4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.6570008285004143,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 5425220239488000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|