fpadovani/nld-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
nld-latn-100mb-after-ppt-sh…/checkpoint-5000/trainer_state.json
ModelHub XC 00e53abac2 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/nld-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407
Source: Original Platform
2026-08-12 22:23:32 +08:00

10145 lines
276 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.54157201272149,
"eval_steps": 500,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.7617779731750485,
"epoch": 0.004543389368468878,
"grad_norm": 12.625,
"learning_rate": 2e-06,
"loss": 14.4213,
"mean_token_accuracy": 0.00039867511950433256,
"num_tokens": 10156.0,
"step": 5
},
{
"entropy": 4.762024164199829,
"epoch": 0.009086778736937756,
"grad_norm": 14.375,
"learning_rate": 4.5e-06,
"loss": 14.3898,
"mean_token_accuracy": 8.278145687654615e-05,
"num_tokens": 20669.0,
"step": 10
},
{
"entropy": 4.801386165618896,
"epoch": 0.013630168105406633,
"grad_norm": 16.875,
"learning_rate": 7e-06,
"loss": 14.1381,
"mean_token_accuracy": 0.00035612244391813874,
"num_tokens": 30952.0,
"step": 15
},
{
"entropy": 5.016745567321777,
"epoch": 0.01817355747387551,
"grad_norm": 28.125,
"learning_rate": 9.5e-06,
"loss": 13.5526,
"mean_token_accuracy": 0.0001651436381507665,
"num_tokens": 41429.0,
"step": 20
},
{
"entropy": 7.075106763839722,
"epoch": 0.02271694684234439,
"grad_norm": 19.75,
"learning_rate": 1.2e-05,
"loss": 11.9297,
"mean_token_accuracy": 0.00024400593247264624,
"num_tokens": 52166.0,
"step": 25
},
{
"entropy": 10.367570781707764,
"epoch": 0.027260336210813266,
"grad_norm": 3.546875,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.747,
"mean_token_accuracy": 0.004502167762257159,
"num_tokens": 61670.0,
"step": 30
},
{
"entropy": 10.692459964752198,
"epoch": 0.03180372557928214,
"grad_norm": 3.0625,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.4764,
"mean_token_accuracy": 0.01658898275345564,
"num_tokens": 71702.0,
"step": 35
},
{
"entropy": 10.684140300750732,
"epoch": 0.03634711494775102,
"grad_norm": 2.8125,
"learning_rate": 1.95e-05,
"loss": 10.1391,
"mean_token_accuracy": 0.01516802804544568,
"num_tokens": 82840.0,
"step": 40
},
{
"entropy": 10.62061595916748,
"epoch": 0.0408905043162199,
"grad_norm": 2.578125,
"learning_rate": 2.2e-05,
"loss": 9.8367,
"mean_token_accuracy": 0.03998391982167959,
"num_tokens": 92290.0,
"step": 45
},
{
"entropy": 10.570560836791993,
"epoch": 0.04543389368468878,
"grad_norm": 2.265625,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.6874,
"mean_token_accuracy": 0.04625674188137054,
"num_tokens": 101842.0,
"step": 50
},
{
"entropy": 10.557878017425537,
"epoch": 0.049977283053157656,
"grad_norm": 1.921875,
"learning_rate": 2.7e-05,
"loss": 9.6545,
"mean_token_accuracy": 0.043805907294154164,
"num_tokens": 114030.0,
"step": 55
},
{
"entropy": 10.589654445648193,
"epoch": 0.05452067242162653,
"grad_norm": 2.0625,
"learning_rate": 2.95e-05,
"loss": 9.5177,
"mean_token_accuracy": 0.04366073869168759,
"num_tokens": 124349.0,
"step": 60
},
{
"entropy": 10.59289197921753,
"epoch": 0.059064061790095414,
"grad_norm": 2.078125,
"learning_rate": 3.2e-05,
"loss": 9.3936,
"mean_token_accuracy": 0.044536291435360906,
"num_tokens": 133534.0,
"step": 65
},
{
"entropy": 10.57084617614746,
"epoch": 0.06360745115856428,
"grad_norm": 2.25,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.3062,
"mean_token_accuracy": 0.041312586329877374,
"num_tokens": 144383.0,
"step": 70
},
{
"entropy": 10.55383882522583,
"epoch": 0.06815084052703317,
"grad_norm": 1.9375,
"learning_rate": 3.7e-05,
"loss": 9.2925,
"mean_token_accuracy": 0.04508417844772339,
"num_tokens": 155467.0,
"step": 75
},
{
"entropy": 10.524444103240967,
"epoch": 0.07269422989550205,
"grad_norm": 2.015625,
"learning_rate": 3.95e-05,
"loss": 9.1809,
"mean_token_accuracy": 0.045108040794730186,
"num_tokens": 165903.0,
"step": 80
},
{
"entropy": 10.497136688232422,
"epoch": 0.07723761926397092,
"grad_norm": 2.203125,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.0687,
"mean_token_accuracy": 0.04266926664859057,
"num_tokens": 175812.0,
"step": 85
},
{
"entropy": 10.480645847320556,
"epoch": 0.0817810086324398,
"grad_norm": 1.984375,
"learning_rate": 4.45e-05,
"loss": 8.9456,
"mean_token_accuracy": 0.0453936118632555,
"num_tokens": 186357.0,
"step": 90
},
{
"entropy": 10.449639892578125,
"epoch": 0.08632439800090867,
"grad_norm": 1.9765625,
"learning_rate": 4.7000000000000004e-05,
"loss": 8.8607,
"mean_token_accuracy": 0.05368485823273659,
"num_tokens": 196124.0,
"step": 95
},
{
"entropy": 10.369837188720703,
"epoch": 0.09086778736937756,
"grad_norm": 1.953125,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.7331,
"mean_token_accuracy": 0.056761689856648444,
"num_tokens": 205655.0,
"step": 100
},
{
"entropy": 10.268289566040039,
"epoch": 0.09541117673784644,
"grad_norm": 1.8203125,
"learning_rate": 5.2e-05,
"loss": 8.6528,
"mean_token_accuracy": 0.05442145057022572,
"num_tokens": 215870.0,
"step": 105
},
{
"entropy": 10.158163261413574,
"epoch": 0.09995456610631531,
"grad_norm": 1.75,
"learning_rate": 5.45e-05,
"loss": 8.522,
"mean_token_accuracy": 0.05480169989168644,
"num_tokens": 225027.0,
"step": 110
},
{
"entropy": 10.10291576385498,
"epoch": 0.10449795547478419,
"grad_norm": 1.6875,
"learning_rate": 5.7e-05,
"loss": 8.411,
"mean_token_accuracy": 0.05560419633984566,
"num_tokens": 236352.0,
"step": 115
},
{
"entropy": 9.994441699981689,
"epoch": 0.10904134484325306,
"grad_norm": 1.6328125,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.2637,
"mean_token_accuracy": 0.05579149127006531,
"num_tokens": 246713.0,
"step": 120
},
{
"entropy": 9.80837516784668,
"epoch": 0.11358473421172194,
"grad_norm": 1.421875,
"learning_rate": 6.2e-05,
"loss": 8.0811,
"mean_token_accuracy": 0.057105731964111325,
"num_tokens": 256950.0,
"step": 125
},
{
"entropy": 9.595962619781494,
"epoch": 0.11812812358019083,
"grad_norm": 1.2109375,
"learning_rate": 6.450000000000001e-05,
"loss": 8.0346,
"mean_token_accuracy": 0.05215160399675369,
"num_tokens": 267345.0,
"step": 130
},
{
"entropy": 9.378811168670655,
"epoch": 0.1226715129486597,
"grad_norm": 1.5859375,
"learning_rate": 6.7e-05,
"loss": 7.8305,
"mean_token_accuracy": 0.059673656895756724,
"num_tokens": 276782.0,
"step": 135
},
{
"entropy": 9.09063777923584,
"epoch": 0.12721490231712856,
"grad_norm": 1.0390625,
"learning_rate": 6.950000000000001e-05,
"loss": 7.7044,
"mean_token_accuracy": 0.06308954879641533,
"num_tokens": 287024.0,
"step": 140
},
{
"entropy": 8.829156970977783,
"epoch": 0.13175829168559747,
"grad_norm": 1.1015625,
"learning_rate": 7.2e-05,
"loss": 7.7479,
"mean_token_accuracy": 0.06074633449316025,
"num_tokens": 296597.0,
"step": 145
},
{
"entropy": 8.660122108459472,
"epoch": 0.13630168105406634,
"grad_norm": 1.0859375,
"learning_rate": 7.45e-05,
"loss": 7.5571,
"mean_token_accuracy": 0.06737130098044872,
"num_tokens": 307709.0,
"step": 150
},
{
"entropy": 8.354702949523926,
"epoch": 0.14084507042253522,
"grad_norm": 1.1171875,
"learning_rate": 7.7e-05,
"loss": 7.5917,
"mean_token_accuracy": 0.07087288200855255,
"num_tokens": 317976.0,
"step": 155
},
{
"entropy": 8.30783452987671,
"epoch": 0.1453884597910041,
"grad_norm": 1.4765625,
"learning_rate": 7.950000000000001e-05,
"loss": 7.462,
"mean_token_accuracy": 0.0719795323908329,
"num_tokens": 327319.0,
"step": 160
},
{
"entropy": 8.110205173492432,
"epoch": 0.14993184915947297,
"grad_norm": 2.015625,
"learning_rate": 8.2e-05,
"loss": 7.4455,
"mean_token_accuracy": 0.07500142268836499,
"num_tokens": 337337.0,
"step": 165
},
{
"entropy": 7.971628332138062,
"epoch": 0.15447523852794184,
"grad_norm": 1.296875,
"learning_rate": 8.450000000000001e-05,
"loss": 7.4838,
"mean_token_accuracy": 0.06365859620273114,
"num_tokens": 347779.0,
"step": 170
},
{
"entropy": 7.962876701354981,
"epoch": 0.15901862789641072,
"grad_norm": 0.93359375,
"learning_rate": 8.7e-05,
"loss": 7.5019,
"mean_token_accuracy": 0.0738484513014555,
"num_tokens": 358088.0,
"step": 175
},
{
"entropy": 7.972894763946533,
"epoch": 0.1635620172648796,
"grad_norm": 1.1171875,
"learning_rate": 8.95e-05,
"loss": 7.4157,
"mean_token_accuracy": 0.08021429628133774,
"num_tokens": 368314.0,
"step": 180
},
{
"entropy": 7.827826261520386,
"epoch": 0.16810540663334847,
"grad_norm": 0.921875,
"learning_rate": 9.2e-05,
"loss": 7.3112,
"mean_token_accuracy": 0.08391624316573143,
"num_tokens": 378406.0,
"step": 185
},
{
"entropy": 7.8609589576721195,
"epoch": 0.17264879600181735,
"grad_norm": 1.1640625,
"learning_rate": 9.45e-05,
"loss": 7.3269,
"mean_token_accuracy": 0.08003454320132733,
"num_tokens": 388912.0,
"step": 190
},
{
"entropy": 7.787754869461059,
"epoch": 0.17719218537028622,
"grad_norm": 1.15625,
"learning_rate": 9.7e-05,
"loss": 7.3277,
"mean_token_accuracy": 0.08325541540980338,
"num_tokens": 399290.0,
"step": 195
},
{
"entropy": 7.774484252929687,
"epoch": 0.18173557473875512,
"grad_norm": 0.89453125,
"learning_rate": 9.95e-05,
"loss": 7.242,
"mean_token_accuracy": 0.08362203016877175,
"num_tokens": 409757.0,
"step": 200
},
{
"entropy": 7.7403558731079105,
"epoch": 0.186278964107224,
"grad_norm": 1.0625,
"learning_rate": 0.000102,
"loss": 7.4135,
"mean_token_accuracy": 0.08218215703964234,
"num_tokens": 420182.0,
"step": 205
},
{
"entropy": 7.780586194992066,
"epoch": 0.19082235347569287,
"grad_norm": 1.046875,
"learning_rate": 0.00010449999999999999,
"loss": 7.2991,
"mean_token_accuracy": 0.07898701690137386,
"num_tokens": 429897.0,
"step": 210
},
{
"entropy": 7.728657341003418,
"epoch": 0.19536574284416175,
"grad_norm": 1.171875,
"learning_rate": 0.000107,
"loss": 7.1992,
"mean_token_accuracy": 0.08197759091854095,
"num_tokens": 441131.0,
"step": 215
},
{
"entropy": 7.609148836135864,
"epoch": 0.19990913221263062,
"grad_norm": 0.859375,
"learning_rate": 0.0001095,
"loss": 7.1791,
"mean_token_accuracy": 0.08322597444057464,
"num_tokens": 451894.0,
"step": 220
},
{
"entropy": 7.640337753295898,
"epoch": 0.2044525215810995,
"grad_norm": 1.03125,
"learning_rate": 0.000112,
"loss": 7.1004,
"mean_token_accuracy": 0.08846881911158562,
"num_tokens": 462613.0,
"step": 225
},
{
"entropy": 7.625602769851684,
"epoch": 0.20899591094956838,
"grad_norm": 1.2734375,
"learning_rate": 0.0001145,
"loss": 7.2065,
"mean_token_accuracy": 0.08791121542453766,
"num_tokens": 471606.0,
"step": 230
},
{
"entropy": 7.539081668853759,
"epoch": 0.21353930031803725,
"grad_norm": 1.265625,
"learning_rate": 0.00011700000000000001,
"loss": 7.1044,
"mean_token_accuracy": 0.09239566922187806,
"num_tokens": 481559.0,
"step": 235
},
{
"entropy": 7.6146509647369385,
"epoch": 0.21808268968650613,
"grad_norm": 1.2890625,
"learning_rate": 0.00011949999999999999,
"loss": 7.123,
"mean_token_accuracy": 0.08494666814804078,
"num_tokens": 492766.0,
"step": 240
},
{
"entropy": 7.549022054672241,
"epoch": 0.222626079054975,
"grad_norm": 0.9765625,
"learning_rate": 0.000122,
"loss": 7.1902,
"mean_token_accuracy": 0.08542409986257553,
"num_tokens": 503357.0,
"step": 245
},
{
"entropy": 7.545317697525024,
"epoch": 0.22716946842344388,
"grad_norm": 1.7109375,
"learning_rate": 0.0001245,
"loss": 7.0221,
"mean_token_accuracy": 0.0891069382429123,
"num_tokens": 513098.0,
"step": 250
},
{
"entropy": 7.549406671524048,
"epoch": 0.23171285779191278,
"grad_norm": 1.1171875,
"learning_rate": 0.000127,
"loss": 7.1016,
"mean_token_accuracy": 0.08548958823084832,
"num_tokens": 522807.0,
"step": 255
},
{
"entropy": 7.560564756393433,
"epoch": 0.23625624716038165,
"grad_norm": 1.1875,
"learning_rate": 0.0001295,
"loss": 7.1193,
"mean_token_accuracy": 0.08555800542235374,
"num_tokens": 532670.0,
"step": 260
},
{
"entropy": 7.523266553878784,
"epoch": 0.24079963652885053,
"grad_norm": 1.09375,
"learning_rate": 0.000132,
"loss": 7.0547,
"mean_token_accuracy": 0.08636701181530952,
"num_tokens": 543647.0,
"step": 265
},
{
"entropy": 7.513870096206665,
"epoch": 0.2453430258973194,
"grad_norm": 1.578125,
"learning_rate": 0.00013450000000000002,
"loss": 7.095,
"mean_token_accuracy": 0.09233981817960739,
"num_tokens": 553047.0,
"step": 270
},
{
"entropy": 7.446543216705322,
"epoch": 0.24988641526578828,
"grad_norm": 1.015625,
"learning_rate": 0.00013700000000000002,
"loss": 7.1552,
"mean_token_accuracy": 0.09209886938333511,
"num_tokens": 563413.0,
"step": 275
},
{
"entropy": 7.474071645736695,
"epoch": 0.25442980463425713,
"grad_norm": 1.1015625,
"learning_rate": 0.0001395,
"loss": 7.1034,
"mean_token_accuracy": 0.08880794793367386,
"num_tokens": 573790.0,
"step": 280
},
{
"entropy": 7.525774335861206,
"epoch": 0.25897319400272606,
"grad_norm": 1.2578125,
"learning_rate": 0.00014199999999999998,
"loss": 7.0109,
"mean_token_accuracy": 0.08925304189324379,
"num_tokens": 582958.0,
"step": 285
},
{
"entropy": 7.388642263412476,
"epoch": 0.26351658337119493,
"grad_norm": 1.46875,
"learning_rate": 0.0001445,
"loss": 6.9668,
"mean_token_accuracy": 0.0980121873319149,
"num_tokens": 592886.0,
"step": 290
},
{
"entropy": 7.4521583080291744,
"epoch": 0.2680599727396638,
"grad_norm": 1.140625,
"learning_rate": 0.000147,
"loss": 7.013,
"mean_token_accuracy": 0.09019459635019303,
"num_tokens": 603692.0,
"step": 295
},
{
"entropy": 7.435794734954834,
"epoch": 0.2726033621081327,
"grad_norm": 1.0390625,
"learning_rate": 0.0001495,
"loss": 7.0484,
"mean_token_accuracy": 0.09026054739952087,
"num_tokens": 613534.0,
"step": 300
},
{
"entropy": 7.424719905853271,
"epoch": 0.27714675147660156,
"grad_norm": 1.421875,
"learning_rate": 0.000152,
"loss": 6.9995,
"mean_token_accuracy": 0.09151231348514557,
"num_tokens": 623862.0,
"step": 305
},
{
"entropy": 7.514718866348266,
"epoch": 0.28169014084507044,
"grad_norm": 1.265625,
"learning_rate": 0.00015450000000000001,
"loss": 7.0863,
"mean_token_accuracy": 0.08810076713562012,
"num_tokens": 633759.0,
"step": 310
},
{
"entropy": 7.361165714263916,
"epoch": 0.2862335302135393,
"grad_norm": 1.234375,
"learning_rate": 0.000157,
"loss": 6.9699,
"mean_token_accuracy": 0.09024499729275703,
"num_tokens": 643553.0,
"step": 315
},
{
"entropy": 7.373637962341308,
"epoch": 0.2907769195820082,
"grad_norm": 1.1484375,
"learning_rate": 0.0001595,
"loss": 6.8834,
"mean_token_accuracy": 0.09637065082788468,
"num_tokens": 653714.0,
"step": 320
},
{
"entropy": 7.3554847717285154,
"epoch": 0.29532030895047706,
"grad_norm": 1.2109375,
"learning_rate": 0.000162,
"loss": 6.9076,
"mean_token_accuracy": 0.09380478709936142,
"num_tokens": 662817.0,
"step": 325
},
{
"entropy": 7.294995498657227,
"epoch": 0.29986369831894594,
"grad_norm": 1.390625,
"learning_rate": 0.00016450000000000001,
"loss": 6.9342,
"mean_token_accuracy": 0.0921212948858738,
"num_tokens": 672381.0,
"step": 330
},
{
"entropy": 7.2491357803344725,
"epoch": 0.3044070876874148,
"grad_norm": 1.375,
"learning_rate": 0.00016700000000000002,
"loss": 6.8159,
"mean_token_accuracy": 0.09338896572589875,
"num_tokens": 682525.0,
"step": 335
},
{
"entropy": 7.309194135665893,
"epoch": 0.3089504770558837,
"grad_norm": 1.25,
"learning_rate": 0.00016950000000000003,
"loss": 6.91,
"mean_token_accuracy": 0.09586632996797562,
"num_tokens": 691826.0,
"step": 340
},
{
"entropy": 7.192427301406861,
"epoch": 0.31349386642435256,
"grad_norm": 1.140625,
"learning_rate": 0.00017199999999999998,
"loss": 6.8849,
"mean_token_accuracy": 0.10212687999010087,
"num_tokens": 701319.0,
"step": 345
},
{
"entropy": 7.242300987243652,
"epoch": 0.31803725579282144,
"grad_norm": 1.296875,
"learning_rate": 0.00017449999999999999,
"loss": 6.9811,
"mean_token_accuracy": 0.0917223945260048,
"num_tokens": 710888.0,
"step": 350
},
{
"entropy": 7.351683664321899,
"epoch": 0.3225806451612903,
"grad_norm": 1.0,
"learning_rate": 0.000177,
"loss": 6.9156,
"mean_token_accuracy": 0.09456708803772926,
"num_tokens": 722516.0,
"step": 355
},
{
"entropy": 7.288315916061402,
"epoch": 0.3271240345297592,
"grad_norm": 1.671875,
"learning_rate": 0.0001795,
"loss": 6.8905,
"mean_token_accuracy": 0.09565110653638839,
"num_tokens": 733058.0,
"step": 360
},
{
"entropy": 7.249158096313477,
"epoch": 0.33166742389822806,
"grad_norm": 1.2265625,
"learning_rate": 0.000182,
"loss": 6.959,
"mean_token_accuracy": 0.09808666259050369,
"num_tokens": 744832.0,
"step": 365
},
{
"entropy": 7.257345390319824,
"epoch": 0.33621081326669694,
"grad_norm": 1.296875,
"learning_rate": 0.0001845,
"loss": 6.8852,
"mean_token_accuracy": 0.09970412328839302,
"num_tokens": 755387.0,
"step": 370
},
{
"entropy": 7.181437587738037,
"epoch": 0.3407542026351658,
"grad_norm": 1.3671875,
"learning_rate": 0.000187,
"loss": 6.8328,
"mean_token_accuracy": 0.0964855819940567,
"num_tokens": 764455.0,
"step": 375
},
{
"entropy": 7.1537618160247805,
"epoch": 0.3452975920036347,
"grad_norm": 1.25,
"learning_rate": 0.0001895,
"loss": 6.8678,
"mean_token_accuracy": 0.09778558984398841,
"num_tokens": 774513.0,
"step": 380
},
{
"entropy": 7.281053781509399,
"epoch": 0.34984098137210357,
"grad_norm": 0.96484375,
"learning_rate": 0.000192,
"loss": 6.9626,
"mean_token_accuracy": 0.09827063530683518,
"num_tokens": 785540.0,
"step": 385
},
{
"entropy": 7.215037727355957,
"epoch": 0.35438437074057244,
"grad_norm": 1.15625,
"learning_rate": 0.0001945,
"loss": 6.8204,
"mean_token_accuracy": 0.09672441035509109,
"num_tokens": 795233.0,
"step": 390
},
{
"entropy": 7.275153923034668,
"epoch": 0.35892776010904137,
"grad_norm": 1.1875,
"learning_rate": 0.00019700000000000002,
"loss": 6.8705,
"mean_token_accuracy": 0.09623145386576652,
"num_tokens": 804855.0,
"step": 395
},
{
"entropy": 7.20860447883606,
"epoch": 0.36347114947751025,
"grad_norm": 1.1171875,
"learning_rate": 0.00019950000000000002,
"loss": 6.8266,
"mean_token_accuracy": 0.10111764296889306,
"num_tokens": 814818.0,
"step": 400
},
{
"entropy": 7.1098264217376705,
"epoch": 0.3680145388459791,
"grad_norm": 1.1328125,
"learning_rate": 0.000202,
"loss": 6.8627,
"mean_token_accuracy": 0.10245592221617698,
"num_tokens": 824622.0,
"step": 405
},
{
"entropy": 7.1824994564056395,
"epoch": 0.372557928214448,
"grad_norm": 1.3046875,
"learning_rate": 0.00020449999999999998,
"loss": 6.7427,
"mean_token_accuracy": 0.10741053223609924,
"num_tokens": 834813.0,
"step": 410
},
{
"entropy": 7.061726331710815,
"epoch": 0.3771013175829169,
"grad_norm": 1.125,
"learning_rate": 0.000207,
"loss": 6.8567,
"mean_token_accuracy": 0.10019884258508682,
"num_tokens": 844887.0,
"step": 415
},
{
"entropy": 7.173247957229615,
"epoch": 0.38164470695138575,
"grad_norm": 1.1640625,
"learning_rate": 0.0002095,
"loss": 6.8447,
"mean_token_accuracy": 0.10034631565213203,
"num_tokens": 855430.0,
"step": 420
},
{
"entropy": 7.076598262786865,
"epoch": 0.3861880963198546,
"grad_norm": 1.234375,
"learning_rate": 0.000212,
"loss": 6.6801,
"mean_token_accuracy": 0.10056375861167907,
"num_tokens": 866052.0,
"step": 425
},
{
"entropy": 7.042910146713257,
"epoch": 0.3907314856883235,
"grad_norm": 1.1015625,
"learning_rate": 0.0002145,
"loss": 6.8356,
"mean_token_accuracy": 0.09405270889401436,
"num_tokens": 877961.0,
"step": 430
},
{
"entropy": 7.111201524734497,
"epoch": 0.3952748750567924,
"grad_norm": 1.1796875,
"learning_rate": 0.00021700000000000002,
"loss": 6.8019,
"mean_token_accuracy": 0.10706538185477257,
"num_tokens": 888386.0,
"step": 435
},
{
"entropy": 7.103235197067261,
"epoch": 0.39981826442526125,
"grad_norm": 1.2265625,
"learning_rate": 0.0002195,
"loss": 6.8377,
"mean_token_accuracy": 0.10170447900891304,
"num_tokens": 898272.0,
"step": 440
},
{
"entropy": 7.130263900756836,
"epoch": 0.4043616537937301,
"grad_norm": 1.3515625,
"learning_rate": 0.000222,
"loss": 6.7814,
"mean_token_accuracy": 0.1046774573624134,
"num_tokens": 908317.0,
"step": 445
},
{
"entropy": 7.089518404006958,
"epoch": 0.408905043162199,
"grad_norm": 1.421875,
"learning_rate": 0.0002245,
"loss": 6.7756,
"mean_token_accuracy": 0.09967579692602158,
"num_tokens": 917970.0,
"step": 450
},
{
"entropy": 7.043964099884033,
"epoch": 0.4134484325306679,
"grad_norm": 1.2890625,
"learning_rate": 0.00022700000000000002,
"loss": 6.8119,
"mean_token_accuracy": 0.0967039294540882,
"num_tokens": 928132.0,
"step": 455
},
{
"entropy": 7.043747615814209,
"epoch": 0.41799182189913675,
"grad_norm": 1.3984375,
"learning_rate": 0.00022950000000000002,
"loss": 6.7437,
"mean_token_accuracy": 0.10632947832345963,
"num_tokens": 938092.0,
"step": 460
},
{
"entropy": 7.126165199279785,
"epoch": 0.4225352112676056,
"grad_norm": 1.0078125,
"learning_rate": 0.00023200000000000003,
"loss": 6.8483,
"mean_token_accuracy": 0.09798347502946854,
"num_tokens": 948650.0,
"step": 465
},
{
"entropy": 6.976731872558593,
"epoch": 0.4270786006360745,
"grad_norm": 1.109375,
"learning_rate": 0.00023449999999999998,
"loss": 6.77,
"mean_token_accuracy": 0.10464980453252792,
"num_tokens": 960195.0,
"step": 470
},
{
"entropy": 7.047196292877198,
"epoch": 0.4316219900045434,
"grad_norm": 1.0625,
"learning_rate": 0.000237,
"loss": 6.7843,
"mean_token_accuracy": 0.10333740636706353,
"num_tokens": 970608.0,
"step": 475
},
{
"entropy": 7.10983419418335,
"epoch": 0.43616537937301225,
"grad_norm": 1.1640625,
"learning_rate": 0.0002395,
"loss": 6.7567,
"mean_token_accuracy": 0.10173246711492538,
"num_tokens": 981053.0,
"step": 480
},
{
"entropy": 6.920445919036865,
"epoch": 0.4407087687414811,
"grad_norm": 1.21875,
"learning_rate": 0.000242,
"loss": 6.6903,
"mean_token_accuracy": 0.10761705115437507,
"num_tokens": 991560.0,
"step": 485
},
{
"entropy": 6.985012483596802,
"epoch": 0.44525215810995,
"grad_norm": 1.2890625,
"learning_rate": 0.0002445,
"loss": 6.7318,
"mean_token_accuracy": 0.09935669675469398,
"num_tokens": 1002008.0,
"step": 490
},
{
"entropy": 7.016558361053467,
"epoch": 0.4497955474784189,
"grad_norm": 1.0703125,
"learning_rate": 0.000247,
"loss": 6.7297,
"mean_token_accuracy": 0.11214710623025895,
"num_tokens": 1012349.0,
"step": 495
},
{
"entropy": 6.973590946197509,
"epoch": 0.45433893684688775,
"grad_norm": 1.4765625,
"learning_rate": 0.0002495,
"loss": 6.7202,
"mean_token_accuracy": 0.10605021119117737,
"num_tokens": 1022886.0,
"step": 500
},
{
"epoch": 0.45433893684688775,
"eval_entropy": 6.977851612623348,
"eval_loss": 6.848958492279053,
"eval_mean_token_accuracy": 0.10486862519539372,
"eval_num_tokens": 1022886.0,
"eval_runtime": 2.6621,
"eval_samples_per_second": 1290.327,
"eval_steps_per_second": 161.526,
"step": 500
},
{
"entropy": 7.035666418075562,
"epoch": 0.45888232621535663,
"grad_norm": 1.25,
"learning_rate": 0.000252,
"loss": 6.7832,
"mean_token_accuracy": 0.09916550442576408,
"num_tokens": 1033129.0,
"step": 505
},
{
"entropy": 6.973564672470093,
"epoch": 0.46342571558382556,
"grad_norm": 1.171875,
"learning_rate": 0.0002545,
"loss": 6.6707,
"mean_token_accuracy": 0.10667844489216805,
"num_tokens": 1043069.0,
"step": 510
},
{
"entropy": 6.935989141464233,
"epoch": 0.46796910495229443,
"grad_norm": 1.0703125,
"learning_rate": 0.000257,
"loss": 6.6902,
"mean_token_accuracy": 0.10451378300786018,
"num_tokens": 1054049.0,
"step": 515
},
{
"entropy": 6.865748262405395,
"epoch": 0.4725124943207633,
"grad_norm": 1.0625,
"learning_rate": 0.0002595,
"loss": 6.6812,
"mean_token_accuracy": 0.10345099791884423,
"num_tokens": 1065155.0,
"step": 520
},
{
"entropy": 6.918096256256104,
"epoch": 0.4770558836892322,
"grad_norm": 1.125,
"learning_rate": 0.000262,
"loss": 6.6756,
"mean_token_accuracy": 0.10925311967730522,
"num_tokens": 1075510.0,
"step": 525
},
{
"entropy": 6.956620407104492,
"epoch": 0.48159927305770106,
"grad_norm": 1.125,
"learning_rate": 0.00026450000000000003,
"loss": 6.6872,
"mean_token_accuracy": 0.10796692073345185,
"num_tokens": 1086494.0,
"step": 530
},
{
"entropy": 6.948440408706665,
"epoch": 0.48614266242616994,
"grad_norm": 1.1328125,
"learning_rate": 0.00026700000000000004,
"loss": 6.7199,
"mean_token_accuracy": 0.10674306005239487,
"num_tokens": 1097311.0,
"step": 535
},
{
"entropy": 6.954690170288086,
"epoch": 0.4906860517946388,
"grad_norm": 1.203125,
"learning_rate": 0.00026950000000000005,
"loss": 6.6821,
"mean_token_accuracy": 0.1063051015138626,
"num_tokens": 1107757.0,
"step": 540
},
{
"entropy": 6.827787971496582,
"epoch": 0.4952294411631077,
"grad_norm": 1.1015625,
"learning_rate": 0.00027200000000000005,
"loss": 6.5914,
"mean_token_accuracy": 0.1098194107413292,
"num_tokens": 1118889.0,
"step": 545
},
{
"entropy": 6.969381999969483,
"epoch": 0.49977283053157656,
"grad_norm": 1.09375,
"learning_rate": 0.0002745,
"loss": 6.7809,
"mean_token_accuracy": 0.10160353481769561,
"num_tokens": 1128951.0,
"step": 550
},
{
"entropy": 6.877148771286011,
"epoch": 0.5043162199000454,
"grad_norm": 1.25,
"learning_rate": 0.000277,
"loss": 6.5911,
"mean_token_accuracy": 0.11018141955137253,
"num_tokens": 1138583.0,
"step": 555
},
{
"entropy": 6.908321237564087,
"epoch": 0.5088596092685143,
"grad_norm": 1.078125,
"learning_rate": 0.0002795,
"loss": 6.6507,
"mean_token_accuracy": 0.11024285778403282,
"num_tokens": 1149572.0,
"step": 560
},
{
"entropy": 6.895721769332885,
"epoch": 0.5134029986369832,
"grad_norm": 1.0625,
"learning_rate": 0.00028199999999999997,
"loss": 6.6124,
"mean_token_accuracy": 0.11112551465630531,
"num_tokens": 1160532.0,
"step": 565
},
{
"entropy": 6.8745800971984865,
"epoch": 0.5179463880054521,
"grad_norm": 1.0546875,
"learning_rate": 0.0002845,
"loss": 6.6244,
"mean_token_accuracy": 0.10947038605809212,
"num_tokens": 1170915.0,
"step": 570
},
{
"entropy": 6.816075086593628,
"epoch": 0.522489777373921,
"grad_norm": 1.171875,
"learning_rate": 0.000287,
"loss": 6.5852,
"mean_token_accuracy": 0.11535143703222275,
"num_tokens": 1180822.0,
"step": 575
},
{
"entropy": 6.798626184463501,
"epoch": 0.5270331667423899,
"grad_norm": 1.1484375,
"learning_rate": 0.0002895,
"loss": 6.6272,
"mean_token_accuracy": 0.11104766130447388,
"num_tokens": 1190447.0,
"step": 580
},
{
"entropy": 6.842454719543457,
"epoch": 0.5315765561108587,
"grad_norm": 1.5,
"learning_rate": 0.000292,
"loss": 6.6738,
"mean_token_accuracy": 0.10826868191361427,
"num_tokens": 1200951.0,
"step": 585
},
{
"entropy": 6.8950550079345705,
"epoch": 0.5361199454793276,
"grad_norm": 1.125,
"learning_rate": 0.0002945,
"loss": 6.5506,
"mean_token_accuracy": 0.11410407572984696,
"num_tokens": 1211846.0,
"step": 590
},
{
"entropy": 6.644092798233032,
"epoch": 0.5406633348477965,
"grad_norm": 1.078125,
"learning_rate": 0.000297,
"loss": 6.5204,
"mean_token_accuracy": 0.11587296798825264,
"num_tokens": 1222074.0,
"step": 595
},
{
"entropy": 6.710137367248535,
"epoch": 0.5452067242162654,
"grad_norm": 1.1328125,
"learning_rate": 0.0002995,
"loss": 6.5777,
"mean_token_accuracy": 0.11183500215411186,
"num_tokens": 1231741.0,
"step": 600
},
{
"entropy": 6.791498470306396,
"epoch": 0.5497501135847342,
"grad_norm": 1.1640625,
"learning_rate": 0.000302,
"loss": 6.5151,
"mean_token_accuracy": 0.11139251589775086,
"num_tokens": 1241078.0,
"step": 605
},
{
"entropy": 6.8775381565094,
"epoch": 0.5542935029532031,
"grad_norm": 1.3125,
"learning_rate": 0.0003045,
"loss": 6.7192,
"mean_token_accuracy": 0.10424678772687912,
"num_tokens": 1251685.0,
"step": 610
},
{
"entropy": 6.692072534561158,
"epoch": 0.558836892321672,
"grad_norm": 1.109375,
"learning_rate": 0.000307,
"loss": 6.5708,
"mean_token_accuracy": 0.11387862861156464,
"num_tokens": 1262044.0,
"step": 615
},
{
"entropy": 6.741009140014649,
"epoch": 0.5633802816901409,
"grad_norm": 1.03125,
"learning_rate": 0.0003095,
"loss": 6.5699,
"mean_token_accuracy": 0.11056023836135864,
"num_tokens": 1273647.0,
"step": 620
},
{
"entropy": 6.8509361743927,
"epoch": 0.5679236710586097,
"grad_norm": 1.09375,
"learning_rate": 0.000312,
"loss": 6.6189,
"mean_token_accuracy": 0.10981747508049011,
"num_tokens": 1284396.0,
"step": 625
},
{
"entropy": 6.829682111740112,
"epoch": 0.5724670604270786,
"grad_norm": 1.09375,
"learning_rate": 0.0003145,
"loss": 6.6044,
"mean_token_accuracy": 0.11069794371724129,
"num_tokens": 1294339.0,
"step": 630
},
{
"entropy": 6.768509197235107,
"epoch": 0.5770104497955475,
"grad_norm": 1.3671875,
"learning_rate": 0.000317,
"loss": 6.5277,
"mean_token_accuracy": 0.11431413069367409,
"num_tokens": 1304572.0,
"step": 635
},
{
"entropy": 6.707783174514771,
"epoch": 0.5815538391640164,
"grad_norm": 1.1953125,
"learning_rate": 0.0003195,
"loss": 6.5716,
"mean_token_accuracy": 0.11335601583123207,
"num_tokens": 1314780.0,
"step": 640
},
{
"entropy": 6.685592889785767,
"epoch": 0.5860972285324852,
"grad_norm": 1.109375,
"learning_rate": 0.000322,
"loss": 6.5905,
"mean_token_accuracy": 0.1090255431830883,
"num_tokens": 1325215.0,
"step": 645
},
{
"entropy": 6.724009275436401,
"epoch": 0.5906406179009541,
"grad_norm": 1.109375,
"learning_rate": 0.00032450000000000003,
"loss": 6.4958,
"mean_token_accuracy": 0.1151325061917305,
"num_tokens": 1334745.0,
"step": 650
},
{
"entropy": 6.831643390655517,
"epoch": 0.595184007269423,
"grad_norm": 1.265625,
"learning_rate": 0.00032700000000000003,
"loss": 6.6123,
"mean_token_accuracy": 0.11255677044391632,
"num_tokens": 1345150.0,
"step": 655
},
{
"entropy": 6.7713886260986325,
"epoch": 0.5997273966378919,
"grad_norm": 1.09375,
"learning_rate": 0.00032950000000000004,
"loss": 6.5504,
"mean_token_accuracy": 0.11500600427389145,
"num_tokens": 1355619.0,
"step": 660
},
{
"entropy": 6.660548448562622,
"epoch": 0.6042707860063607,
"grad_norm": 1.1328125,
"learning_rate": 0.00033200000000000005,
"loss": 6.4861,
"mean_token_accuracy": 0.11048821657896042,
"num_tokens": 1365933.0,
"step": 665
},
{
"entropy": 6.692146348953247,
"epoch": 0.6088141753748296,
"grad_norm": 1.0625,
"learning_rate": 0.00033450000000000005,
"loss": 6.5035,
"mean_token_accuracy": 0.11794941723346711,
"num_tokens": 1376867.0,
"step": 670
},
{
"entropy": 6.744796943664551,
"epoch": 0.6133575647432985,
"grad_norm": 1.03125,
"learning_rate": 0.000337,
"loss": 6.6714,
"mean_token_accuracy": 0.11807657107710838,
"num_tokens": 1386739.0,
"step": 675
},
{
"entropy": 6.708650588989258,
"epoch": 0.6179009541117674,
"grad_norm": 1.1484375,
"learning_rate": 0.0003395,
"loss": 6.5072,
"mean_token_accuracy": 0.11620534211397171,
"num_tokens": 1397138.0,
"step": 680
},
{
"entropy": 6.778751182556152,
"epoch": 0.6224443434802363,
"grad_norm": 1.3203125,
"learning_rate": 0.000342,
"loss": 6.6173,
"mean_token_accuracy": 0.11403155997395516,
"num_tokens": 1408171.0,
"step": 685
},
{
"entropy": 6.662551021575927,
"epoch": 0.6269877328487051,
"grad_norm": 1.25,
"learning_rate": 0.00034449999999999997,
"loss": 6.547,
"mean_token_accuracy": 0.10886397436261178,
"num_tokens": 1418259.0,
"step": 690
},
{
"entropy": 6.684373235702514,
"epoch": 0.631531122217174,
"grad_norm": 1.2734375,
"learning_rate": 0.000347,
"loss": 6.5507,
"mean_token_accuracy": 0.11670795381069184,
"num_tokens": 1428080.0,
"step": 695
},
{
"entropy": 6.766232347488403,
"epoch": 0.6360745115856429,
"grad_norm": 1.171875,
"learning_rate": 0.0003495,
"loss": 6.5778,
"mean_token_accuracy": 0.11337161809206009,
"num_tokens": 1438502.0,
"step": 700
},
{
"entropy": 6.587224006652832,
"epoch": 0.6406179009541118,
"grad_norm": 1.1015625,
"learning_rate": 0.000352,
"loss": 6.5274,
"mean_token_accuracy": 0.1150364950299263,
"num_tokens": 1448396.0,
"step": 705
},
{
"entropy": 6.738826274871826,
"epoch": 0.6451612903225806,
"grad_norm": 1.1796875,
"learning_rate": 0.0003545,
"loss": 6.5893,
"mean_token_accuracy": 0.10938060209155083,
"num_tokens": 1459728.0,
"step": 710
},
{
"entropy": 6.68166880607605,
"epoch": 0.6497046796910495,
"grad_norm": 1.1328125,
"learning_rate": 0.000357,
"loss": 6.4802,
"mean_token_accuracy": 0.11845517382025719,
"num_tokens": 1469733.0,
"step": 715
},
{
"entropy": 6.561370134353638,
"epoch": 0.6542480690595184,
"grad_norm": 1.3046875,
"learning_rate": 0.0003595,
"loss": 6.4669,
"mean_token_accuracy": 0.11939780414104462,
"num_tokens": 1479004.0,
"step": 720
},
{
"entropy": 6.6443627834320065,
"epoch": 0.6587914584279873,
"grad_norm": 1.25,
"learning_rate": 0.000362,
"loss": 6.445,
"mean_token_accuracy": 0.11710177734494209,
"num_tokens": 1489848.0,
"step": 725
},
{
"entropy": 6.573301315307617,
"epoch": 0.6633348477964561,
"grad_norm": 1.0,
"learning_rate": 0.0003645,
"loss": 6.4316,
"mean_token_accuracy": 0.11691953688859939,
"num_tokens": 1500877.0,
"step": 730
},
{
"entropy": 6.733279418945313,
"epoch": 0.667878237164925,
"grad_norm": 1.375,
"learning_rate": 0.000367,
"loss": 6.5895,
"mean_token_accuracy": 0.11292536705732345,
"num_tokens": 1511840.0,
"step": 735
},
{
"entropy": 6.639285326004028,
"epoch": 0.6724216265333939,
"grad_norm": 1.171875,
"learning_rate": 0.0003695,
"loss": 6.4862,
"mean_token_accuracy": 0.11606627479195594,
"num_tokens": 1521839.0,
"step": 740
},
{
"entropy": 6.616071891784668,
"epoch": 0.6769650159018628,
"grad_norm": 1.2421875,
"learning_rate": 0.000372,
"loss": 6.3896,
"mean_token_accuracy": 0.11983591318130493,
"num_tokens": 1531285.0,
"step": 745
},
{
"entropy": 6.439044046401977,
"epoch": 0.6815084052703316,
"grad_norm": 1.1640625,
"learning_rate": 0.0003745,
"loss": 6.4368,
"mean_token_accuracy": 0.11259082779288292,
"num_tokens": 1542027.0,
"step": 750
},
{
"entropy": 6.610742664337158,
"epoch": 0.6860517946388005,
"grad_norm": 1.2265625,
"learning_rate": 0.000377,
"loss": 6.5759,
"mean_token_accuracy": 0.11640218272805214,
"num_tokens": 1552527.0,
"step": 755
},
{
"entropy": 6.7587377548217775,
"epoch": 0.6905951840072694,
"grad_norm": 1.1015625,
"learning_rate": 0.0003795,
"loss": 6.5953,
"mean_token_accuracy": 0.11593217179179191,
"num_tokens": 1562605.0,
"step": 760
},
{
"entropy": 6.547891283035279,
"epoch": 0.6951385733757383,
"grad_norm": 1.1640625,
"learning_rate": 0.000382,
"loss": 6.4772,
"mean_token_accuracy": 0.11769971996545792,
"num_tokens": 1572128.0,
"step": 765
},
{
"entropy": 6.586852359771728,
"epoch": 0.6996819627442071,
"grad_norm": 1.3046875,
"learning_rate": 0.0003845,
"loss": 6.4105,
"mean_token_accuracy": 0.11785594075918197,
"num_tokens": 1582352.0,
"step": 770
},
{
"entropy": 6.539891147613526,
"epoch": 0.704225352112676,
"grad_norm": 1.1328125,
"learning_rate": 0.00038700000000000003,
"loss": 6.5139,
"mean_token_accuracy": 0.11761872917413711,
"num_tokens": 1592887.0,
"step": 775
},
{
"entropy": 6.621896171569825,
"epoch": 0.7087687414811449,
"grad_norm": 1.0078125,
"learning_rate": 0.00038950000000000003,
"loss": 6.4471,
"mean_token_accuracy": 0.1175874799489975,
"num_tokens": 1603564.0,
"step": 780
},
{
"entropy": 6.4781081199646,
"epoch": 0.7133121308496138,
"grad_norm": 1.0078125,
"learning_rate": 0.00039200000000000004,
"loss": 6.2976,
"mean_token_accuracy": 0.12441059201955795,
"num_tokens": 1613270.0,
"step": 785
},
{
"entropy": 6.457367992401123,
"epoch": 0.7178555202180827,
"grad_norm": 1.234375,
"learning_rate": 0.00039450000000000005,
"loss": 6.3618,
"mean_token_accuracy": 0.1260719694197178,
"num_tokens": 1623009.0,
"step": 790
},
{
"entropy": 6.54601092338562,
"epoch": 0.7223989095865516,
"grad_norm": 1.2109375,
"learning_rate": 0.00039700000000000005,
"loss": 6.415,
"mean_token_accuracy": 0.11839432567358017,
"num_tokens": 1633209.0,
"step": 795
},
{
"entropy": 6.550046920776367,
"epoch": 0.7269422989550205,
"grad_norm": 1.25,
"learning_rate": 0.0003995,
"loss": 6.4463,
"mean_token_accuracy": 0.12334003522992135,
"num_tokens": 1643708.0,
"step": 800
},
{
"entropy": 6.514151525497437,
"epoch": 0.7314856883234894,
"grad_norm": 1.0859375,
"learning_rate": 0.000402,
"loss": 6.5074,
"mean_token_accuracy": 0.11595956683158874,
"num_tokens": 1653973.0,
"step": 805
},
{
"entropy": 6.569599771499634,
"epoch": 0.7360290776919582,
"grad_norm": 1.0625,
"learning_rate": 0.0004045,
"loss": 6.4918,
"mean_token_accuracy": 0.11730098500847816,
"num_tokens": 1664856.0,
"step": 810
},
{
"entropy": 6.5198814868927,
"epoch": 0.7405724670604271,
"grad_norm": 1.09375,
"learning_rate": 0.00040699999999999997,
"loss": 6.4048,
"mean_token_accuracy": 0.11975548416376114,
"num_tokens": 1675935.0,
"step": 815
},
{
"entropy": 6.561031484603882,
"epoch": 0.745115856428896,
"grad_norm": 1.1484375,
"learning_rate": 0.0004095,
"loss": 6.4505,
"mean_token_accuracy": 0.11787926480174064,
"num_tokens": 1686597.0,
"step": 820
},
{
"entropy": 6.465208625793457,
"epoch": 0.7496592457973649,
"grad_norm": 1.0234375,
"learning_rate": 0.000412,
"loss": 6.3971,
"mean_token_accuracy": 0.1232870027422905,
"num_tokens": 1697385.0,
"step": 825
},
{
"entropy": 6.63141975402832,
"epoch": 0.7542026351658337,
"grad_norm": 1.0703125,
"learning_rate": 0.0004145,
"loss": 6.4407,
"mean_token_accuracy": 0.11851782277226448,
"num_tokens": 1706680.0,
"step": 830
},
{
"entropy": 6.5177350521087645,
"epoch": 0.7587460245343026,
"grad_norm": 1.109375,
"learning_rate": 0.000417,
"loss": 6.4445,
"mean_token_accuracy": 0.12521564438939095,
"num_tokens": 1716130.0,
"step": 835
},
{
"entropy": 6.452193069458008,
"epoch": 0.7632894139027715,
"grad_norm": 0.99609375,
"learning_rate": 0.0004195,
"loss": 6.3712,
"mean_token_accuracy": 0.11505284011363984,
"num_tokens": 1726835.0,
"step": 840
},
{
"entropy": 6.523750114440918,
"epoch": 0.7678328032712404,
"grad_norm": 1.125,
"learning_rate": 0.000422,
"loss": 6.4201,
"mean_token_accuracy": 0.12676911801099777,
"num_tokens": 1737708.0,
"step": 845
},
{
"entropy": 6.52707142829895,
"epoch": 0.7723761926397092,
"grad_norm": 1.0703125,
"learning_rate": 0.0004245,
"loss": 6.4002,
"mean_token_accuracy": 0.12539136186242103,
"num_tokens": 1747649.0,
"step": 850
},
{
"entropy": 6.448366975784301,
"epoch": 0.7769195820081781,
"grad_norm": 1.1171875,
"learning_rate": 0.000427,
"loss": 6.425,
"mean_token_accuracy": 0.1189683198928833,
"num_tokens": 1758327.0,
"step": 855
},
{
"entropy": 6.485600423812866,
"epoch": 0.781462971376647,
"grad_norm": 1.0234375,
"learning_rate": 0.0004295,
"loss": 6.3848,
"mean_token_accuracy": 0.11789172142744064,
"num_tokens": 1769767.0,
"step": 860
},
{
"entropy": 6.5145751953125,
"epoch": 0.7860063607451159,
"grad_norm": 1.1015625,
"learning_rate": 0.000432,
"loss": 6.472,
"mean_token_accuracy": 0.11863588467240334,
"num_tokens": 1780139.0,
"step": 865
},
{
"entropy": 6.546294212341309,
"epoch": 0.7905497501135847,
"grad_norm": 1.0234375,
"learning_rate": 0.0004345,
"loss": 6.3655,
"mean_token_accuracy": 0.12412618100643158,
"num_tokens": 1790783.0,
"step": 870
},
{
"entropy": 6.424306201934814,
"epoch": 0.7950931394820536,
"grad_norm": 1.109375,
"learning_rate": 0.000437,
"loss": 6.3776,
"mean_token_accuracy": 0.11976175010204315,
"num_tokens": 1801349.0,
"step": 875
},
{
"entropy": 6.453584623336792,
"epoch": 0.7996365288505225,
"grad_norm": 1.0625,
"learning_rate": 0.0004395,
"loss": 6.3463,
"mean_token_accuracy": 0.12058770209550858,
"num_tokens": 1811990.0,
"step": 880
},
{
"entropy": 6.394813871383667,
"epoch": 0.8041799182189914,
"grad_norm": 1.1484375,
"learning_rate": 0.000442,
"loss": 6.3735,
"mean_token_accuracy": 0.12705680951476098,
"num_tokens": 1821725.0,
"step": 885
},
{
"entropy": 6.454415798187256,
"epoch": 0.8087233075874602,
"grad_norm": 1.0703125,
"learning_rate": 0.0004445,
"loss": 6.3369,
"mean_token_accuracy": 0.11955134496092797,
"num_tokens": 1831794.0,
"step": 890
},
{
"entropy": 6.438329124450684,
"epoch": 0.8132666969559291,
"grad_norm": 1.0546875,
"learning_rate": 0.000447,
"loss": 6.4041,
"mean_token_accuracy": 0.1207861989736557,
"num_tokens": 1842002.0,
"step": 895
},
{
"entropy": 6.502096319198609,
"epoch": 0.817810086324398,
"grad_norm": 1.1875,
"learning_rate": 0.00044950000000000003,
"loss": 6.3819,
"mean_token_accuracy": 0.12381160259246826,
"num_tokens": 1852282.0,
"step": 900
},
{
"entropy": 6.393548440933228,
"epoch": 0.8223534756928669,
"grad_norm": 1.1015625,
"learning_rate": 0.00045200000000000004,
"loss": 6.3983,
"mean_token_accuracy": 0.11614296361804008,
"num_tokens": 1862830.0,
"step": 905
},
{
"entropy": 6.391427087783813,
"epoch": 0.8268968650613358,
"grad_norm": 1.0703125,
"learning_rate": 0.00045450000000000004,
"loss": 6.3597,
"mean_token_accuracy": 0.11984742507338524,
"num_tokens": 1872746.0,
"step": 910
},
{
"entropy": 6.431656169891357,
"epoch": 0.8314402544298046,
"grad_norm": 1.0859375,
"learning_rate": 0.00045700000000000005,
"loss": 6.3686,
"mean_token_accuracy": 0.11678105965256691,
"num_tokens": 1884353.0,
"step": 915
},
{
"entropy": 6.500422382354737,
"epoch": 0.8359836437982735,
"grad_norm": 1.03125,
"learning_rate": 0.00045950000000000006,
"loss": 6.4198,
"mean_token_accuracy": 0.11747792363166809,
"num_tokens": 1895719.0,
"step": 920
},
{
"entropy": 6.486808967590332,
"epoch": 0.8405270331667424,
"grad_norm": 1.0625,
"learning_rate": 0.000462,
"loss": 6.3507,
"mean_token_accuracy": 0.13011432066559792,
"num_tokens": 1905905.0,
"step": 925
},
{
"entropy": 6.443813133239746,
"epoch": 0.8450704225352113,
"grad_norm": 1.046875,
"learning_rate": 0.0004645,
"loss": 6.4556,
"mean_token_accuracy": 0.11997624412178994,
"num_tokens": 1916010.0,
"step": 930
},
{
"entropy": 6.399683046340942,
"epoch": 0.8496138119036801,
"grad_norm": 1.0625,
"learning_rate": 0.000467,
"loss": 6.343,
"mean_token_accuracy": 0.1263676293194294,
"num_tokens": 1926429.0,
"step": 935
},
{
"entropy": 6.243683576583862,
"epoch": 0.854157201272149,
"grad_norm": 1.0625,
"learning_rate": 0.0004695,
"loss": 6.2876,
"mean_token_accuracy": 0.1269657112658024,
"num_tokens": 1936322.0,
"step": 940
},
{
"entropy": 6.489983749389649,
"epoch": 0.8587005906406179,
"grad_norm": 1.1640625,
"learning_rate": 0.000472,
"loss": 6.4483,
"mean_token_accuracy": 0.1203050084412098,
"num_tokens": 1947885.0,
"step": 945
},
{
"entropy": 6.486026573181152,
"epoch": 0.8632439800090868,
"grad_norm": 1.03125,
"learning_rate": 0.0004745,
"loss": 6.4107,
"mean_token_accuracy": 0.11928511559963226,
"num_tokens": 1957876.0,
"step": 950
},
{
"entropy": 6.368922376632691,
"epoch": 0.8677873693775556,
"grad_norm": 1.09375,
"learning_rate": 0.000477,
"loss": 6.4371,
"mean_token_accuracy": 0.11843411177396775,
"num_tokens": 1969531.0,
"step": 955
},
{
"entropy": 6.399589204788208,
"epoch": 0.8723307587460245,
"grad_norm": 0.96484375,
"learning_rate": 0.0004795,
"loss": 6.2782,
"mean_token_accuracy": 0.12589425444602967,
"num_tokens": 1980787.0,
"step": 960
},
{
"entropy": 6.333202314376831,
"epoch": 0.8768741481144934,
"grad_norm": 1.0703125,
"learning_rate": 0.000482,
"loss": 6.3289,
"mean_token_accuracy": 0.12418413236737251,
"num_tokens": 1991645.0,
"step": 965
},
{
"entropy": 6.382894659042359,
"epoch": 0.8814175374829623,
"grad_norm": 0.96484375,
"learning_rate": 0.0004845,
"loss": 6.4095,
"mean_token_accuracy": 0.12133717462420464,
"num_tokens": 2002544.0,
"step": 970
},
{
"entropy": 6.457648229598999,
"epoch": 0.8859609268514311,
"grad_norm": 1.0625,
"learning_rate": 0.000487,
"loss": 6.4489,
"mean_token_accuracy": 0.11770377829670906,
"num_tokens": 2012813.0,
"step": 975
},
{
"entropy": 6.422171783447266,
"epoch": 0.8905043162199,
"grad_norm": 0.9453125,
"learning_rate": 0.0004895,
"loss": 6.3112,
"mean_token_accuracy": 0.12627919316291808,
"num_tokens": 2023756.0,
"step": 980
},
{
"entropy": 6.360575008392334,
"epoch": 0.8950477055883689,
"grad_norm": 1.0859375,
"learning_rate": 0.000492,
"loss": 6.3801,
"mean_token_accuracy": 0.12627830281853675,
"num_tokens": 2034245.0,
"step": 985
},
{
"entropy": 6.476523399353027,
"epoch": 0.8995910949568378,
"grad_norm": 1.078125,
"learning_rate": 0.0004945,
"loss": 6.4332,
"mean_token_accuracy": 0.11889058873057365,
"num_tokens": 2045635.0,
"step": 990
},
{
"entropy": 6.360070610046387,
"epoch": 0.9041344843253066,
"grad_norm": 0.93359375,
"learning_rate": 0.000497,
"loss": 6.4018,
"mean_token_accuracy": 0.12363604456186295,
"num_tokens": 2056501.0,
"step": 995
},
{
"entropy": 6.41779899597168,
"epoch": 0.9086778736937755,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995,
"loss": 6.3899,
"mean_token_accuracy": 0.1252191685140133,
"num_tokens": 2066540.0,
"step": 1000
},
{
"epoch": 0.9086778736937755,
"eval_entropy": 6.338110397028369,
"eval_loss": 6.4339213371276855,
"eval_mean_token_accuracy": 0.12323289418289828,
"eval_num_tokens": 2066540.0,
"eval_runtime": 2.7728,
"eval_samples_per_second": 1238.804,
"eval_steps_per_second": 155.076,
"step": 1000
},
{
"entropy": 6.3985589981079105,
"epoch": 0.9132212630622444,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999998223471442,
"loss": 6.2753,
"mean_token_accuracy": 0.12255016267299652,
"num_tokens": 2077008.0,
"step": 1005
},
{
"entropy": 6.28373155593872,
"epoch": 0.9177646524307133,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999991006328981,
"loss": 6.2049,
"mean_token_accuracy": 0.13157252669334413,
"num_tokens": 2087314.0,
"step": 1010
},
{
"entropy": 6.333494234085083,
"epoch": 0.9223080417991822,
"grad_norm": 1.0,
"learning_rate": 0.0004999978237557377,
"loss": 6.3847,
"mean_token_accuracy": 0.12330184951424598,
"num_tokens": 2097928.0,
"step": 1015
},
{
"entropy": 6.362055873870849,
"epoch": 0.9268514311676511,
"grad_norm": 1.03125,
"learning_rate": 0.0004999959917188137,
"loss": 6.3132,
"mean_token_accuracy": 0.12203029170632362,
"num_tokens": 2108662.0,
"step": 1020
},
{
"entropy": 6.352135372161865,
"epoch": 0.93139482053612,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999936045266462,
"loss": 6.3209,
"mean_token_accuracy": 0.12611013650894165,
"num_tokens": 2118546.0,
"step": 1025
},
{
"entropy": 6.3514327049255375,
"epoch": 0.9359382099045889,
"grad_norm": 1.0,
"learning_rate": 0.0004999906621851255,
"loss": 6.3648,
"mean_token_accuracy": 0.1252709671854973,
"num_tokens": 2128012.0,
"step": 1030
},
{
"entropy": 6.32377257347107,
"epoch": 0.9404815992730577,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999871647015115,
"loss": 6.1959,
"mean_token_accuracy": 0.12848494946956635,
"num_tokens": 2138241.0,
"step": 1035
},
{
"entropy": 6.3961326599121096,
"epoch": 0.9450249886415266,
"grad_norm": 0.90625,
"learning_rate": 0.0004999831120844339,
"loss": 6.3662,
"mean_token_accuracy": 0.12063193768262863,
"num_tokens": 2150780.0,
"step": 1040
},
{
"entropy": 6.2805500507354735,
"epoch": 0.9495683780099955,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999785043438922,
"loss": 6.2512,
"mean_token_accuracy": 0.12523388788104056,
"num_tokens": 2161918.0,
"step": 1045
},
{
"entropy": 6.2664916038513185,
"epoch": 0.9541117673784644,
"grad_norm": 0.96875,
"learning_rate": 0.0004999733414912555,
"loss": 6.2842,
"mean_token_accuracy": 0.12542618587613105,
"num_tokens": 2172742.0,
"step": 1050
},
{
"entropy": 6.25712833404541,
"epoch": 0.9586551567469332,
"grad_norm": 1.015625,
"learning_rate": 0.0004999676235392626,
"loss": 6.2735,
"mean_token_accuracy": 0.13171993270516397,
"num_tokens": 2182122.0,
"step": 1055
},
{
"entropy": 6.338843441009521,
"epoch": 0.9631985461154021,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999613505020219,
"loss": 6.3193,
"mean_token_accuracy": 0.12067236453294754,
"num_tokens": 2193703.0,
"step": 1060
},
{
"entropy": 6.3144971370697025,
"epoch": 0.967741935483871,
"grad_norm": 1.15625,
"learning_rate": 0.0004999545223950116,
"loss": 6.2244,
"mean_token_accuracy": 0.12638482749462127,
"num_tokens": 2204041.0,
"step": 1065
},
{
"entropy": 6.339418172836304,
"epoch": 0.9722853248523399,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999471392350795,
"loss": 6.2749,
"mean_token_accuracy": 0.12948141172528266,
"num_tokens": 2214276.0,
"step": 1070
},
{
"entropy": 6.198564529418945,
"epoch": 0.9768287142208087,
"grad_norm": 1.09375,
"learning_rate": 0.0004999392010404426,
"loss": 6.2439,
"mean_token_accuracy": 0.13011023327708243,
"num_tokens": 2225390.0,
"step": 1075
},
{
"entropy": 6.3132896900177,
"epoch": 0.9813721035892776,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999307078306876,
"loss": 6.2116,
"mean_token_accuracy": 0.12767509818077089,
"num_tokens": 2235708.0,
"step": 1080
},
{
"entropy": 6.332637929916382,
"epoch": 0.9859154929577465,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999216596267708,
"loss": 6.23,
"mean_token_accuracy": 0.13385264724493026,
"num_tokens": 2246284.0,
"step": 1085
},
{
"entropy": 6.245004749298095,
"epoch": 0.9904588823262154,
"grad_norm": 1.15625,
"learning_rate": 0.0004999120564510177,
"loss": 6.2549,
"mean_token_accuracy": 0.12341785728931427,
"num_tokens": 2257219.0,
"step": 1090
},
{
"entropy": 6.22976450920105,
"epoch": 0.9950022716946842,
"grad_norm": 0.9453125,
"learning_rate": 0.000499901898327123,
"loss": 6.2626,
"mean_token_accuracy": 0.12642209008336067,
"num_tokens": 2268019.0,
"step": 1095
},
{
"entropy": 6.227673864364624,
"epoch": 0.9995456610631531,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998911852801512,
"loss": 6.2597,
"mean_token_accuracy": 0.13009383752942086,
"num_tokens": 2278382.0,
"step": 1100
},
{
"entropy": 6.392165660858154,
"epoch": 1.003634711494775,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998799173365355,
"loss": 6.1073,
"mean_token_accuracy": 0.13042434884442222,
"num_tokens": 2288515.0,
"step": 1105
},
{
"entropy": 6.195787239074707,
"epoch": 1.008178100863244,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998680945240783,
"loss": 5.9273,
"mean_token_accuracy": 0.14239493906497955,
"num_tokens": 2298396.0,
"step": 1110
},
{
"entropy": 6.091786003112793,
"epoch": 1.0127214902317128,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998557168719513,
"loss": 6.0461,
"mean_token_accuracy": 0.1319786787033081,
"num_tokens": 2308707.0,
"step": 1115
},
{
"entropy": 6.274671411514282,
"epoch": 1.0172648796001817,
"grad_norm": 1.1875,
"learning_rate": 0.0004998427844106954,
"loss": 5.978,
"mean_token_accuracy": 0.1331262059509754,
"num_tokens": 2318981.0,
"step": 1120
},
{
"entropy": 6.118438577651977,
"epoch": 1.0218082689686505,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998292971722197,
"loss": 6.0852,
"mean_token_accuracy": 0.13883541151881218,
"num_tokens": 2329647.0,
"step": 1125
},
{
"entropy": 6.161949110031128,
"epoch": 1.0263516583371195,
"grad_norm": 1.0234375,
"learning_rate": 0.000499815255189803,
"loss": 6.0011,
"mean_token_accuracy": 0.13806321397423743,
"num_tokens": 2339449.0,
"step": 1130
},
{
"entropy": 6.220050716400147,
"epoch": 1.0308950477055883,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998006584980922,
"loss": 6.1039,
"mean_token_accuracy": 0.13446446210145951,
"num_tokens": 2349904.0,
"step": 1135
},
{
"entropy": 6.206082534790039,
"epoch": 1.0354384370740572,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997855071331034,
"loss": 6.0302,
"mean_token_accuracy": 0.14588112831115724,
"num_tokens": 2359833.0,
"step": 1140
},
{
"entropy": 6.221877527236939,
"epoch": 1.039981826442526,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997698011322211,
"loss": 6.0418,
"mean_token_accuracy": 0.13687821179628373,
"num_tokens": 2369857.0,
"step": 1145
},
{
"entropy": 6.155650329589844,
"epoch": 1.044525215810995,
"grad_norm": 1.1328125,
"learning_rate": 0.000499753540534198,
"loss": 6.0586,
"mean_token_accuracy": 0.14001990631222724,
"num_tokens": 2380372.0,
"step": 1150
},
{
"entropy": 6.160613679885865,
"epoch": 1.049068605179464,
"grad_norm": 1.1015625,
"learning_rate": 0.000499736725379156,
"loss": 6.0482,
"mean_token_accuracy": 0.13392850533127784,
"num_tokens": 2391239.0,
"step": 1155
},
{
"entropy": 6.0652158737182615,
"epoch": 1.0536119945479328,
"grad_norm": 1.125,
"learning_rate": 0.0004997193557085844,
"loss": 6.0262,
"mean_token_accuracy": 0.1400870054960251,
"num_tokens": 2401178.0,
"step": 1160
},
{
"entropy": 6.28751277923584,
"epoch": 1.0581553839164017,
"grad_norm": 1.0,
"learning_rate": 0.0004997014315653413,
"loss": 6.1046,
"mean_token_accuracy": 0.13458653688430786,
"num_tokens": 2412496.0,
"step": 1165
},
{
"entropy": 6.150702810287475,
"epoch": 1.0626987732848705,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996829529936527,
"loss": 6.1111,
"mean_token_accuracy": 0.1356681168079376,
"num_tokens": 2423281.0,
"step": 1170
},
{
"entropy": 6.143704080581665,
"epoch": 1.0672421626533395,
"grad_norm": 1.046875,
"learning_rate": 0.0004996639200391126,
"loss": 6.03,
"mean_token_accuracy": 0.14035207256674767,
"num_tokens": 2434078.0,
"step": 1175
},
{
"entropy": 6.086515951156616,
"epoch": 1.0717855520218083,
"grad_norm": 1.1875,
"learning_rate": 0.0004996443327486831,
"loss": 5.9761,
"mean_token_accuracy": 0.1377861127257347,
"num_tokens": 2443808.0,
"step": 1180
},
{
"entropy": 6.055760860443115,
"epoch": 1.0763289413902772,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996241911706937,
"loss": 6.0147,
"mean_token_accuracy": 0.14060620367527008,
"num_tokens": 2453397.0,
"step": 1185
},
{
"entropy": 6.206238746643066,
"epoch": 1.080872330758746,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996034953548418,
"loss": 6.104,
"mean_token_accuracy": 0.13623759970068933,
"num_tokens": 2464295.0,
"step": 1190
},
{
"entropy": 6.191244411468506,
"epoch": 1.085415720127215,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995822453521924,
"loss": 6.1554,
"mean_token_accuracy": 0.13175476416945459,
"num_tokens": 2475311.0,
"step": 1195
},
{
"entropy": 6.2322221279144285,
"epoch": 1.0899591094956838,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995604412151776,
"loss": 6.0449,
"mean_token_accuracy": 0.13293813169002533,
"num_tokens": 2485968.0,
"step": 1200
},
{
"entropy": 6.099966192245484,
"epoch": 1.0945024988641527,
"grad_norm": 1.109375,
"learning_rate": 0.000499538082997597,
"loss": 6.0802,
"mean_token_accuracy": 0.13437789678573608,
"num_tokens": 2496183.0,
"step": 1205
},
{
"entropy": 6.131676864624024,
"epoch": 1.0990458882326215,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995151707546172,
"loss": 6.0191,
"mean_token_accuracy": 0.13535890504717826,
"num_tokens": 2506242.0,
"step": 1210
},
{
"entropy": 6.2055047988891605,
"epoch": 1.1035892776010905,
"grad_norm": 1.0,
"learning_rate": 0.0004994917045427721,
"loss": 6.0469,
"mean_token_accuracy": 0.14065323099493982,
"num_tokens": 2516443.0,
"step": 1215
},
{
"entropy": 6.058293247222901,
"epoch": 1.1081326669695593,
"grad_norm": 1.1484375,
"learning_rate": 0.0004994676844199621,
"loss": 6.0054,
"mean_token_accuracy": 0.1424359381198883,
"num_tokens": 2525795.0,
"step": 1220
},
{
"entropy": 6.132228660583496,
"epoch": 1.1126760563380282,
"grad_norm": 1.09375,
"learning_rate": 0.0004994431104454543,
"loss": 6.0386,
"mean_token_accuracy": 0.13714250698685646,
"num_tokens": 2535046.0,
"step": 1225
},
{
"entropy": 6.124416637420654,
"epoch": 1.117219445706497,
"grad_norm": 1.1875,
"learning_rate": 0.0004994179826798828,
"loss": 6.0256,
"mean_token_accuracy": 0.13614149391651154,
"num_tokens": 2546489.0,
"step": 1230
},
{
"entropy": 5.94155330657959,
"epoch": 1.121762835074966,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993923011852477,
"loss": 5.9764,
"mean_token_accuracy": 0.13668037950992584,
"num_tokens": 2556335.0,
"step": 1235
},
{
"entropy": 6.228255271911621,
"epoch": 1.1263062244434348,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993660660249158,
"loss": 5.9903,
"mean_token_accuracy": 0.13471399769186973,
"num_tokens": 2567647.0,
"step": 1240
},
{
"entropy": 6.143363904953003,
"epoch": 1.1308496138119037,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993392772636194,
"loss": 6.1913,
"mean_token_accuracy": 0.12750165462493895,
"num_tokens": 2578618.0,
"step": 1245
},
{
"entropy": 6.062737894058228,
"epoch": 1.1353930031803725,
"grad_norm": 1.1640625,
"learning_rate": 0.0004993119349674573,
"loss": 6.0101,
"mean_token_accuracy": 0.13999856114387513,
"num_tokens": 2587797.0,
"step": 1250
},
{
"entropy": 6.136137866973877,
"epoch": 1.1399363925488415,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992840392038938,
"loss": 6.0529,
"mean_token_accuracy": 0.13363762199878693,
"num_tokens": 2598403.0,
"step": 1255
},
{
"entropy": 6.147263908386231,
"epoch": 1.1444797819173103,
"grad_norm": 1.0,
"learning_rate": 0.000499255590041759,
"loss": 6.0562,
"mean_token_accuracy": 0.13475360199809075,
"num_tokens": 2608296.0,
"step": 1260
},
{
"entropy": 6.153927946090699,
"epoch": 1.1490231712857792,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992265875512485,
"loss": 6.0793,
"mean_token_accuracy": 0.14075712710618973,
"num_tokens": 2617906.0,
"step": 1265
},
{
"entropy": 6.131653928756714,
"epoch": 1.153566560654248,
"grad_norm": 1.1171875,
"learning_rate": 0.0004991970318039228,
"loss": 6.0092,
"mean_token_accuracy": 0.1413343667984009,
"num_tokens": 2627699.0,
"step": 1270
},
{
"entropy": 6.081873416900635,
"epoch": 1.158109950022717,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991669228727078,
"loss": 5.9884,
"mean_token_accuracy": 0.14022635594010352,
"num_tokens": 2638384.0,
"step": 1275
},
{
"entropy": 5.98123574256897,
"epoch": 1.1626533393911858,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991362608318947,
"loss": 5.9206,
"mean_token_accuracy": 0.1425674617290497,
"num_tokens": 2648279.0,
"step": 1280
},
{
"entropy": 6.091414165496826,
"epoch": 1.1671967287596547,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991050457571384,
"loss": 6.0074,
"mean_token_accuracy": 0.12979530766606331,
"num_tokens": 2657000.0,
"step": 1285
},
{
"entropy": 6.07979564666748,
"epoch": 1.1717401181281235,
"grad_norm": 0.99609375,
"learning_rate": 0.0004990732777254594,
"loss": 6.0393,
"mean_token_accuracy": 0.13943152800202369,
"num_tokens": 2667760.0,
"step": 1290
},
{
"entropy": 6.137358522415161,
"epoch": 1.1762835074965925,
"grad_norm": 0.984375,
"learning_rate": 0.0004990409568152422,
"loss": 5.9685,
"mean_token_accuracy": 0.13984878063201905,
"num_tokens": 2678450.0,
"step": 1295
},
{
"entropy": 6.110213851928711,
"epoch": 1.1808268968650613,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990080831062352,
"loss": 6.0739,
"mean_token_accuracy": 0.13112835064530373,
"num_tokens": 2689695.0,
"step": 1300
},
{
"entropy": 6.112369632720947,
"epoch": 1.1853702862335302,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989746566795512,
"loss": 6.0807,
"mean_token_accuracy": 0.14619563147425652,
"num_tokens": 2700414.0,
"step": 1305
},
{
"entropy": 6.007969236373901,
"epoch": 1.189913675601999,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989406776176665,
"loss": 5.8977,
"mean_token_accuracy": 0.1559330329298973,
"num_tokens": 2709985.0,
"step": 1310
},
{
"entropy": 6.104906463623047,
"epoch": 1.194457064970468,
"grad_norm": 1.1796875,
"learning_rate": 0.000498906146004421,
"loss": 6.0631,
"mean_token_accuracy": 0.14401261359453202,
"num_tokens": 2720990.0,
"step": 1315
},
{
"entropy": 5.987970018386841,
"epoch": 1.1990004543389368,
"grad_norm": 1.0859375,
"learning_rate": 0.0004988710619250181,
"loss": 5.9693,
"mean_token_accuracy": 0.14561515301465988,
"num_tokens": 2731285.0,
"step": 1320
},
{
"entropy": 6.058353090286255,
"epoch": 1.2035438437074057,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988354254660244,
"loss": 6.0477,
"mean_token_accuracy": 0.1376740589737892,
"num_tokens": 2741024.0,
"step": 1325
},
{
"entropy": 6.086800527572632,
"epoch": 1.2080872330758745,
"grad_norm": 1.015625,
"learning_rate": 0.0004987992367153691,
"loss": 5.9733,
"mean_token_accuracy": 0.14504139125347137,
"num_tokens": 2751447.0,
"step": 1330
},
{
"entropy": 6.072046089172363,
"epoch": 1.2126306224443435,
"grad_norm": 1.0625,
"learning_rate": 0.0004987624957623443,
"loss": 5.9849,
"mean_token_accuracy": 0.1435820832848549,
"num_tokens": 2761283.0,
"step": 1335
},
{
"entropy": 5.95879430770874,
"epoch": 1.2171740118128123,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987252026976049,
"loss": 5.9155,
"mean_token_accuracy": 0.14698221608996392,
"num_tokens": 2771354.0,
"step": 1340
},
{
"entropy": 6.137785863876343,
"epoch": 1.2217174011812812,
"grad_norm": 1.078125,
"learning_rate": 0.0004986873576131676,
"loss": 6.1039,
"mean_token_accuracy": 0.139956097304821,
"num_tokens": 2781063.0,
"step": 1345
},
{
"entropy": 6.125413656234741,
"epoch": 1.22626079054975,
"grad_norm": 1.09375,
"learning_rate": 0.0004986489606024116,
"loss": 6.0217,
"mean_token_accuracy": 0.1403006963431835,
"num_tokens": 2791103.0,
"step": 1350
},
{
"entropy": 6.050103759765625,
"epoch": 1.230804179918219,
"grad_norm": 1.0,
"learning_rate": 0.0004986100117600776,
"loss": 6.1096,
"mean_token_accuracy": 0.13651306107640265,
"num_tokens": 2802382.0,
"step": 1355
},
{
"entropy": 6.057727241516114,
"epoch": 1.235347569286688,
"grad_norm": 0.9375,
"learning_rate": 0.000498570511182268,
"loss": 5.9751,
"mean_token_accuracy": 0.13772271424531937,
"num_tokens": 2813740.0,
"step": 1360
},
{
"entropy": 6.108279895782471,
"epoch": 1.2398909586551567,
"grad_norm": 1.046875,
"learning_rate": 0.0004985304589664465,
"loss": 5.9857,
"mean_token_accuracy": 0.1388835422694683,
"num_tokens": 2824121.0,
"step": 1365
},
{
"entropy": 6.027289724349975,
"epoch": 1.2444343480236255,
"grad_norm": 1.1796875,
"learning_rate": 0.0004984898552114381,
"loss": 6.024,
"mean_token_accuracy": 0.14008374959230424,
"num_tokens": 2833356.0,
"step": 1370
},
{
"entropy": 6.076797246932983,
"epoch": 1.2489777373920945,
"grad_norm": 1.078125,
"learning_rate": 0.0004984487000174283,
"loss": 5.9975,
"mean_token_accuracy": 0.13435959443449974,
"num_tokens": 2843839.0,
"step": 1375
},
{
"entropy": 6.056558895111084,
"epoch": 1.2535211267605635,
"grad_norm": 1.0625,
"learning_rate": 0.0004984069934859637,
"loss": 5.9499,
"mean_token_accuracy": 0.141408871114254,
"num_tokens": 2853523.0,
"step": 1380
},
{
"entropy": 5.993710041046143,
"epoch": 1.2580645161290323,
"grad_norm": 1.046875,
"learning_rate": 0.0004983647357199509,
"loss": 6.0104,
"mean_token_accuracy": 0.1399746559560299,
"num_tokens": 2863226.0,
"step": 1385
},
{
"entropy": 6.15961971282959,
"epoch": 1.262607905497501,
"grad_norm": 0.98046875,
"learning_rate": 0.0004983219268236567,
"loss": 6.058,
"mean_token_accuracy": 0.14663353711366653,
"num_tokens": 2873822.0,
"step": 1390
},
{
"entropy": 6.0136792182922365,
"epoch": 1.26715129486597,
"grad_norm": 0.98828125,
"learning_rate": 0.0004982785669027079,
"loss": 5.9099,
"mean_token_accuracy": 0.144786985963583,
"num_tokens": 2883645.0,
"step": 1395
},
{
"entropy": 6.058310556411743,
"epoch": 1.271694684234439,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982346560640907,
"loss": 6.037,
"mean_token_accuracy": 0.13750747442245484,
"num_tokens": 2894397.0,
"step": 1400
},
{
"entropy": 6.104885101318359,
"epoch": 1.2762380736029078,
"grad_norm": 1.0,
"learning_rate": 0.0004981901944161507,
"loss": 6.0125,
"mean_token_accuracy": 0.1382710665464401,
"num_tokens": 2904947.0,
"step": 1405
},
{
"entropy": 5.943738698959351,
"epoch": 1.2807814629713765,
"grad_norm": 0.98046875,
"learning_rate": 0.0004981451820685928,
"loss": 5.9622,
"mean_token_accuracy": 0.14593919217586518,
"num_tokens": 2916183.0,
"step": 1410
},
{
"entropy": 6.031915140151978,
"epoch": 1.2853248523398455,
"grad_norm": 1.171875,
"learning_rate": 0.0004980996191324802,
"loss": 5.8843,
"mean_token_accuracy": 0.13760354071855546,
"num_tokens": 2925429.0,
"step": 1415
},
{
"entropy": 6.027405691146851,
"epoch": 1.2898682417083145,
"grad_norm": 1.0625,
"learning_rate": 0.0004980535057202351,
"loss": 5.9508,
"mean_token_accuracy": 0.1398024417459965,
"num_tokens": 2935142.0,
"step": 1420
},
{
"entropy": 6.043066024780273,
"epoch": 1.2944116310767833,
"grad_norm": 1.140625,
"learning_rate": 0.0004980068419456378,
"loss": 6.0158,
"mean_token_accuracy": 0.1414060413837433,
"num_tokens": 2945151.0,
"step": 1425
},
{
"entropy": 6.113505744934082,
"epoch": 1.2989550204452522,
"grad_norm": 1.1484375,
"learning_rate": 0.0004979596279238265,
"loss": 6.1119,
"mean_token_accuracy": 0.1392623744904995,
"num_tokens": 2956015.0,
"step": 1430
},
{
"entropy": 6.003687238693237,
"epoch": 1.303498409813721,
"grad_norm": 1.0859375,
"learning_rate": 0.0004979118637712969,
"loss": 5.9551,
"mean_token_accuracy": 0.14146706014871596,
"num_tokens": 2965879.0,
"step": 1435
},
{
"entropy": 6.087925672531128,
"epoch": 1.30804179918219,
"grad_norm": 1.015625,
"learning_rate": 0.0004978635496059025,
"loss": 6.0225,
"mean_token_accuracy": 0.13810051381587982,
"num_tokens": 2977021.0,
"step": 1440
},
{
"entropy": 5.9836341381073,
"epoch": 1.3125851885506588,
"grad_norm": 1.0390625,
"learning_rate": 0.0004978146855468537,
"loss": 5.9217,
"mean_token_accuracy": 0.14723185896873475,
"num_tokens": 2987169.0,
"step": 1445
},
{
"entropy": 6.111929321289063,
"epoch": 1.3171285779191277,
"grad_norm": 0.91796875,
"learning_rate": 0.0004977652717147177,
"loss": 6.0299,
"mean_token_accuracy": 0.14329671561717988,
"num_tokens": 2997666.0,
"step": 1450
},
{
"entropy": 5.985518598556519,
"epoch": 1.3216719672875965,
"grad_norm": 1.1015625,
"learning_rate": 0.0004977153082314182,
"loss": 5.8744,
"mean_token_accuracy": 0.1451878011226654,
"num_tokens": 3007021.0,
"step": 1455
},
{
"entropy": 5.98281946182251,
"epoch": 1.3262153566560655,
"grad_norm": 1.1015625,
"learning_rate": 0.000497664795220235,
"loss": 5.9547,
"mean_token_accuracy": 0.14797762855887414,
"num_tokens": 3016557.0,
"step": 1460
},
{
"entropy": 5.992539787292481,
"epoch": 1.3307587460245343,
"grad_norm": 1.0,
"learning_rate": 0.0004976137328058042,
"loss": 5.9177,
"mean_token_accuracy": 0.1424129754304886,
"num_tokens": 3026146.0,
"step": 1465
},
{
"entropy": 6.0835624694824215,
"epoch": 1.3353021353930032,
"grad_norm": 1.0390625,
"learning_rate": 0.000497562121114117,
"loss": 5.9871,
"mean_token_accuracy": 0.13852132931351663,
"num_tokens": 3037234.0,
"step": 1470
},
{
"entropy": 5.974925851821899,
"epoch": 1.339845524761472,
"grad_norm": 1.046875,
"learning_rate": 0.0004975099602725202,
"loss": 5.9224,
"mean_token_accuracy": 0.14474036395549775,
"num_tokens": 3047282.0,
"step": 1475
},
{
"entropy": 5.96318564414978,
"epoch": 1.344388914129941,
"grad_norm": 1.0,
"learning_rate": 0.0004974572504097154,
"loss": 5.9421,
"mean_token_accuracy": 0.14181292355060576,
"num_tokens": 3058648.0,
"step": 1480
},
{
"entropy": 6.146981859207154,
"epoch": 1.3489323034984098,
"grad_norm": 1.0,
"learning_rate": 0.0004974039916557591,
"loss": 6.1159,
"mean_token_accuracy": 0.13969697952270507,
"num_tokens": 3068996.0,
"step": 1485
},
{
"entropy": 5.975324487686157,
"epoch": 1.3534756928668787,
"grad_norm": 1.109375,
"learning_rate": 0.000497350184142062,
"loss": 5.9292,
"mean_token_accuracy": 0.15094703137874604,
"num_tokens": 3078991.0,
"step": 1490
},
{
"entropy": 5.975798654556274,
"epoch": 1.3580190822353475,
"grad_norm": 1.03125,
"learning_rate": 0.0004972958280013886,
"loss": 5.9107,
"mean_token_accuracy": 0.14440180808305741,
"num_tokens": 3090200.0,
"step": 1495
},
{
"entropy": 5.997968149185181,
"epoch": 1.3625624716038165,
"grad_norm": 1.015625,
"learning_rate": 0.0004972409233678575,
"loss": 5.9518,
"mean_token_accuracy": 0.14175319969654082,
"num_tokens": 3101971.0,
"step": 1500
},
{
"epoch": 1.3625624716038165,
"eval_entropy": 5.892982365364253,
"eval_loss": 6.215432167053223,
"eval_mean_token_accuracy": 0.1367779885353737,
"eval_num_tokens": 3101971.0,
"eval_runtime": 2.5782,
"eval_samples_per_second": 1332.319,
"eval_steps_per_second": 166.782,
"step": 1500
},
{
"entropy": 6.005481147766114,
"epoch": 1.3671058609722853,
"grad_norm": 0.99609375,
"learning_rate": 0.0004971854703769401,
"loss": 6.0516,
"mean_token_accuracy": 0.13838631585240363,
"num_tokens": 3113381.0,
"step": 1505
},
{
"entropy": 6.088227128982544,
"epoch": 1.3716492503407542,
"grad_norm": 0.93359375,
"learning_rate": 0.0004971294691654616,
"loss": 6.0106,
"mean_token_accuracy": 0.14234280809760094,
"num_tokens": 3125099.0,
"step": 1510
},
{
"entropy": 5.928353261947632,
"epoch": 1.376192639709223,
"grad_norm": 1.015625,
"learning_rate": 0.000497072919871599,
"loss": 5.9037,
"mean_token_accuracy": 0.14393163919448854,
"num_tokens": 3135007.0,
"step": 1515
},
{
"entropy": 6.051437568664551,
"epoch": 1.380736029077692,
"grad_norm": 1.0703125,
"learning_rate": 0.0004970158226348824,
"loss": 5.9561,
"mean_token_accuracy": 0.14158373177051545,
"num_tokens": 3145031.0,
"step": 1520
},
{
"entropy": 6.030322074890137,
"epoch": 1.3852794184461608,
"grad_norm": 0.953125,
"learning_rate": 0.0004969581775961932,
"loss": 5.9914,
"mean_token_accuracy": 0.14449616745114327,
"num_tokens": 3156044.0,
"step": 1525
},
{
"entropy": 5.899796342849731,
"epoch": 1.3898228078146297,
"grad_norm": 1.140625,
"learning_rate": 0.0004968999848977651,
"loss": 5.889,
"mean_token_accuracy": 0.14498736411333085,
"num_tokens": 3165124.0,
"step": 1530
},
{
"entropy": 5.863690423965454,
"epoch": 1.3943661971830985,
"grad_norm": 1.0703125,
"learning_rate": 0.0004968412446831827,
"loss": 5.8279,
"mean_token_accuracy": 0.14925860613584518,
"num_tokens": 3175246.0,
"step": 1535
},
{
"entropy": 6.081419372558594,
"epoch": 1.3989095865515675,
"grad_norm": 1.0703125,
"learning_rate": 0.0004967819570973816,
"loss": 5.9191,
"mean_token_accuracy": 0.1539956286549568,
"num_tokens": 3184817.0,
"step": 1540
},
{
"entropy": 5.962965250015259,
"epoch": 1.4034529759200365,
"grad_norm": 1.0703125,
"learning_rate": 0.000496722122286648,
"loss": 5.9614,
"mean_token_accuracy": 0.1343929760158062,
"num_tokens": 3195142.0,
"step": 1545
},
{
"entropy": 5.950141382217407,
"epoch": 1.4079963652885052,
"grad_norm": 1.0625,
"learning_rate": 0.0004966617403986185,
"loss": 5.9141,
"mean_token_accuracy": 0.14692503362894058,
"num_tokens": 3205724.0,
"step": 1550
},
{
"entropy": 6.02142071723938,
"epoch": 1.412539754656974,
"grad_norm": 0.9375,
"learning_rate": 0.0004966008115822793,
"loss": 6.0553,
"mean_token_accuracy": 0.13776205033063887,
"num_tokens": 3216919.0,
"step": 1555
},
{
"entropy": 6.137990522384643,
"epoch": 1.417083144025443,
"grad_norm": 0.96484375,
"learning_rate": 0.0004965393359879662,
"loss": 6.0567,
"mean_token_accuracy": 0.1423663817346096,
"num_tokens": 3228929.0,
"step": 1560
},
{
"entropy": 5.951554155349731,
"epoch": 1.421626533393912,
"grad_norm": 1.046875,
"learning_rate": 0.0004964773137673641,
"loss": 6.0115,
"mean_token_accuracy": 0.14155814796686172,
"num_tokens": 3239541.0,
"step": 1565
},
{
"entropy": 6.036742401123047,
"epoch": 1.4261699227623807,
"grad_norm": 1.046875,
"learning_rate": 0.0004964147450735067,
"loss": 5.8767,
"mean_token_accuracy": 0.14599085748195648,
"num_tokens": 3249527.0,
"step": 1570
},
{
"entropy": 5.923075103759766,
"epoch": 1.4307133121308495,
"grad_norm": 0.96875,
"learning_rate": 0.000496351630060776,
"loss": 5.7891,
"mean_token_accuracy": 0.14622468650341033,
"num_tokens": 3260497.0,
"step": 1575
},
{
"entropy": 6.04791111946106,
"epoch": 1.4352567014993185,
"grad_norm": 1.1484375,
"learning_rate": 0.0004962879688849021,
"loss": 5.9794,
"mean_token_accuracy": 0.14303412139415742,
"num_tokens": 3270865.0,
"step": 1580
},
{
"entropy": 5.8683256149292,
"epoch": 1.4398000908677875,
"grad_norm": 0.9921875,
"learning_rate": 0.0004962237617029625,
"loss": 5.9076,
"mean_token_accuracy": 0.14470455199480056,
"num_tokens": 3281531.0,
"step": 1585
},
{
"entropy": 6.022738790512085,
"epoch": 1.4443434802362562,
"grad_norm": 1.09375,
"learning_rate": 0.0004961590086733821,
"loss": 5.8963,
"mean_token_accuracy": 0.1434609390795231,
"num_tokens": 3291970.0,
"step": 1590
},
{
"entropy": 5.9146098613739015,
"epoch": 1.448886869604725,
"grad_norm": 0.9921875,
"learning_rate": 0.0004960937099559326,
"loss": 5.9369,
"mean_token_accuracy": 0.14627966284751892,
"num_tokens": 3301638.0,
"step": 1595
},
{
"entropy": 6.1201530456542965,
"epoch": 1.453430258973194,
"grad_norm": 1.125,
"learning_rate": 0.0004960278657117321,
"loss": 6.0016,
"mean_token_accuracy": 0.14531999602913856,
"num_tokens": 3311780.0,
"step": 1600
},
{
"entropy": 5.894401025772095,
"epoch": 1.457973648341663,
"grad_norm": 1.0703125,
"learning_rate": 0.0004959614761032446,
"loss": 5.8609,
"mean_token_accuracy": 0.1396932542324066,
"num_tokens": 3322096.0,
"step": 1605
},
{
"entropy": 6.111455631256104,
"epoch": 1.4625170377101318,
"grad_norm": 1.1171875,
"learning_rate": 0.00049589454129428,
"loss": 6.0273,
"mean_token_accuracy": 0.13923634737730026,
"num_tokens": 3332324.0,
"step": 1610
},
{
"entropy": 5.867942953109742,
"epoch": 1.4670604270786005,
"grad_norm": 1.125,
"learning_rate": 0.0004958270614499932,
"loss": 5.8085,
"mean_token_accuracy": 0.15766735821962358,
"num_tokens": 3342037.0,
"step": 1615
},
{
"entropy": 6.056097173690796,
"epoch": 1.4716038164470695,
"grad_norm": 0.96875,
"learning_rate": 0.0004957590367368841,
"loss": 5.9165,
"mean_token_accuracy": 0.15122188329696656,
"num_tokens": 3353047.0,
"step": 1620
},
{
"entropy": 5.872474098205567,
"epoch": 1.4761472058155385,
"grad_norm": 0.96484375,
"learning_rate": 0.0004956904673227968,
"loss": 5.8268,
"mean_token_accuracy": 0.14686344414949418,
"num_tokens": 3363841.0,
"step": 1625
},
{
"entropy": 5.953419733047485,
"epoch": 1.4806905951840073,
"grad_norm": 1.0703125,
"learning_rate": 0.0004956213533769195,
"loss": 5.9379,
"mean_token_accuracy": 0.14522140622138976,
"num_tokens": 3374086.0,
"step": 1630
},
{
"entropy": 5.944189929962159,
"epoch": 1.485233984552476,
"grad_norm": 1.1328125,
"learning_rate": 0.0004955516950697843,
"loss": 5.9045,
"mean_token_accuracy": 0.1416383869946003,
"num_tokens": 3384303.0,
"step": 1635
},
{
"entropy": 5.964085388183594,
"epoch": 1.489777373920945,
"grad_norm": 1.046875,
"learning_rate": 0.0004954814925732657,
"loss": 5.8763,
"mean_token_accuracy": 0.15046490281820296,
"num_tokens": 3394858.0,
"step": 1640
},
{
"entropy": 5.888884687423706,
"epoch": 1.494320763289414,
"grad_norm": 0.94921875,
"learning_rate": 0.0004954107460605816,
"loss": 5.8955,
"mean_token_accuracy": 0.15163283944129943,
"num_tokens": 3405252.0,
"step": 1645
},
{
"entropy": 6.067559766769409,
"epoch": 1.4988641526578828,
"grad_norm": 1.09375,
"learning_rate": 0.0004953394557062922,
"loss": 5.9605,
"mean_token_accuracy": 0.14592973738908768,
"num_tokens": 3416007.0,
"step": 1650
},
{
"entropy": 6.0492658615112305,
"epoch": 1.5034075420263515,
"grad_norm": 1.09375,
"learning_rate": 0.0004952676216862989,
"loss": 5.9803,
"mean_token_accuracy": 0.14458779096603394,
"num_tokens": 3426932.0,
"step": 1655
},
{
"entropy": 5.861869144439697,
"epoch": 1.5079509313948205,
"grad_norm": 1.0546875,
"learning_rate": 0.0004951952441778453,
"loss": 5.8586,
"mean_token_accuracy": 0.15005863606929778,
"num_tokens": 3437262.0,
"step": 1660
},
{
"entropy": 5.936305952072144,
"epoch": 1.5124943207632895,
"grad_norm": 0.98828125,
"learning_rate": 0.0004951223233595158,
"loss": 5.8475,
"mean_token_accuracy": 0.15032792538404466,
"num_tokens": 3447525.0,
"step": 1665
},
{
"entropy": 5.97404670715332,
"epoch": 1.5170377101317583,
"grad_norm": 1.03125,
"learning_rate": 0.000495048859411235,
"loss": 5.9417,
"mean_token_accuracy": 0.1464758947491646,
"num_tokens": 3458198.0,
"step": 1670
},
{
"entropy": 5.94585599899292,
"epoch": 1.521581099500227,
"grad_norm": 1.0546875,
"learning_rate": 0.0004949748525142681,
"loss": 5.8301,
"mean_token_accuracy": 0.15196821689605713,
"num_tokens": 3468149.0,
"step": 1675
},
{
"entropy": 5.897303104400635,
"epoch": 1.526124488868696,
"grad_norm": 1.078125,
"learning_rate": 0.0004949003028512195,
"loss": 5.8017,
"mean_token_accuracy": 0.1441194847226143,
"num_tokens": 3478355.0,
"step": 1680
},
{
"entropy": 5.938935089111328,
"epoch": 1.530667878237165,
"grad_norm": 1.0078125,
"learning_rate": 0.0004948252106060334,
"loss": 5.8465,
"mean_token_accuracy": 0.14930058568716048,
"num_tokens": 3488726.0,
"step": 1685
},
{
"entropy": 5.873471784591675,
"epoch": 1.5352112676056338,
"grad_norm": 1.09375,
"learning_rate": 0.0004947495759639923,
"loss": 5.8189,
"mean_token_accuracy": 0.15210621953010559,
"num_tokens": 3498346.0,
"step": 1690
},
{
"entropy": 5.903305149078369,
"epoch": 1.5397546569741025,
"grad_norm": 1.0546875,
"learning_rate": 0.0004946733991117171,
"loss": 5.8781,
"mean_token_accuracy": 0.14939512461423873,
"num_tokens": 3508213.0,
"step": 1695
},
{
"entropy": 6.035619163513184,
"epoch": 1.5442980463425715,
"grad_norm": 1.03125,
"learning_rate": 0.0004945966802371668,
"loss": 6.0111,
"mean_token_accuracy": 0.14754971712827683,
"num_tokens": 3518978.0,
"step": 1700
},
{
"entropy": 5.951667642593383,
"epoch": 1.5488414357110405,
"grad_norm": 0.94921875,
"learning_rate": 0.0004945194195296374,
"loss": 5.9478,
"mean_token_accuracy": 0.14694140776991843,
"num_tokens": 3529328.0,
"step": 1705
},
{
"entropy": 5.887544727325439,
"epoch": 1.5533848250795095,
"grad_norm": 0.94140625,
"learning_rate": 0.000494441617179762,
"loss": 5.8827,
"mean_token_accuracy": 0.14316876903176307,
"num_tokens": 3539347.0,
"step": 1710
},
{
"entropy": 6.013899946212769,
"epoch": 1.5579282144479782,
"grad_norm": 1.09375,
"learning_rate": 0.0004943632733795104,
"loss": 5.9003,
"mean_token_accuracy": 0.14003022536635398,
"num_tokens": 3549798.0,
"step": 1715
},
{
"entropy": 5.915519189834595,
"epoch": 1.562471603816447,
"grad_norm": 1.09375,
"learning_rate": 0.0004942843883221879,
"loss": 5.9884,
"mean_token_accuracy": 0.13486514165997504,
"num_tokens": 3560363.0,
"step": 1720
},
{
"entropy": 5.974686098098755,
"epoch": 1.567014993184916,
"grad_norm": 1.0859375,
"learning_rate": 0.0004942049622024357,
"loss": 5.928,
"mean_token_accuracy": 0.1446959048509598,
"num_tokens": 3570080.0,
"step": 1725
},
{
"entropy": 5.8747306823730465,
"epoch": 1.571558382553385,
"grad_norm": 0.98046875,
"learning_rate": 0.0004941249952162298,
"loss": 5.9056,
"mean_token_accuracy": 0.1480409488081932,
"num_tokens": 3580053.0,
"step": 1730
},
{
"entropy": 6.027320003509521,
"epoch": 1.5761017719218537,
"grad_norm": 0.92578125,
"learning_rate": 0.0004940444875608807,
"loss": 6.0014,
"mean_token_accuracy": 0.14315830692648887,
"num_tokens": 3590119.0,
"step": 1735
},
{
"entropy": 6.038210821151734,
"epoch": 1.5806451612903225,
"grad_norm": 1.0234375,
"learning_rate": 0.0004939634394350332,
"loss": 5.9874,
"mean_token_accuracy": 0.13794291466474534,
"num_tokens": 3601809.0,
"step": 1740
},
{
"entropy": 6.032684993743897,
"epoch": 1.5851885506587915,
"grad_norm": 1.0390625,
"learning_rate": 0.0004938818510386654,
"loss": 5.8442,
"mean_token_accuracy": 0.15077518969774245,
"num_tokens": 3611752.0,
"step": 1745
},
{
"entropy": 5.873696613311767,
"epoch": 1.5897319400272605,
"grad_norm": 1.09375,
"learning_rate": 0.0004937997225730886,
"loss": 5.8985,
"mean_token_accuracy": 0.14551339894533158,
"num_tokens": 3621137.0,
"step": 1750
},
{
"entropy": 5.9400163173675535,
"epoch": 1.5942753293957292,
"grad_norm": 1.0390625,
"learning_rate": 0.0004937170542409464,
"loss": 5.8711,
"mean_token_accuracy": 0.14604211002588272,
"num_tokens": 3631567.0,
"step": 1755
},
{
"entropy": 5.902632474899292,
"epoch": 1.598818718764198,
"grad_norm": 1.046875,
"learning_rate": 0.000493633846246215,
"loss": 5.8428,
"mean_token_accuracy": 0.1430719755589962,
"num_tokens": 3641145.0,
"step": 1760
},
{
"entropy": 5.951475715637207,
"epoch": 1.603362108132667,
"grad_norm": 1.0,
"learning_rate": 0.0004935500987942018,
"loss": 5.8948,
"mean_token_accuracy": 0.14016074314713478,
"num_tokens": 3651716.0,
"step": 1765
},
{
"entropy": 5.927920579910278,
"epoch": 1.607905497501136,
"grad_norm": 1.0390625,
"learning_rate": 0.0004934658120915452,
"loss": 5.833,
"mean_token_accuracy": 0.15086506456136703,
"num_tokens": 3662577.0,
"step": 1770
},
{
"entropy": 5.850854206085205,
"epoch": 1.6124488868696047,
"grad_norm": 1.0234375,
"learning_rate": 0.0004933809863462142,
"loss": 5.8201,
"mean_token_accuracy": 0.1503201723098755,
"num_tokens": 3672984.0,
"step": 1775
},
{
"entropy": 5.90257248878479,
"epoch": 1.6169922762380735,
"grad_norm": 1.078125,
"learning_rate": 0.0004932956217675082,
"loss": 5.9065,
"mean_token_accuracy": 0.14844751954078675,
"num_tokens": 3682634.0,
"step": 1780
},
{
"entropy": 5.9094566822052,
"epoch": 1.6215356656065425,
"grad_norm": 1.1484375,
"learning_rate": 0.0004932097185660555,
"loss": 5.8783,
"mean_token_accuracy": 0.14262584149837493,
"num_tokens": 3693022.0,
"step": 1785
},
{
"entropy": 5.858315515518188,
"epoch": 1.6260790549750115,
"grad_norm": 1.046875,
"learning_rate": 0.0004931232769538138,
"loss": 5.747,
"mean_token_accuracy": 0.15391668528318406,
"num_tokens": 3702234.0,
"step": 1790
},
{
"entropy": 5.847795009613037,
"epoch": 1.6306224443434802,
"grad_norm": 1.0703125,
"learning_rate": 0.0004930362971440692,
"loss": 5.9219,
"mean_token_accuracy": 0.14441194981336594,
"num_tokens": 3712235.0,
"step": 1795
},
{
"entropy": 5.999974346160888,
"epoch": 1.635165833711949,
"grad_norm": 1.0078125,
"learning_rate": 0.0004929487793514358,
"loss": 5.8547,
"mean_token_accuracy": 0.154004468023777,
"num_tokens": 3722745.0,
"step": 1800
},
{
"entropy": 5.845955562591553,
"epoch": 1.639709223080418,
"grad_norm": 1.0390625,
"learning_rate": 0.000492860723791855,
"loss": 5.9095,
"mean_token_accuracy": 0.1449434533715248,
"num_tokens": 3732942.0,
"step": 1805
},
{
"entropy": 5.976245260238647,
"epoch": 1.644252612448887,
"grad_norm": 1.125,
"learning_rate": 0.0004927721306825952,
"loss": 5.8839,
"mean_token_accuracy": 0.14763342440128327,
"num_tokens": 3743806.0,
"step": 1810
},
{
"entropy": 5.88842740058899,
"epoch": 1.6487960018173557,
"grad_norm": 1.0,
"learning_rate": 0.0004926830002422509,
"loss": 5.895,
"mean_token_accuracy": 0.1496370851993561,
"num_tokens": 3755115.0,
"step": 1815
},
{
"entropy": 5.899521589279175,
"epoch": 1.6533393911858245,
"grad_norm": 1.03125,
"learning_rate": 0.000492593332690743,
"loss": 5.853,
"mean_token_accuracy": 0.1544428810477257,
"num_tokens": 3765867.0,
"step": 1820
},
{
"entropy": 5.831729698181152,
"epoch": 1.6578827805542935,
"grad_norm": 1.109375,
"learning_rate": 0.0004925031282493167,
"loss": 5.8164,
"mean_token_accuracy": 0.14891895055770873,
"num_tokens": 3777071.0,
"step": 1825
},
{
"entropy": 5.943818283081055,
"epoch": 1.6624261699227625,
"grad_norm": 1.0390625,
"learning_rate": 0.0004924123871405428,
"loss": 5.8283,
"mean_token_accuracy": 0.14979569017887115,
"num_tokens": 3786554.0,
"step": 1830
},
{
"entropy": 5.918811511993408,
"epoch": 1.6669695592912313,
"grad_norm": 1.1015625,
"learning_rate": 0.0004923211095883161,
"loss": 5.8673,
"mean_token_accuracy": 0.14998457729816436,
"num_tokens": 3797627.0,
"step": 1835
},
{
"entropy": 5.818353986740112,
"epoch": 1.6715129486597,
"grad_norm": 1.046875,
"learning_rate": 0.0004922292958178546,
"loss": 5.8471,
"mean_token_accuracy": 0.14822790026664734,
"num_tokens": 3808841.0,
"step": 1840
},
{
"entropy": 5.9494139671325685,
"epoch": 1.676056338028169,
"grad_norm": 1.15625,
"learning_rate": 0.0004921369460557,
"loss": 5.8922,
"mean_token_accuracy": 0.14491156712174416,
"num_tokens": 3818995.0,
"step": 1845
},
{
"entropy": 5.9480773448944095,
"epoch": 1.680599727396638,
"grad_norm": 1.09375,
"learning_rate": 0.0004920440605297159,
"loss": 5.9488,
"mean_token_accuracy": 0.14302500188350678,
"num_tokens": 3828762.0,
"step": 1850
},
{
"entropy": 5.930938720703125,
"epoch": 1.6851431167651068,
"grad_norm": 1.078125,
"learning_rate": 0.000491950639469088,
"loss": 5.7839,
"mean_token_accuracy": 0.14696138948202134,
"num_tokens": 3838752.0,
"step": 1855
},
{
"entropy": 5.83289942741394,
"epoch": 1.6896865061335755,
"grad_norm": 0.98828125,
"learning_rate": 0.0004918566831043238,
"loss": 5.8273,
"mean_token_accuracy": 0.15819755792617798,
"num_tokens": 3849664.0,
"step": 1860
},
{
"entropy": 5.893629026412964,
"epoch": 1.6942298955020445,
"grad_norm": 1.078125,
"learning_rate": 0.000491762191667251,
"loss": 5.8187,
"mean_token_accuracy": 0.1523476153612137,
"num_tokens": 3859053.0,
"step": 1865
},
{
"entropy": 5.814450216293335,
"epoch": 1.6987732848705135,
"grad_norm": 1.0625,
"learning_rate": 0.000491667165391018,
"loss": 5.8628,
"mean_token_accuracy": 0.14615362882614136,
"num_tokens": 3869654.0,
"step": 1870
},
{
"entropy": 5.983697986602783,
"epoch": 1.7033166742389823,
"grad_norm": 1.1875,
"learning_rate": 0.0004915716045100927,
"loss": 5.8583,
"mean_token_accuracy": 0.15104167461395263,
"num_tokens": 3879678.0,
"step": 1875
},
{
"entropy": 5.821736288070679,
"epoch": 1.707860063607451,
"grad_norm": 1.1640625,
"learning_rate": 0.0004914755092602621,
"loss": 5.7791,
"mean_token_accuracy": 0.15046799182891846,
"num_tokens": 3889950.0,
"step": 1880
},
{
"entropy": 5.858350610733032,
"epoch": 1.71240345297592,
"grad_norm": 1.03125,
"learning_rate": 0.0004913788798786314,
"loss": 5.862,
"mean_token_accuracy": 0.14945125356316566,
"num_tokens": 3900095.0,
"step": 1885
},
{
"entropy": 6.031531238555909,
"epoch": 1.716946842344389,
"grad_norm": 1.203125,
"learning_rate": 0.0004912817166036243,
"loss": 5.9188,
"mean_token_accuracy": 0.1444256231188774,
"num_tokens": 3911034.0,
"step": 1890
},
{
"entropy": 5.851213645935059,
"epoch": 1.7214902317128578,
"grad_norm": 1.0,
"learning_rate": 0.0004911840196749813,
"loss": 5.893,
"mean_token_accuracy": 0.14186890944838523,
"num_tokens": 3922219.0,
"step": 1895
},
{
"entropy": 5.897407054901123,
"epoch": 1.7260336210813265,
"grad_norm": 1.0078125,
"learning_rate": 0.0004910857893337602,
"loss": 5.8646,
"mean_token_accuracy": 0.14644375517964364,
"num_tokens": 3932235.0,
"step": 1900
},
{
"entropy": 5.836635684967041,
"epoch": 1.7305770104497955,
"grad_norm": 0.99609375,
"learning_rate": 0.0004909870258223342,
"loss": 5.7723,
"mean_token_accuracy": 0.1503476046025753,
"num_tokens": 3941845.0,
"step": 1905
},
{
"entropy": 5.936187887191773,
"epoch": 1.7351203998182645,
"grad_norm": 1.015625,
"learning_rate": 0.0004908877293843926,
"loss": 5.8882,
"mean_token_accuracy": 0.14496048614382745,
"num_tokens": 3953392.0,
"step": 1910
},
{
"entropy": 5.838795709609985,
"epoch": 1.7396637891867333,
"grad_norm": 1.0625,
"learning_rate": 0.0004907879002649395,
"loss": 5.728,
"mean_token_accuracy": 0.15308825224637984,
"num_tokens": 3963294.0,
"step": 1915
},
{
"entropy": 5.817688274383545,
"epoch": 1.744207178555202,
"grad_norm": 0.99609375,
"learning_rate": 0.0004906875387102935,
"loss": 5.8129,
"mean_token_accuracy": 0.14528712928295134,
"num_tokens": 3973660.0,
"step": 1920
},
{
"entropy": 5.895379447937012,
"epoch": 1.748750567923671,
"grad_norm": 1.09375,
"learning_rate": 0.0004905866449680864,
"loss": 5.8032,
"mean_token_accuracy": 0.1516289696097374,
"num_tokens": 3983587.0,
"step": 1925
},
{
"entropy": 5.89316234588623,
"epoch": 1.75329395729214,
"grad_norm": 0.96484375,
"learning_rate": 0.0004904852192872638,
"loss": 5.8634,
"mean_token_accuracy": 0.1431760974228382,
"num_tokens": 3994346.0,
"step": 1930
},
{
"entropy": 5.886792945861816,
"epoch": 1.757837346660609,
"grad_norm": 1.0625,
"learning_rate": 0.0004903832619180834,
"loss": 5.8433,
"mean_token_accuracy": 0.1472260557115078,
"num_tokens": 4005234.0,
"step": 1935
},
{
"entropy": 5.8153829097747805,
"epoch": 1.7623807360290777,
"grad_norm": 1.1171875,
"learning_rate": 0.0004902807731121147,
"loss": 5.7834,
"mean_token_accuracy": 0.15148564875125886,
"num_tokens": 4015542.0,
"step": 1940
},
{
"entropy": 5.8969179630279545,
"epoch": 1.7669241253975465,
"grad_norm": 1.125,
"learning_rate": 0.0004901777531222389,
"loss": 5.7768,
"mean_token_accuracy": 0.15424681231379508,
"num_tokens": 4024874.0,
"step": 1945
},
{
"entropy": 5.927392292022705,
"epoch": 1.7714675147660155,
"grad_norm": 1.0078125,
"learning_rate": 0.0004900742022026474,
"loss": 5.9311,
"mean_token_accuracy": 0.14847215712070466,
"num_tokens": 4035880.0,
"step": 1950
},
{
"entropy": 5.898624420166016,
"epoch": 1.7760109041344845,
"grad_norm": 1.0703125,
"learning_rate": 0.0004899701206088419,
"loss": 5.7973,
"mean_token_accuracy": 0.1557096555829048,
"num_tokens": 4046484.0,
"step": 1955
},
{
"entropy": 5.887908458709717,
"epoch": 1.7805542935029532,
"grad_norm": 1.03125,
"learning_rate": 0.0004898655085976333,
"loss": 5.8601,
"mean_token_accuracy": 0.15009021535515785,
"num_tokens": 4057037.0,
"step": 1960
},
{
"entropy": 5.8833253383636475,
"epoch": 1.785097682871422,
"grad_norm": 1.1171875,
"learning_rate": 0.0004897603664271413,
"loss": 5.942,
"mean_token_accuracy": 0.14604452252388,
"num_tokens": 4066169.0,
"step": 1965
},
{
"entropy": 5.934749364852905,
"epoch": 1.789641072239891,
"grad_norm": 1.03125,
"learning_rate": 0.000489654694356794,
"loss": 5.8669,
"mean_token_accuracy": 0.14416440352797508,
"num_tokens": 4076697.0,
"step": 1970
},
{
"entropy": 5.851240968704223,
"epoch": 1.79418446160836,
"grad_norm": 1.015625,
"learning_rate": 0.0004895484926473265,
"loss": 5.7342,
"mean_token_accuracy": 0.15425736904144288,
"num_tokens": 4087372.0,
"step": 1975
},
{
"entropy": 5.897654867172241,
"epoch": 1.7987278509768287,
"grad_norm": 0.984375,
"learning_rate": 0.000489441761560781,
"loss": 5.8422,
"mean_token_accuracy": 0.14519509822130203,
"num_tokens": 4097653.0,
"step": 1980
},
{
"entropy": 5.92055606842041,
"epoch": 1.8032712403452975,
"grad_norm": 0.98046875,
"learning_rate": 0.000489334501360506,
"loss": 5.7835,
"mean_token_accuracy": 0.15067849010229112,
"num_tokens": 4107381.0,
"step": 1985
},
{
"entropy": 5.825592565536499,
"epoch": 1.8078146297137665,
"grad_norm": 0.97265625,
"learning_rate": 0.0004892267123111551,
"loss": 5.8917,
"mean_token_accuracy": 0.14381059408187866,
"num_tokens": 4118762.0,
"step": 1990
},
{
"entropy": 5.882514333724975,
"epoch": 1.8123580190822355,
"grad_norm": 1.0859375,
"learning_rate": 0.0004891183946786873,
"loss": 5.7709,
"mean_token_accuracy": 0.15323199778795243,
"num_tokens": 4129011.0,
"step": 1995
},
{
"entropy": 5.882257509231567,
"epoch": 1.8169014084507042,
"grad_norm": 1.0859375,
"learning_rate": 0.0004890095487303655,
"loss": 5.8611,
"mean_token_accuracy": 0.14659103602170945,
"num_tokens": 4138975.0,
"step": 2000
},
{
"epoch": 1.8169014084507042,
"eval_entropy": 5.651273015488026,
"eval_loss": 6.038098335266113,
"eval_mean_token_accuracy": 0.14571751587439416,
"eval_num_tokens": 4138975.0,
"eval_runtime": 2.5857,
"eval_samples_per_second": 1328.446,
"eval_steps_per_second": 166.297,
"step": 2000
},
{
"entropy": 5.875406503677368,
"epoch": 1.821444797819173,
"grad_norm": 0.9609375,
"learning_rate": 0.0004889001747347562,
"loss": 5.9027,
"mean_token_accuracy": 0.14253540337085724,
"num_tokens": 4150164.0,
"step": 2005
},
{
"entropy": 5.816838121414184,
"epoch": 1.825988187187642,
"grad_norm": 1.2265625,
"learning_rate": 0.000488790272961729,
"loss": 5.7683,
"mean_token_accuracy": 0.15486635118722916,
"num_tokens": 4159872.0,
"step": 2010
},
{
"entropy": 5.900323152542114,
"epoch": 1.830531576556111,
"grad_norm": 1.1171875,
"learning_rate": 0.0004886798436824556,
"loss": 5.8448,
"mean_token_accuracy": 0.1453002981841564,
"num_tokens": 4170665.0,
"step": 2015
},
{
"entropy": 5.87586727142334,
"epoch": 1.8350749659245797,
"grad_norm": 1.0234375,
"learning_rate": 0.0004885688871694091,
"loss": 5.7278,
"mean_token_accuracy": 0.15901888012886048,
"num_tokens": 4180476.0,
"step": 2020
},
{
"entropy": 5.8813841342926025,
"epoch": 1.8396183552930485,
"grad_norm": 1.03125,
"learning_rate": 0.0004884574036963639,
"loss": 5.8659,
"mean_token_accuracy": 0.1450631484389305,
"num_tokens": 4190740.0,
"step": 2025
},
{
"entropy": 5.786100673675537,
"epoch": 1.8441617446615175,
"grad_norm": 1.0625,
"learning_rate": 0.0004883453935383941,
"loss": 5.7719,
"mean_token_accuracy": 0.15286731123924255,
"num_tokens": 4201504.0,
"step": 2030
},
{
"entropy": 5.7496833324432375,
"epoch": 1.8487051340299865,
"grad_norm": 1.1171875,
"learning_rate": 0.0004882328569718738,
"loss": 5.6932,
"mean_token_accuracy": 0.161198627948761,
"num_tokens": 4211614.0,
"step": 2035
},
{
"entropy": 5.850370025634765,
"epoch": 1.8532485233984552,
"grad_norm": 1.125,
"learning_rate": 0.0004881197942744758,
"loss": 5.8285,
"mean_token_accuracy": 0.15054793357849122,
"num_tokens": 4221141.0,
"step": 2040
},
{
"entropy": 5.840492868423462,
"epoch": 1.857791912766924,
"grad_norm": 1.2265625,
"learning_rate": 0.0004880062057251711,
"loss": 5.8705,
"mean_token_accuracy": 0.15193724259734154,
"num_tokens": 4231877.0,
"step": 2045
},
{
"entropy": 5.766042470932007,
"epoch": 1.862335302135393,
"grad_norm": 1.0078125,
"learning_rate": 0.000487892091604228,
"loss": 5.6829,
"mean_token_accuracy": 0.15766026824712753,
"num_tokens": 4242331.0,
"step": 2050
},
{
"entropy": 5.816211318969726,
"epoch": 1.866878691503862,
"grad_norm": 0.9609375,
"learning_rate": 0.00048777745219321186,
"loss": 5.7133,
"mean_token_accuracy": 0.14592440277338029,
"num_tokens": 4252406.0,
"step": 2055
},
{
"entropy": 5.820127296447754,
"epoch": 1.8714220808723308,
"grad_norm": 1.0078125,
"learning_rate": 0.00048766228777498405,
"loss": 5.7833,
"mean_token_accuracy": 0.14930634200572968,
"num_tokens": 4263036.0,
"step": 2060
},
{
"entropy": 5.848497486114502,
"epoch": 1.8759654702407995,
"grad_norm": 1.1484375,
"learning_rate": 0.00048754659863370125,
"loss": 5.7687,
"mean_token_accuracy": 0.15771324187517166,
"num_tokens": 4273032.0,
"step": 2065
},
{
"entropy": 5.815657758712769,
"epoch": 1.8805088596092685,
"grad_norm": 1.0234375,
"learning_rate": 0.00048743038505481503,
"loss": 5.8344,
"mean_token_accuracy": 0.15047525838017464,
"num_tokens": 4283360.0,
"step": 2070
},
{
"entropy": 5.8192830085754395,
"epoch": 1.8850522489777375,
"grad_norm": 1.0546875,
"learning_rate": 0.0004873136473250708,
"loss": 5.7704,
"mean_token_accuracy": 0.15761347264051437,
"num_tokens": 4292623.0,
"step": 2075
},
{
"entropy": 5.8200572490692135,
"epoch": 1.8895956383462063,
"grad_norm": 1.0625,
"learning_rate": 0.00048719638573250715,
"loss": 5.7837,
"mean_token_accuracy": 0.15845565497875214,
"num_tokens": 4302817.0,
"step": 2080
},
{
"entropy": 5.900278472900391,
"epoch": 1.894139027714675,
"grad_norm": 1.015625,
"learning_rate": 0.0004870786005664555,
"loss": 5.7987,
"mean_token_accuracy": 0.1524748384952545,
"num_tokens": 4313049.0,
"step": 2085
},
{
"entropy": 5.7548986911773685,
"epoch": 1.898682417083144,
"grad_norm": 1.1484375,
"learning_rate": 0.0004869602921175392,
"loss": 5.6845,
"mean_token_accuracy": 0.16200295984745025,
"num_tokens": 4322873.0,
"step": 2090
},
{
"entropy": 5.775264549255371,
"epoch": 1.903225806451613,
"grad_norm": 1.046875,
"learning_rate": 0.0004868414606776724,
"loss": 5.8419,
"mean_token_accuracy": 0.14530435353517532,
"num_tokens": 4333339.0,
"step": 2095
},
{
"entropy": 5.864491891860962,
"epoch": 1.9077691958200818,
"grad_norm": 0.98046875,
"learning_rate": 0.00048672210654005996,
"loss": 5.8019,
"mean_token_accuracy": 0.15234252363443374,
"num_tokens": 4344917.0,
"step": 2100
},
{
"entropy": 5.684874629974365,
"epoch": 1.9123125851885505,
"grad_norm": 1.0546875,
"learning_rate": 0.00048660222999919616,
"loss": 5.6508,
"mean_token_accuracy": 0.15912348330020903,
"num_tokens": 4354140.0,
"step": 2105
},
{
"entropy": 5.866669178009033,
"epoch": 1.9168559745570195,
"grad_norm": 1.078125,
"learning_rate": 0.0004864818313508647,
"loss": 5.8984,
"mean_token_accuracy": 0.14722514897584915,
"num_tokens": 4364976.0,
"step": 2110
},
{
"entropy": 5.839132165908813,
"epoch": 1.9213993639254885,
"grad_norm": 1.0625,
"learning_rate": 0.0004863609108921372,
"loss": 5.6634,
"mean_token_accuracy": 0.1591766059398651,
"num_tokens": 4376423.0,
"step": 2115
},
{
"entropy": 5.830103158950806,
"epoch": 1.9259427532939573,
"grad_norm": 1.0703125,
"learning_rate": 0.0004862394689213728,
"loss": 5.8093,
"mean_token_accuracy": 0.15118083506822586,
"num_tokens": 4387865.0,
"step": 2120
},
{
"entropy": 5.763162851333618,
"epoch": 1.930486142662426,
"grad_norm": 0.99609375,
"learning_rate": 0.0004861175057382175,
"loss": 5.7951,
"mean_token_accuracy": 0.15389713793992996,
"num_tokens": 4398275.0,
"step": 2125
},
{
"entropy": 5.88576922416687,
"epoch": 1.935029532030895,
"grad_norm": 1.0703125,
"learning_rate": 0.0004859950216436034,
"loss": 5.6982,
"mean_token_accuracy": 0.15084194988012314,
"num_tokens": 4407396.0,
"step": 2130
},
{
"entropy": 5.9557061195373535,
"epoch": 1.939572921399364,
"grad_norm": 0.94140625,
"learning_rate": 0.0004858720169397478,
"loss": 5.9702,
"mean_token_accuracy": 0.14486352875828742,
"num_tokens": 4419682.0,
"step": 2135
},
{
"entropy": 5.806124067306518,
"epoch": 1.9441163107678328,
"grad_norm": 1.1484375,
"learning_rate": 0.0004857484919301525,
"loss": 5.8464,
"mean_token_accuracy": 0.1488577887415886,
"num_tokens": 4430590.0,
"step": 2140
},
{
"entropy": 5.843648481369018,
"epoch": 1.9486597001363015,
"grad_norm": 0.953125,
"learning_rate": 0.0004856244469196034,
"loss": 5.7546,
"mean_token_accuracy": 0.16025863587856293,
"num_tokens": 4441979.0,
"step": 2145
},
{
"entropy": 5.830633974075317,
"epoch": 1.9532030895047705,
"grad_norm": 1.0546875,
"learning_rate": 0.000485499882214169,
"loss": 5.7642,
"mean_token_accuracy": 0.15131544172763825,
"num_tokens": 4451975.0,
"step": 2150
},
{
"entropy": 5.752586364746094,
"epoch": 1.9577464788732395,
"grad_norm": 1.0625,
"learning_rate": 0.0004853747981212003,
"loss": 5.7148,
"mean_token_accuracy": 0.1495148092508316,
"num_tokens": 4462778.0,
"step": 2155
},
{
"entropy": 5.817465019226074,
"epoch": 1.9622898682417085,
"grad_norm": 1.046875,
"learning_rate": 0.0004852491949493301,
"loss": 5.7616,
"mean_token_accuracy": 0.16084688007831574,
"num_tokens": 4472461.0,
"step": 2160
},
{
"entropy": 5.822881174087525,
"epoch": 1.9668332576101772,
"grad_norm": 1.0546875,
"learning_rate": 0.0004851230730084716,
"loss": 5.7102,
"mean_token_accuracy": 0.1580464079976082,
"num_tokens": 4482507.0,
"step": 2165
},
{
"entropy": 5.805157899856567,
"epoch": 1.971376646978646,
"grad_norm": 1.046875,
"learning_rate": 0.00048499643260981816,
"loss": 5.7338,
"mean_token_accuracy": 0.15504981875419616,
"num_tokens": 4492601.0,
"step": 2170
},
{
"entropy": 5.831399202346802,
"epoch": 1.975920036347115,
"grad_norm": 1.015625,
"learning_rate": 0.0004848692740658424,
"loss": 5.8482,
"mean_token_accuracy": 0.15246382355690002,
"num_tokens": 4503696.0,
"step": 2175
},
{
"entropy": 5.830595779418945,
"epoch": 1.980463425715584,
"grad_norm": 1.1015625,
"learning_rate": 0.0004847415976902954,
"loss": 5.7321,
"mean_token_accuracy": 0.15554610043764114,
"num_tokens": 4513649.0,
"step": 2180
},
{
"entropy": 5.793164587020874,
"epoch": 1.9850068150840527,
"grad_norm": 1.1328125,
"learning_rate": 0.00048461340379820587,
"loss": 5.7705,
"mean_token_accuracy": 0.15155881643295288,
"num_tokens": 4524054.0,
"step": 2185
},
{
"entropy": 5.879954433441162,
"epoch": 1.9895502044525215,
"grad_norm": 1.03125,
"learning_rate": 0.00048448469270587945,
"loss": 5.7402,
"mean_token_accuracy": 0.14864474684000015,
"num_tokens": 4535549.0,
"step": 2190
},
{
"entropy": 5.758612108230591,
"epoch": 1.9940935938209905,
"grad_norm": 1.03125,
"learning_rate": 0.00048435546473089814,
"loss": 5.7742,
"mean_token_accuracy": 0.14887889772653579,
"num_tokens": 4545654.0,
"step": 2195
},
{
"entropy": 5.841834640502929,
"epoch": 1.9986369831894595,
"grad_norm": 1.0859375,
"learning_rate": 0.00048422572019211896,
"loss": 5.7815,
"mean_token_accuracy": 0.1565988466143608,
"num_tokens": 4556230.0,
"step": 2200
},
{
"entropy": 5.775571240319146,
"epoch": 2.0027260336210815,
"grad_norm": 0.94921875,
"learning_rate": 0.00048409545940967375,
"loss": 5.5241,
"mean_token_accuracy": 0.15879998438888127,
"num_tokens": 4564486.0,
"step": 2205
},
{
"entropy": 5.746746397018432,
"epoch": 2.00726942298955,
"grad_norm": 1.0859375,
"learning_rate": 0.000483964682704968,
"loss": 5.4081,
"mean_token_accuracy": 0.16296772062778472,
"num_tokens": 4575026.0,
"step": 2210
},
{
"entropy": 5.681377935409546,
"epoch": 2.011812812358019,
"grad_norm": 1.0546875,
"learning_rate": 0.00048383339040068024,
"loss": 5.4454,
"mean_token_accuracy": 0.16175178587436675,
"num_tokens": 4585902.0,
"step": 2215
},
{
"entropy": 5.86534948348999,
"epoch": 2.016356201726488,
"grad_norm": 1.0703125,
"learning_rate": 0.0004837015828207612,
"loss": 5.5325,
"mean_token_accuracy": 0.16216056123375894,
"num_tokens": 4597609.0,
"step": 2220
},
{
"entropy": 5.694772672653198,
"epoch": 2.020899591094957,
"grad_norm": 1.046875,
"learning_rate": 0.0004835692602904329,
"loss": 5.4215,
"mean_token_accuracy": 0.16129270941019058,
"num_tokens": 4607531.0,
"step": 2225
},
{
"entropy": 5.687169456481934,
"epoch": 2.0254429804634255,
"grad_norm": 1.1171875,
"learning_rate": 0.0004834364231361882,
"loss": 5.442,
"mean_token_accuracy": 0.1637440651655197,
"num_tokens": 4618826.0,
"step": 2230
},
{
"entropy": 5.790351390838623,
"epoch": 2.0299863698318945,
"grad_norm": 1.1015625,
"learning_rate": 0.0004833030716857894,
"loss": 5.451,
"mean_token_accuracy": 0.1670396000146866,
"num_tokens": 4629526.0,
"step": 2235
},
{
"entropy": 5.739242124557495,
"epoch": 2.0345297592003635,
"grad_norm": 1.0859375,
"learning_rate": 0.00048316920626826807,
"loss": 5.4931,
"mean_token_accuracy": 0.16078836470842361,
"num_tokens": 4639930.0,
"step": 2240
},
{
"entropy": 5.695904731750488,
"epoch": 2.0390731485688325,
"grad_norm": 1.1640625,
"learning_rate": 0.00048303482721392363,
"loss": 5.3821,
"mean_token_accuracy": 0.17241068482398986,
"num_tokens": 4649661.0,
"step": 2245
},
{
"entropy": 5.665669393539429,
"epoch": 2.043616537937301,
"grad_norm": 0.97265625,
"learning_rate": 0.00048289993485432325,
"loss": 5.3758,
"mean_token_accuracy": 0.1731106236577034,
"num_tokens": 4660178.0,
"step": 2250
},
{
"entropy": 5.749753999710083,
"epoch": 2.04815992730577,
"grad_norm": 1.015625,
"learning_rate": 0.00048276452952230036,
"loss": 5.4248,
"mean_token_accuracy": 0.1656309887766838,
"num_tokens": 4670417.0,
"step": 2255
},
{
"entropy": 5.645382213592529,
"epoch": 2.052703316674239,
"grad_norm": 1.125,
"learning_rate": 0.00048262861155195407,
"loss": 5.368,
"mean_token_accuracy": 0.17645173668861389,
"num_tokens": 4680221.0,
"step": 2260
},
{
"entropy": 5.680039978027343,
"epoch": 2.057246706042708,
"grad_norm": 1.0859375,
"learning_rate": 0.0004824921812786486,
"loss": 5.4311,
"mean_token_accuracy": 0.1622917652130127,
"num_tokens": 4691136.0,
"step": 2265
},
{
"entropy": 5.602982091903686,
"epoch": 2.0617900954111765,
"grad_norm": 1.0078125,
"learning_rate": 0.000482355239039012,
"loss": 5.2998,
"mean_token_accuracy": 0.17576343119144439,
"num_tokens": 4700879.0,
"step": 2270
},
{
"entropy": 5.68736515045166,
"epoch": 2.0663334847796455,
"grad_norm": 1.0703125,
"learning_rate": 0.0004822177851709356,
"loss": 5.4684,
"mean_token_accuracy": 0.16156385689973832,
"num_tokens": 4711477.0,
"step": 2275
},
{
"entropy": 5.692600727081299,
"epoch": 2.0708768741481145,
"grad_norm": 1.15625,
"learning_rate": 0.00048207982001357327,
"loss": 5.4468,
"mean_token_accuracy": 0.16822416931390763,
"num_tokens": 4721973.0,
"step": 2280
},
{
"entropy": 5.651112842559814,
"epoch": 2.0754202635165835,
"grad_norm": 1.1640625,
"learning_rate": 0.0004819413439073402,
"loss": 5.4316,
"mean_token_accuracy": 0.16672107428312302,
"num_tokens": 4732425.0,
"step": 2285
},
{
"entropy": 5.657487678527832,
"epoch": 2.079963652885052,
"grad_norm": 1.03125,
"learning_rate": 0.0004818023571939125,
"loss": 5.3986,
"mean_token_accuracy": 0.17188587337732314,
"num_tokens": 4743377.0,
"step": 2290
},
{
"entropy": 5.722143316268921,
"epoch": 2.084507042253521,
"grad_norm": 1.1328125,
"learning_rate": 0.000481662860216226,
"loss": 5.4737,
"mean_token_accuracy": 0.16359636634588243,
"num_tokens": 4753534.0,
"step": 2295
},
{
"entropy": 5.649228477478028,
"epoch": 2.08905043162199,
"grad_norm": 1.09375,
"learning_rate": 0.00048152285331847574,
"loss": 5.4316,
"mean_token_accuracy": 0.16490243673324584,
"num_tokens": 4763159.0,
"step": 2300
},
{
"entropy": 5.640550661087036,
"epoch": 2.093593820990459,
"grad_norm": 1.109375,
"learning_rate": 0.0004813823368461148,
"loss": 5.4059,
"mean_token_accuracy": 0.173206128180027,
"num_tokens": 4773497.0,
"step": 2305
},
{
"entropy": 5.650098562240601,
"epoch": 2.098137210358928,
"grad_norm": 1.09375,
"learning_rate": 0.0004812413111458534,
"loss": 5.508,
"mean_token_accuracy": 0.1677454337477684,
"num_tokens": 4784291.0,
"step": 2310
},
{
"entropy": 5.664199924468994,
"epoch": 2.1026805997273965,
"grad_norm": 1.125,
"learning_rate": 0.0004810997765656587,
"loss": 5.4594,
"mean_token_accuracy": 0.16584839820861816,
"num_tokens": 4794762.0,
"step": 2315
},
{
"entropy": 5.764793395996094,
"epoch": 2.1072239890958655,
"grad_norm": 1.09375,
"learning_rate": 0.00048095773345475294,
"loss": 5.5542,
"mean_token_accuracy": 0.16471297442913055,
"num_tokens": 4804449.0,
"step": 2320
},
{
"entropy": 5.695501470565796,
"epoch": 2.1117673784643345,
"grad_norm": 1.234375,
"learning_rate": 0.00048081518216361367,
"loss": 5.4193,
"mean_token_accuracy": 0.16973745822906494,
"num_tokens": 4814172.0,
"step": 2325
},
{
"entropy": 5.684962558746338,
"epoch": 2.1163107678328035,
"grad_norm": 1.078125,
"learning_rate": 0.0004806721230439719,
"loss": 5.5159,
"mean_token_accuracy": 0.16350335478782654,
"num_tokens": 4824909.0,
"step": 2330
},
{
"entropy": 5.735642004013061,
"epoch": 2.120854157201272,
"grad_norm": 1.1484375,
"learning_rate": 0.0004805285564488117,
"loss": 5.5421,
"mean_token_accuracy": 0.16848451942205428,
"num_tokens": 4834546.0,
"step": 2335
},
{
"entropy": 5.663049507141113,
"epoch": 2.125397546569741,
"grad_norm": 1.1328125,
"learning_rate": 0.0004803844827323696,
"loss": 5.4604,
"mean_token_accuracy": 0.16767446398735047,
"num_tokens": 4844809.0,
"step": 2340
},
{
"entropy": 5.663071823120117,
"epoch": 2.12994093593821,
"grad_norm": 1.109375,
"learning_rate": 0.0004802399022501329,
"loss": 5.4448,
"mean_token_accuracy": 0.16737743467092514,
"num_tokens": 4855883.0,
"step": 2345
},
{
"entropy": 5.629608297348023,
"epoch": 2.134484325306679,
"grad_norm": 1.1796875,
"learning_rate": 0.0004800948153588397,
"loss": 5.4765,
"mean_token_accuracy": 0.1688181445002556,
"num_tokens": 4865936.0,
"step": 2350
},
{
"entropy": 5.672809743881226,
"epoch": 2.1390277146751475,
"grad_norm": 1.0859375,
"learning_rate": 0.00047994922241647746,
"loss": 5.3901,
"mean_token_accuracy": 0.17158838659524916,
"num_tokens": 4876621.0,
"step": 2355
},
{
"entropy": 5.512680578231811,
"epoch": 2.1435711040436165,
"grad_norm": 1.1640625,
"learning_rate": 0.00047980312378228244,
"loss": 5.3787,
"mean_token_accuracy": 0.1638313740491867,
"num_tokens": 4886315.0,
"step": 2360
},
{
"entropy": 5.69263334274292,
"epoch": 2.1481144934120855,
"grad_norm": 1.2578125,
"learning_rate": 0.00047965651981673823,
"loss": 5.4641,
"mean_token_accuracy": 0.1665160223841667,
"num_tokens": 4897214.0,
"step": 2365
},
{
"entropy": 5.687677669525146,
"epoch": 2.1526578827805545,
"grad_norm": 1.0703125,
"learning_rate": 0.00047950941088157576,
"loss": 5.4515,
"mean_token_accuracy": 0.16270851343870163,
"num_tokens": 4907360.0,
"step": 2370
},
{
"entropy": 5.614701843261718,
"epoch": 2.157201272149023,
"grad_norm": 1.140625,
"learning_rate": 0.0004793617973397715,
"loss": 5.4225,
"mean_token_accuracy": 0.16925650984048843,
"num_tokens": 4917591.0,
"step": 2375
},
{
"entropy": 5.670958375930786,
"epoch": 2.161744661517492,
"grad_norm": 1.0546875,
"learning_rate": 0.00047921367955554745,
"loss": 5.5176,
"mean_token_accuracy": 0.15314990431070327,
"num_tokens": 4928075.0,
"step": 2380
},
{
"entropy": 5.687479448318482,
"epoch": 2.166288050885961,
"grad_norm": 1.03125,
"learning_rate": 0.0004790650578943693,
"loss": 5.3983,
"mean_token_accuracy": 0.1678956925868988,
"num_tokens": 4937898.0,
"step": 2385
},
{
"entropy": 5.614485836029052,
"epoch": 2.17083144025443,
"grad_norm": 1.0234375,
"learning_rate": 0.00047891593272294636,
"loss": 5.4493,
"mean_token_accuracy": 0.16083518117666246,
"num_tokens": 4948152.0,
"step": 2390
},
{
"entropy": 5.693413019180298,
"epoch": 2.1753748296228985,
"grad_norm": 1.2890625,
"learning_rate": 0.00047876630440923,
"loss": 5.4518,
"mean_token_accuracy": 0.16560257524251937,
"num_tokens": 4957973.0,
"step": 2395
},
{
"entropy": 5.52451524734497,
"epoch": 2.1799182189913675,
"grad_norm": 1.09375,
"learning_rate": 0.0004786161733224134,
"loss": 5.4124,
"mean_token_accuracy": 0.16864717304706572,
"num_tokens": 4967375.0,
"step": 2400
},
{
"entropy": 5.673086023330688,
"epoch": 2.1844616083598365,
"grad_norm": 1.03125,
"learning_rate": 0.00047846553983293006,
"loss": 5.4083,
"mean_token_accuracy": 0.16841582357883453,
"num_tokens": 4977286.0,
"step": 2405
},
{
"entropy": 5.5908843040466305,
"epoch": 2.1890049977283055,
"grad_norm": 1.1171875,
"learning_rate": 0.00047831440431245307,
"loss": 5.4883,
"mean_token_accuracy": 0.16461575627326966,
"num_tokens": 4987930.0,
"step": 2410
},
{
"entropy": 5.722489643096924,
"epoch": 2.193548387096774,
"grad_norm": 1.09375,
"learning_rate": 0.00047816276713389424,
"loss": 5.5292,
"mean_token_accuracy": 0.16635961681604386,
"num_tokens": 4998707.0,
"step": 2415
},
{
"entropy": 5.581833744049073,
"epoch": 2.198091776465243,
"grad_norm": 1.0625,
"learning_rate": 0.0004780106286714034,
"loss": 5.3915,
"mean_token_accuracy": 0.17737244814634323,
"num_tokens": 5008540.0,
"step": 2420
},
{
"entropy": 5.588280296325683,
"epoch": 2.202635165833712,
"grad_norm": 1.1171875,
"learning_rate": 0.000477857989300367,
"loss": 5.4584,
"mean_token_accuracy": 0.16891718804836273,
"num_tokens": 5018199.0,
"step": 2425
},
{
"entropy": 5.707861995697021,
"epoch": 2.207178555202181,
"grad_norm": 1.0703125,
"learning_rate": 0.00047770484939740757,
"loss": 5.462,
"mean_token_accuracy": 0.16908059269189835,
"num_tokens": 5028251.0,
"step": 2430
},
{
"entropy": 5.588924741744995,
"epoch": 2.2117219445706495,
"grad_norm": 1.1015625,
"learning_rate": 0.0004775512093403824,
"loss": 5.4748,
"mean_token_accuracy": 0.15842932015657424,
"num_tokens": 5038290.0,
"step": 2435
},
{
"entropy": 5.648030948638916,
"epoch": 2.2162653339391185,
"grad_norm": 1.2578125,
"learning_rate": 0.00047739706950838327,
"loss": 5.4084,
"mean_token_accuracy": 0.16693121790885926,
"num_tokens": 5048337.0,
"step": 2440
},
{
"entropy": 5.624004697799682,
"epoch": 2.2208087233075875,
"grad_norm": 1.0859375,
"learning_rate": 0.00047724243028173484,
"loss": 5.4686,
"mean_token_accuracy": 0.16390100866556168,
"num_tokens": 5058533.0,
"step": 2445
},
{
"entropy": 5.665524578094482,
"epoch": 2.2253521126760565,
"grad_norm": 1.1953125,
"learning_rate": 0.00047708729204199406,
"loss": 5.4808,
"mean_token_accuracy": 0.16627213954925538,
"num_tokens": 5068556.0,
"step": 2450
},
{
"entropy": 5.591360998153687,
"epoch": 2.229895502044525,
"grad_norm": 1.09375,
"learning_rate": 0.000476931655171949,
"loss": 5.5497,
"mean_token_accuracy": 0.1624269276857376,
"num_tokens": 5079317.0,
"step": 2455
},
{
"entropy": 5.7417370796203615,
"epoch": 2.234438891412994,
"grad_norm": 1.03125,
"learning_rate": 0.0004767755200556183,
"loss": 5.5275,
"mean_token_accuracy": 0.16499654054641724,
"num_tokens": 5089544.0,
"step": 2460
},
{
"entropy": 5.612682771682739,
"epoch": 2.238982280781463,
"grad_norm": 1.1640625,
"learning_rate": 0.0004766188870782497,
"loss": 5.4947,
"mean_token_accuracy": 0.16531910747289658,
"num_tokens": 5099491.0,
"step": 2465
},
{
"entropy": 5.708553075790405,
"epoch": 2.243525670149932,
"grad_norm": 1.078125,
"learning_rate": 0.00047646175662631967,
"loss": 5.5036,
"mean_token_accuracy": 0.16413073986768723,
"num_tokens": 5109523.0,
"step": 2470
},
{
"entropy": 5.644906759262085,
"epoch": 2.2480690595184005,
"grad_norm": 1.109375,
"learning_rate": 0.00047630412908753183,
"loss": 5.4586,
"mean_token_accuracy": 0.17124137431383132,
"num_tokens": 5119940.0,
"step": 2475
},
{
"entropy": 5.602259349822998,
"epoch": 2.2526124488868695,
"grad_norm": 1.1015625,
"learning_rate": 0.00047614600485081657,
"loss": 5.423,
"mean_token_accuracy": 0.17469318360090255,
"num_tokens": 5129789.0,
"step": 2480
},
{
"entropy": 5.621613025665283,
"epoch": 2.2571558382553385,
"grad_norm": 1.09375,
"learning_rate": 0.00047598738430632963,
"loss": 5.4503,
"mean_token_accuracy": 0.1608627885580063,
"num_tokens": 5139570.0,
"step": 2485
},
{
"entropy": 5.5770516872406,
"epoch": 2.2616992276238075,
"grad_norm": 1.046875,
"learning_rate": 0.00047582826784545146,
"loss": 5.4174,
"mean_token_accuracy": 0.16260818243026734,
"num_tokens": 5150526.0,
"step": 2490
},
{
"entropy": 5.608628606796264,
"epoch": 2.266242616992276,
"grad_norm": 1.0546875,
"learning_rate": 0.00047566865586078615,
"loss": 5.379,
"mean_token_accuracy": 0.16596596240997313,
"num_tokens": 5160895.0,
"step": 2495
},
{
"entropy": 5.591861248016357,
"epoch": 2.270786006360745,
"grad_norm": 1.0703125,
"learning_rate": 0.00047550854874616047,
"loss": 5.369,
"mean_token_accuracy": 0.16774842292070388,
"num_tokens": 5171054.0,
"step": 2500
},
{
"epoch": 2.270786006360745,
"eval_entropy": 5.522149945414344,
"eval_loss": 5.9590277671813965,
"eval_mean_token_accuracy": 0.15115891338087792,
"eval_num_tokens": 5171054.0,
"eval_runtime": 2.765,
"eval_samples_per_second": 1242.33,
"eval_steps_per_second": 155.517,
"step": 2500
},
{
"entropy": 5.6621589183807375,
"epoch": 2.275329395729214,
"grad_norm": 1.0703125,
"learning_rate": 0.00047534794689662257,
"loss": 5.5235,
"mean_token_accuracy": 0.1584127202630043,
"num_tokens": 5180870.0,
"step": 2505
},
{
"entropy": 5.65887336730957,
"epoch": 2.279872785097683,
"grad_norm": 1.1484375,
"learning_rate": 0.0004751868507084418,
"loss": 5.4674,
"mean_token_accuracy": 0.16400336474180222,
"num_tokens": 5190388.0,
"step": 2510
},
{
"entropy": 5.6240010261535645,
"epoch": 2.284416174466152,
"grad_norm": 1.0625,
"learning_rate": 0.0004750252605791068,
"loss": 5.444,
"mean_token_accuracy": 0.15800251960754394,
"num_tokens": 5200582.0,
"step": 2515
},
{
"entropy": 5.635478162765503,
"epoch": 2.2889595638346205,
"grad_norm": 1.109375,
"learning_rate": 0.0004748631769073254,
"loss": 5.4065,
"mean_token_accuracy": 0.16952998340129852,
"num_tokens": 5211079.0,
"step": 2520
},
{
"entropy": 5.630209636688233,
"epoch": 2.2935029532030895,
"grad_norm": 1.1328125,
"learning_rate": 0.0004747006000930227,
"loss": 5.5584,
"mean_token_accuracy": 0.15611759200692177,
"num_tokens": 5221481.0,
"step": 2525
},
{
"entropy": 5.721398210525512,
"epoch": 2.2980463425715585,
"grad_norm": 1.09375,
"learning_rate": 0.0004745375305373409,
"loss": 5.4259,
"mean_token_accuracy": 0.17318421304225923,
"num_tokens": 5231387.0,
"step": 2530
},
{
"entropy": 5.52024474143982,
"epoch": 2.302589731940027,
"grad_norm": 1.0703125,
"learning_rate": 0.00047437396864263804,
"loss": 5.388,
"mean_token_accuracy": 0.17229016572237016,
"num_tokens": 5240758.0,
"step": 2535
},
{
"entropy": 5.6235308170318605,
"epoch": 2.307133121308496,
"grad_norm": 1.09375,
"learning_rate": 0.0004742099148124866,
"loss": 5.3977,
"mean_token_accuracy": 0.16649569869041442,
"num_tokens": 5250839.0,
"step": 2540
},
{
"entropy": 5.678149890899658,
"epoch": 2.311676510676965,
"grad_norm": 1.078125,
"learning_rate": 0.0004740453694516734,
"loss": 5.5256,
"mean_token_accuracy": 0.163419970870018,
"num_tokens": 5261554.0,
"step": 2545
},
{
"entropy": 5.608114862442017,
"epoch": 2.316219900045434,
"grad_norm": 1.0546875,
"learning_rate": 0.00047388033296619746,
"loss": 5.4785,
"mean_token_accuracy": 0.1661337912082672,
"num_tokens": 5272422.0,
"step": 2550
},
{
"entropy": 5.620512676239014,
"epoch": 2.320763289413903,
"grad_norm": 1.109375,
"learning_rate": 0.0004737148057632701,
"loss": 5.4149,
"mean_token_accuracy": 0.16315267756581306,
"num_tokens": 5282522.0,
"step": 2555
},
{
"entropy": 5.597893285751343,
"epoch": 2.3253066787823715,
"grad_norm": 1.0625,
"learning_rate": 0.0004735487882513131,
"loss": 5.4357,
"mean_token_accuracy": 0.16960459649562837,
"num_tokens": 5293100.0,
"step": 2560
},
{
"entropy": 5.668226861953736,
"epoch": 2.3298500681508405,
"grad_norm": 1.1875,
"learning_rate": 0.0004733822808399583,
"loss": 5.5032,
"mean_token_accuracy": 0.1607452541589737,
"num_tokens": 5303103.0,
"step": 2565
},
{
"entropy": 5.633330392837524,
"epoch": 2.3343934575193095,
"grad_norm": 1.0390625,
"learning_rate": 0.00047321528394004615,
"loss": 5.479,
"mean_token_accuracy": 0.16839675307273866,
"num_tokens": 5313765.0,
"step": 2570
},
{
"entropy": 5.721703147888183,
"epoch": 2.3389368468877785,
"grad_norm": 1.2265625,
"learning_rate": 0.00047304779796362485,
"loss": 5.5333,
"mean_token_accuracy": 0.16169264763593674,
"num_tokens": 5324816.0,
"step": 2575
},
{
"entropy": 5.578091144561768,
"epoch": 2.343480236256247,
"grad_norm": 1.078125,
"learning_rate": 0.00047287982332394943,
"loss": 5.4357,
"mean_token_accuracy": 0.16642868220806123,
"num_tokens": 5334882.0,
"step": 2580
},
{
"entropy": 5.60108585357666,
"epoch": 2.348023625624716,
"grad_norm": 1.1953125,
"learning_rate": 0.00047271136043548057,
"loss": 5.4373,
"mean_token_accuracy": 0.16962540596723558,
"num_tokens": 5344623.0,
"step": 2585
},
{
"entropy": 5.604558706283569,
"epoch": 2.352567014993185,
"grad_norm": 1.015625,
"learning_rate": 0.0004725424097138838,
"loss": 5.4615,
"mean_token_accuracy": 0.16610558331012726,
"num_tokens": 5355786.0,
"step": 2590
},
{
"entropy": 5.600728702545166,
"epoch": 2.357110404361654,
"grad_norm": 1.1171875,
"learning_rate": 0.00047237297157602816,
"loss": 5.4574,
"mean_token_accuracy": 0.16407135128974915,
"num_tokens": 5366139.0,
"step": 2595
},
{
"entropy": 5.722113180160522,
"epoch": 2.3616537937301225,
"grad_norm": 1.046875,
"learning_rate": 0.0004722030464399854,
"loss": 5.5476,
"mean_token_accuracy": 0.16129898875951768,
"num_tokens": 5376890.0,
"step": 2600
},
{
"entropy": 5.568664264678955,
"epoch": 2.3661971830985915,
"grad_norm": 1.1328125,
"learning_rate": 0.000472032634725029,
"loss": 5.4426,
"mean_token_accuracy": 0.16650206446647645,
"num_tokens": 5386602.0,
"step": 2605
},
{
"entropy": 5.636752033233643,
"epoch": 2.3707405724670605,
"grad_norm": 1.15625,
"learning_rate": 0.00047186173685163273,
"loss": 5.5201,
"mean_token_accuracy": 0.1646490141749382,
"num_tokens": 5397749.0,
"step": 2610
},
{
"entropy": 5.642663669586182,
"epoch": 2.3752839618355295,
"grad_norm": 1.125,
"learning_rate": 0.00047169035324147023,
"loss": 5.4939,
"mean_token_accuracy": 0.15729261189699173,
"num_tokens": 5407666.0,
"step": 2615
},
{
"entropy": 5.60233244895935,
"epoch": 2.379827351203998,
"grad_norm": 1.0625,
"learning_rate": 0.00047151848431741363,
"loss": 5.5776,
"mean_token_accuracy": 0.15825712233781813,
"num_tokens": 5418668.0,
"step": 2620
},
{
"entropy": 5.644424819946289,
"epoch": 2.384370740572467,
"grad_norm": 1.15625,
"learning_rate": 0.0004713461305035322,
"loss": 5.4285,
"mean_token_accuracy": 0.17103398591279984,
"num_tokens": 5429098.0,
"step": 2625
},
{
"entropy": 5.533689212799072,
"epoch": 2.388914129940936,
"grad_norm": 1.1796875,
"learning_rate": 0.000471173292225092,
"loss": 5.3722,
"mean_token_accuracy": 0.17129211127758026,
"num_tokens": 5438997.0,
"step": 2630
},
{
"entropy": 5.592352867126465,
"epoch": 2.393457519309405,
"grad_norm": 1.1953125,
"learning_rate": 0.00047099996990855424,
"loss": 5.4502,
"mean_token_accuracy": 0.16401641368865966,
"num_tokens": 5449315.0,
"step": 2635
},
{
"entropy": 5.580306148529052,
"epoch": 2.3980009086778735,
"grad_norm": 1.1171875,
"learning_rate": 0.00047082616398157456,
"loss": 5.4666,
"mean_token_accuracy": 0.16631554216146469,
"num_tokens": 5460523.0,
"step": 2640
},
{
"entropy": 5.664644241333008,
"epoch": 2.4025442980463425,
"grad_norm": 1.2265625,
"learning_rate": 0.0004706518748730017,
"loss": 5.5217,
"mean_token_accuracy": 0.1689920172095299,
"num_tokens": 5470012.0,
"step": 2645
},
{
"entropy": 5.572385692596436,
"epoch": 2.4070876874148115,
"grad_norm": 1.203125,
"learning_rate": 0.0004704771030128768,
"loss": 5.4625,
"mean_token_accuracy": 0.17034649550914766,
"num_tokens": 5480364.0,
"step": 2650
},
{
"entropy": 5.680575656890869,
"epoch": 2.4116310767832805,
"grad_norm": 1.1796875,
"learning_rate": 0.00047030184883243203,
"loss": 5.5965,
"mean_token_accuracy": 0.1588437870144844,
"num_tokens": 5492138.0,
"step": 2655
},
{
"entropy": 5.6193524360656735,
"epoch": 2.416174466151749,
"grad_norm": 1.15625,
"learning_rate": 0.00047012611276408964,
"loss": 5.359,
"mean_token_accuracy": 0.17521025538444518,
"num_tokens": 5501935.0,
"step": 2660
},
{
"entropy": 5.554859256744384,
"epoch": 2.420717855520218,
"grad_norm": 1.109375,
"learning_rate": 0.0004699498952414609,
"loss": 5.5597,
"mean_token_accuracy": 0.161118221282959,
"num_tokens": 5513993.0,
"step": 2665
},
{
"entropy": 5.696293354034424,
"epoch": 2.425261244888687,
"grad_norm": 1.0625,
"learning_rate": 0.00046977319669934506,
"loss": 5.4655,
"mean_token_accuracy": 0.16867001354694366,
"num_tokens": 5524822.0,
"step": 2670
},
{
"entropy": 5.625522708892822,
"epoch": 2.429804634257156,
"grad_norm": 1.078125,
"learning_rate": 0.00046959601757372814,
"loss": 5.464,
"mean_token_accuracy": 0.16628979593515397,
"num_tokens": 5535583.0,
"step": 2675
},
{
"entropy": 5.597361516952515,
"epoch": 2.4343480236256245,
"grad_norm": 1.09375,
"learning_rate": 0.00046941835830178216,
"loss": 5.4483,
"mean_token_accuracy": 0.17172520905733107,
"num_tokens": 5546199.0,
"step": 2680
},
{
"entropy": 5.57995343208313,
"epoch": 2.4388914129940935,
"grad_norm": 1.421875,
"learning_rate": 0.00046924021932186357,
"loss": 5.4142,
"mean_token_accuracy": 0.17431910037994386,
"num_tokens": 5556409.0,
"step": 2685
},
{
"entropy": 5.58826322555542,
"epoch": 2.4434348023625625,
"grad_norm": 1.125,
"learning_rate": 0.00046906160107351263,
"loss": 5.4798,
"mean_token_accuracy": 0.17440305650234222,
"num_tokens": 5565854.0,
"step": 2690
},
{
"entropy": 5.676161527633667,
"epoch": 2.4479781917310315,
"grad_norm": 1.03125,
"learning_rate": 0.00046888250399745217,
"loss": 5.5125,
"mean_token_accuracy": 0.15484314262866974,
"num_tokens": 5576194.0,
"step": 2695
},
{
"entropy": 5.585902309417724,
"epoch": 2.4525215810995,
"grad_norm": 1.171875,
"learning_rate": 0.00046870292853558646,
"loss": 5.4197,
"mean_token_accuracy": 0.16805924326181412,
"num_tokens": 5586679.0,
"step": 2700
},
{
"entropy": 5.61578893661499,
"epoch": 2.457064970467969,
"grad_norm": 1.0703125,
"learning_rate": 0.00046852287513099997,
"loss": 5.4645,
"mean_token_accuracy": 0.15748845040798187,
"num_tokens": 5597268.0,
"step": 2705
},
{
"entropy": 5.615812826156616,
"epoch": 2.461608359836438,
"grad_norm": 1.1796875,
"learning_rate": 0.0004683423442279566,
"loss": 5.439,
"mean_token_accuracy": 0.16412849575281144,
"num_tokens": 5608046.0,
"step": 2710
},
{
"entropy": 5.583138132095337,
"epoch": 2.466151749204907,
"grad_norm": 1.125,
"learning_rate": 0.0004681613362718985,
"loss": 5.3561,
"mean_token_accuracy": 0.1718935936689377,
"num_tokens": 5617541.0,
"step": 2715
},
{
"entropy": 5.597376203536987,
"epoch": 2.470695138573376,
"grad_norm": 1.1953125,
"learning_rate": 0.0004679798517094447,
"loss": 5.4094,
"mean_token_accuracy": 0.1683037906885147,
"num_tokens": 5628428.0,
"step": 2720
},
{
"entropy": 5.616898536682129,
"epoch": 2.4752385279418445,
"grad_norm": 1.0703125,
"learning_rate": 0.00046779789098839043,
"loss": 5.6047,
"mean_token_accuracy": 0.15547200217843055,
"num_tokens": 5639557.0,
"step": 2725
},
{
"entropy": 5.744855785369873,
"epoch": 2.4797819173103135,
"grad_norm": 1.125,
"learning_rate": 0.00046761545455770555,
"loss": 5.5669,
"mean_token_accuracy": 0.16440719962120057,
"num_tokens": 5650586.0,
"step": 2730
},
{
"entropy": 5.625218391418457,
"epoch": 2.4843253066787825,
"grad_norm": 1.2265625,
"learning_rate": 0.0004674325428675339,
"loss": 5.4893,
"mean_token_accuracy": 0.16572684943675994,
"num_tokens": 5661525.0,
"step": 2735
},
{
"entropy": 5.610922002792359,
"epoch": 2.488868696047251,
"grad_norm": 1.15625,
"learning_rate": 0.0004672491563691919,
"loss": 5.4912,
"mean_token_accuracy": 0.16836876422166824,
"num_tokens": 5671272.0,
"step": 2740
},
{
"entropy": 5.580408716201783,
"epoch": 2.49341208541572,
"grad_norm": 1.234375,
"learning_rate": 0.00046706529551516753,
"loss": 5.4152,
"mean_token_accuracy": 0.1693589746952057,
"num_tokens": 5680384.0,
"step": 2745
},
{
"entropy": 5.579071950912476,
"epoch": 2.497955474784189,
"grad_norm": 1.1875,
"learning_rate": 0.00046688096075911905,
"loss": 5.4525,
"mean_token_accuracy": 0.1694910243153572,
"num_tokens": 5690983.0,
"step": 2750
},
{
"entropy": 5.600765657424927,
"epoch": 2.502498864152658,
"grad_norm": 1.1796875,
"learning_rate": 0.0004666961525558743,
"loss": 5.4536,
"mean_token_accuracy": 0.16386417150497437,
"num_tokens": 5702144.0,
"step": 2755
},
{
"entropy": 5.621084690093994,
"epoch": 2.507042253521127,
"grad_norm": 1.1796875,
"learning_rate": 0.000466510871361429,
"loss": 5.5238,
"mean_token_accuracy": 0.1611946403980255,
"num_tokens": 5713283.0,
"step": 2760
},
{
"entropy": 5.585568809509278,
"epoch": 2.5115856428895955,
"grad_norm": 1.0390625,
"learning_rate": 0.00046632511763294615,
"loss": 5.496,
"mean_token_accuracy": 0.1589619129896164,
"num_tokens": 5724533.0,
"step": 2765
},
{
"entropy": 5.6201286792755125,
"epoch": 2.5161290322580645,
"grad_norm": 1.1015625,
"learning_rate": 0.0004661388918287546,
"loss": 5.4345,
"mean_token_accuracy": 0.16408485770225525,
"num_tokens": 5735186.0,
"step": 2770
},
{
"entropy": 5.633163261413574,
"epoch": 2.5206724216265335,
"grad_norm": 1.1328125,
"learning_rate": 0.0004659521944083481,
"loss": 5.4981,
"mean_token_accuracy": 0.16201248839497567,
"num_tokens": 5746537.0,
"step": 2775
},
{
"entropy": 5.5752369403839115,
"epoch": 2.525215810995002,
"grad_norm": 1.15625,
"learning_rate": 0.0004657650258323838,
"loss": 5.4918,
"mean_token_accuracy": 0.1645279496908188,
"num_tokens": 5756666.0,
"step": 2780
},
{
"entropy": 5.60757155418396,
"epoch": 2.529759200363471,
"grad_norm": 1.109375,
"learning_rate": 0.00046557738656268166,
"loss": 5.4936,
"mean_token_accuracy": 0.16352157443761825,
"num_tokens": 5766870.0,
"step": 2785
},
{
"entropy": 5.603203678131104,
"epoch": 2.53430258973194,
"grad_norm": 1.078125,
"learning_rate": 0.000465389277062223,
"loss": 5.4748,
"mean_token_accuracy": 0.16887758076190948,
"num_tokens": 5776775.0,
"step": 2790
},
{
"entropy": 5.7017662525177,
"epoch": 2.538845979100409,
"grad_norm": 1.1640625,
"learning_rate": 0.00046520069779514917,
"loss": 5.498,
"mean_token_accuracy": 0.16400706321001052,
"num_tokens": 5786993.0,
"step": 2795
},
{
"entropy": 5.554024314880371,
"epoch": 2.543389368468878,
"grad_norm": 1.1796875,
"learning_rate": 0.0004650116492267609,
"loss": 5.4688,
"mean_token_accuracy": 0.1680043414235115,
"num_tokens": 5796230.0,
"step": 2800
},
{
"entropy": 5.651140642166138,
"epoch": 2.5479327578373465,
"grad_norm": 1.1015625,
"learning_rate": 0.0004648221318235165,
"loss": 5.4665,
"mean_token_accuracy": 0.16692867875099182,
"num_tokens": 5807086.0,
"step": 2805
},
{
"entropy": 5.55869312286377,
"epoch": 2.5524761472058155,
"grad_norm": 1.0,
"learning_rate": 0.0004646321460530316,
"loss": 5.5085,
"mean_token_accuracy": 0.16026389002799987,
"num_tokens": 5817897.0,
"step": 2810
},
{
"entropy": 5.630576515197754,
"epoch": 2.5570195365742845,
"grad_norm": 1.109375,
"learning_rate": 0.000464441692384077,
"loss": 5.3969,
"mean_token_accuracy": 0.17501779049634933,
"num_tokens": 5827794.0,
"step": 2815
},
{
"entropy": 5.537441635131836,
"epoch": 2.561562925942753,
"grad_norm": 1.1875,
"learning_rate": 0.0004642507712865784,
"loss": 5.4863,
"mean_token_accuracy": 0.16003730595111848,
"num_tokens": 5837603.0,
"step": 2820
},
{
"entropy": 5.592154407501221,
"epoch": 2.566106315311222,
"grad_norm": 1.140625,
"learning_rate": 0.00046405938323161453,
"loss": 5.3971,
"mean_token_accuracy": 0.17378488332033157,
"num_tokens": 5847013.0,
"step": 2825
},
{
"entropy": 5.588395690917968,
"epoch": 2.570649704679691,
"grad_norm": 1.3828125,
"learning_rate": 0.00046386752869141624,
"loss": 5.3773,
"mean_token_accuracy": 0.16975186318159102,
"num_tokens": 5856783.0,
"step": 2830
},
{
"entropy": 5.544240140914917,
"epoch": 2.57519309404816,
"grad_norm": 1.1875,
"learning_rate": 0.0004636752081393658,
"loss": 5.3943,
"mean_token_accuracy": 0.16941520422697068,
"num_tokens": 5866241.0,
"step": 2835
},
{
"entropy": 5.5188463687896725,
"epoch": 2.579736483416629,
"grad_norm": 1.2265625,
"learning_rate": 0.00046348242204999487,
"loss": 5.4315,
"mean_token_accuracy": 0.1725137710571289,
"num_tokens": 5876135.0,
"step": 2840
},
{
"entropy": 5.550423097610474,
"epoch": 2.5842798727850975,
"grad_norm": 1.1328125,
"learning_rate": 0.00046328917089898396,
"loss": 5.516,
"mean_token_accuracy": 0.16753721833229065,
"num_tokens": 5886841.0,
"step": 2845
},
{
"entropy": 5.6302684307098385,
"epoch": 2.5888232621535665,
"grad_norm": 1.1015625,
"learning_rate": 0.0004630954551631612,
"loss": 5.4524,
"mean_token_accuracy": 0.1689890906214714,
"num_tokens": 5896884.0,
"step": 2850
},
{
"entropy": 5.578263568878174,
"epoch": 2.5933666515220355,
"grad_norm": 1.3203125,
"learning_rate": 0.00046290127532050087,
"loss": 5.4744,
"mean_token_accuracy": 0.1660120502114296,
"num_tokens": 5906194.0,
"step": 2855
},
{
"entropy": 5.586617136001587,
"epoch": 2.5979100408905045,
"grad_norm": 1.046875,
"learning_rate": 0.0004627066318501223,
"loss": 5.3953,
"mean_token_accuracy": 0.17517496943473815,
"num_tokens": 5917217.0,
"step": 2860
},
{
"entropy": 5.537663507461548,
"epoch": 2.602453430258973,
"grad_norm": 1.2265625,
"learning_rate": 0.00046251152523228906,
"loss": 5.3406,
"mean_token_accuracy": 0.1728805810213089,
"num_tokens": 5926307.0,
"step": 2865
},
{
"entropy": 5.527540969848633,
"epoch": 2.606996819627442,
"grad_norm": 1.1015625,
"learning_rate": 0.0004623159559484074,
"loss": 5.46,
"mean_token_accuracy": 0.1611982226371765,
"num_tokens": 5936718.0,
"step": 2870
},
{
"entropy": 5.513413000106811,
"epoch": 2.611540208995911,
"grad_norm": 1.1484375,
"learning_rate": 0.0004621199244810248,
"loss": 5.4089,
"mean_token_accuracy": 0.1685175195336342,
"num_tokens": 5947870.0,
"step": 2875
},
{
"entropy": 5.619717216491699,
"epoch": 2.61608359836438,
"grad_norm": 1.0234375,
"learning_rate": 0.00046192343131382964,
"loss": 5.5763,
"mean_token_accuracy": 0.1631942793726921,
"num_tokens": 5959090.0,
"step": 2880
},
{
"entropy": 5.655989027023315,
"epoch": 2.620626987732849,
"grad_norm": 1.078125,
"learning_rate": 0.00046172647693164926,
"loss": 5.5473,
"mean_token_accuracy": 0.16171808689832687,
"num_tokens": 5970435.0,
"step": 2885
},
{
"entropy": 5.556521701812744,
"epoch": 2.6251703771013175,
"grad_norm": 1.2421875,
"learning_rate": 0.0004615290618204489,
"loss": 5.3878,
"mean_token_accuracy": 0.17438449412584306,
"num_tokens": 5979732.0,
"step": 2890
},
{
"entropy": 5.545953798294067,
"epoch": 2.6297137664697865,
"grad_norm": 1.0234375,
"learning_rate": 0.0004613311864673309,
"loss": 5.5222,
"mean_token_accuracy": 0.17003980576992034,
"num_tokens": 5989965.0,
"step": 2895
},
{
"entropy": 5.657530307769775,
"epoch": 2.6342571558382555,
"grad_norm": 1.28125,
"learning_rate": 0.00046113285136053286,
"loss": 5.5088,
"mean_token_accuracy": 0.17201769798994065,
"num_tokens": 6000752.0,
"step": 2900
},
{
"entropy": 5.574502992630005,
"epoch": 2.638800545206724,
"grad_norm": 1.109375,
"learning_rate": 0.000460934056989427,
"loss": 5.4674,
"mean_token_accuracy": 0.16234221756458284,
"num_tokens": 6011318.0,
"step": 2905
},
{
"entropy": 5.565732717514038,
"epoch": 2.643343934575193,
"grad_norm": 1.0234375,
"learning_rate": 0.0004607348038445187,
"loss": 5.467,
"mean_token_accuracy": 0.1668257936835289,
"num_tokens": 6022024.0,
"step": 2910
},
{
"entropy": 5.529544115066528,
"epoch": 2.647887323943662,
"grad_norm": 1.109375,
"learning_rate": 0.00046053509241744526,
"loss": 5.4304,
"mean_token_accuracy": 0.16876450926065445,
"num_tokens": 6032118.0,
"step": 2915
},
{
"entropy": 5.630978393554687,
"epoch": 2.652430713312131,
"grad_norm": 1.265625,
"learning_rate": 0.00046033492320097483,
"loss": 5.4591,
"mean_token_accuracy": 0.17092808037996293,
"num_tokens": 6042607.0,
"step": 2920
},
{
"entropy": 5.500306272506714,
"epoch": 2.6569741026806,
"grad_norm": 1.1484375,
"learning_rate": 0.0004601342966890049,
"loss": 5.4601,
"mean_token_accuracy": 0.16697472631931304,
"num_tokens": 6053625.0,
"step": 2925
},
{
"entropy": 5.556142330169678,
"epoch": 2.6615174920490685,
"grad_norm": 1.34375,
"learning_rate": 0.00045993321337656163,
"loss": 5.4113,
"mean_token_accuracy": 0.1766616463661194,
"num_tokens": 6063823.0,
"step": 2930
},
{
"entropy": 5.614065027236938,
"epoch": 2.6660608814175375,
"grad_norm": 1.0703125,
"learning_rate": 0.000459731673759798,
"loss": 5.402,
"mean_token_accuracy": 0.16765334010124205,
"num_tokens": 6073914.0,
"step": 2935
},
{
"entropy": 5.549696254730224,
"epoch": 2.6706042707860065,
"grad_norm": 1.0625,
"learning_rate": 0.00045952967833599303,
"loss": 5.3874,
"mean_token_accuracy": 0.17039278745651246,
"num_tokens": 6085121.0,
"step": 2940
},
{
"entropy": 5.484119701385498,
"epoch": 2.675147660154475,
"grad_norm": 1.15625,
"learning_rate": 0.0004593272276035502,
"loss": 5.3093,
"mean_token_accuracy": 0.18039538711309433,
"num_tokens": 6095324.0,
"step": 2945
},
{
"entropy": 5.540274953842163,
"epoch": 2.679691049522944,
"grad_norm": 1.2578125,
"learning_rate": 0.0004591243220619969,
"loss": 5.4293,
"mean_token_accuracy": 0.179240944981575,
"num_tokens": 6104704.0,
"step": 2950
},
{
"entropy": 5.631873559951782,
"epoch": 2.684234438891413,
"grad_norm": 1.109375,
"learning_rate": 0.000458920962211982,
"loss": 5.5444,
"mean_token_accuracy": 0.16498605608940126,
"num_tokens": 6116569.0,
"step": 2955
},
{
"entropy": 5.667964220046997,
"epoch": 2.688777828259882,
"grad_norm": 1.1171875,
"learning_rate": 0.0004587171485552759,
"loss": 5.5176,
"mean_token_accuracy": 0.1636403501033783,
"num_tokens": 6126166.0,
"step": 2960
},
{
"entropy": 5.571827745437622,
"epoch": 2.693321217628351,
"grad_norm": 1.1015625,
"learning_rate": 0.00045851288159476854,
"loss": 5.3881,
"mean_token_accuracy": 0.17198004126548766,
"num_tokens": 6136169.0,
"step": 2965
},
{
"entropy": 5.588191413879395,
"epoch": 2.6978646069968195,
"grad_norm": 1.1015625,
"learning_rate": 0.0004583081618344684,
"loss": 5.4695,
"mean_token_accuracy": 0.1690232589840889,
"num_tokens": 6146249.0,
"step": 2970
},
{
"entropy": 5.685199117660522,
"epoch": 2.7024079963652885,
"grad_norm": 1.0390625,
"learning_rate": 0.000458102989779501,
"loss": 5.5427,
"mean_token_accuracy": 0.16583559662103653,
"num_tokens": 6156892.0,
"step": 2975
},
{
"entropy": 5.563881731033325,
"epoch": 2.7069513857337575,
"grad_norm": 1.2265625,
"learning_rate": 0.0004578973659361081,
"loss": 5.4123,
"mean_token_accuracy": 0.17618827819824218,
"num_tokens": 6166915.0,
"step": 2980
},
{
"entropy": 5.5185730934143065,
"epoch": 2.711494775102226,
"grad_norm": 1.15625,
"learning_rate": 0.00045769129081164596,
"loss": 5.3704,
"mean_token_accuracy": 0.173337659239769,
"num_tokens": 6176837.0,
"step": 2985
},
{
"entropy": 5.5104727268219,
"epoch": 2.716038164470695,
"grad_norm": 1.015625,
"learning_rate": 0.0004574847649145845,
"loss": 5.4539,
"mean_token_accuracy": 0.16636477261781693,
"num_tokens": 6188038.0,
"step": 2990
},
{
"entropy": 5.573533058166504,
"epoch": 2.720581553839164,
"grad_norm": 1.2109375,
"learning_rate": 0.000457277788754506,
"loss": 5.4647,
"mean_token_accuracy": 0.16517086327075958,
"num_tokens": 6199835.0,
"step": 2995
},
{
"entropy": 5.654835844039917,
"epoch": 2.725124943207633,
"grad_norm": 1.203125,
"learning_rate": 0.00045707036284210344,
"loss": 5.3962,
"mean_token_accuracy": 0.16700971275568008,
"num_tokens": 6209122.0,
"step": 3000
},
{
"epoch": 2.725124943207633,
"eval_entropy": 5.43793414359869,
"eval_loss": 5.883479118347168,
"eval_mean_token_accuracy": 0.15563077325391214,
"eval_num_tokens": 6209122.0,
"eval_runtime": 2.5908,
"eval_samples_per_second": 1325.837,
"eval_steps_per_second": 165.971,
"step": 3000
},
{
"entropy": 5.5143434524536135,
"epoch": 2.729668332576102,
"grad_norm": 1.171875,
"learning_rate": 0.0004568624876891796,
"loss": 5.4457,
"mean_token_accuracy": 0.16970110088586807,
"num_tokens": 6219317.0,
"step": 3005
},
{
"entropy": 5.638797378540039,
"epoch": 2.7342117219445705,
"grad_norm": 1.125,
"learning_rate": 0.00045665416380864587,
"loss": 5.5011,
"mean_token_accuracy": 0.16697756946086884,
"num_tokens": 6229561.0,
"step": 3010
},
{
"entropy": 5.609724044799805,
"epoch": 2.7387551113130395,
"grad_norm": 1.2421875,
"learning_rate": 0.0004564453917145206,
"loss": 5.4755,
"mean_token_accuracy": 0.1662609875202179,
"num_tokens": 6240150.0,
"step": 3015
},
{
"entropy": 5.520001983642578,
"epoch": 2.7432985006815085,
"grad_norm": 1.171875,
"learning_rate": 0.0004562361719219283,
"loss": 5.357,
"mean_token_accuracy": 0.16793376505374907,
"num_tokens": 6250641.0,
"step": 3020
},
{
"entropy": 5.588851594924927,
"epoch": 2.747841890049977,
"grad_norm": 1.234375,
"learning_rate": 0.00045602650494709786,
"loss": 5.4381,
"mean_token_accuracy": 0.1629244402050972,
"num_tokens": 6261239.0,
"step": 3025
},
{
"entropy": 5.516488885879516,
"epoch": 2.752385279418446,
"grad_norm": 1.0703125,
"learning_rate": 0.00045581639130736173,
"loss": 5.4572,
"mean_token_accuracy": 0.16792365312576293,
"num_tokens": 6270919.0,
"step": 3030
},
{
"entropy": 5.5905860424041744,
"epoch": 2.756928668786915,
"grad_norm": 1.046875,
"learning_rate": 0.0004556058315211546,
"loss": 5.5349,
"mean_token_accuracy": 0.16173456907272338,
"num_tokens": 6283007.0,
"step": 3035
},
{
"entropy": 5.4845935821533205,
"epoch": 2.761472058155384,
"grad_norm": 1.109375,
"learning_rate": 0.00045539482610801156,
"loss": 5.3776,
"mean_token_accuracy": 0.17222308367490768,
"num_tokens": 6293596.0,
"step": 3040
},
{
"entropy": 5.553569793701172,
"epoch": 2.766015447523853,
"grad_norm": 1.2109375,
"learning_rate": 0.0004551833755885677,
"loss": 5.4216,
"mean_token_accuracy": 0.17449651956558226,
"num_tokens": 6304568.0,
"step": 3045
},
{
"entropy": 5.5662219524383545,
"epoch": 2.7705588368923215,
"grad_norm": 1.125,
"learning_rate": 0.0004549714804845561,
"loss": 5.3998,
"mean_token_accuracy": 0.169203719496727,
"num_tokens": 6315275.0,
"step": 3050
},
{
"entropy": 5.577321100234985,
"epoch": 2.7751022262607905,
"grad_norm": 1.1171875,
"learning_rate": 0.0004547591413188068,
"loss": 5.5116,
"mean_token_accuracy": 0.16704116463661195,
"num_tokens": 6324986.0,
"step": 3055
},
{
"entropy": 5.589274072647095,
"epoch": 2.7796456156292595,
"grad_norm": 1.2421875,
"learning_rate": 0.0004545463586152456,
"loss": 5.5148,
"mean_token_accuracy": 0.16077900826931,
"num_tokens": 6336318.0,
"step": 3060
},
{
"entropy": 5.565379905700683,
"epoch": 2.784189004997728,
"grad_norm": 1.1171875,
"learning_rate": 0.0004543331328988927,
"loss": 5.3866,
"mean_token_accuracy": 0.17265889495611192,
"num_tokens": 6345986.0,
"step": 3065
},
{
"entropy": 5.604962110519409,
"epoch": 2.788732394366197,
"grad_norm": 1.21875,
"learning_rate": 0.0004541194646958614,
"loss": 5.4841,
"mean_token_accuracy": 0.1655448466539383,
"num_tokens": 6357025.0,
"step": 3070
},
{
"entropy": 5.507268619537354,
"epoch": 2.793275783734666,
"grad_norm": 1.1484375,
"learning_rate": 0.00045390535453335666,
"loss": 5.4593,
"mean_token_accuracy": 0.16889591962099076,
"num_tokens": 6367885.0,
"step": 3075
},
{
"entropy": 5.4771851062774655,
"epoch": 2.797819173103135,
"grad_norm": 1.1796875,
"learning_rate": 0.00045369080293967414,
"loss": 5.3863,
"mean_token_accuracy": 0.16707149147987366,
"num_tokens": 6377729.0,
"step": 3080
},
{
"entropy": 5.533416080474853,
"epoch": 2.802362562471604,
"grad_norm": 1.09375,
"learning_rate": 0.00045347581044419846,
"loss": 5.3784,
"mean_token_accuracy": 0.17345982044935226,
"num_tokens": 6387368.0,
"step": 3085
},
{
"entropy": 5.633910608291626,
"epoch": 2.806905951840073,
"grad_norm": 1.109375,
"learning_rate": 0.00045326037757740216,
"loss": 5.4833,
"mean_token_accuracy": 0.16738815158605574,
"num_tokens": 6397771.0,
"step": 3090
},
{
"entropy": 5.55745301246643,
"epoch": 2.8114493412085415,
"grad_norm": 1.1328125,
"learning_rate": 0.0004530445048708447,
"loss": 5.363,
"mean_token_accuracy": 0.17728449255228043,
"num_tokens": 6408238.0,
"step": 3095
},
{
"entropy": 5.568863821029663,
"epoch": 2.8159927305770105,
"grad_norm": 1.09375,
"learning_rate": 0.00045282819285717024,
"loss": 5.5053,
"mean_token_accuracy": 0.16698647141456605,
"num_tokens": 6419196.0,
"step": 3100
},
{
"entropy": 5.586069393157959,
"epoch": 2.8205361199454795,
"grad_norm": 1.125,
"learning_rate": 0.00045261144207010736,
"loss": 5.448,
"mean_token_accuracy": 0.17154122292995452,
"num_tokens": 6430342.0,
"step": 3105
},
{
"entropy": 5.591652727127075,
"epoch": 2.825079509313948,
"grad_norm": 1.1171875,
"learning_rate": 0.00045239425304446684,
"loss": 5.3608,
"mean_token_accuracy": 0.17456261515617372,
"num_tokens": 6440218.0,
"step": 3110
},
{
"entropy": 5.489208936691284,
"epoch": 2.829622898682417,
"grad_norm": 1.140625,
"learning_rate": 0.0004521766263161413,
"loss": 5.369,
"mean_token_accuracy": 0.17575375437736512,
"num_tokens": 6450128.0,
"step": 3115
},
{
"entropy": 5.509318590164185,
"epoch": 2.834166288050886,
"grad_norm": 1.109375,
"learning_rate": 0.000451958562422103,
"loss": 5.3761,
"mean_token_accuracy": 0.17748037576675416,
"num_tokens": 6459916.0,
"step": 3120
},
{
"entropy": 5.584203004837036,
"epoch": 2.838709677419355,
"grad_norm": 1.109375,
"learning_rate": 0.00045174006190040267,
"loss": 5.4018,
"mean_token_accuracy": 0.17300332933664322,
"num_tokens": 6470250.0,
"step": 3125
},
{
"entropy": 5.584368658065796,
"epoch": 2.843253066787824,
"grad_norm": 1.03125,
"learning_rate": 0.000451521125290169,
"loss": 5.502,
"mean_token_accuracy": 0.16230286359786988,
"num_tokens": 6481623.0,
"step": 3130
},
{
"entropy": 5.581066274642945,
"epoch": 2.8477964561562925,
"grad_norm": 1.078125,
"learning_rate": 0.000451301753131606,
"loss": 5.4113,
"mean_token_accuracy": 0.16658567786216735,
"num_tokens": 6492143.0,
"step": 3135
},
{
"entropy": 5.600178337097168,
"epoch": 2.8523398455247615,
"grad_norm": 1.3046875,
"learning_rate": 0.00045108194596599286,
"loss": 5.4134,
"mean_token_accuracy": 0.17637794613838195,
"num_tokens": 6501643.0,
"step": 3140
},
{
"entropy": 5.486820983886719,
"epoch": 2.8568832348932305,
"grad_norm": 1.2421875,
"learning_rate": 0.0004508617043356818,
"loss": 5.3522,
"mean_token_accuracy": 0.17611457407474518,
"num_tokens": 6511442.0,
"step": 3145
},
{
"entropy": 5.505658483505249,
"epoch": 2.861426624261699,
"grad_norm": 1.15625,
"learning_rate": 0.0004506410287840972,
"loss": 5.2864,
"mean_token_accuracy": 0.17741763442754746,
"num_tokens": 6521303.0,
"step": 3150
},
{
"entropy": 5.478482866287232,
"epoch": 2.865970013630168,
"grad_norm": 1.1796875,
"learning_rate": 0.000450419919855734,
"loss": 5.4206,
"mean_token_accuracy": 0.1686701387166977,
"num_tokens": 6531143.0,
"step": 3155
},
{
"entropy": 5.592413854598999,
"epoch": 2.870513402998637,
"grad_norm": 1.140625,
"learning_rate": 0.0004501983780961566,
"loss": 5.4867,
"mean_token_accuracy": 0.1674318328499794,
"num_tokens": 6540899.0,
"step": 3160
},
{
"entropy": 5.682785701751709,
"epoch": 2.875056792367106,
"grad_norm": 1.203125,
"learning_rate": 0.0004499764040519972,
"loss": 5.5307,
"mean_token_accuracy": 0.16565336138010026,
"num_tokens": 6552131.0,
"step": 3165
},
{
"entropy": 5.528329563140869,
"epoch": 2.879600181735575,
"grad_norm": 1.078125,
"learning_rate": 0.00044975399827095467,
"loss": 5.461,
"mean_token_accuracy": 0.1624034345149994,
"num_tokens": 6563275.0,
"step": 3170
},
{
"entropy": 5.498131227493286,
"epoch": 2.8841435711040435,
"grad_norm": 1.15625,
"learning_rate": 0.00044953116130179313,
"loss": 5.3192,
"mean_token_accuracy": 0.17647540271282197,
"num_tokens": 6573109.0,
"step": 3175
},
{
"entropy": 5.495909833908081,
"epoch": 2.8886869604725125,
"grad_norm": 1.046875,
"learning_rate": 0.0004493078936943408,
"loss": 5.4264,
"mean_token_accuracy": 0.16593341231346131,
"num_tokens": 6584978.0,
"step": 3180
},
{
"entropy": 5.485671854019165,
"epoch": 2.8932303498409815,
"grad_norm": 1.1171875,
"learning_rate": 0.0004490841959994882,
"loss": 5.4355,
"mean_token_accuracy": 0.16810845583677292,
"num_tokens": 6595608.0,
"step": 3185
},
{
"entropy": 5.505950784683227,
"epoch": 2.89777373920945,
"grad_norm": 1.1640625,
"learning_rate": 0.00044886006876918716,
"loss": 5.317,
"mean_token_accuracy": 0.1710359960794449,
"num_tokens": 6605983.0,
"step": 3190
},
{
"entropy": 5.569347095489502,
"epoch": 2.902317128577919,
"grad_norm": 1.0078125,
"learning_rate": 0.0004486355125564493,
"loss": 5.435,
"mean_token_accuracy": 0.169928440451622,
"num_tokens": 6616682.0,
"step": 3195
},
{
"entropy": 5.607175588607788,
"epoch": 2.906860517946388,
"grad_norm": 1.1796875,
"learning_rate": 0.0004484105279153449,
"loss": 5.4781,
"mean_token_accuracy": 0.16958287209272385,
"num_tokens": 6628715.0,
"step": 3200
},
{
"entropy": 5.568323945999145,
"epoch": 2.911403907314857,
"grad_norm": 1.0703125,
"learning_rate": 0.000448185115401001,
"loss": 5.4585,
"mean_token_accuracy": 0.16846979558467864,
"num_tokens": 6639219.0,
"step": 3205
},
{
"entropy": 5.492281484603882,
"epoch": 2.915947296683326,
"grad_norm": 1.1640625,
"learning_rate": 0.0004479592755696008,
"loss": 5.349,
"mean_token_accuracy": 0.17246212661266327,
"num_tokens": 6649516.0,
"step": 3210
},
{
"entropy": 5.5872766971588135,
"epoch": 2.9204906860517945,
"grad_norm": 1.1796875,
"learning_rate": 0.00044773300897838137,
"loss": 5.4603,
"mean_token_accuracy": 0.17043966352939605,
"num_tokens": 6659470.0,
"step": 3215
},
{
"entropy": 5.537988948822021,
"epoch": 2.9250340754202635,
"grad_norm": 1.2734375,
"learning_rate": 0.0004475063161856332,
"loss": 5.436,
"mean_token_accuracy": 0.17161599546670914,
"num_tokens": 6669637.0,
"step": 3220
},
{
"entropy": 5.506859159469604,
"epoch": 2.9295774647887325,
"grad_norm": 1.1015625,
"learning_rate": 0.0004472791977506983,
"loss": 5.3825,
"mean_token_accuracy": 0.1703056737780571,
"num_tokens": 6680450.0,
"step": 3225
},
{
"entropy": 5.541793155670166,
"epoch": 2.934120854157201,
"grad_norm": 1.171875,
"learning_rate": 0.00044705165423396855,
"loss": 5.4723,
"mean_token_accuracy": 0.16821575313806533,
"num_tokens": 6691526.0,
"step": 3230
},
{
"entropy": 5.580589962005615,
"epoch": 2.93866424352567,
"grad_norm": 1.125,
"learning_rate": 0.00044682368619688523,
"loss": 5.4175,
"mean_token_accuracy": 0.17500658184289933,
"num_tokens": 6702543.0,
"step": 3235
},
{
"entropy": 5.566064929962158,
"epoch": 2.943207632894139,
"grad_norm": 1.265625,
"learning_rate": 0.0004465952942019367,
"loss": 5.4955,
"mean_token_accuracy": 0.16437314599752426,
"num_tokens": 6712374.0,
"step": 3240
},
{
"entropy": 5.559286451339721,
"epoch": 2.947751022262608,
"grad_norm": 1.1875,
"learning_rate": 0.00044636647881265745,
"loss": 5.3881,
"mean_token_accuracy": 0.16990749388933182,
"num_tokens": 6722091.0,
"step": 3245
},
{
"entropy": 5.495886564254761,
"epoch": 2.952294411631077,
"grad_norm": 1.140625,
"learning_rate": 0.0004461372405936269,
"loss": 5.3644,
"mean_token_accuracy": 0.1799268051981926,
"num_tokens": 6732024.0,
"step": 3250
},
{
"entropy": 5.577452278137207,
"epoch": 2.9568378009995455,
"grad_norm": 1.2109375,
"learning_rate": 0.0004459075801104673,
"loss": 5.4773,
"mean_token_accuracy": 0.16928715109825135,
"num_tokens": 6741480.0,
"step": 3255
},
{
"entropy": 5.536042165756226,
"epoch": 2.9613811903680145,
"grad_norm": 1.125,
"learning_rate": 0.00044567749792984325,
"loss": 5.3619,
"mean_token_accuracy": 0.17672986090183257,
"num_tokens": 6752504.0,
"step": 3260
},
{
"entropy": 5.560869121551514,
"epoch": 2.9659245797364835,
"grad_norm": 1.1484375,
"learning_rate": 0.00044544699461945966,
"loss": 5.4283,
"mean_token_accuracy": 0.16521647348999977,
"num_tokens": 6762945.0,
"step": 3265
},
{
"entropy": 5.542527389526367,
"epoch": 2.970467969104952,
"grad_norm": 1.140625,
"learning_rate": 0.00044521607074806037,
"loss": 5.3836,
"mean_token_accuracy": 0.17187629342079164,
"num_tokens": 6772862.0,
"step": 3270
},
{
"entropy": 5.566279792785645,
"epoch": 2.975011358473421,
"grad_norm": 1.125,
"learning_rate": 0.0004449847268854272,
"loss": 5.4066,
"mean_token_accuracy": 0.16729259938001634,
"num_tokens": 6782912.0,
"step": 3275
},
{
"entropy": 5.579110860824585,
"epoch": 2.97955474784189,
"grad_norm": 1.0234375,
"learning_rate": 0.0004447529636023782,
"loss": 5.4503,
"mean_token_accuracy": 0.17406559735536575,
"num_tokens": 6793249.0,
"step": 3280
},
{
"entropy": 5.563994073867798,
"epoch": 2.984098137210359,
"grad_norm": 1.125,
"learning_rate": 0.00044452078147076614,
"loss": 5.4597,
"mean_token_accuracy": 0.16985790133476258,
"num_tokens": 6803721.0,
"step": 3285
},
{
"entropy": 5.6080903053283695,
"epoch": 2.988641526578828,
"grad_norm": 1.09375,
"learning_rate": 0.0004442881810634774,
"loss": 5.4261,
"mean_token_accuracy": 0.17416889369487762,
"num_tokens": 6813637.0,
"step": 3290
},
{
"entropy": 5.527089834213257,
"epoch": 2.9931849159472965,
"grad_norm": 1.1015625,
"learning_rate": 0.0004440551629544303,
"loss": 5.4286,
"mean_token_accuracy": 0.16837423145771027,
"num_tokens": 6824171.0,
"step": 3295
},
{
"entropy": 5.44291205406189,
"epoch": 2.9977283053157655,
"grad_norm": 1.1171875,
"learning_rate": 0.00044382172771857383,
"loss": 5.2812,
"mean_token_accuracy": 0.1773454487323761,
"num_tokens": 6833668.0,
"step": 3300
},
{
"entropy": 5.45171939002143,
"epoch": 3.0018173557473875,
"grad_norm": 1.046875,
"learning_rate": 0.0004435878759318864,
"loss": 5.2016,
"mean_token_accuracy": 0.17712820735242632,
"num_tokens": 6842297.0,
"step": 3305
},
{
"entropy": 5.592501783370972,
"epoch": 3.0063607451158565,
"grad_norm": 1.1484375,
"learning_rate": 0.0004433536081713738,
"loss": 5.1714,
"mean_token_accuracy": 0.17631011307239533,
"num_tokens": 6853124.0,
"step": 3310
},
{
"entropy": 5.485796403884888,
"epoch": 3.0109041344843255,
"grad_norm": 1.1171875,
"learning_rate": 0.0004431189250150686,
"loss": 5.0419,
"mean_token_accuracy": 0.18753472715616226,
"num_tokens": 6861990.0,
"step": 3315
},
{
"entropy": 5.493206977844238,
"epoch": 3.015447523852794,
"grad_norm": 1.1015625,
"learning_rate": 0.00044288382704202816,
"loss": 5.0513,
"mean_token_accuracy": 0.1917167067527771,
"num_tokens": 6872024.0,
"step": 3320
},
{
"entropy": 5.427991008758545,
"epoch": 3.019990913221263,
"grad_norm": 1.1484375,
"learning_rate": 0.00044264831483233335,
"loss": 5.1137,
"mean_token_accuracy": 0.18667585849761964,
"num_tokens": 6881205.0,
"step": 3325
},
{
"entropy": 5.422624111175537,
"epoch": 3.024534302589732,
"grad_norm": 1.140625,
"learning_rate": 0.00044241238896708716,
"loss": 5.0724,
"mean_token_accuracy": 0.19332612156867982,
"num_tokens": 6891741.0,
"step": 3330
},
{
"entropy": 5.456476736068725,
"epoch": 3.029077691958201,
"grad_norm": 1.2265625,
"learning_rate": 0.0004421760500284131,
"loss": 5.1566,
"mean_token_accuracy": 0.18256397247314454,
"num_tokens": 6902122.0,
"step": 3335
},
{
"entropy": 5.485311603546142,
"epoch": 3.0336210813266695,
"grad_norm": 1.2265625,
"learning_rate": 0.0004419392985994541,
"loss": 5.1085,
"mean_token_accuracy": 0.1815920889377594,
"num_tokens": 6912129.0,
"step": 3340
},
{
"entropy": 5.420976305007935,
"epoch": 3.0381644706951385,
"grad_norm": 1.2578125,
"learning_rate": 0.00044170213526437076,
"loss": 5.0161,
"mean_token_accuracy": 0.1913357898592949,
"num_tokens": 6921409.0,
"step": 3345
},
{
"entropy": 5.42359299659729,
"epoch": 3.0427078600636075,
"grad_norm": 1.1171875,
"learning_rate": 0.00044146456060833994,
"loss": 5.1157,
"mean_token_accuracy": 0.18505289256572724,
"num_tokens": 6932346.0,
"step": 3350
},
{
"entropy": 5.443701648712159,
"epoch": 3.0472512494320765,
"grad_norm": 1.1640625,
"learning_rate": 0.00044122657521755373,
"loss": 5.1104,
"mean_token_accuracy": 0.1851183757185936,
"num_tokens": 6942139.0,
"step": 3355
},
{
"entropy": 5.5190465450286865,
"epoch": 3.051794638800545,
"grad_norm": 1.109375,
"learning_rate": 0.00044098817967921716,
"loss": 5.1291,
"mean_token_accuracy": 0.1860170230269432,
"num_tokens": 6952344.0,
"step": 3360
},
{
"entropy": 5.500512933731079,
"epoch": 3.056338028169014,
"grad_norm": 1.125,
"learning_rate": 0.00044074937458154763,
"loss": 5.1117,
"mean_token_accuracy": 0.18446057736873628,
"num_tokens": 6962430.0,
"step": 3365
},
{
"entropy": 5.474126195907592,
"epoch": 3.060881417537483,
"grad_norm": 1.0546875,
"learning_rate": 0.00044051016051377303,
"loss": 5.1854,
"mean_token_accuracy": 0.17663651704788208,
"num_tokens": 6973213.0,
"step": 3370
},
{
"entropy": 5.449837398529053,
"epoch": 3.065424806905952,
"grad_norm": 1.1953125,
"learning_rate": 0.0004402705380661302,
"loss": 5.109,
"mean_token_accuracy": 0.1824282929301262,
"num_tokens": 6983896.0,
"step": 3375
},
{
"entropy": 5.416739845275879,
"epoch": 3.0699681962744205,
"grad_norm": 1.21875,
"learning_rate": 0.00044003050782986377,
"loss": 5.0833,
"mean_token_accuracy": 0.18850862830877305,
"num_tokens": 6993549.0,
"step": 3380
},
{
"entropy": 5.442818832397461,
"epoch": 3.0745115856428895,
"grad_norm": 1.1015625,
"learning_rate": 0.0004397900703972245,
"loss": 5.1144,
"mean_token_accuracy": 0.1860892131924629,
"num_tokens": 7003893.0,
"step": 3385
},
{
"entropy": 5.5272105693817135,
"epoch": 3.0790549750113585,
"grad_norm": 1.1640625,
"learning_rate": 0.0004395492263614677,
"loss": 5.2164,
"mean_token_accuracy": 0.17726124078035355,
"num_tokens": 7014826.0,
"step": 3390
},
{
"entropy": 5.508368825912475,
"epoch": 3.0835983643798275,
"grad_norm": 1.1640625,
"learning_rate": 0.00043930797631685234,
"loss": 5.1548,
"mean_token_accuracy": 0.18513267934322358,
"num_tokens": 7024758.0,
"step": 3395
},
{
"entropy": 5.403151607513427,
"epoch": 3.088141753748296,
"grad_norm": 1.2265625,
"learning_rate": 0.00043906632085863876,
"loss": 5.1092,
"mean_token_accuracy": 0.19018383026123048,
"num_tokens": 7034063.0,
"step": 3400
},
{
"entropy": 5.4280308246612545,
"epoch": 3.092685143116765,
"grad_norm": 1.265625,
"learning_rate": 0.00043882426058308775,
"loss": 5.1134,
"mean_token_accuracy": 0.17717656791210173,
"num_tokens": 7045137.0,
"step": 3405
},
{
"entropy": 5.332364225387574,
"epoch": 3.097228532485234,
"grad_norm": 1.09375,
"learning_rate": 0.00043858179608745915,
"loss": 5.0576,
"mean_token_accuracy": 0.1812715083360672,
"num_tokens": 7056381.0,
"step": 3410
},
{
"entropy": 5.514891910552978,
"epoch": 3.101771921853703,
"grad_norm": 1.2109375,
"learning_rate": 0.0004383389279700098,
"loss": 5.1888,
"mean_token_accuracy": 0.17273895144462587,
"num_tokens": 7066462.0,
"step": 3415
},
{
"entropy": 5.545683145523071,
"epoch": 3.1063153112221715,
"grad_norm": 1.0859375,
"learning_rate": 0.00043809565682999274,
"loss": 5.2065,
"mean_token_accuracy": 0.1773798570036888,
"num_tokens": 7078438.0,
"step": 3420
},
{
"entropy": 5.4175057888031,
"epoch": 3.1108587005906405,
"grad_norm": 1.1015625,
"learning_rate": 0.00043785198326765535,
"loss": 5.0847,
"mean_token_accuracy": 0.18387235701084137,
"num_tokens": 7088981.0,
"step": 3425
},
{
"entropy": 5.4083209991455075,
"epoch": 3.1154020899591095,
"grad_norm": 1.140625,
"learning_rate": 0.00043760790788423785,
"loss": 5.1268,
"mean_token_accuracy": 0.1803571864962578,
"num_tokens": 7099407.0,
"step": 3430
},
{
"entropy": 5.408180475234985,
"epoch": 3.1199454793275785,
"grad_norm": 1.2109375,
"learning_rate": 0.00043736343128197207,
"loss": 5.0307,
"mean_token_accuracy": 0.19310947358608246,
"num_tokens": 7109454.0,
"step": 3435
},
{
"entropy": 5.372661066055298,
"epoch": 3.124488868696047,
"grad_norm": 1.296875,
"learning_rate": 0.0004371185540640796,
"loss": 5.136,
"mean_token_accuracy": 0.18486130833625794,
"num_tokens": 7119639.0,
"step": 3440
},
{
"entropy": 5.4526002407073975,
"epoch": 3.129032258064516,
"grad_norm": 1.1640625,
"learning_rate": 0.00043687327683477077,
"loss": 5.0955,
"mean_token_accuracy": 0.18959062546491623,
"num_tokens": 7129872.0,
"step": 3445
},
{
"entropy": 5.400248193740845,
"epoch": 3.133575647432985,
"grad_norm": 1.171875,
"learning_rate": 0.0004366276001992427,
"loss": 5.0359,
"mean_token_accuracy": 0.1907562270760536,
"num_tokens": 7140456.0,
"step": 3450
},
{
"entropy": 5.350067186355591,
"epoch": 3.138119036801454,
"grad_norm": 1.28125,
"learning_rate": 0.000436381524763678,
"loss": 5.0679,
"mean_token_accuracy": 0.18641221523284912,
"num_tokens": 7149787.0,
"step": 3455
},
{
"entropy": 5.367957973480225,
"epoch": 3.142662426169923,
"grad_norm": 1.1875,
"learning_rate": 0.0004361350511352435,
"loss": 5.0549,
"mean_token_accuracy": 0.1864156499505043,
"num_tokens": 7160482.0,
"step": 3460
},
{
"entropy": 5.469879150390625,
"epoch": 3.1472058155383915,
"grad_norm": 1.09375,
"learning_rate": 0.0004358881799220882,
"loss": 5.0686,
"mean_token_accuracy": 0.187416909635067,
"num_tokens": 7170818.0,
"step": 3465
},
{
"entropy": 5.355910015106201,
"epoch": 3.1517492049068605,
"grad_norm": 1.2265625,
"learning_rate": 0.00043564091173334236,
"loss": 5.1157,
"mean_token_accuracy": 0.1861570283770561,
"num_tokens": 7180770.0,
"step": 3470
},
{
"entropy": 5.382017660140991,
"epoch": 3.1562925942753295,
"grad_norm": 1.2421875,
"learning_rate": 0.00043539324717911573,
"loss": 5.0474,
"mean_token_accuracy": 0.19021925926208497,
"num_tokens": 7190751.0,
"step": 3475
},
{
"entropy": 5.4344110012054445,
"epoch": 3.1608359836437985,
"grad_norm": 1.2890625,
"learning_rate": 0.00043514518687049587,
"loss": 5.05,
"mean_token_accuracy": 0.18930084854364396,
"num_tokens": 7200093.0,
"step": 3480
},
{
"entropy": 5.431634473800659,
"epoch": 3.165379373012267,
"grad_norm": 1.1484375,
"learning_rate": 0.000434896731419547,
"loss": 5.1505,
"mean_token_accuracy": 0.18467694222927095,
"num_tokens": 7210526.0,
"step": 3485
},
{
"entropy": 5.3451460838317875,
"epoch": 3.169922762380736,
"grad_norm": 1.3046875,
"learning_rate": 0.0004346478814393082,
"loss": 5.1188,
"mean_token_accuracy": 0.18732051700353622,
"num_tokens": 7222127.0,
"step": 3490
},
{
"entropy": 5.3690876960754395,
"epoch": 3.174466151749205,
"grad_norm": 1.09375,
"learning_rate": 0.0004343986375437921,
"loss": 5.0648,
"mean_token_accuracy": 0.1850580394268036,
"num_tokens": 7233418.0,
"step": 3495
},
{
"entropy": 5.425455904006958,
"epoch": 3.179009541117674,
"grad_norm": 1.203125,
"learning_rate": 0.00043414900034798314,
"loss": 5.1328,
"mean_token_accuracy": 0.1841514840722084,
"num_tokens": 7243743.0,
"step": 3500
},
{
"epoch": 3.179009541117674,
"eval_entropy": 5.3034398234167766,
"eval_loss": 5.859730243682861,
"eval_mean_token_accuracy": 0.15904394596988378,
"eval_num_tokens": 7243743.0,
"eval_runtime": 2.5862,
"eval_samples_per_second": 1328.193,
"eval_steps_per_second": 166.266,
"step": 3500
},
{
"entropy": 5.419128513336181,
"epoch": 3.1835529304861425,
"grad_norm": 1.1640625,
"learning_rate": 0.0004338989704678364,
"loss": 5.1711,
"mean_token_accuracy": 0.17648412436246871,
"num_tokens": 7254906.0,
"step": 3505
},
{
"entropy": 5.379770135879516,
"epoch": 3.1880963198546115,
"grad_norm": 1.28125,
"learning_rate": 0.00043364854852027566,
"loss": 5.1185,
"mean_token_accuracy": 0.17789653986692427,
"num_tokens": 7264193.0,
"step": 3510
},
{
"entropy": 5.455638837814331,
"epoch": 3.1926397092230805,
"grad_norm": 1.234375,
"learning_rate": 0.0004333977351231924,
"loss": 5.1789,
"mean_token_accuracy": 0.185332790017128,
"num_tokens": 7273723.0,
"step": 3515
},
{
"entropy": 5.392770671844483,
"epoch": 3.1971830985915495,
"grad_norm": 1.1796875,
"learning_rate": 0.00043314653089544336,
"loss": 5.1879,
"mean_token_accuracy": 0.17546573877334595,
"num_tokens": 7284142.0,
"step": 3520
},
{
"entropy": 5.448967552185058,
"epoch": 3.201726487960018,
"grad_norm": 1.1953125,
"learning_rate": 0.0004328949364568504,
"loss": 5.0786,
"mean_token_accuracy": 0.18860593289136887,
"num_tokens": 7294228.0,
"step": 3525
},
{
"entropy": 5.371697568893433,
"epoch": 3.206269877328487,
"grad_norm": 1.2265625,
"learning_rate": 0.0004326429524281975,
"loss": 5.1018,
"mean_token_accuracy": 0.18420157134532927,
"num_tokens": 7304555.0,
"step": 3530
},
{
"entropy": 5.468740606307984,
"epoch": 3.210813266696956,
"grad_norm": 1.8828125,
"learning_rate": 0.0004323905794312303,
"loss": 5.2123,
"mean_token_accuracy": 0.174720498919487,
"num_tokens": 7315955.0,
"step": 3535
},
{
"entropy": 5.440519094467163,
"epoch": 3.215356656065425,
"grad_norm": 1.2890625,
"learning_rate": 0.00043213781808865414,
"loss": 5.1561,
"mean_token_accuracy": 0.17983912974596022,
"num_tokens": 7326757.0,
"step": 3540
},
{
"entropy": 5.387765121459961,
"epoch": 3.2199000454338935,
"grad_norm": 1.25,
"learning_rate": 0.0004318846690241324,
"loss": 5.1136,
"mean_token_accuracy": 0.19114665538072587,
"num_tokens": 7335969.0,
"step": 3545
},
{
"entropy": 5.361870670318604,
"epoch": 3.2244434348023625,
"grad_norm": 1.3125,
"learning_rate": 0.00043163113286228533,
"loss": 5.1138,
"mean_token_accuracy": 0.1813630685210228,
"num_tokens": 7345824.0,
"step": 3550
},
{
"entropy": 5.478123188018799,
"epoch": 3.2289868241708315,
"grad_norm": 1.1796875,
"learning_rate": 0.0004313772102286882,
"loss": 5.1958,
"mean_token_accuracy": 0.17310854643583298,
"num_tokens": 7356292.0,
"step": 3555
},
{
"entropy": 5.407268142700195,
"epoch": 3.2335302135393005,
"grad_norm": 1.3125,
"learning_rate": 0.0004311229017498698,
"loss": 5.1805,
"mean_token_accuracy": 0.1777238056063652,
"num_tokens": 7366809.0,
"step": 3560
},
{
"entropy": 5.426035833358765,
"epoch": 3.238073602907769,
"grad_norm": 1.296875,
"learning_rate": 0.000430868208053311,
"loss": 5.1027,
"mean_token_accuracy": 0.19172674417495728,
"num_tokens": 7376076.0,
"step": 3565
},
{
"entropy": 5.381820774078369,
"epoch": 3.242616992276238,
"grad_norm": 1.125,
"learning_rate": 0.00043061312976744326,
"loss": 5.1245,
"mean_token_accuracy": 0.17899244874715806,
"num_tokens": 7386338.0,
"step": 3570
},
{
"entropy": 5.403281211853027,
"epoch": 3.247160381644707,
"grad_norm": 1.515625,
"learning_rate": 0.0004303576675216469,
"loss": 5.1012,
"mean_token_accuracy": 0.1824015572667122,
"num_tokens": 7395792.0,
"step": 3575
},
{
"entropy": 5.4239842891693115,
"epoch": 3.251703771013176,
"grad_norm": 1.1484375,
"learning_rate": 0.00043010182194624957,
"loss": 5.1306,
"mean_token_accuracy": 0.18515096753835678,
"num_tokens": 7405765.0,
"step": 3580
},
{
"entropy": 5.3939097881317135,
"epoch": 3.2562471603816445,
"grad_norm": 1.21875,
"learning_rate": 0.0004298455936725248,
"loss": 5.1728,
"mean_token_accuracy": 0.1832568347454071,
"num_tokens": 7415213.0,
"step": 3585
},
{
"entropy": 5.463215160369873,
"epoch": 3.2607905497501135,
"grad_norm": 1.1171875,
"learning_rate": 0.0004295889833326903,
"loss": 5.157,
"mean_token_accuracy": 0.17885514497756957,
"num_tokens": 7425722.0,
"step": 3590
},
{
"entropy": 5.513776779174805,
"epoch": 3.2653339391185825,
"grad_norm": 1.1640625,
"learning_rate": 0.00042933199155990675,
"loss": 5.1996,
"mean_token_accuracy": 0.18279486298561096,
"num_tokens": 7436240.0,
"step": 3595
},
{
"entropy": 5.418988513946533,
"epoch": 3.2698773284870515,
"grad_norm": 1.203125,
"learning_rate": 0.0004290746189882757,
"loss": 5.1705,
"mean_token_accuracy": 0.1762332037091255,
"num_tokens": 7447878.0,
"step": 3600
},
{
"entropy": 5.436759901046753,
"epoch": 3.27442071785552,
"grad_norm": 1.1953125,
"learning_rate": 0.00042881686625283844,
"loss": 5.1302,
"mean_token_accuracy": 0.1849357306957245,
"num_tokens": 7458635.0,
"step": 3605
},
{
"entropy": 5.407280731201172,
"epoch": 3.278964107223989,
"grad_norm": 1.3828125,
"learning_rate": 0.0004285587339895742,
"loss": 5.168,
"mean_token_accuracy": 0.1762292981147766,
"num_tokens": 7468740.0,
"step": 3610
},
{
"entropy": 5.4859508037567135,
"epoch": 3.283507496592458,
"grad_norm": 1.0625,
"learning_rate": 0.00042830022283539863,
"loss": 5.2355,
"mean_token_accuracy": 0.17679316997528077,
"num_tokens": 7480977.0,
"step": 3615
},
{
"entropy": 5.478472805023193,
"epoch": 3.288050885960927,
"grad_norm": 1.140625,
"learning_rate": 0.0004280413334281623,
"loss": 5.2111,
"mean_token_accuracy": 0.18320189267396927,
"num_tokens": 7491758.0,
"step": 3620
},
{
"entropy": 5.44699068069458,
"epoch": 3.292594275329396,
"grad_norm": 1.2734375,
"learning_rate": 0.00042778206640664915,
"loss": 5.1721,
"mean_token_accuracy": 0.17932081073522568,
"num_tokens": 7500486.0,
"step": 3625
},
{
"entropy": 5.347209405899048,
"epoch": 3.2971376646978645,
"grad_norm": 1.1875,
"learning_rate": 0.0004275224224105748,
"loss": 5.1503,
"mean_token_accuracy": 0.18886389434337617,
"num_tokens": 7510371.0,
"step": 3630
},
{
"entropy": 5.463871717453003,
"epoch": 3.3016810540663335,
"grad_norm": 1.1953125,
"learning_rate": 0.00042726240208058484,
"loss": 5.1785,
"mean_token_accuracy": 0.1889471545815468,
"num_tokens": 7520697.0,
"step": 3635
},
{
"entropy": 5.396316289901733,
"epoch": 3.3062244434348025,
"grad_norm": 1.125,
"learning_rate": 0.0004270020060582538,
"loss": 5.1198,
"mean_token_accuracy": 0.18034988939762114,
"num_tokens": 7530757.0,
"step": 3640
},
{
"entropy": 5.421036720275879,
"epoch": 3.310767832803271,
"grad_norm": 1.09375,
"learning_rate": 0.00042674123498608285,
"loss": 5.1389,
"mean_token_accuracy": 0.18343761414289475,
"num_tokens": 7541810.0,
"step": 3645
},
{
"entropy": 5.456188058853149,
"epoch": 3.31531122217174,
"grad_norm": 1.1171875,
"learning_rate": 0.00042648008950749877,
"loss": 5.3251,
"mean_token_accuracy": 0.17192890346050263,
"num_tokens": 7552754.0,
"step": 3650
},
{
"entropy": 5.503060531616211,
"epoch": 3.319854611540209,
"grad_norm": 1.2421875,
"learning_rate": 0.000426218570266852,
"loss": 5.1136,
"mean_token_accuracy": 0.18422657251358032,
"num_tokens": 7562906.0,
"step": 3655
},
{
"entropy": 5.346383142471313,
"epoch": 3.324398000908678,
"grad_norm": 1.234375,
"learning_rate": 0.00042595667790941536,
"loss": 5.1335,
"mean_token_accuracy": 0.18094881922006606,
"num_tokens": 7574149.0,
"step": 3660
},
{
"entropy": 5.38492603302002,
"epoch": 3.328941390277147,
"grad_norm": 1.1484375,
"learning_rate": 0.00042569441308138204,
"loss": 5.1121,
"mean_token_accuracy": 0.1890406534075737,
"num_tokens": 7584610.0,
"step": 3665
},
{
"entropy": 5.361781358718872,
"epoch": 3.3334847796456155,
"grad_norm": 1.171875,
"learning_rate": 0.0004254317764298645,
"loss": 5.0798,
"mean_token_accuracy": 0.18382155895233154,
"num_tokens": 7594954.0,
"step": 3670
},
{
"entropy": 5.457009935379029,
"epoch": 3.3380281690140845,
"grad_norm": 1.1171875,
"learning_rate": 0.0004251687686028927,
"loss": 5.2082,
"mean_token_accuracy": 0.17161625623703003,
"num_tokens": 7606069.0,
"step": 3675
},
{
"entropy": 5.390203285217285,
"epoch": 3.3425715583825535,
"grad_norm": 1.2421875,
"learning_rate": 0.00042490539024941216,
"loss": 5.0619,
"mean_token_accuracy": 0.18886008411645888,
"num_tokens": 7615671.0,
"step": 3680
},
{
"entropy": 5.421680498123169,
"epoch": 3.347114947751022,
"grad_norm": 1.2734375,
"learning_rate": 0.00042464164201928286,
"loss": 5.2506,
"mean_token_accuracy": 0.1733468934893608,
"num_tokens": 7626740.0,
"step": 3685
},
{
"entropy": 5.463884925842285,
"epoch": 3.351658337119491,
"grad_norm": 1.28125,
"learning_rate": 0.00042437752456327734,
"loss": 5.1705,
"mean_token_accuracy": 0.184963558614254,
"num_tokens": 7636459.0,
"step": 3690
},
{
"entropy": 5.437850379943848,
"epoch": 3.35620172648796,
"grad_norm": 1.140625,
"learning_rate": 0.00042411303853307907,
"loss": 5.1275,
"mean_token_accuracy": 0.18153197765350343,
"num_tokens": 7646974.0,
"step": 3695
},
{
"entropy": 5.34959044456482,
"epoch": 3.360745115856429,
"grad_norm": 1.1796875,
"learning_rate": 0.0004238481845812811,
"loss": 5.0787,
"mean_token_accuracy": 0.1830618366599083,
"num_tokens": 7657372.0,
"step": 3700
},
{
"entropy": 5.36639723777771,
"epoch": 3.365288505224898,
"grad_norm": 1.1796875,
"learning_rate": 0.00042358296336138436,
"loss": 5.1186,
"mean_token_accuracy": 0.18323307037353515,
"num_tokens": 7668972.0,
"step": 3705
},
{
"entropy": 5.398306798934937,
"epoch": 3.3698318945933665,
"grad_norm": 1.4375,
"learning_rate": 0.0004233173755277956,
"loss": 5.1439,
"mean_token_accuracy": 0.1980360895395279,
"num_tokens": 7679742.0,
"step": 3710
},
{
"entropy": 5.379440689086914,
"epoch": 3.3743752839618355,
"grad_norm": 1.15625,
"learning_rate": 0.00042305142173582657,
"loss": 5.1714,
"mean_token_accuracy": 0.18570295423269273,
"num_tokens": 7690759.0,
"step": 3715
},
{
"entropy": 5.408399629592895,
"epoch": 3.3789186733303045,
"grad_norm": 1.2109375,
"learning_rate": 0.00042278510264169165,
"loss": 5.1306,
"mean_token_accuracy": 0.1834249660372734,
"num_tokens": 7700482.0,
"step": 3720
},
{
"entropy": 5.377882575988769,
"epoch": 3.3834620626987735,
"grad_norm": 1.2734375,
"learning_rate": 0.00042251841890250696,
"loss": 5.109,
"mean_token_accuracy": 0.180612188577652,
"num_tokens": 7710414.0,
"step": 3725
},
{
"entropy": 5.446986961364746,
"epoch": 3.388005452067242,
"grad_norm": 1.1484375,
"learning_rate": 0.000422251371176288,
"loss": 5.1509,
"mean_token_accuracy": 0.18371910303831102,
"num_tokens": 7721214.0,
"step": 3730
},
{
"entropy": 5.354454469680786,
"epoch": 3.392548841435711,
"grad_norm": 1.09375,
"learning_rate": 0.00042198396012194854,
"loss": 5.1148,
"mean_token_accuracy": 0.18716769963502883,
"num_tokens": 7732551.0,
"step": 3735
},
{
"entropy": 5.409590482711792,
"epoch": 3.39709223080418,
"grad_norm": 1.203125,
"learning_rate": 0.00042171618639929865,
"loss": 5.1997,
"mean_token_accuracy": 0.18170105814933776,
"num_tokens": 7742266.0,
"step": 3740
},
{
"entropy": 5.385374546051025,
"epoch": 3.401635620172649,
"grad_norm": 1.3828125,
"learning_rate": 0.00042144805066904345,
"loss": 5.0564,
"mean_token_accuracy": 0.1858584076166153,
"num_tokens": 7753152.0,
"step": 3745
},
{
"entropy": 5.367702198028565,
"epoch": 3.4061790095411175,
"grad_norm": 1.1171875,
"learning_rate": 0.00042117955359278125,
"loss": 5.1631,
"mean_token_accuracy": 0.18229499757289885,
"num_tokens": 7764420.0,
"step": 3750
},
{
"entropy": 5.408927059173584,
"epoch": 3.4107223989095865,
"grad_norm": 1.140625,
"learning_rate": 0.00042091069583300177,
"loss": 5.1164,
"mean_token_accuracy": 0.18572205007076265,
"num_tokens": 7776044.0,
"step": 3755
},
{
"entropy": 5.316444683074951,
"epoch": 3.4152657882780555,
"grad_norm": 1.2109375,
"learning_rate": 0.0004206414780530849,
"loss": 5.1807,
"mean_token_accuracy": 0.178596331179142,
"num_tokens": 7786737.0,
"step": 3760
},
{
"entropy": 5.359434318542481,
"epoch": 3.4198091776465245,
"grad_norm": 1.2890625,
"learning_rate": 0.0004203719009172987,
"loss": 5.0953,
"mean_token_accuracy": 0.1887235537171364,
"num_tokens": 7796974.0,
"step": 3765
},
{
"entropy": 5.440683364868164,
"epoch": 3.424352567014993,
"grad_norm": 1.1015625,
"learning_rate": 0.000420101965090798,
"loss": 5.2454,
"mean_token_accuracy": 0.1778610572218895,
"num_tokens": 7808221.0,
"step": 3770
},
{
"entropy": 5.34034686088562,
"epoch": 3.428895956383462,
"grad_norm": 1.109375,
"learning_rate": 0.0004198316712396226,
"loss": 5.0937,
"mean_token_accuracy": 0.18953122347593307,
"num_tokens": 7818422.0,
"step": 3775
},
{
"entropy": 5.413815927505493,
"epoch": 3.433439345751931,
"grad_norm": 1.1171875,
"learning_rate": 0.00041956102003069573,
"loss": 5.1069,
"mean_token_accuracy": 0.18646656721830368,
"num_tokens": 7828313.0,
"step": 3780
},
{
"entropy": 5.355943441390991,
"epoch": 3.4379827351204,
"grad_norm": 1.125,
"learning_rate": 0.0004192900121318223,
"loss": 5.1267,
"mean_token_accuracy": 0.19533473551273345,
"num_tokens": 7839493.0,
"step": 3785
},
{
"entropy": 5.428309345245362,
"epoch": 3.4425261244888685,
"grad_norm": 1.234375,
"learning_rate": 0.0004190186482116874,
"loss": 5.1391,
"mean_token_accuracy": 0.18329987972974776,
"num_tokens": 7849796.0,
"step": 3790
},
{
"entropy": 5.382693958282471,
"epoch": 3.4470695138573375,
"grad_norm": 1.1875,
"learning_rate": 0.0004187469289398545,
"loss": 5.2005,
"mean_token_accuracy": 0.18255136013031006,
"num_tokens": 7860865.0,
"step": 3795
},
{
"entropy": 5.326676702499389,
"epoch": 3.4516129032258065,
"grad_norm": 1.1484375,
"learning_rate": 0.00041847485498676385,
"loss": 5.1339,
"mean_token_accuracy": 0.18865247964859008,
"num_tokens": 7871144.0,
"step": 3800
},
{
"entropy": 5.363671588897705,
"epoch": 3.4561562925942755,
"grad_norm": 1.265625,
"learning_rate": 0.00041820242702373094,
"loss": 5.177,
"mean_token_accuracy": 0.18410495221614837,
"num_tokens": 7880916.0,
"step": 3805
},
{
"entropy": 5.352112531661987,
"epoch": 3.460699681962744,
"grad_norm": 1.3515625,
"learning_rate": 0.0004179296457229448,
"loss": 5.102,
"mean_token_accuracy": 0.18739551901817322,
"num_tokens": 7891813.0,
"step": 3810
},
{
"entropy": 5.43652114868164,
"epoch": 3.465243071331213,
"grad_norm": 1.234375,
"learning_rate": 0.0004176565117574659,
"loss": 5.1584,
"mean_token_accuracy": 0.1788448214530945,
"num_tokens": 7902697.0,
"step": 3815
},
{
"entropy": 5.4100196838378904,
"epoch": 3.469786460699682,
"grad_norm": 1.0625,
"learning_rate": 0.0004173830258012255,
"loss": 5.1333,
"mean_token_accuracy": 0.18766922354698182,
"num_tokens": 7912777.0,
"step": 3820
},
{
"entropy": 5.402464246749878,
"epoch": 3.474329850068151,
"grad_norm": 1.2265625,
"learning_rate": 0.0004171091885290226,
"loss": 5.1777,
"mean_token_accuracy": 0.1887985497713089,
"num_tokens": 7923255.0,
"step": 3825
},
{
"entropy": 5.398044681549072,
"epoch": 3.4788732394366195,
"grad_norm": 1.203125,
"learning_rate": 0.0004168350006165238,
"loss": 5.0666,
"mean_token_accuracy": 0.183932563662529,
"num_tokens": 7934268.0,
"step": 3830
},
{
"entropy": 5.300931024551391,
"epoch": 3.4834166288050885,
"grad_norm": 1.2265625,
"learning_rate": 0.00041656046274026034,
"loss": 5.079,
"mean_token_accuracy": 0.19050655663013458,
"num_tokens": 7944526.0,
"step": 3835
},
{
"entropy": 5.368102169036865,
"epoch": 3.4879600181735575,
"grad_norm": 1.2265625,
"learning_rate": 0.0004162855755776272,
"loss": 5.1922,
"mean_token_accuracy": 0.1843766525387764,
"num_tokens": 7954512.0,
"step": 3840
},
{
"entropy": 5.393799591064453,
"epoch": 3.4925034075420265,
"grad_norm": 1.2109375,
"learning_rate": 0.00041601033980688106,
"loss": 5.197,
"mean_token_accuracy": 0.174209626019001,
"num_tokens": 7965703.0,
"step": 3845
},
{
"entropy": 5.46270227432251,
"epoch": 3.497046796910495,
"grad_norm": 1.21875,
"learning_rate": 0.0004157347561071389,
"loss": 5.1816,
"mean_token_accuracy": 0.1791876032948494,
"num_tokens": 7977822.0,
"step": 3850
},
{
"entropy": 5.3717348098754885,
"epoch": 3.501590186278964,
"grad_norm": 1.2109375,
"learning_rate": 0.00041545882515837604,
"loss": 5.095,
"mean_token_accuracy": 0.17957233339548112,
"num_tokens": 7988318.0,
"step": 3855
},
{
"entropy": 5.365765142440796,
"epoch": 3.506133575647433,
"grad_norm": 1.21875,
"learning_rate": 0.00041518254764142477,
"loss": 5.0595,
"mean_token_accuracy": 0.1865748345851898,
"num_tokens": 7998076.0,
"step": 3860
},
{
"entropy": 5.40487961769104,
"epoch": 3.510676965015902,
"grad_norm": 1.21875,
"learning_rate": 0.00041490592423797226,
"loss": 5.1707,
"mean_token_accuracy": 0.18892305344343185,
"num_tokens": 8007913.0,
"step": 3865
},
{
"entropy": 5.381358432769775,
"epoch": 3.515220354384371,
"grad_norm": 1.1953125,
"learning_rate": 0.0004146289556305594,
"loss": 5.1313,
"mean_token_accuracy": 0.18112607449293136,
"num_tokens": 8018340.0,
"step": 3870
},
{
"entropy": 5.460349655151367,
"epoch": 3.5197637437528395,
"grad_norm": 1.2890625,
"learning_rate": 0.0004143516425025786,
"loss": 5.1704,
"mean_token_accuracy": 0.18166618645191193,
"num_tokens": 8028312.0,
"step": 3875
},
{
"entropy": 5.4275665283203125,
"epoch": 3.5243071331213085,
"grad_norm": 1.2578125,
"learning_rate": 0.0004140739855382724,
"loss": 5.1789,
"mean_token_accuracy": 0.18710415810346603,
"num_tokens": 8038580.0,
"step": 3880
},
{
"entropy": 5.443336391448975,
"epoch": 3.5288505224897775,
"grad_norm": 1.265625,
"learning_rate": 0.00041379598542273197,
"loss": 5.1392,
"mean_token_accuracy": 0.18372284322977067,
"num_tokens": 8048365.0,
"step": 3885
},
{
"entropy": 5.3985734462738035,
"epoch": 3.533393911858246,
"grad_norm": 1.1875,
"learning_rate": 0.00041351764284189464,
"loss": 5.1813,
"mean_token_accuracy": 0.18132270276546478,
"num_tokens": 8059638.0,
"step": 3890
},
{
"entropy": 5.431989479064941,
"epoch": 3.537937301226715,
"grad_norm": 1.125,
"learning_rate": 0.0004132389584825433,
"loss": 5.2509,
"mean_token_accuracy": 0.17716899663209915,
"num_tokens": 8070993.0,
"step": 3895
},
{
"entropy": 5.341916847229004,
"epoch": 3.542480690595184,
"grad_norm": 1.234375,
"learning_rate": 0.00041295993303230383,
"loss": 5.0945,
"mean_token_accuracy": 0.1890174850821495,
"num_tokens": 8081203.0,
"step": 3900
},
{
"entropy": 5.352109670639038,
"epoch": 3.547024079963653,
"grad_norm": 1.1953125,
"learning_rate": 0.0004126805671796438,
"loss": 5.1105,
"mean_token_accuracy": 0.18251689076423644,
"num_tokens": 8091564.0,
"step": 3905
},
{
"entropy": 5.405901670455933,
"epoch": 3.551567469332122,
"grad_norm": 1.3203125,
"learning_rate": 0.0004124008616138707,
"loss": 5.0861,
"mean_token_accuracy": 0.1867015466094017,
"num_tokens": 8100675.0,
"step": 3910
},
{
"entropy": 5.359062957763672,
"epoch": 3.5561108587005905,
"grad_norm": 1.265625,
"learning_rate": 0.00041212081702513015,
"loss": 5.1494,
"mean_token_accuracy": 0.18618002086877822,
"num_tokens": 8110605.0,
"step": 3915
},
{
"entropy": 5.35721526145935,
"epoch": 3.5606542480690595,
"grad_norm": 1.1875,
"learning_rate": 0.0004118404341044044,
"loss": 5.1879,
"mean_token_accuracy": 0.18077319115400314,
"num_tokens": 8120910.0,
"step": 3920
},
{
"entropy": 5.430187463760376,
"epoch": 3.5651976374375285,
"grad_norm": 1.1953125,
"learning_rate": 0.00041155971354351034,
"loss": 5.1454,
"mean_token_accuracy": 0.18601136207580565,
"num_tokens": 8131827.0,
"step": 3925
},
{
"entropy": 5.369317722320557,
"epoch": 3.569741026805997,
"grad_norm": 1.21875,
"learning_rate": 0.00041127865603509815,
"loss": 5.1206,
"mean_token_accuracy": 0.182552470266819,
"num_tokens": 8141785.0,
"step": 3930
},
{
"entropy": 5.406180667877197,
"epoch": 3.574284416174466,
"grad_norm": 1.234375,
"learning_rate": 0.0004109972622726493,
"loss": 5.1421,
"mean_token_accuracy": 0.18374724984169005,
"num_tokens": 8151732.0,
"step": 3935
},
{
"entropy": 5.423010540008545,
"epoch": 3.578827805542935,
"grad_norm": 1.15625,
"learning_rate": 0.00041071553295047485,
"loss": 5.231,
"mean_token_accuracy": 0.17640620619058608,
"num_tokens": 8162334.0,
"step": 3940
},
{
"entropy": 5.349823522567749,
"epoch": 3.583371194911404,
"grad_norm": 1.234375,
"learning_rate": 0.0004104334687637139,
"loss": 5.1087,
"mean_token_accuracy": 0.18098020553588867,
"num_tokens": 8172810.0,
"step": 3945
},
{
"entropy": 5.382225418090821,
"epoch": 3.587914584279873,
"grad_norm": 1.1328125,
"learning_rate": 0.00041015107040833187,
"loss": 5.1559,
"mean_token_accuracy": 0.18009609282016753,
"num_tokens": 8183240.0,
"step": 3950
},
{
"entropy": 5.378968524932861,
"epoch": 3.5924579736483415,
"grad_norm": 1.2265625,
"learning_rate": 0.0004098683385811185,
"loss": 5.0912,
"mean_token_accuracy": 0.1828031212091446,
"num_tokens": 8193824.0,
"step": 3955
},
{
"entropy": 5.337375593185425,
"epoch": 3.5970013630168105,
"grad_norm": 1.40625,
"learning_rate": 0.00040958527397968666,
"loss": 5.0792,
"mean_token_accuracy": 0.19234203398227692,
"num_tokens": 8203792.0,
"step": 3960
},
{
"entropy": 5.441889095306396,
"epoch": 3.6015447523852795,
"grad_norm": 1.1875,
"learning_rate": 0.0004093018773024699,
"loss": 5.2588,
"mean_token_accuracy": 0.18046560287475585,
"num_tokens": 8214436.0,
"step": 3965
},
{
"entropy": 5.417685651779175,
"epoch": 3.606088141753748,
"grad_norm": 1.109375,
"learning_rate": 0.0004090181492487215,
"loss": 5.1094,
"mean_token_accuracy": 0.17917166650295258,
"num_tokens": 8224518.0,
"step": 3970
},
{
"entropy": 5.447009038925171,
"epoch": 3.610631531122217,
"grad_norm": 1.28125,
"learning_rate": 0.00040873409051851223,
"loss": 5.2718,
"mean_token_accuracy": 0.17240029126405715,
"num_tokens": 8235234.0,
"step": 3975
},
{
"entropy": 5.373015117645264,
"epoch": 3.615174920490686,
"grad_norm": 1.2734375,
"learning_rate": 0.0004084497018127286,
"loss": 5.1666,
"mean_token_accuracy": 0.17949411123991013,
"num_tokens": 8246612.0,
"step": 3980
},
{
"entropy": 5.4462532043457035,
"epoch": 3.619718309859155,
"grad_norm": 1.3359375,
"learning_rate": 0.0004081649838330717,
"loss": 5.1066,
"mean_token_accuracy": 0.18945149928331376,
"num_tokens": 8256720.0,
"step": 3985
},
{
"entropy": 5.38092155456543,
"epoch": 3.624261699227624,
"grad_norm": 1.1484375,
"learning_rate": 0.00040787993728205464,
"loss": 5.1781,
"mean_token_accuracy": 0.18331068456172944,
"num_tokens": 8267666.0,
"step": 3990
},
{
"entropy": 5.434221839904785,
"epoch": 3.6288050885960925,
"grad_norm": 1.296875,
"learning_rate": 0.00040759456286300154,
"loss": 5.1543,
"mean_token_accuracy": 0.1841464787721634,
"num_tokens": 8277882.0,
"step": 3995
},
{
"entropy": 5.3370462417602536,
"epoch": 3.6333484779645615,
"grad_norm": 1.3359375,
"learning_rate": 0.00040730886128004544,
"loss": 5.0867,
"mean_token_accuracy": 0.18468646705150604,
"num_tokens": 8286968.0,
"step": 4000
},
{
"epoch": 3.6333484779645615,
"eval_entropy": 5.275365178529606,
"eval_loss": 5.815081596374512,
"eval_mean_token_accuracy": 0.16117590587846067,
"eval_num_tokens": 8286968.0,
"eval_runtime": 2.6037,
"eval_samples_per_second": 1319.277,
"eval_steps_per_second": 165.15,
"step": 4000
},
{
"entropy": 5.36260666847229,
"epoch": 3.6378918673330305,
"grad_norm": 1.171875,
"learning_rate": 0.0004070228332381266,
"loss": 5.1551,
"mean_token_accuracy": 0.18651724606752396,
"num_tokens": 8297154.0,
"step": 4005
},
{
"entropy": 5.402426290512085,
"epoch": 3.6424352567014995,
"grad_norm": 1.2265625,
"learning_rate": 0.0004067364794429908,
"loss": 5.1334,
"mean_token_accuracy": 0.1808108851313591,
"num_tokens": 8307954.0,
"step": 4010
},
{
"entropy": 5.297863006591797,
"epoch": 3.646978646069968,
"grad_norm": 1.2734375,
"learning_rate": 0.0004064498006011874,
"loss": 5.1362,
"mean_token_accuracy": 0.18583769649267196,
"num_tokens": 8317714.0,
"step": 4015
},
{
"entropy": 5.329110097885132,
"epoch": 3.651522035438437,
"grad_norm": 1.171875,
"learning_rate": 0.00040616279742006813,
"loss": 5.1335,
"mean_token_accuracy": 0.19293444007635116,
"num_tokens": 8327963.0,
"step": 4020
},
{
"entropy": 5.378097915649414,
"epoch": 3.656065424806906,
"grad_norm": 1.2265625,
"learning_rate": 0.00040587547060778466,
"loss": 5.0621,
"mean_token_accuracy": 0.19223445057868957,
"num_tokens": 8338190.0,
"step": 4025
},
{
"entropy": 5.363200616836548,
"epoch": 3.660608814175375,
"grad_norm": 1.2109375,
"learning_rate": 0.0004055878208732874,
"loss": 5.2063,
"mean_token_accuracy": 0.1818956181406975,
"num_tokens": 8348564.0,
"step": 4030
},
{
"entropy": 5.400208139419556,
"epoch": 3.665152203543844,
"grad_norm": 1.1640625,
"learning_rate": 0.00040529984892632363,
"loss": 5.119,
"mean_token_accuracy": 0.18295660465955735,
"num_tokens": 8359463.0,
"step": 4035
},
{
"entropy": 5.385143804550171,
"epoch": 3.6696955929123125,
"grad_norm": 1.3359375,
"learning_rate": 0.0004050115554774353,
"loss": 5.1842,
"mean_token_accuracy": 0.18372150361537934,
"num_tokens": 8369676.0,
"step": 4040
},
{
"entropy": 5.356810712814331,
"epoch": 3.6742389822807815,
"grad_norm": 1.21875,
"learning_rate": 0.0004047229412379578,
"loss": 5.0884,
"mean_token_accuracy": 0.1979552373290062,
"num_tokens": 8379579.0,
"step": 4045
},
{
"entropy": 5.396504497528076,
"epoch": 3.6787823716492505,
"grad_norm": 1.2890625,
"learning_rate": 0.0004044340069200183,
"loss": 5.1726,
"mean_token_accuracy": 0.1888376295566559,
"num_tokens": 8390164.0,
"step": 4050
},
{
"entropy": 5.315797758102417,
"epoch": 3.683325761017719,
"grad_norm": 1.1953125,
"learning_rate": 0.00040414475323653326,
"loss": 5.0715,
"mean_token_accuracy": 0.19321446269750595,
"num_tokens": 8399220.0,
"step": 4055
},
{
"entropy": 5.363904237747192,
"epoch": 3.687869150386188,
"grad_norm": 1.1875,
"learning_rate": 0.0004038551809012076,
"loss": 5.135,
"mean_token_accuracy": 0.18331129997968673,
"num_tokens": 8409488.0,
"step": 4060
},
{
"entropy": 5.3659923553466795,
"epoch": 3.692412539754657,
"grad_norm": 1.1875,
"learning_rate": 0.0004035652906285322,
"loss": 5.0805,
"mean_token_accuracy": 0.19802087247371675,
"num_tokens": 8418817.0,
"step": 4065
},
{
"entropy": 5.340572166442871,
"epoch": 3.696955929123126,
"grad_norm": 1.0859375,
"learning_rate": 0.00040327508313378244,
"loss": 5.193,
"mean_token_accuracy": 0.1814069166779518,
"num_tokens": 8428742.0,
"step": 4070
},
{
"entropy": 5.402520608901978,
"epoch": 3.701499318491595,
"grad_norm": 1.3828125,
"learning_rate": 0.0004029845591330164,
"loss": 5.0732,
"mean_token_accuracy": 0.19352974444627763,
"num_tokens": 8438016.0,
"step": 4075
},
{
"entropy": 5.450796318054199,
"epoch": 3.7060427078600635,
"grad_norm": 1.203125,
"learning_rate": 0.00040269371934307336,
"loss": 5.1874,
"mean_token_accuracy": 0.1839870646595955,
"num_tokens": 8448428.0,
"step": 4080
},
{
"entropy": 5.304365015029907,
"epoch": 3.7105860972285325,
"grad_norm": 1.1875,
"learning_rate": 0.00040240256448157146,
"loss": 5.0429,
"mean_token_accuracy": 0.19023211151361466,
"num_tokens": 8458444.0,
"step": 4085
},
{
"entropy": 5.326574897766113,
"epoch": 3.7151294865970015,
"grad_norm": 1.2265625,
"learning_rate": 0.0004021110952669065,
"loss": 5.1862,
"mean_token_accuracy": 0.18667993694543839,
"num_tokens": 8467721.0,
"step": 4090
},
{
"entropy": 5.311043930053711,
"epoch": 3.71967287596547,
"grad_norm": 1.234375,
"learning_rate": 0.0004018193124182497,
"loss": 5.09,
"mean_token_accuracy": 0.1905745893716812,
"num_tokens": 8477484.0,
"step": 4095
},
{
"entropy": 5.364870262145996,
"epoch": 3.724216265333939,
"grad_norm": 1.21875,
"learning_rate": 0.0004015272166555462,
"loss": 5.1024,
"mean_token_accuracy": 0.1870442122220993,
"num_tokens": 8486902.0,
"step": 4100
},
{
"entropy": 5.400627326965332,
"epoch": 3.728759654702408,
"grad_norm": 1.3671875,
"learning_rate": 0.0004012348086995132,
"loss": 5.0787,
"mean_token_accuracy": 0.18339478671550752,
"num_tokens": 8497626.0,
"step": 4105
},
{
"entropy": 5.344414806365966,
"epoch": 3.733303044070877,
"grad_norm": 1.2265625,
"learning_rate": 0.0004009420892716385,
"loss": 5.1765,
"mean_token_accuracy": 0.17410700619220734,
"num_tokens": 8507518.0,
"step": 4110
},
{
"entropy": 5.342016506195068,
"epoch": 3.737846433439346,
"grad_norm": 1.2109375,
"learning_rate": 0.0004006490590941781,
"loss": 5.1561,
"mean_token_accuracy": 0.18127464950084687,
"num_tokens": 8517835.0,
"step": 4115
},
{
"entropy": 5.436113357543945,
"epoch": 3.7423898228078145,
"grad_norm": 1.1796875,
"learning_rate": 0.0004003557188901547,
"loss": 5.1783,
"mean_token_accuracy": 0.1828899011015892,
"num_tokens": 8527552.0,
"step": 4120
},
{
"entropy": 5.386102914810181,
"epoch": 3.7469332121762835,
"grad_norm": 1.21875,
"learning_rate": 0.0004000620693833562,
"loss": 5.1716,
"mean_token_accuracy": 0.1865544378757477,
"num_tokens": 8538082.0,
"step": 4125
},
{
"entropy": 5.34010329246521,
"epoch": 3.7514766015447525,
"grad_norm": 1.1875,
"learning_rate": 0.0003997681112983335,
"loss": 5.1016,
"mean_token_accuracy": 0.18389442563056946,
"num_tokens": 8548558.0,
"step": 4130
},
{
"entropy": 5.33505163192749,
"epoch": 3.756019990913221,
"grad_norm": 1.1640625,
"learning_rate": 0.00039947384536039905,
"loss": 5.1149,
"mean_token_accuracy": 0.1849595785140991,
"num_tokens": 8559406.0,
"step": 4135
},
{
"entropy": 5.40301284790039,
"epoch": 3.76056338028169,
"grad_norm": 1.1953125,
"learning_rate": 0.00039917927229562474,
"loss": 5.1542,
"mean_token_accuracy": 0.1759909585118294,
"num_tokens": 8569519.0,
"step": 4140
},
{
"entropy": 5.328449869155884,
"epoch": 3.765106769650159,
"grad_norm": 1.265625,
"learning_rate": 0.00039888439283084035,
"loss": 5.174,
"mean_token_accuracy": 0.18337784558534623,
"num_tokens": 8579658.0,
"step": 4145
},
{
"entropy": 5.305296516418457,
"epoch": 3.769650159018628,
"grad_norm": 1.1953125,
"learning_rate": 0.00039858920769363164,
"loss": 5.0837,
"mean_token_accuracy": 0.18633795380592347,
"num_tokens": 8589681.0,
"step": 4150
},
{
"entropy": 5.396807241439819,
"epoch": 3.774193548387097,
"grad_norm": 1.328125,
"learning_rate": 0.0003982937176123386,
"loss": 5.1494,
"mean_token_accuracy": 0.18492938131093978,
"num_tokens": 8600310.0,
"step": 4155
},
{
"entropy": 5.38708086013794,
"epoch": 3.7787369377555655,
"grad_norm": 1.265625,
"learning_rate": 0.00039799792331605364,
"loss": 5.1713,
"mean_token_accuracy": 0.18235747665166854,
"num_tokens": 8610920.0,
"step": 4160
},
{
"entropy": 5.310511922836303,
"epoch": 3.7832803271240345,
"grad_norm": 1.3203125,
"learning_rate": 0.00039770182553461965,
"loss": 5.0937,
"mean_token_accuracy": 0.1937661811709404,
"num_tokens": 8620536.0,
"step": 4165
},
{
"entropy": 5.348018980026245,
"epoch": 3.7878237164925035,
"grad_norm": 1.2109375,
"learning_rate": 0.00039740542499862866,
"loss": 5.1858,
"mean_token_accuracy": 0.18694871068000793,
"num_tokens": 8631647.0,
"step": 4170
},
{
"entropy": 5.3819011688232425,
"epoch": 3.792367105860972,
"grad_norm": 1.015625,
"learning_rate": 0.0003971087224394194,
"loss": 5.1514,
"mean_token_accuracy": 0.18549743592739104,
"num_tokens": 8642659.0,
"step": 4175
},
{
"entropy": 5.351566934585572,
"epoch": 3.796910495229441,
"grad_norm": 1.28125,
"learning_rate": 0.0003968117185890761,
"loss": 5.1357,
"mean_token_accuracy": 0.19184285998344422,
"num_tokens": 8651775.0,
"step": 4180
},
{
"entropy": 5.342884874343872,
"epoch": 3.80145388459791,
"grad_norm": 1.1953125,
"learning_rate": 0.00039651441418042606,
"loss": 5.1215,
"mean_token_accuracy": 0.18081242740154266,
"num_tokens": 8662520.0,
"step": 4185
},
{
"entropy": 5.387826061248779,
"epoch": 3.805997273966379,
"grad_norm": 1.25,
"learning_rate": 0.0003962168099470384,
"loss": 5.135,
"mean_token_accuracy": 0.18578024655580522,
"num_tokens": 8672630.0,
"step": 4190
},
{
"entropy": 5.416587924957275,
"epoch": 3.810540663334848,
"grad_norm": 1.2265625,
"learning_rate": 0.000395918906623222,
"loss": 5.2363,
"mean_token_accuracy": 0.17437006235122682,
"num_tokens": 8683348.0,
"step": 4195
},
{
"entropy": 5.327307558059692,
"epoch": 3.8150840527033165,
"grad_norm": 1.328125,
"learning_rate": 0.0003956207049440238,
"loss": 5.0578,
"mean_token_accuracy": 0.19229240864515304,
"num_tokens": 8693423.0,
"step": 4200
},
{
"entropy": 5.366624021530152,
"epoch": 3.8196274420717855,
"grad_norm": 1.25,
"learning_rate": 0.0003953222056452265,
"loss": 5.1377,
"mean_token_accuracy": 0.1866310030221939,
"num_tokens": 8703766.0,
"step": 4205
},
{
"entropy": 5.41118426322937,
"epoch": 3.8241708314402545,
"grad_norm": 1.15625,
"learning_rate": 0.0003950234094633479,
"loss": 5.2263,
"mean_token_accuracy": 0.18283589482307433,
"num_tokens": 8715330.0,
"step": 4210
},
{
"entropy": 5.352522182464599,
"epoch": 3.828714220808723,
"grad_norm": 1.25,
"learning_rate": 0.00039472431713563746,
"loss": 5.1315,
"mean_token_accuracy": 0.18953316062688827,
"num_tokens": 8725732.0,
"step": 4215
},
{
"entropy": 5.412807559967041,
"epoch": 3.833257610177192,
"grad_norm": 1.34375,
"learning_rate": 0.000394424929400076,
"loss": 5.0904,
"mean_token_accuracy": 0.19398150146007537,
"num_tokens": 8736377.0,
"step": 4220
},
{
"entropy": 5.409627294540405,
"epoch": 3.837800999545661,
"grad_norm": 1.203125,
"learning_rate": 0.0003941252469953729,
"loss": 5.1224,
"mean_token_accuracy": 0.1868395298719406,
"num_tokens": 8746991.0,
"step": 4225
},
{
"entropy": 5.3006878852844235,
"epoch": 3.84234438891413,
"grad_norm": 1.234375,
"learning_rate": 0.0003938252706609647,
"loss": 5.1445,
"mean_token_accuracy": 0.18721779584884643,
"num_tokens": 8756920.0,
"step": 4230
},
{
"entropy": 5.365189743041992,
"epoch": 3.846887778282599,
"grad_norm": 1.1328125,
"learning_rate": 0.0003935250011370132,
"loss": 5.1969,
"mean_token_accuracy": 0.17555955052375793,
"num_tokens": 8768617.0,
"step": 4235
},
{
"entropy": 5.400828552246094,
"epoch": 3.851431167651068,
"grad_norm": 1.1328125,
"learning_rate": 0.0003932244391644037,
"loss": 5.125,
"mean_token_accuracy": 0.196824848651886,
"num_tokens": 8778427.0,
"step": 4240
},
{
"entropy": 5.309160661697388,
"epoch": 3.8559745570195365,
"grad_norm": 1.1328125,
"learning_rate": 0.0003929235854847428,
"loss": 5.1806,
"mean_token_accuracy": 0.18466379791498183,
"num_tokens": 8789093.0,
"step": 4245
},
{
"entropy": 5.349613809585572,
"epoch": 3.8605179463880055,
"grad_norm": 1.15625,
"learning_rate": 0.00039262244084035724,
"loss": 5.1234,
"mean_token_accuracy": 0.17925288081169127,
"num_tokens": 8799832.0,
"step": 4250
},
{
"entropy": 5.377014398574829,
"epoch": 3.8650613357564745,
"grad_norm": 1.2734375,
"learning_rate": 0.00039232100597429143,
"loss": 5.1468,
"mean_token_accuracy": 0.18501030504703522,
"num_tokens": 8810149.0,
"step": 4255
},
{
"entropy": 5.385255002975464,
"epoch": 3.869604725124943,
"grad_norm": 1.2734375,
"learning_rate": 0.00039201928163030584,
"loss": 5.175,
"mean_token_accuracy": 0.18584203273057937,
"num_tokens": 8820087.0,
"step": 4260
},
{
"entropy": 5.39076566696167,
"epoch": 3.874148114493412,
"grad_norm": 1.203125,
"learning_rate": 0.00039171726855287544,
"loss": 5.2088,
"mean_token_accuracy": 0.18140177875757219,
"num_tokens": 8830959.0,
"step": 4265
},
{
"entropy": 5.405259656906128,
"epoch": 3.878691503861881,
"grad_norm": 1.2109375,
"learning_rate": 0.00039141496748718735,
"loss": 5.1457,
"mean_token_accuracy": 0.18428434282541276,
"num_tokens": 8842522.0,
"step": 4270
},
{
"entropy": 5.313390398025513,
"epoch": 3.88323489323035,
"grad_norm": 1.21875,
"learning_rate": 0.0003911123791791395,
"loss": 5.0596,
"mean_token_accuracy": 0.19310649782419204,
"num_tokens": 8852521.0,
"step": 4275
},
{
"entropy": 5.392964029312134,
"epoch": 3.887778282598819,
"grad_norm": 1.2734375,
"learning_rate": 0.0003908095043753384,
"loss": 5.1786,
"mean_token_accuracy": 0.17746835798025132,
"num_tokens": 8862156.0,
"step": 4280
},
{
"entropy": 5.379548358917236,
"epoch": 3.8923216719672875,
"grad_norm": 1.1796875,
"learning_rate": 0.00039050634382309753,
"loss": 5.0619,
"mean_token_accuracy": 0.18345230221748351,
"num_tokens": 8872772.0,
"step": 4285
},
{
"entropy": 5.380292177200317,
"epoch": 3.8968650613357565,
"grad_norm": 1.265625,
"learning_rate": 0.0003902028982704355,
"loss": 5.1369,
"mean_token_accuracy": 0.19035902619361877,
"num_tokens": 8883229.0,
"step": 4290
},
{
"entropy": 5.329371452331543,
"epoch": 3.9014084507042255,
"grad_norm": 1.203125,
"learning_rate": 0.0003898991684660738,
"loss": 5.1214,
"mean_token_accuracy": 0.18369243741035463,
"num_tokens": 8892403.0,
"step": 4295
},
{
"entropy": 5.302772951126099,
"epoch": 3.905951840072694,
"grad_norm": 1.21875,
"learning_rate": 0.0003895951551594358,
"loss": 5.134,
"mean_token_accuracy": 0.18088584542274475,
"num_tokens": 8902617.0,
"step": 4300
},
{
"entropy": 5.443783187866211,
"epoch": 3.910495229441163,
"grad_norm": 1.2265625,
"learning_rate": 0.000389290859100644,
"loss": 5.1628,
"mean_token_accuracy": 0.1894724652171135,
"num_tokens": 8913099.0,
"step": 4305
},
{
"entropy": 5.305198621749878,
"epoch": 3.915038618809632,
"grad_norm": 1.3359375,
"learning_rate": 0.0003889862810405187,
"loss": 5.0674,
"mean_token_accuracy": 0.18696921020746232,
"num_tokens": 8922469.0,
"step": 4310
},
{
"entropy": 5.365487718582154,
"epoch": 3.919582008178101,
"grad_norm": 1.1953125,
"learning_rate": 0.000388681421730576,
"loss": 5.1627,
"mean_token_accuracy": 0.19289431869983673,
"num_tokens": 8933261.0,
"step": 4315
},
{
"entropy": 5.360935020446777,
"epoch": 3.92412539754657,
"grad_norm": 1.3125,
"learning_rate": 0.0003883762819230259,
"loss": 5.1675,
"mean_token_accuracy": 0.18829834163188935,
"num_tokens": 8943873.0,
"step": 4320
},
{
"entropy": 5.3426697731018065,
"epoch": 3.9286687869150385,
"grad_norm": 1.296875,
"learning_rate": 0.00038807086237077057,
"loss": 5.0597,
"mean_token_accuracy": 0.2000756859779358,
"num_tokens": 8953802.0,
"step": 4325
},
{
"entropy": 5.405356502532959,
"epoch": 3.9332121762835075,
"grad_norm": 1.1484375,
"learning_rate": 0.00038776516382740237,
"loss": 5.2446,
"mean_token_accuracy": 0.18250361382961272,
"num_tokens": 8964309.0,
"step": 4330
},
{
"entropy": 5.359420919418335,
"epoch": 3.9377555656519765,
"grad_norm": 1.3203125,
"learning_rate": 0.00038745918704720205,
"loss": 5.1218,
"mean_token_accuracy": 0.18553209751844407,
"num_tokens": 8973611.0,
"step": 4335
},
{
"entropy": 5.346977376937867,
"epoch": 3.942298955020445,
"grad_norm": 1.046875,
"learning_rate": 0.00038715293278513693,
"loss": 5.1541,
"mean_token_accuracy": 0.1784630611538887,
"num_tokens": 8985783.0,
"step": 4340
},
{
"entropy": 5.404772186279297,
"epoch": 3.946842344388914,
"grad_norm": 1.0546875,
"learning_rate": 0.00038684640179685896,
"loss": 5.2049,
"mean_token_accuracy": 0.17611255198717118,
"num_tokens": 8996736.0,
"step": 4345
},
{
"entropy": 5.408973741531372,
"epoch": 3.951385733757383,
"grad_norm": 1.28125,
"learning_rate": 0.0003865395948387029,
"loss": 5.277,
"mean_token_accuracy": 0.17597326636314392,
"num_tokens": 9007635.0,
"step": 4350
},
{
"entropy": 5.357184791564942,
"epoch": 3.955929123125852,
"grad_norm": 1.140625,
"learning_rate": 0.00038623251266768437,
"loss": 5.1053,
"mean_token_accuracy": 0.18413132429122925,
"num_tokens": 9017892.0,
"step": 4355
},
{
"entropy": 5.379299783706665,
"epoch": 3.960472512494321,
"grad_norm": 1.15625,
"learning_rate": 0.00038592515604149815,
"loss": 5.1501,
"mean_token_accuracy": 0.184952974319458,
"num_tokens": 9027594.0,
"step": 4360
},
{
"entropy": 5.384825992584228,
"epoch": 3.9650159018627895,
"grad_norm": 1.21875,
"learning_rate": 0.0003856175257185162,
"loss": 5.1711,
"mean_token_accuracy": 0.17956896871328354,
"num_tokens": 9037083.0,
"step": 4365
},
{
"entropy": 5.351587915420533,
"epoch": 3.9695592912312585,
"grad_norm": 1.1796875,
"learning_rate": 0.00038530962245778564,
"loss": 5.1059,
"mean_token_accuracy": 0.19460697919130326,
"num_tokens": 9047176.0,
"step": 4370
},
{
"entropy": 5.362721967697143,
"epoch": 3.9741026805997275,
"grad_norm": 1.2265625,
"learning_rate": 0.0003850014470190272,
"loss": 5.0926,
"mean_token_accuracy": 0.18575462996959685,
"num_tokens": 9057810.0,
"step": 4375
},
{
"entropy": 5.40713586807251,
"epoch": 3.978646069968196,
"grad_norm": 1.25,
"learning_rate": 0.0003846930001626332,
"loss": 5.2032,
"mean_token_accuracy": 0.1816106617450714,
"num_tokens": 9067765.0,
"step": 4380
},
{
"entropy": 5.461988306045532,
"epoch": 3.983189459336665,
"grad_norm": 1.1796875,
"learning_rate": 0.0003843842826496656,
"loss": 5.1545,
"mean_token_accuracy": 0.18028418719768524,
"num_tokens": 9078796.0,
"step": 4385
},
{
"entropy": 5.391492509841919,
"epoch": 3.987732848705134,
"grad_norm": 1.3515625,
"learning_rate": 0.000384075295241854,
"loss": 5.0911,
"mean_token_accuracy": 0.19016141444444656,
"num_tokens": 9088368.0,
"step": 4390
},
{
"entropy": 5.312350034713745,
"epoch": 3.992276238073603,
"grad_norm": 1.1953125,
"learning_rate": 0.0003837660387015944,
"loss": 5.0629,
"mean_token_accuracy": 0.18586185127496718,
"num_tokens": 9099082.0,
"step": 4395
},
{
"entropy": 5.360974550247192,
"epoch": 3.996819627442072,
"grad_norm": 1.1484375,
"learning_rate": 0.0003834565137919463,
"loss": 5.1951,
"mean_token_accuracy": 0.18599987030029297,
"num_tokens": 9110543.0,
"step": 4400
},
{
"entropy": 5.390410370296902,
"epoch": 4.000908677873694,
"grad_norm": 1.234375,
"learning_rate": 0.0003831467212766318,
"loss": 5.1053,
"mean_token_accuracy": 0.18587587608231437,
"num_tokens": 9119347.0,
"step": 4405
},
{
"entropy": 5.319525623321534,
"epoch": 4.005452067242163,
"grad_norm": 1.1875,
"learning_rate": 0.0003828366619200331,
"loss": 4.7917,
"mean_token_accuracy": 0.2054760605096817,
"num_tokens": 9129154.0,
"step": 4410
},
{
"entropy": 5.297416019439697,
"epoch": 4.009995456610632,
"grad_norm": 1.1640625,
"learning_rate": 0.0003825263364871908,
"loss": 4.8158,
"mean_token_accuracy": 0.2055535674095154,
"num_tokens": 9139877.0,
"step": 4415
},
{
"entropy": 5.306910133361816,
"epoch": 4.0145388459791,
"grad_norm": 1.2265625,
"learning_rate": 0.0003822157457438022,
"loss": 4.8859,
"mean_token_accuracy": 0.20077282786369324,
"num_tokens": 9150508.0,
"step": 4420
},
{
"entropy": 5.322200441360474,
"epoch": 4.019082235347569,
"grad_norm": 1.1875,
"learning_rate": 0.00038190489045621893,
"loss": 4.8307,
"mean_token_accuracy": 0.2060369685292244,
"num_tokens": 9160403.0,
"step": 4425
},
{
"entropy": 5.388100957870483,
"epoch": 4.023625624716038,
"grad_norm": 1.1328125,
"learning_rate": 0.0003815937713914455,
"loss": 4.8666,
"mean_token_accuracy": 0.19937212616205216,
"num_tokens": 9171026.0,
"step": 4430
},
{
"entropy": 5.305515336990356,
"epoch": 4.028169014084507,
"grad_norm": 1.109375,
"learning_rate": 0.00038128238931713743,
"loss": 4.8457,
"mean_token_accuracy": 0.19893278628587724,
"num_tokens": 9181252.0,
"step": 4435
},
{
"entropy": 5.449811840057373,
"epoch": 4.032712403452976,
"grad_norm": 1.2734375,
"learning_rate": 0.00038097074500159883,
"loss": 4.9344,
"mean_token_accuracy": 0.19561847895383835,
"num_tokens": 9190803.0,
"step": 4440
},
{
"entropy": 5.368840265274048,
"epoch": 4.037255792821445,
"grad_norm": 1.1796875,
"learning_rate": 0.00038065883921378123,
"loss": 4.895,
"mean_token_accuracy": 0.19714238941669465,
"num_tokens": 9201077.0,
"step": 4445
},
{
"entropy": 5.3157508850097654,
"epoch": 4.041799182189914,
"grad_norm": 1.4609375,
"learning_rate": 0.0003803466727232811,
"loss": 4.9248,
"mean_token_accuracy": 0.19727148860692978,
"num_tokens": 9211239.0,
"step": 4450
},
{
"entropy": 5.296138525009155,
"epoch": 4.046342571558383,
"grad_norm": 1.2109375,
"learning_rate": 0.0003800342463003383,
"loss": 4.8878,
"mean_token_accuracy": 0.20535666942596437,
"num_tokens": 9222683.0,
"step": 4455
},
{
"entropy": 5.328317737579345,
"epoch": 4.050885960926851,
"grad_norm": 1.2890625,
"learning_rate": 0.00037972156071583386,
"loss": 4.8615,
"mean_token_accuracy": 0.1979299232363701,
"num_tokens": 9232099.0,
"step": 4460
},
{
"entropy": 5.329192399978638,
"epoch": 4.05542935029532,
"grad_norm": 1.1640625,
"learning_rate": 0.0003794086167412884,
"loss": 4.9234,
"mean_token_accuracy": 0.19783305674791335,
"num_tokens": 9242660.0,
"step": 4465
},
{
"entropy": 5.293742752075195,
"epoch": 4.059972739663789,
"grad_norm": 1.1796875,
"learning_rate": 0.0003790954151488601,
"loss": 4.9368,
"mean_token_accuracy": 0.18643997311592103,
"num_tokens": 9253981.0,
"step": 4470
},
{
"entropy": 5.308784294128418,
"epoch": 4.064516129032258,
"grad_norm": 1.234375,
"learning_rate": 0.0003787819567113427,
"loss": 4.8599,
"mean_token_accuracy": 0.19894113391637802,
"num_tokens": 9264262.0,
"step": 4475
},
{
"entropy": 5.2459859371185305,
"epoch": 4.069059518400727,
"grad_norm": 1.21875,
"learning_rate": 0.0003784682422021638,
"loss": 4.8693,
"mean_token_accuracy": 0.2013252019882202,
"num_tokens": 9273562.0,
"step": 4480
},
{
"entropy": 5.30320029258728,
"epoch": 4.073602907769196,
"grad_norm": 1.25,
"learning_rate": 0.0003781542723953828,
"loss": 4.9012,
"mean_token_accuracy": 0.19877577126026152,
"num_tokens": 9283171.0,
"step": 4485
},
{
"entropy": 5.286655044555664,
"epoch": 4.078146297137665,
"grad_norm": 1.3046875,
"learning_rate": 0.00037784004806568885,
"loss": 4.8815,
"mean_token_accuracy": 0.19889022260904313,
"num_tokens": 9292624.0,
"step": 4490
},
{
"entropy": 5.389542961120606,
"epoch": 4.082689686506134,
"grad_norm": 1.2890625,
"learning_rate": 0.00037752556998839934,
"loss": 4.9416,
"mean_token_accuracy": 0.19593565315008163,
"num_tokens": 9303159.0,
"step": 4495
},
{
"entropy": 5.303213500976563,
"epoch": 4.087233075874602,
"grad_norm": 1.375,
"learning_rate": 0.0003772108389394576,
"loss": 4.8826,
"mean_token_accuracy": 0.20091952234506608,
"num_tokens": 9314340.0,
"step": 4500
},
{
"epoch": 4.087233075874602,
"eval_entropy": 5.188059877794842,
"eval_loss": 5.794461727142334,
"eval_mean_token_accuracy": 0.1632396315072858,
"eval_num_tokens": 9314340.0,
"eval_runtime": 2.6,
"eval_samples_per_second": 1321.145,
"eval_steps_per_second": 165.383,
"step": 4500
},
{
"entropy": 5.369351196289062,
"epoch": 4.091776465243071,
"grad_norm": 1.1953125,
"learning_rate": 0.00037689585569543137,
"loss": 4.9634,
"mean_token_accuracy": 0.2012728273868561,
"num_tokens": 9325381.0,
"step": 4505
},
{
"entropy": 5.321428346633911,
"epoch": 4.09631985461154,
"grad_norm": 1.328125,
"learning_rate": 0.0003765806210335103,
"loss": 4.9469,
"mean_token_accuracy": 0.191410331428051,
"num_tokens": 9336639.0,
"step": 4510
},
{
"entropy": 5.294337749481201,
"epoch": 4.100863243980009,
"grad_norm": 1.3203125,
"learning_rate": 0.00037626513573150473,
"loss": 4.83,
"mean_token_accuracy": 0.2066548377275467,
"num_tokens": 9346794.0,
"step": 4515
},
{
"entropy": 5.265624713897705,
"epoch": 4.105406633348478,
"grad_norm": 1.234375,
"learning_rate": 0.00037594940056784315,
"loss": 4.8889,
"mean_token_accuracy": 0.20270660072565078,
"num_tokens": 9356326.0,
"step": 4520
},
{
"entropy": 5.379350423812866,
"epoch": 4.109950022716947,
"grad_norm": 1.2734375,
"learning_rate": 0.0003756334163215708,
"loss": 4.8833,
"mean_token_accuracy": 0.2025448277592659,
"num_tokens": 9366402.0,
"step": 4525
},
{
"entropy": 5.337375640869141,
"epoch": 4.114493412085416,
"grad_norm": 1.328125,
"learning_rate": 0.00037531718377234723,
"loss": 4.8551,
"mean_token_accuracy": 0.2002864807844162,
"num_tokens": 9377067.0,
"step": 4530
},
{
"entropy": 5.284870052337647,
"epoch": 4.119036801453885,
"grad_norm": 1.125,
"learning_rate": 0.0003750007037004451,
"loss": 4.9016,
"mean_token_accuracy": 0.19602560698986055,
"num_tokens": 9387206.0,
"step": 4535
},
{
"entropy": 5.296796607971191,
"epoch": 4.123580190822353,
"grad_norm": 1.3515625,
"learning_rate": 0.0003746839768867473,
"loss": 4.8793,
"mean_token_accuracy": 0.20161139070987702,
"num_tokens": 9397195.0,
"step": 4540
},
{
"entropy": 5.2543981075286865,
"epoch": 4.128123580190822,
"grad_norm": 1.2421875,
"learning_rate": 0.0003743670041127458,
"loss": 4.8506,
"mean_token_accuracy": 0.19654244184494019,
"num_tokens": 9407611.0,
"step": 4545
},
{
"entropy": 5.372268915176392,
"epoch": 4.132666969559291,
"grad_norm": 1.1953125,
"learning_rate": 0.0003740497861605395,
"loss": 4.9339,
"mean_token_accuracy": 0.19733388125896453,
"num_tokens": 9417897.0,
"step": 4550
},
{
"entropy": 5.344783210754395,
"epoch": 4.13721035892776,
"grad_norm": 1.265625,
"learning_rate": 0.000373732323812832,
"loss": 4.8969,
"mean_token_accuracy": 0.19168096482753755,
"num_tokens": 9428896.0,
"step": 4555
},
{
"entropy": 5.269835710525513,
"epoch": 4.141753748296229,
"grad_norm": 1.2578125,
"learning_rate": 0.00037341461785293025,
"loss": 4.8279,
"mean_token_accuracy": 0.2060396268963814,
"num_tokens": 9438854.0,
"step": 4560
},
{
"entropy": 5.244528675079346,
"epoch": 4.146297137664698,
"grad_norm": 1.3203125,
"learning_rate": 0.0003730966690647421,
"loss": 4.8919,
"mean_token_accuracy": 0.1976166069507599,
"num_tokens": 9447798.0,
"step": 4565
},
{
"entropy": 5.266746091842651,
"epoch": 4.150840527033167,
"grad_norm": 1.265625,
"learning_rate": 0.00037277847823277444,
"loss": 4.8499,
"mean_token_accuracy": 0.2047287181019783,
"num_tokens": 9458930.0,
"step": 4570
},
{
"entropy": 5.26498646736145,
"epoch": 4.155383916401636,
"grad_norm": 1.234375,
"learning_rate": 0.00037246004614213164,
"loss": 4.8341,
"mean_token_accuracy": 0.2063610464334488,
"num_tokens": 9469740.0,
"step": 4575
},
{
"entropy": 5.2268843173980715,
"epoch": 4.159927305770104,
"grad_norm": 1.2578125,
"learning_rate": 0.0003721413735785132,
"loss": 4.8359,
"mean_token_accuracy": 0.20477087199687957,
"num_tokens": 9479823.0,
"step": 4580
},
{
"entropy": 5.28771653175354,
"epoch": 4.164470695138573,
"grad_norm": 1.1953125,
"learning_rate": 0.000371822461328212,
"loss": 4.8214,
"mean_token_accuracy": 0.20891731530427932,
"num_tokens": 9489995.0,
"step": 4585
},
{
"entropy": 5.266953229904175,
"epoch": 4.169014084507042,
"grad_norm": 1.203125,
"learning_rate": 0.00037150331017811233,
"loss": 4.9062,
"mean_token_accuracy": 0.1985541418194771,
"num_tokens": 9501311.0,
"step": 4590
},
{
"entropy": 5.291744613647461,
"epoch": 4.173557473875511,
"grad_norm": 1.1953125,
"learning_rate": 0.0003711839209156878,
"loss": 4.9534,
"mean_token_accuracy": 0.19430545270442962,
"num_tokens": 9511901.0,
"step": 4595
},
{
"entropy": 5.3300055980682375,
"epoch": 4.17810086324398,
"grad_norm": 1.15625,
"learning_rate": 0.0003708642943289999,
"loss": 4.9002,
"mean_token_accuracy": 0.19327285438776015,
"num_tokens": 9522908.0,
"step": 4600
},
{
"entropy": 5.38222336769104,
"epoch": 4.182644252612449,
"grad_norm": 1.3125,
"learning_rate": 0.0003705444312066953,
"loss": 4.9708,
"mean_token_accuracy": 0.1971336454153061,
"num_tokens": 9533363.0,
"step": 4605
},
{
"entropy": 5.265525245666504,
"epoch": 4.187187641980918,
"grad_norm": 1.1796875,
"learning_rate": 0.0003702243323380045,
"loss": 4.8398,
"mean_token_accuracy": 0.1992759719491005,
"num_tokens": 9543302.0,
"step": 4610
},
{
"entropy": 5.232510232925415,
"epoch": 4.191731031349387,
"grad_norm": 1.3515625,
"learning_rate": 0.00036990399851273965,
"loss": 4.8851,
"mean_token_accuracy": 0.19274404942989348,
"num_tokens": 9553143.0,
"step": 4615
},
{
"entropy": 5.365129232406616,
"epoch": 4.196274420717856,
"grad_norm": 1.109375,
"learning_rate": 0.0003695834305212925,
"loss": 4.9149,
"mean_token_accuracy": 0.19661492556333543,
"num_tokens": 9564236.0,
"step": 4620
},
{
"entropy": 5.3101753234863285,
"epoch": 4.200817810086324,
"grad_norm": 1.2578125,
"learning_rate": 0.00036926262915463294,
"loss": 4.8583,
"mean_token_accuracy": 0.2029729127883911,
"num_tokens": 9574363.0,
"step": 4625
},
{
"entropy": 5.252830266952515,
"epoch": 4.205361199454793,
"grad_norm": 1.25,
"learning_rate": 0.00036894159520430616,
"loss": 4.9113,
"mean_token_accuracy": 0.19811587631702424,
"num_tokens": 9585631.0,
"step": 4630
},
{
"entropy": 5.285367250442505,
"epoch": 4.209904588823262,
"grad_norm": 1.2578125,
"learning_rate": 0.00036862032946243173,
"loss": 4.8872,
"mean_token_accuracy": 0.19324180632829666,
"num_tokens": 9596826.0,
"step": 4635
},
{
"entropy": 5.236816167831421,
"epoch": 4.214447978191731,
"grad_norm": 1.1015625,
"learning_rate": 0.00036829883272170087,
"loss": 4.8559,
"mean_token_accuracy": 0.20024349242448808,
"num_tokens": 9608158.0,
"step": 4640
},
{
"entropy": 5.316199731826782,
"epoch": 4.2189913675602,
"grad_norm": 1.1796875,
"learning_rate": 0.00036797710577537486,
"loss": 4.9589,
"mean_token_accuracy": 0.19454145878553392,
"num_tokens": 9619437.0,
"step": 4645
},
{
"entropy": 5.3643382549285885,
"epoch": 4.223534756928669,
"grad_norm": 1.2734375,
"learning_rate": 0.00036765514941728295,
"loss": 4.9258,
"mean_token_accuracy": 0.20042336285114287,
"num_tokens": 9629913.0,
"step": 4650
},
{
"entropy": 5.23789234161377,
"epoch": 4.228078146297138,
"grad_norm": 1.1171875,
"learning_rate": 0.0003673329644418204,
"loss": 4.897,
"mean_token_accuracy": 0.20093750953674316,
"num_tokens": 9640033.0,
"step": 4655
},
{
"entropy": 5.29016227722168,
"epoch": 4.232621535665607,
"grad_norm": 1.2421875,
"learning_rate": 0.00036701055164394664,
"loss": 4.9428,
"mean_token_accuracy": 0.19728004783391953,
"num_tokens": 9651010.0,
"step": 4660
},
{
"entropy": 5.260200071334839,
"epoch": 4.237164925034075,
"grad_norm": 1.234375,
"learning_rate": 0.00036668791181918313,
"loss": 4.8645,
"mean_token_accuracy": 0.20341949462890624,
"num_tokens": 9661549.0,
"step": 4665
},
{
"entropy": 5.311307668685913,
"epoch": 4.241708314402544,
"grad_norm": 1.265625,
"learning_rate": 0.0003663650457636117,
"loss": 4.9051,
"mean_token_accuracy": 0.19370613992214203,
"num_tokens": 9671619.0,
"step": 4670
},
{
"entropy": 5.380846643447876,
"epoch": 4.246251703771013,
"grad_norm": 1.2109375,
"learning_rate": 0.0003660419542738722,
"loss": 4.9453,
"mean_token_accuracy": 0.20350857824087143,
"num_tokens": 9682582.0,
"step": 4675
},
{
"entropy": 5.2221277236938475,
"epoch": 4.250795093139482,
"grad_norm": 1.2890625,
"learning_rate": 0.0003657186381471608,
"loss": 4.866,
"mean_token_accuracy": 0.20088305175304413,
"num_tokens": 9691997.0,
"step": 4680
},
{
"entropy": 5.288985967636108,
"epoch": 4.255338482507951,
"grad_norm": 1.1953125,
"learning_rate": 0.00036539509818122765,
"loss": 4.935,
"mean_token_accuracy": 0.2003515213727951,
"num_tokens": 9702105.0,
"step": 4685
},
{
"entropy": 5.346467351913452,
"epoch": 4.25988187187642,
"grad_norm": 1.265625,
"learning_rate": 0.0003650713351743758,
"loss": 4.9303,
"mean_token_accuracy": 0.20128719210624696,
"num_tokens": 9712347.0,
"step": 4690
},
{
"entropy": 5.29157772064209,
"epoch": 4.264425261244889,
"grad_norm": 1.2890625,
"learning_rate": 0.0003647473499254581,
"loss": 4.9199,
"mean_token_accuracy": 0.2009749487042427,
"num_tokens": 9723116.0,
"step": 4695
},
{
"entropy": 5.268213987350464,
"epoch": 4.268968650613358,
"grad_norm": 1.3125,
"learning_rate": 0.000364423143233876,
"loss": 4.8486,
"mean_token_accuracy": 0.20222790837287902,
"num_tokens": 9732656.0,
"step": 4700
},
{
"entropy": 5.339369440078736,
"epoch": 4.273512039981826,
"grad_norm": 1.171875,
"learning_rate": 0.00036409871589957723,
"loss": 4.9535,
"mean_token_accuracy": 0.19580906629562378,
"num_tokens": 9743312.0,
"step": 4705
},
{
"entropy": 5.286981105804443,
"epoch": 4.278055429350295,
"grad_norm": 1.265625,
"learning_rate": 0.000363774068723054,
"loss": 4.8989,
"mean_token_accuracy": 0.19558375477790832,
"num_tokens": 9753358.0,
"step": 4710
},
{
"entropy": 5.3042456150054935,
"epoch": 4.282598818718764,
"grad_norm": 1.171875,
"learning_rate": 0.00036344920250534106,
"loss": 4.821,
"mean_token_accuracy": 0.20525853782892228,
"num_tokens": 9763472.0,
"step": 4715
},
{
"entropy": 5.2547557830810545,
"epoch": 4.287142208087233,
"grad_norm": 1.234375,
"learning_rate": 0.0003631241180480133,
"loss": 4.8419,
"mean_token_accuracy": 0.2008869990706444,
"num_tokens": 9773700.0,
"step": 4720
},
{
"entropy": 5.246592283248901,
"epoch": 4.291685597455702,
"grad_norm": 1.234375,
"learning_rate": 0.0003627988161531845,
"loss": 4.8696,
"mean_token_accuracy": 0.19719520658254625,
"num_tokens": 9784042.0,
"step": 4725
},
{
"entropy": 5.327698373794556,
"epoch": 4.296228986824171,
"grad_norm": 1.203125,
"learning_rate": 0.0003624732976235046,
"loss": 4.9914,
"mean_token_accuracy": 0.20068394243717194,
"num_tokens": 9794464.0,
"step": 4730
},
{
"entropy": 5.313396787643432,
"epoch": 4.30077237619264,
"grad_norm": 1.171875,
"learning_rate": 0.0003621475632621584,
"loss": 4.9243,
"mean_token_accuracy": 0.1980634704232216,
"num_tokens": 9804459.0,
"step": 4735
},
{
"entropy": 5.25934534072876,
"epoch": 4.305315765561109,
"grad_norm": 1.28125,
"learning_rate": 0.0003618216138728627,
"loss": 4.9049,
"mean_token_accuracy": 0.19715648740530015,
"num_tokens": 9814962.0,
"step": 4740
},
{
"entropy": 5.319570827484131,
"epoch": 4.309859154929577,
"grad_norm": 1.2578125,
"learning_rate": 0.00036149545025986567,
"loss": 4.9427,
"mean_token_accuracy": 0.19676993042230606,
"num_tokens": 9825184.0,
"step": 4745
},
{
"entropy": 5.325601816177368,
"epoch": 4.314402544298046,
"grad_norm": 1.2265625,
"learning_rate": 0.00036116907322794335,
"loss": 4.9835,
"mean_token_accuracy": 0.19220527559518813,
"num_tokens": 9835184.0,
"step": 4750
},
{
"entropy": 5.256272268295288,
"epoch": 4.318945933666515,
"grad_norm": 1.2890625,
"learning_rate": 0.00036084248358239857,
"loss": 4.9076,
"mean_token_accuracy": 0.19772685170173646,
"num_tokens": 9845358.0,
"step": 4755
},
{
"entropy": 5.293330430984497,
"epoch": 4.323489323034984,
"grad_norm": 1.3515625,
"learning_rate": 0.00036051568212905897,
"loss": 4.8454,
"mean_token_accuracy": 0.19976577013731003,
"num_tokens": 9856519.0,
"step": 4760
},
{
"entropy": 5.353240585327148,
"epoch": 4.328032712403453,
"grad_norm": 1.34375,
"learning_rate": 0.0003601886696742745,
"loss": 4.9033,
"mean_token_accuracy": 0.1989097476005554,
"num_tokens": 9867478.0,
"step": 4765
},
{
"entropy": 5.19911036491394,
"epoch": 4.332576101771922,
"grad_norm": 1.2421875,
"learning_rate": 0.000359861447024916,
"loss": 4.8981,
"mean_token_accuracy": 0.19947690963745118,
"num_tokens": 9877709.0,
"step": 4770
},
{
"entropy": 5.25590558052063,
"epoch": 4.337119491140391,
"grad_norm": 1.2421875,
"learning_rate": 0.00035953401498837285,
"loss": 4.8513,
"mean_token_accuracy": 0.2042177215218544,
"num_tokens": 9888134.0,
"step": 4775
},
{
"entropy": 5.270602750778198,
"epoch": 4.34166288050886,
"grad_norm": 1.078125,
"learning_rate": 0.00035920637437255096,
"loss": 4.893,
"mean_token_accuracy": 0.19927174746990203,
"num_tokens": 9899150.0,
"step": 4780
},
{
"entropy": 5.3126648426055905,
"epoch": 4.346206269877328,
"grad_norm": 1.2109375,
"learning_rate": 0.000358878525985871,
"loss": 4.9003,
"mean_token_accuracy": 0.2006280541419983,
"num_tokens": 9909599.0,
"step": 4785
},
{
"entropy": 5.262983894348144,
"epoch": 4.350749659245797,
"grad_norm": 1.109375,
"learning_rate": 0.0003585504706372663,
"loss": 4.8539,
"mean_token_accuracy": 0.20642408281564711,
"num_tokens": 9919607.0,
"step": 4790
},
{
"entropy": 5.293412971496582,
"epoch": 4.355293048614266,
"grad_norm": 1.125,
"learning_rate": 0.0003582222091361808,
"loss": 4.9081,
"mean_token_accuracy": 0.19443069398403168,
"num_tokens": 9930431.0,
"step": 4795
},
{
"entropy": 5.260459470748901,
"epoch": 4.359836437982735,
"grad_norm": 1.1953125,
"learning_rate": 0.0003578937422925671,
"loss": 4.8951,
"mean_token_accuracy": 0.1996706947684288,
"num_tokens": 9940647.0,
"step": 4800
},
{
"entropy": 5.276372957229614,
"epoch": 4.364379827351204,
"grad_norm": 1.296875,
"learning_rate": 0.0003575650709168846,
"loss": 4.9309,
"mean_token_accuracy": 0.19030837416648866,
"num_tokens": 9951110.0,
"step": 4805
},
{
"entropy": 5.31486144065857,
"epoch": 4.368923216719673,
"grad_norm": 1.265625,
"learning_rate": 0.00035723619582009715,
"loss": 4.8907,
"mean_token_accuracy": 0.19809959083795547,
"num_tokens": 9961158.0,
"step": 4810
},
{
"entropy": 5.240279674530029,
"epoch": 4.373466606088142,
"grad_norm": 1.3125,
"learning_rate": 0.00035690711781367137,
"loss": 4.9232,
"mean_token_accuracy": 0.19546207636594773,
"num_tokens": 9971051.0,
"step": 4815
},
{
"entropy": 5.235450363159179,
"epoch": 4.378009995456611,
"grad_norm": 1.3046875,
"learning_rate": 0.0003565778377095744,
"loss": 4.9023,
"mean_token_accuracy": 0.19381988793611526,
"num_tokens": 9981990.0,
"step": 4820
},
{
"entropy": 5.307952642440796,
"epoch": 4.38255338482508,
"grad_norm": 1.3125,
"learning_rate": 0.0003562483563202724,
"loss": 4.8389,
"mean_token_accuracy": 0.20544678270816802,
"num_tokens": 9991668.0,
"step": 4825
},
{
"entropy": 5.202409934997559,
"epoch": 4.387096774193548,
"grad_norm": 1.2109375,
"learning_rate": 0.00035591867445872784,
"loss": 4.7797,
"mean_token_accuracy": 0.20608181059360503,
"num_tokens": 10001519.0,
"step": 4830
},
{
"entropy": 5.235861015319824,
"epoch": 4.391640163562017,
"grad_norm": 1.265625,
"learning_rate": 0.000355588792938398,
"loss": 4.8383,
"mean_token_accuracy": 0.20663538724184036,
"num_tokens": 10010856.0,
"step": 4835
},
{
"entropy": 5.211793565750122,
"epoch": 4.396183552930486,
"grad_norm": 1.296875,
"learning_rate": 0.0003552587125732326,
"loss": 4.8605,
"mean_token_accuracy": 0.20249849408864976,
"num_tokens": 10020647.0,
"step": 4840
},
{
"entropy": 5.204100751876831,
"epoch": 4.400726942298955,
"grad_norm": 1.203125,
"learning_rate": 0.00035492843417767224,
"loss": 4.7832,
"mean_token_accuracy": 0.2101762816309929,
"num_tokens": 10030114.0,
"step": 4845
},
{
"entropy": 5.276265621185303,
"epoch": 4.405270331667424,
"grad_norm": 1.3046875,
"learning_rate": 0.000354597958566646,
"loss": 4.866,
"mean_token_accuracy": 0.2056652709841728,
"num_tokens": 10039863.0,
"step": 4850
},
{
"entropy": 5.205725431442261,
"epoch": 4.409813721035893,
"grad_norm": 1.21875,
"learning_rate": 0.00035426728655556954,
"loss": 4.8465,
"mean_token_accuracy": 0.19329382926225663,
"num_tokens": 10049111.0,
"step": 4855
},
{
"entropy": 5.258617401123047,
"epoch": 4.414357110404362,
"grad_norm": 1.1796875,
"learning_rate": 0.0003539364189603433,
"loss": 4.9182,
"mean_token_accuracy": 0.19670866131782533,
"num_tokens": 10058762.0,
"step": 4860
},
{
"entropy": 5.37601375579834,
"epoch": 4.418900499772831,
"grad_norm": 1.1875,
"learning_rate": 0.00035360535659735006,
"loss": 4.922,
"mean_token_accuracy": 0.19646555036306382,
"num_tokens": 10068587.0,
"step": 4865
},
{
"entropy": 5.3170387744903564,
"epoch": 4.423443889141299,
"grad_norm": 1.2578125,
"learning_rate": 0.0003532741002834534,
"loss": 4.8918,
"mean_token_accuracy": 0.20092277228832245,
"num_tokens": 10079386.0,
"step": 4870
},
{
"entropy": 5.200781726837159,
"epoch": 4.427987278509768,
"grad_norm": 1.40625,
"learning_rate": 0.00035294265083599526,
"loss": 4.8344,
"mean_token_accuracy": 0.20573084354400634,
"num_tokens": 10089490.0,
"step": 4875
},
{
"entropy": 5.206735229492187,
"epoch": 4.432530667878237,
"grad_norm": 1.1875,
"learning_rate": 0.0003526110090727942,
"loss": 4.9131,
"mean_token_accuracy": 0.20527913719415664,
"num_tokens": 10100465.0,
"step": 4880
},
{
"entropy": 5.311049365997315,
"epoch": 4.437074057246706,
"grad_norm": 1.2265625,
"learning_rate": 0.0003522791758121434,
"loss": 4.8558,
"mean_token_accuracy": 0.20936551094055175,
"num_tokens": 10110919.0,
"step": 4885
},
{
"entropy": 5.294192266464234,
"epoch": 4.441617446615175,
"grad_norm": 1.296875,
"learning_rate": 0.00035194715187280835,
"loss": 4.8989,
"mean_token_accuracy": 0.1994428887963295,
"num_tokens": 10121116.0,
"step": 4890
},
{
"entropy": 5.3216118812561035,
"epoch": 4.446160835983644,
"grad_norm": 1.1953125,
"learning_rate": 0.0003516149380740252,
"loss": 5.0015,
"mean_token_accuracy": 0.19395408034324646,
"num_tokens": 10132269.0,
"step": 4895
},
{
"entropy": 5.256423997879028,
"epoch": 4.450704225352113,
"grad_norm": 1.3203125,
"learning_rate": 0.0003512825352354983,
"loss": 4.8806,
"mean_token_accuracy": 0.20602166652679443,
"num_tokens": 10142856.0,
"step": 4900
},
{
"entropy": 5.273519802093506,
"epoch": 4.455247614720582,
"grad_norm": 1.2890625,
"learning_rate": 0.00035094994417739884,
"loss": 4.8636,
"mean_token_accuracy": 0.1966313436627388,
"num_tokens": 10152377.0,
"step": 4905
},
{
"entropy": 5.322879505157471,
"epoch": 4.45979100408905,
"grad_norm": 1.296875,
"learning_rate": 0.000350617165720362,
"loss": 4.9225,
"mean_token_accuracy": 0.2022253915667534,
"num_tokens": 10161961.0,
"step": 4910
},
{
"entropy": 5.3163307189941404,
"epoch": 4.464334393457519,
"grad_norm": 1.390625,
"learning_rate": 0.00035028420068548567,
"loss": 4.8942,
"mean_token_accuracy": 0.2021799713373184,
"num_tokens": 10171624.0,
"step": 4915
},
{
"entropy": 5.355405950546265,
"epoch": 4.468877782825988,
"grad_norm": 1.2734375,
"learning_rate": 0.0003499510498943279,
"loss": 4.9853,
"mean_token_accuracy": 0.19386842846870422,
"num_tokens": 10182642.0,
"step": 4920
},
{
"entropy": 5.298817253112793,
"epoch": 4.473421172194457,
"grad_norm": 1.265625,
"learning_rate": 0.00034961771416890526,
"loss": 4.9403,
"mean_token_accuracy": 0.20447411835193635,
"num_tokens": 10193307.0,
"step": 4925
},
{
"entropy": 5.255819463729859,
"epoch": 4.477964561562926,
"grad_norm": 1.171875,
"learning_rate": 0.0003492841943316903,
"loss": 4.9373,
"mean_token_accuracy": 0.19752543717622756,
"num_tokens": 10203506.0,
"step": 4930
},
{
"entropy": 5.237539672851563,
"epoch": 4.482507950931395,
"grad_norm": 1.1328125,
"learning_rate": 0.0003489504912056103,
"loss": 4.8331,
"mean_token_accuracy": 0.20056360512971877,
"num_tokens": 10214266.0,
"step": 4935
},
{
"entropy": 5.1739089012146,
"epoch": 4.487051340299864,
"grad_norm": 1.1796875,
"learning_rate": 0.0003486166056140445,
"loss": 4.8185,
"mean_token_accuracy": 0.20460565239191056,
"num_tokens": 10224953.0,
"step": 4940
},
{
"entropy": 5.237729740142822,
"epoch": 4.491594729668333,
"grad_norm": 1.203125,
"learning_rate": 0.00034828253838082223,
"loss": 4.8826,
"mean_token_accuracy": 0.1938586562871933,
"num_tokens": 10235435.0,
"step": 4945
},
{
"entropy": 5.234312725067139,
"epoch": 4.496138119036801,
"grad_norm": 1.1953125,
"learning_rate": 0.00034794829033022134,
"loss": 4.8792,
"mean_token_accuracy": 0.20478031933307647,
"num_tokens": 10245677.0,
"step": 4950
},
{
"entropy": 5.182089757919312,
"epoch": 4.50068150840527,
"grad_norm": 1.2734375,
"learning_rate": 0.0003476138622869657,
"loss": 4.8415,
"mean_token_accuracy": 0.20132656544446945,
"num_tokens": 10256135.0,
"step": 4955
},
{
"entropy": 5.27617301940918,
"epoch": 4.505224897773739,
"grad_norm": 1.3046875,
"learning_rate": 0.0003472792550762232,
"loss": 4.9267,
"mean_token_accuracy": 0.1956597864627838,
"num_tokens": 10266465.0,
"step": 4960
},
{
"entropy": 5.32457013130188,
"epoch": 4.509768287142208,
"grad_norm": 1.1875,
"learning_rate": 0.0003469444695236038,
"loss": 4.9488,
"mean_token_accuracy": 0.20120729953050615,
"num_tokens": 10278123.0,
"step": 4965
},
{
"entropy": 5.262253332138061,
"epoch": 4.514311676510677,
"grad_norm": 1.25,
"learning_rate": 0.00034660950645515765,
"loss": 4.8968,
"mean_token_accuracy": 0.1990523084998131,
"num_tokens": 10287486.0,
"step": 4970
},
{
"entropy": 5.272685909271241,
"epoch": 4.518855065879146,
"grad_norm": 1.234375,
"learning_rate": 0.0003462743666973728,
"loss": 4.8778,
"mean_token_accuracy": 0.20325821042060851,
"num_tokens": 10298100.0,
"step": 4975
},
{
"entropy": 5.305084705352783,
"epoch": 4.523398455247615,
"grad_norm": 1.3125,
"learning_rate": 0.0003459390510771734,
"loss": 4.9416,
"mean_token_accuracy": 0.20441449731588363,
"num_tokens": 10308549.0,
"step": 4980
},
{
"entropy": 5.266169643402099,
"epoch": 4.527941844616084,
"grad_norm": 1.3203125,
"learning_rate": 0.0003456035604219172,
"loss": 4.9172,
"mean_token_accuracy": 0.19476551413536072,
"num_tokens": 10319030.0,
"step": 4985
},
{
"entropy": 5.211561870574951,
"epoch": 4.532485233984552,
"grad_norm": 1.2890625,
"learning_rate": 0.0003452678955593942,
"loss": 4.959,
"mean_token_accuracy": 0.2005244344472885,
"num_tokens": 10329613.0,
"step": 4990
},
{
"entropy": 5.329316282272339,
"epoch": 4.537028623353021,
"grad_norm": 1.359375,
"learning_rate": 0.000344932057317824,
"loss": 4.9605,
"mean_token_accuracy": 0.19310607016086578,
"num_tokens": 10339844.0,
"step": 4995
},
{
"entropy": 5.326785230636597,
"epoch": 4.54157201272149,
"grad_norm": 1.1796875,
"learning_rate": 0.0003445960465258542,
"loss": 4.9686,
"mean_token_accuracy": 0.19235819876194,
"num_tokens": 10350906.0,
"step": 5000
},
{
"epoch": 4.54157201272149,
"eval_entropy": 5.178572007112725,
"eval_loss": 5.766829490661621,
"eval_mean_token_accuracy": 0.1660090373005978,
"eval_num_tokens": 10350906.0,
"eval_runtime": 2.6084,
"eval_samples_per_second": 1316.908,
"eval_steps_per_second": 164.853,
"step": 5000
}
],
"logging_steps": 5,
"max_steps": 11000,
"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": 1.504277620992e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink