28185ab911
Model: fpadovani/tur-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.9945300845350572,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3220538609664000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|