28185ab911
Model: fpadovani/tur-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.976628543013426,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.695967006683349,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 14.4375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.4267,
|
|
"mean_token_accuracy": 0.00011848341673612595,
|
|
"num_tokens": 9089.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.677842998504639,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 15.3125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.5334,
|
|
"mean_token_accuracy": 0.00015810276381671428,
|
|
"num_tokens": 21650.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.76645073890686,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 19.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.113,
|
|
"mean_token_accuracy": 0.00020675990963354708,
|
|
"num_tokens": 31067.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.015227270126343,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 34.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.4863,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 40816.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.510676717758178,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 13.875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.7346,
|
|
"mean_token_accuracy": 9.78473573923111e-05,
|
|
"num_tokens": 51065.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.519583129882813,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 3.4375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.7082,
|
|
"mean_token_accuracy": 0.005600619851611555,
|
|
"num_tokens": 60539.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.700261497497559,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.5537,
|
|
"mean_token_accuracy": 0.028791588079184294,
|
|
"num_tokens": 71010.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.671732521057129,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2564,
|
|
"mean_token_accuracy": 0.04515027813613415,
|
|
"num_tokens": 81394.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.483489990234375,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1167,
|
|
"mean_token_accuracy": 0.04342293459922075,
|
|
"num_tokens": 92491.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.399119567871093,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9358,
|
|
"mean_token_accuracy": 0.043518274836242196,
|
|
"num_tokens": 102113.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.48739938735962,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8944,
|
|
"mean_token_accuracy": 0.04352726638317108,
|
|
"num_tokens": 112211.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.497431468963622,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7893,
|
|
"mean_token_accuracy": 0.040843716263771056,
|
|
"num_tokens": 122166.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.438471221923828,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6965,
|
|
"mean_token_accuracy": 0.04762680903077125,
|
|
"num_tokens": 131016.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.383831214904784,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6514,
|
|
"mean_token_accuracy": 0.04989047236740589,
|
|
"num_tokens": 141369.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.395554351806641,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6059,
|
|
"mean_token_accuracy": 0.045283081382513045,
|
|
"num_tokens": 150939.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.431264877319336,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5567,
|
|
"mean_token_accuracy": 0.05196296907961369,
|
|
"num_tokens": 159981.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.404547595977784,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.4901,
|
|
"mean_token_accuracy": 0.05384046658873558,
|
|
"num_tokens": 170139.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.323073673248292,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4153,
|
|
"mean_token_accuracy": 0.05128877349197865,
|
|
"num_tokens": 180746.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.234126758575439,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.2695,
|
|
"mean_token_accuracy": 0.057958731055259706,
|
|
"num_tokens": 190250.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.200442790985107,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3164,
|
|
"mean_token_accuracy": 0.061629551276564595,
|
|
"num_tokens": 201185.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.167423057556153,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2292,
|
|
"mean_token_accuracy": 0.057263224571943286,
|
|
"num_tokens": 211476.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.147284317016602,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1279,
|
|
"mean_token_accuracy": 0.054998861998319624,
|
|
"num_tokens": 220965.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.161615943908691,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0647,
|
|
"mean_token_accuracy": 0.04764294326305389,
|
|
"num_tokens": 231539.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.009593105316162,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9565,
|
|
"mean_token_accuracy": 0.0596030842512846,
|
|
"num_tokens": 241206.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.899549961090088,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8673,
|
|
"mean_token_accuracy": 0.056177464127540586,
|
|
"num_tokens": 250629.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.890707683563232,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8412,
|
|
"mean_token_accuracy": 0.05115443356335163,
|
|
"num_tokens": 260705.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.684529113769532,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7676,
|
|
"mean_token_accuracy": 0.05398021526634693,
|
|
"num_tokens": 270499.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.696455478668213,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.7417,
|
|
"mean_token_accuracy": 0.05484265871345997,
|
|
"num_tokens": 279998.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.45594367980957,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6184,
|
|
"mean_token_accuracy": 0.058821605145931245,
|
|
"num_tokens": 290322.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.307820510864257,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.5725,
|
|
"mean_token_accuracy": 0.06243101172149181,
|
|
"num_tokens": 299893.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.206488990783692,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5498,
|
|
"mean_token_accuracy": 0.06206914372742176,
|
|
"num_tokens": 309191.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.06636447906494,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.4815,
|
|
"mean_token_accuracy": 0.06140335313975811,
|
|
"num_tokens": 318494.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.044106483459473,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4775,
|
|
"mean_token_accuracy": 0.0596194114536047,
|
|
"num_tokens": 328299.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.939890956878662,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.5442,
|
|
"mean_token_accuracy": 0.057165585830807684,
|
|
"num_tokens": 338878.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.874685859680175,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.468,
|
|
"mean_token_accuracy": 0.05600897297263145,
|
|
"num_tokens": 348508.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.797854232788087,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.4283,
|
|
"mean_token_accuracy": 0.06139532402157784,
|
|
"num_tokens": 358233.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.801108264923096,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3868,
|
|
"mean_token_accuracy": 0.06326259039342404,
|
|
"num_tokens": 367297.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.760680103302002,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.3884,
|
|
"mean_token_accuracy": 0.058847269788384435,
|
|
"num_tokens": 376880.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.77523078918457,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.3879,
|
|
"mean_token_accuracy": 0.06135675199329853,
|
|
"num_tokens": 386806.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.665618419647217,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.349,
|
|
"mean_token_accuracy": 0.06269419603049756,
|
|
"num_tokens": 396161.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.676361560821533,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.3986,
|
|
"mean_token_accuracy": 0.06113377064466476,
|
|
"num_tokens": 405643.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.612883186340332,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.3532,
|
|
"mean_token_accuracy": 0.06992297060787678,
|
|
"num_tokens": 414367.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.678384780883789,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4084,
|
|
"mean_token_accuracy": 0.06210144981741905,
|
|
"num_tokens": 424430.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.648359966278075,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.3776,
|
|
"mean_token_accuracy": 0.06364397667348384,
|
|
"num_tokens": 435176.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.576762390136718,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3963,
|
|
"mean_token_accuracy": 0.06432987973093987,
|
|
"num_tokens": 445611.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.57848825454712,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.3247,
|
|
"mean_token_accuracy": 0.0661028627306223,
|
|
"num_tokens": 455143.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.5846848487854,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.4127,
|
|
"mean_token_accuracy": 0.06825519613921642,
|
|
"num_tokens": 465507.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.642884063720704,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.413,
|
|
"mean_token_accuracy": 0.06387550197541714,
|
|
"num_tokens": 476152.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.646145343780518,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.2597,
|
|
"mean_token_accuracy": 0.06847746260464191,
|
|
"num_tokens": 486341.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.544342041015625,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.218,
|
|
"mean_token_accuracy": 0.06961140334606171,
|
|
"num_tokens": 495915.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.424121761322022,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.2949,
|
|
"mean_token_accuracy": 0.07351046912372113,
|
|
"num_tokens": 505666.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.567151832580567,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.2218,
|
|
"mean_token_accuracy": 0.06917047686874866,
|
|
"num_tokens": 515411.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.503114604949952,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2975,
|
|
"mean_token_accuracy": 0.06331334635615349,
|
|
"num_tokens": 525018.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.563315486907959,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.255,
|
|
"mean_token_accuracy": 0.06826438158750534,
|
|
"num_tokens": 535056.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.556770515441894,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.2081,
|
|
"mean_token_accuracy": 0.06950257569551468,
|
|
"num_tokens": 545115.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.51298542022705,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.2583,
|
|
"mean_token_accuracy": 0.06896053329110145,
|
|
"num_tokens": 554429.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.453236293792724,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2007,
|
|
"mean_token_accuracy": 0.07584292404353618,
|
|
"num_tokens": 563357.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.559890079498292,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.3728,
|
|
"mean_token_accuracy": 0.06644579172134399,
|
|
"num_tokens": 574480.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.503904342651367,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.2295,
|
|
"mean_token_accuracy": 0.06927415728569031,
|
|
"num_tokens": 583722.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.454716300964355,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.1738,
|
|
"mean_token_accuracy": 0.07493777833878994,
|
|
"num_tokens": 593644.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.49923391342163,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2147,
|
|
"mean_token_accuracy": 0.07160871513187886,
|
|
"num_tokens": 603547.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.44572868347168,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.2459,
|
|
"mean_token_accuracy": 0.06570485346019268,
|
|
"num_tokens": 613880.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.436261844635009,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.2238,
|
|
"mean_token_accuracy": 0.07062076106667518,
|
|
"num_tokens": 624213.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.490541934967041,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.2437,
|
|
"mean_token_accuracy": 0.07502996549010277,
|
|
"num_tokens": 634350.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.546083354949952,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.2438,
|
|
"mean_token_accuracy": 0.07064831554889679,
|
|
"num_tokens": 644040.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.492809295654297,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.2033,
|
|
"mean_token_accuracy": 0.07537714764475822,
|
|
"num_tokens": 652915.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.426844501495362,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1858,
|
|
"mean_token_accuracy": 0.06924654506146907,
|
|
"num_tokens": 662665.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.427862071990967,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.1955,
|
|
"mean_token_accuracy": 0.07384301200509072,
|
|
"num_tokens": 671667.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.440595722198486,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.2177,
|
|
"mean_token_accuracy": 0.06726174391806125,
|
|
"num_tokens": 680432.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.477052116394043,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.2011,
|
|
"mean_token_accuracy": 0.0729417447000742,
|
|
"num_tokens": 690184.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.372831344604492,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.092,
|
|
"mean_token_accuracy": 0.07851836830377579,
|
|
"num_tokens": 699147.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.329792785644532,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1201,
|
|
"mean_token_accuracy": 0.0765114389359951,
|
|
"num_tokens": 708696.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.382805728912354,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.0897,
|
|
"mean_token_accuracy": 0.07439367957413197,
|
|
"num_tokens": 718438.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.393447971343994,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.2278,
|
|
"mean_token_accuracy": 0.0699504878371954,
|
|
"num_tokens": 728157.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.350840854644776,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.1211,
|
|
"mean_token_accuracy": 0.07720480635762214,
|
|
"num_tokens": 737007.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.357489013671875,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.0726,
|
|
"mean_token_accuracy": 0.07758348956704139,
|
|
"num_tokens": 746873.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.33244686126709,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.9852,
|
|
"mean_token_accuracy": 0.08200759738683701,
|
|
"num_tokens": 755486.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.285024738311767,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.0944,
|
|
"mean_token_accuracy": 0.07533938214182853,
|
|
"num_tokens": 765481.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.345268440246581,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.0464,
|
|
"mean_token_accuracy": 0.07712380215525627,
|
|
"num_tokens": 774335.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.38193359375,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.1296,
|
|
"mean_token_accuracy": 0.07453210800886154,
|
|
"num_tokens": 784307.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.411716270446778,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.152,
|
|
"mean_token_accuracy": 0.07242761701345443,
|
|
"num_tokens": 795058.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.320662117004394,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0567,
|
|
"mean_token_accuracy": 0.07517294511198998,
|
|
"num_tokens": 805687.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.350330257415772,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0842,
|
|
"mean_token_accuracy": 0.07857826426625251,
|
|
"num_tokens": 815138.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.33483533859253,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.0588,
|
|
"mean_token_accuracy": 0.07447804734110833,
|
|
"num_tokens": 825009.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.324224948883057,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.0464,
|
|
"mean_token_accuracy": 0.07719542868435383,
|
|
"num_tokens": 834708.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.2837308883667,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.0994,
|
|
"mean_token_accuracy": 0.07782356329262256,
|
|
"num_tokens": 843888.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.280590438842774,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0652,
|
|
"mean_token_accuracy": 0.07787982597947121,
|
|
"num_tokens": 853057.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.313239860534669,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.1067,
|
|
"mean_token_accuracy": 0.0732115276157856,
|
|
"num_tokens": 862701.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.202727890014648,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0037,
|
|
"mean_token_accuracy": 0.07995785847306251,
|
|
"num_tokens": 872129.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.249781703948974,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.0233,
|
|
"mean_token_accuracy": 0.0809432752430439,
|
|
"num_tokens": 883166.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.261981916427612,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.8287,
|
|
"mean_token_accuracy": 0.09492504075169564,
|
|
"num_tokens": 894159.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.228504180908203,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0265,
|
|
"mean_token_accuracy": 0.07806817516684532,
|
|
"num_tokens": 904588.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.212932109832764,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.0094,
|
|
"mean_token_accuracy": 0.07935516387224198,
|
|
"num_tokens": 915554.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.191864347457885,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.03,
|
|
"mean_token_accuracy": 0.07978576868772506,
|
|
"num_tokens": 925546.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.266866683959961,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.9894,
|
|
"mean_token_accuracy": 0.082433320581913,
|
|
"num_tokens": 936615.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.207193756103516,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.023,
|
|
"mean_token_accuracy": 0.08058698140084744,
|
|
"num_tokens": 947374.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.204709148406982,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.955,
|
|
"mean_token_accuracy": 0.08018621355295182,
|
|
"num_tokens": 957772.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.176499080657958,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.0309,
|
|
"mean_token_accuracy": 0.08311204165220261,
|
|
"num_tokens": 967724.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.139270210266114,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9264,
|
|
"mean_token_accuracy": 0.08439484089612961,
|
|
"num_tokens": 976497.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.097719097137452,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9124,
|
|
"mean_token_accuracy": 0.0810380220413208,
|
|
"num_tokens": 986207.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4972650422675286,
|
|
"eval_entropy": 8.243832484917707,
|
|
"eval_loss": 8.042750358581543,
|
|
"eval_mean_token_accuracy": 0.07886241382623117,
|
|
"eval_num_tokens": 986207.0,
|
|
"eval_runtime": 2.7268,
|
|
"eval_samples_per_second": 1270.725,
|
|
"eval_steps_per_second": 159.162,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.125858736038207,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.8922,
|
|
"mean_token_accuracy": 0.08268110826611519,
|
|
"num_tokens": 995568.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.18411340713501,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9499,
|
|
"mean_token_accuracy": 0.08079156577587128,
|
|
"num_tokens": 1005829.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.089786481857299,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.9077,
|
|
"mean_token_accuracy": 0.08108234629034997,
|
|
"num_tokens": 1015290.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.183901023864745,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.872,
|
|
"mean_token_accuracy": 0.08664105087518692,
|
|
"num_tokens": 1024479.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.893880844116211,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.7961,
|
|
"mean_token_accuracy": 0.08195992782711983,
|
|
"num_tokens": 1034092.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.25371503829956,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 8.0162,
|
|
"mean_token_accuracy": 0.08469353094696999,
|
|
"num_tokens": 1043844.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.041243362426759,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.8648,
|
|
"mean_token_accuracy": 0.08864305019378663,
|
|
"num_tokens": 1052646.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.153491306304932,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.9181,
|
|
"mean_token_accuracy": 0.08187793269753456,
|
|
"num_tokens": 1062345.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.129785537719727,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.9003,
|
|
"mean_token_accuracy": 0.079953121393919,
|
|
"num_tokens": 1071116.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.065836000442506,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8489,
|
|
"mean_token_accuracy": 0.08677002638578415,
|
|
"num_tokens": 1080562.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.08962984085083,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.9092,
|
|
"mean_token_accuracy": 0.08326611518859864,
|
|
"num_tokens": 1091944.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.061810493469238,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8361,
|
|
"mean_token_accuracy": 0.08510831445455551,
|
|
"num_tokens": 1101412.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.109366941452027,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.9156,
|
|
"mean_token_accuracy": 0.08460526764392853,
|
|
"num_tokens": 1110652.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.040374612808227,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8352,
|
|
"mean_token_accuracy": 0.08915429562330246,
|
|
"num_tokens": 1120420.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.038780927658081,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.8934,
|
|
"mean_token_accuracy": 0.0821109727025032,
|
|
"num_tokens": 1130046.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.008350706100464,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.8662,
|
|
"mean_token_accuracy": 0.08604678623378277,
|
|
"num_tokens": 1138946.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.048417139053345,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.6038,
|
|
"mean_token_accuracy": 0.09397232159972191,
|
|
"num_tokens": 1149440.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.026965951919555,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.8972,
|
|
"mean_token_accuracy": 0.08445911929011345,
|
|
"num_tokens": 1158636.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.959886932373047,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.7368,
|
|
"mean_token_accuracy": 0.09059922322630883,
|
|
"num_tokens": 1167776.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.994223403930664,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.719,
|
|
"mean_token_accuracy": 0.08690877109766007,
|
|
"num_tokens": 1177919.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.917198991775512,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.7869,
|
|
"mean_token_accuracy": 0.0975386694073677,
|
|
"num_tokens": 1187363.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.191771316528321,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.9566,
|
|
"mean_token_accuracy": 0.08074369877576829,
|
|
"num_tokens": 1197329.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.946739959716797,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.8811,
|
|
"mean_token_accuracy": 0.08208825774490833,
|
|
"num_tokens": 1207408.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.007125759124756,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.8551,
|
|
"mean_token_accuracy": 0.08459711819887161,
|
|
"num_tokens": 1219258.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.056624031066894,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8472,
|
|
"mean_token_accuracy": 0.08643570393323899,
|
|
"num_tokens": 1228561.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.8858905792236325,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.7861,
|
|
"mean_token_accuracy": 0.0854301854968071,
|
|
"num_tokens": 1239665.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.968693017959595,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7938,
|
|
"mean_token_accuracy": 0.08755213171243667,
|
|
"num_tokens": 1249490.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.931836175918579,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.8103,
|
|
"mean_token_accuracy": 0.08826896399259568,
|
|
"num_tokens": 1259328.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.089329528808594,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.7524,
|
|
"mean_token_accuracy": 0.08410023339092731,
|
|
"num_tokens": 1269523.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.8148744106292725,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.8255,
|
|
"mean_token_accuracy": 0.08130960762500763,
|
|
"num_tokens": 1277961.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.05358533859253,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7896,
|
|
"mean_token_accuracy": 0.08684217259287834,
|
|
"num_tokens": 1288228.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.969742012023926,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7255,
|
|
"mean_token_accuracy": 0.09051954224705697,
|
|
"num_tokens": 1299636.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.837801790237426,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.7596,
|
|
"mean_token_accuracy": 0.08984493017196656,
|
|
"num_tokens": 1310035.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.899158906936646,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7654,
|
|
"mean_token_accuracy": 0.09294063225388527,
|
|
"num_tokens": 1320751.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.864286184310913,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.6445,
|
|
"mean_token_accuracy": 0.0892346404492855,
|
|
"num_tokens": 1331023.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.845469665527344,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7513,
|
|
"mean_token_accuracy": 0.0856916643679142,
|
|
"num_tokens": 1341814.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.911331701278686,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6408,
|
|
"mean_token_accuracy": 0.0899275153875351,
|
|
"num_tokens": 1351026.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.953000831604004,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.855,
|
|
"mean_token_accuracy": 0.07497294135391712,
|
|
"num_tokens": 1361004.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.813990640640259,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7092,
|
|
"mean_token_accuracy": 0.08678068146109581,
|
|
"num_tokens": 1370729.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.900962209701538,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.6649,
|
|
"mean_token_accuracy": 0.08870695382356644,
|
|
"num_tokens": 1379525.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.831973600387573,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7323,
|
|
"mean_token_accuracy": 0.09046533703804016,
|
|
"num_tokens": 1389863.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.800638961791992,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.8508,
|
|
"mean_token_accuracy": 0.08676985278725624,
|
|
"num_tokens": 1399468.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.86581072807312,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6529,
|
|
"mean_token_accuracy": 0.09256174638867379,
|
|
"num_tokens": 1408785.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.747158002853394,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.6904,
|
|
"mean_token_accuracy": 0.09273576214909554,
|
|
"num_tokens": 1418666.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.985917377471924,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.8001,
|
|
"mean_token_accuracy": 0.08961994498968125,
|
|
"num_tokens": 1428699.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.853034973144531,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6893,
|
|
"mean_token_accuracy": 0.09192831739783287,
|
|
"num_tokens": 1437354.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.813079929351806,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.708,
|
|
"mean_token_accuracy": 0.08974602445960045,
|
|
"num_tokens": 1447327.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.871495628356934,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.7563,
|
|
"mean_token_accuracy": 0.09072970226407051,
|
|
"num_tokens": 1457793.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.713533926010132,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6719,
|
|
"mean_token_accuracy": 0.08988175988197326,
|
|
"num_tokens": 1467143.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.857532787322998,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.7148,
|
|
"mean_token_accuracy": 0.08926208540797234,
|
|
"num_tokens": 1476422.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.813498783111572,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.656,
|
|
"mean_token_accuracy": 0.09503143355250358,
|
|
"num_tokens": 1486253.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.716996860504151,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6467,
|
|
"mean_token_accuracy": 0.09296249076724053,
|
|
"num_tokens": 1496080.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.760905981063843,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6075,
|
|
"mean_token_accuracy": 0.09641017168760299,
|
|
"num_tokens": 1505532.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.783897256851196,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.5671,
|
|
"mean_token_accuracy": 0.0936719074845314,
|
|
"num_tokens": 1515705.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.653465604782104,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.6041,
|
|
"mean_token_accuracy": 0.09394907802343369,
|
|
"num_tokens": 1525651.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.739487743377685,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5965,
|
|
"mean_token_accuracy": 0.09230046570301056,
|
|
"num_tokens": 1534576.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.608673191070556,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.4234,
|
|
"mean_token_accuracy": 0.1033897265791893,
|
|
"num_tokens": 1544765.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.687298154830932,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.5766,
|
|
"mean_token_accuracy": 0.09803976342082024,
|
|
"num_tokens": 1553636.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.740061044692993,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5557,
|
|
"mean_token_accuracy": 0.09136517718434334,
|
|
"num_tokens": 1563356.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.585089302062988,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.5708,
|
|
"mean_token_accuracy": 0.09769159182906151,
|
|
"num_tokens": 1572508.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.73101921081543,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.6024,
|
|
"mean_token_accuracy": 0.094760861992836,
|
|
"num_tokens": 1581583.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.694145679473877,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5936,
|
|
"mean_token_accuracy": 0.09348807856440544,
|
|
"num_tokens": 1592114.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.6239536762237545,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.4752,
|
|
"mean_token_accuracy": 0.10072982311248779,
|
|
"num_tokens": 1602417.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.628807640075683,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5729,
|
|
"mean_token_accuracy": 0.09305822476744652,
|
|
"num_tokens": 1611753.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.685155630111694,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.6614,
|
|
"mean_token_accuracy": 0.09417398199439049,
|
|
"num_tokens": 1621959.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.682513904571533,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.5577,
|
|
"mean_token_accuracy": 0.09433014839887618,
|
|
"num_tokens": 1630909.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.712544775009155,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.5309,
|
|
"mean_token_accuracy": 0.09536156728863716,
|
|
"num_tokens": 1641316.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.516957235336304,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.4478,
|
|
"mean_token_accuracy": 0.10105396658182145,
|
|
"num_tokens": 1651506.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.540144920349121,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.4276,
|
|
"mean_token_accuracy": 0.0987382858991623,
|
|
"num_tokens": 1659871.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.714676856994629,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.6448,
|
|
"mean_token_accuracy": 0.0904959000647068,
|
|
"num_tokens": 1670178.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.6358119487762455,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5413,
|
|
"mean_token_accuracy": 0.09731325581669807,
|
|
"num_tokens": 1680131.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.569293355941772,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.4575,
|
|
"mean_token_accuracy": 0.10155909061431885,
|
|
"num_tokens": 1689568.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.681883001327515,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5101,
|
|
"mean_token_accuracy": 0.1035876877605915,
|
|
"num_tokens": 1698039.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.606202030181885,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5097,
|
|
"mean_token_accuracy": 0.10067008286714554,
|
|
"num_tokens": 1707659.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.514298009872436,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.4873,
|
|
"mean_token_accuracy": 0.10052895620465278,
|
|
"num_tokens": 1716952.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.705683279037475,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5457,
|
|
"mean_token_accuracy": 0.0922851774841547,
|
|
"num_tokens": 1726323.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.634090089797974,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5293,
|
|
"mean_token_accuracy": 0.0906990833580494,
|
|
"num_tokens": 1736265.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.553630065917969,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.4983,
|
|
"mean_token_accuracy": 0.0946983590722084,
|
|
"num_tokens": 1745150.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.669640970230103,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5938,
|
|
"mean_token_accuracy": 0.09150914251804351,
|
|
"num_tokens": 1754641.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.481142282485962,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.4017,
|
|
"mean_token_accuracy": 0.10149217993021012,
|
|
"num_tokens": 1763468.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.524608373641968,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.5274,
|
|
"mean_token_accuracy": 0.09709789380431175,
|
|
"num_tokens": 1774845.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.516222858428955,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.4645,
|
|
"mean_token_accuracy": 0.10099468976259232,
|
|
"num_tokens": 1784606.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.644159364700317,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.4937,
|
|
"mean_token_accuracy": 0.09957759752869606,
|
|
"num_tokens": 1793893.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.51349983215332,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4704,
|
|
"mean_token_accuracy": 0.10551093220710754,
|
|
"num_tokens": 1804095.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.605443716049194,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4784,
|
|
"mean_token_accuracy": 0.10306655913591385,
|
|
"num_tokens": 1813475.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.447127199172973,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.3992,
|
|
"mean_token_accuracy": 0.10492947027087211,
|
|
"num_tokens": 1822779.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.615777015686035,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4335,
|
|
"mean_token_accuracy": 0.09923516064882279,
|
|
"num_tokens": 1832239.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.387023115158081,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4093,
|
|
"mean_token_accuracy": 0.09377154260873795,
|
|
"num_tokens": 1843395.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.5788719177246096,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5225,
|
|
"mean_token_accuracy": 0.10159855633974076,
|
|
"num_tokens": 1853861.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.520095157623291,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.4465,
|
|
"mean_token_accuracy": 0.10155273377895355,
|
|
"num_tokens": 1862992.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.476743268966675,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.462,
|
|
"mean_token_accuracy": 0.09680022820830345,
|
|
"num_tokens": 1871684.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.5863439559936525,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.4698,
|
|
"mean_token_accuracy": 0.09647404700517655,
|
|
"num_tokens": 1881956.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.474717378616333,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5371,
|
|
"mean_token_accuracy": 0.09288978725671768,
|
|
"num_tokens": 1891860.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.579680776596069,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.4486,
|
|
"mean_token_accuracy": 0.09725009277462959,
|
|
"num_tokens": 1901521.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.453162288665771,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.4141,
|
|
"mean_token_accuracy": 0.09932923316955566,
|
|
"num_tokens": 1910438.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.387307262420654,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.4446,
|
|
"mean_token_accuracy": 0.1021581083536148,
|
|
"num_tokens": 1922045.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.487140321731568,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.2728,
|
|
"mean_token_accuracy": 0.111979690939188,
|
|
"num_tokens": 1930330.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.266748380661011,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2387,
|
|
"mean_token_accuracy": 0.10433561950922013,
|
|
"num_tokens": 1939485.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.47536039352417,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.3665,
|
|
"mean_token_accuracy": 0.09609432145953178,
|
|
"num_tokens": 1949448.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.554550075531006,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.467,
|
|
"mean_token_accuracy": 0.0964827798306942,
|
|
"num_tokens": 1959923.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.274206740515573,
|
|
"eval_loss": 7.491804599761963,
|
|
"eval_mean_token_accuracy": 0.09507872324763081,
|
|
"eval_num_tokens": 1959923.0,
|
|
"eval_runtime": 2.531,
|
|
"eval_samples_per_second": 1369.031,
|
|
"eval_steps_per_second": 171.475,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.353235244750977,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.3984,
|
|
"mean_token_accuracy": 0.10374441519379615,
|
|
"num_tokens": 1969105.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.457019170125325,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 7.2227,
|
|
"mean_token_accuracy": 0.10618321845928828,
|
|
"num_tokens": 1979124.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.439343357086182,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 7.2425,
|
|
"mean_token_accuracy": 0.10488953292369843,
|
|
"num_tokens": 1987813.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.403989601135254,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 7.2194,
|
|
"mean_token_accuracy": 0.10040127784013748,
|
|
"num_tokens": 1996743.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.332245922088623,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 7.2289,
|
|
"mean_token_accuracy": 0.10728532820940018,
|
|
"num_tokens": 2006499.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.397444581985473,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 7.2181,
|
|
"mean_token_accuracy": 0.10751654133200646,
|
|
"num_tokens": 2015103.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.333607339859009,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 7.1521,
|
|
"mean_token_accuracy": 0.10657267794013023,
|
|
"num_tokens": 2024681.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.361648988723755,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 7.2333,
|
|
"mean_token_accuracy": 0.10235757604241372,
|
|
"num_tokens": 2034388.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.3061192512512205,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 7.1629,
|
|
"mean_token_accuracy": 0.10662485361099243,
|
|
"num_tokens": 2042729.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.468704652786255,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 7.2491,
|
|
"mean_token_accuracy": 0.10004433616995811,
|
|
"num_tokens": 2051669.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.315274286270141,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 7.203,
|
|
"mean_token_accuracy": 0.10867782533168793,
|
|
"num_tokens": 2061002.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.257018661499023,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 7.1119,
|
|
"mean_token_accuracy": 0.11166723147034645,
|
|
"num_tokens": 2070222.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.385891675949097,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 7.2811,
|
|
"mean_token_accuracy": 0.10769186615943908,
|
|
"num_tokens": 2079683.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.287356996536255,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 7.2799,
|
|
"mean_token_accuracy": 0.10403456911444664,
|
|
"num_tokens": 2089659.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.447647762298584,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 7.1489,
|
|
"mean_token_accuracy": 0.11329402998089791,
|
|
"num_tokens": 2099766.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.274552202224731,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 7.3182,
|
|
"mean_token_accuracy": 0.0980657048523426,
|
|
"num_tokens": 2110244.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.460256767272949,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 7.2313,
|
|
"mean_token_accuracy": 0.10618335977196694,
|
|
"num_tokens": 2120068.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.2404412746429445,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 7.2233,
|
|
"mean_token_accuracy": 0.10707401782274246,
|
|
"num_tokens": 2128862.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.458929872512817,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 7.2764,
|
|
"mean_token_accuracy": 0.09662094824016094,
|
|
"num_tokens": 2140618.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.288017702102661,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 7.194,
|
|
"mean_token_accuracy": 0.10394087433815002,
|
|
"num_tokens": 2151890.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.270646715164185,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 7.1826,
|
|
"mean_token_accuracy": 0.11115405187010766,
|
|
"num_tokens": 2160862.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.330387210845947,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 7.2229,
|
|
"mean_token_accuracy": 0.11168113499879836,
|
|
"num_tokens": 2171631.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.2536815166473385,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 7.1376,
|
|
"mean_token_accuracy": 0.10918443202972412,
|
|
"num_tokens": 2181564.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.330545425415039,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 7.1867,
|
|
"mean_token_accuracy": 0.11050143316388131,
|
|
"num_tokens": 2191288.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.261415958404541,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 7.1565,
|
|
"mean_token_accuracy": 0.10519161149859428,
|
|
"num_tokens": 2201254.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.230209684371948,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 7.1055,
|
|
"mean_token_accuracy": 0.11186393052339554,
|
|
"num_tokens": 2210325.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.281809711456299,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 7.2071,
|
|
"mean_token_accuracy": 0.10663560330867768,
|
|
"num_tokens": 2220357.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.23015570640564,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 7.1303,
|
|
"mean_token_accuracy": 0.11602199077606201,
|
|
"num_tokens": 2229208.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.2548665523529055,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 7.1328,
|
|
"mean_token_accuracy": 0.11204535439610482,
|
|
"num_tokens": 2238780.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.141768455505371,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 7.167,
|
|
"mean_token_accuracy": 0.10382938161492347,
|
|
"num_tokens": 2249457.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.285257005691529,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.1068244844675064,
|
|
"num_tokens": 2259565.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.224769735336304,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 7.0625,
|
|
"mean_token_accuracy": 0.11556584015488625,
|
|
"num_tokens": 2268301.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.239184761047364,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 7.1984,
|
|
"mean_token_accuracy": 0.11132131665945053,
|
|
"num_tokens": 2278298.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.2060218334198,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 7.1351,
|
|
"mean_token_accuracy": 0.10933703258633613,
|
|
"num_tokens": 2288442.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.241116619110107,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 7.1301,
|
|
"mean_token_accuracy": 0.10417242571711541,
|
|
"num_tokens": 2298785.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.203887319564819,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 7.19,
|
|
"mean_token_accuracy": 0.10378496199846268,
|
|
"num_tokens": 2308881.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.240233612060547,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 7.1108,
|
|
"mean_token_accuracy": 0.11426399871706963,
|
|
"num_tokens": 2318447.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.245753383636474,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 7.1087,
|
|
"mean_token_accuracy": 0.11610123813152314,
|
|
"num_tokens": 2327162.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.276870441436768,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 7.1426,
|
|
"mean_token_accuracy": 0.10571493431925774,
|
|
"num_tokens": 2337207.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.15098819732666,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 7.1241,
|
|
"mean_token_accuracy": 0.10898757874965667,
|
|
"num_tokens": 2347381.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.295102882385254,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 7.1662,
|
|
"mean_token_accuracy": 0.10816890597343445,
|
|
"num_tokens": 2357942.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.127454805374145,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 7.0819,
|
|
"mean_token_accuracy": 0.11559660658240319,
|
|
"num_tokens": 2367242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.35596718788147,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 7.1466,
|
|
"mean_token_accuracy": 0.11849364861845971,
|
|
"num_tokens": 2376344.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.073469018936157,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 7.1563,
|
|
"mean_token_accuracy": 0.11420170515775681,
|
|
"num_tokens": 2387206.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.336416816711425,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 7.1873,
|
|
"mean_token_accuracy": 0.10830738544464111,
|
|
"num_tokens": 2397532.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.1107689380645756,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 7.0445,
|
|
"mean_token_accuracy": 0.10626394525170327,
|
|
"num_tokens": 2409165.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.190313863754272,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 7.149,
|
|
"mean_token_accuracy": 0.10830143168568611,
|
|
"num_tokens": 2420657.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.180588579177856,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 7.083,
|
|
"mean_token_accuracy": 0.109988534450531,
|
|
"num_tokens": 2429936.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.22004280090332,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 7.0766,
|
|
"mean_token_accuracy": 0.1176595076918602,
|
|
"num_tokens": 2440386.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.133073568344116,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 7.2361,
|
|
"mean_token_accuracy": 0.10853235200047492,
|
|
"num_tokens": 2449086.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.390012121200561,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 7.2336,
|
|
"mean_token_accuracy": 0.10847590416669846,
|
|
"num_tokens": 2459380.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.199709463119507,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 7.0777,
|
|
"mean_token_accuracy": 0.11947175413370133,
|
|
"num_tokens": 2468623.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.199767065048218,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 7.0294,
|
|
"mean_token_accuracy": 0.1127948857843876,
|
|
"num_tokens": 2479817.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.164726114273071,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 7.0859,
|
|
"mean_token_accuracy": 0.11037703976035118,
|
|
"num_tokens": 2490544.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.167629146575928,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 7.0454,
|
|
"mean_token_accuracy": 0.12425570264458656,
|
|
"num_tokens": 2500516.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.172258043289185,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 7.1754,
|
|
"mean_token_accuracy": 0.10876649245619774,
|
|
"num_tokens": 2510508.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.203859663009643,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 7.1213,
|
|
"mean_token_accuracy": 0.10957314744591713,
|
|
"num_tokens": 2520532.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.155911636352539,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 7.0755,
|
|
"mean_token_accuracy": 0.11014840230345727,
|
|
"num_tokens": 2529912.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.191475534439087,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 7.1632,
|
|
"mean_token_accuracy": 0.10724633038043976,
|
|
"num_tokens": 2540700.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.206296586990357,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 7.0413,
|
|
"mean_token_accuracy": 0.11765994355082512,
|
|
"num_tokens": 2549343.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.083246183395386,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 7.0668,
|
|
"mean_token_accuracy": 0.1165860801935196,
|
|
"num_tokens": 2558630.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.1589103698730465,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 7.0718,
|
|
"mean_token_accuracy": 0.11376485601067543,
|
|
"num_tokens": 2568559.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.175602960586548,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 7.1026,
|
|
"mean_token_accuracy": 0.11608376652002335,
|
|
"num_tokens": 2578088.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.084215831756592,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 7.0601,
|
|
"mean_token_accuracy": 0.12005995959043503,
|
|
"num_tokens": 2588518.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.294498062133789,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 7.2039,
|
|
"mean_token_accuracy": 0.10467370301485061,
|
|
"num_tokens": 2598850.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.135764741897583,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 7.1688,
|
|
"mean_token_accuracy": 0.10237733125686646,
|
|
"num_tokens": 2610667.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.185567140579224,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 7.0257,
|
|
"mean_token_accuracy": 0.12162859663367272,
|
|
"num_tokens": 2620430.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.903904771804809,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 6.8987,
|
|
"mean_token_accuracy": 0.12102525010704994,
|
|
"num_tokens": 2629433.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.157380104064941,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 7.0789,
|
|
"mean_token_accuracy": 0.10664870142936707,
|
|
"num_tokens": 2639611.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.200573539733886,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 7.0774,
|
|
"mean_token_accuracy": 0.11245412677526474,
|
|
"num_tokens": 2649494.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.080723524093628,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 7.0022,
|
|
"mean_token_accuracy": 0.11420606151223182,
|
|
"num_tokens": 2658723.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.136773586273193,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 7.0418,
|
|
"mean_token_accuracy": 0.11387985423207284,
|
|
"num_tokens": 2668530.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.171313905715943,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 7.1111,
|
|
"mean_token_accuracy": 0.113609828799963,
|
|
"num_tokens": 2678788.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.13408350944519,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 7.0156,
|
|
"mean_token_accuracy": 0.11711029559373856,
|
|
"num_tokens": 2688308.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.921298694610596,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 6.973,
|
|
"mean_token_accuracy": 0.11835426688194275,
|
|
"num_tokens": 2698146.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.197758626937866,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 7.0236,
|
|
"mean_token_accuracy": 0.11459743902087212,
|
|
"num_tokens": 2707336.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.096755838394165,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 7.0015,
|
|
"mean_token_accuracy": 0.12133366167545319,
|
|
"num_tokens": 2717573.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.0911478996276855,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 6.9963,
|
|
"mean_token_accuracy": 0.12023369744420051,
|
|
"num_tokens": 2727488.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.039147090911865,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 6.9939,
|
|
"mean_token_accuracy": 0.12316683754324913,
|
|
"num_tokens": 2738010.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.114057350158691,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 7.0293,
|
|
"mean_token_accuracy": 0.1164513386785984,
|
|
"num_tokens": 2746984.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.1252563953399655,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 7.0438,
|
|
"mean_token_accuracy": 0.11343937739729881,
|
|
"num_tokens": 2756578.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.112103080749511,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 6.9922,
|
|
"mean_token_accuracy": 0.12000504955649376,
|
|
"num_tokens": 2765773.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.04692587852478,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 6.9818,
|
|
"mean_token_accuracy": 0.12030813694000245,
|
|
"num_tokens": 2775002.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.180310916900635,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 7.0527,
|
|
"mean_token_accuracy": 0.1163438878953457,
|
|
"num_tokens": 2784682.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.036463880538941,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 6.9593,
|
|
"mean_token_accuracy": 0.12377962619066238,
|
|
"num_tokens": 2794371.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.1429462909698485,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 7.0222,
|
|
"mean_token_accuracy": 0.11636790558695793,
|
|
"num_tokens": 2803840.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.9931422710418705,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.9398,
|
|
"mean_token_accuracy": 0.12235129252076149,
|
|
"num_tokens": 2813536.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.061179304122925,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 7.0019,
|
|
"mean_token_accuracy": 0.11406521499156952,
|
|
"num_tokens": 2822741.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.118486928939819,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 6.9595,
|
|
"mean_token_accuracy": 0.1193142868578434,
|
|
"num_tokens": 2832208.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.882198286056519,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.8508,
|
|
"mean_token_accuracy": 0.1252315267920494,
|
|
"num_tokens": 2841672.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.089263820648194,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.9647,
|
|
"mean_token_accuracy": 0.11878615766763687,
|
|
"num_tokens": 2852323.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.0160314559936525,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 7.025,
|
|
"mean_token_accuracy": 0.11984933465719223,
|
|
"num_tokens": 2861863.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.071931934356689,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.12041502445936203,
|
|
"num_tokens": 2871608.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.973705291748047,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.9334,
|
|
"mean_token_accuracy": 0.11892005652189255,
|
|
"num_tokens": 2882091.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.082184648513794,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.8939,
|
|
"mean_token_accuracy": 0.12786136120557784,
|
|
"num_tokens": 2892499.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.019110584259034,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.9454,
|
|
"mean_token_accuracy": 0.11744871884584426,
|
|
"num_tokens": 2902208.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.049245834350586,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 6.9257,
|
|
"mean_token_accuracy": 0.11979928463697434,
|
|
"num_tokens": 2911196.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.953201770782471,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 6.9696,
|
|
"mean_token_accuracy": 0.11339190155267716,
|
|
"num_tokens": 2921339.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.122575187683106,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 6.9165,
|
|
"mean_token_accuracy": 0.117047119140625,
|
|
"num_tokens": 2930795.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.9450513362884525,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.8855,
|
|
"mean_token_accuracy": 0.12101696357131005,
|
|
"num_tokens": 2939647.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.4912978617603183,
|
|
"eval_entropy": 6.896390275471771,
|
|
"eval_loss": 7.153826713562012,
|
|
"eval_mean_token_accuracy": 0.111372493739639,
|
|
"eval_num_tokens": 2939647.0,
|
|
"eval_runtime": 2.5473,
|
|
"eval_samples_per_second": 1360.259,
|
|
"eval_steps_per_second": 170.376,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.024867534637451,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.9212,
|
|
"mean_token_accuracy": 0.12069288194179535,
|
|
"num_tokens": 2948365.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.94093337059021,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.9325,
|
|
"mean_token_accuracy": 0.11558735221624375,
|
|
"num_tokens": 2957447.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.112627077102661,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.9271,
|
|
"mean_token_accuracy": 0.12053594887256622,
|
|
"num_tokens": 2965938.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.008978652954101,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.1205656386911869,
|
|
"num_tokens": 2975830.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.047356414794922,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 6.9197,
|
|
"mean_token_accuracy": 0.1240025982260704,
|
|
"num_tokens": 2985391.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.048643589019775,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.9336,
|
|
"mean_token_accuracy": 0.12923261746764184,
|
|
"num_tokens": 2994397.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.9428564548492435,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 6.953,
|
|
"mean_token_accuracy": 0.11560544520616531,
|
|
"num_tokens": 3005085.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.029836559295655,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.9578,
|
|
"mean_token_accuracy": 0.11529956981539727,
|
|
"num_tokens": 3015500.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.058086109161377,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 7.0565,
|
|
"mean_token_accuracy": 0.11921946704387665,
|
|
"num_tokens": 3024862.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.083318996429443,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.8992,
|
|
"mean_token_accuracy": 0.11814992874860764,
|
|
"num_tokens": 3034775.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.957399702072143,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.9628,
|
|
"mean_token_accuracy": 0.11769941002130509,
|
|
"num_tokens": 3044048.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.14118275642395,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.9626,
|
|
"mean_token_accuracy": 0.11674395352602004,
|
|
"num_tokens": 3054237.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.794984579086304,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.7674,
|
|
"mean_token_accuracy": 0.1281619429588318,
|
|
"num_tokens": 3064787.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.9928360939025875,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.9458,
|
|
"mean_token_accuracy": 0.12229207009077073,
|
|
"num_tokens": 3074002.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.850202512741089,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.7999,
|
|
"mean_token_accuracy": 0.12233345210552216,
|
|
"num_tokens": 3084270.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.043655920028686,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.9182,
|
|
"mean_token_accuracy": 0.12310172468423844,
|
|
"num_tokens": 3095236.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.941450357437134,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 6.9168,
|
|
"mean_token_accuracy": 0.12208462804555893,
|
|
"num_tokens": 3104912.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.970426225662232,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.9082,
|
|
"mean_token_accuracy": 0.12240447178483009,
|
|
"num_tokens": 3114782.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.107024383544922,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.9601,
|
|
"mean_token_accuracy": 0.12547790259122849,
|
|
"num_tokens": 3123606.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.788683700561523,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.7625,
|
|
"mean_token_accuracy": 0.13145018517971038,
|
|
"num_tokens": 3133343.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.108981418609619,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 7.0058,
|
|
"mean_token_accuracy": 0.11305271163582802,
|
|
"num_tokens": 3143137.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.947832918167114,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.8644,
|
|
"mean_token_accuracy": 0.1278637908399105,
|
|
"num_tokens": 3152595.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.939337062835693,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.9363,
|
|
"mean_token_accuracy": 0.11795696318149566,
|
|
"num_tokens": 3162333.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.9427478313446045,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.8453,
|
|
"mean_token_accuracy": 0.12352253571152687,
|
|
"num_tokens": 3172765.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.045589303970337,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.9929,
|
|
"mean_token_accuracy": 0.11614625081419945,
|
|
"num_tokens": 3183613.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.03980655670166,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.8766,
|
|
"mean_token_accuracy": 0.12009989842772484,
|
|
"num_tokens": 3193020.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.980470466613769,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.8784,
|
|
"mean_token_accuracy": 0.11922077238559722,
|
|
"num_tokens": 3202845.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.8759448528289795,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.11566570103168487,
|
|
"num_tokens": 3211195.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.975606393814087,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.9086,
|
|
"mean_token_accuracy": 0.12444080635905266,
|
|
"num_tokens": 3221549.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.922389078140259,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.8006,
|
|
"mean_token_accuracy": 0.12442531585693359,
|
|
"num_tokens": 3232206.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.991663599014283,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.878,
|
|
"mean_token_accuracy": 0.11768145188689232,
|
|
"num_tokens": 3241613.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.811205768585205,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.8443,
|
|
"mean_token_accuracy": 0.11838877126574517,
|
|
"num_tokens": 3251968.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.960154819488525,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.8302,
|
|
"mean_token_accuracy": 0.12369426786899566,
|
|
"num_tokens": 3261969.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.932631254196167,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 6.8695,
|
|
"mean_token_accuracy": 0.12236511632800103,
|
|
"num_tokens": 3270894.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.932890605926514,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.9159,
|
|
"mean_token_accuracy": 0.11275593265891075,
|
|
"num_tokens": 3280949.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.057922506332398,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.8769,
|
|
"mean_token_accuracy": 0.1186174876987934,
|
|
"num_tokens": 3291449.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.8796507835388185,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.8854,
|
|
"mean_token_accuracy": 0.12050152122974396,
|
|
"num_tokens": 3301632.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.933141851425171,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.7486,
|
|
"mean_token_accuracy": 0.12816066890954972,
|
|
"num_tokens": 3310802.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.8287769794464115,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.8149,
|
|
"mean_token_accuracy": 0.12629651874303818,
|
|
"num_tokens": 3320833.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.9207744121551515,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.7941,
|
|
"mean_token_accuracy": 0.129004967212677,
|
|
"num_tokens": 3330030.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.759958076477051,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.5858,
|
|
"mean_token_accuracy": 0.14270602613687516,
|
|
"num_tokens": 3338745.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.899774885177612,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.8278,
|
|
"mean_token_accuracy": 0.12718854397535323,
|
|
"num_tokens": 3348067.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.859093141555786,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.7337,
|
|
"mean_token_accuracy": 0.13416660204529762,
|
|
"num_tokens": 3357061.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.898359632492065,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.7571,
|
|
"mean_token_accuracy": 0.12454862222075462,
|
|
"num_tokens": 3366573.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.931832551956177,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.11833840906620026,
|
|
"num_tokens": 3376996.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.953691625595093,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.8752,
|
|
"mean_token_accuracy": 0.1171141505241394,
|
|
"num_tokens": 3387615.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.756794214248657,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.6557,
|
|
"mean_token_accuracy": 0.1345459997653961,
|
|
"num_tokens": 3397578.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.858054876327515,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.8261,
|
|
"mean_token_accuracy": 0.12104411125183105,
|
|
"num_tokens": 3406697.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.958321046829224,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.12863764241337777,
|
|
"num_tokens": 3416631.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.879059791564941,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.11450941115617752,
|
|
"num_tokens": 3426564.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.018344354629517,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.896,
|
|
"mean_token_accuracy": 0.12408864945173263,
|
|
"num_tokens": 3436200.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.8397636890411375,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.8571,
|
|
"mean_token_accuracy": 0.12426263466477394,
|
|
"num_tokens": 3446460.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.954912996292114,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.742,
|
|
"mean_token_accuracy": 0.13110286071896554,
|
|
"num_tokens": 3457629.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.879143714904785,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.9071,
|
|
"mean_token_accuracy": 0.11985552161931992,
|
|
"num_tokens": 3467754.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.863123798370362,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.8413,
|
|
"mean_token_accuracy": 0.12256991863250732,
|
|
"num_tokens": 3477094.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.883700656890869,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.7897,
|
|
"mean_token_accuracy": 0.12876719385385513,
|
|
"num_tokens": 3487276.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.943530464172364,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.7975,
|
|
"mean_token_accuracy": 0.12761382833123208,
|
|
"num_tokens": 3498118.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.754543590545654,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.6284,
|
|
"mean_token_accuracy": 0.12865251079201698,
|
|
"num_tokens": 3508286.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.811681509017944,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.7476,
|
|
"mean_token_accuracy": 0.13177093118429184,
|
|
"num_tokens": 3517979.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.850424098968506,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.7761,
|
|
"mean_token_accuracy": 0.12440782338380814,
|
|
"num_tokens": 3528003.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.005227136611938,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.8417,
|
|
"mean_token_accuracy": 0.12152071818709373,
|
|
"num_tokens": 3537932.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.810896348953247,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.9556,
|
|
"mean_token_accuracy": 0.1197297677397728,
|
|
"num_tokens": 3548584.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.012252855300903,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.781,
|
|
"mean_token_accuracy": 0.1271994046866894,
|
|
"num_tokens": 3558117.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.8483014583587645,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.9036,
|
|
"mean_token_accuracy": 0.11816685497760773,
|
|
"num_tokens": 3569226.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.974257755279541,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.7629,
|
|
"mean_token_accuracy": 0.12144647464156151,
|
|
"num_tokens": 3578375.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.866056156158447,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.7794,
|
|
"mean_token_accuracy": 0.13146526888012885,
|
|
"num_tokens": 3587356.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.8879882335662845,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.8589,
|
|
"mean_token_accuracy": 0.12173376381397247,
|
|
"num_tokens": 3596785.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.89839768409729,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.7551,
|
|
"mean_token_accuracy": 0.13092348501086234,
|
|
"num_tokens": 3605357.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.862800073623657,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.7526,
|
|
"mean_token_accuracy": 0.12731626331806184,
|
|
"num_tokens": 3614979.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.827274322509766,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.757,
|
|
"mean_token_accuracy": 0.12851415947079659,
|
|
"num_tokens": 3624594.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.864847850799561,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.7487,
|
|
"mean_token_accuracy": 0.12592514380812644,
|
|
"num_tokens": 3633998.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.851823806762695,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.7163,
|
|
"mean_token_accuracy": 0.12974942922592164,
|
|
"num_tokens": 3642406.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.770096349716186,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.7885,
|
|
"mean_token_accuracy": 0.12484269142150879,
|
|
"num_tokens": 3652905.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.815859794616699,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.7069,
|
|
"mean_token_accuracy": 0.12600032165646552,
|
|
"num_tokens": 3662518.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.9039140224456785,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.7629,
|
|
"mean_token_accuracy": 0.13054750561714173,
|
|
"num_tokens": 3672102.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.855216646194458,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.7937,
|
|
"mean_token_accuracy": 0.12748010605573654,
|
|
"num_tokens": 3681821.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.0097535133361815,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.8683,
|
|
"mean_token_accuracy": 0.1209042452275753,
|
|
"num_tokens": 3690753.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.8094244480133055,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.8339,
|
|
"mean_token_accuracy": 0.12141335383057594,
|
|
"num_tokens": 3701694.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.893871736526489,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.7541,
|
|
"mean_token_accuracy": 0.12454791739583015,
|
|
"num_tokens": 3710894.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.784770774841308,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.691,
|
|
"mean_token_accuracy": 0.13386452198028564,
|
|
"num_tokens": 3720081.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.87687315940857,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.8052,
|
|
"mean_token_accuracy": 0.12933627292513847,
|
|
"num_tokens": 3731168.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.751729583740234,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.6062,
|
|
"mean_token_accuracy": 0.13786736354231835,
|
|
"num_tokens": 3740533.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.782799530029297,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.785,
|
|
"mean_token_accuracy": 0.12551278993487358,
|
|
"num_tokens": 3750523.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.876261758804321,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.8015,
|
|
"mean_token_accuracy": 0.12344402298331261,
|
|
"num_tokens": 3760323.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.664627456665039,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.5688,
|
|
"mean_token_accuracy": 0.14272117540240287,
|
|
"num_tokens": 3770970.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.79260401725769,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.8004,
|
|
"mean_token_accuracy": 0.1279281996190548,
|
|
"num_tokens": 3780810.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.853330516815186,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.7076,
|
|
"mean_token_accuracy": 0.13404083177447318,
|
|
"num_tokens": 3791461.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.7495325088500975,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.5249,
|
|
"mean_token_accuracy": 0.13892961144447327,
|
|
"num_tokens": 3800353.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.772555494308472,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.7006,
|
|
"mean_token_accuracy": 0.13255587890744208,
|
|
"num_tokens": 3810633.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.88034839630127,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.842,
|
|
"mean_token_accuracy": 0.1255967289209366,
|
|
"num_tokens": 3820254.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.9175169467926025,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.8171,
|
|
"mean_token_accuracy": 0.12266699597239494,
|
|
"num_tokens": 3829976.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.797274971008301,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.767,
|
|
"mean_token_accuracy": 0.1263387106359005,
|
|
"num_tokens": 3840593.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.817362308502197,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.7407,
|
|
"mean_token_accuracy": 0.13059170991182328,
|
|
"num_tokens": 3850063.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.871044397354126,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.7777,
|
|
"mean_token_accuracy": 0.126081869751215,
|
|
"num_tokens": 3860132.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.796940660476684,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.7346,
|
|
"mean_token_accuracy": 0.130959652364254,
|
|
"num_tokens": 3869914.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.699121189117432,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.674,
|
|
"mean_token_accuracy": 0.14097589403390884,
|
|
"num_tokens": 3879825.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.92541275024414,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.7464,
|
|
"mean_token_accuracy": 0.13307701125741006,
|
|
"num_tokens": 3889318.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.78137845993042,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.7402,
|
|
"mean_token_accuracy": 0.12863369882106782,
|
|
"num_tokens": 3899017.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.874488973617554,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.6951,
|
|
"mean_token_accuracy": 0.13685127720236778,
|
|
"num_tokens": 3907766.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.797482347488403,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.6783,
|
|
"mean_token_accuracy": 0.1312019035220146,
|
|
"num_tokens": 3916323.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.5960938523991315,
|
|
"eval_loss": 6.9205851554870605,
|
|
"eval_mean_token_accuracy": 0.11949630909972751,
|
|
"eval_num_tokens": 3916323.0,
|
|
"eval_runtime": 2.5415,
|
|
"eval_samples_per_second": 1363.394,
|
|
"eval_steps_per_second": 170.769,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.743474960327148,
|
|
"epoch": 1.993535554450522,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004864723636325361,
|
|
"loss": 6.7726,
|
|
"mean_token_accuracy": 0.12707573622465135,
|
|
"num_tokens": 3927639.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.870388746261597,
|
|
"epoch": 1.9985082048731975,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.00048633867439566855,
|
|
"loss": 6.6874,
|
|
"mean_token_accuracy": 0.12734511867165565,
|
|
"num_tokens": 3937452.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.868780401017931,
|
|
"epoch": 2.002983590253605,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004862043484784227,
|
|
"loss": 6.6415,
|
|
"mean_token_accuracy": 0.13130029953188366,
|
|
"num_tokens": 3946737.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.816501617431641,
|
|
"epoch": 2.0079562406762803,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048606938628546975,
|
|
"loss": 6.4258,
|
|
"mean_token_accuracy": 0.12974144965410234,
|
|
"num_tokens": 3957541.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.793483448028565,
|
|
"epoch": 2.0129288910989556,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004859337882233978,
|
|
"loss": 6.5056,
|
|
"mean_token_accuracy": 0.1279767021536827,
|
|
"num_tokens": 3967621.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.799717760086059,
|
|
"epoch": 2.017901541521631,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004857975547007106,
|
|
"loss": 6.5055,
|
|
"mean_token_accuracy": 0.13455797657370566,
|
|
"num_tokens": 3977920.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.578008365631104,
|
|
"epoch": 2.0228741919443065,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048566068612782595,
|
|
"loss": 6.1722,
|
|
"mean_token_accuracy": 0.15097661018371583,
|
|
"num_tokens": 3988128.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.703016328811645,
|
|
"epoch": 2.0278468423669818,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048552318291707533,
|
|
"loss": 6.3662,
|
|
"mean_token_accuracy": 0.14040629491209983,
|
|
"num_tokens": 3998467.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.733289527893066,
|
|
"epoch": 2.032819492789657,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048538504548270177,
|
|
"loss": 6.3358,
|
|
"mean_token_accuracy": 0.13190700858831406,
|
|
"num_tokens": 4007513.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.574963092803955,
|
|
"epoch": 2.0377921432123323,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048524627424085913,
|
|
"loss": 6.4053,
|
|
"mean_token_accuracy": 0.1385435938835144,
|
|
"num_tokens": 4016538.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.686594772338867,
|
|
"epoch": 2.0427647936350075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048510686960961057,
|
|
"loss": 6.3102,
|
|
"mean_token_accuracy": 0.14076657891273497,
|
|
"num_tokens": 4026850.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.68578085899353,
|
|
"epoch": 2.0477374440576828,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004849668320089276,
|
|
"loss": 6.308,
|
|
"mean_token_accuracy": 0.1393464133143425,
|
|
"num_tokens": 4037105.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.621056842803955,
|
|
"epoch": 2.052710094480358,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048482616186068826,
|
|
"loss": 6.3641,
|
|
"mean_token_accuracy": 0.14485840871930122,
|
|
"num_tokens": 4045694.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.611151647567749,
|
|
"epoch": 2.0576827449030333,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048468485958867664,
|
|
"loss": 6.2588,
|
|
"mean_token_accuracy": 0.14561159759759904,
|
|
"num_tokens": 4056425.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.588506889343262,
|
|
"epoch": 2.0626553953257085,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048454292561858083,
|
|
"loss": 6.292,
|
|
"mean_token_accuracy": 0.1491833969950676,
|
|
"num_tokens": 4066257.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.6618874073028564,
|
|
"epoch": 2.0676280457483838,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004844003603779921,
|
|
"loss": 6.3697,
|
|
"mean_token_accuracy": 0.13735560178756714,
|
|
"num_tokens": 4076835.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.6407470703125,
|
|
"epoch": 2.072600696171059,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004842571642964036,
|
|
"loss": 6.492,
|
|
"mean_token_accuracy": 0.13370660170912743,
|
|
"num_tokens": 4087644.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.70997428894043,
|
|
"epoch": 2.0775733465937343,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048411333780520876,
|
|
"loss": 6.4182,
|
|
"mean_token_accuracy": 0.13545572832226754,
|
|
"num_tokens": 4097040.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.660089588165283,
|
|
"epoch": 2.0825459970164095,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048396888133770034,
|
|
"loss": 6.385,
|
|
"mean_token_accuracy": 0.1400207258760929,
|
|
"num_tokens": 4106698.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.67029595375061,
|
|
"epoch": 2.087518647439085,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048382379532906897,
|
|
"loss": 6.3724,
|
|
"mean_token_accuracy": 0.14506559818983078,
|
|
"num_tokens": 4116804.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.649099445343017,
|
|
"epoch": 2.0924912978617605,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004836780802164016,
|
|
"loss": 6.481,
|
|
"mean_token_accuracy": 0.13108737394213676,
|
|
"num_tokens": 4126658.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.663601350784302,
|
|
"epoch": 2.0974639482844357,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048353173643868064,
|
|
"loss": 6.2641,
|
|
"mean_token_accuracy": 0.14222084060311319,
|
|
"num_tokens": 4136323.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.552397966384888,
|
|
"epoch": 2.102436598707111,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004833847644367826,
|
|
"loss": 6.4093,
|
|
"mean_token_accuracy": 0.14488788247108458,
|
|
"num_tokens": 4145714.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.74545841217041,
|
|
"epoch": 2.107409249129786,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048323716465347615,
|
|
"loss": 6.3883,
|
|
"mean_token_accuracy": 0.13914027959108352,
|
|
"num_tokens": 4156059.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.599608993530273,
|
|
"epoch": 2.1123818995524615,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004830889375334215,
|
|
"loss": 6.4329,
|
|
"mean_token_accuracy": 0.13543619066476822,
|
|
"num_tokens": 4166990.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.644566059112549,
|
|
"epoch": 2.1173545499751367,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048294008352316875,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.1432893082499504,
|
|
"num_tokens": 4175716.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.638896894454956,
|
|
"epoch": 2.122327200397812,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004827906030711564,
|
|
"loss": 6.4268,
|
|
"mean_token_accuracy": 0.13586034253239632,
|
|
"num_tokens": 4184832.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.468638515472412,
|
|
"epoch": 2.1272998508204872,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048264049662771053,
|
|
"loss": 6.2182,
|
|
"mean_token_accuracy": 0.15223108381032943,
|
|
"num_tokens": 4195678.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.667506408691406,
|
|
"epoch": 2.1322725012431625,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048248976464504266,
|
|
"loss": 6.4117,
|
|
"mean_token_accuracy": 0.13893484696745872,
|
|
"num_tokens": 4205357.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.682303857803345,
|
|
"epoch": 2.1372451516658377,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048233840757724926,
|
|
"loss": 6.4726,
|
|
"mean_token_accuracy": 0.13612781465053558,
|
|
"num_tokens": 4215416.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.6814305782318115,
|
|
"epoch": 2.142217802088513,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048218642588030943,
|
|
"loss": 6.5098,
|
|
"mean_token_accuracy": 0.1335015468299389,
|
|
"num_tokens": 4225894.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.718389368057251,
|
|
"epoch": 2.1471904525111887,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048203382001208453,
|
|
"loss": 6.4266,
|
|
"mean_token_accuracy": 0.14146106615662574,
|
|
"num_tokens": 4235973.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.5530835628509525,
|
|
"epoch": 2.152163102933864,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048188059043231585,
|
|
"loss": 6.3416,
|
|
"mean_token_accuracy": 0.14535260200500488,
|
|
"num_tokens": 4246411.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.699462938308716,
|
|
"epoch": 2.157135753356539,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048172673760262414,
|
|
"loss": 6.4057,
|
|
"mean_token_accuracy": 0.13765638917684556,
|
|
"num_tokens": 4255754.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.578954648971558,
|
|
"epoch": 2.1621084037792144,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048157226198650735,
|
|
"loss": 6.3426,
|
|
"mean_token_accuracy": 0.14024818763136865,
|
|
"num_tokens": 4265263.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.69226040840149,
|
|
"epoch": 2.1670810542018897,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004814171640493398,
|
|
"loss": 6.5192,
|
|
"mean_token_accuracy": 0.13299463614821433,
|
|
"num_tokens": 4275262.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.60386552810669,
|
|
"epoch": 2.172053704624565,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004812614442583708,
|
|
"loss": 6.4364,
|
|
"mean_token_accuracy": 0.1357870452105999,
|
|
"num_tokens": 4285467.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.650628185272216,
|
|
"epoch": 2.17702635504724,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004811051030827227,
|
|
"loss": 6.3541,
|
|
"mean_token_accuracy": 0.13564409241080283,
|
|
"num_tokens": 4294680.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.605967903137207,
|
|
"epoch": 2.1819990054699154,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048094814099339006,
|
|
"loss": 6.4335,
|
|
"mean_token_accuracy": 0.1353071413934231,
|
|
"num_tokens": 4304605.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.5511720180511475,
|
|
"epoch": 2.1869716558925907,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004807905584632381,
|
|
"loss": 6.2601,
|
|
"mean_token_accuracy": 0.1484084963798523,
|
|
"num_tokens": 4314112.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.510998249053955,
|
|
"epoch": 2.191944306315266,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048063235596700087,
|
|
"loss": 6.3169,
|
|
"mean_token_accuracy": 0.14116548001766205,
|
|
"num_tokens": 4323872.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.636221885681152,
|
|
"epoch": 2.196916956737941,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048047353398128056,
|
|
"loss": 6.2781,
|
|
"mean_token_accuracy": 0.14961938709020614,
|
|
"num_tokens": 4333002.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.5612975597381595,
|
|
"epoch": 2.2018896071606164,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004803140929845452,
|
|
"loss": 6.3989,
|
|
"mean_token_accuracy": 0.14100862443447112,
|
|
"num_tokens": 4342012.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.630680561065674,
|
|
"epoch": 2.2068622575832917,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048015403345712796,
|
|
"loss": 6.335,
|
|
"mean_token_accuracy": 0.1386937290430069,
|
|
"num_tokens": 4351959.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.611843776702881,
|
|
"epoch": 2.2118349080059674,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004799933558812252,
|
|
"loss": 6.3142,
|
|
"mean_token_accuracy": 0.1452218174934387,
|
|
"num_tokens": 4361267.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.592637825012207,
|
|
"epoch": 2.2168075584286426,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047983206074089546,
|
|
"loss": 6.3592,
|
|
"mean_token_accuracy": 0.1453488864004612,
|
|
"num_tokens": 4370104.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.297759294509888,
|
|
"epoch": 2.221780208851318,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004796701485220576,
|
|
"loss": 6.1391,
|
|
"mean_token_accuracy": 0.1663060151040554,
|
|
"num_tokens": 4380724.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.461687994003296,
|
|
"epoch": 2.226752859273993,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004795076197124893,
|
|
"loss": 6.2111,
|
|
"mean_token_accuracy": 0.15564971938729286,
|
|
"num_tokens": 4390521.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.5855005264282225,
|
|
"epoch": 2.2317255096966684,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047934447480182626,
|
|
"loss": 6.39,
|
|
"mean_token_accuracy": 0.14133716821670533,
|
|
"num_tokens": 4400075.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.576947164535523,
|
|
"epoch": 2.2366981601193436,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004791807142815598,
|
|
"loss": 6.3309,
|
|
"mean_token_accuracy": 0.14028072953224183,
|
|
"num_tokens": 4409913.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.559875965118408,
|
|
"epoch": 2.241670810542019,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047901633864503614,
|
|
"loss": 6.3491,
|
|
"mean_token_accuracy": 0.14033740609884263,
|
|
"num_tokens": 4418551.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.563806486129761,
|
|
"epoch": 2.246643460964694,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047885134838745433,
|
|
"loss": 6.5427,
|
|
"mean_token_accuracy": 0.1390278995037079,
|
|
"num_tokens": 4428129.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.722531986236572,
|
|
"epoch": 2.2516161113873694,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004786857440058654,
|
|
"loss": 6.4167,
|
|
"mean_token_accuracy": 0.14224035739898683,
|
|
"num_tokens": 4437311.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.425306606292724,
|
|
"epoch": 2.2565887618100446,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004785195259991701,
|
|
"loss": 6.2592,
|
|
"mean_token_accuracy": 0.15068233460187913,
|
|
"num_tokens": 4447387.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.519154500961304,
|
|
"epoch": 2.26156141223272,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000478352694868118,
|
|
"loss": 6.4323,
|
|
"mean_token_accuracy": 0.1365874394774437,
|
|
"num_tokens": 4456435.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.466250038146972,
|
|
"epoch": 2.266534062655395,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004781852511153058,
|
|
"loss": 6.2613,
|
|
"mean_token_accuracy": 0.15041158795356752,
|
|
"num_tokens": 4466464.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.588946151733398,
|
|
"epoch": 2.271506713078071,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047801719524517573,
|
|
"loss": 6.3699,
|
|
"mean_token_accuracy": 0.14319706484675407,
|
|
"num_tokens": 4476793.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.551668357849121,
|
|
"epoch": 2.276479363500746,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047784852776401403,
|
|
"loss": 6.3408,
|
|
"mean_token_accuracy": 0.14303407892584802,
|
|
"num_tokens": 4485811.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.616492700576782,
|
|
"epoch": 2.2814520139234213,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004776792491799495,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.13649573624134065,
|
|
"num_tokens": 4495018.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.587541723251343,
|
|
"epoch": 2.2864246643460966,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047750936000295213,
|
|
"loss": 6.4237,
|
|
"mean_token_accuracy": 0.13602023273706437,
|
|
"num_tokens": 4505847.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.593376111984253,
|
|
"epoch": 2.291397314768772,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047733886074483115,
|
|
"loss": 6.4319,
|
|
"mean_token_accuracy": 0.13742442131042482,
|
|
"num_tokens": 4516104.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.4319030284881595,
|
|
"epoch": 2.296369965191447,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004771677519192339,
|
|
"loss": 6.1951,
|
|
"mean_token_accuracy": 0.1542932316660881,
|
|
"num_tokens": 4524875.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.522582721710205,
|
|
"epoch": 2.3013426156141223,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004769960340416439,
|
|
"loss": 6.4475,
|
|
"mean_token_accuracy": 0.13546053245663642,
|
|
"num_tokens": 4533836.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.370718002319336,
|
|
"epoch": 2.3063152660367976,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047682370762937976,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.1733056738972664,
|
|
"num_tokens": 4544323.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.515668678283691,
|
|
"epoch": 2.311287916459473,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047665077320159315,
|
|
"loss": 6.3566,
|
|
"mean_token_accuracy": 0.14200845062732698,
|
|
"num_tokens": 4553811.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.400250387191773,
|
|
"epoch": 2.316260566882148,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047647723127926765,
|
|
"loss": 6.1887,
|
|
"mean_token_accuracy": 0.16027804613113403,
|
|
"num_tokens": 4563098.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.615733480453491,
|
|
"epoch": 2.3212332173048233,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004763030823852168,
|
|
"loss": 6.4867,
|
|
"mean_token_accuracy": 0.13570496514439584,
|
|
"num_tokens": 4573589.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.541930723190307,
|
|
"epoch": 2.3262058677274986,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004761283270440829,
|
|
"loss": 6.3805,
|
|
"mean_token_accuracy": 0.1394104018807411,
|
|
"num_tokens": 4583059.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.6634218215942385,
|
|
"epoch": 2.331178518150174,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004759529657823349,
|
|
"loss": 6.5747,
|
|
"mean_token_accuracy": 0.1389290988445282,
|
|
"num_tokens": 4593371.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.5851836681365965,
|
|
"epoch": 2.336151168572849,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047577699912826755,
|
|
"loss": 6.4681,
|
|
"mean_token_accuracy": 0.1362654224038124,
|
|
"num_tokens": 4602862.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.585508203506469,
|
|
"epoch": 2.341123818995525,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047560042761199915,
|
|
"loss": 6.3364,
|
|
"mean_token_accuracy": 0.1408190168440342,
|
|
"num_tokens": 4612827.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.567585754394531,
|
|
"epoch": 2.3460964694182,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004754232517654704,
|
|
"loss": 6.3909,
|
|
"mean_token_accuracy": 0.13844477534294128,
|
|
"num_tokens": 4621706.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.587166833877563,
|
|
"epoch": 2.3510691198408753,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004752454721224424,
|
|
"loss": 6.3494,
|
|
"mean_token_accuracy": 0.1418745554983616,
|
|
"num_tokens": 4630878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.643035554885865,
|
|
"epoch": 2.3560417702635505,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047506708921849535,
|
|
"loss": 6.4744,
|
|
"mean_token_accuracy": 0.13255613371729852,
|
|
"num_tokens": 4640582.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.524079608917236,
|
|
"epoch": 2.361014420686226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004748881035910269,
|
|
"loss": 6.394,
|
|
"mean_token_accuracy": 0.14071099683642388,
|
|
"num_tokens": 4650799.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.599751138687134,
|
|
"epoch": 2.365987071108901,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004747085157792504,
|
|
"loss": 6.3499,
|
|
"mean_token_accuracy": 0.14517031535506247,
|
|
"num_tokens": 4660566.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.573225212097168,
|
|
"epoch": 2.3709597215315763,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004745283263241933,
|
|
"loss": 6.4453,
|
|
"mean_token_accuracy": 0.13409614190459251,
|
|
"num_tokens": 4670234.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.540919303894043,
|
|
"epoch": 2.3759323719542516,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004743475357686957,
|
|
"loss": 6.3862,
|
|
"mean_token_accuracy": 0.14218721091747283,
|
|
"num_tokens": 4681114.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.505688333511353,
|
|
"epoch": 2.380905022376927,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047416614465740843,
|
|
"loss": 6.1865,
|
|
"mean_token_accuracy": 0.16146760135889054,
|
|
"num_tokens": 4691970.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.524224042892456,
|
|
"epoch": 2.385877672799602,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047398415353679167,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.13508354276418685,
|
|
"num_tokens": 4701723.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.523842716217041,
|
|
"epoch": 2.3908503232222773,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004738015629551132,
|
|
"loss": 6.3611,
|
|
"mean_token_accuracy": 0.1428997755050659,
|
|
"num_tokens": 4711356.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.502122688293457,
|
|
"epoch": 2.395822973644953,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047361837346244655,
|
|
"loss": 6.4016,
|
|
"mean_token_accuracy": 0.14505933225154877,
|
|
"num_tokens": 4721765.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.5327929019927975,
|
|
"epoch": 2.4007956240676283,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004734345856106698,
|
|
"loss": 6.2746,
|
|
"mean_token_accuracy": 0.14620665311813355,
|
|
"num_tokens": 4731678.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.482340097427368,
|
|
"epoch": 2.4057682744903035,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047325019995346355,
|
|
"loss": 6.3168,
|
|
"mean_token_accuracy": 0.14606815874576567,
|
|
"num_tokens": 4740960.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.501438188552856,
|
|
"epoch": 2.4107409249129788,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004730652170463091,
|
|
"loss": 6.4316,
|
|
"mean_token_accuracy": 0.13504584059119223,
|
|
"num_tokens": 4751753.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.555155038833618,
|
|
"epoch": 2.415713575335654,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004728796374464875,
|
|
"loss": 6.2847,
|
|
"mean_token_accuracy": 0.14447317868471146,
|
|
"num_tokens": 4761274.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.510864734649658,
|
|
"epoch": 2.4206862257583293,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047269346171307716,
|
|
"loss": 6.4626,
|
|
"mean_token_accuracy": 0.14278751909732817,
|
|
"num_tokens": 4770852.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.688248014450073,
|
|
"epoch": 2.4256588761810045,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004725066904069523,
|
|
"loss": 6.3209,
|
|
"mean_token_accuracy": 0.1462319627404213,
|
|
"num_tokens": 4780857.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.426857948303223,
|
|
"epoch": 2.4306315266036798,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004723193240907814,
|
|
"loss": 6.2768,
|
|
"mean_token_accuracy": 0.1379391722381115,
|
|
"num_tokens": 4790018.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.520555019378662,
|
|
"epoch": 2.435604177026355,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004721313633290257,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.1447875529527664,
|
|
"num_tokens": 4799747.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.63818507194519,
|
|
"epoch": 2.4405768274490303,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000471942808687937,
|
|
"loss": 6.4086,
|
|
"mean_token_accuracy": 0.13961422741413115,
|
|
"num_tokens": 4809710.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.511287641525269,
|
|
"epoch": 2.4455494778717055,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047175366073555654,
|
|
"loss": 6.3511,
|
|
"mean_token_accuracy": 0.14600509107112886,
|
|
"num_tokens": 4819387.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.471326446533203,
|
|
"epoch": 2.4505221282943808,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004715639200417125,
|
|
"loss": 6.29,
|
|
"mean_token_accuracy": 0.1439615435898304,
|
|
"num_tokens": 4829541.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.4300775051116945,
|
|
"epoch": 2.455494778717056,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004713735871780192,
|
|
"loss": 6.2899,
|
|
"mean_token_accuracy": 0.14560722708702087,
|
|
"num_tokens": 4838699.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.544748401641845,
|
|
"epoch": 2.4604674291397313,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004711826627178747,
|
|
"loss": 6.4294,
|
|
"mean_token_accuracy": 0.14071948751807212,
|
|
"num_tokens": 4848295.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.534548664093018,
|
|
"epoch": 2.4654400795624065,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047099114723645944,
|
|
"loss": 6.2891,
|
|
"mean_token_accuracy": 0.14699965268373488,
|
|
"num_tokens": 4857649.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.547018003463745,
|
|
"epoch": 2.470412729985082,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047079904131073434,
|
|
"loss": 6.3242,
|
|
"mean_token_accuracy": 0.1475421205163002,
|
|
"num_tokens": 4867332.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.512684297561646,
|
|
"epoch": 2.4753853804077575,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004706063455194389,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.1304320976138115,
|
|
"num_tokens": 4878726.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.583101511001587,
|
|
"epoch": 2.4803580308304327,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047041306044308986,
|
|
"loss": 6.4034,
|
|
"mean_token_accuracy": 0.13978520780801773,
|
|
"num_tokens": 4888030.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.534812021255493,
|
|
"epoch": 2.485330681253108,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004702191866639792,
|
|
"loss": 6.3693,
|
|
"mean_token_accuracy": 0.1370741046965122,
|
|
"num_tokens": 4897191.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.485330681253108,
|
|
"eval_entropy": 6.471034149969777,
|
|
"eval_loss": 6.821934700012207,
|
|
"eval_mean_token_accuracy": 0.1270321078581332,
|
|
"eval_num_tokens": 4897191.0,
|
|
"eval_runtime": 2.7348,
|
|
"eval_samples_per_second": 1267.006,
|
|
"eval_steps_per_second": 158.696,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.540355014801025,
|
|
"epoch": 2.490303331675783,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004700247247661726,
|
|
"loss": 6.3653,
|
|
"mean_token_accuracy": 0.14543634057044982,
|
|
"num_tokens": 4907154.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.318873071670533,
|
|
"epoch": 2.4952759820984585,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004698296753355072,
|
|
"loss": 6.0402,
|
|
"mean_token_accuracy": 0.1681458830833435,
|
|
"num_tokens": 4917582.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.464159154891968,
|
|
"epoch": 2.5002486325211337,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004696340389595903,
|
|
"loss": 6.2884,
|
|
"mean_token_accuracy": 0.1530615746974945,
|
|
"num_tokens": 4927473.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.559104919433594,
|
|
"epoch": 2.505221282943809,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046943781622779733,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.13506940230727196,
|
|
"num_tokens": 4938037.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.527531147003174,
|
|
"epoch": 2.510193933366484,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046924100773127055,
|
|
"loss": 6.398,
|
|
"mean_token_accuracy": 0.13462310209870337,
|
|
"num_tokens": 4948009.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.499567365646362,
|
|
"epoch": 2.5151665837891595,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046904361406291633,
|
|
"loss": 6.371,
|
|
"mean_token_accuracy": 0.14283587485551835,
|
|
"num_tokens": 4957971.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.621564435958862,
|
|
"epoch": 2.520139234211835,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004688456358174044,
|
|
"loss": 6.3498,
|
|
"mean_token_accuracy": 0.1347363695502281,
|
|
"num_tokens": 4968040.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.478594255447388,
|
|
"epoch": 2.5251118846345104,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004686470735911653,
|
|
"loss": 6.3009,
|
|
"mean_token_accuracy": 0.14506892040371894,
|
|
"num_tokens": 4977142.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.497954607009888,
|
|
"epoch": 2.5300845350571857,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004684479279823891,
|
|
"loss": 6.4449,
|
|
"mean_token_accuracy": 0.14204382449388503,
|
|
"num_tokens": 4986726.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.622348499298096,
|
|
"epoch": 2.535057185479861,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004682481995910232,
|
|
"loss": 6.3631,
|
|
"mean_token_accuracy": 0.13624119982123375,
|
|
"num_tokens": 4996690.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.484410619735717,
|
|
"epoch": 2.540029835902536,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046804788901877084,
|
|
"loss": 6.4084,
|
|
"mean_token_accuracy": 0.14261124581098555,
|
|
"num_tokens": 5005804.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.4933616638183596,
|
|
"epoch": 2.5450024863252114,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046784699686908886,
|
|
"loss": 6.3458,
|
|
"mean_token_accuracy": 0.14723647236824036,
|
|
"num_tokens": 5015666.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.542861223220825,
|
|
"epoch": 2.5499751367478867,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046764552374718656,
|
|
"loss": 6.2782,
|
|
"mean_token_accuracy": 0.15213844776153565,
|
|
"num_tokens": 5025897.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.4771161556243895,
|
|
"epoch": 2.554947787170562,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046744347026002315,
|
|
"loss": 6.3501,
|
|
"mean_token_accuracy": 0.14428422451019288,
|
|
"num_tokens": 5035195.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.505103015899659,
|
|
"epoch": 2.559920437593237,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004672408370163066,
|
|
"loss": 6.3556,
|
|
"mean_token_accuracy": 0.1479845479130745,
|
|
"num_tokens": 5044472.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.518973064422608,
|
|
"epoch": 2.5648930880159124,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000467037624626491,
|
|
"loss": 6.4082,
|
|
"mean_token_accuracy": 0.13811224997043609,
|
|
"num_tokens": 5053593.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.514780187606812,
|
|
"epoch": 2.5698657384385877,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046683383370277554,
|
|
"loss": 6.3632,
|
|
"mean_token_accuracy": 0.1389802947640419,
|
|
"num_tokens": 5063604.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.471893930435181,
|
|
"epoch": 2.574838388861263,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046662946485910226,
|
|
"loss": 6.2686,
|
|
"mean_token_accuracy": 0.14278726428747177,
|
|
"num_tokens": 5073013.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.501459407806396,
|
|
"epoch": 2.579811039283938,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000466424518711154,
|
|
"loss": 6.353,
|
|
"mean_token_accuracy": 0.14196558594703673,
|
|
"num_tokens": 5082590.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.558322525024414,
|
|
"epoch": 2.5847836897066134,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004662189958763531,
|
|
"loss": 6.3299,
|
|
"mean_token_accuracy": 0.14453306496143342,
|
|
"num_tokens": 5091860.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.4649756908416744,
|
|
"epoch": 2.5897563401292887,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004660128969738591,
|
|
"loss": 6.2484,
|
|
"mean_token_accuracy": 0.1543172523379326,
|
|
"num_tokens": 5101164.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.371197319030761,
|
|
"epoch": 2.594728990551964,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004658062226245668,
|
|
"loss": 6.3489,
|
|
"mean_token_accuracy": 0.14326864182949067,
|
|
"num_tokens": 5112885.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.522914695739746,
|
|
"epoch": 2.5997016409746396,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004655989734511052,
|
|
"loss": 6.3226,
|
|
"mean_token_accuracy": 0.14789563715457915,
|
|
"num_tokens": 5124063.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.366082334518433,
|
|
"epoch": 2.604674291397315,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004653911500778343,
|
|
"loss": 6.1761,
|
|
"mean_token_accuracy": 0.1581413634121418,
|
|
"num_tokens": 5134286.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.534527587890625,
|
|
"epoch": 2.60964694181999,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004651827531308443,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.14058353900909423,
|
|
"num_tokens": 5144421.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.610892343521118,
|
|
"epoch": 2.6146195922426654,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004649737832379534,
|
|
"loss": 6.3732,
|
|
"mean_token_accuracy": 0.1419479578733444,
|
|
"num_tokens": 5153466.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.443132543563843,
|
|
"epoch": 2.6195922426653406,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004647642410287057,
|
|
"loss": 6.4243,
|
|
"mean_token_accuracy": 0.14085619822144507,
|
|
"num_tokens": 5164034.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.504536724090576,
|
|
"epoch": 2.624564893088016,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004645541271343696,
|
|
"loss": 6.3538,
|
|
"mean_token_accuracy": 0.14277144595980645,
|
|
"num_tokens": 5173867.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.540040922164917,
|
|
"epoch": 2.629537543510691,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046434344218793566,
|
|
"loss": 6.399,
|
|
"mean_token_accuracy": 0.14028803780674934,
|
|
"num_tokens": 5183182.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.518236303329468,
|
|
"epoch": 2.6345101939333664,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046413218682411485,
|
|
"loss": 6.3741,
|
|
"mean_token_accuracy": 0.14554732367396356,
|
|
"num_tokens": 5192537.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.556453895568848,
|
|
"epoch": 2.6394828443560416,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004639203616793365,
|
|
"loss": 6.3197,
|
|
"mean_token_accuracy": 0.1478041559457779,
|
|
"num_tokens": 5201879.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.537673616409302,
|
|
"epoch": 2.6444554947787173,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046370796739174657,
|
|
"loss": 6.3579,
|
|
"mean_token_accuracy": 0.13904747143387794,
|
|
"num_tokens": 5211555.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.459664249420166,
|
|
"epoch": 2.6494281452013926,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046349500460120556,
|
|
"loss": 6.2529,
|
|
"mean_token_accuracy": 0.15000981390476226,
|
|
"num_tokens": 5220895.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.539349603652954,
|
|
"epoch": 2.654400795624068,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046328147394928684,
|
|
"loss": 6.4012,
|
|
"mean_token_accuracy": 0.14086081832647324,
|
|
"num_tokens": 5230101.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.559596300125122,
|
|
"epoch": 2.659373446046743,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004630673760792741,
|
|
"loss": 6.3814,
|
|
"mean_token_accuracy": 0.13394856974482536,
|
|
"num_tokens": 5239472.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.457580184936523,
|
|
"epoch": 2.6643460964694183,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004628527116361603,
|
|
"loss": 6.3162,
|
|
"mean_token_accuracy": 0.1445642292499542,
|
|
"num_tokens": 5248823.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.471943759918213,
|
|
"epoch": 2.6693187468920936,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004626374812666449,
|
|
"loss": 6.3291,
|
|
"mean_token_accuracy": 0.13978914096951484,
|
|
"num_tokens": 5258551.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.450301027297973,
|
|
"epoch": 2.674291397314769,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046242168561913256,
|
|
"loss": 6.1373,
|
|
"mean_token_accuracy": 0.1550034001469612,
|
|
"num_tokens": 5267933.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.545872449874878,
|
|
"epoch": 2.679264047737444,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046220532534373067,
|
|
"loss": 6.3955,
|
|
"mean_token_accuracy": 0.14434008225798606,
|
|
"num_tokens": 5277856.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.508489274978638,
|
|
"epoch": 2.6842366981601193,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004619884010922477,
|
|
"loss": 6.3651,
|
|
"mean_token_accuracy": 0.13867145925760269,
|
|
"num_tokens": 5288127.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.476871538162231,
|
|
"epoch": 2.6892093485827946,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004617709135181913,
|
|
"loss": 6.3386,
|
|
"mean_token_accuracy": 0.14043029323220252,
|
|
"num_tokens": 5297143.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.433478450775146,
|
|
"epoch": 2.69418199900547,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046155286327676594,
|
|
"loss": 6.315,
|
|
"mean_token_accuracy": 0.14332007616758347,
|
|
"num_tokens": 5307067.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.450258541107178,
|
|
"epoch": 2.699154649428145,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046133425102487134,
|
|
"loss": 6.2566,
|
|
"mean_token_accuracy": 0.14392766803503038,
|
|
"num_tokens": 5317471.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.559765481948853,
|
|
"epoch": 2.7041272998508203,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046111507742110035,
|
|
"loss": 6.3351,
|
|
"mean_token_accuracy": 0.15104418098926545,
|
|
"num_tokens": 5326803.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.460714960098267,
|
|
"epoch": 2.7090999502734956,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000460895343125737,
|
|
"loss": 6.311,
|
|
"mean_token_accuracy": 0.14636903926730155,
|
|
"num_tokens": 5336342.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.327510404586792,
|
|
"epoch": 2.714072600696171,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046067504880075425,
|
|
"loss": 6.1531,
|
|
"mean_token_accuracy": 0.16241607964038848,
|
|
"num_tokens": 5345694.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.50901460647583,
|
|
"epoch": 2.719045251118846,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004604541951098122,
|
|
"loss": 6.2289,
|
|
"mean_token_accuracy": 0.15025691464543342,
|
|
"num_tokens": 5355277.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.397147083282471,
|
|
"epoch": 2.724017901541522,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004602327827182565,
|
|
"loss": 6.2651,
|
|
"mean_token_accuracy": 0.14940724670886993,
|
|
"num_tokens": 5364650.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.462321996688843,
|
|
"epoch": 2.728990551964197,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004600108122931156,
|
|
"loss": 6.2447,
|
|
"mean_token_accuracy": 0.14868565499782563,
|
|
"num_tokens": 5373151.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.449673080444336,
|
|
"epoch": 2.7339632023868723,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00045978828450309907,
|
|
"loss": 6.3625,
|
|
"mean_token_accuracy": 0.14481861144304276,
|
|
"num_tokens": 5383093.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.545788192749024,
|
|
"epoch": 2.7389358528095475,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0004595652000185958,
|
|
"loss": 6.2861,
|
|
"mean_token_accuracy": 0.14802829027175904,
|
|
"num_tokens": 5392060.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.469894981384277,
|
|
"epoch": 2.743908503232223,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045934155951167163,
|
|
"loss": 6.3925,
|
|
"mean_token_accuracy": 0.1392727740108967,
|
|
"num_tokens": 5402119.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.519701528549194,
|
|
"epoch": 2.748881153654898,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004591173636560675,
|
|
"loss": 6.2894,
|
|
"mean_token_accuracy": 0.15115260183811188,
|
|
"num_tokens": 5411804.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.454624652862549,
|
|
"epoch": 2.7538538040775733,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004588926131271974,
|
|
"loss": 6.3411,
|
|
"mean_token_accuracy": 0.1430234581232071,
|
|
"num_tokens": 5423808.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.478226995468139,
|
|
"epoch": 2.7588264545002485,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045866730860214655,
|
|
"loss": 6.3122,
|
|
"mean_token_accuracy": 0.14709860384464263,
|
|
"num_tokens": 5432154.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.300262498855591,
|
|
"epoch": 2.763799104922924,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004584414507596687,
|
|
"loss": 6.0542,
|
|
"mean_token_accuracy": 0.17085577547550201,
|
|
"num_tokens": 5443006.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.417803001403809,
|
|
"epoch": 2.768771755345599,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045821504028018494,
|
|
"loss": 6.2355,
|
|
"mean_token_accuracy": 0.14497338235378265,
|
|
"num_tokens": 5453596.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.4752153873443605,
|
|
"epoch": 2.7737444057682747,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004579880778457811,
|
|
"loss": 6.3689,
|
|
"mean_token_accuracy": 0.14066243916749954,
|
|
"num_tokens": 5463048.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.467712926864624,
|
|
"epoch": 2.77871705619095,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004577605641402056,
|
|
"loss": 6.2947,
|
|
"mean_token_accuracy": 0.14769297540187837,
|
|
"num_tokens": 5472409.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.441144371032715,
|
|
"epoch": 2.7836897066136252,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004575324998488681,
|
|
"loss": 6.2354,
|
|
"mean_token_accuracy": 0.15422018766403198,
|
|
"num_tokens": 5481380.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.496242666244507,
|
|
"epoch": 2.7886623570363005,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004573038856588364,
|
|
"loss": 6.4379,
|
|
"mean_token_accuracy": 0.1381148561835289,
|
|
"num_tokens": 5491157.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.40307035446167,
|
|
"epoch": 2.7936350074589757,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004570747222588354,
|
|
"loss": 6.3196,
|
|
"mean_token_accuracy": 0.14346076548099518,
|
|
"num_tokens": 5500856.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.47411961555481,
|
|
"epoch": 2.798607657881651,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045684501033924443,
|
|
"loss": 6.2046,
|
|
"mean_token_accuracy": 0.14963931888341903,
|
|
"num_tokens": 5509654.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.401515960693359,
|
|
"epoch": 2.8035803083043263,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045661475059209503,
|
|
"loss": 6.2211,
|
|
"mean_token_accuracy": 0.15217607766389846,
|
|
"num_tokens": 5519009.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.397446584701538,
|
|
"epoch": 2.8085529587270015,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004563839437110695,
|
|
"loss": 6.2338,
|
|
"mean_token_accuracy": 0.15214656740427018,
|
|
"num_tokens": 5528316.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.471298694610596,
|
|
"epoch": 2.8135256091496768,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045615259039149826,
|
|
"loss": 6.3488,
|
|
"mean_token_accuracy": 0.1476473778486252,
|
|
"num_tokens": 5538416.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.4038327693939205,
|
|
"epoch": 2.818498259572352,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045592069133035783,
|
|
"loss": 6.2869,
|
|
"mean_token_accuracy": 0.1492191307246685,
|
|
"num_tokens": 5548600.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.528113079071045,
|
|
"epoch": 2.8234709099950273,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004556882472262691,
|
|
"loss": 6.3799,
|
|
"mean_token_accuracy": 0.1402927130460739,
|
|
"num_tokens": 5559454.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.46011061668396,
|
|
"epoch": 2.8284435604177025,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004554552587794948,
|
|
"loss": 6.3378,
|
|
"mean_token_accuracy": 0.13970609605312348,
|
|
"num_tokens": 5567748.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.495553874969483,
|
|
"epoch": 2.8334162108403778,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004552217266919375,
|
|
"loss": 6.4256,
|
|
"mean_token_accuracy": 0.13631942868232727,
|
|
"num_tokens": 5577617.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.5852532386779785,
|
|
"epoch": 2.838388861263053,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004549876516671376,
|
|
"loss": 6.3907,
|
|
"mean_token_accuracy": 0.14207628220319748,
|
|
"num_tokens": 5586600.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.4586786270141605,
|
|
"epoch": 2.8433615116857283,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045475303441027114,
|
|
"loss": 6.2216,
|
|
"mean_token_accuracy": 0.1577283725142479,
|
|
"num_tokens": 5596526.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.426614618301391,
|
|
"epoch": 2.8483341621084035,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004545178756281479,
|
|
"loss": 6.2625,
|
|
"mean_token_accuracy": 0.15038749873638152,
|
|
"num_tokens": 5605780.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.41419792175293,
|
|
"epoch": 2.853306812531079,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045428217602920876,
|
|
"loss": 6.3216,
|
|
"mean_token_accuracy": 0.14313239157199859,
|
|
"num_tokens": 5616109.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.429047727584839,
|
|
"epoch": 2.8582794629537545,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045404593632352385,
|
|
"loss": 6.275,
|
|
"mean_token_accuracy": 0.15239612609148026,
|
|
"num_tokens": 5625423.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.475142240524292,
|
|
"epoch": 2.8632521133764297,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004538091572227909,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.14628186374902724,
|
|
"num_tokens": 5634538.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.36572036743164,
|
|
"epoch": 2.868224763799105,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004535718394403319,
|
|
"loss": 6.1448,
|
|
"mean_token_accuracy": 0.16420113444328308,
|
|
"num_tokens": 5644087.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.331579542160034,
|
|
"epoch": 2.87319741422178,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004533339836910924,
|
|
"loss": 6.1773,
|
|
"mean_token_accuracy": 0.15930747836828232,
|
|
"num_tokens": 5654241.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.4233019828796385,
|
|
"epoch": 2.8781700646444555,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004530955906916382,
|
|
"loss": 6.2268,
|
|
"mean_token_accuracy": 0.1520026817917824,
|
|
"num_tokens": 5663827.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.43547625541687,
|
|
"epoch": 2.8831427150671307,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004528566611601536,
|
|
"loss": 6.3226,
|
|
"mean_token_accuracy": 0.14533980414271355,
|
|
"num_tokens": 5673427.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.36797866821289,
|
|
"epoch": 2.888115365489806,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004526171958164396,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.14768766909837722,
|
|
"num_tokens": 5684762.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.3772743225097654,
|
|
"epoch": 2.893088015912481,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045237719538191107,
|
|
"loss": 6.309,
|
|
"mean_token_accuracy": 0.14853577613830565,
|
|
"num_tokens": 5694938.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.417752361297607,
|
|
"epoch": 2.898060666335157,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000452136660579595,
|
|
"loss": 6.2306,
|
|
"mean_token_accuracy": 0.1528899848461151,
|
|
"num_tokens": 5704910.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.385972118377685,
|
|
"epoch": 2.903033316757832,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045189559213412825,
|
|
"loss": 6.3146,
|
|
"mean_token_accuracy": 0.14834412410855294,
|
|
"num_tokens": 5714325.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.4172662734985355,
|
|
"epoch": 2.9080059671805074,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004516539907717553,
|
|
"loss": 6.2152,
|
|
"mean_token_accuracy": 0.157168248295784,
|
|
"num_tokens": 5724382.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.403017520904541,
|
|
"epoch": 2.9129786176031827,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045141185722032617,
|
|
"loss": 6.2618,
|
|
"mean_token_accuracy": 0.1492059126496315,
|
|
"num_tokens": 5734277.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.416959381103515,
|
|
"epoch": 2.917951268025858,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045116919220929407,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.14605386927723885,
|
|
"num_tokens": 5742772.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.405718898773193,
|
|
"epoch": 2.922923918448533,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004509259964697133,
|
|
"loss": 6.2743,
|
|
"mean_token_accuracy": 0.14779454469680786,
|
|
"num_tokens": 5754011.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.556613206863403,
|
|
"epoch": 2.9278965688712084,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000450682270734237,
|
|
"loss": 6.3732,
|
|
"mean_token_accuracy": 0.13893503844738006,
|
|
"num_tokens": 5765445.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.4057858943939205,
|
|
"epoch": 2.9328692192938837,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004504380157371152,
|
|
"loss": 6.3165,
|
|
"mean_token_accuracy": 0.15030984133481978,
|
|
"num_tokens": 5775529.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.451057004928589,
|
|
"epoch": 2.937841869716559,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045019323221419204,
|
|
"loss": 6.4231,
|
|
"mean_token_accuracy": 0.14229757115244865,
|
|
"num_tokens": 5785594.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.43241868019104,
|
|
"epoch": 2.942814520139234,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044994792090290425,
|
|
"loss": 6.1403,
|
|
"mean_token_accuracy": 0.16009896695613862,
|
|
"num_tokens": 5796071.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.375617933273316,
|
|
"epoch": 2.9477871705619094,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044970208254227836,
|
|
"loss": 6.2772,
|
|
"mean_token_accuracy": 0.1521676555275917,
|
|
"num_tokens": 5805127.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.472930192947388,
|
|
"epoch": 2.9527598209845847,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044945571787292866,
|
|
"loss": 6.3395,
|
|
"mean_token_accuracy": 0.1493378311395645,
|
|
"num_tokens": 5815001.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.419169139862061,
|
|
"epoch": 2.95773247140726,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004492088276370552,
|
|
"loss": 6.2974,
|
|
"mean_token_accuracy": 0.14073021113872528,
|
|
"num_tokens": 5826915.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.408809566497803,
|
|
"epoch": 2.962705121829935,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004489614125784411,
|
|
"loss": 6.3013,
|
|
"mean_token_accuracy": 0.14516500532627105,
|
|
"num_tokens": 5837444.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.355221128463745,
|
|
"epoch": 2.9676777722526104,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004487134734424509,
|
|
"loss": 6.1628,
|
|
"mean_token_accuracy": 0.15315280109643936,
|
|
"num_tokens": 5846514.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.480633068084717,
|
|
"epoch": 2.9726504226752857,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004484650109760276,
|
|
"loss": 6.4218,
|
|
"mean_token_accuracy": 0.14197667092084884,
|
|
"num_tokens": 5857227.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.504110956192017,
|
|
"epoch": 2.9776230730979614,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004482160259276911,
|
|
"loss": 6.4022,
|
|
"mean_token_accuracy": 0.14497765749692917,
|
|
"num_tokens": 5866877.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.437817049026489,
|
|
"epoch": 2.9825957235206366,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004479665190475355,
|
|
"loss": 6.2861,
|
|
"mean_token_accuracy": 0.14751956462860108,
|
|
"num_tokens": 5877127.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9825957235206366,
|
|
"eval_entropy": 6.340936762945993,
|
|
"eval_loss": 6.701125144958496,
|
|
"eval_mean_token_accuracy": 0.13151445259524655,
|
|
"eval_num_tokens": 5877127.0,
|
|
"eval_runtime": 2.5452,
|
|
"eval_samples_per_second": 1361.38,
|
|
"eval_steps_per_second": 170.516,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.398074674606323,
|
|
"epoch": 2.987568373943312,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044771649108722695,
|
|
"loss": 6.3128,
|
|
"mean_token_accuracy": 0.14532861709594727,
|
|
"num_tokens": 5886434.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.389483833312989,
|
|
"epoch": 2.992541024365987,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004474659428000014,
|
|
"loss": 6.2912,
|
|
"mean_token_accuracy": 0.15597790628671646,
|
|
"num_tokens": 5895462.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.451540756225586,
|
|
"epoch": 2.9975136747886624,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044721487494066254,
|
|
"loss": 6.2394,
|
|
"mean_token_accuracy": 0.15187835693359375,
|
|
"num_tokens": 5905274.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.340188609229194,
|
|
"epoch": 3.00198906016907,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000446963288265579,
|
|
"loss": 5.9949,
|
|
"mean_token_accuracy": 0.16156921121809217,
|
|
"num_tokens": 5914164.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.394064807891846,
|
|
"epoch": 3.0069617105917454,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044671118353268257,
|
|
"loss": 5.8749,
|
|
"mean_token_accuracy": 0.15735693871974946,
|
|
"num_tokens": 5923159.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.308403873443604,
|
|
"epoch": 3.0119343610144207,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044645856150146593,
|
|
"loss": 5.8305,
|
|
"mean_token_accuracy": 0.16631314903497696,
|
|
"num_tokens": 5932708.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.30610146522522,
|
|
"epoch": 3.016907011437096,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044620542293297974,
|
|
"loss": 5.8641,
|
|
"mean_token_accuracy": 0.1618802011013031,
|
|
"num_tokens": 5941370.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.352138042449951,
|
|
"epoch": 3.021879661859771,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004459517685898312,
|
|
"loss": 5.9036,
|
|
"mean_token_accuracy": 0.16566911041736604,
|
|
"num_tokens": 5950567.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.388893699645996,
|
|
"epoch": 3.0268523122824464,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044569759923618113,
|
|
"loss": 5.9756,
|
|
"mean_token_accuracy": 0.15771342366933822,
|
|
"num_tokens": 5959242.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.324466896057129,
|
|
"epoch": 3.0318249627051217,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004454429156377419,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.16309747397899627,
|
|
"num_tokens": 5968365.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.359807634353638,
|
|
"epoch": 3.036797613127797,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004451877185617751,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.15564917623996735,
|
|
"num_tokens": 5978407.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.239137601852417,
|
|
"epoch": 3.041770263550472,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004449320087770894,
|
|
"loss": 5.6659,
|
|
"mean_token_accuracy": 0.18490099608898164,
|
|
"num_tokens": 5988865.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.2805784225463865,
|
|
"epoch": 3.046742913973148,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004446757870540376,
|
|
"loss": 5.851,
|
|
"mean_token_accuracy": 0.16899575367569925,
|
|
"num_tokens": 5998873.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.305179262161255,
|
|
"epoch": 3.051715564395823,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004444190541645153,
|
|
"loss": 5.89,
|
|
"mean_token_accuracy": 0.16174780428409577,
|
|
"num_tokens": 6007021.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.248967933654785,
|
|
"epoch": 3.0566882148184984,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00044416181088195776,
|
|
"loss": 5.7122,
|
|
"mean_token_accuracy": 0.17303745299577714,
|
|
"num_tokens": 6015884.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.237709283828735,
|
|
"epoch": 3.0616608652411736,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004439040579813379,
|
|
"loss": 5.9689,
|
|
"mean_token_accuracy": 0.15646934062242507,
|
|
"num_tokens": 6026110.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.072254180908203,
|
|
"epoch": 3.066633515663849,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044364579623916385,
|
|
"loss": 5.7213,
|
|
"mean_token_accuracy": 0.1819382682442665,
|
|
"num_tokens": 6037479.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.319383955001831,
|
|
"epoch": 3.071606166086524,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004433870264334767,
|
|
"loss": 5.8795,
|
|
"mean_token_accuracy": 0.15825727283954621,
|
|
"num_tokens": 6046743.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.270925521850586,
|
|
"epoch": 3.0765788165091994,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004431277493438484,
|
|
"loss": 5.9616,
|
|
"mean_token_accuracy": 0.1608537882566452,
|
|
"num_tokens": 6056982.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.2930597305297855,
|
|
"epoch": 3.0815514669318746,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004428679657513789,
|
|
"loss": 5.9251,
|
|
"mean_token_accuracy": 0.1641475811600685,
|
|
"num_tokens": 6066016.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.192068672180175,
|
|
"epoch": 3.08652411735455,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004426076764386939,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.15714501738548278,
|
|
"num_tokens": 6075149.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.311230945587158,
|
|
"epoch": 3.091496767777225,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044234688218994293,
|
|
"loss": 5.8356,
|
|
"mean_token_accuracy": 0.1684956133365631,
|
|
"num_tokens": 6085736.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.12915735244751,
|
|
"epoch": 3.0964694181999004,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004420855837907967,
|
|
"loss": 5.7768,
|
|
"mean_token_accuracy": 0.16718606948852538,
|
|
"num_tokens": 6096557.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.4019359111785885,
|
|
"epoch": 3.1014420686225757,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044182378202844427,
|
|
"loss": 6.0902,
|
|
"mean_token_accuracy": 0.1507420502603054,
|
|
"num_tokens": 6105735.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.39102954864502,
|
|
"epoch": 3.106414719045251,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044156147769159183,
|
|
"loss": 5.9196,
|
|
"mean_token_accuracy": 0.16611894369125366,
|
|
"num_tokens": 6114132.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.251520824432373,
|
|
"epoch": 3.1113873694679266,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004412986715704591,
|
|
"loss": 5.9318,
|
|
"mean_token_accuracy": 0.15787497460842131,
|
|
"num_tokens": 6124341.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.195437812805176,
|
|
"epoch": 3.116360019890602,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004410353644567777,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.16261177211999894,
|
|
"num_tokens": 6134418.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.287057399749756,
|
|
"epoch": 3.121332670313277,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044077155714378836,
|
|
"loss": 5.9001,
|
|
"mean_token_accuracy": 0.15995267778635025,
|
|
"num_tokens": 6143659.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.270061492919922,
|
|
"epoch": 3.1263053207359524,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004405072504262391,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.1607741340994835,
|
|
"num_tokens": 6152706.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.331454181671143,
|
|
"epoch": 3.1312779711586276,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000440242445100382,
|
|
"loss": 5.9809,
|
|
"mean_token_accuracy": 0.15174245834350586,
|
|
"num_tokens": 6161390.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.332275772094727,
|
|
"epoch": 3.136250621581303,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043997714196397143,
|
|
"loss": 5.9242,
|
|
"mean_token_accuracy": 0.1620955288410187,
|
|
"num_tokens": 6170982.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.339143848419189,
|
|
"epoch": 3.141223272003978,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043971134181626173,
|
|
"loss": 5.9776,
|
|
"mean_token_accuracy": 0.15909559056162834,
|
|
"num_tokens": 6179356.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.3014768123626705,
|
|
"epoch": 3.1461959224266534,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043944504545800396,
|
|
"loss": 5.9115,
|
|
"mean_token_accuracy": 0.16268313229084014,
|
|
"num_tokens": 6188270.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.296312236785889,
|
|
"epoch": 3.1511685728493286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004391782536914447,
|
|
"loss": 5.8739,
|
|
"mean_token_accuracy": 0.16030744910240174,
|
|
"num_tokens": 6197624.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.347575855255127,
|
|
"epoch": 3.156141223272004,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043891096732032257,
|
|
"loss": 5.9272,
|
|
"mean_token_accuracy": 0.15927883833646775,
|
|
"num_tokens": 6207174.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.13545241355896,
|
|
"epoch": 3.161113873694679,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043864318714986625,
|
|
"loss": 5.7614,
|
|
"mean_token_accuracy": 0.17330558001995086,
|
|
"num_tokens": 6217939.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.245046329498291,
|
|
"epoch": 3.1660865241173544,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043837491398679234,
|
|
"loss": 5.9426,
|
|
"mean_token_accuracy": 0.15701133906841278,
|
|
"num_tokens": 6227857.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.2896781921386715,
|
|
"epoch": 3.17105917454003,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004381061486393025,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.15053638741374015,
|
|
"num_tokens": 6236832.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.253927230834961,
|
|
"epoch": 3.1760318249627053,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004378368919170809,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.15767164677381515,
|
|
"num_tokens": 6245879.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.241726875305176,
|
|
"epoch": 3.1810044753853806,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043756714463129244,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.15676286220550537,
|
|
"num_tokens": 6256350.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.198462724685669,
|
|
"epoch": 3.185977125808056,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004372969075945796,
|
|
"loss": 5.7284,
|
|
"mean_token_accuracy": 0.17455279529094697,
|
|
"num_tokens": 6266085.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.287790966033936,
|
|
"epoch": 3.190949776230731,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004370261816210604,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.15546519458293914,
|
|
"num_tokens": 6276499.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.302801609039307,
|
|
"epoch": 3.1959224266534063,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004367549675263258,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.15671393871307374,
|
|
"num_tokens": 6286630.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.248741674423218,
|
|
"epoch": 3.2008950770760816,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004364832661274373,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.15730811208486556,
|
|
"num_tokens": 6295907.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.208601236343384,
|
|
"epoch": 3.205867727498757,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043621107824292444,
|
|
"loss": 5.8547,
|
|
"mean_token_accuracy": 0.15954403430223466,
|
|
"num_tokens": 6304304.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.187569808959961,
|
|
"epoch": 3.210840377921432,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004359384046927824,
|
|
"loss": 5.9667,
|
|
"mean_token_accuracy": 0.1540994629263878,
|
|
"num_tokens": 6315041.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.283197927474975,
|
|
"epoch": 3.2158130283441073,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004356652462984693,
|
|
"loss": 5.8468,
|
|
"mean_token_accuracy": 0.16110004335641862,
|
|
"num_tokens": 6324551.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.227240324020386,
|
|
"epoch": 3.2207856787667826,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043539160388290416,
|
|
"loss": 5.9145,
|
|
"mean_token_accuracy": 0.1652422845363617,
|
|
"num_tokens": 6335512.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.237622880935669,
|
|
"epoch": 3.225758329189458,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004351174782704641,
|
|
"loss": 5.9783,
|
|
"mean_token_accuracy": 0.15737072601914406,
|
|
"num_tokens": 6345987.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.113334465026855,
|
|
"epoch": 3.230730979612133,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043484287028698176,
|
|
"loss": 5.7628,
|
|
"mean_token_accuracy": 0.17646268904209136,
|
|
"num_tokens": 6357617.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.15558614730835,
|
|
"epoch": 3.2357036300348083,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043456778075974306,
|
|
"loss": 5.8489,
|
|
"mean_token_accuracy": 0.1625670999288559,
|
|
"num_tokens": 6366123.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.211847257614136,
|
|
"epoch": 3.240676280457484,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043429221051748487,
|
|
"loss": 5.9587,
|
|
"mean_token_accuracy": 0.1544474706053734,
|
|
"num_tokens": 6375776.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.3595692157745365,
|
|
"epoch": 3.2456489308801593,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043401616039039174,
|
|
"loss": 6.0042,
|
|
"mean_token_accuracy": 0.14871479719877242,
|
|
"num_tokens": 6386235.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.327882146835327,
|
|
"epoch": 3.2506215813028345,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004337396312100944,
|
|
"loss": 5.9654,
|
|
"mean_token_accuracy": 0.1534355416893959,
|
|
"num_tokens": 6396919.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.242443513870239,
|
|
"epoch": 3.2555942317255098,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004334626238096665,
|
|
"loss": 5.92,
|
|
"mean_token_accuracy": 0.1631858080625534,
|
|
"num_tokens": 6406769.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.1702470779418945,
|
|
"epoch": 3.260566882148185,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004331851390236227,
|
|
"loss": 5.9474,
|
|
"mean_token_accuracy": 0.16255069971084596,
|
|
"num_tokens": 6416208.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.277315378189087,
|
|
"epoch": 3.2655395325708603,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043290717768791533,
|
|
"loss": 5.896,
|
|
"mean_token_accuracy": 0.1659083768725395,
|
|
"num_tokens": 6426125.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.2302977561950685,
|
|
"epoch": 3.2705121829935355,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043262874063993285,
|
|
"loss": 5.8787,
|
|
"mean_token_accuracy": 0.15293095484375954,
|
|
"num_tokens": 6436671.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.230917596817017,
|
|
"epoch": 3.2754848334162108,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004323498287184967,
|
|
"loss": 5.9823,
|
|
"mean_token_accuracy": 0.15813344120979309,
|
|
"num_tokens": 6447396.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.268582439422607,
|
|
"epoch": 3.280457483838886,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043207044276385884,
|
|
"loss": 5.9319,
|
|
"mean_token_accuracy": 0.15607068538665772,
|
|
"num_tokens": 6457045.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.140562963485718,
|
|
"epoch": 3.2854301342615613,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043179058361769924,
|
|
"loss": 5.9539,
|
|
"mean_token_accuracy": 0.1630365252494812,
|
|
"num_tokens": 6466741.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.092462396621704,
|
|
"epoch": 3.2904027846842365,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004315102521231237,
|
|
"loss": 5.6903,
|
|
"mean_token_accuracy": 0.17723899930715561,
|
|
"num_tokens": 6477769.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.383918190002442,
|
|
"epoch": 3.295375435106912,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004312294491246609,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.15231009274721147,
|
|
"num_tokens": 6486991.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.201223230361938,
|
|
"epoch": 3.3003480855295875,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004309481754682597,
|
|
"loss": 5.9198,
|
|
"mean_token_accuracy": 0.16021429598331452,
|
|
"num_tokens": 6496772.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.219602870941162,
|
|
"epoch": 3.3053207359522627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043066643200128724,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.15603076964616774,
|
|
"num_tokens": 6507577.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.3069655418396,
|
|
"epoch": 3.310293386374938,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004303842195725259,
|
|
"loss": 5.9816,
|
|
"mean_token_accuracy": 0.15907459557056428,
|
|
"num_tokens": 6518040.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.228147172927857,
|
|
"epoch": 3.3152660367976132,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043010153903217074,
|
|
"loss": 5.9129,
|
|
"mean_token_accuracy": 0.15827998369932175,
|
|
"num_tokens": 6529261.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.192533159255982,
|
|
"epoch": 3.3202386872202885,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004298183912318273,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.16489026099443435,
|
|
"num_tokens": 6537527.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.279222059249878,
|
|
"epoch": 3.3252113376429637,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042953477702450845,
|
|
"loss": 5.9592,
|
|
"mean_token_accuracy": 0.15893186032772064,
|
|
"num_tokens": 6547205.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.2540380477905275,
|
|
"epoch": 3.330183988065639,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004292506972646324,
|
|
"loss": 6.0344,
|
|
"mean_token_accuracy": 0.1568805992603302,
|
|
"num_tokens": 6556328.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.263769674301147,
|
|
"epoch": 3.3351566384883142,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042896615280802006,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.15798300355672837,
|
|
"num_tokens": 6566414.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.155578088760376,
|
|
"epoch": 3.3401292889109895,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042868114451189164,
|
|
"loss": 5.7989,
|
|
"mean_token_accuracy": 0.17771409898996354,
|
|
"num_tokens": 6577365.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.248938512802124,
|
|
"epoch": 3.3451019393336647,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004283956732348655,
|
|
"loss": 5.9143,
|
|
"mean_token_accuracy": 0.15745315849781036,
|
|
"num_tokens": 6586647.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.279117202758789,
|
|
"epoch": 3.35007458975634,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042810973983695426,
|
|
"loss": 6.0087,
|
|
"mean_token_accuracy": 0.1535283401608467,
|
|
"num_tokens": 6596955.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.301529741287231,
|
|
"epoch": 3.3550472401790152,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004278233451795629,
|
|
"loss": 5.8908,
|
|
"mean_token_accuracy": 0.16477383077144622,
|
|
"num_tokens": 6606161.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.083950757980347,
|
|
"epoch": 3.3600198906016905,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000427536490125486,
|
|
"loss": 5.7778,
|
|
"mean_token_accuracy": 0.17313984781503677,
|
|
"num_tokens": 6616540.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.167257642745971,
|
|
"epoch": 3.364992541024366,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042724917553890507,
|
|
"loss": 5.9014,
|
|
"mean_token_accuracy": 0.16311111748218537,
|
|
"num_tokens": 6626109.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.247101974487305,
|
|
"epoch": 3.3699651914470414,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042696140228538616,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.16419617831707,
|
|
"num_tokens": 6635576.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.22693657875061,
|
|
"epoch": 3.3749378418697167,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042667317123187696,
|
|
"loss": 6.0019,
|
|
"mean_token_accuracy": 0.15808911621570587,
|
|
"num_tokens": 6645728.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.271794652938842,
|
|
"epoch": 3.379910492292392,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042638448324670424,
|
|
"loss": 5.8776,
|
|
"mean_token_accuracy": 0.16340596377849578,
|
|
"num_tokens": 6654538.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.158373880386352,
|
|
"epoch": 3.384883142715067,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004260953391995716,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.16797399520874023,
|
|
"num_tokens": 6664811.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.19985933303833,
|
|
"epoch": 3.3898557931377424,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042580573996155646,
|
|
"loss": 5.9396,
|
|
"mean_token_accuracy": 0.16423666924238206,
|
|
"num_tokens": 6674185.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.239108467102051,
|
|
"epoch": 3.3948284435604177,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004255156864051074,
|
|
"loss": 5.9207,
|
|
"mean_token_accuracy": 0.1551107332110405,
|
|
"num_tokens": 6684103.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.03340106010437,
|
|
"epoch": 3.399801093983093,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042522517940404186,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.18921671211719512,
|
|
"num_tokens": 6693631.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.243697547912598,
|
|
"epoch": 3.404773744405768,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004249342198335433,
|
|
"loss": 5.962,
|
|
"mean_token_accuracy": 0.1601146638393402,
|
|
"num_tokens": 6704465.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.178900337219238,
|
|
"epoch": 3.4097463948284434,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042464280857015856,
|
|
"loss": 5.9297,
|
|
"mean_token_accuracy": 0.16500024497509003,
|
|
"num_tokens": 6713592.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.232093381881714,
|
|
"epoch": 3.4147190452511187,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042435094649179525,
|
|
"loss": 5.9642,
|
|
"mean_token_accuracy": 0.15766821950674056,
|
|
"num_tokens": 6726252.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.284474229812622,
|
|
"epoch": 3.419691695673794,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004240586344777191,
|
|
"loss": 6.0436,
|
|
"mean_token_accuracy": 0.15264040231704712,
|
|
"num_tokens": 6736301.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.282935619354248,
|
|
"epoch": 3.4246643460964696,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004237658734085514,
|
|
"loss": 5.9903,
|
|
"mean_token_accuracy": 0.15537417978048323,
|
|
"num_tokens": 6745706.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.241247701644897,
|
|
"epoch": 3.429636996519145,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042347266416626624,
|
|
"loss": 5.9844,
|
|
"mean_token_accuracy": 0.15316892564296722,
|
|
"num_tokens": 6755811.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.303303241729736,
|
|
"epoch": 3.43460964694182,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004231790076341877,
|
|
"loss": 5.9194,
|
|
"mean_token_accuracy": 0.16049495637416838,
|
|
"num_tokens": 6766365.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.1602260112762455,
|
|
"epoch": 3.4395822973644954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004228849046969876,
|
|
"loss": 5.8743,
|
|
"mean_token_accuracy": 0.16943498998880385,
|
|
"num_tokens": 6775806.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.2135313034057615,
|
|
"epoch": 3.4445549477871706,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004225903562406826,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.15922852456569672,
|
|
"num_tokens": 6785996.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.222315454483033,
|
|
"epoch": 3.449527598209846,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004222953631526313,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.17208046913146974,
|
|
"num_tokens": 6795528.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.184564733505249,
|
|
"epoch": 3.454500248632521,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000421999926321532,
|
|
"loss": 5.8814,
|
|
"mean_token_accuracy": 0.16285580843687059,
|
|
"num_tokens": 6804592.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.1636364459991455,
|
|
"epoch": 3.4594728990551964,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004217040466374199,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.16223583221435547,
|
|
"num_tokens": 6813242.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.266526794433593,
|
|
"epoch": 3.4644455494778716,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042140772499166403,
|
|
"loss": 5.9562,
|
|
"mean_token_accuracy": 0.15989834815263748,
|
|
"num_tokens": 6823223.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.076566362380982,
|
|
"epoch": 3.469418199900547,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004211109622769652,
|
|
"loss": 5.774,
|
|
"mean_token_accuracy": 0.16934829950332642,
|
|
"num_tokens": 6832425.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.163863039016723,
|
|
"epoch": 3.474390850323222,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042081375938735283,
|
|
"loss": 5.9707,
|
|
"mean_token_accuracy": 0.15563918948173522,
|
|
"num_tokens": 6842549.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.238685083389282,
|
|
"epoch": 3.4793635007458974,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042051611721818233,
|
|
"loss": 6.0601,
|
|
"mean_token_accuracy": 0.15095847398042678,
|
|
"num_tokens": 6851652.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.4793635007458974,
|
|
"eval_entropy": 6.172711926121866,
|
|
"eval_loss": 6.67270040512085,
|
|
"eval_mean_token_accuracy": 0.13566318958620047,
|
|
"eval_num_tokens": 6851652.0,
|
|
"eval_runtime": 2.5275,
|
|
"eval_samples_per_second": 1370.934,
|
|
"eval_steps_per_second": 171.713,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.201852750778198,
|
|
"epoch": 3.4843361511685726,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042021803666613273,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.16534167975187303,
|
|
"num_tokens": 6860716.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.16337308883667,
|
|
"epoch": 3.489308801591248,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041991951862920365,
|
|
"loss": 5.9095,
|
|
"mean_token_accuracy": 0.15994026511907578,
|
|
"num_tokens": 6870074.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.331715488433838,
|
|
"epoch": 3.4942814520139236,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004196205640067124,
|
|
"loss": 6.1444,
|
|
"mean_token_accuracy": 0.14215195477008818,
|
|
"num_tokens": 6880569.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.314612197875976,
|
|
"epoch": 3.499254102436599,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000419321173699292,
|
|
"loss": 5.9999,
|
|
"mean_token_accuracy": 0.1541007474064827,
|
|
"num_tokens": 6890522.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.205518817901611,
|
|
"epoch": 3.504226752859274,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004190213486088877,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.1579175189137459,
|
|
"num_tokens": 6900388.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.2201203346252445,
|
|
"epoch": 3.5091994032819493,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004187210896387546,
|
|
"loss": 5.9266,
|
|
"mean_token_accuracy": 0.16327783614397048,
|
|
"num_tokens": 6910467.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.187130546569824,
|
|
"epoch": 3.5141720537046246,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004184203976934552,
|
|
"loss": 5.958,
|
|
"mean_token_accuracy": 0.16534241512417794,
|
|
"num_tokens": 6919370.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.0957060813903805,
|
|
"epoch": 3.5191447041273,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041811927367885605,
|
|
"loss": 5.7468,
|
|
"mean_token_accuracy": 0.17413049638271333,
|
|
"num_tokens": 6929688.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.214923286437989,
|
|
"epoch": 3.524117354549975,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004178177185021254,
|
|
"loss": 5.993,
|
|
"mean_token_accuracy": 0.1586630493402481,
|
|
"num_tokens": 6939861.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.277981901168824,
|
|
"epoch": 3.5290900049726504,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004175157330717305,
|
|
"loss": 6.012,
|
|
"mean_token_accuracy": 0.152120178937912,
|
|
"num_tokens": 6949746.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.279894161224365,
|
|
"epoch": 3.5340626553953256,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000417213318297435,
|
|
"loss": 5.9477,
|
|
"mean_token_accuracy": 0.15829975455999373,
|
|
"num_tokens": 6961408.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.155437850952149,
|
|
"epoch": 3.539035305818001,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041691047509029555,
|
|
"loss": 5.9903,
|
|
"mean_token_accuracy": 0.16253448277711868,
|
|
"num_tokens": 6972003.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.238606929779053,
|
|
"epoch": 3.5440079562406765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041660720436265987,
|
|
"loss": 5.8919,
|
|
"mean_token_accuracy": 0.16272868663072587,
|
|
"num_tokens": 6983528.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.24536771774292,
|
|
"epoch": 3.548980606663352,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004163035070281634,
|
|
"loss": 5.9737,
|
|
"mean_token_accuracy": 0.16169923469424247,
|
|
"num_tokens": 6992766.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.18948974609375,
|
|
"epoch": 3.553953257086027,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041599938400172686,
|
|
"loss": 5.8557,
|
|
"mean_token_accuracy": 0.1610173135995865,
|
|
"num_tokens": 7001797.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.242516040802002,
|
|
"epoch": 3.5589259075087023,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041569483619955354,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.16617249846458435,
|
|
"num_tokens": 7011215.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.202178335189819,
|
|
"epoch": 3.5638985579313776,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041538986453912616,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.15492074564099312,
|
|
"num_tokens": 7021513.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.168223142623901,
|
|
"epoch": 3.568871208354053,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004150844699392045,
|
|
"loss": 5.9705,
|
|
"mean_token_accuracy": 0.15568189769983293,
|
|
"num_tokens": 7031654.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.140969467163086,
|
|
"epoch": 3.573843858776728,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041477865331982264,
|
|
"loss": 5.9243,
|
|
"mean_token_accuracy": 0.1648896038532257,
|
|
"num_tokens": 7040614.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.2059790134429935,
|
|
"epoch": 3.5788165091994033,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004144724156022857,
|
|
"loss": 5.9264,
|
|
"mean_token_accuracy": 0.16396153271198272,
|
|
"num_tokens": 7050877.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.350143241882324,
|
|
"epoch": 3.5837891596220786,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041416575770916763,
|
|
"loss": 6.1312,
|
|
"mean_token_accuracy": 0.14578675478696823,
|
|
"num_tokens": 7061330.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.270376348495484,
|
|
"epoch": 3.588761810044754,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041385868056430827,
|
|
"loss": 6.0061,
|
|
"mean_token_accuracy": 0.15612463057041168,
|
|
"num_tokens": 7071290.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.197152137756348,
|
|
"epoch": 3.593734460467429,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004135511850928102,
|
|
"loss": 5.9823,
|
|
"mean_token_accuracy": 0.15387657284736633,
|
|
"num_tokens": 7080112.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.247121429443359,
|
|
"epoch": 3.5987071108901043,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041324327222103655,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.14585825949907302,
|
|
"num_tokens": 7091467.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.221949338912964,
|
|
"epoch": 3.6036797613127796,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00041293494287660785,
|
|
"loss": 5.7527,
|
|
"mean_token_accuracy": 0.17463432848453522,
|
|
"num_tokens": 7101744.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.246480560302734,
|
|
"epoch": 3.608652411735455,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041262619798839914,
|
|
"loss": 5.9563,
|
|
"mean_token_accuracy": 0.16098868250846862,
|
|
"num_tokens": 7112854.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.0680344581604,
|
|
"epoch": 3.61362506215813,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041231703848653746,
|
|
"loss": 5.9483,
|
|
"mean_token_accuracy": 0.16464419662952423,
|
|
"num_tokens": 7121613.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.24086217880249,
|
|
"epoch": 3.6185977125808053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041200746530239903,
|
|
"loss": 5.9938,
|
|
"mean_token_accuracy": 0.15724552422761917,
|
|
"num_tokens": 7131357.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.184099435806274,
|
|
"epoch": 3.623570363003481,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004116974793686061,
|
|
"loss": 5.8765,
|
|
"mean_token_accuracy": 0.17357155233621596,
|
|
"num_tokens": 7140200.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.190071725845337,
|
|
"epoch": 3.6285430134261563,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004113870816190244,
|
|
"loss": 5.9801,
|
|
"mean_token_accuracy": 0.16244304478168486,
|
|
"num_tokens": 7149367.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.152639007568359,
|
|
"epoch": 3.6335156638488315,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004110762729887605,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.16452273428440095,
|
|
"num_tokens": 7158088.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.190124607086181,
|
|
"epoch": 3.6384883142715068,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041076505441415873,
|
|
"loss": 5.9339,
|
|
"mean_token_accuracy": 0.16660187393426895,
|
|
"num_tokens": 7168141.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.254987573623657,
|
|
"epoch": 3.643460964694182,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004104534268327982,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.1626141905784607,
|
|
"num_tokens": 7177646.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.238340234756469,
|
|
"epoch": 3.6484336151168573,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004101413911834904,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.1559474915266037,
|
|
"num_tokens": 7186445.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.2091888904571535,
|
|
"epoch": 3.6534062655395325,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040982894840627614,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.16231908202171325,
|
|
"num_tokens": 7195812.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.149083423614502,
|
|
"epoch": 3.6583789159622078,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004095160994424228,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.1724393531680107,
|
|
"num_tokens": 7204756.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.216005420684814,
|
|
"epoch": 3.663351566384883,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040920284523442125,
|
|
"loss": 5.9843,
|
|
"mean_token_accuracy": 0.16606335788965226,
|
|
"num_tokens": 7214708.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.237343263626099,
|
|
"epoch": 3.6683242168075583,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040888918672598337,
|
|
"loss": 6.0235,
|
|
"mean_token_accuracy": 0.15607056766748428,
|
|
"num_tokens": 7224623.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.20723705291748,
|
|
"epoch": 3.673296867230234,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000408575124862039,
|
|
"loss": 5.925,
|
|
"mean_token_accuracy": 0.16391729712486267,
|
|
"num_tokens": 7235228.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.186968851089477,
|
|
"epoch": 3.678269517652909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004082606605887331,
|
|
"loss": 6.0144,
|
|
"mean_token_accuracy": 0.16542051434516908,
|
|
"num_tokens": 7245318.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.970559120178223,
|
|
"epoch": 3.6832421680755845,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040794579485342304,
|
|
"loss": 5.5855,
|
|
"mean_token_accuracy": 0.19518422335386276,
|
|
"num_tokens": 7255468.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.256650352478028,
|
|
"epoch": 3.6882148184982597,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040763052860467547,
|
|
"loss": 5.9628,
|
|
"mean_token_accuracy": 0.15635033398866655,
|
|
"num_tokens": 7265006.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.160571527481079,
|
|
"epoch": 3.693187468920935,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004073148627922637,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.1624176725745201,
|
|
"num_tokens": 7273835.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.266654777526855,
|
|
"epoch": 3.69816011934361,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004069987983671648,
|
|
"loss": 6.0389,
|
|
"mean_token_accuracy": 0.1518197163939476,
|
|
"num_tokens": 7284617.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.080435180664063,
|
|
"epoch": 3.7031327697662855,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040668233628155664,
|
|
"loss": 5.7677,
|
|
"mean_token_accuracy": 0.17647801116108894,
|
|
"num_tokens": 7293707.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.213000202178955,
|
|
"epoch": 3.7081054201889607,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004063654774888153,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.14929335862398146,
|
|
"num_tokens": 7303227.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.149737310409546,
|
|
"epoch": 3.713078070611636,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004060482229435116,
|
|
"loss": 5.8579,
|
|
"mean_token_accuracy": 0.17009468525648117,
|
|
"num_tokens": 7312782.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.218086004257202,
|
|
"epoch": 3.7180507210343112,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004057305736014089,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.15850173234939574,
|
|
"num_tokens": 7322249.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.258641719818115,
|
|
"epoch": 3.7230233714569865,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004054125304194598,
|
|
"loss": 5.9575,
|
|
"mean_token_accuracy": 0.16373585313558578,
|
|
"num_tokens": 7331076.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.139318656921387,
|
|
"epoch": 3.7279960218796617,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040509409435580355,
|
|
"loss": 5.9144,
|
|
"mean_token_accuracy": 0.16399238407611846,
|
|
"num_tokens": 7340039.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.157795095443726,
|
|
"epoch": 3.732968672302337,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004047752663697627,
|
|
"loss": 5.9026,
|
|
"mean_token_accuracy": 0.16509526073932648,
|
|
"num_tokens": 7349300.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.329801225662232,
|
|
"epoch": 3.7379413227250122,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004044560474218409,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.16038140654563904,
|
|
"num_tokens": 7358363.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.180526304244995,
|
|
"epoch": 3.7429139731476875,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040413643847371923,
|
|
"loss": 5.9073,
|
|
"mean_token_accuracy": 0.1686038687825203,
|
|
"num_tokens": 7369391.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.182810735702515,
|
|
"epoch": 3.7478866235703627,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004038164404882539,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.1539320930838585,
|
|
"num_tokens": 7378272.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.178169631958008,
|
|
"epoch": 3.7528592739930384,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004034960544294731,
|
|
"loss": 5.8548,
|
|
"mean_token_accuracy": 0.16293878555297853,
|
|
"num_tokens": 7387590.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.199071216583252,
|
|
"epoch": 3.7578319244157137,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040317528126257413,
|
|
"loss": 5.9395,
|
|
"mean_token_accuracy": 0.15719935297966003,
|
|
"num_tokens": 7397733.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.15965461730957,
|
|
"epoch": 3.762804574838389,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004028541219539205,
|
|
"loss": 5.8109,
|
|
"mean_token_accuracy": 0.16741475462913513,
|
|
"num_tokens": 7406748.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.16231632232666,
|
|
"epoch": 3.767777225261064,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004025325774710391,
|
|
"loss": 6.0358,
|
|
"mean_token_accuracy": 0.15851458609104158,
|
|
"num_tokens": 7416419.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.185123634338379,
|
|
"epoch": 3.7727498756837394,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040221064878261684,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.16041389405727385,
|
|
"num_tokens": 7425257.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.274077224731445,
|
|
"epoch": 3.7777225261064147,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004018883368584985,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.15457284897565843,
|
|
"num_tokens": 7436914.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.19541301727295,
|
|
"epoch": 3.78269517652909,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040156564266968336,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.16219560652971268,
|
|
"num_tokens": 7447568.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.157040357589722,
|
|
"epoch": 3.787667826951765,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040124256718832214,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.16297836005687713,
|
|
"num_tokens": 7457717.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.241963338851929,
|
|
"epoch": 3.7926404773744404,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004009191113877141,
|
|
"loss": 5.9447,
|
|
"mean_token_accuracy": 0.1600917473435402,
|
|
"num_tokens": 7467123.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.153872013092041,
|
|
"epoch": 3.797613127797116,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004005952762423048,
|
|
"loss": 5.9096,
|
|
"mean_token_accuracy": 0.165580677986145,
|
|
"num_tokens": 7476987.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.235847520828247,
|
|
"epoch": 3.8025857782197914,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004002710627276821,
|
|
"loss": 6.0076,
|
|
"mean_token_accuracy": 0.16175391674041747,
|
|
"num_tokens": 7487024.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.230613899230957,
|
|
"epoch": 3.8075584286424666,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003999464718205739,
|
|
"loss": 6.0186,
|
|
"mean_token_accuracy": 0.16084508448839188,
|
|
"num_tokens": 7496826.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.1878491878509525,
|
|
"epoch": 3.812531079065142,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003996215044988451,
|
|
"loss": 5.9516,
|
|
"mean_token_accuracy": 0.15903948694467546,
|
|
"num_tokens": 7506981.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.100557088851929,
|
|
"epoch": 3.817503729487817,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00039929616174149454,
|
|
"loss": 5.7443,
|
|
"mean_token_accuracy": 0.17773568779230117,
|
|
"num_tokens": 7516925.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.201454401016235,
|
|
"epoch": 3.8224763799104924,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039897044452865207,
|
|
"loss": 5.9044,
|
|
"mean_token_accuracy": 0.16549646556377412,
|
|
"num_tokens": 7525789.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.124452829360962,
|
|
"epoch": 3.8274490303331676,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003986443538415756,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.16523757129907607,
|
|
"num_tokens": 7535907.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.203757095336914,
|
|
"epoch": 3.832421680755843,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003983178906626484,
|
|
"loss": 5.8838,
|
|
"mean_token_accuracy": 0.16845598071813583,
|
|
"num_tokens": 7546985.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.063313245773315,
|
|
"epoch": 3.837394331178518,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039799105597537564,
|
|
"loss": 5.7248,
|
|
"mean_token_accuracy": 0.18246889263391494,
|
|
"num_tokens": 7556912.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.163967084884644,
|
|
"epoch": 3.8423669816011934,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003976638507643818,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.1610849991440773,
|
|
"num_tokens": 7567156.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.187885379791259,
|
|
"epoch": 3.8473396320238686,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003973362760154076,
|
|
"loss": 5.9082,
|
|
"mean_token_accuracy": 0.17026485204696656,
|
|
"num_tokens": 7576423.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.229396629333496,
|
|
"epoch": 3.852312282446544,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003970083327153071,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.16606468260288237,
|
|
"num_tokens": 7585612.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.051190519332886,
|
|
"epoch": 3.857284932869219,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003966800218520445,
|
|
"loss": 5.925,
|
|
"mean_token_accuracy": 0.16799391359090804,
|
|
"num_tokens": 7595581.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.238703393936158,
|
|
"epoch": 3.8622575832918944,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003963513444146914,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.16054464280605316,
|
|
"num_tokens": 7605202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.2128133296966555,
|
|
"epoch": 3.8672302337145696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003960223013934239,
|
|
"loss": 5.8363,
|
|
"mean_token_accuracy": 0.1716415360569954,
|
|
"num_tokens": 7615783.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.20598292350769,
|
|
"epoch": 3.872202884137245,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003956928937795192,
|
|
"loss": 6.0374,
|
|
"mean_token_accuracy": 0.1488328218460083,
|
|
"num_tokens": 7626609.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.1895088195800785,
|
|
"epoch": 3.8771755345599206,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000395363122565353,
|
|
"loss": 5.9054,
|
|
"mean_token_accuracy": 0.16600026860833167,
|
|
"num_tokens": 7636676.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.227712249755859,
|
|
"epoch": 3.882148184982596,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039503298874439655,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.1655281513929367,
|
|
"num_tokens": 7646523.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.089673137664795,
|
|
"epoch": 3.887120835405271,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003947024933112131,
|
|
"loss": 5.8043,
|
|
"mean_token_accuracy": 0.1747158259153366,
|
|
"num_tokens": 7656267.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.189454221725464,
|
|
"epoch": 3.8920934858279463,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003943716372614556,
|
|
"loss": 5.8879,
|
|
"mean_token_accuracy": 0.16677381247282028,
|
|
"num_tokens": 7665529.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.224945831298828,
|
|
"epoch": 3.8970661362506216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003940404215918634,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.16120045632123947,
|
|
"num_tokens": 7675477.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.92595272064209,
|
|
"epoch": 3.902038786673297,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00039370884730025897,
|
|
"loss": 5.5843,
|
|
"mean_token_accuracy": 0.19233729541301728,
|
|
"num_tokens": 7685675.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.115845489501953,
|
|
"epoch": 3.907011437095972,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003933769153855455,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.1654975861310959,
|
|
"num_tokens": 7695821.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.182166719436646,
|
|
"epoch": 3.9119840875186473,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003930446268477034,
|
|
"loss": 5.9437,
|
|
"mean_token_accuracy": 0.16409701406955718,
|
|
"num_tokens": 7706163.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.1722980499267575,
|
|
"epoch": 3.9169567379413226,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003927119826877875,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.16642989367246627,
|
|
"num_tokens": 7715125.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.141762924194336,
|
|
"epoch": 3.921929388363998,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003923789839079238,
|
|
"loss": 5.8855,
|
|
"mean_token_accuracy": 0.16860341429710388,
|
|
"num_tokens": 7724546.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.987500953674316,
|
|
"epoch": 3.9269020387866735,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00039204563151130683,
|
|
"loss": 5.7059,
|
|
"mean_token_accuracy": 0.17774517983198165,
|
|
"num_tokens": 7734829.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.170799255371094,
|
|
"epoch": 3.931874689209349,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003917119265021965,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.1613735541701317,
|
|
"num_tokens": 7744612.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.208910751342773,
|
|
"epoch": 3.936847339632024,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003913778698859146,
|
|
"loss": 5.9705,
|
|
"mean_token_accuracy": 0.1621137499809265,
|
|
"num_tokens": 7754042.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.196197128295898,
|
|
"epoch": 3.9418199900546993,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003910434626688427,
|
|
"loss": 5.8676,
|
|
"mean_token_accuracy": 0.16624239832162857,
|
|
"num_tokens": 7762608.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.097027254104614,
|
|
"epoch": 3.9467926404773745,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003907087058584183,
|
|
"loss": 5.8146,
|
|
"mean_token_accuracy": 0.1681312456727028,
|
|
"num_tokens": 7771477.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.09872784614563,
|
|
"epoch": 3.95176529090005,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00039037360046313186,
|
|
"loss": 5.8922,
|
|
"mean_token_accuracy": 0.17521725371479988,
|
|
"num_tokens": 7782785.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.3034168720245365,
|
|
"epoch": 3.956737941322725,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039003814749252454,
|
|
"loss": 6.0725,
|
|
"mean_token_accuracy": 0.15178293958306313,
|
|
"num_tokens": 7793572.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.195157480239868,
|
|
"epoch": 3.9617105917454003,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038970234795718424,
|
|
"loss": 5.856,
|
|
"mean_token_accuracy": 0.16383777558803558,
|
|
"num_tokens": 7803170.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.1456903457641605,
|
|
"epoch": 3.9666832421680756,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000389366202868743,
|
|
"loss": 6.0241,
|
|
"mean_token_accuracy": 0.1551950082182884,
|
|
"num_tokens": 7813445.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.195073509216309,
|
|
"epoch": 3.971655892590751,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003890297132398739,
|
|
"loss": 6.0116,
|
|
"mean_token_accuracy": 0.16142124235630034,
|
|
"num_tokens": 7823089.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.200736808776855,
|
|
"epoch": 3.976628543013426,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003886928800842879,
|
|
"loss": 5.8244,
|
|
"mean_token_accuracy": 0.1657528504729271,
|
|
"num_tokens": 7832487.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.976628543013426,
|
|
"eval_entropy": 6.082162658190398,
|
|
"eval_loss": 6.595086574554443,
|
|
"eval_mean_token_accuracy": 0.13894746185619436,
|
|
"eval_num_tokens": 7832487.0,
|
|
"eval_runtime": 2.5484,
|
|
"eval_samples_per_second": 1359.685,
|
|
"eval_steps_per_second": 170.304,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.283151583872e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|