": 1.1875,
"learning_rate": 0.000321601850556243,
"loss": 5.577,
"mean_token_accuracy": 0.17663053721189498,
"num_tokens": 9647824.0,
"step": 4925
},
{
"entropy": 6.018051338195801,
"epoch": 4.9010442565887615,
"grad_norm": 1.28125,
"learning_rate": 0.0003212197194419095,
"loss": 5.6134,
"mean_token_accuracy": 0.17575460225343703,
"num_tokens": 9656962.0,
"step": 4930
},
{
"entropy": 5.979993057250977,
"epoch": 4.906016907011437,
"grad_norm": 1.1796875,
"learning_rate": 0.0003208374490857238,
"loss": 5.698,
"mean_token_accuracy": 0.17622910290956498,
"num_tokens": 9666360.0,
"step": 4935
},
{
"entropy": 6.068582439422608,
"epoch": 4.910989557434112,
"grad_norm": 1.1171875,
"learning_rate": 0.0003204550406393164,
"loss": 5.6836,
"mean_token_accuracy": 0.17156701236963273,
"num_tokens": 9676355.0,
"step": 4940
},
{
"entropy": 6.070707035064697,
"epoch": 4.915962207856787,
"grad_norm": 1.296875,
"learning_rate": 0.0003200724952547335,
"loss": 5.6607,
"mean_token_accuracy": 0.1738799810409546,
"num_tokens": 9686361.0,
"step": 4945
},
{
"entropy": 5.980514717102051,
"epoch": 4.9209348582794625,
"grad_norm": 1.0546875,
"learning_rate": 0.0003196898140844341,
"loss": 5.5799,
"mean_token_accuracy": 0.18116140812635423,
"num_tokens": 9696571.0,
"step": 4950
},
{
"entropy": 6.153595018386841,
"epoch": 4.925907508702139,
"grad_norm": 1.21875,
"learning_rate": 0.00031930699828128614,
"loss": 5.7754,
"mean_token_accuracy": 0.1685476392507553,
"num_tokens": 9707595.0,
"step": 4955
},
{
"entropy": 6.052409696578979,
"epoch": 4.930880159124813,
"grad_norm": 1.21875,
"learning_rate": 0.0003189240489985633,
"loss": 5.6612,
"mean_token_accuracy": 0.17375128269195556,
"num_tokens": 9717094.0,
"step": 4960
},
{
"entropy": 5.938237476348877,
"epoch": 4.935852809547489,
"grad_norm": 1.1875,
"learning_rate": 0.0003185409673899412,
"loss": 5.6026,
"mean_token_accuracy": 0.1792594999074936,
"num_tokens": 9727329.0,
"step": 4965
},
{
"entropy": 5.970051383972168,
"epoch": 4.940825459970164,
"grad_norm": 1.25,
"learning_rate": 0.00031815775460949426,
"loss": 5.4714,
"mean_token_accuracy": 0.19122297912836075,
"num_tokens": 9736189.0,
"step": 4970
},
{
"entropy": 6.068285512924194,
"epoch": 4.94579811039284,
"grad_norm": 1.2265625,
"learning_rate": 0.0003177744118116919,
"loss": 5.5869,
"mean_token_accuracy": 0.17852106988430022,
"num_tokens": 9745760.0,
"step": 4975
},
{
"entropy": 6.121452951431275,
"epoch": 4.950770760815515,
"grad_norm": 1.171875,
"learning_rate": 0.0003173909401513954,
"loss": 5.7244,
"mean_token_accuracy": 0.16361072584986686,
"num_tokens": 9755941.0,
"step": 4980
},
{
"entropy": 6.015450668334961,
"epoch": 4.95574341123819,
"grad_norm": 1.25,
"learning_rate": 0.00031700734078385416,
"loss": 5.5903,
"mean_token_accuracy": 0.1792882889509201,
"num_tokens": 9764546.0,
"step": 4985
},
{
"entropy": 6.004361724853515,
"epoch": 4.960716061660865,
"grad_norm": 1.2734375,
"learning_rate": 0.00031662361486470243,
"loss": 5.6437,
"mean_token_accuracy": 0.17401062548160554,
"num_tokens": 9773728.0,
"step": 4990
},
{
"entropy": 5.950984907150269,
"epoch": 4.965688712083541,
"grad_norm": 1.1640625,
"learning_rate": 0.0003162397635499555,
"loss": 5.6174,
"mean_token_accuracy": 0.17360008358955384,
"num_tokens": 9783878.0,
"step": 4995
},
{
"entropy": 6.040719223022461,
"epoch": 4.970661362506216,
"grad_norm": 1.140625,
"learning_rate": 0.00031585578799600653,
"loss": 5.6442,
"mean_token_accuracy": 0.17362129390239717,
"num_tokens": 9793926.0,
"step": 5000
},
{
"epoch": 4.970661362506216,
"eval_entropy": 6.015637426332395,
"eval_loss": 6.523582935333252,
"eval_mean_token_accuracy": 0.14340103832223722,
"eval_num_tokens": 9793926.0,
"eval_runtime": 2.5384,
"eval_samples_per_second": 1365.054,
"eval_steps_per_second": 170.976,
"step": 5000
},
{
"entropy": 6.036915874481201,
"epoch": 4.975634012928891,
"grad_norm": 1.1015625,
"learning_rate": 0.00031547168935962304,
"loss": 5.5738,
"mean_token_accuracy": 0.18685720711946488,
"num_tokens": 9804213.0,
"step": 5005
},
{
"entropy": 5.963804531097412,
"epoch": 4.980606663351566,
"grad_norm": 1.203125,
"learning_rate": 0.0003150874687979433,
"loss": 5.6071,
"mean_token_accuracy": 0.18127841204404832,
"num_tokens": 9814151.0,
"step": 5010
},
{
"entropy": 5.982738256454468,
"epoch": 4.985579313774242,
"grad_norm": 1.2734375,
"learning_rate": 0.00031470312746847284,
"loss": 5.6035,
"mean_token_accuracy": 0.17865185141563417,
"num_tokens": 9822646.0,
"step": 5015
},
{
"entropy": 5.881924486160278,
"epoch": 4.990551964196917,
"grad_norm": 1.140625,
"learning_rate": 0.00031431866652908107,
"loss": 5.369,
"mean_token_accuracy": 0.21032825708389283,
"num_tokens": 9833204.0,
"step": 5020
},
{
"entropy": 5.873883819580078,
"epoch": 4.995524614619592,
"grad_norm": 1.3125,
"learning_rate": 0.0003139340871379977,
"loss": 5.5167,
"mean_token_accuracy": 0.20007498413324357,
"num_tokens": 9843150.0,
"step": 5025
},
{
"entropy": 5.974801434410943,
"epoch": 5.0,
"grad_norm": 2.171875,
"learning_rate": 0.0003135493904538093,
"loss": 5.5878,
"mean_token_accuracy": 0.17905576858255598,
"num_tokens": 9850520.0,
"step": 5030
},
{
"entropy": 6.045516014099121,
"epoch": 5.004972650422675,
"grad_norm": 1.171875,
"learning_rate": 0.0003131645776354559,
"loss": 5.3226,
"mean_token_accuracy": 0.18612450063228608,
"num_tokens": 9860876.0,
"step": 5035
},
{
"entropy": 5.929516458511353,
"epoch": 5.0099453008453505,
"grad_norm": 1.3828125,
"learning_rate": 0.0003127796498422272,
"loss": 5.2215,
"mean_token_accuracy": 0.2104060411453247,
"num_tokens": 9870944.0,
"step": 5040
},
{
"entropy": 5.925146102905273,
"epoch": 5.014917951268026,
"grad_norm": 1.1953125,
"learning_rate": 0.00031239460823375933,
"loss": 5.2832,
"mean_token_accuracy": 0.19128845632076263,
"num_tokens": 9880615.0,
"step": 5045
},
{
"entropy": 5.816671133041382,
"epoch": 5.019890601690701,
"grad_norm": 1.140625,
"learning_rate": 0.00031200945397003157,
"loss": 5.1164,
"mean_token_accuracy": 0.20996299833059312,
"num_tokens": 9889933.0,
"step": 5050
},
{
"entropy": 6.048820495605469,
"epoch": 5.024863252113376,
"grad_norm": 1.1796875,
"learning_rate": 0.0003116241882113623,
"loss": 5.376,
"mean_token_accuracy": 0.18328726142644883,
"num_tokens": 9900166.0,
"step": 5055
},
{
"entropy": 5.900638580322266,
"epoch": 5.0298359025360515,
"grad_norm": 1.109375,
"learning_rate": 0.00031123881211840565,
"loss": 5.2816,
"mean_token_accuracy": 0.1943972259759903,
"num_tokens": 9909747.0,
"step": 5060
},
{
"entropy": 5.944966268539429,
"epoch": 5.034808552958727,
"grad_norm": 1.1953125,
"learning_rate": 0.0003108533268521485,
"loss": 5.3377,
"mean_token_accuracy": 0.19478731453418732,
"num_tokens": 9919039.0,
"step": 5065
},
{
"entropy": 6.0269505977630615,
"epoch": 5.039781203381402,
"grad_norm": 1.25,
"learning_rate": 0.0003104677335739065,
"loss": 5.3512,
"mean_token_accuracy": 0.1993807539343834,
"num_tokens": 9927933.0,
"step": 5070
},
{
"entropy": 5.958450841903686,
"epoch": 5.044753853804077,
"grad_norm": 1.171875,
"learning_rate": 0.00031008203344532067,
"loss": 5.3175,
"mean_token_accuracy": 0.19725027829408645,
"num_tokens": 9939264.0,
"step": 5075
},
{
"entropy": 5.934142780303955,
"epoch": 5.0497265042267525,
"grad_norm": 1.125,
"learning_rate": 0.000309696227628354,
"loss": 5.3567,
"mean_token_accuracy": 0.1859802082180977,
"num_tokens": 9948375.0,
"step": 5080
},
{
"entropy": 5.9629510879516605,
"epoch": 5.054699154649428,
"grad_norm": 1.328125,
"learning_rate": 0.00030931031728528754,
"loss": 5.2344,
"mean_token_accuracy": 0.19987513571977616,
"num_tokens": 9957792.0,
"step": 5085
},
{
"entropy": 5.943561744689942,
"epoch": 5.059671805072103,
"grad_norm": 1.203125,
"learning_rate": 0.0003089243035787178,
"loss": 5.3051,
"mean_token_accuracy": 0.19840159714221955,
"num_tokens": 9967213.0,
"step": 5090
},
{
"entropy": 5.956852817535401,
"epoch": 5.064644455494779,
"grad_norm": 1.1953125,
"learning_rate": 0.00030853818767155225,
"loss": 5.3721,
"mean_token_accuracy": 0.18537731915712358,
"num_tokens": 9976886.0,
"step": 5095
},
{
"entropy": 5.934570980072022,
"epoch": 5.069617105917454,
"grad_norm": 1.203125,
"learning_rate": 0.0003081519707270065,
"loss": 5.2868,
"mean_token_accuracy": 0.19943463057279587,
"num_tokens": 9986702.0,
"step": 5100
},
{
"entropy": 5.973853969573975,
"epoch": 5.07458975634013,
"grad_norm": 1.2109375,
"learning_rate": 0.00030776565390860017,
"loss": 5.2704,
"mean_token_accuracy": 0.20342181771993637,
"num_tokens": 9996597.0,
"step": 5105
},
{
"entropy": 5.955984592437744,
"epoch": 5.079562406762805,
"grad_norm": 1.1796875,
"learning_rate": 0.0003073792383801543,
"loss": 5.3381,
"mean_token_accuracy": 0.19379812479019165,
"num_tokens": 10006440.0,
"step": 5110
},
{
"entropy": 5.959110593795776,
"epoch": 5.08453505718548,
"grad_norm": 1.3203125,
"learning_rate": 0.00030699272530578697,
"loss": 5.3516,
"mean_token_accuracy": 0.19486697912216186,
"num_tokens": 10016198.0,
"step": 5115
},
{
"entropy": 5.999681758880615,
"epoch": 5.089507707608155,
"grad_norm": 1.2734375,
"learning_rate": 0.00030660611584991013,
"loss": 5.3744,
"mean_token_accuracy": 0.18920059353113175,
"num_tokens": 10025795.0,
"step": 5120
},
{
"entropy": 5.9631575584411625,
"epoch": 5.094480358030831,
"grad_norm": 1.09375,
"learning_rate": 0.00030621941117722615,
"loss": 5.2611,
"mean_token_accuracy": 0.2049394354224205,
"num_tokens": 10035448.0,
"step": 5125
},
{
"entropy": 5.925503587722778,
"epoch": 5.099453008453506,
"grad_norm": 1.078125,
"learning_rate": 0.0003058326124527243,
"loss": 5.3247,
"mean_token_accuracy": 0.20077412128448485,
"num_tokens": 10046335.0,
"step": 5130
},
{
"entropy": 6.00296688079834,
"epoch": 5.104425658876181,
"grad_norm": 1.171875,
"learning_rate": 0.0003054457208416771,
"loss": 5.4117,
"mean_token_accuracy": 0.18651205897331238,
"num_tokens": 10056346.0,
"step": 5135
},
{
"entropy": 5.792932558059692,
"epoch": 5.109398309298856,
"grad_norm": 1.3125,
"learning_rate": 0.00030505873750963705,
"loss": 5.1918,
"mean_token_accuracy": 0.21655914485454558,
"num_tokens": 10065948.0,
"step": 5140
},
{
"entropy": 5.867745447158813,
"epoch": 5.114370959721532,
"grad_norm": 1.578125,
"learning_rate": 0.0003046716636224327,
"loss": 5.2612,
"mean_token_accuracy": 0.20049624890089035,
"num_tokens": 10076680.0,
"step": 5145
},
{
"entropy": 5.967780828475952,
"epoch": 5.119343610144207,
"grad_norm": 1.21875,
"learning_rate": 0.0003042845003461657,
"loss": 5.3263,
"mean_token_accuracy": 0.19001180231571196,
"num_tokens": 10086373.0,
"step": 5150
},
{
"entropy": 5.9693605422973635,
"epoch": 5.124316260566882,
"grad_norm": 1.25,
"learning_rate": 0.00030389724884720664,
"loss": 5.3282,
"mean_token_accuracy": 0.19168860614299774,
"num_tokens": 10095415.0,
"step": 5155
},
{
"entropy": 5.7654900550842285,
"epoch": 5.129288910989557,
"grad_norm": 1.2578125,
"learning_rate": 0.00030350991029219236,
"loss": 5.1714,
"mean_token_accuracy": 0.21793798357248306,
"num_tokens": 10104709.0,
"step": 5160
},
{
"entropy": 5.966084718704224,
"epoch": 5.134261561412233,
"grad_norm": 1.4375,
"learning_rate": 0.00030312248584802166,
"loss": 5.3335,
"mean_token_accuracy": 0.1935370609164238,
"num_tokens": 10113539.0,
"step": 5165
},
{
"entropy": 5.829127788543701,
"epoch": 5.139234211834908,
"grad_norm": 1.25,
"learning_rate": 0.0003027349766818521,
"loss": 5.2132,
"mean_token_accuracy": 0.21537336558103562,
"num_tokens": 10123914.0,
"step": 5170
},
{
"entropy": 6.0061119556427,
"epoch": 5.144206862257583,
"grad_norm": 1.2421875,
"learning_rate": 0.0003023473839610965,
"loss": 5.3932,
"mean_token_accuracy": 0.18911180794239044,
"num_tokens": 10133572.0,
"step": 5175
},
{
"entropy": 5.99053635597229,
"epoch": 5.149179512680258,
"grad_norm": 1.1875,
"learning_rate": 0.0003019597088534195,
"loss": 5.4313,
"mean_token_accuracy": 0.186905138194561,
"num_tokens": 10142973.0,
"step": 5180
},
{
"entropy": 5.972963905334472,
"epoch": 5.154152163102934,
"grad_norm": 1.234375,
"learning_rate": 0.0003015719525267338,
"loss": 5.421,
"mean_token_accuracy": 0.18932133615016938,
"num_tokens": 10153322.0,
"step": 5185
},
{
"entropy": 5.9401411533355715,
"epoch": 5.159124813525609,
"grad_norm": 1.3046875,
"learning_rate": 0.0003011841161491969,
"loss": 5.3119,
"mean_token_accuracy": 0.1984267145395279,
"num_tokens": 10163176.0,
"step": 5190
},
{
"entropy": 5.954602956771851,
"epoch": 5.164097463948284,
"grad_norm": 1.21875,
"learning_rate": 0.00030079620088920727,
"loss": 5.3211,
"mean_token_accuracy": 0.19663264602422714,
"num_tokens": 10172616.0,
"step": 5195
},
{
"entropy": 5.929637622833252,
"epoch": 5.169070114370959,
"grad_norm": 1.3046875,
"learning_rate": 0.0003004082079154013,
"loss": 5.3643,
"mean_token_accuracy": 0.18637740463018418,
"num_tokens": 10182243.0,
"step": 5200
},
{
"entropy": 5.981790351867676,
"epoch": 5.174042764793635,
"grad_norm": 1.2109375,
"learning_rate": 0.00030002013839664925,
"loss": 5.3253,
"mean_token_accuracy": 0.1938107892870903,
"num_tokens": 10191394.0,
"step": 5205
},
{
"entropy": 5.787403774261475,
"epoch": 5.17901541521631,
"grad_norm": 1.1484375,
"learning_rate": 0.0002996319935020522,
"loss": 5.1849,
"mean_token_accuracy": 0.21427320688962936,
"num_tokens": 10202167.0,
"step": 5210
},
{
"entropy": 5.928043699264526,
"epoch": 5.183988065638985,
"grad_norm": 1.25,
"learning_rate": 0.0002992437744009381,
"loss": 5.237,
"mean_token_accuracy": 0.2006138324737549,
"num_tokens": 10211250.0,
"step": 5215
},
{
"entropy": 5.99057412147522,
"epoch": 5.188960716061661,
"grad_norm": 1.2265625,
"learning_rate": 0.00029885548226285855,
"loss": 5.415,
"mean_token_accuracy": 0.18949828296899796,
"num_tokens": 10220722.0,
"step": 5220
},
{
"entropy": 6.0179078578948975,
"epoch": 5.193933366484337,
"grad_norm": 1.25,
"learning_rate": 0.0002984671182575851,
"loss": 5.4077,
"mean_token_accuracy": 0.18369015604257583,
"num_tokens": 10229880.0,
"step": 5225
},
{
"entropy": 5.945045137405396,
"epoch": 5.198906016907012,
"grad_norm": 1.28125,
"learning_rate": 0.0002980786835551061,
"loss": 5.3871,
"mean_token_accuracy": 0.1917655274271965,
"num_tokens": 10240939.0,
"step": 5230
},
{
"entropy": 5.975225400924683,
"epoch": 5.203878667329687,
"grad_norm": 1.2734375,
"learning_rate": 0.0002976901793256225,
"loss": 5.3461,
"mean_token_accuracy": 0.19081468135118484,
"num_tokens": 10249383.0,
"step": 5235
},
{
"entropy": 6.0411388874053955,
"epoch": 5.208851317752362,
"grad_norm": 1.34375,
"learning_rate": 0.000297301606739545,
"loss": 5.4094,
"mean_token_accuracy": 0.19023002982139586,
"num_tokens": 10259166.0,
"step": 5240
},
{
"entropy": 5.998856163024902,
"epoch": 5.213823968175038,
"grad_norm": 1.2578125,
"learning_rate": 0.00029691296696748994,
"loss": 5.4681,
"mean_token_accuracy": 0.18406993746757508,
"num_tokens": 10269053.0,
"step": 5245
},
{
"entropy": 5.921797037124634,
"epoch": 5.218796618597713,
"grad_norm": 1.2265625,
"learning_rate": 0.00029652426118027635,
"loss": 5.347,
"mean_token_accuracy": 0.1976258561015129,
"num_tokens": 10279148.0,
"step": 5250
},
{
"entropy": 5.947183418273926,
"epoch": 5.223769269020388,
"grad_norm": 1.15625,
"learning_rate": 0.00029613549054892206,
"loss": 5.2995,
"mean_token_accuracy": 0.1910171702504158,
"num_tokens": 10288719.0,
"step": 5255
},
{
"entropy": 5.9178423404693605,
"epoch": 5.228741919443063,
"grad_norm": 1.1484375,
"learning_rate": 0.0002957466562446401,
"loss": 5.3081,
"mean_token_accuracy": 0.1962686687707901,
"num_tokens": 10300089.0,
"step": 5260
},
{
"entropy": 5.948833322525024,
"epoch": 5.233714569865739,
"grad_norm": 1.15625,
"learning_rate": 0.0002953577594388355,
"loss": 5.336,
"mean_token_accuracy": 0.19297122657299043,
"num_tokens": 10309400.0,
"step": 5265
},
{
"entropy": 5.966468811035156,
"epoch": 5.238687220288414,
"grad_norm": 1.1484375,
"learning_rate": 0.00029496880130310145,
"loss": 5.3664,
"mean_token_accuracy": 0.19562508463859557,
"num_tokens": 10319115.0,
"step": 5270
},
{
"entropy": 5.957953262329101,
"epoch": 5.243659870711089,
"grad_norm": 1.375,
"learning_rate": 0.0002945797830092161,
"loss": 5.2729,
"mean_token_accuracy": 0.195231157541275,
"num_tokens": 10327174.0,
"step": 5275
},
{
"entropy": 5.983389854431152,
"epoch": 5.248632521133764,
"grad_norm": 1.3828125,
"learning_rate": 0.0002941907057291386,
"loss": 5.3506,
"mean_token_accuracy": 0.19135708808898927,
"num_tokens": 10335608.0,
"step": 5280
},
{
"entropy": 5.857269334793091,
"epoch": 5.25360517155644,
"grad_norm": 1.2734375,
"learning_rate": 0.00029380157063500594,
"loss": 5.2711,
"mean_token_accuracy": 0.19114345014095308,
"num_tokens": 10345592.0,
"step": 5285
},
{
"entropy": 5.88392744064331,
"epoch": 5.258577821979115,
"grad_norm": 1.2109375,
"learning_rate": 0.0002934123788991291,
"loss": 5.2916,
"mean_token_accuracy": 0.2016509309411049,
"num_tokens": 10355913.0,
"step": 5290
},
{
"entropy": 5.968164396286011,
"epoch": 5.26355047240179,
"grad_norm": 1.203125,
"learning_rate": 0.00029302313169399,
"loss": 5.3711,
"mean_token_accuracy": 0.1905299261212349,
"num_tokens": 10365419.0,
"step": 5295
},
{
"entropy": 5.938971853256225,
"epoch": 5.268523122824465,
"grad_norm": 1.3046875,
"learning_rate": 0.0002926338301922376,
"loss": 5.3815,
"mean_token_accuracy": 0.18325096219778061,
"num_tokens": 10373939.0,
"step": 5300
},
{
"entropy": 5.979659605026245,
"epoch": 5.273495773247141,
"grad_norm": 1.171875,
"learning_rate": 0.00029224447556668417,
"loss": 5.4103,
"mean_token_accuracy": 0.18754840046167373,
"num_tokens": 10383932.0,
"step": 5305
},
{
"entropy": 5.95255331993103,
"epoch": 5.278468423669816,
"grad_norm": 1.203125,
"learning_rate": 0.0002918550689903022,
"loss": 5.3296,
"mean_token_accuracy": 0.1887911632657051,
"num_tokens": 10393319.0,
"step": 5310
},
{
"entropy": 5.925978517532348,
"epoch": 5.283441074092491,
"grad_norm": 1.1484375,
"learning_rate": 0.00029146561163622076,
"loss": 5.4698,
"mean_token_accuracy": 0.17871202677488326,
"num_tokens": 10404102.0,
"step": 5315
},
{
"entropy": 5.922774839401245,
"epoch": 5.288413724515166,
"grad_norm": 1.1875,
"learning_rate": 0.0002910761046777218,
"loss": 5.3418,
"mean_token_accuracy": 0.19539261013269424,
"num_tokens": 10412718.0,
"step": 5320
},
{
"entropy": 5.987729549407959,
"epoch": 5.293386374937842,
"grad_norm": 1.4296875,
"learning_rate": 0.00029068654928823666,
"loss": 5.4297,
"mean_token_accuracy": 0.18810867369174958,
"num_tokens": 10422546.0,
"step": 5325
},
{
"entropy": 5.9228565216064455,
"epoch": 5.298359025360517,
"grad_norm": 1.2734375,
"learning_rate": 0.00029029694664134284,
"loss": 5.3883,
"mean_token_accuracy": 0.18880436867475509,
"num_tokens": 10431498.0,
"step": 5330
},
{
"entropy": 5.975613260269165,
"epoch": 5.303331675783192,
"grad_norm": 1.2578125,
"learning_rate": 0.0002899072979107597,
"loss": 5.4142,
"mean_token_accuracy": 0.19154437482357026,
"num_tokens": 10441713.0,
"step": 5335
},
{
"entropy": 6.031552648544311,
"epoch": 5.308304326205867,
"grad_norm": 1.125,
"learning_rate": 0.0002895176042703461,
"loss": 5.404,
"mean_token_accuracy": 0.18690674602985383,
"num_tokens": 10451570.0,
"step": 5340
},
{
"entropy": 5.966177225112915,
"epoch": 5.3132769766285435,
"grad_norm": 1.3046875,
"learning_rate": 0.00028912786689409554,
"loss": 5.3876,
"mean_token_accuracy": 0.19130735993385314,
"num_tokens": 10461442.0,
"step": 5345
},
{
"entropy": 6.022460794448852,
"epoch": 5.318249627051218,
"grad_norm": 1.375,
"learning_rate": 0.0002887380869561338,
"loss": 5.4589,
"mean_token_accuracy": 0.184146685898304,
"num_tokens": 10471223.0,
"step": 5350
},
{
"entropy": 5.940879201889038,
"epoch": 5.323222277473894,
"grad_norm": 1.1875,
"learning_rate": 0.0002883482656307145,
"loss": 5.3154,
"mean_token_accuracy": 0.18742220103740692,
"num_tokens": 10480469.0,
"step": 5355
},
{
"entropy": 5.912155199050903,
"epoch": 5.328194927896569,
"grad_norm": 1.125,
"learning_rate": 0.0002879584040922162,
"loss": 5.3634,
"mean_token_accuracy": 0.19065987914800644,
"num_tokens": 10490320.0,
"step": 5360
},
{
"entropy": 5.923424148559571,
"epoch": 5.3331675783192445,
"grad_norm": 1.25,
"learning_rate": 0.0002875685035151386,
"loss": 5.3929,
"mean_token_accuracy": 0.18426522761583328,
"num_tokens": 10500471.0,
"step": 5365
},
{
"entropy": 6.039834594726562,
"epoch": 5.33814022874192,
"grad_norm": 1.1015625,
"learning_rate": 0.00028717856507409886,
"loss": 5.4767,
"mean_token_accuracy": 0.18141003996133803,
"num_tokens": 10510956.0,
"step": 5370
},
{
"entropy": 6.056643438339234,
"epoch": 5.343112879164595,
"grad_norm": 1.375,
"learning_rate": 0.0002867885899438281,
"loss": 5.4614,
"mean_token_accuracy": 0.1897230327129364,
"num_tokens": 10520838.0,
"step": 5375
},
{
"entropy": 6.037986993789673,
"epoch": 5.34808552958727,
"grad_norm": 1.1328125,
"learning_rate": 0.0002863985792991683,
"loss": 5.4043,
"mean_token_accuracy": 0.18920813649892806,
"num_tokens": 10530879.0,
"step": 5380
},
{
"entropy": 5.799209690093994,
"epoch": 5.3530581800099455,
"grad_norm": 1.234375,
"learning_rate": 0.0002860085343150683,
"loss": 5.1917,
"mean_token_accuracy": 0.21452010422945023,
"num_tokens": 10540161.0,
"step": 5385
},
{
"entropy": 5.925060844421386,
"epoch": 5.358030830432621,
"grad_norm": 1.28125,
"learning_rate": 0.00028561845616658033,
"loss": 5.4072,
"mean_token_accuracy": 0.18773865103721618,
"num_tokens": 10551173.0,
"step": 5390
},
{
"entropy": 5.957026720046997,
"epoch": 5.363003480855296,
"grad_norm": 1.234375,
"learning_rate": 0.00028522834602885656,
"loss": 5.3535,
"mean_token_accuracy": 0.18755310624837876,
"num_tokens": 10560882.0,
"step": 5395
},
{
"entropy": 6.040479278564453,
"epoch": 5.367976131277971,
"grad_norm": 1.1796875,
"learning_rate": 0.0002848382050771455,
"loss": 5.494,
"mean_token_accuracy": 0.18573542386293412,
"num_tokens": 10570837.0,
"step": 5400
},
{
"entropy": 5.985017967224121,
"epoch": 5.3729487817006465,
"grad_norm": 1.2109375,
"learning_rate": 0.0002844480344867886,
"loss": 5.3719,
"mean_token_accuracy": 0.19307992309331895,
"num_tokens": 10580335.0,
"step": 5405
},
{
"entropy": 5.935797309875488,
"epoch": 5.377921432123322,
"grad_norm": 1.375,
"learning_rate": 0.0002840578354332165,
"loss": 5.3748,
"mean_token_accuracy": 0.18907978385686874,
"num_tokens": 10589030.0,
"step": 5410
},
{
"entropy": 5.915637016296387,
"epoch": 5.382894082545997,
"grad_norm": 1.2109375,
"learning_rate": 0.0002836676090919456,
"loss": 5.3649,
"mean_token_accuracy": 0.1891477644443512,
"num_tokens": 10599309.0,
"step": 5415
},
{
"entropy": 5.934687328338623,
"epoch": 5.387866732968672,
"grad_norm": 1.2890625,
"learning_rate": 0.0002832773566385746,
"loss": 5.3222,
"mean_token_accuracy": 0.19777963161468506,
"num_tokens": 10608781.0,
"step": 5420
},
{
"entropy": 5.919578647613525,
"epoch": 5.3928393833913475,
"grad_norm": 1.109375,
"learning_rate": 0.00028288707924878056,
"loss": 5.2647,
"mean_token_accuracy": 0.1969473734498024,
"num_tokens": 10618512.0,
"step": 5425
},
{
"entropy": 5.958898878097534,
"epoch": 5.397812033814023,
"grad_norm": 1.3203125,
"learning_rate": 0.0002824967780983162,
"loss": 5.3917,
"mean_token_accuracy": 0.1890107199549675,
"num_tokens": 10629420.0,
"step": 5430
},
{
"entropy": 5.970873165130615,
"epoch": 5.402784684236698,
"grad_norm": 1.265625,
"learning_rate": 0.00028210645436300524,
"loss": 5.3916,
"mean_token_accuracy": 0.19095586389303207,
"num_tokens": 10639625.0,
"step": 5435
},
{
"entropy": 5.906043434143067,
"epoch": 5.407757334659373,
"grad_norm": 1.25,
"learning_rate": 0.00028171610921873995,
"loss": 5.2869,
"mean_token_accuracy": 0.18862376511096954,
"num_tokens": 10648633.0,
"step": 5440
},
{
"entropy": 5.799239778518677,
"epoch": 5.4127299850820485,
"grad_norm": 1.1875,
"learning_rate": 0.0002813257438414766,
"loss": 5.2112,
"mean_token_accuracy": 0.20348195284605025,
"num_tokens": 10657981.0,
"step": 5445
},
{
"entropy": 5.847122478485107,
"epoch": 5.417702635504724,
"grad_norm": 1.078125,
"learning_rate": 0.000280935359407233,
"loss": 5.2228,
"mean_token_accuracy": 0.20272496938705445,
"num_tokens": 10669734.0,
"step": 5450
},
{
"entropy": 5.895791053771973,
"epoch": 5.422675285927399,
"grad_norm": 1.140625,
"learning_rate": 0.0002805449570920838,
"loss": 5.3394,
"mean_token_accuracy": 0.18766908794641496,
"num_tokens": 10679486.0,
"step": 5455
},
{
"entropy": 5.942715787887574,
"epoch": 5.427647936350074,
"grad_norm": 1.1796875,
"learning_rate": 0.00028015453807215786,
"loss": 5.388,
"mean_token_accuracy": 0.18866899311542512,
"num_tokens": 10689551.0,
"step": 5460
},
{
"entropy": 5.95545129776001,
"epoch": 5.4326205867727495,
"grad_norm": 1.203125,
"learning_rate": 0.00027976410352363416,
"loss": 5.3393,
"mean_token_accuracy": 0.1906949758529663,
"num_tokens": 10699173.0,
"step": 5465
},
{
"entropy": 5.885512590408325,
"epoch": 5.437593237195425,
"grad_norm": 1.3125,
"learning_rate": 0.00027937365462273864,
"loss": 5.289,
"mean_token_accuracy": 0.2001232385635376,
"num_tokens": 10709050.0,
"step": 5470
},
{
"entropy": 5.934259843826294,
"epoch": 5.4425658876181,
"grad_norm": 1.2265625,
"learning_rate": 0.00027898319254574043,
"loss": 5.3921,
"mean_token_accuracy": 0.19291134178638458,
"num_tokens": 10718397.0,
"step": 5475
},
{
"entropy": 5.957664155960083,
"epoch": 5.447538538040776,
"grad_norm": 1.0546875,
"learning_rate": 0.0002785927184689483,
"loss": 5.4365,
"mean_token_accuracy": 0.18743063509464264,
"num_tokens": 10729305.0,
"step": 5480
},
{
"entropy": 5.9333860874176025,
"epoch": 5.452511188463451,
"grad_norm": 1.328125,
"learning_rate": 0.0002782022335687073,
"loss": 5.3589,
"mean_token_accuracy": 0.19065942615270615,
"num_tokens": 10738602.0,
"step": 5485
},
{
"entropy": 5.910775661468506,
"epoch": 5.457483838886127,
"grad_norm": 1.1171875,
"learning_rate": 0.00027781173902139485,
"loss": 5.3529,
"mean_token_accuracy": 0.2008916661143303,
"num_tokens": 10748373.0,
"step": 5490
},
{
"entropy": 5.975349569320679,
"epoch": 5.462456489308802,
"grad_norm": 1.2734375,
"learning_rate": 0.0002774212360034176,
"loss": 5.3833,
"mean_token_accuracy": 0.18734287917613984,
"num_tokens": 10756879.0,
"step": 5495
},
{
"entropy": 5.872376585006714,
"epoch": 5.467429139731477,
"grad_norm": 1.2421875,
"learning_rate": 0.0002770307256912078,
"loss": 5.211,
"mean_token_accuracy": 0.2059996396303177,
"num_tokens": 10766969.0,
"step": 5500
},
{
"epoch": 5.467429139731477,
"eval_entropy": 5.857917273648873,
"eval_loss": 6.5340704917907715,
"eval_mean_token_accuracy": 0.1441406408316254,
"eval_num_tokens": 10766969.0,
"eval_runtime": 2.5505,
"eval_samples_per_second": 1358.54,
"eval_steps_per_second": 170.161,
"step": 5500
},
{
"entropy": 5.970714521408081,
"epoch": 5.472401790154152,
"grad_norm": 1.203125,
"learning_rate": 0.00027664020926121944,
"loss": 5.4191,
"mean_token_accuracy": 0.1908913403749466,
"num_tokens": 10776780.0,
"step": 5505
},
{
"entropy": 5.96117262840271,
"epoch": 5.477374440576828,
"grad_norm": 1.2421875,
"learning_rate": 0.0002762496878899251,
"loss": 5.4003,
"mean_token_accuracy": 0.18594695031642913,
"num_tokens": 10786341.0,
"step": 5510
},
{
"entropy": 5.942570209503174,
"epoch": 5.482347090999503,
"grad_norm": 1.328125,
"learning_rate": 0.00027585916275381226,
"loss": 5.2956,
"mean_token_accuracy": 0.20083372741937638,
"num_tokens": 10795754.0,
"step": 5515
},
{
"entropy": 5.942461681365967,
"epoch": 5.487319741422178,
"grad_norm": 1.2109375,
"learning_rate": 0.0002754686350293797,
"loss": 5.4652,
"mean_token_accuracy": 0.18495413810014724,
"num_tokens": 10805796.0,
"step": 5520
},
{
"entropy": 5.854644775390625,
"epoch": 5.492292391844853,
"grad_norm": 1.2890625,
"learning_rate": 0.0002750781058931339,
"loss": 5.2169,
"mean_token_accuracy": 0.20821022093296052,
"num_tokens": 10815427.0,
"step": 5525
},
{
"entropy": 5.937387609481812,
"epoch": 5.497265042267529,
"grad_norm": 1.21875,
"learning_rate": 0.0002746875765215857,
"loss": 5.3401,
"mean_token_accuracy": 0.1959060788154602,
"num_tokens": 10825501.0,
"step": 5530
},
{
"entropy": 5.991639566421509,
"epoch": 5.502237692690204,
"grad_norm": 1.15625,
"learning_rate": 0.00027429704809124665,
"loss": 5.4708,
"mean_token_accuracy": 0.18355388343334197,
"num_tokens": 10834961.0,
"step": 5535
},
{
"entropy": 5.975069379806518,
"epoch": 5.507210343112879,
"grad_norm": 1.2109375,
"learning_rate": 0.0002739065217786255,
"loss": 5.3622,
"mean_token_accuracy": 0.18821591138839722,
"num_tokens": 10845491.0,
"step": 5540
},
{
"entropy": 5.9327311515808105,
"epoch": 5.512182993535554,
"grad_norm": 1.3359375,
"learning_rate": 0.0002735159987602246,
"loss": 5.4261,
"mean_token_accuracy": 0.19062394499778748,
"num_tokens": 10855045.0,
"step": 5545
},
{
"entropy": 5.95606689453125,
"epoch": 5.51715564395823,
"grad_norm": 1.21875,
"learning_rate": 0.0002731254802125362,
"loss": 5.4738,
"mean_token_accuracy": 0.17880992740392684,
"num_tokens": 10864254.0,
"step": 5550
},
{
"entropy": 5.986497116088867,
"epoch": 5.522128294380905,
"grad_norm": 1.296875,
"learning_rate": 0.00027273496731203945,
"loss": 5.3586,
"mean_token_accuracy": 0.19596585035324096,
"num_tokens": 10874768.0,
"step": 5555
},
{
"entropy": 5.914543294906617,
"epoch": 5.52710094480358,
"grad_norm": 1.21875,
"learning_rate": 0.000272344461235196,
"loss": 5.3365,
"mean_token_accuracy": 0.199964939057827,
"num_tokens": 10884820.0,
"step": 5560
},
{
"entropy": 5.804251718521118,
"epoch": 5.532073595226255,
"grad_norm": 1.140625,
"learning_rate": 0.00027195396315844733,
"loss": 5.2544,
"mean_token_accuracy": 0.21133985966444016,
"num_tokens": 10895393.0,
"step": 5565
},
{
"entropy": 5.911523723602295,
"epoch": 5.537046245648931,
"grad_norm": 1.2265625,
"learning_rate": 0.0002715634742582107,
"loss": 5.2806,
"mean_token_accuracy": 0.20161092281341553,
"num_tokens": 10903875.0,
"step": 5570
},
{
"entropy": 6.01757116317749,
"epoch": 5.542018896071606,
"grad_norm": 1.3046875,
"learning_rate": 0.00027117299571087576,
"loss": 5.4343,
"mean_token_accuracy": 0.1855640545487404,
"num_tokens": 10913861.0,
"step": 5575
},
{
"entropy": 5.9324338912963865,
"epoch": 5.546991546494281,
"grad_norm": 1.2734375,
"learning_rate": 0.0002707825286928008,
"loss": 5.403,
"mean_token_accuracy": 0.18071481585502625,
"num_tokens": 10923597.0,
"step": 5580
},
{
"entropy": 5.936261749267578,
"epoch": 5.551964196916956,
"grad_norm": 1.234375,
"learning_rate": 0.0002703920743803097,
"loss": 5.4081,
"mean_token_accuracy": 0.17829694151878356,
"num_tokens": 10934094.0,
"step": 5585
},
{
"entropy": 6.051852226257324,
"epoch": 5.556936847339632,
"grad_norm": 1.3125,
"learning_rate": 0.00027000163394968763,
"loss": 5.4941,
"mean_token_accuracy": 0.1794436290860176,
"num_tokens": 10944121.0,
"step": 5590
},
{
"entropy": 5.88031497001648,
"epoch": 5.561909497762308,
"grad_norm": 1.2421875,
"learning_rate": 0.00026961120857717843,
"loss": 5.2828,
"mean_token_accuracy": 0.20352548360824585,
"num_tokens": 10955117.0,
"step": 5595
},
{
"entropy": 5.819962835311889,
"epoch": 5.566882148184982,
"grad_norm": 1.1640625,
"learning_rate": 0.00026922079943898016,
"loss": 5.2613,
"mean_token_accuracy": 0.2097122400999069,
"num_tokens": 10965248.0,
"step": 5600
},
{
"entropy": 5.994699859619141,
"epoch": 5.571854798607658,
"grad_norm": 1.28125,
"learning_rate": 0.0002688304077112421,
"loss": 5.4648,
"mean_token_accuracy": 0.18445023000240326,
"num_tokens": 10975610.0,
"step": 5605
},
{
"entropy": 5.930114984512329,
"epoch": 5.576827449030333,
"grad_norm": 1.265625,
"learning_rate": 0.0002684400345700613,
"loss": 5.3563,
"mean_token_accuracy": 0.1973711684346199,
"num_tokens": 10984421.0,
"step": 5610
},
{
"entropy": 5.8082794666290285,
"epoch": 5.581800099453009,
"grad_norm": 1.4609375,
"learning_rate": 0.0002680496811914786,
"loss": 5.1949,
"mean_token_accuracy": 0.2062540367245674,
"num_tokens": 10993740.0,
"step": 5615
},
{
"entropy": 5.8984133243560795,
"epoch": 5.586772749875684,
"grad_norm": 1.28125,
"learning_rate": 0.00026765934875147513,
"loss": 5.2752,
"mean_token_accuracy": 0.20063326358795167,
"num_tokens": 11003382.0,
"step": 5620
},
{
"entropy": 5.886960744857788,
"epoch": 5.591745400298359,
"grad_norm": 1.2578125,
"learning_rate": 0.0002672690384259693,
"loss": 5.306,
"mean_token_accuracy": 0.2048925817012787,
"num_tokens": 11013336.0,
"step": 5625
},
{
"entropy": 5.93688383102417,
"epoch": 5.596718050721035,
"grad_norm": 1.1875,
"learning_rate": 0.00026687875139081287,
"loss": 5.3872,
"mean_token_accuracy": 0.1872081145644188,
"num_tokens": 11023800.0,
"step": 5630
},
{
"entropy": 6.056219959259034,
"epoch": 5.60169070114371,
"grad_norm": 1.2578125,
"learning_rate": 0.00026648848882178703,
"loss": 5.5487,
"mean_token_accuracy": 0.18167953342199325,
"num_tokens": 11033250.0,
"step": 5635
},
{
"entropy": 5.909320878982544,
"epoch": 5.606663351566385,
"grad_norm": 1.265625,
"learning_rate": 0.0002660982518945998,
"loss": 5.3434,
"mean_token_accuracy": 0.19531793147325516,
"num_tokens": 11042045.0,
"step": 5640
},
{
"entropy": 5.842851686477661,
"epoch": 5.61163600198906,
"grad_norm": 1.1953125,
"learning_rate": 0.00026570804178488146,
"loss": 5.2825,
"mean_token_accuracy": 0.20330925583839415,
"num_tokens": 11051349.0,
"step": 5645
},
{
"entropy": 5.904132938385009,
"epoch": 5.616608652411736,
"grad_norm": 1.2265625,
"learning_rate": 0.0002653178596681819,
"loss": 5.3289,
"mean_token_accuracy": 0.1930202826857567,
"num_tokens": 11061550.0,
"step": 5650
},
{
"entropy": 5.904772710800171,
"epoch": 5.621581302834411,
"grad_norm": 1.203125,
"learning_rate": 0.0002649277067199665,
"loss": 5.3242,
"mean_token_accuracy": 0.19042237848043442,
"num_tokens": 11071887.0,
"step": 5655
},
{
"entropy": 5.842733669281006,
"epoch": 5.626553953257086,
"grad_norm": 1.5546875,
"learning_rate": 0.0002645375841156127,
"loss": 5.2742,
"mean_token_accuracy": 0.20058438777923585,
"num_tokens": 11081276.0,
"step": 5660
},
{
"entropy": 5.957002401351929,
"epoch": 5.631526603679761,
"grad_norm": 1.1015625,
"learning_rate": 0.0002641474930304067,
"loss": 5.45,
"mean_token_accuracy": 0.18561154454946518,
"num_tokens": 11091761.0,
"step": 5665
},
{
"entropy": 5.816200399398804,
"epoch": 5.636499254102437,
"grad_norm": 1.2890625,
"learning_rate": 0.0002637574346395396,
"loss": 5.1583,
"mean_token_accuracy": 0.2060530036687851,
"num_tokens": 11101986.0,
"step": 5670
},
{
"entropy": 5.991134262084961,
"epoch": 5.641471904525112,
"grad_norm": 1.2421875,
"learning_rate": 0.000263367410118104,
"loss": 5.3733,
"mean_token_accuracy": 0.1980546623468399,
"num_tokens": 11111835.0,
"step": 5675
},
{
"entropy": 5.878189039230347,
"epoch": 5.646444554947787,
"grad_norm": 1.1796875,
"learning_rate": 0.0002629774206410906,
"loss": 5.3194,
"mean_token_accuracy": 0.1970355063676834,
"num_tokens": 11122646.0,
"step": 5680
},
{
"entropy": 5.982857942581177,
"epoch": 5.651417205370462,
"grad_norm": 1.15625,
"learning_rate": 0.0002625874673833844,
"loss": 5.386,
"mean_token_accuracy": 0.1905296891927719,
"num_tokens": 11132693.0,
"step": 5685
},
{
"entropy": 5.958807897567749,
"epoch": 5.656389855793138,
"grad_norm": 1.15625,
"learning_rate": 0.0002621975515197613,
"loss": 5.4341,
"mean_token_accuracy": 0.18375421613454818,
"num_tokens": 11142285.0,
"step": 5690
},
{
"entropy": 5.919147300720215,
"epoch": 5.661362506215813,
"grad_norm": 1.34375,
"learning_rate": 0.00026180767422488466,
"loss": 5.3451,
"mean_token_accuracy": 0.19884450882673263,
"num_tokens": 11150657.0,
"step": 5695
},
{
"entropy": 5.921294641494751,
"epoch": 5.666335156638488,
"grad_norm": 1.2421875,
"learning_rate": 0.00026141783667330156,
"loss": 5.369,
"mean_token_accuracy": 0.19322659075260162,
"num_tokens": 11160391.0,
"step": 5700
},
{
"entropy": 5.983507251739502,
"epoch": 5.671307807061163,
"grad_norm": 1.3984375,
"learning_rate": 0.0002610280400394393,
"loss": 5.5248,
"mean_token_accuracy": 0.18213240057229996,
"num_tokens": 11169919.0,
"step": 5705
},
{
"entropy": 5.9768829345703125,
"epoch": 5.676280457483839,
"grad_norm": 1.2265625,
"learning_rate": 0.00026063828549760195,
"loss": 5.3977,
"mean_token_accuracy": 0.187172931432724,
"num_tokens": 11179948.0,
"step": 5710
},
{
"entropy": 5.85847110748291,
"epoch": 5.681253107906514,
"grad_norm": 1.2421875,
"learning_rate": 0.0002602485742219667,
"loss": 5.1674,
"mean_token_accuracy": 0.21123816967010497,
"num_tokens": 11190875.0,
"step": 5715
},
{
"entropy": 5.772275400161743,
"epoch": 5.686225758329189,
"grad_norm": 1.140625,
"learning_rate": 0.00025985890738658077,
"loss": 5.1923,
"mean_token_accuracy": 0.21412378698587417,
"num_tokens": 11201134.0,
"step": 5720
},
{
"entropy": 5.8833404064178465,
"epoch": 5.691198408751864,
"grad_norm": 1.3046875,
"learning_rate": 0.000259469286165357,
"loss": 5.2869,
"mean_token_accuracy": 0.2004453271627426,
"num_tokens": 11209656.0,
"step": 5725
},
{
"entropy": 5.859318542480469,
"epoch": 5.6961710591745405,
"grad_norm": 1.296875,
"learning_rate": 0.00025907971173207107,
"loss": 5.3322,
"mean_token_accuracy": 0.19251394271850586,
"num_tokens": 11218817.0,
"step": 5730
},
{
"entropy": 5.984533739089966,
"epoch": 5.701143709597215,
"grad_norm": 1.5,
"learning_rate": 0.0002586901852603576,
"loss": 5.4209,
"mean_token_accuracy": 0.1859005570411682,
"num_tokens": 11229291.0,
"step": 5735
},
{
"entropy": 5.940167427062988,
"epoch": 5.706116360019891,
"grad_norm": 1.2578125,
"learning_rate": 0.0002583007079237069,
"loss": 5.4249,
"mean_token_accuracy": 0.19161445051431655,
"num_tokens": 11238877.0,
"step": 5740
},
{
"entropy": 5.928640127182007,
"epoch": 5.711089010442566,
"grad_norm": 1.2578125,
"learning_rate": 0.00025791128089546107,
"loss": 5.4114,
"mean_token_accuracy": 0.19007478952407836,
"num_tokens": 11248885.0,
"step": 5745
},
{
"entropy": 5.957124948501587,
"epoch": 5.7160616608652415,
"grad_norm": 1.1875,
"learning_rate": 0.00025752190534881065,
"loss": 5.3838,
"mean_token_accuracy": 0.19035476744174956,
"num_tokens": 11259423.0,
"step": 5750
},
{
"entropy": 5.9886010646820065,
"epoch": 5.721034311287917,
"grad_norm": 1.28125,
"learning_rate": 0.00025713258245679127,
"loss": 5.4242,
"mean_token_accuracy": 0.18643505722284318,
"num_tokens": 11268900.0,
"step": 5755
},
{
"entropy": 5.8866840362548825,
"epoch": 5.726006961710592,
"grad_norm": 1.2421875,
"learning_rate": 0.00025674331339227966,
"loss": 5.2543,
"mean_token_accuracy": 0.20254520922899247,
"num_tokens": 11277997.0,
"step": 5760
},
{
"entropy": 5.9968218326568605,
"epoch": 5.730979612133267,
"grad_norm": 1.3359375,
"learning_rate": 0.00025635409932799074,
"loss": 5.5302,
"mean_token_accuracy": 0.18124373853206635,
"num_tokens": 11288788.0,
"step": 5765
},
{
"entropy": 5.861563873291016,
"epoch": 5.7359522625559425,
"grad_norm": 1.265625,
"learning_rate": 0.0002559649414364735,
"loss": 5.2764,
"mean_token_accuracy": 0.2064255431294441,
"num_tokens": 11297785.0,
"step": 5770
},
{
"entropy": 5.835377836227417,
"epoch": 5.740924912978618,
"grad_norm": 1.296875,
"learning_rate": 0.0002555758408901076,
"loss": 5.3145,
"mean_token_accuracy": 0.19949227571487427,
"num_tokens": 11307851.0,
"step": 5775
},
{
"entropy": 5.899728775024414,
"epoch": 5.745897563401293,
"grad_norm": 1.1484375,
"learning_rate": 0.00025518679886110023,
"loss": 5.3955,
"mean_token_accuracy": 0.1913009136915207,
"num_tokens": 11318616.0,
"step": 5780
},
{
"entropy": 5.943275737762451,
"epoch": 5.750870213823968,
"grad_norm": 1.3515625,
"learning_rate": 0.0002547978165214822,
"loss": 5.3701,
"mean_token_accuracy": 0.19761822521686553,
"num_tokens": 11328751.0,
"step": 5785
},
{
"entropy": 5.753283977508545,
"epoch": 5.7558428642466435,
"grad_norm": 1.2734375,
"learning_rate": 0.00025440889504310416,
"loss": 5.1504,
"mean_token_accuracy": 0.21352695524692536,
"num_tokens": 11338054.0,
"step": 5790
},
{
"entropy": 5.928505897521973,
"epoch": 5.760815514669319,
"grad_norm": 1.234375,
"learning_rate": 0.00025402003559763393,
"loss": 5.4204,
"mean_token_accuracy": 0.19067077934741974,
"num_tokens": 11348341.0,
"step": 5795
},
{
"entropy": 5.966187763214111,
"epoch": 5.765788165091994,
"grad_norm": 1.2578125,
"learning_rate": 0.00025363123935655214,
"loss": 5.3672,
"mean_token_accuracy": 0.19457206428050994,
"num_tokens": 11357496.0,
"step": 5800
},
{
"entropy": 5.900295400619507,
"epoch": 5.770760815514669,
"grad_norm": 1.1640625,
"learning_rate": 0.0002532425074911491,
"loss": 5.3547,
"mean_token_accuracy": 0.19298552870750427,
"num_tokens": 11366915.0,
"step": 5805
},
{
"entropy": 5.9182164669036865,
"epoch": 5.7757334659373445,
"grad_norm": 1.2890625,
"learning_rate": 0.0002528538411725212,
"loss": 5.4431,
"mean_token_accuracy": 0.18217946141958236,
"num_tokens": 11376920.0,
"step": 5810
},
{
"entropy": 5.774221611022949,
"epoch": 5.78070611636002,
"grad_norm": 1.296875,
"learning_rate": 0.0002524652415715672,
"loss": 5.1914,
"mean_token_accuracy": 0.21514922231435776,
"num_tokens": 11387405.0,
"step": 5815
},
{
"entropy": 6.016244649887085,
"epoch": 5.785678766782695,
"grad_norm": 1.3359375,
"learning_rate": 0.0002520767098589852,
"loss": 5.5175,
"mean_token_accuracy": 0.18416470885276795,
"num_tokens": 11396553.0,
"step": 5820
},
{
"entropy": 6.008527660369873,
"epoch": 5.79065141720537,
"grad_norm": 1.1328125,
"learning_rate": 0.0002516882472052683,
"loss": 5.4717,
"mean_token_accuracy": 0.18571110516786576,
"num_tokens": 11406964.0,
"step": 5825
},
{
"entropy": 5.969941234588623,
"epoch": 5.7956240676280455,
"grad_norm": 1.34375,
"learning_rate": 0.0002512998547807022,
"loss": 5.419,
"mean_token_accuracy": 0.19051730632781982,
"num_tokens": 11416550.0,
"step": 5830
},
{
"entropy": 5.909865474700927,
"epoch": 5.800596718050721,
"grad_norm": 1.3203125,
"learning_rate": 0.00025091153375536026,
"loss": 5.2643,
"mean_token_accuracy": 0.20233702659606934,
"num_tokens": 11425477.0,
"step": 5835
},
{
"entropy": 5.925911903381348,
"epoch": 5.805569368473396,
"grad_norm": 1.3046875,
"learning_rate": 0.00025052328529910135,
"loss": 5.3217,
"mean_token_accuracy": 0.19364791363477707,
"num_tokens": 11435153.0,
"step": 5840
},
{
"entropy": 5.83957052230835,
"epoch": 5.810542018896071,
"grad_norm": 1.15625,
"learning_rate": 0.0002501351105815655,
"loss": 5.3015,
"mean_token_accuracy": 0.20378607213497163,
"num_tokens": 11445635.0,
"step": 5845
},
{
"entropy": 5.97175087928772,
"epoch": 5.8155146693187465,
"grad_norm": 1.2265625,
"learning_rate": 0.00024974701077217057,
"loss": 5.5043,
"mean_token_accuracy": 0.18324391394853592,
"num_tokens": 11455526.0,
"step": 5850
},
{
"entropy": 5.9141051292419435,
"epoch": 5.820487319741423,
"grad_norm": 1.3125,
"learning_rate": 0.00024935898704010875,
"loss": 5.3481,
"mean_token_accuracy": 0.19214192926883697,
"num_tokens": 11464115.0,
"step": 5855
},
{
"entropy": 5.828054428100586,
"epoch": 5.825459970164097,
"grad_norm": 1.171875,
"learning_rate": 0.00024897104055434304,
"loss": 5.264,
"mean_token_accuracy": 0.2073179453611374,
"num_tokens": 11475456.0,
"step": 5860
},
{
"entropy": 5.797663068771362,
"epoch": 5.830432620586773,
"grad_norm": 1.3203125,
"learning_rate": 0.00024858317248360387,
"loss": 5.1744,
"mean_token_accuracy": 0.21666699647903442,
"num_tokens": 11485796.0,
"step": 5865
},
{
"entropy": 5.98051643371582,
"epoch": 5.835405271009448,
"grad_norm": 1.3359375,
"learning_rate": 0.0002481953839963851,
"loss": 5.4202,
"mean_token_accuracy": 0.18856512606143952,
"num_tokens": 11494904.0,
"step": 5870
},
{
"entropy": 5.7613616466522215,
"epoch": 5.840377921432124,
"grad_norm": 1.203125,
"learning_rate": 0.0002478076762609413,
"loss": 5.1756,
"mean_token_accuracy": 0.2155476450920105,
"num_tokens": 11504429.0,
"step": 5875
},
{
"entropy": 5.955584478378296,
"epoch": 5.845350571854799,
"grad_norm": 1.234375,
"learning_rate": 0.0002474200504452834,
"loss": 5.4033,
"mean_token_accuracy": 0.1879813179373741,
"num_tokens": 11515557.0,
"step": 5880
},
{
"entropy": 6.02553014755249,
"epoch": 5.850323222277474,
"grad_norm": 1.2421875,
"learning_rate": 0.00024703250771717576,
"loss": 5.4656,
"mean_token_accuracy": 0.1866442784667015,
"num_tokens": 11525613.0,
"step": 5885
},
{
"entropy": 5.965510749816895,
"epoch": 5.855295872700149,
"grad_norm": 1.2265625,
"learning_rate": 0.00024664504924413224,
"loss": 5.3865,
"mean_token_accuracy": 0.19069099128246308,
"num_tokens": 11535463.0,
"step": 5890
},
{
"entropy": 5.906952857971191,
"epoch": 5.860268523122825,
"grad_norm": 1.2890625,
"learning_rate": 0.0002462576761934129,
"loss": 5.3414,
"mean_token_accuracy": 0.19323182553052903,
"num_tokens": 11546277.0,
"step": 5895
},
{
"entropy": 5.9054313659667965,
"epoch": 5.8652411735455,
"grad_norm": 1.3671875,
"learning_rate": 0.0002458703897320208,
"loss": 5.3869,
"mean_token_accuracy": 0.18860101699829102,
"num_tokens": 11555361.0,
"step": 5900
},
{
"entropy": 5.9093162536621096,
"epoch": 5.870213823968175,
"grad_norm": 1.21875,
"learning_rate": 0.00024548319102669745,
"loss": 5.2535,
"mean_token_accuracy": 0.19613767862319947,
"num_tokens": 11564976.0,
"step": 5905
},
{
"entropy": 5.8248734951019285,
"epoch": 5.87518647439085,
"grad_norm": 1.3046875,
"learning_rate": 0.00024509608124392065,
"loss": 5.2336,
"mean_token_accuracy": 0.1965501695871353,
"num_tokens": 11574574.0,
"step": 5910
},
{
"entropy": 5.918524217605591,
"epoch": 5.880159124813526,
"grad_norm": 1.2265625,
"learning_rate": 0.0002447090615499,
"loss": 5.3788,
"mean_token_accuracy": 0.18683090209960937,
"num_tokens": 11584571.0,
"step": 5915
},
{
"entropy": 6.027804803848267,
"epoch": 5.885131775236201,
"grad_norm": 1.296875,
"learning_rate": 0.00024432213311057383,
"loss": 5.4706,
"mean_token_accuracy": 0.18577160239219664,
"num_tokens": 11594829.0,
"step": 5920
},
{
"entropy": 5.95970401763916,
"epoch": 5.890104425658876,
"grad_norm": 1.40625,
"learning_rate": 0.00024393529709160544,
"loss": 5.4455,
"mean_token_accuracy": 0.19251643419265746,
"num_tokens": 11604520.0,
"step": 5925
},
{
"entropy": 5.969117975234985,
"epoch": 5.895077076081551,
"grad_norm": 1.3125,
"learning_rate": 0.00024354855465837967,
"loss": 5.3403,
"mean_token_accuracy": 0.19593659788370132,
"num_tokens": 11612909.0,
"step": 5930
},
{
"entropy": 5.93201642036438,
"epoch": 5.900049726504227,
"grad_norm": 1.21875,
"learning_rate": 0.00024316190697599955,
"loss": 5.4728,
"mean_token_accuracy": 0.18393043130636216,
"num_tokens": 11622928.0,
"step": 5935
},
{
"entropy": 5.98280897140503,
"epoch": 5.905022376926902,
"grad_norm": 1.3203125,
"learning_rate": 0.00024277535520928257,
"loss": 5.5018,
"mean_token_accuracy": 0.18099606782197952,
"num_tokens": 11632192.0,
"step": 5940
},
{
"entropy": 5.956179428100586,
"epoch": 5.909995027349577,
"grad_norm": 1.2109375,
"learning_rate": 0.00024238890052275742,
"loss": 5.344,
"mean_token_accuracy": 0.1972501501441002,
"num_tokens": 11642346.0,
"step": 5945
},
{
"entropy": 5.932433128356934,
"epoch": 5.914967677772252,
"grad_norm": 1.28125,
"learning_rate": 0.00024200254408066014,
"loss": 5.3154,
"mean_token_accuracy": 0.19756143689155578,
"num_tokens": 11651527.0,
"step": 5950
},
{
"entropy": 5.948135852813721,
"epoch": 5.919940328194928,
"grad_norm": 1.2890625,
"learning_rate": 0.0002416162870469308,
"loss": 5.4199,
"mean_token_accuracy": 0.18299988508224488,
"num_tokens": 11660920.0,
"step": 5955
},
{
"entropy": 5.910482597351074,
"epoch": 5.924912978617603,
"grad_norm": 1.3671875,
"learning_rate": 0.0002412301305852102,
"loss": 5.345,
"mean_token_accuracy": 0.19732747077941895,
"num_tokens": 11670220.0,
"step": 5960
},
{
"entropy": 5.740287160873413,
"epoch": 5.929885629040278,
"grad_norm": 1.1953125,
"learning_rate": 0.00024084407585883588,
"loss": 5.1754,
"mean_token_accuracy": 0.2219235733151436,
"num_tokens": 11680922.0,
"step": 5965
},
{
"entropy": 5.901228332519532,
"epoch": 5.934858279462953,
"grad_norm": 1.2890625,
"learning_rate": 0.0002404581240308391,
"loss": 5.3444,
"mean_token_accuracy": 0.19460434317588807,
"num_tokens": 11691138.0,
"step": 5970
},
{
"entropy": 5.947125959396362,
"epoch": 5.939830929885629,
"grad_norm": 1.2578125,
"learning_rate": 0.000240072276263941,
"loss": 5.3921,
"mean_token_accuracy": 0.18757044821977614,
"num_tokens": 11700379.0,
"step": 5975
},
{
"entropy": 6.043396949768066,
"epoch": 5.944803580308305,
"grad_norm": 1.46875,
"learning_rate": 0.00023968653372054933,
"loss": 5.4017,
"mean_token_accuracy": 0.1809894025325775,
"num_tokens": 11709678.0,
"step": 5980
},
{
"entropy": 5.852338886260986,
"epoch": 5.949776230730979,
"grad_norm": 1.5234375,
"learning_rate": 0.00023930089756275474,
"loss": 5.1907,
"mean_token_accuracy": 0.2039460778236389,
"num_tokens": 11719700.0,
"step": 5985
},
{
"entropy": 5.87944016456604,
"epoch": 5.954748881153655,
"grad_norm": 1.3515625,
"learning_rate": 0.0002389153689523275,
"loss": 5.3541,
"mean_token_accuracy": 0.18902938663959504,
"num_tokens": 11730087.0,
"step": 5990
},
{
"entropy": 5.915482759475708,
"epoch": 5.9597215315763306,
"grad_norm": 1.21875,
"learning_rate": 0.00023852994905071385,
"loss": 5.3845,
"mean_token_accuracy": 0.1909420296549797,
"num_tokens": 11740063.0,
"step": 5995
},
{
"entropy": 6.065284490585327,
"epoch": 5.964694181999006,
"grad_norm": 1.203125,
"learning_rate": 0.00023814463901903238,
"loss": 5.4941,
"mean_token_accuracy": 0.1806158736348152,
"num_tokens": 11750213.0,
"step": 6000
},
{
"epoch": 5.964694181999006,
"eval_entropy": 5.924021175929478,
"eval_loss": 6.492471694946289,
"eval_mean_token_accuracy": 0.14681892322745466,
"eval_num_tokens": 11750213.0,
"eval_runtime": 2.5777,
"eval_samples_per_second": 1344.239,
"eval_steps_per_second": 168.369,
"step": 6000
},
{
"entropy": 5.976996660232544,
"epoch": 5.969666832421681,
"grad_norm": 1.203125,
"learning_rate": 0.00023775944001807077,
"loss": 5.3875,
"mean_token_accuracy": 0.18656652867794038,
"num_tokens": 11759686.0,
"step": 6005
},
{
"entropy": 5.90385422706604,
"epoch": 5.974639482844356,
"grad_norm": 1.3125,
"learning_rate": 0.00023737435320828244,
"loss": 5.3738,
"mean_token_accuracy": 0.19613661170005797,
"num_tokens": 11770199.0,
"step": 6010
},
{
"entropy": 5.641528511047364,
"epoch": 5.979612133267032,
"grad_norm": 1.2578125,
"learning_rate": 0.00023698937974978257,
"loss": 5.0594,
"mean_token_accuracy": 0.23774951845407485,
"num_tokens": 11780585.0,
"step": 6015
},
{
"entropy": 6.0021051406860355,
"epoch": 5.984584783689707,
"grad_norm": 1.2890625,
"learning_rate": 0.00023660452080234473,
"loss": 5.4507,
"mean_token_accuracy": 0.18467377871274948,
"num_tokens": 11789938.0,
"step": 6020
},
{
"entropy": 5.909160232543945,
"epoch": 5.989557434112382,
"grad_norm": 1.2578125,
"learning_rate": 0.00023621977752539785,
"loss": 5.3278,
"mean_token_accuracy": 0.20282323509454728,
"num_tokens": 11799783.0,
"step": 6025
},
{
"entropy": 6.029426527023316,
"epoch": 5.994530084535057,
"grad_norm": 1.3203125,
"learning_rate": 0.0002358351510780223,
"loss": 5.5006,
"mean_token_accuracy": 0.17972384840250016,
"num_tokens": 11809334.0,
"step": 6030
},
{
"entropy": 5.952344703674316,
"epoch": 5.999502734957733,
"grad_norm": 1.1328125,
"learning_rate": 0.00023545064261894623,
"loss": 5.3799,
"mean_token_accuracy": 0.18367550075054168,
"num_tokens": 11820136.0,
"step": 6035
},
{
"entropy": 5.919129265679254,
"epoch": 6.00397812033814,
"grad_norm": 1.25,
"learning_rate": 0.00023506625330654263,
"loss": 5.2138,
"mean_token_accuracy": 0.2042764574289322,
"num_tokens": 11828628.0,
"step": 6040
},
{
"entropy": 5.8930196285247805,
"epoch": 6.008950770760816,
"grad_norm": 1.203125,
"learning_rate": 0.00023468198429882548,
"loss": 5.1425,
"mean_token_accuracy": 0.20828983783721924,
"num_tokens": 11837964.0,
"step": 6045
},
{
"entropy": 5.914318561553955,
"epoch": 6.013923421183491,
"grad_norm": 1.3125,
"learning_rate": 0.0002342978367534462,
"loss": 5.1563,
"mean_token_accuracy": 0.2026321068406105,
"num_tokens": 11847801.0,
"step": 6050
},
{
"entropy": 5.882364702224732,
"epoch": 6.018896071606166,
"grad_norm": 1.1484375,
"learning_rate": 0.0002339138118276905,
"loss": 5.1036,
"mean_token_accuracy": 0.20919743329286575,
"num_tokens": 11857836.0,
"step": 6055
},
{
"entropy": 5.955152130126953,
"epoch": 6.023868722028841,
"grad_norm": 1.1796875,
"learning_rate": 0.00023352991067847456,
"loss": 5.2151,
"mean_token_accuracy": 0.19796516597270966,
"num_tokens": 11867809.0,
"step": 6060
},
{
"entropy": 5.9229330062866214,
"epoch": 6.028841372451517,
"grad_norm": 1.2890625,
"learning_rate": 0.00023314613446234162,
"loss": 5.1557,
"mean_token_accuracy": 0.2022574171423912,
"num_tokens": 11876869.0,
"step": 6065
},
{
"entropy": 5.963329744338989,
"epoch": 6.033814022874192,
"grad_norm": 1.1640625,
"learning_rate": 0.00023276248433545867,
"loss": 5.1496,
"mean_token_accuracy": 0.2083750396966934,
"num_tokens": 11886816.0,
"step": 6070
},
{
"entropy": 5.9732459545135494,
"epoch": 6.038786673296867,
"grad_norm": 1.15625,
"learning_rate": 0.00023237896145361276,
"loss": 5.2926,
"mean_token_accuracy": 0.1916511282324791,
"num_tokens": 11896209.0,
"step": 6075
},
{
"entropy": 5.897648286819458,
"epoch": 6.043759323719542,
"grad_norm": 1.21875,
"learning_rate": 0.00023199556697220758,
"loss": 5.0884,
"mean_token_accuracy": 0.2155316412448883,
"num_tokens": 11906522.0,
"step": 6080
},
{
"entropy": 5.933634853363037,
"epoch": 6.048731974142218,
"grad_norm": 1.3359375,
"learning_rate": 0.00023161230204626015,
"loss": 5.1526,
"mean_token_accuracy": 0.20854526609182358,
"num_tokens": 11916040.0,
"step": 6085
},
{
"entropy": 5.681178092956543,
"epoch": 6.053704624564893,
"grad_norm": 1.09375,
"learning_rate": 0.00023122916783039694,
"loss": 4.8828,
"mean_token_accuracy": 0.2345024362206459,
"num_tokens": 11927138.0,
"step": 6090
},
{
"entropy": 5.902951240539551,
"epoch": 6.058677274987568,
"grad_norm": 1.203125,
"learning_rate": 0.0002308461654788509,
"loss": 5.1082,
"mean_token_accuracy": 0.20991631597280502,
"num_tokens": 11937385.0,
"step": 6095
},
{
"entropy": 5.89009919166565,
"epoch": 6.063649925410243,
"grad_norm": 1.171875,
"learning_rate": 0.00023046329614545758,
"loss": 5.104,
"mean_token_accuracy": 0.2056956857442856,
"num_tokens": 11946621.0,
"step": 6100
},
{
"entropy": 5.919991445541382,
"epoch": 6.068622575832919,
"grad_norm": 1.375,
"learning_rate": 0.00023008056098365183,
"loss": 5.1605,
"mean_token_accuracy": 0.2006782978773117,
"num_tokens": 11956744.0,
"step": 6105
},
{
"entropy": 5.997849035263061,
"epoch": 6.073595226255594,
"grad_norm": 1.28125,
"learning_rate": 0.00022969796114646424,
"loss": 5.2406,
"mean_token_accuracy": 0.20095346122980118,
"num_tokens": 11967702.0,
"step": 6110
},
{
"entropy": 5.9279543399810795,
"epoch": 6.078567876678269,
"grad_norm": 1.3828125,
"learning_rate": 0.00022931549778651783,
"loss": 5.1123,
"mean_token_accuracy": 0.20623803734779358,
"num_tokens": 11976895.0,
"step": 6115
},
{
"entropy": 5.9037830352783205,
"epoch": 6.083540527100944,
"grad_norm": 1.390625,
"learning_rate": 0.00022893317205602444,
"loss": 5.1771,
"mean_token_accuracy": 0.20432715713977814,
"num_tokens": 11985893.0,
"step": 6120
},
{
"entropy": 5.869170904159546,
"epoch": 6.08851317752362,
"grad_norm": 1.1796875,
"learning_rate": 0.00022855098510678125,
"loss": 5.1599,
"mean_token_accuracy": 0.2025202602148056,
"num_tokens": 11995254.0,
"step": 6125
},
{
"entropy": 5.949420595169068,
"epoch": 6.093485827946296,
"grad_norm": 1.25,
"learning_rate": 0.00022816893809016737,
"loss": 5.1538,
"mean_token_accuracy": 0.19390280693769454,
"num_tokens": 12006983.0,
"step": 6130
},
{
"entropy": 5.8580811500549315,
"epoch": 6.098458478368971,
"grad_norm": 1.234375,
"learning_rate": 0.00022778703215714014,
"loss": 5.0814,
"mean_token_accuracy": 0.21563820093870162,
"num_tokens": 12019428.0,
"step": 6135
},
{
"entropy": 5.8901598930358885,
"epoch": 6.103431128791646,
"grad_norm": 1.1875,
"learning_rate": 0.00022740526845823223,
"loss": 5.1298,
"mean_token_accuracy": 0.20294146984815598,
"num_tokens": 12029613.0,
"step": 6140
},
{
"entropy": 6.014317846298217,
"epoch": 6.1084037792143215,
"grad_norm": 1.3046875,
"learning_rate": 0.0002270236481435477,
"loss": 5.2431,
"mean_token_accuracy": 0.19407914131879805,
"num_tokens": 12039328.0,
"step": 6145
},
{
"entropy": 5.9151286602020265,
"epoch": 6.113376429636997,
"grad_norm": 1.3671875,
"learning_rate": 0.00022664217236275847,
"loss": 5.0639,
"mean_token_accuracy": 0.21611939817667009,
"num_tokens": 12048894.0,
"step": 6150
},
{
"entropy": 5.706415033340454,
"epoch": 6.118349080059672,
"grad_norm": 1.2109375,
"learning_rate": 0.00022626084226510124,
"loss": 5.0146,
"mean_token_accuracy": 0.2277321457862854,
"num_tokens": 12059316.0,
"step": 6155
},
{
"entropy": 5.905489444732666,
"epoch": 6.123321730482347,
"grad_norm": 1.328125,
"learning_rate": 0.00022587965899937364,
"loss": 5.1569,
"mean_token_accuracy": 0.2061835765838623,
"num_tokens": 12069142.0,
"step": 6160
},
{
"entropy": 5.880143165588379,
"epoch": 6.1282943809050225,
"grad_norm": 1.2734375,
"learning_rate": 0.0002254986237139312,
"loss": 5.0975,
"mean_token_accuracy": 0.21792755872011185,
"num_tokens": 12077893.0,
"step": 6165
},
{
"entropy": 5.922898769378662,
"epoch": 6.133267031327698,
"grad_norm": 1.2734375,
"learning_rate": 0.00022511773755668342,
"loss": 5.3114,
"mean_token_accuracy": 0.19862927198410035,
"num_tokens": 12087811.0,
"step": 6170
},
{
"entropy": 5.905173921585083,
"epoch": 6.138239681750373,
"grad_norm": 1.3125,
"learning_rate": 0.0002247370016750907,
"loss": 5.1538,
"mean_token_accuracy": 0.20618564188480376,
"num_tokens": 12097172.0,
"step": 6175
},
{
"entropy": 5.963050889968872,
"epoch": 6.143212332173048,
"grad_norm": 1.2109375,
"learning_rate": 0.0002243564172161605,
"loss": 5.2103,
"mean_token_accuracy": 0.20390161126852036,
"num_tokens": 12106787.0,
"step": 6180
},
{
"entropy": 5.912859582901001,
"epoch": 6.1481849825957235,
"grad_norm": 1.3828125,
"learning_rate": 0.00022397598532644441,
"loss": 5.1753,
"mean_token_accuracy": 0.20594002157449723,
"num_tokens": 12116815.0,
"step": 6185
},
{
"entropy": 5.911113262176514,
"epoch": 6.153157633018399,
"grad_norm": 1.25,
"learning_rate": 0.00022359570715203408,
"loss": 5.1976,
"mean_token_accuracy": 0.20201392769813536,
"num_tokens": 12127439.0,
"step": 6190
},
{
"entropy": 5.922432279586792,
"epoch": 6.158130283441074,
"grad_norm": 1.2734375,
"learning_rate": 0.00022321558383855828,
"loss": 5.1431,
"mean_token_accuracy": 0.21000170558691025,
"num_tokens": 12136828.0,
"step": 6195
},
{
"entropy": 5.754469633102417,
"epoch": 6.163102933863749,
"grad_norm": 1.28125,
"learning_rate": 0.0002228356165311791,
"loss": 5.0313,
"mean_token_accuracy": 0.2270607680082321,
"num_tokens": 12146593.0,
"step": 6200
},
{
"entropy": 5.871117019653321,
"epoch": 6.1680755842864246,
"grad_norm": 1.234375,
"learning_rate": 0.00022245580637458883,
"loss": 5.1627,
"mean_token_accuracy": 0.2121468484401703,
"num_tokens": 12157623.0,
"step": 6205
},
{
"entropy": 5.843371200561523,
"epoch": 6.1730482347091,
"grad_norm": 1.28125,
"learning_rate": 0.00022207615451300616,
"loss": 5.1264,
"mean_token_accuracy": 0.21073680818080903,
"num_tokens": 12167353.0,
"step": 6210
},
{
"entropy": 5.930319690704346,
"epoch": 6.178020885131775,
"grad_norm": 1.28125,
"learning_rate": 0.00022169666209017304,
"loss": 5.1688,
"mean_token_accuracy": 0.20581168085336685,
"num_tokens": 12176669.0,
"step": 6215
},
{
"entropy": 5.864517021179199,
"epoch": 6.18299353555445,
"grad_norm": 1.3828125,
"learning_rate": 0.0002213173302493509,
"loss": 5.0732,
"mean_token_accuracy": 0.2192705824971199,
"num_tokens": 12185789.0,
"step": 6220
},
{
"entropy": 5.882683038711548,
"epoch": 6.1879661859771256,
"grad_norm": 1.3046875,
"learning_rate": 0.00022093816013331763,
"loss": 5.2107,
"mean_token_accuracy": 0.20249402076005935,
"num_tokens": 12195289.0,
"step": 6225
},
{
"entropy": 5.876528644561768,
"epoch": 6.192938836399801,
"grad_norm": 1.359375,
"learning_rate": 0.00022055915288436362,
"loss": 5.1595,
"mean_token_accuracy": 0.2097211003303528,
"num_tokens": 12204499.0,
"step": 6230
},
{
"entropy": 5.711639547348023,
"epoch": 6.197911486822476,
"grad_norm": 1.296875,
"learning_rate": 0.00022018030964428897,
"loss": 5.0228,
"mean_token_accuracy": 0.2252396523952484,
"num_tokens": 12214739.0,
"step": 6235
},
{
"entropy": 5.92997899055481,
"epoch": 6.202884137245151,
"grad_norm": 1.421875,
"learning_rate": 0.00021980163155439943,
"loss": 5.0984,
"mean_token_accuracy": 0.21329424679279327,
"num_tokens": 12224027.0,
"step": 6240
},
{
"entropy": 5.918960332870483,
"epoch": 6.207856787667827,
"grad_norm": 1.2578125,
"learning_rate": 0.00021942311975550323,
"loss": 5.1918,
"mean_token_accuracy": 0.20438357293605805,
"num_tokens": 12234317.0,
"step": 6245
},
{
"entropy": 5.886983489990234,
"epoch": 6.212829438090502,
"grad_norm": 1.2890625,
"learning_rate": 0.00021904477538790774,
"loss": 5.0367,
"mean_token_accuracy": 0.212549589574337,
"num_tokens": 12244561.0,
"step": 6250
},
{
"entropy": 5.627741122245789,
"epoch": 6.217802088513178,
"grad_norm": 1.3203125,
"learning_rate": 0.00021866659959141578,
"loss": 4.9026,
"mean_token_accuracy": 0.2377011999487877,
"num_tokens": 12254480.0,
"step": 6255
},
{
"entropy": 5.905136346817017,
"epoch": 6.222774738935853,
"grad_norm": 1.359375,
"learning_rate": 0.0002182885935053225,
"loss": 5.2175,
"mean_token_accuracy": 0.19647449404001235,
"num_tokens": 12263531.0,
"step": 6260
},
{
"entropy": 5.932830333709717,
"epoch": 6.2277473893585285,
"grad_norm": 1.265625,
"learning_rate": 0.00021791075826841166,
"loss": 5.2574,
"mean_token_accuracy": 0.20756605714559556,
"num_tokens": 12273189.0,
"step": 6265
},
{
"entropy": 5.852635908126831,
"epoch": 6.232720039781204,
"grad_norm": 1.3828125,
"learning_rate": 0.0002175330950189523,
"loss": 5.145,
"mean_token_accuracy": 0.20975629091262818,
"num_tokens": 12282241.0,
"step": 6270
},
{
"entropy": 5.912967300415039,
"epoch": 6.237692690203879,
"grad_norm": 1.3359375,
"learning_rate": 0.0002171556048946955,
"loss": 5.11,
"mean_token_accuracy": 0.21054715514183045,
"num_tokens": 12291803.0,
"step": 6275
},
{
"entropy": 5.9338319301605225,
"epoch": 6.242665340626554,
"grad_norm": 1.3359375,
"learning_rate": 0.00021677828903287063,
"loss": 5.2031,
"mean_token_accuracy": 0.20140755027532578,
"num_tokens": 12302086.0,
"step": 6280
},
{
"entropy": 5.900045585632324,
"epoch": 6.2476379910492295,
"grad_norm": 1.2734375,
"learning_rate": 0.00021640114857018205,
"loss": 5.171,
"mean_token_accuracy": 0.2044919967651367,
"num_tokens": 12311651.0,
"step": 6285
},
{
"entropy": 5.840173673629761,
"epoch": 6.252610641471905,
"grad_norm": 1.234375,
"learning_rate": 0.00021602418464280586,
"loss": 5.173,
"mean_token_accuracy": 0.21022891849279404,
"num_tokens": 12321583.0,
"step": 6290
},
{
"entropy": 5.898901796340942,
"epoch": 6.25758329189458,
"grad_norm": 1.3828125,
"learning_rate": 0.00021564739838638615,
"loss": 5.1844,
"mean_token_accuracy": 0.20597055107355117,
"num_tokens": 12331139.0,
"step": 6295
},
{
"entropy": 5.805557346343994,
"epoch": 6.262555942317255,
"grad_norm": 1.1640625,
"learning_rate": 0.00021527079093603202,
"loss": 5.0302,
"mean_token_accuracy": 0.2174318924546242,
"num_tokens": 12341509.0,
"step": 6300
},
{
"entropy": 5.86940221786499,
"epoch": 6.2675285927399305,
"grad_norm": 1.28125,
"learning_rate": 0.00021489436342631352,
"loss": 5.2187,
"mean_token_accuracy": 0.19796469509601594,
"num_tokens": 12350450.0,
"step": 6305
},
{
"entropy": 5.857239723205566,
"epoch": 6.272501243162606,
"grad_norm": 1.21875,
"learning_rate": 0.00021451811699125883,
"loss": 5.0737,
"mean_token_accuracy": 0.21463742405176162,
"num_tokens": 12359168.0,
"step": 6310
},
{
"entropy": 5.8950670719146725,
"epoch": 6.277473893585281,
"grad_norm": 1.40625,
"learning_rate": 0.00021414205276435073,
"loss": 5.1734,
"mean_token_accuracy": 0.202508744597435,
"num_tokens": 12367831.0,
"step": 6315
},
{
"entropy": 5.905054044723511,
"epoch": 6.282446544007956,
"grad_norm": 1.25,
"learning_rate": 0.0002137661718785229,
"loss": 5.2559,
"mean_token_accuracy": 0.19912850111722946,
"num_tokens": 12377804.0,
"step": 6320
},
{
"entropy": 5.94793062210083,
"epoch": 6.2874191944306315,
"grad_norm": 1.2890625,
"learning_rate": 0.00021339047546615675,
"loss": 5.1939,
"mean_token_accuracy": 0.20171410739421844,
"num_tokens": 12387022.0,
"step": 6325
},
{
"entropy": 5.838116931915283,
"epoch": 6.292391844853307,
"grad_norm": 1.2578125,
"learning_rate": 0.00021301496465907794,
"loss": 5.0794,
"mean_token_accuracy": 0.20570251792669297,
"num_tokens": 12395935.0,
"step": 6330
},
{
"entropy": 5.924272108078003,
"epoch": 6.297364495275982,
"grad_norm": 1.328125,
"learning_rate": 0.00021263964058855294,
"loss": 5.2231,
"mean_token_accuracy": 0.20136867761611937,
"num_tokens": 12405760.0,
"step": 6335
},
{
"entropy": 5.703550457954407,
"epoch": 6.302337145698657,
"grad_norm": 1.296875,
"learning_rate": 0.00021226450438528572,
"loss": 4.935,
"mean_token_accuracy": 0.2306780159473419,
"num_tokens": 12415573.0,
"step": 6340
},
{
"entropy": 5.8941466331481935,
"epoch": 6.3073097961213325,
"grad_norm": 1.203125,
"learning_rate": 0.00021188955717941428,
"loss": 5.1941,
"mean_token_accuracy": 0.2038727208971977,
"num_tokens": 12426532.0,
"step": 6345
},
{
"entropy": 5.897432947158814,
"epoch": 6.312282446544008,
"grad_norm": 1.25,
"learning_rate": 0.0002115148001005071,
"loss": 5.2012,
"mean_token_accuracy": 0.20533437430858612,
"num_tokens": 12438319.0,
"step": 6350
},
{
"entropy": 5.8643933773040775,
"epoch": 6.317255096966683,
"grad_norm": 1.203125,
"learning_rate": 0.00021114023427756012,
"loss": 5.1008,
"mean_token_accuracy": 0.2185060679912567,
"num_tokens": 12448470.0,
"step": 6355
},
{
"entropy": 5.931847381591797,
"epoch": 6.322227747389358,
"grad_norm": 1.3359375,
"learning_rate": 0.00021076586083899296,
"loss": 5.2041,
"mean_token_accuracy": 0.20365652143955232,
"num_tokens": 12457647.0,
"step": 6360
},
{
"entropy": 5.9892524719238285,
"epoch": 6.3272003978120335,
"grad_norm": 1.5078125,
"learning_rate": 0.0002103916809126456,
"loss": 5.2633,
"mean_token_accuracy": 0.1989323228597641,
"num_tokens": 12467841.0,
"step": 6365
},
{
"entropy": 5.909793949127197,
"epoch": 6.332173048234709,
"grad_norm": 1.3046875,
"learning_rate": 0.00021001769562577517,
"loss": 5.2494,
"mean_token_accuracy": 0.19474196434020996,
"num_tokens": 12478404.0,
"step": 6370
},
{
"entropy": 5.923398351669311,
"epoch": 6.337145698657384,
"grad_norm": 1.421875,
"learning_rate": 0.0002096439061050523,
"loss": 5.1807,
"mean_token_accuracy": 0.19915954023599625,
"num_tokens": 12486770.0,
"step": 6375
},
{
"entropy": 5.902583837509155,
"epoch": 6.34211834908006,
"grad_norm": 1.25,
"learning_rate": 0.000209270313476558,
"loss": 5.1737,
"mean_token_accuracy": 0.20882674008607865,
"num_tokens": 12496769.0,
"step": 6380
},
{
"entropy": 5.882408857345581,
"epoch": 6.347090999502735,
"grad_norm": 1.34375,
"learning_rate": 0.00020889691886578004,
"loss": 5.1654,
"mean_token_accuracy": 0.19846590757369995,
"num_tokens": 12505552.0,
"step": 6385
},
{
"entropy": 5.718986511230469,
"epoch": 6.352063649925411,
"grad_norm": 1.203125,
"learning_rate": 0.00020852372339760973,
"loss": 4.8983,
"mean_token_accuracy": 0.23832790702581405,
"num_tokens": 12515391.0,
"step": 6390
},
{
"entropy": 5.972678136825562,
"epoch": 6.357036300348086,
"grad_norm": 1.40625,
"learning_rate": 0.0002081507281963383,
"loss": 5.287,
"mean_token_accuracy": 0.1921763926744461,
"num_tokens": 12525001.0,
"step": 6395
},
{
"entropy": 5.880500268936157,
"epoch": 6.362008950770761,
"grad_norm": 1.421875,
"learning_rate": 0.00020777793438565368,
"loss": 5.1199,
"mean_token_accuracy": 0.21290093958377837,
"num_tokens": 12534393.0,
"step": 6400
},
{
"entropy": 5.796476411819458,
"epoch": 6.366981601193436,
"grad_norm": 1.1875,
"learning_rate": 0.00020740534308863717,
"loss": 5.0192,
"mean_token_accuracy": 0.22069818079471587,
"num_tokens": 12544358.0,
"step": 6405
},
{
"entropy": 5.944048023223877,
"epoch": 6.371954251616112,
"grad_norm": 1.1640625,
"learning_rate": 0.00020703295542775992,
"loss": 5.2265,
"mean_token_accuracy": 0.20391181856393814,
"num_tokens": 12555113.0,
"step": 6410
},
{
"entropy": 5.916614246368408,
"epoch": 6.376926902038787,
"grad_norm": 1.3125,
"learning_rate": 0.00020666077252487968,
"loss": 5.1801,
"mean_token_accuracy": 0.20560444742441178,
"num_tokens": 12565204.0,
"step": 6415
},
{
"entropy": 5.90169768333435,
"epoch": 6.381899552461462,
"grad_norm": 1.1328125,
"learning_rate": 0.0002062887955012372,
"loss": 5.0952,
"mean_token_accuracy": 0.2128805473446846,
"num_tokens": 12574789.0,
"step": 6420
},
{
"entropy": 5.815948152542115,
"epoch": 6.386872202884137,
"grad_norm": 1.3671875,
"learning_rate": 0.00020591702547745306,
"loss": 5.0967,
"mean_token_accuracy": 0.21106737852096558,
"num_tokens": 12583865.0,
"step": 6425
},
{
"entropy": 5.832252931594849,
"epoch": 6.391844853306813,
"grad_norm": 1.203125,
"learning_rate": 0.00020554546357352433,
"loss": 5.1804,
"mean_token_accuracy": 0.20326941311359406,
"num_tokens": 12593778.0,
"step": 6430
},
{
"entropy": 5.902464866638184,
"epoch": 6.396817503729488,
"grad_norm": 1.421875,
"learning_rate": 0.000205174110908821,
"loss": 5.1661,
"mean_token_accuracy": 0.20678598284721375,
"num_tokens": 12603571.0,
"step": 6435
},
{
"entropy": 6.000392580032349,
"epoch": 6.401790154152163,
"grad_norm": 1.4609375,
"learning_rate": 0.0002048029686020827,
"loss": 5.294,
"mean_token_accuracy": 0.1913320764899254,
"num_tokens": 12613850.0,
"step": 6440
},
{
"entropy": 5.917159700393677,
"epoch": 6.406762804574838,
"grad_norm": 1.4765625,
"learning_rate": 0.0002044320377714154,
"loss": 5.2131,
"mean_token_accuracy": 0.19892736077308654,
"num_tokens": 12623587.0,
"step": 6445
},
{
"entropy": 5.839389133453369,
"epoch": 6.411735454997514,
"grad_norm": 1.0859375,
"learning_rate": 0.00020406131953428785,
"loss": 5.0766,
"mean_token_accuracy": 0.21031469702720643,
"num_tokens": 12634702.0,
"step": 6450
},
{
"entropy": 5.853838729858398,
"epoch": 6.416708105420189,
"grad_norm": 1.3984375,
"learning_rate": 0.0002036908150075285,
"loss": 5.1015,
"mean_token_accuracy": 0.21247030347585677,
"num_tokens": 12643905.0,
"step": 6455
},
{
"entropy": 5.932968711853027,
"epoch": 6.421680755842864,
"grad_norm": 1.296875,
"learning_rate": 0.0002033205253073218,
"loss": 5.2502,
"mean_token_accuracy": 0.20006133764982223,
"num_tokens": 12653794.0,
"step": 6460
},
{
"entropy": 5.949442911148071,
"epoch": 6.426653406265539,
"grad_norm": 1.3046875,
"learning_rate": 0.00020295045154920528,
"loss": 5.2241,
"mean_token_accuracy": 0.20487441569566728,
"num_tokens": 12663744.0,
"step": 6465
},
{
"entropy": 5.908233976364135,
"epoch": 6.431626056688215,
"grad_norm": 1.328125,
"learning_rate": 0.00020258059484806564,
"loss": 5.1571,
"mean_token_accuracy": 0.20975128412246705,
"num_tokens": 12673503.0,
"step": 6470
},
{
"entropy": 5.777700138092041,
"epoch": 6.43659870711089,
"grad_norm": 1.4296875,
"learning_rate": 0.00020221095631813574,
"loss": 5.0748,
"mean_token_accuracy": 0.21796953231096267,
"num_tokens": 12683534.0,
"step": 6475
},
{
"entropy": 5.870516300201416,
"epoch": 6.441571357533565,
"grad_norm": 1.328125,
"learning_rate": 0.0002018415370729913,
"loss": 5.2076,
"mean_token_accuracy": 0.19739993065595626,
"num_tokens": 12693356.0,
"step": 6480
},
{
"entropy": 5.8438880920410154,
"epoch": 6.44654400795624,
"grad_norm": 1.34375,
"learning_rate": 0.00020147233822554722,
"loss": 5.051,
"mean_token_accuracy": 0.21474214345216752,
"num_tokens": 12702332.0,
"step": 6485
},
{
"entropy": 5.880759572982788,
"epoch": 6.451516658378916,
"grad_norm": 1.4140625,
"learning_rate": 0.00020110336088805463,
"loss": 5.1309,
"mean_token_accuracy": 0.2058056429028511,
"num_tokens": 12710759.0,
"step": 6490
},
{
"entropy": 5.9129866600036625,
"epoch": 6.456489308801591,
"grad_norm": 1.328125,
"learning_rate": 0.00020073460617209732,
"loss": 5.3001,
"mean_token_accuracy": 0.19202208518981934,
"num_tokens": 12720435.0,
"step": 6495
},
{
"entropy": 5.971777248382568,
"epoch": 6.461461959224266,
"grad_norm": 1.3828125,
"learning_rate": 0.00020036607518858824,
"loss": 5.2565,
"mean_token_accuracy": 0.19911165684461593,
"num_tokens": 12730004.0,
"step": 6500
},
{
"epoch": 6.461461959224266,
"eval_entropy": 5.888064303156417,
"eval_loss": 6.494762420654297,
"eval_mean_token_accuracy": 0.14611617657720768,
"eval_num_tokens": 12730004.0,
"eval_runtime": 2.532,
"eval_samples_per_second": 1368.484,
"eval_steps_per_second": 171.406,
"step": 6500
},
{
"entropy": 5.914685487747192,
"epoch": 6.466434609646942,
"grad_norm": 1.3046875,
"learning_rate": 0.00019999776904776646,
"loss": 5.157,
"mean_token_accuracy": 0.2116839572787285,
"num_tokens": 12739647.0,
"step": 6505
},
{
"entropy": 5.95919189453125,
"epoch": 6.471407260069617,
"grad_norm": 1.2109375,
"learning_rate": 0.0001996296888591937,
"loss": 5.2105,
"mean_token_accuracy": 0.20460763126611708,
"num_tokens": 12749055.0,
"step": 6510
},
{
"entropy": 5.844696187973023,
"epoch": 6.476379910492293,
"grad_norm": 1.296875,
"learning_rate": 0.00019926183573175084,
"loss": 5.143,
"mean_token_accuracy": 0.20786870270967484,
"num_tokens": 12759168.0,
"step": 6515
},
{
"entropy": 5.952585506439209,
"epoch": 6.481352560914968,
"grad_norm": 1.375,
"learning_rate": 0.0001988942107736348,
"loss": 5.3888,
"mean_token_accuracy": 0.18240304589271544,
"num_tokens": 12768351.0,
"step": 6520
},
{
"entropy": 5.944849586486816,
"epoch": 6.486325211337643,
"grad_norm": 1.1640625,
"learning_rate": 0.00019852681509235516,
"loss": 5.2709,
"mean_token_accuracy": 0.19185505658388138,
"num_tokens": 12778918.0,
"step": 6525
},
{
"entropy": 5.943400955200195,
"epoch": 6.4912978617603185,
"grad_norm": 1.2109375,
"learning_rate": 0.00019815964979473088,
"loss": 5.1321,
"mean_token_accuracy": 0.20453907549381256,
"num_tokens": 12789332.0,
"step": 6530
},
{
"entropy": 5.464053654670716,
"epoch": 6.496270512182994,
"grad_norm": 1.3359375,
"learning_rate": 0.00019779271598688642,
"loss": 4.7177,
"mean_token_accuracy": 0.2686206981539726,
"num_tokens": 12799299.0,
"step": 6535
},
{
"entropy": 5.840496683120728,
"epoch": 6.501243162605669,
"grad_norm": 1.375,
"learning_rate": 0.00019742601477424937,
"loss": 5.1306,
"mean_token_accuracy": 0.21665511131286622,
"num_tokens": 12808259.0,
"step": 6540
},
{
"entropy": 5.867420768737793,
"epoch": 6.506215813028344,
"grad_norm": 1.0546875,
"learning_rate": 0.0001970595472615463,
"loss": 5.115,
"mean_token_accuracy": 0.21821705400943756,
"num_tokens": 12818774.0,
"step": 6545
},
{
"entropy": 5.912881660461426,
"epoch": 6.5111884634510195,
"grad_norm": 1.2890625,
"learning_rate": 0.00019669331455279989,
"loss": 5.1718,
"mean_token_accuracy": 0.2055032506585121,
"num_tokens": 12828235.0,
"step": 6550
},
{
"entropy": 5.876584386825561,
"epoch": 6.516161113873695,
"grad_norm": 1.2578125,
"learning_rate": 0.00019632731775132522,
"loss": 5.2118,
"mean_token_accuracy": 0.1977726474404335,
"num_tokens": 12838024.0,
"step": 6555
},
{
"entropy": 5.860730457305908,
"epoch": 6.52113376429637,
"grad_norm": 1.4453125,
"learning_rate": 0.00019596155795972698,
"loss": 5.139,
"mean_token_accuracy": 0.21397786140441893,
"num_tokens": 12847863.0,
"step": 6560
},
{
"entropy": 5.892020177841187,
"epoch": 6.526106414719045,
"grad_norm": 1.3671875,
"learning_rate": 0.0001955960362798957,
"loss": 5.1867,
"mean_token_accuracy": 0.20413285493850708,
"num_tokens": 12856734.0,
"step": 6565
},
{
"entropy": 5.851820755004883,
"epoch": 6.5310790651417205,
"grad_norm": 1.3515625,
"learning_rate": 0.0001952307538130045,
"loss": 5.1652,
"mean_token_accuracy": 0.20774354338645934,
"num_tokens": 12867019.0,
"step": 6570
},
{
"entropy": 5.929245185852051,
"epoch": 6.536051715564396,
"grad_norm": 1.3984375,
"learning_rate": 0.0001948657116595059,
"loss": 5.2907,
"mean_token_accuracy": 0.19680215865373613,
"num_tokens": 12876218.0,
"step": 6575
},
{
"entropy": 5.835736227035523,
"epoch": 6.541024365987071,
"grad_norm": 1.34375,
"learning_rate": 0.00019450091091912847,
"loss": 5.1501,
"mean_token_accuracy": 0.215890771150589,
"num_tokens": 12885543.0,
"step": 6580
},
{
"entropy": 5.857462596893311,
"epoch": 6.545997016409746,
"grad_norm": 1.2890625,
"learning_rate": 0.0001941363526908736,
"loss": 5.1381,
"mean_token_accuracy": 0.20982462614774705,
"num_tokens": 12895716.0,
"step": 6585
},
{
"entropy": 5.8829918384552,
"epoch": 6.5509696668324215,
"grad_norm": 1.2421875,
"learning_rate": 0.00019377203807301185,
"loss": 5.1632,
"mean_token_accuracy": 0.2150833487510681,
"num_tokens": 12905370.0,
"step": 6590
},
{
"entropy": 5.87492036819458,
"epoch": 6.555942317255097,
"grad_norm": 1.265625,
"learning_rate": 0.00019340796816308,
"loss": 5.2451,
"mean_token_accuracy": 0.20050168186426162,
"num_tokens": 12915679.0,
"step": 6595
},
{
"entropy": 5.950360679626465,
"epoch": 6.560914967677772,
"grad_norm": 1.3125,
"learning_rate": 0.00019304414405787768,
"loss": 5.2278,
"mean_token_accuracy": 0.20295738130807878,
"num_tokens": 12924519.0,
"step": 6600
},
{
"entropy": 5.934345483779907,
"epoch": 6.565887618100447,
"grad_norm": 1.4453125,
"learning_rate": 0.0001926805668534639,
"loss": 5.2028,
"mean_token_accuracy": 0.2060479298233986,
"num_tokens": 12934571.0,
"step": 6605
},
{
"entropy": 5.906139755249024,
"epoch": 6.5708602685231225,
"grad_norm": 1.3671875,
"learning_rate": 0.00019231723764515397,
"loss": 5.2363,
"mean_token_accuracy": 0.19877939969301223,
"num_tokens": 12943888.0,
"step": 6610
},
{
"entropy": 5.981120252609253,
"epoch": 6.575832918945798,
"grad_norm": 1.3125,
"learning_rate": 0.00019195415752751595,
"loss": 5.2287,
"mean_token_accuracy": 0.19940014332532882,
"num_tokens": 12954058.0,
"step": 6615
},
{
"entropy": 5.832480812072754,
"epoch": 6.580805569368473,
"grad_norm": 1.3125,
"learning_rate": 0.00019159132759436764,
"loss": 5.0357,
"mean_token_accuracy": 0.2192149505019188,
"num_tokens": 12963629.0,
"step": 6620
},
{
"entropy": 5.926331377029419,
"epoch": 6.585778219791148,
"grad_norm": 1.359375,
"learning_rate": 0.00019122874893877294,
"loss": 5.1231,
"mean_token_accuracy": 0.2118276834487915,
"num_tokens": 12973690.0,
"step": 6625
},
{
"entropy": 5.8872088432312015,
"epoch": 6.590750870213824,
"grad_norm": 1.2734375,
"learning_rate": 0.00019086642265303894,
"loss": 5.1317,
"mean_token_accuracy": 0.20181286036968232,
"num_tokens": 12983186.0,
"step": 6630
},
{
"entropy": 5.802744579315186,
"epoch": 6.595723520636499,
"grad_norm": 1.2734375,
"learning_rate": 0.0001905043498287124,
"loss": 5.1316,
"mean_token_accuracy": 0.20310148894786834,
"num_tokens": 12992516.0,
"step": 6635
},
{
"entropy": 5.8013489723205565,
"epoch": 6.600696171059175,
"grad_norm": 1.28125,
"learning_rate": 0.0001901425315565764,
"loss": 5.1799,
"mean_token_accuracy": 0.20374116450548171,
"num_tokens": 13001964.0,
"step": 6640
},
{
"entropy": 5.909906005859375,
"epoch": 6.60566882148185,
"grad_norm": 1.3046875,
"learning_rate": 0.00018978096892664726,
"loss": 5.2342,
"mean_token_accuracy": 0.20258548557758332,
"num_tokens": 13011718.0,
"step": 6645
},
{
"entropy": 5.863251209259033,
"epoch": 6.6106414719045254,
"grad_norm": 1.1953125,
"learning_rate": 0.000189419663028171,
"loss": 5.1464,
"mean_token_accuracy": 0.20894574522972106,
"num_tokens": 13022946.0,
"step": 6650
},
{
"entropy": 5.892551851272583,
"epoch": 6.615614122327201,
"grad_norm": 1.328125,
"learning_rate": 0.00018905861494962034,
"loss": 5.214,
"mean_token_accuracy": 0.19890855848789216,
"num_tokens": 13032555.0,
"step": 6655
},
{
"entropy": 5.875191974639892,
"epoch": 6.620586772749876,
"grad_norm": 1.4140625,
"learning_rate": 0.00018869782577869132,
"loss": 5.1823,
"mean_token_accuracy": 0.20856199711561202,
"num_tokens": 13042447.0,
"step": 6660
},
{
"entropy": 5.71266450881958,
"epoch": 6.625559423172551,
"grad_norm": 1.4140625,
"learning_rate": 0.00018833729660229997,
"loss": 4.9782,
"mean_token_accuracy": 0.2298370346426964,
"num_tokens": 13052210.0,
"step": 6665
},
{
"entropy": 5.889480781555176,
"epoch": 6.6305320735952264,
"grad_norm": 1.390625,
"learning_rate": 0.00018797702850657889,
"loss": 5.2294,
"mean_token_accuracy": 0.19894976317882537,
"num_tokens": 13061701.0,
"step": 6670
},
{
"entropy": 5.8454358100891115,
"epoch": 6.635504724017902,
"grad_norm": 1.4296875,
"learning_rate": 0.0001876170225768744,
"loss": 5.1799,
"mean_token_accuracy": 0.20786878764629363,
"num_tokens": 13070801.0,
"step": 6675
},
{
"entropy": 5.880680847167969,
"epoch": 6.640477374440577,
"grad_norm": 1.2890625,
"learning_rate": 0.0001872572798977429,
"loss": 5.1294,
"mean_token_accuracy": 0.21084130555391312,
"num_tokens": 13080049.0,
"step": 6680
},
{
"entropy": 5.745874500274658,
"epoch": 6.645450024863252,
"grad_norm": 1.25,
"learning_rate": 0.0001868978015529476,
"loss": 4.9753,
"mean_token_accuracy": 0.2308308392763138,
"num_tokens": 13090019.0,
"step": 6685
},
{
"entropy": 5.97193078994751,
"epoch": 6.6504226752859275,
"grad_norm": 1.3984375,
"learning_rate": 0.00018653858862545565,
"loss": 5.2814,
"mean_token_accuracy": 0.1984146162867546,
"num_tokens": 13099735.0,
"step": 6690
},
{
"entropy": 5.940124034881592,
"epoch": 6.655395325708603,
"grad_norm": 1.296875,
"learning_rate": 0.00018617964219743428,
"loss": 5.2814,
"mean_token_accuracy": 0.20315442532300948,
"num_tokens": 13109261.0,
"step": 6695
},
{
"entropy": 5.89940996170044,
"epoch": 6.660367976131278,
"grad_norm": 1.28125,
"learning_rate": 0.00018582096335024813,
"loss": 5.182,
"mean_token_accuracy": 0.19951787889003753,
"num_tokens": 13119381.0,
"step": 6700
},
{
"entropy": 5.818733787536621,
"epoch": 6.665340626553953,
"grad_norm": 1.359375,
"learning_rate": 0.0001854625531644557,
"loss": 5.0845,
"mean_token_accuracy": 0.22000679969787598,
"num_tokens": 13130225.0,
"step": 6705
},
{
"entropy": 5.845886611938477,
"epoch": 6.6703132769766285,
"grad_norm": 1.234375,
"learning_rate": 0.00018510441271980588,
"loss": 5.076,
"mean_token_accuracy": 0.22190400809049607,
"num_tokens": 13140523.0,
"step": 6710
},
{
"entropy": 5.945285320281982,
"epoch": 6.675285927399304,
"grad_norm": 1.3515625,
"learning_rate": 0.00018474654309523508,
"loss": 5.1968,
"mean_token_accuracy": 0.20424776077270507,
"num_tokens": 13150328.0,
"step": 6715
},
{
"entropy": 5.89890341758728,
"epoch": 6.680258577821979,
"grad_norm": 1.296875,
"learning_rate": 0.00018438894536886396,
"loss": 5.2477,
"mean_token_accuracy": 0.1997586026787758,
"num_tokens": 13160095.0,
"step": 6720
},
{
"entropy": 5.909628820419312,
"epoch": 6.685231228244654,
"grad_norm": 1.171875,
"learning_rate": 0.00018403162061799395,
"loss": 5.1562,
"mean_token_accuracy": 0.20301248133182526,
"num_tokens": 13169731.0,
"step": 6725
},
{
"entropy": 5.879189920425415,
"epoch": 6.6902038786673295,
"grad_norm": 1.3671875,
"learning_rate": 0.00018367456991910403,
"loss": 5.1884,
"mean_token_accuracy": 0.20389173775911332,
"num_tokens": 13179065.0,
"step": 6730
},
{
"entropy": 5.936420965194702,
"epoch": 6.695176529090005,
"grad_norm": 1.3203125,
"learning_rate": 0.00018331779434784773,
"loss": 5.2684,
"mean_token_accuracy": 0.19789163917303085,
"num_tokens": 13188873.0,
"step": 6735
},
{
"entropy": 5.935892152786255,
"epoch": 6.70014917951268,
"grad_norm": 1.25,
"learning_rate": 0.00018296129497904955,
"loss": 5.2625,
"mean_token_accuracy": 0.19371768981218337,
"num_tokens": 13198996.0,
"step": 6740
},
{
"entropy": 5.747467088699341,
"epoch": 6.705121829935355,
"grad_norm": 1.1640625,
"learning_rate": 0.0001826050728867021,
"loss": 5.0044,
"mean_token_accuracy": 0.22516107857227324,
"num_tokens": 13208523.0,
"step": 6745
},
{
"entropy": 5.846177864074707,
"epoch": 6.7100944803580305,
"grad_norm": 1.53125,
"learning_rate": 0.00018224912914396253,
"loss": 5.1546,
"mean_token_accuracy": 0.2119359329342842,
"num_tokens": 13217488.0,
"step": 6750
},
{
"entropy": 5.904673624038696,
"epoch": 6.715067130780707,
"grad_norm": 1.296875,
"learning_rate": 0.00018189346482314945,
"loss": 5.194,
"mean_token_accuracy": 0.2024981901049614,
"num_tokens": 13226818.0,
"step": 6755
},
{
"entropy": 5.893219470977783,
"epoch": 6.720039781203381,
"grad_norm": 1.1328125,
"learning_rate": 0.00018153808099573976,
"loss": 5.1371,
"mean_token_accuracy": 0.2112319663167,
"num_tokens": 13237759.0,
"step": 6760
},
{
"entropy": 5.824959659576416,
"epoch": 6.725012431626057,
"grad_norm": 1.46875,
"learning_rate": 0.00018118297873236517,
"loss": 5.0583,
"mean_token_accuracy": 0.2167204961180687,
"num_tokens": 13247656.0,
"step": 6765
},
{
"entropy": 5.898874998092651,
"epoch": 6.729985082048732,
"grad_norm": 1.3203125,
"learning_rate": 0.00018082815910280925,
"loss": 5.2184,
"mean_token_accuracy": 0.1993486240506172,
"num_tokens": 13257055.0,
"step": 6770
},
{
"entropy": 5.924596691131592,
"epoch": 6.734957732471408,
"grad_norm": 1.21875,
"learning_rate": 0.00018047362317600424,
"loss": 5.2041,
"mean_token_accuracy": 0.20187289416790008,
"num_tokens": 13267606.0,
"step": 6775
},
{
"entropy": 5.942571401596069,
"epoch": 6.739930382894083,
"grad_norm": 1.4453125,
"learning_rate": 0.00018011937202002748,
"loss": 5.2518,
"mean_token_accuracy": 0.19588640332221985,
"num_tokens": 13276697.0,
"step": 6780
},
{
"entropy": 5.919096040725708,
"epoch": 6.744903033316758,
"grad_norm": 1.4296875,
"learning_rate": 0.00017976540670209856,
"loss": 5.1748,
"mean_token_accuracy": 0.20403988808393478,
"num_tokens": 13285150.0,
"step": 6785
},
{
"entropy": 5.869894790649414,
"epoch": 6.749875683739433,
"grad_norm": 1.4296875,
"learning_rate": 0.00017941172828857572,
"loss": 5.1787,
"mean_token_accuracy": 0.2006392613053322,
"num_tokens": 13294929.0,
"step": 6790
},
{
"entropy": 5.882103109359742,
"epoch": 6.754848334162109,
"grad_norm": 1.3125,
"learning_rate": 0.00017905833784495325,
"loss": 5.1511,
"mean_token_accuracy": 0.21073561757802964,
"num_tokens": 13305780.0,
"step": 6795
},
{
"entropy": 5.8468077182769775,
"epoch": 6.759820984584784,
"grad_norm": 1.28125,
"learning_rate": 0.00017870523643585769,
"loss": 5.1641,
"mean_token_accuracy": 0.20923701226711272,
"num_tokens": 13314526.0,
"step": 6800
},
{
"entropy": 5.976436281204224,
"epoch": 6.764793635007459,
"grad_norm": 1.265625,
"learning_rate": 0.00017835242512504463,
"loss": 5.1852,
"mean_token_accuracy": 0.21005151122808458,
"num_tokens": 13324034.0,
"step": 6805
},
{
"entropy": 5.886102962493896,
"epoch": 6.769766285430134,
"grad_norm": 1.296875,
"learning_rate": 0.00017799990497539608,
"loss": 5.1688,
"mean_token_accuracy": 0.2024229422211647,
"num_tokens": 13332535.0,
"step": 6810
},
{
"entropy": 5.900266551971436,
"epoch": 6.77473893585281,
"grad_norm": 1.3046875,
"learning_rate": 0.0001776476770489167,
"loss": 5.1927,
"mean_token_accuracy": 0.20756627023220062,
"num_tokens": 13342357.0,
"step": 6815
},
{
"entropy": 5.948041152954102,
"epoch": 6.779711586275485,
"grad_norm": 1.09375,
"learning_rate": 0.00017729574240673069,
"loss": 5.2863,
"mean_token_accuracy": 0.1945098266005516,
"num_tokens": 13352089.0,
"step": 6820
},
{
"entropy": 5.968145751953125,
"epoch": 6.78468423669816,
"grad_norm": 1.5078125,
"learning_rate": 0.00017694410210907906,
"loss": 5.2605,
"mean_token_accuracy": 0.19646771401166915,
"num_tokens": 13361653.0,
"step": 6825
},
{
"entropy": 5.862857294082642,
"epoch": 6.789656887120835,
"grad_norm": 1.3984375,
"learning_rate": 0.0001765927572153156,
"loss": 5.1883,
"mean_token_accuracy": 0.20721648633480072,
"num_tokens": 13370252.0,
"step": 6830
},
{
"entropy": 5.913563442230225,
"epoch": 6.794629537543511,
"grad_norm": 1.3828125,
"learning_rate": 0.0001762417087839046,
"loss": 5.1803,
"mean_token_accuracy": 0.2059689462184906,
"num_tokens": 13379935.0,
"step": 6835
},
{
"entropy": 5.892751407623291,
"epoch": 6.799602187966186,
"grad_norm": 1.2578125,
"learning_rate": 0.00017589095787241688,
"loss": 5.2247,
"mean_token_accuracy": 0.20067763030529023,
"num_tokens": 13389797.0,
"step": 6840
},
{
"entropy": 5.794259023666382,
"epoch": 6.804574838388861,
"grad_norm": 1.375,
"learning_rate": 0.0001755405055375272,
"loss": 5.0977,
"mean_token_accuracy": 0.21113835722208024,
"num_tokens": 13399750.0,
"step": 6845
},
{
"entropy": 5.8526939868927,
"epoch": 6.809547488811536,
"grad_norm": 1.2578125,
"learning_rate": 0.00017519035283501084,
"loss": 5.1904,
"mean_token_accuracy": 0.20704494416713715,
"num_tokens": 13409964.0,
"step": 6850
},
{
"entropy": 5.84862322807312,
"epoch": 6.814520139234212,
"grad_norm": 1.46875,
"learning_rate": 0.00017484050081974004,
"loss": 5.092,
"mean_token_accuracy": 0.21377474814653397,
"num_tokens": 13418714.0,
"step": 6855
},
{
"entropy": 5.981014680862427,
"epoch": 6.819492789656887,
"grad_norm": 1.1328125,
"learning_rate": 0.0001744909505456817,
"loss": 5.2644,
"mean_token_accuracy": 0.19701122343540192,
"num_tokens": 13429314.0,
"step": 6860
},
{
"entropy": 5.857673025131225,
"epoch": 6.824465440079562,
"grad_norm": 1.203125,
"learning_rate": 0.00017414170306589318,
"loss": 5.1448,
"mean_token_accuracy": 0.21436037719249726,
"num_tokens": 13440164.0,
"step": 6865
},
{
"entropy": 5.780084037780762,
"epoch": 6.829438090502237,
"grad_norm": 1.140625,
"learning_rate": 0.0001737927594325201,
"loss": 5.1219,
"mean_token_accuracy": 0.21427245289087296,
"num_tokens": 13450692.0,
"step": 6870
},
{
"entropy": 5.90819525718689,
"epoch": 6.834410740924913,
"grad_norm": 1.3671875,
"learning_rate": 0.00017344412069679257,
"loss": 5.2014,
"mean_token_accuracy": 0.20127684324979783,
"num_tokens": 13460363.0,
"step": 6875
},
{
"entropy": 5.864062261581421,
"epoch": 6.839383391347588,
"grad_norm": 1.2734375,
"learning_rate": 0.0001730957879090218,
"loss": 5.2763,
"mean_token_accuracy": 0.20301908850669861,
"num_tokens": 13469424.0,
"step": 6880
},
{
"entropy": 5.944316864013672,
"epoch": 6.844356041770263,
"grad_norm": 1.3984375,
"learning_rate": 0.00017274776211859778,
"loss": 5.2193,
"mean_token_accuracy": 0.2015284076333046,
"num_tokens": 13478573.0,
"step": 6885
},
{
"entropy": 5.88308277130127,
"epoch": 6.849328692192939,
"grad_norm": 1.234375,
"learning_rate": 0.00017240004437398554,
"loss": 5.1943,
"mean_token_accuracy": 0.2032455638051033,
"num_tokens": 13488488.0,
"step": 6890
},
{
"entropy": 5.833903694152832,
"epoch": 6.854301342615614,
"grad_norm": 1.359375,
"learning_rate": 0.0001720526357227218,
"loss": 5.1739,
"mean_token_accuracy": 0.21241303235292436,
"num_tokens": 13497766.0,
"step": 6895
},
{
"entropy": 5.960533475875854,
"epoch": 6.85927399303829,
"grad_norm": 1.5078125,
"learning_rate": 0.00017170553721141253,
"loss": 5.2531,
"mean_token_accuracy": 0.1900121048092842,
"num_tokens": 13507451.0,
"step": 6900
},
{
"entropy": 5.838674402236938,
"epoch": 6.864246643460965,
"grad_norm": 1.3671875,
"learning_rate": 0.0001713587498857289,
"loss": 5.1067,
"mean_token_accuracy": 0.2122770816087723,
"num_tokens": 13516627.0,
"step": 6905
},
{
"entropy": 5.803294086456299,
"epoch": 6.86921929388364,
"grad_norm": 1.2109375,
"learning_rate": 0.00017101227479040498,
"loss": 5.0474,
"mean_token_accuracy": 0.22415940165519715,
"num_tokens": 13528018.0,
"step": 6910
},
{
"entropy": 5.903308773040772,
"epoch": 6.8741919443063155,
"grad_norm": 1.171875,
"learning_rate": 0.00017066611296923424,
"loss": 5.1619,
"mean_token_accuracy": 0.20667506903409957,
"num_tokens": 13538107.0,
"step": 6915
},
{
"entropy": 5.940898752212524,
"epoch": 6.879164594728991,
"grad_norm": 1.265625,
"learning_rate": 0.00017032026546506595,
"loss": 5.2112,
"mean_token_accuracy": 0.1950005918741226,
"num_tokens": 13546983.0,
"step": 6920
},
{
"entropy": 5.877144241333008,
"epoch": 6.884137245151666,
"grad_norm": 1.25,
"learning_rate": 0.00016997473331980295,
"loss": 5.2591,
"mean_token_accuracy": 0.20484060645103455,
"num_tokens": 13557481.0,
"step": 6925
},
{
"entropy": 5.841855144500732,
"epoch": 6.889109895574341,
"grad_norm": 1.1875,
"learning_rate": 0.0001696295175743977,
"loss": 5.1749,
"mean_token_accuracy": 0.2132147341966629,
"num_tokens": 13567618.0,
"step": 6930
},
{
"entropy": 5.675601434707642,
"epoch": 6.8940825459970165,
"grad_norm": 1.3203125,
"learning_rate": 0.00016928461926884954,
"loss": 4.9068,
"mean_token_accuracy": 0.23901850134134292,
"num_tokens": 13578610.0,
"step": 6935
},
{
"entropy": 5.967743444442749,
"epoch": 6.899055196419692,
"grad_norm": 1.3671875,
"learning_rate": 0.00016894003944220167,
"loss": 5.2495,
"mean_token_accuracy": 0.19981561601161957,
"num_tokens": 13588359.0,
"step": 6940
},
{
"entropy": 5.898694229125977,
"epoch": 6.904027846842367,
"grad_norm": 1.3125,
"learning_rate": 0.0001685957791325375,
"loss": 5.1803,
"mean_token_accuracy": 0.20766668468713761,
"num_tokens": 13598792.0,
"step": 6945
},
{
"entropy": 5.923589515686035,
"epoch": 6.909000497265042,
"grad_norm": 1.3671875,
"learning_rate": 0.0001682518393769782,
"loss": 5.2213,
"mean_token_accuracy": 0.20610079318284988,
"num_tokens": 13607983.0,
"step": 6950
},
{
"entropy": 5.875217247009277,
"epoch": 6.9139731476877175,
"grad_norm": 1.3046875,
"learning_rate": 0.00016790822121167913,
"loss": 5.1138,
"mean_token_accuracy": 0.21502544581890107,
"num_tokens": 13617827.0,
"step": 6955
},
{
"entropy": 5.844430685043335,
"epoch": 6.918945798110393,
"grad_norm": 1.359375,
"learning_rate": 0.00016756492567182674,
"loss": 5.0515,
"mean_token_accuracy": 0.21687419414520265,
"num_tokens": 13627565.0,
"step": 6960
},
{
"entropy": 5.852701997756958,
"epoch": 6.923918448533068,
"grad_norm": 1.234375,
"learning_rate": 0.00016722195379163568,
"loss": 5.1868,
"mean_token_accuracy": 0.21222616583108903,
"num_tokens": 13637828.0,
"step": 6965
},
{
"entropy": 5.86072473526001,
"epoch": 6.928891098955743,
"grad_norm": 1.3671875,
"learning_rate": 0.00016687930660434528,
"loss": 5.1733,
"mean_token_accuracy": 0.20750919431447984,
"num_tokens": 13647757.0,
"step": 6970
},
{
"entropy": 5.929159450531006,
"epoch": 6.9338637493784185,
"grad_norm": 1.40625,
"learning_rate": 0.00016653698514221694,
"loss": 5.2528,
"mean_token_accuracy": 0.19694293886423112,
"num_tokens": 13656309.0,
"step": 6975
},
{
"entropy": 5.802533578872681,
"epoch": 6.938836399801094,
"grad_norm": 1.3125,
"learning_rate": 0.0001661949904365308,
"loss": 5.024,
"mean_token_accuracy": 0.21788535714149476,
"num_tokens": 13666894.0,
"step": 6980
},
{
"entropy": 5.9023772239685055,
"epoch": 6.943809050223769,
"grad_norm": 1.4453125,
"learning_rate": 0.0001658533235175824,
"loss": 5.2279,
"mean_token_accuracy": 0.19733551144599915,
"num_tokens": 13675753.0,
"step": 6985
},
{
"entropy": 5.9096887588500975,
"epoch": 6.948781700646444,
"grad_norm": 1.375,
"learning_rate": 0.00016551198541468,
"loss": 5.105,
"mean_token_accuracy": 0.21092984527349473,
"num_tokens": 13685734.0,
"step": 6990
},
{
"entropy": 5.877429056167602,
"epoch": 6.9537543510691195,
"grad_norm": 1.3984375,
"learning_rate": 0.00016517097715614115,
"loss": 5.217,
"mean_token_accuracy": 0.20380515307188035,
"num_tokens": 13695628.0,
"step": 6995
},
{
"entropy": 5.781128215789795,
"epoch": 6.958727001491795,
"grad_norm": 1.0859375,
"learning_rate": 0.00016483029976928958,
"loss": 5.1826,
"mean_token_accuracy": 0.21405284553766252,
"num_tokens": 13706558.0,
"step": 7000
},
{
"epoch": 6.958727001491795,
"eval_entropy": 5.845373371229743,
"eval_loss": 6.479496955871582,
"eval_mean_token_accuracy": 0.1476145630556455,
"eval_num_tokens": 13706558.0,
"eval_runtime": 2.5463,
"eval_samples_per_second": 1360.791,
"eval_steps_per_second": 170.443,
"step": 7000
},
{
"entropy": 5.917164373397827,
"epoch": 6.96369965191447,
"grad_norm": 1.46875,
"learning_rate": 0.0001644899542804526,
"loss": 5.1491,
"mean_token_accuracy": 0.21135603934526442,
"num_tokens": 13715298.0,
"step": 7005
},
{
"entropy": 5.913910913467407,
"epoch": 6.968672302337145,
"grad_norm": 1.3515625,
"learning_rate": 0.00016414994171495728,
"loss": 5.2545,
"mean_token_accuracy": 0.1995296895503998,
"num_tokens": 13724705.0,
"step": 7010
},
{
"entropy": 5.913446235656738,
"epoch": 6.973644952759821,
"grad_norm": 1.4140625,
"learning_rate": 0.00016381026309712787,
"loss": 5.1793,
"mean_token_accuracy": 0.20715128928422927,
"num_tokens": 13734023.0,
"step": 7015
},
{
"entropy": 5.824751234054565,
"epoch": 6.978617603182496,
"grad_norm": 1.34375,
"learning_rate": 0.00016347091945028273,
"loss": 5.193,
"mean_token_accuracy": 0.20813450068235398,
"num_tokens": 13744634.0,
"step": 7020
},
{
"entropy": 5.874810838699341,
"epoch": 6.983590253605172,
"grad_norm": 1.2109375,
"learning_rate": 0.0001631319117967307,
"loss": 5.2607,
"mean_token_accuracy": 0.1963325023651123,
"num_tokens": 13755420.0,
"step": 7025
},
{
"entropy": 5.916146659851075,
"epoch": 6.988562904027847,
"grad_norm": 1.4375,
"learning_rate": 0.00016279324115776886,
"loss": 5.3043,
"mean_token_accuracy": 0.19515351504087447,
"num_tokens": 13766541.0,
"step": 7030
},
{
"entropy": 5.951808500289917,
"epoch": 6.993535554450522,
"grad_norm": 1.5078125,
"learning_rate": 0.0001624549085536785,
"loss": 5.2409,
"mean_token_accuracy": 0.2009877920150757,
"num_tokens": 13775954.0,
"step": 7035
},
{
"entropy": 5.906681299209595,
"epoch": 6.998508204873198,
"grad_norm": 1.296875,
"learning_rate": 0.00016211691500372304,
"loss": 5.2354,
"mean_token_accuracy": 0.20184094607830047,
"num_tokens": 13787615.0,
"step": 7040
},
{
"entropy": 5.943604681226942,
"epoch": 7.0029835902536055,
"grad_norm": 1.25,
"learning_rate": 0.00016177926152614414,
"loss": 5.1966,
"mean_token_accuracy": 0.19751206702656215,
"num_tokens": 13796765.0,
"step": 7045
},
{
"entropy": 5.9124898433685305,
"epoch": 7.007956240676281,
"grad_norm": 1.34375,
"learning_rate": 0.00016144194913815887,
"loss": 5.0643,
"mean_token_accuracy": 0.21525945216417314,
"num_tokens": 13805980.0,
"step": 7050
},
{
"entropy": 5.909930992126465,
"epoch": 7.012928891098956,
"grad_norm": 1.2890625,
"learning_rate": 0.0001611049788559572,
"loss": 5.1088,
"mean_token_accuracy": 0.21196259111166,
"num_tokens": 13815898.0,
"step": 7055
},
{
"entropy": 5.855438566207885,
"epoch": 7.017901541521631,
"grad_norm": 1.34375,
"learning_rate": 0.00016076835169469796,
"loss": 4.9834,
"mean_token_accuracy": 0.22114828675985337,
"num_tokens": 13826359.0,
"step": 7060
},
{
"entropy": 5.840119218826294,
"epoch": 7.0228741919443065,
"grad_norm": 1.03125,
"learning_rate": 0.00016043206866850652,
"loss": 4.9496,
"mean_token_accuracy": 0.22856925576925277,
"num_tokens": 13836225.0,
"step": 7065
},
{
"entropy": 5.820138883590698,
"epoch": 7.027846842366982,
"grad_norm": 1.5234375,
"learning_rate": 0.0001600961307904717,
"loss": 4.9565,
"mean_token_accuracy": 0.2291437119245529,
"num_tokens": 13846072.0,
"step": 7070
},
{
"entropy": 5.946877098083496,
"epoch": 7.032819492789657,
"grad_norm": 1.3359375,
"learning_rate": 0.00015976053907264208,
"loss": 5.1081,
"mean_token_accuracy": 0.20708346217870713,
"num_tokens": 13854984.0,
"step": 7075
},
{
"entropy": 5.771928215026856,
"epoch": 7.037792143212332,
"grad_norm": 1.3359375,
"learning_rate": 0.00015942529452602378,
"loss": 4.8874,
"mean_token_accuracy": 0.23042453378438948,
"num_tokens": 13864794.0,
"step": 7080
},
{
"entropy": 5.892866086959839,
"epoch": 7.0427647936350075,
"grad_norm": 1.2421875,
"learning_rate": 0.00015909039816057674,
"loss": 5.08,
"mean_token_accuracy": 0.20991730242967604,
"num_tokens": 13874042.0,
"step": 7085
},
{
"entropy": 5.833378410339355,
"epoch": 7.047737444057683,
"grad_norm": 1.234375,
"learning_rate": 0.00015875585098521218,
"loss": 5.0272,
"mean_token_accuracy": 0.2188786342740059,
"num_tokens": 13884884.0,
"step": 7090
},
{
"entropy": 5.8654145240783695,
"epoch": 7.052710094480358,
"grad_norm": 1.3671875,
"learning_rate": 0.00015842165400778942,
"loss": 5.0272,
"mean_token_accuracy": 0.2194517120718956,
"num_tokens": 13894837.0,
"step": 7095
},
{
"entropy": 5.928678417205811,
"epoch": 7.057682744903033,
"grad_norm": 1.21875,
"learning_rate": 0.0001580878082351123,
"loss": 5.0855,
"mean_token_accuracy": 0.19980716109275817,
"num_tokens": 13905723.0,
"step": 7100
},
{
"entropy": 5.826294231414795,
"epoch": 7.0626553953257085,
"grad_norm": 1.265625,
"learning_rate": 0.00015775431467292733,
"loss": 4.9953,
"mean_token_accuracy": 0.21568144410848616,
"num_tokens": 13915993.0,
"step": 7105
},
{
"entropy": 5.845410299301148,
"epoch": 7.067628045748384,
"grad_norm": 1.1875,
"learning_rate": 0.00015742117432591935,
"loss": 4.9961,
"mean_token_accuracy": 0.21465601176023483,
"num_tokens": 13926917.0,
"step": 7110
},
{
"entropy": 5.905386257171631,
"epoch": 7.072600696171059,
"grad_norm": 1.421875,
"learning_rate": 0.00015708838819770922,
"loss": 5.0597,
"mean_token_accuracy": 0.2073136955499649,
"num_tokens": 13936001.0,
"step": 7115
},
{
"entropy": 5.893727779388428,
"epoch": 7.077573346593734,
"grad_norm": 1.265625,
"learning_rate": 0.000156755957290851,
"loss": 5.0071,
"mean_token_accuracy": 0.21896337121725082,
"num_tokens": 13945769.0,
"step": 7120
},
{
"entropy": 5.9003698348999025,
"epoch": 7.0825459970164095,
"grad_norm": 1.328125,
"learning_rate": 0.00015642388260682816,
"loss": 5.059,
"mean_token_accuracy": 0.21080808341503143,
"num_tokens": 13955433.0,
"step": 7125
},
{
"entropy": 5.925618886947632,
"epoch": 7.087518647439085,
"grad_norm": 1.28125,
"learning_rate": 0.00015609216514605136,
"loss": 5.1938,
"mean_token_accuracy": 0.21031879931688308,
"num_tokens": 13965581.0,
"step": 7130
},
{
"entropy": 5.858029413223266,
"epoch": 7.09249129786176,
"grad_norm": 1.359375,
"learning_rate": 0.00015576080590785503,
"loss": 4.9428,
"mean_token_accuracy": 0.22965606898069382,
"num_tokens": 13975913.0,
"step": 7135
},
{
"entropy": 5.824290943145752,
"epoch": 7.097463948284435,
"grad_norm": 1.4140625,
"learning_rate": 0.00015542980589049413,
"loss": 4.9922,
"mean_token_accuracy": 0.2217685341835022,
"num_tokens": 13984626.0,
"step": 7140
},
{
"entropy": 5.823685264587402,
"epoch": 7.1024365987071105,
"grad_norm": 1.4140625,
"learning_rate": 0.00015509916609114188,
"loss": 4.9152,
"mean_token_accuracy": 0.22359603196382521,
"num_tokens": 13993594.0,
"step": 7145
},
{
"entropy": 5.737445497512818,
"epoch": 7.107409249129786,
"grad_norm": 1.3359375,
"learning_rate": 0.00015476888750588585,
"loss": 4.9461,
"mean_token_accuracy": 0.22765136659145355,
"num_tokens": 14004020.0,
"step": 7150
},
{
"entropy": 5.870018672943115,
"epoch": 7.112381899552461,
"grad_norm": 1.203125,
"learning_rate": 0.00015443897112972572,
"loss": 5.0758,
"mean_token_accuracy": 0.21539842933416367,
"num_tokens": 14014614.0,
"step": 7155
},
{
"entropy": 5.809611225128174,
"epoch": 7.117354549975137,
"grad_norm": 1.328125,
"learning_rate": 0.00015410941795657008,
"loss": 5.0034,
"mean_token_accuracy": 0.2205087274312973,
"num_tokens": 14023756.0,
"step": 7160
},
{
"entropy": 5.891222333908081,
"epoch": 7.122327200397812,
"grad_norm": 1.2265625,
"learning_rate": 0.00015378022897923277,
"loss": 5.0262,
"mean_token_accuracy": 0.21592807471752168,
"num_tokens": 14034026.0,
"step": 7165
},
{
"entropy": 5.8637402057647705,
"epoch": 7.127299850820488,
"grad_norm": 1.2578125,
"learning_rate": 0.00015345140518943103,
"loss": 4.9971,
"mean_token_accuracy": 0.2207915723323822,
"num_tokens": 14043362.0,
"step": 7170
},
{
"entropy": 5.859983730316162,
"epoch": 7.132272501243163,
"grad_norm": 1.3125,
"learning_rate": 0.0001531229475777818,
"loss": 5.047,
"mean_token_accuracy": 0.21329281032085418,
"num_tokens": 14052789.0,
"step": 7175
},
{
"entropy": 5.845340633392334,
"epoch": 7.137245151665838,
"grad_norm": 1.359375,
"learning_rate": 0.0001527948571337986,
"loss": 5.0966,
"mean_token_accuracy": 0.21270923018455506,
"num_tokens": 14061750.0,
"step": 7180
},
{
"entropy": 5.828740072250366,
"epoch": 7.142217802088513,
"grad_norm": 1.171875,
"learning_rate": 0.0001524671348458892,
"loss": 4.9932,
"mean_token_accuracy": 0.22220800071954727,
"num_tokens": 14071891.0,
"step": 7185
},
{
"entropy": 5.955962324142456,
"epoch": 7.147190452511189,
"grad_norm": 1.3515625,
"learning_rate": 0.00015213978170135187,
"loss": 5.1038,
"mean_token_accuracy": 0.2052909478545189,
"num_tokens": 14080854.0,
"step": 7190
},
{
"entropy": 5.886085081100464,
"epoch": 7.152163102933864,
"grad_norm": 1.3984375,
"learning_rate": 0.00015181279868637312,
"loss": 5.1127,
"mean_token_accuracy": 0.21340456157922744,
"num_tokens": 14090120.0,
"step": 7195
},
{
"entropy": 5.895999097824097,
"epoch": 7.157135753356539,
"grad_norm": 1.4375,
"learning_rate": 0.00015148618678602433,
"loss": 5.105,
"mean_token_accuracy": 0.21169700473546982,
"num_tokens": 14100582.0,
"step": 7200
},
{
"entropy": 5.838264226913452,
"epoch": 7.162108403779214,
"grad_norm": 1.3515625,
"learning_rate": 0.00015115994698425872,
"loss": 5.0234,
"mean_token_accuracy": 0.21789498180150985,
"num_tokens": 14110710.0,
"step": 7205
},
{
"entropy": 5.857951974868774,
"epoch": 7.16708105420189,
"grad_norm": 1.40625,
"learning_rate": 0.0001508340802639087,
"loss": 4.9144,
"mean_token_accuracy": 0.232724229991436,
"num_tokens": 14121007.0,
"step": 7210
},
{
"entropy": 5.9006446361541744,
"epoch": 7.172053704624565,
"grad_norm": 1.2734375,
"learning_rate": 0.00015050858760668246,
"loss": 5.0015,
"mean_token_accuracy": 0.22241006195545196,
"num_tokens": 14131225.0,
"step": 7215
},
{
"entropy": 5.8567383766174315,
"epoch": 7.17702635504724,
"grad_norm": 1.53125,
"learning_rate": 0.00015018346999316173,
"loss": 4.9748,
"mean_token_accuracy": 0.22460905611515045,
"num_tokens": 14140058.0,
"step": 7220
},
{
"entropy": 5.9116803169250485,
"epoch": 7.181999005469915,
"grad_norm": 1.3046875,
"learning_rate": 0.000149858728402798,
"loss": 5.108,
"mean_token_accuracy": 0.20463975220918657,
"num_tokens": 14151738.0,
"step": 7225
},
{
"entropy": 5.843396329879761,
"epoch": 7.186971655892591,
"grad_norm": 1.28125,
"learning_rate": 0.00014953436381390998,
"loss": 5.0275,
"mean_token_accuracy": 0.22032514959573746,
"num_tokens": 14161145.0,
"step": 7230
},
{
"entropy": 5.8099024295806885,
"epoch": 7.191944306315266,
"grad_norm": 1.3984375,
"learning_rate": 0.00014921037720368085,
"loss": 4.9926,
"mean_token_accuracy": 0.22288767397403716,
"num_tokens": 14169788.0,
"step": 7235
},
{
"entropy": 5.908723211288452,
"epoch": 7.196916956737941,
"grad_norm": 1.4609375,
"learning_rate": 0.00014888676954815484,
"loss": 5.0537,
"mean_token_accuracy": 0.20952604413032533,
"num_tokens": 14178695.0,
"step": 7240
},
{
"entropy": 5.902237319946289,
"epoch": 7.201889607160616,
"grad_norm": 1.3671875,
"learning_rate": 0.0001485635418222347,
"loss": 4.9888,
"mean_token_accuracy": 0.22431717664003373,
"num_tokens": 14186989.0,
"step": 7245
},
{
"entropy": 5.8534644603729244,
"epoch": 7.206862257583292,
"grad_norm": 1.3515625,
"learning_rate": 0.00014824069499967863,
"loss": 5.048,
"mean_token_accuracy": 0.21191126555204393,
"num_tokens": 14196980.0,
"step": 7250
},
{
"entropy": 5.880635786056518,
"epoch": 7.211834908005967,
"grad_norm": 1.4375,
"learning_rate": 0.0001479182300530971,
"loss": 5.0107,
"mean_token_accuracy": 0.21580293476581575,
"num_tokens": 14206066.0,
"step": 7255
},
{
"entropy": 5.912382888793945,
"epoch": 7.216807558428642,
"grad_norm": 1.4296875,
"learning_rate": 0.00014759614795395044,
"loss": 5.0773,
"mean_token_accuracy": 0.21105777621269226,
"num_tokens": 14215106.0,
"step": 7260
},
{
"entropy": 5.934971761703491,
"epoch": 7.221780208851317,
"grad_norm": 1.3828125,
"learning_rate": 0.00014727444967254527,
"loss": 5.0906,
"mean_token_accuracy": 0.21287117898464203,
"num_tokens": 14224114.0,
"step": 7265
},
{
"entropy": 5.914045619964599,
"epoch": 7.226752859273993,
"grad_norm": 1.3046875,
"learning_rate": 0.0001469531361780322,
"loss": 5.0884,
"mean_token_accuracy": 0.21349845826625824,
"num_tokens": 14233489.0,
"step": 7270
},
{
"entropy": 5.935693740844727,
"epoch": 7.231725509696668,
"grad_norm": 1.328125,
"learning_rate": 0.00014663220843840266,
"loss": 5.1457,
"mean_token_accuracy": 0.20920446366071702,
"num_tokens": 14243725.0,
"step": 7275
},
{
"entropy": 5.926391983032227,
"epoch": 7.236698160119343,
"grad_norm": 1.453125,
"learning_rate": 0.0001463116674204855,
"loss": 5.0699,
"mean_token_accuracy": 0.21667914241552352,
"num_tokens": 14252987.0,
"step": 7280
},
{
"entropy": 5.86472635269165,
"epoch": 7.241670810542018,
"grad_norm": 1.40625,
"learning_rate": 0.00014599151408994503,
"loss": 5.0594,
"mean_token_accuracy": 0.22167654186487198,
"num_tokens": 14263010.0,
"step": 7285
},
{
"entropy": 5.826028442382812,
"epoch": 7.246643460964695,
"grad_norm": 1.375,
"learning_rate": 0.00014567174941127752,
"loss": 5.028,
"mean_token_accuracy": 0.22094755321741105,
"num_tokens": 14273064.0,
"step": 7290
},
{
"entropy": 5.948786401748658,
"epoch": 7.25161611138737,
"grad_norm": 1.3671875,
"learning_rate": 0.00014535237434780814,
"loss": 5.1586,
"mean_token_accuracy": 0.20112718641757965,
"num_tokens": 14282374.0,
"step": 7295
},
{
"entropy": 5.912295246124268,
"epoch": 7.256588761810045,
"grad_norm": 1.3984375,
"learning_rate": 0.0001450333898616885,
"loss": 5.0622,
"mean_token_accuracy": 0.2180670127272606,
"num_tokens": 14291838.0,
"step": 7300
},
{
"entropy": 5.902403688430786,
"epoch": 7.26156141223272,
"grad_norm": 1.375,
"learning_rate": 0.00014471479691389343,
"loss": 5.0157,
"mean_token_accuracy": 0.22334802746772767,
"num_tokens": 14302642.0,
"step": 7305
},
{
"entropy": 5.7599140167236325,
"epoch": 7.266534062655396,
"grad_norm": 1.359375,
"learning_rate": 0.00014439659646421843,
"loss": 4.92,
"mean_token_accuracy": 0.23043097406625748,
"num_tokens": 14313829.0,
"step": 7310
},
{
"entropy": 5.893299388885498,
"epoch": 7.271506713078071,
"grad_norm": 1.3515625,
"learning_rate": 0.0001440787894712764,
"loss": 5.0198,
"mean_token_accuracy": 0.2156640335917473,
"num_tokens": 14323296.0,
"step": 7315
},
{
"entropy": 5.903748369216919,
"epoch": 7.276479363500746,
"grad_norm": 1.3125,
"learning_rate": 0.0001437613768924948,
"loss": 5.0754,
"mean_token_accuracy": 0.20699457824230194,
"num_tokens": 14332961.0,
"step": 7320
},
{
"entropy": 5.905309581756592,
"epoch": 7.281452013923421,
"grad_norm": 1.4375,
"learning_rate": 0.00014344435968411317,
"loss": 5.0508,
"mean_token_accuracy": 0.2159063383936882,
"num_tokens": 14342448.0,
"step": 7325
},
{
"entropy": 5.873162174224854,
"epoch": 7.286424664346097,
"grad_norm": 1.328125,
"learning_rate": 0.0001431277388011796,
"loss": 5.0393,
"mean_token_accuracy": 0.21705288141965867,
"num_tokens": 14351491.0,
"step": 7330
},
{
"entropy": 5.885784053802491,
"epoch": 7.291397314768772,
"grad_norm": 1.3125,
"learning_rate": 0.00014281151519754853,
"loss": 5.073,
"mean_token_accuracy": 0.2149481251835823,
"num_tokens": 14360583.0,
"step": 7335
},
{
"entropy": 5.869936418533325,
"epoch": 7.296369965191447,
"grad_norm": 1.4609375,
"learning_rate": 0.0001424956898258774,
"loss": 5.0619,
"mean_token_accuracy": 0.2089579537510872,
"num_tokens": 14370388.0,
"step": 7340
},
{
"entropy": 5.9464024066925045,
"epoch": 7.301342615614122,
"grad_norm": 1.4140625,
"learning_rate": 0.00014218026363762383,
"loss": 5.1022,
"mean_token_accuracy": 0.2150563821196556,
"num_tokens": 14380245.0,
"step": 7345
},
{
"entropy": 5.904611301422119,
"epoch": 7.306315266036798,
"grad_norm": 1.2890625,
"learning_rate": 0.0001418652375830431,
"loss": 5.0398,
"mean_token_accuracy": 0.21168756783008574,
"num_tokens": 14390820.0,
"step": 7350
},
{
"entropy": 5.86742262840271,
"epoch": 7.311287916459473,
"grad_norm": 1.3984375,
"learning_rate": 0.00014155061261118487,
"loss": 4.9596,
"mean_token_accuracy": 0.22206927388906478,
"num_tokens": 14400072.0,
"step": 7355
},
{
"entropy": 5.867152738571167,
"epoch": 7.316260566882148,
"grad_norm": 1.3671875,
"learning_rate": 0.00014123638966989055,
"loss": 5.0077,
"mean_token_accuracy": 0.21573195457458497,
"num_tokens": 14410368.0,
"step": 7360
},
{
"entropy": 5.844783926010132,
"epoch": 7.321233217304823,
"grad_norm": 1.3125,
"learning_rate": 0.0001409225697057904,
"loss": 4.9771,
"mean_token_accuracy": 0.2274682939052582,
"num_tokens": 14420053.0,
"step": 7365
},
{
"entropy": 5.912103366851807,
"epoch": 7.326205867727499,
"grad_norm": 1.3359375,
"learning_rate": 0.0001406091536643006,
"loss": 5.1246,
"mean_token_accuracy": 0.20333026498556137,
"num_tokens": 14430417.0,
"step": 7370
},
{
"entropy": 5.919491100311279,
"epoch": 7.331178518150174,
"grad_norm": 1.1875,
"learning_rate": 0.0001402961424896206,
"loss": 5.1074,
"mean_token_accuracy": 0.21068546324968337,
"num_tokens": 14441412.0,
"step": 7375
},
{
"entropy": 5.897611474990844,
"epoch": 7.336151168572849,
"grad_norm": 1.3046875,
"learning_rate": 0.00013998353712473005,
"loss": 5.024,
"mean_token_accuracy": 0.2191637471318245,
"num_tokens": 14451871.0,
"step": 7380
},
{
"entropy": 5.830190706253052,
"epoch": 7.341123818995524,
"grad_norm": 1.40625,
"learning_rate": 0.00013967133851138603,
"loss": 5.0269,
"mean_token_accuracy": 0.2091280549764633,
"num_tokens": 14461607.0,
"step": 7385
},
{
"entropy": 5.869090366363525,
"epoch": 7.3460964694182,
"grad_norm": 1.265625,
"learning_rate": 0.00013935954759012036,
"loss": 5.0054,
"mean_token_accuracy": 0.22219265550374984,
"num_tokens": 14470761.0,
"step": 7390
},
{
"entropy": 5.964658260345459,
"epoch": 7.351069119840875,
"grad_norm": 1.328125,
"learning_rate": 0.0001390481653002365,
"loss": 5.1511,
"mean_token_accuracy": 0.20993491709232331,
"num_tokens": 14480668.0,
"step": 7395
},
{
"entropy": 5.881209135055542,
"epoch": 7.35604177026355,
"grad_norm": 1.359375,
"learning_rate": 0.0001387371925798069,
"loss": 5.11,
"mean_token_accuracy": 0.20688958317041398,
"num_tokens": 14489120.0,
"step": 7400
},
{
"entropy": 5.89217324256897,
"epoch": 7.361014420686225,
"grad_norm": 1.328125,
"learning_rate": 0.00013842663036567026,
"loss": 5.1104,
"mean_token_accuracy": 0.20964860469102858,
"num_tokens": 14498417.0,
"step": 7405
},
{
"entropy": 5.801281070709228,
"epoch": 7.365987071108901,
"grad_norm": 1.40625,
"learning_rate": 0.00013811647959342834,
"loss": 4.9215,
"mean_token_accuracy": 0.2315659448504448,
"num_tokens": 14507386.0,
"step": 7410
},
{
"entropy": 5.8809069156646725,
"epoch": 7.370959721531577,
"grad_norm": 1.3046875,
"learning_rate": 0.00013780674119744375,
"loss": 5.0531,
"mean_token_accuracy": 0.21399638056755066,
"num_tokens": 14517151.0,
"step": 7415
},
{
"entropy": 5.892367744445801,
"epoch": 7.375932371954252,
"grad_norm": 1.515625,
"learning_rate": 0.0001374974161108364,
"loss": 5.047,
"mean_token_accuracy": 0.21316026598215104,
"num_tokens": 14526581.0,
"step": 7420
},
{
"entropy": 5.913675308227539,
"epoch": 7.380905022376927,
"grad_norm": 1.1484375,
"learning_rate": 0.0001371885052654812,
"loss": 5.1137,
"mean_token_accuracy": 0.20837811529636383,
"num_tokens": 14536830.0,
"step": 7425
},
{
"entropy": 5.767458963394165,
"epoch": 7.3858776727996025,
"grad_norm": 1.28125,
"learning_rate": 0.0001368800095920053,
"loss": 4.8302,
"mean_token_accuracy": 0.24243755936622619,
"num_tokens": 14547156.0,
"step": 7430
},
{
"entropy": 5.927903890609741,
"epoch": 7.390850323222278,
"grad_norm": 1.1640625,
"learning_rate": 0.0001365719300197848,
"loss": 5.0798,
"mean_token_accuracy": 0.20889230221509933,
"num_tokens": 14556611.0,
"step": 7435
},
{
"entropy": 5.869841718673706,
"epoch": 7.395822973644953,
"grad_norm": 1.296875,
"learning_rate": 0.00013626426747694254,
"loss": 5.0371,
"mean_token_accuracy": 0.2155077114701271,
"num_tokens": 14565629.0,
"step": 7440
},
{
"entropy": 5.6679614067077635,
"epoch": 7.400795624067628,
"grad_norm": 1.375,
"learning_rate": 0.00013595702289034473,
"loss": 4.7613,
"mean_token_accuracy": 0.25064645409584047,
"num_tokens": 14575681.0,
"step": 7445
},
{
"entropy": 5.872348213195801,
"epoch": 7.4057682744903035,
"grad_norm": 1.265625,
"learning_rate": 0.00013565019718559854,
"loss": 5.0948,
"mean_token_accuracy": 0.21303518414497374,
"num_tokens": 14585312.0,
"step": 7450
},
{
"entropy": 5.846982097625732,
"epoch": 7.410740924912979,
"grad_norm": 1.03125,
"learning_rate": 0.00013534379128704937,
"loss": 5.0978,
"mean_token_accuracy": 0.21569385081529618,
"num_tokens": 14596725.0,
"step": 7455
},
{
"entropy": 5.894273090362549,
"epoch": 7.415713575335654,
"grad_norm": 1.4375,
"learning_rate": 0.00013503780611777766,
"loss": 5.1042,
"mean_token_accuracy": 0.2031480610370636,
"num_tokens": 14606238.0,
"step": 7460
},
{
"entropy": 5.894358158111572,
"epoch": 7.420686225758329,
"grad_norm": 1.328125,
"learning_rate": 0.0001347322425995966,
"loss": 5.0711,
"mean_token_accuracy": 0.21269853562116622,
"num_tokens": 14615189.0,
"step": 7465
},
{
"entropy": 5.886167383193969,
"epoch": 7.4256588761810045,
"grad_norm": 1.2890625,
"learning_rate": 0.00013442710165304894,
"loss": 5.1097,
"mean_token_accuracy": 0.2068852961063385,
"num_tokens": 14624342.0,
"step": 7470
},
{
"entropy": 5.79851222038269,
"epoch": 7.43063152660368,
"grad_norm": 1.328125,
"learning_rate": 0.00013412238419740444,
"loss": 4.8449,
"mean_token_accuracy": 0.241060371696949,
"num_tokens": 14635050.0,
"step": 7475
},
{
"entropy": 5.669205856323242,
"epoch": 7.435604177026355,
"grad_norm": 1.390625,
"learning_rate": 0.00013381809115065716,
"loss": 4.8246,
"mean_token_accuracy": 0.25167866200208666,
"num_tokens": 14644450.0,
"step": 7480
},
{
"entropy": 5.791391277313233,
"epoch": 7.44057682744903,
"grad_norm": 1.421875,
"learning_rate": 0.00013351422342952228,
"loss": 4.9437,
"mean_token_accuracy": 0.23377273231744766,
"num_tokens": 14654242.0,
"step": 7485
},
{
"entropy": 5.92456316947937,
"epoch": 7.4455494778717055,
"grad_norm": 1.4296875,
"learning_rate": 0.00013321078194943392,
"loss": 5.0713,
"mean_token_accuracy": 0.20616656988859178,
"num_tokens": 14663142.0,
"step": 7490
},
{
"entropy": 5.8545817852020265,
"epoch": 7.450522128294381,
"grad_norm": 1.375,
"learning_rate": 0.0001329077676245421,
"loss": 5.0522,
"mean_token_accuracy": 0.2138432964682579,
"num_tokens": 14673155.0,
"step": 7495
},
{
"entropy": 5.92165584564209,
"epoch": 7.455494778717056,
"grad_norm": 1.34375,
"learning_rate": 0.00013260518136770983,
"loss": 5.1641,
"mean_token_accuracy": 0.2039618358016014,
"num_tokens": 14683967.0,
"step": 7500
},
{
"epoch": 7.455494778717056,
"eval_entropy": 5.869765022383308,
"eval_loss": 6.476799011230469,
"eval_mean_token_accuracy": 0.14863629731851788,
"eval_num_tokens": 14683967.0,
"eval_runtime": 2.5384,
"eval_samples_per_second": 1365.052,
"eval_steps_per_second": 170.976,
"step": 7500
},
{
"entropy": 5.891140651702881,
"epoch": 7.460467429139731,
"grad_norm": 1.4140625,
"learning_rate": 0.00013230302409051066,
"loss": 5.0781,
"mean_token_accuracy": 0.21572350561618805,
"num_tokens": 14693843.0,
"step": 7505
},
{
"entropy": 5.979051160812378,
"epoch": 7.4654400795624065,
"grad_norm": 1.2578125,
"learning_rate": 0.00013200129670322563,
"loss": 5.2308,
"mean_token_accuracy": 0.19771385192871094,
"num_tokens": 14704323.0,
"step": 7510
},
{
"entropy": 5.751616668701172,
"epoch": 7.470412729985082,
"grad_norm": 1.1875,
"learning_rate": 0.00013170000011484084,
"loss": 4.8877,
"mean_token_accuracy": 0.22854024320840835,
"num_tokens": 14714550.0,
"step": 7515
},
{
"entropy": 5.94213171005249,
"epoch": 7.475385380407757,
"grad_norm": 1.5546875,
"learning_rate": 0.00013139913523304458,
"loss": 5.1358,
"mean_token_accuracy": 0.20780053436756135,
"num_tokens": 14723578.0,
"step": 7520
},
{
"entropy": 5.941567277908325,
"epoch": 7.480358030830432,
"grad_norm": 1.1796875,
"learning_rate": 0.00013109870296422436,
"loss": 5.1348,
"mean_token_accuracy": 0.20959407836198807,
"num_tokens": 14733487.0,
"step": 7525
},
{
"entropy": 5.768622350692749,
"epoch": 7.4853306812531075,
"grad_norm": 1.3359375,
"learning_rate": 0.00013079870421346465,
"loss": 4.936,
"mean_token_accuracy": 0.23197107017040253,
"num_tokens": 14743284.0,
"step": 7530
},
{
"entropy": 5.911783885955811,
"epoch": 7.490303331675783,
"grad_norm": 1.28125,
"learning_rate": 0.00013049913988454385,
"loss": 5.0278,
"mean_token_accuracy": 0.21677444130182266,
"num_tokens": 14753222.0,
"step": 7535
},
{
"entropy": 5.9396140575408936,
"epoch": 7.495275982098459,
"grad_norm": 1.375,
"learning_rate": 0.00013020001087993142,
"loss": 5.2049,
"mean_token_accuracy": 0.1987164720892906,
"num_tokens": 14763631.0,
"step": 7540
},
{
"entropy": 5.822959852218628,
"epoch": 7.500248632521133,
"grad_norm": 1.1328125,
"learning_rate": 0.00012990131810078565,
"loss": 4.9624,
"mean_token_accuracy": 0.22951672375202178,
"num_tokens": 14774474.0,
"step": 7545
},
{
"entropy": 5.789368343353272,
"epoch": 7.505221282943809,
"grad_norm": 1.3828125,
"learning_rate": 0.00012960306244695041,
"loss": 4.8754,
"mean_token_accuracy": 0.23570434302091597,
"num_tokens": 14783560.0,
"step": 7550
},
{
"entropy": 5.887509489059449,
"epoch": 7.510193933366485,
"grad_norm": 1.4140625,
"learning_rate": 0.00012930524481695278,
"loss": 5.0144,
"mean_token_accuracy": 0.2228848949074745,
"num_tokens": 14791986.0,
"step": 7555
},
{
"entropy": 5.764388561248779,
"epoch": 7.51516658378916,
"grad_norm": 1.359375,
"learning_rate": 0.00012900786610800047,
"loss": 5.0109,
"mean_token_accuracy": 0.22351184636354446,
"num_tokens": 14801558.0,
"step": 7560
},
{
"entropy": 5.880010461807251,
"epoch": 7.520139234211835,
"grad_norm": 1.328125,
"learning_rate": 0.00012871092721597832,
"loss": 4.968,
"mean_token_accuracy": 0.23157697170972824,
"num_tokens": 14811373.0,
"step": 7565
},
{
"entropy": 5.92024302482605,
"epoch": 7.52511188463451,
"grad_norm": 1.375,
"learning_rate": 0.00012841442903544666,
"loss": 5.1323,
"mean_token_accuracy": 0.2084745556116104,
"num_tokens": 14820912.0,
"step": 7570
},
{
"entropy": 5.955593252182007,
"epoch": 7.530084535057186,
"grad_norm": 1.5234375,
"learning_rate": 0.00012811837245963817,
"loss": 5.1161,
"mean_token_accuracy": 0.20231297016143798,
"num_tokens": 14829731.0,
"step": 7575
},
{
"entropy": 5.864965295791626,
"epoch": 7.535057185479861,
"grad_norm": 1.390625,
"learning_rate": 0.00012782275838045475,
"loss": 5.0605,
"mean_token_accuracy": 0.21228583455085753,
"num_tokens": 14839873.0,
"step": 7580
},
{
"entropy": 5.8445806980133055,
"epoch": 7.540029835902536,
"grad_norm": 1.25,
"learning_rate": 0.00012752758768846557,
"loss": 5.037,
"mean_token_accuracy": 0.21432324796915053,
"num_tokens": 14850334.0,
"step": 7585
},
{
"entropy": 5.903028726577759,
"epoch": 7.545002486325211,
"grad_norm": 1.296875,
"learning_rate": 0.00012723286127290387,
"loss": 5.0946,
"mean_token_accuracy": 0.20986117869615556,
"num_tokens": 14860597.0,
"step": 7590
},
{
"entropy": 5.7573319435119625,
"epoch": 7.549975136747887,
"grad_norm": 1.296875,
"learning_rate": 0.00012693858002166465,
"loss": 4.9342,
"mean_token_accuracy": 0.22743304967880248,
"num_tokens": 14870100.0,
"step": 7595
},
{
"entropy": 5.872517776489258,
"epoch": 7.554947787170562,
"grad_norm": 1.34375,
"learning_rate": 0.00012664474482130145,
"loss": 5.0135,
"mean_token_accuracy": 0.2236061781644821,
"num_tokens": 14879074.0,
"step": 7600
},
{
"entropy": 5.892554092407226,
"epoch": 7.559920437593237,
"grad_norm": 1.3984375,
"learning_rate": 0.00012635135655702446,
"loss": 5.0937,
"mean_token_accuracy": 0.21317134499549867,
"num_tokens": 14887973.0,
"step": 7605
},
{
"entropy": 5.8736512660980225,
"epoch": 7.564893088015912,
"grad_norm": 1.2265625,
"learning_rate": 0.00012605841611269721,
"loss": 5.0326,
"mean_token_accuracy": 0.2203826680779457,
"num_tokens": 14899491.0,
"step": 7610
},
{
"entropy": 5.866386413574219,
"epoch": 7.569865738438588,
"grad_norm": 1.265625,
"learning_rate": 0.0001257659243708341,
"loss": 5.088,
"mean_token_accuracy": 0.21090784519910813,
"num_tokens": 14909585.0,
"step": 7615
},
{
"entropy": 5.642879056930542,
"epoch": 7.574838388861263,
"grad_norm": 1.2265625,
"learning_rate": 0.0001254738822125978,
"loss": 4.8716,
"mean_token_accuracy": 0.23502113968133925,
"num_tokens": 14919654.0,
"step": 7620
},
{
"entropy": 5.949767971038819,
"epoch": 7.579811039283938,
"grad_norm": 1.328125,
"learning_rate": 0.00012518229051779666,
"loss": 5.1978,
"mean_token_accuracy": 0.2074361726641655,
"num_tokens": 14929730.0,
"step": 7625
},
{
"entropy": 5.8492179870605465,
"epoch": 7.584783689706613,
"grad_norm": 1.3828125,
"learning_rate": 0.0001248911501648817,
"loss": 5.0115,
"mean_token_accuracy": 0.22374957352876662,
"num_tokens": 14939540.0,
"step": 7630
},
{
"entropy": 5.911644840240479,
"epoch": 7.589756340129289,
"grad_norm": 1.28125,
"learning_rate": 0.00012460046203094458,
"loss": 5.105,
"mean_token_accuracy": 0.21016863286495208,
"num_tokens": 14949166.0,
"step": 7635
},
{
"entropy": 5.847161149978637,
"epoch": 7.594728990551964,
"grad_norm": 1.3828125,
"learning_rate": 0.0001243102269917143,
"loss": 4.9667,
"mean_token_accuracy": 0.22770235389471055,
"num_tokens": 14959144.0,
"step": 7640
},
{
"entropy": 5.935601282119751,
"epoch": 7.599701640974639,
"grad_norm": 1.3203125,
"learning_rate": 0.00012402044592155497,
"loss": 5.1299,
"mean_token_accuracy": 0.19975300133228302,
"num_tokens": 14968678.0,
"step": 7645
},
{
"entropy": 5.87359299659729,
"epoch": 7.604674291397314,
"grad_norm": 1.2265625,
"learning_rate": 0.00012373111969346323,
"loss": 5.0368,
"mean_token_accuracy": 0.2096355825662613,
"num_tokens": 14978047.0,
"step": 7650
},
{
"entropy": 5.680193376541138,
"epoch": 7.60964694181999,
"grad_norm": 1.234375,
"learning_rate": 0.00012344224917906516,
"loss": 4.879,
"mean_token_accuracy": 0.23537613153457643,
"num_tokens": 14987335.0,
"step": 7655
},
{
"entropy": 5.8360984325408936,
"epoch": 7.614619592242665,
"grad_norm": 1.4140625,
"learning_rate": 0.0001231538352486142,
"loss": 4.9764,
"mean_token_accuracy": 0.22764275670051576,
"num_tokens": 14997131.0,
"step": 7660
},
{
"entropy": 5.849442100524902,
"epoch": 7.619592242665341,
"grad_norm": 1.4765625,
"learning_rate": 0.00012286587877098808,
"loss": 5.0015,
"mean_token_accuracy": 0.2121598407626152,
"num_tokens": 15006282.0,
"step": 7665
},
{
"entropy": 5.852977275848389,
"epoch": 7.624564893088015,
"grad_norm": 1.5,
"learning_rate": 0.00012257838061368657,
"loss": 4.9822,
"mean_token_accuracy": 0.21840211004018784,
"num_tokens": 15014509.0,
"step": 7670
},
{
"entropy": 5.940796566009522,
"epoch": 7.629537543510692,
"grad_norm": 1.390625,
"learning_rate": 0.00012229134164282884,
"loss": 5.1577,
"mean_token_accuracy": 0.2069163516163826,
"num_tokens": 15024922.0,
"step": 7675
},
{
"entropy": 6.000323677062989,
"epoch": 7.634510193933367,
"grad_norm": 1.28125,
"learning_rate": 0.00012200476272315015,
"loss": 5.1782,
"mean_token_accuracy": 0.20802451372146608,
"num_tokens": 15035612.0,
"step": 7680
},
{
"entropy": 5.871282052993775,
"epoch": 7.639482844356042,
"grad_norm": 1.4453125,
"learning_rate": 0.00012171864471800033,
"loss": 5.1086,
"mean_token_accuracy": 0.20978468656539917,
"num_tokens": 15045667.0,
"step": 7685
},
{
"entropy": 5.820932340621948,
"epoch": 7.644455494778717,
"grad_norm": 1.421875,
"learning_rate": 0.00012143298848934063,
"loss": 5.0087,
"mean_token_accuracy": 0.22427080422639847,
"num_tokens": 15055391.0,
"step": 7690
},
{
"entropy": 5.781161642074585,
"epoch": 7.649428145201393,
"grad_norm": 1.15625,
"learning_rate": 0.0001211477948977408,
"loss": 4.963,
"mean_token_accuracy": 0.21792410612106322,
"num_tokens": 15064690.0,
"step": 7695
},
{
"entropy": 5.8914436340332035,
"epoch": 7.654400795624068,
"grad_norm": 1.375,
"learning_rate": 0.00012086306480237723,
"loss": 5.028,
"mean_token_accuracy": 0.2155271738767624,
"num_tokens": 15073002.0,
"step": 7700
},
{
"entropy": 5.842133712768555,
"epoch": 7.659373446046743,
"grad_norm": 1.3046875,
"learning_rate": 0.00012057879906102967,
"loss": 5.0544,
"mean_token_accuracy": 0.2147751495242119,
"num_tokens": 15082680.0,
"step": 7705
},
{
"entropy": 5.861848068237305,
"epoch": 7.664346096469418,
"grad_norm": 1.3046875,
"learning_rate": 0.00012029499853007916,
"loss": 5.1139,
"mean_token_accuracy": 0.21236233562231063,
"num_tokens": 15093704.0,
"step": 7710
},
{
"entropy": 5.865697383880615,
"epoch": 7.669318746892094,
"grad_norm": 1.3828125,
"learning_rate": 0.00012001166406450528,
"loss": 5.0067,
"mean_token_accuracy": 0.2215912789106369,
"num_tokens": 15103852.0,
"step": 7715
},
{
"entropy": 5.875100326538086,
"epoch": 7.674291397314769,
"grad_norm": 1.5078125,
"learning_rate": 0.00011972879651788323,
"loss": 5.0104,
"mean_token_accuracy": 0.222148534655571,
"num_tokens": 15113911.0,
"step": 7720
},
{
"entropy": 5.8967479228973385,
"epoch": 7.679264047737444,
"grad_norm": 1.3125,
"learning_rate": 0.00011944639674238197,
"loss": 5.0711,
"mean_token_accuracy": 0.21271021366119386,
"num_tokens": 15123311.0,
"step": 7725
},
{
"entropy": 5.680833578109741,
"epoch": 7.684236698160119,
"grad_norm": 1.4375,
"learning_rate": 0.00011916446558876088,
"loss": 4.838,
"mean_token_accuracy": 0.24177502244710922,
"num_tokens": 15133300.0,
"step": 7730
},
{
"entropy": 5.725880765914917,
"epoch": 7.689209348582795,
"grad_norm": 1.3515625,
"learning_rate": 0.00011888300390636788,
"loss": 4.8256,
"mean_token_accuracy": 0.2423257678747177,
"num_tokens": 15143703.0,
"step": 7735
},
{
"entropy": 5.893496894836426,
"epoch": 7.69418199900547,
"grad_norm": 1.3359375,
"learning_rate": 0.0001186020125431364,
"loss": 5.0869,
"mean_token_accuracy": 0.2107880175113678,
"num_tokens": 15153260.0,
"step": 7740
},
{
"entropy": 5.878594636917114,
"epoch": 7.699154649428145,
"grad_norm": 1.3828125,
"learning_rate": 0.00011832149234558286,
"loss": 5.0517,
"mean_token_accuracy": 0.21346597671508788,
"num_tokens": 15162370.0,
"step": 7745
},
{
"entropy": 5.828437376022339,
"epoch": 7.70412729985082,
"grad_norm": 1.4453125,
"learning_rate": 0.00011804144415880445,
"loss": 5.0608,
"mean_token_accuracy": 0.21315167397260665,
"num_tokens": 15172412.0,
"step": 7750
},
{
"entropy": 5.881476640701294,
"epoch": 7.709099950273496,
"grad_norm": 1.3203125,
"learning_rate": 0.00011776186882647643,
"loss": 5.0775,
"mean_token_accuracy": 0.21369345635175704,
"num_tokens": 15181607.0,
"step": 7755
},
{
"entropy": 5.801717042922974,
"epoch": 7.714072600696171,
"grad_norm": 1.3515625,
"learning_rate": 0.00011748276719084924,
"loss": 5.0212,
"mean_token_accuracy": 0.21803819090127946,
"num_tokens": 15191275.0,
"step": 7760
},
{
"entropy": 5.804250574111938,
"epoch": 7.719045251118846,
"grad_norm": 1.40625,
"learning_rate": 0.0001172041400927466,
"loss": 5.0489,
"mean_token_accuracy": 0.2167939841747284,
"num_tokens": 15200461.0,
"step": 7765
},
{
"entropy": 5.9288551807403564,
"epoch": 7.724017901541521,
"grad_norm": 1.390625,
"learning_rate": 0.0001169259883715623,
"loss": 5.153,
"mean_token_accuracy": 0.20884290486574172,
"num_tokens": 15210116.0,
"step": 7770
},
{
"entropy": 5.870552635192871,
"epoch": 7.728990551964197,
"grad_norm": 1.4453125,
"learning_rate": 0.0001166483128652583,
"loss": 5.0448,
"mean_token_accuracy": 0.22024185508489608,
"num_tokens": 15218961.0,
"step": 7775
},
{
"entropy": 5.795787286758423,
"epoch": 7.733963202386872,
"grad_norm": 1.3203125,
"learning_rate": 0.0001163711144103617,
"loss": 4.9557,
"mean_token_accuracy": 0.23161430656909943,
"num_tokens": 15229458.0,
"step": 7780
},
{
"entropy": 5.892241144180298,
"epoch": 7.738935852809547,
"grad_norm": 1.2265625,
"learning_rate": 0.00011609439384196255,
"loss": 5.1223,
"mean_token_accuracy": 0.20543610006570817,
"num_tokens": 15241228.0,
"step": 7785
},
{
"entropy": 5.71119635105133,
"epoch": 7.743908503232223,
"grad_norm": 1.4140625,
"learning_rate": 0.0001158181519937113,
"loss": 4.8328,
"mean_token_accuracy": 0.23719814121723176,
"num_tokens": 15251302.0,
"step": 7790
},
{
"entropy": 5.888898420333862,
"epoch": 7.748881153654898,
"grad_norm": 1.2265625,
"learning_rate": 0.00011554238969781585,
"loss": 5.1022,
"mean_token_accuracy": 0.2115231543779373,
"num_tokens": 15260860.0,
"step": 7795
},
{
"entropy": 5.912890958786011,
"epoch": 7.753853804077574,
"grad_norm": 1.1953125,
"learning_rate": 0.00011526710778503977,
"loss": 5.1359,
"mean_token_accuracy": 0.21072917133569719,
"num_tokens": 15270947.0,
"step": 7800
},
{
"entropy": 5.858468532562256,
"epoch": 7.758826454500249,
"grad_norm": 1.390625,
"learning_rate": 0.00011499230708469932,
"loss": 5.0028,
"mean_token_accuracy": 0.22482018768787385,
"num_tokens": 15281163.0,
"step": 7805
},
{
"entropy": 5.857759094238281,
"epoch": 7.763799104922924,
"grad_norm": 1.3671875,
"learning_rate": 0.00011471798842466092,
"loss": 5.0038,
"mean_token_accuracy": 0.22339935153722762,
"num_tokens": 15290846.0,
"step": 7810
},
{
"entropy": 5.806914615631103,
"epoch": 7.7687717553455995,
"grad_norm": 1.3984375,
"learning_rate": 0.00011444415263133905,
"loss": 4.9526,
"mean_token_accuracy": 0.22970549166202545,
"num_tokens": 15299907.0,
"step": 7815
},
{
"entropy": 5.920826482772827,
"epoch": 7.773744405768275,
"grad_norm": 1.4140625,
"learning_rate": 0.00011417080052969314,
"loss": 5.1425,
"mean_token_accuracy": 0.204471592605114,
"num_tokens": 15308604.0,
"step": 7820
},
{
"entropy": 5.9005128860473635,
"epoch": 7.77871705619095,
"grad_norm": 1.3984375,
"learning_rate": 0.00011389793294322582,
"loss": 5.1073,
"mean_token_accuracy": 0.21209308952093125,
"num_tokens": 15317623.0,
"step": 7825
},
{
"entropy": 5.821011066436768,
"epoch": 7.783689706613625,
"grad_norm": 1.3359375,
"learning_rate": 0.00011362555069397988,
"loss": 5.0572,
"mean_token_accuracy": 0.2152262583374977,
"num_tokens": 15327457.0,
"step": 7830
},
{
"entropy": 5.721501779556275,
"epoch": 7.7886623570363005,
"grad_norm": 1.390625,
"learning_rate": 0.00011335365460253592,
"loss": 4.8317,
"mean_token_accuracy": 0.239688877761364,
"num_tokens": 15337025.0,
"step": 7835
},
{
"entropy": 5.853474187850952,
"epoch": 7.793635007458976,
"grad_norm": 1.3203125,
"learning_rate": 0.00011308224548801014,
"loss": 5.0655,
"mean_token_accuracy": 0.2140434741973877,
"num_tokens": 15346936.0,
"step": 7840
},
{
"entropy": 5.906891345977783,
"epoch": 7.798607657881651,
"grad_norm": 1.328125,
"learning_rate": 0.00011281132416805143,
"loss": 5.1064,
"mean_token_accuracy": 0.21187581717967988,
"num_tokens": 15356240.0,
"step": 7845
},
{
"entropy": 5.811156558990478,
"epoch": 7.803580308304326,
"grad_norm": 1.5703125,
"learning_rate": 0.0001125408914588394,
"loss": 5.0347,
"mean_token_accuracy": 0.22242362350225447,
"num_tokens": 15365807.0,
"step": 7850
},
{
"entropy": 5.758872842788696,
"epoch": 7.8085529587270015,
"grad_norm": 1.3359375,
"learning_rate": 0.0001122709481750814,
"loss": 4.9852,
"mean_token_accuracy": 0.22568795680999756,
"num_tokens": 15376077.0,
"step": 7855
},
{
"entropy": 5.872467041015625,
"epoch": 7.813525609149677,
"grad_norm": 1.4140625,
"learning_rate": 0.0001120014951300105,
"loss": 5.077,
"mean_token_accuracy": 0.205927012860775,
"num_tokens": 15385391.0,
"step": 7860
},
{
"entropy": 5.935386276245117,
"epoch": 7.818498259572352,
"grad_norm": 1.40625,
"learning_rate": 0.00011173253313538284,
"loss": 5.0847,
"mean_token_accuracy": 0.21538374125957488,
"num_tokens": 15394998.0,
"step": 7865
},
{
"entropy": 5.892831897735595,
"epoch": 7.823470909995027,
"grad_norm": 1.3984375,
"learning_rate": 0.0001114640630014753,
"loss": 5.0666,
"mean_token_accuracy": 0.21249615848064424,
"num_tokens": 15403766.0,
"step": 7870
},
{
"entropy": 5.652636623382568,
"epoch": 7.8284435604177025,
"grad_norm": 1.0625,
"learning_rate": 0.00011119608553708277,
"loss": 4.8813,
"mean_token_accuracy": 0.24300582855939865,
"num_tokens": 15414963.0,
"step": 7875
},
{
"entropy": 5.6845180034637455,
"epoch": 7.833416210840378,
"grad_norm": 1.4921875,
"learning_rate": 0.00011092860154951618,
"loss": 4.8988,
"mean_token_accuracy": 0.22980903685092927,
"num_tokens": 15425122.0,
"step": 7880
},
{
"entropy": 5.832586336135864,
"epoch": 7.838388861263053,
"grad_norm": 1.25,
"learning_rate": 0.00011066161184459954,
"loss": 4.9901,
"mean_token_accuracy": 0.214211368560791,
"num_tokens": 15436343.0,
"step": 7885
},
{
"entropy": 5.7276082038879395,
"epoch": 7.843361511685728,
"grad_norm": 1.1953125,
"learning_rate": 0.00011039511722666795,
"loss": 4.8961,
"mean_token_accuracy": 0.242788565158844,
"num_tokens": 15446487.0,
"step": 7890
},
{
"entropy": 5.888050127029419,
"epoch": 7.8483341621084035,
"grad_norm": 1.4296875,
"learning_rate": 0.0001101291184985651,
"loss": 5.0492,
"mean_token_accuracy": 0.21491917371749877,
"num_tokens": 15456342.0,
"step": 7895
},
{
"entropy": 5.8608537197113035,
"epoch": 7.853306812531079,
"grad_norm": 1.2421875,
"learning_rate": 0.00010986361646164044,
"loss": 5.035,
"mean_token_accuracy": 0.22237204760313034,
"num_tokens": 15466360.0,
"step": 7900
},
{
"entropy": 5.835109996795654,
"epoch": 7.858279462953754,
"grad_norm": 1.3828125,
"learning_rate": 0.00010959861191574746,
"loss": 4.9855,
"mean_token_accuracy": 0.21936475932598115,
"num_tokens": 15475952.0,
"step": 7905
},
{
"entropy": 5.814520978927613,
"epoch": 7.863252113376429,
"grad_norm": 1.078125,
"learning_rate": 0.00010933410565924061,
"loss": 4.9599,
"mean_token_accuracy": 0.22817686349153518,
"num_tokens": 15486405.0,
"step": 7910
},
{
"entropy": 5.949023580551147,
"epoch": 7.868224763799105,
"grad_norm": 1.359375,
"learning_rate": 0.00010907009848897327,
"loss": 5.1902,
"mean_token_accuracy": 0.20519631057977678,
"num_tokens": 15496311.0,
"step": 7915
},
{
"entropy": 5.727574968338013,
"epoch": 7.87319741422178,
"grad_norm": 1.3515625,
"learning_rate": 0.00010880659120029537,
"loss": 4.8664,
"mean_token_accuracy": 0.23593171536922455,
"num_tokens": 15505373.0,
"step": 7920
},
{
"entropy": 5.861940908432007,
"epoch": 7.878170064644456,
"grad_norm": 1.3203125,
"learning_rate": 0.00010854358458705079,
"loss": 5.0868,
"mean_token_accuracy": 0.21066632270812988,
"num_tokens": 15516035.0,
"step": 7925
},
{
"entropy": 5.9366514682769775,
"epoch": 7.883142715067131,
"grad_norm": 1.40625,
"learning_rate": 0.00010828107944157509,
"loss": 5.1706,
"mean_token_accuracy": 0.20745351314544677,
"num_tokens": 15526374.0,
"step": 7930
},
{
"entropy": 5.702072286605835,
"epoch": 7.888115365489806,
"grad_norm": 1.140625,
"learning_rate": 0.00010801907655469319,
"loss": 4.8842,
"mean_token_accuracy": 0.2336994081735611,
"num_tokens": 15536956.0,
"step": 7935
},
{
"entropy": 5.961058187484741,
"epoch": 7.893088015912482,
"grad_norm": 1.4296875,
"learning_rate": 0.00010775757671571675,
"loss": 5.2334,
"mean_token_accuracy": 0.2000860318541527,
"num_tokens": 15547198.0,
"step": 7940
},
{
"entropy": 5.950444030761719,
"epoch": 7.898060666335157,
"grad_norm": 1.3671875,
"learning_rate": 0.00010749658071244211,
"loss": 5.1537,
"mean_token_accuracy": 0.2071128398180008,
"num_tokens": 15557148.0,
"step": 7945
},
{
"entropy": 5.8046040534973145,
"epoch": 7.903033316757832,
"grad_norm": 1.4140625,
"learning_rate": 0.00010723608933114754,
"loss": 4.9255,
"mean_token_accuracy": 0.2227189227938652,
"num_tokens": 15567515.0,
"step": 7950
},
{
"entropy": 5.81124963760376,
"epoch": 7.908005967180507,
"grad_norm": 1.2734375,
"learning_rate": 0.00010697610335659139,
"loss": 4.8778,
"mean_token_accuracy": 0.23430223613977433,
"num_tokens": 15576806.0,
"step": 7955
},
{
"entropy": 5.704153728485108,
"epoch": 7.912978617603183,
"grad_norm": 1.1953125,
"learning_rate": 0.00010671662357200904,
"loss": 4.8592,
"mean_token_accuracy": 0.24311269968748092,
"num_tokens": 15587057.0,
"step": 7960
},
{
"entropy": 5.926538944244385,
"epoch": 7.917951268025858,
"grad_norm": 1.3515625,
"learning_rate": 0.00010645765075911129,
"loss": 5.1288,
"mean_token_accuracy": 0.20259761214256286,
"num_tokens": 15596550.0,
"step": 7965
},
{
"entropy": 5.821710729598999,
"epoch": 7.922923918448533,
"grad_norm": 1.3203125,
"learning_rate": 0.00010619918569808136,
"loss": 5.0643,
"mean_token_accuracy": 0.21110006272792817,
"num_tokens": 15606124.0,
"step": 7970
},
{
"entropy": 5.893390226364136,
"epoch": 7.927896568871208,
"grad_norm": 1.28125,
"learning_rate": 0.00010594122916757288,
"loss": 5.0442,
"mean_token_accuracy": 0.22030047476291656,
"num_tokens": 15615467.0,
"step": 7975
},
{
"entropy": 5.889110803604126,
"epoch": 7.932869219293884,
"grad_norm": 1.2578125,
"learning_rate": 0.00010568378194470754,
"loss": 5.1315,
"mean_token_accuracy": 0.21186473220586777,
"num_tokens": 15625994.0,
"step": 7980
},
{
"entropy": 5.815444707870483,
"epoch": 7.937841869716559,
"grad_norm": 1.359375,
"learning_rate": 0.00010542684480507278,
"loss": 4.8808,
"mean_token_accuracy": 0.22966034859418868,
"num_tokens": 15636693.0,
"step": 7985
},
{
"entropy": 5.909979629516601,
"epoch": 7.942814520139234,
"grad_norm": 1.28125,
"learning_rate": 0.00010517041852271906,
"loss": 5.1151,
"mean_token_accuracy": 0.20856352597475053,
"num_tokens": 15645537.0,
"step": 7990
},
{
"entropy": 5.81982913017273,
"epoch": 7.947787170561909,
"grad_norm": 1.4140625,
"learning_rate": 0.00010491450387015813,
"loss": 4.9997,
"mean_token_accuracy": 0.22455549687147142,
"num_tokens": 15654902.0,
"step": 7995
},
{
"entropy": 5.869769048690796,
"epoch": 7.952759820984585,
"grad_norm": 1.4453125,
"learning_rate": 0.00010465910161836015,
"loss": 5.025,
"mean_token_accuracy": 0.2156398192048073,
"num_tokens": 15664344.0,
"step": 8000
},
{
"epoch": 7.952759820984585,
"eval_entropy": 5.816498546556394,
"eval_loss": 6.472532272338867,
"eval_mean_token_accuracy": 0.14929654485275667,
"eval_num_tokens": 15664344.0,
"eval_runtime": 2.5413,
"eval_samples_per_second": 1363.481,
"eval_steps_per_second": 170.779,
"step": 8000
},
{
"entropy": 5.866929244995117,
"epoch": 7.95773247140726,
"grad_norm": 1.3125,
"learning_rate": 0.00010440421253675184,
"loss": 5.081,
"mean_token_accuracy": 0.21410890519618989,
"num_tokens": 15674181.0,
"step": 8005
},
{
"entropy": 5.834720802307129,
"epoch": 7.962705121829935,
"grad_norm": 1.3203125,
"learning_rate": 0.00010414983739321382,
"loss": 5.0712,
"mean_token_accuracy": 0.2187820479273796,
"num_tokens": 15683737.0,
"step": 8010
},
{
"entropy": 5.884942197799683,
"epoch": 7.96767777225261,
"grad_norm": 1.2109375,
"learning_rate": 0.00010389597695407843,
"loss": 5.0996,
"mean_token_accuracy": 0.2156265914440155,
"num_tokens": 15693995.0,
"step": 8015
},
{
"entropy": 5.851906204223633,
"epoch": 7.972650422675286,
"grad_norm": 1.4765625,
"learning_rate": 0.00010364263198412746,
"loss": 5.0907,
"mean_token_accuracy": 0.2117193818092346,
"num_tokens": 15704009.0,
"step": 8020
},
{
"entropy": 5.821045398712158,
"epoch": 7.977623073097961,
"grad_norm": 1.40625,
"learning_rate": 0.00010338980324658972,
"loss": 5.002,
"mean_token_accuracy": 0.22041586488485337,
"num_tokens": 15714814.0,
"step": 8025
},
{
"entropy": 5.843690633773804,
"epoch": 7.982595723520636,
"grad_norm": 1.3046875,
"learning_rate": 0.00010313749150313872,
"loss": 4.9568,
"mean_token_accuracy": 0.2156564399600029,
"num_tokens": 15724472.0,
"step": 8030
},
{
"entropy": 5.8362706184387205,
"epoch": 7.987568373943311,
"grad_norm": 1.375,
"learning_rate": 0.00010288569751389082,
"loss": 5.0342,
"mean_token_accuracy": 0.21631285101175307,
"num_tokens": 15735230.0,
"step": 8035
},
{
"entropy": 5.841118335723877,
"epoch": 7.992541024365988,
"grad_norm": 1.3515625,
"learning_rate": 0.00010263442203740217,
"loss": 5.0019,
"mean_token_accuracy": 0.22182606607675553,
"num_tokens": 15744826.0,
"step": 8040
},
{
"entropy": 5.977916240692139,
"epoch": 7.997513674788662,
"grad_norm": 1.296875,
"learning_rate": 0.00010238366583066713,
"loss": 5.1971,
"mean_token_accuracy": 0.19967948496341706,
"num_tokens": 15755763.0,
"step": 8045
},
{
"entropy": 5.853172037336561,
"epoch": 8.00198906016907,
"grad_norm": 1.21875,
"learning_rate": 0.00010213342964911568,
"loss": 5.0194,
"mean_token_accuracy": 0.22359820538096958,
"num_tokens": 15765479.0,
"step": 8050
},
{
"entropy": 5.87145733833313,
"epoch": 8.006961710591746,
"grad_norm": 1.3515625,
"learning_rate": 0.00010188371424661095,
"loss": 5.0192,
"mean_token_accuracy": 0.2122771754860878,
"num_tokens": 15774641.0,
"step": 8055
},
{
"entropy": 5.856323909759522,
"epoch": 8.01193436101442,
"grad_norm": 1.359375,
"learning_rate": 0.00010163452037544745,
"loss": 4.9874,
"mean_token_accuracy": 0.21766557395458222,
"num_tokens": 15784423.0,
"step": 8060
},
{
"entropy": 5.795698738098144,
"epoch": 8.016907011437096,
"grad_norm": 1.40625,
"learning_rate": 0.0001013858487863483,
"loss": 4.9119,
"mean_token_accuracy": 0.21725862920284272,
"num_tokens": 15794542.0,
"step": 8065
},
{
"entropy": 5.8584319114685055,
"epoch": 8.02187966185977,
"grad_norm": 1.1953125,
"learning_rate": 0.0001011377002284633,
"loss": 5.0354,
"mean_token_accuracy": 0.2116153970360756,
"num_tokens": 15805562.0,
"step": 8070
},
{
"entropy": 5.860963201522827,
"epoch": 8.026852312282447,
"grad_norm": 1.28125,
"learning_rate": 0.00010089007544936666,
"loss": 4.9697,
"mean_token_accuracy": 0.22660405784845353,
"num_tokens": 15815796.0,
"step": 8075
},
{
"entropy": 5.865639877319336,
"epoch": 8.031824962705121,
"grad_norm": 1.28125,
"learning_rate": 0.00010064297519505435,
"loss": 4.965,
"mean_token_accuracy": 0.21841677874326706,
"num_tokens": 15824858.0,
"step": 8080
},
{
"entropy": 5.815975904464722,
"epoch": 8.036797613127797,
"grad_norm": 1.1953125,
"learning_rate": 0.00010039640020994241,
"loss": 4.9464,
"mean_token_accuracy": 0.2238340362906456,
"num_tokens": 15835061.0,
"step": 8085
},
{
"entropy": 5.775523281097412,
"epoch": 8.041770263550472,
"grad_norm": 1.125,
"learning_rate": 0.00010015035123686443,
"loss": 4.8423,
"mean_token_accuracy": 0.2400132268667221,
"num_tokens": 15845060.0,
"step": 8090
},
{
"entropy": 5.919277286529541,
"epoch": 8.046742913973148,
"grad_norm": 1.21875,
"learning_rate": 9.990482901706918e-05,
"loss": 5.0448,
"mean_token_accuracy": 0.21610289812088013,
"num_tokens": 15856037.0,
"step": 8095
},
{
"entropy": 5.882541036605835,
"epoch": 8.051715564395822,
"grad_norm": 1.3125,
"learning_rate": 9.965983429021874e-05,
"loss": 4.9761,
"mean_token_accuracy": 0.215408031642437,
"num_tokens": 15865178.0,
"step": 8100
},
{
"entropy": 5.759833335876465,
"epoch": 8.056688214818498,
"grad_norm": 1.390625,
"learning_rate": 9.941536779438584e-05,
"loss": 4.8488,
"mean_token_accuracy": 0.23223868757486343,
"num_tokens": 15874775.0,
"step": 8105
},
{
"entropy": 5.799562740325928,
"epoch": 8.061660865241173,
"grad_norm": 1.296875,
"learning_rate": 9.917143026605201e-05,
"loss": 4.9606,
"mean_token_accuracy": 0.22415116429328918,
"num_tokens": 15884493.0,
"step": 8110
},
{
"entropy": 5.933044290542602,
"epoch": 8.066633515663849,
"grad_norm": 1.4296875,
"learning_rate": 9.892802244010522e-05,
"loss": 5.098,
"mean_token_accuracy": 0.2170291557908058,
"num_tokens": 15893938.0,
"step": 8115
},
{
"entropy": 5.832733821868897,
"epoch": 8.071606166086523,
"grad_norm": 1.3125,
"learning_rate": 9.86851450498375e-05,
"loss": 4.8739,
"mean_token_accuracy": 0.22929019331932068,
"num_tokens": 15903379.0,
"step": 8120
},
{
"entropy": 5.714403390884399,
"epoch": 8.0765788165092,
"grad_norm": 1.3046875,
"learning_rate": 9.844279882694309e-05,
"loss": 4.8072,
"mean_token_accuracy": 0.23706578314304352,
"num_tokens": 15913674.0,
"step": 8125
},
{
"entropy": 5.848007917404175,
"epoch": 8.081551466931876,
"grad_norm": 1.3671875,
"learning_rate": 9.820098450151582e-05,
"loss": 4.9833,
"mean_token_accuracy": 0.22056680619716645,
"num_tokens": 15922565.0,
"step": 8130
},
{
"entropy": 5.9066609859466555,
"epoch": 8.08652411735455,
"grad_norm": 1.3359375,
"learning_rate": 9.795970280204733e-05,
"loss": 5.0102,
"mean_token_accuracy": 0.21353595554828644,
"num_tokens": 15931931.0,
"step": 8135
},
{
"entropy": 5.903261709213257,
"epoch": 8.091496767777226,
"grad_norm": 1.3515625,
"learning_rate": 9.771895445542453e-05,
"loss": 4.9859,
"mean_token_accuracy": 0.22084571421146393,
"num_tokens": 15942197.0,
"step": 8140
},
{
"entropy": 5.872694253921509,
"epoch": 8.0964694181999,
"grad_norm": 1.3671875,
"learning_rate": 9.747874018692752e-05,
"loss": 4.9588,
"mean_token_accuracy": 0.2207825794816017,
"num_tokens": 15951746.0,
"step": 8145
},
{
"entropy": 5.900765371322632,
"epoch": 8.101442068622577,
"grad_norm": 1.3203125,
"learning_rate": 9.723906072022759e-05,
"loss": 4.9941,
"mean_token_accuracy": 0.2157755210995674,
"num_tokens": 15961329.0,
"step": 8150
},
{
"entropy": 5.910271024703979,
"epoch": 8.106414719045251,
"grad_norm": 1.375,
"learning_rate": 9.69999167773847e-05,
"loss": 4.9848,
"mean_token_accuracy": 0.2198132410645485,
"num_tokens": 15970998.0,
"step": 8155
},
{
"entropy": 5.565882062911987,
"epoch": 8.111387369467927,
"grad_norm": 1.40625,
"learning_rate": 9.676130907884564e-05,
"loss": 4.6248,
"mean_token_accuracy": 0.2619062438607216,
"num_tokens": 15981358.0,
"step": 8160
},
{
"entropy": 5.828635120391846,
"epoch": 8.116360019890601,
"grad_norm": 1.375,
"learning_rate": 9.652323834344166e-05,
"loss": 4.9192,
"mean_token_accuracy": 0.2308216869831085,
"num_tokens": 15992194.0,
"step": 8165
},
{
"entropy": 5.862514591217041,
"epoch": 8.121332670313278,
"grad_norm": 1.5078125,
"learning_rate": 9.628570528838626e-05,
"loss": 4.9608,
"mean_token_accuracy": 0.22526509165763856,
"num_tokens": 16001542.0,
"step": 8170
},
{
"entropy": 5.841127109527588,
"epoch": 8.126305320735952,
"grad_norm": 1.296875,
"learning_rate": 9.604871062927328e-05,
"loss": 4.9931,
"mean_token_accuracy": 0.2154684141278267,
"num_tokens": 16012799.0,
"step": 8175
},
{
"entropy": 5.852886486053467,
"epoch": 8.131277971158628,
"grad_norm": 1.34375,
"learning_rate": 9.581225508007441e-05,
"loss": 4.9207,
"mean_token_accuracy": 0.22438742220401764,
"num_tokens": 16021612.0,
"step": 8180
},
{
"entropy": 5.913249778747558,
"epoch": 8.136250621581302,
"grad_norm": 1.3984375,
"learning_rate": 9.557633935313731e-05,
"loss": 5.0078,
"mean_token_accuracy": 0.21595665961503982,
"num_tokens": 16031290.0,
"step": 8185
},
{
"entropy": 5.870132398605347,
"epoch": 8.141223272003979,
"grad_norm": 1.3046875,
"learning_rate": 9.534096415918354e-05,
"loss": 4.98,
"mean_token_accuracy": 0.22096948325634003,
"num_tokens": 16041848.0,
"step": 8190
},
{
"entropy": 5.882846403121948,
"epoch": 8.146195922426653,
"grad_norm": 1.296875,
"learning_rate": 9.51061302073058e-05,
"loss": 5.028,
"mean_token_accuracy": 0.2176157221198082,
"num_tokens": 16051782.0,
"step": 8195
},
{
"entropy": 5.835755920410156,
"epoch": 8.151168572849329,
"grad_norm": 1.28125,
"learning_rate": 9.487183820496664e-05,
"loss": 4.8808,
"mean_token_accuracy": 0.23662614971399307,
"num_tokens": 16062841.0,
"step": 8200
},
{
"entropy": 5.876460123062134,
"epoch": 8.156141223272003,
"grad_norm": 1.28125,
"learning_rate": 9.463808885799582e-05,
"loss": 5.0291,
"mean_token_accuracy": 0.22198223173618317,
"num_tokens": 16073090.0,
"step": 8205
},
{
"entropy": 5.842329502105713,
"epoch": 8.16111387369468,
"grad_norm": 1.21875,
"learning_rate": 9.440488287058819e-05,
"loss": 4.9328,
"mean_token_accuracy": 0.22961543500423431,
"num_tokens": 16083472.0,
"step": 8210
},
{
"entropy": 5.581771373748779,
"epoch": 8.166086524117354,
"grad_norm": 1.4296875,
"learning_rate": 9.417222094530181e-05,
"loss": 4.7281,
"mean_token_accuracy": 0.24689862728118897,
"num_tokens": 16093146.0,
"step": 8215
},
{
"entropy": 5.721676778793335,
"epoch": 8.17105917454003,
"grad_norm": 1.109375,
"learning_rate": 9.39401037830556e-05,
"loss": 4.8147,
"mean_token_accuracy": 0.24677679389715196,
"num_tokens": 16103329.0,
"step": 8220
},
{
"entropy": 5.899397134780884,
"epoch": 8.176031824962704,
"grad_norm": 1.5,
"learning_rate": 9.370853208312732e-05,
"loss": 5.0132,
"mean_token_accuracy": 0.21130330860614777,
"num_tokens": 16112428.0,
"step": 8225
},
{
"entropy": 5.868875646591187,
"epoch": 8.18100447538538,
"grad_norm": 1.265625,
"learning_rate": 9.347750654315165e-05,
"loss": 5.006,
"mean_token_accuracy": 0.21781714409589767,
"num_tokens": 16122697.0,
"step": 8230
},
{
"entropy": 5.9104238986969,
"epoch": 8.185977125808055,
"grad_norm": 1.3515625,
"learning_rate": 9.324702785911762e-05,
"loss": 5.0567,
"mean_token_accuracy": 0.2110861822962761,
"num_tokens": 16132708.0,
"step": 8235
},
{
"entropy": 5.7160561084747314,
"epoch": 8.190949776230731,
"grad_norm": 1.25,
"learning_rate": 9.301709672536701e-05,
"loss": 4.8174,
"mean_token_accuracy": 0.2466488853096962,
"num_tokens": 16143375.0,
"step": 8240
},
{
"entropy": 5.858629655838013,
"epoch": 8.195922426653405,
"grad_norm": 1.40625,
"learning_rate": 9.278771383459189e-05,
"loss": 4.9416,
"mean_token_accuracy": 0.22149503976106644,
"num_tokens": 16153280.0,
"step": 8245
},
{
"entropy": 5.853360652923584,
"epoch": 8.200895077076082,
"grad_norm": 1.40625,
"learning_rate": 9.255887987783287e-05,
"loss": 4.9555,
"mean_token_accuracy": 0.22651243954896927,
"num_tokens": 16164376.0,
"step": 8250
},
{
"entropy": 5.836536455154419,
"epoch": 8.205867727498758,
"grad_norm": 1.515625,
"learning_rate": 9.233059554447674e-05,
"loss": 4.9619,
"mean_token_accuracy": 0.22148529589176177,
"num_tokens": 16173582.0,
"step": 8255
},
{
"entropy": 5.91468071937561,
"epoch": 8.210840377921432,
"grad_norm": 1.375,
"learning_rate": 9.210286152225433e-05,
"loss": 4.9743,
"mean_token_accuracy": 0.22227033376693725,
"num_tokens": 16182731.0,
"step": 8260
},
{
"entropy": 5.818155241012573,
"epoch": 8.215813028344108,
"grad_norm": 1.4296875,
"learning_rate": 9.187567849723889e-05,
"loss": 4.9023,
"mean_token_accuracy": 0.22633808404207229,
"num_tokens": 16191739.0,
"step": 8265
},
{
"entropy": 5.825146150588989,
"epoch": 8.220785678766783,
"grad_norm": 1.3828125,
"learning_rate": 9.164904715384363e-05,
"loss": 4.9044,
"mean_token_accuracy": 0.22773143351078035,
"num_tokens": 16200962.0,
"step": 8270
},
{
"entropy": 5.8469444751739506,
"epoch": 8.225758329189459,
"grad_norm": 1.421875,
"learning_rate": 9.142296817481963e-05,
"loss": 4.9942,
"mean_token_accuracy": 0.22045760750770568,
"num_tokens": 16209731.0,
"step": 8275
},
{
"entropy": 5.656614971160889,
"epoch": 8.230730979612133,
"grad_norm": 1.375,
"learning_rate": 9.11974422412541e-05,
"loss": 4.7272,
"mean_token_accuracy": 0.24805450886487962,
"num_tokens": 16219710.0,
"step": 8280
},
{
"entropy": 5.9306480407714846,
"epoch": 8.23570363003481,
"grad_norm": 1.25,
"learning_rate": 9.097247003256801e-05,
"loss": 5.0909,
"mean_token_accuracy": 0.2133454591035843,
"num_tokens": 16231338.0,
"step": 8285
},
{
"entropy": 5.914089870452881,
"epoch": 8.240676280457484,
"grad_norm": 1.3125,
"learning_rate": 9.074805222651418e-05,
"loss": 5.0683,
"mean_token_accuracy": 0.20729184597730638,
"num_tokens": 16241748.0,
"step": 8290
},
{
"entropy": 5.8338539600372314,
"epoch": 8.24564893088016,
"grad_norm": 1.46875,
"learning_rate": 9.052418949917543e-05,
"loss": 4.9546,
"mean_token_accuracy": 0.22527040392160416,
"num_tokens": 16252004.0,
"step": 8295
},
{
"entropy": 5.8772481918334964,
"epoch": 8.250621581302834,
"grad_norm": 1.3671875,
"learning_rate": 9.030088252496199e-05,
"loss": 5.0285,
"mean_token_accuracy": 0.21613571643829346,
"num_tokens": 16262343.0,
"step": 8300
},
{
"entropy": 5.935594987869263,
"epoch": 8.25559423172551,
"grad_norm": 1.4375,
"learning_rate": 9.007813197661016e-05,
"loss": 5.1156,
"mean_token_accuracy": 0.20527415126562118,
"num_tokens": 16271992.0,
"step": 8305
},
{
"entropy": 5.869619989395142,
"epoch": 8.260566882148185,
"grad_norm": 1.3984375,
"learning_rate": 8.985593852517979e-05,
"loss": 4.9461,
"mean_token_accuracy": 0.22220960557460784,
"num_tokens": 16282173.0,
"step": 8310
},
{
"entropy": 5.761919832229614,
"epoch": 8.26553953257086,
"grad_norm": 1.171875,
"learning_rate": 8.963430284005235e-05,
"loss": 4.8822,
"mean_token_accuracy": 0.23405812978744506,
"num_tokens": 16292472.0,
"step": 8315
},
{
"entropy": 5.815483903884887,
"epoch": 8.270512182993535,
"grad_norm": 1.453125,
"learning_rate": 8.941322558892918e-05,
"loss": 4.8569,
"mean_token_accuracy": 0.23553891927003862,
"num_tokens": 16302107.0,
"step": 8320
},
{
"entropy": 5.8464500427246096,
"epoch": 8.275484833416211,
"grad_norm": 1.2421875,
"learning_rate": 8.919270743782903e-05,
"loss": 4.9244,
"mean_token_accuracy": 0.23084621131420135,
"num_tokens": 16312539.0,
"step": 8325
},
{
"entropy": 5.890255689620972,
"epoch": 8.280457483838886,
"grad_norm": 1.3984375,
"learning_rate": 8.897274905108653e-05,
"loss": 4.9575,
"mean_token_accuracy": 0.2194770857691765,
"num_tokens": 16321914.0,
"step": 8330
},
{
"entropy": 5.916691589355469,
"epoch": 8.285430134261562,
"grad_norm": 1.2109375,
"learning_rate": 8.87533510913499e-05,
"loss": 5.0377,
"mean_token_accuracy": 0.22230081558227538,
"num_tokens": 16332911.0,
"step": 8335
},
{
"entropy": 5.884485673904419,
"epoch": 8.290402784684236,
"grad_norm": 1.40625,
"learning_rate": 8.853451421957887e-05,
"loss": 4.9947,
"mean_token_accuracy": 0.22194119840860366,
"num_tokens": 16342837.0,
"step": 8340
},
{
"entropy": 5.904466867446899,
"epoch": 8.295375435106912,
"grad_norm": 1.484375,
"learning_rate": 8.831623909504304e-05,
"loss": 5.0615,
"mean_token_accuracy": 0.2095898538827896,
"num_tokens": 16352487.0,
"step": 8345
},
{
"entropy": 5.7756751537322994,
"epoch": 8.300348085529587,
"grad_norm": 1.5390625,
"learning_rate": 8.809852637531951e-05,
"loss": 4.8674,
"mean_token_accuracy": 0.24574860483407973,
"num_tokens": 16361743.0,
"step": 8350
},
{
"entropy": 5.706744623184204,
"epoch": 8.305320735952263,
"grad_norm": 1.1015625,
"learning_rate": 8.788137671629121e-05,
"loss": 4.8105,
"mean_token_accuracy": 0.23959992080926895,
"num_tokens": 16373254.0,
"step": 8355
},
{
"entropy": 5.89945912361145,
"epoch": 8.310293386374937,
"grad_norm": 1.3203125,
"learning_rate": 8.766479077214466e-05,
"loss": 5.0192,
"mean_token_accuracy": 0.2210462883114815,
"num_tokens": 16383172.0,
"step": 8360
},
{
"entropy": 5.828373289108276,
"epoch": 8.315266036797613,
"grad_norm": 1.328125,
"learning_rate": 8.744876919536832e-05,
"loss": 4.9113,
"mean_token_accuracy": 0.2308765396475792,
"num_tokens": 16392142.0,
"step": 8365
},
{
"entropy": 5.83174729347229,
"epoch": 8.320238687220288,
"grad_norm": 1.3046875,
"learning_rate": 8.723331263675021e-05,
"loss": 4.9809,
"mean_token_accuracy": 0.221316958963871,
"num_tokens": 16403347.0,
"step": 8370
},
{
"entropy": 5.849996042251587,
"epoch": 8.325211337642964,
"grad_norm": 1.40625,
"learning_rate": 8.701842174537627e-05,
"loss": 4.9943,
"mean_token_accuracy": 0.22452719807624816,
"num_tokens": 16413183.0,
"step": 8375
},
{
"entropy": 5.928778219223022,
"epoch": 8.330183988065638,
"grad_norm": 1.5625,
"learning_rate": 8.680409716862833e-05,
"loss": 5.0288,
"mean_token_accuracy": 0.2135708212852478,
"num_tokens": 16422036.0,
"step": 8380
},
{
"entropy": 5.8775571346282955,
"epoch": 8.335156638488314,
"grad_norm": 1.4296875,
"learning_rate": 8.659033955218221e-05,
"loss": 4.9614,
"mean_token_accuracy": 0.21980690062046052,
"num_tokens": 16431241.0,
"step": 8385
},
{
"entropy": 5.816654396057129,
"epoch": 8.34012928891099,
"grad_norm": 1.3671875,
"learning_rate": 8.637714954000548e-05,
"loss": 4.9221,
"mean_token_accuracy": 0.232638680934906,
"num_tokens": 16440945.0,
"step": 8390
},
{
"entropy": 5.7017683506011965,
"epoch": 8.345101939333665,
"grad_norm": 1.234375,
"learning_rate": 8.616452777435597e-05,
"loss": 4.8077,
"mean_token_accuracy": 0.24582804292440413,
"num_tokens": 16450552.0,
"step": 8395
},
{
"entropy": 5.890775299072265,
"epoch": 8.35007458975634,
"grad_norm": 1.421875,
"learning_rate": 8.595247489577939e-05,
"loss": 5.0504,
"mean_token_accuracy": 0.21481357663869857,
"num_tokens": 16460499.0,
"step": 8400
},
{
"entropy": 5.7632386684417725,
"epoch": 8.355047240179015,
"grad_norm": 1.28125,
"learning_rate": 8.574099154310782e-05,
"loss": 4.8505,
"mean_token_accuracy": 0.2368401512503624,
"num_tokens": 16469829.0,
"step": 8405
},
{
"entropy": 5.845052671432495,
"epoch": 8.360019890601691,
"grad_norm": 1.53125,
"learning_rate": 8.553007835345755e-05,
"loss": 5.0016,
"mean_token_accuracy": 0.214724762737751,
"num_tokens": 16479486.0,
"step": 8410
},
{
"entropy": 5.859267902374268,
"epoch": 8.364992541024366,
"grad_norm": 1.3984375,
"learning_rate": 8.531973596222701e-05,
"loss": 4.9933,
"mean_token_accuracy": 0.2161330461502075,
"num_tokens": 16488644.0,
"step": 8415
},
{
"entropy": 5.932822036743164,
"epoch": 8.369965191447042,
"grad_norm": 1.2265625,
"learning_rate": 8.510996500309531e-05,
"loss": 5.0544,
"mean_token_accuracy": 0.2152727872133255,
"num_tokens": 16500107.0,
"step": 8420
},
{
"entropy": 5.757121992111206,
"epoch": 8.374937841869716,
"grad_norm": 1.2265625,
"learning_rate": 8.490076610801983e-05,
"loss": 4.7148,
"mean_token_accuracy": 0.2418750822544098,
"num_tokens": 16509936.0,
"step": 8425
},
{
"entropy": 5.875438499450683,
"epoch": 8.379910492292392,
"grad_norm": 1.375,
"learning_rate": 8.46921399072346e-05,
"loss": 4.9971,
"mean_token_accuracy": 0.22209224998950958,
"num_tokens": 16519907.0,
"step": 8430
},
{
"entropy": 5.896620893478394,
"epoch": 8.384883142715067,
"grad_norm": 1.4921875,
"learning_rate": 8.448408702924846e-05,
"loss": 4.9808,
"mean_token_accuracy": 0.22669444978237152,
"num_tokens": 16528484.0,
"step": 8435
},
{
"entropy": 5.9206023693084715,
"epoch": 8.389855793137743,
"grad_norm": 1.3671875,
"learning_rate": 8.427660810084292e-05,
"loss": 5.0348,
"mean_token_accuracy": 0.22036661952733994,
"num_tokens": 16538081.0,
"step": 8440
},
{
"entropy": 5.873803949356079,
"epoch": 8.394828443560417,
"grad_norm": 1.4296875,
"learning_rate": 8.406970374707048e-05,
"loss": 4.9743,
"mean_token_accuracy": 0.22271288931369781,
"num_tokens": 16546598.0,
"step": 8445
},
{
"entropy": 5.820176410675049,
"epoch": 8.399801093983093,
"grad_norm": 1.4921875,
"learning_rate": 8.386337459125268e-05,
"loss": 4.9149,
"mean_token_accuracy": 0.22279057204723357,
"num_tokens": 16556312.0,
"step": 8450
},
{
"entropy": 5.793716382980347,
"epoch": 8.404773744405768,
"grad_norm": 1.3203125,
"learning_rate": 8.365762125497814e-05,
"loss": 4.9066,
"mean_token_accuracy": 0.22580559849739074,
"num_tokens": 16566206.0,
"step": 8455
},
{
"entropy": 5.831638669967651,
"epoch": 8.409746394828444,
"grad_norm": 1.3203125,
"learning_rate": 8.345244435810084e-05,
"loss": 4.9693,
"mean_token_accuracy": 0.22660912275314332,
"num_tokens": 16577475.0,
"step": 8460
},
{
"entropy": 5.639604496955871,
"epoch": 8.414719045251118,
"grad_norm": 1.1875,
"learning_rate": 8.324784451873813e-05,
"loss": 4.7325,
"mean_token_accuracy": 0.26055231839418413,
"num_tokens": 16588188.0,
"step": 8465
},
{
"entropy": 5.885994720458984,
"epoch": 8.419691695673794,
"grad_norm": 1.2421875,
"learning_rate": 8.304382235326893e-05,
"loss": 5.0016,
"mean_token_accuracy": 0.21398557126522064,
"num_tokens": 16598000.0,
"step": 8470
},
{
"entropy": 5.81119475364685,
"epoch": 8.424664346096469,
"grad_norm": 1.3203125,
"learning_rate": 8.284037847633189e-05,
"loss": 4.8973,
"mean_token_accuracy": 0.2370400071144104,
"num_tokens": 16608300.0,
"step": 8475
},
{
"entropy": 5.90146803855896,
"epoch": 8.429636996519145,
"grad_norm": 1.3125,
"learning_rate": 8.263751350082345e-05,
"loss": 5.0148,
"mean_token_accuracy": 0.21038416773080826,
"num_tokens": 16618549.0,
"step": 8480
},
{
"entropy": 5.95636739730835,
"epoch": 8.43460964694182,
"grad_norm": 1.3828125,
"learning_rate": 8.243522803789597e-05,
"loss": 5.0668,
"mean_token_accuracy": 0.21344007551670074,
"num_tokens": 16628049.0,
"step": 8485
},
{
"entropy": 5.868502616882324,
"epoch": 8.439582297364495,
"grad_norm": 1.421875,
"learning_rate": 8.22335226969562e-05,
"loss": 4.9473,
"mean_token_accuracy": 0.22385834902524948,
"num_tokens": 16636724.0,
"step": 8490
},
{
"entropy": 5.845301580429077,
"epoch": 8.44455494778717,
"grad_norm": 1.40625,
"learning_rate": 8.203239808566296e-05,
"loss": 4.9727,
"mean_token_accuracy": 0.22365899235010148,
"num_tokens": 16645594.0,
"step": 8495
},
{
"entropy": 5.831522846221924,
"epoch": 8.449527598209846,
"grad_norm": 1.3984375,
"learning_rate": 8.183185480992575e-05,
"loss": 4.9112,
"mean_token_accuracy": 0.23240871876478195,
"num_tokens": 16654219.0,
"step": 8500
},
{
"epoch": 8.449527598209846,
"eval_entropy": 5.852172959235407,
"eval_loss": 6.472639560699463,
"eval_mean_token_accuracy": 0.14877833329003803,
"eval_num_tokens": 16654219.0,
"eval_runtime": 2.7404,
"eval_samples_per_second": 1264.437,
"eval_steps_per_second": 158.374,
"step": 8500
},
{
"entropy": 5.856959009170533,
"epoch": 8.45450024863252,
"grad_norm": 1.375,
"learning_rate": 8.163189347390259e-05,
"loss": 5.0207,
"mean_token_accuracy": 0.21453010588884353,
"num_tokens": 16663187.0,
"step": 8505
},
{
"entropy": 5.890625715255737,
"epoch": 8.459472899055196,
"grad_norm": 1.375,
"learning_rate": 8.14325146799984e-05,
"loss": 4.9953,
"mean_token_accuracy": 0.214973321557045,
"num_tokens": 16673230.0,
"step": 8510
},
{
"entropy": 5.852130126953125,
"epoch": 8.464445549477873,
"grad_norm": 1.3515625,
"learning_rate": 8.123371902886325e-05,
"loss": 4.9047,
"mean_token_accuracy": 0.230030557513237,
"num_tokens": 16682521.0,
"step": 8515
},
{
"entropy": 5.887778043746948,
"epoch": 8.469418199900547,
"grad_norm": 1.375,
"learning_rate": 8.103550711939018e-05,
"loss": 5.025,
"mean_token_accuracy": 0.22103683948516845,
"num_tokens": 16691636.0,
"step": 8520
},
{
"entropy": 5.833680248260498,
"epoch": 8.474390850323223,
"grad_norm": 1.3515625,
"learning_rate": 8.08378795487139e-05,
"loss": 4.8725,
"mean_token_accuracy": 0.23600611984729766,
"num_tokens": 16702141.0,
"step": 8525
},
{
"entropy": 5.848054647445679,
"epoch": 8.479363500745897,
"grad_norm": 1.375,
"learning_rate": 8.064083691220852e-05,
"loss": 5.0266,
"mean_token_accuracy": 0.21496649086475372,
"num_tokens": 16711185.0,
"step": 8530
},
{
"entropy": 5.865668725967407,
"epoch": 8.484336151168574,
"grad_norm": 1.515625,
"learning_rate": 8.044437980348619e-05,
"loss": 4.9261,
"mean_token_accuracy": 0.2310878410935402,
"num_tokens": 16719868.0,
"step": 8535
},
{
"entropy": 5.889631986618042,
"epoch": 8.489308801591248,
"grad_norm": 1.375,
"learning_rate": 8.02485088143949e-05,
"loss": 4.9337,
"mean_token_accuracy": 0.22412277162075042,
"num_tokens": 16730008.0,
"step": 8540
},
{
"entropy": 5.688821458816529,
"epoch": 8.494281452013924,
"grad_norm": 1.3984375,
"learning_rate": 8.005322453501692e-05,
"loss": 4.8523,
"mean_token_accuracy": 0.23872560411691665,
"num_tokens": 16740576.0,
"step": 8545
},
{
"entropy": 5.872801446914673,
"epoch": 8.499254102436598,
"grad_norm": 1.4375,
"learning_rate": 7.985852755366717e-05,
"loss": 5.016,
"mean_token_accuracy": 0.21596289724111556,
"num_tokens": 16749432.0,
"step": 8550
},
{
"entropy": 5.852059984207154,
"epoch": 8.504226752859275,
"grad_norm": 1.4375,
"learning_rate": 7.966441845689102e-05,
"loss": 4.9412,
"mean_token_accuracy": 0.2249598816037178,
"num_tokens": 16759619.0,
"step": 8555
},
{
"entropy": 5.935066223144531,
"epoch": 8.509199403281949,
"grad_norm": 1.1953125,
"learning_rate": 7.947089782946295e-05,
"loss": 5.0682,
"mean_token_accuracy": 0.2175983190536499,
"num_tokens": 16770130.0,
"step": 8560
},
{
"entropy": 5.889373445510865,
"epoch": 8.514172053704625,
"grad_norm": 1.5,
"learning_rate": 7.927796625438466e-05,
"loss": 4.9874,
"mean_token_accuracy": 0.22291228026151658,
"num_tokens": 16780034.0,
"step": 8565
},
{
"entropy": 5.92463321685791,
"epoch": 8.5191447041273,
"grad_norm": 1.203125,
"learning_rate": 7.908562431288303e-05,
"loss": 5.0508,
"mean_token_accuracy": 0.2087842345237732,
"num_tokens": 16790187.0,
"step": 8570
},
{
"entropy": 5.891081047058106,
"epoch": 8.524117354549976,
"grad_norm": 1.359375,
"learning_rate": 7.889387258440884e-05,
"loss": 4.9894,
"mean_token_accuracy": 0.2183925613760948,
"num_tokens": 16799586.0,
"step": 8575
},
{
"entropy": 5.911332130432129,
"epoch": 8.52909000497265,
"grad_norm": 1.4765625,
"learning_rate": 7.870271164663463e-05,
"loss": 5.0156,
"mean_token_accuracy": 0.2206865817308426,
"num_tokens": 16809230.0,
"step": 8580
},
{
"entropy": 5.847852325439453,
"epoch": 8.534062655395326,
"grad_norm": 1.3671875,
"learning_rate": 7.851214207545322e-05,
"loss": 4.9504,
"mean_token_accuracy": 0.22172896862030028,
"num_tokens": 16819087.0,
"step": 8585
},
{
"entropy": 5.78146653175354,
"epoch": 8.539035305818,
"grad_norm": 1.515625,
"learning_rate": 7.832216444497589e-05,
"loss": 4.8415,
"mean_token_accuracy": 0.2411959484219551,
"num_tokens": 16827600.0,
"step": 8590
},
{
"entropy": 5.868714618682861,
"epoch": 8.544007956240677,
"grad_norm": 1.21875,
"learning_rate": 7.813277932753053e-05,
"loss": 5.0393,
"mean_token_accuracy": 0.20716336220502854,
"num_tokens": 16837300.0,
"step": 8595
},
{
"entropy": 5.850877952575684,
"epoch": 8.548980606663351,
"grad_norm": 1.3671875,
"learning_rate": 7.794398729366006e-05,
"loss": 4.9904,
"mean_token_accuracy": 0.2217344105243683,
"num_tokens": 16846575.0,
"step": 8600
},
{
"entropy": 5.808744955062866,
"epoch": 8.553953257086027,
"grad_norm": 1.4296875,
"learning_rate": 7.775578891212076e-05,
"loss": 4.8533,
"mean_token_accuracy": 0.2450559914112091,
"num_tokens": 16856668.0,
"step": 8605
},
{
"entropy": 5.911669540405273,
"epoch": 8.558925907508701,
"grad_norm": 1.28125,
"learning_rate": 7.75681847498803e-05,
"loss": 5.0114,
"mean_token_accuracy": 0.21462003588676454,
"num_tokens": 16867100.0,
"step": 8610
},
{
"entropy": 5.894762086868286,
"epoch": 8.563898557931378,
"grad_norm": 1.421875,
"learning_rate": 7.73811753721164e-05,
"loss": 5.0053,
"mean_token_accuracy": 0.21599971801042556,
"num_tokens": 16876085.0,
"step": 8615
},
{
"entropy": 5.827739667892456,
"epoch": 8.568871208354052,
"grad_norm": 1.4296875,
"learning_rate": 7.719476134221474e-05,
"loss": 4.9454,
"mean_token_accuracy": 0.2258874148130417,
"num_tokens": 16885050.0,
"step": 8620
},
{
"entropy": 5.8250593662261965,
"epoch": 8.573843858776728,
"grad_norm": 1.4375,
"learning_rate": 7.70089432217676e-05,
"loss": 4.9256,
"mean_token_accuracy": 0.23401083946228027,
"num_tokens": 16893838.0,
"step": 8625
},
{
"entropy": 5.818560981750489,
"epoch": 8.578816509199402,
"grad_norm": 1.4296875,
"learning_rate": 7.682372157057196e-05,
"loss": 4.9163,
"mean_token_accuracy": 0.2306785762310028,
"num_tokens": 16903238.0,
"step": 8630
},
{
"entropy": 5.868492317199707,
"epoch": 8.583789159622079,
"grad_norm": 1.34375,
"learning_rate": 7.663909694662783e-05,
"loss": 4.925,
"mean_token_accuracy": 0.22336995154619216,
"num_tokens": 16913151.0,
"step": 8635
},
{
"entropy": 5.923525762557984,
"epoch": 8.588761810044755,
"grad_norm": 1.65625,
"learning_rate": 7.64550699061367e-05,
"loss": 5.02,
"mean_token_accuracy": 0.2161752998828888,
"num_tokens": 16921913.0,
"step": 8640
},
{
"entropy": 5.890104103088379,
"epoch": 8.593734460467429,
"grad_norm": 1.40625,
"learning_rate": 7.627164100349967e-05,
"loss": 4.9963,
"mean_token_accuracy": 0.22021807581186295,
"num_tokens": 16932118.0,
"step": 8645
},
{
"entropy": 5.848131608963013,
"epoch": 8.598707110890105,
"grad_norm": 1.4921875,
"learning_rate": 7.608881079131604e-05,
"loss": 4.8631,
"mean_token_accuracy": 0.23089140355587007,
"num_tokens": 16941100.0,
"step": 8650
},
{
"entropy": 5.766698598861694,
"epoch": 8.60367976131278,
"grad_norm": 1.4296875,
"learning_rate": 7.59065798203813e-05,
"loss": 4.8587,
"mean_token_accuracy": 0.23419751822948456,
"num_tokens": 16950836.0,
"step": 8655
},
{
"entropy": 5.906005573272705,
"epoch": 8.608652411735456,
"grad_norm": 1.5234375,
"learning_rate": 7.572494863968573e-05,
"loss": 5.0149,
"mean_token_accuracy": 0.2223264068365097,
"num_tokens": 16960062.0,
"step": 8660
},
{
"entropy": 5.82679328918457,
"epoch": 8.61362506215813,
"grad_norm": 1.1796875,
"learning_rate": 7.55439177964128e-05,
"loss": 4.9372,
"mean_token_accuracy": 0.23389201909303664,
"num_tokens": 16970319.0,
"step": 8665
},
{
"entropy": 5.692574167251587,
"epoch": 8.618597712580806,
"grad_norm": 1.09375,
"learning_rate": 7.536348783593727e-05,
"loss": 4.8589,
"mean_token_accuracy": 0.24206701219081878,
"num_tokens": 16982193.0,
"step": 8670
},
{
"entropy": 5.826966142654419,
"epoch": 8.62357036300348,
"grad_norm": 1.5390625,
"learning_rate": 7.518365930182368e-05,
"loss": 4.834,
"mean_token_accuracy": 0.23785837292671203,
"num_tokens": 16991786.0,
"step": 8675
},
{
"entropy": 5.934372472763061,
"epoch": 8.628543013426157,
"grad_norm": 1.3515625,
"learning_rate": 7.500443273582478e-05,
"loss": 5.0833,
"mean_token_accuracy": 0.20825429111719132,
"num_tokens": 17001588.0,
"step": 8680
},
{
"entropy": 5.897218179702759,
"epoch": 8.633515663848831,
"grad_norm": 1.3046875,
"learning_rate": 7.482580867787974e-05,
"loss": 4.9515,
"mean_token_accuracy": 0.2214104413986206,
"num_tokens": 17010997.0,
"step": 8685
},
{
"entropy": 5.8544776916503904,
"epoch": 8.638488314271507,
"grad_norm": 1.3828125,
"learning_rate": 7.464778766611266e-05,
"loss": 4.9411,
"mean_token_accuracy": 0.2211606726050377,
"num_tokens": 17019596.0,
"step": 8690
},
{
"entropy": 5.878175544738769,
"epoch": 8.643460964694182,
"grad_norm": 1.4453125,
"learning_rate": 7.447037023683095e-05,
"loss": 4.9943,
"mean_token_accuracy": 0.21957664489746093,
"num_tokens": 17029911.0,
"step": 8695
},
{
"entropy": 5.88730754852295,
"epoch": 8.648433615116858,
"grad_norm": 1.4140625,
"learning_rate": 7.429355692452355e-05,
"loss": 4.9908,
"mean_token_accuracy": 0.21888842731714248,
"num_tokens": 17039300.0,
"step": 8700
},
{
"entropy": 5.854285526275635,
"epoch": 8.653406265539532,
"grad_norm": 1.2109375,
"learning_rate": 7.411734826185953e-05,
"loss": 4.9645,
"mean_token_accuracy": 0.2235252007842064,
"num_tokens": 17049297.0,
"step": 8705
},
{
"entropy": 5.824087715148925,
"epoch": 8.658378915962208,
"grad_norm": 1.3359375,
"learning_rate": 7.394174477968632e-05,
"loss": 4.9755,
"mean_token_accuracy": 0.22011736333370208,
"num_tokens": 17059310.0,
"step": 8710
},
{
"entropy": 5.765740060806275,
"epoch": 8.663351566384883,
"grad_norm": 1.4140625,
"learning_rate": 7.376674700702817e-05,
"loss": 4.8815,
"mean_token_accuracy": 0.22945040017366408,
"num_tokens": 17068421.0,
"step": 8715
},
{
"entropy": 5.7988028049469,
"epoch": 8.668324216807559,
"grad_norm": 1.3515625,
"learning_rate": 7.359235547108469e-05,
"loss": 4.9546,
"mean_token_accuracy": 0.22230857312679292,
"num_tokens": 17078085.0,
"step": 8720
},
{
"entropy": 5.846247577667237,
"epoch": 8.673296867230233,
"grad_norm": 1.359375,
"learning_rate": 7.341857069722897e-05,
"loss": 4.9461,
"mean_token_accuracy": 0.22951264828443527,
"num_tokens": 17087333.0,
"step": 8725
},
{
"entropy": 5.812128353118896,
"epoch": 8.67826951765291,
"grad_norm": 1.4296875,
"learning_rate": 7.324539320900632e-05,
"loss": 4.9111,
"mean_token_accuracy": 0.22971128821372985,
"num_tokens": 17096963.0,
"step": 8730
},
{
"entropy": 5.932662439346314,
"epoch": 8.683242168075584,
"grad_norm": 1.3828125,
"learning_rate": 7.307282352813237e-05,
"loss": 5.0752,
"mean_token_accuracy": 0.20765840858221055,
"num_tokens": 17107332.0,
"step": 8735
},
{
"entropy": 5.835472774505615,
"epoch": 8.68821481849826,
"grad_norm": 1.2890625,
"learning_rate": 7.290086217449176e-05,
"loss": 4.9934,
"mean_token_accuracy": 0.22949409782886504,
"num_tokens": 17117714.0,
"step": 8740
},
{
"entropy": 5.802573919296265,
"epoch": 8.693187468920934,
"grad_norm": 1.40625,
"learning_rate": 7.272950966613659e-05,
"loss": 4.8954,
"mean_token_accuracy": 0.23576511293649674,
"num_tokens": 17127282.0,
"step": 8745
},
{
"entropy": 5.829149580001831,
"epoch": 8.69816011934361,
"grad_norm": 1.453125,
"learning_rate": 7.255876651928453e-05,
"loss": 4.981,
"mean_token_accuracy": 0.21911141276359558,
"num_tokens": 17136978.0,
"step": 8750
},
{
"entropy": 5.923779916763306,
"epoch": 8.703132769766285,
"grad_norm": 1.3125,
"learning_rate": 7.238863324831762e-05,
"loss": 5.0553,
"mean_token_accuracy": 0.21307908296585082,
"num_tokens": 17146930.0,
"step": 8755
},
{
"entropy": 5.915111923217774,
"epoch": 8.70810542018896,
"grad_norm": 1.296875,
"learning_rate": 7.221911036578053e-05,
"loss": 5.0797,
"mean_token_accuracy": 0.21218951940536498,
"num_tokens": 17157463.0,
"step": 8760
},
{
"entropy": 5.806950521469116,
"epoch": 8.713078070611637,
"grad_norm": 1.421875,
"learning_rate": 7.205019838237916e-05,
"loss": 4.8624,
"mean_token_accuracy": 0.2369381755590439,
"num_tokens": 17167881.0,
"step": 8765
},
{
"entropy": 5.87104640007019,
"epoch": 8.718050721034311,
"grad_norm": 1.484375,
"learning_rate": 7.188189780697885e-05,
"loss": 5.0508,
"mean_token_accuracy": 0.2117942601442337,
"num_tokens": 17178077.0,
"step": 8770
},
{
"entropy": 5.926954650878907,
"epoch": 8.723023371456987,
"grad_norm": 1.40625,
"learning_rate": 7.171420914660308e-05,
"loss": 5.026,
"mean_token_accuracy": 0.21826636791229248,
"num_tokens": 17187600.0,
"step": 8775
},
{
"entropy": 5.8885969638824465,
"epoch": 8.727996021879662,
"grad_norm": 1.3984375,
"learning_rate": 7.154713290643195e-05,
"loss": 5.0711,
"mean_token_accuracy": 0.21556093245744706,
"num_tokens": 17197798.0,
"step": 8780
},
{
"entropy": 5.90594072341919,
"epoch": 8.732968672302338,
"grad_norm": 1.28125,
"learning_rate": 7.13806695898006e-05,
"loss": 5.0618,
"mean_token_accuracy": 0.20540596842765807,
"num_tokens": 17208145.0,
"step": 8785
},
{
"entropy": 5.879127073287964,
"epoch": 8.737941322725012,
"grad_norm": 1.375,
"learning_rate": 7.121481969819739e-05,
"loss": 5.0212,
"mean_token_accuracy": 0.21743432581424713,
"num_tokens": 17218468.0,
"step": 8790
},
{
"entropy": 5.889526081085205,
"epoch": 8.742913973147688,
"grad_norm": 1.390625,
"learning_rate": 7.104958373126306e-05,
"loss": 5.0333,
"mean_token_accuracy": 0.21348718106746672,
"num_tokens": 17228361.0,
"step": 8795
},
{
"entropy": 5.876287126541138,
"epoch": 8.747886623570363,
"grad_norm": 1.3203125,
"learning_rate": 7.08849621867885e-05,
"loss": 5.0439,
"mean_token_accuracy": 0.2186755880713463,
"num_tokens": 17238579.0,
"step": 8800
},
{
"entropy": 5.893976306915283,
"epoch": 8.752859273993039,
"grad_norm": 1.2734375,
"learning_rate": 7.072095556071381e-05,
"loss": 5.0215,
"mean_token_accuracy": 0.22282694280147552,
"num_tokens": 17248388.0,
"step": 8805
},
{
"entropy": 5.84912109375,
"epoch": 8.757831924415713,
"grad_norm": 1.3515625,
"learning_rate": 7.055756434712653e-05,
"loss": 4.9217,
"mean_token_accuracy": 0.23460158556699753,
"num_tokens": 17256837.0,
"step": 8810
},
{
"entropy": 5.724826955795288,
"epoch": 8.76280457483839,
"grad_norm": 1.3984375,
"learning_rate": 7.039478903826012e-05,
"loss": 4.7912,
"mean_token_accuracy": 0.24888089150190354,
"num_tokens": 17266636.0,
"step": 8815
},
{
"entropy": 5.917354536056519,
"epoch": 8.767777225261064,
"grad_norm": 1.40625,
"learning_rate": 7.023263012449269e-05,
"loss": 5.0838,
"mean_token_accuracy": 0.21234248876571654,
"num_tokens": 17276514.0,
"step": 8820
},
{
"entropy": 5.881424903869629,
"epoch": 8.77274987568374,
"grad_norm": 1.3984375,
"learning_rate": 7.007108809434527e-05,
"loss": 4.9189,
"mean_token_accuracy": 0.227752947807312,
"num_tokens": 17286079.0,
"step": 8825
},
{
"entropy": 5.923693513870239,
"epoch": 8.777722526106414,
"grad_norm": 1.2578125,
"learning_rate": 6.99101634344805e-05,
"loss": 5.0531,
"mean_token_accuracy": 0.21691745966672898,
"num_tokens": 17296263.0,
"step": 8830
},
{
"entropy": 5.907936382293701,
"epoch": 8.78269517652909,
"grad_norm": 1.6015625,
"learning_rate": 6.974985662970119e-05,
"loss": 4.9983,
"mean_token_accuracy": 0.21573276072740555,
"num_tokens": 17306212.0,
"step": 8835
},
{
"entropy": 5.883978462219238,
"epoch": 8.787667826951765,
"grad_norm": 1.5078125,
"learning_rate": 6.959016816294872e-05,
"loss": 5.0147,
"mean_token_accuracy": 0.2157219260931015,
"num_tokens": 17316333.0,
"step": 8840
},
{
"entropy": 5.884533977508545,
"epoch": 8.79264047737444,
"grad_norm": 1.4609375,
"learning_rate": 6.943109851530168e-05,
"loss": 4.9473,
"mean_token_accuracy": 0.22205858826637268,
"num_tokens": 17326549.0,
"step": 8845
},
{
"entropy": 5.844780349731446,
"epoch": 8.797613127797115,
"grad_norm": 1.53125,
"learning_rate": 6.927264816597448e-05,
"loss": 4.938,
"mean_token_accuracy": 0.22451443672180177,
"num_tokens": 17334928.0,
"step": 8850
},
{
"entropy": 5.871609258651733,
"epoch": 8.802585778219791,
"grad_norm": 1.2578125,
"learning_rate": 6.911481759231566e-05,
"loss": 5.0034,
"mean_token_accuracy": 0.22189999520778655,
"num_tokens": 17345911.0,
"step": 8855
},
{
"entropy": 5.850828504562378,
"epoch": 8.807558428642466,
"grad_norm": 1.3671875,
"learning_rate": 6.895760726980681e-05,
"loss": 4.951,
"mean_token_accuracy": 0.22355148941278458,
"num_tokens": 17355737.0,
"step": 8860
},
{
"entropy": 5.860705137252808,
"epoch": 8.812531079065142,
"grad_norm": 1.46875,
"learning_rate": 6.880101767206079e-05,
"loss": 4.9496,
"mean_token_accuracy": 0.22150055468082427,
"num_tokens": 17364237.0,
"step": 8865
},
{
"entropy": 5.908811712265015,
"epoch": 8.817503729487816,
"grad_norm": 1.3515625,
"learning_rate": 6.864504927082051e-05,
"loss": 5.071,
"mean_token_accuracy": 0.21113202571868897,
"num_tokens": 17374606.0,
"step": 8870
},
{
"entropy": 5.856512975692749,
"epoch": 8.822476379910492,
"grad_norm": 1.453125,
"learning_rate": 6.848970253595759e-05,
"loss": 4.9263,
"mean_token_accuracy": 0.2259698674082756,
"num_tokens": 17384996.0,
"step": 8875
},
{
"entropy": 5.919076538085937,
"epoch": 8.827449030333167,
"grad_norm": 1.46875,
"learning_rate": 6.833497793547059e-05,
"loss": 5.0457,
"mean_token_accuracy": 0.2135239213705063,
"num_tokens": 17394941.0,
"step": 8880
},
{
"entropy": 5.939565563201905,
"epoch": 8.832421680755843,
"grad_norm": 1.390625,
"learning_rate": 6.818087593548391e-05,
"loss": 5.0789,
"mean_token_accuracy": 0.20851074010133744,
"num_tokens": 17404796.0,
"step": 8885
},
{
"entropy": 5.881474590301513,
"epoch": 8.837394331178519,
"grad_norm": 1.3671875,
"learning_rate": 6.802739700024643e-05,
"loss": 4.9958,
"mean_token_accuracy": 0.21569311916828154,
"num_tokens": 17414975.0,
"step": 8890
},
{
"entropy": 5.908678722381592,
"epoch": 8.842366981601193,
"grad_norm": 1.40625,
"learning_rate": 6.787454159212974e-05,
"loss": 5.0922,
"mean_token_accuracy": 0.2121524766087532,
"num_tokens": 17423210.0,
"step": 8895
},
{
"entropy": 5.845994901657105,
"epoch": 8.84733963202387,
"grad_norm": 1.234375,
"learning_rate": 6.772231017162725e-05,
"loss": 4.988,
"mean_token_accuracy": 0.21634145230054855,
"num_tokens": 17433692.0,
"step": 8900
},
{
"entropy": 5.872529363632202,
"epoch": 8.852312282446544,
"grad_norm": 1.3125,
"learning_rate": 6.75707031973523e-05,
"loss": 5.0542,
"mean_token_accuracy": 0.21296356469392777,
"num_tokens": 17443798.0,
"step": 8905
},
{
"entropy": 5.667204093933106,
"epoch": 8.85728493286922,
"grad_norm": 1.4375,
"learning_rate": 6.741972112603719e-05,
"loss": 4.76,
"mean_token_accuracy": 0.24766640812158586,
"num_tokens": 17454009.0,
"step": 8910
},
{
"entropy": 5.909105587005615,
"epoch": 8.862257583291894,
"grad_norm": 1.3125,
"learning_rate": 6.726936441253155e-05,
"loss": 5.0904,
"mean_token_accuracy": 0.21590953022241594,
"num_tokens": 17464323.0,
"step": 8915
},
{
"entropy": 5.7443962574005125,
"epoch": 8.86723023371457,
"grad_norm": 1.421875,
"learning_rate": 6.711963350980106e-05,
"loss": 4.8115,
"mean_token_accuracy": 0.2412910506129265,
"num_tokens": 17474408.0,
"step": 8920
},
{
"entropy": 5.83795166015625,
"epoch": 8.872202884137245,
"grad_norm": 1.296875,
"learning_rate": 6.697052886892614e-05,
"loss": 4.9705,
"mean_token_accuracy": 0.22096491158008574,
"num_tokens": 17483890.0,
"step": 8925
},
{
"entropy": 5.855341482162475,
"epoch": 8.877175534559921,
"grad_norm": 1.453125,
"learning_rate": 6.682205093910044e-05,
"loss": 4.9472,
"mean_token_accuracy": 0.23195761442184448,
"num_tokens": 17493275.0,
"step": 8930
},
{
"entropy": 5.87120885848999,
"epoch": 8.882148184982595,
"grad_norm": 1.3828125,
"learning_rate": 6.667420016762963e-05,
"loss": 5.0371,
"mean_token_accuracy": 0.21926995515823364,
"num_tokens": 17502858.0,
"step": 8935
},
{
"entropy": 5.895511198043823,
"epoch": 8.887120835405272,
"grad_norm": 1.46875,
"learning_rate": 6.652697699993e-05,
"loss": 5.0214,
"mean_token_accuracy": 0.21559280008077622,
"num_tokens": 17511700.0,
"step": 8940
},
{
"entropy": 5.833205366134644,
"epoch": 8.892093485827946,
"grad_norm": 1.46875,
"learning_rate": 6.638038187952705e-05,
"loss": 4.9845,
"mean_token_accuracy": 0.22190120965242385,
"num_tokens": 17521387.0,
"step": 8945
},
{
"entropy": 5.684017181396484,
"epoch": 8.897066136250622,
"grad_norm": 1.390625,
"learning_rate": 6.623441524805433e-05,
"loss": 4.7569,
"mean_token_accuracy": 0.24132902771234513,
"num_tokens": 17530569.0,
"step": 8950
},
{
"entropy": 5.762717962265015,
"epoch": 8.902038786673296,
"grad_norm": 1.3828125,
"learning_rate": 6.608907754525189e-05,
"loss": 4.8882,
"mean_token_accuracy": 0.23049553334712983,
"num_tokens": 17540236.0,
"step": 8955
},
{
"entropy": 5.8401100635528564,
"epoch": 8.907011437095973,
"grad_norm": 1.53125,
"learning_rate": 6.594436920896511e-05,
"loss": 4.9939,
"mean_token_accuracy": 0.21562524288892745,
"num_tokens": 17549257.0,
"step": 8960
},
{
"entropy": 5.873914575576782,
"epoch": 8.911984087518647,
"grad_norm": 1.234375,
"learning_rate": 6.580029067514346e-05,
"loss": 4.9702,
"mean_token_accuracy": 0.22358780801296235,
"num_tokens": 17560089.0,
"step": 8965
},
{
"entropy": 5.722599458694458,
"epoch": 8.916956737941323,
"grad_norm": 1.3984375,
"learning_rate": 6.565684237783881e-05,
"loss": 4.8016,
"mean_token_accuracy": 0.24344651103019715,
"num_tokens": 17569841.0,
"step": 8970
},
{
"entropy": 5.864724254608154,
"epoch": 8.921929388363997,
"grad_norm": 1.4296875,
"learning_rate": 6.551402474920458e-05,
"loss": 4.9438,
"mean_token_accuracy": 0.21990772783756257,
"num_tokens": 17579049.0,
"step": 8975
},
{
"entropy": 5.976896619796753,
"epoch": 8.926902038786674,
"grad_norm": 1.40625,
"learning_rate": 6.537183821949412e-05,
"loss": 5.0934,
"mean_token_accuracy": 0.21207041293382645,
"num_tokens": 17588278.0,
"step": 8980
},
{
"entropy": 5.8569488525390625,
"epoch": 8.931874689209348,
"grad_norm": 1.3671875,
"learning_rate": 6.52302832170596e-05,
"loss": 4.9715,
"mean_token_accuracy": 0.22435979098081588,
"num_tokens": 17598579.0,
"step": 8985
},
{
"entropy": 5.900597763061524,
"epoch": 8.936847339632024,
"grad_norm": 1.4453125,
"learning_rate": 6.508936016835063e-05,
"loss": 4.977,
"mean_token_accuracy": 0.22531002759933472,
"num_tokens": 17607855.0,
"step": 8990
},
{
"entropy": 5.849305868148804,
"epoch": 8.941819990054698,
"grad_norm": 1.3515625,
"learning_rate": 6.494906949791296e-05,
"loss": 4.9824,
"mean_token_accuracy": 0.21982925981283188,
"num_tokens": 17618553.0,
"step": 8995
},
{
"entropy": 5.896419334411621,
"epoch": 8.946792640477375,
"grad_norm": 1.46875,
"learning_rate": 6.480941162838724e-05,
"loss": 4.9985,
"mean_token_accuracy": 0.2228580430150032,
"num_tokens": 17627955.0,
"step": 9000
},
{
"epoch": 8.946792640477375,
"eval_entropy": 5.858961294323618,
"eval_loss": 6.469855785369873,
"eval_mean_token_accuracy": 0.14936679102961095,
"eval_num_tokens": 17627955.0,
"eval_runtime": 2.5465,
"eval_samples_per_second": 1360.698,
"eval_steps_per_second": 170.431,
"step": 9000
},
{
"entropy": 5.802471113204956,
"epoch": 8.951765290900049,
"grad_norm": 1.3984375,
"learning_rate": 6.46703869805078e-05,
"loss": 4.916,
"mean_token_accuracy": 0.22271430343389512,
"num_tokens": 17638155.0,
"step": 9005
},
{
"entropy": 5.911737489700317,
"epoch": 8.956737941322725,
"grad_norm": 1.3984375,
"learning_rate": 6.453199597310119e-05,
"loss": 5.0714,
"mean_token_accuracy": 0.21088473498821259,
"num_tokens": 17647733.0,
"step": 9010
},
{
"entropy": 5.9111415386199955,
"epoch": 8.961710591745401,
"grad_norm": 1.421875,
"learning_rate": 6.439423902308521e-05,
"loss": 5.0414,
"mean_token_accuracy": 0.21871396750211716,
"num_tokens": 17657700.0,
"step": 9015
},
{
"entropy": 5.789221525192261,
"epoch": 8.966683242168076,
"grad_norm": 1.4921875,
"learning_rate": 6.425711654546736e-05,
"loss": 4.8898,
"mean_token_accuracy": 0.23082289099693298,
"num_tokens": 17666151.0,
"step": 9020
},
{
"entropy": 5.850829935073852,
"epoch": 8.971655892590752,
"grad_norm": 1.5390625,
"learning_rate": 6.41206289533438e-05,
"loss": 4.9189,
"mean_token_accuracy": 0.23067329972982406,
"num_tokens": 17675342.0,
"step": 9025
},
{
"entropy": 5.753431415557861,
"epoch": 8.976628543013426,
"grad_norm": 1.40625,
"learning_rate": 6.398477665789803e-05,
"loss": 4.8004,
"mean_token_accuracy": 0.2419154405593872,
"num_tokens": 17685400.0,
"step": 9030
},
{
"entropy": 5.868270063400269,
"epoch": 8.981601193436102,
"grad_norm": 1.2265625,
"learning_rate": 6.384956006839961e-05,
"loss": 5.0163,
"mean_token_accuracy": 0.21563644409179689,
"num_tokens": 17695439.0,
"step": 9035
},
{
"entropy": 5.930341815948486,
"epoch": 8.986573843858777,
"grad_norm": 1.40625,
"learning_rate": 6.371497959220304e-05,
"loss": 5.0787,
"mean_token_accuracy": 0.21244439631700515,
"num_tokens": 17704790.0,
"step": 9040
},
{
"entropy": 5.871539545059204,
"epoch": 8.991546494281453,
"grad_norm": 1.2265625,
"learning_rate": 6.358103563474632e-05,
"loss": 4.9261,
"mean_token_accuracy": 0.23616279810667037,
"num_tokens": 17714496.0,
"step": 9045
},
{
"entropy": 5.873043346405029,
"epoch": 8.996519144704127,
"grad_norm": 1.1640625,
"learning_rate": 6.344772859955005e-05,
"loss": 4.9647,
"mean_token_accuracy": 0.2264036551117897,
"num_tokens": 17724341.0,
"step": 9050
},
{
"entropy": 5.832018587324354,
"epoch": 9.000994530084535,
"grad_norm": 1.2578125,
"learning_rate": 6.331505888821592e-05,
"loss": 4.8664,
"mean_token_accuracy": 0.23368741240766314,
"num_tokens": 17733123.0,
"step": 9055
},
{
"entropy": 5.819580125808716,
"epoch": 9.005967180507211,
"grad_norm": 1.46875,
"learning_rate": 6.318302690042557e-05,
"loss": 4.8491,
"mean_token_accuracy": 0.2382387936115265,
"num_tokens": 17742499.0,
"step": 9060
},
{
"entropy": 5.849063205718994,
"epoch": 9.010939830929885,
"grad_norm": 1.203125,
"learning_rate": 6.305163303393955e-05,
"loss": 4.9875,
"mean_token_accuracy": 0.21512305587530137,
"num_tokens": 17752748.0,
"step": 9065
},
{
"entropy": 5.871422052383423,
"epoch": 9.015912481352562,
"grad_norm": 1.3203125,
"learning_rate": 6.292087768459597e-05,
"loss": 4.9249,
"mean_token_accuracy": 0.23275353610515595,
"num_tokens": 17761938.0,
"step": 9070
},
{
"entropy": 5.938288402557373,
"epoch": 9.020885131775236,
"grad_norm": 1.1484375,
"learning_rate": 6.279076124630922e-05,
"loss": 5.0282,
"mean_token_accuracy": 0.21290740966796876,
"num_tokens": 17771976.0,
"step": 9075
},
{
"entropy": 5.835073471069336,
"epoch": 9.025857782197912,
"grad_norm": 1.171875,
"learning_rate": 6.266128411106915e-05,
"loss": 4.9421,
"mean_token_accuracy": 0.21951935142278672,
"num_tokens": 17781308.0,
"step": 9080
},
{
"entropy": 5.852858161926269,
"epoch": 9.030830432620586,
"grad_norm": 1.328125,
"learning_rate": 6.253244666893933e-05,
"loss": 4.9915,
"mean_token_accuracy": 0.21617018133401872,
"num_tokens": 17791958.0,
"step": 9085
},
{
"entropy": 5.904571676254273,
"epoch": 9.035803083043263,
"grad_norm": 1.34375,
"learning_rate": 6.240424930805648e-05,
"loss": 5.0101,
"mean_token_accuracy": 0.21665038168430328,
"num_tokens": 17802103.0,
"step": 9090
},
{
"entropy": 5.783633375167847,
"epoch": 9.040775733465937,
"grad_norm": 1.6015625,
"learning_rate": 6.227669241462878e-05,
"loss": 4.7649,
"mean_token_accuracy": 0.24207005202770232,
"num_tokens": 17810117.0,
"step": 9095
},
{
"entropy": 5.887320184707642,
"epoch": 9.045748383888613,
"grad_norm": 1.375,
"learning_rate": 6.214977637293502e-05,
"loss": 4.9169,
"mean_token_accuracy": 0.2271772801876068,
"num_tokens": 17820655.0,
"step": 9100
},
{
"entropy": 5.874295377731324,
"epoch": 9.050721034311287,
"grad_norm": 1.4921875,
"learning_rate": 6.202350156532345e-05,
"loss": 4.914,
"mean_token_accuracy": 0.22540174126625062,
"num_tokens": 17830718.0,
"step": 9105
},
{
"entropy": 5.830042123794556,
"epoch": 9.055693684733964,
"grad_norm": 1.390625,
"learning_rate": 6.189786837221042e-05,
"loss": 4.9502,
"mean_token_accuracy": 0.21666134148836136,
"num_tokens": 17840487.0,
"step": 9110
},
{
"entropy": 5.8738508224487305,
"epoch": 9.060666335156638,
"grad_norm": 1.4375,
"learning_rate": 6.177287717207928e-05,
"loss": 4.9617,
"mean_token_accuracy": 0.2270662158727646,
"num_tokens": 17850425.0,
"step": 9115
},
{
"entropy": 5.75973744392395,
"epoch": 9.065638985579314,
"grad_norm": 1.2265625,
"learning_rate": 6.16485283414795e-05,
"loss": 4.7086,
"mean_token_accuracy": 0.24778149425983428,
"num_tokens": 17860538.0,
"step": 9120
},
{
"entropy": 5.924504137039184,
"epoch": 9.070611636001988,
"grad_norm": 1.4453125,
"learning_rate": 6.152482225502518e-05,
"loss": 5.0483,
"mean_token_accuracy": 0.21321818828582764,
"num_tokens": 17868969.0,
"step": 9125
},
{
"entropy": 5.786190605163574,
"epoch": 9.075584286424665,
"grad_norm": 1.3828125,
"learning_rate": 6.140175928539419e-05,
"loss": 4.8391,
"mean_token_accuracy": 0.23929810672998428,
"num_tokens": 17878851.0,
"step": 9130
},
{
"entropy": 5.812713670730591,
"epoch": 9.080556936847339,
"grad_norm": 1.421875,
"learning_rate": 6.127933980332685e-05,
"loss": 4.8609,
"mean_token_accuracy": 0.23598398864269257,
"num_tokens": 17888170.0,
"step": 9135
},
{
"entropy": 5.800144290924072,
"epoch": 9.085529587270015,
"grad_norm": 1.1796875,
"learning_rate": 6.1157564177625e-05,
"loss": 4.8984,
"mean_token_accuracy": 0.23315121829509736,
"num_tokens": 17898699.0,
"step": 9140
},
{
"entropy": 5.724971342086792,
"epoch": 9.09050223769269,
"grad_norm": 1.5234375,
"learning_rate": 6.103643277515075e-05,
"loss": 4.7937,
"mean_token_accuracy": 0.23936803489923478,
"num_tokens": 17908253.0,
"step": 9145
},
{
"entropy": 5.7113439559936525,
"epoch": 9.095474888115366,
"grad_norm": 1.1171875,
"learning_rate": 6.0915945960825393e-05,
"loss": 4.7821,
"mean_token_accuracy": 0.24724831730127333,
"num_tokens": 17918652.0,
"step": 9150
},
{
"entropy": 5.71295132637024,
"epoch": 9.100447538538042,
"grad_norm": 1.3203125,
"learning_rate": 6.079610409762836e-05,
"loss": 4.7176,
"mean_token_accuracy": 0.24649962931871414,
"num_tokens": 17929355.0,
"step": 9155
},
{
"entropy": 5.8472693920135494,
"epoch": 9.105420188960716,
"grad_norm": 1.4140625,
"learning_rate": 6.067690754659605e-05,
"loss": 4.9472,
"mean_token_accuracy": 0.2325298011302948,
"num_tokens": 17937804.0,
"step": 9160
},
{
"entropy": 5.8460310935974125,
"epoch": 9.110392839383392,
"grad_norm": 1.484375,
"learning_rate": 6.055835666682093e-05,
"loss": 5.0186,
"mean_token_accuracy": 0.2140743538737297,
"num_tokens": 17948323.0,
"step": 9165
},
{
"entropy": 5.8350855827331545,
"epoch": 9.115365489806067,
"grad_norm": 1.5234375,
"learning_rate": 6.044045181545013e-05,
"loss": 4.9311,
"mean_token_accuracy": 0.2290230870246887,
"num_tokens": 17957864.0,
"step": 9170
},
{
"entropy": 5.817792224884033,
"epoch": 9.120338140228743,
"grad_norm": 1.2890625,
"learning_rate": 6.032319334768465e-05,
"loss": 4.8945,
"mean_token_accuracy": 0.23186387568712236,
"num_tokens": 17967023.0,
"step": 9175
},
{
"entropy": 5.874756574630737,
"epoch": 9.125310790651417,
"grad_norm": 1.3828125,
"learning_rate": 6.020658161677821e-05,
"loss": 5.0407,
"mean_token_accuracy": 0.21613139808177947,
"num_tokens": 17977762.0,
"step": 9180
},
{
"entropy": 5.702737498283386,
"epoch": 9.130283441074093,
"grad_norm": 1.3671875,
"learning_rate": 6.009061697403609e-05,
"loss": 4.8303,
"mean_token_accuracy": 0.2428809091448784,
"num_tokens": 17986989.0,
"step": 9185
},
{
"entropy": 5.799012136459351,
"epoch": 9.135256091496768,
"grad_norm": 1.421875,
"learning_rate": 5.9975299768814204e-05,
"loss": 4.8229,
"mean_token_accuracy": 0.23819281458854674,
"num_tokens": 17997662.0,
"step": 9190
},
{
"entropy": 5.853587579727173,
"epoch": 9.140228741919444,
"grad_norm": 1.359375,
"learning_rate": 5.9860630348518004e-05,
"loss": 4.9457,
"mean_token_accuracy": 0.22201118171215056,
"num_tokens": 18007014.0,
"step": 9195
},
{
"entropy": 5.8996538639068605,
"epoch": 9.145201392342118,
"grad_norm": 1.515625,
"learning_rate": 5.974660905860135e-05,
"loss": 4.9834,
"mean_token_accuracy": 0.21905974745750428,
"num_tokens": 18016088.0,
"step": 9200
},
{
"entropy": 5.676384687423706,
"epoch": 9.150174042764794,
"grad_norm": 1.359375,
"learning_rate": 5.96332362425656e-05,
"loss": 4.7865,
"mean_token_accuracy": 0.24161888062953948,
"num_tokens": 18026824.0,
"step": 9205
},
{
"entropy": 5.831416940689087,
"epoch": 9.155146693187469,
"grad_norm": 1.3125,
"learning_rate": 5.952051224195852e-05,
"loss": 4.8637,
"mean_token_accuracy": 0.22993663251399993,
"num_tokens": 18036391.0,
"step": 9210
},
{
"entropy": 5.8042909622192385,
"epoch": 9.160119343610145,
"grad_norm": 1.4453125,
"learning_rate": 5.940843739637321e-05,
"loss": 4.9708,
"mean_token_accuracy": 0.23520776331424714,
"num_tokens": 18047151.0,
"step": 9215
},
{
"entropy": 5.9461021900177,
"epoch": 9.165091994032819,
"grad_norm": 1.5859375,
"learning_rate": 5.929701204344712e-05,
"loss": 5.0406,
"mean_token_accuracy": 0.21452815532684327,
"num_tokens": 18057380.0,
"step": 9220
},
{
"entropy": 5.851350498199463,
"epoch": 9.170064644455495,
"grad_norm": 1.4375,
"learning_rate": 5.91862365188611e-05,
"loss": 4.9614,
"mean_token_accuracy": 0.2247861847281456,
"num_tokens": 18067324.0,
"step": 9225
},
{
"entropy": 5.931945085525513,
"epoch": 9.17503729487817,
"grad_norm": 1.375,
"learning_rate": 5.907611115633819e-05,
"loss": 4.9809,
"mean_token_accuracy": 0.21508294641971587,
"num_tokens": 18076175.0,
"step": 9230
},
{
"entropy": 5.923648309707642,
"epoch": 9.180009945300846,
"grad_norm": 1.2265625,
"learning_rate": 5.896663628764294e-05,
"loss": 5.0555,
"mean_token_accuracy": 0.21103509664535522,
"num_tokens": 18084835.0,
"step": 9235
},
{
"entropy": 5.789913940429687,
"epoch": 9.18498259572352,
"grad_norm": 1.4140625,
"learning_rate": 5.885781224258002e-05,
"loss": 4.8102,
"mean_token_accuracy": 0.24715079516172409,
"num_tokens": 18096581.0,
"step": 9240
},
{
"entropy": 5.923751258850098,
"epoch": 9.189955246146196,
"grad_norm": 1.328125,
"learning_rate": 5.8749639348993537e-05,
"loss": 4.9847,
"mean_token_accuracy": 0.22787583470344544,
"num_tokens": 18107080.0,
"step": 9245
},
{
"entropy": 5.826544189453125,
"epoch": 9.19492789656887,
"grad_norm": 1.4921875,
"learning_rate": 5.864211793276597e-05,
"loss": 4.8968,
"mean_token_accuracy": 0.22543938755989074,
"num_tokens": 18115666.0,
"step": 9250
},
{
"entropy": 5.927982473373413,
"epoch": 9.199900546991547,
"grad_norm": 1.2890625,
"learning_rate": 5.853524831781702e-05,
"loss": 4.9395,
"mean_token_accuracy": 0.22255284935235978,
"num_tokens": 18124446.0,
"step": 9255
},
{
"entropy": 5.904932689666748,
"epoch": 9.204873197414221,
"grad_norm": 1.2890625,
"learning_rate": 5.8429030826102886e-05,
"loss": 4.9723,
"mean_token_accuracy": 0.22779402136802673,
"num_tokens": 18135261.0,
"step": 9260
},
{
"entropy": 5.703709697723388,
"epoch": 9.209845847836897,
"grad_norm": 1.1328125,
"learning_rate": 5.83234657776151e-05,
"loss": 4.8269,
"mean_token_accuracy": 0.24903169572353362,
"num_tokens": 18145637.0,
"step": 9265
},
{
"entropy": 5.849071741104126,
"epoch": 9.214818498259572,
"grad_norm": 1.5546875,
"learning_rate": 5.8218553490379686e-05,
"loss": 4.9187,
"mean_token_accuracy": 0.2241338774561882,
"num_tokens": 18155192.0,
"step": 9270
},
{
"entropy": 5.845364618301391,
"epoch": 9.219791148682248,
"grad_norm": 1.4140625,
"learning_rate": 5.8114294280456184e-05,
"loss": 4.8571,
"mean_token_accuracy": 0.23220541924238206,
"num_tokens": 18164570.0,
"step": 9275
},
{
"entropy": 5.92794828414917,
"epoch": 9.224763799104922,
"grad_norm": 1.3984375,
"learning_rate": 5.801068846193656e-05,
"loss": 5.043,
"mean_token_accuracy": 0.21802624613046645,
"num_tokens": 18173618.0,
"step": 9280
},
{
"entropy": 5.900962495803833,
"epoch": 9.229736449527598,
"grad_norm": 1.515625,
"learning_rate": 5.7907736346944444e-05,
"loss": 5.032,
"mean_token_accuracy": 0.21453747451305388,
"num_tokens": 18183150.0,
"step": 9285
},
{
"entropy": 5.720554447174072,
"epoch": 9.234709099950274,
"grad_norm": 1.3359375,
"learning_rate": 5.780543824563411e-05,
"loss": 4.7515,
"mean_token_accuracy": 0.2476479709148407,
"num_tokens": 18193649.0,
"step": 9290
},
{
"entropy": 5.866586065292358,
"epoch": 9.239681750372949,
"grad_norm": 1.390625,
"learning_rate": 5.7703794466189524e-05,
"loss": 5.0084,
"mean_token_accuracy": 0.22014385908842088,
"num_tokens": 18204221.0,
"step": 9295
},
{
"entropy": 5.813675022125244,
"epoch": 9.244654400795625,
"grad_norm": 1.3828125,
"learning_rate": 5.7602805314823495e-05,
"loss": 4.9152,
"mean_token_accuracy": 0.22587153017520906,
"num_tokens": 18212887.0,
"step": 9300
},
{
"entropy": 5.931987476348877,
"epoch": 9.2496270512183,
"grad_norm": 1.34375,
"learning_rate": 5.7502471095776585e-05,
"loss": 4.9825,
"mean_token_accuracy": 0.2224376082420349,
"num_tokens": 18222156.0,
"step": 9305
},
{
"entropy": 5.90490756034851,
"epoch": 9.254599701640975,
"grad_norm": 1.3515625,
"learning_rate": 5.740279211131645e-05,
"loss": 5.0484,
"mean_token_accuracy": 0.21635660529136658,
"num_tokens": 18232610.0,
"step": 9310
},
{
"entropy": 5.83326473236084,
"epoch": 9.25957235206365,
"grad_norm": 1.3828125,
"learning_rate": 5.730376866173662e-05,
"loss": 4.9187,
"mean_token_accuracy": 0.2298435315489769,
"num_tokens": 18243001.0,
"step": 9315
},
{
"entropy": 5.913116121292115,
"epoch": 9.264545002486326,
"grad_norm": 1.375,
"learning_rate": 5.720540104535587e-05,
"loss": 5.037,
"mean_token_accuracy": 0.21489163637161254,
"num_tokens": 18253169.0,
"step": 9320
},
{
"entropy": 5.960994577407837,
"epoch": 9.269517652909,
"grad_norm": 1.5,
"learning_rate": 5.710768955851721e-05,
"loss": 5.0383,
"mean_token_accuracy": 0.21082973331212998,
"num_tokens": 18261770.0,
"step": 9325
},
{
"entropy": 5.805781984329224,
"epoch": 9.274490303331676,
"grad_norm": 1.40625,
"learning_rate": 5.701063449558688e-05,
"loss": 4.9008,
"mean_token_accuracy": 0.23192086070775986,
"num_tokens": 18271351.0,
"step": 9330
},
{
"entropy": 5.8963587284088135,
"epoch": 9.27946295375435,
"grad_norm": 1.390625,
"learning_rate": 5.691423614895373e-05,
"loss": 4.983,
"mean_token_accuracy": 0.22763297110795974,
"num_tokens": 18280431.0,
"step": 9335
},
{
"entropy": 5.964140129089356,
"epoch": 9.284435604177027,
"grad_norm": 1.390625,
"learning_rate": 5.681849480902806e-05,
"loss": 5.0218,
"mean_token_accuracy": 0.20830930918455123,
"num_tokens": 18289426.0,
"step": 9340
},
{
"entropy": 5.85099515914917,
"epoch": 9.289408254599701,
"grad_norm": 1.4140625,
"learning_rate": 5.672341076424086e-05,
"loss": 4.8537,
"mean_token_accuracy": 0.23277246057987214,
"num_tokens": 18298217.0,
"step": 9345
},
{
"entropy": 5.769119071960449,
"epoch": 9.294380905022377,
"grad_norm": 1.3828125,
"learning_rate": 5.662898430104306e-05,
"loss": 4.8914,
"mean_token_accuracy": 0.2372089520096779,
"num_tokens": 18309328.0,
"step": 9350
},
{
"entropy": 5.866606855392456,
"epoch": 9.299353555445052,
"grad_norm": 1.328125,
"learning_rate": 5.653521570390442e-05,
"loss": 4.9415,
"mean_token_accuracy": 0.2241523563861847,
"num_tokens": 18318531.0,
"step": 9355
},
{
"entropy": 5.897594690322876,
"epoch": 9.304326205867728,
"grad_norm": 1.3984375,
"learning_rate": 5.64421052553129e-05,
"loss": 4.9998,
"mean_token_accuracy": 0.2219439134001732,
"num_tokens": 18327138.0,
"step": 9360
},
{
"entropy": 5.8805290222167965,
"epoch": 9.309298856290402,
"grad_norm": 1.453125,
"learning_rate": 5.6349653235773774e-05,
"loss": 4.9928,
"mean_token_accuracy": 0.2173142120242119,
"num_tokens": 18338204.0,
"step": 9365
},
{
"entropy": 5.840798854827881,
"epoch": 9.314271506713078,
"grad_norm": 1.2578125,
"learning_rate": 5.625785992380854e-05,
"loss": 4.9842,
"mean_token_accuracy": 0.22447621673345566,
"num_tokens": 18348626.0,
"step": 9370
},
{
"entropy": 5.854054164886475,
"epoch": 9.319244157135753,
"grad_norm": 1.359375,
"learning_rate": 5.6166725595954425e-05,
"loss": 4.9467,
"mean_token_accuracy": 0.23154101371765137,
"num_tokens": 18358556.0,
"step": 9375
},
{
"entropy": 5.841868305206299,
"epoch": 9.324216807558429,
"grad_norm": 1.3671875,
"learning_rate": 5.6076250526763326e-05,
"loss": 5.0158,
"mean_token_accuracy": 0.21467988938093185,
"num_tokens": 18370159.0,
"step": 9380
},
{
"entropy": 5.716777181625366,
"epoch": 9.329189457981103,
"grad_norm": 1.3984375,
"learning_rate": 5.598643498880108e-05,
"loss": 4.7667,
"mean_token_accuracy": 0.25553106516599655,
"num_tokens": 18380118.0,
"step": 9385
},
{
"entropy": 5.90034236907959,
"epoch": 9.33416210840378,
"grad_norm": 1.34375,
"learning_rate": 5.589727925264665e-05,
"loss": 5.0283,
"mean_token_accuracy": 0.21789306700229644,
"num_tokens": 18390132.0,
"step": 9390
},
{
"entropy": 5.846376276016235,
"epoch": 9.339134758826454,
"grad_norm": 1.3125,
"learning_rate": 5.5808783586891186e-05,
"loss": 4.8739,
"mean_token_accuracy": 0.2290594682097435,
"num_tokens": 18400967.0,
"step": 9395
},
{
"entropy": 5.957685041427612,
"epoch": 9.34410740924913,
"grad_norm": 1.40625,
"learning_rate": 5.5720948258137345e-05,
"loss": 5.005,
"mean_token_accuracy": 0.21899928152561188,
"num_tokens": 18411070.0,
"step": 9400
},
{
"entropy": 5.90081205368042,
"epoch": 9.349080059671804,
"grad_norm": 1.3984375,
"learning_rate": 5.563377353099849e-05,
"loss": 4.8803,
"mean_token_accuracy": 0.22790887653827668,
"num_tokens": 18419308.0,
"step": 9405
},
{
"entropy": 5.89550929069519,
"epoch": 9.35405271009448,
"grad_norm": 1.265625,
"learning_rate": 5.55472596680978e-05,
"loss": 4.9352,
"mean_token_accuracy": 0.22027648985385895,
"num_tokens": 18428698.0,
"step": 9410
},
{
"entropy": 5.866608142852783,
"epoch": 9.359025360517156,
"grad_norm": 1.3828125,
"learning_rate": 5.546140693006755e-05,
"loss": 4.9842,
"mean_token_accuracy": 0.22457075417041777,
"num_tokens": 18438112.0,
"step": 9415
},
{
"entropy": 5.836190509796142,
"epoch": 9.36399801093983,
"grad_norm": 1.46875,
"learning_rate": 5.537621557554831e-05,
"loss": 4.9043,
"mean_token_accuracy": 0.22429714798927308,
"num_tokens": 18447908.0,
"step": 9420
},
{
"entropy": 5.592738389968872,
"epoch": 9.368970661362507,
"grad_norm": 1.375,
"learning_rate": 5.529168586118814e-05,
"loss": 4.6557,
"mean_token_accuracy": 0.2619888514280319,
"num_tokens": 18458646.0,
"step": 9425
},
{
"entropy": 5.733367586135865,
"epoch": 9.373943311785181,
"grad_norm": 1.5234375,
"learning_rate": 5.5207818041641894e-05,
"loss": 4.784,
"mean_token_accuracy": 0.24106484800577163,
"num_tokens": 18467727.0,
"step": 9430
},
{
"entropy": 5.593022894859314,
"epoch": 9.378915962207858,
"grad_norm": 1.3671875,
"learning_rate": 5.5124612369570304e-05,
"loss": 4.7111,
"mean_token_accuracy": 0.25341602861881257,
"num_tokens": 18478934.0,
"step": 9435
},
{
"entropy": 5.8957281589508055,
"epoch": 9.383888612630532,
"grad_norm": 1.21875,
"learning_rate": 5.504206909563937e-05,
"loss": 4.9534,
"mean_token_accuracy": 0.22576880306005478,
"num_tokens": 18489151.0,
"step": 9440
},
{
"entropy": 5.803335475921631,
"epoch": 9.388861263053208,
"grad_norm": 1.28125,
"learning_rate": 5.496018846851957e-05,
"loss": 4.8326,
"mean_token_accuracy": 0.2350559800863266,
"num_tokens": 18498937.0,
"step": 9445
},
{
"entropy": 5.8502373695373535,
"epoch": 9.393833913475882,
"grad_norm": 1.4140625,
"learning_rate": 5.4878970734885066e-05,
"loss": 4.9183,
"mean_token_accuracy": 0.22620916217565537,
"num_tokens": 18509216.0,
"step": 9450
},
{
"entropy": 5.8371213436126705,
"epoch": 9.398806563898559,
"grad_norm": 1.4140625,
"learning_rate": 5.479841613941291e-05,
"loss": 4.9285,
"mean_token_accuracy": 0.22765480279922484,
"num_tokens": 18519297.0,
"step": 9455
},
{
"entropy": 5.864552783966064,
"epoch": 9.403779214321233,
"grad_norm": 1.484375,
"learning_rate": 5.471852492478251e-05,
"loss": 4.9017,
"mean_token_accuracy": 0.22294242382049562,
"num_tokens": 18528150.0,
"step": 9460
},
{
"entropy": 5.833744144439697,
"epoch": 9.408751864743909,
"grad_norm": 1.328125,
"learning_rate": 5.4639297331674625e-05,
"loss": 4.9292,
"mean_token_accuracy": 0.23282832652330399,
"num_tokens": 18538246.0,
"step": 9465
},
{
"entropy": 5.826135540008545,
"epoch": 9.413724515166583,
"grad_norm": 1.234375,
"learning_rate": 5.4560733598770956e-05,
"loss": 5.0019,
"mean_token_accuracy": 0.22367148399353026,
"num_tokens": 18547397.0,
"step": 9470
},
{
"entropy": 5.659270811080932,
"epoch": 9.41869716558926,
"grad_norm": 1.328125,
"learning_rate": 5.4482833962753094e-05,
"loss": 4.7408,
"mean_token_accuracy": 0.2522556260228157,
"num_tokens": 18557117.0,
"step": 9475
},
{
"entropy": 5.836642551422119,
"epoch": 9.423669816011934,
"grad_norm": 1.4296875,
"learning_rate": 5.440559865830207e-05,
"loss": 4.8657,
"mean_token_accuracy": 0.23434685170650482,
"num_tokens": 18567147.0,
"step": 9480
},
{
"entropy": 5.911999750137329,
"epoch": 9.42864246643461,
"grad_norm": 1.453125,
"learning_rate": 5.432902791809747e-05,
"loss": 5.0227,
"mean_token_accuracy": 0.21551745086908342,
"num_tokens": 18576156.0,
"step": 9485
},
{
"entropy": 5.886461925506592,
"epoch": 9.433615116857284,
"grad_norm": 1.2421875,
"learning_rate": 5.425312197281686e-05,
"loss": 4.9789,
"mean_token_accuracy": 0.22305763363838196,
"num_tokens": 18585719.0,
"step": 9490
},
{
"entropy": 5.911331224441528,
"epoch": 9.43858776727996,
"grad_norm": 1.4609375,
"learning_rate": 5.417788105113501e-05,
"loss": 4.9998,
"mean_token_accuracy": 0.21719483584165572,
"num_tokens": 18595506.0,
"step": 9495
},
{
"entropy": 5.888077974319458,
"epoch": 9.443560417702635,
"grad_norm": 1.40625,
"learning_rate": 5.4103305379723265e-05,
"loss": 4.9982,
"mean_token_accuracy": 0.2207440808415413,
"num_tokens": 18606124.0,
"step": 9500
},
{
"epoch": 9.443560417702635,
"eval_entropy": 5.847436030339535,
"eval_loss": 6.470646381378174,
"eval_mean_token_accuracy": 0.14965946644475933,
"eval_num_tokens": 18606124.0,
"eval_runtime": 2.5443,
"eval_samples_per_second": 1361.852,
"eval_steps_per_second": 170.575,
"step": 9500
}
],
"logging_steps": 5,
"max_steps": 10050,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.05261433504e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink