24299 lines
665 KiB
JSON
24299 lines
665 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.942004971002486,
|
|
"eval_steps": 500,
|
|
"global_step": 12000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691956996917725,
|
|
"epoch": 0.004142502071251036,
|
|
"grad_norm": 12.6875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.786,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8553.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69189453125,
|
|
"epoch": 0.008285004142502071,
|
|
"grad_norm": 13.0625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7413,
|
|
"mean_token_accuracy": 0.00027811190811917187,
|
|
"num_tokens": 18133.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691031455993652,
|
|
"epoch": 0.012427506213753107,
|
|
"grad_norm": 9.5625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4928,
|
|
"mean_token_accuracy": 0.023286720411852,
|
|
"num_tokens": 27152.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.681630420684815,
|
|
"epoch": 0.016570008285004142,
|
|
"grad_norm": 6.40625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2019,
|
|
"mean_token_accuracy": 0.042532872036099435,
|
|
"num_tokens": 35844.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.63299388885498,
|
|
"epoch": 0.020712510356255178,
|
|
"grad_norm": 4.46875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8879,
|
|
"mean_token_accuracy": 0.04524078518152237,
|
|
"num_tokens": 44730.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.594239139556885,
|
|
"epoch": 0.024855012427506214,
|
|
"grad_norm": 3.6875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7944,
|
|
"mean_token_accuracy": 0.04270199164748192,
|
|
"num_tokens": 53751.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.599893474578858,
|
|
"epoch": 0.02899751449875725,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6813,
|
|
"mean_token_accuracy": 0.0431942991912365,
|
|
"num_tokens": 62399.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.593089294433593,
|
|
"epoch": 0.033140016570008285,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.675,
|
|
"mean_token_accuracy": 0.045351412147283554,
|
|
"num_tokens": 72384.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.579291725158692,
|
|
"epoch": 0.037282518641259324,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.625,
|
|
"mean_token_accuracy": 0.03919882737100124,
|
|
"num_tokens": 81720.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.570260047912598,
|
|
"epoch": 0.041425020712510356,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5545,
|
|
"mean_token_accuracy": 0.045346799120306966,
|
|
"num_tokens": 91634.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.559380340576173,
|
|
"epoch": 0.045567522783761395,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4963,
|
|
"mean_token_accuracy": 0.048221738263964656,
|
|
"num_tokens": 100359.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.541158962249757,
|
|
"epoch": 0.04971002485501243,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4361,
|
|
"mean_token_accuracy": 0.05027815327048302,
|
|
"num_tokens": 108678.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.408777332305908,
|
|
"epoch": 0.053852526926263466,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3154,
|
|
"mean_token_accuracy": 0.055302799120545386,
|
|
"num_tokens": 117022.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.43354835510254,
|
|
"epoch": 0.0579950289975145,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2758,
|
|
"mean_token_accuracy": 0.05044409707188606,
|
|
"num_tokens": 126489.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.479030323028564,
|
|
"epoch": 0.06213753106876554,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1042,
|
|
"mean_token_accuracy": 0.060464993864297864,
|
|
"num_tokens": 135926.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.352683162689209,
|
|
"epoch": 0.06628003314001657,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0851,
|
|
"mean_token_accuracy": 0.05333640910685063,
|
|
"num_tokens": 144488.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.436153030395507,
|
|
"epoch": 0.07042253521126761,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9888,
|
|
"mean_token_accuracy": 0.05590510331094265,
|
|
"num_tokens": 154180.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.357822036743164,
|
|
"epoch": 0.07456503728251865,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9305,
|
|
"mean_token_accuracy": 0.05672054700553417,
|
|
"num_tokens": 163448.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.390939235687256,
|
|
"epoch": 0.07870753935376967,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8567,
|
|
"mean_token_accuracy": 0.054386957362294194,
|
|
"num_tokens": 173318.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.244186305999756,
|
|
"epoch": 0.08285004142502071,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.745,
|
|
"mean_token_accuracy": 0.05825656205415726,
|
|
"num_tokens": 181540.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.197363090515136,
|
|
"epoch": 0.08699254349627175,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5793,
|
|
"mean_token_accuracy": 0.05952193439006805,
|
|
"num_tokens": 190932.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.11266918182373,
|
|
"epoch": 0.09113504556752279,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4206,
|
|
"mean_token_accuracy": 0.0623247817158699,
|
|
"num_tokens": 199793.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.071755409240723,
|
|
"epoch": 0.09527754763877382,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.4016,
|
|
"mean_token_accuracy": 0.056386180222034454,
|
|
"num_tokens": 208865.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.93977689743042,
|
|
"epoch": 0.09942004971002485,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2482,
|
|
"mean_token_accuracy": 0.06587636284530163,
|
|
"num_tokens": 218987.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.830070209503173,
|
|
"epoch": 0.1035625517812759,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1893,
|
|
"mean_token_accuracy": 0.06974356137216091,
|
|
"num_tokens": 228815.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.623151206970215,
|
|
"epoch": 0.10770505385252693,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.0572,
|
|
"mean_token_accuracy": 0.06595751233398914,
|
|
"num_tokens": 237752.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.47017993927002,
|
|
"epoch": 0.11184755592377796,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9063,
|
|
"mean_token_accuracy": 0.07446046024560929,
|
|
"num_tokens": 246512.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.312911701202392,
|
|
"epoch": 0.115990057995029,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8915,
|
|
"mean_token_accuracy": 0.0656554326415062,
|
|
"num_tokens": 255300.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.113032817840576,
|
|
"epoch": 0.12013256006628004,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7471,
|
|
"mean_token_accuracy": 0.06687054596841335,
|
|
"num_tokens": 265125.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.8920316696167,
|
|
"epoch": 0.12427506213753108,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6598,
|
|
"mean_token_accuracy": 0.06876451000571251,
|
|
"num_tokens": 274516.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.746868896484376,
|
|
"epoch": 0.12841756420878211,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.6736,
|
|
"mean_token_accuracy": 0.07511950880289078,
|
|
"num_tokens": 282994.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.574301624298096,
|
|
"epoch": 0.13256006628003314,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.5639,
|
|
"mean_token_accuracy": 0.07003462612628937,
|
|
"num_tokens": 291703.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.47190055847168,
|
|
"epoch": 0.13670256835128416,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.4587,
|
|
"mean_token_accuracy": 0.0722161963582039,
|
|
"num_tokens": 301596.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.276020717620849,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5039,
|
|
"mean_token_accuracy": 0.075354777649045,
|
|
"num_tokens": 309678.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.170723915100098,
|
|
"epoch": 0.14498757249378624,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4176,
|
|
"mean_token_accuracy": 0.07923398241400718,
|
|
"num_tokens": 319046.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.180229568481446,
|
|
"epoch": 0.1491300745650373,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.07866172268986701,
|
|
"num_tokens": 328004.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.078632307052612,
|
|
"epoch": 0.15327257663628832,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3784,
|
|
"mean_token_accuracy": 0.0761629119515419,
|
|
"num_tokens": 337455.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.00011487007141,
|
|
"epoch": 0.15741507870753935,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3504,
|
|
"mean_token_accuracy": 0.07819610238075256,
|
|
"num_tokens": 345631.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.865444993972778,
|
|
"epoch": 0.1615575807787904,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2621,
|
|
"mean_token_accuracy": 0.07181993611156941,
|
|
"num_tokens": 354790.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.92261233329773,
|
|
"epoch": 0.16570008285004142,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.374,
|
|
"mean_token_accuracy": 0.08448374196887017,
|
|
"num_tokens": 364480.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.90945405960083,
|
|
"epoch": 0.16984258492129245,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4125,
|
|
"mean_token_accuracy": 0.07339793667197228,
|
|
"num_tokens": 372751.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.968420600891113,
|
|
"epoch": 0.1739850869925435,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3969,
|
|
"mean_token_accuracy": 0.07574166990816593,
|
|
"num_tokens": 381140.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.890913581848144,
|
|
"epoch": 0.17812758906379453,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3887,
|
|
"mean_token_accuracy": 0.07703622430562973,
|
|
"num_tokens": 391415.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.813537740707398,
|
|
"epoch": 0.18227009113504558,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3673,
|
|
"mean_token_accuracy": 0.07801026552915573,
|
|
"num_tokens": 400174.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.883461046218872,
|
|
"epoch": 0.1864125932062966,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2726,
|
|
"mean_token_accuracy": 0.08017019852995873,
|
|
"num_tokens": 409355.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.8811296939849855,
|
|
"epoch": 0.19055509527754763,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2959,
|
|
"mean_token_accuracy": 0.077110555768013,
|
|
"num_tokens": 418336.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.754353618621826,
|
|
"epoch": 0.19469759734879868,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.292,
|
|
"mean_token_accuracy": 0.0773789893835783,
|
|
"num_tokens": 427771.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.748405838012696,
|
|
"epoch": 0.1988400994200497,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3225,
|
|
"mean_token_accuracy": 0.07635865435004234,
|
|
"num_tokens": 435576.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.725781488418579,
|
|
"epoch": 0.20298260149130073,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2366,
|
|
"mean_token_accuracy": 0.07985471300780773,
|
|
"num_tokens": 444360.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.684485578536988,
|
|
"epoch": 0.2071251035625518,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1657,
|
|
"mean_token_accuracy": 0.08312897011637688,
|
|
"num_tokens": 452946.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.705522155761718,
|
|
"epoch": 0.2112676056338028,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1751,
|
|
"mean_token_accuracy": 0.08343161344528198,
|
|
"num_tokens": 462105.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.676226043701172,
|
|
"epoch": 0.21541010770505387,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3555,
|
|
"mean_token_accuracy": 0.07977722287178039,
|
|
"num_tokens": 472397.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.740552568435669,
|
|
"epoch": 0.2195526097763049,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2001,
|
|
"mean_token_accuracy": 0.078244648873806,
|
|
"num_tokens": 481595.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.51032190322876,
|
|
"epoch": 0.22369511184755592,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0929,
|
|
"mean_token_accuracy": 0.08391840234398842,
|
|
"num_tokens": 490034.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.6380149841308596,
|
|
"epoch": 0.22783761391880697,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2089,
|
|
"mean_token_accuracy": 0.08496099263429642,
|
|
"num_tokens": 498521.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.588693332672119,
|
|
"epoch": 0.231980115990058,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1605,
|
|
"mean_token_accuracy": 0.07962968274950981,
|
|
"num_tokens": 507251.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.5256298065185545,
|
|
"epoch": 0.23612261806130902,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0795,
|
|
"mean_token_accuracy": 0.08021439425647259,
|
|
"num_tokens": 516507.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.683542346954345,
|
|
"epoch": 0.24026512013256007,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1987,
|
|
"mean_token_accuracy": 0.07887389063835144,
|
|
"num_tokens": 526322.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.589657783508301,
|
|
"epoch": 0.2444076222038111,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0698,
|
|
"mean_token_accuracy": 0.0887388363480568,
|
|
"num_tokens": 534762.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.573185110092163,
|
|
"epoch": 0.24855012427506215,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.128,
|
|
"mean_token_accuracy": 0.08384166359901428,
|
|
"num_tokens": 544279.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.464304256439209,
|
|
"epoch": 0.2526926263463132,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1072,
|
|
"mean_token_accuracy": 0.08207703456282615,
|
|
"num_tokens": 553178.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.532624387741089,
|
|
"epoch": 0.25683512841756423,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0674,
|
|
"mean_token_accuracy": 0.08325572535395623,
|
|
"num_tokens": 562383.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.475009918212891,
|
|
"epoch": 0.2609776304888152,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0478,
|
|
"mean_token_accuracy": 0.08197091519832611,
|
|
"num_tokens": 571104.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.402563810348511,
|
|
"epoch": 0.2651201325600663,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9822,
|
|
"mean_token_accuracy": 0.08756032511591912,
|
|
"num_tokens": 579890.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.563023519515991,
|
|
"epoch": 0.26926263463131733,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1609,
|
|
"mean_token_accuracy": 0.08216102197766303,
|
|
"num_tokens": 589033.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.385844802856445,
|
|
"epoch": 0.27340513670256833,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.031,
|
|
"mean_token_accuracy": 0.08240275979042053,
|
|
"num_tokens": 598040.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.4914178371429445,
|
|
"epoch": 0.2775476387738194,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1792,
|
|
"mean_token_accuracy": 0.08133822605013848,
|
|
"num_tokens": 607618.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.360255098342895,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9847,
|
|
"mean_token_accuracy": 0.08660589605569839,
|
|
"num_tokens": 616107.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.469431829452515,
|
|
"epoch": 0.28583264291632143,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0424,
|
|
"mean_token_accuracy": 0.08697322458028793,
|
|
"num_tokens": 625954.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.43549690246582,
|
|
"epoch": 0.2899751449875725,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1473,
|
|
"mean_token_accuracy": 0.07502747476100921,
|
|
"num_tokens": 635840.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.531662464141846,
|
|
"epoch": 0.29411764705882354,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0981,
|
|
"mean_token_accuracy": 0.08479553908109665,
|
|
"num_tokens": 644597.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.403447055816651,
|
|
"epoch": 0.2982601491300746,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.019,
|
|
"mean_token_accuracy": 0.0871436946094036,
|
|
"num_tokens": 653454.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.331816291809082,
|
|
"epoch": 0.3024026512013256,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0644,
|
|
"mean_token_accuracy": 0.08571388535201549,
|
|
"num_tokens": 662524.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.350220823287964,
|
|
"epoch": 0.30654515327257664,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.037,
|
|
"mean_token_accuracy": 0.08441569060087203,
|
|
"num_tokens": 672741.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.462545442581177,
|
|
"epoch": 0.3106876553438277,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0014,
|
|
"mean_token_accuracy": 0.08310677483677864,
|
|
"num_tokens": 682068.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.483937978744507,
|
|
"epoch": 0.3148301574150787,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.068,
|
|
"mean_token_accuracy": 0.08450363203883171,
|
|
"num_tokens": 691506.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.211332607269287,
|
|
"epoch": 0.31897265948632975,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9459,
|
|
"mean_token_accuracy": 0.07925326749682426,
|
|
"num_tokens": 702355.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.426611804962159,
|
|
"epoch": 0.3231151615575808,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0277,
|
|
"mean_token_accuracy": 0.08726275414228439,
|
|
"num_tokens": 710878.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.349923038482666,
|
|
"epoch": 0.3272576636288318,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0066,
|
|
"mean_token_accuracy": 0.0816049836575985,
|
|
"num_tokens": 719389.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.291773509979248,
|
|
"epoch": 0.33140016570008285,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9038,
|
|
"mean_token_accuracy": 0.08625322580337524,
|
|
"num_tokens": 728109.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.3569948196411135,
|
|
"epoch": 0.3355426677713339,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0035,
|
|
"mean_token_accuracy": 0.08949297964572907,
|
|
"num_tokens": 737617.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.196184492111206,
|
|
"epoch": 0.3396851698425849,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8985,
|
|
"mean_token_accuracy": 0.086442032456398,
|
|
"num_tokens": 746368.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.21246075630188,
|
|
"epoch": 0.34382767191383595,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8542,
|
|
"mean_token_accuracy": 0.09095343127846718,
|
|
"num_tokens": 755561.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.358605766296387,
|
|
"epoch": 0.347970173985087,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9435,
|
|
"mean_token_accuracy": 0.08809717670083046,
|
|
"num_tokens": 764501.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.229580879211426,
|
|
"epoch": 0.352112676056338,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.085,
|
|
"mean_token_accuracy": 0.08461654633283615,
|
|
"num_tokens": 774003.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.327147531509399,
|
|
"epoch": 0.35625517812758906,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.09031773805618286,
|
|
"num_tokens": 782319.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.148540925979614,
|
|
"epoch": 0.3603976801988401,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8587,
|
|
"mean_token_accuracy": 0.08761301413178443,
|
|
"num_tokens": 791467.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.317266750335693,
|
|
"epoch": 0.36454018227009116,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9807,
|
|
"mean_token_accuracy": 0.08628286868333816,
|
|
"num_tokens": 800328.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.229923439025879,
|
|
"epoch": 0.36868268434134216,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9734,
|
|
"mean_token_accuracy": 0.08270182609558105,
|
|
"num_tokens": 808849.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.28882212638855,
|
|
"epoch": 0.3728251864125932,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.09033941775560379,
|
|
"num_tokens": 819090.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.23458833694458,
|
|
"epoch": 0.37696768848384427,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9226,
|
|
"mean_token_accuracy": 0.08801109045743942,
|
|
"num_tokens": 828527.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.252940464019775,
|
|
"epoch": 0.38111019055509526,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9447,
|
|
"mean_token_accuracy": 0.08754967637360096,
|
|
"num_tokens": 837163.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.2431192874908445,
|
|
"epoch": 0.3852526926263463,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9504,
|
|
"mean_token_accuracy": 0.0881062887609005,
|
|
"num_tokens": 846261.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.118913030624389,
|
|
"epoch": 0.38939519469759737,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8595,
|
|
"mean_token_accuracy": 0.08703678250312805,
|
|
"num_tokens": 855655.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.264123439788818,
|
|
"epoch": 0.39353769676884837,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.011,
|
|
"mean_token_accuracy": 0.08637121096253395,
|
|
"num_tokens": 864980.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.17798171043396,
|
|
"epoch": 0.3976801988400994,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8449,
|
|
"mean_token_accuracy": 0.08888368755578994,
|
|
"num_tokens": 874625.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.17276029586792,
|
|
"epoch": 0.40182270091135047,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8597,
|
|
"mean_token_accuracy": 0.08416522145271302,
|
|
"num_tokens": 884104.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.224177455902099,
|
|
"epoch": 0.40596520298260147,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9764,
|
|
"mean_token_accuracy": 0.08516203612089157,
|
|
"num_tokens": 893355.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.136541414260864,
|
|
"epoch": 0.4101077050538525,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.08847377672791482,
|
|
"num_tokens": 901666.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.1944825649261475,
|
|
"epoch": 0.4142502071251036,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8392,
|
|
"mean_token_accuracy": 0.08776508644223213,
|
|
"num_tokens": 911038.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4142502071251036,
|
|
"eval_entropy": 7.029010676117594,
|
|
"eval_loss": 6.951485633850098,
|
|
"eval_mean_token_accuracy": 0.08968717416625287,
|
|
"eval_num_tokens": 911038.0,
|
|
"eval_runtime": 2.9087,
|
|
"eval_samples_per_second": 1288.215,
|
|
"eval_steps_per_second": 161.242,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.090590286254883,
|
|
"epoch": 0.4183927091963546,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7688,
|
|
"mean_token_accuracy": 0.09368073344230651,
|
|
"num_tokens": 920505.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.014606046676636,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7777,
|
|
"mean_token_accuracy": 0.08937770649790763,
|
|
"num_tokens": 929983.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.062945079803467,
|
|
"epoch": 0.4266777133388567,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7857,
|
|
"mean_token_accuracy": 0.08497672230005264,
|
|
"num_tokens": 938920.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.22315354347229,
|
|
"epoch": 0.43082021541010773,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.10367532446980476,
|
|
"num_tokens": 948067.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.985883140563965,
|
|
"epoch": 0.43496271748135873,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8608,
|
|
"mean_token_accuracy": 0.09106278494000435,
|
|
"num_tokens": 957112.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.1854969501495365,
|
|
"epoch": 0.4391052195526098,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8318,
|
|
"mean_token_accuracy": 0.08919779881834984,
|
|
"num_tokens": 966112.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.923943567276001,
|
|
"epoch": 0.44324772162386084,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6632,
|
|
"mean_token_accuracy": 0.09753508195281028,
|
|
"num_tokens": 974791.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.010893726348877,
|
|
"epoch": 0.44739022369511183,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7551,
|
|
"mean_token_accuracy": 0.09772919043898583,
|
|
"num_tokens": 982730.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.060620880126953,
|
|
"epoch": 0.4515327257663629,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7707,
|
|
"mean_token_accuracy": 0.09096172973513603,
|
|
"num_tokens": 992275.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.988409900665284,
|
|
"epoch": 0.45567522783761394,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.09266227409243584,
|
|
"num_tokens": 1001131.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.050741291046142,
|
|
"epoch": 0.45981772990886494,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8893,
|
|
"mean_token_accuracy": 0.08530252054333687,
|
|
"num_tokens": 1010712.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.086013460159302,
|
|
"epoch": 0.463960231980116,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8307,
|
|
"mean_token_accuracy": 0.08732764646410943,
|
|
"num_tokens": 1019726.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.983129787445068,
|
|
"epoch": 0.46810273405136704,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.08867142722010612,
|
|
"num_tokens": 1028838.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.02021803855896,
|
|
"epoch": 0.47224523612261804,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.819,
|
|
"mean_token_accuracy": 0.09228519797325134,
|
|
"num_tokens": 1038839.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.049885368347168,
|
|
"epoch": 0.4763877381938691,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7751,
|
|
"mean_token_accuracy": 0.09855888932943344,
|
|
"num_tokens": 1048201.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.997226238250732,
|
|
"epoch": 0.48053024026512015,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.0835553340613842,
|
|
"num_tokens": 1057636.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.117639541625977,
|
|
"epoch": 0.48467274233637114,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8948,
|
|
"mean_token_accuracy": 0.09270423501729966,
|
|
"num_tokens": 1067776.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.967460298538208,
|
|
"epoch": 0.4888152444076222,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7117,
|
|
"mean_token_accuracy": 0.10028473436832427,
|
|
"num_tokens": 1076436.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.895012378692627,
|
|
"epoch": 0.49295774647887325,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.09727847725152969,
|
|
"num_tokens": 1085907.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.11523928642273,
|
|
"epoch": 0.4971002485501243,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8765,
|
|
"mean_token_accuracy": 0.09036477208137512,
|
|
"num_tokens": 1095617.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.85316572189331,
|
|
"epoch": 0.5012427506213754,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6596,
|
|
"mean_token_accuracy": 0.09437181055545807,
|
|
"num_tokens": 1104463.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.884859657287597,
|
|
"epoch": 0.5053852526926264,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8366,
|
|
"mean_token_accuracy": 0.09331914708018303,
|
|
"num_tokens": 1114594.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.040772867202759,
|
|
"epoch": 0.5095277547638773,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7153,
|
|
"mean_token_accuracy": 0.09341486915946007,
|
|
"num_tokens": 1124413.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.817426300048828,
|
|
"epoch": 0.5136702568351285,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7013,
|
|
"mean_token_accuracy": 0.09107679948210716,
|
|
"num_tokens": 1134392.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.028265714645386,
|
|
"epoch": 0.5178127589063795,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.08923142552375793,
|
|
"num_tokens": 1142563.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.868456792831421,
|
|
"epoch": 0.5219552609776305,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.859,
|
|
"mean_token_accuracy": 0.09302715957164764,
|
|
"num_tokens": 1151325.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.9985448837280275,
|
|
"epoch": 0.5260977630488816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7065,
|
|
"mean_token_accuracy": 0.09792971536517144,
|
|
"num_tokens": 1160817.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.997683715820313,
|
|
"epoch": 0.5302402651201326,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7195,
|
|
"mean_token_accuracy": 0.09700844064354897,
|
|
"num_tokens": 1169913.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.938048315048218,
|
|
"epoch": 0.5343827671913836,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.8268,
|
|
"mean_token_accuracy": 0.09028196930885315,
|
|
"num_tokens": 1179167.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.906701946258545,
|
|
"epoch": 0.5385252692626347,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7851,
|
|
"mean_token_accuracy": 0.09439637437462807,
|
|
"num_tokens": 1189105.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.777953195571899,
|
|
"epoch": 0.5426677713338857,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6017,
|
|
"mean_token_accuracy": 0.09869116172194481,
|
|
"num_tokens": 1197487.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.929132318496704,
|
|
"epoch": 0.5468102734051367,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7028,
|
|
"mean_token_accuracy": 0.09371067583560944,
|
|
"num_tokens": 1206914.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.92956395149231,
|
|
"epoch": 0.5509527754763878,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7621,
|
|
"mean_token_accuracy": 0.1005061112344265,
|
|
"num_tokens": 1214991.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.893949174880982,
|
|
"epoch": 0.5550952775476388,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7168,
|
|
"mean_token_accuracy": 0.09989271685481071,
|
|
"num_tokens": 1223567.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.962380599975586,
|
|
"epoch": 0.5592377796188898,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6499,
|
|
"mean_token_accuracy": 0.0953209526836872,
|
|
"num_tokens": 1232625.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.936891651153564,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7674,
|
|
"mean_token_accuracy": 0.09183648079633713,
|
|
"num_tokens": 1242947.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.9282430648803714,
|
|
"epoch": 0.5675227837613919,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.83,
|
|
"mean_token_accuracy": 0.09135605543851852,
|
|
"num_tokens": 1252231.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.85473780632019,
|
|
"epoch": 0.5716652858326429,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6957,
|
|
"mean_token_accuracy": 0.09633494764566422,
|
|
"num_tokens": 1262287.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.846092748641968,
|
|
"epoch": 0.575807787903894,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5806,
|
|
"mean_token_accuracy": 0.10334836915135384,
|
|
"num_tokens": 1271351.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.770713424682617,
|
|
"epoch": 0.579950289975145,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5851,
|
|
"mean_token_accuracy": 0.09925638064742089,
|
|
"num_tokens": 1280272.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.722933340072632,
|
|
"epoch": 0.584092792046396,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6445,
|
|
"mean_token_accuracy": 0.09788036495447158,
|
|
"num_tokens": 1290040.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.850803089141846,
|
|
"epoch": 0.5882352941176471,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6608,
|
|
"mean_token_accuracy": 0.10229592770338058,
|
|
"num_tokens": 1298814.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.843781042098999,
|
|
"epoch": 0.5923777961888981,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6779,
|
|
"mean_token_accuracy": 0.09635392874479294,
|
|
"num_tokens": 1308426.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.907309722900391,
|
|
"epoch": 0.5965202982601492,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.7811,
|
|
"mean_token_accuracy": 0.09799076840281487,
|
|
"num_tokens": 1316642.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.865832138061523,
|
|
"epoch": 0.6006628003314002,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6458,
|
|
"mean_token_accuracy": 0.10095788836479187,
|
|
"num_tokens": 1325023.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.719896602630615,
|
|
"epoch": 0.6048053024026512,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5506,
|
|
"mean_token_accuracy": 0.10248812884092331,
|
|
"num_tokens": 1333506.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.8264491081237795,
|
|
"epoch": 0.6089478044739023,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6554,
|
|
"mean_token_accuracy": 0.10372467041015625,
|
|
"num_tokens": 1342283.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.7825822830200195,
|
|
"epoch": 0.6130903065451533,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5995,
|
|
"mean_token_accuracy": 0.09992679581046104,
|
|
"num_tokens": 1350570.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.820115232467652,
|
|
"epoch": 0.6172328086164043,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6518,
|
|
"mean_token_accuracy": 0.10142135620117188,
|
|
"num_tokens": 1359390.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.711935329437256,
|
|
"epoch": 0.6213753106876554,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7036,
|
|
"mean_token_accuracy": 0.09649694561958314,
|
|
"num_tokens": 1368179.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.707304859161377,
|
|
"epoch": 0.6255178127589064,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.585,
|
|
"mean_token_accuracy": 0.10146094933152199,
|
|
"num_tokens": 1376739.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.74249963760376,
|
|
"epoch": 0.6296603148301574,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5747,
|
|
"mean_token_accuracy": 0.10194365158677102,
|
|
"num_tokens": 1385654.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.793521308898926,
|
|
"epoch": 0.6338028169014085,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6333,
|
|
"mean_token_accuracy": 0.09613201916217803,
|
|
"num_tokens": 1395811.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.661139059066772,
|
|
"epoch": 0.6379453189726595,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5734,
|
|
"mean_token_accuracy": 0.09805987849831581,
|
|
"num_tokens": 1404985.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.739059400558472,
|
|
"epoch": 0.6420878210439105,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5739,
|
|
"mean_token_accuracy": 0.10577682480216026,
|
|
"num_tokens": 1413306.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.735513067245483,
|
|
"epoch": 0.6462303231151616,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6181,
|
|
"mean_token_accuracy": 0.1049142599105835,
|
|
"num_tokens": 1422572.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.604087400436401,
|
|
"epoch": 0.6503728251864126,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5312,
|
|
"mean_token_accuracy": 0.1017876386642456,
|
|
"num_tokens": 1432743.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.643766403198242,
|
|
"epoch": 0.6545153272576636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5453,
|
|
"mean_token_accuracy": 0.10414253100752831,
|
|
"num_tokens": 1441459.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.806045198440552,
|
|
"epoch": 0.6586578293289147,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.7397,
|
|
"mean_token_accuracy": 0.09327913522720337,
|
|
"num_tokens": 1450900.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.7044349193573,
|
|
"epoch": 0.6628003314001657,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5546,
|
|
"mean_token_accuracy": 0.10626772269606591,
|
|
"num_tokens": 1459983.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.615601396560669,
|
|
"epoch": 0.6669428334714167,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.09967566579580307,
|
|
"num_tokens": 1469197.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.755218744277954,
|
|
"epoch": 0.6710853355426678,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5183,
|
|
"mean_token_accuracy": 0.11426351144909859,
|
|
"num_tokens": 1477422.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.645377492904663,
|
|
"epoch": 0.6752278376139188,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.6683,
|
|
"mean_token_accuracy": 0.09824564009904861,
|
|
"num_tokens": 1486813.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.705157470703125,
|
|
"epoch": 0.6793703396851698,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5939,
|
|
"mean_token_accuracy": 0.10478519424796104,
|
|
"num_tokens": 1495783.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.800464725494384,
|
|
"epoch": 0.6835128417564209,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6916,
|
|
"mean_token_accuracy": 0.10224536880850792,
|
|
"num_tokens": 1503926.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.6785046577453615,
|
|
"epoch": 0.6876553438276719,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5561,
|
|
"mean_token_accuracy": 0.09954166486859321,
|
|
"num_tokens": 1512714.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.569236993789673,
|
|
"epoch": 0.6917978458989229,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4272,
|
|
"mean_token_accuracy": 0.10806796178221703,
|
|
"num_tokens": 1521812.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.603865003585815,
|
|
"epoch": 0.695940347970174,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.6753,
|
|
"mean_token_accuracy": 0.09779459685087204,
|
|
"num_tokens": 1532083.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.691070890426635,
|
|
"epoch": 0.700082850041425,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.10249469205737113,
|
|
"num_tokens": 1541093.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.738386869430542,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5895,
|
|
"mean_token_accuracy": 0.1061245173215866,
|
|
"num_tokens": 1549942.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.638346433639526,
|
|
"epoch": 0.7083678541839271,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5236,
|
|
"mean_token_accuracy": 0.10389250293374061,
|
|
"num_tokens": 1559528.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.567893695831299,
|
|
"epoch": 0.7125103562551781,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4693,
|
|
"mean_token_accuracy": 0.10965005755424499,
|
|
"num_tokens": 1567590.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.621993827819824,
|
|
"epoch": 0.7166528583264291,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6154,
|
|
"mean_token_accuracy": 0.10054041370749474,
|
|
"num_tokens": 1576815.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.813084697723388,
|
|
"epoch": 0.7207953603976802,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6913,
|
|
"mean_token_accuracy": 0.09942748844623565,
|
|
"num_tokens": 1585240.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.702514028549194,
|
|
"epoch": 0.7249378624689312,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5847,
|
|
"mean_token_accuracy": 0.10101142525672913,
|
|
"num_tokens": 1595242.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.623669815063477,
|
|
"epoch": 0.7290803645401823,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5124,
|
|
"mean_token_accuracy": 0.10096012577414512,
|
|
"num_tokens": 1603753.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.524274682998657,
|
|
"epoch": 0.7332228666114333,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4964,
|
|
"mean_token_accuracy": 0.10853204876184464,
|
|
"num_tokens": 1613221.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.586112594604492,
|
|
"epoch": 0.7373653686826843,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.505,
|
|
"mean_token_accuracy": 0.10921112820506096,
|
|
"num_tokens": 1622881.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.622179698944092,
|
|
"epoch": 0.7415078707539354,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5134,
|
|
"mean_token_accuracy": 0.10829928070306778,
|
|
"num_tokens": 1631340.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.599711656570435,
|
|
"epoch": 0.7456503728251864,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5232,
|
|
"mean_token_accuracy": 0.10620987564325332,
|
|
"num_tokens": 1639920.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.625485038757324,
|
|
"epoch": 0.7497928748964374,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.551,
|
|
"mean_token_accuracy": 0.10739724114537239,
|
|
"num_tokens": 1649077.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.574350118637085,
|
|
"epoch": 0.7539353769676885,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.1003094732761383,
|
|
"num_tokens": 1659181.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.611196804046631,
|
|
"epoch": 0.7580778790389395,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4909,
|
|
"mean_token_accuracy": 0.11145598813891411,
|
|
"num_tokens": 1668111.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.5018958568573,
|
|
"epoch": 0.7622203811101905,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5254,
|
|
"mean_token_accuracy": 0.10480432286858558,
|
|
"num_tokens": 1677294.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.6240472316741945,
|
|
"epoch": 0.7663628831814416,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.10766990557312965,
|
|
"num_tokens": 1686187.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.5916759967803955,
|
|
"epoch": 0.7705053852526926,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5396,
|
|
"mean_token_accuracy": 0.1008406363427639,
|
|
"num_tokens": 1695999.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.52704086303711,
|
|
"epoch": 0.7746478873239436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4615,
|
|
"mean_token_accuracy": 0.1080300435423851,
|
|
"num_tokens": 1705292.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.622449254989624,
|
|
"epoch": 0.7787903893951947,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5108,
|
|
"mean_token_accuracy": 0.10377022996544838,
|
|
"num_tokens": 1714807.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.570343255996704,
|
|
"epoch": 0.7829328914664457,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.10502566173672676,
|
|
"num_tokens": 1723383.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.481582975387573,
|
|
"epoch": 0.7870753935376967,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4257,
|
|
"mean_token_accuracy": 0.10356146544218063,
|
|
"num_tokens": 1733151.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.4965410232543945,
|
|
"epoch": 0.7912178956089478,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4716,
|
|
"mean_token_accuracy": 0.11054345145821572,
|
|
"num_tokens": 1742292.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.6443828582763675,
|
|
"epoch": 0.7953603976801988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5643,
|
|
"mean_token_accuracy": 0.11139457151293755,
|
|
"num_tokens": 1751441.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.527015209197998,
|
|
"epoch": 0.7995028997514498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4417,
|
|
"mean_token_accuracy": 0.11440029144287109,
|
|
"num_tokens": 1759833.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.575316333770752,
|
|
"epoch": 0.8036454018227009,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.11134352311491966,
|
|
"num_tokens": 1769258.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.614769506454468,
|
|
"epoch": 0.8077879038939519,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.6136,
|
|
"mean_token_accuracy": 0.09684511870145798,
|
|
"num_tokens": 1779532.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.489955949783325,
|
|
"epoch": 0.8119304059652029,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4826,
|
|
"mean_token_accuracy": 0.10542305111885071,
|
|
"num_tokens": 1788936.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.524744033813477,
|
|
"epoch": 0.816072908036454,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4307,
|
|
"mean_token_accuracy": 0.11505212485790253,
|
|
"num_tokens": 1796486.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.5153892993927,
|
|
"epoch": 0.820215410107705,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4003,
|
|
"mean_token_accuracy": 0.1139255128800869,
|
|
"num_tokens": 1805349.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.511750984191894,
|
|
"epoch": 0.824357912178956,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.4933,
|
|
"mean_token_accuracy": 0.10885351151227951,
|
|
"num_tokens": 1813625.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.511787986755371,
|
|
"epoch": 0.8285004142502072,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.5196,
|
|
"mean_token_accuracy": 0.11211849302053452,
|
|
"num_tokens": 1823465.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8285004142502072,
|
|
"eval_entropy": 6.489772428581709,
|
|
"eval_loss": 6.5978169441223145,
|
|
"eval_mean_token_accuracy": 0.1059395734673497,
|
|
"eval_num_tokens": 1823465.0,
|
|
"eval_runtime": 2.8165,
|
|
"eval_samples_per_second": 1330.384,
|
|
"eval_steps_per_second": 166.52,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.503752756118774,
|
|
"epoch": 0.8326429163214581,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999998547681232,
|
|
"loss": 6.4594,
|
|
"mean_token_accuracy": 0.10898767039179802,
|
|
"num_tokens": 1832701.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.5025763511657715,
|
|
"epoch": 0.8367854183927091,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999992647639447,
|
|
"loss": 6.5037,
|
|
"mean_token_accuracy": 0.1069453164935112,
|
|
"num_tokens": 1841914.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.498456144332886,
|
|
"epoch": 0.8409279204639603,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999982209116618,
|
|
"loss": 6.5033,
|
|
"mean_token_accuracy": 0.1143178328871727,
|
|
"num_tokens": 1851906.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.527345752716064,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999967232133796,
|
|
"loss": 6.4405,
|
|
"mean_token_accuracy": 0.11167196556925774,
|
|
"num_tokens": 1860817.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.471268653869629,
|
|
"epoch": 0.8492129246064622,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999947716721196,
|
|
"loss": 6.4317,
|
|
"mean_token_accuracy": 0.11340907365083694,
|
|
"num_tokens": 1870387.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.53512659072876,
|
|
"epoch": 0.8533554266777134,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999923662918179,
|
|
"loss": 6.4923,
|
|
"mean_token_accuracy": 0.10849138200283051,
|
|
"num_tokens": 1878929.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.381455135345459,
|
|
"epoch": 0.8574979287489644,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999895070773266,
|
|
"loss": 6.5218,
|
|
"mean_token_accuracy": 0.10917675569653511,
|
|
"num_tokens": 1887633.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.555011653900147,
|
|
"epoch": 0.8616404308202155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999861940344129,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.11225893646478653,
|
|
"num_tokens": 1896729.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.461038255691529,
|
|
"epoch": 0.8657829328914665,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999824271697597,
|
|
"loss": 6.4744,
|
|
"mean_token_accuracy": 0.11994748562574387,
|
|
"num_tokens": 1905224.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.426373243331909,
|
|
"epoch": 0.8699254349627175,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499978206490965,
|
|
"loss": 6.4421,
|
|
"mean_token_accuracy": 0.1062840238213539,
|
|
"num_tokens": 1915502.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.507133817672729,
|
|
"epoch": 0.8740679370339686,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999735320065427,
|
|
"loss": 6.5166,
|
|
"mean_token_accuracy": 0.1090014934539795,
|
|
"num_tokens": 1923946.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.391321516036987,
|
|
"epoch": 0.8782104391052196,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999684037259216,
|
|
"loss": 6.3153,
|
|
"mean_token_accuracy": 0.11198925003409385,
|
|
"num_tokens": 1932524.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.523810911178589,
|
|
"epoch": 0.8823529411764706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999628216594456,
|
|
"loss": 6.4342,
|
|
"mean_token_accuracy": 0.11353398859500885,
|
|
"num_tokens": 1941604.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.369825983047486,
|
|
"epoch": 0.8864954432477217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499956785818375,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.11778979599475861,
|
|
"num_tokens": 1950235.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.44530668258667,
|
|
"epoch": 0.8906379453189727,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999502962148843,
|
|
"loss": 6.411,
|
|
"mean_token_accuracy": 0.11221367716789246,
|
|
"num_tokens": 1959597.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.611082124710083,
|
|
"epoch": 0.8947804473902237,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499943352862064,
|
|
"loss": 6.5525,
|
|
"mean_token_accuracy": 0.11493104994297028,
|
|
"num_tokens": 1969032.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.356397533416748,
|
|
"epoch": 0.8989229494614748,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999359557739194,
|
|
"loss": 6.4099,
|
|
"mean_token_accuracy": 0.11556253135204315,
|
|
"num_tokens": 1977691.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.429668664932251,
|
|
"epoch": 0.9030654515327258,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999281049653713,
|
|
"loss": 6.3185,
|
|
"mean_token_accuracy": 0.11952688023447991,
|
|
"num_tokens": 1986118.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.4375769138336185,
|
|
"epoch": 0.9072079536039768,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999198004522557,
|
|
"loss": 6.4391,
|
|
"mean_token_accuracy": 0.11205489486455918,
|
|
"num_tokens": 1995466.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.4425500392913815,
|
|
"epoch": 0.9113504556752279,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999110422513237,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.10964742228388787,
|
|
"num_tokens": 2005219.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.36616268157959,
|
|
"epoch": 0.9154929577464789,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999018303802414,
|
|
"loss": 6.3219,
|
|
"mean_token_accuracy": 0.11532410681247711,
|
|
"num_tokens": 2014078.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.571248149871826,
|
|
"epoch": 0.9196354598177299,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998921648575904,
|
|
"loss": 6.5299,
|
|
"mean_token_accuracy": 0.10419057235121727,
|
|
"num_tokens": 2023456.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.337164688110351,
|
|
"epoch": 0.923777961888981,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998820457028669,
|
|
"loss": 6.3187,
|
|
"mean_token_accuracy": 0.11291450634598732,
|
|
"num_tokens": 2032312.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.432801675796509,
|
|
"epoch": 0.927920463960232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998714729364824,
|
|
"loss": 6.2912,
|
|
"mean_token_accuracy": 0.11874229311943055,
|
|
"num_tokens": 2041580.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.351183366775513,
|
|
"epoch": 0.932062966031483,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998604465797634,
|
|
"loss": 6.5144,
|
|
"mean_token_accuracy": 0.10716686621308327,
|
|
"num_tokens": 2051126.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.538003158569336,
|
|
"epoch": 0.9362054681027341,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998489666549512,
|
|
"loss": 6.5034,
|
|
"mean_token_accuracy": 0.11126186326146126,
|
|
"num_tokens": 2059565.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.341565847396851,
|
|
"epoch": 0.9403479701739851,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004998370331852021,
|
|
"loss": 6.2731,
|
|
"mean_token_accuracy": 0.11285399943590164,
|
|
"num_tokens": 2068113.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.386385202407837,
|
|
"epoch": 0.9444904722452361,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998246461945872,
|
|
"loss": 6.4455,
|
|
"mean_token_accuracy": 0.11150851249694824,
|
|
"num_tokens": 2077227.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.483101797103882,
|
|
"epoch": 0.9486329743164872,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998118057080923,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.11290868446230888,
|
|
"num_tokens": 2086234.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.346535301208496,
|
|
"epoch": 0.9527754763877382,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997985117516183,
|
|
"loss": 6.3818,
|
|
"mean_token_accuracy": 0.11334575861692428,
|
|
"num_tokens": 2094945.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.368511581420899,
|
|
"epoch": 0.9569179784589892,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997847643519803,
|
|
"loss": 6.3666,
|
|
"mean_token_accuracy": 0.12083429247140884,
|
|
"num_tokens": 2104164.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.506369352340698,
|
|
"epoch": 0.9610604805302403,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997705635369085,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.10683630257844925,
|
|
"num_tokens": 2112629.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.375317573547363,
|
|
"epoch": 0.9652029826014913,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997559093350474,
|
|
"loss": 6.4314,
|
|
"mean_token_accuracy": 0.11005270928144455,
|
|
"num_tokens": 2122491.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.416046524047852,
|
|
"epoch": 0.9693454846727423,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997408017759562,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.11960792094469071,
|
|
"num_tokens": 2132115.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.315891075134277,
|
|
"epoch": 0.9734879867439934,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997252408901084,
|
|
"loss": 6.2895,
|
|
"mean_token_accuracy": 0.1188917763531208,
|
|
"num_tokens": 2141663.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.4620068073272705,
|
|
"epoch": 0.9776304888152444,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997092267088922,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.11433440446853638,
|
|
"num_tokens": 2152057.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.459884881973267,
|
|
"epoch": 0.9817729908864954,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996927592646097,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.11693302392959595,
|
|
"num_tokens": 2161554.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.297170400619507,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996758385904776,
|
|
"loss": 6.291,
|
|
"mean_token_accuracy": 0.11896034255623818,
|
|
"num_tokens": 2170553.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.433252429962158,
|
|
"epoch": 0.9900579950289975,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996584647206269,
|
|
"loss": 6.4227,
|
|
"mean_token_accuracy": 0.1149461954832077,
|
|
"num_tokens": 2179996.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.324653291702271,
|
|
"epoch": 0.9942004971002486,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996406376901026,
|
|
"loss": 6.2669,
|
|
"mean_token_accuracy": 0.12070421054959297,
|
|
"num_tokens": 2188882.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.464151000976562,
|
|
"epoch": 0.9983429991714996,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996223575348635,
|
|
"loss": 6.4539,
|
|
"mean_token_accuracy": 0.11858661994338035,
|
|
"num_tokens": 2197304.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.405771923065186,
|
|
"epoch": 1.0024855012427507,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996036242917828,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.12037697657942772,
|
|
"num_tokens": 2206144.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.425885581970215,
|
|
"epoch": 1.0066280033140016,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995844379986476,
|
|
"loss": 6.2114,
|
|
"mean_token_accuracy": 0.12305252626538277,
|
|
"num_tokens": 2215054.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.230459594726563,
|
|
"epoch": 1.0107705053852527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995647986941587,
|
|
"loss": 6.0583,
|
|
"mean_token_accuracy": 0.1284884586930275,
|
|
"num_tokens": 2223578.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.175695371627808,
|
|
"epoch": 1.0149130074565038,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995447064179308,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.1279567815363407,
|
|
"num_tokens": 2234115.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.283312606811523,
|
|
"epoch": 1.0190555095277547,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499524161210492,
|
|
"loss": 6.1152,
|
|
"mean_token_accuracy": 0.12857213839888573,
|
|
"num_tokens": 2243447.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.329511642456055,
|
|
"epoch": 1.0231980115990058,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995031631132844,
|
|
"loss": 6.103,
|
|
"mean_token_accuracy": 0.12532400861382484,
|
|
"num_tokens": 2252730.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.204122638702392,
|
|
"epoch": 1.027340513670257,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004994817121686633,
|
|
"loss": 6.0922,
|
|
"mean_token_accuracy": 0.1204651266336441,
|
|
"num_tokens": 2261809.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.253789472579956,
|
|
"epoch": 1.0314830157415078,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994598084198979,
|
|
"loss": 6.0332,
|
|
"mean_token_accuracy": 0.12827168405056,
|
|
"num_tokens": 2270583.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.270323991775513,
|
|
"epoch": 1.035625517812759,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004994374519111701,
|
|
"loss": 6.1004,
|
|
"mean_token_accuracy": 0.13118301704525948,
|
|
"num_tokens": 2279337.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.257144260406494,
|
|
"epoch": 1.03976801988401,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004994146426875757,
|
|
"loss": 6.0698,
|
|
"mean_token_accuracy": 0.12660571709275245,
|
|
"num_tokens": 2288679.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.186429357528686,
|
|
"epoch": 1.043910521955261,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993913807951232,
|
|
"loss": 6.0641,
|
|
"mean_token_accuracy": 0.12812110856175424,
|
|
"num_tokens": 2297780.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.25863995552063,
|
|
"epoch": 1.048053024026512,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993676662807345,
|
|
"loss": 6.1318,
|
|
"mean_token_accuracy": 0.12341021671891213,
|
|
"num_tokens": 2306424.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.240696907043457,
|
|
"epoch": 1.0521955260977631,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004993434991922443,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.12805933132767677,
|
|
"num_tokens": 2315472.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.19228229522705,
|
|
"epoch": 1.056338028169014,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993188795784002,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.1268403261899948,
|
|
"num_tokens": 2324381.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.129907703399658,
|
|
"epoch": 1.0604805302402651,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004992938074888628,
|
|
"loss": 6.0639,
|
|
"mean_token_accuracy": 0.12428224980831146,
|
|
"num_tokens": 2333629.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.27379355430603,
|
|
"epoch": 1.0646230323115162,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992682829742052,
|
|
"loss": 6.1316,
|
|
"mean_token_accuracy": 0.126251719892025,
|
|
"num_tokens": 2342787.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.154513502120972,
|
|
"epoch": 1.0687655343827671,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992423060859132,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.13422906622290612,
|
|
"num_tokens": 2351684.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.270391750335693,
|
|
"epoch": 1.0729080364540182,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004992158768763849,
|
|
"loss": 6.1692,
|
|
"mean_token_accuracy": 0.12574677392840386,
|
|
"num_tokens": 2361100.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.145001316070557,
|
|
"epoch": 1.0770505385252693,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991889953989308,
|
|
"loss": 6.1219,
|
|
"mean_token_accuracy": 0.12856860533356668,
|
|
"num_tokens": 2370139.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.286712026596069,
|
|
"epoch": 1.0811930405965202,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991616617077741,
|
|
"loss": 6.1773,
|
|
"mean_token_accuracy": 0.12485618069767952,
|
|
"num_tokens": 2378485.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.132142114639282,
|
|
"epoch": 1.0853355426677713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991338758580496,
|
|
"loss": 6.1043,
|
|
"mean_token_accuracy": 0.12516417354345322,
|
|
"num_tokens": 2387821.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.168926572799682,
|
|
"epoch": 1.0894780447390224,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004991056379058045,
|
|
"loss": 6.0131,
|
|
"mean_token_accuracy": 0.13040016889572142,
|
|
"num_tokens": 2396584.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.205351161956787,
|
|
"epoch": 1.0936205468102733,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990769479079976,
|
|
"loss": 6.1349,
|
|
"mean_token_accuracy": 0.12381747737526894,
|
|
"num_tokens": 2405533.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.267547273635865,
|
|
"epoch": 1.0977630488815244,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004990478059225003,
|
|
"loss": 6.1177,
|
|
"mean_token_accuracy": 0.1273566223680973,
|
|
"num_tokens": 2414147.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.170993137359619,
|
|
"epoch": 1.1019055509527755,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990182120080945,
|
|
"loss": 6.1233,
|
|
"mean_token_accuracy": 0.12716805413365365,
|
|
"num_tokens": 2422888.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.229750061035157,
|
|
"epoch": 1.1060480530240264,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989881662244748,
|
|
"loss": 6.0803,
|
|
"mean_token_accuracy": 0.13159133717417718,
|
|
"num_tokens": 2431165.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.236302328109741,
|
|
"epoch": 1.1101905550952775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989576686322468,
|
|
"loss": 6.1354,
|
|
"mean_token_accuracy": 0.12131609246134759,
|
|
"num_tokens": 2440726.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.230614805221558,
|
|
"epoch": 1.1143330571665286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989267192929272,
|
|
"loss": 6.1084,
|
|
"mean_token_accuracy": 0.12925527468323708,
|
|
"num_tokens": 2449393.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.115985679626465,
|
|
"epoch": 1.1184755592377795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004988953182689445,
|
|
"loss": 6.091,
|
|
"mean_token_accuracy": 0.1255219593644142,
|
|
"num_tokens": 2458210.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.1895167350769045,
|
|
"epoch": 1.1226180613090306,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004988634656236378,
|
|
"loss": 6.1111,
|
|
"mean_token_accuracy": 0.13017474710941315,
|
|
"num_tokens": 2466765.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.2510546207427975,
|
|
"epoch": 1.1267605633802817,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004988311614212576,
|
|
"loss": 6.1531,
|
|
"mean_token_accuracy": 0.1353672005236149,
|
|
"num_tokens": 2476517.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.119971084594726,
|
|
"epoch": 1.1309030654515326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004987984057269648,
|
|
"loss": 5.9815,
|
|
"mean_token_accuracy": 0.13376273065805436,
|
|
"num_tokens": 2485073.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.251840734481812,
|
|
"epoch": 1.1350455675227837,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004987651986068311,
|
|
"loss": 6.2087,
|
|
"mean_token_accuracy": 0.1218659982085228,
|
|
"num_tokens": 2493764.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.209574842453003,
|
|
"epoch": 1.1391880695940348,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004987315401278394,
|
|
"loss": 6.0867,
|
|
"mean_token_accuracy": 0.1295430727303028,
|
|
"num_tokens": 2502507.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.127098035812378,
|
|
"epoch": 1.143330571665286,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986974303578819,
|
|
"loss": 6.0602,
|
|
"mean_token_accuracy": 0.13360463380813598,
|
|
"num_tokens": 2510925.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.273448038101196,
|
|
"epoch": 1.1474730737365368,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986628693657621,
|
|
"loss": 6.1665,
|
|
"mean_token_accuracy": 0.1274927906692028,
|
|
"num_tokens": 2519927.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.180120134353638,
|
|
"epoch": 1.151615575807788,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004986278572211933,
|
|
"loss": 6.1406,
|
|
"mean_token_accuracy": 0.12608257159590722,
|
|
"num_tokens": 2529210.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.191981315612793,
|
|
"epoch": 1.1557580778790388,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004985923939947987,
|
|
"loss": 6.0294,
|
|
"mean_token_accuracy": 0.1362480901181698,
|
|
"num_tokens": 2538254.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.151539897918701,
|
|
"epoch": 1.15990057995029,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004985564797581115,
|
|
"loss": 6.0098,
|
|
"mean_token_accuracy": 0.13473647087812424,
|
|
"num_tokens": 2546563.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.353456687927246,
|
|
"epoch": 1.164043082021541,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004985201145835747,
|
|
"loss": 6.3072,
|
|
"mean_token_accuracy": 0.1152469776570797,
|
|
"num_tokens": 2556564.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.18480372428894,
|
|
"epoch": 1.1681855840927922,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498483298544541,
|
|
"loss": 6.1105,
|
|
"mean_token_accuracy": 0.12562097012996673,
|
|
"num_tokens": 2565904.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.174497842788696,
|
|
"epoch": 1.172328086164043,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000498446031715272,
|
|
"loss": 6.0035,
|
|
"mean_token_accuracy": 0.13123355731368064,
|
|
"num_tokens": 2574840.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.230139112472534,
|
|
"epoch": 1.1764705882352942,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984083141709391,
|
|
"loss": 6.2057,
|
|
"mean_token_accuracy": 0.1256824679672718,
|
|
"num_tokens": 2584419.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.165558242797852,
|
|
"epoch": 1.180613090306545,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000498370145987623,
|
|
"loss": 6.1846,
|
|
"mean_token_accuracy": 0.13115376979112625,
|
|
"num_tokens": 2593941.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.197154188156128,
|
|
"epoch": 1.1847555923777962,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004983315272423127,
|
|
"loss": 6.078,
|
|
"mean_token_accuracy": 0.13584018126130104,
|
|
"num_tokens": 2602473.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.174890565872192,
|
|
"epoch": 1.1888980944490473,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004982924580129066,
|
|
"loss": 6.0409,
|
|
"mean_token_accuracy": 0.1310078538954258,
|
|
"num_tokens": 2612287.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.03299913406372,
|
|
"epoch": 1.1930405965202984,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004982529383782116,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.1340833991765976,
|
|
"num_tokens": 2620557.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.140125703811646,
|
|
"epoch": 1.1971830985915493,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982129684179432,
|
|
"loss": 6.1482,
|
|
"mean_token_accuracy": 0.12488306164741517,
|
|
"num_tokens": 2630234.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.186292362213135,
|
|
"epoch": 1.2013256006628004,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004981725482127251,
|
|
"loss": 6.1455,
|
|
"mean_token_accuracy": 0.12422921806573868,
|
|
"num_tokens": 2639500.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.1869282722473145,
|
|
"epoch": 1.2054681027340513,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981316778440892,
|
|
"loss": 6.0985,
|
|
"mean_token_accuracy": 0.12317988276481628,
|
|
"num_tokens": 2648928.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.320038223266602,
|
|
"epoch": 1.2096106048053024,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004980903573944756,
|
|
"loss": 6.2304,
|
|
"mean_token_accuracy": 0.11908927112817765,
|
|
"num_tokens": 2658846.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.055701398849488,
|
|
"epoch": 1.2137531068765535,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004980485869472322,
|
|
"loss": 6.1225,
|
|
"mean_token_accuracy": 0.12987650632858277,
|
|
"num_tokens": 2668126.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.297923517227173,
|
|
"epoch": 1.2178956089478046,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004980063665866143,
|
|
"loss": 6.1422,
|
|
"mean_token_accuracy": 0.13109913021326064,
|
|
"num_tokens": 2676247.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.176990509033203,
|
|
"epoch": 1.2220381110190555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004979636963977853,
|
|
"loss": 6.0856,
|
|
"mean_token_accuracy": 0.13268199265003205,
|
|
"num_tokens": 2684964.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.132426357269287,
|
|
"epoch": 1.2261806130903066,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004979205764668156,
|
|
"loss": 6.1072,
|
|
"mean_token_accuracy": 0.12597387731075288,
|
|
"num_tokens": 2694143.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.088831424713135,
|
|
"epoch": 1.2303231151615575,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004978770068806825,
|
|
"loss": 6.0059,
|
|
"mean_token_accuracy": 0.1387046158313751,
|
|
"num_tokens": 2702243.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.178600311279297,
|
|
"epoch": 1.2344656172328086,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497832987727271,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.12923040315508844,
|
|
"num_tokens": 2710657.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.070341777801514,
|
|
"epoch": 1.2386081193040597,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004977885190953724,
|
|
"loss": 5.9968,
|
|
"mean_token_accuracy": 0.12945516631007195,
|
|
"num_tokens": 2719044.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.265737390518188,
|
|
"epoch": 1.2427506213753108,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004977436010746846,
|
|
"loss": 6.1141,
|
|
"mean_token_accuracy": 0.1338028959929943,
|
|
"num_tokens": 2727990.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2427506213753108,
|
|
"eval_entropy": 6.0073623586056835,
|
|
"eval_loss": 6.336162567138672,
|
|
"eval_mean_token_accuracy": 0.1255791528837513,
|
|
"eval_num_tokens": 2727990.0,
|
|
"eval_runtime": 3.1604,
|
|
"eval_samples_per_second": 1185.623,
|
|
"eval_steps_per_second": 148.401,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.1073040008544925,
|
|
"epoch": 1.2468931234465617,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004976982337558124,
|
|
"loss": 6.1223,
|
|
"mean_token_accuracy": 0.1255759708583355,
|
|
"num_tokens": 2736578.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.173589277267456,
|
|
"epoch": 1.2510356255178128,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976524172302665,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.13060156553983687,
|
|
"num_tokens": 2745693.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.202486658096314,
|
|
"epoch": 1.2551781275890637,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976061515904641,
|
|
"loss": 6.079,
|
|
"mean_token_accuracy": 0.13340094313025475,
|
|
"num_tokens": 2754952.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.176916265487671,
|
|
"epoch": 1.2593206296603148,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004975594369297277,
|
|
"loss": 6.0936,
|
|
"mean_token_accuracy": 0.12923406064510345,
|
|
"num_tokens": 2763981.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.051639413833618,
|
|
"epoch": 1.2634631317315659,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004975122733422861,
|
|
"loss": 6.1086,
|
|
"mean_token_accuracy": 0.12856508418917656,
|
|
"num_tokens": 2772740.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.0949808120727536,
|
|
"epoch": 1.267605633802817,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004974646609232732,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.13504123240709304,
|
|
"num_tokens": 2781489.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.120691251754761,
|
|
"epoch": 1.2717481358740679,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974165997687288,
|
|
"loss": 6.0706,
|
|
"mean_token_accuracy": 0.13792871534824372,
|
|
"num_tokens": 2790112.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.119127416610718,
|
|
"epoch": 1.275890637945319,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004973680899755972,
|
|
"loss": 6.1277,
|
|
"mean_token_accuracy": 0.12491117119789123,
|
|
"num_tokens": 2799940.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.173457908630371,
|
|
"epoch": 1.2800331400165699,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497319131641728,
|
|
"loss": 6.0299,
|
|
"mean_token_accuracy": 0.13059929981827736,
|
|
"num_tokens": 2808369.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.0894286155700685,
|
|
"epoch": 1.284175642087821,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004972697248658755,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.1335919052362442,
|
|
"num_tokens": 2818205.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.195007848739624,
|
|
"epoch": 1.288318144159072,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004972198697476987,
|
|
"loss": 6.1228,
|
|
"mean_token_accuracy": 0.13143828511238098,
|
|
"num_tokens": 2827350.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.121762228012085,
|
|
"epoch": 1.2924606462303232,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004971695663877607,
|
|
"loss": 6.049,
|
|
"mean_token_accuracy": 0.13058406859636307,
|
|
"num_tokens": 2836193.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.9933483600616455,
|
|
"epoch": 1.296603148301574,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004971188148875289,
|
|
"loss": 6.0246,
|
|
"mean_token_accuracy": 0.1318896971642971,
|
|
"num_tokens": 2846016.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.338242387771606,
|
|
"epoch": 1.3007456503728252,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004970676153493748,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.12545630112290382,
|
|
"num_tokens": 2856250.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.093726062774659,
|
|
"epoch": 1.3048881524440763,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004970159678765731,
|
|
"loss": 6.1197,
|
|
"mean_token_accuracy": 0.12585905566811562,
|
|
"num_tokens": 2865423.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.190137577056885,
|
|
"epoch": 1.3090306545153272,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004969638725733028,
|
|
"loss": 6.0842,
|
|
"mean_token_accuracy": 0.12343057692050934,
|
|
"num_tokens": 2874432.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.069796419143676,
|
|
"epoch": 1.3131731565865783,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969113295446455,
|
|
"loss": 6.063,
|
|
"mean_token_accuracy": 0.12247302234172822,
|
|
"num_tokens": 2884645.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.131494760513306,
|
|
"epoch": 1.3173156586578294,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004968583388965863,
|
|
"loss": 6.1312,
|
|
"mean_token_accuracy": 0.12738580405712127,
|
|
"num_tokens": 2894580.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.1513735294342045,
|
|
"epoch": 1.3214581607290803,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968049007360132,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.13449667394161224,
|
|
"num_tokens": 2903234.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.101231431961059,
|
|
"epoch": 1.3256006628003314,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004967510151707169,
|
|
"loss": 6.1014,
|
|
"mean_token_accuracy": 0.12553910091519355,
|
|
"num_tokens": 2912717.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.053020715713501,
|
|
"epoch": 1.3297431648715825,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004966966823093903,
|
|
"loss": 5.987,
|
|
"mean_token_accuracy": 0.13677354380488396,
|
|
"num_tokens": 2921449.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.0964494228363035,
|
|
"epoch": 1.3338856669428334,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004966419022616287,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.12726073935627938,
|
|
"num_tokens": 2930730.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.1492963314056395,
|
|
"epoch": 1.3380281690140845,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004965866751379296,
|
|
"loss": 6.104,
|
|
"mean_token_accuracy": 0.13120296895503997,
|
|
"num_tokens": 2940126.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.140308618545532,
|
|
"epoch": 1.3421706710853356,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004965310010496922,
|
|
"loss": 6.0093,
|
|
"mean_token_accuracy": 0.13938634768128394,
|
|
"num_tokens": 2949023.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.072773408889771,
|
|
"epoch": 1.3463131731565865,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004964748801092168,
|
|
"loss": 6.0634,
|
|
"mean_token_accuracy": 0.13058572858572007,
|
|
"num_tokens": 2957548.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.093048334121704,
|
|
"epoch": 1.3504556752278376,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004964183124297059,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.1446573927998543,
|
|
"num_tokens": 2965914.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.099730920791626,
|
|
"epoch": 1.3545981772990887,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004963612981252625,
|
|
"loss": 6.0614,
|
|
"mean_token_accuracy": 0.1373268634080887,
|
|
"num_tokens": 2974756.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.975049018859863,
|
|
"epoch": 1.3587406793703396,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004963038373108908,
|
|
"loss": 6.0593,
|
|
"mean_token_accuracy": 0.1383127264678478,
|
|
"num_tokens": 2984267.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.220937728881836,
|
|
"epoch": 1.3628831814415907,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004962459301024954,
|
|
"loss": 6.0875,
|
|
"mean_token_accuracy": 0.12818419858813285,
|
|
"num_tokens": 2992981.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.021561861038208,
|
|
"epoch": 1.3670256835128418,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004961875766168816,
|
|
"loss": 6.1065,
|
|
"mean_token_accuracy": 0.13113522082567214,
|
|
"num_tokens": 3002135.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.122888469696045,
|
|
"epoch": 1.371168185584093,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004961287769717547,
|
|
"loss": 6.0753,
|
|
"mean_token_accuracy": 0.12611212581396103,
|
|
"num_tokens": 3011579.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.200709390640259,
|
|
"epoch": 1.3753106876553438,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004960695312857199,
|
|
"loss": 6.1861,
|
|
"mean_token_accuracy": 0.12607147470116614,
|
|
"num_tokens": 3021975.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.122780990600586,
|
|
"epoch": 1.379453189726595,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004960098396782824,
|
|
"loss": 6.0975,
|
|
"mean_token_accuracy": 0.13255783617496492,
|
|
"num_tokens": 3030878.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.164667129516602,
|
|
"epoch": 1.3835956917978458,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004959497022698466,
|
|
"loss": 6.0787,
|
|
"mean_token_accuracy": 0.13239231780171395,
|
|
"num_tokens": 3040021.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.138751554489136,
|
|
"epoch": 1.387738193869097,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004958891191817163,
|
|
"loss": 6.2258,
|
|
"mean_token_accuracy": 0.12937600538134575,
|
|
"num_tokens": 3049944.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.212243986129761,
|
|
"epoch": 1.391880695940348,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000495828090536094,
|
|
"loss": 6.0185,
|
|
"mean_token_accuracy": 0.1379230573773384,
|
|
"num_tokens": 3059327.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.974050140380859,
|
|
"epoch": 1.3960231980115991,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004957666164560815,
|
|
"loss": 6.0064,
|
|
"mean_token_accuracy": 0.13503726869821547,
|
|
"num_tokens": 3068325.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.288743686676026,
|
|
"epoch": 1.40016570008285,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004957046970656784,
|
|
"loss": 6.0853,
|
|
"mean_token_accuracy": 0.13486202135682107,
|
|
"num_tokens": 3076715.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.045103120803833,
|
|
"epoch": 1.4043082021541011,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004956423324897831,
|
|
"loss": 6.15,
|
|
"mean_token_accuracy": 0.13253118470311165,
|
|
"num_tokens": 3086051.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.1374091625213625,
|
|
"epoch": 1.408450704225352,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004955795228541917,
|
|
"loss": 5.9694,
|
|
"mean_token_accuracy": 0.1330952487885952,
|
|
"num_tokens": 3094567.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.041680717468262,
|
|
"epoch": 1.4125932062966031,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000495516268285598,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13208642452955247,
|
|
"num_tokens": 3103821.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.039361143112183,
|
|
"epoch": 1.4167357083678542,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004954525689115935,
|
|
"loss": 5.9978,
|
|
"mean_token_accuracy": 0.13572969138622284,
|
|
"num_tokens": 3113399.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.049833297729492,
|
|
"epoch": 1.4208782104391053,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004953884248606669,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.1403762809932232,
|
|
"num_tokens": 3122863.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.1296131134033205,
|
|
"epoch": 1.4250207125103562,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953238362622034,
|
|
"loss": 6.0345,
|
|
"mean_token_accuracy": 0.14129049256443976,
|
|
"num_tokens": 3131250.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.0656098365783695,
|
|
"epoch": 1.4291632145816073,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004952588032464855,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.14013449773192405,
|
|
"num_tokens": 3140120.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.044841003417969,
|
|
"epoch": 1.4333057166528582,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004951933259446919,
|
|
"loss": 6.0884,
|
|
"mean_token_accuracy": 0.13288939371705055,
|
|
"num_tokens": 3149304.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.077728414535523,
|
|
"epoch": 1.4374482187241093,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004951274044888975,
|
|
"loss": 6.0684,
|
|
"mean_token_accuracy": 0.13015799820423127,
|
|
"num_tokens": 3158386.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.156499576568604,
|
|
"epoch": 1.4415907207953604,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004950610390120731,
|
|
"loss": 6.0927,
|
|
"mean_token_accuracy": 0.12830716595053673,
|
|
"num_tokens": 3166971.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.911584281921387,
|
|
"epoch": 1.4457332228666115,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000494994229648085,
|
|
"loss": 5.9268,
|
|
"mean_token_accuracy": 0.1378532201051712,
|
|
"num_tokens": 3175678.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.382007503509522,
|
|
"epoch": 1.4498757249378624,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004949269765316948,
|
|
"loss": 6.4007,
|
|
"mean_token_accuracy": 0.12360990047454834,
|
|
"num_tokens": 3185151.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.003203964233398,
|
|
"epoch": 1.4540182270091135,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004948592797985599,
|
|
"loss": 5.9886,
|
|
"mean_token_accuracy": 0.12840774431824684,
|
|
"num_tokens": 3193016.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.187309694290161,
|
|
"epoch": 1.4581607290803644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004947911395852317,
|
|
"loss": 6.0342,
|
|
"mean_token_accuracy": 0.1325800023972988,
|
|
"num_tokens": 3201475.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.989087677001953,
|
|
"epoch": 1.4623032311516155,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004947225560291565,
|
|
"loss": 5.9032,
|
|
"mean_token_accuracy": 0.13708751201629638,
|
|
"num_tokens": 3210602.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.112654399871826,
|
|
"epoch": 1.4664457332228666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004946535292686746,
|
|
"loss": 6.2026,
|
|
"mean_token_accuracy": 0.12221089527010917,
|
|
"num_tokens": 3220364.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.232067012786866,
|
|
"epoch": 1.4705882352941178,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004945840594430209,
|
|
"loss": 6.1317,
|
|
"mean_token_accuracy": 0.12892260923981666,
|
|
"num_tokens": 3230019.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.085176420211792,
|
|
"epoch": 1.4747307373653686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004945141466923235,
|
|
"loss": 6.1078,
|
|
"mean_token_accuracy": 0.13496615067124368,
|
|
"num_tokens": 3239211.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.994139719009399,
|
|
"epoch": 1.4788732394366197,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004944437911576041,
|
|
"loss": 5.9044,
|
|
"mean_token_accuracy": 0.1325595885515213,
|
|
"num_tokens": 3248270.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.989656925201416,
|
|
"epoch": 1.4830157415078706,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004943729929807772,
|
|
"loss": 5.9634,
|
|
"mean_token_accuracy": 0.14057789370417595,
|
|
"num_tokens": 3257155.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.030780601501465,
|
|
"epoch": 1.4871582435791217,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004943017523046507,
|
|
"loss": 5.9527,
|
|
"mean_token_accuracy": 0.13465590700507163,
|
|
"num_tokens": 3267242.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.105288457870484,
|
|
"epoch": 1.4913007456503728,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000494230069272925,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.133008673787117,
|
|
"num_tokens": 3276420.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.994972229003906,
|
|
"epoch": 1.495443247721624,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004941579440301923,
|
|
"loss": 5.9439,
|
|
"mean_token_accuracy": 0.13288045302033424,
|
|
"num_tokens": 3285954.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.132870626449585,
|
|
"epoch": 1.4995857497928748,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004940853767219372,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.13481004983186723,
|
|
"num_tokens": 3296448.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.981975412368774,
|
|
"epoch": 1.503728251864126,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004940123674945359,
|
|
"loss": 6.0485,
|
|
"mean_token_accuracy": 0.1355610877275467,
|
|
"num_tokens": 3306084.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.126554489135742,
|
|
"epoch": 1.5078707539353768,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004939389164952561,
|
|
"loss": 6.0361,
|
|
"mean_token_accuracy": 0.13744445368647576,
|
|
"num_tokens": 3314865.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.0176558017730715,
|
|
"epoch": 1.512013256006628,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938650238722564,
|
|
"loss": 5.9756,
|
|
"mean_token_accuracy": 0.13883665949106216,
|
|
"num_tokens": 3323768.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.159342956542969,
|
|
"epoch": 1.516155758077879,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004937906897745861,
|
|
"loss": 6.1306,
|
|
"mean_token_accuracy": 0.1313806004822254,
|
|
"num_tokens": 3333722.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.157751893997192,
|
|
"epoch": 1.5202982601491302,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004937159143521856,
|
|
"loss": 6.0871,
|
|
"mean_token_accuracy": 0.13825847432017327,
|
|
"num_tokens": 3343950.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.122276544570923,
|
|
"epoch": 1.5244407622203813,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004936406977558847,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.1342560924589634,
|
|
"num_tokens": 3352526.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.074772739410401,
|
|
"epoch": 1.5285832642916322,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004935650401374039,
|
|
"loss": 6.0522,
|
|
"mean_token_accuracy": 0.13890262842178344,
|
|
"num_tokens": 3361685.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.141188383102417,
|
|
"epoch": 1.532725766362883,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004934889416493524,
|
|
"loss": 6.012,
|
|
"mean_token_accuracy": 0.13915711641311646,
|
|
"num_tokens": 3369967.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.9735277652740475,
|
|
"epoch": 1.5368682684341342,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004934124024452295,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.12827755585312844,
|
|
"num_tokens": 3379257.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.106106615066528,
|
|
"epoch": 1.5410107705053853,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004933354226794232,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.13620708286762237,
|
|
"num_tokens": 3388157.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.950488042831421,
|
|
"epoch": 1.5451532725766364,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004932580025072098,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1455531969666481,
|
|
"num_tokens": 3397675.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.12842583656311,
|
|
"epoch": 1.5492957746478875,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004931801420847545,
|
|
"loss": 6.0164,
|
|
"mean_token_accuracy": 0.13580601513385773,
|
|
"num_tokens": 3406248.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.97610650062561,
|
|
"epoch": 1.5534382767191384,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004931018415691102,
|
|
"loss": 5.9533,
|
|
"mean_token_accuracy": 0.14214757829904556,
|
|
"num_tokens": 3414600.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.946225595474243,
|
|
"epoch": 1.5575807787903893,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004930231011182176,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.14622760117053984,
|
|
"num_tokens": 3422608.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.0624736785888675,
|
|
"epoch": 1.5617232808616404,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004929439208909048,
|
|
"loss": 5.9276,
|
|
"mean_token_accuracy": 0.13686055019497873,
|
|
"num_tokens": 3431661.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.007996988296509,
|
|
"epoch": 1.5658657829328915,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004928643010468869,
|
|
"loss": 6.0189,
|
|
"mean_token_accuracy": 0.13368499726057054,
|
|
"num_tokens": 3440654.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.140294456481934,
|
|
"epoch": 1.5700082850041426,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004927842417467659,
|
|
"loss": 6.0864,
|
|
"mean_token_accuracy": 0.12963276728987694,
|
|
"num_tokens": 3450041.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.986852550506592,
|
|
"epoch": 1.5741507870753937,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004927037431520302,
|
|
"loss": 6.0653,
|
|
"mean_token_accuracy": 0.1344601631164551,
|
|
"num_tokens": 3459092.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.128014421463012,
|
|
"epoch": 1.5782932891466446,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004926228054250541,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.13942472040653228,
|
|
"num_tokens": 3467789.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.955077886581421,
|
|
"epoch": 1.5824357912178955,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004925414287290979,
|
|
"loss": 5.9321,
|
|
"mean_token_accuracy": 0.13595775440335273,
|
|
"num_tokens": 3476264.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.037222576141358,
|
|
"epoch": 1.5865782932891466,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004924596132283075,
|
|
"loss": 5.9935,
|
|
"mean_token_accuracy": 0.1471448689699173,
|
|
"num_tokens": 3485469.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.157397603988647,
|
|
"epoch": 1.5907207953603977,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004923773590877135,
|
|
"loss": 6.045,
|
|
"mean_token_accuracy": 0.13453733399510384,
|
|
"num_tokens": 3494222.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.979045104980469,
|
|
"epoch": 1.5948632974316488,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004922946664732314,
|
|
"loss": 6.067,
|
|
"mean_token_accuracy": 0.13578267619013787,
|
|
"num_tokens": 3504473.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.982046699523925,
|
|
"epoch": 1.5990057995029,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004922115355516615,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.14055346548557282,
|
|
"num_tokens": 3512704.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.066261434555054,
|
|
"epoch": 1.6031483015741508,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004921279664906877,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.138848215341568,
|
|
"num_tokens": 3521600.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.009872245788574,
|
|
"epoch": 1.6072908036454017,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004920439594588781,
|
|
"loss": 6.118,
|
|
"mean_token_accuracy": 0.12350448817014695,
|
|
"num_tokens": 3531666.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.069752836227417,
|
|
"epoch": 1.6114333057166528,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000491959514625684,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.13175680488348007,
|
|
"num_tokens": 3541145.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.08793363571167,
|
|
"epoch": 1.6155758077879039,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049187463216144,
|
|
"loss": 5.9994,
|
|
"mean_token_accuracy": 0.14488411992788314,
|
|
"num_tokens": 3550151.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.134615421295166,
|
|
"epoch": 1.619718309859155,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491789312237363,
|
|
"loss": 5.9934,
|
|
"mean_token_accuracy": 0.1358059048652649,
|
|
"num_tokens": 3560048.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.870369815826416,
|
|
"epoch": 1.623860811930406,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004917035550255528,
|
|
"loss": 5.8206,
|
|
"mean_token_accuracy": 0.1509326860308647,
|
|
"num_tokens": 3568473.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.949672269821167,
|
|
"epoch": 1.628003314001657,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491617360698991,
|
|
"loss": 5.9349,
|
|
"mean_token_accuracy": 0.14107762947678565,
|
|
"num_tokens": 3577188.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.116447257995605,
|
|
"epoch": 1.6321458160729079,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004915307294315412,
|
|
"loss": 6.1147,
|
|
"mean_token_accuracy": 0.1352829672396183,
|
|
"num_tokens": 3587047.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.995927095413208,
|
|
"epoch": 1.636288318144159,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000491443661397948,
|
|
"loss": 6.0111,
|
|
"mean_token_accuracy": 0.13478146567940713,
|
|
"num_tokens": 3595599.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.004126787185669,
|
|
"epoch": 1.64043082021541,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004913561567738368,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.13491463288664818,
|
|
"num_tokens": 3604539.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.084889984130859,
|
|
"epoch": 1.6445733222866612,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004912682157357146,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1341504581272602,
|
|
"num_tokens": 3613463.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.067630052566528,
|
|
"epoch": 1.6487158243579123,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004911798384609675,
|
|
"loss": 5.9769,
|
|
"mean_token_accuracy": 0.134279103577137,
|
|
"num_tokens": 3622131.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.210787439346314,
|
|
"epoch": 1.6528583264291632,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004910910251278626,
|
|
"loss": 6.1647,
|
|
"mean_token_accuracy": 0.1341046206653118,
|
|
"num_tokens": 3632338.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.020812225341797,
|
|
"epoch": 1.6570008285004143,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004910017759155458,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.13550196066498757,
|
|
"num_tokens": 3640941.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6570008285004143,
|
|
"eval_entropy": 5.842879085906787,
|
|
"eval_loss": 6.168454647064209,
|
|
"eval_mean_token_accuracy": 0.13479597267629242,
|
|
"eval_num_tokens": 3640941.0,
|
|
"eval_runtime": 2.9149,
|
|
"eval_samples_per_second": 1285.449,
|
|
"eval_steps_per_second": 160.896,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.0745893001556395,
|
|
"epoch": 1.6611433305716652,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004909120910040427,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.13120997697114944,
|
|
"num_tokens": 3650699.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.0884589672088625,
|
|
"epoch": 1.6652858326429163,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004908219705742574,
|
|
"loss": 6.055,
|
|
"mean_token_accuracy": 0.13785433918237686,
|
|
"num_tokens": 3660734.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.940933465957642,
|
|
"epoch": 1.6694283347141674,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004907314148079727,
|
|
"loss": 5.9021,
|
|
"mean_token_accuracy": 0.14264713525772094,
|
|
"num_tokens": 3670176.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.014130783081055,
|
|
"epoch": 1.6735708367854185,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004906404238878496,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.14376770704984665,
|
|
"num_tokens": 3679559.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.968165588378906,
|
|
"epoch": 1.6777133388566694,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004905489979974265,
|
|
"loss": 5.8978,
|
|
"mean_token_accuracy": 0.14285829290747643,
|
|
"num_tokens": 3688331.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.978603506088257,
|
|
"epoch": 1.6818558409279205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004904571373211197,
|
|
"loss": 5.9251,
|
|
"mean_token_accuracy": 0.14525610953569412,
|
|
"num_tokens": 3696579.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.980389404296875,
|
|
"epoch": 1.6859983429991714,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000490364842044222,
|
|
"loss": 5.9038,
|
|
"mean_token_accuracy": 0.1411435402929783,
|
|
"num_tokens": 3706019.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.014318990707397,
|
|
"epoch": 1.6901408450704225,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004902721123529031,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.14145262613892556,
|
|
"num_tokens": 3715090.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.995026445388794,
|
|
"epoch": 1.6942833471416736,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000490178948434209,
|
|
"loss": 5.9094,
|
|
"mean_token_accuracy": 0.13903800398111343,
|
|
"num_tokens": 3723721.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.010066366195678,
|
|
"epoch": 1.6984258492129247,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004900853504760614,
|
|
"loss": 6.0361,
|
|
"mean_token_accuracy": 0.13761742487549783,
|
|
"num_tokens": 3732712.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.054876184463501,
|
|
"epoch": 1.7025683512841756,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004899913186672577,
|
|
"loss": 5.9576,
|
|
"mean_token_accuracy": 0.13415518999099732,
|
|
"num_tokens": 3742063.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.035504293441773,
|
|
"epoch": 1.7067108533554267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004898968531974702,
|
|
"loss": 6.0477,
|
|
"mean_token_accuracy": 0.13364452123641968,
|
|
"num_tokens": 3751807.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.036947965621948,
|
|
"epoch": 1.7108533554266776,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004898019542572462,
|
|
"loss": 6.0153,
|
|
"mean_token_accuracy": 0.13318413347005845,
|
|
"num_tokens": 3761316.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.041462182998657,
|
|
"epoch": 1.7149958574979287,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004897066220380073,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.14148937910795212,
|
|
"num_tokens": 3769897.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.976815700531006,
|
|
"epoch": 1.7191383595691798,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004896108567320489,
|
|
"loss": 5.9527,
|
|
"mean_token_accuracy": 0.13763240873813629,
|
|
"num_tokens": 3778678.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.976473712921143,
|
|
"epoch": 1.723280861640431,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048951465853254,
|
|
"loss": 5.9999,
|
|
"mean_token_accuracy": 0.13764951303601264,
|
|
"num_tokens": 3787742.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.105862426757812,
|
|
"epoch": 1.7274233637116818,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004894180276335231,
|
|
"loss": 6.0571,
|
|
"mean_token_accuracy": 0.1314976066350937,
|
|
"num_tokens": 3796300.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.02824821472168,
|
|
"epoch": 1.731565865782933,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004893209642299132,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.13603372126817703,
|
|
"num_tokens": 3805901.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.987863159179687,
|
|
"epoch": 1.7357083678541838,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004892234685174976,
|
|
"loss": 6.0362,
|
|
"mean_token_accuracy": 0.13513321056962013,
|
|
"num_tokens": 3815605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.015270948410034,
|
|
"epoch": 1.739850869925435,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004891255406929359,
|
|
"loss": 5.9213,
|
|
"mean_token_accuracy": 0.13604082316160201,
|
|
"num_tokens": 3824444.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.971472120285034,
|
|
"epoch": 1.743993371996686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004890271809537595,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.13630633503198625,
|
|
"num_tokens": 3833905.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.042605066299439,
|
|
"epoch": 1.7481358740679371,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004889283894983704,
|
|
"loss": 6.0307,
|
|
"mean_token_accuracy": 0.13705341815948485,
|
|
"num_tokens": 3843176.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.8697154998779295,
|
|
"epoch": 1.752278376139188,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000488829166526042,
|
|
"loss": 5.8094,
|
|
"mean_token_accuracy": 0.14387866854667664,
|
|
"num_tokens": 3852263.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.881034851074219,
|
|
"epoch": 1.7564208782104391,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004887295122369176,
|
|
"loss": 5.8988,
|
|
"mean_token_accuracy": 0.14097962900996208,
|
|
"num_tokens": 3861393.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.992901229858399,
|
|
"epoch": 1.76056338028169,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000488629426832011,
|
|
"loss": 5.8526,
|
|
"mean_token_accuracy": 0.13920771479606628,
|
|
"num_tokens": 3870246.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.043943119049072,
|
|
"epoch": 1.7647058823529411,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004885289105132054,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.14054998755455017,
|
|
"num_tokens": 3879486.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.985035562515259,
|
|
"epoch": 1.7688483844241922,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004884279634832532,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.14789656475186347,
|
|
"num_tokens": 3889187.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.948019742965698,
|
|
"epoch": 1.7729908864954433,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004883265859457754,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.14227469339966775,
|
|
"num_tokens": 3898503.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.094852542877197,
|
|
"epoch": 1.7771333885666944,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048822477810526186,
|
|
"loss": 5.9493,
|
|
"mean_token_accuracy": 0.1418413206934929,
|
|
"num_tokens": 3907634.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.974353456497193,
|
|
"epoch": 1.7812758906379453,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048812254016706994,
|
|
"loss": 5.9285,
|
|
"mean_token_accuracy": 0.13976011276245118,
|
|
"num_tokens": 3917057.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.029307699203491,
|
|
"epoch": 1.7854183927091962,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004880198723374247,
|
|
"loss": 6.013,
|
|
"mean_token_accuracy": 0.13228121027350426,
|
|
"num_tokens": 3926854.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.1075136184692385,
|
|
"epoch": 1.7895608947804473,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048791677482341845,
|
|
"loss": 5.991,
|
|
"mean_token_accuracy": 0.1414928138256073,
|
|
"num_tokens": 3935820.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.875960397720337,
|
|
"epoch": 1.7937033968516984,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048781324783301016,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.1361390620470047,
|
|
"num_tokens": 3944804.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.9698957920074465,
|
|
"epoch": 1.7978458989229495,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048770929157502505,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.14194637387990952,
|
|
"num_tokens": 3953534.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.069232940673828,
|
|
"epoch": 1.8019884009942007,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004876049062591543,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.13576488494873046,
|
|
"num_tokens": 3963268.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.971802806854248,
|
|
"epoch": 1.8061309030654515,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048750009209595443,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.1482763409614563,
|
|
"num_tokens": 3972571.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.944825315475464,
|
|
"epoch": 1.8102734051367024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048739484929684714,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.14042048305273055,
|
|
"num_tokens": 3982813.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.9267841339111325,
|
|
"epoch": 1.8144159072079535,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048728917807411856,
|
|
"loss": 5.984,
|
|
"mean_token_accuracy": 0.13693317994475365,
|
|
"num_tokens": 3990743.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.062971305847168,
|
|
"epoch": 1.8185584092792046,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004871830786409191,
|
|
"loss": 5.9275,
|
|
"mean_token_accuracy": 0.1332172743976116,
|
|
"num_tokens": 3999849.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.995508766174316,
|
|
"epoch": 1.8227009113504558,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048707655121126296,
|
|
"loss": 5.9233,
|
|
"mean_token_accuracy": 0.14391714483499526,
|
|
"num_tokens": 4009035.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.967374420166015,
|
|
"epoch": 1.8268434134217069,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048696959600002767,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.1440461315214634,
|
|
"num_tokens": 4018593.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.94206805229187,
|
|
"epoch": 1.8309859154929577,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048686221322295347,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.13509892001748086,
|
|
"num_tokens": 4027744.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.006576442718506,
|
|
"epoch": 1.8351284175642086,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048675440309664314,
|
|
"loss": 5.9717,
|
|
"mean_token_accuracy": 0.13625107258558272,
|
|
"num_tokens": 4036195.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.954492235183716,
|
|
"epoch": 1.8392709196354597,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048664616583856166,
|
|
"loss": 5.9396,
|
|
"mean_token_accuracy": 0.13764657974243164,
|
|
"num_tokens": 4045676.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.978009748458862,
|
|
"epoch": 1.8434134217067109,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048653750166703516,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.14014019593596458,
|
|
"num_tokens": 4055285.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.019568204879761,
|
|
"epoch": 1.847555923777962,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004864284108012513,
|
|
"loss": 5.9945,
|
|
"mean_token_accuracy": 0.13550701290369033,
|
|
"num_tokens": 4065149.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.068925380706787,
|
|
"epoch": 1.851698425849213,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004863188934612583,
|
|
"loss": 6.0048,
|
|
"mean_token_accuracy": 0.13284957855939866,
|
|
"num_tokens": 4074807.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.9438426971435545,
|
|
"epoch": 1.855840927920464,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004862089498679645,
|
|
"loss": 5.863,
|
|
"mean_token_accuracy": 0.14104679301381112,
|
|
"num_tokens": 4083448.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.011416959762573,
|
|
"epoch": 1.8599834299917148,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004860985802431383,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.13921424895524978,
|
|
"num_tokens": 4092810.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.0292891502380375,
|
|
"epoch": 1.864125932062966,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004859877848094071,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.13659395277500153,
|
|
"num_tokens": 4101927.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.995765256881714,
|
|
"epoch": 1.868268434134217,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004858765637902575,
|
|
"loss": 5.8955,
|
|
"mean_token_accuracy": 0.14541091918945312,
|
|
"num_tokens": 4110923.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.908204555511475,
|
|
"epoch": 1.8724109362054682,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004857649174100344,
|
|
"loss": 5.8152,
|
|
"mean_token_accuracy": 0.1410209134221077,
|
|
"num_tokens": 4119939.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.906198358535766,
|
|
"epoch": 1.8765534382767193,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004856528458939409,
|
|
"loss": 5.8779,
|
|
"mean_token_accuracy": 0.14143114984035493,
|
|
"num_tokens": 4129091.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.967985534667969,
|
|
"epoch": 1.8806959403479702,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004855403494680373,
|
|
"loss": 5.8757,
|
|
"mean_token_accuracy": 0.1409454122185707,
|
|
"num_tokens": 4138070.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.897773122787475,
|
|
"epoch": 1.884838442419221,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048542742835924133,
|
|
"loss": 5.8546,
|
|
"mean_token_accuracy": 0.1405012011528015,
|
|
"num_tokens": 4147002.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.997664833068848,
|
|
"epoch": 1.8889809444904722,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048531408279532705,
|
|
"loss": 6.0221,
|
|
"mean_token_accuracy": 0.13353719785809517,
|
|
"num_tokens": 4156525.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.004969120025635,
|
|
"epoch": 1.8931234465617233,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048520031300492507,
|
|
"loss": 5.8583,
|
|
"mean_token_accuracy": 0.14217624366283416,
|
|
"num_tokens": 4165485.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.8434123516082765,
|
|
"epoch": 1.8972659486329744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048508611921752124,
|
|
"loss": 5.8519,
|
|
"mean_token_accuracy": 0.14170566573739052,
|
|
"num_tokens": 4174721.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.057034206390381,
|
|
"epoch": 1.9014084507042255,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048497150166345703,
|
|
"loss": 5.9922,
|
|
"mean_token_accuracy": 0.13678756430745126,
|
|
"num_tokens": 4184981.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.052671194076538,
|
|
"epoch": 1.9055509527754764,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004848564605739287,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.13604575917124748,
|
|
"num_tokens": 4193912.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.900245618820191,
|
|
"epoch": 1.9096934548467275,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048474099618098644,
|
|
"loss": 5.8865,
|
|
"mean_token_accuracy": 0.14115182384848596,
|
|
"num_tokens": 4202904.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.934115266799926,
|
|
"epoch": 1.9138359569179784,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004846251087175348,
|
|
"loss": 5.8392,
|
|
"mean_token_accuracy": 0.1444745846092701,
|
|
"num_tokens": 4212072.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.873389196395874,
|
|
"epoch": 1.9179784589892295,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004845087984173312,
|
|
"loss": 5.7925,
|
|
"mean_token_accuracy": 0.14933519586920738,
|
|
"num_tokens": 4220673.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.9027698040008545,
|
|
"epoch": 1.9221209610604806,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048439206551498644,
|
|
"loss": 5.8719,
|
|
"mean_token_accuracy": 0.14407336413860322,
|
|
"num_tokens": 4229433.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.9854466915130615,
|
|
"epoch": 1.9262634631317317,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004842749102459635,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.1422778360545635,
|
|
"num_tokens": 4239345.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.872208976745606,
|
|
"epoch": 1.9304059652029826,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048415733284657727,
|
|
"loss": 5.7868,
|
|
"mean_token_accuracy": 0.15144341737031936,
|
|
"num_tokens": 4248411.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.881693983078003,
|
|
"epoch": 1.9345484672742337,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004840393335539943,
|
|
"loss": 5.8006,
|
|
"mean_token_accuracy": 0.15395894944667815,
|
|
"num_tokens": 4256745.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.962243127822876,
|
|
"epoch": 1.9386909693454846,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048392091260623195,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.1411220297217369,
|
|
"num_tokens": 4266533.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.138938760757446,
|
|
"epoch": 1.9428334714167357,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048380207024215836,
|
|
"loss": 6.0692,
|
|
"mean_token_accuracy": 0.13505000993609428,
|
|
"num_tokens": 4276598.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.963536787033081,
|
|
"epoch": 1.9469759734879868,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048368280670149145,
|
|
"loss": 5.9542,
|
|
"mean_token_accuracy": 0.13387920409440995,
|
|
"num_tokens": 4285800.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.970139408111573,
|
|
"epoch": 1.951118475559238,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004835631222247989,
|
|
"loss": 5.8984,
|
|
"mean_token_accuracy": 0.1364313505589962,
|
|
"num_tokens": 4295583.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.906875705718994,
|
|
"epoch": 1.9552609776304888,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048344301705349735,
|
|
"loss": 5.7466,
|
|
"mean_token_accuracy": 0.15295102968811988,
|
|
"num_tokens": 4303984.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.84448127746582,
|
|
"epoch": 1.95940347970174,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048332249142985195,
|
|
"loss": 5.907,
|
|
"mean_token_accuracy": 0.13456033021211625,
|
|
"num_tokens": 4312956.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.027677774429321,
|
|
"epoch": 1.9635459817729908,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004832015455969763,
|
|
"loss": 5.8737,
|
|
"mean_token_accuracy": 0.1478977844119072,
|
|
"num_tokens": 4321795.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.860297346115113,
|
|
"epoch": 1.9676884838442419,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048308017979883114,
|
|
"loss": 5.8102,
|
|
"mean_token_accuracy": 0.14385854452848434,
|
|
"num_tokens": 4330579.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.9947404861450195,
|
|
"epoch": 1.971830985915493,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048295839428022454,
|
|
"loss": 5.9063,
|
|
"mean_token_accuracy": 0.1438973106443882,
|
|
"num_tokens": 4339155.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.856879043579101,
|
|
"epoch": 1.975973487986744,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048283618928681127,
|
|
"loss": 5.7939,
|
|
"mean_token_accuracy": 0.14630988240242004,
|
|
"num_tokens": 4348759.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.932731056213379,
|
|
"epoch": 1.980115990057995,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004827135650650923,
|
|
"loss": 5.9188,
|
|
"mean_token_accuracy": 0.1481543578207493,
|
|
"num_tokens": 4358164.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.920789957046509,
|
|
"epoch": 1.984258492129246,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004825905218624139,
|
|
"loss": 5.8305,
|
|
"mean_token_accuracy": 0.14525951594114303,
|
|
"num_tokens": 4366709.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.859962129592896,
|
|
"epoch": 1.988400994200497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048246705992696756,
|
|
"loss": 5.7503,
|
|
"mean_token_accuracy": 0.14968549385666846,
|
|
"num_tokens": 4375055.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.044453191757202,
|
|
"epoch": 1.992543496271748,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048234317950778975,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.13369215801358222,
|
|
"num_tokens": 4384495.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.93978180885315,
|
|
"epoch": 1.9966859983429992,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004822188808547607,
|
|
"loss": 5.8294,
|
|
"mean_token_accuracy": 0.14511772394180297,
|
|
"num_tokens": 4393208.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.007810449600219,
|
|
"epoch": 2.0008285004142503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004820941642186043,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.1328720286488533,
|
|
"num_tokens": 4401882.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.8302782535552975,
|
|
"epoch": 2.0049710024855014,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004819690298508879,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.15972236916422844,
|
|
"num_tokens": 4411434.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.834520959854126,
|
|
"epoch": 2.009113504556752,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048184347800402083,
|
|
"loss": 5.5917,
|
|
"mean_token_accuracy": 0.1504381477832794,
|
|
"num_tokens": 4421519.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.888366079330444,
|
|
"epoch": 2.013256006628003,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004817175089312551,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.16161237508058549,
|
|
"num_tokens": 4430208.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.771408176422119,
|
|
"epoch": 2.0173985086992543,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048159112288668403,
|
|
"loss": 5.6437,
|
|
"mean_token_accuracy": 0.14739402011036873,
|
|
"num_tokens": 4440214.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.905699634552002,
|
|
"epoch": 2.0215410107705054,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048146432012524215,
|
|
"loss": 5.5972,
|
|
"mean_token_accuracy": 0.1480880782008171,
|
|
"num_tokens": 4449406.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.752327966690063,
|
|
"epoch": 2.0256835128417565,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048133710090270446,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.1535774514079094,
|
|
"num_tokens": 4458322.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.858113861083984,
|
|
"epoch": 2.0298260149130076,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048120946547568596,
|
|
"loss": 5.5154,
|
|
"mean_token_accuracy": 0.1561787486076355,
|
|
"num_tokens": 4468132.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.7455836772918705,
|
|
"epoch": 2.0339685169842583,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048108141410164133,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.15598926022648812,
|
|
"num_tokens": 4478846.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.9118971824646,
|
|
"epoch": 2.0381110190555094,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004809529470388642,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.1582340955734253,
|
|
"num_tokens": 4488237.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.821217727661133,
|
|
"epoch": 2.0422535211267605,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004808240645464866,
|
|
"loss": 5.6067,
|
|
"mean_token_accuracy": 0.15413790419697762,
|
|
"num_tokens": 4497568.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.926329755783081,
|
|
"epoch": 2.0463960231980116,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004806947668844787,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.1584421366453171,
|
|
"num_tokens": 4506050.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.811815643310547,
|
|
"epoch": 2.0505385252692627,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048056505431364794,
|
|
"loss": 5.5402,
|
|
"mean_token_accuracy": 0.1506636083126068,
|
|
"num_tokens": 4514734.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.803349113464355,
|
|
"epoch": 2.054681027340514,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048043492709563877,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.1571820303797722,
|
|
"num_tokens": 4524026.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.774190044403076,
|
|
"epoch": 2.0588235294117645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048030438549293193,
|
|
"loss": 5.5649,
|
|
"mean_token_accuracy": 0.156453837454319,
|
|
"num_tokens": 4532209.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.891646766662598,
|
|
"epoch": 2.0629660314830156,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048017342976884424,
|
|
"loss": 5.5948,
|
|
"mean_token_accuracy": 0.15483448803424835,
|
|
"num_tokens": 4542095.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.751562309265137,
|
|
"epoch": 2.0671085335542667,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048004206018752773,
|
|
"loss": 5.5585,
|
|
"mean_token_accuracy": 0.15620076209306716,
|
|
"num_tokens": 4550530.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.836578750610352,
|
|
"epoch": 2.071251035625518,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047991027701396895,
|
|
"loss": 5.6432,
|
|
"mean_token_accuracy": 0.1558740496635437,
|
|
"num_tokens": 4560560.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.071251035625518,
|
|
"eval_entropy": 5.817075447741348,
|
|
"eval_loss": 6.062323093414307,
|
|
"eval_mean_token_accuracy": 0.14113807419279237,
|
|
"eval_num_tokens": 4560560.0,
|
|
"eval_runtime": 2.9157,
|
|
"eval_samples_per_second": 1285.093,
|
|
"eval_steps_per_second": 160.851,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.922831439971924,
|
|
"epoch": 2.075393537696769,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004797780805139892,
|
|
"loss": 5.5875,
|
|
"mean_token_accuracy": 0.15712048560380937,
|
|
"num_tokens": 4570459.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.71963062286377,
|
|
"epoch": 2.07953603976802,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004796454709542432,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.15603303909301758,
|
|
"num_tokens": 4578595.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.822841930389404,
|
|
"epoch": 2.083678541839271,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795124486022189,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.15312176644802095,
|
|
"num_tokens": 4588207.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.7801848411560055,
|
|
"epoch": 2.087821043910522,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047937901372623694,
|
|
"loss": 5.5188,
|
|
"mean_token_accuracy": 0.1585146576166153,
|
|
"num_tokens": 4596819.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.825238847732544,
|
|
"epoch": 2.091963545981773,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047924516659545005,
|
|
"loss": 5.5152,
|
|
"mean_token_accuracy": 0.1578638359904289,
|
|
"num_tokens": 4604741.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.674453163146973,
|
|
"epoch": 2.096106048053024,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004791109074798425,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.15454403012990953,
|
|
"num_tokens": 4614188.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.8349532127380375,
|
|
"epoch": 2.100248550124275,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004789762366502296,
|
|
"loss": 5.5228,
|
|
"mean_token_accuracy": 0.15179868340492247,
|
|
"num_tokens": 4623976.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.823003673553467,
|
|
"epoch": 2.1043910521955262,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004788411543782571,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.1656728833913803,
|
|
"num_tokens": 4632672.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.76104588508606,
|
|
"epoch": 2.108533554266777,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004787056609364008,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.15623851865530014,
|
|
"num_tokens": 4642149.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.675738000869751,
|
|
"epoch": 2.112676056338028,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004785697565979657,
|
|
"loss": 5.3937,
|
|
"mean_token_accuracy": 0.16207751482725144,
|
|
"num_tokens": 4650746.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.702557945251465,
|
|
"epoch": 2.116818558409279,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004784334416370857,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.15734572261571883,
|
|
"num_tokens": 4660154.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.834080362319947,
|
|
"epoch": 2.1209610604805302,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004782967163287231,
|
|
"loss": 5.6194,
|
|
"mean_token_accuracy": 0.1515902802348137,
|
|
"num_tokens": 4668871.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.672147941589356,
|
|
"epoch": 2.1251035625517813,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047815958094866756,
|
|
"loss": 5.5187,
|
|
"mean_token_accuracy": 0.15729935914278032,
|
|
"num_tokens": 4677880.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.909453010559082,
|
|
"epoch": 2.1292460646230325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004780220357735363,
|
|
"loss": 5.7154,
|
|
"mean_token_accuracy": 0.14850281402468682,
|
|
"num_tokens": 4687253.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.834644842147827,
|
|
"epoch": 2.1333885666942836,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047788408108077294,
|
|
"loss": 5.6016,
|
|
"mean_token_accuracy": 0.1510675936937332,
|
|
"num_tokens": 4696392.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.802241468429566,
|
|
"epoch": 2.1375310687655342,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004777457171486471,
|
|
"loss": 5.5617,
|
|
"mean_token_accuracy": 0.1504868522286415,
|
|
"num_tokens": 4704812.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.826993179321289,
|
|
"epoch": 2.1416735708367853,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004776069442562541,
|
|
"loss": 5.5251,
|
|
"mean_token_accuracy": 0.15672111809253692,
|
|
"num_tokens": 4713411.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.7558825492858885,
|
|
"epoch": 2.1458160729080364,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004774677626835138,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.1601279854774475,
|
|
"num_tokens": 4722966.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.761832189559937,
|
|
"epoch": 2.1499585749792876,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047732817271117084,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.14909747391939163,
|
|
"num_tokens": 4731889.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.731906175613403,
|
|
"epoch": 2.1541010770505387,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004771881746207934,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.16156674027442933,
|
|
"num_tokens": 4740231.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.7913134574890135,
|
|
"epoch": 2.1582435791217898,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047704776869477285,
|
|
"loss": 5.6222,
|
|
"mean_token_accuracy": 0.14979231655597686,
|
|
"num_tokens": 4748555.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.844248533248901,
|
|
"epoch": 2.1623860811930404,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047690695521632344,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.1507200427353382,
|
|
"num_tokens": 4757185.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.7017820358276365,
|
|
"epoch": 2.1665285832642915,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047676573446948114,
|
|
"loss": 5.5526,
|
|
"mean_token_accuracy": 0.153288733959198,
|
|
"num_tokens": 4766518.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.749460935592651,
|
|
"epoch": 2.1706710853355426,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047662410673910376,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.15899401307106018,
|
|
"num_tokens": 4776005.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.7959081649780275,
|
|
"epoch": 2.1748135874067938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047648207231087,
|
|
"loss": 5.5733,
|
|
"mean_token_accuracy": 0.15494658946990966,
|
|
"num_tokens": 4785343.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.729545402526855,
|
|
"epoch": 2.178956089478045,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004763396314712786,
|
|
"loss": 5.4946,
|
|
"mean_token_accuracy": 0.16482714265584947,
|
|
"num_tokens": 4794229.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.8528862476348875,
|
|
"epoch": 2.183098591549296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004761967845076484,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.14892114326357841,
|
|
"num_tokens": 4803182.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.812585163116455,
|
|
"epoch": 2.1872410936205466,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004760535317081174,
|
|
"loss": 5.5233,
|
|
"mean_token_accuracy": 0.16473610252141951,
|
|
"num_tokens": 4811820.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.747249031066895,
|
|
"epoch": 2.1913835956917977,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047590987336164207,
|
|
"loss": 5.5679,
|
|
"mean_token_accuracy": 0.15518500953912734,
|
|
"num_tokens": 4821273.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.786758613586426,
|
|
"epoch": 2.195526097763049,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047576580975799705,
|
|
"loss": 5.5832,
|
|
"mean_token_accuracy": 0.15140218138694764,
|
|
"num_tokens": 4830130.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.772381114959717,
|
|
"epoch": 2.1996685998343,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004756213411877743,
|
|
"loss": 5.598,
|
|
"mean_token_accuracy": 0.15639565736055375,
|
|
"num_tokens": 4839238.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.714341497421264,
|
|
"epoch": 2.203811101905551,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.16323678493499755,
|
|
"num_tokens": 4848434.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.684457731246948,
|
|
"epoch": 2.207953603976802,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047533119031404757,
|
|
"loss": 5.5609,
|
|
"mean_token_accuracy": 0.1529301807284355,
|
|
"num_tokens": 4857884.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.672632503509521,
|
|
"epoch": 2.212096106048053,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047518550859580966,
|
|
"loss": 5.4334,
|
|
"mean_token_accuracy": 0.1699458748102188,
|
|
"num_tokens": 4866595.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.8053535461425785,
|
|
"epoch": 2.216238608119304,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004750394230815249,
|
|
"loss": 5.5857,
|
|
"mean_token_accuracy": 0.148395074903965,
|
|
"num_tokens": 4875573.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.796173238754273,
|
|
"epoch": 2.220381110190555,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047489293406586375,
|
|
"loss": 5.5903,
|
|
"mean_token_accuracy": 0.1620958462357521,
|
|
"num_tokens": 4884977.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.723018169403076,
|
|
"epoch": 2.224523612261806,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004747460418443106,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.15953139066696168,
|
|
"num_tokens": 4894091.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.637625503540039,
|
|
"epoch": 2.2286661143330573,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047459874671316305,
|
|
"loss": 5.4194,
|
|
"mean_token_accuracy": 0.16242783069610595,
|
|
"num_tokens": 4903059.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.641981792449951,
|
|
"epoch": 2.2328086164043084,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047445104896953153,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.17116804122924806,
|
|
"num_tokens": 4911969.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.838920593261719,
|
|
"epoch": 2.236951118475559,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047430294891133846,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.15505664348602294,
|
|
"num_tokens": 4920915.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.79048752784729,
|
|
"epoch": 2.24109362054681,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004741544468373179,
|
|
"loss": 5.5957,
|
|
"mean_token_accuracy": 0.15736506879329681,
|
|
"num_tokens": 4931048.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.746155166625977,
|
|
"epoch": 2.2452361226180613,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047400554304701474,
|
|
"loss": 5.5751,
|
|
"mean_token_accuracy": 0.15703048855066298,
|
|
"num_tokens": 4939483.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.6905131340026855,
|
|
"epoch": 2.2493786246893124,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047385623784078427,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.15871881544589997,
|
|
"num_tokens": 4948888.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.770149040222168,
|
|
"epoch": 2.2535211267605635,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004737065315197913,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.15765973925590515,
|
|
"num_tokens": 4956950.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.635520935058594,
|
|
"epoch": 2.2576636288318146,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047355642438600987,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.15296801626682283,
|
|
"num_tokens": 4965968.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.621658706665039,
|
|
"epoch": 2.2618061309030653,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004734059167422225,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.16240133494138717,
|
|
"num_tokens": 4975742.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.73398118019104,
|
|
"epoch": 2.2659486329743164,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047325500889201956,
|
|
"loss": 5.586,
|
|
"mean_token_accuracy": 0.15448082536458968,
|
|
"num_tokens": 4984991.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.708168172836304,
|
|
"epoch": 2.2700911350455675,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004731037011397986,
|
|
"loss": 5.5017,
|
|
"mean_token_accuracy": 0.15908219516277314,
|
|
"num_tokens": 4993574.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.700465059280395,
|
|
"epoch": 2.2742336371168186,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047295199379076403,
|
|
"loss": 5.5119,
|
|
"mean_token_accuracy": 0.16220880597829818,
|
|
"num_tokens": 5001905.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.812096071243286,
|
|
"epoch": 2.2783761391880697,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000472799887150926,
|
|
"loss": 5.6517,
|
|
"mean_token_accuracy": 0.1502576395869255,
|
|
"num_tokens": 5010554.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.749092102050781,
|
|
"epoch": 2.282518641259321,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004726473815271003,
|
|
"loss": 5.5311,
|
|
"mean_token_accuracy": 0.1561155468225479,
|
|
"num_tokens": 5019678.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.719474124908447,
|
|
"epoch": 2.286661143330572,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047249447722690747,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.15179549306631088,
|
|
"num_tokens": 5028180.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.747410917282105,
|
|
"epoch": 2.2908036454018226,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047234117455877215,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.15653812736272812,
|
|
"num_tokens": 5037083.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.838072156906128,
|
|
"epoch": 2.2949461474730737,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004721874738319225,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.15274079591035844,
|
|
"num_tokens": 5046666.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.671936702728272,
|
|
"epoch": 2.299088649544325,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004720333753563898,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.15928964912891388,
|
|
"num_tokens": 5055053.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.686665153503418,
|
|
"epoch": 2.303231151615576,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047187887944300747,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.1539557047188282,
|
|
"num_tokens": 5064049.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.795013904571533,
|
|
"epoch": 2.307373653686827,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004717239864034107,
|
|
"loss": 5.6388,
|
|
"mean_token_accuracy": 0.15225408375263214,
|
|
"num_tokens": 5072420.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.701117706298828,
|
|
"epoch": 2.3115161557580777,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047156869655003554,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.1541515976190567,
|
|
"num_tokens": 5081581.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.693496513366699,
|
|
"epoch": 2.315658657829329,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047141301019611875,
|
|
"loss": 5.5012,
|
|
"mean_token_accuracy": 0.15954914838075637,
|
|
"num_tokens": 5090204.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.7152317523956295,
|
|
"epoch": 2.31980115990058,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004712569276556967,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.15084023624658585,
|
|
"num_tokens": 5099558.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.662888717651367,
|
|
"epoch": 2.323943661971831,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047110044924360484,
|
|
"loss": 5.4811,
|
|
"mean_token_accuracy": 0.16062808632850648,
|
|
"num_tokens": 5107813.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.802162504196167,
|
|
"epoch": 2.328086164043082,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047094357527547725,
|
|
"loss": 5.5867,
|
|
"mean_token_accuracy": 0.15557624846696855,
|
|
"num_tokens": 5117154.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.757091856002807,
|
|
"epoch": 2.332228666114333,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004707863060677459,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.15680305510759354,
|
|
"num_tokens": 5127823.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.655477809906006,
|
|
"epoch": 2.3363711681855843,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047062864193764,
|
|
"loss": 5.5422,
|
|
"mean_token_accuracy": 0.1533115178346634,
|
|
"num_tokens": 5135903.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.664518690109253,
|
|
"epoch": 2.340513670256835,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047047058320318525,
|
|
"loss": 5.4854,
|
|
"mean_token_accuracy": 0.15575041621923447,
|
|
"num_tokens": 5145657.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.735196256637574,
|
|
"epoch": 2.344656172328086,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004703121301832034,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.15849591195583343,
|
|
"num_tokens": 5155061.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.7144616603851315,
|
|
"epoch": 2.348798674399337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047015328319731153,
|
|
"loss": 5.4712,
|
|
"mean_token_accuracy": 0.1625610962510109,
|
|
"num_tokens": 5164302.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.6743780136108395,
|
|
"epoch": 2.3529411764705883,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004699940425659215,
|
|
"loss": 5.6349,
|
|
"mean_token_accuracy": 0.1537710279226303,
|
|
"num_tokens": 5173469.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.824509334564209,
|
|
"epoch": 2.3570836785418394,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046983440861023877,
|
|
"loss": 5.6457,
|
|
"mean_token_accuracy": 0.15466391295194626,
|
|
"num_tokens": 5182548.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.728637599945069,
|
|
"epoch": 2.36122618061309,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046967438165226275,
|
|
"loss": 5.6148,
|
|
"mean_token_accuracy": 0.14389563351869583,
|
|
"num_tokens": 5191802.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.807513904571533,
|
|
"epoch": 2.365368682684341,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004695139620147851,
|
|
"loss": 5.7199,
|
|
"mean_token_accuracy": 0.15426884219050407,
|
|
"num_tokens": 5201986.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.727898979187012,
|
|
"epoch": 2.3695111847555923,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046935315002138994,
|
|
"loss": 5.5067,
|
|
"mean_token_accuracy": 0.1532989665865898,
|
|
"num_tokens": 5211041.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.699389410018921,
|
|
"epoch": 2.3736536868268434,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004691919459964525,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.15786828845739365,
|
|
"num_tokens": 5219838.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.727614879608154,
|
|
"epoch": 2.3777961888980945,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046903035026513894,
|
|
"loss": 5.528,
|
|
"mean_token_accuracy": 0.1572413668036461,
|
|
"num_tokens": 5228282.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.671488904953003,
|
|
"epoch": 2.3819386909693456,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004688683631534056,
|
|
"loss": 5.5662,
|
|
"mean_token_accuracy": 0.15757867842912673,
|
|
"num_tokens": 5238038.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.707324075698852,
|
|
"epoch": 2.3860811930405967,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004687059849879981,
|
|
"loss": 5.5565,
|
|
"mean_token_accuracy": 0.15727738589048385,
|
|
"num_tokens": 5247084.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.7053502082824705,
|
|
"epoch": 2.3902236951118474,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046854321609645093,
|
|
"loss": 5.5825,
|
|
"mean_token_accuracy": 0.15503206253051757,
|
|
"num_tokens": 5256630.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.772836065292358,
|
|
"epoch": 2.3943661971830985,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046838005680708677,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.15751295685768127,
|
|
"num_tokens": 5266854.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.6266319274902346,
|
|
"epoch": 2.3985086992543496,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046821650744901573,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.15903597474098205,
|
|
"num_tokens": 5275823.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.7099535942077635,
|
|
"epoch": 2.4026512013256007,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004680525683521347,
|
|
"loss": 5.5698,
|
|
"mean_token_accuracy": 0.15646034926176072,
|
|
"num_tokens": 5285190.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.8103772640228275,
|
|
"epoch": 2.406793703396852,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004678882398471269,
|
|
"loss": 5.523,
|
|
"mean_token_accuracy": 0.15891006737947463,
|
|
"num_tokens": 5293832.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.590553045272827,
|
|
"epoch": 2.4109362054681025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046772352226546066,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.15144434422254563,
|
|
"num_tokens": 5303514.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.71599235534668,
|
|
"epoch": 2.4150787075393536,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004675584159393895,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.15650684982538224,
|
|
"num_tokens": 5313361.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.752013874053955,
|
|
"epoch": 2.4192212096106047,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046739292120195083,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.1579284265637398,
|
|
"num_tokens": 5323556.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.751535844802857,
|
|
"epoch": 2.423363711681856,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004672270383869657,
|
|
"loss": 5.5814,
|
|
"mean_token_accuracy": 0.15038304030895233,
|
|
"num_tokens": 5333381.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.708028030395508,
|
|
"epoch": 2.427506213753107,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004670607678290379,
|
|
"loss": 5.557,
|
|
"mean_token_accuracy": 0.15642340034246444,
|
|
"num_tokens": 5343722.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.709355688095092,
|
|
"epoch": 2.431648715824358,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004668941098635532,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.16031453162431716,
|
|
"num_tokens": 5352680.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.697698545455933,
|
|
"epoch": 2.435791217895609,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046672706482667897,
|
|
"loss": 5.5378,
|
|
"mean_token_accuracy": 0.15933478623628616,
|
|
"num_tokens": 5361838.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.6182372093200685,
|
|
"epoch": 2.43993371996686,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004665596330553633,
|
|
"loss": 5.5097,
|
|
"mean_token_accuracy": 0.16193565130233764,
|
|
"num_tokens": 5370806.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.677807283401489,
|
|
"epoch": 2.444076222038111,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046639181488733445,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.16008703261613846,
|
|
"num_tokens": 5379690.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.696961450576782,
|
|
"epoch": 2.448218724109362,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004662236106610999,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.15375241935253142,
|
|
"num_tokens": 5388266.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.676030540466309,
|
|
"epoch": 2.452361226180613,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046605502071594604,
|
|
"loss": 5.5166,
|
|
"mean_token_accuracy": 0.15484256520867348,
|
|
"num_tokens": 5396202.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.603984308242798,
|
|
"epoch": 2.4565037282518642,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004658860453919371,
|
|
"loss": 5.4235,
|
|
"mean_token_accuracy": 0.16309440582990647,
|
|
"num_tokens": 5405265.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.765998888015747,
|
|
"epoch": 2.460646230323115,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004657166850299149,
|
|
"loss": 5.6232,
|
|
"mean_token_accuracy": 0.1521397665143013,
|
|
"num_tokens": 5414862.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.677093935012818,
|
|
"epoch": 2.464788732394366,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004655469399714978,
|
|
"loss": 5.5603,
|
|
"mean_token_accuracy": 0.16458463966846465,
|
|
"num_tokens": 5424065.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.699638605117798,
|
|
"epoch": 2.468931234465617,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004653768105590799,
|
|
"loss": 5.5269,
|
|
"mean_token_accuracy": 0.16183326244354249,
|
|
"num_tokens": 5433598.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.756310033798218,
|
|
"epoch": 2.4730737365368682,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046520629713583115,
|
|
"loss": 5.6508,
|
|
"mean_token_accuracy": 0.15125053226947785,
|
|
"num_tokens": 5443370.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.742058801651001,
|
|
"epoch": 2.4772162386081193,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046503540004569555,
|
|
"loss": 5.5286,
|
|
"mean_token_accuracy": 0.15547680854797363,
|
|
"num_tokens": 5452513.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.683076810836792,
|
|
"epoch": 2.4813587406793705,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046486411963339125,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.15302325487136842,
|
|
"num_tokens": 5461212.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.726020097732544,
|
|
"epoch": 2.4855012427506216,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046469245624440955,
|
|
"loss": 5.5638,
|
|
"mean_token_accuracy": 0.1646001785993576,
|
|
"num_tokens": 5470450.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.4855012427506216,
|
|
"eval_entropy": 5.666851568323717,
|
|
"eval_loss": 5.997962474822998,
|
|
"eval_mean_token_accuracy": 0.1460237345620513,
|
|
"eval_num_tokens": 5470450.0,
|
|
"eval_runtime": 2.8811,
|
|
"eval_samples_per_second": 1300.547,
|
|
"eval_steps_per_second": 162.785,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.7583527088165285,
|
|
"epoch": 2.4896437448218722,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046452041022501423,
|
|
"loss": 5.5164,
|
|
"mean_token_accuracy": 0.16106657758355142,
|
|
"num_tokens": 5478839.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.73330397605896,
|
|
"epoch": 2.4937862468931233,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004643479819222411,
|
|
"loss": 5.5609,
|
|
"mean_token_accuracy": 0.15447363406419753,
|
|
"num_tokens": 5487935.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.752573728561401,
|
|
"epoch": 2.4979287489643744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046417517168389664,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.15475835651159286,
|
|
"num_tokens": 5496369.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.710302925109863,
|
|
"epoch": 2.5020712510356256,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004640019798585581,
|
|
"loss": 5.5703,
|
|
"mean_token_accuracy": 0.1553514450788498,
|
|
"num_tokens": 5504263.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.642459678649902,
|
|
"epoch": 2.5062137531068767,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046382840679557233,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.1662412628531456,
|
|
"num_tokens": 5513164.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.786639738082886,
|
|
"epoch": 2.5103562551781273,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046365445284505516,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.15565225183963777,
|
|
"num_tokens": 5522118.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.604031181335449,
|
|
"epoch": 2.5144987572493784,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004634801183578907,
|
|
"loss": 5.5399,
|
|
"mean_token_accuracy": 0.1582293838262558,
|
|
"num_tokens": 5531305.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.727992343902588,
|
|
"epoch": 2.5186412593206295,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046330540368573063,
|
|
"loss": 5.5667,
|
|
"mean_token_accuracy": 0.15120545104146005,
|
|
"num_tokens": 5539474.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.672655773162842,
|
|
"epoch": 2.5227837613918807,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046313030918099365,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.1591261774301529,
|
|
"num_tokens": 5549114.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.6598734855651855,
|
|
"epoch": 2.5269262634631318,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046295483519686437,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.15746332854032516,
|
|
"num_tokens": 5558118.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.661467361450195,
|
|
"epoch": 2.531068765534383,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046277898208729306,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.15640463680028915,
|
|
"num_tokens": 5567635.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.769457387924194,
|
|
"epoch": 2.535211267605634,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004626027502069947,
|
|
"loss": 5.6279,
|
|
"mean_token_accuracy": 0.1561012625694275,
|
|
"num_tokens": 5577272.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.770507669448852,
|
|
"epoch": 2.539353769676885,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046242613991144814,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.16126794517040252,
|
|
"num_tokens": 5586227.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.682786655426026,
|
|
"epoch": 2.5434962717481358,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046224915155689586,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.15951058715581895,
|
|
"num_tokens": 5595648.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.614820337295532,
|
|
"epoch": 2.547638773819387,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046207178550034244,
|
|
"loss": 5.4344,
|
|
"mean_token_accuracy": 0.17111728340387344,
|
|
"num_tokens": 5604187.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.622501564025879,
|
|
"epoch": 2.551781275890638,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004618940420995547,
|
|
"loss": 5.527,
|
|
"mean_token_accuracy": 0.15877681821584702,
|
|
"num_tokens": 5612850.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.678854370117188,
|
|
"epoch": 2.555923777961889,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004617159217130606,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.15891224294900894,
|
|
"num_tokens": 5622902.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.677180290222168,
|
|
"epoch": 2.5600662800331397,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004615374247001483,
|
|
"loss": 5.5319,
|
|
"mean_token_accuracy": 0.14990963488817216,
|
|
"num_tokens": 5631854.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.753375482559204,
|
|
"epoch": 2.564208782104391,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004613585514208659,
|
|
"loss": 5.5185,
|
|
"mean_token_accuracy": 0.16782330721616745,
|
|
"num_tokens": 5640237.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.8014825820922855,
|
|
"epoch": 2.568351284175642,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046117930223602016,
|
|
"loss": 5.6747,
|
|
"mean_token_accuracy": 0.15266044735908507,
|
|
"num_tokens": 5650182.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.6702837467193605,
|
|
"epoch": 2.572493786246893,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004609996775071765,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.1625978335738182,
|
|
"num_tokens": 5659447.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.664971351623535,
|
|
"epoch": 2.576636288318144,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004608196775966575,
|
|
"loss": 5.5933,
|
|
"mean_token_accuracy": 0.1530801922082901,
|
|
"num_tokens": 5668349.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.830552959442139,
|
|
"epoch": 2.5807787903893953,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046063930286754267,
|
|
"loss": 5.6506,
|
|
"mean_token_accuracy": 0.1469719246029854,
|
|
"num_tokens": 5677296.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.737857675552368,
|
|
"epoch": 2.5849212924606464,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046045855368366756,
|
|
"loss": 5.5505,
|
|
"mean_token_accuracy": 0.15921184718608855,
|
|
"num_tokens": 5685822.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.654632902145385,
|
|
"epoch": 2.5890637945318975,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000460277430409623,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.15037565529346467,
|
|
"num_tokens": 5695171.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.782729053497315,
|
|
"epoch": 2.593206296603148,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046009593341075444,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.15687357783317565,
|
|
"num_tokens": 5704559.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.628320407867432,
|
|
"epoch": 2.5973487986743993,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004599140630531612,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.16054383665323257,
|
|
"num_tokens": 5713690.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.740268182754517,
|
|
"epoch": 2.6014913007456504,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004597318197036956,
|
|
"loss": 5.7474,
|
|
"mean_token_accuracy": 0.15208713933825493,
|
|
"num_tokens": 5723792.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.785957098007202,
|
|
"epoch": 2.6056338028169015,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004595492037299624,
|
|
"loss": 5.5389,
|
|
"mean_token_accuracy": 0.16184956580400467,
|
|
"num_tokens": 5732460.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.719061231613159,
|
|
"epoch": 2.6097763048881526,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045936621550031806,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.15228736773133278,
|
|
"num_tokens": 5741770.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.731397724151611,
|
|
"epoch": 2.6139188069594033,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045918285538386973,
|
|
"loss": 5.5834,
|
|
"mean_token_accuracy": 0.15751861929893493,
|
|
"num_tokens": 5750821.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.670954942703247,
|
|
"epoch": 2.6180613090306544,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045899912375047485,
|
|
"loss": 5.5937,
|
|
"mean_token_accuracy": 0.15194432139396669,
|
|
"num_tokens": 5759792.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.701627111434936,
|
|
"epoch": 2.6222038111019055,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045881502097074023,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.15721253752708436,
|
|
"num_tokens": 5769650.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.726138496398926,
|
|
"epoch": 2.6263463131731566,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004586305474160212,
|
|
"loss": 5.592,
|
|
"mean_token_accuracy": 0.1525888927280903,
|
|
"num_tokens": 5779027.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.653416013717651,
|
|
"epoch": 2.6304888152444077,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045844570345842116,
|
|
"loss": 5.5616,
|
|
"mean_token_accuracy": 0.15808800384402275,
|
|
"num_tokens": 5787860.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.8046799659729,
|
|
"epoch": 2.634631317315659,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004582604894707906,
|
|
"loss": 5.6338,
|
|
"mean_token_accuracy": 0.15353483557701111,
|
|
"num_tokens": 5798211.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.68250994682312,
|
|
"epoch": 2.63877381938691,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045807490582672634,
|
|
"loss": 5.5907,
|
|
"mean_token_accuracy": 0.1567721262574196,
|
|
"num_tokens": 5807484.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.7200438499450685,
|
|
"epoch": 2.6429163214581606,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045788895290057083,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.1579717665910721,
|
|
"num_tokens": 5815969.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.740452289581299,
|
|
"epoch": 2.6470588235294117,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045770263106741137,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.15720631256699563,
|
|
"num_tokens": 5825193.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.695869445800781,
|
|
"epoch": 2.651201325600663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004575159407030797,
|
|
"loss": 5.5764,
|
|
"mean_token_accuracy": 0.15763850808143615,
|
|
"num_tokens": 5833628.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.713213205337524,
|
|
"epoch": 2.655343827671914,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004573288821841505,
|
|
"loss": 5.6378,
|
|
"mean_token_accuracy": 0.15480245947837828,
|
|
"num_tokens": 5841562.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.809486722946167,
|
|
"epoch": 2.659486329743165,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004571414558879413,
|
|
"loss": 5.6349,
|
|
"mean_token_accuracy": 0.14864409267902373,
|
|
"num_tokens": 5851145.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.747229051589966,
|
|
"epoch": 2.6636288318144157,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004569536621925114,
|
|
"loss": 5.5888,
|
|
"mean_token_accuracy": 0.15000322833657265,
|
|
"num_tokens": 5860694.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.612951326370239,
|
|
"epoch": 2.667771333885667,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004567655014766612,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.15544012635946275,
|
|
"num_tokens": 5870143.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.695223188400268,
|
|
"epoch": 2.671913835956918,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004565769741199314,
|
|
"loss": 5.5615,
|
|
"mean_token_accuracy": 0.14824787601828576,
|
|
"num_tokens": 5878872.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.685911417007446,
|
|
"epoch": 2.676056338028169,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045638808050260237,
|
|
"loss": 5.5463,
|
|
"mean_token_accuracy": 0.15454385727643966,
|
|
"num_tokens": 5887713.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.738532686233521,
|
|
"epoch": 2.68019884009942,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045619882100569317,
|
|
"loss": 5.5947,
|
|
"mean_token_accuracy": 0.15959140807390212,
|
|
"num_tokens": 5896268.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.7656388759613035,
|
|
"epoch": 2.684341342170671,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004560091960109608,
|
|
"loss": 5.5598,
|
|
"mean_token_accuracy": 0.15922954678535461,
|
|
"num_tokens": 5905584.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.7327958106994625,
|
|
"epoch": 2.6884838442419223,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004558192059008996,
|
|
"loss": 5.6933,
|
|
"mean_token_accuracy": 0.14835613816976548,
|
|
"num_tokens": 5916636.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.779455757141113,
|
|
"epoch": 2.692626346313173,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004556288510587406,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.1529821887612343,
|
|
"num_tokens": 5925629.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.792713212966919,
|
|
"epoch": 2.696768848384424,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045543813186845004,
|
|
"loss": 5.6845,
|
|
"mean_token_accuracy": 0.1453479304909706,
|
|
"num_tokens": 5935316.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.700288534164429,
|
|
"epoch": 2.700911350455675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004552470487147294,
|
|
"loss": 5.5887,
|
|
"mean_token_accuracy": 0.1574153408408165,
|
|
"num_tokens": 5944936.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.704364824295044,
|
|
"epoch": 2.7050538525269263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045505560198301443,
|
|
"loss": 5.5649,
|
|
"mean_token_accuracy": 0.1547586739063263,
|
|
"num_tokens": 5954106.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.689199924468994,
|
|
"epoch": 2.7091963545981774,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045486379205947396,
|
|
"loss": 5.5079,
|
|
"mean_token_accuracy": 0.162031652033329,
|
|
"num_tokens": 5963015.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.812100887298584,
|
|
"epoch": 2.713338856669428,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004546716193310098,
|
|
"loss": 5.711,
|
|
"mean_token_accuracy": 0.1566604346036911,
|
|
"num_tokens": 5972444.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.654112815856934,
|
|
"epoch": 2.717481358740679,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000454479084185255,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.1566864624619484,
|
|
"num_tokens": 5981397.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.636119270324707,
|
|
"epoch": 2.7216238608119303,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004542861870105743,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.1622664988040924,
|
|
"num_tokens": 5989491.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.739955615997315,
|
|
"epoch": 2.7257663628831814,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045409292819606224,
|
|
"loss": 5.6167,
|
|
"mean_token_accuracy": 0.1547279179096222,
|
|
"num_tokens": 5997586.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.695828151702881,
|
|
"epoch": 2.7299088649544325,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045389930813154297,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.17207208275794983,
|
|
"num_tokens": 6006378.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.671674203872681,
|
|
"epoch": 2.7340513670256836,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004537053272075693,
|
|
"loss": 5.5369,
|
|
"mean_token_accuracy": 0.159752693772316,
|
|
"num_tokens": 6016058.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.618702602386475,
|
|
"epoch": 2.7381938690969347,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045351098581542204,
|
|
"loss": 5.5509,
|
|
"mean_token_accuracy": 0.16341104060411454,
|
|
"num_tokens": 6025315.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.689493608474732,
|
|
"epoch": 2.742336371168186,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004533162843471089,
|
|
"loss": 5.5401,
|
|
"mean_token_accuracy": 0.15370434224605561,
|
|
"num_tokens": 6035048.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.669504976272583,
|
|
"epoch": 2.7464788732394365,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004531212231953641,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.15908312052488327,
|
|
"num_tokens": 6043299.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.782401418685913,
|
|
"epoch": 2.7506213753106876,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004529258027536473,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.15550746768712997,
|
|
"num_tokens": 6052396.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.726404809951783,
|
|
"epoch": 2.7547638773819387,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004527300234161428,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.15855898410081865,
|
|
"num_tokens": 6060491.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.562553119659424,
|
|
"epoch": 2.75890637945319,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004525338855777591,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.16163053959608079,
|
|
"num_tokens": 6069770.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.692863273620605,
|
|
"epoch": 2.7630488815244405,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045233738963412746,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.16060031652450563,
|
|
"num_tokens": 6079219.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.6884935855865475,
|
|
"epoch": 2.7671913835956916,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045214053598160173,
|
|
"loss": 5.5913,
|
|
"mean_token_accuracy": 0.15756666362285615,
|
|
"num_tokens": 6088483.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.736902332305908,
|
|
"epoch": 2.7713338856669427,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045194332501725725,
|
|
"loss": 5.6121,
|
|
"mean_token_accuracy": 0.15517180860042573,
|
|
"num_tokens": 6098283.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.713816738128662,
|
|
"epoch": 2.775476387738194,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045174575713889014,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.16259121149778366,
|
|
"num_tokens": 6107657.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.611076307296753,
|
|
"epoch": 2.779618889809445,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004515478327450164,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.17341717928647996,
|
|
"num_tokens": 6116658.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.53763427734375,
|
|
"epoch": 2.783761391880696,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045134955223487093,
|
|
"loss": 5.4839,
|
|
"mean_token_accuracy": 0.16012571901082992,
|
|
"num_tokens": 6125091.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.626507520675659,
|
|
"epoch": 2.787903893951947,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045115091600840757,
|
|
"loss": 5.4702,
|
|
"mean_token_accuracy": 0.16941866427659988,
|
|
"num_tokens": 6133710.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.742727422714234,
|
|
"epoch": 2.7920463960231983,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00045095192446629697,
|
|
"loss": 5.5804,
|
|
"mean_token_accuracy": 0.15803021043539048,
|
|
"num_tokens": 6141986.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.583254098892212,
|
|
"epoch": 2.796188898094449,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000450752578009927,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.16595451682806014,
|
|
"num_tokens": 6150542.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.773436498641968,
|
|
"epoch": 2.8003314001657,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045055287704140115,
|
|
"loss": 5.5513,
|
|
"mean_token_accuracy": 0.1664824068546295,
|
|
"num_tokens": 6158958.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.670917129516601,
|
|
"epoch": 2.804473902236951,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004503528219635381,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.15057535320520402,
|
|
"num_tokens": 6168699.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.750991725921631,
|
|
"epoch": 2.8086164043082023,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004501524131798709,
|
|
"loss": 5.6256,
|
|
"mean_token_accuracy": 0.15223808214068413,
|
|
"num_tokens": 6177472.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.678924322128296,
|
|
"epoch": 2.812758906379453,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000449951651094646,
|
|
"loss": 5.6074,
|
|
"mean_token_accuracy": 0.15856225937604904,
|
|
"num_tokens": 6187098.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.813678789138794,
|
|
"epoch": 2.816901408450704,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044975053611282217,
|
|
"loss": 5.601,
|
|
"mean_token_accuracy": 0.15290533751249313,
|
|
"num_tokens": 6196702.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.672289848327637,
|
|
"epoch": 2.821043910521955,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004495490686400706,
|
|
"loss": 5.5655,
|
|
"mean_token_accuracy": 0.16363370269536973,
|
|
"num_tokens": 6205465.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.612434053421021,
|
|
"epoch": 2.8251864125932062,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044934724908277313,
|
|
"loss": 5.5057,
|
|
"mean_token_accuracy": 0.16250835806131364,
|
|
"num_tokens": 6214967.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.5932745933532715,
|
|
"epoch": 2.8293289146644574,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044914507784802184,
|
|
"loss": 5.3888,
|
|
"mean_token_accuracy": 0.16869354397058486,
|
|
"num_tokens": 6223131.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.679493618011475,
|
|
"epoch": 2.8334714167357085,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004489425553436182,
|
|
"loss": 5.6129,
|
|
"mean_token_accuracy": 0.16268137246370315,
|
|
"num_tokens": 6232404.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.810343885421753,
|
|
"epoch": 2.8376139188069596,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044873968197807213,
|
|
"loss": 5.692,
|
|
"mean_token_accuracy": 0.15346367061138153,
|
|
"num_tokens": 6240666.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.702585172653198,
|
|
"epoch": 2.8417564208782107,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044853645816060156,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.1537062019109726,
|
|
"num_tokens": 6249375.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.691167497634888,
|
|
"epoch": 2.8458989229494613,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044833288430113104,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.16469355970621108,
|
|
"num_tokens": 6259215.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.666302585601807,
|
|
"epoch": 2.8500414250207124,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044812896081029136,
|
|
"loss": 5.5082,
|
|
"mean_token_accuracy": 0.15554405748844147,
|
|
"num_tokens": 6268650.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.616651582717895,
|
|
"epoch": 2.8541839270919636,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004479246880994183,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.1603178933262825,
|
|
"num_tokens": 6277687.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.725367212295533,
|
|
"epoch": 2.8583264291632147,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004477200665805525,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.17011294066905974,
|
|
"num_tokens": 6286940.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.538966131210327,
|
|
"epoch": 2.8624689312344658,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044751509666643777,
|
|
"loss": 5.4376,
|
|
"mean_token_accuracy": 0.16481457650661469,
|
|
"num_tokens": 6295594.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.642479085922242,
|
|
"epoch": 2.8666114333057164,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044730977877052084,
|
|
"loss": 5.5841,
|
|
"mean_token_accuracy": 0.15637488067150115,
|
|
"num_tokens": 6304616.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.710377931594849,
|
|
"epoch": 2.8707539353769675,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044710411330695045,
|
|
"loss": 5.4916,
|
|
"mean_token_accuracy": 0.16140547692775725,
|
|
"num_tokens": 6314449.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.589468860626221,
|
|
"epoch": 2.8748964374482187,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044689810069057615,
|
|
"loss": 5.4644,
|
|
"mean_token_accuracy": 0.1623084232211113,
|
|
"num_tokens": 6323811.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.6096266746521,
|
|
"epoch": 2.8790389395194698,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004466917413369479,
|
|
"loss": 5.4557,
|
|
"mean_token_accuracy": 0.1630224645137787,
|
|
"num_tokens": 6332553.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.631894493103028,
|
|
"epoch": 2.883181441590721,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004464850356623152,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.1654105991125107,
|
|
"num_tokens": 6341934.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.685671377182007,
|
|
"epoch": 2.887323943661972,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044627798408362597,
|
|
"loss": 5.5859,
|
|
"mean_token_accuracy": 0.16682443618774415,
|
|
"num_tokens": 6351032.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.674548244476318,
|
|
"epoch": 2.891466445733223,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004460705870185258,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.15875868052244185,
|
|
"num_tokens": 6360801.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.6504271030426025,
|
|
"epoch": 2.8956089478044738,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004458628448853572,
|
|
"loss": 5.5801,
|
|
"mean_token_accuracy": 0.1501157611608505,
|
|
"num_tokens": 6370359.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.630797243118286,
|
|
"epoch": 2.899751449875725,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004456547581031589,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.16001514345407486,
|
|
"num_tokens": 6379007.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.899751449875725,
|
|
"eval_entropy": 5.561281529570947,
|
|
"eval_loss": 5.929627895355225,
|
|
"eval_mean_token_accuracy": 0.14971610989524867,
|
|
"eval_num_tokens": 6379007.0,
|
|
"eval_runtime": 3.1644,
|
|
"eval_samples_per_second": 1184.11,
|
|
"eval_steps_per_second": 148.211,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.673744583129883,
|
|
"epoch": 2.903893951946976,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004454463270916646,
|
|
"loss": 5.5218,
|
|
"mean_token_accuracy": 0.15965216755867004,
|
|
"num_tokens": 6387494.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.641175699234009,
|
|
"epoch": 2.908036454018227,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004452375522713023,
|
|
"loss": 5.4716,
|
|
"mean_token_accuracy": 0.1585933670401573,
|
|
"num_tokens": 6397182.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.692013549804687,
|
|
"epoch": 2.912178956089478,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004450284340631938,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.15904150605201722,
|
|
"num_tokens": 6407038.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.68268723487854,
|
|
"epoch": 2.916321458160729,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044481897288915334,
|
|
"loss": 5.5973,
|
|
"mean_token_accuracy": 0.15640125796198845,
|
|
"num_tokens": 6416286.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.74196047782898,
|
|
"epoch": 2.92046396023198,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044460916917168697,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.15759674161672593,
|
|
"num_tokens": 6425260.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.644951868057251,
|
|
"epoch": 2.924606462303231,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004443990233339918,
|
|
"loss": 5.6566,
|
|
"mean_token_accuracy": 0.1558603525161743,
|
|
"num_tokens": 6435134.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.651631879806518,
|
|
"epoch": 2.928748964374482,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004441885357999548,
|
|
"loss": 5.4258,
|
|
"mean_token_accuracy": 0.1595441997051239,
|
|
"num_tokens": 6445041.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.561243867874145,
|
|
"epoch": 2.9328914664457333,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004439777069941524,
|
|
"loss": 5.5184,
|
|
"mean_token_accuracy": 0.155023792386055,
|
|
"num_tokens": 6454633.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.6850364208221436,
|
|
"epoch": 2.9370339685169844,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044376653734184935,
|
|
"loss": 5.483,
|
|
"mean_token_accuracy": 0.16251987144351004,
|
|
"num_tokens": 6463268.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.661768388748169,
|
|
"epoch": 2.9411764705882355,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004435550272689979,
|
|
"loss": 5.657,
|
|
"mean_token_accuracy": 0.1528746098279953,
|
|
"num_tokens": 6472157.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.6497681617736815,
|
|
"epoch": 2.945318972659486,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044334317720223706,
|
|
"loss": 5.4755,
|
|
"mean_token_accuracy": 0.1629299998283386,
|
|
"num_tokens": 6481654.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.686532402038575,
|
|
"epoch": 2.9494614747307373,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044313098756889154,
|
|
"loss": 5.6147,
|
|
"mean_token_accuracy": 0.15635916143655776,
|
|
"num_tokens": 6491159.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.639078664779663,
|
|
"epoch": 2.9536039768019884,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004429184587969708,
|
|
"loss": 5.5902,
|
|
"mean_token_accuracy": 0.15843436419963836,
|
|
"num_tokens": 6501058.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.714043283462525,
|
|
"epoch": 2.9577464788732395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004427055913151689,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.16188932508230208,
|
|
"num_tokens": 6509708.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.681045627593994,
|
|
"epoch": 2.9618889809444906,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004424923855528626,
|
|
"loss": 5.5143,
|
|
"mean_token_accuracy": 0.1600479558110237,
|
|
"num_tokens": 6518409.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.741547775268555,
|
|
"epoch": 2.9660314830157413,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004422788419401114,
|
|
"loss": 5.5399,
|
|
"mean_token_accuracy": 0.16325889825820922,
|
|
"num_tokens": 6527541.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.546043062210083,
|
|
"epoch": 2.9701739850869924,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044206496090765595,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.1612915128469467,
|
|
"num_tokens": 6536554.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.697825860977173,
|
|
"epoch": 2.9743164871582435,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044185074288691776,
|
|
"loss": 5.6235,
|
|
"mean_token_accuracy": 0.1581455260515213,
|
|
"num_tokens": 6545800.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.63113842010498,
|
|
"epoch": 2.9784589892294946,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044163618830999794,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.16257798224687575,
|
|
"num_tokens": 6555204.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.639696884155273,
|
|
"epoch": 2.9826014913007457,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044142129760967643,
|
|
"loss": 5.4474,
|
|
"mean_token_accuracy": 0.16219695657491684,
|
|
"num_tokens": 6563846.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.677906131744384,
|
|
"epoch": 2.986743993371997,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004412060712194115,
|
|
"loss": 5.594,
|
|
"mean_token_accuracy": 0.15555257946252823,
|
|
"num_tokens": 6572994.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.587683773040771,
|
|
"epoch": 2.990886495443248,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004409905095733381,
|
|
"loss": 5.4766,
|
|
"mean_token_accuracy": 0.16045906990766526,
|
|
"num_tokens": 6581623.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.626201486587524,
|
|
"epoch": 2.995028997514499,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044077461310626764,
|
|
"loss": 5.4382,
|
|
"mean_token_accuracy": 0.16644880473613738,
|
|
"num_tokens": 6590414.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.629630661010742,
|
|
"epoch": 2.9991714995857497,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000440558382253687,
|
|
"loss": 5.4751,
|
|
"mean_token_accuracy": 0.16550374627113343,
|
|
"num_tokens": 6598974.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.55636076927185,
|
|
"epoch": 3.003314001657001,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004403418174517573,
|
|
"loss": 5.1724,
|
|
"mean_token_accuracy": 0.17068159580230713,
|
|
"num_tokens": 6606661.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.632587480545044,
|
|
"epoch": 3.007456503728252,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044012491913731353,
|
|
"loss": 5.304,
|
|
"mean_token_accuracy": 0.17053095549345015,
|
|
"num_tokens": 6616350.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.65787262916565,
|
|
"epoch": 3.011599005799503,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004399076877478632,
|
|
"loss": 5.1672,
|
|
"mean_token_accuracy": 0.180498468875885,
|
|
"num_tokens": 6625643.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.586923599243164,
|
|
"epoch": 3.015741507870754,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004396901237215858,
|
|
"loss": 5.2091,
|
|
"mean_token_accuracy": 0.17293088585138322,
|
|
"num_tokens": 6634785.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.566254043579102,
|
|
"epoch": 3.019884009942005,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043947222749733173,
|
|
"loss": 5.1796,
|
|
"mean_token_accuracy": 0.18026235550642014,
|
|
"num_tokens": 6643692.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.629177045822144,
|
|
"epoch": 3.024026512013256,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043925399951462153,
|
|
"loss": 5.1909,
|
|
"mean_token_accuracy": 0.17716596275568008,
|
|
"num_tokens": 6653739.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.533583068847657,
|
|
"epoch": 3.028169014084507,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004390354402136449,
|
|
"loss": 5.1198,
|
|
"mean_token_accuracy": 0.17737226486206054,
|
|
"num_tokens": 6662411.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.5834307193756105,
|
|
"epoch": 3.032311516155758,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004388165500352597,
|
|
"loss": 5.2414,
|
|
"mean_token_accuracy": 0.1722160130739212,
|
|
"num_tokens": 6671242.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.554705476760864,
|
|
"epoch": 3.036454018227009,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004385973294209915,
|
|
"loss": 5.1954,
|
|
"mean_token_accuracy": 0.1800830751657486,
|
|
"num_tokens": 6679953.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.609169435501099,
|
|
"epoch": 3.0405965202982603,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004383777788130321,
|
|
"loss": 5.2523,
|
|
"mean_token_accuracy": 0.1689423382282257,
|
|
"num_tokens": 6689134.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.609565591812133,
|
|
"epoch": 3.044739022369511,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004381578986542393,
|
|
"loss": 5.1814,
|
|
"mean_token_accuracy": 0.17327553629875184,
|
|
"num_tokens": 6698653.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.563303661346436,
|
|
"epoch": 3.048881524440762,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004379376893881352,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.17461866289377212,
|
|
"num_tokens": 6708462.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.626731443405151,
|
|
"epoch": 3.053024026512013,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004377171514589059,
|
|
"loss": 5.1954,
|
|
"mean_token_accuracy": 0.17984426617622376,
|
|
"num_tokens": 6717731.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.540272855758667,
|
|
"epoch": 3.0571665285832643,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004374962853114007,
|
|
"loss": 5.1453,
|
|
"mean_token_accuracy": 0.173565112054348,
|
|
"num_tokens": 6725896.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.51949372291565,
|
|
"epoch": 3.0613090306545154,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004372750913911306,
|
|
"loss": 5.1791,
|
|
"mean_token_accuracy": 0.17772143930196763,
|
|
"num_tokens": 6734855.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.581039810180664,
|
|
"epoch": 3.0654515327257665,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004370535701442679,
|
|
"loss": 5.2207,
|
|
"mean_token_accuracy": 0.17997916638851166,
|
|
"num_tokens": 6744728.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.580936050415039,
|
|
"epoch": 3.069594034797017,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043683172201764537,
|
|
"loss": 5.2783,
|
|
"mean_token_accuracy": 0.17166770845651627,
|
|
"num_tokens": 6754061.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.5755539894104,
|
|
"epoch": 3.0737365368682683,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004366095474587547,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.17523306757211685,
|
|
"num_tokens": 6763252.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.567341041564942,
|
|
"epoch": 3.0778790389395194,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00043638704691574646,
|
|
"loss": 5.2062,
|
|
"mean_token_accuracy": 0.17794869095087051,
|
|
"num_tokens": 6771270.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.538587808609009,
|
|
"epoch": 3.0820215410107705,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004361642208374283,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.17380594164133073,
|
|
"num_tokens": 6780923.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.592910575866699,
|
|
"epoch": 3.0861640430820216,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043594106967326495,
|
|
"loss": 5.2442,
|
|
"mean_token_accuracy": 0.17527928948402405,
|
|
"num_tokens": 6789353.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.583430051803589,
|
|
"epoch": 3.0903065451532727,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043571759387337665,
|
|
"loss": 5.1917,
|
|
"mean_token_accuracy": 0.17370658963918686,
|
|
"num_tokens": 6798869.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.605370712280274,
|
|
"epoch": 3.0944490472245234,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004354937938885385,
|
|
"loss": 5.2779,
|
|
"mean_token_accuracy": 0.1647024318575859,
|
|
"num_tokens": 6808282.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.5550957202911375,
|
|
"epoch": 3.0985915492957745,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004352696701701796,
|
|
"loss": 5.1816,
|
|
"mean_token_accuracy": 0.1741710990667343,
|
|
"num_tokens": 6816834.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.479378890991211,
|
|
"epoch": 3.1027340513670256,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004350452231703819,
|
|
"loss": 5.1558,
|
|
"mean_token_accuracy": 0.17557668685913086,
|
|
"num_tokens": 6826018.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.599919939041138,
|
|
"epoch": 3.1068765534382767,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043482045334187954,
|
|
"loss": 5.2131,
|
|
"mean_token_accuracy": 0.17932950854301452,
|
|
"num_tokens": 6834932.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.567425918579102,
|
|
"epoch": 3.111019055509528,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00043459536113805785,
|
|
"loss": 5.2454,
|
|
"mean_token_accuracy": 0.1763409674167633,
|
|
"num_tokens": 6843850.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.496420049667359,
|
|
"epoch": 3.115161557580779,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043436994701295237,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.18245080709457398,
|
|
"num_tokens": 6853062.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.580317163467408,
|
|
"epoch": 3.1193040596520296,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043414421142124804,
|
|
"loss": 5.2782,
|
|
"mean_token_accuracy": 0.17598223090171813,
|
|
"num_tokens": 6861939.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.483883619308472,
|
|
"epoch": 3.1234465617232807,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043391815481827814,
|
|
"loss": 5.1412,
|
|
"mean_token_accuracy": 0.18130695074796677,
|
|
"num_tokens": 6870581.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.444404983520508,
|
|
"epoch": 3.127589063794532,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004336917776600237,
|
|
"loss": 5.1905,
|
|
"mean_token_accuracy": 0.17692520171403886,
|
|
"num_tokens": 6879077.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.583776378631592,
|
|
"epoch": 3.131731565865783,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004334650804031122,
|
|
"loss": 5.2334,
|
|
"mean_token_accuracy": 0.1751287415623665,
|
|
"num_tokens": 6889455.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.554486656188965,
|
|
"epoch": 3.135874067937034,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004332380635048165,
|
|
"loss": 5.2059,
|
|
"mean_token_accuracy": 0.17314982861280442,
|
|
"num_tokens": 6898011.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.556101560592651,
|
|
"epoch": 3.140016570008285,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043301072742305474,
|
|
"loss": 5.2021,
|
|
"mean_token_accuracy": 0.17413138449192048,
|
|
"num_tokens": 6906926.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.515396022796631,
|
|
"epoch": 3.1441590720795363,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004327830726163886,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.1852025032043457,
|
|
"num_tokens": 6915036.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.499240732192993,
|
|
"epoch": 3.148301574150787,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004325550995440226,
|
|
"loss": 5.1223,
|
|
"mean_token_accuracy": 0.18250784426927566,
|
|
"num_tokens": 6923692.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.536622667312622,
|
|
"epoch": 3.152444076222038,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043232680866580356,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.17574603259563445,
|
|
"num_tokens": 6932371.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.55640172958374,
|
|
"epoch": 3.156586578293289,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043209820044221893,
|
|
"loss": 5.2637,
|
|
"mean_token_accuracy": 0.17233470529317857,
|
|
"num_tokens": 6942085.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.510118293762207,
|
|
"epoch": 3.1607290803645403,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043186927533439654,
|
|
"loss": 5.19,
|
|
"mean_token_accuracy": 0.17383171021938323,
|
|
"num_tokens": 6950356.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.569892501831054,
|
|
"epoch": 3.1648715824357914,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043164003380410335,
|
|
"loss": 5.3208,
|
|
"mean_token_accuracy": 0.1764670118689537,
|
|
"num_tokens": 6960371.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.602646017074585,
|
|
"epoch": 3.169014084507042,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004314104763137445,
|
|
"loss": 5.2629,
|
|
"mean_token_accuracy": 0.17266740947961806,
|
|
"num_tokens": 6969878.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.592065906524658,
|
|
"epoch": 3.173156586578293,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043118060332636264,
|
|
"loss": 5.2544,
|
|
"mean_token_accuracy": 0.17003314048051835,
|
|
"num_tokens": 6979665.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.499735593795776,
|
|
"epoch": 3.1772990886495442,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043095041530563665,
|
|
"loss": 5.232,
|
|
"mean_token_accuracy": 0.17545752078294755,
|
|
"num_tokens": 6988794.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.552596855163574,
|
|
"epoch": 3.1814415907207954,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043071991271588104,
|
|
"loss": 5.2834,
|
|
"mean_token_accuracy": 0.17407965511083603,
|
|
"num_tokens": 6997824.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.534604215621949,
|
|
"epoch": 3.1855840927920465,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004304890960220446,
|
|
"loss": 5.2241,
|
|
"mean_token_accuracy": 0.17584522664546967,
|
|
"num_tokens": 7007281.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.493091058731079,
|
|
"epoch": 3.1897265948632976,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043025796568970975,
|
|
"loss": 5.2277,
|
|
"mean_token_accuracy": 0.1785010427236557,
|
|
"num_tokens": 7016248.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.5047351837158205,
|
|
"epoch": 3.1938690969345487,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043002652218509184,
|
|
"loss": 5.2358,
|
|
"mean_token_accuracy": 0.17241031527519227,
|
|
"num_tokens": 7026319.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.5615987300872805,
|
|
"epoch": 3.1980115990057993,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004297947659750376,
|
|
"loss": 5.2448,
|
|
"mean_token_accuracy": 0.16715144962072373,
|
|
"num_tokens": 7035199.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.55068359375,
|
|
"epoch": 3.2021541010770505,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004295626975270246,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.1773977220058441,
|
|
"num_tokens": 7044302.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.490833520889282,
|
|
"epoch": 3.2062966031483016,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042933031730916024,
|
|
"loss": 5.236,
|
|
"mean_token_accuracy": 0.17236902117729186,
|
|
"num_tokens": 7053640.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.5263467788696286,
|
|
"epoch": 3.2104391052195527,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000429097625790181,
|
|
"loss": 5.1684,
|
|
"mean_token_accuracy": 0.17792723923921586,
|
|
"num_tokens": 7062664.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.4697489738464355,
|
|
"epoch": 3.2145816072908038,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004288646234394508,
|
|
"loss": 5.1867,
|
|
"mean_token_accuracy": 0.17260041534900666,
|
|
"num_tokens": 7071928.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.510807609558105,
|
|
"epoch": 3.218724109362055,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042863131072696115,
|
|
"loss": 5.2769,
|
|
"mean_token_accuracy": 0.16995470672845842,
|
|
"num_tokens": 7081283.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.54783296585083,
|
|
"epoch": 3.2228666114333056,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042839768812332905,
|
|
"loss": 5.2716,
|
|
"mean_token_accuracy": 0.17182289361953734,
|
|
"num_tokens": 7090166.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.584614229202271,
|
|
"epoch": 3.2270091135045567,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000428163756099797,
|
|
"loss": 5.2577,
|
|
"mean_token_accuracy": 0.17217337042093278,
|
|
"num_tokens": 7099050.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.502511787414551,
|
|
"epoch": 3.2311516155758078,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004279295151282314,
|
|
"loss": 5.1731,
|
|
"mean_token_accuracy": 0.18214854449033738,
|
|
"num_tokens": 7107676.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.500295972824096,
|
|
"epoch": 3.235294117647059,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004276949656811217,
|
|
"loss": 5.2448,
|
|
"mean_token_accuracy": 0.16818572133779525,
|
|
"num_tokens": 7115943.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.484800100326538,
|
|
"epoch": 3.23943661971831,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042746010823158006,
|
|
"loss": 5.1512,
|
|
"mean_token_accuracy": 0.1799794092774391,
|
|
"num_tokens": 7124302.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.5508044242858885,
|
|
"epoch": 3.243579121789561,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00042722494325333945,
|
|
"loss": 5.2162,
|
|
"mean_token_accuracy": 0.17433089762926102,
|
|
"num_tokens": 7133701.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.580447721481323,
|
|
"epoch": 3.2477216238608118,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004269894712207533,
|
|
"loss": 5.2152,
|
|
"mean_token_accuracy": 0.1773352012038231,
|
|
"num_tokens": 7142282.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.452110958099365,
|
|
"epoch": 3.251864125932063,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004267536926087947,
|
|
"loss": 5.2282,
|
|
"mean_token_accuracy": 0.17310038357973098,
|
|
"num_tokens": 7151418.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.470469045639038,
|
|
"epoch": 3.256006628003314,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004265176078930545,
|
|
"loss": 5.1893,
|
|
"mean_token_accuracy": 0.17587880641222,
|
|
"num_tokens": 7159765.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.50625467300415,
|
|
"epoch": 3.260149130074565,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042628121754974166,
|
|
"loss": 5.2132,
|
|
"mean_token_accuracy": 0.1741190567612648,
|
|
"num_tokens": 7168345.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.530919027328491,
|
|
"epoch": 3.264291632145816,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004260445220556811,
|
|
"loss": 5.3193,
|
|
"mean_token_accuracy": 0.16753941774368286,
|
|
"num_tokens": 7178232.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.50139684677124,
|
|
"epoch": 3.2684341342170673,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042580752188831376,
|
|
"loss": 5.1923,
|
|
"mean_token_accuracy": 0.17186505645513533,
|
|
"num_tokens": 7187145.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.492300796508789,
|
|
"epoch": 3.272576636288318,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004255702175256946,
|
|
"loss": 5.309,
|
|
"mean_token_accuracy": 0.1700039491057396,
|
|
"num_tokens": 7195931.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.522653245925904,
|
|
"epoch": 3.276719138359569,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042533260944649273,
|
|
"loss": 5.2446,
|
|
"mean_token_accuracy": 0.17079851627349854,
|
|
"num_tokens": 7206090.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.52188549041748,
|
|
"epoch": 3.28086164043082,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004250946981299893,
|
|
"loss": 5.2039,
|
|
"mean_token_accuracy": 0.17926215529441833,
|
|
"num_tokens": 7214785.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.536654949188232,
|
|
"epoch": 3.2850041425020713,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042485648405607776,
|
|
"loss": 5.2439,
|
|
"mean_token_accuracy": 0.1726086363196373,
|
|
"num_tokens": 7223112.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.505476570129394,
|
|
"epoch": 3.2891466445733224,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004246179677052617,
|
|
"loss": 5.2537,
|
|
"mean_token_accuracy": 0.17106932252645493,
|
|
"num_tokens": 7231555.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.489013051986694,
|
|
"epoch": 3.2932891466445735,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004243791495586547,
|
|
"loss": 5.2067,
|
|
"mean_token_accuracy": 0.1790871098637581,
|
|
"num_tokens": 7241366.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.469921684265136,
|
|
"epoch": 3.297431648715824,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004241400300979791,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.18227516263723373,
|
|
"num_tokens": 7249927.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.472465944290161,
|
|
"epoch": 3.3015741507870753,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004239006098055651,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.17641822397708892,
|
|
"num_tokens": 7258143.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.609971189498902,
|
|
"epoch": 3.3057166528583264,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004236608891643494,
|
|
"loss": 5.3105,
|
|
"mean_token_accuracy": 0.17450338900089263,
|
|
"num_tokens": 7267177.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.564677095413208,
|
|
"epoch": 3.3098591549295775,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004234208686578748,
|
|
"loss": 5.2047,
|
|
"mean_token_accuracy": 0.17575108259916306,
|
|
"num_tokens": 7275885.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.462838888168335,
|
|
"epoch": 3.3140016570008286,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042318054877028876,
|
|
"loss": 5.3027,
|
|
"mean_token_accuracy": 0.17088000029325484,
|
|
"num_tokens": 7285535.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3140016570008286,
|
|
"eval_entropy": 5.367351619419512,
|
|
"eval_loss": 5.925260543823242,
|
|
"eval_mean_token_accuracy": 0.15130858514100504,
|
|
"eval_num_tokens": 7285535.0,
|
|
"eval_runtime": 2.9138,
|
|
"eval_samples_per_second": 1285.971,
|
|
"eval_steps_per_second": 160.961,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.526497459411621,
|
|
"epoch": 3.3181441590720797,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042293992998634314,
|
|
"loss": 5.1689,
|
|
"mean_token_accuracy": 0.18199221044778824,
|
|
"num_tokens": 7294517.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.524501132965088,
|
|
"epoch": 3.3222866611433304,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004226990127913918,
|
|
"loss": 5.2427,
|
|
"mean_token_accuracy": 0.16842088401317595,
|
|
"num_tokens": 7303944.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.574664497375489,
|
|
"epoch": 3.3264291632145815,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004224577976713915,
|
|
"loss": 5.2822,
|
|
"mean_token_accuracy": 0.17447033077478408,
|
|
"num_tokens": 7314339.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.543520069122314,
|
|
"epoch": 3.3305716652858326,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042221628511289927,
|
|
"loss": 5.3018,
|
|
"mean_token_accuracy": 0.1708614394068718,
|
|
"num_tokens": 7322644.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.417981195449829,
|
|
"epoch": 3.3347141673570837,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004219744756030723,
|
|
"loss": 5.1915,
|
|
"mean_token_accuracy": 0.18130278587341309,
|
|
"num_tokens": 7331609.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.512018775939941,
|
|
"epoch": 3.338856669428335,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042173236962966706,
|
|
"loss": 5.2458,
|
|
"mean_token_accuracy": 0.17276879251003266,
|
|
"num_tokens": 7341200.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.620357036590576,
|
|
"epoch": 3.342999171499586,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004214899676810375,
|
|
"loss": 5.2706,
|
|
"mean_token_accuracy": 0.17080885022878647,
|
|
"num_tokens": 7350579.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.44530987739563,
|
|
"epoch": 3.347141673570837,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004212472702461348,
|
|
"loss": 5.1534,
|
|
"mean_token_accuracy": 0.18294935673475266,
|
|
"num_tokens": 7358912.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.459890460968017,
|
|
"epoch": 3.3512841756420877,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004210042778145064,
|
|
"loss": 5.2087,
|
|
"mean_token_accuracy": 0.16564791649580002,
|
|
"num_tokens": 7368078.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.549344921112061,
|
|
"epoch": 3.355426677713339,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042076099087629443,
|
|
"loss": 5.2652,
|
|
"mean_token_accuracy": 0.17116860598325728,
|
|
"num_tokens": 7377103.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.548752784729004,
|
|
"epoch": 3.35956917978459,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042051740992223534,
|
|
"loss": 5.2641,
|
|
"mean_token_accuracy": 0.17045499980449677,
|
|
"num_tokens": 7386518.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.432188892364502,
|
|
"epoch": 3.363711681855841,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004202735354436586,
|
|
"loss": 5.1811,
|
|
"mean_token_accuracy": 0.17986492067575455,
|
|
"num_tokens": 7395816.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.5598297119140625,
|
|
"epoch": 3.367854183927092,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042002936793248546,
|
|
"loss": 5.3375,
|
|
"mean_token_accuracy": 0.16982727795839309,
|
|
"num_tokens": 7405365.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.507101440429688,
|
|
"epoch": 3.371996685998343,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004197849078812286,
|
|
"loss": 5.1543,
|
|
"mean_token_accuracy": 0.17965258806943893,
|
|
"num_tokens": 7413633.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.54614109992981,
|
|
"epoch": 3.376139188069594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004195401557829907,
|
|
"loss": 5.274,
|
|
"mean_token_accuracy": 0.177698914706707,
|
|
"num_tokens": 7422634.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.4579259872436525,
|
|
"epoch": 3.380281690140845,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041929511213146343,
|
|
"loss": 5.2348,
|
|
"mean_token_accuracy": 0.174661223590374,
|
|
"num_tokens": 7432517.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.57700366973877,
|
|
"epoch": 3.384424192212096,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004190497774209266,
|
|
"loss": 5.3084,
|
|
"mean_token_accuracy": 0.1689162313938141,
|
|
"num_tokens": 7442839.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.579516410827637,
|
|
"epoch": 3.3885666942833472,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041880415214624715,
|
|
"loss": 5.3707,
|
|
"mean_token_accuracy": 0.15963100492954255,
|
|
"num_tokens": 7452391.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.561112928390503,
|
|
"epoch": 3.3927091963545983,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004185582368028781,
|
|
"loss": 5.2482,
|
|
"mean_token_accuracy": 0.16765529066324233,
|
|
"num_tokens": 7461940.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.491196203231811,
|
|
"epoch": 3.3968516984258494,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004183120318868577,
|
|
"loss": 5.1924,
|
|
"mean_token_accuracy": 0.17289134114980698,
|
|
"num_tokens": 7471783.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.438270854949951,
|
|
"epoch": 3.4009942004971,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041806553789480787,
|
|
"loss": 5.2968,
|
|
"mean_token_accuracy": 0.17247532159090043,
|
|
"num_tokens": 7481084.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.552711772918701,
|
|
"epoch": 3.405136702568351,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041781875532393423,
|
|
"loss": 5.1919,
|
|
"mean_token_accuracy": 0.1760552316904068,
|
|
"num_tokens": 7489528.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.478525590896607,
|
|
"epoch": 3.4092792046396023,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041757168467202404,
|
|
"loss": 5.1395,
|
|
"mean_token_accuracy": 0.183062344789505,
|
|
"num_tokens": 7498490.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.5212721824646,
|
|
"epoch": 3.4134217067108534,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041732432643744576,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.1743450343608856,
|
|
"num_tokens": 7508807.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.489108562469482,
|
|
"epoch": 3.4175642087821045,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004170766811191481,
|
|
"loss": 5.2789,
|
|
"mean_token_accuracy": 0.17611340433359146,
|
|
"num_tokens": 7517349.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.478418684005737,
|
|
"epoch": 3.421706710853355,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004168287492166587,
|
|
"loss": 5.2756,
|
|
"mean_token_accuracy": 0.1656208634376526,
|
|
"num_tokens": 7526253.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.62579927444458,
|
|
"epoch": 3.4258492129246063,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041658053123008325,
|
|
"loss": 5.2517,
|
|
"mean_token_accuracy": 0.17204637676477433,
|
|
"num_tokens": 7535450.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.504294109344483,
|
|
"epoch": 3.4299917149958574,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004163320276601046,
|
|
"loss": 5.287,
|
|
"mean_token_accuracy": 0.17244220525026321,
|
|
"num_tokens": 7544626.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.4702620029449465,
|
|
"epoch": 3.4341342170671085,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004160832390079816,
|
|
"loss": 5.298,
|
|
"mean_token_accuracy": 0.16732285767793656,
|
|
"num_tokens": 7553361.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.555963134765625,
|
|
"epoch": 3.4382767191383596,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041583416577554815,
|
|
"loss": 5.3156,
|
|
"mean_token_accuracy": 0.16941359639167786,
|
|
"num_tokens": 7563648.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.576145267486572,
|
|
"epoch": 3.4424192212096107,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041558480846521213,
|
|
"loss": 5.2617,
|
|
"mean_token_accuracy": 0.1738234743475914,
|
|
"num_tokens": 7572976.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.487558603286743,
|
|
"epoch": 3.446561723280862,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00041533516757995455,
|
|
"loss": 5.227,
|
|
"mean_token_accuracy": 0.17593582421541215,
|
|
"num_tokens": 7581660.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.405913877487182,
|
|
"epoch": 3.4507042253521125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004150852436233283,
|
|
"loss": 5.2751,
|
|
"mean_token_accuracy": 0.17128222137689592,
|
|
"num_tokens": 7590170.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.482892656326294,
|
|
"epoch": 3.4548467274233636,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004148350370994574,
|
|
"loss": 5.2459,
|
|
"mean_token_accuracy": 0.17615955471992492,
|
|
"num_tokens": 7599864.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.553196096420288,
|
|
"epoch": 3.4589892294946147,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004145845485130357,
|
|
"loss": 5.236,
|
|
"mean_token_accuracy": 0.17574975937604903,
|
|
"num_tokens": 7609180.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.568481206893921,
|
|
"epoch": 3.463131731565866,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00041433377836932594,
|
|
"loss": 5.2788,
|
|
"mean_token_accuracy": 0.17776157855987548,
|
|
"num_tokens": 7617067.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.444206857681275,
|
|
"epoch": 3.467274233637117,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041408272717415913,
|
|
"loss": 5.2656,
|
|
"mean_token_accuracy": 0.16847834438085557,
|
|
"num_tokens": 7626868.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.549679946899414,
|
|
"epoch": 3.4714167357083676,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004138313954339327,
|
|
"loss": 5.2155,
|
|
"mean_token_accuracy": 0.1831010788679123,
|
|
"num_tokens": 7636439.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.496394252777099,
|
|
"epoch": 3.4755592377796187,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004135797836556106,
|
|
"loss": 5.2497,
|
|
"mean_token_accuracy": 0.17130410969257354,
|
|
"num_tokens": 7646712.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.422447586059571,
|
|
"epoch": 3.47970173985087,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041332789234672094,
|
|
"loss": 5.249,
|
|
"mean_token_accuracy": 0.1735641524195671,
|
|
"num_tokens": 7657158.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.541329336166382,
|
|
"epoch": 3.483844241922121,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004130757220153561,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.18578998148441314,
|
|
"num_tokens": 7665137.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.411342620849609,
|
|
"epoch": 3.487986743993372,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004128232731701712,
|
|
"loss": 5.1445,
|
|
"mean_token_accuracy": 0.18577479124069213,
|
|
"num_tokens": 7674145.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.44687294960022,
|
|
"epoch": 3.492129246064623,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004125705463203833,
|
|
"loss": 5.26,
|
|
"mean_token_accuracy": 0.1733534500002861,
|
|
"num_tokens": 7682733.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.59356541633606,
|
|
"epoch": 3.4962717481358743,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00041231754197576986,
|
|
"loss": 5.3138,
|
|
"mean_token_accuracy": 0.16534519344568252,
|
|
"num_tokens": 7692241.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.522808265686035,
|
|
"epoch": 3.5004142502071254,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004120642606466685,
|
|
"loss": 5.2483,
|
|
"mean_token_accuracy": 0.17329105883836746,
|
|
"num_tokens": 7700749.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.512319469451905,
|
|
"epoch": 3.504556752278376,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004118107028439751,
|
|
"loss": 5.3436,
|
|
"mean_token_accuracy": 0.17029464095830918,
|
|
"num_tokens": 7709810.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.545750141143799,
|
|
"epoch": 3.508699254349627,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041155686907914365,
|
|
"loss": 5.2939,
|
|
"mean_token_accuracy": 0.17339457124471663,
|
|
"num_tokens": 7718953.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.543676090240479,
|
|
"epoch": 3.5128417564208783,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004113027598641845,
|
|
"loss": 5.2843,
|
|
"mean_token_accuracy": 0.1695699006319046,
|
|
"num_tokens": 7727851.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.471398591995239,
|
|
"epoch": 3.5169842584921294,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004110483757116637,
|
|
"loss": 5.2092,
|
|
"mean_token_accuracy": 0.1840727746486664,
|
|
"num_tokens": 7736281.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.498951101303101,
|
|
"epoch": 3.52112676056338,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041079371713470195,
|
|
"loss": 5.2166,
|
|
"mean_token_accuracy": 0.17776271104812622,
|
|
"num_tokens": 7744899.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.5298792839050295,
|
|
"epoch": 3.525269262634631,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041053878464697343,
|
|
"loss": 5.2612,
|
|
"mean_token_accuracy": 0.17422301769256593,
|
|
"num_tokens": 7754229.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.502022218704224,
|
|
"epoch": 3.5294117647058822,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004102835787627047,
|
|
"loss": 5.2949,
|
|
"mean_token_accuracy": 0.17208452224731446,
|
|
"num_tokens": 7763530.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.576735353469848,
|
|
"epoch": 3.5335542667771334,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004100280999966741,
|
|
"loss": 5.2782,
|
|
"mean_token_accuracy": 0.1697841614484787,
|
|
"num_tokens": 7772572.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.568362045288086,
|
|
"epoch": 3.5376967688483845,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040977234886421007,
|
|
"loss": 5.406,
|
|
"mean_token_accuracy": 0.1635054439306259,
|
|
"num_tokens": 7782842.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.579652595520019,
|
|
"epoch": 3.5418392709196356,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040951632588119064,
|
|
"loss": 5.2533,
|
|
"mean_token_accuracy": 0.17181454449892045,
|
|
"num_tokens": 7792984.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.479238653182984,
|
|
"epoch": 3.5459817729908867,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004092600315640422,
|
|
"loss": 5.2012,
|
|
"mean_token_accuracy": 0.17903612703084945,
|
|
"num_tokens": 7801782.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.518893384933472,
|
|
"epoch": 3.550124275062138,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004090034664297382,
|
|
"loss": 5.2567,
|
|
"mean_token_accuracy": 0.16952574998140335,
|
|
"num_tokens": 7810526.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.475725221633911,
|
|
"epoch": 3.5542667771333885,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040874663099579883,
|
|
"loss": 5.2955,
|
|
"mean_token_accuracy": 0.16559477001428605,
|
|
"num_tokens": 7819916.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.477356100082398,
|
|
"epoch": 3.5584092792046396,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040848952578028915,
|
|
"loss": 5.2016,
|
|
"mean_token_accuracy": 0.17308269143104554,
|
|
"num_tokens": 7828808.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.558059883117676,
|
|
"epoch": 3.5625517812758907,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004082321513018183,
|
|
"loss": 5.3688,
|
|
"mean_token_accuracy": 0.1688382439315319,
|
|
"num_tokens": 7838564.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.3860174179077145,
|
|
"epoch": 3.566694283347142,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040797450807953897,
|
|
"loss": 5.1937,
|
|
"mean_token_accuracy": 0.17987646758556367,
|
|
"num_tokens": 7847813.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.484470844268799,
|
|
"epoch": 3.5708367854183924,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004077165966331454,
|
|
"loss": 5.2976,
|
|
"mean_token_accuracy": 0.16886635273694992,
|
|
"num_tokens": 7856818.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.51791820526123,
|
|
"epoch": 3.5749792874896436,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004074584174828735,
|
|
"loss": 5.2059,
|
|
"mean_token_accuracy": 0.17356480062007903,
|
|
"num_tokens": 7866220.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.471795320510864,
|
|
"epoch": 3.5791217895608947,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040719997114949856,
|
|
"loss": 5.2705,
|
|
"mean_token_accuracy": 0.16564323604106904,
|
|
"num_tokens": 7875648.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.513032722473144,
|
|
"epoch": 3.5832642916321458,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040694125815433533,
|
|
"loss": 5.1727,
|
|
"mean_token_accuracy": 0.1779831498861313,
|
|
"num_tokens": 7884460.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.555390787124634,
|
|
"epoch": 3.587406793703397,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004066822790192359,
|
|
"loss": 5.3197,
|
|
"mean_token_accuracy": 0.17107179015874863,
|
|
"num_tokens": 7893501.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.470081377029419,
|
|
"epoch": 3.591549295774648,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040642303426658967,
|
|
"loss": 5.2692,
|
|
"mean_token_accuracy": 0.18066452890634538,
|
|
"num_tokens": 7902867.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.442861175537109,
|
|
"epoch": 3.595691797845899,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040616352441932163,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.1710597515106201,
|
|
"num_tokens": 7911769.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.428875494003296,
|
|
"epoch": 3.59983429991715,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040590375000089146,
|
|
"loss": 5.097,
|
|
"mean_token_accuracy": 0.18330689072608947,
|
|
"num_tokens": 7919527.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.515513372421265,
|
|
"epoch": 3.603976801988401,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040564371153529256,
|
|
"loss": 5.309,
|
|
"mean_token_accuracy": 0.17409183084964752,
|
|
"num_tokens": 7929726.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.455484056472779,
|
|
"epoch": 3.608119304059652,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004053834095470509,
|
|
"loss": 5.1902,
|
|
"mean_token_accuracy": 0.17837373316287994,
|
|
"num_tokens": 7939001.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.461891746520996,
|
|
"epoch": 3.612261806130903,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040512284456122407,
|
|
"loss": 5.2057,
|
|
"mean_token_accuracy": 0.17613703310489653,
|
|
"num_tokens": 7948782.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.522659873962402,
|
|
"epoch": 3.616404308202154,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040486201710340016,
|
|
"loss": 5.2475,
|
|
"mean_token_accuracy": 0.17603346705436707,
|
|
"num_tokens": 7958015.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.446337366104126,
|
|
"epoch": 3.6205468102734053,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040460092769969654,
|
|
"loss": 5.2323,
|
|
"mean_token_accuracy": 0.17861249595880507,
|
|
"num_tokens": 7966297.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.534082412719727,
|
|
"epoch": 3.624689312344656,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004043395768767592,
|
|
"loss": 5.31,
|
|
"mean_token_accuracy": 0.1688306972384453,
|
|
"num_tokens": 7974998.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.51026930809021,
|
|
"epoch": 3.628831814415907,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004040779651617612,
|
|
"loss": 5.3008,
|
|
"mean_token_accuracy": 0.17100179344415664,
|
|
"num_tokens": 7984882.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.5321870803833,
|
|
"epoch": 3.632974316487158,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040381609308240207,
|
|
"loss": 5.2946,
|
|
"mean_token_accuracy": 0.1682634100317955,
|
|
"num_tokens": 7994163.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.392755746841431,
|
|
"epoch": 3.6371168185584093,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040355396116690634,
|
|
"loss": 5.1406,
|
|
"mean_token_accuracy": 0.18378142416477203,
|
|
"num_tokens": 8003633.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.435786581039428,
|
|
"epoch": 3.6412593206296604,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004032915699440227,
|
|
"loss": 5.2814,
|
|
"mean_token_accuracy": 0.16911471039056777,
|
|
"num_tokens": 8012865.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.596218681335449,
|
|
"epoch": 3.6454018227009115,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000403028919943023,
|
|
"loss": 5.3363,
|
|
"mean_token_accuracy": 0.16400040686130524,
|
|
"num_tokens": 8022979.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.438734769821167,
|
|
"epoch": 3.6495443247721626,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004027660116937009,
|
|
"loss": 5.2388,
|
|
"mean_token_accuracy": 0.17290958613157273,
|
|
"num_tokens": 8031772.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.450803899765015,
|
|
"epoch": 3.6536868268434133,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004025028457263712,
|
|
"loss": 5.2318,
|
|
"mean_token_accuracy": 0.17455596029758452,
|
|
"num_tokens": 8041398.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.476619005203247,
|
|
"epoch": 3.6578293289146644,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004022394225718684,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.175406576693058,
|
|
"num_tokens": 8050607.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.495734119415284,
|
|
"epoch": 3.6619718309859155,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004019757427615456,
|
|
"loss": 5.2899,
|
|
"mean_token_accuracy": 0.17273638099431993,
|
|
"num_tokens": 8060018.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.479266786575318,
|
|
"epoch": 3.6661143330571666,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004017118068272739,
|
|
"loss": 5.2007,
|
|
"mean_token_accuracy": 0.17613022178411483,
|
|
"num_tokens": 8069740.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.454047203063965,
|
|
"epoch": 3.6702568351284177,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040144761530144096,
|
|
"loss": 5.2438,
|
|
"mean_token_accuracy": 0.17386065274477006,
|
|
"num_tokens": 8078792.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.41513295173645,
|
|
"epoch": 3.6743993371996684,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040118316871695,
|
|
"loss": 5.1731,
|
|
"mean_token_accuracy": 0.18286127597093582,
|
|
"num_tokens": 8086835.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.46587438583374,
|
|
"epoch": 3.6785418392709195,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040091846760721857,
|
|
"loss": 5.285,
|
|
"mean_token_accuracy": 0.1675587072968483,
|
|
"num_tokens": 8096084.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.532276725769043,
|
|
"epoch": 3.6826843413421706,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040065351250617774,
|
|
"loss": 5.1728,
|
|
"mean_token_accuracy": 0.18539456129074097,
|
|
"num_tokens": 8105332.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.471866178512573,
|
|
"epoch": 3.6868268434134217,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040038830394827107,
|
|
"loss": 5.239,
|
|
"mean_token_accuracy": 0.17389686703681945,
|
|
"num_tokens": 8113572.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.5083342552185055,
|
|
"epoch": 3.690969345484673,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000400122842468453,
|
|
"loss": 5.2398,
|
|
"mean_token_accuracy": 0.17414215356111526,
|
|
"num_tokens": 8122457.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.423385524749756,
|
|
"epoch": 3.695111847555924,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003998571286021884,
|
|
"loss": 5.1889,
|
|
"mean_token_accuracy": 0.17862726897001266,
|
|
"num_tokens": 8131397.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.472986793518066,
|
|
"epoch": 3.699254349627175,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003995911628854512,
|
|
"loss": 5.2382,
|
|
"mean_token_accuracy": 0.16506683677434922,
|
|
"num_tokens": 8140949.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.467649698257446,
|
|
"epoch": 3.7033968516984257,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003993249458547234,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.17652828693389894,
|
|
"num_tokens": 8149097.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.461728191375732,
|
|
"epoch": 3.707539353769677,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039905847804699363,
|
|
"loss": 5.1947,
|
|
"mean_token_accuracy": 0.18033552914857864,
|
|
"num_tokens": 8157671.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.460736560821533,
|
|
"epoch": 3.711681855840928,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039879175999975677,
|
|
"loss": 5.3368,
|
|
"mean_token_accuracy": 0.17144761979579926,
|
|
"num_tokens": 8166924.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.468470668792724,
|
|
"epoch": 3.715824357912179,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039852479225101217,
|
|
"loss": 5.1618,
|
|
"mean_token_accuracy": 0.17559899538755416,
|
|
"num_tokens": 8175452.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.530250835418701,
|
|
"epoch": 3.71996685998343,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039825757533926305,
|
|
"loss": 5.309,
|
|
"mean_token_accuracy": 0.16716831028461457,
|
|
"num_tokens": 8184024.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.504942846298218,
|
|
"epoch": 3.724109362054681,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039799010980351494,
|
|
"loss": 5.2763,
|
|
"mean_token_accuracy": 0.17509255856275557,
|
|
"num_tokens": 8193177.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.464646863937378,
|
|
"epoch": 3.728251864125932,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039772239618327535,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.18288797587156297,
|
|
"num_tokens": 8202160.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.728251864125932,
|
|
"eval_entropy": 5.3644926217573285,
|
|
"eval_loss": 5.881169319152832,
|
|
"eval_mean_token_accuracy": 0.1548894850302861,
|
|
"eval_num_tokens": 8202160.0,
|
|
"eval_runtime": 2.9022,
|
|
"eval_samples_per_second": 1291.098,
|
|
"eval_steps_per_second": 161.603,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.559306049346924,
|
|
"epoch": 3.732394366197183,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039745443501855156,
|
|
"loss": 5.2974,
|
|
"mean_token_accuracy": 0.17328668683767318,
|
|
"num_tokens": 8212165.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.476095485687256,
|
|
"epoch": 3.736536868268434,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039718622684985075,
|
|
"loss": 5.3062,
|
|
"mean_token_accuracy": 0.1719233885407448,
|
|
"num_tokens": 8222303.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.465471267700195,
|
|
"epoch": 3.7406793703396852,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000396917772218178,
|
|
"loss": 5.1798,
|
|
"mean_token_accuracy": 0.17783670872449875,
|
|
"num_tokens": 8231154.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.550866031646729,
|
|
"epoch": 3.7448218724109363,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003966490716650357,
|
|
"loss": 5.2927,
|
|
"mean_token_accuracy": 0.1665860414505005,
|
|
"num_tokens": 8240501.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.517164468765259,
|
|
"epoch": 3.7489643744821874,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039638012573242225,
|
|
"loss": 5.2304,
|
|
"mean_token_accuracy": 0.17294808775186538,
|
|
"num_tokens": 8248494.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.424460935592651,
|
|
"epoch": 3.7531068765534386,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003961109349628309,
|
|
"loss": 5.0358,
|
|
"mean_token_accuracy": 0.18549548387527465,
|
|
"num_tokens": 8256614.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.4096883773803714,
|
|
"epoch": 3.757249378624689,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000395841499899249,
|
|
"loss": 5.1846,
|
|
"mean_token_accuracy": 0.17832193821668624,
|
|
"num_tokens": 8265866.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.559077882766724,
|
|
"epoch": 3.7613918806959403,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039557182108515624,
|
|
"loss": 5.2997,
|
|
"mean_token_accuracy": 0.17162225842475892,
|
|
"num_tokens": 8274286.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.492881774902344,
|
|
"epoch": 3.7655343827671914,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039530189906452457,
|
|
"loss": 5.3423,
|
|
"mean_token_accuracy": 0.16214823126792907,
|
|
"num_tokens": 8282722.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.480157470703125,
|
|
"epoch": 3.7696768848384425,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00039503173438181605,
|
|
"loss": 5.2603,
|
|
"mean_token_accuracy": 0.17920465320348739,
|
|
"num_tokens": 8290850.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.499615907669067,
|
|
"epoch": 3.773819386909693,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039476132758198236,
|
|
"loss": 5.2799,
|
|
"mean_token_accuracy": 0.16742433309555055,
|
|
"num_tokens": 8300020.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.44893479347229,
|
|
"epoch": 3.7779618889809443,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039449067921046353,
|
|
"loss": 5.1373,
|
|
"mean_token_accuracy": 0.18598946928977966,
|
|
"num_tokens": 8308383.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.529224252700805,
|
|
"epoch": 3.7821043910521954,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039421978981318703,
|
|
"loss": 5.2428,
|
|
"mean_token_accuracy": 0.16686344891786575,
|
|
"num_tokens": 8316911.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.407128429412841,
|
|
"epoch": 3.7862468931234465,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039394865993656625,
|
|
"loss": 5.213,
|
|
"mean_token_accuracy": 0.17910367399454116,
|
|
"num_tokens": 8326013.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.471581697463989,
|
|
"epoch": 3.7903893951946976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003936772901274999,
|
|
"loss": 5.231,
|
|
"mean_token_accuracy": 0.18024682998657227,
|
|
"num_tokens": 8336138.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.402024364471435,
|
|
"epoch": 3.7945318972659487,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003934056809333705,
|
|
"loss": 5.1917,
|
|
"mean_token_accuracy": 0.17773695886135102,
|
|
"num_tokens": 8344886.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.456485223770142,
|
|
"epoch": 3.7986743993372,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039313383290204333,
|
|
"loss": 5.2071,
|
|
"mean_token_accuracy": 0.1724173456430435,
|
|
"num_tokens": 8354309.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.430463218688965,
|
|
"epoch": 3.802816901408451,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003928617465818658,
|
|
"loss": 5.1755,
|
|
"mean_token_accuracy": 0.18038210719823838,
|
|
"num_tokens": 8363073.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.45398530960083,
|
|
"epoch": 3.8069594034797016,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039258942252166554,
|
|
"loss": 5.2873,
|
|
"mean_token_accuracy": 0.1756221666932106,
|
|
"num_tokens": 8372002.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.5177818775177006,
|
|
"epoch": 3.8111019055509527,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003923168612707502,
|
|
"loss": 5.2873,
|
|
"mean_token_accuracy": 0.1761231541633606,
|
|
"num_tokens": 8381182.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.453721380233764,
|
|
"epoch": 3.815244407622204,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003920440633789053,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.18097713440656663,
|
|
"num_tokens": 8390144.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.458373498916626,
|
|
"epoch": 3.819386909693455,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039177102939639424,
|
|
"loss": 5.3532,
|
|
"mean_token_accuracy": 0.1726341038942337,
|
|
"num_tokens": 8399293.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.571560430526733,
|
|
"epoch": 3.8235294117647056,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003914977598739563,
|
|
"loss": 5.3382,
|
|
"mean_token_accuracy": 0.1679692968726158,
|
|
"num_tokens": 8410013.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.414801025390625,
|
|
"epoch": 3.8276719138359567,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003912242553628059,
|
|
"loss": 5.1392,
|
|
"mean_token_accuracy": 0.17811594903469086,
|
|
"num_tokens": 8419194.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.49128212928772,
|
|
"epoch": 3.831814415907208,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003909505164146318,
|
|
"loss": 5.2603,
|
|
"mean_token_accuracy": 0.1782832309603691,
|
|
"num_tokens": 8428540.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.489205312728882,
|
|
"epoch": 3.835956917978459,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000390676543581595,
|
|
"loss": 5.2615,
|
|
"mean_token_accuracy": 0.17321666777133943,
|
|
"num_tokens": 8437454.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.486815071105957,
|
|
"epoch": 3.84009942004971,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003904023374163289,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.17493631243705748,
|
|
"num_tokens": 8447025.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.432703828811645,
|
|
"epoch": 3.844241922120961,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003901278984719372,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.18728309869766235,
|
|
"num_tokens": 8455936.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.441501140594482,
|
|
"epoch": 3.8483844241922123,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038985322730199326,
|
|
"loss": 5.1998,
|
|
"mean_token_accuracy": 0.17771358042955399,
|
|
"num_tokens": 8464991.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.478009748458862,
|
|
"epoch": 3.8525269262634634,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003895783244605389,
|
|
"loss": 5.2367,
|
|
"mean_token_accuracy": 0.1771668627858162,
|
|
"num_tokens": 8474236.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.5974479675292965,
|
|
"epoch": 3.856669428334714,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038930319050208315,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.1756897822022438,
|
|
"num_tokens": 8483197.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.5113684177398685,
|
|
"epoch": 3.860811930405965,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038902782598160127,
|
|
"loss": 5.2999,
|
|
"mean_token_accuracy": 0.17589164525270462,
|
|
"num_tokens": 8493341.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.489524078369141,
|
|
"epoch": 3.8649544324772163,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038875223145453346,
|
|
"loss": 5.2577,
|
|
"mean_token_accuracy": 0.17299377918243408,
|
|
"num_tokens": 8502277.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.468378257751465,
|
|
"epoch": 3.8690969345484674,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003884764074767842,
|
|
"loss": 5.1842,
|
|
"mean_token_accuracy": 0.1811079576611519,
|
|
"num_tokens": 8511266.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.4820366382598875,
|
|
"epoch": 3.873239436619718,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003882003546047205,
|
|
"loss": 5.2211,
|
|
"mean_token_accuracy": 0.17966129183769225,
|
|
"num_tokens": 8519869.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.514575862884522,
|
|
"epoch": 3.877381938690969,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038792407339517123,
|
|
"loss": 5.3667,
|
|
"mean_token_accuracy": 0.16565651297569275,
|
|
"num_tokens": 8528835.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.483935832977295,
|
|
"epoch": 3.8815244407622203,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038764756440542574,
|
|
"loss": 5.2208,
|
|
"mean_token_accuracy": 0.17118781208992004,
|
|
"num_tokens": 8538166.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.482791376113892,
|
|
"epoch": 3.8856669428334714,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00038737082819323285,
|
|
"loss": 5.2134,
|
|
"mean_token_accuracy": 0.18406185507774353,
|
|
"num_tokens": 8546920.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.471239519119263,
|
|
"epoch": 3.8898094449047225,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038709386531679974,
|
|
"loss": 5.2085,
|
|
"mean_token_accuracy": 0.17829401940107345,
|
|
"num_tokens": 8556459.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.434108781814575,
|
|
"epoch": 3.8939519469759736,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038681667633479086,
|
|
"loss": 5.2925,
|
|
"mean_token_accuracy": 0.1692568451166153,
|
|
"num_tokens": 8565461.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.54814043045044,
|
|
"epoch": 3.8980944490472247,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038653926180632665,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.18008081763982772,
|
|
"num_tokens": 8574357.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.517384004592896,
|
|
"epoch": 3.902236951118476,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038626162229098257,
|
|
"loss": 5.2519,
|
|
"mean_token_accuracy": 0.1762557953596115,
|
|
"num_tokens": 8584063.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.520116090774536,
|
|
"epoch": 3.9063794531897265,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003859837583487877,
|
|
"loss": 5.3771,
|
|
"mean_token_accuracy": 0.16987684667110442,
|
|
"num_tokens": 8593367.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.535309791564941,
|
|
"epoch": 3.9105219552609776,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003857056705402241,
|
|
"loss": 5.3299,
|
|
"mean_token_accuracy": 0.16806533932685852,
|
|
"num_tokens": 8603575.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.393655633926391,
|
|
"epoch": 3.9146644573322287,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003854273594262253,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.17735397964715957,
|
|
"num_tokens": 8612271.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.521361923217773,
|
|
"epoch": 3.91880695940348,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003851488255681752,
|
|
"loss": 5.2462,
|
|
"mean_token_accuracy": 0.17396413087844848,
|
|
"num_tokens": 8621438.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.518579816818237,
|
|
"epoch": 3.922949461474731,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000384870069527907,
|
|
"loss": 5.3138,
|
|
"mean_token_accuracy": 0.16041608452796935,
|
|
"num_tokens": 8631579.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.4228527545928955,
|
|
"epoch": 3.9270919635459816,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038459109186770213,
|
|
"loss": 5.1614,
|
|
"mean_token_accuracy": 0.18738928586244583,
|
|
"num_tokens": 8640666.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.457975006103515,
|
|
"epoch": 3.9312344656172327,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038431189315028905,
|
|
"loss": 5.2527,
|
|
"mean_token_accuracy": 0.17120098769664766,
|
|
"num_tokens": 8650422.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.561083650588989,
|
|
"epoch": 3.9353769676884838,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003840324739388419,
|
|
"loss": 5.3432,
|
|
"mean_token_accuracy": 0.17369136959314346,
|
|
"num_tokens": 8660770.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.485694551467896,
|
|
"epoch": 3.939519469759735,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038375283479698,
|
|
"loss": 5.2028,
|
|
"mean_token_accuracy": 0.17698052674531936,
|
|
"num_tokens": 8669289.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.44639482498169,
|
|
"epoch": 3.943661971830986,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003834729762887659,
|
|
"loss": 5.2152,
|
|
"mean_token_accuracy": 0.1800999715924263,
|
|
"num_tokens": 8678639.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.462325716018677,
|
|
"epoch": 3.947804473902237,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003831928989787049,
|
|
"loss": 5.197,
|
|
"mean_token_accuracy": 0.17724782973527908,
|
|
"num_tokens": 8687359.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.495435762405395,
|
|
"epoch": 3.951946975973488,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003829126034317435,
|
|
"loss": 5.3156,
|
|
"mean_token_accuracy": 0.17422839552164077,
|
|
"num_tokens": 8695921.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.557425928115845,
|
|
"epoch": 3.956089478044739,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003826320902132684,
|
|
"loss": 5.2619,
|
|
"mean_token_accuracy": 0.1737216591835022,
|
|
"num_tokens": 8704890.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.524425983428955,
|
|
"epoch": 3.96023198011599,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038235135988910545,
|
|
"loss": 5.299,
|
|
"mean_token_accuracy": 0.17786565721035003,
|
|
"num_tokens": 8715085.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.53675537109375,
|
|
"epoch": 3.964374482187241,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038207041302551844,
|
|
"loss": 5.2738,
|
|
"mean_token_accuracy": 0.16570378839969635,
|
|
"num_tokens": 8724361.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.399000549316407,
|
|
"epoch": 3.968516984258492,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003817892501892078,
|
|
"loss": 5.1744,
|
|
"mean_token_accuracy": 0.17498714923858644,
|
|
"num_tokens": 8733078.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.425567960739135,
|
|
"epoch": 3.9726594863297433,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038150787194730977,
|
|
"loss": 5.1722,
|
|
"mean_token_accuracy": 0.17695580124855043,
|
|
"num_tokens": 8742459.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.4331988334655765,
|
|
"epoch": 3.976801988400994,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000381226278867395,
|
|
"loss": 5.1929,
|
|
"mean_token_accuracy": 0.1812575191259384,
|
|
"num_tokens": 8751300.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.434637403488159,
|
|
"epoch": 3.980944490472245,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038094447151746746,
|
|
"loss": 5.1928,
|
|
"mean_token_accuracy": 0.18481962382793427,
|
|
"num_tokens": 8759646.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.463468313217163,
|
|
"epoch": 3.985086992543496,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003806624504659635,
|
|
"loss": 5.3334,
|
|
"mean_token_accuracy": 0.17498264461755753,
|
|
"num_tokens": 8768810.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.480513191223144,
|
|
"epoch": 3.9892294946147473,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003803802162817503,
|
|
"loss": 5.186,
|
|
"mean_token_accuracy": 0.18265987187623978,
|
|
"num_tokens": 8777248.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.499728441238403,
|
|
"epoch": 3.9933719966859984,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000380097769534125,
|
|
"loss": 5.2837,
|
|
"mean_token_accuracy": 0.16867271214723586,
|
|
"num_tokens": 8787027.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.409338140487671,
|
|
"epoch": 3.9975144987572495,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003798151107928136,
|
|
"loss": 5.0845,
|
|
"mean_token_accuracy": 0.17988889962434768,
|
|
"num_tokens": 8795426.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.317444229125977,
|
|
"epoch": 4.001657000828501,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037953224062796967,
|
|
"loss": 5.0188,
|
|
"mean_token_accuracy": 0.1928405523300171,
|
|
"num_tokens": 8803753.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.537591743469238,
|
|
"epoch": 4.005799502899752,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003792491596101733,
|
|
"loss": 5.0516,
|
|
"mean_token_accuracy": 0.18323733210563659,
|
|
"num_tokens": 8813203.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.44191575050354,
|
|
"epoch": 4.009942004971003,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003789658683104298,
|
|
"loss": 4.9438,
|
|
"mean_token_accuracy": 0.19332674890756607,
|
|
"num_tokens": 8821920.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.4816694259643555,
|
|
"epoch": 4.014084507042254,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037868236730016866,
|
|
"loss": 4.9697,
|
|
"mean_token_accuracy": 0.18597270995378495,
|
|
"num_tokens": 8831761.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.498815965652466,
|
|
"epoch": 4.018227009113504,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037839865715124234,
|
|
"loss": 5.0388,
|
|
"mean_token_accuracy": 0.191935957968235,
|
|
"num_tokens": 8841096.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.526767206192017,
|
|
"epoch": 4.022369511184755,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037811473843592534,
|
|
"loss": 5.1382,
|
|
"mean_token_accuracy": 0.18287116438150405,
|
|
"num_tokens": 8850554.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.420525550842285,
|
|
"epoch": 4.026512013256006,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00037783061172691273,
|
|
"loss": 4.9689,
|
|
"mean_token_accuracy": 0.19099729210138322,
|
|
"num_tokens": 8858578.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.433285093307495,
|
|
"epoch": 4.0306545153272575,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003775462775973191,
|
|
"loss": 4.9244,
|
|
"mean_token_accuracy": 0.19511812031269074,
|
|
"num_tokens": 8868458.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.411444568634034,
|
|
"epoch": 4.034797017398509,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003772617366206773,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.19730500429868697,
|
|
"num_tokens": 8877634.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.403138589859009,
|
|
"epoch": 4.03893951946976,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00037697698937093783,
|
|
"loss": 4.9839,
|
|
"mean_token_accuracy": 0.18978192806243896,
|
|
"num_tokens": 8886033.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.364664554595947,
|
|
"epoch": 4.043082021541011,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037669203642246685,
|
|
"loss": 4.8932,
|
|
"mean_token_accuracy": 0.2053120642900467,
|
|
"num_tokens": 8895558.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.364785861968994,
|
|
"epoch": 4.047224523612262,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037640687835004567,
|
|
"loss": 4.9603,
|
|
"mean_token_accuracy": 0.1945762738585472,
|
|
"num_tokens": 8904573.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.415970182418823,
|
|
"epoch": 4.051367025683513,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003761215157288692,
|
|
"loss": 5.0044,
|
|
"mean_token_accuracy": 0.1893035128712654,
|
|
"num_tokens": 8914933.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.391638612747192,
|
|
"epoch": 4.055509527754764,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037583594913454514,
|
|
"loss": 4.9287,
|
|
"mean_token_accuracy": 0.1958758443593979,
|
|
"num_tokens": 8924997.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.427448654174805,
|
|
"epoch": 4.059652029826015,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003755501791430925,
|
|
"loss": 4.9428,
|
|
"mean_token_accuracy": 0.19656527787446976,
|
|
"num_tokens": 8933114.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.346042776107788,
|
|
"epoch": 4.063794531897266,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037526420633094044,
|
|
"loss": 4.934,
|
|
"mean_token_accuracy": 0.1955525815486908,
|
|
"num_tokens": 8942142.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.403874158859253,
|
|
"epoch": 4.067937033968517,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00037497803127492757,
|
|
"loss": 4.9747,
|
|
"mean_token_accuracy": 0.19537484496831894,
|
|
"num_tokens": 8951774.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.447066402435302,
|
|
"epoch": 4.072079536039768,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003746916545523001,
|
|
"loss": 4.9517,
|
|
"mean_token_accuracy": 0.19272928535938263,
|
|
"num_tokens": 8961195.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.376217031478882,
|
|
"epoch": 4.076222038111019,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003744050767407113,
|
|
"loss": 4.939,
|
|
"mean_token_accuracy": 0.1912950396537781,
|
|
"num_tokens": 8970440.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.421129751205444,
|
|
"epoch": 4.08036454018227,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003741182984182198,
|
|
"loss": 4.9848,
|
|
"mean_token_accuracy": 0.19086510688066483,
|
|
"num_tokens": 8980327.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.425602149963379,
|
|
"epoch": 4.084507042253521,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003738313201632889,
|
|
"loss": 4.9854,
|
|
"mean_token_accuracy": 0.18635470867156984,
|
|
"num_tokens": 8988817.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.419035530090332,
|
|
"epoch": 4.088649544324772,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003735441425547851,
|
|
"loss": 4.9031,
|
|
"mean_token_accuracy": 0.2012096792459488,
|
|
"num_tokens": 8997240.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.3494363784790036,
|
|
"epoch": 4.092792046396023,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037325676617197697,
|
|
"loss": 4.8943,
|
|
"mean_token_accuracy": 0.19664187878370284,
|
|
"num_tokens": 9005706.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.305064344406128,
|
|
"epoch": 4.096934548467274,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003729691915945342,
|
|
"loss": 4.9611,
|
|
"mean_token_accuracy": 0.18814149200916291,
|
|
"num_tokens": 9015053.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.3199138164520265,
|
|
"epoch": 4.1010770505385254,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000372681419402526,
|
|
"loss": 4.9808,
|
|
"mean_token_accuracy": 0.18428089320659638,
|
|
"num_tokens": 9023619.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.445406055450439,
|
|
"epoch": 4.105219552609777,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003723934501764203,
|
|
"loss": 4.9386,
|
|
"mean_token_accuracy": 0.19274009764194489,
|
|
"num_tokens": 9032680.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.428956031799316,
|
|
"epoch": 4.109362054681028,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003721052844970826,
|
|
"loss": 5.0108,
|
|
"mean_token_accuracy": 0.19121325761079788,
|
|
"num_tokens": 9042379.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.346081924438477,
|
|
"epoch": 4.113504556752279,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037181692294577457,
|
|
"loss": 4.9007,
|
|
"mean_token_accuracy": 0.19392535090446472,
|
|
"num_tokens": 9050831.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.355849742889404,
|
|
"epoch": 4.117647058823529,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00037152836610415284,
|
|
"loss": 4.8587,
|
|
"mean_token_accuracy": 0.195468071103096,
|
|
"num_tokens": 9059086.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.410044097900391,
|
|
"epoch": 4.12178956089478,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00037123961455426836,
|
|
"loss": 4.9784,
|
|
"mean_token_accuracy": 0.18813973367214204,
|
|
"num_tokens": 9068092.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.318189239501953,
|
|
"epoch": 4.125932062966031,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037095066887856434,
|
|
"loss": 4.8618,
|
|
"mean_token_accuracy": 0.20042508691549302,
|
|
"num_tokens": 9076140.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.437709808349609,
|
|
"epoch": 4.130074565037282,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003706615296598758,
|
|
"loss": 5.0985,
|
|
"mean_token_accuracy": 0.1856787994503975,
|
|
"num_tokens": 9086021.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.442448711395263,
|
|
"epoch": 4.134217067108533,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003703721974814282,
|
|
"loss": 5.0627,
|
|
"mean_token_accuracy": 0.1781926155090332,
|
|
"num_tokens": 9096572.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.427663516998291,
|
|
"epoch": 4.1383595691797845,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00037008267292683615,
|
|
"loss": 4.8923,
|
|
"mean_token_accuracy": 0.18733982741832733,
|
|
"num_tokens": 9105428.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.461180067062378,
|
|
"epoch": 4.142502071251036,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036979295658010227,
|
|
"loss": 5.019,
|
|
"mean_token_accuracy": 0.1952150955796242,
|
|
"num_tokens": 9114676.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.142502071251036,
|
|
"eval_entropy": 5.230287625083029,
|
|
"eval_loss": 5.872674465179443,
|
|
"eval_mean_token_accuracy": 0.15736299255954178,
|
|
"eval_num_tokens": 9114676.0,
|
|
"eval_runtime": 2.7583,
|
|
"eval_samples_per_second": 1358.462,
|
|
"eval_steps_per_second": 170.034,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.3767101764678955,
|
|
"epoch": 4.146644573322287,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003695030490256161,
|
|
"loss": 4.944,
|
|
"mean_token_accuracy": 0.1982339709997177,
|
|
"num_tokens": 9123484.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.436297988891601,
|
|
"epoch": 4.150787075393538,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000369212950848153,
|
|
"loss": 4.9755,
|
|
"mean_token_accuracy": 0.19222400784492494,
|
|
"num_tokens": 9132318.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.358932065963745,
|
|
"epoch": 4.154929577464789,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003689226626328725,
|
|
"loss": 5.0334,
|
|
"mean_token_accuracy": 0.18922119140625,
|
|
"num_tokens": 9141621.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.407165050506592,
|
|
"epoch": 4.15907207953604,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036863218496531767,
|
|
"loss": 5.0218,
|
|
"mean_token_accuracy": 0.1867724373936653,
|
|
"num_tokens": 9149598.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.401487922668457,
|
|
"epoch": 4.163214581607291,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003683415184314139,
|
|
"loss": 4.8901,
|
|
"mean_token_accuracy": 0.19928769767284393,
|
|
"num_tokens": 9157680.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.4233259677886965,
|
|
"epoch": 4.167357083678542,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036805066361746716,
|
|
"loss": 4.9924,
|
|
"mean_token_accuracy": 0.19483090490102767,
|
|
"num_tokens": 9166715.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.405722141265869,
|
|
"epoch": 4.1714995857497925,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036775962111016346,
|
|
"loss": 4.9949,
|
|
"mean_token_accuracy": 0.19161402732133864,
|
|
"num_tokens": 9176045.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.385818815231323,
|
|
"epoch": 4.175642087821044,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00036746839149656734,
|
|
"loss": 4.9106,
|
|
"mean_token_accuracy": 0.19114130586385727,
|
|
"num_tokens": 9183989.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.364512777328491,
|
|
"epoch": 4.179784589892295,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003671769753641207,
|
|
"loss": 5.018,
|
|
"mean_token_accuracy": 0.18578292280435563,
|
|
"num_tokens": 9193868.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.40920786857605,
|
|
"epoch": 4.183927091963546,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036688537330064187,
|
|
"loss": 4.9908,
|
|
"mean_token_accuracy": 0.1905068650841713,
|
|
"num_tokens": 9203016.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.397743463516235,
|
|
"epoch": 4.188069594034797,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000366593585894324,
|
|
"loss": 4.9166,
|
|
"mean_token_accuracy": 0.19907285869121552,
|
|
"num_tokens": 9211321.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.378936338424682,
|
|
"epoch": 4.192212096106048,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003663016137337341,
|
|
"loss": 4.9751,
|
|
"mean_token_accuracy": 0.18562365472316741,
|
|
"num_tokens": 9220260.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.382741975784302,
|
|
"epoch": 4.196354598177299,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036600945740781197,
|
|
"loss": 5.0517,
|
|
"mean_token_accuracy": 0.18590944409370422,
|
|
"num_tokens": 9230501.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.419135284423828,
|
|
"epoch": 4.20049710024855,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00036571711750586896,
|
|
"loss": 5.0228,
|
|
"mean_token_accuracy": 0.1889002189040184,
|
|
"num_tokens": 9238956.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.337713289260864,
|
|
"epoch": 4.204639602319801,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003654245946175864,
|
|
"loss": 4.9215,
|
|
"mean_token_accuracy": 0.19794601202011108,
|
|
"num_tokens": 9246921.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.453129720687866,
|
|
"epoch": 4.2087821043910525,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036513188933301504,
|
|
"loss": 5.0072,
|
|
"mean_token_accuracy": 0.19138323068618773,
|
|
"num_tokens": 9256474.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.381913137435913,
|
|
"epoch": 4.212924606462304,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003648390022425734,
|
|
"loss": 5.0361,
|
|
"mean_token_accuracy": 0.18918444365262985,
|
|
"num_tokens": 9266238.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.369112920761109,
|
|
"epoch": 4.217067108533554,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003645459339370468,
|
|
"loss": 4.9743,
|
|
"mean_token_accuracy": 0.197220878303051,
|
|
"num_tokens": 9275293.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.32979154586792,
|
|
"epoch": 4.221209610604805,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00036425268500758594,
|
|
"loss": 4.7503,
|
|
"mean_token_accuracy": 0.2165217161178589,
|
|
"num_tokens": 9284214.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.468719387054444,
|
|
"epoch": 4.225352112676056,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00036395925604570614,
|
|
"loss": 4.9137,
|
|
"mean_token_accuracy": 0.19313290268182753,
|
|
"num_tokens": 9293011.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.334768295288086,
|
|
"epoch": 4.229494614747307,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003636656476432854,
|
|
"loss": 5.0256,
|
|
"mean_token_accuracy": 0.1883176565170288,
|
|
"num_tokens": 9302387.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.32701621055603,
|
|
"epoch": 4.233637116818558,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003633718603925642,
|
|
"loss": 4.9673,
|
|
"mean_token_accuracy": 0.19103406220674515,
|
|
"num_tokens": 9311292.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.356899547576904,
|
|
"epoch": 4.237779618889809,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003630778948861434,
|
|
"loss": 4.9356,
|
|
"mean_token_accuracy": 0.19458310902118683,
|
|
"num_tokens": 9319507.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.377451658248901,
|
|
"epoch": 4.2419221209610605,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003627837517169836,
|
|
"loss": 4.9828,
|
|
"mean_token_accuracy": 0.18844300508499146,
|
|
"num_tokens": 9328705.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.393976831436158,
|
|
"epoch": 4.246064623032312,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003624894314784037,
|
|
"loss": 4.958,
|
|
"mean_token_accuracy": 0.1927539050579071,
|
|
"num_tokens": 9339160.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.410513782501221,
|
|
"epoch": 4.250207125103563,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003621949347640797,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.18613501489162446,
|
|
"num_tokens": 9349088.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.335210704803467,
|
|
"epoch": 4.254349627174814,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003619002621680439,
|
|
"loss": 4.881,
|
|
"mean_token_accuracy": 0.20249925106763839,
|
|
"num_tokens": 9358320.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.3401085376739506,
|
|
"epoch": 4.258492129246065,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003616054142846829,
|
|
"loss": 4.9863,
|
|
"mean_token_accuracy": 0.19585341066122056,
|
|
"num_tokens": 9367873.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.331043338775634,
|
|
"epoch": 4.262634631317316,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003613103917087371,
|
|
"loss": 4.9902,
|
|
"mean_token_accuracy": 0.19353567510843278,
|
|
"num_tokens": 9376777.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.370423078536987,
|
|
"epoch": 4.266777133388567,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003610151950352994,
|
|
"loss": 4.9592,
|
|
"mean_token_accuracy": 0.1841620087623596,
|
|
"num_tokens": 9386125.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.341997337341309,
|
|
"epoch": 4.270919635459817,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003607198248598137,
|
|
"loss": 4.9699,
|
|
"mean_token_accuracy": 0.19745617657899855,
|
|
"num_tokens": 9394351.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.374164199829101,
|
|
"epoch": 4.2750621375310685,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003604242817780739,
|
|
"loss": 5.0167,
|
|
"mean_token_accuracy": 0.1880358412861824,
|
|
"num_tokens": 9403603.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.4496626377105715,
|
|
"epoch": 4.27920463960232,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036012856638622273,
|
|
"loss": 5.0453,
|
|
"mean_token_accuracy": 0.18761429637670518,
|
|
"num_tokens": 9412562.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.3720136165618895,
|
|
"epoch": 4.283347141673571,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003598326792807504,
|
|
"loss": 4.9605,
|
|
"mean_token_accuracy": 0.19091594368219375,
|
|
"num_tokens": 9420976.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.360024309158325,
|
|
"epoch": 4.287489643744822,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035953662105849357,
|
|
"loss": 4.9715,
|
|
"mean_token_accuracy": 0.20033646374940872,
|
|
"num_tokens": 9429978.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.333786106109619,
|
|
"epoch": 4.291632145816073,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000359240392316634,
|
|
"loss": 4.9573,
|
|
"mean_token_accuracy": 0.20098181813955307,
|
|
"num_tokens": 9440640.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.457873106002808,
|
|
"epoch": 4.295774647887324,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003589439936526974,
|
|
"loss": 4.9931,
|
|
"mean_token_accuracy": 0.18610698431730271,
|
|
"num_tokens": 9449248.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.385212850570679,
|
|
"epoch": 4.299917149958575,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003586474256645525,
|
|
"loss": 5.0109,
|
|
"mean_token_accuracy": 0.18972010761499405,
|
|
"num_tokens": 9459281.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.398737239837646,
|
|
"epoch": 4.304059652029826,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00035835068895040905,
|
|
"loss": 5.0266,
|
|
"mean_token_accuracy": 0.18628641217947006,
|
|
"num_tokens": 9467886.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.398097658157349,
|
|
"epoch": 4.308202154101077,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035805378410881753,
|
|
"loss": 4.9997,
|
|
"mean_token_accuracy": 0.18839718103408815,
|
|
"num_tokens": 9477146.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.481663417816162,
|
|
"epoch": 4.312344656172328,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003577567117386675,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.18959626853466033,
|
|
"num_tokens": 9485996.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.38418493270874,
|
|
"epoch": 4.3164871582435795,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003574594724391863,
|
|
"loss": 5.0166,
|
|
"mean_token_accuracy": 0.185970439016819,
|
|
"num_tokens": 9494881.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.389380884170532,
|
|
"epoch": 4.32062966031483,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003571620668099382,
|
|
"loss": 5.0627,
|
|
"mean_token_accuracy": 0.18504559397697448,
|
|
"num_tokens": 9503539.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.457156467437744,
|
|
"epoch": 4.324772162386081,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003568644954508227,
|
|
"loss": 5.0194,
|
|
"mean_token_accuracy": 0.18831649720668792,
|
|
"num_tokens": 9512831.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.450788068771362,
|
|
"epoch": 4.328914664457332,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003565667589620739,
|
|
"loss": 5.0764,
|
|
"mean_token_accuracy": 0.1861792117357254,
|
|
"num_tokens": 9522153.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.379431819915771,
|
|
"epoch": 4.333057166528583,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003562688579442588,
|
|
"loss": 4.9945,
|
|
"mean_token_accuracy": 0.19255556017160416,
|
|
"num_tokens": 9532524.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.3680661678314205,
|
|
"epoch": 4.337199668599834,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035597079299827625,
|
|
"loss": 4.9462,
|
|
"mean_token_accuracy": 0.1935371458530426,
|
|
"num_tokens": 9542436.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.350747394561767,
|
|
"epoch": 4.341342170671085,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00035567256472535585,
|
|
"loss": 4.9669,
|
|
"mean_token_accuracy": 0.18399111181497574,
|
|
"num_tokens": 9551814.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.360168933868408,
|
|
"epoch": 4.345484672742336,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003553741737270566,
|
|
"loss": 5.017,
|
|
"mean_token_accuracy": 0.1892317444086075,
|
|
"num_tokens": 9560877.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.387518882751465,
|
|
"epoch": 4.3496271748135875,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035507562060526574,
|
|
"loss": 4.9772,
|
|
"mean_token_accuracy": 0.18372887521982192,
|
|
"num_tokens": 9570167.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.336313772201538,
|
|
"epoch": 4.353769676884839,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035477690596219763,
|
|
"loss": 4.9318,
|
|
"mean_token_accuracy": 0.19445303827524185,
|
|
"num_tokens": 9579022.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.441338396072387,
|
|
"epoch": 4.35791217895609,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035447803040039225,
|
|
"loss": 4.9792,
|
|
"mean_token_accuracy": 0.18699862509965898,
|
|
"num_tokens": 9588965.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.4083739757537845,
|
|
"epoch": 4.362054681027341,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003541789945227143,
|
|
"loss": 5.0166,
|
|
"mean_token_accuracy": 0.19081446826457976,
|
|
"num_tokens": 9598945.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.437520837783813,
|
|
"epoch": 4.366197183098592,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035387979893235184,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.18957430124282837,
|
|
"num_tokens": 9608728.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.395904970169068,
|
|
"epoch": 4.370339685169842,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003535804442328151,
|
|
"loss": 4.9436,
|
|
"mean_token_accuracy": 0.2000102147459984,
|
|
"num_tokens": 9617271.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.326337289810181,
|
|
"epoch": 4.374482187241093,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035328093102793506,
|
|
"loss": 5.0224,
|
|
"mean_token_accuracy": 0.1848796486854553,
|
|
"num_tokens": 9627247.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.364673852920532,
|
|
"epoch": 4.378624689312344,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035298125992186283,
|
|
"loss": 5.0087,
|
|
"mean_token_accuracy": 0.18895388543605804,
|
|
"num_tokens": 9636799.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.428055620193481,
|
|
"epoch": 4.3827671913835955,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035268143151906764,
|
|
"loss": 5.0371,
|
|
"mean_token_accuracy": 0.1856800302863121,
|
|
"num_tokens": 9646080.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.339316987991333,
|
|
"epoch": 4.386909693454847,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035238144642433616,
|
|
"loss": 4.9451,
|
|
"mean_token_accuracy": 0.18947637230157852,
|
|
"num_tokens": 9655011.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.403146505355835,
|
|
"epoch": 4.391052195526098,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000352081305242771,
|
|
"loss": 5.0219,
|
|
"mean_token_accuracy": 0.18335327208042146,
|
|
"num_tokens": 9664108.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.424431657791137,
|
|
"epoch": 4.395194697597349,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035178100857978994,
|
|
"loss": 5.0373,
|
|
"mean_token_accuracy": 0.18917354494333266,
|
|
"num_tokens": 9672869.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.342849588394165,
|
|
"epoch": 4.3993371996686,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035148055704112405,
|
|
"loss": 5.0447,
|
|
"mean_token_accuracy": 0.19195736795663834,
|
|
"num_tokens": 9681759.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.483038282394409,
|
|
"epoch": 4.403479701739851,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003511799512328168,
|
|
"loss": 5.052,
|
|
"mean_token_accuracy": 0.18776207715272902,
|
|
"num_tokens": 9691141.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.531901121139526,
|
|
"epoch": 4.407622203811102,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003508791917612232,
|
|
"loss": 5.0283,
|
|
"mean_token_accuracy": 0.18074734508991241,
|
|
"num_tokens": 9700575.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.373521900177002,
|
|
"epoch": 4.411764705882353,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00035057827923300785,
|
|
"loss": 4.9753,
|
|
"mean_token_accuracy": 0.20194365829229355,
|
|
"num_tokens": 9708538.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.304724597930909,
|
|
"epoch": 4.415907207953604,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035027721425514414,
|
|
"loss": 4.9417,
|
|
"mean_token_accuracy": 0.1946966215968132,
|
|
"num_tokens": 9717581.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.357331895828247,
|
|
"epoch": 4.4200497100248555,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003499759974349131,
|
|
"loss": 4.9572,
|
|
"mean_token_accuracy": 0.19250614792108536,
|
|
"num_tokens": 9726038.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.3921912670135494,
|
|
"epoch": 4.424192212096106,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00034967462937990194,
|
|
"loss": 4.9506,
|
|
"mean_token_accuracy": 0.19206922948360444,
|
|
"num_tokens": 9733980.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.354183626174927,
|
|
"epoch": 4.428334714167357,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000349373110698003,
|
|
"loss": 4.9794,
|
|
"mean_token_accuracy": 0.19183162748813629,
|
|
"num_tokens": 9742729.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.463667011260986,
|
|
"epoch": 4.432477216238608,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003490714419974124,
|
|
"loss": 5.1317,
|
|
"mean_token_accuracy": 0.18081195801496505,
|
|
"num_tokens": 9752096.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.434142303466797,
|
|
"epoch": 4.436619718309859,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034876962388662885,
|
|
"loss": 4.9832,
|
|
"mean_token_accuracy": 0.18835036009550093,
|
|
"num_tokens": 9761668.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.317730951309204,
|
|
"epoch": 4.44076222038111,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003484676569744525,
|
|
"loss": 4.9516,
|
|
"mean_token_accuracy": 0.19123376458883284,
|
|
"num_tokens": 9770836.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.350033092498779,
|
|
"epoch": 4.444904722452361,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003481655418699836,
|
|
"loss": 4.9605,
|
|
"mean_token_accuracy": 0.1869571775197983,
|
|
"num_tokens": 9779154.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.38888726234436,
|
|
"epoch": 4.449047224523612,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034786327918262126,
|
|
"loss": 5.0287,
|
|
"mean_token_accuracy": 0.18287905007600785,
|
|
"num_tokens": 9787956.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.392240571975708,
|
|
"epoch": 4.4531897265948635,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034756086952206255,
|
|
"loss": 4.9918,
|
|
"mean_token_accuracy": 0.19192148894071578,
|
|
"num_tokens": 9799665.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.446460819244384,
|
|
"epoch": 4.457332228666115,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003472583134983005,
|
|
"loss": 5.0169,
|
|
"mean_token_accuracy": 0.18665772825479507,
|
|
"num_tokens": 9809382.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.313643121719361,
|
|
"epoch": 4.461474730737366,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00034695561172162396,
|
|
"loss": 4.9705,
|
|
"mean_token_accuracy": 0.19682691544294356,
|
|
"num_tokens": 9817168.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.41845006942749,
|
|
"epoch": 4.465617232808617,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034665276480261526,
|
|
"loss": 4.9655,
|
|
"mean_token_accuracy": 0.1940222531557083,
|
|
"num_tokens": 9825612.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.360441875457764,
|
|
"epoch": 4.469759734879867,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003463497733521499,
|
|
"loss": 4.9931,
|
|
"mean_token_accuracy": 0.18545621186494826,
|
|
"num_tokens": 9835301.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.356918573379517,
|
|
"epoch": 4.473902236951118,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003460466379813947,
|
|
"loss": 4.9998,
|
|
"mean_token_accuracy": 0.19060030430555344,
|
|
"num_tokens": 9844713.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.388667488098145,
|
|
"epoch": 4.478044739022369,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003457433593018068,
|
|
"loss": 4.9783,
|
|
"mean_token_accuracy": 0.19133242815732956,
|
|
"num_tokens": 9853607.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.435423564910889,
|
|
"epoch": 4.48218724109362,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00034543993792513243,
|
|
"loss": 5.0067,
|
|
"mean_token_accuracy": 0.1896097868680954,
|
|
"num_tokens": 9862206.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.336844873428345,
|
|
"epoch": 4.486329743164871,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003451363744634057,
|
|
"loss": 4.9325,
|
|
"mean_token_accuracy": 0.19103709161281585,
|
|
"num_tokens": 9870900.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.341982126235962,
|
|
"epoch": 4.4904722452361225,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034483266952894725,
|
|
"loss": 4.9509,
|
|
"mean_token_accuracy": 0.19399509131908416,
|
|
"num_tokens": 9879382.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.365136909484863,
|
|
"epoch": 4.494614747307374,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 5.0037,
|
|
"mean_token_accuracy": 0.18651852905750274,
|
|
"num_tokens": 9888689.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.32116322517395,
|
|
"epoch": 4.498757249378625,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003442248376925436,
|
|
"loss": 4.8845,
|
|
"mean_token_accuracy": 0.19514576047658921,
|
|
"num_tokens": 9897563.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.377099180221558,
|
|
"epoch": 4.502899751449876,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00034392071201666167,
|
|
"loss": 4.8616,
|
|
"mean_token_accuracy": 0.19910115152597427,
|
|
"num_tokens": 9906404.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.30635929107666,
|
|
"epoch": 4.507042253521127,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000343616447320172,
|
|
"loss": 4.9297,
|
|
"mean_token_accuracy": 0.196750707924366,
|
|
"num_tokens": 9915035.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.346225166320801,
|
|
"epoch": 4.511184755592378,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003433120442168097,
|
|
"loss": 5.0711,
|
|
"mean_token_accuracy": 0.18608085066080093,
|
|
"num_tokens": 9924357.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.4952003955841064,
|
|
"epoch": 4.515327257663629,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034300750332058924,
|
|
"loss": 5.0521,
|
|
"mean_token_accuracy": 0.19098345786333085,
|
|
"num_tokens": 9935514.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.465622329711914,
|
|
"epoch": 4.51946975973488,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003427028252458028,
|
|
"loss": 5.1054,
|
|
"mean_token_accuracy": 0.17619496136903762,
|
|
"num_tokens": 9944847.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.40270767211914,
|
|
"epoch": 4.5236122618061305,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003423980106070194,
|
|
"loss": 5.0786,
|
|
"mean_token_accuracy": 0.18635642230510713,
|
|
"num_tokens": 9953158.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.372509574890136,
|
|
"epoch": 4.527754763877382,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003420930600190834,
|
|
"loss": 5.015,
|
|
"mean_token_accuracy": 0.18395156562328338,
|
|
"num_tokens": 9962817.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.397466945648193,
|
|
"epoch": 4.531897265948633,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003417879740971135,
|
|
"loss": 4.9716,
|
|
"mean_token_accuracy": 0.18653742372989654,
|
|
"num_tokens": 9972454.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.351282119750977,
|
|
"epoch": 4.536039768019884,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003414827534565014,
|
|
"loss": 5.0309,
|
|
"mean_token_accuracy": 0.18586624264717103,
|
|
"num_tokens": 9981127.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.398399353027344,
|
|
"epoch": 4.540182270091135,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034117739871291036,
|
|
"loss": 5.0621,
|
|
"mean_token_accuracy": 0.18715079128742218,
|
|
"num_tokens": 9990071.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.383011150360107,
|
|
"epoch": 4.544324772162386,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034087191048227445,
|
|
"loss": 4.9959,
|
|
"mean_token_accuracy": 0.19656106531620027,
|
|
"num_tokens": 9998191.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.3476831912994385,
|
|
"epoch": 4.548467274233637,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003405662893807967,
|
|
"loss": 4.962,
|
|
"mean_token_accuracy": 0.18991767317056657,
|
|
"num_tokens": 10007503.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.35564341545105,
|
|
"epoch": 4.552609776304888,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003402605360249482,
|
|
"loss": 4.9654,
|
|
"mean_token_accuracy": 0.1986487790942192,
|
|
"num_tokens": 10016148.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.438293552398681,
|
|
"epoch": 4.556752278376139,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000339954651031467,
|
|
"loss": 5.2091,
|
|
"mean_token_accuracy": 0.18034602105617523,
|
|
"num_tokens": 10025520.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.556752278376139,
|
|
"eval_entropy": 5.29549301916094,
|
|
"eval_loss": 5.839051246643066,
|
|
"eval_mean_token_accuracy": 0.1591346115985913,
|
|
"eval_num_tokens": 10025520.0,
|
|
"eval_runtime": 2.8596,
|
|
"eval_samples_per_second": 1310.335,
|
|
"eval_steps_per_second": 164.01,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.434541320800781,
|
|
"epoch": 4.5608947804473905,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033964863501735647,
|
|
"loss": 4.9995,
|
|
"mean_token_accuracy": 0.190106800198555,
|
|
"num_tokens": 10034636.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.424151039123535,
|
|
"epoch": 4.565037282518642,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033934248859988453,
|
|
"loss": 5.0247,
|
|
"mean_token_accuracy": 0.18851222544908525,
|
|
"num_tokens": 10044290.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.4405842304229735,
|
|
"epoch": 4.569179784589892,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003390362123965817,
|
|
"loss": 5.0178,
|
|
"mean_token_accuracy": 0.18946292251348495,
|
|
"num_tokens": 10053492.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.374919605255127,
|
|
"epoch": 4.573322286661144,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003387298070252407,
|
|
"loss": 5.0131,
|
|
"mean_token_accuracy": 0.18401849418878555,
|
|
"num_tokens": 10062692.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.351923847198487,
|
|
"epoch": 4.577464788732394,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003384232731039146,
|
|
"loss": 5.0412,
|
|
"mean_token_accuracy": 0.18709686249494553,
|
|
"num_tokens": 10072948.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.337179708480835,
|
|
"epoch": 4.581607290803645,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033811661125091584,
|
|
"loss": 5.0406,
|
|
"mean_token_accuracy": 0.18689691424369811,
|
|
"num_tokens": 10081523.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.455275344848633,
|
|
"epoch": 4.585749792874896,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033780982208481494,
|
|
"loss": 5.0153,
|
|
"mean_token_accuracy": 0.18922292441129684,
|
|
"num_tokens": 10091224.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.446336555480957,
|
|
"epoch": 4.589892294946147,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003375029062244392,
|
|
"loss": 5.1624,
|
|
"mean_token_accuracy": 0.17954717427492142,
|
|
"num_tokens": 10101312.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.357786321640015,
|
|
"epoch": 4.5940347970173985,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033719586428887125,
|
|
"loss": 5.0135,
|
|
"mean_token_accuracy": 0.19209006577730178,
|
|
"num_tokens": 10110302.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.42117829322815,
|
|
"epoch": 4.59817729908865,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033688869689744836,
|
|
"loss": 5.0147,
|
|
"mean_token_accuracy": 0.18905038386583328,
|
|
"num_tokens": 10119896.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.412507772445679,
|
|
"epoch": 4.602319801159901,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003365814046697608,
|
|
"loss": 5.0157,
|
|
"mean_token_accuracy": 0.18720900267362595,
|
|
"num_tokens": 10129189.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.408407306671142,
|
|
"epoch": 4.606462303231152,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003362739882256504,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.19595430046319962,
|
|
"num_tokens": 10137148.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.349532747268677,
|
|
"epoch": 4.610604805302403,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003359664481852099,
|
|
"loss": 4.9083,
|
|
"mean_token_accuracy": 0.19885521084070207,
|
|
"num_tokens": 10145083.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.400881004333496,
|
|
"epoch": 4.614747307373654,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033565878516878103,
|
|
"loss": 5.04,
|
|
"mean_token_accuracy": 0.19142535775899888,
|
|
"num_tokens": 10154413.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.4506223678588865,
|
|
"epoch": 4.618889809444905,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003353509997969537,
|
|
"loss": 5.0311,
|
|
"mean_token_accuracy": 0.18476530462503432,
|
|
"num_tokens": 10162793.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.403977298736573,
|
|
"epoch": 4.623032311516155,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033504309269056484,
|
|
"loss": 5.0881,
|
|
"mean_token_accuracy": 0.1786352276802063,
|
|
"num_tokens": 10172077.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.490953731536865,
|
|
"epoch": 4.6271748135874065,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003347350644706965,
|
|
"loss": 5.1155,
|
|
"mean_token_accuracy": 0.17577841281890869,
|
|
"num_tokens": 10181724.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.379710483551025,
|
|
"epoch": 4.631317315658658,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003344269157586754,
|
|
"loss": 4.9388,
|
|
"mean_token_accuracy": 0.1997591257095337,
|
|
"num_tokens": 10190679.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.378966379165649,
|
|
"epoch": 4.635459817729909,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033411864717607114,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.18446184545755387,
|
|
"num_tokens": 10200953.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.372676038742066,
|
|
"epoch": 4.63960231980116,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003338102593446951,
|
|
"loss": 4.9525,
|
|
"mean_token_accuracy": 0.19693396538496016,
|
|
"num_tokens": 10209662.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.413083839416504,
|
|
"epoch": 4.643744821872411,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033350175288659925,
|
|
"loss": 5.046,
|
|
"mean_token_accuracy": 0.19485325813293458,
|
|
"num_tokens": 10218486.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.376834058761597,
|
|
"epoch": 4.647887323943662,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033319312842407483,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.18680172562599182,
|
|
"num_tokens": 10229237.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.396331167221069,
|
|
"epoch": 4.652029826014913,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003328843865796511,
|
|
"loss": 5.0232,
|
|
"mean_token_accuracy": 0.18656206578016282,
|
|
"num_tokens": 10237802.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.392931270599365,
|
|
"epoch": 4.656172328086164,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033257552797609426,
|
|
"loss": 5.0039,
|
|
"mean_token_accuracy": 0.19311116486787797,
|
|
"num_tokens": 10247145.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.299695587158203,
|
|
"epoch": 4.660314830157415,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033226655323640553,
|
|
"loss": 4.9022,
|
|
"mean_token_accuracy": 0.20144017189741134,
|
|
"num_tokens": 10256333.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.272337865829468,
|
|
"epoch": 4.664457332228666,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000331957462983821,
|
|
"loss": 4.8557,
|
|
"mean_token_accuracy": 0.19807905107736587,
|
|
"num_tokens": 10265899.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.411304378509522,
|
|
"epoch": 4.668599834299917,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003316482578418094,
|
|
"loss": 5.0512,
|
|
"mean_token_accuracy": 0.18164961636066437,
|
|
"num_tokens": 10275974.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.414994382858277,
|
|
"epoch": 4.672742336371169,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033133893843407126,
|
|
"loss": 4.9961,
|
|
"mean_token_accuracy": 0.2038290396332741,
|
|
"num_tokens": 10284882.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.382686614990234,
|
|
"epoch": 4.676884838442419,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033102950538453776,
|
|
"loss": 4.9615,
|
|
"mean_token_accuracy": 0.19490918666124343,
|
|
"num_tokens": 10293502.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.366381502151489,
|
|
"epoch": 4.68102734051367,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033071995931736897,
|
|
"loss": 5.0145,
|
|
"mean_token_accuracy": 0.19174712896347046,
|
|
"num_tokens": 10302570.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.326482772827148,
|
|
"epoch": 4.685169842584921,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00033041030085695323,
|
|
"loss": 4.88,
|
|
"mean_token_accuracy": 0.19980028122663498,
|
|
"num_tokens": 10311014.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.340007352828979,
|
|
"epoch": 4.689312344656172,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003301005306279055,
|
|
"loss": 5.0233,
|
|
"mean_token_accuracy": 0.18317166715860367,
|
|
"num_tokens": 10320356.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.4053020000457765,
|
|
"epoch": 4.693454846727423,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032979064925506627,
|
|
"loss": 5.0382,
|
|
"mean_token_accuracy": 0.19510855078697203,
|
|
"num_tokens": 10329447.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.357555484771728,
|
|
"epoch": 4.697597348798674,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003294806573634999,
|
|
"loss": 5.0722,
|
|
"mean_token_accuracy": 0.18092973828315734,
|
|
"num_tokens": 10338533.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.323354959487915,
|
|
"epoch": 4.7017398508699255,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003291705555784942,
|
|
"loss": 4.9508,
|
|
"mean_token_accuracy": 0.19252579808235168,
|
|
"num_tokens": 10347088.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.352200508117676,
|
|
"epoch": 4.705882352941177,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032886034452555814,
|
|
"loss": 4.8961,
|
|
"mean_token_accuracy": 0.20022098571062089,
|
|
"num_tokens": 10355533.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.374318552017212,
|
|
"epoch": 4.710024855012428,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003285500248304214,
|
|
"loss": 4.9744,
|
|
"mean_token_accuracy": 0.18975879549980162,
|
|
"num_tokens": 10364435.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.364798212051392,
|
|
"epoch": 4.714167357083679,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003282395971190328,
|
|
"loss": 5.0239,
|
|
"mean_token_accuracy": 0.18888593912124635,
|
|
"num_tokens": 10374276.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.358923816680909,
|
|
"epoch": 4.71830985915493,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032792906201755876,
|
|
"loss": 4.9145,
|
|
"mean_token_accuracy": 0.1939673438668251,
|
|
"num_tokens": 10383149.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.362069797515869,
|
|
"epoch": 4.72245236122618,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032761842015238256,
|
|
"loss": 5.0092,
|
|
"mean_token_accuracy": 0.1894121840596199,
|
|
"num_tokens": 10392904.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.384946203231811,
|
|
"epoch": 4.726594863297431,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003273076721501029,
|
|
"loss": 4.9832,
|
|
"mean_token_accuracy": 0.19319511353969573,
|
|
"num_tokens": 10401618.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.3807319641113285,
|
|
"epoch": 4.730737365368682,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003269968186375325,
|
|
"loss": 4.9666,
|
|
"mean_token_accuracy": 0.20137769877910613,
|
|
"num_tokens": 10410438.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.432919025421143,
|
|
"epoch": 4.7348798674399335,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003266858602416967,
|
|
"loss": 5.029,
|
|
"mean_token_accuracy": 0.1950489416718483,
|
|
"num_tokens": 10419292.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.401694440841675,
|
|
"epoch": 4.739022369511185,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032637479758983255,
|
|
"loss": 4.978,
|
|
"mean_token_accuracy": 0.19333263784646987,
|
|
"num_tokens": 10428581.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.390124273300171,
|
|
"epoch": 4.743164871582436,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00032606363130938766,
|
|
"loss": 5.0547,
|
|
"mean_token_accuracy": 0.190736848115921,
|
|
"num_tokens": 10437613.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.322377777099609,
|
|
"epoch": 4.747307373653687,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003257523620280181,
|
|
"loss": 4.9904,
|
|
"mean_token_accuracy": 0.18828242123126984,
|
|
"num_tokens": 10446480.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.422512006759644,
|
|
"epoch": 4.751449875724938,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032544099037358834,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.18698132932186126,
|
|
"num_tokens": 10454543.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.372480297088623,
|
|
"epoch": 4.755592377796189,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003251295169741688,
|
|
"loss": 4.9718,
|
|
"mean_token_accuracy": 0.18937113136053085,
|
|
"num_tokens": 10463600.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.3019531726837155,
|
|
"epoch": 4.75973487986744,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032481794245803536,
|
|
"loss": 4.955,
|
|
"mean_token_accuracy": 0.19356780201196672,
|
|
"num_tokens": 10472594.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.436426591873169,
|
|
"epoch": 4.763877381938691,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00032450626745366806,
|
|
"loss": 5.0672,
|
|
"mean_token_accuracy": 0.18057732731103898,
|
|
"num_tokens": 10481374.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.439925479888916,
|
|
"epoch": 4.768019884009942,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032419449258974924,
|
|
"loss": 5.0193,
|
|
"mean_token_accuracy": 0.1930960848927498,
|
|
"num_tokens": 10490444.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.381526708602905,
|
|
"epoch": 4.7721623860811935,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000323882618495163,
|
|
"loss": 5.0635,
|
|
"mean_token_accuracy": 0.1872535914182663,
|
|
"num_tokens": 10500623.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.251973581314087,
|
|
"epoch": 4.776304888152444,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003235706457989934,
|
|
"loss": 4.8886,
|
|
"mean_token_accuracy": 0.19448312669992446,
|
|
"num_tokens": 10509361.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.425653600692749,
|
|
"epoch": 4.780447390223695,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032325857513052354,
|
|
"loss": 5.0382,
|
|
"mean_token_accuracy": 0.18274351507425307,
|
|
"num_tokens": 10518390.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.37084927558899,
|
|
"epoch": 4.784589892294946,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000322946407119234,
|
|
"loss": 4.9745,
|
|
"mean_token_accuracy": 0.18968174904584884,
|
|
"num_tokens": 10527826.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.388440895080566,
|
|
"epoch": 4.788732394366197,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032263414239480187,
|
|
"loss": 5.0055,
|
|
"mean_token_accuracy": 0.18278948217630386,
|
|
"num_tokens": 10537445.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.405228328704834,
|
|
"epoch": 4.792874896437448,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000322321781587099,
|
|
"loss": 4.9994,
|
|
"mean_token_accuracy": 0.1862544298171997,
|
|
"num_tokens": 10546810.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.396326160430908,
|
|
"epoch": 4.797017398508699,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003220093253261915,
|
|
"loss": 5.0113,
|
|
"mean_token_accuracy": 0.1809608206152916,
|
|
"num_tokens": 10555764.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.418875503540039,
|
|
"epoch": 4.80115990057995,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003216967742423377,
|
|
"loss": 5.0309,
|
|
"mean_token_accuracy": 0.1846464082598686,
|
|
"num_tokens": 10565371.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.389196443557739,
|
|
"epoch": 4.8053024026512015,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003213841289659872,
|
|
"loss": 5.0273,
|
|
"mean_token_accuracy": 0.19190154522657393,
|
|
"num_tokens": 10573897.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.364774560928344,
|
|
"epoch": 4.809444904722453,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003210713901277799,
|
|
"loss": 5.0229,
|
|
"mean_token_accuracy": 0.19107308536767958,
|
|
"num_tokens": 10582665.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.376154899597168,
|
|
"epoch": 4.813587406793704,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000320758558358544,
|
|
"loss": 5.0237,
|
|
"mean_token_accuracy": 0.19199632108211517,
|
|
"num_tokens": 10591606.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.3657880306243895,
|
|
"epoch": 4.817729908864955,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003204456342892955,
|
|
"loss": 4.9085,
|
|
"mean_token_accuracy": 0.19690821021795274,
|
|
"num_tokens": 10600884.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.334457206726074,
|
|
"epoch": 4.821872410936205,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032013261855123637,
|
|
"loss": 4.9146,
|
|
"mean_token_accuracy": 0.19975414127111435,
|
|
"num_tokens": 10609310.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.3085558891296385,
|
|
"epoch": 4.826014913007457,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00031981951177575366,
|
|
"loss": 4.9105,
|
|
"mean_token_accuracy": 0.1986094146966934,
|
|
"num_tokens": 10617819.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.349913835525513,
|
|
"epoch": 4.830157415078707,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003195063145944178,
|
|
"loss": 4.9443,
|
|
"mean_token_accuracy": 0.19787970185279846,
|
|
"num_tokens": 10626082.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.332488346099853,
|
|
"epoch": 4.834299917149958,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003191930276389819,
|
|
"loss": 5.0369,
|
|
"mean_token_accuracy": 0.18846531361341476,
|
|
"num_tokens": 10634585.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.372330617904663,
|
|
"epoch": 4.838442419221209,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031887965154137987,
|
|
"loss": 5.0242,
|
|
"mean_token_accuracy": 0.18838179260492324,
|
|
"num_tokens": 10643709.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.38498740196228,
|
|
"epoch": 4.8425849212924605,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003185661869337255,
|
|
"loss": 4.8849,
|
|
"mean_token_accuracy": 0.19826163202524186,
|
|
"num_tokens": 10652879.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.4045037746429445,
|
|
"epoch": 4.846727423363712,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00031825263444831137,
|
|
"loss": 4.9569,
|
|
"mean_token_accuracy": 0.1875440865755081,
|
|
"num_tokens": 10661757.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.297507333755493,
|
|
"epoch": 4.850869925434963,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003179389947176069,
|
|
"loss": 5.0132,
|
|
"mean_token_accuracy": 0.1910937875509262,
|
|
"num_tokens": 10670233.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.3350501537323,
|
|
"epoch": 4.855012427506214,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003176252683742578,
|
|
"loss": 5.0799,
|
|
"mean_token_accuracy": 0.18518346548080444,
|
|
"num_tokens": 10680732.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.367203187942505,
|
|
"epoch": 4.859154929577465,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003173114560510843,
|
|
"loss": 4.9504,
|
|
"mean_token_accuracy": 0.1950843408703804,
|
|
"num_tokens": 10689729.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.4245607376098635,
|
|
"epoch": 4.863297431648716,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031699755838108027,
|
|
"loss": 5.0663,
|
|
"mean_token_accuracy": 0.18715314269065858,
|
|
"num_tokens": 10698828.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.442091703414917,
|
|
"epoch": 4.867439933719967,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003166835759974115,
|
|
"loss": 5.1052,
|
|
"mean_token_accuracy": 0.1857593312859535,
|
|
"num_tokens": 10708142.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.4029889583587645,
|
|
"epoch": 4.871582435791218,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031636950953341486,
|
|
"loss": 5.0355,
|
|
"mean_token_accuracy": 0.1902431935071945,
|
|
"num_tokens": 10716708.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.331116342544556,
|
|
"epoch": 4.8757249378624685,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031605535962259666,
|
|
"loss": 4.8903,
|
|
"mean_token_accuracy": 0.20048092305660248,
|
|
"num_tokens": 10724800.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.34693865776062,
|
|
"epoch": 4.87986743993372,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003157411268986316,
|
|
"loss": 5.017,
|
|
"mean_token_accuracy": 0.18317356705665588,
|
|
"num_tokens": 10733553.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.4325127601623535,
|
|
"epoch": 4.884009942004971,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003154268119953615,
|
|
"loss": 5.0367,
|
|
"mean_token_accuracy": 0.19113590568304062,
|
|
"num_tokens": 10743501.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.307816410064698,
|
|
"epoch": 4.888152444076222,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003151124155467938,
|
|
"loss": 4.9916,
|
|
"mean_token_accuracy": 0.188314950466156,
|
|
"num_tokens": 10752307.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.306516122817993,
|
|
"epoch": 4.892294946147473,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031479793818710056,
|
|
"loss": 4.9227,
|
|
"mean_token_accuracy": 0.19994584769010543,
|
|
"num_tokens": 10760547.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.463548946380615,
|
|
"epoch": 4.896437448218724,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00031448338055061685,
|
|
"loss": 5.0293,
|
|
"mean_token_accuracy": 0.19401468634605407,
|
|
"num_tokens": 10769522.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.357081317901612,
|
|
"epoch": 4.900579950289975,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00031416874327183994,
|
|
"loss": 4.9543,
|
|
"mean_token_accuracy": 0.19806358218193054,
|
|
"num_tokens": 10778292.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.372450304031372,
|
|
"epoch": 4.904722452361226,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031385402698542753,
|
|
"loss": 5.0711,
|
|
"mean_token_accuracy": 0.1864228755235672,
|
|
"num_tokens": 10787927.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.373929691314697,
|
|
"epoch": 4.908864954432477,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031353923232619674,
|
|
"loss": 5.0207,
|
|
"mean_token_accuracy": 0.19643336534500122,
|
|
"num_tokens": 10796781.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.337554121017456,
|
|
"epoch": 4.9130074565037285,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003132243599291229,
|
|
"loss": 4.9218,
|
|
"mean_token_accuracy": 0.19281717985868455,
|
|
"num_tokens": 10805118.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.3288023471832275,
|
|
"epoch": 4.91714995857498,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031290941042933797,
|
|
"loss": 4.943,
|
|
"mean_token_accuracy": 0.1897909387946129,
|
|
"num_tokens": 10814189.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.429559898376465,
|
|
"epoch": 4.92129246064623,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031259438446212953,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.1792408987879753,
|
|
"num_tokens": 10824236.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.402896070480347,
|
|
"epoch": 4.925434962717482,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00031227928266293933,
|
|
"loss": 4.9779,
|
|
"mean_token_accuracy": 0.19006628394126893,
|
|
"num_tokens": 10833005.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.334726190567016,
|
|
"epoch": 4.929577464788732,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031196410566736215,
|
|
"loss": 4.9789,
|
|
"mean_token_accuracy": 0.18983061611652374,
|
|
"num_tokens": 10842385.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.226403284072876,
|
|
"epoch": 4.933719966859983,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031164885411114453,
|
|
"loss": 4.8769,
|
|
"mean_token_accuracy": 0.19544212520122528,
|
|
"num_tokens": 10851606.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.429797077178955,
|
|
"epoch": 4.937862468931234,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031133352863018315,
|
|
"loss": 5.1776,
|
|
"mean_token_accuracy": 0.1851620852947235,
|
|
"num_tokens": 10861097.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.411133718490601,
|
|
"epoch": 4.942004971002485,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031101812986052406,
|
|
"loss": 4.9313,
|
|
"mean_token_accuracy": 0.20171212255954743,
|
|
"num_tokens": 10869621.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.34544153213501,
|
|
"epoch": 4.9461474730737365,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031070265843836107,
|
|
"loss": 5.0126,
|
|
"mean_token_accuracy": 0.18955454975366592,
|
|
"num_tokens": 10878841.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.390292310714722,
|
|
"epoch": 4.950289975144988,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003103871150000345,
|
|
"loss": 4.9738,
|
|
"mean_token_accuracy": 0.1922559916973114,
|
|
"num_tokens": 10887679.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.319056653976441,
|
|
"epoch": 4.954432477216239,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00031007150018202983,
|
|
"loss": 4.9405,
|
|
"mean_token_accuracy": 0.19547946602106095,
|
|
"num_tokens": 10896591.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.405071210861206,
|
|
"epoch": 4.95857497928749,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030975581462097673,
|
|
"loss": 5.0093,
|
|
"mean_token_accuracy": 0.19332779049873353,
|
|
"num_tokens": 10906144.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.419402313232422,
|
|
"epoch": 4.962717481358741,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030944005895364754,
|
|
"loss": 5.112,
|
|
"mean_token_accuracy": 0.17851991653442384,
|
|
"num_tokens": 10915602.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.369933986663819,
|
|
"epoch": 4.966859983429992,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030912423381695584,
|
|
"loss": 5.0511,
|
|
"mean_token_accuracy": 0.18782127946615218,
|
|
"num_tokens": 10925358.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.42211880683899,
|
|
"epoch": 4.971002485501243,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030880833984795554,
|
|
"loss": 5.0457,
|
|
"mean_token_accuracy": 0.18852933049201964,
|
|
"num_tokens": 10934587.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 4.971002485501243,
|
|
"eval_entropy": 5.2326576165807275,
|
|
"eval_loss": 5.807068347930908,
|
|
"eval_mean_token_accuracy": 0.16080639087187965,
|
|
"eval_num_tokens": 10934587.0,
|
|
"eval_runtime": 2.7605,
|
|
"eval_samples_per_second": 1357.357,
|
|
"eval_steps_per_second": 169.896,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.326371860504151,
|
|
"epoch": 4.975144987572493,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003084923776838392,
|
|
"loss": 4.9928,
|
|
"mean_token_accuracy": 0.19638096541166306,
|
|
"num_tokens": 10944386.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.426962518692017,
|
|
"epoch": 4.9792874896437445,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030817634796193705,
|
|
"loss": 4.9795,
|
|
"mean_token_accuracy": 0.19106023013591766,
|
|
"num_tokens": 10953322.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.36447057723999,
|
|
"epoch": 4.983429991714996,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030786025131971555,
|
|
"loss": 4.8686,
|
|
"mean_token_accuracy": 0.19176887571811677,
|
|
"num_tokens": 10962922.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.297856760025025,
|
|
"epoch": 4.987572493786247,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003075440883947764,
|
|
"loss": 5.0155,
|
|
"mean_token_accuracy": 0.1848055377602577,
|
|
"num_tokens": 10972532.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.3936295986175535,
|
|
"epoch": 4.991714995857498,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003072278598248544,
|
|
"loss": 5.0899,
|
|
"mean_token_accuracy": 0.17636170238256454,
|
|
"num_tokens": 10982625.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.373222303390503,
|
|
"epoch": 4.995857497928749,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030691156624781734,
|
|
"loss": 5.0597,
|
|
"mean_token_accuracy": 0.17941006422042846,
|
|
"num_tokens": 10991933.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.4459065914154055,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000306595208301664,
|
|
"loss": 5.0099,
|
|
"mean_token_accuracy": 0.17807235717773437,
|
|
"num_tokens": 11000005.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.361193704605102,
|
|
"epoch": 5.004142502071251,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00030627878662452273,
|
|
"loss": 4.7274,
|
|
"mean_token_accuracy": 0.20889190584421158,
|
|
"num_tokens": 11008283.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.3238592624664305,
|
|
"epoch": 5.008285004142502,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003059623018546508,
|
|
"loss": 4.7625,
|
|
"mean_token_accuracy": 0.2126881569623947,
|
|
"num_tokens": 11017165.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.436987638473511,
|
|
"epoch": 5.012427506213753,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003056457546304326,
|
|
"loss": 4.8683,
|
|
"mean_token_accuracy": 0.20007142573595046,
|
|
"num_tokens": 11025842.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.379302501678467,
|
|
"epoch": 5.016570008285004,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003053291455903782,
|
|
"loss": 4.8009,
|
|
"mean_token_accuracy": 0.1963910847902298,
|
|
"num_tokens": 11034142.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.30905704498291,
|
|
"epoch": 5.0207125103562555,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00030501247537312293,
|
|
"loss": 4.7291,
|
|
"mean_token_accuracy": 0.20740348398685454,
|
|
"num_tokens": 11041821.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.289470767974853,
|
|
"epoch": 5.024855012427507,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003046957446174251,
|
|
"loss": 4.7233,
|
|
"mean_token_accuracy": 0.20378533154726028,
|
|
"num_tokens": 11050574.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.378904581069946,
|
|
"epoch": 5.028997514498757,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030437895396216524,
|
|
"loss": 4.7349,
|
|
"mean_token_accuracy": 0.21469833254814147,
|
|
"num_tokens": 11059875.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.308254384994507,
|
|
"epoch": 5.033140016570008,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030406210404634475,
|
|
"loss": 4.6847,
|
|
"mean_token_accuracy": 0.21084672808647156,
|
|
"num_tokens": 11068126.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.30233678817749,
|
|
"epoch": 5.037282518641259,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003037451955090846,
|
|
"loss": 4.712,
|
|
"mean_token_accuracy": 0.21777195036411284,
|
|
"num_tokens": 11076571.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.276848268508911,
|
|
"epoch": 5.04142502071251,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030342822898962377,
|
|
"loss": 4.7169,
|
|
"mean_token_accuracy": 0.21495039463043214,
|
|
"num_tokens": 11085788.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.343266487121582,
|
|
"epoch": 5.045567522783761,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003031112051273185,
|
|
"loss": 4.7931,
|
|
"mean_token_accuracy": 0.20406606793403625,
|
|
"num_tokens": 11095838.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.327229833602905,
|
|
"epoch": 5.049710024855012,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003027941245616405,
|
|
"loss": 4.7871,
|
|
"mean_token_accuracy": 0.20866283178329467,
|
|
"num_tokens": 11105791.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.418568706512451,
|
|
"epoch": 5.0538525269262635,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003024769879321761,
|
|
"loss": 4.9043,
|
|
"mean_token_accuracy": 0.19387636035680772,
|
|
"num_tokens": 11114990.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.3424934387207035,
|
|
"epoch": 5.057995028997515,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003021597958786243,
|
|
"loss": 4.7173,
|
|
"mean_token_accuracy": 0.2181010901927948,
|
|
"num_tokens": 11124452.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.290416240692139,
|
|
"epoch": 5.062137531068766,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003018425490407962,
|
|
"loss": 4.7204,
|
|
"mean_token_accuracy": 0.21197246462106706,
|
|
"num_tokens": 11133729.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.265803289413452,
|
|
"epoch": 5.066280033140017,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030152524805861344,
|
|
"loss": 4.7317,
|
|
"mean_token_accuracy": 0.2222418427467346,
|
|
"num_tokens": 11143084.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.252679061889649,
|
|
"epoch": 5.070422535211268,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030120789357210667,
|
|
"loss": 4.6327,
|
|
"mean_token_accuracy": 0.21469631493091584,
|
|
"num_tokens": 11151288.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.30620059967041,
|
|
"epoch": 5.074565037282519,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030089048622141453,
|
|
"loss": 4.7863,
|
|
"mean_token_accuracy": 0.20942170470952987,
|
|
"num_tokens": 11159990.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.346095418930053,
|
|
"epoch": 5.078707539353769,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030057302664678243,
|
|
"loss": 4.7574,
|
|
"mean_token_accuracy": 0.20667121559381485,
|
|
"num_tokens": 11170254.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.381305074691772,
|
|
"epoch": 5.08285004142502,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00030025551548856105,
|
|
"loss": 4.8169,
|
|
"mean_token_accuracy": 0.20504982322454451,
|
|
"num_tokens": 11179937.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.398670482635498,
|
|
"epoch": 5.0869925434962715,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002999379533872051,
|
|
"loss": 4.7799,
|
|
"mean_token_accuracy": 0.19865742921829224,
|
|
"num_tokens": 11188352.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.285335206985474,
|
|
"epoch": 5.091135045567523,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002996203409832719,
|
|
"loss": 4.7109,
|
|
"mean_token_accuracy": 0.20933137685060502,
|
|
"num_tokens": 11196767.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.321043062210083,
|
|
"epoch": 5.095277547638774,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002993026789174205,
|
|
"loss": 4.731,
|
|
"mean_token_accuracy": 0.20679521560668945,
|
|
"num_tokens": 11205289.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.40277247428894,
|
|
"epoch": 5.099420049710025,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002989849678304099,
|
|
"loss": 4.7513,
|
|
"mean_token_accuracy": 0.2035286992788315,
|
|
"num_tokens": 11214287.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.336764335632324,
|
|
"epoch": 5.103562551781276,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029866720836309817,
|
|
"loss": 4.7097,
|
|
"mean_token_accuracy": 0.21619651913642884,
|
|
"num_tokens": 11223953.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.279402875900269,
|
|
"epoch": 5.107705053852527,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029834940115644086,
|
|
"loss": 4.8052,
|
|
"mean_token_accuracy": 0.2001069575548172,
|
|
"num_tokens": 11233446.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.354794597625732,
|
|
"epoch": 5.111847555923778,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002980315468514898,
|
|
"loss": 4.8249,
|
|
"mean_token_accuracy": 0.20808946639299392,
|
|
"num_tokens": 11243301.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.353132486343384,
|
|
"epoch": 5.115990057995029,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029771364608939196,
|
|
"loss": 4.7748,
|
|
"mean_token_accuracy": 0.2021961435675621,
|
|
"num_tokens": 11252375.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.302717208862305,
|
|
"epoch": 5.12013256006628,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002973956995113878,
|
|
"loss": 4.8009,
|
|
"mean_token_accuracy": 0.20373091995716094,
|
|
"num_tokens": 11261739.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.26840386390686,
|
|
"epoch": 5.1242750621375315,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002970777077588103,
|
|
"loss": 4.788,
|
|
"mean_token_accuracy": 0.2127612680196762,
|
|
"num_tokens": 11271324.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.295388555526733,
|
|
"epoch": 5.128417564208782,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002967596714730837,
|
|
"loss": 4.742,
|
|
"mean_token_accuracy": 0.20528868734836578,
|
|
"num_tokens": 11280808.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.329572582244873,
|
|
"epoch": 5.132560066280033,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029644159129572185,
|
|
"loss": 4.8181,
|
|
"mean_token_accuracy": 0.20702140927314758,
|
|
"num_tokens": 11290309.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.34248538017273,
|
|
"epoch": 5.136702568351284,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002961234678683272,
|
|
"loss": 4.7466,
|
|
"mean_token_accuracy": 0.2036570727825165,
|
|
"num_tokens": 11299105.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.324543237686157,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002958053018325896,
|
|
"loss": 4.7919,
|
|
"mean_token_accuracy": 0.20606669038534164,
|
|
"num_tokens": 11308072.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.3552039623260494,
|
|
"epoch": 5.144987572493786,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002954870938302846,
|
|
"loss": 4.8559,
|
|
"mean_token_accuracy": 0.1987327292561531,
|
|
"num_tokens": 11317169.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.250323486328125,
|
|
"epoch": 5.149130074565037,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002951688445032726,
|
|
"loss": 4.6986,
|
|
"mean_token_accuracy": 0.21553778648376465,
|
|
"num_tokens": 11325137.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.287027597427368,
|
|
"epoch": 5.153272576636288,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002948505544934971,
|
|
"loss": 4.7961,
|
|
"mean_token_accuracy": 0.21625454872846603,
|
|
"num_tokens": 11334640.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.2852967262268065,
|
|
"epoch": 5.1574150787075395,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000294532224442984,
|
|
"loss": 4.772,
|
|
"mean_token_accuracy": 0.20753686726093293,
|
|
"num_tokens": 11343829.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.26735372543335,
|
|
"epoch": 5.161557580778791,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029421385499383976,
|
|
"loss": 4.6753,
|
|
"mean_token_accuracy": 0.2112901821732521,
|
|
"num_tokens": 11352079.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.354950428009033,
|
|
"epoch": 5.165700082850042,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029389544678825035,
|
|
"loss": 4.8005,
|
|
"mean_token_accuracy": 0.19910743236541747,
|
|
"num_tokens": 11362947.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.370255327224731,
|
|
"epoch": 5.169842584921293,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00029357700046847985,
|
|
"loss": 4.7832,
|
|
"mean_token_accuracy": 0.2012724071741104,
|
|
"num_tokens": 11371950.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.388430404663086,
|
|
"epoch": 5.173985086992544,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002932585166768694,
|
|
"loss": 4.976,
|
|
"mean_token_accuracy": 0.18903279900550843,
|
|
"num_tokens": 11381057.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.358771896362304,
|
|
"epoch": 5.178127589063794,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029293999605583545,
|
|
"loss": 4.8016,
|
|
"mean_token_accuracy": 0.1964482307434082,
|
|
"num_tokens": 11389715.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.48654727935791,
|
|
"epoch": 5.182270091135045,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00029262143924786905,
|
|
"loss": 4.8711,
|
|
"mean_token_accuracy": 0.20317432433366775,
|
|
"num_tokens": 11398843.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.44269585609436,
|
|
"epoch": 5.186412593206296,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00029230284689553404,
|
|
"loss": 4.875,
|
|
"mean_token_accuracy": 0.20180386751890184,
|
|
"num_tokens": 11408658.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.276189136505127,
|
|
"epoch": 5.190555095277547,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029198421964146605,
|
|
"loss": 4.7827,
|
|
"mean_token_accuracy": 0.2000276193022728,
|
|
"num_tokens": 11417633.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.303118085861206,
|
|
"epoch": 5.1946975973487985,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002916655581283709,
|
|
"loss": 4.7872,
|
|
"mean_token_accuracy": 0.20193638354539872,
|
|
"num_tokens": 11426359.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.347332525253296,
|
|
"epoch": 5.19884009942005,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029134686299902397,
|
|
"loss": 4.7748,
|
|
"mean_token_accuracy": 0.21312018632888793,
|
|
"num_tokens": 11436047.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.31416335105896,
|
|
"epoch": 5.202982601491301,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002910281348962679,
|
|
"loss": 4.8034,
|
|
"mean_token_accuracy": 0.2082650691270828,
|
|
"num_tokens": 11446079.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.3646527290344235,
|
|
"epoch": 5.207125103562552,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00029070937446301225,
|
|
"loss": 4.7937,
|
|
"mean_token_accuracy": 0.20683704018592836,
|
|
"num_tokens": 11455765.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.33660569190979,
|
|
"epoch": 5.211267605633803,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002903905823422317,
|
|
"loss": 4.7397,
|
|
"mean_token_accuracy": 0.21072338670492172,
|
|
"num_tokens": 11464037.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.37380633354187,
|
|
"epoch": 5.215410107705054,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002900717591769646,
|
|
"loss": 4.8146,
|
|
"mean_token_accuracy": 0.2108115077018738,
|
|
"num_tokens": 11473536.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.39555549621582,
|
|
"epoch": 5.219552609776305,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002897529056103122,
|
|
"loss": 4.9076,
|
|
"mean_token_accuracy": 0.2061152160167694,
|
|
"num_tokens": 11482689.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.334890460968017,
|
|
"epoch": 5.223695111847556,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002894340222854371,
|
|
"loss": 4.7875,
|
|
"mean_token_accuracy": 0.20945750772953034,
|
|
"num_tokens": 11491755.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.232196903228759,
|
|
"epoch": 5.2278376139188065,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002891151098455617,
|
|
"loss": 4.7469,
|
|
"mean_token_accuracy": 0.20961148291826248,
|
|
"num_tokens": 11500816.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.325902509689331,
|
|
"epoch": 5.231980115990058,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002887961689339672,
|
|
"loss": 4.7972,
|
|
"mean_token_accuracy": 0.19985350370407104,
|
|
"num_tokens": 11509528.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.299388313293457,
|
|
"epoch": 5.236122618061309,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002884772001939924,
|
|
"loss": 4.6854,
|
|
"mean_token_accuracy": 0.21243221014738084,
|
|
"num_tokens": 11518420.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.26543960571289,
|
|
"epoch": 5.24026512013256,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000288158204269032,
|
|
"loss": 4.7606,
|
|
"mean_token_accuracy": 0.2050339847803116,
|
|
"num_tokens": 11528395.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.327279424667358,
|
|
"epoch": 5.244407622203811,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00028783918180253567,
|
|
"loss": 4.8375,
|
|
"mean_token_accuracy": 0.20284549295902252,
|
|
"num_tokens": 11537798.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.4203307151794435,
|
|
"epoch": 5.248550124275062,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002875201334380066,
|
|
"loss": 4.8774,
|
|
"mean_token_accuracy": 0.1951832354068756,
|
|
"num_tokens": 11546848.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.30546088218689,
|
|
"epoch": 5.252692626346313,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002872010598190002,
|
|
"loss": 4.7413,
|
|
"mean_token_accuracy": 0.20688969045877456,
|
|
"num_tokens": 11556124.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.271653842926026,
|
|
"epoch": 5.256835128417564,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002868819615891229,
|
|
"loss": 4.7694,
|
|
"mean_token_accuracy": 0.20747413635253906,
|
|
"num_tokens": 11564890.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.30571141242981,
|
|
"epoch": 5.260977630488815,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002865628393920305,
|
|
"loss": 4.7373,
|
|
"mean_token_accuracy": 0.20634693652391434,
|
|
"num_tokens": 11573760.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.3646307468414305,
|
|
"epoch": 5.2651201325600665,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002862436938714276,
|
|
"loss": 4.913,
|
|
"mean_token_accuracy": 0.19641882926225662,
|
|
"num_tokens": 11583459.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.246608066558838,
|
|
"epoch": 5.269262634631318,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002859245256710654,
|
|
"loss": 4.7337,
|
|
"mean_token_accuracy": 0.20982263386249542,
|
|
"num_tokens": 11593116.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.324666261672974,
|
|
"epoch": 5.273405136702569,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002856053354347411,
|
|
"loss": 4.8094,
|
|
"mean_token_accuracy": 0.20848099142313004,
|
|
"num_tokens": 11602814.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.2771008014678955,
|
|
"epoch": 5.27754763877382,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002852861238062963,
|
|
"loss": 4.7217,
|
|
"mean_token_accuracy": 0.21635773181915283,
|
|
"num_tokens": 11612596.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.320498895645142,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002849668914296157,
|
|
"loss": 4.7599,
|
|
"mean_token_accuracy": 0.2122380629181862,
|
|
"num_tokens": 11620687.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.272141981124878,
|
|
"epoch": 5.285832642916321,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000284647638948626,
|
|
"loss": 4.7481,
|
|
"mean_token_accuracy": 0.20964501798152924,
|
|
"num_tokens": 11631013.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.30500898361206,
|
|
"epoch": 5.289975144987572,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002843283670072942,
|
|
"loss": 4.7703,
|
|
"mean_token_accuracy": 0.21014779061079025,
|
|
"num_tokens": 11640206.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.335312986373902,
|
|
"epoch": 5.294117647058823,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002840090762496268,
|
|
"loss": 4.8055,
|
|
"mean_token_accuracy": 0.21077920198440553,
|
|
"num_tokens": 11648679.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.336844825744629,
|
|
"epoch": 5.2982601491300745,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002836897673196681,
|
|
"loss": 4.7816,
|
|
"mean_token_accuracy": 0.20302717685699462,
|
|
"num_tokens": 11658200.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.330560827255249,
|
|
"epoch": 5.302402651201326,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028337044086149913,
|
|
"loss": 4.8129,
|
|
"mean_token_accuracy": 0.20258800089359283,
|
|
"num_tokens": 11667900.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.269723033905029,
|
|
"epoch": 5.306545153272577,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00028305109751923616,
|
|
"loss": 4.8071,
|
|
"mean_token_accuracy": 0.20401435494422912,
|
|
"num_tokens": 11677267.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.324369192123413,
|
|
"epoch": 5.310687655343828,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028273173793702967,
|
|
"loss": 4.7887,
|
|
"mean_token_accuracy": 0.20086884796619414,
|
|
"num_tokens": 11687026.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.3616804599761965,
|
|
"epoch": 5.314830157415079,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002824123627590629,
|
|
"loss": 4.8458,
|
|
"mean_token_accuracy": 0.20515393316745759,
|
|
"num_tokens": 11696127.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.301491069793701,
|
|
"epoch": 5.31897265948633,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002820929726295504,
|
|
"loss": 4.7323,
|
|
"mean_token_accuracy": 0.21000828444957734,
|
|
"num_tokens": 11704592.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.326675367355347,
|
|
"epoch": 5.323115161557581,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002817735681927371,
|
|
"loss": 4.7682,
|
|
"mean_token_accuracy": 0.2125292167067528,
|
|
"num_tokens": 11713441.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.248379945755005,
|
|
"epoch": 5.327257663628831,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028145415009289644,
|
|
"loss": 4.7566,
|
|
"mean_token_accuracy": 0.20854133516550064,
|
|
"num_tokens": 11722665.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.334552574157715,
|
|
"epoch": 5.3314001657000825,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002811347189743297,
|
|
"loss": 4.7899,
|
|
"mean_token_accuracy": 0.19762199968099595,
|
|
"num_tokens": 11731654.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.324225330352784,
|
|
"epoch": 5.335542667771334,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00028081527548136446,
|
|
"loss": 4.8084,
|
|
"mean_token_accuracy": 0.2041504606604576,
|
|
"num_tokens": 11740072.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.339646100997925,
|
|
"epoch": 5.339685169842585,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002804958202583531,
|
|
"loss": 4.7614,
|
|
"mean_token_accuracy": 0.21011618077754973,
|
|
"num_tokens": 11748982.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.3193443775177,
|
|
"epoch": 5.343827671913836,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002801763539496718,
|
|
"loss": 4.7938,
|
|
"mean_token_accuracy": 0.2058758944272995,
|
|
"num_tokens": 11757961.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.305312824249268,
|
|
"epoch": 5.347970173985087,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002798568771997188,
|
|
"loss": 4.7867,
|
|
"mean_token_accuracy": 0.2002536326646805,
|
|
"num_tokens": 11767722.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.314368724822998,
|
|
"epoch": 5.352112676056338,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002795373906529137,
|
|
"loss": 4.796,
|
|
"mean_token_accuracy": 0.20524593442678452,
|
|
"num_tokens": 11777466.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.327050828933716,
|
|
"epoch": 5.356255178127589,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002792178949536959,
|
|
"loss": 4.8347,
|
|
"mean_token_accuracy": 0.19547835141420364,
|
|
"num_tokens": 11787617.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.346150493621826,
|
|
"epoch": 5.36039768019884,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000278898390746523,
|
|
"loss": 4.8353,
|
|
"mean_token_accuracy": 0.20629490166902542,
|
|
"num_tokens": 11796313.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.335114002227783,
|
|
"epoch": 5.364540182270091,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027857887867587005,
|
|
"loss": 4.814,
|
|
"mean_token_accuracy": 0.21321912109851837,
|
|
"num_tokens": 11805450.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.3299966812133786,
|
|
"epoch": 5.368682684341342,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002782593593862276,
|
|
"loss": 4.8627,
|
|
"mean_token_accuracy": 0.21006193161010742,
|
|
"num_tokens": 11815264.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.3146003723144535,
|
|
"epoch": 5.3728251864125935,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002779398335221011,
|
|
"loss": 4.7091,
|
|
"mean_token_accuracy": 0.2059704542160034,
|
|
"num_tokens": 11823958.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.237430047988892,
|
|
"epoch": 5.376967688483845,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002776203017280092,
|
|
"loss": 4.7055,
|
|
"mean_token_accuracy": 0.21215224266052246,
|
|
"num_tokens": 11831621.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.281644010543824,
|
|
"epoch": 5.381110190555095,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00027730076464848225,
|
|
"loss": 4.8255,
|
|
"mean_token_accuracy": 0.20646308809518815,
|
|
"num_tokens": 11840885.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.350199556350708,
|
|
"epoch": 5.385252692626346,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027698122292806174,
|
|
"loss": 4.9236,
|
|
"mean_token_accuracy": 0.20531702041625977,
|
|
"num_tokens": 11850913.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.385252692626346,
|
|
"eval_entropy": 5.225421832314432,
|
|
"eval_loss": 5.818324565887451,
|
|
"eval_mean_token_accuracy": 0.1620604004591767,
|
|
"eval_num_tokens": 11850913.0,
|
|
"eval_runtime": 2.8112,
|
|
"eval_samples_per_second": 1332.875,
|
|
"eval_steps_per_second": 166.832,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.391539907455444,
|
|
"epoch": 5.389395194697597,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000276661677211298,
|
|
"loss": 4.8784,
|
|
"mean_token_accuracy": 0.20058761537075043,
|
|
"num_tokens": 11860379.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.369897937774658,
|
|
"epoch": 5.393537696768848,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002763421281427497,
|
|
"loss": 4.8261,
|
|
"mean_token_accuracy": 0.20628753304481506,
|
|
"num_tokens": 11869765.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.274693155288697,
|
|
"epoch": 5.397680198840099,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002760225763669823,
|
|
"loss": 4.7395,
|
|
"mean_token_accuracy": 0.21269467175006868,
|
|
"num_tokens": 11879490.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.289733123779297,
|
|
"epoch": 5.40182270091135,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027570302252856666,
|
|
"loss": 4.748,
|
|
"mean_token_accuracy": 0.2016552582383156,
|
|
"num_tokens": 11888187.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.235450315475464,
|
|
"epoch": 5.4059652029826015,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00027538346727207767,
|
|
"loss": 4.7556,
|
|
"mean_token_accuracy": 0.20766783207654954,
|
|
"num_tokens": 11897105.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.294193983078003,
|
|
"epoch": 5.410107705053853,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002750639112420934,
|
|
"loss": 4.7158,
|
|
"mean_token_accuracy": 0.21099428385496138,
|
|
"num_tokens": 11905556.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.240210676193238,
|
|
"epoch": 5.414250207125104,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00027474435508319304,
|
|
"loss": 4.6659,
|
|
"mean_token_accuracy": 0.21623111218214036,
|
|
"num_tokens": 11913442.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.37037296295166,
|
|
"epoch": 5.418392709196355,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027442479943995645,
|
|
"loss": 4.9185,
|
|
"mean_token_accuracy": 0.19152672290802003,
|
|
"num_tokens": 11923247.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.245935773849487,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002741052449569624,
|
|
"loss": 4.7379,
|
|
"mean_token_accuracy": 0.20472141951322556,
|
|
"num_tokens": 11932696.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.403725433349609,
|
|
"epoch": 5.426677713338857,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002737856922787869,
|
|
"loss": 4.8655,
|
|
"mean_token_accuracy": 0.19575051367282867,
|
|
"num_tokens": 11942333.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.349066972732544,
|
|
"epoch": 5.430820215410108,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027346614205000277,
|
|
"loss": 4.8677,
|
|
"mean_token_accuracy": 0.2052873194217682,
|
|
"num_tokens": 11951136.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.267371606826782,
|
|
"epoch": 5.434962717481358,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00027314659491517777,
|
|
"loss": 4.798,
|
|
"mean_token_accuracy": 0.2051367312669754,
|
|
"num_tokens": 11960317.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.292479944229126,
|
|
"epoch": 5.4391052195526095,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002728270515188733,
|
|
"loss": 4.8072,
|
|
"mean_token_accuracy": 0.20061399042606354,
|
|
"num_tokens": 11969712.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.258834457397461,
|
|
"epoch": 5.443247721623861,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002725075125056435,
|
|
"loss": 4.6971,
|
|
"mean_token_accuracy": 0.21074152290821074,
|
|
"num_tokens": 11979143.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.233439159393311,
|
|
"epoch": 5.447390223695112,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002721879785200333,
|
|
"loss": 4.7584,
|
|
"mean_token_accuracy": 0.2076289877295494,
|
|
"num_tokens": 11988659.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.285871458053589,
|
|
"epoch": 5.451532725766363,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002718684502065777,
|
|
"loss": 4.7282,
|
|
"mean_token_accuracy": 0.20997810512781143,
|
|
"num_tokens": 11997474.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.273050308227539,
|
|
"epoch": 5.455675227837614,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027154892820980024,
|
|
"loss": 4.8728,
|
|
"mean_token_accuracy": 0.1999638557434082,
|
|
"num_tokens": 12007708.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.330840015411377,
|
|
"epoch": 5.459817729908865,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00027122941317421175,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.19912724196910858,
|
|
"num_tokens": 12017843.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.443169641494751,
|
|
"epoch": 5.463960231980116,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000270909905744309,
|
|
"loss": 4.9356,
|
|
"mean_token_accuracy": 0.19400988668203353,
|
|
"num_tokens": 12027094.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.246637678146362,
|
|
"epoch": 5.468102734051367,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00027059040656457316,
|
|
"loss": 4.7036,
|
|
"mean_token_accuracy": 0.21273018270730973,
|
|
"num_tokens": 12035228.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.391880702972412,
|
|
"epoch": 5.472245236122618,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002702709162794692,
|
|
"loss": 4.9118,
|
|
"mean_token_accuracy": 0.19574049413204192,
|
|
"num_tokens": 12044357.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.400943565368652,
|
|
"epoch": 5.4763877381938695,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026995143553344394,
|
|
"loss": 4.7063,
|
|
"mean_token_accuracy": 0.21196494698524476,
|
|
"num_tokens": 12052888.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.211304092407227,
|
|
"epoch": 5.48053024026512,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026963196497092483,
|
|
"loss": 4.7269,
|
|
"mean_token_accuracy": 0.20890222638845443,
|
|
"num_tokens": 12063121.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.306645202636719,
|
|
"epoch": 5.484672742336371,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026931250523631905,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.20921180099248887,
|
|
"num_tokens": 12072642.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.239311790466308,
|
|
"epoch": 5.488815244407622,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026899305697401175,
|
|
"loss": 4.728,
|
|
"mean_token_accuracy": 0.20660368949174882,
|
|
"num_tokens": 12081640.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.23077073097229,
|
|
"epoch": 5.492957746478873,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002686736208283649,
|
|
"loss": 4.7355,
|
|
"mean_token_accuracy": 0.21483805030584335,
|
|
"num_tokens": 12090176.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.289255857467651,
|
|
"epoch": 5.497100248550124,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026835419744371627,
|
|
"loss": 4.7666,
|
|
"mean_token_accuracy": 0.2045901134610176,
|
|
"num_tokens": 12098865.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.329091501235962,
|
|
"epoch": 5.501242750621375,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002680347874643775,
|
|
"loss": 4.6954,
|
|
"mean_token_accuracy": 0.21893252283334733,
|
|
"num_tokens": 12106699.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.416130399703979,
|
|
"epoch": 5.505385252692626,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002677153915346336,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.19688885062932968,
|
|
"num_tokens": 12116001.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.327429676055909,
|
|
"epoch": 5.5095277547638775,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002673960102987412,
|
|
"loss": 4.8504,
|
|
"mean_token_accuracy": 0.20442483574151993,
|
|
"num_tokens": 12124967.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.3236244201660154,
|
|
"epoch": 5.513670256835129,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026707664440092676,
|
|
"loss": 4.7487,
|
|
"mean_token_accuracy": 0.21059839874505998,
|
|
"num_tokens": 12133916.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.304289293289185,
|
|
"epoch": 5.51781275890638,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026675729448538655,
|
|
"loss": 4.7699,
|
|
"mean_token_accuracy": 0.21062902808189393,
|
|
"num_tokens": 12141941.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.29205813407898,
|
|
"epoch": 5.521955260977631,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026643796119628404,
|
|
"loss": 4.7697,
|
|
"mean_token_accuracy": 0.21584970206022264,
|
|
"num_tokens": 12151441.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.256274557113647,
|
|
"epoch": 5.526097763048882,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002661186451777495,
|
|
"loss": 4.6735,
|
|
"mean_token_accuracy": 0.21815446019172668,
|
|
"num_tokens": 12159954.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.280739688873291,
|
|
"epoch": 5.530240265120133,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002657993470738781,
|
|
"loss": 4.8018,
|
|
"mean_token_accuracy": 0.20665960907936096,
|
|
"num_tokens": 12168579.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.342403221130371,
|
|
"epoch": 5.534382767191383,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00026548006752872915,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.2086340844631195,
|
|
"num_tokens": 12177168.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.363294792175293,
|
|
"epoch": 5.538525269262634,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002651608071863244,
|
|
"loss": 4.8799,
|
|
"mean_token_accuracy": 0.19662356972694398,
|
|
"num_tokens": 12185894.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.326411342620849,
|
|
"epoch": 5.542667771333885,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026484156669064665,
|
|
"loss": 4.779,
|
|
"mean_token_accuracy": 0.206169493496418,
|
|
"num_tokens": 12195298.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.3812727451324465,
|
|
"epoch": 5.5468102734051365,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000264522346685639,
|
|
"loss": 4.8621,
|
|
"mean_token_accuracy": 0.20506574511528014,
|
|
"num_tokens": 12204602.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.3329614162445065,
|
|
"epoch": 5.550952775476388,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002642031478152031,
|
|
"loss": 4.8499,
|
|
"mean_token_accuracy": 0.1928146868944168,
|
|
"num_tokens": 12213667.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.318030738830567,
|
|
"epoch": 5.555095277547639,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026388397072319786,
|
|
"loss": 4.7958,
|
|
"mean_token_accuracy": 0.20818064063787461,
|
|
"num_tokens": 12222298.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.309712028503418,
|
|
"epoch": 5.55923777961889,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002635648160534385,
|
|
"loss": 4.8041,
|
|
"mean_token_accuracy": 0.20435244888067244,
|
|
"num_tokens": 12232284.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.367008352279663,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026324568444969473,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.20195558965206145,
|
|
"num_tokens": 12240693.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.285096645355225,
|
|
"epoch": 5.567522783761392,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00026292657655569,
|
|
"loss": 4.8724,
|
|
"mean_token_accuracy": 0.20037627667188646,
|
|
"num_tokens": 12249841.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.334894275665283,
|
|
"epoch": 5.571665285832643,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026260749301509954,
|
|
"loss": 4.7939,
|
|
"mean_token_accuracy": 0.2031836286187172,
|
|
"num_tokens": 12257932.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.303253936767578,
|
|
"epoch": 5.575807787903894,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026228843447155005,
|
|
"loss": 4.7271,
|
|
"mean_token_accuracy": 0.21309324204921723,
|
|
"num_tokens": 12266251.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.303086614608764,
|
|
"epoch": 5.5799502899751445,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026196940156861736,
|
|
"loss": 4.8423,
|
|
"mean_token_accuracy": 0.1988528177142143,
|
|
"num_tokens": 12275659.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.305685663223267,
|
|
"epoch": 5.584092792046396,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026165039494982555,
|
|
"loss": 4.7695,
|
|
"mean_token_accuracy": 0.2088609367609024,
|
|
"num_tokens": 12284527.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.3052887439727785,
|
|
"epoch": 5.588235294117647,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026133141525864605,
|
|
"loss": 4.819,
|
|
"mean_token_accuracy": 0.20199260860681534,
|
|
"num_tokens": 12293856.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.347426080703736,
|
|
"epoch": 5.592377796188898,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026101246313849557,
|
|
"loss": 4.8845,
|
|
"mean_token_accuracy": 0.20065800994634628,
|
|
"num_tokens": 12303235.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.3439243793487545,
|
|
"epoch": 5.596520298260149,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002606935392327356,
|
|
"loss": 4.8305,
|
|
"mean_token_accuracy": 0.19738183915615082,
|
|
"num_tokens": 12312817.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.313336706161499,
|
|
"epoch": 5.6006628003314,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026037464418467035,
|
|
"loss": 4.7698,
|
|
"mean_token_accuracy": 0.21235544234514236,
|
|
"num_tokens": 12321823.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.351697111129761,
|
|
"epoch": 5.604805302402651,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002600557786375459,
|
|
"loss": 4.8433,
|
|
"mean_token_accuracy": 0.2005123034119606,
|
|
"num_tokens": 12330782.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.362290954589843,
|
|
"epoch": 5.608947804473902,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025973694323454916,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.20430914908647538,
|
|
"num_tokens": 12339931.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.298960399627686,
|
|
"epoch": 5.613090306545153,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025941813861880575,
|
|
"loss": 4.7913,
|
|
"mean_token_accuracy": 0.20193053483963014,
|
|
"num_tokens": 12350148.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.369911861419678,
|
|
"epoch": 5.6172328086164045,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025909936543337957,
|
|
"loss": 4.9085,
|
|
"mean_token_accuracy": 0.2010484129190445,
|
|
"num_tokens": 12359378.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.297723007202149,
|
|
"epoch": 5.621375310687656,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002587806243212709,
|
|
"loss": 4.7397,
|
|
"mean_token_accuracy": 0.20590801537036896,
|
|
"num_tokens": 12368384.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.304005527496338,
|
|
"epoch": 5.625517812758907,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002584619159254154,
|
|
"loss": 4.8067,
|
|
"mean_token_accuracy": 0.21065375357866287,
|
|
"num_tokens": 12377830.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.343201446533203,
|
|
"epoch": 5.629660314830158,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00025814324088868276,
|
|
"loss": 4.781,
|
|
"mean_token_accuracy": 0.20493562817573546,
|
|
"num_tokens": 12386463.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.276494121551513,
|
|
"epoch": 5.633802816901408,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025782459985387526,
|
|
"loss": 4.7116,
|
|
"mean_token_accuracy": 0.20999046117067338,
|
|
"num_tokens": 12394608.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.319973087310791,
|
|
"epoch": 5.637945318972659,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002575059934637268,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.19504383504390715,
|
|
"num_tokens": 12403016.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.298243236541748,
|
|
"epoch": 5.64208782104391,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002571874223609012,
|
|
"loss": 4.7963,
|
|
"mean_token_accuracy": 0.20408849269151688,
|
|
"num_tokens": 12411752.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.2814184665679935,
|
|
"epoch": 5.646230323115161,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025686888718799136,
|
|
"loss": 4.7942,
|
|
"mean_token_accuracy": 0.20867660343647004,
|
|
"num_tokens": 12420803.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.2650855541229244,
|
|
"epoch": 5.6503728251864125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002565503885875173,
|
|
"loss": 4.7858,
|
|
"mean_token_accuracy": 0.2077450633049011,
|
|
"num_tokens": 12429742.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.319002151489258,
|
|
"epoch": 5.654515327257664,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00025623192720192564,
|
|
"loss": 4.7954,
|
|
"mean_token_accuracy": 0.20256923586130143,
|
|
"num_tokens": 12438642.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.297832727432251,
|
|
"epoch": 5.658657829328915,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002559135036735877,
|
|
"loss": 4.7975,
|
|
"mean_token_accuracy": 0.2033243417739868,
|
|
"num_tokens": 12447250.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.241038227081299,
|
|
"epoch": 5.662800331400166,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025559511864479864,
|
|
"loss": 4.6786,
|
|
"mean_token_accuracy": 0.21368588507175446,
|
|
"num_tokens": 12456713.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.330965566635132,
|
|
"epoch": 5.666942833471417,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002552767727577758,
|
|
"loss": 4.7899,
|
|
"mean_token_accuracy": 0.19689203798770905,
|
|
"num_tokens": 12465436.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.265569734573364,
|
|
"epoch": 5.671085335542668,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025495846665465774,
|
|
"loss": 4.7901,
|
|
"mean_token_accuracy": 0.2031988173723221,
|
|
"num_tokens": 12474930.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.241378688812256,
|
|
"epoch": 5.675227837613919,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002546402009775024,
|
|
"loss": 4.7316,
|
|
"mean_token_accuracy": 0.2084824562072754,
|
|
"num_tokens": 12483599.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.291776943206787,
|
|
"epoch": 5.679370339685169,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00025432197636828663,
|
|
"loss": 4.7961,
|
|
"mean_token_accuracy": 0.20324424058198928,
|
|
"num_tokens": 12492494.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.30623869895935,
|
|
"epoch": 5.683512841756421,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002540037934689042,
|
|
"loss": 4.8367,
|
|
"mean_token_accuracy": 0.2022312805056572,
|
|
"num_tokens": 12501145.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.375609111785889,
|
|
"epoch": 5.687655343827672,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025368565292116475,
|
|
"loss": 4.8042,
|
|
"mean_token_accuracy": 0.20826269388198854,
|
|
"num_tokens": 12511312.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.435929155349731,
|
|
"epoch": 5.691797845898923,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002533675553667927,
|
|
"loss": 4.8289,
|
|
"mean_token_accuracy": 0.20321155488491058,
|
|
"num_tokens": 12520348.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.313596963882446,
|
|
"epoch": 5.695940347970174,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00025304950144742523,
|
|
"loss": 4.7869,
|
|
"mean_token_accuracy": 0.20051283091306688,
|
|
"num_tokens": 12528868.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.195265054702759,
|
|
"epoch": 5.700082850041425,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002527314918046122,
|
|
"loss": 4.7655,
|
|
"mean_token_accuracy": 0.2143235981464386,
|
|
"num_tokens": 12538358.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.259504985809326,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002524135270798136,
|
|
"loss": 4.7773,
|
|
"mean_token_accuracy": 0.20917656272649765,
|
|
"num_tokens": 12546887.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.354610013961792,
|
|
"epoch": 5.708367854183927,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025209560791439914,
|
|
"loss": 4.8203,
|
|
"mean_token_accuracy": 0.20545507669448854,
|
|
"num_tokens": 12555396.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.315736103057861,
|
|
"epoch": 5.712510356255178,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002517777349496465,
|
|
"loss": 4.7721,
|
|
"mean_token_accuracy": 0.2027284860610962,
|
|
"num_tokens": 12565171.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.290043449401855,
|
|
"epoch": 5.716652858326429,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002514599088267401,
|
|
"loss": 4.7865,
|
|
"mean_token_accuracy": 0.2137683093547821,
|
|
"num_tokens": 12573847.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.354130411148072,
|
|
"epoch": 5.72079536039768,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025114213018677004,
|
|
"loss": 4.8142,
|
|
"mean_token_accuracy": 0.20738655030727388,
|
|
"num_tokens": 12581979.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.313599300384522,
|
|
"epoch": 5.7249378624689315,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002508243996707305,
|
|
"loss": 4.8066,
|
|
"mean_token_accuracy": 0.20306238085031508,
|
|
"num_tokens": 12591175.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.36643123626709,
|
|
"epoch": 5.729080364540183,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025050671791951887,
|
|
"loss": 4.7899,
|
|
"mean_token_accuracy": 0.21273463666439058,
|
|
"num_tokens": 12600117.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.329094409942627,
|
|
"epoch": 5.733222866611433,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002501890855739337,
|
|
"loss": 4.8472,
|
|
"mean_token_accuracy": 0.20101388841867446,
|
|
"num_tokens": 12609939.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.387216424942016,
|
|
"epoch": 5.737365368682684,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00024987150327467414,
|
|
"loss": 4.9161,
|
|
"mean_token_accuracy": 0.19303029924631118,
|
|
"num_tokens": 12619653.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.314742517471314,
|
|
"epoch": 5.741507870753935,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024955397166233855,
|
|
"loss": 4.7705,
|
|
"mean_token_accuracy": 0.20892754346132278,
|
|
"num_tokens": 12628601.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.384041404724121,
|
|
"epoch": 5.745650372825186,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024923649137742276,
|
|
"loss": 4.8374,
|
|
"mean_token_accuracy": 0.20546534210443496,
|
|
"num_tokens": 12638154.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.333579158782959,
|
|
"epoch": 5.749792874896437,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002489190630603194,
|
|
"loss": 4.8674,
|
|
"mean_token_accuracy": 0.20757418870925903,
|
|
"num_tokens": 12647593.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.311374616622925,
|
|
"epoch": 5.753935376967688,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002486016873513159,
|
|
"loss": 4.8884,
|
|
"mean_token_accuracy": 0.19325120896100997,
|
|
"num_tokens": 12657067.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.337721443176269,
|
|
"epoch": 5.7580778790389395,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002482843648905937,
|
|
"loss": 4.7609,
|
|
"mean_token_accuracy": 0.20660501718521118,
|
|
"num_tokens": 12666288.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.360270977020264,
|
|
"epoch": 5.762220381110191,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002479670963182271,
|
|
"loss": 4.8663,
|
|
"mean_token_accuracy": 0.19935106337070466,
|
|
"num_tokens": 12676150.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.266315031051636,
|
|
"epoch": 5.766362883181442,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024764988227418114,
|
|
"loss": 4.8487,
|
|
"mean_token_accuracy": 0.20336804091930388,
|
|
"num_tokens": 12685405.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.191412830352784,
|
|
"epoch": 5.770505385252693,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002473327233983116,
|
|
"loss": 4.7133,
|
|
"mean_token_accuracy": 0.21085067838430405,
|
|
"num_tokens": 12694214.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.29790153503418,
|
|
"epoch": 5.774647887323944,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002470156203303622,
|
|
"loss": 4.7734,
|
|
"mean_token_accuracy": 0.21520546078681946,
|
|
"num_tokens": 12703054.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.330405950546265,
|
|
"epoch": 5.778790389395195,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002466985737099649,
|
|
"loss": 4.8798,
|
|
"mean_token_accuracy": 0.19415688812732695,
|
|
"num_tokens": 12713263.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.324331378936767,
|
|
"epoch": 5.782932891466446,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002463815841766371,
|
|
"loss": 4.8125,
|
|
"mean_token_accuracy": 0.19346748292446136,
|
|
"num_tokens": 12722274.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.333473205566406,
|
|
"epoch": 5.787075393537696,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002460646523697813,
|
|
"loss": 4.7906,
|
|
"mean_token_accuracy": 0.20946188122034073,
|
|
"num_tokens": 12732608.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.3331263065338135,
|
|
"epoch": 5.7912178956089475,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024574777892868386,
|
|
"loss": 4.8185,
|
|
"mean_token_accuracy": 0.2014390915632248,
|
|
"num_tokens": 12742481.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.291971254348755,
|
|
"epoch": 5.795360397680199,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00024543096449251286,
|
|
"loss": 4.724,
|
|
"mean_token_accuracy": 0.21113979071378708,
|
|
"num_tokens": 12751974.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.3013286113739015,
|
|
"epoch": 5.79950289975145,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002451142097003177,
|
|
"loss": 4.7969,
|
|
"mean_token_accuracy": 0.2025451183319092,
|
|
"num_tokens": 12761358.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.79950289975145,
|
|
"eval_entropy": 5.21366401013535,
|
|
"eval_loss": 5.794280052185059,
|
|
"eval_mean_token_accuracy": 0.16286438558973484,
|
|
"eval_num_tokens": 12761358.0,
|
|
"eval_runtime": 2.8496,
|
|
"eval_samples_per_second": 1314.901,
|
|
"eval_steps_per_second": 164.582,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.30829553604126,
|
|
"epoch": 5.803645401822701,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002447975151910276,
|
|
"loss": 4.7158,
|
|
"mean_token_accuracy": 0.2141193613409996,
|
|
"num_tokens": 12769798.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.370926332473755,
|
|
"epoch": 5.807787903893952,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024448088160344974,
|
|
"loss": 4.8394,
|
|
"mean_token_accuracy": 0.20802996605634688,
|
|
"num_tokens": 12778560.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.298399591445923,
|
|
"epoch": 5.811930405965203,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002441643095762688,
|
|
"loss": 4.83,
|
|
"mean_token_accuracy": 0.20863751620054244,
|
|
"num_tokens": 12788394.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.345996856689453,
|
|
"epoch": 5.816072908036454,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000243847799748045,
|
|
"loss": 4.8638,
|
|
"mean_token_accuracy": 0.20390648096799852,
|
|
"num_tokens": 12797468.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.339472103118896,
|
|
"epoch": 5.820215410107705,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024353135275721356,
|
|
"loss": 4.7132,
|
|
"mean_token_accuracy": 0.21080717742443084,
|
|
"num_tokens": 12806233.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.336239004135132,
|
|
"epoch": 5.824357912178956,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024321496924208235,
|
|
"loss": 4.8507,
|
|
"mean_token_accuracy": 0.20134634226560594,
|
|
"num_tokens": 12814933.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.241674470901489,
|
|
"epoch": 5.8285004142502075,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024289864984083166,
|
|
"loss": 4.7614,
|
|
"mean_token_accuracy": 0.20401494354009628,
|
|
"num_tokens": 12823666.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.270673418045044,
|
|
"epoch": 5.832642916321458,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002425823951915122,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.20177935361862182,
|
|
"num_tokens": 12832012.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.269376182556153,
|
|
"epoch": 5.836785418392709,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002422662059320442,
|
|
"loss": 4.7705,
|
|
"mean_token_accuracy": 0.20922453701496124,
|
|
"num_tokens": 12841023.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.253552103042603,
|
|
"epoch": 5.84092792046396,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024195008270021595,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.20516045093536378,
|
|
"num_tokens": 12850320.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.3117328643798825,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024163402613368252,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.20046598166227342,
|
|
"num_tokens": 12859096.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.3454653263092045,
|
|
"epoch": 5.849212924606462,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024131803686996457,
|
|
"loss": 4.8127,
|
|
"mean_token_accuracy": 0.20158441811800004,
|
|
"num_tokens": 12869086.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.299232292175293,
|
|
"epoch": 5.853355426677713,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024100211554644696,
|
|
"loss": 4.7871,
|
|
"mean_token_accuracy": 0.20782023221254348,
|
|
"num_tokens": 12877850.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.345463800430298,
|
|
"epoch": 5.857497928748964,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024068626280037753,
|
|
"loss": 4.8091,
|
|
"mean_token_accuracy": 0.20316458195447923,
|
|
"num_tokens": 12886256.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.264532566070557,
|
|
"epoch": 5.8616404308202155,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024037047926886578,
|
|
"loss": 4.7874,
|
|
"mean_token_accuracy": 0.20838468372821808,
|
|
"num_tokens": 12895534.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.292352914810181,
|
|
"epoch": 5.865782932891467,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024005476558888172,
|
|
"loss": 4.8024,
|
|
"mean_token_accuracy": 0.20062680095434188,
|
|
"num_tokens": 12905017.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.3796710014343265,
|
|
"epoch": 5.869925434962718,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002397391223972541,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.20445189923048018,
|
|
"num_tokens": 12914416.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.32456636428833,
|
|
"epoch": 5.874067937033969,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023942355033066993,
|
|
"loss": 4.7354,
|
|
"mean_token_accuracy": 0.2126239776611328,
|
|
"num_tokens": 12922783.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.355032539367675,
|
|
"epoch": 5.87821043910522,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023910805002567244,
|
|
"loss": 4.8121,
|
|
"mean_token_accuracy": 0.21395953893661498,
|
|
"num_tokens": 12931154.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.359413480758667,
|
|
"epoch": 5.882352941176471,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002387926221186603,
|
|
"loss": 4.8179,
|
|
"mean_token_accuracy": 0.20713991671800613,
|
|
"num_tokens": 12940864.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.210391521453857,
|
|
"epoch": 5.886495443247721,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023847726724588608,
|
|
"loss": 4.6872,
|
|
"mean_token_accuracy": 0.21473013162612914,
|
|
"num_tokens": 12949265.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.244672584533691,
|
|
"epoch": 5.890637945318972,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002381619860434549,
|
|
"loss": 4.7825,
|
|
"mean_token_accuracy": 0.20627081096172334,
|
|
"num_tokens": 12959061.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.278915691375732,
|
|
"epoch": 5.894780447390223,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023784677914732356,
|
|
"loss": 4.8208,
|
|
"mean_token_accuracy": 0.19842978566884995,
|
|
"num_tokens": 12967704.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.38638653755188,
|
|
"epoch": 5.8989229494614746,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023753164719329868,
|
|
"loss": 4.8398,
|
|
"mean_token_accuracy": 0.2045602470636368,
|
|
"num_tokens": 12976616.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.373883485794067,
|
|
"epoch": 5.903065451532726,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023721659081703595,
|
|
"loss": 4.8215,
|
|
"mean_token_accuracy": 0.20100327730178832,
|
|
"num_tokens": 12986124.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.368735074996948,
|
|
"epoch": 5.907207953603977,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023690161065403838,
|
|
"loss": 4.8482,
|
|
"mean_token_accuracy": 0.2111173078417778,
|
|
"num_tokens": 12995904.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.303588485717773,
|
|
"epoch": 5.911350455675228,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002365867073396555,
|
|
"loss": 4.783,
|
|
"mean_token_accuracy": 0.20996139496564864,
|
|
"num_tokens": 13005346.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.2844610691070555,
|
|
"epoch": 5.915492957746479,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023627188150908174,
|
|
"loss": 4.817,
|
|
"mean_token_accuracy": 0.19675445258617402,
|
|
"num_tokens": 13014856.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.3592236042022705,
|
|
"epoch": 5.91963545981773,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00023595713379735494,
|
|
"loss": 4.8606,
|
|
"mean_token_accuracy": 0.2017446681857109,
|
|
"num_tokens": 13024010.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.367470216751099,
|
|
"epoch": 5.923777961888981,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023564246483935602,
|
|
"loss": 4.8901,
|
|
"mean_token_accuracy": 0.19699376225471496,
|
|
"num_tokens": 13032783.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.32507004737854,
|
|
"epoch": 5.927920463960232,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002353278752698063,
|
|
"loss": 4.7926,
|
|
"mean_token_accuracy": 0.201738204061985,
|
|
"num_tokens": 13041768.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.4142005920410154,
|
|
"epoch": 5.9320629660314825,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002350133657232676,
|
|
"loss": 4.9004,
|
|
"mean_token_accuracy": 0.19654888957738875,
|
|
"num_tokens": 13050154.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.261559963226318,
|
|
"epoch": 5.9362054681027345,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002346989368341399,
|
|
"loss": 4.824,
|
|
"mean_token_accuracy": 0.1959752693772316,
|
|
"num_tokens": 13059199.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.3489843845367435,
|
|
"epoch": 5.940347970173985,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023438458923666057,
|
|
"loss": 4.841,
|
|
"mean_token_accuracy": 0.20398749262094498,
|
|
"num_tokens": 13068363.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.279772090911865,
|
|
"epoch": 5.944490472245236,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002340703235649032,
|
|
"loss": 4.8093,
|
|
"mean_token_accuracy": 0.2032819762825966,
|
|
"num_tokens": 13077642.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.2986469745635985,
|
|
"epoch": 5.948632974316487,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023375614045277593,
|
|
"loss": 4.8531,
|
|
"mean_token_accuracy": 0.19490729719400407,
|
|
"num_tokens": 13086758.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.355213308334351,
|
|
"epoch": 5.952775476387738,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023344204053402047,
|
|
"loss": 4.7824,
|
|
"mean_token_accuracy": 0.20358290076255797,
|
|
"num_tokens": 13096407.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.329388618469238,
|
|
"epoch": 5.956917978458989,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023312802444221066,
|
|
"loss": 4.867,
|
|
"mean_token_accuracy": 0.19691111147403717,
|
|
"num_tokens": 13105657.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.256456708908081,
|
|
"epoch": 5.96106048053024,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002328140928107511,
|
|
"loss": 4.7992,
|
|
"mean_token_accuracy": 0.20688048005104065,
|
|
"num_tokens": 13115225.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.35938138961792,
|
|
"epoch": 5.965202982601491,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002325002462728766,
|
|
"loss": 4.8988,
|
|
"mean_token_accuracy": 0.2027306318283081,
|
|
"num_tokens": 13124711.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.302161359786988,
|
|
"epoch": 5.9693454846727425,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023218648546164945,
|
|
"loss": 4.7525,
|
|
"mean_token_accuracy": 0.21453772038221358,
|
|
"num_tokens": 13133587.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.3076964855194095,
|
|
"epoch": 5.973487986743994,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002318728110099599,
|
|
"loss": 4.7336,
|
|
"mean_token_accuracy": 0.20199378877878188,
|
|
"num_tokens": 13142088.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.326257562637329,
|
|
"epoch": 5.977630488815245,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00023155922355052339,
|
|
"loss": 4.8319,
|
|
"mean_token_accuracy": 0.21207345575094222,
|
|
"num_tokens": 13151550.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.3169924259185795,
|
|
"epoch": 5.981772990886496,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023124572371588,
|
|
"loss": 4.8149,
|
|
"mean_token_accuracy": 0.20328260958194733,
|
|
"num_tokens": 13160703.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.3462910652160645,
|
|
"epoch": 5.985915492957746,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00023093231213839344,
|
|
"loss": 4.8378,
|
|
"mean_token_accuracy": 0.19642047584056854,
|
|
"num_tokens": 13169452.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.30769910812378,
|
|
"epoch": 5.990057995028997,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002306189894502488,
|
|
"loss": 4.7609,
|
|
"mean_token_accuracy": 0.21082856506109238,
|
|
"num_tokens": 13179094.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.373433065414429,
|
|
"epoch": 5.994200497100248,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023030575628345246,
|
|
"loss": 4.8515,
|
|
"mean_token_accuracy": 0.20574633181095123,
|
|
"num_tokens": 13187966.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.30269193649292,
|
|
"epoch": 5.998342999171499,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022999261326982985,
|
|
"loss": 4.7969,
|
|
"mean_token_accuracy": 0.20804344564676286,
|
|
"num_tokens": 13196741.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.349225330352783,
|
|
"epoch": 6.0024855012427505,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022967956104102445,
|
|
"loss": 4.7415,
|
|
"mean_token_accuracy": 0.21540724337100983,
|
|
"num_tokens": 13205067.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.364014148712158,
|
|
"epoch": 6.006628003314002,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022936660022849714,
|
|
"loss": 4.6181,
|
|
"mean_token_accuracy": 0.21370948404073714,
|
|
"num_tokens": 13213274.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.2984740257263185,
|
|
"epoch": 6.010770505385253,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022905373146352382,
|
|
"loss": 4.6461,
|
|
"mean_token_accuracy": 0.21742259711027145,
|
|
"num_tokens": 13222137.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.334527587890625,
|
|
"epoch": 6.014913007456504,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002287409553771952,
|
|
"loss": 4.6639,
|
|
"mean_token_accuracy": 0.21406797766685487,
|
|
"num_tokens": 13232744.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.303852272033692,
|
|
"epoch": 6.019055509527755,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022842827260041455,
|
|
"loss": 4.6132,
|
|
"mean_token_accuracy": 0.22277068793773652,
|
|
"num_tokens": 13241371.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.33689169883728,
|
|
"epoch": 6.023198011599006,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022811568376389745,
|
|
"loss": 4.6383,
|
|
"mean_token_accuracy": 0.21806675791740418,
|
|
"num_tokens": 13250404.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.336875104904175,
|
|
"epoch": 6.027340513670257,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022780318949816963,
|
|
"loss": 4.6577,
|
|
"mean_token_accuracy": 0.22686232924461364,
|
|
"num_tokens": 13260678.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.301117181777954,
|
|
"epoch": 6.031483015741508,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022749079043356623,
|
|
"loss": 4.6217,
|
|
"mean_token_accuracy": 0.21875354051589965,
|
|
"num_tokens": 13269762.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.293356609344483,
|
|
"epoch": 6.0356255178127585,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022717848720023022,
|
|
"loss": 4.6457,
|
|
"mean_token_accuracy": 0.2241827592253685,
|
|
"num_tokens": 13279226.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.369172859191894,
|
|
"epoch": 6.03976801988401,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022686628042811142,
|
|
"loss": 4.6871,
|
|
"mean_token_accuracy": 0.2100516989827156,
|
|
"num_tokens": 13289152.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.231834411621094,
|
|
"epoch": 6.043910521955261,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000226554170746965,
|
|
"loss": 4.5725,
|
|
"mean_token_accuracy": 0.21603020280599594,
|
|
"num_tokens": 13298143.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.31619119644165,
|
|
"epoch": 6.048053024026512,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022624215878635023,
|
|
"loss": 4.5633,
|
|
"mean_token_accuracy": 0.22952798157930374,
|
|
"num_tokens": 13307064.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.282352352142334,
|
|
"epoch": 6.052195526097763,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022593024517562945,
|
|
"loss": 4.63,
|
|
"mean_token_accuracy": 0.2215061143040657,
|
|
"num_tokens": 13316352.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.282910108566284,
|
|
"epoch": 6.056338028169014,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022561843054396636,
|
|
"loss": 4.6772,
|
|
"mean_token_accuracy": 0.2133912444114685,
|
|
"num_tokens": 13325868.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.358627986907959,
|
|
"epoch": 6.060480530240265,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022530671552032521,
|
|
"loss": 4.7278,
|
|
"mean_token_accuracy": 0.21394070386886596,
|
|
"num_tokens": 13335817.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.307431364059449,
|
|
"epoch": 6.064623032311516,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002249951007334693,
|
|
"loss": 4.5393,
|
|
"mean_token_accuracy": 0.2272082343697548,
|
|
"num_tokens": 13344186.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.26256160736084,
|
|
"epoch": 6.068765534382767,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022468358681195966,
|
|
"loss": 4.6469,
|
|
"mean_token_accuracy": 0.21850877553224562,
|
|
"num_tokens": 13353601.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.18783221244812,
|
|
"epoch": 6.072908036454018,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022437217438415393,
|
|
"loss": 4.5301,
|
|
"mean_token_accuracy": 0.231480011343956,
|
|
"num_tokens": 13362841.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.362886667251587,
|
|
"epoch": 6.0770505385252696,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00022406086407820503,
|
|
"loss": 4.6258,
|
|
"mean_token_accuracy": 0.23286102414131166,
|
|
"num_tokens": 13372112.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.321625995635986,
|
|
"epoch": 6.081193040596521,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022374965652205975,
|
|
"loss": 4.647,
|
|
"mean_token_accuracy": 0.21555265635252,
|
|
"num_tokens": 13380418.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.290751981735229,
|
|
"epoch": 6.085335542667772,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022343855234345796,
|
|
"loss": 4.6891,
|
|
"mean_token_accuracy": 0.20590780228376387,
|
|
"num_tokens": 13389213.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.305972671508789,
|
|
"epoch": 6.089478044739022,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022312755216993065,
|
|
"loss": 4.612,
|
|
"mean_token_accuracy": 0.21789441257715225,
|
|
"num_tokens": 13398558.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.309966611862182,
|
|
"epoch": 6.093620546810273,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022281665662879903,
|
|
"loss": 4.6554,
|
|
"mean_token_accuracy": 0.2209019511938095,
|
|
"num_tokens": 13407258.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.3165764808654785,
|
|
"epoch": 6.097763048881524,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022250586634717362,
|
|
"loss": 4.6523,
|
|
"mean_token_accuracy": 0.22093119323253632,
|
|
"num_tokens": 13416009.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.287519741058349,
|
|
"epoch": 6.101905550952775,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022219518195195216,
|
|
"loss": 4.5349,
|
|
"mean_token_accuracy": 0.22534932494163512,
|
|
"num_tokens": 13424297.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.297530698776245,
|
|
"epoch": 6.106048053024026,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022188460406981925,
|
|
"loss": 4.5822,
|
|
"mean_token_accuracy": 0.21721828132867813,
|
|
"num_tokens": 13433496.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.298411560058594,
|
|
"epoch": 6.1101905550952775,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022157413332724436,
|
|
"loss": 4.7258,
|
|
"mean_token_accuracy": 0.2098759263753891,
|
|
"num_tokens": 13442315.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.2143714427948,
|
|
"epoch": 6.114333057166529,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022126377035048072,
|
|
"loss": 4.5267,
|
|
"mean_token_accuracy": 0.23844918459653855,
|
|
"num_tokens": 13449982.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.223819875717163,
|
|
"epoch": 6.11847555923778,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002209535157655646,
|
|
"loss": 4.5747,
|
|
"mean_token_accuracy": 0.22771441340446472,
|
|
"num_tokens": 13459605.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.300343561172485,
|
|
"epoch": 6.122618061309031,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022064337019831322,
|
|
"loss": 4.623,
|
|
"mean_token_accuracy": 0.21384273767471312,
|
|
"num_tokens": 13468443.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.384988069534302,
|
|
"epoch": 6.126760563380282,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022033333427432422,
|
|
"loss": 4.6874,
|
|
"mean_token_accuracy": 0.21775298565626144,
|
|
"num_tokens": 13477315.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.233644676208496,
|
|
"epoch": 6.130903065451533,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002200234086189738,
|
|
"loss": 4.578,
|
|
"mean_token_accuracy": 0.22483548521995544,
|
|
"num_tokens": 13486367.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.330848312377929,
|
|
"epoch": 6.135045567522784,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021971359385741608,
|
|
"loss": 4.6982,
|
|
"mean_token_accuracy": 0.2137132927775383,
|
|
"num_tokens": 13495756.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.378737306594848,
|
|
"epoch": 6.139188069594034,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021940389061458115,
|
|
"loss": 4.6853,
|
|
"mean_token_accuracy": 0.2186382681131363,
|
|
"num_tokens": 13504970.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.310305786132813,
|
|
"epoch": 6.1433305716652855,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002190942995151743,
|
|
"loss": 4.5914,
|
|
"mean_token_accuracy": 0.22214196771383285,
|
|
"num_tokens": 13513605.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.29168930053711,
|
|
"epoch": 6.147473073736537,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002187848211836746,
|
|
"loss": 4.634,
|
|
"mean_token_accuracy": 0.21668605655431747,
|
|
"num_tokens": 13523552.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.2700779914855955,
|
|
"epoch": 6.151615575807788,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002184754562443337,
|
|
"loss": 4.6716,
|
|
"mean_token_accuracy": 0.21091263741254807,
|
|
"num_tokens": 13532940.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.328031778335571,
|
|
"epoch": 6.155758077879039,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021816620532117448,
|
|
"loss": 4.6394,
|
|
"mean_token_accuracy": 0.22433526813983917,
|
|
"num_tokens": 13541227.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.326112508773804,
|
|
"epoch": 6.15990057995029,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002178570690379899,
|
|
"loss": 4.7061,
|
|
"mean_token_accuracy": 0.21365715712308883,
|
|
"num_tokens": 13550845.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.337257814407349,
|
|
"epoch": 6.164043082021541,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021754804801834144,
|
|
"loss": 4.6519,
|
|
"mean_token_accuracy": 0.2187626764178276,
|
|
"num_tokens": 13559058.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.224030303955078,
|
|
"epoch": 6.168185584092792,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002172391428855585,
|
|
"loss": 4.5644,
|
|
"mean_token_accuracy": 0.22548640072345733,
|
|
"num_tokens": 13568224.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.278336381912231,
|
|
"epoch": 6.172328086164043,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021693035426273633,
|
|
"loss": 4.6273,
|
|
"mean_token_accuracy": 0.2183200642466545,
|
|
"num_tokens": 13577416.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.3287252426147464,
|
|
"epoch": 6.176470588235294,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021662168277273547,
|
|
"loss": 4.6977,
|
|
"mean_token_accuracy": 0.2080409899353981,
|
|
"num_tokens": 13586908.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.3359479904174805,
|
|
"epoch": 6.1806130903065455,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002163131290381799,
|
|
"loss": 4.6067,
|
|
"mean_token_accuracy": 0.2202505111694336,
|
|
"num_tokens": 13595476.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.27282338142395,
|
|
"epoch": 6.184755592377797,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021600469368145624,
|
|
"loss": 4.6425,
|
|
"mean_token_accuracy": 0.21951130479574205,
|
|
"num_tokens": 13604347.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.264153909683228,
|
|
"epoch": 6.188898094449047,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002156963773247124,
|
|
"loss": 4.6301,
|
|
"mean_token_accuracy": 0.22206076383590698,
|
|
"num_tokens": 13612622.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.339120483398437,
|
|
"epoch": 6.193040596520298,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002153881805898561,
|
|
"loss": 4.7141,
|
|
"mean_token_accuracy": 0.21635956466197967,
|
|
"num_tokens": 13621799.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.3252605438232425,
|
|
"epoch": 6.197183098591549,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002150801040985539,
|
|
"loss": 4.675,
|
|
"mean_token_accuracy": 0.2100287690758705,
|
|
"num_tokens": 13630542.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.21474256515503,
|
|
"epoch": 6.2013256006628,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021477214847222958,
|
|
"loss": 4.5498,
|
|
"mean_token_accuracy": 0.22742502093315126,
|
|
"num_tokens": 13640183.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.274860858917236,
|
|
"epoch": 6.205468102734051,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00021446431433206348,
|
|
"loss": 4.634,
|
|
"mean_token_accuracy": 0.22505085468292235,
|
|
"num_tokens": 13648552.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.268704462051391,
|
|
"epoch": 6.209610604805302,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002141566022989907,
|
|
"loss": 4.6617,
|
|
"mean_token_accuracy": 0.21855499148368834,
|
|
"num_tokens": 13656964.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.366129636764526,
|
|
"epoch": 6.2137531068765535,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002138490129936998,
|
|
"loss": 4.7368,
|
|
"mean_token_accuracy": 0.20757783502340316,
|
|
"num_tokens": 13666908.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.2137531068765535,
|
|
"eval_entropy": 5.196083771394514,
|
|
"eval_loss": 5.789187908172607,
|
|
"eval_mean_token_accuracy": 0.1640587144378406,
|
|
"eval_num_tokens": 13666908.0,
|
|
"eval_runtime": 2.9466,
|
|
"eval_samples_per_second": 1271.626,
|
|
"eval_steps_per_second": 159.165,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.321217346191406,
|
|
"epoch": 6.217895608947805,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002135415470366323,
|
|
"loss": 4.6825,
|
|
"mean_token_accuracy": 0.20877762138843536,
|
|
"num_tokens": 13675479.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.345216274261475,
|
|
"epoch": 6.222038111019056,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021323420504798047,
|
|
"loss": 4.6722,
|
|
"mean_token_accuracy": 0.2186599612236023,
|
|
"num_tokens": 13684880.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.261148452758789,
|
|
"epoch": 6.226180613090307,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021292698764768675,
|
|
"loss": 4.5656,
|
|
"mean_token_accuracy": 0.2279446855187416,
|
|
"num_tokens": 13693696.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.340752601623535,
|
|
"epoch": 6.230323115161558,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021261989545544218,
|
|
"loss": 4.6673,
|
|
"mean_token_accuracy": 0.21050670892000198,
|
|
"num_tokens": 13703752.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.356611728668213,
|
|
"epoch": 6.234465617232809,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021231292909068534,
|
|
"loss": 4.7286,
|
|
"mean_token_accuracy": 0.2143424019217491,
|
|
"num_tokens": 13713220.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.272091674804687,
|
|
"epoch": 6.238608119304059,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021200608917260082,
|
|
"loss": 4.6694,
|
|
"mean_token_accuracy": 0.21681212931871413,
|
|
"num_tokens": 13722688.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.286226558685303,
|
|
"epoch": 6.24275062137531,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021169937632011838,
|
|
"loss": 4.6374,
|
|
"mean_token_accuracy": 0.22034697681665422,
|
|
"num_tokens": 13732034.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.321897220611572,
|
|
"epoch": 6.2468931234465614,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021139279115191123,
|
|
"loss": 4.6483,
|
|
"mean_token_accuracy": 0.2145317405462265,
|
|
"num_tokens": 13739934.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.212859678268432,
|
|
"epoch": 6.2510356255178126,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021108633428639534,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.21781880110502244,
|
|
"num_tokens": 13749352.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.269030284881592,
|
|
"epoch": 6.255178127589064,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002107800063417275,
|
|
"loss": 4.635,
|
|
"mean_token_accuracy": 0.22059867680072784,
|
|
"num_tokens": 13758521.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.344493007659912,
|
|
"epoch": 6.259320629660315,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002104738079358048,
|
|
"loss": 4.68,
|
|
"mean_token_accuracy": 0.21227535754442214,
|
|
"num_tokens": 13767469.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.334861516952515,
|
|
"epoch": 6.263463131731566,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002101677396862628,
|
|
"loss": 4.6403,
|
|
"mean_token_accuracy": 0.2231946185231209,
|
|
"num_tokens": 13775923.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.3412223815917965,
|
|
"epoch": 6.267605633802817,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020986180221047463,
|
|
"loss": 4.7258,
|
|
"mean_token_accuracy": 0.21173176318407058,
|
|
"num_tokens": 13785220.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.247770023345947,
|
|
"epoch": 6.271748135874068,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020955599612554966,
|
|
"loss": 4.6442,
|
|
"mean_token_accuracy": 0.21562438905239106,
|
|
"num_tokens": 13793670.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.263835477828979,
|
|
"epoch": 6.275890637945319,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020925032204833206,
|
|
"loss": 4.6076,
|
|
"mean_token_accuracy": 0.21849464923143386,
|
|
"num_tokens": 13802668.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.294063663482666,
|
|
"epoch": 6.28003314001657,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020894478059539995,
|
|
"loss": 4.6404,
|
|
"mean_token_accuracy": 0.22222468852996827,
|
|
"num_tokens": 13811374.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.344520378112793,
|
|
"epoch": 6.284175642087821,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020863937238306375,
|
|
"loss": 4.6744,
|
|
"mean_token_accuracy": 0.21950925290584564,
|
|
"num_tokens": 13820812.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.3712992668151855,
|
|
"epoch": 6.2883181441590725,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020833409802736535,
|
|
"loss": 4.7788,
|
|
"mean_token_accuracy": 0.20978112667798995,
|
|
"num_tokens": 13830637.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.215086460113525,
|
|
"epoch": 6.292460646230323,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020802895814407636,
|
|
"loss": 4.5969,
|
|
"mean_token_accuracy": 0.22417790591716766,
|
|
"num_tokens": 13839954.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.268776798248291,
|
|
"epoch": 6.296603148301574,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002077239533486972,
|
|
"loss": 4.5965,
|
|
"mean_token_accuracy": 0.21614333987236023,
|
|
"num_tokens": 13848000.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.26701169013977,
|
|
"epoch": 6.300745650372825,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020741908425645606,
|
|
"loss": 4.6524,
|
|
"mean_token_accuracy": 0.2238873228430748,
|
|
"num_tokens": 13857353.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.242597007751465,
|
|
"epoch": 6.304888152444076,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020711435148230707,
|
|
"loss": 4.5468,
|
|
"mean_token_accuracy": 0.23241741806268693,
|
|
"num_tokens": 13866026.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.287490558624268,
|
|
"epoch": 6.309030654515327,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020680975564092968,
|
|
"loss": 4.6285,
|
|
"mean_token_accuracy": 0.21387007385492324,
|
|
"num_tokens": 13875789.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.216368389129639,
|
|
"epoch": 6.313173156586578,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020650529734672686,
|
|
"loss": 4.6481,
|
|
"mean_token_accuracy": 0.21199302822351457,
|
|
"num_tokens": 13885625.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.283020114898681,
|
|
"epoch": 6.317315658657829,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020620097721382424,
|
|
"loss": 4.6495,
|
|
"mean_token_accuracy": 0.21890244483947754,
|
|
"num_tokens": 13895273.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.339339303970337,
|
|
"epoch": 6.3214581607290805,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002058967958560689,
|
|
"loss": 4.714,
|
|
"mean_token_accuracy": 0.21472427994012833,
|
|
"num_tokens": 13904565.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.211856746673584,
|
|
"epoch": 6.325600662800332,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020559275388702775,
|
|
"loss": 4.5466,
|
|
"mean_token_accuracy": 0.22738529294729232,
|
|
"num_tokens": 13913864.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.302509546279907,
|
|
"epoch": 6.329743164871583,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020528885191998682,
|
|
"loss": 4.707,
|
|
"mean_token_accuracy": 0.21997793316841124,
|
|
"num_tokens": 13923522.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.2530739307403564,
|
|
"epoch": 6.333885666942834,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020498509056794952,
|
|
"loss": 4.6238,
|
|
"mean_token_accuracy": 0.22173805981874467,
|
|
"num_tokens": 13932930.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.31190299987793,
|
|
"epoch": 6.338028169014084,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020468147044363556,
|
|
"loss": 4.6546,
|
|
"mean_token_accuracy": 0.21905871778726577,
|
|
"num_tokens": 13941242.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.310239219665528,
|
|
"epoch": 6.342170671085335,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020437799215948017,
|
|
"loss": 4.633,
|
|
"mean_token_accuracy": 0.21343540102243425,
|
|
"num_tokens": 13949722.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.268713188171387,
|
|
"epoch": 6.346313173156586,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002040746563276319,
|
|
"loss": 4.6566,
|
|
"mean_token_accuracy": 0.21334510892629624,
|
|
"num_tokens": 13958902.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.381603145599366,
|
|
"epoch": 6.350455675227837,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020377146355995257,
|
|
"loss": 4.745,
|
|
"mean_token_accuracy": 0.21202935576438903,
|
|
"num_tokens": 13968533.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.3131969451904295,
|
|
"epoch": 6.3545981772990885,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020346841446801495,
|
|
"loss": 4.6307,
|
|
"mean_token_accuracy": 0.21661254912614822,
|
|
"num_tokens": 13977006.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.2711097240448,
|
|
"epoch": 6.35874067937034,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020316550966310222,
|
|
"loss": 4.5828,
|
|
"mean_token_accuracy": 0.22178242355585098,
|
|
"num_tokens": 13986223.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.180701208114624,
|
|
"epoch": 6.362883181441591,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002028627497562065,
|
|
"loss": 4.5371,
|
|
"mean_token_accuracy": 0.22446803152561187,
|
|
"num_tokens": 13994778.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.241506481170655,
|
|
"epoch": 6.367025683512842,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020256013535802748,
|
|
"loss": 4.6208,
|
|
"mean_token_accuracy": 0.21906774044036864,
|
|
"num_tokens": 14003629.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.338815593719483,
|
|
"epoch": 6.371168185584093,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002022576670789717,
|
|
"loss": 4.6752,
|
|
"mean_token_accuracy": 0.2154947593808174,
|
|
"num_tokens": 14013702.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.2692523956298825,
|
|
"epoch": 6.375310687655344,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002019553455291505,
|
|
"loss": 4.5581,
|
|
"mean_token_accuracy": 0.22790279984474182,
|
|
"num_tokens": 14022170.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.146282148361206,
|
|
"epoch": 6.379453189726595,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002016531713183797,
|
|
"loss": 4.5279,
|
|
"mean_token_accuracy": 0.23489146083593368,
|
|
"num_tokens": 14031983.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.297533655166626,
|
|
"epoch": 6.383595691797846,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00020135114505617752,
|
|
"loss": 4.6769,
|
|
"mean_token_accuracy": 0.21216924786567687,
|
|
"num_tokens": 14040432.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.2439323425292965,
|
|
"epoch": 6.387738193869097,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020104926735176393,
|
|
"loss": 4.5431,
|
|
"mean_token_accuracy": 0.224224990606308,
|
|
"num_tokens": 14049183.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.259787225723267,
|
|
"epoch": 6.391880695940348,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020074753881405949,
|
|
"loss": 4.5609,
|
|
"mean_token_accuracy": 0.22817756086587906,
|
|
"num_tokens": 14057853.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.227885580062866,
|
|
"epoch": 6.396023198011599,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002004459600516833,
|
|
"loss": 4.5914,
|
|
"mean_token_accuracy": 0.22620867043733597,
|
|
"num_tokens": 14066817.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.221218156814575,
|
|
"epoch": 6.40016570008285,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020014453167295298,
|
|
"loss": 4.5813,
|
|
"mean_token_accuracy": 0.2227261781692505,
|
|
"num_tokens": 14075450.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.260425567626953,
|
|
"epoch": 6.404308202154101,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019984325428588229,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.212201352417469,
|
|
"num_tokens": 14084612.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.224593448638916,
|
|
"epoch": 6.408450704225352,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019954212849818077,
|
|
"loss": 4.4831,
|
|
"mean_token_accuracy": 0.23297636210918427,
|
|
"num_tokens": 14093571.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.295340919494629,
|
|
"epoch": 6.412593206296603,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019924115491725197,
|
|
"loss": 4.6798,
|
|
"mean_token_accuracy": 0.2140097588300705,
|
|
"num_tokens": 14103419.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.2957884788513185,
|
|
"epoch": 6.416735708367854,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019894033415019253,
|
|
"loss": 4.687,
|
|
"mean_token_accuracy": 0.20762077569961548,
|
|
"num_tokens": 14113468.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.22314600944519,
|
|
"epoch": 6.420878210439105,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019863966680379086,
|
|
"loss": 4.6262,
|
|
"mean_token_accuracy": 0.2112201914191246,
|
|
"num_tokens": 14123256.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.284819650650024,
|
|
"epoch": 6.4250207125103564,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001983391534845258,
|
|
"loss": 4.5847,
|
|
"mean_token_accuracy": 0.2226705551147461,
|
|
"num_tokens": 14131824.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.262083721160889,
|
|
"epoch": 6.4291632145816076,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019803879479856547,
|
|
"loss": 4.6847,
|
|
"mean_token_accuracy": 0.21154463440179824,
|
|
"num_tokens": 14141461.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.290934610366821,
|
|
"epoch": 6.433305716652859,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019773859135176626,
|
|
"loss": 4.683,
|
|
"mean_token_accuracy": 0.21248364299535752,
|
|
"num_tokens": 14150924.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.295626306533814,
|
|
"epoch": 6.43744821872411,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001974385437496713,
|
|
"loss": 4.6634,
|
|
"mean_token_accuracy": 0.22182252407073974,
|
|
"num_tokens": 14160292.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.418256521224976,
|
|
"epoch": 6.44159072079536,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019713865259750945,
|
|
"loss": 4.7902,
|
|
"mean_token_accuracy": 0.20880477726459504,
|
|
"num_tokens": 14170070.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.311098098754883,
|
|
"epoch": 6.445733222866611,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019683891850019391,
|
|
"loss": 4.6143,
|
|
"mean_token_accuracy": 0.22034519165754318,
|
|
"num_tokens": 14178941.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.305362319946289,
|
|
"epoch": 6.449875724937862,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019653934206232098,
|
|
"loss": 4.6377,
|
|
"mean_token_accuracy": 0.212357759475708,
|
|
"num_tokens": 14188192.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.264620161056518,
|
|
"epoch": 6.454018227009113,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019623992388816935,
|
|
"loss": 4.6506,
|
|
"mean_token_accuracy": 0.21911658048629762,
|
|
"num_tokens": 14197995.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.244184684753418,
|
|
"epoch": 6.458160729080364,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001959406645816979,
|
|
"loss": 4.6349,
|
|
"mean_token_accuracy": 0.22354553043842315,
|
|
"num_tokens": 14206664.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.284440326690674,
|
|
"epoch": 6.4623032311516155,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019564156474654554,
|
|
"loss": 4.6886,
|
|
"mean_token_accuracy": 0.21510905474424363,
|
|
"num_tokens": 14215263.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.261671447753907,
|
|
"epoch": 6.466445733222867,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019534262498602928,
|
|
"loss": 4.6644,
|
|
"mean_token_accuracy": 0.21712061762809753,
|
|
"num_tokens": 14224718.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.317324447631836,
|
|
"epoch": 6.470588235294118,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019504384590314318,
|
|
"loss": 4.6818,
|
|
"mean_token_accuracy": 0.2202373191714287,
|
|
"num_tokens": 14233590.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.37575249671936,
|
|
"epoch": 6.474730737365369,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019474522810055745,
|
|
"loss": 4.7291,
|
|
"mean_token_accuracy": 0.2100374549627304,
|
|
"num_tokens": 14242536.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.263584995269776,
|
|
"epoch": 6.47887323943662,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019444677218061673,
|
|
"loss": 4.6292,
|
|
"mean_token_accuracy": 0.2164051651954651,
|
|
"num_tokens": 14252125.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.159062671661377,
|
|
"epoch": 6.483015741507871,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001941484787453393,
|
|
"loss": 4.524,
|
|
"mean_token_accuracy": 0.23839974850416185,
|
|
"num_tokens": 14261218.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.2152821063995365,
|
|
"epoch": 6.487158243579122,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019385034839641558,
|
|
"loss": 4.6106,
|
|
"mean_token_accuracy": 0.22316165268421173,
|
|
"num_tokens": 14270243.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.294852828979492,
|
|
"epoch": 6.491300745650372,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001935523817352071,
|
|
"loss": 4.7164,
|
|
"mean_token_accuracy": 0.21525251716375352,
|
|
"num_tokens": 14280124.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.254275178909301,
|
|
"epoch": 6.4954432477216235,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019325457936274516,
|
|
"loss": 4.5792,
|
|
"mean_token_accuracy": 0.2217668190598488,
|
|
"num_tokens": 14289098.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.278547191619873,
|
|
"epoch": 6.499585749792875,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019295694187972955,
|
|
"loss": 4.6649,
|
|
"mean_token_accuracy": 0.21588271260261535,
|
|
"num_tokens": 14297097.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.206375646591186,
|
|
"epoch": 6.503728251864126,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019265946988652783,
|
|
"loss": 4.5639,
|
|
"mean_token_accuracy": 0.23070597201585769,
|
|
"num_tokens": 14305249.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.29442458152771,
|
|
"epoch": 6.507870753935377,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001923621639831733,
|
|
"loss": 4.6758,
|
|
"mean_token_accuracy": 0.21983738094568253,
|
|
"num_tokens": 14313969.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.282436943054199,
|
|
"epoch": 6.512013256006628,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019206502476936467,
|
|
"loss": 4.5783,
|
|
"mean_token_accuracy": 0.22526358515024186,
|
|
"num_tokens": 14322642.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.298219108581543,
|
|
"epoch": 6.516155758077879,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019176805284446414,
|
|
"loss": 4.6965,
|
|
"mean_token_accuracy": 0.20873862653970718,
|
|
"num_tokens": 14331158.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.288141918182373,
|
|
"epoch": 6.52029826014913,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019147124880749633,
|
|
"loss": 4.65,
|
|
"mean_token_accuracy": 0.22162335813045503,
|
|
"num_tokens": 14340623.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.241119241714477,
|
|
"epoch": 6.524440762220381,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019117461325714768,
|
|
"loss": 4.5761,
|
|
"mean_token_accuracy": 0.22655788958072662,
|
|
"num_tokens": 14349418.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.247488117218017,
|
|
"epoch": 6.528583264291632,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019087814679176436,
|
|
"loss": 4.5983,
|
|
"mean_token_accuracy": 0.2192952200770378,
|
|
"num_tokens": 14358556.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.3020223617553714,
|
|
"epoch": 6.5327257663628835,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019058185000935175,
|
|
"loss": 4.6543,
|
|
"mean_token_accuracy": 0.2139863923192024,
|
|
"num_tokens": 14367845.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.283290004730224,
|
|
"epoch": 6.536868268434135,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00019028572350757278,
|
|
"loss": 4.6367,
|
|
"mean_token_accuracy": 0.21764726787805558,
|
|
"num_tokens": 14376700.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.297294473648071,
|
|
"epoch": 6.541010770505386,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018998976788374682,
|
|
"loss": 4.6885,
|
|
"mean_token_accuracy": 0.21217481344938277,
|
|
"num_tokens": 14385993.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.249093961715698,
|
|
"epoch": 6.545153272576636,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018969398373484898,
|
|
"loss": 4.6118,
|
|
"mean_token_accuracy": 0.2169957473874092,
|
|
"num_tokens": 14394616.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.31406569480896,
|
|
"epoch": 6.549295774647887,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00018939837165750796,
|
|
"loss": 4.6841,
|
|
"mean_token_accuracy": 0.20811054706573487,
|
|
"num_tokens": 14403844.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.184445524215699,
|
|
"epoch": 6.553438276719138,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018910293224800584,
|
|
"loss": 4.5902,
|
|
"mean_token_accuracy": 0.21901727169752122,
|
|
"num_tokens": 14413201.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.3348406791687015,
|
|
"epoch": 6.557580778790389,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018880766610227608,
|
|
"loss": 4.7241,
|
|
"mean_token_accuracy": 0.21072426438331604,
|
|
"num_tokens": 14422898.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.322178888320923,
|
|
"epoch": 6.56172328086164,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018851257381590264,
|
|
"loss": 4.6762,
|
|
"mean_token_accuracy": 0.2121797487139702,
|
|
"num_tokens": 14432186.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.2864867687225345,
|
|
"epoch": 6.5658657829328915,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018821765598411917,
|
|
"loss": 4.6836,
|
|
"mean_token_accuracy": 0.20689743310213088,
|
|
"num_tokens": 14441195.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.2478076934814455,
|
|
"epoch": 6.570008285004143,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018792291320180693,
|
|
"loss": 4.5729,
|
|
"mean_token_accuracy": 0.21737836599349974,
|
|
"num_tokens": 14450397.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.292281866073608,
|
|
"epoch": 6.574150787075394,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018762834606349455,
|
|
"loss": 4.7187,
|
|
"mean_token_accuracy": 0.2202954426407814,
|
|
"num_tokens": 14459639.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.292811441421509,
|
|
"epoch": 6.578293289146645,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018733395516335588,
|
|
"loss": 4.7058,
|
|
"mean_token_accuracy": 0.21762485206127166,
|
|
"num_tokens": 14468476.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.305009365081787,
|
|
"epoch": 6.582435791217896,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018703974109520976,
|
|
"loss": 4.643,
|
|
"mean_token_accuracy": 0.22237767428159713,
|
|
"num_tokens": 14477980.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.323591089248657,
|
|
"epoch": 6.586578293289147,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018674570445251798,
|
|
"loss": 4.7087,
|
|
"mean_token_accuracy": 0.21048482209444047,
|
|
"num_tokens": 14487311.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.271584558486938,
|
|
"epoch": 6.590720795360397,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018645184582838465,
|
|
"loss": 4.619,
|
|
"mean_token_accuracy": 0.22284981459379197,
|
|
"num_tokens": 14495665.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.303694534301758,
|
|
"epoch": 6.594863297431648,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018615816581555466,
|
|
"loss": 4.6978,
|
|
"mean_token_accuracy": 0.2111795499920845,
|
|
"num_tokens": 14505002.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.317273426055908,
|
|
"epoch": 6.5990057995028994,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001858646650064128,
|
|
"loss": 4.6857,
|
|
"mean_token_accuracy": 0.2120526447892189,
|
|
"num_tokens": 14514737.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.31736798286438,
|
|
"epoch": 6.603148301574151,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018557134399298215,
|
|
"loss": 4.6046,
|
|
"mean_token_accuracy": 0.23250542879104613,
|
|
"num_tokens": 14523399.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.224788475036621,
|
|
"epoch": 6.607290803645402,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001852782033669233,
|
|
"loss": 4.5517,
|
|
"mean_token_accuracy": 0.22345021218061448,
|
|
"num_tokens": 14532242.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.279419708251953,
|
|
"epoch": 6.611433305716653,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00018498524371953295,
|
|
"loss": 4.7078,
|
|
"mean_token_accuracy": 0.2123997077345848,
|
|
"num_tokens": 14541098.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.263177347183228,
|
|
"epoch": 6.615575807787904,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018469246564174275,
|
|
"loss": 4.6445,
|
|
"mean_token_accuracy": 0.22179186195135117,
|
|
"num_tokens": 14549911.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.268185091018677,
|
|
"epoch": 6.619718309859155,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000184399869724118,
|
|
"loss": 4.6571,
|
|
"mean_token_accuracy": 0.21856060177087783,
|
|
"num_tokens": 14558470.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.212279319763184,
|
|
"epoch": 6.623860811930406,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018410745655685662,
|
|
"loss": 4.5642,
|
|
"mean_token_accuracy": 0.22103830575942993,
|
|
"num_tokens": 14567495.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.245939064025879,
|
|
"epoch": 6.628003314001657,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018381522672978806,
|
|
"loss": 4.6571,
|
|
"mean_token_accuracy": 0.21503240317106248,
|
|
"num_tokens": 14576886.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.628003314001657,
|
|
"eval_entropy": 5.173422715811333,
|
|
"eval_loss": 5.782319068908691,
|
|
"eval_mean_token_accuracy": 0.16456795215352513,
|
|
"eval_num_tokens": 14576886.0,
|
|
"eval_runtime": 2.7676,
|
|
"eval_samples_per_second": 1353.866,
|
|
"eval_steps_per_second": 169.459,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.370665645599365,
|
|
"epoch": 6.632145816072908,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018352318083237167,
|
|
"loss": 4.6838,
|
|
"mean_token_accuracy": 0.2196650817990303,
|
|
"num_tokens": 14586660.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.223986387252808,
|
|
"epoch": 6.636288318144159,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018323131945369603,
|
|
"loss": 4.567,
|
|
"mean_token_accuracy": 0.2226695030927658,
|
|
"num_tokens": 14595937.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.276063251495361,
|
|
"epoch": 6.6404308202154105,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00018293964318247735,
|
|
"loss": 4.596,
|
|
"mean_token_accuracy": 0.22161204516887664,
|
|
"num_tokens": 14604965.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.250232172012329,
|
|
"epoch": 6.644573322286661,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018264815260705863,
|
|
"loss": 4.6172,
|
|
"mean_token_accuracy": 0.21619130969047545,
|
|
"num_tokens": 14614552.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.266603708267212,
|
|
"epoch": 6.648715824357912,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018235684831540826,
|
|
"loss": 4.6462,
|
|
"mean_token_accuracy": 0.21381857395172119,
|
|
"num_tokens": 14624260.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.347579526901245,
|
|
"epoch": 6.652858326429163,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018206573089511862,
|
|
"loss": 4.7024,
|
|
"mean_token_accuracy": 0.21514545530080795,
|
|
"num_tokens": 14634702.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.224014806747436,
|
|
"epoch": 6.657000828500414,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00018177480093340564,
|
|
"loss": 4.5948,
|
|
"mean_token_accuracy": 0.22303812205791473,
|
|
"num_tokens": 14643390.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.220793724060059,
|
|
"epoch": 6.661143330571665,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001814840590171065,
|
|
"loss": 4.6034,
|
|
"mean_token_accuracy": 0.22505060583353043,
|
|
"num_tokens": 14652075.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.220482635498047,
|
|
"epoch": 6.665285832642916,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018119350573267979,
|
|
"loss": 4.6521,
|
|
"mean_token_accuracy": 0.20736779272556305,
|
|
"num_tokens": 14661252.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.3052215576171875,
|
|
"epoch": 6.669428334714167,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018090314166620302,
|
|
"loss": 4.6744,
|
|
"mean_token_accuracy": 0.21989104002714158,
|
|
"num_tokens": 14670652.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.318071985244751,
|
|
"epoch": 6.6735708367854185,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018061296740337225,
|
|
"loss": 4.6652,
|
|
"mean_token_accuracy": 0.2193156436085701,
|
|
"num_tokens": 14680561.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.300692415237426,
|
|
"epoch": 6.67771333885667,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018032298352950077,
|
|
"loss": 4.7194,
|
|
"mean_token_accuracy": 0.21450931429862977,
|
|
"num_tokens": 14689317.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.3064271926879885,
|
|
"epoch": 6.681855840927921,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018003319062951756,
|
|
"loss": 4.7093,
|
|
"mean_token_accuracy": 0.2127823680639267,
|
|
"num_tokens": 14699184.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.246539640426636,
|
|
"epoch": 6.685998342999172,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017974358928796682,
|
|
"loss": 4.6307,
|
|
"mean_token_accuracy": 0.22148862481117249,
|
|
"num_tokens": 14708158.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.2467413425445555,
|
|
"epoch": 6.690140845070422,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017945418008900594,
|
|
"loss": 4.5529,
|
|
"mean_token_accuracy": 0.2329227775335312,
|
|
"num_tokens": 14716865.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.293979787826538,
|
|
"epoch": 6.694283347141674,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001791649636164049,
|
|
"loss": 4.6285,
|
|
"mean_token_accuracy": 0.22072465121746063,
|
|
"num_tokens": 14727121.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.316159343719482,
|
|
"epoch": 6.698425849212924,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000178875940453545,
|
|
"loss": 4.7072,
|
|
"mean_token_accuracy": 0.20998934358358384,
|
|
"num_tokens": 14736994.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.200163984298706,
|
|
"epoch": 6.702568351284175,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00017858711118341742,
|
|
"loss": 4.5999,
|
|
"mean_token_accuracy": 0.2264136090874672,
|
|
"num_tokens": 14745994.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.253979778289795,
|
|
"epoch": 6.7067108533554265,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017829847638862255,
|
|
"loss": 4.6158,
|
|
"mean_token_accuracy": 0.2292914494872093,
|
|
"num_tokens": 14754560.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.271402835845947,
|
|
"epoch": 6.710853355426678,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017801003665136823,
|
|
"loss": 4.5773,
|
|
"mean_token_accuracy": 0.2141339287161827,
|
|
"num_tokens": 14764054.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.240528392791748,
|
|
"epoch": 6.714995857497929,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017772179255346883,
|
|
"loss": 4.6141,
|
|
"mean_token_accuracy": 0.22719011306762696,
|
|
"num_tokens": 14772826.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.277284812927246,
|
|
"epoch": 6.71913835956918,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017743374467634433,
|
|
"loss": 4.6634,
|
|
"mean_token_accuracy": 0.21183691173791885,
|
|
"num_tokens": 14782171.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.211915731430054,
|
|
"epoch": 6.723280861640431,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017714589360101862,
|
|
"loss": 4.5609,
|
|
"mean_token_accuracy": 0.23358217179775237,
|
|
"num_tokens": 14791767.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.226518821716309,
|
|
"epoch": 6.727423363711682,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000176858239908119,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.2316247820854187,
|
|
"num_tokens": 14800614.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.238487195968628,
|
|
"epoch": 6.731565865782933,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017657078417787414,
|
|
"loss": 4.633,
|
|
"mean_token_accuracy": 0.22251679599285126,
|
|
"num_tokens": 14809371.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.330119037628174,
|
|
"epoch": 6.735708367854184,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017628352699011396,
|
|
"loss": 4.7522,
|
|
"mean_token_accuracy": 0.21364987939596175,
|
|
"num_tokens": 14817988.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.278370046615601,
|
|
"epoch": 6.739850869925435,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001759964689242674,
|
|
"loss": 4.6559,
|
|
"mean_token_accuracy": 0.2217027574777603,
|
|
"num_tokens": 14826825.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.358547353744507,
|
|
"epoch": 6.743993371996686,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001757096105593619,
|
|
"loss": 4.6079,
|
|
"mean_token_accuracy": 0.22096496671438218,
|
|
"num_tokens": 14835809.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.291381978988648,
|
|
"epoch": 6.748135874067937,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017542295247402234,
|
|
"loss": 4.6157,
|
|
"mean_token_accuracy": 0.20849885642528534,
|
|
"num_tokens": 14844508.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.307956695556641,
|
|
"epoch": 6.752278376139188,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00017513649524646919,
|
|
"loss": 4.6029,
|
|
"mean_token_accuracy": 0.22766001671552658,
|
|
"num_tokens": 14851818.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.228078079223633,
|
|
"epoch": 6.756420878210439,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001748502394545181,
|
|
"loss": 4.6288,
|
|
"mean_token_accuracy": 0.22155047953128815,
|
|
"num_tokens": 14861538.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.21757960319519,
|
|
"epoch": 6.76056338028169,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001745641856755783,
|
|
"loss": 4.5732,
|
|
"mean_token_accuracy": 0.21470511257648467,
|
|
"num_tokens": 14870169.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.23885145187378,
|
|
"epoch": 6.764705882352941,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017427833448665136,
|
|
"loss": 4.5686,
|
|
"mean_token_accuracy": 0.21864869445562363,
|
|
"num_tokens": 14878868.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.210526752471924,
|
|
"epoch": 6.768848384424192,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017399268646433065,
|
|
"loss": 4.6019,
|
|
"mean_token_accuracy": 0.21890079081058503,
|
|
"num_tokens": 14887613.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.206032180786133,
|
|
"epoch": 6.772990886495443,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017370724218479912,
|
|
"loss": 4.6646,
|
|
"mean_token_accuracy": 0.21488202810287477,
|
|
"num_tokens": 14895800.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.280506658554077,
|
|
"epoch": 6.7771333885666944,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017342200222382934,
|
|
"loss": 4.708,
|
|
"mean_token_accuracy": 0.21405385583639144,
|
|
"num_tokens": 14905105.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.305052185058594,
|
|
"epoch": 6.781275890637946,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017313696715678134,
|
|
"loss": 4.6535,
|
|
"mean_token_accuracy": 0.21793029457330704,
|
|
"num_tokens": 14914357.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.277199935913086,
|
|
"epoch": 6.785418392709197,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000172852137558602,
|
|
"loss": 4.6526,
|
|
"mean_token_accuracy": 0.21921917647123337,
|
|
"num_tokens": 14923315.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.307702970504761,
|
|
"epoch": 6.789560894780448,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017256751400382384,
|
|
"loss": 4.6205,
|
|
"mean_token_accuracy": 0.22228205054998398,
|
|
"num_tokens": 14931953.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.389893484115601,
|
|
"epoch": 6.793703396851699,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017228309706656354,
|
|
"loss": 4.8168,
|
|
"mean_token_accuracy": 0.20108604729175567,
|
|
"num_tokens": 14940647.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.2408195495605465,
|
|
"epoch": 6.797845898922949,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001719988873205213,
|
|
"loss": 4.5818,
|
|
"mean_token_accuracy": 0.2250256657600403,
|
|
"num_tokens": 14949402.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.304331731796265,
|
|
"epoch": 6.8019884009942,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017171488533897916,
|
|
"loss": 4.6896,
|
|
"mean_token_accuracy": 0.2166202872991562,
|
|
"num_tokens": 14958931.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.2377660274505615,
|
|
"epoch": 6.806130903065451,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017143109169480005,
|
|
"loss": 4.561,
|
|
"mean_token_accuracy": 0.22895603030920028,
|
|
"num_tokens": 14967614.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.297658634185791,
|
|
"epoch": 6.810273405136702,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00017114750696042695,
|
|
"loss": 4.6626,
|
|
"mean_token_accuracy": 0.22159790992736816,
|
|
"num_tokens": 14975910.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.237723350524902,
|
|
"epoch": 6.8144159072079535,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017086413170788107,
|
|
"loss": 4.5484,
|
|
"mean_token_accuracy": 0.22901702523231507,
|
|
"num_tokens": 14984204.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.229389572143555,
|
|
"epoch": 6.818558409279205,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017058096650876143,
|
|
"loss": 4.5778,
|
|
"mean_token_accuracy": 0.22689877301454545,
|
|
"num_tokens": 14992769.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.2746611595153805,
|
|
"epoch": 6.822700911350456,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017029801193424313,
|
|
"loss": 4.6907,
|
|
"mean_token_accuracy": 0.2099005937576294,
|
|
"num_tokens": 15001722.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.259488582611084,
|
|
"epoch": 6.826843413421707,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017001526855507632,
|
|
"loss": 4.5737,
|
|
"mean_token_accuracy": 0.224118672311306,
|
|
"num_tokens": 15011148.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.252034044265747,
|
|
"epoch": 6.830985915492958,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016973273694158555,
|
|
"loss": 4.6419,
|
|
"mean_token_accuracy": 0.21621179580688477,
|
|
"num_tokens": 15020449.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.314564895629883,
|
|
"epoch": 6.835128417564209,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016945041766366773,
|
|
"loss": 4.6614,
|
|
"mean_token_accuracy": 0.21861369907855988,
|
|
"num_tokens": 15029813.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.2483312606811525,
|
|
"epoch": 6.83927091963546,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00016916831129079187,
|
|
"loss": 4.7578,
|
|
"mean_token_accuracy": 0.2152937039732933,
|
|
"num_tokens": 15038304.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.315652179718017,
|
|
"epoch": 6.84341342170671,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001688864183919973,
|
|
"loss": 4.6673,
|
|
"mean_token_accuracy": 0.21041538417339326,
|
|
"num_tokens": 15048037.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.329968070983886,
|
|
"epoch": 6.8475559237779615,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016860473953589266,
|
|
"loss": 4.7167,
|
|
"mean_token_accuracy": 0.2182564914226532,
|
|
"num_tokens": 15057763.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.359201240539551,
|
|
"epoch": 6.851698425849213,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016832327529065522,
|
|
"loss": 4.6461,
|
|
"mean_token_accuracy": 0.2221391662955284,
|
|
"num_tokens": 15067418.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.3020414352417,
|
|
"epoch": 6.855840927920464,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001680420262240289,
|
|
"loss": 4.6346,
|
|
"mean_token_accuracy": 0.2141245648264885,
|
|
"num_tokens": 15076907.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.252104139328003,
|
|
"epoch": 6.859983429991715,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016776099290332397,
|
|
"loss": 4.6106,
|
|
"mean_token_accuracy": 0.2266027644276619,
|
|
"num_tokens": 15086448.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.334567403793335,
|
|
"epoch": 6.864125932062966,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016748017589541525,
|
|
"loss": 4.7109,
|
|
"mean_token_accuracy": 0.2106695517897606,
|
|
"num_tokens": 15096277.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.315297269821167,
|
|
"epoch": 6.868268434134217,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001671995757667414,
|
|
"loss": 4.6403,
|
|
"mean_token_accuracy": 0.2208337664604187,
|
|
"num_tokens": 15104657.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.271635723114014,
|
|
"epoch": 6.872410936205468,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016691919308330343,
|
|
"loss": 4.5868,
|
|
"mean_token_accuracy": 0.21768613904714584,
|
|
"num_tokens": 15113096.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.266825389862061,
|
|
"epoch": 6.876553438276719,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016663902841066404,
|
|
"loss": 4.6526,
|
|
"mean_token_accuracy": 0.2212653636932373,
|
|
"num_tokens": 15122556.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.273133707046509,
|
|
"epoch": 6.88069594034797,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016635908231394587,
|
|
"loss": 4.6832,
|
|
"mean_token_accuracy": 0.21701868027448654,
|
|
"num_tokens": 15132011.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.332955455780029,
|
|
"epoch": 6.8848384424192215,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001660793553578307,
|
|
"loss": 4.6756,
|
|
"mean_token_accuracy": 0.21488548666238785,
|
|
"num_tokens": 15141881.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.30683708190918,
|
|
"epoch": 6.888980944490473,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016579984810655856,
|
|
"loss": 4.6966,
|
|
"mean_token_accuracy": 0.2110757350921631,
|
|
"num_tokens": 15152318.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.219453191757202,
|
|
"epoch": 6.893123446561724,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016552056112392594,
|
|
"loss": 4.5845,
|
|
"mean_token_accuracy": 0.2309218689799309,
|
|
"num_tokens": 15161287.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.230193662643432,
|
|
"epoch": 6.897265948632974,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001652414949732854,
|
|
"loss": 4.6793,
|
|
"mean_token_accuracy": 0.22260308861732483,
|
|
"num_tokens": 15170658.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.2259667873382565,
|
|
"epoch": 6.901408450704225,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016496265021754365,
|
|
"loss": 4.5441,
|
|
"mean_token_accuracy": 0.23273635655641556,
|
|
"num_tokens": 15180046.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.277891778945923,
|
|
"epoch": 6.905550952775476,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016468402741916124,
|
|
"loss": 4.5725,
|
|
"mean_token_accuracy": 0.2279438704252243,
|
|
"num_tokens": 15188909.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.311296272277832,
|
|
"epoch": 6.909693454846727,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001644056271401506,
|
|
"loss": 4.6803,
|
|
"mean_token_accuracy": 0.21608312875032426,
|
|
"num_tokens": 15198655.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.275272464752197,
|
|
"epoch": 6.913835956917978,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016412744994207573,
|
|
"loss": 4.6888,
|
|
"mean_token_accuracy": 0.2144805982708931,
|
|
"num_tokens": 15208149.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.317964696884156,
|
|
"epoch": 6.9179784589892295,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016384949638605025,
|
|
"loss": 4.7037,
|
|
"mean_token_accuracy": 0.21555749028921128,
|
|
"num_tokens": 15217529.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.317746353149414,
|
|
"epoch": 6.922120961060481,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016357176703273696,
|
|
"loss": 4.7464,
|
|
"mean_token_accuracy": 0.21130641251802446,
|
|
"num_tokens": 15226607.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.280500650405884,
|
|
"epoch": 6.926263463131732,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001632942624423463,
|
|
"loss": 4.6271,
|
|
"mean_token_accuracy": 0.21781034171581268,
|
|
"num_tokens": 15235673.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.327365207672119,
|
|
"epoch": 6.930405965202983,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016301698317463525,
|
|
"loss": 4.6897,
|
|
"mean_token_accuracy": 0.21666931062936784,
|
|
"num_tokens": 15244735.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.311723899841309,
|
|
"epoch": 6.934548467274234,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016273992978890658,
|
|
"loss": 4.6901,
|
|
"mean_token_accuracy": 0.21131074279546738,
|
|
"num_tokens": 15254330.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.299632930755616,
|
|
"epoch": 6.938690969345485,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001624631028440071,
|
|
"loss": 4.6017,
|
|
"mean_token_accuracy": 0.2292412906885147,
|
|
"num_tokens": 15263136.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.283994054794311,
|
|
"epoch": 6.942833471416735,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001621865028983271,
|
|
"loss": 4.6293,
|
|
"mean_token_accuracy": 0.21863681077957153,
|
|
"num_tokens": 15272355.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.243247222900391,
|
|
"epoch": 6.946975973487987,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001619101305097988,
|
|
"loss": 4.6165,
|
|
"mean_token_accuracy": 0.22120813578367232,
|
|
"num_tokens": 15281855.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.296655225753784,
|
|
"epoch": 6.9511184755592375,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016163398623589563,
|
|
"loss": 4.7466,
|
|
"mean_token_accuracy": 0.20605631470680236,
|
|
"num_tokens": 15292328.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.307953882217407,
|
|
"epoch": 6.955260977630489,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016135807063363074,
|
|
"loss": 4.6716,
|
|
"mean_token_accuracy": 0.21860750317573546,
|
|
"num_tokens": 15302093.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.258717727661133,
|
|
"epoch": 6.95940347970174,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016108238425955584,
|
|
"loss": 4.6501,
|
|
"mean_token_accuracy": 0.22871047258377075,
|
|
"num_tokens": 15311506.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.202821397781372,
|
|
"epoch": 6.963545981772991,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001608069276697608,
|
|
"loss": 4.5225,
|
|
"mean_token_accuracy": 0.2219666585326195,
|
|
"num_tokens": 15320268.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.226770544052124,
|
|
"epoch": 6.967688483844242,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001605317014198714,
|
|
"loss": 4.5669,
|
|
"mean_token_accuracy": 0.22790412306785585,
|
|
"num_tokens": 15329528.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.210282611846924,
|
|
"epoch": 6.971830985915493,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016025670606504926,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.2133000135421753,
|
|
"num_tokens": 15338731.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.253313302993774,
|
|
"epoch": 6.975973487986744,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001599819421599899,
|
|
"loss": 4.635,
|
|
"mean_token_accuracy": 0.22574678659439087,
|
|
"num_tokens": 15347704.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.250069189071655,
|
|
"epoch": 6.980115990057995,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015970741025892233,
|
|
"loss": 4.6401,
|
|
"mean_token_accuracy": 0.22157265841960908,
|
|
"num_tokens": 15356717.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.2906712055206295,
|
|
"epoch": 6.984258492129246,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015943311091560721,
|
|
"loss": 4.733,
|
|
"mean_token_accuracy": 0.20928027331829072,
|
|
"num_tokens": 15366436.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.343661308288574,
|
|
"epoch": 6.988400994200497,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015915904468333635,
|
|
"loss": 4.6965,
|
|
"mean_token_accuracy": 0.21900769025087358,
|
|
"num_tokens": 15375854.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.3276725769042965,
|
|
"epoch": 6.9925434962717485,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00015888521211493135,
|
|
"loss": 4.748,
|
|
"mean_token_accuracy": 0.20656630545854568,
|
|
"num_tokens": 15385174.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.237693881988525,
|
|
"epoch": 6.996685998342999,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001586116137627423,
|
|
"loss": 4.4846,
|
|
"mean_token_accuracy": 0.22596284598112107,
|
|
"num_tokens": 15393792.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.250106334686279,
|
|
"epoch": 7.00082850041425,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015833825017864711,
|
|
"loss": 4.5889,
|
|
"mean_token_accuracy": 0.22144968062639236,
|
|
"num_tokens": 15401921.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.225988149642944,
|
|
"epoch": 7.004971002485501,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015806512191405,
|
|
"loss": 4.4909,
|
|
"mean_token_accuracy": 0.2232973873615265,
|
|
"num_tokens": 15410937.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.3065837860107425,
|
|
"epoch": 7.009113504556752,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001577922295198803,
|
|
"loss": 4.5466,
|
|
"mean_token_accuracy": 0.22927301973104477,
|
|
"num_tokens": 15419608.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.319969654083252,
|
|
"epoch": 7.013256006628003,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015751957354659206,
|
|
"loss": 4.5969,
|
|
"mean_token_accuracy": 0.2271721124649048,
|
|
"num_tokens": 15428741.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.280915689468384,
|
|
"epoch": 7.017398508699254,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015724715454416206,
|
|
"loss": 4.5359,
|
|
"mean_token_accuracy": 0.2312479332089424,
|
|
"num_tokens": 15438473.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.2156414031982425,
|
|
"epoch": 7.021541010770505,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015697497306208926,
|
|
"loss": 4.5341,
|
|
"mean_token_accuracy": 0.22636407911777495,
|
|
"num_tokens": 15448188.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.199126529693603,
|
|
"epoch": 7.0256835128417565,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001567030296493934,
|
|
"loss": 4.4442,
|
|
"mean_token_accuracy": 0.2398556724190712,
|
|
"num_tokens": 15457116.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.289102935791016,
|
|
"epoch": 7.029826014913008,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015643132485461395,
|
|
"loss": 4.4445,
|
|
"mean_token_accuracy": 0.23127604871988297,
|
|
"num_tokens": 15466107.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.3634435653686525,
|
|
"epoch": 7.033968516984259,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001561598592258094,
|
|
"loss": 4.6541,
|
|
"mean_token_accuracy": 0.2178284153342247,
|
|
"num_tokens": 15475158.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.290414476394654,
|
|
"epoch": 7.03811101905551,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00015588863331055541,
|
|
"loss": 4.5483,
|
|
"mean_token_accuracy": 0.22435671389102935,
|
|
"num_tokens": 15483529.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.274207878112793,
|
|
"epoch": 7.042253521126761,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001556176476559444,
|
|
"loss": 4.4435,
|
|
"mean_token_accuracy": 0.2442246526479721,
|
|
"num_tokens": 15493072.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.042253521126761,
|
|
"eval_entropy": 5.1849700281106585,
|
|
"eval_loss": 5.774664402008057,
|
|
"eval_mean_token_accuracy": 0.1654605287065638,
|
|
"eval_num_tokens": 15493072.0,
|
|
"eval_runtime": 2.839,
|
|
"eval_samples_per_second": 1319.823,
|
|
"eval_steps_per_second": 165.198,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.295238924026489,
|
|
"epoch": 7.046396023198011,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015534690280858405,
|
|
"loss": 4.5218,
|
|
"mean_token_accuracy": 0.23220051378011702,
|
|
"num_tokens": 15502522.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.299421977996826,
|
|
"epoch": 7.050538525269262,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015507639931459615,
|
|
"loss": 4.5102,
|
|
"mean_token_accuracy": 0.23622628599405288,
|
|
"num_tokens": 15511641.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.200384330749512,
|
|
"epoch": 7.054681027340513,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000154806137719616,
|
|
"loss": 4.4419,
|
|
"mean_token_accuracy": 0.2457203224301338,
|
|
"num_tokens": 15520045.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.212980270385742,
|
|
"epoch": 7.0588235294117645,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001545361185687906,
|
|
"loss": 4.4402,
|
|
"mean_token_accuracy": 0.23710457235574722,
|
|
"num_tokens": 15528635.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.248097896575928,
|
|
"epoch": 7.062966031483016,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015426634240677824,
|
|
"loss": 4.5268,
|
|
"mean_token_accuracy": 0.22719627022743225,
|
|
"num_tokens": 15538708.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.1817912578582765,
|
|
"epoch": 7.067108533554267,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00015399680977774689,
|
|
"loss": 4.4094,
|
|
"mean_token_accuracy": 0.23902855068445206,
|
|
"num_tokens": 15547202.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.2307576656341555,
|
|
"epoch": 7.071251035625518,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015372752122537318,
|
|
"loss": 4.5633,
|
|
"mean_token_accuracy": 0.22459081560373306,
|
|
"num_tokens": 15555875.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.1603546142578125,
|
|
"epoch": 7.075393537696769,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015345847729284168,
|
|
"loss": 4.4137,
|
|
"mean_token_accuracy": 0.24392253458499907,
|
|
"num_tokens": 15564511.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.215771436691284,
|
|
"epoch": 7.07953603976802,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00015318967852284335,
|
|
"loss": 4.4711,
|
|
"mean_token_accuracy": 0.23276307433843613,
|
|
"num_tokens": 15573389.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.268087387084961,
|
|
"epoch": 7.083678541839271,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015292112545757476,
|
|
"loss": 4.5419,
|
|
"mean_token_accuracy": 0.22359508723020555,
|
|
"num_tokens": 15582884.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.296631097793579,
|
|
"epoch": 7.087821043910522,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015265281863873663,
|
|
"loss": 4.5476,
|
|
"mean_token_accuracy": 0.22547755390405655,
|
|
"num_tokens": 15591959.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.202712488174439,
|
|
"epoch": 7.091963545981773,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015238475860753338,
|
|
"loss": 4.4529,
|
|
"mean_token_accuracy": 0.23556775897741317,
|
|
"num_tokens": 15600849.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.304818534851075,
|
|
"epoch": 7.096106048053024,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015211694590467118,
|
|
"loss": 4.5955,
|
|
"mean_token_accuracy": 0.22399300187826157,
|
|
"num_tokens": 15610424.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.305905961990357,
|
|
"epoch": 7.100248550124275,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001518493810703575,
|
|
"loss": 4.5176,
|
|
"mean_token_accuracy": 0.22609855532646178,
|
|
"num_tokens": 15618875.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.249867343902588,
|
|
"epoch": 7.104391052195526,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00015158206464430002,
|
|
"loss": 4.5412,
|
|
"mean_token_accuracy": 0.22556609511375428,
|
|
"num_tokens": 15627746.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.2521843910217285,
|
|
"epoch": 7.108533554266777,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000151314997165705,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.22658670991659163,
|
|
"num_tokens": 15636840.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.236520910263062,
|
|
"epoch": 7.112676056338028,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015104817917327694,
|
|
"loss": 4.518,
|
|
"mean_token_accuracy": 0.22916755229234695,
|
|
"num_tokens": 15646035.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.297655916213989,
|
|
"epoch": 7.116818558409279,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00015078161120521676,
|
|
"loss": 4.475,
|
|
"mean_token_accuracy": 0.24264889508485793,
|
|
"num_tokens": 15654786.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.183017349243164,
|
|
"epoch": 7.12096106048053,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001505152937992211,
|
|
"loss": 4.4643,
|
|
"mean_token_accuracy": 0.23951448053121566,
|
|
"num_tokens": 15663992.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.198861026763916,
|
|
"epoch": 7.125103562551781,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015024922749248143,
|
|
"loss": 4.4638,
|
|
"mean_token_accuracy": 0.23570709228515624,
|
|
"num_tokens": 15672809.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.246333646774292,
|
|
"epoch": 7.1292460646230325,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014998341282168243,
|
|
"loss": 4.5546,
|
|
"mean_token_accuracy": 0.22476116865873336,
|
|
"num_tokens": 15682105.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.334461641311646,
|
|
"epoch": 7.133388566694284,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001497178503230015,
|
|
"loss": 4.6262,
|
|
"mean_token_accuracy": 0.21646377593278884,
|
|
"num_tokens": 15690840.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.3045751571655275,
|
|
"epoch": 7.137531068765535,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014945254053210712,
|
|
"loss": 4.5553,
|
|
"mean_token_accuracy": 0.21858414709568025,
|
|
"num_tokens": 15699953.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.266128253936768,
|
|
"epoch": 7.141673570836786,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00014918748398415801,
|
|
"loss": 4.4685,
|
|
"mean_token_accuracy": 0.23337865769863128,
|
|
"num_tokens": 15707707.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.235085201263428,
|
|
"epoch": 7.145816072908037,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014892268121380244,
|
|
"loss": 4.48,
|
|
"mean_token_accuracy": 0.23441701680421828,
|
|
"num_tokens": 15716575.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.2462553024292,
|
|
"epoch": 7.149958574979287,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014865813275517637,
|
|
"loss": 4.5484,
|
|
"mean_token_accuracy": 0.22458526045083999,
|
|
"num_tokens": 15725960.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.221630573272705,
|
|
"epoch": 7.154101077050538,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014839383914190302,
|
|
"loss": 4.4317,
|
|
"mean_token_accuracy": 0.24238034933805466,
|
|
"num_tokens": 15735793.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.2364397048950195,
|
|
"epoch": 7.158243579121789,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001481298009070915,
|
|
"loss": 4.5312,
|
|
"mean_token_accuracy": 0.22333418428897858,
|
|
"num_tokens": 15745131.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.221724796295166,
|
|
"epoch": 7.16238608119304,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014786601858333566,
|
|
"loss": 4.5512,
|
|
"mean_token_accuracy": 0.22646010369062425,
|
|
"num_tokens": 15754431.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.239632320404053,
|
|
"epoch": 7.1665285832642915,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014760249270271341,
|
|
"loss": 4.4842,
|
|
"mean_token_accuracy": 0.2330325022339821,
|
|
"num_tokens": 15763181.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.19738187789917,
|
|
"epoch": 7.170671085335543,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014733922379678513,
|
|
"loss": 4.4657,
|
|
"mean_token_accuracy": 0.23221879303455353,
|
|
"num_tokens": 15772888.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.290270519256592,
|
|
"epoch": 7.174813587406794,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014707621239659312,
|
|
"loss": 4.5383,
|
|
"mean_token_accuracy": 0.22485556751489638,
|
|
"num_tokens": 15783244.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.199770832061768,
|
|
"epoch": 7.178956089478045,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001468134590326601,
|
|
"loss": 4.4758,
|
|
"mean_token_accuracy": 0.23595971018075942,
|
|
"num_tokens": 15791920.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.267360019683838,
|
|
"epoch": 7.183098591549296,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001465509642349881,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.22791072130203247,
|
|
"num_tokens": 15800517.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.280459403991699,
|
|
"epoch": 7.187241093620547,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014628872853305808,
|
|
"loss": 4.5636,
|
|
"mean_token_accuracy": 0.23035934418439866,
|
|
"num_tokens": 15810417.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.23087100982666,
|
|
"epoch": 7.191383595691798,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014602675245582795,
|
|
"loss": 4.4975,
|
|
"mean_token_accuracy": 0.23260946124792098,
|
|
"num_tokens": 15819660.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.334040546417237,
|
|
"epoch": 7.195526097763049,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014576503653173219,
|
|
"loss": 4.5095,
|
|
"mean_token_accuracy": 0.2279452234506607,
|
|
"num_tokens": 15828618.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.270238065719605,
|
|
"epoch": 7.1996685998342995,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014550358128868036,
|
|
"loss": 4.4931,
|
|
"mean_token_accuracy": 0.22704765796661378,
|
|
"num_tokens": 15837384.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.289967632293701,
|
|
"epoch": 7.203811101905551,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014524238725405615,
|
|
"loss": 4.5468,
|
|
"mean_token_accuracy": 0.23718589842319487,
|
|
"num_tokens": 15846495.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.227925157546997,
|
|
"epoch": 7.207953603976802,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014498145495471664,
|
|
"loss": 4.4725,
|
|
"mean_token_accuracy": 0.23716235607862474,
|
|
"num_tokens": 15854936.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.321663808822632,
|
|
"epoch": 7.212096106048053,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014472078491699076,
|
|
"loss": 4.6266,
|
|
"mean_token_accuracy": 0.22154373973608016,
|
|
"num_tokens": 15864120.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.3266431331634525,
|
|
"epoch": 7.216238608119304,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001444603776666784,
|
|
"loss": 4.5878,
|
|
"mean_token_accuracy": 0.22720413655042648,
|
|
"num_tokens": 15872503.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.234654426574707,
|
|
"epoch": 7.220381110190555,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014420023372904944,
|
|
"loss": 4.5308,
|
|
"mean_token_accuracy": 0.22646201252937317,
|
|
"num_tokens": 15881824.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.345166254043579,
|
|
"epoch": 7.224523612261806,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00014394035362884274,
|
|
"loss": 4.5398,
|
|
"mean_token_accuracy": 0.23385821431875228,
|
|
"num_tokens": 15891000.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.245229864120484,
|
|
"epoch": 7.228666114333057,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014368073789026482,
|
|
"loss": 4.4725,
|
|
"mean_token_accuracy": 0.22802259922027587,
|
|
"num_tokens": 15899414.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.358830499649048,
|
|
"epoch": 7.232808616404308,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014342138703698882,
|
|
"loss": 4.6692,
|
|
"mean_token_accuracy": 0.21760265678167343,
|
|
"num_tokens": 15908681.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.297170114517212,
|
|
"epoch": 7.2369511184755595,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000143162301592154,
|
|
"loss": 4.608,
|
|
"mean_token_accuracy": 0.22596510052680968,
|
|
"num_tokens": 15918245.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.31043643951416,
|
|
"epoch": 7.241093620546811,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00014290348207836385,
|
|
"loss": 4.5504,
|
|
"mean_token_accuracy": 0.22361586838960648,
|
|
"num_tokens": 15927512.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.279439783096313,
|
|
"epoch": 7.245236122618062,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001426449290176857,
|
|
"loss": 4.5378,
|
|
"mean_token_accuracy": 0.22593795955181123,
|
|
"num_tokens": 15936449.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.192220306396484,
|
|
"epoch": 7.249378624689312,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00014238664293164934,
|
|
"loss": 4.4353,
|
|
"mean_token_accuracy": 0.2399519518017769,
|
|
"num_tokens": 15944456.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.123786306381225,
|
|
"epoch": 7.253521126760563,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001421286243412459,
|
|
"loss": 4.3581,
|
|
"mean_token_accuracy": 0.23915824443101882,
|
|
"num_tokens": 15952273.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.186490249633789,
|
|
"epoch": 7.257663628831814,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014187087376692724,
|
|
"loss": 4.5058,
|
|
"mean_token_accuracy": 0.23586500883102418,
|
|
"num_tokens": 15961439.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.238715410232544,
|
|
"epoch": 7.261806130903065,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001416133917286043,
|
|
"loss": 4.4819,
|
|
"mean_token_accuracy": 0.22957350462675094,
|
|
"num_tokens": 15970236.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.219631481170654,
|
|
"epoch": 7.265948632974316,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001413561787456466,
|
|
"loss": 4.509,
|
|
"mean_token_accuracy": 0.23185957372188568,
|
|
"num_tokens": 15980781.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.199354648590088,
|
|
"epoch": 7.2700911350455675,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014109923533688068,
|
|
"loss": 4.4988,
|
|
"mean_token_accuracy": 0.23691075891256333,
|
|
"num_tokens": 15990875.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.256020259857178,
|
|
"epoch": 7.274233637116819,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001408425620205897,
|
|
"loss": 4.5334,
|
|
"mean_token_accuracy": 0.2256982818245888,
|
|
"num_tokens": 15999732.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.246872520446777,
|
|
"epoch": 7.27837613918807,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001405861593145116,
|
|
"loss": 4.5206,
|
|
"mean_token_accuracy": 0.2306825891137123,
|
|
"num_tokens": 16008955.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.2469298362731935,
|
|
"epoch": 7.282518641259321,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014033002773583872,
|
|
"loss": 4.4435,
|
|
"mean_token_accuracy": 0.2337464764714241,
|
|
"num_tokens": 16017076.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.2458672523498535,
|
|
"epoch": 7.286661143330572,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001400741678012164,
|
|
"loss": 4.5658,
|
|
"mean_token_accuracy": 0.23537755757570267,
|
|
"num_tokens": 16025561.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.257904720306397,
|
|
"epoch": 7.290803645401823,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013981858002674223,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.22268954515457154,
|
|
"num_tokens": 16034853.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.304434967041016,
|
|
"epoch": 7.294946147473074,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001395632649279645,
|
|
"loss": 4.5766,
|
|
"mean_token_accuracy": 0.22462111562490464,
|
|
"num_tokens": 16043830.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.316950893402099,
|
|
"epoch": 7.299088649544325,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013930822301988177,
|
|
"loss": 4.5456,
|
|
"mean_token_accuracy": 0.22120891213417054,
|
|
"num_tokens": 16053327.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.300349330902099,
|
|
"epoch": 7.3032311516155755,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013905345481694147,
|
|
"loss": 4.555,
|
|
"mean_token_accuracy": 0.22759606540203095,
|
|
"num_tokens": 16061882.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.243546056747436,
|
|
"epoch": 7.307373653686827,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001387989608330388,
|
|
"loss": 4.5452,
|
|
"mean_token_accuracy": 0.23280268013477326,
|
|
"num_tokens": 16070645.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.262579154968262,
|
|
"epoch": 7.311516155758078,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013854474158151602,
|
|
"loss": 4.5626,
|
|
"mean_token_accuracy": 0.22421094924211502,
|
|
"num_tokens": 16079689.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.26329231262207,
|
|
"epoch": 7.315658657829329,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013829079757516106,
|
|
"loss": 4.5496,
|
|
"mean_token_accuracy": 0.23165889978408813,
|
|
"num_tokens": 16088265.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.231566143035889,
|
|
"epoch": 7.31980115990058,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013803712932620681,
|
|
"loss": 4.5021,
|
|
"mean_token_accuracy": 0.22283203154802322,
|
|
"num_tokens": 16097921.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.229958581924438,
|
|
"epoch": 7.323943661971831,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001377837373463298,
|
|
"loss": 4.5373,
|
|
"mean_token_accuracy": 0.22989922463893891,
|
|
"num_tokens": 16107204.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.242744255065918,
|
|
"epoch": 7.328086164043082,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001375306221466494,
|
|
"loss": 4.4969,
|
|
"mean_token_accuracy": 0.234531269967556,
|
|
"num_tokens": 16115588.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.246748065948486,
|
|
"epoch": 7.332228666114333,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013727778423772654,
|
|
"loss": 4.5108,
|
|
"mean_token_accuracy": 0.2347363755106926,
|
|
"num_tokens": 16124469.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.2705645084381105,
|
|
"epoch": 7.336371168185584,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013702522412956304,
|
|
"loss": 4.4983,
|
|
"mean_token_accuracy": 0.23824732303619384,
|
|
"num_tokens": 16133175.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.315527296066284,
|
|
"epoch": 7.340513670256835,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013677294233160016,
|
|
"loss": 4.5118,
|
|
"mean_token_accuracy": 0.2352664813399315,
|
|
"num_tokens": 16142260.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.314030456542969,
|
|
"epoch": 7.3446561723280865,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013652093935271776,
|
|
"loss": 4.591,
|
|
"mean_token_accuracy": 0.21988280415534972,
|
|
"num_tokens": 16151409.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.2716185569763185,
|
|
"epoch": 7.348798674399337,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013626921570123357,
|
|
"loss": 4.5301,
|
|
"mean_token_accuracy": 0.22848975360393525,
|
|
"num_tokens": 16160555.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.2927894115448,
|
|
"epoch": 7.352941176470588,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013601777188490156,
|
|
"loss": 4.5426,
|
|
"mean_token_accuracy": 0.22406074404716492,
|
|
"num_tokens": 16169421.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.285395860671997,
|
|
"epoch": 7.357083678541839,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001357666084109115,
|
|
"loss": 4.5883,
|
|
"mean_token_accuracy": 0.2271425649523735,
|
|
"num_tokens": 16178308.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.249518013000488,
|
|
"epoch": 7.36122618061309,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001355157257858875,
|
|
"loss": 4.4965,
|
|
"mean_token_accuracy": 0.22820567041635514,
|
|
"num_tokens": 16187534.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.253032922744751,
|
|
"epoch": 7.365368682684341,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013526512451588718,
|
|
"loss": 4.4799,
|
|
"mean_token_accuracy": 0.23708704113960266,
|
|
"num_tokens": 16196903.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.248826742172241,
|
|
"epoch": 7.369511184755592,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001350148051064008,
|
|
"loss": 4.5131,
|
|
"mean_token_accuracy": 0.22983570843935014,
|
|
"num_tokens": 16205363.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.297190999984741,
|
|
"epoch": 7.373653686826843,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013476476806234983,
|
|
"loss": 4.6289,
|
|
"mean_token_accuracy": 0.21601468175649643,
|
|
"num_tokens": 16215005.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.266012525558471,
|
|
"epoch": 7.3777961888980945,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001345150138880865,
|
|
"loss": 4.5795,
|
|
"mean_token_accuracy": 0.22895101010799407,
|
|
"num_tokens": 16224556.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.212702798843384,
|
|
"epoch": 7.381938690969346,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013426554308739214,
|
|
"loss": 4.4653,
|
|
"mean_token_accuracy": 0.2364416167140007,
|
|
"num_tokens": 16232856.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.2203985214233395,
|
|
"epoch": 7.386081193040597,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00013401635616347655,
|
|
"loss": 4.532,
|
|
"mean_token_accuracy": 0.23128040432929992,
|
|
"num_tokens": 16242099.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.276306438446045,
|
|
"epoch": 7.390223695111848,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013376745361897712,
|
|
"loss": 4.6206,
|
|
"mean_token_accuracy": 0.21572913527488707,
|
|
"num_tokens": 16251617.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.263050699234009,
|
|
"epoch": 7.394366197183099,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013351883595595743,
|
|
"loss": 4.5233,
|
|
"mean_token_accuracy": 0.22834506779909133,
|
|
"num_tokens": 16260126.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.257452392578125,
|
|
"epoch": 7.39850869925435,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001332705036759065,
|
|
"loss": 4.5411,
|
|
"mean_token_accuracy": 0.23098128587007521,
|
|
"num_tokens": 16268844.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.162498855590821,
|
|
"epoch": 7.4026512013256,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001330224572797377,
|
|
"loss": 4.4368,
|
|
"mean_token_accuracy": 0.23918474614620208,
|
|
"num_tokens": 16277645.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.273772048950195,
|
|
"epoch": 7.406793703396851,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013277469726778756,
|
|
"loss": 4.5974,
|
|
"mean_token_accuracy": 0.22762052863836288,
|
|
"num_tokens": 16287265.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.230074310302735,
|
|
"epoch": 7.4109362054681025,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001325272241398153,
|
|
"loss": 4.4596,
|
|
"mean_token_accuracy": 0.2383972242474556,
|
|
"num_tokens": 16295310.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.362758922576904,
|
|
"epoch": 7.415078707539354,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013228003839500113,
|
|
"loss": 4.5933,
|
|
"mean_token_accuracy": 0.21690630316734313,
|
|
"num_tokens": 16303933.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.256103420257569,
|
|
"epoch": 7.419221209610605,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001320331405319458,
|
|
"loss": 4.5619,
|
|
"mean_token_accuracy": 0.2280977860093117,
|
|
"num_tokens": 16312993.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.266288900375367,
|
|
"epoch": 7.423363711681856,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013178653104866926,
|
|
"loss": 4.5353,
|
|
"mean_token_accuracy": 0.2314622789621353,
|
|
"num_tokens": 16322191.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.333379411697388,
|
|
"epoch": 7.427506213753107,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013154021044260966,
|
|
"loss": 4.5202,
|
|
"mean_token_accuracy": 0.2317146047949791,
|
|
"num_tokens": 16330680.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.26063346862793,
|
|
"epoch": 7.431648715824358,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013129417921062275,
|
|
"loss": 4.5304,
|
|
"mean_token_accuracy": 0.2289079800248146,
|
|
"num_tokens": 16339889.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.2644750595092775,
|
|
"epoch": 7.435791217895609,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001310484378489803,
|
|
"loss": 4.5197,
|
|
"mean_token_accuracy": 0.22929859757423401,
|
|
"num_tokens": 16348775.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.272098588943481,
|
|
"epoch": 7.43993371996686,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013080298685336957,
|
|
"loss": 4.5922,
|
|
"mean_token_accuracy": 0.2221032962203026,
|
|
"num_tokens": 16357972.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.292725706100464,
|
|
"epoch": 7.444076222038111,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013055782671889205,
|
|
"loss": 4.6467,
|
|
"mean_token_accuracy": 0.21627185493707657,
|
|
"num_tokens": 16368327.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.268766069412232,
|
|
"epoch": 7.448218724109362,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001303129579400624,
|
|
"loss": 4.4891,
|
|
"mean_token_accuracy": 0.2233644723892212,
|
|
"num_tokens": 16377128.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.185982513427734,
|
|
"epoch": 7.452361226180613,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001300683810108078,
|
|
"loss": 4.4704,
|
|
"mean_token_accuracy": 0.24111500084400178,
|
|
"num_tokens": 16386019.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.260823202133179,
|
|
"epoch": 7.456503728251864,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001298240964244666,
|
|
"loss": 4.4908,
|
|
"mean_token_accuracy": 0.2425772503018379,
|
|
"num_tokens": 16395146.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.456503728251864,
|
|
"eval_entropy": 5.151675126700005,
|
|
"eval_loss": 5.778080940246582,
|
|
"eval_mean_token_accuracy": 0.16587539516023989,
|
|
"eval_num_tokens": 16395146.0,
|
|
"eval_runtime": 2.9731,
|
|
"eval_samples_per_second": 1260.29,
|
|
"eval_steps_per_second": 157.746,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.284982347488404,
|
|
"epoch": 7.460646230323115,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001295801046737876,
|
|
"loss": 4.5915,
|
|
"mean_token_accuracy": 0.22223802357912065,
|
|
"num_tokens": 16404328.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.268152523040771,
|
|
"epoch": 7.464788732394366,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012933640625092867,
|
|
"loss": 4.4968,
|
|
"mean_token_accuracy": 0.2311706244945526,
|
|
"num_tokens": 16414611.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.269545269012451,
|
|
"epoch": 7.468931234465617,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012909300164745636,
|
|
"loss": 4.5406,
|
|
"mean_token_accuracy": 0.23454982787370682,
|
|
"num_tokens": 16423789.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.233197832107544,
|
|
"epoch": 7.473073736536868,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00012884989135434423,
|
|
"loss": 4.5345,
|
|
"mean_token_accuracy": 0.23545969724655152,
|
|
"num_tokens": 16433491.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.207718849182129,
|
|
"epoch": 7.477216238608119,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012860707586197218,
|
|
"loss": 4.4168,
|
|
"mean_token_accuracy": 0.23975736647844315,
|
|
"num_tokens": 16442233.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.265835905075074,
|
|
"epoch": 7.4813587406793705,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012836455566012586,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.22493952810764312,
|
|
"num_tokens": 16450897.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.298757600784302,
|
|
"epoch": 7.485501242750622,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012812233123799481,
|
|
"loss": 4.5611,
|
|
"mean_token_accuracy": 0.22628814578056336,
|
|
"num_tokens": 16459779.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.261304235458374,
|
|
"epoch": 7.489643744821873,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012788040308417232,
|
|
"loss": 4.546,
|
|
"mean_token_accuracy": 0.2214462697505951,
|
|
"num_tokens": 16469309.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.236078357696533,
|
|
"epoch": 7.493786246893124,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012763877168665383,
|
|
"loss": 4.5024,
|
|
"mean_token_accuracy": 0.23161682784557341,
|
|
"num_tokens": 16477391.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.271702575683594,
|
|
"epoch": 7.497928748964375,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012739743753283617,
|
|
"loss": 4.5018,
|
|
"mean_token_accuracy": 0.23550426065921784,
|
|
"num_tokens": 16485992.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.256993341445923,
|
|
"epoch": 7.502071251035625,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001271564011095169,
|
|
"loss": 4.5186,
|
|
"mean_token_accuracy": 0.2326687127351761,
|
|
"num_tokens": 16495260.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.1895592212677,
|
|
"epoch": 7.506213753106876,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001269156629028927,
|
|
"loss": 4.4047,
|
|
"mean_token_accuracy": 0.24649101346731186,
|
|
"num_tokens": 16504045.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.270913600921631,
|
|
"epoch": 7.510356255178127,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001266752233985589,
|
|
"loss": 4.528,
|
|
"mean_token_accuracy": 0.2298363909125328,
|
|
"num_tokens": 16513920.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.258766508102417,
|
|
"epoch": 7.514498757249378,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012643508308150822,
|
|
"loss": 4.5648,
|
|
"mean_token_accuracy": 0.2255310982465744,
|
|
"num_tokens": 16522099.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.28293023109436,
|
|
"epoch": 7.5186412593206295,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001261952424361299,
|
|
"loss": 4.551,
|
|
"mean_token_accuracy": 0.22689804881811143,
|
|
"num_tokens": 16530801.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.230189609527588,
|
|
"epoch": 7.522783761391881,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001259557019462088,
|
|
"loss": 4.5145,
|
|
"mean_token_accuracy": 0.22977649867534639,
|
|
"num_tokens": 16539783.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.263046216964722,
|
|
"epoch": 7.526926263463132,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012571646209492417,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.23383537232875823,
|
|
"num_tokens": 16548614.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.243165016174316,
|
|
"epoch": 7.531068765534383,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00012547752336484908,
|
|
"loss": 4.5292,
|
|
"mean_token_accuracy": 0.2340427115559578,
|
|
"num_tokens": 16558340.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.276821517944336,
|
|
"epoch": 7.535211267605634,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012523888623794892,
|
|
"loss": 4.5473,
|
|
"mean_token_accuracy": 0.22760409563779832,
|
|
"num_tokens": 16566925.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.202182817459106,
|
|
"epoch": 7.539353769676885,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001250005511955809,
|
|
"loss": 4.5116,
|
|
"mean_token_accuracy": 0.23640167117118835,
|
|
"num_tokens": 16576338.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.262622404098511,
|
|
"epoch": 7.543496271748136,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001247625187184929,
|
|
"loss": 4.5558,
|
|
"mean_token_accuracy": 0.22133441120386124,
|
|
"num_tokens": 16585306.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.237573432922363,
|
|
"epoch": 7.547638773819386,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001245247892868223,
|
|
"loss": 4.5697,
|
|
"mean_token_accuracy": 0.22780365943908693,
|
|
"num_tokens": 16593987.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.220071315765381,
|
|
"epoch": 7.551781275890638,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001242873633800955,
|
|
"loss": 4.4834,
|
|
"mean_token_accuracy": 0.23567230254411697,
|
|
"num_tokens": 16602553.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.255688333511353,
|
|
"epoch": 7.555923777961889,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012405024147722643,
|
|
"loss": 4.5174,
|
|
"mean_token_accuracy": 0.22360992878675462,
|
|
"num_tokens": 16611272.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.193259906768799,
|
|
"epoch": 7.56006628003314,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001238134240565158,
|
|
"loss": 4.4836,
|
|
"mean_token_accuracy": 0.23239105343818664,
|
|
"num_tokens": 16620127.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.241860723495483,
|
|
"epoch": 7.564208782104391,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001235769115956504,
|
|
"loss": 4.5844,
|
|
"mean_token_accuracy": 0.2254987731575966,
|
|
"num_tokens": 16630069.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.340811824798584,
|
|
"epoch": 7.568351284175642,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012334070457170151,
|
|
"loss": 4.6789,
|
|
"mean_token_accuracy": 0.2234554573893547,
|
|
"num_tokens": 16639102.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.211176061630249,
|
|
"epoch": 7.572493786246893,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012310480346112472,
|
|
"loss": 4.4661,
|
|
"mean_token_accuracy": 0.23987812846899031,
|
|
"num_tokens": 16647859.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.269927167892456,
|
|
"epoch": 7.576636288318144,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012286920873975815,
|
|
"loss": 4.5604,
|
|
"mean_token_accuracy": 0.23633040636777877,
|
|
"num_tokens": 16657116.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.262793159484863,
|
|
"epoch": 7.580778790389395,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012263392088282226,
|
|
"loss": 4.6048,
|
|
"mean_token_accuracy": 0.21968501657247544,
|
|
"num_tokens": 16666470.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.23599328994751,
|
|
"epoch": 7.584921292460646,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012239894036491833,
|
|
"loss": 4.5055,
|
|
"mean_token_accuracy": 0.2266399919986725,
|
|
"num_tokens": 16675784.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.236028623580933,
|
|
"epoch": 7.5890637945318975,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012216426766002762,
|
|
"loss": 4.5265,
|
|
"mean_token_accuracy": 0.22461990118026734,
|
|
"num_tokens": 16685438.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.291051006317138,
|
|
"epoch": 7.593206296603149,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012192990324151079,
|
|
"loss": 4.5497,
|
|
"mean_token_accuracy": 0.22482420802116393,
|
|
"num_tokens": 16694834.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.260760736465454,
|
|
"epoch": 7.5973487986744,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00012169584758210635,
|
|
"loss": 4.5717,
|
|
"mean_token_accuracy": 0.22383186221122742,
|
|
"num_tokens": 16703996.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.2278906345367435,
|
|
"epoch": 7.60149130074565,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012146210115393025,
|
|
"loss": 4.5446,
|
|
"mean_token_accuracy": 0.22601110786199569,
|
|
"num_tokens": 16713674.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.282411336898804,
|
|
"epoch": 7.605633802816901,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012122866442847455,
|
|
"loss": 4.561,
|
|
"mean_token_accuracy": 0.22372008115053177,
|
|
"num_tokens": 16723783.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.251124858856201,
|
|
"epoch": 7.609776304888152,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012099553787660654,
|
|
"loss": 4.4917,
|
|
"mean_token_accuracy": 0.23510961681604386,
|
|
"num_tokens": 16732478.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.2333142280578615,
|
|
"epoch": 7.613918806959403,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012076272196856808,
|
|
"loss": 4.4851,
|
|
"mean_token_accuracy": 0.2298592820763588,
|
|
"num_tokens": 16741269.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.2707812786102295,
|
|
"epoch": 7.618061309030654,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001205302171739742,
|
|
"loss": 4.59,
|
|
"mean_token_accuracy": 0.2280466228723526,
|
|
"num_tokens": 16751214.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.288453769683838,
|
|
"epoch": 7.6222038111019055,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012029802396181256,
|
|
"loss": 4.628,
|
|
"mean_token_accuracy": 0.2185559540987015,
|
|
"num_tokens": 16760563.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.334579372406006,
|
|
"epoch": 7.626346313173157,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012006614280044225,
|
|
"loss": 4.6549,
|
|
"mean_token_accuracy": 0.22408310025930406,
|
|
"num_tokens": 16769931.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.260780477523804,
|
|
"epoch": 7.630488815244408,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011983457415759274,
|
|
"loss": 4.6227,
|
|
"mean_token_accuracy": 0.22169588655233383,
|
|
"num_tokens": 16779748.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.2880706787109375,
|
|
"epoch": 7.634631317315659,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011960331850036352,
|
|
"loss": 4.5296,
|
|
"mean_token_accuracy": 0.2334790989756584,
|
|
"num_tokens": 16789999.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.283678340911865,
|
|
"epoch": 7.63877381938691,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011937237629522233,
|
|
"loss": 4.5239,
|
|
"mean_token_accuracy": 0.23558811396360396,
|
|
"num_tokens": 16799990.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.249262809753418,
|
|
"epoch": 7.642916321458161,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011914174800800501,
|
|
"loss": 4.5085,
|
|
"mean_token_accuracy": 0.24060285687446595,
|
|
"num_tokens": 16808677.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.250159120559692,
|
|
"epoch": 7.647058823529412,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011891143410391387,
|
|
"loss": 4.5037,
|
|
"mean_token_accuracy": 0.22584034204483033,
|
|
"num_tokens": 16817040.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.226622533798218,
|
|
"epoch": 7.651201325600663,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011868143504751732,
|
|
"loss": 4.4778,
|
|
"mean_token_accuracy": 0.23347067832946777,
|
|
"num_tokens": 16825138.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.279749011993408,
|
|
"epoch": 7.6553438276719135,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011845175130274848,
|
|
"loss": 4.48,
|
|
"mean_token_accuracy": 0.23995605558156968,
|
|
"num_tokens": 16834663.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.204747247695923,
|
|
"epoch": 7.659486329743165,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011822238333290475,
|
|
"loss": 4.3876,
|
|
"mean_token_accuracy": 0.2502914160490036,
|
|
"num_tokens": 16842777.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.210317325592041,
|
|
"epoch": 7.663628831814416,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011799333160064617,
|
|
"loss": 4.5538,
|
|
"mean_token_accuracy": 0.2226479694247246,
|
|
"num_tokens": 16852305.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.141436290740967,
|
|
"epoch": 7.667771333885667,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001177645965679953,
|
|
"loss": 4.4989,
|
|
"mean_token_accuracy": 0.2285989060997963,
|
|
"num_tokens": 16861736.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.210664892196656,
|
|
"epoch": 7.671913835956918,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011753617869633561,
|
|
"loss": 4.5301,
|
|
"mean_token_accuracy": 0.22809041440486907,
|
|
"num_tokens": 16871098.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.180535745620728,
|
|
"epoch": 7.676056338028169,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011730807844641085,
|
|
"loss": 4.478,
|
|
"mean_token_accuracy": 0.23257430642843246,
|
|
"num_tokens": 16879477.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.266501426696777,
|
|
"epoch": 7.68019884009942,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011708029627832431,
|
|
"loss": 4.5607,
|
|
"mean_token_accuracy": 0.23377658426761627,
|
|
"num_tokens": 16888735.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.291167449951172,
|
|
"epoch": 7.684341342170671,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011685283265153732,
|
|
"loss": 4.5221,
|
|
"mean_token_accuracy": 0.22680869698524475,
|
|
"num_tokens": 16897628.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.259503602981567,
|
|
"epoch": 7.688483844241922,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011662568802486905,
|
|
"loss": 4.5757,
|
|
"mean_token_accuracy": 0.2234109088778496,
|
|
"num_tokens": 16906771.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.291567659378051,
|
|
"epoch": 7.692626346313173,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011639886285649489,
|
|
"loss": 4.5683,
|
|
"mean_token_accuracy": 0.227107672393322,
|
|
"num_tokens": 16916446.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.269523334503174,
|
|
"epoch": 7.6967688483844245,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011617235760394614,
|
|
"loss": 4.4721,
|
|
"mean_token_accuracy": 0.23378842175006867,
|
|
"num_tokens": 16925941.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.232016849517822,
|
|
"epoch": 7.700911350455675,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001159461727241085,
|
|
"loss": 4.5307,
|
|
"mean_token_accuracy": 0.23323894441127777,
|
|
"num_tokens": 16935142.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.323593044281006,
|
|
"epoch": 7.705053852526927,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001157203086732217,
|
|
"loss": 4.6096,
|
|
"mean_token_accuracy": 0.22642216980457305,
|
|
"num_tokens": 16944737.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.296499347686767,
|
|
"epoch": 7.709196354598177,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011549476590687811,
|
|
"loss": 4.5383,
|
|
"mean_token_accuracy": 0.22358610928058625,
|
|
"num_tokens": 16953909.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.341109800338745,
|
|
"epoch": 7.713338856669428,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011526954488002227,
|
|
"loss": 4.5903,
|
|
"mean_token_accuracy": 0.2233624666929245,
|
|
"num_tokens": 16963282.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.248440980911255,
|
|
"epoch": 7.717481358740679,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011504464604694953,
|
|
"loss": 4.4604,
|
|
"mean_token_accuracy": 0.23559881001710892,
|
|
"num_tokens": 16970908.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.248945999145508,
|
|
"epoch": 7.72162386081193,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011482006986130533,
|
|
"loss": 4.4744,
|
|
"mean_token_accuracy": 0.22936770915985108,
|
|
"num_tokens": 16979209.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.253082036972046,
|
|
"epoch": 7.725766362883181,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011459581677608453,
|
|
"loss": 4.5604,
|
|
"mean_token_accuracy": 0.22624849677085876,
|
|
"num_tokens": 16988352.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.224340772628784,
|
|
"epoch": 7.7299088649544325,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011437188724362997,
|
|
"loss": 4.5016,
|
|
"mean_token_accuracy": 0.2345532402396202,
|
|
"num_tokens": 16997398.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.251537227630616,
|
|
"epoch": 7.734051367025684,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011414828171563214,
|
|
"loss": 4.5146,
|
|
"mean_token_accuracy": 0.22609912604093552,
|
|
"num_tokens": 17006641.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.262172222137451,
|
|
"epoch": 7.738193869096935,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011392500064312773,
|
|
"loss": 4.6004,
|
|
"mean_token_accuracy": 0.21589517444372178,
|
|
"num_tokens": 17015879.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.240403747558593,
|
|
"epoch": 7.742336371168186,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000113702044476499,
|
|
"loss": 4.5727,
|
|
"mean_token_accuracy": 0.22586589008569719,
|
|
"num_tokens": 17025305.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.309038305282593,
|
|
"epoch": 7.746478873239437,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011347941366547303,
|
|
"loss": 4.5329,
|
|
"mean_token_accuracy": 0.22893053144216538,
|
|
"num_tokens": 17034440.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.284827804565429,
|
|
"epoch": 7.750621375310688,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001132571086591204,
|
|
"loss": 4.555,
|
|
"mean_token_accuracy": 0.22764674425125123,
|
|
"num_tokens": 17044118.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.271107816696167,
|
|
"epoch": 7.754763877381938,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001130351299058547,
|
|
"loss": 4.5809,
|
|
"mean_token_accuracy": 0.22725985944271088,
|
|
"num_tokens": 17052693.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.314342641830445,
|
|
"epoch": 7.758906379453189,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011281347785343127,
|
|
"loss": 4.609,
|
|
"mean_token_accuracy": 0.2260008841753006,
|
|
"num_tokens": 17062106.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.27345290184021,
|
|
"epoch": 7.7630488815244405,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011259215294894647,
|
|
"loss": 4.5357,
|
|
"mean_token_accuracy": 0.23755454570055007,
|
|
"num_tokens": 17071709.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.232308578491211,
|
|
"epoch": 7.767191383595692,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011237115563883693,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.23762645423412324,
|
|
"num_tokens": 17080606.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.257081747055054,
|
|
"epoch": 7.771333885666943,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011215048636887825,
|
|
"loss": 4.5176,
|
|
"mean_token_accuracy": 0.2282841295003891,
|
|
"num_tokens": 17089696.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.2346333980560305,
|
|
"epoch": 7.775476387738194,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001119301455841846,
|
|
"loss": 4.4544,
|
|
"mean_token_accuracy": 0.2354099839925766,
|
|
"num_tokens": 17099175.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.2944286346435545,
|
|
"epoch": 7.779618889809445,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011171013372920735,
|
|
"loss": 4.6027,
|
|
"mean_token_accuracy": 0.22485320270061493,
|
|
"num_tokens": 17108299.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.302739095687866,
|
|
"epoch": 7.783761391880696,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011149045124773441,
|
|
"loss": 4.5746,
|
|
"mean_token_accuracy": 0.22643714398145676,
|
|
"num_tokens": 17117107.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.231918096542358,
|
|
"epoch": 7.787903893951947,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011127109858288945,
|
|
"loss": 4.5572,
|
|
"mean_token_accuracy": 0.22558478862047196,
|
|
"num_tokens": 17127175.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.248330736160279,
|
|
"epoch": 7.792046396023198,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011105207617713065,
|
|
"loss": 4.5463,
|
|
"mean_token_accuracy": 0.22825556695461274,
|
|
"num_tokens": 17136283.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.230691814422608,
|
|
"epoch": 7.796188898094449,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011083338447225027,
|
|
"loss": 4.5215,
|
|
"mean_token_accuracy": 0.233886282145977,
|
|
"num_tokens": 17145762.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.298114252090454,
|
|
"epoch": 7.8003314001657,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011061502390937331,
|
|
"loss": 4.5539,
|
|
"mean_token_accuracy": 0.22342685461044312,
|
|
"num_tokens": 17154756.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.286074161529541,
|
|
"epoch": 7.804473902236952,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001103969949289568,
|
|
"loss": 4.5252,
|
|
"mean_token_accuracy": 0.23229531049728394,
|
|
"num_tokens": 17164089.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.2408825874328615,
|
|
"epoch": 7.808616404308202,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011017929797078919,
|
|
"loss": 4.4885,
|
|
"mean_token_accuracy": 0.23738452345132827,
|
|
"num_tokens": 17173126.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.241729116439819,
|
|
"epoch": 7.812758906379453,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010996193347398886,
|
|
"loss": 4.5291,
|
|
"mean_token_accuracy": 0.2290113627910614,
|
|
"num_tokens": 17182698.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.176688575744629,
|
|
"epoch": 7.816901408450704,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010974490187700394,
|
|
"loss": 4.4342,
|
|
"mean_token_accuracy": 0.23970996588468552,
|
|
"num_tokens": 17191062.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.262580680847168,
|
|
"epoch": 7.821043910521955,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001095282036176107,
|
|
"loss": 4.579,
|
|
"mean_token_accuracy": 0.22438587099313737,
|
|
"num_tokens": 17200330.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.18640718460083,
|
|
"epoch": 7.825186412593206,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010931183913291335,
|
|
"loss": 4.4937,
|
|
"mean_token_accuracy": 0.23648798018693923,
|
|
"num_tokens": 17209761.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.24940333366394,
|
|
"epoch": 7.829328914664457,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010909580885934265,
|
|
"loss": 4.5173,
|
|
"mean_token_accuracy": 0.23395260274410248,
|
|
"num_tokens": 17219378.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.26439208984375,
|
|
"epoch": 7.8334714167357085,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010888011323265517,
|
|
"loss": 4.5631,
|
|
"mean_token_accuracy": 0.2271890327334404,
|
|
"num_tokens": 17230286.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.207281398773193,
|
|
"epoch": 7.83761391880696,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00010866475268793275,
|
|
"loss": 4.4447,
|
|
"mean_token_accuracy": 0.2451467588543892,
|
|
"num_tokens": 17238115.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.297142457962036,
|
|
"epoch": 7.841756420878211,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00010844972765958096,
|
|
"loss": 4.5687,
|
|
"mean_token_accuracy": 0.23140267729759217,
|
|
"num_tokens": 17247041.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.222435760498047,
|
|
"epoch": 7.845898922949462,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010823503858132894,
|
|
"loss": 4.4908,
|
|
"mean_token_accuracy": 0.23347440958023072,
|
|
"num_tokens": 17256012.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.303977155685425,
|
|
"epoch": 7.850041425020713,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010802068588622793,
|
|
"loss": 4.657,
|
|
"mean_token_accuracy": 0.2124525472521782,
|
|
"num_tokens": 17266054.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.28911862373352,
|
|
"epoch": 7.854183927091963,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010780667000665067,
|
|
"loss": 4.5213,
|
|
"mean_token_accuracy": 0.223433718085289,
|
|
"num_tokens": 17276003.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.2641688823699955,
|
|
"epoch": 7.858326429163214,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010759299137429073,
|
|
"loss": 4.508,
|
|
"mean_token_accuracy": 0.23626709282398223,
|
|
"num_tokens": 17285912.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.281970071792602,
|
|
"epoch": 7.862468931234465,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010737965042016116,
|
|
"loss": 4.5978,
|
|
"mean_token_accuracy": 0.22393255531787873,
|
|
"num_tokens": 17295136.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.2496092319488525,
|
|
"epoch": 7.866611433305716,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010716664757459404,
|
|
"loss": 4.4841,
|
|
"mean_token_accuracy": 0.2225743427872658,
|
|
"num_tokens": 17304063.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.251446104049682,
|
|
"epoch": 7.8707539353769675,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010695398326723938,
|
|
"loss": 4.5164,
|
|
"mean_token_accuracy": 0.2325739488005638,
|
|
"num_tokens": 17313061.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 7.8707539353769675,
|
|
"eval_entropy": 5.1662215393489355,
|
|
"eval_loss": 5.76571798324585,
|
|
"eval_mean_token_accuracy": 0.16708771126674438,
|
|
"eval_num_tokens": 17313061.0,
|
|
"eval_runtime": 2.7701,
|
|
"eval_samples_per_second": 1352.675,
|
|
"eval_steps_per_second": 169.31,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.263190269470215,
|
|
"epoch": 7.874896437448219,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010674165792706423,
|
|
"loss": 4.5111,
|
|
"mean_token_accuracy": 0.2285247877240181,
|
|
"num_tokens": 17322222.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.269113779067993,
|
|
"epoch": 7.87903893951947,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010652967198235214,
|
|
"loss": 4.5866,
|
|
"mean_token_accuracy": 0.22473605424165727,
|
|
"num_tokens": 17331253.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.235748815536499,
|
|
"epoch": 7.883181441590721,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001063180258607018,
|
|
"loss": 4.6057,
|
|
"mean_token_accuracy": 0.22360503375530244,
|
|
"num_tokens": 17341462.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.271676921844483,
|
|
"epoch": 7.887323943661972,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010610671998902671,
|
|
"loss": 4.5953,
|
|
"mean_token_accuracy": 0.21584573835134507,
|
|
"num_tokens": 17351056.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.196175527572632,
|
|
"epoch": 7.891466445733223,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010589575479355379,
|
|
"loss": 4.5159,
|
|
"mean_token_accuracy": 0.233812215924263,
|
|
"num_tokens": 17360123.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.269428730010986,
|
|
"epoch": 7.895608947804474,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00010568513069982289,
|
|
"loss": 4.6018,
|
|
"mean_token_accuracy": 0.22042047828435898,
|
|
"num_tokens": 17369379.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.299595785140991,
|
|
"epoch": 7.899751449875725,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010547484813268594,
|
|
"loss": 4.5964,
|
|
"mean_token_accuracy": 0.22401643991470338,
|
|
"num_tokens": 17379014.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.214079999923706,
|
|
"epoch": 7.903893951946976,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00010526490751630568,
|
|
"loss": 4.4441,
|
|
"mean_token_accuracy": 0.2417148694396019,
|
|
"num_tokens": 17388388.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.284358549118042,
|
|
"epoch": 7.908036454018227,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010505530927415544,
|
|
"loss": 4.5042,
|
|
"mean_token_accuracy": 0.23135959804058076,
|
|
"num_tokens": 17397112.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.206902551651001,
|
|
"epoch": 7.912178956089478,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010484605382901772,
|
|
"loss": 4.5163,
|
|
"mean_token_accuracy": 0.23637685179710388,
|
|
"num_tokens": 17406166.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.269112777709961,
|
|
"epoch": 7.916321458160729,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010463714160298354,
|
|
"loss": 4.5608,
|
|
"mean_token_accuracy": 0.23420423418283462,
|
|
"num_tokens": 17415357.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.315996837615967,
|
|
"epoch": 7.92046396023198,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00010442857301745184,
|
|
"loss": 4.7019,
|
|
"mean_token_accuracy": 0.21716914027929307,
|
|
"num_tokens": 17424853.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.371943521499634,
|
|
"epoch": 7.924606462303231,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001042203484931281,
|
|
"loss": 4.6447,
|
|
"mean_token_accuracy": 0.226338754594326,
|
|
"num_tokens": 17433941.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.306926918029785,
|
|
"epoch": 7.928748964374482,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010401246845002407,
|
|
"loss": 4.5618,
|
|
"mean_token_accuracy": 0.22942230701446534,
|
|
"num_tokens": 17442733.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.26545147895813,
|
|
"epoch": 7.932891466445733,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010380493330745641,
|
|
"loss": 4.4774,
|
|
"mean_token_accuracy": 0.23628817200660707,
|
|
"num_tokens": 17452295.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.288152265548706,
|
|
"epoch": 7.937033968516984,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010359774348404626,
|
|
"loss": 4.5434,
|
|
"mean_token_accuracy": 0.22995532155036927,
|
|
"num_tokens": 17461646.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.258202791213989,
|
|
"epoch": 7.9411764705882355,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010339089939771814,
|
|
"loss": 4.5201,
|
|
"mean_token_accuracy": 0.23450240790843963,
|
|
"num_tokens": 17470028.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.315814352035522,
|
|
"epoch": 7.945318972659487,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010318440146569904,
|
|
"loss": 4.5472,
|
|
"mean_token_accuracy": 0.22288650572299956,
|
|
"num_tokens": 17479815.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.344705724716187,
|
|
"epoch": 7.949461474730738,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010297825010451806,
|
|
"loss": 4.6244,
|
|
"mean_token_accuracy": 0.22105477005243301,
|
|
"num_tokens": 17489733.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.291326951980591,
|
|
"epoch": 7.953603976801988,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001027724457300048,
|
|
"loss": 4.5715,
|
|
"mean_token_accuracy": 0.2292417526245117,
|
|
"num_tokens": 17498921.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.19606728553772,
|
|
"epoch": 7.957746478873239,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010256698875728938,
|
|
"loss": 4.4613,
|
|
"mean_token_accuracy": 0.23985940963029861,
|
|
"num_tokens": 17508285.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.203891849517822,
|
|
"epoch": 7.96188898094449,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010236187960080084,
|
|
"loss": 4.4962,
|
|
"mean_token_accuracy": 0.22880738824605942,
|
|
"num_tokens": 17517322.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.263271903991699,
|
|
"epoch": 7.966031483015741,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010215711867426678,
|
|
"loss": 4.5009,
|
|
"mean_token_accuracy": 0.22785776555538179,
|
|
"num_tokens": 17525445.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.317650508880615,
|
|
"epoch": 7.970173985086992,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010195270639071243,
|
|
"loss": 4.6146,
|
|
"mean_token_accuracy": 0.22779650390148162,
|
|
"num_tokens": 17534638.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.269761419296264,
|
|
"epoch": 7.9743164871582435,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001017486431624596,
|
|
"loss": 4.5721,
|
|
"mean_token_accuracy": 0.22240918427705764,
|
|
"num_tokens": 17545367.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.267766857147217,
|
|
"epoch": 7.978458989229495,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010154492940112624,
|
|
"loss": 4.5566,
|
|
"mean_token_accuracy": 0.23049682080745698,
|
|
"num_tokens": 17553863.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.255144929885864,
|
|
"epoch": 7.982601491300746,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010134156551762519,
|
|
"loss": 4.5865,
|
|
"mean_token_accuracy": 0.22993125021457672,
|
|
"num_tokens": 17562773.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.200955200195312,
|
|
"epoch": 7.986743993371997,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010113855192216359,
|
|
"loss": 4.4494,
|
|
"mean_token_accuracy": 0.24445375055074692,
|
|
"num_tokens": 17571634.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.206564140319824,
|
|
"epoch": 7.990886495443248,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010093588902424217,
|
|
"loss": 4.517,
|
|
"mean_token_accuracy": 0.22749946415424346,
|
|
"num_tokens": 17580749.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.216109991073608,
|
|
"epoch": 7.995028997514499,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010073357723265398,
|
|
"loss": 4.5036,
|
|
"mean_token_accuracy": 0.2284680113196373,
|
|
"num_tokens": 17589402.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.225698709487915,
|
|
"epoch": 7.99917149958575,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010053161695548418,
|
|
"loss": 4.5994,
|
|
"mean_token_accuracy": 0.2174151510000229,
|
|
"num_tokens": 17598929.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.242027235031128,
|
|
"epoch": 8.003314001657001,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010033000860010866,
|
|
"loss": 4.4802,
|
|
"mean_token_accuracy": 0.23776553869247435,
|
|
"num_tokens": 17607779.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.225143814086914,
|
|
"epoch": 8.007456503728251,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001001287525731934,
|
|
"loss": 4.445,
|
|
"mean_token_accuracy": 0.2415912389755249,
|
|
"num_tokens": 17617031.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.2578761100769045,
|
|
"epoch": 8.011599005799503,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.992784928069397e-05,
|
|
"loss": 4.457,
|
|
"mean_token_accuracy": 0.2254150003194809,
|
|
"num_tokens": 17625643.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.199114656448364,
|
|
"epoch": 8.015741507870754,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.972729912785411e-05,
|
|
"loss": 4.4091,
|
|
"mean_token_accuracy": 0.2443189039826393,
|
|
"num_tokens": 17634821.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.217359018325806,
|
|
"epoch": 8.019884009942006,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.95271025192055e-05,
|
|
"loss": 4.4268,
|
|
"mean_token_accuracy": 0.23860696256160735,
|
|
"num_tokens": 17643457.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.23916392326355,
|
|
"epoch": 8.024026512013256,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.93272598585665e-05,
|
|
"loss": 4.4233,
|
|
"mean_token_accuracy": 0.24197322130203247,
|
|
"num_tokens": 17651441.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.241343545913696,
|
|
"epoch": 8.028169014084508,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.912777154904171e-05,
|
|
"loss": 4.4591,
|
|
"mean_token_accuracy": 0.2312661737203598,
|
|
"num_tokens": 17660911.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.296669435501099,
|
|
"epoch": 8.032311516155758,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.89286379930207e-05,
|
|
"loss": 4.4931,
|
|
"mean_token_accuracy": 0.23905631750822068,
|
|
"num_tokens": 17670253.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.24735279083252,
|
|
"epoch": 8.036454018227008,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.872985959217779e-05,
|
|
"loss": 4.4841,
|
|
"mean_token_accuracy": 0.23937708437442778,
|
|
"num_tokens": 17680369.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.175539922714234,
|
|
"epoch": 8.04059652029826,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.853143674747053e-05,
|
|
"loss": 4.334,
|
|
"mean_token_accuracy": 0.24430516958236695,
|
|
"num_tokens": 17688234.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.22845983505249,
|
|
"epoch": 8.04473902236951,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.833336985913965e-05,
|
|
"loss": 4.4226,
|
|
"mean_token_accuracy": 0.24006450027227402,
|
|
"num_tokens": 17697569.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.26625337600708,
|
|
"epoch": 8.048881524440763,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.813565932670767e-05,
|
|
"loss": 4.4876,
|
|
"mean_token_accuracy": 0.23214493244886397,
|
|
"num_tokens": 17706672.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.178779125213623,
|
|
"epoch": 8.053024026512013,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.793830554897826e-05,
|
|
"loss": 4.4225,
|
|
"mean_token_accuracy": 0.23367810398340225,
|
|
"num_tokens": 17716033.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.339380264282227,
|
|
"epoch": 8.057166528583265,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.774130892403566e-05,
|
|
"loss": 4.5836,
|
|
"mean_token_accuracy": 0.22804645746946334,
|
|
"num_tokens": 17725984.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.253745794296265,
|
|
"epoch": 8.061309030654515,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.754466984924348e-05,
|
|
"loss": 4.5106,
|
|
"mean_token_accuracy": 0.2278538629412651,
|
|
"num_tokens": 17735878.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.282870054244995,
|
|
"epoch": 8.065451532725767,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.734838872124432e-05,
|
|
"loss": 4.4823,
|
|
"mean_token_accuracy": 0.23122033774852752,
|
|
"num_tokens": 17744910.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.273496770858765,
|
|
"epoch": 8.069594034797017,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.715246593595859e-05,
|
|
"loss": 4.5041,
|
|
"mean_token_accuracy": 0.2354026436805725,
|
|
"num_tokens": 17753907.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.310891914367676,
|
|
"epoch": 8.07373653686827,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.695690188858408e-05,
|
|
"loss": 4.4691,
|
|
"mean_token_accuracy": 0.23567755669355392,
|
|
"num_tokens": 17762913.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.247167301177979,
|
|
"epoch": 8.07787903893952,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.676169697359469e-05,
|
|
"loss": 4.4001,
|
|
"mean_token_accuracy": 0.2403183862566948,
|
|
"num_tokens": 17771262.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.278298044204712,
|
|
"epoch": 8.08202154101077,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.656685158474018e-05,
|
|
"loss": 4.4662,
|
|
"mean_token_accuracy": 0.23878922164440156,
|
|
"num_tokens": 17780200.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.234626770019531,
|
|
"epoch": 8.086164043082022,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.63723661150449e-05,
|
|
"loss": 4.3967,
|
|
"mean_token_accuracy": 0.2406259760260582,
|
|
"num_tokens": 17788838.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.257783079147339,
|
|
"epoch": 8.090306545153272,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.617824095680742e-05,
|
|
"loss": 4.4409,
|
|
"mean_token_accuracy": 0.23529001623392104,
|
|
"num_tokens": 17797928.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.212325382232666,
|
|
"epoch": 8.094449047224524,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.598447650159931e-05,
|
|
"loss": 4.4608,
|
|
"mean_token_accuracy": 0.2285444512963295,
|
|
"num_tokens": 17807094.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.241442012786865,
|
|
"epoch": 8.098591549295774,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.579107314026459e-05,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.22618237435817717,
|
|
"num_tokens": 17815965.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.227851343154907,
|
|
"epoch": 8.102734051367026,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.559803126291911e-05,
|
|
"loss": 4.5232,
|
|
"mean_token_accuracy": 0.22967969626188278,
|
|
"num_tokens": 17825504.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.184619379043579,
|
|
"epoch": 8.106876553438276,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.540535125894931e-05,
|
|
"loss": 4.4427,
|
|
"mean_token_accuracy": 0.2376345068216324,
|
|
"num_tokens": 17834175.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.288371229171753,
|
|
"epoch": 8.111019055509528,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.521303351701193e-05,
|
|
"loss": 4.5241,
|
|
"mean_token_accuracy": 0.23711019158363342,
|
|
"num_tokens": 17843622.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.284456443786621,
|
|
"epoch": 8.115161557580779,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.502107842503282e-05,
|
|
"loss": 4.4431,
|
|
"mean_token_accuracy": 0.241398486495018,
|
|
"num_tokens": 17852297.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.221557760238648,
|
|
"epoch": 8.11930405965203,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.48294863702063e-05,
|
|
"loss": 4.4163,
|
|
"mean_token_accuracy": 0.23962014466524123,
|
|
"num_tokens": 17860833.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.289076042175293,
|
|
"epoch": 8.12344656172328,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.463825773899465e-05,
|
|
"loss": 4.5144,
|
|
"mean_token_accuracy": 0.22108684629201888,
|
|
"num_tokens": 17869234.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.329522132873535,
|
|
"epoch": 8.127589063794533,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.444739291712673e-05,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.23654535561800002,
|
|
"num_tokens": 17878672.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.239313125610352,
|
|
"epoch": 8.131731565865783,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.425689228959791e-05,
|
|
"loss": 4.4063,
|
|
"mean_token_accuracy": 0.24596261829137803,
|
|
"num_tokens": 17887509.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.217898321151734,
|
|
"epoch": 8.135874067937033,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.406675624066868e-05,
|
|
"loss": 4.3834,
|
|
"mean_token_accuracy": 0.2421853721141815,
|
|
"num_tokens": 17896194.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.2356531620025635,
|
|
"epoch": 8.140016570008285,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.38769851538642e-05,
|
|
"loss": 4.4782,
|
|
"mean_token_accuracy": 0.23908120393753052,
|
|
"num_tokens": 17905647.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.208623266220092,
|
|
"epoch": 8.144159072079535,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.368757941197356e-05,
|
|
"loss": 4.399,
|
|
"mean_token_accuracy": 0.23890700042247773,
|
|
"num_tokens": 17915371.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.322758579254151,
|
|
"epoch": 8.148301574150787,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.34985393970488e-05,
|
|
"loss": 4.5077,
|
|
"mean_token_accuracy": 0.2305398941040039,
|
|
"num_tokens": 17924101.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.243159389495849,
|
|
"epoch": 8.152444076222038,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.330986549040432e-05,
|
|
"loss": 4.39,
|
|
"mean_token_accuracy": 0.24642235189676284,
|
|
"num_tokens": 17933264.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.227045965194702,
|
|
"epoch": 8.15658657829329,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.312155807261602e-05,
|
|
"loss": 4.5094,
|
|
"mean_token_accuracy": 0.23021458983421325,
|
|
"num_tokens": 17941913.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.284433507919312,
|
|
"epoch": 8.16072908036454,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.293361752352045e-05,
|
|
"loss": 4.4545,
|
|
"mean_token_accuracy": 0.23619781136512757,
|
|
"num_tokens": 17951637.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.299141693115234,
|
|
"epoch": 8.164871582435792,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.274604422221433e-05,
|
|
"loss": 4.5189,
|
|
"mean_token_accuracy": 0.2278443843126297,
|
|
"num_tokens": 17960783.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.227235269546509,
|
|
"epoch": 8.169014084507042,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.255883854705344e-05,
|
|
"loss": 4.4582,
|
|
"mean_token_accuracy": 0.24150545746088029,
|
|
"num_tokens": 17969984.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.199880790710449,
|
|
"epoch": 8.173156586578294,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.237200087565217e-05,
|
|
"loss": 4.3996,
|
|
"mean_token_accuracy": 0.24653828144073486,
|
|
"num_tokens": 17977769.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.305277299880982,
|
|
"epoch": 8.177299088649544,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.218553158488238e-05,
|
|
"loss": 4.5129,
|
|
"mean_token_accuracy": 0.2307430699467659,
|
|
"num_tokens": 17988041.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.172802877426148,
|
|
"epoch": 8.181441590720794,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.199943105087319e-05,
|
|
"loss": 4.3536,
|
|
"mean_token_accuracy": 0.24522697925567627,
|
|
"num_tokens": 17996619.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.257208919525146,
|
|
"epoch": 8.185584092792046,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.181369964900957e-05,
|
|
"loss": 4.4455,
|
|
"mean_token_accuracy": 0.23698629289865494,
|
|
"num_tokens": 18004678.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.293122911453247,
|
|
"epoch": 8.189726594863297,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.162833775393209e-05,
|
|
"loss": 4.5493,
|
|
"mean_token_accuracy": 0.23128005713224412,
|
|
"num_tokens": 18013763.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.304418706893921,
|
|
"epoch": 8.193869096934549,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.1443345739536e-05,
|
|
"loss": 4.502,
|
|
"mean_token_accuracy": 0.23888981938362122,
|
|
"num_tokens": 18022875.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.244324159622193,
|
|
"epoch": 8.198011599005799,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.12587239789703e-05,
|
|
"loss": 4.4506,
|
|
"mean_token_accuracy": 0.2375013306736946,
|
|
"num_tokens": 18032229.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.308739948272705,
|
|
"epoch": 8.202154101077051,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.10744728446374e-05,
|
|
"loss": 4.5003,
|
|
"mean_token_accuracy": 0.23199772834777832,
|
|
"num_tokens": 18041294.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.314047479629517,
|
|
"epoch": 8.206296603148301,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.089059270819187e-05,
|
|
"loss": 4.4528,
|
|
"mean_token_accuracy": 0.23331775814294814,
|
|
"num_tokens": 18050156.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.253266668319702,
|
|
"epoch": 8.210439105219553,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.070708394054005e-05,
|
|
"loss": 4.4488,
|
|
"mean_token_accuracy": 0.2378482222557068,
|
|
"num_tokens": 18059620.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.248688697814941,
|
|
"epoch": 8.214581607290803,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.052394691183921e-05,
|
|
"loss": 4.4763,
|
|
"mean_token_accuracy": 0.2308114066720009,
|
|
"num_tokens": 18067839.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.2583513259887695,
|
|
"epoch": 8.218724109362055,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.03411819914967e-05,
|
|
"loss": 4.4712,
|
|
"mean_token_accuracy": 0.24067655056715012,
|
|
"num_tokens": 18076366.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.259646701812744,
|
|
"epoch": 8.222866611433306,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.015878954816941e-05,
|
|
"loss": 4.5377,
|
|
"mean_token_accuracy": 0.23090635240077972,
|
|
"num_tokens": 18085724.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.352771282196045,
|
|
"epoch": 8.227009113504558,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.997676994976281e-05,
|
|
"loss": 4.5901,
|
|
"mean_token_accuracy": 0.2312619462609291,
|
|
"num_tokens": 18095565.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.274063444137573,
|
|
"epoch": 8.231151615575808,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.979512356343023e-05,
|
|
"loss": 4.508,
|
|
"mean_token_accuracy": 0.2383059471845627,
|
|
"num_tokens": 18105314.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.226291990280151,
|
|
"epoch": 8.235294117647058,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.961385075557243e-05,
|
|
"loss": 4.3963,
|
|
"mean_token_accuracy": 0.24217721521854402,
|
|
"num_tokens": 18114792.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.206581449508667,
|
|
"epoch": 8.23943661971831,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.943295189183632e-05,
|
|
"loss": 4.3825,
|
|
"mean_token_accuracy": 0.2441452831029892,
|
|
"num_tokens": 18124090.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.254038333892822,
|
|
"epoch": 8.24357912178956,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.925242733711484e-05,
|
|
"loss": 4.4192,
|
|
"mean_token_accuracy": 0.2368018239736557,
|
|
"num_tokens": 18132631.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.246567344665527,
|
|
"epoch": 8.247721623860812,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.90722774555456e-05,
|
|
"loss": 4.4292,
|
|
"mean_token_accuracy": 0.23851138800382615,
|
|
"num_tokens": 18141950.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.224877738952637,
|
|
"epoch": 8.251864125932062,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.889250261051063e-05,
|
|
"loss": 4.4348,
|
|
"mean_token_accuracy": 0.23533212393522263,
|
|
"num_tokens": 18150721.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.187523317337036,
|
|
"epoch": 8.256006628003314,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.871310316463554e-05,
|
|
"loss": 4.3506,
|
|
"mean_token_accuracy": 0.24384420067071916,
|
|
"num_tokens": 18159706.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.207602167129517,
|
|
"epoch": 8.260149130074565,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.853407947978845e-05,
|
|
"loss": 4.4456,
|
|
"mean_token_accuracy": 0.23555804491043092,
|
|
"num_tokens": 18168221.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.294720697402954,
|
|
"epoch": 8.264291632145817,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.835543191707982e-05,
|
|
"loss": 4.5029,
|
|
"mean_token_accuracy": 0.22814472168684005,
|
|
"num_tokens": 18177396.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.2606728076934814,
|
|
"epoch": 8.268434134217067,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.81771608368613e-05,
|
|
"loss": 4.4787,
|
|
"mean_token_accuracy": 0.22940047085285187,
|
|
"num_tokens": 18186190.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.26011381149292,
|
|
"epoch": 8.272576636288319,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.799926659872507e-05,
|
|
"loss": 4.4602,
|
|
"mean_token_accuracy": 0.23847721964120866,
|
|
"num_tokens": 18195019.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.281919670104981,
|
|
"epoch": 8.276719138359569,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.782174956150335e-05,
|
|
"loss": 4.5188,
|
|
"mean_token_accuracy": 0.2311836987733841,
|
|
"num_tokens": 18204528.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.282897424697876,
|
|
"epoch": 8.280861640430821,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.764461008326723e-05,
|
|
"loss": 4.4498,
|
|
"mean_token_accuracy": 0.23748137354850768,
|
|
"num_tokens": 18213189.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.218039608001709,
|
|
"epoch": 8.285004142502071,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.746784852132661e-05,
|
|
"loss": 4.462,
|
|
"mean_token_accuracy": 0.23884083479642867,
|
|
"num_tokens": 18222432.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 8.285004142502071,
|
|
"eval_entropy": 5.158710529555136,
|
|
"eval_loss": 5.770650386810303,
|
|
"eval_mean_token_accuracy": 0.16699034202772417,
|
|
"eval_num_tokens": 18222432.0,
|
|
"eval_runtime": 2.7765,
|
|
"eval_samples_per_second": 1349.535,
|
|
"eval_steps_per_second": 168.917,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.299756813049316,
|
|
"epoch": 8.289146644573322,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.729146523222877e-05,
|
|
"loss": 4.4762,
|
|
"mean_token_accuracy": 0.23852387219667434,
|
|
"num_tokens": 18230630.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.22135910987854,
|
|
"epoch": 8.293289146644574,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.711546057175804e-05,
|
|
"loss": 4.4368,
|
|
"mean_token_accuracy": 0.23945680558681487,
|
|
"num_tokens": 18238903.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.2395326614379885,
|
|
"epoch": 8.297431648715824,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.69398348949351e-05,
|
|
"loss": 4.4682,
|
|
"mean_token_accuracy": 0.23651341646909713,
|
|
"num_tokens": 18248123.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.2905601978302,
|
|
"epoch": 8.301574150787076,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 8.676458855601612e-05,
|
|
"loss": 4.4569,
|
|
"mean_token_accuracy": 0.24139965772628785,
|
|
"num_tokens": 18257966.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.2831768035888675,
|
|
"epoch": 8.305716652858326,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.658972190849215e-05,
|
|
"loss": 4.5215,
|
|
"mean_token_accuracy": 0.228233177959919,
|
|
"num_tokens": 18267955.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.25753264427185,
|
|
"epoch": 8.309859154929578,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.641523530508827e-05,
|
|
"loss": 4.4578,
|
|
"mean_token_accuracy": 0.24362254589796067,
|
|
"num_tokens": 18277091.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.300469970703125,
|
|
"epoch": 8.314001657000828,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.624112909776314e-05,
|
|
"loss": 4.5161,
|
|
"mean_token_accuracy": 0.22749851047992706,
|
|
"num_tokens": 18286194.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.240508413314819,
|
|
"epoch": 8.31814415907208,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.606740363770788e-05,
|
|
"loss": 4.4811,
|
|
"mean_token_accuracy": 0.23397946804761888,
|
|
"num_tokens": 18295693.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.227995252609253,
|
|
"epoch": 8.32228666114333,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 8.589405927534571e-05,
|
|
"loss": 4.4541,
|
|
"mean_token_accuracy": 0.2426013395190239,
|
|
"num_tokens": 18303549.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.288485145568847,
|
|
"epoch": 8.326429163214582,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.57210963603313e-05,
|
|
"loss": 4.4746,
|
|
"mean_token_accuracy": 0.23579823970794678,
|
|
"num_tokens": 18312964.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.32373194694519,
|
|
"epoch": 8.330571665285833,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.554851524154961e-05,
|
|
"loss": 4.5635,
|
|
"mean_token_accuracy": 0.2303529217839241,
|
|
"num_tokens": 18321142.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.265218067169189,
|
|
"epoch": 8.334714167357085,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.537631626711573e-05,
|
|
"loss": 4.4624,
|
|
"mean_token_accuracy": 0.23054206222295762,
|
|
"num_tokens": 18330978.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.2279633522033695,
|
|
"epoch": 8.338856669428335,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.520449978437372e-05,
|
|
"loss": 4.4509,
|
|
"mean_token_accuracy": 0.24146103113889694,
|
|
"num_tokens": 18340309.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.254219007492066,
|
|
"epoch": 8.342999171499585,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.50330661398962e-05,
|
|
"loss": 4.4636,
|
|
"mean_token_accuracy": 0.23530360609292983,
|
|
"num_tokens": 18349103.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.252362871170044,
|
|
"epoch": 8.347141673570837,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.486201567948363e-05,
|
|
"loss": 4.5258,
|
|
"mean_token_accuracy": 0.23006199300289154,
|
|
"num_tokens": 18359151.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.2486168384552006,
|
|
"epoch": 8.351284175642087,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.469134874816342e-05,
|
|
"loss": 4.5166,
|
|
"mean_token_accuracy": 0.22913163900375366,
|
|
"num_tokens": 18368928.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.231261110305786,
|
|
"epoch": 8.35542667771334,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.45210656901895e-05,
|
|
"loss": 4.4275,
|
|
"mean_token_accuracy": 0.2332036688923836,
|
|
"num_tokens": 18379761.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.234804582595825,
|
|
"epoch": 8.35956917978459,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.435116684904136e-05,
|
|
"loss": 4.4921,
|
|
"mean_token_accuracy": 0.23260947614908217,
|
|
"num_tokens": 18389579.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.249290561676025,
|
|
"epoch": 8.363711681855841,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.418165256742346e-05,
|
|
"loss": 4.481,
|
|
"mean_token_accuracy": 0.2392916798591614,
|
|
"num_tokens": 18398375.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.265042734146118,
|
|
"epoch": 8.367854183927092,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.401252318726473e-05,
|
|
"loss": 4.5134,
|
|
"mean_token_accuracy": 0.22839679270982743,
|
|
"num_tokens": 18408015.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.261607122421265,
|
|
"epoch": 8.371996685998344,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 8.384377904971753e-05,
|
|
"loss": 4.3889,
|
|
"mean_token_accuracy": 0.24720900654792785,
|
|
"num_tokens": 18415923.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.305111646652222,
|
|
"epoch": 8.376139188069594,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 8.367542049515733e-05,
|
|
"loss": 4.5093,
|
|
"mean_token_accuracy": 0.2298623278737068,
|
|
"num_tokens": 18423878.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.300043344497681,
|
|
"epoch": 8.380281690140846,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.350744786318168e-05,
|
|
"loss": 4.4868,
|
|
"mean_token_accuracy": 0.2324920266866684,
|
|
"num_tokens": 18432822.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.2468890190124515,
|
|
"epoch": 8.384424192212096,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.33398614926097e-05,
|
|
"loss": 4.4661,
|
|
"mean_token_accuracy": 0.23152823448181153,
|
|
"num_tokens": 18441934.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.26579098701477,
|
|
"epoch": 8.388566694283346,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.317266172148145e-05,
|
|
"loss": 4.5027,
|
|
"mean_token_accuracy": 0.23406290113925934,
|
|
"num_tokens": 18451259.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.275383758544922,
|
|
"epoch": 8.392709196354598,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.300584888705709e-05,
|
|
"loss": 4.4891,
|
|
"mean_token_accuracy": 0.2315078541636467,
|
|
"num_tokens": 18461085.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.173043251037598,
|
|
"epoch": 8.396851698425849,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.283942332581644e-05,
|
|
"loss": 4.4517,
|
|
"mean_token_accuracy": 0.23731214106082915,
|
|
"num_tokens": 18469821.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.270431280136108,
|
|
"epoch": 8.4009942004971,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.267338537345792e-05,
|
|
"loss": 4.471,
|
|
"mean_token_accuracy": 0.23261656612157822,
|
|
"num_tokens": 18479795.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.269517278671264,
|
|
"epoch": 8.40513670256835,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.250773536489833e-05,
|
|
"loss": 4.5045,
|
|
"mean_token_accuracy": 0.2284645229578018,
|
|
"num_tokens": 18489193.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.237123680114746,
|
|
"epoch": 8.409279204639603,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.234247363427174e-05,
|
|
"loss": 4.4493,
|
|
"mean_token_accuracy": 0.23930226862430573,
|
|
"num_tokens": 18498147.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.309970855712891,
|
|
"epoch": 8.413421706710853,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.217760051492926e-05,
|
|
"loss": 4.6179,
|
|
"mean_token_accuracy": 0.22286109626293182,
|
|
"num_tokens": 18507318.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.292797231674195,
|
|
"epoch": 8.417564208782105,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.201311633943788e-05,
|
|
"loss": 4.4605,
|
|
"mean_token_accuracy": 0.240536130964756,
|
|
"num_tokens": 18516087.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.2840173721313475,
|
|
"epoch": 8.421706710853355,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.184902143958013e-05,
|
|
"loss": 4.4422,
|
|
"mean_token_accuracy": 0.23593308180570602,
|
|
"num_tokens": 18524336.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.2332299709320065,
|
|
"epoch": 8.425849212924607,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.168531614635347e-05,
|
|
"loss": 4.4616,
|
|
"mean_token_accuracy": 0.2356266662478447,
|
|
"num_tokens": 18533970.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.207690238952637,
|
|
"epoch": 8.429991714995857,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.152200078996924e-05,
|
|
"loss": 4.383,
|
|
"mean_token_accuracy": 0.24017399549484253,
|
|
"num_tokens": 18543285.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.2513281345367435,
|
|
"epoch": 8.434134217067108,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.135907569985248e-05,
|
|
"loss": 4.4389,
|
|
"mean_token_accuracy": 0.23822331726551055,
|
|
"num_tokens": 18552020.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.2607708930969235,
|
|
"epoch": 8.43827671913836,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.119654120464077e-05,
|
|
"loss": 4.4797,
|
|
"mean_token_accuracy": 0.2413047358393669,
|
|
"num_tokens": 18561056.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.27950963973999,
|
|
"epoch": 8.44241922120961,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.103439763218411e-05,
|
|
"loss": 4.498,
|
|
"mean_token_accuracy": 0.23119530975818633,
|
|
"num_tokens": 18570197.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.237138319015503,
|
|
"epoch": 8.446561723280862,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.087264530954362e-05,
|
|
"loss": 4.4434,
|
|
"mean_token_accuracy": 0.24342181533575058,
|
|
"num_tokens": 18579079.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.16762866973877,
|
|
"epoch": 8.450704225352112,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.071128456299159e-05,
|
|
"loss": 4.418,
|
|
"mean_token_accuracy": 0.23662289530038833,
|
|
"num_tokens": 18587672.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.261301517486572,
|
|
"epoch": 8.454846727423364,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.055031571801013e-05,
|
|
"loss": 4.481,
|
|
"mean_token_accuracy": 0.23190022110939026,
|
|
"num_tokens": 18597190.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.256229639053345,
|
|
"epoch": 8.458989229494614,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.038973909929112e-05,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.23834038376808167,
|
|
"num_tokens": 18606051.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.194650077819825,
|
|
"epoch": 8.463131731565866,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.022955503073502e-05,
|
|
"loss": 4.4277,
|
|
"mean_token_accuracy": 0.2443808987736702,
|
|
"num_tokens": 18614412.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.272135019302368,
|
|
"epoch": 8.467274233637117,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.006976383545071e-05,
|
|
"loss": 4.453,
|
|
"mean_token_accuracy": 0.23779590874910356,
|
|
"num_tokens": 18622895.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.315969705581665,
|
|
"epoch": 8.471416735708369,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.991036583575448e-05,
|
|
"loss": 4.5507,
|
|
"mean_token_accuracy": 0.23082175105810165,
|
|
"num_tokens": 18632915.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.259643697738648,
|
|
"epoch": 8.475559237779619,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.975136135316941e-05,
|
|
"loss": 4.414,
|
|
"mean_token_accuracy": 0.2516813114285469,
|
|
"num_tokens": 18642017.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.234678840637207,
|
|
"epoch": 8.47970173985087,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.959275070842502e-05,
|
|
"loss": 4.4408,
|
|
"mean_token_accuracy": 0.23919759392738343,
|
|
"num_tokens": 18650385.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.200926160812378,
|
|
"epoch": 8.483844241922121,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.943453422145626e-05,
|
|
"loss": 4.4258,
|
|
"mean_token_accuracy": 0.23791312724351882,
|
|
"num_tokens": 18660020.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.204594612121582,
|
|
"epoch": 8.487986743993371,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.927671221140317e-05,
|
|
"loss": 4.4232,
|
|
"mean_token_accuracy": 0.2406802698969841,
|
|
"num_tokens": 18668880.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.268185377120972,
|
|
"epoch": 8.492129246064623,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.911928499660986e-05,
|
|
"loss": 4.557,
|
|
"mean_token_accuracy": 0.23018244355916978,
|
|
"num_tokens": 18678474.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.307393884658813,
|
|
"epoch": 8.496271748135873,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.896225289462426e-05,
|
|
"loss": 4.5823,
|
|
"mean_token_accuracy": 0.22468128353357314,
|
|
"num_tokens": 18687614.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.257604694366455,
|
|
"epoch": 8.500414250207125,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.880561622219735e-05,
|
|
"loss": 4.516,
|
|
"mean_token_accuracy": 0.23317746371030806,
|
|
"num_tokens": 18697440.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.288731718063355,
|
|
"epoch": 8.504556752278376,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.864937529528232e-05,
|
|
"loss": 4.4796,
|
|
"mean_token_accuracy": 0.23478027284145356,
|
|
"num_tokens": 18706659.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.257978296279907,
|
|
"epoch": 8.508699254349628,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.849353042903428e-05,
|
|
"loss": 4.3901,
|
|
"mean_token_accuracy": 0.23892433494329451,
|
|
"num_tokens": 18715041.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.230406761169434,
|
|
"epoch": 8.512841756420878,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.833808193780934e-05,
|
|
"loss": 4.4353,
|
|
"mean_token_accuracy": 0.23951757550239564,
|
|
"num_tokens": 18723532.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.193192768096924,
|
|
"epoch": 8.51698425849213,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.818303013516401e-05,
|
|
"loss": 4.4456,
|
|
"mean_token_accuracy": 0.23859017789363862,
|
|
"num_tokens": 18731939.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.249845933914185,
|
|
"epoch": 8.52112676056338,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.802837533385482e-05,
|
|
"loss": 4.4354,
|
|
"mean_token_accuracy": 0.23805483132600785,
|
|
"num_tokens": 18740071.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.152613115310669,
|
|
"epoch": 8.525269262634632,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.787411784583733e-05,
|
|
"loss": 4.3771,
|
|
"mean_token_accuracy": 0.23865023702383042,
|
|
"num_tokens": 18749024.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.243736410140992,
|
|
"epoch": 8.529411764705882,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.772025798226582e-05,
|
|
"loss": 4.4088,
|
|
"mean_token_accuracy": 0.23425072580575942,
|
|
"num_tokens": 18758511.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.2582190990447994,
|
|
"epoch": 8.533554266777134,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.756679605349241e-05,
|
|
"loss": 4.4715,
|
|
"mean_token_accuracy": 0.2339015409350395,
|
|
"num_tokens": 18768729.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.282213115692139,
|
|
"epoch": 8.537696768848384,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.741373236906658e-05,
|
|
"loss": 4.4815,
|
|
"mean_token_accuracy": 0.23057300448417664,
|
|
"num_tokens": 18778399.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.272563791275024,
|
|
"epoch": 8.541839270919635,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.726106723773452e-05,
|
|
"loss": 4.5631,
|
|
"mean_token_accuracy": 0.22452725172042848,
|
|
"num_tokens": 18787845.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.274098348617554,
|
|
"epoch": 8.545981772990887,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.710880096743845e-05,
|
|
"loss": 4.5638,
|
|
"mean_token_accuracy": 0.22270353138446808,
|
|
"num_tokens": 18797607.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.264918375015259,
|
|
"epoch": 8.550124275062137,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.695693386531612e-05,
|
|
"loss": 4.4655,
|
|
"mean_token_accuracy": 0.23435032218694687,
|
|
"num_tokens": 18806583.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.233798551559448,
|
|
"epoch": 8.554266777133389,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.680546623770001e-05,
|
|
"loss": 4.4873,
|
|
"mean_token_accuracy": 0.2384819805622101,
|
|
"num_tokens": 18816456.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.268015432357788,
|
|
"epoch": 8.55840927920464,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.665439839011701e-05,
|
|
"loss": 4.4623,
|
|
"mean_token_accuracy": 0.23856367468833922,
|
|
"num_tokens": 18826457.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.198453187942505,
|
|
"epoch": 8.562551781275891,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.650373062728735e-05,
|
|
"loss": 4.4524,
|
|
"mean_token_accuracy": 0.23532486110925674,
|
|
"num_tokens": 18836612.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.2620970726013185,
|
|
"epoch": 8.566694283347141,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.635346325312437e-05,
|
|
"loss": 4.5429,
|
|
"mean_token_accuracy": 0.237743978202343,
|
|
"num_tokens": 18846088.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.2506757259368895,
|
|
"epoch": 8.570836785418393,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.62035965707339e-05,
|
|
"loss": 4.4503,
|
|
"mean_token_accuracy": 0.23594015836715698,
|
|
"num_tokens": 18854874.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.246778535842895,
|
|
"epoch": 8.574979287489644,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.605413088241325e-05,
|
|
"loss": 4.4782,
|
|
"mean_token_accuracy": 0.2415588155388832,
|
|
"num_tokens": 18864320.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.281873989105224,
|
|
"epoch": 8.579121789560896,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.59050664896512e-05,
|
|
"loss": 4.4939,
|
|
"mean_token_accuracy": 0.23168267905712128,
|
|
"num_tokens": 18873625.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.213468456268311,
|
|
"epoch": 8.583264291632146,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.575640369312684e-05,
|
|
"loss": 4.4391,
|
|
"mean_token_accuracy": 0.238327993452549,
|
|
"num_tokens": 18881836.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.201680278778076,
|
|
"epoch": 8.587406793703398,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.560814279270928e-05,
|
|
"loss": 4.4376,
|
|
"mean_token_accuracy": 0.23110086619853973,
|
|
"num_tokens": 18891002.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.222077703475952,
|
|
"epoch": 8.591549295774648,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.5460284087457e-05,
|
|
"loss": 4.5123,
|
|
"mean_token_accuracy": 0.22924143373966216,
|
|
"num_tokens": 18901100.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.247162103652954,
|
|
"epoch": 8.595691797845898,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.53128278756171e-05,
|
|
"loss": 4.4956,
|
|
"mean_token_accuracy": 0.2297233149409294,
|
|
"num_tokens": 18910021.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.200025510787964,
|
|
"epoch": 8.59983429991715,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.516577445462493e-05,
|
|
"loss": 4.4074,
|
|
"mean_token_accuracy": 0.2421842485666275,
|
|
"num_tokens": 18919000.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.120391225814819,
|
|
"epoch": 8.6039768019884,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.501912412110331e-05,
|
|
"loss": 4.3301,
|
|
"mean_token_accuracy": 0.25499629229307175,
|
|
"num_tokens": 18927954.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.234947299957275,
|
|
"epoch": 8.608119304059652,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.487287717086193e-05,
|
|
"loss": 4.4441,
|
|
"mean_token_accuracy": 0.23775298297405242,
|
|
"num_tokens": 18937098.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.256605529785157,
|
|
"epoch": 8.612261806130903,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.472703389889695e-05,
|
|
"loss": 4.5028,
|
|
"mean_token_accuracy": 0.22914714962244034,
|
|
"num_tokens": 18945501.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.270039510726929,
|
|
"epoch": 8.616404308202155,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.458159459939007e-05,
|
|
"loss": 4.5334,
|
|
"mean_token_accuracy": 0.23301581889390946,
|
|
"num_tokens": 18955191.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.220399951934814,
|
|
"epoch": 8.620546810273405,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.443655956570842e-05,
|
|
"loss": 4.433,
|
|
"mean_token_accuracy": 0.24217931032180787,
|
|
"num_tokens": 18963778.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.249224233627319,
|
|
"epoch": 8.624689312344657,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.429192909040339e-05,
|
|
"loss": 4.511,
|
|
"mean_token_accuracy": 0.23294726610183716,
|
|
"num_tokens": 18972286.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.382362174987793,
|
|
"epoch": 8.628831814415907,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.414770346521043e-05,
|
|
"loss": 4.6948,
|
|
"mean_token_accuracy": 0.22690797597169876,
|
|
"num_tokens": 18982138.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.211470365524292,
|
|
"epoch": 8.632974316487159,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.400388298104851e-05,
|
|
"loss": 4.4325,
|
|
"mean_token_accuracy": 0.23553294092416763,
|
|
"num_tokens": 18990785.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.259786987304688,
|
|
"epoch": 8.63711681855841,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.386046792801911e-05,
|
|
"loss": 4.4608,
|
|
"mean_token_accuracy": 0.2389303505420685,
|
|
"num_tokens": 19000183.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.252145624160766,
|
|
"epoch": 8.64125932062966,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.371745859540622e-05,
|
|
"loss": 4.4683,
|
|
"mean_token_accuracy": 0.23253835886716842,
|
|
"num_tokens": 19009039.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.326835298538208,
|
|
"epoch": 8.645401822700912,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.357485527167522e-05,
|
|
"loss": 4.5806,
|
|
"mean_token_accuracy": 0.22435685247182846,
|
|
"num_tokens": 19017446.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.221448707580566,
|
|
"epoch": 8.649544324772162,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 7.343265824447254e-05,
|
|
"loss": 4.4513,
|
|
"mean_token_accuracy": 0.23768773227930068,
|
|
"num_tokens": 19026222.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.250608444213867,
|
|
"epoch": 8.653686826843414,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.329086780062522e-05,
|
|
"loss": 4.4569,
|
|
"mean_token_accuracy": 0.24067220687866211,
|
|
"num_tokens": 19035765.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.283592605590821,
|
|
"epoch": 8.657829328914664,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 7.314948422613998e-05,
|
|
"loss": 4.4663,
|
|
"mean_token_accuracy": 0.237451434135437,
|
|
"num_tokens": 19044159.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.210062456130982,
|
|
"epoch": 8.661971830985916,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.30085078062031e-05,
|
|
"loss": 4.3968,
|
|
"mean_token_accuracy": 0.2410138502717018,
|
|
"num_tokens": 19053273.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.234121656417846,
|
|
"epoch": 8.666114333057166,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.286793882517922e-05,
|
|
"loss": 4.4682,
|
|
"mean_token_accuracy": 0.23808223754167557,
|
|
"num_tokens": 19062405.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.260992002487183,
|
|
"epoch": 8.670256835128418,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.272777756661152e-05,
|
|
"loss": 4.4251,
|
|
"mean_token_accuracy": 0.23846110850572586,
|
|
"num_tokens": 19070839.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.276519393920898,
|
|
"epoch": 8.674399337199668,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.258802431322047e-05,
|
|
"loss": 4.5127,
|
|
"mean_token_accuracy": 0.24230123907327653,
|
|
"num_tokens": 19079674.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.216338586807251,
|
|
"epoch": 8.67854183927092,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.244867934690364e-05,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.23942160308361055,
|
|
"num_tokens": 19088813.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.149845695495605,
|
|
"epoch": 8.68268434134217,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.230974294873513e-05,
|
|
"loss": 4.3272,
|
|
"mean_token_accuracy": 0.2522314369678497,
|
|
"num_tokens": 19096805.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.225519418716431,
|
|
"epoch": 8.68682684341342,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.217121539896476e-05,
|
|
"loss": 4.4726,
|
|
"mean_token_accuracy": 0.24243174344301224,
|
|
"num_tokens": 19105423.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.204275465011596,
|
|
"epoch": 8.690969345484673,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.203309697701784e-05,
|
|
"loss": 4.3673,
|
|
"mean_token_accuracy": 0.2505881696939468,
|
|
"num_tokens": 19113481.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.248758602142334,
|
|
"epoch": 8.695111847555923,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.189538796149424e-05,
|
|
"loss": 4.4973,
|
|
"mean_token_accuracy": 0.230351784825325,
|
|
"num_tokens": 19122987.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.255853271484375,
|
|
"epoch": 8.699254349627175,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.175808863016806e-05,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.23196303695440293,
|
|
"num_tokens": 19133180.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 8.699254349627175,
|
|
"eval_entropy": 5.152034001055557,
|
|
"eval_loss": 5.768686294555664,
|
|
"eval_mean_token_accuracy": 0.16696232431796568,
|
|
"eval_num_tokens": 19133180.0,
|
|
"eval_runtime": 2.9882,
|
|
"eval_samples_per_second": 1253.922,
|
|
"eval_steps_per_second": 156.949,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 5.2888922691345215,
|
|
"epoch": 8.703396851698425,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.162119925998714e-05,
|
|
"loss": 4.5287,
|
|
"mean_token_accuracy": 0.23070713579654695,
|
|
"num_tokens": 19142305.0,
|
|
"step": 10505
|
|
},
|
|
{
|
|
"entropy": 5.27380518913269,
|
|
"epoch": 8.707539353769677,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.148472012707225e-05,
|
|
"loss": 4.4442,
|
|
"mean_token_accuracy": 0.2338199570775032,
|
|
"num_tokens": 19152360.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 5.186902618408203,
|
|
"epoch": 8.711681855840927,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.134865150671676e-05,
|
|
"loss": 4.4342,
|
|
"mean_token_accuracy": 0.24185711592435838,
|
|
"num_tokens": 19162132.0,
|
|
"step": 10515
|
|
},
|
|
{
|
|
"entropy": 5.330926418304443,
|
|
"epoch": 8.71582435791218,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.121299367338593e-05,
|
|
"loss": 4.5521,
|
|
"mean_token_accuracy": 0.22707564532756805,
|
|
"num_tokens": 19171224.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 5.282039165496826,
|
|
"epoch": 8.71996685998343,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.107774690071635e-05,
|
|
"loss": 4.4663,
|
|
"mean_token_accuracy": 0.23552273064851761,
|
|
"num_tokens": 19180096.0,
|
|
"step": 10525
|
|
},
|
|
{
|
|
"entropy": 5.225061321258545,
|
|
"epoch": 8.724109362054682,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.094291146151568e-05,
|
|
"loss": 4.4335,
|
|
"mean_token_accuracy": 0.23152414709329605,
|
|
"num_tokens": 19189115.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 5.224990653991699,
|
|
"epoch": 8.728251864125932,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.080848762776159e-05,
|
|
"loss": 4.4518,
|
|
"mean_token_accuracy": 0.2368248000741005,
|
|
"num_tokens": 19197328.0,
|
|
"step": 10535
|
|
},
|
|
{
|
|
"entropy": 5.191633653640747,
|
|
"epoch": 8.732394366197184,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.067447567060178e-05,
|
|
"loss": 4.4075,
|
|
"mean_token_accuracy": 0.24210347086191178,
|
|
"num_tokens": 19207198.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 5.261967134475708,
|
|
"epoch": 8.736536868268434,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.054087586035295e-05,
|
|
"loss": 4.446,
|
|
"mean_token_accuracy": 0.24023890346288682,
|
|
"num_tokens": 19216296.0,
|
|
"step": 10545
|
|
},
|
|
{
|
|
"entropy": 5.263440036773682,
|
|
"epoch": 8.740679370339684,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.040768846650046e-05,
|
|
"loss": 4.5035,
|
|
"mean_token_accuracy": 0.233292655646801,
|
|
"num_tokens": 19225351.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 5.277057886123657,
|
|
"epoch": 8.744821872410936,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.027491375769795e-05,
|
|
"loss": 4.4272,
|
|
"mean_token_accuracy": 0.23754440248012543,
|
|
"num_tokens": 19234431.0,
|
|
"step": 10555
|
|
},
|
|
{
|
|
"entropy": 5.26463475227356,
|
|
"epoch": 8.748964374482187,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.014255200176643e-05,
|
|
"loss": 4.4717,
|
|
"mean_token_accuracy": 0.23711207062005996,
|
|
"num_tokens": 19244475.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 5.195476102828979,
|
|
"epoch": 8.753106876553439,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.001060346569408e-05,
|
|
"loss": 4.42,
|
|
"mean_token_accuracy": 0.24640779346227645,
|
|
"num_tokens": 19253882.0,
|
|
"step": 10565
|
|
},
|
|
{
|
|
"entropy": 5.193240022659301,
|
|
"epoch": 8.757249378624689,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.987906841563552e-05,
|
|
"loss": 4.4109,
|
|
"mean_token_accuracy": 0.2377583384513855,
|
|
"num_tokens": 19262569.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 5.24358229637146,
|
|
"epoch": 8.76139188069594,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.974794711691125e-05,
|
|
"loss": 4.4467,
|
|
"mean_token_accuracy": 0.24153018295764922,
|
|
"num_tokens": 19270767.0,
|
|
"step": 10575
|
|
},
|
|
{
|
|
"entropy": 5.264425945281983,
|
|
"epoch": 8.765534382767191,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.961723983400733e-05,
|
|
"loss": 4.4574,
|
|
"mean_token_accuracy": 0.24082808941602707,
|
|
"num_tokens": 19280105.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 5.231091690063477,
|
|
"epoch": 8.769676884838443,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.948694683057456e-05,
|
|
"loss": 4.4681,
|
|
"mean_token_accuracy": 0.23813995122909545,
|
|
"num_tokens": 19289600.0,
|
|
"step": 10585
|
|
},
|
|
{
|
|
"entropy": 5.217465400695801,
|
|
"epoch": 8.773819386909693,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.935706836942826e-05,
|
|
"loss": 4.4054,
|
|
"mean_token_accuracy": 0.24610555768013,
|
|
"num_tokens": 19299053.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 5.24623851776123,
|
|
"epoch": 8.777961888980945,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.922760471254738e-05,
|
|
"loss": 4.4388,
|
|
"mean_token_accuracy": 0.23524802029132844,
|
|
"num_tokens": 19308862.0,
|
|
"step": 10595
|
|
},
|
|
{
|
|
"entropy": 5.258311748504639,
|
|
"epoch": 8.782104391052195,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.909855612107437e-05,
|
|
"loss": 4.518,
|
|
"mean_token_accuracy": 0.23038132935762407,
|
|
"num_tokens": 19318055.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 5.256415700912475,
|
|
"epoch": 8.786246893123447,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.896992285531424e-05,
|
|
"loss": 4.4815,
|
|
"mean_token_accuracy": 0.23155483305454255,
|
|
"num_tokens": 19327992.0,
|
|
"step": 10605
|
|
},
|
|
{
|
|
"entropy": 5.23081226348877,
|
|
"epoch": 8.790389395194698,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.884170517473444e-05,
|
|
"loss": 4.4127,
|
|
"mean_token_accuracy": 0.23641562312841416,
|
|
"num_tokens": 19336401.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 5.244655561447144,
|
|
"epoch": 8.794531897265948,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.871390333796398e-05,
|
|
"loss": 4.4667,
|
|
"mean_token_accuracy": 0.22823416441679,
|
|
"num_tokens": 19345469.0,
|
|
"step": 10615
|
|
},
|
|
{
|
|
"entropy": 5.301026344299316,
|
|
"epoch": 8.7986743993372,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.85865176027931e-05,
|
|
"loss": 4.4476,
|
|
"mean_token_accuracy": 0.23840482234954835,
|
|
"num_tokens": 19354201.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 5.240279960632324,
|
|
"epoch": 8.80281690140845,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.845954822617284e-05,
|
|
"loss": 4.536,
|
|
"mean_token_accuracy": 0.22835701256990432,
|
|
"num_tokens": 19363470.0,
|
|
"step": 10625
|
|
},
|
|
{
|
|
"entropy": 5.267053508758545,
|
|
"epoch": 8.806959403479702,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.833299546421423e-05,
|
|
"loss": 4.4672,
|
|
"mean_token_accuracy": 0.2311469852924347,
|
|
"num_tokens": 19371924.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 5.306638431549072,
|
|
"epoch": 8.811101905550952,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.820685957218808e-05,
|
|
"loss": 4.5448,
|
|
"mean_token_accuracy": 0.23532919138669967,
|
|
"num_tokens": 19382118.0,
|
|
"step": 10635
|
|
},
|
|
{
|
|
"entropy": 5.289222478866577,
|
|
"epoch": 8.815244407622204,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.80811408045242e-05,
|
|
"loss": 4.5139,
|
|
"mean_token_accuracy": 0.233585125207901,
|
|
"num_tokens": 19391115.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 5.232711982727051,
|
|
"epoch": 8.819386909693455,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.795583941481113e-05,
|
|
"loss": 4.4582,
|
|
"mean_token_accuracy": 0.23468625247478486,
|
|
"num_tokens": 19400289.0,
|
|
"step": 10645
|
|
},
|
|
{
|
|
"entropy": 5.224986219406128,
|
|
"epoch": 8.823529411764707,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.783095565579546e-05,
|
|
"loss": 4.4758,
|
|
"mean_token_accuracy": 0.2367667466402054,
|
|
"num_tokens": 19409548.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 5.329962825775146,
|
|
"epoch": 8.827671913835957,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.77064897793814e-05,
|
|
"loss": 4.5724,
|
|
"mean_token_accuracy": 0.22351539134979248,
|
|
"num_tokens": 19418695.0,
|
|
"step": 10655
|
|
},
|
|
{
|
|
"entropy": 5.198006343841553,
|
|
"epoch": 8.831814415907209,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.758244203663015e-05,
|
|
"loss": 4.4103,
|
|
"mean_token_accuracy": 0.23261348009109498,
|
|
"num_tokens": 19428515.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 5.181308317184448,
|
|
"epoch": 8.835956917978459,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.745881267775967e-05,
|
|
"loss": 4.4056,
|
|
"mean_token_accuracy": 0.2418215438723564,
|
|
"num_tokens": 19437247.0,
|
|
"step": 10665
|
|
},
|
|
{
|
|
"entropy": 5.228489017486572,
|
|
"epoch": 8.840099420049711,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.733560195214384e-05,
|
|
"loss": 4.4675,
|
|
"mean_token_accuracy": 0.23592644333839416,
|
|
"num_tokens": 19446533.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 5.243283319473266,
|
|
"epoch": 8.844241922120961,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.721281010831209e-05,
|
|
"loss": 4.4965,
|
|
"mean_token_accuracy": 0.23029326200485228,
|
|
"num_tokens": 19456116.0,
|
|
"step": 10675
|
|
},
|
|
{
|
|
"entropy": 5.280758810043335,
|
|
"epoch": 8.848384424192211,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.709043739394909e-05,
|
|
"loss": 4.4666,
|
|
"mean_token_accuracy": 0.23763754814863206,
|
|
"num_tokens": 19464668.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 5.27049150466919,
|
|
"epoch": 8.852526926263463,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.696848405589389e-05,
|
|
"loss": 4.4246,
|
|
"mean_token_accuracy": 0.2404986873269081,
|
|
"num_tokens": 19473521.0,
|
|
"step": 10685
|
|
},
|
|
{
|
|
"entropy": 5.255083513259888,
|
|
"epoch": 8.856669428334714,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.68469503401398e-05,
|
|
"loss": 4.4387,
|
|
"mean_token_accuracy": 0.2356901302933693,
|
|
"num_tokens": 19482633.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 5.2859635829925535,
|
|
"epoch": 8.860811930405966,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.672583649183346e-05,
|
|
"loss": 4.5474,
|
|
"mean_token_accuracy": 0.2289673313498497,
|
|
"num_tokens": 19492121.0,
|
|
"step": 10695
|
|
},
|
|
{
|
|
"entropy": 5.280924081802368,
|
|
"epoch": 8.864954432477216,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.66051427552749e-05,
|
|
"loss": 4.4704,
|
|
"mean_token_accuracy": 0.22637060582637786,
|
|
"num_tokens": 19501117.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 5.184842157363891,
|
|
"epoch": 8.869096934548468,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.648486937391651e-05,
|
|
"loss": 4.4074,
|
|
"mean_token_accuracy": 0.239145191013813,
|
|
"num_tokens": 19509727.0,
|
|
"step": 10705
|
|
},
|
|
{
|
|
"entropy": 5.249301958084106,
|
|
"epoch": 8.873239436619718,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.636501659036278e-05,
|
|
"loss": 4.5418,
|
|
"mean_token_accuracy": 0.23166395127773284,
|
|
"num_tokens": 19519740.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 5.2324120044708256,
|
|
"epoch": 8.87738193869097,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.624558464636999e-05,
|
|
"loss": 4.4394,
|
|
"mean_token_accuracy": 0.238249671459198,
|
|
"num_tokens": 19529427.0,
|
|
"step": 10715
|
|
},
|
|
{
|
|
"entropy": 5.17536039352417,
|
|
"epoch": 8.88152444076222,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.612657378284528e-05,
|
|
"loss": 4.3998,
|
|
"mean_token_accuracy": 0.2424028381705284,
|
|
"num_tokens": 19537842.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 5.252032041549683,
|
|
"epoch": 8.885666942833472,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.600798423984671e-05,
|
|
"loss": 4.4923,
|
|
"mean_token_accuracy": 0.23236556351184845,
|
|
"num_tokens": 19546500.0,
|
|
"step": 10725
|
|
},
|
|
{
|
|
"entropy": 5.192902135848999,
|
|
"epoch": 8.889809444904722,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.588981625658227e-05,
|
|
"loss": 4.4435,
|
|
"mean_token_accuracy": 0.23461371660232544,
|
|
"num_tokens": 19554960.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 5.1956158638000485,
|
|
"epoch": 8.893951946975973,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.577207007140969e-05,
|
|
"loss": 4.4816,
|
|
"mean_token_accuracy": 0.235395647585392,
|
|
"num_tokens": 19564133.0,
|
|
"step": 10735
|
|
},
|
|
{
|
|
"entropy": 5.260041236877441,
|
|
"epoch": 8.898094449047225,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.565474592183594e-05,
|
|
"loss": 4.532,
|
|
"mean_token_accuracy": 0.2310117229819298,
|
|
"num_tokens": 19573879.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 5.2741374492645265,
|
|
"epoch": 8.902236951118475,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.553784404451656e-05,
|
|
"loss": 4.488,
|
|
"mean_token_accuracy": 0.23017262518405915,
|
|
"num_tokens": 19583137.0,
|
|
"step": 10745
|
|
},
|
|
{
|
|
"entropy": 5.291891145706177,
|
|
"epoch": 8.906379453189727,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.542136467525555e-05,
|
|
"loss": 4.4746,
|
|
"mean_token_accuracy": 0.23570170849561692,
|
|
"num_tokens": 19592289.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 5.313775205612183,
|
|
"epoch": 8.910521955260977,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.530530804900441e-05,
|
|
"loss": 4.5401,
|
|
"mean_token_accuracy": 0.22770173698663712,
|
|
"num_tokens": 19600885.0,
|
|
"step": 10755
|
|
},
|
|
{
|
|
"entropy": 5.232036066055298,
|
|
"epoch": 8.91466445733223,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.518967439986209e-05,
|
|
"loss": 4.3475,
|
|
"mean_token_accuracy": 0.25298545211553575,
|
|
"num_tokens": 19609320.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 5.240622663497925,
|
|
"epoch": 8.91880695940348,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.507446396107429e-05,
|
|
"loss": 4.4423,
|
|
"mean_token_accuracy": 0.23575087934732436,
|
|
"num_tokens": 19618796.0,
|
|
"step": 10765
|
|
},
|
|
{
|
|
"entropy": 5.249802780151367,
|
|
"epoch": 8.922949461474731,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.495967696503303e-05,
|
|
"loss": 4.4497,
|
|
"mean_token_accuracy": 0.23797554522752762,
|
|
"num_tokens": 19628144.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 5.250053071975708,
|
|
"epoch": 8.927091963545982,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.484531364327629e-05,
|
|
"loss": 4.4407,
|
|
"mean_token_accuracy": 0.24095406830310823,
|
|
"num_tokens": 19636750.0,
|
|
"step": 10775
|
|
},
|
|
{
|
|
"entropy": 5.248784494400025,
|
|
"epoch": 8.931234465617234,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.473137422648728e-05,
|
|
"loss": 4.5036,
|
|
"mean_token_accuracy": 0.2321176201105118,
|
|
"num_tokens": 19646697.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 5.246744155883789,
|
|
"epoch": 8.935376967688484,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.461785894449434e-05,
|
|
"loss": 4.4464,
|
|
"mean_token_accuracy": 0.23951565176248552,
|
|
"num_tokens": 19655305.0,
|
|
"step": 10785
|
|
},
|
|
{
|
|
"entropy": 5.2878687381744385,
|
|
"epoch": 8.939519469759734,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.45047680262702e-05,
|
|
"loss": 4.4905,
|
|
"mean_token_accuracy": 0.23326742351055146,
|
|
"num_tokens": 19664092.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 5.263401079177856,
|
|
"epoch": 8.943661971830986,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.439210169993153e-05,
|
|
"loss": 4.5034,
|
|
"mean_token_accuracy": 0.2256308302283287,
|
|
"num_tokens": 19673393.0,
|
|
"step": 10795
|
|
},
|
|
{
|
|
"entropy": 5.183784437179566,
|
|
"epoch": 8.947804473902236,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.427986019273873e-05,
|
|
"loss": 4.3618,
|
|
"mean_token_accuracy": 0.24535836577415465,
|
|
"num_tokens": 19680934.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 5.236357069015503,
|
|
"epoch": 8.951946975973488,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.416804373109507e-05,
|
|
"loss": 4.4951,
|
|
"mean_token_accuracy": 0.2379722088575363,
|
|
"num_tokens": 19689758.0,
|
|
"step": 10805
|
|
},
|
|
{
|
|
"entropy": 5.216552448272705,
|
|
"epoch": 8.956089478044738,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.405665254054666e-05,
|
|
"loss": 4.4467,
|
|
"mean_token_accuracy": 0.24092562943696977,
|
|
"num_tokens": 19698422.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 5.241431140899659,
|
|
"epoch": 8.96023198011599,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.394568684578165e-05,
|
|
"loss": 4.4376,
|
|
"mean_token_accuracy": 0.24312170892953872,
|
|
"num_tokens": 19708075.0,
|
|
"step": 10815
|
|
},
|
|
{
|
|
"entropy": 5.222333860397339,
|
|
"epoch": 8.96437448218724,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.383514687063e-05,
|
|
"loss": 4.4528,
|
|
"mean_token_accuracy": 0.23652777671813965,
|
|
"num_tokens": 19718346.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 5.289671611785889,
|
|
"epoch": 8.968516984258493,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.372503283806291e-05,
|
|
"loss": 4.531,
|
|
"mean_token_accuracy": 0.232435967028141,
|
|
"num_tokens": 19728303.0,
|
|
"step": 10825
|
|
},
|
|
{
|
|
"entropy": 5.257823514938354,
|
|
"epoch": 8.972659486329743,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.36153449701924e-05,
|
|
"loss": 4.4227,
|
|
"mean_token_accuracy": 0.24222785383462905,
|
|
"num_tokens": 19737001.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 5.308825874328614,
|
|
"epoch": 8.976801988400995,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.350608348827096e-05,
|
|
"loss": 4.5875,
|
|
"mean_token_accuracy": 0.223843951523304,
|
|
"num_tokens": 19746719.0,
|
|
"step": 10835
|
|
},
|
|
{
|
|
"entropy": 5.27318844795227,
|
|
"epoch": 8.980944490472245,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.339724861269085e-05,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.23224152475595475,
|
|
"num_tokens": 19757336.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 5.169971466064453,
|
|
"epoch": 8.985086992543497,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.328884056298393e-05,
|
|
"loss": 4.3983,
|
|
"mean_token_accuracy": 0.2398223266005516,
|
|
"num_tokens": 19766188.0,
|
|
"step": 10845
|
|
},
|
|
{
|
|
"entropy": 5.246915864944458,
|
|
"epoch": 8.989229494614747,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.31808595578211e-05,
|
|
"loss": 4.4894,
|
|
"mean_token_accuracy": 0.23585894405841829,
|
|
"num_tokens": 19774887.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 5.198973751068115,
|
|
"epoch": 8.993371996685998,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.307330581501183e-05,
|
|
"loss": 4.4524,
|
|
"mean_token_accuracy": 0.23435194790363312,
|
|
"num_tokens": 19783760.0,
|
|
"step": 10855
|
|
},
|
|
{
|
|
"entropy": 5.291239929199219,
|
|
"epoch": 8.99751449875725,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.29661795515038e-05,
|
|
"loss": 4.6043,
|
|
"mean_token_accuracy": 0.2314337059855461,
|
|
"num_tokens": 19794907.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 5.234113121032715,
|
|
"epoch": 9.0016570008285,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.285948098338238e-05,
|
|
"loss": 4.4624,
|
|
"mean_token_accuracy": 0.23397717326879502,
|
|
"num_tokens": 19803320.0,
|
|
"step": 10865
|
|
},
|
|
{
|
|
"entropy": 5.2232214450836185,
|
|
"epoch": 9.005799502899752,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.27532103258702e-05,
|
|
"loss": 4.423,
|
|
"mean_token_accuracy": 0.24048043191432952,
|
|
"num_tokens": 19812704.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 5.2392534732818605,
|
|
"epoch": 9.009942004971002,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.26473677933269e-05,
|
|
"loss": 4.3742,
|
|
"mean_token_accuracy": 0.2451390579342842,
|
|
"num_tokens": 19821718.0,
|
|
"step": 10875
|
|
},
|
|
{
|
|
"entropy": 5.276467943191529,
|
|
"epoch": 9.014084507042254,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.254195359924827e-05,
|
|
"loss": 4.4245,
|
|
"mean_token_accuracy": 0.24329400956630706,
|
|
"num_tokens": 19830391.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 5.209226417541504,
|
|
"epoch": 9.018227009113504,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.24369679562664e-05,
|
|
"loss": 4.3577,
|
|
"mean_token_accuracy": 0.2449675038456917,
|
|
"num_tokens": 19838799.0,
|
|
"step": 10885
|
|
},
|
|
{
|
|
"entropy": 5.263902473449707,
|
|
"epoch": 9.022369511184756,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.233241107614875e-05,
|
|
"loss": 4.5225,
|
|
"mean_token_accuracy": 0.2288858786225319,
|
|
"num_tokens": 19847828.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 5.23580379486084,
|
|
"epoch": 9.026512013256006,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.222828316979798e-05,
|
|
"loss": 4.4133,
|
|
"mean_token_accuracy": 0.24425291419029235,
|
|
"num_tokens": 19856561.0,
|
|
"step": 10895
|
|
},
|
|
{
|
|
"entropy": 5.232066202163696,
|
|
"epoch": 9.030654515327258,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.212458444725146e-05,
|
|
"loss": 4.4885,
|
|
"mean_token_accuracy": 0.23170086294412612,
|
|
"num_tokens": 19865801.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 5.19902400970459,
|
|
"epoch": 9.034797017398509,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.202131511768079e-05,
|
|
"loss": 4.387,
|
|
"mean_token_accuracy": 0.2440118134021759,
|
|
"num_tokens": 19874749.0,
|
|
"step": 10905
|
|
},
|
|
{
|
|
"entropy": 5.2941333770751955,
|
|
"epoch": 9.03893951946976,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.19184753893916e-05,
|
|
"loss": 4.5523,
|
|
"mean_token_accuracy": 0.22943517565727234,
|
|
"num_tokens": 19884036.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 5.297849845886231,
|
|
"epoch": 9.04308202154101,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.181606546982274e-05,
|
|
"loss": 4.4389,
|
|
"mean_token_accuracy": 0.24659695029258727,
|
|
"num_tokens": 19893335.0,
|
|
"step": 10915
|
|
},
|
|
{
|
|
"entropy": 5.195471668243409,
|
|
"epoch": 9.047224523612261,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.171408556554633e-05,
|
|
"loss": 4.3293,
|
|
"mean_token_accuracy": 0.24293112754821777,
|
|
"num_tokens": 19902862.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 5.223734617233276,
|
|
"epoch": 9.051367025683513,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.16125358822669e-05,
|
|
"loss": 4.3705,
|
|
"mean_token_accuracy": 0.24530320465564728,
|
|
"num_tokens": 19911679.0,
|
|
"step": 10925
|
|
},
|
|
{
|
|
"entropy": 5.227439069747925,
|
|
"epoch": 9.055509527754763,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.151141662482126e-05,
|
|
"loss": 4.4027,
|
|
"mean_token_accuracy": 0.24040811210870744,
|
|
"num_tokens": 19920551.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 5.247386074066162,
|
|
"epoch": 9.059652029826015,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.141072799717805e-05,
|
|
"loss": 4.4547,
|
|
"mean_token_accuracy": 0.23670902401208876,
|
|
"num_tokens": 19930187.0,
|
|
"step": 10935
|
|
},
|
|
{
|
|
"entropy": 5.235815477371216,
|
|
"epoch": 9.063794531897265,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.131047020243717e-05,
|
|
"loss": 4.4198,
|
|
"mean_token_accuracy": 0.23990993201732635,
|
|
"num_tokens": 19939988.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 5.251574087142944,
|
|
"epoch": 9.067937033968517,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.121064344282962e-05,
|
|
"loss": 4.4282,
|
|
"mean_token_accuracy": 0.24241002053022384,
|
|
"num_tokens": 19949536.0,
|
|
"step": 10945
|
|
},
|
|
{
|
|
"entropy": 5.247794151306152,
|
|
"epoch": 9.072079536039768,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.111124791971687e-05,
|
|
"loss": 4.4568,
|
|
"mean_token_accuracy": 0.24499829709529877,
|
|
"num_tokens": 19958499.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 5.248871850967407,
|
|
"epoch": 9.07622203811102,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.1012283833590465e-05,
|
|
"loss": 4.419,
|
|
"mean_token_accuracy": 0.24602134078741072,
|
|
"num_tokens": 19967235.0,
|
|
"step": 10955
|
|
},
|
|
{
|
|
"entropy": 5.261237764358521,
|
|
"epoch": 9.08036454018227,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.091375138407189e-05,
|
|
"loss": 4.4642,
|
|
"mean_token_accuracy": 0.23958979696035385,
|
|
"num_tokens": 19975829.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 5.282030296325684,
|
|
"epoch": 9.084507042253522,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.0815650769911774e-05,
|
|
"loss": 4.4788,
|
|
"mean_token_accuracy": 0.23591329008340836,
|
|
"num_tokens": 19984522.0,
|
|
"step": 10965
|
|
},
|
|
{
|
|
"entropy": 5.246661376953125,
|
|
"epoch": 9.088649544324772,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.0717982188989884e-05,
|
|
"loss": 4.4438,
|
|
"mean_token_accuracy": 0.238225257396698,
|
|
"num_tokens": 19993981.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 5.265883207321167,
|
|
"epoch": 9.092792046396022,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.062074583831432e-05,
|
|
"loss": 4.5099,
|
|
"mean_token_accuracy": 0.22879938930273055,
|
|
"num_tokens": 20002574.0,
|
|
"step": 10975
|
|
},
|
|
{
|
|
"entropy": 5.336284255981445,
|
|
"epoch": 9.096934548467274,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.052394191402145e-05,
|
|
"loss": 4.5789,
|
|
"mean_token_accuracy": 0.22337686717510224,
|
|
"num_tokens": 20012937.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 5.213770532608033,
|
|
"epoch": 9.101077050538525,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.042757061137545e-05,
|
|
"loss": 4.4145,
|
|
"mean_token_accuracy": 0.2386445254087448,
|
|
"num_tokens": 20021686.0,
|
|
"step": 10985
|
|
},
|
|
{
|
|
"entropy": 5.312621641159057,
|
|
"epoch": 9.105219552609777,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.033163212476762e-05,
|
|
"loss": 4.4959,
|
|
"mean_token_accuracy": 0.23091989010572433,
|
|
"num_tokens": 20031435.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 5.284722900390625,
|
|
"epoch": 9.109362054681027,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.023612664771651e-05,
|
|
"loss": 4.4774,
|
|
"mean_token_accuracy": 0.23144399076700212,
|
|
"num_tokens": 20040692.0,
|
|
"step": 10995
|
|
},
|
|
{
|
|
"entropy": 5.262555551528931,
|
|
"epoch": 9.113504556752279,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.014105437286705e-05,
|
|
"loss": 4.4298,
|
|
"mean_token_accuracy": 0.23655308187007903,
|
|
"num_tokens": 20049938.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"epoch": 9.113504556752279,
|
|
"eval_entropy": 5.154066635855734,
|
|
"eval_loss": 5.767139911651611,
|
|
"eval_mean_token_accuracy": 0.16739037327929093,
|
|
"eval_num_tokens": 20049938.0,
|
|
"eval_runtime": 2.763,
|
|
"eval_samples_per_second": 1356.148,
|
|
"eval_steps_per_second": 169.745,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"entropy": 5.320124387741089,
|
|
"epoch": 9.117647058823529,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.0046415491990355e-05,
|
|
"loss": 4.445,
|
|
"mean_token_accuracy": 0.24189643412828446,
|
|
"num_tokens": 20058315.0,
|
|
"step": 11005
|
|
},
|
|
{
|
|
"entropy": 5.262517356872559,
|
|
"epoch": 9.121789560894781,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.99522101959835e-05,
|
|
"loss": 4.4237,
|
|
"mean_token_accuracy": 0.23968904465436935,
|
|
"num_tokens": 20066702.0,
|
|
"step": 11010
|
|
},
|
|
{
|
|
"entropy": 5.248305225372315,
|
|
"epoch": 9.125932062966031,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.9858438674868744e-05,
|
|
"loss": 4.4294,
|
|
"mean_token_accuracy": 0.24182839542627335,
|
|
"num_tokens": 20075400.0,
|
|
"step": 11015
|
|
},
|
|
{
|
|
"entropy": 5.220608377456665,
|
|
"epoch": 9.130074565037283,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.976510111779359e-05,
|
|
"loss": 4.409,
|
|
"mean_token_accuracy": 0.24118502289056779,
|
|
"num_tokens": 20084697.0,
|
|
"step": 11020
|
|
},
|
|
{
|
|
"entropy": 5.262684965133667,
|
|
"epoch": 9.134217067108533,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.9672197713030044e-05,
|
|
"loss": 4.4365,
|
|
"mean_token_accuracy": 0.2379970446228981,
|
|
"num_tokens": 20093551.0,
|
|
"step": 11025
|
|
},
|
|
{
|
|
"entropy": 5.28862829208374,
|
|
"epoch": 9.138359569179785,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.9579728647974416e-05,
|
|
"loss": 4.4564,
|
|
"mean_token_accuracy": 0.23762071430683135,
|
|
"num_tokens": 20102964.0,
|
|
"step": 11030
|
|
},
|
|
{
|
|
"entropy": 5.282498741149903,
|
|
"epoch": 9.142502071251036,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.948769410914694e-05,
|
|
"loss": 4.4072,
|
|
"mean_token_accuracy": 0.2522799581289291,
|
|
"num_tokens": 20111439.0,
|
|
"step": 11035
|
|
},
|
|
{
|
|
"entropy": 5.17955756187439,
|
|
"epoch": 9.146644573322286,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.939609428219131e-05,
|
|
"loss": 4.4048,
|
|
"mean_token_accuracy": 0.23311796486377717,
|
|
"num_tokens": 20119712.0,
|
|
"step": 11040
|
|
},
|
|
{
|
|
"entropy": 5.22906460762024,
|
|
"epoch": 9.150787075393538,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.930492935187446e-05,
|
|
"loss": 4.4472,
|
|
"mean_token_accuracy": 0.2519031226634979,
|
|
"num_tokens": 20130024.0,
|
|
"step": 11045
|
|
},
|
|
{
|
|
"entropy": 5.210365962982178,
|
|
"epoch": 9.154929577464788,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.921419950208592e-05,
|
|
"loss": 4.4103,
|
|
"mean_token_accuracy": 0.23966276198625563,
|
|
"num_tokens": 20139394.0,
|
|
"step": 11050
|
|
},
|
|
{
|
|
"entropy": 5.284750509262085,
|
|
"epoch": 9.15907207953604,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.912390491583776e-05,
|
|
"loss": 4.477,
|
|
"mean_token_accuracy": 0.23916612416505814,
|
|
"num_tokens": 20148100.0,
|
|
"step": 11055
|
|
},
|
|
{
|
|
"entropy": 5.263777112960815,
|
|
"epoch": 9.16321458160729,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.9034045775264016e-05,
|
|
"loss": 4.4358,
|
|
"mean_token_accuracy": 0.23507414311170577,
|
|
"num_tokens": 20157361.0,
|
|
"step": 11060
|
|
},
|
|
{
|
|
"entropy": 5.225469923019409,
|
|
"epoch": 9.167357083678542,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.894462226162042e-05,
|
|
"loss": 4.4606,
|
|
"mean_token_accuracy": 0.2319960117340088,
|
|
"num_tokens": 20165788.0,
|
|
"step": 11065
|
|
},
|
|
{
|
|
"entropy": 5.209522485733032,
|
|
"epoch": 9.171499585749793,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.8855634555283954e-05,
|
|
"loss": 4.3473,
|
|
"mean_token_accuracy": 0.24744935929775239,
|
|
"num_tokens": 20174220.0,
|
|
"step": 11070
|
|
},
|
|
{
|
|
"entropy": 5.241843938827515,
|
|
"epoch": 9.175642087821045,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.876708283575254e-05,
|
|
"loss": 4.4575,
|
|
"mean_token_accuracy": 0.23421479016542435,
|
|
"num_tokens": 20183017.0,
|
|
"step": 11075
|
|
},
|
|
{
|
|
"entropy": 5.224777889251709,
|
|
"epoch": 9.179784589892295,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.867896728164468e-05,
|
|
"loss": 4.4628,
|
|
"mean_token_accuracy": 0.23608822822570802,
|
|
"num_tokens": 20192238.0,
|
|
"step": 11080
|
|
},
|
|
{
|
|
"entropy": 5.283325433731079,
|
|
"epoch": 9.183927091963547,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.859128807069905e-05,
|
|
"loss": 4.4821,
|
|
"mean_token_accuracy": 0.2398526757955551,
|
|
"num_tokens": 20201044.0,
|
|
"step": 11085
|
|
},
|
|
{
|
|
"entropy": 5.251319169998169,
|
|
"epoch": 9.188069594034797,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.8504045379774246e-05,
|
|
"loss": 4.4614,
|
|
"mean_token_accuracy": 0.23395148664712906,
|
|
"num_tokens": 20210832.0,
|
|
"step": 11090
|
|
},
|
|
{
|
|
"entropy": 5.2750688076019285,
|
|
"epoch": 9.192212096106047,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.841723938484826e-05,
|
|
"loss": 4.4098,
|
|
"mean_token_accuracy": 0.24749137461185455,
|
|
"num_tokens": 20219637.0,
|
|
"step": 11095
|
|
},
|
|
{
|
|
"entropy": 5.251072263717651,
|
|
"epoch": 9.1963545981773,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.833087026101834e-05,
|
|
"loss": 4.4237,
|
|
"mean_token_accuracy": 0.23806550204753876,
|
|
"num_tokens": 20230312.0,
|
|
"step": 11100
|
|
},
|
|
{
|
|
"entropy": 5.289454984664917,
|
|
"epoch": 9.20049710024855,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.824493818250034e-05,
|
|
"loss": 4.5246,
|
|
"mean_token_accuracy": 0.2333722934126854,
|
|
"num_tokens": 20240871.0,
|
|
"step": 11105
|
|
},
|
|
{
|
|
"entropy": 5.221672677993775,
|
|
"epoch": 9.204639602319801,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.815944332262878e-05,
|
|
"loss": 4.4257,
|
|
"mean_token_accuracy": 0.2510790377855301,
|
|
"num_tokens": 20249685.0,
|
|
"step": 11110
|
|
},
|
|
{
|
|
"entropy": 5.242713165283203,
|
|
"epoch": 9.208782104391052,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.8074385853856035e-05,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.2489153727889061,
|
|
"num_tokens": 20258426.0,
|
|
"step": 11115
|
|
},
|
|
{
|
|
"entropy": 5.267001342773438,
|
|
"epoch": 9.212924606462304,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.7989765947752384e-05,
|
|
"loss": 4.4696,
|
|
"mean_token_accuracy": 0.2297951102256775,
|
|
"num_tokens": 20268065.0,
|
|
"step": 11120
|
|
},
|
|
{
|
|
"entropy": 5.2815351486206055,
|
|
"epoch": 9.217067108533554,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.790558377500537e-05,
|
|
"loss": 4.5321,
|
|
"mean_token_accuracy": 0.22692357301712035,
|
|
"num_tokens": 20277326.0,
|
|
"step": 11125
|
|
},
|
|
{
|
|
"entropy": 5.18825249671936,
|
|
"epoch": 9.221209610604806,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.7821839505419636e-05,
|
|
"loss": 4.4004,
|
|
"mean_token_accuracy": 0.23578597754240035,
|
|
"num_tokens": 20285933.0,
|
|
"step": 11130
|
|
},
|
|
{
|
|
"entropy": 5.327149486541748,
|
|
"epoch": 9.225352112676056,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.773853330791659e-05,
|
|
"loss": 4.5452,
|
|
"mean_token_accuracy": 0.22299625277519225,
|
|
"num_tokens": 20295236.0,
|
|
"step": 11135
|
|
},
|
|
{
|
|
"entropy": 5.217405319213867,
|
|
"epoch": 9.229494614747308,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.765566535053383e-05,
|
|
"loss": 4.4387,
|
|
"mean_token_accuracy": 0.23313387334346772,
|
|
"num_tokens": 20305717.0,
|
|
"step": 11140
|
|
},
|
|
{
|
|
"entropy": 5.203705883026123,
|
|
"epoch": 9.233637116818558,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.757323580042519e-05,
|
|
"loss": 4.4318,
|
|
"mean_token_accuracy": 0.23677320331335067,
|
|
"num_tokens": 20315472.0,
|
|
"step": 11145
|
|
},
|
|
{
|
|
"entropy": 5.277881908416748,
|
|
"epoch": 9.23777961888981,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.7491244823860026e-05,
|
|
"loss": 4.4681,
|
|
"mean_token_accuracy": 0.2391311675310135,
|
|
"num_tokens": 20324127.0,
|
|
"step": 11150
|
|
},
|
|
{
|
|
"entropy": 5.240291881561279,
|
|
"epoch": 9.24192212096106,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.7409692586223105e-05,
|
|
"loss": 4.3826,
|
|
"mean_token_accuracy": 0.24853167831897735,
|
|
"num_tokens": 20333207.0,
|
|
"step": 11155
|
|
},
|
|
{
|
|
"entropy": 5.287311124801636,
|
|
"epoch": 9.24606462303231,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.7328579252014204e-05,
|
|
"loss": 4.4072,
|
|
"mean_token_accuracy": 0.2483006238937378,
|
|
"num_tokens": 20342250.0,
|
|
"step": 11160
|
|
},
|
|
{
|
|
"entropy": 5.2331396579742435,
|
|
"epoch": 9.250207125103563,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.724790498484778e-05,
|
|
"loss": 4.4105,
|
|
"mean_token_accuracy": 0.23652527183294297,
|
|
"num_tokens": 20351464.0,
|
|
"step": 11165
|
|
},
|
|
{
|
|
"entropy": 5.2211085796356205,
|
|
"epoch": 9.254349627174813,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.7167669947452654e-05,
|
|
"loss": 4.38,
|
|
"mean_token_accuracy": 0.2477695479989052,
|
|
"num_tokens": 20360089.0,
|
|
"step": 11170
|
|
},
|
|
{
|
|
"entropy": 5.224429273605347,
|
|
"epoch": 9.258492129246065,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.708787430167163e-05,
|
|
"loss": 4.3623,
|
|
"mean_token_accuracy": 0.2437489226460457,
|
|
"num_tokens": 20369728.0,
|
|
"step": 11175
|
|
},
|
|
{
|
|
"entropy": 5.21832914352417,
|
|
"epoch": 9.262634631317315,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.700851820846129e-05,
|
|
"loss": 4.4179,
|
|
"mean_token_accuracy": 0.23716152757406234,
|
|
"num_tokens": 20378512.0,
|
|
"step": 11180
|
|
},
|
|
{
|
|
"entropy": 5.211008501052857,
|
|
"epoch": 9.266777133388567,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.69296018278915e-05,
|
|
"loss": 4.4298,
|
|
"mean_token_accuracy": 0.2376560389995575,
|
|
"num_tokens": 20387383.0,
|
|
"step": 11185
|
|
},
|
|
{
|
|
"entropy": 5.267211103439331,
|
|
"epoch": 9.270919635459817,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.685112531914531e-05,
|
|
"loss": 4.5292,
|
|
"mean_token_accuracy": 0.23249492645263672,
|
|
"num_tokens": 20396834.0,
|
|
"step": 11190
|
|
},
|
|
{
|
|
"entropy": 5.218554592132568,
|
|
"epoch": 9.27506213753107,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.677308884051833e-05,
|
|
"loss": 4.4341,
|
|
"mean_token_accuracy": 0.23480421155691147,
|
|
"num_tokens": 20405235.0,
|
|
"step": 11195
|
|
},
|
|
{
|
|
"entropy": 5.1613787651062015,
|
|
"epoch": 9.27920463960232,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.669549254941865e-05,
|
|
"loss": 4.3633,
|
|
"mean_token_accuracy": 0.24047209769487382,
|
|
"num_tokens": 20414369.0,
|
|
"step": 11200
|
|
},
|
|
{
|
|
"entropy": 5.234552574157715,
|
|
"epoch": 9.283347141673572,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.661833660236657e-05,
|
|
"loss": 4.4407,
|
|
"mean_token_accuracy": 0.2397792011499405,
|
|
"num_tokens": 20423512.0,
|
|
"step": 11205
|
|
},
|
|
{
|
|
"entropy": 5.136560297012329,
|
|
"epoch": 9.287489643744822,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.6541621154993956e-05,
|
|
"loss": 4.3409,
|
|
"mean_token_accuracy": 0.2423170119524002,
|
|
"num_tokens": 20432615.0,
|
|
"step": 11210
|
|
},
|
|
{
|
|
"entropy": 5.22599458694458,
|
|
"epoch": 9.291632145816074,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.646534636204434e-05,
|
|
"loss": 4.4428,
|
|
"mean_token_accuracy": 0.2316483050584793,
|
|
"num_tokens": 20440937.0,
|
|
"step": 11215
|
|
},
|
|
{
|
|
"entropy": 5.142677640914917,
|
|
"epoch": 9.295774647887324,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.6389512377372245e-05,
|
|
"loss": 4.3314,
|
|
"mean_token_accuracy": 0.2411670595407486,
|
|
"num_tokens": 20449775.0,
|
|
"step": 11220
|
|
},
|
|
{
|
|
"entropy": 5.243865585327148,
|
|
"epoch": 9.299917149958574,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.6314119353943155e-05,
|
|
"loss": 4.4329,
|
|
"mean_token_accuracy": 0.24143245369195937,
|
|
"num_tokens": 20458876.0,
|
|
"step": 11225
|
|
},
|
|
{
|
|
"entropy": 5.20656452178955,
|
|
"epoch": 9.304059652029826,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.6239167443833004e-05,
|
|
"loss": 4.4196,
|
|
"mean_token_accuracy": 0.24094328582286834,
|
|
"num_tokens": 20467594.0,
|
|
"step": 11230
|
|
},
|
|
{
|
|
"entropy": 5.222805976867676,
|
|
"epoch": 9.308202154101076,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.6164656798227973e-05,
|
|
"loss": 4.345,
|
|
"mean_token_accuracy": 0.241726653277874,
|
|
"num_tokens": 20476417.0,
|
|
"step": 11235
|
|
},
|
|
{
|
|
"entropy": 5.274118280410766,
|
|
"epoch": 9.312344656172328,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.6090587567424247e-05,
|
|
"loss": 4.4895,
|
|
"mean_token_accuracy": 0.23779649287462234,
|
|
"num_tokens": 20486068.0,
|
|
"step": 11240
|
|
},
|
|
{
|
|
"entropy": 5.244985103607178,
|
|
"epoch": 9.316487158243579,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.60169599008275e-05,
|
|
"loss": 4.4145,
|
|
"mean_token_accuracy": 0.2295830503106117,
|
|
"num_tokens": 20495953.0,
|
|
"step": 11245
|
|
},
|
|
{
|
|
"entropy": 5.201765012741089,
|
|
"epoch": 9.32062966031483,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.594377394695277e-05,
|
|
"loss": 4.4283,
|
|
"mean_token_accuracy": 0.23919679671525956,
|
|
"num_tokens": 20505719.0,
|
|
"step": 11250
|
|
},
|
|
{
|
|
"entropy": 5.252166080474853,
|
|
"epoch": 9.32477216238608,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.587102985342416e-05,
|
|
"loss": 4.4681,
|
|
"mean_token_accuracy": 0.23162125945091247,
|
|
"num_tokens": 20514553.0,
|
|
"step": 11255
|
|
},
|
|
{
|
|
"entropy": 5.2548744678497314,
|
|
"epoch": 9.328914664457333,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.579872776697443e-05,
|
|
"loss": 4.4348,
|
|
"mean_token_accuracy": 0.23620375394821166,
|
|
"num_tokens": 20522973.0,
|
|
"step": 11260
|
|
},
|
|
{
|
|
"entropy": 5.233001708984375,
|
|
"epoch": 9.333057166528583,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.572686783344483e-05,
|
|
"loss": 4.4132,
|
|
"mean_token_accuracy": 0.24101911932229997,
|
|
"num_tokens": 20532606.0,
|
|
"step": 11265
|
|
},
|
|
{
|
|
"entropy": 5.23349609375,
|
|
"epoch": 9.337199668599835,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.565545019778468e-05,
|
|
"loss": 4.4445,
|
|
"mean_token_accuracy": 0.22929153293371202,
|
|
"num_tokens": 20541307.0,
|
|
"step": 11270
|
|
},
|
|
{
|
|
"entropy": 5.2646949768066404,
|
|
"epoch": 9.341342170671085,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.558447500405111e-05,
|
|
"loss": 4.4336,
|
|
"mean_token_accuracy": 0.24302127510309218,
|
|
"num_tokens": 20550908.0,
|
|
"step": 11275
|
|
},
|
|
{
|
|
"entropy": 5.259866523742676,
|
|
"epoch": 9.345484672742336,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.5513942395408883e-05,
|
|
"loss": 4.5247,
|
|
"mean_token_accuracy": 0.23462703973054885,
|
|
"num_tokens": 20559758.0,
|
|
"step": 11280
|
|
},
|
|
{
|
|
"entropy": 5.216423225402832,
|
|
"epoch": 9.349627174813588,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.544385251412998e-05,
|
|
"loss": 4.4295,
|
|
"mean_token_accuracy": 0.23885958939790725,
|
|
"num_tokens": 20568890.0,
|
|
"step": 11285
|
|
},
|
|
{
|
|
"entropy": 5.242618131637573,
|
|
"epoch": 9.353769676884838,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.537420550159335e-05,
|
|
"loss": 4.4076,
|
|
"mean_token_accuracy": 0.2509371846914291,
|
|
"num_tokens": 20577324.0,
|
|
"step": 11290
|
|
},
|
|
{
|
|
"entropy": 5.264973449707031,
|
|
"epoch": 9.35791217895609,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.530500149828463e-05,
|
|
"loss": 4.449,
|
|
"mean_token_accuracy": 0.23654330968856813,
|
|
"num_tokens": 20586371.0,
|
|
"step": 11295
|
|
},
|
|
{
|
|
"entropy": 5.257686805725098,
|
|
"epoch": 9.36205468102734,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.523624064379585e-05,
|
|
"loss": 4.3859,
|
|
"mean_token_accuracy": 0.23951853960752487,
|
|
"num_tokens": 20595180.0,
|
|
"step": 11300
|
|
},
|
|
{
|
|
"entropy": 5.284074354171753,
|
|
"epoch": 9.366197183098592,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.516792307682516e-05,
|
|
"loss": 4.5024,
|
|
"mean_token_accuracy": 0.2300515815615654,
|
|
"num_tokens": 20603967.0,
|
|
"step": 11305
|
|
},
|
|
{
|
|
"entropy": 5.276690053939819,
|
|
"epoch": 9.370339685169842,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.510004893517655e-05,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.23941688239574432,
|
|
"num_tokens": 20612649.0,
|
|
"step": 11310
|
|
},
|
|
{
|
|
"entropy": 5.137055540084839,
|
|
"epoch": 9.374482187241094,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.503261835575965e-05,
|
|
"loss": 4.2422,
|
|
"mean_token_accuracy": 0.2491842821240425,
|
|
"num_tokens": 20622326.0,
|
|
"step": 11315
|
|
},
|
|
{
|
|
"entropy": 5.255376529693604,
|
|
"epoch": 9.378624689312344,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.4965631474589205e-05,
|
|
"loss": 4.3944,
|
|
"mean_token_accuracy": 0.24388679563999177,
|
|
"num_tokens": 20631368.0,
|
|
"step": 11320
|
|
},
|
|
{
|
|
"entropy": 5.218535137176514,
|
|
"epoch": 9.382767191383596,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.4899088426785145e-05,
|
|
"loss": 4.4133,
|
|
"mean_token_accuracy": 0.24054847955703734,
|
|
"num_tokens": 20640874.0,
|
|
"step": 11325
|
|
},
|
|
{
|
|
"entropy": 5.255773591995239,
|
|
"epoch": 9.386909693454847,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.4832989346572046e-05,
|
|
"loss": 4.4495,
|
|
"mean_token_accuracy": 0.2333657220005989,
|
|
"num_tokens": 20650129.0,
|
|
"step": 11330
|
|
},
|
|
{
|
|
"entropy": 5.1083272933959964,
|
|
"epoch": 9.391052195526099,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.476733436727895e-05,
|
|
"loss": 4.2863,
|
|
"mean_token_accuracy": 0.2556468576192856,
|
|
"num_tokens": 20658327.0,
|
|
"step": 11335
|
|
},
|
|
{
|
|
"entropy": 5.250888395309448,
|
|
"epoch": 9.395194697597349,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.47021236213392e-05,
|
|
"loss": 4.4843,
|
|
"mean_token_accuracy": 0.23321465998888016,
|
|
"num_tokens": 20668223.0,
|
|
"step": 11340
|
|
},
|
|
{
|
|
"entropy": 5.237976789474487,
|
|
"epoch": 9.399337199668599,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.463735724028991e-05,
|
|
"loss": 4.3838,
|
|
"mean_token_accuracy": 0.25005229711532595,
|
|
"num_tokens": 20677476.0,
|
|
"step": 11345
|
|
},
|
|
{
|
|
"entropy": 5.189510536193848,
|
|
"epoch": 9.403479701739851,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.457303535477202e-05,
|
|
"loss": 4.4369,
|
|
"mean_token_accuracy": 0.24486345201730728,
|
|
"num_tokens": 20686408.0,
|
|
"step": 11350
|
|
},
|
|
{
|
|
"entropy": 5.277032804489136,
|
|
"epoch": 9.407622203811101,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.450915809452976e-05,
|
|
"loss": 4.4784,
|
|
"mean_token_accuracy": 0.236481212079525,
|
|
"num_tokens": 20694679.0,
|
|
"step": 11355
|
|
},
|
|
{
|
|
"entropy": 5.230095672607422,
|
|
"epoch": 9.411764705882353,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.444572558841053e-05,
|
|
"loss": 4.4842,
|
|
"mean_token_accuracy": 0.24030174165964127,
|
|
"num_tokens": 20703946.0,
|
|
"step": 11360
|
|
},
|
|
{
|
|
"entropy": 5.223361825942993,
|
|
"epoch": 9.415907207953603,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.4382737964364674e-05,
|
|
"loss": 4.4296,
|
|
"mean_token_accuracy": 0.22832178622484206,
|
|
"num_tokens": 20713930.0,
|
|
"step": 11365
|
|
},
|
|
{
|
|
"entropy": 5.241746425628662,
|
|
"epoch": 9.420049710024855,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.432019534944507e-05,
|
|
"loss": 4.4731,
|
|
"mean_token_accuracy": 0.2337207704782486,
|
|
"num_tokens": 20723485.0,
|
|
"step": 11370
|
|
},
|
|
{
|
|
"entropy": 5.261400461196899,
|
|
"epoch": 9.424192212096106,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 5.4258097869807054e-05,
|
|
"loss": 4.4287,
|
|
"mean_token_accuracy": 0.236363185942173,
|
|
"num_tokens": 20731906.0,
|
|
"step": 11375
|
|
},
|
|
{
|
|
"entropy": 5.361881303787231,
|
|
"epoch": 9.428334714167358,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.4196445650708e-05,
|
|
"loss": 4.5732,
|
|
"mean_token_accuracy": 0.23503864109516143,
|
|
"num_tokens": 20741808.0,
|
|
"step": 11380
|
|
},
|
|
{
|
|
"entropy": 5.25352373123169,
|
|
"epoch": 9.432477216238608,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.4135238816507154e-05,
|
|
"loss": 4.4088,
|
|
"mean_token_accuracy": 0.24084127098321914,
|
|
"num_tokens": 20751302.0,
|
|
"step": 11385
|
|
},
|
|
{
|
|
"entropy": 5.1855066299438475,
|
|
"epoch": 9.43661971830986,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.407447749066542e-05,
|
|
"loss": 4.3616,
|
|
"mean_token_accuracy": 0.24526970088481903,
|
|
"num_tokens": 20760425.0,
|
|
"step": 11390
|
|
},
|
|
{
|
|
"entropy": 5.248808908462524,
|
|
"epoch": 9.44076222038111,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.401416179574502e-05,
|
|
"loss": 4.4297,
|
|
"mean_token_accuracy": 0.23940862119197845,
|
|
"num_tokens": 20769583.0,
|
|
"step": 11395
|
|
},
|
|
{
|
|
"entropy": 5.205909729003906,
|
|
"epoch": 9.44490472245236,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.395429185340929e-05,
|
|
"loss": 4.3862,
|
|
"mean_token_accuracy": 0.2475704148411751,
|
|
"num_tokens": 20777773.0,
|
|
"step": 11400
|
|
},
|
|
{
|
|
"entropy": 5.293411111831665,
|
|
"epoch": 9.449047224523612,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.389486778442248e-05,
|
|
"loss": 4.4788,
|
|
"mean_token_accuracy": 0.23852578550577164,
|
|
"num_tokens": 20787990.0,
|
|
"step": 11405
|
|
},
|
|
{
|
|
"entropy": 5.26765284538269,
|
|
"epoch": 9.453189726594863,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.3835889708649374e-05,
|
|
"loss": 4.4853,
|
|
"mean_token_accuracy": 0.2340528577566147,
|
|
"num_tokens": 20797675.0,
|
|
"step": 11410
|
|
},
|
|
{
|
|
"entropy": 5.270390224456787,
|
|
"epoch": 9.457332228666115,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.3777357745055276e-05,
|
|
"loss": 4.4549,
|
|
"mean_token_accuracy": 0.24124883264303207,
|
|
"num_tokens": 20805977.0,
|
|
"step": 11415
|
|
},
|
|
{
|
|
"entropy": 5.222946834564209,
|
|
"epoch": 9.461474730737365,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.371927201170547e-05,
|
|
"loss": 4.4022,
|
|
"mean_token_accuracy": 0.24467591047286988,
|
|
"num_tokens": 20814996.0,
|
|
"step": 11420
|
|
},
|
|
{
|
|
"entropy": 5.316894388198852,
|
|
"epoch": 9.465617232808617,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.3661632625765284e-05,
|
|
"loss": 4.5459,
|
|
"mean_token_accuracy": 0.2271300494670868,
|
|
"num_tokens": 20824098.0,
|
|
"step": 11425
|
|
},
|
|
{
|
|
"entropy": 5.182884740829468,
|
|
"epoch": 9.469759734879867,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.360443970349965e-05,
|
|
"loss": 4.3056,
|
|
"mean_token_accuracy": 0.25463185608386996,
|
|
"num_tokens": 20832351.0,
|
|
"step": 11430
|
|
},
|
|
{
|
|
"entropy": 5.285925006866455,
|
|
"epoch": 9.473902236951119,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.3547693360272904e-05,
|
|
"loss": 4.454,
|
|
"mean_token_accuracy": 0.2455856129527092,
|
|
"num_tokens": 20841309.0,
|
|
"step": 11435
|
|
},
|
|
{
|
|
"entropy": 5.247221183776856,
|
|
"epoch": 9.47804473902237,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.349139371054865e-05,
|
|
"loss": 4.4024,
|
|
"mean_token_accuracy": 0.24143542498350143,
|
|
"num_tokens": 20850320.0,
|
|
"step": 11440
|
|
},
|
|
{
|
|
"entropy": 5.280381393432617,
|
|
"epoch": 9.482187241093621,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 5.343554086788945e-05,
|
|
"loss": 4.4648,
|
|
"mean_token_accuracy": 0.24243502467870712,
|
|
"num_tokens": 20859993.0,
|
|
"step": 11445
|
|
},
|
|
{
|
|
"entropy": 5.258852910995484,
|
|
"epoch": 9.486329743164871,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.338013494495658e-05,
|
|
"loss": 4.3991,
|
|
"mean_token_accuracy": 0.24164223074913024,
|
|
"num_tokens": 20868610.0,
|
|
"step": 11450
|
|
},
|
|
{
|
|
"entropy": 5.261618947982788,
|
|
"epoch": 9.490472245236123,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.3325176053509826e-05,
|
|
"loss": 4.495,
|
|
"mean_token_accuracy": 0.23136062622070314,
|
|
"num_tokens": 20877791.0,
|
|
"step": 11455
|
|
},
|
|
{
|
|
"entropy": 5.19754729270935,
|
|
"epoch": 9.494614747307374,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.3270664304407256e-05,
|
|
"loss": 4.3969,
|
|
"mean_token_accuracy": 0.2410440355539322,
|
|
"num_tokens": 20886732.0,
|
|
"step": 11460
|
|
},
|
|
{
|
|
"entropy": 5.258463287353516,
|
|
"epoch": 9.498757249378624,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.32165998076051e-05,
|
|
"loss": 4.4163,
|
|
"mean_token_accuracy": 0.24568518996238708,
|
|
"num_tokens": 20896090.0,
|
|
"step": 11465
|
|
},
|
|
{
|
|
"entropy": 5.2362847328186035,
|
|
"epoch": 9.502899751449876,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.316298267215728e-05,
|
|
"loss": 4.4434,
|
|
"mean_token_accuracy": 0.24002688825130464,
|
|
"num_tokens": 20905260.0,
|
|
"step": 11470
|
|
},
|
|
{
|
|
"entropy": 5.295423555374145,
|
|
"epoch": 9.507042253521126,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.310981300621547e-05,
|
|
"loss": 4.4289,
|
|
"mean_token_accuracy": 0.2354574054479599,
|
|
"num_tokens": 20915044.0,
|
|
"step": 11475
|
|
},
|
|
{
|
|
"entropy": 5.2275755405426025,
|
|
"epoch": 9.511184755592378,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.305709091702865e-05,
|
|
"loss": 4.4508,
|
|
"mean_token_accuracy": 0.23762622773647307,
|
|
"num_tokens": 20924604.0,
|
|
"step": 11480
|
|
},
|
|
{
|
|
"entropy": 5.209408330917358,
|
|
"epoch": 9.515327257663628,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.30048165109431e-05,
|
|
"loss": 4.3817,
|
|
"mean_token_accuracy": 0.24619587361812592,
|
|
"num_tokens": 20933119.0,
|
|
"step": 11485
|
|
},
|
|
{
|
|
"entropy": 5.204202461242676,
|
|
"epoch": 9.51946975973488,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.295298989340196e-05,
|
|
"loss": 4.4207,
|
|
"mean_token_accuracy": 0.24292295724153518,
|
|
"num_tokens": 20942217.0,
|
|
"step": 11490
|
|
},
|
|
{
|
|
"entropy": 5.287106323242187,
|
|
"epoch": 9.52361226180613,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.290161116894518e-05,
|
|
"loss": 4.4976,
|
|
"mean_token_accuracy": 0.23548925966024398,
|
|
"num_tokens": 20951268.0,
|
|
"step": 11495
|
|
},
|
|
{
|
|
"entropy": 5.209807300567627,
|
|
"epoch": 9.527754763877383,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.285068044120927e-05,
|
|
"loss": 4.4483,
|
|
"mean_token_accuracy": 0.2437433511018753,
|
|
"num_tokens": 20960060.0,
|
|
"step": 11500
|
|
},
|
|
{
|
|
"epoch": 9.527754763877383,
|
|
"eval_entropy": 5.148745706594829,
|
|
"eval_loss": 5.770114898681641,
|
|
"eval_mean_token_accuracy": 0.16742923058299367,
|
|
"eval_num_tokens": 20960060.0,
|
|
"eval_runtime": 3.0848,
|
|
"eval_samples_per_second": 1214.66,
|
|
"eval_steps_per_second": 152.035,
|
|
"step": 11500
|
|
},
|
|
{
|
|
"entropy": 5.239894866943359,
|
|
"epoch": 9.531897265948633,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.280019781292707e-05,
|
|
"loss": 4.4335,
|
|
"mean_token_accuracy": 0.23547269254922867,
|
|
"num_tokens": 20969024.0,
|
|
"step": 11505
|
|
},
|
|
{
|
|
"entropy": 5.260391092300415,
|
|
"epoch": 9.536039768019885,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.27501633859276e-05,
|
|
"loss": 4.454,
|
|
"mean_token_accuracy": 0.2334306389093399,
|
|
"num_tokens": 20978482.0,
|
|
"step": 11510
|
|
},
|
|
{
|
|
"entropy": 5.248519372940064,
|
|
"epoch": 9.540182270091135,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.270057726113572e-05,
|
|
"loss": 4.4237,
|
|
"mean_token_accuracy": 0.2447550818324089,
|
|
"num_tokens": 20987799.0,
|
|
"step": 11515
|
|
},
|
|
{
|
|
"entropy": 5.203705930709839,
|
|
"epoch": 9.544324772162387,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.265143953857205e-05,
|
|
"loss": 4.449,
|
|
"mean_token_accuracy": 0.22941320687532424,
|
|
"num_tokens": 20996213.0,
|
|
"step": 11520
|
|
},
|
|
{
|
|
"entropy": 5.281027746200562,
|
|
"epoch": 9.548467274233637,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.2602750317352814e-05,
|
|
"loss": 4.442,
|
|
"mean_token_accuracy": 0.23822075873613358,
|
|
"num_tokens": 21005691.0,
|
|
"step": 11525
|
|
},
|
|
{
|
|
"entropy": 5.290962553024292,
|
|
"epoch": 9.552609776304887,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.25545096956894e-05,
|
|
"loss": 4.4963,
|
|
"mean_token_accuracy": 0.23200150281190873,
|
|
"num_tokens": 21015487.0,
|
|
"step": 11530
|
|
},
|
|
{
|
|
"entropy": 5.251545667648315,
|
|
"epoch": 9.55675227837614,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.250671777088849e-05,
|
|
"loss": 4.3999,
|
|
"mean_token_accuracy": 0.24684842228889464,
|
|
"num_tokens": 21024253.0,
|
|
"step": 11535
|
|
},
|
|
{
|
|
"entropy": 5.2650390625,
|
|
"epoch": 9.56089478044739,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.245937463935159e-05,
|
|
"loss": 4.4121,
|
|
"mean_token_accuracy": 0.23628033250570296,
|
|
"num_tokens": 21032996.0,
|
|
"step": 11540
|
|
},
|
|
{
|
|
"entropy": 5.250713348388672,
|
|
"epoch": 9.565037282518642,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.241248039657496e-05,
|
|
"loss": 4.4195,
|
|
"mean_token_accuracy": 0.23175226300954818,
|
|
"num_tokens": 21042308.0,
|
|
"step": 11545
|
|
},
|
|
{
|
|
"entropy": 5.378061103820801,
|
|
"epoch": 9.569179784589892,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.236603513714943e-05,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.22443914860486985,
|
|
"num_tokens": 21052297.0,
|
|
"step": 11550
|
|
},
|
|
{
|
|
"entropy": 5.1775572299957275,
|
|
"epoch": 9.573322286661144,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.232003895476016e-05,
|
|
"loss": 4.3145,
|
|
"mean_token_accuracy": 0.2472367614507675,
|
|
"num_tokens": 21060665.0,
|
|
"step": 11555
|
|
},
|
|
{
|
|
"entropy": 5.24647274017334,
|
|
"epoch": 9.577464788732394,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.227449194218648e-05,
|
|
"loss": 4.4282,
|
|
"mean_token_accuracy": 0.24510374963283538,
|
|
"num_tokens": 21069454.0,
|
|
"step": 11560
|
|
},
|
|
{
|
|
"entropy": 5.277973318099976,
|
|
"epoch": 9.581607290803646,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.222939419130167e-05,
|
|
"loss": 4.5304,
|
|
"mean_token_accuracy": 0.22821399420499802,
|
|
"num_tokens": 21078563.0,
|
|
"step": 11565
|
|
},
|
|
{
|
|
"entropy": 5.209210395812988,
|
|
"epoch": 9.585749792874896,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.218474579307282e-05,
|
|
"loss": 4.4198,
|
|
"mean_token_accuracy": 0.23446668237447738,
|
|
"num_tokens": 21087071.0,
|
|
"step": 11570
|
|
},
|
|
{
|
|
"entropy": 5.238626766204834,
|
|
"epoch": 9.589892294946148,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.214054683756069e-05,
|
|
"loss": 4.4735,
|
|
"mean_token_accuracy": 0.23034384250640869,
|
|
"num_tokens": 21097024.0,
|
|
"step": 11575
|
|
},
|
|
{
|
|
"entropy": 5.169147539138794,
|
|
"epoch": 9.594034797017398,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.2096797413919344e-05,
|
|
"loss": 4.344,
|
|
"mean_token_accuracy": 0.251152066886425,
|
|
"num_tokens": 21105938.0,
|
|
"step": 11580
|
|
},
|
|
{
|
|
"entropy": 5.254090261459351,
|
|
"epoch": 9.59817729908865,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.205349761039613e-05,
|
|
"loss": 4.4517,
|
|
"mean_token_accuracy": 0.23605589121580123,
|
|
"num_tokens": 21115691.0,
|
|
"step": 11585
|
|
},
|
|
{
|
|
"entropy": 5.223546981811523,
|
|
"epoch": 9.6023198011599,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.201064751433156e-05,
|
|
"loss": 4.3913,
|
|
"mean_token_accuracy": 0.24444385170936583,
|
|
"num_tokens": 21123691.0,
|
|
"step": 11590
|
|
},
|
|
{
|
|
"entropy": 5.266046667098999,
|
|
"epoch": 9.606462303231151,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.19682472121589e-05,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.23368379175662995,
|
|
"num_tokens": 21132947.0,
|
|
"step": 11595
|
|
},
|
|
{
|
|
"entropy": 5.262412261962891,
|
|
"epoch": 9.610604805302403,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.192629678940421e-05,
|
|
"loss": 4.4803,
|
|
"mean_token_accuracy": 0.24401004165410994,
|
|
"num_tokens": 21142326.0,
|
|
"step": 11600
|
|
},
|
|
{
|
|
"entropy": 5.189616632461548,
|
|
"epoch": 9.614747307373653,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.188479633068607e-05,
|
|
"loss": 4.4553,
|
|
"mean_token_accuracy": 0.23606438487768172,
|
|
"num_tokens": 21151546.0,
|
|
"step": 11605
|
|
},
|
|
{
|
|
"entropy": 5.2343785762786865,
|
|
"epoch": 9.618889809444905,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.1843745919715485e-05,
|
|
"loss": 4.3836,
|
|
"mean_token_accuracy": 0.245500285923481,
|
|
"num_tokens": 21159714.0,
|
|
"step": 11610
|
|
},
|
|
{
|
|
"entropy": 5.275112962722778,
|
|
"epoch": 9.623032311516155,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.180314563929555e-05,
|
|
"loss": 4.4709,
|
|
"mean_token_accuracy": 0.23328327387571335,
|
|
"num_tokens": 21169201.0,
|
|
"step": 11615
|
|
},
|
|
{
|
|
"entropy": 5.2364133358001705,
|
|
"epoch": 9.627174813587407,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.176299557132151e-05,
|
|
"loss": 4.4146,
|
|
"mean_token_accuracy": 0.23679282367229462,
|
|
"num_tokens": 21178270.0,
|
|
"step": 11620
|
|
},
|
|
{
|
|
"entropy": 5.24067325592041,
|
|
"epoch": 9.631317315658658,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.172329579678042e-05,
|
|
"loss": 4.4213,
|
|
"mean_token_accuracy": 0.2426633968949318,
|
|
"num_tokens": 21186654.0,
|
|
"step": 11625
|
|
},
|
|
{
|
|
"entropy": 5.2697618961334225,
|
|
"epoch": 9.63545981772991,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.168404639575108e-05,
|
|
"loss": 4.4673,
|
|
"mean_token_accuracy": 0.23336609452962875,
|
|
"num_tokens": 21196074.0,
|
|
"step": 11630
|
|
},
|
|
{
|
|
"entropy": 5.259697818756104,
|
|
"epoch": 9.63960231980116,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.1645247447403826e-05,
|
|
"loss": 4.4582,
|
|
"mean_token_accuracy": 0.23122403770685196,
|
|
"num_tokens": 21204908.0,
|
|
"step": 11635
|
|
},
|
|
{
|
|
"entropy": 5.247566747665405,
|
|
"epoch": 9.64374482187241,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.1606899030000355e-05,
|
|
"loss": 4.4269,
|
|
"mean_token_accuracy": 0.23623718619346618,
|
|
"num_tokens": 21212941.0,
|
|
"step": 11640
|
|
},
|
|
{
|
|
"entropy": 5.25924162864685,
|
|
"epoch": 9.647887323943662,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.156900122089365e-05,
|
|
"loss": 4.4992,
|
|
"mean_token_accuracy": 0.2360633686184883,
|
|
"num_tokens": 21222837.0,
|
|
"step": 11645
|
|
},
|
|
{
|
|
"entropy": 5.208449745178223,
|
|
"epoch": 9.652029826014912,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.153155409652768e-05,
|
|
"loss": 4.4035,
|
|
"mean_token_accuracy": 0.241214719414711,
|
|
"num_tokens": 21232380.0,
|
|
"step": 11650
|
|
},
|
|
{
|
|
"entropy": 5.169385051727295,
|
|
"epoch": 9.656172328086164,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.1494557732437435e-05,
|
|
"loss": 4.4345,
|
|
"mean_token_accuracy": 0.2431327611207962,
|
|
"num_tokens": 21241462.0,
|
|
"step": 11655
|
|
},
|
|
{
|
|
"entropy": 5.221358585357666,
|
|
"epoch": 9.660314830157414,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.14580122032486e-05,
|
|
"loss": 4.4277,
|
|
"mean_token_accuracy": 0.23850308060646058,
|
|
"num_tokens": 21250488.0,
|
|
"step": 11660
|
|
},
|
|
{
|
|
"entropy": 5.2776342868804935,
|
|
"epoch": 9.664457332228666,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.142191758267751e-05,
|
|
"loss": 4.5076,
|
|
"mean_token_accuracy": 0.2412742555141449,
|
|
"num_tokens": 21260684.0,
|
|
"step": 11665
|
|
},
|
|
{
|
|
"entropy": 5.256235361099243,
|
|
"epoch": 9.668599834299917,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.138627394353095e-05,
|
|
"loss": 4.4888,
|
|
"mean_token_accuracy": 0.23341884315013886,
|
|
"num_tokens": 21270191.0,
|
|
"step": 11670
|
|
},
|
|
{
|
|
"entropy": 5.28235092163086,
|
|
"epoch": 9.672742336371169,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.135108135770603e-05,
|
|
"loss": 4.4457,
|
|
"mean_token_accuracy": 0.24000017940998078,
|
|
"num_tokens": 21280617.0,
|
|
"step": 11675
|
|
},
|
|
{
|
|
"entropy": 5.203524112701416,
|
|
"epoch": 9.676884838442419,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.131633989619008e-05,
|
|
"loss": 4.4444,
|
|
"mean_token_accuracy": 0.23211785405874252,
|
|
"num_tokens": 21289700.0,
|
|
"step": 11680
|
|
},
|
|
{
|
|
"entropy": 5.2637604713439945,
|
|
"epoch": 9.68102734051367,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.12820496290604e-05,
|
|
"loss": 4.4518,
|
|
"mean_token_accuracy": 0.24259986728429794,
|
|
"num_tokens": 21299197.0,
|
|
"step": 11685
|
|
},
|
|
{
|
|
"entropy": 5.292748165130615,
|
|
"epoch": 9.685169842584921,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.124821062548417e-05,
|
|
"loss": 4.5601,
|
|
"mean_token_accuracy": 0.22433626502752305,
|
|
"num_tokens": 21308392.0,
|
|
"step": 11690
|
|
},
|
|
{
|
|
"entropy": 5.208738374710083,
|
|
"epoch": 9.689312344656173,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.121482295371841e-05,
|
|
"loss": 4.405,
|
|
"mean_token_accuracy": 0.24426733702421188,
|
|
"num_tokens": 21316683.0,
|
|
"step": 11695
|
|
},
|
|
{
|
|
"entropy": 5.258275556564331,
|
|
"epoch": 9.693454846727423,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.1181886681109676e-05,
|
|
"loss": 4.4689,
|
|
"mean_token_accuracy": 0.2406328484416008,
|
|
"num_tokens": 21324931.0,
|
|
"step": 11700
|
|
},
|
|
{
|
|
"entropy": 5.214004421234131,
|
|
"epoch": 9.697597348798674,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.1149401874094035e-05,
|
|
"loss": 4.4037,
|
|
"mean_token_accuracy": 0.23817749619483947,
|
|
"num_tokens": 21333855.0,
|
|
"step": 11705
|
|
},
|
|
{
|
|
"entropy": 5.324123573303223,
|
|
"epoch": 9.701739850869926,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.111736859819692e-05,
|
|
"loss": 4.4825,
|
|
"mean_token_accuracy": 0.23803267031908035,
|
|
"num_tokens": 21342764.0,
|
|
"step": 11710
|
|
},
|
|
{
|
|
"entropy": 5.271431398391724,
|
|
"epoch": 9.705882352941176,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.10857869180329e-05,
|
|
"loss": 4.417,
|
|
"mean_token_accuracy": 0.2371239870786667,
|
|
"num_tokens": 21352010.0,
|
|
"step": 11715
|
|
},
|
|
{
|
|
"entropy": 5.236468076705933,
|
|
"epoch": 9.710024855012428,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.105465689730571e-05,
|
|
"loss": 4.4035,
|
|
"mean_token_accuracy": 0.2366127222776413,
|
|
"num_tokens": 21361287.0,
|
|
"step": 11720
|
|
},
|
|
{
|
|
"entropy": 5.3099555492401125,
|
|
"epoch": 9.714167357083678,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.1023978598808e-05,
|
|
"loss": 4.5321,
|
|
"mean_token_accuracy": 0.236946003139019,
|
|
"num_tokens": 21371244.0,
|
|
"step": 11725
|
|
},
|
|
{
|
|
"entropy": 5.285102987289429,
|
|
"epoch": 9.71830985915493,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.09937520844212e-05,
|
|
"loss": 4.4456,
|
|
"mean_token_accuracy": 0.24117545187473297,
|
|
"num_tokens": 21380524.0,
|
|
"step": 11730
|
|
},
|
|
{
|
|
"entropy": 5.246137189865112,
|
|
"epoch": 9.72245236122618,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.0963977415115586e-05,
|
|
"loss": 4.4075,
|
|
"mean_token_accuracy": 0.2410462498664856,
|
|
"num_tokens": 21388960.0,
|
|
"step": 11735
|
|
},
|
|
{
|
|
"entropy": 5.313761186599732,
|
|
"epoch": 9.726594863297432,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.0934654650949854e-05,
|
|
"loss": 4.5024,
|
|
"mean_token_accuracy": 0.23790204226970674,
|
|
"num_tokens": 21398841.0,
|
|
"step": 11740
|
|
},
|
|
{
|
|
"entropy": 5.257930660247803,
|
|
"epoch": 9.730737365368682,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.090578385107121e-05,
|
|
"loss": 4.5107,
|
|
"mean_token_accuracy": 0.2357464015483856,
|
|
"num_tokens": 21408074.0,
|
|
"step": 11745
|
|
},
|
|
{
|
|
"entropy": 5.280686664581299,
|
|
"epoch": 9.734879867439934,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.0877365073715285e-05,
|
|
"loss": 4.5149,
|
|
"mean_token_accuracy": 0.22782277911901475,
|
|
"num_tokens": 21417725.0,
|
|
"step": 11750
|
|
},
|
|
{
|
|
"entropy": 5.1883868217468265,
|
|
"epoch": 9.739022369511185,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.0849398376205797e-05,
|
|
"loss": 4.3832,
|
|
"mean_token_accuracy": 0.24534595906734466,
|
|
"num_tokens": 21427411.0,
|
|
"step": 11755
|
|
},
|
|
{
|
|
"entropy": 5.24030327796936,
|
|
"epoch": 9.743164871582437,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.082188381495463e-05,
|
|
"loss": 4.5208,
|
|
"mean_token_accuracy": 0.23463494926691056,
|
|
"num_tokens": 21437092.0,
|
|
"step": 11760
|
|
},
|
|
{
|
|
"entropy": 5.23886284828186,
|
|
"epoch": 9.747307373653687,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.0794821445461716e-05,
|
|
"loss": 4.3754,
|
|
"mean_token_accuracy": 0.24499955475330354,
|
|
"num_tokens": 21445532.0,
|
|
"step": 11765
|
|
},
|
|
{
|
|
"entropy": 5.258833885192871,
|
|
"epoch": 9.751449875724937,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.076821132231474e-05,
|
|
"loss": 4.4528,
|
|
"mean_token_accuracy": 0.23483196198940276,
|
|
"num_tokens": 21454888.0,
|
|
"step": 11770
|
|
},
|
|
{
|
|
"entropy": 5.272388029098511,
|
|
"epoch": 9.755592377796189,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.074205349918927e-05,
|
|
"loss": 4.4736,
|
|
"mean_token_accuracy": 0.24114804118871688,
|
|
"num_tokens": 21463615.0,
|
|
"step": 11775
|
|
},
|
|
{
|
|
"entropy": 5.258852767944336,
|
|
"epoch": 9.75973487986744,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.071634802884852e-05,
|
|
"loss": 4.4326,
|
|
"mean_token_accuracy": 0.24518507868051528,
|
|
"num_tokens": 21471722.0,
|
|
"step": 11780
|
|
},
|
|
{
|
|
"entropy": 5.159647369384766,
|
|
"epoch": 9.763877381938691,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.0691094963143157e-05,
|
|
"loss": 4.3886,
|
|
"mean_token_accuracy": 0.24702022969722748,
|
|
"num_tokens": 21480689.0,
|
|
"step": 11785
|
|
},
|
|
{
|
|
"entropy": 5.293799352645874,
|
|
"epoch": 9.768019884009941,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.066629435301143e-05,
|
|
"loss": 4.4778,
|
|
"mean_token_accuracy": 0.23406366854906083,
|
|
"num_tokens": 21489925.0,
|
|
"step": 11790
|
|
},
|
|
{
|
|
"entropy": 5.195842218399048,
|
|
"epoch": 9.772162386081193,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.0641946248478876e-05,
|
|
"loss": 4.3554,
|
|
"mean_token_accuracy": 0.24134149253368378,
|
|
"num_tokens": 21498744.0,
|
|
"step": 11795
|
|
},
|
|
{
|
|
"entropy": 5.223667049407959,
|
|
"epoch": 9.776304888152444,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.061805069865827e-05,
|
|
"loss": 4.409,
|
|
"mean_token_accuracy": 0.24548505991697311,
|
|
"num_tokens": 21507358.0,
|
|
"step": 11800
|
|
},
|
|
{
|
|
"entropy": 5.338182163238526,
|
|
"epoch": 9.780447390223696,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.059460775174955e-05,
|
|
"loss": 4.5843,
|
|
"mean_token_accuracy": 0.22162350118160248,
|
|
"num_tokens": 21517539.0,
|
|
"step": 11805
|
|
},
|
|
{
|
|
"entropy": 5.275141334533691,
|
|
"epoch": 9.784589892294946,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.057161745503972e-05,
|
|
"loss": 4.4733,
|
|
"mean_token_accuracy": 0.23614558279514314,
|
|
"num_tokens": 21527423.0,
|
|
"step": 11810
|
|
},
|
|
{
|
|
"entropy": 5.229053783416748,
|
|
"epoch": 9.788732394366198,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.0549079854902665e-05,
|
|
"loss": 4.4751,
|
|
"mean_token_accuracy": 0.24238475859165193,
|
|
"num_tokens": 21536648.0,
|
|
"step": 11815
|
|
},
|
|
{
|
|
"entropy": 5.263770484924317,
|
|
"epoch": 9.792874896437448,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.052699499679925e-05,
|
|
"loss": 4.4218,
|
|
"mean_token_accuracy": 0.23765358030796052,
|
|
"num_tokens": 21545364.0,
|
|
"step": 11820
|
|
},
|
|
{
|
|
"entropy": 5.27674674987793,
|
|
"epoch": 9.7970173985087,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.050536292527702e-05,
|
|
"loss": 4.4533,
|
|
"mean_token_accuracy": 0.24444084614515305,
|
|
"num_tokens": 21554198.0,
|
|
"step": 11825
|
|
},
|
|
{
|
|
"entropy": 5.297759056091309,
|
|
"epoch": 9.80115990057995,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.0484183683970235e-05,
|
|
"loss": 4.46,
|
|
"mean_token_accuracy": 0.23340023159980774,
|
|
"num_tokens": 21563410.0,
|
|
"step": 11830
|
|
},
|
|
{
|
|
"entropy": 5.244086694717407,
|
|
"epoch": 9.8053024026512,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.046345731559973e-05,
|
|
"loss": 4.5431,
|
|
"mean_token_accuracy": 0.2344498723745346,
|
|
"num_tokens": 21573139.0,
|
|
"step": 11835
|
|
},
|
|
{
|
|
"entropy": 5.3083147525787355,
|
|
"epoch": 9.809444904722453,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.044318386197286e-05,
|
|
"loss": 4.5193,
|
|
"mean_token_accuracy": 0.23404059410095215,
|
|
"num_tokens": 21583205.0,
|
|
"step": 11840
|
|
},
|
|
{
|
|
"entropy": 5.230082082748413,
|
|
"epoch": 9.813587406793703,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.042336336398341e-05,
|
|
"loss": 4.4275,
|
|
"mean_token_accuracy": 0.24171337038278579,
|
|
"num_tokens": 21592013.0,
|
|
"step": 11845
|
|
},
|
|
{
|
|
"entropy": 5.236507987976074,
|
|
"epoch": 9.817729908864955,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.040399586161144e-05,
|
|
"loss": 4.4245,
|
|
"mean_token_accuracy": 0.23745862245559693,
|
|
"num_tokens": 21600664.0,
|
|
"step": 11850
|
|
},
|
|
{
|
|
"entropy": 5.209987354278565,
|
|
"epoch": 9.821872410936205,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.038508139392338e-05,
|
|
"loss": 4.3975,
|
|
"mean_token_accuracy": 0.24147771596908568,
|
|
"num_tokens": 21609869.0,
|
|
"step": 11855
|
|
},
|
|
{
|
|
"entropy": 5.220817613601684,
|
|
"epoch": 9.826014913007457,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.0366619999071734e-05,
|
|
"loss": 4.4218,
|
|
"mean_token_accuracy": 0.2405165046453476,
|
|
"num_tokens": 21619016.0,
|
|
"step": 11860
|
|
},
|
|
{
|
|
"entropy": 5.1729076385498045,
|
|
"epoch": 9.830157415078707,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.0348611714295205e-05,
|
|
"loss": 4.4296,
|
|
"mean_token_accuracy": 0.2387221336364746,
|
|
"num_tokens": 21628095.0,
|
|
"step": 11865
|
|
},
|
|
{
|
|
"entropy": 5.2461161613464355,
|
|
"epoch": 9.83429991714996,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.03310565759184e-05,
|
|
"loss": 4.375,
|
|
"mean_token_accuracy": 0.24048026502132416,
|
|
"num_tokens": 21636824.0,
|
|
"step": 11870
|
|
},
|
|
{
|
|
"entropy": 5.234218883514404,
|
|
"epoch": 9.83844241922121,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.0313954619351996e-05,
|
|
"loss": 4.4462,
|
|
"mean_token_accuracy": 0.23448495268821717,
|
|
"num_tokens": 21645043.0,
|
|
"step": 11875
|
|
},
|
|
{
|
|
"entropy": 5.2532164573669435,
|
|
"epoch": 9.842584921292461,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.029730587909253e-05,
|
|
"loss": 4.4775,
|
|
"mean_token_accuracy": 0.23598712235689162,
|
|
"num_tokens": 21653688.0,
|
|
"step": 11880
|
|
},
|
|
{
|
|
"entropy": 5.2885498046875,
|
|
"epoch": 9.846727423363712,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.028111038872228e-05,
|
|
"loss": 4.4611,
|
|
"mean_token_accuracy": 0.24499110877513885,
|
|
"num_tokens": 21662277.0,
|
|
"step": 11885
|
|
},
|
|
{
|
|
"entropy": 5.249349927902221,
|
|
"epoch": 9.850869925434964,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 5.026536818090935e-05,
|
|
"loss": 4.4512,
|
|
"mean_token_accuracy": 0.23113121539354325,
|
|
"num_tokens": 21670778.0,
|
|
"step": 11890
|
|
},
|
|
{
|
|
"entropy": 5.244719791412353,
|
|
"epoch": 9.855012427506214,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.0250079287407486e-05,
|
|
"loss": 4.4501,
|
|
"mean_token_accuracy": 0.24167153984308243,
|
|
"num_tokens": 21681058.0,
|
|
"step": 11895
|
|
},
|
|
{
|
|
"entropy": 5.1548279285430905,
|
|
"epoch": 9.859154929577464,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.023524373905606e-05,
|
|
"loss": 4.3442,
|
|
"mean_token_accuracy": 0.25028724372386935,
|
|
"num_tokens": 21689673.0,
|
|
"step": 11900
|
|
},
|
|
{
|
|
"entropy": 5.217740678787232,
|
|
"epoch": 9.863297431648716,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.0220861565779966e-05,
|
|
"loss": 4.3549,
|
|
"mean_token_accuracy": 0.2478707477450371,
|
|
"num_tokens": 21697792.0,
|
|
"step": 11905
|
|
},
|
|
{
|
|
"entropy": 5.342480611801148,
|
|
"epoch": 9.867439933719966,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.020693279658967e-05,
|
|
"loss": 4.5254,
|
|
"mean_token_accuracy": 0.23265190720558165,
|
|
"num_tokens": 21706472.0,
|
|
"step": 11910
|
|
},
|
|
{
|
|
"entropy": 5.264366292953492,
|
|
"epoch": 9.871582435791218,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.0193457459580984e-05,
|
|
"loss": 4.492,
|
|
"mean_token_accuracy": 0.23719860464334488,
|
|
"num_tokens": 21716031.0,
|
|
"step": 11915
|
|
},
|
|
{
|
|
"entropy": 5.1936369895935055,
|
|
"epoch": 9.875724937862469,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.0180435581935156e-05,
|
|
"loss": 4.374,
|
|
"mean_token_accuracy": 0.2493791937828064,
|
|
"num_tokens": 21725072.0,
|
|
"step": 11920
|
|
},
|
|
{
|
|
"entropy": 5.2594566822052,
|
|
"epoch": 9.87986743993372,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.0167867189918727e-05,
|
|
"loss": 4.4757,
|
|
"mean_token_accuracy": 0.23802968114614487,
|
|
"num_tokens": 21734064.0,
|
|
"step": 11925
|
|
},
|
|
{
|
|
"entropy": 5.334070253372192,
|
|
"epoch": 9.88400994200497,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.015575230888351e-05,
|
|
"loss": 4.4997,
|
|
"mean_token_accuracy": 0.23724851459264756,
|
|
"num_tokens": 21743567.0,
|
|
"step": 11930
|
|
},
|
|
{
|
|
"entropy": 5.260899829864502,
|
|
"epoch": 9.888152444076223,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.014409096326654e-05,
|
|
"loss": 4.4011,
|
|
"mean_token_accuracy": 0.24281606823205948,
|
|
"num_tokens": 21752206.0,
|
|
"step": 11935
|
|
},
|
|
{
|
|
"entropy": 5.187197875976563,
|
|
"epoch": 9.892294946147473,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.0132883176590055e-05,
|
|
"loss": 4.3688,
|
|
"mean_token_accuracy": 0.24284715652465821,
|
|
"num_tokens": 21761602.0,
|
|
"step": 11940
|
|
},
|
|
{
|
|
"entropy": 5.248630523681641,
|
|
"epoch": 9.896437448218723,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.0122128971461365e-05,
|
|
"loss": 4.4629,
|
|
"mean_token_accuracy": 0.23442771136760712,
|
|
"num_tokens": 21770282.0,
|
|
"step": 11945
|
|
},
|
|
{
|
|
"entropy": 5.240181589126587,
|
|
"epoch": 9.900579950289975,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.011182836957287e-05,
|
|
"loss": 4.4407,
|
|
"mean_token_accuracy": 0.23720859736204147,
|
|
"num_tokens": 21780116.0,
|
|
"step": 11950
|
|
},
|
|
{
|
|
"entropy": 5.1811751365661625,
|
|
"epoch": 9.904722452361225,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.010198139170204e-05,
|
|
"loss": 4.3681,
|
|
"mean_token_accuracy": 0.24757199734449387,
|
|
"num_tokens": 21789656.0,
|
|
"step": 11955
|
|
},
|
|
{
|
|
"entropy": 5.206674337387085,
|
|
"epoch": 9.908864954432477,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.009258805771126e-05,
|
|
"loss": 4.4307,
|
|
"mean_token_accuracy": 0.24286702871322632,
|
|
"num_tokens": 21798981.0,
|
|
"step": 11960
|
|
},
|
|
{
|
|
"entropy": 5.284504365921021,
|
|
"epoch": 9.913007456503728,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.008364838654795e-05,
|
|
"loss": 4.3957,
|
|
"mean_token_accuracy": 0.23985276967287064,
|
|
"num_tokens": 21808472.0,
|
|
"step": 11965
|
|
},
|
|
{
|
|
"entropy": 5.2924974918365475,
|
|
"epoch": 9.91714995857498,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.007516239624437e-05,
|
|
"loss": 4.5482,
|
|
"mean_token_accuracy": 0.22847945243120193,
|
|
"num_tokens": 21818708.0,
|
|
"step": 11970
|
|
},
|
|
{
|
|
"entropy": 5.248195838928223,
|
|
"epoch": 9.92129246064623,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.006713010391773e-05,
|
|
"loss": 4.4328,
|
|
"mean_token_accuracy": 0.24362633526325225,
|
|
"num_tokens": 21828364.0,
|
|
"step": 11975
|
|
},
|
|
{
|
|
"entropy": 5.214263677597046,
|
|
"epoch": 9.925434962717482,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 5.005955152577002e-05,
|
|
"loss": 4.3765,
|
|
"mean_token_accuracy": 0.24795337915420532,
|
|
"num_tokens": 21837453.0,
|
|
"step": 11980
|
|
},
|
|
{
|
|
"entropy": 5.281455850601196,
|
|
"epoch": 9.929577464788732,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.005242667708806e-05,
|
|
"loss": 4.4976,
|
|
"mean_token_accuracy": 0.23217789381742476,
|
|
"num_tokens": 21846661.0,
|
|
"step": 11985
|
|
},
|
|
{
|
|
"entropy": 5.256623077392578,
|
|
"epoch": 9.933719966859984,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.0045755572243416e-05,
|
|
"loss": 4.4072,
|
|
"mean_token_accuracy": 0.24344514161348343,
|
|
"num_tokens": 21855627.0,
|
|
"step": 11990
|
|
},
|
|
{
|
|
"entropy": 5.2600541591644285,
|
|
"epoch": 9.937862468931234,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.0039538224692496e-05,
|
|
"loss": 4.4172,
|
|
"mean_token_accuracy": 0.2433106541633606,
|
|
"num_tokens": 21864392.0,
|
|
"step": 11995
|
|
},
|
|
{
|
|
"entropy": 5.260848665237427,
|
|
"epoch": 9.942004971002486,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.003377464697636e-05,
|
|
"loss": 4.4994,
|
|
"mean_token_accuracy": 0.22476268857717513,
|
|
"num_tokens": 21872876.0,
|
|
"step": 12000
|
|
},
|
|
{
|
|
"epoch": 9.942004971002486,
|
|
"eval_entropy": 5.153130344236329,
|
|
"eval_loss": 5.768707752227783,
|
|
"eval_mean_token_accuracy": 0.16707063359873636,
|
|
"eval_num_tokens": 21872876.0,
|
|
"eval_runtime": 2.7755,
|
|
"eval_samples_per_second": 1350.016,
|
|
"eval_steps_per_second": 168.977,
|
|
"step": 12000
|
|
}
|
|
],
|
|
"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": 3.26105597088e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|