67936528b5
Model: fpadovani/eng-latn-10mb-after-ppt-shuff-dyck-disjoint-tok-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.5408328826392644,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 259611278376960.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|