67936528b5
Model: fpadovani/eng-latn-10mb-after-ppt-shuff-dyck-disjoint-tok-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.6219578150351541,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.216819477081298,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 3.125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 11.4704,
|
|
"mean_token_accuracy": 0.002567927382187918,
|
|
"num_tokens": 10942.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.167101860046387,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 3.03125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 11.4695,
|
|
"mean_token_accuracy": 0.002131140569690615,
|
|
"num_tokens": 21887.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.220389270782471,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 3.171875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 11.4794,
|
|
"mean_token_accuracy": 0.0033514668932184575,
|
|
"num_tokens": 34345.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.243446731567383,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 3.53125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 11.4553,
|
|
"mean_token_accuracy": 0.0024542414117604493,
|
|
"num_tokens": 45445.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.28340311050415,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 3.390625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.4364,
|
|
"mean_token_accuracy": 0.0033778926474042235,
|
|
"num_tokens": 58237.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.235557365417481,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 3.859375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.2493,
|
|
"mean_token_accuracy": 0.0032004193868488072,
|
|
"num_tokens": 68098.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.295963096618653,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 3.96875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 11.1958,
|
|
"mean_token_accuracy": 0.006015744962496683,
|
|
"num_tokens": 79867.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.354150295257568,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 3.609375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.9911,
|
|
"mean_token_accuracy": 0.014508522395044566,
|
|
"num_tokens": 91257.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.366622924804688,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.8589,
|
|
"mean_token_accuracy": 0.02584882704541087,
|
|
"num_tokens": 103474.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.452950763702393,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.6675,
|
|
"mean_token_accuracy": 0.03272002637386322,
|
|
"num_tokens": 114830.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.536296272277832,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.5343,
|
|
"mean_token_accuracy": 0.03166484478861094,
|
|
"num_tokens": 126017.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.5467098236084,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.4777,
|
|
"mean_token_accuracy": 0.03441474046558142,
|
|
"num_tokens": 138509.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.520569229125977,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.3714,
|
|
"mean_token_accuracy": 0.03507332652807236,
|
|
"num_tokens": 148353.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.503596687316895,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 10.3324,
|
|
"mean_token_accuracy": 0.041431845724582674,
|
|
"num_tokens": 160269.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.511462497711182,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 10.2291,
|
|
"mean_token_accuracy": 0.03545696176588535,
|
|
"num_tokens": 171568.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.526496124267577,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 10.1765,
|
|
"mean_token_accuracy": 0.037500371038913724,
|
|
"num_tokens": 184243.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.54613437652588,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 10.0784,
|
|
"mean_token_accuracy": 0.03511844500899315,
|
|
"num_tokens": 194887.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.554924297332764,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.9905,
|
|
"mean_token_accuracy": 0.03714885897934437,
|
|
"num_tokens": 206476.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.555365371704102,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.8628,
|
|
"mean_token_accuracy": 0.04239063672721386,
|
|
"num_tokens": 216578.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.5505690574646,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.7904,
|
|
"mean_token_accuracy": 0.0391782708466053,
|
|
"num_tokens": 226941.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.55742769241333,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.6689,
|
|
"mean_token_accuracy": 0.041145208105444905,
|
|
"num_tokens": 238060.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.557588005065918,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.5515,
|
|
"mean_token_accuracy": 0.040022865682840345,
|
|
"num_tokens": 249137.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.552389335632324,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.4573,
|
|
"mean_token_accuracy": 0.0363481679931283,
|
|
"num_tokens": 260712.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.550541019439697,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.2967,
|
|
"mean_token_accuracy": 0.03613805994391441,
|
|
"num_tokens": 271039.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.543986415863037,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.1758,
|
|
"mean_token_accuracy": 0.03463791813701391,
|
|
"num_tokens": 281353.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.52222499847412,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.0799,
|
|
"mean_token_accuracy": 0.0336436340585351,
|
|
"num_tokens": 292221.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.495210266113281,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.9689,
|
|
"mean_token_accuracy": 0.03860460445284843,
|
|
"num_tokens": 303353.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.463899612426758,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.8343,
|
|
"mean_token_accuracy": 0.04367244802415371,
|
|
"num_tokens": 314165.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.42198667526245,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6741,
|
|
"mean_token_accuracy": 0.03774042520672083,
|
|
"num_tokens": 326387.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.378685760498048,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.6144,
|
|
"mean_token_accuracy": 0.03910857178270817,
|
|
"num_tokens": 339155.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 10.322171592712403,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.442,
|
|
"mean_token_accuracy": 0.04052340723574162,
|
|
"num_tokens": 350643.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 10.243664264678955,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.3197,
|
|
"mean_token_accuracy": 0.03984756488353014,
|
|
"num_tokens": 361977.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 10.13938636779785,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.2581,
|
|
"mean_token_accuracy": 0.03751962631940842,
|
|
"num_tokens": 373604.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 10.010906314849853,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.1311,
|
|
"mean_token_accuracy": 0.03923973124474287,
|
|
"num_tokens": 384306.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.842521572113037,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.9844,
|
|
"mean_token_accuracy": 0.0418755117803812,
|
|
"num_tokens": 396479.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.654963970184326,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.8669,
|
|
"mean_token_accuracy": 0.043255141377449034,
|
|
"num_tokens": 408489.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.446887016296387,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7954,
|
|
"mean_token_accuracy": 0.04363793805241585,
|
|
"num_tokens": 419339.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.211331462860107,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7041,
|
|
"mean_token_accuracy": 0.043199213966727254,
|
|
"num_tokens": 430486.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.930498027801514,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6534,
|
|
"mean_token_accuracy": 0.04089142791926861,
|
|
"num_tokens": 440414.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.629734897613526,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.565,
|
|
"mean_token_accuracy": 0.044371229410171506,
|
|
"num_tokens": 451475.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.362038612365723,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5923,
|
|
"mean_token_accuracy": 0.0473127044737339,
|
|
"num_tokens": 462053.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.213934516906738,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 0.6484375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5438,
|
|
"mean_token_accuracy": 0.04470139406621456,
|
|
"num_tokens": 473207.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.11230001449585,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 0.6796875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5204,
|
|
"mean_token_accuracy": 0.04156735874712467,
|
|
"num_tokens": 484999.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.029958534240723,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 0.6328125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5247,
|
|
"mean_token_accuracy": 0.045830757543444636,
|
|
"num_tokens": 496772.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.934043312072754,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 0.6875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.498,
|
|
"mean_token_accuracy": 0.043144963681697845,
|
|
"num_tokens": 507782.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.86778154373169,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.4267,
|
|
"mean_token_accuracy": 0.052391791343688966,
|
|
"num_tokens": 518950.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.84511079788208,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 0.65234375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4877,
|
|
"mean_token_accuracy": 0.055992310866713524,
|
|
"num_tokens": 530795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.866034030914307,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4986,
|
|
"mean_token_accuracy": 0.054794903844594955,
|
|
"num_tokens": 542878.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.8260335445404055,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 0.63671875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.4095,
|
|
"mean_token_accuracy": 0.06074765659868717,
|
|
"num_tokens": 553682.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.705665588378906,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 0.62890625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4244,
|
|
"mean_token_accuracy": 0.06472517400979996,
|
|
"num_tokens": 564433.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.750538444519043,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 0.6484375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.4362,
|
|
"mean_token_accuracy": 0.06701958850026131,
|
|
"num_tokens": 576001.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.732649183273315,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 0.59765625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3795,
|
|
"mean_token_accuracy": 0.0689960490912199,
|
|
"num_tokens": 587138.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.777127647399903,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.3437,
|
|
"mean_token_accuracy": 0.07119518890976906,
|
|
"num_tokens": 597572.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.722668266296386,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3154,
|
|
"mean_token_accuracy": 0.07119804993271828,
|
|
"num_tokens": 608879.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.613141679763794,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 0.65625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.3541,
|
|
"mean_token_accuracy": 0.06755611971020699,
|
|
"num_tokens": 620439.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.718517589569092,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.3973,
|
|
"mean_token_accuracy": 0.07045769393444061,
|
|
"num_tokens": 632543.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.743256139755249,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.4428,
|
|
"mean_token_accuracy": 0.07540748417377471,
|
|
"num_tokens": 643369.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.690731143951416,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.3153,
|
|
"mean_token_accuracy": 0.07570057697594165,
|
|
"num_tokens": 655328.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.686294317245483,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3655,
|
|
"mean_token_accuracy": 0.07609904557466507,
|
|
"num_tokens": 665780.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.696232032775879,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2842,
|
|
"mean_token_accuracy": 0.0769994892179966,
|
|
"num_tokens": 676566.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.6681499004364015,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2739,
|
|
"mean_token_accuracy": 0.07856337130069732,
|
|
"num_tokens": 687407.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.742949819564819,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.204,
|
|
"mean_token_accuracy": 0.08214914500713348,
|
|
"num_tokens": 698162.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.601278686523438,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.201,
|
|
"mean_token_accuracy": 0.07435800954699516,
|
|
"num_tokens": 708743.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.678054904937744,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.3656,
|
|
"mean_token_accuracy": 0.08484909981489182,
|
|
"num_tokens": 719828.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.735585069656372,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 0.671875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.3253,
|
|
"mean_token_accuracy": 0.07948748283088207,
|
|
"num_tokens": 732514.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.664690399169922,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1636,
|
|
"mean_token_accuracy": 0.08530994355678559,
|
|
"num_tokens": 742391.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.560717058181763,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1707,
|
|
"mean_token_accuracy": 0.07897167801856994,
|
|
"num_tokens": 753638.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.659924840927124,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.3564,
|
|
"mean_token_accuracy": 0.08199642971158028,
|
|
"num_tokens": 764850.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.605685758590698,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.143,
|
|
"mean_token_accuracy": 0.08915193602442742,
|
|
"num_tokens": 776192.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.580406904220581,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0854,
|
|
"mean_token_accuracy": 0.08997164443135261,
|
|
"num_tokens": 787051.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.528575992584228,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.1132,
|
|
"mean_token_accuracy": 0.08350502476096153,
|
|
"num_tokens": 799281.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.5134275436401365,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.124,
|
|
"mean_token_accuracy": 0.08074113614857197,
|
|
"num_tokens": 810003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.528310012817383,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0824,
|
|
"mean_token_accuracy": 0.08352047577500343,
|
|
"num_tokens": 819749.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.540182399749756,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0848,
|
|
"mean_token_accuracy": 0.08753013238310814,
|
|
"num_tokens": 830552.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.491492366790771,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1871,
|
|
"mean_token_accuracy": 0.08871370777487755,
|
|
"num_tokens": 840495.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.468461799621582,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1005,
|
|
"mean_token_accuracy": 0.08377788811922074,
|
|
"num_tokens": 852915.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.483887434005737,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0612,
|
|
"mean_token_accuracy": 0.09137825071811675,
|
|
"num_tokens": 862762.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.449963474273682,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0182,
|
|
"mean_token_accuracy": 0.08982106223702431,
|
|
"num_tokens": 873023.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.487195777893066,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1135,
|
|
"mean_token_accuracy": 0.08343466520309448,
|
|
"num_tokens": 885281.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.404913902282715,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1066,
|
|
"mean_token_accuracy": 0.09114829823374748,
|
|
"num_tokens": 896210.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.482153987884521,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.022,
|
|
"mean_token_accuracy": 0.09178107455372811,
|
|
"num_tokens": 909028.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.419927978515625,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9525,
|
|
"mean_token_accuracy": 0.09246442466974258,
|
|
"num_tokens": 919710.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.486433124542236,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0043,
|
|
"mean_token_accuracy": 0.09004321321845055,
|
|
"num_tokens": 930019.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.388433170318604,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9292,
|
|
"mean_token_accuracy": 0.09662804156541824,
|
|
"num_tokens": 940535.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.219930601119995,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8966,
|
|
"mean_token_accuracy": 0.09820497632026673,
|
|
"num_tokens": 951390.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.334374332427979,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.9157,
|
|
"mean_token_accuracy": 0.0975546158850193,
|
|
"num_tokens": 963010.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.2931358337402346,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8682,
|
|
"mean_token_accuracy": 0.09842227175831794,
|
|
"num_tokens": 974032.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.399092769622802,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9194,
|
|
"mean_token_accuracy": 0.10364327356219291,
|
|
"num_tokens": 983954.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.277565431594849,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9341,
|
|
"mean_token_accuracy": 0.10381620079278946,
|
|
"num_tokens": 994135.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.305540657043457,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8908,
|
|
"mean_token_accuracy": 0.10408909693360328,
|
|
"num_tokens": 1006275.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.190761947631836,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7296,
|
|
"mean_token_accuracy": 0.11000855416059493,
|
|
"num_tokens": 1017767.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.202107238769531,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8501,
|
|
"mean_token_accuracy": 0.1077425830066204,
|
|
"num_tokens": 1028286.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.2581945896148685,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8519,
|
|
"mean_token_accuracy": 0.10369035303592682,
|
|
"num_tokens": 1040555.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.310864782333374,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9436,
|
|
"mean_token_accuracy": 0.10092839002609252,
|
|
"num_tokens": 1053668.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.198493099212646,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8919,
|
|
"mean_token_accuracy": 0.10193371847271919,
|
|
"num_tokens": 1066042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.229844808578491,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8927,
|
|
"mean_token_accuracy": 0.10288028344511986,
|
|
"num_tokens": 1077843.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.212245082855224,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7852,
|
|
"mean_token_accuracy": 0.10650102347135544,
|
|
"num_tokens": 1089448.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.108809137344361,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8789,
|
|
"mean_token_accuracy": 0.10020507723093033,
|
|
"num_tokens": 1100989.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.3636938571929935,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9124,
|
|
"mean_token_accuracy": 0.10427305847406387,
|
|
"num_tokens": 1113012.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.121536731719971,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8686,
|
|
"mean_token_accuracy": 0.10281460434198379,
|
|
"num_tokens": 1123551.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5408328826392644,
|
|
"eval_entropy": 7.235462856292725,
|
|
"eval_loss": 7.000098705291748,
|
|
"eval_mean_token_accuracy": 0.10328642907975749,
|
|
"eval_num_tokens": 1123551.0,
|
|
"eval_runtime": 3.6492,
|
|
"eval_samples_per_second": 963.223,
|
|
"eval_steps_per_second": 120.574,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.183961009979248,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.781,
|
|
"mean_token_accuracy": 0.10652425587177276,
|
|
"num_tokens": 1133838.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.226875591278076,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.814,
|
|
"mean_token_accuracy": 0.10197880044579506,
|
|
"num_tokens": 1146005.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.15813946723938,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8962,
|
|
"mean_token_accuracy": 0.10023846477270126,
|
|
"num_tokens": 1157328.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.096353960037232,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7064,
|
|
"mean_token_accuracy": 0.11231652200222016,
|
|
"num_tokens": 1168592.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.194019794464111,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8205,
|
|
"mean_token_accuracy": 0.10801674649119378,
|
|
"num_tokens": 1178680.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.094686985015869,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7643,
|
|
"mean_token_accuracy": 0.11055622547864914,
|
|
"num_tokens": 1188825.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.117889642715454,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.802,
|
|
"mean_token_accuracy": 0.10735909044742584,
|
|
"num_tokens": 1201375.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.267568588256836,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9092,
|
|
"mean_token_accuracy": 0.10683771148324013,
|
|
"num_tokens": 1211915.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.125114440917969,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8284,
|
|
"mean_token_accuracy": 0.11069979071617127,
|
|
"num_tokens": 1223151.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.086154651641846,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7564,
|
|
"mean_token_accuracy": 0.1088243767619133,
|
|
"num_tokens": 1234229.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.090384912490845,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7558,
|
|
"mean_token_accuracy": 0.11664472594857216,
|
|
"num_tokens": 1245298.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.033158445358277,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7494,
|
|
"mean_token_accuracy": 0.1083575077354908,
|
|
"num_tokens": 1256154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.113464641571045,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6708,
|
|
"mean_token_accuracy": 0.11131817921996116,
|
|
"num_tokens": 1267404.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.056666231155395,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.11155360341072082,
|
|
"num_tokens": 1278894.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.0011985301971436,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.698,
|
|
"mean_token_accuracy": 0.11561334431171418,
|
|
"num_tokens": 1290408.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.103354215621948,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7649,
|
|
"mean_token_accuracy": 0.11274085119366646,
|
|
"num_tokens": 1301508.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.132675695419311,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7385,
|
|
"mean_token_accuracy": 0.11607649326324462,
|
|
"num_tokens": 1312551.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.114047622680664,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8501,
|
|
"mean_token_accuracy": 0.10954068154096604,
|
|
"num_tokens": 1324876.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.035396909713745,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7259,
|
|
"mean_token_accuracy": 0.11332784742116928,
|
|
"num_tokens": 1335401.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.929111957550049,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6573,
|
|
"mean_token_accuracy": 0.113351521641016,
|
|
"num_tokens": 1348265.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.042233514785766,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6259,
|
|
"mean_token_accuracy": 0.11777078881859779,
|
|
"num_tokens": 1359078.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.8026378631591795,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5128,
|
|
"mean_token_accuracy": 0.11638505533337593,
|
|
"num_tokens": 1369499.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.9526762008667,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6241,
|
|
"mean_token_accuracy": 0.11522940248250961,
|
|
"num_tokens": 1380329.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.96287784576416,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6931,
|
|
"mean_token_accuracy": 0.11117734163999557,
|
|
"num_tokens": 1393047.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.956924057006836,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6917,
|
|
"mean_token_accuracy": 0.11352601423859596,
|
|
"num_tokens": 1405600.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.947170162200928,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6206,
|
|
"mean_token_accuracy": 0.1178322158753872,
|
|
"num_tokens": 1416670.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.968237590789795,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6379,
|
|
"mean_token_accuracy": 0.11947860792279244,
|
|
"num_tokens": 1428929.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.879861640930176,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5854,
|
|
"mean_token_accuracy": 0.11590865403413772,
|
|
"num_tokens": 1439298.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.971583414077759,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6593,
|
|
"mean_token_accuracy": 0.12206053733825684,
|
|
"num_tokens": 1449853.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.976396465301514,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7355,
|
|
"mean_token_accuracy": 0.11254098638892174,
|
|
"num_tokens": 1460816.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.860588884353637,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5909,
|
|
"mean_token_accuracy": 0.1176596887409687,
|
|
"num_tokens": 1473003.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.978535223007202,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7073,
|
|
"mean_token_accuracy": 0.11083203330636024,
|
|
"num_tokens": 1484745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.940826559066773,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5861,
|
|
"mean_token_accuracy": 0.11994393318891525,
|
|
"num_tokens": 1495803.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.849512815475464,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5984,
|
|
"mean_token_accuracy": 0.1257324993610382,
|
|
"num_tokens": 1507504.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.007606935501099,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6877,
|
|
"mean_token_accuracy": 0.11458093971014023,
|
|
"num_tokens": 1518765.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.8483922481536865,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5605,
|
|
"mean_token_accuracy": 0.12220490351319313,
|
|
"num_tokens": 1529481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.830742645263672,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5844,
|
|
"mean_token_accuracy": 0.12169603258371353,
|
|
"num_tokens": 1539612.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.952711391448974,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7252,
|
|
"mean_token_accuracy": 0.1164122186601162,
|
|
"num_tokens": 1551370.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.95903000831604,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6168,
|
|
"mean_token_accuracy": 0.1276045598089695,
|
|
"num_tokens": 1561662.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.883334970474243,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5939,
|
|
"mean_token_accuracy": 0.12063917517662048,
|
|
"num_tokens": 1573479.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.813182020187378,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5471,
|
|
"mean_token_accuracy": 0.12155145853757858,
|
|
"num_tokens": 1583684.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.886851835250854,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5924,
|
|
"mean_token_accuracy": 0.11929144784808159,
|
|
"num_tokens": 1595694.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.864528703689575,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5786,
|
|
"mean_token_accuracy": 0.11884318962693215,
|
|
"num_tokens": 1606104.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.854226732254029,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6524,
|
|
"mean_token_accuracy": 0.12053779140114784,
|
|
"num_tokens": 1617368.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.926800298690796,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6771,
|
|
"mean_token_accuracy": 0.11467185989022255,
|
|
"num_tokens": 1630485.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.893377113342285,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5096,
|
|
"mean_token_accuracy": 0.12203627899289131,
|
|
"num_tokens": 1641411.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.7226519107818605,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5406,
|
|
"mean_token_accuracy": 0.1242325022816658,
|
|
"num_tokens": 1653737.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.925417041778564,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.7243,
|
|
"mean_token_accuracy": 0.11538803800940514,
|
|
"num_tokens": 1665910.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.879580736160278,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4739,
|
|
"mean_token_accuracy": 0.1237889863550663,
|
|
"num_tokens": 1676927.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.77749228477478,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5252,
|
|
"mean_token_accuracy": 0.1222883902490139,
|
|
"num_tokens": 1688167.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.693407392501831,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6017,
|
|
"mean_token_accuracy": 0.12450111284852028,
|
|
"num_tokens": 1699899.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.863288259506225,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5193,
|
|
"mean_token_accuracy": 0.12551374658942221,
|
|
"num_tokens": 1709862.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.761876726150513,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4988,
|
|
"mean_token_accuracy": 0.12482719644904136,
|
|
"num_tokens": 1719902.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.813378047943115,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5149,
|
|
"mean_token_accuracy": 0.12510257288813592,
|
|
"num_tokens": 1730351.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.691298294067383,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4878,
|
|
"mean_token_accuracy": 0.12052481323480606,
|
|
"num_tokens": 1741783.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.732113790512085,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5123,
|
|
"mean_token_accuracy": 0.12696686387062073,
|
|
"num_tokens": 1752952.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.788561725616455,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5427,
|
|
"mean_token_accuracy": 0.1241848573088646,
|
|
"num_tokens": 1764014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.777952671051025,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5151,
|
|
"mean_token_accuracy": 0.12806738466024398,
|
|
"num_tokens": 1776154.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.786651563644409,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5609,
|
|
"mean_token_accuracy": 0.1211055114865303,
|
|
"num_tokens": 1787292.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.721488761901855,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5258,
|
|
"mean_token_accuracy": 0.12567294761538506,
|
|
"num_tokens": 1799469.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.660627269744873,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4041,
|
|
"mean_token_accuracy": 0.1303122013807297,
|
|
"num_tokens": 1810091.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.775736951828003,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5669,
|
|
"mean_token_accuracy": 0.12215637862682342,
|
|
"num_tokens": 1820835.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.75370192527771,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5196,
|
|
"mean_token_accuracy": 0.12661589086055755,
|
|
"num_tokens": 1831868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.739461183547974,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4906,
|
|
"mean_token_accuracy": 0.12791995331645012,
|
|
"num_tokens": 1842040.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.805172634124756,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5726,
|
|
"mean_token_accuracy": 0.12422277256846428,
|
|
"num_tokens": 1853765.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.756449556350708,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5716,
|
|
"mean_token_accuracy": 0.12897725030779839,
|
|
"num_tokens": 1866081.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.707052612304688,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.469,
|
|
"mean_token_accuracy": 0.1314797580242157,
|
|
"num_tokens": 1876754.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.833410263061523,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5774,
|
|
"mean_token_accuracy": 0.1273835837841034,
|
|
"num_tokens": 1888299.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.547404336929321,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3498,
|
|
"mean_token_accuracy": 0.13612620383501053,
|
|
"num_tokens": 1899027.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.767558288574219,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.12598962038755418,
|
|
"num_tokens": 1911043.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.700322914123535,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4153,
|
|
"mean_token_accuracy": 0.12619624137878419,
|
|
"num_tokens": 1921295.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.681533527374268,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5112,
|
|
"mean_token_accuracy": 0.12896694540977477,
|
|
"num_tokens": 1932944.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.636178922653198,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4548,
|
|
"mean_token_accuracy": 0.12324619740247726,
|
|
"num_tokens": 1944072.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.706494665145874,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4454,
|
|
"mean_token_accuracy": 0.13123676255345346,
|
|
"num_tokens": 1954798.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.63202395439148,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4004,
|
|
"mean_token_accuracy": 0.13771043270826339,
|
|
"num_tokens": 1965304.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.687908601760864,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4533,
|
|
"mean_token_accuracy": 0.12809950411319732,
|
|
"num_tokens": 1976797.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.68499231338501,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.465,
|
|
"mean_token_accuracy": 0.13124867752194405,
|
|
"num_tokens": 1987448.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.453022909164429,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.12852491885423661,
|
|
"num_tokens": 2000948.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.707898950576782,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4053,
|
|
"mean_token_accuracy": 0.13699749186635019,
|
|
"num_tokens": 2011807.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.635624694824219,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.485,
|
|
"mean_token_accuracy": 0.12594227939844133,
|
|
"num_tokens": 2022835.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.493094444274902,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3887,
|
|
"mean_token_accuracy": 0.13212062120437623,
|
|
"num_tokens": 2034331.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.793362283706665,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4011,
|
|
"mean_token_accuracy": 0.12509146854281425,
|
|
"num_tokens": 2047002.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.472154426574707,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3309,
|
|
"mean_token_accuracy": 0.13486048579216003,
|
|
"num_tokens": 2056299.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.6369061946868895,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4731,
|
|
"mean_token_accuracy": 0.12534581273794174,
|
|
"num_tokens": 2068031.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.65595801671346,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3882,
|
|
"mean_token_accuracy": 0.13124301036198935,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.586441087722778,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2891,
|
|
"mean_token_accuracy": 0.13704440146684646,
|
|
"num_tokens": 2087316.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.637719631195068,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3095,
|
|
"mean_token_accuracy": 0.12899292260408401,
|
|
"num_tokens": 2098522.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.479368448257446,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2677,
|
|
"mean_token_accuracy": 0.13769578710198402,
|
|
"num_tokens": 2110015.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.62018051147461,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3805,
|
|
"mean_token_accuracy": 0.1309829458594322,
|
|
"num_tokens": 2121333.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.606573867797851,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2564,
|
|
"mean_token_accuracy": 0.14038568511605262,
|
|
"num_tokens": 2131385.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.51173095703125,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3255,
|
|
"mean_token_accuracy": 0.13718775436282157,
|
|
"num_tokens": 2143274.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.524426651000977,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1832,
|
|
"mean_token_accuracy": 0.14308830797672273,
|
|
"num_tokens": 2154529.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.507630968093872,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3404,
|
|
"mean_token_accuracy": 0.13590393140912055,
|
|
"num_tokens": 2165949.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.510678100585937,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.301,
|
|
"mean_token_accuracy": 0.13782548978924752,
|
|
"num_tokens": 2176710.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.57047290802002,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3054,
|
|
"mean_token_accuracy": 0.1405063934624195,
|
|
"num_tokens": 2189470.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.468301725387573,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2417,
|
|
"mean_token_accuracy": 0.1447782188653946,
|
|
"num_tokens": 2200266.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.555663108825684,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2552,
|
|
"mean_token_accuracy": 0.13506530076265336,
|
|
"num_tokens": 2210736.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.565949821472168,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3632,
|
|
"mean_token_accuracy": 0.13319521099328996,
|
|
"num_tokens": 2221057.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.469114494323731,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3534,
|
|
"mean_token_accuracy": 0.13533424213528633,
|
|
"num_tokens": 2232352.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.635223293304444,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2648,
|
|
"mean_token_accuracy": 0.14019849002361298,
|
|
"num_tokens": 2244655.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.261687904054468,
|
|
"eval_loss": 6.522247314453125,
|
|
"eval_mean_token_accuracy": 0.13498601920225403,
|
|
"eval_num_tokens": 2244655.0,
|
|
"eval_runtime": 3.5336,
|
|
"eval_samples_per_second": 994.742,
|
|
"eval_steps_per_second": 124.52,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.392566728591919,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.2215,
|
|
"mean_token_accuracy": 0.14575545713305474,
|
|
"num_tokens": 2255699.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.515384817123413,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.215,
|
|
"mean_token_accuracy": 0.14548230916261673,
|
|
"num_tokens": 2266627.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.574292850494385,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.3975,
|
|
"mean_token_accuracy": 0.13272227719426155,
|
|
"num_tokens": 2277814.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.601232385635376,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.13475676402449607,
|
|
"num_tokens": 2287344.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.318281745910644,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 6.1409,
|
|
"mean_token_accuracy": 0.1440238431096077,
|
|
"num_tokens": 2298332.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.4605019092559814,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 6.2482,
|
|
"mean_token_accuracy": 0.13886248245835303,
|
|
"num_tokens": 2309161.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.49772047996521,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.2395,
|
|
"mean_token_accuracy": 0.14152603298425676,
|
|
"num_tokens": 2319516.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.387350988388062,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 6.1364,
|
|
"mean_token_accuracy": 0.14436752647161483,
|
|
"num_tokens": 2330441.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.414741897583008,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.1448,
|
|
"mean_token_accuracy": 0.14739955514669417,
|
|
"num_tokens": 2340524.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.413249206542969,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 6.1574,
|
|
"mean_token_accuracy": 0.1391792982816696,
|
|
"num_tokens": 2353379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.415913724899292,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.3375,
|
|
"mean_token_accuracy": 0.13021384850144385,
|
|
"num_tokens": 2365708.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.4221429347991945,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 6.2353,
|
|
"mean_token_accuracy": 0.14104826971888543,
|
|
"num_tokens": 2375630.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.410649967193604,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.2265,
|
|
"mean_token_accuracy": 0.13970739394426346,
|
|
"num_tokens": 2386885.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.3925048351287845,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 6.2582,
|
|
"mean_token_accuracy": 0.14038242176175117,
|
|
"num_tokens": 2397676.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.471703100204468,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.2718,
|
|
"mean_token_accuracy": 0.13743244409561156,
|
|
"num_tokens": 2409039.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.462762975692749,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.3006,
|
|
"mean_token_accuracy": 0.13881117701530457,
|
|
"num_tokens": 2420103.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.446409559249878,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.14210361912846564,
|
|
"num_tokens": 2430792.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.448412847518921,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.2317,
|
|
"mean_token_accuracy": 0.1363034226000309,
|
|
"num_tokens": 2442165.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.411017274856567,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 6.1392,
|
|
"mean_token_accuracy": 0.14295777827501296,
|
|
"num_tokens": 2454231.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.395845222473144,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 6.1973,
|
|
"mean_token_accuracy": 0.14889837354421614,
|
|
"num_tokens": 2464828.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.362335443496704,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 6.247,
|
|
"mean_token_accuracy": 0.13055480644106865,
|
|
"num_tokens": 2476144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.556964492797851,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 6.4243,
|
|
"mean_token_accuracy": 0.13321975469589234,
|
|
"num_tokens": 2489116.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.495721340179443,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.2709,
|
|
"mean_token_accuracy": 0.1425547793507576,
|
|
"num_tokens": 2501609.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.316252708435059,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.1623,
|
|
"mean_token_accuracy": 0.14786348417401313,
|
|
"num_tokens": 2513338.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.502756214141845,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 6.1935,
|
|
"mean_token_accuracy": 0.14928646236658097,
|
|
"num_tokens": 2523199.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.430925273895264,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 6.2268,
|
|
"mean_token_accuracy": 0.13952379450201988,
|
|
"num_tokens": 2535136.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.347667121887207,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.1183,
|
|
"mean_token_accuracy": 0.1378582313656807,
|
|
"num_tokens": 2546428.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.522240066528321,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 6.2987,
|
|
"mean_token_accuracy": 0.13986383080482484,
|
|
"num_tokens": 2558304.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.215605926513672,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 5.9978,
|
|
"mean_token_accuracy": 0.14998608231544494,
|
|
"num_tokens": 2568717.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.445933389663696,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.2162,
|
|
"mean_token_accuracy": 0.13953494057059287,
|
|
"num_tokens": 2578748.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.396315002441407,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 6.1849,
|
|
"mean_token_accuracy": 0.1391325779259205,
|
|
"num_tokens": 2590095.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.419918870925903,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 6.2344,
|
|
"mean_token_accuracy": 0.14044754207134247,
|
|
"num_tokens": 2600929.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.391916799545288,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 6.2334,
|
|
"mean_token_accuracy": 0.14248308837413787,
|
|
"num_tokens": 2611830.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.340531587600708,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 6.1622,
|
|
"mean_token_accuracy": 0.14207693040370942,
|
|
"num_tokens": 2623177.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.31855001449585,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 6.1629,
|
|
"mean_token_accuracy": 0.14008935093879699,
|
|
"num_tokens": 2635563.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.418999481201172,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 6.1535,
|
|
"mean_token_accuracy": 0.15044649839401245,
|
|
"num_tokens": 2647157.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.273387622833252,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 6.1151,
|
|
"mean_token_accuracy": 0.15164786130189895,
|
|
"num_tokens": 2658175.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.34501576423645,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 6.0843,
|
|
"mean_token_accuracy": 0.1456107623875141,
|
|
"num_tokens": 2669458.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.4375499248504635,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.2834,
|
|
"mean_token_accuracy": 0.14490200877189635,
|
|
"num_tokens": 2679727.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.378272485733032,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 6.2372,
|
|
"mean_token_accuracy": 0.1412258230149746,
|
|
"num_tokens": 2691031.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.2971197128295895,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.15455402433872223,
|
|
"num_tokens": 2701699.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.32627272605896,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 6.0922,
|
|
"mean_token_accuracy": 0.14536757618188859,
|
|
"num_tokens": 2713038.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.282141542434692,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 6.095,
|
|
"mean_token_accuracy": 0.144093357026577,
|
|
"num_tokens": 2723685.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.338719892501831,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 6.1619,
|
|
"mean_token_accuracy": 0.14463287889957427,
|
|
"num_tokens": 2735341.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.319677209854126,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 6.0942,
|
|
"mean_token_accuracy": 0.15070197135210037,
|
|
"num_tokens": 2746052.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.301837205886841,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 6.0911,
|
|
"mean_token_accuracy": 0.14471486508846282,
|
|
"num_tokens": 2757155.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.300720977783203,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.1586,
|
|
"mean_token_accuracy": 0.14146095067262648,
|
|
"num_tokens": 2769631.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.46447868347168,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 6.2484,
|
|
"mean_token_accuracy": 0.1465391218662262,
|
|
"num_tokens": 2780797.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.2404039859771725,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 6.0971,
|
|
"mean_token_accuracy": 0.1423661321401596,
|
|
"num_tokens": 2792721.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.312399816513062,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 6.1613,
|
|
"mean_token_accuracy": 0.15470626205205917,
|
|
"num_tokens": 2804531.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.421265363693237,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 6.1487,
|
|
"mean_token_accuracy": 0.1429950214922428,
|
|
"num_tokens": 2816666.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.304442501068115,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 6.1735,
|
|
"mean_token_accuracy": 0.151274611055851,
|
|
"num_tokens": 2827977.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.348642826080322,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 6.1448,
|
|
"mean_token_accuracy": 0.1445183590054512,
|
|
"num_tokens": 2839069.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.365104818344117,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 6.2275,
|
|
"mean_token_accuracy": 0.13933413326740265,
|
|
"num_tokens": 2850632.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.286058044433593,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 6.1057,
|
|
"mean_token_accuracy": 0.1429712951183319,
|
|
"num_tokens": 2862204.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.296426582336426,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 6.0879,
|
|
"mean_token_accuracy": 0.15390856564044952,
|
|
"num_tokens": 2874010.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.331337833404541,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.15620552152395248,
|
|
"num_tokens": 2884315.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.407461166381836,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.2916,
|
|
"mean_token_accuracy": 0.1400587186217308,
|
|
"num_tokens": 2896327.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.271218967437744,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 6.0528,
|
|
"mean_token_accuracy": 0.14465109407901763,
|
|
"num_tokens": 2907007.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.302019453048706,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 6.1003,
|
|
"mean_token_accuracy": 0.1535784885287285,
|
|
"num_tokens": 2918852.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.250105190277099,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.15070367753505706,
|
|
"num_tokens": 2929948.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.244026756286621,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 6.0828,
|
|
"mean_token_accuracy": 0.15208643525838852,
|
|
"num_tokens": 2941299.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.181199932098389,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 6.0191,
|
|
"mean_token_accuracy": 0.14539152830839158,
|
|
"num_tokens": 2952283.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.554185962677002,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 6.3791,
|
|
"mean_token_accuracy": 0.1393312931060791,
|
|
"num_tokens": 2962241.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.257846307754517,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 6.1338,
|
|
"mean_token_accuracy": 0.14794275015592576,
|
|
"num_tokens": 2973408.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.329844570159912,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 6.1686,
|
|
"mean_token_accuracy": 0.14748868122696876,
|
|
"num_tokens": 2985877.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.380424976348877,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 6.1645,
|
|
"mean_token_accuracy": 0.1469645656645298,
|
|
"num_tokens": 2996935.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.20194149017334,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 6.044,
|
|
"mean_token_accuracy": 0.15571995675563813,
|
|
"num_tokens": 3007932.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.381639242172241,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 6.1396,
|
|
"mean_token_accuracy": 0.14735481142997742,
|
|
"num_tokens": 3019368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.361180973052979,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 6.1941,
|
|
"mean_token_accuracy": 0.1442886658012867,
|
|
"num_tokens": 3029787.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.15062255859375,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 6.1244,
|
|
"mean_token_accuracy": 0.1482379414141178,
|
|
"num_tokens": 3042100.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.291181373596191,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 6.1695,
|
|
"mean_token_accuracy": 0.14743669778108598,
|
|
"num_tokens": 3054160.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.321239900588989,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.14769912362098694,
|
|
"num_tokens": 3065401.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.224961471557617,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 6.0394,
|
|
"mean_token_accuracy": 0.1538393899798393,
|
|
"num_tokens": 3076137.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.356675624847412,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 6.1769,
|
|
"mean_token_accuracy": 0.15344239473342897,
|
|
"num_tokens": 3087274.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.313038015365601,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.14906611144542695,
|
|
"num_tokens": 3098476.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.2663798332214355,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 6.0968,
|
|
"mean_token_accuracy": 0.14780760034918786,
|
|
"num_tokens": 3110060.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.143105459213257,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.991,
|
|
"mean_token_accuracy": 0.1502829059958458,
|
|
"num_tokens": 3121032.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.303813886642456,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 6.0284,
|
|
"mean_token_accuracy": 0.1546885773539543,
|
|
"num_tokens": 3131999.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.118764638900757,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.9459,
|
|
"mean_token_accuracy": 0.15407799631357194,
|
|
"num_tokens": 3142693.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.180855655670166,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.14824497997760772,
|
|
"num_tokens": 3154179.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.272782230377198,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.15759131461381912,
|
|
"num_tokens": 3165314.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.201319456100464,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 6.1189,
|
|
"mean_token_accuracy": 0.15706945806741715,
|
|
"num_tokens": 3177646.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.315565681457519,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 6.1126,
|
|
"mean_token_accuracy": 0.15531182885169983,
|
|
"num_tokens": 3188401.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.154540348052978,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.9911,
|
|
"mean_token_accuracy": 0.15018533393740655,
|
|
"num_tokens": 3200206.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.307532501220703,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 6.1091,
|
|
"mean_token_accuracy": 0.14888520240783693,
|
|
"num_tokens": 3211981.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.18949179649353,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 6.1167,
|
|
"mean_token_accuracy": 0.14888323992490768,
|
|
"num_tokens": 3221997.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.322359466552735,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 6.0598,
|
|
"mean_token_accuracy": 0.14826754182577134,
|
|
"num_tokens": 3234333.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.197854804992676,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 6.1046,
|
|
"mean_token_accuracy": 0.1452392354607582,
|
|
"num_tokens": 3245083.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.279910755157471,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.9312,
|
|
"mean_token_accuracy": 0.157687509059906,
|
|
"num_tokens": 3255787.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.1504504680633545,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 6.1119,
|
|
"mean_token_accuracy": 0.14946537464857101,
|
|
"num_tokens": 3267793.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.191984701156616,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.982,
|
|
"mean_token_accuracy": 0.15891378968954087,
|
|
"num_tokens": 3279100.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.246756458282471,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 6.1383,
|
|
"mean_token_accuracy": 0.15445440858602524,
|
|
"num_tokens": 3290251.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.197033500671386,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 6.0118,
|
|
"mean_token_accuracy": 0.15319959670305253,
|
|
"num_tokens": 3301817.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.270974349975586,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 6.0553,
|
|
"mean_token_accuracy": 0.15082636773586272,
|
|
"num_tokens": 3313435.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.237086534500122,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 6.1823,
|
|
"mean_token_accuracy": 0.14484680742025374,
|
|
"num_tokens": 3325489.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.332687664031982,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 6.0399,
|
|
"mean_token_accuracy": 0.1546601414680481,
|
|
"num_tokens": 3337030.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.088555431365966,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 6.0114,
|
|
"mean_token_accuracy": 0.15409026592969893,
|
|
"num_tokens": 3347766.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.178443479537964,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.1566903904080391,
|
|
"num_tokens": 3360379.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.149486923217774,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 6.0586,
|
|
"mean_token_accuracy": 0.15657030493021012,
|
|
"num_tokens": 3370962.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6219578150351541,
|
|
"eval_entropy": 6.046384253285148,
|
|
"eval_loss": 6.249795913696289,
|
|
"eval_mean_token_accuracy": 0.15230790820311416,
|
|
"eval_num_tokens": 3370962.0,
|
|
"eval_runtime": 3.5261,
|
|
"eval_samples_per_second": 996.847,
|
|
"eval_steps_per_second": 124.783,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"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": 775440484761600.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|