7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.899751449875725,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691956996917725,
|
|
"epoch": 0.004142502071251036,
|
|
"grad_norm": 12.6875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.786,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8553.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69189453125,
|
|
"epoch": 0.008285004142502071,
|
|
"grad_norm": 13.0625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7413,
|
|
"mean_token_accuracy": 0.00027811190811917187,
|
|
"num_tokens": 18133.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691031455993652,
|
|
"epoch": 0.012427506213753107,
|
|
"grad_norm": 9.5625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4928,
|
|
"mean_token_accuracy": 0.023286720411852,
|
|
"num_tokens": 27152.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.681630420684815,
|
|
"epoch": 0.016570008285004142,
|
|
"grad_norm": 6.40625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2019,
|
|
"mean_token_accuracy": 0.042532872036099435,
|
|
"num_tokens": 35844.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.63299388885498,
|
|
"epoch": 0.020712510356255178,
|
|
"grad_norm": 4.46875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8879,
|
|
"mean_token_accuracy": 0.04524078518152237,
|
|
"num_tokens": 44730.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.594239139556885,
|
|
"epoch": 0.024855012427506214,
|
|
"grad_norm": 3.6875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7944,
|
|
"mean_token_accuracy": 0.04270199164748192,
|
|
"num_tokens": 53751.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.599893474578858,
|
|
"epoch": 0.02899751449875725,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6813,
|
|
"mean_token_accuracy": 0.0431942991912365,
|
|
"num_tokens": 62399.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.593089294433593,
|
|
"epoch": 0.033140016570008285,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.675,
|
|
"mean_token_accuracy": 0.045351412147283554,
|
|
"num_tokens": 72384.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.579291725158692,
|
|
"epoch": 0.037282518641259324,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.625,
|
|
"mean_token_accuracy": 0.03919882737100124,
|
|
"num_tokens": 81720.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.570260047912598,
|
|
"epoch": 0.041425020712510356,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5545,
|
|
"mean_token_accuracy": 0.045346799120306966,
|
|
"num_tokens": 91634.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.559380340576173,
|
|
"epoch": 0.045567522783761395,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4963,
|
|
"mean_token_accuracy": 0.048221738263964656,
|
|
"num_tokens": 100359.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.541158962249757,
|
|
"epoch": 0.04971002485501243,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4361,
|
|
"mean_token_accuracy": 0.05027815327048302,
|
|
"num_tokens": 108678.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.408777332305908,
|
|
"epoch": 0.053852526926263466,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3154,
|
|
"mean_token_accuracy": 0.055302799120545386,
|
|
"num_tokens": 117022.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.43354835510254,
|
|
"epoch": 0.0579950289975145,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2758,
|
|
"mean_token_accuracy": 0.05044409707188606,
|
|
"num_tokens": 126489.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.479030323028564,
|
|
"epoch": 0.06213753106876554,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1042,
|
|
"mean_token_accuracy": 0.060464993864297864,
|
|
"num_tokens": 135926.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.352683162689209,
|
|
"epoch": 0.06628003314001657,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0851,
|
|
"mean_token_accuracy": 0.05333640910685063,
|
|
"num_tokens": 144488.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.436153030395507,
|
|
"epoch": 0.07042253521126761,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9888,
|
|
"mean_token_accuracy": 0.05590510331094265,
|
|
"num_tokens": 154180.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.357822036743164,
|
|
"epoch": 0.07456503728251865,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9305,
|
|
"mean_token_accuracy": 0.05672054700553417,
|
|
"num_tokens": 163448.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.390939235687256,
|
|
"epoch": 0.07870753935376967,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8567,
|
|
"mean_token_accuracy": 0.054386957362294194,
|
|
"num_tokens": 173318.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.244186305999756,
|
|
"epoch": 0.08285004142502071,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.745,
|
|
"mean_token_accuracy": 0.05825656205415726,
|
|
"num_tokens": 181540.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.197363090515136,
|
|
"epoch": 0.08699254349627175,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5793,
|
|
"mean_token_accuracy": 0.05952193439006805,
|
|
"num_tokens": 190932.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.11266918182373,
|
|
"epoch": 0.09113504556752279,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4206,
|
|
"mean_token_accuracy": 0.0623247817158699,
|
|
"num_tokens": 199793.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.071755409240723,
|
|
"epoch": 0.09527754763877382,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.4016,
|
|
"mean_token_accuracy": 0.056386180222034454,
|
|
"num_tokens": 208865.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.93977689743042,
|
|
"epoch": 0.09942004971002485,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2482,
|
|
"mean_token_accuracy": 0.06587636284530163,
|
|
"num_tokens": 218987.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.830070209503173,
|
|
"epoch": 0.1035625517812759,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1893,
|
|
"mean_token_accuracy": 0.06974356137216091,
|
|
"num_tokens": 228815.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.623151206970215,
|
|
"epoch": 0.10770505385252693,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.0572,
|
|
"mean_token_accuracy": 0.06595751233398914,
|
|
"num_tokens": 237752.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.47017993927002,
|
|
"epoch": 0.11184755592377796,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9063,
|
|
"mean_token_accuracy": 0.07446046024560929,
|
|
"num_tokens": 246512.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.312911701202392,
|
|
"epoch": 0.115990057995029,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8915,
|
|
"mean_token_accuracy": 0.0656554326415062,
|
|
"num_tokens": 255300.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.113032817840576,
|
|
"epoch": 0.12013256006628004,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7471,
|
|
"mean_token_accuracy": 0.06687054596841335,
|
|
"num_tokens": 265125.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.8920316696167,
|
|
"epoch": 0.12427506213753108,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6598,
|
|
"mean_token_accuracy": 0.06876451000571251,
|
|
"num_tokens": 274516.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.746868896484376,
|
|
"epoch": 0.12841756420878211,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.6736,
|
|
"mean_token_accuracy": 0.07511950880289078,
|
|
"num_tokens": 282994.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.574301624298096,
|
|
"epoch": 0.13256006628003314,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.5639,
|
|
"mean_token_accuracy": 0.07003462612628937,
|
|
"num_tokens": 291703.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.47190055847168,
|
|
"epoch": 0.13670256835128416,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.4587,
|
|
"mean_token_accuracy": 0.0722161963582039,
|
|
"num_tokens": 301596.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.276020717620849,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5039,
|
|
"mean_token_accuracy": 0.075354777649045,
|
|
"num_tokens": 309678.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.170723915100098,
|
|
"epoch": 0.14498757249378624,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4176,
|
|
"mean_token_accuracy": 0.07923398241400718,
|
|
"num_tokens": 319046.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.180229568481446,
|
|
"epoch": 0.1491300745650373,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.07866172268986701,
|
|
"num_tokens": 328004.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.078632307052612,
|
|
"epoch": 0.15327257663628832,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3784,
|
|
"mean_token_accuracy": 0.0761629119515419,
|
|
"num_tokens": 337455.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.00011487007141,
|
|
"epoch": 0.15741507870753935,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3504,
|
|
"mean_token_accuracy": 0.07819610238075256,
|
|
"num_tokens": 345631.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.865444993972778,
|
|
"epoch": 0.1615575807787904,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2621,
|
|
"mean_token_accuracy": 0.07181993611156941,
|
|
"num_tokens": 354790.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.92261233329773,
|
|
"epoch": 0.16570008285004142,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.374,
|
|
"mean_token_accuracy": 0.08448374196887017,
|
|
"num_tokens": 364480.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.90945405960083,
|
|
"epoch": 0.16984258492129245,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4125,
|
|
"mean_token_accuracy": 0.07339793667197228,
|
|
"num_tokens": 372751.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.968420600891113,
|
|
"epoch": 0.1739850869925435,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3969,
|
|
"mean_token_accuracy": 0.07574166990816593,
|
|
"num_tokens": 381140.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.890913581848144,
|
|
"epoch": 0.17812758906379453,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3887,
|
|
"mean_token_accuracy": 0.07703622430562973,
|
|
"num_tokens": 391415.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.813537740707398,
|
|
"epoch": 0.18227009113504558,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3673,
|
|
"mean_token_accuracy": 0.07801026552915573,
|
|
"num_tokens": 400174.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.883461046218872,
|
|
"epoch": 0.1864125932062966,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2726,
|
|
"mean_token_accuracy": 0.08017019852995873,
|
|
"num_tokens": 409355.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.8811296939849855,
|
|
"epoch": 0.19055509527754763,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2959,
|
|
"mean_token_accuracy": 0.077110555768013,
|
|
"num_tokens": 418336.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.754353618621826,
|
|
"epoch": 0.19469759734879868,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.292,
|
|
"mean_token_accuracy": 0.0773789893835783,
|
|
"num_tokens": 427771.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.748405838012696,
|
|
"epoch": 0.1988400994200497,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3225,
|
|
"mean_token_accuracy": 0.07635865435004234,
|
|
"num_tokens": 435576.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.725781488418579,
|
|
"epoch": 0.20298260149130073,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2366,
|
|
"mean_token_accuracy": 0.07985471300780773,
|
|
"num_tokens": 444360.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.684485578536988,
|
|
"epoch": 0.2071251035625518,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1657,
|
|
"mean_token_accuracy": 0.08312897011637688,
|
|
"num_tokens": 452946.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.705522155761718,
|
|
"epoch": 0.2112676056338028,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1751,
|
|
"mean_token_accuracy": 0.08343161344528198,
|
|
"num_tokens": 462105.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.676226043701172,
|
|
"epoch": 0.21541010770505387,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3555,
|
|
"mean_token_accuracy": 0.07977722287178039,
|
|
"num_tokens": 472397.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.740552568435669,
|
|
"epoch": 0.2195526097763049,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2001,
|
|
"mean_token_accuracy": 0.078244648873806,
|
|
"num_tokens": 481595.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.51032190322876,
|
|
"epoch": 0.22369511184755592,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0929,
|
|
"mean_token_accuracy": 0.08391840234398842,
|
|
"num_tokens": 490034.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.6380149841308596,
|
|
"epoch": 0.22783761391880697,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2089,
|
|
"mean_token_accuracy": 0.08496099263429642,
|
|
"num_tokens": 498521.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.588693332672119,
|
|
"epoch": 0.231980115990058,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1605,
|
|
"mean_token_accuracy": 0.07962968274950981,
|
|
"num_tokens": 507251.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.5256298065185545,
|
|
"epoch": 0.23612261806130902,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0795,
|
|
"mean_token_accuracy": 0.08021439425647259,
|
|
"num_tokens": 516507.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.683542346954345,
|
|
"epoch": 0.24026512013256007,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1987,
|
|
"mean_token_accuracy": 0.07887389063835144,
|
|
"num_tokens": 526322.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.589657783508301,
|
|
"epoch": 0.2444076222038111,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0698,
|
|
"mean_token_accuracy": 0.0887388363480568,
|
|
"num_tokens": 534762.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.573185110092163,
|
|
"epoch": 0.24855012427506215,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.128,
|
|
"mean_token_accuracy": 0.08384166359901428,
|
|
"num_tokens": 544279.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.464304256439209,
|
|
"epoch": 0.2526926263463132,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1072,
|
|
"mean_token_accuracy": 0.08207703456282615,
|
|
"num_tokens": 553178.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.532624387741089,
|
|
"epoch": 0.25683512841756423,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0674,
|
|
"mean_token_accuracy": 0.08325572535395623,
|
|
"num_tokens": 562383.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.475009918212891,
|
|
"epoch": 0.2609776304888152,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0478,
|
|
"mean_token_accuracy": 0.08197091519832611,
|
|
"num_tokens": 571104.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.402563810348511,
|
|
"epoch": 0.2651201325600663,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9822,
|
|
"mean_token_accuracy": 0.08756032511591912,
|
|
"num_tokens": 579890.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.563023519515991,
|
|
"epoch": 0.26926263463131733,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1609,
|
|
"mean_token_accuracy": 0.08216102197766303,
|
|
"num_tokens": 589033.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.385844802856445,
|
|
"epoch": 0.27340513670256833,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.031,
|
|
"mean_token_accuracy": 0.08240275979042053,
|
|
"num_tokens": 598040.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.4914178371429445,
|
|
"epoch": 0.2775476387738194,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1792,
|
|
"mean_token_accuracy": 0.08133822605013848,
|
|
"num_tokens": 607618.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.360255098342895,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9847,
|
|
"mean_token_accuracy": 0.08660589605569839,
|
|
"num_tokens": 616107.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.469431829452515,
|
|
"epoch": 0.28583264291632143,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0424,
|
|
"mean_token_accuracy": 0.08697322458028793,
|
|
"num_tokens": 625954.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.43549690246582,
|
|
"epoch": 0.2899751449875725,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1473,
|
|
"mean_token_accuracy": 0.07502747476100921,
|
|
"num_tokens": 635840.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.531662464141846,
|
|
"epoch": 0.29411764705882354,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0981,
|
|
"mean_token_accuracy": 0.08479553908109665,
|
|
"num_tokens": 644597.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.403447055816651,
|
|
"epoch": 0.2982601491300746,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.019,
|
|
"mean_token_accuracy": 0.0871436946094036,
|
|
"num_tokens": 653454.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.331816291809082,
|
|
"epoch": 0.3024026512013256,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0644,
|
|
"mean_token_accuracy": 0.08571388535201549,
|
|
"num_tokens": 662524.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.350220823287964,
|
|
"epoch": 0.30654515327257664,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.037,
|
|
"mean_token_accuracy": 0.08441569060087203,
|
|
"num_tokens": 672741.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.462545442581177,
|
|
"epoch": 0.3106876553438277,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0014,
|
|
"mean_token_accuracy": 0.08310677483677864,
|
|
"num_tokens": 682068.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.483937978744507,
|
|
"epoch": 0.3148301574150787,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.068,
|
|
"mean_token_accuracy": 0.08450363203883171,
|
|
"num_tokens": 691506.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.211332607269287,
|
|
"epoch": 0.31897265948632975,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9459,
|
|
"mean_token_accuracy": 0.07925326749682426,
|
|
"num_tokens": 702355.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.426611804962159,
|
|
"epoch": 0.3231151615575808,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0277,
|
|
"mean_token_accuracy": 0.08726275414228439,
|
|
"num_tokens": 710878.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.349923038482666,
|
|
"epoch": 0.3272576636288318,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0066,
|
|
"mean_token_accuracy": 0.0816049836575985,
|
|
"num_tokens": 719389.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.291773509979248,
|
|
"epoch": 0.33140016570008285,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9038,
|
|
"mean_token_accuracy": 0.08625322580337524,
|
|
"num_tokens": 728109.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.3569948196411135,
|
|
"epoch": 0.3355426677713339,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0035,
|
|
"mean_token_accuracy": 0.08949297964572907,
|
|
"num_tokens": 737617.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.196184492111206,
|
|
"epoch": 0.3396851698425849,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8985,
|
|
"mean_token_accuracy": 0.086442032456398,
|
|
"num_tokens": 746368.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.21246075630188,
|
|
"epoch": 0.34382767191383595,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8542,
|
|
"mean_token_accuracy": 0.09095343127846718,
|
|
"num_tokens": 755561.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.358605766296387,
|
|
"epoch": 0.347970173985087,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9435,
|
|
"mean_token_accuracy": 0.08809717670083046,
|
|
"num_tokens": 764501.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.229580879211426,
|
|
"epoch": 0.352112676056338,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.085,
|
|
"mean_token_accuracy": 0.08461654633283615,
|
|
"num_tokens": 774003.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.327147531509399,
|
|
"epoch": 0.35625517812758906,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.09031773805618286,
|
|
"num_tokens": 782319.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.148540925979614,
|
|
"epoch": 0.3603976801988401,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8587,
|
|
"mean_token_accuracy": 0.08761301413178443,
|
|
"num_tokens": 791467.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.317266750335693,
|
|
"epoch": 0.36454018227009116,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9807,
|
|
"mean_token_accuracy": 0.08628286868333816,
|
|
"num_tokens": 800328.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.229923439025879,
|
|
"epoch": 0.36868268434134216,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9734,
|
|
"mean_token_accuracy": 0.08270182609558105,
|
|
"num_tokens": 808849.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.28882212638855,
|
|
"epoch": 0.3728251864125932,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.09033941775560379,
|
|
"num_tokens": 819090.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.23458833694458,
|
|
"epoch": 0.37696768848384427,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9226,
|
|
"mean_token_accuracy": 0.08801109045743942,
|
|
"num_tokens": 828527.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.252940464019775,
|
|
"epoch": 0.38111019055509526,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9447,
|
|
"mean_token_accuracy": 0.08754967637360096,
|
|
"num_tokens": 837163.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.2431192874908445,
|
|
"epoch": 0.3852526926263463,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9504,
|
|
"mean_token_accuracy": 0.0881062887609005,
|
|
"num_tokens": 846261.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.118913030624389,
|
|
"epoch": 0.38939519469759737,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8595,
|
|
"mean_token_accuracy": 0.08703678250312805,
|
|
"num_tokens": 855655.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.264123439788818,
|
|
"epoch": 0.39353769676884837,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.011,
|
|
"mean_token_accuracy": 0.08637121096253395,
|
|
"num_tokens": 864980.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.17798171043396,
|
|
"epoch": 0.3976801988400994,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8449,
|
|
"mean_token_accuracy": 0.08888368755578994,
|
|
"num_tokens": 874625.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.17276029586792,
|
|
"epoch": 0.40182270091135047,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8597,
|
|
"mean_token_accuracy": 0.08416522145271302,
|
|
"num_tokens": 884104.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.224177455902099,
|
|
"epoch": 0.40596520298260147,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9764,
|
|
"mean_token_accuracy": 0.08516203612089157,
|
|
"num_tokens": 893355.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.136541414260864,
|
|
"epoch": 0.4101077050538525,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.08847377672791482,
|
|
"num_tokens": 901666.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.1944825649261475,
|
|
"epoch": 0.4142502071251036,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8392,
|
|
"mean_token_accuracy": 0.08776508644223213,
|
|
"num_tokens": 911038.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4142502071251036,
|
|
"eval_entropy": 7.029010676117594,
|
|
"eval_loss": 6.951485633850098,
|
|
"eval_mean_token_accuracy": 0.08968717416625287,
|
|
"eval_num_tokens": 911038.0,
|
|
"eval_runtime": 2.9087,
|
|
"eval_samples_per_second": 1288.215,
|
|
"eval_steps_per_second": 161.242,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.090590286254883,
|
|
"epoch": 0.4183927091963546,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7688,
|
|
"mean_token_accuracy": 0.09368073344230651,
|
|
"num_tokens": 920505.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.014606046676636,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7777,
|
|
"mean_token_accuracy": 0.08937770649790763,
|
|
"num_tokens": 929983.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.062945079803467,
|
|
"epoch": 0.4266777133388567,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7857,
|
|
"mean_token_accuracy": 0.08497672230005264,
|
|
"num_tokens": 938920.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.22315354347229,
|
|
"epoch": 0.43082021541010773,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.10367532446980476,
|
|
"num_tokens": 948067.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.985883140563965,
|
|
"epoch": 0.43496271748135873,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8608,
|
|
"mean_token_accuracy": 0.09106278494000435,
|
|
"num_tokens": 957112.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.1854969501495365,
|
|
"epoch": 0.4391052195526098,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8318,
|
|
"mean_token_accuracy": 0.08919779881834984,
|
|
"num_tokens": 966112.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.923943567276001,
|
|
"epoch": 0.44324772162386084,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6632,
|
|
"mean_token_accuracy": 0.09753508195281028,
|
|
"num_tokens": 974791.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.010893726348877,
|
|
"epoch": 0.44739022369511183,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7551,
|
|
"mean_token_accuracy": 0.09772919043898583,
|
|
"num_tokens": 982730.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.060620880126953,
|
|
"epoch": 0.4515327257663629,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7707,
|
|
"mean_token_accuracy": 0.09096172973513603,
|
|
"num_tokens": 992275.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.988409900665284,
|
|
"epoch": 0.45567522783761394,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.09266227409243584,
|
|
"num_tokens": 1001131.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.050741291046142,
|
|
"epoch": 0.45981772990886494,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8893,
|
|
"mean_token_accuracy": 0.08530252054333687,
|
|
"num_tokens": 1010712.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.086013460159302,
|
|
"epoch": 0.463960231980116,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8307,
|
|
"mean_token_accuracy": 0.08732764646410943,
|
|
"num_tokens": 1019726.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.983129787445068,
|
|
"epoch": 0.46810273405136704,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.08867142722010612,
|
|
"num_tokens": 1028838.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.02021803855896,
|
|
"epoch": 0.47224523612261804,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.819,
|
|
"mean_token_accuracy": 0.09228519797325134,
|
|
"num_tokens": 1038839.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.049885368347168,
|
|
"epoch": 0.4763877381938691,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7751,
|
|
"mean_token_accuracy": 0.09855888932943344,
|
|
"num_tokens": 1048201.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.997226238250732,
|
|
"epoch": 0.48053024026512015,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.0835553340613842,
|
|
"num_tokens": 1057636.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.117639541625977,
|
|
"epoch": 0.48467274233637114,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8948,
|
|
"mean_token_accuracy": 0.09270423501729966,
|
|
"num_tokens": 1067776.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.967460298538208,
|
|
"epoch": 0.4888152444076222,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7117,
|
|
"mean_token_accuracy": 0.10028473436832427,
|
|
"num_tokens": 1076436.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.895012378692627,
|
|
"epoch": 0.49295774647887325,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.09727847725152969,
|
|
"num_tokens": 1085907.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.11523928642273,
|
|
"epoch": 0.4971002485501243,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8765,
|
|
"mean_token_accuracy": 0.09036477208137512,
|
|
"num_tokens": 1095617.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.85316572189331,
|
|
"epoch": 0.5012427506213754,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6596,
|
|
"mean_token_accuracy": 0.09437181055545807,
|
|
"num_tokens": 1104463.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.884859657287597,
|
|
"epoch": 0.5053852526926264,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8366,
|
|
"mean_token_accuracy": 0.09331914708018303,
|
|
"num_tokens": 1114594.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.040772867202759,
|
|
"epoch": 0.5095277547638773,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7153,
|
|
"mean_token_accuracy": 0.09341486915946007,
|
|
"num_tokens": 1124413.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.817426300048828,
|
|
"epoch": 0.5136702568351285,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7013,
|
|
"mean_token_accuracy": 0.09107679948210716,
|
|
"num_tokens": 1134392.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.028265714645386,
|
|
"epoch": 0.5178127589063795,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.08923142552375793,
|
|
"num_tokens": 1142563.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.868456792831421,
|
|
"epoch": 0.5219552609776305,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.859,
|
|
"mean_token_accuracy": 0.09302715957164764,
|
|
"num_tokens": 1151325.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.9985448837280275,
|
|
"epoch": 0.5260977630488816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7065,
|
|
"mean_token_accuracy": 0.09792971536517144,
|
|
"num_tokens": 1160817.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.997683715820313,
|
|
"epoch": 0.5302402651201326,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7195,
|
|
"mean_token_accuracy": 0.09700844064354897,
|
|
"num_tokens": 1169913.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.938048315048218,
|
|
"epoch": 0.5343827671913836,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.8268,
|
|
"mean_token_accuracy": 0.09028196930885315,
|
|
"num_tokens": 1179167.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.906701946258545,
|
|
"epoch": 0.5385252692626347,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7851,
|
|
"mean_token_accuracy": 0.09439637437462807,
|
|
"num_tokens": 1189105.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.777953195571899,
|
|
"epoch": 0.5426677713338857,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6017,
|
|
"mean_token_accuracy": 0.09869116172194481,
|
|
"num_tokens": 1197487.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.929132318496704,
|
|
"epoch": 0.5468102734051367,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7028,
|
|
"mean_token_accuracy": 0.09371067583560944,
|
|
"num_tokens": 1206914.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.92956395149231,
|
|
"epoch": 0.5509527754763878,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7621,
|
|
"mean_token_accuracy": 0.1005061112344265,
|
|
"num_tokens": 1214991.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.893949174880982,
|
|
"epoch": 0.5550952775476388,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7168,
|
|
"mean_token_accuracy": 0.09989271685481071,
|
|
"num_tokens": 1223567.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.962380599975586,
|
|
"epoch": 0.5592377796188898,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6499,
|
|
"mean_token_accuracy": 0.0953209526836872,
|
|
"num_tokens": 1232625.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.936891651153564,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7674,
|
|
"mean_token_accuracy": 0.09183648079633713,
|
|
"num_tokens": 1242947.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.9282430648803714,
|
|
"epoch": 0.5675227837613919,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.83,
|
|
"mean_token_accuracy": 0.09135605543851852,
|
|
"num_tokens": 1252231.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.85473780632019,
|
|
"epoch": 0.5716652858326429,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6957,
|
|
"mean_token_accuracy": 0.09633494764566422,
|
|
"num_tokens": 1262287.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.846092748641968,
|
|
"epoch": 0.575807787903894,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5806,
|
|
"mean_token_accuracy": 0.10334836915135384,
|
|
"num_tokens": 1271351.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.770713424682617,
|
|
"epoch": 0.579950289975145,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5851,
|
|
"mean_token_accuracy": 0.09925638064742089,
|
|
"num_tokens": 1280272.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.722933340072632,
|
|
"epoch": 0.584092792046396,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6445,
|
|
"mean_token_accuracy": 0.09788036495447158,
|
|
"num_tokens": 1290040.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.850803089141846,
|
|
"epoch": 0.5882352941176471,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6608,
|
|
"mean_token_accuracy": 0.10229592770338058,
|
|
"num_tokens": 1298814.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.843781042098999,
|
|
"epoch": 0.5923777961888981,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6779,
|
|
"mean_token_accuracy": 0.09635392874479294,
|
|
"num_tokens": 1308426.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.907309722900391,
|
|
"epoch": 0.5965202982601492,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.7811,
|
|
"mean_token_accuracy": 0.09799076840281487,
|
|
"num_tokens": 1316642.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.865832138061523,
|
|
"epoch": 0.6006628003314002,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6458,
|
|
"mean_token_accuracy": 0.10095788836479187,
|
|
"num_tokens": 1325023.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.719896602630615,
|
|
"epoch": 0.6048053024026512,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5506,
|
|
"mean_token_accuracy": 0.10248812884092331,
|
|
"num_tokens": 1333506.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.8264491081237795,
|
|
"epoch": 0.6089478044739023,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6554,
|
|
"mean_token_accuracy": 0.10372467041015625,
|
|
"num_tokens": 1342283.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.7825822830200195,
|
|
"epoch": 0.6130903065451533,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5995,
|
|
"mean_token_accuracy": 0.09992679581046104,
|
|
"num_tokens": 1350570.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.820115232467652,
|
|
"epoch": 0.6172328086164043,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6518,
|
|
"mean_token_accuracy": 0.10142135620117188,
|
|
"num_tokens": 1359390.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.711935329437256,
|
|
"epoch": 0.6213753106876554,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7036,
|
|
"mean_token_accuracy": 0.09649694561958314,
|
|
"num_tokens": 1368179.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.707304859161377,
|
|
"epoch": 0.6255178127589064,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.585,
|
|
"mean_token_accuracy": 0.10146094933152199,
|
|
"num_tokens": 1376739.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.74249963760376,
|
|
"epoch": 0.6296603148301574,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5747,
|
|
"mean_token_accuracy": 0.10194365158677102,
|
|
"num_tokens": 1385654.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.793521308898926,
|
|
"epoch": 0.6338028169014085,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6333,
|
|
"mean_token_accuracy": 0.09613201916217803,
|
|
"num_tokens": 1395811.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.661139059066772,
|
|
"epoch": 0.6379453189726595,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5734,
|
|
"mean_token_accuracy": 0.09805987849831581,
|
|
"num_tokens": 1404985.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.739059400558472,
|
|
"epoch": 0.6420878210439105,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5739,
|
|
"mean_token_accuracy": 0.10577682480216026,
|
|
"num_tokens": 1413306.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.735513067245483,
|
|
"epoch": 0.6462303231151616,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6181,
|
|
"mean_token_accuracy": 0.1049142599105835,
|
|
"num_tokens": 1422572.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.604087400436401,
|
|
"epoch": 0.6503728251864126,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5312,
|
|
"mean_token_accuracy": 0.1017876386642456,
|
|
"num_tokens": 1432743.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.643766403198242,
|
|
"epoch": 0.6545153272576636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5453,
|
|
"mean_token_accuracy": 0.10414253100752831,
|
|
"num_tokens": 1441459.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.806045198440552,
|
|
"epoch": 0.6586578293289147,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.7397,
|
|
"mean_token_accuracy": 0.09327913522720337,
|
|
"num_tokens": 1450900.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.7044349193573,
|
|
"epoch": 0.6628003314001657,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5546,
|
|
"mean_token_accuracy": 0.10626772269606591,
|
|
"num_tokens": 1459983.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.615601396560669,
|
|
"epoch": 0.6669428334714167,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.09967566579580307,
|
|
"num_tokens": 1469197.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.755218744277954,
|
|
"epoch": 0.6710853355426678,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5183,
|
|
"mean_token_accuracy": 0.11426351144909859,
|
|
"num_tokens": 1477422.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.645377492904663,
|
|
"epoch": 0.6752278376139188,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.6683,
|
|
"mean_token_accuracy": 0.09824564009904861,
|
|
"num_tokens": 1486813.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.705157470703125,
|
|
"epoch": 0.6793703396851698,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5939,
|
|
"mean_token_accuracy": 0.10478519424796104,
|
|
"num_tokens": 1495783.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.800464725494384,
|
|
"epoch": 0.6835128417564209,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6916,
|
|
"mean_token_accuracy": 0.10224536880850792,
|
|
"num_tokens": 1503926.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.6785046577453615,
|
|
"epoch": 0.6876553438276719,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5561,
|
|
"mean_token_accuracy": 0.09954166486859321,
|
|
"num_tokens": 1512714.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.569236993789673,
|
|
"epoch": 0.6917978458989229,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4272,
|
|
"mean_token_accuracy": 0.10806796178221703,
|
|
"num_tokens": 1521812.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.603865003585815,
|
|
"epoch": 0.695940347970174,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.6753,
|
|
"mean_token_accuracy": 0.09779459685087204,
|
|
"num_tokens": 1532083.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.691070890426635,
|
|
"epoch": 0.700082850041425,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.10249469205737113,
|
|
"num_tokens": 1541093.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.738386869430542,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5895,
|
|
"mean_token_accuracy": 0.1061245173215866,
|
|
"num_tokens": 1549942.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.638346433639526,
|
|
"epoch": 0.7083678541839271,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5236,
|
|
"mean_token_accuracy": 0.10389250293374061,
|
|
"num_tokens": 1559528.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.567893695831299,
|
|
"epoch": 0.7125103562551781,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4693,
|
|
"mean_token_accuracy": 0.10965005755424499,
|
|
"num_tokens": 1567590.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.621993827819824,
|
|
"epoch": 0.7166528583264291,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6154,
|
|
"mean_token_accuracy": 0.10054041370749474,
|
|
"num_tokens": 1576815.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.813084697723388,
|
|
"epoch": 0.7207953603976802,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6913,
|
|
"mean_token_accuracy": 0.09942748844623565,
|
|
"num_tokens": 1585240.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.702514028549194,
|
|
"epoch": 0.7249378624689312,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5847,
|
|
"mean_token_accuracy": 0.10101142525672913,
|
|
"num_tokens": 1595242.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.623669815063477,
|
|
"epoch": 0.7290803645401823,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5124,
|
|
"mean_token_accuracy": 0.10096012577414512,
|
|
"num_tokens": 1603753.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.524274682998657,
|
|
"epoch": 0.7332228666114333,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4964,
|
|
"mean_token_accuracy": 0.10853204876184464,
|
|
"num_tokens": 1613221.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.586112594604492,
|
|
"epoch": 0.7373653686826843,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.505,
|
|
"mean_token_accuracy": 0.10921112820506096,
|
|
"num_tokens": 1622881.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.622179698944092,
|
|
"epoch": 0.7415078707539354,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5134,
|
|
"mean_token_accuracy": 0.10829928070306778,
|
|
"num_tokens": 1631340.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.599711656570435,
|
|
"epoch": 0.7456503728251864,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5232,
|
|
"mean_token_accuracy": 0.10620987564325332,
|
|
"num_tokens": 1639920.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.625485038757324,
|
|
"epoch": 0.7497928748964374,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.551,
|
|
"mean_token_accuracy": 0.10739724114537239,
|
|
"num_tokens": 1649077.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.574350118637085,
|
|
"epoch": 0.7539353769676885,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.1003094732761383,
|
|
"num_tokens": 1659181.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.611196804046631,
|
|
"epoch": 0.7580778790389395,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4909,
|
|
"mean_token_accuracy": 0.11145598813891411,
|
|
"num_tokens": 1668111.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.5018958568573,
|
|
"epoch": 0.7622203811101905,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5254,
|
|
"mean_token_accuracy": 0.10480432286858558,
|
|
"num_tokens": 1677294.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.6240472316741945,
|
|
"epoch": 0.7663628831814416,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.10766990557312965,
|
|
"num_tokens": 1686187.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.5916759967803955,
|
|
"epoch": 0.7705053852526926,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5396,
|
|
"mean_token_accuracy": 0.1008406363427639,
|
|
"num_tokens": 1695999.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.52704086303711,
|
|
"epoch": 0.7746478873239436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4615,
|
|
"mean_token_accuracy": 0.1080300435423851,
|
|
"num_tokens": 1705292.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.622449254989624,
|
|
"epoch": 0.7787903893951947,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5108,
|
|
"mean_token_accuracy": 0.10377022996544838,
|
|
"num_tokens": 1714807.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.570343255996704,
|
|
"epoch": 0.7829328914664457,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.10502566173672676,
|
|
"num_tokens": 1723383.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.481582975387573,
|
|
"epoch": 0.7870753935376967,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4257,
|
|
"mean_token_accuracy": 0.10356146544218063,
|
|
"num_tokens": 1733151.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.4965410232543945,
|
|
"epoch": 0.7912178956089478,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4716,
|
|
"mean_token_accuracy": 0.11054345145821572,
|
|
"num_tokens": 1742292.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.6443828582763675,
|
|
"epoch": 0.7953603976801988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5643,
|
|
"mean_token_accuracy": 0.11139457151293755,
|
|
"num_tokens": 1751441.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.527015209197998,
|
|
"epoch": 0.7995028997514498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4417,
|
|
"mean_token_accuracy": 0.11440029144287109,
|
|
"num_tokens": 1759833.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.575316333770752,
|
|
"epoch": 0.8036454018227009,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.11134352311491966,
|
|
"num_tokens": 1769258.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.614769506454468,
|
|
"epoch": 0.8077879038939519,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.6136,
|
|
"mean_token_accuracy": 0.09684511870145798,
|
|
"num_tokens": 1779532.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.489955949783325,
|
|
"epoch": 0.8119304059652029,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4826,
|
|
"mean_token_accuracy": 0.10542305111885071,
|
|
"num_tokens": 1788936.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.524744033813477,
|
|
"epoch": 0.816072908036454,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4307,
|
|
"mean_token_accuracy": 0.11505212485790253,
|
|
"num_tokens": 1796486.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.5153892993927,
|
|
"epoch": 0.820215410107705,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4003,
|
|
"mean_token_accuracy": 0.1139255128800869,
|
|
"num_tokens": 1805349.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.511750984191894,
|
|
"epoch": 0.824357912178956,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.4933,
|
|
"mean_token_accuracy": 0.10885351151227951,
|
|
"num_tokens": 1813625.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.511787986755371,
|
|
"epoch": 0.8285004142502072,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.5196,
|
|
"mean_token_accuracy": 0.11211849302053452,
|
|
"num_tokens": 1823465.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8285004142502072,
|
|
"eval_entropy": 6.489772428581709,
|
|
"eval_loss": 6.5978169441223145,
|
|
"eval_mean_token_accuracy": 0.1059395734673497,
|
|
"eval_num_tokens": 1823465.0,
|
|
"eval_runtime": 2.8165,
|
|
"eval_samples_per_second": 1330.384,
|
|
"eval_steps_per_second": 166.52,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.503752756118774,
|
|
"epoch": 0.8326429163214581,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999998547681232,
|
|
"loss": 6.4594,
|
|
"mean_token_accuracy": 0.10898767039179802,
|
|
"num_tokens": 1832701.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.5025763511657715,
|
|
"epoch": 0.8367854183927091,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999992647639447,
|
|
"loss": 6.5037,
|
|
"mean_token_accuracy": 0.1069453164935112,
|
|
"num_tokens": 1841914.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.498456144332886,
|
|
"epoch": 0.8409279204639603,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999982209116618,
|
|
"loss": 6.5033,
|
|
"mean_token_accuracy": 0.1143178328871727,
|
|
"num_tokens": 1851906.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.527345752716064,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999967232133796,
|
|
"loss": 6.4405,
|
|
"mean_token_accuracy": 0.11167196556925774,
|
|
"num_tokens": 1860817.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.471268653869629,
|
|
"epoch": 0.8492129246064622,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999947716721196,
|
|
"loss": 6.4317,
|
|
"mean_token_accuracy": 0.11340907365083694,
|
|
"num_tokens": 1870387.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.53512659072876,
|
|
"epoch": 0.8533554266777134,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999923662918179,
|
|
"loss": 6.4923,
|
|
"mean_token_accuracy": 0.10849138200283051,
|
|
"num_tokens": 1878929.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.381455135345459,
|
|
"epoch": 0.8574979287489644,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999895070773266,
|
|
"loss": 6.5218,
|
|
"mean_token_accuracy": 0.10917675569653511,
|
|
"num_tokens": 1887633.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.555011653900147,
|
|
"epoch": 0.8616404308202155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999861940344129,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.11225893646478653,
|
|
"num_tokens": 1896729.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.461038255691529,
|
|
"epoch": 0.8657829328914665,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999824271697597,
|
|
"loss": 6.4744,
|
|
"mean_token_accuracy": 0.11994748562574387,
|
|
"num_tokens": 1905224.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.426373243331909,
|
|
"epoch": 0.8699254349627175,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499978206490965,
|
|
"loss": 6.4421,
|
|
"mean_token_accuracy": 0.1062840238213539,
|
|
"num_tokens": 1915502.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.507133817672729,
|
|
"epoch": 0.8740679370339686,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999735320065427,
|
|
"loss": 6.5166,
|
|
"mean_token_accuracy": 0.1090014934539795,
|
|
"num_tokens": 1923946.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.391321516036987,
|
|
"epoch": 0.8782104391052196,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999684037259216,
|
|
"loss": 6.3153,
|
|
"mean_token_accuracy": 0.11198925003409385,
|
|
"num_tokens": 1932524.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.523810911178589,
|
|
"epoch": 0.8823529411764706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999628216594456,
|
|
"loss": 6.4342,
|
|
"mean_token_accuracy": 0.11353398859500885,
|
|
"num_tokens": 1941604.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.369825983047486,
|
|
"epoch": 0.8864954432477217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499956785818375,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.11778979599475861,
|
|
"num_tokens": 1950235.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.44530668258667,
|
|
"epoch": 0.8906379453189727,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999502962148843,
|
|
"loss": 6.411,
|
|
"mean_token_accuracy": 0.11221367716789246,
|
|
"num_tokens": 1959597.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.611082124710083,
|
|
"epoch": 0.8947804473902237,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499943352862064,
|
|
"loss": 6.5525,
|
|
"mean_token_accuracy": 0.11493104994297028,
|
|
"num_tokens": 1969032.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.356397533416748,
|
|
"epoch": 0.8989229494614748,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999359557739194,
|
|
"loss": 6.4099,
|
|
"mean_token_accuracy": 0.11556253135204315,
|
|
"num_tokens": 1977691.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.429668664932251,
|
|
"epoch": 0.9030654515327258,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999281049653713,
|
|
"loss": 6.3185,
|
|
"mean_token_accuracy": 0.11952688023447991,
|
|
"num_tokens": 1986118.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.4375769138336185,
|
|
"epoch": 0.9072079536039768,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999198004522557,
|
|
"loss": 6.4391,
|
|
"mean_token_accuracy": 0.11205489486455918,
|
|
"num_tokens": 1995466.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.4425500392913815,
|
|
"epoch": 0.9113504556752279,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999110422513237,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.10964742228388787,
|
|
"num_tokens": 2005219.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.36616268157959,
|
|
"epoch": 0.9154929577464789,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999018303802414,
|
|
"loss": 6.3219,
|
|
"mean_token_accuracy": 0.11532410681247711,
|
|
"num_tokens": 2014078.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.571248149871826,
|
|
"epoch": 0.9196354598177299,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998921648575904,
|
|
"loss": 6.5299,
|
|
"mean_token_accuracy": 0.10419057235121727,
|
|
"num_tokens": 2023456.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.337164688110351,
|
|
"epoch": 0.923777961888981,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998820457028669,
|
|
"loss": 6.3187,
|
|
"mean_token_accuracy": 0.11291450634598732,
|
|
"num_tokens": 2032312.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.432801675796509,
|
|
"epoch": 0.927920463960232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998714729364824,
|
|
"loss": 6.2912,
|
|
"mean_token_accuracy": 0.11874229311943055,
|
|
"num_tokens": 2041580.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.351183366775513,
|
|
"epoch": 0.932062966031483,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998604465797634,
|
|
"loss": 6.5144,
|
|
"mean_token_accuracy": 0.10716686621308327,
|
|
"num_tokens": 2051126.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.538003158569336,
|
|
"epoch": 0.9362054681027341,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998489666549512,
|
|
"loss": 6.5034,
|
|
"mean_token_accuracy": 0.11126186326146126,
|
|
"num_tokens": 2059565.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.341565847396851,
|
|
"epoch": 0.9403479701739851,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998370331852021,
|
|
"loss": 6.2731,
|
|
"mean_token_accuracy": 0.11285399943590164,
|
|
"num_tokens": 2068113.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.386385202407837,
|
|
"epoch": 0.9444904722452361,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998246461945872,
|
|
"loss": 6.4455,
|
|
"mean_token_accuracy": 0.11150851249694824,
|
|
"num_tokens": 2077227.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.483101797103882,
|
|
"epoch": 0.9486329743164872,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998118057080923,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.11290868446230888,
|
|
"num_tokens": 2086234.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.346535301208496,
|
|
"epoch": 0.9527754763877382,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997985117516183,
|
|
"loss": 6.3818,
|
|
"mean_token_accuracy": 0.11334575861692428,
|
|
"num_tokens": 2094945.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.368511581420899,
|
|
"epoch": 0.9569179784589892,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997847643519803,
|
|
"loss": 6.3666,
|
|
"mean_token_accuracy": 0.12083429247140884,
|
|
"num_tokens": 2104164.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.506369352340698,
|
|
"epoch": 0.9610604805302403,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997705635369085,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.10683630257844925,
|
|
"num_tokens": 2112629.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.375317573547363,
|
|
"epoch": 0.9652029826014913,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997559093350474,
|
|
"loss": 6.4314,
|
|
"mean_token_accuracy": 0.11005270928144455,
|
|
"num_tokens": 2122491.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.416046524047852,
|
|
"epoch": 0.9693454846727423,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997408017759562,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.11960792094469071,
|
|
"num_tokens": 2132115.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.315891075134277,
|
|
"epoch": 0.9734879867439934,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997252408901084,
|
|
"loss": 6.2895,
|
|
"mean_token_accuracy": 0.1188917763531208,
|
|
"num_tokens": 2141663.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.4620068073272705,
|
|
"epoch": 0.9776304888152444,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997092267088922,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.11433440446853638,
|
|
"num_tokens": 2152057.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.459884881973267,
|
|
"epoch": 0.9817729908864954,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996927592646097,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.11693302392959595,
|
|
"num_tokens": 2161554.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.297170400619507,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996758385904776,
|
|
"loss": 6.291,
|
|
"mean_token_accuracy": 0.11896034255623818,
|
|
"num_tokens": 2170553.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.433252429962158,
|
|
"epoch": 0.9900579950289975,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996584647206269,
|
|
"loss": 6.4227,
|
|
"mean_token_accuracy": 0.1149461954832077,
|
|
"num_tokens": 2179996.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.324653291702271,
|
|
"epoch": 0.9942004971002486,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996406376901026,
|
|
"loss": 6.2669,
|
|
"mean_token_accuracy": 0.12070421054959297,
|
|
"num_tokens": 2188882.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.464151000976562,
|
|
"epoch": 0.9983429991714996,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996223575348635,
|
|
"loss": 6.4539,
|
|
"mean_token_accuracy": 0.11858661994338035,
|
|
"num_tokens": 2197304.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.405771923065186,
|
|
"epoch": 1.0024855012427507,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996036242917828,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.12037697657942772,
|
|
"num_tokens": 2206144.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.425885581970215,
|
|
"epoch": 1.0066280033140016,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995844379986476,
|
|
"loss": 6.2114,
|
|
"mean_token_accuracy": 0.12305252626538277,
|
|
"num_tokens": 2215054.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.230459594726563,
|
|
"epoch": 1.0107705053852527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995647986941587,
|
|
"loss": 6.0583,
|
|
"mean_token_accuracy": 0.1284884586930275,
|
|
"num_tokens": 2223578.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.175695371627808,
|
|
"epoch": 1.0149130074565038,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995447064179308,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.1279567815363407,
|
|
"num_tokens": 2234115.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.283312606811523,
|
|
"epoch": 1.0190555095277547,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499524161210492,
|
|
"loss": 6.1152,
|
|
"mean_token_accuracy": 0.12857213839888573,
|
|
"num_tokens": 2243447.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.329511642456055,
|
|
"epoch": 1.0231980115990058,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995031631132844,
|
|
"loss": 6.103,
|
|
"mean_token_accuracy": 0.12532400861382484,
|
|
"num_tokens": 2252730.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.204122638702392,
|
|
"epoch": 1.027340513670257,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004994817121686633,
|
|
"loss": 6.0922,
|
|
"mean_token_accuracy": 0.1204651266336441,
|
|
"num_tokens": 2261809.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.253789472579956,
|
|
"epoch": 1.0314830157415078,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994598084198979,
|
|
"loss": 6.0332,
|
|
"mean_token_accuracy": 0.12827168405056,
|
|
"num_tokens": 2270583.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.270323991775513,
|
|
"epoch": 1.035625517812759,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004994374519111701,
|
|
"loss": 6.1004,
|
|
"mean_token_accuracy": 0.13118301704525948,
|
|
"num_tokens": 2279337.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.257144260406494,
|
|
"epoch": 1.03976801988401,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004994146426875757,
|
|
"loss": 6.0698,
|
|
"mean_token_accuracy": 0.12660571709275245,
|
|
"num_tokens": 2288679.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.186429357528686,
|
|
"epoch": 1.043910521955261,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993913807951232,
|
|
"loss": 6.0641,
|
|
"mean_token_accuracy": 0.12812110856175424,
|
|
"num_tokens": 2297780.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.25863995552063,
|
|
"epoch": 1.048053024026512,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993676662807345,
|
|
"loss": 6.1318,
|
|
"mean_token_accuracy": 0.12341021671891213,
|
|
"num_tokens": 2306424.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.240696907043457,
|
|
"epoch": 1.0521955260977631,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004993434991922443,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.12805933132767677,
|
|
"num_tokens": 2315472.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.19228229522705,
|
|
"epoch": 1.056338028169014,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993188795784002,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.1268403261899948,
|
|
"num_tokens": 2324381.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.129907703399658,
|
|
"epoch": 1.0604805302402651,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004992938074888628,
|
|
"loss": 6.0639,
|
|
"mean_token_accuracy": 0.12428224980831146,
|
|
"num_tokens": 2333629.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.27379355430603,
|
|
"epoch": 1.0646230323115162,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992682829742052,
|
|
"loss": 6.1316,
|
|
"mean_token_accuracy": 0.126251719892025,
|
|
"num_tokens": 2342787.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.154513502120972,
|
|
"epoch": 1.0687655343827671,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992423060859132,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.13422906622290612,
|
|
"num_tokens": 2351684.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.270391750335693,
|
|
"epoch": 1.0729080364540182,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004992158768763849,
|
|
"loss": 6.1692,
|
|
"mean_token_accuracy": 0.12574677392840386,
|
|
"num_tokens": 2361100.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.145001316070557,
|
|
"epoch": 1.0770505385252693,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991889953989308,
|
|
"loss": 6.1219,
|
|
"mean_token_accuracy": 0.12856860533356668,
|
|
"num_tokens": 2370139.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.286712026596069,
|
|
"epoch": 1.0811930405965202,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991616617077741,
|
|
"loss": 6.1773,
|
|
"mean_token_accuracy": 0.12485618069767952,
|
|
"num_tokens": 2378485.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.132142114639282,
|
|
"epoch": 1.0853355426677713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991338758580496,
|
|
"loss": 6.1043,
|
|
"mean_token_accuracy": 0.12516417354345322,
|
|
"num_tokens": 2387821.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.168926572799682,
|
|
"epoch": 1.0894780447390224,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004991056379058045,
|
|
"loss": 6.0131,
|
|
"mean_token_accuracy": 0.13040016889572142,
|
|
"num_tokens": 2396584.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.205351161956787,
|
|
"epoch": 1.0936205468102733,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990769479079976,
|
|
"loss": 6.1349,
|
|
"mean_token_accuracy": 0.12381747737526894,
|
|
"num_tokens": 2405533.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.267547273635865,
|
|
"epoch": 1.0977630488815244,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004990478059225003,
|
|
"loss": 6.1177,
|
|
"mean_token_accuracy": 0.1273566223680973,
|
|
"num_tokens": 2414147.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.170993137359619,
|
|
"epoch": 1.1019055509527755,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990182120080945,
|
|
"loss": 6.1233,
|
|
"mean_token_accuracy": 0.12716805413365365,
|
|
"num_tokens": 2422888.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.229750061035157,
|
|
"epoch": 1.1060480530240264,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989881662244748,
|
|
"loss": 6.0803,
|
|
"mean_token_accuracy": 0.13159133717417718,
|
|
"num_tokens": 2431165.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.236302328109741,
|
|
"epoch": 1.1101905550952775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989576686322468,
|
|
"loss": 6.1354,
|
|
"mean_token_accuracy": 0.12131609246134759,
|
|
"num_tokens": 2440726.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.230614805221558,
|
|
"epoch": 1.1143330571665286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989267192929272,
|
|
"loss": 6.1084,
|
|
"mean_token_accuracy": 0.12925527468323708,
|
|
"num_tokens": 2449393.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.115985679626465,
|
|
"epoch": 1.1184755592377795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004988953182689445,
|
|
"loss": 6.091,
|
|
"mean_token_accuracy": 0.1255219593644142,
|
|
"num_tokens": 2458210.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.1895167350769045,
|
|
"epoch": 1.1226180613090306,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004988634656236378,
|
|
"loss": 6.1111,
|
|
"mean_token_accuracy": 0.13017474710941315,
|
|
"num_tokens": 2466765.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.2510546207427975,
|
|
"epoch": 1.1267605633802817,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004988311614212576,
|
|
"loss": 6.1531,
|
|
"mean_token_accuracy": 0.1353672005236149,
|
|
"num_tokens": 2476517.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.119971084594726,
|
|
"epoch": 1.1309030654515326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004987984057269648,
|
|
"loss": 5.9815,
|
|
"mean_token_accuracy": 0.13376273065805436,
|
|
"num_tokens": 2485073.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.251840734481812,
|
|
"epoch": 1.1350455675227837,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004987651986068311,
|
|
"loss": 6.2087,
|
|
"mean_token_accuracy": 0.1218659982085228,
|
|
"num_tokens": 2493764.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.209574842453003,
|
|
"epoch": 1.1391880695940348,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004987315401278394,
|
|
"loss": 6.0867,
|
|
"mean_token_accuracy": 0.1295430727303028,
|
|
"num_tokens": 2502507.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.127098035812378,
|
|
"epoch": 1.143330571665286,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986974303578819,
|
|
"loss": 6.0602,
|
|
"mean_token_accuracy": 0.13360463380813598,
|
|
"num_tokens": 2510925.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.273448038101196,
|
|
"epoch": 1.1474730737365368,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986628693657621,
|
|
"loss": 6.1665,
|
|
"mean_token_accuracy": 0.1274927906692028,
|
|
"num_tokens": 2519927.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.180120134353638,
|
|
"epoch": 1.151615575807788,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004986278572211933,
|
|
"loss": 6.1406,
|
|
"mean_token_accuracy": 0.12608257159590722,
|
|
"num_tokens": 2529210.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.191981315612793,
|
|
"epoch": 1.1557580778790388,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004985923939947987,
|
|
"loss": 6.0294,
|
|
"mean_token_accuracy": 0.1362480901181698,
|
|
"num_tokens": 2538254.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.151539897918701,
|
|
"epoch": 1.15990057995029,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004985564797581115,
|
|
"loss": 6.0098,
|
|
"mean_token_accuracy": 0.13473647087812424,
|
|
"num_tokens": 2546563.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.353456687927246,
|
|
"epoch": 1.164043082021541,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004985201145835747,
|
|
"loss": 6.3072,
|
|
"mean_token_accuracy": 0.1152469776570797,
|
|
"num_tokens": 2556564.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.18480372428894,
|
|
"epoch": 1.1681855840927922,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498483298544541,
|
|
"loss": 6.1105,
|
|
"mean_token_accuracy": 0.12562097012996673,
|
|
"num_tokens": 2565904.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.174497842788696,
|
|
"epoch": 1.172328086164043,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000498446031715272,
|
|
"loss": 6.0035,
|
|
"mean_token_accuracy": 0.13123355731368064,
|
|
"num_tokens": 2574840.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.230139112472534,
|
|
"epoch": 1.1764705882352942,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984083141709391,
|
|
"loss": 6.2057,
|
|
"mean_token_accuracy": 0.1256824679672718,
|
|
"num_tokens": 2584419.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.165558242797852,
|
|
"epoch": 1.180613090306545,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000498370145987623,
|
|
"loss": 6.1846,
|
|
"mean_token_accuracy": 0.13115376979112625,
|
|
"num_tokens": 2593941.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.197154188156128,
|
|
"epoch": 1.1847555923777962,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004983315272423127,
|
|
"loss": 6.078,
|
|
"mean_token_accuracy": 0.13584018126130104,
|
|
"num_tokens": 2602473.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.174890565872192,
|
|
"epoch": 1.1888980944490473,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004982924580129066,
|
|
"loss": 6.0409,
|
|
"mean_token_accuracy": 0.1310078538954258,
|
|
"num_tokens": 2612287.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.03299913406372,
|
|
"epoch": 1.1930405965202984,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004982529383782116,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.1340833991765976,
|
|
"num_tokens": 2620557.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.140125703811646,
|
|
"epoch": 1.1971830985915493,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982129684179432,
|
|
"loss": 6.1482,
|
|
"mean_token_accuracy": 0.12488306164741517,
|
|
"num_tokens": 2630234.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.186292362213135,
|
|
"epoch": 1.2013256006628004,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004981725482127251,
|
|
"loss": 6.1455,
|
|
"mean_token_accuracy": 0.12422921806573868,
|
|
"num_tokens": 2639500.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.1869282722473145,
|
|
"epoch": 1.2054681027340513,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981316778440892,
|
|
"loss": 6.0985,
|
|
"mean_token_accuracy": 0.12317988276481628,
|
|
"num_tokens": 2648928.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.320038223266602,
|
|
"epoch": 1.2096106048053024,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004980903573944756,
|
|
"loss": 6.2304,
|
|
"mean_token_accuracy": 0.11908927112817765,
|
|
"num_tokens": 2658846.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.055701398849488,
|
|
"epoch": 1.2137531068765535,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004980485869472322,
|
|
"loss": 6.1225,
|
|
"mean_token_accuracy": 0.12987650632858277,
|
|
"num_tokens": 2668126.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.297923517227173,
|
|
"epoch": 1.2178956089478046,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004980063665866143,
|
|
"loss": 6.1422,
|
|
"mean_token_accuracy": 0.13109913021326064,
|
|
"num_tokens": 2676247.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.176990509033203,
|
|
"epoch": 1.2220381110190555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004979636963977853,
|
|
"loss": 6.0856,
|
|
"mean_token_accuracy": 0.13268199265003205,
|
|
"num_tokens": 2684964.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.132426357269287,
|
|
"epoch": 1.2261806130903066,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004979205764668156,
|
|
"loss": 6.1072,
|
|
"mean_token_accuracy": 0.12597387731075288,
|
|
"num_tokens": 2694143.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.088831424713135,
|
|
"epoch": 1.2303231151615575,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004978770068806825,
|
|
"loss": 6.0059,
|
|
"mean_token_accuracy": 0.1387046158313751,
|
|
"num_tokens": 2702243.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.178600311279297,
|
|
"epoch": 1.2344656172328086,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497832987727271,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.12923040315508844,
|
|
"num_tokens": 2710657.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.070341777801514,
|
|
"epoch": 1.2386081193040597,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004977885190953724,
|
|
"loss": 5.9968,
|
|
"mean_token_accuracy": 0.12945516631007195,
|
|
"num_tokens": 2719044.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.265737390518188,
|
|
"epoch": 1.2427506213753108,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004977436010746846,
|
|
"loss": 6.1141,
|
|
"mean_token_accuracy": 0.1338028959929943,
|
|
"num_tokens": 2727990.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2427506213753108,
|
|
"eval_entropy": 6.0073623586056835,
|
|
"eval_loss": 6.336162567138672,
|
|
"eval_mean_token_accuracy": 0.1255791528837513,
|
|
"eval_num_tokens": 2727990.0,
|
|
"eval_runtime": 3.1604,
|
|
"eval_samples_per_second": 1185.623,
|
|
"eval_steps_per_second": 148.401,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.1073040008544925,
|
|
"epoch": 1.2468931234465617,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004976982337558124,
|
|
"loss": 6.1223,
|
|
"mean_token_accuracy": 0.1255759708583355,
|
|
"num_tokens": 2736578.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.173589277267456,
|
|
"epoch": 1.2510356255178128,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976524172302665,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.13060156553983687,
|
|
"num_tokens": 2745693.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.202486658096314,
|
|
"epoch": 1.2551781275890637,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976061515904641,
|
|
"loss": 6.079,
|
|
"mean_token_accuracy": 0.13340094313025475,
|
|
"num_tokens": 2754952.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.176916265487671,
|
|
"epoch": 1.2593206296603148,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004975594369297277,
|
|
"loss": 6.0936,
|
|
"mean_token_accuracy": 0.12923406064510345,
|
|
"num_tokens": 2763981.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.051639413833618,
|
|
"epoch": 1.2634631317315659,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004975122733422861,
|
|
"loss": 6.1086,
|
|
"mean_token_accuracy": 0.12856508418917656,
|
|
"num_tokens": 2772740.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.0949808120727536,
|
|
"epoch": 1.267605633802817,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004974646609232732,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.13504123240709304,
|
|
"num_tokens": 2781489.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.120691251754761,
|
|
"epoch": 1.2717481358740679,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974165997687288,
|
|
"loss": 6.0706,
|
|
"mean_token_accuracy": 0.13792871534824372,
|
|
"num_tokens": 2790112.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.119127416610718,
|
|
"epoch": 1.275890637945319,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004973680899755972,
|
|
"loss": 6.1277,
|
|
"mean_token_accuracy": 0.12491117119789123,
|
|
"num_tokens": 2799940.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.173457908630371,
|
|
"epoch": 1.2800331400165699,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497319131641728,
|
|
"loss": 6.0299,
|
|
"mean_token_accuracy": 0.13059929981827736,
|
|
"num_tokens": 2808369.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.0894286155700685,
|
|
"epoch": 1.284175642087821,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004972697248658755,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.1335919052362442,
|
|
"num_tokens": 2818205.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.195007848739624,
|
|
"epoch": 1.288318144159072,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004972198697476987,
|
|
"loss": 6.1228,
|
|
"mean_token_accuracy": 0.13143828511238098,
|
|
"num_tokens": 2827350.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.121762228012085,
|
|
"epoch": 1.2924606462303232,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004971695663877607,
|
|
"loss": 6.049,
|
|
"mean_token_accuracy": 0.13058406859636307,
|
|
"num_tokens": 2836193.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.9933483600616455,
|
|
"epoch": 1.296603148301574,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004971188148875289,
|
|
"loss": 6.0246,
|
|
"mean_token_accuracy": 0.1318896971642971,
|
|
"num_tokens": 2846016.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.338242387771606,
|
|
"epoch": 1.3007456503728252,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004970676153493748,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.12545630112290382,
|
|
"num_tokens": 2856250.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.093726062774659,
|
|
"epoch": 1.3048881524440763,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004970159678765731,
|
|
"loss": 6.1197,
|
|
"mean_token_accuracy": 0.12585905566811562,
|
|
"num_tokens": 2865423.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.190137577056885,
|
|
"epoch": 1.3090306545153272,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004969638725733028,
|
|
"loss": 6.0842,
|
|
"mean_token_accuracy": 0.12343057692050934,
|
|
"num_tokens": 2874432.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.069796419143676,
|
|
"epoch": 1.3131731565865783,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969113295446455,
|
|
"loss": 6.063,
|
|
"mean_token_accuracy": 0.12247302234172822,
|
|
"num_tokens": 2884645.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.131494760513306,
|
|
"epoch": 1.3173156586578294,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004968583388965863,
|
|
"loss": 6.1312,
|
|
"mean_token_accuracy": 0.12738580405712127,
|
|
"num_tokens": 2894580.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.1513735294342045,
|
|
"epoch": 1.3214581607290803,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968049007360132,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.13449667394161224,
|
|
"num_tokens": 2903234.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.101231431961059,
|
|
"epoch": 1.3256006628003314,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004967510151707169,
|
|
"loss": 6.1014,
|
|
"mean_token_accuracy": 0.12553910091519355,
|
|
"num_tokens": 2912717.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.053020715713501,
|
|
"epoch": 1.3297431648715825,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004966966823093903,
|
|
"loss": 5.987,
|
|
"mean_token_accuracy": 0.13677354380488396,
|
|
"num_tokens": 2921449.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.0964494228363035,
|
|
"epoch": 1.3338856669428334,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004966419022616287,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.12726073935627938,
|
|
"num_tokens": 2930730.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.1492963314056395,
|
|
"epoch": 1.3380281690140845,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004965866751379296,
|
|
"loss": 6.104,
|
|
"mean_token_accuracy": 0.13120296895503997,
|
|
"num_tokens": 2940126.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.140308618545532,
|
|
"epoch": 1.3421706710853356,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004965310010496922,
|
|
"loss": 6.0093,
|
|
"mean_token_accuracy": 0.13938634768128394,
|
|
"num_tokens": 2949023.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.072773408889771,
|
|
"epoch": 1.3463131731565865,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004964748801092168,
|
|
"loss": 6.0634,
|
|
"mean_token_accuracy": 0.13058572858572007,
|
|
"num_tokens": 2957548.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.093048334121704,
|
|
"epoch": 1.3504556752278376,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004964183124297059,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.1446573927998543,
|
|
"num_tokens": 2965914.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.099730920791626,
|
|
"epoch": 1.3545981772990887,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004963612981252625,
|
|
"loss": 6.0614,
|
|
"mean_token_accuracy": 0.1373268634080887,
|
|
"num_tokens": 2974756.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.975049018859863,
|
|
"epoch": 1.3587406793703396,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004963038373108908,
|
|
"loss": 6.0593,
|
|
"mean_token_accuracy": 0.1383127264678478,
|
|
"num_tokens": 2984267.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.220937728881836,
|
|
"epoch": 1.3628831814415907,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004962459301024954,
|
|
"loss": 6.0875,
|
|
"mean_token_accuracy": 0.12818419858813285,
|
|
"num_tokens": 2992981.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.021561861038208,
|
|
"epoch": 1.3670256835128418,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004961875766168816,
|
|
"loss": 6.1065,
|
|
"mean_token_accuracy": 0.13113522082567214,
|
|
"num_tokens": 3002135.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.122888469696045,
|
|
"epoch": 1.371168185584093,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004961287769717547,
|
|
"loss": 6.0753,
|
|
"mean_token_accuracy": 0.12611212581396103,
|
|
"num_tokens": 3011579.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.200709390640259,
|
|
"epoch": 1.3753106876553438,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004960695312857199,
|
|
"loss": 6.1861,
|
|
"mean_token_accuracy": 0.12607147470116614,
|
|
"num_tokens": 3021975.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.122780990600586,
|
|
"epoch": 1.379453189726595,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004960098396782824,
|
|
"loss": 6.0975,
|
|
"mean_token_accuracy": 0.13255783617496492,
|
|
"num_tokens": 3030878.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.164667129516602,
|
|
"epoch": 1.3835956917978458,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004959497022698466,
|
|
"loss": 6.0787,
|
|
"mean_token_accuracy": 0.13239231780171395,
|
|
"num_tokens": 3040021.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.138751554489136,
|
|
"epoch": 1.387738193869097,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004958891191817163,
|
|
"loss": 6.2258,
|
|
"mean_token_accuracy": 0.12937600538134575,
|
|
"num_tokens": 3049944.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.212243986129761,
|
|
"epoch": 1.391880695940348,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000495828090536094,
|
|
"loss": 6.0185,
|
|
"mean_token_accuracy": 0.1379230573773384,
|
|
"num_tokens": 3059327.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.974050140380859,
|
|
"epoch": 1.3960231980115991,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004957666164560815,
|
|
"loss": 6.0064,
|
|
"mean_token_accuracy": 0.13503726869821547,
|
|
"num_tokens": 3068325.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.288743686676026,
|
|
"epoch": 1.40016570008285,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004957046970656784,
|
|
"loss": 6.0853,
|
|
"mean_token_accuracy": 0.13486202135682107,
|
|
"num_tokens": 3076715.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.045103120803833,
|
|
"epoch": 1.4043082021541011,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004956423324897831,
|
|
"loss": 6.15,
|
|
"mean_token_accuracy": 0.13253118470311165,
|
|
"num_tokens": 3086051.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.1374091625213625,
|
|
"epoch": 1.408450704225352,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004955795228541917,
|
|
"loss": 5.9694,
|
|
"mean_token_accuracy": 0.1330952487885952,
|
|
"num_tokens": 3094567.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.041680717468262,
|
|
"epoch": 1.4125932062966031,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000495516268285598,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13208642452955247,
|
|
"num_tokens": 3103821.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.039361143112183,
|
|
"epoch": 1.4167357083678542,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004954525689115935,
|
|
"loss": 5.9978,
|
|
"mean_token_accuracy": 0.13572969138622284,
|
|
"num_tokens": 3113399.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.049833297729492,
|
|
"epoch": 1.4208782104391053,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004953884248606669,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.1403762809932232,
|
|
"num_tokens": 3122863.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.1296131134033205,
|
|
"epoch": 1.4250207125103562,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953238362622034,
|
|
"loss": 6.0345,
|
|
"mean_token_accuracy": 0.14129049256443976,
|
|
"num_tokens": 3131250.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.0656098365783695,
|
|
"epoch": 1.4291632145816073,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004952588032464855,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.14013449773192405,
|
|
"num_tokens": 3140120.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.044841003417969,
|
|
"epoch": 1.4333057166528582,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004951933259446919,
|
|
"loss": 6.0884,
|
|
"mean_token_accuracy": 0.13288939371705055,
|
|
"num_tokens": 3149304.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.077728414535523,
|
|
"epoch": 1.4374482187241093,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004951274044888975,
|
|
"loss": 6.0684,
|
|
"mean_token_accuracy": 0.13015799820423127,
|
|
"num_tokens": 3158386.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.156499576568604,
|
|
"epoch": 1.4415907207953604,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004950610390120731,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.12830716595053673,
|
|
"num_tokens": 3166971.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.911584281921387,
|
|
"epoch": 1.4457332228666115,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000494994229648085,
|
|
"loss": 5.9268,
|
|
"mean_token_accuracy": 0.1378532201051712,
|
|
"num_tokens": 3175678.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.382007503509522,
|
|
"epoch": 1.4498757249378624,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004949269765316948,
|
|
"loss": 6.4007,
|
|
"mean_token_accuracy": 0.12360990047454834,
|
|
"num_tokens": 3185151.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.003203964233398,
|
|
"epoch": 1.4540182270091135,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004948592797985599,
|
|
"loss": 5.9886,
|
|
"mean_token_accuracy": 0.12840774431824684,
|
|
"num_tokens": 3193016.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.187309694290161,
|
|
"epoch": 1.4581607290803644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004947911395852317,
|
|
"loss": 6.0342,
|
|
"mean_token_accuracy": 0.1325800023972988,
|
|
"num_tokens": 3201475.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.989087677001953,
|
|
"epoch": 1.4623032311516155,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004947225560291565,
|
|
"loss": 5.9032,
|
|
"mean_token_accuracy": 0.13708751201629638,
|
|
"num_tokens": 3210602.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.112654399871826,
|
|
"epoch": 1.4664457332228666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004946535292686746,
|
|
"loss": 6.2026,
|
|
"mean_token_accuracy": 0.12221089527010917,
|
|
"num_tokens": 3220364.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.232067012786866,
|
|
"epoch": 1.4705882352941178,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004945840594430209,
|
|
"loss": 6.1317,
|
|
"mean_token_accuracy": 0.12892260923981666,
|
|
"num_tokens": 3230019.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.085176420211792,
|
|
"epoch": 1.4747307373653686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004945141466923235,
|
|
"loss": 6.1078,
|
|
"mean_token_accuracy": 0.13496615067124368,
|
|
"num_tokens": 3239211.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.994139719009399,
|
|
"epoch": 1.4788732394366197,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004944437911576041,
|
|
"loss": 5.9044,
|
|
"mean_token_accuracy": 0.1325595885515213,
|
|
"num_tokens": 3248270.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.989656925201416,
|
|
"epoch": 1.4830157415078706,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004943729929807772,
|
|
"loss": 5.9634,
|
|
"mean_token_accuracy": 0.14057789370417595,
|
|
"num_tokens": 3257155.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.030780601501465,
|
|
"epoch": 1.4871582435791217,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004943017523046507,
|
|
"loss": 5.9527,
|
|
"mean_token_accuracy": 0.13465590700507163,
|
|
"num_tokens": 3267242.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.105288457870484,
|
|
"epoch": 1.4913007456503728,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000494230069272925,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.133008673787117,
|
|
"num_tokens": 3276420.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.994972229003906,
|
|
"epoch": 1.495443247721624,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004941579440301923,
|
|
"loss": 5.9439,
|
|
"mean_token_accuracy": 0.13288045302033424,
|
|
"num_tokens": 3285954.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.132870626449585,
|
|
"epoch": 1.4995857497928748,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004940853767219372,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.13481004983186723,
|
|
"num_tokens": 3296448.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.981975412368774,
|
|
"epoch": 1.503728251864126,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004940123674945359,
|
|
"loss": 6.0485,
|
|
"mean_token_accuracy": 0.1355610877275467,
|
|
"num_tokens": 3306084.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.126554489135742,
|
|
"epoch": 1.5078707539353768,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004939389164952561,
|
|
"loss": 6.0361,
|
|
"mean_token_accuracy": 0.13744445368647576,
|
|
"num_tokens": 3314865.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.0176558017730715,
|
|
"epoch": 1.512013256006628,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938650238722564,
|
|
"loss": 5.9756,
|
|
"mean_token_accuracy": 0.13883665949106216,
|
|
"num_tokens": 3323768.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.159342956542969,
|
|
"epoch": 1.516155758077879,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004937906897745861,
|
|
"loss": 6.1306,
|
|
"mean_token_accuracy": 0.1313806004822254,
|
|
"num_tokens": 3333722.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.157751893997192,
|
|
"epoch": 1.5202982601491302,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004937159143521856,
|
|
"loss": 6.0871,
|
|
"mean_token_accuracy": 0.13825847432017327,
|
|
"num_tokens": 3343950.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.122276544570923,
|
|
"epoch": 1.5244407622203813,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004936406977558847,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.1342560924589634,
|
|
"num_tokens": 3352526.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.074772739410401,
|
|
"epoch": 1.5285832642916322,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004935650401374039,
|
|
"loss": 6.0522,
|
|
"mean_token_accuracy": 0.13890262842178344,
|
|
"num_tokens": 3361685.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.141188383102417,
|
|
"epoch": 1.532725766362883,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004934889416493524,
|
|
"loss": 6.012,
|
|
"mean_token_accuracy": 0.13915711641311646,
|
|
"num_tokens": 3369967.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.9735277652740475,
|
|
"epoch": 1.5368682684341342,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004934124024452295,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.12827755585312844,
|
|
"num_tokens": 3379257.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.106106615066528,
|
|
"epoch": 1.5410107705053853,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004933354226794232,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.13620708286762237,
|
|
"num_tokens": 3388157.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.950488042831421,
|
|
"epoch": 1.5451532725766364,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004932580025072098,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1455531969666481,
|
|
"num_tokens": 3397675.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.12842583656311,
|
|
"epoch": 1.5492957746478875,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004931801420847545,
|
|
"loss": 6.0164,
|
|
"mean_token_accuracy": 0.13580601513385773,
|
|
"num_tokens": 3406248.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.97610650062561,
|
|
"epoch": 1.5534382767191384,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004931018415691102,
|
|
"loss": 5.9533,
|
|
"mean_token_accuracy": 0.14214757829904556,
|
|
"num_tokens": 3414600.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.946225595474243,
|
|
"epoch": 1.5575807787903893,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004930231011182176,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.14622760117053984,
|
|
"num_tokens": 3422608.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.0624736785888675,
|
|
"epoch": 1.5617232808616404,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004929439208909048,
|
|
"loss": 5.9276,
|
|
"mean_token_accuracy": 0.13686055019497873,
|
|
"num_tokens": 3431661.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.007996988296509,
|
|
"epoch": 1.5658657829328915,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004928643010468869,
|
|
"loss": 6.0189,
|
|
"mean_token_accuracy": 0.13368499726057054,
|
|
"num_tokens": 3440654.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.140294456481934,
|
|
"epoch": 1.5700082850041426,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004927842417467659,
|
|
"loss": 6.0864,
|
|
"mean_token_accuracy": 0.12963276728987694,
|
|
"num_tokens": 3450041.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.986852550506592,
|
|
"epoch": 1.5741507870753937,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004927037431520302,
|
|
"loss": 6.0653,
|
|
"mean_token_accuracy": 0.1344601631164551,
|
|
"num_tokens": 3459092.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.128014421463012,
|
|
"epoch": 1.5782932891466446,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004926228054250541,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.13942472040653228,
|
|
"num_tokens": 3467789.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.955077886581421,
|
|
"epoch": 1.5824357912178955,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004925414287290979,
|
|
"loss": 5.9321,
|
|
"mean_token_accuracy": 0.13595775440335273,
|
|
"num_tokens": 3476264.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.037222576141358,
|
|
"epoch": 1.5865782932891466,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004924596132283075,
|
|
"loss": 5.9935,
|
|
"mean_token_accuracy": 0.1471448689699173,
|
|
"num_tokens": 3485469.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.157397603988647,
|
|
"epoch": 1.5907207953603977,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004923773590877135,
|
|
"loss": 6.045,
|
|
"mean_token_accuracy": 0.13453733399510384,
|
|
"num_tokens": 3494222.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.979045104980469,
|
|
"epoch": 1.5948632974316488,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004922946664732314,
|
|
"loss": 6.067,
|
|
"mean_token_accuracy": 0.13578267619013787,
|
|
"num_tokens": 3504473.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.982046699523925,
|
|
"epoch": 1.5990057995029,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004922115355516615,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.14055346548557282,
|
|
"num_tokens": 3512704.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.066261434555054,
|
|
"epoch": 1.6031483015741508,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004921279664906877,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.138848215341568,
|
|
"num_tokens": 3521600.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.009872245788574,
|
|
"epoch": 1.6072908036454017,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004920439594588781,
|
|
"loss": 6.118,
|
|
"mean_token_accuracy": 0.12350448817014695,
|
|
"num_tokens": 3531666.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.069752836227417,
|
|
"epoch": 1.6114333057166528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000491959514625684,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.13175680488348007,
|
|
"num_tokens": 3541145.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.08793363571167,
|
|
"epoch": 1.6155758077879039,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049187463216144,
|
|
"loss": 5.9994,
|
|
"mean_token_accuracy": 0.14488411992788314,
|
|
"num_tokens": 3550151.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.134615421295166,
|
|
"epoch": 1.619718309859155,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491789312237363,
|
|
"loss": 5.9934,
|
|
"mean_token_accuracy": 0.1358059048652649,
|
|
"num_tokens": 3560048.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.870369815826416,
|
|
"epoch": 1.623860811930406,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004917035550255528,
|
|
"loss": 5.8206,
|
|
"mean_token_accuracy": 0.1509326860308647,
|
|
"num_tokens": 3568473.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.949672269821167,
|
|
"epoch": 1.628003314001657,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491617360698991,
|
|
"loss": 5.9349,
|
|
"mean_token_accuracy": 0.14107762947678565,
|
|
"num_tokens": 3577188.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.116447257995605,
|
|
"epoch": 1.6321458160729079,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004915307294315412,
|
|
"loss": 6.1147,
|
|
"mean_token_accuracy": 0.1352829672396183,
|
|
"num_tokens": 3587047.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.995927095413208,
|
|
"epoch": 1.636288318144159,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000491443661397948,
|
|
"loss": 6.0111,
|
|
"mean_token_accuracy": 0.13478146567940713,
|
|
"num_tokens": 3595599.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.004126787185669,
|
|
"epoch": 1.64043082021541,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004913561567738368,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.13491463288664818,
|
|
"num_tokens": 3604539.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.084889984130859,
|
|
"epoch": 1.6445733222866612,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004912682157357146,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1341504581272602,
|
|
"num_tokens": 3613463.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.067630052566528,
|
|
"epoch": 1.6487158243579123,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004911798384609675,
|
|
"loss": 5.9769,
|
|
"mean_token_accuracy": 0.134279103577137,
|
|
"num_tokens": 3622131.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.210787439346314,
|
|
"epoch": 1.6528583264291632,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004910910251278626,
|
|
"loss": 6.1647,
|
|
"mean_token_accuracy": 0.1341046206653118,
|
|
"num_tokens": 3632338.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.020812225341797,
|
|
"epoch": 1.6570008285004143,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004910017759155458,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.13550196066498757,
|
|
"num_tokens": 3640941.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6570008285004143,
|
|
"eval_entropy": 5.842879085906787,
|
|
"eval_loss": 6.168454647064209,
|
|
"eval_mean_token_accuracy": 0.13479597267629242,
|
|
"eval_num_tokens": 3640941.0,
|
|
"eval_runtime": 2.9149,
|
|
"eval_samples_per_second": 1285.449,
|
|
"eval_steps_per_second": 160.896,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.0745893001556395,
|
|
"epoch": 1.6611433305716652,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004909120910040427,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.13120997697114944,
|
|
"num_tokens": 3650699.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.0884589672088625,
|
|
"epoch": 1.6652858326429163,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004908219705742574,
|
|
"loss": 6.055,
|
|
"mean_token_accuracy": 0.13785433918237686,
|
|
"num_tokens": 3660734.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.940933465957642,
|
|
"epoch": 1.6694283347141674,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004907314148079727,
|
|
"loss": 5.9021,
|
|
"mean_token_accuracy": 0.14264713525772094,
|
|
"num_tokens": 3670176.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.014130783081055,
|
|
"epoch": 1.6735708367854185,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004906404238878496,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.14376770704984665,
|
|
"num_tokens": 3679559.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.968165588378906,
|
|
"epoch": 1.6777133388566694,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004905489979974265,
|
|
"loss": 5.8978,
|
|
"mean_token_accuracy": 0.14285829290747643,
|
|
"num_tokens": 3688331.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.978603506088257,
|
|
"epoch": 1.6818558409279205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004904571373211197,
|
|
"loss": 5.9251,
|
|
"mean_token_accuracy": 0.14525610953569412,
|
|
"num_tokens": 3696579.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.980389404296875,
|
|
"epoch": 1.6859983429991714,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000490364842044222,
|
|
"loss": 5.9038,
|
|
"mean_token_accuracy": 0.1411435402929783,
|
|
"num_tokens": 3706019.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.014318990707397,
|
|
"epoch": 1.6901408450704225,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004902721123529031,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.14145262613892556,
|
|
"num_tokens": 3715090.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.995026445388794,
|
|
"epoch": 1.6942833471416736,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000490178948434209,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.13903800398111343,
|
|
"num_tokens": 3723721.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.010066366195678,
|
|
"epoch": 1.6984258492129247,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004900853504760614,
|
|
"loss": 6.0361,
|
|
"mean_token_accuracy": 0.13761742487549783,
|
|
"num_tokens": 3732712.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.054876184463501,
|
|
"epoch": 1.7025683512841756,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004899913186672577,
|
|
"loss": 5.9576,
|
|
"mean_token_accuracy": 0.13415518999099732,
|
|
"num_tokens": 3742063.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.035504293441773,
|
|
"epoch": 1.7067108533554267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004898968531974702,
|
|
"loss": 6.0477,
|
|
"mean_token_accuracy": 0.13364452123641968,
|
|
"num_tokens": 3751807.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.036947965621948,
|
|
"epoch": 1.7108533554266776,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004898019542572462,
|
|
"loss": 6.0153,
|
|
"mean_token_accuracy": 0.13318413347005845,
|
|
"num_tokens": 3761316.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.041462182998657,
|
|
"epoch": 1.7149958574979287,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004897066220380073,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.14148937910795212,
|
|
"num_tokens": 3769897.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.976815700531006,
|
|
"epoch": 1.7191383595691798,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004896108567320489,
|
|
"loss": 5.9527,
|
|
"mean_token_accuracy": 0.13763240873813629,
|
|
"num_tokens": 3778678.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.976473712921143,
|
|
"epoch": 1.723280861640431,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048951465853254,
|
|
"loss": 5.9999,
|
|
"mean_token_accuracy": 0.13764951303601264,
|
|
"num_tokens": 3787742.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.105862426757812,
|
|
"epoch": 1.7274233637116818,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004894180276335231,
|
|
"loss": 6.0571,
|
|
"mean_token_accuracy": 0.1314976066350937,
|
|
"num_tokens": 3796300.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.02824821472168,
|
|
"epoch": 1.731565865782933,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004893209642299132,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.13603372126817703,
|
|
"num_tokens": 3805901.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.987863159179687,
|
|
"epoch": 1.7357083678541838,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004892234685174976,
|
|
"loss": 6.0362,
|
|
"mean_token_accuracy": 0.13513321056962013,
|
|
"num_tokens": 3815605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.015270948410034,
|
|
"epoch": 1.739850869925435,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004891255406929359,
|
|
"loss": 5.9213,
|
|
"mean_token_accuracy": 0.13604082316160201,
|
|
"num_tokens": 3824444.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.971472120285034,
|
|
"epoch": 1.743993371996686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004890271809537595,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.13630633503198625,
|
|
"num_tokens": 3833905.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.042605066299439,
|
|
"epoch": 1.7481358740679371,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004889283894983704,
|
|
"loss": 6.0307,
|
|
"mean_token_accuracy": 0.13705341815948485,
|
|
"num_tokens": 3843176.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.8697154998779295,
|
|
"epoch": 1.752278376139188,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000488829166526042,
|
|
"loss": 5.8094,
|
|
"mean_token_accuracy": 0.14387866854667664,
|
|
"num_tokens": 3852263.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.881034851074219,
|
|
"epoch": 1.7564208782104391,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004887295122369176,
|
|
"loss": 5.8988,
|
|
"mean_token_accuracy": 0.14097962900996208,
|
|
"num_tokens": 3861393.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.992901229858399,
|
|
"epoch": 1.76056338028169,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000488629426832011,
|
|
"loss": 5.8526,
|
|
"mean_token_accuracy": 0.13920771479606628,
|
|
"num_tokens": 3870246.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.043943119049072,
|
|
"epoch": 1.7647058823529411,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004885289105132054,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.14054998755455017,
|
|
"num_tokens": 3879486.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.985035562515259,
|
|
"epoch": 1.7688483844241922,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004884279634832532,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.14789656475186347,
|
|
"num_tokens": 3889187.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.948019742965698,
|
|
"epoch": 1.7729908864954433,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004883265859457754,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.14227469339966775,
|
|
"num_tokens": 3898503.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.094852542877197,
|
|
"epoch": 1.7771333885666944,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048822477810526186,
|
|
"loss": 5.9493,
|
|
"mean_token_accuracy": 0.1418413206934929,
|
|
"num_tokens": 3907634.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.974353456497193,
|
|
"epoch": 1.7812758906379453,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048812254016706994,
|
|
"loss": 5.9285,
|
|
"mean_token_accuracy": 0.13976011276245118,
|
|
"num_tokens": 3917057.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.029307699203491,
|
|
"epoch": 1.7854183927091962,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004880198723374247,
|
|
"loss": 6.013,
|
|
"mean_token_accuracy": 0.13228121027350426,
|
|
"num_tokens": 3926854.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.1075136184692385,
|
|
"epoch": 1.7895608947804473,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048791677482341845,
|
|
"loss": 5.991,
|
|
"mean_token_accuracy": 0.1414928138256073,
|
|
"num_tokens": 3935820.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.875960397720337,
|
|
"epoch": 1.7937033968516984,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048781324783301016,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.1361390620470047,
|
|
"num_tokens": 3944804.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.9698957920074465,
|
|
"epoch": 1.7978458989229495,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048770929157502505,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.14194637387990952,
|
|
"num_tokens": 3953534.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.069232940673828,
|
|
"epoch": 1.8019884009942007,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004876049062591543,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.13576488494873046,
|
|
"num_tokens": 3963268.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.971802806854248,
|
|
"epoch": 1.8061309030654515,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048750009209595443,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.1482763409614563,
|
|
"num_tokens": 3972571.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.944825315475464,
|
|
"epoch": 1.8102734051367024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048739484929684714,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.14042048305273055,
|
|
"num_tokens": 3982813.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.9267841339111325,
|
|
"epoch": 1.8144159072079535,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048728917807411856,
|
|
"loss": 5.984,
|
|
"mean_token_accuracy": 0.13693317994475365,
|
|
"num_tokens": 3990743.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.062971305847168,
|
|
"epoch": 1.8185584092792046,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004871830786409191,
|
|
"loss": 5.9275,
|
|
"mean_token_accuracy": 0.1332172743976116,
|
|
"num_tokens": 3999849.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.995508766174316,
|
|
"epoch": 1.8227009113504558,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048707655121126296,
|
|
"loss": 5.9233,
|
|
"mean_token_accuracy": 0.14391714483499526,
|
|
"num_tokens": 4009035.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.967374420166015,
|
|
"epoch": 1.8268434134217069,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048696959600002767,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.1440461315214634,
|
|
"num_tokens": 4018593.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.94206805229187,
|
|
"epoch": 1.8309859154929577,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048686221322295347,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.13509892001748086,
|
|
"num_tokens": 4027744.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.006576442718506,
|
|
"epoch": 1.8351284175642086,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048675440309664314,
|
|
"loss": 5.9717,
|
|
"mean_token_accuracy": 0.13625107258558272,
|
|
"num_tokens": 4036195.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.954492235183716,
|
|
"epoch": 1.8392709196354597,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048664616583856166,
|
|
"loss": 5.9396,
|
|
"mean_token_accuracy": 0.13764657974243164,
|
|
"num_tokens": 4045676.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.978009748458862,
|
|
"epoch": 1.8434134217067109,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048653750166703516,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.14014019593596458,
|
|
"num_tokens": 4055285.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.019568204879761,
|
|
"epoch": 1.847555923777962,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004864284108012513,
|
|
"loss": 5.9945,
|
|
"mean_token_accuracy": 0.13550701290369033,
|
|
"num_tokens": 4065149.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.068925380706787,
|
|
"epoch": 1.851698425849213,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004863188934612583,
|
|
"loss": 6.0048,
|
|
"mean_token_accuracy": 0.13284957855939866,
|
|
"num_tokens": 4074807.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.9438426971435545,
|
|
"epoch": 1.855840927920464,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004862089498679645,
|
|
"loss": 5.863,
|
|
"mean_token_accuracy": 0.14104679301381112,
|
|
"num_tokens": 4083448.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.011416959762573,
|
|
"epoch": 1.8599834299917148,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004860985802431383,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.13921424895524978,
|
|
"num_tokens": 4092810.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.0292891502380375,
|
|
"epoch": 1.864125932062966,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004859877848094071,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.13659395277500153,
|
|
"num_tokens": 4101927.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.995765256881714,
|
|
"epoch": 1.868268434134217,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004858765637902575,
|
|
"loss": 5.8955,
|
|
"mean_token_accuracy": 0.14541091918945312,
|
|
"num_tokens": 4110923.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.908204555511475,
|
|
"epoch": 1.8724109362054682,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004857649174100344,
|
|
"loss": 5.8152,
|
|
"mean_token_accuracy": 0.1410209134221077,
|
|
"num_tokens": 4119939.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.906198358535766,
|
|
"epoch": 1.8765534382767193,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004856528458939409,
|
|
"loss": 5.8779,
|
|
"mean_token_accuracy": 0.14143114984035493,
|
|
"num_tokens": 4129091.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.967985534667969,
|
|
"epoch": 1.8806959403479702,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004855403494680373,
|
|
"loss": 5.8757,
|
|
"mean_token_accuracy": 0.1409454122185707,
|
|
"num_tokens": 4138070.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.897773122787475,
|
|
"epoch": 1.884838442419221,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048542742835924133,
|
|
"loss": 5.8546,
|
|
"mean_token_accuracy": 0.1405012011528015,
|
|
"num_tokens": 4147002.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.997664833068848,
|
|
"epoch": 1.8889809444904722,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048531408279532705,
|
|
"loss": 6.0221,
|
|
"mean_token_accuracy": 0.13353719785809517,
|
|
"num_tokens": 4156525.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.004969120025635,
|
|
"epoch": 1.8931234465617233,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048520031300492507,
|
|
"loss": 5.8583,
|
|
"mean_token_accuracy": 0.14217624366283416,
|
|
"num_tokens": 4165485.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.8434123516082765,
|
|
"epoch": 1.8972659486329744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048508611921752124,
|
|
"loss": 5.8519,
|
|
"mean_token_accuracy": 0.14170566573739052,
|
|
"num_tokens": 4174721.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.057034206390381,
|
|
"epoch": 1.9014084507042255,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048497150166345703,
|
|
"loss": 5.9922,
|
|
"mean_token_accuracy": 0.13678756430745126,
|
|
"num_tokens": 4184981.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.052671194076538,
|
|
"epoch": 1.9055509527754764,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004848564605739287,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.13604575917124748,
|
|
"num_tokens": 4193912.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.900245618820191,
|
|
"epoch": 1.9096934548467275,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048474099618098644,
|
|
"loss": 5.8865,
|
|
"mean_token_accuracy": 0.14115182384848596,
|
|
"num_tokens": 4202904.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.934115266799926,
|
|
"epoch": 1.9138359569179784,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004846251087175348,
|
|
"loss": 5.8392,
|
|
"mean_token_accuracy": 0.1444745846092701,
|
|
"num_tokens": 4212072.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.873389196395874,
|
|
"epoch": 1.9179784589892295,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004845087984173312,
|
|
"loss": 5.7925,
|
|
"mean_token_accuracy": 0.14933519586920738,
|
|
"num_tokens": 4220673.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.9027698040008545,
|
|
"epoch": 1.9221209610604806,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048439206551498644,
|
|
"loss": 5.8719,
|
|
"mean_token_accuracy": 0.14407336413860322,
|
|
"num_tokens": 4229433.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.9854466915130615,
|
|
"epoch": 1.9262634631317317,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004842749102459635,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.1422778360545635,
|
|
"num_tokens": 4239345.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.872208976745606,
|
|
"epoch": 1.9304059652029826,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048415733284657727,
|
|
"loss": 5.7868,
|
|
"mean_token_accuracy": 0.15144341737031936,
|
|
"num_tokens": 4248411.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.881693983078003,
|
|
"epoch": 1.9345484672742337,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004840393335539943,
|
|
"loss": 5.8006,
|
|
"mean_token_accuracy": 0.15395894944667815,
|
|
"num_tokens": 4256745.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.962243127822876,
|
|
"epoch": 1.9386909693454846,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048392091260623195,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.1411220297217369,
|
|
"num_tokens": 4266533.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.138938760757446,
|
|
"epoch": 1.9428334714167357,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048380207024215836,
|
|
"loss": 6.0692,
|
|
"mean_token_accuracy": 0.13505000993609428,
|
|
"num_tokens": 4276598.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.963536787033081,
|
|
"epoch": 1.9469759734879868,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048368280670149145,
|
|
"loss": 5.9542,
|
|
"mean_token_accuracy": 0.13387920409440995,
|
|
"num_tokens": 4285800.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.970139408111573,
|
|
"epoch": 1.951118475559238,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004835631222247989,
|
|
"loss": 5.8984,
|
|
"mean_token_accuracy": 0.1364313505589962,
|
|
"num_tokens": 4295583.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.906875705718994,
|
|
"epoch": 1.9552609776304888,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048344301705349735,
|
|
"loss": 5.7466,
|
|
"mean_token_accuracy": 0.15295102968811988,
|
|
"num_tokens": 4303984.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.84448127746582,
|
|
"epoch": 1.95940347970174,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048332249142985195,
|
|
"loss": 5.907,
|
|
"mean_token_accuracy": 0.13456033021211625,
|
|
"num_tokens": 4312956.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.027677774429321,
|
|
"epoch": 1.9635459817729908,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004832015455969763,
|
|
"loss": 5.8737,
|
|
"mean_token_accuracy": 0.1478977844119072,
|
|
"num_tokens": 4321795.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.860297346115113,
|
|
"epoch": 1.9676884838442419,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048308017979883114,
|
|
"loss": 5.8102,
|
|
"mean_token_accuracy": 0.14385854452848434,
|
|
"num_tokens": 4330579.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.9947404861450195,
|
|
"epoch": 1.971830985915493,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048295839428022454,
|
|
"loss": 5.9063,
|
|
"mean_token_accuracy": 0.1438973106443882,
|
|
"num_tokens": 4339155.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.856879043579101,
|
|
"epoch": 1.975973487986744,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048283618928681127,
|
|
"loss": 5.7939,
|
|
"mean_token_accuracy": 0.14630988240242004,
|
|
"num_tokens": 4348759.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.932731056213379,
|
|
"epoch": 1.980115990057995,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004827135650650923,
|
|
"loss": 5.9188,
|
|
"mean_token_accuracy": 0.1481543578207493,
|
|
"num_tokens": 4358164.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.920789957046509,
|
|
"epoch": 1.984258492129246,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004825905218624139,
|
|
"loss": 5.8305,
|
|
"mean_token_accuracy": 0.14525951594114303,
|
|
"num_tokens": 4366709.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.859962129592896,
|
|
"epoch": 1.988400994200497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048246705992696756,
|
|
"loss": 5.7503,
|
|
"mean_token_accuracy": 0.14968549385666846,
|
|
"num_tokens": 4375055.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.044453191757202,
|
|
"epoch": 1.992543496271748,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048234317950778975,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.13369215801358222,
|
|
"num_tokens": 4384495.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.93978180885315,
|
|
"epoch": 1.9966859983429992,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004822188808547607,
|
|
"loss": 5.8294,
|
|
"mean_token_accuracy": 0.14511772394180297,
|
|
"num_tokens": 4393208.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.007810449600219,
|
|
"epoch": 2.0008285004142503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004820941642186043,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.1328720286488533,
|
|
"num_tokens": 4401882.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.8302782535552975,
|
|
"epoch": 2.0049710024855014,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004819690298508879,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.15972236916422844,
|
|
"num_tokens": 4411434.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.834520959854126,
|
|
"epoch": 2.009113504556752,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048184347800402083,
|
|
"loss": 5.5917,
|
|
"mean_token_accuracy": 0.1504381477832794,
|
|
"num_tokens": 4421519.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.888366079330444,
|
|
"epoch": 2.013256006628003,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004817175089312551,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.16161237508058549,
|
|
"num_tokens": 4430208.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.771408176422119,
|
|
"epoch": 2.0173985086992543,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048159112288668403,
|
|
"loss": 5.6437,
|
|
"mean_token_accuracy": 0.14739402011036873,
|
|
"num_tokens": 4440214.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.905699634552002,
|
|
"epoch": 2.0215410107705054,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048146432012524215,
|
|
"loss": 5.5972,
|
|
"mean_token_accuracy": 0.1480880782008171,
|
|
"num_tokens": 4449406.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.752327966690063,
|
|
"epoch": 2.0256835128417565,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048133710090270446,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.1535774514079094,
|
|
"num_tokens": 4458322.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.858113861083984,
|
|
"epoch": 2.0298260149130076,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048120946547568596,
|
|
"loss": 5.5154,
|
|
"mean_token_accuracy": 0.1561787486076355,
|
|
"num_tokens": 4468132.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.7455836772918705,
|
|
"epoch": 2.0339685169842583,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048108141410164133,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.15598926022648812,
|
|
"num_tokens": 4478846.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.9118971824646,
|
|
"epoch": 2.0381110190555094,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004809529470388642,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.1582340955734253,
|
|
"num_tokens": 4488237.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.821217727661133,
|
|
"epoch": 2.0422535211267605,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004808240645464866,
|
|
"loss": 5.6067,
|
|
"mean_token_accuracy": 0.15413790419697762,
|
|
"num_tokens": 4497568.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.926329755783081,
|
|
"epoch": 2.0463960231980116,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004806947668844787,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.1584421366453171,
|
|
"num_tokens": 4506050.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.811815643310547,
|
|
"epoch": 2.0505385252692627,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048056505431364794,
|
|
"loss": 5.5402,
|
|
"mean_token_accuracy": 0.1506636083126068,
|
|
"num_tokens": 4514734.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.803349113464355,
|
|
"epoch": 2.054681027340514,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048043492709563877,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.1571820303797722,
|
|
"num_tokens": 4524026.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.774190044403076,
|
|
"epoch": 2.0588235294117645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048030438549293193,
|
|
"loss": 5.5649,
|
|
"mean_token_accuracy": 0.156453837454319,
|
|
"num_tokens": 4532209.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.891646766662598,
|
|
"epoch": 2.0629660314830156,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048017342976884424,
|
|
"loss": 5.5948,
|
|
"mean_token_accuracy": 0.15483448803424835,
|
|
"num_tokens": 4542095.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.751562309265137,
|
|
"epoch": 2.0671085335542667,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048004206018752773,
|
|
"loss": 5.5585,
|
|
"mean_token_accuracy": 0.15620076209306716,
|
|
"num_tokens": 4550530.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.836578750610352,
|
|
"epoch": 2.071251035625518,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047991027701396895,
|
|
"loss": 5.6432,
|
|
"mean_token_accuracy": 0.1558740496635437,
|
|
"num_tokens": 4560560.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.071251035625518,
|
|
"eval_entropy": 5.817075447741348,
|
|
"eval_loss": 6.062323093414307,
|
|
"eval_mean_token_accuracy": 0.14113807419279237,
|
|
"eval_num_tokens": 4560560.0,
|
|
"eval_runtime": 2.9157,
|
|
"eval_samples_per_second": 1285.093,
|
|
"eval_steps_per_second": 160.851,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.922831439971924,
|
|
"epoch": 2.075393537696769,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004797780805139892,
|
|
"loss": 5.5875,
|
|
"mean_token_accuracy": 0.15712048560380937,
|
|
"num_tokens": 4570459.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.71963062286377,
|
|
"epoch": 2.07953603976802,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004796454709542432,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.15603303909301758,
|
|
"num_tokens": 4578595.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.822841930389404,
|
|
"epoch": 2.083678541839271,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795124486022189,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.15312176644802095,
|
|
"num_tokens": 4588207.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.7801848411560055,
|
|
"epoch": 2.087821043910522,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047937901372623694,
|
|
"loss": 5.5188,
|
|
"mean_token_accuracy": 0.1585146576166153,
|
|
"num_tokens": 4596819.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.825238847732544,
|
|
"epoch": 2.091963545981773,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047924516659545005,
|
|
"loss": 5.5152,
|
|
"mean_token_accuracy": 0.1578638359904289,
|
|
"num_tokens": 4604741.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.674453163146973,
|
|
"epoch": 2.096106048053024,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004791109074798425,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.15454403012990953,
|
|
"num_tokens": 4614188.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.8349532127380375,
|
|
"epoch": 2.100248550124275,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004789762366502296,
|
|
"loss": 5.5228,
|
|
"mean_token_accuracy": 0.15179868340492247,
|
|
"num_tokens": 4623976.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.823003673553467,
|
|
"epoch": 2.1043910521955262,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004788411543782571,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.1656728833913803,
|
|
"num_tokens": 4632672.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.76104588508606,
|
|
"epoch": 2.108533554266777,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004787056609364008,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.15623851865530014,
|
|
"num_tokens": 4642149.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.675738000869751,
|
|
"epoch": 2.112676056338028,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004785697565979657,
|
|
"loss": 5.3937,
|
|
"mean_token_accuracy": 0.16207751482725144,
|
|
"num_tokens": 4650746.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.702557945251465,
|
|
"epoch": 2.116818558409279,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004784334416370857,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.15734572261571883,
|
|
"num_tokens": 4660154.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.834080362319947,
|
|
"epoch": 2.1209610604805302,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004782967163287231,
|
|
"loss": 5.6194,
|
|
"mean_token_accuracy": 0.1515902802348137,
|
|
"num_tokens": 4668871.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.672147941589356,
|
|
"epoch": 2.1251035625517813,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047815958094866756,
|
|
"loss": 5.5187,
|
|
"mean_token_accuracy": 0.15729935914278032,
|
|
"num_tokens": 4677880.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.909453010559082,
|
|
"epoch": 2.1292460646230325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004780220357735363,
|
|
"loss": 5.7154,
|
|
"mean_token_accuracy": 0.14850281402468682,
|
|
"num_tokens": 4687253.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.834644842147827,
|
|
"epoch": 2.1333885666942836,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047788408108077294,
|
|
"loss": 5.6016,
|
|
"mean_token_accuracy": 0.1510675936937332,
|
|
"num_tokens": 4696392.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.802241468429566,
|
|
"epoch": 2.1375310687655342,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004777457171486471,
|
|
"loss": 5.5617,
|
|
"mean_token_accuracy": 0.1504868522286415,
|
|
"num_tokens": 4704812.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.826993179321289,
|
|
"epoch": 2.1416735708367853,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004776069442562541,
|
|
"loss": 5.5251,
|
|
"mean_token_accuracy": 0.15672111809253692,
|
|
"num_tokens": 4713411.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.7558825492858885,
|
|
"epoch": 2.1458160729080364,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004774677626835138,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.1601279854774475,
|
|
"num_tokens": 4722966.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.761832189559937,
|
|
"epoch": 2.1499585749792876,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047732817271117084,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.14909747391939163,
|
|
"num_tokens": 4731889.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.731906175613403,
|
|
"epoch": 2.1541010770505387,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004771881746207934,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.16156674027442933,
|
|
"num_tokens": 4740231.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.7913134574890135,
|
|
"epoch": 2.1582435791217898,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047704776869477285,
|
|
"loss": 5.6222,
|
|
"mean_token_accuracy": 0.14979231655597686,
|
|
"num_tokens": 4748555.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.844248533248901,
|
|
"epoch": 2.1623860811930404,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047690695521632344,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.1507200427353382,
|
|
"num_tokens": 4757185.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.7017820358276365,
|
|
"epoch": 2.1665285832642915,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047676573446948114,
|
|
"loss": 5.5526,
|
|
"mean_token_accuracy": 0.153288733959198,
|
|
"num_tokens": 4766518.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.749460935592651,
|
|
"epoch": 2.1706710853355426,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047662410673910376,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.15899401307106018,
|
|
"num_tokens": 4776005.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.7959081649780275,
|
|
"epoch": 2.1748135874067938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047648207231087,
|
|
"loss": 5.5733,
|
|
"mean_token_accuracy": 0.15494658946990966,
|
|
"num_tokens": 4785343.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.729545402526855,
|
|
"epoch": 2.178956089478045,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004763396314712786,
|
|
"loss": 5.4946,
|
|
"mean_token_accuracy": 0.16482714265584947,
|
|
"num_tokens": 4794229.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.8528862476348875,
|
|
"epoch": 2.183098591549296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004761967845076484,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.14892114326357841,
|
|
"num_tokens": 4803182.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.812585163116455,
|
|
"epoch": 2.1872410936205466,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004760535317081174,
|
|
"loss": 5.5233,
|
|
"mean_token_accuracy": 0.16473610252141951,
|
|
"num_tokens": 4811820.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.747249031066895,
|
|
"epoch": 2.1913835956917977,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047590987336164207,
|
|
"loss": 5.5679,
|
|
"mean_token_accuracy": 0.15518500953912734,
|
|
"num_tokens": 4821273.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.786758613586426,
|
|
"epoch": 2.195526097763049,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047576580975799705,
|
|
"loss": 5.5832,
|
|
"mean_token_accuracy": 0.15140218138694764,
|
|
"num_tokens": 4830130.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.772381114959717,
|
|
"epoch": 2.1996685998343,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004756213411877743,
|
|
"loss": 5.598,
|
|
"mean_token_accuracy": 0.15639565736055375,
|
|
"num_tokens": 4839238.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.714341497421264,
|
|
"epoch": 2.203811101905551,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.16323678493499755,
|
|
"num_tokens": 4848434.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.684457731246948,
|
|
"epoch": 2.207953603976802,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047533119031404757,
|
|
"loss": 5.5609,
|
|
"mean_token_accuracy": 0.1529301807284355,
|
|
"num_tokens": 4857884.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.672632503509521,
|
|
"epoch": 2.212096106048053,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047518550859580966,
|
|
"loss": 5.4334,
|
|
"mean_token_accuracy": 0.1699458748102188,
|
|
"num_tokens": 4866595.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.8053535461425785,
|
|
"epoch": 2.216238608119304,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004750394230815249,
|
|
"loss": 5.5857,
|
|
"mean_token_accuracy": 0.148395074903965,
|
|
"num_tokens": 4875573.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.796173238754273,
|
|
"epoch": 2.220381110190555,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047489293406586375,
|
|
"loss": 5.5903,
|
|
"mean_token_accuracy": 0.1620958462357521,
|
|
"num_tokens": 4884977.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.723018169403076,
|
|
"epoch": 2.224523612261806,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004747460418443106,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.15953139066696168,
|
|
"num_tokens": 4894091.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.637625503540039,
|
|
"epoch": 2.2286661143330573,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047459874671316305,
|
|
"loss": 5.4194,
|
|
"mean_token_accuracy": 0.16242783069610595,
|
|
"num_tokens": 4903059.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.641981792449951,
|
|
"epoch": 2.2328086164043084,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047445104896953153,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.17116804122924806,
|
|
"num_tokens": 4911969.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.838920593261719,
|
|
"epoch": 2.236951118475559,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047430294891133846,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.15505664348602294,
|
|
"num_tokens": 4920915.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.79048752784729,
|
|
"epoch": 2.24109362054681,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004741544468373179,
|
|
"loss": 5.5957,
|
|
"mean_token_accuracy": 0.15736506879329681,
|
|
"num_tokens": 4931048.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.746155166625977,
|
|
"epoch": 2.2452361226180613,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047400554304701474,
|
|
"loss": 5.5751,
|
|
"mean_token_accuracy": 0.15703048855066298,
|
|
"num_tokens": 4939483.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.6905131340026855,
|
|
"epoch": 2.2493786246893124,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047385623784078427,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.15871881544589997,
|
|
"num_tokens": 4948888.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.770149040222168,
|
|
"epoch": 2.2535211267605635,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004737065315197913,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.15765973925590515,
|
|
"num_tokens": 4956950.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.635520935058594,
|
|
"epoch": 2.2576636288318146,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047355642438600987,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.15296801626682283,
|
|
"num_tokens": 4965968.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.621658706665039,
|
|
"epoch": 2.2618061309030653,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004734059167422225,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.16240133494138717,
|
|
"num_tokens": 4975742.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.73398118019104,
|
|
"epoch": 2.2659486329743164,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047325500889201956,
|
|
"loss": 5.586,
|
|
"mean_token_accuracy": 0.15448082536458968,
|
|
"num_tokens": 4984991.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.708168172836304,
|
|
"epoch": 2.2700911350455675,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004731037011397986,
|
|
"loss": 5.5017,
|
|
"mean_token_accuracy": 0.15908219516277314,
|
|
"num_tokens": 4993574.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.700465059280395,
|
|
"epoch": 2.2742336371168186,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047295199379076403,
|
|
"loss": 5.5119,
|
|
"mean_token_accuracy": 0.16220880597829818,
|
|
"num_tokens": 5001905.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.812096071243286,
|
|
"epoch": 2.2783761391880697,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000472799887150926,
|
|
"loss": 5.6517,
|
|
"mean_token_accuracy": 0.1502576395869255,
|
|
"num_tokens": 5010554.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.749092102050781,
|
|
"epoch": 2.282518641259321,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004726473815271003,
|
|
"loss": 5.5311,
|
|
"mean_token_accuracy": 0.1561155468225479,
|
|
"num_tokens": 5019678.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.719474124908447,
|
|
"epoch": 2.286661143330572,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047249447722690747,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.15179549306631088,
|
|
"num_tokens": 5028180.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.747410917282105,
|
|
"epoch": 2.2908036454018226,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047234117455877215,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.15653812736272812,
|
|
"num_tokens": 5037083.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.838072156906128,
|
|
"epoch": 2.2949461474730737,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004721874738319225,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.15274079591035844,
|
|
"num_tokens": 5046666.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.671936702728272,
|
|
"epoch": 2.299088649544325,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004720333753563898,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.15928964912891388,
|
|
"num_tokens": 5055053.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.686665153503418,
|
|
"epoch": 2.303231151615576,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047187887944300747,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.1539557047188282,
|
|
"num_tokens": 5064049.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.795013904571533,
|
|
"epoch": 2.307373653686827,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004717239864034107,
|
|
"loss": 5.6388,
|
|
"mean_token_accuracy": 0.15225408375263214,
|
|
"num_tokens": 5072420.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.701117706298828,
|
|
"epoch": 2.3115161557580777,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047156869655003554,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.1541515976190567,
|
|
"num_tokens": 5081581.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.693496513366699,
|
|
"epoch": 2.315658657829329,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047141301019611875,
|
|
"loss": 5.5012,
|
|
"mean_token_accuracy": 0.15954914838075637,
|
|
"num_tokens": 5090204.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.7152317523956295,
|
|
"epoch": 2.31980115990058,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004712569276556967,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.15084023624658585,
|
|
"num_tokens": 5099558.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.662888717651367,
|
|
"epoch": 2.323943661971831,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047110044924360484,
|
|
"loss": 5.4811,
|
|
"mean_token_accuracy": 0.16062808632850648,
|
|
"num_tokens": 5107813.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.802162504196167,
|
|
"epoch": 2.328086164043082,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047094357527547725,
|
|
"loss": 5.5867,
|
|
"mean_token_accuracy": 0.15557624846696855,
|
|
"num_tokens": 5117154.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.757091856002807,
|
|
"epoch": 2.332228666114333,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004707863060677459,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.15680305510759354,
|
|
"num_tokens": 5127823.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.655477809906006,
|
|
"epoch": 2.3363711681855843,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047062864193764,
|
|
"loss": 5.5422,
|
|
"mean_token_accuracy": 0.1533115178346634,
|
|
"num_tokens": 5135903.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.664518690109253,
|
|
"epoch": 2.340513670256835,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047047058320318525,
|
|
"loss": 5.4854,
|
|
"mean_token_accuracy": 0.15575041621923447,
|
|
"num_tokens": 5145657.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.735196256637574,
|
|
"epoch": 2.344656172328086,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004703121301832034,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.15849591195583343,
|
|
"num_tokens": 5155061.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.7144616603851315,
|
|
"epoch": 2.348798674399337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047015328319731153,
|
|
"loss": 5.4712,
|
|
"mean_token_accuracy": 0.1625610962510109,
|
|
"num_tokens": 5164302.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.6743780136108395,
|
|
"epoch": 2.3529411764705883,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004699940425659215,
|
|
"loss": 5.6349,
|
|
"mean_token_accuracy": 0.1537710279226303,
|
|
"num_tokens": 5173469.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.824509334564209,
|
|
"epoch": 2.3570836785418394,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046983440861023877,
|
|
"loss": 5.6457,
|
|
"mean_token_accuracy": 0.15466391295194626,
|
|
"num_tokens": 5182548.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.728637599945069,
|
|
"epoch": 2.36122618061309,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046967438165226275,
|
|
"loss": 5.6148,
|
|
"mean_token_accuracy": 0.14389563351869583,
|
|
"num_tokens": 5191802.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.807513904571533,
|
|
"epoch": 2.365368682684341,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004695139620147851,
|
|
"loss": 5.7199,
|
|
"mean_token_accuracy": 0.15426884219050407,
|
|
"num_tokens": 5201986.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.727898979187012,
|
|
"epoch": 2.3695111847555923,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046935315002138994,
|
|
"loss": 5.5067,
|
|
"mean_token_accuracy": 0.1532989665865898,
|
|
"num_tokens": 5211041.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.699389410018921,
|
|
"epoch": 2.3736536868268434,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004691919459964525,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.15786828845739365,
|
|
"num_tokens": 5219838.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.727614879608154,
|
|
"epoch": 2.3777961888980945,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046903035026513894,
|
|
"loss": 5.528,
|
|
"mean_token_accuracy": 0.1572413668036461,
|
|
"num_tokens": 5228282.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.671488904953003,
|
|
"epoch": 2.3819386909693456,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004688683631534056,
|
|
"loss": 5.5662,
|
|
"mean_token_accuracy": 0.15757867842912673,
|
|
"num_tokens": 5238038.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.707324075698852,
|
|
"epoch": 2.3860811930405967,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004687059849879981,
|
|
"loss": 5.5565,
|
|
"mean_token_accuracy": 0.15727738589048385,
|
|
"num_tokens": 5247084.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.7053502082824705,
|
|
"epoch": 2.3902236951118474,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046854321609645093,
|
|
"loss": 5.5825,
|
|
"mean_token_accuracy": 0.15503206253051757,
|
|
"num_tokens": 5256630.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.772836065292358,
|
|
"epoch": 2.3943661971830985,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046838005680708677,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.15751295685768127,
|
|
"num_tokens": 5266854.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.6266319274902346,
|
|
"epoch": 2.3985086992543496,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046821650744901573,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.15903597474098205,
|
|
"num_tokens": 5275823.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.7099535942077635,
|
|
"epoch": 2.4026512013256007,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004680525683521347,
|
|
"loss": 5.5698,
|
|
"mean_token_accuracy": 0.15646034926176072,
|
|
"num_tokens": 5285190.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.8103772640228275,
|
|
"epoch": 2.406793703396852,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004678882398471269,
|
|
"loss": 5.523,
|
|
"mean_token_accuracy": 0.15891006737947463,
|
|
"num_tokens": 5293832.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.590553045272827,
|
|
"epoch": 2.4109362054681025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046772352226546066,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.15144434422254563,
|
|
"num_tokens": 5303514.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.71599235534668,
|
|
"epoch": 2.4150787075393536,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004675584159393895,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.15650684982538224,
|
|
"num_tokens": 5313361.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.752013874053955,
|
|
"epoch": 2.4192212096106047,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046739292120195083,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.1579284265637398,
|
|
"num_tokens": 5323556.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.751535844802857,
|
|
"epoch": 2.423363711681856,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004672270383869657,
|
|
"loss": 5.5814,
|
|
"mean_token_accuracy": 0.15038304030895233,
|
|
"num_tokens": 5333381.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.708028030395508,
|
|
"epoch": 2.427506213753107,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004670607678290379,
|
|
"loss": 5.557,
|
|
"mean_token_accuracy": 0.15642340034246444,
|
|
"num_tokens": 5343722.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.709355688095092,
|
|
"epoch": 2.431648715824358,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004668941098635532,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.16031453162431716,
|
|
"num_tokens": 5352680.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.697698545455933,
|
|
"epoch": 2.435791217895609,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046672706482667897,
|
|
"loss": 5.5378,
|
|
"mean_token_accuracy": 0.15933478623628616,
|
|
"num_tokens": 5361838.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.6182372093200685,
|
|
"epoch": 2.43993371996686,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004665596330553633,
|
|
"loss": 5.5097,
|
|
"mean_token_accuracy": 0.16193565130233764,
|
|
"num_tokens": 5370806.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.677807283401489,
|
|
"epoch": 2.444076222038111,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046639181488733445,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.16008703261613846,
|
|
"num_tokens": 5379690.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.696961450576782,
|
|
"epoch": 2.448218724109362,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004662236106610999,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.15375241935253142,
|
|
"num_tokens": 5388266.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.676030540466309,
|
|
"epoch": 2.452361226180613,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046605502071594604,
|
|
"loss": 5.5166,
|
|
"mean_token_accuracy": 0.15484256520867348,
|
|
"num_tokens": 5396202.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.603984308242798,
|
|
"epoch": 2.4565037282518642,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004658860453919371,
|
|
"loss": 5.4235,
|
|
"mean_token_accuracy": 0.16309440582990647,
|
|
"num_tokens": 5405265.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.765998888015747,
|
|
"epoch": 2.460646230323115,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004657166850299149,
|
|
"loss": 5.6232,
|
|
"mean_token_accuracy": 0.1521397665143013,
|
|
"num_tokens": 5414862.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.677093935012818,
|
|
"epoch": 2.464788732394366,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004655469399714978,
|
|
"loss": 5.5603,
|
|
"mean_token_accuracy": 0.16458463966846465,
|
|
"num_tokens": 5424065.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.699638605117798,
|
|
"epoch": 2.468931234465617,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004653768105590799,
|
|
"loss": 5.5269,
|
|
"mean_token_accuracy": 0.16183326244354249,
|
|
"num_tokens": 5433598.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.756310033798218,
|
|
"epoch": 2.4730737365368682,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046520629713583115,
|
|
"loss": 5.6508,
|
|
"mean_token_accuracy": 0.15125053226947785,
|
|
"num_tokens": 5443370.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.742058801651001,
|
|
"epoch": 2.4772162386081193,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046503540004569555,
|
|
"loss": 5.5286,
|
|
"mean_token_accuracy": 0.15547680854797363,
|
|
"num_tokens": 5452513.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.683076810836792,
|
|
"epoch": 2.4813587406793705,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046486411963339125,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.15302325487136842,
|
|
"num_tokens": 5461212.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.726020097732544,
|
|
"epoch": 2.4855012427506216,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046469245624440955,
|
|
"loss": 5.5638,
|
|
"mean_token_accuracy": 0.1646001785993576,
|
|
"num_tokens": 5470450.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.4855012427506216,
|
|
"eval_entropy": 5.666851568323717,
|
|
"eval_loss": 5.997962474822998,
|
|
"eval_mean_token_accuracy": 0.1460237345620513,
|
|
"eval_num_tokens": 5470450.0,
|
|
"eval_runtime": 2.8811,
|
|
"eval_samples_per_second": 1300.547,
|
|
"eval_steps_per_second": 162.785,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.7583527088165285,
|
|
"epoch": 2.4896437448218722,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046452041022501423,
|
|
"loss": 5.5164,
|
|
"mean_token_accuracy": 0.16106657758355142,
|
|
"num_tokens": 5478839.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.73330397605896,
|
|
"epoch": 2.4937862468931233,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004643479819222411,
|
|
"loss": 5.5609,
|
|
"mean_token_accuracy": 0.15447363406419753,
|
|
"num_tokens": 5487935.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.752573728561401,
|
|
"epoch": 2.4979287489643744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046417517168389664,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.15475835651159286,
|
|
"num_tokens": 5496369.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.710302925109863,
|
|
"epoch": 2.5020712510356256,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004640019798585581,
|
|
"loss": 5.5703,
|
|
"mean_token_accuracy": 0.1553514450788498,
|
|
"num_tokens": 5504263.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.642459678649902,
|
|
"epoch": 2.5062137531068767,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046382840679557233,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.1662412628531456,
|
|
"num_tokens": 5513164.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.786639738082886,
|
|
"epoch": 2.5103562551781273,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046365445284505516,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.15565225183963777,
|
|
"num_tokens": 5522118.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.604031181335449,
|
|
"epoch": 2.5144987572493784,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004634801183578907,
|
|
"loss": 5.5399,
|
|
"mean_token_accuracy": 0.1582293838262558,
|
|
"num_tokens": 5531305.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.727992343902588,
|
|
"epoch": 2.5186412593206295,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046330540368573063,
|
|
"loss": 5.5667,
|
|
"mean_token_accuracy": 0.15120545104146005,
|
|
"num_tokens": 5539474.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.672655773162842,
|
|
"epoch": 2.5227837613918807,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046313030918099365,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.1591261774301529,
|
|
"num_tokens": 5549114.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.6598734855651855,
|
|
"epoch": 2.5269262634631318,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046295483519686437,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.15746332854032516,
|
|
"num_tokens": 5558118.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.661467361450195,
|
|
"epoch": 2.531068765534383,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046277898208729306,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.15640463680028915,
|
|
"num_tokens": 5567635.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.769457387924194,
|
|
"epoch": 2.535211267605634,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004626027502069947,
|
|
"loss": 5.6279,
|
|
"mean_token_accuracy": 0.1561012625694275,
|
|
"num_tokens": 5577272.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.770507669448852,
|
|
"epoch": 2.539353769676885,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046242613991144814,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.16126794517040252,
|
|
"num_tokens": 5586227.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.682786655426026,
|
|
"epoch": 2.5434962717481358,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046224915155689586,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.15951058715581895,
|
|
"num_tokens": 5595648.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.614820337295532,
|
|
"epoch": 2.547638773819387,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046207178550034244,
|
|
"loss": 5.4344,
|
|
"mean_token_accuracy": 0.17111728340387344,
|
|
"num_tokens": 5604187.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.622501564025879,
|
|
"epoch": 2.551781275890638,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004618940420995547,
|
|
"loss": 5.527,
|
|
"mean_token_accuracy": 0.15877681821584702,
|
|
"num_tokens": 5612850.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.678854370117188,
|
|
"epoch": 2.555923777961889,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004617159217130606,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.15891224294900894,
|
|
"num_tokens": 5622902.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.677180290222168,
|
|
"epoch": 2.5600662800331397,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004615374247001483,
|
|
"loss": 5.5319,
|
|
"mean_token_accuracy": 0.14990963488817216,
|
|
"num_tokens": 5631854.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.753375482559204,
|
|
"epoch": 2.564208782104391,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004613585514208659,
|
|
"loss": 5.5185,
|
|
"mean_token_accuracy": 0.16782330721616745,
|
|
"num_tokens": 5640237.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.8014825820922855,
|
|
"epoch": 2.568351284175642,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046117930223602016,
|
|
"loss": 5.6747,
|
|
"mean_token_accuracy": 0.15266044735908507,
|
|
"num_tokens": 5650182.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.6702837467193605,
|
|
"epoch": 2.572493786246893,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004609996775071765,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.1625978335738182,
|
|
"num_tokens": 5659447.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.664971351623535,
|
|
"epoch": 2.576636288318144,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004608196775966575,
|
|
"loss": 5.5933,
|
|
"mean_token_accuracy": 0.1530801922082901,
|
|
"num_tokens": 5668349.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.830552959442139,
|
|
"epoch": 2.5807787903893953,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046063930286754267,
|
|
"loss": 5.6506,
|
|
"mean_token_accuracy": 0.1469719246029854,
|
|
"num_tokens": 5677296.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.737857675552368,
|
|
"epoch": 2.5849212924606464,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046045855368366756,
|
|
"loss": 5.5505,
|
|
"mean_token_accuracy": 0.15921184718608855,
|
|
"num_tokens": 5685822.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.654632902145385,
|
|
"epoch": 2.5890637945318975,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000460277430409623,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.15037565529346467,
|
|
"num_tokens": 5695171.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.782729053497315,
|
|
"epoch": 2.593206296603148,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046009593341075444,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.15687357783317565,
|
|
"num_tokens": 5704559.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.628320407867432,
|
|
"epoch": 2.5973487986743993,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004599140630531612,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.16054383665323257,
|
|
"num_tokens": 5713690.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.740268182754517,
|
|
"epoch": 2.6014913007456504,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004597318197036956,
|
|
"loss": 5.7474,
|
|
"mean_token_accuracy": 0.15208713933825493,
|
|
"num_tokens": 5723792.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.785957098007202,
|
|
"epoch": 2.6056338028169015,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004595492037299624,
|
|
"loss": 5.5389,
|
|
"mean_token_accuracy": 0.16184956580400467,
|
|
"num_tokens": 5732460.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.719061231613159,
|
|
"epoch": 2.6097763048881526,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045936621550031806,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.15228736773133278,
|
|
"num_tokens": 5741770.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.731397724151611,
|
|
"epoch": 2.6139188069594033,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045918285538386973,
|
|
"loss": 5.5834,
|
|
"mean_token_accuracy": 0.15751861929893493,
|
|
"num_tokens": 5750821.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.670954942703247,
|
|
"epoch": 2.6180613090306544,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045899912375047485,
|
|
"loss": 5.5937,
|
|
"mean_token_accuracy": 0.15194432139396669,
|
|
"num_tokens": 5759792.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.701627111434936,
|
|
"epoch": 2.6222038111019055,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045881502097074023,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.15721253752708436,
|
|
"num_tokens": 5769650.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.726138496398926,
|
|
"epoch": 2.6263463131731566,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004586305474160212,
|
|
"loss": 5.592,
|
|
"mean_token_accuracy": 0.1525888927280903,
|
|
"num_tokens": 5779027.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.653416013717651,
|
|
"epoch": 2.6304888152444077,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045844570345842116,
|
|
"loss": 5.5616,
|
|
"mean_token_accuracy": 0.15808800384402275,
|
|
"num_tokens": 5787860.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.8046799659729,
|
|
"epoch": 2.634631317315659,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004582604894707906,
|
|
"loss": 5.6338,
|
|
"mean_token_accuracy": 0.15353483557701111,
|
|
"num_tokens": 5798211.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.68250994682312,
|
|
"epoch": 2.63877381938691,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045807490582672634,
|
|
"loss": 5.5907,
|
|
"mean_token_accuracy": 0.1567721262574196,
|
|
"num_tokens": 5807484.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.7200438499450685,
|
|
"epoch": 2.6429163214581606,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045788895290057083,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.1579717665910721,
|
|
"num_tokens": 5815969.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.740452289581299,
|
|
"epoch": 2.6470588235294117,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045770263106741137,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.15720631256699563,
|
|
"num_tokens": 5825193.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.695869445800781,
|
|
"epoch": 2.651201325600663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004575159407030797,
|
|
"loss": 5.5764,
|
|
"mean_token_accuracy": 0.15763850808143615,
|
|
"num_tokens": 5833628.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.713213205337524,
|
|
"epoch": 2.655343827671914,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004573288821841505,
|
|
"loss": 5.6378,
|
|
"mean_token_accuracy": 0.15480245947837828,
|
|
"num_tokens": 5841562.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.809486722946167,
|
|
"epoch": 2.659486329743165,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004571414558879413,
|
|
"loss": 5.6349,
|
|
"mean_token_accuracy": 0.14864409267902373,
|
|
"num_tokens": 5851145.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.747229051589966,
|
|
"epoch": 2.6636288318144157,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004569536621925114,
|
|
"loss": 5.5888,
|
|
"mean_token_accuracy": 0.15000322833657265,
|
|
"num_tokens": 5860694.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.612951326370239,
|
|
"epoch": 2.667771333885667,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004567655014766612,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.15544012635946275,
|
|
"num_tokens": 5870143.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.695223188400268,
|
|
"epoch": 2.671913835956918,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004565769741199314,
|
|
"loss": 5.5615,
|
|
"mean_token_accuracy": 0.14824787601828576,
|
|
"num_tokens": 5878872.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.685911417007446,
|
|
"epoch": 2.676056338028169,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045638808050260237,
|
|
"loss": 5.5463,
|
|
"mean_token_accuracy": 0.15454385727643966,
|
|
"num_tokens": 5887713.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.738532686233521,
|
|
"epoch": 2.68019884009942,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045619882100569317,
|
|
"loss": 5.5947,
|
|
"mean_token_accuracy": 0.15959140807390212,
|
|
"num_tokens": 5896268.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.7656388759613035,
|
|
"epoch": 2.684341342170671,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004560091960109608,
|
|
"loss": 5.5598,
|
|
"mean_token_accuracy": 0.15922954678535461,
|
|
"num_tokens": 5905584.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.7327958106994625,
|
|
"epoch": 2.6884838442419223,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004558192059008996,
|
|
"loss": 5.6933,
|
|
"mean_token_accuracy": 0.14835613816976548,
|
|
"num_tokens": 5916636.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.779455757141113,
|
|
"epoch": 2.692626346313173,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004556288510587406,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.1529821887612343,
|
|
"num_tokens": 5925629.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.792713212966919,
|
|
"epoch": 2.696768848384424,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045543813186845004,
|
|
"loss": 5.6845,
|
|
"mean_token_accuracy": 0.1453479304909706,
|
|
"num_tokens": 5935316.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.700288534164429,
|
|
"epoch": 2.700911350455675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004552470487147294,
|
|
"loss": 5.5887,
|
|
"mean_token_accuracy": 0.1574153408408165,
|
|
"num_tokens": 5944936.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.704364824295044,
|
|
"epoch": 2.7050538525269263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045505560198301443,
|
|
"loss": 5.5649,
|
|
"mean_token_accuracy": 0.1547586739063263,
|
|
"num_tokens": 5954106.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.689199924468994,
|
|
"epoch": 2.7091963545981774,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045486379205947396,
|
|
"loss": 5.5079,
|
|
"mean_token_accuracy": 0.162031652033329,
|
|
"num_tokens": 5963015.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.812100887298584,
|
|
"epoch": 2.713338856669428,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004546716193310098,
|
|
"loss": 5.711,
|
|
"mean_token_accuracy": 0.1566604346036911,
|
|
"num_tokens": 5972444.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.654112815856934,
|
|
"epoch": 2.717481358740679,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000454479084185255,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.1566864624619484,
|
|
"num_tokens": 5981397.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.636119270324707,
|
|
"epoch": 2.7216238608119303,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004542861870105743,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.1622664988040924,
|
|
"num_tokens": 5989491.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.739955615997315,
|
|
"epoch": 2.7257663628831814,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045409292819606224,
|
|
"loss": 5.6167,
|
|
"mean_token_accuracy": 0.1547279179096222,
|
|
"num_tokens": 5997586.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.695828151702881,
|
|
"epoch": 2.7299088649544325,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045389930813154297,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.17207208275794983,
|
|
"num_tokens": 6006378.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.671674203872681,
|
|
"epoch": 2.7340513670256836,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004537053272075693,
|
|
"loss": 5.5369,
|
|
"mean_token_accuracy": 0.159752693772316,
|
|
"num_tokens": 6016058.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.618702602386475,
|
|
"epoch": 2.7381938690969347,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045351098581542204,
|
|
"loss": 5.5509,
|
|
"mean_token_accuracy": 0.16341104060411454,
|
|
"num_tokens": 6025315.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.689493608474732,
|
|
"epoch": 2.742336371168186,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004533162843471089,
|
|
"loss": 5.5401,
|
|
"mean_token_accuracy": 0.15370434224605561,
|
|
"num_tokens": 6035048.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.669504976272583,
|
|
"epoch": 2.7464788732394365,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004531212231953641,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.15908312052488327,
|
|
"num_tokens": 6043299.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.782401418685913,
|
|
"epoch": 2.7506213753106876,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004529258027536473,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.15550746768712997,
|
|
"num_tokens": 6052396.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.726404809951783,
|
|
"epoch": 2.7547638773819387,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004527300234161428,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.15855898410081865,
|
|
"num_tokens": 6060491.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.562553119659424,
|
|
"epoch": 2.75890637945319,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004525338855777591,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.16163053959608079,
|
|
"num_tokens": 6069770.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.692863273620605,
|
|
"epoch": 2.7630488815244405,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045233738963412746,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.16060031652450563,
|
|
"num_tokens": 6079219.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.6884935855865475,
|
|
"epoch": 2.7671913835956916,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045214053598160173,
|
|
"loss": 5.5913,
|
|
"mean_token_accuracy": 0.15756666362285615,
|
|
"num_tokens": 6088483.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.736902332305908,
|
|
"epoch": 2.7713338856669427,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045194332501725725,
|
|
"loss": 5.6121,
|
|
"mean_token_accuracy": 0.15517180860042573,
|
|
"num_tokens": 6098283.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.713816738128662,
|
|
"epoch": 2.775476387738194,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045174575713889014,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.16259121149778366,
|
|
"num_tokens": 6107657.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.611076307296753,
|
|
"epoch": 2.779618889809445,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004515478327450164,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.17341717928647996,
|
|
"num_tokens": 6116658.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.53763427734375,
|
|
"epoch": 2.783761391880696,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045134955223487093,
|
|
"loss": 5.4839,
|
|
"mean_token_accuracy": 0.16012571901082992,
|
|
"num_tokens": 6125091.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.626507520675659,
|
|
"epoch": 2.787903893951947,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045115091600840757,
|
|
"loss": 5.4702,
|
|
"mean_token_accuracy": 0.16941866427659988,
|
|
"num_tokens": 6133710.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.742727422714234,
|
|
"epoch": 2.7920463960231983,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00045095192446629697,
|
|
"loss": 5.5804,
|
|
"mean_token_accuracy": 0.15803021043539048,
|
|
"num_tokens": 6141986.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.583254098892212,
|
|
"epoch": 2.796188898094449,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000450752578009927,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.16595451682806014,
|
|
"num_tokens": 6150542.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.773436498641968,
|
|
"epoch": 2.8003314001657,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045055287704140115,
|
|
"loss": 5.5513,
|
|
"mean_token_accuracy": 0.1664824068546295,
|
|
"num_tokens": 6158958.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.670917129516601,
|
|
"epoch": 2.804473902236951,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004503528219635381,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.15057535320520402,
|
|
"num_tokens": 6168699.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.750991725921631,
|
|
"epoch": 2.8086164043082023,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004501524131798709,
|
|
"loss": 5.6256,
|
|
"mean_token_accuracy": 0.15223808214068413,
|
|
"num_tokens": 6177472.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.678924322128296,
|
|
"epoch": 2.812758906379453,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000449951651094646,
|
|
"loss": 5.6074,
|
|
"mean_token_accuracy": 0.15856225937604904,
|
|
"num_tokens": 6187098.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.813678789138794,
|
|
"epoch": 2.816901408450704,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044975053611282217,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.15290533751249313,
|
|
"num_tokens": 6196702.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.672289848327637,
|
|
"epoch": 2.821043910521955,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004495490686400706,
|
|
"loss": 5.5655,
|
|
"mean_token_accuracy": 0.16363370269536973,
|
|
"num_tokens": 6205465.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.612434053421021,
|
|
"epoch": 2.8251864125932062,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044934724908277313,
|
|
"loss": 5.5057,
|
|
"mean_token_accuracy": 0.16250835806131364,
|
|
"num_tokens": 6214967.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.5932745933532715,
|
|
"epoch": 2.8293289146644574,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044914507784802184,
|
|
"loss": 5.3888,
|
|
"mean_token_accuracy": 0.16869354397058486,
|
|
"num_tokens": 6223131.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.679493618011475,
|
|
"epoch": 2.8334714167357085,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004489425553436182,
|
|
"loss": 5.6129,
|
|
"mean_token_accuracy": 0.16268137246370315,
|
|
"num_tokens": 6232404.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.810343885421753,
|
|
"epoch": 2.8376139188069596,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044873968197807213,
|
|
"loss": 5.692,
|
|
"mean_token_accuracy": 0.15346367061138153,
|
|
"num_tokens": 6240666.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.702585172653198,
|
|
"epoch": 2.8417564208782107,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044853645816060156,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.1537062019109726,
|
|
"num_tokens": 6249375.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.691167497634888,
|
|
"epoch": 2.8458989229494613,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044833288430113104,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.16469355970621108,
|
|
"num_tokens": 6259215.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.666302585601807,
|
|
"epoch": 2.8500414250207124,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044812896081029136,
|
|
"loss": 5.5082,
|
|
"mean_token_accuracy": 0.15554405748844147,
|
|
"num_tokens": 6268650.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.616651582717895,
|
|
"epoch": 2.8541839270919636,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004479246880994183,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.1603178933262825,
|
|
"num_tokens": 6277687.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.725367212295533,
|
|
"epoch": 2.8583264291632147,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004477200665805525,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.17011294066905974,
|
|
"num_tokens": 6286940.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.538966131210327,
|
|
"epoch": 2.8624689312344658,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044751509666643777,
|
|
"loss": 5.4376,
|
|
"mean_token_accuracy": 0.16481457650661469,
|
|
"num_tokens": 6295594.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.642479085922242,
|
|
"epoch": 2.8666114333057164,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044730977877052084,
|
|
"loss": 5.5841,
|
|
"mean_token_accuracy": 0.15637488067150115,
|
|
"num_tokens": 6304616.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.710377931594849,
|
|
"epoch": 2.8707539353769675,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044710411330695045,
|
|
"loss": 5.4916,
|
|
"mean_token_accuracy": 0.16140547692775725,
|
|
"num_tokens": 6314449.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.589468860626221,
|
|
"epoch": 2.8748964374482187,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044689810069057615,
|
|
"loss": 5.4644,
|
|
"mean_token_accuracy": 0.1623084232211113,
|
|
"num_tokens": 6323811.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.6096266746521,
|
|
"epoch": 2.8790389395194698,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004466917413369479,
|
|
"loss": 5.4557,
|
|
"mean_token_accuracy": 0.1630224645137787,
|
|
"num_tokens": 6332553.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.631894493103028,
|
|
"epoch": 2.883181441590721,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004464850356623152,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.1654105991125107,
|
|
"num_tokens": 6341934.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.685671377182007,
|
|
"epoch": 2.887323943661972,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044627798408362597,
|
|
"loss": 5.5859,
|
|
"mean_token_accuracy": 0.16682443618774415,
|
|
"num_tokens": 6351032.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.674548244476318,
|
|
"epoch": 2.891466445733223,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004460705870185258,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.15875868052244185,
|
|
"num_tokens": 6360801.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.6504271030426025,
|
|
"epoch": 2.8956089478044738,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004458628448853572,
|
|
"loss": 5.5801,
|
|
"mean_token_accuracy": 0.1501157611608505,
|
|
"num_tokens": 6370359.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.630797243118286,
|
|
"epoch": 2.899751449875725,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004456547581031589,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.16001514345407486,
|
|
"num_tokens": 6379007.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.899751449875725,
|
|
"eval_entropy": 5.561281529570947,
|
|
"eval_loss": 5.929627895355225,
|
|
"eval_mean_token_accuracy": 0.14971610989524867,
|
|
"eval_num_tokens": 6379007.0,
|
|
"eval_runtime": 3.1644,
|
|
"eval_samples_per_second": 1184.11,
|
|
"eval_steps_per_second": 148.211,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12060,
|
|
"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": 9512917697664000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|